issn: 2180-1053 vol. 4 no. 1 january-june 2012 comparison between grinding and lapping of machined part surface roughness in micro and nano scale 91 comparison between grinding and lapping of machined part surface roughness in micro and nano scale walid mahmoud shewakh production technology department -industrial education college beni-suef university, industrial engineering department –jazan university. email: waleedshewakh@hotmail.com abstract micro and nano surface finish has become an important parameter in semiconductor, optical, electrical and mechanical industries. in this work a comparison between two traditional finishing processes grinding and lapping was made. machined parts surface roughness in micro and nano scale has been measured using two different devises in two different directions normal and perpendicular to the machining direction. results show that the traditional finishing processes are not suitable for nano scale surface finish. there is a significant difference between the normal and perpendicular measured surface roughness in nano and micro scale. keywords: lapping, grinding, surface roughness, micro, nano scale. 1.0 introduction final finishing operations in manufacturing of precise parts are always of concern owing to their most critical, intensive labor and least controllable nature. in the era of nanotechnology, deterministic high precision finishing methods are of utmost importance. the need for high precision in manufacturing was felt by manufacturers worldwide to improve interchangeability of components, improve quality control and longer fatigue life [1]. taniguchi [2] reviewed the historical progress of achievable machining accuracy during the last century. the machining processes were classifieds into three categories on the basis of achievable accuracy conventional machining, precision machining and ultraprecision machining. ultraprecision machining are the processes by which the highest possible dimensional accuracy is achieved at a given point of time. this is a relative definition which varies with time. it was predicted that by 2000 ad, the machining issn: 2180-1053 vol. 4 no. 1 january-june 2012 journal of mechanical engineering and technology 92 accuracies in conventional processes would reach 1 µm, while in precision and ultraprecision machining would reach 0.01µm (10 nm) and 0.001µm (1 nm) respectively [2]. the study of micro and nano surface metrology is becoming common in industrial and research environments as structures and surface features become smaller and smaller [3,4,5,]. scanning interferometry is becoming increasingly important in metrology analysis because of various factors such as ; the possibility of non-destructive measurement no sample contact or preparations are required [6]; its accurate and quantitative surface characterization; the fast and convenient sample loading and set-up; the capability of measuring a wide range of materials; high resolution; highly repeatable measurements; fully automated measurement – ideal for process control; performing roughness and step height analysis within a single measurement; the possibility of surface coating measurement – film thickness and real surface roughness measurement. it can address many of the challenging measurement problems that exist when studying samples at the micro and nano scale [7]. these include the measurement of critical dimensions, heights, angles, surface roughness, solving etch rate/time problems, measuring stress gradients, etc [8]. roughness is an important parameter for sample properties control. various roughness ranges are normally studied in order to define the overall properties of the surface and one of the limitations to the analysis is the bandwidth of the measurement method [9]. it is very important to accurately evaluate the quantities values of surface roughness, to determine the possibility of their usage and quality of products, to measure the effective height of surface roughness, a scanning microscope is used [10, 11]. comparing the surface roughness machined with traditional finishing, grinding and lapping in small scale can help in choosing the finishing methods used in production of small parts. in this work a comparison between two traditional finishing processes, grinding and lapping was made. machined part surface roughness in micro and nano scale using two different devises was measured in two different directions normal and perpendicular to the machining direction. to show how suitable grinding and lapping is suitable in finishing parts in nano scale. issn: 2180-1053 vol. 4 no. 1 january-june 2012 comparison between grinding and lapping of machined part surface roughness in micro and nano scale 93 2.0 experiential work a set of experiments were carried out to compare the measurement results of micro and nano grinding surface roughness. samples were processed implementing various methods and their surface roughness were measured using a special measurement device “surfcorder se 1200 fig(1) and a multi-microscopic scanner cmm-2000 fig(2)”. during this experiment the main difficulty was the selection of samples’ parameter. initially they should be big enough to be stably mounted during the machining process, as well as fitting the space of measurement devices. in this case, we have used samples of steel 45 with dimensions of 10 mm in length, 8 mm in width and a thickness of 2 mm were used. these parameters were selected according to microscopic scanner capability. the workpice has been machined on a shaping machine then abrasive processing processes grinding or lapping were used. silicon carbide 55c and grain size of 20 µm. were used in grinding and lapping. 78 45 with dimensions of 10 mm in length, 8 mm in width and a thickness of 2 mm were used. these parameters were selected according to microscopic scanner capability. the workpice has been machined on a shaping machine then abrasive processing processes grinding or lapping were used. silicon carbide 55c and grain size of 20 µm. were used in grinding and lapping. fig.1. surfcorder se 1200 fig.2. multi-microscopic scanner cmm-2000 3.0 results analysis a “surfcorder se 1200” of kosaka lab (japan) was used to measure the micro surface roughness. to define the surface nano characteristics a scanning microscope was used “cmm 2000” manufactured by proton –miet (russia). the workpiece surface roughness (rz) after shaping process of the profilemeter was 2,998 µm, in the direction of machining, and 3,311 µm perpendicular. this value shows the significant effect of shaping cutting tool. it is not possible to test it on the cmm-2000 scanner, as the resulted values of (rz) are higher that allowed measurement rang (2 µm). the results of surface roughness after the abrasive processing with different direction are stated in the table (1) and the surface roughness in micro and nano scale after grinding and lapping process are given in table (1) 78 45 with dimensions of 10 mm in length, 8 mm in width and a thickness of 2 mm were used. these parameters were selected according to microscopic scanner capability. the workpice has been machined on a shaping machine then abrasive processing processes grinding or lapping were used. silicon carbide 55c and grain size of 20 µm. were used in grinding and lapping. fig.1. surfcorder se 1200 fig.2. multi-microscopic scanner cmm-2000 3.0 results analysis a “surfcorder se 1200” of kosaka lab (japan) was used to measure the micro surface roughness. to define the surface nano characteristics a scanning microscope was used “cmm 2000” manufactured by proton –miet (russia). the workpiece surface roughness (rz) after shaping process of the profilemeter was 2,998 µm, in the direction of machining, and 3,311 µm perpendicular. this value shows the significant effect of shaping cutting tool. it is not possible to test it on the cmm-2000 scanner, as the resulted values of (rz) are higher that allowed measurement rang (2 µm). the results of surface roughness after the abrasive processing with different direction are stated in the table (1) and the surface roughness in micro and nano scale after grinding and lapping process are given in table (1) fig.1. surfcorder se 1200 fig.2. multi-microscopic scanner cmm-2000 3.0 results analysis a “surfcorder se 1200” of kosaka lab (japan) was used to measure the micro surface roughness. to define the surface nano characteristics a scanning microscope was used “cmm 2000” manufactured by proton –miet (russia). the workpiece surface roughness (rz) after shaping process of the profile-meter was 2,998 µm, in the direction of machining, and 3,311 µm perpendicular. this value shows the significant effect of shaping cutting tool. it is not possible to test it on the cmm-2000 scanner, as the resulted values of (rz) are higher that allowed measurement rang (2 µm). the results of surface roughness after the abrasive processing with different direction are stated in the table (1) and the surface roughness in micro and nano scale after grinding and lapping process are given in table (1) issn: 2180-1053 vol. 4 no. 1 january-june 2012 journal of mechanical engineering and technology 94 table 1: measuring surface roughness after abrasive machining in the different direction. 79 table 1: measuring surface roughness after abrasive machining in the different direction. process measurement direction profile-meter «surfcorder se 1200» microscope смм-2000 ra, µm rz, µm ra, nm rz, nm grinding along 0.410 2.120 34.71 150.7 across 0.431 2.577 43.88 lapping along 0.270 1.450 39.31 149.5 across 0.395 2.152 57.38 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 1 2 grinding lapping r a µ m along across 0 10 20 30 40 50 60 70 1 2 grinding lapping r a n m along across a) b) fig.3. ra after grinding and lapping process. a) in micro scale, b) in nano scale fig (3) shown that in the micro surface roughness the lapping process has better surface than grinding, while in nano scale grinding process has better surface. fig 4. shows the results according to “cmm 2000” microscope а) after grinding 79 table 1: measuring surface roughness after abrasive machining in the different direction. process measurement direction profile-meter «surfcorder se 1200» microscope смм-2000 ra, µm rz, µm ra, nm rz, nm grinding along 0.410 2.120 34.71 150.7 across 0.431 2.577 43.88 lapping along 0.270 1.450 39.31 149.5 across 0.395 2.152 57.38 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 1 2 grinding lapping r a µ m along across 0 10 20 30 40 50 60 70 1 2 grinding lapping r a n m along across a) b) fig.3. ra after grinding and lapping process. a) in micro scale, b) in nano scale fig (3) shown that in the micro surface roughness the lapping process has better surface than grinding, while in nano scale grinding process has better surface. fig 4. shows the results according to “cmm 2000” microscope а) after grinding fig.3. ra after grinding and lapping process. a) in micro scale, b) in nano scale fig (3) shown that in the micro surface roughness the lapping process has better surface than grinding, while in nano scale grinding process has better surface. fig 4. shows the results according to “cmm 2000” microscope 79 table 1: measuring surface roughness after abrasive machining in the different direction. process measurement direction profile-meter «surfcorder se 1200» microscope смм-2000 ra, µm rz, µm ra, nm rz, nm grinding along 0.410 2.120 34.71 150.7 across 0.431 2.577 43.88 lapping along 0.270 1.450 39.31 149.5 across 0.395 2.152 57.38 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 1 2 grinding lapping r a µ m along across 0 10 20 30 40 50 60 70 1 2 grinding lapping r a n m along across a) b) fig.3. ra after grinding and lapping process. a) in micro scale, b) in nano scale fig (3) shown that in the micro surface roughness the lapping process has better surface than grinding, while in nano scale grinding process has better surface. fig 4. shows the results according to “cmm 2000” microscope а) after grinding а) after grinding 80 b) after lapping fig. 4. the results according to “cmm 2000” microscope in nano scale left – scanned image ; right – profile measurements the results of the surface roughness in nano scale which measured with cmm 2000 microscope and the scanned profile of the grinding and lapping surface show that the surface after grinding process is smother than lapping process. the comparison between the grinding and lapping surfaces gives a different result in micro and nano, the lapping process is better in micro while the grinding in nano scale, the cutoff distance in measuring surface roughness may be the reasons of this result because in nano scale the cutoff distance very small about 2-4 nm while in micro 0.8-1 mm. 4.0 conclusion the present work has led to the following conclusions: the surface roughness by the two devices has qualitative value, quantities comparisons is not possible to define in nano scale surface roughness because it differs from one place to anther on the machined surface. lapping process gives better surface than grinding in micro scale, while in nano scale the grinding process is better. measuring direction has an effect in the micro and nano surface roughness. the research result shows that the traditional finishing processes it not suitable in nano scale machined part. 5.0 acknowledgements the author is grateful to the production technology, machine tool and instruments department, peoples’ friendship university of russian, for extending the facilities of metal cutting laboratory and nanotechnology laboratory to carry out this investigation. b) after lapping fig. 4. the results according to “cmm 2000” microscope in nano scale left – scanned image ; right – profile measurements issn: 2180-1053 vol. 4 no. 1 january-june 2012 comparison between grinding and lapping of machined part surface roughness in micro and nano scale 95 the results of the surface roughness in nano scale which measured with cmm 2000 microscope and the scanned profile of the grinding and lapping surface show that the surface after grinding process is smother than lapping process. the comparison between the grinding and lapping surfaces gives a different result in micro and nano, the lapping process is better in micro while the grinding in nano scale, the cutoff distance in measuring surface roughness may be the reasons of this result because in nano scale the cutoff distance very small about 2-4 nm while in micro 0.8-1 mm. 4.0 conclusion the present work has led to the following conclusions: the surface roughness by the two devices has qualitative value, quantities comparisons is not possible to define in nano scale surface roughness because it differs from one place to anther on the machined surface. lapping process gives better surface than grinding in micro scale, while in nano scale the grinding process is better. measuring direction has an effect in the micro and nano surface roughness. the research result shows that the traditional finishing processes it not suitable in nano scale machined part. 5.0 acknowledgements the author is grateful to the production technology, machine tool and instruments department, peoples’ friendship university of russian, for extending the facilities of metal cutting laboratory and nanotechnology laboratory to carry out this investigation. 6.0 references [1] mc. keown, p. a., “the role of precision engineering in manufacturing of the future”, annals of cirp, vol. 36/2, 1987, pp 495-501 [2] taniguchi, n., 1994, “the state of the art of nanotechnology for processing of ultraprecision and ultrafine products”, precision engineering, 16(1), pp. 5–24. [3] chae j., park s.s. and freiheit t. investigation of micro cutting operations. int. journal of machine tools and manufacture, 2006, 46 ( 3-4 ), pp. 313332. issn: 2180-1053 vol. 4 no. 1 january-june 2012 journal of mechanical engineering and technology 96 [4] gowri s., ranjith p., vijayaraj r., balan a.s.s. micromachining: technology for the future. int. journal of materials and structure integrity, 2007, 1 (1,2), pp. 161179(19). [5] ehmann f. a synopsis of u.s. micro-manufacturin research and development activities and trends. multi-material micro manufacture conference. borovets, bulgaria, 2007, pp. 7-13. [6] b. bhushan (ed.). springer handbook of nanotechnology. springer, 2004. [7] gao, w., hocken, r.j., patten, j.a., lovingood, j. and lucca d., 2000, “construction and testing of a nano-machining instrument”, precision engineering, 24(4), pp. 320–328. [8] ikawa n, shimada s, tanaka h (1992) nanotechnology3. [9] http://www.cemmnt.co.uk [10] http:// www.agilent.com [11] gao, w., kudo, y., kiyono, s. and patten, j. a, 2004, “an instrument for nano-machining and nanometrology of free-form surface profiles with a diamond turning machine”, journal of chinese society of mechanical engineers, 25(5), pp. 449–456. [12] g. bissacco, h. n. hansen, l. de chiffre: “size effects on surface generation in micro milling of hardened tool steel”, cirp annals manufacturing technology, vol. 55, issue 1, pp. 593-596, 2006 [13] i.s. jawahir, e. brinksmeier, r. m'saoubi, d.k. aspinwall, j.c. outeiro, d. meyer, d. umbrello, a.d. jayal:” surface integrity in material removal processes recent advances”, cirp annals manufacturing technology, volume 60, issue 2, 2011, pages 603-626 issn: 2180-1053 vol. 1 no. 1 july-december 2009 vibration control of a gantry crane system using dynamic feedback swing controller 63 vibration control of a gantry crane system using dynamic feedback swing controller azdiana1, noor asyikin1, sharatul izah1, nur alisa1 1faculty of electronics engineering and computer engineering, universiti teknikal malaysia melaka, locked bag 1752, pejabat pos durian tunggal, 76109 durian tunggal, melaka. abstract the use of gantry crane system for transporting payload is very common in industrial application. however, moving the payload using the crane is not easy task especially when strict specifications on the swing angle and on the transfer time need to be satisfied. to overcome this problem a dynamic feedback swing controller is designed for the gantry position and speed, as well as the load angle and angular velocity using pid controller. simulated responses of the position of the trolley and sway angle of the mass are presented using matlab. the performance of the bangbang torque input function and the feedback swing controller are compared. from the simulation results, satisfactory vibration reduction of a crane system has been achieved using the proposed method. keywords: pid controller, dynamic feedback swing controller, gantry crane system 1.0 introduction a gantry crane system is a crane carrying the trolley or trolley with a movable or fixed hoisting mechanism, that the bridge is rigidly supported on two or more legs running on fixed rails or other runway. the fundamental motions of a gantry crane consist of traversing, load hosting and load lowering. like other crane types, gantry cranes met with some dissatisfactory due to its natural characteristics. nowadays, industrial uses gantry crane to transfer the loads but not too safety because of the load always swing and might be any incident occurs. the operator, by skillful manual drive of the gantry controls, ensures that this unavoidable pendulum motion subsides as quickly as possible, since extended loading and unloading time costly. as mentioned, the fundamental motions of a gantry crane consist of traversing, load hosting and load lowering. these significant characteristic is that all motions are performed simultaneous at relatively high speed. crane traversing motions, particularly when starting or stopping; induce undesirable swinging of the suspended load. this creates another problem that the swing could cause the hosting rope to leave its groove which could lead to over wrapping and damage. one of the characteristics of these cranes is the flexible hoisting ropes used as a part of the structure for the reduction of system mass, which result in favorable features of high payload ratio, high motion speed and low power consumption. however, the flexible hoisting create serious problems, that is the crane acceleration which required for motion will generate undesirable load swing, which is frequently aggravated by load hoisting. therefore, such load swing should be suppressed as rapidly as possible to maximize the operations. issn: 2180-1053 vol. 1 no. 1 july-december 2009 journal of mechanical engineering and technology 64 several methods of open-loop and closed-loop solutions have been proposed in order to control the vibration. for example, open loop time optimal strategies were applied to the crane by many researchers such as discussed in [g.a. manson,1992]. they came out with poor results because open loop strategy is sensitive to the system parameters (e.g. rope length) and could not compensate for wind disturbances [wahyudi and j. jalani, 2005]. the most popular technique for input shaping is to convolve a sequence of impulses and various methods for shaping impulse sequence of impulses have been testified and applied to crane system as in [j.k. cho and y. s. park, 1995]. m n sahinkaya in his paper [m.n. sahinkaya, 2001] also has reported the same inverse dynamic technique in spring-mass-damper system. however all the above method is still an open-loop approach that avoid the system from become less sensitive to disturbances. increasingly however, feedback control which is well known to be less sensitive to disturbances and parameter variations also adopted to control the gantry crane system. work that has been presented by omar [h.m. omar, 2003] had proposed pd (proportional-derivative) controllers for both position and anti-swing controls. moreover, a fuzzy logic controller had been introduced by wahyudi and jamaludin [wahyudi and j. jalani, 2005]. fuzzy logic controllers were designed and implemented for controlling payload position as well as the swing angle of the gantry crane. this paper will focus on a feedback control system based on the dynamic model of the gantry crane system. the main idea is to produce vibration free system using pid as a controller algorithm. the improvement of the output response will be investigated by comparing the dynamic feedback swing controller and bang-bang torque input function. 2.0 modelling of a gantry crane system this section will emphasize on the modelling of a gantry crane that includes certain assumptions to make the design more sustainable. the model of a gantry crane is shown in figure 1. generally, the configuration of the gantry crane model is specified by the horizontal position of the trolley, x, the length of the hosting rope, l, and the swing angle of the rope, θ. the payload, which is suspended from the point of suspension, s, is assumed to be a rigid body symmetric about its axis with mass m and centre point , g of mass m. issn: 2180-1053 vol. 1 no. 1 july-december 2009 vibration control of a gantry crane system using dynamic feedback swing controller 65 figure 1 model of a gantry crane before the derivation of the equations of motion, some assumptions are made for simplicity. firstly, friction force that may exist in the trolley is ignored. the trolley and the payload can be considered as point masses. besides, the tension force that may cause the hoisting rope elongate is also neglected. the trolley and the payload are assumed to move in x-y plane, which means a study of two dimensional. for the dynamic behaviors of the gantry crane model, the centre point g and the position vector of point of suspension, s with respected to the fixed axes coordinate have to be determined. the position of the load is given by: based on the above assumptions, the equations of motion for the gantry crane system depicted in figure 1 are derived as follows. after linearization, then equation of motion of the gantry crane becomes: generally, θ is very small of its value. this makes . this suggests that the hoisting motion of the gantry crane is decoupled from the trolley if such approximation is issn: 2180-1053 vol. 1 no. 1 july-december 2009 journal of mechanical engineering and technology 66 made. 3.0 dynamic feedback swing controller dynamic refers to the motion of the trolley as induced by internal forces and external forces. internal forces include the gravity forces while the external forces include the applied force, normal force, tension force, friction force, and air resistance force. some of these forces are shown in figure 1. feedback is the path that leads from the initial generation of the feedback signal to the subsequent modification of the system. it means for this system, the initial generation is unstable. so, to overcome this problem it needs to be modified so that the system achieves a stable system. gains are determined to complete this feedback for gantry crane system. whereas, swing control refers to payload that carried by the trolley. the trolley is controlled with feedback system to ensure minimal swing of payload. this method able to control the swing of payload until the trolley reaches target position without vibration. by conclusion, dynamic feedback swing control means the trolley moves along the track with internal forces and external forces from the initial generation of the feedback signal to the subsequent modification of the system by controlling the swing of payload to reduce vibration. dynamic feedback swing control is a method used to drive a system of gantry crane uncertain mass from initial position to a target position as quickly, accurately and safely as possible without vibration by reducing the sway angle of payload. this dynamic feedback is designed by using pid controller. the value of certain system parameters such as rope length and payload mass may significantly during the operation. this project will focus on closed loop control system where a dynamic model is designed by using simulink in the matlab software. it will be analyzed by referring to the derivation of the dynamic feedback input function. the first step to design pid controller is to determine the type of controller whether siso type or mimo type. siso type also known as single input single output while mimo is known as multiple input multiple output. based on the plant of gantry crane, two forces are needed to move the trolley until it reach target position (1 meter) without vibration. it means two inputs and two outputs are considered in order to reduce the vibration. so, this cascade pid controller design is called as mimo type controller. since it is a mimo type controller, the output needs to be determined whether it should be placed as inner loop or outer loop. sway angle is defined as inner loop while trolley position as outer loop. the sway angle is set up as inner loop because its output response is easier to change compared to output response of trolley position. figure 2 shows the inner loop and outer loop for this response. issn: 2180-1053 vol. 1 no. 1 july-december 2009 vibration control of a gantry crane system using dynamic feedback swing controller 67 figure 2 inner loop and outer loop both outputs can be controlled by using pid tuning based on process response. when a process response rapidly change in the manipulate variable, tuning can be determined by observing the response to change in controller parameters. in general the parameters are tuned in order proportional (kp), integral (ti), derivative (td). each parameter is increased until a sustained oscillation is observed in process variable response. 4.0 results and analysis in this work, the following gantry crane parameters are used. the parameters correspond to the experimental crane system presented by yong-seok kim, han-suk seo and seung-ki in [y.s. kim et al., 2001]. system. i. trolley mass, m=1 kg ii. payload mass, m=0.8 kg iii. length of the hoisting rope, l=0.305 meter a comparison with input shaping technique has been executed in order to prove the effectiveness of the approach presented in this paper. the dynamic feedback swing controller has been compared with the bang-bang torque input function that is assumed to be 1.8 nm. figures 3 and 4 show the sway motion of the payload when an input force is applied to the trolley. the motion of the payload is a pendulum like motion where the payload will swing from its initial position to a final position. by definition, the initial position has an angle of zero. when the input force is positive, as shown in figure 3, the sway angle will be a negative value, which swings clockwise by definition. vice versa, when the applied force is negative, the sway angle, which swings anticlockwise, will be positive in value, as shown in figure 4. from the perspective of the transformation of energy, the initial position point has a maximum value of kinetic energy whereas the final position point has the maximum potential energy. issn: 2180-1053 vol. 1 no. 1 july-december 2009 journal of mechanical engineering and technology 68 from the following simulation results in figure 6, which the bang-bang torque input function is applied to the trolley, it can be seen that the sway angle in radian have a trend related to the applied force directions. however, when the applied force is taken off (fx = 0n), it is found that the load still oscillate and is having a large oscillation. this result indicates that the payload motion has a very significant response to acceleration or deceleration commands without any feedback control. it is noted that the motion of the payload is affected by the applied force which it will swing according to its path defined earlier. in bang-bang torque input function, the payload continues to oscillate although the force is taken off by the time 2 seconds. the payload oscillates with a maximum value 0.28 radian, which is approximately 16.04 degree from its initial position. this value is become significant when the length of hoisting rope become larger. this is due to the chord length, which is nearly equal to the arc length that proportionally increases with the length of the hoisting rope as illustrated in figure 5. figure 5 arc length and chord length issn: 2180-1053 vol. 1 no. 1 july-december 2009 vibration control of a gantry crane system using dynamic feedback swing controller 69 figure 6 payload sway angle by using bang-bang torque input function thus, from the simulation result in figure 6, the 0.28 radian will create a 8.4 cm chord length, which is considered as a large swing in this case. for industrial use gantry crane, this will cause nearly 1 meter chord length which will create safety problems and damages to the surrounding environment. hence, by using dynamic feedback swing controller adopted in figure 7, the payload stops oscillate with no vibration. it will not create any chord length, resulting the vibration reduction of sway angle in this technique is 100% compared to bang-bang torque input function. table 1 shows the summary results of the sway angle of the payload by using both techniques. figure 7 payload sway angle by using dynamic feedback swing controller table 1 sway angle of the hoisting rope the simulation results of the positioning of the trolley with the bang-bang torque input function and dynamic feedback swing controller are shown in figure 8 and figure 9 respectively. from the simulation results, it can be clearly seen that the vibration occur when using bang-bang torque input function. it reaches about 1.22 meter at 2 seconds. issn: 2180-1053 vol. 1 no. 1 july-december 2009 journal of mechanical engineering and technology 70 the dynamic feedback swing controller provides no vibration during positioning the trolley along its path and reach 1 meter at 7.4 seconds where the force is taken off. this is the dissatisfactory of this technique where its performance is slower than bang-bang torque input function. figure 8 trolley positioning by using bang-bang torque input function figure 9 trolley positioning by using dynamic feedback swing controller the time response of the trolley position has been simplified in table 2. although the settling time and rise time of bang-bang torque input function is faster than dynamic feedback swing controller, in terms of vibrations, the performance of dynamic feedback swing controller is better than bang-bang input force. table 2 time response of trolley positioning issn: 2180-1053 vol. 1 no. 1 july-december 2009 vibration control of a gantry crane system using dynamic feedback swing controller 71 5.0 conclusions by using the developed model, the dynamic behaviors of the controller has been evaluated using matlab and simulink. the performance of the controller are examined in terms of vibration reduction of sway angle and a stable positioning. simulation results have shown that a dynamic feedback swing controller can be applied to control vibration. satisfactory vibration reduction of a crane system has been achieved using the proposed technique. 6.0 acknowledgement this research is financially supported by crim, universiti teknikal malaysia melaka. 7.0 references a. piazzi, and a. visioli. 2000. minimum time system inversion-based motion planning for residual vibration reduction. ieee/asme trans. mechatronics, pp. 12-22. d.w. frakes. 2001. input-shaped control of gantry cranes: simulation and curriculum developmen. proceedings of the eighteenth asme biennial conference on mechanical vibration and noise. g.a. manson.1992.time-optimal control of an overhead crane model. optimal control applications & methods, vol. 3, no. 2, pp.115-120. h.m. omar. 2003. control of gantry and tower cranes. ph.d. thesis. m.s. virginia tec. 12-25. j.k. cho and y.s. park. 1995. vibration reduction in flexible systems using a timevarying impulse sequence. robotica, 13, pp. 305-313. m. kenison. and w. singhose. 1999. input shaper design for double-pendulum planar gantry cranes. control applications, pp. 539-544; 1999. m. n sahinkaya. 2001. input shaping for vibration-free positioning of flexible systems. proceedings of institution mechanical engineers, vol 215, pp. 467-481. wahyudi and j. jalani. 2005. design and implementation of fuzzy logic controller for an intelligent gantry crane system. proceedings of the 2nd international conference on mechatronics, pp. 345-351. wildi and theodore. 2000. electrical machines, drives and power systems. fourth edition. prentice-hall inc. 417-433; 2000. y.s. kim, h.s. seo, and s.k. sul. 2001. a new anti-sway control scheme for trolley crane system. proceedings of industry applications conference, vol. 1, pp. 548-552. issn: 2180-1053 vol. 1 no. 1 july-december 2009 journal of mechanical engineering and technology 72 _________________________________________ *corresponding author e-mail : nilghuge@gmail.com issn: 2180-1053 vol. 8 no. 2 july – december 2016 95 optimization of cutting parameter during turning using different cutting fluids nilesh c. ghuge1*, dr.ashish m. mahalle1 1laxminarayan institute of technology, nagpur, rtmn university, nagpur, india abstract efforts were made to completely eliminate the toxic cutting fluids. in this regards interest in vegetable oil is growing. minimum quantity lubrication using vegetable based cutting fluids can be used to improve the productivity and avoid environmental detriment. the main objectives of this research are to develop low cost mql system and to evaluate performance of vegetable oil based on surface roughness. the performance of different vegetable oils such as blasocut-4000, soyabean oil, sunflower oil, groundnut oil and coconut oil is compared during turning aisi 4130. analysis of the experimental results is performed using response surface methodology. mathematical model for each performance parameter is developed showing relation between significant parameters such as cutting speed, depth of cut and feed rate. anova test is performed to check the competency of the developed model. optimization of the process parameter is carried out using response surface optimizer. desirability is calculated to show the feasibility of optimization. the experimental results show that vegetable oil outperforms the mineral based fluid. among all the vegetable oil soyabean oil gave the best result. keywords: mql; surface roughness; vegetable oil 1.0 introduction due to high heat, poor surface finish, ecological concern and government protocols, immense efforts are made to reduce the petroleum based cutting fluids. dry cutting is one of the replacement but increased wear rate, raised temperature are major concern issues. practice of using mineral oil is potentially perilous. increased use of cutting fluids results into environmental degradation like soil pollution, water contamination, disposal and dumping problems. recycling cost of the waste cutting fluid is high. it requires separate setup for waste disposal and management. most significant issue is health of the operator. the operator may suffer from dermatological, respiratory disease that may lead to cancer. the use of the mineral oil is hazardous. the mineral oils which are used as cutting fluids are not renewable sources. they are depleting. gases exerted from theses oils also results into ozone depletion and contribute to global warming. (sokovic & mijanovic, 2001) thus mineral oil based cutting fluids are accountable for upsetting the eco-system balance. this disruption has long term effect on mankind and next generation. in order to maintain the natural equilibrium, every country is now imposing strict restrictions for the use of mineral oil based cutting fluids. (niosh,1998) the minimization of cutting fluid leads to saving of the cutting fluid, reduces the journal of mechanical engineering and technology 96 issn: 2180-1053 vol. 8 no. 2 july– december 2016 machining cycle time hence reduces the chances of the contact of the operator with cutting fluid. minimum quantity lubrication is emerged as substitute for dry machining and flood cutting. the minimization of cutting fluids lead to saving of lubricant cost. reduced quantity of cutting fluid lessens exposure level of the cutting fluid. conventional flood lubrication system uses cutting fluid at the rate of one-liter to ten-liter per min. mql uses very small amount of cutting fluid (heisel & lutz, 1994). mql consist of mixture of high-pressure air and cutting fluid applied directly into the interface of cutting tool and work piece the vegetable oils are considered as viable alternative to petroleum based cutting fluids due to high flash point, high boiling point and greater molecular weight. (ghuge & mahalle, 2016). for turning operations, the input parameters like depth of cut, speed, feed, tool geometry etc. will decide the performance parameter. proper selection on input parameter is very essential to get the desired output. performing experiment trial and deciding the optimum values from the experience is very skilled, time consuming and costly affair. the selection of optimal cutting parameters such as cutting speed, feed rate and depth of cut for every machining process is a very important issue in order to maintain the quality of machined products to reduce the machining costs and to increase the production rate. taguchi and response surface methodology is one tool for obtaining the optimum values for anticipated goal (satish, salve, netake, more, kendre & kumar, 2014). 2.0 experimentation since number of input and output parameters are more, there are huge number of possibility of combination to perform the experiments. 33 full factorial design by one replica is selected for investigation. testing is conducted on aisi 4130 bar of 60 mm diameter. carbide tipped single point cutting tool is used for the investigation. testing is done for three different cutting conditions namely dry cutting, flood cutting and mql cutting. figure 1. experimental set up optimization of cutting parameter during turning using different cutting fluids issn: 2180-1053 vol. 8 no.2 july – december 2016 97 blasocut-4000, soyabean oil, sunflower oil, ground nut oil and coconut oils are used as cutting fluid during mql cutting. a portable stylus-type profilometer is used for surface roughness measurement. figure 1 shows experimental set up consisting of medium duty lathe and mql system. turning experiment were performed varying cutting speed, feed and depth of cut as shown in table 1.analysis of the experimental results was carried out using response surface methodology. regression equations for surface roughness were formulated on the basis of the experimental data. in rsm, generally regression model are fitted to experimental data and anova test is used to check the adequacy of the regression model. software minitab17 is used for rsm analysis. table 1. experimental condition and machining parameter experimental condition description parameter cutting speed v (m/min)= 34.27,53,79.73 feed rate ,f (mm/rev)=0.35,0.40,0.45 depth of cut ,d(mm)=0.5,1,1.5 coolant condition dry, flood cut-1 l/hr., mql-50ml/hr. cutting insert carbide tipped single point cutting tool, rake angle=12° 3.0 mathematical modeling and optimization regression analysis is a statistical instrument for the study of relationships between variables. regression is mainly used for prediction and contributing inference. regression equations 1 to 7 are formulated for surface roughness under different cutting condition. regression equations for dry cutting radry=12.380.0439 v49.2 f+ 0.586 dp0.000061 v*v+ 68.9 f*f 0.338 dp*dp+ 0.1054 v*f+ 0.00248 v*dp+ 0.93 f*dp … (1) regression equations for flood cutting (flood cutting) raflood=10.300.2347 v8.2 f0.261 dp+ 0.001377 v*v+ 9.3 f*f+ 0.080 dp*dp + 0.1157 v*f0.00090 v*dp+ 0.80 f*dp … (2) regression equations for mql cutting (blaso cut-4000) ra blassocut = 3.38 0.1171 v + 4.79 f + 0.817 dp + 0.000886 v*v + 4.4 f*f 0.089 dp*dp 0.0125 v*f 0.00249 v*dp 0.567 f*dp … (3) journal of mechanical engineering and technology 98 issn: 2180-1053 vol. 8 no. 2 july– december 2016 regression equations for mql cutting (soyabean oil) ra soyabean = -1.76 0.0973 v + 25.1 f + 0.447 dp + 0.000863 v*v 18.4 f*f + 0.036 dp*dp 0.0404 v*f 0.00196 v*dp 0.367 f*dp … (4) regression equations for mql cutting (sunflower oil) ra sunflower = 1.36 0.09825 v + 10.75 f + 0.094 dp + 0.000852 v*v 0.89 f*f + 0.0978 dp*dp0.0396 v*f0.00066 v*dp+ 0.033 f*dp … (5) regression equations for mql cutting (ground nut oil) ragroundnut = 3.04 0.0778 v 0.8 f + 0.328 dp + 0.000641 v*v + 13.6 f*f + 0.016 dp*dp 0.0297 v*f0.00137 v*dp0.000 f*dp … (6) regression equations for mql cutting (coconut oil) racoconut = 2.76 0.10913 v + 8.48 f + 0.381 dp + 0.000750 v*v 2.00 f*f 0.0600 dp*dp 0.0009 v*f + 0.00220 v*dp 0.400 f*dp … (7) 3.1 significance of f value, p value, r2, r2 adjusted and r2 predicted the f value tells us how much we are deviated from hypothesis. the statistic f is compared to the critical fcrit = f (α, fterm, ferror), where α is significance level (generally it is 0.05). if observed f-value is greater than the critical f, model is acceptable and accurate. the value of fcrit is taken from statistical table. for, f (0.05, 2, 24), fcrit value is 3.40. larger value of f for parameter indicated that it has significant effect on the response. percentage contribution of individual parameter can be calculated by dividing f value of each term by total sum of the f value of the all parameter. p value is the probability value. smaller value of the p value indicates that the parameters has influence on the response. r2 measures percentage of the variation of response, as per regression equation. for better assessment of regression equation to fit the trial data, r2should be closer to one. the higher value of r2, means the model fits data the better. adjusted r2 accounts for the number of predictors in model and is useful for comparing models with different numbers of predictors. predicted r2, indicates how well the model predicts responses for new observations. larger values of predicted r2 suggest models of greater predictive ability. (satish chinchanikar et al., 2014) 3.2 competency of developed model using minitab-17, the competency of the developed regression equations were verified by analysis of variance (anova).r-squared is a correlation coefficient squaredmeasure the variations in the experimental results. the r squared values close to 100% indicate that the model is significant .i.e. experimental model provides an excellent relationship between factor and response. in this study the depth of cut, feed and cutting speed are the factors. surface roughness is the responses. anova results for each cutting conditions are shown in table.2.it can be seen that r-squared values for all optimization of cutting parameter during turning using different cutting fluids issn: 2180-1053 vol. 8 no.2 july – december 2016 99 the developed equations are approaching to one. adjusted and predicted values are in reasonable agreement. the model f-values obtained for the all the equations indicates that the model is significant. table 3 shows the percentage contribution of the individual parameter on surface roughness. table 2. anova result for surface roughness (ra) factor/cutting conditions dry cutting flood cutting blaso cut soyabean oil sunflower oil coconut oil groundnut oil r2 (%) 97.8 99 98.8 98.1 99.1 99.1 98.3 adj. r2 (%) 96.7 98.4 98.3 97.1 98.7 98.6 97.5 pre.r2 (%) 94.1 97.1 96.6 95.3 98.1 97.6 96.1 model fvalue 86.1 187.7 167.5 99.9 230.8 211.4 114.6 table 3. anova result for contribution (%) of different parameter on surface roughness parameter /condition dry cutting flood cutting blaso cut soyabean oil sunflower oil coconut oil groundnut oil v 4.1 65.4 57.9 39.7 21.9 60.1 42.4 f 82.2 13.3 25.1 38.9 32.0 25.2 43.7 d 9.0 0.8 3.7 4.4 30.3 3.1 5.0 v*v 0.1 19.0 13.1 16.3 3.5 11.4 8.6 f*f 2.0 0.0 0.0 0.2 12.0 0.0 0.1 d*d 0.5 0.0 0.0 0.0 0.0 0.0 0.0 v*f 1.9 1.4 0.0 0.4 0.3 0.0 0.2 v*d 0.1 0.0 0.1 0.1 0.0 0.1 0.0 f*d 0.1 0.0 0.0 0.0 0.0 0.0 0.0 total 100 100 100 100 100 100 100 table 3 shows for different cutting fluid, the influencing factor is different. for dry cutting, feed (almost 82% contribution) is the most influencing factor. as there is no cutting fluid used during dry cutting, feed and depth of the cut affects the surface roughness remarkably. increasing feed increases temperature and thus worsen the product quality. surface roughness during flood cutting is mainly affected by velocity. as velocity increases frictional resistance decreases, this results into decreased roughness values while feed is most contributing factor for blasocut. when soyabean is used as coolant surface roughness is affected by velocity and feed. the contribution of velocity and feed is approximately equal. when sunflower is used as a cutting fluid, surface roughness is influenced by feed rate and depth of cut. sunflower oil and soyabean oil are vegetable oil with tri glyceride00 chain. this results into less frictional force as well less cutting forces. heat generated at tool-work piece interface is less. increasing feed rate and depth of cut affects the surface roughness values. journal of mechanical engineering and technology 100 issn: 2180-1053 vol. 8 no. 2 july– december 2016 4.0 results and discussions surface finish is a key factor of machinability because it affects the performance and service life of the machined component. variation of surface roughness with respect to different cutting condition is shown in figure 2.points 1 to 27 in the figure represents the observation number of experiment. observation 1 to 9 indicates observation at speed 34.27 m/min. observation while observation 10 to 18 shows cutting forces at 53 m/min. 19 to 27 observation numbers are at speed 79.27 m/min. the blue line and orange line corresponds to dry cutting and flood cutting as shown in figure 2. both the lines are along outward side of the circle. surface roughness decreases as speed decreases. the lubricating action of the dipolar molecure of soyabean oil reduces the frictional force. this decreases the temperature and results in the lesser tool wear, thus resulting in surface quality improvement.machining with soyabean oil shows 4% ,8% and 15% less roughness values as comapred to sunflower oil, ground nut oil and cocnout oil respectively figure 2. surface roughness variation of at different cutting condition. surface roughness is more for dry cutting; flood cutting is more as compared to mql cutting. there is average 8%, and 38% decrease in surface roughness in mql cutting as compared to flood cutting and dry cutting respectively. it is noticed from figure 2 that there is drastic decrease in roughness value when speed is changed from 53 m/min to 79.72 m/min. when vegetable oils are used as cutting fluid with mql,the sufrace roughness values are greatly reduced.mql machining with soyabean oil shows average 16 % decrease in surface roughness value.this is due to less cutitng forces in case of soyabean oil 0 2 4 6 1 2 3 4 5 6 7 8 9 10 11 12 13 1415 16 17 18 19 20 21 22 23 24 25 26 27 ra dry ra flood ra blassocut ra soyabean ra groundnut ra sunflower ra coconut optimization of cutting parameter during turning using different cutting fluids issn: 2180-1053 vol. 8 no.2 july – december 2016 101 5.0 optimization and confirmation experiment response optimizer is used to find the optimum value. the goal decided is to minimize the surface roughness value. the optimum cutting speed is in between 50-60 m/min. the depth of cut and feed are 0.5 mm and 0.35 mm/rev respectively. desirability analysis carried out. optimized model showed more than 90% level for all cutting conditions. this shows that optimized values are feasible. the aim of the confirmation test is to validate inferences drawn during the analysis phase. once the optimum level of the process parameters is selected, the final step is to predict and verify the improvement of the performance characteristics using the optimum level of the process parameters. in order to validate the results obtained from response optimizer, confirmation experiment were conducted at optimum values of the process parameter determined. experimental results and optimum values are in good agreement with error within permissible limit as shown in table 4 table 4. optimum cutting condition and desirability condition v(m/min) f(mm) dp(mm) ra pred. ra expt desirability dry cutting 52.6377 0.35 0.5 3.49442 3.54 0.9078 flood cutting 61.8215 0.35 0.5 1.82166 1.62 0.9235 blaso cut 58.6072 0.35 0.5 1.74043 1.75 0.9348 soyabean oil 50.8009 0.35 0.5 1.45677 1.57 0.9540 sunflower oil 52.1785 0.35 0.5 1.54195 1.64 0.943 coconut oil 60.9031 0.35 0.5 1.76844 1.87 0.9315 groundnut oil 57.2296 0.35 0.5 1.59412 1.72 0.9382 6.0 conclusions based on the results of the experiments and analyses carried out the following general conclusions are drawn:  there is extensive difference in surface roughness produced by dry, flood and mql cutting conditions. there is an average 8% reduction in roughness values for mql as compared to flood cutting  mql technique is substitute for dry and flood cutting not only in terms of performance but also in terms of it is cost effectiveness and environment friendliness.  result shows that sunflower oil, groundnut oil ,blasocut and coconut oil have surface roughness values,4%,8%,15% and 16% more that of soyabean oil respectively  mathematical models are developed to validate the experimental data. anova test carried out to check the adequacy of the model. mathematical models developed for all parameters are accurate and acceptable. journal of mechanical engineering and technology 102 issn: 2180-1053 vol. 8 no. 2 july– december 2016  multi-response optimization is carried out to get optimal values. the optimum value of depth of cut is 0.5mm, feed rate is 0.35mm /rev and speed value is between 50-60 m/min, which gives minimum surface roughness values. desirability analysis indicates that the optimization process is feasible.  speed is most substantial factor while predicting surface roughness values. references heisel, u. & lutz, m. (1994). application of minimum quantity cooling lubrication technology in cutting processes. prod. eng,. ii (1), 49-54. nilesh, g. & mahalle, a.m. (2016). comparative study of different vegetable oil during turning in terms of cutting force and power consumption. journal of biological science, 2 (5), 91-94. nilesh, g. & mahalle, a.m. (2016). performance evaluation of vegetable oil using mql during turning of aisi 4130 in terms of temperature and surface roughness. international journal of engineering science and computing (ijesc), 6 (5), 58795781, nilesh, g. & mahalle, a.m (2016). experimental investigation on the performance of soyabean oil and blasocut-4000 during turning of aisi 4130 in terms of cutting forces. international journal of scientific research in science, engineering and technology (ijsrset), 2(3), 330-333, satish, c. & salve, a.v., netake, p., more, a., kendre, s. & kumar, r. (2014). evaluation of surface roughness during hard turning under dry and with water based and vegetable based fluids. procedia material science, 5, 1966-1975, sokovic, m. & mijanovic, k. (2001). ecological aspects of the cutting fluids and its influence on the quantifiable parameters of the cutting processes. journal of materials processing technology, 109, 181-189. u.s. department of health and human services, january 1998, occupational exposure to metalworking fluids”. niosh publication, 98-102. issn: 2180-1053 vol. 1 no. 1 july-december 2009 hybrid palm oil mills maintenance system 53 hybrid palm oil mills maintenance system abd. samad hasan basari1, nanna suryana herman2, mohammad ishak desa3 1,2faculty of information and communication technology, universiti teknikal malaysia melaka, locked bag 1752, pejabat pos durian tunggal, 76109 durian tunggal, melaka 3faculty of computer science and information system, universiti teknologi malaysia, 83100 skudai, johor 1abdsamad@utem.edu.my, 2nsuryana@utem.edu.my abstract this paper proposes a technique that enhances snapshot model for cause of failure and decision analysis in order to easily assist maintenance engineers during identification and definition of the actual maintenance problem. the technique is a hybrid of failure mode, effect and criticality analysis, information technology and decision analysis into the snapshot model. a tool that automates the hybrid of snapshot modelling for cause of failure and decision analysis is also developed. this tool aims to ensure maintenance engineers can conduct snapshot modelling with little or without the help of operation research experts to facilitate in the cause of failure and decision analysis process. keywords: hybrid system, maintenance, snapshot modelling 1.0 introduction the cause of failure analysis is the process of identifying, defining, and diagnosing the maintenance problem. the main purpose of cause of failure analysis is to avoid tackling the wrong problem. generally this analysis involves (liu, 1997): a) identifying the existence and location of the problem: which are recognising the symptoms, seriousness of the problem from the aspect of cost, downtime as well as the size, and the areas of the fault in the plant’s machines where the problems are most developed. b) determining the problem’s causes: the analysis of the problem’s causes can be at structural or functional level. consequently depending on the level of causal analysis, different solution strategies may be generated. c) generating and determining possible solution strategies. having identified problem and its nature, location, causes and consequences, then possible solution strategies could be developed or generated. however, the data specified above are difficult to be found in any organisation and also very tedious to be collected on a dynamic basis if maintenance management information system is supposed to be used. for this reason the usages of a survey form for collecting such type of data on periodic basis is suggested. at each failure or maintenance intervention, the engineer registers the data related to the snapshot model issn: 2180-1053 vol. 1 no. 1 july-december 2009 journal of mechanical engineering and technology 54 in survey form. once finished, or analyst collects back the survey form and starts the analysis process. the results of the analysis, will be reported back to the maintenance engineers which reveal the true status of the plant under the study. despite the usefulness of the snapshot model as one of the important tools for cause of failure and decision analysis, the implementation of the model in large scale is doubtful. this is mostly due to, the scarcity and the reliability of the data related to snapshot model, the problem of analysing the data, and the problem of interpreting the results of the analysis to the users (maintenance engineers). 2.0 the proposed hybrid system the proposed approach of automating and augmenting snapshot model aimed to complement such type of modelling. enriched the various techniques that have proven appropriate and possible in combining with snapshot model could give a more effective, ease of use and practically applied to the real world maintenance problems. 2.1 elements of enhancement the enhancement require the utilisation of the emerging information technology (it) and failure mode, effect and criticality analysis (fmeca). in theory, it and fmeca can be utilised to produce an enhanced snapshot model. once the data collection done, the analysis process needs further techniques to be enriched. the technique called decision analysis is introduced. the decision analysis will use analytic hierarchy process (ahp) method and decision making grid (dmg) by utilising fuzzy logic rule base (flrb) method. figure 1 shows the conceptual merger of above mentioned techniques into the current snapshot model. figure 1 the conceptual hybrid of fmeca, it and decision analysis into snapshot model 2.2 information technology the computer technology can increase the involvement of maintenance engineers in the development of the snapshot model by allowing the replacement of the survey form with a more general computer form that contains feature of checking the validity and consistency of the data and can be applied for different machines. it can also permit maintenance engineers to carry out the snapshot analysis with little assistance issn: 2180-1053 vol. 1 no. 1 july-december 2009 hybrid palm oil mills maintenance system 55 of or without or analysts. 2.3 failure mode, effect and criticality analysis the failure mode and effect analysis that could combined with the snapshot analysis include: • major fault areas and their modes. this analysis will analyse all number of failures for each component according to their mode. • failure mode and their cause analysis. this analysis provides guidelines and directions to which is need to be done for specific failure mode. • failure modes and their cost analysis. this analysis identifies the consequences of each failure mode in term of the cost. • failure modes and their downtime. this analysis will lead to identify the failure mode, which frequently disrupt the operation of the machines. • failure modes and means of prevention analysis. this analysis identifies the viable means of preventing each type of failure mode. criticality analysis (ca) is a procedure by which each potential failure mode is ranked according to the combined influence of severity and probability of occurrence. the procedure for obtaining the criticality analysis is as follows (kececioglu, 1991): • the number of failure for each mode will be calculated from the collected data. • the total number for all the failure of the machine will be calculated. • the failure mode frequency ratio (fmfr) will be calculated by dividing the number of failure for each mode by the total number of failure for the machines. • obtain the estimated probability of stopping, ps, of the machine if the failure in a given mode should occur. • obtain the component unreliability q by subtracting the component’s predicted reliability from 1 or 100 (if calculated in %). • calculated the criticality cr = (fmfr) x (ps) x (q). by using the above steps, criticality ranking will be conducted for the components of any machine under the study. 2.4 decision analysis decision analysis is particularly a techniques which is part of the framework to achieve world class maintenance (labib, 1998). among the established method to implement decision analysis are ahp and dmg based on flrb method. the ahp is a decision support tool, which can lead the decision makers to model a complex problem in a hierarchical structure showing the relationship of the goal, objective (criteria), subobjectives and alternatives (saaty, 1977). figure 2 show the workflow of ahp process. issn: 2180-1053 vol. 1 no. 1 july-december 2009 journal of mechanical engineering and technology 56 figure 2 analytic hierarchy process (ahp) workflow there are four major steps to calculate ahp which are: 1) setting up the hierarchy: the first step in ahp is to develop hierarchy by breaking the problem down into components. this level is also known as design phase. the three major level of hierarchy are the goal, objectives and alternatives. 2) comparison of characteristics and establish priority vector: characteristics refer to the objectives or criteria that located in the second level of the hierarchy. in this phase, it is known as evaluation phase. decision maker needs to perform comparison between each objective in a one-to-one (n x n) matrix form. pair wise comparison is used to determine the relative importance of each alternative in term of each criterion. the pair wise comparison expresses the qualitative answer of a decision maker into some numbers, which is easy to manipulate in the calculation and solve the problem of inconsistency unit of measurement for each criterion. table 1 showed the proposed scale where the scale member set is {9, 8, 7, 6, 5, 4, 3, 2, 1, 1/2, 1/3, 1/4, 1/5, 1/6, 1/7, 1/8, 1/9}. by referring to the above standard scale, a matrix of characteristic (objectives) can be constructed. for consistency, it is necessary to set aji =1/aij (this state the obvious fact that if objective 1 is slightly more important than objective 3, than the objective 3 is slightly less important than objective 1). issn: 2180-1053 vol. 1 no. 1 july-december 2009 hybrid palm oil mills maintenance system 57 table 1 scale of relative importance hence the concept of putting values in a matrix conform the following rules: a) the equal attribute in the matrix is put as 1 (diagonal). b) the decision maker only needs to fill the upper right triangle of the matrix. c) for the lower left triangle of the matrix, the value should be the inverse of the corresponding cell in upper right. 3) comparison of alternatives and establish priority vector for alternatives: the previous steps are determined the weight of each objectives, so the next step is to determine how well each alternative score on each objective. the process of calculation is almost similar with the previous step where a pair wise comparison matrix for each objective is constructed by referring to the scale. 4) obtaining the overall ranking: the final step is to obtain a vector of overall scores for each alternative, which can accomplish by multiplying the weight calculated by each alternative associated to each of the criteria. the first ranked alternative will have the highest weight (highest priority). one foundation of the ahp is the observation that the human decision-making is not always consistent. consistency suffers when the criteria being compared are subjectively in nature. the ahp provides a standard by which the degree of consistency can be measured. if inconsistency exceeds an established threshold, then participants can re-examine their judgements. in the ahp, the pair wise comparisons in a judgement matrix are considered to be adequately consistent if the corresponding consistency ratio (cr) is less than 10%. first, the columns in the judgement matrix a, multiply with the resulting vector priority, w, and the averaging the ratio of each element to yield an approximation of the maximum eigenvalue, denoted by (an eigenvalue of a square matrix a is a scalar c such that aw = cw holds for some nonzero vector w). then the consistency index (ci) value is calculated by using formula ci = ( – n)/ (n-1). next, the consistency ratio (cr) is computed by dividing the ci value by the random index (ri). the cr is the average ci of sets of judgements (from a 1 to 9 scale) for randomly generated reciprocal matrices. the consistency index is shown in table 2. issn: 2180-1053 vol. 1 no. 1 july-december 2009 journal of mechanical engineering and technology 58 for a perfectly consistent decision maker, each ratio in step 2 equal to n. this implies that a perfectly consistent decision maker has ci = 0. the values of ri in table 2 give the average value of ci if the entries in, for example a were chosen at random (subject to the constraints that aij’s must equal 1, and aij = 1/aji). if the ratio of ci to ri is sufficiently small, then the decision maker’s comparison is probably consistent enough to be useful. if ci/ri<0.10, then the degree of consistency is satisfactory, whereas if ci/ri > 0.10, serious inconsistencies exist and ahp may not yield any meaningful results (saaty, 1990). the features to enhance the snapshot model are: 1) first level-criteria evaluation: this steps need the decision maker prioritises his/her preferences on different criteria such as fault mode, effect, major fault, fault cause and consequences. table 2 random index/random consistency index for different value of n 2) second level-sub criteria evaluation: this steps need the decision maker prioritises his/her preferences on different sub criteria such as number of fault, machine downtime, cost and criticality. 3) third level-alternatives selection: the machines are ranked according to their weights. weights are obtained through running an ahp algorithm in an absolute mode and hence a consistency ratio of value zero is assured. the above mentioned three level of ahp method is a complimentary of three type of analysis provided by snapshot model which are major fault analysis, cause of fault analysis and consequences of fault analysis. once the fmeca features called fault mode, effect and criticality analysis embedded to snapshot model, they also will be added features to decision analysis process. the three steps of the fuzzy controller are fuzzification, rule evaluation (inference) and defuzzification (sharma et.al., 2007). each of these steps is described below: issn: 2180-1053 vol. 1 no. 1 july-december 2009 hybrid palm oil mills maintenance system 59 1) first step-fuzzification: the first step in the fuzzy controller is the fuzzification process. the membership function, universe of discourse u, is the classifications that are considered in the problem. it is assumed that both frequency and downtime can be classified into `high’, `medium’ and `low’. 2) second step-rule evaluation: the rule evaluation step can also be explained as an input-output system. in this step, inputs are expert rules, and fuzzy inputs obtained from the first step (that is values of m), while outputs are fuzzy values of maintenance actions to be carried out. given two variables of frequency and downtime with each having three subsets of low, medium, and high, then one needs at least nine (3x3) rules to describe the model (system). these rules are in the form of if . . . then . . . statements. examples of maintenance prescriptions are as follows: a) operate to failure (otf) b) fixed time maintenance (ftm) c) skill levels upgrade (slu) d) condition base monitoring (cbm) e) design out maintenance (dom) a summary of the application of each action, based on the values of frequency (fr) and downtime (dt), is given in table 3. an example of a rule can be `if downtime is low and frequency is high, then improve operators skill. this rule can be written as follows: if frequency is high and downtime is low then s. l. u (rule 7) rule 7 is shown in the third row, and first column in table 3. the summary of rules is presented in table 3. table 3 summary of rules for maintenance actions once rules are constructed, and given the values of the fuzzy inputs for (mfl, mfm, mfh, mdl, mdm, mdh) one can apply the minimum and maximum (and & or zadeh) inference computations. 3) third step-defuzzification: this is the final step in the fuzzy controller. this process is based on the idea of deriving a crisp value from a fuzzy function. the defuzzification can be performed by deriving the centre of gravity of the area under the curve of the function. given the cost function of each maintenance action, one can arrange the maintenance actions, the fuzzy output, and the cost scale function. the feedback mechanism offered by the rules grid or dmg of fuzzy logic, as shown in table 3, in addition to the feedback already offered in ahp in the form of consistency ratio, provides an effective performance. the above-mentioned flrb method will be used as an enhancement of snapshot model features called prevention action analysis. issn: 2180-1053 vol. 1 no. 1 july-december 2009 journal of mechanical engineering and technology 60 3.0 the case study and result this case study demonstrates the application of the above-mentioned techniques and its effect on maintenance performance. this company is used as a pilot study in order to test whether the system meets the user expectation and preference. a number of experts were interviewed and proposed during the design and develop the targeted system. 3.1 company background the company is a palm oil mills (pom), which the main job is extracting the fresh fruit bunches (ffb) to crude palm oil (cpo). in this particular company there are about 50 major machines or plants. since the aim of this tool is to assist maintenance engineers establish an appropriate maintenance action, the case study related to an old pom, which are operated more than ten years and use a conventional method of cause of failure and decision analysis techniques. 3.2 the result most of the maintenance information found at pom at the time of the study commenced is originated from the unstructured daily and lubricant report. the unstructured daily report only has the date of the report, time of the report and the description of works. the lubricant report just gives the machine that need a top up or change the lubricant oil, the quantity of oil needed and a description of work or problem occurs that might cause the need to top up or change the lubricant oil. the snapshot model will be built based on the data collected from pom concerning the most problematic machine namely, screw press. example of the result using the hybrid cause of failure and decision analysis techniques are shown in table 4, 5 and 6. table 4 the combined major fault area with the number of fault, criticality, cost and downtime analysis for the period from 1.8.05 to 30.9.05 issn: 2180-1053 vol. 1 no. 1 july-december 2009 hybrid palm oil mills maintenance system 61 table 5 final result based on the ahp method in the decision analysis technique to select the most critical components for the period from 1.8.05 to 30.9.05 table 6 final result based on the flrb method in the decision analysis technique to select prevention action for the period from 1.8.05 to 30.9.05 4.0 conclusion it is recognised the importance of the snapshot model as a tool for cause of failure and decision analysis. the recent development in the computer technology in terms of speed and capacity coupled with the successful research in the human computer interaction play considerable role in the development of a successful tool that capable of constructing snapshot model. from the result, it shows that the use of hybrid maintenance cause of failure and decision analysis could significantly improve the decision context by adding the features of snapshot model. in term of efficiency of decision-making process, the result shows that the reduction of time to reach decision among the decision makers. the hybrid cause of failure and decision analysis techniques could not deny the use of human judgments during the survey that have been conducted. further enhancement could be done by embedding the techniques issn: 2180-1053 vol. 1 no. 1 july-december 2009 journal of mechanical engineering and technology 62 with the computerised maintenance management system (cmms). the quality of data also could be the major issues and it could be done by using the automated data capturing techniques such as using condition monitoring method. 5.0 acknowledgement the project is part of phd research funded from utem scholarship. 6.0 references a.w. labib. 1998. world class maintenance using computerised maintenance management system, journal of quality in maintenance engineering, vol. 4, pp. 66-75. d. kececioglu. 1991. reliability engineering handbook, englewood cliffs, nj: prentice hall. l. liu. 1997. development of an integrated system for machine fault diagosis, phd in industrial engineering, university of houstan. r.k. sharma, d. kumar and p. kumar. 2007., fm a pragmatic tool to model, analyse and predict complex behaviour of industrial systems, engineering computations: international journal for computer-aided engineering and software, vol. 24, no. 4, pp. 319-346. t.l. saaty. 1977. a scaling method for priorities in hierarchical structures, journal of mathematical psychology, vol. 15, pp. 234-281. t.l. saaty. 1990. how to make a decision: the analytic hierarchy process, european journal of operational research, vol. 48, pp. 9-26. issn: 2180-1053 vol. 3 no. 1 january-june 2011 characterization of injection molded 17-4ph stainless steel prepared with waste rubber binder 11 characterization of injection molded 17-4ph stainless steel prepared with waste rubber binder a.r. jeefferie1, s. nurhashima1, m.y., yuhazri1, haeryip sihombing1, s., mohd shukor1, n.s., abdullah2, m.a., omar2 1faculty of manufacturing engineering, universiti teknikal malaysia melaka, hang tuah jaya, 76100, durian tunggal, melaka, malaysia 2structural materials division, advanced materials research center -sirim amrec, 09000, kulim, kedah, malaysia email: 1jeefferie@utem.edu.my abstract this study is to investigate the sintering characteristics and to establish the best heating rate and soaking time used for sintering process, by determining the physical, mechanical, and microstructural properties of the injection molded 17-4ph stainless steel using waste rubber as a new developed binder system. by using the feedstock which having 65 vol.% of metal powder, the molding are injected into the tensile test bar and immediately processed with two stage debinding process that involves of solvent extraction and thermal pyrolisis to remove the binder. the specimens were sintered at 1360°c under vacuum atmosphere and tested for a critical property analysis of tensile test. later, the observation on tensile tested specimens fracture surface are done to understand the fracture behavior, distribution of grain and porosity and the significant correlation of fracture morphology to the mechanical properties. from this study, it is found that the combination of 50c/min heating rate and 60 minutes of soaking period resulted in higher density value, higher tensile strength, less porosity and homogenous grains distribution of the sintered specimens. keywords: injection molded; sintering; binder system; stainless steel powder; waste rubber. 1.0 introduction there are many improvisation and development in the field of powder metallurgy. newly invented and the very recent method of powder metallurgy is by combining the concept of plastic injection molding issn: 2180-1053 vol. 3 no. 1 january-june 2011 journal of mechanical engineering and technology 12 with the utilization of metal powder feedstock (rodriguez-senin et.al., 2005). however, this process is more complicated than regular plastic injection molding process. this is due to the process based on the use of fine powder particles mixed with the small quantity of wax binders and/or thermoplastic polymer to form the molded feedstock. as the novel binder system in metal injection molding (mim) processing, the use of waste rubber to replace natural rubber is receiving great attention due to the advantage of renewability, thermal stability and high shear viscosity (tan et.al., 2008). in addition, a new binder system used in metal injection molding (mim) processing also exhibit economical and environmental friendly characteristics especially for the application in automotive, tooling, medical and hardware component. in this study, the waste rubber was utilized, formulated and evaluated as a new binder system for the mim processing due to its advantageous. 2.0 experiment 2.1 materials basically, every process involved in metal injection molding (mim) has significant impacts to the characteristics of starting materials since the powder and the binder are critically important to the overall success of this process. in this research, the main raw material used is 17-4ph stainless steel and the binder systems consist of paraffin wax (pw), thermoplastic waste rubber (tpwr) and stearic acid (sa) combination, while the feedstock consists of 65 vol. % of powder loading and 35 vol. % of the binder system refers to the proportion selected based on previous study (omar and subuki, 2006). 2.2 specimen preparation the specimen samples were prepared through mim process which involved the mixing, injection molding, debinding and sintering process. prior to the mim operation, the mixing process must be undertaken to produce a homogenous feedstock. by using z-blade mixer type, the powder and binder were mixed together all in one batch. the mixing process was operated at temperature condition of 1750c for about two hours to produce a granulated feedstock according to the parameters used on the previous study (omar and subuki, 2006; gulsoy et.al., 2007). this granulated feedstock was injected into a tool with tensile bar-shaped cavity using a vertically aligned and pneumatically operated plunger machine (mcp hek-gmbh vertical issn: 2180-1053 vol. 3 no. 1 january-june 2011 characterization of injection molded 17-4ph stainless steel prepared with waste rubber binder 13 injection molding) operating at temperature of 1700c, 6 bar of injection pressure, and 0.25 seconds of cycle time in order to produce good surface finish of green body. the two stage operations, namely solvent extraction and thermal pyrolisis (figure 1a), must be performed in order to remove the binder system through the process of debinding within the feedstock. solvent extraction was first performed using water bath brand memmert and then the samples were immersed in the heptane in order to remove all soluble binder. the dimension of green body specimens were measured prior to the solvent extraction process. thereafter, the green body specimens were arranged in a glass container containing heptanes that must be heated at 600c and half immersed in the tap water (omar and subuki, 2006). the specimens were then immersed for about five hours. to prevent the evaporation of the heptane solvent during the extraction operation, the glass container should be covered. when the extraction completed at about five hours, the specimens were removed and left for drying in the oven at the temperature of 400c for four hours to remove the remaining solvent. next, the thermal pyrolisis or debinding is carried out in linn furnace to remove the remaining binder that cannot be completely removed during the solvent extraction operation. thermal debinding operation used two different heating rates which are 50c/min up to temperature 2500c for the first stage and 50c/min up to temperature 4500c for the second stage. about one hour of soaking period is allocated for each stage, before the cooling operation. 2.3 sintering sintering is a processing technique to produce density-controlled materials and components from the metal powder by applying thermal energy (kang, 2005). therefore, the study on the effect of heating rate and soaking period of sintering was based on the operation in hot wall multi-atmosphere sintering furnace in the vacuum condition with sintering temperature at about 13600c. the sintering profile can be found in the figure 1(b). issn: 2180-1053 vol. 3 no. 1 january-june 2011 journal of mechanical engineering and technology 14 11 heating rate and soaking period of sintering was based on the operation in hot wall multiatmosphere sintering furnace in the vacuum condition with sintering temperature at about 13600c. the sintering profile can be found in the figure 1(b). figure 1 (a) the heating profile of thermal pyrolisis process (b) the heating profile of sintering operation 2.4 density test astm 328 is a density determination standard to measure the density of sintered specimens by determining the specific gravity using the archimedes principle. the density measurement was carried by taking the weight of sintered specimens in the air and in the water, without impregnation done to the samples, prior to the measurement. later, the calculation of the density was done by referring to the following formula. time (min) 5 0c / min (44 min) 250 0c 5 0c / min 450 0c 1hr temperature (0c) time (min) 300c cooling 50c / min (1h 24 min) 4500c 1hr heating rate 13600c soaking time temperature (0c) 300c cooling (a) (b) 11 heating rate and soaking period of sintering was based on the operation in hot wall multiatmosphere sintering furnace in the vacuum condition with sintering temperature at about 13600c. the sintering profile can be found in the figure 1(b). figure 1 (a) the heating profile of thermal pyrolisis process (b) the heating profile of sintering operation 2.4 density test astm 328 is a density determination standard to measure the density of sintered specimens by determining the specific gravity using the archimedes principle. the density measurement was carried by taking the weight of sintered specimens in the air and in the water, without impregnation done to the samples, prior to the measurement. later, the calculation of the density was done by referring to the following formula. time (min) 5 0c / min (44 min) 250 0c 5 0c / min 450 0c 1hr temperature (0c) time (min) 300c cooling 50c / min (1h 24 min) 4500c 1hr heating rate 13600c soaking time temperature (0c) 300c cooling (a) (b) figure 1: (a) the heating profile of thermal pyrolisis process (b) the heating profile of sintering operation 2.4 density test astm 328 is a density determination standard to measure the density of sintered specimens by determining the specific gravity using the archimedes principle. the density measurement was carried by taking the weight of sintered specimens in the air and in the water, without impregnation done to the samples, prior to the measurement. later, the calculation of the density was done by referring to the following formula. density, ρ = [wair / (wair – wwater)] x ρwater [1] issn: 2180-1053 vol. 3 no. 1 january-june 2011 characterization of injection molded 17-4ph stainless steel prepared with waste rubber binder 15 where: wair = weight of specimen in air wwater = weight of specimen in water ρwater = density of water 2.5 tensile test the tensile properties of sintered samples are measured using the universal testing machine utm model ag–1/100 kn (shimadzu corporation, japan). the standard test for tensile testing of metallic materials is based on astm e8. prior of the test, specimen’s initial length were marked and measured. then, the specimens were gripped and tensile stress was imposed onto the sample until the fracture occurs. the tensile strength and tensile modulus were measured to determine the mechanical properties of the prepared specimens. 2.6 hardness test micro vickers hardness test with diamond indenter was used as the indentation hardness test against the surface of the material, by using an established machine setup to force a diamond spheroconical indenter under the specified condition. this test is to measure the difference in depth of the indentation according to the mpif standard 51 (under the specified conditions of preliminary and total test forces) to determine the micro hardness of the sintered specimens. 2.7 scanning electron microscope (sem) to understand the properties of 17-4ph stainless steel, the microstructures studies were carried out. this was done by observing through the scanning electron microscope (sem). the cross sectional surface of the highest density fractured samples was selected and the sem micrographs were taken at 1000-times, 3000-times and 50000-times of magnifications by using a secondary and backscattered electron mode image detector. 3.0 results and discussion 3.1 17-4 ph stainless steel powder and green body characterization in this study, the metal powder used is 17-4 ph stainless steel issn: 2180-1053 vol. 3 no. 1 january-june 2011 journal of mechanical engineering and technology 16 (precipitated hardened martensitic stainless steel with cu and nb/ cb additions). the 17-4ph stainless steel metal powder had nearly spherical particle shape with broad size distribution within 1µm into 22µm, as depicted in the sem micrograph of figure 2. this powder shape exhibits that gas atomization powder had higher possibility to be packed in higher density. the micrograph shows that the particles are polydispersed in nature where the surface of particles looks smooth. the particles size with less than 22µm indicates largest particles size (figure 2a). figure 2b shows the morphology of green body at fracture surface with 1000x of magnification using the sem observation. it can be clearly seen that the binders are homogeneously filled at all the interstitial spaces between the powder particles. the formation of pores occurred because of air trapped during the injection molding or binder system shrinked during the cooling process. the powder particles are linked to each other by a network of binders. this proved that the function of binder system in mim processing gives a temporary vehicle for a homogenous powder packing into desired shape and then holding the particles in that shape until the debinding process carried out to remove the binders (german, 1990; german and bose, 1997). 13 binder system shrinked during the cooling process. the powder particles are linked to each other by a network of binders. this proved that the function of binder system in mim processing gives a temporary vehicle for a homogenous powder packing into desired shape and then holding the particles in that shape until the debinding process carried out to remove the binders (german, 1990; german and bose, 1997). figure 2 sem micrograph of the (a) 17-4ph stainless steel used in the feedstock preparation and; (b) the green body after injection molding at fracture (a) (b) (b) soaking period soaking period (a) figure 2: sem micrograph of the (a) 17-4ph stainless steel used in the feedstock preparation and; (b) the green body after injection molding at fracture 13 binder system shrinked during the cooling process. the powder particles are linked to each other by a network of binders. this proved that the function of binder system in mim processing gives a temporary vehicle for a homogenous powder packing into desired shape and then holding the particles in that shape until the debinding process carried out to remove the binders (german, 1990; german and bose, 1997). figure 2 sem micrograph of the (a) 17-4ph stainless steel used in the feedstock preparation and; (b) the green body after injection molding at fracture (a) (b) (b) soaking period soaking period (a) issn: 2180-1053 vol. 3 no. 1 january-june 2011 characterization of injection molded 17-4ph stainless steel prepared with waste rubber binder 17 13 binder system shrinked during the cooling process. the powder particles are linked to each other by a network of binders. this proved that the function of binder system in mim processing gives a temporary vehicle for a homogenous powder packing into desired shape and then holding the particles in that shape until the debinding process carried out to remove the binders (german, 1990; german and bose, 1997). figure 2 sem micrograph of the (a) 17-4ph stainless steel used in the feedstock preparation and; (b) the green body after injection molding at fracture (a) (b) (b) soaking period soaking period (a) figure 3: (a) tensile strength and (b) young modulus for each sample set 3.2 tensile test results there were nine sample groups tested with the tensile test to obtain the tensile strength and modulus of elasticity (young’s modulus). the maximum strength and young’s modulus are clearly depicted as in the figure 3. the lowest value of tensile strength is samples that soaked for 30 minutes. the samples sintered using 150c/min shows the higher tensile strength for each difference soaking period, except for the sample that soaked at 60 minutes. at the initial of sintering process, a constant heating rate of 50c/min is used for varied soaking period of 30, 60 and 120 minutes. thus, the increasing of soaking period increased the tensile strength before it reduced at 120 minutes of soaking period. this is likely due to the presence of pores in the sintered parts which tend to allow further growth of grain boundaries as a result of excessive or abnormal grain growth during the long period of sintering time. excessive grain growth will result grain coarsening which will lower the strength of the sintered parts (klar and samal, 2007). 3.3 fracture surface morphology of tensile specimens tensile properties obtained from tensile testing are correlated with the morphological observation, where it provides direct evidence of the strengthening or reinforcing mechanism to the fabricated metal composites. figure 4 depicts the sem micrographs of vacuum sintered specimens. through the observation by using 1000x magnification on the samples soaked for 30 minutes, there are large numbers of pores presence between the particles, where the original shape of the powder can still be discerned even though the particles have been fused together. when the soaking period or sintering time increased to the 60 minutes, there is less porosity of microstructure, existed. this is near to fully densification of sintered specimens, although the original shape issn: 2180-1053 vol. 3 no. 1 january-june 2011 journal of mechanical engineering and technology 18 of powder can still be detected, where the powder boundaries were replaced by grain boundaries development. however, if the soaking period was increased into 120 minutes, the microstructures began to coarse and the grains began to grow excessively. there are small pores still occurred in that specimens with the dimple rupture phenomenon obviously revealed. the fracture morphology of 17-4 ph stainless steel specimens sintered at 100c/min and 150c/min of heating rates for three different soaking periods are relatively having similar microstructure evolution to specimens that were treated at 50c/min. however, most of specimens exhibit dimple rupture and pores. this is caused by the δ-ferrite formation occurred during the sintering that decreased the amount of porosity and exhibits ductile fracture, where the morphologies of the tensile fracture surface of gas atomized exhibits the dimple rupture with spherical shape of pores (gulsoy et.al., 2007). the pores existed in the sintered parts tend to provide extra space for further growth of the grain boundaries that resulted the excessive or abnormal grain growth during the long period of sintering time. excessive grain growth will result larger grain formation which particularly lower the strength of the sintered parts (klar and samal, 2007). 15 figure 4 the sem images of fractured surface of the samples used 50c/min of heating rate for three different soaking period: (a) 30 minutes; (b) 60 minutes; (c) 120 minutes (a) (b) (c) (a) (b) figure 4: the sem images of fractured surface of the samples used 50c/min of heating rate for three different soaking period: (a) 30 minutes; (b) 60 minutes; (c) 120 minutes issn: 2180-1053 vol. 3 no. 1 january-june 2011 characterization of injection molded 17-4ph stainless steel prepared with waste rubber binder 19 15 figure 4 the sem images of fractured surface of the samples used 50c/min of heating rate for three different soaking period: (a) 30 minutes; (b) 60 minutes; (c) 120 minutes (a) (b) (c) (a) (b) 15 figure 4 the sem images of fractured surface of the samples used 50c/min of heating rate for three different soaking period: (a) 30 minutes; (b) 60 minutes; (c) 120 minutes (a) (b) (c) (a) (b) figure 5: (a) actual density and (b) hardness; of 17-4ph sintered samples for different heating rates and soaking periods 3.4 density and hardness evaluation the theoretical density of sintered 17-4ph stainless steel is 7.5 g/cm3. in this study, the actual densities of sintered 17-4ph stainless steel specimens are shown as in the figure 5a. the graph shows that samples were treated at 60 minutes of soaking periods for each heating rates gives the higher value of densification, where the sample treated at 30 minutes of soaking periods exhibit the lower density values. previous studies found that the heating rates for vacuum sintering will affect the density of the sintered specimens (liu et.al., 2000). the actual value of hardness for sintered sample of 17-4ph stainless steel at different heating rate and soaking period is presented in figure 5b. the hardness was increased with the soaking period for the entire heating rate before it reduced at 120 minutes of the soaking period. it was found that the decrease of hardness value at 120 minutes of soaking period applies for specimens treated at 100c/min and 150c/min of the issn: 2180-1053 vol. 3 no. 1 january-june 2011 journal of mechanical engineering and technology 20 heating rate. this phenomenon occurred due to long soaking duration after rapid heating of sample treatment. for the specimens treated at 50c/ min of heating rate and 120 minutes of soaking period, the grains were constantly growing started from the lower sintering temperature until the temperature of 13600c. at this condition, the grain is continuously growing in the constant mode. this phenomenon was caused by the reduction of free energy that concurrently minimizing the interfacial energy per unit volume against the grain growth where the hardness decreased by larger size of grains formation (german, 1996). 4.0 conclusion combination of 17-4ph stainless steel metal powder with waste rubber binder provides an increase to the mechanical properties of the injected specimens that sintered at below than 120 minutes of soaking period with optimum heating rate of 50c/min. the combination of these parameters during the sintering process gives the higher density, higher tensile strength, less porosity and homogenous grain shape morphology of the sintered samples. although the introduction of waste rubber as binder system in mim fabricated product would not affect to the properties of sintered component due to achievement of the nearly theoretical values of the resulted mechanical properties, but the application of innovative and novel binder system is compatible and very promising for the 17-4ph stainless steel metal powder product fabricated through the mim process. utilization of waste rubber provides an alternative binder for metal injection molding process. this alternative gives the advantageous of an economical and environmental friendly material as well as to solve the problems of waste rubber disposal. 5.0 acknowledgement special thanks for sirim amrec and universiti teknikal malaysia melaka (utem) for funding this research. also, special appreciation to miss nurhashima shafie for her dedicated support in experimental works and result analysis. 6.0 references e. rodriguez-senin, a. varez, b. levenfeld, j.m. torralba, m.a. paris. 2005. processing of mn-zn ferrites using mould casting with acrylic thermosetting binder. powder metallurgy, vol. 48, p. 249-253. issn: 2180-1053 vol. 3 no. 1 january-june 2011 characterization of injection molded 17-4ph stainless steel prepared with waste rubber binder 21 k. tan, c. li, h. meng, z. wang. 2008. preparation and characterization of thermoplastic elastomer of poly (vinyl chloride) and chlorinated waste rubberpolymer testing, vol. 28, p. 2-7. m.a. omar, i. subuki. 2006. microstructure evolution during rapid debinding of mim compactmaterial letters, vol. 2, p. 42-47. h.o. gulsoy, s. ozbek, t. baykara. 2007. microstructural and mechanical properties of injection moulded gas and water atomized 17-4 ph stainless steel powder. powder metallurgy, vol. 50, no. 2, p. 120-126. s.j.l. kang. 2005. sintering-densification, grain growth & microstructure. elsevier butterworth-heinemann, burlington. r.m. german. 1990. powder injection molding. metal powder, new york, p. 3-124. r.m. german, a. bose. 1997. injection molding of metal and ceramic. metal powder industries federation, new jersey. e. klar, p. samal. 2007. powder metallurgy stainless steel processing, microstructures and properties. asm international, new jersey , p. 39, 49-53. z.y. liu, n.h. loh, k.a. khor, s.b. tor. 2000. sintering of injection molded m2 high-speedteel materials letters, vol. 45, p.32-38. r.m. german. 1996. sintering theory and practice. john wiley & sons. inc, new york,p.8-10, 96-99, 166-169. preparation of papers in a two column model paper format issn: 2180-1053 vol. 9 no.2 july – december 2017 59 experimental and numerical investigation on hydro-forming of stepped tubes for aluminum alloy 6063 m. safari 1* , j. salimi 2 , s. hamidipour 3 1,2,3 department of mechanical engineering, arak university of technology, arak 38181-41167, iran abstract in this paper, hydro-forming process of stepped tube is studied experimentally and numerically. the material is aluminum alloy 6063. a new bush driving mechanism has been used in the tube hydroforming die set and consequently a stepped tube with sharp corners and high expansion ratio can be produced. in order to more investigation, effects of some process parameters such as internal pressure, die stroke and friction coefficient on the wall thickness of manufactured specimen and die filling are surveyed. the results show that the die is betted filled with an increase in the die stroke and internal pressure due to more material flow into the die corners. it is concluded that the reduction of wall thickness is increased with increasing the die stroke, internal pressure and friction coefficient because of more contact surface area of the tube with the die and consequently more friction force in the contact pairs. keywords: hydroforming process, stepped tube, experimental and numerical investigation 1.0 introduction automotive parts currently under development or in production include seat frames, engine cradles, rails, exhaust manifolds and space frame components. interest in the tube hydroforming process by the automotive industry is due to the possibility of replacing many multi-piece stamped and welded assemblies in body, frame or chassis components with one-piece hydroformed components. thus, there is a great potential for not only weight saving but also for tooling and labor cost saving that may occur due to the elimination of multi-stage stamping and assembly processes through part consolidation. additional benefits of tube hydroforming over stamping are improved dimensional accuracy, improved structural strength and stiffness, and consistent dimensional repeatability (dohmann & hartl, 1996), (smith et al., 2003), (stoughton & yoon, 2004). in recent years, pre-form processes such as bending, crushing and mechanical forming during the hydro-forming process have been used in order to produce a part with required mechanical properties or complex shapes. with these mechanical operations * corresponding author email: m.safari@arakut.ac.ir journal of mechanical engineering and technology issn: 2180-1053 vol. 9 no.2 july – december 2017 60 before or during the hydroforming process, required hydroforming pressure is decreased and also a part with more uniform thickness distribution can be produced. in recent years, many researches have been reported in the field of tube hydroforming process (kim & hwang, 2002), (tabatabaei et al., 2013). (kim et al., 2017) studied the finite element modeling of the hydroforming process for niobium tubes intended for use in superconducting radio frequency (srf) cavities. in their work, crystal plasticity (cp) model was constructed that included the evolution of crystallographic orientation during deformation as well as the anisotropy of tubes in all directions and loading conditions. the results showed that high quality predictions of the deformation under hydroforming of nb tubes can be obtained using cp-fem based on their known texture and the results of tensile tests. (shi et al., 2017) studied the necking and fracture in hydroforming of tubes under internal pressure through using the gtn model numerically. they investigated the effect of superimposed hydrostatic pressure on necking (both uniform stain and localized necking), fracture initiation and fracture surface formation. their results showed that superimposed hydrostatic pressure has a great impact on the onset of fracture with the increase of superimposed hydrostatic pressure, but insignificant influence on the uniform strain. (hashemi et al., 2015) predicted bulge height of aluminum tubes aa6063 using ductile fracture criteria at high temperatures. they calibrated ductile fracture criteria by performing several uniaxial tensile tests at different temperatures and strain rates. free bulging process of tubes was simulated using finite element method and different loading curves were used to bulge the tubes. in their work, prediction of ductile fracture was compared with the experimental results measured on a warm free bulging set-up. their results showed that ayada ductile fracture criterion was able to predict the bulge height of aluminum tubes at high temperatures. (hajializadeh & mashhadi, 2015) investigated the finite element analysis of impulsive hydroforming on the sheet and tube using an explicit scheme. the studied the effect of discharge energy, die radius and friction coefficient. it was observed that the discharge energy value has major effect on the process and the friction coefficient has minor effect relative to the others. their results showed that al6061-t6 tube did not sustain any damage even by experiencing stresses near the ultimate strength stress due to the high strain rate of the process. (bihamta et al., 2015) optimized the variable thickness tube drawing and two-step bending in tube hydroforming process in order to obtain parts without any problems like bursting or un-filled zones at the end of the forming processes. (liu et al., 2014) studied austenite-to-martensite transformation and microcrack initiation and propagation of the tube during t-shape hydroforming using electron backscattering diffraction, scanning electron microscopy, and transmission electron microscopy. the results showed that compared to the compressive stress, metastable austenite with similar strain surrounding or inside the grains transformed easier under tensile loading conditions. the inclusions were responsible for microcrack initiation. the propagation of the cracks was hindered by martensite/austenite constituent due to transformation induced plasticity effect. (cui et al., 2014) studied the effect of external pressure on the critical effective strain theoretically, numerically and experimentally in order to explore the deformation behavior of double-sided tube hydroforming in square-section die. it was shown that increasing of external pressure has an effect on the fraction of grain boundaries, the number and size of the microvoids and the microhardness in the transition zone, and experimental and numerical investigation on hydro-forming of stepped tubes for aluminum alloy 6063 issn: 2180-1053 vol. 9 no.2 july – december 2017 61 thus increases the critical effective strain in the transition zone. it was concluded that the deformation ability of the transition zone is improved by the external pressure in double-sided tube hydroforming of square-section. one of the main limitations in tube hydroforming process is production of stepped tubes with sharp cornesrs and also high expantion ratios (expanded tube diameter∕initial tube diameter). for production of stepped tubes with sharp corners or high expansion ratios, high forming pressures are needed. using high pressures, necking and tearing are occurred in the produced part. therefore, production of stepped tubes with sharp corners of high expansion ratios is very difficult in industries. in recent years some researchers have focused on hydroforming process of stepped tubes with sharp corners. (kridli et al., 2003) investigated the thickness variation and corner filling in tube hydroforming process. using commercial finite element code abaqus/standard they simulated twodimensional plane-strain finite element models of the tube hydroforming process. the interaction of material properties and die geometry on the selection of hydroforming process parameters was examined. it was concluded that the thickness distribution is a function of the die corner radius and strain-hardening behavior of the material. in addition, the thickness variation distribution could be reduced if a larger corner radius was used. (hwang & chen, 2005) have examined the die filling in a square cross sectional die by analytical, numerical and experimental methods. it was proved that higher pressure was required to fill the die corner if the corner radius was decreased. (loh-mousavi et al., 2007) have studied the filling of the die corner in hydroforming of a tube with box die and a t-shape die. a pulsating pressure path was used to improve the die filling. although this path could improve the filling of the die corners, but the dies were not filled completely. on the other hand, producing a pulsating pressure path is very difficult than linear on constant pressures. as it was mentioned above production of stepped tubes with sharp corners of high expansion ratios is very difficult and therefore there has been done few researchers in this field. however, in this paper a new mechanism is proposed for production a stepped tube with sharp corners and high expansion ratio. using this new mechanism, bending process is associated with hydroforming process and combination of these two forming processes leads to production of a stepped tube with desirable features. in addition, the effects of some process parameters such as internal pressure, friction coefficient and die stroke on the thickness distribution of produced stepped tube and also filing of die is investigated. 2.0 experimental work the tubes were made of aluminum alloy 6063, and have 5 mm thickness, 26 mm outside diameter, and 170 mm length as initial dimensions. in figure1, the schematic view of hydroformed tube with its dimensions is shown. journal of mechanical engineering and technology issn: 2180-1053 vol. 9 no.2 july – december 2017 62 figure 1. schematic view of final hydroformed tube with its dimensions a conventional hydroforming die for stepped tubes is composed of two halves. after placing the tube in the lower die, the upper and lower dies are closed, the tube is filled with liquid, and the punches seal the tube. by applying simultaneously the internal pressure and axial feeding, the tube is formed into the shape of the die cavity. however, with the conventional dies, at the end of the process, the corner of the cavity was not filled completely. in figure 2 the schematics of die set for hydroforming of stepped tube are shown. it should be noted that the die set contains two additional bushes, compared with the common tube hydroforming dies [9-13]. (a) experimental and numerical investigation on hydro-forming of stepped tubes for aluminum alloy 6063 issn: 2180-1053 vol. 9 no.2 july – december 2017 63 (b) figure 2. (a) the die set for hydroforming of stepped tube, (b) cut-out view of the die with positioning the tube the forming stages of the tube in this die set is such that initially the tube is placed in the die, filled with liquid, and sealed with the punches. then, by increasing the internal pressure, the tube is bulged and contacts the die walls that are fixed. by maintaining the internal pressure, the two bushes move until the die cavity is filled completely. it should be explained that in the common dies, the punches that are in contact with the tube ends, exert the axial feeding on the tube and the die corners cannot be filled completely. however, in the proposed die set in each side of the tube, even though the punch is similarly in contact with the tube end, but the bush gradually contacts the tube. at the end of the forming stage, both the punch and bush will be in complete contact with the tube and will exert axial feeding on the tube. this will lead to complete filling of the die corners. figure3 shows the die set mounted on the test machine. as it is seen in figure 3, in this work a single action universal press machine 250 kn is used. however, in order to hydroforming a stepped tube a press machine with double action is needed. for this purpose, using a new mechanism, single action of press machine is converted to double action. the pressure generating system was a hydraulic unit with a maximum capacity of 30 mpa. the working pressure is regulated by a pressure relief valve. in the fem simulation, it was necessary to introduce the true stress-strain curve of the tube material. thus, tension tests are performed. figure 4 shows the true stress-strain curve obtained for the material. in figure 5, hydroformed stepped tube with the proposed die is shown. as it is shown in this paper, using the proposed die set a stepped tube with approximately sharp corners can be produced. journal of mechanical engineering and technology issn: 2180-1053 vol. 9 no.2 july – december 2017 64 figure 3. the die set mounted on the test machine figure 4. the true stress-strain curve obtained for aluminum alloy 6063 tube experimental and numerical investigation on hydro-forming of stepped tubes for aluminum alloy 6063 issn: 2180-1053 vol. 9 no.2 july – december 2017 65 figure 5. the hydroformed stepped tube with the proposed die and mechanism 3.0 numerical work in order to simulate the hydroforming process of stepped tube, commercial software, abaqus 6.14 is used. the simulation conditions, such as boundary conditions, interactions, internal pressure and loading paths are modeled as same as experiments. it should be noted that due to axial symmetry of parts and simulation conditions, 2d model is simulated in finite element method. the tube is modeled as a 2d axisymmetric with cax4r element. the die components are modeled as 2d axisymmetric analytical rigid elements. based on reference [9], the coefficient of friction between the workpiece and the die surfaces is considered to be 0.06 in the simulations. in the numerical simulations, two steps are considered for hydroforming. in the first step, the internal pressure increases until the tube bulges. in the second step, by axial feeding of punches and the bushes, and by remaining the internal pressure constant, the tube completely fills the die cavities. in figure 6, hydroformed stepped tube in the numerical simulations is seen. figure 6. the hydroformed stepped tube in the numerical simulation journal of mechanical engineering and technology issn: 2180-1053 vol. 9 no.2 july – december 2017 66 4.0 results and discussion in hydroforming simulations of stepped tube, the mesh size is important. using coarse elements reduces the accuracy of simulations and very fine elements will increase simulations time without improving the accuracy of results. in order to find an optimum mesh size, internal energy for whole model is calculated at the end of hydroforming process for various numbers of elements (figure7). from figure 7, it is concluded that optimum number of elements for these simulations is about 2200 elements. figure 7. internal energy of whole model versus number of elements at the end of hydroforming process in figure8, thickness distributions of a hydroformed stepped tube that is completely formed in the die obtained from experimental and numerical works are shown. it should be mentioned that for this experiment, the applied internal pressure and die stroke are 118 mpa and 8 mm, respectively. figure 8. thickness distributions of a completely hydroformed stepped tube obtained from experimental and numerical works experimental and numerical investigation on hydro-forming of stepped tubes for aluminum alloy 6063 issn: 2180-1053 vol. 9 no.2 july – december 2017 67 as it is seen from figure 8, the thickness reduction in a completely hydroformed stepped tube with sharp corners is approximately 40% in both experimental and numerical results. the results of figure8 show that the maximum thickness reduction is happened in the expanded area of stepped tube due to effects of internal pressure. in addition, it is proved from figure8 that there is a good and suitable agreement between experimental measurements and numerical results. as the numerical simulations have been verified with experimental measurements, in the following the effects of some process parameters such as internal pressure, friction coefficient and die stroke on the thickness distribution of produced stepped tube and also die filling are investigated numerically. 4.1 effect of internal pressure in order to investigate the effect of internal pressure on thickness distribution and die filling, in the numerical simulations the initial tube will be hydroformed with three different internal pressures. for this purpose five different pressures such as 90, 100, 110, 115 and 118 mpa are applied in the numerical simulations. axial feeding and friction coefficient are hold as 0.05 and 8 mm respectively. in table 1, the results of die filling and minimum wall thickness for hydroformed stepped tube are presented. the results show that the initial tube cannot fill the die with internal pressures of 90 and 100 mpa. however, the die is partially filled with the initial tube for internal pressure of 110 and 115 mpa and completely filled with the internal pressure of 118 mpa. the reason is that with increasing the internal pressure the material flow into the die corners is increased and hence die is more filled. the results indicated that with increasing in the internal pressure, reduction of wall thickness of stepped tube is increased because of more contact surface area of the tube with the die and consequently more friction force in the contact pairs. table 1. the results of die filling and minimum wall thickness of hydroformed stepped tube for different internal pressures internal pressure (mpa) 90 100 110 115 118 filling of die not filled not filled partially filled partially filled completely filled minimum wall thickness (mm) --------1.76 1.72 1.69 4.2 effect of die stroke in this section, effect of die stroke on filling of the die and minimum wall thickness of hydroformed stepped tube is investigated. for this purpose, five different die strokes such as 3, 5, 6, 7 and 8 mm are considered in the numerical simulations while internal pressure and friction coefficient are hold as 118 mpa and 0.05, respectively. the results of variation of die stroke on filling of the die and minimum wall thickness are presented in table 2. journal of mechanical engineering and technology issn: 2180-1053 vol. 9 no.2 july – december 2017 68 table 2. the effect of die stroke on filling of the die and also minimum wall thickness of hydroformed stepped tube die stroke (mm) 3 5 6 7 8 filling of die not filled not filled partially filled partially filled completely filled minimum wall thickness (mm) --------1.93 1.80 1.66 as it is seen in table 2, better filling of the die is happened with an increase in the die stroke. table 2 shows that the reduction of wall thickness is increased with increasing the die stroke. the material flow into the die is increased with increasing in the die stroke. therefore, the die is better filled. however, with an interaction between die stroke, internal pressure and friction coefficient and due to increasing in the contact surfaces between die and the tube, more thickness reduction is seen in the manufactured tubes. 4.3 effect of friction coefficient in figure 9, the effect of friction coefficient (between die and tube) on the minimum wall thickness of hydroformed stepped tube is shown. it should be noted that in the investigation of friction coefficient, the internal pressure and die stroke have been adjusted as 118 mpa and 8 mm respectively. in this state, the die can be filled completely by the tube. however, in this section only the effect of friction coefficient of the minimum wall thickness of hydroformed stepped tube is investigated. figure 9. effect of friction coefficient of the minimum wall thickness of hydroformed stepped tube as it is seen in figure9, with an increase in the friction coefficient, reduction of wall thickness of hydroformed stepped tube is increased because of more contact surface area of the tube with the die and consequently more friction force in the contact pairs. experimental and numerical investigation on hydro-forming of stepped tubes for aluminum alloy 6063 issn: 2180-1053 vol. 9 no.2 july – december 2017 69 5.0 conclusions in this work, hydroforming process of a stepped tube from of al 6063 was investigated experimentally and numerically. a new mechanism was proposed for production a stepped tube with sharp corners and high expansion ratio. the effects of some process parameters such as internal pressure, die stroke and friction coefficient on filling of the die and also minimum wall thickness were investigated. the results showed that using the proposed mechanism in this paper, a stepped tube with sharp corners and high expansion ratio could be manufacture. it was concluded that better filling of the die was happened with an increase in the die stroke and internal pressure due to more material flow into the die corners. the reduction of wall thickness was increased with increasing the die stroke, internal pressure and friction coefficient because of more contact surface area of the tube with the die and consequently more friction force in the contact pairs. 6.0 references bihamta, r., bui, q. h., guillot, m., d’amours, g., rahem, a., & fafard, m., (2015). global optimisation of the production of complex aluminium tubes by the hydroforming process. cirp journal of manufacturing science and technology, 9, 1–11. cui, x. l., wang, x. s. & yuan, sh. j., (2014). deformation analysis of double-sided tube hydroforming in square-section die. journal of materials processing technology, 214(7), 1341–1351. dohmann, f., & hartl, ch., (1996). hydroforming – a method to manufacture lightweight parts. journal of materials processing technology, 60, 669–676. hajializadeh, f., & mashhadi, m. m., (2015). investigation and numerical analysis of impulsive hydroforming of aluminum 6061-t6 tube, journal of manufacturing processes, 20(1), 257–273. hashemi, s. j., moslemi naeini, h., liaghat, g. h., & azizi tafti, r., (2015). prediction of bulge height in warm hydroforming of aluminum tubes using ductile fracture criteria, archives of civil and mechanical engineering, 15(1), 19–29. hwang, y. m., & chen, w. c., (2005). analysis of tube hydroforming in a square cross-section die. international journal of plasticity, 21, 1815–1833. kim, h. s., sumption, m. d., bong, h. j., lim, h., & collings, e. w., (2017). development of a multi-scale simulation model of tube hydroforming for superconducting rf cavities, materials science and engineering: a, 679, 104– 115. http://www.sciencedirect.com/science/article/pii/s1755581715000115 http://www.sciencedirect.com/science/article/pii/s1755581715000115 http://www.sciencedirect.com/science/article/pii/s1755581715000115 http://www.sciencedirect.com/science/article/pii/s1755581715000115 http://www.sciencedirect.com/science/article/pii/s1755581715000115 http://www.sciencedirect.com/science/journal/17555817 http://www.sciencedirect.com/science/journal/17555817/9/supp/c http://www.sciencedirect.com/science/journal/09240136 http://www.sciencedirect.com/science/journal/09240136 http://www.sciencedirect.com/science/journal/09240136/214/7 http://www.sciencedirect.com/science/article/pii/s1526612515000742 http://www.sciencedirect.com/science/article/pii/s1526612515000742 http://www.sciencedirect.com/science/journal/15266125 http://www.sciencedirect.com/science/journal/15266125 http://www.sciencedirect.com/science/journal/15266125/20/supp/p1 http://www.sciencedirect.com/science/article/pii/s1644966514001071 http://www.sciencedirect.com/science/article/pii/s1644966514001071 http://www.sciencedirect.com/science/article/pii/s1644966514001071 http://www.sciencedirect.com/science/article/pii/s1644966514001071 http://www.sciencedirect.com/science/journal/16449665 http://www.sciencedirect.com/science/article/pii/s0921509316312308 http://www.sciencedirect.com/science/article/pii/s0921509316312308 http://www.sciencedirect.com/science/article/pii/s0921509316312308 http://www.sciencedirect.com/science/article/pii/s0921509316312308 http://www.sciencedirect.com/science/article/pii/s0921509316312308 http://www.sciencedirect.com/science/journal/09215093 http://www.sciencedirect.com/science/journal/09215093/679/supp/c journal of mechanical engineering and technology issn: 2180-1053 vol. 9 no.2 july – december 2017 70 kim, j., & hwang, s. m., (2002). preform design in hydroforming by three-dimensional backward tracing scheme of the fem, journal of materials processing technology, 130-131, 100-106. kridli, g. t., bao, l., mallick, p. k., & tian, y., (2003). investigation of thickness variation and corner filling in tube hydroforming. journal of materials processing technology, 133, 287–296. liu, j., zhang, z., manabe, k. i., li, y., & misra, r. d. k., (2014). microstructure evolution in trip-aided seamless steel tube during t-shape hydroforming process. materials characterization, 94, 149–160. loh-mousavi, m., mori, k., hayashi, k., & bakhshi-jooybari, m., (2007). improvement of filling die corners in box-shaped tube hydro-forming by control of wrinkling. key engineering materials, 334, 461–467. shi, y., jin, h., wu, p. d., & lloyd, d. j., (2017). effects of superimposed hydrostatic pressure on necking and fracture of tube under hydroforming, international journal of solids and structures, 113–114, 209–217. smith, l.m., averill, r.c., lucas, j.p., stoughton, t.b., & matin, p.h., (2003). influence of transverse normal stress on sheet metal formability. international journal of plasticity, 19, 1567–1583. stoughton, t.b., & yoon, j.w., (2004). a pressure-sensitive yield criterion under a non-associated flow rule for sheet metal forming. international journal of plasticity, 20, 705–731. tabatabaei, s. a., shariat panahi, m., mosavi mashhadi, m., tabatabee, s. m., & aghajanzadeh, m., (2013). optimum design of preform geometry and forming pressure in tube hydroforming using the equipotential lines method, the international journal of advanced manufacturing technology, 69(9-12), 27872792. http://www.sciencedirect.com/science/journal/09240136 http://www.sciencedirect.com/science/journal/09240136 http://www.sciencedirect.com/science/article/pii/s1044580314001648 http://www.sciencedirect.com/science/article/pii/s1044580314001648 http://www.sciencedirect.com/science/article/pii/s1044580314001648 http://www.sciencedirect.com/science/article/pii/s1044580314001648 http://www.sciencedirect.com/science/article/pii/s1044580314001648 http://www.sciencedirect.com/science/journal/10445803 http://www.sciencedirect.com/science/article/pii/s0020768317300951 http://www.sciencedirect.com/science/article/pii/s0020768317300951 http://www.sciencedirect.com/science/article/pii/s0020768317300951 http://www.sciencedirect.com/science/article/pii/s0020768317300951 http://www.sciencedirect.com/science/journal/00207683 http://www.sciencedirect.com/science/journal/00207683 http://www.sciencedirect.com/science/journal/00207683/113/supp/c issn: 2180-1053 vol. 9 no.2 july – december 2017 21 analysis of nonlinear dynamic behaviour of nanobeam resting on winkler and pasternak foundations using variational iteration method m. g. sobamowo 1 * and g. a. oguntala 2 1 department of mechanical engineering, faculty of engineering, university of lagos, akoka, lagos, nigeria. 2 school of electrical engineering and computer science, faculty of engineering and informatics, university of bradford, uk abstract dynamic modeling of nanobeam under stretching and two-parameter foundations effects result in nonlinear equations that are very difficult to find exact analytical solutions. in this study, variational iteration method is used to develop approximate analytical solutions to nonlinear vibration analysis of nanobeam under the effects of stretching, winkler and pasternak foundations. the governing equation of motion for the nanobeam was derived based on euler-bernoulli beam theory. the developed approximate analytical solutions for the governing equation are used to study the effects of the model parameters on the dynamic behaviour of the nanobeam. the results show that increase in the beam length decreases the natural frequency of vibration while the diameter of the nanobeam increases as the natural frequency increases. as the spring constant increases, the nonlinear frequency ratio decreases. at a high stiffness media, the carbon nanobeam behavior can be modeled as a linear system whose geometric nonlinearity becomes negligible. the nonlinear frequency of nanobeam increases with increase in the vibration amplitude and the discrepancy between the linear and nonlinear responses tends to increase as time evolves. also, it is found that as the foundation parameter increases, the nonlinear vibration frequency ratio increases and the difference between the nonlinear and linear frequency becomes pronounced. these analytical solutions can serve as a starting point for a better understanding of the relationship between the physical quantities of the problems as they provide clearer insights to understanding the problems in comparison with numerical methods. keyword: nonlinear vibration; variational iteration method; nanobeam; winkler and pasternak foundations. *corresponding author e-mail: mikegbeminiyiprof@yahoo.com journal of mechanical engineering and technology issn: 2180-1053 vol. 9 no.2 july – december 2017 22 1.0 introduction the study of dynamic behaviour of nanobeam is an important research area due to small scale of carbon nanobeam (cnb) and their important applications in sensitive devices. the vibration behaviour and properties of the cnb have been investigated in the past few decades. many of the past researches are based on linear vibration analysis of the cnb. however, owing to the small scale of carbon nanobeams, the linear assumptions cannot provide an accurate prediction and analysis of the vibration of the cnb. furthermore, the assumption that carbon nanobeam rests on linear foundations shows apparently that the linear foundation is not very precise approximation for the tiny instruments, and so the obtained past estimations either by numerical or analytical approaches cannot accurately predict the dynamic behaviour of the cnb. therefore, in order to increase the level of prediction or accurately predict the dynamic behaviour of cnb, it is very essential to develop mathematical model for the cnb, which contains geometrical nonlinearity and nonlinear foundations. modeling the nanobeam under such considerations results in nonlinear dynamic equations which are difficult to solve exactly and analytically. however, in many cases under different scenarios, recourse is always made to numerical methods to solve the nonlinear or approximate analytical methods are often applied in which their accuracies largely depend on the number of terms included in the solutions. in some cases, where decomposition procedures into spatial and temporal parts are carried out, the resulting nonlinear equation for the temporal part comes in form of duffing equation. application of exact analytical methods to the nonlinear equation is limited as many of the cases where the exact solutions are generated are not practicable and the solutions hardly provide an all-encompassing understanding of the nature of systems in response to parameters affecting nonlinearity. however, the classical way for finding analytical solution either exact or approximated is obviously still very important since it serves as an accurate benchmark for numerical solutions. although, different approximate analytical methods such as perturbation method (regular or singular perturbation method), homotopy perturbation method (hpm), homotopy analysis method (ham), variational iterative method (vim), differential transformation method (dtm), harmonic balancing method, adomian decomposition method etc. zhou (1986); liao, (1992), (1995); chen and ho. (1996), he, (1998), momani, (2004), el-shahed, (2008); liao and tan, (2007), fernandez (2009); rafiepour et al. (2014). these approximate analytical methods solve nonlinear differential equations without linearization, without discretization or approximation of the derivatives. however, most of the approximate methods give accurate predictions only when the nonlinearities are weak and they fail to predict accurate solutions for strong nonlinear models. also, when they are routinely implemented, they can sometimes lead to erroneous results (sobamowo, 2016). additionally, some of them require more mathematical manipulations and are not applicable to all problems, and thus suffer a lack of generality. for example, dtm proved to be more effective than most of the other analysis of nonlinear dynamic behaviour of nanobeam resting on winkler and pasternak foundations using variational iteration method issn: 2180-1053 vol. 9 no.2 july – december 2017 23 approximate analytical solutions as it does not require many computations as carried out in adm, ham, hpm, and vim. however, the transformation of the nonlinear equations and the development of equivalent recurrence equations for the nonlinear equations using dtm proved somehow difficult in some nonlinear system such as in rational duffing oscillator, irrational nonlinear duffing oscillator, finite extensibility nonlinear oscillator. therefore, the quest for comparatively simple, flexible, generic and high accurate analytical solutions continues. moreover, the determination of adomian polynomials as carried out in adm, the restrictions of hpm to weakly nonlinear problems, the lack of rigorous theories or proper guidance for choosing initial approximation, auxiliary linear operators, auxiliary functions, and auxiliary parameters in ham, operational restrictions to small domains and the search for a particular value for the auxiliary parameter that will satisfy second the boundary condition which leads to additional computational cost in using dtm, ham, adm. in the class of the approximate analytical methods, the relative simplicity and flexibility of vim makes it a desirable and promising method for the analysis of nonlinear problems. the method has been applied to solve many nonlinear problems (he, 1998a, 1998b, 1999a, 1999b, 2000, 2006, 2007a, 2007b, 2011, 2012a, 2012b; 2012c; rafei et al. ,2007, marinca and herisanu, 2006; ganji, et al., 2008; hesameddini and latifizadeh, 2009; wu, 2012). therefore, in this work, variation iteration method (vim) is applied to develop approximate analytical solutions for nonlinear vibration analysis of single-walled carbon nanobeam under the effects of stretching and winkler and pasternak foundations. variational iteration method has shown to be the one of the most effective, accurate, flexible, convenient approximate analytical methods for large class of weakly and strongly nonlinear equations. it is a user-friendly method with reduced size of calculation, direct and straightforward iteration and generates solution with a rapid rate of convergent and without any restrictive assumptions or transformations. in vim, the initial solution can be freely chosen with some unknown parameters and the unknown parameters in the initial solution can be achieved easily. although, there is a rigour of step-by-step integrations coupled with the problem of determination of lagrange multiplier in application of vim, with few number of iteration, even, in some cases, a single iteration of vim can converge to correct solutions or results. the analytical solutions as developed in this work can serve as a starting point for a better understanding of the relationship between the physical quantities of the problems as it provides continuous physical insights into the problem than pure numerical or computation methods. 2.0 problem formulation consider a single-walled carbon nanobeam under the stretching effects and resting on linear and nonlinear elastic foundations (pasternak, linear and nonlinear winkler foundations) as shown in figure 1. assuming the nanobeam to have homogeneous mass density and crossjournal of mechanical engineering and technology issn: 2180-1053 vol. 9 no.2 july – december 2017 24 sectional area along its length. also, the nanobeam material is assumed to be isotropic and the mechanical properties of the foundation are uniform along the length of the nanobeam. based on the assumptions, the governing differential equation are developed as figure 1 a nanobeam resting on winkler and pasternak foundations 24 2 2 2 3 1 34 2 2 2 0 0 2 l p w w w w ea w w ei m k k w k w dx x t t x l x x                             (1) subject to the following initial and the boundary conditions (simply-supported nanobeam) ( , 0) o w x w ( , 0) 0w x & (2) (0, ) ''(0, ) 0w t w t  ( , ) ''( , ) 0w l t w l t  the derivation of the governing equation is shown in the appendix. using the galerkin’s decomposition procedure to separate the spatial and temporal parts of the lateral displacement functions as 1 ( , ) ( ) ( ) n n n n w x t x q t    (3) where ( )q t the generalized coordinate of the system and ( )x is a trial/comparison function that will satisfy both the geometric and natural boundary conditions. for the simply-supported nanobeam considered in this work ( ) n x sin x  (4) analysis of nonlinear dynamic behaviour of nanobeam resting on winkler and pasternak foundations using variational iteration method issn: 2180-1053 vol. 9 no.2 july – december 2017 25 where 0 n n sin l l      n=1, 2, 3, 4….n therefore, eq. (3) becomes 2 1 ( , ) ( ) n n n w x t q t sin x    (5) on substituting eq. (5) into eq. (1) and apply orthogonal principle of the mode shapes, we arrived at 2 3 2 1 1 1 1 1 1 3 1 2 0q q q q q q       && & (6a) 2 3 2 2 2 2 2 2 2 4 2 1 0q q q q q q       && & (6b) while for the undamped nanobeam, we have 2 3 2 1 1 1 1 1 3 1 2 0q q q q q     && (8a) 2 3 2 2 2 2 2 2 4 2 1 0q q q q q     && (8b) where 4 2 2 1 1 1 p ei k k m l l                       4 2 2 2 1 1 2 2 p ei k k m l l                       4 1 3 1 3 4 4 ei k m l               4 3 3 1 3 2 eia k m l               4 2 3 1 3 4 4 eia k m l               4 4 3 1 3 4 2 ei k m l               the initial conditions are 1 0 1(0) (0) 0q x q & 2 0 2(0) (0) 0q y q & (9) journal of mechanical engineering and technology issn: 2180-1053 vol. 9 no.2 july – december 2017 26 3.0 method of solution: variational iteration method in finding direct and practical solutions to the problem, variational iteration method is applied to the simultaneous nonlinear equations. as pointed previously, the variational iteration method is an approximate analytical method for solving differential equations. the basic definitions of the method are as follows the differential equation to be solved can be written in the form ( )lu nu g t  (10) where l is a linear operator, n is a nonlinear operator and g(t) is an inhomogeneous term in the differential equation. following vim procedure, we have a correction functional as  1 0 ( ) ( ) ( ) ( ) ( ) t n n n u t u t lu nu g t d         % (11)  is a general lagrange multiplier, the subscript n is the nth approximation and u% is a restricted variation 0u % applying the above vim procedures to eqs. (7a) and (7b), the following iteration formulations are constructed, letting 1 2 u q v q    2 2 3 2 1 1 1 1 32 1 0 1 t n n n n n n n d u u u sin t u u u v d d                      (12a)   2 2 3 2 1 2 2 2 42 2 0 1 t n n n n n n n d v v v sin t v v v u d d                      (12b) in order to find the periodic solution of eq. (12), we assume an initial approximation for zero-order deformation as 1 2 ( ) ( ) o o u acos t v bcos t    (13) for the first iteration, i.e. n=0   2 2 3 20 1 0 1 1 0 1 0 3 0 02 1 0 1 t d u u u sin t u u u v d d                     (14a) analysis of nonlinear dynamic behaviour of nanobeam resting on winkler and pasternak foundations using variational iteration method issn: 2180-1053 vol. 9 no.2 july – december 2017 27   2 2 3 20 1 0 2 2 0 2 0 4 0 02 2 0 1 t d v v v sin t v v v u d d                     (14b) on substituting the corresponding terms in eq. (13) into eq. (14a) and (14b), we have               2 2 3 3 1 1 1 1 1 1 1 1 2 2 1 0 3 1 1 1 t a cos a cos u acos sin t d ab cos cos                               (15a)               2 2 3 3 2 2 2 2 2 1 1 2 2 2 2 0 4 2 2 1 t b cos b cos v bcos sin t d b acos cos                               (15b) it should be pointed out that 1  and 2  are the nonlinear natural frequencies. after mathematical calculations and simplifications of eq. (15a) and (15b), we have                   2 2 3 3 2 22 2 2 2 1 1 1 2 1 1 2 1 1 1 3 1 2 2 2 2 1 1 1 1 2 1 1 1 113 2 2 2 22 2 1 1 1 2 1 1 2 1 1 1 42 2 2 3 1 4 9 2 2 22 4 2 2 ab ab a u cos t a cos t cos tcos tab                                                                                      3 1 11 2 2 2 2 1 1 1 1 3 3 4 9 cos t cos ta                     (16a) journal of mechanical engineering and technology issn: 2180-1053 vol. 9 no.2 july – december 2017 28                   2 2 4 4 2 22 2 2 2 2 2 2 1 2 1 2 2 1 2 3 2 2 2 2 2 2 2 2 2 2 1 2 1 224 2 2 2 22 2 2 2 2 1 2 1 1 2 1 1 42 2 2 3 1 4 9 2 2 22 4 2 2 ba ba b v cos t b cos t cos tcos tba                                                                                      3 2 22 2 2 2 2 2 2 2 2 3 3 4 9 cos t cos tb                     (16b) we should recall from eq. (5), 2 1 1 2 2 1 1 1 2 1 ( , ) ( ) ( ) ( ) n n n w x t q t sin x q t sin x q t sin x u sin x v sin x           (17) therefore,                 2 2 3 3 2 22 2 2 2 1 1 1 2 1 1 2 1 1 3 1 2 2 2 2 1 1 1 1 2 1 1 1 113 2 2 22 2 1 1 1 2 1 1 2 1 1 42 2 2 3 1 ( , ) 4 9 2 2 22 4 2 2 ab ab a cos t a w x t cos t cos tcos tab                                                                                                   1 3 1 11 2 2 2 22 1 1 1 11 2 2 4 4 2 22 2 2 2 2 2 2 1 2 1 2 2 3 2 2 2 2 2 2 2 2 2 3 3 4 9 1 1 42 2 2 3 1 4 9 sin x cos t cos ta ba ba b b                                                                                                                  2 2 2 3 1 2 1 22 2 24 2 2 2 2 2 2 22 22 2 2 2 2 2 2 22 1 2 1 1 2 2 2 22 3 3 4 4 92 2 cos t sin x cos t cos tcos t cos t cos tba b                                                                            (18) analysis of nonlinear dynamic behaviour of nanobeam resting on winkler and pasternak foundations using variational iteration method issn: 2180-1053 vol. 9 no.2 july – december 2017 29 4.0 determination of natural frequency of the vibration in order to find the natural frequency of the vibration, we have to eliminate the secular term. after eliminating the secular term in u and v, we have       2 2 3 3 2 22 2 2 2 1 1 1 2 1 1 2 1 3 1 2 2 2 2 1 1 1 1 1 1 42 2 2 3 1 0 4 9 ab ab a a                                         (19a) and       2 2 4 4 2 22 2 2 2 2 2 2 1 2 1 2 2 3 2 2 2 2 2 2 2 2 2 1 1 42 2 2 3 1 0 4 9 ba ba b b                                         (19b) it should be noted that from eq. (3), eq. (5) and the initial conditions in eq. (9) that 1 2 ( , 0) o o o w x u sin x v sin x w    (20) which can be written as 1 1 2 2 ( ) ( ) o acos t sin x bcos t sin x w     (21) for the general case of , 0a b  , eq. (19a) and eq. (19b) implicitly generate the main frequencies of the symmetric and un-symmetric modes of oscillations. however, for the special case of 0, 0a b  , after simplifications, we arrived at    4 2 2 2 2 2 21 1 1 1 1 1 19 10 7 0a a           (22) on solving eq. (22), we arrived at journal of mechanical engineering and technology issn: 2180-1053 vol. 9 no.2 july – december 2017 30       2 2 2 2 2 2 2 2 1 1 1 1 1 1 1 1 10 7 10 7 36 18 a a a             (23) a further simplification gives  2 2 4 2 2 2 41 1 1 1 1 1 1 10 7 64 104 49 18 a a a           (24) and      3 3 1 11 1 1 12 2 2 2 2 2 2 2 1 1 1 1 1 1 1 1 3 33 1 4 9 4 9 cos t cos ta a u a cos t                                   (25a) 1 0v  (25b) therefore,      3 3 1 11 1 1 12 2 2 2 2 2 2 2 1 1 1 1 1 1 1 1 3 33 1 ( , ) 4 9 4 9 cos t cos ta a w x t a cos t sin x                                       (26) where 1 1 ( ) o a w sec t cosec x  also, for the special case of 0, 0a b  , after simplifications, we arrived at    4 2 2 2 2 2 22 2 2 2 2 2 29 10 7 0b b           (27) which gives       2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 10 7 10 7 36 18 b b b             (28) a further simplification gives analysis of nonlinear dynamic behaviour of nanobeam resting on winkler and pasternak foundations using variational iteration method issn: 2180-1053 vol. 9 no.2 july – december 2017 31  2 2 4 2 2 2 42 2 2 2 2 2 2 10 7 64 104 49 18 b b b           (29) and 1 0u  (30a)      3 3 2 22 2 1 22 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 3 33 1 4 9 4 9 cos t cos tb b v b cos t                                   (30b) therefore      3 3 2 22 2 2 22 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 3 33 1 ( , ) 4 9 4 9 cos t cos tb b w x t b cos t sin x                                       (31) where 2 2 ( ) o b w sec t cosec x  also, it can easily be seen that as the nonlinear term tends to zero, the frequency ratio of the nonlinear frequency to the linear frequency, 1,2 1,2   tends to 1. 1,2 ,3,4 1,2 0 1,2 1lim     (32) also, as the amplitudes a and b tend to zero, the frequency ratio of the nonlinear frequency to the linear frequency, 1,2 1,2   tends to 1. 1,2 , 0 1,2 1 a b lim    (33) for very large values of the amplitudes a, b, we have journal of mechanical engineering and technology issn: 2180-1053 vol. 9 no.2 july – december 2017 32 1,2 , 1,2 a b lim     (34) table: parameters used for the simulation s/n parameter symbol values used 1. modulus of elasticity e 1-1.2×10 12 pa 2. density of the nanobeam ρ 1.2-2.3 ×10 3 kg/m 3 3. winkler foundation constant, k1 0-10 6 n/m 2 4. pasternak linear foundation constant, kp 0-10 -5 n/m 2 5. pasternak nonlinear foundation constant k3 0-10 15 n/m 2 6. length of the nanobeam l 10-100 nm 7. diameter of the nanobeam d 0.56 nm 5.0 results and discussions the first-five normalized mode shapes of the simple-simple beam are shown in figure 2. also, the figure shows the deflections of the beam along the beams’ span at five different buckled and mode shapes. from the first mode shape, the highest deflection occurs at the mid-span of the beam due to the symmetrical nature of the boundary conditions of the simply-simply support beam. figure 2. the first five normalized mode shaped of the 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 -2 -1.5 -1 -0.5 0 0.5 1 1.5 2 2.5 3 dimensionless beam lenght m o d e s h a p e f u n c ti o n 1 st mode shape 2 nd mode shape 3 rd mode shape 4 th mode shape 5 th mode shape analysis of nonlinear dynamic behaviour of nanobeam resting on winkler and pasternak foundations using variational iteration method issn: 2180-1053 vol. 9 no.2 july – december 2017 33 figure 3. variation of nanotube length on natural frequency under simple-simple supports figure 4. variation of nanotube length on natural frequency 10 20 30 40 50 60 70 80 90 100 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 lenght of nanobeam (nm) n o rm a li z e d n a tu ra l f re q u e n c y first natural frequency second natural frequency 10 15 20 25 30 35 40 45 50 55 0 1 2 3 4 5 6 7 8 diameter of nanobeam (nm) n o rm a li z e d n a tu ra l f re q u e n c y first natural frequency second natural frequency journal of mechanical engineering and technology issn: 2180-1053 vol. 9 no.2 july – december 2017 34 figure 5. effects of winkler foundation parameter on natural frequency figure 3 and 4 show the effects of nanobeam length and diameter on the normalized natural frequencies of the beam, respectively. the increase in the beam length decreases the natural frequency of vibration while as the diameter of the nanobeam increases, the natural frequency increases. the observations are in good agreements with the established results in literature. figure 5 is associated with the variation of the nonlinear frequency ratios of the cnb with spring constant with spring constant of the foundation/surrounding medium. from the figure, it could be seen that with the increase of the spring constant, the nonlinear frequency ratio decreases. it is observed that by increasing the spring constant of the surrounding medium, the nonlinear frequencies get close to the linear frequencies so that nonlinearity becomes less evident for the spring constants of large enough i.e. the influence of nonlinearity is more prominent for low stiffness of elastic media. however, for high stiffness media, nonlinear vibration frequencies are very close to linear ones. this establishes that at a high stiffness media, the cnb behavior can be modeled as a linear system whose geometric nonlinearity becomes negligible. also, the variation of nonlinear frequency with the non-dimensional amplitude for cnb is depicted in the figure. in contrast to linear systems, the nonlinear frequency ratio is strongly dependent on amplitude so that the larger the amplitude, the more pronounced the discrepancy between the linear and nonlinear frequencies becomes. this mean that the nonlinear frequency of nanobeam increases with increase in the vibration amplitude. also, it was found that the discrepancy between the linear and nonlinear responses tends to increase as time evolves. figure 6 depicts that as the foundation parameter increases, the nonlinear vibration frequency ratio 0 50 100 150 200 250 300 3.2 3.4 3.6 3.8 4 4.2 4.4 4.6 4.8 5 amplitube (nm) n o rm a li z e d n a tu ra l f re q u e n c y k 1 =100000 n/m k 1 =200000 n/m analysis of nonlinear dynamic behaviour of nanobeam resting on winkler and pasternak foundations using variational iteration method issn: 2180-1053 vol. 9 no.2 july – december 2017 35 increases and the difference between the nonlinear and linear frequency becomes pronounced figure 6. effects of foundation parameter on natural frequency 0 1 2 3 4 5 6 x 10 -5 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 foundation parameter, k p (n/m. 2 ) n o rm a li z e d n a tu ra l f re q u e n c y first natural frequency second natural frequency journal of mechanical engineering and technology issn: 2180-1053 vol. 9 no.2 july – december 2017 36 . figure 7. effects of foundation parameter on the midpoint displacement figure 8. effects of foundation parameter on midpoint displacement 0 10 20 30 40 50 60 70 80 90 100 -1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 1 time (sec.) n o rm a li z e d d is p la c e m e n t 0 10 20 30 40 50 60 70 80 90 100 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 time (sec.) n o rm a li z e d d is p la c e m e n t analysis of nonlinear dynamic behaviour of nanobeam resting on winkler and pasternak foundations using variational iteration method issn: 2180-1053 vol. 9 no.2 july – december 2017 37 effects of foundation parameter on the midpoint deflection time history are illustrated in figure 7 and 8 . figure 7 displays the midpoint deflection time history for the nonlinear analysis of carbon nanobeam when kp=0.01 while figure 8 presents the midpoint deflection time history for the nonlinear analysis of carbon nanobeam when kp = 0.03 6.0 conclusions in this work, nonlinear vibration analysis of nanobeam has been studied under the effects of stretching and winkler and pasternak foundations using variational iteration method. the increase in the beam length decreases the natural frequency of vibration while as the diameter of the nanobeam increases, the natural frequency increase. the increase of the spring constant, the nonlinear frequency ratio decreases. it was established that at a high stiffness media, the cnb behavior can be modeled as a linear system whose geometric nonlinearity becomes negligible. the nonlinear frequency of nanobeam increases with increase in the vibration amplitude and the discrepancy between the linear and nonlinear responses tends to increase when time evolved. as the foundation parameter increases, the nonlinear vibration frequency ratio increases and the difference between the nonlinear and linear frequency becomes pronounced. these analytical solutions can serve as a starting point for a better understanding of the relationship between the physical quantities in the problems as it provides clearer insights to understanding the problems in comparison with numerical methods. 7.0 references chen, c. k. and s.h. ho, s. h. (1996). application of differential transformation to eigenvalue problems. journal of applied mathematics and computation, 79, 173188. el-shahed, m. (2008). application of differential transform method to non-linear oscillatory systems. communic. nonlin. scien. numer. simul. 13, 1714-1720. fernandez. a. (2009). on some approximate methods for nonlinear models. appl math comput., 215. :168-74. ganji, s. s., ganji, d. d., babazadeh, h. and karimpour, s. (2008). variational approach method for nonlinear oscillations of the motion of a rigid rod rocking back and cubic-quinticduffing oscillators. prog. electromagn. res. m 4, 23–32. he j. h. (1998a). approximate analytical solution for seepage flow with fractional derivatives in porous media, computer methods in applied mechanics and engineering 167 (1–2), 57–68. journal of mechanical engineering and technology issn: 2180-1053 vol. 9 no.2 july – december 2017 38 he j. h. (1998b)., approximate solution of nonlinear differential equations with convolution product nonlinearities, computer methods in applied mechanics and engineering 167 (1–2), 69–73. he j. h. (1999). variational iteration method — a kind of non-linear analytical technique: some examples, international journal of non-linear mechanics 34 (4), 699–708. he j. h. (2000). variational iteration method for autonomous ordinary differential systems, applied mathematics and computation 118 (2–3) ,115–123. he j. h. (2006). x.h.wu, construction of solitary solution and compacton-like solution by variational iteration method, chaos, solitons and fractals, 29 (1), 108–113. he j. h. (2007a). variational iteration method—some recent results and new interpretations, journal of computational and applied mathematics, 207 (1), 3–17. he j. h. and wu x. h. (2007b). variational iteration method: new development and applications, comput. meth. appl. mech. eng, 54(7-8): 881-894. he j. h. (2011). a short remark on fractional variational iteration method, phys. lett. a, 375(38) 3362-3364 he j. h. (2012a). an approximation to solution of space and time fractional telegraph equations by the variational iteration method , mathematical problems in engineering, 394212 he j. h. (2012b). comment on "variational iteration method for fractional calculus using he's polynomials" , abstract and applied analysis, 964-974 hesameddini e, latifizadeh h. (2009). reconstruction of variational iteration algorithms using the laplace transform, int. j. nonlin. sci. num., 10(11-12): 1377-1382 liao, s. j. (1992) the proposed homotopy analysis technique for the solution of nonlinear problems, ph. d. dissertation, shanghai jiao tong university. liao, s. j. (1995) an approximate solution technique not depending on small parameters: a special example. int. j. non-linear mech. 30(3), 371–380. liao, s. j. and tan, y. a. (2007). general approach to obtain series solutions of nonlinear differential equations, studies appl. math. 119(4), 297–354. marinca, v. and herisanu, n. (2006). a modified iteration perturbation method for some nonlinear oscillation problems. acta mech. 184, 231–242. momani, s. s. (2004). analytical approximate solutions of non-linear oscillators by the modified decomposition method. int. j. modern. phys. c, 15(7): 967-979. analysis of nonlinear dynamic behaviour of nanobeam resting on winkler and pasternak foundations using variational iteration method issn: 2180-1053 vol. 9 no.2 july – december 2017 39 rafei., m., ganji, d. d., daniali, h. and pashaei, h. (2007). the variational iteration method for nonlinear oscillators with discontinuities. j. sound vib. 305, 614–620. rafiepour, h., tabatabaei, s. h. and abbaspour. m. (2014) a novel approximate analytical method for nonlinear vibration analysis of euler-bernoulli and rayleigh beams on the nonlinear foundation. arab j. sci eng. sobamowo, m. g. (2016). thermal analysis of longitudinal fin with temperature-dependent properties and internal heat generation using galerkin’s method of weighted residual. applied thermal engineering 99 (2016) 1316–1330. wu g. c. (2012). laplace transform overcoming principal drawbacks in application of the variational iteration method to fractional heat equations, thermal science, 16(4): 1257-1261 zhou, j. k. (1986) differential transformation and its applications for electrical circuits. huazhong university press: wuhan, china. nomenclature a area of the structure e young modulus of elasticity i moment of area k1, k2, k3 foundation constants l length of the nanobeam mp mass of the nanobeam n axial/longitudinal force t time ( )u t generalized coordinate of the system w transverse displacement/deflection x axial coordinate σv tangential moment accommodation coefficient ( )x trial/comparison function appendix using euler-bernoulli theory,the governing equation of motion as derived as follows. the bending moment for the euler-bernoulli beam is given as ( , ) xx a m x t z da  (a1) journal of mechanical engineering and technology issn: 2180-1053 vol. 9 no.2 july – december 2017 40 where a, z, σxx is the cross sectional area of the nanotube, distance from the neutral axis and the axial stress on the nanotube, respectively it should be noted that xx xx e  (a2) ɛxx is the axial strain of the nanotube on substituting eq. (a1) into eq. (a2), we have ( , ) xx a m x t ze da  (a3) following von karman strain, we have 2 2xx w z x      (a4) where wis the displacement of the nanotube on substituting eq. (a4) into eq. (a3), we have 2 2 2 ( , ) a w m x t e z da x      (a5) but the second moment of area, 2 a i z da  (a6) therefore, 2 2 ( , ) w m x t ei x     (a7) by incorporating von karman’s nonlinearity, the internal shear force on the structural cross section must satisfy themoment equilibrium relation ( , ) ( , ) m w v x t n x t x x       (a8) it should be pointed out that the internal membrane force, n is constant along the beam as analysis of nonlinear dynamic behaviour of nanobeam resting on winkler and pasternak foundations using variational iteration method issn: 2180-1053 vol. 9 no.2 july – december 2017 41 0 ( , ) ( ) n n x t n t x      (a9) therefore, eq. (a9) becomes ( , ) ( ) m w v x t n t x x       (a10) differentiating eq. (a10) with respect to spatial variable x considering the absence of external axial load on the beam 2 2 2 2 ( ) m v w n t x x x         (a11) using newton’s law, the governing equation of motion for the free vibration of the nanotube can be expressed as 2 2 3 1 32 2p v w w w m k k w k w x t t x               (a12) substituting eq. (a12) into eq. (a11), we have 2 2 2 2 3 1 32 2 2 2 ( ) p m w w w w m k k w k w n t x t t x x                  (a13) for the immovable supports, the internal membrane force is given as 2 0 ( ) 2 lea w n t dx l x         (a14) therefore, eq. (a13) can be expressed 22 2 2 2 3 1 32 2 2 202 l p m w w w ea w w m k k w k w dx x t t x l x x                             (a15) where k1, k3 and kp are the pasternak, linear and nonlinear winkler foundation constants from equ. (a7), we have journal of mechanical engineering and technology issn: 2180-1053 vol. 9 no.2 july – december 2017 42 2 4 2 4 m w ei x x       (a16) if we substitute eq. (a16) into eq. (a15), we obtained the governing equation as motion for the nanotube as 24 2 2 2 3 1 34 2 2 2 0 0 2 l p w w w w ea w w ei m k k w k w dx x t t x l x x                             (a17) issn: 2180-1053 vol. 3 no. 2 july-december 2011 preparation and characterization of untreated waste palm oil/diesel fuel blend 1 preparation and characterization of untreated waste palm oil/diesel fuel blend m. i. ali, s. abdullah,t. i mohamad, w. r. w. daud faculty of engineering and built environment, universiti kebangsaan malaysia, 43600 bangi, selangor, malaysia email: misra63@yahoo.com abstract some diesel engines can run on some kinds of vegetable oil under some conditions without problems. to use vegetable oils in diesel engine without modification, it is necessary to make sure that the vegetable oils properties must be similar to diesel fuel. in this study, palm oil that has been used several times for frying purposes is investigated for the utilization as an alternative fuel for diesel engines. the waste palm oil has a variety of qualities, possess properties different from that of neat oils. higher impurities of the used oils make them different from neat vegetable oil. the high viscosity of the waste palm oil was decreased by blending with diesel. two different previous uses of waste palm oil were blend with diesel. the blends of varying proportions of waste palm oil and diesel were prepared, analyzed and compared with diesel fuel and the waste palm oil ethyl ester. the properties of the blends such as heating value, viscosity, specific gravity, etc. were determined. it was found that blending waste palm oil with diesel reduces the viscosity and different previous uses of waste palm oil significantly affected the properties of the blended fuels. from the properties test results it has been established that blends containing 5 to 40% of waste palm oil in diesel yielded the properties closely matching that of diesel. keywords: alternative fuel; waste frying oil; renewable energy. 1.0 introduction the increasing production of waste frying oils from household and industrial sources is a growing problem all around the world. the united states produces about 3 billion gallons (roughly 11 billion liters) per year of waste frying oil (pugazhvadivu, 2005). in malaysia, the average domestic consumption for cooking oil is 40,000 to 50,000 tonnes per month and even exceeded the normal monthly demand to reach over 70,000 tonnes, especially during festive seasons. if 5% issn: 2180-1053 vol. 3 no. 2 july-december 2011 journal of mechanical engineering and technology 2 of these could be considered as waste frying oil, it might cause big problem to dispose the waste frying oil. nowadays, most of the waste frying oil is poured into the sewer system. this practice contributes to the pollution of rivers, lakes, seas and underground water, which is very harmful for environment and human health. in this situation, the waste frying oil must either be disposed of or recycled in some way. from the viewpoint of sustainability, the used of waste frying oil as fuel offers a plausible means by which it can be recycled. waste frying oil is produced after repeated frying of a variety of foods in vegetable oils and it offers a significant potential as an alternative low-cost fuel. many researchers have shown that vegetable oils (neat or used) can be used as diesel substitutes. several vegetable oils have been evaluated including soybean oil, sunflower oil, cotton-seed oil, peanut, coconut, palm, etc. (ramadhas, et.al., 2004; altin, et.al., 2001; kalam, et.al., 2003; rakopoulus et.al., 2006). most of the researchers found that vegetable oils pose some problems when subjected to prolonged usage in compression ignition engines because of their high viscosity. the common problems are poor atomization, carbon deposits, ring sticking, fuel pump failure, etc. (choo et.al., 2005). the literature also shows that a large amount works have done into evaluating the conversion the waste frying oil to biodiesel. the use of biodiesel as fuel for compression ignition engines has many environmental advantages, however, the production of biodiesel involves the use of a toxic, flammable liquid methanol and caustic compounds like sodium hydroxide or potassium hydroxide. there are some other methods to reduce the viscosity of vegetable oils. fuel blending is one of the methods. it has the advantages of improving the use of vegetable oils with minimal processing and economic. waste frying oil used in industrial or household frying undergo degradation by thermolytic, hydrolytic and oxidative reactions (mangesh, 2006). these process being responsible for changes in the chemical and physical properties, as compared to neat oil. most frying oil reported in the literature used various origin of vegetable oils or waste frying oil and in many cases the waste frying oil are collected after frying a wide variety of meat, fish or vegetable products. the waste frying oil is then blended with other waste oils and being processed for further application. no study has been reported focus on the waste frying oil variation of specific cooking habit and foods. to use waste frying oil in diesel engine without any modification, it is necessary to make sure that the waste frying oil properties must issn: 2180-1053 vol. 3 no. 2 july-december 2011 preparation and characterization of untreated waste palm oil/diesel fuel blend 3 be similar to diesel fuel. in this study, waste frying oil with different previous uses was blend with diesel and compares their respective fuel properties with waste palm oil ethyl ester and malaysian petroleum diesel. the main objective of this study were to decrease the viscosity of waste frying oil by blending with diesel, to examine the influence of different types of food on reduction in viscosities of the blend and other important properties of the fuel. future work will involve test on engine performance and emissions. 2.0 experimental procedure samples of waste frying oil (wfo) were collected from restaurants and from local domestic consumer. all the wfo samples used in this study were from fryer palm oil since most local restaurants and consumer used palm oil. the fatty acid composition of palm oil is dominated by palmitic, oleic, linoleic, and stearic fatty acids plus much less proportions of myristic, lauric, linolenic, and capric acids (alwidyan et.al., 2002). two types of wfo were used in the experiments. the first one was mainly used for frying flour-based food and referred as wfoa. the second one was mainly used for frying chicken or fish and referred as wfob. the wfo utilized in the present study has no additional chemical treatments. the wfo has been filtered to remove food residues and solid precipitate in the oil. to ensure that the oil is clean from water the oil is heated above 100oc to evaporate the moisture. blend of wfo with diesel have been prepared in the laboratory for experiment measurements. different blend ratios have been selected for measurements and evaluation. the blend ratios include 5, 10, 15, 20, 30, 40, 50, 60 and 70 percent by volume of wfo in a mixture of wfo in diesel. they are referred to as diesel, for example 5wfoa-95d means 5% wfoa and 95% diesel. the reference fuel is a petroleum diesel fuel similar to those available in petronas petrol station. the pure diesel fuel was tested to establish the 0% blending point and pure wfo were tested to establish the 100% wfo points. several tests were conducted to characterize the blended fuel, diesel and the pure wfo according to astm standard methods. this is to compare various physical, chemical and thermal properties of the fuels. various procedures followed and the instruments used are given in table 1. issn: 2180-1053 vol. 3 no. 2 july-december 2011 journal of mechanical engineering and technology 4 table 1 astm methods and instrument to measure various properties 3 table 1 astm methods and instrument to measure various properties      property  astm  method  instrument  gross heating value  astm d240  ika werke bomb calorimeter  viscosity  astm d445  haake viscometer  specific gravity  astm d1298  hydrometer pour point  astm d97  pour point apparatus  cloud point  astm d2500  cloud point apparatus  flash point  astm d93  setaflash tester  carbon (%wt)  ‐  elemental analyzer  hydrogen (%wt)  ‐  elemental analyzer  nitrogen (%wt)  ‐  elemental analyzer  oxygen (%wt)  ‐  elemental analyzer sulfur (%wt)   ‐  elemental analyzer  3.0 result and discussion the important chemical and physical properties of the pure wfo were determined by standard methods and compared with diesel and its ethyl ester. the analytical results are shown in table 2. the ester of wfo was found in the literature (al-widyan, 2002). the results show that the gross heating value of the wfoa and wfob are comparable to the diesel fuel and its ethyl ester. the measured heating value of both wfo is approximately 13.5% lower than diesel and 0.09% higher than its ethyl ester. as expected, the viscosity of both wfoa and wfob are significantly higher than those of other types of fuel. the viscosity of the diesel fuel is 3.743 mpa.s at 40oc while the viscosity of both wfo is 9.2-9.8 times that of diesel fuel. blending with other fuel might reduce the viscosity and therefore eliminates this problem. the specific gravity of both wfoa and wfob were also found higher than that of the diesel fuel and varied from 0.904-0.905, while the specific gravity of the ester was significantly lower than that of both wfo. in comparison, the wfo had an average of 7.3% higher specific gravity than that of diesel fuel. both wfo contained approximately 13.5% less heat energy on a mass basis. since the specific gravity of the wfo was 7.3% higher than diesel fuel, the heat energy of the wfo therefore was about 6.2% lower on a volume basis. 3.0 result and discussion the important chemical and physical properties of the pure wfo were determined by standard methods and compared with diesel and its ethyl ester. the analytical results are shown in table 2. the ester of wfo was found in the literature (al-widyan, 2002). the results show that the gross heating value of the wfoa and wfob are comparable to the diesel fuel and its ethyl ester. the measured heating value of both wfo is approximately 13.5% lower than diesel and 0.09% higher than its ethyl ester. as expected, the viscosity of both wfoa and wfob are significantly higher than those of other types of fuel. the viscosity of the diesel fuel is 3.743 mpa.s at 40oc while the viscosity of both wfo is 9.2-9.8 times that of diesel fuel. blending with other fuel might reduce the viscosity and therefore eliminates this problem. the specific gravity of both wfoa and wfob were also found higher than that of the diesel fuel and varied from 0.904-0.905, while the specific gravity of the ester was significantly lower than that of both wfo. in comparison, the wfo had an average of 7.3% higher specific issn: 2180-1053 vol. 3 no. 2 july-december 2011 preparation and characterization of untreated waste palm oil/diesel fuel blend 5 gravity than that of diesel fuel. both wfo contained approximately 13.5% less heat energy on a mass basis. since the specific gravity of the wfo was 7.3% higher than diesel fuel, the heat energy of the wfo therefore was about 6.2% lower on a volume basis. the pour point and cloud point of both wfo were found higher than diesel by 4oc and 6oc, respectively. higher pour point and cloud point reflect unsuitability of wfo as diesel fuel in cold climate conditions. the flash points of both wfo were also found quite high compared to diesel. hence the wfo is extremely safe to handle. carbon, hydrogen, nitrogen, sulfur and oxygen (chnso) content were also measured for diesel and wfo. it has been observed that the value of carbon and hydrogen content of pure wfo are much lower compared to the values of diesel. thus the carbon/hydrogen ratio, which depends upon the degree of unsaturated of wfo, is different from diesel. presence of nitrogen in the fuel might contribute nox emissions and presence of oxygen in the fuel will improves the combustion properties and emissions but reduces the heating value of the fuels. the sulfur contents for the wfo are very low and could not be detected and it is assumed zero. low sulfur content of both wfo will results in lower sox emissions. table 2 physical and chemical properties of fuels 4 the pour point and cloud point of both wfo were found higher than diesel by 4oc and 6oc, respectively. higher pour point and cloud point reflect unsuitability of wfo as diesel fuel in cold climate conditions. the flash points of both wfo were also found quite high compared to diesel. hence the wfo is extremely safe to handle. carbon, hydrogen, nitrogen, sulfur and oxygen (chnso) content were also measured for diesel and wfo. it has been observed that the value of carbon and hydrogen content of pure wfo are much lower compared to the values of diesel. thus the carbon/hydrogen ratio, which depends upon the degree of unsaturated of wfo, is different from diesel. presence of nitrogen in the fuel might contribute nox emissions and presence of oxygen in the fuel will improves the combustion properties and emissions but reduces the heating value of the fuels. the sulfur contents for the wfo are very low and could not be detected and it is assumed zero. low sulfur content of both wfo will results in lower sox emissions. table 2 physical and chemical properties of fuels property  diesel  wfoa  wfob  ethyl  ester  gross hv (kj/kg)  45609  39340  39349  39305  viscosity (mpa.s)  3.743  36.700  34.700  13.05  specific gravity  0.838  0.905  0.904  0.874  pour point ( oc)  8  12  12  0  cloud point ( oc)  15  21  21  0  flash point ( oc))  84.8  >124.0  >124.1  109  carbon (%wt)  84.10  72.96  72.41  na  hydrogen (%wt)  12.80  11.82  12.15  na  nitrogen (%wt)  0.30  0.49  0.45  na  oxygen (%wt)  2.61  14.73  14.99  na  sulfur (%wt)  0.19  0  0  na  na indicates not available tests were also conducted using blends of wfo with petroleum diesel. figure 1 shows the gross heating value (ghv) of various blends of the wfo. ghv of wfo is lower compared to that of diesel, therefore it was observed that the ghv of the blends were decreased with the increasing percentage of wfo in the blends. the blends maintained a similar trend for both wfo, but the ghv were slightly higher for wfob compared to wfoa. na indicates not available tests were also conducted using blends of wfo with petroleum diesel. figure 1 shows the gross heating value (ghv) of various blends of the wfo. ghv of wfo is lower compared to that of diesel, therefore it was observed that the ghv of the blends were decreased with the issn: 2180-1053 vol. 3 no. 2 july-december 2011 journal of mechanical engineering and technology 6 increasing percentage of wfo in the blends. the blends maintained a similar trend for both wfo, but the ghv were slightly higher for wfob compared to wfoa. figure1 effect of blends on gross heating value 39000 40000 41000 42000 43000 44000 45000 46000 0 10 20 30 40 50 60 70 80 90 100 wfo p ercentage in the fuel blend (%) g ro ss h ea tin g va lu e (k j/ kg ) wfoa wfob figure 1: effect of blends on gross heating value in general, the ghv of the blends are lower than that of diesel. the ghv is expected to affect the fuel consumptions of the engine. higher viscosity is a major problem in using waste frying oil as fuel in compression ignition engines. the viscosity of various blends of wfo and diesel were evaluated at 40oc. it shows that when wfo mixed with diesel to make blends, it is found that the more percentage of wfo in the blends, the higher of the viscosity. this happens to both of wfo, but wfob gives better reduction in viscosity compared to wfoa at any percentage of blends. for example, a reduction of 87.3% and 88.4% for wfoa and wfob was achieved with 5wfoa-95d and 5wfob-95d, respectively. the injection system of a diesel engine is designed to operate with diesel fuels, which has a viscosity of 3 – 8 mpa.s (bari et.al, 2002) and the astm limit for viscosity of diesel fuel is 5 cst. (deepak, 2007). it can be seen that, the viscosity of both wfo up to 40% blend ratio can be used as diesel substitute as the viscosity are within the design limit. it is always desirable for the wfo to have a viscosity value nearer to that of diesel. this is because the higher the viscosity, the more difficult it is to atomize for the wfo and its blends. blending of wfo with diesel seem to be an effective tool to overcome engine problems associated with the high viscosity of wfo. the variation of the specific gravity with various wfo/diesel blends were also evaluated and is shown in figure 3. it has been observed that up to 40% wfo/diesel blends for both wfo produce slightly higher than diesel. this however, is not important, as this would only cause figure 1 effect of blends on gross heating value issn: 2180-1053 vol. 3 no. 2 july-december 2011 preparation and characterization of untreated waste palm oil/diesel fuel blend 7 a slight increase of fuel consumption. the specific gravity of wfoa is found higher than wfob at all level of blends. more dense fuel will provide greater energy per gallon of fuel. figure 2 effect of blending on viscosity 0 5 10 15 20 25 30 35 40 0 10 20 30 40 50 60 70 80 90 100 wco p ercentage in the fuel blend (%) v is co si ty (m pa .s ) wcoa wcob figure 3 effect of blending on specific gravity 0.83 0.84 0.85 0.86 0.87 0.88 0.89 0.90 0.91 0 10 20 30 40 50 60 70 80 90 100 wfo percentage in the fuel blend (%) sp ec if ic g ra vi ty wfoa wfob effect of blending on pour point is shown in figure 4. pour point is defined as the lower temperature that the fuel can be poured by gravity. as the wfo concentration in the blend was increased, the pour point was increased. wfob provide slightly lower cold flow properties than wfoa. the wfo/diesel blends were also evaluated for their cloud points. cloud point is the temperature of which a cloud of wax crystal first appears in the fuel when it is cooled. the effect of blending on cloud figure 2 effect of blending on viscosity figure 3 effect of blending on specific gravity issn: 2180-1053 vol. 3 no. 2 july-december 2011 journal of mechanical engineering and technology 8 point is shown in figure 5. both wfo exhibits cloud points ranging from 15 to 19oc for the blending ratio from 5 to 60% of wfo. the differences of cloud point between wfoa and wfob are very small at all percentage in the fuel blend. a key property determining the flammability of a fuel is the flash point. it is a measure of the tendency of a sample to form a flammable mixture with air. the flash point temperature is critical from a safety viewpoint. the flash point must be as high as practical. blending of wfo with diesel reduces the value of the flash point of the blend. however, the flash point of the different blend ratio for both wfo is relatively higher than that of diesel. as can be observed in figure 6, wfoa exhibit greater flash points than wfob at all blends. figure 4 effect of blending on pour point 4 5 6 7 8 9 10 11 12 13 0 10 20 30 40 50 60 70 80 90 100 wfo percentage in the fuel blend (%) po ur p oi nt (o c ) wfoa wfob figure 5 effect of blending on cloud point 10 12 14 16 18 20 22 0 10 20 30 40 50 60 70 80 90 100 wfo percentage in the fuel blend (%) c lo ud p oi nt (o c ) wfoa wfob figure 4 effect of blending on pour point figure 5 effect of blending on cloud point issn: 2180-1053 vol. 3 no. 2 july-december 2011 preparation and characterization of untreated waste palm oil/diesel fuel blend 9 figure 6 effect of blending on flash point 80 90 100 110 120 130 0 10 20 30 40 50 60 70 80 90 100 wfo percentage in the fuel blend (%) fl as h po in t ( oc ) wfoa wfob blending of wfo with diesel is also found reduces the value of carbon and hydrogen content in the blends (figure 7 and figure 8). this is expected to the reduction in co2 emissions for the blended fuels because of the lower amount of carbon content in the fuel as shown in fuel properties of wfo/diesel blends. figure 7 effect of blending on carbon content 70 72 74 76 78 80 82 84 86 0 10 20 30 40 50 60 70 80 90 100 wfo percentage in the fuel blend (%) c ar bo n co nt en t ( % w t) wfoa wfob figure 6 effect of blending on flash point figure 7 effect of blending on carbon content issn: 2180-1053 vol. 3 no. 2 july-december 2011 journal of mechanical engineering and technology 10 figure 8 effect of blending on hydrogen content 11.00 11.20 11.40 11.60 11.80 12.00 12.20 12.40 12.60 12.80 13.00 0 10 20 30 40 50 60 70 80 90 100 wfo percentage in the fuel blend (%) h yd ro ge n co nt en t ( % w t) wfoa wfob 4.0 conclusion the experimental results show that the use of wfo/diesel blend is possible. blending of wfo with diesel seems to be an effective technique to decrease the viscosity of the wfo. reasonable viscosity values have been obtained using blending ratios as high as 40%wfo and 60% diesel. other fuel properties such as heating value, specific gravity and flash point are comparable. the blends containing 5-40% of wfo yielded the properties closely matching that of diesel. different previous uses of wfo influence the properties of the fuel blends. this may due to changes of the molecular structure of the wfo as well as the nutrients, aromatics or other compounds that may have been introduced during frying. in general, wfob gives better properties and comparable to that of diesel. wfo/diesel blend possessing good fuel properties. since they are from palm oil, they are environment friendly, biodegradable and renewable. wfo may be preferred choice in terms of cost, since it is essentially a waste product and it is cheaper than unused oils. the above observations indicate a good potential of using wfo/ diesel blends as an alternative compression ignition engine fuel and encourage continuation of the present experiment program. figure 8 effect of blending on hydrogen content issn: 2180-1053 vol. 3 no. 2 july-december 2011 preparation and characterization of untreated waste palm oil/diesel fuel blend 11 5.0 acknowledgement the authors would like to thank department of chemical and process engineering, universiti kebangsaan malaysia for kindly providing the facilities and extending necessary help to conduct the tests for the fuels and various blends under study. the authors also wish to express their appreciation for the funding from the research grant ukm-gupbtt-07-25-024 which supported this study. 6.0 references pugazhvadivu. m., jeyachandran, k. 2005. investigations on the performance and exhaust emissions of a diesel engine using preheated waste frying oil as fuel. renewable energy. 30: 2189-2200. ramadhas, a.s., jayaraj, s., muraleedharan, c. 2004. use of vegetables oil as i.c engine fuels – a review. renewable energy. 29: 727-742. altin, r., cetinkaya, s., yucesu, h.s. 2001. the potential of using vegetables oil fuels for diesel engines. energy conversion and management. 42: 529-538. kalam, m.a., husnawan, m., masjuki, h.h. 2003. exhaust emission and combustion evaluation of coconut oil-powered indirect injection diesel engine. renewable energy. 28: 2405-2415. rakopoulos, c.d., antonopoulos, k.a., rakopoulos, d.c., hountalas, d.t., giakoumis, e.g. 2006. comparative performance and emissions study of a direct injection diesel engine using blends of diesel fuel with vegetable oils or bio-diesels of various origin. energy conversion and management. 47: 3272-3287. choo, y.m., yung, c.l., cheng, s.f., ma, a.n., chuah, c.h., basiron, y. 2005. key fuel properties of palm oil alkyl esters. fuel. 84: 1717-1720. mangesh, g. k., ajay, k. d. 2006. waste cooking oil – an economical source for biodiesel : a review. ind. eng. chem. res. 45: 2901-2913 praminik, k. 2003. properties and use of jatropha curcas oil and diesel fuel blends in compression ignition engine. renewable energy. 28: 239248. al-widyan, m., tashtoush, g., abu-qudais, m. 2002. utilization of ethyl ester of waste vegetable oils as fuels in diesel engines. fuel processing technology. 76: 91-103. . al-widyan, m., al-shyoukh, a.o. 2002. experimental evaluation of the transesterification of waste palm oil into biodiesel. bioresource technology. 85: 253-256. issn: 2180-1053 vol. 3 no. 2 july-december 2011 journal of mechanical engineering and technology 12 bari, s., lim, t.h., yu, c.w. 2002. effects of preheating of crude palm oil (cpo) on injection system, performance and emission of a diesel engine. renewable energy. 27: 339-351. deepak, a., avinash kumar, a. 2007. performance and emissions characteristics of jatropha oil (preheated and blends) in a direct injection compression ignition engine. applied thermal engineering. article in press. 01(1-12).pdf issn: 2180-1053 vol. 9 no.2 july – december 2017 71 hybrid position and vibration control of nonlinear crane system n. m. tahir 1, 2* , a. g. ibrahim 1 , h. liman 1 1 faculty of engineering, abu bakar tafawa balewa university, pmb 0248, bauchi, nigeria 2 faculty of electrical engineering, universiti teknologi malaysia, 81310 utm, johor, abstract this paper presents comparative assessments of input shaping techniques using two different approaches, for sway reduction of cranes system. first, the shaper was designed at maximum load hoisting length while the second was designed at average load hoisting length. these were accomplished using curve fitting toolbox in matlab. in both case; zero vibration (zv), zero vibration derivative (zvd) and zero vibration derivative derivatives (zvdd) were designed. average hoisting length (ahl) shapers performed better than the maximum hoisting length (mhl) shapers. proportional integral derivative (pid) was incorporated for position control. after successful implementation, simulation results show that a precise payload positioning was achieved. ahlzvdd has superior performances in sway reduction and robustness. keywords: crane system; hoisting length; sway reduction; zero vibration derivative shaper 1.0 introduction developments in large-scale manufacturing have seen the rise of crane systems deployed in industries. other sectors of the economy such as transportation have also benefitted from the use of cranes. because of the huge importance attached to crane use in industries, it has become necessary to have the cranes operating with high speed, and minimum sway (masoud et al, 2001). high-speed operation of cranes results in unwanted motions such as swinging, bouncing and twisting. it also has serious safety concerns and could affect the rate of production due to unwanted downtime and inaccurate positioning of the payload (singhose, 2009). because of the aforementioned challenges, it has become necessary to apply techniques that could limit these unwanted sways in the crane system. a lot of researchers have focused on the control of oscillations in crane systems. ha & kang (2013) stated that this can be categorized into two; feedforward control and feedback control. it also inferred that feedforward techniques can be employed to cancel system oscillations while feedback control can be used to achieve precision in load positioning. it can also be used to reduce unwanted oscillations. bartulovi & zu (2014) showed that although a combination of the two control strategies could result in a more *corresponding author e-mail: nuratahir85@gmail.com journal of mechanical engineering and technology issn: 2180-1053 vol. 9 no.2 july – december 2017 72 efficient system, the feedforward control helps in reducing the complexity and cost of feedback control. several attempts have been made to design and implement control strategies that could minimize or eliminate these unwanted motions so that the safety and efficiency of the crane systems can be guaranteed. numerous techniques ranging from classical control to modern control have been presented. uchiyama et al (2013) proposed residual load sway suppression using open loop control. similarly, ahmad et al (2009) compared the performance of feedforward input shaping and low pass filtering (lpf) for anti-sway control, they discovered that input shaping was more robust compared to lpf for erroneous natural frequency. mohamed et al (2015) used command shaping technique to reduce vibrations in a single link flexible manipulator. this technique results in the suppression of oscillatory response. in addition, maleki & singhose (2010), ahmad et al (2009) and sorensen et al (2007) have shown that input shapers can be applied to crane systems to reduce oscillations. vaughan et al (2008) and schaper et al (2013) showed that, although the use of open loop controllers alone makes the crane systems vulnerable to external disturbances, a combination of feedforward and feedback control can result in an efficient control system. terashima et al (2007) proposed control of rotary crane using the straight transfer transformation method (stt), for sway and position control of the system. le et al (2013) presented a partial feedback linearization (pfl) and adaptive sliding mode control (smc) for sway suppression of a rotary crane in a situation of inaccurate model or poor parameter representation. though it is simple to design and implement, pfl is highly affected by parameter variations. bartolini et al (2002), tuan & lee (2013) and tuan et al (2013) also presented sliding mode control for position and sway reductions of the crane system. but tai & andrew (2015) showed that smc is not very popular due to the fact that it dissipates a lot of energy which leads to system burn-out. furthermore, nakazono et al (2008) presented three-layered neural networks with genetic algorithm for vibration control of the rotary crane. ahmad et al (2010) carried out performance investigation of sway control using lqr and pd-type fuzzy controller in the presence of a disturbance. al-mousa & pratt (2000) presented a combined fuzzy logic and a delayed feedback controller for oscillation reduction of the rotary crane. 2.0 crane dynamics a crane system is a machine designed for the transportation of heavy and large amounts of a load from one position to another. it is mostly used in construction sites and large industries. as shown in figure.1, a laboratory scale 3d system consists of three parts namely; the cart, the rail, and the payload, giving three directions of motion. hybrid position and vibration control of nonlinear crane system issn: 2180-1053 vol. 9 no.2 july – december 2017 73 figure 1. laboratory scale 3d crane system that is, the cart moving along x-axis, the rail moving along y-axis and payload hoisting in z-axis. the schematic diagram of the 3d crane system is shown in figure 2. xyz represents the coordinates; other system parameters are defined as; (maghsoudi et al, 2014). α angle of lift-line with y-axis β angle between – z and projection of the payload onto the xz plane figure 2. schematic diagram and forces journal of mechanical engineering and technology issn: 2180-1053 vol. 9 no.2 july – december 2017 74 t reaction force in the payload cable acting on the cart fx, fy forces driving the rail and cart respectively fz force lifting the payload fx, fy, fz corresponding frictional forces by definition, rt p t p mm m m m   21 ,  p z rt y t x m f u mm f u m f u    321 ,, p z rt y t x m f f mm f f m f f    321 ,, 1 1 1 2 2 2 3 3 3 , ,k u f k u f k u f      where; and are the payload mass, trolley mass (including gearbox, encoders, and dc motor) and moving rail respectively and l represents the length of the lift-line. the dynamic equations of motion of the crane can be obtained as given in (maghsoudi et al, 2014).  sinsin 322 kkx t  (1)  cos 311 kky t  (2)   cossin)2(sincos)2( coscos2sinsin)( 22   llll llllxx tp   (3)  sin)2(cos)( 2  llllyy tp  (4)   sinsin)2(coscos)2( sincos2cossin)( 22   llll llllz p   (5) where, p x , p y and p z are position of payload in x, y and z axes respectively. t x and t y are positions of trolley in x and y axes. dots represent derivative of the respective quantities. p m tm rm hybrid position and vibration control of nonlinear crane system issn: 2180-1053 vol. 9 no.2 july – december 2017 75 table 1 shows the parameters used for simulation and experiment which correspond to the crane system (maghsoudi et al, 2014). table 1. system parameters variables values length of cable, l 0.72 m trolley mass, t m 1.155 kg payload mass, p m 1 kg mass of rail, r m 2.2 kg frictional forces, zyx fff ,, 100, 82, 75 ns/m acceleration due to gravity, g 9.8 m/s 3.0 hybrid controller design in this section, ahl-shapers and mhl-shapers ware designed. figure 3 shows the block diagram of the control schemes, where 𝑥 is the trolley position and 𝜃 is the sway angle. zv, zvd, and zvdd are designed in each situation to suppress payload sways and pid was incorporated for trolley position control. comparative assessments are presented. 3.1 input shaping in this section, zv, zvd and zvdd ware designed. using curve fitting toolbox in the matlab software to determine the damping ratio and natural frequency of the system, with the input and output data obtained from the nonlinear crane system. the shapers ware designed using different approaches that is, taking data at average load hoisting length and at maximum load hoisting length. figure.4 shows an input shaping process while the parameters of this design are as recorded in table 2 and table 3. figure 3. block diagram of hybrid control input input shaper pid controller crane system + 𝑥 𝑥 𝜃 𝜃 journal of mechanical engineering and technology issn: 2180-1053 vol. 9 no.2 july – december 2017 76 the crane system was considered as 2 nd order under-damped system as in (ha & kang, 2013). 2 2 2 ( ) 2 g s s s       (6) in which, and  is the damping ratio and natural frequency of the system respectively. the system response was expressed in time domain as ( maleki & singhose, 2010).     0( ) 20 2 ( ) sin ( ) 1 1 t ta y t e t t           (7) where o t and a are the time instant and amplitude of the impulse. the response to an impulse sequence was obtained using superposition as;     ( ) 2 2 1 ( ) sin ( ) 1 1 n i n t ti n i i a y t e t t                    (8) the residual vibration amplitude was obtained, with the following trigonometric.   1 sin sin( ) n i i i b t a t        (9) in which;     2 2 1 1 cos sin n n i i i i i i a b b                   (10) comparing the (8) and (9), we got   ( ) 2 1 n it ti n i a b e       (11) * a1 0 a 2 a3 t 2 t 3 0 unshaped input input shaper shaped input a t 2 t 3 a 0 figure 4. input shaping process hybrid position and vibration control of nonlinear crane system issn: 2180-1053 vol. 9 no.2 july – december 2017 77 rearranging (10) and (11) gives (singhose, 2009).   2 2 1 2 2 1 nta e r r       (12) where;   21 1 sin 1i n t i i i r a e t         22 1 cos 1i n t i i i r a e t       at t=0, the residual oscillation amplitude from a unity magnitude was obtained as in (blackburn et al, 2010).  21 a      (13) in addition, the percentage residual vibration was obtained by, dividing (12) by (13) as; ( ) 2 2 1 2 nt a r e r r a      (14) the zero vibration (zv) constraint after the last impulse can be obtained by setting r1 and r2 of (14) to zero. the impulse amplitudes should be one hence the summation constraints are as (ahmad et al, 2009). 1 1 n i i a   (15) however, the first impulse time instant is set as, 1 0t  thus, the zv parameters are obtained using its constraints by solving (14) and (15) as; 1 1 1 0 i i d k a k k t                 (16) in which  21 d       and  21 k e      however, to increased robustness to frequency errors, r1, and r2 derivatives are set to zero as; 1 0 i i r     and 2 0 i i r     (17) journal of mechanical engineering and technology issn: 2180-1053 vol. 9 no.2 july – december 2017 78 hence, the constraints equations (14), (15) and (17) are solved to obtain the zvd parameters as; 2 2 2 2 1 2 (1 ) (1 ) (1 ) 0 2 i i d d k k a k k k t                     (18) also, by using the second derivative of (17) and solving the constraints equations, the zvd parameters were obtained as; 2 3 3 3 3 3 1 3 3 (1 ) (1 ) (1 ) (1 ) 0 2 3 i i d d d k k k a k k k k t                       (19) equation (16), (18) and (19) were used to calculate the shapers parameters table 2. shapers parameters table 3. shapers parameters 𝑙(𝑚) 𝑤𝑛 (𝑟𝑎𝑑 /𝑠) 𝜁 1a 2 a 3a 4a 1t 2t 3t 4t mhlzv 0.72 3.73 0.006 0.5047 0.4953 0 0 0 0.8423 0 0 mhlzvd 0.72 3.73 0.006 0.2547 0.5192 0.2453 0 0 0.8423 1.6846 0 mhlzvdd 072 3.73 0.006 0.1286 0.3785 0.3714 0.125 0 0.8423 1.6846 2.5269 𝑙(𝑚) 𝑤𝑛(𝑟𝑎𝑑 /𝑠) 𝜁 1a 2 a 3a 4a 1t 2t 3t 4t ahlzv 0.47 4.57 0.008 0.5063 0.4937 0 0 0 0.6874 0 0 ahlzvd 0.47 4.57 0.008 0.2563 0.4999 0.2438 0 0 0.6874 1.3748 0 ahlzvdd 0.47 4.57 0.008 0.1298 0.3769 0.3702 0.1204 0 0.6874 1.3748 2.0622 hybrid position and vibration control of nonlinear crane system issn: 2180-1053 vol. 9 no.2 july – december 2017 79 4.0 results and discussions in this section, the nonlinear crane system was simulated using step input to assess the ahl-shapers and mhl-shapers performance in vibrations suppression. pid incorporated with both ahl-shapers and mhl-shapers are also assessed for setpoint tracking. a mean absolute error was employed as the performance index, is a very good measure of average error, level of sway reduction was measured and compared for various shapers. this is accomplished using the mean absolute errors of the shaped and unshaped responses. also, the time response of the control algorithms was discussed and analyzed. the simulation results are presented and compared in this section. 4.1 simulation result of nonlinear crane system with mhl-shapers the simulation results of nonlinear crane system with mhl-shapers are as shown in figure.5.mhl-zv, mhl-zvd, and mhl-zvdd are presented and their results ware compared. as shown in table 4, mhl-zvdd has shown a better performance in sways suppression, based on the percentage of sway reductions using mean absolute error. 4.2 simulation result of nonlinear crane system with ahl-shapers the simulation results of nonlinear crane system with ahl-shapers are as shown in figure.6.ahl-zv, ahl-zvd, and ahl-zvdd are also designed and presented, their performance is compared and assessed. it was observed that the performance of the shapers in sway reduction increased as the order of the derivatives increased as shown in table 4. 4.3 simulation result comparing ahl-shapers with mhl-shapers. the performance of shapers from section 4.1 and 4.2 above ware compared and evaluated. these were presented in figure.7, 8 and 9. it was observed that ahl-shapers shown a superior performance as compared with mhl-shapers, this was as recorded in table 4. 4.4 simulation results using pid with the shapers incorporated the nonlinear crane system was also simulated using pid with ahl-shapers and pid with mhl-shapers, for both set point tracking and vibrations control. figure 10 and 11 show the performances of the hybrid algorithms, using the pid gains of kp=2.21, ki=2.01, and kd=0.53. from the two figures, it was observed that a good setpoint tracking was achieved. but increased in the derivative order of the shapers, increases delay in the system as shown in table 5. journal of mechanical engineering and technology issn: 2180-1053 vol. 9 no.2 july – december 2017 80 figure 6. trolley sways with average payload hoisting length 0 2 4 6 8 10 -0.1 -0.08 -0.06 -0.04 -0.02 0 0.02 0.04 0.06 0.08 0.1 time(s) s w a y (r a d ) mhl-zv mhl-zvd mhl-zvdd unshaped figure 5. trolley sways with maximum payload hoisting length 0 2 4 6 8 10 -0.1 -0.08 -0.06 -0.04 -0.02 0 0.02 0.04 0.06 0.08 0.1 time(s) s w a y (r a d ) ahl-zv ahl-zvd ahl-zvdd unshaped hybrid position and vibration control of nonlinear crane system issn: 2180-1053 vol. 9 no.2 july – december 2017 81 figure 7. trolley sways comparison using ahl-zv and mhl-zv figure 8. trolley sways comparison using ahl-zvd and mhl-zvd 0 2 4 6 8 10 -0.1 -0.08 -0.06 -0.04 -0.02 0 0.02 0.04 0.06 0.08 0.1 time(s) s w a y (r a d ) ahl-zvd mhl-zvd unshaped 0 2 4 6 8 10 -0.1 -0.08 -0.06 -0.04 -0.02 0 0.02 0.04 0.06 0.08 0.1 time(s) s w a y (r a d ) ahl-zv mhl-zv unshaped journal of mechanical engineering and technology issn: 2180-1053 vol. 9 no.2 july – december 2017 82 figure 9. trolley sway using comparison using ahl-zvdd and mhl-zvdd figure 10. trolley position with maximum payload hoisting 0 2 4 6 8 10 -0.1 -0.08 -0.06 -0.04 -0.02 0 0.02 0.04 0.06 0.08 0.1 time(s) s w a y (r a d ) ahl-zvdd mhl-zvdd unshaped 0 2 4 6 8 10 0 0.2 0.4 0.6 0.8 1 1.2 1.4 time(s) t ro ll e y p o s it io n (m ) pid-mhlzvdd pid-zvd pid-mhlzv hybrid position and vibration control of nonlinear crane system issn: 2180-1053 vol. 9 no.2 july – december 2017 83 figure 11. trolley position with average payload hoisting table 4. level of sway reduction shaper percentage of sway reduction ahl mhl zv 84.2% 78.1% zvd 87.3% 83.4% zvdd 89.6% 86.2% table 5. response time specifications shaper settling time ahl mhl zv 1.85sec 2.20sec zvd 3.00sec 3.30sec zvdd 4.35sec 4.50sec 0 2 4 6 8 10 0 0.2 0.4 0.6 0.8 1 1.2 1.4 time(s) t ro ll e y p o s it io n (m ) pid-ahlzvdd pid-ahlzvd pid-ahlzv journal of mechanical engineering and technology issn: 2180-1053 vol. 9 no.2 july – december 2017 84 5.0 conclusions this paper has presented hybrid control of nonlinear crane system. ahl-shapers and mhl-shapers, ware designed to suppress vibrations while pid was incorporated for set point tracking control of the nonlinear crane system. the performances of the shapers are compared and matlab simulation results show that ahl-shapers outperformed the mhl-shapers in vibrations reduction. it was also observed that the higher the derivatives the better in vibrations suppression but the more the delay in the system. the control schemes have shown very good performances in payload positioning with load hoisting of crane system. 6.0 acknowledgments the authors gratefully acknowledged ministry of science, technology, and innovation (project no. 03-01-06-sf1213) and utm (vote no. 4s103) for the financial support through science fund research grant. 7.0 references masoud, z.n.,nayfeh, a.h abdel-rahman e.m (2001). dynamics and control of cranes : a review.journal of vibration and control,vol.9 pp. 863–908,. singhose, w, (2009). command shaping for flexible systems: a review of the first 50 years. international journal of precision engineering and manufacturing, 10(4), pp. 153-168 ha, m. and kang, c., (2013). experimental analysis of natural frequency error to residual vibration in zv, zvd, and zvdd shapers,"10 th international conference on ubiquitous robots and ambient intelligence jeju south korea, pp. 195–199. bartulovi, m. and zuzic, g., (2014).ˇnonlinear predictive control of a tower crane using reference shaping approach,”16 th international conference of power electronic and motion control,antalya turkey. no. 6, pp. 872–876, al-mousa, a., (2015). delayed position-feedback controller for the reduction of payload pendulations of rotary cranes,” j vib. control vol. 9, pp. 257–277. uchiyama, n., ouyang, h. and sano, s. (2013). simple rotary crane dynamics modeling and open-loop control for residual load sway suppression by only horizontal boom motion,” mechatronics, vol. 23, no. 8, pp. 1223–1236. ahmad, m. a., raja ismail, r. m. t., ramli, m. s., zakaria, n. f., and abd ghani, n. m., (2009). robust feed-forward schemes for anti-sway control of rotary crane,” cssim 2009 1st int. conf. comput. intell. model. simul., pp. 17–22. hybrid position and vibration control of nonlinear crane system issn: 2180-1053 vol. 9 no.2 july – december 2017 85 mohamed, z., chee, a. k., hashim, a. w. i. m., tokhi, m. o., amin, s. h. m.,mamat, r.,(2015).techniques for vibration control of a flexible robot manipulator,” robotica, vol. 24, no. january 2006, pp. 499–511. maleki e. and singhose, w.,(2010). dynamics and zero vibration input shaping control of a small-scale boom crane,”2010 american control conference,baltimore, usa. pp. 2296–2301. ahmad, m.a., mohd, r., raja, t. and ramli, m. s.(2009). input shaping techniques for anti-sway control of a 3-d gantry crane system, proc. ieee int. conf. mechatronics autom. chang. china, pp. 2876–2881, august 2009. sorensen, k. l., w. s. ã, and dickerson, s. (2007). a controller enabling precise positioning and sway reduction in bridge and gantry cranes,” control eng. pract. 15 825–837, vol. 15, pp. 825–837. vaughan, j., yano, a., and singhose, w. (2008). performance comparison of robust negative input shapers, proc. am. control conf., vol. 00, pp. 3257–3262. schaper, u., arnold, e., sawodny, o., and schneider k.(2013). constrained real-time model-predictive reference trajectory planning for rotary cranes,” 2013 ieee/asme int. conf. adv. intell. mechatronics hum. wellbeing, aim 2013, pp. 680–685. terashima, k., shen, y., and yano k. (2007). modeling and optimal control of a rotary crane using the straight transfer transformation method,” control eng. pract., vol. 15, no. 9, pp. 1179–1192. le, t. a., dang, v., ko, d. h., and an, t. n. (2013). nonlinear controls of a rotating tower crane in conjunction with trolley motion," journal of system and control engineering, vol. 227, no. 5, pp. 451–460. bartolini, g., pisano, a., and usai, e. (2002). second-order sliding-mode control of container cranes,” automatica, vol. 38, no. 10, pp. 1783–1790. tuan l. a. and lee, s. (2013). sliding mode controls of double-pendulum crane systems†,” journal of mechanical s cience and technology,vol. 27, no. 6, pp. 1863–1873. tuan, l. a., moon, s., lee, w. g., and lee, s. (2013). adaptive sliding mode control of overhead cranes with varying cable length †,” journal of mechanical science and technology, vol. 27, no. 3, pp. 885–893. tai, c. and andrew, k. (2015). review of control and sensor system of flexible manipulator,” journal of intelligent and robotic system, vol. 77 (1), pp. 187–213. nakazono, k., ohnishi, k., kinjo, h., and yamamoto, t. (2008). vibration control of load for rotary crane system using neural network with ga-based training,” artif. life robot., vol. 13, no. 1, pp. 98–101, journal of mechanical engineering and technology issn: 2180-1053 vol. 9 no.2 july – december 2017 86 ahmad, m. a., samin, r. e., and zawawi, m. a. (2010). comparison of optimal and intelligent sway control for a lab-scale rotary crane system,” 2010 second int. conf. comput. eng. appl., pp. 229–234. al-mousa, a. a., and pratt, t. (2000). control of rotary cranes using fuzzy logic and time-delayed position feedback control,”http:/hdl.handle.net/10919/36024. maghsoudi, m. j., mohamed, z., husain, a. r., and jaafar, h. i. (2014). improved input shaping technique for a nonlinear system,” 2014 ieee international conference on control system, computing and engineering, 28 30 november 2014, penang, malaysia blackburn, d., singhose, w., kitchen, j., patrangenaru, v., lawrence, j., and kamoi, t. (2010). command shaping for nonlinear crane dynamics,” j. vib. control. issn: 2180-1053 vol. 3 no. 2 july-december 2011 development of a stand-alone solar powered bus stop 55 development of a stand-alone solar powered bus stop mohd afzanizam mohd rosli1, mohd zaid akop 2, muhd ridzuan mansor3, sivarao s.4 1,2,3faculty of mechanical engineering, universiti teknikal malaysia melaka, 76100 durian tunggal, melaka. 4faculty of manufacturing engineering, universiti teknikal malaysia melaka, 76100 durian tunggal, melaka. email: 1afzanizam@utem.edu.my abstract this paper presents the development of a stand-alone solar photovoltaic (pv) system for bus stop at universiti teknikal malaysia melaka, malaysia. the design intent for the bus stop was to provide lighting and information to the bus stop users using reliable renewable energy system as well as to promote green technology awareness to the university residences. the stand-alone pv system was designed to power two units of cfl lamps and an led display unit installed at the bus stop. five units of polycrystalline photovoltaic modules with 110w rating each and four deep cycle battery units were utilized to provide three days of autonomy period for system operation. a part from that, 15 degree of tilt angle was selected for pv module placement to provide optimum energy generation as well as self cleaning for the modules. after the bus stop structure construction, the pv system was installed and commissioned. final results from the commissioning process showed that the system is able to operate successfully as per design requirement. keywords: solar photovoltaic, bus stop, stand-alone system development of a stand-alone solar powered bus stop mohd afzanizam mohd rosli1, mohd zaid akop 2, muhd ridzuan mansor3, sivarao s.4 1,2,3faculty of mechanical engineering, universiti teknikal malaysia melaka, 76100 durian tunggal, melaka. 4faculty of manufacturing engineering, universiti teknikal malaysia melaka, 76100 durian tunggal, melaka. email: 1afzanizam@utem.edu.my abstract this paper presents the development of a stand-alone solar photovoltaic (pv) system for bus stop at universiti teknikal malaysia melaka, malaysia. the design intent for the bus stop was to provide lighting and information to the bus stop users using reliable renewable energy system as well as to promote green technology awareness to the university residences. the stand-alone pv system was designed to power two units of cfl lamps and an led display unit installed at the bus stop. five units of polycrystalline photovoltaic modules with 110w rating each and four deep cycle battery units were utilized to provide three days of autonomy period for system operation. a part from that, 15 degree of tilt angle was selected for pv module placement to provide optimum energy generation as well as self cleaning for the modules. after the bus stop structure construction, the pv system was installed and commissioned. final results from the commissioning process showed that the system is able to operate successfully as per design requirement. keywords: solar photovoltaic, bus stop, stand-alone system 1.0 introduction photovoltaic or pv is currently one of the most attractive options for renewable energy resources in the world. malaysia which is blessed with yearly average solar irradiance of 1400 to 1900 kwh/m2 is considered to be in a very advantageous position to harness the unlimited energy towards pv applications to cater current domestic energy demand (malaysian pv handbook, 2009). the technology offers free and renewable energy, no emission effects during energy production, help to reduce dependency to conventional power supply, as well as easy to install and maintain due to its modular characteristics (chow, 2010). up to date, solar photovoltaic (pv) system application is divided into two distinctive categories, which are grid-connected system and stand-alone or off-grid system. the grid connected pv system operates by linking the solar pv with the utility-grid connection, where as the stand alone pv system operates without connection to the utility grid and utilize a battery system to store the excess energy generated by the solar pv as well as supplying the needed energy when no power from the sun is available. both system are widely applied especially in malaysia, however the stand-alone system offer notable issn: 2180-1053 vol. 3 no. 2 july-december 2011 journal of mechanical engineering and technology 56 advantage of able to operate in remote areas where utility-grid connection are not viable (zekai, 2008). in this project, an innovative design of bus stops is proposed where the infrastructure is to be equips with lamps and electrical signboard powered by solar pv energy system. the new system is aimed to provide higher level human comfort as well as information to user, and the prototype is targeted to be applied in main kampus, universiti teknikal malaysia melaka. currently, bus stops in utem is designed without any equipment to provide proper notice board and led display to give better information for the public. this paper described the overall design and development process for the utem solar powered bus stop such as pv system sizing, structure design and development as well as pv system installation and commissioning. 2.0 methodology the overall research methodology implement in this project is shown in fgure 1 below. development of a stand-alone solar powered bus stop mohd afzanizam mohd rosli1, mohd zaid akop 2, muhd ridzuan mansor3, sivarao s.4 1,2,3faculty of mechanical engineering, universiti teknikal malaysia melaka, 76100 durian tunggal, melaka. 4faculty of manufacturing engineering, universiti teknikal malaysia melaka, 76100 durian tunggal, melaka. email: 1afzanizam@utem.edu.my abstract this paper presents the development of a stand-alone solar photovoltaic (pv) system for bus stop at universiti teknikal malaysia melaka, malaysia. the design intent for the bus stop was to provide lighting and information to the bus stop users using reliable renewable energy system as well as to promote green technology awareness to the university residences. the stand-alone pv system was designed to power two units of cfl lamps and an led display unit installed at the bus stop. five units of polycrystalline photovoltaic modules with 110w rating each and four deep cycle battery units were utilized to provide three days of autonomy period for system operation. a part from that, 15 degree of tilt angle was selected for pv module placement to provide optimum energy generation as well as self cleaning for the modules. after the bus stop structure construction, the pv system was installed and commissioned. final results from the commissioning process showed that the system is able to operate successfully as per design requirement. keywords: solar photovoltaic, bus stop, stand-alone system 1.0 introduction photovoltaic or pv is currently one of the most attractive options for renewable energy resources in the world. malaysia which is blessed with yearly average solar irradiance of 1400 to 1900 kwh/m2 is considered to be in a very advantageous position to harness the unlimited energy towards pv applications to cater current domestic energy demand (malaysian pv handbook, 2009). the technology offers free and renewable energy, no emission effects during energy production, help to reduce dependency to conventional power supply, as well as easy to install and maintain due to its modular characteristics (chow, 2010). up to date, solar photovoltaic (pv) system application is divided into two distinctive categories, which are grid-connected system and stand-alone or off-grid system. the grid connected pv system operates by linking the solar pv with the utility-grid connection, where as the stand alone pv system operates without connection to the utility grid and utilize a battery system to store the excess energy generated by the solar pv as well as supplying the needed energy when no power from the sun is available. both system are widely applied especially in malaysia, however the stand-alone system offer notable issn: 2180-1053 vol. 3 no. 2 july-december 2011 development of a stand-alone solar powered bus stop 57 figure 1 overall research flow chart among the important stage in this project was the system development where the sizing of the pv standalone system is conducted. in order to attain an optimum design for the system, several sizing criteria such as location selection, design specification, determination of pv modules and battery capacity etc. were carefully calculated based on ms1837:2005 pv standard requirements (malaysian standard, 2005). the structure was later designed to accommodate the system sizing results using solidworks cad software. system testing and commissioning were performed in the final stage of the project to ensure that the pv system able to operate as per requirement. 3.0 system sizing not good good start project planning evaluation system installations modify system testing performance monitoring end system development evaluation literature review on current design of solar system not good good modify figure 1 overall research flow chart issn: 2180-1053 vol. 3 no. 2 july-december 2011 journal of mechanical engineering and technology 58 3.1 selection of bus stop location the first approach in designing the utem solar bus stop was to choose the suitable location for the structure. this is very crucial in order to ensure that the solar modules installed later on will have exposed to adequate sunray throughout the entire day for maximum power generation. the surrounding of the bus stop location was also ensured to be clear from shading sources such as tress and nearby structure especially to the solar modules. the bus stop must also be easily accessible to the users to serve its main function. the proposed location of the utem solar bus stop is shown in figure 2 below. figure 2 utem solar bus stop location 3.2 load determination the details of the electrical appliances designed for the bus stop and its power rating is shown in table 1 below. based on table 1, the majority of the loads selected operated on ac current. this is mainly because of low cost of acquiring the component compared to dc current components. a part from that, the system load is also divided into two cases which are with and without spare loads. the spare loads are incorporated into the design to cater for unexpected higher power usage in the future. the total load per hours in the case of without spare load is 953 watt-hours daily and in the case of spare loads is 1269 watt-hours daily. table 1 load profile for utem solar bus stop in a day no load detail type quantity power/unit (w) operating time operating hour/day total power (wh) 1 cfl lamp ac 2 18 7pm – 1am 6 216 2 led ac 1 49 8am – 14 686 display 10pm 3 internal system dc & ac 1 8.5 6 51 4 *13a socket ac 1 100 1 100 5 *spare loads ac 2 18 6 216 3.3 system specification and tilt angle several parameters have been determined for the design specification of the solar bus stop. first and foremost, the new utem solar bus stop is set to operate as a standalone system, where the power source is generated through pv modules only. the system is also set to be operated using 12v system voltage and must be able to operate up to 3 days of autonomy (in the case where zero sunlight occurred). the estimated achievable maximum solar radiation per day for the system is set to 3 peak sun hours and because the bus stop is a standalone photovoltaic system, the maximum depth of discharge for the battery used is the system is set to 80% capacity. the solar module is set to be placed on top of the bus stop roof to avoid any shading and to optimize the bus stop area. hence, another important parameter that was determined was the tilt angle for the solar module installation. the location for the bus stop is roughly at the latitude 2°18'51.61"n and longitude 102°19'8.17"e, where, optimum tilt angle for the solar module is from 2° to 3° facing south. however, it is recommended that for sites at latitudes between 15°s and 15°n, a tilt angle of 15° is used. hence, the tilt angle selected for the bus stop design is 15° to enable optimum energy generation as well as self cleaning for the modules. based on solar irradiation data for malaysia, the approximate average solar irradiance for the given latitude is 1400 w/m2 (sulaiman et al., 2008). 3.4 battery and module sizing based on load data in table 1 and design specifications, the sizing of the battery unit and the solar pv modules were performed. the determination for the number of batteries and pv modules were carefully conducted to ensure optimum sizing of the system can be achieved, which will benefits in lowering down the overall system cost as well as ensuring the system is able to operate efficiently. the sizing approach for standalone system is more delicate because the system operates solely on solar energy and energy from the batteries, compared with grid connected system. the battery must be able to supply the required power for minimum 3 days autonomy while the pv modules must be able to supply the power during sunlight as well as charging the battery to its maximum capacity continuously. table 2 below shows the final result for the battery and pv module sizing. figure 1 overall research flow chart among the important stage in this project was the system development where the sizing of the pv standalone system is conducted. in order to attain an optimum design for the system, several sizing criteria such as location selection, design specification, determination of pv modules and battery capacity etc. were carefully calculated based on ms1837:2005 pv standard requirements (malaysian standard, 2005). the structure was later designed to accommodate the system sizing results using solidworks cad software. system testing and commissioning were performed in the final stage of the project to ensure that the pv system able to operate as per requirement. 3.0 system sizing not good good start project planning evaluation system installations modify system testing performance monitoring end system development evaluation literature review on current design of solar system not good good modify 3.1 selection of bus stop location the first approach in designing the utem solar bus stop was to choose the suitable location for the structure. this is very crucial in order to ensure that the solar modules installed later on will have exposed to adequate sunray throughout the entire day for maximum power generation. the surrounding of the bus stop location was also ensured to be clear from shading sources such as tress and nearby structure especially to the solar modules. the bus stop must also be easily accessible to the users to serve its main function. the proposed location of the utem solar bus stop is shown in figure 2 below. figure 2 utem solar bus stop location 3.2 load determination the details of the electrical appliances designed for the bus stop and its power rating is shown in table 1 below. based on table 1, the majority of the loads selected operated on ac current. this is mainly because of low cost of acquiring the component compared to dc current components. a part from that, the system load is also divided into two cases which are with and without spare loads. the spare loads are incorporated into the design to cater for unexpected higher power usage in the future. the total load per hours in the case of without spare load is 953 watt-hours daily and in the case of spare loads is 1269 watt-hours daily. table 1 load profile for utem solar bus stop in a day no load detail type quantity power/unit (w) operating time operating hour/day total power (wh) 1 cfl lamp ac 2 18 7pm – 1am 6 216 2 led ac 1 49 8am – 14 686 figure 2 utem solar bus stop location table 1 load profile for utem solar bus stop in a day issn: 2180-1053 vol. 3 no. 2 july-december 2011 development of a stand-alone solar powered bus stop 59 display 10pm 3 internal system dc & ac 1 8.5 6 51 4 *13a socket ac 1 100 1 100 5 *spare loads ac 2 18 6 216 3.3 system specification and tilt angle several parameters have been determined for the design specification of the solar bus stop. first and foremost, the new utem solar bus stop is set to operate as a standalone system, where the power source is generated through pv modules only. the system is also set to be operated using 12v system voltage and must be able to operate up to 3 days of autonomy (in the case where zero sunlight occurred). the estimated achievable maximum solar radiation per day for the system is set to 3 peak sun hours and because the bus stop is a standalone photovoltaic system, the maximum depth of discharge for the battery used is the system is set to 80% capacity. the solar module is set to be placed on top of the bus stop roof to avoid any shading and to optimize the bus stop area. hence, another important parameter that was determined was the tilt angle for the solar module installation. the location for the bus stop is roughly at the latitude 2°18'51.61"n and longitude 102°19'8.17"e, where, optimum tilt angle for the solar module is from 2° to 3° facing south. however, it is recommended that for sites at latitudes between 15°s and 15°n, a tilt angle of 15° is used. hence, the tilt angle selected for the bus stop design is 15° to enable optimum energy generation as well as self cleaning for the modules. based on solar irradiation data for malaysia, the approximate average solar irradiance for the given latitude is 1400 w/m2 (sulaiman et al., 2008). 3.4 battery and module sizing based on load data in table 1 and design specifications, the sizing of the battery unit and the solar pv modules were performed. the determination for the number of batteries and pv modules were carefully conducted to ensure optimum sizing of the system can be achieved, which will benefits in lowering down the overall system cost as well as ensuring the system is able to operate efficiently. the sizing approach for standalone system is more delicate because the system operates solely on solar energy and energy from the batteries, compared with grid connected system. the battery must be able to supply the required power for minimum 3 days autonomy while the pv modules must be able to supply the power during sunlight as well as charging the battery to its maximum capacity continuously. table 2 below shows the final result for the battery and pv module sizing. issn: 2180-1053 vol. 3 no. 2 july-december 2011 journal of mechanical engineering and technology 60 table 2 sizing for battery and pv modules sizing details value a. battery i. amp-hour required excluding spare load (ah) 297.8 ii. amp-hour required including spare load (ah) 396.6 iii. no of battery need 4 b. pv module i. maximum power, pm (w) 110 ii. maximum voltage, vm (v) 17.2 iii. maximum current, im (a) 6.4 iv. short-circuit voltage, voc (v) 21.7 v. short-circuit current, isc (a) 6.9 vi total module required excluding spare loads 4 vii. total module required including spare loads 5 viii. array configuration 5 x 1 3.5 balance-of-system (bos) sizing the final procedure for the sizing is to determine the specification for charge controller and inverter. the charge controller is used to regulate the power from the pv modules to the loads and vice versa efficiently. the inverter in the other hand is selected to convert the dc voltage produced from the pv modules to ac voltage needed for the load operations (cfl lamps, led display etc.). the results of the balance of system sizing are shown in table 3 below. table 3 balance-of-system (bos) for utem solar bus stop bos sizing details value a. charge controller i. system voltage : minimum/maximum (v) 12/24 ii. total rated solar current (a) 18.5 iii. total maximum array current (a) 26.5 iv. total rated load current (a) 7.8 v. total maximum load current (a) 19.2 b. inverter i. ac system voltage (v) 240 ii. ac system frequency (hz) 50 iii. power factor (cos(pi)) 0.85 iv. rated va based on excluding spare loads 110 v. rated va based on including spare loads 270 table 1 sizing for battery and pv modules table 3 balance-of-system (bos) for utem solar bus stop issn: 2180-1053 vol. 3 no. 2 july-december 2011 development of a stand-alone solar powered bus stop 61 a part from that, another important parameter for the bos is the appropriate selection of cable size to minimize the power transmission lost in the cables. due to the small amount of energy generated by the pv modules, thus the sizing process in this project was also focused to reduce any losses that may be created in the system which can deteriorate the system performance. the diameter of cable size selected is 4mm for running from the charge controller to the loads as well as from the batteries to the charge controller and 6mm cable diameter for cable designed from the pv modules to the charge controller. the selected cable sizes for the whole system were carefully calculated to permit maximum cabling losses of 4% based on requirement stated in the malaysian standard (malaysian standard, 2005). 3.6 results of system sizing based on the system sizing, the appropriate components for the utem solar bus stop system were selected. the design specific yield of the system is 1037.1 kwh/kwp. the installation quality of the designed bus stop measured in term of performance ratio was found to be approximately 74% which satisfy the ms1837:2005 pv standard requirements (malaysian standard, 2005). a timer was also incorporated in the system to control precisely the operation hours of the loads in the system. table 4 and figure 3 below show the description of the selected equipments for the standalone pv system and the system overall schematic diagram. table 4 equipments for utem solar bus stop no item quantity desciptions 1 solar module 5 110 w, 12v, polycrystalline. brand: solartif 2 battery 4 100 ah, 12v, jxh100-12 valve regulated lead acid. brand: mpower 3 charge controller 1 model pl40, 12-48v, 30a. brand: phocos 4 inverter 1 standalone sinewave inverter, model picollo.brand: asp table 4 equipments for utem solar bus stop issn: 2180-1053 vol. 3 no. 2 july-december 2011 journal of mechanical engineering and technology 62 figure 3 schematic diagram of the utem solar bus stop 4.0 bus stop design and construction the design intent is for the bus stop to provide lighting and visual display to the users of the bus stop. a part from that, the bus stop structure must also be able to accommodate the solar modules as well as the rest of the balance of system components. acknowledging the above requirements, a design for the utem bus stop was proposed and shown in figure 4 below. figure 4 utem solar bus stop design the utem bus stop structure encompassed several features. the roof of the structure is tilted at 15 degree from horizontal plane to enable tilt angle of 15 degree for the module installation. the bus stop foundation is needed to be flat to ensure the tilt angle for the modules can be achieved. the roof structure also houses the cfl lamps and led display unit. due to the location of the led display unit where it is placed on the front of the bus stop roof facing the main road, the display is required to be fabricated based on ip65 casing requirement for ensuring reliable outdoor performance. the overall structure of the bus stop will be made from concrete as well as for the base of the bus stop. the balance-of-system components are located separately in a special housing behind the bus stop structure for ease of maintenance and security control. figure 5 below show the utem bus stop in the end of the construction process. figure 3 schematic diagram of the utem solar bus stop 4.0 bus stop design and construction the design intent is for the bus stop to provide lighting and visual display to the users of the bus stop. a part from that, the bus stop structure must also be able to accommodate the solar modules as well as the rest of the balance of system components. acknowledging the above requirements, a design for the utem bus stop was proposed and shown in figure 4 below. figure 3 schematic diagram of the utem solar bus stop figure 4 utem solar bus stop design issn: 2180-1053 vol. 3 no. 2 july-december 2011 development of a stand-alone solar powered bus stop 63 figure 4 utem solar bus stop design the utem bus stop structure encompassed several features. the roof of the structure is tilted at 15 degree from horizontal plane to enable tilt angle of 15 degree for the module installation. the bus stop foundation is needed to be flat to ensure the tilt angle for the modules can be achieved. the roof structure also houses the cfl lamps and led display unit. due to the location of the led display unit where it is placed on the front of the bus stop roof facing the main road, the display is required to be fabricated based on ip65 casing requirement for ensuring reliable outdoor performance. the overall structure of the bus stop will be made from concrete as well as for the base of the bus stop. the balance-of-system components are located separately in a special housing behind the bus stop structure for ease of maintenance and security control. figure 5 below show the utem bus stop in the end of the construction process. figure 5 utem bus stop after construction 5.0 system installation and commissioning after the bus stop structure was constructed, the stand-alone pv system was later installed at the location. simple l-shape brackets were first placed on top of the structure roof acting as the mounting points for the pv modules. five units of pv modules were installed in 5x1 array configuration as per sizing requirement at 15 degree tilt angle. later, the cfl lights and led display unit were installed on the bus stop. the batteries and balance-of-system components were installed separately on a special housing behind the bus stop. wires connecting the modules, loads and the rest of the pv system are secured inside conduits to ensure safety to the users as well as to protect from surrounding effect such as rain water that may cause it to deteriorate prematurely. the wires connecting the modules and the balance-of-system housing is also burried in ground for safety purposes. the installed pv system on the bus stop and the pv balance of system are shown in figure 6 and figure 7 below. figure 5 utem bus stop after construction 5.0 system installation and commissioning after the bus stop structure was constructed, the stand-alone pv system was later installed at the location. simple l-shape brackets were first placed on top of the structure roof acting as the mounting points for the pv modules. five units of pv modules were installed in 5x1 array configuration as per sizing requirement at 15 degree tilt angle. later, the cfl lights and led display unit were installed on the bus stop. the batteries and balance-of-system components were installed separately on a special housing behind the bus stop. wires connecting the modules, loads and the rest of the pv system are secured inside conduits to ensure safety to the users as well as to protect from surrounding effect such as rain water that may cause it to deteriorate prematurely. the wires connecting the modules and the balance-of-system housing is also burried in ground for safety purposes. the installed pv system on the bus stop and the pv balance of system are shown in figure 6 and figure 7 below. figure 5 utem bus stop after construction issn: 2180-1053 vol. 3 no. 2 july-december 2011 journal of mechanical engineering and technology 64 figure 6 utem bus stop with stand-alone pv system figure 5 utem bus stop after construction 5.0 system installation and commissioning after the bus stop structure was constructed, the stand-alone pv system was later installed at the location. simple l-shape brackets were first placed on top of the structure roof acting as the mounting points for the pv modules. five units of pv modules were installed in 5x1 array configuration as per sizing requirement at 15 degree tilt angle. later, the cfl lights and led display unit were installed on the bus stop. the batteries and balance-of-system components were installed separately on a special housing behind the bus stop. wires connecting the modules, loads and the rest of the pv system are secured inside conduits to ensure safety to the users as well as to protect from surrounding effect such as rain water that may cause it to deteriorate prematurely. the wires connecting the modules and the balance-of-system housing is also burried in ground for safety purposes. the installed pv system on the bus stop and the pv balance of system are shown in figure 6 and figure 7 below. figure 6 utem stop with stand-alone pv system issn: 2180-1053 vol. 3 no. 2 july-december 2011 development of a stand-alone solar powered bus stop 65 figure 7 housing for batteries and balance-of-system components 6.0 conclusion in conclusion, the standalone solar pv system for utem green bus stop was successfully developed in this project. the solar pv system comprised of five polycrystalline modules in 5x1 array configuration and four deep cycle batteries which give power to two units of cfl lamps and a led display unit. the bus stop structure was design and constructed to accommodate the module tilt angle of 15 degree. results from the testing and commissioning process performed show that the solar pv system was able to perform as per design requirement. the solar powered bus stop developed in this project was proven not only to perform as conventional bus stop, but also able to provide human comfort and information to the users as well as promoting green technology within the university and its residence. 7.0 acknowledgement the authors would like to thank universiti teknikal malaysia melaka for providing the support and fund for accomplishment of this project. the project was funded under utem top-down research grant pjp/2010/fkm(15a)-s715. figure 7 housing for batteries and balance-of-system components issn: 2180-1053 vol. 3 no. 2 july-december 2011 journal of mechanical engineering and technology 66 8.0 references pv industry handbook. 2009. pusat tenaga malaysia. malaysia. t.t. chow. 2010. a review on photovoltaic/thermal hybrid solar technology, journal of applied energy, vol. 87, pp. 365-379. s. zekai. 2008. solar energy fundamental and modeling technique: atmosphere, environment, climate change and renewable energy. 1st ed. ginora, spain: springer installation of grid connected photovoltaic (pv) system, malaysian standard ms1837:2005, sirim, malaysia. sulaiman s., kamaruzzaman s., and ahmad m.o. 2008. solar irradiation handbook for photovoltaic systems design in malaysia. solar energy research institute, universiti kebangsaan malaysia, malaysia. isbn: 9789675048326. 05(55-66).pdf issn: 2180-1053 vol. 9 no.1 january – june 2017 53 the effect of ply orientation on the vibration characteristics of ‘t’ stiffen composites panel: a finite element study opukuro s david-west1* division of automotive, mechanical and mechatronics engineering, school of engineering and technology, university of hertfordshire, hatfield, united kingdom, al10 9ab abstract aircraft producers have extensively adopted the use of t-shaped stiffened fibre reinforced composite panels in the thin walled structures such as the fuselage and wings. the composite materials present the advantage of high specific strength and stiffness ratios, coupled with weight reduction compare to traditional materials. this report presents a numerical study about the free-free vibration analysis of t-stiffened carbon fibre reinforced epoxy composite panels with surface and identical ply orientations of 0°, 15°, 30°, 45°, 60°, 75° and 90° using ansys 17 finite element code. these changes has effect on the element stiffness matrix and hence the dynamic characteristics of the panels. the fundamental frequencies increase to a peak and then decrease taking the form a half sine curve. the dynamic analysis was realized using the lanczos tool to extract the mode shapes and natural frequencies. keywords: stiffen pane; modal analysis; finite element analysis; composite structure 1.0 introduction the desire of light weight structures in the aerospace industries in the present day advanced technology has increasingly admitted the use of composite materials in the secondary structures such as the fuselage, wing, engine cowl, landing gear cover, rudder, control surfaces, cabinets etc. stiffened composite panels reinforced with ‘t’ stiffeners are used in the construction of aircraft wings. the composite skin and stiffener laminates consist of layers stacked at different fibre orientation angles, which are often limited to 0°, 45°, and 90° (liu, et al 2010). the 45° plies are suitable to absorb the shear loads. the use of stiffened panels contributes to reduction in the weight of the panel without compromising the strength and stiffness. (york & williams, 1998), determined the critical buckling loads of prismatic benchmark metal and composite panels representative of typical aircraft wing panel configurations, with in-plane shear and compression load combinations using exact' and *corresponding author e-mail: o.david-west@herts.ac.uk journal of mechanical engineering and technology 54 issn: 2180-1053 vol. 9 no.1 january – june 2017 approximate methods. while (rikards, et al 2001) developed triangular finite element model for buckling and vibration analysis of laminated composite stiffened shells. also (suh, et al 2003) investigated the damage tolerance of stitched stiffened composite panels with clearly visible impact damage and discussed about the effects of stitching using unstitched, selectively stitched and fully stitched panels. (chiarelli, et al 1996) discussed about the post critical characteristics of stiffened panels loaded in compression and design strategies in this regime. while (chen, et al 2003) presented an analytical method for ultimate longitudinal strength calculation and reliability analysis of a ship’s hull made of composite materials. the definition of a post-buckling optimisation procedure for the design of composite stiffened panels subjected to compression loads was reported by (bisagni & lanzi, 2002) based on a global approximation strategy, where the structure response is given by a system of neural networks. (bedon & amadio, 2012) presented an analytical formulation proposed for the estimation of the buckling resistance of flat laminated glass panels under in-plane compression or shear, using two different design approaches one directly derived from the theory of sandwich panels and the other based on the approximate concept of equivalent thickness. (hwang & huang, 2005) employed nonlinear buckling analysis using the finite element method to investigate buckling and postbuckling behaviour of unidirectional composite materials with two delaminations under uniaxial compression and reported that with a long delamination close to the surface of the laminate, the inner and short delamination has no effect on the buckling stress. however, the presences of inner, short delamination significantly change the behaviour of delamination growth. (lanzi & giavotto, 2006) presented a multi-objective optimization procedure for the design of composite stiffened panels based on genetic algorithms and three different methods of global optimization: neural networks, radial basis functions and kriging approximation, and the response surfaces were used to approximate the post-buckling characteristics. research of fatigue and stability of the highly stressed thin plate structures is an ongoing activity that provides valuable information for the structural engineers. (jia & ulfvarson, 2004) investigated about the static and dynamic behaviour of a lightweight ship deck and reported about the improvement that can be achieved by replacing a conventional steel structure with lightweight material using finite element. the free vibrations characteristics of simply supported anisotropic composite laminates were investigated using analytical approach (ganapathi, et al 2009) using the first-order shear deformation theory and the shear correction factors. (yang, et al 2008) investigated the reliability of orthogonally stiffened composite plate with boundary conditions of all four edges simply supported to uniform transverse load using grillage model assumptions and sensitivity of the variables observed. (sliseris & rocens, 2013) proposed an optimization technique for composite plates with discrete varying stiffness consisting of three steps for structural compliance and stress field differences, size optimization with genetic algorithm and dimension optimization of plates internal structure by neural network. (herencia, et al 2008) presented an initial sizing optimisation of anisotropic composite panels with t-shaped stiffeners using mathematical programming to model the skin and the stiffeners with the laminate the effect of ply orientation on the vibration characteristics of ‘t’ stiffen composites panel: a finite element study issn: 2180-1053 vol. 9 no.1 january – june 2017 55 parameters and genetic algorithms to account for manufacturability and design practices. (xue, et al 2011) investigated about the optimization of top-hat stiffened composite panels with the objective to minimize the weight and concluded from the parametric study that the plate aspect ratio and in-plane loading ratio have coupling effects on the critical buckling load. (marín, et al 2012) reported the optimization procedure for a geometric design of a composite material stiffened panel by a neural network system using the results of finite element analyses and genetic algorithm. the composite panels considered in this investigation have the longitudinal stiffeners, loaded in the in-plane direction, with changes to the surface and identical ply orientation being 0°, 15°, 30°, 45°, 60°, 75° and 90°, which has effect on the laminate stiffness and hence the vibration characteristics. the angle plies are known to support the structure against shear loading, but commonly reported in the literatures are laminates that contains the ±45° plies; in this investigation other additional angle plies have been considered i.e. ±15⁰, ±30⁰, ±60⁰ and ±75⁰. the results of this study will be useful to the manufacturers and users of stiffened panels such as the aerospace industries. 2.0 description of the panel the structure being used for this investigation has the special design such as the ones application for use at the wings of the aircraft. the sketch of the panel is shown in figure 1 showing the skin, web and foot-flange as parts of the structure. the stringer consists of the web and the foot-flange. all the parts of the panel are made of carbon fibre reinforced epoxy composite material. the stringer being perfectly bonded to the skin with epoxy as the adhesive which is the same material as the base resin of the composites laminate that comprises all the component parts. the stringers strengthen the skin and keeping it stable so it can absorb in-plane load. the laminate configurations of the components of the panels are as described in table 1. the thickness of a ply was 0.125 mm. the strength of the panel will depend on the laminate configuration of the component parts, bonding strength between the footflange and skin, and the manufacturing process. these carbon fibre reinforced composite panels have superior strength and stiffness compared to ones made with traditional materials such as steel or aluminium. also it is lighter in weight, making it a good candidate for air vehicles. the uniqueness of the panel construction provides resistance to impact blow, vibration and breakage. about 50% of aircraft structures are made of carbon fibre reinforced composite materials (roeseler, et al 2007). also (tanasa & zanoaga, 2013) has document reasons as regards the continuous use of carbon fibre-reinforced composites in the aviation industry. in boeing 787, (nayak, 2014) some of the components of the wings are made of carbon fibre reinforced composites; the stacking plies typically comprises of 0°, 90°, 45° and -45° directions (liu, et al 2010). the 0° and 90° plies are useful to take the in-plane loads, while the ±45° plies supports the shear strength. journal of mechanical engineering and technology 56 issn: 2180-1053 vol. 9 no.1 january – june 2017 figure 1. features of the t-stiffened composite panel. table 1: laminate configurations of the panels description lay up skin web foot flange [a°/0°/-a°/a⁰/90°/0°/±a°/0°/-a°/a°/90°/-a°]s [a°/0°/90°/-a°/90°/0°2/90°/-a°/90°/0°/a°]s [a°/0°/90°/-a°/90°/0°]s where a = 0°,15°, 30°, 45°, 60°, 75° and 90° 3.0 material properties the material used for the composite structure is carbon fibre reinforced / epoxy lamina. the unidirectional layer orthotropic properties for the material are given as in table 2, as obtained from reference (hou, et al 2000) were used in the finite element analysis. table 2: material properties for cfrp composite lamina property value ex (gpa) ey (gpa) ez (gpa) gxy (gpa) 132 10.3 10.3 6.5 all dimensions in mm the effect of ply orientation on the vibration characteristics of ‘t’ stiffen composites panel: a finite element study issn: 2180-1053 vol. 9 no.1 january – june 2017 57 gxz (gpa) gyz (gpa) vxy vxz vyz ρ (kg/m3) 6.5 3.91 0.25 0.25 0.38 1570 the material properties are given with reference to the ply coordinate axes where index ‘x’ denotes the ply principal axis that coincides with the direction of maximum in-plane young’s modulus (fiber direction). index ‘y’ denotes the direction transverse to the fiber in the plane of the lamina and index ‘z’ the direction perpendicular to the plane of the lamina. the lamina is a transversely isotropic structure, hence where ex represent the longitudinal modulus, ey and ez the transverse modulus, vxy is the major poisson’s ratio and gxy, gxz and gyz are in-plane shear modulus. table 3: the specifications of epoxy properties property value modulus, e (gpa) poisson’s ratio, v density, ρ (kg/m3) 10.5 0.3 1560 in table 3 are the mechanical properties of the resin used for the analysis used to characterise the solely resin rich sections of the model. 4.0 concept of dynamic analysis a general dynamic analysis will solve the equation of motion which gives the time dependent response of every node point in the structure by including inertial forces and damping forces in the equation. }{}]{[}]{[}]{[ ... fxkxcxm  (1) where [m] represents the structural mass matrix,{ .. x }the nodal acceleration vector, [c] the structural damping matrix,{ . x }the node velocity vector, [k] the structure stiffness matrix, {x}the node displacement vector and{f}is the applied time varying load. most engineering systems designers are interested in the natural frequencies and mode shapes of vibration of the system. however, for vibration modal analysis, the damping is generally ignored and considering a free vibration multi-degree of freedom system, the dynamic equation becomes. }0{}]{[}]{[ ..  xkxm (2) journal of mechanical engineering and technology 58 issn: 2180-1053 vol. 9 no.1 january – june 2017 if the displacement vector }{x , has the form txx sin}{}{  , then the acceleration vector is txx  sin}{}{ 2 ..  and substituting into equation (2), gives the eigenvalue equation. }0{}]){[]([ 2  xmk  (3) each eigenvalue has a corresponding eigenvector and the eigenvectors cannot be null vectors. 0][][ 2  mk  (4) equation (3), represent an eigenvalue problem, where 2  is the eigenvalue and }{x the eigenvector (or the mode shape). the eigenvalue is the square of the natural frequency of the system. 5.0 geometry of the shell element and the stress – strain relationship ansys shell281 element was primarily used to model the composites plates; it is primarily used to model composite shells or sandwich construction. the accuracy in modelling composite shells is governed by the first-order shear-deformation theory (usually referred to as mindlin-reissner shell theory). the element is formulation based on logarithmic strain and true stress. the element has eight nodes with six degrees of freedom at each node, i.e. translations in the x, y, and z axes, and rotations about the x, y, and z-axes. figure 2 shows the geometry and node locations of the element. it is defined by shell section information and eight nodes (i, j, k, l, m, n, o and p). a triangular-shaped element may be formed by defining the same node number for nodes k, l and o as shown in figure 2. figure 2. geometric description of shell281 (ansys, 2103). the stress is related to the strains by: the effect of ply orientation on the vibration characteristics of ‘t’ stiffen composites panel: a finite element study issn: 2180-1053 vol. 9 no.1 january – june 2017 59      d or       1 d (5) this is in accordance with the hooke’s law. the flexibility or compliance matrix, [d]-1 is (ansys, 2013)                                    xz yz xy zz zy z zx y yz yy yx x xz x xy x g g g eee eee eee d 100000 010000 001000 0001 0001 0001 1    (6) ex = young’s modulus in the x-direction; ey = young’s modulus in the y-direction ez = young’s modulus in the z-direction; ʋxy =major poisson’s ratio; ʋyx =minor poisson’s ratio; ʋxz = major poisson’s ratio x-z plane; gxy = shear modulus in the xy plane gyz = shear modulus in the xy plane; gxz = shear modulus in the xy plane also, the [d]-1 matrix is assumed to be a symmetric matrix, so that: x xy y yx ee   ; x xz z zx ee   ; y yz z zy ee   in fibre reinforced composite structures the stiffness is affected by the stacking sequence and hence the stiffness matrix of equation (4) for the eigenvalue analysis. a component of the stiffness is the [d] which consist of the modulus and the poison’s ratio. 6.0 finite element modelling and the mesh discretization a three-dimensional linear elastic finite element model of the composite panel was constructed using shell 281 (element with mid-side node) in ansys 17 finite element code. this element has six degrees of freedom at each node. the discretisation is the splitting of the continuous composite panel into the attached separated units. the finite element model has 5960 shell elements and 360 solid elements for all the panels considered in this study. journal of mechanical engineering and technology 60 issn: 2180-1053 vol. 9 no.1 january – june 2017 figure 3 shows the finite element discretisation for the panel with the display of the shell element thickness. the element size was 10 mm. the epoxy resin rich region between the web, foot flange and skin was characterised using a solid element with corner and mid-side nodes (solid 186) and the contact between the foot flange and skin as bonded model. figure 3. mesh discretization of the panel showing the display of the elements the shell element allows for finite rotations and membrane strains. the models were discretized into sufficient number of elements to allow for adequate representation of the deformation that coincides with the natural frequencies of the panels. 7.0 modal analysis of composite panels the aim of the modal analysis is to determine the natural mode shapes and frequencies of the panel during free vibration. the mode shapes describe the deformation of the structure at the natural frequencies. this resonant vibration is usually due to the interaction between the inertial and elastic properties of the panel materials. the analysis was performed using modal analysis tool available in ansys 17 finite element code. the first twenty-one modes of vibration of the panels were extracted using the block lanczos method. as this is a free vibration analysis i.e. no boundary condition, and the elements in it description have six degrees of freedom at each node, the first six modes were rigid body modes with zero frequency value. the finite element analysis yields a stiffness matrix and a mass matrix; with these two matrices (while the damping effect is ignored) an eigenvalue problem is generated and solved for the mode shapes and natural frequencies of the panels. usually, it is a good idea to design the panels operating below the fundamental frequency (i.e. the first mode). the effect of ply orientation on the vibration characteristics of ‘t’ stiffen composites panel: a finite element study issn: 2180-1053 vol. 9 no.1 january – june 2017 61 8.0 description and comparison of fundamental mode shapes the mode shapes were affected by the characteristics of the structure such as the stiffness, deflection distribution and fiber orientation. the laminates that constitute the panels are symmetrical in the stacking sequence, but the difference being the orientation of the surface ply and identical lamina, which are 0°, 15°, 30°, 45°, 60°, 75°, and 90°, table 1 shows the stacking sequence. in figures 4 to 10 are the fundamental modes of vibration with the natural frequencies of 41 hz, 54 hz, 65 hz, 74 hz, 71 hz, 55 hz and 42 hz respectively. figure 4. 0° degree surface ply orientation figure 5. 15° degree surface ply orientation figure 6. 30° degree surface ply orientation figure 7. 45° degree surface ply orientation figure 8. 60° degree surface ply orientation figure 9. 75° degree surface ply orientation journal of mechanical engineering and technology 62 issn: 2180-1053 vol. 9 no.1 january – june 2017 figure 10. 90° degree surface ply orientation exploring the mode shapes of the stiffened plate it could be seen that the fundamental modes for the panels with 0°, 15°, 60°, 75° and 90° surface ply orientation are similar. much of the vibration energies seems to have been trapped at the corners where the maximum displacements where attained. this implies that the corners of the panels were compliant with the energy at the fundamental frequencies. the shear effects were pronounced in the panels with the 30° and 45° surface ply directions, hence suitable configuration for the absorption of shear loads. the maximum displacements are trapped at two opposite corners of the panels. 9.0 the effect of surface ply orientation the vibration of the composite panels depends highly on the extremely large number of material permutations, such as the stacking sequence, architecture and the interface properties. hence, the strategies to achieve the design requirements must take into account the complexities of composite material structure and the geometry of the structural elements. axially biased composite laminates such as the ones of the panels consider in this study are an important class of laminates as they display good characteristics in both the axial and shear directions. they are favoured in the design of structures in the aerospace, marine and automobile industries, depending on how the designers want to tailor the strength properties. the effect of ply orientation on the vibration characteristics of ‘t’ stiffen composites panel: a finite element study issn: 2180-1053 vol. 9 no.1 january – june 2017 63 figure 11. the relation between the fundamental frequency and the surface ply orientation figure 12. comparison of the natural frequencies in figure 11 is presented the plot of the fundamental frequencies of the panels as against the surface ply fibre direction. the curve takes the form of an approximate half sine 30 35 40 45 50 55 60 65 70 75 80 0 20 40 60 80 100 f u n d a m e n ta l f re q u e n cy ( h z) surface ply orientation (degrees) 0 100 200 300 400 500 600 700 800 900 1000 0 2 4 6 8 10 12 14 16 n a tu ra l f re q u e n ci e s (h z) modes 0 degree surface orientation 15 degree surface orientation 30 degree surface orientation 45 degree surface orientation journal of mechanical engineering and technology 64 issn: 2180-1053 vol. 9 no.1 january – june 2017 curve with fundamental frequencies increase from 0° to 45° surface ply panels and decrease from the 60° to 90° surface ply direction panels. the highest value of the first natural frequency being 74 hz, attributed to the panel with 45° direction for the surface lamina is indicative of the fact that it has the highest stiffness. the relationships between the natural frequencies and the modes for all the panels are ploted and superimposed in figure 12. it is an approximate step-wise rise from the lowest to the highest value. the differences become significant as the resonant frequencies increases, this is thought be because of the laminate stacking sequence that has significant effect on the panel stiffnesses and the complex vibration energy absorption characteristics at high frequencies. these fourteen natural frequencies of the panels ranging from 41 – 761 hz, 53 – 775 hz, 65 – 785 hz, 74 – 810 hz, 71 – 832 hz, 55 – 859 hz and 42 – 871 hz for the 0°, 15°, 30°, 45°, 60°, 75° and 90° surface ply direction panels respectively. 10.0 conclusions this study has shown that surface ply orientation of carbon fibre reinforced composite panels and stacking sequence have significant effect on the vibration performance of the structure. the natural frequencies of vibration and the mode shapes of the panels were obtained by using the lanczos tool to extract the dynamic characteristics. some of the modes have the vibration energies trapped at certain locations of the panel, while others were globally. the first natural frequencies of the panels changes as the ply direction at the surface of the panels and identical lamina changes from 15° to 90°. an increase to a peak value of 74 hz for the panel with 45° on the surface and then gradually decreases. also considering the relationship of how the first fifteen natural frequencies increases with the modes an approximate step-wise rise from the lowest to the highest value was observed. as the configurations of the composite panels used for this study are typical for the aerospace industries, these results from this investigation will be of interest to such industries. further investigation in this study will be to see the effect of hybridization. this work has implications in the selection of composite laminate lay up for optimum combinations stiffness, vibration, compression and shear behaviour. references ansys mechanical apdl theory reference 2013. bedon, c., & amadio, c. (2012). buckling of flat laminated glass panels under in-plane compression or shear. engineering structures 36 185–197. bisagni, c., & lanzi, l. (2002). post-buckling optimisation of composite stiffened panels using neural networks. composite structures 58 237–247. the effect of ply orientation on the vibration characteristics of ‘t’ stiffen composites panel: a finite element study issn: 2180-1053 vol. 9 no.1 january – june 2017 65 chen, n. z., sun, h., & guedes soares, c. (2003). reliability analysis of a ship hull in composite material. composite structures 62, 59–66. chiarelli, m., lanciotti, a., & lazzeri, l. (1996). compression behaviour of flat stiffened panels made of composite material. composite structures 36, 16 1 – 169. ganapathi, m., kalyani, a., mondal, b., & prakash, t. (2009). free vibration analysis of simply supported composite laminated panels. composite structures 90, 100–103. herencia, j. e., weaver, p. m., & friswell, m. i. (2008). initial sizing optimisation of anisotropic composite panels with t-shaped stiffeners. thin-walled structures 46, 399–412. hou, j. p., petrinic, n., ruiz, c., & hallett, s. r. (2000). prediction of impact damage in composite plates. composites science and technology 60 273 281. hwang, s., & huang, s. (2005). postbuckling behavior of composite laminates with two delaminations under uniaxial compression. composite structures 68, 157–165. jia, j., & ulfvarson, a. (2004). a parametric study for the structural behaviour of a lightweight deck. engineering structures 26, 963–977. lanzi, l., & giavotto, v. (2006). post-buckling optimization of composite stiffened panels: computations and experiments. composite structures 73, 208–220. liu, d., toropov, v. v., zhou, m., barton, d. c., & querin, o. m. (2010). optimization of blended composite wing panels using smeared stiffness technique and lamination parameters. in proceedings of the 51st aiaa/asme/asce/ahs/asc structures, structural dynamics, and materials conference, orlando, florida, 12 april 2010 15 april 2010. marín, l., trias, d., badalló, p., rus, g., & mayugo, j.a. (2012). optimization of composite stiffened panels under mechanical and hygrothermal loads using neural networks and genetic algorithms. composite structures 94, 3321–3326. nayak, n. v. (2014). composite materials in aerospace applications. international journal of scientific and research publications, volume 4, (9), 2250-3153. rikards, r., chate, a., & ozolinsh, o. (2001). analysis for buckling and vibrations of composite stiffened shells and plates. composite structures 51 361 – 370. roeseler, w. g., sarh, b., & kismarton, m. u. (2007). composite structures: the first 100 years. in proceedings of the 16th international conference on composite materials, kyoto, japan, 8 – 13 july 2007. journal of mechanical engineering and technology 66 issn: 2180-1053 vol. 9 no.1 january – june 2017 sliseris, j., & rocens, k. (2013). optimal design of composite plates with discrete variable stiffness. composite structures 98, 15–23. suh, s. s., han, n. l., yang, j. m., & hahn, h. t. (2003). compression behavior of stitched stiffened panel with a clearly visible stiffener impact damage. composite structures 62, 213–221. tanasa, f., & zanoaga, m. (2013). fibre-reinforced polymer composites as structural materials for aeronautics. in proceedings of the international conference of scientific paper, brasov, 23 – 25 may 2013. xue, x. g., li, g. x., shenoi, r. a., & sobey, a. j. (2011). the application of reliabilitybased optimization of tophat stiffened composite panels under bidirectional buckling load. in proceedings of the 18th international conference on composite materials, south korea, jeju island, 21 – 26 august 2011. yang, n., das, p. k., & yao, x. (2008). reliability analysis of stiffened composite panel”. in proceedings of the 4th international asranet colloquium, athens, 2008. york, c. b., & williams, f. w. (1998). aircraft wing panel buckling analysis: efficiency by approximations. computers and structures 68, 665 – 676. issn: 2180-1053 vol. 6 no. 2 july-december 2014 1 a numerical study of steady and unsteady flow and heat transfer from a confined slot jet impinging on a constant heat flux wall d. u. lawal 1* , a. a. abubakar 2 , m. b. alharbi 3 , r. ben-mansour 4 1,2,3,4 department of mechanical engineering, king fahd university of petroleum & minerals, dhahran, 31261, saudi arabia abstract impinging jets have been used effectively in several applications including films and foods, rapid cooling and heating processes, tempering of glass and metal, drying of papers, coating, and freezing of tissue. in this work, a numerical simulation of steady and unsteady flow and heat transfer due to a confined 2-d slot jet impinging on constant heat flux plate is presented. two cases of problem were considered. in the first case, jet-to-plate spacing was varied from 2 to 5 at a fixed jet reynolds number of 500. in the second case, jet reynolds number was varied from 200 to 750 at fixed jet-to-plate spacing of 5. in the steady regime, the stagnation nusselt number was found to increase linearly with increasing reynolds number, and the distribution of heat transfer in the wall jet region was found to be highly influenced by flow characteristics of the jet. a strong correlation between pressure distribution and nusselt number was noticed. the critical reynolds number at which the symmetry of the flow in the formation of vortex sheets is highly disrupted was determine d. it was observed that, at the critical reynolds number, the area-averaged heat transfer coefficient is high and influence the drastic changes of the nusselt number in the unsteady regime. keywords: jet impingement; steady flow; unsteady flow; heat flux; critical reynolds number 1.0 introduction impinging jets are jets of fluid that are directed on a surface that is required to be dried, heated or cooled. due to their high mass and heat transfer rates, impinging jet have received its due attention (vadiraj & prabhu, 2008; vadiraj et al., 2011). because of their high efficiency and high heat transfer rate, industries and engineering firms had applied jet impingement to the applications like freezing of tissue, rapid heating or cooling process, metal and drying of papers. among numerous advantages of jet impingement is the simplicity to move the jet to a desired location (chattopadhyay & saha, 2002) and the jet ability to remove large amount of heat and mass transfer from the surface of impingement. in electrical and electronics components, jet impingement is used to quench the heat generated in the components (guarino & manno, 2002). compared to other regions, the impingement region has the largest mass and heat transfer rates (sergey et al., 2007). prandtl number, roughness of the target plate, reynolds number, target plate inclination, jet-to-plate spacing, nozzle geometry, and many more are the parameters that influence the jet impingement on a surface (vadiraj et al., 2011). chen et al. (2000) carried out numerical and experimental investigation of high schmidt number mass transfer in a laminar impinging slot jet flow at reynolds number ranging from 220 to 690. their results showed that the point of maximum mass transfer rate is located at 1 1 2 width from the point of stagnation. chiriac and ortega (2002) modelled the laminar fluid issn: 2180-1053 vol. 6 no. 2 july-december 2014 2 flow and heat transfer associated with impinging slot jets on an isothermal plate. the jet reynolds number was varied up to the unsteady regime at fixed prandtl number and jet-toplate spacing. the unsteadiness was discovered at reynolds number between 585 and 610. contrary to the heat transfer characteristics of the plate under steady state, the unsteady regime shows an abrupt rise in stagnation nusselt number and area-averaged heat transfer coefficient. the unsteadiness was also observed to cause the asymmetry of flow fields, lateral jet instability, and formation of shear layer vortices at the jet exit. similarly, sahoo and sherif (2004) carried out a numerical investigation of the heat transfer process in the mixed convection-laminar regime due to jet impingement on a constant heat flux surface. the resulting flow fields and isotherms were carefully studied as the reynolds number, richardson number, and domain aspect ratio were varied. as expected, the results shows that the average nusselt number increases with increasing reynolds number up to a certain value of reynolds number and domain aspect ratio, where the average nusselt number does not change significantly. they concluded that, buoyancy effects are not very significant on the heat transfer characteristics of laminar-flow slot impingement jets. chen et al. (2005) also carried out theoretical investigation of laminar flow and heat transfer due to slot jets impinging on arbitrary-heat-flux surface. expressions of the heat transfer coefficients of the various regions were determined, which were all found to be in good agreement with previous experiments. aldaddagh & sezai (2002) numerically studied the characteristic of heat transfer and laminar flows of multiple square jet impingements. several jets-to-plate spacing ratio were considered. results revealed that the variation of jet-to-plate spacing significantly affect the flow structure of jet impingement on the heated plate and had negligible effect on peak nusselt number. heat transfer characteristics due to turbulent flow of an impinging slot jet was also largely studied in the literature. vadiraj et al. (2011) studied reynolds number and jet-to-plate spacing effect on the local heat transfer distribution between smooth flat surface and impinging air jet. imaging technique was used to capture the flow fields. the surface temperature at various points was obtained using infra-red radiometry techniques. they obtained correlations for the nusselt number at various points on the domain, which were all found to be in agreement with experiments. san & chen (2014) conduct an investigation on the effects of jet-to-jet spacing and jet height on heat transfer characteristics of an impinging jet array. nusselt number due to five confined circular air jets impinging on a flat surface were measured. the jet-to-jet to jet diameter (s/d) ratio and jet height to jet diameter (h/d) ratio were varied at a constant reynold number of 20,000. small s/d and h/d ratios were found to result in large nusselt number due to the jets interaction. however, reduction in nusselt number was observed for both intermediate and large ratios due to weaker jets interaction and interference. caggese et al. (2013) carried out an experimental and numerical investigation of the heat transfer characteristics of a flat plate under a fully confined impingement jet. the experiments were carried out over a range of reynolds varying between 16,500 and 41,800. both experimental and numerical results were found to be in agreement. their results revealed that for the target plate, low dependence of the local and average heat transfer level with z/d was observed. mohammadpour et al. (2014) carried out optimization of an impingement system that is composed of both pulsating and steady submerged slot jets using numerical techniques. the simulation was carried out at varying temperatures, amplitudes, and frequencies of the slot jets. the results show that uniform as well as significantly high nusselt number can be obtained using combination of pulsed and steady jets. in view of this, a considerable increase issn: 2180-1053 vol. 6 no. 2 july-december 2014 3 in convective heat transfer was found at the stagnation point. they also show that a relatively higher heat transfer rate can even be obtained in a system that is composed of both intermittent-steady jets rather than the sinusoidal-steady ones. from the foregoing literature review, it can be concluded that no study has investigated the unsteady flow analysis and heat transfer from a confined slot jet impinging on a constant heat flux wall (target). to the best of the author’s knowledge, the nature of the flow fields and isotherms is not fully exploited during the unsteady regime. therefore, the objective of present research is to study the effect of geometrical changes and jet inlet velocity on heat transfer characteristics of impinging plates at at constant heat flux and under laminar flow condition. 2.0 problem description the problem description is illustrated in figure 1 below. it is a uniform velocity 2dimensional jet of air entering a nozzle of width (w) located above a channel of height (h) and length (l). the two cases below further described the problem statement. case 1 at a constant reynolds number of 500 and plate length of 25cm, the jet-to-plate spacing (h/w) was varied from 2 to 5. case 2 at a constant jet-to-plate spacing (h/w) of 5 and fixed plate length of 50cm, the reynolds number was varied from 200 to 750. in both cases, the upper confining wall is considered to be adiabatic while the lower target wall was kept at constant heat flux of 1000w/m 2 . the above problems was solved using commercial computational fluid dynamics package (fluent 6.2). the entire domain was considered in order to fully detect any unsteadiness and asymmetry in our analysis. 3.0 model assumptions using air of low viscosity as a working fluid, the following assumptions are made:  flow is incompressible  unsteady  2-dimensional modelling  all physical properties are constant issn: 2180-1053 vol. 6 no. 2 july-december 2014 4 figure 1. the geometry of a 2-d confined impinging jet 3.1 governing equations with aforementioned assumptions, the continuity, the momentum, and the energy equations are reduced to equations (1) (3) respectively, and are considered to be the basic equations in our analysis: ∇. �̅� = 0 (1) 𝜕𝑢 𝜕𝑡 + �̅� . ∇u̅ = − 1 ρ ∇p + 𝜈∇2u̅ (2) 𝜕t 𝜕𝑡 + �̅� . ∇t = α∇2t (3) 3.2 boundary conditions (b.c) the b.c used are as follows: top plate: left and right walls, ν = 0, u̅ = 0 and 𝜕𝑇 𝜕𝑦 = 0 jet inlet, ν = vj, u̅ = 0 and t = 300k bottom plate: bottom wall, ν = 0, u̅ = 0 and 𝑞 = 1000𝑊/𝑚2 left and right outlet: 𝜕𝑢 𝜕𝑥 = 0, 𝜕𝑢 𝜕𝑦 = 0, 𝜕𝑇 𝜕𝑥 = 0, 𝑎𝑛𝑑 𝑃 = 𝑃𝑎𝑡𝑚 in other words, the boundary conditions are,  top plate is fixed and insulated  jet flow is isothermal with constant velocity  bottom plate is fixed with constant heat dissipation/flux  left and right outlets are far from the control volume, and considered to be at atmospheric conditions issn: 2180-1053 vol. 6 no. 2 july-december 2014 5 4.0 modeling in solving the above equations using the computational fluid dynamics code (fluent 6.2), the main procedure involved are: in gambit, create the geometry, mesh the domain, name the boundary conditions and export the output to fluent. in fluent, the boundary conditions and the material properties are set. the solver is then used to compute the equations governing the flow. the end result to the problem is achieved when the solution converged. 4.1 geometry the geometry of our problem consists of the target wall, outlets (left and right), top walls (left and right), and the jet inlet. the face of the geometry is depicted in figure 2. figure 2. the geometry as drawn in gambit 4.2 mesh the domain considered is meshed with quadrilateral elements of aspect ratio 0.98. a mesh of high density was used at regions of steep gradients, i.e. near the jet inlet and the target wall. this allows more accurate results. the meshed geometry is as shown in figure 3. figure 3. the mesh as built from gambit 4.3 material used and properties air was used as the working medium and its properties is tabulated in table 1. issn: 2180-1053 vol. 6 no. 2 july-december 2014 6 table 1. the properties of air at 300 k and 101325 pa properties of air at 300 k and 101,325 pa density 1.225 kg/m 3 specific heat capacity at constant pressure 1006.43 j/kg-k thermal conductivity 0.0242w/m-k viscosity 0.00001789 kg/m-s 4.4 mesh independence test six sets of mesh size were used to determine the optimum mesh size. the results are tabulated in table 2 according to the decreasing number of grids with s6 having the lowest number of cells. the velocity magnitude of h/w= 5 for different mesh scheme are illustrated in fig 4. the plot revealed no clear differences among the six mesh schemes. for compromise between computational cost and accuracy of results, s4 mesh scheme was selected. time independent test was then conducted for different h/w ratio (h/w=2, h/w= 3 and h/w= 4). table 2. the number of grids in the domain for h/w=5. mesh scheme number of cells s1 10,200 s2 9,000 s3 7,500 s4 6,200 s5 4,800 s6 3,900 figure 4. plot of velocity for various mesh schemes for h/w=5 0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16 0.18 0.2 0 0.05 0.1 0.15 0.2 0.25 v e lo c it y ( m /s ) h/w=5 s1 s2 s3 s4 s5 s6 issn: 2180-1053 vol. 6 no. 2 july-december 2014 7 5.0 results and discussions 5.1 validation the variation of time-averaged nusselt number at the point of stagnation for different reynolds number is depicted in figure 5. a direct relation between nusselt number and of reynolds number was observed in the steady regime, which is an indication that the rate heat transfer is due to convectional rise in jet inlet velocity. however, a significant departure from the linear relationship was observed in the unsteady regime. the sudden change in slope of the graph occurred at re = 600. at this point, nusselt number was noticed to rise sharply. in unsteady regime, the rate of convective heat transfer is greater at higher reynolds number than that at lower reynolds numbers due to instability (unsteadiness) encountered by the fluid. figure 5. time-averaged stagnation nusselt number for h/w=5 and different reynolds number available for validation is the work of chiriac & ortega (2002) which is presented in figure 6. from their work, the critical reynolds number was found to be at re =750. this value was determined experimentally and numerically. in the current work, the critical reynolds number was also found to be around re = 750. however, the nusselt number obtained in this study is quite higher than those from chiriac & ortega (2002). the observed differences may be attributed to the fact that chiriac & ortega (2002) based their analysis on an isothermal target plate temperature of 300k, whereas our analysis was based on a target plate at constant heat flux of 1000 w/m 2 . 0 200 400 600 800 1000 1200 1400 0 100 200 300 400 500 600 700 800 n u s ta g re stagnation nusselt number issn: 2180-1053 vol. 6 no. 2 july-december 2014 8 figure 6. transitional and critical reynolds number for the same type of flow as presented by chiriac & ortega (2002) 5.2 case 1 it is known that with increasing distance from exit and increasing momentum exchange between the jet and the surrounding, the free boundary of jet widens while the potential core contacts on the impingement surface and the wall jets are formed and spread laterally. thus, the overall structure of jet impingement on a surface as observed from figure 7 consists of potential core, wall jet region, free jet region, shear layer, stagnation point and uniform jet velocity inlet. the figure present the velocity contours of different h/w ratios (h/w = 2, 3, 4 and 5), for l = 25cm and reynolds number of 500. the formation of vortices near the lower wall and upper walls was observed in all cases. however, at h/w ratio of 4 and 5, the vortices are larger, well-formed and far away from the jet inlet. the size and strength of the vortices formed at h/w = 4 and 5 are noticed to be greater than those at h/w = 2 and 3. this is perhaps due to the fact that the jet entrainment near the jet inlet and the flow instability on the lower wall increases due to increasing domain size (h/w). the formation of secondary vortices in the upper wall surrounded by the main recirculating vortices was observed in all the cases. the formation of secondary vortices grew bigger with increasing h/w ratio. this may be attributed to the variation in momentum exchange between the fluid jets and the surrounding. issn: 2180-1053 vol. 6 no. 2 july-december 2014 9 figure 7. the velocity contours for re = 500 and (a) h/w=2, (b) h/w=3, (c) h/w=4, and (d) h/w=5 figure 8. the temperature contours for re = 500 and (a) h/w=2, (b) h/w=3, (c) h/w=4, and (d) h/w=5 issn: 2180-1053 vol. 6 no. 2 july-december 2014 10 figure 9. the surface nusselt number for re = 500 and h/w = 2, h/w = 3, h/w = 4, h/w = 5. figure 10. the wall pressure coefficients for re = 500 and h/w = 2, h/w = 3, h/w = 4, and h/w = 5. 0 200 400 600 800 1000 0 0.05 0.1 0.15 0.2 0.25 n u s s e lt n u m b e r( n u ) l(m) h/w=2 h/w=3 h/w=4 h/w=5 -0.2 0 0.2 0.4 0.6 0.8 1 0 0.05 0.1 0.15 0.2 0.25 p re s u re c o e ff ic ie n t( c p ) l(m) h/w=2 h/w=3 h/w=4 h/w=5 issn: 2180-1053 vol. 6 no. 2 july-december 2014 11 figure 11. target wall friction coefficient for re = 500 and h/w = 2, h/w = 3, h/w = 4, and h/w = 5. the steady state temperature contour for the given flow is depicted figure 8. it can be noticed that h/w ratio has a strong influence on the heat transfer characteristics of the plate. as expected, rate of heat transfer was observed to be higher at stagnation point and lower at flow separation point. moreover, figure 9 confirms that increasing the h/w ratio leads to considerable increment in nusselt number at the stagnation region, thus leading to higher heat transfer rate on the target plate. the higher heat transfer rate on the target plate is attributed to increase in the size of the shear layer generated by the jet nozzle exit. the effect of h/w ratio on normalized wall pressure coefficient is displayed in figure 10. the peak value was found to be located at the stagnation points. it can also be observed that the normalized wall pressure coefficient values increases with increasing h/w ratio. however, there is mild pressure recovery in a lateral direction away from the stagnation point (wall jet region) as a result of flow separation that is induced by the adverse pressure gradient introduced by the tendency for the flow to be re-entrained by the jet. figure 11 present the surface skin friction coefficients on the target plate. the observed trend in the surface skin friction coefficients on the target plate is similar to that observed in figure 10. it can be noticed that decreasing the domain size (h/w ratio) leads to reduction in skin friction coefficient. this indicated that the wall shear stress on the fluid decrease with h/w ratio. thus, magnitude of skin friction coefficients depend on the distance from stagnation point to the first separation point, the size and strength of the secondary vortices formed on the lower wall and the location. 5.3 case 2 the velocity contour of instantaneous velocities for re = 200, 300, 400, 500, 600 and 750 is illustrated in figure 12. it was noticed that at re = 600, the symmetry of the flow begin to diminish till re = 750 where the flow field is completely asymmetrical and the jet core is distorted. this point represent the critical regime in which unsteady (periodic) flow was observed. hence, re = 600 is the transitional reynolds number to the unsteady regime while re=750 indicates the critical reynolds number at which unsteadiness is fully developed. 0 0.03 0.06 0.09 0 0.05 0.1 0.15 0.2 0.25 s k in f ri c ti o n c o e ff ic ie n t( c f) l(m) h/w=2 h/w=3 h/w=4 h/w=5 issn: 2180-1053 vol. 6 no. 2 july-december 2014 12 figure 12. velocity contour plot for h/w=5 and (a) re=200 (b) re=300 (c) re=400 (d) re=500 (e) re=600 and (f) re=750 figure 13. velocity contour plot for h/w=5 and (a) t=0 s (b) t=2 s (c) t=4 s and (d) t=6 s issn: 2180-1053 vol. 6 no. 2 july-december 2014 13 in order to quantify the flow instabilities in the unsteady regime, the velocity contour at the critical reynolds number were sampled over time as presented in figure 13. it can be noticed that the pattern of the velocity field is neither constant nor symmetrical with time, as the pattern of the velocity field is neither symmetrical nor constant with time. the coefficients of pressure for different reynolds numbers is depicted in figure 14. it can be noticed that, as the reynolds number increases, the symmetry of the flow is disrupted. the peak point of the coefficient of pressure as expected was found to be situated at the point of stagnation. careful observation revealed that the wall static pressure drops rapidly as the flow turns and accelerate. it was noticed that the coefficient of pressure at the point of stagnation increases with increasing reynolds numbers. this is attributed to the overall increment in the flow velocity due to increment in reynolds number. issn: 2180-1053 vol. 6 no. 2 july-december 2014 14 figure 14. normalized pressure coefficients for h/w=5 and (a) re=200 (b) re=300 (c) re=400 (d) re=500 (e) re=600 and (f) re=750. issn: 2180-1053 vol. 6 no. 2 july-december 2014 15 figure 15. normalized pressure coefficients for h/w=5, re=750 and (a) t=0 s (b) t=2 s (c) t=4 s and (d) t=6 s in the unsteady regime moreover, figure 14 shows that rise in reynolds number cause the symmetry of the flow to be disrupted. figure 15 present the pressure coefficient for h/w = 5 and re = 750 in unsteady regime (t = 0 s, t = 2 s, t = 4 s and t = 6 s). at critical reynolds number, the stagnation coefficient of pressure changes (oscillates) with time. this clearly confirm the fact that re = 750 is indeed the critical reynolds number at which the flow unsteadiness and instabilities fully developed. 5.4 findings the following are some of the findings as related to the current work,  height-to-width ratio and jet inlet velocity have noticeable impact on the flow and heat transfer characteristics of jet impingement on the flat plate.  larger domain sizes are found to generate greater heat transfer rate and flow instabilities.  higher jet inlet velocity leads to the development of unsteadiness in flow, and drastically increases the heat transfer rate on the target plates. issn: 2180-1053 vol. 6 no. 2 july-december 2014 16 6.0 conclusions the effects height-to-width ratio and jet inlet velocity on the heat and flow characteristics of slot jets impinging on a flat target wall of constant heat flux have been presented. hence the following conclusion can be drawn from the study.  increasing the domain size leads to the formation of vortices (hence, instabilities) in the flow characteristics of the jet.  the transition and critical reynolds numbers were found to be 600 and 750 respectively.  at the critical reynolds number, the unsteadiness of the flow was fully observed and was found to have major impact on the heat transfer characteristics of the target plates.  increasing the jet inlet speed triggered the flow transition from steady to unsteady regime.  the heat transfer rate was noticed to be greater at unsteady regimes. acknowledgment the authors would like acknowledge the support received from king fahd university of petroleum & minerals (kfupm). nomenclature cf friction coefficient cp specific heat [j/kg.k] h heat transfer coefficient [w m -2 k -l ] h nozzle–plate spacing (m) k thermal conductivity [w m -1 k -l ] l channel length (m) nu nusselt number p static pressure [n m -2 ] q heat flux at target wall [w/m 2 ] re reynolds number based on hydraulic diameter (2w) t temperature of fluid [ o c] t time [s] u velocity in jet transverse direction (m/s) v velocity in jet stream wise direction (m/s) vj jet inlet velocity (m/s) w jet nozzle width (m) x jet transverse coordinate (m) y jet stream wise coordinate (m) greek symbols ρ density [kg.m -3 ] ν kinematic viscosity [m 2 s -1 ] α thermal diffusivity [m 2 s -1 ] τw wall shear stress [nm -2 ] subscripts j jet stag stagnation w target wall x local value of parameter on target wall issn: 2180-1053 vol. 6 no. 2 july-december 2014 17 references aldabbagh, l. b. y. & sezai, i. (2002). numerical simulation of three-dimensional laminar multiple impinging square jets. international journal of heat and fluid flow , 23, 509-518. caggese, o., gnaegi, g., hannema, g., terzis , a., & ott, p. (2013). experimental and numerical investigation of a fully confined impingement round jet. international journal of heat and mass transfer, 65, 873882. chattopadhyay, h. & saha, s. k. (2002). simulation of laminar slot jets impinging on a moving surface. journal of heat transfer, asme, 124, 1049-1055. chen, y. c., ma, c. f., qin, m., & li, y. x. (2005). theoretical study on impingement heat transfer with single-phase free-surface slot jets. international journal of heat and mass transfer, 48(16), 3381-3386. chen, m., chalupa, r., west, a. c. & modi, v. (2000). high schmidt mass transfer in a laminar impinging slot jet flow. international journal of heat and mass transfer, 43, 3907–3915. chiriac, v. a. & ortega, a. (2002). a numerical study of the unsteady flow and heat transfer in a transitional confined slot jet impinging on an isothermal surface. international journal of heat and mass transfer, 45, 1237-1248. guarino, j. r. & manno, v. p. (2002). characterization of laminar jet impingement cooling in portable computer applications . ieee transactions on components and pack aging technologies, 25(3). mohammadpour, j., zolfagharian, m. m., mujumdar, a. s., zargarabadi, m. r., & abdulah zadeh, m. (2014). heat transfer under composite arrangement of pulsed and steady turbulent submerged multiple jets impinging on a flat surface. international journal of thermal sciences, 86, 139-147. sahoo, d. & sharif, m. a. r. (2004). numerical modeling of slot-jet impingement cooling of a constant heat flux surface confined by a parallel wall. international journal of thermal sciences, 43(9), 877-887. san, j. y. & chen, j. j. (2014). effects of jet-to-jet spacing and jet height on heat transfer characteristics of an impinging jet array. international journal of heat and mass transfer, 71, 8-17. sergey, v. a., artur, v. b., vladimir, m., & dmitriy, m. m. (2007). experimental study of an impinging jet with different swirl rates. heat and fluid flow, 28, 1340-1359. vadiraj, k., & prabhu, s. v. (2008). experimental study and theoretical analysis of local heat transfer distribution between smooth flat surface and impinging air jet fro m circular straight pipe nozzle. heat and mass transfer, 51, 4480-4495. vadiraj, v., katti, s., & nagesh, y. (2011). local heat transfer distribution between smooth flat surface and impinging air jet from a circular nozzle at low reynolds numbers . heat mass transfer, 47, 237-244. http://www.sciencedirect.com/science/article/pii/s0017931005002024?np=y http://www.sciencedirect.com/science/article/pii/s0017931005002024?np=y http://www.sciencedirect.com/science/article/pii/s0017931005002024?np=y http://www.sciencedirect.com/science/article/pii/s0017931005002024?np=y http://www.sciencedirect.com/science/journal/00179310 http://www.sciencedirect.com/science/journal/00179310/48/16 http://www.sciencedirect.com/science/article/pii/s1290072904000316 http://www.sciencedirect.com/science/article/pii/s1290072904000316 http://www.sciencedirect.com/science/journal/12900729 http://www.sciencedirect.com/science/journal/12900729/43/9 issn: 2180-1053 vol. 10 no.2 june – december 2018 1 production of fuel briquettes from hybrid waste (blend of saw-dust and groundnut shell) modestus o. okwu 1* , olusegun d. samuel 2 , ikuobase emovon 3 1,2,3 department of mechanical engineering, federal university of petroleum resources, effurun, warri, delta state, nigeria abstract there is a great increase in the demand and cost of energy globally. the cost of purchasing petrol fuel is quite high; this has led the rural dwellers to concentrate on firewood as an alternative means of cooking, action which has greatly affected our ecosystem. this research is focused on the production of briquettes using waste sawdust and groundnut shell with starch and condemn oil as binder. these waste materials can be converted into a product that will provide alternative energy to the people rather than constituting environmental problems. the briquettes were produced using a modified briquetting machine developed at the federal university of petroleum resources engineering workshop. it was observed that the density of the briquette produced increased with blend ratio, amount of binder and compaction time. optimisation technique predicted briquette density of 278.1 kg/m 3 at the optimal level of 8.0, 10% and 50 minutes for the blend, binder and compaction time respectively. combustion related properties of the briquette at the optimum level were density(277.9 kg/m 3 ), ash content (25.29%), moisture content (4.68 %wt), bulk density (3.03) g/cm 3 ) and burning rate (0.46 g/min) and these values are in accordance with previous research studies. conclusively, the briquette fuel formed is effective, affordable and can be used as solid fuels to support heating in local and industrialized settings. in addition, establishing a small briquetting firm will serve as an alternative source of energy for cooking, create job opportunities and raise the standard of living of youths. keywords: briquette, density, ash content, moisture content, bulk density, burning rate 1.0 introduction the production of fossil fuel globally will start depreciating in 20 to 30 years’ time. this has been the major concern of scientist and engineers worldwide (adegoke and kuti, 2005; adegoke and mohammed, 2002). in emerging countries, organic waste offers approximately 20 to 33% of aggregate energy demand (vargas-moreno et al., 2012). biomass, a general name for all dry plant materials and organic waste is well appreciated in nigeria. this is because it is easy to access and can be obtained at no cost. * corresponding author e-mail: mechanicalmodestus@yahoo.com mailto:mechanicalmodestus@yahoo.com journal of mechanical engineering and technology 2 issn: 2180-1053 vol. 10 no.2 june – december 2018 they are commonly used in rural communities. most rural dwellers cannot afford the price of prevalent fuels such as kerosene and cooking gas necessary to meet their daily domestic needs. electric bill is on the high side in nigeria and the populace often experience constant epileptic power supply thus making life very difficult especially for those who use electric stoves. hence, the need for firewood as an alternative source of fuel. the use of firewood for cooking has negative effect on the populace as a result of the smoke produced during burning. (raymer 2006; rehfuess 2006). research is ongoing on a daily basis in search of substitute for firewood as a source of cooking and heating in homes. this solution has been obtained from by-products of agricultural waste which are available in large quantities and very useful in rural and urban region; developing and developed countries (yank et al., 2016). researchers in energy fields are giving so much attention to this area of research. also, investigation on the production of eco-friendly briquettes is seen in fuel and energy literature (thao et al., 2011). 2.0 review of literature on bio-briquettes however, a lot has been reported on briquette production though most researchers focus on briquettes development using single waste material. some of the research conducted include: briquette production from cotton stalk (abakr et al. 2006); briquette production from rice husk (sisman and gezer 2011 ); briquette production from sawdust (arinola et al., 2013, chinyere et al. 2014); briquette production from hyacinth plant (davies and abolude 2013, supatata et al. 2013, rezania et al., 2016, rezania et al., 2017); briquette production from rice (brand et al., 2017); briquette production from pterocarpus indicus leaves (anggono et al. 2017). it is possible to formulate briquettes from hybrid waste. briquettes developed from hybrid waste will offer a cheap, eco-friendly and readily available solid fuel for domestic and commercial purposes. for instance, olorunisola (2007) conducted research on briquette production using waste paper and coconut husk. rajkumar and venkatachalam (2013) researched on briquette production by combining cutton, soya beans and pigeon pea stalk. lubwama and yiga (2017) developed a system for production of briquettes by binding together groundnut shell and bagasse. nwabue et al., (2017) researched on briquette production from plastic and other waste materials with proper carbonization. lubwama and yiga (2018) researched on briquette production from rice and coffee husk. wu et al., (2018) looked at the development of briquettes from waste material by combining wood sawdust and cotton stalk. sawdust and groundnut shell are popular waste materials in nigeria. thus, the production of briquettes from these waste will go a long way to discourage deforestation, pollution and environmental degradation (okello et al., 2013). also, very few studies exist in the literature on the utilization of condemned oil as a binder. most researchers in previously reviewed work used cassava and starch as binder during briquette production. in this study, hybrid waste materials (sawdust and groundnut shell) is considered with condemned oil is used as binder. these materials were considered because they are prevailing and readily available waste materials. for instance, condemned oil as a waste material is usually disposed in hotels and restaurants, saw dust as waste are very much available in wood industries and groundnut shells are found at market place and farm land. production of fuel briquettes from hybrid waste (blend of saw-dust and groundnut shell) issn: 2180-1053 vol. 10 no.2 june – december 2018 3 2.1 sawdust as a briquette material according to kuti 2009, of all available biomass materials, wood contribute to the prevalent possible source of biomass which is very relevant in briquetting technology. to effectively address waste disposal issues in the timber and wood industry, it is important to convert the waste materials or sawdust into useful briquettes for domestic cooking and heating and for commercial purposes. this would greatly address the problem of pollution and unemployment (tembe et al., 2014). sawdust materials or wood waste are quite easy to come by and tend to be very relevant in briquette production (kuti, 2009). sawdust generated in nigeria is very high in volume and sawmills dispose these waste in a careless manner. in the past, sawdust generated as waste material in sawmill were gathered together in an open area or field and set on fire. this act is hazardous and environmentally unfriendly. presently, the act of burning sawdust in open places is discouraged. research has shown that these waste materials are useful form of renewable or alternative energy which can serve as solid briquettes for heating in homes and industries. kiss and alexa (2015) researched further on sawdust and briquetting and observed that sawdust contains natural lignin which is present in wood. this helps in bonding process and as such sawdust require very little or no binder during the formation of briquettes. though in the process of briquetting, there is need for greater compression of the raw material (sawdust and water) to form solid useful briquettes. kiss and alexa (2015) further stated that there are two types of briquettes produced from sawdust: solid briquettes and hollow briquettes. solid briquettes are briquettes bonded together using piston press while hollow briquettes are briquettes produced with hole at the center using screw press. hollow briquettes burn better than solid briquettes because of the presence of space for efficient combustion. the waste sawdust can serve as a source of wealth to the nation. (oyelaran et al., 2015). briquetting of sawdust is a traditional way of regulating the amount of heat released during burning process in homes and industries (adetogun, 2014). kiss and alexa (2015) are still of the opinion that raw sawdust have low burning efficiency. conversely, when the materials are bonded together to form briquettes, they become solid biofuel with high burning efficiency and higher calorific value than firewood. the combustion process of briquettes made from sawdust is highly efficient compared to the combustion process of firewood during heating. this is as a result of the low moisture content of briquettes made from sawdust (4%). the moisture content of firewood is quite high (65%) (kiss and alexa, 2015). 2.2 groundnut as a briquette material nigeria rank top on the list of foremost producer of groundnut globally. research has shown that the average annual production of groundnut is about 1,000,000 tons (oyelaran et al., 2015). one of the residue present in harvested groundnut is groundnut shell disposed as a waste material after shelling. it is important to note that this waste can be converted into useful briquette for cooking traditionally and industrially. the waste material is considered perfect for briquetting because of the low moist and ash content present (10% or lower and 2-3%) (ajobo, 2014). therefore, groundnut shell briquettes can serve as a substitute to firewood for cooking in homes and industries. a blend of sawdust and groundnut shell for briquette production is expected to produce journal of mechanical engineering and technology 4 issn: 2180-1053 vol. 10 no.2 june – december 2018 highly efficient and effective briquettes for domestic cooking in rural and urban setting. this is the ultimate goal of the research. 3.0 materials and methods 3.1 materials, reagents and equipment sawdust (sd), groundnut shell (gns) and starch were obtained from okuokoko sawmill, agbarho market and ugbmro market, uvwie local government area in delta state nigeria. 200ml granulated cylinder, 200ml pyrex beaker, electronic weighing balance (pesson, sweden), crucible (logat, finland) and bunsen burner were utilized for the development of briquette synthesized from sd and gns blends. 3.2 processing of briquette from blending ratio of groundnut shell and sawdust a modified briquetting machine developed in the university workshop and was employed to conduct preliminary production of fuel briquettes from grinding of raw materials in figure 1, to production of briquettes in figure 2. figure 1. prepared sample of sawdust : groundnut shell : starch ratio; a (90% groundnut shell : 8% sawdust : 2% starch); b (70% groundnut shell : 24% sawdust : 6% starch); c (50% groundnut shell : 40% sawdust : 10% starch) a b c production of fuel briquettes from hybrid waste (blend of saw-dust and groundnut shell) issn: 2180-1053 vol. 10 no.2 june – december 2018 5 the briquetting process entailed the grinding of the groundnut shell and smashing of sawdust. the smashing reduced the sizes of the feedstock into powdery form ranging from 3.5 to 5.5 meshes. the smashed feedstock was properly mixed with waste cooking oil and starch as binder prepared in the ratio 90:8:2, 70:24:6, 50:40:10 by mass percentage which are ratios of sawdust to groundnut shell and starch mixed with waste oil (sawdust: groundnut shell: starch with waste oil). presented in figure 2 is the prepared sample of the feedstock blending ratios. with adequate proportion of water added to the content, the briquetting operation continued as the mixture rolled into the mixing chamber, then to the barrel through the connector to the mould. the mould shaped the mixture into the desired briquette architecture and moved them directly below the hydraulic press to compress the briquettes while the heating element dry up the wet briquettes appropriately as they roll out of the chamber. good quality briquettes were obtained thereafter, see figure 2. figure 2. solid briquettes from the blending ratios. a, b and c are 90:8:2, 70:24:6 and 50:40:10 4.0 statistical analysis of briquetting production the coded variables for the design of the experiment for briquette production from sawdust and groundnut shell with starch as additive are presented in the table 1. mixtures were prepared according to the parameters provided in table 1 which was effectively computed using rsm software. the polynomial/quadratic equation of the response models is highlighted by equations [1] and [2]. where ρ is the density of the briquette produced (kg/m 3 ), a, b and c are the coded experimental input for saw dust: groundnut shell (a), starch amount (b) and compaction time (c), respectively. the a b c journal of mechanical engineering and technology 6 issn: 2180-1053 vol. 10 no.2 june – december 2018 quadratic model is obtained to estimate the density of the briquette including all experimental parameters. the variation of experimental density and predicted density is depicted by figure 3 and figure 4. the contours with response surfaces also generated by the model are shown in figures 5 for all experimental variables. table 1. design with briquetting machine and density from experiments as predicted by response surface methodology. actual a coded a actual b coded b actual c coded c experimental density predicted density residual saw dust or ground nut shell starch quantity compaction time wt. % min (kg/m 3 ) (kg/m 3 ) 8 -1 2 -1 10 -1 116.1 116.11 -0.007 40 1 2 1 10 -1 170.5 170.51 -0.007 8 -1 10 1 10 -1 155.7 155.71 -0.007 40 1 10 -1 10 1 179.6 179.61 -0.007 8 -1 2 -1 50 1 156.0 156.01 -0.007 40 1 2 1 50 1 112.9 112.91 -0.007 8 -1 10 1 50 1 278.1 278.11 -0.007 40 1 10 0 50 0 204.5 204.51 -0.007 8 -1 6 0 30 0 210.2 210.17 0.027 40 1 6 -1 30 0 200.6 200.57 0.027 24 0 2 1 30 0 93.6 93.57 0.027 24 0 10 0 30 -1 159.2 159.17 0.027 24 0 6 0 10 1 113.2 113.17 0.027 24 0 6 0 50 0 145.6 145.57 0.027 24 0 6 0 30 0 144.7 144.72 -0.018 24 0 6 0 30 0 144.7 144.72 -0.018 24 0 6 0 30 0 144.7 144.72 -0.018 24 0 6 0 30 0 144.7 144.72 -0.018 24 0 6 0 30 0 144.7 144.72 -0.018 24 0 6 0 30 0 144.7 144.72 -0.018 production of fuel briquettes from hybrid waste (blend of saw-dust and groundnut shell) issn: 2180-1053 vol. 10 no.2 june – december 2018 7 figure 3. contour plots and response surface for density of briquette as a function of sawdust-grounut shell and starch. density (kgm -3) a: saw dust: groundnut b : s ta rc h ( w t. % ) 8 .0 0 1 6 .0 0 2 4 .0 0 3 2 .0 0 4 0 .0 0 2 .0 0 4 .0 0 6 .0 0 8 .0 0 1 0 .0 0 116.33 139.49 162.65 162.65 185.80 185.80 208.96 666666 93.18 127.92 162.65 197.38 232.12 d e n s it y ( k g m -3 ) 8.00 16.00 24.00 32.00 40.00 2.00 4.00 6.00 8.00 10.00 a: saw dust: groundnut b: starch (wt.%) journal of mechanical engineering and technology 8 issn: 2180-1053 vol. 10 no.2 june – december 2018 figure 4. contour plots and response surface for density of briquette as a function of sawdust-grounut shell and compaction time. density (kgm -3) a: saw dust: groundnut c : c o m p a c ti o n t im e ( m in ) 8 .0 0 1 6 .0 0 2 4 .0 0 3 2 .0 0 4 0 .0 0 1 0 .0 0 2 0 .0 0 3 0 .0 0 4 0 .0 0 5 0 .0 0 116.33 139.49 162.65 162.65 185.80 185.80 208.96 666666 111.24 142.24 173.25 204.26 235.27 d e n s it y ( k g m -3 ) 8.00 16.00 24.00 32.00 40.00 10.00 20.00 30.00 40.00 50.00 a: saw dust: groundnut c: compaction time (min) production of fuel briquettes from hybrid waste (blend of saw-dust and groundnut shell) issn: 2180-1053 vol. 10 no.2 june – december 2018 9 figure 5. contour plots and response surface for density of briquette as a function of compacttion time and starch. density (kgm -3) b: starch (wt.%) c : c o m p a c ti o n t im e ( m in ) 2 .0 0 4 .0 0 6 .0 0 8 .0 0 1 0 .0 0 1 0 .0 0 2 0 .0 0 3 0 .0 0 4 0 .0 0 5 0 .0 0 116.33 139.49 162.65 666666 73.67 100.38 127.09 153.80 180.52 d e n s it y ( k g m -3 ) 2.00 4.00 6.00 8.00 10.00 10.00 20.00 30.00 40.00 50.00 b: starch (wt.%) c: compaction time (min) journal of mechanical engineering and technology 10 issn: 2180-1053 vol. 10 no.2 june – december 2018 4.1 optimisation of modified quadratic density model the predicted density obtained from the optimization technique of design expert 6.06 is shown in table 2. the optimized values ranges from 93.6 to 278 kg/m 3 at various variable parameters. the predicted density is 4.87 kg/m 3 . the estimated optimization values as a result of the experimental parameter is depicted in table 3. 4.2 characterisation of briquette produced the table 4 shows the characterised properties of a briquette sample. the briquette sample was produced as a result of the values obtained after optimization. = 144.75 – 4.8a+32.8b + 16.2 c 2 + 60.88 a 2 – 18.12 b 2 – 15.12 c 2 – 7.63ab – 24.38 ac + 20.63bc (1) = 114.44 – 8.72a + 16.91b + 33.6 c – 0.238 a 2 1.13 b 2 – 0.038c 2 – 0.12ab – 0.076 ac + 0.26bc (2) table 2. optimization summary for modified yield response model solution number saw dust: ground nut shell (a) starch amount (w/w%) (b) compaction time (min) (c) estimated density of briquette (kg/m 3 ) desirability (% w/w) (%) 1 8.00 9.98 50.00 278.16* 1.00 2 8.74 10.00 50.00 270.96 0.961 3 8.00 10.00 43.93 267.41 0.942 4 8.00 7.26 49.47 252.27 0.860 5 8.00 10.00 32.31 238.98 0.788 6 40.00 9.53 38.66 209.28 0.483 7 40.00 6.25 17.71 200.51 0.442 8 40.00 6.57 15.77 199.75 0.439 *optimal value selected table 3. estimated optimization values of experimental parameters experiment saw dust: ground nut shell starch amount (w/w%) compaction time (min) experimental density (kg/m 3 ) predicted density (kg/m 3 ) % error 1 8.00 10 50 5 4.87 0.2 production of fuel briquettes from hybrid waste (blend of saw-dust and groundnut shell) issn: 2180-1053 vol. 10 no.2 june – december 2018 11 table 4. characterised properties of the produced briquettes sample number density (g/cm 3 ) ash content (%) moisture content (%wb) bulk density (g/cm 3 ) durability burning rate (g/min) ignition time (sec) 1 281.6 25.27 4.65 3.012 0.84 0.43 35.27 2 274.2 25.30 4.7 3.045 0.88 0.48 29.78 mean 277.9 25.29 4.68 3.03 0.86 0.46 32.53 standard deviation 0.01 0.02 0.03 0.02 0.02 0.03 2.75 5.0 discussions 5.1 physical properties of sawdust-groundnut shell blend the physical properties of the feedstock taken from the individual blend ratio shows that moisture content increased with an increase in starch percentage from 4.18 to 11.62%. the ash content also increased with an increase in sawdust and groundnut shell ratio from 36.02 to 48.68%. bulk density of the sawdust shifted from 3.77 to 3.85g/cm 3 as the starch percentage increased from 2 to 10%. the second degree order equations were found adequate to correlate the physical properties (moisture content, ash content and bulk density) with the starch content. the regression coefficient the high r 2 (1) implies that 100% of the physical characterisation was captured by the regression model equation. 5.2 response surface methodology the rsm results shows that the maximum density ranges from 93.6 to 278 kg/m 3 . the input variables for briquette fuel for minimum density were 2% starch, sawdust/groundnut shell quantity of 24 and compaction time of 30 minutes while the maximum density of briquette was obtained at 10% starch, sawdust/groundnut shell quantity of 10% and a compaction time of 50 minutes. the contour plots and response surface graph shows sawdust-groundnut ratio varies within the range of 8 to 40 for density and starch fraction was varied from 2 to 10% with a briquetting density of 116.33 to 208.96 kg/m 3 . in this present study, the optimum process was obtained at a ratio 24 for sawdust-groundnut shell at 6% of starch content, this was observed to be decreasing beyond 6% of starch and waste oil. 5.3 optimization of response parameters the numerical optimization of the modified density produced set of optimised solution for the maximization of density within experimental domain of sawdust-groundnut ratio (8), starch content (9.98wt%) and compaction time (50 minutes) with a predicted density of 278.16 kg/m 3 was chosen as the optimised parameter point of variables. the validation test using the optimised experimental variable yielded an experimental density journal of mechanical engineering and technology 12 issn: 2180-1053 vol. 10 no.2 june – december 2018 of 5 kg/m 3 . the average error between the predicted and experimental variable was found to be 2%. the model was accurate, since the percentage error was in good agreement. 6.0 conclusions at the end of this study, solid fuel briquette was produced from waste hybrid materials (sd and gns) using the improved briquetting machine developed at federal university of petroleum resources workshop in delta state, nigeria. the raw material is a blend of sawdust and groundnut shell with starch and waste oil as binder. the production of the desired briquette was actualized by effective determination of the physical properties of the feedstock. response surface methodology (rsm) was used to find the range of values necessary to obtain quality briquettes from hybrid agro-waste. the briquettes produced from these values are of high quality with low moisture content and faster ignition time which symbolises a good briquette. finally, this research can be expanded further via mass production. the briquette produced is: environmentally friendly; a good source employment opportunity; raise the standard of living of youths especially in rural area of nigeria and the world at large. this will also reduce the expenditure on electricity as an alternative source of energy for cooking and warming the rooms in cold region. 6.0 references abakr a.a., a. e. abasaeed (2006) experimental evaluation of a conical-screw briquetting machine for the briquetting of carbonized cotton stalks in sudan. journal of engineering science and technology. 1 (2) 212220 adetogun, a. c. (2014). density of briquettes produced from bambara groundnut shells and it ’ s binary and tertiary combinations with rice husk and peanut shells, journal of natural sciences research, 4(24). ajobo, j. a. (2014). densification characteristics of groundnut shell. international journal of mechanical and industrial technology, 2(1), 150–154. anggono w., sutrisno, suprianto f.d.; j. evander. (2017) biomass briquette investigation from pterocarpus indicus leaves waste as an alternative renewable energy. iop conf. series: materials science and engineering 24, 10 12-43. arinola b. ajayi, justina i. osumune. (2013) design of sawdust briquette machine. innovative systems design and engineering, 4 (10), brand m.a., jacinto r.c., antunes r., cunha a.b. (2017), production of briquettes as a tool to optimize the use of waste from rice cultivation and industrial processing. renewable energy, 111, 116-123 production of fuel briquettes from hybrid waste (blend of saw-dust and groundnut shell) issn: 2180-1053 vol. 10 no.2 june – december 2018 13 chinyere d.c., asoegbu s.n., nwadikom g.i. an evaluation of briquettes from sawdust and corn starch binder. the international journal of science & technology, 2(7), 149-157 davies r.m. and d.s. abolude (2013). ignition and burning rate of water hyacinth briquettes. journal of scientific research & reports 2(1): 111-120, article no. jsrr.009 ighodalo o.a., zoukumor k., egbon c., okoh s., and odu k. (2011) processing water hyacinth into biomass briquettes for cooking purposes. journal of emerging trends in engineering and applied sciences, 2 (2), 305-307 kiss, i., and alexa, v. (2015). short introspections regarding sawdust briquetting. journal of engineering technology, 8(2), 72–79. kuti, o.a (2009). performance of composite sawdust briquette fuel in a biomass stove under simulated condition. au j.t. 12(4): 284-288 lubwama m., yiga v.a. (2018) characteristics of briquettes developed from rice and coffee husks for domestic cooking applications in uganda. renewable energy 118, 43-55 lubwama m., yiga v.a. (2017) development of groundnut shells and bagasse briquettes as sustainable fuel sources for domestic cooking applications in uganda. renewable energy, 111, 532-542 nwabue f.i., u. unah, e.j. itumoh (2017). production and characterization of smokeless bio-coal briquettes incorporating plastic waste materials. environmental technology & innovation, 8, 233–245 okello c., pindozzi s., faugno s., boccia l. (2013) development of bioenergy technologies in uganda: a review of progress. renew. sustain. energy rev, 18, 55-63. olorunnisola a. (2007). production of fuel briquettes from waste paper and coconut husk mixtures. agricultural engineering international: the cigr e. journal. manuscript ee 06 006 (4). oyelaran, o. a., bolaji, b. o., waheed, m. a., & adekunle, m. f. (2015). characterization of briquettes produced from groundnut shell and waste paper admixture. iranica journal of energy and environment, 6(1), 34–38. raymer a.k.p. (2006). a comparison of avoided greenhouse gas emissions when using different kinds of wood energy biomass bioenergy, 30 (7) 605-617. rajkumar d. and venkatachalam p. (2013) physical properties of agro residual briquettes produced from cotton, soybean and pigeon pea stalks. international journal on power engineering and energy. (4) 414-417 journal of mechanical engineering and technology 14 issn: 2180-1053 vol. 10 no.2 june – december 2018 rehfuess e., (2006) fuel for life. household energy and health, world health organization, geneva. switzerland. rezania s., din m., kamaruddin s.f., taib s.m., singh l., yong e.l., dahalan f.a. (2016) evaluation of water hyacinth (eichhornia crassipes) as a potential raw material source for briquette production. energy, 111, 768-773 rezania; m. f. m. din, s. f. kamaruddin, s. m. taib, l. singh, e. l. yong, f. a. dahalan (2017). corrigendum to evaluation of water hyacinth (eichhornia crassipes) as a potential raw material source for briquette production. journal of energy, 111, 768-773. sisman c.b. and gezer e. (2013). effects of rice husk ash on characteristics of the briquette produced for masonry units. scientific research and essays, 6(4), pp. 984-992 supatata n., buates j., and hariyanont p. (2013) characterization of fuel briquettes made from sewage sludge mixed with water hyacinth and sewage sludge mixed with sedge. international journal of environmental science and development, 4(2). tembe, e. t., otache, p. o., & ekhuemelo, d. o. (2014). density, shatter index, and combustion properties of briquettes produced from groundnut shells, rice husks and saw dust of daniellia oliveri. journal of applied biosciences, 82(1), 7372– 7378. thao p.t.m., kurisu k.h., k. hanaki (2011) greenhouse gas emission mitigation potential of rice husks for an giang province. vietnam biomass bioenergy, 35 (8) 3656-3666. vargas-moreno j.m., a.j. callej_on-ferre, j. p_erez-alonso, b. vel_azquez-martí (2012) a review of the mathematical models for predicting the heating value of biomass materials. renew. sustain. energy rev, 16 (5) 3065-3083. wu s., s. zhang, c. wang, c. mu, x. huang (2018). high-strength charcoal briquette preparation from hydrothermal pretreated biomass wastes. fuel processing technology, 171, 293–300 yank a., m. ngadi, r. kok (2016) physical properties of rice husk and bran briquettes under low pressure densification for rural applications. biomass bioenergy, 84, 22-30 issn: 2180-1053 vol. 10 no.2 june – december 2018 77 finite element analysis of spreader bar by utilizing the arrangement and connection of padeyes che zaid bin zakaria 1 * and nuraini abdul aziz 1 1 department of mechanical and manufacturing, faculty of engineering, universiti putra malaysia, malaysia phone: +60389464382; fax: +603 86567122 abstract heavy lifting is one of several methods used for marine installation of heavy equipment while spreader bar (sb) is widely used in heavy lifting. the application of sb is mainly to avoid an overstress in the structure when being lifts which due to sling arrangement in bridle. sb is typically made of high strength tubular pipe with padeye/trunnion attached. comparison between 3 types of padeye arrangements on sb is made based on its strength properties as reflected in api rp 2a 22 nd edition to ensure its optimum design centred on material’s weight and welding work criteria. the buckling load for lightest pipe among 3 types of sb is then calculated. finite element analysis (fea) is performed to verify design stresses and buckling load of selected pipe. from observation, the thickness of tubular pipe can be reduced up to 50 percent compared to other sb types by setting the centre line (cl) of upper padeye to be in line with tubular pipe axis. keywords: spreader bar, heavy lifting, euler stress, padeye, finite element 1.0 introduction nowadays, major drilling and production facilities of oil and gas industry are located offshore and thus installing this equipment necessitate efficiency and safety. the industry has developed a number of ways to overcome heavy lift challenges through experience and innovation. heavy lifting is among main methods of marine installation for heavy equipment. the conventional way to install major facilities, such as topsides and production equipment, is through heavylift vessel (www.rigzone.com). heerema's thialf and saipem's s7000, (by then renamed) were upgraded such that the combined lifting capacity of two cranes on each vessel is 14200 tonnes and 14000 tonnes respectively. the balder and hermod semi-submersible crane vessels (sscv) were each fitted with two enormous cranes. s7000 is well-known for holding world record for an actual lift of 12150 tonnes and for lifted 9500 tonnes jacket in 2007 for pemex in dynamic positioning mode. also, thialf has a staggering lifting capacity of 14200 tonne *corresponding author e-mail: chezaid@gmail.com journal of mechanical engineering and technology 78 issn: 2180-1053 vol. 10 no.2 june – december 2018 or equivalent to weight of more than 1180 fully laden london buses (offshoretechnology.com). during heavy lifting, the structure is subjected to higher stress due to its self-weight and dynamic load from variation of hoisting speeds, crane, vessel motions, cargo barge movements, object movements and others. (dnv vmo, 2014). thicker and higher strength of steels are used in designing installation aids for heavy lifting. the use of slings in bridle arrangement configuration will induce stress in the structure. this additional stress can be eliminated using sb. by utilizing sb to vertically line up the sling on top of structure’s lifting point, this will allow a straight pull movement. sb is a structure designed to resist compression forces induced by angled slings, by altering line of action’s force on lift point into the vertical plane. (gl nobel denton, 2013). tubular pipe is commonly used in designing sb, as shown in figure 1, due to its constant properties in any sectional direction if compared to i beam which have a combination of strong and weak axis. there are two types of sb padeye design usually found in offshore lifting (figure 2) where the differences are on bottom padeye’s design. where, bottom part of type 2 padeye is fabricated in form of delta plate whereas bottom of type 1 padeye is sharing same plate with its top. type 1 is using less material and welding filler because of simplicity of lifting point (padeye) connection design. though by sharing a same plate, bottom shackle’s need to be de-rated (depend on manufacture) since side loading applied to the shackle will caused larger size of shackle (figure 3). increased size of shackle will lead to larger plate needed and stimulate weight gain of lifting system, whereas type 2 will require more material to be used in fabricating delta plate. by introducing delta plate, top padeye will need to be shifted away from bottom padeye to avoid clashes between top shackle and delta plate. this arrangement will result in a bigger in-plane moment hence thicker pipe need to be used. spreader bar (sb) upper padeyes lower padeyes delta plate upper sling lower sling stiffener plate figure 1. typical rigging arrangement using single sb figure 1. common sb padeye design finite element analysis of spreader bar by utilizing the arrangement and connection of padeyes issn: 2180-1053 vol. 10 no.2 june – december 2018 79 figure 3. position of shackle at spreader beam (product datasheet, http://www.nli.no) the plate used to fabricate padeye is made from through-thickness (wang, et al., 2015) property material to avoid failure due to lamellar tearing during lifting. lamellar tearing is a separation in parent or based metal caused by through-thickness strains. such strains are induced primarily by weld metal shrinkage under conditions of high strain. when detected, lamellar tearing can result in often difficult and costly repairs and subsequent construction delays (ship structure committee, 1979). furthermore, required type of steel is barely available in the market therefore require to be procure within at least 3 months in advance, depending on its availability. therefore, if any defect detected during fabrication, schedule of offshore installation campaign will be drag forward until procurement of material is completed. also, the limitation on rigging weight due to operational/material handling issue or as per client’s requirement on “not to exceed weight” need to be considered. hence, designing sb using relatively thinner material for weight reduction and lessen welding work to reduce risk of defect in welding is essential. in this study, an advanced sb padeye design (type 3) and comparison analysis with other two common sb padeye design is introduce where main plate of padeye (figure 4) is slotted through tubular pipe in order to get maximum lifting capacity by transferring load through weld at joint. tubular joint without stiffener inside the chord will cause chord to experience punching stress thus reducing its capacity. therefore, slot in connection is ideal for designing connection between lifting point and tubular pipe. lifting point on sb is installed right above lifting point of module to ensure that vertical pull conditions can be achieve without overstress due to sling arrangement occurred in lifted structure. upper shackles lower shackles journal of mechanical engineering and technology 80 issn: 2180-1053 vol. 10 no.2 june – december 2018 figure 2. new spreader bar padeye design 2.0 material and methodology 2.1 material for the purpose of this study, sbs are design to lift the 19620n structure. lifting arrangement is shown in figure 1. the length of sb is fixed to 800mm based on distance between two lifting point of structure and dimensions of sb’s geometry is detailed in table 1 below. table 1. geometry details for each sb items type 1 type 2 type 3 outer diameter (mm) 42.40 48.30 26.9 wall thickness (mm) 2.60 4.90 2.0 cross sectional area(mm 2 ) 325.10 668.10 156.45 slenderness ratio kl/r 56.73 51.81 120.77 total weight (n) 54.76 70.49 26.33 material properties selected for fabrication of plates and tubular are mild steel which in compliance to astm a36 for plate and astm a106 grade b for pipe. calculations are based on minimum yield strength of relevant material grade as specified in table 3. table 2. material properties density in air 7.74e-5 n/mm 3 young’s modulus, e 210000 mpa shear modulus, g 80000 mpa poisson’s ratio 0.3 tubular pipe length of sb, l a a view a-a (cross sectional area) outer diameter wall thickness finite element analysis of spreader bar by utilizing the arrangement and connection of padeyes issn: 2180-1053 vol. 10 no.2 june – december 2018 81 table 3. strength of material steel grade yield stress (mpa) tensile strength (mpa) remark astm a36 248 400-550 for lifting point (padeye) astm a106 grade b 240 415 min. spreader 2.2 tubular design criteria mathematical modelling force in horizontal component produce a compressive stress on tubular and assume to be matched with euler’s theory of buckling. euler formula is derived for an ideal or perfect column which the theory is simple enough to be applied. though, the formula is constructs on a couple of assumptions that rarely comply with real conditions as highlighted below (mckenzie, 2006):  the compression load acts through absolute centre of columns cross sectional area.  the column is completely a long, slender, straight and homogeneous even before concentric axial compressive load is applied. slenderness is defined as the ratio between height and cross-sectional dimensions of column. slender columns which subject to buckling will produces additional moment resulting in significant reduction of column capacity.  the column’s material is elastic and follows hooke’s law.  there are no imperfections in the column.  lateral deflections of the column are small compared to overall length (the column’s displacement is small).  the column is pin-jointed at each end and restrained against lateral loading.  there are no residual stresses in the column.  there is no strain hardening on the material. since self-weight of sb is relatively too small compared to sling load, it is possible to neglect the weight from here. common design of sb normally produced moment due to padeye eccentricity as shown in figure 5. figure 3. free body diagram for sb type 1 and 2 journal of mechanical engineering and technology 82 issn: 2180-1053 vol. 10 no.2 june – december 2018 total moment at a point x from either end (boundary condition) gives: 𝑀(𝑥) − 𝑃𝑦 − 𝐹𝑥 = 0 0 ≤ 𝑥 ≤ 𝑎 (1) 𝑀(𝑥) − 𝑃𝑦 − 𝐹𝑎 = 0 𝑎 ≤ 𝑥 ≤ (𝐿 − 𝑎) (2) 𝑀(𝑥) − 𝑃𝑦 − 𝐹𝐿 + 𝐹𝑥 = 0 (𝐿 − 𝑎) ≤ 𝑥 ≤ 𝐿 (3) where: m = total moment, p = axial load, f = vertical load p and f are derived from calculated sling load based on rigging arrangement as shown in figure 1. from equation (1), (2) and (3), it shows that sb is needed to resist axial compressive force and bending moment. thus, figure 5 shows the eccentricity on geometry that generates moment in sb where the bending moment created will reduced the capacity of sb. any force that applied to sb at neutral axis resulted in a purely compression force in tubular pipe. therefore, forces that are not lined up with neutral axis generate bending force or bending moment. sb that is subjected to bending forces and/or bending moments is more difficult to be properly design and will no longer be simple and light weight construction as preferred (heavyliftnews.com). to eliminate eccentricity, padeye’s top is moved to cl of tubular pipe and left euler buckling as an only mode of failure for sb. this concept is illustrated in figure 6. figure 4. free body diagram for new sb design when lateral displacement is y, summation of moments on beam section is (chen et al., 1999): 𝑀 + 𝑃𝑦 = 0, 𝑤ℎ𝑒𝑟𝑒 𝑀 = 𝐸𝐼 𝑑2𝑦 𝑑𝑥2 , 𝑡ℎ𝑢𝑠 𝐸𝐼 𝑑2𝑦 𝑑𝑥2 + 𝑃𝑦 = 0 (4) solution for equation for is 𝑦 = 𝐵𝑠𝑖𝑛 (𝑘𝑥) (5) by solving equation (4), smallest value of p is known as critical load, buckling load, or euler formula: 𝑃𝐸 = 𝑃𝑐𝑟 = 𝜋2𝐸𝐼 𝐿2 = 𝜋2𝐸 ( 𝑘𝐿 𝑟 ) 2 (6) according to api rp2a wsd 22 nd edition, allowable axial compressive stress, 𝐹𝑎 must be determined from the following aisc formulas for members with d/t ratio or less than 60. in the (aisc asd, 9 th edition) equations for allowable compressive stresses, various imperfections such as effect of residual stresses, actual end restraint conditions, crookedness, and small unavoidable eccentricities are empirically taken into account. finite element analysis of spreader bar by utilizing the arrangement and connection of padeyes issn: 2180-1053 vol. 10 no.2 june – december 2018 83 the api code which is based on aisc 9th edition requirement assumes arbitrarily that the elastic buckling holds valid when stress in the column is not greater than one-half of the yield stress (fy/2) (j.s.arora, http://user.engineering.uiowa.edu). for column having effective length less than cc, it is assuming the failure by crushing of the material induced by predominantly axial compressive stresses. failure occurs when stress over cross-section reaches yield or crushing value for the material. figure 5: variation of critical stress and allowable stress as specified by the api code 3.0 finite element analysis finite element analysis is widely used in offshore industry to design offshore and subsea structure. all structures were modelled using space claim (ansys package). in spaceclaim, spreader bar components i.e tubular pipe, main plate of padeye, cheek plate and stiffener are group together and shared topology is activated. shared topology occurs is triggered when bodies are grouped into multibody parts. it allows for a continuous mesh across common regions where bodies touch, instead of having to define contact regions in the ansys workbench. these bodies share topology in the region where they are in contact with, so the mesh is continuous across part as shown in figure 8. it is often, but not always, more desirable for analysis to have a continuous mesh across parts than to use contact (ansys 14.5 user guide). 𝑘𝐿 𝑟 𝐶𝑐 0.5𝐹𝑦 𝐹𝑦 𝐸𝑞. (6) 𝐸𝑞. (8) 𝐸𝑞. (7) 𝐹𝑎 [1 − ( 𝑘𝑙 𝑟 ) 2 2𝐶𝑐 2 ] 𝐹𝑦 journal of mechanical engineering and technology 84 issn: 2180-1053 vol. 10 no.2 june – december 2018 figure 6. typical details of sb components and continuous meshing across the parts figure 7. solid186 homogeneous structural solid geometry 3 x 3 x 2 x 4 x 1 x 1 x 4 x legend: 1-tubular pipe 2padeye-main plate 3. padeye-cheek plate 4. stiffener plate finite element analysis of spreader bar by utilizing the arrangement and connection of padeyes issn: 2180-1053 vol. 10 no.2 june – december 2018 85 the model then exported to ansys for assignment of materials and contact details. the meshing is performed with element size of 5mm as shown in figure 9 and solid186 element is assigned to the geometry. solid186 is a higher order 3-d 20node solid element that exhibits quadratic displacement behaviour where the element is defined by 20 nodes having three degrees of freedom per node: translations in the nodal x, y, and z directions. the geometry, node locations, and element coordinate system for this element are shown in figure 9. a prism-shaped element may be formed by defining the same node numbers for nodes k, l, and s; nodes a and b; and nodes o, p, and w. a tetrahedral-shaped element and a pyramid-shaped element may also be formed (ansys 14.5 user guide) as shown in figure 9. summary of mesh statistic for each type of sb is as shown in table 4. the static structural analysis is performed to determine the stress level for imposed load. then the data is exported to linear buckling module to find critical buckling load for each type of sb. table 4. mesh statistic for each type of tubular pipe item type 1 type 2 type 3 nodes 12251 124060 74675 elements 8766 25018 28507 mesh metric (average aspect ratio) 4.78 4.19 2.83 bearing load is then assigned to the surface of upper padeye’s pinhole representing dynamic sling load (dsl) applied to sb as shown in figure 10, 11 and 12 for sb type 1, 2, and 3 respectively. as defined by ansys, the bearing load simulates radial forces only and applied on interior of cylinder in the radial direction by using a coordinate system (ansys 14.5 user guide). in addition, torsion is imposed to the surface of pinhole for upper and bottom padeye to distributes moment "about" (the vector of) an axis curved faces where right-hand rule is applied to determine sense of moment. remote displacement applied to the surface of bottom padeye’s pinhole as a boundary condition allows displacements and rotations application at an arbitrary remote location in space. the origin of remote location can be specified under scope in details view by selecting or entering the xyz coordinates. the default location is at the centroid of the geometry. these remote boundary conditions are all based on the use of a remote point, be it created by the boundary condition itself, or by being scoped to remote point object (ansys 14.5 user guide). details of loading are as specified in table 5. table 5. detail of loading condition for each of sb items type 1 type 2 type 3 horizontal force (n) 10035.39 9679.53 10498.02 vertical force (n) 18149.11 30248.52 30248.52 lateral force (n) 1036.94 1587.98 1600.92 journal of mechanical engineering and technology 86 issn: 2180-1053 vol. 10 no.2 june – december 2018 note: loadings are derived from rigging arrangement as shown in figure 1. the maximum loading in upper sling is separated into horizontal (parallel to longitudinal axis of sb) and vertical component. latera load is assuming 5 percent of dsl. figure 8 (a). loading and boundary condition applied to the sb type 1 figure 10(b). detail of loading condition and boundary condition applied to the sb type 1 finite element analysis of spreader bar by utilizing the arrangement and connection of padeyes issn: 2180-1053 vol. 10 no.2 june – december 2018 87 figure 9 (a). loading and boundary condition applied to the sb type 2 figure 11(b). detail of loading and boundary condition applied to the sb type 2 journal of mechanical engineering and technology 88 issn: 2180-1053 vol. 10 no.2 june – december 2018 figure 10 (a). loading and boundary condition applied to the sb type 3 figure 12(b). detail of loading and boundary condition applied to the sb type 3 once stress result is found, linearization of stress is performed. the linearized stress results calculate membrane, bending, peak, and total stress along a straight-line path in the ansys workbench. when result is evaluated (stress linearization), component stress values at the path points are interpolated from appropriate element's average corner nodal values. stress components through the section are linearized by a line integral method and are separated into constant membrane stresses, bending stresses varying linearly between end points, and peak stresses (defined as differences between actual (total) stress and membrane plus bending combination). the details view shows membrane, bending, membrane + bending, peak, and total stresses. the bending stresses are calculated such that neutral axis is at midpoint of the path. principal stresses are recalculated from the component stresses and are invariant with the coordinate system as long as stress is in the same direction at all points along the defined path (ansys 14.5 user guide). finite element analysis of spreader bar by utilizing the arrangement and connection of padeyes issn: 2180-1053 vol. 10 no.2 june – december 2018 89 4.0 experiment the objective of the testing program is to determine critical buckling loads for tubular under compressive force which conducted according to astm e9. only tubular pipe for sb type 3 is used in the experiment. in total, 5 specimens are used in this testing. figure 11. experiment set-up 5.0 results and discussions the tubular size selected as in table 1 is based on loading applied to the sb structure. due to huge moment for sb type 2, bigger diameter needs to be selected. total weight of sb type 1 and 2 are more than 50 percent heavier than sb type 3. the weight ratios (sb’s weight/weight of lifted structure) are 0.30, 0.41 and 0.15 percent for sb type 1, 2 and 3 respectively. the differences in sling loads for sb type 1 and 2 are due to the eccentricity of padeyes (figure 5) where eccentricity is varying on padeye’s arrangement (figure 2). for sb type 1 and 2, the eccentricities are 10mm and 30mm respectively. no eccentricity should be considered for sb type 3 since the cl of top padeye lay at similar line with cl of tubular pipe. sb type 2 having a biggest eccentricity is since delta plate arrangement requires it to be installed far away from upper padeye to ensure that it will not clash with shackle that was installed at the top of padeye. therefore, padeye type 2 produce highest moment among other types of sb thus require thickest material in design. the buckling load as specified in equation (6) is depending upon geometry and elastic modulus of column and not upon the strength of it. however, in aisc asd, 9 th edition, the equations for allowable compressive stresses, various imperfections such as effect of residual stresses, the actual end restraint conditions, crookedness, and small unavoidable eccentricities are empirically taken into account. to get a buckling stress, buckling load is divided by area of cross section. considering that buckling stress is found as above, it is noted that allowable compressive stress is fall beyond elastic region for type 1 and 2 journal of mechanical engineering and technology 90 issn: 2180-1053 vol. 10 no.2 june – december 2018 of the sb therefore, buckling load did not applied since excessive yielding occurs before reaching the buckling. for type 1, maximum moment allowed is 5.49e+05n.mm which equivalent to 75 percent of tubular yield strength. design moment is 2.24e+05n.mm which at 41 percent of allowable moment (based on api 2a stress criteria). the maximum moment allowed is 1.19e+06n.mm which equivalent to 75 percent of pipe strength. design moment is 9.93e+05n.mm which at 84 percent of allowable moment. therefore, by using sb type 1, 31 percent of the pipe strength is used to resist bending moment while for sb type 2, 63 percent of the pipe strength is used to resist bending moment. fea is performed to obtain the von-mises stress. the stress linearization is performed for each type of sb to separate primary (structural) and secondary (geometry) stress. figure 12 (a). von-mises stress for sb type 1 figure 14(b). stress linearization for sb type 1 finite element analysis of spreader bar by utilizing the arrangement and connection of padeyes issn: 2180-1053 vol. 10 no.2 june – december 2018 91 figure 14(c). buckling load for sb type 1 figure 13 (a). von-mises stress for sb type 2 figure 15(b). stress linearization for sb type 2 journal of mechanical engineering and technology 92 issn: 2180-1053 vol. 10 no.2 june – december 2018 figure 15(c). buckling load for sb type 2 figure 14 (a). von-mises stress for sb type 3 figure 16 (b). stress linearization for sb type 3 finite element analysis of spreader bar by utilizing the arrangement and connection of padeyes issn: 2180-1053 vol. 10 no.2 june – december 2018 93 figure 16(c). buckling load for sb type 3 as shown in table 6, mathematical modelling for value of stress is consistent with fea analysis. table 6. von-mises stress on tubular sb type von-mises stress (mpa) theory fea percentage of different (%) type 1 106.80 103.59 3 type 2 170.43 167.22 1.9 type 3 85.82 84.5 1.5 the experiment results for 5 specimens of pipe used to design sb type 3 shows that the pipe will be buckled at an average load of 23447.12n. table 7. buckling load for specimens of sb type 3 specimen buckling load (n) s1 23781.80 s2 22128.31 s3 24368.38 s4 26203.11 s5 20754.02 journal of mechanical engineering and technology 94 issn: 2180-1053 vol. 10 no.2 june – december 2018 figure 15. buckling load for sb type 3 (experiment) table 8. buckling load and stress for tubular sb buckling load (n) buckling stress (mpa) yield stress (mpa) theory fea experiment theory fea experiment type 1 209349.03 210642.90 na 643.97 647.95 na 240 type 2 515899.66 590393.07 na 772.20 883.70 na 240 type 3 22914.11 21520.93 23347.80 142.10 133.46 149.23 240 table 9. axial stress for tubular sb type design axial stress (mpa) allowable axial stress (mpa) type 1 30.87 119.68 type 2 14.49 122.51 type 3 65.10 72.44 finite element analysis of spreader bar by utilizing the arrangement and connection of padeyes issn: 2180-1053 vol. 10 no.2 june – december 2018 95 result as shown in table 9 indicate that for sb type 1, the utilization of axial stress is only 26 percent of allowable while type 2 is only 12 percent. remaining tubular pipe strength is used to resist design stress due to bending moment. for sb type 3, the utilization of axial stress is at 90 percent capacity. table 10. details of lifting point geometry on sb padeye characteristic advantage disadvantage type 1 cl of upper and bottom padeyes at a distance from cl of tubular. top and bottom padeye at a horizontal distance. only 2 shackles required. moment due to eccentricity will reduce capacity of tubular. bottom shackle require bigger capacity due to de-rated capacity when pulling at certain angle from vertical axis of shackle. by increasing the size of shackle, the size of padeye will increase to accommodate shackle geometries. type 2 cl of upper padeye at a distance from cl of tubular. upper and bottom padeye at a horizontal distance. end plate using delta plate to connect two shackles at bottom padeye. smaller shackles used for bottom padeye due to shackle capacity are not required to be derated thus reduce the size of padeye. delta plate is installed away from upper padeye to avoid clashed with upper shackle, thus moment due to eccentricity will be higher compared to type 1 and 2 thus reduce the capacity of tubular. 3 shackles required more material required to fabricate delta plate. more welding jobs. type 3 cl of upper padeye at similar elevation of tubular cl. no moment due to eccentricity, therefore reduce the thickness of tubular. only 2 shackles required. padeye main plate must check for axial buckling due to horizontal load. bottom shackle require bigger capacity due to de-rated while pulling at certain angle from vertical axis of shackle. by increasing the size of shackle, padeye’s size will vary to accommodate the shackle geometries. journal of mechanical engineering and technology 96 issn: 2180-1053 vol. 10 no.2 june – december 2018 6.0 conclusions since sb type 1 and 2 are not categorised as long column as shown in figure 7, its failure is mainly caused by excessive yielding instead of buckling. therefore, sb type 1 and 2 is considered as not buckling sensitive. sb type 1and 2 use respective 31 and 63 percent of the pipe strength to resist bending moment which resulting in heavier section needed to be used. weight of sb type 1 and 2 are more than 50 percent heavier compared to sb type 3. due to highest moment induced for sb type 2, it is not recommended for used on heavy lifting since thicker material and relatively more welding works and non-destructive test) ndt are required that may increase chances for cracks in the weldment or plate. the moment induced in sb type 1 and 2 resulted in increased of length slot that require more filler material for welding and therefore increase chances of having crack in the structure. 7.0 acknowledgements the authors would like to thank universiti putra malaysia for providing facility to perform this study. 8.0 references dnv-os-h205lifting operations (vmo standard part 2-5), april 2014 gl nobel denton, guideline for marine lifting & lowering operations, june 2013 api rp 2a-wsd 22nd edition recommended practice for planning, designing and constructing fixed offshore platforms-working stress design, december 2014. ship structure committee, significance and control of lamellar tearing of steel plate in the shipbuilding industry, 1979. william m.c.mckenzie examples in structural analysis, taylor & francis e-library, 2006 taylor & francis, 2 park square, milton park, abingdon, oxon ox14 4rn. the knowledge: spreader bar design and application. retrieved from http://www.heavyliftnews.com/news/the-knowledge--spreader-bar-design-andapplication. (access on 14-aug-2015) rigzone. training. how does heavy lift work?. retrieved from http://www.rigzone. com/training/insight.asp?insight_id=321&c_id=20 (access on 15-jun-2015) passionate about… cranes and lifting. retrieved from http://www.offshoretechnology.com/features/feature1538/ (access on 15-jun-2015) chen w.f. and duan, l. “effective length factors of compression members”, structural engineering handbook, boca raton: crc press llc, 1999 http://www.heavyliftnews.com/news/the-knowledge--spreader-bar-design-and-application http://www.heavyliftnews.com/news/the-knowledge--spreader-bar-design-and-application http://www.offshore-technology.com/features/feature1538/ http://www.offshore-technology.com/features/feature1538/ finite element analysis of spreader bar by utilizing the arrangement and connection of padeyes issn: 2180-1053 vol. 10 no.2 june – december 2018 97 lifting arrangement. retrieved from http://www.nli.no/index.php?groupid=1831 (access on 15-jun-2015) wang yuan-qing, liao xiao-wei, zhang yuan-yuan & shi yong-jiu (2015). experimental study on the through-thickness properties of structural steel thick plate and its heat-affected zone at low temperatures. journal of zhejiang universityscience a (applied physics & engineering),16(3), 217-228. j.s.arora,http://user.engineering.uiowa.edu/~design1/structuraldesignii/compresiondesi gn.pdf (access on 15-jun-2015) ansys workbench 14.5 –user guide http://www.nli.no/index.php?groupid=1831 experimental investigation on performance of single cylinder diesel engine with mullite as thermal barrier coating issn: 2180-1053 vol. 3 no. 1 january-june 2011 45 experimental investigation on performance of single cylinder diesel engine with mullite as thermal barrier coating p. n. shrirao1, a. n. pawar2 1j d college of engineering, yavatmal445001, maharashtra, india 2government polytechnic, yavatmal445001, maharashtra, india email: 1pn_shrirao@yahoo.co.in abstract tests were performed on a single cylinder, four stroke, direct injection, diesel engine whose piston crown, cylinder head and valves were coated with a 0.5 mm thickness of 3al2o3 .2sio2 (mullite) (al2o3= 60%, sio2= 40%) over a 150 µm thickness of nicraly bond coat. the working conditions for the standard engine (uncoated) and low heat rejection (lhr) engine were kept exactly same to ensure a comparison between the two configurations of the engine. this paper is intended to emphasis on energy balance with and without ceramic insulation coating at identical conditions. tests were carried out at same a/f ratio for both standard and low heat rejection engine at different engine load and engine speed conditions. the results showed that there was 1.07 % decreasing on specific fuel consumption value of low heat rejection (lhr) engine compared to standard engine at full load. however, there was as much as 16 % decreasing on heat amount to coolant of lhr engine compared to standard engine at full load. there was as much as 22 % increasing on heat amount to exhaust of lhr engine compared to standard engine at full load. keywords: energy balance, ceramic coating, mullite, lhr, se 1.0 introduction ceramics have a higher thermal durability than metals; therefore it is usually not necessary to cool them as fast as metals. low thermal conductivity ceramics can be used to control temperature distribution and heat flow in a structure (alkidas, 1989 and uzun., 1999). thermal barrier coatings (tbc) provide the potential for higher thermal efficiencies of the engine, improved combustion and reduced emissions. in addition, ceramics show better wear characteristics than conventional issn: 2180-1053 vol. 3 no. 1 january-june 2011 journal of mechanical engineering and technology 46 materials. lower heat rejection from combustion chamber through thermally insulated components causes an increase in available energy that would increase the in-cylinder work and the amount of energy carried by the exhaust gases, which could be also utilized (hejwowski et.al., 2002 and toyama et.al., 1983). a major breakthrough in diesel engine technology has been achieved by the pioneering work done by (kamo et.al., 1978 and kamo et.al., 1979). kamo and bryzik used thermally insulating materials such as silicon nitride for insulting different surfaces of combustion chamber. an improvement of 7% in the performance was observed (kamo et.al., 1978). sekar et.al, 1984 developed an adiabatic engine for passenger cars and reported an improvement in performance to the maximum extent of 12%. the experimental results of (morel et.al., 1985) indicate that the higher temperatures of the insulated engine cause reduction in the incylinder heat rejection, which is in accordance with the conventional knowledge of convective heat transfer. (woschni et.al., 1987) state that 5% of the input fuel energy cannot be accounted for which is of the order of the expected improvements. (havstad et.al., 1986) developed a semi-adiabatic diesel engine and reported an improvement ranging from 5 to 9% in isfc, about 30% reduction in the in-cylinder heat rejection. (prasad et.al., 1990) used thermally insulating material, namely partially stabilized zirconia (psz), on the piston crown face and reported a 19% reduction in heat loss through the piston. among possible alternative materials, one of the most promising is mullite. mullite is an important ceramic material because of its low density, high thermal stability, stability in severe chemical environments, low thermal conductivity and favorable strength and creep behavior. it is a compound of sio2 and al2o3 with composition 3al2o3.2sio2. compared with yttria-stabilized zirconia (ysz), mullite has a much lower thermal expansion coefficient and higher thermal conductivity, and is much more oxygen-resistant than ysz. for the applications such as diesel engines where the surface temperatures are lower than those encountered in gas turbines and where the temperature variations across the coating are large, mullite is an excellent alternative to zirconia as a tbc material. engine tests performed with both materials show that the life of the mullite coating in the engine is significantly longer than that of zirconia (kokini et.al., 1996 and yonushonis, 1997). above 1273 k, the thermal cycling life of mullite coating is much shorter than that of ysz (ramaswamy et.al., 1999). mullite coating crystallizes at 1023–1273 k, accompanied by a volume contraction, causing cracking and de-bonding. mullite has excellent thermo-mechanical behavior; experimental investigation on performance of single cylinder diesel engine with mullite as thermal barrier coating issn: 2180-1053 vol. 3 no. 1 january-june 2011 47 however its low thermal expansion coefficient creates a large mismatch with the substrate (samadi et.al., 2005). to address this problem, a 150 µm thickness of nicraly bond coat was used. table 1 properties of tbc materials (cao et.al., 2004) 38 input fuel energy cannot be accounted for which is of the order of the expected improvements. (havstad et.al., 1986) developed a semi-adiabatic diesel engine and reported an improvement ranging from 5 to 9% in isfc, about 30% reduction in the in-cylinder heat rejection. (prasad et al., 1990) used thermally insulating material, namely partially stabilized zirconia (psz), on the piston crown face and reported a 19% reduction in heat loss through the piston. among possible alternative materials, one of the most promising is mullite. mullite is an important ceramic material because of its low density, high thermal stability, stability in severe chemical environments, low thermal conductivity and favorable strength and creep behavior. it is a compound of sio2 and al2o3 with composition 3al2o3.2sio2. compared with yttria-stabilized zirconia (ysz), mullite has a much lower thermal expansion coefficient and higher thermal conductivity, and is much more oxygen-resistant than ysz. for the applications such as diesel engines where the surface temperatures are lower than those encountered in gas turbines and where the temperature variations across the coating are large, mullite is an excellent alternative to zirconia as a tbc material. engine tests performed with both materials show that the life of the mullite coating in the engine is significantly longer than that of zirconia (kokini et.al., 1996 and yonushonis, 1997). above 1273 k, the thermal cycling life of mullite coating is much shorter than that of ysz (ramaswamy et.al., 1999). mullite coating crystallizes at 1023–1273 k, accompanied by a volume contraction, causing cracking and de-bonding. mullite has excellent thermomechanical behavior; however its low thermal expansion coefficient creates a large mismatch with the substrate (samadi et.al., 2005). to address this problem, a 150 µm thickness of nicraly bond coat was used. table 1 properties of tbc materials (cao et.al., 2004) materials properties mullite melting point (tm) =2123 k thermal conductivity(λ) =3.3 w m-1 k-1 (1400 k) young’s modulus (e)=30 gpa (293 k) thermal expansion coefficient (α) =5.3x10-6 k-1 (293–1273 k) poisson’s number (υ) =0.25 nicraly (bond coat of tbc) young’s modulus (e) =86gpa(293k) thermal expansion coefficient (α) =17.5x10-6k-1 (293–1273k) poisson’s number (υ)=0.3 table 2 thermal conductivity and thermal resistance for mullite at various temperatures (gilbert et.al.,2008) material thermal conductivity (w/m k) thermal resistance rth (k/w) 296k 773k 1273k 296k 773k 1273k mullite 1.05 1.23 1.39 8.8 7.6 6.7 table 2 thermal conductivity and thermal resistance for mullite at various temperatures (gilbert et.al.,2008) 38 input fuel energy cannot be accounted for which is of the order of the expected improvements. (havstad et.al., 1986) developed a semi-adiabatic diesel engine and reported an improvement ranging from 5 to 9% in isfc, about 30% reduction in the in-cylinder heat rejection. (prasad et al., 1990) used thermally insulating material, namely partially stabilized zirconia (psz), on the piston crown face and reported a 19% reduction in heat loss through the piston. among possible alternative materials, one of the most promising is mullite. mullite is an important ceramic material because of its low density, high thermal stability, stability in severe chemical environments, low thermal conductivity and favorable strength and creep behavior. it is a compound of sio2 and al2o3 with composition 3al2o3.2sio2. compared with yttria-stabilized zirconia (ysz), mullite has a much lower thermal expansion coefficient and higher thermal conductivity, and is much more oxygen-resistant than ysz. for the applications such as diesel engines where the surface temperatures are lower than those encountered in gas turbines and where the temperature variations across the coating are large, mullite is an excellent alternative to zirconia as a tbc material. engine tests performed with both materials show that the life of the mullite coating in the engine is significantly longer than that of zirconia (kokini et.al., 1996 and yonushonis, 1997). above 1273 k, the thermal cycling life of mullite coating is much shorter than that of ysz (ramaswamy et.al., 1999). mullite coating crystallizes at 1023–1273 k, accompanied by a volume contraction, causing cracking and de-bonding. mullite has excellent thermomechanical behavior; however its low thermal expansion coefficient creates a large mismatch with the substrate (samadi et.al., 2005). to address this problem, a 150 µm thickness of nicraly bond coat was used. table 1 properties of tbc materials (cao et.al., 2004) materials properties mullite melting point (tm) =2123 k thermal conductivity(λ) =3.3 w m-1 k-1 (1400 k) young’s modulus (e)=30 gpa (293 k) thermal expansion coefficient (α) =5.3x10-6 k-1 (293–1273 k) poisson’s number (υ) =0.25 nicraly (bond coat of tbc) young’s modulus (e) =86gpa(293k) thermal expansion coefficient (α) =17.5x10-6k-1 (293–1273k) poisson’s number (υ)=0.3 table 2 thermal conductivity and thermal resistance for mullite at various temperatures (gilbert et.al.,2008) material thermal conductivity (w/m k) thermal resistance rth (k/w) 296k 773k 1273k 296k 773k 1273k mullite 1.05 1.23 1.39 8.8 7.6 6.7 table 3 tbc materials and their characteristics 39 table 3 tbc materials and their characteristics material advantages disadvantages mullite (1) high corrosion resistance (2) low thermal conductivity (3) good thermal-shock resistance below 1273 k (4) not oxygen-transparent (1)crystallization (1023-1273 k) (2) very low thermal expansion coefficient the main purpose of this study was to evaluate the energy balance at different engine loads and speeds with and without ceramic-coated diesel engine. experiments were conducted with single cylinder, direct injected, inter-cooled diesel engine to evaluate heat losses to oil, ambient and cooling system of ceramic coated engine (lhr). 2.0 experimental setup a four stroke, direct injected, water-cooled, single cylinder, naturally aspirated diesel engine was used for investigation. details of the engine specifications are given in table 4. table 4 engine specifications engine type kirloskar av1, di stroke number 4 cylinder number 1 bore (mm) 80 stroke (mm) 110 compression ratio 16.5:1 maximum engine power (kw) 3.7 maximum engine speed (rpm) 1500 specific fuel consumption (g/kwh) 245 injection timing 20 before top dead center(btdc) static the main purpose of this study was to evaluate the energy balance at different engine loads and speeds with and without ceramic-coated diesel engine. experiments were conducted with single cylinder, direct injected, inter-cooled diesel engine to evaluate heat losses to oil, ambient and cooling system of ceramic coated engine (lhr). 2.0 experimental setup a four stroke, direct injected, water-cooled, single cylinder, naturally aspirated diesel engine was used for investigation. details of the engine issn: 2180-1053 vol. 3 no. 1 january-june 2011 journal of mechanical engineering and technology 48 specifications are given in table 4. table 4 engine specifications 39 table 3 tbc materials and their characteristics material advantages disadvantages mullite (1) high corrosion resistance (2) low thermal conductivity (3) good thermal-shock resistance below 1273 k (4) not oxygen-transparent (1)crystallization (1023-1273 k) (2) very low thermal expansion coefficient the main purpose of this study was to evaluate the energy balance at different engine loads and speeds with and without ceramic-coated diesel engine. experiments were conducted with single cylinder, direct injected, inter-cooled diesel engine to evaluate heat losses to oil, ambient and cooling system of ceramic coated engine (lhr). 2.0 experimental setup a four stroke, direct injected, water-cooled, single cylinder, naturally aspirated diesel engine was used for investigation. details of the engine specifications are given in table 4. table 4 engine specifications engine type kirloskar av1, di stroke number 4 cylinder number 1 bore (mm) 80 stroke (mm) 110 compression ratio 16.5:1 maximum engine power (kw) 3.7 maximum engine speed (rpm) 1500 specific fuel consumption (g/kwh) 245 injection timing 20 before top dead center(btdc) static 40 figure 1 experimental set up the first stage tests were performed at different engine loads. the experiments were conducted at five load levels, viz. 0, 25, 50, 75% of full load and full load. the required engine load percentage was adjusted by using the rope break dynamometer. the second stage concerned an investigation of heat losses when combustion chamber insulation was applied. a piston crown, cylinder head and valves were coated with ceramic material over super alloy bond coating (nicraly). the bond coat was first applied to these engine components to avoid mismatch in thermal expansion between substrate and ceramic material. a piston crown, cylinder head and valves were coated with 0.5 mm coating of mullite is commonly denoted as 3al2o3 .2sio2 (i.e. 60 mol% al2o3). however it is actually a solid solution with the equilibrium composition limits of 60 – 63mol % al2o3 below 1600o c. the ceramic material was coated by using plasma-spray technique. the engine were insulated and tested at baseline conditions to see the effect of insulated surfaces on engine heat losses. the ceramic-coated engine (lhre) was compared to standard engine. 3.0 plasma spray technique figure 2 photographic view of cylinder head, cylinder valves and piston crown after ceramic coating. the gas tunnel type plasma spraying torch was used. the experimental method to produce ceramic coating by means of the gas tunnel type plasma spraying is as follows. after igniting plasma gun, note: 1. engine 2. rope break dynamometer 3. damping box with orifice 4. fuel tank 5. burette with measuring scale 6.water tank 7. rotameter 8. reciprrocating compressor as pressure booster 9. calorimeter figure 1 experimental set up the first stage tests were performed at different engine loads. the experiments were conducted at five load levels, viz. 0, 25, 50, 75% of full load and full load. the required engine load percentage was adjusted by using the rope break dynamometer. the second stage concerned an investigation of heat losses when combustion chamber insulation was applied. a piston crown, cylinder head and valves were coated with ceramic material over super alloy bond coating (nicraly). the bond coat was first applied to these engine components to avoid mismatch in thermal expansion between substrate and ceramic material. a piston crown, cylinder head and valves were coated with 0.5 mm coating of mullite is commonly denoted as 3al2o3 .2sio2 (i.e. 60 mol% al2o3). however it is actually a solid solution with the equilibrium composition limits of 60 – 63mol% al2o3 below 1600 oc. the ceramic material was coated by using plasma-spray technique. the engine were insulated and tested at baseline conditions to see the effect of insulated surfaces on engine heat losses. the ceramic-coated engine (lhre) was compared to standard engine. experimental investigation on performance of single cylinder diesel engine with mullite as thermal barrier coating issn: 2180-1053 vol. 3 no. 1 january-june 2011 49 3.0 plasma spray technique 40 figure 1 experimental set up the first stage tests were performed at different engine loads. the experiments were conducted at five load levels, viz. 0, 25, 50, 75% of full load and full load. the required engine load percentage was adjusted by using the rope break dynamometer. the second stage concerned an investigation of heat losses when combustion chamber insulation was applied. a piston crown, cylinder head and valves were coated with ceramic material over super alloy bond coating (nicraly). the bond coat was first applied to these engine components to avoid mismatch in thermal expansion between substrate and ceramic material. a piston crown, cylinder head and valves were coated with 0.5 mm coating of mullite is commonly denoted as 3al2o3 .2sio2 (i.e. 60 mol% al2o3). however it is actually a solid solution with the equilibrium composition limits of 60 – 63mol % al2o3 below 1600o c. the ceramic material was coated by using plasma-spray technique. the engine were insulated and tested at baseline conditions to see the effect of insulated surfaces on engine heat losses. the ceramic-coated engine (lhre) was compared to standard engine. 3.0 plasma spray technique figure 2 photographic view of cylinder head, cylinder valves and piston crown after ceramic coating. the gas tunnel type plasma spraying torch was used. the experimental method to produce ceramic coating by means of the gas tunnel type plasma spraying is as follows. after igniting plasma gun, note: 1. engine 2. rope break dynamometer 3. damping box with orifice 4. fuel tank 5. burette with measuring scale 6.water tank 7. rotameter 8. reciprrocating compressor as pressure booster 9. calorimeter figure 2 photographic view of cylinder head, cylinder valves and piston crown after ceramic coating. the gas tunnel type plasma spraying torch was used. the experimental method to produce ceramic coating by means of the gas tunnel type plasma spraying is as follows. after igniting plasma gun, the main vortex plasma jet is produced in the low pressure gas tunnel. the spraying powder is fed from central inlet of plasma gun. the coating was formed on the substrate traversed at the spraying distance l. the power input to the plasma torch was about p= 25 kw. the current and voltage applied was about 837 amp and 37.3 volts respectively. the inputs were given by miller thermal, lnc. model 3702. the power input to the pilot plasma torch, which was supplied by power supply ps1, was turned off after starting of the gas tunnel type plasma jet. the spraying distance was short distance of l=40 mm. the working gas was argon gas, and the flow rate for gas tunnel type plasma spraying torch was q= 180 l/min, and gas flow rate of carrier gas was 10 l/min (arata et.al., 1986). 4.0 result and discussion a long term experimental study has been conducted on a single cylinder, direct injection diesel engine. both the standard engine (without tbc) and its lhr version have been used in the experiments. for lhr engine a reciprocating compressor has been installed between air box and engine to boost the air pressure and to maintain constant air fuel ratio (a/f) as in standard engine. a comparative evaluation for both cases has been made based upon engine performance; brake specific fuel consumption (bsfc); exhaust gas temperature and energy balance. table 5 and 6 shows energy balance at various loads for standard and low heat rejection diesel engine respectively. it can be observed from the tables that the percentage of heat lost to the coolant and miscellaneous heat losses are less for lhr engine as compared with standard engine issn: 2180-1053 vol. 3 no. 1 january-june 2011 journal of mechanical engineering and technology 50 due to coating of low thermal conductivity material on substrate. the percentage of heat lost to the exhaust gases is more for lhr engine as compared with standard engine due to increase of combustion temperature. table 5 energy balance at various loads for standard diesel engine (se) 41 the main vortex plasma jet is produced in the low pressure gas tunnel. the spraying powder is fed from central inlet of plasma gun. the coating was formed on the substrate traversed at the spraying distance l. the power input to the plasma torch was about p= 25 kw. the current and voltage applied was about 837 amp and 37.3 volts respectively. the inputs were given by miller thermal, lnc. model 3702. the power input to the pilot plasma torch, which was supplied by power supply ps1, was turned off after starting of the gas tunnel type plasma jet. the spraying distance was short distance of l=40 mm. the working gas was argon gas, and the flow rate for gas tunnel type plasma spraying torch was q= 180 l/min, and gas flow rate of carrier gas was 10 l/min (arata et.al., 1986). 4.0 result and discussion a long term experimental study has been conducted on a single cylinder, direct injection diesel engine. both the standard engine (without tbc) and its lhr version have been used in the experiments. for lhr engine a reciprocating compressor has been installed between air box and engine to boost the air pressure and to maintain constant air fuel ratio (a/f) as in standard engine. a comparative evaluation for both cases has been made based upon engine performance; brake specific fuel consumption (bsfc); exhaust gas temperature and energy balance. table 5 and 6 shows energy balance at various loads for standard and low heat rejection diesel engine respectively. it can be observed from the tables that the percentage of heat lost to the coolant and miscellaneous heat losses are less for lhr engine as compared with standard engine due to coating of low thermal conductivity material on substrate. the percentage of heat lost to the exhaust gases is more for lhr engine as compared with standard engine due to increase of combustion temperature. table 5 energy balance at various loads for standard diesel engine (se) no load ¼ load ½ load ¾ load full load energy supplied, kw 4.28 6.83 8.93 10.49 17.29 brake power,kw 0 0.68 1.65 2.4 2.8 heat lost to exhaust gases, kw 0.6206 1.1935 1.5979 2.525 3.35205 heat lost to coolant, kw 1.7976 2.8686 3.2148 3.3568 4.8412 miscellaneouslosses, kw 1.8618 2.0879 2.4673 2.2082 6.29675 bsfc, kg/kwh 0 0.860924 0.463896 0.374643 0.529286 table 6 energy balance at various loads for low heat rejection (lhr) diesel engine 42 table 6 energy balance at various loads for low heat rejection (lhr) diesel engine no load ¼ load ½ load ¾ load full load energy supplied, kw 4.28 6.83 8.93 10.49 17.29 brake power,kw 0 0.685 1.6665 2.427 2.83 heat lost to exhaust gases,kw 0.6448 1.3243 1.9646 3.21 4.31 heat lost to coolant, kw 1.7976 2.8686 3.0362 2.9372 4.1496 miscellaneous losses, kw 1.8376 1.9521 2.2642 1.9158 6.0004 bsfc, kg/kwh 0 0.85464 0.459717 0.370475 0.523675 at all load levels, the reduction in heat rejection mostly resulted in an increase in exhaust energy. the exhaust energy was increased by 9, 18, 21 and 22% with lhr engine at 25, 50, 75% of full load and full load condition respectively compared to standard engine. figure 3 load vs heat lost to exhaust gas ceramic coated combustion chamber reduced heat transfer to the coolant. lhr engine resulted 0, 5, 14 and 16% reduction in heat transfer to the coolant for 25, 50, 75% of full load and full load condition respectively compared to standard engine. this is due to fact that ceramics have a much lower thermal conductivity than metals so that the energy flow to the coolant will be reduced which results in higher combustion temperature. at all load levels, the reduction in heat rejection mostly resulted in an increase in exhaust energy. the exhaust energy was increased by 9, 18, 21 and 22% with lhr engine at 25, 50, 75% of full load and full load condition respectively compared to standard engine. 42 table 6 energy balance at various loads for low heat rejection (lhr) diesel engine no load ¼ load ½ load ¾ load full load energy supplied, kw 4.28 6.83 8.93 10.49 17.29 brake power,kw 0 0.685 1.6665 2.427 2.83 heat lost to exhaust gases,kw 0.6448 1.3243 1.9646 3.21 4.31 heat lost to coolant, kw 1.7976 2.8686 3.0362 2.9372 4.1496 miscellaneous losses, kw 1.8376 1.9521 2.2642 1.9158 6.0004 bsfc, kg/kwh 0 0.85464 0.459717 0.370475 0.523675 at all load levels, the reduction in heat rejection mostly resulted in an increase in exhaust energy. the exhaust energy was increased by 9, 18, 21 and 22% with lhr engine at 25, 50, 75% of full load and full load condition respectively compared to standard engine. figure 3 load vs heat lost to exhaust gas ceramic coated combustion chamber reduced heat transfer to the coolant. lhr engine resulted 0, 5, 14 and 16% reduction in heat transfer to the coolant for 25, 50, 75% of full load and full load condition respectively compared to standard engine. this is due to fact that ceramics have a much lower thermal conductivity than metals so that the energy flow to the coolant will be reduced which results in higher combustion temperature. figure 3 load vs heat lost to exhaust gas ceramic coated combustion chamber reduced heat transfer to the coolant. lhr engine resulted 0, 5, 14 and 16% reduction in heat transfer to the coolant for 25, 50, 75% of full load and full load condition experimental investigation on performance of single cylinder diesel engine with mullite as thermal barrier coating issn: 2180-1053 vol. 3 no. 1 january-june 2011 51 respectively compared to standard engine. this is due to fact that ceramics have a much lower thermal conductivity than metals so that the energy flow to the coolant will be reduced which results in higher combustion temperature. 43 0 1 2 3 4 5 6 no load 1/4 load 1/2 load 3/4 load f ull load load(kgf) h ea t l os t t o co ol an t( k w ) se lhr figure 4 load vs heat lost to coolant the higher combustion temperature will lead to more expansion work. the increase of combustion temperature causes the brake power to increase up to 1.06% with lhr engine at full load condition compared to standard engine. it can be seen that the values of brake power are slightly higher for lhr engine as compared to standard engine. figure 5 load vs bp a comparison of bsfc for standard and lhr engine for all loads is as shown in fig. because of higher surface temperatures of combustion chamber of lhr engine, the bsfc values of lhr engine were lower than those of standard engine. it was observed that bsfc value was decreased by 1.07% for lhr engine as compared to standard engine at full load. figure 4 load vs heat lost to coolant the higher combustion temperature will lead to more expansion work. the increase of combustion temperature causes the brake power to increase up to 1.06% with lhr engine at full load condition compared to standard engine. it can be seen that the values of brake power are slightly higher for lhr engine as compared to standard engine. 43 0 1 2 3 4 5 6 no load 1/4 load 1/2 load 3/4 load f ull load load(kgf) h ea t l os t t o co ol an t( k w ) se lhr figure 4 load vs heat lost to coolant the higher combustion temperature will lead to more expansion work. the increase of combustion temperature causes the brake power to increase up to 1.06% with lhr engine at full load condition compared to standard engine. it can be seen that the values of brake power are slightly higher for lhr engine as compared to standard engine. figure 5 load vs bp a comparison of bsfc for standard and lhr engine for all loads is as shown in fig. because of higher surface temperatures of combustion chamber of lhr engine, the bsfc values of lhr engine were lower than those of standard engine. it was observed that bsfc value was decreased by 1.07% for lhr engine as compared to standard engine at full load. figure 5 load vs bp a comparison of bsfc for standard and lhr engine for all loads is as shown in fig. because of higher surface temperatures of combustion chamber of lhr engine, the bsfc values of lhr engine were lower than those of standard engine. it was observed that bsfc value was decreased by 1.07% for lhr engine as compared to standard engine at full load. issn: 2180-1053 vol. 3 no. 1 january-june 2011 journal of mechanical engineering and technology 52 44 figure 6 load vs bsfc the combustion chamber insulation causes the qmisc (heat rejected to the oil plus convection and radiation from the engine’s external surface) to drop at medium and high load conditions for the lhr engine. percentage heat rejection to ambient and oil decreases from 18% for lhr engine as compared to standard engine. 5.0 conclusion the following conclusions were drawn from this investigation that used a single cylinder, directinjected, inter-cooled lhr diesel engine (mullite coated). 1. in case of standard engine (without coating), the heat loss to exhaust gas was about 22 % higher than that of the lhr engine at full load. 2. the heat loss to the coolant gas was about 16 % less for lhr engine than the standard engine at full load. 3. thermal efficiency was slightly more for lhr engine than the standard engine. 4. insulation increases the cylinder wall temperature which increases exhaust gas energy. it can be then harnessed to increase the net power output of the system, thus raising the thermal efficiency and decreasing specific fuel consumption. turbo compounding can be incorporated in order to effectively recover the exhaust energy increased by heat insulation, as the shaft output. 5. drop in volumetric efficiency for lhr engine than the standard engine. it can be increased by using turbocharger. 6.0 acknowledgement we are very thankful to mr. m.nageswara rao, managing director, sai surface coating technologies pvt. ltd., hyderabad, for coating the diesel engine components using plasma spray technique. we are also thankful to j.d. college of engineering yavatmal for providing research recognized i. c. engine lab for testing. figure 6 load vs bsfc the combustion chamber insulation causes the qmisc (heat rejected to the oil plus convection and radiation from the engine’s external surface) to drop at medium and high load conditions for the lhr engine. percentage heat rejection to ambient and oil decreases from 1 8% for lhr engine as compared to standard engine. 5.0 conclusion the following conclusions were drawn from this investigation that used a single cylinder, direct-injected, inter-cooled lhr diesel engine (mullite coated). 1. in case of standard engine (without coating), the heat loss to exhaust gas was about 22 % higher than that of the lhr engine at full load. 2. the heat loss to the coolant gas was about 16 % less for lhr engine than the standard engine at full load. 3. thermal efficiency was slightly more for lhr engine than the standard engine. 4. insulation increases the cylinder wall temperature which increases exhaust gas energy. it can beharnessed to increase the net power output of the system, thus raising the thermal efficiency and decreasing specific fuel consumption. turbo compounding can be incorporated in order to effectively recover the exhaust energy increased by heat insulation, as the shaft output. 5. drop in volumetric efficiency for lhr engine than the standard engine. it can be increased by using turbocharger. experimental investigation on performance of single cylinder diesel engine with mullite as thermal barrier coating issn: 2180-1053 vol. 3 no. 1 january-june 2011 53 6.0 acknowledgement we are very thankful to mr. m.nageswara rao, managing director, sai surface coating technologies pvt. ltd., hyderabad, for coating the diesel engine components using plasma spray technique. we are also thankful to j.d. college of engineering yavatmal for providing research recognized i. c. engine lab for testing. 7.0 references a.c. alkidas, 1989. performance and emissions achievements with an uncooled heavy duty, single cylinder diesel engine. sae paper 890141. a.uzun, i. cevik, and m.akcil, 1999. effects of thermal barrier coating material on turbocharged diesel engine performance. surf. coat. technol. pp.116–119. t.hejwowski, and a. weronski, 2002. the effect of thermal barrier coatings on diesel engine performance, vacuum 65. k.toyama, t. yoshimitsu, and t.nishiyama, 1983. heat insulated turbo compound engine. sae transactions, vol. 92, pp. 3.1086. r.kamo, and w. bryzik, 1978. adiabatic turbocompound engine performance prediction. sae paper 780068. r. kamo, and w. bryzik, 1979. ceramics in heat engines. sae paper 790645. r. kamo, and w.bryzik, 1978. adiabatic turbocompound engine performance prediction. sae paper 780068. rr.sekar, and r.kamo, 1984. advanced adiabatic diesel engine for passenger cars. sae paper 840434. t. morel, ef.fort, and pn.bulumberg, 1985. effect of insulation strategy and design parameters on diesel engine heat rejection and performance. sae paper 850506. g.woschni, w. spindler, and k.kolesa, 1987. heat insulation of combustion chamber walls—a measure to decrease the fuel consumption of i.c. engines. sae paper 870339. ph.havstad, ij. gervin, and wr.wade, 1986. a ceramic insert uncooled diesel engine. sae paper 860447. r.prasad, and nk.samria, 1990. heat transfer and stress fields in the inlet and exhaust valves of a semi-adiabatic diesel engine.comput. struct., 34(5), pp.765–77. issn: 2180-1053 vol. 3 no. 1 january-june 2011 journal of mechanical engineering and technology 54 k.kokini, y.r. takeuchi, and b.d.choules, 1996. surface thermal cracking of thermal barrier coatings owing to stress relaxation: zirconia vs mullite. surf. coat. technol., 82, pp.77–82. t.m.yonushonis, 1997. overview of thermal barrier coatings for diesel engines, j. therm. spray technol., 6(1), pp. 50–56. p.ramaswamy,s. seetharamu, k.b.r.varma, and k.j. rao, 1999. thermal shock characteristics of plasma sprayed mullite coatings. j. therm. spray technol., 7(4), pp. 497–504. h.samadi, and t.w.coyle, 2005. alternative thermal barrier coatings for diesel engines. x.q.cao, r. vassen, and d. stoever, 2004. journal of the european society 24, pp.1-10. a.gilbert, k. kokini, and s. sankarasubramanian, 2008. thermal fracture of zirconia mullite composite thermal barrier coatings under thermal shock: a numerical study. surface and coating technology , 203,pp. 91-98 y.arata, a. kobayashi, y. habara, and s.jing, 1986. gas tunnel type plasma spraying // trans. of jwri., vol.15-2, pp.227-231. issn: 2180-1053 vol. 3 no. 2 july-december 2011 the interlaminar fracture properties of fibre reinforced thermoplastic composites:the effect of processing temperature and time 13 the interlaminar fracture properties of fibre reinforced thermoplastic composites:the effect of processing temperature and time s. h. sheikh md fadzullah 1,2 and w.j.cantwell1 1 school of engineering, university of liverpool, liverpool l69 3gh, united kingdom 2 faculty of mechanical engineering, universiti teknikal malaysia melaka (utem), 76109 durian tunggal, melaka, malaysia e-mail:shsmf@liv.ac.uk abstract an experimental study has been conducted to study the effect of processing temperature and dwell time on the critical energy release rate, gc of unidirectional carbon fibre-reinforced poly ether imide (cf/pei) under mode ii and mixed-mode i/ii loading. under mode ii loading, the value of giic, increased as a function of both processing temperature as well as with dwell time at the required temperature. a linear relationship was observed between the value of giic and the logarithm of the dwell time, t. under mixed-mode loading conditions, the r-curves showed a continuous increase in the value of gi/iic as a function of crack length, possibly due to the effect of fibre bridging. a comparison between the fracture toughness under mode ii shear loading and mixed-mode i/ii loading indicated that mode ii loading yielded higher fracture energies for the same temperature and dwell time, possibly due to the occurrence of additional toughening mechanisms. the results suggest that the optimum processing temperature and dwell time for this material are 300°c and 60 minutes respectively. it is planned that these processing parameters will be employed in the repair of impact-damaged panels based on this thermoplastic material. keywords: interlaminar fracture, damage, thermoplastic composites 11 the interlaminar fracture properties of fibre reinforced thermoplastic composites:the effect of processing temperature and time s. h. sheikh md fadzullah 1,2 and w.j.cantwell1 1 school of engineering, university of liverpool, liverpool l69 3gh, united kingdom 2 faculty of mechanical engineering, universiti teknikal malaysia melaka (utem), 76109 durian tunggal, melaka, malaysia e-mail:shsmf@liv.ac.uk abstract an experimental study has been conducted to study the effect of processing temperature and dwell time on the critical energy release rate, gc of unidirectional carbon fibre-reinforced poly ether imide (cf/pei) under mode ii and mixed-mode i/ii loading. under mode ii loading, the value of giic, increased as a function of both processing temperature as well as with dwell time at the required temperature. a linear relationship was observed between the value of giic and the logarithm of the dwell time, t. under mixed-mode loading conditions, the r-curves showed a continuous increase in the value of gi/iic as a function of crack length, possibly due to the effect of fibre bridging. a comparison between the fracture toughness under mode ii shear loading and mixed-mode i/ii loading indicated that mode ii loading yielded higher fracture energies for the same temperature and dwell time, possibly due to the occurrence of additional toughening mechanisms. the results suggest that the optimum processing temperature and dwell time for this material are 300°c and 60 minutes respectively. it is planned that these processing parameters will be employed in the repair of impact-damaged panels based on this thermoplastic material. keywords: interlaminar fracture, damage, thermoplastic composites 1.0 introduction with the continuous increase in the use of polymers and structural composites in a wide range of applications, such as the automotive, aerospace, defence and construction industries, appropriate techniques are required for repairing damage that may occur during the operational lifetime of the structure. however, according to wu et al. (2008), many of the existing repair methods are costly, time-consuming and require reliable detection techniques as well as a skilled workforce. in addition, these procedures are mainly applied to external repairs and accessible damage, instead of internal and invisible microcracks. in contrast, self-healing polymeric materials offer an attractive alternative for repairing damage. in principle, self-healing polymeric materials offer the potential to substantially recover their load-carrying ability after repair. such recovery can occur autonomously (selfsufficiently) or be activated through the application of a specific stimulus, such as heat or radiation. issn: 2180-1053 vol. 3 no. 2 july-december 2011 journal of mechanical engineering and technology 14 12 according to the comprehensive review on repair by wu et al. (2008), the conventional methods for the repair of advanced composites include welding, patching and in-situ curing of new resins. techniques for the repair of thermoplastics include (i) fusion bonding through resistance heating, infrared welding, dielectric and microwave welding, ultrasonic welding, vibration welding, induction welding and thermobond interlayer bonding, (ii) adhesive bonding and mechanical fastening such as riveting. however, as previously mentioned, there are some limitations to these traditional methods, including their cost and the fact that they can be time-consuming. in addition, there is a need for reliable detection techniques and a skilled workforce. it is also worth noting that existing methods are limited to the repair of external and accessible damage instead of internal and visible microcracks. in an attempt to fill this gap, a number of self-healing polymers have been recently introduced. jud et.al. (1981) investigated the mechanisms of crack-healing in glassy polymers (pmmapmma, san-san and pmma-san) using the compact tension test specimen (ct). in this study, two batches of specimens were considered; these being (i) specimens that were not completely broken and were exposed to a crack-healing treatment at elevated temperatures and (ii) samples in which the fracture surfaces of the broken ct specimens were polished and then subjected to a welding treatment at elevated temperatures. from the experimental results, it was found that at temperatures above the glass transition temperature, tg, the fracture toughness, kii varied from very low values, of the order of the surface free energy of the polymer , γ, to kio, i.e. the fracture toughness of the original material, as a function of healing time, tp, and temperature, tp. based on a diffusion model, a relationship in the form of kii t1/4 was developed and applied. in a subsequent study by davies et.al (1989), the process of crack healing in carbon fibrereinforced peek composites was investigated. double cantilever beam (dcb) specimens were repaired at temperatures between 320°c and 380°c and the degree of recovery in the delamination resistance was assessed. this initial study suggested that the mode i delamination resistance can be fully recovered in these carbon fibre reinforced thermoplastic composites. reyes and sharma (2010) studied the reparability of impact-damaged woven glass fibre reinforced polypropylene composites. a simple compression moulding process was employed to repair the damaged panels. following four-point-bend tests, on the repaired samples, a significant recovery in the flexural strength and modulus of the thermoplastic matrix composites was reported. wu et al. (2008) reviewed the literature on self-healing polymers and reported that one of the commonly-observed healing techniques include molecular inter-diffusion via thermal action. in the current research programme, a similar healing technique is used to join a fibre-reinforced thermoplastic at different temperatures and dwell times. 2.0 experimental procedure 11 the interlaminar fracture properties of fibre reinforced thermoplastic composites:the effect of processing temperature and time s. h. sheikh md fadzullah 1,2 and w.j.cantwell1 1 school of engineering, university of liverpool, liverpool l69 3gh, united kingdom 2 faculty of mechanical engineering, universiti teknikal malaysia melaka (utem), 76109 durian tunggal, melaka, malaysia e-mail:shsmf@liv.ac.uk abstract an experimental study has been conducted to study the effect of processing temperature and dwell time on the critical energy release rate, gc of unidirectional carbon fibre-reinforced poly ether imide (cf/pei) under mode ii and mixed-mode i/ii loading. under mode ii loading, the value of giic, increased as a function of both processing temperature as well as with dwell time at the required temperature. a linear relationship was observed between the value of giic and the logarithm of the dwell time, t. under mixed-mode loading conditions, the r-curves showed a continuous increase in the value of gi/iic as a function of crack length, possibly due to the effect of fibre bridging. a comparison between the fracture toughness under mode ii shear loading and mixed-mode i/ii loading indicated that mode ii loading yielded higher fracture energies for the same temperature and dwell time, possibly due to the occurrence of additional toughening mechanisms. the results suggest that the optimum processing temperature and dwell time for this material are 300°c and 60 minutes respectively. it is planned that these processing parameters will be employed in the repair of impact-damaged panels based on this thermoplastic material. keywords: interlaminar fracture, damage, thermoplastic composites 1.0 introduction with the continuous increase in the use of polymers and structural composites in a wide range of applications, such as the automotive, aerospace, defence and construction industries, appropriate techniques are required for repairing damage that may occur during the operational lifetime of the structure. however, according to wu et al. (2008), many of the existing repair methods are costly, time-consuming and require reliable detection techniques as well as a skilled workforce. in addition, these procedures are mainly applied to external repairs and accessible damage, instead of internal and invisible microcracks. in contrast, self-healing polymeric materials offer an attractive alternative for repairing damage. in principle, self-healing polymeric materials offer the potential to substantially recover their load-carrying ability after repair. such recovery can occur autonomously (selfsufficiently) or be activated through the application of a specific stimulus, such as heat or radiation. issn: 2180-1053 vol. 3 no. 2 july-december 2011 the interlaminar fracture properties of fibre reinforced thermoplastic composites:the effect of processing temperature and time 15 12 according to the comprehensive review on repair by wu et al. (2008), the conventional methods for the repair of advanced composites include welding, patching and in-situ curing of new resins. techniques for the repair of thermoplastics include (i) fusion bonding through resistance heating, infrared welding, dielectric and microwave welding, ultrasonic welding, vibration welding, induction welding and thermobond interlayer bonding, (ii) adhesive bonding and mechanical fastening such as riveting. however, as previously mentioned, there are some limitations to these traditional methods, including their cost and the fact that they can be time-consuming. in addition, there is a need for reliable detection techniques and a skilled workforce. it is also worth noting that existing methods are limited to the repair of external and accessible damage instead of internal and visible microcracks. in an attempt to fill this gap, a number of self-healing polymers have been recently introduced. jud et.al. (1981) investigated the mechanisms of crack-healing in glassy polymers (pmmapmma, san-san and pmma-san) using the compact tension test specimen (ct). in this study, two batches of specimens were considered; these being (i) specimens that were not completely broken and were exposed to a crack-healing treatment at elevated temperatures and (ii) samples in which the fracture surfaces of the broken ct specimens were polished and then subjected to a welding treatment at elevated temperatures. from the experimental results, it was found that at temperatures above the glass transition temperature, tg, the fracture toughness, kii varied from very low values, of the order of the surface free energy of the polymer , γ, to kio, i.e. the fracture toughness of the original material, as a function of healing time, tp, and temperature, tp. based on a diffusion model, a relationship in the form of kii t1/4 was developed and applied. in a subsequent study by davies et.al (1989), the process of crack healing in carbon fibrereinforced peek composites was investigated. double cantilever beam (dcb) specimens were repaired at temperatures between 320°c and 380°c and the degree of recovery in the delamination resistance was assessed. this initial study suggested that the mode i delamination resistance can be fully recovered in these carbon fibre reinforced thermoplastic composites. reyes and sharma (2010) studied the reparability of impact-damaged woven glass fibre reinforced polypropylene composites. a simple compression moulding process was employed to repair the damaged panels. following four-point-bend tests, on the repaired samples, a significant recovery in the flexural strength and modulus of the thermoplastic matrix composites was reported. wu et al. (2008) reviewed the literature on self-healing polymers and reported that one of the commonly-observed healing techniques include molecular inter-diffusion via thermal action. in the current research programme, a similar healing technique is used to join a fibre-reinforced thermoplastic at different temperatures and dwell times. 2.0 experimental procedure 13 2.1 materials in this study, a carbon fibre-reinforced polyether imide (cf/pei) from ten cate advanced composites has been investigated. the material was a unidirectional composite supplied in prepreg form with a nominal thickness of approximately 0.25 mm. table 1 presents some of the fundamental properties of this material. table 1 fundamental properties of the unidirectional carbon fibre reinforced pei properties value density (kg/m 3 ) 1510 t g (°c) 217 flexural strength (mpa) 870 flexural modulus (gpa) 50 2.2 experimental test methods interlaminar fracture testing interlaminar fracture test specimens were prepared by stacking twelve plies of unidirectional carbon fibre-reinforced polyether imide (cf/pei) in a picture frame mould with an opening size of 200 x 240 mm. to introduce a pre-crack, a folded aluminium alloys with a nominal thickness of 20 μm was inserted at the mid-plane of the stacked prepreg. the stacked prepreg was then heated to different processing temperatures (220, 240, 260, 280 and 300 °c). two different dwell times were considered at each processing temperatures; these being 5 and 60 minutes. a fixed pressure of 6.5 bar was employed during processing. this was followed by cold pressing to room temperature. after consolidation, the moulded panels had a thickness of approximately 3 mm. mode ii end notched flexure (enf) testing mode ii end notched flexure (enf) testing was carried out in accordance with european structural integrity society (esis) protocol. the nominal specimen thickness width (b) was 20 mm, the initial crack length (ao) was 30 mm and the total specimen length was 120 mm as illustrated in figure 1. table 1 table 1 issn: 2180-1053 vol. 3 no. 2 july-december 2011 journal of mechanical engineering and technology 16 14 figure 1 schematic of the test configuration for the mode ii enf test. based on modified beam theory, the critical energy release rate, giic was calculated using (1) whereby giic = the critical strain energy release rate, p = load, δ = displacement of the cross-head, a = effective crack length, b = width of the specimen and l = half distance between the supports. the mode ii fracture tests were performed at a crosshead displacement rate of 1 mm/min using an instron 4505 universal testing machine. the load-displacement data were measured using a 5 kn load-cell. mixed-mode interlaminar fracture testing the mixed-mode i/ii interlaminar fracture properties of the composites was characterised using the mixed-mode flexure geometry (mmf) shown in figure 2. this test is similar to the enf test configuration, with the difference being that the load is applied to only one arm. the ratio of the length of the starter defect (ao) to the half span (l) was fixed at 0.5. in this test, the load is applied at the mid-span, yielding a ratio of mode i (gi) to mode ii strain energy release rate (giic) of 4/3. figure 2 schematic of test configuration for mixed mode flexure testing issn: 2180-1053 vol. 3 no. 2 july-december 2011 the interlaminar fracture properties of fibre reinforced thermoplastic composites:the effect of processing temperature and time 17 15 the test was conducted at a cross-head displacement rate of 1 mm/min using an instron 4505 screw-driven universal testing machine. the mixed-mode critical strain energy release rate, gi/iic was calculated via: (2) where p is the applied load and c is the specimen compliance. 3.0 results and discussion figure 3 shows representative mode ii load-displacement traces following tests at temperatures between 220°c and 300°c. the load increases in a linear fashion until a maximum value is reached, associated with crack propagation from the starter defect. from the plots, it is clear that the maximum load increases as a function of processing temperature, reflecting a gradual increase in the mode ii critical energy release rate with increasing processing temperature. over the temperature range of 220°c to 300°c, there is a continuous increase in the critical strain energy release rate, as shown in figure 4. it is also evident that the value of giic reaches a maximum at 320°c. also, it is also clear that, for a given processing temperature, there is an increase in the critical strain energy release rate as the dwell time is increased, as is also shown in figure 4. table 2 summarizes the critical energy release rate for different temperatures and dwell. it is clear that the mode ii fracture toughness of the fullyconsolidated composite is very high, being in excess of 3000 j/m2. these high value are significantly above values for other thermoplastics, such as peek, and all thermosets, such as epoxy resins. a plot of giic as a function of the logarithm of dwell time is presented in figure 5. from the figure, there is evidence that there is a linear relationship between the critical energy release rate and the logarithm of log time. similar observations have been reported by jud et.al (1981) in the study of crack healing in pmma and san polymers where it was observed that the fracture toughness, kii increased with contact time, according to kii t1/4, at temperatures above the glass transition temperature, tg , of the polymer. in general, under mode ii loading conditions, this unidirectional carbon fibre reinforced poly ether imide (cf/pei) exhibited an increase in giic with increasing processing temperature and dwell time. in figure 6, it is evident that there is an increase in the value of gi/iic with processing temperature, from 220°c up to 300°c for a constant dwell time of 5 minutes. it is interesting to note that the values of gi/iic are much lower than the mode ii values, with the maximum value reaching only 1200 j/m2. this suggests that this form of loading is more critical for these composites. figure 6, issn: 2180-1053 vol. 3 no. 2 july-december 2011 journal of mechanical engineering and technology 18 16 to understand the effect of dwell time on gi/iic, typical load-displacement traces are plotted for specimen processed at two temperatures (240°c and 300°c) at dwell times of 5 minutes and 60 minutes, in figure 7. from the figure, there is a noticeable increase in the maximum load at the temperature is increased and as the dwell time is increased from 5 to 60 minutes at a given temperature. a summary of the experimental results under mixed-mode i/ii tests is given in table 3. figure 8 shows typical r-curves for different temperatures at a dwell time of 5 minutes. from the plots, it is clear that there is an increase in the value of gi/iic as a function of crack length. this is likely due to the presence of fibre bridging, along the crack path, as depicted in figure 9. figure 10 presents a comparison of the critical strain energy release rate under mode ii shear loading and the results of mixed-mode i/ii loading conditions. as reported above, it is clear that mode ii loading yielded higher values of gc at the same temperature and dwell time, suggesting that there is an additional toughening mechanism under mode ii shear loading. these findings are in agreement with the work by kim et al. (2004) figure 3 (a) typical load-displacement traces for mode ii samples, processed at 220°c, 240°c and 260°c respectively with 5 minutes of dwell time. 0 50 100 150 200 250 300 350 400 0 0.5 1 1.5 2 2.5 3 l oa d (n ) displacement (mm) 220°c 260°c 240°c figure 8 figure 10 issn: 2180-1053 vol. 3 no. 2 july-december 2011 the interlaminar fracture properties of fibre reinforced thermoplastic composites:the effect of processing temperature and time 19 17 figure 3 (b) typical load-displacement traces, for mode ii samples processed at 280°c, 300°c and 320°c respectively with 5 minutes of dwell time. figure 4 plot of giic as a function of processing temperature and dwell time. 0 100 200 300 400 500 600 700 800 900 0 2 4 6 8 l oa d (n ) displacement (mm) 280°c 300°c 320°c 0 500 1000 1500 2000 2500 3000 3500 4000 220 240 260 280 300 320 g ii c (j /m 2 ) temperature (°c) 5minutes 60minutes issn: 2180-1053 vol. 3 no. 2 july-december 2011 journal of mechanical engineering and technology 20 18 figure 5 plot of giic versus log time table 2 average values of giic for different processing temperatures and dwell times. temperature (°c) dwell time (mins) giic (j/m2) 220 0 0 5 139(46) 60 283(75) 240 0 0 5 14(46) 60 468(95) 260 0 0 5 580(30) 60 840(231) 280 0 0 5 1162(170) 60 2271(343) 300 0 0 5 2880(298) 60 3101(649) 0 500 1000 1500 2000 2500 3000 3500 4000 0 0.5 1 1.5 2 g ii c ( j/ m 2 ) log time (minutes) 220°c 240°c 260°c 280°c 300°c 320°c 19 320 0 0 5 3260(174) note: *the values in the bracket correspond to the standard deviations figure 6 plot of gi/iic at different processing temperatures for the mixed-mode fracture specimens manufactured using a dwell time of 5 minutes figure 7 (a) typical mixed-mode load-displacement traces following processing at 240°c with dwell times of 5 and 60 minutes. 0 200 400 600 800 1000 1200 1400 1600 1800 220 240 260 280 300 g i/ ii c ( j/ m 2 ) temperature (°c) 0 50 100 150 200 250 0 1 2 3 4 5 l oa d( n ) displacement(mm) 5 minutes 60 minutes 18 figure 5 plot of giic versus log time table 2 average values of giic for different processing temperatures and dwell times. temperature (°c) dwell time (mins) giic (j/m2) 220 0 0 5 139(46) 60 283(75) 240 0 0 5 14(46) 60 468(95) 260 0 0 5 580(30) 60 840(231) 280 0 0 5 1162(170) 60 2271(343) 300 0 0 5 2880(298) 60 3101(649) 0 500 1000 1500 2000 2500 3000 3500 4000 0 0.5 1 1.5 2 g ii c ( j/ m 2 ) log time (minutes) 220°c 240°c 260°c 280°c 300°c 320°c 18 figure 5 plot of giic versus log time table 2 average values of giic for different processing temperatures and dwell times. temperature (°c) dwell time (mins) giic (j/m2) 220 0 0 5 139(46) 60 283(75) 240 0 0 5 14(46) 60 468(95) 260 0 0 5 580(30) 60 840(231) 280 0 0 5 1162(170) 60 2271(343) 300 0 0 5 2880(298) 60 3101(649) 0 500 1000 1500 2000 2500 3000 3500 4000 0 0.5 1 1.5 2 g ii c ( j/ m 2 ) log time (minutes) 220°c 240°c 260°c 280°c 300°c 320°c issn: 2180-1053 vol. 3 no. 2 july-december 2011 the interlaminar fracture properties of fibre reinforced thermoplastic composites:the effect of processing temperature and time 21 19 320 0 0 5 3260(174) note: *the values in the bracket correspond to the standard deviations figure 6 plot of gi/iic at different processing temperatures for the mixed-mode fracture specimens manufactured using a dwell time of 5 minutes figure 7 (a) typical mixed-mode load-displacement traces following processing at 240°c with dwell times of 5 and 60 minutes. 0 200 400 600 800 1000 1200 1400 1600 1800 220 240 260 280 300 g i/ ii c ( j/ m 2 ) temperature (°c) 0 50 100 150 200 250 0 1 2 3 4 5 l oa d( n ) displacement(mm) 5 minutes 60 minutes issn: 2180-1053 vol. 3 no. 2 july-december 2011 journal of mechanical engineering and technology 22 20 figure 7 (b) typical mmf load-displacement traces following processing at 300° with dwell times of 5 and 60 minutes. table 3 average values of gi/iic for mixed mode interlaminar fracture tests at different processing temperatures and dwell times. processing temperature (°c) dwell time (minutes) gi/iic (j/m2) 220 5 291 (46) 60 509 (57) 240 5 420 (126) 60 1191 (168) 260 5 800(161) 60 686(69) 280 5 873(61) 60 1088(211) 300 5 1264(310) 60 1256(255) note: *the values in the bracket corresponds to the standard deviations 0 100 200 300 400 500 600 0 1 2 3 4 5 6 l oa d (n ) displacement (mm) 5 minutes 60 minutes issn: 2180-1053 vol. 3 no. 2 july-december 2011 the interlaminar fracture properties of fibre reinforced thermoplastic composites:the effect of processing temperature and time 23 21 figure 8 representative r-curves for mmf test specimens processed at different processing temperatures, manufactured using a dwell time of 5 minutes. figure 9 photograph of a mixed-mode test specimen showing fibre bridging as the crack propagates from the initial crack to the mid span position. 0 100 200 300 400 500 600 700 800 900 15 17 19 21 23 25 27 29 31 g i/ ii c (j /m 2 ) crack length, a (mm) 220° 240° marked crack point fibre bridging issn: 2180-1053 vol. 3 no. 2 july-december 2011 journal of mechanical engineering and technology 24 22 figure 10 gc versus processing temperature following both mode ii and mixed-mode i/ii interlaminar fracture tests on specimens manufactured using a dwell time of 5 minutes. 4.0 conclusions an experimental study has been conducted to investigate the effect of varying the processing temperature on the interlaminar fracture properties of a carbon fibre reinforced polyether imide (cf/pei). the tests have been conducted at four different repair temperatures above the glass transition temperature of the poly ether imide, pei matrix (tg of 217°c); these being 220°c, 240°c, 260°c and 300°c, all of which are. data from both end-notched flexure (enf) as well as mixed-mode interlaminar fracture testing suggest that the optimum processing temperature in terms of achieving the required critical energy release rate, gc is 300°c, with a dwell time of 60 minutes. further work will investigate the reparability of damaged cf/pei following lowvelocity impact, in which the optimum processing parameters identified in this current work will be considered. the test laminates will be repaired using a one-step compression moulding procedure using the hot press machine. the laminate stiffness values after impact as well as an effect of repair temperatures and time will be 0 500 1000 1500 2000 2500 3000 3500 4000 220 240 260 280 300 320 g c ( j/ m 2 ) temperature (°c) mixed mode mode ii 22 figure 10 gc versus processing temperature following both mode ii and mixed-mode i/ii interlaminar fracture tests on specimens manufactured using a dwell time of 5 minutes. 4.0 conclusions an experimental study has been conducted to investigate the effect of varying the processing temperature on the interlaminar fracture properties of a carbon fibre reinforced polyether imide (cf/pei). the tests have been conducted at four different repair temperatures above the glass transition temperature of the poly ether imide, pei matrix (tg of 217°c); these being 220°c, 240°c, 260°c and 300°c, all of which are. data from both end-notched flexure (enf) as well as mixed-mode interlaminar fracture testing suggest that the optimum processing temperature in terms of achieving the required critical energy release rate, gc is 300°c, with a dwell time of 60 minutes. further work will investigate the reparability of damaged cf/pei following lowvelocity impact, in which the optimum processing parameters identified in this current work will be considered. the test laminates will be repaired using a one-step compression moulding procedure using the hot press machine. the laminate stiffness values after impact as well as an effect of repair temperatures and time will be 0 500 1000 1500 2000 2500 3000 3500 4000 220 240 260 280 300 320 g c ( j/ m 2 ) temperature (°c) mixed mode mode ii 23 assessed via indentation tests. a detailed analysis of the repaired sites will be carried out using optical microscopy, looking at impact sites repaired at different temperatures, to investigate the presence of any residual damage after repair. overall, the findings of this research work suggest that it is possible to repair damage in fibre-reinforced thermoplastics, such as carbon fibre-reinforced poly ether imide (cf/pei). 5.0 acknowledgements the authors would like to thank to universiti teknikal malaysia melaka (utem) and ministry of higher education (mohe) of malaysia for the financial support of this study. 6.0 references p.davies, w.j.cantwell, h.h.kausch. 1989. healing of cracks in carbon fibre-peek composites. journal of materials science letters. 8.1247-1248. k. jud, h.h.kausch, j.g.williams. 1981. fracture mechanics studies of crack healing and welding of polymers. journal of materials science. 16.204-210. k.y.kim and l.ye. 2004. interlaminar fracture toughness of cf/pei composites at elevated temperatures: roles of matrix toughness and fibre/matrix adhesion. composites: part a. 35.477-487. k.y.kim, l.ye and k.m.phoa. 2004. interlaminar fracture toughness of cf/pei and gf/pei composites at elevated temperatures. applied composite materials.11.173-190. g.reyes and u.sharma. 2010. modelling and damage repair of woven thermoplastic composites subjected to low velocity impact. composite structures. 92. 523-531. d.y.wu, s.meure and d.solomon. 2008. self-healing polymeric materials: a review of recent developments. prog. polym.sci. 33.479-522. issn: 2180-1053 vol. 3 no. 2 july-december 2011 the interlaminar fracture properties of fibre reinforced thermoplastic composites:the effect of processing temperature and time 25 23 assessed via indentation tests. a detailed analysis of the repaired sites will be carried out using optical microscopy, looking at impact sites repaired at different temperatures, to investigate the presence of any residual damage after repair. overall, the findings of this research work suggest that it is possible to repair damage in fibre-reinforced thermoplastics, such as carbon fibre-reinforced poly ether imide (cf/pei). 5.0 acknowledgements the authors would like to thank to universiti teknikal malaysia melaka (utem) and ministry of higher education (mohe) of malaysia for the financial support of this study. 6.0 references p.davies, w.j.cantwell, h.h.kausch. 1989. healing of cracks in carbon fibre-peek composites. journal of materials science letters. 8.1247-1248. k. jud, h.h.kausch, j.g.williams. 1981. fracture mechanics studies of crack healing and welding of polymers. journal of materials science. 16.204-210. k.y.kim and l.ye. 2004. interlaminar fracture toughness of cf/pei composites at elevated temperatures: roles of matrix toughness and fibre/matrix adhesion. composites: part a. 35.477-487. k.y.kim, l.ye and k.m.phoa. 2004. interlaminar fracture toughness of cf/pei and gf/pei composites at elevated temperatures. applied composite materials.11.173-190. g.reyes and u.sharma. 2010. modelling and damage repair of woven thermoplastic composites subjected to low velocity impact. composite structures. 92. 523-531. d.y.wu, s.meure and d.solomon. 2008. self-healing polymeric materials: a review of recent developments. prog. polym.sci. 33.479-522. 23 assessed via indentation tests. a detailed analysis of the repaired sites will be carried out using optical microscopy, looking at impact sites repaired at different temperatures, to investigate the presence of any residual damage after repair. overall, the findings of this research work suggest that it is possible to repair damage in fibre-reinforced thermoplastics, such as carbon fibre-reinforced poly ether imide (cf/pei). 5.0 acknowledgements the authors would like to thank to universiti teknikal malaysia melaka (utem) and ministry of higher education (mohe) of malaysia for the financial support of this study. 6.0 references p.davies, w.j.cantwell, h.h.kausch. 1989. healing of cracks in carbon fibre-peek composites. journal of materials science letters. 8.1247-1248. k. jud, h.h.kausch, j.g.williams. 1981. fracture mechanics studies of crack healing and welding of polymers. journal of materials science. 16.204-210. k.y.kim and l.ye. 2004. interlaminar fracture toughness of cf/pei composites at elevated temperatures: roles of matrix toughness and fibre/matrix adhesion. composites: part a. 35.477-487. k.y.kim, l.ye and k.m.phoa. 2004. interlaminar fracture toughness of cf/pei and gf/pei composites at elevated temperatures. applied composite materials.11.173-190. g.reyes and u.sharma. 2010. modelling and damage repair of woven thermoplastic composites subjected to low velocity impact. composite structures. 92. 523-531. d.y.wu, s.meure and d.solomon. 2008. self-healing polymeric materials: a review of recent developments. prog. polym.sci. 33.479-522. 02(13-26).pdf investigation on the mechanical characteristics of sawdust and chipwood filled epoxy issn: 2180-1053 vol. 3 no. 1 january-june 2011 71 investigation on the mechanical characteristics of sawdust and chipwood filled epoxy puvanasvaran, a.p1., hisham, s2., kamil sued, m1 1faculty of manufacturing engineering, universiti teknikal malaysia melaka, locked bag 1752, 76109, melaka, malaysia 2universiti kuala lumpur 1016, jalan sultan ismail, 50250 kuala lumpur, malaysia email: 1punesh@utem.edu.my abstract in furniture and paper industries, huge amount of wood flakes and wood flours in the form of chip wood and sawdust are always found as waste. this is known as natural fiber, a renewable source that available at low cost. this study was carried out for three different sizes of fiber which categories into “soft”, “rough” and “coarse” particles that derived from chip wood (cw) and sawdust (sw). the sw and cw fiber were blended with epoxy by using hand tools machine respectively, which then open molding was employed to form a fiber composite and specimens’ accordance to the astm standards. it was found that the strength of tensile value for the rough size particles of sw were higher than the cw. the works presented a good quality of sw and cw fiber composite had been produced which can be used for home furniture utilities. keywords: sawdust and chip wood, wood composites, tensile and flexural properties. 1.0 introduction polymer composites have replaced conventional engineering materials such as metals, plastics and ceramics in many engineering applications in the recent years. nowadays, composite materials are chosen as materials in engineering products for a variety of reasons, including lightweight, high stiffness, high strength, low thermal expansion, corrosion resistance, and long fatigue life. in recent years, natural fibre composites are used in many non-structural and semi-structural applications due to their low cost, being renewable and abundance and the examples include coconut, oil palm, pineapple leaf and wood fibre reinforced polymer composites. in the past there were numerous issn: 2180-1053 vol. 3 no. 1 january-june 2011 journal of mechanical engineering and technology 72 work have been conducted in the area of natural fibre composites using different types of fibres such as bamboo (chen et.al., 1998, ismail et.al., 2002), oil palm empty fruit bunch (rozman et.al., 2003) and wood (jayaraman and bhattacharyya, 2004, elinwa and mahmood, 2002). in furniture and paper industries, huge amount of wood flakes and wood flours in the form of chip wood and sawdust are always found as wastes. because of high accessibility and low cost of the chip wood and sawdust, they can easily be incorporated with polymers by compounding process in order to improve the properties of reinforced polymers such as high strength and ease of processing compared to wood or neat polymer. wood is a unique material that has many characteristics such as aesthetics and light weight compared with man-made materials such as concrete and brick. wood can be used in conjunction with polymers to forms glued laminated members and can be used alone in the form of a sheet of paper or to make a toy (jozsef,1982). wood continues to be the raw material for the large number of products in the modern time, even thought other materials such as metals, cement and plastics are highly used as raw materials to produce consumer products (george, 1991). wood fibre reinforced polymer composites have been known for many years. historically, most of woods are used in the form of wood flour to produce filled composites. the wood flour reduces the cost of composite materials, but was not usually intended to substantially improve the performance of composites. more recently, the use of wood fibers to provide a reinforcing mechanism in thermoplastics has been of substantial interest. in fact, several companies have manufactured wood fibre reinforced thermoplastic composite materials for use as synthetic lumber in applications such as decking and window frames (susan et.al., 2004). the demand from the end user for furniture especially in modern furniture design, required the use of wood composites with high strength (eckelman ,1997). moreover, the use of natural fibre in furniture is not just limited to wood fibres but also in other plant based fibres such as banana pseudo-stem faibres (sapuan and maleque, 2006). in this paper, a study on mechanical properties of chip wood and sawdust fibre reinforced epoxy composites is presented. mechanical properties studied include tensile strength, flexural strength, modulus of elasticity and modulus of rapture. investigation on the mechanical characteristics of sawdust and chipwood filled epoxy issn: 2180-1053 vol. 3 no. 1 january-june 2011 73 2.0 methodology materials and methods the waste wood fibers used in this study were collected from a saw mill in sungai buluh, selangor, malaysia. in all cases waste woods were dried overnight to remove moisture at room temperature and were kept in the polyethylene bags to prevent fungus growth. both sawdust and wood chip were sieved using three different sizes of sieves. the timber species dominant in the sawdust and chip woods was not known to the authors and perhaps they were simply a blend of various hardwoods. chip wood (cw) fibres collected from each sieve were separated into three different sizes namely, “coarse” for 150 meshes, “rough” for 100 meshes and “soft” for 50 meshes. the sawdust (sw) were divided “coarse” for 1.0 mesh, “rough” for 0.5 meshes and “soft” for 0.3 meshes (see figure 1). the sieving process was carried out manually and the whole process was carried out in 30 minutes. the resins used in the study were epoxy type asasin 142 a used as matrix and asasin 142 b used as hardener. the mixing ratio was for 100 parts of asasin 142a there was 40 parts of asasin 142b. the amount of sw/epoxy and cw/epoxy composites was determined by their weight ratios. different amount of sw, cw and epoxy liquid were blended using hand tool machine. the fibre of the size classified as ‘coarse’, ‘rough’ and ‘soft’ with 14 % ratio by weight each was loaded to epoxy with the weight of 100 gm and together with hardener with the weight of 40gm were prepared. the material used to prepare tensile test specimen was plaster of paris and for flexural test was glass. the parts to be molded were coated with release agent to prevent them from sticking to the mould. the test specimens were prepared using astm d 638 (tahun) for tensile testing and astm d 790-97 (tahun) for flexural testing. the stress-strain plots and moduli of elasticity the composite were obtained using an instron universal testing machine, model 6566 during the tensile and flexural testing. the load applied was specified at 10kn at 2.5 mm/min cross-head speed. the thickness of both tensile and flexural testing specimens was 6 mm. the experiment was conducted at room temperature. the specimens prepared took 24 hours to dry. six specimens were tested for both tensile and flexural testing. issn: 2180-1053 vol. 3 no. 1 january-june 2011 journal of mechanical engineering and technology 74 61 the test specimens were prepared using astm d 638 (tahun) for tensile testing and astm d 790-97 (tahun) for flexural testing. the stress-strain plots and moduli of elasticity the composite were obtained using an instron universal testing machine, model 6566 during the tensile and flexural testing. the load applied was specified at 10kn at 2.5 mm/min cross-head speed. the thickness of both tensile and flexural testing specimens was 6 mm. the experiment was conducted at room temperature. the specimens prepared took 24 hours to dry. six specimens were tested for both tensile and flexural testing. (a) cw coarse fibre (b) cw rough fibre (c) cw soft fibre (a) sw course fibre (b) sw rough fibre (c) sw soft fibre figure 1: fibre used in the study 3.0 results and discussions the results of measurement of the tensile properties and flexural properties of the composites for different particle sizes in the case of sawdust and chip wood are shown in tables 1 and 2 respectively. the comparison of the mean of each property of three categories of fibres is analyzed using t-test. the analysis of variance (anova) is used to compare the mean scores for all properties of during the tensile and flexural tests for each category of fibres. table 1: tensile properties of waste wood fibre reinforced epoxy composites property sawdust (sw) soft rough coarse chip wood (cw) soft rough coarse displacement at peak mm 8.626 12.968 12.037 1.906 11.447 11.086 strain at peak ( %) 17.030 26.103 24.407 3.846 22.893 22.256 stress at peak (mpa) 10.921 28.917 24.767 14.224 19.867 17.922 young’s modulus (mpa) 386.873 889.988 860.116 594.453 902.331 737.815 load at peak (kn) 0.852 2.297 1.932 1.117 1.550 1.398 61 the test specimens were prepared using astm d 638 (tahun) for tensile testing and astm d 790-97 (tahun) for flexural testing. the stress-strain plots and moduli of elasticity the composite were obtained using an instron universal testing machine, model 6566 during the tensile and flexural testing. the load applied was specified at 10kn at 2.5 mm/min cross-head speed. the thickness of both tensile and flexural testing specimens was 6 mm. the experiment was conducted at room temperature. the specimens prepared took 24 hours to dry. six specimens were tested for both tensile and flexural testing. (a) cw coarse fibre (b) cw rough fibre (c) cw soft fibre (a) sw course fibre (b) sw rough fibre (c) sw soft fibre figure 1: fibre used in the study 3.0 results and discussions the results of measurement of the tensile properties and flexural properties of the composites for different particle sizes in the case of sawdust and chip wood are shown in tables 1 and 2 respectively. the comparison of the mean of each property of three categories of fibres is analyzed using t-test. the analysis of variance (anova) is used to compare the mean scores for all properties of during the tensile and flexural tests for each category of fibres. table 1: tensile properties of waste wood fibre reinforced epoxy composites property sawdust (sw) soft rough coarse chip wood (cw) soft rough coarse displacement at peak mm 8.626 12.968 12.037 1.906 11.447 11.086 strain at peak ( %) 17.030 26.103 24.407 3.846 22.893 22.256 stress at peak (mpa) 10.921 28.917 24.767 14.224 19.867 17.922 young’s modulus (mpa) 386.873 889.988 860.116 594.453 902.331 737.815 load at peak (kn) 0.852 2.297 1.932 1.117 1.550 1.398 61 the test specimens were prepared using astm d 638 (tahun) for tensile testing and astm d 790-97 (tahun) for flexural testing. the stress-strain plots and moduli of elasticity the composite were obtained using an instron universal testing machine, model 6566 during the tensile and flexural testing. the load applied was specified at 10kn at 2.5 mm/min cross-head speed. the thickness of both tensile and flexural testing specimens was 6 mm. the experiment was conducted at room temperature. the specimens prepared took 24 hours to dry. six specimens were tested for both tensile and flexural testing. (a) cw coarse fibre (b) cw rough fibre (c) cw soft fibre (a) sw course fibre (b) sw rough fibre (c) sw soft fibre figure 1: fibre used in the study 3.0 results and discussions the results of measurement of the tensile properties and flexural properties of the composites for different particle sizes in the case of sawdust and chip wood are shown in tables 1 and 2 respectively. the comparison of the mean of each property of three categories of fibres is analyzed using t-test. the analysis of variance (anova) is used to compare the mean scores for all properties of during the tensile and flexural tests for each category of fibres. table 1: tensile properties of waste wood fibre reinforced epoxy composites property sawdust (sw) soft rough coarse chip wood (cw) soft rough coarse displacement at peak mm 8.626 12.968 12.037 1.906 11.447 11.086 strain at peak ( %) 17.030 26.103 24.407 3.846 22.893 22.256 stress at peak (mpa) 10.921 28.917 24.767 14.224 19.867 17.922 young’s modulus (mpa) 386.873 889.988 860.116 594.453 902.331 737.815 load at peak (kn) 0.852 2.297 1.932 1.117 1.550 1.398 61 the test specimens were prepared using astm d 638 (tahun) for tensile testing and astm d 790-97 (tahun) for flexural testing. the stress-strain plots and moduli of elasticity the composite were obtained using an instron universal testing machine, model 6566 during the tensile and flexural testing. the load applied was specified at 10kn at 2.5 mm/min cross-head speed. the thickness of both tensile and flexural testing specimens was 6 mm. the experiment was conducted at room temperature. the specimens prepared took 24 hours to dry. six specimens were tested for both tensile and flexural testing. (a) cw coarse fibre (b) cw rough fibre (c) cw soft fibre (a) sw course fibre (b) sw rough fibre (c) sw soft fibre figure 1: fibre used in the study 3.0 results and discussions the results of measurement of the tensile properties and flexural properties of the composites for different particle sizes in the case of sawdust and chip wood are shown in tables 1 and 2 respectively. the comparison of the mean of each property of three categories of fibres is analyzed using t-test. the analysis of variance (anova) is used to compare the mean scores for all properties of during the tensile and flexural tests for each category of fibres. table 1: tensile properties of waste wood fibre reinforced epoxy composites property sawdust (sw) soft rough coarse chip wood (cw) soft rough coarse displacement at peak mm 8.626 12.968 12.037 1.906 11.447 11.086 strain at peak ( %) 17.030 26.103 24.407 3.846 22.893 22.256 stress at peak (mpa) 10.921 28.917 24.767 14.224 19.867 17.922 young’s modulus (mpa) 386.873 889.988 860.116 594.453 902.331 737.815 load at peak (kn) 0.852 2.297 1.932 1.117 1.550 1.398 figure 1: fibre used in the study 3.0 results and discussions the results of measurement of the tensile properties and flexural properties of the composites for different particle sizes in the case of sawdust and chip wood are shown in tables 1 and 2 respectively. the comparison of the mean of each property of three categories of fibres is analyzed using t-test. the analysis of variance (anova) is used to compare the mean scores for all properties of during the tensile and flexural tests for each category of fibres. table 1: tensile properties of waste wood fibre reinforced epoxy composites 61 the test specimens were prepared using astm d 638 (tahun) for tensile testing and astm d 790-97 (tahun) for flexural testing. the stress-strain plots and moduli of elasticity the composite were obtained using an instron universal testing machine, model 6566 during the tensile and flexural testing. the load applied was specified at 10kn at 2.5 mm/min cross-head speed. the thickness of both tensile and flexural testing specimens was 6 mm. the experiment was conducted at room temperature. the specimens prepared took 24 hours to dry. six specimens were tested for both tensile and flexural testing. (a) cw coarse fibre (b) cw rough fibre (c) cw soft fibre (a) sw course fibre (b) sw rough fibre (c) sw soft fibre figure 1: fibre used in the study 3.0 results and discussions the results of measurement of the tensile properties and flexural properties of the composites for different particle sizes in the case of sawdust and chip wood are shown in tables 1 and 2 respectively. the comparison of the mean of each property of three categories of fibres is analyzed using t-test. the analysis of variance (anova) is used to compare the mean scores for all properties of during the tensile and flexural tests for each category of fibres. table 1: tensile properties of waste wood fibre reinforced epoxy composites property sawdust (sw) soft rough coarse chip wood (cw) soft rough coarse displacement at peak mm 8.626 12.968 12.037 1.906 11.447 11.086 strain at peak ( %) 17.030 26.103 24.407 3.846 22.893 22.256 stress at peak (mpa) 10.921 28.917 24.767 14.224 19.867 17.922 young’s modulus (mpa) 386.873 889.988 860.116 594.453 902.331 737.815 load at peak (kn) 0.852 2.297 1.932 1.117 1.550 1.398 from table 1, it is shown that the tensile strength of rough sw is higher than the chip wood cw with a maximum reading of 29.917 mpa. the mechanical properties for both composites (sw and cw) revealed an optimum values at rough particle size. moreover, by comparing sw and cw, it is found that the tensile strength for rough and coarse sw is higher than the rough and coarse of cw. on the other hand, for soft chip wood the tensile strength is higher than the soft sawdust. it is believed that the material properties differences might originate from investigation on the mechanical characteristics of sawdust and chipwood filled epoxy issn: 2180-1053 vol. 3 no. 1 january-june 2011 75 the compositions and active surface areas for the sawdust and chip wood. table 2: flexural test results for sw and cw 62 from table 1, it is shown that the tensile strength of rough sw is higher than the chip wood cw with a maximum reading of 29.917 mpa. the mechanical properties for both composites (sw and cw) revealed an optimum values at rough particle size. moreover, by comparing sw and cw, it is found that the tensile strength for rough and coarse sw is higher than the rough and coarse of cw. on the other hand, for soft chip wood the tensile strength is higher than the soft sawdust. it is believed that the material properties differences might originate from the compositions and active surface areas for the sawdust and chip wood. table 2: flexural test results for sw and cw property sawdust (sw) soft rough coarse chip wood (cw) soft rough coarse maximum load (n) 102.550 121.145 62.395 78.827 102.740 145.755 extension at compression load (mm) 4.408 6.215 14.552 4.225 4.083 8.548 energy at compression load (j) 0.247 0.395 0.625 0.180 0.218 0.757 strain at compression load (mm) 0.102 0.167 0.145 0.043 0.040 0.192 stress at compression load (mpa) 21.910 25.891 13.333 16.843 21.952 31.141 in term of flexural properties, the highest and lowest properties between sawdust and chip wood can be found in table 2. stress compression load (mpa) is found decreased with the increments of the particles sizes for sw but a vise versa behavior is shown by the chip wood composite. the results in flexural test for both composites can be ascribed because of the differences in particles sizes and fiber surfaces. a three-point bend configuration was employed for the determination of modulus of rupture (mor) and modulus of elasticity (moe). equations for mor and moe are as shown in equation 1 (eq. 1) and equation 2 (eq. 2). where p is the maximum load carried by the specimen, l the support span, b and d are the specimen breadth and depth, respectively, measured at the nearest undisturbed location to the region of failure, and m is the slope of the load–deflection curve during elastic deformation savastano et al. (2000). table 3: the means of mor for sawdust and chip wood fiber mean sd particle size mean sd saw dust course 19.998 3.364 30.565 8.669 rough 38.829 1.534 soft 32.869 4.421 chipwood course 46.716 3.543 34.970 10.403 rough 32.930 5.309 soft 25.265 6.607 mor=3pl/2bd2 (1) moe=ml3/4bd3 (2) in term of flexural properties, the highest and lowest properties between sawdust and chip wood can be found in table 2. stress compression load (mpa) is found decreased with the increments of the particles sizes for sw but a vise versa behavior is shown by the chip wood composite. the results in flexural test for both composites can be ascribed because of the differences in particles sizes and fiber surfaces. a three-point bend configuration was employed for the determination of modulus of rupture (mor) and modulus of elasticity (moe). equations for mor and moe are as shown in equation 1 (1) and equation 2 (2). where p is the maximum load carried by the specimen, l the support span, b and d are the specimen breadth and depth, respectively, measured at the nearest undisturbed location to the region of failure, and m is the slope of the load–deflection curve during elastic deformation savastano et.al. (2000). 62 from table 1, it is shown that the tensile strength of rough sw is higher than the chip wood cw with a maximum reading of 29.917 mpa. the mechanical properties for both composites (sw and cw) revealed an optimum values at rough particle size. moreover, by comparing sw and cw, it is found that the tensile strength for rough and coarse sw is higher than the rough and coarse of cw. on the other hand, for soft chip wood the tensile strength is higher than the soft sawdust. it is believed that the material properties differences might originate from the compositions and active surface areas for the sawdust and chip wood. table 2: flexural test results for sw and cw property sawdust (sw) soft rough coarse chip wood (cw) soft rough coarse maximum load (n) 102.550 121.145 62.395 78.827 102.740 145.755 extension at compression load (mm) 4.408 6.215 14.552 4.225 4.083 8.548 energy at compression load (j) 0.247 0.395 0.625 0.180 0.218 0.757 strain at compression load (mm) 0.102 0.167 0.145 0.043 0.040 0.192 stress at compression load (mpa) 21.910 25.891 13.333 16.843 21.952 31.141 in term of flexural properties, the highest and lowest properties between sawdust and chip wood can be found in table 2. stress compression load (mpa) is found decreased with the increments of the particles sizes for sw but a vise versa behavior is shown by the chip wood composite. the results in flexural test for both composites can be ascribed because of the differences in particles sizes and fiber surfaces. a three-point bend configuration was employed for the determination of modulus of rupture (mor) and modulus of elasticity (moe). equations for mor and moe are as shown in equation 1 (eq. 1) and equation 2 (eq. 2). where p is the maximum load carried by the specimen, l the support span, b and d are the specimen breadth and depth, respectively, measured at the nearest undisturbed location to the region of failure, and m is the slope of the load–deflection curve during elastic deformation savastano et al. (2000). table 3: the means of mor for sawdust and chip wood fiber mean sd particle size mean sd saw dust course 19.998 3.364 30.565 8.669 rough 38.829 1.534 soft 32.869 4.421 chipwood course 46.716 3.543 34.970 10.403 rough 32.930 5.309 soft 25.265 6.607 mor=3pl/2bd2 (1) moe=ml3/4bd3 (2) where p is the maximum load carried by the specimen, l the support span, b and d are the specimen breadth and depth, respectively, measured at the nearest undisturbed location to the region of failure, and m is the slope of the load–deflection curve during elastic deformation savastano et.al. (2000). issn: 2180-1053 vol. 3 no. 1 january-june 2011 journal of mechanical engineering and technology 76 table 3: the means of mor for sawdust and chip wood 62 from table 1, it is shown that the tensile strength of rough sw is higher than the chip wood cw with a maximum reading of 29.917 mpa. the mechanical properties for both composites (sw and cw) revealed an optimum values at rough particle size. moreover, by comparing sw and cw, it is found that the tensile strength for rough and coarse sw is higher than the rough and coarse of cw. on the other hand, for soft chip wood the tensile strength is higher than the soft sawdust. it is believed that the material properties differences might originate from the compositions and active surface areas for the sawdust and chip wood. table 2: flexural test results for sw and cw property sawdust (sw) soft rough coarse chip wood (cw) soft rough coarse maximum load (n) 102.550 121.145 62.395 78.827 102.740 145.755 extension at compression load (mm) 4.408 6.215 14.552 4.225 4.083 8.548 energy at compression load (j) 0.247 0.395 0.625 0.180 0.218 0.757 strain at compression load (mm) 0.102 0.167 0.145 0.043 0.040 0.192 stress at compression load (mpa) 21.910 25.891 13.333 16.843 21.952 31.141 in term of flexural properties, the highest and lowest properties between sawdust and chip wood can be found in table 2. stress compression load (mpa) is found decreased with the increments of the particles sizes for sw but a vise versa behavior is shown by the chip wood composite. the results in flexural test for both composites can be ascribed because of the differences in particles sizes and fiber surfaces. a three-point bend configuration was employed for the determination of modulus of rupture (mor) and modulus of elasticity (moe). equations for mor and moe are as shown in equation 1 (eq. 1) and equation 2 (eq. 2). where p is the maximum load carried by the specimen, l the support span, b and d are the specimen breadth and depth, respectively, measured at the nearest undisturbed location to the region of failure, and m is the slope of the load–deflection curve during elastic deformation savastano et al. (2000). table 3: the means of mor for sawdust and chip wood fiber mean sd particle size mean sd saw dust course 19.998 3.364 30.565 8.669 rough 38.829 1.534 soft 32.869 4.421 chipwood course 46.716 3.543 34.970 10.403 rough 32.930 5.309 soft 25.265 6.607 mor=3pl/2bd2 (1) moe=ml3/4bd3 (2) table 3 shows the means of mechanical properties for the particle sizes in modulus of rupture (mor) according to the fiber types. it can be indicate that the mean of cw is higher than sw, at 34.970 kgf/mm2 and 30.565 kgf/mm2 respectively. further test was done using turkey’s honestly significant difference with a significant level of 0.05 in identifying the lies. the p value and the significant level of physical properties of the composite are listed in tables 4 and 5. table 4: independent t-test for mean mor between types of fiber 63 table 3 shows the means of mechanical properties for the particle sizes in modulus of rupture (mor) according to the fiber types. it can be indicate that the mean of cw is higher than sw, at 34.970 kgf/mm2 and 30.565 kgf/mm2 respectively. further test was done using turkey’s honestly significant difference with a significant level of 0.05 in identifying the lies. the p value and the significant level of physical properties of the composite are listed in tables 4 and 5. table 4: independent t-test for mean mor between types of fiber variable t-value p-value fiber -1.380 0.177 significant level, α = 0.05 from the independent t-test the p-values are greater than α = 0.05 (refer to table 4). this indicates that there is no significant difference in the mean of mor between sw and cw. table 5: the means of moe for sw and cw fibre mean sd particle size mean sd saw dust 4254.195 874.771 course 324.3288 44.0466 rough 10185.66 14675.49 soft 2252.595 450.6688 chip wood 1993.023 421.217 course 1373.762 337.5417 rough 2682.045 354.8565 soft 1923.25 1075.615 table 5 shows the means of mechanical properties in modulus of elasticity (moe) according to fiber respectively to the particle sizes. the means and the test direction used multiple comparison of tukey’s studentized range test. from table 5 indicates that the mean of sw (4254.195 kgf/mm2) is higher than cw (1993.023 kgf/mm2). table 6: independent t-test for mean moe between types of fiber variable t-value p-value fiber 1.05 0.301 significant level, α=0.05 the comparison of moe between sw and cw were used an independent t-test, shows that all the p-values are greater than α = 0.05 (refer to table 6).this indicate that there is no significant difference in the mean of moe between chips wood and saw dust. significant level, α = 0.05 from the independent t-test the p-values are greater than α = 0.05 (refer to table 4). this indicates that there is no significant difference in the mean of mor between sw and cw. table 5: the means of moe for sw and cw 63 table 3 shows the means of mechanical properties for the particle sizes in modulus of rupture (mor) according to the fiber types. it can be indicate that the mean of cw is higher than sw, at 34.970 kgf/mm2 and 30.565 kgf/mm2 respectively. further test was done using turkey’s honestly significant difference with a significant level of 0.05 in identifying the lies. the p value and the significant level of physical properties of the composite are listed in tables 4 and 5. table 4: independent t-test for mean mor between types of fiber variable t-value p-value fiber -1.380 0.177 significant level, α = 0.05 from the independent t-test the p-values are greater than α = 0.05 (refer to table 4). this indicates that there is no significant difference in the mean of mor between sw and cw. table 5: the means of moe for sw and cw fibre mean sd particle size mean sd saw dust 4254.195 874.771 course 324.3288 44.0466 rough 10185.66 14675.49 soft 2252.595 450.6688 chip wood 1993.023 421.217 course 1373.762 337.5417 rough 2682.045 354.8565 soft 1923.25 1075.615 table 5 shows the means of mechanical properties in modulus of elasticity (moe) according to fiber respectively to the particle sizes. the means and the test direction used multiple comparison of tukey’s studentized range test. from table 5 indicates that the mean of sw (4254.195 kgf/mm2) is higher than cw (1993.023 kgf/mm2). table 6: independent t-test for mean moe between types of fiber variable t-value p-value fiber 1.05 0.301 significant level, α=0.05 the comparison of moe between sw and cw were used an independent t-test, shows that all the p-values are greater than α = 0.05 (refer to table 6).this indicate that there is no significant difference in the mean of moe between chips wood and saw dust. table 5 shows the means of mechanical properties in modulus of elasticity (moe) according to fiber respectively to the particle sizes. the means and the test direction used multiple comparison of tukey’s studentized range test. from table 5 indicates that the mean of sw (4254.195 kgf/mm2) is higher than cw (1993.023 kgf/mm2). investigation on the mechanical characteristics of sawdust and chipwood filled epoxy issn: 2180-1053 vol. 3 no. 1 january-june 2011 77 table 6: independent t-test for mean moe between types of fiber 63 table 3 shows the means of mechanical properties for the particle sizes in modulus of rupture (mor) according to the fiber types. it can be indicate that the mean of cw is higher than sw, at 34.970 kgf/mm2 and 30.565 kgf/mm2 respectively. further test was done using turkey’s honestly significant difference with a significant level of 0.05 in identifying the lies. the p value and the significant level of physical properties of the composite are listed in tables 4 and 5. table 4: independent t-test for mean mor between types of fiber variable t-value p-value fiber -1.380 0.177 significant level, α = 0.05 from the independent t-test the p-values are greater than α = 0.05 (refer to table 4). this indicates that there is no significant difference in the mean of mor between sw and cw. table 5: the means of moe for sw and cw fibre mean sd particle size mean sd saw dust 4254.195 874.771 course 324.3288 44.0466 rough 10185.66 14675.49 soft 2252.595 450.6688 chip wood 1993.023 421.217 course 1373.762 337.5417 rough 2682.045 354.8565 soft 1923.25 1075.615 table 5 shows the means of mechanical properties in modulus of elasticity (moe) according to fiber respectively to the particle sizes. the means and the test direction used multiple comparison of tukey’s studentized range test. from table 5 indicates that the mean of sw (4254.195 kgf/mm2) is higher than cw (1993.023 kgf/mm2). table 6: independent t-test for mean moe between types of fiber variable t-value p-value fiber 1.05 0.301 significant level, α=0.05 the comparison of moe between sw and cw were used an independent t-test, shows that all the p-values are greater than α = 0.05 (refer to table 6).this indicate that there is no significant difference in the mean of moe between chips wood and saw dust. significant level, α=0.05 the comparison of moe between sw and cw were used an independent t-test, shows that all the p-values are greater than α = 0.05 (refer to table 6).this indicate that there is no significant difference in the mean of moe between chips wood and saw dust. 4.0 conclusion based six epoxy composite that reinforced by chip wood and sawdust, the rough sawdust composite exhibit greater mechanical properties compared to chip wood. mechanical properties data were subjected to two-way analysis of variance using tukey’s multiple comparison method to determine the significance of differences in sample means at the 95% confidence level (p=0.05). from the tests results, evidence shows that the factors influenced the natural composite are depending on the fiber source, size, shape, processes, separation in matrix, bonding in matrixes. the other potential factor is defects that have in the fiber. 5.0 references chen, x., gue, q. and. mi, y. bamboo fiber reinforced polypropylene composites: a study of the mechanical properties, j. appl. polym. sci., 1998.69: 1891-1899. eckelman, c.a., a look at the strength design of furniture. forest product journal. 1966. 16. 3: pp. 21-24. elinwa, a.u. and mahmood, y.a. ash from timber waste as cement replacement material, cement and concrete composite 2002. 24 (2),pp. 219-222. george tsoumis, (1991) science and technology of wood. structure, properties, utilization. ed. van nostrand reinhold pp ix. new york: library of congress cataloging-in publication data. ismail,h., edyham, m.r. and wirjosentono, b. bamboo fibre filled natural rubber composites the effects of filler loading and bonding agent, polymer testing, 2002; 21, (2), pp 139-144. jozsef bodig, benjamin a. jayne, ( 1982). mechanics of wood and wood issn: 2180-1053 vol. 3 no. 1 january-june 2011 journal of mechanical engineering and technology 78 composites, ed. van nostrand reinhold company pp 1 library of congress cataloging in publication data. khrisnan jayaraman, debes bhattacharyya,(2004) mechanical performance of woodfiber-waste plastic composite materials. resources, conservation & recucling 41,307-319 rozman, h.d., saad, m.j. and ishak, z.a.m., flexural and impact properties of oil palm empty fruit bunch (efb)-polypropylene composites the effect of maleic anhydride chemical modification of efb, polym. test. 2003. 22: pp 335-341. sapuan, s.m. and maleque, m.a. design and fabrication of natural woven fabric reinforced epoxy composite for household telephone stand, material and design 2005. 26: 65-71. savastano, h, warden, p.g. and coutts, r.s.p., .brazilian waste fibres as reinforcement for cement-based composites cement and concrete composites, 2000. 22, 5, pp 379 384. susan e. selke,indrek wichman. (2004) wood fiber/polyolefin compisite,composite part a: applied science and manufacturing. 35, 321-326 kinematics assignment issn: 2180-1053 vol. 9 no.2 july – december 2017 1 forward and inverse kinematics analysis and validation of the abb irb 140 industrial robot mohammed almaged 1* , omar ismael 2 1,2 nineveh university, college of electronics engineering, department of systems & control engineering, nineveh, iraq abstract the main goal of this paper is to derive the forward and inverse kinematics model of the abb irb 140 industrial manipulator. this work provides essential kinematics information that could be a useful reference for future research on the robot. it can also serve as teaching material for students in the area of robotics, especially forward and inverse kinematics, to aid students' understanding of these subjects. denavit-hartenberg analysis (dh) is presented to write the forward kinematic equations. initially, a coordinate system is attached to each of the six links of the manipulator. then, the corresponding four link parameters are determined for each link to construct the six transformation matrices ( 𝑇𝑖 𝑖−1 ) that define each frame relative to the previous one. while, to develop the kinematics that calculates the required joint angles (𝜃1 − 𝜃6), both geometrical and analytical approaches are used to solve the inverse kinematic problem. after introducing the forward and inverse kinematic models, a matlab code is written to obtain the solutions of these models. then, the forward kinematics is validated by examining a set of known positions of the robot arm, while the inverse kinematics is checked by comparing the results obtained in matlab with a simulation in robot studio. keywords: robotics; forward kinematics; inverse kinematics; irb 140 manipulator 1.0 introduction 'kinematics is the science of geometry in motion' (jazar, 2010). this means it deals only with geometrical issues of motion such as the position and orientation regardless the force that causes them. there are two types of kinematics, the forward and inverse kinematics. forward kinematic analysis is concerned with the relationship between the joint angle of the robot manipulator and the position and orientation of the end-effector (spong, hutchinson, & vidyasagar, 2006) (paul, 1981) . in other words, it deals with finding the homogeneous transformation matrix that describes the position and orientation of the tool frame with respect to the global reference frame. on the other hand, inverse kinematics is used to calculate the joint angles required to achieve the desired position and orientation. the same transformation matrix which resulted from the forward kinematics in order to describe the position and the orientation of the tool frame relative to the robot base frame is used here in the inverse kinematics to solve for the joint angles. several academic studies investigating the kinematics of the robot manipulators have been carried out to increase their intelligence and usability. various *corresponding author e-mail: m_engineer89@yahoo.com journal of mechanical engineering and technology issn: 2180-1053 vol. 9 no.2 july – december 2017 2 approaches have been introduced for the analysis. in his book, selig (selig, 2013) has discussed several ways of analyzing robots using geometrical approach. jazar (jazar, 2010) also reviewed a number of analytical methods for the analysis of serial robots. the concept of the homogeneous transformation matrix is very old in the area of kinematic analysis. however, it is still very popular and valuable. several authors (craig, 2005; jazar, 2010; shahinpoor, 1987; uicker, pennock, & shigley, 2011) have discussed the formulation of the homogeneous transformation matrix. in 2012, k. mitra introduced a different procedure for the formulation of the homogeneous matrix. his method was based on motion transfer at the joints from the base to the end effector. this technique was validated through a numerical study on a 5 dof robot (mitra, 2012). also, a. khatamian produced a new analytical method for solving the forward kinematics of a six dof manipulator (khatamian, 2015). several approaches have been used to solve the inverse kinematic problem. some researchers have investigated the inverse kinematics of robot manipulators using standard techniques such as geometric, algebraic, etc. in 2012, deshpande and george presented an analytical solution for the inverse kinematics derived from the d-h homogeneous transformation matrix (deshpande & george, 2012). in the same vein, neppalli et al developed a closed-form analytical approach to solve the inverse kinematics for multi-section robots. in this novel approach, the problem is decomposed into several easier sub-problems. then, an algorithm is employed to produce a complete solution to the inverse kinematic problem (neppalli, csencsits, jones, & walker, 2009). s. yahya et al proposed a new geometrical method to find the inverse kinematics of the planar manipulators (samer y, 2009). other researchers have solved the inverse kinematic problem using advanced techniques such as artificial neural network and biomimetic approach. in 2014, feng et al produced a novel learning algorithm, called extreme learning machine, based on a neural network to generate the inverse kinematic solution of robot manipulator (feng, yao-nan, & yimin, 2014). the findings of this advanced method revealed that the extreme learning machine has not only significantly reduced the computation time but also enhanced the precision. 2.0 robot specifications figure 1 shows the compact six degree of freedom industrial abb irb 140 manipulator. the robot has six revolute joints controlled by ac-motors. it is designed specifically for manufacturing industries to perform a wide range of applications such as welding, packing, assembly, etc. the specifications, axes and dimensions of the robot manipulator are shown below in table 1. forward and inverse kinematics analysis and validation of the abb irb 140 industrial robot issn: 2180-1053 vol. 8 no.2 july – december 2017 3 table 1. the abb irb 140 specifications (abb, 2000) manipulator weight 98 kg tool centre point tcp max. speed 250 mm/s endurance load in xy direction ± 1300 n endurance load in z direction ± 1000 n endurance torque in xy direction ± 1300 n.m endurance torque in z direction ± 300 n.m figure 1. the abb irb 140 manipulator (abb, 2000) 3.0 forward kinematics to mathematically model a robot and hence determine the position and orientation of the end effector with respect to the base or any other point, it is necessary to assign a global coordinate frame to the base of the robot and a local reference frame at each joint. then, the denavit-hartenberg analysis (dh) is presented to build the homogeneous transformation matrices between the robot joint axes (craig, 2005) (siciliano, sciavicco, villani, & oriolo, 2010). these matrices are a function of four parameters resulted from a series of translations and rotations around different axes. the illustration of how frame {i} is related to the previous frame {i −1} and the description of the frame parameters are shown in figure 2 below. journal of mechanical engineering and technology issn: 2180-1053 vol. 9 no.2 july – december 2017 4 figure 2. the description of frame {i} with respect to frame {i −1}(craig, 2005) these modified d-h parameters can be described as:  αi-1: twist angle between the joint axes zi and zi-1 measured about xi-1.  ai-1: distance between the two joint axes zi and zi-1 measured along the common normal.  θi: joint angle between the joint axes xi and xi-1 measured about zi.  di: link offset between the axes xi and xi-1 measured along zi. thus, the four transformations between the two axes can be defined as: after finishing the multiplication of these four transformations, the homogeneous transform can be obtained as: ti i−1 = ( cθi −sθi 0 ai−1 sθicαi−1 cθicαi−1 −sαi−1 −disαi−1 sθisαi−1 cθisαi−1 cαi−1 dicαi−1 0 0 0 1 ) (1.1) the abb irb 140 frames assignment is shown below in figure 3. 𝑇𝑖 𝑖−1 = 𝑅𝑜𝑡(𝑋𝑖−1,𝛼𝑖−1 ) x 𝑇𝑟𝑎𝑛𝑠(𝑋𝑖−1,𝑎𝑖−1) x 𝑅𝑜𝑡(𝑍𝑖 ,𝜃𝑖) x 𝑇𝑟𝑎𝑛𝑠(0,0,𝑑𝑖) forward and inverse kinematics analysis and validation of the abb irb 140 industrial robot issn: 2180-1053 vol. 8 no.2 july – december 2017 5 figure 3. abb irb140 frames assignment according to this particular frame assignment, the modified d-h parameters are defined in table 2 below. table 2. the abb irb 140 d-h parameters for the simplicity of calculations and matrix product, it can be assumed that s2 = sin (θ2-90), c2 = cos (θ2-90). after achieving the d-h table, the individual transformation matrix for each link is achieved by substituting the link parameters into the general homogeneous transform derived in matrix (1.1) above. axis (i) αi-1 ai-1 di θi 1 0 0 d1 = 352 θ1 2 -90 a1 = 70 0 θ2-90 3 0 a2 = 360 0 θ3 4 -90 0 d4 = 380 θ4 5 90 0 0 θ5 6 -90 0 0 θ6 journal of mechanical engineering and technology issn: 2180-1053 vol. 9 no.2 july – december 2017 6 𝑇1 0 = ( 𝑐𝜃1 −𝑠𝜃1 0 𝑎0 𝑠𝜃1𝑐𝛼0 𝑐𝜃𝑖𝑐𝛼01 −𝑠𝛼0 −𝑑1𝑠𝛼0 𝑠𝜃1𝑠𝛼0 𝑐𝜃1𝑠𝛼0 𝑐𝛼0 𝑑1𝑐𝛼0 0 0 0 1 ) 𝑇1 0 = ( 𝑐1 −𝑠1 0 0 𝑠1 𝑐1 0 0 0 0 1 𝑑1 0 0 0 1 ) 𝑇2 1 = ( 𝑐𝜃2 −𝑠𝜃2 0 𝑎1 𝑠𝜃2𝑐𝛼1 𝑐𝜃2𝑐𝛼1 −𝑠𝛼1 −𝑑2𝑠𝛼1 𝑠𝜃2𝑠𝛼1 𝑐𝜃2𝑠𝛼1 𝑐𝛼1 𝑑2𝑐𝛼1 0 0 0 1 ) 𝑇2 01 = ( 𝑐2 −𝑠2 0 𝑎1 0 0 1 0 −𝑠2 −𝑐2 0 0 0 0 0 1 ) 𝑇3 2 = ( 𝑐𝜃3 −𝑠𝜃3 0 𝑎2 𝑠𝜃3𝑐𝛼2 𝑐𝜃3𝑐𝛼2 −𝑠𝛼2 −𝑑3𝑠𝛼2 𝑠𝜃3𝑠𝛼2 𝑐𝜃3𝑠𝛼2 𝑐𝛼2 𝑑3𝑐𝛼2 0 0 0 1 ) 𝑇3 2 = ( 𝑐3 −𝑠3 0 𝑎2 𝑠3 𝑐3 0 0 0 0 1 0 0 0 0 1 ) 𝑇4 3 = ( 𝑐𝜃4 −𝑠𝜃4 0 𝑎3 𝑠𝜃4𝑐𝛼3 𝑐𝜃4𝑐𝛼3 −𝑠𝛼3 −𝑑4𝑠𝛼3 𝑠𝜃4𝑠𝛼3 𝑐𝜃4𝑠𝛼3 𝑐𝛼3 𝑑4𝑐𝛼3 0 0 0 1 ) 𝑇4 3 = ( 𝑐4 −𝑠4 0 0 0 0 1 𝑑4 −𝑠4 −𝑐4 0 0 0 0 0 1 ) 𝑇5 4 = ( 𝑐𝜃5 −𝑠𝜃5 0 𝑎4 𝑠𝜃5𝑐𝛼4 𝑐𝜃5𝑐𝛼4 −𝑠𝛼4 −𝑑5𝑠𝛼4 𝑠𝜃5𝑠𝛼4 𝑐𝜃5𝑠𝛼4 𝑐𝛼4 𝑑5𝑐𝛼4 0 0 0 1 ) 𝑇5 4 = ( 𝑐5 −𝑠5 0 0 0 0 −1 0 𝑠5 𝑐5 0 0 0 0 0 1 ) 𝑇6 5 = ( 𝑐𝜃6 −𝑠𝜃6 0 𝑎5 𝑠𝜃6𝑐𝛼5 𝑐𝜃6𝑐𝛼5 −𝑠𝛼5 −𝑑6𝑠𝛼5 𝑠𝜃6𝑠𝛼5 𝑐𝜃5𝑠𝛼5 𝑐𝛼5 𝑑6𝑐𝛼5 0 0 0 1 ) 𝑇6 5 = ( 𝑐6 −𝑠6 0 0 0 0 1 0 −𝑠6 −𝑐6 0 0 0 0 0 1 ) once the homogeneous transformation matrix of each link is obtained, forward kinematic chain can be applied to achieve the position and orientation of the robot endeffector with respect to the global reference frame (robot base). 𝑇2 0 = 𝑇1 0 x 𝑇2 1 𝑇2 0 = ( 𝑐1 −𝑠1 0 0 𝑠1 𝑐1 0 0 0 0 1 𝑑1 0 0 0 1 ) x ( 𝑐2 −𝑠2 0 𝑎1 0 0 1 0 −𝑠2 −𝑐2 0 0 0 0 0 1 ) = ( 𝑐1𝑐2 −𝑐1𝑠2 −𝑠1 𝑐1𝑎1 𝑠1𝑐2 −𝑠1𝑠2 𝑐1 𝑠1𝑎1 −𝑠2 −𝑐2 0 𝑑1 0 0 0 1 ) 𝑇3 0 = 𝑇2 0 x 𝑇3 2 forward and inverse kinematics analysis and validation of the abb irb 140 industrial robot issn: 2180-1053 vol. 8 no.2 july – december 2017 7 𝑇3 0 = ( 𝑐1𝑐2 −𝑐1𝑠2 −𝑠1 𝑐1𝑎1 𝑠1𝑐2 −𝑠1𝑠2 𝑐1 𝑠1𝑎1 −𝑠2 −𝑐2 0 𝑑1 0 0 0 1 )𝑋( 𝑐3 −𝑠3 0 𝑎2 𝑠3 𝑐3 0 0 0 0 1 0 0 0 0 1 ) 𝑇3 0 = ( 𝑐1𝑐2𝑐3 −𝑐1𝑠2𝑠3 −(𝑐1𝑐2𝑠3 +𝑐1𝑠2𝑐3) −𝑠1 𝑐1𝑐2𝑎2 +𝑐1𝑎1 ) 𝑠1𝑐2𝑐3 −𝑠1𝑠2𝑠3 −(𝑠1𝑐2𝑠3 +𝑠1𝑠2𝑐3) 𝑐1 𝑠1𝑐2𝑎2 +𝑠1𝑎1 ) −(𝑠2𝑐3 +𝑐2𝑠3) 𝑠2𝑠3 −𝑐2𝑐3 0 −𝑠2𝑎2 +𝑑1 0 0 0 1 ) 𝑇 3 0 = ( 𝑐1𝑐23 −𝑐1𝑠23 −𝑠1 𝑐1(𝑐2𝑎2 +𝑎1 ) 𝑠1𝑐23 −𝑠1𝑠23 𝑐1 𝑠1(𝑐2𝑎2 +𝑎1 ) −𝑠23 −𝑐23 0 −𝑠2𝑎2 +𝑑1 0 0 0 1 ) 𝑇6 4 = 𝑇 5 4 x 𝑇6 5 𝑇 =6 4 ( 𝑐5 −𝑠5 0 0 0 0 −1 0 𝑠5 𝑐5 0 0 0 0 0 1 )𝑋( 𝑐6 −𝑠6 0 0 0 0 1 0 −𝑠6 −𝑐6 0 0 0 0 0 1 ) = ( 𝑐5𝑐6 −𝑐5𝑠6 −𝑠5 0 𝑠6 𝑐6 0 0 𝑠5 𝑐6 −𝑠5 𝑠6 𝑐5 0 0 0 0 1 ) 𝑇6 3 = 𝑇 4 3 x 𝑇6 4 𝑇6 3 = ( 𝑐4 −𝑠4 0 0 0 0 1 𝑑4 −𝑠4 −𝑐4 0 0 0 0 0 1 )𝑋( 𝑐5𝑐6 −𝑐5𝑠6 −𝑠5 0 𝑠6 𝑐6 0 0 𝑠5 𝑐6 −𝑠5 𝑠6 𝑐5 0 0 0 0 1 ) 𝑇6 3 = ( 𝑐4𝑐5𝑐6−𝑠4𝑠6 −𝑐4𝑐5𝑠6−𝑠4𝑐6 −𝑐4𝑠5 0 𝑠5 𝑐6 −𝑠5 𝑠6 𝑐5 𝑑4 −𝑠4𝑐5𝑐6−𝑐4𝑠6 𝑠4𝑐5𝑠6−c4𝑐6 𝑠4𝑠5 0 0 0 0 1 ) 𝑇6 0 = 𝑇 x3 0 𝑇6 3 𝑇 =6 0 ( 𝑐1𝑐23 −𝑐1𝑠23 −𝑠1 𝑐1(𝑐2𝑎2 +𝑎1 ) 𝑠1𝑐23 −𝑠1𝑠23 𝑐1 𝑠1(𝑐2𝑎2 +𝑎1 ) −𝑠23 −𝑐23 0 −𝑠2𝑎2 +𝑑1 0 0 0 1 )𝑋( 𝑐4𝑐5𝑐6−𝑠4𝑠6 −𝑐4𝑐5𝑠6−𝑠4𝑐6 −𝑐4𝑠5 0 𝑠5 𝑐6 −𝑠5 𝑠6 𝑐5 𝑑4 −𝑠4𝑐5𝑐6−𝑐4𝑠6 𝑠4𝑐5𝑠6−c4𝑐6 𝑠4𝑠5 0 0 0 0 1 ) 𝑇 6 0 = ( 𝑟11 r12 r13 x r21 r22 r23 y r31 r32 r33 𝑧 0 0 0 1 ) r11 = 𝑐1𝑐23 (𝑐4𝑐5𝑐6−𝑠4𝑠6)−𝑐1𝑠23𝑠5 𝑐6 +𝑠1(𝑠4𝑐5𝑐6+𝑐4𝑠6) r12 = 𝑐1𝑐23(−𝑐4𝑐5𝑠6−𝑠4𝑐6)+ 𝑐1𝑠23𝑠5 𝑠6 −𝑠1(𝑠4𝑐5𝑠6−c4𝑐6) r13 = −𝑐1𝑐23𝑐4𝑠5 − 𝑐1𝑠23𝑐5 −𝑠1𝑠4𝑠5 journal of mechanical engineering and technology issn: 2180-1053 vol. 9 no.2 july – december 2017 8 r21 = 𝑠1𝑐23 (𝑐4𝑐5𝑐6−𝑠4𝑠6)−𝑠1𝑠23𝑠5 𝑐6 −𝑐1(𝑠4𝑐5𝑐6+𝑐4𝑠6) r22 = 𝑠1𝑐23(−𝑐4𝑐5𝑠6−𝑠4𝑐6)+𝑠1𝑠23𝑠5 𝑠6 +𝑐1(𝑠4𝑐5𝑠6−c4𝑐6) r23 = −𝑠1𝑐23𝑐4𝑠5 − 𝑠1𝑠23𝑐5 +𝑐1𝑠4𝑠5 r31 = −𝑠23 (𝑐4𝑐5𝑐6−𝑠4𝑠6)−𝑐23𝑠5 𝑐6 r32 = −𝑠23 (−𝑐4𝑐5𝑠6−𝑠4𝑐6)+𝑐23𝑠5 𝑠6 r33 = 𝑠23𝑐4𝑠5 −𝑐23𝑐5 x = − 𝑑4𝑐1𝑠23 + 𝑐1(𝑐2𝑎2 +𝑎1 ) y = − 𝑑4𝑠1𝑠23 + 𝑠1(𝑐2𝑎2 +𝑎1 ) z = − 𝑠2𝑎2 +𝑑1 − 𝑑4𝑐23 it is also possible to find the position of the tool centre point (tcp) with respect to the robot base. according to the robot frame assignment, it is simply a transition along the z axis of frame {6} by d6 (distance from joint 6 to the tcp). therefore, the final position of the end effector with respect to the robot global reference frame can be expressed as: ptcp = 𝑇6 0 𝑋 p6 ptcp = ( 𝑟11 r12 r13 x r21 r22 r23 y r31 r32 r33 𝑧 0 0 0 1 )𝑋( 0 0 d6 1 ) = ( d6 r13 +x d6 r23 +y d6 r33 +z 1 ) 4.0 forward kinematic validation after finding the homogeneous transformation matrix ( 𝑇)6 0 that describes the end effector position and orientation with respect to the robot global reference frame, the position of the robot in space is expressed by the vector 0 p6org which gives the values of x, y and z vectors as follow: 𝑥 = −𝑑4𝑐1𝑠23 + 𝑐1(𝑐2𝑎2 +𝑎1 ) 𝑦 = −𝑑4𝑠1𝑠23 + 𝑠1(𝑐2𝑎2 +𝑎1 ) 𝑧 = −𝑠2𝑎2 +𝑑1 − 𝑑4𝑐23 given: s2 = sin (θ2-90), c2 = cos (θ2-90), d1 = 352 mm, d4 = 380 mm, a1 = 70 mm a2 = 360 mm. forward and inverse kinematics analysis and validation of the abb irb 140 industrial robot issn: 2180-1053 vol. 8 no.2 july – december 2017 9 the above equations are programmed in matlab and a set of eight positions, illustrated below in figure 4, were chosen randomly to validate the forward kinematic model. the joint angles of each position are entered manually by the user to obtain the x, y and z vectors as shown in table 3 below. it can be clearly seen that there is no y component corresponding to these particular positions because ɵ1 is always given to be zero. then these joint angle values were entered through the robot operating software (teach pendant) in the lab. for each case, the actual robot position was similar to the x, y and z vector obtained from matlab which proves the validity of the matlab code. figure 4. set of robot’s positions table 3. matlab results of each position position joint angles x vector y vector z vector 0 ɵ1 = 0, ɵ2 = 0, ɵ3 = 0 450 0 712 1 ɵ1 = 0, ɵ2 = 0, ɵ3 = -90 70 0 1092 2 ɵ1 = 0, ɵ2 = 0, ɵ3 = 50 314 0 420.9 3 ɵ1 = 0, ɵ2 = 110, ɵ3 = -90 765 0 98.9 6 ɵ1 = 0, ɵ2 = -90, ɵ3 = 50 1.1 0 596 7 ɵ1 = 0, ɵ2 = 110, ɵ3 =-230 218 0 558 8 ɵ1 = 0, ɵ2 = -90, ɵ3 = -90 -670 0 352 journal of mechanical engineering and technology issn: 2180-1053 vol. 9 no.2 july – december 2017 10 5.0 inverse kinematics inverse kinematics is used to calculate the joint angles required to achieve the desired position and orientation in the robot workspace. the configuration of the robot governs the selection of the solution method. since three consecutive axes of the robot intersect at a common point, pieper's solution can be applied which provides a huge simplification of the inverse kinematic problem. an algebraic solution can also be implemented through the use of the inverse trigonometric functions. however, piper's solution is chosen because it can be easily coded in matlab. pieper's approach works on the principle of separating the position solution for ɵ1, ɵ2 and ɵ3 from the orientation solution to solve for ɵ4, ɵ5 and ɵ6 (pires, 2007). in general, there are two methods of solution, the analytical and geometrical approaches. a geometrical approach is initially implemented to find the joint variables ɵ1, ɵ2 and ɵ3 that define the end effector position in space, while an analytical solution is applied to calculate the angles ɵ4, ɵ5 and ɵ6 which describe the end-effector orientation. 5.1 geometrical solution according to the frame assignment shown in figure one, x and y components of frame {1} is the same as frame {0} because there is only a z-directional offset between the two frames. therefore, the projection of the wrist components on x-y plane of frame {0} has the same components on frame {1} (carter, 2009; vicente, 2007). in addition, since both link two and three are planar, the position vector in y direction changes with respect to θ1 only. thus, two possible solutions for θ1 can be achieved by simply applying the arctangent function. 𝜃1 = 𝑎𝑡𝑎𝑛2 (𝑃𝑦𝑡𝑐𝑝,𝑃𝑥𝑡𝑐𝑝), (5.1) 𝜃1 ′ = 𝛱 + 𝜃1 (5.2) the solutions of θ2 and θ3 are obtained by considering the plane, shown in figure 5 below, formed by the second and third planar links with respect to the robot reference frame. forward and inverse kinematics analysis and validation of the abb irb 140 industrial robot issn: 2180-1053 vol. 8 no.2 july – december 2017 11 figure 5. projection of links two and three onto the x y plane the cosine low is used to solve for θ3 as follow: ℎ2 = 𝐿2 2 + 𝐿3 2 −2 𝑥 𝐿2 𝑥 𝐿3 𝑥 𝑐𝑜𝑠 (180− 𝜁) since the position is given with respect to robot’s tool center point (tcp), l3 should be equal to d4+d6, where d6 is the distance from joint 6 to the tcp. while, 𝐿2 = 𝑎2 ,ℎ 2 = 𝑠2 +𝑟2 ,𝑐𝑜𝑠 (180 −𝜁) = − 𝑐𝑜𝑠 (𝜁) 𝑠2 + 𝑟2 = 𝑎2 2 + (𝑑4 + 𝑑6) 2 +2 𝑥 𝑎2 𝑥 (𝑑4 + 𝑑6 ) 𝑐𝑜𝑠 (𝜁) 𝐶𝑜𝑠 (𝜁) = 𝑠2 + 𝑟2 — 𝑎2 2 — (𝑑4 + 𝑑6) 2 2 𝑥 𝑎2 𝑥 (𝑑4 + 𝑑6 ) (5.3) now, we should have the value of (s) and (r) in term of pxtcp, pytcp, pztcp and θ1. 𝑆 = (𝑃𝑧𝑡𝑐𝑝 — 𝑑1) 𝑟 = ± √ (𝑃𝑥𝑡𝑐𝑝 — 𝑎1 𝑐𝑜𝑠 (𝜃1)) 2 + (𝑃𝑦𝑡𝑐𝑝 — 𝑎1 𝑠𝑖𝑛 (𝜃1)) 2 sub. (s) and (r) in (5.3) yield: 𝐶𝑜𝑠 (𝜁) = (𝑃𝑧𝑡𝑐𝑝 − 𝑑1) 2 +(𝑃𝑥𝑡𝑐𝑝 − 𝑎1 𝐶𝑜𝑠 𝜃1) 2 +(𝑃𝑦𝑡𝑐𝑝 −𝑎1 𝑆𝑖𝑛 𝜃1) 2 −𝑎2 2 −(𝑑4 + 𝑑6) 2 2 𝑥 𝑎2 𝑥 (𝑑4 + 𝑑6 ) journal of mechanical engineering and technology issn: 2180-1053 vol. 9 no.2 july – december 2017 12 𝑆𝑖𝑛 (𝜁) = ±√1 − 𝐶𝑜𝑠2 (𝜁) 𝜁 = 𝑎𝑡𝑎𝑛2 (𝑆𝑖𝑛 (𝜁),𝐶𝑜𝑠 (𝜁)) 𝐹𝑖𝑛𝑎𝑙𝑙𝑦,𝜃3 = − (90 + 𝜁) (5.4) the negative sign in θ3 indicates that the rotation occurred in the opposite direction. likewise, we can follow the same procedure to solve for θ2 using similar trigonometric relationships. 𝜃2 = 𝛺 – 𝜆 𝛺 = 𝑎𝑡𝑎𝑛2 (𝑠,𝑟) 𝜆 = 𝑎𝑡𝑎𝑛2 ((𝑑4 + 𝑑6 )sin(𝜁) ,𝑎2 +(𝑑4 + 𝑑6 )cos(𝜁)) 𝜃2 = 𝑎𝑡𝑎𝑛2 (𝑠,𝑟) – 𝑎𝑡𝑎𝑛2 ((𝑑4 + 𝑑6 )sin(𝜁) ,𝑎2 +(𝑑4 + 𝑑6 )cos(𝜁)) substitute the values of (s) and (r) yield: 𝜃2 = 𝑎𝑡𝑎𝑛2 ((𝑃𝑧𝑡𝑐𝑝 −𝑑1),±√ (𝑃𝑥𝑡𝑐𝑝 — 𝑎1 cos(𝜃1)) 2 +(𝑃𝑦𝑡𝑐𝑝 — 𝑎1 sin(𝜃1)) 2 ) − 𝑎𝑡𝑎𝑛2 ((𝑑4 + 𝑑6 ) 𝑠𝑖𝑛 (𝜁) ,𝑎2 + (𝑑4 + 𝑑6 ) 𝑐𝑜𝑠 (𝜁)) again the rotation occurred in the opposite direction of the z axis as well as there are an initial rotation of 90 0 between axis 1 and axis 2. thus, the final value of θ2 is equal to: 𝜃2 = – ((𝛺 – 𝜆) – 90) (5.5) it is important to say that any position within the robot workspace can be achieved with many orientations. therefore, multiple solutions exist for the variables ɵ1, ɵ2 and ɵ3 due to the nature of trigonometric functions. in general, the problem of inverse kinematics may have eight solutions for the most six dof manipulators (nicolescu, ilie, & alexandru, 2015). as noticed above, every solution step resulted in two values that will be used in the next step, and so on. for example, there are four solutions for ζ that resulted from two different values of ɵ1 [ɵ1 and ɵ1'], this procedure gives four solutions for θ3 [ɵ3 ɵ3' ɵ3a ɵ3a'] and eight solutions for ɵ2 [ɵ2 ɵ2' ɵ2a ɵ2a' ɵ2b ɵ2b' ɵ2c ɵ2a'], each set of solution corresponds to different robot configurations of elbow-up and elbow-down representations. these values are listed in table 4 below to illustrate all the possible solution set. forward and inverse kinematics analysis and validation of the abb irb 140 industrial robot issn: 2180-1053 vol. 8 no.2 july – december 2017 13 table 4. possible solution set solution theta1 theta3 theta2 set 1 ɵ1 ɵ3 ɵ2 set 1 2 ɵ2' 3 ɵ3' ɵ2a set 2 4 ɵ2a' 5 ɵ1' ɵ3a ɵ2b set 3 6 ɵ2b' 7 ɵ3a' ɵ2c set 4 8 ɵ2c' 5.2 analytical solution after solving the first inverse kinematic sub-problem which gives the required position of the end effector, the next step of the inverse kinematic solution will deal with the procedure of solving the orientation sub-problem to find the joint angles ɵ4, ɵ5 and ɵ6. this can be done using z-y-x euler's formula. as the orientation of the tool frame with respect to the robot base frame is described in term of z-y-x euler's rotation, this means that each rotation will take place about an axis whose location depends on the previous rotation (craig, 2005). the z-y-x euler's rotation is shown below in figure 6. figure 6. z—y—x euler rotation the final orientation matrix that results from these three consecutive rotations will be as follow: 𝑅6 0 = rz'y'x' = rz (α) ry (β) rx (γ) 𝑅6 0 = ( 𝑐α −𝑠α 0 𝑠α 𝑐α 0 0 0 1 ) 𝑋( 𝑐β 0 𝑠β 0 1 0 −𝑠β 0 𝑐β ) 𝑋( 1 0 0 0 𝑐γ −𝑠γ 0 𝑠γ 𝑐γ ) 𝑅6 0 = ( 𝑐α𝑐β 𝑐α𝑠β𝑠γ − 𝑠α𝑐γ 𝑐α𝑠β𝑐γ +𝑠α𝑠γ 𝑠α𝑐β 𝑠α𝑠β𝑠γ +𝑐α𝑐γ 𝑠α𝑠β𝑐γ −𝑐α𝑠γ −𝑠β 𝑐β𝑠γ 𝑐β𝑐γ ) journal of mechanical engineering and technology issn: 2180-1053 vol. 9 no.2 july – december 2017 14 recall the forward kinematic equation, 𝑅3 0 = ( 𝑐1𝑐23 −𝑐1𝑠23 −𝑠1 𝑠1𝑐23 −𝑠1𝑠23 𝑐1 −𝑠23 −𝑐23 0 ) 𝑅6 3 = ( 𝑅) 𝑅6 𝑇 0 3 0 𝑅6 3 = ( 𝑐1𝑐23 𝑠1𝑐23 −𝑠23 −𝑐1𝑠23 −𝑠1𝑠23 −𝑐23 −𝑠1 𝑐1 0 )𝑥( 𝑐α𝑐β 𝑐α𝑠β𝑠γ − 𝑠α𝑐γ 𝑐α𝑠β𝑐γ +𝑠α𝑠γ 𝑠α𝑐β 𝑠α𝑠β𝑠γ +𝑐α𝑐γ 𝑠α𝑠β𝑐γ −𝑐α𝑠γ −𝑠β 𝑐β𝑠γ 𝑐β𝑐γ ) 𝑅6 3 = ( 𝑔11 𝑔12 𝑔13 𝑔21 𝑔22 𝑔23 𝑔31 𝑔32 𝑔33 ) however, it can be concluded that the last three intersected joints form a set of zyz euler angles with respect to frame {3}. therefore, these rotations can be expressed as: rz'y'z' = 𝑅6 3 = rz (α) ry (β) rz (γ) 𝑅6 3 = ( 𝑐α −𝑠α 0 𝑠α 𝑐α 0 0 0 1 ) 𝑋 ( 𝑐β 0 𝑠β 0 1 0 −𝑠β 0 𝑐β ) 𝑋( 𝑐γ −𝑠γ 0 𝑠γ 𝑐γ 0 0 0 1 ) 𝑅6 3 = ( 𝑐α𝑐β𝑐γ −𝑠α𝑠γ −𝑐α𝑐β𝑠γ −𝑠α𝑐γ 𝑐α𝑠β 𝑠α𝑐β𝑐γ +𝑐α𝑠γ −𝑠α𝑐β𝑠γ +𝑐α𝑐γ 𝑠α𝑠β −𝑠β𝑐γ 𝑠β𝑠γ 𝑐β ) where 𝑅6 3 is given above as 𝑅6 3 = ( 𝑔11 𝑔12 𝑔13 𝑔21 𝑔22 𝑔23 𝑔31 𝑔32 𝑔33 ) it is possible now to use the zyz euler's angles formula to obtain the solutions for ɵ4, ɵ5 and ɵ6 where 𝜃5 = 𝛽 = 𝑎𝑡𝑎𝑛2(+√𝑔31 2 +𝑔32 2 ,𝑔33) (5.6) 𝜃4 = 𝛼 = 𝑎𝑡𝑎𝑛2( 𝑔32 𝑠𝛽 , −𝑔31 𝑠𝛽 ) (5.7) 𝜃6 = 𝛾 = 𝑎𝑡𝑎𝑛2( 𝑔23 𝑠𝛽 , 𝑔13 𝑠𝛽 ) (5.8) forward and inverse kinematics analysis and validation of the abb irb 140 industrial robot issn: 2180-1053 vol. 8 no.2 july – december 2017 15 for each of the eight solutions achieved from the geometric approach for ɵ1, ɵ2 and ɵ3, there is another flipped solution of ɵ4, ɵ5 and ɵ6 that can be obtained as: 𝜃5 ′ = 𝛽′ = 𝑎𝑡𝑎𝑛2(−√𝑔31 2 +𝑔32 2 ,𝑔33), 𝜃4 ′ = 𝛼′ = 𝑎𝑡𝑎𝑛2( 𝑔32 𝑠𝛽′ , −𝑔31 𝑠𝛽′ ), 𝜃6 ′ = 𝛾′ = 𝑎𝑡𝑎𝑛2( 𝑔23 𝑠𝛽′ , 𝑔13 𝑠𝛽′ ) however, if β = 0 or 180, this means that the robot in a singular configuration where the joint axes 4 and 6 are parallel. this results in a similar motion of the last three intersection links of the robot manipulator. alternatively: if β = 𝜃5 = 0, the solution will be 𝜃4 = 𝛼 = 0 𝜃6 = 𝛾 = 𝑎𝑡𝑎𝑛2 (−𝑔12,𝑔11) and if β = 𝜃5 = 180, the solution will be 𝜃4 = 𝛼 = 0 𝜃6 = 𝛾 = 𝑎𝑡𝑎𝑛2 (𝑔12,−𝑔11) 6.0 inverse kinematic validation the home position of the robot in space is chosen to check the validity of the inverse kinematic solution. this position can be represented by a point (ptcp) in the robot workspace. this point describes the position of the tool centre point (tcp) with respect to the robot base frame. by applying the inverse kinematic equations derived above, a set of joint angles is achieved. however, some of these angles do not yield a valid solution which is simply due to the fact that not all the joints can be rotated by 360 0 . ptcp (home position) = [𝑝𝑥𝑡𝑐𝑝 𝑝𝑦𝑡𝑐𝑝 𝑝𝑧𝑡𝑐𝑝] t = [515 0 712] t after performing the calculations in matlab, four sets of solution were obtained as shown in table 5 below: or simply 𝜃4 ′ = 180+ 𝜃4 or simply 𝜃5 ′ = − 𝜃5 or simply 𝜃6 ′ = 180+ 𝜃6 journal of mechanical engineering and technology issn: 2180-1053 vol. 9 no.2 july – december 2017 16 table 5. all possible inverse kinematics solutions ɵ1 ɵ3 ɵ2 set 0 -180 102 set 1 0 0 0 set 2 -102 180 -153 93.7 set 3 -23 -27 23 set 4 -93.7 however, because of the limitation on the joint angle range of movement, especially joints 2 and 3, some of these solutions are not valid. nevertheless, they are shown above only to illustrate the calculation process. after that, these possible solutions are compared with the joint angle limits, listed below, and only valid solutions are presented in matlab. table 6. abb irb 140 joint angle limits (abb, 2000) joint angle max min ɵ1 180 -180 ɵ2 110 -90 ɵ3 50 -230 ɵ4 200 -200 ɵ5 115 -115 ɵ6 400 -400 after filtering all the possible solutions according to the joint angle limitation, only three valid solutions were achieved as shown in table 7. table 7. the valid inverse kinematics solutions ɵ1 ɵ2 ɵ3 set 0 0 0 1 st 180 -23 -153 2 nd 0 102 -180 3 rd the three solutions, shown above, actually represent different robot configurations of the home position. these are default, elbow-up and elbow-down representations. the elbow-up configuration that corresponds to joint angles (180, -23, -153) is shown in figure 7 below, while figure 8 shows the elbow-down configuration that corresponds to joint angles (0, 102, 180). finally, the set (0, 0, 0) represents the default home position. it is important to note that the position vector in robot studio is given for the tcp with respect to the robot global reference frame. thus to match our solution with the simulation in robot studio, the inverse kinematics was solved with respect to the robot’s tcp. forward and inverse kinematics analysis and validation of the abb irb 140 industrial robot issn: 2180-1053 vol. 8 no.2 july – december 2017 17 figure 7. elbow-up configuration figure 8. elbow-down configuration 7.0 conclusions this work was undertaken to build the forward and inverse kinematic models of the abb irb 140 industrial manipulator. the denavit-hartenberg analysis (dh) is introduced to form the homogeneous transformation matrices. from the derived kinematic equations, it can be concluded that the position of the robot is given as a function of ɵ1, ɵ2 and ɵ3 only, while the three last intersection joint angles (ɵ4, ɵ5 and ɵ6) are used to give the desired orientation in space. the position vectors (x, y and z) obtained from the kinematic equations were matched with the actual robot position in the lab for the same joint angle input. therefore, it can be declared that the kinematic derivation was carried out successfully. two approaches have been presented to solve the inverse kinematic problem. those were the geometrical and analytical approaches. multiple solutions have been produced due to the nature of trigonometric functions. however, it has been shown that not all the solutions that resulted from the inverse kinematics were valid. this is basically due to the physical restrictions on the joint angle range of movement. a simulation of the manipulator in robot studio has been introduced to prove the validity of the inverse kinematic model. it is also used to validate the written matlab code. journal of mechanical engineering and technology issn: 2180-1053 vol. 9 no.2 july – december 2017 18 8.0 references abb. (2000). irb 140 m2000 product specification. retrieved december 01, 2015 http://www.abb.com carter, t. j. (2009). the modeling of a six degree-of-freedom industrial robot for the purpose of efficient path planning. the pennsylvania state university. craig, j. j. (2005). introduction to robotics: mechanics and control: pearson/prentice hall upper saddle river, nj, usa:. deshpande, v. a., & george, p. (2012). analytical solution for inverse kinematics of scorbot-er-vplus robot. international journal of emerging technology and advanced engineering, 2(3). feng, y., yao-nan, w., & yi-min, y. (2014). inverse kinematics solution for robot manipulator based on neural network under joint subspace. international journal of computers communications & control, 7(3), 459-472. jazar, r. n. (2010). theory of applied robotics (vol. 1): springer. khatamian, a. (2015). solving kinematics problems of a 6-dof robot manipulator. paper presented at the proceedings of the international conference on scientific computing (csc). mitra, a. k. (2012). joint motion-based homogeneous matrix method for forward kinematic analysis of serial mechanisms. int. j. emerg. technol. adv. eng, 2, 111-122. neppalli, s., csencsits, m. a., jones, b. a., & walker, i. d. (2009). closed-form inverse kinematics for continuum manipulators. advanced robotics, 23(15), 2077-2091. nicolescu, a.-f., ilie, f.-m., & alexandru, t.-g. (2015). forward and inverse kinematics study of industrial robots taking into account constructive and functional parameter's moduling. proceedings in manufacturing systems, 10(4), 157. paul, r. p. (1981). robot manipulators: mathematics, programming, and control: the computer control of robot manipulators: richard paul. pires, j. n. (2007). industrial robots programming: building applications for the factories of the future: springer. samer y, h. a., moghavvemi m. (2009). a new geometrical approach for the inverse kinematics of the hyper redundant equal length links planar manipulators. http://www.abb.com/ forward and inverse kinematics analysis and validation of the abb irb 140 industrial robot issn: 2180-1053 vol. 8 no.2 july – december 2017 19 selig, j. (2013). geometrical methods in robotics: springer science & business media. shahinpoor, m. (1987). a robot engineering textbook: harper & row publishers, inc. siciliano, b., sciavicco, l., villani, l., & oriolo, g. (2010). robotics: modelling, planning and control: springer science & business media. spong, m. w., hutchinson, s., & vidyasagar, m. (2006). robot modeling and control (vol. 3): wiley new york. uicker, j. j., pennock, g. r., & shigley, j. e. (2011). theory of machines and mechanisms (vol. 1): oxford university press new york. vicente, d. b. (2007). modeling and balancing of spherical pendulum using a parallel kinematic manipulator. journal of mechanical engineering and technology issn: 2180-1053 vol. 9 no.2 july – december 2017 20 issn: 2180-1053 vol. 9 no.1 january – june 2017 33 anisotropic damage modelling of composite plates and shells d. benzerga1*, a. chouiter1, a. haddi2 and a. lavie2 1university of sciences and technology of oran, mechanical department, b.p. 1505, 31000 oran, algeria 2univ. artois, ea 4515, laboratoire de génie civil et géo-environnement (lgcge), béthune, f-62400, france abstract this paper deals with the analysis of geometrically nonlinear structures (plates, shells) of laminated composite materials by taking into account delamination phenomenon. we suggest a contribution to the modelling of the fibre reinforced composite materials damage in order to be able to predict the delamination of structures made of this type of material. the damage behaviour in large displacements was limited to the elastic domain with hypotheses of the moderate rotation theory including delamination phenomenon in the constitutive equations based on a damage model. this damage model was based on the use of damage mechanics considering three modes of interlaminar degradation which are associated with three modes of crack opening. cracking has been described by a weakening of three stiffnesses acting in the three directions and damage variables are associated with the degradation of the stiffness matrix. numerical simulations based on the finite element method were carried out predict the damage initiation and propagation of composite structures. the numerical results are compared with a number of similar results reported in the literature. keywords: modelling; anisotropic damage; delamination, laminate composite 1.0 introduction for many years, a large number of damage models have been developed for multiple applications, from ductile materials to almost brittle behaviour. the mechanics of damage have changed considerably since kachanov's first work (kachanov, 1958). this formalism has shown its advantages in many applications. without being exhaustive, we can cite (chaboche, 1989), (krajcinovic, 2010), and (babu, 2010). many models written in this framework are dedicated to the behaviour of composites. different approaches can be distinguished: some study the microscopic scale and, by homogenization technique, deduce the behaviour on a larger scale (rami, 2007 blassiau, 2005), others are placed on the mesoscopic scale (that of the plies ud). in this latter group, several very satisfactory models of damage have been developed. we can cite the mesodel of cachan developed by (ladevèze et al, 2000) or the onera model (laurin et al, 2007). composite materials are inhomogeneous and generally anisotropic solids consisting of two or more materials of different natures. the model presented here is placed on a mesoscopic scale and is based on that developed by (bui et al, 2017), * corresponding author e-mail: djeb_benz@yahoo.fr journal of mechanical engineering and technology 34 issn: 2180-1053 vol. 9 no.1 january – june 2017 (babu, 2010), (hinton et al, 2004) for composite materials. it is based on the theory of representation of tensor functions (boehler, 1978), which are particularly suited to the modelling of anisotropic materials. it is thus desired to explore the capacities of these tools to reproduce the degradation of laminated composite shell subjected to stresses in large displacements. a formulation for anisotropic damage is established in the framework of the principle of strain equivalence. the damage variable is still related to the surface density of microcracks and microvoids and is represented by a second-order tensor. the coupling of the damage with the elasticity is written in tensor form. the evolution law is an extension of the classical law of isotropic damage. the damage tensor allows to link the actual stresses to their nominal quantities which are measurable externally (observable variables). the continuous description of the damage makes it possible to represent finely the initiation and propagation of the delamination in the stratified composite structures. the numerical results of delamination simulations are compared with those obtained from the literature give a good validation of the anisotropic model developed in this study. 2.0 laminated composite shell the laminated shell is composed of a finite number of layers parallel to a reference surface ω, chosen to coincide with the average surface of the first layer (figure 1). each layer has different physico-mechanical properties and different fiber orientations. the behavior of each layer is linearly elastic and anisotropic. the delamination is due to the interlaminar stresses which are exerted on weak interfaces. the material constituting each layer is assumed to be homogeneous and anisotropic, the anisotropy being symmetrical with respect to the reference surface ω ( 0 3  ). our material possesses three planes of symmetry, it is orthotropic (kreja & schmidt, 2005). the composite shell consists of orthotropic materials reinforced with fibers embedded in layers. each layer is characterized by its orthotropy reference  i such that the axis 1 is aligned with the direction of the fibers, while the axis 3  is normal to the average surface (figure 1). in the system of axes i , the components of the elasticity tensor relative to the reference axes of the fibers of the layer l are dcba abc d ˆˆˆˆc ~ eeee  ll c (1) the tensor of the orthotropic material l c ~ can be written in matrix form   6x6mijkl l c ~ :   66 1212 1313 2323 333322331133 223322221122 113311221111 00000 00000 00000 000 000 000 xl l l l ll lll lll ijk l l c ~ c ~ c ~ c ~ c ~ c ~ c ~ c ~ c ~ c ~ c ~ c ~ c ~                      (2) anisotropic damage modelling of composite plates and shells issn: 2180-1053 vol. 9 no.1 january – june 2017 35 where the components ijk l l c ~ take the following values (barbero et al, 1990): . eee νν2ννννννν1 δ gc ~ ,gc ~ ,gc ~ , δee νν1 c ~ , δee ννν δee ννν c ~ δee νν1 c ~ , δee ννν δee ννν c ~ δee ννν δee ννν c ~ , δee νν1 c ~ 321 133221133132232112 12 1212 l13 1313 l23 2323 l 21 21123333 l 21 132123 31 3112322233 l 31 31132222 l 31 131213 32 3221311133 l 31 133212 32 2331121122 l 32 32231111 l                      avec (3) the elasticity tensor l c is determined by the 9 constants independent of the relations (3). the tensor l c must also be expressed in the coordinate system   i g using following transformations (figure 1) l ggggeeee  kji ijkl dcba abcd cˆˆˆˆc ~ lll c . (4) which leads to      abc d dcba ijk l c ~ ˆˆˆˆc ll lkji egegegeg  . (5) the basic vectors in the coordinate systems i  and i  are linked by the following relation: ia i a ˆ ge      (6) for the layer l, the relation (5) becomes, then abcd d l c k b j a i ijkl c ~ c ll                  (7) journal of mechanical engineering and technology 36 issn: 2180-1053 vol. 9 no.1 january – june 2017 figure 1. main coordinates of a layer  i and the curvilinear coordinates  i . the coordinate 3  always remains normal to the reference surface ω and since 33   we obtain .ˆ,ˆ,ˆ a 100 3 3 3 3  egegeg  (8) when the vectors of the space base   i g are orthogonal, one obtains 22 2 2 11 2 1 22 1 2 11 1 1 g cos ˆ, g sin ˆ, g sin ˆ, g cos ˆ   egegegeg (9) where  g (no summation) are the components of the metric tensor and  the angle indicating the orientation of the fibres (figure 1). for an orthotropic material we have 0 3   l c and 0 333   l c (merzouki et al, 2007), therefore .ececs ecs ececs ll l ll 33 33333333 3 333 33 33 2           (10) greek indices take the values 1 and 2. the above relationships can be written in matrix form by defining the matrix   6x1mijs and  6x1mije . then, anisotropic damage modelling of composite plates and shells issn: 2180-1053 vol. 9 no.1 january – june 2017 37                                                              12 13 23 33 22 11 1212331222121112 13132313 23132323 3312333322331133 2212223322221122 1112113311221111 12 13 23 33 22 11 2 2 2 00 0000 0000 00 00 00 e e e e e e cccc cc cc cccc cccc cccc s s s s s s llll ll ll l l ll llll llll . (11) the matrix   6x6mijkl l c in the curvilinear coordinate system  i can also be determined using the base change matrix [t], such that      tc~tc ijkl l ijkl l  (12) where   6x6mt is denoted by           661221221122122111 1112 2122 2221 2 22 2 21 1211 2 12 2 11 000 0000 0000 000100 2000 2000 x dddddddd dd dd dddd dddd t                       (13) and j i eg ˆd ij  (see relationships (8) and (9). 3.0 anisotropic damage the damage in composite materials is due to the heterogeneities that give rise to stress concentrations. this case occurs at the interface between the fibre and the matrix, where decohesions can appear. anisotropy also causes stress concentrations, especially at the interface between adjacent plies of different orientations, causing delamination. it is difficult to define a typical scenario which would lead to the failure of laminated composite structures, as the mechanisms of damage are numerous and complex [ladeveze & lubineau, 2003]. however, the phenomenon of delamination remains one of the most important problems faced by laminated composite materials. the phenomenon of delamination constitutes a particular case of anisotropic damage. taking into account that the geometric effect of cavities and cracks, we can introduce on each element of area, spotted by its normal n  , an area reduction  n   and the state of damage is characterized by a tensor of second order which is expressed in its main coordinate system by jj j j nn     . 3 1  (14) journal of mechanical engineering and technology 38 issn: 2180-1053 vol. 9 no.1 january – june 2017 where j and jn  are the principal values and the unit vectors of the principal directions (directions which coincide with the axes of the material) of the tensor respectively. symmetric tensor of damage of second order are most commonly used because of their mathematical simplicity compared to the tensor of higher order. they can describe most of anisotropic damage. however, these tensors of second order cannot represent a complicated state of damage such as orthotropic damage. the second order damage tensor were often used in the development of theories of anisotropic damage (bui et al, 2017bielski et al, 2006chandra et al, 2008rajhi et all, 2014). damage variables used to link the effective stresses at their nominal quantities that are measurable externally (observable variables). the law taking into account the damage behaviour is then written     ees  c  (15) where  c  is the hooke elasticity tensor for damaged material. for a general state of deformation and damage, the nominal stress tensor can be connected to the tensor of the effective stresses by the following linear transformation sms   (16) which m is an operator of damage (order tensor 4). following the shape used for tensor m , it is clear that from equation (3), the stress tensor s is generally not symmetrical. for the symmetry of tensor s , (bui et al, 2017) use an energy equivalence instead of deformation equivalence and propose the following expression 2 1 2 1    ss  . (17) the fourth order tensor of the damaged material corresponding to equation (16) can be defined by 2 1 2 1    m (18) then the tensor is symmetric ijlkjikllklijijkl mmmm  (19) for a virgin material (not damaged), i which corresponds to the identity transformation  ss   . in the literature, one can find the damage tensor denoted by the variable d such that   1 di (20) anisotropic damage modelling of composite plates and shells issn: 2180-1053 vol. 9 no.1 january – june 2017 39 tensors det have the same principal axes and principal values are such that (andrew et al, 2008) j j d1 1   . (21) jd can be interpreted as the ratio of the area reduction in the plane perpendicular to jn  caused by the development of cracks (ganczarski et al, 2007 ). jd varies between 0 and 1 while  varies between 1 (virgin material) and  (completely damaged material).  is considered as an internal state variable which characterizes the anisotropy of distribution of the microcracks in the material. to determine the relationship between the strain tensor and stresses three approaches are used: strain equivalence, stress equivalence and energy equivalence. unlike the equity deformation and stress, energy equivalence induces the symmetry of the tensor of rigidity and flexibility. this approach recognizes that the elastic energy stored in the damaged material is the same as the elastic energy stored in the equivalent virgin material, where in the nominal quantities by the effective amounts is replaced (rouabhi, 2004): 1 1 : : 2 2 s e s e (22) from relations (17) and (22) is easily obtained 2 1 2 1   ee  (23) the degradation can be considered as the average effect of any microcracks developed in the material (marriage, 2003). in this context, it is assumed that the material obeys the general law of hooke in the damaged state (ghosh & sinha, 2005), then we can write   eces   . (24) combining equations (22) and (23) with equations (24) and (15), one obtain em :   cm e:cs 11  (25) where 1  is expressed by (rohwer, 2014) jj j j nn ω     3 1 11  (26) journal of mechanical engineering and technology 40 issn: 2180-1053 vol. 9 no.1 january – june 2017 taking into account the relation (21), equation (26) is written yet   jj j j nnd     3 1 1 1  (27) the fourth tensor of equation m then becomes                                                  2 1 21 00000 02 1 31 0000 002 1 32 000 000 1 3 00 0000 1 2 0 00000 1 1 2 1 2 1       m (28) with jj j j nn ω      3 1 2 1 1  . the coupling between the damages j  ( j variables representing the damage in the principal directions of the tensor) makes it possible to write the fourth order tensor m in the following form:                                                  2 1 12 00000 02 1 13 0000 002 1 23 000 000 1 33 00 0000 1 22 0 00000 1 11 2 1 2 1       m . (29) considering the relation (27), the fourth order tensor m of the relation (29) can be written as anisotropic damage modelling of composite plates and shells issn: 2180-1053 vol. 9 no.1 january – june 2017 41              .12,13,23,33,22,11ijpour1 ij d0 2 1 12 d100000 02 1 13 d10000 002 1 23 d1000 0002 1 33 d100 00002 1 22 d10 000002 1 11 d1 2 1 2 1 m                                         (30) we will then use the voigt notation for the components of the elasticity tensor such as: 612513423333222111  ;;;;; the orthotropic material elasticity tensor can be written as:   66 66 55 44 332313 232212 131211 00000 00000 00000 000 000 000 xl l l l ll lll lll ijk l l c ~ c ~ c ~ c ~ c ~ c ~ c ~ c ~ c ~ c ~ c ~ c ~ c ~                      (31) considering the relation (30), the constitutive law (equation 25) of the damaged orthotropic material can be written as: journal of mechanical engineering and technology 42 issn: 2180-1053 vol. 9 no.1 january – june 2017                         . e e e e e e c ~ d c ~ d c ~ d c ~ dc ~ dc ~ d c ~ dc ~ dc ~ d c ~ dc ~ dc ~ d s s s s s s l l l l ll lll lll                                                                    12 13 23 33 22 11 66 12 55 13 44 23 33 3 23 2 13 1 23 3 22 2 12 1 13 3 12 2 11 1 12 13 23 33 22 11 2 2 2 100 010 001 000 000 000 000 000 000 111 111 111 (32) 4.0 delamination the interface between two plies may break under local stress peel and / or shear 1 3 s and 2 3 s (figure 2). it then creates an interlaminar fracture called delamination. figure 2. interlaminar stresses responsible of delamination delamination only affects terms of shear and normal of the part above the plane strain field. therefore the behaviour of law (10) for an orthotropic material, is written 33 33 33 3 3 3 3 3 33 33 3333 33 33 (1 ) 2(1 ) (1 ) . l l l l l s c e d c e s d c e s c e d c e                    (33) greek indices take the values 1 and 2. the damages are closely coupled because the same microcracks are participating in the delamination phenomenon. therefore, (gupta et al, 2005) introduced an evolution law as: anisotropic damage modelling of composite plates and shells issn: 2180-1053 vol. 9 no.1 january – june 2017 43       33 33 33 13 1 13 33 13 23 2 23 33 23 1 1 1 1 1 1 1 1 d y if d d otherwise d y if d and d d otherwise d y if d and d d otherwise                     (34) where   n f yy yy n n y             0 0 1    with       1 33 1 13 2 23 y y y y            energy release rate equivalent [gornet & sinha, 2000]. 1 and 2 are coupling parameters between shear and transverse energy, and  another material parameter used to describe the shape of the fracture surface. in fracture mechanics  is determined by delamination tests in mixed mode [aiello et al., 2003] and the rate of energy release in the three modes of delamination failures are defined as:   20 13 13 13 1 2 y k u   20 23 23 23 1 2 y k u   20 33 33 33 1 2 y k u (35) mode iii mode ii mode i 0 3ik are the interface stiffness and  3,2,1 0 3 iui represent the relative displacements at the interface. we can observe that the more the interface is ‘strong’ and the more the coefficient n is important. 0y and fy are initiation and rupture threshold delamination (see benzerga, et al, 2014). 5.0 application examples in this section, numerical investigations are presented and our results are compared to analytical solutions found in literature in order to validate our model. 5.1 composite plate (0/90) the first plate is composed of two layers (0/90) cantilever subjected to a shearing force evenly distributed at its free end (figure 3). this plate has 100 mm of length (l), 10 mm of width (b) and 3 mm of thickness (h). the mechanical properties of laminated composite are presented in table 1. table 1. mechanical characteristics of composite young’s modulus shear modulus poisson coefficient e11 e22=e33 g12=g13 g23 12=13=23 1.0 x 103 gpa 0.3 x 103 gpa 0.15 x 103 gpa 0.12 x 103 gpa 0.25 journal of mechanical engineering and technology 44 issn: 2180-1053 vol. 9 no.1 january – june 2017 figure 3. composite plate (0/90) cantilevered the second plate is composed of two layers (0/90) subjected to a uniform pressure (figure 4). this plate has 228.6 mm of length (l), 38.1 mm of width (b) and 1.016 mm of thickness (h). the mechanical properties of laminated composite are presented in table 2. table 2. mechanical characteristics of composite young’s modulus shear modulus poisson coefficient e11 e22=e33 g12=g13 g23 12=13=23 140 gpa 9.8 gpa 4.9 gpa 0.84 gpa 0.3 figure 4. composite plate (0/90) under pressure figure 5 presents the evolution of the applied force or pressure versus displacement. this figure illustrates the comparison between the present study and analytical results from (kreja & schmidt, 2005) and (arciniega, 2005). analytical results correspond to the shell theory without taking into account the damage phenomenon. it should be noted that, from zero to a, the two curves have similar behaviour with an absence of any damage phenomena. from point a, the two curve diverge resulting from the damage initiation and propagation at the composite interface. however, the nature of applied load has an influence on the initiation of delamination. the results obtained show that for applied force the damage initiation starts at 60 mm of displacement (figure 5a), whereas for applied pressure the damage starts at 10 mm of displacement (figure 5b). anisotropic damage modelling of composite plates and shells issn: 2180-1053 vol. 9 no.1 january – june 2017 45 (a) (b) figure 5. displacement of the center of the plate (a) shearing force and (b) under pressure loading journal of mechanical engineering and technology 46 issn: 2180-1053 vol. 9 no.1 january – june 2017 5.2. stretching of an open thin cylinder this example is the most requested test in the literature for the analysis of isotropic shells moderate rotations. currently, composite variations of this test have been proposed. a laminate cylinder (0/90) of short length subjected to two opposing forces in its central section, both ends remain free (see figure 6a). this cylinder has 131.445 mm of length (l), 125.8062 mm of radius (r) and 2.3876 mm of thickness (h). the cylinder is characterized by the geometrical data and the following mechanical properties (table 3). table 3. mechanical characteristics of cylinder young’s modulus shear modulus poisson coefficient e11 e22=e33 g12=g13 g23 12=13=23 213.5 mpa 73.5 mpa 28 mpa 28 mpa 0.3125 (a) a laminate cylinder (0/90) subjected to two opposing forces in its central section (b) displacement curve of laminate cylinder center a figure 6. behaviour cylinder (0/90) under two opposing forces in its central section anisotropic damage modelling of composite plates and shells issn: 2180-1053 vol. 9 no.1 january – june 2017 47 5.3 laminated cylindrical panel secured at both ends subjected to a concentrated load in the middle. figure 7 shows a cylindrical laminate panel (0/90/0/90) is submitted to concentrated force in its centre. the mechanical and geometrical properties are presented in table 4 and 5. table 4. mechanical characteristics of cylinder young’s modulus shear modulus (mpa) poisson coefficient e11 e22=e33 g12=g13 g23 12=13=23 143.22 x 103 mpa 34.44 x 103 gpa 17.76 x 103 mpa 17.76 x 103 mpa 0.313 table 5. geometry of cylinder length radius angle thickness (mm) l (mm) r (mm)  h (mm) 139.7 304.8 0.5 1.016 (a) laminated cylindrical panel submitted to concentrated force in its centre 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 0 200 400 600 800 1000 kreja. i, schmidt. r. 2005 our model a p p li e d f o rc e a t c e n tr a l p o in t, n displacement of central point, mm (b) displacement of laminated cylindrical panel center figure 7. behaviour of laminated cylindrical panel subjected to a concentrated load at its centre. journal of mechanical engineering and technology 48 issn: 2180-1053 vol. 9 no.1 january – june 2017 figure 7 shows the result of the central displacement of cylindrical panel according to the central load. it can be seen that the buckling load predicted by our approach is close to that obtained by (kreja and schmidt 2005). however, a difference appears in the post-buckling region when the delamination begins to appear. 5.4. semi-cylindrical shell figure 8 shows the laminated half-cylinder (0/90/0) subjected to a concentrated load applied to its free end. this cylinder has 304 mm of length (l), 101.6 mm of radius (r) and 3.3 of thickness (h). the material composite properties are presented in table 6. table 6. mechanical characteristics of cylinder young’s modulus shear modulus poisson coefficient e11 e22=e33 g12=g13 g23 12=13=23 2068.5 mpa 517.125 mpa 795.6 mpa 198.89 mpa 0.3 figure 8 shows the displacements of the point a compared to those of reference [arciniega aleman, 2005]. the results are quite close before the appearance of delamination phenomenon. (a) laminated half-cylinder subjected to force in its center anisotropic damage modelling of composite plates and shells issn: 2180-1053 vol. 9 no.1 january – june 2017 49 (b) displacement of the centre a of laminated half-cylinder figure 8. behaviour of the laminated half-cylinder submitted to load in its centre 6.0 conclusions an anisotropic damage model, based on the concept of continuum damage mechanics, was developed to simulate the delamination phenomenon in laminated composite shells composed of unidirectional plies subjected to large displacement. these composite structures have the enormous advantage of being able to adapt to any loading by orienting the fibers according to the direction of stresses. the behavior was limited to the elastic domain with the assumptions of the moderate rotational theory (mrt5) including the delamination phenomenon in the constitutive equations. using the ansys programmable langauage, subroutine was developed and implemented in the main code. different geometries have been used to validate the damage model presented in this study for predicting delamination initiation and propagation. the results of these first simulations, are in good agreement with the results obtained from the literature, indicate that the proposed model is able to describe the degradation modes in composite structures. they open the way to many perspectives. in the first stage, it is now necessary to compare the simulations on composite structures, whose stacking sequence is more complex. it is also desired to introduce into the developed model all the types of degradation to which a laminated shell could be confronted. journal of mechanical engineering and technology 50 issn: 2180-1053 vol. 9 no.1 january – june 2017 references andrew t. r., richard, b., & giles w. h. (2008). post-buckled propagation model for compressive fatigue of impact damaged laminates. international journal of solids and structures. 45, 4349–4361 arciniega, a. r. a. (2005). on a tensor-based finite element model for the analysis of shell structures. dessertation submitted to the office of graduate of texas a & m university, in partial fulfilment of requirements for degree of phd. babu, r.r., benipal g. s., & singh a. k. (2010) constitutive model for bimodular elastic damage of concrete. latin american journal of solids and structures. 7(2). barbero, e. j., reddy, j. n., & teply, j. l. (1990). general two-dimensional theory of laminated cylindrical shells. american institute of aeronautics and astronautics journals. 28 (3) 544–553. benzerga, d., haddi, a., lavie, a. (2014). delamination model using damage mechanics applied to new composite for orthopaedic use, international journal of materials engineering, 4(3). 103–113. bielski, j., skrzypek, j. j., & kuna-ciskal, h. (2006). implementation of a model of coupled elastic-plastic unilateral damage material to finite element code. international journal of damage mechanics, sage publications, 15 (1). 5–39. blassiau, s. (2005). modélisation des phénomènes microstructuraux au sein du composite unidirectionnel carbone/époxy et prédiction de durée de vie : contrôle et qualification de réservoirs bobinés. rapport de thèse, ecole des mines de paris. boehler, p. (1978). lois de comportement anisotrope des milieux continus. j meca, vol.17, pp 153 – 190. bui, t.a., wong, h., deleruyelle, f., xie, l.z., &tran, d.t. (2017). a thermodynamically consistent model accounting for viscoplastic creep and anisotropic damage in unsaturated rocks », international journal of solids and structures, volume 117, pages 26-38. chaboche, j. l. (1989). continuum damage mechanics, parts ii and i. asme j appl mech, pp. 55-59 and 55-65. chandra, v. s., & ramesh, t. (2008). analysis of multiple off-axis ply cracks in composite laminates. international journal of solids and structures. volume 45, issue 16, 1 august, pages 4574-4589 http://www.sciencedirect.com/science/article/pii/s0020768308001121 http://www.sciencedirect.com/science/article/pii/s0020768308001121 http://www.sciencedirect.com/science/article/pii/s0020768308001121 http://www.sciencedirect.com/science/article/pii/s002076831730166x http://www.sciencedirect.com/science/article/pii/s002076831730166x http://www.sciencedirect.com/science/article/pii/s002076831730166x http://www.sciencedirect.com/science/article/pii/s0020768308001728 http://www.sciencedirect.com/science/article/pii/s0020768308001728 http://www.sciencedirect.com/science/journal/00207683/45/16 http://www.sciencedirect.com/science/journal/00207683/45/16 anisotropic damage modelling of composite plates and shells issn: 2180-1053 vol. 9 no.1 january – june 2017 51 ganczarski, a., barwacz, l., & ganczarski, a. (2007). low cycle fatigue based on unilateral damage evolution’, international journal of damage mechanics, vol 16, issue 2. ghosh, a., & sinha, p.k. (2005). initiation and propagation of damage in laminated composite shells due to low velocity impact. woodhead publishing ltd, u crash, vol. 10. gornet, l., leveque, d., & perret, p. (2000). modélisation, identification et simulations éléments finis des phénomènes de délaminage dans les structures composites stratifiées. mécanique and industries / mécanique et industries, edp sciences, numéro spécial matériaux composites. 1(3), 267-276. gupta, a.k., patel, b. p., & nath, y. (2015). progressive damage of laminated cylindrical/conical panels under meridional compression. european journal of mechanics a/solids. 53, 329-341. hinton, m.j., kaddour, a.s., & soden, p.d. (2004). evaluation of failure prediction in composite laminates: background to ‘part c’ of the exercise. computer science and technology. 64, 32 – 327. kachanov, l. m. (1958). on the time to failure under creep conditions. izv an sssr, otd teknh. 31,8-31. krajcinovic, d. (2010). continuum damage mechanics. applied mechanics review. 37. kreja, i., & schmidt, r. (2006). large rotations in first-order shear deformation fe analysis of laminated shells. international journal of nonlinear mechanics, 41(1), 101-123 ladevèze, p., & lubineau. g. (2003). on a damage mesomodel for laminates: micromechanics basis and improvement. mechanics of materials. 35(8), 763775. ladeveze, p., allix, o., deu, j-f., & eveque, d. (2000). a mesomodel for localization and damage computation in laminates. computer methods in applied mechanics and engineering. 183,105-122. laurin f., carrere, n., & maire, j-f. (2007). a multiscale progressive failure approach for composite laminates based on thermodynamical viscoelastic and damage models. composites: part a, 38,198-209. mariage, j. f. (2003). simulation numérique de l’endommagement ductile en forgeage de pièces massives. thèse de l’université de technologie de troyes. merzouki,t., chalal, h., & meraghni, f. (2007). identification of orthotropic material behaviour using the error in constitutive equation. experimental analysis of nano and engineering materials and structures proceedings of the 13th http://journals.sagepub.com/author/ganczarski%2c+a http://journals.sagepub.com/author/barwacz%2c+l http://journals.sagepub.com/author/ganczarski%2c+a http://www.sciencedirect.com/science/article/pii/s0997753815000571 http://www.sciencedirect.com/science/article/pii/s0997753815000571 journal of mechanical engineering and technology 52 issn: 2180-1053 vol. 9 no.1 january – june 2017 international conference on experimental mechanics, alexandroupolis, greece, july 1–6, 2007. rajhi, w., saanouni, k., & sidhom, h. (2014). anisotropic ductile damage fully coupled with anisotropic plastic flow: modeling, experimental validation, and application to metal forming simulation. international journal of damage mechanics. 23(8). rami, h. a., kilic, h., & mulina, a. (2007). nested nonlinear micromechanical and structural models for the analysis of thick –section composite materials and structures. computer science and technology. 67, 1993 – 2004. rohwer, k. (2014). predicting fiber composite damage and failure. journal of composite materials. doi: 10.1177/0021998314553885 issn 0021-9983. rouabhi, a. (2004). comportement et fragmentation dynamique des matériaux quasi fragiles. thèse de l’ecole nationale supérieure des mines de paris. http://journals.sagepub.com/author/rajhi%2c+w http://journals.sagepub.com/author/saanouni%2c+k http://journals.sagepub.com/author/sidhom%2c+h issn: 2180-1053 vol. 3 no. 1 january-june 2011 a study on the deposition of tin on the contacts subjected to high frequency impact loading in semiconductor device testing 1 a study on the deposition of tin on the contacts subjected to high frequency impact loading in semiconductor device testing k.y. ong1, md. radzai said2 1,2faculty of mechanical engineering, universiti teknikal malaysia melaka, hang tuah jaya, 76100 durian tunggal, melaka, malaysia email: 1okaywhy@gmail.com, 2radzai@utem.edu.my abstract semiconductor device leads were previously plated with lead (pb). due to environmental concern, the plating material has been switched to tin (sn). however, due to the softness of tin element plating, it often wears off and its residues deposited to the surface of the contacts used in semiconductor device testing. as a result, significant drop of contact performance has been observed. this study intends to find out a theoretical explanation to the problem of tin deposition. as soon as the root cause is identified, studies go into the details of how to solve this problem. one of the economical suggestions is to apply lubricants on the contact. a theoretical calculation is derived to verify if the lubricant would significantly altered the overall resistance. keywords: contact, tin, adhesive wear, semiconductor device testing. 1.0 introduction to date, the phenomenon of tin deposition on the electric contacts used in semiconductor device testing is still not very well understood in the industry. prior to going into the phenomenon itself, it is convenient to first understand the problem from the basics. we know that semiconductor devices are used in every electronic device around us. some of the notable semiconductor devices are diode, transistor, integrated circuits (ic), microprocessor etc. however, in reality, there are always mixtures of good and bad semiconductor devices in mass production. a “good” semiconductor device is one that is having product specifications within the tolerance limit and vice versa for a “bad” one. to verify if a particular semiconductor device is good or bad, it needs to be tested. issn: 2180-1053 vol. 3 no. 1 january-june 2011 journal of mechanical engineering and technology 2 there are many testing methods available in the market. some uses electro-pneumatic machine to pick semiconductor devices up and place them on electric contacts (or contactors) connected to a tester machine, the tester machine will read the semiconductor device and determine whether it is “good” or “bad”, this type of testing is often called the pick and place method in the industry. another popular type of testing is using gravitational force to draw the semiconductor devices downwards, as they passes through the electric contacts at the bottom which are also connected to the tester machine; they will be verified and sorted out. this type of testing method is industrially known as gravity fed method. common to all testing methods, the semiconductor device’ leads or pads must touches with the electric contacts (which can be either contacting blades or pogo pins depends on application) so that the current could flow through the device. the technology today has enable the process of picking, placing semiconductor devices on electric contacts for testing and picking it up from electric contacts to be done in merely 0.09 seconds. in this particular case, an electric contact is subjected to approximately 11 impact loadings from the leads of the semiconductor device in just one second. although achieving very high unit per hour (uph) performance, there is still one serious issue bothering the engineers. it is found that there are always some tin residues deposited on the electric contacts after several cycles of testing. the tin residues come from the plating material of the semiconductor device leads or pads. the increase of tin deposition on the contacts brings about an increase of electrical contact resistance as the contacting cycles increases. the usual practice has been to clean the contacts by grinding off the tin residues. however, this measure causes considerable damage to the contacts itself. thus, this paper studies about the possible root cause of the deposition or adhesion of tin on contacts used in semiconductor testing. 2.0 literature review before venturing further, it is important to first knowing what materials these contacting surfaces are made of. the lead of semiconductor device is often made of alloy 42 plated with pure tin (epistola, 2009). alloy 42 is an iron alloy consists of 41% of nickel, 0.8% of manganese, 0.5% of cobalt. as for the electric contacts used in semiconductor device testing as shown in figure 1, there are a huge variety of alloys used issn: 2180-1053 vol. 3 no. 1 january-june 2011 a study on the deposition of tin on the contacts subjected to high frequency impact loading in semiconductor device testing 3 for different applications, such as copper beryllium, nickel beryllium, tungsten, and some with gold or rhodium plating. 2 contacts to be done in merely 0.09 seconds. in this particular case, an electric contact is subjected to approximately 11 impact loadings from the leads of the semiconductor device in just one second. although achieving very high unit per hour (uph) performance, there is still one serious issue bothering the engineers. it is found that there are always some tin residues deposited on the electric contacts after several cycles of testing. the tin residues come from the plating material of the semiconductor device leads or pads. the increase of tin deposition on the contacts brings about an increase of electrical contact resistance as the contacting cycles increases. the usual practice has been to clean the contacts by grinding off the tin residues. however, this measure causes considerable damage to the contacts itself. thus, this paper studies about the possible root cause of the deposition or adhesion of tin on contacts used in semiconductor testing. 2.0 literature review before venturing further, it is important to first knowing what materials these contacting surfaces are made of. the lead of semiconductor device is often made of alloy 42 plated with pure tin (epistola, 2009). alloy 42 is an iron alloy consists of 41% of nickel, 0.8% of manganese, 0.5% of cobalt. as for the electric contacts used in semiconductor device testing as shown in figure 1, there are a huge variety of alloys used for different applications, such as copper beryllium, nickel beryllium, tungsten, and some with gold or rhodium plating. figure 1 contact materials there are various perspectives are being looked upon to identify why tin deposited on the contacts. it bears resemblance to solid-state welding, it can be seen as an adhesive wear, and the volume of wear can be calculated. the first concept to be taken serious attention is the concept of asperity. it has to be accepted that no surface are perfect. in practice, there are no ideal surfaces which are completely flat. so, it should be accepted that in certain level of magnification, the truth that surfaces are all rough and uneven. such uneven surfaces are called asperity. alloy 42 copper beryllium tin plating gold plating moving downwards and upwards figure 1: contact materials there are various perspectives are being looked upon to identify why tin deposited on the contacts. it bears resemblance to solid-state welding, it can be seen as an adhesive wear, and the volume of wear can be calculated. the first concept to be taken serious attention is the concept of asperity. it has to be accepted that no surface are perfect. in practice, there are no ideal surfaces which are completely flat. so, it should be accepted that in certain level of magnification, the truth that surfaces are all rough and uneven. such uneven surfaces are called asperity. beginning with the problem of tin deposition, the direct contact of the electric contacts with the semiconductor device leads can be understood in three levels of complexity, namely: • simple lead-tocontact contacting • lead-tocontact impact contacting • lead-tocontact impact contacting with current also, there is a high possibility of sliding contact between the lead to electric contact contacting, which will also be taken into consideration. for the first type of contact which neither involve repeating cycles of impact loading nor the interference of current flow, bares resemblance to solid-state welding. the principle of solid-state welding is demonstrated best with two clean surfaces being brought into atomic contact with each other under sufficient pressure, they form bonds and produce a joint. issn: 2180-1053 vol. 3 no. 1 january-june 2011 journal of mechanical engineering and technology 4 according to kalpakjian (2006), it is essential that the interface be free of oxide films, residues, metalworking fluids, other contaminants to form a strong bond. based on the criteria mentioned, it is easy to imagine that heat generated through friction improves the transfer of atoms across an interface. relative interfacial movements of the contacting surfaces occur, even at very small amplitudes will disturb the mating surfaces which will break up any oxide films, and generate new clean surfaces which improve the strength of the bond. if the pressure is high enough, plastic deformation will occur at the interface also promote stronger bonding. experiments also reveal that when adhesion measurements are performed in vacuum, where the degree of surface contamination is reduced and adhesion between metals become very large (bowden., and rowe, 1956). hence, summarizing from the experience gained from solid-state welding, the following criteria influences the bond strength of solid-state welding: • pressure applied to the interface of contacting surfaces • cleanliness of the contacting surfaces • heat generated from internal friction between contacting surface • relative interfacial movements that breaks up the contaminants • air density (strongest adhesion occurs in vacuum) the deposition of tin on the electric contacts could also be a result of adhesion or adhesive wear which is already being studied extensively in the field of tribology. adhesive wear occurs as a result of the transfer of softer metal (weak material) to the harder metal (stronger material) by means of friction. (stachowiak and batchelor, 2005). the question of practical interest is what properties of metal that favour such adhesive wear? buckley (1981) discovered that greatest adhesion occur between like materials. iron to iron adhesion for example exhibits the ratio of adhesion force to contact force which can be as high as 20. jellium model (ziman, 1963) explains the metal to metal adhesion as a result of electron transfer between contacting surfaces. when the distance between two surfaces become sufficiently close, i.e. <1nm, electrons which are not bound by the rigid structure can move from one body to another. this theorem relates the strength of adhesion to the electron density of metal. issn: 2180-1053 vol. 3 no. 1 january-june 2011 a study on the deposition of tin on the contacts subjected to high frequency impact loading in semiconductor device testing 5 crystal structure also influences the adhesive strength. metals with hexagonal close packed structure show much less adhesion than other crystal structures (sikorski, 1963). this is because hexagonal closed pack metals have far fewer slip systems and are therefore less ductile than face-centered and body centered metals. high hardness, large elastic moduli and surface energy of metals also suppress adhesion too (sikorski, 1963). apart from that, adhesive strength is also a function of the chemical reactivity of metals. chemical reactivity is often defined as values in electropositivity. chemically active metals such as aluminium bond more readily and there show stronger adhesion than noble metals. summarizing from the study conducted the material properties of practical importance in contributing to the effect of adhesive wear are divided into mechanical properties and chemical properties. the mechanical properties are • crystal structure • hardness • elastic modulus • surface energy meanwhile, the chemical properties are: • electron density • chemical reactivity the severity of wear can be determined by how much volume of material is rubbed off or carried off from the origin material. there is a mathematical relationship described by archard equation as a means to determine wear volume as shown in equation 1. (harris, 2002) 4 crystal structure also influences the adhesive strength. metals with hexagonal close packed structure show much less adhesion than other crystal structures (sikorski, 1963). this is because hexagonal closed pack metals have far fewer slip systems and are therefore less ductile than facecentered and body centered metals. high hardness, large elastic moduli and surface energy of metals also suppress adhesion too (sikorski, 1963). apart from that, adhesive strength is also a function of the chemical reactivity of metals. chemical reactivity is often defined as values in electropositivity. chemically active metals such as aluminium bond more readily and there show stronger adhesion than noble metals. summarizing from the study conducted the material properties of practical importance in contributing to the effect of adhesive wear are divided into mechanical properties and chemical properties. the mechanical properties are • crystal structure • hardness • elastic modulus • surface energy meanwhile, the chemical properties are: • electron density • chemical reactivity the severity of wear can be determined by how much volume of material is rubbed off or carried off from the origin material. there is a mathematical relationship described by archard equation as a means to determine wear volume as shown in equation 1. (harris, 2002) h kwl q = (1) 3.0 discussion using the technical computing software, wolfram mathematica documentation 7, the following table of material properties is constructed. the properties of the material are arranged in the order of from the lowest electrical resistivity. 4 crystal structure also influences the adhesive strength. metals with hexagonal close packed structure show much less adhesion than other crystal structures (sikorski, 1963). this is because hexagonal closed pack metals have far fewer slip systems and are therefore less ductile than facecentered and body centered metals. high hardness, large elastic moduli and surface energy of metals also suppress adhesion too (sikorski, 1963). apart from that, adhesive strength is also a function of the chemical reactivity of metals. chemical reactivity is often defined as values in electropositivity. chemically active metals such as aluminium bond more readily and there show stronger adhesion than noble metals. summarizing from the study conducted the material properties of practical importance in contributing to the effect of adhesive wear are divided into mechanical properties and chemical properties. the mechanical properties are • crystal structure • hardness • elastic modulus • surface energy meanwhile, the chemical properties are: • electron density • chemical reactivity the severity of wear can be determined by how much volume of material is rubbed off or carried off from the origin material. there is a mathematical relationship described by archard equation as a means to determine wear volume as shown in equation 1. (harris, 2002) h kwl q = (1) 3.0 discussion using the technical computing software, wolfram mathematica documentation 7, the following table of material properties is constructed. the properties of the material are arranged in the order of from the lowest electrical resistivity. 3.0 discussion using the technical computing software, wolfram mathematica documentation 7, the following table of material properties is constructed. the properties of the material are arranged in the order of from the lowest electrical resistivity. issn: 2180-1053 vol. 3 no. 1 january-june 2011 journal of mechanical engineering and technology 6 table 1: material properties 5 table 1 material properties element group ρ (mω) e (gpa) g (gpa) hb (mpa) mt (k) x crystal structure ag 11 1.60 810−× 83 30 24.5 1234.93 1.93 fcc cu 11 1.70 810−× 130 48 874 1357.77 1.9 fcc au 11 2.20 810−× 78 27 2450 1337.33 2.54 fcc al 13 2.60 810−× 70 26 245 933.47 1.61 fcc ca 2 3.40 810−× 20 7.4 167 1115.00 1 fcc be 2 4.00 810−× 287 132 600 1560.00 1.57 hexagonal rh 9 4.30 810−× 275 150 1100 2237.00 2.28 fcc mg 2 4.40 810−× 45 17 260 923.00 1.31 hexagonal ir 9 4.70 810−× 528 210 1670 2739.00 2.2 fcc na 1 4.70 810−× 10 3.3 0.69 370.87 0.93 bcc mo 6 5.00 810−× 329 20 1500 2896.00 2.16 bcc w 6 5.00 810−× 411 161 2570 3695.00 2.36 bcc since, the hardness is of central importance of a contact, gold, iridium and mobylenum may well be the alternative substitutes for copper. as seen from the study above, the principal reason for wear is a result of the softness of tin plating itself. however, the tin plating is already a standard plating material for semiconductor device lead. hence, the only thing can be altered is the contact itself, be it by changing the material, changing the surface treatment or using lubricants. instead of changing the building material or surface treatment of the material, it is easier if not better to try lubrication. lubrication can be a good solution as it is known that lubricants reduce wear rate. however, it is of engineering interest that whether lubrication would reduce the overall electrical conductivity of the contact. since lubricant constitute a thin layer of electric contact itself, one important aspect to consider is whether such thin layer of lubricant affect the overall electrical resistance of the contact as the electrical resistance is of primary importance for the system itself. since resistance elements are in series of each other, the total resistance of the system before lubricant is added would be cubeausnalloy42 rrrrrt +++= (2) since, the hardness is of central importance of a contact, gold, iridium and mobylenum may well be the alternative substitutes for copper. as seen from the study above, the principal reason for wear is a result of the softness of tin plating itself. however, the tin plating is already a standard plating material for semiconductor device lead. hence, the only thing can be altered is the contact itself, be it by changing the material, changing the surface treatment or using lubricants. instead of changing the building material or surface treatment of the material, it is easier if not better to try lubrication. lubrication can be a good solution as it is known that lubricants reduce wear rate. however, it is of engineering interest that whether lubrication would reduce the overall electrical conductivity of the contact. since lubricant constitute a thin layer of electric contact itself, one important aspect to consider is whether such thin layer of lubricant affect the overall electrical resistance of the contact as the electrical resistance is of primary importance for the system itself. since resistance elements are in series of each other, the total resistance of the system before lubricant is added would be 5 table 1 material properties element group ρ (mω) e (gpa) g (gpa) hb (mpa) mt (k) x crystal structure ag 11 1.60 810−× 83 30 24.5 1234.93 1.93 fcc cu 11 1.70 810−× 130 48 874 1357.77 1.9 fcc au 11 2.20 810−× 78 27 2450 1337.33 2.54 fcc al 13 2.60 810−× 70 26 245 933.47 1.61 fcc ca 2 3.40 810−× 20 7.4 167 1115.00 1 fcc be 2 4.00 810−× 287 132 600 1560.00 1.57 hexagonal rh 9 4.30 810−× 275 150 1100 2237.00 2.28 fcc mg 2 4.40 810−× 45 17 260 923.00 1.31 hexagonal ir 9 4.70 810−× 528 210 1670 2739.00 2.2 fcc na 1 4.70 810−× 10 3.3 0.69 370.87 0.93 bcc mo 6 5.00 810−× 329 20 1500 2896.00 2.16 bcc w 6 5.00 810−× 411 161 2570 3695.00 2.36 bcc since, the hardness is of central importance of a contact, gold, iridium and mobylenum may well be the alternative substitutes for copper. as seen from the study above, the principal reason for wear is a result of the softness of tin plating itself. however, the tin plating is already a standard plating material for semiconductor device lead. hence, the only thing can be altered is the contact itself, be it by changing the material, changing the surface treatment or using lubricants. instead of changing the building material or surface treatment of the material, it is easier if not better to try lubrication. lubrication can be a good solution as it is known that lubricants reduce wear rate. however, it is of engineering interest that whether lubrication would reduce the overall electrical conductivity of the contact. since lubricant constitute a thin layer of electric contact itself, one important aspect to consider is whether such thin layer of lubricant affect the overall electrical resistance of the contact as the electrical resistance is of primary importance for the system itself. since resistance elements are in series of each other, the total resistance of the system before lubricant is added would be cubeausnalloy42 rrrrrt +++= (2) 5 table 1 material properties element group ρ (mω) e (gpa) g (gpa) hb (mpa) mt (k) x crystal structure ag 11 1.60 810−× 83 30 24.5 1234.93 1.93 fcc cu 11 1.70 810−× 130 48 874 1357.77 1.9 fcc au 11 2.20 810−× 78 27 2450 1337.33 2.54 fcc al 13 2.60 810−× 70 26 245 933.47 1.61 fcc ca 2 3.40 810−× 20 7.4 167 1115.00 1 fcc be 2 4.00 810−× 287 132 600 1560.00 1.57 hexagonal rh 9 4.30 810−× 275 150 1100 2237.00 2.28 fcc mg 2 4.40 810−× 45 17 260 923.00 1.31 hexagonal ir 9 4.70 810−× 528 210 1670 2739.00 2.2 fcc na 1 4.70 810−× 10 3.3 0.69 370.87 0.93 bcc mo 6 5.00 810−× 329 20 1500 2896.00 2.16 bcc w 6 5.00 810−× 411 161 2570 3695.00 2.36 bcc since, the hardness is of central importance of a contact, gold, iridium and mobylenum may well be the alternative substitutes for copper. as seen from the study above, the principal reason for wear is a result of the softness of tin plating itself. however, the tin plating is already a standard plating material for semiconductor device lead. hence, the only thing can be altered is the contact itself, be it by changing the material, changing the surface treatment or using lubricants. instead of changing the building material or surface treatment of the material, it is easier if not better to try lubrication. lubrication can be a good solution as it is known that lubricants reduce wear rate. however, it is of engineering interest that whether lubrication would reduce the overall electrical conductivity of the contact. since lubricant constitute a thin layer of electric contact itself, one important aspect to consider is whether such thin layer of lubricant affect the overall electrical resistance of the contact as the electrical resistance is of primary importance for the system itself. since resistance elements are in series of each other, the total resistance of the system before lubricant is added would be cubeausnalloy42 rrrrrt +++= (2) issn: 2180-1053 vol. 3 no. 1 january-june 2011 a study on the deposition of tin on the contacts subjected to high frequency impact loading in semiconductor device testing 7 6 figure 2 resistance model of contact materials with an addition of lubricant when a layer of lubricant is added as shown in figure 2, the total resistance would change to cubeaulubricantsnalloy42 rrrrrrt ++++= ′ (3) according to bueche (1995), resistance is mathematically defined as a l r ρ= (4) since the thickness of the lubricant, l in common sense would be very much smaller compared to the thicknesses of contact and lead, it is assumed that l is close to zero, i.e. mm0lubricant ≅l (5) when l is of negligible length, the resistance of the lubricant would be ( ) 0 mm0lubricant lubricant ≅ ≅ = a l r ρ (6) hence ( ) tt rrrrrrrrrrr ≅+++≅++≅++= ′ cubeausnalloy42cubeaulubricantsnalloy42 0 (7) tt rr ≅ ′∴ (8) the overall resistance of the system is not significantly altered. alloy 42 copper beryllium tin plating gold plating lubricant lubricantr snr alloy42r aur cuber figure 2: resistance model of contact materials with an addition of lubricant when a layer of lubricant is added as shown in figure 2, the total resistance would change to 6 figure 2 resistance model of contact materials with an addition of lubricant when a layer of lubricant is added as shown in figure 2, the total resistance would change to cubeaulubricantsnalloy42 rrrrrrt ++++= ′ (3) according to bueche (1995), resistance is mathematically defined as a l r ρ= (4) since the thickness of the lubricant, l in common sense would be very much smaller compared to the thicknesses of contact and lead, it is assumed that l is close to zero, i.e. mm0lubricant ≅l (5) when l is of negligible length, the resistance of the lubricant would be ( ) 0 mm0lubricant lubricant ≅ ≅ = a l r ρ (6) hence ( ) tt rrrrrrrrrrr ≅+++≅++≅++= ′ cubeausnalloy42cubeaulubricantsnalloy42 0 (7) tt rr ≅ ′∴ (8) the overall resistance of the system is not significantly altered. alloy 42 copper beryllium tin plating gold plating lubricant lubricantr snr alloy42r aur cuber 6 figure 2 resistance model of contact materials with an addition of lubricant when a layer of lubricant is added as shown in figure 2, the total resistance would change to cubeaulubricantsnalloy42 rrrrrrt ++++= ′ (3) according to bueche (1995), resistance is mathematically defined as a l r ρ= (4) since the thickness of the lubricant, l in common sense would be very much smaller compared to the thicknesses of contact and lead, it is assumed that l is close to zero, i.e. mm0lubricant ≅l (5) when l is of negligible length, the resistance of the lubricant would be ( ) 0 mm0lubricant lubricant ≅ ≅ = a l r ρ (6) hence ( ) tt rrrrrrrrrrr ≅+++≅++≅++= ′ cubeausnalloy42cubeaulubricantsnalloy42 0 (7) tt rr ≅ ′∴ (8) the overall resistance of the system is not significantly altered. alloy 42 copper beryllium tin plating gold plating lubricant lubricantr snr alloy42r aur cuber according to bueche (1995), resistance is mathematically defined as 6 figure 2 resistance model of contact materials with an addition of lubricant when a layer of lubricant is added as shown in figure 2, the total resistance would change to cubeaulubricantsnalloy42 rrrrrrt ++++= ′ (3) according to bueche (1995), resistance is mathematically defined as a l r ρ= (4) since the thickness of the lubricant, l in common sense would be very much smaller compared to the thicknesses of contact and lead, it is assumed that l is close to zero, i.e. mm0lubricant ≅l (5) when l is of negligible length, the resistance of the lubricant would be ( ) 0 mm0lubricant lubricant ≅ ≅ = a l r ρ (6) hence ( ) tt rrrrrrrrrrr ≅+++≅++≅++= ′ cubeausnalloy42cubeaulubricantsnalloy42 0 (7) tt rr ≅ ′∴ (8) the overall resistance of the system is not significantly altered. alloy 42 copper beryllium tin plating gold plating lubricant lubricantr snr alloy42r aur cuber 6 figure 2 resistance model of contact materials with an addition of lubricant when a layer of lubricant is added as shown in figure 2, the total resistance would change to cubeaulubricantsnalloy42 rrrrrrt ++++= ′ (3) according to bueche (1995), resistance is mathematically defined as a l r ρ= (4) since the thickness of the lubricant, l in common sense would be very much smaller compared to the thicknesses of contact and lead, it is assumed that l is close to zero, i.e. mm0lubricant ≅l (5) when l is of negligible length, the resistance of the lubricant would be ( ) 0 mm0lubricant lubricant ≅ ≅ = a l r ρ (6) hence ( ) tt rrrrrrrrrrr ≅+++≅++≅++= ′ cubeausnalloy42cubeaulubricantsnalloy42 0 (7) tt rr ≅ ′∴ (8) the overall resistance of the system is not significantly altered. alloy 42 copper beryllium tin plating gold plating lubricant lubricantr snr alloy42r aur cuber since the thickness of the lubricant, l in common sense would be very much smaller compared to the thicknesses of contact and lead, it is assumed that l is close to zero, i.e. 6 figure 2 resistance model of contact materials with an addition of lubricant when a layer of lubricant is added as shown in figure 2, the total resistance would change to cubeaulubricantsnalloy42 rrrrrrt ++++= ′ (3) according to bueche (1995), resistance is mathematically defined as a l r ρ= (4) since the thickness of the lubricant, l in common sense would be very much smaller compared to the thicknesses of contact and lead, it is assumed that l is close to zero, i.e. mm0lubricant ≅l (5) when l is of negligible length, the resistance of the lubricant would be ( ) 0 mm0lubricant lubricant ≅ ≅ = a l r ρ (6) hence ( ) tt rrrrrrrrrrr ≅+++≅++≅++= ′ cubeausnalloy42cubeaulubricantsnalloy42 0 (7) tt rr ≅ ′∴ (8) the overall resistance of the system is not significantly altered. alloy 42 copper beryllium tin plating gold plating lubricant lubricantr snr alloy42r aur cuber 6 figure 2 resistance model of contact materials with an addition of lubricant when a layer of lubricant is added as shown in figure 2, the total resistance would change to cubeaulubricantsnalloy42 rrrrrrt ++++= ′ (3) according to bueche (1995), resistance is mathematically defined as a l r ρ= (4) since the thickness of the lubricant, l in common sense would be very much smaller compared to the thicknesses of contact and lead, it is assumed that l is close to zero, i.e. mm0lubricant ≅l (5) when l is of negligible length, the resistance of the lubricant would be ( ) 0 mm0lubricant lubricant ≅ ≅ = a l r ρ (6) hence ( ) tt rrrrrrrrrrr ≅+++≅++≅++= ′ cubeausnalloy42cubeaulubricantsnalloy42 0 (7) tt rr ≅ ′∴ (8) the overall resistance of the system is not significantly altered. alloy 42 copper beryllium tin plating gold plating lubricant lubricantr snr alloy42r aur cuber when l is of negligible length, the resistance of the lubricant would be 6 figure 2 resistance model of contact materials with an addition of lubricant when a layer of lubricant is added as shown in figure 2, the total resistance would change to cubeaulubricantsnalloy42 rrrrrrt ++++= ′ (3) according to bueche (1995), resistance is mathematically defined as a l r ρ= (4) since the thickness of the lubricant, l in common sense would be very much smaller compared to the thicknesses of contact and lead, it is assumed that l is close to zero, i.e. mm0lubricant ≅l (5) when l is of negligible length, the resistance of the lubricant would be ( ) 0 mm0lubricant lubricant ≅ ≅ = a l r ρ (6) hence ( ) tt rrrrrrrrrrr ≅+++≅++≅++= ′ cubeausnalloy42cubeaulubricantsnalloy42 0 (7) tt rr ≅ ′∴ (8) the overall resistance of the system is not significantly altered. alloy 42 copper beryllium tin plating gold plating lubricant lubricantr snr alloy42r aur cuber 6 figure 2 resistance model of contact materials with an addition of lubricant when a layer of lubricant is added as shown in figure 2, the total resistance would change to cubeaulubricantsnalloy42 rrrrrrt ++++= ′ (3) according to bueche (1995), resistance is mathematically defined as a l r ρ= (4) since the thickness of the lubricant, l in common sense would be very much smaller compared to the thicknesses of contact and lead, it is assumed that l is close to zero, i.e. mm0lubricant ≅l (5) when l is of negligible length, the resistance of the lubricant would be ( ) 0 mm0lubricant lubricant ≅ ≅ = a l r ρ (6) hence ( ) tt rrrrrrrrrrr ≅+++≅++≅++= ′ cubeausnalloy42cubeaulubricantsnalloy42 0 (7) tt rr ≅ ′∴ (8) the overall resistance of the system is not significantly altered. alloy 42 copper beryllium tin plating gold plating lubricant lubricantr snr alloy42r aur cuber hence 6 figure 2 resistance model of contact materials with an addition of lubricant when a layer of lubricant is added as shown in figure 2, the total resistance would change to cubeaulubricantsnalloy42 rrrrrrt ++++= ′ (3) according to bueche (1995), resistance is mathematically defined as a l r ρ= (4) since the thickness of the lubricant, l in common sense would be very much smaller compared to the thicknesses of contact and lead, it is assumed that l is close to zero, i.e. mm0lubricant ≅l (5) when l is of negligible length, the resistance of the lubricant would be ( ) 0 mm0lubricant lubricant ≅ ≅ = a l r ρ (6) hence ( ) tt rrrrrrrrrrr ≅+++≅++≅++= ′ cubeausnalloy42cubeaulubricantsnalloy42 0 (7) tt rr ≅ ′∴ (8) the overall resistance of the system is not significantly altered. alloy 42 copper beryllium tin plating gold plating lubricant lubricantr snr alloy42r aur cuber issn: 2180-1053 vol. 3 no. 1 january-june 2011 journal of mechanical engineering and technology 8 6 figure 2 resistance model of contact materials with an addition of lubricant when a layer of lubricant is added as shown in figure 2, the total resistance would change to cubeaulubricantsnalloy42 rrrrrrt ++++= ′ (3) according to bueche (1995), resistance is mathematically defined as a l r ρ= (4) since the thickness of the lubricant, l in common sense would be very much smaller compared to the thicknesses of contact and lead, it is assumed that l is close to zero, i.e. mm0lubricant ≅l (5) when l is of negligible length, the resistance of the lubricant would be ( ) 0 mm0lubricant lubricant ≅ ≅ = a l r ρ (6) hence ( ) tt rrrrrrrrrrr ≅+++≅++≅++= ′ cubeausnalloy42cubeaulubricantsnalloy42 0 (7) tt rr ≅ ′∴ (8) the overall resistance of the system is not significantly altered. alloy 42 copper beryllium tin plating gold plating lubricant lubricantr snr alloy42r aur cuber 6 figure 2 resistance model of contact materials with an addition of lubricant when a layer of lubricant is added as shown in figure 2, the total resistance would change to cubeaulubricantsnalloy42 rrrrrrt ++++= ′ (3) according to bueche (1995), resistance is mathematically defined as a l r ρ= (4) since the thickness of the lubricant, l in common sense would be very much smaller compared to the thicknesses of contact and lead, it is assumed that l is close to zero, i.e. mm0lubricant ≅l (5) when l is of negligible length, the resistance of the lubricant would be ( ) 0 mm0lubricant lubricant ≅ ≅ = a l r ρ (6) hence ( ) tt rrrrrrrrrrr ≅+++≅++≅++= ′ cubeausnalloy42cubeaulubricantsnalloy42 0 (7) tt rr ≅ ′∴ (8) the overall resistance of the system is not significantly altered. alloy 42 copper beryllium tin plating gold plating lubricant lubricantr snr alloy42r aur cuber the overall resistance of the system is not significantly altered. 4.0 conclusion as seen from the study above, the principal reason for wear is hypnotized as a result of the softness of tin plating itself. however, the tin plating is already a standard plating material for semiconductor device lead. hence, the only thing can be altered is the contact itself, be it by changing the material, changing the surface treatment or using lubricants. since changing the surface treatment and material takes a very long time and not necessarily an economical way, lubricant may be recommended. testing will be conducted in the future to verify the proposed hypothesis. 5.0 symbols q total volume of wear debris produced w total normal load h hardness of the softest contacting surfaces k dimensionless constant l sliding length electrical resistance before lubrication electrical resistance after lubrication electrical resistance of alloy 42 electrical resistance of tin electrical resistance of lubricant electrical resistance of gold electrical resistance of copper beryllium electrical resistance resistivity length/thickness ross-sectional area lubricant thickness 6.0 acknowledgement the foremost gratitude is dedicated to prof. dr. md radzai bin said for guiding and giving duly advice throughout the project. he has been issn: 2180-1053 vol. 3 no. 1 january-june 2011 a study on the deposition of tin on the contacts subjected to high frequency impact loading in semiconductor device testing 9 tolerant and wise mentor. thanks to ismeca, especially mr. sh chong for initiating this collaboration between ismeca and utem. special thanks to mr. masjuri for his discussion which lead to the birth of such collaboration. 7.0 references buckley, d. h. 1981. surface effects in adhesion, friction, wear and lubrication. elsevier. epistola, e. 2006. what is a semiconductor. silicon far east. retrieved from april 9, 2010 from http://www.siliconfareast.com/whatissemicon.htm. harris, c. k., broussard, j. p. and whitcomb, p. j 2002. determination of wear in a tribo-system. proceedings of the 2002 asee gulf-southwestern annual conference. kalpakjian, s. 2006. manufacturing engineering and technology (5th edition). singapore: prentice hall. sikorski,m.e. 1963. correlation of the coefficient of adhesion with various physical and mechanical properties of metals, transactions asme, series d journal of basic engineering, vol. 85, pp. 279-285. stachowiak, g. w. and batchelor, a. w. 2005. engineering tribology (3rd edition). burlington:elsevier butterworth-heinemann. ziman, j. m. 1963 electrons in metals a short guide to the fermi surface. london: taylor and francis. issn: 2180-1053 vol. 10 no.1 january – june 2018 59 greening the existing building (chancellery building university technical malaysia melaka) m. k. hussein1*,r.b. mat dan2 1mechanical engineering department, al-mustansiriayah university, baghdad, iraq. 2faculty of mechanical engineering, universiti teknikal malaysia melaka (utem), melaka, malaysia. abstract buildings contribute significantly to the environmental and economic issues, as they consume a high amount of energy and water. as a building consumes energy, it contributes to emissions of carbon dioxide which lead to environmental pollution. these factors have a negative impact on the environment and the economy among other issues. green building practices and approaches can considerably reduce or eliminate negative ecological and economic impacts. this study aims to “greening the existing building” and achieve the “certified” rating level according to the gbi classification with a low budget, taking into consideration estimated cost. the green building audit results show the total current building rating level is only 18 off 100 points based on the major six criteria that shows the existing chancellery building achieves a low rating level when evaluated according to the gbi rating system. to achieve a “certified rating level” of (50 points) this study proposes short-term improvements and medium-term evaluation durations of existing building’s criteria (retrofitting). the economic analysis involves the estimation of costs included the “greening existing building” and the potential savings acquired from “retrofitting” and it shows the calculation of “payback period and return on investment (roi)”. keywords: green building; energy audit; retrofit; economic analysis, green building index 1.0 introduction “greening buildings” is considered one of the solutions proposed to address global climate changes and economic issues due to unbalanced energy consumption in various types of infrastructure (green building council, 2009) (carroon, 2010). the global annual energy consumption of buildings is a high impact as it accounts for more than 60% of the total electricity consumption and 40% of the aggregate energy consumed. moreover, the use of water in buildings is more than 16% of the aggregate water consumption (zaid et al., 2013). buildings are considered one of the causes of the global warming phenomenon since it accounts for over 40% of total carbon dioxide co2 emissions. while currently, united states, canada, western europe, and japan are the major contributors to greenhouse gas emissions, this situation is going to change radically in the upcoming years. (yudelson, 2007) the “green building index (gbi)” is the recognized “rating tool” for green buildings in malaysia. this encourages sustainability in the built environment, and increase *corresponding author e-mail: mustafakhudhur@uomustansiriyah.edu.iq mailto:mustafakhudhur@uomustansiriyah.edu.iq journal of mechanical engineering and technology 60 issn: 2180-1053 vol. 10 no.1 january – june 2018 awareness of these matters among related stakeholders, including developers, contractors, and architects. the evaluation of residential and commercial properties using the “gbi rating tool”, depends on six main criteria: “indoor environment quality (eq), sustainable site planning & management (sm), materials and resources (mr), energy efficiency (ee), innovation (in), and water efficiency (we)” alternatively, buildings are divided into the following categories: “non-residential new construction (nrnc), non-residential existing building (nreb), residential new construction (rnc), and residential existing building (reb)”. there are four levels of gbi certification: “(more than 86 points) platinum; (76 to 85 points) gold; (66 to 75 points) silver; and (50 to 65 points) certified” (gbi malaysia, 2011). this study aims to “greening” the nonresidential existing building (chancellery building university technical malaysia melaka) and achieves the “certified” rating level of (gbi) classification. also, estimate the cost of greening procedure of the building and to identify the assessment criteria that most effect on the overall cost and determine the potential overall savings from retrofitting. 2.0 methodology the methodology which is implemented in this study consists of three phases; the first phase is the description of the building. the second phase is the energy audit procedure that is used to assess current building rating level according to (gbi). energy audit includes the data collection, walk-through tour, check list according to (gbi), calculation of building criteria, physical measurement, and the energy audit tools. the last phase is a proposed potential improvement of building criteria (retrofit). 2.1 description of the building the case study building for this study is the “chancellery building” the building consists of three blocks: chancellery, plaza, and hepa. figure 1. location of chancellery building [source: utem-development office] a greening the existing building (chancellery building university technical malaysia melaka) issn: 2180-1053 vol. 10 no.1 january – june 2018 61 the building located in the main campus of the “university technical malaysia melaka” “hang tuah jaya, 76100 durian tunggal, melaka, malaysia”, and the total building area is about 11212 m2, figure 1 below show the location of chancellery building. 2.2 energy audit procedure energy audit includes the data collection, walk-through tour, check list according to (gbi), calculation of building criteria, and physical measurement. (natural resources canada, 2009) (fallis, 2013) 2.2.1. data collection data collection is the first step of energy audit, the data collected in this study includes electricity bill, water bill, building layout (civil layout, architecture layout, electrical layout and mechanical layout), and the material used in the construction of the building. it also includes data collected by the interview with the many engineers in the utemdevelopment office. 2.2.2. walk-through tour the second step is walk-through tour, the main objective of the tour is to obtain general information about the case study building, including building envelope, acmv system, lighting system, and operation & maintenance practices. that seeks to identify potential retrofit opportunities which can be improved the building to achieve study target. 2.2.3. check list according to (gbi) the third step is fill up the gbi checklist for non-residential existing building. the checklist filled up by conducting an interview with the many engineers in the utemdevelopment office, analysis of the collected data, walk-through tour, calculation of some parameters and the physical measurement for many parameters. 2.2.4. calculation of building criteria the calculation of building criteria includes the “overall thermal transfer value (ottv)” and “building energy intensity (bei)”. the (ottv) was calculated through the building involve (walls and windows) and is given by the equation (1) as shown below: ottv = ao1∗ottv1+ao2∗ottv2+aon∗ottvn ao1+ao2+aon (1) where a1 is the total exterior wall area for orientation one and ottv1 is the “ottv” value for orientation one. ottv of any given wall orientation is given by the equation (2) as shown below: ottvi = 15 (1 − wwr)uw + 6(wwr)uf + (194 ∗ cf ∗ wwr ∗ sc) (2) journal of mechanical engineering and technology 62 issn: 2180-1053 vol. 10 no.1 january – june 2018 where wwr is the window-to-total exterior wall area ratio for the orientation; α the “solar absorptivity” of opaque wall; uw is the “thermal transmittance” of opaque wall (w/m 2 k); uf is the “thermal transmittance” of windows (w/m 2 k); sc is the “shading coefficient” of the fenestration system; cf is the “solar correction factor” for building orientation. (sukri, m. f., 2012) as well as the (bei) was calculated by performing analysis of the electrical consumption and building floor area. the (bei) given by the equation (3) as shown below (noranai and kammalluden, 2012): bei = total energy consumed in one year (kwh/yr) total floor space area of the building (m2) (3) 2.2.5 physical measurement the physical measurement of indoor environmental quality (ieq) was conducted. which includes indoor air quality (iaq), mold prevention, daylighting, daylight glare control, electric lighting levels, and internal noise levels. the indoor air quality (iaq) and mold prevention include the flowing parameter: average air velocity (m/s), average fresh air flow (cfm), average operating temperature (°c), average relative humidity (%), and average co2 (ppm). therefore, gathering field data by doing measurements of physical parameters in order to evaluate indoor air quality for each floor in the chancellery building. the selection of measurement points was conducted in a random manner in such a way that it covered all areas. the number of points is (5-8) points in different places on the same floor. the time spent to conduct the measurement (8:30 am to 5 pm). every point is an average of (60) points in the same zone, and the height of the devices is 1.5 m. the role of this measurement is to characterize and compare iaq results with gbi requirements. where the gbi requirements are compare the results with the minimum requirements of ventilation rate in “ashrae 62.1:2007” (ashrae standard, 2007) and/or malaysian’s standards such as “ms 1525: 2007” (malaysia, 2007) and “malaysian industry code of practice on indoor air quality 2010” (dosh, 2010). while the measurement of average fresh air flow (cfm), was conducted in all ahu by measuring the air velocity of fresh air and measure the dimension of the fresh air duct. (tuan et al., 2015) the daylighting measurement is conducted by measure the indoor illumination (li), the (li) were taken at middle points of north, south, east and west rooms by using lux meter probes located 0.8 m above floor level. readings were logged every one hour from (10:00 am to 3:00 pm). while the outdoor illumination (lo) was taken from privies study was conducted in malaysia (fadzil et al., 2015). and then compare the result with gbi requirements. heat conduction through windows heat conduction through walls solar heat gain through windows ottv= + + a greening the existing building (chancellery building university technical malaysia melaka) issn: 2180-1053 vol. 10 no.1 january – june 2018 63 the daylight glare control is conducted by measure the indoor illumination using lux meter probes located 0.8 m above floor level during periods of low angle sun (early mornings and late afternoons) and during periods with a bright sky. the measurement was conducted from (10:00 am to 3:00 pm), in two cases. the first case without using the manual blinds and the second case by using the manual blinds, and then compare the result with gbi requirements. the electric lighting levels measurement was conducted by dividing the work area into a small area and then take the average electric lighting illumination levels. after that compare with gbi requirements. the measurement conducted by using lux meter probes located 0.8 m above floor level. the measurement of the internal noise levels was conducted by using the sound level meter, by measure the (laeq) (average sound level, equivalent continuous sound level). the number of points is (5-8) points in different places on the same floor and the time duration of conducting the measurement too from (8:30 am to 5 pm). the measurement involves the offices, lobby and the area near to the ahu. the measurement devices that were used in this study are classified as follows:  tsi 7545 indoor air quality meter  rs avm-0.1 anemometer  center 337 light meter  rion na-28 sound level meter 2.3 improvement of building criteria (retrofit) after conducting the energy audit and obtaining the current gbi rating level. the final step is greening the existing building and achieve a “certified rating level” of (50 points) according to gbi rating system, this study proposes a short-term improvements duration (4 years) of existing building’s criteria (retrofitting) and medium-term duration (8 years) to evaluate the result of implementation of the improvement to determine their effectiveness in moving towards the greening the existing building. the potential improvement includes six-part: “indoor environment quality (eq), energy efficiency (ee), sustainable site planning & management (sm), water efficiency (we) materials & resources (mr) and innovation (in)”. below some of the potential improvement: encourage sustainable maintenance, upgrades the traditional energy lighting system, installation rainwater harvesting equipment, provide more flexible light control system like daylight sensor and motion sensor, promote innovation and environmental initiative, provide more greenery and roof to reduce heat island effects, provide co2 sensor to monitor indoor air quality, installation renewable energy like photovoltaic (pv), install self-cleaning façade and use more green product/material in building. (zakaria et al., 2012). the economic analysis involves the estimation of costs included in “greening existing building” and the potential savings acquired from “retrofitting”. the potential savings include cutting costs from lighting system and building integrated photo voltage. journal of mechanical engineering and technology 64 issn: 2180-1053 vol. 10 no.1 january – june 2018 3. results and discussions in this study, the “energy audit” was conducted to assess the building assessment criteria based on the major six criteria. energy audit includes the data collection, walk-through tour, check list according to (gbi), calculation of building criteria, and physical measurement. 3.1 results of energy efficiency (ee) the energy audit results show the current building rating level for (ee) is (9 from 38) points. as shown in table 1 below table 1: results of energy efficiency (ee) 3.1.1 overall thermal transfer value the “overall thermal transfer value (ottv)” is an amount of the average rate of heat gain into a building through the building envelope and includes walls and windows for all sides (sukri, m. f., 2012). the results show the average ottv for chancellery block, plaza block, and hepa block is (64.23, 86.88 and 70.19) w/m2 respectively. according to the gbi requirement, all blocks do not achieve the recommendation of ms 1525: 2007 (ottv≤50 w/m2) (malaysia, 2007), the building got (0 from 1 point) according to green building index (gbi/nreb). 3.1.2. advanced or improved ee performance – bei building energy intensity is the amount of electric energy consumed per year per meter square, where electric energy in the kwh unit and the unit of building energy intensity is kwh/m²/yr.(noranai and kammalluden, 2012) in this study the building consists of three blocks: chancellery, plaza, and hepa, the total floor space area of the building are (11212 m2). the analysis of electrical consumption was based on a monthly summary report for one year (january 2015 – december 2015). the report prepared by the “development office in utem” by used electrical meter, while the electrical bill involves assessment points score assessment points score minimum energy efficiency performance 2 0 enhanced commissioning of building energy systems 4 4 lighting zoning 3 0 on-going post occupancy commissioning 2 0 electrical submetering 2 0 energy efficiency improvement and monitoring 2 0 renewable energy 5 0 sustainable maintenance 3 0 improved energy efficiency performance –“ bei” 15 5 total score = 9 a greening the existing building (chancellery building university technical malaysia melaka) issn: 2180-1053 vol. 10 no.1 january – june 2018 65 the whole university. figure 2 shows a summary of the estimated electricity for chancellery block, hepa block, and plaza block as shown below: figure 2. electricity consumption from (january 2015) to (december 2015) the results show the building energy intensity (bei) is 127.55 kwh/m2/yr. and when to compare the result with the recommendations of the (gbi) building’s assessment criteria the value of bei≤130. from that the building assessment criteria for this part got (5 points from 15). 3.1.3 enhanced commissioning this part includes develop a commissioning plan for the building’s “energy-related systems”, update the building “operating plan”, and provide “training” for management staff to build. the energy audit results the building’s management office (development office) is already implemented the gbi requirements. from that the building assessment criteria for this part (4 from 4) points. 3.2 results of indoor environment quality (eq) the energy audit results show the current building rating level for (eq) is (5 from 21) points. as shown in table 2 below: table 2. results of indoor environment quality (eq) assessment points score assessment points score minimum indoor air quality “iaq” performance 1 1 daylighting factor (df) 2 1 environmental tobacco smoke control 1 1 daylight glare control 1 1 carbon dioxide (co2) monitoring and control 1 0 electric lighting levels (lux) 1 0 indoor air pollutants 2 0 high frequency ballasts 1 0 journal of mechanical engineering and technology 66 issn: 2180-1053 vol. 10 no.1 january – june 2018 3.2.1 indoor air quality (iaq) the aim of this part is to analyze data collected from the field measurements to determine and characterize the real physical indoor air quality (iaq) of the building. figure 3 shows the fluctuation of the average air velocity (m/s), average operating temperature (°c), average relative humidity (%) and average co2 (ppm) at the whole building (chancellery block, plaza block, and hepa block). (a) indoor air quality average air velocity mold prevention 1 1 external views 2 0 controllability of systems 2 0 internal noise levels 1 0 air change effectiveness (ace) 1 0 occupancy comfort survey 2 0 iaq test during occupancy 2 0 total score = 5 a greening the existing building (chancellery building university technical malaysia melaka) issn: 2180-1053 vol. 10 no.1 january – june 2018 67 (b) indoor air quality average operating temperature (c) indoor air quality average relative humidity (d) indoor air quality average co2 figure 3. average iaq parameters at the whole building journal of mechanical engineering and technology 68 issn: 2180-1053 vol. 10 no.1 january – june 2018 the measurement of average fresh air flow (cfm), was conducted in all ahu of building by measure the air velocity of fresh air and measure the dimension of the fresh air duct. figure 4 below shows the average fresh air flow (cfm) for each building blocks, occupancy number and people outdoor air rate (rp) cfm/person. (a) average fresh air flow (b) occupancy number (c) people outdoor air rate figure 4. average fresh air flow (cfm) for each building blocks, occupancy number and people outdoor air rate (rp) cfm/person a greening the existing building (chancellery building university technical malaysia melaka) issn: 2180-1053 vol. 10 no.1 january – june 2018 69 according to the gbi requirement the results of iaq compared with the flowing standards:  malaysian “standard ms 1525: 2007” the “indoor air quality” original design should be as follows: recommended design “dry bulb temperature” 23 º c to 26 °c, minimum “dry bulb temperature” 22 ° c, “relative humidity” (rh) 55 % to 70 %, “air velocity” 0.15 m/s to 0.50 m/s and maximum “air velocity” 0.7 m/s.  “ashrae standard 62.1: 2007” the purpose of the standard has to specify “minimum ventilation rates” and “indoor air quality (iaq)” that will be suitable to building occupants, the recommended parameters should be as flows: people outdoor fresh air flow rate (rp) 5 cfm and carbon dioxide (co2) 500-700 ppm.  malaysian “industry code of practice on indoor air quality 2010” the purpose of this “code” is to improving the “indoor air quality (iaq)” in the building and to set minimum value of “iaq”. the physical parameters are given as follows: “air temperature” 23 º c to 26 °c, “relative humidity” (rh) 40 % to 70 %, “air velocity” 0.15 m/s to 0.50 m/s and “carbon dioxide” co2 1000 ppm. the results show the iaq parameters, within the recommended standards. the building got (1 from 1 point) according to green building index (gbi/nreb). 3.2.2 mold prevention the (gbi) has recommended, the “mechanical ventilation system” must maintain indoor air “relative humidity” below 70% rh without using active regulator system. the results show that the average “overall relative humidity” (%) is 65.5, from that can be seen the mold prevention’s requirements, within the recommendations of gbi. from that the building got (1 from 1 point) according to green building index (gbi/nreb). 3.2.3 daylighting factor the daylighting measurement is conducted by measure the indoor illumination (li) and the outdoor illumination (lo). figure 5 shows the average value of li (lux) for the (n, s, e, and w) orientations, and the daylight factor for each orientation. however, the total area of rooms has daylight about (4117.33 m2), the percentage area of rooms has daylight to the total area about (36.72 %). the cbi recommended the area of rooms ≥ 30 % or 50 % of the total area, has daylight factor in the range of (1 % 3.5 %). from that the building got (1 from 2 points) according to green building index (gbi/nreb). journal of mechanical engineering and technology 70 issn: 2180-1053 vol. 10 no.1 january – june 2018 (a) average value of li (lux) (b) daylight factor (d.f) figure 5. average value of li (lux) and daylight factor (d.f) for each orientation 3.2.4 daylighting glare control the measurement was conducted from (10:00 am to 3:00 pm), in two cases. the first case without using the manual blinds and the second case by using the manual blinds. figure 6 below, shows the fluctuation of the indoor illumination (lux) during the time change and (with & without) using (manual blinds). it can be seen that the indoor illumination (lux) with using manual blinds smaller than without using manual blinds. as well as the maximum indoor illumination (lux) at 3:00 pm in both cases. the results show the average luminance level is (1718.33lux) below (2000 lux). from that the building got (1 from 1 point) according to green building index (gbi/nreb). a greening the existing building (chancellery building university technical malaysia melaka) issn: 2180-1053 vol. 10 no.1 january – june 2018 71 figure 6. results of daylight glare control figure 7. average electric lighting illumination levels 3.2.5 electric lighting levels the electric lighting levels measurement was conducted by divided the work area into a small area and then take the average electric lighting illumination levels. figure 7 shows the fluctuation of the average electric lighting illumination levels at the whole building (chancellery block, plaza block, and hepa block). the (gbi) has recommended that original design of electrical lighting system must maintain a “luminance level” not in excess of specified in ms1525 for 90% of the net lettable area (nla), the recommended average illuminance levels for the general offices is (300-400) lux. when to compare the average overall value (518.36) lux with recommended value, the average overall value > recommended value. from that, the building got (0 from 1 point) according to green building index (gbi/nreb). journal of mechanical engineering and technology 72 issn: 2180-1053 vol. 10 no.1 january – june 2018 3.2.6 internal noise levels the measurement of the internal noise levels was conducted, by measure the (laeq) (db) (average sound level, equivalent continuous sound level). figure 8 shows the fluctuation of the average (laeq) (db) (average sound level, equivalent continuous sound level) at the whole building (chancellery block, plaza block, and hepa block). figure 8. average internal noise levels the gbi recommended, within the entire building general office, space noise does not exceed 40 dba. the results show the overall average internal noise levels > the recommendation. from that, the building got (0 from 1 point) according to green building index (gbi/nreb). 3.3 results of sustainable site planning & management (sm) the energy audit results show the current building rating level for (sm) is (1 from 10) points. as shown in table 3 below: table 3. results of sustainable site planning & management (sm) assessment points score assessment points score “gbi” rated design and construction 1 0 parking capacity 1 1 building exterior management 1 0 roof and greenery 2 0 landscape management, integrated pest management, and erosion control 1 0 building user manual 1 0 green vehicle priority 1 0 total score = 1 a greening the existing building (chancellery building university technical malaysia melaka) issn: 2180-1053 vol. 10 no.1 january – june 2018 73 3.3.1 parking capacity parking capacity, the (gbi) has recommended that size of parking capacity not more than the minimum local zoning requirements. the energy audit shows the original design of the parking meets the requirements. from that the building assessment criteria for this part (1 from 1) point. 3.4 results of materials & resources (mr) the energy audit results show the current building rating level for (sm) is (3 from 9) points. as shown in table 4 below: table 4. results of materials & resources (mr) 3.4.1 waste management waste management includes the collection & disposal of recyclables. the “gbi” has recommended facilitating minimizing the waste generated within retrofitting construction and the occupancy of the building that is hauled and disposed of in landfills. also, “gbi” recommended providing recycling facilities/infrastructure for separating and sorting a recyclable waste collection of for recycling, such as paper, glass consumables, equipment, and metal. the energy audit show building’s management office is already meet the gbi requirements. from that the building assessment criteria for this part (3 from 3) point. 3.5 results of water efficiency (we) the energy audit results show the current building rating level for (we) is (0 from 12) points. as shown in table 5 below: table 5. results of water efficiency (we) assessment points score assessment points score selection and materials reuse 1 0 disposal of recyclables, storage, and collection 3 3 recycled content materials 1 0 clean agents and refrigerants 2 0 sustainable timber 1 0 sustainable purchasing policy 1 0 total score = 3 assessment points score assessment points score rainwater harvesting 3 0 water efficient fittings 3 0 water recycling 2 0 metering & leak detection system 2 0 water efficient irrigation/landscaping 2 0 total score = 0 journal of mechanical engineering and technology 74 issn: 2180-1053 vol. 10 no.1 january – june 2018 3.5.1. water consumption analysis the compliance audit results show the water system in the building served the cooling tower, toilet, kitchen, and irrigation. the utem / main campus was provided one water meter for all building. the water consumption for chancellery building has been estimated according to the technical information was collected from the utem / development office, as shown below : the number of occupants = 1820, total building floor area = 11212 m2 water consumption = 8 l / day / m2 water consumption = 89,696 l / day = 1,973,312 l / month = 23,679,744 l / year water consumption = 23679 m3 / year 3.6 results of innovation (in) the energy audit results show the current building rating level for (in) is (0 from 10) points. as shown in table 6 below: table 6. results of innovation (in) 3.7 summary of energy audit’s results table 7 below shows the final energy audit results according to gbi. the results indicate an existing 18 points only, while the study target must obtain 50 to 66 points to get (certified) rating level. table 7. summary of energy audit’s results 4.0 greening the existing building greening the existing building, involves the potential short-term improvements duration (4 years) of building criteria (retrofit). to achieve (certified) (50 points) rating level. the proposed potential improvements based on the major six criteria, include “energy assessment points score assessment points score innovation & environmental initiatives (max. nine points) 9 0 green building index facilitator 1 0 total score = 0 no. item maximum points existing points 1 energy efficiency (ee) 38 9 2 indoor environment quality (eq) 21 5 3 sustainable site planning & management (sm) 10 1 4 materials & resources (mr) 9 3 5 water efficiency (we) 12 0 6 innovation (in) 10 0 total score 100 18 a greening the existing building (chancellery building university technical malaysia melaka) issn: 2180-1053 vol. 10 no.1 january – june 2018 75 efficiency (ee)” involve labeling all lighting switches, install (50) lighting motion sensors, install (11) electrical sub-metering, install (bipv), reducing building energy intensity (bei) by retrofit the lighting system, and sustainable maintenance plan. “indoor environment quality (eq)” involve installation (18) co2 sensors, develop and implement an “indoor air quality (iaq) management plan”, and conduct “occupancy comfort survey”. “management and sustainable site planning (sm)” involve employing environmentally sensitive building exterior management plan, specify 5% (12 cars) of the parking capacity for low-emitting and fuel efficient vehicles, and provide a building user manual. “materials & resources (mr)” involve reused products/materials in retrofit process, used at least 75% certified wood-based materials in the retrofit process, and develop a sustainable purchasing policy. “water efficiency (we)” involve install rainwater harvesting complete system and install (5) water sub-meters. “innovation (in)” involves using “self – cleaning façade” paint ≥ 10% of total façade area (432 m2), install “electrochromic glazed façade” ≥ 10% of total façade area (120 m2) and taking the advantage of the service of “green building index facilitator”. figure 9 below shows the comparison of the building assessment criteria before and after proposed greening existing building. it can be seen the “energy efficiency (ee)” and “indoor environment quality (eq)” are the heights points. the medium-term duration (8 years) is to evaluate the result of implementing the improvement to determine their effectiveness in moving towards the greening the existing building. 5.0 economic analysis 5.1 cost of greening existing building (retrofit) the estimated cost of greening the chancellery building includes the improvement of building assessment criteria. the total cost of greening the existing building is (800,764 rm). the cost involves the improvement of building assessment criteria to get additional (32 points) to achieve the study target. in addition, (ee) is the most effective part of the total cost. figure 10 below shows the comparison between building assessment criteria cost. figure 9. comparison of the building assessment criteria before and after proposed greening existing building existing points proposed points total points 0 5 10 15 20 25 (ee) (eq) (sm) (mr) (we) (in) 9 5 1 3 0 0 15 5 3 4 2 3 24 10 4 7 2 3 existing points proposed points total points journal of mechanical engineering and technology 76 issn: 2180-1053 vol. 10 no.1 january – june 2018 figure 10. comparison between building assessment criteria cost 5.2 potential saving from retrofit the potential savings include cutting costs from “lighting system” and “building integrated photo voltage”. the potential savings from lighting system is (67345.9 rm/year) and the payback period is (1.68 years). also, the potential saving from building integrated photo voltage is (30492rm/year) and the payback period is (4.9 years). moreover, the total saving is (97837.9rm/year) and the payback period is (2.69 years). figure 11 shows the current electricity consumption and the proposed energy saving from retrofitting the lighting system and installation building integrated photovoltaic (bipv). it can be seen the total energy savings from retrofit the lighting system and installing bipv is 296478 kwh/year. the return of investment (roi) calculation results show the roi for retrofitting lighting system is (59%), roi for installation building integrated photovoltaic is (20%) and the roi for retrofitting lighting system & installation building integrated photovoltaic is (37%). figure 11. power saving from the lighting system retrofit and the proposed power generated from (bipv) 0 100,000 200,000 300,000 400,000 500,000 c o st r m energy efficiency (ee) indoor environment quality (eq) sustainable site planning & management (sm) materials & resources (mr) water efficiency (we) a greening the existing building (chancellery building university technical malaysia melaka) issn: 2180-1053 vol. 10 no.1 january – june 2018 77 6.0 conclusions lack of control and proper power management in modern infrastructure has a negative impact on economy and environment. in a specific case, when it lacks an appropriate organization of its resources such as energy and water, buildings are considered a major key factor of this negative impact. energy consumption in a building has many downsides, in addition to the accumulated cost of power supplied, the emissions of gasses such as carbon dioxide, as a result, is considered one of the factors of the overall environmental pollution. as a solution to these issues, green building practices and approaches can significantly reduce or eradicate negative ecological and economic impacts and offer human comfort, through innovative design, construction, siting, operation, and maintenance. this study involves the methods and techniques for greening the non-residential existing building (chancellery building universiti teknikal malaysia melaka) according to “green building index malaysia (gbi)”. this study concludes that existing chancellery building achieves a low rating level when evaluated according to the gbi rating system. hence, it requires several improvements aspects that are considered applicable under a low budget strategy. 7.0 references ashrae standard. (2007). ventilation for acceptable iaq. ansi/ashrae standard 62.1-2007, 62(1), 41 carroon, j., 2010. sustainable preservation: greening existing buildings (google ebook) dosh, 2010. industry code of practice on indoor air quality. ministry of human resources department of occupational safety and health, 1–50. fadzil, s.f.s., harun, w.m.w. and abdullah, a., (2015). a method to determine a single point percentage daylight factor (%df) value from field work data. advances in environmental biology, 9, 39–42. retrieved from : http://www.aensiweb.com/aeb/. albert t., terry n. and william j. y. (2013). handbook of energy audit. lilburn, ga : fairmont press ; boca raton, fl : distributed by taylor & francis, ©2013.. gbi malaysia, (2011). non-residential existing building [online]. retrived 5 december 2017 from: http://new.greenbuildingindex.org/how/tools green building council, 2009. green building design and construction for the design construction, u.s. green building council. malaysia, d. of s., 2007. ms 1525:2007: code of practice on energy efficiency and use of renewable energy for non-residential building, natural resources canada (2009). energy saving toolbox – an energy audit manual journal of mechanical engineering and technology 78 issn: 2180-1053 vol. 10 no.1 january – june 2018 and tool. ecoenergy, 277. noranai, z. and kammalluden, m.n. (2012). study of building energy index in universiti tun hussein onn malaysia. academic journal of science, 1(2), 429– 433. reterive from: http://universitypublications.net/ajs/0102/html/trn168.xml. sukri, m. f. (2012). an analytical investigation of overall thermal transfer value on commercial building in malaysia. international review of mechanical engineering, 6(5), 1050. tuan, t.b. (2015). energy analysis for lighting and air-conditioning system of an academic building. jurnal teknologi, 5, 53–56. yudelson, j. (2007). green building a to z: understanding the language of green building, new society publishers. zaid, s.m., myeda, n.e., mahyuddin, n. and sulaiman, r. (2013). the need for energy efficiency legislation in the malaysian building sector. the 3rd international building control conference 9,19. zakaria, r.b. et al., 2012. potential retrofitting of existing campus buildings to green buildings. applied mechanics and materials, 178–181.42, 42–45. issn: 2180-1053 vol. 7 no. 2 july december 2015 design and validation of a device to aid in extension ladder setup 1 design and validation of a device to aid in extension ladder setup joseph c. musto1*, adam c. resnick1, michael c. fricke1 1mechanical engineering department, milwaukee school of engineering, 1025 n. broadway, milwaukee, wi 53202-3109, usa abstract the problem of ladder base slippage is a leading cause of workplace injuries and causes a number of annual deaths. research has shown that ladder users tend to set up extension ladders at an angle between 66° and 69° above horizontal, which is much shallower than the specified standard of 75.5°. this results in an increase in the friction required at the base of the ladder to support the weight of the ladder and its user, and leads to an increased likelood of a slideout accident. to counteract the problem of ladder base slipping, a device was developed to aid the user in achieving a proper setup angle. the device uses a mechanical switch to wired to leds that provide the user feedback on setup angle. the device was tested in a laboratory environment, and was shown to positively impact the ability of the user to erect the ladder at a proper angle. keywords: extension ladder, safety engineering 1.0 introduction the problem of ladder base slippage is a leading cause of workplace injuries and causes a number of annual deaths. in 2008, ladders were responsible for 119 fatalities and over 17,500 serious injuries (simeonov, et al. 2012) . it has been reported that 23% to 33% of straight ladder accidents result from slipping of the ladder’s base (chang, et al. 2005). slipping of a ladder’s base is due primarily to improper angling of ladders above horizontal (i.e., setting the ladder to shallow), which causes an increase in the friction required at the base of the ladder to support the weight of the ladder and its user. research indicates that ladder users tend to set up extension ladders at an angle between 66° and 69° above horizontal, which is much shallower than the specified standard of 75.5° (simeonov, et al. 2012). in static loading, setting a ladder at 69° requires a coefficient of friction 50% greater than a * corresponding author email: musto@msoe.edu issn: 2180-1053 vol. 7 no. 2 july december 2015 journal of mechanical engineering and technology 2 properly installed ladder and may easily induce slipping of the base (wilson, 1990). the research of campbell and pagano indicates that instruction and training were not sufficient to ensure proper setup of an extension ladder (campbell and pagano, 2014). this was also indicated by simeonov et al. (2012) which concluded that devices could be used to aid the user in obtaining a proper setup angle (simeonov, et al. 2012). a patented device was developed by the simeonov group. other work on active warning systems for ladder safety have shown them to be effective (musto, et al. 2013). while the work of the simeonov group has shown that active warning devices are more accurate and faster to use than traditional methods, the technology has not yet been commercialized. current commercialized solutions are limited to standard l-labels, which are generally factoryapplied, and bubble level devices, which can be found in the $15 us price range. the l-label method has been shown to be difficult and confusing to use, and results in a wide deviation in setup angle in user testing (campbell, 2012). bubble levels have been shown to be more effective than l-labels or unaided setup (young and wogalter, 2000), cambell (2012), but are subject to human interpretation; they therefore require significant iteration and increase setup time (simeonov, et al. 2013). a newer approach is provided by the free or low-cost smart phone applications available for ladder safety. these products use the accelerometers in a smart phone to indicte ladder angle, and warn the user of a deviation from the proper setup angle. while these show promise as a safety device (simeonov, et al. 2013), they require access to a smart phone, and are handheld devices. since they are not attachable to the ladder, they may not be consistently used, and they cannot give feedback once climbing has begun. in this paper, a novel device for aiding the user in obtaining a safe setup angle is detailed. this device was developed by undergraduate mechanical engineering students as part of the senior design sequence at the milwaukee school of engineering. the details of the device will be shown, including how the range of acceptable deviation from the nominal accepted angle was selected. testing of the device, which demonstrates that it successfully aids a user iin achieving a safe setup angle, will be detailed. issn: 2180-1053 vol. 7 no. 2 july december 2015 design and validation of a device to aid in extension ladder setup 3 2.0 design of the device the issue of extension ladders slipping at the base causes numerous significant injuries and several deaths each year. a leading cause for slipping of ladders is improper ladder usage; this often involves users setting the ladder at an improper angle. several devices have been created and proposed to alleviate either the risk of slipping by increasing friction at the base or to mitigate user error in the setup of ladders. many devices that measure the angle of ladders have been created, including smartphone apps written to measure angles of surfaces, bubble levels designed to specifically fit within a ladder’s rung to aid the user in ladder setup, as well as more complex electronic devices developed to provide precise angle readouts to the user. ladders with bubble level indicators have been shown to be significantly more time consuming to setup than continuous-feedback electronic devices used to setup the ladders (simeonov, et al. 2013). however, the currently available electronic devices consist of several parts, contain complicated microprocessor circuitry, and have large displays that lead to high costs. the purpose of this project was to develop a low-cost, easy-to-use sensor to aid users in the setup of extension ladders. the sensor allows the user to know whether the ladder is properly set within a certain range of the ansi-specified angle of 75.5°. the overall goal of the project was to develop a sensor that will provide fast, accurate, and precise measurement of ladder angle to recommend whether or not a user should use the ladder as it is set, or to change the angle of the ladder. the project scope included development of the device, construction of a prototype, and testing of the prototype’s functionality. the design of the device utilizes a pendulum-like mechanical switch to determine whether the ladder is set properly, too steep, or too shallow. contact is made with one of two terminals if the angle is incorrect, and no contact is made when the angle is correct. this allowed for electrical circuitry to be designed to allow for the logic shown in figure 1. 3 the design of the device utilizes a pendulum-like mechanical switch to determine whether the ladder is set properly, too steep, or too shallow. contact is made with one of two terminals if the angle is incorrect, and no contact is made when the angle is correct. this allowed for electrical circuitry to be designed to allow for the logic shown in figure 1. turn device on if terminal a receives current if terminal b receives current false turn on red led #1 true turn on red led #2 true turn on green led false figure 1. switching logic as shown in the logic diagram, the user interface consists of 3 leds that alert the user whether the ladder is set properly or improperly. the prototype designed for testing purposes is shown in figure 2. figure 2. mechanical design of the device a key parameter in the design is the swing angle of the sensing pendulum. the swing angle of the pendulum corresponds to the allowable tolerance on ladder setup angle. to determine the acceptable tolerance, static calculations were performed on various ladder setup conditions. these analyses follow previous published analyses (wilson, 1990, barnett and liber, 2004). for analysis, four setup scenarios were analyzed, assuming no figure 1. switching logic issn: 2180-1053 vol. 7 no. 2 july december 2015 journal of mechanical engineering and technology 4 as shown in the logic diagram, the user interface consists of 3 leds that alert the user whether the ladder is set properly or improperly. the prototype designed for testing purposes is shown in figure 2. 3 the design of the device utilizes a pendulum-like mechanical switch to determine whether the ladder is set properly, too steep, or too shallow. contact is made with one of two terminals if the angle is incorrect, and no contact is made when the angle is correct. this allowed for electrical circuitry to be designed to allow for the logic shown in figure 1. turn device on if terminal a receives current if terminal b receives current false turn on red led #1 true turn on red led #2 true turn on green led false figure 1. switching logic as shown in the logic diagram, the user interface consists of 3 leds that alert the user whether the ladder is set properly or improperly. the prototype designed for testing purposes is shown in figure 2. figure 2. mechanical design of the device a key parameter in the design is the swing angle of the sensing pendulum. the swing angle of the pendulum corresponds to the allowable tolerance on ladder setup angle. to determine the acceptable tolerance, static calculations were performed on various ladder setup conditions. these analyses follow previous published analyses (wilson, 1990, barnett and liber, 2004). for analysis, four setup scenarios were analyzed, assuming no figure 2. mechanical design of the device a key parameter in the design is the swing angle of the sensing pendulum. the swing angle of the pendulum corresponds to the allowable tolerance on ladder setup angle. to determine the acceptable tolerance, static calculations were performed on various ladder setup conditions. these analyses follow previous published analyses (wilson, 1990, barnett and liber, 2004). for analysis, four setup scenarios were analyzed, assuming no additional devices were to be used to aid in the support of the ladder. ladders and their feet may be made of several materials, including rubber (similar to that in a tire) and wood. various materials may act as the ground support as well, including asphalt, grass, wood, concrete, and stone. table 1 shows various friction coefficients for possible combinations of foot and ground materials (engineers’s handbook, 2006) table 1. friction coefficients for various material combinations for a typical ladder 4 additional devices were to be used to aid in the support of the ladder. ladders and their feet may be made of several materials, including rubber (similar to that in a tire) and wood. various materials may act as the ground support as well, including asphalt, grass, wood, concrete, and stone. table 1 shows various friction coefficients for possible combinations of foot and ground materials (engineers’s handbook, 2006) table 1. friction coefficients for various material combinations for a typical ladder based on the friction coefficient values in table, the minimum friction coefficient at the base was assumed to be 𝜇𝜇𝐴𝐴 = 0.25. 2.1 case i analysis a static analysis was performed given a ladder set up on level ground, top supported by vertical wall, as shown in figure 3. figure 3. setup of the ladder for case i for the setup, a free body diagram may be drawn and static equilibrium conditions may be applied, as shown in figure 4, to determine the friction required at the base of the ladder. ladder foot material ground material friction coefficient rubber asphalt, dry 0.9 rubber asphalt, wet 0.25 0.75 rubber dry concrete 0.6 0.85 rubber wet concrete 0.45 0.75 wood clean wood 0.25 0.5 wood wet wood 0.2 wood stone 0.2 0.4 wood concrete 0.62 issn: 2180-1053 vol. 7 no. 2 july december 2015 design and validation of a device to aid in extension ladder setup 5 based on the friction coefficient values in table 3, the minimum friction coefficient at the base was assumed to be μa=0.25. 2.1 case i analysis a static analysis was performed given a ladder set up on level ground, top supported by vertical wall, as shown in figure 3. 4 additional devices were to be used to aid in the support of the ladder. ladders and their feet may be made of several materials, including rubber (similar to that in a tire) and wood. various materials may act as the ground support as well, including asphalt, grass, wood, concrete, and stone. table 1 shows various friction coefficients for possible combinations of foot and ground materials (engineers’s handbook, 2006) table 1. friction coefficients for various material combinations for a typical ladder based on the friction coefficient values in table, the minimum friction coefficient at the base was assumed to be 𝜇𝜇𝐴𝐴 = 0.25. 2.1 case i analysis a static analysis was performed given a ladder set up on level ground, top supported by vertical wall, as shown in figure 3. figure 3. setup of the ladder for case i for the setup, a free body diagram may be drawn and static equilibrium conditions may be applied, as shown in figure 4, to determine the friction required at the base of the ladder. ladder foot material ground material friction coefficient rubber asphalt, dry 0.9 rubber asphalt, wet 0.25 0.75 rubber dry concrete 0.6 0.85 rubber wet concrete 0.45 0.75 wood clean wood 0.25 0.5 wood wet wood 0.2 wood stone 0.2 0.4 wood concrete 0.62 figure 3. setup of the ladder for case i for the setup, a free body diagram may be drawn and static equilibrium conditions may be applied, as shown in figure 4, to determine the friction required at the base of the ladder. 5 figure 4. free body diagram for the ladder case i the weight vector as shown in is the resultant of the weight of the ladder and the weight of the user, located between the center of the ladder (assumed to be the ladder’s center of gravity) and the location of the user. thus, the location l of the center of gravity of the system is given by equation 1, where 𝑚𝑚1 and 𝑙𝑙1 are the mass of the ladder and location of its center of gravity, and 𝑚𝑚2 and 𝑙𝑙2 are the mass and location of the user. 𝑙𝑙 = ∑ 𝑚𝑚𝑖𝑖𝑙𝑙𝑖𝑖 𝑛𝑛 𝑖𝑖=1 σ𝑚𝑚 = 𝑚𝑚1𝑙𝑙1+𝑚𝑚2𝑙𝑙2 𝑚𝑚1+𝑚𝑚2 (1) 𝑚𝑚 = 𝑚𝑚1 + 𝑚𝑚2 (2) thus, the equilibrium equations are, σ𝐹𝐹𝑥𝑥 = 0 = 𝜇𝜇𝐴𝐴𝐴𝐴𝑦𝑦 − 𝐵𝐵𝑥𝑥 (3) 𝐹𝐹𝑦𝑦 = 0 = 𝐴𝐴𝑦𝑦 + 𝜇𝜇𝐵𝐵 𝐵𝐵𝑥𝑥 − 𝑚𝑚𝑚𝑚 (4) σ𝑀𝑀𝐴𝐴 = 0 = −(𝑙𝑙 cos 𝜃𝜃)𝑚𝑚𝑚𝑚 + 𝐵𝐵𝑥𝑥 [(𝐿𝐿 cos 𝜃𝜃)𝜇𝜇𝐵𝐵 + (𝐿𝐿 sin 𝜃𝜃)] (5) where 𝜇𝜇𝐴𝐴 and 𝜇𝜇𝐵𝐵 are the coefficients ot friction at the bottom and top of the ladder respectively. simplifying and solving for the required coefficient of friction yields, 𝜇𝜇𝐴𝐴 = 1 𝐿𝐿 𝑙𝑙 (𝜇𝜇𝐵𝐵+tan 𝜃𝜃)−𝜇𝜇𝐵𝐵 = [ 𝐿𝐿(𝑚𝑚1+𝑚𝑚2) 𝑚𝑚1(0.5𝐿𝐿)+𝑚𝑚2𝑙𝑙2 (𝜇𝜇𝐵𝐵 + tan 𝜃𝜃) − 𝜇𝜇𝐵𝐵 ] −1 (6) figure 4. free body diagram for the ladder case i issn: 2180-1053 vol. 7 no. 2 july december 2015 journal of mechanical engineering and technology 6 the weight vector as shown in is the resultant of the weight of the ladder and the weight of the user, located between the center of the ladder (assumed to be the ladder’s center of gravity) and the location of the user. thus, the location l of the center of gravity of the system is given by equation 1, where m1 and l1 are the mass of the ladder and location of its center of gravity, and m2 and l2 are the mass and location of the user. 5 figure 4. free body diagram for the ladder case i the weight vector as shown in is the resultant of the weight of the ladder and the weight of the user, located between the center of the ladder (assumed to be the ladder’s center of gravity) and the location of the user. thus, the location l of the center of gravity of the system is given by equation 1, where 𝑚𝑚1 and 𝑙𝑙1 are the mass of the ladder and location of its center of gravity, and 𝑚𝑚2 and 𝑙𝑙2 are the mass and location of the user. 𝑙𝑙 = ∑ 𝑚𝑚𝑖𝑖𝑙𝑙𝑖𝑖 𝑛𝑛 𝑖𝑖=1 σ𝑚𝑚 = 𝑚𝑚1𝑙𝑙1+𝑚𝑚2𝑙𝑙2 𝑚𝑚1+𝑚𝑚2 (1) 𝑚𝑚 = 𝑚𝑚1 + 𝑚𝑚2 (2) thus, the equilibrium equations are, σ𝐹𝐹𝑥𝑥 = 0 = 𝜇𝜇𝐴𝐴𝐴𝐴𝑦𝑦 − 𝐵𝐵𝑥𝑥 (3) 𝐹𝐹𝑦𝑦 = 0 = 𝐴𝐴𝑦𝑦 + 𝜇𝜇𝐵𝐵 𝐵𝐵𝑥𝑥 − 𝑚𝑚𝑚𝑚 (4) σ𝑀𝑀𝐴𝐴 = 0 = −(𝑙𝑙 cos 𝜃𝜃)𝑚𝑚𝑚𝑚 + 𝐵𝐵𝑥𝑥 [(𝐿𝐿 cos 𝜃𝜃)𝜇𝜇𝐵𝐵 + (𝐿𝐿 sin 𝜃𝜃)] (5) where 𝜇𝜇𝐴𝐴 and 𝜇𝜇𝐵𝐵 are the coefficients ot friction at the bottom and top of the ladder respectively. simplifying and solving for the required coefficient of friction yields, 𝜇𝜇𝐴𝐴 = 1 𝐿𝐿 𝑙𝑙 (𝜇𝜇𝐵𝐵+tan 𝜃𝜃)−𝜇𝜇𝐵𝐵 = [ 𝐿𝐿(𝑚𝑚1+𝑚𝑚2) 𝑚𝑚1(0.5𝐿𝐿)+𝑚𝑚2𝑙𝑙2 (𝜇𝜇𝐵𝐵 + tan 𝜃𝜃) − 𝜇𝜇𝐵𝐵 ] −1 (6) thus, the equilibrium equations are, 5 figure 4. free body diagram for the ladder case i the weight vector as shown in is the resultant of the weight of the ladder and the weight of the user, located between the center of the ladder (assumed to be the ladder’s center of gravity) and the location of the user. thus, the location l of the center of gravity of the system is given by equation 1, where 𝑚𝑚1 and 𝑙𝑙1 are the mass of the ladder and location of its center of gravity, and 𝑚𝑚2 and 𝑙𝑙2 are the mass and location of the user. 𝑙𝑙 = ∑ 𝑚𝑚𝑖𝑖𝑙𝑙𝑖𝑖 𝑛𝑛 𝑖𝑖=1 σ𝑚𝑚 = 𝑚𝑚1𝑙𝑙1+𝑚𝑚2𝑙𝑙2 𝑚𝑚1+𝑚𝑚2 (1) 𝑚𝑚 = 𝑚𝑚1 + 𝑚𝑚2 (2) thus, the equilibrium equations are, σ𝐹𝐹𝑥𝑥 = 0 = 𝜇𝜇𝐴𝐴𝐴𝐴𝑦𝑦 − 𝐵𝐵𝑥𝑥 (3) 𝐹𝐹𝑦𝑦 = 0 = 𝐴𝐴𝑦𝑦 + 𝜇𝜇𝐵𝐵 𝐵𝐵𝑥𝑥 − 𝑚𝑚𝑚𝑚 (4) σ𝑀𝑀𝐴𝐴 = 0 = −(𝑙𝑙 cos 𝜃𝜃)𝑚𝑚𝑚𝑚 + 𝐵𝐵𝑥𝑥 [(𝐿𝐿 cos 𝜃𝜃)𝜇𝜇𝐵𝐵 + (𝐿𝐿 sin 𝜃𝜃)] (5) where 𝜇𝜇𝐴𝐴 and 𝜇𝜇𝐵𝐵 are the coefficients ot friction at the bottom and top of the ladder respectively. simplifying and solving for the required coefficient of friction yields, 𝜇𝜇𝐴𝐴 = 1 𝐿𝐿 𝑙𝑙 (𝜇𝜇𝐵𝐵+tan 𝜃𝜃)−𝜇𝜇𝐵𝐵 = [ 𝐿𝐿(𝑚𝑚1+𝑚𝑚2) 𝑚𝑚1(0.5𝐿𝐿)+𝑚𝑚2𝑙𝑙2 (𝜇𝜇𝐵𝐵 + tan 𝜃𝜃) − 𝜇𝜇𝐵𝐵 ] −1 (6) where μa and μb are the coefficients ot friction at the bottom and top of the ladder respectively. simplifying and solving for the required coefficient of friction yields, 5 figure 4. free body diagram for the ladder case i the weight vector as shown in is the resultant of the weight of the ladder and the weight of the user, located between the center of the ladder (assumed to be the ladder’s center of gravity) and the location of the user. thus, the location l of the center of gravity of the system is given by equation 1, where 𝑚𝑚1 and 𝑙𝑙1 are the mass of the ladder and location of its center of gravity, and 𝑚𝑚2 and 𝑙𝑙2 are the mass and location of the user. 𝑙𝑙 = ∑ 𝑚𝑚𝑖𝑖𝑙𝑙𝑖𝑖 𝑛𝑛 𝑖𝑖=1 σ𝑚𝑚 = 𝑚𝑚1𝑙𝑙1+𝑚𝑚2𝑙𝑙2 𝑚𝑚1+𝑚𝑚2 (1) 𝑚𝑚 = 𝑚𝑚1 + 𝑚𝑚2 (2) thus, the equilibrium equations are, σ𝐹𝐹𝑥𝑥 = 0 = 𝜇𝜇𝐴𝐴𝐴𝐴𝑦𝑦 − 𝐵𝐵𝑥𝑥 (3) 𝐹𝐹𝑦𝑦 = 0 = 𝐴𝐴𝑦𝑦 + 𝜇𝜇𝐵𝐵 𝐵𝐵𝑥𝑥 − 𝑚𝑚𝑚𝑚 (4) σ𝑀𝑀𝐴𝐴 = 0 = −(𝑙𝑙 cos 𝜃𝜃)𝑚𝑚𝑚𝑚 + 𝐵𝐵𝑥𝑥 [(𝐿𝐿 cos 𝜃𝜃)𝜇𝜇𝐵𝐵 + (𝐿𝐿 sin 𝜃𝜃)] (5) where 𝜇𝜇𝐴𝐴 and 𝜇𝜇𝐵𝐵 are the coefficients ot friction at the bottom and top of the ladder respectively. simplifying and solving for the required coefficient of friction yields, 𝜇𝜇𝐴𝐴 = 1 𝐿𝐿 𝑙𝑙 (𝜇𝜇𝐵𝐵+tan 𝜃𝜃)−𝜇𝜇𝐵𝐵 = [ 𝐿𝐿(𝑚𝑚1+𝑚𝑚2) 𝑚𝑚1(0.5𝐿𝐿)+𝑚𝑚2𝑙𝑙2 (𝜇𝜇𝐵𝐵 + tan 𝜃𝜃) − 𝜇𝜇𝐵𝐵 ] −1 (6) 2.2 case ii analysis certain situations require a ladder to be set with its top supported by an edge somewhere in the middle of the ladder. a schematic depicting the setup of the ladder of total length lt on level ground set at angle θ is shown in figure 5. issn: 2180-1053 vol. 7 no. 2 july december 2015 design and validation of a device to aid in extension ladder setup 7 6 2.2 case ii analysis certain situations require a ladder to be set with its top supported by an edge somewhere in the middle of the ladder. a schematic depicting the setup of the ladder of total length lt on level ground set at angle 𝜃𝜃 is shown in figure 5. figure 5. setup of the ladder for case ii for the setup, the free body diagram is shown in figure 6. figure 6. free body diagram for the ladder in case ii performing equilibrium analysis as in the case i analysis, and solving for the required coefficient of friction, yields, figure 5. setup of the ladder for case ii for the setup, the free body diagram is shown in figure 6. 6 2.2 case ii analysis certain situations require a ladder to be set with its top supported by an edge somewhere in the middle of the ladder. a schematic depicting the setup of the ladder of total length lt on level ground set at angle 𝜃𝜃 is shown in figure 5. figure 5. setup of the ladder for case ii for the setup, the free body diagram is shown in figure 6. figure 6. free body diagram for the ladder in case ii performing equilibrium analysis as in the case i analysis, and solving for the required coefficient of friction, yields, figure 6. free body diagram for the ladder in case ii performing equilibrium analysis as in the case i analysis, and solving for the required coefficient of friction, yields, 7 𝜇𝜇𝐴𝐴 = (𝑚𝑚1(0.5𝐿𝐿𝑇𝑇)+𝑚𝑚2𝑙𝑙2)cos𝜃𝜃(sin𝜃𝜃−𝜇𝜇𝐵𝐵 cos𝜃𝜃) 𝐿𝐿(𝑚𝑚1+𝑚𝑚2)−(𝑚𝑚1(0.5𝐿𝐿𝑇𝑇)+𝑚𝑚2𝑙𝑙2)cos𝜃𝜃(𝜇𝜇𝐵𝐵 sin𝜃𝜃+cos𝜃𝜃) (7) 2.3 case iii analysis in certain cases, ladder users improperly set a ladder on sloped ground even though it is warned against by ladder manufacturers and standards. understanding the effects of ground surfaces that may not be level is important in understanding usage of a device used to predict safe use of a ladder. figure 7 shows the setup of a ladder supported by a wall on a sloped ground surface. figure 7. setup of the ladder for case iii following the analysis of barnett and liber (barnett and liber, 2004), the static analysis was performed for this case. the required coefficient of friction that resulted is, 𝜇𝜇𝐴𝐴 = tan𝜙𝜙(𝑚𝑚1+𝑚𝑚2)(𝐿𝐿𝜇𝜇𝐵𝐵+𝐿𝐿tan𝜃𝜃)+(𝑚𝑚1𝑙𝑙1+𝑚𝑚2𝑙𝑙2)(1−𝜇𝜇𝐵𝐵 tan𝜙𝜙) (𝑚𝑚1+𝑚𝑚2)(𝐿𝐿𝜇𝜇𝐵𝐵+𝐿𝐿 tan𝜃𝜃)−(𝑚𝑚1𝑙𝑙1+𝑚𝑚2𝑙𝑙2)(tan𝜙𝜙+𝝁𝝁𝑩𝑩) (8) 2.4 case iv analysis mathematical analysis was performed on a ladder setup with sloped ground, supported by an edge, as shown in figure 8. issn: 2180-1053 vol. 7 no. 2 july december 2015 journal of mechanical engineering and technology 8 2.3 case iii analysis in certain cases, ladder users improperly set a ladder on sloped ground even though it is warned against by ladder manufacturers and standards. understanding the effects of ground surfaces that may not be level is important in understanding usage of a device used to predict safe use of a ladder. figure 7 shows the setup of a ladder supported by a wall on a sloped ground surface. 7 𝜇𝜇𝐴𝐴 = (𝑚𝑚1(0.5𝐿𝐿𝑇𝑇)+𝑚𝑚2𝑙𝑙2)cos𝜃𝜃(sin𝜃𝜃−𝜇𝜇𝐵𝐵 cos𝜃𝜃) 𝐿𝐿(𝑚𝑚1+𝑚𝑚2)−(𝑚𝑚1(0.5𝐿𝐿𝑇𝑇)+𝑚𝑚2𝑙𝑙2)cos𝜃𝜃(𝜇𝜇𝐵𝐵 sin𝜃𝜃+cos𝜃𝜃) (7) 2.3 case iii analysis in certain cases, ladder users improperly set a ladder on sloped ground even though it is warned against by ladder manufacturers and standards. understanding the effects of ground surfaces that may not be level is important in understanding usage of a device used to predict safe use of a ladder. figure 7 shows the setup of a ladder supported by a wall on a sloped ground surface. figure 7. setup of the ladder for case iii following the analysis of barnett and liber (barnett and liber, 2004), the static analysis was performed for this case. the required coefficient of friction that resulted is, 𝜇𝜇𝐴𝐴 = tan𝜙𝜙(𝑚𝑚1+𝑚𝑚2)(𝐿𝐿𝜇𝜇𝐵𝐵+𝐿𝐿tan𝜃𝜃)+(𝑚𝑚1𝑙𝑙1+𝑚𝑚2𝑙𝑙2)(1−𝜇𝜇𝐵𝐵 tan𝜙𝜙) (𝑚𝑚1+𝑚𝑚2)(𝐿𝐿𝜇𝜇𝐵𝐵+𝐿𝐿 tan𝜃𝜃)−(𝑚𝑚1𝑙𝑙1+𝑚𝑚2𝑙𝑙2)(tan𝜙𝜙+𝝁𝝁𝑩𝑩) (8) 2.4 case iv analysis mathematical analysis was performed on a ladder setup with sloped ground, supported by an edge, as shown in figure 8. figure 7. setup of the ladder for case iii following the analysis of barnett and liber (barnett and liber, 2004), the static analysis was performed for this case. the required coefficient of friction that resulted is, 7 𝜇𝜇𝐴𝐴 = (𝑚𝑚1(0.5𝐿𝐿𝑇𝑇)+𝑚𝑚2𝑙𝑙2)cos𝜃𝜃(sin𝜃𝜃−𝜇𝜇𝐵𝐵 cos𝜃𝜃) 𝐿𝐿(𝑚𝑚1+𝑚𝑚2)−(𝑚𝑚1(0.5𝐿𝐿𝑇𝑇)+𝑚𝑚2𝑙𝑙2)cos𝜃𝜃(𝜇𝜇𝐵𝐵 sin𝜃𝜃+cos𝜃𝜃) (7) 2.3 case iii analysis in certain cases, ladder users improperly set a ladder on sloped ground even though it is warned against by ladder manufacturers and standards. understanding the effects of ground surfaces that may not be level is important in understanding usage of a device used to predict safe use of a ladder. figure 7 shows the setup of a ladder supported by a wall on a sloped ground surface. figure 7. setup of the ladder for case iii following the analysis of barnett and liber (barnett and liber, 2004), the static analysis was performed for this case. the required coefficient of friction that resulted is, 𝜇𝜇𝐴𝐴 = tan𝜙𝜙(𝑚𝑚1+𝑚𝑚2)(𝐿𝐿𝜇𝜇𝐵𝐵+𝐿𝐿tan𝜃𝜃)+(𝑚𝑚1𝑙𝑙1+𝑚𝑚2𝑙𝑙2)(1−𝜇𝜇𝐵𝐵 tan𝜙𝜙) (𝑚𝑚1+𝑚𝑚2)(𝐿𝐿𝜇𝜇𝐵𝐵+𝐿𝐿 tan𝜃𝜃)−(𝑚𝑚1𝑙𝑙1+𝑚𝑚2𝑙𝑙2)(tan𝜙𝜙+𝝁𝝁𝑩𝑩) (8) 2.4 case iv analysis mathematical analysis was performed on a ladder setup with sloped ground, supported by an edge, as shown in figure 8. 2.4 case iv analysis mathematical analysis was performed on a ladder setup with sloped ground, supported by an edge, as shown in figure 8. issn: 2180-1053 vol. 7 no. 2 july december 2015 design and validation of a device to aid in extension ladder setup 9 8 figure 8. setup of the ladder for case iv again following barnett and liber (barnett and liber, 2004), the required coefficient of friction was determined to be, 𝜇𝜇𝐴𝐴 = 𝐿𝐿 sin 𝜙𝜙(𝑚𝑚1+𝑚𝑚2)+(𝑚𝑚1(0.5𝐿𝐿𝑇𝑇)+𝑚𝑚2𝑙𝑙2) cos 𝜃𝜃(sin(𝜃𝜃−𝜙𝜙)−𝜇𝜇𝐵𝐵 cos(𝜃𝜃−𝜙𝜙)) 𝐿𝐿 cos 𝜙𝜙(𝑚𝑚1+𝑚𝑚2)+(𝑚𝑚1(0.5𝐿𝐿𝑇𝑇)+𝑚𝑚2𝑙𝑙2) cos 𝜃𝜃(−𝜇𝜇𝐵𝐵 sin(𝜃𝜃−𝜙𝜙)−cos(𝜃𝜃−𝜙𝜙)) (9) 2.5 comparison of loading cases in order to determine allowable deviation from the nominal setup angle for the sensor, these four cases were analyzed using some typical ladder values. it was assumed that the ladder user had a weight of 200 lb, a ladder weight of 40 lb, ladder length of 24 ft, 𝜇𝜇𝐵𝐵 = 0.6, and user location of 20.25 ft up the ladder, which, on most extension ladders, would be about the maximum recommended climbing height. for cases iii and iv, ground slopes of 1° and 5° were used. the resulting minimum values for ground friction can be seen in figure 9. figure 8. setup of the ladder for case iv again following barnett and liber (barnett and liber, 2004), the required coefficient of friction was determined to be, 8 figure 8. setup of the ladder for case iv again following barnett and liber (barnett and liber, 2004), the required coefficient of friction was determined to be, 𝜇𝜇𝐴𝐴 = 𝐿𝐿 sin 𝜙𝜙(𝑚𝑚1+𝑚𝑚2)+(𝑚𝑚1(0.5𝐿𝐿𝑇𝑇)+𝑚𝑚2𝑙𝑙2) cos 𝜃𝜃(sin(𝜃𝜃−𝜙𝜙)−𝜇𝜇𝐵𝐵 cos(𝜃𝜃−𝜙𝜙)) 𝐿𝐿 cos 𝜙𝜙(𝑚𝑚1+𝑚𝑚2)+(𝑚𝑚1(0.5𝐿𝐿𝑇𝑇)+𝑚𝑚2𝑙𝑙2) cos 𝜃𝜃(−𝜇𝜇𝐵𝐵 sin(𝜃𝜃−𝜙𝜙)−cos(𝜃𝜃−𝜙𝜙)) (9) 2.5 comparison of loading cases in order to determine allowable deviation from the nominal setup angle for the sensor, these four cases were analyzed using some typical ladder values. it was assumed that the ladder user had a weight of 200 lb, a ladder weight of 40 lb, ladder length of 24 ft, 𝜇𝜇𝐵𝐵 = 0.6, and user location of 20.25 ft up the ladder, which, on most extension ladders, would be about the maximum recommended climbing height. for cases iii and iv, ground slopes of 1° and 5° were used. the resulting minimum values for ground friction can be seen in figure 9. 2.5 comparison of loading cases in order to determine allowable deviation from the nominal setup angle for the sensor, these four cases were analyzed using some typical ladder values. it was assumed that the ladder user had a weight of 200 lb, a ladder weight of 40 lb, ladder length of 24 ft, μb=0.6, and user location of 20.25 ft up the ladder, which, on most extension ladders, would be about the maximum recommended climbing height. for cases iii and iv, ground slopes of 1° and 5° were used. the resulting minimum values for ground friction can be seen in figure 9. issn: 2180-1053 vol. 7 no. 2 july december 2015 journal of mechanical engineering and technology 10 9 figure 9. plot comparing the minimum required friction at the base at 73.5°, the minimum required friction coefficient is about 0.25. based on the values in table, the minimum friction coefficient between rubber and wet asphalt is about 0.25. thus, the sensor must be able to sense the angle within 2.0° of the proper setup angle. as can be seen through the analysis, decreasing the angle to 69°, as one study suggests is within the range that untrained users tend to set ladders (wilson, 1990), requires about 50% more friction than a properly set ladder. this justifies the need for a product to measure the setup angle within the specified accuracy of the device under development. to meet the required angle sensing constraints, the system utilizes a pendulum-like switch, similar to the test switch developed, as shown in figure 10. figure 10. pendulum-type switch developed for testing its feasibility figure 9. plot comparing the minimum required friction at the base at 73.5°, the minimum required friction coefficient is about 0.25. based on the values in table 3, the minimum friction coefficient between rubber and wet asphalt is about 0.25. thus, the sensor must be able to sense the angle within 2.0° of the proper setup angle. as can be seen through the analysis, decreasing the angle to 69°, as one study suggests is within the range that untrained users tend to set ladders (wilson, 1990), requires about 50% more friction than a properly set ladder. this justifies the need for a product to measure the setup angle within the specified accuracy of the device under development. to meet the required angle sensing constraints, the system utilizes a pendulum-like switch, similar to the test switch developed, as shown in figure 10. 9 figure 9. plot comparing the minimum required friction at the base at 73.5°, the minimum required friction coefficient is about 0.25. based on the values in table, the minimum friction coefficient between rubber and wet asphalt is about 0.25. thus, the sensor must be able to sense the angle within 2.0° of the proper setup angle. as can be seen through the analysis, decreasing the angle to 69°, as one study suggests is within the range that untrained users tend to set ladders (wilson, 1990), requires about 50% more friction than a properly set ladder. this justifies the need for a product to measure the setup angle within the specified accuracy of the device under development. to meet the required angle sensing constraints, the system utilizes a pendulum-like switch, similar to the test switch developed, as shown in figure 10. figure 10. pendulum-type switch developed for testing its feasibility figure 10. pendulum-type switch developed for testing its feasibility issn: 2180-1053 vol. 7 no. 2 july december 2015 design and validation of a device to aid in extension ladder setup 11 the pendulum creates contact with one of the two terminals when rotated, closing one of two circuits. a bearing at the pivot of the pendulum allows for low-friction rotation. the constraints for the device required the switch to allow for accurate reading of a high or low signal by the analog logic circuitry implemented into the device. testing of the prototype based on this switch mechanism will be detailed in the following section. 3.0 laboratory testing of the device the prototype ladder angle sensor was tested to determine whether or not it aided ladder users in a significant and positive manner. the test was setup to mimic a previously published test in which data was collected to determine the effectiveness of various ladder setup techniques. this test included unassisted setup and assisted setup using various methods, including a bubble level, another electronic sensor device, and anthropometric methods (simeonov, et al. 2013) the test was performed with fifteen participants of varying experience and skill level. all participants were required to answer a short survey regarding their ladder usage habits in order to gain an understanding of the experience level of testers. questions included on the survey were, • have you ever had on-the-job training or experience in extension ladder safety? • have you received classroom education in extension ladder safety? • rate your knowledge of and experience with ladder safety (0 = none, 5 = expert) • have you ever used any device to aid in extension ladder setup? • what is the proper straight ladder setup angle? • how often do you use extension ladders? (daily, weekly, monthly, rarely) • which concerns you more? (ladder base slipping, ladder tipping backwards) answers to survey questions are attached at the end of the report with other test data. out of the fifteen participants, one knew the proper setup angle, one has received classroom training on ladder safety, all participants rated themselves as novices or intermediate ladder users, two participants use ladders more than a few times a year, and the majority (73%) were most concerned about ladders slipping than tipping backward. there were no significant peculiarities in the issn: 2180-1053 vol. 7 no. 2 july december 2015 journal of mechanical engineering and technology 12 experience or knowledge of the device testers, but it should be noted that most participants were engineering students who deal with angular measurements on a weekly basis. this is a somewhat different demographic than other published tests, but the overall experience and skill matched closely to the previously published study. the test consisted of three individual sets of tests, using both extended and retracted ladders. the test was split into three sets of testing, which were performed in the order below, • test 1 – the test participants set the ladder without being educated on the proper setup angle, both extended and retracted positions tested, not timed • test 2 – the test participants set the ladder with knowledge that the proper setup angle is 75.5°, both extended and retracted positions tested, not timed • test 3 – the test participants set the ladder using the aid of the prototype device, both extended and retracted positions, timed each test was set up with the ladder starting in a near vertical position, with the base located 0.1 m from the wall. the user was then asked to set the ladder in the correct position. time was recorded for the user to set the ladder for test 3 in order to compare to values in the previously published test. the angle of the ladder was measured using a digital inclinometer with a precision of 0.1°. images of the prototype on the ladder used for the test are shown in figure 11. 11 each test was set up with the ladder starting in a near vertical position, with the base located 0.1 m from the wall. the user was then asked to set the ladder in the correct position. time was recorded for the user to set the ladder for test 3 in order to compare to values in the previously published test. the angle of the ladder was measured using a digital inclinometer with a precision of 0.1°. images of the prototype on the ladder used for the test are shown in figure 11. figure 11. prototype device on the ladder used for testing the distribution of the data is shown in figure 12, and the averaged results for the setup angle with one standard deviation are shown in figure 13. figure 12. distribution of the data from the test 0 2 4 6 8 10 12 n um be r of o cc ur an ce s setup angle (deg) test 1 retracted test 1 extended test 2 retracted test 2 extended test 3 retracted test 3 extended figure 11. prototype device on the ladder used for testing the distribution of the data is shown in figure 12, and the averaged results for the setup angle with one standard deviation are shown in figure 13. issn: 2180-1053 vol. 7 no. 2 july december 2015 design and validation of a device to aid in extension ladder setup 13 11 each test was set up with the ladder starting in a near vertical position, with the base located 0.1 m from the wall. the user was then asked to set the ladder in the correct position. time was recorded for the user to set the ladder for test 3 in order to compare to values in the previously published test. the angle of the ladder was measured using a digital inclinometer with a precision of 0.1°. images of the prototype on the ladder used for the test are shown in figure 11. figure 11. prototype device on the ladder used for testing the distribution of the data is shown in figure 12, and the averaged results for the setup angle with one standard deviation are shown in figure 13. figure 12. distribution of the data from the test 0 2 4 6 8 10 12 n um be r of o cc ur an ce s setup angle (deg) test 1 retracted test 1 extended test 2 retracted test 2 extended test 3 retracted test 3 extended figure 12. distribution of the data from the test 12 figure 13. results for the three tests, extended and retracted, with one standard deviate error bars from the results of the test, it appears that the prototype device yielded a significant improvement in the setup of the ladder, with no data being outside of the acceptable range limits when the device was used. the averages of the tests only fell out of the acceptable range for test 1 – extended, but the range of the data was widespread for the data sets from tests 1 and 2. to further analyze the data, f-tests were performed to validate the effectiveness of the device. the f-test was used to determine the probability that the data contained within two tests were statistically similar. the results of the f-tests are shown below in table 2, showing the probability that data sets are statistically similar. table 2. probability that two tests are statistically similar as determined by the f-test from the f-test results, it appears that the prototype device significantly changed the setup of the ladder. to further analyze the data, t-tests were performed to find p-values for hypotheses regarding the correlation between data sets to determine. the null hypotheses tested were,  hypothesis 1 – extending the ladder has no effect – compare all data for extended setup to corresponding data from retracted 69.0 70.0 71.0 72.0 73.0 74.0 75.0 76.0 77.0 78.0 79.0 a ve ra ge s et up a ng le [d eg ] figure 13. results for the three tests, extended and retracted, with one standard deviate error bars from the results of the test, it appears that the prototype device yielded a significant improvement in the setup of the ladder, with no data being outside of the acceptable range limits when the device was used. the averages of the tests only fell out of the acceptable range for test 1 – extended, but the range of the data was widespread for the data sets from tests 1 and 2. to further analyze the data, f-tests were performed to validate the effectiveness of the device. the f-test was used to determine the probability that the data contained within two tests were statistically similar. the results of the f-tests are shown below in table 2, showing the probability that data sets are statistically similar. issn: 2180-1053 vol. 7 no. 2 july december 2015 journal of mechanical engineering and technology 14 table 2. probability that two tests are statistically similar as determined by the f-test 12 figure 13. results for the three tests, extended and retracted, with one standard deviate error bars from the results of the test, it appears that the prototype device yielded a significant improvement in the setup of the ladder, with no data being outside of the acceptable range limits when the device was used. the averages of the tests only fell out of the acceptable range for test 1 – extended, but the range of the data was widespread for the data sets from tests 1 and 2. to further analyze the data, f-tests were performed to validate the effectiveness of the device. the f-test was used to determine the probability that the data contained within two tests were statistically similar. the results of the f-tests are shown below in table 2, showing the probability that data sets are statistically similar. table 2. probability that two tests are statistically similar as determined by the f-test from the f-test results, it appears that the prototype device significantly changed the setup of the ladder. to further analyze the data, t-tests were performed to find p-values for hypotheses regarding the correlation between data sets to determine. the null hypotheses tested were,  hypothesis 1 – extending the ladder has no effect – compare all data for extended setup to corresponding data from retracted 69.0 70.0 71.0 72.0 73.0 74.0 75.0 76.0 77.0 78.0 79.0 a ve ra ge s et up a ng le [d eg ] from the f-test results, it appears that the prototype device significantly changed the setup of the ladder. to further analyze the data, t-tests were performed to find p-values for hypotheses regarding the correlation between data sets to determine. the null hypotheses tested were, • hypothesis 1 – extending the ladder has no effect – compare all data for extended setup to corresponding data from retracted • hypothesis 2 – education on the proper setup angle has no effect – compare all test 1 data to corresponding test 2 data • hypothesis 3 – use of the prototype device has no effect on ladder setup – compare all test 1 data to corresponding test 3 data, and compare all test 2 data to corresponding test 3 data hypothesis 1 was tested to determine if all the data could be grouped together for analysis in addition to providing a better understanding of the setup of ladders. hypothesis 2 was tested to determine whether or not significant improvement could be detected once a ladder user was instructed on the proper setup angle. hypothesis 3 was tested to determine the probability that the device produces a significant change in the setup of the ladder. the t-test analysis utilized the assumption of a one-tailed distribution, as the data in figure 12 shows a skew in the data, biased to the left of the peak occurrence. the results of the analyses performed are shown in the table below. a 95% confidence interval (p=0.05) was used as the threshold for rejecting a hypothesis. the results are shown in table 3. issn: 2180-1053 vol. 7 no. 2 july december 2015 design and validation of a device to aid in extension ladder setup 15 table 3. results of t-tests performed on the three hypotheses 13  hypothesis 2 – education on the proper setup angle has no effect – compare all test 1 data to corresponding test 2 data  hypothesis 3 – use of the prototype device has no effect on ladder setup – compare all test 1 data to corresponding test 3 data, and compare all test 2 data to corresponding test 3 data hypothesis 1 was tested to determine if all the data could be grouped together for analysis in addition to providing a better understanding of the setup of ladders. hypothesis 2 was tested to determine whether or not significant improvement could be detected once a ladder user was instructed on the proper setup angle. hypothesis 3 was tested to determine the probability that the device produces a significant change in the setup of the ladder. the t-test analysis utilized the assumption of a one-tailed distribution, as the data in figure 12 shows a skew in the data, biased to the left of the peak occurrence. the results of the analyses performed are shown in the table below. a 95% confidence interval (𝑝𝑝 = 0.05) was used as the threshold for rejecting a hypothesis. the results are shown in table 3. table 3. results of t-tests performed on the three hypotheses as seen in table 3, education on the proper setup angle cannot be proven to have any significant effect on the setup of the ladder, and length of the ladder made no detectable difference in ladder setup. however, use of the prototype angle sensor was shown with over 99.9% confidence to improve the setup of the ladder from uneducated setup, and over 99.2% percent confidence for improvement over educated setup. in addition to the angle measurements, time measurements were taken for test 3. these time measurements are shown in comparison to setup time for other methods of determining ladder angle. the methods of setup used in the previously published test were,  no instruction – nearly identical conditions to this experiment’s no instruction test  anthropometric – utilization of the geometry of the human body to determine proper setup angle  bubble indicator – use of a bubble level specifically designed to show proper ladder setup angle as seen in table 3, education on the proper setup angle cannot be proven to have any significant effect on the setup of the ladder, and length of the ladder made no detectable difference in ladder setup. however, use of the prototype angle sensor was shown with over 99.9% confidence to improve the setup of the ladder from uneducated setup, and over 99.2% percent confidence for improvement over educated setup. in addition to the angle measurements, time measurements were taken for test 3. these time measurements are shown in comparison to setup time for other methods of determining ladder angle. the methods of setup used in the previously published test were, • no instruction – nearly identical conditions to this experiment’s no instruction test • anthropometric – utilization of the geometry of the human body to determine proper setup angle • bubble indicator – use of a bubble level specifically designed to show proper ladder setup angle • multimodal indicator – use of an expensive, high-precision electronic device designed specifically for determining proper ladder setup angle a plot with showing the data from this test compared with the results of previous published testing (simeonov, et al. 2013) is shown in figure 14. issn: 2180-1053 vol. 7 no. 2 july december 2015 journal of mechanical engineering and technology 16 14  multimodal indicator – use of an expensive, high-precision electronic device designed specifically for determining proper ladder setup angle a plot with showing the data from this test compared with the results of previous published testing (simeonov, et al. 2013) is shown in figure 14. figure 14. setup time required for the prototype device compared to other ladder setup methods from previously published data (simeonov, et al. 2013) the prototype device developed was shown to be comparable in accuracy and setup time (about 3 seconds) to the multimodal indicator, which performed best in the previously published study. from the testing of the prototype it can be determined that the device performs functionally and produces the desired results. the performance is comparable to or better than all other ladder setup methods that have been tested previously. 4.0 conclusions in this paper, a new device for aided in the proper setup of extension ladders has been presented. the device was designed to be more reliable and simpler to use than other passive devices (e.g. bubble levels), and to be a low-cost alternative to the active sensing alternatives available. the design was detailed, and the acceptable range of sensor accuracy was justified using static analysis. prototype testing was also detailed; this testing demonstrated that the prototype device dramatically improves the likelihood of proper setup angle, with a confidence level of over 99%. in addition, the testing demonstrated an accuracy and setup time comparable with the more expensive multimodal indicator devices. the prototype shows great promise as a device to improve industrial ladder safety. the cost of the prototype was approximately $175.00 us. major cost drivers were the cost of the prototype housing, and the cost of the plating process required for switch contacts. these costs will be greatly reduced in mass production; 69 70 71 72 73 74 75 76 77 78 0 5 10 15 se tu p a ng le [d eg ] time [s] no instruction anthropometric bubble indicator multimodal indicator prototype device figure 14. setup time required for the prototype device compared to other ladder setup methods from previously published data (simeonov, et al. 2013) the prototype device developed was shown to be comparable in accuracy and setup time (about 3 seconds) to the multimodal indicator, which performed best in the previously published study. from the testing of the prototype it can be determined that the device performs functionally and produces the desired results. the performance is comparable to or better than all other ladder setup methods that have been tested previously. 4.0 conclusions in this paper, a new device for aided in the proper setup of extension ladders has been presented. the device was designed to be more reliable and simpler to use than other passive devices (e.g. bubble levels), and to be a low-cost alternative to the active sensing alternatives available. the design was detailed, and the acceptable range of sensor accuracy was justified using static analysis. prototype testing was also detailed; this testing demonstrated that the prototype device dramatically improves the likelihood of proper setup angle, with a confidence level of over 99%. in addition, the testing demonstrated an accuracy and setup time comparable with the more expensive multimodal indicator devices. the prototype shows great promise as a device to improve industrial ladder safety. the cost of the prototype was approximately $175.00 us. major cost drivers were the cost of the prototype housing, and the cost of the plating process required for switch contacts. these costs will be greatly reduced in mass production; initial cost estimates issn: 2180-1053 vol. 7 no. 2 july december 2015 design and validation of a device to aid in extension ladder setup 17 show that the device could be produced in the range of $14.00 us in large quantities. future work is continuing on the optimization of the design for mass production. references barnett, r., and liber, t. (2004). sloped surfaces – ladder slide out. triodyne safety brief, 25(1), 1-6. campbell, a., and pagano, c. (2014). the effects of instructions on potential slide-out failures during portable extension ladder postioning. accident analysis and prevention, 67, 30-39. campbell, a., (2012). an engineering psychology based analysis of ladder setup procedures. master theses. clemson university. usa. chang, w.-r., chien-chi, c., and matz. s. (2005). available friction of ladder shoes and slip potenal for climbing on a straight ladder. ergonomics, 48(9), 1169-1182. engineer’s handbook (2006). reference tables coefficient of friction. retrieved november 10, 2013, from http//www. engineershandbook. com /tables/ frictioncoefficients.htm. musto, j., karsten, f., resnick, a., and berg, k. (2013). design of an active warning system for fall protection. iron and steel technology, 10(13), 58-62. simeonov, p., hsiao, h., powers, j., and kau, t.y. (2012). factors affecting extension ladder angular positioning. human factors. the journal of the human factors and ergonomics society, 54(3), 334-345. simeonov, p., hsiao, h., powers, j., kim, i.j., kau, t.-y., and weaver, d., (2013). research to improve extention ladder angular positioning. applied ergonomics, 44, 496-502. wilson, j. (1990). a generalized gnalysis of the ladder slippage problem. proceedings of the the american society of mechanical engineers winter annual meeting, dallas, texas. young, s. l., and wogalter, m. s. (2000). on improving set-up angle accuracy for extension ladders. proceedings of the human factors and ergonomics society annual meeting, 44(25), 111-114. issn: 2180-1053 vol. 9 no.1 january – june 2017 1 thermal performance and optimum design analysis of fin with variable thermal conductivity using double decomposition method gbeminiyi m. sobamowo1* department of mechanical engineering, university of lagos, akoka, lagos, nigeria. abstract in this paper, thermal performance study and optimum design analysis of straight fin with variable thermal conductivity are carried out using double decomposition method. the developed heat transfer models are used to analyze the thermal performance, establish the optimum thermal design parameters and also, investigate the effects of thermo-geometric parameters and thermal conductivity (non-linear) parameters on the temperature distribution, heat transfer and thermal performance of the longitudinal rectangular fin. from the results, it is established that the fin temperature distribution, the total heat transfer, the fin effectiveness, and the fin efficiency are significantly affected by the thermo-geometric and thermal parameters of the fin. also, it is established that the optimum fin length increases as the non-linear thermal conductivity term, increases. therefore, the operational parameters must be carefully chosen to ensure that the fin retains its primary purpose of removing heat from the primary surface. the results obtained in this analysis provides platform for improvement in the design of fin in heat transfer equipment. keywords: performance analysis; convective optimal design; longitudinal fin; double decomposition method; temperature-dependent thermal conductivity. 1.0 introduction high-performance heat transfer components with progressively small weights, volume and costs are continuously demanded in large numbers of thermal systems. consequently, fins are widely employed in the design and construction of various types of heat-transfer equipment and components such as air conditioning, refrigeration, superheaters, automobile, power plants, heat exchangers, convectional furnaces, economizers, gas turbines, chemical processing equipment, oil carrying pipelines, computer processors, electrical chips etc. the extended surfaces are used to increase the rate of heat transfer *corresponding author e-mail: mikegebeminiyi@gmail.com mailto:mikegebeminiyi@gmail.com journal of mechanical engineering and technology 2 issn: 2180-1053 vol. 9 no.1 january – june 2017 between the primary surface and the fin. in practice, various types of fins with different geometries are used, but due to simplicity of its design and ease of construction and manufacturing process, the rectangular fins are widely used. also, for ordinary fins problem, the thermal properties of the fin thermal conductivity is assumed to be constant, but if large temperature difference exists within the fin, typically, between tip and the base of the fin (such as heat pipe, space radiator etc.), the thermal conductivity is temperaturedependent. these facts attest that for many engineering applications, the thermal conductivity is temperature-dependent. therefore, while analyzing the fin under such situations, effects of the temperature-dependent thermal properties must be taken into consideration. in carrying out such analysis, the thermal conductivity may be modelled for such and other many engineering applications by power law and by linear dependency on temperature (khani & aziz, 2010; ndlovu & moitsheki, 2013). such dependency of thermal conductivity renders the problem non-linear and difficult to solve exactly. over the past few decades, the solution of the governing non-linear differential equations has been constructed using different techniques. aziz and enamul-huq (1973) applied regular perturbation expansion to study a pure convection fin with temperature dependent thermal conductivity. aziz (1977) extended the previous analysis to include a uniform internal heat generation in the fin. few years later, campo and spaulding (1999) applied method of successive approximation to predict the thermal behaviour of uniform circumferential fins. chiu and chen (2002) and arslanturk (2005) adopted the adomian decomposition method (adm) to obtain the temperature distribution in a pure convection fin with variable thermal conductivity. the same problem was also solved by ganji (2006) with the aid of the homotopy perturbation method originally proposed by he (1999). chowdhury and hashim (2008) applied the adomian decomposition method to evaluate the temperature distribution of straight rectangular fin with temperature dependent surface flux for all possible types of heat transfer. in the following year, rajabi (2007) employed homotopy perturbation method (hpm) to calculate the efficiency of straight fins with temperature-dependent thermal conductivity. a year later, mustapha (2008) adopted homotopy analysis method (ham) to find the efficiency of straight fins with temperature-dependent thermal conductivity. also, coskun and atay (2007) utilized variational iteration method (vim) for the analysis of convective straight and radial fins with temperature-dependent thermal conductivity while languri et al. (2008) applied both variation iteration and homotopy perturbation methods for the evaluation of efficiency of straight fins with temperaturedependent thermal conductivity. coskun and atay (2008) applied variational iteration method to analyse the efficiency of convective straight fins with temperature-dependent thermal conductivity. in the same year, atay and coskum (2008) employed variation iteration and finite element methods to carry out comparative analysis of power-law-fin type problems. domairry and fazeli (2009) used homotopy analysis method to determine the efficiency of straight fins with temperature-dependent thermal conductivity. chowdhury et al.(2009) investigated a rectangular fin with power law surface heat flux and made a comparative assessment of results predicted by ham, hpm and adm. khani et al. (2009) used adomian decomposition method (adm) to provide series solution to fin problem with a temperature-dependent thermal conductivity. moitsheki et al. (2010) thermal performance and optimum design analysis of fin with variable thermal conductivity using double decomposition method issn: 2180-1053 vol. 9 no.1 january – june 2017 3 applied the lie symmetry analysis to provide exact solutions of the fin problem with a power-law temperature-dependent thermal conductivity. also, in the same year, hosseini et al. (2012) applied homotopy analysis method to provide approximate but accurate solution of heat transfer in fin with temperature-dependent internal heat generation and thermal conductivity. to the best of the authors’ knowledge, very few studies were actually directed to the analysis of heat transfer in fins with temperature-dependent thermal properties while the study of fin with temperature-dependent internal heat generation, thermal conductivity and heat transfer coefficient are very limited or scarcely carried out in literature. furthermore, differential transform method (dtm) solves the differential equations without linearization, discretization or no approximation, linearization restrictive assumptions or perturbation, complexity of expansion of derivatives and computation of derivatives symbolically. this method was applied by joneidi et al. (2009), moradi and ahmadikia (2010) as well as moradi (2010) presented analytical solution for fin with temperature dependent thermal coefficient. the method was also used by mosayebidorcheh et al. (2014), ghasemi et al. (2014), sandri et al. (2012), ganji and dogonchi (2014) also applied the dtm to solve the fin problem but the search for an auxiliary value that will satisfy the second boundary condition necessitated the use of maple software and such results in additional computational costs and efforts in the generation of solution to the problem. this drawback is not only peculiar to dtm, other approximate analytical methods such as hpm, ham, adm and vim also required additional computational cost, time and efforts for the determination of auxiliary parameters which could lead to tedious and very complicated work to do. also, dtm only provides acceptable approximation for small range i.e. it does not exhibit a good approximation in large domain. this is because a boundary condition is satisfied via the method, and the remaining unsatisfied boundary condition plays no roles in the final results. this deficiency limits the efficiency and the applications of dtm over wide range of problems. hpm, ham, adm and vim often involved complex mathematical analysis leading to analytic expression involving a large number terms and when such methods are routinely implemented, they can sometimes lead to erroneous results (fernandez, 2009) and (aziz and bouaziz, 2011). in practice, approximate analytical solutions with large number of terms are not convenient for use by designers and engineers. inevitably, cost effective and accurate expressions are required to analyse the fin. in order to meet this demand, adomian and rach (1993) modified the adomian decomposition method and introduced the double decomposition method (ddm). yang et al. (2008, 2010) solved the periodic base temperature in convective longitudinal fins using ddm, while chiu and chen (2003) applied the ddm to analyze convectiveradiative fins.the method was found to have more advantages than the adomian decomposition method, including faster convergence,reduced calculations, higher accuracy andprovision of a direct scheme for solving the non-linear problem without the need of linearization and iteration and most importantly, it gives an explicit form of solution to nonlinear problem.it solves non-linear problems without linearization, perturbation, closure approximations, or discretization methods that could result in massive numerical journal of mechanical engineering and technology 4 issn: 2180-1053 vol. 9 no.1 january – june 2017 computations. therefore, in this present work, double decomposition method is applied to analyze thermal performance and optimum thermal design of convective straight fin with temperature-dependent thermal conductivity. the dmm is computationally convenient, provides analytic, direct scheme, verifiable solutions not requiring perturbation, linearization, or discretization and resulting massive computation. it gives faster convergence, reduced calculations, higher accuracy than adm and more importantly, it gives an explicit form of solution to non-linear problem. also, golberg (1999) has shown that adm does not converge in general, in particular, when the method is applied to linear operator equations. furthermore, it was shown that adomian’s decomposition method is equivalent to picard iteration method, and therefore it might diverge. from the previous studies and analysis, it was revealed that the ddm provides a very powerful, novel and accurate approximate analytical solution procedure that is applicable to a wide variety of linear and non-linear problems and thus makes it unnecessary to search for an auxiliary value that will satisfy second the boundary condition as in the case of hpm, ham, adm and vim, and without searching for variational formulations in order to apply the finite element method for the problems and the difficulties associated with proper construction of the approximating functions for arbitrary domains or geometry of interest as in galerkin weighted residual method (gwrm), least square method (lsm) and collocation method (cm) are overcome. although, the method presents its own difficulty in determining the adomian polynomials, am, the resulting solutions from the method are more physically realistic. it would be desirable to find easier ways of generating the adomian polynomials and to study their properties to reduce the computational effort. from the present analysis, the results obtained by the method for solving the problem under investigation are compared with the exact solution for the linear problem and also with the numerical solution for the non-linear case and very good agreements were established. 2.0 problem formulation consider a straight fin of temperature-dependent thermal conductivity k(t), length l and thickness δ that is exposed on both faces to a convective environment at temperature  t and with heat transfer co-efficient h shown in figure1, assuming that the heat flow in the fin and its temperatures remain constant with time, the temperature of the medium surrounding the fin is uniform, the fin base temperature is uniform., there is no contact resistance where the base of the fin joins the prime surface, also the fin thickness is small compared with its width and length, so that temperature gradients acrossthe fin thickness and heat transfer from the edges of the fin may be neglected. the dimension x pertains to the length coordinatewhich has its origin at the tip of the finand has a positive orientation from the fin tip to the fin base. following the model assumptions, the governing differential equation for the problem is shown in equation (1). ( ) ( ) 0 c d dt h k t p t t dx dx a          (1) thermal performance and optimum design analysis of fin with variable thermal conductivity using double decomposition method issn: 2180-1053 vol. 9 no.1 january – june 2017 5 figure 1. schematic of the longitudinal straight fin geometry the boundary conditions are , 0, 0 b x b t t dt x dx     (2) for many engineering applications, the thermal conductivity and the coefficient of heat transfer are temperature-dependent. therefore, the temperature-dependent thermal properties and internal heat generation are given by   [1 ( )]ak t k t t    (3) substituting equation (3) into equation (1), we have ( ) [1 ( )] 0 a c hp t td dt k t t dx dx a             (4) introducing the following dimensionless parameters into equation (4); 2 2 , , , , ( ) b b a c a t tx k phl x k m t t b t t k a k              (5) one arrives at the dimensionless governing differential equation (4) and the boundary conditions 2 (1 ) 0 d d m dx dx            (6) journal of mechanical engineering and technology 6 issn: 2180-1053 vol. 9 no.1 january – june 2017 equation (6) could be written in expanded form as 22 2 2 2 2 0 d d d m dx dx dx                (7) where the boundary conditions are 1, 1 0, 0 x d x dx       (8) 3.0 method of solution: double decomposition method the nonlinearity in the governing equation (7) makes it very difficult to generate a closed form solution for equation (7). therefore, recourse has to be made to either approximation analytical methods, semi-numerical methods or numerical methods of solution. in this work, an approximate analytical method of solution, double decomposition method is used. it makes the calculation accuracy much higher than the adomian decomposition and lowers the computational load. the double decomposition method uses the same operator as the adomian decomposition method, but decomposes the first undefined parameters. to do this, the zero-order decomposition formula is set into the boundary conditions and then evaluates the undefined parameters. the procedure of the method is described as follows: the general nonlinear equation is in the form lu nu ru g   (9) the linear terms are decomposed into l + r, with l taken as the highest order derivative which is easily invertible and r as the remainder of the linear operator of less order than l. where g is the system input or the source term and u is the system output, nu represents the nonlinear terms, which is assumed to be analytic. l-1 is regarded as the inverse operator of l and is defined by a definite integration from 0 to x, i.e. 1 0 [l ]( ) ( ) x f x f v dv    (10) if l is a second-order operator, then l-1 is a two ford indefinite integral i.e. l-1 could be expressed as 1 1 0 [l ]( ) ( ) x x f x f v dvdv     (11) thermal performance and optimum design analysis of fin with variable thermal conductivity using double decomposition method issn: 2180-1053 vol. 9 no.1 january – june 2017 7 applying the inverse operator l-1 to the both sides of equation (9), and using the given conditions, the resulting equation could be written as 1 1 ( )u x l ru l nu      (12) where 1( ) x x l g     and λx represents the term arising from integrating the source term g(x). the adomian methods decomposes the solution u(x) into a series 0 m m u u     (13) and the nonlinear term into a series 0 m m nu a     (14) where am’s are adomian’s polynomials of u0 ,u1, . . ., um and are obtained for the nonlinearity nu = f(u)from the recursive formula 00 0 1 1 [ ( )] 0,1, 2, 3,... ! ! m m i m im m i d d a fu f y m m d m d                            (15) where ζ is a grouping parameter of convenience. using the double decomposition, the integral term λxcould be further decompose as , 0 x x m m       (16) substituting eqs. (13), (14) and (16) into equation (12), we have 1 1 1 ,m 0 0 0 0 m x m m m m m m u l g l r u l a                   (17) assuming that , , 1,x m o m m a xa   .the constants of integration 0, 1,m m a and a can be found from the boundary conditions journal of mechanical engineering and technology 8 issn: 2180-1053 vol. 9 no.1 january – june 2017 therefore, from the established recursive relation in equation (17), one can write the double decompositions solution as 1 0 0,0 1,0 u a xa l g     1 1 1 0,1 1,1 0 0 u a xa l ru l a       1 1 2 0,2 1,2 1 1 u a xa l ru l a       1 1 3 0,3 1,3 2 2 u a xa l ru l a       1 1 4 0,4 1,4 3 3 u a xa l ru l a       . . . 1 1 0,n 1,n 1 1n n n u a xa l ru l a         (18) while the solution obtained by decomposition is generally an infinite series, an (n+1) terms approximation φm to θ usually serves as the practical solution. this could be written as 1 0 1 2 3 0 ... 1 n m n m u u u u u u n            (19) such that 1 lim n n      3.1 the fin temperature distribution from the adomian decomposition analysis, the linear operator is defined as x d l dx  (20) substituting equation (20) into equation (7), we have 22 2 2x d d l m dx dx                (21) equation (21) could also be written as 2 x l m na nb      (22) where the nonlinear terms thermal performance and optimum design analysis of fin with variable thermal conductivity using double decomposition method issn: 2180-1053 vol. 9 no.1 january – june 2017 9 2 2 0 m m d na a dx        (23a) 2 0 m m d nb b dx            (23b) using equation (15) the ai’s and bi’s are expressed as 2 0 0 0 2 d a dx   2 2 0 1 1 1 02 2 d d a dx dx      2 2 2 0 1 2 2 2 1 02 2 2 d d d a dx dx dx         2 22 2 0 31 2 3 3 2 1 02 2 2 2 d dd d a dx dx dx dx           2 22 2 2 0 31 2 4 4 4 3 2 1 02 2 2 2 2 d dd d d a dx dx dx dx dx              . . . 2 2 22 2 0 31 2 1 2 3 02 2 2 2 2 ... m m m m m m d d dd d a dx dx dx dx dx                   (24) and 2 0 0 d b dx        0 1 1 2 d d b dx dx    journal of mechanical engineering and technology 10 issn: 2180-1053 vol. 9 no.1 january – june 2017 2 01 2 2 2 dd d b dx dx dx         0 31 2 3 2 2 d dd d b dx dx dx dx      2 3 02 1 4 4 2 2 d dd d d b dx dx dx dx dx            . . . (25) if we operate 1 x l  on both sides of equation (22), we obtained 1 2 1 1 1 x x x x x l l m l l na l nb           (26) which gives 2 1 1 1 0 x x x m l l na l nb            (27) where the inverse operator 1 1 0 l ( ) ( ) x x dxdx  •  •  the value of the first term can be determined as 0 0,0 1,0 a xa   (28) where the constants 0,0 1,0 a and a can be found from the boundary conditions in equation (8). 2 1 1 1 1 0,1 1,1 0 0 0x x x a xa m l l na l nb            2 1 1 1 2 0,2 1,2 1 1 1x x x a xa m l l na l nb            2 1 1 1 3 0,3 1,3 2 2 2x x x a xa m l l na l nb            (29) generally, the recursive relationship in equation (30) can be used to determine the iterates 2 1 1 1 1 0,m 1 1,m 1m x m x m x m a xa m l l na l nb               (30) thermal performance and optimum design analysis of fin with variable thermal conductivity using double decomposition method issn: 2180-1053 vol. 9 no.1 january – june 2017 11 from eqs. (28) and (29), the first four iterations are given as 0 1  2 2 2 1 2 2 m x m    4 2 4 4 4 2 2 2 2 5 24 2 24 4 2 m m m x m x m x        6 6 6 4 6 6 4 4 4 2 3 4 2 4 2 2 2 2 2 2 2 2 7 5 13 3 720 48 360 24 24 4 5 5 48 48 2 2 2 2 m x m x m m x m m x m x m m x m m x m                    . . . (31) summing up the iterates, gives 1 0 1 2 3 1 0 ... m m m m m                  (32) therefore, the components of θ are determined and are written as m-terms approximation such that lim m m     4.0 fin parameters for thermal performance indications the performance indication parameters for fin includes heat transfer rate at the base of the fin, the total heat flux from the fin, the efficiency and the effectiveness of the fin. in this section, each thermal performance indication parameter is analyzed as follows. 4.1 heat flux of the fin the fin base heat flux is given by journal of mechanical engineering and technology 12 issn: 2180-1053 vol. 9 no.1 january – june 2017 ( ) bn c dt q a k t dx  (33) the dimensionless heat transfer rate at the base of the fin (x=1) is obtained as (1 ) ( ) bn b a c b q l d q k a t t dx        (34) the total heat flux of the fin is given by )( bc b t ttha q q   (35) substituting equation (34) and introducing the dimensionless parameters in equation (5) into equation (35), gives dx d bidx d h k bi q t    )1( 1)(1  (36) 4.2 fin efficiency the amount of heat dissipated from the entire fin is found by using newton’s law of cooling as 1 0 ( ) f q ph t t dx    (37) the maximum heat dissipated is obtained if the fin base temperature is kept throughout the fin max ( ) b q phl t t    (38) fin efficiency is defined as the ratio of the fin heat transfer rate to the rate that would be if the entire fin were at the base temperature and is given by 0 max ( ) ( ) l f b ph t t dx q q phl t t         (39) therefore, the fin efficiency in dimensionless variables is given by thermal performance and optimum design analysis of fin with variable thermal conductivity using double decomposition method issn: 2180-1053 vol. 9 no.1 january – june 2017 13 1 0 (x)dx   (40) it is very important to point out that the thermo-geometric parameter or the fin performance factor, m could be written in terms of biot number, bi and the aspect ratio, ar as shown in equation (41). 22 2 2 2 2 2 (2 ) 2 2 2 ( ) b b b b r c a a a a ph l l h l h l h l m bia a k l k k k                (41) where ,b r a h l bi a k     from equation (41), it implies that biam r 2 (42) 4.3 fin effectiveness the removal number or fin effectiveness is the ratio of the fin dissipation (equal, in the steady state, to the heat passing through the base of the fin by conduction) to the heat passing through the fin footprint of the base or prime surface if the fin were not present. following the definition, the effectiveness of the fin could be expressed mathematically as f fb q q   (43) where qfb is the amount of heat dissipation from the area of the fin base and is given by ( ) 2 fb b b q ph t t     (44) substituting equation (37) and (44) into equation (43), gives 0 2 ( ) ( ) l f fb b ph t t q q ph t t          (45) therefore, the fin effectiveness in dimensionless variables is given by journal of mechanical engineering and technology 14 issn: 2180-1053 vol. 9 no.1 january – june 2017 1 0 2 (x) r a dx   (46) 5.0 fin optimization the optimization of the fin could be achieved either by minimizing the volume (weight) for any required heat dissipation or maximizing the heat dissipation for any given fin volume. the later approach is adopted in this work. the constant fin volume is defined as v=acb. following equation (37), one can therefore write the heat dissipation per unit volume as 0 ( ) l f c ph t t dx q v a b     (47) the dimensionless form of equation (47) is given as 1 0 (t t ) p f p f a b a a c pha dx q a q k v k a           (48) equation (48) could be written as 1 2/3 0 f q m dx   (49) where 2/3 2 p p a h a a b k            the maximum heat dissipation value occurs at the condition when the optimum fin characteristics have been achieved. the fin dimensions in this situation represent the optimum fin configuration per unit volume. with the volume constant, the optimization procedure is also carried out to fix the profile area ap by first expressing f q  as a function of the thermo-geometric parameter, m (or fin length, b) and then searching for the optimum value of m or b. thermal performance and optimum design analysis of fin with variable thermal conductivity using double decomposition method issn: 2180-1053 vol. 9 no.1 january – june 2017 15 6.0 results and discussions figures 2a and 2b show the variation of dimensionless temperature with dimensionless length of the fin and also depict the effect of the thermogeometric parameter on the straight fin with an insulated tip. it is shown that as the thermogeometric parameter increases, the rate of heat transfer (the convective heat transfer) through the fin increases as the temperature in the fin drops faster (becomes steeper reflecting high base heat flow rates) as depicted in the figures. it can be inferred from the results that the ratio of convective heat transfer to conductive heat transfer has much effect on the temperature distribution, rate of heat transfer at the base of the fin, efficiency and effectiveness of the fin. as h increases (or kb decreases), the ratio h/kb increases at the base of the fin and consequently the temperature along the fin, especially at the tip of the fin decreases i.e. the tip end temperature decrease as m increases. the profile has steepest temperature gradient at m=1.0, but it is much higher value gotten from the lower value of thermal conductivity than the other values of m in the profiles produces a lower heat-transfer rate. 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 d im e n s io n le s s t e m p e ra tu re ,  dimensionless lenght, x m= 0.5 m= 1.0 m= 1.5 m= 2.0 (a) journal of mechanical engineering and technology 16 issn: 2180-1053 vol. 9 no.1 january – june 2017 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 d im e n s io n le s s t e m p e ra tu re ,  dimensionless lenght, x m= 0.5 m= 1.0 m= 1.5 m= 2.0 (b) 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 d im e n s io n le s s t e m p e ra tu re ,  dimensionless lenght, x m= 0.5 m= 1.0 m= 1.5 m= 2.0 (c) thermal performance and optimum design analysis of fin with variable thermal conductivity using double decomposition method issn: 2180-1053 vol. 9 no.1 january – june 2017 17 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 d im e n s io n le s s t e m p e ra tu re ,  dimensionless lenght, x m= 0.5 m= 1.0 m= 1.5 m= 2.0 (d) figure 2. effects of thermo-geometric parameter on the temperature distribution in the fin when (a) β= -0.3 (b) β=-0.1(c) β=0.1 (d) β=0.3 this shows that the thermal performance or efficiency of the fin is favoured at low values of thermogeometric parameter since the aim (high effective use of the fin) is to minimize the temperature decrease along the fin length, where the best possible scenario is when t=tb everywhere. one of the major important analyses in the fin problem is the determination of the rate of heat transfer at the base of the fin. figure 3a shows the effects of no-linear or thermal conductivity term on the dimensionless heat transfer rate at the base of the fin while figure 4 shows the effects of no-linear or thermal conductivity term on the dimensionless total heat flux of the fin. also, the figures depict the variation of the rate of heat transfer with the thermo-geometric parameter. it could be deducted that the thermal conductivity and the thermo-geometric parameter have direct and significant effects on the rate of heat transfer at the base of the fin. thus, the operational parameters must be carefully chosen to ensure that the fin retains its primary purpose of removing heat from the primary surface. journal of mechanical engineering and technology 18 issn: 2180-1053 vol. 9 no.1 january – june 2017 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0 0.5 1 1.5 2 2.5 3 3.5 d im e n s io n le s s h e a t fl u x a t th e b a s e o f th e f in , q b thermo-geometric parameter, m = -0.3 = -0.1 = 0.1 = 0.3 (a) 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 d im e n s io n le s s h e a t fl u x a t th e b a s e o f th e f in , q b thermo-geometric parameter, m = 0.6 = 0.4 = 0.2 = 0.0 (b) figure 3. effects of thermal conductivity parameter on heat transfer rate at the base of the fin thermal performance and optimum design analysis of fin with variable thermal conductivity using double decomposition method issn: 2180-1053 vol. 9 no.1 january – june 2017 19 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 0 10 20 30 40 50 60 d im e n s io n le s s t o ta l h e a t fl u x a t th e b a s e o f th e f in , q b thermo-geometric parameter, m = 0.6 = 0.4 = 0.2 = 0.0 (a) 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 0 5 10 15 20 25 30 d im e n s io n le s s t o ta l h e a t fl u x a t th e b a s e o f th e f in , q b thermo-geometric parameter, m = 0.6 = 0.4 = 0.2 = 0.0 (b) figure 4. effects of non-linear thermal conductivity parameter on the dimensionless total heat flux of the fin (a), bi=0.04 (b) bi=0.08 journal of mechanical engineering and technology 20 issn: 2180-1053 vol. 9 no.1 january – june 2017 figure 5a and 5b show the effects of non-linear thermal conductivity and thermo-geometric parameters on the fin efficiency. the figures show that the fin efficiency decreases monotonically with increasing thermogeometric parameter. also, it shows the variation of fin efficiency with thermogeometric in longitudinal convecting fin with insulated tip. from the figures, it is shown that as the thermogeometric parameter increases, the efficiency of the fin decreases. when the thermogeometric fin parameter equals to zero, the fin efficiency is 100%, which implies that there is no conduction resistance or no presence of fin at all. as the convective heat transfer coefficient to thermal conductivity ratio approaches zero, the temperature at every point in the fin is equal to the temperature of the base. the inverse variation in the fin efficiency with the thermo-geometric parameter is due to the fact that as more material is attached to the prime surface, the resistance to heat flow increases thereby reducing the fin efficiency. upon further increase in the fin thermogeometric parameter, the effect of reducing the resistance becomes visible in the sense that the fin efficiency starts to normalize. therefore, high efficiency of the fin could be achieved by using small values of thermogeometric parameter, which could be realized using a fin of small length or by using a material of better thermal conductivity. moreover, the results depicted that care must be taken in the choice of length of fin used during applications. this is because, thermogeometric parameter (which increases as the fin length increases) tends to infinity, and the fin efficiency tends to zero. the fin to a large extent of its length will remain at ambient. this consequently results in weak conduction limit. the extended area is largely useless in the heat transfer process and hence inefficient. therefore, very long fins are to be avoided in practice. 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 0.4 0.5 0.6 0.7 0.8 0.9 1  m = -0.3 = -0.1 = 0.1 = 0.3 (a) thermal performance and optimum design analysis of fin with variable thermal conductivity using double decomposition method issn: 2180-1053 vol. 9 no.1 january – june 2017 21 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1  m = -0.6 = -0.2 = 0.2 = 0.6 (b) figure 5. effects of non-linear thermal conductivity and thermo-geometric parameters on the efficiency of the fin also, as shown in the figures 5a and 5b, the fin efficiency is unity in the limit m→0. in this limit, the actual heat transfer rate from the fin is zero. this fin parameter (the thermogeometric parameter) plays a very important role in determining the amount of heat transfer from the fin as it accounts for the effects of decrease in temperature on the heat transfer from the fin. since, the fin temperature drops along the fin length, the fin efficiency decreases with increase in fin length. therefore, in practice required fin length should be properly determined because the fin length that causes the fin efficiency to drop below 60% usually cannot be justified economically and should be avoided. figures 6a and 6b show the effects of non-linear thermal conductivity and thermosgeometric parameters (under the aspect ratio of 20) on the effectiveness of the fin. as the aspect ratio increases, higher local temperature is produced in the fin, thereby increases the effectiveness of the fin. also, it is shown that high effectiveness of fin could be achieved by using small values of thermogeometric parameter and this could be realized using a fin of small length or by using a material of better thermal conductivity. journal of mechanical engineering and technology 22 issn: 2180-1053 vol. 9 no.1 january – june 2017 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 15 20 25 30 35 40  m = -0.3 = -0.1 = 0.1 = -0.3 (a) 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 15 20 25 30 35 40  m = -0.6 = -0.2 = 0.2 = -0.6 (b) figure 6. effects of non-linear thermal conductivity and thermo-geometric parameters on the effectiveness of the fin thermal performance and optimum design analysis of fin with variable thermal conductivity using double decomposition method issn: 2180-1053 vol. 9 no.1 january – june 2017 23 0 5 10 15 20 25 30 35 40 45 50 0 2 4 6 8 10 12 14 16 18 20 t h e rm o -g e o m e ti c p a ra m e te r, m aspect ratio, a bi= 0.02 bi= 0.04 bi= 0.06 bi= 0.08 (a) 0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.1 0 2 4 6 8 10 12 14 16 18 t h e rm o -g e o m e ti c p a ra m e te r, m biot number, bi aspect ratio, a= 10 aspect ratio, a= 20 aspect ratio, a= 30 aspect ratio, a= 40 (b) figure 7. effects of biot number on the thermo-geometric parameter of the fin journal of mechanical engineering and technology 24 issn: 2180-1053 vol. 9 no.1 january – june 2017 the effects of biot number and aspect ratio on the thermo-geometric parameter (the fin performance factor) are shown in figure 7. the fin performance factor increases as the aspect ratio and biot number increase. however, the thermal performance or efficiency of the fin is favoured at low values of thermogeometric parameter since the aim (high effective use of the fin) is to minimize the temperature decrease along the fin length, where the best possible scenario is when t=tb everywhere. it must be pointed out that equation (42) shows the direct relationship between the thermogeometric parameter, m and the biot number, bi which directly depends on the fin length. a small value of m corresponds to a relatively short and thick fin of poor thermal conductivity and a high value of m implies a long fin or fin with low value of thermal conductivity. since, the thermal performance or efficiency of the fin is favoured at low values of thermogeometric fin parameter, very long fins are to be avoided in practice. a compromise is reached for one-dimensional analysis of fins 0 < bi <0.1. when the biot number is greater than 0.1, two dimensional analysis of the fin is recommended as one-dimensional analysis predicts unreliable results for such limit. figure 8a shows the nondimensional heat transfer q/ζ (for a unit fin volume) varying with m from 1 and 2 for specified values of non-linear thermal conductivity terms, β, under a given profile area, ap, the heat transfer first rises and then falls as the fin length increases. also, as the optimum fin length (at which q/ζ reaches a maximum value) increases as the non-linear thermal conductivity term, β increases. it also shows that the optimum value of m can be obtained based upon the value of non-linear term. therefore, from the analysis the optimum dimensions of the convection fin with variable thermal conductivity is established and the relative values of optimum m and β are shown in figure 8b. the approximate analytical method of solution was validated by the exact solution in figure 9a and 9b for the linear thermal model of the fin problem and the non-linear problem was validated with numerical solution as shown in figures 10a and 10b. it is depicted that the double decomposition method is highly accurate and agrees very well with the exact solution for the linear problem and also with the numerical solution for non-linear problem. thermal performance and optimum design analysis of fin with variable thermal conductivity using double decomposition method issn: 2180-1053 vol. 9 no.1 january – june 2017 25 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 q f /  m = -0.6 = -0.2 = 0.2 = -0.6 (a) -0.6 -0.4 -0.2 0 0.2 0.4 0.6 1.1 1.2 1.3 1.4 1.5 1.6 1.7 m o p ti m u m  (b) figure 8 (a) effects of non-linear thermal conductivity and thermo-geometric parameters on the dimensionless heat transfer, qf/ζ (b) effects of non-linear thermal conductivity parameter on the optimum thermo-geometric parameter journal of mechanical engineering and technology 26 issn: 2180-1053 vol. 9 no.1 january – june 2017 6.1 validation of results and thermal stability analysis of the fin 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0.88 0.9 0.92 0.94 0.96 0.98 1  x exact solution ddm solution (a) 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0.97 0.975 0.98 0.985 0.99 0.995 1  x exact solution ddm solution (b) figure 9. validation of the results when (a) m=0.25, β=0 (a) m=0.5, β=0 thermal performance and optimum design analysis of fin with variable thermal conductivity using double decomposition method issn: 2180-1053 vol. 9 no.1 january – june 2017 27 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0.84 0.86 0.88 0.9 0.92 0.94 0.96 0.98 1  x ddm solution numerical solution (a) 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0.7 0.75 0.8 0.85 0.9 0.95 1  x ddm solution numerical solution (b) figure 10. validation of the results when (a) m=0.5, β=-0.3 (b) m=1.0, β= 0.4 journal of mechanical engineering and technology 28 issn: 2180-1053 vol. 9 no.1 january – june 2017 7.0 conclusions in this paper, thermal performance study and optimum design analysis of straight fin with variable thermal conductivity have been carried out using double decomposition method. the analysis revealed that the fin temperature distribution, the total heat transfer, the fin effectiveness, and the fin efficiency are significantly affected by the thermo-geometric and thermal parameters of the fin. also, it is established that the optimum fin length increases as the non-linear thermal conductivity term, increases. it also shows that the optimum value of m can be obtained based upon the value of non-linear term.therefore, the operational parameters must be carefully chosen to ensure that the fin retains its primary purpose of removing heat from the primary surface. the results obtained in this analysis provides platform for improvement in the design of fin in heat transfer equipment. nomenclature ar aspect ratio ac cross sectional area of the fins ap profile area of the fins b lenght of the fin bi biot number h heat transfer coefficient k thermal conductivity of the fin material ka thermal conductivity of the fin material at ambient temperature kb thermal conductivity of the fin material at the base temperature of the fin k dimensionless thermal conductivity of the fin material m dimensionless thermo-geometric fin parameter m2 thermo-geometric fin parameter p perimeter of the fin t temperature t∞ ambient temperature tb temperature at the base of the fin x fin axial distance, m x dimensionless length of the fin q rate of heat transfer qf dimensionless heat transfer greek symbols β thermal conductivity parameter or non-linear parameter δ thickness of the fin, m θ dimensionless temperature θb dimensionless temperature at the base of the fin η efficiency of the fin ε effectiveness of the fin thermal performance and optimum design analysis of fin with variable thermal conductivity using double decomposition method issn: 2180-1053 vol. 9 no.1 january – june 2017 29 references adomian g and rach r (1993). analytic solution of nonlinear boundary-value problems in several dimensions by decomposition. journal of mathematical analysis and applications. 174, 118–37. atay, m. t. and coskun, s. b. (2008). comparative analysis of power-law fin-type problems using variational iteration method and finite element method, mathematical problems in engineering. 1–9. arslanturk, a. (2005). a decomposition method for fin efficiency of convective straight fin with temperature dependent thermal conductivity, international communication in heat and mass transfer. 32, 831–841. aziz, a and enamul-huq, s. m. (1973). perturbation solution for convecting fin with temperature dependent thermal conductivity, journal of heat transfer. 97, 300 – 301. aziz, a. and bouaziz, m. n. (2011). a least squares method for a longitudinal fin with temperature dependent internal heat generation and thermal conductivity, energy conversion and management. 52, 2876–2882. aziz, a (1977). perturbation solution for convecting fin with internal heat generation and temperature dependent thermal conductivity, international journal of heat and mass transfer. 20, 1253-1255. campo, a. and spaulding, r. j. (1999). “coupling of the methods of successive approximations and undetermined coefficients for the prediction of the thermal behaviour of uniform circumferential fins,” heat and mass transfer, 34(6), 461– 468, ching-huang chiu, cha’o-kuang chen (2002). a decomposition method for solving the convectice longitudinal fins with variable thermal conductivity, international journal of heat and mass transfer 45, 2067-2075. chiu, c. h. and chen, c. k. (2003). application of adomian decomposition procedure to analysis of convective-radiative fins, journal of heat transfer. 125, 312–316. chowdhury m. s. h., hashim i., abdulaziz o (2009).comparison of homotopy analysis method and homotopy-permutation method for purely nonlinear fin-type problems, communications in nonlinear science and numerical simulation 14, 371-378. journal of mechanical engineering and technology 30 issn: 2180-1053 vol. 9 no.1 january – june 2017 coskun, s. b. and atay m. t. (2007). analysis of convective straight and radial fins with temperature dependent thermal conductivity using variational iteration method with comparison with respect to finite element analysis. mathematical problem in engineering. 1–5 coskun, s. b. and atay, m. t. (2008). fin efficiency analysis of convective straight fin with temperature dependent thermal conductivity using variational iteration method, applied thermal engineering. 28, 2345–2352. chowdhury m. s. h., hashim i. (2008). analytical solutions to heat transfer equations by homotopy-perturbation method revisited, physical letters a. 372, 1240–1243. domairry, g and fazeli, m. (2009) homotopy analysis method to determine the fin efficiency of convective straight fins with temperature dependent thermal conductivity. communication in nonlinear science and numerical simulation. 14, 489–499. fernandez, a. (2009). on some approximate methods for nonlinear models. applied mathematics and computation. 215, 168–174. ganji, d. d. (2006). the application of he’s homotopy perturbation method to nonlinear equations arising in heat transfer, physical letters a. 355,337–341. ganji d. d. and a. s. dogonchi (2014). analytical investigation of convective heat transfer of a longitudinal fin with temperature-dependent thermal conductivity, heat transfer coefficient and heat generation, international journal of physical sciences. 9(21), 466-474. ghasemi, s. e., hatami, m. and ganji, d. d. (2014). thermal analysis of convective fin with temperature-dependent thermal conductivity and heat generation, cases studies in thermal engineering. 4, 1-8. he, j. h. (1999). homotopy perturbation method, computer methods in applied mechanics and engineering. 178, 257–262. joneidi, a.a., ganji, d.d., babaelahi, m. (2009) differential transformation method to determine fin efficiency of convective straight fins with temperature dependent thermal conductivity. international communication in heat and mass transfer. 36,757-762 khani f., ahmadzadeh raji m. and hamedi nejad h. (2009). analytical solutions and efficiency of the nonlinear fin problem with temperature-dependent thermal conductivity and heat transfer coefficient, communications in nonlinear science and numerical simulation. 14, 3327–3338. thermal performance and optimum design analysis of fin with variable thermal conductivity using double decomposition method issn: 2180-1053 vol. 9 no.1 january – june 2017 31 khani, f. and aziz, a. (2010). thermal analysis of a longitudinal trapezoidal fin with temperature dependent thermal conductivity and heat transfer coefficient, communications in nonlinear science and numerical simulation. 15, 590–601. k. hosseini, b. daneshian, n. amanifard, r. ansari. (2012). homotopy analysis method for a fin with temperature dependent internal heat generation and thermal conductivity. international journal of nonlinear science. 14(2), 201–210. languri, e. m., ganji, d.d, jamshidi, n. (2008). variational iteration and homotopy perturbation methods for fin efficiency of convective straight fins with temperature dependent thermal conductivity. the 5th wseas international conference on fluid mechanics (fluids 08) acapulco, mexico january 25–27. m.a. golberg. (1999). a note on the decomposition method for operator equations, applied mathematics and computation. 106 (2–3) 215–220. moradi. a and ahmadikia, h. (2010). analytical solution for different profiles of fin with temperature dependent thermal conductivity. mathematical problem in engineering. 2010, article id 568263, 15. moradi, a and ahmadikia, h (2011). investigation of effect thermal conductivity on straight fin performance with dtm, international journal of engineering and applied sciences.1, 42–54 mosayebidorcheh s., ganji d. d., masoud farzinpoor (2014). approximate solution of the nonlinear heat transfer equation of a fin with the power-law temperature-dependent thermal conductivity and heat transfer coefficient, propulsion and power research. 41–47. moitheki, r.j. hayat, t. and malik, m.y. (2010). some exact solutions of the fin problem with a power law temperature dependent thermal conductivity. nonlinear analysis: real world application. 11, 3287–3294. mustafa inc. (2008). application of homotopy analysis method for fin efficiency of convective straight fin with temperature dependent thermal conductivity. mathematics and computers simulation. 79, 189 – 200. ndlovu, p. l. and moitsheki , j. r. (2013). analytical solutions for steady heat transfer in longitudinal fins with temperature-dependent properties, mathematical problems in engineering. 2013, 1–14. journal of mechanical engineering and technology 32 issn: 2180-1053 vol. 9 no.1 january – june 2017 rajabi. a. (2007). homotopy perturbation method for fin efficiency of convective straight fins with temperature dependent thermal conductivity. physical letters a. 364, 33– 37 sadri, s.; raveshi, m. r., and amiri, s. (2012). efficiency analysis of straight fin with variable heat transfer coefficient and thermal conductivity. journal of mechanical science and technology. 26(4), 1283–1290. yue-tzu yang, shih-kai chien and cha’o-kuang chen (2008). a double decomposition method for solving the periodic base temperature in convective longitudinal fins. energy conversion and management. 49, 2910–2916 yue-tzu yang, shih-kai chien and cha’o-kuang chen (2010). a double decomposition method for solving the annular hyperbolic profile fins with variable thermal conductivity. heat transfer engineering. 31, 1165–1172. foundry metallurgy of tungsten carbide and aluminium silicate particulate reinforced lm6 alloy hybrid composites issn: 2180-1053 vol. 3 no. 1 january-june 2011 55 foundry metallurgy of tungsten carbide and aluminium silicate particulate reinforced lm6 alloy hybrid composites thoguluva raghavan vijayaram1 1faculty of manufacturing engineering, universiti teknikal malaysia melaka, hang tuah jaya, 76100 durian tunggal, melaka, malaysia email: 1thoguluva@utem.edu.my abstract hybrid composites are advanced composite materials in which a combination of two or more second phase particulates or fibers are reinforced in a base matrix. in this research paper, liquid metallurgical processing of a new hybrid composite material containing tungsten carbide particulate and aluminium silicate particulate combined at different weight fraction percentage is discussed. manufacturing of such combined tungsten carbide particulate and aluminium silicate particulate reinforced aluminium-11.8% silicon alloy matrix composites by metal casting technology has some advantages of processing the composites by near net shape techniques. turbulence generated by the liquid metallurgical vortex mixing technique is the easiest technique for processing these hybrid composites. aluminium-11.8% silicon alloy hybrid particulate, combined tungsten carbide and aluminium silicate reinforced composites is related to their higher strength, lightweight, hardness, higher temperature resistance and wear resistance than that of any conventional monolithic materials. in this experimental work, aluminum-silicon alloy composites containing tungsten carbide and aluminium silicate combined particulate combination of 2.5%, 5.0%, 7.5% and 10.0% on weight fraction basis are produced by using the liquid stirring method. the size of the tungsten carbide particulate is 47.30 micron supplied by aldrich, usa and the size of the aluminium silicate particulate is equal to 157.10 micron supplied by fluka, usa. this paper discusses the vortex stirring process to produce these hybrid composite castings. these are processed in the form of slab containing 2.5%, 5%, 7.5% and 10% weight fraction of the two combination of particulate equally reinforced in lm6 alloy. a grey cast iron metallic mold is used to pour the hybrid composite slurry mixture. mechanical, electrical and thermal properties are determined and all these aspects are discussed in this paper. the microstructures of the processed hybrid composites are studied at different magnifications and photomicrographs are captured to identify the presence of the two different reinforced particulates and its distribution uniformity. fracture surface analysis has been performed to study the failure mechanisms. issn: 2180-1053 vol. 3 no. 1 january-june 2011 journal of mechanical engineering and technology 56 keywords: liquid-stirring technique, aluminium-11.8% silicon alloy, metallography, fracture surface, fractography. 1.0 introduction hybrid composites are usually used when a combination of properties of different types of particulates wants to be achieved, or when longitudinal as well as lateral mechanical performances are required [1]. aluminium oxide and silicon carbide reinforced aluminum alloy matrix composites are applied in the automotive and aircraft industries as engine pistons and cylinder heads, where the tribological properties of these materials are considered important. therefore, the development of aluminum matrix composites is receiving considerable emphasis in meeting the requirements of various industries. it has also reduced the thermal insulation requirements because of its lower thermal conductivity. precision components in the missile guidance systems demands dimensional stability and the geometries of the components cannot change during usage [2]. 2.0 literature review today’s search is for composite materials with ever-higher service characteristics such as wear and heat resistance, hightemperature strength, antifriction, and cutting properties [3]. different shapes of particulates are reinforced in the matrix alloy and they are characterized as acicular, irregular rod like, flake, dendritic, spherical, rounded, irregular, angular, sub angular, fibrous, granular, lamellar, nodular, crystalline and porous type. particle shape has a major influence on processing characteristics [4, 5]. the shape is usually described in terms of the aspect ratio or shape factor [2]. aspect ratio is the ratio of the largest dimension to the smallest dimension of the particle. this ratio ranges from unity, for a spherical particle to about 10 for flake like or needlelike particles. shape factor or shape index of the particulate is a measure of the ratio of the surface area of the particle to its volume, normalized by reference to a spherical particle of equivalent volume [2]. the size distribution of particulates is an important consideration, because its affects the processing characteristics of the powder. normally the fibers have a definite aspect ratio, which is defined by length/diameter ratio. composite materials have found increasingly wider applications in aircraft, space vehicles, offshore structures, piping, electronic, automobiles, boats and sporting good. foundry metallurgy of tungsten carbide and aluminium silicate particulate reinforced lm6 alloy hybrid composites issn: 2180-1053 vol. 3 no. 1 january-june 2011 57 on a weight adjusted basis, many aluminium and aluminium alloy based composite materials can outperform the conventional ferrous and non-ferrous materials like cast iron, steel, aluminum, magnesium and virtually any other reinforced metal or alloy in a wide variety of applications [3]. hence, probably, metal matrix composites will replace the conventional materials in many commercial and industrial applications in the near future. special interest on particulate-reinforced metal matrix composites is due to the several merits offered by them. researchers have reported that some composites exhibit a higher compressive strength than tensile strength due to the dominance of the presence of the reinforcement phase since the compressive strength of the reinforcement particulate is very high. in the processing of metal matrix composite, one of the subjects of interest is to choose a suitable matrix and a reinforcement material. it influences the mechanical properties, shear modulus and shear strength and its processing characteristics. reinforcement phase is the principal load-carrying member in a composite. therefore, the orientation, of the reinforcement phase decides the properties of the composite. the reinforcing phase may be a particulate or a fiber, continuous type or discontinuous type. the size of the particulate is expressed in microns, micrometer. but the discontinuous fiber is defined by a term called as ‘aspect ratio’. it is expressed as the ratio of length to the diameter of the fiber. to improve the wettabilty with the liquid alloy or metal matrix material, the reinforcement phase is always preheated [4]. the interface between the matrix and the reinforcement plays an important role for deciding and explaining the toughening mechanism in the metal matrix composites. the interface between the matrix and the reinforcement should be organized in such a way that the bond in between the interface and the matrix should not be either strong or weak. while the load is acting on the composite, it has been distributed to the matrix and the reinforcement phase through the matrix interface. the reinforcement is effective in strengthening the matrix only if a strong interfacial bond exists between them [5]. the interfacial properties also influence the resistance to crack propagation in a composite and therefore its fracture toughness. the two most important energy-absorbing failure mechanisms in a composite are debonding and particle pull out at the particle matrix interface [3]. if the interface between the matrix and reinforcement issn: 2180-1053 vol. 3 no. 1 january-june 2011 journal of mechanical engineering and technology 58 debonds, then the crack propagation is interrupted by the debonding process and instead of moving through the particle, the crack move along the particle surface allowing the particle to carry higher load. particles can be spherical, disk-shaped, rod shaped, and plate shaped [5]. 3.0 materials and methods in this work, hybrid composite slab castings of dimension 190 mm* 30mm*20mm and weighing approximately 302.10 grams are produced. the processed hybrid composite castings are subjected to various testing to assess its mechanical properties, hardness, impact resistance and microstructure. then electrical and thermal properties are determined and metallographic analysis has been performed to study the tungsten carbide particulate and aluminium silicate particulate distribution uniformity. besides, fracture surface analysis has been conducted to study the failure and interfacial reaction characteristics. permanent grey cast iron metallic mold is used to pour 2.5%, 5%, 7.5% and 10% weight fraction of combined tungsten carbide particulate and aluminium silicate particulate reinforced lm6 alloy hybrid composite slurry mixture to make hybrid composite slab castings. the mold material used in this process is grey cast iron and it is reused to process more number of test castings. testing for mechanical properties, thermal properties and electrical properties of the newly developed combined tungsten carbide particulate and aluminium silicate particulate reinforced aluminium-11.8percentage silicon alloy matrix hybrid composites has been performed by using standard equipments and hence data are generated. electrical resistivity and electrical conductivity values of the processed hybrid composites are calculated by finding the resistance in between the hybrid composite slab casting material. an indigenously designed circuit is used to determine the resistance value and hence the electrical conductivity and resistivity values are determined. thermal diffusivity of the processed hybrid composites are measured by photo flash method and hence thermal conductivity values are determined. metallographic analysis has been performed with the aid of a metallurgical microscope to study the tungsten carbide and aluminium silicate particulates distribution uniformity and hence to characterize the phases present in the processed new type of hybrid composites. lastly, fracture surface analysis has been conducted to study the failure of composites and interfacial reaction characteristics [6]. foundry metallurgy of tungsten carbide and aluminium silicate particulate reinforced lm6 alloy hybrid composites issn: 2180-1053 vol. 3 no. 1 january-june 2011 59 4.0 properties determinatiaon, microstructural characterization, and fracture surface studies testing for mechanical properties, thermal properties and electrical properties of the newly developed combined tungsten carbide particulate and aluminium silicate particulate reinforced aluminium11.8percentage silicon alloy matrix hybrid composites has been performed by using standard equipments and hence data are generated. tensile specimens are carried out by using an instron universal testing machine 8500 to determine the tensile properties of the material such as tensile strength, yield stress, fracture stress, young’s modulus, percentage of elongation, ductility, specific strength and specific stiffness. mitutoyo atk-600 model hardness testing machine is used to determine the hardness values of the hybrid composites expressed in rockwell superficial scale with 15 n-s by applying a load of 100kg. in this innovative project, impact charpy values of the machined samples from the processed hybrid composite castings are determined from the reading taken in an impact testing machine. electrical resistivity and electrical conductivity values of the processed hybrid composites are calculated by finding the resistance in between the hybrid composite slab casting material. an indigenously designed circuit is used to determine the resistance value and hence the electrical conductivity and resistivity values are determined. thermal diffusivity of the processed hybrid composites are measured by photo flash method and hence thermal conductivity values are determined. metallographic analysis has been performed with the aid of a metallurgical microscope to study the tungsten carbide and aluminium silicate particulates distribution uniformity and hence to characterize the phases present in the processed new type of hybrid composites. lastly, fracture surface analysis has been conducted to study the failure of composites and interfacial reaction characteristics [6]. 5.0 results and discussions results and data are obtained from the tested samples taken from the combined tungsten carbide and aluminium silicate particulate reinforced lm6 alloy slab composite castings made in grey cast iron mold. the values are reported for the mechanical, thermal and electrical properties as well as the density, hardness, impact strength and microstructural features of the tungsten carbide and aluminium silicate particulate distribution for each weight fraction percentage addition to the lm6 alloy matrix. in this section, the above composites issn: 2180-1053 vol. 3 no. 1 january-june 2011 journal of mechanical engineering and technology 60 made in the grey cast iron are analyzed and the results are presented. it is found that the two combined tungsten carbide particulate and aluminium silicate particulate reinforced lm6 alloy matrix hybrid composite casting properties are superior to the lm6 alloy with and without grain refiner addition and no particulate reinforcement. in this innovative hybrid composite material development research work, the two combined particulates are reinforced in the alloy matrix are processed by liquid vortex metallurgical melt stirring technique. microstructures of the processed hybrid composites based on the metallographic studies have confirmed the uniformity of tungsten carbide particulate and aluminium silicate particulate distribution in the aluminium-11.8% silicon alloy matrix. sufficient amount of turbulence during the mixing of the particulates with the liquid alloy is necessary to get uniform particulate distribution during its solidification processing. the impeller blade type and its rotational speed have not shown any effect on the distribution uniformity. but, faster pouring of the hybrid composite slurry mixture into the grey cast iron mold immediately after the mixing by vortex method has played a significant role in the distribution of the particulates used in this project. a small amount of tungsten particulate segregation has been observed due to its higher density. but, due to combination with aluminium silicate particulate, the effect of segregation is protected. in this section, the above composites made in the grey cast iron are analyzed and the results are presented in the corresponding graphs that are shown in figure 1 to figure 7. 5.1 mechanical properties the tensile strength, yield stress and fracture stress values of combined tungsten carbide and aluminium silicate particulate composites are determined. the tensile strength of 10% and 2.5% weight fraction of the above combined particulate composite is 89.16 mpa and 173.63 mpa respectively. from this, it is clear that the tensile strength value decreases with the increase on the weight fraction % of combined tungsten carbide and aluminium silicate in the alloy matrix. the tensile strength value decreased gradually when above combination of particulate weight fraction addition of lm6 alloy matrix is increased and it is shown in the graph as figure 1. foundry metallurgy of tungsten carbide and aluminium silicate particulate reinforced lm6 alloy hybrid composites issn: 2180-1053 vol. 3 no. 1 january-june 2011 61 50 tensile strength value decreased gradually when above combination of particulate weight fraction addition of lm6 alloy matrix is increased and it is shown in the graph as figure 1. tensile strength vs weight fraction % of tungsten carbide and aluminium silicate (grey cast iron mold) 0 50 100 150 200 0.00% 5.00% 10.00% 15.00% weight fraction % of tungsten carbide and aluminium silicate (grams) t en si le s tr en gt h (m p a) figure 1 tensile strength vs weight fraction % of combined tungsten carbide and aluminium silicate specific strength vs weight fraction % of tungsten carbide and aluminium silicate (grey cast iron mold) 0 2 4 6 8 0.00% 5.00% 10.00% 15.00% weight fraction % of tungsten carbide and aluminm silicate (grams) s pe ci fic s tr en gt h (m *1 00 0) figure 2 specific strength vs weight fraction percentage of combined tungsten carbide and aluminium silicate from the graph shown in figure 2, it is observed that the values of specific strength gradually increase with an increased addition of combined particulate up to the addition of 7.5% weight fraction, after which the values start decreasing. therefore, it is understood that the optimum value of adding the combined particulate to the alloy matrix is 7.5% by weight fraction. data on the hardness of combined particulate reinforced composites made in grey cast iron mold is analyzed. it is found that the hardness value increases gradually with the increased addition % by weight and it is shown below in the graph as figure 3. the maximum hardness value based rockwell superficial 15n-s scale is 67.13 for 10% weight fraction addition. figure 1 tensile strength vs weight fraction % of combined tungsten carbide and aluminium silicate 50 tensile strength value decreased gradually when above combination of particulate weight fraction addition of lm6 alloy matrix is increased and it is shown in the graph as figure 1. tensile strength vs weight fraction % of tungsten carbide and aluminium silicate (grey cast iron mold) 0 50 100 150 200 0.00% 5.00% 10.00% 15.00% weight fraction % of tungsten carbide and aluminium silicate (grams) t en si le s tr en gt h (m p a) figure 1 tensile strength vs weight fraction % of combined tungsten carbide and aluminium silicate specific strength vs weight fraction % of tungsten carbide and aluminium silicate (grey cast iron mold) 0 2 4 6 8 0.00% 5.00% 10.00% 15.00% weight fraction % of tungsten carbide and aluminm silicate (grams) s pe ci fic s tr en gt h (m *1 00 0) figure 2 specific strength vs weight fraction percentage of combined tungsten carbide and aluminium silicate from the graph shown in figure 2, it is observed that the values of specific strength gradually increase with an increased addition of combined particulate up to the addition of 7.5% weight fraction, after which the values start decreasing. therefore, it is understood that the optimum value of adding the combined particulate to the alloy matrix is 7.5% by weight fraction. data on the hardness of combined particulate reinforced composites made in grey cast iron mold is analyzed. it is found that the hardness value increases gradually with the increased addition % by weight and it is shown below in the graph as figure 3. the maximum hardness value based rockwell superficial 15n-s scale is 67.13 for 10% weight fraction addition. figure 2 specific strength vs weight fraction percentage of combined tungsten carbide and aluminium silicate from the graph shown in figure 2, it is observed that the values of specific strength gradually increase with an increased addition of combined particulate up to the addition of 7.5% weight fraction, after which the values start decreasing. therefore, it is understood that the optimum value of adding the combined particulate to the alloy matrix is 7.5% by weight fraction. data on the hardness of combined particulate reinforced composites made in grey cast iron mold is analyzed. it is found that the hardness value increases gradually with the increased addition % by weight and it is shown below in the graph as figure 3. the maximum hardness value based rockwell superficial 15n-s scale is 67.13 for 10% weight fraction addition. issn: 2180-1053 vol. 3 no. 1 january-june 2011 journal of mechanical engineering and technology 62 51 hardness vs weight f raction % of tungsten carbide and aluminium silicate (grey cast iron mold) 0 20 40 60 80 0.00% 5.00% 10.00% 15.00% weight f raction % of tungsten carbide and aluminium silicate (grams) h ar dn es s r oc kw el l s up er fic ia l 1 5n -s figure 3 hardness vs weight fraction percentage of combined tungsten carbide and aluminium silicate the density values of the combined particulate composite castings poured grey cast iron mold is observed from the data. based on the plotted graph as shown in figure 4, it is found that there is no remarkable variation or changes observed in the determined density values of the processed combined particulate hybrid composite casting from the grey cast iron mold. density vs weight fraction % of tungsten carbide and aluminium silicate (grey cast iron mold) 2.3 2.4 2.5 2.6 2.7 0.00% 5.00% 10.00% 15.00% weight fraction % of tungsten carbide and aluminium silicate (grams) d en si ty ( gm s/ cc ) figure 4 density vs weight fraction % of combined tungsten carbide and aluminium silicate impact strength values of the hybrid composite castings processed in grey cast iron mold is determined. from the analysis and plotted graphs, it is found that the impact strength values are gradually increasing up to a certain extent and then starts decreasing sharply, with the increasing addition of the two different particulates in the alloy matrix. based on the plotted graph as shown in figure 5, it is found that the optimum amount of tungsten carbide and aluminium silicate to be added is nearly 5 % weight fraction and further addition is not effective in the impact strength improvement due to the reinforcement. the optimum value of impact strength is 8.838 n-m for 5% weight fraction addition of combined particulate to the alloy matrix. figure 3 hardness vs weight fraction percentage of combined tungsten carbide and aluminium silicate the density values of the combined particulate composite castings poured grey cast iron mold is observed from the data. based on the plotted graph as shown in figure 4, it is found that there is no remarkable variation or changes observed in the determined density values of the processed combined particulate hybrid composite casting from the grey cast iron mold. 51 hardness vs weight f raction % of tungsten carbide and aluminium silicate (grey cast iron mold) 0 20 40 60 80 0.00% 5.00% 10.00% 15.00% weight f raction % of tungsten carbide and aluminium silicate (grams) h ar dn es s r oc kw el l s up er fic ia l 1 5n -s figure 3 hardness vs weight fraction percentage of combined tungsten carbide and aluminium silicate the density values of the combined particulate composite castings poured grey cast iron mold is observed from the data. based on the plotted graph as shown in figure 4, it is found that there is no remarkable variation or changes observed in the determined density values of the processed combined particulate hybrid composite casting from the grey cast iron mold. density vs weight fraction % of tungsten carbide and aluminium silicate (grey cast iron mold) 2.3 2.4 2.5 2.6 2.7 0.00% 5.00% 10.00% 15.00% weight fraction % of tungsten carbide and aluminium silicate (grams) d en si ty ( gm s/ cc ) figure 4 density vs weight fraction % of combined tungsten carbide and aluminium silicate impact strength values of the hybrid composite castings processed in grey cast iron mold is determined. from the analysis and plotted graphs, it is found that the impact strength values are gradually increasing up to a certain extent and then starts decreasing sharply, with the increasing addition of the two different particulates in the alloy matrix. based on the plotted graph as shown in figure 5, it is found that the optimum amount of tungsten carbide and aluminium silicate to be added is nearly 5 % weight fraction and further addition is not effective in the impact strength improvement due to the reinforcement. the optimum value of impact strength is 8.838 n-m for 5% weight fraction addition of combined particulate to the alloy matrix. figure 4 density vs weight fraction % of combined tungsten carbide and aluminium silicate impact strength values of the hybrid composite castings processed in grey cast iron mold is determined. from the analysis and plotted graphs, it is found that the impact strength values are gradually increasing up to a certain extent and then starts decreasing sharply, with the increasing addition of the two different particulates in the alloy matrix. based on the plotted graph as shown in figure 5, it is found that the optimum amount of tungsten carbide and aluminium silicate to be added is nearly 5 % weight fraction and further addition is not effective in the impact strength improvement due to the reinforcement. the optimum value of impact strength is 8.838 n-m for 5% weight fraction addition of combined particulate to the alloy matrix. foundry metallurgy of tungsten carbide and aluminium silicate particulate reinforced lm6 alloy hybrid composites issn: 2180-1053 vol. 3 no. 1 january-june 2011 63 52 impact strength vs weight f raction % of tungsten carbide and aluminium silicate (grey cast iron mold) 0 5 10 0.00% 5.00% 10.00% 15.00% weight f raction % of tungsten carbide and aluminium silicate (grams) im pa ct s tr en gt h (n -m ) figure 5 impact strength vs weight fraction % of combined tungsten carbide and aluminium silicate 5.2 electrical properties the processed hybrid composite castings made in grey cast iron metal mold are tested for electrical properties. graph is plotted between the weight fraction % addition of combined particulate and electrical resistivity and it is shown in figure 6. from the analysis, it is found that the electrical resistivities of the hybrid composites are decreased with the increased addition in the alloy matrix. electrical resistivity vs weight f raction % of tungsten carbide and aluminium silicate (grey cast iron mold) 0.75 0.8 0.85 0.9 0.00% 5.00% 10.00% 15.00% weight f raction % of tungsten carbide and aluminium silicate (grams) e le ct ric al r es is tiv ity (o hm -m ) figure 6 electrical resistivity vs weight fraction % of combined tungsten carbide and aluminium silicate the average electrical resistivity value for 10 percentage weight fraction of combined particulate addition to the alloy matrix is 0.825765 ohm-m and the electrical conductivity value is 1.2124911/ (ohm-m). 5.3 thermal properties the processed combined hybrid composite castings made in grey cast iron mold are tested for thermal properties. graph is plotted for the weight fraction % addition of combined tungsten carbide and aluminium silicate particulate between thermal conductivity values. from the analysis, it is found that the thermal conductivity of the combined composites is decreased with the increased addition of combined particulate in the alloy matrix. the data for thermal conductivity of grey cast iron mold is analyzed and this is illustrated in the plotted graph as shown in figure 7.the thermal conductivity values for 10% weight fraction addition is 21.035 w/ m k respectively. figure 5 impact strength vs weight fraction % of combined tungsten carbide and aluminium silicate 5.2 electrical properties the processed hybrid composite castings made in grey cast iron metal mold are tested for electrical properties. graph is plotted between the weight fraction % addition of combined particulate and electrical resistivity and it is shown in figure 6. from the analysis, it is found that the electrical resistivities of the hybrid composites are decreased with the increased addition in the alloy matrix. 52 impact strength vs weight f raction % of tungsten carbide and aluminium silicate (grey cast iron mold) 0 5 10 0.00% 5.00% 10.00% 15.00% weight f raction % of tungsten carbide and aluminium silicate (grams) im pa ct s tr en gt h (n -m ) figure 5 impact strength vs weight fraction % of combined tungsten carbide and aluminium silicate 5.2 electrical properties the processed hybrid composite castings made in grey cast iron metal mold are tested for electrical properties. graph is plotted between the weight fraction % addition of combined particulate and electrical resistivity and it is shown in figure 6. from the analysis, it is found that the electrical resistivities of the hybrid composites are decreased with the increased addition in the alloy matrix. electrical resistivity vs weight f raction % of tungsten carbide and aluminium silicate (grey cast iron mold) 0.75 0.8 0.85 0.9 0.00% 5.00% 10.00% 15.00% weight f raction % of tungsten carbide and aluminium silicate (grams) e le ct ric al r es is tiv ity (o hm -m ) figure 6 electrical resistivity vs weight fraction % of combined tungsten carbide and aluminium silicate the average electrical resistivity value for 10 percentage weight fraction of combined particulate addition to the alloy matrix is 0.825765 ohm-m and the electrical conductivity value is 1.2124911/ (ohm-m). 5.3 thermal properties the processed combined hybrid composite castings made in grey cast iron mold are tested for thermal properties. graph is plotted for the weight fraction % addition of combined tungsten carbide and aluminium silicate particulate between thermal conductivity values. from the analysis, it is found that the thermal conductivity of the combined composites is decreased with the increased addition of combined particulate in the alloy matrix. the data for thermal conductivity of grey cast iron mold is analyzed and this is illustrated in the plotted graph as shown in figure 7.the thermal conductivity values for 10% weight fraction addition is 21.035 w/ m k respectively. figure 6 electrical resistivity vs weight fraction % of combined tungsten carbide and aluminium silicate the average electrical resistivity value for 10 percentage weight fraction of combined particulate addition to the alloy matrix is 0.825765 ohm-m and the electrical conductivity value is 1.2124911/ (ohm-m). 5.3 thermal properties the processed combined hybrid composite castings made in grey cast iron mold are tested for thermal properties. graph is plotted for the weight fraction % addition of combined tungsten carbide and aluminium silicate particulate between thermal conductivity values. from the analysis, it is found that the thermal conductivity of the combined composites is decreased with the increased addition of combined particulate in the alloy matrix. the data for thermal conductivity of grey cast iron mold is analyzed and this is illustrated issn: 2180-1053 vol. 3 no. 1 january-june 2011 journal of mechanical engineering and technology 64 in the plotted graph as shown in figure 7.the thermal conductivity values for 10% weight fraction addition is 21.035 w/ m k respectively. 53 thermal conductivity vs weight fraction % of tungsten carbide (grey cast iron mold) 0 20 40 60 80 0.00% 5.00% 10.00% 15.00% weight fraction % of tungsten carbide and aluminium silicate (grams) t he rm al c on du ct iv ity [1 /( oh m -m )] figure 7 thermal conductivity vs weight fraction percentage of combined tungsten carbide and aluminium silicate 5.4 metallography of combined tungsten carbide and aluminium silicate particulate reinforced aluminium-11.8% silicon alloy hybrid composites microstructural observation at different magnifications of the processed combined tungsten carbide and aluminium silicate particulate reinforced lm6 alloy composite test specimens made in grey cast iron mold are analyzed by a metallurgical microscope and hence it is employed to obtain some qualitative evidences on the combined tungsten carbide and aluminium silicate particulate distribution in the alloy matrix and bonding quality between the two particulates and the matrix. metallographic samples of the combined composites are prepared under the standard procedures and hf, hydrofluoric acid is used as an etchant to reveal the phases present in the lm6 alloy matrix. the samples are viewed at different magnifications such as at 50x, and 100x and photomicrographs are captured to predict the confirmation of the presence of the two particulates in the alloy matrix. then, it is further studied to identify the particulate distribution. from the in-depth research on this, it is confirmed the presence and distribution of embedded two particulates in the matrix is uniform. the alloy matrix grains are finer and the bonding between particulate surface and the matrix material is satisfactory. it is found that, the morphological distribution of combined particulate for every weight fraction % addition increases. no interfacial reaction products are observed superficially. from this analysis, it is confirmed that the two different particulates reinforced lm6 alloy hybrid composite casting properties are superior to the lm6 alloy with and without grain refiner addition and no particulate reinforcement. in this section, a number of captured photomicrographs are shown in the figure 8 to figure 10 for better understanding. figure 8 2.5 % tungsten carbide and aluminium silicate mixed particulate hybrid composite magnified at 50x figure 7 thermal conductivity vs weight fraction percentage of combined tungsten carbide and aluminium silicate 5.4 metallography of combined tungsten carbide and aluminium silicate particulate reinforced aluminium-11.8% silicon alloy hybrid composites microstructural observation at different magnifications of the processed combined tungsten carbide and aluminium silicate particulate reinforced lm6 alloy composite test specimens made in grey cast iron mold are analyzed by a metallurgical microscope and hence it is employed to obtain some qualitative evidences on the combined tungsten carbide and aluminium silicate particulate distribution in the alloy matrix and bonding quality between the two particulates and the matrix. metallographic samples of the combined composites are prepared under the standard procedures and hf, hydrofluoric acid is used as an etchant to reveal the phases present in the lm6 alloy matrix. the samples are viewed at different magnifications such as at 50x, and 100x and photomicrographs are captured to predict the confirmation of the presence of the two particulates in the alloy matrix. then, it is further studied to identify the particulate distribution. from the indepth research on this, it is confirmed the presence and distribution of embedded two particulates in the matrix is uniform. the alloy matrix grains are finer and the bonding between particulate surface and the matrix material is satisfactory. it is found that, the morphological distribution of combined particulate for every weight fraction % addition increases. no interfacial reaction products are observed superficially. from this analysis, it is confirmed that the two different particulates reinforced lm6 alloy hybrid composite casting properties are superior to the lm6 alloy with and without grain refiner addition and no particulate reinforcement. in this section, a number of captured photomicrographs are shown in the figure 8 to figure 10 for better understanding. foundry metallurgy of tungsten carbide and aluminium silicate particulate reinforced lm6 alloy hybrid composites issn: 2180-1053 vol. 3 no. 1 january-june 2011 65 53 thermal conductivity vs weight fraction % of tungsten carbide (grey cast iron mold) 0 20 40 60 80 0.00% 5.00% 10.00% 15.00% weight fraction % of tungsten carbide and aluminium silicate (grams) t he rm al c on du ct iv ity [1 /( oh m -m )] figure 7 thermal conductivity vs weight fraction percentage of combined tungsten carbide and aluminium silicate 5.4 metallography of combined tungsten carbide and aluminium silicate particulate reinforced aluminium-11.8% silicon alloy hybrid composites microstructural observation at different magnifications of the processed combined tungsten carbide and aluminium silicate particulate reinforced lm6 alloy composite test specimens made in grey cast iron mold are analyzed by a metallurgical microscope and hence it is employed to obtain some qualitative evidences on the combined tungsten carbide and aluminium silicate particulate distribution in the alloy matrix and bonding quality between the two particulates and the matrix. metallographic samples of the combined composites are prepared under the standard procedures and hf, hydrofluoric acid is used as an etchant to reveal the phases present in the lm6 alloy matrix. the samples are viewed at different magnifications such as at 50x, and 100x and photomicrographs are captured to predict the confirmation of the presence of the two particulates in the alloy matrix. then, it is further studied to identify the particulate distribution. from the in-depth research on this, it is confirmed the presence and distribution of embedded two particulates in the matrix is uniform. the alloy matrix grains are finer and the bonding between particulate surface and the matrix material is satisfactory. it is found that, the morphological distribution of combined particulate for every weight fraction % addition increases. no interfacial reaction products are observed superficially. from this analysis, it is confirmed that the two different particulates reinforced lm6 alloy hybrid composite casting properties are superior to the lm6 alloy with and without grain refiner addition and no particulate reinforcement. in this section, a number of captured photomicrographs are shown in the figure 8 to figure 10 for better understanding. figure 8 2.5 % tungsten carbide and aluminium silicate mixed particulate hybrid composite magnified at 50x figure 8 2.5 % tungsten carbide and aluminium silicate mixed particulate hybrid composite magnified at 50x 54 figure 9 5 % tungsten carbide and aluminium silicate mixed particulate hybrid composite magnified at 50x figure10 10 % tungsten carbide and aluminium silicate mixed particulate hybrid composite magnified at 50 x 5.5 fracture surface analysis and interfacial bonding characterization by sem investigation of hybrid composite test samples is performed by using the leo variable pressure sem 1455 vp series. by using it, the fractures surfaces of the tensile tested samples are observed at higher magnifications to characterize the failure. then, studies on the interphase and bonding are performed to observe the formation of interfacial reaction products and to predict the type of bonding between the particulate surface and the matrix surface. this research consists of two parts. 1. fracture surface analysis of tensile tested composite samples by sem. 2. interphase studies and bonding characteristics between the particulate surface and the matrix by sem. figure 9 5 % tungsten carbide and aluminium silicate mixed particulate hybrid composite magnified at 50x 54 figure 9 5 % tungsten carbide and aluminium silicate mixed particulate hybrid composite magnified at 50x figure10 10 % tungsten carbide and aluminium silicate mixed particulate hybrid composite magnified at 50 x 5.5 fracture surface analysis and interfacial bonding characterization by sem investigation of hybrid composite test samples is performed by using the leo variable pressure sem 1455 vp series. by using it, the fractures surfaces of the tensile tested samples are observed at higher magnifications to characterize the failure. then, studies on the interphase and bonding are performed to observe the formation of interfacial reaction products and to predict the type of bonding between the particulate surface and the matrix surface. this research consists of two parts. 1. fracture surface analysis of tensile tested composite samples by sem. 2. interphase studies and bonding characteristics between the particulate surface and the matrix by sem. figure10 10 % tungsten carbide and aluminium silicate mixed particulate hybrid composite magnified at 50 x 5.5 fracture surface analysis and interfacial bonding characterization by sem investigation of hybrid composite test samples is performed by using the leo variable pressure sem 1455 vp series. by using it, issn: 2180-1053 vol. 3 no. 1 january-june 2011 journal of mechanical engineering and technology 66 the fractures surfaces of the tensile tested samples are observed at higher magnifications to characterize the failure. then, studies on the interphase and bonding are performed to observe the formation of interfacial reaction products and to predict the type of bonding between the particulate surface and the matrix surface. this research consists of two parts. 1. fracture surface analysis of tensile tested composite samples by sem. 2. interphase studies and bonding characteristics between the particulate surface and the matrix by sem. 5.5.1 fracture surface analysis of tensile tested composite samples by sem 55 5.5.1 fracture surface analysis of tensile tested composite samples by sem figure 11 tensile fracture surface of 2.5% weight fraction of combined tungsten carbide and aluminium silicate particulate reinforced hybrid composite magnified at 2000 x figure 11 shows the fractograph of the tensile fracture surface of 2.5% weight fraction of combined tungsten carbide and aluminium silicate particulate hybrid composite magnified at 2000x by sem. the fracture is of brittle type and the particle is pulled out due to poor debonding. a cluster of aluminium silicate particulates are seen in the top right side of the fractograph. a few silicon needles with undeformed condition are also visible. cleavage of the matrix is observed and the tungsten carbide particulate is covered by the silicon particles. figure 12 tensile fracture surface of 5% weight fraction of combined tungsten carbide and aluminium silicate particulate reinforced hybrid composite magnified at 2000-x the above figure 12 shows the fractograph of the fracture surface of 5% combined hybrid particulate composite and it is magnified at 2000x. it reveals the uniform distribution of the two particulates in the lm6 alloy matrix and the fracture has taken place along the interface region of the particulates. a very low deformation zone is seen in the fractograph. figure 11 tensile fracture surface of 2.5% weight fraction of combined tungsten carbide and aluminium silicate particulate reinforced hybrid composite magnified at 2000 x figure 11 shows the fractograph of the tensile fracture surface of 2.5% weight fraction of combined tungsten carbide and aluminium silicate particulate hybrid composite magnified at 2000x by sem. the fracture is of brittle type and the particle is pulled out due to poor debonding. a cluster of aluminium silicate particulates are seen in the top right side of the fractograph. a few silicon needles with undeformed condition are also visible. cleavage of the matrix is observed and the tungsten carbide particulate is covered by the silicon particles. foundry metallurgy of tungsten carbide and aluminium silicate particulate reinforced lm6 alloy hybrid composites issn: 2180-1053 vol. 3 no. 1 january-june 2011 67 55 5.5.1 fracture surface analysis of tensile tested composite samples by sem figure 11 tensile fracture surface of 2.5% weight fraction of combined tungsten carbide and aluminium silicate particulate reinforced hybrid composite magnified at 2000 x figure 11 shows the fractograph of the tensile fracture surface of 2.5% weight fraction of combined tungsten carbide and aluminium silicate particulate hybrid composite magnified at 2000x by sem. the fracture is of brittle type and the particle is pulled out due to poor debonding. a cluster of aluminium silicate particulates are seen in the top right side of the fractograph. a few silicon needles with undeformed condition are also visible. cleavage of the matrix is observed and the tungsten carbide particulate is covered by the silicon particles. figure 12 tensile fracture surface of 5% weight fraction of combined tungsten carbide and aluminium silicate particulate reinforced hybrid composite magnified at 2000-x the above figure 12 shows the fractograph of the fracture surface of 5% combined hybrid particulate composite and it is magnified at 2000x. it reveals the uniform distribution of the two particulates in the lm6 alloy matrix and the fracture has taken place along the interface region of the particulates. a very low deformation zone is seen in the fractograph. figure 12 tensile fracture surface of 5% weight fraction of combined tungsten carbide and aluminium silicate particulate reinforced hybrid composite magnified at 2000-x the above figure 12 shows the fractograph of the fracture surface of 5% combined hybrid particulate composite and it is magnified at 2000x. it reveals the uniform distribution of the two particulates in the lm6 alloy matrix and the fracture has taken place along the interface region of the particulates. a very low deformation zone is seen in the fractograph. 56 figure 13 tensile fracture surface of 10% weight fraction of combined tungsten carbide and aluminium silicate particulate reinforced hybrid composite magnified at 2000-x the above figure 13 shows the fractograph, which reveals the fracture surface of a tensile tested 10% weight fraction of combined particulate hybrid composite. the uncracked long deformed silicon needles are observed. transgranular type of fracture is observed and an interphase region is clearly visible from the fractograph. 5.5.2 interphase studies and bonding characteristics between the particulate surface and the matrix by sem figure 14 interface and bonding in 2.5% weight fraction of combined tungsten carbide and aluminium silicate particulate reinforced composite magnified at 1200-x the above displayed figure 14 shows the microstructure of 2.5% combined particulate hybrid composite magnified at 1200x. a small aluminium silicate particulate is visible at the top right side of the micrograph and a bigger tungsten carbide particulate is seen clearly and well embedded in the lm6 alloy matrix. the bonding between the tungsten carbide particulate is excellently seen in the micrograph and it is not surrounded by any interfacial reaction products. figure 13 tensile fracture surface of 10% weight fraction of combined tungsten carbide and aluminium silicate particulate reinforced hybrid composite magnified at 2000-x the above figure 13 shows the fractograph, which reveals the fracture surface of a tensile tested 10% weight fraction of combined particulate hybrid composite. the uncracked long deformed silicon needles are observed. transgranular type of fracture is observed and an interphase region is clearly visible from the fractograph. issn: 2180-1053 vol. 3 no. 1 january-june 2011 journal of mechanical engineering and technology 68 5.5.2 interphase studies and bonding characteristics between the particulate surface and the matrix by sem 56 figure 13 tensile fracture surface of 10% weight fraction of combined tungsten carbide and aluminium silicate particulate reinforced hybrid composite magnified at 2000-x the above figure 13 shows the fractograph, which reveals the fracture surface of a tensile tested 10% weight fraction of combined particulate hybrid composite. the uncracked long deformed silicon needles are observed. transgranular type of fracture is observed and an interphase region is clearly visible from the fractograph. 5.5.2 interphase studies and bonding characteristics between the particulate surface and the matrix by sem figure 14 interface and bonding in 2.5% weight fraction of combined tungsten carbide and aluminium silicate particulate reinforced composite magnified at 1200-x the above displayed figure 14 shows the microstructure of 2.5% combined particulate hybrid composite magnified at 1200x. a small aluminium silicate particulate is visible at the top right side of the micrograph and a bigger tungsten carbide particulate is seen clearly and well embedded in the lm6 alloy matrix. the bonding between the tungsten carbide particulate is excellently seen in the micrograph and it is not surrounded by any interfacial reaction products. figure 14 interface and bonding in 2.5% weight fraction of combined tungsten carbide and aluminium silicate particulate reinforced composite magnified at 1200-x the above displayed figure 14 shows the microstructure of 2.5% combined particulate hybrid composite magnified at 1200x. a small aluminium silicate particulate is visible at the top right side of the micrograph and a bigger tungsten carbide particulate is seen clearly and well embedded in the lm6 alloy matrix. the bonding between the tungsten carbide particulate is excellently seen in the micrograph and it is not surrounded by any interfacial reaction products. 57 figure 15 interface and bonding in 5% weight fraction of combined tungsten carbide and aluminium silicate particulate reinforced composite magnified at 1200-x the above figure 15 shows the microstructure of 5% weight fraction of combined particulate hybrid composite magnified at 1200x and it reveals the tungsten carbide particulate covered by the dissolute silicon particles from the matrix alloy. figure 16 interface and bonding in 10% weight fraction of combined tungsten carbide and aluminium silicate particulate reinforced composite magnified at 1500-x the micrograph shown in figure 16 shows the microstructure of 10% weight fraction of combined tungsten carbide and aluminium silicate particulate hybrid composite. it reveals the presence of aluminium silicate particulates in the matrix material. a void is present at the center and it contains a cluster of particulates embedded on it. 6.0 conclusion it is concluded that the two combined tungsten carbide particulate and aluminium silicate particulate reinforced lm6 alloy matrix hybrid composite casting properties are superior to the lm6 alloy with and without grain refiner addition and no particulate reinforcement. in this innovative hybrid composite material development research work, the two combined particulates are reinforced in the alloy matrix are processed by liquid vortex metallurgical melt stirring technique. microstructures of the processed hybrid composites based on the metallographic studies have confirmed the uniformity of tungsten carbide particulate and aluminium silicate particulate distribution in the aluminium-11.8% silicon alloy matrix. sufficient amount of turbulence during the mixing of the particulates with the liquid alloy is necessary to get uniform particulate distribution during its solidification processing. the impeller blade type and its rotational speed have not shown any effect on the distribution uniformity. but, faster pouring of the hybrid figure 15 interface and bonding in 5% weight fraction of combined tungsten carbide and aluminium silicate particulate reinforced composite magnified at 1200-x the above figure 15 shows the microstructure of 5% weight fraction of combined particulate hybrid composite magnified at 1200x and it reveals the tungsten carbide particulate covered by the dissolute silicon particles from the matrix alloy. foundry metallurgy of tungsten carbide and aluminium silicate particulate reinforced lm6 alloy hybrid composites issn: 2180-1053 vol. 3 no. 1 january-june 2011 69 57 figure 15 interface and bonding in 5% weight fraction of combined tungsten carbide and aluminium silicate particulate reinforced composite magnified at 1200-x the above figure 15 shows the microstructure of 5% weight fraction of combined particulate hybrid composite magnified at 1200x and it reveals the tungsten carbide particulate covered by the dissolute silicon particles from the matrix alloy. figure 16 interface and bonding in 10% weight fraction of combined tungsten carbide and aluminium silicate particulate reinforced composite magnified at 1500-x the micrograph shown in figure 16 shows the microstructure of 10% weight fraction of combined tungsten carbide and aluminium silicate particulate hybrid composite. it reveals the presence of aluminium silicate particulates in the matrix material. a void is present at the center and it contains a cluster of particulates embedded on it. 6.0 conclusion it is concluded that the two combined tungsten carbide particulate and aluminium silicate particulate reinforced lm6 alloy matrix hybrid composite casting properties are superior to the lm6 alloy with and without grain refiner addition and no particulate reinforcement. in this innovative hybrid composite material development research work, the two combined particulates are reinforced in the alloy matrix are processed by liquid vortex metallurgical melt stirring technique. microstructures of the processed hybrid composites based on the metallographic studies have confirmed the uniformity of tungsten carbide particulate and aluminium silicate particulate distribution in the aluminium-11.8% silicon alloy matrix. sufficient amount of turbulence during the mixing of the particulates with the liquid alloy is necessary to get uniform particulate distribution during its solidification processing. the impeller blade type and its rotational speed have not shown any effect on the distribution uniformity. but, faster pouring of the hybrid figure 16 interface and bonding in 10% weight fraction of combined tungsten carbide and aluminium silicate particulate reinforced composite magnified at 1500-x the micrograph shown in figure 16 shows the microstructure of 10% weight fraction of combined tungsten carbide and aluminium silicate particulate hybrid composite. it reveals the presence of aluminium silicate particulates in the matrix material. a void is present at the center and it contains a cluster of particulates embedded on it. 6.0 conclusion it is concluded that the two combined tungsten carbide particulate and aluminium silicate particulate reinforced lm6 alloy matrix hybrid composite casting properties are superior to the lm6 alloy with and without grain refiner addition and no particulate reinforcement. in this innovative hybrid composite material development research work, the two combined particulates are reinforced in the alloy matrix are processed by liquid vortex metallurgical melt stirring technique. microstructures of the processed hybrid composites based on the metallographic studies have confirmed the uniformity of tungsten carbide particulate and aluminium silicate particulate distribution in the aluminium-11.8% silicon alloy matrix. sufficient amount of turbulence during the mixing of the particulates with the liquid alloy is necessary to get uniform particulate distribution during its solidification processing. the impeller blade type and its rotational speed have not shown any effect on the distribution uniformity. but, faster pouring of the hybrid composite slurry mixture into the grey cast iron mold immediately after the mixing by vortex method has played a significant role in the distribution of the particulates used in this project. a small amount of tungsten particulate segregation has been observed due to its higher density. but, due to combination with aluminium silicate particulate, the effect of segregation is protected and future research work on this mentioned problem can be continued by a metallurgical engineering researcher. 7.0 references aldrich handbook of fine chemicals and laboratory equipment, the sigmaaldrich family; 2004. arthur e. hawkins, the shape of powder-particle outlines, materials science and technology series, research studies press ltd, england, uk; 1993. doru .michael .stefanescu, issues in liquid processing of particulate metal matrix composites, key engineering materials, 1993; 79-80:75-90. donald f adams, leif a carlson, r byron pipes, experimental characterization of advanced composite materials, 3rd edition, crc press llc, florida, usa; 2003. e.a.feest, metal matrix composites for industrial application, materials & design 1996; 7(2):58-64. thoguluva raghavan vijayaram: processing and properties studies of cast particulate reinforced aluminium -11.8% silicon alloy based metal matrix composites, phd thesis in mechanical engineering, faculty of engineering, universiti putra malaysia, upm 43400 serdang, selangor darul ehsan, malaysia 2003-2006. issn: 2180-1053 vol. 5 no. 1 january-june 2013 investigating a power tiller vibration transmissibility using diesel-biodiesel fuel blends on stationary conditions 19 investigating a power tiller vibration transmissibility using diesel-biodiesel fuel blends on stationary conditions bahareh heidary1, seyed reza hassan-beygi2, barat ghobadian3 1,2department of agrotechnology, college of aboureihan, university of tehran, tehran, iran. 3department of mechanics of agricultural machinery engineering, college of agriculture, university of tarbiat modarres, tehran, iran. corresponding email: bahareh_celestial@yahoo.com abstract the wide use of fossil fuel in internal combustion engine cause reduction in these fuel resources and also increase in greenhouse gasses and environmental pollution, for conquering these problems, so many researches have done for finding the renewable herbaceous fuel. between these different kinds of fuel, biodiesel seems to be appropriate because of the nonexistence of air pollutions and the existence of similar trait with diesel fuel. in this research, vibration of 13hp power tiller in 5 levels of engine speed and 6 levels of consuming fuel blends for investigating the power tiller engine vibration behavior and the vibration transmissibility. results showed that vibration transmissibility is decreased by increasing the engine speed. the maximum of the vibration transmissibility is happened in b20 and b5 and minimum of it is happened in b15 and d respectively. also it is observed that the amount of vibration acceleration in longitudinal axis is much more than other two axes. the vibration acceleration value, in the frequency range of 8 to 100 hz was higher than a dangerous frequency range of hand-arm vibration transmission and the total weighted acceleration has the maximum value at the frequency of 20, 31.5, 40 and 63 hz with the vibration amplitude reach up to 20, 12.6, 18.6 and 40 m/s2 respectively. the best engine speed are 1400, 1600 and 2000 rpm, in this engine speed using the b20 and b5 seems appropriate. keywords: vibration, power tiller, fuel, diesel and biodiesel blends, vibration transmissibility 1.0 introduction the wide use of fossil fuel in internal combustion engines causes reduction the available reserves of oil resources, increase in greenhouse gasses, ozone layer destruction, high environmental pollution, different outbreak disease and respiratory disease in big cities. for conquering these problems many research issn: 2180-1053 vol. 5 no. 1 january-june 2013 journal of mechanical engineering and technology 20 works have been carried out to find alternative energies which have being renewable, non-toxic, safe and reduce environmental pollutions. biodiesel as an alternative liquefied biofuel has made significant progress and access to it is possible. the biodiesel could be produced by natural resources such as plant oils, animal fats, waste oils and algae (carratto et.al., 2004; ghobadian and khatamifar, 2006). the specifications of biodiesel are similar to diesel fuel but it has not some disorder such as sulfurs, nitrogen and aromatic polycyclic. the biodiesel can be used pure or blend with diesel fuel in transportation systems, heating buildings and factories, and also in industrial processes without any change in fuel supply systems (zenoozi., 2007; lee et.al., 2004). biodiesel during combustion produces less pollution compared to fossil fuel (dorado et.al., 2004). power tillers are agricultural machines which are fitted with small diesel engines. economical features and user capabilities of power tillers in various conditions have caused these machines are used as a main source of power in small and medium size farms (sam and kathirvel, 2006; hassan-beygi et.al., 2005). single cylinder diesel engine of power tiller has not a good balance. the forces acting on the piston during compression and power strokes transmitted to the crankshaft and engine block. due to lack of using vibration dampers between engine and power tiller chassis the engine forces entering to the tractor chassis as shocks and then through the chassis transmitted to the power tiller handle. power tiller handle also acts like cantilever beam so that one end is attached to the tractor chassis and free vibration of the other end is high (salokhe et.al., 1995). operators of the walking type power tillers are exposed to high levels of noise and vibration because these machines are guided entirely by operator hands. long time working with these machines might be caused damage to various organs of the body including hearing loss, spine and gastrointestinal disorders and even neurological disorders. furthermore it causes decrease in work efficiency and quality (goglia et.al., 2006). in order to reduce the risks of working with these machines the regulations have been developed by international organizations to limit working hours and duration of vibration exposure. regarding to fossil fuel limitations application of renewable energies would be a necessity. change in kind of fuel cause for changing in combustion process of internal combustion engines. since the main portion of power tiller vibration is due to the diesel engine, so using the diesel-biodiesel fuel blends could be varied the vibration behavior of this type of tractors. literature survey showed that some researchers were evaluated sound and vibration of power tillers by using conventional fossil fuel (hassan-beygi and ghobadian, 2005; taghizadeh-alisaraei, 2007; sam and kathrivel, 2006; dewangan and tewari, 2009; sam and kathrivel, 2009). however, the vibration behavior of power tiller fuelled by diesel-biodiesel fuel blends did not yet investigate . in this study, the vibration of a power tiller was measured on stationary mode for various engine speeds and diesel-biodiesel fuel blends at the power tiller handle position and vibration transmissibility to operator hand was evaluated. issn: 2180-1053 vol. 5 no. 1 january-june 2013 investigating a power tiller vibration transmissibility using diesel-biodiesel fuel blends on stationary conditions 21 the data obtained from this research could be used for choosing appropriate diesel-biodiesel fuel blends in order to reduce vibration harmful effects on the operator. 2.0 materials and methods the power tiller used for this research study was fitted with a single cylinder, four-stroke, naturally aspirated, water-cooled, indirect injection diesel engine, providing 13-hp power at rated engine speed of 2200 rpm. the vibration of the power tiller was measured on stationary conditions on asphalt surface in open area. the instruments used in this study, consisted of three accelerometers, a tachometer, a lap-top computer and a few other devices. the detailed specifications of the instruments are given in table 1. table 1 specifications of the used instruments table 1 specifications of the used instruments. name of instrument model accuracy/ resolution range accelerometer ctc-ac102 ±3 db 0.4 to 14000 hz tachometer lutron dt2268 1 rpm 1-20000 rpm laptop computer sony vpc-f12-lgx a/d converter advantech 4711 12 bits 150000 hzsignals capture software lab view 2009 in this research the selected variables were engine speeds, diesel-biodiesel fuel blends and directions of vibration measurement. the range of variables considered to perform the test could cover the normal and safe operating range of the power tiller during operation. table 2 shows the test matrix for this study. all of the experiments were replicated three times. table 2 matrix of experimentations. parameters levels of parameters 1 2 3 4 5 6 engine speed (rpm) 1400 1600 1800 200 22000 diesel-biodiesel fuel blends b0 (0% biodiesel) b5 (5% biodiesel) b10 (10% biodiesel) b15 (15% biodiesel) b20 (20% biodiesel) b100 (100% biodiesel) position of sensors handle chest of operator orientations of measurement lateral longitude vertical vibration occurs in lateral, longitudinal and vertical translational axes so in order to measure the vibration of the power tiller at positions in front of engine, handle grip and chest of operator, three accelerometers were screwed on a steel cube with dimensions of 2×2×2 cm3 according to iso 5349 standard directions (fig. 1). the cube was installed on the power tiller handle by a strong metallic grip, also it glued on a wide leather belt, which was fastened on the user chest so tightly with a little pain and upset in user (fig. 2). the required power for the accelerometers was supplied from a 24-volt power supply and an electronic circuit. using an a/d converter which was recognized and controlled by labview software program, the accelerometer analog output voltage was converted to digital ones with 40000 hz sampling rate and recorded on laptop computer hard disk for each accelerometer separately. the equipments that used for this research was shown in fig. 3. human response to vibration is dependent on the frequency of vibration so for a detailed investigation of the vibration signals, the recorded time domain digital signals were converted to frequency domain narrow band signals by fast fourier transform (fft) algorithm using matlab software program. due to sudden change and uncertainty of the narrow band signals as well as complication of comparing the independent parameters, as shown in table 2, on the 3 in this research the selected variables were engine speeds, diesel-biodiesel fuel blends and directions of vibration measurement. the range of variables considered to perform the test could cover the normal and safe operating range of the power tiller during operation. table 2 shows the test matrix for this study. all of the experiments were replicated three times. table 2 matrix of experimentations table 1 specifications of the used instruments. name of instrument model accuracy/ resolution range accelerometer ctc-ac102 ±3 db 0.4 to 14000 hz tachometer lutron dt2268 1 rpm 1-20000 rpm laptop computer sony vpc-f12-lgx a/d converter advantech 4711 12 bits 150000 hzsignals capture software lab view 2009 in this research the selected variables were engine speeds, diesel-biodiesel fuel blends and directions of vibration measurement. the range of variables considered to perform the test could cover the normal and safe operating range of the power tiller during operation. table 2 shows the test matrix for this study. all of the experiments were replicated three times. table 2 matrix of experimentations. parameters levels of parameters 1 2 3 4 5 6 engine speed (rpm) 1400 1600 1800 200 22000 diesel-biodiesel fuel blends b0 (0% biodiesel) b5 (5% biodiesel) b10 (10% biodiesel) b15 (15% biodiesel) b20 (20% biodiesel) b100 (100% biodiesel) position of sensors handle chest of operator orientations of measurement lateral longitude vertical vibration occurs in lateral, longitudinal and vertical translational axes so in order to measure the vibration of the power tiller at positions in front of engine, handle grip and chest of operator, three accelerometers were screwed on a steel cube with dimensions of 2×2×2 cm3 according to iso 5349 standard directions (fig. 1). the cube was installed on the power tiller handle by a strong metallic grip, also it glued on a wide leather belt, which was fastened on the user chest so tightly with a little pain and upset in user (fig. 2). the required power for the accelerometers was supplied from a 24-volt power supply and an electronic circuit. using an a/d converter which was recognized and controlled by labview software program, the accelerometer analog output voltage was converted to digital ones with 40000 hz sampling rate and recorded on laptop computer hard disk for each accelerometer separately. the equipments that used for this research was shown in fig. 3. human response to vibration is dependent on the frequency of vibration so for a detailed investigation of the vibration signals, the recorded time domain digital signals were converted to frequency domain narrow band signals by fast fourier transform (fft) algorithm using matlab software program. due to sudden change and uncertainty of the narrow band signals as well as complication of comparing the independent parameters, as shown in table 2, on the 3 vibration occurs in lateral, longitudinal and vertical translational axes so in order to measure the vibration of the power tiller at positions in front of engine, handle grip and chest of operator, three accelerometers were screwed on a steel cube with dimensions of 2×2×2 cm3 according to iso 5349 standard directions (fig. 1). the cube was installed on the power tiller handle by a strong metallic grip, also it glued on a wide leather belt, which was fastened on the issn: 2180-1053 vol. 5 no. 1 january-june 2013 journal of mechanical engineering and technology 22 user chest so tightly with a little pain and upset in user (fig. 2). the required power for the accelerometers was supplied from a 24-volt power supply and an electronic circuit. using an a/d converter which was recognized and controlled by labview software program, the accelerometer analog output voltage was converted to digital ones with 40000 hz sampling rate and recorded on laptop computer hard disk for each accelerometer separately. the equipments that used for this research was shown in fig. 3. human response to vibration is dependent on the frequency of vibration so for a detailed investigation of the vibration signals, the recorded time domain digital signals were converted to frequency domain narrow band signals by fast fourier transform (fft) algorithm using matlab software program. due to sudden change and uncertainty of the narrow band signals as well as complication of comparing the independent parameters, as shown in table 2, on the narrow band signals of vibration acceleration, the narrow band signals were converted to 1/3rd octave frequency band signals by using a subroutine computer program. different parts of fig. 5 shows a recorded vibration signal of the power tiller in time domain (part a) and corresponding narrow band frequency domain signal (part b) as well as 1/3rd octave broadband frequency signal (part c). human perception of vibration is the most at low frequency and the perception greatly decreases with frequency, so the weighting factor for vibration varies with frequency. the 1/3rd octave frequency band weights were defined on iso 5349 standard. the 1/3rd octave spectra of the power tiller vibration signals were weighted in accordance with iso 5349 standard (dewangan 2009, goglia et.al. 2006, griffin1996). the weighted acceleration value, ahw, was calculated as: (iso 5349, 2001; dewangan 2009, goglia et.al., 2006) narrow band signals of vibration acceleration, the narrow band signals were converted to 1/3rd octave frequency band signals by using a subroutine computer program. different parts of fig. 5 shows a recorded vibration signal of the power tiller in time domain (part a) and corresponding narrow band frequency domain signal (part b) as well as 1/3rd octave broadband frequency signal (part c). human perception of vibration is the most at low frequency and the perception greatly decreases with frequency, so the weighting factor for vibration varies with frequency. the 1/3rd octave frequency band weights were defined on iso 5349 standard. the 1/3rd octave spectra of the power tiller vibration signals were weighted in accordance with iso 5349 standard (dewangan 2009, goglia et al. 2006, griffin1996). the weighted acceleration value, ahw, was calculated as: (iso 5349, 2001; dewangan 2009, goglia et al., 2006) (1) where: kj is the weighting factor for j-the frequency according to iso 5349 standard factors, ah,j is the rms value of measured vibration in 1/3rd octave frequency band (m/s2) and n is the number of frequencies used in one third octave band (from 6.3 to 1250). the power tiller vibration acceleration rms (root mean square) values calculated by equation (2): ( ) 1/2 t 0 2 rms dttat 1 a     = ∫ (2) where: arms are the root mean square of vibration acceleration (m/s2), a(t) is measured vibration acceleration amplitude (m/s2), and t is the duration of measured vibration acceleration (s). the evaluation of vibration transmissibility in accordance with iso 5349 standard is based on three axes vibration combination named total weighted vibration acceleration, ahv. the vibration total value, ahv, was determined as: (iso 5349, 2001; dewangan, 2009; goglia et al., 2006) (3) where: eve is the total rms value of vibration acceleration (m/s2), ahwx is total weighted vibration acceleration in x axis (m/s2), ahwy is total weighted vibration acceleration in y axis (m/s2) and ahwz is total weighted vibration acceleration in z axis (m/s2) (mansfield, 2005). vibration transmissibility is defined as the ratio of the vibration measured on the hand–arm system to the input vibration on the handle of power tiller. it can be represented as (dewangan 2008): (4) where: am is the total weighted reams' the measured accelerationion at the hand or chest of the user. 4 where: kj is the weighting factor for j-the frequency according to iso 5349 standard factors, ah,j is the rms value of measured vibration in 1/3rd octave frequency band (m/s2) and n is the number of frequencies used in one third octave band (from 6.3 to 1250). the power tiller vibration acceleration rms (root mean square) values calculated by equation (2): narrow band signals of vibration acceleration, the narrow band signals were converted to 1/3rd octave frequency band signals by using a subroutine computer program. different parts of fig. 5 shows a recorded vibration signal of the power tiller in time domain (part a) and corresponding narrow band frequency domain signal (part b) as well as 1/3rd octave broadband frequency signal (part c). human perception of vibration is the most at low frequency and the perception greatly decreases with frequency, so the weighting factor for vibration varies with frequency. the 1/3rd octave frequency band weights were defined on iso 5349 standard. the 1/3rd octave spectra of the power tiller vibration signals were weighted in accordance with iso 5349 standard (dewangan 2009, goglia et al. 2006, griffin1996). the weighted acceleration value, ahw, was calculated as: (iso 5349, 2001; dewangan 2009, goglia et al., 2006) (1) where: kj is the weighting factor for j-the frequency according to iso 5349 standard factors, ah,j is the rms value of measured vibration in 1/3rd octave frequency band (m/s2) and n is the number of frequencies used in one third octave band (from 6.3 to 1250). the power tiller vibration acceleration rms (root mean square) values calculated by equation (2): ( ) 1/2 t 0 2 rms dttat 1 a     = ∫ (2) where: arms are the root mean square of vibration acceleration (m/s2), a(t) is measured vibration acceleration amplitude (m/s2), and t is the duration of measured vibration acceleration (s). the evaluation of vibration transmissibility in accordance with iso 5349 standard is based on three axes vibration combination named total weighted vibration acceleration, ahv. the vibration total value, ahv, was determined as: (iso 5349, 2001; dewangan, 2009; goglia et al., 2006) (3) where: eve is the total rms value of vibration acceleration (m/s2), ahwx is total weighted vibration acceleration in x axis (m/s2), ahwy is total weighted vibration acceleration in y axis (m/s2) and ahwz is total weighted vibration acceleration in z axis (m/s2) (mansfield, 2005). vibration transmissibility is defined as the ratio of the vibration measured on the hand–arm system to the input vibration on the handle of power tiller. it can be represented as (dewangan 2008): (4) where: am is the total weighted reams' the measured accelerationion at the hand or chest of the user. 4 where: arms are the root mean square of vibration acceleration (m/s2), a(t) is measured vibration acceleration amplitude (m/s2), and t is the duration of measured vibration acceleration (s). issn: 2180-1053 vol. 5 no. 1 january-june 2013 investigating a power tiller vibration transmissibility using diesel-biodiesel fuel blends on stationary conditions 23 the evaluation of vibration transmissibility in accordance with iso 5349 standard is based on three axes vibration combination named total weighted vibration acceleration, ahv. the vibration total value, ahv, was determined as: (iso 5349, 2001; dewangan, 2009; goglia et.al., 2006) narrow band signals of vibration acceleration, the narrow band signals were converted to 1/3rd octave frequency band signals by using a subroutine computer program. different parts of fig. 5 shows a recorded vibration signal of the power tiller in time domain (part a) and corresponding narrow band frequency domain signal (part b) as well as 1/3rd octave broadband frequency signal (part c). human perception of vibration is the most at low frequency and the perception greatly decreases with frequency, so the weighting factor for vibration varies with frequency. the 1/3rd octave frequency band weights were defined on iso 5349 standard. the 1/3rd octave spectra of the power tiller vibration signals were weighted in accordance with iso 5349 standard (dewangan 2009, goglia et al. 2006, griffin1996). the weighted acceleration value, ahw, was calculated as: (iso 5349, 2001; dewangan 2009, goglia et al., 2006) (1) where: kj is the weighting factor for j-the frequency according to iso 5349 standard factors, ah,j is the rms value of measured vibration in 1/3rd octave frequency band (m/s2) and n is the number of frequencies used in one third octave band (from 6.3 to 1250). the power tiller vibration acceleration rms (root mean square) values calculated by equation (2): ( ) 1/2 t 0 2 rms dttat 1 a     = ∫ (2) where: arms are the root mean square of vibration acceleration (m/s2), a(t) is measured vibration acceleration amplitude (m/s2), and t is the duration of measured vibration acceleration (s). the evaluation of vibration transmissibility in accordance with iso 5349 standard is based on three axes vibration combination named total weighted vibration acceleration, ahv. the vibration total value, ahv, was determined as: (iso 5349, 2001; dewangan, 2009; goglia et al., 2006) (3) where: eve is the total rms value of vibration acceleration (m/s2), ahwx is total weighted vibration acceleration in x axis (m/s2), ahwy is total weighted vibration acceleration in y axis (m/s2) and ahwz is total weighted vibration acceleration in z axis (m/s2) (mansfield, 2005). vibration transmissibility is defined as the ratio of the vibration measured on the hand–arm system to the input vibration on the handle of power tiller. it can be represented as (dewangan 2008): (4) where: am is the total weighted reams' the measured accelerationion at the hand or chest of the user. 4 where: eve is the total rms value of vibration acceleration (m/s2), ahwx is total weighted vibration acceleration in x axis (m/s2), ahwy is total weighted vibration acceleration in y axis (m/s2) and ahwz is total weighted vibration acceleration in z axis (m/s2) (mansfield, 2005). vibration transmissibility is defined as the ratio of the vibration measured on the hand–arm system to the input vibration on the handle of power tiller. it can be represented as (dewangan, 2008): narrow band signals of vibration acceleration, the narrow band signals were converted to 1/3rd octave frequency band signals by using a subroutine computer program. different parts of fig. 5 shows a recorded vibration signal of the power tiller in time domain (part a) and corresponding narrow band frequency domain signal (part b) as well as 1/3rd octave broadband frequency signal (part c). human perception of vibration is the most at low frequency and the perception greatly decreases with frequency, so the weighting factor for vibration varies with frequency. the 1/3rd octave frequency band weights were defined on iso 5349 standard. the 1/3rd octave spectra of the power tiller vibration signals were weighted in accordance with iso 5349 standard (dewangan 2009, goglia et al. 2006, griffin1996). the weighted acceleration value, ahw, was calculated as: (iso 5349, 2001; dewangan 2009, goglia et al., 2006) (1) where: kj is the weighting factor for j-the frequency according to iso 5349 standard factors, ah,j is the rms value of measured vibration in 1/3rd octave frequency band (m/s2) and n is the number of frequencies used in one third octave band (from 6.3 to 1250). the power tiller vibration acceleration rms (root mean square) values calculated by equation (2): ( ) 1/2 t 0 2 rms dttat 1 a     = ∫ (2) where: arms are the root mean square of vibration acceleration (m/s2), a(t) is measured vibration acceleration amplitude (m/s2), and t is the duration of measured vibration acceleration (s). the evaluation of vibration transmissibility in accordance with iso 5349 standard is based on three axes vibration combination named total weighted vibration acceleration, ahv. the vibration total value, ahv, was determined as: (iso 5349, 2001; dewangan, 2009; goglia et al., 2006) (3) where: eve is the total rms value of vibration acceleration (m/s2), ahwx is total weighted vibration acceleration in x axis (m/s2), ahwy is total weighted vibration acceleration in y axis (m/s2) and ahwz is total weighted vibration acceleration in z axis (m/s2) (mansfield, 2005). vibration transmissibility is defined as the ratio of the vibration measured on the hand–arm system to the input vibration on the handle of power tiller. it can be represented as (dewangan 2008): (4) where: am is the total weighted reams' the measured accelerationion at the hand or chest of the user. 4 where: am is the total weighted reams' the measured accelerationion at the hand or chest of the user. figure 1 orientation of the axes for the vibration measurement on the power tiller (iso 5349, 2001). figure 2 monitoring of accelerometers and orientation of measurement axes. figure 3 equipments used for the vibration measurement. the reason of investigating of vibration acceleration in stationary mode was its stability, this mode could be used for investigating and comparing with other modalities, furthermore it could be used as a basis of vibration assay. the total weighted vibration acceleration (eve) was calculated and vibration transmissibility was calculated by 6 levels of fuel blends and 5 level of engine speed by relation 3 and in different frequencies in relation 4. the vibration transmissibility were analyzed statistically by using factorial tests with completely random design in sas and excell software, for obtaining the effect of speed and fuel blends. 5 figure 1 orientation of the axes for the vibration measurement on the power tiller (iso 5349, 2001)figure 1 orientation of the axes for the vibration measurement on the power tiller (iso 5349, 2001). figure 2 monitoring of accelerometers and orientation of measurement axes. figure 3 equipments used for the vibration measurement. the reason of investigating of vibration acceleration in stationary mode was its stability, this mode could be used for investigating and comparing with other modalities, furthermore it could be used as a basis of vibration assay. the total weighted vibration acceleration (eve) was calculated and vibration transmissibility was calculated by 6 levels of fuel blends and 5 level of engine speed by relation 3 and in different frequencies in relation 4. the vibration transmissibility were analyzed statistically by using factorial tests with completely random design in sas and excell software, for obtaining the effect of speed and fuel blends. 5 figure 2 monitoring of accelerometers and orientation of measurement axes issn: 2180-1053 vol. 5 no. 1 january-june 2013 journal of mechanical engineering and technology 24 figure 1 orientation of the axes for the vibration measurement on the power tiller (iso 5349, 2001). figure 2 monitoring of accelerometers and orientation of measurement axes. figure 3 equipments used for the vibration measurement. the reason of investigating of vibration acceleration in stationary mode was its stability, this mode could be used for investigating and comparing with other modalities, furthermore it could be used as a basis of vibration assay. the total weighted vibration acceleration (eve) was calculated and vibration transmissibility was calculated by 6 levels of fuel blends and 5 level of engine speed by relation 3 and in different frequencies in relation 4. the vibration transmissibility were analyzed statistically by using factorial tests with completely random design in sas and excell software, for obtaining the effect of speed and fuel blends. 5 figure 3 equipments used for the vibration measurement the reason of investigating of vibration acceleration in stationary mode was its stability, this mode could be used for investigating and comparing with other modalities, furthermore it could be used as a basis of vibration assay. the total weighted vibration acceleration (eve) was calculated and vibration transmissibility was calculated by 6 levels of fuel blends and 5 level of engine speed by relation 3 and in different frequencies in relation 4. the vibration transmissibility were analyzed statistically by using factorial tests with completely random design in sas and excell software, for obtaining the effect of speed and fuel blends. 3.0 results and discussion this paper presents the results of hand-transmitted vibration in a stationary condition for engine position and handle and chest of user in three axes of vertical, longitudinal and lateral axes and in 6 levels of engine speed and 6 levels of diesel and biodiesel fuel blends. these levels of parameter are independent variable, and the tests were carried out with 3 replications. data was analyzed for vibration acceleration in rms (ah,j) at 1/3 octave band in the frequency range between 2.15 and 20000 hz for each test fuel mixture and each axis. for each axis the frequency-weighted rms acceleration (ahwx, ahwy and ahwz) was calculated using the filter suggested by (iso 5349-2, 2001). an average of the three trials to test was calculated. vibration total value (ahv) was calculated for each test fuel mixture. vibration exposure during operation of the hand tractor was calculated for each test fuel blend as (iso 5349-2, 2001). finally an optimum percentage of biodiesel and diesel mixture was represented. 3.1 weighted vibration acceleration (rms) the total weighted vibration acceleration (ahv) at the power tiller handle position is shown in figure 4. it could be also observed from this figure that with increasing engine speed from 1400 to 2200 rpm, the weighted vibration acceleration was increased in the range of 7.44 m/s2 to 21.52 m/s2. the maximum of vibration acceleration was 21.52 m/s2 belongs to b15 in 1800 rpm and 15.39 m/s2 belongs to b20 in 1800 rpm and 15.23 m/s2 belongs to b100 in 1800 rpm. the least of the vibration acceleration happened in b10, b15, b20 and b100 respectively. issn: 2180-1053 vol. 5 no. 1 january-june 2013 investigating a power tiller vibration transmissibility using diesel-biodiesel fuel blends on stationary conditions 25 3.0 results and discussion this paper presents the results of hand-transmitted vibration in a stationary condition for engine position and handle and chest of user in three axes of vertical, longitudinal and lateral axes and in 6 levels of engine speed and 6 levels of diesel and biodiesel fuel blends. these levels of parameter are independent variable, and the tests were carried out with 3 replications. data was analyzed for vibration acceleration in rms (ah,j) at 1/3 octave band in the frequency range between 2.15 and 20000 hz for each test fuel mixture and each axis. for each axis the frequency-weighted rms acceleration (ahwx, ahwy and ahwz) was calculated using the filter suggested by (iso 5349-2, 2001). an average of the three trials to test was calculated. vibration total value (ahv) was calculated for each test fuel mixture. vibration exposure during operation of the hand tractor was calculated for each test fuel blend as (iso 5349-2, 2001). finally an optimum percentage of biodiesel and diesel mixture was represented. 3.1 weighted vibration acceleration (rms) the total weighted vibration acceleration (ahv ) at the power tiller handle position is shown in figure 4. it could be also observed from this figure that with increasing engine speed from 1400 to 2200 rpm, the weighted vibration acceleration was increased in the range of 7.44 m/s2 to 21.52 m/s2. the maximum of vibration acceleration was 21.52 m/s2 belongs to b15 in 1800 rpm and 15.39 m/s2 belongs to b20 in 1800 rpm and 15.23 m/s2 belongs to b100 in 1800 rpm. the least of the vibration acceleration happened in b10, b15, b20 and b100 respectively. figure 4 the total weighted vibration acceleration (ahv). as the figure 5 shows, comparing the vibration acceleration values in three axes of vertical, lateral and longitudinal directions showed that the rms vibration acceleration in the vertical direction was the maximum and in longitudinal axis was minimum. in almost all the engine speeds the fuel blends b5 and d showed more vibration acceleration rms than b15, b20 and b100 and b10 respectively. 6 figure 4 the total weighted vibration acceleration (ahv). as the figure 5 shows, comparing the vibration acceleration values in three axes of vertical, lateral and longitudinal directions showed that the rms vibration acceleration in the vertical direction was the maximum and in longitudinal axis was minimum. in almost all the engine speeds the fuel blends b5 and d showed more vibration acceleration rms than b15, b20 and b100 and b10 respectively. figure 5 the weighted vibration acceleration in 2000 rpm engine speed for different fuel blends. 3.2 vibration transmissibility in considering the effect of engine speed and fuel blends variants at vibration transmissibility (year), the variance analysis in sas software is done and the results are shown in table 3. the parameter of engine speed and fuel blends and the interaction of engine speed and fuel blends are prominent in 1% level as has been shown in table 4. comparing the mean squares is shown in table 4 and figure 6 and 7. it is observed that the least of the vibration transmissibility from handle to the hand and body of user happened in b5, b10 and b20. the vibration transmissibility is decreased by increasing the engine speed, just one exception exists and that was 1800 rpm engine speed, the reason was intensification that happened at this speed and the handle has severe vibration but it had no effect on transmitted vibration. also it is observed that vibration transmissibility is most in vertical and longitudinal axes. table 3 intraction of fuel blends and engine speeds. sl. no. variant source freedom degree mean square 1 replication 2 0.00182529ns 2 engine speed 4 0.00757560* 3 fuel blend 5 0.00322901* 4 engine speed * fuel blend 20 0.00020926* 5 error 58 0.00013761ns *means significant at 1% level 7 figure 5 the weighted vibration acceleration in 2000 rpm engine speed for different fuel blends. 3.2 vibration transmissibility in considering the effect of engine speed and fuel blends variants at vibration transmissibility (year), the variance analysis in sas software is done and the results are shown in table 3. the parameter of engine speed and fuel blends and the interaction of engine speed and fuel blends are prominent in 1% level as has been shown in table 4. comparing the mean squares is shown in table 4 and figure 6 and 7. it is observed that the least of the vibration transmissibility from handle to the hand and body of user happened in b5, b10 and b20. the vibration transmissibility is decreased by increasing the engine speed, just one exception exists and that was 1800 rpm engine speed, the reason was intensification that happened at this speed and the handle has severe vibration but it had no effect issn: 2180-1053 vol. 5 no. 1 january-june 2013 journal of mechanical engineering and technology 26 on transmitted vibration. also it is observed that vibration transmissibility is most in vertical and longitudinal axes. table 3 intraction of fuel blends and engine speeds figure 5 the weighted vibration acceleration in 2000 rpm engine speed for different fuel blends. 3.2 vibration transmissibility in considering the effect of engine speed and fuel blends variants at vibration transmissibility (year), the variance analysis in sas software is done and the results are shown in table 3. the parameter of engine speed and fuel blends and the interaction of engine speed and fuel blends are prominent in 1% level as has been shown in table 4. comparing the mean squares is shown in table 4 and figure 6 and 7. it is observed that the least of the vibration transmissibility from handle to the hand and body of user happened in b5, b10 and b20. the vibration transmissibility is decreased by increasing the engine speed, just one exception exists and that was 1800 rpm engine speed, the reason was intensification that happened at this speed and the handle has severe vibration but it had no effect on transmitted vibration. also it is observed that vibration transmissibility is most in vertical and longitudinal axes. table 3 intraction of fuel blends and engine speeds. sl. no. variant source freedom degree mean square 1 replication 2 0.00182529ns 2 engine speed 4 0.00757560* 3 fuel blend 5 0.00322901* 4 engine speed * fuel blend 20 0.00020926* 5 error 58 0.00013761ns *means significant at 1% level 7 table 4 vibration transmissibility on different fuel blend at various engine speed table 4 vibration transmissibility on different fuel blend at various engine speed fuel blends db5b10b15b20b100speed(rpm) 0.05 30.0470.0520.0730.0430.062 1400 0.06 40.0480.0520.0820.0640.059 1600 0.11 60.070.0790.1310.0990.112 1800 0.120.0850.0760.1070.0930.0952000 0.08 40.0520.0560.1040.0720.085 2200 figure 6 and 7 show that the vibration transmissibility is decreased by increasing the engine speed, and it is decreased significantly in 1400, 1600, 2000 and 2200 respectively. just in 1800 rpm engine speed the vibration transmissibility decrease severly and the reason was vibration intensification that happened in this speed and the handle has severe vibration but the hand of operator damp the vibration so no change in vibration were observed at the hand and chest of operator. the maximum of vibration transsmissibility happened in engine speed of 1400 rpm and the minimum happened in 1800 rpm. the maximum of vibration transmissibility in 1400 rpm engine speed happened in b20 and b5. the minimum of vibration transmissibility in 1800 rpm of engine speed happened in b15 and d respectively but the magnitude of vibration acceleration in this engine speed was high on engine and handle of powertiller. it shows that the body of operator damp this vibration and so using the engine speed of 1800 rpm may damage to operators health and put them in to high risk of inducing white finger and vascular disease. the best engine speed are 1400, 1600 and 2000 rpm, in this engine speed using the b20 and b5 seems appropriate. also it is observed that the amount of vibration acceleration in longitudinal axis is much more than other two axes and in there is no difference between vertical axis and lateral axis. the results of this research were corresponded with the results of (dewangan,et al., 2008) research were corresponded the results of this research, he measured the acceleration at metacarpal, wrist, elbow and acromion, in transportation and rota-puddling. he showed the maximum transmissibility was observed during the rota-tilling operation with the mean values of 0.91, 0.47, 0.30 and 0.21 at the metacarpal, wrist, elbow and acromion, respectively. he also showed the resonance at the metacarpal was observed in the frequency range of 31.5–100 hz high-frequency vibration (100–1250 hz) was primarily localized to the hands. he showed with increasing in engine speed transmissibility increased and he also showed the maximum transmissibility was occurring in x-axis. 8 figure 6 and 7 show that the vibration transmissibility is decreased by increasing the engine speed, and it is decreased significantly in 1400, 1600, 2000 and 2200 respectively. just in 1800 rpm engine speed the vibration transmissibility decrease severly and the reason was vibration intensification that happened in this speed and the handle has severe vibration but the hand of operator damp the vibration so no change in vibration were observed at the hand and chest of operator. the maximum of vibration transsmissibility happened in engine speed of 1400 rpm and the minimum happened in 1800 rpm. the maximum of vibration transmissibility in 1400 rpm engine speed happened in b20 and b5. the minimum of vibration transmissibility in 1800 rpm of engine speed happened in b15 and d respectively but the magnitude of vibration acceleration in this engine speed was high on engine and handle of powertiller. it shows that the body of operator damp this vibration and so using the engine speed of 1800 rpm may damage to operators health and put them in to high risk of inducing white finger and vascular disease. the best engine speed are 1400, 1600 and 2000 rpm, in this engine speed using the b20 and b5 seems appropriate. also it is observed that the amount of vibration acceleration in longitudinal axis is much more than other two axes and in there is no difference between vertical axis and lateral axis. issn: 2180-1053 vol. 5 no. 1 january-june 2013 investigating a power tiller vibration transmissibility using diesel-biodiesel fuel blends on stationary conditions 27 the results of this research were corresponded with the results of (dewangan, et.al., 2008) research were corresponded the results of this research, he measured the acceleration at metacarpal, wrist, elbow and acromion, in transportation and rota-puddling. he showed the maximum transmissibility was observed during the rota-tilling operation with the mean values of 0.91, 0.47, 0.30 and 0.21 at the metacarpal, wrist, elbow and acromion, respectively. he also showed the resonance at the metacarpal was observed in the frequency range of 31.5–100 hz high-frequency vibration (100–1250 hz) was primarily localized to the hands. he showed with increasing in engine speed transmissibility increased and he also showed the maximum transmissibility was occurring in x-axis. figure 6 the engine speeds and fuel blends intraction. figure 7 the engine speeds and fuel blends intraction. the envelope curve of total weighted acceleration (m/s2) of the power tiller handle at different engine speeds was illustrated in figure 8. according to this figure, the vibration acceleration value, in the frequency range of 8 to 100 hz was higher than a dangerous frequency range of hand-arm vibration transmission according to figure 1 the total weighted acceleration has the maximum value at the frequency of 20, 31.5, 40 and 63 hz with the vibration amplitude reach up to 20, 12.6, 18.6 and 40 m/s2 respectively (figure 8). according to this figure, the vibration acceleration value, in the frequency range of 12.5 to 125 hz was higher than exposure limit, 2 m/s2, for hand-arm vibration exposure for 8-hour working per day. however, the mean value of total vibration acceleration in low frequencies, which is the most sensitive frequency of handarm system, has considerable values. at frequencies above 40 hz for all of the fuel blends and engine speeds, the acceleration value is less than 1 m/s2. the frequency less than 1 hz is approximately negligible so the motion sickness doesn’t appear in operators of 13 hp power 9 figure 6 the engine speeds and fuel blends intraction figure 6 the engine speeds and fuel blends intraction. figure 7 the engine speeds and fuel blends intraction. the envelope curve of total weighted acceleration (m/s2) of the power tiller handle at different engine speeds was illustrated in figure 8. according to this figure, the vibration acceleration value, in the frequency range of 8 to 100 hz was higher than a dangerous frequency range of hand-arm vibration transmission according to figure 1 the total weighted acceleration has the maximum value at the frequency of 20, 31.5, 40 and 63 hz with the vibration amplitude reach up to 20, 12.6, 18.6 and 40 m/s2 respectively (figure 8). according to this figure, the vibration acceleration value, in the frequency range of 12.5 to 125 hz was higher than exposure limit, 2 m/s2, for hand-arm vibration exposure for 8-hour working per day. however, the mean value of total vibration acceleration in low frequencies, which is the most sensitive frequency of handarm system, has considerable values. at frequencies above 40 hz for all of the fuel blends and engine speeds, the acceleration value is less than 1 m/s2. the frequency less than 1 hz is approximately negligible so the motion sickness doesn’t appear in operators of 13 hp power 9 figure 7 the engine speeds and fuel blends intraction the envelope curve of total weighted acceleration (m/s2) of the power tiller handle at different engine speeds was illustrated in figure 8. according to this figure, the vibration acceleration value, in the frequency range of 8 to 100 hz was higher than a dangerous frequency range of hand-arm vibration transmission according to figure 1 the total weighted acceleration has the maximum value at the frequency of 20, 31.5, 40 and 63 hz with the vibration amplitude reach up to 20, 12.6, 18.6 and 40 m/s2 respectively (figure 8). according to this figure, issn: 2180-1053 vol. 5 no. 1 january-june 2013 journal of mechanical engineering and technology 28 the vibration acceleration value, in the frequency range of 12.5 to 125 hz was higher than exposure limit, 2 m/s2, for hand-arm vibration exposure for 8-hour working per day. however, the mean value of total vibration acceleration in low frequencies, which is the most sensitive frequency of handarm system, has considerable values. at frequencies above 40 hz for all of the fuel blends and engine speeds, the acceleration value is less than 1 m/s2. the frequency less than 1 hz is approximately negligible so the motion sickness doesn’t appear in operators of 13 hp power tiller. in this study it was found that hand-arm acts like a low pass filter and high frequency range of vibration energy, decreased by the fingers and wrist joints and the amplitude of vibration is reduced. this trend is also observed by sam and kathirvel (2006), while they were studding 13 hp power tiller with an empty trailer on transportation. tiller. in this study it was found that hand-arm acts like a low pass filter and high frequency range of vibration energy, decreased by the fingers and wrist joints and the amplitude of vibration is reduced. this trend is also observed by sam and kathirvel (2006), while they were studding 13 hp power tiller with an empty trailer on transportation. figure 8 frequency ranges and magnitudes of hand-transmitted vibration. 5.0 conclusions the following major conclusions can be drawn from the present study: 1. vibration transmissibility is decreased by increasing the engine speed. 2. it shows that the body of operator damps this vibration and so using the engine speed of 1800 rpm may damage to operator health and put them in a high risk of inducing white finger and vascular disease. 3. the amount of vibration acceleration in longitudinal axis is much more than other two axes. 4. the vibration acceleration value, in the frequency range of 8 to 100 hz was higher than a dangerous frequency range of hand-arm vibration transmission. 5. the total weighted acceleration has the maximum value at the frequency of 20, 31.5, 40 and 63 hz with the vibration amplitude reach up to 20, 12.6, 18.6 and 40 m/s2 respectively. 6. the best engine speed are 1400, 1600 and 2000 rpm, in this engine speed using the b20 and b5 seems appropriate. 10 figure 8 frequency ranges and magnitudes of hand-transmitted vibration 5.0 conclusions the following major conclusions can be drawn from the present study: 1. vibration transmissibility is decreased by increasing the engine speed. 2. it shows that the body of operator damps this vibration and so using the engine speed of 1800 rpm may damage to operator health and put them in a high risk of inducing white finger and vascular disease. 3. the amount of vibration acceleration in longitudinal axis is much more than other two axes. 4. the vibration acceleration value, in the frequency range of 8 to 100 hz was higher than a dangerous frequency range of hand-arm vibration transmission. 5. the total weighted acceleration has the maximum value at the frequency of 20, 31.5, 40 and 63 hz with the vibration amplitude reach up to 20, 12.6, 18.6 and 40 m/s2 respectively. 6. the best engine speed are 1400, 1600 and 2000 rpm, in this engine speed using the b20 and b5 seems appropriate. issn: 2180-1053 vol. 5 no. 1 january-june 2013 investigating a power tiller vibration transmissibility using diesel-biodiesel fuel blends on stationary conditions 29 6.0 references agilent technologies, (2002). understanding dynamic signal analysis, fundamental of signal analysis series, application note 1405-2. sam, b., kathirvel, k., (2009). development and evaluation of vibration isolators for reducing hand transmitted vibration of walking and riding type power tillers, biosystems engineering, vol.103, 427– 437. dewangan, k.n, tewari, v.k., (2009). characteristics of hand-transmitted vibration of a hand tractor used in three operational modes, international journal of industrial ergonomics, vol.39:239–245 dewangan, k.n, tewari, v.k., (2008). characteristics of vibration transmission in the hand–arm system and subjective response during field operation of a hand tractor, international journal of biosystem engineering vol.100: 535 – 546. dorado, m.p., ballesteros, e., arnal, j.m., lopez, f.j., (2003). exhaust emissions from a diesel engine fueled with transesterified waste olive oil. journal of fuel, vol.82: 1311-1315. grau, b., bernat, e., antoni, r., jordi-roger, r., rita, p., (2009). small-scale production of straight vegetable oil from rapeseed and its use as biofuel in the spanish territory, energy policy. griffin, m.j., (1996). hand book of human vibration. academic press, london. griffin, m.j., bovenzi, m., nelson, c.m., 2003. dose–response patterns for vibrationinduced white finger. occupational and environmental medicine 60,16–26. goglia,v., gospodaric, z., filipovic, d., djukic, i., (2006). influence on operator’shealth of hand-transmitted vibrations from handles of a single-axle tractor. annals of agricultural and environmental medicine 13, 33–38. ghobadian, b., hassan-beygi, s., (2005). acquisition and processing sound pressure signals of a power tiller, conference of cukurova university, adana, turkey hassan-beygi, s., ghobadian, b., (2005).noise attenuation characteristics of different road surfaces during power tiller transport, agricultural engineering international, cigr e-journal. vol. vii. manuscript pm 04009. hassan-beygi, s., ghobadian, b., amiri, r., kianmehr, m,h., (2009). prediction of power tiller noise levels using a back propagation algorithm, vol.11, p 147-160 hassan-beygi, s., ghobadian, b., kianmehr,m,h., amiri,r., (2007). prediction of a power tiller sound pressure levels in octave frequency bands using artificial neural networks, international journal of agriculture and biology, vol.3, p 494–498 iso 5349-1, (2001). mechanical vibration – guidelines for the measurement and assessment of human exposure to hand-transmitted vibration. part – 1:general requirements. international standard organization, geneva. iso 5349-2, (2001). mechanical vibration – guidelines for the measurement and assessment of human exposure to hand-transmitted vibration. part – 2: practical guidance for measurement at the workplace. international standard organization, geneva. issn: 2180-1053 vol. 5 no. 1 january-june 2013 journal of mechanical engineering and technology 30 iso 8041, (1993). human response to vibrationmeasuring instrumentation standard. international standard organization, geneva. mansfield, n.j., (2005). human response to vibration, crc press, p.244 nurun nabi, md., shamim-akhter, md., shahadat, z., mhia, md., (2006). improvement of engine emissions with conventional diesel feul and diesel-biodiesel blends. journal of biosource technology, vol.97,p 372-378. sarmidi a., (2005). review on biofuel oil and gas production processes from microalgae, energy conversion and management, vol.50, p 1834–1840, 20 taghizadeh, a, hashjin, t. t, ghobadian, b, nikbakht, a. m(2007). evaluation of vibration in power tiller on the asphalt surface, proceedings of the international agricultural engineering conference, bangkok, thailand, 3-6 december 2007. cutting edge technologies and innovations on sustainable resources for world food sufficiency a. taghizadeh, t. tavakoli hashjin, b. ghobadian, a. m. nikbakht, evaluation of vibration in power tiller of the asphalt surface, proceedings of the international agricultural engineering cutting edge technologies and innovations on sustainable resources for world food sufficiency conference, bangkok, thailand, 3-6 december 2007. b. sam, k. kathirvel, development and evaluation of vibration isolators for reducing hand transmitted vibration of walking and riding type power tillers, biosystem journal, 103 427– 437, 2009. b. sam, k. kathirvel, vibration characteristics of walking and riding type power tillers, biosystem engineering, 95:517–528, 2006. b. ghobadian, m. khatamifar, biodiesel fuel production using transesterification of waste vegetable oils, the journal of engine research, p.8-9, 2006. c. carraretto, a. mac, a. mirandola, a. stoppato, s. tonon, biodiesel as alternative fuel: experimental analysis and energetic evaluations, elsevier science, vol.29, p.2195-2211, 2004. c.r. mehta, p.s. tiwari, a.c. varshney, ride vibrations on a 7.5kw rotary power tiller, agricultural engineering, vol.66, p.169-176, 1997. k.n. dewangan, v.k. tewari, characteristics of hand transmitted vibration of a hand tractor used in three operational modes. international journal of industrial ergonomics, 39: 239-245. 2009. k. n. dewangan, v. k. tewari, characteristics of vibration transmission in the hand– arm system and subjective response during field operation of a hand tractor, international journal of biosystem engineering, 100, 535 – 546, 2008. m. p. dorado, e. ballesteros, j. m. arnal, f. j. lopez, exhaust emissions from a diesel engine fuelled with transesterified waste olive oil. journal of fuel, 82, 13111315, 2004. m. j. griffin, hand book of human vibration. academic press, london, 1996. v. goglia, z. gospodaric, d. filipovic, i. djukic, influence on operator’shealth of handtransmitted vibrations from the handles of a single-axle tractor. annals of agricultural and environmental medicine 13 33–38, 2006. issn: 2180-1053 vol. 5 no. 1 january-june 2013 investigating a power tiller vibration transmissibility using diesel-biodiesel fuel blends on stationary conditions 31 s. hassan-beygi, b. ghobadian, noise attenuation characteristics of different road surfaces during power tiller transport, agricultural engineering international, cigr e-journal. vol. vii. manuscript pm 04009, 2005. s. hassan-beygi, b. ghobadian, r. amiri, m. h. kianmehr, prediction of power tiller noise levels using a back propagation algorithm, 11 147-160, 2009. s. hassan-beygi, b. ghobadian, m. h. kianmehr, r. amiri, prediction of a power tiller sound pressure levels in octave frequency bands using artificial neural networks, international journal of agriculture and biology, 3 494–498, 2007. iso 5349-2, mechanical vibration – guidelines for the measurement and assessment of human exposure to hand-transmitted vibration. part – 2: practical guidance for measurement in the workplace. international standard organization, geneva, 2001. j. a. raff, r. d. h. perry, review of vehicle noise studies carried out at the institude of sound and vibration research with a reference to some recent research on petrol engine noise, journal of sound and vibration, 28 (3) 433–470, 1973. n. j. mansfield, human response to vibration, crc press, 244, 2005. o.o. okunribido, m. magnusson, m.h. pope, low back pain in drivers: the relative role of whole-body vibration, posture and manual materials handling, journal of sound and vibration, 298 540–555, 2006. p.s. tiwari, l.p. gite, physiological responses during operation of a rotary tiller, biosystems engineering, vol.82 (2), p.161–168, 2002. s. w. lee, t. her age, b. young, emission reduction potential from the combustion of soy methyl ester fuel blended with petroleum distillate fuel, canment energy technology control, vol.83, p.1607-1613, 2004. v.k. tewari, k.n. dewangan, k.subrata, operator’s fatigue in field operation of hand tractors, biosystem engineering, 89:1-11, 2004. v.m. salokhe, b. majumder, m.s. islam, vibration characteristics of a power tiller. journal of terramechanics, 32:181-196, 1995. issn: 2180-1053 vol. 3 no. 2 july-december 2011 position tracking of slider crank mechanism using pid controller optimized by ziegler nichol’s method 27 position tracking of slider crank mechanism using pid controller optimized by ziegler nichol’s method fauzi ahmad1, ahmad lukman hitam2, khisbullah hudha3, and hishamuddin jamaluddin4 1, 2,3) smart material and automotive control (smac) group center of vehicle research & development (cevred) faculty of mechanical engineering universiti teknikal malaysia melaka (utem) karung berkunci 1200, hang tuah jaya, ayer keroh 75450 melaka, malaysia e-mail1): fauzi.ahmad@utem.edu.my, e-mail2): lukh2u@yahoo.com, e-mail3): khisbullah@utem.edu.my 4) faculty of mechanical engineering, universiti teknologi malaysia (utm), 81310 utm skudai, johor, malaysia e-mail: hishamj@fkm.utm.my abstract this paper presents a study on the position tracking response of a propotional-integral-derivative (pid) controlledslider crank mechanism, which is driven by a two phase stepper motor. in this study, the rod and crank are assumed to be rigid where the newton second law is applied to formulate the equation of motion. a position tracking control of the slider crank mechanism is then developed by using pid controller. several tests such as saw tooth, step function and square function are used in order to examine the performance of the proposed control structure. the results show that, the proposed control structure is able to tracking the desired position with a good response. the slider crank mechanism rig is then developed to investigate experimentally the ability of the proposed controller structure. the results show that the proposed control structure is able to track the desired displacements with acceptable error. keyword: pid control, slider crank mechanism, stepper motor. 23 position tracking of slider crank mechanism using pid controller optimized by ziegler nichol’s method fauzi ahmad1, ahmad lukman hitam2, khisbullah hudha3, and hishamuddin jamaluddin4 1, 2,3) smart material and automotive control (smac) group center of vehicle research & development (cevred) faculty of mechanical engineering universiti teknikal malaysia melaka (utem) karung berkunci 1200, hang tuah jaya, ayer keroh 75450 melaka, malaysia e-mail1): fauzi.ahmad@utem.edu.my, e-mail2): lukh2u@yahoo.com, e-mail3): khisbullah@utem.edu.my 4) faculty of mechanical engineering, universiti teknologi malaysia (utm), 81310 utm skudai, johor, malaysia e-mail: hishamj@fkm.utm.my abstract this paper presents a study on the position tracking response of a propotional-integralderivative (pid) controlledslider crank mechanism, which is driven by a two phase stepper motor. in this study, the rod and crank are assumed to be rigid where the newton second law is applied to formulate the equation of motion. a position tracking control of the slider crank mechanism is then developed by using pid controller. several tests such as saw tooth, step function and square function are used in order to examine the performance of the proposed control structure. the results show that, the proposed control structure is able to tracking the desired position with a good response. the slider crank mechanism rig is then developed to investigate experimentally the ability of the proposed controller structure. the results show that the proposed control structure is able to track the desired displacements with acceptable error. keyword: pid control, slider crank mechanism, stepper motor. 1.0 introduction in mechanical engineering, the slider crank mechanism is a basic structure that have been applied in many usage such as fretsaws, petrol and diesel engines. due to its mechanical coupling, the physical sense is not enough to derive its dynamic equations. jasinski et al.(1971); zhu and chen (1983) and badlani et al. (1979) have solved the steady state solutions of a slider crank few years ago. according to viscomi et al. (1971), the response of slider crank is dependent on length, mass, damping, external piston force and frequency. based on the viewpoints of the ratios, length and speeds of the crank to the connecting rod, the transient responses have been investigated (fung, 1996). recently, issn: 2180-1053 vol. 3 no. 2 july-december 2011 journal of mechanical engineering and technology 28 24 the slider crank mechanisms are actuated by the field oriented control pm synchronous (leonard, 1996; novotny et al. 1996 and lin et al. 1998). the slider crank mechanism driven by pm synchronous is used to transfer rotational motion to translation motion. nowadays, advancements in magnetic materials, semiconductor power devices, and control theory have made the pm synchronous servo motor drive plays a vitally important role in motion-control applications in the low-to-medium power range. the desirable features of the pm synchronous servo motor are its compact structure, high airgap flux density, high power density, high torque-to-inertia ratio, and high torque capability. moreover, compared with an induction servo motor, a pm synchronous servo motor has such advantages as higher efficiency, due to the absence of rotor losses and lower no-load current below the rated speed; and its decoupling control performance is much less sensitive to the parametric variation of the motor (leonard, 1996 and novotny et al. 1996 ). to achieve fast four-quadrant operation and smooth starting and acceleration, the field-oriented control or vector control, is used in the design of the pm synchronous servo motor drive (lin et al. 1998),. however the control performance of the pm synchronous servo motor drive is still influenced by the uncertainties of the controlled plant, which usually comprise unpredictable plant parametric variation, external load disturbances, unmodelled and nonlinear dynamics. during the past decades controlling of slider crank position have resulted with various control strategies to be developed. numerous control methods such as: adaptive control; neural control; and fuzzy control have been studied (visioli, 2001; seng et al. 1999; krohling et al. 2001; mitsukura et al. 1999 and kawabe t et al. 1997). among these the best known is the proportional integral derivative (pid) controller, which has been widely used in the industry because of its simple structure and robust performance within a wide range of operating conditions (huang hp et al. 2002; cominos p et al. 2002; chuang et al. (2006); ahmad et al. (2010) and kristiansson et al. 2002). in this paper, the formulation and dynamic behavior of a pm synchronous motor coupled with a complexity mechanical system is introduced where a slider crank mechanism system actuated by a pm synchronous servo motor is investigated. matlab-simulink software is chosen as a computer simulation tool used to simulate the system‟s behavior and evaluate the performance of the control structure. in controlling the mechanical system with a good response, a pid controller is designed to control the position of the coupled mechanism roller the proposed control structure based feed back control is consists of inner loop controller and outer loop controller (ahmad et al. (2010); kristiansson et al. (2002 yukitom et al., 2004). the inner loop controller is used for position tracking control of the motor actuator while the outer loop controller is used to track the position of the slider shaft. simulation studies for the slider crank mechanism model are presented in order to demonstrate the effectiveness of using the proposed controller. several test have been performed namely sine wave function test, square function test, step function test and saw tooth function test. the simulation results 23 position tracking of slider crank mechanism using pid controller optimized by ziegler nichol’s method fauzi ahmad1, ahmad lukman hitam2, khisbullah hudha3, and hishamuddin jamaluddin4 1, 2,3) smart material and automotive control (smac) group center of vehicle research & development (cevred) faculty of mechanical engineering universiti teknikal malaysia melaka (utem) karung berkunci 1200, hang tuah jaya, ayer keroh 75450 melaka, malaysia e-mail1): fauzi.ahmad@utem.edu.my, e-mail2): lukh2u@yahoo.com, e-mail3): khisbullah@utem.edu.my 4) faculty of mechanical engineering, universiti teknologi malaysia (utm), 81310 utm skudai, johor, malaysia e-mail: hishamj@fkm.utm.my abstract this paper presents a study on the position tracking response of a propotional-integralderivative (pid) controlledslider crank mechanism, which is driven by a two phase stepper motor. in this study, the rod and crank are assumed to be rigid where the newton second law is applied to formulate the equation of motion. a position tracking control of the slider crank mechanism is then developed by using pid controller. several tests such as saw tooth, step function and square function are used in order to examine the performance of the proposed control structure. the results show that, the proposed control structure is able to tracking the desired position with a good response. the slider crank mechanism rig is then developed to investigate experimentally the ability of the proposed controller structure. the results show that the proposed control structure is able to track the desired displacements with acceptable error. keyword: pid control, slider crank mechanism, stepper motor. 1.0 introduction in mechanical engineering, the slider crank mechanism is a basic structure that have been applied in many usage such as fretsaws, petrol and diesel engines. due to its mechanical coupling, the physical sense is not enough to derive its dynamic equations. jasinski et al.(1971); zhu and chen (1983) and badlani et al. (1979) have solved the steady state solutions of a slider crank few years ago. according to viscomi et al. (1971), the response of slider crank is dependent on length, mass, damping, external piston force and frequency. based on the viewpoints of the ratios, length and speeds of the crank to the connecting rod, the transient responses have been investigated (fung, 1996). recently, issn: 2180-1053 vol. 3 no. 2 july-december 2011 position tracking of slider crank mechanism using pid controller optimized by ziegler nichol’s method 29 25 show that the use of the proposed pid control technique proved to be effective in controlling the position of the slider crank with a good accuracy. since the explorations of the proposed controller have been done in the simulation study, the slider crank mechanism rig is needed to investigate experimentally the capability of the proposed controller. the mechanism of the slider crank system is consists of an actuator such as stepper motor to actuate the slider crank in a real slider mechanism system. the result of the experiment studies show that the control technique is able to track the desired position with a small deviation and acceptable error. the remainder of this paper is organized as follows: the first section contains the introduction and the review of some related works, followed by mathematical derivations of slider crank kinematics model with stepper motor model in the second section. the third section presents the proposed control structure for the position tracking control of slider crank mechanism system. the fourth section will explain about the performance evaluation of the proposed control structure. the following section will discuss about the setup and the performance of the slider crank mechanism in the experimental study and the last section contains some conclusion. 2.0 slider crank mechanism modeling the slider crank mechanism is a basic structure in mechanical application. it is also widely used in practical application (nagchaudhuri, 2002; fung et al., 1999; ranjbarkohan et al., 2011). for examples, fretsaws, petrol and diesel engines are the typical application of velocity control. hence, the slider crank mechanism considered in this study is based on the basic operation of engine which consists of a crankshaft, r, connecting rod, l, and piston. the purpose of the slider-crank mechanism is to convert rotational motion of the crankshaft to the linear motion of the piston. like shown in figure 1, the kinematic of slider crank mechanism can be described in equation 1 to 7. the piston displacement from top dead centre, x, can be determined from the geometry of the mechanism, in terms of the lengths of the con-rod, l, and crank, r, and the crank angle, . from the geometry and noting that = = 0 when x = 0, x can be expressed as: (1) issn: 2180-1053 vol. 3 no. 2 july-december 2011 journal of mechanical engineering and technology 30 26 also from the geometry, it can be seen that (2) and [ ] [ ] (3) substituting for from equation 2 in equation 3 and leaves as the only variable on the right hand side of the expression, [ ] [ ] (4) equation 4 can be substituted into equation 1 to obtain the kinematic equation for the slider crank mechanism such as equation 5, √ [ ] (5) equation 5 can then be rearranged by introducing another parameter, n, the ratio of the length of the conrod, l, to the radius of crankshaft, r, as: { [ √ ( ) ]} (6) where (7) the values of parameters r and n are determined by measurement of the slider crank mechanism. the technical specifications of the slider crank mechanism are listed in table 1. table 1: slider crank mechanism parameters parameters value r 100mm l 300mm table 2: stepper motor model parameters table 1. table 1 issn: 2180-1053 vol. 3 no. 2 july-december 2011 position tracking of slider crank mechanism using pid controller optimized by ziegler nichol’s method 31 27 parameters value operating value u 24 v static holding torque tm 0.5 nm winding resistance r 100 ohms electric time constant e 5 ms stepping angle 1.8 degree numbers of pole pairs n 50 load inertia j 1.0 kgm2 2.1 stepper motor modeling in this study, the slider crank mechanism is driven by two phase stepper motor which is used to provide the rotational motion to the crank shaft. the stepper motor is consisting of one stator side and one rotor with one pole pair that function as the permanent magnet. when the windings of one phase are energized, a magnetic dipole is generated on the stator side. the basic principle of the stepper motor is given in figure 2. for example, if phase 2 is active, winding 3 produces an electrical south pole and winding 4 produces an electrical north pole. the number of steps per revolution of the rotor is calculated as: (8) where n is the number of rotor pole pair and m is the number of stator phases. for the hybrid stepper motor, n is half number of rotor teeth. the stepping angle is (9) for example, if n=1 and m=2, it will have 4 steps per revolution and stepping angle is 90 degrees. if sinusoidal characteristic of the magnetic field in the air gap is assumed, the contribution of each phase j on the motor torque tmj can be written as [ ( ) ] ( ) (10) where motor constant, depending on the design of the rotor 26 also from the geometry, it can be seen that (2) and [ ] [ ] (3) substituting for from equation 2 in equation 3 and leaves as the only variable on the right hand side of the expression, [ ] [ ] (4) equation 4 can be substituted into equation 1 to obtain the kinematic equation for the slider crank mechanism such as equation 5, √ [ ] (5) equation 5 can then be rearranged by introducing another parameter, n, the ratio of the length of the conrod, l, to the radius of crankshaft, r, as: { [ √ ( ) ]} (6) where (7) the values of parameters r and n are determined by measurement of the slider crank mechanism. the technical specifications of the slider crank mechanism are listed in table 1. table 1: slider crank mechanism parameters parameters value r 100mm l 300mm table 2: stepper motor model parameters 28 ( ) actual motor position location of the coil j in the stator ( ) the current in the coil as function of time. the current ( ) in the coil is a function of the supplied voltage and the coil properties. a general between and ( ) is given by: ( ) ( ) (11) where the electronic force induced in the phase j r the resistance of the coil l the inductance of the coil the emf in each coil can be expressed as: [ ( ) ] (12) resistance and inductance of all coils in the motor are the same so that no indices are required. the differential equation can be expressed in the laplace domain as shown in the equation below: (13) the total torque produced by the stepper is : ∑ (14) considering the equation of motion of the stepper motor ∑ (15) where j the inertia of the rotor and the load d the viscous damping constant tf frictional load torque where is the rotational velocity of the rotor and the data of the stepper motor is given in the table 2. table 2 issn: 2180-1053 vol. 3 no. 2 july-december 2011 journal of mechanical engineering and technology 32 28 ( ) actual motor position location of the coil j in the stator ( ) the current in the coil as function of time. the current ( ) in the coil is a function of the supplied voltage and the coil properties. a general between and ( ) is given by: ( ) ( ) (11) where the electronic force induced in the phase j r the resistance of the coil l the inductance of the coil the emf in each coil can be expressed as: [ ( ) ] (12) resistance and inductance of all coils in the motor are the same so that no indices are required. the differential equation can be expressed in the laplace domain as shown in the equation below: (13) the total torque produced by the stepper is : ∑ (14) considering the equation of motion of the stepper motor ∑ (15) where j the inertia of the rotor and the load d the viscous damping constant tf frictional load torque where is the rotational velocity of the rotor and the data of the stepper motor is given in the table 2. 29 2.2 description of the simulation model the slider crank simulation model was developed based on the mathematical equations presented in the previous section by using matlab simulink software. the relationship between slider crank mechanism and stepper motor are clearly described in figure 3. there are two inputs that can be used in the analysis of the slider crank namely torque input and position input which come from the stepper motor. but in this study, the position of rotor is used to become the input to the slider crank mechanism. it simply explains that the model created is able to perform the position tracking control analysis of the slider crank mechanism. figure 3: slider crank mechanism model in matlab simulink software 3.0 position tracking control of slider crank mechanism table 2. issn: 2180-1053 vol. 3 no. 2 july-december 2011 position tracking of slider crank mechanism using pid controller optimized by ziegler nichol’s method 33 29 2.2 description of the simulation model the slider crank simulation model was developed based on the mathematical equations presented in the previous section by using matlab simulink software. the relationship between slider crank mechanism and stepper motor are clearly described in figure 3. there are two inputs that can be used in the analysis of the slider crank namely torque input and position input which come from the stepper motor. but in this study, the position of rotor is used to become the input to the slider crank mechanism. it simply explains that the model created is able to perform the position tracking control analysis of the slider crank mechanism. figure 3: slider crank mechanism model in matlab simulink software 3.0 position tracking control of slider crank mechanism 30 there are two loops used in the controller structure which are inner loop and outer loop controller. the inner loop controller is used to evaluate the deviation from the commanded position and the encoder which detects the position of the rotor with robust and accuracy tracking performance and an outer loop controller is used as the position controller for the tracking of periodic reference inputs. the proposed control structure is shown in figure 4. figure 4: the proposed control structure for position tracking control of slider crank mechanism in this control scheme, the proportional plus integral plus derivative (pid) controller is used in the outer loop control to isolate the piston displacement from top dead centre, x. the reason for using pid controller is because the pid controller has already proven effective in many applications where it is easy to maintain and easy to implement in the real system (chuang et al., 2006). the pid controller that applied in the system mathematically can be described by equation (16). ( ) ( ) ( )∫ ( ) ( ) ( ) (16) where ( ) ( ) ( ) and is the piston actual displacement, is the piston desired displacement, the proportional gain, ( ), integral gain, ( ) and derivative gain, ( ) are the function of the position error piston displacement. even pid controllers are probably the most commonly used controller structures in industry, however it present some challenges to control and instrumentation in the aspect of tuning of the gains required for stability and good transient performance. because of that, there are several prescriptive rules used in pid tuning such as ziegler-nichole‟s method. 3.1 ziegler nicholes for auto tuning 29 2.2 description of the simulation model the slider crank simulation model was developed based on the mathematical equations presented in the previous section by using matlab simulink software. the relationship between slider crank mechanism and stepper motor are clearly described in figure 3. there are two inputs that can be used in the analysis of the slider crank namely torque input and position input which come from the stepper motor. but in this study, the position of rotor is used to become the input to the slider crank mechanism. it simply explains that the model created is able to perform the position tracking control analysis of the slider crank mechanism. figure 3: slider crank mechanism model in matlab simulink software 3.0 position tracking control of slider crank mechanism figure 3 figure 4. issn: 2180-1053 vol. 3 no. 2 july-december 2011 journal of mechanical engineering and technology 34 30 there are two loops used in the controller structure which are inner loop and outer loop controller. the inner loop controller is used to evaluate the deviation from the commanded position and the encoder which detects the position of the rotor with robust and accuracy tracking performance and an outer loop controller is used as the position controller for the tracking of periodic reference inputs. the proposed control structure is shown in figure 4. figure 4: the proposed control structure for position tracking control of slider crank mechanism in this control scheme, the proportional plus integral plus derivative (pid) controller is used in the outer loop control to isolate the piston displacement from top dead centre, x. the reason for using pid controller is because the pid controller has already proven effective in many applications where it is easy to maintain and easy to implement in the real system (chuang et al., 2006). the pid controller that applied in the system mathematically can be described by equation (16). ( ) ( ) ( )∫ ( ) ( ) ( ) (16) where ( ) ( ) ( ) and is the piston actual displacement, is the piston desired displacement, the proportional gain, ( ), integral gain, ( ) and derivative gain, ( ) are the function of the position error piston displacement. even pid controllers are probably the most commonly used controller structures in industry, however it present some challenges to control and instrumentation in the aspect of tuning of the gains required for stability and good transient performance. because of that, there are several prescriptive rules used in pid tuning such as ziegler-nichole‟s method. 3.1 ziegler nicholes for auto tuning 31 in 1942 ziegler and nichols, both employees of taylor instruments, described simple mathematical procedures for tuning pid controller. the procedures are now accepted as standard in control systems practice. the ziegler-nichols formulae for specifying the controllers are based on plant step responses. the method is applied to plants with step responses of the form displayed in figure 5. this type of response is typical of a first order system with transportation delay, such as that induced by fluid flow from a tank along a pipe line. it is also typical of a plant made up of a series of first order systems. the response is characterized by two parameters, l the delay time and t the time constant. these are found by drawing a tangent to the step response at its point of inflection and noting its intersections with the time axis and the steady state value. the plant model is therefore: ( ) (17) figure 5: response curve for ziegler-nichols method based on the equation, ziegler and nichols derived the control parameters such as in table 3: table 3: ziegler and nichols control parameters pid type kp ti= kp/ki td=kd/kp p ∞ 0 pi 0.9 0 pid 1.2 2l 0.5l figure 4 issn: 2180-1053 vol. 3 no. 2 july-december 2011 position tracking of slider crank mechanism using pid controller optimized by ziegler nichol’s method 35 31 in 1942 ziegler and nichols, both employees of taylor instruments, described simple mathematical procedures for tuning pid controller. the procedures are now accepted as standard in control systems practice. the ziegler-nichols formulae for specifying the controllers are based on plant step responses. the method is applied to plants with step responses of the form displayed in figure 5. this type of response is typical of a first order system with transportation delay, such as that induced by fluid flow from a tank along a pipe line. it is also typical of a plant made up of a series of first order systems. the response is characterized by two parameters, l the delay time and t the time constant. these are found by drawing a tangent to the step response at its point of inflection and noting its intersections with the time axis and the steady state value. the plant model is therefore: ( ) (17) figure 5: response curve for ziegler-nichols method based on the equation, ziegler and nichols derived the control parameters such as in table 3: table 3: ziegler and nichols control parameters pid type kp ti= kp/ki td=kd/kp p ∞ 0 pi 0.9 0 pid 1.2 2l 0.5l 32 4.0 performance assesment of the proposed control structure to the position tracking control of slider crank mechanism the performance of the pid controller in tracking the desired position is examined through simulation studies using simulink toolbox of the matlab software package. for comparison purposes, the performance of the proposed pid control structure is compared with the desired displacement which is the reference. the desired specifications are settling time = 0.5 sec, rising time =0.25 sec, maximum overshoot < 5% and steady state error < 1%. 4.1 simulation parameters the simulation study was performed for a period of 10 seconds using heun solver with a fixed step size of 0.01 second. the controller parameters are obtained using trial and error technique as shown in table 4. the numerical values of the slider crank model are defined in table 1 and the stepper motor model parameters are as in table 2 adopted from morar, (2003). table 4: controller parameters pid value kp 995 ki 79030 kd 0 4.2 simulation result figures 6(a) to figure 6(b) show the response for system with the pid controller. the parameters of the pid controller is optimized using ziegler nichols method for optimal performance under various condition. several test procedures such as step function, sine wave function, square function and saw tooth function are applied to verified the effectiveness of the control structure. all the input signals are generated from „step function‟ and „signal generator‟ block with the amplitude set at 0.03m and frequency at 0.03 hz.the results of the simulation runs corresponding to the trajectory profiles of the slider as discussed before. with appropriate tuning of the pid gains, excellent results are achieved as illustrated in figures 6(a) to figures 6(d). in the graphs, the dashed line corresponds to the desired motion of the slider position at the 0.03m and the solid line indicates the actual motion achieved by controlling the position piston of the slider crank. it can be seen that the proposed control structure with pid controller in driving the crank is very encouraging as shown in step function response in figure 6(a). in term of sine wave function, figure 6(b), saw tooth function, figure 6(c), and square function, figure 6(d) the controller structure shows it‟s figure 5 table 3 issn: 2180-1053 vol. 3 no. 2 july-december 2011 journal of mechanical engineering and technology 36 32 4.0 performance assesment of the proposed control structure to the position tracking control of slider crank mechanism the performance of the pid controller in tracking the desired position is examined through simulation studies using simulink toolbox of the matlab software package. for comparison purposes, the performance of the proposed pid control structure is compared with the desired displacement which is the reference. the desired specifications are settling time = 0.5 sec, rising time =0.25 sec, maximum overshoot < 5% and steady state error < 1%. 4.1 simulation parameters the simulation study was performed for a period of 10 seconds using heun solver with a fixed step size of 0.01 second. the controller parameters are obtained using trial and error technique as shown in table 4. the numerical values of the slider crank model are defined in table 1 and the stepper motor model parameters are as in table 2 adopted from morar, (2003). table 4: controller parameters pid value kp 995 ki 79030 kd 0 4.2 simulation result figures 6(a) to figure 6(b) show the response for system with the pid controller. the parameters of the pid controller is optimized using ziegler nichols method for optimal performance under various condition. several test procedures such as step function, sine wave function, square function and saw tooth function are applied to verified the effectiveness of the control structure. all the input signals are generated from „step function‟ and „signal generator‟ block with the amplitude set at 0.03m and frequency at 0.03 hz.the results of the simulation runs corresponding to the trajectory profiles of the slider as discussed before. with appropriate tuning of the pid gains, excellent results are achieved as illustrated in figures 6(a) to figures 6(d). in the graphs, the dashed line corresponds to the desired motion of the slider position at the 0.03m and the solid line indicates the actual motion achieved by controlling the position piston of the slider crank. it can be seen that the proposed control structure with pid controller in driving the crank is very encouraging as shown in step function response in figure 6(a). in term of sine wave function, figure 6(b), saw tooth function, figure 6(c), and square function, figure 6(d) the controller structure shows it‟s 33 capability in achieving control design criteria such as discussed in section 4.0 via providing a good response in tracking the desired position.. (a). step function (b). sine wave function (c). saw tooth function (d). square function figure 6: simulation responses of the position tracking control of the slider crank mechanism 5. performance assesment of the proposed control structure using hardware in the loop simulations of slider crank mechanism in order to demonstrate the effectiveness of the proposed control rule, slider crank mechanism rig has been setup as shown in figure 7.the experimental instrument of slider crank is divided into five parts such as actuator, slider crank, controller, host pc and target pc. in this system, the simulation of the slider crank is simulated in the host pc, while the host pc is used to give the direction to the target pc to interact with the hardware which means the slider crank mechanism. in this study, the slider crank is coupled with the stepper motor that consists of a stepper motor driver. the driver is worked on 2-phase, 220 v and 60 hz. to measure the translation position, a sensor namely linear variable displacement transducer (lvdt) is used. the output of the lvdt is 0~5v, which mapped to real translation position is 0~0.2m. on the other hand, another table 4 table 1 figures issn: 2180-1053 vol. 3 no. 2 july-december 2011 position tracking of slider crank mechanism using pid controller optimized by ziegler nichol’s method 37 33 capability in achieving control design criteria such as discussed in section 4.0 via providing a good response in tracking the desired position.. (a). step function (b). sine wave function (c). saw tooth function (d). square function figure 6: simulation responses of the position tracking control of the slider crank mechanism 5. performance assesment of the proposed control structure using hardware in the loop simulations of slider crank mechanism in order to demonstrate the effectiveness of the proposed control rule, slider crank mechanism rig has been setup as shown in figure 7.the experimental instrument of slider crank is divided into five parts such as actuator, slider crank, controller, host pc and target pc. in this system, the simulation of the slider crank is simulated in the host pc, while the host pc is used to give the direction to the target pc to interact with the hardware which means the slider crank mechanism. in this study, the slider crank is coupled with the stepper motor that consists of a stepper motor driver. the driver is worked on 2-phase, 220 v and 60 hz. to measure the translation position, a sensor namely linear variable displacement transducer (lvdt) is used. the output of the lvdt is 0~5v, which mapped to real translation position is 0~0.2m. on the other hand, another 34 one sensor is employed to measure the angular position of the crank namely angular encoder. the used of the encoder is to give feed back control in the inner loop control. the data acquisition system namely national instrument (ni) interface card (advantech co., pcl-1800) is installed in the isa bus to handle the a/d and d/a process. figure 7: slider crank mechanism experimental setup 5.1 experimental result based on the same requirement of simulation, the experimental results are shown in figures 8(a) to 8(d) with the differences signal applied for the test. from the graphs, it is necessary to note that the dashed line is the desired position, while the solid line is the response of the system. figure 8(a) shows the step function responses of the system. it can be seen that the trends between the desired position and actual responses are slightly similar but didn‟t fulfill the control design criteria as discussed in section 4.0. the control strategy is over damp where the rise time of the system is more then 0.25 second, no over shoot and the steady state error is more then 1 %. from observation, this is happened because of the frictions that happened between the piston with the bearing where the surface of the piston is rough. other than that, the control optimization was also the most factors contributing to this problem. the parameters of the controller were optimized (with assume) the system is in ideal conditions and no friction happen at all the joints and the touching surface. in term of sine wave function in figure 8(b) and saw tooth function in figure 8(c), the system shows the tendencies to follow the desired position with the similar shape but have a little bit differences in term of magnitude. these differences are caused by the friction that lowering down the speed of the system where the system was not yet finished implementing the old instruction but already arrived the new order from target figure 6 issn: 2180-1053 vol. 3 no. 2 july-december 2011 journal of mechanical engineering and technology 38 34 one sensor is employed to measure the angular position of the crank namely angular encoder. the used of the encoder is to give feed back control in the inner loop control. the data acquisition system namely national instrument (ni) interface card (advantech co., pcl-1800) is installed in the isa bus to handle the a/d and d/a process. figure 7: slider crank mechanism experimental setup 5.1 experimental result based on the same requirement of simulation, the experimental results are shown in figures 8(a) to 8(d) with the differences signal applied for the test. from the graphs, it is necessary to note that the dashed line is the desired position, while the solid line is the response of the system. figure 8(a) shows the step function responses of the system. it can be seen that the trends between the desired position and actual responses are slightly similar but didn‟t fulfill the control design criteria as discussed in section 4.0. the control strategy is over damp where the rise time of the system is more then 0.25 second, no over shoot and the steady state error is more then 1 %. from observation, this is happened because of the frictions that happened between the piston with the bearing where the surface of the piston is rough. other than that, the control optimization was also the most factors contributing to this problem. the parameters of the controller were optimized (with assume) the system is in ideal conditions and no friction happen at all the joints and the touching surface. in term of sine wave function in figure 8(b) and saw tooth function in figure 8(c), the system shows the tendencies to follow the desired position with the similar shape but have a little bit differences in term of magnitude. these differences are caused by the friction that lowering down the speed of the system where the system was not yet finished implementing the old instruction but already arrived the new order from target 35 pc. in term of square function, (figure 8(d)), the result shows that, the control structure try to force the system (following) the desired position and as a result, there is a small difference in term of trends and magnitude. again, this is due to the frictions that occur in the system. this is due to the fact that, during the simulation, the efficiency of the stepper motor‟s rotating shaft, the mass of connecting rod and the sliding shaft or piston were ignored and neglected while in the actual condition, all of these should be taken into account. (a). step functions (b) sine wave function (c). saw tooth function (d). square function figure 8: response of hardware in the lop simulation of position tracking control of slider crank mechanism 6.0 conclusion as a conclusion, the kinematics model of slider crank mechanism has been developed and integrated with stepper motor model by using matlab simulink software. the position tracking control of the slider crank mechanism is then has been developed which consist of two close loop function namely inner loop and outer loop controller. the inner loop controller is used to command the stepper motor to give the necessary motion such desired by the reference, while the outer loop control structure is used to instruct the slider crank to give the required position as needed. in this study, the proportionalintegral-derivative (pid) controller is used as the controller strategy. the reason for figures issn: 2180-1053 vol. 3 no. 2 july-december 2011 position tracking of slider crank mechanism using pid controller optimized by ziegler nichol’s method 39 35 pc. in term of square function, (figure 8(d)), the result shows that, the control structure try to force the system (following) the desired position and as a result, there is a small difference in term of trends and magnitude. again, this is due to the frictions that occur in the system. this is due to the fact that, during the simulation, the efficiency of the stepper motor‟s rotating shaft, the mass of connecting rod and the sliding shaft or piston were ignored and neglected while in the actual condition, all of these should be taken into account. (a). step functions (b) sine wave function (c). saw tooth function (d). square function figure 8: response of hardware in the lop simulation of position tracking control of slider crank mechanism 6.0 conclusion as a conclusion, the kinematics model of slider crank mechanism has been developed and integrated with stepper motor model by using matlab simulink software. the position tracking control of the slider crank mechanism is then has been developed which consist of two close loop function namely inner loop and outer loop controller. the inner loop controller is used to command the stepper motor to give the necessary motion such desired by the reference, while the outer loop control structure is used to instruct the slider crank to give the required position as needed. in this study, the proportionalintegral-derivative (pid) controller is used as the controller strategy. the reason for 36 using pid controller is because the pid controller has already proven effective in many applications where it is easy to maintain and easy to implement in the online system. simulation studies for the slider crank mechanism model are presented to demonstrate the effectiveness of using the proposed controller. several tests have been performed in order to verified the effectiveness of the proposed controller namely sine wave function test, square function test, step function test and saw tooth function test. the simulation results show that the use of the proposed pid control technique proved to be effective in controlling the position of the slider crank with a good accuracy. a slider crank mechanism test rig is then has been developed to validate experimentally the efficiency of the proposed control technique. same testing methods have been implemented and the results show that the control technique is able to track the desired position with a small deviation and acceptable error. 7.0 references ahmad, f.,imaduddin, f., hudha, k. and jamaluddin, h. (2010). modelling, validation and adaptive pid control with pitch moment rejection of active suspension system for reducing unwanted vehicle motion in longitudinal direction. international journal of vehicle system modeling and testing (ijvsmt) vol. 5 no. 4 pp. 312-346, inderscience publishers, uk viscomi, b. v. and arye, r. s. (1971). nonlinear dynamic response of elastic slider crank mechanism,” asme j. eng. ind., pp. 251-262. boot, r., richert, j. (1998). automated test of ecus in a hardware-in-the-loop simulation environment. asim 1998 12th symposium on simulation technology, chuang, c. w., lee, c. d. and huang, c. l. (2006). applying experienced self-tuning pid control to position control of slider crank mechanisms. international symposium on power electronics, electrical drives, automation and motion, 2006. (speedam 2006). taormina, sicily may 2006, pp. 652 – 657, isbn: 1-4244-0193-3 cominos, p, munro n. (2002). pid controllers: recent tuning methods and design to specification. iee proceeding on control theory application. pp. 46–53. novotny, d. w. and lipo, t. a. (1996). slider crank mechanism control using adaptive computed torque technique “vector control and dynamics of ac drivers,” oxford university, lin f.i., lin y.s. and chiu s.l. slider-crank mechanism control using adaptive computed iorque technique, jee proceedings control theory and applications, 1998, pp.364-276. hanselmann, h. (1993). hardware-in-the-loop simulation as a standard approach for development, customization, and production test. sae 1993-930207 huang, h.p., roan, m.l. and jeng, j.c. (2002). on-line adaptive tuning for pid controllers”. ieee proceeding on control theory application 2002, pp. 60–72. kawabe, t. and tagami, t. (1997). a real coded genetic algorithm for matrix inequality design approach of robust pid controller with two degrees of freedom”. ieee proceeding on 12th international symposium intelligent control. istanbul, turkey, july 1997. pp. 119– 24. issn: 2180-1053 vol. 3 no. 2 july-december 2011 journal of mechanical engineering and technology 40 36 using pid controller is because the pid controller has already proven effective in many applications where it is easy to maintain and easy to implement in the online system. simulation studies for the slider crank mechanism model are presented to demonstrate the effectiveness of using the proposed controller. several tests have been performed in order to verified the effectiveness of the proposed controller namely sine wave function test, square function test, step function test and saw tooth function test. the simulation results show that the use of the proposed pid control technique proved to be effective in controlling the position of the slider crank with a good accuracy. a slider crank mechanism test rig is then has been developed to validate experimentally the efficiency of the proposed control technique. same testing methods have been implemented and the results show that the control technique is able to track the desired position with a small deviation and acceptable error. 7.0 references ahmad, f.,imaduddin, f., hudha, k. and jamaluddin, h. (2010). modelling, validation and adaptive pid control with pitch moment rejection of active suspension system for reducing unwanted vehicle motion in longitudinal direction. international journal of vehicle system modeling and testing (ijvsmt) vol. 5 no. 4 pp. 312-346, inderscience publishers, uk viscomi, b. v. and arye, r. s. (1971). nonlinear dynamic response of elastic slider crank mechanism,” asme j. eng. ind., pp. 251-262. boot, r., richert, j. (1998). automated test of ecus in a hardware-in-the-loop simulation environment. asim 1998 12th symposium on simulation technology, chuang, c. w., lee, c. d. and huang, c. l. (2006). applying experienced self-tuning pid control to position control of slider crank mechanisms. international symposium on power electronics, electrical drives, automation and motion, 2006. (speedam 2006). taormina, sicily may 2006, pp. 652 – 657, isbn: 1-4244-0193-3 cominos, p, munro n. (2002). pid controllers: recent tuning methods and design to specification. iee proceeding on control theory application. pp. 46–53. novotny, d. w. and lipo, t. a. (1996). slider crank mechanism control using adaptive computed torque technique “vector control and dynamics of ac drivers,” oxford university, lin f.i., lin y.s. and chiu s.l. slider-crank mechanism control using adaptive computed iorque technique, jee proceedings control theory and applications, 1998, pp.364-276. hanselmann, h. (1993). hardware-in-the-loop simulation as a standard approach for development, customization, and production test. sae 1993-930207 huang, h.p., roan, m.l. and jeng, j.c. (2002). on-line adaptive tuning for pid controllers”. ieee proceeding on control theory application 2002, pp. 60–72. kawabe, t. and tagami, t. (1997). a real coded genetic algorithm for matrix inequality design approach of robust pid controller with two degrees of freedom”. ieee proceeding on 12th international symposium intelligent control. istanbul, turkey, july 1997. pp. 119– 24. 37 kristiansson, b. and lennartson. b. (2002). robust and optimal tuning of pi and pid controllers. ieee proceedingon control theory application. pp. 46–53. krohling, r. a. and rey, j.p. (2001).design of optimal disturbance rejection pid controllers using genetic algorithm. ieee transaction on evolution computer 2001;vol. 5, pp.78–82. badlani, m. and kleinhenz, w. (1979). dynamic stability of elastic mechanisms. journal ofmechanism des. pp. 149-153. mitsukura, y., yamamoto, t. and kaneda, m. (1999). a design of self-tuning pid controllers using a genetic algorithm. proceeding of am control confefernce, san diego, ca, june 1999. vol. 5. pp. 1361. morar, a., (2003). stepper motor model for dynamic simulation. acta electrotehnica. vol. 44(2), pp. 117-122. jasinski, p. w., lee, h.c. and sandor, g. n. (1971).vibrations of elastic connecting rod of a high-speed slider crank mechanism. asme j. eng. ind., pp. 636-644. park, t. j., oh, s. w., jang, j. h. and han, c.s. (2002). the design of a controller for the steerby-wire system using the hardware-in-the-loop-simulation system. sae automotive dynamics and stability.no.01, pp.1596. fung, r. f. (1996). dynamic analysis of the flexible connecting rod of a slider crank mechanism. asme j. vibration acoust. pp. 687-689, seng, t. l., khalid, m. b., yusof, r. (1999). tuning of a neuro-fuzzy controller by genetic algorithm. ieee trans on syst man cybern b 1999. no.29, pp.226–36. terwiesch, p., keller, t. and scheiben, e. (1999). rail vehicle control system integration testing using digital hardware-in-the-loop simulation. ieee trans control syst technol 1999;7(3):352–62. visioli a. (2001) tuning of pid controllers with fuzzy logic. proceeding of electrical engineering control theory application.vol.148, no. 1, pp.1–8. yukitomo, m., shigemasa, t. and baba, y. and kojima, f. (2004). a two degrees of freedom pid control system, its features and applications. 2004 5th asian control conference. melbourne, australia. vol. 1, pp: 456isbn: 0-7803-8873-9, july 2004 zhu, z. g. and chen, y. (1983). the stability of the motion of a connecting rod,” j. mechanisms, transmissions, automation des. pp. 637-640, nagchaudhuri, a. (2002). mechatronic redesign of slider crank mechanism. proceeding of imece2002, asme international mechanical engineering congress & exposition. no. imece2002-32482. november 17-22, new orleans, louisiana united state of america. fung, r. f., chen, k. w. and yen, j. y. (1999). fuzzy sliding mode controlled slider crank mechanism using a pm synchronous servo motor drive. international journal of mechanical sciences . no. 41(1999), pp. 337-355. ranjbarkohan, m., rasekh, m., hoseini, a. h., kheiralipour, k. and asadi, m. r. (2011). kinematics and kinetic analysis of the slider-crank mechanism in otto linear four cylinder z24 engine. journal of mechanical engineering research vol. 3(3), pp. 85-95, march 2011 issn: 2180-1053 vol. 3 no. 2 july-december 2011 position tracking of slider crank mechanism using pid controller optimized by ziegler nichol’s method 41 37 kristiansson, b. and lennartson. b. (2002). robust and optimal tuning of pi and pid controllers. ieee proceedingon control theory application. pp. 46–53. krohling, r. a. and rey, j.p. (2001).design of optimal disturbance rejection pid controllers using genetic algorithm. ieee transaction on evolution computer 2001;vol. 5, pp.78–82. badlani, m. and kleinhenz, w. (1979). dynamic stability of elastic mechanisms. journal ofmechanism des. pp. 149-153. mitsukura, y., yamamoto, t. and kaneda, m. (1999). a design of self-tuning pid controllers using a genetic algorithm. proceeding of am control confefernce, san diego, ca, june 1999. vol. 5. pp. 1361. morar, a., (2003). stepper motor model for dynamic simulation. acta electrotehnica. vol. 44(2), pp. 117-122. jasinski, p. w., lee, h.c. and sandor, g. n. (1971).vibrations of elastic connecting rod of a high-speed slider crank mechanism. asme j. eng. ind., pp. 636-644. park, t. j., oh, s. w., jang, j. h. and han, c.s. (2002). the design of a controller for the steerby-wire system using the hardware-in-the-loop-simulation system. sae automotive dynamics and stability.no.01, pp.1596. fung, r. f. (1996). dynamic analysis of the flexible connecting rod of a slider crank mechanism. asme j. vibration acoust. pp. 687-689, seng, t. l., khalid, m. b., yusof, r. (1999). tuning of a neuro-fuzzy controller by genetic algorithm. ieee trans on syst man cybern b 1999. no.29, pp.226–36. terwiesch, p., keller, t. and scheiben, e. (1999). rail vehicle control system integration testing using digital hardware-in-the-loop simulation. ieee trans control syst technol 1999;7(3):352–62. visioli a. (2001) tuning of pid controllers with fuzzy logic. proceeding of electrical engineering control theory application.vol.148, no. 1, pp.1–8. yukitomo, m., shigemasa, t. and baba, y. and kojima, f. (2004). a two degrees of freedom pid control system, its features and applications. 2004 5th asian control conference. melbourne, australia. vol. 1, pp: 456isbn: 0-7803-8873-9, july 2004 zhu, z. g. and chen, y. (1983). the stability of the motion of a connecting rod,” j. mechanisms, transmissions, automation des. pp. 637-640, nagchaudhuri, a. (2002). mechatronic redesign of slider crank mechanism. proceeding of imece2002, asme international mechanical engineering congress & exposition. no. imece2002-32482. november 17-22, new orleans, louisiana united state of america. fung, r. f., chen, k. w. and yen, j. y. (1999). fuzzy sliding mode controlled slider crank mechanism using a pm synchronous servo motor drive. international journal of mechanical sciences . no. 41(1999), pp. 337-355. ranjbarkohan, m., rasekh, m., hoseini, a. h., kheiralipour, k. and asadi, m. r. (2011). kinematics and kinetic analysis of the slider-crank mechanism in otto linear four cylinder z24 engine. journal of mechanical engineering research vol. 3(3), pp. 85-95, march 2011 03(27-42).pdf issn: 2180-1053 vol. 3 no. 2 july-december 2011 abrasive wear behavior of al6061frit particulate composites 43 abrasive wear behavior of al6061frit particulate composites d.ramesh1*, r.p.swamy2 & t.k.chandrashekar3 1research scholar, sri siddhartha institute of technology, tumkur-572105, india 2 department of studies in mechanical engineering, university b.d.t. college of engineering, davangere-577004, india 3 department of mechanical engineering, sri siddhartha institute of technology, tumkur-572105, india *corresponding author: ph. +919739620150, fax: (0816) 2200270, email: srirameshg@gmail.com abstract in recent decades, aluminium alloy based metal matrix composites are gaining important role in several engineering applications. al6061 alloy has been used as matrix material due to its good formability, excellent mechanical properties and etc., wide spectrum of the applications in industrial sectors. inclusion of frit particulates as reinforcement in al6061 alloy material system improves its hardness, tensile strength, wear resistance. in this present investigation al6061-frit particulate composites were produced by ‘vortex’ method with varying percentages of frit particulate from 0 wt% to 10 wt% in steps of 2.the as-cast matrix alloy and its composites have been subjected to solutionizing treatment at a temperature of 5300c for 2 hours followed by quenching in ice. the quenched specimens were subjected to both natural and artificial ageing. microstructure studies were conducted on as cast and composites in order to investigate the distribution of frit particles retained in matrix material system. densities of al6061 matrix alloy and al6061-frit particulate composites were measured. mechanical properties such as hardness and sand abrasive wear test have been conducted on both al6061 alloy matrix and al6061-frit particulate composite before and after treatment. it has been observed that under identical treatment conditions adopted, a al6061frit particulate composites exhibited significant improvement in hardness, wear resistance and reduced density when compared with al6061 matrix alloy. keywords: mmc’s, solutionizing, ageing, hardness, sand abrasive wear, density 38 abrasive wear behavior of al6061frit particulate composites d.ramesh1*, r.p.swamy2 & t.k.chandrashekar3 1research scholar, sri siddhartha institute of technology, tumkur-572105, india 2 department of studies in mechanical engineering, university b.d.t. college of engineering, davangere-577004, india 3 department of mechanical engineering, sri siddhartha institute of technology, tumkur-572105, india * corresponding author: ph. +919739620150, fax: (0816) 2200270, email: srirameshg@gmail.com abstract in recent decades, aluminium alloy based metal matrix composites are gaining important role in several engineering applications. al6061 alloy has been used as matrix material due to its good formability, excellent mechanical properties and etc., wide spectrum of the applications in industrial sectors. inclusion of frit particulates as reinforcement in al6061 alloy material system improves its hardness, tensile strength, wear resistance. in this present investigation al6061-frit particulate composites were produced by „vortex‟ method with varying percentages of frit particulate from 0 wt% to 10 wt% in steps of 2.the as-cast matrix alloy and its composites have been subjected to solutionizing treatment at a temperature of 5300c for 2 hours followed by quenching in ice. the quenched specimens were subjected to both natural and artificial ageing. microstructure studies were conducted on as cast and composites in order to investigate the distribution of frit particles retained in matrix material system. densities of al6061 matrix alloy and al6061-frit particulate composites were measured. mechanical properties such as hardness and sand abrasive wear test have been conducted on both al6061 alloy matrix and al6061-frit particulate composite before and after treatment. it has been observed that under identical treatment conditions adopted, a al6061-frit particulate composites exhibited significant improvement in hardness, wear resistance and reduced density when compared with al6061 matrix alloy. keywords: mmc‟s, solutionizing, ageing, hardness, sand abrasive wear, density 1.0 introduction improvement in the mechanical properties of wear resistance of aluminium matrix composites can be achieved by adopting suitable treatment; (das et al., 2008) have reported the abrasive wear behavior of as-cast and treated sic reinforced al-si composites. they have reported that un-reinforced matrix material suffers from higher wear rates then that of al-si/sic composites in both as-cast and heat treated conditions. further, heat treated alsi/sic composites exhibits better performance under all studied conditions. (modi et al., 2001) have reported the three body abrasive wear behavior of aluminium-zinc/ al2o3 composites exhibited excellent wear resistance under all the test conditions employed. issn: 2180-1053 vol. 3 no. 2 july-december 2011 journal of mechanical engineering and technology 44 39 a comparative study by (s.das et al. 2007) on wear resistance of zircon sand and alumina reinforced amc`s, revealed improved abrasive wear resistance with the decrease in particle size. adhesive wear behavior of cast al6061-tio2 composites studied by (ramesh et al., 2005) reported that, the wear resistance of composites is superior when compared to al6061 matrix alloy. further, it increases with increase in tio2 particle content. s.das (2004) reported the effect of load on abrasive wear rate of lm13-alloy and lm13 – sic composites, results revealed that wear rates increases as the applied load increases for both as-cast alloy and its composites. an extensive review work on the dry sliding wear characteristics of composites based on aluminium alloys have been under taken by (sannio et al., 1995) and abrasive wear behavior by (deuis et al., 1996).in their studies and discussions, the effect of reinforcement volume fraction, reinforcement size, sliding distance, applied load, sliding speed, hardness of the counter face and properties of the reinforcement phase which influence the wear behavior of this group of composites are examined in detail. reinforcement of hard particles in al matrix protects the matrix alloy surface against destructive action of the abrasive during the abrasive wear behavior and rake angle of the abrasive affects the behavior (zumghar k.h., 1979, hutchings.j.m, 1987, kulik.t et al., 1989, jain-main.t 1985, axen n, 1992). (wang et al., 1989) reported that coarse abrasive particles and high volume fraction of reinforcement results in decreased resistance; this is attributed mainly due to fragmentation of reinforcement phase. on the other hand, it was mentioned with decrease in the abrasive particle size. 2.0 experimental 2.1 material selection al6061 matrix alloy was choosen as matrix material owing to its many advantages like excellent casting properties, strength, formability, and heat treatable. table 1 shows the chemical composition of al6061 matrix alloy material used in this present study. frit particles size around 50 µm was used as reinforcement material in al6061 matrix material. table 2 shows the chemical composition of frit particle reinforcement used in present study. 40 table 1. chemical composition of al6061 (wt %). si cu fe mn mg zn pb ti sn al 0.809 0.355 0.155 0.027 0.8 0.008 0.023 0.010 0.010 97.390 table 2. chemical composition of frit (wt %). sio2 al2o3 fe2o3 cao mgo na2o k2o b2o3 68.90 9.41 0.40 15.22 4.30 0.75 0.42 <0.05 2.2 composite production al6061-frit composites were prepared using liquid metallurgy route (vortex).particulate mmc`s are most commonly manufactured either by melt incorporation and casting technique or by powder blending and consolidation (clyne t.w., 2001). amc`s are synthesized via variety of manufacturing routes. these techniques include stir casting (s.skolians. 1996, kang c.g. et al., 1997, xuy et al., 1998), liquid metal infiltration (seo y.h. et al., 1995), squeeze casting (lee j.c., et al.,1998) and spray co-deposition (bar j. et al.,1993). stir casting route is generally practiced commercially (skolianos s., et al., 1993, banerji a. et al., 1982, surappa m.k, et al., 1982) .its advantage lies in its simplicity, flexibility and applicability to large quantity of production. al6061 matrix alloy material was melted using 6 kw electrical resistance furnace. pre heated frit particles were slowly added into the molten matrix alloy material and mixed thoroughly by means of mechanical stirrer. thoroughly mixed composite melt maintained at a temperature of 7100c was poured into the preheated metallic molds. the proportion of frit particles was varied from 2 wt% to 10 wt% in steps of 2 wt%. however al6061 matrix alloy material was also casted for comparison. cast al6061 matrix alloy material and al6061frit particulate composites were machined to test standards. 2.3 heat treatment al6061 matrix alloy and al6061-frit particulate composites were subjected to thermal treatment by solutionizing at a temperature of 5300c followed by ice quenching. both artificial and natural ageing (0 h) were employed on the quenched specimens. artificial ageing was performed at a temperature of 1750c for duration of 2 h to 10 h in steps of 2 h. 2.4 microstructure al6061 matrix alloy and al6061-frit particulate composites were subjected to microstructural studies. the standard metallographic technique was adopted on al6061 matrix alloy and al6061-frit particulate composites for microstructural characterization. the polished specimens were etched with keller’s reagent. 2.5 density test 38 abrasive wear behavior of al6061frit particulate composites d.ramesh1*, r.p.swamy2 & t.k.chandrashekar3 1research scholar, sri siddhartha institute of technology, tumkur-572105, india 2 department of studies in mechanical engineering, university b.d.t. college of engineering, davangere-577004, india 3 department of mechanical engineering, sri siddhartha institute of technology, tumkur-572105, india * corresponding author: ph. +919739620150, fax: (0816) 2200270, email: srirameshg@gmail.com abstract in recent decades, aluminium alloy based metal matrix composites are gaining important role in several engineering applications. al6061 alloy has been used as matrix material due to its good formability, excellent mechanical properties and etc., wide spectrum of the applications in industrial sectors. inclusion of frit particulates as reinforcement in al6061 alloy material system improves its hardness, tensile strength, wear resistance. in this present investigation al6061-frit particulate composites were produced by „vortex‟ method with varying percentages of frit particulate from 0 wt% to 10 wt% in steps of 2.the as-cast matrix alloy and its composites have been subjected to solutionizing treatment at a temperature of 5300c for 2 hours followed by quenching in ice. the quenched specimens were subjected to both natural and artificial ageing. microstructure studies were conducted on as cast and composites in order to investigate the distribution of frit particles retained in matrix material system. densities of al6061 matrix alloy and al6061-frit particulate composites were measured. mechanical properties such as hardness and sand abrasive wear test have been conducted on both al6061 alloy matrix and al6061-frit particulate composite before and after treatment. it has been observed that under identical treatment conditions adopted, a al6061-frit particulate composites exhibited significant improvement in hardness, wear resistance and reduced density when compared with al6061 matrix alloy. keywords: mmc‟s, solutionizing, ageing, hardness, sand abrasive wear, density 1.0 introduction improvement in the mechanical properties of wear resistance of aluminium matrix composites can be achieved by adopting suitable treatment; (das et al., 2008) have reported the abrasive wear behavior of as-cast and treated sic reinforced al-si composites. they have reported that un-reinforced matrix material suffers from higher wear rates then that of al-si/sic composites in both as-cast and heat treated conditions. further, heat treated alsi/sic composites exhibits better performance under all studied conditions. (modi et al., 2001) have reported the three body abrasive wear behavior of aluminium-zinc/ al2o3 composites exhibited excellent wear resistance under all the test conditions employed. table 1 issn: 2180-1053 vol. 3 no. 2 july-december 2011 abrasive wear behavior of al6061frit particulate composites 45 40 table 1. chemical composition of al6061 (wt %). si cu fe mn mg zn pb ti sn al 0.809 0.355 0.155 0.027 0.8 0.008 0.023 0.010 0.010 97.390 table 2. chemical composition of frit (wt %). sio2 al2o3 fe2o3 cao mgo na2o k2o b2o3 68.90 9.41 0.40 15.22 4.30 0.75 0.42 <0.05 2.2 composite production al6061-frit composites were prepared using liquid metallurgy route (vortex).particulate mmc`s are most commonly manufactured either by melt incorporation and casting technique or by powder blending and consolidation (clyne t.w., 2001). amc`s are synthesized via variety of manufacturing routes. these techniques include stir casting (s.skolians. 1996, kang c.g. et al., 1997, xuy et al., 1998), liquid metal infiltration (seo y.h. et al., 1995), squeeze casting (lee j.c., et al.,1998) and spray co-deposition (bar j. et al.,1993). stir casting route is generally practiced commercially (skolianos s., et al., 1993, banerji a. et al., 1982, surappa m.k, et al., 1982) .its advantage lies in its simplicity, flexibility and applicability to large quantity of production. al6061 matrix alloy material was melted using 6 kw electrical resistance furnace. pre heated frit particles were slowly added into the molten matrix alloy material and mixed thoroughly by means of mechanical stirrer. thoroughly mixed composite melt maintained at a temperature of 7100c was poured into the preheated metallic molds. the proportion of frit particles was varied from 2 wt% to 10 wt% in steps of 2 wt%. however al6061 matrix alloy material was also casted for comparison. cast al6061 matrix alloy material and al6061frit particulate composites were machined to test standards. 2.3 heat treatment al6061 matrix alloy and al6061-frit particulate composites were subjected to thermal treatment by solutionizing at a temperature of 5300c followed by ice quenching. both artificial and natural ageing (0 h) were employed on the quenched specimens. artificial ageing was performed at a temperature of 1750c for duration of 2 h to 10 h in steps of 2 h. 2.4 microstructure al6061 matrix alloy and al6061-frit particulate composites were subjected to microstructural studies. the standard metallographic technique was adopted on al6061 matrix alloy and al6061-frit particulate composites for microstructural characterization. the polished specimens were etched with keller’s reagent. 2.5 density test 41 the theoretical density was calculated using rule-of-mixture and experimentally, the density measurements were carried out on the base alloy and reinforced samples using archimedes principle. the buoyant force on submerged object is equal to the weight of the fluid displaced. this principle is useful for determining the volume, by measuring its mass in air and its effective mass when submerged in water (density=1 g/cc).this effective mass under water will be its actual mass minus the mass of the fluid displaced. the difference between the real and effective masses therefore gives the mass of the displaced water and allows the volume of the object to be calculated. mass divided by the volume thus determined gives a measure of the average density of the object (ramachandra m. et al., 2006). the density of material, which is ratio of weight to volume (bermudeza m.d. et al., 2001, ahmad s.n. et al., 2005), theoretical density was derived from halpin-tsai equation. density (theoretical) =ρ (drmmc) = ρm vf(m)+ ρpvf (p) ------------(1) ρ (drmmc)= density of the drmmcs, ρm= density of matrix alloy, ρp= density of reinforcing particle, vf(m)= volume fraction of the matrix, vf(p)= volume fraction of reinforcing particle. the densities of the respective specimen were determined basically by measuring the mass and the volume by using the balance and the measuring cylinder respectively. it is then estimated from the formula given below (aigbodin v.s. et al., 2007, ogucha i. a, 1997). density (gm/cc) =mass (gm)/volume (cc) -------------(2) density of reinforcement material (frit) =2.52 g/cc. density of matrix material (al6061) = 2.70 g/cc. 2.4 hardness test hardness is one important property which effects wear resistance of any metal or alloy, hardness measurements were carried out on al6061 matrix alloy and al6061-frit particulate composite specimens of both as-cast and treated. brinell hardness measurements were carried out in order to investigate the influence of frit particulate on the matrix alloy hardness. the applied load was 500 kgs and an indenter of 10 mm diameter steel ball (hb500). round specimens of 20 mm in diameter were prepared and polished on different grits of emery paper. the polished specimens were tested using brinell hardness tester. the test was carried out at five different locations to controvert the possible effect of indenter resting on the harder particles. hardness was determined by measuring the indentations diameter produced. the average of all the five readings was taken as the hardness of as-cast and composite specimens. figure 1 & 2 shows the hardness test specimen before and after indentation. table 1 issn: 2180-1053 vol. 3 no. 2 july-december 2011 journal of mechanical engineering and technology 46 41 the theoretical density was calculated using rule-of-mixture and experimentally, the density measurements were carried out on the base alloy and reinforced samples using archimedes principle. the buoyant force on submerged object is equal to the weight of the fluid displaced. this principle is useful for determining the volume, by measuring its mass in air and its effective mass when submerged in water (density=1 g/cc).this effective mass under water will be its actual mass minus the mass of the fluid displaced. the difference between the real and effective masses therefore gives the mass of the displaced water and allows the volume of the object to be calculated. mass divided by the volume thus determined gives a measure of the average density of the object (ramachandra m. et al., 2006). the density of material, which is ratio of weight to volume (bermudeza m.d. et al., 2001, ahmad s.n. et al., 2005), theoretical density was derived from halpin-tsai equation. density (theoretical) =ρ (drmmc) = ρm vf(m)+ ρpvf (p) ------------(1) ρ (drmmc)= density of the drmmcs, ρm= density of matrix alloy, ρp= density of reinforcing particle, vf(m)= volume fraction of the matrix, vf(p)= volume fraction of reinforcing particle. the densities of the respective specimen were determined basically by measuring the mass and the volume by using the balance and the measuring cylinder respectively. it is then estimated from the formula given below (aigbodin v.s. et al., 2007, ogucha i. a, 1997). density (gm/cc) =mass (gm)/volume (cc) -------------(2) density of reinforcement material (frit) =2.52 g/cc. density of matrix material (al6061) = 2.70 g/cc. 2.4 hardness test hardness is one important property which effects wear resistance of any metal or alloy, hardness measurements were carried out on al6061 matrix alloy and al6061-frit particulate composite specimens of both as-cast and treated. brinell hardness measurements were carried out in order to investigate the influence of frit particulate on the matrix alloy hardness. the applied load was 500 kgs and an indenter of 10 mm diameter steel ball (hb500). round specimens of 20 mm in diameter were prepared and polished on different grits of emery paper. the polished specimens were tested using brinell hardness tester. the test was carried out at five different locations to controvert the possible effect of indenter resting on the harder particles. hardness was determined by measuring the indentations diameter produced. the average of all the five readings was taken as the hardness of as-cast and composite specimens. figure 1 & 2 shows the hardness test specimen before and after indentation. 42 2.7 sand abrasion test the three body abrasion test was carried out at room temperature on al6061 matrix alloy and al6061-frit particulate composites of as-cast and thermal treated conditions. the tests were performed using standard rubber wheel abrasion testing apparatus as per astm g6581 standards. figure 3 is the photograph of the sand abrasion tester. details of the sand abrasion tester employed in this present study are reported in table 3. the test specimens of size 75 x 25 x 8 mm were metallographically prepared and polished. loads were varied from 2 n to 10 n in steps of 2 n at constant wheel speed of 200 rpm. silica sand of grain size 50 µm was used as the abrasive media. test duration of 30 minutes was adopted for all specimens. using digital weighing balance of accuracy 0.1 mg wear loss was measured. table 3.details of the sand abrasion testing apparatus sl. no. description particulars 1 abrasive material afs 50-70 test sand 2 rubber wheel speed 200 rpm through a helical geared motor of 1.5 kw (3 phase) 3 test load 1 to 45 n 4 rubber wheel diameter 228 mm 5 power 430 v ac (3 phase) 6 specimen dimension 75 x 24 x 8 mm 7 erodent afs3080 figure 1. pictorial view of hardness test specimens figure 2. pictorial view of hardness test specimens after indentation 42 2.7 sand abrasion test the three body abrasion test was carried out at room temperature on al6061 matrix alloy and al6061-frit particulate composites of as-cast and thermal treated conditions. the tests were performed using standard rubber wheel abrasion testing apparatus as per astm g6581 standards. figure 3 is the photograph of the sand abrasion tester. details of the sand abrasion tester employed in this present study are reported in table 3. the test specimens of size 75 x 25 x 8 mm were metallographically prepared and polished. loads were varied from 2 n to 10 n in steps of 2 n at constant wheel speed of 200 rpm. silica sand of grain size 50 µm was used as the abrasive media. test duration of 30 minutes was adopted for all specimens. using digital weighing balance of accuracy 0.1 mg wear loss was measured. table 3.details of the sand abrasion testing apparatus sl. no. description particulars 1 abrasive material afs 50-70 test sand 2 rubber wheel speed 200 rpm through a helical geared motor of 1.5 kw (3 phase) 3 test load 1 to 45 n 4 rubber wheel diameter 228 mm 5 power 430 v ac (3 phase) 6 specimen dimension 75 x 24 x 8 mm 7 erodent afs3080 figure 1. pictorial view of hardness test specimens figure 2. pictorial view of hardness test specimens after indentation figure 1 & 2 shows the hardness test specimen before and after figure 2 issn: 2180-1053 vol. 3 no. 2 july-december 2011 abrasive wear behavior of al6061frit particulate composites 47 42 2.7 sand abrasion test the three body abrasion test was carried out at room temperature on al6061 matrix alloy and al6061-frit particulate composites of as-cast and thermal treated conditions. the tests were performed using standard rubber wheel abrasion testing apparatus as per astm g6581 standards. figure 3 is the photograph of the sand abrasion tester. details of the sand abrasion tester employed in this present study are reported in table 3. the test specimens of size 75 x 25 x 8 mm were metallographically prepared and polished. loads were varied from 2 n to 10 n in steps of 2 n at constant wheel speed of 200 rpm. silica sand of grain size 50 µm was used as the abrasive media. test duration of 30 minutes was adopted for all specimens. using digital weighing balance of accuracy 0.1 mg wear loss was measured. table 3.details of the sand abrasion testing apparatus sl. no. description particulars 1 abrasive material afs 50-70 test sand 2 rubber wheel speed 200 rpm through a helical geared motor of 1.5 kw (3 phase) 3 test load 1 to 45 n 4 rubber wheel diameter 228 mm 5 power 430 v ac (3 phase) 6 specimen dimension 75 x 24 x 8 mm 7 erodent afs3080 figure 1. pictorial view of hardness test specimens figure 2. pictorial view of hardness test specimens after indentation 43 8 sand mass flow rate 0.25 kg/min or 2.45 n/min 9 rubber hardness 60-62 shore a 10 duration 30 min 11 pressure 5.88 n/mm2 12 load 12.75 n figure 3. photograph of sand abrasion testing apparatus 3.0 results and discussion 3.1 microstructure the optical micrograph of the cast al6061 and al6061-frit particulate composites are shown in fig.4. a & b. the micrograph clearly indicates the distribution of frit particles is fairly uniform. further the micrograph reveals an excellent bond between the matrix alloy and the reinforcement as shown in fig.5. table 3 figure 3 issn: 2180-1053 vol. 3 no. 2 july-december 2011 journal of mechanical engineering and technology 48 44 3.2 density the fig. 6. is presented with comparison of theoretical density obtained by rule of mixture and measured density values by experiment for al6061 matrix alloy and al6061frit particulate composites studied. the results reveal that the presence of frit particles has an effect on the density of the mmc`s. from figure it can be concluded that the experimental and theoretical density values are in line with each other and confirms the suitability of the liquid metallurgy route for successful composite fabrication. this result is similar to that of other researchers (aigbodion v.s. et al., 2010). m.ramachandra et.al. (2006) reported the density of the reinforced particulates has effect on the density of the mmc`s and variation of density with reinforcement content is in line with experimental density values of al6061un-heat treated heat treated heat treated un-heat treated figure 4.b. microstructure of 6 wt % of al6061frit particulate composite figure 5. optical micrograph of aluminium /frit composite indicating good bond between the matrix and frit particles. figure. 4.a. microstructure of as cast al6061 matrix alloy 43 8 sand mass flow rate 0.25 kg/min or 2.45 n/min 9 rubber hardness 60-62 shore a 10 duration 30 min 11 pressure 5.88 n/mm2 12 load 12.75 n figure 3. photograph of sand abrasion testing apparatus 3.0 results and discussion 3.1 microstructure the optical micrograph of the cast al6061 and al6061-frit particulate composites are shown in fig.4. a & b. the micrograph clearly indicates the distribution of frit particles is fairly uniform. further the micrograph reveals an excellent bond between the matrix alloy and the reinforcement as shown in fig.5. figure 4.a. figure 4.b. figure 5 issn: 2180-1053 vol. 3 no. 2 july-december 2011 abrasive wear behavior of al6061frit particulate composites 49 44 3.2 density the fig. 6. is presented with comparison of theoretical density obtained by rule of mixture and measured density values by experiment for al6061 matrix alloy and al6061frit particulate composites studied. the results reveal that the presence of frit particles has an effect on the density of the mmc`s. from figure it can be concluded that the experimental and theoretical density values are in line with each other and confirms the suitability of the liquid metallurgy route for successful composite fabrication. this result is similar to that of other researchers (aigbodion v.s. et al., 2010). m.ramachandra et.al. (2006) reported the density of the reinforced particulates has effect on the density of the mmc`s and variation of density with reinforcement content is in line with experimental density values of al6061un-heat treated heat treated heat treated un-heat treated figure 4.b. microstructure of 6 wt % of al6061frit particulate composite figure 5. optical micrograph of aluminium /frit composite indicating good bond between the matrix and frit particles. figure. 4.a. microstructure of as cast al6061 matrix alloy 45 frit composites. it is observed that as the weight % of the reinforcement increases the density of the composites decreases. 3.3 hardness fig. 7 and 8. shows the variation of hardness of al6061 matrix alloy and al6061-frit particulate composites in as-cast and treated conditions. it is observed that hardness increases with increase in weight percentage of frit particles in matrix alloy in both as-cast and treated conditions. a maximum of around 29 % and 44 % improvement is observed in as-cast and treated al60616 wt % frit composite respectively. increased hardness with increased weight percentage of frit particles in the al6061 matrix alloy can be attributed to the following reasons.  higher hardness of frit particles in ductile and soft al6061 matrix alloy enhances the hardness in general (ganesh v.v. et al., 2002).  good interfacial bond between matrix alloy and particle reinforcement as shown in fig. 4.  reinforcement content weight percentage increase in the matrix alloy. density v/s weight % of reinforcement 2.66 2.665 2.67 2.675 2.68 2.685 2.69 2.695 2.7 2.705 0 2 4 6 8 10 w e i gh t % of re i n force me n t d en si ty ( g/ cc ) t heoret ical densit y experiment al densit y figure 6. effect of reinforcement on density of the composites. 0 10 20 30 40 50 60 70 0 2 4 6 8 10 w e i gh t % of re i n force me n t h a rd n es s (b h n ) figure 7. variation of hardness with increase in weight % of reinforcement for as cast al6061 matrix alloy and al 6061-frit particulate composites 45 frit composites. it is observed that as the weight % of the reinforcement increases the density of the composites decreases. 3.3 hardness fig. 7 and 8. shows the variation of hardness of al6061 matrix alloy and al6061-frit particulate composites in as-cast and treated conditions. it is observed that hardness increases with increase in weight percentage of frit particles in matrix alloy in both as-cast and treated conditions. a maximum of around 29 % and 44 % improvement is observed in as-cast and treated al60616 wt % frit composite respectively. increased hardness with increased weight percentage of frit particles in the al6061 matrix alloy can be attributed to the following reasons.  higher hardness of frit particles in ductile and soft al6061 matrix alloy enhances the hardness in general (ganesh v.v. et al., 2002).  good interfacial bond between matrix alloy and particle reinforcement as shown in fig. 4.  reinforcement content weight percentage increase in the matrix alloy. density v/s weight % of reinforcement 2.66 2.665 2.67 2.675 2.68 2.685 2.69 2.695 2.7 2.705 0 2 4 6 8 10 w e i gh t % of re i n force me n t d en si ty ( g/ cc ) t heoret ical densit y experiment al densit y figure 6. effect of reinforcement on density of the composites. 0 10 20 30 40 50 60 70 0 2 4 6 8 10 w e i gh t % of re i n force me n t h a rd n es s (b h n ) figure 7. variation of hardness with increase in weight % of reinforcement for as cast al6061 matrix alloy and al 6061-frit particulate composites figure 6 issn: 2180-1053 vol. 3 no. 2 july-december 2011 journal of mechanical engineering and technology 50 46 3.4 sand abrasion 3.4.1 effect of reinforcement fig.9. shows the variation of abrasive wear loss (weight loss) of as-cast and heat-treated al6061 matrix alloy and al6061frit particulate composites. it is noticed that weight loss decreases with increase in reinforcement content in matrix alloy in both as-cast and thermally treated conditions. in the all the specimens studied, heat-treated specimens exhibited better abrasive wear resistance than as-cast matrix alloy and its composites. decrease in weight loss with in increase in weight percentage of reinforcement indicates higher hardness of composites. the inclusion of hard frit particles in soft ductile matrix alloy protects and reduces the extent of penetration of the abrasive particles on the surface. (mondal d.p. et al., 2006). higher hardness results in better abrasive wear resistance of the materials. the major factor that influences the wear behavior of composites is good interfacial bond between the matrix alloy and the reinforcement. a large wear results in absence of good interfacial bond, under three body abrasion wear conditions (zhang z.f. et al., 1994, ramesh c.s. et al., 2007 ). the three body abrasion wear results of all specimens studied indicated no plucking of frit particles from matrix alloy. this shows the good interfacial bond between frit particles and the matrix alloy. it is reported that the wear behavior of particle reinforced composites mainly depend on the type of interfacial bond between the al matrix alloy and the reinforcement. (kok m., 2006). load 4 n 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 0% 2% 4% 6% 8% 10% wt % of reinforcement w t l os s in m g un-heat treated heat treated figure 9. variation of weight loss of al6061 matrix alloy and l6061frit particulate composites. 45 frit composites. it is observed that as the weight % of the reinforcement increases the density of the composites decreases. 3.3 hardness fig. 7 and 8. shows the variation of hardness of al6061 matrix alloy and al6061-frit particulate composites in as-cast and treated conditions. it is observed that hardness increases with increase in weight percentage of frit particles in matrix alloy in both as-cast and treated conditions. a maximum of around 29 % and 44 % improvement is observed in as-cast and treated al60616 wt % frit composite respectively. increased hardness with increased weight percentage of frit particles in the al6061 matrix alloy can be attributed to the following reasons.  higher hardness of frit particles in ductile and soft al6061 matrix alloy enhances the hardness in general (ganesh v.v. et al., 2002).  good interfacial bond between matrix alloy and particle reinforcement as shown in fig. 4.  reinforcement content weight percentage increase in the matrix alloy. density v/s weight % of reinforcement 2.66 2.665 2.67 2.675 2.68 2.685 2.69 2.695 2.7 2.705 0 2 4 6 8 10 w e i gh t % of re i n force me n t d en si ty ( g/ cc ) t heoret ical densit y experiment al densit y figure 6. effect of reinforcement on density of the composites. 0 10 20 30 40 50 60 70 0 2 4 6 8 10 w e i gh t % of re i n force me n t h a rd n es s (b h n ) figure 7. variation of hardness with increase in weight % of reinforcement for as cast al6061 matrix alloy and al 6061-frit particulate composites figure 7 issn: 2180-1053 vol. 3 no. 2 july-december 2011 abrasive wear behavior of al6061frit particulate composites 51 46 3.4 sand abrasion 3.4.1 effect of reinforcement fig.9. shows the variation of abrasive wear loss (weight loss) of as-cast and heat-treated al6061 matrix alloy and al6061frit particulate composites. it is noticed that weight loss decreases with increase in reinforcement content in matrix alloy in both as-cast and thermally treated conditions. in the all the specimens studied, heat-treated specimens exhibited better abrasive wear resistance than as-cast matrix alloy and its composites. decrease in weight loss with in increase in weight percentage of reinforcement indicates higher hardness of composites. the inclusion of hard frit particles in soft ductile matrix alloy protects and reduces the extent of penetration of the abrasive particles on the surface. (mondal d.p. et al., 2006). higher hardness results in better abrasive wear resistance of the materials. the major factor that influences the wear behavior of composites is good interfacial bond between the matrix alloy and the reinforcement. a large wear results in absence of good interfacial bond, under three body abrasion wear conditions (zhang z.f. et al., 1994, ramesh c.s. et al., 2007 ). the three body abrasion wear results of all specimens studied indicated no plucking of frit particles from matrix alloy. this shows the good interfacial bond between frit particles and the matrix alloy. it is reported that the wear behavior of particle reinforced composites mainly depend on the type of interfacial bond between the al matrix alloy and the reinforcement. (kok m., 2006). load 4 n 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 0% 2% 4% 6% 8% 10% wt % of reinforcement w t l os s in m g un-heat treated heat treated figure 9. variation of weight loss of al6061 matrix alloy and l6061frit particulate composites. 47 3.4.2 effect of load the variation of weight loss of al6061 matrix alloy and al6061-frit composites with load in as-cast and heat-treated condition is shown in fig. 10 and 11. there is a steady increase in wear up to a load of 8 n and a steep increase is noticed at 8 n for all the specimens studied. tim e 30 m inutes 0 0.5 1 1.5 2 2.5 2n 4n 6n 8n 10n load in n w t lo s s i n m g 0 wt % un-heat treated 2 wt % un-heat treated 4 wt % un-heat treated 6 wt % un-heat treated 8 wt % un-heat treated 10 wt % un-heat treated figure 10. variation of weight loss with increase in load for al6061 matrix alloy and al 6061-frit particulate composites under un-heat treated conditions time 30 minutes 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 2n 4n 6n 8n 10n load in n w t l os s in m g 0 wt % heat treated 2 wt % heat treated 4 wt % heat treated 6 wt % heat treated 8 wt % heat treated 10 wt % heat treated figure 11.variation of weight loss with increase in load for al6061 matrix alloy and al 6061-frit particulate composites under heat treatment conditions figure 8 figure 9 issn: 2180-1053 vol. 3 no. 2 july-december 2011 journal of mechanical engineering and technology 52 47 3.4.2 effect of load the variation of weight loss of al6061 matrix alloy and al6061-frit composites with load in as-cast and heat-treated condition is shown in fig. 10 and 11. there is a steady increase in wear up to a load of 8 n and a steep increase is noticed at 8 n for all the specimens studied. tim e 30 m inutes 0 0.5 1 1.5 2 2.5 2n 4n 6n 8n 10n load in n w t lo s s i n m g 0 wt % un-heat treated 2 wt % un-heat treated 4 wt % un-heat treated 6 wt % un-heat treated 8 wt % un-heat treated 10 wt % un-heat treated figure 10. variation of weight loss with increase in load for al6061 matrix alloy and al 6061-frit particulate composites under un-heat treated conditions time 30 minutes 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 2n 4n 6n 8n 10n load in n w t l os s in m g 0 wt % heat treated 2 wt % heat treated 4 wt % heat treated 6 wt % heat treated 8 wt % heat treated 10 wt % heat treated figure 11.variation of weight loss with increase in load for al6061 matrix alloy and al 6061-frit particulate composites under heat treatment conditions 48 4.0 conclusion microstructural studies clearly indicate a fairly uniform distribution of frit particles in the al6061 matrix alloy with a good interfacial bond between the reinforcement and al6061 matrix alloy. density measurement studies reveals as the weight percentage of reinforcement increases the density of the composites decreases. hardness increases with ageing duration reaches a peak value at 6 h, and with further increase in ageing duration, there is a decrease in hardness. al6061frit particulate composites exhibit superior abrasive wear resistance than al6061 matrix alloy in as-cast and heat-treated conditions. 5.0 references ahmad s.n., hashim j. and ghazali m.t., (2005), “the effect of porosity on mechanical properties of cast discontinuous reinforced metal-matrix composites”, journal of composite materials, vol.39, no 5, (pp 451-466). aigbodin v.s. and hassan s.b., (2007) “effects of silicon carbide reinforcement on microstructure and properties of cast al – si – fe / sic particulate composites”, journal of material sciences and engineering a, 447, (pp 355-360). aigbodion v.s., agunsoye j.o., kalu v., asuke f., ola s., (2010) “microstructure and mechanical properties of ceramic composites”, jmmce, vol.9, no.6, (pp527-538). axen n, zumghar k.h. , wear, 1992, 157,189. banerji a., prasad s.v., surappa m.k. and rohatgi p.k., (1982) “abrasive wear of cast aluminium alloy-zircon particle composites”, wear, 82, (pp 141-151). bar j., klubman h.k., and gudladt h.a., (1993), “influence of fiber reinforcement on microstructure of an al-si based alloy”, scripta materialia 23, (pp 787-792). bermudeza m.d., martinez-niccolas g., carrion f.j, martin-mateo i., rodriguez j.a., herrera e.j., (2001) “dry and lubricated wear resistance of mechanically-alloyed aluminium-base sintered composites”, wear, 248, (pp 178-186). clyne t.w., (2001) metal matrix composites; matrices and processing, “composites: mmc, cmc, pmc”, a mortensen (ed), elsevier, (pp 7-14). das s. and das k., (2007)“abrasive wear of zircon sand and alumina reinforced al-4.5 wt% cu alloy matrix compositesa comparative study”, compos. sci. technol., 67, (pp 746-751). das s., (2004) “development of aluminium alloy composites for engineering applications,trans. indian inst. met, vol 57, no 4, (pp 325-334). das s., mondal d.p., sawla s., and ramakrishnan n., (2008) “synergic effect of reinforcement and heat treatment on the two body abrasive wear of an al-si alloy under varying loads and abrasive size”, wear, 264, (pp 47-59). deuis r.l., subramanian c. and yellup j.m., (1996) “abrasive wear of aluminium compositesa review”, wear , 201, 1-2, p132-144. ganesh v.v., lee c.k. and gupta m., (2002) “enhancing the tensile modulus and strength of an aluminum”, mater. sci. eng. a, 333, (1-2), (pp 103-198). hutchings.j.m, (1987) chem. engg. sci. 42. jain-main.t, ye-yang s,hua-ji z,chingan z, zianwu x., (1985) tribo int., 18, 101. kang c.g., and yoon j.h., (1997) “the upsetting behavior of semi-solid aluminium material fabricated by a mechanical stirring process”, journal of material production technology, 66, (pp 30-38). 48 4.0 conclusion microstructural studies clearly indicate a fairly uniform distribution of frit particles in the al6061 matrix alloy with a good interfacial bond between the reinforcement and al6061 matrix alloy. density measurement studies reveals as the weight percentage of reinforcement increases the density of the composites decreases. hardness increases with ageing duration reaches a peak value at 6 h, and with further increase in ageing duration, there is a decrease in hardness. al6061frit particulate composites exhibit superior abrasive wear resistance than al6061 matrix alloy in as-cast and heat-treated conditions. 5.0 references ahmad s.n., hashim j. and ghazali m.t., (2005), “the effect of porosity on mechanical properties of cast discontinuous reinforced metal-matrix composites”, journal of composite materials, vol.39, no 5, (pp 451-466). aigbodin v.s. and hassan s.b., (2007) “effects of silicon carbide reinforcement on microstructure and properties of cast al – si – fe / sic particulate composites”, journal of material sciences and engineering a, 447, (pp 355-360). aigbodion v.s., agunsoye j.o., kalu v., asuke f., ola s., (2010) “microstructure and mechanical properties of ceramic composites”, jmmce, vol.9, no.6, (pp527-538). axen n, zumghar k.h. , wear, 1992, 157,189. banerji a., prasad s.v., surappa m.k. and rohatgi p.k., (1982) “abrasive wear of cast aluminium alloy-zircon particle composites”, wear, 82, (pp 141-151). bar j., klubman h.k., and gudladt h.a., (1993), “influence of fiber reinforcement on microstructure of an al-si based alloy”, scripta materialia 23, (pp 787-792). bermudeza m.d., martinez-niccolas g., carrion f.j, martin-mateo i., rodriguez j.a., herrera e.j., (2001) “dry and lubricated wear resistance of mechanically-alloyed aluminium-base sintered composites”, wear, 248, (pp 178-186). clyne t.w., (2001) metal matrix composites; matrices and processing, “composites: mmc, cmc, pmc”, a mortensen (ed), elsevier, (pp 7-14). das s. and das k., (2007)“abrasive wear of zircon sand and alumina reinforced al-4.5 wt% cu alloy matrix compositesa comparative study”, compos. sci. technol., 67, (pp 746-751). das s., (2004) “development of aluminium alloy composites for engineering applications,trans. indian inst. met, vol 57, no 4, (pp 325-334). das s., mondal d.p., sawla s., and ramakrishnan n., (2008) “synergic effect of reinforcement and heat treatment on the two body abrasive wear of an al-si alloy under varying loads and abrasive size”, wear, 264, (pp 47-59). deuis r.l., subramanian c. and yellup j.m., (1996) “abrasive wear of aluminium compositesa review”, wear , 201, 1-2, p132-144. ganesh v.v., lee c.k. and gupta m., (2002) “enhancing the tensile modulus and strength of an aluminum”, mater. sci. eng. a, 333, (1-2), (pp 103-198). hutchings.j.m, (1987) chem. engg. sci. 42. jain-main.t, ye-yang s,hua-ji z,chingan z, zianwu x., (1985) tribo int., 18, 101. kang c.g., and yoon j.h., (1997) “the upsetting behavior of semi-solid aluminium material fabricated by a mechanical stirring process”, journal of material production technology, 66, (pp 30-38). figure 10 issn: 2180-1053 vol. 3 no. 2 july-december 2011 abrasive wear behavior of al6061frit particulate composites 53 48 4.0 conclusion microstructural studies clearly indicate a fairly uniform distribution of frit particles in the al6061 matrix alloy with a good interfacial bond between the reinforcement and al6061 matrix alloy. density measurement studies reveals as the weight percentage of reinforcement increases the density of the composites decreases. hardness increases with ageing duration reaches a peak value at 6 h, and with further increase in ageing duration, there is a decrease in hardness. al6061frit particulate composites exhibit superior abrasive wear resistance than al6061 matrix alloy in as-cast and heat-treated conditions. 5.0 references ahmad s.n., hashim j. and ghazali m.t., (2005), “the effect of porosity on mechanical properties of cast discontinuous reinforced metal-matrix composites”, journal of composite materials, vol.39, no 5, (pp 451-466). aigbodin v.s. and hassan s.b., (2007) “effects of silicon carbide reinforcement on microstructure and properties of cast al – si – fe / sic particulate composites”, journal of material sciences and engineering a, 447, (pp 355-360). aigbodion v.s., agunsoye j.o., kalu v., asuke f., ola s., (2010) “microstructure and mechanical properties of ceramic composites”, jmmce, vol.9, no.6, (pp527-538). axen n, zumghar k.h. , wear, 1992, 157,189. banerji a., prasad s.v., surappa m.k. and rohatgi p.k., (1982) “abrasive wear of cast aluminium alloy-zircon particle composites”, wear, 82, (pp 141-151). bar j., klubman h.k., and gudladt h.a., (1993), “influence of fiber reinforcement on microstructure of an al-si based alloy”, scripta materialia 23, (pp 787-792). bermudeza m.d., martinez-niccolas g., carrion f.j, martin-mateo i., rodriguez j.a., herrera e.j., (2001) “dry and lubricated wear resistance of mechanically-alloyed aluminium-base sintered composites”, wear, 248, (pp 178-186). clyne t.w., (2001) metal matrix composites; matrices and processing, “composites: mmc, cmc, pmc”, a mortensen (ed), elsevier, (pp 7-14). das s. and das k., (2007)“abrasive wear of zircon sand and alumina reinforced al-4.5 wt% cu alloy matrix compositesa comparative study”, compos. sci. technol., 67, (pp 746-751). das s., (2004) “development of aluminium alloy composites for engineering applications,trans. indian inst. met, vol 57, no 4, (pp 325-334). das s., mondal d.p., sawla s., and ramakrishnan n., (2008) “synergic effect of reinforcement and heat treatment on the two body abrasive wear of an al-si alloy under varying loads and abrasive size”, wear, 264, (pp 47-59). deuis r.l., subramanian c. and yellup j.m., (1996) “abrasive wear of aluminium compositesa review”, wear , 201, 1-2, p132-144. ganesh v.v., lee c.k. and gupta m., (2002) “enhancing the tensile modulus and strength of an aluminum”, mater. sci. eng. a, 333, (1-2), (pp 103-198). hutchings.j.m, (1987) chem. engg. sci. 42. jain-main.t, ye-yang s,hua-ji z,chingan z, zianwu x., (1985) tribo int., 18, 101. kang c.g., and yoon j.h., (1997) “the upsetting behavior of semi-solid aluminium material fabricated by a mechanical stirring process”, journal of material production technology, 66, (pp 30-38). 49 kok m., (2006),“abrasive wear behavior of al2o3 particle reinforced 2024 aluminum alloy composites fabricated by vortex method”, composites a, 37, (pp 457-464). kulik.t, kosel.t.h, yu.k, (1989) proc. int conf, wear mater.ed. k.c.ludema usa, denver,asme, 23. lee j.c., byun j.y., oh c.s., seok h.k., and lee h.i, (1998), “effect of various processing methods on the interfacial reaction in sicp/2024 al composites”, acta materials, 45, (pp 5303-5315). modi o.p., yadav r.p., prasad b.k., jha a.k., das s., and yagneswaran a.h., (2001) “three body abrasion of cast zinc aluminium alloy: influence of al2o3 dispersoid and abrasive medium”, wear, 249, (pp 792-799). mondal d.p. and das s., (2006) “high stress abrasive wear behavior of aluminum hard particle composites; effect of experimental parameters, particle size, and volume fraction”, tribol. int., 39, (pp 470-478). ogucha i. a., (1997) “characterization of aluminium alloy 2618 and its composites containing alumina particles”, ph.d. dissertation, department of mechanical engineering, university of saskatchewan, saskanoon, (pp 1-200). ramachandra m.and radhakrishna k., (2006) “sliding wear, slurry erosive wear, and corrosive wear of aluminium/sic composite, materials science-poland, vol.24, no 2/1. ramesh c.s. and mir safiulla, (2007) “wear behavior of hot extruded al6061 based composites”, wear, 263, (pp 629-635). ramesh c.s., answar khan a.r., ravikumar n., and savaprabhu p.,(2005) “prediction of wear co-efficient of al6061-tio2 composites”, wear, 259, (pp 602-608). s.skolians., (1996), “mechanical behavior of cast sicp-reinforced al-4.5%cu-1.5%mg alloy”, material science engineering, a 210, (pp76-82). sannio a.p. and rack h.j., (1995) “dry sliding wear of discontinuously reinforced aluminium composites: review and discussion”, wear, 189, (pp 1-19). seo y.h., and kang c.g. (1995), “the effect of applied pressure on particle dispersion characteristics and mechanical properties in melt-stirring squeezecast sicp/ al composites”, journal of materials production technology, 55, (pp 370-379). skolianos s., and kattamis t.z., (1993), “tribological properties of sicp-reinforced al4.5%cu1.5% mg alloy composites”, material science engineering, a 163, (pp107-113). surappa m.k., prasad s.v., and rohatgi p.k., (1982) “wear and abrasion of cast al-alumina particle composites”, wear, 77, (pp 295-302). wang a.g, and hutchings.j.m, (1989) mater. sci. technol, 5, 71. xuy and chung d.d.l, (1998) “low volume fraction particulate performs for making metalmatrix composites by liquid metal infiltration”, journal of material science, 33, (pp 47074709). zhang z.f., zhang l.c. and mai y.w., “wear of ceramic particle-reinforced metal matrix composites”, part i, wear mechanisms, j. mater. sci., 1994, 30, p 1961-1966. zumghar k.h., (1979), met prog. 116, 4. issn: 2180-1053 vol. 3 no. 2 july-december 2011 journal of mechanical engineering and technology 54 49 kok m., (2006),“abrasive wear behavior of al2o3 particle reinforced 2024 aluminum alloy composites fabricated by vortex method”, composites a, 37, (pp 457-464). kulik.t, kosel.t.h, yu.k, (1989) proc. int conf, wear mater.ed. k.c.ludema usa, denver,asme, 23. lee j.c., byun j.y., oh c.s., seok h.k., and lee h.i, (1998), “effect of various processing methods on the interfacial reaction in sicp/2024 al composites”, acta materials, 45, (pp 5303-5315). modi o.p., yadav r.p., prasad b.k., jha a.k., das s., and yagneswaran a.h., (2001) “three body abrasion of cast zinc aluminium alloy: influence of al2o3 dispersoid and abrasive medium”, wear, 249, (pp 792-799). mondal d.p. and das s., (2006) “high stress abrasive wear behavior of aluminum hard particle composites; effect of experimental parameters, particle size, and volume fraction”, tribol. int., 39, (pp 470-478). ogucha i. a., (1997) “characterization of aluminium alloy 2618 and its composites containing alumina particles”, ph.d. dissertation, department of mechanical engineering, university of saskatchewan, saskanoon, (pp 1-200). ramachandra m.and radhakrishna k., (2006) “sliding wear, slurry erosive wear, and corrosive wear of aluminium/sic composite, materials science-poland, vol.24, no 2/1. ramesh c.s. and mir safiulla, (2007) “wear behavior of hot extruded al6061 based composites”, wear, 263, (pp 629-635). ramesh c.s., answar khan a.r., ravikumar n., and savaprabhu p.,(2005) “prediction of wear co-efficient of al6061-tio2 composites”, wear, 259, (pp 602-608). s.skolians., (1996), “mechanical behavior of cast sicp-reinforced al-4.5%cu-1.5%mg alloy”, material science engineering, a 210, (pp76-82). sannio a.p. and rack h.j., (1995) “dry sliding wear of discontinuously reinforced aluminium composites: review and discussion”, wear, 189, (pp 1-19). seo y.h., and kang c.g. (1995), “the effect of applied pressure on particle dispersion characteristics and mechanical properties in melt-stirring squeezecast sicp/ al composites”, journal of materials production technology, 55, (pp 370-379). skolianos s., and kattamis t.z., (1993), “tribological properties of sicp-reinforced al4.5%cu1.5% mg alloy composites”, material science engineering, a 163, (pp107-113). surappa m.k., prasad s.v., and rohatgi p.k., (1982) “wear and abrasion of cast al-alumina particle composites”, wear, 77, (pp 295-302). wang a.g, and hutchings.j.m, (1989) mater. sci. technol, 5, 71. xuy and chung d.d.l, (1998) “low volume fraction particulate performs for making metalmatrix composites by liquid metal infiltration”, journal of material science, 33, (pp 47074709). zhang z.f., zhang l.c. and mai y.w., “wear of ceramic particle-reinforced metal matrix composites”, part i, wear mechanisms, j. mater. sci., 1994, 30, p 1961-1966. zumghar k.h., (1979), met prog. 116, 4. 04(43-54).pdf issn: 2180-1053 vol. 4 no. 1 january-june 2012 a theoretical model of pitting corrosion using a general purpose finite element package 1 a theoretical model of pitting corrosion using a general purpose finite element package suhaila salleh1, nicholas p.c. stevens2 1faculty of mechanical engineering, universiti teknikal malaysia melaka, hang tuah jaya, 76100 durian tunggal, melaka, malaysia. 2materials performance centre, university of manchester, manchester m13 9pl, united kingdom email: 1suhaila@utem.edu.my, 2nicholas.stevens@manchester.ac.uk abstract pitting corrosion is one of the most destructive types of metal loss. this paper presents the mathematical model of the propagation of pitting corrosion using a commercial finite element program. in view of the chemical and electrochemical reactions inside a single pit in steel, a two dimensional model that allows the prediction of pit evolution is developed. the results are discussed in comparison to pourbaix diagram of iron and also discussed in the light of results obtained from published work reported in literature. keywords: pitting corrosion, model, steel, simulation. 1.0 introduction pitting is one of many types of highly localized corrosion where small holes or cavities are formed on a metal surface but the bulk of the surface remains unattacked. a pit can be difficult to detect given that it has a tendency to undercut the metal surface and is usually covered by corrosion product. pitting can be very destructive due to the fact that pits can cause perforation to the metal. pits can occur in isolated locations or be so concentrated that it looks like uniform attack (fontana 1987; shreir 2000; shreir 2000; perez 2004; ahmad 2006). besides aqueous solution, studies have found that pitting is induced by the presence of halides, such as chlorides or bromides (cui 2001). it is also found that pitting often occurs in situations where general corrosion is prevented by passive oxide film formed on the surface of metal (cheng 2000). the oxide film can break down and is able to self-repair. however, sites that are not able to repair itself are liable for localized corrosion to issn: 2180-1053 vol. 4 no. 1 january-june 2012 journal of mechanical engineering and technology 2 happen. with the presence of halides, rigorous corrosion occurs and it is established that the ph inside a pit drops to a value way below that of the bulk solution. it is also established that there are three stages of pitting (burstein 1993; laycock 2001; perez 2004; ahmad 2006): (i.) initiation of pits, (ii) formation of metastable pits and (iii) stable pitting. even though there is a great body of literature on pitting corrosion of iron, the propagation of pitting is not fully understood. the present paper describes a modelling approach to illustrate stage (iii), which is the evolution of a single, stable pit in steel after the initiation stage. the initial objective of this study is to look at the corrosion activities inside a pit at different potentials and ph values. to do this, a commercial program, comsol multiphysics, is used as a tool and the results are discussed in relation to pourbaix diagram of iron. 2.0 methodology in particular, this model is developed for the evolution of single corrosion pit in carbon steel material in aqueous sodium chloride solution. according to sharland (sharland 1989), six aqueous chemical species are necessary to construct a basic corrosion model of iron : fe2+, h+, oh-, cl-, na+ and feoh+. pickering (pickering 1984) and al-khamis and pickering (al-khamis 2001) have reported experimental evidence of hydrogen gas, h2, formation inside propagating pits. this is also included in the model of crevice corrosion done by vuillemin et.al. (vuillemin 2007). due to the fact that the model is to be discussed in comparison to pourbaix diagram of iron (as shown in figure 1), more ionic species are included, namely fe3+, feo2and hfeo2-. the model applies the chemical reactions as stated below (turnbull 1982; sharland 1988; sharland 1988): 2 i m c a r i c i t o a i e 2.0 met in particular material in a chemical spe and feoh+. reported exp included in fact that the 1), more io chemical rea + + ohfe 2 2 +→ hoh2 in the case o + +→ fefe 2 this model occurrence o h →+ −+ e as mentione ions, na+, electroneutr →nacl (aq) thodolo r, this mode aqueous sod ecies are nec . pickering perimental ev the model o model is to onic species actions as sta + +→ feoho −+ oh of active wal −+ e2 l allows the of proton red 2h2 1 ed earlier, th are not inv ality across t −+ + clna ogy el is develo dium chlorid cessary to co (pickering vidence of h of crevice co be discussed are includ ated below ( ++ h lls, the metal e occurrenc duction (cath he model st volved in t the mouth o fig oped for the e solution. a onstruct a ba 1984) and hydrogen ga orrosion don d in compari ed, namely (turnbull 19 l dissolution e of gaseou hodic) proce tudies a syst the corrosio f pit. gure 1 pourb e evolution according to asic corrosio al-khamis s, h2, forma ne by vuille ison to pourb fe3+, feo2 982; sharland n (anodic) pr us hydrogen ess, which is tem in sodiu on reaction, baix diagram of single co o sharland ( on model of i and pickeri ation inside p emin et al. (v baix diagram 2 and hfeo d 1988; shar ocess is give n formation given by: um chloride but are in m of iron orrosion pit (sharland 19 iron : fe2+, h ing (al-kha propagating vuillemin 2 m of iron (as o2-. the mo rland 1988): en by (sharla n through th e solution. h nvolved in t in carbon 989), six aqu h+, oh-, clamis 2001) pits. this is 007). due to shown in fi odel applies and 1988): he simultan however, sod maintaining steel ueous , na+ have s also o the igure s the (1) (2) (3) neous (4) dium g the (5) (1) 2 i m c a r i c i t o a i e 2.0 met in particular material in a chemical spe and feoh+. reported exp included in fact that the 1), more io chemical rea + + ohfe 2 2 +→ hoh2 in the case o + +→ fefe 2 this model occurrence o h →+ −+ e as mentione ions, na+, electroneutr →nacl (aq) thodolo r, this mode aqueous sod ecies are nec . pickering perimental ev the model o model is to onic species actions as sta + +→ feoho −+ oh of active wal −+ e2 l allows the of proton red 2h2 1 ed earlier, th are not inv ality across t −+ + clna ogy el is develo dium chlorid cessary to co (pickering vidence of h of crevice co be discussed are includ ated below ( ++ h lls, the metal e occurrenc duction (cath he model st volved in t the mouth o fig oped for the e solution. a onstruct a ba 1984) and hydrogen ga orrosion don d in compari ed, namely (turnbull 19 l dissolution e of gaseou hodic) proce tudies a syst the corrosio f pit. gure 1 pourb e evolution according to asic corrosio al-khamis s, h2, forma ne by vuille ison to pourb fe3+, feo2 982; sharland n (anodic) pr us hydrogen ess, which is tem in sodiu on reaction, baix diagram of single co o sharland ( on model of i and pickeri ation inside p emin et al. (v baix diagram 2 and hfeo d 1988; shar ocess is give n formation given by: um chloride but are in m of iron orrosion pit (sharland 19 iron : fe2+, h ing (al-kha propagating vuillemin 2 m of iron (as o2-. the mo rland 1988): en by (sharla n through th e solution. h nvolved in t in carbon 989), six aqu h+, oh-, clamis 2001) pits. this is 007). due to shown in fi odel applies and 1988): he simultan however, sod maintaining steel ueous , na+ have s also o the igure s the (1) (2) (3) neous (4) dium g the (5) (2) in the case of active walls, the metal dissolution (anodic) process is given by (sharland 1988): 2 i m c a r i c i t o a i e 2.0 met in particular material in a chemical spe and feoh+. reported exp included in fact that the 1), more io chemical rea + + ohfe 2 2 +→ hoh2 in the case o + +→ fefe 2 this model occurrence o h →+ −+ e as mentione ions, na+, electroneutr →nacl (aq) thodolo r, this mode aqueous sod ecies are nec . pickering perimental ev the model o model is to onic species actions as sta + +→ feoho −+ oh of active wal −+ e2 l allows the of proton red 2h2 1 ed earlier, th are not inv ality across t −+ + clna ogy el is develo dium chlorid cessary to co (pickering vidence of h of crevice co be discussed are includ ated below ( ++ h lls, the metal e occurrenc duction (cath he model st volved in t the mouth o fig oped for the e solution. a onstruct a ba 1984) and hydrogen ga orrosion don d in compari ed, namely (turnbull 19 l dissolution e of gaseou hodic) proce tudies a syst the corrosio f pit. gure 1 pourb e evolution according to asic corrosio al-khamis s, h2, forma ne by vuille ison to pourb fe3+, feo2 982; sharland n (anodic) pr us hydrogen ess, which is tem in sodiu on reaction, baix diagram of single co o sharland ( on model of i and pickeri ation inside p emin et al. (v baix diagram 2 and hfeo d 1988; shar ocess is give n formation given by: um chloride but are in m of iron orrosion pit (sharland 19 iron : fe2+, h ing (al-kha propagating vuillemin 2 m of iron (as o2-. the mo rland 1988): en by (sharla n through th e solution. h nvolved in t in carbon 989), six aqu h+, oh-, clamis 2001) pits. this is 007). due to shown in fi odel applies and 1988): he simultan however, sod maintaining steel ueous , na+ have s also o the igure s the (1) (2) (3) neous (4) dium g the (5) (3) issn: 2180-1053 vol. 4 no. 1 january-june 2012 a theoretical model of pitting corrosion using a general purpose finite element package 3 this model allows the occurrence of gaseous hydrogen formation through the simultaneous occurrence of proton reduction (cathodic) process, which is given by: 2 i m c a r i c i t o a i e 2.0 met in particular material in a chemical spe and feoh+. reported exp included in fact that the 1), more io chemical rea + + ohfe 2 2 +→ hoh2 in the case o + +→ fefe 2 this model occurrence o h →+ −+ e as mentione ions, na+, electroneutr →nacl (aq) thodolo r, this mode aqueous sod ecies are nec . pickering perimental ev the model o model is to onic species actions as sta + +→ feoho −+ oh of active wal −+ e2 l allows the of proton red 2h2 1 ed earlier, th are not inv ality across t −+ + clna ogy el is develo dium chlorid cessary to co (pickering vidence of h of crevice co be discussed are includ ated below ( ++ h lls, the metal e occurrenc duction (cath he model st volved in t the mouth o fig oped for the e solution. a onstruct a ba 1984) and hydrogen ga orrosion don d in compari ed, namely (turnbull 19 l dissolution e of gaseou hodic) proce tudies a syst the corrosio f pit. gure 1 pourb e evolution according to asic corrosio al-khamis s, h2, forma ne by vuille ison to pourb fe3+, feo2 982; sharland n (anodic) pr us hydrogen ess, which is tem in sodiu on reaction, baix diagram of single co o sharland ( on model of i and pickeri ation inside p emin et al. (v baix diagram 2 and hfeo d 1988; shar ocess is give n formation given by: um chloride but are in m of iron orrosion pit (sharland 19 iron : fe2+, h ing (al-kha propagating vuillemin 2 m of iron (as o2-. the mo rland 1988): en by (sharla n through th e solution. h nvolved in t in carbon 989), six aqu h+, oh-, clamis 2001) pits. this is 007). due to shown in fi odel applies and 1988): he simultan however, sod maintaining steel ueous , na+ have s also o the igure s the (1) (2) (3) neous (4) dium g the (5) (4) as mentioned earlier, the model studies a system in sodium chloride solution. however, sodium ions, na+, are not involved in the corrosion reaction, but are involved in maintaining the electroneutrality across the mouth of pit. 2 i m c a r i c i t o a i e 2.0 met in particular material in a chemical spe and feoh+. reported exp included in fact that the 1), more io chemical rea + + ohfe 2 2 +→ hoh2 in the case o + +→ fefe 2 this model occurrence o h →+ −+ e as mentione ions, na+, electroneutr →nacl (aq) thodolo r, this mode aqueous sod ecies are nec . pickering perimental ev the model o model is to onic species actions as sta + +→ feoho −+ oh of active wal −+ e2 l allows the of proton red 2h2 1 ed earlier, th are not inv ality across t −+ + clna ogy el is develo dium chlorid cessary to co (pickering vidence of h of crevice co be discussed are includ ated below ( ++ h lls, the metal e occurrenc duction (cath he model st volved in t the mouth o fig oped for the e solution. a onstruct a ba 1984) and hydrogen ga orrosion don d in compari ed, namely (turnbull 19 l dissolution e of gaseou hodic) proce tudies a syst the corrosio f pit. gure 1 pourb e evolution according to asic corrosio al-khamis s, h2, forma ne by vuille ison to pourb fe3+, feo2 982; sharland n (anodic) pr us hydrogen ess, which is tem in sodiu on reaction, baix diagram of single co o sharland ( on model of i and pickeri ation inside p emin et al. (v baix diagram 2 and hfeo d 1988; shar ocess is give n formation given by: um chloride but are in m of iron orrosion pit (sharland 19 iron : fe2+, h ing (al-kha propagating vuillemin 2 m of iron (as o2-. the mo rland 1988): en by (sharla n through th e solution. h nvolved in t in carbon 989), six aqu h+, oh-, clamis 2001) pits. this is 007). due to shown in fi odel applies and 1988): he simultan however, sod maintaining steel ueous , na+ have s also o the igure s the (1) (2) (3) neous (4) dium g the (5) (5) 2 i m c a r i c i t o a i e 2.0 met in particular material in a chemical spe and feoh+. reported exp included in fact that the 1), more io chemical rea + + ohfe 2 2 +→ hoh2 in the case o + +→ fefe 2 this model occurrence o h →+ −+ e as mentione ions, na+, electroneutr →nacl (aq) thodolo r, this mode aqueous sod ecies are nec . pickering perimental ev the model o model is to onic species actions as sta + +→ feoho −+ oh of active wal −+ e2 l allows the of proton red 2h2 1 ed earlier, th are not inv ality across t −+ + clna ogy el is develo dium chlorid cessary to co (pickering vidence of h of crevice co be discussed are includ ated below ( ++ h lls, the metal e occurrenc duction (cath he model st volved in t the mouth o fig oped for the e solution. a onstruct a ba 1984) and hydrogen ga orrosion don d in compari ed, namely (turnbull 19 l dissolution e of gaseou hodic) proce tudies a syst the corrosio f pit. gure 1 pourb e evolution according to asic corrosio al-khamis s, h2, forma ne by vuille ison to pourb fe3+, feo2 982; sharland n (anodic) pr us hydrogen ess, which is tem in sodiu on reaction, baix diagram of single co o sharland ( on model of i and pickeri ation inside p emin et al. (v baix diagram 2 and hfeo d 1988; shar ocess is give n formation given by: um chloride but are in m of iron orrosion pit (sharland 19 iron : fe2+, h ing (al-kha propagating vuillemin 2 m of iron (as o2-. the mo rland 1988): en by (sharla n through th e solution. h nvolved in t in carbon 989), six aqu h+, oh-, clamis 2001) pits. this is 007). due to shown in fi odel applies and 1988): he simultan however, sod maintaining steel ueous , na+ have s also o the igure s the (1) (2) (3) neous (4) dium g the (5) figure 1 pourbaix diagram of iron the reason for adding fecl+ is to accurately model situations in which the bulk chloride concentration is large and ion pairing is likely to be significant. furthermore, fecl+ is a stable species in standard conditions. the chemical reaction is taken as (sharland 1988; cottis 2004): 3 the reason for adding fecl+ is to accurately model situations in which the bulk chloride concentration is large and ion pairing is likely to be significant. furthermore, fecl+ is a stable species in standard conditions. the chemical reaction is taken as (sharland 1988; cottis 2004): ++ =+ feclclfe -2 (6) to obtain the respective currents for metal dissolution and reduction of h+, the model applies tafel expression (turnbull 1982; turnbull 1982; galvele 2005) and hence, the current density for iron oxidation, i1, and the current density for proton reduction, i2, are, respectively, given as: ( ) ⎥⎦ ⎤ ⎢⎣ ⎡ − = rt vv faii m1011 exp (7) [ ] ( )⎥⎦ ⎤ ⎢⎣ ⎡ − = + rt vv faii m2022 exph (8) where vm is the metal potential, v is the electrostatic potential in solution. ]h[ + is the concentration of h+, a1 and a2 are rate constants. the metal dissolution rate and the hydrogen ions reduction rate follow fick’s first law of diffusion (smith 2006) which stated that the kinetics in equations and produce their respective fluxes (turnbull 1987), as stated below: flux of metal ions +2fe (dissolution rate), f i j diss 2 1= (9) flux of hydrogen ions +h (reduction rate), f i j h 2−=+ (10) the fluxes above are now stated as: ( ) ⎥⎦ ⎤ ⎢⎣ ⎡ −× == rt vv fa f i f i j mdiss 1 011 exp 25.0 2 (11) [ ] ( ) ⎥⎦ ⎤ ⎢⎣ ⎡ −× =−= + + rt vv fa f i f i j m h 2 022 exp h (12) where -21101 ma107.2 ×=i , -17 02 molma102 ×=i , 11 =a and 5.02 =a (vuillemin 2007). the chemical reactions mentioned above apply the equilibrium constants at 25oc obtained from the hsc chemistry program. it is a chemical reaction and equilibrium software with extensive thermochemical databases. as mentioned earlier, the initial objective of this study is to look at the corrosion activities inside a pit at different potentials and ph values. for the decrease in ph, the model examines the ph (6) to obtain the respective currents for metal dissolution and reduction of h+, the model applies tafel expression (turnbull 1982; turnbull 1982; galvele 2005) and hence, the current density for iron oxidation, i1, and the current density for proton reduction, i2, are, respectively, given as: 3 the reason for adding fecl+ is to accurately model situations in which the bulk chloride concentration is large and ion pairing is likely to be significant. furthermore, fecl+ is a stable species in standard conditions. the chemical reaction is taken as (sharland 1988; cottis 2004): ++ =+ feclclfe -2 (6) to obtain the respective currents for metal dissolution and reduction of h+, the model applies tafel expression (turnbull 1982; turnbull 1982; galvele 2005) and hence, the current density for iron oxidation, i1, and the current density for proton reduction, i2, are, respectively, given as: ( ) ⎥⎦ ⎤ ⎢⎣ ⎡ − = rt vv faii m1011 exp (7) [ ] ( )⎥⎦ ⎤ ⎢⎣ ⎡ − = + rt vv faii m2022 exph (8) where vm is the metal potential, v is the electrostatic potential in solution. ]h[ + is the concentration of h+, a1 and a2 are rate constants. the metal dissolution rate and the hydrogen ions reduction rate follow fick’s first law of diffusion (smith 2006) which stated that the kinetics in equations and produce their respective fluxes (turnbull 1987), as stated below: flux of metal ions +2fe (dissolution rate), f i j diss 2 1= (9) flux of hydrogen ions +h (reduction rate), f i j h 2−=+ (10) the fluxes above are now stated as: ( ) ⎥⎦ ⎤ ⎢⎣ ⎡ −× == rt vv fa f i f i j mdiss 1 011 exp 25.0 2 (11) [ ] ( ) ⎥⎦ ⎤ ⎢⎣ ⎡ −× =−= + + rt vv fa f i f i j m h 2 022 exp h (12) where -21101 ma107.2 ×=i , -17 02 molma102 ×=i , 11 =a and 5.02 =a (vuillemin 2007). the chemical reactions mentioned above apply the equilibrium constants at 25oc obtained from the hsc chemistry program. it is a chemical reaction and equilibrium software with extensive thermochemical databases. as mentioned earlier, the initial objective of this study is to look at the corrosion activities inside a pit at different potentials and ph values. for the decrease in ph, the model examines the ph (7) issn: 2180-1053 vol. 4 no. 1 january-june 2012 journal of mechanical engineering and technology 4 3 the reason for adding fecl+ is to accurately model situations in which the bulk chloride concentration is large and ion pairing is likely to be significant. furthermore, fecl+ is a stable species in standard conditions. the chemical reaction is taken as (sharland 1988; cottis 2004): ++ =+ feclclfe -2 (6) to obtain the respective currents for metal dissolution and reduction of h+, the model applies tafel expression (turnbull 1982; turnbull 1982; galvele 2005) and hence, the current density for iron oxidation, i1, and the current density for proton reduction, i2, are, respectively, given as: ( ) ⎥⎦ ⎤ ⎢⎣ ⎡ − = rt vv faii m1011 exp (7) [ ] ( )⎥⎦ ⎤ ⎢⎣ ⎡ − = + rt vv faii m2022 exph (8) where vm is the metal potential, v is the electrostatic potential in solution. ]h[ + is the concentration of h+, a1 and a2 are rate constants. the metal dissolution rate and the hydrogen ions reduction rate follow fick’s first law of diffusion (smith 2006) which stated that the kinetics in equations and produce their respective fluxes (turnbull 1987), as stated below: flux of metal ions +2fe (dissolution rate), f i j diss 2 1= (9) flux of hydrogen ions +h (reduction rate), f i j h 2−=+ (10) the fluxes above are now stated as: ( ) ⎥⎦ ⎤ ⎢⎣ ⎡ −× == rt vv fa f i f i j mdiss 1 011 exp 25.0 2 (11) [ ] ( ) ⎥⎦ ⎤ ⎢⎣ ⎡ −× =−= + + rt vv fa f i f i j m h 2 022 exp h (12) where -21101 ma107.2 ×=i , -17 02 molma102 ×=i , 11 =a and 5.02 =a (vuillemin 2007). the chemical reactions mentioned above apply the equilibrium constants at 25oc obtained from the hsc chemistry program. it is a chemical reaction and equilibrium software with extensive thermochemical databases. as mentioned earlier, the initial objective of this study is to look at the corrosion activities inside a pit at different potentials and ph values. for the decrease in ph, the model examines the ph (8) where vm is the metal potential, v is the electrostatic potential in solution. ]h[ + is the concentration of h+, a1 and a2 are rate constants. the metal dissolution rate and the hydrogen ions reduction rate follow fick’s first law of diffusion (smith 2006) which stated that the kinetics in equations and produce their respective fluxes (turnbull 1987), as stated below: 3 the reason for adding fecl+ is to accurately model situations in which the bulk chloride concentration is large and ion pairing is likely to be significant. furthermore, fecl+ is a stable species in standard conditions. the chemical reaction is taken as (sharland 1988; cottis 2004): ++ =+ feclclfe -2 (6) to obtain the respective currents for metal dissolution and reduction of h+, the model applies tafel expression (turnbull 1982; turnbull 1982; galvele 2005) and hence, the current density for iron oxidation, i1, and the current density for proton reduction, i2, are, respectively, given as: ( ) ⎥⎦ ⎤ ⎢⎣ ⎡ − = rt vv faii m1011 exp (7) [ ] ( )⎥⎦ ⎤ ⎢⎣ ⎡ − = + rt vv faii m2022 exph (8) where vm is the metal potential, v is the electrostatic potential in solution. ]h[ + is the concentration of h+, a1 and a2 are rate constants. the metal dissolution rate and the hydrogen ions reduction rate follow fick’s first law of diffusion (smith 2006) which stated that the kinetics in equations and produce their respective fluxes (turnbull 1987), as stated below: flux of metal ions +2fe (dissolution rate), f i j diss 2 1= (9) flux of hydrogen ions +h (reduction rate), f i j h 2−=+ (10) the fluxes above are now stated as: ( ) ⎥⎦ ⎤ ⎢⎣ ⎡ −× == rt vv fa f i f i j mdiss 1 011 exp 25.0 2 (11) [ ] ( ) ⎥⎦ ⎤ ⎢⎣ ⎡ −× =−= + + rt vv fa f i f i j m h 2 022 exp h (12) where -21101 ma107.2 ×=i , -17 02 molma102 ×=i , 11 =a and 5.02 =a (vuillemin 2007). the chemical reactions mentioned above apply the equilibrium constants at 25oc obtained from the hsc chemistry program. it is a chemical reaction and equilibrium software with extensive thermochemical databases. as mentioned earlier, the initial objective of this study is to look at the corrosion activities inside a pit at different potentials and ph values. for the decrease in ph, the model examines the ph (9) 3 the reason for adding fecl+ is to accurately model situations in which the bulk chloride concentration is large and ion pairing is likely to be significant. furthermore, fecl+ is a stable species in standard conditions. the chemical reaction is taken as (sharland 1988; cottis 2004): ++ =+ feclclfe -2 (6) to obtain the respective currents for metal dissolution and reduction of h+, the model applies tafel expression (turnbull 1982; turnbull 1982; galvele 2005) and hence, the current density for iron oxidation, i1, and the current density for proton reduction, i2, are, respectively, given as: ( ) ⎥⎦ ⎤ ⎢⎣ ⎡ − = rt vv faii m1011 exp (7) [ ] ( )⎥⎦ ⎤ ⎢⎣ ⎡ − = + rt vv faii m2022 exph (8) where vm is the metal potential, v is the electrostatic potential in solution. ]h[ + is the concentration of h+, a1 and a2 are rate constants. the metal dissolution rate and the hydrogen ions reduction rate follow fick’s first law of diffusion (smith 2006) which stated that the kinetics in equations and produce their respective fluxes (turnbull 1987), as stated below: flux of metal ions +2fe (dissolution rate), f i j diss 2 1= (9) flux of hydrogen ions +h (reduction rate), f i j h 2−=+ (10) the fluxes above are now stated as: ( ) ⎥⎦ ⎤ ⎢⎣ ⎡ −× == rt vv fa f i f i j mdiss 1 011 exp 25.0 2 (11) [ ] ( ) ⎥⎦ ⎤ ⎢⎣ ⎡ −× =−= + + rt vv fa f i f i j m h 2 022 exp h (12) where -21101 ma107.2 ×=i , -17 02 molma102 ×=i , 11 =a and 5.02 =a (vuillemin 2007). the chemical reactions mentioned above apply the equilibrium constants at 25oc obtained from the hsc chemistry program. it is a chemical reaction and equilibrium software with extensive thermochemical databases. as mentioned earlier, the initial objective of this study is to look at the corrosion activities inside a pit at different potentials and ph values. for the decrease in ph, the model examines the ph (10) the fluxes above are now stated as: 3 the reason for adding fecl+ is to accurately model situations in which the bulk chloride concentration is large and ion pairing is likely to be significant. furthermore, fecl+ is a stable species in standard conditions. the chemical reaction is taken as (sharland 1988; cottis 2004): ++ =+ feclclfe -2 (6) to obtain the respective currents for metal dissolution and reduction of h+, the model applies tafel expression (turnbull 1982; turnbull 1982; galvele 2005) and hence, the current density for iron oxidation, i1, and the current density for proton reduction, i2, are, respectively, given as: ( ) ⎥⎦ ⎤ ⎢⎣ ⎡ − = rt vv faii m1011 exp (7) [ ] ( )⎥⎦ ⎤ ⎢⎣ ⎡ − = + rt vv faii m2022 exph (8) where vm is the metal potential, v is the electrostatic potential in solution. ]h[ + is the concentration of h+, a1 and a2 are rate constants. the metal dissolution rate and the hydrogen ions reduction rate follow fick’s first law of diffusion (smith 2006) which stated that the kinetics in equations and produce their respective fluxes (turnbull 1987), as stated below: flux of metal ions +2fe (dissolution rate), f i j diss 2 1= (9) flux of hydrogen ions +h (reduction rate), f i j h 2−=+ (10) the fluxes above are now stated as: ( ) ⎥⎦ ⎤ ⎢⎣ ⎡ −× == rt vv fa f i f i j mdiss 1 011 exp 25.0 2 (11) [ ] ( ) ⎥⎦ ⎤ ⎢⎣ ⎡ −× =−= + + rt vv fa f i f i j m h 2 022 exp h (12) where -21101 ma107.2 ×=i , -17 02 molma102 ×=i , 11 =a and 5.02 =a (vuillemin 2007). the chemical reactions mentioned above apply the equilibrium constants at 25oc obtained from the hsc chemistry program. it is a chemical reaction and equilibrium software with extensive thermochemical databases. as mentioned earlier, the initial objective of this study is to look at the corrosion activities inside a pit at different potentials and ph values. for the decrease in ph, the model examines the ph (11) 3 the reason for adding fecl+ is to accurately model situations in which the bulk chloride concentration is large and ion pairing is likely to be significant. furthermore, fecl+ is a stable species in standard conditions. the chemical reaction is taken as (sharland 1988; cottis 2004): ++ =+ feclclfe -2 (6) to obtain the respective currents for metal dissolution and reduction of h+, the model applies tafel expression (turnbull 1982; turnbull 1982; galvele 2005) and hence, the current density for iron oxidation, i1, and the current density for proton reduction, i2, are, respectively, given as: ( ) ⎥⎦ ⎤ ⎢⎣ ⎡ − = rt vv faii m1011 exp (7) [ ] ( )⎥⎦ ⎤ ⎢⎣ ⎡ − = + rt vv faii m2022 exph (8) where vm is the metal potential, v is the electrostatic potential in solution. ]h[ + is the concentration of h+, a1 and a2 are rate constants. the metal dissolution rate and the hydrogen ions reduction rate follow fick’s first law of diffusion (smith 2006) which stated that the kinetics in equations and produce their respective fluxes (turnbull 1987), as stated below: flux of metal ions +2fe (dissolution rate), f i j diss 2 1= (9) flux of hydrogen ions +h (reduction rate), f i j h 2−=+ (10) the fluxes above are now stated as: ( ) ⎥⎦ ⎤ ⎢⎣ ⎡ −× == rt vv fa f i f i j mdiss 1 011 exp 25.0 2 (11) [ ] ( ) ⎥⎦ ⎤ ⎢⎣ ⎡ −× =−= + + rt vv fa f i f i j m h 2 022 exp h (12) where -21101 ma107.2 ×=i , -17 02 molma102 ×=i , 11 =a and 5.02 =a (vuillemin 2007). the chemical reactions mentioned above apply the equilibrium constants at 25oc obtained from the hsc chemistry program. it is a chemical reaction and equilibrium software with extensive thermochemical databases. as mentioned earlier, the initial objective of this study is to look at the corrosion activities inside a pit at different potentials and ph values. for the decrease in ph, the model examines the ph (12) 3 the reason for adding fecl+ is to accurately model situations in which the bulk chloride concentration is large and ion pairing is likely to be significant. furthermore, fecl+ is a stable species in standard conditions. the chemical reaction is taken as (sharland 1988; cottis 2004): ++ =+ feclclfe -2 (6) to obtain the respective currents for metal dissolution and reduction of h+, the model applies tafel expression (turnbull 1982; turnbull 1982; galvele 2005) and hence, the current density for iron oxidation, i1, and the current density for proton reduction, i2, are, respectively, given as: ( ) ⎥⎦ ⎤ ⎢⎣ ⎡ − = rt vv faii m1011 exp (7) [ ] ( )⎥⎦ ⎤ ⎢⎣ ⎡ − = + rt vv faii m2022 exph (8) where vm is the metal potential, v is the electrostatic potential in solution. ]h[ + is the concentration of h+, a1 and a2 are rate constants. the metal dissolution rate and the hydrogen ions reduction rate follow fick’s first law of diffusion (smith 2006) which stated that the kinetics in equations and produce their respective fluxes (turnbull 1987), as stated below: flux of metal ions +2fe (dissolution rate), f i j diss 2 1= (9) flux of hydrogen ions +h (reduction rate), f i j h 2−=+ (10) the fluxes above are now stated as: ( ) ⎥⎦ ⎤ ⎢⎣ ⎡ −× == rt vv fa f i f i j mdiss 1 011 exp 25.0 2 (11) [ ] ( ) ⎥⎦ ⎤ ⎢⎣ ⎡ −× =−= + + rt vv fa f i f i j m h 2 022 exp h (12) where -21101 ma107.2 ×=i , -17 02 molma102 ×=i , 11 =a and 5.02 =a (vuillemin 2007). the chemical reactions mentioned above apply the equilibrium constants at 25oc obtained from the hsc chemistry program. it is a chemical reaction and equilibrium software with extensive thermochemical databases. as mentioned earlier, the initial objective of this study is to look at the corrosion activities inside a pit at different potentials and ph values. for the decrease in ph, the model examines the ph (vuillemin 2007). the chemical reactions mentioned above apply the equilibrium constants at 25oc obtained from the hsc chemistry program. it is a chemical reaction and equilibrium software with extensive thermochemical databases. as mentioned earlier, the initial objective of this study is to look at the corrosion activities inside a pit at different potentials and ph values. for the decrease in ph, the model examines the ph changes by incorporating the equation below and as a result, the model observes this decrease in ph: issn: 2180-1053 vol. 4 no. 1 january-june 2012 a theoretical model of pitting corrosion using a general purpose finite element package 5 4 changes by incorporating the equation below and as a result, the model observes this decrease in ph: [ ]( )+×−= habs001.0logph 10 (13) the term inside the bracket is multiplied by 0.001 because of unit conversion from litre to cubic meter. figure 2 geometry representing a single pit with a curved bottom working with comsol multiphysics requires the specification of initial geometry. to represent a pit, the geometry chosen is as shown in figure 2. region 1 and 3 represent the passive walls, region 2 represent the mouth of the pit and regions 4 and 5 are the active bottom of the pit where metal dissolution is expected to be observed. the model applies the neutral environment of ph 7 and takes the initial chloride concentration to be 1000 mol per cubic meter. it is solved using the nernstplanck resolution for the range of potential between -1.5 and 1 volt. 3.0 results and discussion corrosion rates of a metal can be measured by applying a current (the movement of electrical charges carried by electrons) to produce a polarization curve. the polarization curve is the degree of potential change as a function of the amount of current applied. the degree of polarization is a measure of how the rates of the anodic and the cathodic reactions are affected by various environmental conditions. figure 3 shows that, the corrosion potential is about -0.85 volt. as potential increases, the current density, log i, increases and this indicates active dissolution of metal. further down the pit, the ph is expected to be lower than the ph at the mouth. the higher the concentration of h+, the lower the ph is going to be. figure 4 shows that as the potential increases, hydrogen ions gradually accumulate at the bottom of the pit. this is shown in the experiment done by lee et al. (lee 1981). the increase in hydrogen ion concentration from almost non-existent (concentration 4101 −× mol m-3) to 0.0218 mol m-3 will result in ph change. this agrees with the model which is shown in figure 5 where the ph inside the pit drops to 4.7. mouth of pit bottom of pit 1 2 3 4 5 (13) the term inside the bracket is multiplied by 0.001 because of unit conversion from litre to cubic meter. 4 changes by incorporating the equation below and as a result, the model observes this decrease in ph: [ ]( )+×−= habs001.0logph 10 (13) the term inside the bracket is multiplied by 0.001 because of unit conversion from litre to cubic meter. figure 2 geometry representing a single pit with a curved bottom working with comsol multiphysics requires the specification of initial geometry. to represent a pit, the geometry chosen is as shown in figure 2. region 1 and 3 represent the passive walls, region 2 represent the mouth of the pit and regions 4 and 5 are the active bottom of the pit where metal dissolution is expected to be observed. the model applies the neutral environment of ph 7 and takes the initial chloride concentration to be 1000 mol per cubic meter. it is solved using the nernstplanck resolution for the range of potential between -1.5 and 1 volt. 3.0 results and discussion corrosion rates of a metal can be measured by applying a current (the movement of electrical charges carried by electrons) to produce a polarization curve. the polarization curve is the degree of potential change as a function of the amount of current applied. the degree of polarization is a measure of how the rates of the anodic and the cathodic reactions are affected by various environmental conditions. figure 3 shows that, the corrosion potential is about -0.85 volt. as potential increases, the current density, log i, increases and this indicates active dissolution of metal. further down the pit, the ph is expected to be lower than the ph at the mouth. the higher the concentration of h+, the lower the ph is going to be. figure 4 shows that as the potential increases, hydrogen ions gradually accumulate at the bottom of the pit. this is shown in the experiment done by lee et al. (lee 1981). the increase in hydrogen ion concentration from almost non-existent (concentration 4101 −× mol m-3) to 0.0218 mol m-3 will result in ph change. this agrees with the model which is shown in figure 5 where the ph inside the pit drops to 4.7. mouth of pit bottom of pit 1 2 3 4 5 figure 2 geometry representing a single pit with a curved bottom working with comsol multiphysics requires the specification of initial geometry. to represent a pit, the geometry chosen is as shown in figure 2. region 1 and 3 represent the passive walls, region 2 represent the mouth of the pit and regions 4 and 5 are the active bottom of the pit where metal dissolution is expected to be observed. the model applies the neutral environment of ph 7 and takes the initial chloride concentration to be 1000 mol per cubic meter. it is solved using the nernst-planck resolution for the range of potential between -1.5 and 1 volt. 3.0 results and discussion corrosion rates of a metal can be measured by applying a current (the movement of electrical charges carried by electrons) to produce a polarization curve. the polarization curve is the degree of potential change as a function of the amount of current applied. the degree of polarization is a measure of how the rates of the anodic and the cathodic reactions are affected by various environmental conditions. figure 3 shows that, the corrosion potential is about -0.85 volt. as potential increases, the current density, log i, increases and this indicates active dissolution of metal. further down the pit, the ph is expected to be lower than the ph at the mouth. the higher the concentration of h+, the lower the ph is going to be. figure 4 shows that as the potential increases, hydrogen issn: 2180-1053 vol. 4 no. 1 january-june 2012 journal of mechanical engineering and technology 6 ions gradually accumulate at the bottom of the pit. this is shown in the experiment done by lee et.al. (lee 1981). the increase in hydrogen ion concentration from almost non-existent (concentration 4101 −× mol m-3) to 0.0218 mol m-3 will result in ph change. this agrees with the model which is shown in figure 5 where the ph inside the pit drops to 4.7. 5 figure 3 graph showing the polarization curve at the bottom of pit (i) (ii) figure 3 graph showing the polarization curve at the bottom of pit 5 figure 3 graph showing the polarization curve at the bottom of pit (i) (ii) (i) 5 figure 3 graph showing the polarization curve at the bottom of pit (i) (ii) (ii) issn: 2180-1053 vol. 4 no. 1 january-june 2012 a theoretical model of pitting corrosion using a general purpose finite element package 7 6 (iii) figure 4 graphs (i)-(iii) show the distribution of h+ inside the pit. (i) concentration of h+ at -0.7 volt (cathodic). (ii) concentration of h+ at 0 volt (anodic) (iii) concentration of h+ at 0.5 volt figure 5 graph shows difference in ph at three points along the pit ph change still occurs at the point where the active region meets the passive region. however, the ph is shown to be slightly higher compared to the ph further down the pit. it can also be seen from the graph that at very low potential (around -1.5 volt), the condition inside the pit is very alkaline compared to the condition at the mouth. relating figure 3 and 5, the metal is seen to act as cathode at potential less than -0.85 volt. this is the corrosion potential, ecorr. at potential higher than -0.85 volt, increase in current density occurs at the region where the metal acts as cathode, as indicated by the region at the bottom of the pit. since the increase in current density indicates corrosion activities, the ph of the purported region is expected to decrease. sato (sato 1995) stated that the 4.7 (iii) figure 4 graphs (i)-(iii) show the distribution of h+ inside the pit. (i) concentration of h+ at -0.7 volt (cathodic). (ii) concentration of h+ at 0 volt (anodic) (iii) concentration of h+ at 0.5 volt 6 (iii) figure 4 graphs (i)-(iii) show the distribution of h+ inside the pit. (i) concentration of h+ at -0.7 volt (cathodic). (ii) concentration of h+ at 0 volt (anodic) (iii) concentration of h+ at 0.5 volt figure 5 graph shows difference in ph at three points along the pit ph change still occurs at the point where the active region meets the passive region. however, the ph is shown to be slightly higher compared to the ph further down the pit. it can also be seen from the graph that at very low potential (around -1.5 volt), the condition inside the pit is very alkaline compared to the condition at the mouth. relating figure 3 and 5, the metal is seen to act as cathode at potential less than -0.85 volt. this is the corrosion potential, ecorr. at potential higher than -0.85 volt, increase in current density occurs at the region where the metal acts as cathode, as indicated by the region at the bottom of the pit. since the increase in current density indicates corrosion activities, the ph of the purported region is expected to decrease. sato (sato 1995) stated that the 4.7 figure 5 graph shows difference in ph at three points along the pit ph change still occurs at the point where the active region meets the passive region. however, the ph is shown to be slightly higher compared to the ph further down the pit. it can also be seen from the graph that at very low potential (around -1.5 volt), the condition inside the pit is very alkaline compared to the condition at the mouth. relating figure 3 and 5, the metal is seen to act as cathode at potential less than -0.85 volt. this is the corrosion potential, ecorr. at potential higher than -0.85 volt, increase in current density occurs at the region where the metal acts as cathode, as indicated by the region at the bottom of the pit. since the increase in current density indicates corrosion activities, the ph of the purported region is expected to decrease. sato (sato 1995) stated that the ph in the pit is the major factor that determines the evolution (passivation or propagation) of an active pit. this is shown true by the model. issn: 2180-1053 vol. 4 no. 1 january-june 2012 journal of mechanical engineering and technology 8 it can also be seen that as the potential increases, the slope of the current density starts to decrease starting from around -0.3 volt. this is the result of the changes in concentrations of ionic species inside the pit. it is interesting to see how one ionic concentration effects or relates to the concentration of another ionic species. figure 6 illustrates the profiles of species concentration and shows the possible reaction occurring at different potentials. it is remarkable to see that more corrosion activities occur once water dissociates through the process of reduction. water dissociates at potential -0.65 volt. this is explained later in this paragraph using the pourbaix diagram in figure 7. when this happens, the environment now has hydrogen and hydroxyl ions. the existence of hydrogen ions inside the pit attracts the chloride ions, cl-, into the pit and thus, creating an acidic environment. this reaction causes the ph value to decrease (sato 1995). when metal is placed in an acidic solution, a vigorous reaction occurs. immediately, metal ion, fe2+, is oxidized and simultaneously, reduction of hydrogen ions occurs. this is a rapid process and the hydrogen gas is expected to be seen at this stage, as shown in the experiment done by pickering (pickering 1984) and al-khamis and pickering (al-khamis 2001). this is indicated in the figure below, which shows a sudden ‘jump’ in the ‘concentration’ of hydrogen gas. meanwhile, the ohconcentration is seen to decrease after -0.65 volt. this is in line with the theory of equilibrium constant of water. 7 ph in the pit is the major factor that determines the evolution (passivation or propagation) of an active pit. this is shown true by the model. it can also be seen that as the potential increases, the slope of the current density starts to decrease starting from around -0.3 volt. this is the result of the changes in concentrations of ionic species inside the pit. it is interesting to see how one ionic concentration effects or relates to the concentration of another ionic species. figure 6 illustrates the profiles of species concentration and shows the possible reaction occurring at different potentials. it is remarkable to see that more corrosion activities occur once water dissociates through the process of reduction. water dissociates at potential -0.65 volt. this is explained later in this paragraph using the pourbaix diagram in figure 7. when this happens, the environment now has hydrogen and hydroxyl ions. the existence of hydrogen ions inside the pit attracts the chloride ions, cl-, into the pit and thus, creating an acidic environment. this reaction causes the ph value to decrease (sato 1995). when metal is placed in an acidic solution, a vigorous reaction occurs. immediately, metal ion, fe2+, is oxidized and simultaneously, reduction of hydrogen ions occurs. this is a rapid process and the hydrogen gas is expected to be seen at this stage, as shown in the experiment done by pickering (pickering 1984) and al-khamis and pickering (al-khamis 2001). this is indicated in the figure below, which shows a sudden ‘jump’ in the ‘concentration’ of hydrogen gas. meanwhile, the ohconcentration is seen to decrease after -0.65 volt. this is in line with the theory of equilibrium constant of water. figure 6 profiles of concentration of all species at the bottom of the pit the phenomenon can also be related to the pourbaix diagram in figure 7 shown below. at ph 7 and potential -0.65 volt, the system is crossing the pourbaix line where the metal is crossing between two zones: zone ‘immunity’ into the zone ‘corrosion’. this indicates that fe2+ is produced at these potential and ph condition. the presence of fe2+ immediately triggers ionization of water and thus, this dissociation produces h+ and oh-. from the concentration profiles plot above, it can be seen that the amount of hydrogen gas starts to reduce almost immediately at potential about -0.3 volts. figure 6 profiles of concentration of all species at the bottom of the pit the phenomenon can also be related to the pourbaix diagram in figure 7 shown below. at ph 7 and potential -0.65 volt, the system is crossing the pourbaix line where the metal is crossing between two zones: zone ‘immunity’ into the zone ‘corrosion’. this indicates that fe2+ is produced issn: 2180-1053 vol. 4 no. 1 january-june 2012 a theoretical model of pitting corrosion using a general purpose finite element package 9 at these potential and ph condition. the presence of fe2+ immediately triggers ionization of water and thus, this dissociation produces h+ and oh-. from the concentration profiles plot above, it can be seen that the amount of hydrogen gas starts to reduce almost immediately at potential about -0.3 volts. f a g j a v t figure7 p from figure about 5, the gas. below just as indic figure 8 po as mentione volts. even that corrosio pourbaix diag e 8, it can be system is cr line f, water ated by the m ourbaix diagr ed above, th though this on of metal s -0.65 -0.3 gram showin e seen that w rossing line f r is reduced model. ram showing h he slope of t happens, th still occurrin 5 ng the point ion when the pot f, where wat to hydrogen g the point y h2, when sys the current he current de ng. from fig 5 oy x where the ns, fe2+ tential is abo ter is thermo n and hence y where the m stem passes density start ensity still m gure 9, the c o x e f 7 e model indic out -0.3 volt odynamically , hydrogen i model indica line f ts to decreas maintain at a oncentration cates the pre t and the ph y stable rath is thermodyn ates the loss se starting f a high level n of fe2+ at t esence of me h of the syste er than hydr namically st of hydrogen from around l which indi the bottom o etal em is rogen table, n gas, d -0.3 cates of the figure7 pourbaix diagram showing the point x where the model indicates the presence of metal ions, fe2+ from figure 8, it can be seen that when the potential is about -0.3 volt and the ph of the system is about 5, the system is crossing line f, where water is thermodynamically stable rather than hydrogen gas. below line f, water is reduced to hydrogen and hence, hydrogen is thermodynamically stable, just as indicated by the model. f a g j a v t figure7 p from figure about 5, the gas. below just as indic figure 8 po as mentione volts. even that corrosio pourbaix diag e 8, it can be system is cr line f, water ated by the m ourbaix diagr ed above, th though this on of metal s -0.65 -0.3 gram showin e seen that w rossing line f r is reduced model. ram showing h he slope of t happens, th still occurrin 5 ng the point ion when the pot f, where wat to hydrogen g the point y h2, when sys the current he current de ng. from fig 5 oy x where the ns, fe2+ tential is abo ter is thermo n and hence y where the m stem passes density start ensity still m gure 9, the c o x e f 7 e model indic out -0.3 volt odynamically , hydrogen i model indica line f ts to decreas maintain at a oncentration cates the pre t and the ph y stable rath is thermodyn ates the loss se starting f a high level n of fe2+ at t esence of me h of the syste er than hydr namically st of hydrogen from around l which indi the bottom o etal em is rogen table, n gas, d -0.3 cates of the figure 8 pourbaix diagram showing the point y where the model indicates the loss of hydrogen gas, h2, when system passes line f issn: 2180-1053 vol. 4 no. 1 january-june 2012 journal of mechanical engineering and technology 10 as mentioned above, the slope of the current density starts to decrease starting from around -0.3 volts. even though this happens, the current density still maintain at a high level which indicates that corrosion of metal still occurring. from figure 9, the concentration of fe2+ at the bottom of the pit is indicated to be slightly higher than that at the side of the pit where the active region meets the passive region. metal ions, fe2+, starts to accumulate gradually at the bottom of the pit. this is the region where the model is set to be active. increasing concentration of fe2+ at the bottom of the pit indicates that the metal dissolution is occurring continuously. from the figure also, it can be observed that fe2+ starting to accumulate as we go higher up the pit. this agrees with the theory of diffusion where species move under the action of concentration gradient. this process involves species moving from high to low concentration until even concentration is achieved for all species. in this case, the concentration of fe2+ has reached a limit where the metal ions diffuse to the area where the concentration of fe2+ is lower which is the passive region higher up the pit. it is expected that fe2+ will reach a point where it will diffuse out of the pit, as stated by grimm and landolt (grimm 1994), and thus, promotes dissolution of metal, forming general corrosion to the metal surface. 9 pit is indicated to be slightly higher than that at the side of the pit where the active region meets the passive region. metal ions, fe2+, starts to accumulate gradually at the bottom of the pit. this is the region where the model is set to be active. increasing concentration of fe2+ at the bottom of the pit indicates that the metal dissolution is occurring continuously. from the figure also, it can be observed that fe2+ starting to accumulate as we go higher up the pit. this agrees with the theory of diffusion where species move under the action of concentration gradient. this process involves species moving from high to low concentration until even concentration is achieved for all species. in this case, the concentration of fe2+ has reached a limit where the metal ions diffuse to the area where the concentration of fe2+ is lower which is the passive region higher up the pit. it is expected that fe2+ will reach a point where it will diffuse out of the pit, as stated by grimm and landolt (grimm 1994), and thus, promotes dissolution of metal, forming general corrosion to the metal surface. (i) (ii) issn: 2180-1053 vol. 4 no. 1 january-june 2012 a theoretical model of pitting corrosion using a general purpose finite element package 11 10 (iii) (iv) figure 9 figure (i)-(iv) show the stages with respect to potential, where accumulation of fe2+ occurs at the bottom of the pit with the accumulation of fe2+ inside the pit, this gives a positively charged environment inside the pit. this phenomenon attracts negatively charged anions into the pit through the process of migration. in this case, negatively charged chloride ions, cl-, migrate into the pit and this is shown by the model in figure 10. from the figure, it is observed that the concentration of clis half the concentration of fe2+ when comparing their concentration at each potential. this also agrees with the fact that fe2+ has valence of 2 compared to cl-, which has valence 1. with increasing concentration of clinside the pit, this in turn gives a negatively charged environment inside the pit and thus, attracts the h+ ions through the dissociation of water in the bulk solution. accumulation of h+ occurs inside the pit, as shown by figure 11error! reference source not found.. figure 9 figure (i)-(iv) show the stages with respect to potential, where accumulation of fe2+ occurs at the bottom of the pit with the accumulation of fe2+ inside the pit, this gives a positively charged environment inside the pit. this phenomenon attracts negatively charged anions into the pit through the process of migration. in this case, negatively charged chloride ions, cl-, migrate into the pit and this is shown by the model in figure 10. from the figure, it is observed that the concentration of clis half the concentration of fe2+ when comparing their concentration at each potential. this also agrees with the fact that fe2+ has valence of 2 compared to cl-, which has valence 1. with increasing concentration of clinside the pit, this in turn gives a negatively charged environment inside the pit and thus, attracts the h+ ions through the dissociation of water in the bulk solution. accumulation of h+ occurs inside the pit, as shown by figure 11error! reference source not found. issn: 2180-1053 vol. 4 no. 1 january-june 2012 journal of mechanical engineering and technology 12 11 figure 10 graph showing concentration of clat three points along the pit figure 11 graph showing concentration of h+ at three points along the pit 4.0 conclusions the model is constructed as a first step to develop a model that can represent a propagating pit. it is suggested that salt film formation and passivity are considered in the next stage of modelling. figure 10 graph showing concentration of clat three points along the pit 11 figure 10 graph showing concentration of clat three points along the pit figure 11 graph showing concentration of h+ at three points along the pit 4.0 conclusions the model is constructed as a first step to develop a model that can represent a propagating pit. it is suggested that salt film formation and passivity are considered in the next stage of modelling. figure 11 graph showing concentration of h+ at three points along the pit 4.0 conclusions the model is constructed as a first step to develop a model that can represent a propagating pit. it is suggested that salt film formation and passivity are considered in the next stage of modelling. issn: 2180-1053 vol. 4 no. 1 january-june 2012 a theoretical model of pitting corrosion using a general purpose finite element package 13 5.0 references ahmad, z. 2006. principles of corrosion engineering and corrosion control, butterworth-heinemann. al-khamis, j. n., pickering,h.w. 2001. “ir mechanism of crevice corrosion for alloy t-2205 duplex stainless steel in acidic-chloride media.” journal of electrochemical society 148(8): b314-b321. burstein, g. t., pistorius, p.c., and mattin, s.p. 1993. “the nucleation and growth of corrosion pits on stainless steel.” corrosion science 35(1-4): 57-62. cheng, y. f., luo, j.l. 2000. “a comparison of the pitting susceptibility and semiconducting properties of the passive films on carbon steel in chromate and bicarbonate solutions.” applied surface science 167: 113-121. cottis, r. a., mousson,j.l., vuillemin,b., and oltra,r. 2004. “use of a general purpose finite element package for modeling of crevice corrosion.” corrosion 04066. cui, n., ma, h.y., luo, j.l., and chiovelli, s. 2001. “use of general reference electrode technique for characterizing pitting and general corrosion of carbon steel in neutral media.” electrochemistry communications 3: 716-721. fontana, m. g. 1987. corrosion engineering. singapore, mcgraw-hill. galvele, j. r. 2005. “tafel’s law in pitting corrosion and crevice corrosion susceptibility.” corrosion science 47: 3053-3067. grimm, r. d., landolt, d. 1994. “salt films formed during mass transport controlled dissolution of iron-chromium alloys in concentrated chloride media.” corrosion science 36(11): 1874-1868. laycock, n. j., white,s.p. 2001. “computer simulation of single pit propagation in stainless steel under potentiostatic control.” journal of the electrochemistry society 148(7): b264-b275. lee, y. h., takehara, z. and yoshizawa, s. 1981. “the enrichment of hydrogen and chloride ions in the crevice corrosion of steels.” corrosion science 21(5): 391-397. perez, n. 2004. electrochemistry and corrosion science. united states of america, kluwer academic publisher. pickering, h. w. 1984. role of gas bubbles and cavity dimensions on the e-phion concentrations inside cavities. proceeding electrochemical society. sato, n. 1995. “the stability of localized corrosion.” corrosion science 37(12): 1947-1967. issn: 2180-1053 vol. 4 no. 1 january-june 2012 journal of mechanical engineering and technology 14 sharland, s. m. 1988. “a mathematical model of crevice and pitting corrosion ii. the mathematical solution.” corrosion science 28(6): 621-630. sharland, s. m., jackson, c.p., and diver,a.j. 1989. “a finite-element model of the propagation of corrosion crevices and pits.” corrosion science 29(9): 1149-1166. sharland, s. m., tasker,p.w 1988. “a mathematical model of crevice and pitting corrosion i. the physical model.” corrosion science 28(6): 603-620. shreir, l. l., jarman, r.a, burstein, g.t. 2000. corrosion : corrosion control, butterworth-heinemann. shreir, l. l., jarman, r.a., burstein,g.t. 2000. corrosion : metal/environment reactions, butterworth-heinemann. smith, w. f., hashemi, j. 2006. foundations of materials science and engineering. singapore, mcgraw-hill. turnbull, a. 1987. “mathematical modelling of the electrochemistry in corrosion fatigue cracks in steel corroding in marine environments.” corrosion science 27(12): 1323-1350. turnbull, a., gardner, m.k. 1982. “electrochemical polarization studies of bs 4360 50d steel in 3.5% nacl.” corrosion science 22(7): 661-673. turnbull, a., thomas,j.g.n 1982. “a model of crack electrochemistry for steels in the active state based on mass transport by diffusion and ion migration.” journal of electrochemical society 129(7). vuillemin, b., oltra, r., cottis, r., and crusset, d. 2007. “consideration of the formation of solids and gases in steady state modelling of crevice corrosion propagation.” electrochemica acta 52: 7570-7576. issn: 2180-1053 vol. 4 no. 1 january-june 2012 counterflow combustion of micro organic particles 97 counterflow combustion of micro organic particles mehdi bidabadi1, ali esmaeilnejad2, sirousfarshadi production technology department -industrial education college beni-suef university, industrial engineering department –jazan university. email: waleedshewakh@hotmail.com abstract the structure ofcounterflowpremixed flames in an axisymmetric configuration, containing uniformly distributed volatile fuel particles, with nonunity lewis number of the fuel are examined. it is presumed that the gaseous fuel, produced from vaporization of the fuel particles, oxidizes in the gas phase and the fuel particles do not participate in the reaction. the analysis is carried out in the asymptotic limit for large values of zeldovich number.a one-step reaction 82 counterflow combustion of micro organic particles mehdi bidabadi1, ali esmaeilnejad2, sirousfarshadi faculty of mechanical engineering, iran university of science and technology, tehran, iran corresponding email: sirous.farshadi@gmail.com abstract the structure ofcounterflowpremixed flames in an axisymmetric configuration, containing uniformly distributed volatile fuel particles, with nonunity lewis number of the fuel are examined. it is presumed that the gaseous fuel, produced from vaporization of the fuel particles, oxidizes in the gas phase and the fuel particles do not participate in the reaction. the analysis is carried out in the asymptotic limit for large values of zeldovich number.a one-step reaction is assumed. the flame position is determined andthe effect of lewis number change on the gaseous fuel mass fraction distribution is investigated. keywords: counterflow combustion; nonunity lewis number; organic particles; particle combustion 1.0 introduction many studies of dust clouds combustion have been published over the last few years; see, for example, introduction of the article on flame propagation through micro-organic dust particles (bidabadi et al., 2010). but from the fact that in many practical applications the flow field is appreciably strained, to yield realistic flame prediction under such conditions, counterflow configuration is suitable for studying these cases.over the last few decades, the counterflow configuration has been extensively adopted in theoretical, experimental and numerical studies as a means to investigate various physical effects on real flames on real flames, such as stretch, preferential diffusion, radiation and chemical kinetics (daou, 2011), (thatcher et al.) and (wang et al., 2007). but these studies are done for gaseous fuels. eckhoff clarified the differences and similarities between dust and gases (eckhof, 2006). it has been concluded that there are two basic differences between dusts and gases which are of substantially greater significance in design of safety standards than these similarities. firstly, the physics of generation and up-keeping of dust clouds and premixed gas/vapor clouds are substantially different. secondly, contrary to premixed gas flame propagation, the propagation of flames in dust/air mixtures is not limited to the flammable dust concentration range of dynamic clouds. thus here we modeled counterflow combustion of dust clouds and the effect of lewis number on gaseous fuel mass fraction distribution is investigated. a model is developed for describing a one dimensional, axisymmetric, premixed flame in a counterflow configuration. uniformly distributed volatile fuel particles in air are considered as the entering materials. the initial number density of the particles, (number of particles per unit volume) and the initial radius are presumed to be known. vproduct [p] is assumed. the flame position is determined andthe effect of lewis number change on the gaseous fuel mass fraction distribution is investigated. keywords: lapping, grinding, surface roughness, micro, nano scale. 1.0 introduction many studies of dust clouds combustion have been published over the last few years; see, for example, introduction of the article on flame propagation through micro-organic dust particles (bidabadi et.al., 2010). but from the fact that in many practical applications the flow field is appreciably strained, to yield realistic flame prediction under such conditions, counterflow configuration is suitable for studying these cases. over the last few decades, the counterflow configuration has been extensively adopted in theoretical, experimental and numerical studies as a means to investigate various physical effects on real flames on real flames, such as stretch, preferential diffusion, radiation and chemical kinetics (daou, 2011), (thatcher et.al.) and (wang et.al., 2007). but these studies are done for gaseous fuels. eckhoff clarified the differences and similarities between dust and gases (eckhof, 2006). it has been concluded that there are two basic differences between issn: 2180-1053 vol. 4 no. 1 january-june 2012 journal of mechanical engineering and technology 98 dusts and gases which are of substantially greater significance in design of safety standards than these similarities. firstly, the physics of generation and up-keeping of dust clouds and premixed gas/vapor clouds are substantially different. secondly, contrary to premixed gas flame propagation, the propagation of flames in dust/air mixtures is not limited to the flammable dust concentration range of dynamic clouds. thus here we modeled counterflow combustion of dust clouds and the effect of lewis number on gaseous fuel mass fraction distribution is investigated. a model is developed for describing a one dimensional, axisymmetric, premixed flame in a counterflow configuration. uniformly distributed volatile fuel particles in air are considered as the entering materials. the initial number density of the particles, nu (number of particles per unit volume) and the initial radius ru are presumed to be known. in the analysis it is presumed that the fuel particles vaporize to form a known gaseous compound which is then oxidized. in other words particles do not participate in the reaction. the kinetics of vaporization are presumed to be of the form: 83 in the analysis it is presumed that the fuel particles vaporize to form a known gaseous compound which is then oxidized. in other words particles do not participate in the reaction. the kinetics of vaporization are presumed to be of the form: where the units of are mass of gaseous fuel vaporized per unit volume per second. the quantity is constant which is presumed to be known, and denotes the gas temperature.for simplicity it is assumed that gas and particle have same temperature. the combustion process is represented by one –step irreversible reaction of the form where the symbols , and denote the fuel, oxygen and product, respectively, and the quantities , , and denote their respective stoichiometric coefficients. the zeldovich number is presumed to be large and is defined as in this paper subscript denotes conditions in the flame and the subscript denotes conditions at the inlet. the quantities and r represent respectively the activation energy and the universal gas number. figure 1 the counterflow configuration and twin planar premixed flames in figure 1 we have illustrated the considered configuration including the planar twin flames. reactants enter from direction and exhaust gases exit in the direction. the flame structure consists of a broad preheat-vaporization zone, a thin reaction zone and a post flame zone. in the formulation,subscripts , and are respectively used to show these zones. to determine the structure of these zones, a number of approximations are introduced in the conservation equations governing their structure.in the preheat-vaporization zone the rate of reaction between the fuel andoxidizer is presumed to be small and the structure of this layer is determined from a balance between the convective, diffusive, and vaporization terms in the conservation equation. in the thin reaction zone the convective and vaporization terms are presumed to be small in comparison with (1) where the units of wv are mass of gaseous fuel vaporized per unit volume per second. the quantity ais constant which is presumed to be known, and t denotes the gas temperature. for simplicity it is assumed that gas and particle have same temperature. the combustion process is represented by one–step irreversible reaction of the form 83 in the analysis it is presumed that the fuel particles vaporize to form a known gaseous compound which is then oxidized. in other words particles do not participate in the reaction. the kinetics of vaporization are presumed to be of the form: where the units of are mass of gaseous fuel vaporized per unit volume per second. the quantity is constant which is presumed to be known, and denotes the gas temperature.for simplicity it is assumed that gas and particle have same temperature. the combustion process is represented by one –step irreversible reaction of the form where the symbols , and denote the fuel, oxygen and product, respectively, and the quantities , , and denote their respective stoichiometric coefficients. the zeldovich number is presumed to be large and is defined as in this paper subscript denotes conditions in the flame and the subscript denotes conditions at the inlet. the quantities and r represent respectively the activation energy and the universal gas number. figure 1 the counterflow configuration and twin planar premixed flames in figure 1 we have illustrated the considered configuration including the planar twin flames. reactants enter from direction and exhaust gases exit in the direction. the flame structure consists of a broad preheat-vaporization zone, a thin reaction zone and a post flame zone. in the formulation,subscripts , and are respectively used to show these zones. to determine the structure of these zones, a number of approximations are introduced in the conservation equations governing their structure.in the preheat-vaporization zone the rate of reaction between the fuel andoxidizer is presumed to be small and the structure of this layer is determined from a balance between the convective, diffusive, and vaporization terms in the conservation equation. in the thin reaction zone the convective and vaporization terms are presumed to be small in comparison with (2) where the symbols f, o2 and p denote the fuel, oxygen and product, respectively, and the quantities vf, vo2, and vproduct denote their respective stoichiometric coefficients. the zeldovich number is presumed to be large and is defined as 83 in the analysis it is presumed that the fuel particles vaporize to form a known gaseous compound which is then oxidized. in other words particles do not participate in the reaction. the kinetics of vaporization are presumed to be of the form: where the units of are mass of gaseous fuel vaporized per unit volume per second. the quantity is constant which is presumed to be known, and denotes the gas temperature.for simplicity it is assumed that gas and particle have same temperature. the combustion process is represented by one –step irreversible reaction of the form where the symbols , and denote the fuel, oxygen and product, respectively, and the quantities , , and denote their respective stoichiometric coefficients. the zeldovich number is presumed to be large and is defined as in this paper subscript denotes conditions in the flame and the subscript denotes conditions at the inlet. the quantities and r represent respectively the activation energy and the universal gas number. figure 1 the counterflow configuration and twin planar premixed flames in figure 1 we have illustrated the considered configuration including the planar twin flames. reactants enter from direction and exhaust gases exit in the direction. the flame structure consists of a broad preheat-vaporization zone, a thin reaction zone and a post flame zone. in the formulation,subscripts , and are respectively used to show these zones. to determine the structure of these zones, a number of approximations are introduced in the conservation equations governing their structure.in the preheat-vaporization zone the rate of reaction between the fuel andoxidizer is presumed to be small and the structure of this layer is determined from a balance between the convective, diffusive, and vaporization terms in the conservation equation. in the thin reaction zone the convective and vaporization terms are presumed to be small in comparison with (3) in this paper subscript f denotes conditions in the flame and the subscript issn: 2180-1053 vol. 4 no. 1 january-june 2012 counterflow combustion of micro organic particles 99 u denotes conditions at the inlet. the quantities e and r represent respectively the activation energy and the universal gas number. 83 in the analysis it is presumed that the fuel particles vaporize to form a known gaseous compound which is then oxidized. in other words particles do not participate in the reaction. the kinetics of vaporization are presumed to be of the form: where the units of are mass of gaseous fuel vaporized per unit volume per second. the quantity is constant which is presumed to be known, and denotes the gas temperature.for simplicity it is assumed that gas and particle have same temperature. the combustion process is represented by one –step irreversible reaction of the form where the symbols , and denote the fuel, oxygen and product, respectively, and the quantities , , and denote their respective stoichiometric coefficients. the zeldovich number is presumed to be large and is defined as in this paper subscript denotes conditions in the flame and the subscript denotes conditions at the inlet. the quantities and r represent respectively the activation energy and the universal gas number. figure 1 the counterflow configuration and twin planar premixed flames in figure 1 we have illustrated the considered configuration including the planar twin flames. reactants enter from direction and exhaust gases exit in the direction. the flame structure consists of a broad preheat-vaporization zone, a thin reaction zone and a post flame zone. in the formulation,subscripts , and are respectively used to show these zones. to determine the structure of these zones, a number of approximations are introduced in the conservation equations governing their structure.in the preheat-vaporization zone the rate of reaction between the fuel andoxidizer is presumed to be small and the structure of this layer is determined from a balance between the convective, diffusive, and vaporization terms in the conservation equation. in the thin reaction zone the convective and vaporization terms are presumed to be small in comparison with figure 1 the counterflow configuration and twin planar premixed flames in figure 1 we have illustrated the considered configuration including the planar twin flames. reactants enter from ±x direction and exhaust gases exit in the ±y direction. the flame structure consists of a broad preheat-vaporization zone, a thin reaction zone and a post flame zone. in the formulation,subscripts1, 2 and 3are respectively used to show these zones. to determine the structure of these zones, a number of approximations are introduced in the conservation equations governing their structure.in the preheat-vaporization zone the rate of reaction between the fuel andoxidizer is presumed to be small and the structure of this layer is determined from a balance between the convective, diffusive, and vaporization terms in the conservation equation. in the thin reaction zone the convective and vaporization terms are presumed to be small in comparison with the diffusive and reactive terms.it is assumed that all of the particles vaporizes just before the flame, thus in the post flame zone the vaporization term is not considered. as it has been mentioned in (seshadri et.al,. 1992) for large values of nu, ϕu>0.7, where ϕu is equivalence ratio based on fuel available in the particles in the ambient reactant stream, the standoff distance of the envelope flame surrounding each particle is much larger than the characteristic separation distance between the particles. thus, the analysis developed here is only valid for ϕu>0.7. the velocity field has components (-ax,ay,0)in the cartesian directions, where a is the (dimensional) strain rate. for small values of strain rate we can consider the problem as one dimensional. all external forces are issn: 2180-1053 vol. 4 no. 1 january-june 2012 journal of mechanical engineering and technology 100 assumed to be negligible. also diffusion caused by pressure gradient is neglected. it is also assumed that the ratio 84 the diffusive and reactive terms.it is assumed that all of the particles vaporizes just before the flame, thus in the post flame zone the vaporization term is not considered. as it has been mentioned in (seshadri et al,. 1992) for large values of , , where is equivalence ratio based on fuel available in the particles in the ambient reactant stream, the standoff distance of the envelope flame surrounding each particle is much larger than the characteristic separation distance between the particles. thus, the analysis developed here is only valid for . the velocity field has components in the cartesian directions, where is the (dimensional) strain rate. for small values of strain rate we can consider the problem as one dimensional. all external forces are assumed to be negligible. also diffusion caused by pressure gradient is neglected. it is also assumed that the ratio is constant, where is the density of the mixture of gas and the fuel particles. the one dimensional governing equations are in eqs. 4-7, denotes density, and represent diffusion coefficients for heat and fuel respectively, is the reaction rate and it's unit is mass of gaseous fuel consumed per unit volume per second, stands for the heat release per unit mass of the fuel burned, is the heat associated with vaporizing unit mass of the fuel, is the mass fraction of the fuel, is the combined heat capacity of the gas and of the particles and denotes mass fraction of the particles. further approximation introduced in eqs. 4-7 are that the mean molecular weight do not vary and that the thermal conductivity of the mixture, , is constant and the diffusion coefficient, , is proportional to . also particles diffusion is neglected and the density of a fuel particle, , is presumed to be constant. given the symmetry of the configuration about the plane , we only solve the problem for with the boundary conditions is constant, where ρis the density of the mixture of gas and the fuel particles. the one dimensional governing equations are 84 the diffusive and reactive terms.it is assumed that all of the particles vaporizes just before the flame, thus in the post flame zone the vaporization term is not considered. as it has been mentioned in (seshadri et al,. 1992) for large values of , , where is equivalence ratio based on fuel available in the particles in the ambient reactant stream, the standoff distance of the envelope flame surrounding each particle is much larger than the characteristic separation distance between the particles. thus, the analysis developed here is only valid for . the velocity field has components in the cartesian directions, where is the (dimensional) strain rate. for small values of strain rate we can consider the problem as one dimensional. all external forces are assumed to be negligible. also diffusion caused by pressure gradient is neglected. it is also assumed that the ratio is constant, where is the density of the mixture of gas and the fuel particles. the one dimensional governing equations are in eqs. 4-7, denotes density, and represent diffusion coefficients for heat and fuel respectively, is the reaction rate and it's unit is mass of gaseous fuel consumed per unit volume per second, stands for the heat release per unit mass of the fuel burned, is the heat associated with vaporizing unit mass of the fuel, is the mass fraction of the fuel, is the combined heat capacity of the gas and of the particles and denotes mass fraction of the particles. further approximation introduced in eqs. 4-7 are that the mean molecular weight do not vary and that the thermal conductivity of the mixture, , is constant and the diffusion coefficient, , is proportional to . also particles diffusion is neglected and the density of a fuel particle, , is presumed to be constant. given the symmetry of the configuration about the plane , we only solve the problem for with the boundary conditions (4) 84 the diffusive and reactive terms.it is assumed that all of the particles vaporizes just before the flame, thus in the post flame zone the vaporization term is not considered. as it has been mentioned in (seshadri et al,. 1992) for large values of , , where is equivalence ratio based on fuel available in the particles in the ambient reactant stream, the standoff distance of the envelope flame surrounding each particle is much larger than the characteristic separation distance between the particles. thus, the analysis developed here is only valid for . the velocity field has components in the cartesian directions, where is the (dimensional) strain rate. for small values of strain rate we can consider the problem as one dimensional. all external forces are assumed to be negligible. also diffusion caused by pressure gradient is neglected. it is also assumed that the ratio is constant, where is the density of the mixture of gas and the fuel particles. the one dimensional governing equations are in eqs. 4-7, denotes density, and represent diffusion coefficients for heat and fuel respectively, is the reaction rate and it's unit is mass of gaseous fuel consumed per unit volume per second, stands for the heat release per unit mass of the fuel burned, is the heat associated with vaporizing unit mass of the fuel, is the mass fraction of the fuel, is the combined heat capacity of the gas and of the particles and denotes mass fraction of the particles. further approximation introduced in eqs. 4-7 are that the mean molecular weight do not vary and that the thermal conductivity of the mixture, , is constant and the diffusion coefficient, , is proportional to . also particles diffusion is neglected and the density of a fuel particle, , is presumed to be constant. given the symmetry of the configuration about the plane , we only solve the problem for with the boundary conditions (5) 84 the diffusive and reactive terms.it is assumed that all of the particles vaporizes just before the flame, thus in the post flame zone the vaporization term is not considered. as it has been mentioned in (seshadri et al,. 1992) for large values of , , where is equivalence ratio based on fuel available in the particles in the ambient reactant stream, the standoff distance of the envelope flame surrounding each particle is much larger than the characteristic separation distance between the particles. thus, the analysis developed here is only valid for . the velocity field has components in the cartesian directions, where is the (dimensional) strain rate. for small values of strain rate we can consider the problem as one dimensional. all external forces are assumed to be negligible. also diffusion caused by pressure gradient is neglected. it is also assumed that the ratio is constant, where is the density of the mixture of gas and the fuel particles. the one dimensional governing equations are in eqs. 4-7, denotes density, and represent diffusion coefficients for heat and fuel respectively, is the reaction rate and it's unit is mass of gaseous fuel consumed per unit volume per second, stands for the heat release per unit mass of the fuel burned, is the heat associated with vaporizing unit mass of the fuel, is the mass fraction of the fuel, is the combined heat capacity of the gas and of the particles and denotes mass fraction of the particles. further approximation introduced in eqs. 4-7 are that the mean molecular weight do not vary and that the thermal conductivity of the mixture, , is constant and the diffusion coefficient, , is proportional to . also particles diffusion is neglected and the density of a fuel particle, , is presumed to be constant. given the symmetry of the configuration about the plane , we only solve the problem for with the boundary conditions (6) 84 the diffusive and reactive terms.it is assumed that all of the particles vaporizes just before the flame, thus in the post flame zone the vaporization term is not considered. as it has been mentioned in (seshadri et al,. 1992) for large values of , , where is equivalence ratio based on fuel available in the particles in the ambient reactant stream, the standoff distance of the envelope flame surrounding each particle is much larger than the characteristic separation distance between the particles. thus, the analysis developed here is only valid for . the velocity field has components in the cartesian directions, where is the (dimensional) strain rate. for small values of strain rate we can consider the problem as one dimensional. all external forces are assumed to be negligible. also diffusion caused by pressure gradient is neglected. it is also assumed that the ratio is constant, where is the density of the mixture of gas and the fuel particles. the one dimensional governing equations are in eqs. 4-7, denotes density, and represent diffusion coefficients for heat and fuel respectively, is the reaction rate and it's unit is mass of gaseous fuel consumed per unit volume per second, stands for the heat release per unit mass of the fuel burned, is the heat associated with vaporizing unit mass of the fuel, is the mass fraction of the fuel, is the combined heat capacity of the gas and of the particles and denotes mass fraction of the particles. further approximation introduced in eqs. 4-7 are that the mean molecular weight do not vary and that the thermal conductivity of the mixture, , is constant and the diffusion coefficient, , is proportional to . also particles diffusion is neglected and the density of a fuel particle, , is presumed to be constant. given the symmetry of the configuration about the plane , we only solve the problem for with the boundary conditions (7) in eqs. 4-7, ρ denotes density, dt and df represent diffusion coefficients for heat and fuel respectively, ωf is the reaction rate and it's unit is mass of gaseous fuel consumed per unit volume per second, q stands for the heat release per unit mass of the fuel burned, qv is the heat associated with vaporizing unit mass of the fuel,yf is the mass fraction of the fuel, c is the combined heat capacity of the gas and of the particles and ys denotes mass fraction of the particles. further approximation introduced in eqs. 4-7 are that the mean molecular weight do not vary and that the thermal conductivity of the mixture, λ, is constant and the diffusion coefficient, d, is proportional to t. also particles diffusion is neglected and the density of a fuel particle, ρs, is presumed to be constant. given the symmetry of the configuration about the plane x=0, we only solve the problem for x>0 with the boundary conditions 84 the diffusive and reactive terms.it is assumed that all of the particles vaporizes just before the flame, thus in the post flame zone the vaporization term is not considered. as it has been mentioned in (seshadri et al,. 1992) for large values of , , where is equivalence ratio based on fuel available in the particles in the ambient reactant stream, the standoff distance of the envelope flame surrounding each particle is much larger than the characteristic separation distance between the particles. thus, the analysis developed here is only valid for . the velocity field has components in the cartesian directions, where is the (dimensional) strain rate. for small values of strain rate we can consider the problem as one dimensional. all external forces are assumed to be negligible. also diffusion caused by pressure gradient is neglected. it is also assumed that the ratio is constant, where is the density of the mixture of gas and the fuel particles. the one dimensional governing equations are in eqs. 4-7, denotes density, and represent diffusion coefficients for heat and fuel respectively, is the reaction rate and it's unit is mass of gaseous fuel consumed per unit volume per second, stands for the heat release per unit mass of the fuel burned, is the heat associated with vaporizing unit mass of the fuel, is the mass fraction of the fuel, is the combined heat capacity of the gas and of the particles and denotes mass fraction of the particles. further approximation introduced in eqs. 4-7 are that the mean molecular weight do not vary and that the thermal conductivity of the mixture, , is constant and the diffusion coefficient, , is proportional to . also particles diffusion is neglected and the density of a fuel particle, , is presumed to be constant. given the symmetry of the configuration about the plane , we only solve the problem for with the boundary conditions where yfu denotes the mass fraction of fuel available in the particles. nondimensionalization of governing equations we define the following rescaled variables issn: 2180-1053 vol. 4 no. 1 january-june 2012 counterflow combustion of micro organic particles 101 85 where denotes the mass fraction of fuel available in the particles. nondimensionalization of governing equations we define the following rescaled variables where is the maximum temperature attained in the reaction zone, calculated neglecting the heat of vaporization of the particles. the quantity is chosen such that in eq. 8, is mixing layer thickness and is defined by . with introducing eqs. 8 and 9 into governing equations, we obtain their non-dimensional form where the radius has been rewritten in terms of using the relation further assumption used in eqs. 10-12 is that the rate of vaporization is expected to be dominant near the reaction zone, for , the vaporization rate shown in eq. was presumed to be proportional to . the quantities and used above are given by (8) where tf is the maximum temperature attained in the reaction zone, calculated neglecting the heat of vaporization of the particles. the quantity yfc is chosen such that 85 where denotes the mass fraction of fuel available in the particles. nondimensionalization of governing equations we define the following rescaled variables where is the maximum temperature attained in the reaction zone, calculated neglecting the heat of vaporization of the particles. the quantity is chosen such that in eq. 8, is mixing layer thickness and is defined by . with introducing eqs. 8 and 9 into governing equations, we obtain their non-dimensional form where the radius has been rewritten in terms of using the relation further assumption used in eqs. 10-12 is that the rate of vaporization is expected to be dominant near the reaction zone, for , the vaporization rate shown in eq. was presumed to be proportional to . the quantities and used above are given by (9) in eq. 8, l is mixing layer thickness and is defined by 85 where denotes the mass fraction of fuel available in the particles. nondimensionalization of governing equations we define the following rescaled variables where is the maximum temperature attained in the reaction zone, calculated neglecting the heat of vaporization of the particles. the quantity is chosen such that in eq. 8, is mixing layer thickness and is defined by . with introducing eqs. 8 and 9 into governing equations, we obtain their non-dimensional form where the radius has been rewritten in terms of using the relation further assumption used in eqs. 10-12 is that the rate of vaporization is expected to be dominant near the reaction zone, for , the vaporization rate shown in eq. was presumed to be proportional to . the quantities and used above are given by with introducing eqs. 8 and 9 into governing equations, we obtain their non-dimensional form 85 where denotes the mass fraction of fuel available in the particles. nondimensionalization of governing equations we define the following rescaled variables where is the maximum temperature attained in the reaction zone, calculated neglecting the heat of vaporization of the particles. the quantity is chosen such that in eq. 8, is mixing layer thickness and is defined by . with introducing eqs. 8 and 9 into governing equations, we obtain their non-dimensional form where the radius has been rewritten in terms of using the relation further assumption used in eqs. 10-12 is that the rate of vaporization is expected to be dominant near the reaction zone, for , the vaporization rate shown in eq. was presumed to be proportional to . the quantities and used above are given by (10) 85 where denotes the mass fraction of fuel available in the particles. nondimensionalization of governing equations we define the following rescaled variables where is the maximum temperature attained in the reaction zone, calculated neglecting the heat of vaporization of the particles. the quantity is chosen such that in eq. 8, is mixing layer thickness and is defined by . with introducing eqs. 8 and 9 into governing equations, we obtain their non-dimensional form where the radius has been rewritten in terms of using the relation further assumption used in eqs. 10-12 is that the rate of vaporization is expected to be dominant near the reaction zone, for , the vaporization rate shown in eq. was presumed to be proportional to . the quantities and used above are given by (11) 85 where denotes the mass fraction of fuel available in the particles. nondimensionalization of governing equations we define the following rescaled variables where is the maximum temperature attained in the reaction zone, calculated neglecting the heat of vaporization of the particles. the quantity is chosen such that in eq. 8, is mixing layer thickness and is defined by . with introducing eqs. 8 and 9 into governing equations, we obtain their non-dimensional form where the radius has been rewritten in terms of using the relation further assumption used in eqs. 10-12 is that the rate of vaporization is expected to be dominant near the reaction zone, for , the vaporization rate shown in eq. was presumed to be proportional to . the quantities and used above are given by (12) where the radius r has been rewritten in terms of ys using the relation 85 where denotes the mass fraction of fuel available in the particles. nondimensionalization of governing equations we define the following rescaled variables where is the maximum temperature attained in the reaction zone, calculated neglecting the heat of vaporization of the particles. the quantity is chosen such that in eq. 8, is mixing layer thickness and is defined by . with introducing eqs. 8 and 9 into governing equations, we obtain their non-dimensional form where the radius has been rewritten in terms of using the relation further assumption used in eqs. 10-12 is that the rate of vaporization is expected to be dominant near the reaction zone, for , the vaporization rate shown in eq. was presumed to be proportional to . the quantities and used above are given by (13) further assumption used in eqs. 10-12 is that the rate of vaporization is expected to be dominant near the reaction zone, for 85 where denotes the mass fraction of fuel available in the particles. nondimensionalization of governing equations we define the following rescaled variables where is the maximum temperature attained in the reaction zone, calculated neglecting the heat of vaporization of the particles. the quantity is chosen such that in eq. 8, is mixing layer thickness and is defined by . with introducing eqs. 8 and 9 into governing equations, we obtain their non-dimensional form where the radius has been rewritten in terms of using the relation further assumption used in eqs. 10-12 is that the rate of vaporization is expected to be dominant near the reaction zone, for , the vaporization rate shown in eq. was presumed to be proportional to . the quantities and used above are given by , the vaporization rate shown in eq. 1 was presumed to be proportional to 85 where denotes the mass fraction of fuel available in the particles. nondimensionalization of governing equations we define the following rescaled variables where is the maximum temperature attained in the reaction zone, calculated neglecting the heat of vaporization of the particles. the quantity is chosen such that in eq. 8, is mixing layer thickness and is defined by . with introducing eqs. 8 and 9 into governing equations, we obtain their non-dimensional form where the radius has been rewritten in terms of using the relation further assumption used in eqs. 10-12 is that the rate of vaporization is expected to be dominant near the reaction zone, for , the vaporization rate shown in eq. was presumed to be proportional to . the quantities and used above are given by . the quantities γ and ω used above are given by 85 where denotes the mass fraction of fuel available in the particles. nondimensionalization of governing equations we define the following rescaled variables where is the maximum temperature attained in the reaction zone, calculated neglecting the heat of vaporization of the particles. the quantity is chosen such that in eq. 8, is mixing layer thickness and is defined by . with introducing eqs. 8 and 9 into governing equations, we obtain their non-dimensional form where the radius has been rewritten in terms of using the relation further assumption used in eqs. 10-12 is that the rate of vaporization is expected to be dominant near the reaction zone, for , the vaporization rate shown in eq. was presumed to be proportional to . the quantities and used above are given by (14) 85 where denotes the mass fraction of fuel available in the particles. nondimensionalization of governing equations we define the following rescaled variables where is the maximum temperature attained in the reaction zone, calculated neglecting the heat of vaporization of the particles. the quantity is chosen such that in eq. 8, is mixing layer thickness and is defined by . with introducing eqs. 8 and 9 into governing equations, we obtain their non-dimensional form where the radius has been rewritten in terms of using the relation further assumption used in eqs. 10-12 is that the rate of vaporization is expected to be dominant near the reaction zone, for , the vaporization rate shown in eq. was presumed to be proportional to . the quantities and used above are given by (15) issn: 2180-1053 vol. 4 no. 1 january-june 2012 journal of mechanical engineering and technology 102 85 where denotes the mass fraction of fuel available in the particles. nondimensionalization of governing equations we define the following rescaled variables where is the maximum temperature attained in the reaction zone, calculated neglecting the heat of vaporization of the particles. the quantity is chosen such that in eq. 8, is mixing layer thickness and is defined by . with introducing eqs. 8 and 9 into governing equations, we obtain their non-dimensional form where the radius has been rewritten in terms of using the relation further assumption used in eqs. 10-12 is that the rate of vaporization is expected to be dominant near the reaction zone, for , the vaporization rate shown in eq. was presumed to be proportional to . the quantities and used above are given by (16) where the quantity q is the ratio of the heat required to vaporize the fuel particles to the overall heat release in the flame and is presumed to be small in this analysis. if θ0 represents the nondimensionalized temperature for q=0, then non-dimensional equations transform into 86 where the quantity is the ratio of the heat required to vaporize the fuel particles to the overall heat release in the flame and is presumed to be small in this analysis. if represents the nondimensionalized temperature for , then non-dimensional equations transform into with the boundary conditions where . 2.0 asymptotic analysis 2.1. preheat-vaporization zone in the asymptotic limit , we can neglect chemical reaction between the gaseous fuel and oxidizer in the preheat-vaporization zone. thus the energy equation in this zone reduces to since, by definition is the flame temperature in the reaction zone, at . where is the flame position. using boundary condition we have introducing eq. 21 into the eq. 19 using the boundary condition yields introducing eqs. 21 and 22 into eq. 18 and integratingit once, results in (17) 86 where the quantity is the ratio of the heat required to vaporize the fuel particles to the overall heat release in the flame and is presumed to be small in this analysis. if represents the nondimensionalized temperature for , then non-dimensional equations transform into with the boundary conditions where . 2.0 asymptotic analysis 2.1. preheat-vaporization zone in the asymptotic limit , we can neglect chemical reaction between the gaseous fuel and oxidizer in the preheat-vaporization zone. thus the energy equation in this zone reduces to since, by definition is the flame temperature in the reaction zone, at . where is the flame position. using boundary condition we have introducing eq. 21 into the eq. 19 using the boundary condition yields introducing eqs. 21 and 22 into eq. 18 and integratingit once, results in (18) 86 where the quantity is the ratio of the heat required to vaporize the fuel particles to the overall heat release in the flame and is presumed to be small in this analysis. if represents the nondimensionalized temperature for , then non-dimensional equations transform into with the boundary conditions where . 2.0 asymptotic analysis 2.1. preheat-vaporization zone in the asymptotic limit , we can neglect chemical reaction between the gaseous fuel and oxidizer in the preheat-vaporization zone. thus the energy equation in this zone reduces to since, by definition is the flame temperature in the reaction zone, at . where is the flame position. using boundary condition we have introducing eq. 21 into the eq. 19 using the boundary condition yields introducing eqs. 21 and 22 into eq. 18 and integratingit once, results in with the boundary conditions 86 where the quantity is the ratio of the heat required to vaporize the fuel particles to the overall heat release in the flame and is presumed to be small in this analysis. if represents the nondimensionalized temperature for , then non-dimensional equations transform into with the boundary conditions where . 2.0 asymptotic analysis 2.1. preheat-vaporization zone in the asymptotic limit , we can neglect chemical reaction between the gaseous fuel and oxidizer in the preheat-vaporization zone. thus the energy equation in this zone reduces to since, by definition is the flame temperature in the reaction zone, at . where is the flame position. using boundary condition we have introducing eq. 21 into the eq. 19 using the boundary condition yields introducing eqs. 21 and 22 into eq. 18 and integratingit once, results in where 86 where the quantity is the ratio of the heat required to vaporize the fuel particles to the overall heat release in the flame and is presumed to be small in this analysis. if represents the nondimensionalized temperature for , then non-dimensional equations transform into with the boundary conditions where . 2.0 asymptotic analysis 2.1. preheat-vaporization zone in the asymptotic limit , we can neglect chemical reaction between the gaseous fuel and oxidizer in the preheat-vaporization zone. thus the energy equation in this zone reduces to since, by definition is the flame temperature in the reaction zone, at . where is the flame position. using boundary condition we have introducing eq. 21 into the eq. 19 using the boundary condition yields introducing eqs. 21 and 22 into eq. 18 and integratingit once, results in (19) 2.0 asymptotic analysis 2.1 preheat-vaporization zone in the asymptotic limit 86 where the quantity is the ratio of the heat required to vaporize the fuel particles to the overall heat release in the flame and is presumed to be small in this analysis. if represents the nondimensionalized temperature for , then non-dimensional equations transform into with the boundary conditions where . 2.0 asymptotic analysis 2.1. preheat-vaporization zone in the asymptotic limit , we can neglect chemical reaction between the gaseous fuel and oxidizer in the preheat-vaporization zone. thus the energy equation in this zone reduces to since, by definition is the flame temperature in the reaction zone, at . where is the flame position. using boundary condition we have introducing eq. 21 into the eq. 19 using the boundary condition yields introducing eqs. 21 and 22 into eq. 18 and integratingit once, results in , we can neglect chemical reaction between the gaseous fuel and oxidizer in the preheat-vaporization zone. thus the energy equation in this zone reduces to 86 where the quantity is the ratio of the heat required to vaporize the fuel particles to the overall heat release in the flame and is presumed to be small in this analysis. if represents the nondimensionalized temperature for , then non-dimensional equations transform into with the boundary conditions where . 2.0 asymptotic analysis 2.1. preheat-vaporization zone in the asymptotic limit , we can neglect chemical reaction between the gaseous fuel and oxidizer in the preheat-vaporization zone. thus the energy equation in this zone reduces to since, by definition is the flame temperature in the reaction zone, at . where is the flame position. using boundary condition we have introducing eq. 21 into the eq. 19 using the boundary condition yields introducing eqs. 21 and 22 into eq. 18 and integratingit once, results in (20) issn: 2180-1053 vol. 4 no. 1 january-june 2012 counterflow combustion of micro organic particles 103 since, by definition tf is the flame temperature in the reaction zone, 86 where the quantity is the ratio of the heat required to vaporize the fuel particles to the overall heat release in the flame and is presumed to be small in this analysis. if represents the nondimensionalized temperature for , then non-dimensional equations transform into with the boundary conditions where . 2.0 asymptotic analysis 2.1. preheat-vaporization zone in the asymptotic limit , we can neglect chemical reaction between the gaseous fuel and oxidizer in the preheat-vaporization zone. thus the energy equation in this zone reduces to since, by definition is the flame temperature in the reaction zone, at . where is the flame position. using boundary condition we have introducing eq. 21 into the eq. 19 using the boundary condition yields introducing eqs. 21 and 22 into eq. 18 and integratingit once, results in is the flame position. using boundary condition we have 86 where the quantity is the ratio of the heat required to vaporize the fuel particles to the overall heat release in the flame and is presumed to be small in this analysis. if represents the nondimensionalized temperature for , then non-dimensional equations transform into with the boundary conditions where . 2.0 asymptotic analysis 2.1. preheat-vaporization zone in the asymptotic limit , we can neglect chemical reaction between the gaseous fuel and oxidizer in the preheat-vaporization zone. thus the energy equation in this zone reduces to since, by definition is the flame temperature in the reaction zone, at . where is the flame position. using boundary condition we have introducing eq. 21 into the eq. 19 using the boundary condition yields introducing eqs. 21 and 22 into eq. 18 and integratingit once, results in (21) introducing eq. 21 into the eq. 19 using the boundary condition yields 86 where the quantity is the ratio of the heat required to vaporize the fuel particles to the overall heat release in the flame and is presumed to be small in this analysis. if represents the nondimensionalized temperature for , then non-dimensional equations transform into with the boundary conditions where . 2.0 asymptotic analysis 2.1. preheat-vaporization zone in the asymptotic limit , we can neglect chemical reaction between the gaseous fuel and oxidizer in the preheat-vaporization zone. thus the energy equation in this zone reduces to since, by definition is the flame temperature in the reaction zone, at . where is the flame position. using boundary condition we have introducing eq. 21 into the eq. 19 using the boundary condition yields introducing eqs. 21 and 22 into eq. 18 and integratingit once, results in (22) introducing eqs. 21 and 22 into eq. 18 and integratingit once, results in 87 where boundary conditions are used in writing eq. 23. also as the value of is sufficiently high, it is assumed that , where is gaseous fuel mass fraction in the reaction zone. in the post flame zoneas we have assumed that all of the particles vaporizes just before the flame, the reaction and vaporization terms are not considered. thus where the subscript denotes conditions at the interface between post flame zone and the reaction zone. the jump conditions across the reaction zone which are used in finding flame position are where the subscript denotes conditions at the interface between preheat-vaporization zone and the reaction zone. 3.0 results and discussion for purpose of illustration it will be assumed that the gaseous fuel that evolves from the fuel particles is methane. the values of , , , and are considered to be , , , and respectively, based on (seshadri et al,. 1992). also , and are chosen from (seshadri et al,. 1992)based on and assumption. where is equivalence ratio based on fuel available in the ambient reactant stream and is radius of fuel particle in the ambient reactant stream. the value of is calculated with trial and error method. a value is guessed. then equation 22 is solved and the answer is substituted in 18 neglecting reaction term. this equation is solved numerically to find the first term of equation 23.at last eq. 24 is used to find the next guess.this is done until convergence. the value that is obtained is , which results in (23) where boundary conditions are used in writing eq. 23. also as the value of 87 where boundary conditions are used in writing eq. 23. also as the value of is sufficiently high, it is assumed that , where is gaseous fuel mass fraction in the reaction zone. in the post flame zoneas we have assumed that all of the particles vaporizes just before the flame, the reaction and vaporization terms are not considered. thus where the subscript denotes conditions at the interface between post flame zone and the reaction zone. the jump conditions across the reaction zone which are used in finding flame position are where the subscript denotes conditions at the interface between preheat-vaporization zone and the reaction zone. 3.0 results and discussion for purpose of illustration it will be assumed that the gaseous fuel that evolves from the fuel particles is methane. the values of , , , and are considered to be , , , and respectively, based on (seshadri et al,. 1992). also , and are chosen from (seshadri et al,. 1992)based on and assumption. where is equivalence ratio based on fuel available in the ambient reactant stream and is radius of fuel particle in the ambient reactant stream. the value of is calculated with trial and error method. a value is guessed. then equation 22 is solved and the answer is substituted in 18 neglecting reaction term. this equation is solved numerically to find the first term of equation 23.at last eq. 24 is used to find the next guess.this is done until convergence. the value that is obtained is , which results in is sufficiently high, it is assumed that 87 where boundary conditions are used in writing eq. 23. also as the value of is sufficiently high, it is assumed that , where is gaseous fuel mass fraction in the reaction zone. in the post flame zoneas we have assumed that all of the particles vaporizes just before the flame, the reaction and vaporization terms are not considered. thus where the subscript denotes conditions at the interface between post flame zone and the reaction zone. the jump conditions across the reaction zone which are used in finding flame position are where the subscript denotes conditions at the interface between preheat-vaporization zone and the reaction zone. 3.0 results and discussion for purpose of illustration it will be assumed that the gaseous fuel that evolves from the fuel particles is methane. the values of , , , and are considered to be , , , and respectively, based on (seshadri et al,. 1992). also , and are chosen from (seshadri et al,. 1992)based on and assumption. where is equivalence ratio based on fuel available in the ambient reactant stream and is radius of fuel particle in the ambient reactant stream. the value of is calculated with trial and error method. a value is guessed. then equation 22 is solved and the answer is substituted in 18 neglecting reaction term. this equation is solved numerically to find the first term of equation 23.at last eq. 24 is used to find the next guess.this is done until convergence. the value that is obtained is , which results in where 87 where boundary conditions are used in writing eq. 23. also as the value of is sufficiently high, it is assumed that , where is gaseous fuel mass fraction in the reaction zone. in the post flame zoneas we have assumed that all of the particles vaporizes just before the flame, the reaction and vaporization terms are not considered. thus where the subscript denotes conditions at the interface between post flame zone and the reaction zone. the jump conditions across the reaction zone which are used in finding flame position are where the subscript denotes conditions at the interface between preheat-vaporization zone and the reaction zone. 3.0 results and discussion for purpose of illustration it will be assumed that the gaseous fuel that evolves from the fuel particles is methane. the values of , , , and are considered to be , , , and respectively, based on (seshadri et al,. 1992). also , and are chosen from (seshadri et al,. 1992)based on and assumption. where is equivalence ratio based on fuel available in the ambient reactant stream and is radius of fuel particle in the ambient reactant stream. the value of is calculated with trial and error method. a value is guessed. then equation 22 is solved and the answer is substituted in 18 neglecting reaction term. this equation is solved numerically to find the first term of equation 23.at last eq. 24 is used to find the next guess.this is done until convergence. the value that is obtained is , which results in is gaseous fuel mass fraction in the reaction zone. in the post flame zoneas we have assumed that all of the particles vaporizes just before the flame, the reaction and vaporization terms are not considered. thus 87 where boundary conditions are used in writing eq. 23. also as the value of is sufficiently high, it is assumed that , where is gaseous fuel mass fraction in the reaction zone. in the post flame zoneas we have assumed that all of the particles vaporizes just before the flame, the reaction and vaporization terms are not considered. thus where the subscript denotes conditions at the interface between post flame zone and the reaction zone. the jump conditions across the reaction zone which are used in finding flame position are where the subscript denotes conditions at the interface between preheat-vaporization zone and the reaction zone. 3.0 results and discussion for purpose of illustration it will be assumed that the gaseous fuel that evolves from the fuel particles is methane. the values of , , , and are considered to be , , , and respectively, based on (seshadri et al,. 1992). also , and are chosen from (seshadri et al,. 1992)based on and assumption. where is equivalence ratio based on fuel available in the ambient reactant stream and is radius of fuel particle in the ambient reactant stream. the value of is calculated with trial and error method. a value is guessed. then equation 22 is solved and the answer is substituted in 18 neglecting reaction term. this equation is solved numerically to find the first term of equation 23.at last eq. 24 is used to find the next guess.this is done until convergence. the value that is obtained is , which results in where the subscript, denotes conditions at the interface between post flame zone and the reaction zone. the jump conditions across the reaction zone which are used in finding flame position are 87 where boundary conditions are used in writing eq. 23. also as the value of is sufficiently high, it is assumed that , where is gaseous fuel mass fraction in the reaction zone. in the post flame zoneas we have assumed that all of the particles vaporizes just before the flame, the reaction and vaporization terms are not considered. thus where the subscript denotes conditions at the interface between post flame zone and the reaction zone. the jump conditions across the reaction zone which are used in finding flame position are where the subscript denotes conditions at the interface between preheat-vaporization zone and the reaction zone. 3.0 results and discussion for purpose of illustration it will be assumed that the gaseous fuel that evolves from the fuel particles is methane. the values of , , , and are considered to be , , , and respectively, based on (seshadri et al,. 1992). also , and are chosen from (seshadri et al,. 1992)based on and assumption. where is equivalence ratio based on fuel available in the ambient reactant stream and is radius of fuel particle in the ambient reactant stream. the value of is calculated with trial and error method. a value is guessed. then equation 22 is solved and the answer is substituted in 18 neglecting reaction term. this equation is solved numerically to find the first term of equation 23.at last eq. 24 is used to find the next guess.this is done until convergence. the value that is obtained is , which results in (24) where the subscript + denotes conditions at the interface between preheat-vaporization zone and the reaction zone. issn: 2180-1053 vol. 4 no. 1 january-june 2012 journal of mechanical engineering and technology 104 3.0 results and discussion for purpose of illustration it will be assumed that the gaseous fuel that evolves from the fuel particles is methane. the values of a, nu, ρu, ρs and tu are considered to be 87 where boundary conditions are used in writing eq. 23. also as the value of is sufficiently high, it is assumed that , where is gaseous fuel mass fraction in the reaction zone. in the post flame zoneas we have assumed that all of the particles vaporizes just before the flame, the reaction and vaporization terms are not considered. thus where the subscript denotes conditions at the interface between post flame zone and the reaction zone. the jump conditions across the reaction zone which are used in finding flame position are where the subscript denotes conditions at the interface between preheat-vaporization zone and the reaction zone. 3.0 results and discussion for purpose of illustration it will be assumed that the gaseous fuel that evolves from the fuel particles is methane. the values of , , , and are considered to be , , , and respectively, based on (seshadri et al,. 1992). also , and are chosen from (seshadri et al,. 1992)based on and assumption. where is equivalence ratio based on fuel available in the ambient reactant stream and is radius of fuel particle in the ambient reactant stream. the value of is calculated with trial and error method. a value is guessed. then equation 22 is solved and the answer is substituted in 18 neglecting reaction term. this equation is solved numerically to find the first term of equation 23.at last eq. 24 is used to find the next guess.this is done until convergence. the value that is obtained is , which results in 87 where boundary conditions are used in writing eq. 23. also as the value of is sufficiently high, it is assumed that , where is gaseous fuel mass fraction in the reaction zone. in the post flame zoneas we have assumed that all of the particles vaporizes just before the flame, the reaction and vaporization terms are not considered. thus where the subscript denotes conditions at the interface between post flame zone and the reaction zone. the jump conditions across the reaction zone which are used in finding flame position are where the subscript denotes conditions at the interface between preheat-vaporization zone and the reaction zone. 3.0 results and discussion for purpose of illustration it will be assumed that the gaseous fuel that evolves from the fuel particles is methane. the values of , , , and are considered to be , , , and respectively, based on (seshadri et al,. 1992). also , and are chosen from (seshadri et al,. 1992)based on and assumption. where is equivalence ratio based on fuel available in the ambient reactant stream and is radius of fuel particle in the ambient reactant stream. the value of is calculated with trial and error method. a value is guessed. then equation 22 is solved and the answer is substituted in 18 neglecting reaction term. this equation is solved numerically to find the first term of equation 23.at last eq. 24 is used to find the next guess.this is done until convergence. the value that is obtained is , which results in and 300k respectively, based on (seshadri et.al,. 1992). also 87 where boundary conditions are used in writing eq. 23. also as the value of is sufficiently high, it is assumed that , where is gaseous fuel mass fraction in the reaction zone. in the post flame zoneas we have assumed that all of the particles vaporizes just before the flame, the reaction and vaporization terms are not considered. thus where the subscript denotes conditions at the interface between post flame zone and the reaction zone. the jump conditions across the reaction zone which are used in finding flame position are where the subscript denotes conditions at the interface between preheat-vaporization zone and the reaction zone. 3.0 results and discussion for purpose of illustration it will be assumed that the gaseous fuel that evolves from the fuel particles is methane. the values of , , , and are considered to be , , , and respectively, based on (seshadri et al,. 1992). also , and are chosen from (seshadri et al,. 1992)based on and assumption. where is equivalence ratio based on fuel available in the ambient reactant stream and is radius of fuel particle in the ambient reactant stream. the value of is calculated with trial and error method. a value is guessed. then equation 22 is solved and the answer is substituted in 18 neglecting reaction term. this equation is solved numerically to find the first term of equation 23.at last eq. 24 is used to find the next guess.this is done until convergence. the value that is obtained is , which results in are chosen from (seshadri et.al,. 1992) based on ϕu=1 and ru=20μm assumption. where ϕu is equivalence ratio based on fuel available in the ambient reactant stream and ru is radius of fuel particle in the ambient reactant stream. the value of 87 where boundary conditions are used in writing eq. 23. also as the value of is sufficiently high, it is assumed that , where is gaseous fuel mass fraction in the reaction zone. in the post flame zoneas we have assumed that all of the particles vaporizes just before the flame, the reaction and vaporization terms are not considered. thus where the subscript denotes conditions at the interface between post flame zone and the reaction zone. the jump conditions across the reaction zone which are used in finding flame position are where the subscript denotes conditions at the interface between preheat-vaporization zone and the reaction zone. 3.0 results and discussion for purpose of illustration it will be assumed that the gaseous fuel that evolves from the fuel particles is methane. the values of , , , and are considered to be , , , and respectively, based on (seshadri et al,. 1992). also , and are chosen from (seshadri et al,. 1992)based on and assumption. where is equivalence ratio based on fuel available in the ambient reactant stream and is radius of fuel particle in the ambient reactant stream. the value of is calculated with trial and error method. a value is guessed. then equation 22 is solved and the answer is substituted in 18 neglecting reaction term. this equation is solved numerically to find the first term of equation 23.at last eq. 24 is used to find the next guess.this is done until convergence. the value that is obtained is , which results in is calculated with trial and error method. a value is guessed. then equation 22 is solved and the answer is substituted in 18 neglecting reaction term. this equation is solved numerically to find the first term of equation 23.at last eq. 24 is used to find the next guess. this is done until convergence. the value that is obtained is 87 where boundary conditions are used in writing eq. 23. also as the value of is sufficiently high, it is assumed that , where is gaseous fuel mass fraction in the reaction zone. in the post flame zoneas we have assumed that all of the particles vaporizes just before the flame, the reaction and vaporization terms are not considered. thus where the subscript denotes conditions at the interface between post flame zone and the reaction zone. the jump conditions across the reaction zone which are used in finding flame position are where the subscript denotes conditions at the interface between preheat-vaporization zone and the reaction zone. 3.0 results and discussion for purpose of illustration it will be assumed that the gaseous fuel that evolves from the fuel particles is methane. the values of , , , and are considered to be , , , and respectively, based on (seshadri et al,. 1992). also , and are chosen from (seshadri et al,. 1992)based on and assumption. where is equivalence ratio based on fuel available in the ambient reactant stream and is radius of fuel particle in the ambient reactant stream. the value of is calculated with trial and error method. a value is guessed. then equation 22 is solved and the answer is substituted in 18 neglecting reaction term. this equation is solved numerically to find the first term of equation 23.at last eq. 24 is used to find the next guess.this is done until convergence. the value that is obtained is , which results in =1.71, which results in 87 where boundary conditions are used in writing eq. 23. also as the value of is sufficiently high, it is assumed that , where is gaseous fuel mass fraction in the reaction zone. in the post flame zoneas we have assumed that all of the particles vaporizes just before the flame, the reaction and vaporization terms are not considered. thus where the subscript denotes conditions at the interface between post flame zone and the reaction zone. the jump conditions across the reaction zone which are used in finding flame position are where the subscript denotes conditions at the interface between preheat-vaporization zone and the reaction zone. 3.0 results and discussion for purpose of illustration it will be assumed that the gaseous fuel that evolves from the fuel particles is methane. the values of , , , and are considered to be , , , and respectively, based on (seshadri et al,. 1992). also , and are chosen from (seshadri et al,. 1992)based on and assumption. where is equivalence ratio based on fuel available in the ambient reactant stream and is radius of fuel particle in the ambient reactant stream. the value of is calculated with trial and error method. a value is guessed. then equation 22 is solved and the answer is substituted in 18 neglecting reaction term. this equation is solved numerically to find the first term of equation 23.at last eq. 24 is used to find the next guess.this is done until convergence. the value that is obtained is , which results in (25) 88 figure 2 effect of lewis number on gaseous mass fraction distribution in preheat-vaporization zone eqs. 25 and 26 are used to find gaseous fuel mass fraction distribution in the preheat-vaporization zone, which the result is plotted in figure 2. this figure shows that lewis number has a dual effect on gaseous fuel mass fraction distribution. first, since heat diffusion increases with increasing values of lewis number, gaseous fuel mass fraction can be expected to become larger near the reaction zone, as lewis number increases, because pyrolysis phenomena is increased. on the other side, larger lewis number means lower mass diffusion, which decreases gaseous fuel mass fraction far from the reaction zone. 4.0 conclusions in this work the structure of a one dimensional, axisymmetric, premixed flame in a counterflow configuration containing uniformly distributed volatile fuel particles is examined. effect of lewis number on gaseous mass fraction distribution in preheat-vaporization zone is investigated, which shows increasing in lewis number has dual effect on gaseous mass fraction distribution based on increasing in heat diffusion and decreasing in mass diffusion. this dual effect results a higher value of gaseous mass fraction near the reaction zone and a lower value far from the reaction zone, in a higher lewis number value. 5.0 references m. bidabadi, a.haghiri, a. rahbari. 2010, the effect of lewis and damk hler numbers on the flame propagation through micro-organic dust particles, int. j. thermal sci 49, pp. 534-542. j. daou. 2011, strained premixed flames: effect of heat loss, preferential diffusion and reversibility of the reaction, combust. theory model. 15:4, pp. 437-454. r. w. thatcher, e.alsarairah, steady and unsteady flame propagation in a premixed counterflow, combust. theory model, 11:4, pp. 569-583. h. y. wang, w. h. chen, and c. k. law. 2007, extinction of counterflow diffusion flames with radiative heat loss and nonunity lewis numbers, combust. flame 148, pp. 100-116. r. k. eckhof.2006 , differences and similarities of gas and dust explosions: a critical evaluation of the european 'atex' directives in relation to dusts. j. loss prev. process ind. 19, pp. 553560. (26) 88 figure 2 effect of lewis number on gaseous mass fraction distribution in preheat-vaporization zone eqs. 25 and 26 are used to find gaseous fuel mass fraction distribution in the preheat-vaporization zone, which the result is plotted in figure 2. this figure shows that lewis number has a dual effect on gaseous fuel mass fraction distribution. first, since heat diffusion increases with increasing values of lewis number, gaseous fuel mass fraction can be expected to become larger near the reaction zone, as lewis number increases, because pyrolysis phenomena is increased. on the other side, larger lewis number means lower mass diffusion, which decreases gaseous fuel mass fraction far from the reaction zone. 4.0 conclusions in this work the structure of a one dimensional, axisymmetric, premixed flame in a counterflow configuration containing uniformly distributed volatile fuel particles is examined. effect of lewis number on gaseous mass fraction distribution in preheat-vaporization zone is investigated, which shows increasing in lewis number has dual effect on gaseous mass fraction distribution based on increasing in heat diffusion and decreasing in mass diffusion. this dual effect results a higher value of gaseous mass fraction near the reaction zone and a lower value far from the reaction zone, in a higher lewis number value. 5.0 references m. bidabadi, a.haghiri, a. rahbari. 2010, the effect of lewis and damk hler numbers on the flame propagation through micro-organic dust particles, int. j. thermal sci 49, pp. 534-542. j. daou. 2011, strained premixed flames: effect of heat loss, preferential diffusion and reversibility of the reaction, combust. theory model. 15:4, pp. 437-454. r. w. thatcher, e.alsarairah, steady and unsteady flame propagation in a premixed counterflow, combust. theory model, 11:4, pp. 569-583. h. y. wang, w. h. chen, and c. k. law. 2007, extinction of counterflow diffusion flames with radiative heat loss and nonunity lewis numbers, combust. flame 148, pp. 100-116. r. k. eckhof.2006 , differences and similarities of gas and dust explosions: a critical evaluation of the european 'atex' directives in relation to dusts. j. loss prev. process ind. 19, pp. 553560. figure 2 effect of lewis number on gaseous mass fraction distribution in preheat-vaporization zone eqs. 25 and 26 are used to find gaseous fuel mass fraction distribution in the preheat-vaporization zone, which the result is plotted in figure 2. this figure shows that lewis number has a dual effect on gaseous fuel mass fraction distribution. first, since heat diffusion increases with increasing values of lewis number, gaseous fuel mass fraction can be expected to become larger near the reaction zone, as lewis number increases, because pyrolysis phenomena is increased. on the other side, issn: 2180-1053 vol. 4 no. 1 january-june 2012 counterflow combustion of micro organic particles 105 larger lewis number means lower mass diffusion, which decreases gaseous fuel mass fraction far from the reaction zone. 4.0 conclusions in this work the structure of a one dimensional, axisymmetric, premixed flame in a counterflow configuration containing uniformly distributed volatile fuel particles is examined. effect of lewis number on gaseous mass fraction distribution in preheat-vaporization zone is investigated, which shows increasing in lewis number has dual effect on gaseous mass fraction distribution based on increasing in heat diffusion and decreasing in mass diffusion. this dual effect results a higher value of gaseous mass fraction near the reaction zone and a lower value far from the reaction zone, in a higher lewis number value. 5.0 references m. bidabadi, a.haghiri, a. rahbari. 2010, the effect of lewis and damkohler numbers on the flame propagation through micro-organic dust particles, int. j. thermal sci 49, pp. 534-542. j. daou. 2011, strained premixed flames: effect of heat loss, preferential diffusion and reversibility of the reaction, combust. theory model. 15:4, pp. 437-454. r. w. thatcher, e.alsarairah, steady and unsteady flame propagation in a premixed counterflow, combust. theory model, 11:4, pp. 569-583. h. y. wang, w. h. chen, and c. k. law. 2007, extinction of counterflow diffusion flames with radiative heat loss and nonunity lewis numbers, combust. flame 148, pp. 100-116. r. k. eckhof.2006 , differences and similarities of gas and dust explosions: a critical evaluation of the european 'atex' directives in relation to dusts. j. loss prev. process ind. 19, pp. 553-560. k. seshadri, a. l. berlad, and v. tangirala. 1992, the structure of premixed particle-cloud flames, combust. flame 89, pp. 333-342. issn: 2180-1053 vol. 10 no.1 january – june 2018 31 on the prediction of peak temperatures in tool of different profiles in friction stir welding using improved heat generation models l. o. jayesimi1* and 2m. a. waheed 1 works and physical planning department, university of lagos, akoka, lagos, nigeria. 2 department of mechanical engineering, federal university of agriculture, abeokuta, nigeria. abstract the process of frictional stir welding in which non-consumable rotating tool with a specially designed pin and shoulder which are inserted into the abutting edges of sheets or plates to be joined and traversed along the line of joint. the initial modeling approaches of the thermomechanical process in the frictional stir welding (fsw) focused on the estimation of heat generation during the process. in this work, new model for the prediction of the peak temperatures in tools of different profiles fsw tools is presented through an improved analytical heat generation models. the developed models take into considerations that the welding process is a combination or mixture of the pure sliding and the pure sticking. from the obtained results, it is observed that increasing the tool rotational speed at constant weld speed increases the heat input, whereas the heat input decreases with an increase in the weld speed at constant tool rotational speed. also, it was observed that the rate of heat generation at the shoulder is more in flat shoulder that the conical shoulder. the results in this work agreed with the experimental results. therefore, the improved models could be used to estimate the heat generation in fsw tool. keywords: peak temperature; heat generation model; tool of different profiles; friction stir welding. 1.0 introduction welding as a fabrication process involves joining of materials, usually metals or thermoplastics, by causing fusion. the methods of welding include oxy-fuel welding, shielded metal arc welding (smaw), gas tungsten arc welding (gtaw), gas metal arc welding (gmaw), fluxcored arc welding (fcaw), submerged arc welding (saw), electroslag welding (esw), electric resistance welding (erw). although less common, there are also solid state welding processes such as friction welding in which metal does not melt. the friction stir welding (fsw), which is a contemporary relatively efficient novel solid-state welding method invented at the welding institute (twi) of uk in 1991 has shown to be remarkably simple welding technique. it is considered to be the most significant development in metal joining in a decade and is a‘‘green’’ technology due to its energy efficiency, environment friendliness, and versatility. however, in such significant welding technique, the fundamental knowledge of the *corresponding author e-mail: lawrenceunilag@yahoo.com https://en.wikipedia.org/wiki/oxy-fuel_welding https://en.wikipedia.org/wiki/shielded_metal_arc_welding https://en.wikipedia.org/wiki/shielded_metal_arc_welding https://en.wikipedia.org/wiki/gas_tungsten_arc_welding https://en.wikipedia.org/wiki/gas_metal_arc_welding https://en.wikipedia.org/wiki/flux-cored_arc_welding https://en.wikipedia.org/wiki/flux-cored_arc_welding https://en.wikipedia.org/wiki/submerged_arc_welding https://en.wikipedia.org/wiki/electroslag_welding https://en.wikipedia.org/wiki/electric_resistance_welding https://en.wikipedia.org/wiki/friction_welding journal of mechanical engineering and technology 32 issn: 2180-1053 vol. 10 no.1 january – june 2018 thermal impact and thermomechanical processes of the technique is still not completely understood (thomas et al., 1991, schneider 2007). understanding the heat generation and the temperature history during the fsw process is the first step towards understanding the thermomechanical interaction taking place during the welding process. modelling of heat generation and the optimum parameters during fsw can potentially accelerate the development of the welding process since the central issue in all cases is the determination of the heat input. in addressing the issue, several methods involving experimental analysis have been adopted to calibrate heat flow and maximum temperature but none of these approaches enable the heat generation and welding temperature to be predicted without an experimental measurement of some kind and in most cases, trial and error method is adopted. the determination of precise amount of heat generated during friction stir welding process is complicated since there are various uncertainties, assumptions and simplifications of mathematical model that describes welding process. various experiments conducted around the planet, from the very beginning of the fsw method’s application gave dispersive results about the generated heat. a more accurate and predictive approach uses the 3-dimensional flow field to calculate the heat generation from the material viscous dissipation. even with these more sophisticated models there is conjecture over the best ways to describe the material behaviour, and the interface between the workpiece material and the tool, i.e. is there stick or slip. the analytical heat generation estimate correlates with the experimental heat generation, by assuming either a sliding or a sticking condition. however, the main uncertainties about process are when welding condition is a mixture of sliding and sticking. in this situation ambiguity of the value of the friction coefficient in every moment of the welding process, contact pressure between weld tool and weld pieces and shear stress are main reasons for difference between analytical and experimental result. the process of heat generation and peak temperature during fsw are complex and challenging tasks that require a multidisciplinary approach. therefore, the seemingly simple task of predicting the weld heat generation and peak temperature has proved beyond the ability of most models. previous works on modelling the fsw process for heat generation and peak temperature from the fsw tool are based on assumptions regarding the interface condition, which led some limitations and inaccuracies. in the model by chao et al. (2003) developed heat generation model based on the assumption of sliding friction, where coulomb’s law is used to estimate the shear or friction force at the interface. also, in their model, the pressure at the tool interface is assumed to be constant, thereby enabling a radially dependent surface heat flux distribution as a representation of the friction heat generated by the tool shoulder, but neglecting that generated by the probe surface. frigaard et al. (2001) modelled the heat input from the tool shoulder and probe as fluxes on squared surfaces at the top and sectional planes on a three-dimensional model and control the maximum allowed temperature by adjustment of the friction coefficient at elevated temperatures. russell and shercliff (1999) based the heat generation on a constant friction stress at the interface, equal to the shear yield stress at elevated temperature, which is set to 5% of the yield stress at room temperature. colegrove et al. (2007) used an advanced analytical estimation of the heat generation for tools with a threaded probe to estimate the heat generation distribution. the fraction of heat generated by the probe is estimated to be as high as 20%, which leads to the conclusion that the analytical estimated probe heat generation contribution is not negligible. also, the real situation during the welding process is a combination of the pure sliding and the pure sticking. therefore, in this work improved analytical models are developed for the predictions of heat generation in different profiles fsw tools. the developed models take into on the prediction of peak temperatures in tool of different profiles in friction stir welding using improved heat generation models issn: 2180-1053 vol. 10 no.1 january – june 2017 33 considerations that the welding process is a combination or mixture of the pure sliding and the pure sticking. the results in this work agreed with the conclusion of the past work. therefore, the improved models could be used to estimate the heat generation in fsw tool. it should be noted that the heat generation in fsw is a complex transformation process where one part of the mechanical energy is delivered to the workpiece, which is consumed in welding, while another is used for the deformation process and the rest of the energy is transformed into heat (gadakh and kumar, 2013). ulysse (2002) had earlier pointed out that 80-90% of the mechanical power delivered to the welding tool transforms into heat while recently, pala and phaniraj, (2015) showed that 10-20% of the total heat generated is transfer to the tool as shown in table 1. table 1. heat generation and heat partitioning with change in welding speed. in the process, heat is generated by friction (frictional heat) and by plastic deformation (plastic deformation). both types of heat appear simultaneously on the fsw and they influence each other. also, it should be noted that this heat is conducted to both the tool and the workpiece. the amount of the heat conducted into the workpiece dictates a successful fsw process, the quality of the weld, shape of the weld, micro-structure of the weld, as well as the residual stress and the distortion of the workpiece. the amount of the heat gone to the tool dictates the life of the tool and the capability of the tool for the joining process. insufficient heat from the friction could lead to breakage of the pin of the tool since the material is not soft enough. therefore, understanding the heat generation phenomena and the heat transfer aspects of the fsw process is fundamental to all other aspects of the welding process. moreover, the influences of the tool geometry on thermal cycles, peak temperatures, power requirements, and torque during fsw processes are complex and remain to be fully understood. consequently, the tool design is currently carried out by trial and error methods. the current effort in this work is directed towards development of mathematical models that will predict the maximum temperature during frictional stir welding. 2.0 development of heat generation models for the friction stir welding consider the friction stir welding (fsw) shown schematically in figure 2. during the fsw, a rotating tool moves along the joint interface. as the tool moves along the joint and into the workpiece, heat generated at surface and near the interface between the tool and the work-piece is transported into the workpiece and the tool (figure 2). the total heat generated at different portions of the tool is the summation of the heat generated at the tool shoulder surface, heat generated at the tool pin/probe side and the heat generated at the tool pin/probe tip. also, during the frictional stir welding, heat is generated by pure sliding (adhesion) and puresticking (deformation). in pure sliding condition, it is assumed that there is shear in the contact interface journal of mechanical engineering and technology 34 issn: 2180-1053 vol. 10 no.1 january – june 2018 and can be described as fully coulomb friction condition. in the assumption, the contact pressure between tool and weld piece p and friction coefficient μ are constant or linearly dependable values from various variables. the shear stress becomes equals to dynamic contact shear stress. in the pure sticking, it is assumed shearing in the layer of the material of weld pieces very close to the contact surface and uniformity of the shear stress τ. in this situation, surface of the weld piece will stick to the moving tool’s surface only if friction shear stress exceeds the yield shear stress of the weld piece. the real situation during welding process gives combination of the pure sliding and the pure sticking. therefore, it is absolutely correct to say that heat generating during friction stir welding is product of pure sliding, pure sticking and combination of sliding and sticking durdanovic et. al., (2009). figure 2. heat generation in fsw (mijajlović and milčić , 2012) 2.1 model development for heat generation for flat circular/straight cylindrical tool as pointed out, in this work, the heat generated in fsw was considered to be due to friction (due total sliding condition only), but practically, it is due to friction as well as deformation (due to sticking condition) (gadakh and kumar, 2013). considering both types of heat and their influence on each other, the total amount of heat generated on the pin tip, pin side and the shoulder tip are respectively given by (1 ) fr def pt pt pt pt pt q q q    (1) (1 ) fr def ps ps ps ps ps q q q    (2) (1 ) fr def st st st st st q q q    (3) where the heat indexed with fr represents frictional heat, heat indexed with def represents deformation heat, δpt, δps, δst are dimensionless contact state variable (extension of slip) at the pin on the prediction of peak temperatures in tool of different profiles in friction stir welding using improved heat generation models issn: 2180-1053 vol. 10 no.1 january – june 2017 35 tip, pin side and shoulder tip, respectively. it should be noted that δpt=0.1, δps= 0.2 and δst= 0.1 (jauhari, 2012). it should be noted that if  is 1, full sticking condition is applied, and all the heat is generated by plastics deformation. when 0 , heat is generated only by friction. the analytical estimation based on a general assumption of uniform contact shear stress τcontact is considered. i. weld cycle excludes plunging; first, second dwell, and retract cycles. ii. tool inclination angle was not considered. iii. no heat flows into the workpiece if the local temperature reaches the material melting temperature. iv. the axial pressure is evenly distributed along z-axis v. due to friction and deformation interface conditions, the frictional and deformation shear stresses are considered. vi. the thread on the pin side of the welding tool was neglected (a) (b) (c) figure 3. active surfaces in fsw: (a) pin tip, (b) pin side, (c) shoulder tip the general expression for an infinitesimal amount of heat generation at each of the different zones of the tool/workpiece interface is given by dq rdf (4) where dq is the heat generated per unit time, df is the force acting on the surface at a distance r from the tool centerline and ω is the angular velocity of the tool. contact df da (5) where τcontact is the contact shear stress and da rdrd is the area of the infinitesimal segment on the surface. the frictional and deformation amount of heat with respect to the contact shear stress is given by journal of mechanical engineering and technology 36 issn: 2180-1053 vol. 10 no.1 january – june 2018 (coulumb's friction law) cont yield p for frictional heat generation for deformational heat generation        (6) where μ is the frictional coefficients, p is the contact pressure, τyield is the yield strength of the workpiece. following arora et al., (2009, 2011) frictional coefficients can be calculated as p o slip o s r exp r             (7) where μo is the static friction coefficient and it is taken as 0.45 (el-tayeb et al., 2009), (devaraju et al., 2013). slip  is the slipping factor, ω is the rotating speed and the reference rotation speed ωois taken as 400 rpm. rp and rs are the radii of the tool pin and the shoulder, respectively. substituting equations (5) and (6) into equation (4) and integrate, for the shoulder tip frictional heat generation, gives 2 0 s p r fr st r q u prdrd      (8)   2 0 s p r fr st w r q p r v sin r drd        (9) where w u r v sin   , the positive and the negative sign denote advancing and retracting movement of the tool. vwis the welding velocity 2 2 0 ( ) s p r fr st w r q p r rv sin drd        (10) 2 2 2 0 0 s s p p r r fr st w r r q p r drd rv sin drd                   (11) after the integration, one arrives at  3 3 2 3 fr st s p q p r r  (12) similarly, for the shoulder tip deformational heat generation,  3 3 2 3 def st yield s p q r r  (13) the total heat generationat the shoulder tip is on the prediction of peak temperatures in tool of different profiles in friction stir welding using improved heat generation models issn: 2180-1053 vol. 10 no.1 january – june 2017 37    3 3 3 3 2 2 (1 ) 3 3 st s p st yield st s p q p r r r r        (14) it should be noted that not all the mechanical energy is converted to frictional and deformational heat.    3 3 3 3 2 (1 ) 3 st fd s p st yield st s p q p r r r r              (15) the total heat generation at the interfaces is the summation of the total heat generation at the shoulder tip, total heat generation at the pin tip and total heat generation at the pin side i.e. total st pt ps q q q q       (16) for the flat shoulder and flat pin, total heat generation at the interfaces is given as    3 3 3 3 3 3 2 2 (1 ) (1 )2 3 (1 ) s p st yield st s p p pt total fd yield pt p p p ps yield p p ps p r r r r pr q r pr l r l                             (17) the energy per unit length of the weld of the flat shoulder and flat pin tool is    3 3 3 3 3 ' 2 3 2 2 (1 ) (1 )2 3 (1 ) s p st yield st s p p ptfd enery s yield pt p p p ps yield p p ps p r r r r pr q vr r pr l r l                            (18) 2.2 model development for heat generation for conical circular/straight cylindrical tool (a) (b) figure 4. fsw tool (a) conical shoulder with flat pin (b) conical shoulder with conical pin, d=rp and d= rs (roy et al., 2006) journal of mechanical engineering and technology 38 issn: 2180-1053 vol. 10 no.1 january – june 2018 also, the heat generation models for conical/tapered tool are derived in similar way as shown above in equations (4), (5) and (8-18). for the conical shoulder and flat pin, total heat generation at the interfaces is given as    3 3 3 3 3 3 2 2 (1 )(1 ) (1 )2 3 (1 ) (1 ) s p st yield st s p total fd p pt yield pt p p p ps yield p p ps p r r tan r r tan q pr r pr l r l                                 (19) if the shoulder is flat and the pin conical, total heat generation at the interfaces    3 3 3 3 3 3 2 2 (1 ) (1 ) 2 3 1 (1 ) 1 2 2 s p st yield st s p p pt yield pt p total fd p p ps yield p p ps p r r r r pr r q pr l tan r l tan                                              (20) the energy per unit length of the weld of the flat shoulder conical pin tool is    3 3 3 3 3 3 ' 2 2 2 (1 ) (1 ) 2 3 1 (1 ) 1 2 2 s p st yield st s p p pt yield pt p fd energy s p p ps yield p p ps p r r r r pr r q vr pr l tan r l tan                                              (21) if the shoulder and the pin are conical with different conical angles, total heat generation at the interfaces    3 3 3 3 3 ' 3 2 2 (1 )(1 ) (1 ) (1 ) 2 3 1 (1 ) 1 2 2 s p st yield st s p p pt energy fd yield pt p p p ps yield p p ps p r r tan r r tan pr q r pr l tan r l tan                                                  (22) the energy per unit length of the weld for the conical shoulder tool is    3 3 3 3 3 ' 2 3 2 2 (1 )(1 ) (1 ) (1 ) 2 3 1 (1 ) 1 2 2 s p st yield st s p p pt fd energy s yield pt p p p ps yield p p ps p r r tan r r tan pr q vr r pr l tan r l tan                                                  (23) 2.3 development of heat generation models for different pin profiles the pin geometry plays a vital role for material flow, temperature history, grain size, and mechanical properties in the fsw processes (gadakh and kumar, 2013). therefore, in this on the prediction of peak temperatures in tool of different profiles in friction stir welding using improved heat generation models issn: 2180-1053 vol. 10 no.1 january – june 2017 39 section, we developed heat generation models are developed for different pin profiles with flat and conical shoulder. (a) (b) (c) (d) figure 5. different profiles pin used in fsw. (a) triangular pin. (b) square pin (c). pentagonal pin. (d) hexagonal pin 2.3.1 triangular profile pin following the same procedure of derivation as in the previous section, we arrived at the total heat generation for the triangular profile pin with flat shoulder is 3 3 3 3 3 2 2 3 3 (1 ) 3 9 9 2 3 3 (1 ) 3 9 9 1 (1 ) 2 s st yield st s total fd pt yield pt p ps yeild ps a a p r r q a p a l p                                                                    (24) the energy per unit length of the weld for the flat shoulder tool is 3 3 3 3 ' 2 2 2 2 3 3 (1 ) 3 9 9 2 3 3 1 (1 ) (1 ) 3 9 9 2 s st yield st s fd energy s pt yield pt p ps yeild ps a a p r r q vr a pa a l p                                                             (25) journal of mechanical engineering and technology 40 issn: 2180-1053 vol. 10 no.1 january – june 2018 while the total heat generation for the triangular profile pin with conical shoulder is     3 3 3 3 2 2 2 3 3 (1 ) 1 1 3 9 9 2 3 3 1 (1 ) (1 ) 3 9 9 2 s st yield st s total fd pt yield pt p ps yeild ps a a p r tan r tan q a pa a l p                                                                  (26) the energy per unit length of the weld for the conical shoulder tool is     3 3 3 3 ' 2 2 2 2 3 3 (1 ) 1 1 3 9 9 2 3 3 1 (1 ) (1 ) 3 9 9 2 s st yield st s fd energy s pt yield pt p ps yeild ps a a p r tan r tan q vr a pa a l p                                                                 (27) 2.3.2 square profile pin the total heat generation for the square profile pin with flat shoulder is 3 3 3 3 3 3 2 2 2 2 (1 ) 3 4 4 2 2 2 (1 ) 3 4 4 1 (1 ) 4 s st yield st s total fd pt yield pt p ps yeild ps a a p r r a a q p a l p                                                                    (28) 3 3 2 3 3 3 3 3 3 2 2 2 2 1 (1 ) (1 ) 3 4 4 4 2 2 2 (1 ) 3 4 4 2 2 2 (1 ) 3 4 4 1 (1 ) 4 pt yield pt p ps yeild ps p p total s st yield st s pt yield pt p ps yeild ps a a p a l p q f q a a p r r a a p a l p                                                                                           (29) the energy per unit length of the weld for the flat shoulder tool is on the prediction of peak temperatures in tool of different profiles in friction stir welding using improved heat generation models issn: 2180-1053 vol. 10 no.1 january – june 2017 41 3 3 3 3 3 3 ' 2 2 2 2 2 (1 ) 3 4 4 2 2 2 (1 ) 3 4 4 1 (1 ) 4 s st yield st s fd energy pt yield pt s p ps yeild ps a a p r r a a q p vr a l p                                                                   (30) the total heat generation for the square profile pin with conical shoulder is     3 3 3 3 3 3 2 2 2 2 (1 ) 1 1 3 4 4 2 2 2 1 (1 ) (1 ) 3 4 4 4 s st yield st s total fd pt yield pt p ps yeild ps a a p r tan r tan q a a p a l p                                                                  (31) the energy per unit length of the weld for the conical shoulder tool is     3 3 3 3 ' 2 3 3 2 2 2 2 (1 ) 1 1 3 4 4 2 2 2 1 (1 ) (1 ) 3 4 4 4 s st yield st s fd energy s pt yield pt p ps yeild ps a a p r tan r tan q vr a a p a l p                                                                 (32) 2.3.3 pentagonal profile pin the total heat generation for the pentagonal profile pin with flat shoulder is    3 3 3 3 3 3 2 2 2 0.6155 (1 ) 0.6155 3 3 2 2 (0.6155 )(1 ) (0.6155 ) 3 3 1.8088 (1 ) s st yield s total fd pt yield pt fd p ps yield ps p r a r a q p a a pa l p                                           (33) the energy per unit length of the weld for the flat shoulder tool is journal of mechanical engineering and technology 42 issn: 2180-1053 vol. 10 no.1 january – june 2018    3 3 3 3 ' 3 3 2 2 2 2 0.6155 (1 ) 0.6155 3 3 2 2 (0.6155 )(1 ) (0.6155 ) 3 3 1.8088 (1 ) s st yield s fd energy pt yield pt s fd p ps yield ps p r a r a q p a a vr pa l p                                          (34) and the total heat generation for the pentagonal profile pin with conical shoulder is        3 3 3 3 3 3 2 0.6155 (1 ) 12 2 (0.6155 )(1 ) 3 30.6155 1 2 (0.6155 ) 1.8088 (1 ) 3 s st fd pt yield s total fd yield pt p ps yield ps p r a tan p a r a tanq a pa l p                                              (35) the energy per unit length of the weld for the conical shoulder tool is       3 3 3 3 ' 3 3 2 2 2 2 0.6155 (1 ) 1 0.6155 1 3 3 2 2 (0.6155 )(1 ) (0.6155 ) 3 3 1.8088 (1 ) s st yield s fd energy pt yield pt s fd p ps yield ps p r a tan r a tan q p a a vr pa l p                                              (36) 2.3.4 hexagonal profile pin the total heat generation for the hexagonal profile pin with flat shoulder is    3 3 3 3 3 3 2 2 2 2 (1 ) (1 ) 3 3 3 2 3 (1 ) 3 s st yield s pt total fd yield pt p ps yield ps p r a r a pa q a pa l p                                    (37) the energy per unit length of the weld for the conical shoulder tool is    3 3 3 3 ' 3 3 2 2 2 (1 ) 3 2 2 (1 ) 3 3 3 (1 ) s st yield s fd energy pt yield pt s p ps yield ps p r a r a q pa a vr pa l p                                            (38) on the prediction of peak temperatures in tool of different profiles in friction stir welding using improved heat generation models issn: 2180-1053 vol. 10 no.1 january – june 2017 43 the total heat generation for the hexagonal profile pin with conical shoulder is       3 3 3 3 3 3 2 2 (1 ) 1 1 3 2 2 (1 ) 3 (1 ) 3 3 s st yield s total fd pt yield pt p ps yield ps p r a tan r a tan q pa a pa l p                                           (39) the energy per unit length of the weld for the conical shoulder tool is       3 3 3 3 ' 2 3 3 2 2 (1 ) 1 1 3 2 2 (1 ) 3 (1 ) 3 3 s st yield s fd energy s pt yield pt p ps yield ps p r a tan r a tan q vr pa a pa l p                                          (40) where ηfd represents the fraction of the mechanical energy that is converted to frictional heat and deformational heat. which could be as high as 0.99 based on the assumptions of previous work the boundary value of the yield shear stress from the von misses yield criterion in uniaxial tension and pure shear is given by ( , ) 3 yield yield t    (41) the yield strength of the workpiece’s material ( , ) yield t  is highly dependent on temperature, t and strain rate, ε. the analysis of the tangential stresses within fsw requires the full temperature and strain history in the workpiece in a wide zone around the welding tool. sheppard and wright [17] elastic-plastic model may be used to evaluate the temperature-strain dependent yield strength of the workpiece’s material, ( , ) yield t  . 1 11 ( , t) ( , ) n yield z t sinh a                  (42) where a, α, and n are material constants and z(ε,t) is the zener-hollomon parameter that represents the temperature-compensated effective strain rate by ( , ) xp q z t e rt          (43) journal of mechanical engineering and technology 44 issn: 2180-1053 vol. 10 no.1 january – june 2018 where . , , ,q r and t are strain rate, activation energy, universal gas constant and absolute temperature, respectively. sheppard and jackson (1997) developed the elastic-plastic model for yield strength of the workpiece’s material as 1 1 2 2 1 ( , ) ( , ) ( , ) 1 n n yield z t z t t in a a                                (44) it was stated that the lack of the detailed material constitutive information and other thermal and physical properties at conditions such as very high strain rates and elevated temperatures seems to be the limiting factor while modeling the fsw process (uyyuru and kallas, 2006). consequently, colegrove and sherchiff (2006) and wang et al. (2013) pointed out that sheppard and jackson’s elastic-plastic model is not applicable at the melting of the material. although, su et al. (2013) modified the sheppard and jackson’s elastic-plastic model as 1 1 2 2 1 273 ( , ) ( , ) ( , ) 1 1 273 n n yield o t z t z t t in t a a                                               (45) however, analysis of heat generation in fsw can neglect the influence of strain on the decrease of yield strength and still maintain sufficient precision (schmidt et al., 2004). neglecting strain effects on the yield strength is possible since the maximal temperatures of the material reach about 80% of the melting temperature when the strain has significant values due to near melting conditions in the material (arora et al., 2011),(arora et al.,2009). therefore, equation (16) becomes ( ) 3 yield yield t   (46) for stainless steel, yield stress as developed in this work 2 3 4 5 1 2 3 4 5 ( ) yield o t t t t t t            (47) where 1 2 5 1 2 3 8 8 4 5 240, 7.3583 10 , 7.1333 10 , 2.163 10 , 2.7292 10 1.1849 10 o and                         on the prediction of peak temperatures in tool of different profiles in friction stir welding using improved heat generation models issn: 2180-1053 vol. 10 no.1 january – june 2017 45 3.0 modeling the peak temperature in friction stir welding the published works in literature point to the fact that there is an optimum temperature range to obtain defect-free joints and such a range has not been specified. however, a number of researchers as shown that there is a linear regression of the temperature ratio tpeak/ts (where tpeak is the peak/maximum temperature and ts is the solidus temperature) on tswas derived: where this temperature range can be thought of as the optimum temperature range, i.e. tpeak = topt. the correlation has a standard deviation of 0.024. the calculation results in a temperature range topt = (0.8–0.9) ts. the rationality of this assumption was verified by experiments (khandkar et al., 2003). the linear relationship is given by 1 2 peak opt s s s t t t t t     (48) which gives 1 2 ( ) peak s s t t t   (49) where 1 2 and  are to be determined from experiment e.g. for aluminum alloy 1100-h14 and 2024-t3 rolled plates 8 and 3.2 mm in thick, 1 2 1.344 0.0005917and   . also, empirical model developed by hamilton et al. (2008) shown in equation (50) could be adopted 0.54 0.000156 q peak max s t t   (50) where ' q q max energy  (51)  is the ratio of the pin length lp to the workpiece thickness, t. ' q energy are defined in previous equations for different tool profiles. on this work, our analysis establishes a non-linear regression of the peak temperature and maximum heat generation, qmax (q q) peak max t   (52) also, q is the non-dimensional heat input, which is defined by roy et al., (2006). journal of mechanical engineering and technology 46 issn: 2180-1053 vol. 10 no.1 january – june 2018 8 2 q w s c k v     (53) where, for conformity of calculation, the unit of ω changes from rpm to rad m/s and v from mm/min to m/s, 8  is the yield stress of the material at a temperature of 0.8ts, s is the crosssectional area of the tool shoulder, c is the specific heat capacity of the workpiece material, kwis the thermal conductivity of the workpiece, and η is the ratio according to which heat generated at the shoulder–workpiece interface is transported between the tool and the workpiece, and is defined as ( ) ( ) p w p t k c k c     (54) also 30 ( , ) yield nr v n t     (55) substituting equation. (55) into equation. (53), one arrives at 8 2 30 q 30 ( , ) w yield n s c nr k n t                      (56) also, on substituting equation. (56) into equation. (53), gives 8 2 30 q 30 ( , ) peak max w yield n s c t nr k n t                                (57) on the prediction of peak temperatures in tool of different profiles in friction stir welding using improved heat generation models issn: 2180-1053 vol. 10 no.1 january – june 2017 47 4.0 results and discussions 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0 200 400 600 800 1000 1200 1400 q , t o ta l h e a t h e a t g e n e ra ti o n r a te a t th e s h o u ld e r ( k w ) dimensionless contact state variable flat/circular pin triangular pin square pin pentagonal pin hexagonal pin figure 6. effects of pin shape on the total heat generation rate at the interfaces figure 6 shows the effects of tool pin geometry on the total heat generation rate at the interfaces. from the results, it is depicted that by increasing the number of edges, the amount of heat generation initially increases from the square pin to hexagonal pin profile and then decreases to the triangular pin profile. furthermore, increasing the tool rotational speed under constant weld speed, heat input increases, and increasing the weld speed under constant tool rotational speed, heat input decreases. table 2: comparison of results of heat generated and peak temperature for different tool profiles table 2 shows the effects of tool pin geometry on the total heat generation and the peak temperature. from the results, it is depicted that by increasing the number of edges, the peak temperature increases. also, the present models developed in this work provide improved and journal of mechanical engineering and technology 48 issn: 2180-1053 vol. 10 no.1 january – june 2018 more accurate results than the other previous models as shown in the table. figure 7 shows the influence of shoulder radius on the rate of heat generation at the shoulder-workpiece, aluminum alloys (aa-6061-t6). 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.1 0.11 0 500 1000 1500 2000 2500 3000 3500 4000 4500 q , t o ta l h e a t h e a t g e n e ra ti o n r a te a t th e i n te rf a c e s (k w ) shoulder radius(m) figure 7. variations total heat generation rate at the interfaces with shoulder radius 0 0.001 0.002 0.003 0.004 0.005 0.006 0.007 0.008 0.009 0.01 365 370 375 380 385 390 395 400 405 410 q , t o ta l h e a t h e a t g e n e ra ti o n r a te a t th e i n te rf a c e s (k w ) pin lenght(m) figure 8. variations total heat generation rate at the interfaces with pin length. on the prediction of peak temperatures in tool of different profiles in friction stir welding using improved heat generation models issn: 2180-1053 vol. 10 no.1 january – june 2017 49 it could be inferred from the results that the shoulder radius is directly proportional to the total heat generated rate at the interface. i.e. as the shoulder radius increases, the rate at which heat is generated at the interface increases. the same trend was noticed in figure 8 and 9 where the total heat generation rate increases with increase in pin length and pin radius. this heat propagates either through conduction in the various parts of the workpiece and the tool or through convection to the environment. in addition, higher heat generation due to plastic deformation and smaller interfacial contact area with the workpiece leads to lower frictional heat generation relative to the pin. the failure of friction stir welded joints takes place at the heat-affected zone (haz) where the density of the needle-shaped precipitate is less. from the fractional heat generation rate analysis carried out in this study, it is shown that depending on the welding conditions, between 80 to 90% heat is generated at the tool shoulder and the remaining amount at other tool surfaces. this fact has also been confirmed in the experimental work carried out by nandan et al. (2006). indisputably, the proportion of the heat generated at the tool shoulder and the pin surfaces is determined by the tool geometry and the welding variables nandan et al. 2006. also, from the reported literature, it is understood that the pin geometry plays a vital role for material flow, temperature history, grain size, and mechanical properties in the fsw process (gadakh and kumar, 2013). figure 10 and 11 effects of angle of rotation on rate of heat generation when the extent of sticking are 0.65 (sticking and sliding condition) and 1 (full sticking condition). the non-uniformity in the heat generation pattern results from the difference in the relative velocity at different angular locations on the pin surface, which arises due to the variation in term usinθ. 0 0.001 0.002 0.003 0.004 0.005 0.006 0.007 0.008 0.009 0.01 360 380 400 420 440 460 480 500 520 q , t o ta l h e a t h e a t g e n e ra ti o n r a te a t th e i n te rf a c e s (k w ) pin radius(m) figure 9. variations total heat generation rate at the interfaces with pin radius journal of mechanical engineering and technology 50 issn: 2180-1053 vol. 10 no.1 january – june 2018 0 50 100 150 200 250 300 350 465 470 475 480 485 490 495 500 505 510 515 520 q , h e a t g e n e ra ti o n r a te (k w /m 2 ) angle (degree) figure 10. variations heat generate with angle of rotation when the extent of sticking is 0.65 0 50 100 150 200 250 300 350 -40 -30 -20 -10 0 10 20 30 40 q , h e a t g e n e ra ti o n r a te (k w /m 2 ) angle (degree) figure 11. variations heat generate with angle of rotation when the extent of sticking is 1 (full sticking condition) on the prediction of peak temperatures in tool of different profiles in friction stir welding using improved heat generation models issn: 2180-1053 vol. 10 no.1 january – june 2017 51 the angular variations of temperature on the tool surface results from the local differences in the heat generation rates. therefore, meaningful modeling of temperature and plastic flow fields must consider 3d heat transfer. 0 50 100 150 200 250 300 350 1396 1396.2 1396.4 1396.6 1396.8 1397 1397.2 1397.4 1397.6 1397.8 1398 q , h e a t g e n e ra ti o n r a te (k w /m 2 ) angle (degree) figure 12. variations heat generate with angle of rotation when the extent of sticking is 0 (full sliding condition) figure 12 shows the variations of heat generation with angle of rotation when the extent of sticking is 0 (full sliding condition). the result depicts that angle of rotation has no effect on the rate of heat generation when the extent of sticking is 0 as a constant value line is shown in the figure. journal of mechanical engineering and technology 52 issn: 2180-1053 vol. 10 no.1 january – june 2018 300 320 340 360 380 400 420 440 460 480 500 3000 4000 5000 6000 7000 8000 9000 10000 11000 12000 13000 q s , s h o u ld e r h e a t g e n e ra ti o n p e r m m l e n g h t o f w e ld (j /m m ) n, rotation speed of the shoulder (rpm) v=101 mm/min v=150 mm/min v=200 mm/min figure 13. effects of welding speed on the heat generation at the shoulder figure 13 to 15 present the effects of shoulder rotation speed, conical angle and contact conditions on heat generation. figure 13 shows variation of shoulder heat generation rate with welding rotational speed at different welding velocities of 101, 150 and 200 mm/minute while figure 14 shows the variation of shoulder heat generation with rotational speed of the shoulder at for conical and flat shoulders. 425 430 435 440 445 2500 2520 2540 2560 2580 2600 2620 q s , s h o u ld e r h e a t g e n e ra ti o n ( w ) n, rotation speed of the shoulder (rpm) conical shoulder flat shoulder figure 14. effects of shoulder conical angle on the heat generation at the shoulder on the prediction of peak temperatures in tool of different profiles in friction stir welding using improved heat generation models issn: 2180-1053 vol. 10 no.1 january – june 2017 53 0 10 20 30 40 50 60 70 80 5.4 5.5 5.6 5.7 5.8 5.9 6 6.1 6.2 t o ta l h e a t fl u x t o t h e w o rk p ic e a t s h o u ld e rw o rk p ie c e i n te rf a c e ( k w ) cone angle (degree) figure 15. effects of conical angle on the heat generation at the shoulder-workpiece interface as it could be seen from figure 13, that the rate of heat generated at the shoulder varies inversely proportional with the welding speed. this is due to the fact that at higher welding velocity, the heat input per unit length decreases as heat is dissipated over a wider region of the workpiece. at high rotational speed, the relative velocity between the tool and workpiece is high, and consequently the heat generation rate and the temperatures are also high. the rate of heat generation at the shoulder is more in flat shoulder that the conical shoulder as shown in figure 14. figure 16. bar chart for comparison of results for maximum temperature in aluminum alloy 0 200 400 600 800 1000 1200 t e st 1 t e st 2 t e st 3 t e st 4 t e st 5 t e st 6 t e st 7 t e st 8 t e st 9 m a x im u m t e m p e ra tu re ( k ) measured predicted journal of mechanical engineering and technology 54 issn: 2180-1053 vol. 10 no.1 january – june 2018 figure 17. line graph for comparison of results for maximum temperature in aluminum alloy this is because the flatness of the shoulder tip increases the tool-workpiece contact surfaces and thereby creating more friction during the process to generate frictional heat and consequently, increases the rate of heat generation. the influence of contact condition variables on the rate of heat generation at the shoulder and the pin as displayed in figure 15. as expected, the heat generation rate increases with the increase in contact condition variables because more heat are generated due to friction due to increased contacts between the tool and the workpiece. for the experimental conditions studied by nandan et al. (2006), the computed heat generation rates at the shoulder and the pin surfaces are presented in tables 3. the results show that depending on the welding conditions, between 80 to 90% heat is generated at the tool shoulder and the remaining amount at other tool surfaces. as shown in the results, the heat inputs from the shoulder and the pin as well as the maximum temperature of the workpiece increase with the weld and rotational speeds. from the analysis, it was found that the average absolute error between the experimental and the predicted maximum temperature is 0.090799, while average bias error of correlation is 0.00006446, the normalized standard deviation is 0.12047, correlation coefficient is 0.99961 and the coefficient of multiple determination is 0.99953. good agreements between the experimentally determined and the computed results at different monitoring locations indicate that the model can be used to examine the temperature profiles and cooling rates. 1 2 3 4 5 6 7 8 9 10 600 650 700 750 800 850 900 m a x im u m t e m p e ra tu re ( k ) serial number as appear in table measured predicted on the prediction of peak temperatures in tool of different profiles in friction stir welding using improved heat generation models issn: 2180-1053 vol. 10 no.1 january – june 2017 55 table 3 variation of heat generated and peak temperature at the tool shoulder and the pin surfaces the welding variables. 5.0 conclusions in this work, new models for the prediction of the peak temperatures in tools of different profiles fsw tools have been developed through an improved analytical heat generation models. the developed models take into considerations that the welding process is a combination or mixture of the pure sliding and the pure sticking. the results agreed with experimental results. therefore, the improved models could be used to estimate the heat generation in fsw tool. 6.0 references arora, a. debroy, t. and bhadeshia. h. k. d. h. (2011). back-of-the-envelope calculations in friction stir welding – velocities, peak temperature, torque, and hardness. acta materialia 592020–8. arora, a. nandan, r. reynolds, a. p. debroy. t. (2009), torque, power requirement and stir zone geometry in friction stir welding through modeling and experiments. scripta materialia, 60, 13–16. s/n weld speed (mm/s) rotational speed (rpm) heat input from shoulder (kw) heat input from tool pin (w) heat input from tool bottom (w) maximum temperature (k) (measured) maximum temperature (k) (predicted) 1 0.5 200 2.97 250.1 45.6 700.2 700.311 2 1 200 3.05 252.8 46.1 694.4 694.434 3 1.5 200 3.17 258.5 47.7 688.2 687.914 4 0.5 400 3.72 213.9 61.9 762.7 761.789 5 1 400 3.72 215.5 62.1 756 755.573 6 1.5 400 3.88 216.2 63.4 749.6 751.346 7 0.5 600 4.23 164.3 76.2 807.4 808.654 8 1 600 4.31 168.1 77 801.5 800.795 9 1.5 600 4.47 172.3 77.4 797.3 796.480 journal of mechanical engineering and technology 56 issn: 2180-1053 vol. 10 no.1 january – june 2018 chao, y.j., qi, x. and tang, w. (2003). heat transfer in friction stir welding: experimental and numerical studies, asme journal of manufacturing science and engineering 125, 138– 145. colegrove, p. a. shercliff h. r., zettler. r. (2007). a model for predicting the heat generation and temperature in friction stir welding from the material properties. science and technology of welding and joining 12, 284–297. devaraju, a. kumar, a. and kotiveerachari. b. (2013). influence of addition of grp/al2o3p with sicp on wear properties of aluminum alloy 6061-t6 hybrid composites via friction stir processing. the transactions of nonferrous metals society of china 23), 1275–1280 đurdanovic, m. b., mijajlovic, m. m. milcic, d. s. and stamenkovic d. s., (2009). heat generation during friction stir welding process, tribology in industry, 31, 8-14. el-tayeb, n. s. m. low, k. o. and brevern. p. v. (2009), on the surface and tribological characteristics of burnished cylindrical al-6061. tribology international 42, 320–326 frigaard, o., grong, o. and midling o. t. (2001). a process model for friction stir welding of age hardening aluminium alloys. metallurgical and materials transactions a. 321189– 1200. gadakh, v. s. and kumar, a. (2013). heat generation model for taper cylindrical pin profile in friction stir welding. journal of materials research and technology 2(4), 370–375. hamilton, c., dymek, s. and sommer, e. (2008). a thermal model for friction stir welding in aluminum alloys. int j mach tool manuf. 2811201130. jauhari. t. k. (2012). development of multi-component device for load measurement and temperature profile for friction stir welding process [m.sc thesis]. penang: universiti sains malaysia; unpublished. khandkar, m. z. h., khan j. a. and reynolds. r. a. (2003). prediction of temperature distribution and thermal history during friction stir welding: input torque based model, science and technology of welding and joining 8(3), 165-174. mijajlović, m. and milčić d., (2012). analytical model for estimating the amount of heat generated during friction stir welding: application on plates made of aluminium alloy. intech, open science 2024-t351, chapter 11 247–274. nandan, r., roy, g. and debroy, t. (2006), numerical simulation of three-dimensional heat transfer and plastic flow during friction stir welding. metallurgical and materials transactions a 37(4), 1247-1259. http://www.intechopen.com/books/references/welding-processes/principles-and-thermo-mechanical-model-of-friction-stir-welding#b57 http://www.intechopen.com/books/references/welding-processes/principles-and-thermo-mechanical-model-of-friction-stir-welding#b57 on the prediction of peak temperatures in tool of different profiles in friction stir welding using improved heat generation models issn: 2180-1053 vol. 10 no.1 january – june 2017 57 pala, s. and phanirajb, m.p. (2015). determination of heat partition between tool and workpiece during fsw of ss304 using 3d cfd modeling. journal of materials processing technology 222, 280–286 p. a. colegrove, h. r. shercliff. cfd modelling of the friction stir welding of thick plate 7449 aluminium alloy. sci. technol. weld. joining 11 (4) (2006), 429–441. roy, g. g., nandan, r. and debroy, t (2006). dimensionless correlation to estimate peak temperature during friction stir welding. scitechnol weld join; 11:606–8. russell, m. j. and shercliff, h. r. (1999).1st int. symp. on friction stir welding (thousand oaks, california, usa), schneider, j. a. (2007). temperature distribution and resulting metal flow. in: mishra rs, mahoney mw, editors. friction stir welding and processing. materials park, oh (usa): asm international, 71-110. sheppard t. and wright. d. (1979). determination of flow stress. part 1 constitutive equation for aluminum alloys at elevated temperatures, metals technology, 6, 215–223. sheppard t. and jackson. a. (1997). constitutive equations for use in prediction of flow stress during extrusion of aluminium alloys, materials science and technology, 13(3), 203– 209. su h., wu c. and chen. m. (2013). analysis of material flow and heat transfer in friction stir welding of aluminium alloys. china weld (engl ed). 22, 6–10. schmidt, h. hattel, j. and wert. j. (2004). an analytical model for the heat generation in friction stir welding. modelling and simulation in materials science and engineering, 12, 143– 157. thomas, w. m., nicholas, e. d., needham, j. c., murch temple-smith m. g., p. and dawes, c. j. (1991). friction-stir butt welding, gb patent no. 9125978.8, international patent application no. pct/ gb92/02203, ulysse. p. (2002). three-dimensional modeling of the friction stir-welding process. int’l journal of machine tools and manufacture 42, 1549-1557. uyyuru, r. k. and kallas. s. v. (2006).numerical analysis of friction stir welding process. journal of materials engineering and performance, 15, 505–518. wang, h. colegrove, p. a. and dos santos. j. f (2013). numerical investigation of the tool contact condition during friction stir welding of aerospace aluminium alloy. computational materials science, 71, 101–108. journal of mechanical engineering and technology 58 issn: 2180-1053 vol. 10 no.1 january – june 2018 nomenclature a area, m2 f axial force/n fs shoulder heat generation ratio from shoulder fps probe side heat generation ratio from probe side fpt probe tip heat generation ratio from probe tip hprobe tool probe height/mm hp tool probe height, m q heat generation, w qs heat generation from the shoulder side, w qp heat generation from the probe side, w qt heat generation from the tip, w r shoulder tool shoulder radius, m rp tool probe radius, m qtotal total heat generation/w τ friction shear stress, pa p contact pressure, pa σ contact pressure, pa μ friction coefficient ω tool angular rotation speed, rad/s δ contact state variable r position along tool radius, m v tool tool speed of ωr, ms−1 slip  slip rate at interface, m/s θ angle, deg z dimension along rotation axis, m α tool shoulder cone angle, deg τy yield shear stress, pa σy yield yield stress, pa issn: 2180-1053 vol. 4 no. 1 january-june 2012 advances in plasma arc welding: a review 35 advances in plasma arc welding: a review kondapalli siva prasad1*, chalamalasetti srinivasa rao2, damera nageswara rao3 1assistant professor, department of mechanical engineering, anil neerukonda institute of technology & sciences , visakhapatnam, india 2associate professor, department of mechanical engineering, au college of engineering, andhra university, visakhapatnam, india 3vice chancellor, centurion university of technology & management, odisha, india email: 1kspanits@gmail.com abstract the nature of welding in the aeronautical industry is characterized by low unit production, high unit cost, extreme reliability and severe service conditions. these characteristics point towards more expensive and more concentrated heat sources such as plasma arc, laser beam and electron beam welding as the processes of choice for welding of critical components. among various precision welding processes, plasma arc welding has gained importance in small and medium scale industries manufacturing bellows , diaphragms etc because of less expensive and easy to operate. this paper reviews the works on plasma arc welding and associated phenomena such as micro plasma arc welding, variable polarity plasma arc welding and keyhole plasma arc welding. the review covers works carried out by various researchers on various metals using different modes of plasma arc. keywords: plasma arc welding, micro plasma, variable polarity, keyhole. 1.0 introduction welding is almost as old as the processing of metals by humans. for most of history, it has been regarded as an obscure art or a crude construction technique. until the end of the 19th century, sections of metal were joined together by a heating and hammering process called forge welding. new discoveries and the availability of electric current in the nineteenth century pushed the development of modern welding issn: 2180-1053 vol. 4 no. 1 january-june 2012 journal of mechanical engineering and technology 36 with an ever-accelerating rate. welding processes are classified by the intensity of the heat source [1]. the penetration measured as the ratio of width of the weld cross section increases dramatically with the intensity of the heat source. this makes the welding process more efficient and allows for higher welding speeds. the more efficient process requires less heat input for the same joint, resulting in a stronger weld. a smaller heat source moving at a faster speed also implies a much reduced dwell time at any particular point. if the dwell time is too short, the process cannot be manually controlled and must be automated. the minimum dwell time that can still be controlled manually corresponds to arc welding (approximately 0.3 seconds). heat sources more intense than arcs have stronger dwell times: therefore they must be automated. welding processes with a more concentrated heat source create a smaller heat affected zone (haz) and lower post weld distortions. however the capital cost of the equipment is roughly proportional to the intensity of the heat source. today, a variety of different welding processes are available, such that welding is extensively used as a fabrication process for joining materials in a wide range of compositions, part shapes and sizes. many types of manufacturing industries make use of a wide variety of welding processes:  aircraft and aerospace industries e.g. wings and fuselages.  shipbuilding and marine industries e.g. panels for decks and superstructures.  land transportation / automotive industries.  oil and petrochemicals industries e.g. off shore production platforms and pipelines.  domestic e.g. white goods and metal furniture. the nature of welding in the aeronautical industry is characterized by low unit production, high unit cost, extreme reliability and severe service conditions. these characteristics point towards the more expensive and more concentrated heat sources such as plasma arc, laser beam and electron beam welding as the processes of choice for welding of critical components. welds are replacing rivets in a variety of components in both military and commercial airplanes to improve both cost and structural integrity. diffusion, laser and electron beam welding are preferred in commercial issn: 2180-1053 vol. 4 no. 1 january-june 2012 advances in plasma arc welding: a review 37 aircraft, while electron beam welding is continually gaining ground for the joining of titanium alloys in military airplanes. in large commercial airplanes laser beam welds are posed to replace rivets in large parts of the fuse large. some new processes developed for the space industry also show promise for the aeronautics industry. these include friction stir welding and variable polarity plasma arc welding, which are already being used for critical applications in rockets. one process that does not gained wide spread application is the diffusion welding of aluminum alloys. 2.0 milestones in plasma arc welding the detailed milestones related to paw was given below [2]. 1973-1975: plasma process understanding characterisation and keyhole stability conditions and in parallel first preliminary applications. 1981-1986: plasma process understanding molten pool movement and industrial applications in pressure vessel manufacturing, first nasa development in vppa of aluminium and trials in orbital welding. 1998-1999: observations of the keyhole and industrial equipment of the plasma welding of aluminium (basics of the nasa in 1984) 2002-2007: modeling of the plasma and some plasma adaptations to comply new applications. 3.0 review of various plasma arc welding processes a thorough review was carried out on plasma arc welding, micro plasma arc welding, variable polarity plasma arc welding and keyhole plasma arc welding process by various researchers and presented in the following paragraphs. 3.1 plasma arc welding (paw) kimiyuki nishiguchi et.al [3] investigated factors which predominate series arcing in plasma arc welding. the results reveals that when the nozzle end of the plasma torch is covered with oxide film , the cathods spot of series arc is easily formed with the help of oxide film, resulting issn: 2180-1053 vol. 4 no. 1 january-june 2012 journal of mechanical engineering and technology 38 in the great reduction of the current capacity. k.tsuchiya et.al [4] carried out preliminary research for further development of some plasma arc welding methods for thick plate above 10mm.large plasma torch and the control equipment designed to be proof against up to 1000a with straight polarity connection have been fabricated. in plasma arc welding of 16mm thick mild steel plates, weld beads were produced as burn through or incomplete penetration beads. when the plates were backed up with copper plates, unstable plasma arc and sometimes series arcing occurred and resulted in a defective bead. kunio narita [5] investigated the effect of different welding parameters of plasma arc welding process on the shape of welds and consistency of defects in the flat, vertical and overhead positions of mild steel pipes of thickness 6.4mm and outer diameter 406.4mm. v.i.astakhin et.al [6] developed plasma arc generators (plasmatrons) operating without replacement of the rapidly wearing parts in less than 200hrs which are simple in service and provide good protection of the weld zone and stabilization of the plasma arc. katsunori inoue et.al [7] presented the method of measurement for the penetration and the control circuit which adjusts adaptively the pulsed current duration depending on the results of penetration quality with on-line measurement. welding experiment and theoretical analysis indicate that the system can ensure uniform penetration quality under the disturbing condition. t. s. baker [8] reported tensile, fracture toughness and fatigue crack propagation (fcp) data for a plasma arc weld (paw) in 4mm thick ti-6al-4v alloy sheet. in addition, fcp data is reported for a weld in 9.6mm thickti6a1-4v alloy produced by a paw root weld and tig filler runs. t. ishida [9] investigated the interfacial microstructures and intermetallic compounds produced by plasma arc butt fusion welding of aluminium to mild steel. experiments were carried out on 5mm thick mild steel and aluminium plates. an intermetallic compound alloy layer formed at the interface region between mild steel and aluminium was determined using quantitative metallography and the mechanism of the intermetallic layer formation and growth was elucidated. s.c.tam et.al [10] studied the process of mechanized plasma arc butt welding of thin gauge mild steel sheets both theoretically and experimentally. the transient temperature distribution has been computed using an analytical model due to rosenthal. the results were compared with those generated by finite element analysis using the commercial package pafec. john w,mckelliget [11-12] developed a mathematical model to predict the velocity, temperature and electromagnetic fields inside an issn: 2180-1053 vol. 4 no. 1 january-june 2012 advances in plasma arc welding: a review 39 inductively coupled plasma torch , as well as the motion and thermal histories of particle injected into the torch. it is demonstrated that high particle feed rates which are vital for industrial scale materials processing applications have an adverse effect on the degree of particle melting. russell g. keanini [13-14] presented a three dimensional finite element model of the plasma arc welding process. the model allows calculation of the weld pools approximate capillary and solid -liquid phase boundaries; the weld pools three dimensional flow and temperature fields and solid phase temperature distribution. the results reveal that flow in vertical cross sections is dominated by a large jet driven vortex, competition between surface tension and jet shear produces a stagnation region near the top of the pool, flow in horizontal planes are largely determined by the plate’s motion and buoyancy is a secondary driving force within the plasma arc weld pool. v. n. startsev [15] investigated interaction between laser radiation and the plasma of a welding arc. the equations of continuity, momentum and energy of viscous flow and the equations of electric current flow and laser radiation transfer were employed. ph. bertrand et.al [16] a set of pyrometers was developed and applied for surface temperature monitoring in thermal plasma processing: a 1-spot monochromatic, a 1-spot multiwavelength and a bi-dimensional monochromatic system. measurements of solidand liquid-phase temperature were carried out for plasma-arc waste treatment and welding. d.k.zhang et.al [17] studied the influence of welding current, arc voltage, welding speed, wire feed rate and magnitude of ion gas flow on front melting width, back melting width and weld reinforcement of alternating current plasma arc welding process using artificial neural networkback propagation algorithm. he used lf6 aluminum alloy of size 300 x 80 x 3mm. orthogonalising design matrix was used to perform the experiments. sheng-chai chi et.al [18] develop an intelligent decision support system for plasma arc welding based on fuzzy radial basis function (rbf) neural network. the system solved problems relating to time-consuming of learning in back-propagation neural network, fluctuation of the values of parameters during welding, and fuzzy linguistic-term judgment for welding quality. based on the results obtained from the taguchi experiments, the developed fuzzy neural network can be trained to establish a quality prediction system for plasma arc welding. g. ravichandran [19] carried out thermal analysis of molten pool formation and solidification for keyhole welding using plasma arc welding has been done using finite element method. yaowen wang et.al [20] addressed the problems involved in the automatic monitoring of the weld quality produced by plasmaarc keyhole welding. the acoustic signal of plasma arc welding was issn: 2180-1053 vol. 4 no. 1 january-june 2012 journal of mechanical engineering and technology 40 acquired by using a condenser microphone at high speed and analyzed with the aid of computers. it is shown that the overall ac power of the acoustic analysis, especially the low frequency part (0±100 hz) of the acoustic signal power spectra, greatly varies with the variation of the statuses of the weld pool. takeshi kawachi [21] presented an application of numerical calculation of thermal plasma for the computational analysis of the anode used in atmospheric pressure. yaowen wang et.al [22] acquired weld voltage and current simultaneously at high speed and investigated with the aid of computers and reported that the overall ac power of the arc signals, especially the low frequency part (0-100hz) of the arc signal power spectra, varies greatly with the variation of the status of the weld pool. b. b. nayak et.al [23] reported that microhardness is found to increase significantly in arc plasma melted tungsten carbide. an attempt has been made to understand the reason behind the enhancement in microhardness. g shanmugav elayutham et.al [24] evaluated the electrothermal efficiency of a dc arc plasma torch and temperature and thermal conductivity of plasma jet in the torch. the effect of nitrogen in combination with argon as plasma gas on the above properties were investigated. casper van der eijk et.al [25] presented the results form welding experiments of niti to niti, stainless steel and hastelloy c276. the welds were characterized by dsc, optical microscopy and field emission sem. the investigation of the niti-niti welds show that there is no compositional variation of the material through the weld. the mechanical properties were however significantly deteriorated after welding. the investigation of the dissimilar welds shows that the mixed zone of these welds contains a number of brittle phases, deteriorating the quality of the weld. micrograph of the niti-niti plasma weld is shown in figure-1. 37 figure 1 optical micrograph of the niti-niti plasma weld [25] w. lu et. al [26] adopted a power module to cut off the main arc current periodically for a very short period of time to acquire accurate information for monitoring the weld pool surface intrinsic characteristic of the non-transferred arc and eliminate the influence of the transferred arc in a normal paw process. pavel kotalik [27] developed a model to analyse the flow of argon plasma inside and outside the discharge chamber of the cascaded plasma torch. woeishyan lee et.al [28] uses a split-hopkinson pressure bar to investigate the effects of strain rate in the range of 103 s_1 to 8x 103 s_1 and welding current mode upon the dynamic impact behavior of plasma arc-welded (paw) 304l stainless steel (ss) weldments. reported the strain rate and the welding current mode have a significant influence upon the dynamic impact behavior and microstructure evolution of 304l ss weldments. the study concerns the paw butt welding of cold-rolled 304l ss plates of 9-mm thickness a. abdellah el-hadj et. al [29] presented the data in two parts. in the first part of, an appropriate inflow turbulent boundary condition is chosen. then, a comparison is made between two turbulence models for a plasma jet discharged into air atmosphere. the plasma jet gas phase flow is predicted with the standard k−ε model and the rng model of turbulence. particles behavior is modeled using stochastic particles trajectories. a validation of the plasma jet model is made by comparison with experimental data. the second part is concerned with the effect of the substrate movement on the gas flow field. this is performed in order to simulate a realistic coatings process where a relative movement between the torch and the substrate always exists. three substrate velocities have been used and it is found that the flow fields are affected only very near the substrate wall. t. matsumoto et.al [30] measured surface tension and the density of 304 stainless steels with the sulfur contents of 10, 100 and 250 ppm under low pressure arc plasma conditions in the temperature range of 1823–2073 k. the measurements were carried out by the sessile drop method and a (la0.9ca0.1)cro3 substrate was used. no significant influence of the plasma was observed on the surface tension and its temperature coefficient. jingguo geet et.al [31] developed a plasma arc welding (paw) seam tracking system, which senses the molten and the seam in a frame using a vision sensor, and then detects the seam deviation to adjust the work piece motion adaptively to the seam position sensed by the vision sensor. proposed a novel molten pool area image-processing algorithm based on machine vision. the algorithm processes each image at a speed of 20 frames/s in real-time to extract three feature variables to get the seam deviation and it was proved experimentally that the algorithm is very fast and effective. kai cheng et.al [32] compared the characteristics of laminar and turbulent argon thermal plasma jets issuing into ambient air. the combined-diffusion-coefficient method and the turbulenceenhanced combined-diffusion-coefficient method were employed to treat the diffusion of ambient air into the laminar and turbulent argon plasma jets, respectively. j. mirapeix et.al [33-36] presented an optimized technique for real-time spectral analysis of thermal plasmas, with application in the monitoring and defect detection of industrial welding processes, particularly arc-welding. the calculation of the plasma electronic temperature by means of a sub-pixel algorithm permits on-line quality assessment of the welds, allowing the figure 1 optical micrograph of the niti-niti plasma weld [25] w. lu et.al [26] adopted a power module to cut off the main arc current periodically for a very short period of time to acquire accurate information for monitoring the weld pool surface intrinsic characteristic of the non-transferred arc and eliminate the influence of the transferred arc in a normal paw process. pavel kotalik [27] developed a model issn: 2180-1053 vol. 4 no. 1 january-june 2012 advances in plasma arc welding: a review 41 to analyse the flow of argon plasma inside and outside the discharge chamber of the cascaded plasma torch. woei-shyan lee et.al [28] uses a split-hopkinson pressure bar to investigate the effects of strain rate in the range of 103 s_1 to 8x 103 s_1 and welding current mode upon the dynamic impact behavior of plasma arc-welded (paw) 304l stainless steel (ss) weldments. reported the strain rate and the welding current mode have a significant influence upon the dynamic impact behavior and microstructure evolution of 304l ss weldments. the study concerns the paw butt welding of cold-rolled 304l ss plates of 9-mm thickness a. abdellah el-hadj et.al [29] presented the data in two parts. in the first part of, an appropriate inflow turbulent boundary condition is chosen. then, a comparison is made between two turbulence models for a plasma jet discharged into air atmosphere. the plasma jet gas phase flow is predicted with the standard k−ε model and the rng model of turbulence. particles behavior is modeled using stochastic particles trajectories. a validation of the plasma jet model is made by comparison with experimental data. the second part is concerned with the effect of the substrate movement on the gas flow field. this is performed in order to simulate a realistic coatings process where a relative movement between the torch and the substrate always exists. three substrate velocities have been used and it is found that the flow fields are affected only very near the substrate wall. t. matsumoto et.al [30] measured surface tension and the density of 304 stainless steels with the sulfur contents of 10, 100 and 250 ppm under low pressure arc plasma conditions in the temperature range of 1823–2073 k. the measurements were carried out by the sessile drop method and a (la0.9ca0.1)cro3 substrate was used. no significant influence of the plasma was observed on the surface tension and its temperature coefficient. jingguo geet et.al [31] developed a plasma arc welding (paw) seam tracking system, which senses the molten and the seam in a frame using a vision sensor, and then detects the seam deviation to adjust the work piece motion adaptively to the seam position sensed by the vision sensor. proposed a novel molten pool area image-processing algorithm based on machine vision. the algorithm processes each image at a speed of 20 frames/s in real-time to extract three feature variables to get the seam deviation and it was proved experimentally that the algorithm is very fast and effective. kai cheng et.al [32] compared the characteristics of laminar and turbulent argon thermal plasma jets issuing into ambient air. the combined-diffusion-coefficient method and the turbulence-enhanced combined-diffusion-coefficient method were employed to treat the diffusion of ambient air into the laminar and turbulent argon plasma jets, respectively. issn: 2180-1053 vol. 4 no. 1 january-june 2012 journal of mechanical engineering and technology 42 j. mirapeix et.al [33-36] presented an optimized technique for real-time spectral analysis of thermal plasmas, with application in the monitoring and defect detection of industrial welding processes, particularly arcwelding. the calculation of the plasma electronic temperature by means of a sub-pixel algorithm permits on-line quality assessment of the welds, allowing the detection of common defects to be found in the welding seam, such as oxidation due to insufficient shielding gas flux or lack of penetration caused by current fluctuations of the welding power source. the proposed technique was successfully checked in a real-time arc-welding monitoring system and experimental results of stainless-steel welds were also reported. presented a novel system which allows arc-welding defect detection and classification. proposed a new approach that allows automatic weld defect detection and classification based in the combined use of principal component analysis (pca) and an artificial neural network (ann). the plasma spectra captured from the welding process is processed with pca, which reduces the processing complexity, by performing a data compression in the spectral dimension. the designed ann, after the selection of a proper data training set, allows automatic detection of weld defects. the proposed technique was successfully checked. arcweld tests on stainless steel are reported, showing a good correlation between the ann outputs and the classical interpretation of the electronic temperature profile. developed a new plasma spectroscopy analysis technique based on the generation of synthetic spectra by means of optimization processes. the technique was developed for its application in arc-welding quality assurance. the new approach was checked through several experimental tests, yielding results in reasonably good agreement with the ones offered by the traditional spectroscopic analysis technique. v. rajamani et.al [37] conducted experimental measurements and computational analysis of heat transfer in atmospheric pressure, mid temperature range (1200 to 1600 k) plasma flow over an aluminum cylinder. a heat transfer problem is computationally modeled by using available experimental measurements of temperature rise in the cylinder to determine the degree of ionization in the plasma flow. r. bini et.al [38] investigated the influence of two nozzle geometries and three process parameters (arc current, arc length and plasma sheath gas flow rate) on the energy distribution for an argon transferred arc. measurements were reported for a straight bore cylindrical and for a convergent nozzle, with arc currents of 100 a and 200 a and electrode gaps of 10 mm and 20 mm. the results obtained from this study show that the shape of the cathode torch nozzle has an important influence on arc behaviour and on the energy distribution between the different issn: 2180-1053 vol. 4 no. 1 january-june 2012 advances in plasma arc welding: a review 43 system components. a convergent nozzle results in higher arc voltages, and consequently, in higher powers being generated in the discharge for the same applied arc current, when compared to the case of a straight bore nozzle. a. urena et.al [39] reported the optimum welding conditions (welding intensity and travel speed) for butt joints of 2205 duplex stainless steel sheets of 3mm and 4mm using plasma-arc welding (paw). minimum net energy input for proper operative and metallurgical weldabilities were studied using two different welding modes: the melt-in or conduction mode and the keyhole mode. the influence of the welding parameter for each mode on the dimensions and shape of the welds and on their ferrite contents was investigated. a. dudek et.al [40] proposed a research method for diagnostics and determination of temperature and shape of plasma arc used for surface treatment of 40cr4 steel with tio2 coating. the surface of samples, previously coated with ceramic coating was remelted with plasma arc. for investigations of arc shape the high-resolution modern visible light camera and thermovision camera was used. the temperature distribution in plasma arc with percentage quantity of temperature fields was determined. the arc limiting profiles with isotherms was also determined. lei yu-cheng et.al [41] adopted plasma arc welding to join sicp/al composite with titanium as alloying filler material. microstructure of the weld was characterized by an optical microscope. the results show that the harmful needle-like phase a14c3 is completely eliminated in the weld of sicp/ai metal matrix composite (mmc) by in-situ weldalloying/plasma arc welding with titanium as the alloying element. the wetting property between reinforced phase and a1 matrix is improved, a stable weld puddle is gotten and a novel composite-material welded joint reinforced by tin, aln and tic is produced. tensile-strength and malleability of the welded joints were improved effectively because of the use of titanium. microstructure of sicp/ai metal matrix composite and tensile specimen is shown in figure-2 39 between reinforced phase and a1 matrix is improved, a stable weld puddle is gotten and a novel composite-material welded joint reinforced by tin, aln and tic is produced. tensilestrength and malleability of the welded joints were improved effectively because of the use of titanium. microstructure of sicp/ai metal matrix composite and tensile specimen is shown in figure-2 figure 2 microstructure of sicp/al metal matrix composite [41] y. f. hsiao et.al [42] studied the optimal parameters process of plasma arc welding (paw) by the taguchi method with grey relational analysis was studied. sus316 stainless steel plate of thickness 4mm and the test piece of 250mm x 220mm without groove was used for welding. torch stand-off, welding current, welding speed, and plasma gas flow rate (argon) were chosen as input variables and welding groove root penetration, welding groove width, front-side undercut were measured as output parameters. emel taban et.al [43] welded en 1.4410 (uns s32750) superduplex stainless steel (sdss) of thickness of 6.5 mm using paw process with different heat inputs. mechanical properties, impact toughness testing at subzero temperatures starting from -20 °c down to -60 °c was carried out while fractographs were examined by scanning electron microscopy (sem). maco and micro graphs of paw welded sample is shown in figure-3. figure 3 macro and microphotographs of weld 1:(a) macro (b) base metal (c) haz and(d) weld metal [42] lei yu-cheng et.al [44] investigated the effect of ti-al on microstructures and mechanical properties of sicp/al mmc joints produced by plasma arc in-situ weld-alloying, in which argon-nitrogen mixture was used as plasma gases and ti-al alloy as filling composite. the results show that the formation of needle-like harmful phase al4c3 is effectively prevented in the weld by in-situ weld-alloying/plasma arc welding with ti-al alloy sheet filler whose titanium content is more than 20%. the fluidity of molten pool is improved, and stable molten pool is gained for the addition of the ti-al alloy. the mechanical properties of welded joint are effectively enhanced by the compact-grain structure and the new reinforced composites figure 2 microstructure of sicp/al metal matrix composite [41] issn: 2180-1053 vol. 4 no. 1 january-june 2012 journal of mechanical engineering and technology 44 y. f. hsiao et.al [42] studied the optimal parameters process of plasma arc welding (paw) by the taguchi method with grey relational analysis was studied. sus316 stainless steel plate of thickness 4mm and the test piece of 250mm x 220mm without groove was used for welding. torch stand-off, welding current, welding speed, and plasma gas flow rate (argon) were chosen as input variables and welding groove root penetration, welding groove width, front-side undercut were measured as output parameters. emel taban et.al [43] welded en 1.4410 (uns s32750) superduplex stainless steel (sdss) of thickness of 6.5 mm using paw process with different heat inputs. mechanical properties, impact toughness testing at subzero temperatures starting from -20°c down to -60°c was carried out while fractographs were examined by scanning electron microscopy (sem). maco and micro graphs of paw welded sample is shown in figure-3. 39 between reinforced phase and a1 matrix is improved, a stable weld puddle is gotten and a novel composite-material welded joint reinforced by tin, aln and tic is produced. tensilestrength and malleability of the welded joints were improved effectively because of the use of titanium. microstructure of sicp/ai metal matrix composite and tensile specimen is shown in figure-2 figure 2 microstructure of sicp/al metal matrix composite [41] y. f. hsiao et.al [42] studied the optimal parameters process of plasma arc welding (paw) by the taguchi method with grey relational analysis was studied. sus316 stainless steel plate of thickness 4mm and the test piece of 250mm x 220mm without groove was used for welding. torch stand-off, welding current, welding speed, and plasma gas flow rate (argon) were chosen as input variables and welding groove root penetration, welding groove width, front-side undercut were measured as output parameters. emel taban et.al [43] welded en 1.4410 (uns s32750) superduplex stainless steel (sdss) of thickness of 6.5 mm using paw process with different heat inputs. mechanical properties, impact toughness testing at subzero temperatures starting from -20 °c down to -60 °c was carried out while fractographs were examined by scanning electron microscopy (sem). maco and micro graphs of paw welded sample is shown in figure-3. figure 3 macro and microphotographs of weld 1:(a) macro (b) base metal (c) haz and(d) weld metal [42] lei yu-cheng et.al [44] investigated the effect of ti-al on microstructures and mechanical properties of sicp/al mmc joints produced by plasma arc in-situ weld-alloying, in which argon-nitrogen mixture was used as plasma gases and ti-al alloy as filling composite. the results show that the formation of needle-like harmful phase al4c3 is effectively prevented in the weld by in-situ weld-alloying/plasma arc welding with ti-al alloy sheet filler whose titanium content is more than 20%. the fluidity of molten pool is improved, and stable molten pool is gained for the addition of the ti-al alloy. the mechanical properties of welded joint are effectively enhanced by the compact-grain structure and the new reinforced composites figure 3 macro and microphotographs of weld 1:(a) macro (b) base metal (c) haz and(d) weld metal [42] lei yu-cheng et.al [44] investigated the effect of ti-al on microstructures and mechanical properties of sicp/al mmc joints produced by plasma arc in-situ weld-alloying, in which argon-nitrogen mixture was used as plasma gases and ti-al alloy as filling composite. the results show that the formation of needle-like harmful phase al4c3 is effectively prevented in the weld by in-situ weld-alloying/plasma arc welding with ti-al alloy sheet filler whose titanium content is more than 20%. the fluidity of molten pool is improved, and stable molten pool is gained for the addition of the ti-al alloy. the mechanical properties of welded joint are effectively enhanced by the compact-grain structure and the new reinforced composites such as al3ti, tin, aln and tic welded joint. the test results of mechanical property show that the maximum tensile strength of welded joint gained by adding ti-60al issn: 2180-1053 vol. 4 no. 1 january-june 2012 advances in plasma arc welding: a review 45 alloy is up to 235 mpa. the factors influencing the tensile strength were also investigated. tashiro shinichi et.al [45] reported numerical simulation result of heat source property of ac plasma arc welding. the results reveal that the maximum electrode temperature gradually increases during ep peak current due to heating caused mainly by electron condensation to the electrode surface and decreases during en peak current due to cooling caused mainly by thermionic electron emission. the maximum electrode temperature becomes higher with small en ratio. the electrode temperature increases especially near the electrode tip surface during ep peak current. r. sanchez-tovar et.al [46] analysed the corrosion under flowing conditions of four kinds of aisi 316l materials welded by the micro-plasma arc welding technique in different media: a basic (libr) and an acidic (h3po4) solution by means of polarization measurements. corrosion parameters revealed that, among the materials welded with backing gas, the alloy which presented better corrosion behavior was the one welded without filler alloy. however, this kind of material could undergo several corrosion problems if a crack is formed or due to an inadequate joint penetration. 3.2 micro plasma arc welding (mpaw) n.m.voropai et.al [47] developed a technique of pulsed micro plasma butt welding the shells of asbestors-metal gaskets made of aluminium of thickness 0.2-0.3mm. pulsed micro plasma welding results in steady burning of the arc on low current and in the destruction of the oxide film on the joined metal. in this method, argon of a purity of not less than 99.8% is used as plasma forming gas and helium of a purity of not less than 99.5% as proctective gas. a.s.sepokurov et.al [48] developed a a-1255 source which permits the smooth control of welding current over the 1-10a range, so that it can be used for welding thin components and for hardfacing small components.w. luo et.al [49] analysed the surface microstructure and the anodic polarization curves in a 1 n h2so4 solution of a 0cr19ni9 steel submerged arc welded joint before and after surface melting using a 4-a micro-plasma arc. the results showed that both the heat-affected zone and the weld metal of the as-welded joint had a lower corrosion resistance than the as-received parent material, while the arc melted joint had a significantly increased corrosion resistance. this increase in corrosion resistance is attributed to a rapid solidification of the melted layer. rapid solidification of the melted layer refines its microstructure, decreases microsegregation and inhibits the precipitation of chromium carbides at the grain boundaries. issn: 2180-1053 vol. 4 no. 1 january-june 2012 journal of mechanical engineering and technology 46 f. karimzadeh et.al [50-51] investigated the effect of micro plasma arc welding (mpaw) process parameters on grain growth and porosity distribution of thin sheet ti6al4v alloy weldment. the mpaw procedure was performed at different current, welding speed and flow rates of shielding & plasma gas. square-butt welding in a single pass, using direct current and straight polarity (dcen) was selected for the welding process. the titanium alloy studied in the present experiment is a thin sheet of ti6al4v alloy with a thickness of 0.8mm. examined the effect of epitaxial growth on microstructure of ti–6al– 4v alloy weldment by artificial neural networks (anns). the mpaw procedure was performed at different currents, welding speeds and flow rates of shielding & plasma gas. microstructural characterizations were studied by optical and scanning electron microscopy (sem) and the image was shown in figure-4. finally, an artificial neural network was developed to predict grain size of fusion zone (fz) at different currents and welding speeds. the results showed that a coarse primary β phase develops in the fusion zone as a result of epitaxial nucleation on coarsened β grains near the heat affected zone (nhaz) which grow competitively into the molten weld pool. based on anns analyses, a map of current and welding speed for α→β transformation in the haz can be constructed. for a lower energy input, grain growth of β phase in the haz could be restricted by α phase. the presence of small quantities of this phase at high peak temperatures in the weld cycle is sufficient to prevent the grain growth of β phase in haz and fz. microstructure of heat affected zone and fusion zone are shown in figure-5. 41 haz can be constructed. for a lower energy input, grain growth of β phase in the haz could be restricted by α phase. the presence of small quantities of this phase at high peak temperatures in the weld cycle is sufficient to prevent the grain growth of β phase in haz and fz. microstructure of heat affected zone and fusion zone are shown in figure-5. figure 4 microstructure of annealed base metal [50]. figure 5 microstructure of: (a) heat affected zone and (b) fusion zone (the arrow shows the grain boundary of primary coarse grains) [50]. pei-quan xu et.al [52] developed a model to simulate the electromagnetic phenomena and fluid field in plasma arc occurring during the low-current micro plasma arc welding process. the effects of the nozzle neck-in and welding current of micro-plasma arc on the arc electromagnetic field distribution were discussed. three types of micro plasma arc, namely, needle plasma arc, columnar plasma arc and opening model plasma arc are founded by experiment. based on the unified model, a thorough investigation of the low-current micro plasma arc characteristics during the micro-paw process was conducted. it was found that the process parameters have significant effects on the micro plasma arc and the distributions of current density and electromagnet force distribution. experiments were conducted on fine sheet of 0.1-mm thickness. kondapalli siva prasad et.al [53-59] developed empirical relations to predict weld geometry parameters of aluminium alloy using statistical techniques. studied the weld quality characteristics of ss304l stainless steel sheets and developed mathematical models to predict the weld pool geometry, grain size and hardness of ss304l stainless steel sheets. developed empirical relations to predict grain size, hardness and ultimate tensile strength of mpaw welded inconel 625 sheets. 3.3 variable polarity plasma arc welding (vppaw) b. zheng et.al [60] investigated the front image sensing of the keyhole puddle invariable polarity plasma arc welding of aluminum alloys to extract the characteristically geometrical size of the keyhole and to realize the feedback control for weld formation in the welding process. the results show that the approach of composite arc light filtering with narrow-band spectrum can be applied to take the image of the keyhole puddle of aluminum alloys. zhonghua liu et.al [61] investigated double side image sensing of the keyhole puddle in the variable polarity plasma arc welding (vppaw) of aluminium alloys to extract the characteristic geometrical size of the keyhole and to realize feedback controlling for weld formation in the welding process. an artificial neural network is applied to establish the figure 4 microstructure of annealed base metal [50]. 41 haz can be constructed. for a lower energy input, grain growth of β phase in the haz could be restricted by α phase. the presence of small quantities of this phase at high peak temperatures in the weld cycle is sufficient to prevent the grain growth of β phase in haz and fz. microstructure of heat affected zone and fusion zone are shown in figure-5. figure 4 microstructure of annealed base metal [50]. figure 5 microstructure of: (a) heat affected zone and (b) fusion zone (the arrow shows the grain boundary of primary coarse grains) [50]. pei-quan xu et.al [52] developed a model to simulate the electromagnetic phenomena and fluid field in plasma arc occurring during the low-current micro plasma arc welding process. the effects of the nozzle neck-in and welding current of micro-plasma arc on the arc electromagnetic field distribution were discussed. three types of micro plasma arc, namely, needle plasma arc, columnar plasma arc and opening model plasma arc are founded by experiment. based on the unified model, a thorough investigation of the low-current micro plasma arc characteristics during the micro-paw process was conducted. it was found that the process parameters have significant effects on the micro plasma arc and the distributions of current density and electromagnet force distribution. experiments were conducted on fine sheet of 0.1-mm thickness. kondapalli siva prasad et.al [53-59] developed empirical relations to predict weld geometry parameters of aluminium alloy using statistical techniques. studied the weld quality characteristics of ss304l stainless steel sheets and developed mathematical models to predict the weld pool geometry, grain size and hardness of ss304l stainless steel sheets. developed empirical relations to predict grain size, hardness and ultimate tensile strength of mpaw welded inconel 625 sheets. 3.3 variable polarity plasma arc welding (vppaw) b. zheng et.al [60] investigated the front image sensing of the keyhole puddle invariable polarity plasma arc welding of aluminum alloys to extract the characteristically geometrical size of the keyhole and to realize the feedback control for weld formation in the welding process. the results show that the approach of composite arc light filtering with narrow-band spectrum can be applied to take the image of the keyhole puddle of aluminum alloys. zhonghua liu et.al [61] investigated double side image sensing of the keyhole puddle in the variable polarity plasma arc welding (vppaw) of aluminium alloys to extract the characteristic geometrical size of the keyhole and to realize feedback controlling for weld formation in the welding process. an artificial neural network is applied to establish the figure 5 microstructure of: (a) heat affected zone and (b) fusion zone (the arrow shows the grain boundary of primary coarse grains) [50]. issn: 2180-1053 vol. 4 no. 1 january-june 2012 advances in plasma arc welding: a review 47 pei-quan xu et.al [52] developed a model to simulate the electromagnetic phenomena and fluid field in plasma arc occurring during the lowcurrent micro plasma arc welding process. the effects of the nozzle neckin and welding current of micro-plasma arc on the arc electromagnetic field distribution were discussed. three types of micro plasma arc, namely, needle plasma arc, columnar plasma arc and opening model plasma arc are founded by experiment. based on the unified model, a thorough investigation of the low-current micro plasma arc characteristics during the micro-paw process was conducted. it was found that the process parameters have significant effects on the micro plasma arc and the distributions of current density and electromagnet force distribution. experiments were conducted on fine sheet of 0.1mm thickness. kondapalli siva prasad et.al [53-59] developed empirical relations to predict weld geometry parameters of aluminium alloy using statistical techniques. studied the weld quality characteristics of ss304l stainless steel sheets and developed mathematical models to predict the weld pool geometry, grain size and hardness of ss304l stainless steel sheets. developed empirical relations to predict grain size, hardness and ultimate tensile strength of mpaw welded inconel 625 sheets. 3.3 variable polarity plasma arc welding (vppaw) b. zheng et.al [60] investigated the front image sensing of the keyhole puddle invariable polarity plasma arc welding of aluminum alloys to extract the characteristically geometrical size of the keyhole and to realize the feedback control for weld formation in the welding process. the results show that the approach of composite arc light filtering with narrow-band spectrum can be applied to take the image of the keyhole puddle of aluminum alloys. zhonghua liu et.al [61] investigated double side image sensing of the keyhole puddle in the variable polarity plasma arc welding (vppaw) of aluminium alloys to extract the characteristic geometrical size of the keyhole and to realize feedback controlling for weld formation in the welding process. an artificial neural network is applied to establish the steady model for predicting the geometrical size of the back keyhole puddle. the model can be used to control the stability of the keyhole and the weld formation. the experiments reveal that neural networks are capable of modeling parameters of the vppaw process. s. ganguly et.al [62] measured the residual stresses in a 12-mm-thick vppa-welded aluminum 2024-t351 alloy plate have been measured using neutron diffraction. the stresses were then remeasured by a combination of neutron and synchrotron x-ray diffraction after the plate had been reduced in thickness (or, skimmed) to 7 mm by machining both sides of the weld, mimicking issn: 2180-1053 vol. 4 no. 1 january-june 2012 journal of mechanical engineering and technology 48 the likely manufacturing operation, should such welds be used in aerospace structures. emad saad et.al [63] investigated relationships between the acoustic signal and the modes of the welding pool such as no-keyhole (meltin), keyhole and cutting in variable polarity plasma arc welding (vppaw). a 5256 aluminum alloy plate with the dimensions of 76.2mm×178.0mm×4.8mm is used to study the effect of the welding pool mode on the signature of the acoustic signal in the real vppaw process. welch power spectral density (psd) estimate is used for preprocessing the sound data. a neural network (nn) is used to distinguish the keyhole mode from the cutting mode. the results show that the keyhole mode can be distinguished from the cutting mode under the experiment conditions. h.x. wang et.al [64] developed threedimensional transient governing equations based on conservation laws of energy, momentum and mass. these equations described physical phenomena of convection in weld pool and heat transfer in workpiece during variable polarity vertical-up plasma arc welding process. boundary conditions for the developed governing equations were given. welding energy input for variable polarity verticalup plasma arc welding process was quantitatively expressed. free surface deformation of the keyhole molten pool was coupled into calculation. effect of wire filling on the geometry of molten pool and weld reinforcement was considered in the simulation. correlations of temperature and thermophysical properties for aluminum alloy 2219 were quantitatively established. a control volume based finite difference method was used to solve the discrete governing equations. b. zheng et.al [65] presented a technique for real-time, closed-loop feedback control of weld penetration based on the front image signal of the weld pool in variable polarity plasma arc welding (vppaw) of aluminum alloys. the results achieved show a feasible way to implement the real-time weld formation control into the aluminum vppaw. 3.4 keyhole plasma arc welding y. f. hsu et.al [66] proposed a two dimensional, quasi stationary finite element numerical model to study the fluid flow and the heat transfer phenomena which occur during constant travel speed, keyhole plasma arc welding of metal plates. a newton raphson iteration procedure was developed in this model to accurately identify the solid-liquid interface location during welding. the finite element method was applied to the study of a typical keyhole welding processes of an issn: 2180-1053 vol. 4 no. 1 january-june 2012 advances in plasma arc welding: a review 49 aisi 304 stainless steel plate. the results show that the method can be used to predict the shape of the welding pool as a function of welding parameters and that the widths of both the fusion zone and the heat affected zone decrease as the welding speed increases while the power required for welding increases with an increase in welding speed. alkhalidy nehad [67] carried out a computational method to predict the transient development of the weld pool and the temperature histories in the workpiece during plasma arc welding of aisi 304. the influence of the different parameters such as, welding speed, keyhole diameter, thermal properties, distance between the nodes in the region of interest and the time steps on the welding quality and numerical consideration are studied. a stable results is obtained by the use of control volume enthalpy technique without under realaxation or over relaxiation method. the mechanism which affected the transient shape of the welding pool are, the velocity of the welding torch, the keyhole diameter and thermal properties. jukka martihainen [68] investigated the possibilities and the technological conditions for welding structural steels, especially high strength steels, reproducibly and with high quality. the investigation comprises butt welding with an i-groove in the flat, horizontal – vertical and vertical positions and root welding of thick plates in the flat position. it was shown that mechanized plasma keyhole welding is a very useful method for structural steels. breton e.losch et.al [69] developed a methodology for analyzing the reflected plasma arc in keyhole arc welding (kaw) in real time. images obtained from a high speed imaging system were stored and analysed off-line. fuzzy logic was examined as a means of accomplishing the image analysis, due to its ability to mimic the qualitative approach taken by human operators of welding equipment. y. m. zhang et.al [70-72] monitored the keyhole and the weld pools simultaneously from the back side of the workpiece as shown in figure-6. bead-on-plate and butt-joint welds were made on 3 mm thick stainless steel (304) plates in the flat position. it was found that once the keyhole is established, the width of the keyhole does not change with an increasing welding current and a decreasing welding speed. issn: 2180-1053 vol. 4 no. 1 january-june 2012 journal of mechanical engineering and technology 50 43 the welding pool are, the velocity of the welding torch, the keyhole diameter and thermal properties. jukka martihainen [68] investigated the possibilities and the technological conditions for welding structural steels, especially high strength steels, reproducibly and with high quality. the investigation comprises butt welding with an i-groove in the flat, horizontal –vertical and vertical positions and root welding of thick plates in the flat position. it was shown that mechanized plasma keyhole welding is a very useful method for structural steels. breton e.losch et.al [69] developed a methodology for analyzing the reflected plasma arc in keyhole arc welding (kaw) in real time. images obtained from a high speed imaging system were stored and analysed off-line. fuzzy logic was examined as a means of accomplishing the image analysis, due to its ability to mimic the qualitative approach taken by human operators of welding equipment. y. m. zhang et.al [70-72] monitored the keyhole and the weld pools simultaneously from the back side of the workpiece as shown in figure-6. bead-on-plate and butt-joint welds were made on 3 mm thick stainless steel (304) plates in the flat position. it was found that once the keyhole is established, the width of the keyhole does not change with an increasing welding current and a decreasing welding speed. figure 6 simultaneous imaging of weld pool and keyhole [72] w. lu et.al [73] addressed the development of a nonlinear model based interval model control system for the quasi-keyhole arc welding process, a novel arc welding process which has advantages over the laser welding process and conventional arc welding processes. the structure of the nonlinear model chosen was proposed based on an analysis of the quasikeyhole process to be controlled. experiments suggested that the nonlinear model-based interval model control has advantages over the linear model-based interval model control and the linear model-based adaptive predictive control in terms of fluctuations of the control signal and output and of the system response speed. experiments have also verified the effectiveness of the developed system as a robust control which requires no readjustment and can function properly when fluctuations/variations in manufacturing conditions, and thus the process dynamics, change, vary or fluctuate.c s wu et.al [74-76] developed a numerical model for examining and simulating the dynamic keyhole establishment process, which will be a key in developing an effective control technology for double sided arc welding (dsaw). the model is used to determine the geometrical shape of the keyhole and the weld pool, and the temperature distribution in the work piece. the dsaw experiments show that the predicted weld cross-section is in agreement with the measured one. john zhang et.al [77] developed an adaptive interval model control system for keyhole plasma arc welding process. the developed system identifies the process parameters online, converts the identification results to the intervals in zhang and kovacevic’s algorithm and uses a prefilter to eliminate the effect of the keyhole process’ fluctuation on the control system. experiments comparing the adaptive interval model control system with its nonadaptive counterpart have been figure 6 simultaneous imaging of weld pool and keyhole [72] w. lu et.al [73] addressed the development of a nonlinear model based interval model control system for the quasi-keyhole arc welding process, a novel arc welding process which has advantages over the laser welding process and conventional arc welding processes. the structure of the nonlinear model chosen was proposed based on an analysis of the quasi-keyhole process to be controlled. experiments suggested that the nonlinear model-based interval model control has advantages over the linear model-based interval model control and the linear modelbased adaptive predictive control in terms of fluctuations of the control signal and output and of the system response speed. experiments have also verified the effectiveness of the developed system as a robust control which requires no readjustment and can function properly when fluctuations/variations in manufacturing conditions, and thus the process dynamics, change, vary or fluctuate. c s wu et.al [74-76] developed a numerical model for examining and simulating the dynamic keyhole establishment process, which will be a key in developing an effective control technology for double sided arc welding (dsaw). the model is used to determine the geometrical shape of the keyhole and the weld pool, and the temperature distribution in the work piece. the dsaw experiments show that the predicted weld cross-section is in agreement with the measured one. john zhang et.al [77] developed an adaptive interval model control system for keyhole plasma arc welding process. the developed system identifies the process parameters online, converts the identification results to the intervals in zhang and kovacevic’s algorithm and uses a prefilter to eliminate the effect of the keyhole process’ fluctuation on the control system. experiments comparing the adaptive interval model control system with its nonadaptive counterpart have been conducted to verify the effectiveness of the former in achieving fast response speed when the manufacturing conditions or the set-point vary. dong honggang et.al [78] developed a three-dimensional steady numerical model for the heat transfer and fluid flow in plasma arc issn: 2180-1053 vol. 4 no. 1 january-june 2012 advances in plasma arc welding: a review 51 (pa)–gas tungsten arc (gta) double-sided keyhole welding process. the model considers the surface tension gradient, electromagnetic force and buoyancy force. a double-v-shaped keyhole geometry as shown in figure-7 is proposed and its characteristic parameters are derived from the images and cross-section of weld bead. based on the numerical model, the distributions of the fluid flow and temperature field are calculated. a comparison of cross-section of the weld bead with the experimental result shows that the numerical model’s accuracy is reasonable. 44 conducted to verify the effectiveness of the former in achieving fast response speed when the manufacturing conditions or the set-point vary. dong honggang .et.al [78] developed a three-dimensional steady numerical model for the heat transfer and fluid flow in plasma arc (pa)–gas tungsten arc (gta) double-sided keyhole welding process. the model considers the surface tension gradient, electromagnetic force and buoyancy force. a double-v-shaped keyhole geometry as shown in figure-7 is proposed and its characteristic parameters are derived from the images and cross-section of weld bead. based on the numerical model, the distributions of the fluid flow and temperature field are calculated. a comparison of cross-section of the weld bead with the experimental result shows that the numerical model’s accuracy is reasonable. figure 7 experimental cross-section in keyhole double-sided arc welding [78] e.o. correa et.al [79] investigated the weldability of three different iron-based powder metal alloys (pure fe, fe–ni and fe–p–ni alloys) using the keyhole pulsed plasma arc process (paw). the work undertaken included the effect of pulsed welding parameters on the microstructure and mechanical characteristics of the welded joints. microstructural examination results revealed that for the pure fe and fe–ni alloys, the fusion-welded zone was free of porosity and cracks. however, the fe–p–ni powder metal alloy with high level of phosphorus content (0.25 wt%) and 7mm thickness specimen presented solidification cracks and tunneling failure as a result of high shrinking stress due to the higher volume of molten metal and faster cooling rates. jia chuan-bao et.al [80] developed a new sensing and control system for monitoring and controlling the keyhole condition during plasma arc welding (paw). through sensing and processing the efflux plasma voltage signals, the quantitative relationship among the welding current, efflux plasma voltage and backside weld width of the weld were established. 4 future prospects of plasma arc research and development activities for future applications, especially regarding metals which so far cannot be welded well is going on. in the near future, the different possibilities of constricting the arc mechanically or magnetically, together with the use of variable electric currents, will allow practically total command of the arc. this means that it will become possible to weld and cut all metals. the manufacturing of the instruments essential for scientific research and the application of new inventions, create ever more difficult problems in the joining of materials. meanwhile, the solutions based on welding become more and more numerous and are contributing significantly to the progress of science and industry. the plasma welding technique offers the advantage of ensuring a good continuity of the material between the parts to be joined. figure 7 experimental cross-section in keyhole double-sided arc welding [78] e.o. correa et.al [79] investigated the weldability of three different ironbased powder metal alloys (pure fe, fe–ni and fe–p–ni alloys) using the keyhole pulsed plasma arc process (paw). the work undertaken included the effect of pulsed welding parameters on the microstructure and mechanical characteristics of the welded joints. microstructural examination results revealed that for the pure fe and fe–ni alloys, the fusion-welded zone was free of porosity and cracks. however, the fe–p–ni powder metal alloy with high level of phosphorus content (0.25 wt%) and 7mm thickness specimen presented solidification cracks and tunneling failure as a result of high shrinking stress due to the higher volume of molten metal and faster cooling rates. jia chuan-bao et.al [80] developed a new sensing and control system for monitoring and controlling the keyhole condition during plasma arc welding (paw). through sensing and processing the efflux plasma voltage signals, the quantitative relationship among the welding current, efflux plasma voltage and backside weld width of the weld were established. 4.0 future prospects of plasma arc research and development activities for future applications, especially regarding metals which so far cannot be welded well is going on. in the near future, the different possibilities of constricting the arc mechanically or magnetically, together with the use of variable electric currents, will allow practically total command of the arc. this means that issn: 2180-1053 vol. 4 no. 1 january-june 2012 journal of mechanical engineering and technology 52 it will become possible to weld and cut all metals. the manufacturing of the instruments essential for scientific research and the application of new inventions, create ever more difficult problems in the joining of materials. meanwhile, the solutions based on welding become more and more numerous and are contributing significantly to the progress of science and industry. the plasma welding technique offers the advantage of ensuring a good continuity of the material between the parts to be joined. 5.0 conclusions this paper reviews the development of plasma arc welding and associated phenomena.  it was understood from the earlier works that most of the works in plasma arc welding and associated phenomena are towards modeling of plasma arc, temperature & heat transformation and process parameter optimization to get the desired weld quality. very few works had happened especially in micro plasma arc welding of thin sheets.  in most of the works welding current, arc voltage, welding speed, magnitude of ionic gas , torch stand of are considered for predicting and optimizing the weld bead geometry. however certain other factors like peak current, background current, pulse, pulse width, purging gas magnitude are not concentrated much especially while welding thin sheets using micro plasma arc welding.  from the literature review it was understood that many works were carried out on stainless steels, aluminum, nickel based alloys, titanium etc. on can try for welding dissimilar materials and new materials using plasma arc welding and associated phenomena. also one can try for grain refinement techniques such as pulsed current welding, magnetic arc oscillation etc to obtain better weld quality characteristics. issn: 2180-1053 vol. 4 no. 1 january-june 2012 advances in plasma arc welding: a review 53 6.0 references [1] patricio f. mendez, thomas w.eagar, welding processes for aeronatics, advanced materials & processes, may 2001. [2] jean marie fortain, plasma welding evolution & challenges, air liquide ctas, welding and cutting research center, 95315 cergy pontoise france. [3] kimiyuki nishiguchi and kazumasa tashiro,(1970) , series arcing in plasma arc welding, japan welding society, pp 59-69. [4] k.tsuchiya, k.kishimoto, t.matsunaga, e.nakano,(1973), plasma arc welding for thick plate (part-1), japan welding society, pp 554-566. [5] kunio narita, (1975), plasma arc welding of pipelines: a study to optimize welding conditions for horizontal fixed joints of mild steel pipes, int. j. pres. ves. & piping 3: pp 233-266. [6] v. i. astakhin, a. s. bychkov, v. a. konovalov and r. m. meirov, (1977), plasma arc welding of aluminium alloy cryogenic piping, japan welding society, no.2, pp 26-28. [7] katsunori inoue and de-fu he,(1984), penetartion–self-asaptive free frequency pulsed plasma arc welding process controlled with photocell sensor, transactions of jwri, vol. 13,no. 1:pp 7-11. [8] t. s. baker, (1985), fatigue crack propagation and fracture toughness of plasma arc welded ti-6al-4v alloy, royal aircraft establishment, technical report no: 85066. [9] t. ishida, (1987), interfacial phenomena of plasma arc welding of mild steel and aluminum, journal of materials science, 22 : pp 10611066. [10] s. c. tam, l. e. lindgren and l. j. yang, (1989), computer simulation of tempaerture fields in mechanized plasma arc welding, journal of mechanical working technology,19:pp 23-33. [11] john w, mckelliget,(1990), numerical computation of coupled heat transfer, fluid flow and electromagnetism: the inductivity coupled plasma torch, advanced computational methods in heat transfer, vol.3. [12] john w, mckelliget, (1992), a mathematical model of the spheroidization of porous agglomerate particles in thermal plasma torches, thermal plasma applications in materials and metallurgical processing”, pp. 337-349. [13] russell g. keanini, (1993), simuataion of weld pool flow and capillary interface shapes associated with the plasma arc welding process, finite elemets in analysis and design 15:pp 83-92. issn: 2180-1053 vol. 4 no. 1 january-june 2012 journal of mechanical engineering and technology 54 [14] russell g. keanini and boris rubinsky, (1993), three dimensional simulation of the plasma arc welding process, int. j. heat mass transfer, vol.36, no.13, pp 3283-3298. [15] v. n. startsev, (1999), numerical analysis of the effect of laser radiation on the plasma of a welding arc, journal of engineering physics and thermophysics, vol.72, no.5,pp 920-926. [16] ph. bertrand, m. ignatiev, g. flamant, i. smurov, (2000), pyrometry applications in thermal plasma processing, vacuum 56:pp 71-76. [17] d. k. zhang and j. t. niu (2000) application of artificial neural network modeling to plasma arc welding of aluminum alloys, journal of advanced metallurgical sciences, vol. 13 no. 1 pp 194-200 . [18] sheng-chai chi and li-chang hsu, (2001), a fuzzy radial basis function neural network for predicting multiple quality characteristics of plasma arc welding, ieee, pp 2807-2812. [19] g. ravichandran, (2001), solidification behavior in plasma arc welding, sadana, vol. 26, parts i & ii, pp 199-211. [20] yaowen wang, pensheng zhaob, (2001), noncontact acoustic analysis monitoring of plasma arc welding, international journal of pressure vessels and piping 78:pp 43-47. [21] takeshi kawachi, (2002), the computational analysis of the anode using numerical method of thermal plasma, fifth world congress on computational mechanics, vienna, austria. [22] yaowen wang, qiang chen, (2002), on-line quality monitoring in plasma arc welding, journal of materials processing technology, 120:pp 270-274. [23] b. b. nayak, (2003), enhancement in the microhardness of arc plasma melted tungsten carbide, journal of materials science 38, pp 2717 – 2721. [24] g shanmugavelayutham and v selvarajan, (2003), electrothermal efficiency, temperature and thermal conductivity of plasma jet in a dc plasma spray torch, pramana journal of physics, indian academy of sciences vol. 61, no. 6, pp 1109–1119. [25] casper van der eijk, hans fostervoll, zuhair k. sallom and odd m. akselsen, (2003), plasma welding of niti to niti, stainless steel and hastelloy c276, asm materials solutions 2003 conference, pittsburgh, pennsylvania, usa. [26] w. lu, y m zhang and john emmerson, (2004), sensing of weld pool surface using non-transferred plasma charge sensor, meas. sci. technol. 15 :pp 991–999. issn: 2180-1053 vol. 4 no. 1 january-june 2012 advances in plasma arc welding: a review 55 [27] pavel kotalik , (2004), modelling of an argon plasma flow, czechoslovak journal of physics, vol. 55 , no. 2,pp 173-188. [28] woei-shyan lee, chi-feng lin, chen-yang liu, and chin-wei cheng, (2004), the effects of strain rate and welding current mode on the dynamic impact behavior of plasma-arc-welded 304l stainless steel weldments, metallurgical and materials transactions, volume 35a, pp 1501-1515. [29] a. abdellah el-hadj and n. ait-messaoudene, (2005), comparison between two turbulence models and analysis of the effect of the substrate movement on the flow field of a plasma jet, plasma chemistry and plasma processing, vol. 25, no. 6,pp 699-722. [30] t. matsumoto, t. misono, h. fujii, k. nogi, (2005), surface tension of molten stainless steels under plasma conditions, journal of materials science 40: pp 2197 – 2200. [31] jingguo ge , zhengqiang zhu, defu he, ligong chen, (2005), a vision-based algorithm for seam detection in a paw process for largediameter stainless steel pipes, int j adv manuf technol 26: pp 1006– 1011. [32] kai cheng, xi chen, wenxia pan, (2006), comparison of laminar and turbulent thermal plasma jet characteristics— a modeling study, plasma chem plasma process 26:pp 211–235. [33] j. mirapeix, a. cobo , o.m. conde , c. jauregui, j.m. lopez-higuera, (2006), real-time arc welding defect detection technique by means of plasma spectrum optical analysis, ndt&e international 39 : pp 356– 360. [34] j. mirapeix, p.b. garcia-allende, a. cobo, o.m. conde, j.m. lopezhiguera, (2007), real-time arc-welding defect detection and classification with principal component analysis and artificial neural networks, ndt&e international 40 : pp 315–323. [35] j. mirapeix, a. cobo, d. a. gonzalez and j. m. lopez-higuera,(2007), plasma spectroscopy analysis technique based on optimization algorithms and spectral synthesis for arc-welding quality assurance, optical express, vol. 15, no. 4, pp 1884 -1897. [36] j. mirapeix, a. cobo, d. a. gonzález and j. m.lopez-higuera,(2007), plasma spectroscopy analysis technique based on optimization algorithms and spectral synthesis for arc-welding quality assurance, optics express, vol. 15, no. 4 ,pp 184-1897. [37] v. rajamani, r. anand, g.s. reddy, j.a. sekhar, and m.a. jog, (2006), heat-transfer enhancement using weakly ionized, atmospheric pressure plasma in metallurgical applications, metallurgical and materials transactions b, volume 37b, pp 565-570. issn: 2180-1053 vol. 4 no. 1 january-june 2012 journal of mechanical engineering and technology 56 [38] r. bini , m. monno , m. i. boulos, (2007), effect of cathode nozzle geometry and process parameters on the energy distribution for an argon transferred arc, plasma chem plasma process 27:pp 359–380. [39] a. urena , e. otero, m.v. utrilla, c.j. munez, (2007), weldability of a 2205 duplex stainless steel using plasma arc welding, journal of materials processing technology 182 : pp 624–631. [40] a. dudek, z. nitkiewicz, (2007), diagnostics of plasma arc during the process of remelting of surface layer in 40cr4 steel, international scientific journal, volume 28,issue 6,pp 369-372. [41] lei yu-cheng, yuan wei-jin,chen xi-zhang, zhu fei cheng xiao-nong, (2007), in-situ weld-alloying plasma arc welding of sicp/ a1 mmc, trans. nonferrous met. soc. china 17: pp 3133 17. [42] y. f. hsiao, y. s. tarng, and w. j. huang, (2008), optimization of plasma arc welding parameters by using the taguchi method with the grey relational analysis, journal of materials and manufacturing processes, 23: pp 51–58. [43] emel taban, (2008), toughness and microstructural analysis of superduplex stainless steel joined by plasma arc welding, j mater sci 43:pp 4309–4315. [44] lei yu-cheng, zhang zhen, nie jia-jun, chen xi-zhang, (2008), effect of ti-al on microstructures and mechanical properties of plasma arc in-situ welded joint of sicp/al mmcs, transacations of nonferrous metals society of china, 18:pp 809-813. [45] tashiro shinichi, miyata minoru, tanaka manabu, (2009), numerical simulation of ac plasma arc welding, volume 27, no.2,pp 1s-3s. [46] r. sanchez-tovar, m.t. montanes, j. garcia-anton, (2010), effect of different micro-plasma arc welding (mpaw) processes on the corrosion of aisi 316l ss tubes in libr and h3po4 solutions under flowing conditions, journal of corrosion science 52 :pp 1508–1519. [47] n. m. voropai, v. v. shcherbak and a. a. grigorev,(1971), pulsed microplasma welding of thin aluminum gaskets, equipment manufacturing technology, no.11. pp 19. [48] a. s. sepokurov, g. i. sergatskii and a. p. alikin, (1971), use of microplasma welding in component construction, japan welding society, no.11, pp 20. [49] w. luo, (2002), effect of micro-plasma arc melting on the corrosion resistance of a 0cr19ni9 stainless steel saw joint, materials letters 55: pp 290–295. issn: 2180-1053 vol. 4 no. 1 january-june 2012 advances in plasma arc welding: a review 57 [50] karimzadeh, f, salehi, m , saatchi. a and meratian. m, (2005), ‘effect of microplasma arc welding process parameters on grain growth and porosity distribution of thin sheet ti6al4v alloyweldment’, materials and manufacturing processes, 20: 2, pp 205 — 219. [51] f. karimzadeh , a. ebnonnasir , a. foroughi , (2006), artificial neural network modeling for evaluating of epitaxial growth of ti6al4v weldment, materials science and engineering a 432 :pp 184–190. [52] pei-quan xu , shun yao, jian-ping he, chun-wei ma & jiang-wei ren, (2009), numerical analysis for effect of process parameters of lowcurrent micro-paw on constricted arc, int j adv manuf technol 44:pp 255–264. [53] kondapalli siva prasad, srinivasa rao. ch, nageswara rao.d, (2010), prediction of weld quality in plasma arc welding using statistical approach, aijstpme, 3(4),pp.29-35. [54] kondapalli siva prasad , srinivasa rao. ch, nageswara rao.d, (2011), prediction of weld bead geometry in plasma arc welding using factorial design approach, journal of minerals & m a t e r i a l s characterization & engineering, vol. 10, no.10, pp.875-886. [55] kondapalli siva prasad, ch. srinivasa rao, d. nageswara rao, (2011), prediction of weld pool geometry in pulsed current micro plasma arc welding of ss304l stainless steel sheets, international transaction journal of engineering, management & applied sciences & technologies, volume 2 no.3,p.325-336. [56] kondapalli siva prasad, ch. srinivasa rao, d. nageswara rao, (2011), a study on weld quality characteristics of pulsed current micro plasma arc welding of ss304l sheets, international transaction journal of engineering, management & applied sciences & technologies, volume 2 no.3,pp.437-446. [57] kondapalli siva prasad, ch. srinivasa rao, d. nageswara rao, (2011), establishing empirical relations to predict grain size and hardness of pulsed current micro plasma arc welded ss 304l sheets, american transactions on engineering & applied sciences, volume 1 no.1, pp. 57-74. [58] kondapalli siva prasad, ch. srinivasa rao, d. nageswara rao, (2011), optimizing pulsed current micro plasma arc welding parameters to maximize ultimate tensile strength of inconel625 nickel alloy using response surface method, international journal of engineering, science and technology, vol. 3, no. 6, , pp. 226-236. [59] kondapalli siva prasad, ch. srinivasa rao, d. nageswara rao, (2011), establishing empirical relationships to predict grain size and hardness of pulsed current micro plasma arc welded inconel 625 issn: 2180-1053 vol. 4 no. 1 january-june 2012 journal of mechanical engineering and technology 58 sheets, journal of materials & metallurgical engineering volume 1 issue 3, pp.1-10. [60] b. zheng , h.j. wang, q.l. wang, (1998), front image sensing of the keyhole puddle in the variable polarity paw of aluminum alloys, journal of materials processing technology 83 : pp 286–298. [61] zhonghua liu, qilong wang, bing zheng, (2001), process control based on double side image sensing of the keyhole in vppa welding, journal of materials processing technology115:373-379. [62] s. ganguly, m. e. fitzpatrick, and l. edwards, (2006), use of neutron and synchrotron x-ray diffraction for evaluation of residual stresses in a 2024-t351 aluminum alloy variable-polarity plasma-arc weld, metallurgical and materials transactions a, volume 37a, pp 411-420. [63] emad saad, huijun wang, radovan kovacevic, (2006), classification of molten pool modes in variable polarity plasma arc welding based on acoustic signature, journal of materials processing technology 174: pp 127–136. [64] h. x. wang, y.h. wei, c. l. yang ,(2007), numerical simulation of variable polarity vertical-up plasma arc welding process, computational materials science 38: pp 571–587. [65] b. zheng, h. j. wang, q. i. wang and r. kovacevic, (2009),control for weld penetration invppaw of aluminum alloys using the front weld pool image signal, welding research supplement, pp 363 to 371. [66] y. f. hsu and b. rubinsky, (1988), two dimensional heat transfer study on the keyhole plasma arc welding process, int.j.heat mass transfer, vol.31, no.7, pp1409-1421. [67] al-khalidy nehad, (1995), enthalpy technique for solution of stefan problems: application to the keyhole plasma arc welding process involving moving heat source, int.comm.heat mass transfer, vol.22, no.6, pp.779-790. [68] jukka martihainen, (1995), conditions for achieving high quality welds in the plasma keyhole welding of structural steels, journal of materials processing technology 52 :pp 68-75. [69] breton e. losch and yuming zhang, (2002), fuzzy classification of plasma reflection for keyhole sensing and control, proceedings of the 2002 ieee international conference on control applications, glasgow, scotland, uk, pp 31-36. [70] y. m. zhang & b. zhang, (1999), observation of the keyhole during plasma arc welding, aws welding research supplement, pp 53-58. [71] y m zhang and y ma , (2001), stochastic modelling of plasma reflection during keyhole arc welding, meas. sci. technol. 12 : pp 1964–1975. issn: 2180-1053 vol. 4 no. 1 january-june 2012 advances in plasma arc welding: a review 59 [72] y.m. zhang, y.c. liu, (2003), modeling and control of quasi-keyhole arc welding process control engineering practice 11 : pp 1401–1411. [73] w. lu, y. m. zhang, w. -y. lin, (2004), nonlinear interval model control of quasi-keyhole arc welding process, automatica 40: pp 805 – 813. [74] c. s. wu , j s sun and y m zhang, (2004), numerical simulation of dynamic development of keyhole in double-sided arc welding, journal of modelling simul. mater. sci. eng. 12 :pp 423–442. [75] c. s. wu, h. g. wang, and y. m. zhang, (2006), a new heat source model for keyhole plasma arc welding in fem analysis of the temperature profile, welding journal, pp 284 to 291. [76] c.s. wu, q.x. hu, j.q. gao, (2009), an adaptive heat source model for finite-element analysis of keyhole plasma arc welding, computational materials science 46: pp 167–172. [77] john zhang and bruce l. walcott, (2006), adaptive interval model control of arc welding process,ieee transactions on control systems technology, vol. 14, no. 6, pp 1127-1134. [78] dong honggang, gao hongming and wu lin, (2006), numerical simulation of fluid flow and temperature field in keyhole doublesided arc welding process on stainless steel, int. j. numer. meth. engg 65:pp 1673–1687. [79] e.o. correa, s.c. costa, j. n. santos,(2008), weldability of iron-based powder & metal materials using pulsed plasma arc welding process, journal of materials processing technology 1 9 8:pp 323–329. [80] jia chuan-bao, wu chuan-song, zhang yu-ming, (2009), sensing controlled pulse key-holing condition in plasma arc welding, transactions of nonferrous metals society of china,19: pp341-346. issn: 2180-1053 vol. 4 no. 1 january-june 2012 implementation of nonlinear finite element using object–oriented design patterns 61 implementation of nonlinear finite element using object–oriented design patterns a. yaghoobi faculty of mechanical engineering, university of tabriz, tabriz, iran email: amin.yaghoobi@gmail.com abstract this paper concerns with the aspects of the object–oriented programming used to develop a nonlinear finite element for the analysis of plates based on reissner–mindlin theory. to study the shear locking problem in thin plates which occurs in the case of using full integration method, three kinds of finite elements namely bilinear, serendipity and lagrange with full, reduced and selective reduced integration methods, are used. by implementing three design patterns of model–analysis separation, model– ui separation and modular analyzer in the code, the reusability and the extendibility of the program in adding new elements with different number of nodes and integration methods have been increased. keywords: nonlinear finite element, object–oriented programming, design pattern, reissner–mindlin plates 1.0 introduction most existing finite element software packages are developed in procedural–based programming languages. these packages are normally monolithic and difficult for a programmer to maintain and extend, though some of them are quite rich in terms of functionality. extensibility usually requires access to, and manipulation of internal data structures. due to the lack of data encapsulation and protection, small changes in one piece of code can ripple through the rest of the software system. for example, to add a new element to an existing procedural–based finite element analysis software package, the programmer is usually required to specify, at the element level, the memory pointers to global arrays. exposing such unnecessary implementation details increases the software complexity and adds a burden to a programmer. issn: 2180-1053 vol. 4 no. 1 january-june 2012 journal of mechanical engineering and technology 62 object oriented (oo) design principles and programming techniques can be utilized in finite element analysis programs to support better data encapsulation and to facilitate code reuse. a number of object– oriented finite element program designs and implementations have been presented in the literature over the past decade [1–4]. as shown in mentioned papers, the modularity, reusability and extendibility capacities of object–oriented finite element codes are the major characteristics of the approach. the object–oriented methodology has been most successfully applied to various domains of interest in finite element developments. description becomes easier for algorithms and more natural for basic mathematical equations. thus, the object– oriented paradigm has been shown to be more appropriate for the easy description of complex phenomena. software engineering researchers are developing sets of organizational concepts for designing qualified object–oriented software. these concepts called design patterns. design patterns in software engineering have been proven to offer great benefits. especially as engineering software becomes more object–oriented, the importance of design patterns cannot be underestimated. in this research a set of design patterns for engineering finite element program is implemented using an object–oriented framework in c# to an example. this example is the elastic–plastic analysis of bending plates based on reissner–mindlin plate theory. to study the behavior of these plates, three kinds of finite element, namely bilinear, serendipity and lagrange, are used. to overcome the shear locking problem, three integration methods for each element are used to determine the element stiffness matrix. this example shows by applying the design patterns in object–oriented framework, the usability, extensibility, flexibility and maintainability of the code has been increased. 2.0 the object–oriented programming approach a traditional (non–oo) program can be viewed as a logical procedure that takes input data, processes it and returns the output. the main program is built around simpler procedures or functions. in designing a procedural code, one focuses on how to define the logic rather than how to define the data and its organization. in contrast, an oo program is built around objects which encapsulate both the data and the operations on the data. an object can be viewed as an abstraction which relates variables and methods. issn: 2180-1053 vol. 4 no. 1 january-june 2012 implementation of nonlinear finite element using object–oriented design patterns 63 therefore, the first step in building an oo program is to identify the objects and how they relate to each other. once the objects are identified they can be generalized to a class of objects. a group of objects with the same character is called a class (see figure 1). the software only contains classes. these encapsulate data and data methods. the generic procedures are called methods. the methods represent the behavior of an object and constitute its external interface. 53 software engineering researchers are developing sets of organizational concepts for designing qualified object–oriented software. these concepts called design patterns. design patterns in software engineering have been proven to offer great benefits. especially as engineering software becomes more object–oriented, the importance of design patterns cannot be underestimated. in this research a set of design patterns for engineering finite element program is implemented using an object–oriented framework in c# to an example. this example is the elastic–plastic analysis of bending plates based on reissner–mindlin plate theory. to study the behavior of these plates, three kinds of finite element, namely bilinear, serendipity and lagrange, are used. to overcome the shear locking problem, three integration methods for each element are used to determine the element stiffness matrix. this example shows by applying the design patterns in object–oriented framework, the usability, extensibility, flexibility and maintainability of the code has been increased. 2.0 the object–oriented programming approach a traditional (non–oo) program can be viewed as a logical procedure that takes input data, processes it and returns the output. the main program is built around simpler procedures or functions. in designing a procedural code, one focuses on how to define the logic rather than how to define the data and its organization. in contrast, an oo program is built around objects which encapsulate both the data and the operations on the data. an object can be viewed as an abstraction which relates variables and methods. therefore, the first step in building an oo program is to identify the objects and how they relate to each other. once the objects are identified they can be generalized to a class of objects. a group of objects with the same character is called a class (see figure 1). the software only contains classes. these encapsulate data and data methods. the generic procedures are called methods. the methods represent the behavior of an object and constitute its external interface. element number nodes stiffness matrix givenode compute stiffness matrix …. …. name of the class attributes message methodes figure. 1 class concept. object–oriented programming can be said to have four key concepts: abstraction, encapsulation, inheritance and polymorphism. detailed descriptions of the main concepts of oo programming can be found in many papers [5-7]. here we merely provide reminders. abstraction consists of extracting the most relevant features of the system to be modeled. it provides adequate generalization and eliminates irrelevant details. in oop, abstraction means figure. 1 class concept. object–oriented programming can be said to have four key concepts: abstraction, encapsulation, inheritance and polymorphism. detailed descriptions of the main concepts of oo programming can be found in many papers [5-7]. here we merely provide reminders. abstraction consists of extracting the most relevant features of the system to be modeled. it provides adequate generalization and eliminates irrelevant details. in oop, abstraction means to list the defining characteristics of the classes. it also means to state the public interface of the classes, i.e., how their objects will interact with other objects. the encapsulation concept means hiding the class internal implementation while the class interface is visible. interaction among objects is controlled by the message mechanism. when an object receives a message, it performs an associated method. the implementation details are not known by the client code. this means that information is hidden outside the class and its derived classes. information hiding is very useful, for example, if the class public interface is unaltered, the internal implementation can be changed without affecting how the other classes and application programs access that class. the class information can be specialized using the inheritance principle. subclasses inherit data and methods of their super–classes. in this way, issn: 2180-1053 vol. 4 no. 1 january-june 2012 journal of mechanical engineering and technology 64 it is possible to reuse codes in many applications with consequent reduction in development time and costs. the inheritance principle with c# virtual classes introduces an important generalization feature. pointers to higher level objects of the class hierarchy can represent lower level ones in application programs. this characteristic allows developing type independent code with dynamic binding at runtime. new classes can be added to the hierarchy and the application code will still work with this new type (see figure. 2). 54 to list the defining characteristics of the classes. it also means to state the public interface of the classes, i.e., how their objects will interact with other objects. the encapsulation concept means hiding the class internal implementation while the class interface is visible. interaction among objects is controlled by the message mechanism. when an object receives a message, it performs an associated method. the implementation details are not known by the client code. this means that information is hidden outside the class and its derived classes. information hiding is very useful, for example, if the class public interface is unaltered, the internal implementation can be changed without affecting how the other classes and application programs access that class. the class information can be specialized using the inheritance principle. subclasses inherit data and methods of their super–classes. in this way, it is possible to reuse codes in many applications with consequent reduction in development time and costs. the inheritance principle with c# virtual classes introduces an important generalization feature. pointers to higher level objects of the class hierarchy can represent lower level ones in application programs. this characteristic allows developing type independent code with dynamic binding at runtime. new classes can be added to the hierarchy and the application code will still work with this new type (see figure. 2). figure. 2 uml diagram of the element class. polymorphism means that objects will answer differently for a same message. for instance, the message "+" may mean concatenation and sum, respectively, for string and matrix classes. polymorphism and inheritance allow achieving a fairly generic code that selects the methods to be performed at runtime. 3.0 design patterns since object–oriented programming implemented in finite element method for the first time, numerous approaches have been proposed. the design of an oo finite element program is affected by a number of factors, including software requirements, language features, executing environment, etc, that cause to make some differences between the programs. of course, there are similarities too that reflect consensus among researchers. as this field of research continues to mature, best practices in program design will begin to emerge. it would be useful to capture the key features of these practices in a language–independent and figure. 2 uml diagram of the element class. polymorphism means that objects will answer differently for a same message. for instance, the message "+" may mean concatenation and sum, respectively, for string and matrix classes. polymorphism and inheritance allow achieving a fairly generic code that selects the methods to be performed at runtime. 3.0 design patterns since object–oriented programming implemented in finite element method for the first time, numerous approaches have been proposed. the design of an oo finite element program is affected by a number of factors, including software requirements, language features, executing environment, etc, that cause to make some differences between the programs. of course, there are similarities too that reflect consensus among researchers. as this field of research continues to mature, best practices in program design will begin to emerge. it would be useful to capture the key features of these practices in a language–independent and reusable format. design patterns are a means of achieving this goal. liu et.al. [8] and fenves et.al. [9] explicitly used some of these patterns in their finite element systems. heng and mackie [10] used issn: 2180-1053 vol. 4 no. 1 january-june 2012 implementation of nonlinear finite element using object–oriented design patterns 65 design patterns to identify best practice in object–oriented finite element program design. in this study, three of them will use in a nonlinear finite element method. 3.1. model–analysis separation this pattern decomposes a finite element program into model and analysis subsystems. rucki and miller [11] were among the first to explicitly separate analysis classes from the model subsystem. dubois– pèlerin and pegon [12] believed that a clear distinction between analysis– related classes and those related to the model is vital to implementing a flexible program. 3.2. model–ui separation the model–ui separation pattern separates methods and data related to the user interface (ui) from model classes. most modern finite element systems have integrated graphical user interfaces. in oo finite element programming, a graphical user interface could be implemented by adding ui–related responsibilities to the model classes [13]. in oo finite element programming, the principle of separating graphical classes from model classes was proposed by ju and hosain [14]. 3.3. modular analyzer this pattern decomposes the analysis subsystem into components. marczak [15] decomposed his analysis subsystem along different lines and also added components representing analysis types, integration schemes, and equation solution algorithms. in the next sections object oriented finite element implementation will be illustrated using an example of a plate. first, a review of a plate formulation for finite elements in bending– shear based on the theory of reissner–mindlin plate is carried out. 4.0 reissner–mindlin plate theory in reissner–mindlin plates, normal to the mid–surface (z=0) remains straight but not necessary normal to the mid–surface after deformation [16]. so the effect of shear deformation is considered in this plate unlike kirchhoff formulation. issn: 2180-1053 vol. 4 no. 1 january-june 2012 journal of mechanical engineering and technology 66 4.1. fundamental relation the displacement components at a typical point in a mindlin plate may be represented as: 55 reusable format. design patterns are a means of achieving this goal. liu et al. [8] and fenves et al. [9] explicitly used some of these patterns in their finite element systems. heng and mackie [10] used design patterns to identify best practice in object–oriented finite element program design. in this study, three of them will use in a nonlinear finite element method. 3.1. model–analysis separation this pattern decomposes a finite element program into model and analysis subsystems. rucki and miller [11] were among the first to explicitly separate analysis classes from the model subsystem. dubois–pèlerin and pegon [12] believed that a clear distinction between analysis– related classes and those related to the model is vital to implementing a flexible program. 3.2. model–ui separation the model–ui separation pattern separates methods and data related to the user interface (ui) from model classes. most modern finite element systems have integrated graphical user interfaces. in oo finite element programming, a graphical user interface could be implemented by adding ui–related responsibilities to the model classes [13]. in oo finite element programming, the principle of separating graphical classes from model classes was proposed by ju and hosain [14]. 3.3. modular analyzer this pattern decomposes the analysis subsystem into components. marczak [15] decomposed his analysis subsystem along different lines and also added components representing analysis types, integration schemes, and equation solution algorithms. in the next sections object oriented finite element implementation will be illustrated using an example of a plate. first, a review of a plate formulation for finite elements in bending– shear based on the theory of reissner–mindlin plate is carried out. 4. reissner–mindlin plate theory in reissner–mindlin plates, normal to the mid–surface (z=0) remains straight but not necessary normal to the mid–surface after deformation [16]. so the effect of shear deformation is considered in this plate unlike kirchhoff formulation. 4.1. fundamental relation the displacement components at a typical point in a mindlin plate may be represented as: ),(),,( ),(),,( ),(),,( yxwzyxw yxzzyxv yxzzyxu y x = −= −= θ θ (1) where ),( yxw is normal displacement and ),( yxxθ and ),( yxyθ are the normal rotations of the mid–plane in xz and yz planes, respectively. the rotations xθ and yθ can be expressed in the form: (1) where 55 reusable format. design patterns are a means of achieving this goal. liu et al. [8] and fenves et al. [9] explicitly used some of these patterns in their finite element systems. heng and mackie [10] used design patterns to identify best practice in object–oriented finite element program design. in this study, three of them will use in a nonlinear finite element method. 3.1. model–analysis separation this pattern decomposes a finite element program into model and analysis subsystems. rucki and miller [11] were among the first to explicitly separate analysis classes from the model subsystem. dubois–pèlerin and pegon [12] believed that a clear distinction between analysis– related classes and those related to the model is vital to implementing a flexible program. 3.2. model–ui separation the model–ui separation pattern separates methods and data related to the user interface (ui) from model classes. most modern finite element systems have integrated graphical user interfaces. in oo finite element programming, a graphical user interface could be implemented by adding ui–related responsibilities to the model classes [13]. in oo finite element programming, the principle of separating graphical classes from model classes was proposed by ju and hosain [14]. 3.3. modular analyzer this pattern decomposes the analysis subsystem into components. marczak [15] decomposed his analysis subsystem along different lines and also added components representing analysis types, integration schemes, and equation solution algorithms. in the next sections object oriented finite element implementation will be illustrated using an example of a plate. first, a review of a plate formulation for finite elements in bending– shear based on the theory of reissner–mindlin plate is carried out. 4. reissner–mindlin plate theory in reissner–mindlin plates, normal to the mid–surface (z=0) remains straight but not necessary normal to the mid–surface after deformation [16]. so the effect of shear deformation is considered in this plate unlike kirchhoff formulation. 4.1. fundamental relation the displacement components at a typical point in a mindlin plate may be represented as: ),(),,( ),(),,( ),(),,( yxwzyxw yxzzyxv yxzzyxu y x = −= −= θ θ (1) where ),( yxw is normal displacement and ),( yxxθ and ),( yxyθ are the normal rotations of the mid–plane in xz and yz planes, respectively. the rotations xθ and yθ can be expressed in the form: is normal displacement and 55 reusable format. design patterns are a means of achieving this goal. liu et al. [8] and fenves et al. [9] explicitly used some of these patterns in their finite element systems. heng and mackie [10] used design patterns to identify best practice in object–oriented finite element program design. in this study, three of them will use in a nonlinear finite element method. 3.1. model–analysis separation this pattern decomposes a finite element program into model and analysis subsystems. rucki and miller [11] were among the first to explicitly separate analysis classes from the model subsystem. dubois–pèlerin and pegon [12] believed that a clear distinction between analysis– related classes and those related to the model is vital to implementing a flexible program. 3.2. model–ui separation the model–ui separation pattern separates methods and data related to the user interface (ui) from model classes. most modern finite element systems have integrated graphical user interfaces. in oo finite element programming, a graphical user interface could be implemented by adding ui–related responsibilities to the model classes [13]. in oo finite element programming, the principle of separating graphical classes from model classes was proposed by ju and hosain [14]. 3.3. modular analyzer this pattern decomposes the analysis subsystem into components. marczak [15] decomposed his analysis subsystem along different lines and also added components representing analysis types, integration schemes, and equation solution algorithms. in the next sections object oriented finite element implementation will be illustrated using an example of a plate. first, a review of a plate formulation for finite elements in bending– shear based on the theory of reissner–mindlin plate is carried out. 4. reissner–mindlin plate theory in reissner–mindlin plates, normal to the mid–surface (z=0) remains straight but not necessary normal to the mid–surface after deformation [16]. so the effect of shear deformation is considered in this plate unlike kirchhoff formulation. 4.1. fundamental relation the displacement components at a typical point in a mindlin plate may be represented as: ),(),,( ),(),,( ),(),,( yxwzyxw yxzzyxv yxzzyxu y x = −= −= θ θ (1) where ),( yxw is normal displacement and ),( yxxθ and ),( yxyθ are the normal rotations of the mid–plane in xz and yz planes, respectively. the rotations xθ and yθ can be expressed in the form: and 55 reusable format. design patterns are a means of achieving this goal. liu et al. [8] and fenves et al. [9] explicitly used some of these patterns in their finite element systems. heng and mackie [10] used design patterns to identify best practice in object–oriented finite element program design. in this study, three of them will use in a nonlinear finite element method. 3.1. model–analysis separation this pattern decomposes a finite element program into model and analysis subsystems. rucki and miller [11] were among the first to explicitly separate analysis classes from the model subsystem. dubois–pèlerin and pegon [12] believed that a clear distinction between analysis– related classes and those related to the model is vital to implementing a flexible program. 3.2. model–ui separation the model–ui separation pattern separates methods and data related to the user interface (ui) from model classes. most modern finite element systems have integrated graphical user interfaces. in oo finite element programming, a graphical user interface could be implemented by adding ui–related responsibilities to the model classes [13]. in oo finite element programming, the principle of separating graphical classes from model classes was proposed by ju and hosain [14]. 3.3. modular analyzer this pattern decomposes the analysis subsystem into components. marczak [15] decomposed his analysis subsystem along different lines and also added components representing analysis types, integration schemes, and equation solution algorithms. in the next sections object oriented finite element implementation will be illustrated using an example of a plate. first, a review of a plate formulation for finite elements in bending– shear based on the theory of reissner–mindlin plate is carried out. 4. reissner–mindlin plate theory in reissner–mindlin plates, normal to the mid–surface (z=0) remains straight but not necessary normal to the mid–surface after deformation [16]. so the effect of shear deformation is considered in this plate unlike kirchhoff formulation. 4.1. fundamental relation the displacement components at a typical point in a mindlin plate may be represented as: ),(),,( ),(),,( ),(),,( yxwzyxw yxzzyxv yxzzyxu y x = −= −= θ θ (1) where ),( yxw is normal displacement and ),( yxxθ and ),( yxyθ are the normal rotations of the mid–plane in xz and yz planes, respectively. the rotations xθ and yθ can be expressed in the form: are the normal rotations of the mid–plane in xz and yz planes, respectively. the rotations 55 reusable format. design patterns are a means of achieving this goal. liu et al. [8] and fenves et al. [9] explicitly used some of these patterns in their finite element systems. heng and mackie [10] used design patterns to identify best practice in object–oriented finite element program design. in this study, three of them will use in a nonlinear finite element method. 3.1. model–analysis separation this pattern decomposes a finite element program into model and analysis subsystems. rucki and miller [11] were among the first to explicitly separate analysis classes from the model subsystem. dubois–pèlerin and pegon [12] believed that a clear distinction between analysis– related classes and those related to the model is vital to implementing a flexible program. 3.2. model–ui separation the model–ui separation pattern separates methods and data related to the user interface (ui) from model classes. most modern finite element systems have integrated graphical user interfaces. in oo finite element programming, a graphical user interface could be implemented by adding ui–related responsibilities to the model classes [13]. in oo finite element programming, the principle of separating graphical classes from model classes was proposed by ju and hosain [14]. 3.3. modular analyzer this pattern decomposes the analysis subsystem into components. marczak [15] decomposed his analysis subsystem along different lines and also added components representing analysis types, integration schemes, and equation solution algorithms. in the next sections object oriented finite element implementation will be illustrated using an example of a plate. first, a review of a plate formulation for finite elements in bending– shear based on the theory of reissner–mindlin plate is carried out. 4. reissner–mindlin plate theory in reissner–mindlin plates, normal to the mid–surface (z=0) remains straight but not necessary normal to the mid–surface after deformation [16]. so the effect of shear deformation is considered in this plate unlike kirchhoff formulation. 4.1. fundamental relation the displacement components at a typical point in a mindlin plate may be represented as: ),(),,( ),(),,( ),(),,( yxwzyxw yxzzyxv yxzzyxu y x = −= −= θ θ (1) where ),( yxw is normal displacement and ),( yxxθ and ),( yxyθ are the normal rotations of the mid–plane in xz and yz planes, respectively. the rotations xθ and yθ can be expressed in the form: and 55 reusable format. design patterns are a means of achieving this goal. liu et al. [8] and fenves et al. [9] explicitly used some of these patterns in their finite element systems. heng and mackie [10] used design patterns to identify best practice in object–oriented finite element program design. in this study, three of them will use in a nonlinear finite element method. 3.1. model–analysis separation this pattern decomposes a finite element program into model and analysis subsystems. rucki and miller [11] were among the first to explicitly separate analysis classes from the model subsystem. dubois–pèlerin and pegon [12] believed that a clear distinction between analysis– related classes and those related to the model is vital to implementing a flexible program. 3.2. model–ui separation the model–ui separation pattern separates methods and data related to the user interface (ui) from model classes. most modern finite element systems have integrated graphical user interfaces. in oo finite element programming, a graphical user interface could be implemented by adding ui–related responsibilities to the model classes [13]. in oo finite element programming, the principle of separating graphical classes from model classes was proposed by ju and hosain [14]. 3.3. modular analyzer this pattern decomposes the analysis subsystem into components. marczak [15] decomposed his analysis subsystem along different lines and also added components representing analysis types, integration schemes, and equation solution algorithms. in the next sections object oriented finite element implementation will be illustrated using an example of a plate. first, a review of a plate formulation for finite elements in bending– shear based on the theory of reissner–mindlin plate is carried out. 4. reissner–mindlin plate theory in reissner–mindlin plates, normal to the mid–surface (z=0) remains straight but not necessary normal to the mid–surface after deformation [16]. so the effect of shear deformation is considered in this plate unlike kirchhoff formulation. 4.1. fundamental relation the displacement components at a typical point in a mindlin plate may be represented as: ),(),,( ),(),,( ),(),,( yxwzyxw yxzzyxv yxzzyxu y x = −= −= θ θ (1) where ),( yxw is normal displacement and ),( yxxθ and ),( yxyθ are the normal rotations of the mid–plane in xz and yz planes, respectively. the rotations xθ and yθ can be expressed in the form: can be expressed in the form: 56 yzy xzx y w x w γθ γθ −∂ ∂= −∂ ∂= (2) where x w ∂ ∂ and y w ∂ ∂ are the slopes of the middle surface in the x and y directions and xzγ and yzγ are the shear strains. 4.2. strain–displacement relations for mindlin plate theory, the strain components may be written in terms of the displacements of the middle surface as follows: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ −= ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂ ∂∂ −= ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂ ∂∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = xy y x yx y x xy y x b z xy y x z xvyu yv xu κ κ κ θθ θ θ γ ε ε ε (3) ⎥ ⎦ ⎤ ⎢ ⎣ ⎡ −∂∂ −∂∂ =⎥ ⎦ ⎤ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂+∂∂ =⎥ ⎦ ⎤ ⎢ ⎣ ⎡ = y x yz xz s yw xw ywzv xwzu θ θ γ γ ε (4) where bε is bending strains vector, sε is shear strains vector and xκ , yκ are bending curvatures and xyκ is shear curvature. 4.3. stress– strain relations the moment–curvature relationships and the bending moments are [16]: ]][[][ bbb d εσ = (5) dzz m m m h h xy y x xy y x b ∫− ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = 2 2 ][ σ σ σ σ (6) where h is the plate thickness and ][ bd is the flexural rigidity matrix for an isotropic material which may be expressed as follows: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ − − = 2)1(00 01 01 )1(12 ][ 2 3 ν ν ν ν eh db (7) where e is young’s modulus, ν is poisson’s ratio. the shear force–shear strain relationships and also the shear forces are given as: ]][[][ sss d εσ = (8) (2) where 56 yzy xzx y w x w γθ γθ −∂ ∂= −∂ ∂= (2) where x w ∂ ∂ and y w ∂ ∂ are the slopes of the middle surface in the x and y directions and xzγ and yzγ are the shear strains. 4.2. strain–displacement relations for mindlin plate theory, the strain components may be written in terms of the displacements of the middle surface as follows: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ −= ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂ ∂∂ −= ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂ ∂∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = xy y x yx y x xy y x b z xy y x z xvyu yv xu κ κ κ θθ θ θ γ ε ε ε (3) ⎥ ⎦ ⎤ ⎢ ⎣ ⎡ −∂∂ −∂∂ =⎥ ⎦ ⎤ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂+∂∂ =⎥ ⎦ ⎤ ⎢ ⎣ ⎡ = y x yz xz s yw xw ywzv xwzu θ θ γ γ ε (4) where bε is bending strains vector, sε is shear strains vector and xκ , yκ are bending curvatures and xyκ is shear curvature. 4.3. stress– strain relations the moment–curvature relationships and the bending moments are [16]: ]][[][ bbb d εσ = (5) dzz m m m h h xy y x xy y x b ∫− ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = 2 2 ][ σ σ σ σ (6) where h is the plate thickness and ][ bd is the flexural rigidity matrix for an isotropic material which may be expressed as follows: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ − − = 2)1(00 01 01 )1(12 ][ 2 3 ν ν ν ν eh db (7) where e is young’s modulus, ν is poisson’s ratio. the shear force–shear strain relationships and also the shear forces are given as: ]][[][ sss d εσ = (8) and 56 yzy xzx y w x w γθ γθ −∂ ∂= −∂ ∂= (2) where x w ∂ ∂ and y w ∂ ∂ are the slopes of the middle surface in the x and y directions and xzγ and yzγ are the shear strains. 4.2. strain–displacement relations for mindlin plate theory, the strain components may be written in terms of the displacements of the middle surface as follows: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ −= ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂ ∂∂ −= ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂ ∂∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = xy y x yx y x xy y x b z xy y x z xvyu yv xu κ κ κ θθ θ θ γ ε ε ε (3) ⎥ ⎦ ⎤ ⎢ ⎣ ⎡ −∂∂ −∂∂ =⎥ ⎦ ⎤ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂+∂∂ =⎥ ⎦ ⎤ ⎢ ⎣ ⎡ = y x yz xz s yw xw ywzv xwzu θ θ γ γ ε (4) where bε is bending strains vector, sε is shear strains vector and xκ , yκ are bending curvatures and xyκ is shear curvature. 4.3. stress– strain relations the moment–curvature relationships and the bending moments are [16]: ]][[][ bbb d εσ = (5) dzz m m m h h xy y x xy y x b ∫− ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = 2 2 ][ σ σ σ σ (6) where h is the plate thickness and ][ bd is the flexural rigidity matrix for an isotropic material which may be expressed as follows: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ − − = 2)1(00 01 01 )1(12 ][ 2 3 ν ν ν ν eh db (7) where e is young’s modulus, ν is poisson’s ratio. the shear force–shear strain relationships and also the shear forces are given as: ]][[][ sss d εσ = (8) are the slopes of the middle surface in the x and y directions and 56 yzy xzx y w x w γθ γθ −∂ ∂= −∂ ∂= (2) where x w ∂ ∂ and y w ∂ ∂ are the slopes of the middle surface in the x and y directions and xzγ and yzγ are the shear strains. 4.2. strain–displacement relations for mindlin plate theory, the strain components may be written in terms of the displacements of the middle surface as follows: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ −= ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂ ∂∂ −= ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂ ∂∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = xy y x yx y x xy y x b z xy y x z xvyu yv xu κ κ κ θθ θ θ γ ε ε ε (3) ⎥ ⎦ ⎤ ⎢ ⎣ ⎡ −∂∂ −∂∂ =⎥ ⎦ ⎤ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂+∂∂ =⎥ ⎦ ⎤ ⎢ ⎣ ⎡ = y x yz xz s yw xw ywzv xwzu θ θ γ γ ε (4) where bε is bending strains vector, sε is shear strains vector and xκ , yκ are bending curvatures and xyκ is shear curvature. 4.3. stress– strain relations the moment–curvature relationships and the bending moments are [16]: ]][[][ bbb d εσ = (5) dzz m m m h h xy y x xy y x b ∫− ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = 2 2 ][ σ σ σ σ (6) where h is the plate thickness and ][ bd is the flexural rigidity matrix for an isotropic material which may be expressed as follows: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ − − = 2)1(00 01 01 )1(12 ][ 2 3 ν ν ν ν eh db (7) where e is young’s modulus, ν is poisson’s ratio. the shear force–shear strain relationships and also the shear forces are given as: ]][[][ sss d εσ = (8) and 56 yzy xzx y w x w γθ γθ −∂ ∂= −∂ ∂= (2) where x w ∂ ∂ and y w ∂ ∂ are the slopes of the middle surface in the x and y directions and xzγ and yzγ are the shear strains. 4.2. strain–displacement relations for mindlin plate theory, the strain components may be written in terms of the displacements of the middle surface as follows: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ −= ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂ ∂∂ −= ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂ ∂∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = xy y x yx y x xy y x b z xy y x z xvyu yv xu κ κ κ θθ θ θ γ ε ε ε (3) ⎥ ⎦ ⎤ ⎢ ⎣ ⎡ −∂∂ −∂∂ =⎥ ⎦ ⎤ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂+∂∂ =⎥ ⎦ ⎤ ⎢ ⎣ ⎡ = y x yz xz s yw xw ywzv xwzu θ θ γ γ ε (4) where bε is bending strains vector, sε is shear strains vector and xκ , yκ are bending curvatures and xyκ is shear curvature. 4.3. stress– strain relations the moment–curvature relationships and the bending moments are [16]: ]][[][ bbb d εσ = (5) dzz m m m h h xy y x xy y x b ∫− ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = 2 2 ][ σ σ σ σ (6) where h is the plate thickness and ][ bd is the flexural rigidity matrix for an isotropic material which may be expressed as follows: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ − − = 2)1(00 01 01 )1(12 ][ 2 3 ν ν ν ν eh db (7) where e is young’s modulus, ν is poisson’s ratio. the shear force–shear strain relationships and also the shear forces are given as: ]][[][ sss d εσ = (8) are the shear strains. 4.2. strain–displacement relations for mindlin plate theory, the strain components may be written in terms of the displacements of the middle surface as follows: 56 yzy xzx y w x w γθ γθ −∂ ∂= −∂ ∂= (2) where x w ∂ ∂ and y w ∂ ∂ are the slopes of the middle surface in the x and y directions and xzγ and yzγ are the shear strains. 4.2. strain–displacement relations for mindlin plate theory, the strain components may be written in terms of the displacements of the middle surface as follows: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ −= ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂ ∂∂ −= ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂ ∂∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = xy y x yx y x xy y x b z xy y x z xvyu yv xu κ κ κ θθ θ θ γ ε ε ε (3) ⎥ ⎦ ⎤ ⎢ ⎣ ⎡ −∂∂ −∂∂ =⎥ ⎦ ⎤ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂+∂∂ =⎥ ⎦ ⎤ ⎢ ⎣ ⎡ = y x yz xz s yw xw ywzv xwzu θ θ γ γ ε (4) where bε is bending strains vector, sε is shear strains vector and xκ , yκ are bending curvatures and xyκ is shear curvature. 4.3. stress– strain relations the moment–curvature relationships and the bending moments are [16]: ]][[][ bbb d εσ = (5) dzz m m m h h xy y x xy y x b ∫− ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = 2 2 ][ σ σ σ σ (6) where h is the plate thickness and ][ bd is the flexural rigidity matrix for an isotropic material which may be expressed as follows: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ − − = 2)1(00 01 01 )1(12 ][ 2 3 ν ν ν ν eh db (7) where e is young’s modulus, ν is poisson’s ratio. the shear force–shear strain relationships and also the shear forces are given as: ]][[][ sss d εσ = (8) (3) 56 yzy xzx y w x w γθ γθ −∂ ∂= −∂ ∂= (2) where x w ∂ ∂ and y w ∂ ∂ are the slopes of the middle surface in the x and y directions and xzγ and yzγ are the shear strains. 4.2. strain–displacement relations for mindlin plate theory, the strain components may be written in terms of the displacements of the middle surface as follows: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ −= ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂ ∂∂ −= ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂ ∂∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = xy y x yx y x xy y x b z xy y x z xvyu yv xu κ κ κ θθ θ θ γ ε ε ε (3) ⎥ ⎦ ⎤ ⎢ ⎣ ⎡ −∂∂ −∂∂ =⎥ ⎦ ⎤ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂+∂∂ =⎥ ⎦ ⎤ ⎢ ⎣ ⎡ = y x yz xz s yw xw ywzv xwzu θ θ γ γ ε (4) where bε is bending strains vector, sε is shear strains vector and xκ , yκ are bending curvatures and xyκ is shear curvature. 4.3. stress– strain relations the moment–curvature relationships and the bending moments are [16]: ]][[][ bbb d εσ = (5) dzz m m m h h xy y x xy y x b ∫− ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = 2 2 ][ σ σ σ σ (6) where h is the plate thickness and ][ bd is the flexural rigidity matrix for an isotropic material which may be expressed as follows: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ − − = 2)1(00 01 01 )1(12 ][ 2 3 ν ν ν ν eh db (7) where e is young’s modulus, ν is poisson’s ratio. the shear force–shear strain relationships and also the shear forces are given as: ]][[][ sss d εσ = (8) (4) where 56 yzy xzx y w x w γθ γθ −∂ ∂= −∂ ∂= (2) where x w ∂ ∂ and y w ∂ ∂ are the slopes of the middle surface in the x and y directions and xzγ and yzγ are the shear strains. 4.2. strain–displacement relations for mindlin plate theory, the strain components may be written in terms of the displacements of the middle surface as follows: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ −= ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂ ∂∂ −= ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂ ∂∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = xy y x yx y x xy y x b z xy y x z xvyu yv xu κ κ κ θθ θ θ γ ε ε ε (3) ⎥ ⎦ ⎤ ⎢ ⎣ ⎡ −∂∂ −∂∂ =⎥ ⎦ ⎤ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂+∂∂ =⎥ ⎦ ⎤ ⎢ ⎣ ⎡ = y x yz xz s yw xw ywzv xwzu θ θ γ γ ε (4) where bε is bending strains vector, sε is shear strains vector and xκ , yκ are bending curvatures and xyκ is shear curvature. 4.3. stress– strain relations the moment–curvature relationships and the bending moments are [16]: ]][[][ bbb d εσ = (5) dzz m m m h h xy y x xy y x b ∫− ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = 2 2 ][ σ σ σ σ (6) where h is the plate thickness and ][ bd is the flexural rigidity matrix for an isotropic material which may be expressed as follows: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ − − = 2)1(00 01 01 )1(12 ][ 2 3 ν ν ν ν eh db (7) where e is young’s modulus, ν is poisson’s ratio. the shear force–shear strain relationships and also the shear forces are given as: ]][[][ sss d εσ = (8) is bending strains vector, 56 yzy xzx y w x w γθ γθ −∂ ∂= −∂ ∂= (2) where x w ∂ ∂ and y w ∂ ∂ are the slopes of the middle surface in the x and y directions and xzγ and yzγ are the shear strains. 4.2. strain–displacement relations for mindlin plate theory, the strain components may be written in terms of the displacements of the middle surface as follows: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ −= ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂ ∂∂ −= ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂ ∂∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = xy y x yx y x xy y x b z xy y x z xvyu yv xu κ κ κ θθ θ θ γ ε ε ε (3) ⎥ ⎦ ⎤ ⎢ ⎣ ⎡ −∂∂ −∂∂ =⎥ ⎦ ⎤ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂+∂∂ =⎥ ⎦ ⎤ ⎢ ⎣ ⎡ = y x yz xz s yw xw ywzv xwzu θ θ γ γ ε (4) where bε is bending strains vector, sε is shear strains vector and xκ , yκ are bending curvatures and xyκ is shear curvature. 4.3. stress– strain relations the moment–curvature relationships and the bending moments are [16]: ]][[][ bbb d εσ = (5) dzz m m m h h xy y x xy y x b ∫− ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = 2 2 ][ σ σ σ σ (6) where h is the plate thickness and ][ bd is the flexural rigidity matrix for an isotropic material which may be expressed as follows: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ − − = 2)1(00 01 01 )1(12 ][ 2 3 ν ν ν ν eh db (7) where e is young’s modulus, ν is poisson’s ratio. the shear force–shear strain relationships and also the shear forces are given as: ]][[][ sss d εσ = (8) is shear strains vector and 56 yzy xzx y w x w γθ γθ −∂ ∂= −∂ ∂= (2) where x w ∂ ∂ and y w ∂ ∂ are the slopes of the middle surface in the x and y directions and xzγ and yzγ are the shear strains. 4.2. strain–displacement relations for mindlin plate theory, the strain components may be written in terms of the displacements of the middle surface as follows: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ −= ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂ ∂∂ −= ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂ ∂∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = xy y x yx y x xy y x b z xy y x z xvyu yv xu κ κ κ θθ θ θ γ ε ε ε (3) ⎥ ⎦ ⎤ ⎢ ⎣ ⎡ −∂∂ −∂∂ =⎥ ⎦ ⎤ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂+∂∂ =⎥ ⎦ ⎤ ⎢ ⎣ ⎡ = y x yz xz s yw xw ywzv xwzu θ θ γ γ ε (4) where bε is bending strains vector, sε is shear strains vector and xκ , yκ are bending curvatures and xyκ is shear curvature. 4.3. stress– strain relations the moment–curvature relationships and the bending moments are [16]: ]][[][ bbb d εσ = (5) dzz m m m h h xy y x xy y x b ∫− ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = 2 2 ][ σ σ σ σ (6) where h is the plate thickness and ][ bd is the flexural rigidity matrix for an isotropic material which may be expressed as follows: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ − − = 2)1(00 01 01 )1(12 ][ 2 3 ν ν ν ν eh db (7) where e is young’s modulus, ν is poisson’s ratio. the shear force–shear strain relationships and also the shear forces are given as: ]][[][ sss d εσ = (8) , are bending curvatures and 56 yzy xzx y w x w γθ γθ −∂ ∂= −∂ ∂= (2) where x w ∂ ∂ and y w ∂ ∂ are the slopes of the middle surface in the x and y directions and xzγ and yzγ are the shear strains. 4.2. strain–displacement relations for mindlin plate theory, the strain components may be written in terms of the displacements of the middle surface as follows: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ −= ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂ ∂∂ −= ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂ ∂∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = xy y x yx y x xy y x b z xy y x z xvyu yv xu κ κ κ θθ θ θ γ ε ε ε (3) ⎥ ⎦ ⎤ ⎢ ⎣ ⎡ −∂∂ −∂∂ =⎥ ⎦ ⎤ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂+∂∂ =⎥ ⎦ ⎤ ⎢ ⎣ ⎡ = y x yz xz s yw xw ywzv xwzu θ θ γ γ ε (4) where bε is bending strains vector, sε is shear strains vector and xκ , yκ are bending curvatures and xyκ is shear curvature. 4.3. stress– strain relations the moment–curvature relationships and the bending moments are [16]: ]][[][ bbb d εσ = (5) dzz m m m h h xy y x xy y x b ∫− ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = 2 2 ][ σ σ σ σ (6) where h is the plate thickness and ][ bd is the flexural rigidity matrix for an isotropic material which may be expressed as follows: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ − − = 2)1(00 01 01 )1(12 ][ 2 3 ν ν ν ν eh db (7) where e is young’s modulus, ν is poisson’s ratio. the shear force–shear strain relationships and also the shear forces are given as: ]][[][ sss d εσ = (8) is shear curvature. issn: 2180-1053 vol. 4 no. 1 january-june 2012 implementation of nonlinear finite element using object–oriented design patterns 67 4.3. stress– strain relations the moment–curvature relationships and the bending moments are [16]: 56 yzy xzx y w x w γθ γθ −∂ ∂= −∂ ∂= (2) where x w ∂ ∂ and y w ∂ ∂ are the slopes of the middle surface in the x and y directions and xzγ and yzγ are the shear strains. 4.2. strain–displacement relations for mindlin plate theory, the strain components may be written in terms of the displacements of the middle surface as follows: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ −= ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂ ∂∂ −= ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂ ∂∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = xy y x yx y x xy y x b z xy y x z xvyu yv xu κ κ κ θθ θ θ γ ε ε ε (3) ⎥ ⎦ ⎤ ⎢ ⎣ ⎡ −∂∂ −∂∂ =⎥ ⎦ ⎤ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂+∂∂ =⎥ ⎦ ⎤ ⎢ ⎣ ⎡ = y x yz xz s yw xw ywzv xwzu θ θ γ γ ε (4) where bε is bending strains vector, sε is shear strains vector and xκ , yκ are bending curvatures and xyκ is shear curvature. 4.3. stress– strain relations the moment–curvature relationships and the bending moments are [16]: ]][[][ bbb d εσ = (5) dzz m m m h h xy y x xy y x b ∫− ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = 2 2 ][ σ σ σ σ (6) where h is the plate thickness and ][ bd is the flexural rigidity matrix for an isotropic material which may be expressed as follows: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ − − = 2)1(00 01 01 )1(12 ][ 2 3 ν ν ν ν eh db (7) where e is young’s modulus, ν is poisson’s ratio. the shear force–shear strain relationships and also the shear forces are given as: ]][[][ sss d εσ = (8) (5) 56 yzy xzx y w x w γθ γθ −∂ ∂= −∂ ∂= (2) where x w ∂ ∂ and y w ∂ ∂ are the slopes of the middle surface in the x and y directions and xzγ and yzγ are the shear strains. 4.2. strain–displacement relations for mindlin plate theory, the strain components may be written in terms of the displacements of the middle surface as follows: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ −= ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂ ∂∂ −= ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂ ∂∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = xy y x yx y x xy y x b z xy y x z xvyu yv xu κ κ κ θθ θ θ γ ε ε ε (3) ⎥ ⎦ ⎤ ⎢ ⎣ ⎡ −∂∂ −∂∂ =⎥ ⎦ ⎤ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂+∂∂ =⎥ ⎦ ⎤ ⎢ ⎣ ⎡ = y x yz xz s yw xw ywzv xwzu θ θ γ γ ε (4) where bε is bending strains vector, sε is shear strains vector and xκ , yκ are bending curvatures and xyκ is shear curvature. 4.3. stress– strain relations the moment–curvature relationships and the bending moments are [16]: ]][[][ bbb d εσ = (5) dzz m m m h h xy y x xy y x b ∫− ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = 2 2 ][ σ σ σ σ (6) where h is the plate thickness and ][ bd is the flexural rigidity matrix for an isotropic material which may be expressed as follows: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ − − = 2)1(00 01 01 )1(12 ][ 2 3 ν ν ν ν eh db (7) where e is young’s modulus, ν is poisson’s ratio. the shear force–shear strain relationships and also the shear forces are given as: ]][[][ sss d εσ = (8) (6) where h is the plate thickness and 56 yzy xzx y w x w γθ γθ −∂ ∂= −∂ ∂= (2) where x w ∂ ∂ and y w ∂ ∂ are the slopes of the middle surface in the x and y directions and xzγ and yzγ are the shear strains. 4.2. strain–displacement relations for mindlin plate theory, the strain components may be written in terms of the displacements of the middle surface as follows: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ −= ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂ ∂∂ −= ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂ ∂∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = xy y x yx y x xy y x b z xy y x z xvyu yv xu κ κ κ θθ θ θ γ ε ε ε (3) ⎥ ⎦ ⎤ ⎢ ⎣ ⎡ −∂∂ −∂∂ =⎥ ⎦ ⎤ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂+∂∂ =⎥ ⎦ ⎤ ⎢ ⎣ ⎡ = y x yz xz s yw xw ywzv xwzu θ θ γ γ ε (4) where bε is bending strains vector, sε is shear strains vector and xκ , yκ are bending curvatures and xyκ is shear curvature. 4.3. stress– strain relations the moment–curvature relationships and the bending moments are [16]: ]][[][ bbb d εσ = (5) dzz m m m h h xy y x xy y x b ∫− ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = 2 2 ][ σ σ σ σ (6) where h is the plate thickness and ][ bd is the flexural rigidity matrix for an isotropic material which may be expressed as follows: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ − − = 2)1(00 01 01 )1(12 ][ 2 3 ν ν ν ν eh db (7) where e is young’s modulus, ν is poisson’s ratio. the shear force–shear strain relationships and also the shear forces are given as: ]][[][ sss d εσ = (8) is the flexural rigidity matrix for an isotropic material which may be expressed as follows: 56 yzy xzx y w x w γθ γθ −∂ ∂= −∂ ∂= (2) where x w ∂ ∂ and y w ∂ ∂ are the slopes of the middle surface in the x and y directions and xzγ and yzγ are the shear strains. 4.2. strain–displacement relations for mindlin plate theory, the strain components may be written in terms of the displacements of the middle surface as follows: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ −= ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂ ∂∂ −= ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂ ∂∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = xy y x yx y x xy y x b z xy y x z xvyu yv xu κ κ κ θθ θ θ γ ε ε ε (3) ⎥ ⎦ ⎤ ⎢ ⎣ ⎡ −∂∂ −∂∂ =⎥ ⎦ ⎤ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂+∂∂ =⎥ ⎦ ⎤ ⎢ ⎣ ⎡ = y x yz xz s yw xw ywzv xwzu θ θ γ γ ε (4) where bε is bending strains vector, sε is shear strains vector and xκ , yκ are bending curvatures and xyκ is shear curvature. 4.3. stress– strain relations the moment–curvature relationships and the bending moments are [16]: ]][[][ bbb d εσ = (5) dzz m m m h h xy y x xy y x b ∫− ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = 2 2 ][ σ σ σ σ (6) where h is the plate thickness and ][ bd is the flexural rigidity matrix for an isotropic material which may be expressed as follows: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ − − = 2)1(00 01 01 )1(12 ][ 2 3 ν ν ν ν eh db (7) where e is young’s modulus, ν is poisson’s ratio. the shear force–shear strain relationships and also the shear forces are given as: ]][[][ sss d εσ = (8) (7) where e is young’s modulus, 56 yzy xzx y w x w γθ γθ −∂ ∂= −∂ ∂= (2) where x w ∂ ∂ and y w ∂ ∂ are the slopes of the middle surface in the x and y directions and xzγ and yzγ are the shear strains. 4.2. strain–displacement relations for mindlin plate theory, the strain components may be written in terms of the displacements of the middle surface as follows: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ −= ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂ ∂∂ −= ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂ ∂∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = xy y x yx y x xy y x b z xy y x z xvyu yv xu κ κ κ θθ θ θ γ ε ε ε (3) ⎥ ⎦ ⎤ ⎢ ⎣ ⎡ −∂∂ −∂∂ =⎥ ⎦ ⎤ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂+∂∂ =⎥ ⎦ ⎤ ⎢ ⎣ ⎡ = y x yz xz s yw xw ywzv xwzu θ θ γ γ ε (4) where bε is bending strains vector, sε is shear strains vector and xκ , yκ are bending curvatures and xyκ is shear curvature. 4.3. stress– strain relations the moment–curvature relationships and the bending moments are [16]: ]][[][ bbb d εσ = (5) dzz m m m h h xy y x xy y x b ∫− ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = 2 2 ][ σ σ σ σ (6) where h is the plate thickness and ][ bd is the flexural rigidity matrix for an isotropic material which may be expressed as follows: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ − − = 2)1(00 01 01 )1(12 ][ 2 3 ν ν ν ν eh db (7) where e is young’s modulus, ν is poisson’s ratio. the shear force–shear strain relationships and also the shear forces are given as: ]][[][ sss d εσ = (8) is poisson’s ratio. the shear force– shear strain relationships and also the shear forces are given as: 56 yzy xzx y w x w γθ γθ −∂ ∂= −∂ ∂= (2) where x w ∂ ∂ and y w ∂ ∂ are the slopes of the middle surface in the x and y directions and xzγ and yzγ are the shear strains. 4.2. strain–displacement relations for mindlin plate theory, the strain components may be written in terms of the displacements of the middle surface as follows: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ −= ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂ ∂∂ −= ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂ ∂∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = xy y x yx y x xy y x b z xy y x z xvyu yv xu κ κ κ θθ θ θ γ ε ε ε (3) ⎥ ⎦ ⎤ ⎢ ⎣ ⎡ −∂∂ −∂∂ =⎥ ⎦ ⎤ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂+∂∂ =⎥ ⎦ ⎤ ⎢ ⎣ ⎡ = y x yz xz s yw xw ywzv xwzu θ θ γ γ ε (4) where bε is bending strains vector, sε is shear strains vector and xκ , yκ are bending curvatures and xyκ is shear curvature. 4.3. stress– strain relations the moment–curvature relationships and the bending moments are [16]: ]][[][ bbb d εσ = (5) dzz m m m h h xy y x xy y x b ∫− ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = 2 2 ][ σ σ σ σ (6) where h is the plate thickness and ][ bd is the flexural rigidity matrix for an isotropic material which may be expressed as follows: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ − − = 2)1(00 01 01 )1(12 ][ 2 3 ν ν ν ν eh db (7) where e is young’s modulus, ν is poisson’s ratio. the shear force–shear strain relationships and also the shear forces are given as: ]][[][ sss d εσ = (8) (8) 57 dz q q h h yz xz y x s ∫− ⎥⎦ ⎤ ⎢ ⎣ ⎡ =⎥ ⎦ ⎤ ⎢ ⎣ ⎡ = 2 2 ][ σ σ σ (9) where ][ sd is the matrix of shear rigidity for an isotropic material which may be expressed as: ⎥ ⎦ ⎤ ⎢ ⎣ ⎡ + = 10 01 )1(2 ][ ν keh ds (10) here, k is the shear modification factor and is normally set equal to 5/6 for homogeneous isotropic plates. 4.4. finite element formulation the normal displacement and normal rotation of the mid–plane at a typical point with local coordinates, ),( ηξ in an element with n nodes are obtained using: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∑ = yi xi in i i i i y x w n n nw θ θ θ θ 1 00 00 00 (11) or i n i i unu rr ∑ = = 1 (12) where in is the shape function corresponded node i, and iw , xiθ and yiθ are the displacement and rotation values of node i. the curvatures in eq. (3) are expressed as: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂∂∂ ∂∂ ∂∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂ ∂∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∑ = yi xi in i iii i i yx y x xy y x w xnyn yn xn xy y x θ θ θθ θ θ κ κ κ 1 0 00 00 (13) or i n i bib ub r ∑ = = 1 ε (14) where bib is the curvature–displacement matrix. the shear strains in eq. (4) can be expressed as: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ⎥ ⎦ ⎤ ⎢ ⎣ ⎡ −∂∂ −∂∂ =⎥ ⎦ ⎤ ⎢ ⎣ ⎡ −∂∂ −∂∂ = ∑ = yi xi in i ii ii y x s w nyn nxn yw xw θ θ θ θ ε 1 0 0 (15) or (9) where 57 dz q q h h yz xz y x s ∫− ⎥⎦ ⎤ ⎢ ⎣ ⎡ =⎥ ⎦ ⎤ ⎢ ⎣ ⎡ = 2 2 ][ σ σ σ (9) where ][ sd is the matrix of shear rigidity for an isotropic material which may be expressed as: ⎥ ⎦ ⎤ ⎢ ⎣ ⎡ + = 10 01 )1(2 ][ ν keh ds (10) here, k is the shear modification factor and is normally set equal to 5/6 for homogeneous isotropic plates. 4.4. finite element formulation the normal displacement and normal rotation of the mid–plane at a typical point with local coordinates, ),( ηξ in an element with n nodes are obtained using: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∑ = yi xi in i i i i y x w n n nw θ θ θ θ 1 00 00 00 (11) or i n i i unu rr ∑ = = 1 (12) where in is the shape function corresponded node i, and iw , xiθ and yiθ are the displacement and rotation values of node i. the curvatures in eq. (3) are expressed as: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂∂∂ ∂∂ ∂∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂ ∂∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∑ = yi xi in i iii i i yx y x xy y x w xnyn yn xn xy y x θ θ θθ θ θ κ κ κ 1 0 00 00 (13) or i n i bib ub r ∑ = = 1 ε (14) where bib is the curvature–displacement matrix. the shear strains in eq. (4) can be expressed as: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ⎥ ⎦ ⎤ ⎢ ⎣ ⎡ −∂∂ −∂∂ =⎥ ⎦ ⎤ ⎢ ⎣ ⎡ −∂∂ −∂∂ = ∑ = yi xi in i ii ii y x s w nyn nxn yw xw θ θ θ θ ε 1 0 0 (15) or is the matrix of shear rigidity for an isotropic material which may be expressed as: 57 dz q q h h yz xz y x s ∫− ⎥⎦ ⎤ ⎢ ⎣ ⎡ =⎥ ⎦ ⎤ ⎢ ⎣ ⎡ = 2 2 ][ σ σ σ (9) where ][ sd is the matrix of shear rigidity for an isotropic material which may be expressed as: ⎥ ⎦ ⎤ ⎢ ⎣ ⎡ + = 10 01 )1(2 ][ ν keh ds (10) here, k is the shear modification factor and is normally set equal to 5/6 for homogeneous isotropic plates. 4.4. finite element formulation the normal displacement and normal rotation of the mid–plane at a typical point with local coordinates, ),( ηξ in an element with n nodes are obtained using: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∑ = yi xi in i i i i y x w n n nw θ θ θ θ 1 00 00 00 (11) or i n i i unu rr ∑ = = 1 (12) where in is the shape function corresponded node i, and iw , xiθ and yiθ are the displacement and rotation values of node i. the curvatures in eq. (3) are expressed as: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂∂∂ ∂∂ ∂∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂ ∂∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∑ = yi xi in i iii i i yx y x xy y x w xnyn yn xn xy y x θ θ θθ θ θ κ κ κ 1 0 00 00 (13) or i n i bib ub r ∑ = = 1 ε (14) where bib is the curvature–displacement matrix. the shear strains in eq. (4) can be expressed as: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ⎥ ⎦ ⎤ ⎢ ⎣ ⎡ −∂∂ −∂∂ =⎥ ⎦ ⎤ ⎢ ⎣ ⎡ −∂∂ −∂∂ = ∑ = yi xi in i ii ii y x s w nyn nxn yw xw θ θ θ θ ε 1 0 0 (15) or (10) here, k is the shear modification factor and is normally set equal to 5/6 for homogeneous isotropic plates. issn: 2180-1053 vol. 4 no. 1 january-june 2012 journal of mechanical engineering and technology 68 4.4. finite element formulation the normal displacement and normal rotation of the mid–plane at a typical point with local coordinates, 57 dz q q h h yz xz y x s ∫− ⎥⎦ ⎤ ⎢ ⎣ ⎡ =⎥ ⎦ ⎤ ⎢ ⎣ ⎡ = 2 2 ][ σ σ σ (9) where ][ sd is the matrix of shear rigidity for an isotropic material which may be expressed as: ⎥ ⎦ ⎤ ⎢ ⎣ ⎡ + = 10 01 )1(2 ][ ν keh ds (10) here, k is the shear modification factor and is normally set equal to 5/6 for homogeneous isotropic plates. 4.4. finite element formulation the normal displacement and normal rotation of the mid–plane at a typical point with local coordinates, ),( ηξ in an element with n nodes are obtained using: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∑ = yi xi in i i i i y x w n n nw θ θ θ θ 1 00 00 00 (11) or i n i i unu rr ∑ = = 1 (12) where in is the shape function corresponded node i, and iw , xiθ and yiθ are the displacement and rotation values of node i. the curvatures in eq. (3) are expressed as: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂∂∂ ∂∂ ∂∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂ ∂∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∑ = yi xi in i iii i i yx y x xy y x w xnyn yn xn xy y x θ θ θθ θ θ κ κ κ 1 0 00 00 (13) or i n i bib ub r ∑ = = 1 ε (14) where bib is the curvature–displacement matrix. the shear strains in eq. (4) can be expressed as: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ⎥ ⎦ ⎤ ⎢ ⎣ ⎡ −∂∂ −∂∂ =⎥ ⎦ ⎤ ⎢ ⎣ ⎡ −∂∂ −∂∂ = ∑ = yi xi in i ii ii y x s w nyn nxn yw xw θ θ θ θ ε 1 0 0 (15) or in an element with n nodes are obtained using: 57 dz q q h h yz xz y x s ∫− ⎥⎦ ⎤ ⎢ ⎣ ⎡ =⎥ ⎦ ⎤ ⎢ ⎣ ⎡ = 2 2 ][ σ σ σ (9) where ][ sd is the matrix of shear rigidity for an isotropic material which may be expressed as: ⎥ ⎦ ⎤ ⎢ ⎣ ⎡ + = 10 01 )1(2 ][ ν keh ds (10) here, k is the shear modification factor and is normally set equal to 5/6 for homogeneous isotropic plates. 4.4. finite element formulation the normal displacement and normal rotation of the mid–plane at a typical point with local coordinates, ),( ηξ in an element with n nodes are obtained using: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∑ = yi xi in i i i i y x w n n nw θ θ θ θ 1 00 00 00 (11) or i n i i unu rr ∑ = = 1 (12) where in is the shape function corresponded node i, and iw , xiθ and yiθ are the displacement and rotation values of node i. the curvatures in eq. (3) are expressed as: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂∂∂ ∂∂ ∂∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂ ∂∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∑ = yi xi in i iii i i yx y x xy y x w xnyn yn xn xy y x θ θ θθ θ θ κ κ κ 1 0 00 00 (13) or i n i bib ub r ∑ = = 1 ε (14) where bib is the curvature–displacement matrix. the shear strains in eq. (4) can be expressed as: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ⎥ ⎦ ⎤ ⎢ ⎣ ⎡ −∂∂ −∂∂ =⎥ ⎦ ⎤ ⎢ ⎣ ⎡ −∂∂ −∂∂ = ∑ = yi xi in i ii ii y x s w nyn nxn yw xw θ θ θ θ ε 1 0 0 (15) or (11) or 57 dz q q h h yz xz y x s ∫− ⎥⎦ ⎤ ⎢ ⎣ ⎡ =⎥ ⎦ ⎤ ⎢ ⎣ ⎡ = 2 2 ][ σ σ σ (9) where ][ sd is the matrix of shear rigidity for an isotropic material which may be expressed as: ⎥ ⎦ ⎤ ⎢ ⎣ ⎡ + = 10 01 )1(2 ][ ν keh ds (10) here, k is the shear modification factor and is normally set equal to 5/6 for homogeneous isotropic plates. 4.4. finite element formulation the normal displacement and normal rotation of the mid–plane at a typical point with local coordinates, ),( ηξ in an element with n nodes are obtained using: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∑ = yi xi in i i i i y x w n n nw θ θ θ θ 1 00 00 00 (11) or i n i i unu rr ∑ = = 1 (12) where in is the shape function corresponded node i, and iw , xiθ and yiθ are the displacement and rotation values of node i. the curvatures in eq. (3) are expressed as: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂∂∂ ∂∂ ∂∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂ ∂∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∑ = yi xi in i iii i i yx y x xy y x w xnyn yn xn xy y x θ θ θθ θ θ κ κ κ 1 0 00 00 (13) or i n i bib ub r ∑ = = 1 ε (14) where bib is the curvature–displacement matrix. the shear strains in eq. (4) can be expressed as: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ⎥ ⎦ ⎤ ⎢ ⎣ ⎡ −∂∂ −∂∂ =⎥ ⎦ ⎤ ⎢ ⎣ ⎡ −∂∂ −∂∂ = ∑ = yi xi in i ii ii y x s w nyn nxn yw xw θ θ θ θ ε 1 0 0 (15) or (12) where 57 dz q q h h yz xz y x s ∫− ⎥⎦ ⎤ ⎢ ⎣ ⎡ =⎥ ⎦ ⎤ ⎢ ⎣ ⎡ = 2 2 ][ σ σ σ (9) where ][ sd is the matrix of shear rigidity for an isotropic material which may be expressed as: ⎥ ⎦ ⎤ ⎢ ⎣ ⎡ + = 10 01 )1(2 ][ ν keh ds (10) here, k is the shear modification factor and is normally set equal to 5/6 for homogeneous isotropic plates. 4.4. finite element formulation the normal displacement and normal rotation of the mid–plane at a typical point with local coordinates, ),( ηξ in an element with n nodes are obtained using: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∑ = yi xi in i i i i y x w n n nw θ θ θ θ 1 00 00 00 (11) or i n i i unu rr ∑ = = 1 (12) where in is the shape function corresponded node i, and iw , xiθ and yiθ are the displacement and rotation values of node i. the curvatures in eq. (3) are expressed as: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂∂∂ ∂∂ ∂∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂ ∂∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∑ = yi xi in i iii i i yx y x xy y x w xnyn yn xn xy y x θ θ θθ θ θ κ κ κ 1 0 00 00 (13) or i n i bib ub r ∑ = = 1 ε (14) where bib is the curvature–displacement matrix. the shear strains in eq. (4) can be expressed as: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ⎥ ⎦ ⎤ ⎢ ⎣ ⎡ −∂∂ −∂∂ =⎥ ⎦ ⎤ ⎢ ⎣ ⎡ −∂∂ −∂∂ = ∑ = yi xi in i ii ii y x s w nyn nxn yw xw θ θ θ θ ε 1 0 0 (15) or is the shape function corresponded node i, and 57 dz q q h h yz xz y x s ∫− ⎥⎦ ⎤ ⎢ ⎣ ⎡ =⎥ ⎦ ⎤ ⎢ ⎣ ⎡ = 2 2 ][ σ σ σ (9) where ][ sd is the matrix of shear rigidity for an isotropic material which may be expressed as: ⎥ ⎦ ⎤ ⎢ ⎣ ⎡ + = 10 01 )1(2 ][ ν keh ds (10) here, k is the shear modification factor and is normally set equal to 5/6 for homogeneous isotropic plates. 4.4. finite element formulation the normal displacement and normal rotation of the mid–plane at a typical point with local coordinates, ),( ηξ in an element with n nodes are obtained using: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∑ = yi xi in i i i i y x w n n nw θ θ θ θ 1 00 00 00 (11) or i n i i unu rr ∑ = = 1 (12) where in is the shape function corresponded node i, and iw , xiθ and yiθ are the displacement and rotation values of node i. the curvatures in eq. (3) are expressed as: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂∂∂ ∂∂ ∂∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂ ∂∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∑ = yi xi in i iii i i yx y x xy y x w xnyn yn xn xy y x θ θ θθ θ θ κ κ κ 1 0 00 00 (13) or i n i bib ub r ∑ = = 1 ε (14) where bib is the curvature–displacement matrix. the shear strains in eq. (4) can be expressed as: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ⎥ ⎦ ⎤ ⎢ ⎣ ⎡ −∂∂ −∂∂ =⎥ ⎦ ⎤ ⎢ ⎣ ⎡ −∂∂ −∂∂ = ∑ = yi xi in i ii ii y x s w nyn nxn yw xw θ θ θ θ ε 1 0 0 (15) or , 57 dz q q h h yz xz y x s ∫− ⎥⎦ ⎤ ⎢ ⎣ ⎡ =⎥ ⎦ ⎤ ⎢ ⎣ ⎡ = 2 2 ][ σ σ σ (9) where ][ sd is the matrix of shear rigidity for an isotropic material which may be expressed as: ⎥ ⎦ ⎤ ⎢ ⎣ ⎡ + = 10 01 )1(2 ][ ν keh ds (10) here, k is the shear modification factor and is normally set equal to 5/6 for homogeneous isotropic plates. 4.4. finite element formulation the normal displacement and normal rotation of the mid–plane at a typical point with local coordinates, ),( ηξ in an element with n nodes are obtained using: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∑ = yi xi in i i i i y x w n n nw θ θ θ θ 1 00 00 00 (11) or i n i i unu rr ∑ = = 1 (12) where in is the shape function corresponded node i, and iw , xiθ and yiθ are the displacement and rotation values of node i. the curvatures in eq. (3) are expressed as: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂∂∂ ∂∂ ∂∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂ ∂∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∑ = yi xi in i iii i i yx y x xy y x w xnyn yn xn xy y x θ θ θθ θ θ κ κ κ 1 0 00 00 (13) or i n i bib ub r ∑ = = 1 ε (14) where bib is the curvature–displacement matrix. the shear strains in eq. (4) can be expressed as: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ⎥ ⎦ ⎤ ⎢ ⎣ ⎡ −∂∂ −∂∂ =⎥ ⎦ ⎤ ⎢ ⎣ ⎡ −∂∂ −∂∂ = ∑ = yi xi in i ii ii y x s w nyn nxn yw xw θ θ θ θ ε 1 0 0 (15) or are the displacement and rotation values of node i. the curvatures in eq. (3) are expressed as: 57 dz q q h h yz xz y x s ∫− ⎥⎦ ⎤ ⎢ ⎣ ⎡ =⎥ ⎦ ⎤ ⎢ ⎣ ⎡ = 2 2 ][ σ σ σ (9) where ][ sd is the matrix of shear rigidity for an isotropic material which may be expressed as: ⎥ ⎦ ⎤ ⎢ ⎣ ⎡ + = 10 01 )1(2 ][ ν keh ds (10) here, k is the shear modification factor and is normally set equal to 5/6 for homogeneous isotropic plates. 4.4. finite element formulation the normal displacement and normal rotation of the mid–plane at a typical point with local coordinates, ),( ηξ in an element with n nodes are obtained using: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∑ = yi xi in i i i i y x w n n nw θ θ θ θ 1 00 00 00 (11) or i n i i unu rr ∑ = = 1 (12) where in is the shape function corresponded node i, and iw , xiθ and yiθ are the displacement and rotation values of node i. the curvatures in eq. (3) are expressed as: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂∂∂ ∂∂ ∂∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂ ∂∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∑ = yi xi in i iii i i yx y x xy y x w xnyn yn xn xy y x θ θ θθ θ θ κ κ κ 1 0 00 00 (13) or i n i bib ub r ∑ = = 1 ε (14) where bib is the curvature–displacement matrix. the shear strains in eq. (4) can be expressed as: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ⎥ ⎦ ⎤ ⎢ ⎣ ⎡ −∂∂ −∂∂ =⎥ ⎦ ⎤ ⎢ ⎣ ⎡ −∂∂ −∂∂ = ∑ = yi xi in i ii ii y x s w nyn nxn yw xw θ θ θ θ ε 1 0 0 (15) or (13) or 57 dz q q h h yz xz y x s ∫− ⎥⎦ ⎤ ⎢ ⎣ ⎡ =⎥ ⎦ ⎤ ⎢ ⎣ ⎡ = 2 2 ][ σ σ σ (9) where ][ sd is the matrix of shear rigidity for an isotropic material which may be expressed as: ⎥ ⎦ ⎤ ⎢ ⎣ ⎡ + = 10 01 )1(2 ][ ν keh ds (10) here, k is the shear modification factor and is normally set equal to 5/6 for homogeneous isotropic plates. 4.4. finite element formulation the normal displacement and normal rotation of the mid–plane at a typical point with local coordinates, ),( ηξ in an element with n nodes are obtained using: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∑ = yi xi in i i i i y x w n n nw θ θ θ θ 1 00 00 00 (11) or i n i i unu rr ∑ = = 1 (12) where in is the shape function corresponded node i, and iw , xiθ and yiθ are the displacement and rotation values of node i. the curvatures in eq. (3) are expressed as: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂∂∂ ∂∂ ∂∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂ ∂∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∑ = yi xi in i iii i i yx y x xy y x w xnyn yn xn xy y x θ θ θθ θ θ κ κ κ 1 0 00 00 (13) or i n i bib ub r ∑ = = 1 ε (14) where bib is the curvature–displacement matrix. the shear strains in eq. (4) can be expressed as: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ⎥ ⎦ ⎤ ⎢ ⎣ ⎡ −∂∂ −∂∂ =⎥ ⎦ ⎤ ⎢ ⎣ ⎡ −∂∂ −∂∂ = ∑ = yi xi in i ii ii y x s w nyn nxn yw xw θ θ θ θ ε 1 0 0 (15) or (14) where 57 dz q q h h yz xz y x s ∫− ⎥⎦ ⎤ ⎢ ⎣ ⎡ =⎥ ⎦ ⎤ ⎢ ⎣ ⎡ = 2 2 ][ σ σ σ (9) where ][ sd is the matrix of shear rigidity for an isotropic material which may be expressed as: ⎥ ⎦ ⎤ ⎢ ⎣ ⎡ + = 10 01 )1(2 ][ ν keh ds (10) here, k is the shear modification factor and is normally set equal to 5/6 for homogeneous isotropic plates. 4.4. finite element formulation the normal displacement and normal rotation of the mid–plane at a typical point with local coordinates, ),( ηξ in an element with n nodes are obtained using: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∑ = yi xi in i i i i y x w n n nw θ θ θ θ 1 00 00 00 (11) or i n i i unu rr ∑ = = 1 (12) where in is the shape function corresponded node i, and iw , xiθ and yiθ are the displacement and rotation values of node i. the curvatures in eq. (3) are expressed as: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂∂∂ ∂∂ ∂∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂ ∂∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∑ = yi xi in i iii i i yx y x xy y x w xnyn yn xn xy y x θ θ θθ θ θ κ κ κ 1 0 00 00 (13) or i n i bib ub r ∑ = = 1 ε (14) where bib is the curvature–displacement matrix. the shear strains in eq. (4) can be expressed as: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ⎥ ⎦ ⎤ ⎢ ⎣ ⎡ −∂∂ −∂∂ =⎥ ⎦ ⎤ ⎢ ⎣ ⎡ −∂∂ −∂∂ = ∑ = yi xi in i ii ii y x s w nyn nxn yw xw θ θ θ θ ε 1 0 0 (15) or is the curvature–displacement matrix. the shear strains in eq. (4) can be expressed as: 57 dz q q h h yz xz y x s ∫− ⎥⎦ ⎤ ⎢ ⎣ ⎡ =⎥ ⎦ ⎤ ⎢ ⎣ ⎡ = 2 2 ][ σ σ σ (9) where ][ sd is the matrix of shear rigidity for an isotropic material which may be expressed as: ⎥ ⎦ ⎤ ⎢ ⎣ ⎡ + = 10 01 )1(2 ][ ν keh ds (10) here, k is the shear modification factor and is normally set equal to 5/6 for homogeneous isotropic plates. 4.4. finite element formulation the normal displacement and normal rotation of the mid–plane at a typical point with local coordinates, ),( ηξ in an element with n nodes are obtained using: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∑ = yi xi in i i i i y x w n n nw θ θ θ θ 1 00 00 00 (11) or i n i i unu rr ∑ = = 1 (12) where in is the shape function corresponded node i, and iw , xiθ and yiθ are the displacement and rotation values of node i. the curvatures in eq. (3) are expressed as: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂∂∂ ∂∂ ∂∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂+∂∂ ∂∂ ∂∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∑ = yi xi in i iii i i yx y x xy y x w xnyn yn xn xy y x θ θ θθ θ θ κ κ κ 1 0 00 00 (13) or i n i bib ub r ∑ = = 1 ε (14) where bib is the curvature–displacement matrix. the shear strains in eq. (4) can be expressed as: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ⎥ ⎦ ⎤ ⎢ ⎣ ⎡ −∂∂ −∂∂ =⎥ ⎦ ⎤ ⎢ ⎣ ⎡ −∂∂ −∂∂ = ∑ = yi xi in i ii ii y x s w nyn nxn yw xw θ θ θ θ ε 1 0 0 (15) or (15) issn: 2180-1053 vol. 4 no. 1 january-june 2012 implementation of nonlinear finite element using object–oriented design patterns 69 58 i n i sis ub r ∑ = = 1 ε (16) where sib is the shear strain–displacement matrix. the bending and shear stiffness matrices can be written as: dabdbk bb a t b e b ][][][][ ∫= (17) dabdbk ss a t s e s ][][][][ ∫= (18) where ][][ 21 bnbbb bbbb k= , ][][ 21 snsss bbbb k= and e bk ][ and e sk ][ are element bending and shear stiffness matrices, respectively. so the total element matrix is: e s e b e kkk ][][][ += (19) the elemental load vector is also obtained from: daqnf a i e ∫= ][ (20) where q is the uniformly distributed load. the element jacobian matrix is: ⎥ ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂∂∂ ∂∂∂∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂ ∂ ∂ ∂ ∂ ∂ ∂ ∂ = ∑∑ ∑∑ == == n i ii n i ii n i ii n i ii ynxn ynxn yx yx j 11 11 )()( )()( ηη ξξ ηη ξξ (21) and its inverse is: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂ ∂ ∂ ∂− ∂ ∂−∂ ∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂ ∂ ∂ ∂ ∂ ∂ ∂ ∂ =− ξη ξη ηξ ηξ xx yy j yy xx j det 11 (22) where jdet is the determinant of jacobian matrix. to obtain shape functions derivative respect to x, y chair rule can be applied, so: x n x n x n iii ∂ ∂ ∂ ∂ + ∂ ∂ ∂ ∂ = ∂ ∂ η η ξ ξ (23) y n y n y n iii ∂ ∂ ∂ ∂ + ∂ ∂ ∂ ∂ = ∂ ∂ η η ξ ξ (24) the relation between cartesian coordinate and local coordinate may be written as: ηξ ddjdydx det= (25) (16) where 58 i n i sis ub r ∑ = = 1 ε (16) where sib is the shear strain–displacement matrix. the bending and shear stiffness matrices can be written as: dabdbk bb a t b e b ][][][][ ∫= (17) dabdbk ss a t s e s ][][][][ ∫= (18) where ][][ 21 bnbbb bbbb k= , ][][ 21 snsss bbbb k= and e bk ][ and e sk ][ are element bending and shear stiffness matrices, respectively. so the total element matrix is: e s e b e kkk ][][][ += (19) the elemental load vector is also obtained from: daqnf a i e ∫= ][ (20) where q is the uniformly distributed load. the element jacobian matrix is: ⎥ ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂∂∂ ∂∂∂∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂ ∂ ∂ ∂ ∂ ∂ ∂ ∂ = ∑∑ ∑∑ == == n i ii n i ii n i ii n i ii ynxn ynxn yx yx j 11 11 )()( )()( ηη ξξ ηη ξξ (21) and its inverse is: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂ ∂ ∂ ∂− ∂ ∂−∂ ∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂ ∂ ∂ ∂ ∂ ∂ ∂ ∂ =− ξη ξη ηξ ηξ xx yy j yy xx j det 11 (22) where jdet is the determinant of jacobian matrix. to obtain shape functions derivative respect to x, y chair rule can be applied, so: x n x n x n iii ∂ ∂ ∂ ∂ + ∂ ∂ ∂ ∂ = ∂ ∂ η η ξ ξ (23) y n y n y n iii ∂ ∂ ∂ ∂ + ∂ ∂ ∂ ∂ = ∂ ∂ η η ξ ξ (24) the relation between cartesian coordinate and local coordinate may be written as: ηξ ddjdydx det= (25) is the shear strain–displacement matrix. the bending and shear stiffness matrices can be written as: 58 i n i sis ub r ∑ = = 1 ε (16) where sib is the shear strain–displacement matrix. the bending and shear stiffness matrices can be written as: dabdbk bb a t b e b ][][][][ ∫= (17) dabdbk ss a t s e s ][][][][ ∫= (18) where ][][ 21 bnbbb bbbb k= , ][][ 21 snsss bbbb k= and e bk ][ and e sk ][ are element bending and shear stiffness matrices, respectively. so the total element matrix is: e s e b e kkk ][][][ += (19) the elemental load vector is also obtained from: daqnf a i e ∫= ][ (20) where q is the uniformly distributed load. the element jacobian matrix is: ⎥ ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂∂∂ ∂∂∂∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂ ∂ ∂ ∂ ∂ ∂ ∂ ∂ = ∑∑ ∑∑ == == n i ii n i ii n i ii n i ii ynxn ynxn yx yx j 11 11 )()( )()( ηη ξξ ηη ξξ (21) and its inverse is: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂ ∂ ∂ ∂− ∂ ∂−∂ ∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂ ∂ ∂ ∂ ∂ ∂ ∂ ∂ =− ξη ξη ηξ ηξ xx yy j yy xx j det 11 (22) where jdet is the determinant of jacobian matrix. to obtain shape functions derivative respect to x, y chair rule can be applied, so: x n x n x n iii ∂ ∂ ∂ ∂ + ∂ ∂ ∂ ∂ = ∂ ∂ η η ξ ξ (23) y n y n y n iii ∂ ∂ ∂ ∂ + ∂ ∂ ∂ ∂ = ∂ ∂ η η ξ ξ (24) the relation between cartesian coordinate and local coordinate may be written as: ηξ ddjdydx det= (25) (17) 58 i n i sis ub r ∑ = = 1 ε (16) where sib is the shear strain–displacement matrix. the bending and shear stiffness matrices can be written as: dabdbk bb a t b e b ][][][][ ∫= (17) dabdbk ss a t s e s ][][][][ ∫= (18) where ][][ 21 bnbbb bbbb k= , ][][ 21 snsss bbbb k= and e bk ][ and e sk ][ are element bending and shear stiffness matrices, respectively. so the total element matrix is: e s e b e kkk ][][][ += (19) the elemental load vector is also obtained from: daqnf a i e ∫= ][ (20) where q is the uniformly distributed load. the element jacobian matrix is: ⎥ ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂∂∂ ∂∂∂∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂ ∂ ∂ ∂ ∂ ∂ ∂ ∂ = ∑∑ ∑∑ == == n i ii n i ii n i ii n i ii ynxn ynxn yx yx j 11 11 )()( )()( ηη ξξ ηη ξξ (21) and its inverse is: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂ ∂ ∂ ∂− ∂ ∂−∂ ∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂ ∂ ∂ ∂ ∂ ∂ ∂ ∂ =− ξη ξη ηξ ηξ xx yy j yy xx j det 11 (22) where jdet is the determinant of jacobian matrix. to obtain shape functions derivative respect to x, y chair rule can be applied, so: x n x n x n iii ∂ ∂ ∂ ∂ + ∂ ∂ ∂ ∂ = ∂ ∂ η η ξ ξ (23) y n y n y n iii ∂ ∂ ∂ ∂ + ∂ ∂ ∂ ∂ = ∂ ∂ η η ξ ξ (24) the relation between cartesian coordinate and local coordinate may be written as: ηξ ddjdydx det= (25) (18) where 58 i n i sis ub r ∑ = = 1 ε (16) where sib is the shear strain–displacement matrix. the bending and shear stiffness matrices can be written as: dabdbk bb a t b e b ][][][][ ∫= (17) dabdbk ss a t s e s ][][][][ ∫= (18) where ][][ 21 bnbbb bbbb k= , ][][ 21 snsss bbbb k= and e bk ][ and e sk ][ are element bending and shear stiffness matrices, respectively. so the total element matrix is: e s e b e kkk ][][][ += (19) the elemental load vector is also obtained from: daqnf a i e ∫= ][ (20) where q is the uniformly distributed load. the element jacobian matrix is: ⎥ ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂∂∂ ∂∂∂∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂ ∂ ∂ ∂ ∂ ∂ ∂ ∂ = ∑∑ ∑∑ == == n i ii n i ii n i ii n i ii ynxn ynxn yx yx j 11 11 )()( )()( ηη ξξ ηη ξξ (21) and its inverse is: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂ ∂ ∂ ∂− ∂ ∂−∂ ∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂ ∂ ∂ ∂ ∂ ∂ ∂ ∂ =− ξη ξη ηξ ηξ xx yy j yy xx j det 11 (22) where jdet is the determinant of jacobian matrix. to obtain shape functions derivative respect to x, y chair rule can be applied, so: x n x n x n iii ∂ ∂ ∂ ∂ + ∂ ∂ ∂ ∂ = ∂ ∂ η η ξ ξ (23) y n y n y n iii ∂ ∂ ∂ ∂ + ∂ ∂ ∂ ∂ = ∂ ∂ η η ξ ξ (24) the relation between cartesian coordinate and local coordinate may be written as: ηξ ddjdydx det= (25) , 58 i n i sis ub r ∑ = = 1 ε (16) where sib is the shear strain–displacement matrix. the bending and shear stiffness matrices can be written as: dabdbk bb a t b e b ][][][][ ∫= (17) dabdbk ss a t s e s ][][][][ ∫= (18) where ][][ 21 bnbbb bbbb k= , ][][ 21 snsss bbbb k= and e bk ][ and e sk ][ are element bending and shear stiffness matrices, respectively. so the total element matrix is: e s e b e kkk ][][][ += (19) the elemental load vector is also obtained from: daqnf a i e ∫= ][ (20) where q is the uniformly distributed load. the element jacobian matrix is: ⎥ ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂∂∂ ∂∂∂∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂ ∂ ∂ ∂ ∂ ∂ ∂ ∂ = ∑∑ ∑∑ == == n i ii n i ii n i ii n i ii ynxn ynxn yx yx j 11 11 )()( )()( ηη ξξ ηη ξξ (21) and its inverse is: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂ ∂ ∂ ∂− ∂ ∂−∂ ∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂ ∂ ∂ ∂ ∂ ∂ ∂ ∂ =− ξη ξη ηξ ηξ xx yy j yy xx j det 11 (22) where jdet is the determinant of jacobian matrix. to obtain shape functions derivative respect to x, y chair rule can be applied, so: x n x n x n iii ∂ ∂ ∂ ∂ + ∂ ∂ ∂ ∂ = ∂ ∂ η η ξ ξ (23) y n y n y n iii ∂ ∂ ∂ ∂ + ∂ ∂ ∂ ∂ = ∂ ∂ η η ξ ξ (24) the relation between cartesian coordinate and local coordinate may be written as: ηξ ddjdydx det= (25) and 58 i n i sis ub r ∑ = = 1 ε (16) where sib is the shear strain–displacement matrix. the bending and shear stiffness matrices can be written as: dabdbk bb a t b e b ][][][][ ∫= (17) dabdbk ss a t s e s ][][][][ ∫= (18) where ][][ 21 bnbbb bbbb k= , ][][ 21 snsss bbbb k= and e bk ][ and e sk ][ are element bending and shear stiffness matrices, respectively. so the total element matrix is: e s e b e kkk ][][][ += (19) the elemental load vector is also obtained from: daqnf a i e ∫= ][ (20) where q is the uniformly distributed load. the element jacobian matrix is: ⎥ ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂∂∂ ∂∂∂∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂ ∂ ∂ ∂ ∂ ∂ ∂ ∂ = ∑∑ ∑∑ == == n i ii n i ii n i ii n i ii ynxn ynxn yx yx j 11 11 )()( )()( ηη ξξ ηη ξξ (21) and its inverse is: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂ ∂ ∂ ∂− ∂ ∂−∂ ∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂ ∂ ∂ ∂ ∂ ∂ ∂ ∂ =− ξη ξη ηξ ηξ xx yy j yy xx j det 11 (22) where jdet is the determinant of jacobian matrix. to obtain shape functions derivative respect to x, y chair rule can be applied, so: x n x n x n iii ∂ ∂ ∂ ∂ + ∂ ∂ ∂ ∂ = ∂ ∂ η η ξ ξ (23) y n y n y n iii ∂ ∂ ∂ ∂ + ∂ ∂ ∂ ∂ = ∂ ∂ η η ξ ξ (24) the relation between cartesian coordinate and local coordinate may be written as: ηξ ddjdydx det= (25) and 58 i n i sis ub r ∑ = = 1 ε (16) where sib is the shear strain–displacement matrix. the bending and shear stiffness matrices can be written as: dabdbk bb a t b e b ][][][][ ∫= (17) dabdbk ss a t s e s ][][][][ ∫= (18) where ][][ 21 bnbbb bbbb k= , ][][ 21 snsss bbbb k= and e bk ][ and e sk ][ are element bending and shear stiffness matrices, respectively. so the total element matrix is: e s e b e kkk ][][][ += (19) the elemental load vector is also obtained from: daqnf a i e ∫= ][ (20) where q is the uniformly distributed load. the element jacobian matrix is: ⎥ ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂∂∂ ∂∂∂∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂ ∂ ∂ ∂ ∂ ∂ ∂ ∂ = ∑∑ ∑∑ == == n i ii n i ii n i ii n i ii ynxn ynxn yx yx j 11 11 )()( )()( ηη ξξ ηη ξξ (21) and its inverse is: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂ ∂ ∂ ∂− ∂ ∂−∂ ∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂ ∂ ∂ ∂ ∂ ∂ ∂ ∂ =− ξη ξη ηξ ηξ xx yy j yy xx j det 11 (22) where jdet is the determinant of jacobian matrix. to obtain shape functions derivative respect to x, y chair rule can be applied, so: x n x n x n iii ∂ ∂ ∂ ∂ + ∂ ∂ ∂ ∂ = ∂ ∂ η η ξ ξ (23) y n y n y n iii ∂ ∂ ∂ ∂ + ∂ ∂ ∂ ∂ = ∂ ∂ η η ξ ξ (24) the relation between cartesian coordinate and local coordinate may be written as: ηξ ddjdydx det= (25) are element bending and shear stiffness matrices, respectively. so the total element matrix is: 58 i n i sis ub r ∑ = = 1 ε (16) where sib is the shear strain–displacement matrix. the bending and shear stiffness matrices can be written as: dabdbk bb a t b e b ][][][][ ∫= (17) dabdbk ss a t s e s ][][][][ ∫= (18) where ][][ 21 bnbbb bbbb k= , ][][ 21 snsss bbbb k= and e bk ][ and e sk ][ are element bending and shear stiffness matrices, respectively. so the total element matrix is: e s e b e kkk ][][][ += (19) the elemental load vector is also obtained from: daqnf a i e ∫= ][ (20) where q is the uniformly distributed load. the element jacobian matrix is: ⎥ ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂∂∂ ∂∂∂∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂ ∂ ∂ ∂ ∂ ∂ ∂ ∂ = ∑∑ ∑∑ == == n i ii n i ii n i ii n i ii ynxn ynxn yx yx j 11 11 )()( )()( ηη ξξ ηη ξξ (21) and its inverse is: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂ ∂ ∂ ∂− ∂ ∂−∂ ∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂ ∂ ∂ ∂ ∂ ∂ ∂ ∂ =− ξη ξη ηξ ηξ xx yy j yy xx j det 11 (22) where jdet is the determinant of jacobian matrix. to obtain shape functions derivative respect to x, y chair rule can be applied, so: x n x n x n iii ∂ ∂ ∂ ∂ + ∂ ∂ ∂ ∂ = ∂ ∂ η η ξ ξ (23) y n y n y n iii ∂ ∂ ∂ ∂ + ∂ ∂ ∂ ∂ = ∂ ∂ η η ξ ξ (24) the relation between cartesian coordinate and local coordinate may be written as: ηξ ddjdydx det= (25) (19) the elemental load vector is also obtained from: 58 i n i sis ub r ∑ = = 1 ε (16) where sib is the shear strain–displacement matrix. the bending and shear stiffness matrices can be written as: dabdbk bb a t b e b ][][][][ ∫= (17) dabdbk ss a t s e s ][][][][ ∫= (18) where ][][ 21 bnbbb bbbb k= , ][][ 21 snsss bbbb k= and e bk ][ and e sk ][ are element bending and shear stiffness matrices, respectively. so the total element matrix is: e s e b e kkk ][][][ += (19) the elemental load vector is also obtained from: daqnf a i e ∫= ][ (20) where q is the uniformly distributed load. the element jacobian matrix is: ⎥ ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂∂∂ ∂∂∂∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂ ∂ ∂ ∂ ∂ ∂ ∂ ∂ = ∑∑ ∑∑ == == n i ii n i ii n i ii n i ii ynxn ynxn yx yx j 11 11 )()( )()( ηη ξξ ηη ξξ (21) and its inverse is: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂ ∂ ∂ ∂− ∂ ∂−∂ ∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂ ∂ ∂ ∂ ∂ ∂ ∂ ∂ =− ξη ξη ηξ ηξ xx yy j yy xx j det 11 (22) where jdet is the determinant of jacobian matrix. to obtain shape functions derivative respect to x, y chair rule can be applied, so: x n x n x n iii ∂ ∂ ∂ ∂ + ∂ ∂ ∂ ∂ = ∂ ∂ η η ξ ξ (23) y n y n y n iii ∂ ∂ ∂ ∂ + ∂ ∂ ∂ ∂ = ∂ ∂ η η ξ ξ (24) the relation between cartesian coordinate and local coordinate may be written as: ηξ ddjdydx det= (25) (20) where 58 i n i sis ub r ∑ = = 1 ε (16) where sib is the shear strain–displacement matrix. the bending and shear stiffness matrices can be written as: dabdbk bb a t b e b ][][][][ ∫= (17) dabdbk ss a t s e s ][][][][ ∫= (18) where ][][ 21 bnbbb bbbb k= , ][][ 21 snsss bbbb k= and e bk ][ and e sk ][ are element bending and shear stiffness matrices, respectively. so the total element matrix is: e s e b e kkk ][][][ += (19) the elemental load vector is also obtained from: daqnf a i e ∫= ][ (20) where q is the uniformly distributed load. the element jacobian matrix is: ⎥ ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂∂∂ ∂∂∂∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂ ∂ ∂ ∂ ∂ ∂ ∂ ∂ = ∑∑ ∑∑ == == n i ii n i ii n i ii n i ii ynxn ynxn yx yx j 11 11 )()( )()( ηη ξξ ηη ξξ (21) and its inverse is: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂ ∂ ∂ ∂− ∂ ∂−∂ ∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂ ∂ ∂ ∂ ∂ ∂ ∂ ∂ =− ξη ξη ηξ ηξ xx yy j yy xx j det 11 (22) where jdet is the determinant of jacobian matrix. to obtain shape functions derivative respect to x, y chair rule can be applied, so: x n x n x n iii ∂ ∂ ∂ ∂ + ∂ ∂ ∂ ∂ = ∂ ∂ η η ξ ξ (23) y n y n y n iii ∂ ∂ ∂ ∂ + ∂ ∂ ∂ ∂ = ∂ ∂ η η ξ ξ (24) the relation between cartesian coordinate and local coordinate may be written as: ηξ ddjdydx det= (25) is the uniformly distributed load. the element jacobian matrix is: 58 i n i sis ub r ∑ = = 1 ε (16) where sib is the shear strain–displacement matrix. the bending and shear stiffness matrices can be written as: dabdbk bb a t b e b ][][][][ ∫= (17) dabdbk ss a t s e s ][][][][ ∫= (18) where ][][ 21 bnbbb bbbb k= , ][][ 21 snsss bbbb k= and e bk ][ and e sk ][ are element bending and shear stiffness matrices, respectively. so the total element matrix is: e s e b e kkk ][][][ += (19) the elemental load vector is also obtained from: daqnf a i e ∫= ][ (20) where q is the uniformly distributed load. the element jacobian matrix is: ⎥ ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂∂∂ ∂∂∂∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂ ∂ ∂ ∂ ∂ ∂ ∂ ∂ = ∑∑ ∑∑ == == n i ii n i ii n i ii n i ii ynxn ynxn yx yx j 11 11 )()( )()( ηη ξξ ηη ξξ (21) and its inverse is: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂ ∂ ∂ ∂− ∂ ∂−∂ ∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂ ∂ ∂ ∂ ∂ ∂ ∂ ∂ =− ξη ξη ηξ ηξ xx yy j yy xx j det 11 (22) where jdet is the determinant of jacobian matrix. to obtain shape functions derivative respect to x, y chair rule can be applied, so: x n x n x n iii ∂ ∂ ∂ ∂ + ∂ ∂ ∂ ∂ = ∂ ∂ η η ξ ξ (23) y n y n y n iii ∂ ∂ ∂ ∂ + ∂ ∂ ∂ ∂ = ∂ ∂ η η ξ ξ (24) the relation between cartesian coordinate and local coordinate may be written as: ηξ ddjdydx det= (25) (21) and its inverse is: 58 i n i sis ub r ∑ = = 1 ε (16) where sib is the shear strain–displacement matrix. the bending and shear stiffness matrices can be written as: dabdbk bb a t b e b ][][][][ ∫= (17) dabdbk ss a t s e s ][][][][ ∫= (18) where ][][ 21 bnbbb bbbb k= , ][][ 21 snsss bbbb k= and e bk ][ and e sk ][ are element bending and shear stiffness matrices, respectively. so the total element matrix is: e s e b e kkk ][][][ += (19) the elemental load vector is also obtained from: daqnf a i e ∫= ][ (20) where q is the uniformly distributed load. the element jacobian matrix is: ⎥ ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂∂∂ ∂∂∂∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂ ∂ ∂ ∂ ∂ ∂ ∂ ∂ = ∑∑ ∑∑ == == n i ii n i ii n i ii n i ii ynxn ynxn yx yx j 11 11 )()( )()( ηη ξξ ηη ξξ (21) and its inverse is: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂ ∂ ∂ ∂− ∂ ∂−∂ ∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂ ∂ ∂ ∂ ∂ ∂ ∂ ∂ =− ξη ξη ηξ ηξ xx yy j yy xx j det 11 (22) where jdet is the determinant of jacobian matrix. to obtain shape functions derivative respect to x, y chair rule can be applied, so: x n x n x n iii ∂ ∂ ∂ ∂ + ∂ ∂ ∂ ∂ = ∂ ∂ η η ξ ξ (23) y n y n y n iii ∂ ∂ ∂ ∂ + ∂ ∂ ∂ ∂ = ∂ ∂ η η ξ ξ (24) the relation between cartesian coordinate and local coordinate may be written as: ηξ ddjdydx det= (25) (22) issn: 2180-1053 vol. 4 no. 1 january-june 2012 journal of mechanical engineering and technology 70 where 58 i n i sis ub r ∑ = = 1 ε (16) where sib is the shear strain–displacement matrix. the bending and shear stiffness matrices can be written as: dabdbk bb a t b e b ][][][][ ∫= (17) dabdbk ss a t s e s ][][][][ ∫= (18) where ][][ 21 bnbbb bbbb k= , ][][ 21 snsss bbbb k= and e bk ][ and e sk ][ are element bending and shear stiffness matrices, respectively. so the total element matrix is: e s e b e kkk ][][][ += (19) the elemental load vector is also obtained from: daqnf a i e ∫= ][ (20) where q is the uniformly distributed load. the element jacobian matrix is: ⎥ ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂∂∂ ∂∂∂∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂ ∂ ∂ ∂ ∂ ∂ ∂ ∂ = ∑∑ ∑∑ == == n i ii n i ii n i ii n i ii ynxn ynxn yx yx j 11 11 )()( )()( ηη ξξ ηη ξξ (21) and its inverse is: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂ ∂ ∂ ∂− ∂ ∂−∂ ∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂ ∂ ∂ ∂ ∂ ∂ ∂ ∂ =− ξη ξη ηξ ηξ xx yy j yy xx j det 11 (22) where jdet is the determinant of jacobian matrix. to obtain shape functions derivative respect to x, y chair rule can be applied, so: x n x n x n iii ∂ ∂ ∂ ∂ + ∂ ∂ ∂ ∂ = ∂ ∂ η η ξ ξ (23) y n y n y n iii ∂ ∂ ∂ ∂ + ∂ ∂ ∂ ∂ = ∂ ∂ η η ξ ξ (24) the relation between cartesian coordinate and local coordinate may be written as: ηξ ddjdydx det= (25) is the determinant of jacobian matrix. to obtain shape functions derivative respect to x, y chair rule can be applied, so: 58 i n i sis ub r ∑ = = 1 ε (16) where sib is the shear strain–displacement matrix. the bending and shear stiffness matrices can be written as: dabdbk bb a t b e b ][][][][ ∫= (17) dabdbk ss a t s e s ][][][][ ∫= (18) where ][][ 21 bnbbb bbbb k= , ][][ 21 snsss bbbb k= and e bk ][ and e sk ][ are element bending and shear stiffness matrices, respectively. so the total element matrix is: e s e b e kkk ][][][ += (19) the elemental load vector is also obtained from: daqnf a i e ∫= ][ (20) where q is the uniformly distributed load. the element jacobian matrix is: ⎥ ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂∂∂ ∂∂∂∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂ ∂ ∂ ∂ ∂ ∂ ∂ ∂ = ∑∑ ∑∑ == == n i ii n i ii n i ii n i ii ynxn ynxn yx yx j 11 11 )()( )()( ηη ξξ ηη ξξ (21) and its inverse is: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂ ∂ ∂ ∂− ∂ ∂−∂ ∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂ ∂ ∂ ∂ ∂ ∂ ∂ ∂ =− ξη ξη ηξ ηξ xx yy j yy xx j det 11 (22) where jdet is the determinant of jacobian matrix. to obtain shape functions derivative respect to x, y chair rule can be applied, so: x n x n x n iii ∂ ∂ ∂ ∂ + ∂ ∂ ∂ ∂ = ∂ ∂ η η ξ ξ (23) y n y n y n iii ∂ ∂ ∂ ∂ + ∂ ∂ ∂ ∂ = ∂ ∂ η η ξ ξ (24) the relation between cartesian coordinate and local coordinate may be written as: ηξ ddjdydx det= (25) (23) 58 i n i sis ub r ∑ = = 1 ε (16) where sib is the shear strain–displacement matrix. the bending and shear stiffness matrices can be written as: dabdbk bb a t b e b ][][][][ ∫= (17) dabdbk ss a t s e s ][][][][ ∫= (18) where ][][ 21 bnbbb bbbb k= , ][][ 21 snsss bbbb k= and e bk ][ and e sk ][ are element bending and shear stiffness matrices, respectively. so the total element matrix is: e s e b e kkk ][][][ += (19) the elemental load vector is also obtained from: daqnf a i e ∫= ][ (20) where q is the uniformly distributed load. the element jacobian matrix is: ⎥ ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂∂∂ ∂∂∂∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂ ∂ ∂ ∂ ∂ ∂ ∂ ∂ = ∑∑ ∑∑ == == n i ii n i ii n i ii n i ii ynxn ynxn yx yx j 11 11 )()( )()( ηη ξξ ηη ξξ (21) and its inverse is: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂ ∂ ∂ ∂− ∂ ∂−∂ ∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂ ∂ ∂ ∂ ∂ ∂ ∂ ∂ =− ξη ξη ηξ ηξ xx yy j yy xx j det 11 (22) where jdet is the determinant of jacobian matrix. to obtain shape functions derivative respect to x, y chair rule can be applied, so: x n x n x n iii ∂ ∂ ∂ ∂ + ∂ ∂ ∂ ∂ = ∂ ∂ η η ξ ξ (23) y n y n y n iii ∂ ∂ ∂ ∂ + ∂ ∂ ∂ ∂ = ∂ ∂ η η ξ ξ (24) the relation between cartesian coordinate and local coordinate may be written as: ηξ ddjdydx det= (25) (24) the relation between cartesian coordinate and local coordinate may be written as: 58 i n i sis ub r ∑ = = 1 ε (16) where sib is the shear strain–displacement matrix. the bending and shear stiffness matrices can be written as: dabdbk bb a t b e b ][][][][ ∫= (17) dabdbk ss a t s e s ][][][][ ∫= (18) where ][][ 21 bnbbb bbbb k= , ][][ 21 snsss bbbb k= and e bk ][ and e sk ][ are element bending and shear stiffness matrices, respectively. so the total element matrix is: e s e b e kkk ][][][ += (19) the elemental load vector is also obtained from: daqnf a i e ∫= ][ (20) where q is the uniformly distributed load. the element jacobian matrix is: ⎥ ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎢ ⎣ ⎡ ∂∂∂∂ ∂∂∂∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂ ∂ ∂ ∂ ∂ ∂ ∂ ∂ = ∑∑ ∑∑ == == n i ii n i ii n i ii n i ii ynxn ynxn yx yx j 11 11 )()( )()( ηη ξξ ηη ξξ (21) and its inverse is: ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂ ∂ ∂ ∂− ∂ ∂−∂ ∂ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ∂ ∂ ∂ ∂ ∂ ∂ ∂ ∂ =− ξη ξη ηξ ηξ xx yy j yy xx j det 11 (22) where jdet is the determinant of jacobian matrix. to obtain shape functions derivative respect to x, y chair rule can be applied, so: x n x n x n iii ∂ ∂ ∂ ∂ + ∂ ∂ ∂ ∂ = ∂ ∂ η η ξ ξ (23) y n y n y n iii ∂ ∂ ∂ ∂ + ∂ ∂ ∂ ∂ = ∂ ∂ η η ξ ξ (24) the relation between cartesian coordinate and local coordinate may be written as: ηξ ddjdydx det= (25) (25) substituting equation (25) into equation (17) and equation (18), we can obtain: 59 substituting equation (25) into equation (17) and equation (18), we can obtain: ∫ ∫ + − + − = 1 1 1 1 det][][][][ ηξ ddjbdbk bb t b e b (26) ∫ ∫ + − + − = 1 1 1 1 det][][][][ ηξ ddjbdbk ss t s e s (27) 4.5. shear locking plate finite elements based on mindlin theory require only c0 continuity for displacement and independent rotations, unlike kirchhoff theory. therefore, the behavior of the mindlin plate elements is usually very good for a moderately thick plate situation. however, when a thin plate is considered, these displacement–based elements cause a problem known as ‘‘shear locking’’. when the full integration of the stiffness matrices is used with standard mindlin finite elements, very stiff results may be obtained in application to thin plates. this means that the bending energy, which should dominate the shear terms, will be incorrectly estimated to be zero in thin plate problems. to avoid the shear locking problem in thin plates, the reduced and selective reduced integration techniques were proposed in early 1970s [17, 18]. the reduced integration procedure is the reduction in the order of integration in computing the stiffness matrix of the finite element. similarly, the selective integration procedure is also a kind of reduced integration rule which is used to evaluate the stiffness matrix associated with the shear strain energy. that is to say, this has been adopted to the shear stiffness matrix only and full integration is used on the remaining terms [19]. therefore, the ek ][ element stiffness matrix can be obtained by separating into bending terms and shear terms. with these definitions, ek ][ element stiffness matrix is given by simplifying the equation (19). table 1 shows the full, reduced, and selective reduced integration rules used to test the shear locking response of the bilinear, serendipity and lagrange elements in this paper. 4.6. plasticity in the mindlin plate relations, yield function can be written as a function of bending and shear moments [16]. in this case, we can assume that the whole section of the plate be plastic. then, we use a yield criterion expressed in terms of bending and shear moments, similar to the iliushin’s yield function [20]. the iliushin’s yield function f can be written as: 0 )( 2 0 2 0 2 =−= σ ky m m f (28) where m are the stress intensities given by: 2222 3 xyyxyx mmmmmm +−+= (29) (26) 59 substituting equation (25) into equation (17) and equation (18), we can obtain: ∫ ∫ + − + − = 1 1 1 1 det][][][][ ηξ ddjbdbk bb t b e b (26) ∫ ∫ + − + − = 1 1 1 1 det][][][][ ηξ ddjbdbk ss t s e s (27) 4.5. shear locking plate finite elements based on mindlin theory require only c0 continuity for displacement and independent rotations, unlike kirchhoff theory. therefore, the behavior of the mindlin plate elements is usually very good for a moderately thick plate situation. however, when a thin plate is considered, these displacement–based elements cause a problem known as ‘‘shear locking’’. when the full integration of the stiffness matrices is used with standard mindlin finite elements, very stiff results may be obtained in application to thin plates. this means that the bending energy, which should dominate the shear terms, will be incorrectly estimated to be zero in thin plate problems. to avoid the shear locking problem in thin plates, the reduced and selective reduced integration techniques were proposed in early 1970s [17, 18]. the reduced integration procedure is the reduction in the order of integration in computing the stiffness matrix of the finite element. similarly, the selective integration procedure is also a kind of reduced integration rule which is used to evaluate the stiffness matrix associated with the shear strain energy. that is to say, this has been adopted to the shear stiffness matrix only and full integration is used on the remaining terms [19]. therefore, the ek ][ element stiffness matrix can be obtained by separating into bending terms and shear terms. with these definitions, ek ][ element stiffness matrix is given by simplifying the equation (19). table 1 shows the full, reduced, and selective reduced integration rules used to test the shear locking response of the bilinear, serendipity and lagrange elements in this paper. 4.6. plasticity in the mindlin plate relations, yield function can be written as a function of bending and shear moments [16]. in this case, we can assume that the whole section of the plate be plastic. then, we use a yield criterion expressed in terms of bending and shear moments, similar to the iliushin’s yield function [20]. the iliushin’s yield function f can be written as: 0 )( 2 0 2 0 2 =−= σ ky m m f (28) where m are the stress intensities given by: 2222 3 xyyxyx mmmmmm +−+= (29) (27) 4.5. shear locking plate finite elements based on mindlin theory require only c0 continuity for displacement and independent rotations, unlike kirchhoff theory. therefore, the behavior of the mindlin plate elements is usually very good for a moderately thick plate situation. however, when a thin plate is considered, these displacement–based elements cause a problem known as ‘‘shear locking’’. when the full integration of the stiffness matrices is used with standard mindlin finite elements, very stiff results may be obtained in application to thin plates. this means that the bending energy, which should dominate the shear terms, will be incorrectly estimated to be zero in thin plate problems. to avoid the shear locking problem in thin plates, the reduced and selective reduced integration techniques were proposed in early 1970s [17, 18]. the reduced integration procedure is the reduction in issn: 2180-1053 vol. 4 no. 1 january-june 2012 implementation of nonlinear finite element using object–oriented design patterns 71 the order of integration in computing the stiffness matrix of the finite element. similarly, the selective integration procedure is also a kind of reduced integration rule which is used to evaluate the stiffness matrix associated with the shear strain energy. that is to say, this has been adopted to the shear stiffness matrix only and full integration is used on the remaining terms [19]. therefore, the 59 substituting equation (25) into equation (17) and equation (18), we can obtain: ∫ ∫ + − + − = 1 1 1 1 det][][][][ ηξ ddjbdbk bb t b e b (26) ∫ ∫ + − + − = 1 1 1 1 det][][][][ ηξ ddjbdbk ss t s e s (27) 4.5. shear locking plate finite elements based on mindlin theory require only c0 continuity for displacement and independent rotations, unlike kirchhoff theory. therefore, the behavior of the mindlin plate elements is usually very good for a moderately thick plate situation. however, when a thin plate is considered, these displacement–based elements cause a problem known as ‘‘shear locking’’. when the full integration of the stiffness matrices is used with standard mindlin finite elements, very stiff results may be obtained in application to thin plates. this means that the bending energy, which should dominate the shear terms, will be incorrectly estimated to be zero in thin plate problems. to avoid the shear locking problem in thin plates, the reduced and selective reduced integration techniques were proposed in early 1970s [17, 18]. the reduced integration procedure is the reduction in the order of integration in computing the stiffness matrix of the finite element. similarly, the selective integration procedure is also a kind of reduced integration rule which is used to evaluate the stiffness matrix associated with the shear strain energy. that is to say, this has been adopted to the shear stiffness matrix only and full integration is used on the remaining terms [19]. therefore, the ek ][ element stiffness matrix can be obtained by separating into bending terms and shear terms. with these definitions, ek ][ element stiffness matrix is given by simplifying the equation (19). table 1 shows the full, reduced, and selective reduced integration rules used to test the shear locking response of the bilinear, serendipity and lagrange elements in this paper. 4.6. plasticity in the mindlin plate relations, yield function can be written as a function of bending and shear moments [16]. in this case, we can assume that the whole section of the plate be plastic. then, we use a yield criterion expressed in terms of bending and shear moments, similar to the iliushin’s yield function [20]. the iliushin’s yield function f can be written as: 0 )( 2 0 2 0 2 =−= σ ky m m f (28) where m are the stress intensities given by: 2222 3 xyyxyx mmmmmm +−+= (29) element stiffness matrix can be obtained by separating into bending terms and shear terms. with these definitions, 59 substituting equation (25) into equation (17) and equation (18), we can obtain: ∫ ∫ + − + − = 1 1 1 1 det][][][][ ηξ ddjbdbk bb t b e b (26) ∫ ∫ + − + − = 1 1 1 1 det][][][][ ηξ ddjbdbk ss t s e s (27) 4.5. shear locking plate finite elements based on mindlin theory require only c0 continuity for displacement and independent rotations, unlike kirchhoff theory. therefore, the behavior of the mindlin plate elements is usually very good for a moderately thick plate situation. however, when a thin plate is considered, these displacement–based elements cause a problem known as ‘‘shear locking’’. when the full integration of the stiffness matrices is used with standard mindlin finite elements, very stiff results may be obtained in application to thin plates. this means that the bending energy, which should dominate the shear terms, will be incorrectly estimated to be zero in thin plate problems. to avoid the shear locking problem in thin plates, the reduced and selective reduced integration techniques were proposed in early 1970s [17, 18]. the reduced integration procedure is the reduction in the order of integration in computing the stiffness matrix of the finite element. similarly, the selective integration procedure is also a kind of reduced integration rule which is used to evaluate the stiffness matrix associated with the shear strain energy. that is to say, this has been adopted to the shear stiffness matrix only and full integration is used on the remaining terms [19]. therefore, the ek ][ element stiffness matrix can be obtained by separating into bending terms and shear terms. with these definitions, ek ][ element stiffness matrix is given by simplifying the equation (19). table 1 shows the full, reduced, and selective reduced integration rules used to test the shear locking response of the bilinear, serendipity and lagrange elements in this paper. 4.6. plasticity in the mindlin plate relations, yield function can be written as a function of bending and shear moments [16]. in this case, we can assume that the whole section of the plate be plastic. then, we use a yield criterion expressed in terms of bending and shear moments, similar to the iliushin’s yield function [20]. the iliushin’s yield function f can be written as: 0 )( 2 0 2 0 2 =−= σ ky m m f (28) where m are the stress intensities given by: 2222 3 xyyxyx mmmmmm +−+= (29) element stiffness matrix is given by simplifying the equation (19). table 1 shows the full, reduced, and selective reduced integration rules used to test the shear locking response of the bilinear, serendipity and lagrange elements in this paper. 4.6. plasticity in the mindlin plate relations, yield function can be written as a function of bending and shear moments [16]. in this case, we can assume that the whole section of the plate be plastic. then, we use a yield criterion expressed in terms of bending and shear moments, similar to the iliushin’s yield function [20]. the iliushin’s yield function f can be written as: 59 substituting equation (25) into equation (17) and equation (18), we can obtain: ∫ ∫ + − + − = 1 1 1 1 det][][][][ ηξ ddjbdbk bb t b e b (26) ∫ ∫ + − + − = 1 1 1 1 det][][][][ ηξ ddjbdbk ss t s e s (27) 4.5. shear locking plate finite elements based on mindlin theory require only c0 continuity for displacement and independent rotations, unlike kirchhoff theory. therefore, the behavior of the mindlin plate elements is usually very good for a moderately thick plate situation. however, when a thin plate is considered, these displacement–based elements cause a problem known as ‘‘shear locking’’. when the full integration of the stiffness matrices is used with standard mindlin finite elements, very stiff results may be obtained in application to thin plates. this means that the bending energy, which should dominate the shear terms, will be incorrectly estimated to be zero in thin plate problems. to avoid the shear locking problem in thin plates, the reduced and selective reduced integration techniques were proposed in early 1970s [17, 18]. the reduced integration procedure is the reduction in the order of integration in computing the stiffness matrix of the finite element. similarly, the selective integration procedure is also a kind of reduced integration rule which is used to evaluate the stiffness matrix associated with the shear strain energy. that is to say, this has been adopted to the shear stiffness matrix only and full integration is used on the remaining terms [19]. therefore, the ek ][ element stiffness matrix can be obtained by separating into bending terms and shear terms. with these definitions, ek ][ element stiffness matrix is given by simplifying the equation (19). table 1 shows the full, reduced, and selective reduced integration rules used to test the shear locking response of the bilinear, serendipity and lagrange elements in this paper. 4.6. plasticity in the mindlin plate relations, yield function can be written as a function of bending and shear moments [16]. in this case, we can assume that the whole section of the plate be plastic. then, we use a yield criterion expressed in terms of bending and shear moments, similar to the iliushin’s yield function [20]. the iliushin’s yield function f can be written as: 0 )( 2 0 2 0 2 =−= σ ky m m f (28) where m are the stress intensities given by: 2222 3 xyyxyx mmmmmm +−+= (29) (28) where m are the stress intensities given by: 59 substituting equation (25) into equation (17) and equation (18), we can obtain: ∫ ∫ + − + − = 1 1 1 1 det][][][][ ηξ ddjbdbk bb t b e b (26) ∫ ∫ + − + − = 1 1 1 1 det][][][][ ηξ ddjbdbk ss t s e s (27) 4.5. shear locking plate finite elements based on mindlin theory require only c0 continuity for displacement and independent rotations, unlike kirchhoff theory. therefore, the behavior of the mindlin plate elements is usually very good for a moderately thick plate situation. however, when a thin plate is considered, these displacement–based elements cause a problem known as ‘‘shear locking’’. when the full integration of the stiffness matrices is used with standard mindlin finite elements, very stiff results may be obtained in application to thin plates. this means that the bending energy, which should dominate the shear terms, will be incorrectly estimated to be zero in thin plate problems. to avoid the shear locking problem in thin plates, the reduced and selective reduced integration techniques were proposed in early 1970s [17, 18]. the reduced integration procedure is the reduction in the order of integration in computing the stiffness matrix of the finite element. similarly, the selective integration procedure is also a kind of reduced integration rule which is used to evaluate the stiffness matrix associated with the shear strain energy. that is to say, this has been adopted to the shear stiffness matrix only and full integration is used on the remaining terms [19]. therefore, the ek ][ element stiffness matrix can be obtained by separating into bending terms and shear terms. with these definitions, ek ][ element stiffness matrix is given by simplifying the equation (19). table 1 shows the full, reduced, and selective reduced integration rules used to test the shear locking response of the bilinear, serendipity and lagrange elements in this paper. 4.6. plasticity in the mindlin plate relations, yield function can be written as a function of bending and shear moments [16]. in this case, we can assume that the whole section of the plate be plastic. then, we use a yield criterion expressed in terms of bending and shear moments, similar to the iliushin’s yield function [20]. the iliushin’s yield function f can be written as: 0 )( 2 0 2 0 2 =−= σ ky m m f (28) where m are the stress intensities given by: 2222 3 xyyxyx mmmmmm +−+= (29) (29) and 60 and 0m is the moment capacity of the cross section. when the cross section is fully plastic, and the moment capacity of the cross section given by: 4 2 0 0 h m σ = (30) the symbol 0σ is the uniaxial yield stress; )(ky is a material parameter, which depends on the isotropic hardening parameter k and h is the thickness of the plate. mx, my and mxy are stress couples defined by (6). 5.0 application in this study, three design patterns are used to increase the reusability and extensibility of the nonlinear finite element program. for this aim, the elastic–plastic analysis of plates based on mindlin theory is selected. as shown in section 4.5, to overcome the shear locking problem, the element stiffness matrix is divided in two terms and for each term, different integration method is used. in this paper three kinds of elements which have 4, 8 and 9 nodes, namely, bilinear, serendipity and lagrange, respectively (see figure. 3), and three integration methods are used to analysis the mindlin's plate (see table 1). firstly, model–analysis separation is used to decompose the finite element program to model and analysis packages. there are essentially two stages in finite element analysis. the first stage involves modeling the problem domain. the second stage involves analyzing the finite element model. it is natural therefore to decompose a finite element program into two major subsystems, one for modeling and the other for analysis. model classes represent finite element entities such as elements, nodes, and degrees–of–freedom (d.o.f.). the analysis subsystem is responsible for forming and solving the system of equations. the two subsystems should be loosely coupled. this means minimizing dependencies across subsystem boundaries. in a procedural code, using many kinds of element with different number of nodes and d.o.fs persuade the programmer to manipulate the analysis section. but in an object–oriented program, by using the model–analysis separation pattern, programmer is able to change the element without manipulation of the analysis packages. on the other hand, to implement the different integration method, programmer changes just the analysis packages. is the moment capacity of the cross section. when the cross section is fully plastic, and the moment capacity of the cross section given by: 60 and 0m is the moment capacity of the cross section. when the cross section is fully plastic, and the moment capacity of the cross section given by: 4 2 0 0 h m σ = (30) the symbol 0σ is the uniaxial yield stress; )(ky is a material parameter, which depends on the isotropic hardening parameter k and h is the thickness of the plate. mx, my and mxy are stress couples defined by (6). 5.0 application in this study, three design patterns are used to increase the reusability and extensibility of the nonlinear finite element program. for this aim, the elastic–plastic analysis of plates based on mindlin theory is selected. as shown in section 4.5, to overcome the shear locking problem, the element stiffness matrix is divided in two terms and for each term, different integration method is used. in this paper three kinds of elements which have 4, 8 and 9 nodes, namely, bilinear, serendipity and lagrange, respectively (see figure. 3), and three integration methods are used to analysis the mindlin's plate (see table 1). firstly, model–analysis separation is used to decompose the finite element program to model and analysis packages. there are essentially two stages in finite element analysis. the first stage involves modeling the problem domain. the second stage involves analyzing the finite element model. it is natural therefore to decompose a finite element program into two major subsystems, one for modeling and the other for analysis. model classes represent finite element entities such as elements, nodes, and degrees–of–freedom (d.o.f.). the analysis subsystem is responsible for forming and solving the system of equations. the two subsystems should be loosely coupled. this means minimizing dependencies across subsystem boundaries. in a procedural code, using many kinds of element with different number of nodes and d.o.fs persuade the programmer to manipulate the analysis section. but in an object–oriented program, by using the model–analysis separation pattern, programmer is able to change the element without manipulation of the analysis packages. on the other hand, to implement the different integration method, programmer changes just the analysis packages. (30) the symbol 60 and 0m is the moment capacity of the cross section. when the cross section is fully plastic, and the moment capacity of the cross section given by: 4 2 0 0 h m σ = (30) the symbol 0σ is the uniaxial yield stress; )(ky is a material parameter, which depends on the isotropic hardening parameter k and h is the thickness of the plate. mx, my and mxy are stress couples defined by (6). 5.0 application in this study, three design patterns are used to increase the reusability and extensibility of the nonlinear finite element program. for this aim, the elastic–plastic analysis of plates based on mindlin theory is selected. as shown in section 4.5, to overcome the shear locking problem, the element stiffness matrix is divided in two terms and for each term, different integration method is used. in this paper three kinds of elements which have 4, 8 and 9 nodes, namely, bilinear, serendipity and lagrange, respectively (see figure. 3), and three integration methods are used to analysis the mindlin's plate (see table 1). firstly, model–analysis separation is used to decompose the finite element program to model and analysis packages. there are essentially two stages in finite element analysis. the first stage involves modeling the problem domain. the second stage involves analyzing the finite element model. it is natural therefore to decompose a finite element program into two major subsystems, one for modeling and the other for analysis. model classes represent finite element entities such as elements, nodes, and degrees–of–freedom (d.o.f.). the analysis subsystem is responsible for forming and solving the system of equations. the two subsystems should be loosely coupled. this means minimizing dependencies across subsystem boundaries. in a procedural code, using many kinds of element with different number of nodes and d.o.fs persuade the programmer to manipulate the analysis section. but in an object–oriented program, by using the model–analysis separation pattern, programmer is able to change the element without manipulation of the analysis packages. on the other hand, to implement the different integration method, programmer changes just the analysis packages. is the uniaxial yield stress; 60 and 0m is the moment capacity of the cross section. when the cross section is fully plastic, and the moment capacity of the cross section given by: 4 2 0 0 h m σ = (30) the symbol 0σ is the uniaxial yield stress; )(ky is a material parameter, which depends on the isotropic hardening parameter k and h is the thickness of the plate. mx, my and mxy are stress couples defined by (6). 5.0 application in this study, three design patterns are used to increase the reusability and extensibility of the nonlinear finite element program. for this aim, the elastic–plastic analysis of plates based on mindlin theory is selected. as shown in section 4.5, to overcome the shear locking problem, the element stiffness matrix is divided in two terms and for each term, different integration method is used. in this paper three kinds of elements which have 4, 8 and 9 nodes, namely, bilinear, serendipity and lagrange, respectively (see figure. 3), and three integration methods are used to analysis the mindlin's plate (see table 1). firstly, model–analysis separation is used to decompose the finite element program to model and analysis packages. there are essentially two stages in finite element analysis. the first stage involves modeling the problem domain. the second stage involves analyzing the finite element model. it is natural therefore to decompose a finite element program into two major subsystems, one for modeling and the other for analysis. model classes represent finite element entities such as elements, nodes, and degrees–of–freedom (d.o.f.). the analysis subsystem is responsible for forming and solving the system of equations. the two subsystems should be loosely coupled. this means minimizing dependencies across subsystem boundaries. in a procedural code, using many kinds of element with different number of nodes and d.o.fs persuade the programmer to manipulate the analysis section. but in an object–oriented program, by using the model–analysis separation pattern, programmer is able to change the element without manipulation of the analysis packages. on the other hand, to implement the different integration method, programmer changes just the analysis packages. is a material parameter, which depends on the isotropic hardening parameter k and h is the thickness of the plate. mx, my and mxy are stress couples defined by (6). issn: 2180-1053 vol. 4 no. 1 january-june 2012 journal of mechanical engineering and technology 72 5.0 application in this study, three design patterns are used to increase the reusability and extensibility of the nonlinear finite element program. for this aim, the elastic–plastic analysis of plates based on mindlin theory is selected. as shown in section 4.5, to overcome the shear locking problem, the element stiffness matrix is divided in two terms and for each term, different integration method is used. in this paper three kinds of elements which have 4, 8 and 9 nodes, namely, bilinear, serendipity and lagrange, respectively (see figure. 3), and three integration methods are used to analysis the mindlin's plate (see table 1). firstly, model– analysis separation is used to decompose the finite element program to model and analysis packages. there are essentially two stages in finite element analysis. the first stage involves modeling the problem domain. the second stage involves analyzing the finite element model. it is natural therefore to decompose a finite element program into two major subsystems, one for modeling and the other for analysis. model classes represent finite element entities such as elements, nodes, and degrees–of–freedom (d.o.f.). the analysis subsystem is responsible for forming and solving the system of equations. the two subsystems should be loosely coupled. this means minimizing dependencies across subsystem boundaries. in a procedural code, using many kinds of element with different number of nodes and d.o.fs persuade the programmer to manipulate the analysis section. but in an object– oriented program, by using the model–analysis separation pattern, programmer is able to change the element without manipulation of the analysis packages. on the other hand, to implement the different integration method, programmer changes just the analysis packages. 61 ( )( )ηηξξ iiin ++= 114 1 ( )( )ηηξξ iii n +−== 112 1 ,0 2 ( )( )211 2 1 ,0 ηξξη −+== iii n ( )( )ηηξξ iii n +−== 112 1 ,0 2 ( )( )211 2 1 ,0 ηξξη −+== iii n ( )( )229 112 1 ηξ −−=n ( )( )( )111 4 1 −+++= ηηξξηηξξ iiiiin ( )( )( )1114 1 −+++= ηηξξηηξξ iiiiin figure. 3 the finite elements and the shape functions (a) bilinear (b) quadratic serendipity (c) quadratic lagrange. table 1 integration rules. [ ]ebk [ ]esk[ ]ebk [ ] e sk[ ] e bk [ ] e sk figure. 4 shows the packages participating in this pattern and their dependencies. the model package contains model classes, while the calculation and solvers packages together form the analysis subsystem. calculation classes represent different types of analysis. the solvers package consists of mathematical classes for solving system equations. there is no coupling between solvers and model. figure. 3 the finite elements and the shape functions (a) bilinear (b) quadratic serendipity (c) quadratic lagrange. issn: 2180-1053 vol. 4 no. 1 january-june 2012 implementation of nonlinear finite element using object–oriented design patterns 73 table 1 integration rules. 61 ( )( )ηηξξ iiin ++= 114 1 ( )( )ηηξξ iii n +−== 112 1 ,0 2 ( )( )211 2 1 ,0 ηξξη −+== iii n ( )( )ηηξξ iii n +−== 112 1 ,0 2 ( )( )211 2 1 ,0 ηξξη −+== iii n ( )( )229 112 1 ηξ −−=n ( )( )( )111 4 1 −+++= ηηξξηηξξ iiiiin ( )( )( )1114 1 −+++= ηηξξηηξξ iiiiin figure. 3 the finite elements and the shape functions (a) bilinear (b) quadratic serendipity (c) quadratic lagrange. table 1 integration rules. [ ]ebk [ ]esk[ ]ebk [ ] e sk[ ] e bk [ ] e sk figure. 4 shows the packages participating in this pattern and their dependencies. the model package contains model classes, while the calculation and solvers packages together form the analysis subsystem. calculation classes represent different types of analysis. the solvers package consists of mathematical classes for solving system equations. there is no coupling between solvers and model. figure. 4 shows the packages participating in this pattern and their dependencies. the model package contains model classes, while the calculation and solvers packages together form the analysis subsystem. calculation classes represent different types of analysis. the solvers package consists of mathematical classes for solving system equations. there is no coupling between solvers and model. 62 figure. 4 packages in the model–analysis separation pattern. since for each element, numbers of nodes are different, fem classes are also different. but the ui models of elements are the same. therefore, the ui–model separation pattern is used. ui–related responsibilities should be assigned to ui objects. ui classes should be grouped together in a subsystem that is dependent on the model subsystem. this allows the more volatile ui subsystem to be changed without affecting model classes. naturally, there should be no coupling between the analysis and ui subsystems. figure. 5 shows the dependencies between the ui, mesh and fe packages. ui contains classes such as struct, curve, and point. a struct represents a (sub) domain of the finite element model. the curve of a struct describes its boundary. each curve is in turn defined by its point. these ui classes are used to build and manipulate a model on screen. classes in the mesh package are responsible for generating the mesh based on the on–screen model, creating element and node objects in the process. the class diagram in figure. 6 shows some model and ui classes. feui model mesh figure. 5 packages in the model–ui separation pattern. in the nonlinear analysis, loading is applied as a linear increment. by using the modular analysis pattern, the linear increment is carried out by elastic package and results involve in the plastic package. this increases the reusability and extensibility of the code, so in other nonlinear analyses, the elastic packages remain fix. figure. 4 packages in the model–analysis separation pattern. since for each element, numbers of nodes are different, fem classes are also different. but the ui models of elements are the same. therefore, the ui–model separation pattern is used. ui–related responsibilities should be assigned to ui objects. ui classes should be grouped together in a subsystem that is dependent on the model subsystem. this allows the more volatile ui subsystem to be changed without affecting model classes. naturally, there should be no coupling between the analysis and ui subsystems. figure. 5 shows the dependencies between the ui, mesh and fe packages. ui contains classes such as struct, curve, and point. a struct represents a (sub) domain of the finite element model. the curve of a struct describes its boundary. each curve is in turn defined by its point. these ui classes are used to build and manipulate a model on screen. classes in the mesh package are responsible for generating the mesh issn: 2180-1053 vol. 4 no. 1 january-june 2012 journal of mechanical engineering and technology 74 based on the on–screen model, creating element and node objects in the process. the class diagram in figure. 6 shows some model and ui classes. 62 figure. 4 packages in the model–analysis separation pattern. since for each element, numbers of nodes are different, fem classes are also different. but the ui models of elements are the same. therefore, the ui–model separation pattern is used. ui–related responsibilities should be assigned to ui objects. ui classes should be grouped together in a subsystem that is dependent on the model subsystem. this allows the more volatile ui subsystem to be changed without affecting model classes. naturally, there should be no coupling between the analysis and ui subsystems. figure. 5 shows the dependencies between the ui, mesh and fe packages. ui contains classes such as struct, curve, and point. a struct represents a (sub) domain of the finite element model. the curve of a struct describes its boundary. each curve is in turn defined by its point. these ui classes are used to build and manipulate a model on screen. classes in the mesh package are responsible for generating the mesh based on the on–screen model, creating element and node objects in the process. the class diagram in figure. 6 shows some model and ui classes. feui model mesh figure. 5 packages in the model–ui separation pattern. in the nonlinear analysis, loading is applied as a linear increment. by using the modular analysis pattern, the linear increment is carried out by elastic package and results involve in the plastic package. this increases the reusability and extensibility of the code, so in other nonlinear analyses, the elastic packages remain fix. figure. 5 packages in the model–ui separation pattern. in the nonlinear analysis, loading is applied as a linear increment. by using the modular analysis pattern, the linear increment is carried out by elastic package and results involve in the plastic package. this increases the reusability and extensibility of the code, so in other nonlinear analyses, the elastic packages remain fix. 63 figure. 6 model and ui classes. classes that implement calculation (figure. 7) represent various types of analysis. for example, static analyzes a finite element model statically. extending the program to perform, say, transient response analysis can be achieved by implementing a new subtype. figure. 7: calculation classes. 6.0 example the numerical example considered for validation is an isotropic square plate of constant thickness, simply supported on its four sides, subjected to a uniformly distributed load q = 1.0kpa. the stress–strain behavior of the plate is elastic–perfect plastic with young’s modulus of e = 10.92kpa, poisson’s ratio of ν = 0.3 and yield stress of σ = 1600kpa. the geometry and material properties are shown in figure. 8. figure. 6 model and ui classes. classes that implement calculation (figure. 7) represent various types of analysis. for example, static analyzes a finite element model statically. extending the program to perform, say, transient response analysis can be achieved by implementing a new subtype. issn: 2180-1053 vol. 4 no. 1 january-june 2012 implementation of nonlinear finite element using object–oriented design patterns 75 63 figure. 6 model and ui classes. classes that implement calculation (figure. 7) represent various types of analysis. for example, static analyzes a finite element model statically. extending the program to perform, say, transient response analysis can be achieved by implementing a new subtype. figure. 7: calculation classes. 6.0 example the numerical example considered for validation is an isotropic square plate of constant thickness, simply supported on its four sides, subjected to a uniformly distributed load q = 1.0kpa. the stress–strain behavior of the plate is elastic–perfect plastic with young’s modulus of e = 10.92kpa, poisson’s ratio of ν = 0.3 and yield stress of σ = 1600kpa. the geometry and material properties are shown in figure. 8. figure. 7: calculation classes. 6.0 example the numerical example considered for validation is an isotropic square plate of constant thickness, simply supported on its four sides, subjected to a uniformly distributed load q = 1.0kpa. the stress–strain behavior of the plate is elastic–perfect plastic with young’s modulus of e = 10.92kpa, poisson’s ratio of ν = 0.3 and yield stress of σ = 1600kpa. the geometry and material properties are shown in figure. 8. 64 figure. 8 an isotropic square plate, simply supported on its four sides, subjected to a uniformly distributed load. we compare our results obtained for bilinear, serendipity and lagrange finite elements using full, reduced and selective reduced integration methods, with those published by owen and hinton [21]. as shown in figure. 9, the full and reduced integration methods for bilinear element give very stiff results, but selective reduced is comparably good. figure. 9 the load–deflection relation for bilinear element. for analysis of thin plate based on mindlin theory, as shown in figure. 10 and figure. 11, the serendipity and lagrange elements using both reduced and selective reduced integration methods are suitable. figure. 8 an isotropic square plate, simply supported on its four sides, subjected to a uniformly distributed load. we compare our results obtained for bilinear, serendipity and lagrange finite elements using full, reduced and selective reduced integration methods, with those published by owen and hinton [21]. as shown in figure. 9, the full and reduced integration methods for bilinear element give very stiff results, but selective reduced is comparably good. issn: 2180-1053 vol. 4 no. 1 january-june 2012 journal of mechanical engineering and technology 76 64 figure. 8 an isotropic square plate, simply supported on its four sides, subjected to a uniformly distributed load. we compare our results obtained for bilinear, serendipity and lagrange finite elements using full, reduced and selective reduced integration methods, with those published by owen and hinton [21]. as shown in figure. 9, the full and reduced integration methods for bilinear element give very stiff results, but selective reduced is comparably good. figure. 9 the load–deflection relation for bilinear element. for analysis of thin plate based on mindlin theory, as shown in figure. 10 and figure. 11, the serendipity and lagrange elements using both reduced and selective reduced integration methods are suitable. figure. 9 the load–deflection relation for bilinear element. for analysis of thin plate based on mindlin theory, as shown in figure. 10 and figure. 11, the serendipity and lagrange elements using both reduced and selective reduced integration methods are suitable. 65 figure. 10 the load–deflection relation for serendipity element. figure. 11 the load–deflection relation for lagrange element. 7.0 conclusion the object–oriented approach is shown to offer undeniable advantages compared to earlier programming structures (procedural based). the encapsulation of the data largely improves the modularity, and thus the reliability and legibility of the code. inheritance allows an automatic reusability of the already developed methods, and polymorphism is a powerful means to raise the level of abstraction. through the example of nonlinear analyzing the mindlin plates, this paper has shown how using of design patterns into an object oriented finite element code, the reusability, extensibility and maintainability of the code increase. in determining of the nonlinear behavior of plates based on mindlin theory, to access of better accuracy, using element with figure. 10 the load–deflection relation for serendipity element. issn: 2180-1053 vol. 4 no. 1 january-june 2012 implementation of nonlinear finite element using object–oriented design patterns 77 65 figure. 10 the load–deflection relation for serendipity element. figure. 11 the load–deflection relation for lagrange element. 7.0 conclusion the object–oriented approach is shown to offer undeniable advantages compared to earlier programming structures (procedural based). the encapsulation of the data largely improves the modularity, and thus the reliability and legibility of the code. inheritance allows an automatic reusability of the already developed methods, and polymorphism is a powerful means to raise the level of abstraction. through the example of nonlinear analyzing the mindlin plates, this paper has shown how using of design patterns into an object oriented finite element code, the reusability, extensibility and maintainability of the code increase. in determining of the nonlinear behavior of plates based on mindlin theory, to access of better accuracy, using element with figure. 11 the load–deflection relation for lagrange element. 7.0 conclusion the object–oriented approach is shown to offer undeniable advantages compared to earlier programming structures (procedural based). the encapsulation of the data largely improves the modularity, and thus the reliability and legibility of the code. inheritance allows an automatic reusability of the already developed methods, and polymorphism is a powerful means to raise the level of abstraction. through the example of nonlinear analyzing the mindlin plates, this paper has shown how using of design patterns into an object oriented finite element code, the reusability, extensibility and maintainability of the code increase. in determining of the nonlinear behavior of plates based on mindlin theory, to access of better accuracy, using element with higher degree of freedoms is proposed. also, to overcome the shear locking problem, different integration methods are used. then, the code must be able to add new elements or integration methods. to achieve this aim, three design patterns are used in the object oriented code. by using model–analysis separation pattern, programmer can be able to add new elements to model subsystem and new integration methods to analysis subsystem without any manipulation of the other subsystem. the clear division of responsibilities makes both maintenance and subsequent extensions of the system easier. also, decomposing the model subsystem to ui and fem subsystems help to add new elements to fem subsystems without change the ui subsystem. finally, decomposing the analysis subsystem into components facilitates code reuse without complicating the main hierarchy. issn: 2180-1053 vol. 4 no. 1 january-june 2012 journal of mechanical engineering and technology 78 references [1] miller g. r. an object–oriented approach to structural analysis and design. computers and structures 1991;40(1):75–82. [2] dubois–pelerin y, bomme p, zimmermann t. object–oriented finite element programming concepts. proceedings of european conference on new advances in computational structural mechanics, elsevier, amsterdam 1991;95–101. [3] dubois–pelerin y, zimmermann t, bomme p. object–oriented finite element programming: ii. a prototype program in smalltalk. comput methods appl mech eng 1992;98:361–397. [4] commend s, zimmermann t. object–oriented nonlinear finite element programming: a primer. adv in engng software 2001;32(8):611–628. [5] mackie r. i. object–oriented finite element programming – the importance of data modeling. advances in engineering software 1999;32(9–11): 775–782. [6] mackerle j. object–oriented techniques in fem and bem, a bibliography (1996–1999). finite elements in analysis and design 2000;36:189–196. [7] patzak b, bittnar z. design of object oriented finite element code, advances in engineering software 2001;32:759–767. [8] liu w, tong m, wu x, lee gc. object oriented modeling of structural analysis and design with application to damping device configureuration. j comput civil eng 2003;17(2):113–22. [9] fenves gl, mckenna f, scott mh, takahashi y. an object oriented software environment for collaborative network simulation. in: proceedings of the 13th world conference on earthquake engineering, vancouver, canada; 2004. [10] heng b, mackie r.i. using design patterns in object–oriented finite element programming. comput and struct, doi:10.10161j.compstruc. 2008.04.016. [11] rucki md, miller gr. an algorithmic framework for flexible finite element based structural modeling. comput methods appl mech eng 1996;136:363–84. [12] dubois–pèlerin y, pegon p. improving modularity in object–oriented finite element programming. commun numer methods eng 1997;13:193–8. [13] bettig bp, han rps. an object oriented framework for interactive numerical analysis in a graphical user interface environment. int j numer methods eng 1996;39(17):2945–72. issn: 2180-1053 vol. 4 no. 1 january-june 2012 implementation of nonlinear finite element using object–oriented design patterns 79 [14] ju j, hosain mu. finite element graphic objects in c++. j comput civil eng 1996;10(3):258–60. [15] marczak rj. an object–oriented programming framework for boundary integral equation methods. comput struct 2004;82:1237– 1257. [16] mindlin rd. influence of rotatory inertia and shear on flexural motions of isotropic elastic plates. j. appl. mech.1951;18(1):31–38. [17] zienkiewicz oc, taylor rl, too jm. reduced integration technique in general analysis of plates and shells. international journal for numerical methods in engineering 1971;3:275–90. [18] pawsey sf, clough rw. improved numerical integration of thick shell finite elements. international journal for numerical methods in engineering 1971;3:575–86. [19] briassoulis d. on the basics of the shear locking problem of cv isoparametric plate elements. comput structs 1989;33:169–85. [20] iliushin a. plastichnost. moscow: gostekhizdat, 1965(in russian). [21] owen d, hinton e. finite elements in plasticity: theory and practice. swansea: pineridge press, 1980. issn: 2180-1053 vol. 2 no. 1 january-june 2010 an improvement on mechanism of motorcycle rim adjusting jig 1 an improvement on mechanism of motorcycle rim adjusting jig masjuri bin musa @ othman1, mohd. ruzi bin harun2, wan mohammad farid bin wan mohammad3, sulaiman bin sabikan4 1faculty of mechanical engineering, universiti teknikal malaysia melaka, locked bag 1752, pejabat pos durian tunggal, 76109 durian tunggal, melaka. 2faculty of electrical engineering, universiti teknikal malaysia melaka, locked bag 1752, pejabat pos durian tunggal, 76109 durian tunggal, melaka. email: 1masjuri@utem.edu.my abstract motorcycle rim adjusting jig has been widely used as an instrument for wheel alignment of motorcycle’s tires, which is between rims and its hub. by using this jig, it may help the mechanic to do the adjustment of “high-low” and “side-run-out” problems which is normally related to the alignment of the motorcycle rim. however, the existing motorcycle rim adjusting jig which is already in the market, is not really user friendly. it is very difficult to operate and it takes longer time to complete the alignment. furthermore, the measurement of the “run-out” is not very precise and this may affect time during the adjustment process.to overcome this problem, the improvement and enhancement of the existing jig’s design mechanism must be done in order to reduce the operating time and facilitate the rim adjustment techniques. the concept of using the dial gauge to adjust the rim tolerance (“high-low” and “side-run-out” problems) as well as the presence of the driven motor which is connected to the hub’s holders can make the product function effectively.the alignment process can be done automatically hence the detection of the run out problems become more easily and the alignment process will become faster with the existence of “present sensor” which will be located near to the hub’s holder. keywords: mechanism improvement, motorcycle rim, adjusting jig. 1.0 introduction a motorcycle rim adjusting jig or also known as truing jig, is a specialized tool or equipment for the purpose of straightening the wheels. the jig consists of an axle stand on which the wheel can be rotated. the other issn: 2180-1053 vol. 2 no. 1 january-june 2010 journal of mechanical engineering and technology 2 component, calipers which act as an indicator, is used to measure the deviations of the wheel’s rim from ideal alignment. figure 1 shows the common motorcycle rim adjusting jig which widely used in the market. the stand is used in conjuction with an appropriately sized spoke wrench to loosen and tighten the spokes that hold the rim in its own position. common rim consist of frame, hub, spokes and nipples (hoeppner, 2006) as shown in figure 2. a good wheel alignment will ensure that the wheel is ideal in two ways which are high-low (roundness of the rim) and side-run-out (sideways wobble). ideally, spokes have equal tension although the two sides will be different if a wheel facing uneven bracing angle of spokes on some multi speed wheels with tension high enough to give a rigid wheel and retain some tension under all loads but not so high as to lead the failure of spokes or the rim. spokes should have no residual twist from tightening the nipples. axle stand caliper figure 1 motorcycle rim adjusting jig. spoke hub frame nipple figure 2 common parts of motorcycle rim issn: 2180-1053 vol. 2 no. 1 january-june 2010 an improvement on mechanism of motorcycle rim adjusting jig 3 current motorcycle rim adjusting jig which widely used in the market nowadays, are difficult to operate. operating time, and alignment accuracy are the main concerned factors in order to complete the alignment process. the operator skills who handle these particular jig and their concentration are very essential as well. based on the current scenario, required time to finish up the alignment for a wheel is within 30 minutes to 40 minutes. in addition, these alignment operation is totally depends on the operator skills, and this lead to poor accuracy of the overall alignment process. full concentration is required in handling this jig, and this causes the operator not being able to perform other tasks until the alignment process is done completely. therefore, the existing design mechanism need to be reviewed to overcome those problems that heve been raised as to ensure that the jig can be optimized for more effective use. the concept of the using dial gauges, sensors, and driven motor which can be intergrated together with the jig, will enable the process alignment implemented automatically, where by the run-out problems can be detected more easily and quickly. the purpose of this paper is to discuss about the conceptual idea of how to improve and overcome the problems especially those which are related to the difficulties faced by the end users as stated earlier. improvement in terms of mechanism design will be emphasized in this work. 2.0 run-out alignment problems in the process of wheel alignment and balancing, there are two types of alignment which should be concerned with; high-low and side runout problems. typically, the side run-out problem is more common compared with the high-low problem. 2.1 side run-out alignment problem the calipers which act as a gauge for side shaking are located on both side of the u-shaped frame. rim’s position should be in the middle between the calipers. when rim is rotated slowly, observe the clearance between the tip of the calipers and the rim surface (www.cyclingnews. com). sometimes, the tip of the caliper will touch the rim or will away from the rim. this is called side run-out alignment problem. figure 3 shows the side run-out alignment problem, where the side to side issn: 2180-1053 vol. 2 no. 1 january-june 2010 journal of mechanical engineering and technology 4 wobble of the rim can be seen as the wheel spins. side surface of the rim caliper’s tip figure 3 side run-out alignment problem 2.2 high-low alignment problem high-low alignment problem is the amount of up and down wobble. if the wheel becomes out-of-round, it wobbles up and down with each revolution. this high-low alignment prolem can be affected by spoke tension (www.cyclingnews.com). sections of rim can be moved toward the hub by tightening spokes. alternatively, sections of rim can move slightly outward by loosening spokes. figure 4 shows the high-low alignment problem of the rim. similarly like side run-out alignment prolem, when rotating the rim slowly, the clearance between the gauge plate and the rim surface should be given full attention. sometimes the gauge plate will touch the rim, and will away from the rim. when this scenario appeared, the high-low alignment problem need to be addressed. rim’s sidewall surface gauge plate figure 4 high-low alignment problem issn: 2180-1053 vol. 2 no. 1 january-june 2010 an improvement on mechanism of motorcycle rim adjusting jig 5 2.3 how to fix the run-out alignment problems? prior to discuss further about this particular matter, it is better to know the position of the spokes on the motorcycle rim. figure 5 shows the location of spokes on the common motorcycle rim. the spoke nipples labelled a, c, and e are on the left side of the rim, actually come from the right side flange. spoke nipples b, d, and f are on the right, actually come from the right side of the flange. left side spokes tend to pull the rim toward the left. their pulling is offset by the pull of spokes on the right (www.cyclingnews.com). figure 5 location of spokes (seen from the top view) (www.cyclingnews.com). each nipple affects a relatively wide area of the rim. spoke c will pulls area mainly adjacent to its location, but will also effect the rim up to and even past a and f. tightening nipple c will increases spoke tension and tend to move that section of rim to the left. tightening nipple d will tend to move the rim to the right. loosening nipple c, will also tend to move the rim to the right, due to the constant pull of d (www. cyclingnews.com). in case for side run-out alignment problem, if rim touches left side of issn: 2180-1053 vol. 2 no. 1 january-june 2010 journal of mechanical engineering and technology 6 the caliper (reference point), find out the closest nipple to the center deviation coming from right side of hub flange. on the other hand, if rim touches right side of the caliper, find out the closest nipple to the center deviation coming from left side of hub flange. this can be done by tightening the nipple half turn and rotate the rim back and forth in this area and check the deviation again. repeat the steps for both situations until rim position is in aligned position. that means the right and left side of rim will never touch the caliper. in order to get good adjustment, both side of rim must always in center location from right and left calipers (www.cyclingnews.com). for the high-low alignment problem, if the gauge touching the rim sidewall, tighten the two spokes in the middle of the hub with same amount, beginning with half turn. rotate the rim again back and forth in this particular areas and check the deviation once again. repeat the steps until rim become completely true (www.cyclingnews.com). 3.0 methodology as has been discussed before, the problems faced when carrying out wheel alignment and balancing are the time consumption, ease of operating the jig, requires skilled labour to handle as well as his full concentration when doing the alignment. therefore, to overcome or at least to reduce the problems faced by this situations, few suggested solutions to replace this conventional method has been made, where the jig’s mechanism design itself need to be changed and improved. the proposed methods including the use of dial gauges which are connected with sensors (korth, 1983), intended to detect “high-low” or “side runout” automatically; involving the use of a motor which placed at the rim holder – to rotates the rim, and stop automatically whenever the sensors detect the presence of the misalignment problems. the current motorcycle rim adjusting jigs do not have features as proposed. 3.1 the implementation of dial gauge, limit switch sensor, and driven motor. dial gauge is used to replace the caliper where its main function is to detect the existance of “high-low” or “side run-out” problems. on the other hand, limit switch sensor acts as a flagman for the driving motor to stop rotating when the dial gauge detect a problem. meanwhile, the function of the driving motor is to automatically rotate the rim. in the proposed method, the position of dial gauge and sensor on the adjusting jig stand is as shown in figure 6. this method is to do adjustment for issn: 2180-1053 vol. 2 no. 1 january-june 2010 an improvement on mechanism of motorcycle rim adjusting jig 7 the “side run-out” alignment problem. as indicated in the figure 6, the sensor is located at the upper part of the dial gauge. when the tip of the dial gauge touches the rim, the needle of the dial gauge moving from zero reference, and this shows that the rim is facing “side run-out” alignment problem. when this scenario occurs, the cap of the dial gauge will be in contact with the sensor and this sensor will give command to the driving motor to stop the rim from rotate. in this particular case, observations should be done either the rim touches the left or right hand side of the dial guage’s tip. therefore, from the observation, the spoke’s adjustment need to be done as described in subsection 2.2. nipple of the spoke will be adjusted until the dial gauge needle back to zero reference. the same method will be used for “highlow” alignment problem. limit switch sensor dial gauge figure 6 the location of the sensor and dial gauge 4.0 results and discussion before producing the prototype, all components which are related to the jig, will bw graphically constructed in three dimensional view by using solidworks® software. the aim is to ensure that all components are free of any problem such as dimension problem, before the assembly process can be done. figure 7 shows the new jig after modification has been made. issn: 2180-1053 vol. 2 no. 1 january-june 2010 journal of mechanical engineering and technology 8 driven motor dial gauge sensor control box jig’s stand figure 7 modification jig after the fabrication as well as installation process has been completed, the next step is to perform some experiments on a prototype produced. experiments to be conducted will involve the existing jig as well as the modification jig (prototype) based on proposed methods. two parameters have been taken into account to implement this particular experiment which are the operating time and ease of handling the jig. table 1 comparison on operating time for both types of jig (abdul ghani, 2009) to conduct this experiment, six units of motorcycle rims have been used, and before the operation starts, all spokes on the rims will be loosened. first of all, the experiment will be done on existing jig (jig without modification), and operating time will be taken from start until end of the alignment process. the same steps will be repeated again for the new jig (abdul ghani, 2009). table 1 shows the comparison of the experimental results for six units of motorcycle rim. from table 1, we can conclude that: time saving for modification jig: 42 – 14 = 28 minutes percentage of time saving: (28/42) x 100% = 67% issn: 2180-1053 vol. 2 no. 1 january-june 2010 an improvement on mechanism of motorcycle rim adjusting jig 9 based from the data collection in table 1, approximately 67 per cent of operating time can be saved by using new jig. for the existing jig, the average operating time is approximately 42 minutes, and on the other hand for new jig, time taken is 14 minutes (abdul ghani, 2009). during the alignment operation for the new jig, the operator can perform other tasks, without having to wait and give full attention to the operation of the alignment. the reason is because the rim is controlled by the driving motor and it is rotated automatically. in addition, the presence of dial gauges and limit switch sensors can detect the alignment problems by itself. using the old jig the operator must give full attention when doing the alignment process. this is because, the rim should manually be rotated, while the alignment problems should be detected by operator himself (abdul ghani, 2009). from the observations on the handling activities for both jigs, it is clearly shown that the existing jig somehow is difficult to operate, while for the new jig, the handling activities is much more easier due to the implementation of the dial gauges, limit switch sensors and driving motor. indirectly it can reduce the use of operator’s skill and allows the jig to function more effectively. 5.0 conclusion the performance of the mechanism suggested for the motorcycle rim adjusting jig has been evaluated. comparison between the existing jig and new modification jig (after improvement has been made) has been carried out. based on the experiment’s result, clearly shows that the implementation of dial gauges, limit switch sensors, and driving motor has great influenced on the operating time as well as facilitate the handling activities. for future works, more various other types of mechanism can be proposed and studied in order to find out the best solution to facilitate and expedite the process of using the existing jig. 6.0 acknowledgement this research work is supported by short term grant scheme. the authors are very greatful to the ministry of higher education malaysia and utem for supporting the present works. issn: 2180-1053 vol. 2 no. 1 january-june 2010 journal of mechanical engineering and technology 10 7.0 references charles c. mccloskey. 2008. manual clamping device for static balancing a wheel. u.s. patent no. 2008/0163959 a1 john k. korth. 1983. apparatus and method for detecting and indicating misalignments of vehicle wire spoke wheels. u.s. patent no. 4417237. kirk hoeppner. 2006. wire spoke wheel, and components for same. u.s. patent no. 2006/0250021 a1. mohamad harith bin abdul ghani.2009. bsc.mechanical engineering (design & innovation) theses, universiti teknikal malaysia melaka, 71-72. richard j. gessler jr. 2007. tire balancing devices and methods. u.s. patent no. 2007/0000322 a1. website: www.cyclingnews.com/tech/fix/?id=howfix_truing. (access 20 august 2009.) issn: 2180-1053 vol. 2 no. 1 january-june 2010 solving tracking problem of a nonholonomic wheel mobile robot using backstepping technique 85 solving tracking problem of a nonholonomic wheel mobile robot using backstepping technique noor asyikin binti sulaiman1, azdiana binti md. yusop1, sharatul izah binti samsudin1 1faculty of electronic and computer engineering, universiti teknikal malaysia melaka, locked bag 1752, pejabat pos durian tunggal, 76109 durian tunggal, melaka. abstract nonholonomic system is a mechanical system that is subject to nonholonomic constraints. they are the constraints on the velocity of the system which can not be integrated into position constraints that can be used to reduce the number of generalized coordinates. mobile robots constitute a typical example of non-holonomic systems. this project attempts to control a nonholonomic mobile robot to track the desired trajectories. in this project, the combination of kinematics and dynamics of the mobile robot are used to control the robot using backstepping technique. two types of input are presented in this paperwork. from the simulation results, the controller is able to control a non-holonomic mobile robot to track the desired trajectories. all simulations are performed using simulink/matlab. keywords: backstepping technique, non-holonomic system, mobile robot 1.0 introduction mobile robots have the capability to move around in their environment and are not fixed to one physical location. the mobile robot can be broken down into holonomic and nonholonomic mobile robot. nonholonomic mobile robot means that a mobile robot that cannot move laterally. this is due to the velocity of the mobile robot possess two degree of freedom which cannot be integrated into positioning constraint since it has three degree of freedom. several methods of control techniques have been studied and proposed considering the system kinematics and dynamics model. in early years, many researchers have been done using kinematics control. in issn: 2180-1053 vol. 2 no. 1 january-june 2010 journal of mechanical engineering and technology 86 this technique, the dynamic model is neglected to simplify the work. it is always assumed that the mobile robot systems fulfill the perfect velocity tracking. in general, these controllers have successfully driven the trajectory tracking error to converge to zero asymptotically. nevertheless, the kinematics control is inadequate to provide good stability, maneuverability and robustness of the mobile robot. however, the simplification is acceptable when the velocities are low, as in most mobile robot applications (t. c. lee, 2001). in contrast, the dynamic control technique is approaching closer to the real mobile robot system compared to the kinematics control because it includes dynamic environments of the system such as the mass and inertia factor. the main equation of motion employed in dynamic control model is derived from the euler-langrange method. in later years, many researchers have investigated the application of dynamic control incorporate with elements of adaptive, intelligent, robust control and many more. r. fierro has proposed a dynamic control that is extended to integrate the kinematics controller with a torque controller using a backstepping method. this method combines both kinematics and torque control laws. it is asymptotically stable and guarantees to converge through the derivative of a lyapunov function. this paper will use the controller developed by (r. fiero, 1995) to observe the performance of the mobile robot with the input of straight line and circular. 2.0 nonholonomic wheeled mobile robot the model of a nonholonomic mobile robot is shown in figure 1. it has two active wheels mounted on the same axis at rear and a passive wheel at front. the active wheels will drive and steer the mobile robot. previous research shown that linearization control technique failed at point p and a new reference point, c is used to develop the mathematical model (r. fierro, 1995). issn: 2180-1053 vol. 2 no. 1 january-june 2010 solving tracking problem of a nonholonomic wheel mobile robot using backstepping technique 87 r. fierro has proposed a dynamic control that is extended to integrate the kinematics controller with a torque controller using a backstepping method. this method combines both kinematics and torque control laws. it is asymptotically stable and guarantees to converge through the derivative of a lyapunov function. this paper will use the controller developed by (r. fiero, 1995) to observe the performance of the mobile robot with the input of straight line and circular. 2.0 nonholonomic wheeled mobile robot the model of a nonholonomic mobile robot is shown in figure 1. it has two active wheels mounted on the same axis at rear and a passive wheel at front. the active wheels will drive and steer the mobile robot. previous research shown that linearization control technique failed at point p and a new reference point, c is used to develop the mathematical model (r. fierro, 1995). figure 1 model of a nonholonomic mobile robot the steering system derived from non-holonomic constraint is known as tvqsq (1) with s(q) as 10 cossin sincos )( d d qs (2) therefore, (1) can be written as v d d y x c c c 10 cossin sincos    (3) figure 1 model of a nonholonomic mobile robot the steering system derived from non-holonomic constraint is known as with s(q) as therefore, (1) can be written as the dynamic equation of the mobile robot is the definitions of (4) can be obtained in (r. fierro, 1995) and (shen lin, 2000) as: the dynamic equation of the mobile robot is )()(),( qaqbqgqfqqqvqqm td (4) the definitions of (4) can be obtained in (r. fierro, 1995) and (shen lin, 2000) as:     sincos,,0)( sinsin coscos 1 )( 0 sin cos ),( cossin cos0 sin0 )( 2 2 yxmqg rr r qb md md qqqv imdmd mdm mdm qm c l r (5) in this case, g(q)=0, because the trajectory of the robot base is constrained to the horizontal plane, since the system cannot change its vertical position (r. fierro, 1995). 3.0 backstepping control design control theory is a combination of engineering and mathematics that deals with the behaviour of dynamical systems. in control theory, backstepping is a technique (p. v. kokotovic, 1992) for designing stabilizing controls for a special class of nonlinear dynamical systems. it breaks a design problem for a full system into a sequence of design problems. because of this recursive structure, the design process can be started at the known-stable system and "back out" new controllers that progressively stabilize each outer subsystem. the process terminates when the final external control is reached. hence, this process is known as backstepping (h. k. khalil, 2002). r. fierro has proposed to convert the velocity control into a torque control for the actual physical cart. the selection of the torque control is obtained from the dynamic equation of the mobile robot, so that the steering system will behave in the same manner of the desired velocity. the dynamic equation of the mobile robot is )()(),( qaqbqgqfqqqvqqm td (4) the definitions of (4) can be obtained in (r. fierro, 1995) and (shen lin, 2000) as:     sincos,,0)( sinsin coscos 1 )( 0 sin cos ),( cossin cos0 sin0 )( 2 2 yxmqg rr r qb md md qqqv imdmd mdm mdm qm c l r (5) in this case, g(q)=0, because the trajectory of the robot base is constrained to the horizontal plane, since the system cannot change its vertical position (r. fierro, 1995). 3.0 backstepping control design control theory is a combination of engineering and mathematics that deals with the behaviour of dynamical systems. in control theory, backstepping is a technique (p. v. kokotovic, 1992) for designing stabilizing controls for a special class of nonlinear dynamical systems. it breaks a design problem for a full system into a sequence of design problems. because of this recursive structure, the design process can be started at the known-stable system and "back out" new controllers that progressively stabilize each outer subsystem. the process terminates when the final external control is reached. hence, this process is known as backstepping (h. k. khalil, 2002). r. fierro has proposed to convert the velocity control into a torque control for the actual physical cart. the selection of the torque control is obtained from the dynamic equation of the mobile robot, so that the steering system will behave in the same manner of the desired velocity. issn: 2180-1053 vol. 2 no. 1 january-june 2010 journal of mechanical engineering and technology 88 the dynamic equation of the mobile robot is )()(),( qaqbqgqfqqqvqqm td (4) the definitions of (4) can be obtained in (r. fierro, 1995) and (shen lin, 2000) as:     sincos,,0)( sinsin coscos 1 )( 0 sin cos ),( cossin cos0 sin0 )( 2 2 yxmqg rr r qb md md qqqv imdmd mdm mdm qm c l r (5) in this case, g(q)=0, because the trajectory of the robot base is constrained to the horizontal plane, since the system cannot change its vertical position (r. fierro, 1995). 3.0 backstepping control design control theory is a combination of engineering and mathematics that deals with the behaviour of dynamical systems. in control theory, backstepping is a technique (p. v. kokotovic, 1992) for designing stabilizing controls for a special class of nonlinear dynamical systems. it breaks a design problem for a full system into a sequence of design problems. because of this recursive structure, the design process can be started at the known-stable system and "back out" new controllers that progressively stabilize each outer subsystem. the process terminates when the final external control is reached. hence, this process is known as backstepping (h. k. khalil, 2002). r. fierro has proposed to convert the velocity control into a torque control for the actual physical cart. the selection of the torque control is obtained from the dynamic equation of the mobile robot, so that the steering system will behave in the same manner of the desired velocity. in this case, g(q)=0, because the trajectory of the robot base is constrained to the horizontal plane, since the system cannot change its vertical position (r. fierro, 1995). 3.0 backstepping control design control theory is a combination of engineering and mathematics that deals with the behaviour of dynamical systems. in control theory, backstepping is a technique (p. v. kokotovic, 1992) for designing stabilizing controls for a special class of nonlinear dynamical systems. it breaks a design problem for a full system into a sequence of design problems. because of this recursive structure, the design process can be started at the known-stable system and “back out” new controllers that progressively stabilize each outer subsystem. the process terminates when the final external control is reached. hence, this process is known as backstepping (h. k. khalil, 2002). r. fierro has proposed to convert the velocity control into a torque control for the actual physical cart. the selection of the torque control is obtained from the dynamic equation of the mobile robot, so that the steering system will behave in the same manner of the desired velocity. figure 2 shows the structure of the complete system. it starts with calculating the errors position and then continues with control law that calculates the target velocities. next, the target velocities will be converted to the desired torque to drive the mobile robot. the current position of the mobile robot is then obtained from the steering system. since this is a closed loop system, the system tries to nullify the errors position until it is able to follow the desired tracking. issn: 2180-1053 vol. 2 no. 1 january-june 2010 solving tracking problem of a nonholonomic wheel mobile robot using backstepping technique 89 figure 2 the structure of the system the control law of the system is: where kx, ky and k are positive consonant. while the accelaration control input is: where k is a positive definite, diagonal matrix given by: to prove it’s stability, let a scalar function v be a lyapunov function candidate as below; and the derivative is; by considering , then; issn: 2180-1053 vol. 2 no. 1 january-june 2010 journal of mechanical engineering and technology 90 by assuming . therefore, the equilibrium point e = 0 is uniformly asymtotically stable. while the torque equation is; 4.0 result and discussion in this work, all the simulations are performed using the matlab/ simulink. the following is the wheel mobile robot’s parameters used in this work (didik, 2003). m = 31kg d = 0.1m r = 0.15m r = 0.8m initial position coordinates [0,0] figure 3 the performance of the straight line input figure 3 the performance of the straight line input figure 3 the performance of the straight line input issn: 2180-1053 vol. 2 no. 1 january-june 2010 solving tracking problem of a nonholonomic wheel mobile robot using backstepping technique 91 figure 4 the performance of circular input. there are two types of input presented here, which are straight line and circular input. for all the input, vr is fixed to 0.03 m/s to keep low system velocity as in most mobile robot applications. the performances of both inputs are depicted in figure 3 and figure 4. for the straight line input, the r is set to 0 rad/s while the circular input, the r is set to 0.6 rad/s. both input show that the system is able to track the trajectory input with all the velocities are converging to the reference values. the torque for both wheels in the straight line trajectory is the same when the mobile robot totally followed the trajectory. meanwhile, for circular input, the torque for the right wheel is always higher than the left wheel since the input is in circular motion and the mobile robot is moving in counter clockwise direction. figure 4 the performance of circular input. there are two types of input presented here, which are straight line and circular input. for all the input, vr is fixed to 0.03 m/s to keep low system velocity as in most mobile robot applications. the performances of both inputs are depicted in figure 3 and figure 4. for the straight line input, the is set to 0 rad/s while the circular input, the is set to 0.6 rad/s. both input show that the system is able to track the trajectory input with all the velocities are converging to the reference values. the torque for both wheels in the straight line trajectory is the same when the mobile robot totally followed the trajectory. meanwhile, for circular input, the torque for the right wheel is always higher than the left wheel since the input is in circular motion and the mobile robot is moving in counter clockwise direction. 5.0 conclusion in this paperwork, the performance of a nonholonomic wheel mobile robot has been discussed. all the simulations are performed using simulink/ matlab. the results show that the system is able to track issn: 2180-1053 vol. 2 no. 1 january-june 2010 journal of mechanical engineering and technology 92 the reference trajectories and the stability of the system is proved since all the errors have converged to zero. this shows that the backstepping method can be applied to the system. 6.0 acknowledgement the authors would like to thank universiti teknikal malaysia melaka for sponsoring this research study. 7.0 references didik setyo purnomo and musa mailah. (2003). trajectory tracking control of a nonholonomic mobile robot using adaptive active force control with fuzzylogic. advanced technology congress. h. k. khalil. (2002). nonlinear systems. prentice hall. new jersey. 3rd edition. p. v. kokotovic. (1992). the joy of feedback: nonlinear and adaptive. control system magazine. ieee. pp. 7-17 r. fierro and f. l. lewis. (1995). control of a non-holonomic mobile robot: backstepping kinematics into dynamics. proceeding of the 34th conference on decision & control. pp. 3805-3810. shen lin and a. goldenberg. (2000). robust damping control of wheeled mobile robots. proceeding of the 2000 ieee international conference on robotics & automation. pp. 2919-2924. t. c. lee, k. t. song, c. h. lee and c. c. teng. (2001). tracking control of unicycle-modeled mobile robots using a saturation feedback controller. ieee transactions on control systems technology. volume 9. pp. 305 – 318 issn: 2180-1053 vol. 7 no. 2 july december 2015 study of failure loads of carbon epoxy composite plates with single pin holes 33 study of failure loads of carbon epoxy composite plates with single pin holes k. sridevi department of mechanical engineering, faculty of technical and engineering, the m.s. university of baroda, vadodara, gujarat, india abstract this paper deals with the study of failure loads of carbon epoxy composite plates with a circular hole subjected to a traction force by a rigid pin using mathematical model. these are investigated for two variables, the ratio of distance from the free edge of the plate (e) to the diameter of the hole (d) and the ratio of width of the plate (w) to the diameter of the hole (d). the effect of joint geometry on the failure loads has been studied and a comparison of experimental, numerical and mathematical models is made. the results obtained by mathematical model are found to be close to the experimental results. keywords: failure loads, carbon epoxy composite plates, single pin holes. 1.0 introduction composite materials are popularly used because of their light weight, high strength to weight ratio, good fatigue resistance, corrosion resistance etc. compared to metals. fiber-reinforced laminated composite materials have been gaining a wide application area in aircraft, aerospace, and marine industries because of their advanced properties. pinned connections are commonly used in joining composites either to composites or to metal. but the presence of a hole in a laminated plate subjected to external loading introduces a disturbance in the stress field. stress concentrations are generated in the vicinity of the hole making the joint a weak one. the knowledge of failure strength of a joint helps in selecting the appropriate joint size in a given application. the capability of a composite structure to withstand any physical load can be evaluated either by physical testing or by any advanced computational method. performing physical tests on composites is destructive and costly. so, implementing advanced computational techniques to determine the failure loads and failure modes are preferred after some experiments are done. * corresponding author email: srikavirayani@gmail.com issn: 2180-1053 vol. 7 no. 2 july december 2015 journal of mechanical engineering and technology 34 nanda et al. (2009) studied the effects of various geometric parameters on the behaviour of three and four-pin joints in glass fiber/epoxy composite laminate with emphasis on pitch-to-diameter ratio. numerical analysis was performed using a two-dimensional finite element model to study the propagation of damage by implementing tsai–wu failure criteria to predict failure load and to differentiate failure modes. experiments were conducted to validate the results obtained from finite element analysis. aktas (2011) has done experimental and numerical study to determine the failure behaviour of glass epoxy composite plates with single pinned hole and two serial pinned holes. the numerical study was performed by using ansys and yamada-sun failure criteria were used. ozen and sayman (2011) investigated experimentally and numerically the first failure load and the bearing strength behaviour of pinned joints of glass fibre reinforced woven epoxy composite prepregs with two serial holes subjected to traction forces by two serial rigid pins. soykok et al. (2013) have carried experiments to understand the effect of thermal condition and tightening torque on the failure load and failure behavior of glass epoxy composite joints. it was observed that the load carrying capacity of the joint decreased by increasing the temperature level. the tightening torque was observed to increase the joint strength. khashaba et al. (2013) has dealt with the failure and reliability analysis of composite pinned-joints using theoretical models based on weibull distribution functions with experimental results for a guideline of safe design strength. sridevi and satyadevi (2013) have studied the failure of glass vinylester composite plates for different geometries using ansys and also through mathematical modeling. here the results obtained by mathematical models were found to be close to the experimental results when compared with the ansys results. kadir turan et al. (2014) have studied experimentally and numerically the failure loads of carbon epoxy composite plates for different geometric parameters. in the present work mathematical model is developed to predict the failure loads of different geometric specimens. the results obtained from mathematical model are found to be close to the experimental results. 2.0 problem definition in the present work a composite rectangular plate, shown in figure. 1, of length l+e, width w and thickness t is considered. a hole of diameter d is present at a distance e from one edge of the plate. a rigid pin is located at the centre of the hole. a load p is applied to the plate along the longitudinal axis. the plate is symmetric with respect to the longitudinal axis. the diameter of the hole is taken as 6mm and thickness of the plate t as 1.235mm. issn: 2180-1053 vol. 7 no. 2 july december 2015 study of failure loads of carbon epoxy composite plates with single pin holes 35 woven epoxy composite prepregs with two serial holes subjected to traction forces by two serial rigid pins. soykok et al. (2013) have carried experiments to understand the effect of thermal condition and tightening torque on the failure load and failure behavior of glass epoxy composite joints. it was observed that the load carrying capacity of the joint decreased by increasing the temperature level. the tightening torque was observed to increase the joint strength. khashaba et al. (2013) has dealt with the failure and reliability analysis of composite pinned-joints using theoretical models based on weibull distribution functions with experimental results for a guideline of safe design strength. sridevi and satyadevi (2013) have studied the failure of glass vinylester composite plates for different geometries using ansys and also through mathematical modeling. here the results obtained by mathematical models were found to be close to the experimental results when compared with the ansys results. kadir turan et al. (2014) have studied experimentally and numerically the failure loads of carbon epoxy composite plates for different geometric parameters. in the present work mathematical model is developed to predict the failure loads of different geometric specimens. the results obtained from mathematical model are found to be close to the experimental results. 2.0 problem definition in the present work a composite rectangular plate, shown in figure. 1, of length l+e, width w and thickness t is considered. a hole of diameter d is present at a distance e from one edge of the plate. a rigid pin is located at the centre of the hole. a load p is applied to the plate along the longitudinal axis. the plate is symmetric with respect to the longitudinal axis. the diameter of the hole is taken as 6mm and thickness of the plate t as 1.235mm. figure 2. geometry of the specimen figure 2. geometry of the specimen the material properties considered are shown in table 1 given by k. turan et al. (2014). different models are obtained by varying e/d and w/d but keeping the parameters d and t as constant. a mathematical model is developed to obtain the failure loads of different specimens. a comparison of results obtained from mathematical models with experimental and numerical results is made and correlations are observed. table 1. material properties of the plate. longitudinal young’s module e1 (mpa) 172,891 transverse young’s module e2 (mpa) 10,797 shear module g12 (mpa) 3638 poisson’s ratio v12 0.32 longitudinal tensile strength xt (mpa) 1441 transverse tensile strength yt (mpa) 37 longitudinal compressive strength xc (mpa) 420 transverse compressive strength yc (mpa) 116 shear strength s (mpa 57 2.1 mathematical modeling a mathematical model has been developed to predict the failure loads of specimens with different geometries using curve expert. the model has been built with the available experimental results of kadir turan et al. (2014). the equation has two independent variables in w/d ratio as x1 and e/d ratio as x2. the dependent variable considered here is the failure load p. the thickness of the specimen and the diameter of the hole are constant for all the specimens. a full cubic polynomial equation is found to be best suited to determine the failure loads for the issn: 2180-1053 vol. 7 no. 2 july december 2015 journal of mechanical engineering and technology 36 existing problem. this equation can be used to predict the failure load of specimens with other geometric parameters within the given range i.e. for e/d and w/d ratios for which experiments have not been done. the limitation with the model is that the failure mode is not predicted. but, we know that for higher values of e/d and w/d the shear strength and normal strength of the plates increase, so specimen tends to fail in bearing mode only. hence, the mathematical model is best suited to obtain the results for failure loads. the equation developed is found to be the material properties considered are shown in table 1 given by k. turan et al. (2014). different models are obtained by varying e/d and w/d but keeping the parameters d and t as constant. a mathematical model is developed to obtain the failure loads of different specimens. a comparison of results obtained from mathematical models with experimental and numerical results is made and correlations are observed. table 1. material properties of the plate. longitudinal young’s module e1 (mpa) 172,891 transverse young’s module e2 (mpa) 10,797 shear module g12 (mpa) 3638 poisson’s ratio v12 0.32 longitudinal tensile strength xt (mpa) 1441 transverse tensile strength yt (mpa) 37 longitudinal compressive strength xc (mpa) 420 transverse compressive strength yc (mpa) 116 shear strength s (mpa 57 2.1 mathematical modeling a mathematical model has been developed to predict the failure loads of specimens with different geometries using curve expert. the model has been built with the available experimental results of kadir turan et al. (2014). the equation has two independent variables in w/d ratio as x1 and e/d ratio as x2. the dependent variable considered here is the failure load p. the thickness of the specimen and the diameter of the hole are constant for all the specimens. a full cubic polynomial equation is found to be best suited to determine the failure loads for the existing problem. this equation can be used to predict the failure load of specimens with other geometric parameters within the given range i.e. for e/d and w/d ratios for which experiments have not been done. the limitation with the model is that the failure mode is not predicted. but, we know that for higher values of e/d and w/d the shear strength and normal strength of the plates increase, so specimen tends to fail in bearing mode only. hence, the mathematical model is best suited to obtain the results for failure loads. the equation developed is found to be p = a + b*x1 + c*x2 + d*x12 +….. …..+ e*x22 + f*x13 + g*x23 + h*x1*x2 + ….. (1) ……+ i*x12*x2 + j*x1*x22 where in the values of the co-efficients a, b, c, d, e, f, g, h, i and j are given in table 2 table 2. co-efficients of the full cubic model developed a b c d e f g h i j 896.12 -1037.76 465.34 267.51 -123.72 -20.01 7.45 183.98 -21.23 -2.68 where in the values of the co-efficients a, b, c, d, e, f, g, h, i and j are given in table 2 table 2. co-efficients of the full cubic model developed the material properties considered are shown in table 1 given by k. turan et al. (2014). different models are obtained by varying e/d and w/d but keeping the parameters d and t as constant. a mathematical model is developed to obtain the failure loads of different specimens. a comparison of results obtained from mathematical models with experimental and numerical results is made and correlations are observed. table 1. material properties of the plate. longitudinal young’s module e1 (mpa) 172,891 transverse young’s module e2 (mpa) 10,797 shear module g12 (mpa) 3638 poisson’s ratio v12 0.32 longitudinal tensile strength xt (mpa) 1441 transverse tensile strength yt (mpa) 37 longitudinal compressive strength xc (mpa) 420 transverse compressive strength yc (mpa) 116 shear strength s (mpa 57 2.1 mathematical modeling a mathematical model has been developed to predict the failure loads of specimens with different geometries using curve expert. the model has been built with the available experimental results of kadir turan et al. (2014). the equation has two independent variables in w/d ratio as x1 and e/d ratio as x2. the dependent variable considered here is the failure load p. the thickness of the specimen and the diameter of the hole are constant for all the specimens. a full cubic polynomial equation is found to be best suited to determine the failure loads for the existing problem. this equation can be used to predict the failure load of specimens with other geometric parameters within the given range i.e. for e/d and w/d ratios for which experiments have not been done. the limitation with the model is that the failure mode is not predicted. but, we know that for higher values of e/d and w/d the shear strength and normal strength of the plates increase, so specimen tends to fail in bearing mode only. hence, the mathematical model is best suited to obtain the results for failure loads. the equation developed is found to be p = a + b*x1 + c*x2 + d*x12 +….. …..+ e*x22 + f*x13 + g*x23 + h*x1*x2 + ….. (1) ……+ i*x12*x2 + j*x1*x22 where in the values of the co-efficients a, b, c, d, e, f, g, h, i and j are given in table 2 table 2. co-efficients of the full cubic model developed a b c d e f g h i j 896.12 -1037.76 465.34 267.51 -123.72 -20.01 7.45 183.98 -21.23 -2.68 the results obtained by mathematical model are compared with the experimental and numerical results. the graphical representation is shown in figure.2. issn: 2180-1053 vol. 7 no. 2 july december 2015 study of failure loads of carbon epoxy composite plates with single pin holes 37 the results obtained by mathematical model are compared with the experimental and numerical results. the graphical representation is shown in figure.2. figure. 2 comparison of results for constant w/d ratios figure. 2 comparison of results for constant w/d 3.0 results and discussions the failure loads of specimens obtained by varying the e/d and w/d ratios are found from mathematical model. results from numerical models found using ansys are available. a comparison of the failure loads obtained from experimental, numerical and mathematical model have been made. shown in the figure. 2 are the graphs plotted for the failure loads of specimens obtained from experimental, numerical and mathematical models for different w/d ratios and e/d ratios. all the three methods show the same trend in the failure loads but the results obtained by the mathematical model are close to the experimental results. it is observed from the results obtained that ● for a constant w/d ratio, the failure strength of the specimen increases with increase in e/d. this is because keeping the diameter of hole constant, when e/d increases, the distance of the hole from one edge of the plate increases and so the shear strength of the specimen increases. ● with the increase in w/d ratio, keeping e/d ratio constant, the failure load of the specimens increase. this is because as w/d increases and hence the width of the specimen, the normal strength issn: 2180-1053 vol. 7 no. 2 july december 2015 journal of mechanical engineering and technology 38 of specimen increases. ● both numerical and mathematical models show the same trend in failure loads with varying e/d and w/d ratios. ● a comparison of the experimental results with the mathematical model shows that the mathematical model gives results with correlation co-efficient 0.9956, maximum absolute error as 176n and root mean square error as 80n. 4.0 conclusions in the present work, failure loads of carbon epoxy composite plates with singe pin holes were studied mathematically for the geometrical parameters w/d and e/d. a comparison with the experimental results show that ● the specimen is weak for lower values of e/d and w/d and so failure occurs at small loads. ● e/d ratio has a greater effect on the failure load of the specimen. ● as for higher values of e/d and w/d ratios, the specimen generally fails in bearing mode only, and hence the failure loads are the important parameters to be analyzed, wherein the mathematical models prove to be more efficient. ● mathematical models show the same trend in the failure loads of specimens, when compared with the experimental models. so, for estimation of the failure loads within the range considered for the study, the mathematical models developed, i.e., full cubic models proves to be efficient with the given values of correlation coefficient, maximum absolute error and root mean square error. issn: 2180-1053 vol. 7 no. 2 july december 2015 study of failure loads of carbon epoxy composite plates with single pin holes 39 references nanda, a.k., malhotra, s.k. and prasad, n.s., (2009). failure analysis of multi-pin joints in glass fibre/epoxy composite laminates. composite structures, 91. 266-277 aktas, a., (2011). failure analysis of serial pinned joints in composite plates, indian journal of engineering and material sciences, 18. 102-110 ozen, m. and sayman, o., (2011). failure loads of mechanical fastened pinned and bolted composite joints with two serial holes, composites: part b 42, 264-274 soykok, i.f., sayman, o., ozen, m., korkmaz, b., (2013). failure analysis of mechanically fastened glass fiber/epoxy composite joints under thermal effects, composites: part b 45, 192-199 khashaba, u.a., sebaey, t.a., alnefaie, k.a., (2013). failure and reliability analysis of pinned-joints composite laminates: effects of stacking sequences, composites: part b, 45 (2013) 1694-1703 sridevi, k., and satyadevi, a., (2013). model development for estimation of failure loads: a case study of composite plates with single pin-loaded hole, international journal for scientific research & development 1(05), 1268-1271 kadir turan., mustafa gur., meta onur kaman., (2014). progressivef analysis of pin-loaded unidirectional carbon-epoxy laminated composites, mechanics of advanced materials and structures 21, 98-106 issn: 2180-1053 vol. 10 no.2 june – december 2018 67 numerical investigation of the water/alumina nanofluid within a microchannel with baffles mehdi jahangiri 1* , rouhollah yadollahi farsani 1 , akbar alidadi shamsabadi 2 1 department of mechanic, shahrekord branch, islamic azad university, shahrekord, iran 2 young research and elite, shahrekord branch, islamic azad university, shahrekord, iran abstract the study of heat transfer phenomenon in microchannels has attracted researchers’ attention as they have many advantages in the cooling of electronic components. in this numerical study, the effect of adding alumina nanoparticles to the water flow through a microchannel with some baffles embedded on the top and bottom walls is discussed. the several cases including the effect of various volume fraction of nanoparticles (2, 4, 6, and 10%), reynolds number of the inlet flow (10, 20, 30, 40, and 50), and the number of baffles and their heights on the heat transfer phenomena are investigated. the local nusselt number, the average outlet temperature, and the streamlines are presented for representing the results. the results show that increasing the reynolds number decreases the average outlet temperature. moreover, the increase in the number of baffles causes an increase in the average outlet temperature since the formation of vorticities just behind of each baffle and results in a large heat transfer rate. as the baffles height increase, the strength and the area of the vortices increase and hence the heat transfer rate increases. however, an increase in the volume fraction of the nanoparticle increases the average outlet temperature which is due to the increase in conduction heat transfer of nanofluid. keywords: microchannel, heat transfer rate, nanofluids. 1.0 introduction microchannels have many industrial and engineering applications including electronic cooling, medical industries, chemical engineering, automotive heat exchangers, laser equipment and aerospace technology. the microchannels were introduced firstly by tuckerman and pease (tuckerman & pease, 1981). microchannels with liquid coolant like water are widely used to prevent overheating of electronic components and circuits. therefore, the thermal management of microelectronics has become a promising field of research (bar-cohen, 2013; colgan et al., 2007; lee & mudawar, 2009). the microchannels create a higher heat transfer surface per unit volume, as well as a higher heat transfer rate. however, the smaller size of the channel, the more pressure drop takes * corresponding author e-mail: jahangiri.m@iaushk.ac.ir mailto:jahangiri.m@iaushk.ac.ir journal of mechanical engineering and technology 68 issn: 2180-1053 vol. 10 no.2 june – december 2018 places on the flow. higher pumping power is needed as a penalty of high inlet velocity occurs and significantly falls as hydraulic diameter increases (sakanova et al., 2014). the study of heat transfer in microchannels using conventional liquids has been reported by many studies (zhang et al., 2015; lewis & wang, 2018; dixit & ghosh, 2015). basically, the heat transfer of the fluid flow is limited to their thermal properties. however, the methods of heat transfer augmentation in microchannels have been considered and introduced, recently. the most recent one is associated with the high thermal conductor nanoparticles suspended in base fluids and increase the thermal conductivity of the medium (ganvir et al., 2017; hajmohammadi et al., 2018; chari & kleinstreuer, 2018). these particles are generally metal, metal oxide or carbide with the diameters of 1-100 nm (minkowycz et al., 2016). due to the very low flow rate, the microchannel flow is characterized by a very low reynolds number. therefore, it is difficult to achieve an effective turbulent flow. manay and shahin in 2016 investigated the effect of titanium oxide nanoparticles suspended in water on heat transfer in a microchannel. they concluded that increasing the volume fraction of nanoparticles and also decreasing the microchannel height would increase the heat transfer rate. azizi et al. (2016) investigated the effects of water/copper nanofluid on the heat transfer rate and friction coefficient in a microchannel. they showed that by increasing the volume fraction of nanoparticles, the heat transfer rate increases. moreover, the local nusselt number and friction coefficient increase significantly by adding nanoparticles compared to the base fluid. alfaryjat et al. (2018) numerically studied the enhancement of heat transfer using various nanofluids in hexagonal microchannel as a heat sink. they separately examed three types of nanoparticles including aluminum oxide, copper oxide, and silicon oxide. they found that the aluminum oxide gives the highest heat transfer rate compared to the other nanoparticles. ambreem and kim (2018) studied the effect of nanoparticle size on the hydrothermal characteristics of nanofluids in a microchannel which is under a constant heat flux. they used water with aluminum and titanium oxide nanoparticles. the size of the nanoparticles was considered to be 20 to 200 nm. eventually, they realized that the heat transfer rate increases by reducing the size of the nanoparticles. reviewing previous studies leads us to conduct a study on the numerical simulation of two-dimensional flow and heat transfer of water-alumina nanofluid in microchannels using finite element method. the effect of various geometric parameters and flow conditions, including the different arrangement of the baffles, the height and the distance between them, the reynolds number, and the volume fraction of nanoparticles, have been investigated. 2.0 theoretical theory figure 1 shows the geometric configuration of the microchannel. the microchannel with the height l=1mm and a length of s =13l. the six baffles with the height of e1=0.5mm are embedded on the top and bottom walls. the distance between inlet and the first baffles on the top and bottom walls are the sb1 and sb2, respectively. the baffles are assumed to be adiabatic and with zero thickness in numerical simulation. numerical investigation of the water/alumina nanofluid within a microchannel with baffles issn: 2180-1053 vol. 10 no.2 june – december 2018 69 figure 1. schematic geometry of the microchannel. the physical properties of the nanofluids are assumed to be constant and initially given at the inlet flow temperature. the flow is assumed to be laminar and steady. the inlet fluid flow is fully developed. also, the wall temperature is uniform. some physical properties of nanoparticles are shown in table 1. the inlet fluid flow has the temperature of 21°c and the constant temperature of 57°c is assumed on the wall. li et al. (2006) reported that the conventional navier-stokes and energy equations with no-slip boundary condition based on the continuum assumption are still valid and could precisely predict the fluid flow and the heat transfer characteristics in microchannels. table 1. thermophysical properties of nanoparticles and base fluid at 27°c (akbarinia et al., 2011). alumina water properties 3890 998.2 density (kg/m 3 ) 880 4240 heat capacity (j/kg k) 35 0.608 thermal conductivity (w/m.k) 36 -diameter (nm) the governing equations include the continuity, momentum, and energy equation; (1) 𝜌∇. 𝑢 = 0 (2 ) ∇. [−𝑝 + 𝜇(∇𝑢) + (∇𝑢)𝑇] = 0 (3) 𝜌𝐶𝑃.∇𝑇 = ∇. (𝑘∇𝑇) + 𝑄 to calculate the density and thermal capacity of the water/alumina nanofluid, the following relationships are used: (4) 𝜌𝑛𝑓 = (1 − ∅)𝜌𝑓 + ∅𝜌𝑃 (5 ) (𝜌𝐶𝑃 )𝑛𝑓 = (1 − ∅)(𝜌𝐶𝑃 )𝑓 + ∅(𝜌𝐶𝑃 )𝑝 the equation 6 is used to calculate the viscosity of the nanofluid. this correlation is based on the experimental results of meiga et al. (2004) for nanofluid of water/alumina. (6) 𝜇𝑛𝑓 = (1 + 2.5∅ + 150∅ 2)𝜇𝑓 the thermal conductivity of the nanofluid is determined by chein, & huang in 2005. this relationship takes into account the effect of brownian motion and the average diameter of nanoparticles, which is as follows: inlet outlet journal of mechanical engineering and technology 70 issn: 2180-1053 vol. 10 no.2 june – december 2018 (7) 𝑘𝑛𝑓 𝑘𝑓 =1+64.7 ∅0.7460 [ 𝑑𝑓 𝑑𝑛𝑝 ] 0.3690 [ 𝑘𝑝 𝑘𝑓 ] 0.7476 𝑃𝑟0.9955𝑅𝑒1.2321 the special reynolds number (re) and prandtle (pr) are defined as: (8) 𝑃𝑟𝑓 = 𝜂 𝜌𝑓 𝛼𝑓 , 𝑅𝑒𝑓 = 𝜌𝑓 𝐾𝐵𝑇 3𝜋𝜂2𝜆𝑓 kb is the boltzmann constant equal to 1.3807×10 -23 j/k and λf is the mean free path of the water molecule equal to 0.17 nm and η are also calculated by equation 6 : 𝜂 = 𝐴 × 10 𝐵 𝑇−𝐶, a=2.414×10 -5 , b=247.8, c=140 (9) the amount of heat absorbed by the fluid through the pipe is equal to the amount of heat that pass through the walls. therefore, the method for calculating the heat transfer coefficient is as follows: (10) ℎ = [ 𝜌𝑄𝐶𝑃 (𝑇𝑜𝑢𝑡 − 𝑇𝑖𝑛) 𝐴(𝑇𝑤𝑎𝑙𝑙 − 𝑇) ] the local nusselt number on the walls of the microchannel calculated as follows. 𝑁𝑢𝑠𝑠𝑒𝑙𝑡 𝑛𝑢𝑚𝑏𝑒𝑟 = ℎ𝐿 𝑘𝑓 = 𝑞𝑊𝐿 (𝑇𝑊−𝑇𝑏)𝑘𝑓 (11) 3.0 results and discussions the set of equations of continuity and momentum and energy are discretized on the network including the triangular elements shown in the figure 2. the numerical solution has been done with ansys-fluent software. to improve the accuracy of the solution, the mesh is refined in the vicinity of each baffle. to get accurate of numerical simulation a mesh study was accomplished by calculation the nusselt number in the heated walls versus the number of grids. finally, the network includes the number of 12531 cells have been selected for evaluating the results. figure 2. networking of the model to validate the numerical method, the local nusselt number of the nanofluid flow with the reynolds number of 6.9 and volume fraction of 5% was calculated and represented in figure 3. the results were compromised with the data of akbarinia et al. (2011) work. as numerical investigation of the water/alumina nanofluid within a microchannel with baffles issn: 2180-1053 vol. 10 no.2 june – december 2018 71 it can be observed the results are in a good agreement with the laboratory data, so the numerical model would be suitable for the modelling of the problem. figure 3. comparison between the present results of nusselt number and the data of akbarnia et al. (2011). figure 4. the average outlet temperature with various reynolds and baffles number. the effect of baffles on the flow pattern for various reynolds numbers is shown in figure 4. as the reynolds number increases, the average outlet temperature decreases. for the microchannel of a single-baffle at the reynolds of 10 and 40, the outlet temperature is equal to 326.45 k and 314.9 k, respectively. as the reynolds number increases, the volumetric flow rate through the channel increases. therefore, the convective heat journal of mechanical engineering and technology 72 issn: 2180-1053 vol. 10 no.2 june – december 2018 transfer coefficient increases and the average outlet temperature decreases. the effect of the number of baffles on the average outlet temperature in various reynolds number also depicts at figure4. the greater the number of the baffles, the more the rejoin where the vortices form, which leads to an increase in the heat transfer. as it is observable, for reynolds of 40, the outlet temperature changes from 315 for one baffle to about 322 for six baffles. the other point is that for the higher reynolds number, the affection of the number of the baffles is more significant. the reason lays on the augmentation of the strength of vortices which are formed adjacent each of the baffles. the local nusselt number along the length of microchannel on the top and bottom wall is shown at the figure 5. as it can be observed, the local nusselt number fluctuates along the microchannel and the trend is downward. as the flow pass through the microchannel, the temperature difference between the nanofluid and the walls reduces; therefore, the local nusselt gets decreases along the microchannel length. since the first baffle is embedded on the top wall, the first pick is observed on the top local nusselt number trend. the picks are located at the rejoin where the vortices are formed. figure 5. local nusselt number along the microchannel on the top and bottom walls for the case of six-baffles and reynolds of 10. the effect of baffles height including the 0.33 l, 0.5 l, and 0.5 l on the average outlet temperature, for various reynolds numbers, in a three baffles involved microchannel, is provided at figure 6. as the baffles height increases, the more strong vortices are generated becomes larger and so the more nanofluid gets stuck behind the baffle. subsequently, the heat transfer between the wall and the nanofluid increases and the outlet temperature increases. x numerical investigation of the water/alumina nanofluid within a microchannel with baffles issn: 2180-1053 vol. 10 no.2 june – december 2018 73 figure 6. average outlet temperature of the three-baffle involved microchannel for various reynolds at different baffles height. figure 7. shows the average output of microchannel with a volume fraction of different nanoparticles in the number of baffles in re = 10. the effect of addition of alumina nanoparticles to the heat transfer characteristics for various numbers of baffles is shown at figure 7. as the volume fraction of nanoparticle increases from 0.02 to 0.1, the outlet average temperature of the microchannel output increases about 1 to 2 degree centigrade. adding nanoparticles increases the conduction heat transfer coefficient as it was predicted by equation (7). when the nanofluid flows through the microchannel the both convection and conduction heat transfer play roles to convey heat from the walls to the nanofluid. adjacent the walls, where the no-slip condition is applied, the conduction heat transfer is dominant; therefore, the nanoparticles journal of mechanical engineering and technology 74 issn: 2180-1053 vol. 10 no.2 june – december 2018 with a higher heat conduction coefficient improve the heat transfer in this rejoin. however, near the microchannel center line, where the wall affection is negligible, the convection heat transfer plays the main role. for this, addition of nanoparticles to the base fluid results in the viscosity increasing (see equation 6) which in turn, reduces the strength of vortices circulation and the convective heat transfer. the decreasing in the slope of outlet average temperature versus the volume fraction is because of the affection of viscosity increment and its influence on the strength of vortices. 4.0 conclusions in this study, a numerical modelling of the heat transfer of nanofluids flow of water/alumina through a microchannel includes fins was investigated. the effect of numbers and height of the baffles, reynolds number, and the volume fraction of nanoparticles in the nanofluid on the heat transfer characteristic were studied. the governing equations were solved by finite element method used by ansys fluent software. the results show that the increase in the number of baffles leads to an increase in the number of vortices, which augments the heat transfer between the nanofluid and the microchannel walls. the results also show that an increasing in the reynolds number causes a decreasing in the microchannel average outlet temperature. as the height and number of the baffles increases, the extent of the nanofluid involved in the vortices zone increases and, subsequently; the heat transfer increases. finally, it was observed that an increase in the volume fraction of nanoparticles increases the average outlet temperature as the result of the heat transfer conduction increase. 5.0 references akbarinia, a., abdolzadeh, m. & laur, r. (2011). critical investigation of heat transfer enhancement using nanofluids in microchannels with slip and non-slip flow regimes. applied thermal engineering, 31(4), 556-565. alfaryjat, a.a., mohammed, h.a., adam, n.m., stanciu, d. & dobrovicescu, a. (2018). numerical investigation of heat transfer enhancement using various nanofluids in hexagonal microchannel heat sink. thermal science and engineering progress, 5, 252-262. ambreen, t. & kim, m.h. (2018). effects of variable particle sizes on hydrothermal characteristics of nanofluids in a microchannel. international journal of heat and mass transfer, 120, 490-498. azizi, z., alamdari, a. & malayeri, m.r. (2016). thermal performance and friction factor of a cylindrical microchannel heat sink cooled by cu-water nanofluid. applied thermal engineering, 99, 970-978. bar-cohen, a. (2013(. gen-3 thermal management technology: role of microchannels and nanostructures in an embedded cooling paradigm. journal of nanotechnology in engineering and medicine, 4(2), 020907. numerical investigation of the water/alumina nanofluid within a microchannel with baffles issn: 2180-1053 vol. 10 no.2 june – december 2018 75 chari, s. & kleinstreuer, c. (2018). convective mass and heat transfer enhancement of nanofluid streams in bifurcating microchannels. international journal of heat and mass transfer, 125, 1212-1229. chein, r. & huang, g. (2005). analysis of microchannel heat sink performance using nanofluids. applied thermal engineering, 25(17-18), 3104-3114. colgan, e.g., furman, b., gaynes, m., graham, w.s., labianca, n.c., magerlein, j.h., polastre, r.j., rothwell, m.b., bezama, r.j., choudhary, r. & marston, k.c. (2007). a practical implementation of silicon microchannel coolers for high power chips. ieee transactions on components and packaging technologies, 30(2), 218-225. dixit, t. & ghosh, i. (2015). review of micro-and mini-channel heat sinks and heat exchangers for single phase fluids. renewable and sustainable energy reviews, 41, 1298-1311. ganvir, r.b., walke, p.v. & kriplani, v.m. (2017). heat transfer characteristics in nanofluid—a review. renewable and sustainable energy reviews, 75, 451-460. hajmohammadi, m.r., alipour, p. & parsa, h. (2018). microfluidic effects on the heat transfer enhancement and optimal design of microchannels heat sinks. international journal of heat and mass transfer, 126, 808-815. lee, j. & mudawar, i. (2009). low-temperature two-phase microchannel cooling for high-heat-flux thermal management of defense electronics. ieee transactions on components and packaging technologies, 32(2), 453-465. lewis, j.m. & wang, y. (2018). two-phase frictional pressure drop and water film thickness in a thin hydrophilic microchannel. international journal of heat and mass transfer, 127, 813-828. li, z., tao, w.q. & he, y.l. (2006). a numerical study of laminar convective heat transfer in microchannel with non-circular cross-section☆. international journal of thermal sciences, 45(12), 1140-1148. maı̈ga, s.e.b., nguyen, c.t., galanis, n. & roy, g. (2004). heat transfer behaviours of nanofluids in a uniformly heated tube. superlattices and microstructures, 35(3), 543-557. manay, e. & sahin, b. (2016). the effect of microchannel height on performance of nanofluids. international journal of heat and mass transfer, 95, 307-320. minkowycz, w.j., sparrow, e.m. and abraham, j.p. (2016). nanoparticle heat transfer and fluid flow. crc press. sakanova, a., yin, s., zhao, j., wu, j.m. & leong, k.c. (2014). optimization and comparison of double-layer and double-side micro-channel heat sinks with journal of mechanical engineering and technology 76 issn: 2180-1053 vol. 10 no.2 june – december 2018 nanofluid for power electronics cooling. applied thermal engineering, 65(1-2), 124-134. tuckerman, d.b. & pease, r.f.w. (1981(. high-performance heat sinking for vlsi. ieee electron device letters, 2(5), 126-129. zhang, r., chen, z., xie, g. & sunden, b. (2015). numerical analysis of constructal water-cooled microchannel heat sinks with multiple bifurcations in the entrance region. numerical heat transfer, part a: applications, 67(6), 632-650. preparation of papers in a two column model paper format journal of mechanical and engineering technology issn 2180-1053 vol. 12 no.1 1 june – december 2020 the influence of current on laser welding on mechanical properties and microstructures of dissimilar metal joints syamsul ma’arif11, bambang irjanto2, rena juwita sari3 1 department of mechanical engineering, proklamasi 45 university, yogyakarta, indonesia 2,3 department of petroleum engineering, proklamasi 45 university, yogyakarta, indonesia abstract the joints of dissimilar metals have several problems because they have different mechanical properties and microstructures. this problem can be solved by using laser welding because the resulting energy focus and the hot zone around the welding joint are minimal. the parameters that are important in welding either using welding laser or other welding process is the current. thus, this study examined the joint of dissimilar metals, i.e., low carbon steel (mild steel a36) and stainless steel (stainless steel 304) using laser welding. the result of the laser welding joint with a variation of current tested its mechanical properties with tensile test and seen its microstructure with photo scanning electron microscopy (sem). tensile test results on the current variable 410 440 ampere obtained fracture is on the side of mild steel a36, while at 450 ampere current obtained a fault on the laser welding joints, where the point of the break is below the point of broken mild steel a36. based on the tensile test, the suitable current for nonlinear steel joint of a36 mild steel with stainless steel 304 is 410 440 ampere. in the result of the sem photo, it is seen that on all variables of the current is obtained a photo of the porosity. based on the microstructure, it is obtained that the results of laser welding are not perfect. this causes easy corrosion and cracks occur so that the laser weld tensile test results are initially good, after some time the results will go down. keywords: dissimilar metal, laser welding, current, tensile test, scanning electron microscopy (sem) 1.0 introduction the development of material technology requires a durable but lightweight material, not easily corroded and can be formed under industry needs (noh, zin, alnasser, yusoff, & yusof, 2017). currently, there are many material joints, especially in dissimilar metallic materials. for example, the use of non-metal joints is in the application of heat exchangers, i.e., there is a metal joint between carbon steel and stainless steel (corleto & argade, 2017). the use of non-aliphatic metals is in the automotive and aircraft industries, i.e., the use of non-metal joints between aluminum and steel (zhao, ren, zhao, pan, & guo, 2017). the use of a joint of other unlike metals is an aluminum joint with titanium (tomashchuk, et al., 2017). in the marine industry, the stainless-steel materials from seawater using a dissimilar metal joint between austenite material and duplex (ramkumar, et al., 2017). dissimilar metals have different material properties, including melting point differences, tensile strength differences, differences in corrosion resistance, and so on. 1 corresponding author. email: arief.syams@up45.ac.id / arief.up45@gmail.com mailto:arief.syams@up45.ac.id mailto:arief.up45@gmail.com journal of mechanical and engineering technology issn 2180-1053 vol. 12 no.1 1 june – december 2020 uncoated metal joint technology will result in changes in mechanical, chemical, and thermal properties of the material (martinsen, hu, & carlson, 2015). connecting two different materials poses several problems, that are the metallurgical and thermophysical problems that can arise from connecting (oliveira, et al., 2017). to know the mechanical properties of the nonwoven metal joints with the laser weld, a tensile test (enz, et al., 2017) was used, while the microstructure of the dissimilar metal joint was performed with a scanning electron microscopy (sem) (xu & zhang, 2016). to reduce the change in the properties of dissimilar metal joints, the process of welding uses lasers. the advantages of using laser welding versus arc welding are the minimal heat-exposed zones, rapid cooling and solidification poses, can reach the problem dimensions, the energy that the focus generates and are able to melt two types of material at different points of melting (meco, cozzolino, ganguly, williams, & mcpherson, 2017). in the process of laser welding and other welding processes, weld geometry (penetration depth and weld width) can be controlled with parameters of laser power and travel speed (ayoola, suder, & williams, 2017). in this research, we will connect dissimilar metals such as low carbon steel (mild steel a36) and stainless steel (stainless steel 304) using laser welding. to know the mechanical properties of different types of laser weld joints is done by tensile test, while to see the microstructure is done with photo scanning electron microscopy (sem). the laser welding machine used is the ht-wy 180-mk automatic laser welding machine with current variables: 410 a, 420 a, 430 a, 440 a, and 450 a, while the plate thickness is 2 mm. 2.0 materials and methodology 2.1 materials the materials used in this study are a36 stainless steel (mild steel) and stainless steel 304. the carbon steel used is a36 carbon steel is often also called black steel carbon. the stainless steel used is austenitic stainless steel, which is stainless steel 304. both materials are a sheet-shaped plate with a thickness of 2 mm. the raw materials can be seen in figure 1. a. mild steel a36 b. stainless steel 304 figure 1. raw materials journal of mechanical and engineering technology issn 2180-1053 vol. 12 no.1 1 june – december 2020 the preparation of raw materials into specimens with dimensions is based on the standard of astm e 646 98 or as per standard tensile testing specimen in mm. details of the workpiece size can be seen in figure 2. plate with a thickness of 2 mm is tough to be made by a frais machine. so the process of making specimens uses laser cutting. neater laser cutting results also result in a high degree of precision. samples that have been made with laser cutting can be seen in figure 3. figure 2. standard dimension of the tensile testing specimen figure 3. the specimen result from the laser cutting process after the specimen is ready, then the next stage is the process of grafting two dissimilar materials, that are mild steel a36 with stainless steel 304 using laser welding. the laser welding machine used is the ht-wy 180-mk automatic laser welding machine, as shown in figure 4. this laser welding device uses a water cooling medium, so every few times the welding process, the cooling water must be replaced. figure 4. automatic laser welding machine ht-wy 180-mk during the welding process, the welding material uses is stainless steel. the selection of welding materials based on the metal that has a better tensile strength that is stainless steel ( journal of mechanical and engineering technology issn 2180-1053 vol. 12 no.1 1 june – december 2020 2.2 methodology briefly, after the implementation of research after the welding process with laser welding, then the next stage is the test tensile strength (tensile test) and seeing the microstructure with photo scanning electron microscopy (sem). the results of tensile tests and photos of microstructure were then analyzed and concluded. the welding process using laser welding is carried out with variations of current, ranging from 410 a, 420 a, 430 a, 440 a, and 450 a. each variation of current is performed three times. the process of laser welding can be seen in figure 5. as for the welded metals, the result of the joint of dissimilar metal can be seen in figure 6. figure 5. laser welding process using a laser welding machine figure 6. the result of the lasers on dissimilar metal joints the size of the specimen, as shown in figure 6, refers to the standard dimension of astm e 646 98 specimens tensile testing, as shown in figure 2. the next step is the test. to know the mechanical properties of dissimilar metal joints to the laser welding machine. it is done by tensile test. meanwhile, to see the quality of the joint is by looking at the microstructure on the joint of dissimilar metal using sem photo. in addition to knowing the joint condition, sem photos can be used to view changes in microstructure and chemical element changes due to heat during the welding process. this test equipment can be seen in figure 7. a. tensile test machine b. sem photograph equipment figure 7. the testing equipment of mechanical and microstructure journal of mechanical and engineering technology issn 2180-1053 vol. 12 no.1 1 june – december 2020 3.0 result and discussion 3.1 tensile test results and discussion based on the test results through the tensile test, it can be known that the maximum stress (ultimate strength) and the stress when the specimen starts to crack (fracture). the location of ultimate strength and fracture is shown in figure 8 (reviewer c). the maximum stress (σmax) of stainless steel specimens (without joints), low carbon steel (without joints), and the joining of stainless steel and low carbon steel with variations of current can be seen in table 1. as for knowing the result of fracture stress (σf), see table 2. the average of table 1 and table 2 is showed in table 3. figure 8. the location of ultimate strength and fracture in the stress-strain diagram table 1. the value of maximum stress (σmax) of tensile test results no material current (a) σmax (n/m 2) test 1 test 2 test 3 1 stainless steel (ss 304) 450 16,483 2 mild steel (a36) 450 8,568 3 ss 304 + a36 410 8,744 8,603 8,621 4 ss 304 + a36 420 8,656 8,799 8,850 5 ss 304 + a36 430 8,674 8,621 8,842 6 ss 304 + a36 440 8,780 9,009 8,921 7 ss 304 + a36 450 8,285 8,674 8,250 table 2. the value of fracture stress (σf) of tensile test results no material current (a) σf (n/m 2) test 1 test 2 test 3 1 stainless steel (ss 304) 450 12,723 2 mild steel (a36) 450 5,178 3 ss 304 + a36 410 5,529 5,542 5,025 4 ss 304 + a36 420 5,053 5,433 5,335 5 ss 304 + a36 430 5,187 5,184 5,450 6 ss 304 + a36 440 4,982 5,475 5,411 7 ss 304 + a36 450 4,940 5,385 4,954 σmax σf ultimate strength fracture yield strength journal of mechanical and engineering technology issn 2180-1053 vol. 12 no.1 1 june – december 2020 table 3. average of maximum stress (σmax) and fracture stress (σf) no material current (a) average σmax (n/m 2) σf (n/m 2) 1 stainless steel (ss 304) 450 16,483 12,723 2 mild steel (a36) 450 8,568 5,178 3 ss 304 + a36 410 8,656 5,365 4 ss 304 + a36 420 8,768 5,274 5 ss 304 + a36 430 8,712 5,274 6 ss 304 + a36 440 8,903 5,289 7 ss 304 + a36 450 8,403 5,093 based on table 3, we can make a graph, as shown in figure 9 for the value of maximum stress (σmax) and figure 10 for the value of the fracture stress (σf). figure 9. comparison of maximum stress (σmax) to currents figure 10. comparison of fracture stress (σf) to current journal of mechanical and engineering technology issn 2180-1053 vol. 12 no.1 1 june – december 2020 based on figure 9, it can be seen that in the current of 410 a up to 440 a, the maximum stress value of the uniform metal joint (blue line) is higher than the maximum stress value on mild steel a36 (red line). while at a current of 450 a, its maximum stress value falls to lower than the maximum stress on mild steel a36. all metal joints are not similar from variations of current; their maximum stress value is far below the maximum stainless steel 304 value. thus it can be said that the joints between stainless steel and low carbon steel, based on its tensile strength is lower than stainless steel but higher than low carbon steel. based on figure 10, the fracture stress values of the current from 410 a up to 440 a are rated above the fracture stress at mild steel a36. while at a current of 450 a, the stress when it breaks below the stress when broken mild steel a36. thus, based on the discussions of figures 9 and figure 10, the recommended current for laser welding machines on unbonded metal joints with a 2 mm thick plate is 410 a up to 440 a. 3.2 the sem photo result and discussion these are the results of micro-photo structure using photo scanning electron microscopy (sem) with a sample of each variation of current. there are five (5) photos, as shown in figure 11. a. current 410 a b. current 420 a c. current 430 a d. current 440 a e. current 450 a figure 11. results of sem photographs of laser welding seen from various variations of current during the welding process based on figure 11, it can be seen that all laser welding results of various current variations have porosity. this can be caused by the power of the laser welding machine is less stable, or protective gas is used less than the maximum. the porosity that occurs causes the power to withstand the voltage (tensile test) to drop (drop). therefore, the result of the tensile test of stainless steel joint with low carbon steel obtained lower voltage very far below the voltage on stainless steel. journal of mechanical and engineering technology issn 2180-1053 vol. 12 no.1 1 june – december 2020 4.0 conclusions laser welding machines require a very large current, but the resulting melting power is minimal. proven with currents of 410 a up to 450 a is only able to provide penetration (melt) welding and filling added materials for specimens with a thickness of 2 mm. the results of the laser welding tensile test on the joint of different type of material (stainless steel and low carbon steel) resulted that the current of 410 a up to the current of 440 a has the maximum stress and fracture stress above low carbon steel (a36), but still far below stainless steel (ss 304). while the tensile test results on a current of 450 a, the maximum stress and fracture stress is under low carbon steel (a36) and stainless steel (ss 304). recommended current for different type material joints, i.e., stainless steel (ss 304) with low carbon steel (a36) is 410 a up to 440 a. 5.0 acknowledgement the authors would like to thank the ministry of research, technology, and higher education of the republic of indonesia for funding this activity through the beginner lecturers research (pdp) scheme in 2017. the authors also thank the inlastek welding institute surakarta who has assisted the implementation of this research. 6.0 references ayoola, w. a., suder, w. j., & williams, s. w. (2017, june). parameters controlling weld bead profile in conduction laser welding. journal of materials processing tech., 249, 522 530. corleto, c. r., & argade, g. r. (2017, may 24). failure analysis of dissimilar weld in heat exchanger. case studies in engineering failure analysis, 9, 27 34. enz, j., kumar, m., riekehr, s., ventzke, v., huber, n., & kashaev, n. (2017, october). mechanical properties of laser beam welded similar and dissimilar aluminum alloys. journal of manufacturing processed, 29, 272 280. martinsen, k., hu, s. j., & carlson, b. e. (2015). joining of dissimilar materials. cirp annals, 64(2), 679 699. meco, s., cozzolino, l., ganguly, s., williams, s., & mcpherson, n. (2017). laser welding of steel to aluminium: thermal modelling and joint strength analysis. journal of materials processing tech., 247, 121 133. noh, f. s., zin, h. m., alnasser, k., yusoff, n., & yusof, f. (2017). optimization of laser lap joining between stainless steel 304 and acrylonitrile butadiene styrene (abs). procedia engineering, 184, 246 250. oliveira, j. p., zeng, z., andrei, c., fernandes, f. m., miranda, r. m., ramirez, a. j., zhou, n. (2017, august). dissimilar laser welding of superelastic niti and cualmn shape memory alloys. materials & design, 128, 166 175. ramkumar, k. d., dagur, a. h., kartha, a. a., subodh, m. a., vishnu, c., arun, d., abraham, j. (2017, december). microstructure, mechanical properties and biocorrosion behavior of dissimilar welds of aisi 904l and uns s32750. journal of manufacturing processes, 30, 27 40. journal of mechanical and engineering technology issn 2180-1053 vol. 12 no.1 1 june – december 2020 tomashchuk, i., sallamand, p., measson, a., cicala, e., duband, m., & peyre, p. (2017, july). aluminum to titanium laser welding-brazing in v-shaped groove. journal of materials processing technology, 245, 24 36. xu, w.-f., & zhang, z.-l. (2016, december). microstructure and mechanical properties of laser beam welded tc4/ta15 dissimilar joints. transactions of nonferrous metals society of china, 26, 3135 3146. zhao, d., ren, d., zhao, k., pan, s., & guo, x. (2017, december). effect of welding parameters on tensile strength of ultrasonic spot welded joints of aluminum to steel – by experimentation and artificial neural network. journal of manufacturing processes, 30, 63 74. issn: 2180-1053 vol. 2 no. 1 january-june 2010 adaptive gain controller using model reference adaptive control method stability approach for road vehicle following system 55 adaptive gain controller using model reference adaptive control method stability approach for road vehicle following system m. r. sapiee1, a. noordin1 1faculty of electrical engineering, universiti teknikal malaysia melaka, locked bag 1752, pejabat pos durian tunggal, 76109 durian tunggal, melaka. email: 1mohd.razali@utem.edu.my abstract in order to maintain stability and satisfy operating constraints, the control system on the following vehicle needs information about the motion of preceding vehicle. a one-vehicle look-ahead control strategy is proposed and will be investigated for this operation. a mathematical model for this control strategy is obtained and simulated. this paper describes the process of designing an adaptive gain controller for a road vehicle following system. this is done through simulations and is further discussed to find the effectiveness of the method. keywords: one-vehicle look-ahead control, model reference adaptive control, stability approach. 1.0 introduction in general, malaysian drivers tend to follow another vehicle closely. hence, platoons or convoys appeared to develop rapidly. this normally happens when spacing between vehicles is close and there is no chance for the following vehicle to overtake the preceding vehicle. the vehicle at the back or the following vehicle will have to adjust its speed and spacing with respect to the preceding vehicle. if this is not taken into careful consideration, collision between them may occur. in order to avoid any collisions between the following vehicle and the preceding vehicle, it is necessary to maintain some safe distances between both vehicles at any speed. so, the following vehicle needs to have information regarding to the speed and the distance between them. in a normal driving, the driver of the following vehicle will estimate the information and adjust his vehicle’s speed and position to issn: 2180-1053 vol. 2 no. 1 january-june 2010 journal of mechanical engineering and technology 56 have a nominal speed and a safe distance between his vehicle and the preceding vehicle. nowadays, all the information does not need to be estimated. sensors are available to measure the speed of the preceding vehicle and the position of the preceding vehicle with respect to the following vehicle. the information from the sensors will be used and processed by the following vehicle controller to produce the required speed and the safe spacing distance. this is where autonomous control can take place. the controller can automatically ensure safe distance based on the information obtained from the preceding vehicle. the autonomous controller on the following vehicle can activate the vehicle cruise control mode, where the driver does not need to hold the steering nor press the fuel pedal, and automatically apply the brake when necessary in order to ensure the safety of the vehicle. in developed nations, the autonomous concept leads to the intelligent vehicle highway system (ivhs). as a vehicle enters the highway, his vehicle automatically takes-over the control of the vehicle while following the preceding vehicle. this feature also gives rise to steering less technology where during the autonomous control in action, the driver does not need to hold the steering wheel. all the driving tasks are taken care by the vehicle intelligent system. one of the autonomous features is the adaptive type control based on certain control strategy which gives rise to adaptive cruise control (acc). an acc controlled vehicle will follow the front vehicle at a safe distance. a model reference adaptive control (mrac) can be used in this type of control where the vehicle controller has the ability to adapt to the variation of speed and position of the preceding vehicle. 2.0 one-vehicle look-ahead control strategy when a vehicle follows another vehicle in front of it, a vehicle convoy or a vehicle following system is formed as shown in figure 1. issn: 2180-1053 vol. 2 no. 1 january-june 2010 adaptive gain controller using model reference adaptive control method stability approach for road vehicle following system 57 following vehicle front vehicle figure 1 a vehicle following system the following vehicle can be controlled in such a way that it will maintain either the same speed to that of the immediate preceding vehicle, which is the front vehicle or maintain a safe distance in order to avoid collision between them. this is where the string stability plays an important role by having a string stable vehicle following system. the system is said to be stable if the range errors decrease as they propagate along the vehicle stream. in this control strategy, the controlled vehicle only refers the information from the preceding vehicle. so, the control system on the following vehicle needs information about the motion of preceding vehicles. yanakiev, d. and kanellakopoulos, i. used a simple spring-massdamper system to demonstrate the idea of string stability and show the string-stability criterion for constant time-headway and variable timeheadway policies. a mathematical modeling for this control strategy is shown in figure 2 where i denotes the following vehicle and i-1 denotes the preceding vehicle. vehicle i vehicle i-1 kp kv ix •• 1− •• ix ix 1−ix m i mi-1 figure 2 mathematical model of a vehicle following system control strategy performing the mathematical modelling on only the following vehicle i and applying the newton’s second law results in the following equation (1). assuming unit mass for equation (1) and taking laplace transform, gives the following transfer function. issn: 2180-1053 vol. 2 no. 1 january-june 2010 journal of mechanical engineering and technology 58 the transfer function in equation (2) depends on the following spacing policy. the aim of this strategy is to maintain string stability for longitudinal motion within the vehicle following system or the vehicle convoy, particularly between a vehicle with a vehicle or between a vehicle with a following vehicle. this strategy is adopted in order to design a controller by investigating the following two policies. 2.1 spacing policy researchers are trying to get close inter-vehicular spacing between vehicles, in order to have an effective following system. spacing policy is defined as a rule that dictates how the speed of an automatically controlled vehicle must regulate as a function of the following distance. a control system should be designed such that it regulates the vehicle speed according to the designed spacing policy. there are three basic spacing policies employed by many researchers thus far. those spacing policies are the fixed spacing policy, the constant time headway policy and the variable time policy. only the first two will be discussed and investigated in this paper. 2.1.1 fixed spacing policy under this policy, a fixed inter-vehicular spacing is implemented regardless of the vehicle’s speed. a well-known result states that it is impossible to achieve string stability in an autonomous operation when this spacing policy is adopted. this is mainly due to the relative spacing error that does not attenuate as it propagates down the string at all frequencies. spacing error attenuation will only occur for frequencies above certain level. in addition, keeping the same fixed spacing at different convoy speed would risk the safety and comfort of passengers, especially when the vehicles are closely separated. obviously, at higher speed, faster vehicle reaction time is needed in an emergency situation to avoid collision. nevertheless, the fixed spacing policy can give guaranteed string stability if the front vehicle provides its information on its speed and or position to the rear vehicle in the convoy. this can be done through radio communication or the rear issn: 2180-1053 vol. 2 no. 1 january-june 2010 adaptive gain controller using model reference adaptive control method stability approach for road vehicle following system 59 vehicle having sensors to detect the above two parameters. equation (2) shown before is the transfer function for the fixed spacing policy. 2.1.2 fixed headway spacing policy this spacing policy keeps a fixed time interval, called time headway or headway, h, between the preceding vehicle and the following vehicle. it is a speed dependent policy where the inter-vehicular spacing will vary according to the preceding vehicle speed. at higher speed, vehicles will be separated in a greater distance but always maintains a fixed time interval between vehicles. most researchers used this spacing policy in designing controllers to ensure string stability as this policy mimics the behavior of human drivers. as vehicle speed is increased, a human driver will keep a safe inter-vehicular spacing with the immediate preceding vehicle. the performance of the fixed headway spacing policy used in autonomous and cooperative vehicles following systems has been studied. it is found that there exists minimum possible fixed headway spacing before the string stability of a convoy collapses which is related to the actual dynamics of the vehicle. the effect of this fixed headway spacing policy is equivalent to the introduction of additional damping in the transfer function, which allows the poles of the transfer function to be moved independently from the zeros of the same transfer function. with the addition of the fixed headway spacing, equation (2) then becomes with the control law developed as to simplify the control law and at the same time ensure stability, a pole-zero cancellation technique is chosen. this can be achieved by introducing the constraint issn: 2180-1053 vol. 2 no. 1 january-june 2010 journal of mechanical engineering and technology 60 thus, figure 3 shows the pole-zero maps with pole and zero cancelling each other. through the pole-zero cancellation technique, equation (3) is thus reduced to which is a first order transfer function with a stable response with time delay. while kv must always be positive, there exists a pole which is always in the left hand side of the s-plane. thus, the system is always stable for this spacing policy. thus, h k 1 v (6) figure 3 shows the pole-zero maps with pole and zero cancelling each other. through the pole-zero cancellation technique, equation (3) is thus reduced to v v 1i i 2 ks k sx sx sg (7) which is a first order transfer function with a stable response with time delay. while kv must always be positive, there exists a pole which is always in the left hand side of the s-plane. thus, the system is always stable for this spacing policy. figure 3 pole-zero cancellation map 2.2 inclusion of vehicle dynamics after proving that the fixed time headway policy is suitable to be adopted, a simplified vehicle dynamics model is introduced in order to mimic the actual vehicle internal dynamics. in this case, the external dynamics is not considered. in the simplified model, the internal dynamics is represented as a lag function i.e., the actual vehicle acceleration is obtained after a certain time delay . this is given by the relation in equation (8). uaa (8) -2 -1.8 -1.6 -1.4 -1.2 -1 -0.8 -0.6 -0.4 -0.2 0 -1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 1 pole-zero map real axis im ag in ar y a xi s pole and zero cancelling each other figure 3 pole-zero cancellation map 2.2 inclusion of vehicle dynamics after proving that the fixed time headway policy is suitable to be adopted, a simplified vehicle dynamics model is introduced in order to mimic the actual vehicle internal dynamics. in this case, the external dynamics is not considered. in the simplified model, the internal dynamics is represented as a lag function i.e., the actual vehicle acceleration is obtained after a certain time delay . this is given by the relation in equation (8). issn: 2180-1053 vol. 2 no. 1 january-june 2010 adaptive gain controller using model reference adaptive control method stability approach for road vehicle following system 61 equation (3) is modified to include the vehicle dynamics part and this gives a transfer function in equation (9). having designed the control strategy and by including the vehicle dynamics, the block diagram of the one-vehicle look-ahead vehicle following system is shown in figure 4. control strategy vehicle dynamics vi-1 ai vi xi ui xi-1 figure 4 block diagram consisting of the control strategy and vehicle dynamics if τ is so small (as in an ideal vehicle) i.e. , then . hence, the transfer function is reduced back to a second order transfer function. equation (10) is then simulated by using matlab simulink giving a speed input for h=1s and kp values of 0.5, 1, 2 and 3. shown in figure 5 and figure 6 are the speed and acceleration responses, respectively. further analysis of figure 5 shows that kp value of 2 gives the best response. issn: 2180-1053 vol. 2 no. 1 january-june 2010 journal of mechanical engineering and technology 62 figure 5 speed response of fixed headway with various kp values figure 6 acceleration response of fixed headway with various kp values 3.0 model reference adaptive control an adaptive controller can modify its behavior in response to changes in the dynamics of a system and the character of any disturbance. it is a controller with adjustable parameter and a mechanism for adjusting the parameter. an adaptive control system consists of two loops, normal issn: 2180-1053 vol. 2 no. 1 january-june 2010 adaptive gain controller using model reference adaptive control method stability approach for road vehicle following system 63 feedback loop with plant and controller and an adaptive parameter mechanism loop. figure 7 illustrates the general structure of a model reference adaptive control (mrac) system. figure 7 general structure of an mrac system the basic mrac system consists of four main components: i) plant to be controlled ii) reference model to generate desired closed loop output response iii) controller that is time-varying and whose coefficients are adjusted by adaptive mechanism iv) adaptive mechanism that uses ‘error’ (the difference between the plants and the desired model output) to produce controller coefficient regardless of the actual process parameters, adaptation in mrac takes the form of adjustment of some or all of the controller coefficients so as to force the response of the resulting closed-loop control system to that of the reference model. therefore, the actual parameter values of the controlled system do not really matter. 3.1 the stability approach the mrac can be designed such that the globally asymptotic stability of the equilibrium point of the error difference equation is guaranteed. to do this, the lyapunov second approach or stability approach is used. the term stability approach is used throughout this paper. it requires an appropriate lyapunov function to be chosen, which could be difficult. this approach has stability consideration in mind and is also known as the lyapunov approach. in designing the mrac controller, we would like the output of the closed-loop system (y) to follow the output of the reference model (ym). therefore, we aim to minimise the error (e=y-ym) by designing a controller that has one or more adjustable parameters such that a certain cost function is minimized. e(t) ym(t) y(t) u(t) r(t) figure 7 general structure of an mrac system the basic mrac system consists of four main components: i) plant to be controlled ii) reference model to generate desired closed loop output response iii) controller that is time-varying and whose coefficients are adjusted by adaptive mechanism iv) adaptive mechanism that uses ‘error’ (the difference between the plants and the desired model output) to produce controller coefficient regardless of the actual process parameters, adaptation in mrac takes the form of adjustment of some or all of the controller coefficients so as to force the response of the resulting closed-loop control system to that of the reference model. therefore, the actual parameter values of the controlled system do not really matter. 3.1 the stability approach the mrac can be designed such that the globally asymptotic stability of the equilibrium point of the error difference equation is guaranteed. to do this, the lyapunov second approach or stability approach is used. the term stability approach is used throughout this paper. it requires an appropriate lyapunov function to be chosen, which could be difficult. this approach has stability consideration in mind and is issn: 2180-1053 vol. 2 no. 1 january-june 2010 journal of mechanical engineering and technology 64 also known as the lyapunov approach. in designing the mrac controller, we would like the output of the closed-loop system (y) to follow the output of the reference model (ym). therefore, we aim to minimise the error (e=y-ym) by designing a controller that has one or more adjustable parameters such that a certain cost function is minimized. 4.0 adaptive gain controller design an adaptive gain controller is to be designed for the one-vehicle lookahead control strategy with fixed time headway and vehicle dynamics by applying a model reference adaptive control (mrac). this section presents a direct adaptive controller design which adapts the unknown vehicle parameter kp. the advantage of the adaptive approach is that unpredictable changes in the value of kp can be easily accommodated. figure 8 on the other hand shows the one-vehicle look-ahead control strategy based controller with the gain parameter to be adjusted, kp is the gain adjustment mechanism to be designed in the control strategy. 4.0 adaptive gain controller design an adaptive gain controller is to be designed for the one-vehicle look-ahead control strategy with fixed time headway and vehicle dynamics by applying a model reference adaptive control (mrac). this section presents a direct adaptive controller design which adapts the unknown vehicle parameter kp. the advantage of the adaptive approach is that unpredictable changes in the value of kp can be easily accommodated. figure 8 on the other hand shows the one-vehicle lookahead control strategy based controller with the gain parameter to be adjusted, kp is the gain adjustment mechanism to be designed in the control strategy. figure 8 a one-vehicle look-ahead controller with adjustable gain 4.1 the stability approach design from the analysis of figure 6, kp value of 2 gives the best response. so, it will be used in equation (10) to give a reference model to be used in designing the adaptive gain controller. the vehicle dynamic has been included in the control law to form the plant. plant: r ksk1s ks y pp 2 p (11) reference model: r 2s3s 2s y 2m (12) 1 x''i 1 h product 1 kvadd2 add1 add 5 kp 4 xi 3 x'i 2 x'i-1 1 xi-1 xi x'i x"i figure 8 a one-vehicle look-ahead controller with adjustable gain 4.1 the stability approach design from the analysis of figure 6, kp value of 2 gives the best response. so, it will be used in equation (10) to give a reference model to be used in designing the adaptive gain controller. the vehicle dynamic has been included in the control law to form the plant. plant: issn: 2180-1053 vol. 2 no. 1 january-june 2010 adaptive gain controller using model reference adaptive control method stability approach for road vehicle following system 65 reference model: a closed-loop system with a controller has the following parameters: r(t) = reference input signal u(t) = control signal y(t) = plant output ym(t) = reference model output e(t) = difference between plant and reference model output = y(t) ym(t) in designing an mrac using stability approach, the following steps should be followed: i) derive a differential equation for error, e = y − ym that contains the adjustable parameter, . from equations (11) and (12), after replacing, the differential equation becomes substituting equations (13) and (14) into , thus let , so that . thus ii) find a suitable lyapunov function, usually in a quadratic form (to ensure positive definiteness). issn: 2180-1053 vol. 2 no. 1 january-june 2010 journal of mechanical engineering and technology 66 the lyapunov function, , is based on equation (16). , where is positive definite. the derivative of v becomes where for stability must be negative i.e. . iii) derive an adaptation mechanism based on such that e goes to zero. therefore, the block diagram implementation is given in figure 9 where is denoted by gamma in the simulation diagram and the red dotted line is the adaptive mechanism. iii) derive an adaptation mechanism based on x,e,ev such that e goes to zero. rex 3 1 (18) therefore, erk p (19) the block diagram implementation is given in figure 9 where is denoted by gamma in the simulation diagram and the red dotted line is the adaptive mechanism. figure 9 the stability approach adaptive gain controller simulation diagram. the mrac stability approach adaptive gain controller design is then simulated again using matlab simulink. both the output of the system responses (y and ym) are shown in figure 10 and figure 11. figure 10 shows a perfect model following output while figure 11 on the other hand shows the acceleration response of y where it does not follow a sharp change in input acceleration. ui x'i xi vehicle dynamics -kgamma xi-1 x'i-1 x'i xi kp x''i control strategy s+2 s +3s+22 t ransfer fcn product 1 s integrator2 1 s integrator du/dt derivative add 1 x'i-1 y m e y yy figure 9 the stability approach adaptive gain controller simulation diagram. the mrac stability approach adaptive gain controller design is then simulated again using matlab simulink. both the output of the system responses (y and ym) are shown in figure 10 and figure 11. figure 10 shows a perfect model following output while figure 11 on the other hand shows the acceleration response of y where it does not issn: 2180-1053 vol. 2 no. 1 january-june 2010 adaptive gain controller using model reference adaptive control method stability approach for road vehicle following system 67 follow a sharp change in input acceleration. figure 10 comparison of y and ym for speed from the stability approach figure 11 comparison of y and ym for acceleration from the stability approach it can be seen that the output response for plant y perfectly follows the reference model ym. it can be said that the system is a perfect model following system. the adaptive gain controller is again simulated but this time with gamma γ values of 0.1,0.01 0.001 and 0.0001 while the value of h is fixed at 1s. 0 20 40 60 80 100 120 140 160 180 200 0 5 10 15 20 25 time (s) s pe ed ( m /s ) ym y 0 20 40 60 80 100 120 140 160 180 200 -1.5 -1 -0.5 0 0.5 1 1.5 time (s) a cc el er at io n (m /s 2 ) ym y figure 10 comparison of y and ym for speed from the stability approach figure 10 comparison of y and ym for speed from the stability approach figure 11 comparison of y and ym for acceleration from the stability approach it can be seen that the output response for plant y perfectly follows the reference model ym. it can be said that the system is a perfect model following system. the adaptive gain controller is again simulated but this time with gamma γ values of 0.1,0.01 0.001 and 0.0001 while the value of h is fixed at 1s. 0 20 40 60 80 100 120 140 160 180 200 0 5 10 15 20 25 time (s) s pe ed ( m /s ) ym y 0 20 40 60 80 100 120 140 160 180 200 -1.5 -1 -0.5 0 0.5 1 1.5 time (s) a cc el er at io n (m /s 2 ) ym y figure 11 comparison of y and ym for acceleration from the stability approach it can be seen that the output response for plant y perfectly follows the reference model ym. it can be said that the system is a perfect model following system. the adaptive gain controller is again simulated but this time with gamma γ values of 0.1,0.01 0.001 and 0.0001 while the value of h is fixed at 1s. issn: 2180-1053 vol. 2 no. 1 january-june 2010 journal of mechanical engineering and technology 68 figure 12 speed response with various gamma values for adaptive gain controller figure 13 acceleration response with various gamma values for adaptive gain controller shown in figure 12 and figure 13 are the speed and acceleration responses for the adaptive gain controller, respectively. further analysis of figure 13 gamma value of between 0.01 and 0.001 gives the best response by almost fitting the reference model with smooth curve and with jerk of less than 5 m/s3. 0 20 40 60 80 100 120 140 0 5 10 15 20 25 time (s) s pe ed ( m /s ) ym gamma=0.1 gamma=0.01 gamma=0.001 gamma=0.0001 0 20 40 60 80 100 120 140 -1.5 -1 -0.5 0 0.5 1 1.5 time (s) a cc el er at io n (m /s 2 ) ym gamma=0.1 gamma=0.01 gamma=0.001 gamma=0.0001 figure 12 speed response with various gamma values for adaptive gain controller figure 12 speed response with various gamma values for adaptive gain controller figure 13 acceleration response with various gamma values for adaptive gain controller shown in figure 12 and figure 13 are the speed and acceleration responses for the adaptive gain controller, respectively. further analysis of figure 13 gamma value of between 0.01 and 0.001 gives the best response by almost fitting the reference model with smooth curve and with jerk of less than 5 m/s3. 0 20 40 60 80 100 120 140 0 5 10 15 20 25 time (s) s pe ed ( m /s ) ym gamma=0.1 gamma=0.01 gamma=0.001 gamma=0.0001 0 20 40 60 80 100 120 140 -1.5 -1 -0.5 0 0.5 1 1.5 time (s) a cc el er at io n (m /s 2 ) ym gamma=0.1 gamma=0.01 gamma=0.001 gamma=0.0001 figure 13 acceleration response with various gamma values for adaptive gain controller shown in figure 12 and figure 13 are the speed and acceleration responses for the adaptive gain controller, respectively. further analysis of figure 13 gamma value of between 0.01 and 0.001 gives the best response by almost fitting the reference model with smooth curve and with jerk of less than 5 m/s3. a one-vehicle look-ahead control strategy with fixed headway policy has been adopted in designing a controller to produce an output which can respond immediately to the change in input; in this case, the input issn: 2180-1053 vol. 2 no. 1 january-june 2010 adaptive gain controller using model reference adaptive control method stability approach for road vehicle following system 69 is the speed with varying speed conditions. with normal controller, the response does not quite match perfectly with the input. with the introduction of the mrac adaptive gain controller, the response can be made to follow the input by choosing a suitable reference model. furthermore, using mrac adaptive gain controller produces a smooth output as compared to the other one. 5.0 conclusion the adaptive gain controller tuning has been investigated using mrac concepts through the stability approach. simple adaptation law for the controller parameters has been presented assuming that the process under control can be approximated by a second order transfer function. the developed adaptation rule has been applied and simulated. the results obtained show the effectiveness of the technique. the resulting performance could be improved by a better choice of the length of the adaptation period. the stability approach is used to provide guaranteed nominal stability. however, the stability approach controller can only have very small gain with 1 is the maximum limit. a further limitation of the approach is the assumption of a structure for the nominal system. in this paper, a second order transfer function has been assumed resulting from the assumption of a very small time delay between the command signal and the vehicle dynamics as in ideal vehicle. 6.0 acknowledgments the authors would like to acknowledge their gratitude to faculty of electrical engineering, universiti teknikal malaysia melaka for providing the resources and support in the completion of this paper. 7.0 references k. pirabakaran, v. m. becerra. (2001). automatic tuning of pid controllers using model reference adaptive control techniques. iecon’01, the 27th annual conference of the ieee industrial electronics society. pp. 736-740. issn: 2180-1053 vol. 2 no. 1 january-june 2010 journal of mechanical engineering and technology 70 liang c. y. and peng h. (1999). optimal adaptive cruise control with guaranteed string stability’. vehicle system dynamics. volume 31. pp 313-330. m. r. sapiee, h. selamat, a. noordin, a. n. jahari (2008). pi controller design using model reference adaptive control approaches for a chemical process. proceedings of 2008 student conference on research and development (scored 2008). 26-27 nov. 2008. pp. 155-1 to 155-4. p.a. cook and s. sudin. (2003). convoy dynamics with bidirectional flow of control information. 10th, ifac symposium on control in transportation systems, tokyo, japan. 4-6 august 2003. pp. 433-438. p.a. cook and s. sudin. (2002). dynamics of convoy control systems. 10th ieee mediterranean conference on control and automation, lisbon, portugal. 9-12 july 2002. wp7-2. s. sudin and p.a. cook. (2003). dynamics of convoy control systems with twovehicle look-ahead strategy. international conference on robotics, vision, information and signal processing, universiti sains malaysia, penang, malaysia. 22-24 january 2003. pp. 327-332. s. sudin and p.a. cook. (2004). two-vehicle look-ahead convoy control systems. 59th, ieee vehicular technology conference, milan, italy. 17-19 may 2004. volume 5. pp. 2935-2939. s. sudin (2005). dynamics and control of vehicle convoy systems. phd thesis. school of electrical and electronic engineering, faculty of engineering and physical sciences, the university of manchester. yanakiev, d. and kanellakopoulos, i. (1996). a simplified framework for string stability analysis in ahs. proc. of the 13th ifac world congress. volume q. pp.177-182. the effect of carbon nanotube wall thickness on elastic modulus of nanocomposite n. kordani1*, r. adibipour2, a. sadough vanini2 1department of mechanical engineering, university of mazandaran, mazandaran, iran 2department of mechanical engineering, amirkabir university of technology, tehran, iran abstractthis paper focuses on effect of carbon nanotubes physical parameter on elastic modulus of nanocomposites. the remarkable properties of at least some of nano particles have led to high research in the field of nanocomposites, especially carbon nanotubes. in this paper, polymer matrixes and carbon nanotubes are interest. at the nano scale, the structure of the carbon nanotube strongly influences the overall properties of the composite. some well-known theories such as halpin-tsai equation, shear lag model and modified mixture of low were employed to consider the efficient of carbon nanotubes physical parameter on elastic modulus of nanocomposites. according to the results, addition of volume fraction of carbon nanotubes caused a reduction of elastic modulus. the nanocomposite elastic properties are particularly sensitive to the nanotube diameter, with increasing on diameter and wall thickness of carbon nanotube the elastic modulus decreases and when length of carbon nanotube is increasing, the elastic modulus increases. keywords: mechanical properties, nanocomposites, polymer matrix composites, aspect ratio, carbon nanotube 1. introduction the development of nano-particle reinforced polymer composites is newly one of the most favorable approaches in the field of future engineering applications. the remarkable properties of at least some of these nano particles have led to high research in the field of nanocomposites, especially carbon nanotubes (sreejarani & ray, 2011; baharvandi & et. al., 2017; kordani & sadough, 2014). of the various nano-particles, carbon nanotubes (cnts) have attracted great interest newly as structural reinforcements because of their unique properties. cnts with their notable mechanical properties such as low density, high aspect ratio, high strength and stiffness, excellent electrical and chemical resistance are a potential candidate as reinforcement for polymeric materials. the addition of only small quantity of nano particle (specially cnts) leads to improved mechanical properties of matrix (baharvandi & et. al., 2016; guozhong, 2004; harris, 1999). the most important of properties of single-walled carbon nanotubes (swcnt) and multi-walled carbon nanotubes (mwcnt) are collected on table 1. table 1 typical properties of swcnt and mwcnt. property swcnt mwcnt ref. diameter (nm) 0.4-5 5–50 (micah & et.al, 2009) aspect ratio 100–10,000 100–10,000 (saleh & sundararaj, 2011) density (𝑔 𝑐𝑚3⁄ ) ~1.3 ~1.75 (saleh & sundararaj, 2011) tensile strength (gpa) 50–500 10–60 (saleh & sundararaj, 2011) elastic modulus (tpa) ~1 (from 1 to 5) 0.2-0.95 (meo & rossi, 2006; yu & et.al, 2000) journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 2 june – december 2019 17 corresponding author:naser.kordani@umz.ac.ir mailto:naser.kordani@umz.ac.ir failure strain (%) ~5 − 10 10.5, up to 12 (liu & et.al, 2005; chowdhury & et.al, 2012) cnts have many structures, differing in length, diameter, thickness, spiral types and number of layers that are used to modify the other materials. mwnts and swnts are the most popular type of cnts. in 1991, mwnts consist of many coaxial graphite cylindrical tubes and in 1993, swnts with one graphite cylindrical tube were discovered by iijima. mwcnts and swcnts were discovered in the soot of the arc-discharge method and using of metal catalysts in the arc-discharge method (natsuki & tantrakan, 2004; khare & bose, 2005). tem micrograph of a mwnt is shown in figure 1. figure 1 tem micrograph of a mwnt with measurements of outside diameter, inside diameter and wall thickness (thostenson & et. al, 2001). 2. influence of physical parameters of cnts on elastic modulus of nanocomposite at the nano scale, the structure of the carbon nanotube strongly influences the overall properties of the composite. for design purposes, we need to have simple and rapid calculative procedures for estimating the effective properties. some wellknown theories such as halpin-tsai equation, shear lag model and modified mixture of low were used. their equations depend on a parameter which is considered, table. 2. table 2 theories on mechanical properties of nanocomposite with their parameter model equation ref. halpin-sai 𝐸𝑐 = 𝐸𝑚 ( 1+𝜁𝜂𝑣𝑓 1−𝜂𝑣𝑓 ), 𝜂 = 𝐸𝑓 (𝐸𝑚) −1 𝐸𝑓 (𝐸𝑚) +𝜁 , 𝜁 = 2𝑙 𝑑⁄ (halpin & tsai, 1967) shear-lag 𝐸𝑐 = 𝜂𝐸𝑓 𝑣𝑓 + 𝐸𝑚 . (1 − 𝑣𝑓 ), 𝜂 = 1 − tanh (𝑘 𝑙 𝑑 ) 𝑘 𝑙 𝑑 , 𝑘 = √ 2𝐸𝑚 𝐸𝑓(1+𝜈)ln ( 1 𝑉𝑓⁄ ) (kashyap & et.al, 2011) modified mixture low 𝐸𝑐 = 𝑋1𝜂𝐸𝑓 𝑣𝑓 + 𝐸𝑚 𝑣𝑚, 𝜂 = 1 − 𝑡𝑎𝑛ℎ (𝑘 𝑙 𝑑 ) 𝑘 𝑙 𝑑 , 𝑋1(2𝐷, 3𝐷) = 3/8,1/5 (crutis & et.al, 1978) journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 2 june – december 2019 18 where 𝐸𝑐 , 𝐸𝑓 , 𝐸𝑚 are the modulus of nanocomposite, cnts and matrix. ζ is called shape factor that is depend on the particular elastic property. l is average length and d is average diameter of nanotube and 𝜈 is poisson ratio. 𝑣𝑓 is volume fraction. base on experimental results, thostenson and chou ploted a linear line through the data that shows relationship between the nanotube diameter and wall thickness, figure 2. according their study at smaller nanotube diameters this relationship between the nanotube diameter and wall thickness begins to deviate from the linear curve fit (thostenson & et. al, 2001). figure 2 linear relationship between wall thickness and nanotube diameter by (thostenson & et. al, 2001). 𝑇 = −3.0793 + 0.4796𝑑, 𝑅 = 0.98687 (1) in which, t is wall thickness and d is nanotube diameter. r is the error of this curve fitting. eq. (1), substituted into shape factor of halpin-tsai equation to show the effective of wall thikness on elastic modulus of nanocomposite. 𝜁 = 0.9592𝑙 𝑇 + 3.0793 (2) we also used eq. (2) for other theories to consider the effect of wall thickness of cnts. halpin-tsai equation and experimental results by montazeri et al. (2010), were used to consider the influence of nanotube diameter, length and volume fraction on the elastic modulus of nanocomposite. results are shown in figures 3 and 4. as shown in figures 3 and 4, increasing on diameter and wall thickness make the elastic modulus of nanocomposite decreases and increasing on weight percent and wall length of cnts make the elastic modulus of nanocomposite increases. journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 2 june – december 2019 19 figure 3 the effect of nanotube diameter, d, length, l and weight percent, wt% on the elastic modulus of nanocomposite. figure 4 the effect of nanotube wall thickness, t, length, l and weight percent, wt% on the elastic modulus of nanocomposite. by using the theories on table 2 and experimental results by (montazeri & et al., 2010), effect of diameter on the composite elastic modulus of nanocomposite are predicted, and results are shown in figure 5. modified mixture of low in 2d and 3d are inefficient to consider the wall thickness of cnts on elastic modulus of nanocomposite. journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 2 june – december 2019 20 figure 5 effect of diameter on the elastic modulus of nanocomposite by using halpin-tsai equation (h-t), shear-lag theory (sh-l), and modified mixture of low in 2 dimension(mml,2d) and 3 dimension(mml,3d). 3. conclusion in this paper by using the experimental data, halpin-tsai equation, shear lag model and modified mixture of low, effects of reinforcement, length and diameter of carbon nanotube on the mechanical properties were investigated. according to the results, additional cnts weight percent caused an increase on elastic modulus. the elastic properties of nanocomposite are particularly sensitive to the nanotube diameter. by increasing the diameter and wall thickness of carbon nanotube, the elastic modulus decreases and when length of carbon nanotube is increasing, the elastic modulus increases. effect of geometry parameter on elastic properties of nanocomposite is better shown by halpin-tsai equation and shear-lag theory. references al-saleh, m.h., sundararaj, u. (2011). review of the mechanical properties of carbon nanofiber/polymer composites. composites, part a 42, 2126–2142. baharvandi, h., alebooyeh, m., alizadeh, m., saeedi heydari, m., kordani, n. & khaksari, p. (2016). the influences of particle–particle interaction and viscosity of carrier fluid on characteristics of silica and calcium carbonate suspensions-coated twaron® composite. journal of experimental nanoscience, vol. 11, no. 7, 550_563. baharvandi, h., saeedi heydari, m., kordani, n., alebooyeh, m., alizadeh, m., khaksari, p. (2017). characterization of the rheological and mechanical properties of shear thickening fluid-coated twaron® composite. the journal of the textile institute, vol. 108, no. 3, 397-407. chowdhury, s.c., haque, b.z., okabe, t., gillespie, j.w. (2012). modeling the effect of statistical variations in length and diameter of randomly oriented cnts on the properties of cnt reinforced nanocomposites. composites,part b 43,1756–1762. journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 2 june – december 2019 21 crutis, p.t., bader, m.g., bailey, j.e. (1978). the stiffness and strength of a polyamide thermoplastic reinforced with glass and carbon fibers. journal of material science, 13, 377-390. guozhong, c. (2004). nanostructures and nanomaterials: synthesis, properties and applications. imperial college press, london. halpin, j. c., tsai, s.w. (1967). environmental factors in composite materials design. us air force technical report afml tr, 67–423. harris, p., f. j. (1999). carbon nanotubes and related structures: new materials for the 21st century. cambridge university press. kashyap, k.t., koppad praveennath, g., puneeth, k.b., aniruddha ram, h.r., mallikarjuna, h.m. (2011). elastic modulus of multi-walled carbon nanotubes reinforced aluminium matrix nanocomposite. computational materials science, 8, 2493-2495. khare, r., bose, s. (2005). carbon nanotube based compositesa review. journal of minerals & materials characterization & engineering, 4, 31-46. kordani, n., sadough, a. (2014). different method to make laminates by shear thickening fluid. science and engineering of composite materials (secm), vol. 21, no. 3, 421-425. liu, l., barber, a.h., nuriel, s., wanger, h.d. (2005). mechanical properties of functionalized single-walled carbon-nanotube/poly (vinyl alcohol) nanocomposites. advanced functional materials, 15, 975-980. meo, m., rossi, m. (2006). prediction of young's modulus of single wall carbon nanotubes by molecular-mechanics-based finite element modeling. composites science and technology, 66, 1597–1605. micah, j., green behabtu n., pasquali, m., wade adams, w. (2009). nanotubes as polymers. polymer, 50, 4979–4997. montazeri, a., javadpour, j., khavandi, a., tcharkhtchi, a., mohajeri, a. (2010). mechanical properties of multi-walled carbon nanotube/epoxy composites. materials and design, 31, 4202–4208. natsuki, t., tantrakan, k.m. (2004). effects of carbon nanotube structures on mechanical properties. applied physics, 79, 117-124. sreejarani, k., pillai ray, s.s. (2011). epoxy-based carbon nanotubes reinforced composite. national centre for nano-structured materials, csir, south africa, 726792. journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 2 june – december 2019 22 thostenson, e.t., ren, z., chou, t.w. (2001). advances in the science and technology of carbon nanotubes and their composites: a review. composites science and technology, 61, 1899–912. yu, m.f., lourie, o., dyer, m.j., moloni, k., kelly, t.f., ruoff, r.s. (2000). strength and breaking mechanism of multiwalled carbon nanotubes under tensile load. science, 287, 637–640. journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 2 june – december 2019 23 http://www.ingentaconnect.com/content/els/02663538;jsessionid=1op4kp203avl9.victoria http://www.ingentaconnect.com/content/els/02663538;jsessionid=1op4kp203avl9.victoria preparation of papers in a two column model paper format journal of mechanical engineering and technology *corresponding author. email: azmmi@utem.edu.my issn 2180-1053 vol. 11 no. 1 july-december 2019 30 effect of tensile load on electrical resistivity of stretchable conductive ink (sci) n. a. b. masripan1,2*, s. j. lim1, g. omar1,2, m.a. salim1,2, m.z. akop1,2 , a.a kamarolzaman1,2, n. tamaldin1, , a. nurfaizey1,2, r. nadlene1,2, s. fadzullah1,2, a.m. saad1,2 , s. jasmee1, m.b. ramli1,2, m. n. a. nordin1,2 1 faculty of mechanical engineering, universiti teknikal malaysia melaka, hang tuah jaya, 76100 durian tunggal, melaka 2, centre for advanced research on energy, universiti teknikal malaysia melaka, hang tuah jaya, 76100 durian tunggal, melaka abstract to date, research has tended to focus on emerging electrical conductive adhesive (eca) with stretchable and flexible substrate or known as stretchable conductive ink (sci). sci is more flexible, stretchable and multi-purpose compare with the traditional printed circuit. limitation on the chatacreization of sci performance especially on it electrical performane under tensile stress has motivate this study. the aim of this research is to investigate the conductivity of the conductive ink under tensile stress at different elongation. the conductive ink carbon black was used to print on the thermoplastic polyurethane (tpu) and cure in the oven at 120°c for 30 minutes. the conductive ink was clamp using in-house stretching equipment with different elongation. the resistivity was measured by four-point probe while surface structure was observed by using axioscope 2 mat microscope. the result shows that the resistance increased when the elongation increased. for 40mm length of conductive ink, the initial resistance is 0.562 kω and its become 1.217 kω when stretch until 18% of its initial length. the sheet resistance of the conductive ink also increased due to the defection (porosity) on the surface of conductive ink after stretching. the strain level for 40mm and 60mm also increase form 0.14 to 0.16 that cause incerase in resistance. however, since there are no crack/defection observes at 80mm after maximum elongaton, the resistance start to decrease that cause increase in sci conductivity. keywords: conductive ink, resistance, tensile load 1.0 introduction the revolution of technology has created many opportunities for the discovery of many multi-functional devices that made lives easier, faster and better. new devices are being developed every day, and each of these devices may be able to handle variations in size and functionality. with all of these revolutions, it is always an interest among manufacturer to create a more complex device that have high functionality but smaller in size. in electronic industry, the introduction of modern technology like conductive ink to replace the use of copper wire has able to create a device that is more flexible, smaller and multi-purpose. example of conductive that has been used in flexible circuit device are metal-based inks, conductive polymers and carbon complex (tran, dutta, & choudhury, 2018). for metal-based ink, there are a few type of metal that has been using as a filler for the conductive ink such as copper and silver. these types of ink journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 1 july-december 2019 31 have high conductivity and they are commonly used in traditional solid-state electronic. however, metal-based ink are high-cost and will be oxidize under ambient condition (pekarovicova & husovska, 2016). the research to date has tended to focus on emerging eca with stretchable and flexible substrate or known as stretchable conductive ink (sci) by using conductive filler especially form carbon based like carbon nano tube (cnt), graphene and carbon black. it have very good electrical conductivity by using high charge mobility to conduct electricity (grandea, et. al., 2012). conductive polymers are the creation of the mobility to charge on the polymers backbone so that it can conduct electricity. the example of conductive polymers is polyacetylene. it compress pellets to arrange as conjugated structure to exhibit the electronic conductivity (ramakrishnan, 2011). these several types of conductive ink are needed to be printed on the surface so that it can connect the electronic product together. few kinds of printing techniques have been developed to achieve the fabrication process which are ink-jet printing, screen printing, and gravure printing (tran, dutta, & choudhury, 2018). among these types of printing, screen printing is the most common printing technique in the industries process due to its compatibility. screen printing is also low in cost, scalable and able to produce both fixed and flexible thin-film compare with the others printing technique (cao, et. al, 2014). although ink-jet printing is high cost but it has high registration accuracy so that it can produce a fine product (sirringhaus, et. al, 2000). although the development of conductive ink had growth rapidly in the electronic industry, this technology is still not fully replacing the conventional soldering method due to the limitation of low electrical conductivity, low life-cycle, and low stretchability. therefore, the use of stretchable and flexible polymer like thermoplastic polyurethane (tpu) has gain a lot of attention to use as medium to apply conductive ink. tpu is known for its flexibility and stretchability properties that may exhibit strain up to 100%. thus, combining the mechanical properties of tpu with conductive ink is believed to able to provide a more flexible. researcher had been continuously researching about conductive ink under stretching condition and improve the stretching ability without affect the electrical conductivity but still the study still limited and not widely explored yet. besides, the conductivity of the sci will affect with the elongation of the conductive ink during stretching. this is because, the resistance will increase after many stretching cycles due to the deformation of the stretchable conductive ink (su, 2017). hence, the aim of this study is to figure out the conductivity of the stretchable conductive ink under tensile stress and improve the stretchability without changing the resistivity of the conductive ink. 2.0 methodology the thermoplastic polyurethane (tpu) used in this study is commercially available tpu that purchased from takeda sangyo that have thickness of 100µm with optically transparent polyester film. the bare conductive ink, carbon black is printed on the tpu with stencil by using screen printing technique. the conductive ink is printed at different length which is 40 mm, 60 mm and 80 mm but with same width (5.0 mm) and thickness (0.2 mm). later, the substrate undergoes curing process by heating at temperature of 120°c in the oven for 30 minutes. after the curing process, the tpu is stretch until reach 28% from its original length by using inhouse stretching test as journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 1 july-december 2019 32 shown on figure 1 and the resistivity was measured at every 2% interval by using fourpoint probe at four different point. further analysis was done on all the samples in order to qualitatively observe the morphology of the sample surface using image analyzer (axioscope 2mat microscope). figure 1. in-house streching test set-up 3.0 result and discussion figure 2 shows the results for resistivity of conductive ink carbon black printed on tpu at different length (40, 60 and 80)mm. the result reveals that the longer length of printed conductive ink on tpu substrate, the lower the resistivity of the sci. this is due to the fact that, the rate of cracking and broken is decreased as the length of printed conductive ink increased. this occurance can be observe physically with naked eye in figure 3. during the experiment, 40 mm length of conductive ink cracked at 14% (5.6 mm) of elongation and broke at 20% (8 mm) of elongation while for 60 mm length, the conductive ink cracked at 16% and does not break until reach 28% of its initial length. for 80 mm of conductive ink, there are no crack occur even after stretch up to 28% of its initial length. this show that the formation of crack on printed conductive ink after strecthing affect the rate of electron flow between the carbon black particles to decrease thus resulting lower conductivity. figure 2. resistivity of 40 mm, 60 mm and 80 mm length at different elongation journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 1 july-december 2019 34 figure 3. physical appearance of printed conductive ink on tpu after strecthing further analysis to study the structural behaviour of the sci when stretching up to 28% of elongation was done by using image analyser (axioscope 2mat microscope). the microstructure of the conductive ink at length of 40 mm shows a smooth surface without any imperfection at initial state as shown in table 1(a) . after stretching, there are lot of porosity that clearly found on the surface of the conductive ink due to the deformation of conductive ink when stretching as shown in table 1(b). moreover, there are a crack in the middle of the conductive ink which represent the broken part of the conductive ink. all of the imperfections are affecting the conductivity of conductive ink. at 60 mm length of conductive ink, some porosity also could be found after stretching up to 28% of its intial length as shown in table 1(d). however, the porosity are not as distinct as founded in 40 mm length. while for length of 80 mm, there are no defection occur at the surface of the conductive ink when comparing the result of microtrusture before (table 1(e)) and after (table 1(f)) stretching. this phenomenon shows that a surface with no defect will give small gap between the particles that create conductive paths allowing the ease of electron movements as explained by merilampi (merilampi, laine-ma, & ruuskanen, 2009). therefore, the longer the length of printed conductive ink, the higher the conductivity of the conductive ink. table 1. microstructure image of printed conductive ink before and after stretching sci length 40 60 80 before (a) (c) (e) 40 mm 60 mm 80 mm journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 1 july-december 2019 34 after figure 4 is the comparison study of strain level between 40 mm, 60 mm and 80 mm length of printed conductive ink. strain was measured to evaluate the amount an object deforms as a result of a force. it was measured when the change in a dimension (length) is divided by the original value of that length. the cracking point of 40 mm length of printed conductive ink is at 5.6 mm of elongation with the strain level of 0.14. meanwhile, the cracking point of 60 mm length of printed conductive ink is at 9.6mm of elongation with the strain level of 0.16. based on the previous study using four-point probe analyser to measure the resistivity of the stretchable conductive ink under stretching condition, as the strain level increase, the sheet resistance incresae as well due to the cracking on the surface of the conductive ink (park, lee, kwon, nam, & choa, 2018). however, as for 80 mm length, no cracking occur after elongation up to 28% and the resistance was decrease eventhough the strain was predicted to occur at more higher of strain level which is 0.2 and above. figure 4. graph of strain against elongation between different length of conductive ink 4.0 conclusion in summary, this paper succesfully investigate the effect of tensile stress on the different length of conductive ink printed on the tpu. the sheet resistance increase as the the length of printed conductive ink decrese due to the effect of porosity formation at the conductive ink surfaces. the formation of porosity cause the printed conductive ink to crack and reduce the conductive path. however, when the length of sci is increase, the (b) (d) (f) porosity porosity no porosity journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 1 july-december 2019 35 resistivity was decrease that cause the conductiviti of sci increase since there are no crack observed. 5.0 acknowledgement the authors gratefully acknowledge to universiti teknikal malaysia melaka (utem) for providing laboratory facilities. 6.0 references cao, x., chen, h., gu, x., liu, b., wang, w., cao, y., zhou, c. (2014). screen printing as a scalable and low-cost approach for rigid and flexible thin-film transistors using separated carbon nanotubes. acs nano, 8(12), 12769-12776. grandea, l., chundi, v. t., wei, d., bower, c., andrew, p., & ryhänen, t. (2012). graphene for energy harvesting/storage devices and printed electronics. particuology, 10(1), 1-8. merilampi, s., laine-ma, t., & ruuskanen, p. (2009). the characterization of electrically conductive silver ink patterns on flexible substrates. microelectronics reliability, 49(7), 782-790. park, j. y., lee, w. j., kwon, b. s., nam, s. y., & choa, s. h. (2018). highly stretchable and conductive conductors based on ag flakes and polyester composites. microelectronic engineering, 199, 16-23. pekarovicova, a., & husovska, v. (2016). printing ink formulations. in j. izdebska, & s. thomas, printing on polymers: fundamentals and applications (pp. 41-55). william andrew. ramakrishnan, s. (2011). from a laboratory curiosity to the market place. resonance, 16(12), 1254-1265. sirringhaus, h., kawase, t., friend, r. h., shimoda, t., inbasekaran, m., wu, w., & woo, e. p. (2000). high-resolution inkjet printing of all-polymer transistor circuits. science, 290(5499), 2123-2126. su, c. (2017). environmental implications and applications of engineered nanoscale magnetite and its hybrid nanocomposites: a review of recent literature. journal of hazardous materials, 322(part a), 48-84. tran, t. s., dutta, n. k., & choudhury, n. r. (2018). graphene inks for printed flexible electronics: graphene dispersions, ink formulations, printing techniques and applications. advances in colloid and interface science, 261, 41-61. preparation of papers in a two column model paper format journal of mechanical engineering and technology *corresponding author. email: ridzuan@utem.edu.my issn 2180-1053 vol. 11 no.1 july – december 2019 1 effect of line width and thickness on flexible printed electronic circuit electrical performance m. r. mansor1*, m. k. sulaiman2, r. n. h. raja norazli3, s. a. azli4, s. h. s. m. fadzullah5 and g. omar6 1,2,3,4,5,6, faculty of mechanical engineering, universiti teknikal malaysia melaka, hang tuah jaya, 76100 durian tunggal, melaka, malaysia 1 centre for advanced research on energy, universiti teknikal malaysia melaka, hang tuah jaya, 76100 durian tunggal, melaka, malaysia abstract flexible and printable electronics is among the rapidly growing field in many applications. their performances are affected by many factors such as the interaction between the conductive ink circuit and the type of flexible substrate used as the printed board. in this paper, the effect of the conductive ink circuitry line width and thickness to the flexible printed electronic (fpe) electrical performance is investigated. commercial type carbon based conductive ink and polyethylene terephthalate (pet) flexible substrate were applied to formulate the fpe circuit, using screen printing technique and cured at room temperature, with varying circuitry line width (between 1.00 mm to 3.00 mm) and thickness (between 0.05 mm to 0.25 mm). square shape circuit pattern was also utilized. the final resistivity for all samples were later tested using digital multimeter. results for the experiments showed that the electrical resistivity of the fpe samples were alost inversely proportional to the dimension of the circuit thickness and width. the results obtained shall be used in the next project stage as benchmarking data to establish design guidelines related to circuitry geometrical parameters to obtain optimum fpe electrical performance in actual application. keywords: flexible printed electronic circuit, conductive ink, line width, line thickness, electrical performance 1.0 introduction flexible printed electronics (fpe) can be defined as printing of circuits and electrical components on flexible substrate such as film, paper or textiles. as stated in previous study by mitsui et al. (2015), the electrical connections that used flexible circuits stands as one amongst the most advanced and important technologies as because of their useful and beneficial properties as example is their marginal weight, flexible form factors and thinness. for the application in whole devices, it helps in boosting the device functionality by provide portability, freedom of form or flexible, impact resistance, reducing the product weight and thickness. similar study by chang et al. (2014) stated that the development of printed electronic technology is prompted by its quality potential of being low-cost, high volume, high-throughput production of electronic components or devices which are lightweight and small, think and flexible, and inexpensive. in time to come, printed electronics will emerge innovative and journal of mechanical engineering and technology issn 2180-1053 vol. 11 no.1 july – december 2019 2 disposable products such as point-of-care applications, diagnosis, power sources, biosensors, and smart/interactive packaging. khirotdin et al. (2016) reported that the ability of a material to conduct an electric current is measured by the electrical conductivity. a material of higher resistivity will present a lower conductivity and vice versa. there are few important factors that affect the mechanical and electrical properties and must be taken into use such as the polymer matrix of the ink and electrical conductivity. all these factors play a big role on the quality of the ink and on the control of the production parameters. however, the most influential production parameters affecting conductivity and mechanical properties are the substrate material surface structure, curing parameters (time and temperature), materials composition, and the cross-section area of the conductive layer. up to date, there are several finding reported on the performance of fpe at varying type of conductive paste, subtrates and circuit geometry. happonen (2016) reported on the reliability performance of fpe subjected to cyclic bending load. silver based conductive ink was used as the conductor, while plastic and paper were used as the substrates. the silver based fpe electrical performance was also studied in term of varying circuitry width and thickness. elsewhere, janeczek et al. (2012) studied the effect of different silver nanoparticle based conductive filler size to the electrical performance of fpe. they concluded that smaller nanoparticle size provided higher durability compared to micro-meter particle size when subjected to varying cyclic bending load. furthermore, merilampi et al. (2009) reported on the effect of varying ink patterns to the electrical behaviour of fpe under cyclic bending load. their study showed that higher number of cyclic bending raised the fpe resistivity. in addition, they also stated that composition of the conductive ink and the size of the conductive filler also affected both electrical and mechanical properties of fpe. the failure mechanism of fpe was also reported by dai et al. (2015). they concluded that there are mainly four (4) interfacial failure modes associated with fpe which are cracking, slipping, delamination in the slipping zone and delamination. the cracking failure occurred due to rupture on the conductive thin film, where as the remainder of the failure modes occurred at the interface between the substrate and the conductive thin film. another potential type of filler in producing conductive paste for fpe is carbon black. carbon balck offers many advantages especially in term of balance between cost and performance (good electrical conductivity and good mechanical properties) which are comparable to conventional silver-based nanoparticles. however, based on current literature review, there are still limited reports on studies involving the relationship between conductive layer geometrical parameters to the electrical performance of fpe made from carbon-based conductive paste. rozali et al. (2018) used carbon based conductive paste to produce fpe using stretchable thermoplastic polyurethane (tpu) substrate using single line circuit geometry with fix circuit thickness and width. in another report, suhaimi et al. (2018) used carbon based conductive paste to produce fpe using stretchable thermoplastic polyurethane (tpu) substrate using also single line circuit geometry but with varying circuit thickness. hence, in this paper, the effect of carbon based conductive layer line width and line thickness to the fpe electrical behaviour is studied for square shaped circuit pattern. varying square shaped circuit line width and line thickness were formulated using the material onto thin flexible substrate. screen printing technique was employed to fabricate the test samples and the electrical behaviour of the fpe was measured in term of their electrical resistivity journal of mechanical engineering and technology issn 2180-1053 vol. 11 no.1 july – december 2019 3 2.0 methodology 2.1 sample preparation the optically clear polyester film pet of 100µm thickness was obtained from lohmann technologies uk ltd. the carbon based conductive ink was obtained from bare conductive ltd, uk, with density of 1.16 g/cm3 and surface resistance of 55 /sq. the sample preparation consists of two sets which are set a (varies in thickness) and set b (varies in width) as shown in table 1. set-a setup, the fix parameter was the line width and variable parameter was line thickness (0.05 mm until 0.25 mm, with 0.05 mm interval). for set-b setup, the fix parameter was the line thickness and variable parameter was line width (1.0 mm until 3.0 mm, with 0.5 mm interval). the readily engrave circuit design adhesive tape was pasted on pet substrate. prior to the screenprinting process, the substrate was wiped using isopropyl alcohol (ipa) to remove contaminants. after that, conductive ink was poured onto the substrate surface and aligned evenly. screen printing process conducted were based on astm d2739 (astm, 2015). all samples were cured at room temperature for approximately 30 minutes. figure 1 shows the geometrical circuit design parameter and sample of the fabricated fpe. table 1. sample preparation test plan set a set b width (mm) thickness (mm) width (mm) thickness (mm) 2.0 2.0 2.0 2.0 2.0 0.05 0.10 0.15 0.20 0.25 1.0 1.5 2.0 2.5 3.0 1.0 0.05 0.05 0.05 0.05 2.0 2.0 2.0 2.0 2.0 0.10 0.10 0.15 0.20 0.25 1.0 1.5 2.0 2.5 3.0 1.5 0.05 0.05 0.05 0.05 2.0 2.0 2.0 2.0 2.0 0.15 0.10 0.15 0.20 0.25 1.0 1.5 2.0 2.5 3.0 2.0 0.05 0.05 0.05 0.05 2.0 2.0 2.0 2.0 2.0 0.20 0.10 0.15 0.20 0.25 1.0 1.5 2.0 2.5 3.0 2.5 0.05 0.05 0.05 0.05 2.0 2.0 2.0 2.0 2.0 0.25 0.10 0.15 0.20 0.25 1.0 1.5 2.0 2.5 3.0 3.0 0.05 0.05 0.05 0.05 (a) journal of mechanical engineering and technology issn 2180-1053 vol. 11 no.1 july – december 2019 4 (b) figure 1. (a) square shaped geometrical circuit design parameter (happonen et al., 2016), and (b) sample of square shaped flexible printed circuit. 2.2 electrical testing the digital multimeter was used to read the resistivity of the circuit with accuracy of ± (1.0% + 1). the resistance was first measured by connecting test lead to “com” terminal and the red test lead to the “vohm” input terminal, followed by setting the function range switch to the ohm range. finally, test leads were connected across the resistance under measurement and the final resistance value was taken based on the given reading. the initial resistance was measured by taking 3 readings for each sample to get the average value. all tests were performed at room temperature. figure 2 shows the resistivity measurement location taken on the samples (at the termination points of the sample). figure 2. resistivity measurement location on sample 3.0 results and discussion figure 3 and figure 4 show the comparison of resistivity between thickness and width of conductive ink circuit for square shaped fpe circuit pattern. based on figure 3, the sample with thicker layer conductive ink (0.25 mm) showed lowest resistivity (0.751 kω) compared to the sample with a thinner layer of conductive ink (0.05mm) that has higher resistivity (3.766 kω). this indicates that increasing conductive ink thickness resulted in lower resistance reading. meanwhile based on figure 4, the largest width circuit (3.0 mm) gave the lowest resistance value (1.301 kω) while the smallest journal of mechanical engineering and technology issn 2180-1053 vol. 11 no.1 july – december 2019 6 width circuit (1.0 mm) showed highest resistance value (6.961 kω), which indicates a wider circuit line caused less resistance compared to thinner circuit line. from both graph, the average resistance of conductive ink sample was found to be almost inversely proportional to the dimension of the circuit line width and thickness. figure 3. average resistance of varies conductive layer thickness figure 4. average resistance of varies conductive layer width similar pattern of performance as shown in figure 3 for fpe from carbon black at varying circuit thickness was also reported by suhaimi et al. (2018). in their study, single straight line fpe using tpu subtrate were fabricated with varying circuit thickness between 1 layer until 10 layers of conductive paste. they concluded that as the conductive circuit thickness increase, the resistivity and sheet resistivity decreased. 4.0 summary in conclusion, it was verified that the width and thickness of carbon conductive ink on the pet substrate affect the resistivity of the overall fpe. from both results, the average resistance of conductive ink sample was found to be almost inversely proportional to the dimension of the circuit line, which provided good correlation information especially for fpe circuit design performance estimation. further works shall be carried out to further characterize the electrical and thermal performance of the carbon based conductive ink on pet polymer substrate especially when subjected to varying cyclic bending load journal of mechanical engineering and technology issn 2180-1053 vol. 11 no.1 july – december 2019 7 5.0 acknowledgement the authors acknowledge with thanks to the fund and support from the universiti teknikal malaysia melaka (utem) and ministry of education, malaysia. this project was funded by utem industrial research grant gluar/jabil/2016/fkm-care/i00017. 6.0 references astm d2739 (2015). standard test method for volume resistivity of conductive adhesives 1, american standard testing method, 97(4306), 2010–2012. chang, j., zhang, x., ge, t. & zhou, j. (2014). fully printed electronics on flexible substrates: high gain amplifiers and dac. organic electronics, 15(3), 701–710. dai, l., huang, y., chen, h., feng, x. & fang, d. (2015). transition among failure modes of the bending system with a stiff film on a soft substrate. applied physics letter, 106(2), 021905. happonen, t. (2016). reliability studies on printed conductors on flexible substrates under cyclic bending. (ph.d thesis), university of oulu. happonen, t. vouitilanen, j-v., hakkinen, j. & fabritius, t. (2016). the effect of width and thickness on cyclic bending reliability of screen-printed silver traces on a plastic substrate. in ieee transactions on component, packaging and manufacturing technology, may 2016, 6(5), (pp. 722-728). janeczek, k., jakubowska, m., koziol, g., mlozniak, a. & arazna, a. (2012). investigation of ultra-high-frequency antennas printed with polymer pastes on flexible substrates. iet microwaves antennas propagation, 6(5), 549-554. khirotdin, r. k., cheng, t. s. & mokhtar, k. a. (2016). printing of conductive ink tracks on textiles using silkscreen printing. arpn journal of engineering and applied sciences, 11(10), 6619–6624. merilampi, s., laine-ma, t. & ruuskanen, p. (2009). the characterization of electrically conductive silver ink patterns on flexible substrates. microelectronics reliability, 49(7), 782-790. mitsui, r., sato, j., takahashi, s. & nakajima, s. (2015). electrical reliability of a filmtype connection during bending. electron. 4(4), 827–846. rozali, n. s., sobri, n. h., suhaimi, m. a., azmi, m. z., & akop, m. z. (2018). effects of carbon black to electrical properties on stretchable printed circuit. in 1st colloquium paper: advanced materials and mechanical engineering research (cammer'18), penerbit universiti, universiti teknikal malaysia melaka, 1, 5859. journal of mechanical engineering and technology issn 2180-1053 vol. 11 no.1 july – december 2019 7 suhaimi, m. a., azmi, m. z., rozali, n. s., sobri, n. h., & akop, m. z. (2018). effect of line thickness cross-sectional geometry to stretchable printed circuit. in 1st colloquium paper: advanced materials and mechanical engineering research (cammer'18), penerbit universiti, universiti teknikal malaysia melaka, 1, 4445. journal of mechanical engineering and technology *corresponding author. email: sivakumard@utem.edu.my issn 2180-1053 vol. 11 no. 1 july-december 2019 36 application of fuzzy vikor in automotive brake pad material n. m. ishak1, d. sivakumar 1*, m. r. mansor1, i.siva2 1centre for advanced research on energy, fakulti kejuruteraan mekanikal, universiti teknikal malaysia melaka, hang tuah jaya, 76100 durian tunggal, melaka, malaysia. 2centre for composite materials, kalasalingam academy of research and education, anand nagar, krishnankoil-626126, tamil nadu, india. abstract multi criteria decision making methods is one of the most common methods used to determine the most appropriate material. in the decision making process, there are dissimilarities to elicit, specify and analyse the information on alternatives, criteria and relative significance of the criteria. fuzzy set has been utilised in multi criteria decision making methods to optimise the method and created an extended approach to deal with uncertainty and increase the accuracy of decision making. as for many years’ asbestos was viewed as having an optimal performance as a brake pad. however, this material has been banned by the environmental protection agency. due to the increasing awareness on environmental impact and subsequently the need towards sustainability, selection of the appropriate material for a brake pad that complies with the environment and regulations is vital and natural fibre reinforced composite has potential to replace the asbestos in the automotive brake pad application. therefore, the objective of this study is to apply the fuzzy vikor to select the best natural fibre reinforced composite for the automotive brake pad to replace the asbestos. four alternatives of natural fibre reinforced composite with five criteria have been evaluated by three decision maker. the results of the fuzzy vikor shows that the date palm fibre is selected as the best material for the automotive brake pad. keywords: material selection; fuzzy vikor; mcdm; brake pad 1.0 introduction material selection is one of the crucial processes in engineering design to fulfil the requirement in product design. multi criteria decision making (mcdm) methods is one of the material selection process that has many different methods such as the elimination and et choice translating reality (electre) method, vlse kriterijumska optimizacija kompromisno rejense (vikor) method, technique for order preference by similarity to ideal solution (topsis) method, analytical hierarchy process (ahp) and preference journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 1 july-december 2019 37 ranking organisation method for enrichment of evaluations (promethee) and many more. however, there are dissimilarities to elicit, specify and analyse the information on alternatives, criteria and relative significance of the criteria (belton and stewart, 2002). therefore, fuzzy set has been utilised in mcdm methods to optimise the mcdm methods and created an extended approach to deal with uncertainty and increase the accuracy of decision making (asemi et al., 2014), especially in the material selection process. there are several studies examining material selection that extend the mcdm method with fuzzy sets. ishak et al., (2017) studied the selection of thermoplastic matrix for fibre metal laminate using fuzzy vikor and entropy. anojkumar et al., (2014) studied the pipe material selection in sugar industry using the fuzzy ahp. rathod and kanzaria, (2011) studied the material selection of solar domestic hot water system using fuzzy tosis. yang et al., (2017) studied the material selection for automotive products design using fuzzy topsis. xue et al., (2016) studied the material selection for the automotive instrument panel using fuzzy mabac. ishak et al., (2016) studied the material selection of natural fibre reinforced composites using fuzzy vikor for car front hood. brake is a device that stops motion. for many years asbestos was viewed as having an optimal performance as a brake pad. however, this material has been banned by the environmental protection agency (epa) as this material is very poisonousness and could affect human health and the environment (ramazzini, 2010). due to the increasing awareness on environmental impact and subsequently the need towards sustainability, selection of the appropriate material for a brake pad that complies with the environment and regulations is vital. nowadays, natural fibre reinforced composite have gained interest among researchers due to its potential in reducing weight, cost-effective, environmentally friendly, a renewable source, biodegradable and recyclable (tong et al., 2017). natural fibre reinforced composite has high possibility to substitute the asbestos. therefore, the objective of this study is to apply the fuzzy vikor to select the best natural fibre reinforced composite for the automotive brake pad to replace the asbestos. 2.0 methodology 2.1 fuzzy vikor vikor is the serbian abbreviation which stands for “vlsekriterijumska optimizacija i kompromisno resenje” which means multi criteria optimization and compromise solution method. integration of vikor; one of the mcdm methods with fuzzy set produced fuzzy vikor method. to utilise the fuzzy set, linguistic variables constitute evaluation were used to calculate the importance of criteria and the ratings of alternatives with various respects to various criteria. table 1 shows the linguistic terms and their corresponding fuzzy numbers. linguistic terms will be used by the decision maker to evaluate the respective alternatives and criteria. trapezoidal fuzzy numbers were implemented since this function can perform calculation easily journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 1 july-december 2019 38 table 1: linguistic terms and corresponding fuzzy numbers for each criterion and alternatives linguistic variable for criteria linguistic variable for alternatives fuzzy number unimportant (ui) very poor (vp) (0.0, 0.0, 0.1, 0.2) low importance (li) poor (p) (0.1, 0.2, 0.2, 0.3) slightly important (si) medium poor (mp) (0.2, 0.3, 0.4, 0.5) moderate importance (mi) fair (f) (0.4, 0.5, 0.5, 0.6) important (i) medium good (mg) (0.5, 0.6, 0.7, 0.8) very important (vi) good (g) (0.7, 0.8, 0.8, 0.9) extremely important (ei) very good (vg) (0.8, 0.9, 1.0, 1.0) the membership function is determined ( )                            − −   − − = otherwisex nnx nn xn nnx nnx nn nx x a 0 4 , 3 43 4 3 , 2 ,1 2 , 1 , 12 1 ~ (1) the aggregated fuzzy weight j w of each criterion  s j s j s j s j s j wwwwwj 4321 ,,;= (2) where,  s jk s j ww 11 min= = s jk s j w k w 22 1 = s jk s j w k w 33 1  s jk s j ww 44 max= =k decision makers the aggregated fuzzy ratings ij x of alternatives with respect to each criterion  4321 ,,; ijijijijij xxxxx = (3) where,  ,min 11 ijkij xx = journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 1 july-december 2019 39 = 22 1 ijkij x k x = 33 1 ijkijk x k x   44 max ijkijk xx = =k decision makers defuzzify the fuzzy decision matrix and fuzzy weight of each criterion into crisp value ( ) ( ) ( )  = dxx xdxx xdefuzz ij   . (4) the best * j f and the worst − j f value of all criterion ratings   iji ff max * = (5)   iji ff min= − (6) the utility (si), regret (ri) and vikor index (qi) ( ) ( )= −− − = n j ii iji s j i ff ffw s 1 * * (7) ( ) ( )         − − = − ii iji s j ii ff ffw r * * max (8) ( ) ( )( ) * * * * 1 rr rrv ss ssv q ii i − −− +         − − = −− (9)               − − ++         − −         − − ++         − − = 2 1 3 2 4 3 2 1 3 2 4 3 12 1 12 1 34 4 12 1 .. ij ij ij ij ij ij ij ij ij ij ij ij x x x x x x ijij ij ijij ij x x x x x x ijij ij ijij ij dx xx xx dxdx xx xx xdx xx xx xdxxdx xx xx ( ) ( ) 4321 2 12 2 344321 3 1 3 1 ijijijij ijijijijijijijij xxxx xxxxxxxx ++−− −−−++− = journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 1 july-december 2019 40 compromise solution if and only satisfy two conditions 1 and 2 are satisfied. the set of compromise solutions are composed of: condition 1: acceptable advantage: ( )( ) ( )( ) ( )1/112 −− maqaq , where ( )2a is the second position in the alternatives ranked by q . condition 2: acceptable stability in decision making: alternative ( )1a must also be the best ranked by s or/and r. when one of the conditions is not satisfied, a set of compromise solution is selected. the set of compromise solutions are composed of: (1) alternatives ( )1a and ( )2a if only condition 2 is not satisfied (or) (2) alternatives ( )1a , ( )2a ,…, ( )ma if condition 1 is not satisfied. ( )ma is calculated using the relation ( )( ) ( )( )1aqaq m − < ( )1/1 −m for maximum m . 3.0 case study four (4) alternatives of natural fibre reinforced composite have been designated for the automotive brake pad to replace the asbestos which are palm kernel fibre (m1), date palm fibre (m2), sisal fibre (m3) and bamboo fibre (m4). five (5) criteria; coefficient of friction (c1), thermal conductivity (c2), hardness (c3), tensile strength (c4) and wear (c5) will be evaluated by three (3) decision makers (dm). table 2 shows the mechanical properties of the candidate materials. table 2: mechanical properties of the candidate materials through linguistic terms, decision makers determine the importance of each criterion and then analyse and evaluate each alternative with respect to evaluation criteria. table 3 and table 4 shows the linguistic variables and the fuzzy value assessed by the decision makers. coefficient of friction (μ) thermal conductivity (w.𝐦−𝟏.𝐊−𝟏) hardness (h) tensile strength (mpa) wear (%) palm kernel fibre 0.33 0.70 30 28.7 4.0 date palm fibre 0.32 0.74 54.2 37.2 2.0 sisal fibre 0.43 0.25 52 36.6 1.4 bamboo fibre 0.31 0.20 22.3 26.4 3.0 journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 1 july-december 2019 41 table 3: importance weight of criteria assessed by decision makers (linguistic variable) c1 c2 c3 c4 c5 dm 1 vi e i i vi dm 2 vi vi vi i e dm 3 vi e i i e table 4: importance weight of criteria assessed by decision makers (fuzzy value) c1 c2 c3 c4 c5 dm 1 (0.7, 0.8, 0.8, 0.9) (0.8, 0.9, 1.0, 1.0) (0.5, 0.6, 0.7, 0.8) (0.5, 0.6, 0.7, 0.8) (0.7, 0.8, 0.8, 0.9) dm 2 (0.7, 0.8, 0.8, 0.9) (0.7, 0.8, 0.8, 0.9) (0.7, 0.8, 0.8, 0.9) (0.5, 0.6, 0.7, 0.8) (0.8, 0.9, 1.0, 1.0) dm 3 (0.7, 0.8, 0.8, 0.9) (0.8, 0.9, 1.0, 1.0) (0.5, 0.6, 0.7, 0.8) (0.5, 0.6, 0.7, 0.8) (0.8, 0.9, 1.0, 1.0) based on equation 2, table 5 shows the aggregated fuzzy value of natural fibre criterion weights assessments. table 5: the aggregated fuzzy value of natural fibre criterion weights assessments c1 c2 c3 c4 c5 w vi e i i vi table 6 and 7 shows the evaluation of the decision makers on the importance of material with respect to criteria of the automotive brake pad. table 6: importance of material with respect to criteria (linguistic variable) c1 c2 c3 c4 c5 dm 1 m1 mg mg f mp mp m2 mg mg g f mg m3 g f g f g m4 mg f mp mp f dm 2 m1 f g p f f m2 f g f f g m3 mg p f f g m4 f p p f f dm 3 m1 g g g mg mg m2 g g g mg mg m3 g mg g mg g m4 g mg g mg mg journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 1 july-december 2019 42 table 7: importance of material with respect to criteria (fuzzy value) c1 c2 c3 c4 c5 dm 1 m1 (0.5, 0.6, 0.7, 0.8) (0.5, 0.6, 0.7, 0.8) (0.4, 0.5, 0.5, 0.6) (0.2, 0.3, 0.4, 0.5) (0.2, 0.3, 0.4, 0.5) m2 (0.5, 0.6, 0.7, 0.8) (0.5, 0.6, 0.7, 0.8) (0.7, 0.8, 0.8, 0.9) (0.4, 0.5, 0.5, 0.6) (0.5, 0.6, 0.7, 0.8) m3 (0.7, 0.8, 0.8, 0.9) (0.4, 0.5, 0.5, 0.6) (0.7, 0.8, 0.8, 0.9) (0.4, 0.5, 0.5, 0.6) (0.7, 0.8, 0.8, 0.9) m4 (0.5, 0.6, 0.7, 0.8) (0.4, 0.5, 0.5, 0.6) (0.2, 0.3, 0.4, 0.5) (0.2, 0.3, 0.4, 0.5) (0.4, 0.5, 0.5, 0.6) dm 2 m1 (0.4, 0.5, 0.5, 0.6) (0.7, 0.8, 0.8, 0.9) (0.1, 0.2, 0.2, 0.3) (0.4, 0.5, 0.5, 0.6) (0.4, 0.5, 0.5, 0.6) m2 (0.4, 0.5, 0.5, 0.6) (0.7, 0.8, 0.8, 0.9) (0.4, 0.5, 0.5, 0.6) (0.4, 0.5, 0.5, 0.6) (0.7, 0.8, 0.8, 0.9) m3 (0.5, 0.6, 0.7, 0.8) (0.1, 0.2, 0.2, 0.3) (0.4, 0.5, 0.5, 0.6) (0.4, 0.5, 0.5, 0.6) (0.7, 0.8, 0.8, 0.9) m4 (0.4, 0.5, 0.5, 0.6) (0.1, 0.2, 0.2, 0.3) (0.1, 0.2, 0.2, 0.3) (0.4, 0.5, 0.5, 0.6) (0.4, 0.5, 0.5, 0.6) dm 3 m1 (0.7, 0.8, 0.8, 0.9) (0.7, 0.8, 0.8, 0.9) (0.7, 0.8, 0.8, 0.9) (0.5, 0.6, 0.7, 0.8) (0.5, 0.6, 0.7, 0.8) m2 (0.7, 0.8, 0.8, 0.9) (0.7, 0.8, 0.8, 0.9) (0.7, 0.8, 0.8, 0.9) (0.5, 0.6, 0.7, 0.8) (0.5, 0.6, 0.7, 0.8) m3 (0.7, 0.8, 0.8, 0.9) (0.5, 0.6, 0.7, 0.8) (0.7, 0.8, 0.8, 0.9) (0.5, 0.6, 0.7, 0.8) (0.7, 0.8, 0.8, 0.9) m4 (0.7, 0.8, 0.8, 0.9) (0.5, 0.6, 0.7, 0.8) (0.7, 0.8, 0.8, 0.9) (0.5, 0.6, 0.7, 0.8) (0.5, 0.6, 0.7, 0.8) the aggregated fuzzy value for the importance of material with respect to criteria assessments was calculated using equation 3. table 8: the aggregated fuzzy value of the importance of material with respect to criteria assessments c1 c2 c3 c4 c5 m1 (0.4, 0.63, 0.67, 0.9) (0.5, 0.73, 0.77, 0.9) (0.1, 0.50, 0.50, 0.9) (0.2, 0.47, 0.53, 0.8) (0.2, 0.47, 0.53, 0.8) m2 (0.4, 0.63, 0.67, 0.9) (0.4, 0.73, 0.77, 0.9) (0.4, 0.70, 0.70, 0.9) (0.4, 0.53, 0.57, 0.8) (0.5, 0.67, 0.73, 0.9) m3 (0.4, 0.63, 0.73, 0.9) (0.1, 0.43, 0.47, 0.8) (0.4, 0.70, 0.70, 0.9) (0.4, 0.53, 0.57, 0.8) (0.7, 0.80, 0.80, 0.9) m4 (0.4, 0.63, 0.67, 0.9) (0.1, 0.43, 0.47, 0.8) (0.1, 0.43, 0.47, 0.9) (0.2, 0.43, 0.53, 0.8) (0.4, 0.53, 0.57, 0.8) journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 1 july-december 2019 43 the aggregated fuzzy value for the weight and importance of material with respect to criteria assessments were then defuzzified to derive their crisp value using equation 4 shown in table 9. table 9: crisp value for weight and importance of material ratings c1 c2 c3 c4 c5 w 0.08 0.87 0.70 0.65 0.87 m1 0.65 0.72 0.50 0.50 0.50 m2 0.65 0.69 0.67 0.58 0.70 m3 0.72 0.45 0.67 0.58 0.80 m4 0.65 0.45 0.48 0.50 0.58 then, the best value * j f and worst value − j f of crisp material values are identified and they are shown in table 10. c1 c2 c3 c4 c5 𝑓∗ 0.72 0.72 0.67 0.58 0.80 𝑓− 0.65 0.45 0.48 0.50 0.50 utility index (si) and regret index (ri) were then defined using equation 7 and equation 8. the comprehensive utility value or the vikor value (𝑄𝑖) was calculated using equation 9. table 10 shows the utility, regret measure and vikor index value. the value of 𝑣 is taken as 0.5 to avoid bias (mandal et al., 2015). table 10: utility, regret measure and vikor index value (si) (ri) (𝑄𝑖) m1 2.95 0.87 0.87 m2 0.87 0.80 0.06 m3 1.20 0.87 0.50 m4 3.65 0.87 1.00 rank the preferences in an ascending order to determine the best alternatives as per the vikor method; the smallest alternative value was determined to be the best solution. table 11 shows the ranking of the material. table 11: ranking of the natural fibre reinforced composite 1 2 3 4 (si) m2 m3 m1 m4 (ri) m2 m1, m3 & m4 (𝑄𝑖) m2 m3 m1 m4 both conditions are satisfied in this context; therefore, the material with least vikor index which is m2 which is date palm fibre is selected as the best material for the automotive brake pad. journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 1 july-december 2019 44 4.0 conclusion based on the result of the fuzzy vikor analyses, the ascending rank suggested that m2 has the best criteria among the other four candidate materials. m2 (date palm fibre) has been selected as the best natural fibre by satisfying both conditions 1 and condition 2 with validation using least vikor index, where the m2 has the lowest vikor index (𝑄𝑖) value which is 0.06. m3 (sisal fibre) was in the second ranking with 0.50 scores, followed by m1 (kernel palm) with 0.87 scores and m4 (bamboo fibre) is the last choice of natural fibre in the automotive brake pad to replace the asbestos with the 1.00 scores. 5.0 references anojkumar, l., ilangkumaran, m., and sasirekha, v. (2014). comparative analysis of mcdm methods for pipe material selection in sugar industry. expert systems with applications, 41(6), 2964–2980. asemi, a., sapiyan, m., asemi, a., and haji, r.b. (2014). fuzzy multi criteria decision making applications : a review study. 344–351. belton, v., and stewart, t. (2002). multiple criteria decision analysis: and integrated approach., boston: kluwer academic publishers. ishak, n.m., malingam, s.d., and mansor, m.r. (2016). selection of natural fibre reinforced composites using fuzzy vikor for car front hood. international journal of materials and product technology, 53(3/4), 267–285. ishak, n.m., sivakumar, d., and mansor, m.r. (2017). thermoplastic matrix selection for fibre metal laminate using fuzzy vikor and entropy measure for objective weighting. journal of engineering science and technology, 12(10), 2792–2804. mandal, s., singh, k., behera, r.k., sahu, s.k., raj, n., and maiti, j. (2015). human error identification and risk prioritization in overhead crane operations using hta, sherpa and fuzzy vikor method. expert systems with applications, 42(20), 7195–7206. ramazzini, c. (2010). asbestos is still with us: repeat call for a universal ban. archives of environmental & occupational health, 65(3), 121–126. rathod, m.k., and kanzaria, h. v. (2011). a methodological concept for phase change material selection based on multiple criteria decision analysis with and without fuzzy environment. materials & design, 32(6), 3578–3585. tong, f.s., chin, s.c., doh, s.i., and gimbun, j. (2017). natural fiber composites as potential external strengthening material – a review. indian journal of science and technology, 10(2). journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 1 july-december 2019 45 xue, y.x., you, j.x., lai, x.d., and liu, h.c. (2016). an interval-valued intuitionistic fuzzy mabac approach for material selection with incomplete weight information. applied soft computing, 38, 703–713. yang, s.s., nasr, n., ong, s.k., and nee, a.y.c. (2017). designing automotive products for remanufacturing from material selection perspective. journal of cleaner production, 153, 570–579. issn: 2180-1053 vol. 7 no. 1 january june 2015 effect of atmospheric temperature on the performance of a petrol car 23 effect of atmospheric temperature on the performance of a petrol car r.k.pal1* 1department of mechanical engineering, panjab university swami sarvanand giri regional centre, hoshiarpur, india abstract the present work is to see the effect of atmospheric temperature on performance parameters like mileage, brake specific fuel consumption and thermal efficiency of a vehicle. experiments were conducted on a car and parameters like mileage, thermal efficiency and specific fuel consumption were computed at various values of atmospheric temperature and speed throughout the year. the maximum mileage and thermal efficiency of the car was in the month of march. the minimum specific fuel consumption was for the month of march as compared to other months of the year. the maximum mileage and thermal efficiency obtained were 22.6 km/litre and 29.48% respectively at speed 1550 rpm, torque 38 nm and average atmospheric temperature of 290 k. the minimum specific fuel consumption obtained was 0.275 kg/kwh at 1550 rpm speed, 38 nm torque and average atmospheric temperature of 290 k. the performance of the car was best in the month of march. keywords: automobile, specific fuel consumption, thermal efficiency, atmospheric temperature, torque 1.0 introduction the demand of energy is increasing very fast, but the main source of energy is still fossil fuels which has very limited reserves. around half of the world energy consumption is fulfilled by fossil fuel oil. transport sector consumes a major share of the world energy utilization. around 60% of world oil consumption is in transport sector, within this sector 80% of the total oil consumption is in road transport alone (silitonga et al. 2012). this sector is growing very fast due to steady increase in motorization and urbanization in developing countries. increase in energy consumption for non-road transport modes was 13% between 1990 and 2005 while that in road transport energy consumption was 41% (daniel, manfred & felix, 2010). estimates tell that the energy use * corresponding author email: ravinder_75@yahoo.com issn: 2180-1053 vol. 7 no. 1 january june 2015 journal of mechanical engineering and technology 24 for road vehicle will increase by 1.4% annually up to 2030 (silitonga et al. 2012 and takao, 2008). transportation sector produces 25% of total world carbon dioxide (co2) emissions; in this sector road transport produces 10% of greenhouse gases (ghg) emissions globally (silitonga et al. 2012). these greenhouse gases cause global warming and other health related problems. estimates tell that the co2 emissions for road vehicle will increase by 1.3% annually up to 2030 (silitonga et al. 2012 and takao, 2008). the annual fossil fuel consumption in transport sector can be decreased by improving the fuel efficiency of the transport vehicles. the fuel efficiency can be increased by increasing the engine efficiency which in turn will lower the co2 and other pollutants emission. the fuel efficiency of vehicles can be enhanced by utilizing existing improved engine & vehicle technologies (bezdek and wendling 2005 and weiss et al. 2000). the methods to enhance engine efficiency are improving thermal efficiency, using continuously variable transmission and reducing aerodynamic drag & vehicle weight etc. all of these methods except thermal efficiency need redesigning of the car. therefore these methods cannot be implemented on existing cars. the thermal efficiency of a vehicle can be improved by means of better radiators which loses more or less heat to surroundings depending on weather conditions. air inlet temperature is one of the factors which influence radiator performance (amrutkar and patil, 2013). determining the effect of atmospheric temperature on thermal efficiency is main task for design of such radiators. in summers the heat transfer and cooling capacity decreases with the increase in air inlet temperature (oliet et al. 2007). in winters the heat transfer and cooling capacity increases with the decrease in atmospheric air temperature. so we can say that cooling capacity keeps on changing throughput the year. keeping this literature review in mind the present work is to see the effect of atmospheric temperature on performance of the vehicle in terms of parameters like the mileage, brake specific fuel consumption and thermal efficiency. the parameters like fuel consumption, distance covered, atmospheric temperature, weight of passengers, speed of car were noted down and from these values, parameters like mileage of vehicle, power developed by engine, thermal efficiency of vehicle and break specific fuel consumption etc. were computed. 2.0 material and methods the experiments were conducted on a petrol engine car throughout the year at various speeds. the atmospheric temperatures, speed of car, distance travelled, time, fuel consumption were noted down. the other parameters like temperature after compression, maximum temperature and exhaust temperature of the engine were computed issn: 2180-1053 vol. 7 no. 1 january june 2015 effect of atmospheric temperature on the performance of a petrol car 25 from the inlet temperature using the formulae and data available in the literature and given in the appendix. the index of compression is taken as 1.35 (pulkrabek, 1997). the speed of the engine (in rpm) was calculated from the velocity of the car using the formula given in the appendix. the torque was computed from the normal force which was calculated from the weight of the car and that of passengers using formulae given in the appendix. the road/tyre friction factor was taken as 0.9 (ghandour et al. 2010). the brake power was computed from the speed of engine and torque using the formula given in appendix. mass of fuel (in kg) was calculated from the volumetric fuel consumption (in litre). the thermal efficiency of the engine was calculated from the calculated values of brake power, mass of fuel and calorific value of fuel available in the literature using the formula given in appendix. brake specific fuel consumption was calculated from the brake power and mass of the fuel consumed using the formula given in the appendix. 3.0 results and discussions the average atmospheric temperature (figure 1) increased from the month of january to june and decreased afterwards up to december. the maximum average atmospheric temperature was for the month of june. the temperature after compression, maximum temperature and exhaust gas temperature (figure 2) increased from the month of january to june and then decreased up to the month of december. this is because the atmospheric temperature increased from the month of january to june and decreased afterwards. therefore the heat transfer from the engine in the form of unaccounted loss decreased up to june which in turn increased all these temperatures. the heat transfer from the engine increased afterwards which decreased all these temperatures. 3 decreased up to june which in turn increased all these temperatures. the heat transfer from the engine increased afterwards which decreased all these temperatures. figure 1. monthwise average atmospheric temperature variation figure 2. monthwise variation of various temperature the thermal efficiency (figure 3) and mileage (figure 4) of the car increased from the month of january to march due to increase in the maximum engine temperature due to which thermal efficiency of the car increased which in turn increased the mileage. the thermal efficiency and mileage decreased from march to june due to more increase in figure 1. monthwise average atmospheric temperature variation issn: 2180-1053 vol. 7 no. 1 january june 2015 journal of mechanical engineering and technology 26 3 decreased up to june which in turn increased all these temperatures. the heat transfer from the engine increased afterwards which decreased all these temperatures. figure 1. monthwise average atmospheric temperature variation figure 2. monthwise variation of various temperature the thermal efficiency (figure 3) and mileage (figure 4) of the car increased from the month of january to march due to increase in the maximum engine temperature due to which thermal efficiency of the car increased which in turn increased the mileage. the thermal efficiency and mileage decreased from march to june due to more increase in figure 2. monthwise variation of various temperature the thermal efficiency (figure 3) and mileage (figure 4) of the car increased from the month of january to march due to increase in the maximum engine temperature due to which thermal efficiency of the car increased which in turn increased the mileage. the thermal efficiency and mileage decreased from march to june due to more increase in exhaust temperature although maximum temperature also increased. after that the thermal efficiency and mileage increased up to october due to decrease in lower temperature although higher temperature also decreased. these two parameters remained constant from july to august due to almost same amount of fall in upper and lower temperature. these two parameters decreased from october to december due to more decrease in higher temperature although lower temperature also decreased. 4 exhaust temperature although maximum temperature also increased. after that the thermal efficiency and mileage increased up to october due to decrease in lower temperature although higher temperature also decreased. these two parameters remained constant from july to august due to almost same amount of fall in upper and lower temperature. these two parameters decreased from october to december due to more decrease in higher temperature although lower temperature also decreased. figure 3. monthwise car thermal efficiency at different speeds figure 4. monthwise car mileage at different speeds figure 3. monthwise car thermal efficiency at different speeds issn: 2180-1053 vol. 7 no. 1 january june 2015 effect of atmospheric temperature on the performance of a petrol car 27 4 exhaust temperature although maximum temperature also increased. after that the thermal efficiency and mileage increased up to october due to decrease in lower temperature although higher temperature also decreased. these two parameters remained constant from july to august due to almost same amount of fall in upper and lower temperature. these two parameters decreased from october to december due to more decrease in higher temperature although lower temperature also decreased. figure 3. monthwise car thermal efficiency at different speeds figure 4. monthwise car mileage at different speeds figure 4. monthwise car mileage at different speeds the bsfc increased due to decrease in thermal efficiency from march to june. after that the bsfc decreased from june to october due increase in thermal efficiency. then the bsfc increased from october to december due to decrease in thermal efficiency of the car engine. 5 the bsfc increased due to decrease in thermal efficiency from march to june. after that the bsfc decreased from june to october due increase in thermal efficiency. then the bsfc increased from october to december due to decrease in thermal efficiency of the car engine. figure 5. monthwise car brake specific fuel consumption at different speeds 4.0 conclusions the maximum thermal efficiency and mileage of the car was for the month of march. the maximum value of the thermal efficiency and mileage of the car were 29.48% and 22.6 km/litre respectively at speed 1550 rpm, torque 38 nm and average atmospheric temperature of 290 k. the minimum brake specific fuel consumption of the engine at various values of speed and at various values of torque was for the month of march. the minimum value of bsfc was 0.275 kg/kwh at 1550 rpm speed, 38 nm torque and average atmospheric temperature of 290 k. the performance of the car was best in the month of march. nomenclature various temperatures inlet temperature = t1 temperature after compression = t2 maximum temperature = t3 exhaust temperature = t4 compression ratio = r index of compression = γ figure 5. monthwise car brake specific fuel consumption at different speeds 4.0 conclusions the maximum thermal efficiency and mileage of the car was for the month of march. the maximum value of the thermal efficiency and mileage of the car were 29.48% and 22.6 km/litre respectively at speed 1550 rpm, torque 38 nm and average atmospheric temperature of 290 k. the minimum brake specific fuel consumption of the engine at various values of speed and at various values of torque was for the month of march. the minimum value of bsfc was 0.275 kg/kwh at 1550 issn: 2180-1053 vol. 7 no. 1 january june 2015 journal of mechanical engineering and technology 28 rpm speed, 38 nm torque and average atmospheric temperature of 290 k. the performance of the car was best in the month of march. nomenclature 5 the bsfc increased due to decrease in thermal efficiency from march to june. after that the bsfc decreased from june to october due increase in thermal efficiency. then the bsfc increased from october to december due to decrease in thermal efficiency of the car engine. figure 5. monthwise car brake specific fuel consumption at different speeds 4.0 conclusions the maximum thermal efficiency and mileage of the car was for the month of march. the maximum value of the thermal efficiency and mileage of the car were 29.48% and 22.6 km/litre respectively at speed 1550 rpm, torque 38 nm and average atmospheric temperature of 290 k. the minimum brake specific fuel consumption of the engine at various values of speed and at various values of torque was for the month of march. the minimum value of bsfc was 0.275 kg/kwh at 1550 rpm speed, 38 nm torque and average atmospheric temperature of 290 k. the performance of the car was best in the month of march. nomenclature various temperatures inlet temperature = t1 temperature after compression = t2 maximum temperature = t3 exhaust temperature = t4 compression ratio = r index of compression = γ 6 pressure ratio = t2 = t1 t3 = t2 t4 = t3* torque (t) weight on each tyre = w radius of tyre = r road/tire friction factor = µ normal force on tyre = fn friction force = f w = total weight/4 fn = w f = µ * fn t = f * r volume of fuel = vf distance travelled by the car = d mileage = m m = speed of engine (n) gear ratio of transmission = grt gear ratio of differential = grd circumference of the tyre = ct velocity of car = v n = brake power (bp) bp = thermal efficiency (η) mass of fuel = mf calorific value of fuel = cv η = brake specific fuel consumption (bsfc) bsfc = issn: 2180-1053 vol. 7 no. 1 january june 2015 effect of atmospheric temperature on the performance of a petrol car 29 6 pressure ratio = t2 = t1 t3 = t2 t4 = t3* torque (t) weight on each tyre = w radius of tyre = r road/tire friction factor = µ normal force on tyre = fn friction force = f w = total weight/4 fn = w f = µ * fn t = f * r volume of fuel = vf distance travelled by the car = d mileage = m m = speed of engine (n) gear ratio of transmission = grt gear ratio of differential = grd circumference of the tyre = ct velocity of car = v n = brake power (bp) bp = thermal efficiency (η) mass of fuel = mf calorific value of fuel = cv η = brake specific fuel consumption (bsfc) bsfc = references bezdek, r.h. & wendling, r.m. (2005). potential long-term impacts of changes in us vehicle fuel efficiency standards. energy policy, 33(3), 407. daniel, b., manfred, b. & felix, c. (2010). beyond the fossil city: towards low carbon transport and green growth. 5th regional environmentally sustainable transport (est) forum in asia, 23-25 august 2010, bangkok, thailand. ghandour raymond, victorino alessandro, doumiati moustapha and charara ali, (2010). tire/road friction coefficient estimation applied to road safety. 18th mediterranean conference on control & automation congress palace hotel, marrakech, morocco june 23-25, 2010, 1488. pulkrabek, willard w. (1997). engineering fundamentals of the internal combustion. prentice hall, upper saddle river, new jersey. silitonga, a.s., atabania, a.e. & mahlia, t.m.i. (2012). review on fuel economy standard and label for vehicle in selected asean countries. renewable and sustainable energy, 16 (2012), 1683-1695. takao onoda, (2008). review of international policies for vehicle fuel efficiency; international energy agency (iea). weiss, m. a., heywood, j. b., drake, e. m., schafer, a. & auyeung, f. f. (2000). on the road in 2020: a life-cycle analysis of new automobile technologies, mit energy laboratory report 00-003. amrutkar, pawan s., patil, sangram r. (2013). automotive radiator performance – review, international journal of engineering and advanced technology (ijeat), 2(3), 563-565. oliet, c., oliva, a., castro, j., pe´rez-segarra,c.d. (2007). parametric studies on automotive radiators, applied thermal engineering, 27. journal of mechanical engineering and technology corresponding author. email: patrick.amiolemhen@uniben.edu issn 2180-1053 vol. 11 no. 1 july-december 2019 61 production of a connecting rod of an 8hp diesel engine by reverse engineering technique using lost wax casting method p. e. amiolemhen1*, f. c. nwosa1 1department of production engineering, faculty of engineering, university of benin, pmb 1154, benin city, nigeria. abstract the aim of this work is to produce by reverse engineering method, the connecting rod of a single-cylinder, four-stroke, 8-hp diesel engine using lost wax casting and machining processes. the connecting rod to be produced was sourced from an 8hp diesel engine generator. the dimensions of the connecting rod were taken and recorded and a hand sketch of the connecting rod was made. an engineering drawing was produced from the connecting rod sketch that was made. a pattern made of wax was produced from a pattern-mould box made of ceramics. the mould that was used to cast the connecting rod was made of green sand. a replica the an 8hp diesel engine connecting rod has been produced by lot wax casting using reverse engineering method. the values obtained for engine speeds and engine temperatures for the control engine and the engine with the manufactured connecting rod were very close under the two experimental conditions, which confirmed that the efficiency of the manufactured connecting rod is close to that of the original connecting rod. keywords: reversed engineering; lost wax casting method; connecting rod; connecting rod cap, internal combustion, engine speed 1.0 introduction the major problem bedeviling the nigerian manufacturing sector is the dearth of technology, which is as result of the advanced nations not willing to transfer technological know-how to developing nations like nigeria. this challenged has been taken up by some indigenous researchers on the need to locally develop the manufacturing sector by first developing the technological know-how through reverse engineering technique (or copy creativity). copy creativity of this kind requires a good knowledge of manufacturing processes, ingenuity, patience and a strong will to succeed (ibhadode, 2008). the internal combustion (ic) engine consists of many different components and parts that are assembled together to perform its intended function. the connecting rod is a major link inside of a combustion engine. it connects the piston to the crankshaft and is responsible for journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 1 july-december 2019 62 transferring power from the piston to the crankshaft and sending it to the transmission. the connecting rod is the most common cause of catastrophic engine failure. it is under an enormous amount of load pressure and is often the recipient of special care to ensure that it does not fail prematurely. a lot of contributions have been made to the definition of reverse engineering by different researchers manzoor et al, 2008; abdullahi and umar (2006); ibhadode, 2001; ebhojiaye and ibhadode, 2013). there are different types of materials and production methods used in the creation of connecting rods. the most common types of connecting rods materials are steel and aluminum and the most common type of manufacturing processes are casting. in lost wax casting, mould-maker creates an original pattern from wax (ravi, 2003). this research study is concerned with the production (by lost wax casting method) of connecting rod of an 8hp internal combustion (ic) engine by reverse engineering method. 2.0 methodology 2.1 design and materials selection this research study is centered on the development of local manufacturing sector by the use of locally sourced materials, labour, skills and tools to carry out a reverse engineering production of the connecting rod of an internal combustion engine. material used for the production of the connecting rod is cast iron, while bronze was used to produce the journal bearing that was force-fitted into the connecting rod end that the piston rod will go through with mild steel connecting rod bolts. 2.2 development of connecting rod pattern the sample of the connecting rod shown in figure 1 was sourced from an 8hp diesel. the wax pattern was formed to shape by ceramic mould constructed from the dimensions obtained from the engineering drawing of the sample-connecting rod as shown in figure 2. figure 1: sample-connecting rod sourced from the 8hp diesel engine http://en.wikipedia.org/wiki/wax journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 1 july-december 2019 63 figure 2: sketch of the sample-connecting rod 2.3 determination of pattern dimensions the following pattern allowances were considered during the development of the pattern size: linear shrinkage allowance; machining allowance and draft allowance. ibhadode (2001) provided the recommended values for a cast iron with the dimensions of the connecting rod as: linear shrinkage allowance = 0.01042 mm; machining allowance = 3 mm and draft allowance =1mm. hence, the dimensions of the connecting rod pattern and the connecting rod cap pattern were computed from the dimensions of the sample of connecting rod shown in figure as follows: 2.3.1 connecting rod 1. length of pattern : 340 + 0.01042 + 3 + 1 = 344.01mm 2. inner diameter of bearing shell: 70 + 0.01042 + 3 + 1 = 74.01mm 3. outer diameter of bearing shell: 91 + 0.01042 + 3 + 1 = 95.01mm 4. inner diameter of the short-arm: 40 + 0.01042 + 3 + 1 = 44.01mm 5. outer diameter of the short-arm: 55 + 0.01042 + 3 + 1 = 59.01mm 6. length of connecting rod shaft: 232 + 0.01042 + 3 + 1 = 236.01mm 7. breadth of the long-arm of the connecting rod shaft: 44 + 0.01042 + 3 + 1 = 48.01mm 8. breadth of the short-arm of the connecting rod shaft: 36 + 0.01042 + 3 + 1 = 40.01mm 9. width of the long-arm of the connecting rod shaft: 35 + 0.01042 + 3 + 1 = 39.01mm 10. width of the long-arm of the connecting rod shaft: 25 + 0.01042 + 3 + 1 = 29.01mm 11. width of the of the connecting rod bearing shell: 54 + 0.01042 + 3 + 1 = 58.01mm 12. width of the short-arm of the connecting rod : 48 + 0.01042 + 3 + 1 = 52.01mm 13. length of the thread that attaches the cap to the connecting rod: 64 + 0.01042 + 3 + 1 = 68.01mm. 14. diameter of the thread that attaches the cap to the connecting rod: 30 + 0.01042 + 3 + 1 = 34.01mm. journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 1 july-december 2019 64 2.3.2 connecting rod cap 1. length of connecting rod cap: 47 + 0.01042 + 3 + 1 = 51.01mm 2. breadth of connecting rod cap: 115 + 0.01042 + 3 + 1 = 119.01mm 3. width of connecting rod cap: 54 + 0.01042 + 3 + 1 = 58.01mm 2.3.3 sampled connecting rod parameters 1. mass of the sample-connecting rod: 2.981 kg 2. density of cast iron: 7.15 x 103 kg/m3 3. specific heat capacity of cast iron: 500 j/kg oc 4. volume of sample-connecting rod: 4.17 x 10-4 m3 2.4 pattern fabrication two mould boxes were fabricated for the connecting rod pattern and the connecting rod cap pattern, as shown in figures 3 and 4, respectively. figure 3: pattern mould box for the connecting rod journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 1 july-december 2019 65 figure. 4: pattern mould box for the connecting rod cap 2.5 design of the mould parameters mould parameters, such as diameter of riser; diameter of sprue; mould filling time and molten metal pouring velocity were computed as follows: 2.5.1 design of riser size according to ravi (2003), the feeder compensates solidification shrinkage of the hot spot region. hence, the volume of the feeder was computed as 5.36 x 10-5m3 from eqn. (1): vf = αvc ɳf−αvc (1) where: vc= volume of casting = 4.17 x 10 -4m3; 𝜂f = efficiency of feeder (riser) = 0.14 and α = volumetric solidification shrinkage of cast metal = 0.018 (for cast iron) the best shape of the riser for optimum run of casting is a cylindrical shape and the ratio of riser height to diameter usually varies from 1:1 to 3:2, ravi [8]. hence, the ratio of riser height to diameter used for the casting was eqn. (2): riser height, 𝐻 = 1.5𝐷 (2) journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 1 july-december 2019 66 the diameter and the height of the cylindrical riser were computed as 0.0357m and 0.054m, respectively from eqns. (2) and (3): d = √ 4 vf 1.5π 3 (3) 2.5.2 determination of mould filling time the mould filling time was computed as 7.39sec from the generalized empirical equation for filling time [8], stated as eqn. (5): τt = ko ( k flf 1000 ) [k s + ( k tt 20 )] (k w w) p (4) where: ko = overall coefficient = 1.0; kf = coefficients for fluidity = 1.0; ks = coefficients for size = 1.1; kt = coefficients for thickness = 1.4 (for wall thickness up to 10 mm); lf = fluidity length = 500 mm (for cast iron); kw = coefficients for weight = 1.0; w = weight of casting = 29.21 n; t = section thickness = 39.01 mm and p = 0.4 2.5.3 determination of pouring velocity the liquid metal pouring velocity was computed as 1.05m/s from eqn. (5): v = √2ghs (5) where: hs = mould metalostatic height (i.e. cope height + ½ ingate diameter) = 56mm and g = acceleration due to gravity = 9.81m/s2. 2.5.4 design of the sprue diameter the sprue exit area and size (usually selected to control the pouring rate), were determined as 1.75 x 10-4m2 and 47.20mm, respectively from eqns. (6) and (7), as given by ravi (2003): the formula for finding the sprue choke area ƒn, is given as: fn = m ρ τfμv (6) where: ρ = molten metal density = 7.15 x 103 kg/m3; v = liquid metal pouring velocity =1.05m/s; m = mass of casting in the mould (including risers, runners ingates and sprue well) = 2.981 kg and μ = discharge coefficient of the metal, usually 0 < μ=0.54 < 1 d = 2√ fn π (7) 2.6 production of patterns the two patterns were produced: one wax pattern for the connecting rod and the other for the connecting rod cap as shown in figure 5. journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 1 july-december 2019 67 figure 5: pictorial view of wax pattern produced 2.7 composition of the cast iron the chemical composition of the cast iron by percentage weight (wt. %) is: carbon – 2.50; silicon – 1.80; manganese – 0.80; sulfur – 0.15; phosphorus – 0.15 2.8 lost wax casting the connecting rod was produced by reverse engineering of the original connecting rod, which was acquired from 8hp diesel generator engine. a wax pattern was produced from the pattern mould that was made from ceramic material. the wax pattern was embedded into a green sand mould and heated to allow the wax to flow out of the mould thereby creating cavity of the shape of connecting rod. the same procedure was used to make the green sand mould of the connecting rod cap. lost wax (investment casting) process was used to produce the connecting rod and its cap as shown in figure 6, with pouring temperature in the range of 1430oc (kanno et al., 2006; escobar et al., 2015). the produced pieces were machined to the dimensions of the original connecting rod and its cap. figure 6: connecting rod and cap produced by lost wax casting before machining journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 1 july-december 2019 68 2.9 connecting rod bushing the connecting rod bushing which serves as a journal bearing inside the short arm of the connecting rod was produced by machining a cylindrical block into the required shape of a hollow ring. bronze was selected as the material for this item because the original connecting rod had its bushing made of mild steel coated with bronze. 2.10 inspection check of the produced connecting rod and cap the dimensions of the cast connecting rod (prior to machining) were checked to see if its shrinkage values (mm/mm) were within the recommended range of values by the following analysis: 1. full length of original connecting rod = 387 mm full length of designed pattern of connecting rod = 395.02 mm full length of cast connecting rod before machining = 392.21 mm contraction of connecting rod = 2.81 mm true contraction (mm/mm) = 0.007 mm 2. width of the original connecting rod cap = 54 mm width of designed pattern of connecting rod cap = 58.01 mm width of cast connecting rod cap before machining = 57.45 mm contraction of connecting rod cap= 0.56 mm true contraction (mm/mm) = 0.0096 mm. the computations of the true contractions of the full length and width of the connecting rod that was cast showed an approximate value of 0.01 mm each. this value is within the recommended shrinkage value of cast iron of this size, found in standard texts. this shows that the pattern that was produced is satisfactory. 2.11 cast connecting rod parameters 1. mass of the cast-connecting rod: 2.9892kg 2. density of cast iron: 7.16 x 103 kg/m3 3. specific heat capacity of cast iron: 500 j/kg oc 4. volume of sample-connecting rod: 4.18 x 10-4 m3 2.12 performance tests the connecting rod produced from this study (shown in figure 7), was subjected to performance tests by interchanging it with the original connecting rod inside the diesel generator engine. journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 1 july-december 2019 69 figure 7: connecting rod produced 2.12.1 test for engine temperature this test was carried out to measure the temperature rise of the diesel engine with respect to time using a digital read-out thermocouple. during this test the engine was run for duration of one hour under no-load and another one hour under a load of 3hp, and the engine temperature were recorded after an interval of 5 mins. first, engine temperature readings were taken when running the engine with the original connecting rod (i.e. control engine readings). secondly, engine temperature readings were taken when the original connecting rod was interchanged for the reverse engineered connecting rod. 2.12.2 engine speed test the engine speeds were measured with a tachometer under the specified experimental conditions above and the readings were recorded. during this test the engine was run for duration of one hour under no-load and another one hour under a load of 3hp, and engine speed values were recorded after an interval of 5 mins. 3.0 results and discussion 3.1 results 3.1.1 engine temperatures the readings were taken for the control engine (i.e. engine with original connecting rod) and engine with produced connecting rod temperatures with respect to time as shown in figure 8. journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 1 july-december 2019 70 figure 8: engine temperature against engine runs time 3.1.2 engine speeds the speed of the engine was measured using the tachometer. readings were taken for the control engine and engine with the manufactured connecting rod under the no-load condition and under a load of 3hp. this is shown in table 1. table 1: speeds of control engine operational condition control connecting rod produced connecting rod engine speed (rpm) engine speed (rpm) under no-load condition 850 793 under a load of 3hp 824 786 3.2 discussion figure 8 shows the plot of engine temperatures against run times, under the no-load condition and under a load of 3 hp for the control connecting rod and the cast connecting rod. the figure shows that there was a steady rise in temperature under the no-load condition and under a load of 3 hp for the first 15 mins of running the engine. after 20 mins of running the engine, the temperature dipped with 2oc under the no-load condition and 1oc under a load of 3hp for the control connecting rod as well as the cast connecting rod. the engine temperature began to rise again after 25 mins under both experimental conditions. however, both experimental conditions recorded the highest temperature difference at 45 mins, with the engine under no-load condition recording an 0 50 100 150 200 0 5 10 15 20 25 30 35 40 45 50 55 60 e n g in e t e m p e ra tu re s (o c ) engine run times (mins) temperature (oc) with load of 3hp (cast rod) temperature (oc) under no-load condition (control rod) temperature (oc) under no-load condition (cast rod) temperature (oc) with load of 3hp (control rod) journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 1 july-december 2019 71 increase of 47oc and the engine under a load of 3hp recording a temperature rise of 49oc within the same space of 5 mins. again, the result showed that after 1 hour of running the control engine, the engine under a load of 3hp recorded the highest temperature of 200oc while the engine under no-load condition recorded a temperature of 190oc. this figure showed a curve with sharp increase in temperature from 20 mins to 50 mins. the temperature curve became mildly slant between 50 mins and the 1 hour mark, indicating a closer temperature difference at that range of time. for the cast connecting rod the engine temperature did not dip under a no-load condition but dipped with 1oc under a load of 3hp at 20 mins. however, the engine under no-load condition recoded its lowest temperature rise of 2oc at 20 mins. again, under the no-load condition, there was 0oc temperature rise between 25 mins and 30 mins as the temperature was stable at 69oc. the engine under load of 3hp recorded the highest temperature of 201oc after i hour while the engine under no-load condition recorded a temperature of 198oc. this figure showed that the engine under load of 3hp had the steepest slope when compared with engine under no-load. table 1showed the values of the engine speeds under the two experimental conditions for the control engine. from the table, the engine speed reduced with 3.06% under 3hp load condition as against the value of 250 rpm when the control engine was under no-load condition. table 1 also shows the same decrease in value for the engine with the manufactured connecting rod. the speed of the engine reduced with 0.88% under 3hp load condition from the 793 rpm recorded under no-load condition. however, the reduction in speed was more for control engine than for engine with the manufactured piston. 4.0 conclusion the values obtained for engine speeds and engine temperatures for the control engine and the engine with the manufactured connecting rod were very close under the two experimental conditions, which confirmed that the efficiency of the manufactured connecting rod is close to that of the original connecting rod. journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 1 july-december 2019 72 5.0 references abdullahi, i., umar, a. a. (2006). manufacturing of 1hrc 230f transmission coupling using reverse engineering. nigerian journal of research and development. vol. 5. no. 1. pp. 36 39. degarmo, e.p., black, j. t. and kohser, r. a. (2003). materials and processes in manufacturing. 9th edition, wiley. ebhojiaye, r. s. and, ibhadode, a. o. a. (2013). production of a piston for a single cylinder four stroke 8hp diesel engine by reverse engineering technique, the journal of the nigerian institution of production engineers. 2013. vol. 15, 80 – 88. escobar, a., celentano, d., cruchaga, m. and schutz, b. (2015). on the effect of pouring temperature on spheroidal graphite cast iron solidification. metals. vol. pp. 628 – 647. ibhadode, a. o. a. (2001). progress report on development of 3-hp petrol engine, university research and publication committee, university of benin, benin city. ibhadode, a. o. a. (2004. progress report on development of 3-hp petrol engine.university research and publication committee, university of benin, benin city. kanno, t., kang, i., fukuda, y., mizuki, t. and kiguchi, s. (2006). effect of pouring temperature and composition on shrinkage cavity in spheroidal graphite cast iron. transaction of the american foundry society: annual metalcasting congress. 110. 114. pp. 525 – 534. manzoor, h. m., sambasiva, r. c. h. and prasad, k. e. (2008). reverse engineering: point cloud generation with cmm for part modeling and error analysis, arpnjournal of engineering and applied sciences. vol. 3. no.4.pp. 3740. ravi, b. (2003). casting design and analysis, indian institute of technology, bombay. issn: 2180-1053 vol. 7 no. 1 january june 2015 prototype development of wireless pneumatic gear shifter 1 prototype development of wireless pneumatic gear shifter m.a. abdullah1*, m.a mohd ajwad1 a.h. muhammad ali imran1 and f.r ramli 1centre for advanced research on energy, universiti teknikal malaysia melaka, hang tuah jaya, 76100 durian tunggal, melaka, malaysia abstract conventional design of transmission gear shifter is basically consists of mechanical linkage of gear lever attached on the side of motorcycle engine with pivot directly connected to the gear shifting mechanism. shifting up and down of gear positions are performed by moving foot or toe upward and downward on the lever respectively. for disable rider (rider requires special need), shifting gear for manual transmission type of motorcycle with or without clutch system is difficult. in this research, a pneumatic gear shifter is designed and fabricated. it consists of pneumatic circuit with actuator, tanks and air compressor. the movement of gear lever is performed by the pneumatic actuator. pneumatic tanks are installed to achieve optimum pressure. it also controlled by wireless system for convenient purpose and buttons installed at the handle. simple experiment is performed to measure the force for each gear position. keywords: pneumatic gear shifter, wireless gear shifter, gear changer, gear mechanism, gear leve 1.0 introduction the transmission of a light vehicle is determined by the number of force applied to the gearshift. most motorcycle gearshift assemblies in recent years have been fabricated with a foot pedal that is shifted upwardly and downwardly by the bottom and top surfaces of the toe or foot. the heel portion of the rider’s foot normally rests on the stationary stirrup of foot rest which bears most of the weight of the leg and foot of the rider, while the pedal mounted on the end of a foot rest at a location where the rider could merely depress or lift the toe portion of his foot with pivotal movement about his ankle joint to shift the gears of the motorcycle (herbert, 2005)(bosch, 1975). in transmission system, conventionally, it has a mechanical linkage that connects the gear lever * corresponding author email: azman@utem.edu.my issn: 2180-1053 vol. 7 no. 1 january june 2015 journal of mechanical engineering and technology 2 to the gear switching mechanism. the mechanism for gear changing of the transmission still remains the same that is the reliance towards the gear lever situated on the left leg of the rider (cengel et al. 2009). a normal healthy person would not have any problem to carry out this action but the situation is vice versa for the elderly and leg disable people. hence, the research hopes to give the data and a new light so that people that fall into the previously mentioned category could also enjoy the pleasure of riding this economical means of transport (steve, 2007). the transmission system size, weight and type are varied from one manufacturer to another. nevertheless, its basic principle on how the system works remain constant although it is produced by different manufacturers. for the simplest form of this system, it will only contain a centrifugal clutch attached to the crankshaft and then redirected to the sprocket via chain. as the engine speed increases, the clutch activates and propels the rear wheel (jaap, 2000),(molly and pautot, 1992). this is a perfect example for a single speed transmission system which is consider to be the most efficient system available nowadays (lin and costello,1983),(salonidis, 2001). a solenoid gear shifter or an electric solenoid shifter is an invention that is equipped onto motorcycle or car for the gear changing process (lee, 1995). this technology, mainly for motorcycle, is only used for clutchless shifting of the motorcycle by only pressing or pushing a button that is mounted on the handlebars of the selected motorcycle. it also includes with a solenoid mounting plate for ensuring that the solenoid is fasten securely to the motorcycle and a micro-switch which is used to be operable linked to the solenoid and the mounted motorcycles (kevin, 2007). solenoid gear shifting mechanism uses magnet to move upwardly and downwardly (gerald, 1996). this movement depends on the magnetic field that is produce by the magnet when power is supply through it. this paper is meant to provide a better understanding of motorcycle transmission system and how the system could be simply or improve in order to ensure that elderly and leg disable people could ride the vehicle. the component relation to the human safety is given extra attention in order to avoid unnecessary expenditure in the maintenance and repair works of the vehicle, this is due mechanical failure during the test of this technology could means fatality if no contingency plan present issn: 2180-1053 vol. 7 no. 1 january june 2015 prototype development of wireless pneumatic gear shifter 3 2.0 methodology the pneumatic gear changer requires a very precise precision and accuracy so that it will not fail during the real working condition. each of the specification used in this phase will be taken serious consideration and any flaw in the system during this phase will be enhance or otherwise change by a new part. theoretical calculation also will be taken into consideration as it will help to further understand how the system actually works and it flaw was detected, it can be solve in a fast pace. the pneumatic gear changer owns novelty which is the wireless system, requires two main components which is the transmitter which will relay any information that the user input and the receiver. the receiver is the most vital component for the wireless system as all the information transmitted by the transmitter via radio frequency will be translated by this component and it will signal the information towards the related parts for operation. 2.1 prototype design figure 1 shows the image for the final configuration of the circuit board for both the transmitter and the receiver. the schematic diagram was later then used as a guideline in order to produce the actual wireless communication system. a jig is a type of custom made tool used to control the location and motion of another tool. its primary purpose is to provide repeatability, accuracy and interchangeability in the manufacturing process. in addition it also used for conducting experiment that used the same material over and over without changing any part of the system. as a result it enabled the user to conduct a controlled experiment without any further delay for to production or creating a place to hold the same material or apparatus over and over again. since the jig will be used for the testing of a pneumatic actuator; it will be made from plywood with thickness of 15 mm. figure 2 (a) was used as the base guideline for designing of the jig. the design parameter is quite different with the actual product as there has been difficulty in obtaining the most suitable material during the production process. figure 2 (a) shows the cad file for the design that will be used for creating the most suitable for the pneumatic actuator experiment. the gimmick box is manipulating all the moving part of the research project which is the movement of the gear lever for engaging and disengaging gear so that it will attract and at the same time make the audience to get better understanding of what being explained to them. it also served as a storage compartment which will house the entire vital component for this project. figure 2 (b) shows the setup of the gimmick in one of exhibition that this project took part in. component such as gas tank, issn: 2180-1053 vol. 7 no. 1 january june 2015 journal of mechanical engineering and technology 4 circuit board and even wiring is placed inside the gimmick box in order to remove most of the nuisance from the audience vision. by having the gear lever stick outside of the box, explanation could be done more easier as both side of the box is pasted with picture depicting the actual positioning of the pneumatic gear changer on a light vehicle. 3 parameter is quite different with the actual product as there has been difficulty in obtaining the most suitable material during the production process. figure 2 (a) shows the cad file for the design that will be used for creating the most suitable for the pneumatic actuator experiment. the gimmick box is manipulating all the moving part of the research project which is the movement of the gear lever for engaging and disengaging gear so that it will attract and at the same time make the audience to get better understanding of what being explained to them. it also served as a storage compartment which will house the entire vital component for this project. figure 2 (b) shows the setup of the gimmick in one of exhibition that this project took part in. component such as gas tank, circuit board and even wiring is placed inside the gimmick box in order to remove most of the nuisance from the audience vision. by having the gear lever stick outside of the box, explanation could be done more easier as both side of the box is pasted with picture depicting the actual positioning of the pneumatic gear changer on a light vehicle. (a) (b) figure 1. transmitter (a) and receiver schematic diagrams (b) figure 1. transmitter (a) and receiver schematic diagrams (b) 4 (a) (b) figure 2. test jig (a) and prototype box (b) 3.0 results and discussions gear lever is the main or the most vital part for the transmission of a light vehicle. it helps the vehicle to translate the movement of up and down of the rider’s leg into kinetic energy which will then either engage or disengage a gear. an experiment was conducted in order to find out the exact value of pressure and force required to be exerted by any rider in order to change gear of the light vehicle itself. in addition, the data obtained also will help this research with giving an insight for how much pressure does an actuator need to be able carry out the same task as the rider does. table 1 shows issn: 2180-1053 vol. 7 no. 1 january june 2015 prototype development of wireless pneumatic gear shifter 5 4 (a) (b) figure 2. test jig (a) and prototype box (b) 3.0 results and discussions gear lever is the main or the most vital part for the transmission of a light vehicle. it helps the vehicle to translate the movement of up and down of the rider’s leg into kinetic energy which will then either engage or disengage a gear. an experiment was conducted in order to find out the exact value of pressure and force required to be exerted by any rider in order to change gear of the light vehicle itself. in addition, the data obtained also will help this research with giving an insight for how much pressure does an actuator need to be able carry out the same task as the rider does. table 1 shows figure 2. test jig (a) and prototype box (b) 3.0 results and discussions gear lever is the main or the most vital part for the transmission of a light vehicle. it helps the vehicle to translate the movement of up and down of the rider’s leg into kinetic energy which will then either engage or disengage a gear. an experiment was conducted in order to find out the exact value of pressure and force required to be exerted by any rider in order to change gear of the light vehicle itself. in addition, the data obtained also will help this research with giving an insight for how much pressure does an actuator need to be able carry out the same task as the rider does. table 1 shows the result acquired from test done in order to find out the amount of force required in order to engage and disengage the transmission of a light vehicle. the test was conducted by an analog scale as shown in figure 3. the result of the test was done been double checked with a spring balance in order to ensure the accuracy and the reliability of the data. in addition, each of the data was taken 20 times in order to eliminate human error during the data reading through the usage of average data. the data for each of the gear transmission have been translated into graphical view which can be seen below. figure 3 (a) shows line graph for the data of the first gear, figure 3 (b) represent the graphical translation of the tabulated data for the second gear. figure 3 (c) and (d) show the representation of the tabulated data for the third and the last gear of the light vehicle. the average value of mass needed for engaging and disengaging is around 6 kg and 4 kg respectively. do take note that, the experiment was conducted by using light vehicle which still using the default part obtain from the manufacturer. in addition, the data might issn: 2180-1053 vol. 7 no. 1 january june 2015 journal of mechanical engineering and technology 6 be varying slightly for different light vehicle as it might use different type of configuration in its transmission system. all the result obtained from this task was then tabulated into table 1 where the value will then be compare with the value of pneumatic actuator. the system used a gear lever that helps to translate the human input into mechanical movement which will then engaging or disengaging the light vehicle transmission depending on the human input. upward movement is for shifting while downward movement is for downshifting. the amount of force required to be exerted by the riders depend on the speed of his leg; experiment result as tabulated in table 1 shows the trend of force required for shifting on a light vehicle which is honda wave 110. the engagement and disengagement of gear n mean shifting from gear n-1 to gear n or from gear n+1 to gear n. from the table, it can be seen that based on all 20 data taken for each gear shifting, the amount of force required remain in the range of 45 n to 55 n. this value change significantly depending on how fast the user exerted the force onto the gear lever. the faster the movement of the rider leg, the less amount of force required to engage and disengage the transmission system. yet, different light vehicle uses different kind of tools for the construction of the transmission system; the stiffness of the transmission spring inside the system will be different for different vehicle. 5 the result acquired from test done in order to find out the amount of force required in order to engage and disengage the transmission of a light vehicle. the test was conducted by an analog scale as shown in figure 3. the result of the test was done been double checked with a spring balance in order to ensure the accuracy and the reliability of the data. in addition, each of the data was taken 20 times in order to eliminate human error during the data reading through the usage of average data. the data for each of the gear transmission have been translated into graphical view which can be seen below. figure 3 (a) shows line graph for the data of the first gear, figure 3 (b) represent the graphical translation of the tabulated data for the second gear. figure 3 (c) and (d) show the representation of the tabulated data for the third and the last gear of the light vehicle. the average value of mass needed for engaging and disengaging is around 6 kg and 4 kg respectively. do take note that, the experiment was conducted by using light vehicle which still using the default part obtain from the manufacturer. in addition, the data might be varying slightly for different light vehicle as it might use different type of configuration in its transmission system. all the result obtained from this task was then tabulated into table 1 where the value will then be compare with the value of pneumatic actuator. the system used a gear lever that helps to translate the human input into mechanical movement which will then engaging or disengaging the light vehicle transmission depending on the human input. upward movement is for shifting while downward movement is for downshifting. the amount of force required to be exerted by the riders depend on the speed of his leg; experiment result as tabulated in table 1 shows the trend of force required for shifting on a light vehicle which is honda wave 110. the engagement and disengagement of gear n mean shifting from gear n-1 to gear n or from gear n+1 to gear n. from the table, it can be seen that based on all 20 data taken for each gear shifting, the amount of force required remain in the range of 45 n to 55 n. this value change significantly depending on how fast the user exerted the force onto the gear lever. the faster the movement of the rider leg, the less amount of force required to engage and disengage the transmission system. yet, different light vehicle uses different kind of tools for the construction of the transmission system; the stiffness of the transmission spring inside the system will be different for different vehicle. (a) issn: 2180-1053 vol. 7 no. 1 january june 2015 prototype development of wireless pneumatic gear shifter 7 6 (b) (c) 6 (b) (c) 7 (d) figure 3. load during gears' engage () and disengage (■) for first gear (a), second gear(b), third gear(c) and fourth gear (d). table 1: pressure and force for all gears gear n gear 1 gear 2 gear 3 gear 4 action engage disengage engage disengage engage disengage engage disengage average (kg) 4.94 3.54 5.09 3.51 4.93 3.32 5.04 3.44 force (n) 48.46 34.73 49.93 34.43 48.36 32.57 49.44 33.75 pressure (pa) 5.19 3.72 5.35 4.01 5.18 3.49 5.30 3.61 4.0 conclusions the prototype of wireless pneumatic gear changer has been developed. the prototype has proven experimentally produced suitable amount of force and pressure to engage and disengage gears using pneumatic actuator on the gear lever. the prototype can be used by elder riders and special need riders who cannot shift the gear ordinarily using foot. the prototype is available for future improvement and installation on actual motorcycle. acknowledgement the authors gratefully acknowledged the advanced vehicle technology (active) research group of centre for advanced research on energy (care), the financial support from universiti teknikal malaysia melaka and the ministry of education, malaysia (moe) under fundamental research grant scheme (frgs), grant no.: frgs/2013/fkm/tk06/02/2/f00165. figure 3. load during gears’ engage (♦) and disengage (■) for first gear (a), second gear(b), third gear(c) and fourth gear (d). issn: 2180-1053 vol. 7 no. 1 january june 2015 journal of mechanical engineering and technology 8 table 1: pressure and force for all gears 7 (d) figure 3. load during gears' engage () and disengage (■) for first gear (a), second gear(b), third gear(c) and fourth gear (d). table 1: pressure and force for all gears gear n gear 1 gear 2 gear 3 gear 4 action engage disengage engage disengage engage disengage engage disengage average (kg) 4.94 3.54 5.09 3.51 4.93 3.32 5.04 3.44 force (n) 48.46 34.73 49.93 34.43 48.36 32.57 49.44 33.75 pressure (pa) 5.19 3.72 5.35 4.01 5.18 3.49 5.30 3.61 4.0 conclusions the prototype of wireless pneumatic gear changer has been developed. the prototype has proven experimentally produced suitable amount of force and pressure to engage and disengage gears using pneumatic actuator on the gear lever. the prototype can be used by elder riders and special need riders who cannot shift the gear ordinarily using foot. the prototype is available for future improvement and installation on actual motorcycle. acknowledgement the authors gratefully acknowledged the advanced vehicle technology (active) research group of centre for advanced research on energy (care), the financial support from universiti teknikal malaysia melaka and the ministry of education, malaysia (moe) under fundamental research grant scheme (frgs), grant no.: frgs/2013/fkm/tk06/02/2/f00165. 4.0 conclusions the prototype of wireless pneumatic gear changer has been developed. the prototype has proven experimentally produced suitable amount of force and pressure to engage and disengage gears using pneumatic actuator on the gear lever. the prototype can be used by elder riders and special need riders who cannot shift the gear ordinarily using foot. the prototype is available for future improvement and installation on actual motorcycle. acknowledgement the authors gratefully acknowledged the advanced vehicle technology (active) research group of centre for advanced research on energy (care), the financial support from universiti teknikal malaysia melaka and the ministry of education, malaysia (moe) under fundamental research grant scheme (frgs), grant no.: frgs/2013/fkm/tk06/02/2/ f00165. references herbert, l. n. (2005). moving the earth: the workbook excavation. 5th edition. mc-graw hill professional. bosch (1975). the bosch book of the motor car. st martin’s press. library of congress. 206-207. cengel, y.a., boles, m.a., and yaling h. (2009). thermodynamics an engineering approach. mc-graw hill companies. steve, r. (2007). wireless networking technology: from principles to successful implementation. jordon hill oxford. jaap, c.h. (2000). the bluetooth radio system. ericsson radio system b.v. molly, m. and pautot, m.b. (1992). the gsm system for mobile communications. university of michigan. lin, s. and costello, d.j. (1983). error control coding. prentice hall. issn: 2180-1053 vol. 7 no. 1 january june 2015 prototype development of wireless pneumatic gear shifter 9 salonidis, t. (2001). distributed topology construction of bluetooth personal area networks. ieee communication society. lee, g.h. (1995). a study on the speed control of a closed-loop hydrostatic transmission. seoul university. kevin, p.(2007). control of a manual transmission in an electric land speed vehicle. thesis. ohio state university. gerald, a.t. (1996). electronically controlled gear shift mechanism particularly suited for racing cars. united states patent. journal of mechanical engineering and technology *corresponding author. email: ghazali@utem.edu.my issn 2180-1053 vol. 11 no. 1 july-december 2019 22 the effect of different shape pattern of metal interconnects on the electrical and mechanical properties of stretchable conductive circuit a.m. yunos1, g. omar1, 2, n.a.b. masripan1, 2 1 faculty of mechanical engineering, universiti teknikal malaysia melaka, hang tuah jaya, 76100 durian tunggal, melaka, malaysia 2 centre for advanced research on energy, faculty of mechanical engineering, universiti teknikal malaysia melaka, hang tuah jaya, 76100 durian tunggal, melaka, malaysia abstract electrically conductive adhesive (eca) had been extensively studied to replace the sn/pb solder mainly found in printed circuit boards (pcbs) because of their harmful action towards human health and environment. in the production of stretchable pcbs, eca mainly comprises of metallic filler and polymer matrix should perform good electrical and mechanical properties when straining being loaded. therefore, determining the optimum shape pattern to be printed will contribute toward the desired traits of stretchable pcbs. in this study, commercial silver ink and thermoplastic polyurethane (tpu) as substrate was used. the ink was printed on the substrate by doctor-blade technique with different shape patterns with varies widths (1mm, 2mm and 3mm): (a) straight, (b) zigzag, (c) square and (d) sinusoidal. then measurement of sheet resistance by four-point measurement was conducted on unloaded and loaded straining of shape pattern. this study exhibited that 3mm width zig zag shape pattern can elongate the highest straining (5% strained) compare than others patterns. in the meanwhile, straight and square shape pattern did not tolerate to any deformation which when straining at a minimum elongation of 0.1mm, the conductivity already lost. in conclusion, further study purpose, more analysis were suggested like analysis on the silver composition, curing temperature variation as well as the distribution of stress in printed shape pattern by 3d finite element analysis (fea) can be done for the more reliable study. keywords: shape, sheet resistance, elongation, silver, thermoplastic polyurethane 1.0 introduction globally, electronic industry had been leading in term of the fastest growth since the last two decades (poh‐kam, 1995). the rapid growth and advancement of the electronic sector make their availability widespread to the public. however, they reach limitation when it comes to environmental context. for instances, printed circuit boards (pcbs) are known to contained heavy metal reported causing harmful towards human health and environment when improper disposable practice worldwide (hadi, xu, lin, hui, & mckay, 2015). therefore, proper studied on the electrically conductive adhesive (eca) growing interest among researchers as they had been observed to potentially substitute the sn/pb solders in pcbs (jagt, 1998; lee, chou, & shih, 2005). eca comprise of conductive filler which serves as an electrical conductor and polymer matrix for mechanical support in them as well as solvent and additives also included as part of eca journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 1 july-december 2019 23 components. stretchable polymer, thermoplastic polyurethane (tpu) has low elastic modulus and high stretchability was used in this studied. conductive filler had been extensively studied for example carbonous material (graphene (liang et al., 2009), carbon nanotube (cnt) (sandler, kirk, kinloch, shaffer, & windle, 2003) and metallic material (copper (ho et al., 2010; lin & chiu, 2008), silver (lee et al., 2005; merilampi, lainema, & ruuskanen, 2009)). among all the conductive fillers, silver is the most robust and excellent conductivity with sheet resistance typically 0.01-0.04 ω/□ at dry thickness 25 µm of ink layer (merilampi et al., 2009) and also perform good chemical durability (lee et al., 2005). fabricating electronic circuits can be made by several printing techniques like screen printing, gravure printing and inkjet printing (khan, lorenzelli, & dahiya, 2015; yoon et al., 2011), which all these techniques allowed integration on several materials such as paper, plastic and fabrics. however, for this work, printing was done by doctor-blade printing as it is simple, fast and low cost, besides this technique is favoured for small scale experimental as low material consumption compared to screen printing and others (ghediya & chaudhuri, 2014). there are many studied had been successfully investigated on the important parameters (particle content, size and form) affecting electrical and mechanical properties of the pattern (dziedzic, 2007; hicks, allington, & johnson, 1980; lin & chiu, 2004). however, main challenges are when it involved the production of stretchable pcbs, which definitely need conductive patterns and substrate to withstand several degrees of stretching manner before lost its conductivity. therefore, a printed pattern should free from any defects such as porosity and cracks as well as must perform good adhesion between substrate and pattern. the design of shape pattern comparison had been studied by the previous researchers, but mainly they only focusing into the sinusoidal and horseshoe shape pattern (abu-khalaf, saraireh, eisa, & alhalhouli, 2018; gonzalez et al., 2008). in their comparative studied, the parameters were varies based on the amplitude, width of lines, cycles and inner radius. meanwhile, in this study, we are focusing into a comparison of definitely different shape (straight, zigzag, square and sinusoidal) to determine the optimum shape suitable for stretchable electronic application. thus, the objective of this study is to investigate the effect of different shapes patterns with different width ranging between 1mm to 3mm on the conductivity performance upon stretching. 2.0 experimental 2.1 test samples and patterns the doctor-blade printing was used to print conductive silver ink pattern on the polymeric substrate: thermoplastic polyurethane (tpu). the commercial silver ink used in this study and the curing condition was at 120oc by an oven. the prepared test patterns are presented in figure 1 (a) straight, (b) zigzag, (c) square and (d) sinusoidal. all the patterns have length 6 cm and vary line widths: 1mm, 2mm and 3 mm. journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 1 july-december 2019 24 2.2 measurements sheet resistance was measured in unloaded conditions by using four-point measurement method for 6cm length of the sample. the measurement was conducted by dc power by supplying 10 ma of current along the pattern by connecting alligator clips at each end of the pattern. then, the measurement of voltage was performed by using accuracy multimeter. the sheet resistance was obtained by integrated measured voltage results into equation (1): 𝑅𝑠𝑞𝑢𝑎𝑟𝑒 = 𝑐𝑓 × 𝑉 𝐼 (1) where cf is correction factor, rsquare is sheet resistance, v is a voltage between inner probes and i is applied current. the correction factor was assumed to be π ln 2 = 4.53 (kalavagunta, a., & weller, 2005). the resistance of the conductor was measure when tensile test performed to investigate the electrical performance during stretchability. each sample was strained until the conductivity was lost. the straining was conducted by manually stretching, which the stretched sample was fixed on the glass slide by clipped each end by using paper clipper and further measurement on the specific point of each pattern was conducted in the same manner during the unloaded condition. (c) figure 1. shape and dimensions of prepared patterns (b) (d) (a) journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 1 july-december 2019 25 3.0 results and discussion the sheet resistance of the different shapes with different widths: 1mm, 2mm and 3mm are presented in figure 2. the sheet resistance of the samples at unloaded condition was smaller for the sample with a straight line. in this study, the length of each pattern is being fixed to 6cm (refer to figure 1) for consistent straining of the shape pattern. however, if a total of pattern length is being considered, hence square, sinusoidal and zigzag having much longer length when measured. according to pouillet's law, the resistance of the material is directly proportional to the length. thus, square, sinusoidal and zigzag displayed larger resistances than straight shape pattern. besides, the plots exhibited incline trend of measured sheet resistance when increasing the width in all shape patterns. this due to the more metallic silver particle content in a larger width than smaller, hence showed good conductivity performance. however, in this studies, observation on the ability of each shape pattern to stretch is much interested which further discussed in figure 3 and 4. figure 3 gives a picture of the relation between the shape of the pattern and the maximum elongation for a different shape. 3mm width of zigzag shape showed the highest elongation (0.3mm/5% strain) before it fails. these plots show a clear trend which is a directly proportional relationship between elongation and reading sheet resistance. this explained that shape pattern cannot withstand any further elasticity hence lead to plastic deformation that expressed in term of fracture presence. the fracture cause restriction of currents from transpassing along the printed circuit. besides, straight and square shape pattern giving early failure as they not allowed deformation to occur. this is due to the 0 5 10 15 20 25 30 0 0.1 0.2 0.3 s h e e t r e si st a n ce (m ω /□ ) displacement (mm) weavy-1mm weavy-3mm zigzag-2mm zigzag-3mm figure 3. sheet resistance of different shape when elongate at different displacement 0 5 10 15 20 25 30 1 2 3 1 2 3 1 2 3 1 2 3 straight square sinusoidal zig zag s h e e t re si st a n ce s (m ω /□ ) width (mm) figure 2. sheet resistance of different shapes with different widths at the unloaded condition journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 1 july-december 2019 26 high concentration of stress exist in the region of printed shape pattern that parallels to the applied load. moreover, it also supported with the properties of metal conductors that typically have limited elastic ranges, therefore they required proper shape design in fabricating the conductive circuit for the stretching to be possible (gonzalez et al., 2008). furthermore, larger width of shape pattern (3mm) can allow the production of conductivity up till 0.2mm and 0.3mm elongation for sinusoidal and zigzag shape pattern respectively before failed. we believe this observation result from the small fractures in the printed shape pattern upon stretching do not individually span the width of the circuit, instead of, the fractures are spread out over a large area, hence allowing conductivity around their edge. this small fracture allowed to spread in the long distance manner when shape pattern having larger width but not in the smaller width of shape pattern due to the fractures tends to concentrate or spread in short distance manner because of small area present. many intensive studied had been done on the sinusoidal shape (abu-khalaf et al., 2018; gonzalez et al., 2008). the evaluation was done by mario and his team on the horseshoe design, present that semicircle design (θ=0o) will have small resistivity as well as smaller elongation (figure 5). besides that, they also described the width of metal track was defined as an important parameter in allowing high deformation of the structure (gonzalez et al., 2008). thus, this explained the reason why sinusoidal shape pattern in this work (figure 4) perform low deformation when stretching and elongation possible only in the highest width (3mm) of sinusoidal shape pattern. furthermore, figure 4 and 5 present that zigzag shape showed the highest elongation compare than other shapes while maintaining the smallest measurement of sheet resistance. this counter the theory discussed before on the parameter (radius and joining angle) that are needed to consider for designing a shape pattern which zigzag shape did not have. instead of, zig-zag shape pattern showed high deformation performance due to the slanted shape (45o) which allowed the shape to has less stress in that reading point region compared to sinusoidal shape which accumulated plastic strain located mostly at the meander part of the shape which leads to fracture when load applied which can refer in figure 6. 0 1 2 3 4 5 6 7 0 0.1 0.2 0.3 0.4 0.5 s h e e t r e si st a m ce ( m ω /□ ) displacement (mm) sinusoidal3mm zig zag-1mm figure 4. sheet resistance of different shape at point parallel to the applied stretch journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 1 july-december 2019 27 4.0 further outcomes although zigzag pattern showed better performance of stretchability when comparing with different shape pattern in these studies, maximum 5% strained is too little for it to be integrated into a daily application. when comparing with other study, the maximum strain can reach about 80% strained which indicate better mechanical performance and able to maintained low sheet resistivity (merilampi et al., 2009). further studied needed in term of ink characterization for silver composition percentage and also curing temperature variation as these greatly affected sheet resistance of the pattern. the silver composition (at a concentration about 10-30 vol%) is important to analyze for determining the percolation threshold for electrical transportation phenomena (‘tunneling') to be performed (hu, karube, yan, masuda, & fukunaga, 2008; sevkat, li, liaw, & delale, 2008). the curing temperature also affected the sheet resistance as it contributes toward different evaporation rates of potentially available solvent in ink composition. moreover, performing 3d finite element analysis (fea) will give thermomechanical modelling of sample pattern which displayed the distribution of stress or strain in different parts of a structure, thus making studies more reliable. figure 5. definition of horseshoe design by its inner radius (r), joining angle (θ) and width of metal track (w) (gonzalez et al., 2008) figure 6. plastic strain distribution in a multi-track horseshoe conductor design (gonzalez et al., 2008) journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 1 july-december 2019 28 5.0 references abu-khalaf, j., saraireh, r., eisa, s., & al-halhouli, a. (2018). experimental characterization of inkjet-printed stretchable circuits for wearable sensor applications. sensors, 18(10), 3476. https://doi.org/10.3390/s18103476 dziedzic, a. (2007). carbon/polyesterimide thick-film resistive composites – experimental determination and theoretical analysis of physicochemical , electrical and stability properties. microelectronics reliability, 47, 354–362. ghediya, p., & chaudhuri, t. (2014). doctor-blade printing of cu 2 znsns 4 films from microwave-processed ink doctor-blade printing of cu 2 znsns 4 films from microwave-processed ink. journal of materials science: materials in electronics, 23(3), 1908–1912. https://doi.org/10.1007/s10854-014-2628-1 gonzalez, m., axisa, f., vanden, m., brosteaux, d., vandevelde, b., & vanfleteren, j. (2008). design of metal interconnects for stretchable electronic circuits. microelectronics reliability, 48, 825–832. https://doi.org/10.1016/j.microrel.2008.03.025 hadi, p., xu, m., lin, c. s. k., hui, c., & mckay, g. (2015). waste printed circuit board recycling techniques and product utilization. journal of hazardous materials, 283, 234–243. hicks, w. t., allington, t. r., & johnson, v. (1980). membrane touch switches: thickfilm materials systems and processing options. ieee transactions on components, hybrids, and manufacturing technology, 3(4), 518–524. https://doi.org/10.1109/tchmt.1980.1135649 ho, l., nishikawa, h., natsume, n., takemoto, t., miyake, k., & fujita, m. (2010). effects of trace elements in copper fillers on the electrical properties of conductive adhesives. journal of electronic mater, 39(1), 115–123. https://doi.org/10.1007/s11664-009-0946-5 hu, n., karube, y., yan, c., masuda, z., & fukunaga, h. (2008). tunneling effect in a polymer/carbon nanotube nanocomposite strain sensor. acta materialia, 56(13), 2929–2936. jagt, j. c. (1998). reliability of electrically conductive adhesive joints for surface mount applications: a summary of the state of the art. ieee transactions on components packaging and manufacturing technology part a, 21(2), 215–225. https://doi.org/10.1109/95.705467 kalavagunta, a., & weller, r. a. (2005). accurate geometry factor estimation for the four point probe method using comsol multiphysics. in proceedings of the comsol users conference. boston. khan, s., lorenzelli, l., & dahiya, r. s. (2015). technologies for printing sensors and electronics over large flexible substrates: a review. ieee sensors journal, 15(6), 3164–3185. https://doi.org/10.1109/jsen.2014.2375203 journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 1 july-december 2019 29 lee, h., chou, k., & shih, z. (2005). effect of nano-sized silver particles on the resistivity of polymeric conductive adhesives. international journal of adhesion and adhesives, 25, 437–441. https://doi.org/10.1016/j.ijadhadh.2004.11.008 liang, j., wang, y., huang, y., ma, y., liu, z., & cai, j. (2009). electromagnetic interference shielding of graphene / epoxy composites. carbon, 47(3), 922–925. https://doi.org/10.1016/j.carbon.2008.12.038 lin, y., & chiu, s. (2004). effects of oxidation and particle shape on critical volume fractions of silver-coated copper powders in conductive adhesives for microelectronic applications. polymer engineering and science, 44(11), 2075– 2082. https://doi.org/10.1002/pen.20212 lin, y., & chiu, s. (2008). electrical properties of copper-filled electrically conductive adhesives and pressure-dependent conduction behavior of copper particles. journal of adhesion science and technology, 22(14), 1673–1697. https://doi.org/10.1163/156856108x320537 merilampi, s., laine-ma, t., & ruuskanen, p. (2009). the characterization of electrically conductive silver ink patterns on flexible substrates. microelectronics reliability, 49(7), 782–790. https://doi.org/10.1016/j.microrel.2009.04.004 poh‐kam, w. (1995). competing in the global electronics industry: a comparative study of the innovation networks of singapore and taiwan. journal of industry studies, 2(2), 35–61. sandler, j. k. w., kirk, j. e., kinloch, i. a., shaffer, m. s. p., & windle, a. h. (2003). ultra-low electrical percolation threshold in carbon-nanotube-epoxy composites. polymer, 44, 5893–5899. https://doi.org/10.1016/s0032-3861(03)00539-1 sevkat, e., li, j., liaw, b., & delale, f. (2008). a statistical model of electrical resistance of carbon fiber reinforced composites under tensile loading. composites science and technology, 68, 2214–2219. https://doi.org/10.1016/j.compscitech.2008.04.011 yoon, b., ham, d., yarimaga, o., an, h., lee, c. w., & kim, j. (2011). inkjet printing of conjugated polymer precursors on paper substrates for colorimetric sensing and flexible electrothermochromic display. advanced materials, 23, 5492–5497. https://doi.org/10.1002/adma.201103471 issn: 2180-1053 vol. 2 no. 1 january-june 2010 pd-fuzzy logic controlled on a magnetic bearing system 71 pd-fuzzy logic controlled on a magnetic bearing system s. i. samsudin1, h. r. a. rahim2, and a. n. m. jahari3 1,2,3faculty of electronic and computer engineering, universiti teknikal malaysia melaka, locked bag 1752, pejabat pos durian tunggal, 76109 durian tunggal, melaka. abstract a magnetic bearing system is a device that uses electromagnetic forces to support a rotor without mechanical contact. the force exerted on the rotor is determined by the current flow in the magnet coil. this project will be focused on the stability and control of the mbc 500 system test bed constructed by magnetic moments incorporated. the mbc 500 system contains a stainless steel shaft or rotor which can be levitated using eight horseshoe electromagnets. a controller which is able to stabilize the position of the rotor by varying the electromagnet force produced by the electromagnets at each end of the shaft will be developed. here, direct fuzzy logic controller and proportional derivative fuzzy logic controller with mamdani’s inference method are designed. keywords: fuzzy control, magnetic bearing system. 1.0 background on magnetic bearing system 1.1 introduction to mbc 500 the mbc 500 is a desktop test bed made by magnetic moments incorporation and the system can be seen as shown as in figure 1. it consist of a horizontal stainless steel shaft or rotor which can be levitated using eight horseshoe electromagnets, four at each end of the rotor (paden et al.) (morse et al., 1996). the shaft never even touches the bearings when it is operating since there are two silver housings that hold the electromagnetic bearings which levitate the spindle. issn: 2180-1053 vol. 2 no. 1 january-june 2010 journal of mechanical engineering and technology 72 figure 1 mbc 500 magnetic bearing systems figure 2 shows the location of magnetic bearing. the copper-colored are the coils of wire that make up the electromagnet. meanwhile, figure 3 represents an attractive force exerted by electromagnet. figure 2 magnetic bearing figure 3 attractive force exerted by electromagnet one problem when using electromagnets is that they can only produce an attractive force. the force is stronger when the spindle is closer to the magnet, which in turn brings the spindle more close and makes the force even stronger. this leads to an unstable system. as a result, the magnets must be arranged radially around a spindle, so that a magnet on one side of the spindle can counteract the force exerted by a magnet issn: 2180-1053 vol. 2 no. 1 january-june 2010 pd-fuzzy logic controlled on a magnetic bearing system 73 on the other side of the spindle. moreover, a control system is required to make the spindle levitate. the force exerted by the magnets can be controlled by changing the current flowing through the coils of wire. equation (1) shows that a greater current increases the force exerted by the magnet. 2 figure 2 magnetic bearing figure 3 attractive force exerted by electromagnet one problem when using electromagnets is that they can only produce an attractive force. the force is stronger when the spindle is closer to the magnet, which in turn brings the spindle more close and makes the force even stronger. this leads to an unstable system. as a result, the magnets must be arranged radially around a spindle, so that a magnet on one side of the spindle can counteract the force exerted by a magnet on the other side of the spindle. moreover, a control system is required to make the spindle levitate. the force exerted by the magnets can be controlled by changing the current flowing through the coils of wire. equation (1) shows that a greater current increases the force exerted by the magnet. 2 2 g i f  (1) where f = exerted force i = current flowing through the coils of wire g = distance between spindle and magnetic bearing 1.2 advantages and disadvantages of a magnetic bearing system magnetic bearing offer significant advantages because they do not come into contact with other parts during operation, which can reduce the maintenance. higher speeds, no friction, no lubrication, weight reduction, precise position control, and active damping make them far superior to conventional contact bearings (rebecca et al., 2000). however, there are rephrase that limit the application of the magnetic bearing such as to balance the electromagnets forces which are exerted on the magnetic bearing and to maintain the position of the rotor at the equilibrium point (rebecca where f = exerted force i = current flowing through the coils of wire g = distance between spindle and magnetic bearing 1.2 advantages and disadvantages of a magnetic bearing system magnetic bearing offer significant advantages because they do not come into contact with other parts during operation, which can reduce the maintenance. higher speeds, no friction, no lubrication, weight reduction, precise position control, and active damping make them far superior to conventional contact bearings (rebecca et al., 2000). however, there are rephrase that limit the application of the magnetic bearing such as to balance the electromagnets forces which are exerted on the magnetic bearing and to maintain the position of the rotor at the equilibrium point (rebecca et al., 2000). thus, magnetic bearing needs a controller which is able to stabilize the position of the rotor during operation before it works effectively. 2.0 analysis and system modeling for magnetic bearing system the target system for this project is the magnetic moments mbc 500 magnetic bearing system. a diagram of this system is shown in figure 4. it has a stainless steel shaft or rotor which is levitated using eight horseshoe electromagnets at each end of the rotor. hall effect sensors are placed just outside of the electromagnets at each end of the rotor to measure the rotor end displacement. this system is a four degree of freedom system with two degrees of freedom at each end of the rotor. these two degrees of freedom are translation in the horizontal direction, perpendicular to the z axis (x1 and x2) and translation in the issn: 2180-1053 vol. 2 no. 1 january-june 2010 journal of mechanical engineering and technology 74 vertical direction (y1 and y2) (paden et al.) (morse et al., 1996). in this project, the rotor is assumed as as a rigid body. a rigid body is considered as it does not change its shape. therefore, it is assumed that the rotor does not bend but rather experiences only translational or rotational motion. in addition, it is assumed that the horizontal and vertical dynamics (x and y directions) are uncoupled. the system in theory operates identically in the x and y directions except for the additional constant force due to gravity acting in the y direction. here, gravity in the linear y direction analysis is neglected. thus, analysis of the x and y directions is identical and the focus of analysis is strictly on the horizontal or x direction motion (morse et al., 1996). figure 4 mbc500 system figure 5 shows rotor configuration of mbc 500 magnetic bearing system. meanwhile, table 1 and 2 represent system variables and parameters. the nominal or desired rotor position corresponds to x1=0, and x2=0 or (equivalently x1=0 and x2=0 or x0=0 and =0). in this position, the rotor is centered horizontally with respect to the front and back electromagnets on each end and its long axis is parallel to the z axis (morse et al., 1996). figure 5 rotor configuration issn: 2180-1053 vol. 2 no. 1 january-june 2010 pd-fuzzy logic controlled on a magnetic bearing system 75 table 1 system variables table 2 system parameters the state-space model of the magnetic bearing system can be represented as 4 figure 5 rotor configuration table 1 system variables symbol description x0 the displacement of center of mass of rotor x1 and x2 the displacement of rotor at left and right bearings x1 and x2 the displacement of rotor at hall effect sensor  the angle that the long axis of the rotor makes with the z axis f1 and f2 the forces exerted on the rotor by left and right bearing table 2 system parameters symbol description value l total length of the rotor 0.269m l distance from each bearing to the end of the rotor 0.024m 2l distance from each hall-effect sensor to the end of the rotor 0.0028m 0l moment of inertia of the rotor with respect to rotation about an axis in the y direction 1.5884 x 10-3 kg m2 m mass of the rotor 0.2629kg the state-space model of the magnetic bearing system can be represented as where (2) where and the output equation can be written as 5 and the output equation can be written as in control engineering, experimental determination of a system model is an important part of the modeling process. this is referred as a system identification. the transfer function of the mbc 500 magnetic bearing system was obtained by measuring the frequency response of the closed loop system (shi et al., 2002). the toolboxes of matlab are used as to fit the transfer function model from the data collected. figure 6 shows a general block diagram for a closed loop system identification. here, r is the input signal, u is the input signal to the plant g, n is the measurement of noise, and y is the output signal. figure 6 closed loop system identification for negative feedback of a closed loop system, the input signal, u and the output signal, y can be described as in equation (4) and (5). (5) hence, the transfer function of the mbc 500 magnetic bearing system can be represented as (3) in control engineering, experimental determination of a system model is an important part of the modeling process. this is referred as a issn: 2180-1053 vol. 2 no. 1 january-june 2010 journal of mechanical engineering and technology 76 system identification. the transfer function of the mbc 500 magnetic bearing system was obtained by measuring the frequency response of the closed loop system (shi et al., 2002). the toolboxes of matlab are used as to fit the transfer function model from the data collected. figure 6 shows a general block diagram for a closed loop system identification. here, r is the input signal, u is the input signal to the plant g, n is the measurement of noise, and y is the output signal. 5 and the output equation can be written as in control engineering, experimental determination of a system model is an important part of the modeling process. this is referred as a system identification. the transfer function of the mbc 500 magnetic bearing system was obtained by measuring the frequency response of the closed loop system (shi et al., 2002). the toolboxes of matlab are used as to fit the transfer function model from the data collected. figure 6 shows a general block diagram for a closed loop system identification. here, r is the input signal, u is the input signal to the plant g, n is the measurement of noise, and y is the output signal. figure 6 closed loop system identification for negative feedback of a closed loop system, the input signal, u and the output signal, y can be described as in equation (4) and (5). (5) hence, the transfer function of the mbc 500 magnetic bearing system can be represented as (3) figure 6 closed loop system identification for negative feedback of a closed loop system, the input signal, u and the output signal, y can be described as in equation (4) and (5). hence, the transfer function of the mbc 500 magnetic bearing system can be represented as 3.0 fuzzy logic control design approach 3.1 introduction to fuzzy logic controller fuzzy concepts derive from fuzzy phenomena that commonly occur in the natural world. the concepts formed in human brains for perceiving, recognizing, and categorizing natural phenomena are often fuzzy concepts. one such control strategy is the use of fuzzy logic based control. figure 7 shows the typical structure of a fuzzy logic controller. basically, a fuzzy logic controller consists of fuzzifier, knowledge base, issn: 2180-1053 vol. 2 no. 1 january-june 2010 pd-fuzzy logic controlled on a magnetic bearing system 77 inference engine and defuzzifier. 6 3.0 fuzzy logic control design approach 3.1 introduction to fuzzy logic controller fuzzy concepts derive from fuzzy phenomena that commonly occur in the natural world. the concepts formed in human brains for perceiving, recognizing, and categorizing natural phenomena are often fuzzy concepts. one such control strategy is the use of fuzzy logic based control. figure 7 shows the typical structure of a fuzzy logic controller. basically, a fuzzy logic controller consists of fuzzifier, knowledge base, inference engine and defuzzifier. figure 7 block diagram of fuzzy controller 3.2 fuzzy logic controller fuzzy controllers are being used in various control schemes. the most obvious one is a direct control, where the fuzzy controller is in the forward path in a feedback control system. the process output is compared with a reference, and if there is a deviation, the controller takes an action according to the control strategy. the block diagram of a direct fuzzy control can be shown as in figure 8. figure 8 fuzzy logic controllers figure 7 block diagram of fuzzy controller 3.2 fuzzy logic controller fuzzy controllers are being used in various control schemes. the most obvious one is a direct control, where the fuzzy controller is in the forward path in a feedback control system. the process output is compared with a reference, and if there is a deviation, the controller takes an action according to the control strategy. the block diagram of a direct fuzzy control can be shown as in figure 8. 6 3.0 fuzzy logic control design approach 3.1 introduction to fuzzy logic controller fuzzy concepts derive from fuzzy phenomena that commonly occur in the natural world. the concepts formed in human brains for perceiving, recognizing, and categorizing natural phenomena are often fuzzy concepts. one such control strategy is the use of fuzzy logic based control. figure 7 shows the typical structure of a fuzzy logic controller. basically, a fuzzy logic controller consists of fuzzifier, knowledge base, inference engine and defuzzifier. figure 7 block diagram of fuzzy controller 3.2 fuzzy logic controller fuzzy controllers are being used in various control schemes. the most obvious one is a direct control, where the fuzzy controller is in the forward path in a feedback control system. the process output is compared with a reference, and if there is a deviation, the controller takes an action according to the control strategy. the block diagram of a direct fuzzy control can be shown as in figure 8. figure 8 fuzzy logic controllers figure 8 fuzzy logic controllers 4.0 result on fuzzy controlled 4.1 stability test on magnetic bearing system before starting this project, the stability of the magnetic bearing system is identified. for this purpose, the locations of poles of the system were considered. location of poles can be determined from the eigen values issn: 2180-1053 vol. 2 no. 1 january-june 2010 journal of mechanical engineering and technology 78 of system matrix. it was observed that system is unstable since there have positive poles. it was observed too that the uncontrolled plant of the system is unstable since the output response goes to infinity. these proved that the system under test is unstable and requires a controller. 4.2 description of fuzzy logic controller the project aims to present the implementation of a fuzzy logic control, flc strategy for stabilizing the unstable response in mbc 500. this control strategy is expected to stabilize the position of the rotor of magnetic bearing system. simulation of the non-linear system shows that for certain operating parameters, the mbc 500 exhibits unstable response for a rotor position. however, the use of fuzzy logic control has been able to eliminate this instability and improve the rotor stability performance. uncontrolled mbc 500 plant was developed using matlab. since mbc 500 is a non-linear system, the deployment of flc is highly commendable. the non-linearity of the system is expressed using fuzzy principles in linguistic variable descriptions. the implementation of fuzzy control in matlab was done in two stages analysis. the first stage covers the controlled action of direct fuzzy logic controller, flc while the second analysis covers the controlled action of proportional derivative fuzzy logic controller, pdflc. figure 9 shows the block diagram of direct flc while figure 10 shows the pd-flc. 7 4.0 result on fuzzy controlled 4.1 stability test on magnetic bearing system before starting this project, the stability of the magnetic bearing system is identified. for this purpose, the locations of poles of the system were considered. location of poles can be determined from the eigen values of system matrix. it was observed that system is unstable since there have positive poles. it was observed too that the uncontrolled plant of the system is unstable since the output response goes to infinity. these proved that the system under test is unstable and requires a controller. 4.2 description of fuzzy logic controller the project aims to present the implementation of a fuzzy logic control, flc strategy for stabilizing the unstable response in mbc 500. this control strategy is expected to stabilize the position of the rotor of magnetic bearing system. simulation of the non-linear system shows that for certain operating parameters, the mbc 500 exhibits unstable response for a rotor position. however, the use of fuzzy logic control has been able to eliminate this instability and improve the rotor stability performance. uncontrolled mbc 500 plant was developed using matlab. since mbc 500 is a non-linear system, the deployment of flc is highly commendable. the non-linearity of the system is expressed using fuzzy principles in linguistic variable descriptions. the implementation of fuzzy control in matlab was done in two stages analysis. the first stage covers the controlled action of direct fuzzy logic controller, flc while the second analysis covers the controlled action of proportional derivative fuzzy logic controller, pd-flc. figure 9 shows the block diagram of direct flc while figure 10 shows the pd-flc. figure 9 direct fuzzy logic controllers figure 9 direct fuzzy logic controllers issn: 2180-1053 vol. 2 no. 1 january-june 2010 pd-fuzzy logic controlled on a magnetic bearing system 79 8 figure 10 pd fuzzy logic controllers for this purpose, input for the fuzzy properties referring as error, e(t) and the output referring as vcontrol, are applied in direct fuzzy controlled. on the other hand, the change-of-error, d(e)/d(t) is applied as derivative input in pd-flc. the output, vcontrol will regulates the current into the bearing which then regulate the magnetic bearing force. channel x1 is considered. the displacement output is sensed by the hall-effect sensor with the output voltage, vsense. for the magnetic bearing stabilization problem, the reference input is set to 0. figure 11 shows the interface of direct flc while figure 12 indicates the interface of pd-flc. figure 11 interface of direct flc figure 12 interface of pd-flc figure 10 pd fuzzy logic controllers for this purpose, input for the fuzzy properties referring as error, e(t) and the output referring as vcontrol, are applied in direct fuzzy controlled. on the other hand, the change-of-error, d(e)/d(t) is applied as derivative input in pd-flc. the output, vcontrol will regulates the current into the bearing which then regulate the magnetic bearing force. channel x1 is considered. the displacement output is sensed by the hall-effect sensor with the output voltage, vsense. for the magnetic bearing stabilization problem, the reference input is set to 0. figure 11 shows the interface of direct flc while figure 12 indicates the interface of pd-flc. figure 11 interface of direct flc figure 12 interface of pd-flc issn: 2180-1053 vol. 2 no. 1 january-june 2010 journal of mechanical engineering and technology 80 the linguistic values have been used to describe the inputs and output of the flc to specify a set of rules how to control the plant. initially, five linguistic values are considered in fuzzy controlled analysis includes negative big, negative small, zero, positive small and positive big. in addition, seven linguistic values have been used include negative big, negative medium, negative small, zero, positive small, positive medium and positive big. the triangular membership function (trimf) is used in mapping the input point to membership values and it was normalized universe of discourse at [-1, 1]. the if-then rules are generated to infer the proper value for each of the output variables. for this purpose, 5 rules based on mamdani’s approach are applied to direct flc while 25 rules are applied to pdflc. it then followed by developing 7 rules based on direct flc and 49 rules based on pd-flc. the 5 rules base developed in direct flc include; if error is nb then force is nb if error is ns then force is ns if error is ze then force is ze if error is pb then force is pb if error is ps then force is ps the 25 rules base developed in pd-flc represented as in table 3; ‘ table 3 25 rules base of pd-flc 9 the linguistic values have been used to describe the inputs and output of the flc to specify a set of rules how to control the plant. initially, five linguistic values are considered in fuzzy controlled analysis includes negative big, negative small, zero, positive small and positive big. in addition, seven linguistic values have been used include negative big, negative medium, negative small, zero, positive small, positive medium and positive big. the triangular membership function (trimf) is used in mapping the input point to membership values and it was normalized universe of discourse at [-1, 1]. the if-then rules are generated to infer the proper value for each of the output variables. for this purpose, 5 rules based on mamdani’s approach are applied to direct flc while 25 rules are applied to pd-flc. it then followed by developing 7 rules based on direct flc and 49 rules based on pdflc. the 5 rules base developed in direct flc include; if error is nb then force is nb if error is ns then force is ns if error is ze then force is ze if error is pb then force is pb if error is ps then force is ps the 25 rules base developed in pd-flc represented as in table 3; table 3 25 rules base of pd-flc e(t) δe(t) nb ns ze ps pb nb nb nb nb ns ze ns nb nb ns ze ps ze nb ns ze ps pb ps ns ze ps pb pb pb ze ps pb pb pb meanwhile, the 7 rules base developed in direct flc include: if error is nb then force is nb if error is nm then force is nm if error is ns then force is ns if error is ze then force is ze if error is ps then force is ps if error is pm then force is pm if error is pb then force is pb meanwhile, the 7 rules base developed in direct flc include: if error is nb then force is nb if error is nm then force is nm if error is ns then force is ns if error is ze then force is ze if error is ps then force is ps issn: 2180-1053 vol. 2 no. 1 january-june 2010 pd-fuzzy logic controlled on a magnetic bearing system 81 if error is pm then force is pm if error is pb then force is pb table 4 represents the 49 rules base developed in pd-flc; table 4 49 rules base of pd-flc 10 table 4 represents the 49 rules base developed in pd-flc; table 4 49 rules base of pd-flc e(t) δe(t) nb nm ns ze ps pm pb nb nb nb nb nb nm ns ze nm nb nb nb nm ns ze ps ns nb nb nm ns ze ps pm ze nb nm ns ze ps pm pb ps nm ns ze ps pm pb pb pm ns ze ps pm pb pb pb pb ze ps pm pb pb pb pb figure 13 shows the response of a direct flc with 5 linguistic values includes negative big, negative small, zero, positive small and positive big. this response indicates that the position of the rotor is able to be stabilized at 0.145s. on the other hand, figure 14 represents the response of controlled pd-flc and the system was able to be controlled at 0.1s. these were proved that the unstable position of the rotor of mbc 500 is able to be stabilized at desired position using fuzzy logic controller. thus, the position of rotor would be maintained and capable to return to its reference position successfully. figure 13 output controlled by direct flc with 5 linguistic values figure 13 shows the response of a direct flc with 5 linguistic values includes negative big, negative small, zero, positive small and positive big. this response indicates that the position of the rotor is able to be stabilized at 0.145s. on the other hand, figure 14 represents the response of controlled pd-flc and the system was able to be controlled at 0.1s. these were proved that the unstable position of the rotor of mbc 500 is able to be stabilized at desired position using fuzzy logic controller. thus, the position of rotor would be maintained and capable to return to its reference position successfully. 10 table 4 represents the 49 rules base developed in pd-flc; table 4 49 rules base of pd-flc e(t) δe(t) nb nm ns ze ps pm pb nb nb nb nb nb nm ns ze nm nb nb nb nm ns ze ps ns nb nb nm ns ze ps pm ze nb nm ns ze ps pm pb ps nm ns ze ps pm pb pb pm ns ze ps pm pb pb pb pb ze ps pm pb pb pb pb figure 13 shows the response of a direct flc with 5 linguistic values includes negative big, negative small, zero, positive small and positive big. this response indicates that the position of the rotor is able to be stabilized at 0.145s. on the other hand, figure 14 represents the response of controlled pd-flc and the system was able to be controlled at 0.1s. these were proved that the unstable position of the rotor of mbc 500 is able to be stabilized at desired position using fuzzy logic controller. thus, the position of rotor would be maintained and capable to return to its reference position successfully. figure 13 output controlled by direct flc with 5 linguistic values figure 13 output controlled by direct flc with 5 linguistic values issn: 2180-1053 vol. 2 no. 1 january-june 2010 journal of mechanical engineering and technology 82 11 figure 14 output controlled by pd-flc with 25 linguistic values figure 15 shows the response of a direct flc with 7 linguistic values includes negative big, negative medium, negative small, zero, positive small, positive medium and positive big. this response indicates that the position of the rotor is able to be stabilized at 0.08s. on the other hand, figure 16 represents the response of controlled pd-flc and the system was able to be controlled at 0.18s. again, these were proved that both fuzzy logic control strategy are able to stabilize an unstable position of the rotor of mbc 500. thus, the position of rotor would be maintained and capable to return to its reference position effectively. figure 15 output controlled by direct flc with 7 linguistic values figure 14 output controlled by pd-flc with 25 linguistic values figure 15 shows the response of a direct flc with 7 linguistic values includes negative big, negative medium, negative small, zero, positive small, positive medium and positive big. this response indicates that the position of the rotor is able to be stabilized at 0.08s. on the other hand, figure 16 represents the response of controlled pd-flc and the system was able to be controlled at 0.18s. again, these were proved that both fuzzy logic control strategy are able to stabilize an unstable position of the rotor of mbc 500. thus, the position of rotor would be maintained and capable to return to its reference position effectively. 11 figure 14 output controlled by pd-flc with 25 linguistic values figure 15 shows the response of a direct flc with 7 linguistic values includes negative big, negative medium, negative small, zero, positive small, positive medium and positive big. this response indicates that the position of the rotor is able to be stabilized at 0.08s. on the other hand, figure 16 represents the response of controlled pd-flc and the system was able to be controlled at 0.18s. again, these were proved that both fuzzy logic control strategy are able to stabilize an unstable position of the rotor of mbc 500. thus, the position of rotor would be maintained and capable to return to its reference position effectively. figure 15 output controlled by direct flc with 7 linguistic values figure 15 output controlled by direct flc with 7 linguistic values issn: 2180-1053 vol. 2 no. 1 january-june 2010 pd-fuzzy logic controlled on a magnetic bearing system 83 12 figure 16 output controlled by pd-flc with 49 linguistic values by referring to figure 13-16, these were proved that the rotor’s position was able to be controlled at 0.145s with direct flc and 5 linguistic values compared to 0.08s with 7 linguistic values. on the other hand, the position of the rotor is stabilized at 0.1s with pd-flc and 25 linguistic values compared to 0.18s with 49 linguistic values. 5.0 conclusion the system wish to be controlled is mbc 500 magnetic bearing system with a fuzzy logic controller. both techniques are explored which direct flc and proportional derivative flc. the incorporation of developed fuzzy logic controllers into the existing system shows successful results in stabilization a non-linear behavior in the rotor response of the mbc 500 within the range of operating parameters. it was observed that the shaft is returned to its reference position at the end of control process and these were proved by the responses of controlled output of direct flc and pd flc. 6.0 references a. e. hartavi and o. ustu. 2003. a comparative approach on pd and fuzzy control of amb using rcp. ieee. b. paden. operating manual for the mbc500 magnetic moments inc. santa barbara ca. d. l., trumper and s.m. olson. 1997. linearizing control of magnetic suspension systems. ieeetrans. on control system technology. f. matsumura and t. yoshimoto. 1986. system modeling and control design of a horizontal-shaft magnetic bearing system. ieee transaction on magnetics. g.j. ballas and j. c. doyle. 1995. matlab μ-analysis and syntesis toolbox. natick, ma, usa. the mathworks. figure 16 output controlled by pd-flc with 49 linguistic values by referring to figure 13-16, these were proved that the rotor’s position was able to be controlled at 0.145s with direct flc and 5 linguistic values compared to 0.08s with 7 linguistic values. on the other hand, the position of the rotor is stabilized at 0.1s with pd-flc and 25 linguistic values compared to 0.18s with 49 linguistic values. 5.0 conclusion the system wish to be controlled is mbc 500 magnetic bearing system with a fuzzy logic controller. both techniques are explored which direct flc and proportional derivative flc. the incorporation of developed fuzzy logic controllers into the existing system shows successful results in stabilization a non-linear behavior in the rotor response of the mbc 500 within the range of operating parameters. it was observed that the shaft is returned to its reference position at the end of control process and these were proved by the responses of controlled output of direct flc and pdflc. 6.0 references a. e. hartavi and o. ustu. (2003). a comparative approach on pd and fuzzy control of amb using rcp. ieee. b. paden. operating manual for the mbc500 magnetic moments inc. santa barbara ca. d. l., trumper and s.m. olson. (1997). linearizing control of magnetic suspension systems. ieeetrans. on control system technology. f. matsumura and t. yoshimoto. (1986). system modeling and control design issn: 2180-1053 vol. 2 no. 1 january-june 2010 journal of mechanical engineering and technology 84 of a horizontal-shaft magnetic bearing system. ieee transaction on magnetics. g.j. ballas and j. c. doyle. (1995). matlab μ-analysis and syntesis toolbox. natick, ma, usa. the mathworks. h. chen. (2001). fuzzy neural intelligent systems: mathematical foundation and the applications in engineering. crc press llc. florida. j. shi and j. revell. (2002). system identification and re-engineering controller for a magnetic bearing system. ieee tencon. pp 1591-1594. j.y. hung. (1991). nonlinear control of electromagnetic systems. conf. ieee ind. electron soc. j.y. hung. (1995). magnetic bearing control using fuzzy logic. ieee transactions on industry applications. vol. 31. pp 1492-1497. k. ogata. (2002). modern control engineering. pearson education international. 4th edition. m. chen and c.l. knospe. feedback linearization of active magnetic bearing: currentmode implementation. ieee/asme transaction on mechatronic. pp 632-639. m. k. habib, and j. i. inayat-hussain. (2003). control of dual acting magnetic bearing actuator system using fuzzy logic. proceedings 2003 ieee international symposium on intelligence. n. morse, r. smith and b. paden. (1996). mbc 500 analytical modeling of a magnetic bearing system. mbc 500 magnetic system operating instruction. pp 1-14. p. rebecca and p. gordon. (2000). disturbance rejection control of an electromagnetic bearing spindle. r. c. dorf and r.h.bishop. modern control systems. addisonwesley publishing company. s. k. hong and r. langari. (1997). fuzzy modelling control of a nonlinear magnetic bearing system. proc. of the 1997 ieee international conference on control application. issn: 2180-1053 vol. 2 no. 1 january-june 2010 effect of pvd process parameters on the tialn coating roughness 41 effect of pvd process parameters on the tialn coating roughness a.r md nizam1, p. swanson2, m. mohd razali1, b. esmar1, h. abdul hakim3 1faculty of manufacturing engineering, universiti teknikal malaysia melaka, locked bag 1752, pejabat pos durian tunggal, 76109 durian tunggal, melaka. 2faculty of engineering and computing, coventry university, priory street, coventry cv1 5fb. united kingdom. 3advanced materials research center, sirim berhad, no 34, jalan hi tech ¾, kulim high tech park 09000 kulim kedah abstract coating on cutting tools has been proven to improve tool life significantly. physical vapour deposition sputtering process is one on the main techniques to deposit coating on cutting tool. one of the coating characteristics that influence its performance is coating surface roughness. extensive research has been carried out to understand the effect of process parameters on the resulting coating roughness. however holistic study to understand the combination of parameters and their interactions are lacking. the objective of this study is to evaluate the effect of coating process parameters (substrate bias voltage, substrate temperature, and sputtering power) on the deposited tialn coating roughness using response surface methodology. coating roughness characterization was done using atomic force microscopy apparatus. aside from that coating microstructure was also investigated using xrd and sem. finding from this research suggested that sputtering power, interaction between sputtering power and substrate temperature, and substrate bias quadratic term significantly influence the deposited coating surface roughness. keywords: sputtering, tialn, surface roughness, pvd, interaction, rsm 1.0 introduction the application of thin film coating on cutting tools can significantly improve cutting tool performance by enhancing the surface properties of the tool. the improved performance of coated cutting tools has been proven and documented (k. laing et.al., 1999) (o. gekonde haron et.al., issn: 2180-1053 vol. 2 no. 1 january-june 2010 journal of mechanical engineering and technology 42 2002) (g. byrne et. al., 1993) (k. tuffy et. al., 2004). one particular study done by tuffy et. al. indicated that coated tool wear performance was forty times better than the uncoated tools (k. tuffy et. al., 2004). aside form prolonging tool life, coated tools can also enable the implementation of minimum quantity lubrication (mql) and pursuant of dry machining. this can drastically reduce manufacturing cost associated with cutting fluids, which attribute to about 15% of metal cutting manufacturing costs, and minimize environmental impact associated with disposal of cutting fluid (g. byrne et. al., 1993). one of the coating characteristics that have significant influence on cutting tool performance is surface roughness of the developed coating. surface roughness can influence the friction level and material pick-up behaviour of cutting tool upon sliding with other material (b. podgornik et. al., 2004). two main techniques in depositing coating on cutting tool are physical vapour deposition (pvd) and chemical vapour deposition (cvd). the fundamental difference between the two processes is the vapour source. as the name indicates, the vapour source for pvd originates from a solid target from which atoms are displaced and vapour source for cvd originated form a chemical vapour precursor. in pvd process, the vaporization of the solid target may be done through heating or sputtering; this work focuses on the pvd sputtering process only. the pvd sputtering process involves the ejection of particles from target material due to the collision of highly energetic projectile particles (e.g. argon ions) with the target surface (bunshah et al., 1994). coating process optimization requires good understanding of the factors that influence coating characteristics. pvd process parameters have a significant influence on the resultant coating characteristics including the coating roughness (musil et al., 1998), (smith et al., 1995), (mayrhofer et al., 2006). some of the experimental work done on pvd process indicated that substrate bias voltage, sputter power, and substrate temperature could have significant effects the deposited coating roughness (barshilia et al., 2004), (xu et al., 2006). the objective of this study is to investigate the effect of substrate bias voltage, substrate temperature and sputtering power on the deposited coating roughness using response surface methodology (rsm). this holistic approach in assessing the behaviour of the pvd process is lacking in the previous studies. issn: 2180-1053 vol. 2 no. 1 january-june 2010 effect of pvd process parameters on the tialn coating roughness 43 2.0 experiment the tialn coating was deposited onto the tungsten carbide (wc) cutting tool insert using an unbalanced magnetron sputtering system made by vactec korea, model vtc pvd 1000. the target was ti-al alloy targets (50 % ti: 50 % al). prior to coating, the substrates were cleaned using an ultrasonic cleaner with alcohol bath for 20 minutes. the coating process throughout the experiment consisted of three stages; in-situ substrate etching for 30 minutes, interlayer coating (tial of approximately 0.2 micron), and coating deposition (tialn). the base pressure before the initiation of coating process was set at 5.0 x 10-5 mbar. the deposition of tialn was done in the ar and n2 partial pressure environment of 4.0x10-3 mbar and 0.4 x 10-3 mbar respectively for 90 minutes. the experimental matrix and data analysis were based on the on rsm centre cubic design, using design expert version 7.0.3 software. it consisted of 8 factorial points, 4 axial points and 6 central points to enable an estimation of process variability as illustrated by figure 1. the experimental matrix was designed based on assigning the extreme points (operating window) as the +/alpha value, refer to table 1. based on the defined extreme point values, the software then assigned the high and low settings for the factorial points. this was to ensure the characterization could be performed covering the widest range of operating window possible for respective parameters. because of this the value of factorial points were not nicely rounded. the developed experimental matrix based on the rsm central composite design and the +/alpha values defined in table 1, are as shown in table 2. figure 1 rsm central composite design for 3 factors at two levels issn: 2180-1053 vol. 2 no. 1 january-june 2010 journal of mechanical engineering and technology 44 table 1 extreme operating window for respective process parameters 2.1 coating procedures coating roughness and microstructure analysis technique were performed following these procedures: 2.1.1 atomic force microscopy (afm) the roughness of the developed tialn thin film coatings were analysed using shimadzu spm-9500j2 afm apparatus. the detection mode used was contact mode using a commercial si3n4 cantilever and the scanning area was set at 5x5 microns (25 μm2). 2.1.2 x-ray diffraction (xrd) the xrd analyses were performed using bruker d-8 xrd apparatus. due to the thin film sample, a grazing incidence angle (gia) feature was utilized with a grazing angle of 1 degree. the analysis was done using cukα radiation with λ = 0.15406 nm with ni filter, operated at 40 kv and 40 ma. the 2θ scanning range was set between 30 to 60 degrees with a step size of 0.020 degree and a dwell time of 1 second. the 2θ scanning range was selected to capture two main peaks appeared for the developed coating, tialn (111) and (200). the identification of tialn (111) and (200) peaks are based on standard jcpds no: 37-1140; the peaks at 37.7° and 43.8° correspond to diffraction along 111 and 200 planes respectively. the quantitative data extracted from the xrd analysis are the i(111)/ i(200) and the grain size. the grain size (dp) data was collected on dominant xrd peak of either (111) or (200) using scherrer’s equation dp = 0.9 λ/ β2θcosθ (culity et al., 1972); where λ is the wavelength of the x-ray, θ is the bragg’s angle and β2θ is the fwhm of 111 or 200 peak of the xrd pattern. 2.1.3 scanning electron microscopy (sem) the analyses were performed using sem/edx leo-1525. sem image captured the cross section view of fractured coating deposited on wc substrate. issn: 2180-1053 vol. 2 no. 1 january-june 2010 effect of pvd process parameters on the tialn coating roughness 45 3.0 results and discussions the twenty experimental run results are listed in table 2. the roughness data, in nanometres (nm), for the developed coating of each experimental run is tabulated in table 2. table 2 experimental run and results of coating roughness determination of significant factors influencing resultant coating roughness and the presence of interactions affecting the surface roughness were done by carrying out analysis of variance (anova) on the experimental data. the anova analysis is shown in table 3. based on the p-value of less than 0.1, sputtering power, interaction between sputtering power and substrate temperature, and substrate bias quadratic term are the significant influencing factors of the resultant surface roughness. issn: 2180-1053 vol. 2 no. 1 january-june 2010 journal of mechanical engineering and technology 46 table 3 anova for coating roughness model discussions on the influence of sputtering power, interaction between sputtering power and substrate temperature, and substrate bias quadratic term are as the following: 3.1 sputtering power: as the sputtering power increases from 4.81kw to 7.19kw, coating roughness reduced from 56.2 nm to 44.4 nm. (figure 2). this is aligned with findings by wuhrer and yeung who reported a decrease in roughness with the increase in sputtering power (wuhrer et al., 2004). figure 2 behaviour of coating roughness in response to variation of sputtering power the decrease in roughness as the sputter power increases can be explained by investigating the microstructure of the developed coating. the 2θ vs. intensity xrd curves for sputter power of 4kw, 6kw and 8kw are shown in figure 3. the quantitative data from the xrd analysis such as i111/i200 and grain size are tabulated in table 4. the xrd curves in figure 3 indicates the shift in dominant peak of crystal orientation form (111) to (200) as the sputter power increases from 4kw to 8 kw. this is also reflected by the peak intensity ratio of the two, i111/i200, in table 4. the i111/i200 ratio indicates that no significant peak of (200) can be found at the low sputter power of 4 kw. however the (200) peak become dominant at the sputter power of 8 kw level. this trend was also reported by xu et al. (xu et al., 2006) and wuhrer and yeung (wuhrer et al., 2004) based on their study of tin coating and tialn coating deposited using the same pvd sputtering technique. design-expert® software roughness x1 = a: sputter power actual factors b: bias voltage = 175.00 c: substrate temperature = 400.00 4.8 1 5.4 1 6.0 0 6.5 9 7.1 9 38.63 79 53.97 84 69.31 9 84.65 95 10 0 a: sputter power powervoltage r o u g h n e s s one factor warning! factor involved in an interaction. 100 figure 2 behaviour of coating roughness in response to variation of sputtering power issn: 2180-1053 vol. 2 no. 1 january-june 2010 effect of pvd process parameters on the tialn coating roughness 47 the decrease in roughness as the sputter power increases can be explained by investigating the microstructure of the developed coating. the 2θ vs. intensity xrd curves for sputter power of 4kw, 6kw and 8kw are shown in figure 3. the quantitative data from the xrd analysis such as i111/i200 and grain size are tabulated in table 4. the xrd curves in figure 3 indicates the shift in dominant peak of crystal orientation form (111) to (200) as the sputter power increases from 4kw to 8 kw. this is also reflected by the peak intensity ratio of the two, i111/i200, in table 4. the i111/i200 ratio indicates that no significant peak of (200) can be found at the low sputter power of 4 kw. however the (200) peak become dominant at the sputter power of 8 kw level. this trend was also reported by xu et al. (xu et al., 2006) and wuhrer and yeung (wuhrer et al., 2004) based on their study of tin coating and tialn coating deposited using the same pvd sputtering technique. figure 3 the 2θ vs intensity xrd curves for sputtering power of 4 kw, 6 kw and 8 kw. the grain size of the deposited tialn coating decreases as the sputter power increases as reflected in table 4. the decrease in grain size can be attributed to higher number and greater energy of the depositing atoms onto the substrate surface. this condition is more favourable for the nucleation of new grains than the growth of existing ones (wuhrer et al., 2004). this reduction in grain size corresponding to the reduction in surface roughness as reflected in table 4. table 4 the quantitative data from the xrd and edx analysis for tialn coating as the sputtering power increases. run sputtering power (kw) i111/i200 dp (nm) 18 4 * 24.59 19 6 3.65 12.31 15 8 0.78 8.75 * no i200 peak 3.2 substrate bias as reflected in figure 4, as the substrate bias increases from 100.67 to 175 v the coating roughness decreases from 66.4nm to 55nm and as the substrate bias increases from 175v to 249v, the coating roughness increases from 55nm to 67nm. this quadratic behaviour can also be inferred from the anova analysis in table 4 where the quadratic term of substrate bias is one of the significant terms that influence coating roughness. this finding is aligned with study by barshilia and rajam (barshilia et al., 2004) that indicated as the substrate bias increased from 0v to 200v, the developed coating roughness reduced significantly. the upward trend of coating roughness beyond certain substrate bias level, as indicated in this study, was also reported by cheng et al. (cheng et al., 0 10 20 30 40 50 60 32 37 42 47 52 57 2 teta in te ns ity run 18: ps 4 kw run 19: ps 6 kw run 15: ps 8 kw tialn (200) tialn (111) figure 3 the 2θ vs intensity xrd curves for sputtering power of 4 kw, 6 kw and 8 kw. the grain size of the deposited tialn coating decreases as the sputter power increases as reflected in table 4. the decrease in grain size can be attributed to higher number and greater energy of the depositing atoms onto the substrate surface. this condition is more favourable for the nucleation of new grains than the growth of existing ones (wuhrer et al., 2004). this reduction in grain size corresponding to the reduction in surface roughness as reflected in table 4. issn: 2180-1053 vol. 2 no. 1 january-june 2010 journal of mechanical engineering and technology 48 table 4 the quantitative data from the xrd and edx analysis for tialn coating as the sputtering power increases. figure 3 the 2θ vs intensity xrd curves for sputtering power of 4 kw, 6 kw and 8 kw. the grain size of the deposited tialn coating decreases as the sputter power increases as reflected in table 4. the decrease in grain size can be attributed to higher number and greater energy of the depositing atoms onto the substrate surface. this condition is more favourable for the nucleation of new grains than the growth of existing ones (wuhrer et al., 2004). this reduction in grain size corresponding to the reduction in surface roughness as reflected in table 4. table 4 the quantitative data from the xrd and edx analysis for tialn coating as the sputtering power increases. run sputtering power (kw) i111/i200 dp (nm) 18 4 * 24.59 19 6 3.65 12.31 15 8 0.78 8.75 * no i200 peak 3.2 substrate bias as reflected in figure 4, as the substrate bias increases from 100.67 to 175 v the coating roughness decreases from 66.4nm to 55nm and as the substrate bias increases from 175v to 249v, the coating roughness increases from 55nm to 67nm. this quadratic behaviour can also be inferred from the anova analysis in table 4 where the quadratic term of substrate bias is one of the significant terms that influence coating roughness. this finding is aligned with study by barshilia and rajam (barshilia et al., 2004) that indicated as the substrate bias increased from 0v to 200v, the developed coating roughness reduced significantly. the upward trend of coating roughness beyond certain substrate bias level, as indicated in this study, was also reported by cheng et al. (cheng et al., 0 10 20 30 40 50 60 32 37 42 47 52 57 2 teta in te ns ity run 18: ps 4 kw run 19: ps 6 kw run 15: ps 8 kw tialn (200) tialn (111) 3.2 substrate bias as reflected in figure 4, as the substrate bias increases from 100.67 to 175 v the coating roughness decreases from 66.4nm to 55nm and as the substrate bias increases from 175v to 249v, the coating roughness increases from 55nm to 67nm. this quadratic behaviour can also be inferred from the anova analysis in table 4 where the quadratic term of substrate bias is one of the significant terms that influence coating roughness. this finding is aligned with study by barshilia and rajam (barshilia et al., 2004) that indicated as the substrate bias increased from 0v to 200v, the developed coating roughness reduced significantly. the upward trend of coating roughness beyond certain substrate bias level, as indicated in this study, was also reported by cheng et al. (cheng et al., 2002). this could be due to imperfection of coating surface caused by bombardment of ions with excessively high energy level above certain substrate bias voltage (hultman et al., 1987).2002). this could be due to imperfection of coating surface caused by bombardment of ions with excessively high energy level above certain substrate bias voltage (hultman et al., 1987). figure 4 behavior of coating roughness in response to variation of substrate bias voltage. figure 5 2θ vs. intensity curves for the xrd analysis for substrate bias voltage of 50v, 175v and 300v design-expert® software roughness 5x5 x1 = b: bias voltage actual factors a: sputter voltage = 6.00 c: substrate temperature = 400.00 100.67 137.84 175.00 212.16 249.33 40 55 70 85 100 b: bias voltage r ou gh ne ss 5 x5 one factor warning! factor involved in an interaction. design-expert® software roughness 5x5 x1 = b: bias voltage actual factors a: sputter voltage = 6.00 c: substrate temperature = 400.00 100.67 137.84 175.00 212.16 249.33 40 55 70 85 100 b: bias voltage r ou gh ne ss 5 x5 one factor warning! factor involved in an interaction. design-expert® software roughness 5x5 x1 = b: bias voltage actual factors a: sputter voltage = 6.00 c: substrate temperature = 400.00 100.67 137.84 175.00 212.16 249.33 40 55 70 85 100 b: bias voltage r ou gh ne ss 5 x5 one factor warning! factor involved in an interaction. 0 10 20 30 40 50 60 70 32 37 42 47 52 57 2 teta in te ns ity tialn (200) tialn (111) run 1: vs 50 v run 19: vs 175v run 16: vs 300v figure 4 behavior of coating roughness in response to variation of substrate bias voltage. issn: 2180-1053 vol. 2 no. 1 january-june 2010 effect of pvd process parameters on the tialn coating roughness 49 2002). this could be due to imperfection of coating surface caused by bombardment of ions with excessively high energy level above certain substrate bias voltage (hultman et al., 1987). figure 4 behavior of coating roughness in response to variation of substrate bias voltage. figure 5 2θ vs. intensity curves for the xrd analysis for substrate bias voltage of 50v, 175v and 300v design-expert® software roughness 5x5 x1 = b: bias voltage actual factors a: sputter voltage = 6.00 c: substrate temperature = 400.00 100.67 137.84 175.00 212.16 249.33 40 55 70 85 100 b: bias voltage r ou gh ne ss 5 x5 one factor warning! factor involved in an interaction. design-expert® software roughness 5x5 x1 = b: bias voltage actual factors a: sputter voltage = 6.00 c: substrate temperature = 400.00 100.67 137.84 175.00 212.16 249.33 40 55 70 85 100 b: bias voltage r ou gh ne ss 5 x5 one factor warning! factor involved in an interaction. design-expert® software roughness 5x5 x1 = b: bias voltage actual factors a: sputter voltage = 6.00 c: substrate temperature = 400.00 100.67 137.84 175.00 212.16 249.33 40 55 70 85 100 b: bias voltage r ou gh ne ss 5 x5 one factor warning! factor involved in an interaction. 0 10 20 30 40 50 60 70 32 37 42 47 52 57 2 teta in te ns ity tialn (200) tialn (111) run 1: vs 50 v run 19: vs 175v run 16: vs 300v figure 5 2θ vs. intensity curves for the xrd analysis for substrate bias voltage of 50v, 175v and 300v the 2θ vs. intensity curves for the xrd analysis for substrate bias voltage of 50v, 175v and 300v are shown in the figure 5. quantitative data from the xrd analysis such as i111/i200 and grain size are tabulated in table 5. table 5 tialn coating characteristics and microstructure data as substrate bias voltage varies the intensity ratio data indicates that as the substrate bias increases from 50v to 175v, the i111/i200 increases significantly from 2.429 to 3.654 reflecting shift in crystal orientation from (200) plane towards (111) plane. subsequent incremental increases in substrate bias from 175 v to 300 v resulted in minimal changes in i111/i200 value. a similar trend in crystal orientation behavior under influence of substrate bias voltage variation was also reported in studies by matsue et al. (2004) and ahlgren and blomqvist (2005). significant grain size reduction was observed from 59.299 nm to 12.309 nm as the voltage increased from 50v to 175 v. further increase in substrate bias from 175v to 300 v resulted less significant reduction in grain size. the afm images shown in figure 6a and 6b provide visual evidence of the reduction of grain size and smoother surface morphology of tialn coating as issn: 2180-1053 vol. 2 no. 1 january-june 2010 journal of mechanical engineering and technology 50 the substrate bias voltage increases and the respective sem images of fractured cross section indicate a reduction in porosity and formation of a dense columnar structure at higher bias voltage. the reduction in grain size can be attributed to increases in ion bombardment as a result from substrate bias incremental changes. this is due to higher nucleation density resulting in fine-grained morphology (barshilia et al., 2004). the energy impacted upon the growing coating, due to ion bombardment, also helps to anneal out imperfections in the coating. however above certain ion bombardment energy level, the damaged induced by ion bombardment is more detrimental than the benefits (hultman et al., 1987). this is evidence in figure 6c as the substrate bias was further increased to 300v where imperfection of coating morphology occurred possibly due to resputtering of the deposited coating. figure 6 afm image (with imbedded sem image) indicating the transformation of grain size and morphology of tialn coating as the substrate bias increases 3.3 interaction between sputtering power and substrate temperature: the anova analysis also revealed that one of the significant factors influencing the coating roughness is the interaction between sputtering power and the substrate temperature. as shown in figure 7, at low substrate temperature level, changes in sputtering power does not significantly affect coating roughness. however at the high levels of substrate temperature, increases in sputtering power significantly reduce coating roughness. this indicates strong interaction exists between these two parameters that affect coating roughness. a) vs: 50 volts b) vs: 175 volts finer grains with faint boundary line imperfection possibly due to excessive ion bombardment c) vs: 300 volts figure 6 afm image (with imbedded sem image) indicating the transformatioof grain size and morphology of tialn coating as the substrate bias increases issn: 2180-1053 vol. 2 no. 1 january-june 2010 effect of pvd process parameters on the tialn coating roughness 51 3.3 interaction between sputtering power and substrate temperature: the anova analysis also revealed that one of the significant factors influencing the coating roughness is the interaction between sputtering power and the substrate temperature. as shown in figure 7, at low substrate temperature level, changes in sputtering power does not significantly affect coating roughness. however at the high levels of substrate temperature, increases in sputtering power significantly reduce coating roughness. this indicates strong interaction exists between these two parameters that affect coating roughness. figure 7 behaviour of coating roughness relative to interaction between sputtering power and substrate temperature. at high substrate temperate the change in sputtering power has an insignificant effect on the roughness of the tialn coating because it suppressed preferential crystal growth which resulted in smoother surfaces (lugscheider et al., 1996). the lack of preferential growth can be observed in figure 8 where the peaks of xrd curves for high temperature samples (run 2 and run 17) are much less pronounce compared to that of the lower temperature level (run 11 and run 6). the afm images in figure 9 shows the tialn coating morphology, where the coating with high substrate temperature level, run 2 and run 17, has a rounded and smoother surface compared to the coating of lower substrate temperature level, run 6 and run 11. design-expert® software roughness 5x5 c281.079 c+ 518.921 x1 = a: sputter power x2 = c: substrate temperature actual factor b: bias voltage = 175.00 c: substrate temperature 4.81 5.41 6.00 6.59 7.19 interaction a: sputter power r ou gh ne ss 5 x5 30 47.5 65 82.5 100 design-expert® software roughness 5x5 c281.079 c+ 518.921 x1 = a: sputter power x2 = c: substrate temperature actual factor b: bias voltage = 175.00 c: substrate temperature 4.81 5.41 6.00 6.59 7.19 interaction a: sputter power r ou gh ne ss 5 x5 30 47.5 65 82.5 100 design-expert® software roughness 5x5 c281.079 c+ 518.921 x1 = a: sputter power x2 = c: substrate temperature actual factor b: bias voltage = 175.00 c: substrate temperature 4.81 5.41 6.00 6.59 7.19 interaction a: sputter power r ou gh ne ss 5 x5 30 47.5 65 82.5 100 ts= 518 °c ts= 281 °c figure 7 behaviour of coating roughness relative to interaction between sputtering power and substrate temperature. at high substrate temperate the change in sputtering power has an insignificant effect on the roughness of the tialn coating because it suppressed preferential crystal growth which resulted in smoother surfaces (lugscheider et al., 1996). the lack of preferential growth can be observed in figure 8 where the peaks of xrd curves for high temperature samples (run 2 and run 17) are much less pronounce compared to that of the lower temperature level (run 11 and run 6). the afm images in figure 9 shows the tialn coating morphology, where the coating with high substrate temperature level, run 2 and run 17, has a rounded and smoother surface compared to the coating of lower substrate temperature level, run 6 and run 11. issn: 2180-1053 vol. 2 no. 1 january-june 2010 journal of mechanical engineering and technology 52 figure 8 xrd curves to compare the effect of interaction between sputtering power and substrate temperature figure 9 interaction effect between substrate temperature and sputtering power on the tialn coating morphology. 0 10 20 30 40 50 60 70 80 30 35 40 45 50 55 60 in te ns ity ( cp s) 2 teta run 6: ts: 281c ps: 4 kw run 2: ts: 518c ps: 4.8kw run 11: ts: 281c ps: 7kw run 17: ts: 518c ps: 7kw run 6: ts: 281c ps: 4 kw : run 17: ts: 518c ps: 7kw run 2 : ts: 518c ps: 4.8kw run 11: ts: 281c ps: 7kw tialn (111) tialn(200) 2 theta figure 8 xrd curves to compare the effect of interaction between sputtering power and substrate temperature figure 8 xrd curves to compare the effect of interaction between sputtering power and substrate temperature figure 9 interaction effect between substrate temperature and sputtering power on the tialn coating morphology. 0 10 20 30 40 50 60 70 80 30 35 40 45 50 55 60 in te ns ity ( cp s) 2 teta run 6: ts: 281c ps: 4 kw run 2: ts: 518c ps: 4.8kw run 11: ts: 281c ps: 7kw run 17: ts: 518c ps: 7kw run 6: ts: 281c ps: 4 kw : run 17: ts: 518c ps: 7kw run 2 : ts: 518c ps: 4.8kw run 11: ts: 281c ps: 7kw tialn (111) tialn(200) 2 theta figure 9 interaction effect between substrate temperature and sputtering power on the tialn coating morphology. issn: 2180-1053 vol. 2 no. 1 january-june 2010 effect of pvd process parameters on the tialn coating roughness 53 4.0 conclusion tialn coatings were deposited using pvd sputtering process at different levels of substrate bias voltages, substrate temperatures, and sputtering powers following the experimental matrix developed based on rsm approach. findings from this study indicated that sputtering power, interaction between sputtering power and substrate temperature, and substrate bias quadratic term are the significant process parameters that influence the deposited tialn coating roughness. increase in sputtering power resulted in decrease in surface roughness due to finer grain size formation. the interaction between sputtering power and substrate temperature indicated that at lower substrate temperature level, the change in sputtering power resulted in insignificant change in coating roughness attributed to the suppressed preferential crystal growth which resulted in smoother surfaces. the substrate bias voltage influenced the coating roughness in a quadratic behaviour where increase in substrate bias voltage up to 175v resulted in lower roughness value; however increment beyond that value resulted in higher surface roughness. this was attributed to the resputtering phenomenon which could occur if the ion bombardment energy is excessively high. 5.0 references b. podgornik,, s. hogmark, o. sandberg. 2004. influence of surface roughness and coating type on the galling properties of coated forming tool steel. surface and coatings technology 184. pp 338–348. barshilia harish c. and rajam k. s. 2004. nanoindentation and atomic force microscopy measurements on reactively sputtered tin coatings. bull. mater. sci. 27. pp (1) 35–41. bunshah rointan f. 1994. handbook of deposition technologies for films and coatings. second edition. new jersey. noyes publication. byrne g. and scholta e. 1993. environmentally clean manufacturing processes/a strategic approach. ann. cirp 42. pp 471–474. cheng y. h., tay b. k. and lau s. p. 2002. substrate bias dependence of the structure and internal stress of tin films deposited by the filtered cathodic vacuum arc. j. vac. sci. technol. a 20. pp (4) 1327-1331. cullity b. d. 1972. elements of x-ray diffraction. ma: addison-wesley. issn: 2180-1053 vol. 2 no. 1 january-june 2010 journal of mechanical engineering and technology 54 gekonde haron o. and subramanian s. v. (2002) ‘tribology of tool–chip interface and tool wear mechanisms’ surface and coatings technology 149. pp (2-3) 151-160. hultman l., helmersson u., barnett s. a., sundgren j. e. and greene j. e. 1987. low-energy ion irradiation during film growth for reducing defect densities in epitaxial tin(100) films deposited by reactive-magnetron sputtering. j. appl. phys. 61. pp 552. laing k., hampshire j., teer d., and chester g.1999. the effect of ion current density on the adhesion and structure of coatings deposited by magnetron sputter ion plating. surf. coat. technol. 112. pp 177–180. lugscheider e. , barimani c., wolff c., guerreiro s., doepper g. 1996. comparison of the structure of pvd-thin films deposited with different deposition energies. surface and coatings technology 86-87. pp 177-183. mayrhofer paul h., mitterer christian, hultman lars, clemens helmut. 2006. microstructural design of hard coatings. progress in materials science 51. pp 1032-1114. musil j and vlcek j. 1998. magnetron sputtering of films with controlled texture and grain size. materials chemistry and physics 54. pp 116-122. smith d.l. 1995. thin film deposition: principle & practice. new york: mcgraw hill. tuffy k., byrne g. and dowling d. 2004. determination of the optimum tin coating thickness on wc inserts for machining carbon steels. journal of materials processing technology 155-156. pp 1861-1866. wuhrer r. and yeung w. y. 2004. grain refinement with increasing magnetron discharge power in sputter deposition of nanostructured titanium aluminium nitride coatings. scripta materialia 50,pp. 813–818. xu xuan-qian, ye hui, zou tong. 2006. characterization of dc magnetron sputtering deposited thin films of tin for sbn/mgo/tin/si structural waveguide. j zhejiang univ science a 7. pp (3) 472-476. preparation of papers in a two column model paper format journal of mechanical engineering and technology corresponding author. email: fatimah@utem.edu.my issn 2180-1053 vol. 11 no. 1 july-december 2019 46 experimental and cfd studies of a thermoacoustic apparatus w. t. lee1, f. a. z. mohd saat1,2* 1)fakulti kejuruteraan mekanikal, universiti teknikal malaysia melaka, hang tuah jaya, 76100 durian tunggal, melaka, malaysia. 2)centre for advanced research on energy, universiti teknikal malaysia melaka, hang tuah jaya, 76100 durian tunggal, melaka, malaysia. abstract this paper reports a study of a thermoacoustic system by both experimentation and simulation works. a small scale thermoacoustic prototype was built for experimentation purposes. the prototype was built based on a thermoacoustic cooler setup. in addition, a two-dimensional transient thermoacoustic model was solved by using computational fluid dynamics (cfd) software. in experimentation, a small thermoacoustic rig operated at a frequency of 133.45 hz was built and the system was tested using two different types of stack; acrylonitrile butadiene styrene (abs) and a stainless-steel scrubber. at atmospheric pressure and at a relatively low frequency, a small temperature drop of approximately 1°c was recorded. similar result (with an error of 0.1%) was obtained using a simple cfd model that was designed based on the actual operating parameters of the experimental rig with abs as a stack. the study shows that thermoacoustic principles can be achieved using simple experimental as well as numerical setups. future works are focussing on optimising the apparatus so that better performance can be achieved. in optimised condition, thermoacoustic principle can create a sustainable and green alternative technology for a refrigerator or an engine. keywords: thermoacoustic; computational fluid dynamics (cfd); temperature difference; temperature drop 1.0 introduction thermoacoustic is a study that focuses on the interaction between thermodynamic and acoustic which are about the energy conversion from sound kinetic to heat (thermoacoustic refrigerator) or heat to sound kinetic (thermoacoustic engine) (ibrahim et al., 2011). figure 1 shows a schematic diagram of a thermoacoustic cycle on how the air particles in a thermoacoustic stack perform heat transfer (dyatmika et al., 2015). figure 1: schematic diagram of thermoacoustic cycle (dyatmika et al., 2015) journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 1 july-december 2019 47 the objectives of this research are to design a small scale thermoacoustic prototype for experimentation, test the small scale thermoacoustic system using suitable measurement method, and model a simple thermoacoustic flow field by using computational fluid dynamics (cfd) software. this research focuses only on the temperature at both ends of a structure known as ‘stack’ at certain pressure and velocity conditions inside a thermoacoustic refrigeration laboratory setup. some formulas and terminologies should be understood to investigate the flow in a thermoacoustic system. generally, there are few terminologies and formulas that are important in thermoacoustic system such as wavelength of sound, λ in equation (1), thermal penetration depth, δk in equation (2), viscous penetration depth, δv in equation (3) while the square of the viscous penetration depth divided by thermal penetration depth is equivalent to prandtl number, σ as shown in equation (4) (swift, 2001). λ = a / f (1) δk = = (2) δv = = (3) ( )2 = = σ ≤ 1 (4) where a is speed of sound, f is oscillation of frequency, k is thermal conductivity, ω is angular frequency which is equivalent to 2πf, ρ is density of gas, cp is specific heat per unit mass at constant pressure, κ is diffusivity of the gas, µ is gas dynamic viscosity, and υ is gas kinematic viscosity. equation (1-4) are used to determine the fundamental parameters for thermoacoustic refrigerator. the inlet pressure and mass flux equations are as shown in equation (5) and equation (6) (mohd saat & jaworski., 2013). equation (5) and equation (6) are interpreted in the computational fluid dynamics (cfd) for the inlet pressure and mass flux. they are also used to find the theoretical velocity of the flow and to validate the simulation result. p1 = pa cos(j x) cos(2πft) (5) m1 = (pa / c) sin(j x) cos(2πft + θ) (6) where p1 is the inlet pressure, pa is the antinode pressure, j is the wave number, x is the length of the resonator, f is the frequency of flowing air, t is the time taken, θ is the phase difference, and m1 is the mass flux. since the acrylonitrile butadiene styrene (abs) stack is usually made by cubepro 3d printer, it is necessary to understand its printing setup in order to obtain a good quality abs stack. it is found that the shorter the layer thickness, the greater the interlayer bonding but product printed by 200 μm layer thickness produces reasonable mechanical response because the strength of long flat plateau is good (rosli et al, 2017). journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 1 july-december 2019 49 figure 2 shows the build setting of a cubepro software. figure 2: build settings of cubepro software 2.0 methodology this research works consist of both experimentation and simulation analyses and the methodology for experimentation and simulation are discussed in subsection 2.1 and subsection 2.2 respectively. 2.1 experimentation to study the flow inside thermoacoustic system, an simple experimental rig of thermoacoustic refrigerator is built. figure 3 shows a schematic diagram thermoacoustic refrigerator while figure 4 shows the experimental rig of thermoacoustic refrigerator that is built in the laboratory. figure 3: schematic diagram of thermoacoustic setup the thermoacoustic refrigerator apparatus consists of resonator, sound driver, driver box, resonator, stack, function generator, and amplifier. the speaker box is made of 30cm x 30 cm x 10 cm customized acrylic box with a 3.1 cm center hole on top surface (30 cm x 30 cm) and 20 cm x 20 cm lockable door at the bottom surface (30 cm x journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 1 july-december 2019 49 30 cm). the resonator is made from acrylic tube with 3.5 cm outer diameter, 3.1cm inner diameter, and 0.65 m length. the acrylic material has thermal conductivity, k in range of 0.187 w m-1 k-1 to 0.209 w m-1 k-1 and specific heat, cp in range of 1.46 j g-1 °c-1 to 2.16 j g-1 °c-1. figure 4: experimental thermoacoustic laboratory rig by considering the portability factor, it is considered as too long and therefore, it needs to be cut down to a shorter length of 0.65 m (quarter wavelength). this corresponds to an operation parameter of 133.46 hz of flow frequency. the sound driver used in the thermoacoustic refrigerator is a used car loudspeaker of hyundai i10 2008. the loudspeaker’s maximum power is 300 w. the cross-section of the 3d printed acrylonitrile butadiene styrene (abs) stack is initially drawn by using solidworks software and saved in stereolithography (stl) format so that cubepro 3d printer software can read the drawing. the cross-section of the abs stack is as shown in figure 5. its area is 5.04 cm2 which gives a porosity of 0.3326. the acoustic wave number is calculated by formula, j = 2π / λ and for a resonator with a 0.65 m long (quarter wavelength), the acoustic wave number, j is 2.42 m-1. figure 5: cross-section of 3d printed abs stack the amplifier used in this thermoacoustic refrigerator apparatus is flepcher flpmt1201 while the function generator used in the thermoacoustic refrigerator apparatus is mcp sg1005. they are fabricated and assembled by using appropriate tools such as table saw, silicone, screws, nuts, et cetera. the instruments used to journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 1 july-december 2019 51 measure the temperature and velocity of the thermoacoustic standing wave inside the resonator are the picolog data logger tc 08 with association with picolog 6 software and sentry st-730 hot wire as shown in figure 6 (a) and (b). the thermocouple used is type k thermocouples as shown in figure 3.6 (c). since the hot wire is more sensitive and accurate, the hot wire is used to calibrate the thermocouple readings. (a) (b) (c) figure 6: (a) picolog data logger tc 08, (b) sentry st-730 hot wire and (c) type k thermocouples thermoacoustic refrigerator as shown earlier in figure 4 was used for implementing an experiment. there are two types of stack; 3d-printed abs stack and stainless-steel scrubber stack and they are used in the experiment to obtain different temperature reading for comparison purpose. the operating parameters, gas parameters and stack parameters for thermoacoustic refrigerator’s parameters are summarized and listed in table 1 for three-dimensional (3d) acrylonitrile butadiene styrene (abs) printed stack and table 2 for stainless-steel scrubber stack. the calculation is based on the formulas in section 1. most of the operating parameters are according the optimized characteristics. however, due to limitation of cubepro 3d printer, the plate thickness was set to 1.2 mm. the abs and stainless-steel scrubber materials are supplied by manufacturer without any specification details. their thermal properties in terms of specific heat, cp and thermal conductivity, k is therefore assumed. journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 1 july-december 2019 52 table 1: thermoacoustic refrigerator’s parameter (3d-printed abs stack) operating parameters gas parameters quarter wavelength, λ / 4 0.65 m speed of sound, a 347 m s-1 wavelength, λ 2.6 m thermal conductivity, k 0.02566 w m-1 k1 acoustic wave number, j 2.42 m-1 thermal penetration depth, δk 0.233 mm frequency, f 133.45 hz specific heat, cp 1007 j kg -1 k-1 mean pressure, pm 101 kpa mean density, ρm 1.176 kg m -3 mean temperature, tm 300 k gas displacement, ξ 0.47 m stack parameters (abs) stack length, ls 0.04 m stack center position, xs 0.43 cm thermal conductivity, k 0.195 w m-1 k-1 specific heat, cp 1420 j kg -1 k-1 spacing, 2yo 0.75 mm plate thickness 1.2 mm stack diameter, di 3.1 cm porosity, b 0.3326 table 2: thermoacoustic refrigerator’s parameter (stainless steel scrubber stack) operating parameters gas parameters quarter wavelength, λ / 4 0.65 m speed of sound, a 347 m s-1 wavelength, λ 2.6 m thermal conductivity, k 0.02566 w m-1 k1 acoustic wave number, j 2.42 m-1 thermal penetration depth, δk 0.233 mm frequency, f 133.45 hz specific heat, cp 1007 j kg -1 k-1 mean pressure, pm 101 kpa mean density, ρm 1.176 kg m -3 mean temperature, tm 300 k gas displacement, ξ 0.47 m stack parameters (abs) stack length, ls 0.12 m stack center position, xs 0.53 mm thermal conductivity, k 15 w m-1 k-1 specific heat, cp 500 j kg -1 k-1 stack diameter, di 3.1 cm porosity, b 0.748 2.2 computational fluid dynamics (cfd) simulation ansys computational fluid dynamics (cfd) software is used for two dimensional simulation of thermoacoustic refrigerator with 3d printed abs stack. journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 1 july-december 2019 53 figure 7 shows the geometry drawing of the cfd model. a “multizone quad/tri” mesh method is applied for mesh. figure 7: 2d geometry drawing in ansys the model used pressure-based solver, absolute velocity formulation, transient time, and planar two-dimensional (2d) space with gravitational acceleration acted on negative y direction, g = -9.81 m s-2 and floating operating pressure of 101, 325 pa. in the models’ material setup, there are two materials defined which are air as a working medium and acrylonitrile butadiene styrene (abs) as the stack’s material. the specific heat, cp and thermal conductivity, k of the air were set using polynomial function as shown in equations (7) and (8), respectively. the correlation coefficient, r2 of equation (7) and (8) are 0.99961 and 0.99993 respectively. cp = 1.9327x10 -10t4–7.9999x10-7t3+1.1407x10-3t2–4.4890x10-1t+1.0575x103 (7) k = 1.5207x10-11t3–4.8574x10-8t2+1.0184x10-4t–3.9333x10-4 (8) the viscosity of air was set using power-law formula. the density, ρ, specific heat, cp, and thermal conductivity, k of abs are 1215 kg m -3, 1.42 kj kg-1 k-1, and 0.195 w m-1 k-1 respectively. the air was set as an ideal gas for this case and t is the temperature of air. the mass flux is at 0.588 kg m-2 amplitude with 90o phase difference to the pressure and the flow frequency is 133.45 hz. the velocity of the air is measured by the sentry hot wire anemometer and the value was recorded as 0.5 m s-1 and this value is used to calculate the mass flux amplitude. in the solution methods configuration, pressure-implicit with splitting of operators (piso) scheme was used for dealing with large number of time steps as it would run up to more than 1,000 time steps for each case while others are following the default configurations. calculation is run at 7.5 x 10-6 s of time step size where 1,000 number of time steps is needed to complete one wave cycle for a flow frequency of 133.45 hz and a flow period of 7.5 ms. 3.0 results and discussions the results are discussed thorugh analysis related to the temperature drop obtained through the thermoacoustic setup. in addition, the numerical model is used to produce the pressure and velocity distribution along the resonator of the thermoacoustic apparatus. these are presented in two subsections to separate the findings of the experimental results from the findings of the numerical results. 3.1 experimental results two different stacks were used for the experiment which are acrylonitrile butadiene styrene (abs) and stainless-steel scrubber and their respective operating parameters were as discussed earlier using tables 1 and 2. figure 8 shows graph for journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 1 july-december 2019 54 temperature at two end points of three-dimension (3d) printed abs stack against time. temperature at one end of the stack is presented by blue line while the temperature at another end is given by the red line. temperature drop is observed at one end of the stack (red-line) after approximately 4 minutes of operation. the temperature drop between the two ends is sustained for about 25 minutes before the temperature different becomes smaller again. this probably happen because of the absence of heat exchanger to help the system in sustaining the obtained temperature difference. figure 8: graph of temperature at two end points of 3d-printed abs stack against time figure 9 shows a graph of temperature at two ends of stack against time when the stainless-steel scrubber is used as a stack within the same resonator. the stack is now a 12 cm long stainless-steel scrubber as compared to the earlier used of a 4 cm long 3d-printed abs stack with a stack center position, xs = 0.53 cm. when abs stack is used, the highest temperature gradient recorded is 0.9 ºc which was measured using type-k thermocouple that is connected to picolog tc 08 signal conditioner. the highest temperature gradient recorded using scrubber as stack is 0.6 ºc. the curve of the graph against temperature in stainless steel scrubber stack is more stable compared to the 3d-printed abs stack because the flow is steadier due to higher porosity. it is shown that the thermoacoustic effect of stainless-steel scrubber stack is more stable than the abs stack as the graph of temperature against time graph in stainless-steel is more likely uniform than abs. however, the temperature difference created across the stack is not high enough because optimization of parameters was not done before the experiment is carried out. nevertheless, the results show that temperature different was successfully achieved between the two ends of the stack inside the resonator. the temperature different is expected to be bigger if the rig is built at optimum condition. journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 1 july-december 2019 55 figure 9: graph of temperature at two end points of stainless-steel scrubber stack against time 3.2 computational fluid dynamics (cfd) results velocity changed with time at the center of the computational domain for different mesh density are as shown in figure 10. the low mesh density is the mesh with low mesh element while the high mesh density is the mesh with high mesh element which has doubled number of divisions of all edge sizing. after the mesh independency is checked, the simulation is proceeded to next phase, which is verification of the model. the verification method is to obtain the velocity profile at the centre of the model and compare the simulated velocity profile with the theoretical velocity profile. the solver setting is also described earlier as can be found in subsection 3.2. since the amplitude of velocity at the inlet is experimentally measured as 0.5 m s-1, the antinode pressure, pa can be obtained from calculation and it is found to be 204.04 pa by using the equations equation (5) and equation (6). therefore, the drive ratio is 0.2% for the resonator apparatus that functioned at a frequency of 133.45 hz and a length of the 0.65 m of the quarter wavelength. figure 10: velocity at the centre of the model for different mesh densities journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 1 july-december 2019 56 figure 11 shows the x-velocity comparison between the simulated and calculated amplitude results at a point located in the centre of the model. figure 11: comparison of x-velocity between the theoretical and simulation results table 3 shows the percentage of error by comparing the theoretical calculated result and the simulated result. table 3: percentage of error of amplitude amplitude theoretical value (m s-1) simulation value (m s-1) percentage of error (%) maximum 0.324752 0.35372 8.15 minimum -0.40379 -0.35357 14.20 a graph of pressure amplitude against axial direction is plotted as shown in figure 12. the maximum temperature at the antinode is 218.88 pa and if it is compared to the theoretical calculation, the value from numerical model differs from the theoretical prediction by 7.27%. based on the numerical results, it can be concluded that the air particles are vibrating in a range of 0 pa to 218.88 pa of pressure in the resonator environment. journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 1 july-december 2019 57 figure 12: graph of pressure along x direction with inlet wall located at the right end. the temperature across axial direction of the resonator is also discussed. the temperature data is captured at the time step of 8,524. figure 13 and table 4 show the graph of temperature against x direction, and average temperature at both end points of abs stack respectively. with reference to figure 13, the environment temperature for the fluid is 300 k. as the acoustic wave passes through the stack, there is a visible temperature increment at the other end of the stack. the effect of temperature drops is not strong due to parameters which was not optimized, the unsteady state condition of the flow, and the low drive ratio. along the x direction, there is a temperature increment from inlet boundary to wall boundary from 299.01 k to 300.07 k. journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 1 july-december 2019 58 figure 13: graph of temperature along x direction with inlet wall located at the right end. two points named as point 1 and point 2 are created in the cartesian coordinate of (-0.02, 0.000975) and (-0.06, 0.000975) in meter. the points are at the two points of stack ends near the gap. although the temperature differences recorded is 1 k and the acoustic effect is not strong, but this shows that the principle of thermoacoustic is feasible as there is temperature different observed at the two ends of the stack and similar observation was also gained from experimental result as reported earlier in subsection 3.1 where a temperature drop of approximately 1 oc was recorded. table 4: average temperature at both end points of abs stack probe average temperature (k) point 1 (-0.02, 0.000975) 301 point 2 (-0.06, 0.000975) 302 figure 14 shows the graph of the velocity against x direction. the wave particles at each of the points along the x-direction vibrate in sinusoidal patterns as was shown earlier in figure 11. however, as particles move from inlet boundary to the wall boundary at the hard end of the resonator, the amplitude of velocity of the wave particles decreases as it gets closer to the hard end of the resonator following the journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 1 july-december 2019 59 standing wave criterion. figure 14: graph of velocity along x direction with inlet wall located at the right end. hence, it is observable that the velocity of the air particles along x direction is decreasing from inlet boundary (maximum x) and hit zero at the hard end (minimum x) of the resonator while the air particles in each point along the x direction before reaching the hard end vibrate at their amplitude along their orientation line. 4.0 conclusionh an experimental test rig was built and a cfd model was solved for a low frequency range thermoacoustic environment. the temperature difference across the both ends of the stack was discussed by comparing the temperature drops at the stack ends. from experimental result, it is shown that a small temperature difference of 0.9 oc were recorded for the abs stack while the cfd simulation predicts a temperature difference of 1 oc. the low temperature difference was obtained due to non-optimised conditions of the thermoacoustic environment. amongst the parameters that need to be optimised are the resonance frequency, stack’s dimension, stack’s location (swift, 2001), (zolpakar, 2016), (ibrahim et al., 2011). a bigger temperature difference is expected should the rig be operated at optimised condition and at higher operating conditions (swift, 2001). nevertheless, the study shows that a preliminary stage of thermoacoustic rig and cfd models were succesfully developed and further investigation should be carried out to increase the performance to an acceptable level. journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 1 july-december 2019 60 5.0 acknowledgement this work has been carried out using facilities at universiti teknikal malaysia melaka. 6.0 references dyatmika, h. s., achmadin, w. n., murti, p., setiawan, i. & utomo, a. b. s. (2015). development of the thermoacoustic refrigerator system using a stack made of some stainless steel mesh and a hot heat exchanger. indonesian journal of physics, 26(01). 5 8. ibrahim, a., arafa, n. & khalil, e., (2011). geometrical optimization of thermoacoustic heat engines. in: proceedings of the forty-nineth aiaa aerospace sciences meeting including the new horizons forum and aerospace exposition. mohd saat, f. a. z. & jaworski, a. j. (2013). oscillatory flow and heat transfer within parallel-plate heat exchangers of thermoacoustic systems. proceeding of the world congresson engineering iii. rosli, n. a., hasan, r., alkahari, m. r. & tokoroyama, t. (2017, november 17). effect of process parameters on the geometrical quality of abs polymer lattice structure. proceedings of sakura symposium on mechanical science and engineering 2017, pp. 3-5. swift, g. (2001). theromoacoustics: a unifying perspective for some engines and refrigerators. zhurnal eksperimental’noi i teoreticheskoi fiziki (fifth draf). los alamos national laboratory. zolpakar, n. a., mohd-ghazali, n. & hassan, e. f. m. (2016). performance analysis of the standing wave thermoacoustic refrigerator: a review. renewable and sustainable energy reviews. elsevier, 54, pp. 626-634. doi: 10.1016/j.rser.2015.10.0 journal of mechanical engineering and technology *corresponding author e-mail: o.david-west@herts.ac.uk issn 2180-1053 vol.12 no.1 1 june – december 2020 structural dynamic modelling of a multi-storey shear frame using mass and stiffness addition opukuro s david-west centre for engineering research materials & structures university of hertfordshire, hatfield, al10 9ab, united kingdom. . abstract measurements of stress/strain properties are not sufficient for high-speed operation systems and lightweight structures, instead, dynamic measurement/analysis are necessary for a comprehensive understanding of their characteristics. a shear frame structure was modelled using solid elements (ansys solid 187) and the discrepancy between the experimental and initial numerical results were very high. the three experimental modes were observed, and the suspected areas of high local stiffness were noted; these being the areas of connection between the floor plates and vertical pillars and ansys shell 181 was used to adjust the stiffness locally. also, with appropriate engineering judgements, omitted masses compared with the physical structure were added locally using ansys mass 21 element type. in addition, the finite element model boundary conditions were carefully manipulated to predict the experiment condition. the process of model updating was good as the difference between the experimental and finite element results were reduced. keywords: model updating, finite element model, structural dynamics, ansys, vibration 1.0 introduction simulation of the behaviour of structure is now a very important tool in the present advanced technology, in the design and manufacturing sectors. nowadays finite element techniques are commonly used to understand the characteristics of engineering structures; however, these numerical models are usually analysed based on idealized engineering properties. the precise prediction is not usually achieved with the first simulation, so there exist some discrepancies between experimental and numerical results. mailto:o.david-west@herts.ac.uk journal of mechanical engineering and technology issn 2180-1053 vol.12 no.1 1 june – december 2020 the accuracy of the finite element result is often obtained by comparing with experimental result tested on the physical system. if the correlation between the two is poor, then the numerical model must be adjusted so that the agreement between prediction and test result could improve; hence model updating. parker, (2008) have observed that there are challenges in trying to identify the differences between the simulation and experimental results. the primary task in model updating is the choice of parameters to modify the mass, stiffness, and damping matrices in order to obtain better agreement between numerical and test results, and a survey of updating methods were given by mottershead & friswell, (1993). mottershead & friswell, (1993) mention that for complex structures at least the first third of the eigenvalues of the finite element result should be accurate enough for design purpose and have highlighted a couple of model updating methods. mares, et al (2003) use natural frequency errors and physical reasoning to update the finite element model of a garteur sm-ag19 test structure. the modeling uncertainties were concentrated at the joints and constrained visco-elastic layers. friswell, et al (2001), applied the methods of regularization, singular value decomposition, l-curves and crossvalidation to model updating. the parameters of the model were adjusted using residuals between a measurement set and the corresponding model predictions. nalitolela, et al (1993) have presented the idea to use imaginary stiffness addition and simple structural modification to perturb a model and predict the dynamic characteristics; while cha & de pillis, (2001) used experimental data and the inclusion of masses to conduct updating of an analytical model. bridges are indispensable components of the infrastructure of modern society. zapico, et al (2003) added to this debate by updating an experimental bridge model with a geometric scale of 1:50 representing a continuous-deck motorway bridge, using the technique of mass addition. brownjohn & xia, (2000), used the sensitivity-based model updating method for the dynamic assessment of a curved cable-stayed bridge and bien, et al (2002), highlighted the disadvantages of the conventional approach of exciting a bridge i.e. movement of large vehicles and reported that the use of an inertial vibration exciter gives a better result. ren & chen, (2010) adopted the response surface approach to update the dynamic model of engineering structure; an objective function created between the experimental and numerically obtained natural frequencies was implemented in the optimisation algorithm to get the updated model. the process was observed to be efficient with faster convergence, compared to the sensitivity-based model approach; while gordis & papagiannakis, (2011), mentioned that the sensitivity-based model error localization and damage detection is limited by the relative differences in sensitivity magnitude among updating parameters and that artificial boundary conditions may be used to reduce this limitation. shan, et al (2015) used the response surface characteristics and finite element analysis to obtain an updated model of a cable suspension bridge; the sensitivity parameters were extracted on the bases of variance analysis. khodaparast, et al (2008) addressed the issues of model adjustment of a test structure by perturbation technique on random variables of measured model response. zang, et al (2004) reported using a combination of independent component analysis extraction of time domain data and artificial neural networks to detect structural damage. an inverse problem requires the use of a model and the uncertain parameters, friswell, (2007) has highlighted the use of this approach for several issues such as health monitoring, including modeling error, environmental effects, damage localization and regularization. titurus, et al (2003) proposed the use of generic elements as a viable tool journal of mechanical engineering and technology issn 2180-1053 vol.12 no.1 1 june – december 2020 for damage detection and sinha & friswell, (2002) summarized the use of eigensensitivity approach in model updating and structural health monitoring. friswell, et al (1998) have discussed the use of damping and stiffness matrices in model updating by minimizes of the changes in the damping and stiffness matrices, with the objective that the measured data is reproduced. ahmadian, et al (1998) used the inverse approach to parameters for the element model within allowable mass and stiffness matrices in model updating. kozak, et al (2009) presented the model updating procedure based on minimization of an index called miscorrelation index, which are introduced at error locations in the finite element model. min, et al (2014) presented a sensitivity-based model updating technique for damped structures and the process involves data extracted from experiment and degree of freedom reduction in the finite element analysis. zapico-valle, et al (2010), introduced a minimization of an error function for finite element model updating carried out by an adaptive sampling algorithm. chouksey, et al (2014), updated the finite element model of a rotor shaft using the inverse eigensensitivity approach. khanmirza, et al (2011) presented two techniques for the identification of mass–damping–stiffness of shear buildings; in the first method the dynamic parameters were obtained from a forced vibration test and the second technique was an inverse analysis to identify the dynamic characteristics. lepoittevin & kress, (2011) simulated the free – free boundary conditions of bare and damped samples and predicted the resonance frequencies and modal loss factors from the numerical analysis. in this study experimental modal analysis was performed on a three storey shear frame using the shaker, accelerometer, charge amplifier and fast fourier analyser. the three expected natural frequencies were obtained, and mode shapes observed with a stroboscope. these results were compared with the ones obtained from finite element analysis using ansys 17 finite element code. the discrepancies between the experiment and numerical results were reduced by updating the numerical model. the mass and stiffness matrices of the finite element model were manipulated respectively by appropriate engineering judgement with a dot element (mass 21) and shell element (shell 181) as ghost or imaginary element, that is elements that have contributed to the final results but are not seen when the elements/mode shapes are displayed. in addition, the boundary conditions of the finite element model were carefully controlled to match that of the experiment. 2.0 governing equations the dynamic properties of a system can be represented if the basic properties are assumed to be discretised and considered separately [ maia & silva, 1997]. the spatial distribution of mass, stiffness and damping properties are illustrated in terms of matrixes of mass[𝑀], stiffness [𝐾] and damping[𝐶] (for a viscously damped model) or [𝐷] (for hysterically damped model). if the degree of freedom (dof) is illustrated by the time-dependent displacement 𝑥𝑖 (𝑡) with a time-dependent applied force, 𝑓𝑖(𝑡). a general dynamic analysis will solve the equation of motion which gives the time dependent response of every node point in the structure by including inertial and damping forces in the equation. the model can be illustrated using the newton’s second law of motion and for hysterically damped model, [𝑀]{�̈�(𝑡)} + [𝐾]{𝑥(𝑡)} + 𝑖[𝐷]{𝑥(𝑡)} = {𝑓(𝑡)} (1) journal of mechanical engineering and technology issn 2180-1053 vol.12 no.1 1 june – december 2020 taking consideration of viscous damping, [𝑀]{�̈�(𝑡)} + [𝐶]{�̇�(𝑡)} + [𝐾]{𝑥(𝑡)} = {𝑓(𝑡)} (2) if the solution is as illustrated in equation (3), {𝑥(𝑡)} = {𝑋}𝑒𝑖𝜔𝑡 (3) where, {𝑋} is a time-dependent complex amplitudes (maia & silva, 1997), then spatial model can be presented as a generalized eigenvalue problem (maia & silva, 1997) as in equation (4). [[𝐾] − 𝜔2[𝑀] + 𝑖[𝐷]]{𝑋} = {0} (4) where, d is the structural damping matrix. most engineering systems designers are interested in the natural frequencies and mode shapes of vibration of the system. if the damping is ignored and considering a free vibration multi-degree of freedom system, the dynamic equation becomes. [𝑀]{�̈�} + [𝐾]{𝑥} = {0} (5) if the displacement vector has the form {𝑥} = {𝑋} 𝑆𝑖𝑛 𝜔𝑡 , then the acceleration vector is {�́�} = −{𝑋}𝜔2 𝑆𝑖𝑛 𝜔𝑡 and substituting into equation (5) gives the eigenvalue equation. ([𝐾] − 𝜔2[𝑀]){𝑋} = {0} (6) each eigenvalue has a corresponding eigenvector and the eigenvectors cannot be null vectors, hence, |[𝐾] − 𝜔2[𝑀]| = {0} (7) equation (6), represent an eigenvalue problem, where is the eigenvalue and the eigenvector (or the mode shape). the eigenvalue is the square of the natural frequency of the system. the structural damping matrix in ansys finite element code is analysed using the relationship, 𝐷 = 𝛼𝑠 ∗ 𝑀 + 𝛽𝑠 ∗ 𝐾 (8) where  and  are the damping proportionality constants corresponding to mass and stiffness matrix respectively. the subscript ‘s’ refers to structural damping. structural damping coefficient of 0.05 used for both 𝛼𝑠 and 𝛽𝑠 for purpose of finite model updating of the shear frame structure. }{x 2  }{x journal of mechanical engineering and technology issn 2180-1053 vol.12 no.1 1 june – december 2020 3.0 structural description the shear frame is made of steel. it consists of two metal pillars connected by three steel metal strips that are at three different levels. the connection from the pillars to the metallic strips was with metallic tabs screwed through the pillars into the metallic strips. the bases of the pillars were firmly bolted to the table foundation. figure 1: photograph of the shear frame structure. journal of mechanical engineering and technology issn 2180-1053 vol.12 no.1 1 june – december 2020 figure 2: sketch of the shear frame structure with all the dimensions in mm. the photograph of the shear frame is shown in figure (1) and the schematic of the shear frame with the dimension in millimetres presented in figure (2). this three-storey system is expected to present three modes of vibration because it possesses three degree of freedom, hence we may presume that the number of degrees of freedom is equal to the number of storeys on a building for this illustration. 4.0 experimental procedure and results the vibration response of the three-storey shear frame with bolted boundary conditions under the excitation of a shaker was measured with a magnetic accelerometer connected to a charge amplifier and the fourier analyser. journal of mechanical engineering and technology issn 2180-1053 vol.12 no.1 1 june – december 2020 figure 3: set-up of the experimental arrangement. the schematic of the test arrangement is as shown in figure (3). the input frequency of the shaker via the stinger and force transducer was set by the signal generator. the accelerometer with magnetic base attached to the top of the three-storey frame; transfers the measured data to be display on the screen of the spectrum analyser via the charge amplifier and the resonant frequencies determined. as vibration of the shear frame was varied through the input from the shaker, a stroboscope was used to confirm the nodes and anti-nodes of the oscillation along the sides of the frame. the photograph of the stroboscope is shown in figure (4). figure 4: stroboscope journal of mechanical engineering and technology issn 2180-1053 vol.12 no.1 1 june – december 2020 figure 5: experimental mode shapes figure 6: frequency response function – screen shut of oscilloscope fourier analyser. table 1: natural frequencies from test result mode 1 mode 2 mode 3 16.75 hz 50.00 hz 75.75 hz the three modes shape of the shear frame observed with the stroboscope are sketched and presented in figure (5) and frequency response function extracted with the accelerometer shown in figure (6). each peak in the frequency response plot correspond to the natural frequency of the mode’s shapes in the respective order and the values of the natural frequencies shown in table 1. journal of mechanical engineering and technology issn 2180-1053 vol.12 no.1 1 june – december 2020 5.0 initial finite element simulation the initial finite element model of the three-storey shear frame was created as a threedimensional linear elastic model using the mechanical properties of steel shown in table 2 in ansys 17 code. the element used for meshing was solid187; it is a higher order 3-d element, defined by 10 nodes and three translational (x, y and z) degree of freedom at each node. table 2: material properties of steel item density modulus poisson ratio steel 7830 kg/m3 210 g pa 0.33 the three storey steel shear frame structure was bolted to the base structure (a table) via the blocks as shown in the picture of figure 1 and this was represented in the finite element simulation as fixed condition for the nodes attached to the bottom lines i.e. by setting the degrees of freedom in the x, y and z directions to zero. the finite element model has 7461 elements, with the element edge length of 6 mm and figure (7) shows the finite element discretisation for the shear frame and the global coordinate system, which by default will corresponds to the element coordinate system. figure 7: mesh discretisation of the three-storey shear frame. journal of mechanical engineering and technology issn 2180-1053 vol.12 no.1 1 june – december 2020 table 3: natural frequencies obtained for initial finite element model mode number frequency (hz) 1 18.4 2 54.5 3 55.2 the realization of the elastic modes and natural frequencies were achieved with lanczos tool in ansys 17 finite element code to conduct the dynamic analysis. table 3 shows the natural frequencies obtained for the first three modes. the fundamental frequency was obtained as 18 hz and the other two seems to be very close modes in the vicinity of 55 hz. that is from this initial finite element results, the second and third eigenvalues were repeated and hence the corresponding mode shapes were not unique. 6.0 initial results and localisation of updating parameters the results as regards the natural frequencies obtained from experiment and finite element are presented in table 4. the initial finite element model seems to be relatively stiff with respect to the fundamental mode and the second one. also, the finite element second and third modes are seen to be closed from the results of the initial analysis. table 4: comparison of experimental and initial numerical results mode number frequency (hz) percentage change [%] |( 𝑓2 − 𝑓1 𝑓1 )| × 100 experiment, f1 finite element analysis, f2 1 16.75 18.4 9.9 2 50.00 54.5 9.0 3 75.75 55.2 27.1 in order to successfully match the simulation results with that of the experimental ones, it is important to fully consider all aspects of the experimental setup. this includes ensuring dimensional accuracy of the finite element model, and accounting for additional masses of components within the system that would cause the vibration characteristics to be altered. there also exist changes in the structural stiffness at various locations which was not taken into considerations. table 5 summarized the assumptions of the initial finite element model. journal of mechanical engineering and technology issn 2180-1053 vol.12 no.1 1 june – december 2020 table 5: initial model assumptions. type description geometric shape the parts such as the hook, taps and screws were not included in the geometric model. connections connections between parts were treated as one full member. boundary conditions based boundary was taken as fixed points nodes. modulus young’s modulus of steel material could vary between 190 – 215 g pa; used 210 gpa density density of steel material could vary between 7700 – 8000 kg/m3; used 7830 kg/m3 thickness assumed uniform thickness along the cross-section of each part. stiffness variations at different locations due to connections of parts not taken into account. it can therefore be concluded that there was a problem with inadequate modelling which overestimated the stiffness as is evident from the results of the first and second natural frequencies obtained from the initial finite element model. the variation of the masses such as addition of the metallic tabs / hook with the use of dot elements were not adequately taken into consideration. also, the changes in local stiffness due to the different levels of the shear frame will need to be considered in the process of updating. 7.0 model updating procedure the primary task in finite element model updating is the ability to apply engineering judgements to tune the model and satisfactorily predict the characteristics of the structure; hence it is very important to identify sufficient model updating parameters. these parameters are usually the physical variables of the model which are adjusted at the element level. as a preliminary adjustment, the finite element model was material properties were modified globally as in table 6 to improve the model. table 6: justification for mechanical properties selection type of check action justification young’s modulus used 200 g pa steel stated modulus between 190 – 215 g pa. density used material density of 7830 kg/m3. steel material density range between 7700 – 8000 kg/m3. journal of mechanical engineering and technology issn 2180-1053 vol.12 no.1 1 june – december 2020 the photograph of the metallic tab with hook and the screws connecting to the level plates (floors) of the shear frame is as shown in figure (8). the weight of the solid metallic tab was obtained using ‘mettler ae200’ digital weighing instrument and use for purpose of the model adjustment. the mass was added to the finite element model across 4 nodes in the regions where tabs / hooks are meant to be present. ansys mass 21 element was used to implement this. figure 8: shear frame hook at one-tab location the mass added per node was 0.003 kg for the four nodes the total is 0.012 kg which is approximately the mass of a solid tab. after adding the masses to the model and running a second modal analysis, it was found that the addition of the masses resulted in a negligible effect on the natural frequencies of the structure. due to this, it was determined that the stiffness of the model should be modified in order to alter the natural frequencies. 8.0 model adjustment – mass and stiffness addition to add stiffness to the current finite element model ‘shell 181’ element was used and applied with negligible element thickness in ansys 17 finite element code. this element type was selected as it well suited to both linear and non-linear strain applications; it has six degrees of freedom at each node: translations in the x, y, and z directions, and rotations about the x, y, and z axes. considering the initial modes shapes of the shear frame, it was observed that the stiffness around the joints of the vertical pillars and the level plates (floors), and the area near the fixed base are quite different from other parts of the structure. in structural finite element analysis, the stiffness matrices are formulated by equation (9), which is available in most literatures on finite element analysis. [𝑘]𝑒 = ∫[𝐵] 𝑇 [𝐸][𝐵]𝑑𝑣 (9) where, b is the strain – displacement transformation matrix and e is material property matrix. hence, as can be seen from this expression the material young’s modulus is an important parameter in the formation of the element stiffness matrices. local stiffness was added at the various levels shown in figure (9) with the ansys shell 181 element. journal of mechanical engineering and technology issn 2180-1053 vol.12 no.1 1 june – december 2020 figure 9: discretisation of model, mass addition with local stiffness adjustment. figure 10: discretisation of model, mass addition and coordinate system. level a level b level c level d journal of mechanical engineering and technology issn 2180-1053 vol.12 no.1 1 june – december 2020 in this study the lost mass in the initial model were added with the dot element of ansys 17 known at ‘mass 21’. that is mass of the metallic / hook obtained by weighing with a digital scale to be approximately 0.012 kg, was added using ansys mass 21 element as distributed mass (0.003 kg per node on four nodes) along the appropriate locations as shown in figure (10) and the local stiffness at certain locations of the shear frame model change using ansys ‘shell 181’ element type as presented figure (9). also in figure (10) is the global coordinate system for the updated finite element model and it also corresponds to the element coordinate system; as the vibration energy is transmitted through the shaker to the shear frame in the z-direction, the x-direction for the finite element model was set to zero. table 7 shows the comparison between the test natural frequencies and the ones obtained from finite element analysis. the percentage differences between the results with respect to the experimental results have significantly reduced. the percentage difference for the fundamental, second and third modes were 6.6, 4.0 and 5.2 respectively, as can be seen in table 7. table 7: comparison of experimental and updated numerical results natural frequency (hz) percentage change [%] |( 𝑓2 − 𝑓1 𝑓1 )| × 100 experiment, f1 finite element analysis, f2 mode 1 16.75 17.85 6.6 mode 2 50.00 52.70 4.0 mode 3 75.75 79.68 5.2 table 8: properties of the updated model modulus (g pa) density (kg/m3) damping constant method mass addition kg additional modulus location modulus (n/m2) 200 7830 0.05 mass and stiffness sensitivity analysis 0.012 a 200 b 50 c 50 d 50 the material properties of the updated numerical model are as presented in table 8, that is the modulus of 200 x 109 n/m2 and density of 7830 kg/m3; based on engineering judgement of the physical structure and the experimental mode shapes the mass and stiffness matrices were adjusted at the locations shown in figures (9) and (10). the mass was added to four nodes at each location, the load per node was 0.003 kg. the addition of stiffness was implemented at the locations a, b, c and d shown in figure (10) using ansys shell181 with the moduli values of 200 n/m2, 50 n/m2, 50 n/m2 and 50 n/m2 at the respective locations. journal of mechanical engineering and technology issn 2180-1053 vol.12 no.1 1 june – december 2020 table 9: experimental and finite element analysis mode shapes. experiment finite element analysis mode 1 mode 1 mode 2 mode 2 mode 3 mode 3 table 9 show good similarity between the experimental and finite element mode shapes. the masses such as that of the metallic taps omitted in the initial model were added with the use of dot elements and with the variation by trial and error of local stiffness at the joints and the lower parts of the frames with the shell elements. 9.0 summary when creating the finite element model of a structure based on available data and appropriate engineering judgements where necessary, there is no guarantee that the initial model can reasonably predict the modal properties (natural frequencies and mode shapes). journal of mechanical engineering and technology issn 2180-1053 vol.12 no.1 1 june – december 2020 according to mottershead, et al (2011) common sources of discrepancies between these results are: o idealisation errors o discretisation errors o erroneous assumptions in this report the author has evaluated the accuracy of a finite element model, with the use of elements not displayed in the final results to adjust the mass and stiffness matrices. the three mode shapes and natural frequencies of a three-storey shear frame were located experimentally using a stroboscope and the fast fourier analyser respectively. a comparison between the experimental results and the initial finite element analysis results revealed errors in the dynamics properties (i.e. frequencies and mode shapes). to reconcile or reduce the degree of mismatch, manually tuning of the finite element model was conducted after critical observation of the mode shapes with the aim to identify zones of differences in the stiffness. shell 181 element of ansys finite element code was used to locally adjust the model stiffness around the zones of connections between the shear frame floors and the pillars. also, the boundary conditions of the numerical model were controlled to closely match that of the experimental performance. this technique of stiffness addition significantly improved the correlation between the numerical and the experimental results. 10.0 references ahmadian, h., friswell, m. i. & mottershead, j. e. (1998). minimization of the discretization error in mass and stiffness formulations by an inverse method. int. journal for numerical methods in engineering, vol. 41, 371 -387. bien, j., krzyzanowski, j., poprawski, w., skoczynski, w. & szymkowski. j. (2002). experimental study of bridge structure dynamics characteristics using periodic excitation. proceedings of isma, vol. 2, pp 555 – 562. brownjohn, j. m. w. & pin-qi xia (2000). dynamic assessment of curved cable-stayed bridge by model updating. journal of structural engineering, pp 252 – 206. cha, p. d. & de pillis, l. g. (2001). model updating by adding known masses, int. j. numer. meth. engng; 50:2547 – 2571. chouksey, m., dutt, j. k. & modak, s. v. (2014). model updating of rotors supported on journal bearings. mechanism and machine theory 71, 52–63. friswell, m. i., inman, d. j. pilkey, & d. f. (1998). the direct updating of damping and stiffness matrices. aiaa journal, vol. 36, no. 3, march 1998, pp. 491-493. friswell, m. i. (2007). damage identification using inverse methods, phil. trans. r. soc. a 365, 393–410. journal of mechanical engineering and technology issn 2180-1053 vol.12 no.1 1 june – december 2020 friswell, m i., mottershead, j e. & ahmadian, h. (2001). finite-element model updating using experimental test data: parametrization and regularization. phil. trans. r. soc. lond. a, 359, 169 – 186. gordis, j. h. & papagiannakis, k. (2011). optimal selection of artificial boundary conditions for model update and damage detection. mechanical systems and signal processing, vol 25, (5), 1451 – 1468. khanmirza, e., khaji, n. & majd, v. j. (2011). model updating of multistory shear buildings for simultaneous identification of mass, stiffness and damping matrices using two different soft-computing methods. expert systems with applications 38, 5320–5329. khodaparast, h. h., mottershead, j. e. & friswell, m. i. (2008). perturbation methods for the estimation of parameter variability in stochastic model updating. mechanical systems and signal processing 22, 1751– 1773. kozak, m. t., ozturk, m. & ozguven, h. n. (2009). a method in model updating using miscorrelation index sensitivity, mechanical systems and signal processing 23, 1747– 1758. lepoittevin, g. & kress, g. (2011). finite element model updating of vibrating structures under free–free boundary conditions for modal damping prediction, mechanical systems and signal processing 25, 2203–2218. maia, n. m. m. & silva, j. m. m. (1997). theoretical and experimental modal analysis. taunton: research studies press and john wiley and sons, somerset. mares, c., mottershead, j. e. & friswell, m. i. (2003). results obtained by minimising natural-frequency errors and using physical reasoning. mechanical systems and signal processing, 17(1), 39–46. min, c. h., hong, s., park, s. y. & park, d. c. (2014). sensitivity-based finite element model updating with natural frequencies and zero frequencies for damped beam structures, int. j. nav. archit. ocean eng. 6:904~921. mottershead, j. e. & friswell, m. i. (1993). model updating in structural dynamics: a survey, journal of sound and vibration 167(2), 347 – 375. mottershead, j., link, m. & friswell, m. (2011). the sensitivity method in finite element model updating: a tutorial, mechanical systems and signal processing, vol. 25, no. 7, pp. 2275-2296. nalitolela, n, penny, j. e. t. & friswell, m. i. (1993). updating model parameters by adding an imagined stiffness to the structure, mechanical systems and signal processing 7(2), 161 – 172. journal of mechanical engineering and technology issn 2180-1053 vol.12 no.1 1 june – december 2020 parker, w. s. (2008) does matter really matter? computer simulations, experiments, and materiality, synthese, vol. 169, no. 3, pp. 483-496. ren, w. & chen, h. (2010). finite element model updating in structural dynamics by using the response surface method. engineering structures 32, 2455 -2465. shan, d., li, q., khan, i., & zhou, x. (2015). a novel finite element model updating method based on substructure and response surface model. engineering structures 103, 147–156. sinha, j. k. & friswell, m. i. (2002). model updating: a tool for reliable modeling, design modification and diagnosis. the shock and vibration digest, vol. 34, no. 1, january 2002, pp 27-35. titurus, b., friswell, m. i. & starek, l. (2003). damage detection using generic elements: part i. model updating. computers and structures 81, 2273–2286. zang, c., friswell, m. i. & imregun, m. (2004) structural damage detection using independent component analysis, structural health monitoring, vol 3(1), pp: 69–83. zapico-valle, j. l., alonso-camblor, r., gonzalez-martinez, m. p. & garcia-dieguez, m., (2010). a new method for finite element model updating in structural dynamics mechanical systems and signal processing 24, 2137–2159. zapico, j l., gonzalez, m p., friswell, m i., taylor, c a. & crewe, a j. (2003). finite element model updating of a small-scale bridge, journal of sound and vibration 268, 993–1012 journal of mechanical engineering and technology *corresponding author. email: qielahsan@gmail.com issn 2180-1053 vol. 12 no.1 1 june – december 2020 non-catalytic microwave assisted transesterification of palm oil with dimethyl carbonate aqilah s1; ong my*2 and saifuddin n.3 institute sustainable energy, universiti tenaga nasional, 43000 kajang, selangor, malaysia. email: 1qielahsan@gmail.com; *2me089475@hotmail.com, 3saifuddin@uniten.edu.my (* corresponding author) abstract: a reactor-condenser microwave (600w) was modified as an assisted method for continuous transesterification of palm oil. the high free fatty acid oil was simultaneously neutralized and trans esterified with dimethyl carbonate. with the dmc to oil molar ratio of 10:1, 7:1 and 5:1, with temperature range of 150 degrees to 250 degrees, 2 to 4-hour residence time, the continuous conversion of palm oil to ethyl ester was over 90%. the palm oil biodiesel was analyses using ftir analysis to determine the conversion yield. most ideal ratio was figured out to be 1:7 (oil to dmc) and it continue to next 4 hour of heating to obtain the best result. the maximum conversion yield achieve was 95.9% and the density, viscosity also fuel properties achieve astm standard. keywords: microwave reactor; non-catalytic; palm oil; ftir analysis 1 introduction the world is evolving, creating a new developed civilization which make the petroleum or fossil fuel as an important aspect. nevertheless, the source was limited because it will be depleted. yet, the energy demand keeps increasing while the power supply industry was neck on neck to fulfill the demand. even worse, the main feedstock used to generate energy is mostly nonrenewable resources. hence, the research nowadays have been shifted to renewable energy in order to support sustainable development, especially biofuel (lin & chen, 2017). biofuel is getting more attention, journal of mechanical engineering and technology issn 2180-1053 vol. 12 no.1 1 june – december 2020 particularly biodiesel, to reduce the greenhouse effect and air pollution. biodiesel, also known as fatty acid methyl esters (fame), chemically can be synthesized through the chemical reaction between oil (usually vegetable oil) and alcohol in the presence or absence of catalyst (nomanbhay & mei yin, 2017, sahar et al., 2018). the feedstock that usually used in producing biodiesel are palm oil, rapeseed oil, canola oil, bacteria, algae, waste cooking oil (wco) etc. as one of the huge producer of palm oil, the research in malaysia has seemed take interest in palm oil industry. the palm oil industry was developed massively as the demand was increasing for last 25 years, and now, enhancement has been made by focusing on the quality improvement, disease resistant and high increment on palm oil (masani et al., 2018). a study was done by el-araby et al., (2018) on the palm oil–biodiesel–diesel fuel blend properties (viscosity, density and flash point) in diesel engine where they found that palm oil properties in engine performance has no difference with regular diesel oil. hence, this has made the palm oil and blended diesel-palm oil have huge potential in transportation industry. there were many methods can be used to convert the feedstock into biodiesel or biofuel. several conversion method and technology have been studied and validated its effectiveness. among the available technologies, the chemical conversion method has been proven as one of the simplest and widely-used process in producing biofuel. xiang et al., (2017) studied on the effect of modified coal fly ash as catalyst in increasing the biodiesel yield of waste cooking oil. the highest biodiesel yield of 94.91% was obtained with the ratio of 1:9.67 (oil to methanol). the yield was increased by 90% if the catalyst was used for 8 times. besides, sahar et al. (2018) also conducted transesterification on waste cooking oil using reactor equipped with reflux condenser and the result getting the achievement of fame yield up to 94% with 1:3 oil-methanol ratio and 1 percent of catalyst. the microwave-assisted (thermochemical) conversion have become more demanding than other method and it was evolving through quite some time as some study shows its effectiveness in achieve high conversion yield for short of time (lertsathapornsuk et al., 2008) furthermore, superior advantages in term of time and product yield was also observed when microwave processing technology was implemented into transesterification reaction (li et al., 2018). electromagnetic waves that emitted from magnetron has the ability to make the material absorb the energy and convert it into heat, providing volumetric heating effect. study done by ding et al., (2018) using microwave irritation condition on transesterification process for biodiesel journal of mechanical engineering and technology issn 2180-1053 vol. 12 no.1 1 june – december 2020 production concluded that biodiesel yield can be achieved up to 98.93% with molar ratio of 1:11 oil to methanol. this has also proven that microwave-assisted transesterification is an efficient and environmental friendly biodiesel production method. the research keep continues to find the best way to convert feedstock to biodiesel but not much on palm oil itself. this paper will focus on the non-catalytic conversion of palm oil into biodiesel with microwave-assisted method at milder condition using dimethyl carbonate. this paper focus on non-catalytic transesterification process as it provides simpler purification and environmentally friendly post-processes (cho et al., 2012). besides, without catalyst, the problems, such as the formation of unwanted soap instead of biodiesel due to the high free fatty acid content in the feedstock, will be eliminated. diasakou et al., (1998) investigated the transesterification of soybean oil with methanol for non-catalytic thermal method at 220-235°c, they found that diglycexide and triglyceride conversion rates are much higher than the conversion rates of monoglyceride to glycerol. a non-catalytic with supercritical methanol allows a simple process and obtained high yield in biodiesel production despite also having likely same properties with biodiesel and petrodiesel make the transesterified product of vegetable oil considered the most promising one to substitute diesel (demirbas, 2006). there were also two-step method suggest by minami and saka (2006) which include the hydrolysis step in the transesterification process to against the presence of water in oil/fats and it was high tolerant. asri et al., (2013) figured out that non-catalytic of transesterification process for vegetable oil in supercritical methanol can overcome the flaw of homogeneous catalyst process. ilham and saka (2011) in other hand study on the potential of dimethyl carbonate as non-catalytic reactant and found out that it can be a good candidate in the supercritical condition. however, the studies mentioned above required high temperature and pressure operating condition, which lead to high energy input and production cost. hence, non-catalytic transesterification under milder condition is needed to be explored further. to the best knowledge of the author, there are little information about the noncatalytic microwaveassisted transesterification under subcritical condition. in this paper, the effect of oil-to-dimethyl carbonate (dmc) molar ratio and reaction temperature on biodiesel conversion yield was investigated under microwave irradiation using non-catalytic transesterification method. journal of mechanical engineering and technology issn 2180-1053 vol. 12 no.1 1 june – december 2020 2 material and methods this study was devoted to convert the palm oil mixed with dimethyl carbonate (dmc) to biodiesel using most economic method, which is by microwave-assisted reaction. this noncatalytic experiment was conducted at optimum and ideal measured where the transesterification process in the microwave reactor also been studied. the gained sample was been analyzed using fouriertransform infrared spectroscopy (ftir) to determine the conversion of yield of ester and decided which sample give the highest percentage of conversion yield. technically, the palm oil was heated in the microwave before pouring in the dmc according to the decided molar ratio and will in constant heat in the range of 150 degree to 250 degree celsius. the sample was been drawn out at the end of each experiment and was analyzed. 2.1 materials refined commercial palm oil was obtained from local store located at bangi, selangor. this oil has main fatty acid composition: myristic acid 1.1%, lauric acid 0.2%, stearic acid 4.5%, linoleic acid 10.1%, oleic acid 39.2%, palmitic acid 44.0%, arachidic acid 0.1%, and linolenic acid 0.4%. while for the dimethyl carbonate (dmc) was purely analytical and was purchased from sigma aldrich company (99%, mw : 90.08 g/mol) 2.2 preparation of biodiesel the biodiesel was prepared in 2l beaker with a magnetic stirrer, a thermometer and watercooled condenser for transesterification reaction. the palm oil (762.57g) and dmc (810.81g) was displaced into the beaker at predetermined temperature with different palm oil and dmc mole ratio (1:5, 1:7 and 1:10). 2.3 palm oil transesterification the total reaction volume of 1.58 liters was considered as constant for transesterification process. the sample was heated up in range of 150 degrees to 250 degrees for about 2 hours while the microwave was set to 600 watts. sample was then withdrawn from microwave using pump at every journal of mechanical engineering and technology issn 2180-1053 vol. 12 no.1 1 june – december 2020 two hours at determined temperature. the sample was then analyze before decided the ideal ratio blend of oil and dimethyl carbonate for heat it up again up to 4 hours. 2.4 ftir analysis the sample was undergone fourier-transform infrared spectroscopy (ftir) to obtained infrared spectrum of emission of absorption of a sample. the measurement was performed simply by pouring a droplet of liquid sample onto crystal surface. before that, background spectrum has to be obtained to avoid any disruption from the outside factor on the result. cleaning was done with trisolvent mixture of acetone – toluene – methanol, this was employed to clean the sample crystal before background scan. the scan results were obtained on the incorporated computer system as spectra. total require time for spectral collection takes approximately 5 minutes per sample. the spectra were recorded within the range of 4000 to 600 cm-1 with 4 cm-1 and happgenzel appodization (rabelo et al., 2015). 20 scans were calculated on each spectrum subjected to background subtraction. analyses were carried out in triplicates and the average of the three were used to construct the models of spectra 2.5 analysis of sample sample prepared as described above were analyzed to determine its conversion yield. mixture of fatty acid methyl ester, fame was calculated using calibration method with the sample gain and find the percentage difference between actual value of biodiesel and fame (1) 𝑌𝑖𝑒𝑙𝑑 (%) = 100% − 𝑃𝑒𝑟𝑐𝑒𝑛𝑡𝑎𝑔𝑒 𝐷𝑖𝑓𝑓𝑒𝑟𝑒𝑛𝑐𝑒 (2) 3 result and discussion 3.1 effect of molar ratio of oil to dmc the molar ratio between dmc and palm oil played an important measure in determine the ester conversion yield. the stoichiometry of this reaction requires at least three moles of methanol or two moles of dmc per mole of vegetable oil to yield three moles of fatty ester and one mole of journal of mechanical engineering and technology issn 2180-1053 vol. 12 no.1 1 june – december 2020 glycerol (anastopoulos et al., 2009). it was proven that increasing of dmc concentration in molar ratio helps the mass transfer to be increase as the reaction mixture overall viscosity decreased (panadare & rathod, 2016). nevertheless, it required more power to increase the working volume of overall reaction. in nature itself, fame are reversible which leads to more reactant need to be added to counter this forward reaction. considering this facts, the sample was made of three different molar ratio palm oil to dmc with range of 1:5, 1:7 and 1:10 where dmc should be more than oil ratio itself. logically, glycerol will be formed as impart product because of the enzyme binding active site when using other solvent. to combat this, dmc was used to partially solved this problem by making immediate conversion of glycerol to glycerol carbonate the effect of molar ratio of oil to dmc was studied at various amounts of dmc (molar ratios of methanol to oil = 5-10), and the others variables were fixed (1 molar of palm oil, temperature range of 150 250°c, 2 hour reaction time, 600-watt microwave power, and 300 rpm stirring rate). graphical presentation of the results from figure 1, figure 2 and figure 3 shows result of ftir analysis and all of the sample has been compared to biodiesel. the peaks at 1427 cm -1 indicated the deformation vibrations of ch2. the highest conversion yield achieved for each ratio were 90.1%, 95.6% and 89.6% respectively at the peaks of 1427 cm-1 after 2 hour of heating. result tells that with the increase dimethyl carbonate to palm oil mole ratio, the yield increased but it decreased back after 1 to 7 ratio. at higher molar ratios, excessive dmc dilutes the concentration of oil and reduces the collision frequency of reactants and catalyst. therefore 7:1 was considered as optimum ratio for given process so as to give maximum yield of 95.6%. reaction studied without application of microwave was reported 7:1 as the optimized ratio which is more than the optimized value for microwave assisted reaction. as discussed before, power consumption is low at lower viscosity, which is maintained by high reactant ratio. in addition, as dmc was more polar than the oil, it improved the dielectric constant in if using it in high ratio which will be favor the absorption of electromagnetic vibration and heat transfer. result for this section will be summarized in table 1, table 2 and table 3 where shows every conversion yield at each temperature for all ratio in this experiment journal of mechanical engineering and technology issn 2180-1053 vol. 12 no.1 1 june – december 2020 table 1: conversion yield at each determined temperature for oil to dimethyl carbonate ratio of 1:5 temperature (°c) conversion yield at peaks of 1427 cm-1 (%) 200 92.6 250 90.1 table 2 : conversion yield at each determined temperature for oil to dimethyl carbonate ratio of 1:7 temperature (°c) conversion yield at peaks of 1427 cm-1 (%) 150 72.4 200 89.3 250 95.6 table 3 : conversion yield at each determined temperature for oil to dimethyl carbonate ratio of 1:10 temperature (°c) conversion yield at peaks of 1427 cm-1 (%) 150 88.8 200 89.6 250 82.4 journal of mechanical engineering and technology issn 2180-1053 vol. 12 no.1 1 june – december 2020 figure 1: spectrum of ftir analysis on the palm oil biodiesel sample with oil to dmc ratio 1:5 figure 2: spectrum of ftir analysis on the palm oil biodiesel sample with oil to dmc ratio 1:7 journal of mechanical engineering and technology issn 2180-1053 vol. 12 no.1 1 june – december 2020 figure 3: spectrum of ftir analysis on the palm oil biodiesel sample with oil to dmc ratio 1:10 3.2 effect of reaction time on yield next experiment have been made to few sample while prolong the period of heating. the influence of reaction time on the biodiesel yield was investigated, while the reaction time was prolonging from 2 hours to 4 hours, mole ratio of dmc to oil was 7:1 (optimum ratio for microwave assisted reaction) and microwave power was constant to 600 watts. according to figure 4, the yield increased slightly when the reaction time prolonged and it achieve 95.9% of conversion yield compare to before. at 2 hour of heating, the conversion yield achieve was just 95.6% and it was increase by 0.3 percent by prolong the heating to 4 hours. therefore, it could be considering the longer the heating period, it may not effect much on the conversion yield as it also may be ideal for reaction time below 2 hour as based on previous study stated that reaction time of 50 min was found suitable for higher fame yield and by increasing the reaction time, there no significant change in fame yield (sahar et al., 2018) journal of mechanical engineering and technology issn 2180-1053 vol. 12 no.1 1 june – december 2020 table 4 : change in conversion yield after prolong the heating for 4 hour for ratio of 1:7 temperature (°c) conversion yield at peaks of 1427 cm-1 (%) 150 88.3 200 70.3 250 95.9 figure 4: spectra from ftir analysis that was done on sample ratio of 1:7 after prolong the heating to 4 hours 3.2 effect on viscosity and density conversion of palm oil to methyl ester will be effect on its viscosity and density as we are using different method. the table 5 presented the data gained after the sample has been measured and compared to the diesel and biodiesel based on astm standard (el-araby et al., 2018). for palm oil methyl ester, the density and viscosity result was fall in between the range of biodiesel properties while for flash point was bit off from the range. because of the molecular weight of the palm oil itself make the flash point of palm oil methyl ester is higher than diesel and biodiesel. as the flash point are higher it may takes time for palm oil methyl ester to ignite in combustion engine. journal of mechanical engineering and technology issn 2180-1053 vol. 12 no.1 1 june – december 2020 table 5: properties of diesel fuel, biodiesel and palm oil biodiesel. fuel property diesel biodiesel palm oil methyl ester density @15 °c, g/ml 0.848 0.978 0.878 kin. viscosity @40 °c, mm2/s 1.3–4.1 1.9–6 4.55 flash point, °c 60–80 100–170 197 previous study has been made on different molar ratio, heated under 100 degrees for short period of time. the conversion yield achieves almost the same with this study. specifically, the yield of biodiesel was only 52.69% after 1 hour while the yield reached 97.85% after 6 hour (ding et al., 2018). however, the yield was maintained at approximately 98%, even when the reaction time continuously enhanced to 8 hours. the temperature not playing big role in manipulate the conversion yield whilst it may affect the viscosity and density of the product itself. fuels with higher viscosity increases the problems in atomization and damages the fuel injector, thus ultimately results in incomplete combustion and poor engine performance leads to damaging of the engine and also the deposition of solid unburned particles. the fuels with lower viscosity lacks in providing lubrication to the pump and injector, so there also damaging takes place hence optimum viscosity is needed that lies within the range prescribed by astm and en standards. the viscosity of the biodiesel must be at optimum level for it to work ideally. while for density, it seems impossible to achieve 100% of exact conventional diesel but for biodiesel will be varied 37.27 mj/kg for its calorific value. this is 9% lower than regular number 2 petro diesel. variations in biodiesel energy density is more dependent on the feedstock used than the production process. at the end, we can conclude even though the process was varied by temperature and time, it may effect on viscosity difference for biodiesel produce even if it achieves 100 percent conversion yield. as being said, assumption was made that better viscosity of biodiesel can be produce at higher temperature of reaction. journal of mechanical engineering and technology issn 2180-1053 vol. 12 no.1 1 june – december 2020 3.3 fuel properties biodiesel that was produced through the conversion from palm oil using dimethyl carbonate method must ensure that it fulfill the standard properties for biodiesel. fuel properties were studied as shown in table 6. the biodiesel conversation in this was prepared in ideal condition (150°c250°c/120 min/1:10, 1:7, 1:5 molar ratio of oil to dimethyl carbonate). the result was obtained been compared with eu, us and japanese standard (ilham & saka, 2012). generally, the result from this method following all the requirement based on international standard excluding oxidation stability. oxidation stability was indicator for chemical reaction that happened when there were interaction between oxygen and lubricating oil. high temperature, acids, water and catalyst can affect the rate of oxidation (machinerylubrication.com, 2018). it is one of the very important measure to avoid deterioration and it could be enhance by utilizing the oils with lower level of unsaturated fatty acid or by adding the antioxidant. in non-catalytic transesterification process, especially using microwave assisted method, can be more efficient to support the reaction due to minimizing the mass transfer resistance as to commercialize grade biodiesel product with little lost (tran et al., 2017). low grade fatty materials such as waste cooking oil can be used as feedstock where traditional transesterification cannot be permitted due to high content of fatty acid. nevertheless, quality of biodiesel produced from fame may be varied in each method of conversation. quality problem may be resulted ineffective and low performance of the biodiesel itself or in some cases may broke the engine. the comparison between the properties gained from the sample with the international standard for biodiesel are very important to ensure the biodiesel gained from this conversion process can be used safely and perform same as conventional biodiesel. non-catalytic transesterification also have to be commercialized as to reduce the cost of biodiesel production thus this study was very important. due to this, next study should be focus on effectiveness of this method in large production for it to be commercialize in the oil market as well as how it going to effect the plantation of the palm oil itself. in other similar studies in supercritical condition, it has been found that using dimethyl carbonate as in non-catalytic transesterification will produce no glycerol for biodiesel production as it was more convenient rather than normal transesterification (ilham & saka, 2011). journal of mechanical engineering and technology issn 2180-1053 vol. 12 no.1 1 june – december 2020 table 6: fuel properties of fatty acid methyl esters (fame) from palm oil as treated in supercritical dimethyl carbonate method compared with international standards. properties method unit fame (palm oil) international standard japan (jis k2390) eu (en 14214) us (astm d6751 07) kinematic viscosity (40 °c) astm d445 mm2/s 4.5 3.5–5.0 3.5–5.0 1.9–6.0 carbon residue astm d4530 wt% 0.09 ⩽0.30 ⩽0.30 ⩽0.05 pour point astm d2500 °c −7.0 – – – cold filter plugging point astm d6371 °c −7.2 – – – ignition point astm d93 °c 163.6 ⩾120 ⩾101 ⩾130 cloud point astm d6749 °c −7.0 – – – oxidation stability en 11442 h 5.7 – ⩾6 – ester content en 14103 wt% 98.5 >96.5 >96.5 – monoglyceride en 14105 wt% 0.9 <0.80 <0.80 – diglyceride en 14105 wt% 0.07 <0.20 <0.20 – total glycerol content en 14105 wt% 0.04 <0.25 <0.25 <0.24 water content en iso12937 mg/kg 230 <500 <500 <500 acid number en 14104 mg(koh)/g 0.18 <0.50 <0.50 <0.50 iodine value en 14111 g(i2)/100 g 110 <120 <120 – 4 conclusion in this paper, three different ratio of oil to dmc was put into test to determine the conversion yield compare to the original biodiesel. it was reveal that using microwave-assisted method could possibly achieve higher conversion yield among other conversion method as the entire sample achieves more than 85%. ratio of 1:5 achieve 89.5%, for 1:7 ratio achieve 95.5% while for 1:10 ratio gained 98.85% of conversion yield for 2 hour of heating. the next phase using 1:7 ratio as journal of mechanical engineering and technology issn 2180-1053 vol. 12 no.1 1 june – december 2020 ideal ratio achieved the highest conversion yield which is 99.7% make it more similar to original biodiesel after being heat for 4 hour. density, viscosity and flash point where been compared with astm standard and it achieve between the range except for flash point for it has more molecular weight. fuel properties also been studied and most of it attain the astm standard except for oxidation stability. real test on combustion engine was suggested to be made for further effectiveness of biodiesel that was extract from palm oil to be commercialize in the market especially transportation. acknowledgements the authors would like to thank the university tenaga nasional (uniten) for the research facilities. the funding from e-sc research grant from the ministry of science technology and innovation malaysia (e-sc-03-02-03-sf0287) supported this work. competing interests: the authors declare that they have no competing interests. 5 references anastopoulos, g., zannikou, y., stournas, s. and kalligeros, s. (2009). transesterification of vegetable oils with ethanol and characterization of the key fuel properties of ethyl esters. energies, 2(2), pp.362-376. asri, n., machmudah, s., wahyudiono, w., suprapto, s., budikarjono, k., roesyadi, a. and goto, m. (2013). non catalytic transesterification of vegetables oil to biodiesel in sub-and supercritical methanol: a kinetic’s study. bulletin of chemical reaction engineering & catalysis, 7(3). cho, h., kim, s., hong, s. and yeo, y. (2012). a single step non-catalytic esterification of palm fatty acid distillate (pfad) for biodiesel production. fuel, 93, pp.373-380. demirbas, a. (2006). biodiesel production via non-catalytic scf method and biodiesel fuel characteristics. energy conversion and management, 47(15-16), pp.2271-2282. journal of mechanical engineering and technology issn 2180-1053 vol. 12 no.1 1 june – december 2020 diasakou, m., louloudi, a. and papayannakos, n. (1998). kinetics of the non-catalytic transesterification of soybean oil. fuel, 77(12), pp.1297-1302. ding, h., ye, w., wang, y., wang, x., li, l., liu, d., gui, j., song, c. and ji, n. (2018). process intensification of transesterification for biodiesel production from palm oil: microwave irradiation on transesterification reaction catalyzed by acidic imidazolium ionic liquids. energy, 144, pp.957967. el-araby, r., m. amin, a., el morsi, a., el-ibiari, n. and el-diwani, g. (2018). study on the characteristics of palm oil–biodiesel–diesel fuel blend. egyptian journal of petroleum, 27(2), pp.187-194. ilham, z. and saka, s. (2012). optimization of supercritical dimethyl carbonate method for biodiesel production. fuel, 97, pp.670-677. ilham, z. and saka, s. (2011). production of biodiesel with glycerol carbonate by non-catalytic supercritical dimethyl carbonate. lipid technology, 23(1), pp.10-13. lertsathapornsuk, v., pairintra, r., aryusuk, k. and krisnangkura, k. (2008). microwave assisted in continuous biodiesel production from waste frying palm oil and its performance in a 100 kw diesel generator. fuel processing technology, 89(12), pp.1330-1336. li, s., chen, c., zhang, d., zhang, x., sun, b. and lv, s. (2018). microwave-assisted fast and efficient dissolution of silkworm silk for constructing fibroin-based biomaterials. chemical engineering science, 189, pp.286-295. lin, j. and chen, y. (2017). production of biodiesel by transesterification of jatropha oil with microwave heating. journal of the taiwan institute of chemical engineers, 75, pp.43-50. machinerylubrication.com. (2018). the importance of oil oxidation stability. [online] available at: https://www.machinerylubrication.com/read/28966/oil-oxidation-stability [accessed 16 nov. 2018]. masani, m., izawati, a., rasid, o. and parveez, g. (2018). biotechnology of oil palm: current status of oil palm genetic transformation. biocatalysis and agricultural biotechnology, 15, pp.335347. journal of mechanical engineering and technology issn 2180-1053 vol. 12 no.1 1 june – december 2020 minami, e. and saka, s. (2006). kinetics of hydrolysis and methyl esterification for biodiesel production in two-step supercritical methanol process. fuel, 85(17-18), pp.2479-2483. nomanbhay, s. and mei yin, o. (2017). a review of microwave-assisted reactions for biodiesel production. bioengineering, 4(4), p.57. panadare, d. and rathod, v. (2016). microwave assisted enzymatic synthesis of biodiesel with waste cooking oil and dimethyl carbonate. journal of molecular catalysis b: enzymatic, 133, pp.s518-s524. rabelo, s., ferraz, v., oliveira, l., & franca, a. (2015). ftir analysis for quantification of fatty acid methyl esters in biodiesel produced by microwaveassistedtransesterification. international journal of environmental science and development, 6(12), 964-969. doi: 10.7763/ijesd.2015.v6.730 sahar, sadaf, s., iqbal, j., ullah, i., bhatti, h., nouren, s., habib-ur-rehman, nisar, j. and iqbal, m. (2018). biodiesel production from waste cooking oil: an efficient technique to convert waste into biodiesel. sustainable cities and society, 41, pp.220-226. tran, d., chang, j. and lee, d. (2017). recent insights into continuous-flow biodiesel production via catalytic and non-catalytic transesterification processes. applied energy, 185, pp.376-409. xiang, y., xiang, y. and wang, l. (2017). microwave radiation improves biodiesel yields from waste cooking oil in the presence of modified coal fly ash. journal of taibah university for science, 11(6), pp.1019-1029. exergo-economic analysis of performance of a gas turbine power generation system with a solar air preheater masoud valizadeh*1 1 department of energy engineering, college of environment and energy, science and research branch, islamic azad university, p.o. box 14515-775, tehran, iran. abstract the power generation sector, especially the gas turbine, is one of the most critical sources in order to eliminate greenhouse gases worldwide. in a thermo-dynamic system, exergo-economic analysis is utilized as a means to specify the inefficient thermo-dynamic points, where the highest loss of exergy arises. in this paper, engineering equation solver (ees) software and exergo-economic analysis, which uses both the second-law of thermodynamics and economic principles, are utilized to evaluate the economical and exergetical performance of the gas turbine with solar air preheater. the gas turbine without preheating of the air entering the combustion chamber is first investigated. then, based on three concepts including relative difference, exergo-economic coefficient and exergetic efficiency, a comparison study is performed between the gas turbine with and without solar air preheater. the results clearly reveal that by increasing the inlet temperature of the combustion chamber from 620˚k to 820˚k, the exergy factor increases from 0.41% to 0.68%. also, the consumption of gas turbine with solar air preheater is reduced from 8.99 kg/s to 7.84 kg/s by raising the inlet temperature of the combustion chamber. as a result, it is noteworthy to express that the exergetic efficiency is increased from 58.4% to 63.4%. keywords: exergo-economic analysis; solar air preheater; gas turbine; ees software. 1. introduction nowadays, the use of renewable energy sources is rapidly progressing. the researchers believe that supply of energy using renewable sources such as water, wind and sun should be a priority * corresponding author: m.masoud.valizadeh@gmail.com journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 2 june – december 2019 1 mailto:m.masoud.valizadeh@gmail.com instead of oil. among them, sun has become more appealing to the researchers. generally, hybrid power plants use a fossil fuel such as diesel or gas, supplemented with a renewable energy source such as solar and wind. as shown in fig. 1, a solar power plant is a set of facilities which collect radiation energy from the sun, or by focusing it gives high temperature. the energy collected through a heat exchanger, turbines or steam engines will be converted into electrical energy, leading to reduction in cost as any other conventional plant [1-3]. fig. 1 schematic of a solar power plant, [4, 5]. in general, solar power plants correspond to the concept of focusing solar radiation to produce steam or hot air which can then be utilized for electricity generation. there are mainly four known solar power plant based on the receiver, and among them only one that is investigated in this paper. in solar tower power plant, also known as central receiver systems, sunrays are concentrated by a field consisting of reflectors, called heliostats, on a receiver which stands on the top of the tower. heliostats are flat or slightly concave mirrors which follow the sun in a two axis tracking. the central receiver at the top of the tower converts solar energy into thermal energy and transfer the heat generated to the fluid flowing through it. the fluid becomes steam after receiving heat which generates electricity. moreover, the use of heliostats and placing a receiver prior to the combustion chamber results in increasing the temperature of the intake air into the combustion chamber and thus reduction in the fuel consumption. however, the use of solar energy in the gas turbine cycle is one of the new methods for increasing efficiency [4-7]. optimization is one of the most important issues for design of energy systems. in large thermal systems, which have many design variables, conventional mathematical optimization methods are journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 2 june – december 2019 2 not efficient. in recent decades, due to the rise in energy costs and restrictions of non-renewable energy, the optimal performance of the system has attracted special importance from the point of view of energy and production costs. thus, the combination of the second law of thermodynamics with the economic concept has led to the formation of a powerful tool known as exergo-economic analysis in order to optimize the thermal systems and design efficient and cost-effective systems [8-10]. such concept of optimization became popular among researchers, with studies performed by antonio valero [11], richard gaggioli [12], and el-sayed [13]. in recent times, comprehensive works have been performed on the application of exergoeconomic concept for analysis and optimization of energy systems [14-20]. the purpose of work done by khaljani and his associates [21] was thermodynamic, exergo-economic and environmental evaluation of heat and power cycle. in their work, the three objective functions of first and second law efficiencies and the total cost rates of the system were considered. the main result of their assessment was that combustion chamber, and heat recovery steam generator and gas turbine had the most exergy destruction rate, respectively. the exergetic sustainability indicators were extended by aydin [22] in order to analyze gas turbine engine based power plant and specify sustainability aspects of it. mousafarash et al. [23, 24] investigated energy, exergy and exergoeconomic analyses of a gas turbine power generation system. the results obtained from this study represented that the combustion chamber, where the high temperature difference is the main source of the irreversibility, had the greatest exergy destruction rate. in order to evaluate the cost rate related to all the exergy streams at cycle state points, engineering equation solver software and exergo-economic methods were employed to analyze a 100 mw gas turbine power plant at ughell, nigeria [25]. in another work, exergo-economic assessment of solar hybrid power generation systems combining with thermo-chemical fuel conversion was studied by yue and lior [26]. a gas turbine power plant was simulated by ahmadi and dincer [27] according to thermodynamic and exergo-economic approaches, in which the results were compared with one of the largest gas turbine power plants in iran in order to verify their thermodynamic model. 2. exergo-economic analysis of gas turbine with solar air preheater the gas turbine mechanism with gas and solar air preheaters discussed in this study is depicted in fig. 2. in this section, using the air thermodynamic table and the ees library functions, enthalpy journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 2 june – december 2019 3 and entropy values are calculated at each point of the flow path (points 1-10) of table 1. following eq. (1): 𝐸𝑥 = (ℎ − ℎ0) − 𝑇0(𝑆 − 𝑆0) (1) ℎ0, 𝑆0 and 𝑇0 = 298 𝐾 are enthalpy, entropy and temperature at the reference point, respectively. also, ℎ and 𝑆 are enthalpy and entropy at considered points, respectively. after identifying the exergy of products, 𝐸𝑥𝑝, and fuel exergy, 𝐸𝑥𝑓 , the exergy loss, 𝐸𝑥𝐷 , is evaluated as follows: 𝐸𝑥𝐷 = 𝐸𝑥𝑓 − 𝐸𝑥𝑝 (2) by determining the exergy of each stream, the actual energy loss or, in other words, the thermodynamic inefficiencies of exergy loss and exergetic efficiency are determined for each component of the system. fig. 2 schematic of a gas turbine with gas and solar air preheaters table 1 calculation of exergy values for gas turbine with solar air preheater position fluid mat. pressure (bar) mass (kg/s) temperature (k) exergy (mw) exergy destroid 1 air 1.013 427 298 0 0 2 air 10.47 427 620 130.98 25.2 journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 2 june – december 2019 4 2’ air 10.15 427 720 155.15 9 3 air 9.84 427 810 163 73.23 4 gas-pro 9.54 436 1320 365 259.9 5 gas-pro 1.08 436 810 100.84 38.62 5’ gas-pro 1.032 436 700 67.67 6 fuel 30 8.99 308 462.2 7 molten salt 1.013 120 820 135.1 8 molten salt 1.013 120 298 46.2 3. ge-f9 gas turbine model in our case study, the ge-f9 gas turbine (100 mw) is considered with solar air preheating system. there are important variables as exergo-economic parameters, discussed as follows:  the pressure ratio 𝑟𝑝 = 𝑃2 𝑃1⁄ (output pressure of compressor on inlet pressure of compressor), which depends on the position of the gas turbine installation and varies from 10.5 to 12, which is assumed to be 11 in this study.  isentropic efficiency of compressor and turbine, which according to the documentation of power plant was calculated 88% and 89%, respectively [5, 25].  the outlet temperature of the solar air preheater is assumed to be 820 k.  the outlet temperature of the combustion chamber and the inlet temperature entering the turbine are considered 1400 k. 4. economic analysis it is necessary to examine each component of the turbine in order to determine the economic situation of a gas turbine. these costs include the cost of ownership and exploitation, and each of journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 2 june – december 2019 5 them depends on the factors such as unit life, investment conditions, and financing structure, calculated for each component according to following. 6.1. purchased equipment cost (pec) calculation the cost of purchasing and investing for an equipment is calculated based on the following model. a) cost evaluation for compressor 𝑃𝐸𝐶𝑎𝑐 = ( 71.1 𝑚𝑎 0.9 − 𝜂𝑎𝑐 ) . ( 𝑃2 𝑃1 ) . 𝑙𝑛( 𝑃2 𝑃1 ) (3) b) cost evaluation for combustion chamber 𝑃𝐸𝐶𝑐𝑐 = ( 46.08 𝑚𝑎 0.995 − 𝑃3 𝑃2 ) . (1 + 𝑒𝑥𝑝 (0.081𝑇3 − 26.4)) (4) c) cost evaluation for gas air preheater 𝑃𝐸𝐶𝑎𝑝ℎ = 4122 ( 𝑚𝑔(ℎ5 − ℎ5 ′ ) 𝑈∆t𝑚,𝑎𝑝ℎ ) 0.6 (5) d) cost evaluation for gas turbine 𝑃𝐸𝐶𝑔𝑡 = ( 479.34 𝑚𝑔 0.92 − 𝜂𝑔𝑡 ) . ( 𝑃3 𝑃4 ) . (1 + 𝑒𝑥𝑝 (0.036𝑇3 − 56.4)) (6) table 2 calculation of pec components of gas turbine pec (m$) purchase cost of compressor 19 purchase cost of gas air preheater 0.46 purchase cost of heliostat 70.318 purchase cost of tower and receiver construction 13.104 purchase cost of concrete tower construction 3.7306 purchase cost of tank 14.7347 journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 2 june – december 2019 6 purchase cost of combustion chamber 0.83 purchase cost of turbine 15.18 6.2. calculation of �̇�𝑘 using the equilibrium equation and the effect of the inflation rate i = 13%, and the time period of depreciation of n = 20 years, �̇�𝑘 is calculated based on the following relation: �̇�𝑘 = ( 𝑃𝐸𝐶 − 0.1 (𝑖 + 1)𝑛 ) (𝑖 − 1 (𝑖 + 1)𝑛 ) (7) 6.3. calculation of 𝑍𝑘 in order to calculate 𝑍𝑘, which includes repair and maintenance costs, the coefficient 𝜙𝑘 = 1.06 and 𝐻 are respectively considered as the cost correction factor and the unit operating hours per year (approximately 𝐻 = 24 × 365 × 0.85 = 7446 hours). 𝑍𝑘 = 𝜙𝑘 �̇�𝑘 𝐻 (8) table 3 economic calculation of gas turbine with solar air preheater component purchased equipment cost pec ($) annual levelized cost �̇�𝑘 ($) capital cost rate �̇�𝑘 ($/h) air compressor 19 × 106 2.66 × 106 378.6 air preheater 0.46 × 106 0.0641 × 106 9.11 solar air preheater 101.8 × 106 9.6 × 106 1367 combustion chamber 0.83 × 106 0.12 × 105 16.5 gas turbine 15.18 × 106 2.12 × 106 302 journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 2 june – december 2019 7 6.4. exergo-economic analysis given that the cost analysis in each system and in each component of the system is different, we use an equilibrium equation of exergy flow. ∑ �̇�𝑂,𝑘 𝑜𝑢𝑡 = ∑ �̇�𝑖,𝑘 𝑖𝑛 + 𝑍𝑘 (9) using equation (11), the exergo-economic relations of the gas turbine with preheater are determined in accordance with table 4. table 4 exergo-economic equations for gas turbine with solar air preheater components the main equations of energy balance the equivalent equations of energy balance air compressor �̇�1 + �̇�9 + �̇�𝑐𝑜𝑚𝑝 = �̇�2 𝐶1 = 0; 𝐶𝑤,9 = 𝐶𝑤,10; �̇�9/𝐸�̇�9 = �̇�10/𝐸�̇�10 combustion chamber �̇�3 + �̇�6 + �̇�𝑐𝑐 = �̇�4 gas turbine �̇�4 + �̇�𝑔𝑡 = �̇�5 + �̇�𝑤,9 + �̇�𝑤,10 𝐶4 = 𝐶5; �̇�4/𝐸�̇�4 = �̇�5/𝐸�̇�5 gas air preheater �̇�2 + �̇�5 + �̇�𝑎𝑝 = �̇�2′ + �̇�5′ 𝐶5 = 𝐶5′ ; �̇�5/𝐸�̇�5 = �̇�5′ /𝐸�̇�5′ solar air preheater �̇�2′ + �̇�7 + �̇�𝑠𝑎𝑝 = �̇�3 + �̇�8 𝐶7 = 𝐶8; �̇�7/𝐸�̇�7 = �̇�8/𝐸�̇�8 according to table (4), which is presented for 12 exergy streams of the gas turbine with a preheating system based on eq. (10), we need to solve these equations. each component of the system consists of multi-input and multi-output. therefore, the use of equivalent equations is necessary in order to solve the mentioned equations, which is obtained a 12 × 12 matrix. journal of mechanical engineering and technology 8 [𝐴][𝐶] = [𝑍] (10) in which [𝐶] = {�̇�𝑖 = 1, 2, 2 ′, 3, 4, 5, 5′, 6, 7, 8, 9, 10} (11) [𝑍] = {�̇�𝑖 = 1, 2, 2 ′, 3, 4, 5, 5′, 6, 7, 8, 9, 10} the matrix [𝐴], which is the coefficient matrix, is adjusted according to the following matrix. �̇�𝑖 is calculated as follows: [𝐶] = [𝐴]−1[𝑍] (12) 𝑐𝑘 is evaluated according to 𝑐𝑘 = �̇�𝑘 /𝐸�̇�𝑘 . table 5 calculation of 𝒄𝒌 position �̇�𝒌 ($) 𝒄𝒌 ($/kwh) 𝒄𝒌 ($/gj) 1 0 0 0 2 1938.6 0.0148 4.11 2’ 2191.93 0.0141 3.92 journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 2 june – december 2019 9 3 6988.25 0.0428 11.89 4 2704.75 0.0074 2.05 5 737.66 0.0073 1.03 5’ 493.43 0.00729 2.02 6 4300 0.0093 2.58 7 5167.46 0.0038 10.6 8 1738.15 0.0376 10.44 9 1560 0.01 2.77 10 2269.09 0.01 2.77 5. results and discussion the results of all calculations performed in exergy analysis are recorded in tables 6 and 7 for gas turbine with and without solar air preheater. �̇�𝑝 and �̇�𝑓 are the average cost of the exergy of products and fuel of each component of the gas turbine, respectively. according to the data of tables 6 and 7 and the exergo-economic evaluation method, the highest value of �̇�𝐷,𝑘 + �̇�𝑘 is related to the solar air preheater, combustion chamber and gas turbine, which are the most important units to examine and optimize from the point of view of exergo-economic. table 6 calculation of exergo-economic parameters of gas turbine with preheater component �̇�𝒙𝒇 (mj/s) �̇�𝒙𝒑 (mj/s) �̇�𝒙𝑫 (mj/s) 𝒚𝒅 (%) �̇�𝒑 ($/gj) �̇�𝒇 ($/gj) �̇�𝑫 ($/h) �̇�𝒌 ($/h) �̇�𝑫 + �̇�𝒌 ($/h) 𝒓𝒌 (%) 𝜺 (%) 𝒇𝒌 (%) air compressor 156.2 130.98 25.2 6 4.11 2.77 251.3 378.6 629.9 48.3 83.8 60.1 journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 2 june – december 2019 10 air preheater 33.17 24.17 9 2.2 3.92 1.03 33.37 9.11 42.48 28 72.8 21.4 solar air preheater 88.9 7.65 81.05 19.6 11.89 10.6 3092.8 1367 4459.8 12.2 8.6 30.6 combustion chamber 625.23 365.26 259.9 63 2.05 2.58 2413.9 16.5 2430.4 20.5 63.4 0.68 turbine 264.42 225.8 38.6 9.3 2.77 2.05 284.8 302 586.8 35.1 84.1 51.5 total 1167.72 753.86 413.7 100 24.74 19.03 6040.2 2073.2 8149.3 30.1 64.6 47.3 table 7 calculation of exergo-economic parameters of gas turbine without preheater component �̇�𝒙𝒇 (mj/s) �̇�𝒙𝒑 (mj/s) �̇�𝒙𝑫 (mj/s) 𝒚𝒅 (%) �̇�𝒑 ($/gj) �̇�𝒇 ($/gj) �̇�𝑫 ($/h) �̇�𝒌 ($/h) �̇�𝑫 + �̇�𝒌 ($/h) 𝒓𝒌 (%) 𝜺 (%) 𝒇𝒌 (%) compressor 156.2 130.98 25.2 8.8 2.75 1.672 151.7 378 529.7 64.6 83.8 71.4 combustion chamber 593.18 376.26 216.8 76 4.05 5.28 4120.9 17 4137.9 23.3 58.4 0.41 turbine 268.56 225 43.56 15.3 1.68 3.969 622.4 310.9 933.3 57.7 85.4 33.3 total 1017.9 732.24 285.26 100 8.43 10.92 4895 705.9 1480.2 22.3 71.9 44.5 7.1. solar air preheater from the exergo-economic point of view, and tables 7 and 6, the highest �̇�𝐷,𝑘 + �̇�𝑘 occurs in solar air preheater due to high cost of investment and loss of exergy. the efficiency of the system should be investigated with regard to the high exergo-economic factor, 𝑓𝑘 = 30.6. also, it is merely a reduction in the investment cost regarding the lowness of 𝑟𝑘 = 12% in solar air preheater. as a solution, it is proposed to replace the molten salt with a new composition which has the capability of absorption temperature above 860 k, (decomposition point of the molten salt structure). this method reduces the flow rate (molten salt) and finally the dimensions of the journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 2 june – december 2019 11 facilities include storage tanks, heat exchangers, tower and receiver, etc., which results in a reduction in the cost of solar air preheater investment. 7.2. combustion chamber according to tables 6 and 7, the exergo-economic factor of the combustion chamber in two conditions with solar air preheater (𝑓𝑘 = 0.68) and without solar air preheater (𝑓𝑘 = 0.41) indicates that the change in cost of combustion chamber depends on exergy loss variation. increasing the inlet temperature of the combustion chamber is one of the factors which reduces the loss of exergy. it is observed from tables 6 and 7 that the cost of exergy loss without solar air preheater is 4120.9 $/h which is reduced to 2413.9 $/h by using solar preheater. in other words, the cost of exergy loss is reduced 41% by utilizing solar preheater. table 8 results of �̇�𝐷 for combustion chamber unit 𝑻𝟑 (k) �̇�𝑫 ($/h) combustion chamber without preheater with preheater without preheater with preheater 620 820 4120.9 2413.9 the computational values of tables 6 and 7 show that the highest values of �̇�𝐷,𝑘 + �̇�𝑘 are related to combustion chamber with and without solar air preheater, which is the most important unit for the examination and optimization. in the exergo-economics analysis, when 𝑓𝑘 is a small number, it should be tried to improve the efficiency of the components by increasing the cost of investment. in this study, 𝑓𝑘 is very small for combustion chamber either with solar air preheater or without solar air preheater. journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 2 june – december 2019 12 table 9 results of 𝑓𝑘 for combustion chamber unit 𝑻𝟑 (k) 𝒇𝒌 combustion chamber without preheater with preheater without preheater with preheater 620 820 0.41 0.68 in addition to increase of inlet temperature of the combustion chamber, which improves the exergo-economical factor, other methods such as improved fuel nozzle spraying, improved fuel feedback control systems, and control of fuel / air ratio can be used. 7.3. gas turbine the gas turbine with compressors and solar preheating has the least amount of exergy loss. the gas turbine needs to be optimized about 38.6 $/h due to high �̇�𝐷,𝑘 + �̇�𝑘 and 𝑟𝑘 = 35.1%. tables 6 and 7 show that increasing the inlet temperature of the combustion chamber reduces the cost of exergy loss to 54%. also, the cost of exergy loss without solar air preheater is 622.4 $/h which is reduced to 284.8 $/h by using solar air preheater. therefore, the use of solar air preheater improves the exergo-economic factor from 33.3% to 51.5%. it should be tried to reduce the cost of investment and maintenance regarding the high exergo-economics factor 𝑓𝑘 = 51.5% and 𝑟𝑘 = 35.1% for gas turbine with solar air preheater, which depends on inlet flow rate of gas turbine considering the relations of the cost of investment. therefore, reducing the inlet flow rate of the turbine will reduce the cost of exergy of products for gas turbine, which is obtained as follows: 𝐶𝑝,𝑔𝑡 = ( �̇�4 − �̇�5 + �̇�𝑡𝑜𝑡𝑎𝑙 𝑊𝑔𝑡 ) (13) as a second method to reduce the investment cost based on the following equation, a decrease in isentropic efficiency of the turbine (𝜂𝑔𝑡) will diminish investment costs. 𝑃𝐸𝐶𝑔𝑡 = ( 79.34 𝑚𝑔 0.92 − 𝜂𝑔𝑡 ) . ( 𝑃3 𝑃4 ) . (1 + 𝑒𝑥𝑝 (0.036𝑇3 − 56.4)) (14) journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 2 june – december 2019 13 7.4. the effect of using pre-regulator to reduce fuel consumption the fuel exergy in this study, which is considered methane gas, contains two parts of the chemical exergy and physical exergy. however, total fuel exergy is determined based on the following equation. 𝐸𝑥𝑓 = 𝑚𝑓 [𝑐𝑝(𝑇𝑓 − 𝑇0) − 𝑇0𝑐𝑝𝑙𝑛(𝑇𝑓 𝑇0⁄ ) + ∑(𝑥𝑘 �̅�𝑘 𝐶𝐻 ) + �̅�𝑇0 ∑(𝑥𝑘 𝑙𝑛(𝑥𝑘 )) 𝑁 𝑘=1 𝑁 𝑘=1 ] (15) it should be pointed out that the fuel exergy of methane gas is directly related to the mass flow rate of fuel. furthermore, assuming that the output exergy of combustion chamber remains constant, and according to the exergy changes in the combustion chamber, the fuel flow rate decreases from 8.99 to 7.84 kg/s. also, it can be observed from tables (6) and (7) that the exergy efficiency (𝜀) of combustion chamber is increased from 58.4% (without preheater) to 63.4% (with preheater) in the non-progressive state with a pre-igniter. 6. conclusion in this research, engineering equation solver (ees) software and exergo-economic analysis were employed to investigate the economical and exergetical performance of the gas turbine with and without solar air preheater the inlet air entering to the combustion chamber was initially heated by exhaust gas and then used at the stage of a solar power plant, which utilized molten salt to store the extracted energy from the sun. in this mechanism, the air was preheated using a heat exchanger prior to entering the combustion chamber up to 820 k. the results obtained from this study show that by increasing the inlet temperature of the combustion chamber from 620˚k to 820˚k, the exergy factor increases from 0.41% to 0.68% and the cost of exergy loss decreases from 4120.9 $/h to 2413.9 $/h. also, the consumption of gas turbine with solar air preheater is reduced from 8.99 kg/s to 7.84 kg/s by raising the inlet temperature of the combustion chamber. as a result, the exergetic efficiency is enriched from 58.4% to 63.4%. therefore, due to the variety of gas turbines used in the power plants, all available turbines can be assessed based on exergo-economic analysis in order to examine the inefficient points and the sources of exergy loss. journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 2 june – december 2019 14 references [1] kribus, a., zaibel, r., carey, d., segal, a., and karni, j., 1998, "a solar-driven combined cycle power plant," solar energy, 62(2), pp. 121-129. [2] buck, r., abele, m., kunberger, j., denk, t., heller, p., and lüpfert, e., 1999, "receiver for solar-hybrid gas turbine and combined cycle systems," le journal de physique iv, 9(pr3), pp. pr3-537-pr533-544. [3] buck, r., brauning, t., denk, t., pfänder, m., schwarzbözl, p., and tellez, f., 2002, "solarhybrid gas turbine-based power tower systems (refos)," journal of solar energy engineering, 124(1), pp. 2-9. [4] ávila-marín, a. l., 2011, "volumetric receivers in solar thermal power plants with central receiver system technology: a review," solar energy, 85(5), pp. 891-910. [5] augsburger, g., 2013, "thermo-economic optimisation of large solar tower power plants." [6] zhang, h. l., baeyens, j., degrève, j., and cacères, g., 2013, "concentrated solar power plants: review and design methodology," renewable and sustainable energy reviews, 22, pp. 466-481. [7] lozano, m., and valero, a., 1993, "thermoeconomic analysis of gas turbine cogeneration systems," asme, new york, ny,(usa). 30, pp. 311-320. [8] gorji-bandpy, m., and ebrahimian, v., 2006, "exergoeconomic analysis of gas turbine power plants," international energy journal, 7(1). [9]tsatsaronis, g., 2007, "definitions and nomenclature in exergy analysis and exergoeconomics," energy, 32(4), pp. 249-253. [10] gorji-bandpy, m., goodarzian, h., and biglari, m., 2010, "the cost-effective analysis of a gas turbine power plant," energy sources, part b: economics, planning, and policy, 5(4), pp. 348-358. [11] valero, a., lozano, m., serra, l., and torres, c., 1994, "application of the exergetic cost theory to the cgam problem," energy, 19(3), pp. 365-381. [12] gaggioli, r., and el-sayed, y., 1989, "a critical review of second law costing methods—ii: calculus procedures," journal of energy resources technology, 111(1), pp. 8-15. [13] el-sayed, y., and gaggioli, r., 1989, "a critical review of second law costing methods—i: background and algebraic procedures," journal of energy resources technology, 111(1), pp. 1-7. [14] baghernejad, a., and yaghoubi, m., 2011, "exergoeconomic analysis and optimization of an integrated solar combined cycle system (isccs) using genetic algorithm," energy conversion and management, 52(5), pp. 2193-2203. [15] bagdanavicius, a., and jenkins, n., 2014, "exergy and exergoeconomic analysis of a compressed air energy storage combined with a district energy system," energy conversion and management, 77, pp. 432-440. [16] elsafi, a. m., 2015, "exergy and exergoeconomic analysis of sustainable direct steam generation solar power plants," energy conversion and management, 103, pp. 338-347. journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 2 june – december 2019 15 [17] mohammadkhani, f., shokati, n., mahmoudi, s., yari, m., and rosen, m., 2014, "exergoeconomic assessment and parametric study of a gas turbine-modular helium reactor combined with two organic rankine cycles," energy, 65, pp. 533-543. [18] cavalcanti, e. j. c., and motta, h. p., 2015, "exergoeconomic analysis of a solarpowered/fuel assisted rankine cycle for power generation," energy, 88, pp. 555-562. [19] ahmadi, r., pourfatemi, s. m., and ghaffari, s., 2017, "exergoeconomic optimization of hybrid system of gt, sofc and med implementing genetic algorithm," desalination, 411, pp. 76-88. [20] lee, y. d., ahn, k. y., morosuk, t., and tsatsaronis, g., 2018, "exergetic and exergoeconomic evaluation of an sofc-engine hybrid power generation system," energy, 145, pp. 810-822. [21] khaljani, m., khoshbakhti saray, r., and bahlouli, k., 2015, "comprehensive analysis of energy, exergy and exergo-economic of cogeneration of heat and power in a combined gas turbine and organic rankine cycle," energy conversion and management, 97, pp. 154-165. [22] aydin, h., 2013, "exergetic sustainability analysis of lm6000 gas turbine power plant with steam cycle," energy, 57, pp. 766-774. [23] mousafarash, a., and ahmadi, p., 2014, "exergy and exergo-economic based analysis of a gas turbine power generation system," progress in sustainable energy technologies vol ii, springer, pp. 97-108. [24] mousafarash, a., and ameri, m., 2013, "exergy and exergo-economic based analyses of a gas turbine power generation system," journal of power technologies, 93(1), p. 44. [25] igbong, d., and fakorede, d., 2014, "exergoeconomic analysis of a 100 mw unit ge frame 9 gas turbine plant in ughelli, nigeria," international journal of engineering and technology, 4(8), pp. 463-468. [26] yue, t., and lior, n., 2017, "exergo economic analysis of solar-assisted hybrid power generation systems integrated with thermochemical fuel conversion," applied energy, 191, pp. 204-222. [27] ahmadi, p., and dincer, i., 2011, "thermodynamic and exergoenvironmental analyses, and multi-objective optimization of a gas turbine power plant," applied thermal engineering, 31(14), pp. 2529-2540. journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 2 june – december 2019 16 issn: 2180-1053 vol. 10 no.2 june – december 2018 39 thermal performance analysis of nano enhanced paraffin wax and myristic acid ajay m nair 1* , p vinod kumar naidu 2 , rajeev kukreja 3 1,2,3 department of mechanical engineering, dr.b.r ambedkar national institute of technology, jalandhar, india abstract in the present study, paraffin wax and myristic acid were chosen as phase change materials for the analysis. further, nanoparticles of copper and micro particles of aluminium and graphite were used for the analysis of enhancement in thermal performance of phase change materials. charging and storing period analysis have also been conducted to compare the performance of different composites. from the charging period analysis, it has been found that graphite composites are performing better than copper and aluminium composites. from the both charging and storing period analysis, among composites of paraffin wax, both 1% copper and 1% graphite composites are giving more satisfactory results than all other composites of paraffin wax and among composites of myristic acid, it can be concluded that graphite composites are giving more satisfactory results than all other composite, in which 3% graphite is the optimum composition. keywords: phase change materials, energy storage, nano composite 1.0 introduction the standard of living in a society is measured by the amount energy consumed. the development of an efficient non-conventional energy harvesting system is an important approach to reduce the climatic changes and dependence on conventional fuels. since the availability of some the main renewable energy sources is not continuous and the demand and the supply of the energy normally does not occur simultaneously,there is need of efficient thermal energy storage system. phase change materials (pcms) are the better option to store these excess energy when their availability is more can be made use when demand arises. abhat (1993) compared the performance of sensible heat and latent heat thermal energy storage, concluded that pcm can achieve a significant reduction in storage volume. bjurstrom, carlsson (1985) and adebiyi, russell (1987) studied on charging of pcm and concluded that when the melting temperature of pcm is equal to the geometric mean of the ambient temperature and temperature of htf, maximum exergy would be recovered from htf. velraj et al. (1999) performed an experiment on heat transfer *corresponding author e-mail: ajaymnair4491@gmail.com journal of mechanical engineering and technology 40 issn: 2180-1053 vol. 10 no.2 june – december 2018 enhancement in lhtes using bubble agitation in this they poured small amount of water in a tube which was evacuated so that the saturation temperature of water is equal to the phase change temperature of pcm, this help in heat transfer enhancement by creating steam bubbles. hoshi et al. (2005) classified the storage materials based on their melting points, such as low temperature materials (mp<220°c), medium temperature materials (mp up to 420°c) and high temperature materials (mp>420°c). zalba et al. (2003) compared the properties of organic and inorganic pcms and found that high latent heat as the positive aspect and least corrosion resistance, highly thermal/ chemical instability are negative aspects of inorganic pcms. lecuona et al. (2013) performed experiments on portable solar cooker of the standard concentrating parabolic type by incorporating technical grade paraffin and erythritol as pcms. al-kayie et al. (2014) presented effects of various inclination (10,20 and 30) on solar water heater integrated with a pcm nanocomposite (without tes , with paraffin wax pcm and with copper nano composite of paraffin wax) and results indicated that there was a considerable enhancement in performance of the system with paraffin wax, while in the case of nanocomposite no significant enhancement was observed. mahfuz et al. (2014) analysed the performance of paraffin wax based pcm thermal energy storage shell and tube water heating system by the exergy, energy and cost analyses. the charging of the paraffin wax pcm, paraffin wax was done during the day time. the energy, exergy efficiencies of the thermal storage system and total life cycle cost were developed for different flow rate of the heat transfer fluid. chaichan and kazem (2015) investigated the use of phase change material for extracting solar energy from the concentrated solar heater for water distillation. the solar energy sored in the paraffin wax pcm during the day time is used for the distillation purpose which increased the distillation efficiency of the system and there by the productivity increased by about 180%. zhao et al. (2013) investigated the high temperature energy storage using encapsulated phase change materials for storing solar energy and conducted transient two-dimensional heat transfer analysis. nano3 encapsulated with stainless steel in a cylindrical shaped capsule is used as phase change material. elmozughi et al. (2014) did the analysis of encapsulated phase change materials by considering 20% void and buoyancy driven convection in a stainless steel capsule. the effect of thermal and volume expansions of the potential pcm sodium nitrate during charging and discharging were evaluated. memona et al. (2015) prepared macro-encapsulated paraffin light weight aggregate for the development of normal weight aggregate concrete by the introduction of paraffin into porous light weight aggregate through vacuum impregnation. sancheza et al. (2015) did studies on a thermal storage system for solar power plant consists of a thermocline tank with pcm capsules together with filler materials. it was based on the principle of multilayered solid-pcm thermocline. the encapsulation was done by developing an external coating over a pcm pellet. cai et al. (2011) studied the effect of nano sio2 addition on the morphology, thermal energy storage, thermal stability and combustion properties of electro spun lauric acid/ polyethylene terephthalate (pet) ultrafine composite fibres as form stable phase change materials. the results showed that the la/petsio2 composite fibres with nano-sio2 had desired morphologies with reduced average fibre diameters as compared to the la/pet fibres without nano-sio2 due to the increased conductivity of the spin dopes and the strong hydrogen bonding among the components in the fibres and the incorporation of nano-sio2 increased the thermal degradation temperature. li (2013) analysed the effect of addition of nano graphite on the thermal conductivity of paraffin based pcm. the dispersed nano graphite enhanced the heat transfer and the thermal performance analysis of nano enhanced paraffin wax and myristic acid issn: 2180-1053 vol. 10 no.2 june – december 2018 41 charging performance in terms of efficiency of pcm. yang et al. (2014) explored the effect of nano si3n4 on the thermal behaviour of si3n4/ paraffin composites. the melting processes were monitored by 3d optical microscopy. transient hot-bridge method (thb-instrument, linseis, inc.) was approached to measure the thermal conductivities. with the nano-si3n4 addition the thermal conductivity of the pcm was enhanced by 35% and the thermal diffusivity of the pcm by 47%. jiang et al. (2015) did the synthesis, characterisation and analysis of the morphology, phase change behaviour, thermal energy storage and thermoregulation performance, thermal stability, as well as heat-transfer properties of paraffin micro capsules modified with nano al2o3. nourani et al. (2016) investigated the phase change temperature, heating rate, latent heat and effective thermal conductivity of paraffin – nano –al2o3 stabilized by sodium stearoyl lactylate as a stable phase change material with a high thermal conductivity for application in solar energy storage systems where sodium stearoyl lactylate was used as a surfactant. . the results showed that the effective thermal conductivity of the pcms thus prepared increased by increasing nano-al2o3 mass fraction, showing a nonlinear relationship with mass fraction in both the solid and liquid states. sharma et al. studied the thermal behaviour of composite of palmitic acid and tio2 nano particles for thermal energy storage system and the thermal energy storage behaviour, morphology, chemical compatibility and other thermal properties of these composites were investigated. in the present work we have chosen paraffin wax and myristic acid as pcm. since the melting temperature of both the materials are above 50, these are suitable for applications like solar water heating, solar stills, solar dryers, solar air heaters etc. but the low thermal conductivity (0.2w/mk) hampers its application, though both have very high latent heat of fusion. the addition of nano particle can overcome the shortcomings of the selected pcm, like low thermal conductivity and supercooling. in the present work, designing and fabrication of an economical system which uses low melting point pcms for storing energy and analysing the effect of addition of different nano sized metallic particles in the charging and discharging period of pcm. 2.0 methodology 2.1 experimental setup the experimental apparatus consists of insulated hot water bath, hot water pump, parabolic solar concentrator with copper water circulation coil, copper helical coil heat exchanger, pcm cylinder, data acquisition system (agilent 34972a lxi), t type thermocouples and insulation tanks for storage. the main components of the experimental setup are shown in its pictorial view in figure 1(a&b). water was stored in hot water tank which is having multi-layer insulation such as black spray paint, cardboard, thermocol covered with aluminium foil and black sheet. water was pumped into the coil attached with solar concentrator using the hot water pump (model star rs25/6 wilo company germany, capacity 800 -3400 litre per hour, head of 2-6 m) connected to the tank. parabolic solar concentrator fabricated from aluminium sheet and multi loop 3/8 in copper coil was placed just below the focus (focal length of 50 cm) so that the entire area of the coil was covered by the concentrated solar radiation. the hot water was made to flow through the helical copper coil heat exchanger constructed from copper (1/4” diameter) tube having outer diameter 8cm and length journal of mechanical engineering and technology 42 issn: 2180-1053 vol. 10 no.2 june – december 2018 10cm, it was placed in a borosilicate 1 l container containing pcm (paraffin wax and myristic acid composites) is shown in figure 2(a), this container along with heat exchanger was kept in insulation box insulated similar as that of hot water tank finally the water flow back to hot water tank from the heat exchanger. three insulation boxes were constructed for the storage of the charged (a) (b) figure 1. parabolic collector (a) and the photograph of the experimental set-up (b) thermal performance analysis of nano enhanced paraffin wax and myristic acid issn: 2180-1053 vol. 10 no.2 june – december 2018 43 table1. properties of paraffin wax and myristic acid material melting point(°c) latent heat of fusion(kj/kgk) thermal conductivity(w/mk) density(kg/m³) paraffin wax 58-60 189 0.21 (s) 795 (l, 70°c) 920 (s,20°c) myristic acid 52-54 204 0.17 (s) 861 (l, 55°c) 990 (s,24°c) pcms is shown in figure 2(b). two t type thermocouples were fixed, one at the inside of the coil and other at the periphery of the container. pcms used are the paraffin wax and myristic acid. micro particles used are the copper, graphite and aluminium, with compositions 1%, 3%, and 5%. the properties of pcms and nano particles used in the present study are given in table 1 and table 2 (a) (b) figure 2. insulated charging box with copper coil heat exchanger and pcm container (a) and insulated storage box with sample for storing (b). table 2. properties of added materials to paraffin wax and myristic acid material size thermal conductivity(w/mk) purity (%) copper 30-50nm 350-380 99.5 graphite 400-1200nm 150-180 99.8 aluminium 40µm 200-240 98 journal of mechanical engineering and technology 44 issn: 2180-1053 vol. 10 no.2 june – december 2018 2.2 experimentation procedure 500g of required pcm was kept in hot water bath to ensure the complete melting, this molten pcm was taken out of the hot water and kept on the magnetic stirrer followed by ultra sonicator for around 3-4 hours. the required quantities of micro particles are added to the molten pcms, later it was removed from the sonicator and kept in cold water bath to solidify. after ensuring the complete solidification of the composites, it is removed from the container and grinded into powder with the help of dry grinder then the grinded powder is kept in the charging container. twenty composites samples of paraffin wax and myristic acid are shown in figure 3(a)-3(t). fill the powdered sample in the container kept in the storage box and ensure that the thermocouple positions are maintained. connect the data acquisition system to the laptop, with the keysight connection expert software make the interface between data acquisition system and laptop. then launch the keysight benchvue software to select the thermocouple whose values are to be recorded. in order to have similar operating condition water from hot water tank is pumped to the coil kept at the focus of solar concentrator and this water is bypassed to tank without allowing it to pass through test coil. the procedure will continue till the required operating condition is reached. once the required temperature is attained the water from solar coil is made to pass through the test coil till the pcm is completely melted. after the complete melting the pcm container is removed from the charging box and kept in storage box and the pcm temperature is recorded for eight hours. (a) pure paraffin wax (b) 1% copper (c) 3% copper (d) 5% copper (e) 1% al (f) 3% al (g) 5% al (h) 1% graphite thermal performance analysis of nano enhanced paraffin wax and myristic acid issn: 2180-1053 vol. 10 no.2 june – december 2018 45 (i) 3% graphite (j) 5% graphite (k) pure myristic acid (l)1% graphite (m) 3% graphite (n)5% graphite (o) 1% cu (p) 3% cu (q) 5% cu (r)1% al (s) 3% al (t) 5% al figure 3. samples of paraffin wax and its composites (a) to (j), and myristic acid and its composites (k) to (t). journal of mechanical engineering and technology 46 issn: 2180-1053 vol. 10 no.2 june – december 2018 3.0 results and discussions 3.1 charging period analysis 3.1.1 pure paraffin wax and pure myristic acid during experimentation it is found that pure paraffin wax took 3h 5min and pure myristic acid took 2h 40min for complete charging. the variation of pcm temperatures with time during charging of pure paraffin wax and puremyristic acid is shown in figure 4(a)-4(b). (a) (b) figure 4 (a). charging time vs pcm temp of pure paraffin wax (b).charging time vs pcm temp of pure myristic acid 3.1.2 with graphite micro particles thermal performance analysis of nano enhanced paraffin wax and myristic acid issn: 2180-1053 vol. 10 no.2 june – december 2018 47 the variation of pcm temperatures with time during charging of different compositions of graphite micro composites of paraffin wax andmyristic acid is shown in figure 5(a)5(f). (a) (b) (c) journal of mechanical engineering and technology 48 issn: 2180-1053 vol. 10 no.2 june – december 2018 (d) (e) (f) thermal performance analysis of nano enhanced paraffin wax and myristic acid issn: 2180-1053 vol. 10 no.2 june – december 2018 49 figure 5 (a) charging time vs pcm temp of 5g graphite added in paraffin wax (b) in myristic acid(c), of 15g graphite added in paraffin wax (d) of 15g graphite added in myristic acid(e) charging time vs pcm temp of 25g graphite added in paraffin wax (f) charging time vs pcm temp of 25g graphite added in myristic acid on addition of 1% graphite micro particles to pure paraffin wax,the charging time duration is reduced to 2 h 15 min , with 3% graphite micro particles it took 2h 7 min and with 5% graphite micro particles found to be completely melted in 2h 2 min. in the case of myristic acid, 1% graphite composite took 2h 12 mins for complete melting and 3% graphite composite charging time is found to be 1h 56 min and for 5% graphite composite of myristic acid it was 1h 44 min. since the density of graphite micro particles is low, it dispersed uniformly during the stirring and it is found that deposition rate of graphite micro particles is less and moreover high thermal conducting graphite particles made the composite conductive. 3.1.3 with aluminium micro particles the variation of pcm temperatures with time during charging of different compositions of aluminium micro composites of paraffin wax and myristic acid is shown in figure 6(a)-6(f) (a) journal of mechanical engineering and technology 50 issn: 2180-1053 vol. 10 no.2 june – december 2018 (b) (c) (d) thermal performance analysis of nano enhanced paraffin wax and myristic acid issn: 2180-1053 vol. 10 no.2 june – december 2018 51 (e) (f) figure 6 charging time vs pcm temp of 5g al added in paraffin wax (a) charging time vs pcm temp of 5g al added in myristic acid (b) charging time vs pcm temp of 15g al added in paraffin wax (c) charging time vs pcm temp of 15g al added in myristic acid (d) charging time vs pcm temp of 25g al added in paraffin wax (e) charging time vs pcm temp of 25g al added in myristic acid(f) considering aluminium micro particles, on addition of 1% al micro particles to the pure paraffin wax, the charging duration is reduced to 2h 30 min which is not as significant as when compared to that of 3% al micro particles which melted in 2h 5 min and the charging duration of 5% al micro particles added paraffin wax is 2h 7min. beyond 3% there is no much reduction in charging duration due to the significant deposition of al micro particles. from the analysis of 1%3%5% of al composites of myristic acid it is found that , by addition of 1% al micro particles the composite melted in 2h 13min but 3% and 5% al composites were melted in 2h 22min and 2h 31 min respectively. it is found that beyond 1% of al micro particles sedimentation rates are more because the extra added particles might have disturbed already dispersed particles and made them to settle down. 3.1.4 with copper nano particles from the analysis copper composites of paraffin wax it is found that 1% copper nanoparticles added to the pure paraffin wax resulted in decrease in the charging duration i.e. to 2h 25 min but when 3% copper nano particles were added, there was a journal of mechanical engineering and technology 52 issn: 2180-1053 vol. 10 no.2 june – december 2018 considerable enhancement in charging duration to 2 h 6 min and with 5% copper nanocomposite there was no significant reduction in the charging duration i.e. it took 2h for complete charging. when the composition is increased from 1% to 3% there is considerable reduction in charging time due to increase in the conductivity of the composite whereas when it is increased to 5% due to deposition and increase in the viscosity of the composite there was no much reduction in charging duration.the variation of pcm temperatures with time during charging of different compositions of copper nano composites of paraffin wax and myristic acid is shown in figure 7(a)7(f) (a) (b) thermal performance analysis of nano enhanced paraffin wax and myristic acid issn: 2180-1053 vol. 10 no.2 june – december 2018 53 (c) (d) (e) journal of mechanical engineering and technology 54 issn: 2180-1053 vol. 10 no.2 june – december 2018 (f) figure 7. charging time vs pcm temp of 5g cu added in paraffin wax (a) charging time vs pcm temp of 5g cu added in myristic acid (b) charging time vs pcm temp of 15g cu added in paraffin wax (c) charging time vs pcm temp of 15g cu added in myristic acid (d) charging time vs pcm temp of 5g cu added in paraffin wax (e) charging time vs pcm temp of 5g cu added in myristic acid (f) the results of study of 1%, 3%, 5% copper composite of myristic acid showed that 1% composite melted in 2h 4min but in the case of both 3% and 5% of copper composite there was a significant increase in viscosity of the composite due to which there was no proper phase change was found even after 3h 30min of observation. the paste form of 15g and 25g cu composite is shown in figure 8(a) and 8(b). (a) (b) figure 8. 25g cu added myristic acid after 3h 30 minutes (a) 25g cu myristic acid paste (b). 3.2 storing period analysis 3.2.1 pure paraffin wax and pure myristic acid thermal performance analysis of nano enhanced paraffin wax and myristic acid issn: 2180-1053 vol. 10 no.2 june – december 2018 55 for storing period analysis different samples were kept in the storage box for 8h and the variation in temperature was analysed. the temperature drop of pure paraffin wax during 8h of storage was found to be 12.578°c while the temperature drop for pure myristic acid was observed to be 10°c.the variation of pcm temperatures with time during storing of pure paraffin wax and myristic acid is shown in figure 9(a)-9(b). 3.2.2 with graphite micro particles during the storage period analysis of graphite composites of paraffin wax, for 1% graphite added paraffin wax 13.406°c temperature drop was found but wherein 3% added paraffin wax it was about 16.349°c and for 5% graphite composite the drop was 16.856°c. the increase in temperature drop for 1% to 5% graphite added paraffin wax revealed the enhancement of thermal conductivity. (a) (b) journal of mechanical engineering and technology 56 issn: 2180-1053 vol. 10 no.2 june – december 2018 figure 9 storing time vs pcm temp pure paraffin wax (a) storing time vs pcm temp of pure myristic acid (b) in the case of 1% and 3% graphite added composite, temperature drop was around 11°c in eight hours and 5% graphite added myristic acid composite showed a temperature drop of 13°c. the reduced performance of graphite composite of myristic acid is due to the increased thermal conductivity of the composites.the variation of pcm temperatures with time during charging of different compositions of graphite micro composites of paraffin wax and myristic acid is shown in figure 10(a)-10(f). (a) (b) thermal performance analysis of nano enhanced paraffin wax and myristic acid issn: 2180-1053 vol. 10 no.2 june – december 2018 57 (c) (d) journal of mechanical engineering and technology 58 issn: 2180-1053 vol. 10 no.2 june – december 2018 (e) (f) figure 10 storing time vs pcm temp of 5g graphite added in paraffin wax (a) storing time vs pcm temp of 5g graphite added in myristic acid (b) storing time vs pcm temp of 15g graphite added in paraffin wax (c) storing time vs pcm temp of 15g graphite added in myristic acid (d) storing time vs pcm temp of 25g graphite added in paraffin wax (e) storing time vs pcm temp of 25g graphite added in myristic acid (f). 3.2.3 with aluminium micro particles storage period analysis of aluminium added paraffin wax composites showed that 1% composite was having a drop of 15.159°c while 3% composite showed 17.543°c and the temperature drop of 20.38°c was for 5% composite. the storage period analysis of al composite of myristic acid revealed that for 1% al composite 13°c temperature drop was observed and 17.131 °c for 3% composite whereas in case of 5% composite 19.341°c temperature drop was observed. the variation of pcm temperatures with time during charging of different compositions of aluminium micro composites of paraffin wax and myristic acid is shown in figure 11(a)-11(f) thermal performance analysis of nano enhanced paraffin wax and myristic acid issn: 2180-1053 vol. 10 no.2 june – december 2018 59 (a) (b) (c) journal of mechanical engineering and technology 60 issn: 2180-1053 vol. 10 no.2 june – december 2018 (d) (e) (f) figure 11 storing time vs pcm temp of 5g al added in paraffin wax (a) storing time vs pcm temp of 5g al added in myristic acid(b) storing time vs pcm temp of 15g al added in paraffin wax (c) storing time vs pcm temp of 15g al added in myristic acid (d) (e) storing time vs pcm temp of 25g al added in paraffin wax (f) storing time vs pcm temp of 25g al added in myristic acid 3.2.4 with copper nano particles thermal performance analysis of nano enhanced paraffin wax and myristic acid issn: 2180-1053 vol. 10 no.2 june – december 2018 61 in copper nanocomposites, the temperature drop of 16.024°c was found in 1% copper added paraffin wax and it was around 18.097°c for 3% composite whereas in 5% composite temperature drop was found to be less than that of 3% composite i.e. 16.349°c the storage period analysis of copper composite of myristic acid, it was found that a temperature drop of 13°c for 1% composite and for 3% composite it was around 17°c and for 5% composite the 30°c temperature drop were observed. this large increase in the drop might be due to improper phase change from solid to liquid. the variation of pcm temperatures with time during charging of different compositions of copper nano composites of paraffin wax and myristic acid is shown in figure 12(a)-12(f) (a) journal of mechanical engineering and technology 62 issn: 2180-1053 vol. 10 no.2 june – december 2018 (b) (c) (d) thermal performance analysis of nano enhanced paraffin wax and myristic acid issn: 2180-1053 vol. 10 no.2 june – december 2018 63 (e) (f) figure 12 (a) storing time vs pcm temp of 5g cu added in paraffin wax (b) storing time vs pcm temp of 5g cu added in myristic acid (c) storing time vs pcm temp of 15g cu added in paraffin wax (d) storing time vs pcm temp of 15g cu added in myristic acid (e) storing time vs pcm temp of 25g cu added in paraffin wax (f) storing time vs pcm temp of 25g cu added in myristic acid 4.0 conclusions the performance analysis of pure paraffin wax and pure myristic acid was done and compared with that of the various compositions of nanocomposites of paraffin wax and myristic acid. the conclusions of present work are summarized as follows  from the charging period analysis of graphite composites, there was no significant enhancement in the performance by adding beyond 1% graphite to the paraffin wax. whereas in the case of myristic acid there was a clear enhancement in thermal performance from 1% to 5% of graphite composites.  charging period analysis of aluminium composites, revealed that 3% aluminium added paraffin wax is better performer than 1% and 5% composites and whereas in aluminium composite of myristic acid 1% composite was found have shorter charging duration than remaining.  from the experimental analysis, 5% copper paraffin wax composite showed marginal better charging performance than 1% and 3%.in the case of copper composites of myristic acid, there is no significant enhancement in performance by adding beyond 1% of copper to pure myristic acid. journal of mechanical engineering and technology 64 issn: 2180-1053 vol. 10 no.2 june – december 2018  from the storage period analysis of graphite composite 1% graphite composite is marginally better performer than other two composites of paraffin wax. among myristic acid composites, 1% and 3% composites have almost similar enhancement in performance and that was better than 5%.  storage period analysis of aluminium composites of both paraffin wax and myristic acidrevealed that 1% aluminium added composite was best in storing energy than3% and 5% composite. among 3% and 5% composite, 3% composite is marginally better performer.  among the copper composite of paraffin wax and myristic acid, 1% copper composite showedbetter performance with significantly less temperature drop compared to that of 3%and 5%.  from the both charging and storing period analysis, among composites of paraffin wax, both 1% copper and 1% graphite composites are giving more satisfactory results than all other composites of paraffin wax and among composites of myristic acid, it can be concluded that graphite composites are giving more satisfactory results than all other composite, in which 3% graphite is the optimum composition. 5.0 references abhat, a. (1983). low temperature latent heat thermal energy storage: heat storage materials. solar energy, 30(4), 313-332. bjurström, h., & carlsson, b. (1985). an exergy analysis of sensible and latent heat storage. journal of heat recovery systems, 5(3), 233-250. adebiyi, g. a., & russell, l. d. (1987). a second law analysis of phase-change thermal energy storage systems. asme htd, 80, 9-20. velraj, r. v. s. r., seeniraj, r. v., hafner, b., faber, c., & schwarzer, k. (1999). heat transfer enhancement in a latent heat storage system1. solar energy, 65(3), 171180. hoshi, a., mills, d. r., bittar, a., & saitoh, t. s. (2005). screening of high melting point phase change materials (pcm) in solar thermal concentrating technology based on clfr. solar energy, 79(3), 332-339. zalba, b., marın, j. m., cabeza, l. f., & mehling, h. (2003). review on thermal energy storage with phase change: materials, heat transfer analysis and applications. applied thermal engineering, 23(3), 251-283. lecuona, a., nogueira, j. i., ventas, r., & legrand, m. (2013). solar cooker of the portable parabolic type incorporating heat storage based on pcm. applied energy, 111, 1136-1146. al-kayiem, h. h., & lin, s. c. (2014). performance evaluation of a solar water heater integrated with a pcm nanocomposite tes at various inclinations. solar energy, 109, 82-92. thermal performance analysis of nano enhanced paraffin wax and myristic acid issn: 2180-1053 vol. 10 no.2 june – december 2018 65 mahfuz, m. h., anisur, m. r., kibria, m. a., saidur, r., & metselaar, i. h. s. c. (2014). performance investigation of thermal energy storage system with phase change material (pcm) for solar water heating application. international communications in heat and mass transfer, 57, 132-139. chaichan, m. t., & kazem, h. a. (2015). water solar distiller productivity enhancement using concentrating solar water heater and phase change material (pcm). case studies in thermal engineering, 5, 151-159. zhao, w., elmozughi, a. f., oztekin, a., & neti, s. (2013). heat transfer analysis of encapsulated phase change material for thermal energy storage. international journal of heat and mass transfer, 63, 323-335. elmozughi, a. f., solomon, l., oztekin, a., & neti, s. (2014). encapsulated phase change material for high temperature thermal energy storage–heat transfer analysis. international journal of heat and mass transfer, 78, 1135-1144. memon, s. a., cui, h. z., zhang, h., & xing, f. (2015). utilization of macro encapsulated phase change materials for the development of thermal energy storage and structural lightweight aggregate concrete. applied energy, 139, 4355. muñoz-sánchez, b., iparraguirre-torres, i., madina-arrese, v., izagirre-etxeberria, u., unzurrunzaga-iturbe, a., & garcía-romero, a. (2015). encapsulated high temperature pcm as active filler material in a thermocline-based thermal storage system. energy procedia, 69, 937-946. cai, y., ke, h., dong, j., wei, q., lin, j., zhao, y., & fong, h. (2011). effects of nano-sio2 on morphology, thermal energy storage, thermal stability, and combustion properties of electrospun lauric acid/pet ultrafine composite fibers as form-stable phase change materials. applied energy, 88(6), 2106-2112. li, m. (2013). a nano-graphite/paraffin phase change material with high thermal conductivity. applied energy, 106, 25-30. yang, y., luo, j., song, g., liu, y., & tang, g. (2014). the experimental exploration of nano-si3n4/paraffin on thermal behavior of phase change materials. thermochimica acta, 597, 101-106. jiang, x., luo, r., peng, f., fang, y., akiyama, t., & wang, s. (2015). synthesis, characterization and thermal properties of paraffin microcapsules modified with nano-al2o3. applied energy, 137, 731-737. nourani, m., hamdami, n., keramat, j., moheb, a., & shahedi, m. (2016). thermal behavior of paraffin-nano-al2o3 stabilized by sodium stearoyl lactylate as a stable phase change material with high thermal conductivity. renewable energy, 88, 474-482. journal of mechanical engineering and technology 66 issn: 2180-1053 vol. 10 no.2 june – december 2018 journal of mechanical engineering and technology *corresponding author. email: azli@utem.edu.my issn 2180-1053 vol. 11 no. 1 july – december 2019 8 mechanical and electrical characterization of nanocomposites liquidsolid conductive ink on polyethylene terephthalate (pet) substrate norhisham ismail1, , mohd azli salim1* adzni md saad1, nor azmmi masripan1 and ghazali omar1 1universiti teknikal malaysia melaka, hang tuah jaya, durian tunggal, melaka, malaysia abstract with drastic development of wearable electronics have urged the studies on the conductive ink and flexible substrate. wearable electronics consist of nanocomposites liquid-solid conductive ink and flexible substrate such as polyethylene terephthalate (pet). they were produced by using stencil printing method. this paper presents the mechanical and electrical characteristics of conductive ink with unloaded condition. the conductive ink was printed with four patterns, which were straight, curve, square and zig-zag patterns. then, all four patterns were tested for their surface morphology, surface roughness, sheet resistivity and bulk resistivity. surface morphology showed that conductive ink with 3 mm width had less granular particle formed than conductive ink with 1 mm width. surface roughness of conductive ink with 3 mm width was smoother compared to 2 mm width and 1 mm width. sheet resistivity and bulk resistivity results indicated that resistivity of all four patterns decreased with the increase of the conductive ink width. from the result, it showed that conductive ink with straight pattern has the best performance. meanwhile, individual result for each pattern had its own function inside the circuit track. keywords: liquid-solid conductive ink, mechanical and electrical characteristic, polyethylene terephthalate, stencil printing method, wearable electronics. 1. introduction nanocomposites liquid-solid conductive ink had become the focus and mostly researched material nowadays because it can update and enhance the properties and multi-functionalities of existing material. with the combination of flexible substrate, this advanced material technology also benefits electronic industry like electronic packaging, flexible display, wearable electronic, clothing, and sensor [1][2] which most of them need integration on surface, micro size and thin thickness devices. these flexible and wearable devices use substrate that allows roll-up, roll-toroll, stretchable, twist and bend behavior. by replacing the thick and hard substrate, the advantages of flexibility, lightweight, and toughness can be achieved with flexible substrate [3][4]. polyethylene terephthalate (pet) is one of the flexible material that suitable to be used as a substrate. pet is a strong, stiff and belongs to the group in the polyester family of polymers [5]. they aree highly resistance to deformation which is hard to wrinkle. pet is usually used in durable-press blends with other fibers, which provides the fiber or fabric to recover from wrinkling. in other case, with slightly higher molecular weight, high-strength plastic can be produced by all the other thermoplastic methods. liquid-solid conductive ink is material that converts insulative polymers into conductive material. there are several types of conductive filler that have been developed such as metalbased inks, carbon complexes, and conductive polymers [6] insulative polymers usually can be grouped into thermosetting resins and thermoplastic resins, where they were formed in threedimensional, which won’t melt again even they were heated repeatedly. whereas, the later are chain polymers, which can melt again after heated even after molded. this prove that thermoplastic resins can be used to thermally bond the nanofiller and insulative polymer. direct bonding [7] can be used to bond nanofiller and insulative polymer together using chemical and thermal process method. chemical bonding method modifies the bonding surfaces into contact journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 1 july – december 2019 9 chemically and then bonding them together. for thermal bonding method, it uses thermal to softening and welds the contact surfaces and then cooling it after contact surfaces are welded. insulative polymer were added with conductive nanofiller until the conductive network was formed inside the polymer matrix. in other case, with randomly dispersed and high loading of conductive nanofiller may not form a network between insulative polymers and conductive nanofiller, which causing the material to become relatively weak brittle structure [3], [8]. also, there are quite a challenge to control and manipulate the nanoparticles in order to produce remarkable and effective electric properties [9]. however, nanocomposites show excellent properties, where they have excellent high surface to volume ratio of the nanofiller or high aspect ratio, exceptional strength and toughness, also electrical and thermal conductivity, combine with their low cost and easy of processability[6]. electrical conductivity that can fulfill the needs of the application and mechanical properties of the ink were important [10]. so the ink must be defect free such as pore, delamination, porosity and crack, and had good adhesion between substrate and ink [8], [10]. in this paper, four patterns of conductive ink were printed on pet substrate with three different track width such as 1 mm, 2 mm, and 3 mm. these patterns were commonly used inside circuit track. after that, the samples were tested with several tests such as surface morphology, surface roughness, sheet resistivity, and bulk resistivity to characterize their mechanical and electrical characteristics under static condition or unloaded condition. from the test, the characteristic for each pattern and effect of the parameter can be observed and analyzed. 2. experimental arrangement surface morphology, surface roughness, bulk resistance, and sheet resistance of prepared samples were measured with respective devices. 2.1 sample preparation stencil printing method was used as printing method to print the samples on polyethylene terephthalate (pet) substrate. this method required the conductive ink to be direct-write on top of the stencil to acquire the desired shape as figure 1. carbon based liquid-solid ink was used as the material and printed into four different test patterns as shown in figure 1. these test patterns were chosen by considering of many shapes and turns inside an electronic circuit track. figure 1: conductive ink’s test patterns 9 cm 2 cm test pattern 1 test pattern 3 test pattern 2 test pattern 4 journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 1 july – december 2019 10 in this project, test pattern 1 was straight line pattern, test pattern 2 was curve pattern, test pattern 3 was square pattern, and test pattern 4 was zig-zag pattern. they were directly printed with stencil jig. all the patterns were designed with 9 cm x 2 cm of length, and width, and 1 mm of thickness as in figure 1. each test pattern was printed with track width of 1 mm, 2 mm, and 3 mm, and cured at room temperature for at least 15 minutes. pet was used as ink substrate due to its flexibility, good chemical resistance to alkalis and good mechanical properties such as stiffness, absorb very small amount of water, and strong with less cost and thin thickness [11]. also, the crystallinity of the pet varies from amorphous to fairly high crystalline. four test patterns that had been printed will undergo a few characterization tests to characterize their mechanical and electrical characteristics such as surface morphology, surface roughness, bulk resistivity, and sheet resistivity. in order to determine the testing point and to ensure consistency of the data, all the samples were marked at three points like figure 2 and all the tests was done on the marked point. figure 2: measurement point on the sample 2.2 measurements sample’s mechanical characteristic such as surface morphology and surface roughness, and electrical characteristic such as bulk resistivity and sheet resistivity were observed and measured with 3d non-contact profilometer, multimeter and four-point probe. all these tests were done on the samples with unloaded condition. 2.2.1 surface morphology and surface roughness surface morphology and surface roughness were measured with 3d non-contact profilometer, which has interference microscopes with wavelength of light as the ruler to measure height variations like surface roughness with high precision. it can be used for single point, a line, and three dimensional scan. this profilometer measured the optical path differences of the samples, which were the height variances on the samples surface. in this experiment, sample was put on the profilometer platform and pinned with tape to keep it from shaking and remain stationary and flat during 3d scanning process and image taking process. the marked point as in figure 2 was focused with fine focus with 20x magnification objective lens. 3d non-contact profilometer scanned and measured height, which was z-axis over an area of x and y. lateral dimensions, which the lowest height of focus area was set as start measuring point and the highest height of focus area as end of measuring point. with the help of software, which was connected live with the profilometer, it showed surface condition of focus area including peaks, valleys, steps, voids, and flat surface and all the images were collected into table 1. for surface roughness or arithmetic average height, ra was measured by continuing the surface morphology process with 3d non-contact profilometer. after the focus area was scanned, the 3d image was transferred and projected inside 3d profile section. the analysis process was done in 3d profile as in figure 3, where 3d roughness profile calculates an area of the scanned surface instead of a single line. generally, inside the software, the 3d profile was divide 2 cm 4.5 cm 7 cm p1 p2 p3 journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 1 july – december 2019 11 number of sections over one sampling length, ɩ, from this, the parameters of several 2d roughness profiles that projecting a number of consequents profiles from surface was calculated for each sections. when 3d profile was scanned, consequent profiles were projected with the direction of x-axis and magnification at y-axis. there was mean line between peaks and valleys of the profile. in 2d roughness profile, area of peaks and valleys from mean line over each section, yi, were taken. then, the roughness was obtained when average of parameter for each section was taken. therefore, mathematical definition and digital implementation were shown as below; 𝑅𝑎 = 1 𝑙 ∫ |𝑦(𝑥)| 𝑑𝑥 𝑙 0 𝑅𝑎 = ∑|𝑦𝑖 | 𝑛 𝑖=1 three lines or 2d profile across the focus area were set and measured to get their average surface roughness, ra. ra used a set of individual measurements of peaks and valleys from each lines into average value. each of the data of ra was recorded into table, which consisted of three lines at every point of the samples and also their averages. figure 3: arithmetic average height, ra [12] it important to do surface morphology and surface roughness because it can evaluate and determine the amount of filler scattering or filler agglomerate on the conductive ink, which can affect the surface roughness and then affect the current resistance inside the ink. 2.2.2 bulk resistivity bulk resistivity was measured by using digital multimeter at four points, which were p1, p2, p3, and p4. p1 until p3 like mentioned in figure 2 and p4 is end-to-end of the ink. most digital multimeter measured down to 0.1 ω and some of it might go as high as 300 mω. firstly, negative probe was placed at the start of conductive ink track and positive probe was placed on four marked points starting from p1 until p4. this test was done to measure bulk resistivity that journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 1 july – december 2019 12 flowed along the ink track for each mark distance. the track or circuit must be powered off when measuring the resistivity or must be tested with the absence of voltage to avoid damages on the circuit. from this test, changes of resistance in straight line, turn or edge for each patterns can be observed. bulk resistivity value that was obtained from multimeter was live reading, which needed to be collected manually. all these tests were conducted with three repetitions. 2.2.3 sheet resistivity sheet resistivity was performed using four point probe machine due to its independent of the square and fairly low resistivity of thin film[10]. all the samples were measured on marked points as in figure 2. firstly, before starting the measuring process, calibration by using reference sample, an indium-tin-oxide (ito) coated glass was done to ensure the device was functioning well and accurate. referenced sample was put on the device base under probe pin which had four probes and the height of probe pin was lowered until it touched the referenced sample. all the four probes must be in parallel and touched the referenced sample because the current, i flows through outer probes and induces a voltage, v in the inner voltage probes as shown in figure 4. from the set-up as in figure 4, sheet resistivity can also be defined with mathematical equation as below; 𝜌 = 2𝜋𝑠 𝑉 𝐼 when all the probes touched the referenced sample, they started to take live reading of the sample. due to live measurement, rc (resistance-capacitance) delay was needed as the current inside the sample needed some time to climb up or down to reach saturation value. to measure the experimental samples, all calibration steps were repeated on it. before that, probe pin height needed to be adjusted again to gain suitable height for the experimental samples because pet substrate and referenced sample have different thickness. when the value was obtained, enter button was pushed to save the measured value. lastly, the collected data was imported into its software. moreover, all tests were conducted with three repetitions figure 4: four-point probe schematic set-up 3. results and discussion s s s t journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 1 july – december 2019 13 3.1 surface morphology and roughness table 1 shows conductive ink surface microstructure condition for each samples. from the image in table 1, granular particles with big quantity and size can be observed on surface of all sample with 1 mm track width. granular particles occurred when there were conglomeration of discrete solid such as filler particle in one point. this phenomenon may obstruct current conductivity inside the ink. according to kim d., & moon j. in 2005, granular particle that had been formed needed to have 3d connection that caused the particle necking and growth into continuous connection. the granular particle can be conductive even though it was still porous. but, in this case there were formation of granular particle only with no continuous connection. this situation increased the resistance inside the ink. furthermore, the printed ink for 1 mm width was formed with many layers, which gave resistance changes when current flow through it and also gave higher value of surface roughness. image microstructure for 2 mm of track width had shown there were less granular particle formed and smoother than 1 mm width. moreover, 3 mm of track had lesser granular particle and smoothest surface roughness. table 1. sample’s surface morphology p a tt e rn width surface morphology point 1 point 2 point 3 s tr a ig h t 1 mm 2 mm 3 mm c u rv e 1 mm s tr a ig h t journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 1 july – december 2019 14 2 mm 3 mm s q u a re 1 mm 2 mm 3 mm z ig -z a g 1 mm 2 mm 3 mm c u rv e z ig -z a g s q u a re journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 1 july – december 2019 13 table 2. result of surface roughness pattern width surface roughness, ra (µm) p1 p2 p3 straight 1 mm 0.33 0.30 0.63 2 mm 0.33 0.57 0.30 3 mm 0.33 0.43 0.33 curve 1 mm 0.53 0.43 0.57 2 mm 0.63 0.23 0.30 3 mm 0.27 0.33 0.27 square 1 mm 0.43 1.00 0.63 2 mm 0.53 0.27 0.27 3 mm 0.17 0.43 0.33 zig-zag 1 mm 1.20 0.93 0.23 2 mm 0.30 0.70 0.23 3 mm 0.27 0.13 0.27 other than surface morphology, average of surface roughness, ra was also measured from 3d non-contact profilometer. ra can also be defined as arithmetic average of mean line of absolute value of profile height deviations. all the obtained data was tabulated into table 2 and represented into the graph that analyze the relationship between surface roughness and each point at every width size and types of patterns. figure 5 shows the surface images of four patterns with 1 mm, 2 mm, and 3 mm of width. the surface roughness of zig-zag pattern had highest surface roughness that decreased from 0.79 µm to 0.41 µm then dropped to the lowest of four pattern to 0.22 µm with the increase of ink width from 1 mm to 3 mm. square pattern had the second highest surface roughness at 1 mm, which was 0.69 µm and dropped rapidly at 2 mm. it meant that filler had concentrated at observed point as compared to decreased value from width 2 mm to 3 mm. meanwhile, surface roughness for curve pattern had decreased uniformly and straight pattern had lowest surface roughness and they decreased constantly with small gap between each width. therefore, figure 4 shows decrement in surface roughness for all patterns with the increase of ink width. with increasing of ink’s width, it gave more space for filler to disperse during printing process and avoid the filler from agglomeration at one point. straight and curve patterns showed uniform and small changes between each width, which may due to no turn or less turn in ink track as compared to square and zig-zag patterns, which had sharp edges and turn. surface roughness is commonly controlled by process parameter such as substrate temperature, working pressure and gas pressure[3]. in this experiment, the differences of ink track width curve square zig-zag journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 1 july – december 2019 16 affected the changes of surface roughness since the other parameters were either consistent or not being considered for conductive ink. figure 5: surface roughness graph for 1mm, 2mm and 3mm line width for all patterns 3.2 bulk resistivity table 3: data for bulk resistivity pattern width (mm) bulk resistivity (kω) p1 p2 p3 p4 straight 1 0.341 0.579 0.981 1.240 2 0.181 0.348 0.579 0.753 3 0.152 0.271 0.378 0.457 curve 1 0.283 0.677 1.360 1.591 2 0.193 0.454 0.872 0.981 3 0.114 0.243 0.475 0.573 square 1 0.346 0.762 1.207 1.551 2 0.242 0.528 0.836 1.047 3 0.187 0.331 0.533 0.675 zig-zag 1 0.260 0.755 1.420 1.633 2 0.186 0.425 0.745 0.884 3 0.110 0.298 0.524 0.615 straight curve square zig-zag journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 1 july – december 2019 17 table 3 above shows collected data for bulk resistivity for each sample and these data then translate into of graph resistivity over distance point for each width size and pattern as in figure 6. figure 6 shows the bulk resistivity over point on the track, begins at p1 (2 mm), p2 (4.5 mm), p3 (7 mm), and p4 (9 mm) for four patterns like straight pattern (a), curve pattern (b), square pattern (c), and zig-zag pattern (d). in all four graphs show bulk resistivity for 1 mm width had highest resistance followed with 2 mm and 3 mm had the lowest resistance that flowed through the track. it proved that bulk resistance decreased with the increase of track width size. with bigger track width size, the more space was available inside the track to make the filler to disperse effectively, then eased the current flow and lowered the resistance existence inside the track. other than that, all four graphs show that bulk resistivity increased with longer distance of electricity flow (a) (b) (c) (d) figure 6: graph of bulk resistivity with straight pattern (a), curve pattern (b), square pattern(c), and zig-zag pattern (d) for 1 mm, 2 mm, and 3 mm of track width. meanwhile, figure 7 shows the graphs of bulk resistivity over point distance for all four patterns at 1 mm track width (a), 2 mm (b), and 3 mm (c). for graph 7(a), straight pattern had the lowest bulk resistance over point distance followed with square pattern, then curve pattern and zig-zag pattern had the highest bulk resistance. also, the graph shows that curve and zig-zag journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 1 july – december 2019 18 patterns show the lowest bulk resistance, 0.346 kω and 0.260 kω at the beginning but rise to the highest bulk resistance at p3, 1.207 kω and 1.420 kω and p4, 1.551 kω and 1.633 kω. for graph 7(b), straight pattern had the lowest bulk resistance over point distance followed with zig-zag pattern, then curve pattern and square pattern had the highest bulk resistance. all points had constant increment of bulk resistivity except for curve pattern’s bulk resistivity at point p3, which increased sharply with the value of 0.872 kω. it exceeded the other three data at p3. lastly, graph 7(c) shows that curve pattern had the lowest resistance at the beginning, then zig-zag, straight, and square patterns. but, curve, zig-zag, and square patterns had sharp rise with square pattern as the highest. meanwhile, straight pattern had increment of resistance consistently with small difference gap. (a) (b) (c) figure 7: graph of bulk resistivity with 1 mm track width (a), 2 mm track width (b), and 3 mm track width(c) for all four patterns. journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 1 july – december 2019 19 3.3 sheet resistivity table 4: data for sheet resistivity pattern width (mm) sheet resistivity (ω/sq) straight 1 122.10 2 65.34 3 33.13 1 79.94 2 50.12 3 23.40 1 72.91 2 38.90 3 25.49 1 105.94 2 48.05 3 26.90 sheet resistivity was measured with four point probe and the data was tabulated into table, then represented into graph of each pattern’s sheet resistivity over point for every track width and graph of sheet resistivity over track width. figure 8 shows sheet resistivity over point at every track width. it shows that track width of 1 mm had the highest sheet resistance, followed by 2 mm and then 3 mm width with lowest sheet resistance for all four patterns. other than that, figure 8 shows that track with 1 mm width had unstable and sharp increase or decrease of resistance values as compared to track with 2 mm and 3 mm width, which were more stable and had value increment with small differences. in these cases, sheet resistivity was highest with small track width and decreased as the width increased due to filler particle that had dispersed excellently. when the width became small or narrow, the filler particles were tended to agglomerate together and hindered the electrical conductivity. (a) (b) straight curve square zig-zag journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 1 july – december 2019 20 (c) (d) figure 8: graph of sheet resistivity with straight pattern (a), curve pattern (b), square pattern(c), and zig-zag pattern (d) for 1 mm, 2 mm, and 3 mm of track width. results of sheet resistivity over track width for every patterns were shown in figure 9. square pattern had the lowest sheet resistivity with the increase of the track width. then, it followed with curve, and zig-zag patterns. straight pattern showed the highest sheet resistivity for all four patterns figure 9: graph of sheet resistivity over track width for all four patterns journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 1 july – december 2019 21 4. conclusion stencil printing method was used to make the liquid-solid conductive ink on pet as substrate. results was obtained with different parameters such as track width, and track pattern. furthermore, the samples were prepared and measured under unloaded condition. from observation with profilometer, granular particles were formed less and far apart when the track width increased. surface roughness of the ink also became smoother when track width was increased and straight pattern had small decrement and changes at three points as compared to zig-zag pattern, which surface roughness decreased sharply at three points. there were also sheet resistivity and bulk resistivity that were measured with four point probe and multimeter. these two results had the same outcome, which the resistivity decreased with the increment of track width. zig-zag pattern had higher sheet resistivity and straight pattern had lowest sheet resistivity. meanwhile, for all three track width, straight pattern had lowest bulk resistivity and zig-zag pattern had highest bulk resistivity for 1 mm track width. meanwhile, square pattern had the highest bulk resistivity for 2 mm and 3 mm track width. from all the obtained results, straight pattern showed the best result as compared to other patterns. this was because other pattern had edge and sharp corner, which hindered the electrical conductivity and increased their resistivity. references [1] z. wang, w. wang, z. jiang, and d. yu, “low temperature sintering nano-silver conductive ink printed on cotton fabric as printed electronics,” prog. org. coatings, vol. 101, pp. 604–611, 2016. [2] j. kastner, t. faury, h. m. außerhuber, t. obermüller, h. leichtfried, m. j. haslinger, e. liftinger, j. innerlohinger, i. gnatiuk, d. holzinger, and t. lederer, “silver-based reactive ink for inkjet-printing of conductive lines on textiles,” microelectron. eng., vol. 176, pp. 84–88, 2017. [3] w. tang, y. chao, x. weng, l. deng, and k. xu, “optical property and the relationship between resistivity and surface roughness of indium tin oxide thin films,” vol. 32, pp. 680–686, 2012. [4] s. son, y. cho, j. rha, and c. choi, “fabrication of metal electrodes on fl exible substrates by controlled deposition of conductive nano-ink,” mater. lett., vol. 117, pp. 179–183, 2014. [5] e. britannica, “polyethylene terephthalate.” encyclopaedia britannica, inc, pp. 1–3, 2019. [6] t. s. tran, n. k. dutta, and n. r. choudhury, “graphene inks for printed flexible electronics: graphene dispersions, ink formulations, printing techniques and applications,” adv. colloid interface sci., p. #pagerange#, 2018. [7] h. mekaru, “thermal and ultrasonic bonding between planar polyethylene terephthalate , acrylonitrile butadiene styrene , and polycarbonate substrates,” int. j. adhes. adhes., vol. 84, no. april, pp. 394–405, 2018. [8] s. h. kim, t. min, j. w. choi, s. h. baek, j. choi, and c. aranas, “ternary bi2te3–in2te3–ga2te3 (ntype) thermoelectric film on a flexible pet substrate for use in wearables sang,” energy, vol. 3, 2018. [9] b. marinho, m. ghislandi, e. tkalya, c. e. koning, and g. de with, “electrical conductivity of compacts of graphene , multi-wall carbon nanotubes , carbon black , and graphite powder,” powder technol., vol. 221, pp. 351–358, 2012. [10] s. merilampi and p. ruuskanen, “the characterization of electrically conductive silver ink patterns on flexible substrates,” microelectron. reliab., vol. 49, no. 7, pp. 782–790, 2009. [11] j. wei, t. vo, and f. inam, “epoxy/graphene nanocomposites – processing and properties: a review,” rsc adv., vol. 5, pp. 73510–73524, 2015. [12] t. m. a. maksoud, i. m. elewa, h. soliman, and p. media, “roughness parameters,” vol. 8, no. 2, pp. 263–276, 2016. issn: 2180-1053 vol. 3 no. 2 july-december 2011 study on mechanical properties and microstructure analysis of aisi 304l stainless steel weldments 71 study on mechanical properties and microstructure analysis of aisi 304l stainless steel weldments mohd shukor salleh1, mohd irman ramli2, saifudin hafiz yahaya3 1,2,3faculty of mechanical engineering, universiti teknikal malaysia melaka, locked bag 1752, pejabat pos durian tunggal, 76109 durian tunggal, melaka email : 1shukor@utem.edu.my abstract manufacturing operations require joining process in a way that it is considered as an important process to be applied in almost every operation or process that involves fabricating of products. the aim of this research is to evaluate mechanical properties and analyzed heat affected zone (haz) of austenic stainless steel aisi 304lweldments. the welding was conducted based on three different sizes of filler wire 0.8mm, 1.0mm and 1.2mm respectively. the arc voltage used also consists of three different values 30v, 60v and 90v and the current flow for metal inert gas (mig) welding was set to constant value of 100a. the specimens were divided into five groups to undergo tensile test, hardness test, impact test, haz temperature variation study and followed by microstructure observation. the experimental result showed that tensile strength, hardness and impact resistance were increased with the used of biggest size of filler wire which is 1.2 mm. the relations then were compared with haz temperature variation analysis and the image analyzer showed that the transformation from austenite to martensite at haz created a hard and brittle structure near the fusion zone. the results revealed that different filler wire size and different arc voltage applied could enforce the austenitic stainless steel structure. key words: mig, haz, mig, austenic, martensite 74 study on mechanical properties and microstructure analysis of aisi 304l stainless steel weldments mohd shukor salleh1, mohd irman ramli2, saifudin hafiz yahaya3 1,2,3faculty of mechanical engineering, universiti teknikal malaysia melaka, locked bag 1752, pejabat pos durian tunggal, 76109 durian tunggal, melaka email : 1shukor@utem.edu.my abstract manufacturing operations require joining process in a way that it is considered as an important process to be applied in almost every operation or process that involves fabricating of products. the aim of this research is to evaluate mechanical properties and analyzed heat affected zone (haz) of austenic stainless steel aisi 304lweldments. the welding was conducted based on three different sizes of filler wire 0.8mm, 1.0mm and 1.2mm respectively. the arc voltage used also consists of three different values 30v, 60v and 90v and the current flow for metal inert gas (mig) welding was set to constant value of 100a. the specimens were divided into five groups to undergo tensile test, hardness test, impact test, haz temperature variation study and followed by microstructure observation. the experimental result showed that tensile strength, hardness and impact resistance were increased with the used of biggest size of filler wire which is 1.2 mm. the relations then were compared with haz temperature variation analysis and the image analyzer showed that the transformation from austenite to martensite at haz created a hard and brittle structure near the fusion zone. the results revealed that different filler wire size and different arc voltage applied could enforce the austenitic stainless steel structure. key words: mig, haz, mig, austenic, martensite 1.0 introduction welding is a fabrication or sculptural process that joints materials, usually metals or thermoplastics. welding involves in bringing the surfaces of metals to be joined close enough together for atomic bonding to occur as the natural consequence of atoms seeking to create for themselves a stable electron configuration. in general, welding includes any process that causes materials to join through the attractive action of inter-atomic or inter-molecular forces as opposed to purely macroscopic or even microscopic mechanical interlocking forces. welding has become a prevalent mechanical joining methodology in various industries because of its advantage over other joining methods including design flexibility, cost savings, overall weight reduction and structural performance enhancement (song et. al., 2003). mainly, in order to gain issn: 2180-1053 vol. 3 no. 2 july-december 2011 journal of mechanical engineering and technology 72 75 an acceptable weldments outcome, the recommended approaches such like welding type selection, controlling welding process parameters and modifying the structural configuration must be considered (song et.al., 2003). stainless steels belong to iron-base alloys family. the steels have excellent resistance to corrosion. normally, the stainless steels have good low temperature toughness and ductility. most of the steels contain good strength properties and corrosion resistance. all stainless steel contain iron as the main element and chromium (11% to 30%). the chromium has the basic corrosion resistance that supplements the trademark of stainless steels. welded structures made of stainless steel are commonly used in the power generation, oil and gas, marine transportation, petrochemical industries due to their higher mechanical strength and better corrosion resistance (hsiao et. al., 2008). there are only a few studies have been done to evaluate the effect of filler wire diameter and voltage setting on weldments of austenitic stainless steel. besides, it is hardly found the study on the microstructure of heat affected zone (haz) in the austenitic stainless steel weldment area. thus, this work is undertaken with the aim to study the effect of different stainless steel filler wire diameter on stainless steel weldments. in addition this study also investigates the effect of different voltage settings, mechanical properties in weldments area and heat affected zone (haz). 2.0 experimental details an experimental activity was conducted to determine the mechanical properties and also weldment area and haz microstructure of aisi 304l weldment. the concentration initially is on preparing the plate specimen (total 9 specimens) by cutting them into pieces of 20mm x 15mm. after that a 300 v-groove shape was prepared in order to obtain an acceptable welding result. the process was done by using milling machine. then the welding activity was taken place and followed by mechanical properties evaluation, and haz analysis. a. specimens preparation the stainless steel plate was cut into desired size of 20mm x 15mm accordingly. altogether there were 18 pieces of plate. laser cutting machine (model: lvd helius 2513) was used to cut the plates. laser cutting machine was used due to its accurateness in cutting the stainless steel in appropriate time consumed (figures 2.1, 2.2). 76 figure 2.1 figure 2.2 aisi stainless steel plate before cutting aisi 304l stianless steel after cutting process b. v-groove preparation the v-groove shape was prepared by using milling machine (model: bridgeport turret milling, aem br 2j2). the groove angle was 300 in order to obtain good welding quality thus saving the weld cost. a root opening was adjusted to fit with the sizes of filler wire to obtain good fusion at the root (lincoln, 2000). c. mig welding operation the welding process was conducted by using mig weld 210s (wim). the welding was set continuously by using pure argon as shielding gas. the setting for arc voltage is 30v, 60v and 90v respectively. meanwhile, the filler wire speed was set at 3.5 seconds for every voltage value. the filler wire (er308l) diameter consists of three different sizes (0.8mm, 1.0mm and 1.2mm). each filler wire required different contact tip to fit on the mig gun. d. welding parameters and voltage setting the experiment was started based on three different sizes of filler wire and three different arc voltages value. in order to obtain good welding quality, a v-groove shape was done in advance to allow acceptable penetration of weld metal between the plates. a root face of 3mm was determined to avoid melt through of the weldment metal and reduce distortion and contraction on weldment area (lincoln, 2000). table1 shows a welding parameter. 75 an acceptable weldments outcome, the recommended approaches such like welding type selection, controlling welding process parameters and modifying the structural configuration must be considered (song et.al., 2003). stainless steels belong to iron-base alloys family. the steels have excellent resistance to corrosion. normally, the stainless steels have good low temperature toughness and ductility. most of the steels contain good strength properties and corrosion resistance. all stainless steel contain iron as the main element and chromium (11% to 30%). the chromium has the basic corrosion resistance that supplements the trademark of stainless steels. welded structures made of stainless steel are commonly used in the power generation, oil and gas, marine transportation, petrochemical industries due to their higher mechanical strength and better corrosion resistance (hsiao et. al., 2008). there are only a few studies have been done to evaluate the effect of filler wire diameter and voltage setting on weldments of austenitic stainless steel. besides, it is hardly found the study on the microstructure of heat affected zone (haz) in the austenitic stainless steel weldment area. thus, this work is undertaken with the aim to study the effect of different stainless steel filler wire diameter on stainless steel weldments. in addition this study also investigates the effect of different voltage settings, mechanical properties in weldments area and heat affected zone (haz). 2.0 experimental details an experimental activity was conducted to determine the mechanical properties and also weldment area and haz microstructure of aisi 304l weldment. the concentration initially is on preparing the plate specimen (total 9 specimens) by cutting them into pieces of 20mm x 15mm. after that a 300 v-groove shape was prepared in order to obtain an acceptable welding result. the process was done by using milling machine. then the welding activity was taken place and followed by mechanical properties evaluation, and haz analysis. a. specimens preparation the stainless steel plate was cut into desired size of 20mm x 15mm accordingly. altogether there were 18 pieces of plate. laser cutting machine (model: lvd helius 2513) was used to cut the plates. laser cutting machine was used due to its accurateness in cutting the stainless steel in appropriate time consumed (figures 2.1, 2.2). issn: 2180-1053 vol. 3 no. 2 july-december 2011 study on mechanical properties and microstructure analysis of aisi 304l stainless steel weldments 73 76 figure 2.1 figure 2.2 aisi stainless steel plate before cutting aisi 304l stianless steel after cutting process b. v-groove preparation the v-groove shape was prepared by using milling machine (model: bridgeport turret milling, aem br 2j2). the groove angle was 300 in order to obtain good welding quality thus saving the weld cost. a root opening was adjusted to fit with the sizes of filler wire to obtain good fusion at the root (lincoln, 2000). c. mig welding operation the welding process was conducted by using mig weld 210s (wim). the welding was set continuously by using pure argon as shielding gas. the setting for arc voltage is 30v, 60v and 90v respectively. meanwhile, the filler wire speed was set at 3.5 seconds for every voltage value. the filler wire (er308l) diameter consists of three different sizes (0.8mm, 1.0mm and 1.2mm). each filler wire required different contact tip to fit on the mig gun. d. welding parameters and voltage setting the experiment was started based on three different sizes of filler wire and three different arc voltages value. in order to obtain good welding quality, a v-groove shape was done in advance to allow acceptable penetration of weld metal between the plates. a root face of 3mm was determined to avoid melt through of the weldment metal and reduce distortion and contraction on weldment area (lincoln, 2000). table1 shows a welding parameter. 77 table 1 welding parameter voltage current travel speed plate thickness sample amount (v) (a) (cm/min) (mm) (pieces) 30 100 60 5 1 30 100 60 5 1 30 100 60 5 1 60 100 60 5 1 60 100 60 5 1 60 100 60 5 1 90 100 60 5 1 90 100 60 5 1 90 100 60 5 1 e. microstructure analysis microstructure analysis was conducted by using image analyzer (model: buehler omniment).this equipment was specialized in analyzing the metallography structure of materials. areas that involve in this analysis were weldment area and heat affected zone (haz). in addition, the boundary between weldment area and haz also were investigated. two types of magnification are set. the magnifications are 100x and 200x. f. temperature variations measurement the variations measurement is taken using infrared thermometer (model: tmirl, cpstempseeker). the temperature is measured in a distance of approximately 10cm from the heat affected zone (haz) area. there were five readings captured in a distance of 8mm, 16mm, 24mm, 32mm and 40mm from the weld metal. the distance is decided after determining the haz area that almost penetrates approximately until 4cm from the fusion zone. the reading was taken in respect of different filler wire and arc voltage effect to temperature and mechanical properties behavior. g. tensile test measurement the tensile specimens were cut into length of 70mm and width of 15mm. to obtain the dog bone shape, milling process was conducted to machine the designated area. maximum stress was determined by dividing the maximum forces applied (kn) versus specimen area (mm2). the stress and strain data were obtained and analyzed. table 1 table 1 shows a ding parameter. issn: 2180-1053 vol. 3 no. 2 july-december 2011 journal of mechanical engineering and technology 74 77 table 1 welding parameter voltage current travel speed plate thickness sample amount (v) (a) (cm/min) (mm) (pieces) 30 100 60 5 1 30 100 60 5 1 30 100 60 5 1 60 100 60 5 1 60 100 60 5 1 60 100 60 5 1 90 100 60 5 1 90 100 60 5 1 90 100 60 5 1 e. microstructure analysis microstructure analysis was conducted by using image analyzer (model: buehler omniment).this equipment was specialized in analyzing the metallography structure of materials. areas that involve in this analysis were weldment area and heat affected zone (haz). in addition, the boundary between weldment area and haz also were investigated. two types of magnification are set. the magnifications are 100x and 200x. f. temperature variations measurement the variations measurement is taken using infrared thermometer (model: tmirl, cpstempseeker). the temperature is measured in a distance of approximately 10cm from the heat affected zone (haz) area. there were five readings captured in a distance of 8mm, 16mm, 24mm, 32mm and 40mm from the weld metal. the distance is decided after determining the haz area that almost penetrates approximately until 4cm from the fusion zone. the reading was taken in respect of different filler wire and arc voltage effect to temperature and mechanical properties behavior. g. tensile test measurement the tensile specimens were cut into length of 70mm and width of 15mm. to obtain the dog bone shape, milling process was conducted to machine the designated area. maximum stress was determined by dividing the maximum forces applied (kn) versus specimen area (mm2). the stress and strain data were obtained and analyzed. 78 h. hardness measurement the hardness specimens were cut into length of 60mm and width 10mm. milling process is conducted to obtain the dog bone shape. there were three indentation points recorded on the weldment area and haz. i. impact test measurement the impact specimens were cut into length of 60mm and width 10mm. the angle degree for the pendulum was set 900. the data were collected and analyzed for each specimen according to different filler wire diameter and arc voltage setting. 3.0 results and discussion in this experiment, parameter of filler wire diameter 0.8mm and 1.0mm with 30v, 60v and 90v arc voltage, the weldment condition were satisfying. there were no crack and porosity detected. on the other hand, for weldment using 1.2mm filler wire, crack and porosity were found (figure 2.3). the haz width was monitored occurring in a distance of maximum 4cm from the weld metal or fusion zone for every filler wire diameter and arc voltage. for example, haz minimum width (3cm) was monitored when using 0.8mm filler wire (30v) while maximum width (4cm) when using 1.2mm filler wire (90v). the haz was also observed clearly when using 0.8mm than 1.2mm filler wire (regardless of arc voltage value). figure 2.3 weldment of 1.2mm filler wire diameter meanwhile, figure 2.4 shows the penetration of weldment for different filler wire but with same arc voltage of 60v. from the figure, the behavior of weld pool proved better during 0.8mm filler wire than 1.2mm filler wire. this concluded that in this experiment, based on weld metal porosity for each filler wire diameter, filler wire (er308l) with diameter size 1.2mm issn: 2180-1053 vol. 3 no. 2 july-december 2011 study on mechanical properties and microstructure analysis of aisi 304l stainless steel weldments 75 78 h. hardness measurement the hardness specimens were cut into length of 60mm and width 10mm. milling process is conducted to obtain the dog bone shape. there were three indentation points recorded on the weldment area and haz. i. impact test measurement the impact specimens were cut into length of 60mm and width 10mm. the angle degree for the pendulum was set 900. the data were collected and analyzed for each specimen according to different filler wire diameter and arc voltage setting. 3.0 results and discussion in this experiment, parameter of filler wire diameter 0.8mm and 1.0mm with 30v, 60v and 90v arc voltage, the weldment condition were satisfying. there were no crack and porosity detected. on the other hand, for weldment using 1.2mm filler wire, crack and porosity were found (figure 2.3). the haz width was monitored occurring in a distance of maximum 4cm from the weld metal or fusion zone for every filler wire diameter and arc voltage. for example, haz minimum width (3cm) was monitored when using 0.8mm filler wire (30v) while maximum width (4cm) when using 1.2mm filler wire (90v). the haz was also observed clearly when using 0.8mm than 1.2mm filler wire (regardless of arc voltage value). figure 2.3 weldment of 1.2mm filler wire diameter meanwhile, figure 2.4 shows the penetration of weldment for different filler wire but with same arc voltage of 60v. from the figure, the behavior of weld pool proved better during 0.8mm filler wire than 1.2mm filler wire. this concluded that in this experiment, based on weld metal porosity for each filler wire diameter, filler wire (er308l) with diameter size 1.2mm 79 was not suitable for welding aisi 304l base metal. further study and adjustment need to be taken in order to get the acceptable outcome if the 1.2mm filler wire needed in application. figure 2.4 weld pool 0.8mm, 60v the trend of temperature variations is shown in graphs below (figure 2.52.8). from the graphs, the trend of temperature increased was tended to follow the arc voltage value. the higher the arc voltage, the higher temperature recorded (correia et. al., 2005). furthermore, the thicker filler wire diameter, the higher temperature recorded (cui et. al., 2006). it can be concluded, at 0.8mm filler wire size for 30v arc voltage, the temperature effect on the haz is the lowest while at 1.2mm filler wire size for 90v arc voltage, the temperature effect on the haz is the highest. according to current flow activity, it is understood that at 1.2mm filler wire, the current flow energy is the highest compare to 0.8mm and 1.0mm filler wire respectively. the temperature also felt drastically when using 30v arc voltage power source compare to 90v. figure 2.5 temperature variations for 0.8mm, 30v 0 100 200 300 400 500 1 2 3 4 5 te m pe ra tu re 0 c haz temperature effect (0.8mm, 30v) 0cm 5cm 10cm 15cm 20cm issn: 2180-1053 vol. 3 no. 2 july-december 2011 journal of mechanical engineering and technology 76 80 figure 2.6 temperature variations for 0.8mm, 90v figure 2.7 temperature variations for 1.2mm, 30v figure 2.8 temperature variations for 1.2mm, 90v the average of the maximum force and maximum stress for every 2 sample were taken for analysis. by referring to figure 2.9 which shown the result for 0.8mm filler wire, the maximum force getting higher following the increasing of arc voltage. the gap getting bigger when the 0 100 200 300 400 500 1 2 3 4 5 te m pe ra tu re 0 c haz temperature effect (0.8mm, 90v) 0cm 5cm 10cm 15cm 20cm 0 100 200 300 400 1 2 3 4 5 te m pe ra tu re 0 c haz temperature effect (1.2mm, 30v) 0cm 5cm 10cm 15cm 20cm 0 100 200 300 400 500 1 2 3 4 5 te m pe ra tu re 0 c haz temperature effect (1.2mm, 90v) 0cm 5cm 10cm 15cm 20cm 79 was not suitable for welding aisi 304l base metal. further study and adjustment need to be taken in order to get the acceptable outcome if the 1.2mm filler wire needed in application. figure 2.4 weld pool 0.8mm, 60v the trend of temperature variations is shown in graphs below (figure 2.52.8). from the graphs, the trend of temperature increased was tended to follow the arc voltage value. the higher the arc voltage, the higher temperature recorded (correia et. al., 2005). furthermore, the thicker filler wire diameter, the higher temperature recorded (cui et. al., 2006). it can be concluded, at 0.8mm filler wire size for 30v arc voltage, the temperature effect on the haz is the lowest while at 1.2mm filler wire size for 90v arc voltage, the temperature effect on the haz is the highest. according to current flow activity, it is understood that at 1.2mm filler wire, the current flow energy is the highest compare to 0.8mm and 1.0mm filler wire respectively. the temperature also felt drastically when using 30v arc voltage power source compare to 90v. figure 2.5 temperature variations for 0.8mm, 30v 0 100 200 300 400 500 1 2 3 4 5 te m pe ra tu re 0 c haz temperature effect (0.8mm, 30v) 0cm 5cm 10cm 15cm 20cm 80 figure 2.6 temperature variations for 0.8mm, 90v figure 2.7 temperature variations for 1.2mm, 30v figure 2.8 temperature variations for 1.2mm, 90v the average of the maximum force and maximum stress for every 2 sample were taken for analysis. by referring to figure 2.9 which shown the result for 0.8mm filler wire, the maximum force getting higher following the increasing of arc voltage. the gap getting bigger when the 0 100 200 300 400 500 1 2 3 4 5 te m pe ra tu re 0 c haz temperature effect (0.8mm, 90v) 0cm 5cm 10cm 15cm 20cm 0 100 200 300 400 1 2 3 4 5 te m pe ra tu re 0 c haz temperature effect (1.2mm, 30v) 0cm 5cm 10cm 15cm 20cm 0 100 200 300 400 500 1 2 3 4 5 te m pe ra tu re 0 c haz temperature effect (1.2mm, 90v) 0cm 5cm 10cm 15cm 20cm 79 was not suitable for welding aisi 304l base metal. further study and adjustment need to be taken in order to get the acceptable outcome if the 1.2mm filler wire needed in application. figure 2.4 weld pool 0.8mm, 60v the trend of temperature variations is shown in graphs below (figure 2.52.8). from the graphs, the trend of temperature increased was tended to follow the arc voltage value. the higher the arc voltage, the higher temperature recorded (correia et. al., 2005). furthermore, the thicker filler wire diameter, the higher temperature recorded (cui et. al., 2006). it can be concluded, at 0.8mm filler wire size for 30v arc voltage, the temperature effect on the haz is the lowest while at 1.2mm filler wire size for 90v arc voltage, the temperature effect on the haz is the highest. according to current flow activity, it is understood that at 1.2mm filler wire, the current flow energy is the highest compare to 0.8mm and 1.0mm filler wire respectively. the temperature also felt drastically when using 30v arc voltage power source compare to 90v. figure 2.5 temperature variations for 0.8mm, 30v 0 100 200 300 400 500 1 2 3 4 5 te m pe ra tu re 0 c haz temperature effect (0.8mm, 30v) 0cm 5cm 10cm 15cm 20cm 80 figure 2.6 temperature variations for 0.8mm, 90v figure 2.7 temperature variations for 1.2mm, 30v figure 2.8 temperature variations for 1.2mm, 90v the average of the maximum force and maximum stress for every 2 sample were taken for analysis. by referring to figure 2.9 which shown the result for 0.8mm filler wire, the maximum force getting higher following the increasing of arc voltage. the gap getting bigger when the 0 100 200 300 400 500 1 2 3 4 5 te m pe ra tu re 0 c haz temperature effect (0.8mm, 90v) 0cm 5cm 10cm 15cm 20cm 0 100 200 300 400 1 2 3 4 5 te m pe ra tu re 0 c haz temperature effect (1.2mm, 30v) 0cm 5cm 10cm 15cm 20cm 0 100 200 300 400 500 1 2 3 4 5 te m pe ra tu re 0 c haz temperature effect (1.2mm, 90v) 0cm 5cm 10cm 15cm 20cm issn: 2180-1053 vol. 3 no. 2 july-december 2011 study on mechanical properties and microstructure analysis of aisi 304l stainless steel weldments 77 80 figure 2.6 temperature variations for 0.8mm, 90v figure 2.7 temperature variations for 1.2mm, 30v figure 2.8 temperature variations for 1.2mm, 90v the average of the maximum force and maximum stress for every 2 sample were taken for analysis. by referring to figure 2.9 which shown the result for 0.8mm filler wire, the maximum force getting higher following the increasing of arc voltage. the gap getting bigger when the 0 100 200 300 400 500 1 2 3 4 5 te m pe ra tu re 0 c haz temperature effect (0.8mm, 90v) 0cm 5cm 10cm 15cm 20cm 0 100 200 300 400 1 2 3 4 5 te m pe ra tu re 0 c haz temperature effect (1.2mm, 30v) 0cm 5cm 10cm 15cm 20cm 0 100 200 300 400 500 1 2 3 4 5 te m pe ra tu re 0 c haz temperature effect (1.2mm, 90v) 0cm 5cm 10cm 15cm 20cm 81 arc voltage is 60v with the maximum stress value recorded at 204.55 n/mm2. the graph trend shows a same pattern when using 1.2mm filler wire. but the value of maximum stress recorded at 90v is higher at 318.66 n/mm2. whereas by using 1.0mm filler wire, the maximum stress increased radically at 90v with value of 257.62 n/mm2. from the observation, the higher the arc voltage, the penetration of the weld metal is getting good. bigger filler wire diameter also contribute in strengthened the weldment area. the filler wire size and arc voltage value worked in parallel to ensure the strength of weldment structure. although the porosity and crack were observed for 1.2mm filler wire, due to its penetration is better than 0.8mm and 1.0mm filler wire, the structure of weldment is the strongest compare to others. figure 2.9 average max force and max stress for 0.8mm filler wire the hardness value for weldment area (fusion zone) is always lower than haz. this is because at haz, the microstructure is the hardest due to the transformation of microstructure from austenite to martensite caused by heat treatment during welding. whereas at weldment area, the structure hardness is the lowest due to overheat treatment received inside the fusion zone. besides, the microstructure of weldment area shows solidification process that formed ferrite at high temperature. meanwhile, hardness for base metal stands between haz and weldment area due to the structure didn’t experience any heat treatment process and stay as austenite. figure 2.10 shows the increasing of hardness not as radically as during 1.2mm filler wire. the observation is almost the same for 1.0mm. here the trend shows that for every different filler wire size, the bigger the size, the hardness increased radically. this can be aligned with the data from tensile test that mentioned the max stress getting higher with the increasing in filler wire size. 30v 60v 90v arc voltage issn: 2180-1053 vol. 3 no. 2 july-december 2011 journal of mechanical engineering and technology 78 82 figure 2.10 weld area, haz and base metal hardness (0.8mm) the impact test was conducted on two samples for each specimen. based on the result, the higher arc voltage supplied, the higher energy absorbed. the trend showed that the energy absorbed increased radically. meanwhile, same observation also appeared for 1.0mm filler wire. the energy absorbed is perpendicular with the voltage applied. at 90v, the energy absorbed tend to reduce a bit. the possibility here is the energy absorbed getting lower due to weldment structure for 1.2mm filler wire content with porosity and crack. figure 2.11meanwhile shows the average energy absorbed for every filler wire diameter concerning the arc voltage applied. from the graph, suggestion can be made that higher arc voltage with bigger filler wire diameter tend to absorb higher energy. although the 1.2mm filler wire contribute to weldment that full with porosity and crack, high penetration ratio that occurred when welding helps in strengthened the structure. weldment distortion and contraction that occurred due to higher current flow has further tightened the weld area. figure 2.11 average impact test result 30v 60v 90v volt hrd 1.0mm 1.2mm 0.8mm 1.0mm 1.2mm volt hrd issn: 2180-1053 vol. 3 no. 2 july-december 2011 study on mechanical properties and microstructure analysis of aisi 304l stainless steel weldments 79 83 for microstructure analysis, all haz and weldement area microstructures exhibit almost the same behavior (figure 2.12-2.17). the grain looked coarser. meanwhile, at 0.8mm and 1.0mm, the grain exhibit skeletal morphology structure. this occurred due to when weld cooling rates are moderate, or when the cr is low but still within ferrite austenite (fa) range, skeletal ferrite morphology appeared. this is a consequence of the advance of the austenite consuming the ferrite until the ferrite is sufficiently enriched in ferrite promoting elements (chromium and molybdenum) and depleted austenite promoting elements (nickel, carbon and nitrogen) (jang et.al., 2005). it is stable at lower temperatures where diffusion is limited. at 1.2mm, where the heat is the highest during welding, the weldment area showed a solidification subgrain boundary (ssgb). this occurred in ferrite austenite (fa) and ferrite matrix (f) modes (lee et. al., 2006). figure 2.12 haz (100x magnification: 0.8mm, 30v) figure 2.13 weldment area (100x magnification : 0.8mm, 30v) 84 figure 2.14 haz (100x magnification: 1.0mm, 30v) figure 2.15 weldment area (100x magnification : 1.0mm, 30v) figure 2.16 haz (100x magnification: 1.2mm, 30v) issn: 2180-1053 vol. 3 no. 2 july-december 2011 journal of mechanical engineering and technology 80 84 figure 2.14 haz (100x magnification: 1.0mm, 30v) figure 2.15 weldment area (100x magnification : 1.0mm, 30v) figure 2.16 haz (100x magnification: 1.2mm, 30v) 85 figure 2.17 weldment area (100x magnification : 1.2mm, 30v) 4.0 conclusion the effect of welding parameters as stated previously that has link with arc voltage and filler wire diameter on aisi 304l stainless steel have been investigated. all the observations and analysis for the welded specimen and samples are subjected to normal inspection and testing methods. overall, three main investigations related with mechanical properties (consists of tensile and impact test) and haz temperature effect analysis were conducted accordingly. the haz temperature variations analysis showed that filler wire diameter had a significant role in weldment strength. this is because, different filler wire size can produce different heat that gives effect to the weldment distortion and contraction thus delaying the cooling time. as for an example, in 1.2mm filler wire, the penetration of weld metal through v-groove area is relatively better compared with 0.8mm and 1.0mm filler wire. the cooling ratio is also tent to reduce steadily and this has contributed to stable microstructure transition. meanwhile, tensile test proved that penetration of weld metal acts as the main subject that contributed to the increasing of weldment strength. regardless of porosity and crack, but with the assist of higher voltage (90v) and stable cooling rate, with additional bigger filler wire size (1.2mm), maximum stress is the highest contributing by well penetration. this can be concluded that melting rate of filler wire reduce the possibility of weldment failure due to porosity and crack. the result of hardness test proved that the filler wire size and arc voltage contribute in strengthening the structure of haz. this is because the existing of amount of heat during weldment according to filler wire size and arc voltage applied further increased the heat treatment process received by haz thus contributed in strengthening the structure (al-haidary et. al., 2006). whereas the microstructure analysis revealed that the ferrite to austenite formation contribute in increasing the weldment strength, whereby at weldment area (fusion zone) ferrite form occurred due high temperature during welding. this makes the structure strength at weldment area weaker compare to haz and base metal (lippold et. al., 2005). as for impact test, the result also shown that energy absorbed for 1.2mm filler wire is the highest. 85 figure 2.17 weldment area (100x magnification : 1.2mm, 30v) 4.0 conclusion the effect of welding parameters as stated previously that has link with arc voltage and filler wire diameter on aisi 304l stainless steel have been investigated. all the observations and analysis for the welded specimen and samples are subjected to normal inspection and testing methods. overall, three main investigations related with mechanical properties (consists of tensile and impact test) and haz temperature effect analysis were conducted accordingly. the haz temperature variations analysis showed that filler wire diameter had a significant role in weldment strength. this is because, different filler wire size can produce different heat that gives effect to the weldment distortion and contraction thus delaying the cooling time. as for an example, in 1.2mm filler wire, the penetration of weld metal through v-groove area is relatively better compared with 0.8mm and 1.0mm filler wire. the cooling ratio is also tent to reduce steadily and this has contributed to stable microstructure transition. meanwhile, tensile test proved that penetration of weld metal acts as the main subject that contributed to the increasing of weldment strength. regardless of porosity and crack, but with the assist of higher voltage (90v) and stable cooling rate, with additional bigger filler wire size (1.2mm), maximum stress is the highest contributing by well penetration. this can be concluded that melting rate of filler wire reduce the possibility of weldment failure due to porosity and crack. the result of hardness test proved that the filler wire size and arc voltage contribute in strengthening the structure of haz. this is because the existing of amount of heat during weldment according to filler wire size and arc voltage applied further increased the heat treatment process received by haz thus contributed in strengthening the structure (al-haidary et. al., 2006). whereas the microstructure analysis revealed that the ferrite to austenite formation contribute in increasing the weldment strength, whereby at weldment area (fusion zone) ferrite form occurred due high temperature during welding. this makes the structure strength at weldment area weaker compare to haz and base metal (lippold et. al., 2005). as for impact test, the result also shown that energy absorbed for 1.2mm filler wire is the highest. issn: 2180-1053 vol. 3 no. 2 july-december 2011 study on mechanical properties and microstructure analysis of aisi 304l stainless steel weldments 81 85 figure 2.17 weldment area (100x magnification : 1.2mm, 30v) 4.0 conclusion the effect of welding parameters as stated previously that has link with arc voltage and filler wire diameter on aisi 304l stainless steel have been investigated. all the observations and analysis for the welded specimen and samples are subjected to normal inspection and testing methods. overall, three main investigations related with mechanical properties (consists of tensile and impact test) and haz temperature effect analysis were conducted accordingly. the haz temperature variations analysis showed that filler wire diameter had a significant role in weldment strength. this is because, different filler wire size can produce different heat that gives effect to the weldment distortion and contraction thus delaying the cooling time. as for an example, in 1.2mm filler wire, the penetration of weld metal through v-groove area is relatively better compared with 0.8mm and 1.0mm filler wire. the cooling ratio is also tent to reduce steadily and this has contributed to stable microstructure transition. meanwhile, tensile test proved that penetration of weld metal acts as the main subject that contributed to the increasing of weldment strength. regardless of porosity and crack, but with the assist of higher voltage (90v) and stable cooling rate, with additional bigger filler wire size (1.2mm), maximum stress is the highest contributing by well penetration. this can be concluded that melting rate of filler wire reduce the possibility of weldment failure due to porosity and crack. the result of hardness test proved that the filler wire size and arc voltage contribute in strengthening the structure of haz. this is because the existing of amount of heat during weldment according to filler wire size and arc voltage applied further increased the heat treatment process received by haz thus contributed in strengthening the structure (al-haidary et. al., 2006). whereas the microstructure analysis revealed that the ferrite to austenite formation contribute in increasing the weldment strength, whereby at weldment area (fusion zone) ferrite form occurred due high temperature during welding. this makes the structure strength at weldment area weaker compare to haz and base metal (lippold et. al., 2005). as for impact test, the result also shown that energy absorbed for 1.2mm filler wire is the highest. 86 the result has relation with tensile test as stated previously. only at 1.2mm filler wire, the energy absorbed was reduced a bit at 90v was due to crack and porosity existence. but in general, the energy absorbed for 1.2mm filler wire was the highest compare with 0.8mm and 1.0mm filler wire regardless of arc voltage supplied. here, the current affect that melt-up the filler wire supplied permissible energy for the weld metal to penetrate smoothly. 5.0 recommendation there are some activities that highly recommended in the future work: a) to study the effect of different groove angle on the weldment strength. the results obtained from this experiment are only focused on 300 groove angle. in the future, various angles can be tested and observation on the weldment strength and its microstructures can be further investigated. the result shall be compiled and identification of the appropriate groove angle that contributes to the better strength of all could be understood. the relation of melt through, filler wire melting rate also can be studied. for further expanding the scope, root face for every groove angle can be set as parameter to determine the weldment strength obtained. b) to study the effect of welding time and speed on the haz by using robot welding. the welding time and speed are easy controlled if the application using robot welding. the haz is smoothly obtained and the result accuracy will be increased. systematic approach by using robot welding will enhanced the data collected thus better comparison can be clearly made. welding time and speed will act as parameters and analysis will be conducted on the haz obtained by referring to the microstructures transformation and behavior. all the observations then can be related to weldment strength, penetration effect and weld metal porosity by conducting mechanical and microstructural analysis following standard recommended. 6.0 acknowledgements the authors wish to thank and gratitude to the faculty of manufacturing engineering, university teknikal malaysia melaka (utem) and also the ministry of higher education malaysia for their financial support to the above project through the frgs funding frgs/2007/fkp(17)-f0042. issn: 2180-1053 vol. 3 no. 2 july-december 2011 journal of mechanical engineering and technology 82 87 7.0 references j. song, j. peters, a. noor and p. michaleris. 2003. “sensitivity analysis of the thermomechanical response of welded joints”, international journal of solids and structures, vol. 40, issue 16, pp. 4167-4180, 2003. w.y. hsiao, s.h. wang, c.y. chen, and w.s. lee. 2008. “effect of dynamic impact on mechanical properties and microstructure of special stainless steel weldments”, materials chemistry and physics, vol. 111, issue 1, pp. 172-179. james f. lincoln arc welding foundation. 2000. “the procedure handbook of arc welding: fourteenth edition”, james f. lincoln arc welding foundation, p.o. box 17035, cleveland, ohio, usa. d.s. correia, c.v. goncalves, s.s. da cunha jr. and v.a. ferraresi. 2005. “comparison between genetic algorithm and response surface methodology in gmaw welding optimization”, journal of materials processing technology, vol. 160, issue 1, pp. 70-76. y. cui, c.d. lundinand, v. hariharan. 2006. “mechanical behavior of austenitic stainless steel weld metals with microfissures’. journal of materials processing technology, vol. 171, issue 1, pp. 150-155. k.c. jang, d.g. lee, j.m. kuk, and i.s. kim.2005. “welding and environmental test condition effect in weldability and strength of al alloy”, journal of materials processing technology, vol. 164-165, pp. 1038-1045. w.s. lee, c.f. lin, c.y. liu, and c.w. cheng. 2006.“effects of strain rate and welding current mode on microstructural properties of sus 304l paw welds”, journal of materials processing technology, vol. 183, issues 2-3, pp. 183-193. j.t. al-haidary, a.a. wahab, and e.h. abdul salam. 2006. “fatigue crack propagation in austenitic stainless steel weldments” metallurgical and materials transactions a, vol. 37, issue 11, pp. 3205-3214. j. c. lippold, and d. j. kotecki “welding metallurgy and weldability of stainless steels”. 2005. john wiley & sons, inc., hoboken, new jersey. 86 the result has relation with tensile test as stated previously. only at 1.2mm filler wire, the energy absorbed was reduced a bit at 90v was due to crack and porosity existence. but in general, the energy absorbed for 1.2mm filler wire was the highest compare with 0.8mm and 1.0mm filler wire regardless of arc voltage supplied. here, the current affect that melt-up the filler wire supplied permissible energy for the weld metal to penetrate smoothly. 5.0 recommendation there are some activities that highly recommended in the future work: a) to study the effect of different groove angle on the weldment strength. the results obtained from this experiment are only focused on 300 groove angle. in the future, various angles can be tested and observation on the weldment strength and its microstructures can be further investigated. the result shall be compiled and identification of the appropriate groove angle that contributes to the better strength of all could be understood. the relation of melt through, filler wire melting rate also can be studied. for further expanding the scope, root face for every groove angle can be set as parameter to determine the weldment strength obtained. b) to study the effect of welding time and speed on the haz by using robot welding. the welding time and speed are easy controlled if the application using robot welding. the haz is smoothly obtained and the result accuracy will be increased. systematic approach by using robot welding will enhanced the data collected thus better comparison can be clearly made. welding time and speed will act as parameters and analysis will be conducted on the haz obtained by referring to the microstructures transformation and behavior. all the observations then can be related to weldment strength, penetration effect and weld metal porosity by conducting mechanical and microstructural analysis following standard recommended. 6.0 acknowledgements the authors wish to thank and gratitude to the faculty of manufacturing engineering, university teknikal malaysia melaka (utem) and also the ministry of higher education malaysia for their financial support to the above project through the frgs funding frgs/2007/fkp(17)-f0042. 07(71-82).pdf preparation of papers in a two column model paper format issn: 2180-1053 vol. 10 no.1 january – june 2018 79 simulation of compressible flow using a semiimplicit tvd scheme e. salimipour1*, a. salimipour2 1 department of mechanical engineering, quchan university of technology, quchan, iran 2 department of mathematics, quchan university of technology, quchan, iran abstract total variation diminishing (tvd) scheme is a kind of robust high-resolution approach, which removes the undesirable oscillations generated by numerical solution. the present work proposes a new implementation of the tvd scheme into a density-based semi-implicit finite-volume procedure to solve the inviscid and viscous flow equations. the proposed algorithm uses a simple linearization technique for convective fluxes. in order to enhance the accuracy of the algorithm, a high-resolution tvd scheme is employed in the discretization of the governing equations. this procedure has a simple implementation compared to other explicit and implicit schemes. the present scheme is first examined for some inviscid and viscous steady-state flows at several mach numbers from subsonic to the supersonic regime. in addition, the inviscid and viscous unsteady flows are simulated and compared with experimental and numerical results, so that an acceptable correspondence was obtained. results from this study indicate that the proposed algorithm is accurate for a wide range of mach numbers. keywords: density-based method; semi-implicit; finite-volume; navier-stokes equations; total variation diminishing. 1.0 introduction the flow equations typically solved in two general techniques of explicit and implicit in density-based methods. implementations of the flow solution algorithm for implicit methods are commonly more difficult than the explicit ones, and for this reason, the flow simulations began with explicit procedures. jameson was one of the earliest researchers who used explicit methods to solve the steady-state compressible inviscid flows (jameson, 1981; 1983; 1991; jameson & yoon, 1986). he was commonly used dissipative terms for eliminating spurious oscillations, which were triggered by discontinuities in the solution. however, more stability and consequently, using the larger time steps in implicit methods cause to be attended by researchers. another significant advantage of implicit schemes is their notable robustness and the rate of convergence in stiff equation systems or source terms, which are usually encountered in the simulations *corresponding author e-mail: esalimipour@qiet.ac.ir journal of mechanical engineering and technology 80 issn: 2180-1053 vol. 10 no.1 january – june 2018 of the real gases, turbulence modeling, or in highly stretched grids cases namely, high reynolds number flows (blazek, 2001). difficulties of the implicit methods are associated with the existence of nonlinear fluxes in the flow equations which need to be linearized. a popular method for linearization is the local taylor expansion about the current time level. in one of the earliest works by jespersen & pulliam (1983), some linearizations for flux-vector splitting were analyzed using a numerical fixed-point iteration analysis. their analysis showed that the use of approximate linearizations could be quite detrimental to stability and convergence of the numerical scheme. yoon & jameson (1986) developed a relaxation method using a multigrid method for the solution of the euler equations. their solution was based on a central difference scheme and did not need flux splitting. the total variation diminishing (tvd) scheme is a high-resolution scheme, which acts very well on the unfavorable numerical errors in the solution of the flow equations, especially near the part where variables are discontinuous. those numerical errors generally include numerical diffusion and oscillations when the numerical scheme is within the framework of the finite-volume method (hou, simons & hinkelmann, 2012). yee, warming & harten (1983) presented a detailed implementation of the implicit tvd scheme for the steady-state oneand two-dimensional compressible inviscid equations of gas dynamics. their numerical results also indicated that the convergence rate is susceptible to the courant–friedrichs–lewy (cfl) number. the iteration count had overgrown when the calculation was carried out away from an optimal time-step. a spatial discretization with third-order accuracy for the inviscid flux terms of the euler equations used by ravichandran (1997). he combined a runge-kutta time-stepping algorithm with the high-order spatial discretization to produce effective integration schemes for steady-state euler computations. teixeira & alves (2012) carried out a procedure that generated steady-states with accurate far-field entrainment. an efficient and robust explicit time integration procedure for a high-order discontinuous galerkin method was proposed by renac et al. (2013) to solve the unsteady compressible navier– stokes equations. kapen & tchuen (2015) investigated an easy implementation method for the solution of the multi-dimensional riemann problem for gas dynamics by use of the literal extension of the toro vazquez-harten lax leer scheme. a high-order tvd scheme is also a kind of robust high-resolution scheme, which removes the numerical errors by employing a limited flux to maintain monotonicity as well as high accuracy. the tvd concept introduced by harten (1983) for ensuring monotonicity. this concept was then developed by sweby (1984) using a flux limiter. this limiter consists of a limiter variable r and a limiter function φ, which are presented as an r-φ diagram. however, the tvd conditions are met only within a fixed part of this diagram. that means a group of schemes fulfill this law by laying the r-φ into this tvd region, such as minmod, van leer, van albada, superbee, and other schemes (waterson & deconinck, 2007). all of these schemes are derived based on the uniform grids. hou et al. (2012) presented the tvd schemes for both uniform and unstructured grids. they also proposed an improved tvd scheme named wahy. zhang et al. (2015) proposed a refined r‐ factor algorithm for implementing the tvd schemes on arbitrary unstructured meshes based on the previous researches. the above-mentioned studies were carried out for the steady-state solutions. another disadvantage of explicit methods was appeared in solving unsteady problems, because of the lack of convergence in time steps. jameson (2009) first implemented the dual-time simulation of compressible flow using a semi-implicit tvd scheme issn: 2180-1053 vol. 10 no.1 january – june 2018 81 method for unsteady flows using an explicit multistage scheme accelerated by local timestepping and multigrid. the significant advantage of this approach is that the physical time step is not restricted as usual in explicit methods. this method caused the convergence in each physical time step using an internal iteration loop. however, the dual-time stepping method increased cpu time. some of the existing implicit procedures have high speed to solve the steady-state flows due to using the large time steps. however, there are lots of unsteady problems, which do not need the large time steps and thus, implementation difficulties and use of huge computer memory are not affordable for these problems. this issue is especially crucial for solving three-dimensional problems or two-dimensional ones, which need the large computational grids. in this study, the implementation of a semi-implicit tvd scheme with a simple linearization technique is described in details. this procedure has some benefits as below: 1ease of implementation even against the existing explicit procedures 2use of high-resolution tvd schemes to prevent the numerical oscillations 3ease of coupling with other equations such as turbulence models, solid equations of motion (fluid-structure interaction) and two-phase flows. 4no need to solve additional relations such as dissipative terms used in the central methods, roe matrix, … to evaluate the present procedure, the navier-stokes equations are solved on a twodimensional, unsteady, compressible flow by writing a computer code. since an accurate validation is required for any numerical solver, several code validation studies are presented, including some cases of pressure and mach number distributions and variations of lift, drag and pitching moment coefficients. governing equations in this section, the numerical procedure used to compute the unsteady compressible flow is briefly described. the integral formulations for mass, momentum and energy conservation, in the non-dimensional form, are expressed as follows (blazek, 2001): (1) 𝜕 𝜕𝜏 ∫ �⃗⃗� 𝛺 𝑑𝛺 + ∮(𝐹 𝑐 − 𝐹 𝑣)𝑑𝑆 = 0 𝜕𝛺 where ω is the control volume, bounded by the closed surface ∂ω, �⃗⃗� denotes the vector of the so-called conservative variables, and 𝐹 𝑐 and 𝐹 𝑣 are convection and viscous fluxes expressed as follows: journal of mechanical engineering and technology 82 issn: 2180-1053 vol. 10 no.1 january – june 2018 (2) �⃗⃗� = [ 𝜌 𝜌𝑢 𝜌𝑣 𝜌𝐸 ] ; 𝐹 𝑐 = [ 𝜌𝑉 𝜌𝑢𝑉 + 𝑛𝑥𝑝 𝜌𝑣𝑉 + 𝑛𝑦𝑝 𝑉(𝜌𝐸 + 𝑝) ] ; 𝐹 𝑣 = [ 0 𝑛𝑥𝜏𝑥𝑥 + 𝑛𝑦𝜏𝑥𝑦 𝑛𝑥𝜏𝑦𝑥 + 𝑛𝑦𝜏𝑦𝑦 𝑛𝑥𝛩𝑥 + 𝑛𝑦𝛩𝑦 ] where v is defined as the scalar product of the velocity vector and the unit normal vector as follows: (3) 𝑉 ≡ 𝑣 . �⃗� = 𝑛𝑥𝑢 + 𝑛𝑦𝑣 with nx and ny being the components of the outward facing unit normal vector of the control surface ∂ω. e is the total energy per unit mass and is defined as (4) 𝐸 = 𝑝 𝜌(𝛾 − 1) + ( 𝑢2 + 𝑣2 2 ) the shear stress components, θx and θy are expressed as follows: (5) 𝜏𝑥𝑥 = 2𝜇 𝑀∞ re ( 𝜕𝑢 𝜕𝑥 − ∇⃗⃗ · 𝑣 3 ) (6) 𝜏𝑦𝑦 = 2𝜇 𝑀∞ re ( 𝜕𝑣 𝜕𝑦 − ∇⃗⃗ · 𝑣 3 ) (7) 𝜏𝑥𝑦 = 𝜏𝑦𝑥 = 𝜇 𝑀∞ re ( 𝜕𝑢 𝜕𝑦 + 𝜕𝑣 𝜕𝑥 ) (8) 𝛩𝑥 = 𝑢𝜏𝑥𝑥 + 𝑣𝜏𝑥𝑦 (9) 𝛩𝑦 = 𝑢𝜏𝑦𝑥 + 𝑣𝜏𝑦𝑦 all geometrical lengths are normalized with the size of characteristic length l, velocities with the free-stream speed of sound a∞, physical time (τ) with l/a∞, viscosity (μ) with μ∞, density (ρ) with ρ∞, and pressure (p) with ρ∞a∞ 2. implicit discretization of the governing equations the present semi-implicit algorithm is based on an iterative procedure. for simplicity, all the equations are developed in two-dimensional cartesian coordinates. however, the algorithm and the underlying principle are general and can be applied to all structured (staggered or collocated) or unstructured grid systems. performing spatial integration on the governing equation, over the control volume presented in figure 1, leads to the following discrete equations: simulation of compressible flow using a semi-implicit tvd scheme issn: 2180-1053 vol. 10 no.1 january – june 2018 83 figure 1. control volume for the two-dimensional situation. (10) �⃗⃗� 𝑛+1𝑃 − �⃗⃗� 𝑛 𝑃 ∆𝑡 ∆𝑥∆𝑦 + 𝐴 𝑒 − 𝐴 𝑤 + �⃗� 𝑛 − �⃗� 𝑠 = 𝑆 where 𝐴 and �⃗� are the nonlinear convective flux vectors and 𝑆 denotes the source term vector expressed as follows: (11) 𝐴 = [ 𝜌𝑢∆𝑦 𝜌𝑢𝑢∆𝑦 𝜌𝑢𝑣∆𝑦 𝜌𝑢𝐸∆𝑦 ] ; �⃗� = [ 𝜌𝑣∆𝑥 𝜌𝑣𝑢∆𝑥 𝜌𝑣𝑣∆𝑥 𝜌𝑣𝐸∆𝑥 ] ; 𝑆 = [ 0 [(𝜏𝑥𝑥 − 𝑝)𝑒 − (𝜏𝑥𝑥 − 𝑝)𝑤]∆𝑦 + (𝜏𝑥𝑦𝑛 − 𝜏𝑥𝑦𝑠 ) ∆𝑥 (𝜏𝑦𝑥𝑒 − 𝜏𝑦𝑥𝑤 ) ∆𝑦 + [(𝜏𝑦𝑦 − 𝑝)𝑛 − (𝜏𝑦𝑦 − 𝑝)𝑠 ] ∆𝑥 [(𝛩𝑥 − 𝑢𝑝)𝑒 − (𝛩𝑥 − 𝑢𝑝)𝑤]∆𝑦 + [(𝛩𝑦 − 𝑣𝑝)𝑛 − (𝛩𝑦 − 𝑣𝑝)𝑠 ] ∆𝑥] equation (10) can be linearized as sollow: (12) �⃗⃗� 𝑃 𝑛+1 − �⃗⃗� 𝑃 𝑛 ∆𝑡 ∆𝑥∆𝑦 + 𝑢𝑒 𝑛�⃗⃗� 𝑒 𝑛+1 ∆𝑦 − 𝑢𝑤 𝑛�⃗⃗� 𝑤 𝑛+1 ∆𝑦 + 𝑣𝑛 𝑛�⃗⃗� 𝑛 𝑛+1 ∆𝑥 − 𝑣𝑠 𝑛�⃗⃗� 𝑠 𝑛+1 ∆𝑥 = 𝑆 𝑃 𝑛 with known faces velocities ue, uw, vn and vs from the previous iteration. subscripts e, w, n, s and p in equation (12) are the locations as described in figure 1. to solve the linear equation (12), the conservative variables on the faces of the control volume, i.e., �⃗⃗� 𝑒, �⃗⃗� 𝑤, �⃗⃗� 𝑛 and �⃗⃗� 𝑠 must be approximated by the cell centers ones with highorder accuracy. there are various high-resolution limiter schemes such as essentially nonjournal of mechanical engineering and technology 84 issn: 2180-1053 vol. 10 no.1 january – june 2018 oscillatory (eno), weighted essentially non-oscillatory (weno), normalized variable diagram (nvd) and total variation diminishing (tvd). the last case is one of the most robust schemes for capturing severe gradients in the flow-field. 3.1 applying the tvd scheme for uniform grids, the face value �⃗⃗� 𝑒 (between the cell centers p and e) is written in roe’s way (roe, 1985). it consists of a diffusive first order upwind term and an anti-diffusive term: (13) �⃗⃗� 𝑒 = �⃗⃗� 𝑃 + 1 2 𝛷(𝑟𝑒)(�⃗⃗� 𝐸 − �⃗⃗� 𝑃) ; 𝑢𝑒 > 0 �⃗⃗� 𝑒 = �⃗⃗� 𝐸 + 1 2 𝛷(𝑟𝑒)(�⃗⃗� 𝑃 − �⃗⃗� 𝐸) ; 𝑢𝑒 < 0 the upwind ratio 𝑟𝑒 of consecutive differences of �⃗⃗� , can be expressed as follows: (14) 𝑟𝑒 = �⃗⃗� 𝑃 − �⃗⃗� 𝑊 �⃗⃗� 𝐸 − �⃗⃗� 𝑃 ; 𝑢𝑒 > 0 𝑟𝑒 = �⃗⃗� 𝐸 − �⃗⃗� 𝐸𝐸 �⃗⃗� 𝑃 − �⃗⃗� 𝐸 ; 𝑢𝑒 < 0 indeed, the anti-diffusive term in equation (13) is a second-order correction term; however, the tvd conditions necessitate that the function φ stays within the secondorder tvd region and pass through the point (1,1) (sweby, 1984) as shown in figure 2. some conventional tvd schemes are listed in below and also plotted in figure 2. ∅(𝑟) = 𝑚𝑎𝑥[0, 𝑚𝑖𝑛(𝑟, 1)] minmod (harten, 1983) ∅(𝑟) = 𝑚𝑎𝑥[0, 𝑚𝑖𝑛(2𝑟, 1), 𝑚𝑖𝑛(𝑟, 2)] superbee (harten, 1983) ∅(𝑟) = (𝑟 + |𝑟|) (1 + 𝑟)⁄ van leer (waterson & deconinck, 1983) ∅(𝑟) = 𝑚𝑎𝑥[0, (𝑟 + 𝑟2) (1 + 𝑟2)⁄ ] van albada (blazek, 2001) simulation of compressible flow using a semi-implicit tvd scheme issn: 2180-1053 vol. 10 no.1 january – june 2018 85 figure 2. second-order tvd region (grayed area) and some schemes. 3.2 deriving linear equations system with replacing the equation (13) into the equation (12) and considering the flow directions, the set of linear flow equations will be derived as follow: (15) 𝑎𝑃�⃗⃗� 𝑃 𝑛+1 = 𝑎𝐸�⃗⃗� 𝐸 𝑛+1 + 𝑎𝑊�⃗⃗� 𝑊 𝑛+1 + 𝑎𝑁�⃗⃗� 𝑁 𝑛+1 + 𝑎𝑆�⃗⃗� 𝑆 𝑛+1 + �⃗� 𝑛 where (16) 𝑎𝐸 = 𝑚𝑎𝑥[0, −(1 − ∅(𝑟𝑒) 2⁄ )𝑢𝑒∆𝑦] − 𝑚𝑎𝑥[0, 𝑢𝑒∆𝑦 ∅(𝑟𝑒) 2⁄ ] (17) 𝑎𝑊 = 𝑚𝑎𝑥[0, (1 − ∅(𝑟𝑤) 2⁄ )𝑢𝑤∆𝑦] − 𝑚𝑎𝑥[0, −𝑢𝑤∆𝑦 ∅(𝑟𝑤) 2⁄ ] (18) 𝑎𝑁 = 𝑚𝑎𝑥[0, −(1 − ∅(𝑟𝑛) 2⁄ )𝑣𝑛∆𝑥] − 𝑚𝑎𝑥[0, 𝑣𝑛∆𝑥 ∅(𝑟𝑛) 2⁄ ] (19) 𝑎𝑆 = 𝑚𝑎𝑥[0, (1 − ∅(𝑟𝑠) 2⁄ )𝑣𝑠∆𝑥] − 𝑚𝑎𝑥[0, −𝑣𝑠∆𝑥 ∅(𝑟𝑠) 2⁄ ] (20) 𝑎𝑃 = ∆𝑥∆𝑦 ∆𝑡⁄ + 𝑎𝐸 + 𝑎𝑊 + 𝑎𝑁 + 𝑎𝑆 + (𝑢𝑒 − 𝑢𝑤)∆𝑦 + (𝑣𝑛 − 𝑣𝑠)∆𝑥 (21) �⃗� 𝑛 = 𝑆 𝑃 𝑛 + �⃗⃗� 𝑃 𝑛 ∆𝑥∆𝑦 ∆𝑡⁄ where ae, aw, … , ass and ap are the 4 x 4 diagonal matrices for each grid cell. the set of equations (15) can be solved by various schemes such as alternating direction implicit (adi), lower-upper symmetric gauss-seidel (lu-sgs), generalized minimal residual (gmres), etc. in this work, an adi method is used. to dispose of the above procedure, an algorithm is presented for one time step as below: 1generate a computational grid, 2determine the distances of the grid cells (δx and δy), 3calculate the faces velocities (ue, uw, vn and vs), 4choose a tvd scheme (φ(r)), journal of mechanical engineering and technology 86 issn: 2180-1053 vol. 10 no.1 january – june 2018 5calculate the coefficients (relations (16-20)), 6calculate the source term (�⃗� 𝑛), 7form coefficient matrix (a), 8solve the set of equations 𝐴�⃗⃗� 𝑛+1 = �⃗� 𝑛, 9obtain the pressure from equation (4) and, iterate the steps (3-9) to reach convergence criteria. results and discussions for the implementation of the numerical solution in the computational space, the present procedure is tested for some steady-state and unsteady flows. in all cases, the van albada tvd scheme was used. for steady-state and unsteady flows, the cfl number was set to 10 and 5, respectively. 4.1 steady-state inviscid flows the results of inviscid subsonic, transonic and supersonic flow calculations with the tvd scheme over a bump in a channel and on a naca 0012 airfoil are presented. for the bump test, at the inlet of the channel, all flow variables are specified if a supersonic flow is considered. for subsonic inlet flow, stagnation pressure p0, stagnation temperature t0 and the inlet angle are specified. at the outlet boundary, all the flow variables are extrapolated for the supersonic regime. the pressure is fixed for the subsonic outlet flows. also, the slip boundary conditions are used on the upper and lower walls. a non-uniform grid of 90×30 in which the grid lines are closely packed in and near the bump region is shown in figure 3. figure 3. bump geometry. for the airfoil case, far-field boundary condition at the outlet and slip boundary conditions on the airfoil are used. a 340×40 c-type grid with excellent orthogonality is used in this case, so that the far-field boundaries were at least twenty chord lengths away. figure 4 shows a close-up view of this grid. simulation of compressible flow using a semi-implicit tvd scheme issn: 2180-1053 vol. 10 no.1 january – june 2018 87 figure 4. a view of c-type grid used in flow computations. in the first test, static to stagnation pressure ratio was used to give a mach number of 0.5 at the inlet of 10% thick bump. the mach number distribution on the lower and upper walls are compared to tvd results obtained by eidelman, colella & shreeve (1984) in figure 5. it is seen that two mach number distributions are very similar. figure 5. mach number values on lower and upper walls of bump for m∞=0.5. for the transonic test, flow past a naca 0012 airfoil with free-stream mach number of 0.8 and angle of attack α=1.25 deg is solved and the pressure distribution is compared with tvd results by pulliam & steger (1985) in figure 6. the results are very similar. only the minimum pressure coefficient on the lower surface of the airfoil in the present study is 6% smaller. locations of the shock waves can be seen on mach contour distributions as shown in figure 7. journal of mechanical engineering and technology 88 issn: 2180-1053 vol. 10 no.1 january – june 2018 figure 6. pressure distribution on naca 0012 airfoil for m∞=0.8 and α=1.25 o). figure 7. contours of mach number on naca 0012 airfoil for m∞=0.8 and α=1.25 o. the third case is supersonic flow with m∞=1.4 over 4% thick bumps on a channel wall. figure 8 shows the mach number distribution on the upper and lower surfaces. these results are compared with the results of tvd scheme obtained by djavareshkian & rezazadeh (2007). this comparison shows that the resolution of the leading edge shock, the reflection of leading edge shock at the upper wall and trailing edge shock for two schemes are very close together. this verifies the validity of the present high-resolution scheme for supersonic flow. simulation of compressible flow using a semi-implicit tvd scheme issn: 2180-1053 vol. 10 no.1 january – june 2018 89 figure 8. mach number values on lower and upper walls of bump for m∞=1.4. 4.2 steady-state viscous flows the viscous flow calculations with the tvd scheme over a naca 0008 airfoil and a circular cylinder are carried out at subsonic regime with m∞=0.1. no-slip conditions on the solid boundary are used. a 340×40 c-type grid and a 150×60 o-type grid are used in naca 0008 airfoil and cylinder case, respectively. in the first viscous flow test, flow around a circular cylinder with re=40 is simulated and the surface pressure distribution is compared to experimental data (grove et al., 1964) and numerical results from (sen, mittal & biswas, 2009) as shown in figure 9. figure 9. pressure distribution on a circular cylinder for re=40 and m∞=0.1. the second viscous case is the flow past a naca 0008 airfoil with chord-base re=6000 and α=4 deg. figure 10 shows the temporal variation of the drag and lift coefficients. the simulations are run for a relatively large time duration of τ*=τm∞=20 at which point the force on the foil reaches a nearly constant value. these lift and drag coefficients are journal of mechanical engineering and technology 90 issn: 2180-1053 vol. 10 no.1 january – june 2018 compared with the numerical simulations by mittal et al. (2008), and it is found that the current methodology provides a reasonably good prediction of these fundamental quantities. figure 10. temporal variation of drag and lift coefficients for naca 0008 airfoil at α=4° for re=6000. 4.3 unsteady inviscid flow the present method is used to calculate the flow over a naca 0012 airfoil pitching around its quarter-chord point. experimental data were provided by landon (1982). the following equation describes the pitching motion of the airfoil: (22) 𝛼(𝑡) = 𝛼𝑚 + 𝛼0𝑠𝑖𝑛(𝜔𝑡) where αm is the main angle of attack and α0 is the angular amplitude. the angular frequency ω is related to the reduced frequency defined as follow: (23) 𝑘 = 𝜔𝑐/2𝑈∞ where c is the airfoil chord length. the grid used in the unsteady-flow calculations is the same as those used in the other steady-flow computations. the criterion for the convergence of the computations is that the maximum magnitude of the normalized residual must be reduced by more than six orders of magnitude as follow: (24) 𝑅(𝑛) = 𝑚𝑎𝑥 (| �⃗⃗� 𝑘+1 − �⃗⃗� 𝑘 �⃗⃗� 𝑟𝑒𝑓 |) < 10−6 where �⃗⃗� 𝑟𝑒𝑓 is a reference value for �⃗⃗� and subscript k denotes the previous iteration. the present method is validated by comparing with numerical results obtained by gao et al. (2005). results are also compared with experimental data (landon 1982).the airfoil is a simulation of compressible flow using a semi-implicit tvd scheme issn: 2180-1053 vol. 10 no.1 january – june 2018 91 naca 0012 pitching at the free-stream mach number m∞ = 0.755, αm = 0.016 deg, α0 = 2.51 deg, and k = 0.0814. the experimental re = 5.5×106. the comparisons of the present inviscid computations, numerical results (gao et al., 2005) and the experimental data (landon, 1982) of the instantaneous lift and moment coefficients versus the instantaneous angle of attack are presented in figure 11. in order to observe the convergence process, the normalized residual is plotted in figure 12. it can be seen that the convergence is well taken. figure 11. comparison of lift and moment coefficients on naca 0012 airfoil m∞ = 0.755, αm = 0.016 deg, α0 = 2.51 deg, k = 0.0814. figure 12. convergence process for naca 0012 airfoil. journal of mechanical engineering and technology 92 issn: 2180-1053 vol. 10 no.1 january – june 2018 4.4 unsteady viscous flow in order to test the accuracy of the viscous flow solutions for the unsteady problem, a rotating circular cylinder is studied. some calculations are carried out for re=200 and ratio of the surface speed to free-stream velocity, k =0.5 with an impulsive start. streamlines for non-dimensional time τ*=τm∞ are compared with the experimental results from coutanceau & menard (1985) as shown in figure 13. good agreement is seen between the computational and experimental results for all times. the next study is to compare the lift coefficients of a rotating cylinder with numerical results from (teymourtash & salimipour 2017). since in this reference, an incompressible flow is simulated, a free-stream mach number of 0.05 is used to prevent the compressibility effects. figure 14 shows these comparisons with respect to the nondimensional time τ* for 0 ≤ k ≤ 5 and re=200. the results show excellent agreement. figure 13. comparison of instantaneous streamlines past a rotating cylinder for α=0.5, re=200 and m∞=0.1. simulation of compressible flow using a semi-implicit tvd scheme issn: 2180-1053 vol. 10 no.1 january – june 2018 93 figure 14. comparison of cylinder lift coefficients for k=0-5 and re=200. conclusions a high-resolution scheme has been implemented in a density-based, finite-volume procedure which uses a semi-implicit solution algorithm. the mentioned scheme based on total variation diminishing (tvd) is developed to compute the fluxes of the convected quantities, including mass flux. the tvd scheme removes the undesirable oscillations generated by numerical errors, with preserving the solution accuracy. the method is applied to subsonic, transonic and supersonic flows, and the results have been compared with predicted data by the other tvd schemes based on the characteristic variable or experimental data. these comparisons show that the present high-resolution scheme predicts shock waves and unsteady boundary layer with high accuracy for both steady and unsteady flows. the discretization and the deriving linear equations system show that the present procedure can be easily implemented to solve the inviscid and viscous flows for a wide range of mach numbers. journal of mechanical engineering and technology 94 issn: 2180-1053 vol. 10 no.1 january – june 2018 references blazek, j. (2001). computational fluid dynamics: principles and applications. amsterdam, london, new york: elsevier. coutanceau, m. & menard, c. (1985). influence of rotation on the near-wake development behind an impulsively started circular cylinder. journal of fluid mechanics, 158, 399-466. djavareshkian, m.h. & reza-zadeh, s. (2007). application of normalized flux in pressure-based algorithm. computers & fluids, 36, 1224–1234. eidelman, s., colella, p. & shreeve, r.p. (1984). application of the godunov method and its second-order extension to cascade flow modeling. aiaa journal, 22(11), 1609–1615. gao, c., yang, s., luo, s., liu, f. & schuster, d. (2005). calculation of airfoil flutter by an euler method with approximate boundary conditions. aiaa journal, 43(2), 295–305. grove, a.s., shair, f.h., peterson, e.e. & acrivos, a. (1964). an experimental investigation of the steady separated flow past a circular cylinder. journal of fluid mechanics, 19, 60–80. harten a. (1983). high resolution schemes for hyperbolic conservation laws. journal of computational physics, 49, 357–393. hou, j., simons, f. & hinkelmann, r. (2012). improved total variation diminishing schemes for advection simulation on arbitrary grids. international ournal for numeical methods in fluids, 70, 359–382. jameson, a. & yoon, s. (1986). lu implicit schemes with multiple grids for the euler equations. aiaa 24th aerospace sciences meeting, 1-12. jameson, a. (1981). steady-state solution of the euler equations for transonic flow. advances in scientific computing, 37-70. jameson, a. (1983). numerical solution of the euler equation for compressible inviscid fluids, princeton university, new jersy 08544, usa, 1-50. jameson, a. (1991). time-dependent calculations using multigrid, with applications to unsteady flows past airfoils and wings. aiaa 10th computational fluid dynamics conference, 1-12. jameson, a. (2009). an assessment of dual-time stepping, time spectral and artificial compressibility based numerical algorithms for unsteady flow with applications to flapping wings. 19th aiaa computational fluid dynamics conference, 1-20. simulation of compressible flow using a semi-implicit tvd scheme issn: 2180-1053 vol. 10 no.1 january – june 2018 95 jespersen, d.c. & pulliam, t.h. (1983). approximate newton methods and flux vector splitting. aiaa paper, 83-1899. kapen, p.t. & tchuen, g. (2015). an extension of the tv-hll scheme for multidimensional compressible flows. international journal of computational fluid dynamics, 29(3-5), 303–312. landon, r.h. (1982). naca 0012 oscillating and transient pitching, compendium of unsteady aerodynamic measurements, data set 3. agard, rept. r-702. mittal, r., dong, h., bozkurttas, m., najjar, f.m., vargas, a. & loebbecke, a. (2008). a versatile sharp interface immersed boundary method for incompressible flows with complex boundaries. journal of computational physics, 227, 4825–4852. pulliam, t.h. & steger, j.l. (1985). recent improvements in efficiency, accuracy, and convergence for implicit approximate factorization algorithms. aiaa 23th aerospace sciences meeting, 1-37. ravichandran, k.s. (1997). explicit third order compact upwind difference schemes for compressible flow calculations. international journal of computational fluid dynamics, 8, 311–316. renac, f., gérald, s., marmignon, c. & coquel, f. (2013). fast time implicit–explicit discontinuous galerkin method for the compressible navier–stokes equations. journal of computational physics, 251, 272–291. roe p.l. (1985). some contributions to the modeling of discontinuous flows. lectures in applied mathematics, 22, 163–193. sen, s., mittal, s. & biawas, g. (2009). steady separated flow past a circular cylinder at low reynolds numbers. journal of fluid mechanics, 620, 89–119. sweby p.k. (1984). high resolution schemes using flux limiters for hyperbolic conservation laws. siam journal on numerical analysis, 21, 995–1011. teixeira, r.s. & alves, l.s.b. (2012). modelling far-field entrainment in compressible flows. international journal of computational fluid dynamics, 26(1), 67–78. teymourtash, a.r. & salimipour, s.e. (2017). compressibility effects on the flow past a rotating cylinder. physics of fluids, 29, 016101. waterson, n.p. & deconinck, h. (2007). design principles for bounded higher-order convection schemes – a unified approach. journal of computational physics, 224, 182–207. journal of mechanical engineering and technology 96 issn: 2180-1053 vol. 10 no.1 january – june 2018 yee, h.c., warming, r.f. & harten, a. (1983). implicit total variation diminishing (tvd) schemes for steady-state calculations. nasa technical memorandum 84342. yoon, s. & jameson, a. (1986). a multigrid lu-ssor scheme for approximate newton iteration applied to the euler equations. nasa-cr-179524 zhang, di., jiang, ch., cheng, l. & liang d. (2015). a refined r‐factor algorithm for tvd schemes on arbitrary unstructured meshes. international journal for numerical methods in fluids, 80(2), 105-139. nomenclature a∞ free-stream speed of sound u, v cartesian velocity components c airfoil chord length m∞ freestream mach number cd drag coefficient α angle of attack (deg) cl lift coefficient θ surface angle respect to cylinder center cm moment coefficient k ratio of surface speed to free-stream velocity re reynolds number τ time cp pressure coefficient __ mean value superscript p pressure issn: 2180-1053 vol. 7 no. 1 january june 2015 correlation of lateral and yaw analysis responses to tracking of linearlized rail wheelset model 31 correlation of lateral and yaw analysis responses to tracking of linearlized rail wheelset model z. a. soomro1* 1directorate of post-graduate studies 1mehran university of enggineering and technology, jamshoro, pakistan abstract controlling analysis for lateral and yaw motion has great influence on rail wheelset model during running. to avoid slip due to lower adhesion, sliding due to higher speeds and balance of creep forces is attributed by degree of freedom (dof). here simple dynamics of lateral and yaw motions in terms of displacement, velocity and acceleration has been discussed. this correlation depending upon creep co-efficient is shaped into matrix form too. in this paper an effort has been consumed to observe the correlation and behavior of the lateral and yaw motion analysis based upon its geometry. the railway wheelset is modeled depending upon preliminary values of parameters. the step response of lateral and yaw motion has been performed to study attitude of each other based upon time variant. keywords: creep co-efficient, conicity, forward velocity, angular velocity, degree of freedom 1.0 introduction it has been observed that mechanical systems need explanation of concerned elemental motion of bodies realizing their large movements, pertaining complicated interaction with their surrounding environment. the interaction for multi-body systems is conceived as multiple-body systems connected by different types of kinematic pairs, and forces acting upon it allow them to study the dynamic phenomena occurring in the during dynamic position (polach, 2005). for correct modeling, all eigen damping properties of rubber elements, as well as other parasitic damping have to be considered in the model parameters. the modes and nomenclature used for car body eigen behavior are shown necessary (lee, 2005). the sway mode, as combined lateral movement and rotation about longitudinal axis, is present in two forms with different heights of the rotation center lower sway mode and upper * corresponding author email: zulfiqarali_s@yahoo.com issn: 2180-1053 vol. 7 no. 1 january june 2015 journal of mechanical engineering and technology 32 sway mode (anant, 2004). the difference in the rolling radii of the two wheels is created when the wheelset is moved to one side due to curved path. since the wheels are rigidly connected together by the solid axle. these wheels usually spin at the same rate. the forward velocity of the first wheel becomes larger than the forward velocity of the second wheel on diversion of path. this creates a rotation of the axle towards the center of the track position, with the angle of yaw. this continues to increase until the axle center goes back to the middle of the track position. the motion changes the solid axle oscillation from side to side accompanied by lateral and yaw motion referred as hunting during the running (baldovin, 2012). the hunting motion caused by railway vehicles is combined by lateral and yaw self-oscillatory motion, which can largely be determined by wheel–rail contact geometry of rail wheelset. the stability of this occurred motion is an important dynamic problem that solely depends upon the railway vehicles speed to determine the maximum operating speed of the rail vehicle (yabuno, 2002). the lateral movement of the wheelset creates the rolling radii difference that allows the wheelset to roll through the curve. as the curvature increases, the wheelset displaces more and generates a larger difference of the wheel radii, increasing the rate of yaw rotation. the yaw rotation rate, and hence the curvature that the wheelset can smoothly roll through, is limited by the maximum wheel radii difference. this, in turn, is limited by the clearance between the wheel flanges and the rail as the gauge clearance (dukkipatti, 2002). the path curvature of rail wheelset cannot directly be measured, and lateral acceleration is due to disturbances created like body roll; also, the lateral acceleration depends upon sensor location, different to the yaw rate, which is only sensitive to sensor orientation installed (snen et al. 2006). it is also important to note that due to simple conditions of constant speed, the minimal rail vehicle sideslip occurs when the vehicle is in a normal stable condition and due to the negligible roll angle of body, the three mentioned variables path curvature, yaw rate and lateral acceleration are actually proportional to each other on observation (zelenka et al.2010) in this paper, the behavior of lateral and yaw dynamics has huge influence upon the control and proper running of rail wheelset both on straight and curved track. these are fundamental motions based upon adhesion level and creep forces affecting braking system of railway vehicle system are also enumerated. issn: 2180-1053 vol. 7 no. 1 january june 2015 correlation of lateral and yaw analysis responses to tracking of linearlized rail wheelset model 33 2.0 rail wheelset dynamics the fundamental dynamics railway wheelset model is here classified by two clauses. one pertains to geometry of wheelset model to prepare proper modeling and other relates to basic movement of the railway wheelset during running. 2.1 mathematical modeling if track is considered to be rigid then the wheelset has three degrees of freedom, lateral displacement y, yaw movement ψ, and longitudinal motions x, as shown in figure 1. lateral and yaw motion are very small as compare to longitudinal motion but play an important role in stability and ride comfort of the vehicle (soomro, 2014) 3 figure 1. rail wheelset geometry showing lateral and yaw motion from figure 1, we extract the relation as v=ω.ro , where vl=ω.rl , vr=ω.rr , thus v=(vl + vr)/2 = ω.ro. the constraints in lateral (y) and yaw ( ) are y ̇ = v sin  = v. , and  = y ̇.v, after rearranging we get  =(vl + vr)/2lg = (ω.∆r)/2lg ,  = [v/(lg.ro)] * λ .y (1) where   v y y  , thus v y m f m f y ww   2222 22   (2) where yyff /22  and xxff /22  .    w w t wo g wo g w g w log w rog i k y ir fl y ir fl vi fl vi rfl vi rfl  111111 2 1111 222  (3) t 11 11 2 11 2222 2 0 0 0 2 0 2 0 22 0 1000 0100 y ir fly y vi fl i k ir fl vm f m f y y dt d wo g w g w w wo g ww                                                                     (4) figure 1. rail wheelset geometry showing lateral and yaw motion 3 figure 1. rail wheelset geometry showing lateral and yaw motion from figure 1, we extract the relation as v=ω.ro , where vl=ω.rl , vr=ω.rr , thus v=(vl + vr)/2 = ω.ro. the constraints in lateral (y) and yaw ( ) are y ̇ = v sin  = v. , and  = y ̇.v, after rearranging we get  =(vl + vr)/2lg = (ω.∆r)/2lg ,  = [v/(lg.ro)] * λ .y (1) where   v y y  , thus v y m f m f y ww   2222 22   (2) where yyff /22  and xxff /22  .    w w t wo g wo g w g w log w rog i k y ir fl y ir fl vi fl vi rfl vi rfl  111111 2 1111 222  (3) t 11 11 2 11 2222 2 0 0 0 2 0 2 0 22 0 1000 0100 y ir fly y vi fl i k ir fl vm f m f y y dt d wo g w g w w wo g ww                                                                     (4) issn: 2180-1053 vol. 7 no. 1 january june 2015 journal of mechanical engineering and technology 34 3 figure 1. rail wheelset geometry showing lateral and yaw motion from figure 1, we extract the relation as v=ω.ro , where vl=ω.rl , vr=ω.rr , thus v=(vl + vr)/2 = ω.ro. the constraints in lateral (y) and yaw ( ) are y ̇ = v sin  = v. , and  = y ̇.v, after rearranging we get  =(vl + vr)/2lg = (ω.∆r)/2lg ,  = [v/(lg.ro)] * λ .y (1) where   v y y  , thus v y m f m f y ww   2222 22   (2) where yyff /22  and xxff /22  .    w w t wo g wo g w g w log w rog i k y ir fl y ir fl vi fl vi rfl vi rfl  111111 2 1111 222  (3) t 11 11 2 11 2222 2 0 0 0 2 0 2 0 22 0 1000 0100 y ir fly y vi fl i k ir fl vm f m f y y dt d wo g w g w w wo g ww                                                                     (4) for a single wheelset, the second-order differential equations that represent the relationship of creep damping and creep stiffness coefficient with lateral and yaw displacement can been shown below considering the mass, lateral force and external yaw torque. 4 for a single wheelset, the second-order differential equations that represent the relationship of creep damping and creep stiffness coefficient with lateral and yaw displacement can been shown below considering the mass, lateral force and external yaw torque. (5) (6) these mathematical wheelset model parameters will be used for simulating results through table 1 values to observe behavior of lateral and yaw motions (soomro, 2015). 2.2 degree of freedom railway model the figure 2, shows the possible movements of railway vehicle with respect to ‘x’ in horizontal plane and ‘y’ both in lateral and yawing motions. while rolling wrt ‘y’ horizontal plane and ‘z’ in vertical plane and bouncing and pitching occur in ‘z’ vertical plane and ‘x’ horizontal planes. the railway wheelset has basic three degree of freedom (longitudinal, lateral and yawing), but here in figure 2, six possible dof are shown which are related with eachother. figure 2. possible types of motions created by rail vehicle during running these mathematical wheelset model parameters will be used for simulating results through table 1 values to observe behavior of lateral and yaw motions (soomro, 2015). 2.2 degree of freedom railway model the figure 2, shows the possible movements of railway vehicle with respect to ‘x’ in horizontal plane and ‘y’ both in lateral and yawing motions. while rolling wrt ‘y’ horizontal plane and ‘z’ in vertical plane and bouncing and pitching occur in ‘z’ vertical plane and ‘x’ horizontal planes. the railway wheelset has basic three degree of freedom (longitudinal, lateral and yawing), but here in figure 2, six possible dof are shown which are related with eachother. issn: 2180-1053 vol. 7 no. 1 january june 2015 correlation of lateral and yaw analysis responses to tracking of linearlized rail wheelset model 35 4 for a single wheelset, the second-order differential equations that represent the relationship of creep damping and creep stiffness coefficient with lateral and yaw displacement can been shown below considering the mass, lateral force and external yaw torque. (5) (6) these mathematical wheelset model parameters will be used for simulating results through table 1 values to observe behavior of lateral and yaw motions (soomro, 2015). 2.2 degree of freedom railway model the figure 2, shows the possible movements of railway vehicle with respect to ‘x’ in horizontal plane and ‘y’ both in lateral and yawing motions. while rolling wrt ‘y’ horizontal plane and ‘z’ in vertical plane and bouncing and pitching occur in ‘z’ vertical plane and ‘x’ horizontal planes. the railway wheelset has basic three degree of freedom (longitudinal, lateral and yawing), but here in figure 2, six possible dof are shown which are related with eachother. figure 2. possible types of motions created by rail vehicle during running figure 2. possible types of motions created by rail vehicle during running 2.3 proposed linearized modelling a schematic idea for having simplified model is invented as under. 5 2.3 proposed linearized modelling a schematic idea for having simplified model is invented as under. figure 3 . schematic diagram from non-linear to linear modeling here in figure 3, non-linear model railway wheel set model is linearized and simplified to limit it up to lateral and yaw motion analysis. as these are easily calculated and computed to check the behavior of railway locomotive lateral and yaw motion analysis through running. table 1 preliminary parameters with values used for modeling parameter value mass of vehicle (mv) 15000 kg right wheel moment of inertia (ir) 133.2 kgm2 left wheel moment of inertia (il) 62.8 kgm2 wheels radius (ro) 0.5 m torsional stiffness (kt) 6063260 n/m conicity of wheel(γ) 0.15 mass of wheelset (mw) 1250kg curve radius (ro) 100m figure 3 . schematic diagram from non-linear to linear modeling here in figure 3, non-linear model railway wheel set model is linearized and simplified to limit it up to lateral and yaw motion analysis. as these are easily calculated and computed to check the behavior of railway locomotive lateral and yaw motion analysis through running. issn: 2180-1053 vol. 7 no. 1 january june 2015 journal of mechanical engineering and technology 36 table 1 preliminary parameters with values used for modeling 5 2.3 proposed linearized modelling a schematic idea for having simplified model is invented as under. figure 3 . schematic diagram from non-linear to linear modeling here in figure 3, non-linear model railway wheel set model is linearized and simplified to limit it up to lateral and yaw motion analysis. as these are easily calculated and computed to check the behavior of railway locomotive lateral and yaw motion analysis through running. table 1 preliminary parameters with values used for modeling parameter value mass of vehicle (mv) 15000 kg right wheel moment of inertia (ir) 133.2 kgm2 left wheel moment of inertia (il) 62.8 kgm2 wheels radius (ro) 0.5 m torsional stiffness (kt) 6063260 n/m conicity of wheel(γ) 0.15 mass of wheelset (mw) 1250kg curve radius (ro) 100m the different parameters used in the modeling of simplified rail wheelset are enumerated in above table 1. the values of different parameters are mentioned. these values are used by matlab code and simulink blocks to study behavior of lateral and yaw analysis. thus the correlation between these dynamical terms is sketched by concerned graphical diagrams. 3.0 graphic simulation results the lateral and yaw motions created due to running of rail wheelset locomotive are numerated by matlab simulation. following are the various simulation results denoting the correlation of lateral and yaw dynamics. the values mentioned in table 1 are used. in above figure 4, the lateral and yaw displacements have been shown in vertical plane and time in horizontal plane. the black ‘+’ curve shows lateral distance while blue diamond line shows the yaw distance. here lateral curve shows behavior in zigzag making like triangle by points on 3, 6 and 7 on vertical plane inclined within time periods 0.4, 0.7 and 1 sec. whereas yaw line is marked 1, 3, 5 and 8 in 0.1, 0.4, 0.7 and 1 sec with horizontal. this conceives the concept that when yaw reaches in 0.1 to 1 sec in horizontal to 8cm, the lateral displaces upward direction in 0.4, 0.7 and 1 sec to complete 7cm, yaw goes upward and laterally in triangular direction. from this graph, it can be assumed that both lateral and yaw are proportional with each in different directions on running rail wheelset. the lateral and yaw distances are usually smaller during the running of railway train to increase in minor steps within span of time. from below graph it is observed, that yaw distance is greater than lateral within same time. issn: 2180-1053 vol. 7 no. 1 january june 2015 correlation of lateral and yaw analysis responses to tracking of linearlized rail wheelset model 37 6 the different parameters used in the modeling of simplified rail wheelset are enumerated in above table 1. the values of different parameters are mentioned. these values are used by matlab code and simulink blocks to study behavior of lateral and yaw analysis. thus the correlation between these dynamical terms is sketched by concerned graphical diagrams. 3.0 graphic simulation results the lateral and yaw motions created due to running of rail wheelset locomotive are numerated by matlab simulation. following are the various simulation results denoting the correlation of lateral and yaw dynamics. the values mentioned in table 1 are used. in above figure 4, the lateral and yaw displacements have been shown in vertical plane and time in horizontal plane. the black ‘+’ curve shows lateral distance while blue diamond line shows the yaw distance. here lateral curve shows behavior in zigzag making like triangle by points on 3, 6 and 7 on vertical plane inclined within time periods 0.4, 0.7 and 1 sec. whereas yaw line is marked 1, 3, 5 and 8 in 0.1, 0.4, 0.7 and 1 sec with horizontal. this conceives the concept that when yaw reaches in 0.1 to 1 sec in horizontal to 8cm, the lateral displaces upward direction in 0.4, 0.7 and 1 sec to complete 7cm, yaw goes upward and laterally in triangular direction. from this graph, it can be assumed that both lateral and yaw are proportional with each in different directions on running rail wheelset. the lateral and yaw distances are usually smaller during the running of railway train to increase in minor steps within span of time. from below graph it is observed, that yaw distance is greater than lateral within same time. 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1 2 3 4 5 6 7 8 time(sec) d is ta nc e lateral and yaw distance lateral(cm) yaw(rad) figure 4 relation of lateral and yaw displacement of rail wheelset figure 4 relation of lateral and yaw displacement of rail wheelset in figure 5 below, the relation between lateral velocity and yaw velocity has been shown in vertical plane with respect to time in horizontal plane. the black ‘+’ curve shows lateral velocity while blue diamond line shows the yaw velocity. here lateral curve shows behavior in zigzag making like triangle by points from 0 0n 0.1 sec to reach upward 4000 on 0.4 sec then declines below 0 on 0.7 sec and end up to zero in 1 sec on horizontal plane. whereas yaw line is marked below -1000 on 0.1 to -3000 in 0.4 sec and upwards to 0 in 0.7 sec with horizontal plane and ends below 0 on -200 rad/sec in 1.0 second on horizontal side. this conceives the concept that yaw velocity varies differently to the lateral velocity on 4000 cm/sec in upward direction in 04 seconds. yaw goes upward to zero with a little bit downfall of lateral velocity on -200 cm/ sec in 0.7 seconds. thus lateral velocity ends at zero in one second. from this graph, it can be assumed that both lateral and yaw are initially more inversely proportional with each in different directions and after 0.7 second its inverse proportion decreases to end after one second. the lateral and yaw velocities are expressed in ‘cm/sec’ and ‘rad/sec’ respectively. issn: 2180-1053 vol. 7 no. 1 january june 2015 journal of mechanical engineering and technology 38 7 in figure 5 below, the relation between lateral velocity and yaw velocity has been shown in vertical plane with respect to time in horizontal plane. the black ‘+’ curve shows lateral velocity while blue diamond line shows the yaw velocity. here lateral curve shows behavior in zigzag making like triangle by points from 0 0n 0.1 sec to reach upward 4000 on 0.4 sec then declines below 0 on 0.7 sec and end up to zero in 1 sec on horizontal plane. whereas yaw line is marked below -1000 on 0.1 to -3000 in 0.4 sec and upwards to 0 in 0.7 sec with horizontal plane and ends below 0 on -200 rad/sec in 1.0 second on horizontal side. this conceives the concept that yaw velocity varies differently to the lateral velocity on 4000 cm/sec in upward direction in 04 seconds. yaw goes upward to zero with a little bit downfall of lateral velocity on -200 cm/sec in 0.7 seconds. thus lateral velocity ends at zero in one second. from this graph, it can be assumed that both lateral and yaw are initially more inversely proportional with each in different directions and after 0.7 second its inverse proportion decreases to end after one second. the lateral and yaw velocities are expressed in ‘cm/sec’ and ‘rad/sec’ respectively. . 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 -3000 -2000 -1000 0 1000 2000 3000 4000 time(sec) ve lo ci ty lateral and yaw velocities lateral(cm/s) yaw(rad/s) figure 5. lateral and yaw velocity behavior with each other on running wheelset figure 5. lateral and yaw velocity behavior with each other on running wheelset 8 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 -8 -6 -4 -2 0 2 4 6 8 10 time (sec) ac ce ler at ion lateral and yaw acceleration lateral(cm/s2) yaw(rad/s2) figure 6. acceleration of lateral and yaw of rail wheelset in figure 6, the relation between later acceleration denoted by black ‘+’ curve and yaw acceleration by ‘blue diamond’ are shown. here lateral acceleration starts from 0 on 0.1 sec hyperabolly upto nearby below 10 0n 0.4 sec and ends on zero on 0.7 sec to 1 sec in straight line. while yaw blue line starts from -4 0n 0.1 sec to 0.4 sec on below -6 on vertical plane and jumps upto 0 on 0.7 sec and ends on 1 sec in straight line. from this it can be assesed that these acceleration curves initialy behave inversely and then both end to go in straight line on zero from 0.7 sec to 1 sec. this displays unstability, where both lateral and yaw do not oppose each other. the lateral acceleration is expressed in terms of ‘cm/sec2’ and yaw or spin acceleration is denoted by ‘rad/sec2’ as displayed in figure 6. the below figure 7, has been extracted by simulink block diagram. in this figure 7, step response has been shown for lateral motion in vertical direction to time in horizontal plane. here lateral curve starts initialy and slighly in straight line to jump up parabolicaly 2e-3 m/s (20 cm/s) on vertical scale in 1 sec. then it goes downward to below zero on -0.6e-3 m/sec in 2 sec and then travels to zero path in 3 sec to end in 10 sec. this shows again unstability of lateral motion in straight line. thus acceleration expressed in ‘cm/sec2’ is represented by lateral parameter, whereas yaw shows as deceleration in opposite direction. these both then travel in straight path on increase of speed and span of time. figure 6. acceleration of lateral and yaw of rail wheelset in figure 6, the relation between later acceleration denoted by black ‘+’ curve and yaw acceleration by ‘blue diamond’ are shown. here lateral acceleration starts from 0 on 0.1 sec hyperabolly upto nearby below 10 0n 0.4 sec and ends on zero on 0.7 sec to 1 sec in straight line. while yaw blue line starts from -4 0n 0.1 sec to 0.4 sec on below -6 on vertical plane and jumps upto 0 on 0.7 sec and ends on 1 sec in straight line. from this it can be assesed that these acceleration curves initialy behave inversely and then both end to go in straight line on zero from 0.7 sec to 1 sec. this displays unstability, where both lateral and yaw do not oppose each other. the lateral acceleration is expressed in terms of ‘cm/sec2’ and yaw or spin acceleration is denoted by ‘rad/sec2’ as displayed in issn: 2180-1053 vol. 7 no. 1 january june 2015 correlation of lateral and yaw analysis responses to tracking of linearlized rail wheelset model 39 figure 6. the below figure 7, has been extracted by simulink block diagram. in this figure 7, step response has been shown for lateral motion in vertical direction to time in horizontal plane. here lateral curve starts initialy and slighly in straight line to jump up parabolicaly 2e-3 m/s (20 cm/s) on vertical scale in 1 sec. then it goes downward to below zero on -0.6e-3 m/sec in 2 sec and then travels to zero path in 3 sec to end in 10 sec. this shows again unstability of lateral motion in straight line. thus acceleration expressed in ‘cm/sec2’ is represented by lateral parameter, whereas yaw shows as deceleration in opposite direction. these both then travel in straight path on increase of speed and span of time. 9 0 1 2 3 4 5 6 7 8 9 10 -1 -0.5 0 0.5 1 1.5 2 2.5 x 10 -3 time la te ra l m ot io n step response figure 7. step response of lateral motion of rail vehicle wheelset 0 1 2 3 4 5 6 7 8 9 10 -4 -3 -2 -1 0 1 2 3 4 x 10 -3 time (sec) lateral step responses of railway wheelset ya w m ot io n (ra d) constrained un-constrained figure 8. constrained and unconstrained yaw motion of rail locomototive figure 8 above, shows the yaw motion of rail wheelset by simulink through step response. here yaw motion is displayed both by unconstrained mode by green lines in zigzag form and constrained by blue line. the constrained mode starts initialy wih slight parabolicaly figure 7. step response of lateral motion of rail vehicle wheelset 9 0 1 2 3 4 5 6 7 8 9 10 -1 -0.5 0 0.5 1 1.5 2 2.5 x 10 -3 time la te ra l m ot io n step response figure 7. step response of lateral motion of rail vehicle wheelset 0 1 2 3 4 5 6 7 8 9 10 -4 -3 -2 -1 0 1 2 3 4 x 10 -3 time (sec) lateral step responses of railway wheelset ya w m ot io n (ra d) constrained un-constrained figure 8. constrained and unconstrained yaw motion of rail locomototive figure 8 above, shows the yaw motion of rail wheelset by simulink through step response. here yaw motion is displayed both by unconstrained mode by green lines in zigzag form and constrained by blue line. the constrained mode starts initialy wih slight parabolicaly figure 8. constrained and unconstrained yaw motion of rail locomototive issn: 2180-1053 vol. 7 no. 1 january june 2015 journal of mechanical engineering and technology 40 figure 8 above, shows the yaw motion of rail wheelset by simulink through step response. here yaw motion is displayed both by unconstrained mode by green lines in zigzag form and constrained by blue line. the constrained mode starts initialy wih slight parabolicaly upto two steps of green line in same paths lower than green curve. thus both go in straight path to end in 10 seconds. the unconstrained parameter moves in the range of the 3e-3 to -3e-3 radians. while the constrained line moves upward 2e-3 to -1e-3 radians in range, which shows its stability better that of unconstrained parameter for yaw motion in vertical direction. 4.0 conclusions from above analysis it is concluded that rail wheelset model is linearlised and simplified to the dynamics of lateral and yaw motion analysis related to longitudinal and later creep of co-efficient, skipping other affecting factors application to railway wheelset model. a suitable matlab code has been created to observe the behavior of lateral motion analysis to yaw motion analysis through curves with each other. a simulink block program has also been established to study step response of lateral and yaw motions separately. from the modeling the dynamics of lateral and yaw motions, it is accessed that both lateral and yaw vary in zigzag making triangles whereas yaw displacement starts initially horizontally. while velocity analysis of both lateral and yaw for wheelset vary inversely with each other. this inverse variation minimizes after some time before end. the acceleration analysis starts with same behavior and after some this inverse proportion is ended to travel in same path overlapping each other. in simulink the individual behavior of both lateral motion and yaw with constrained and unconstrained is displayed. from this whole phenomenon it can be conceived that lateral motion and yaw behavior verses in opposite directions in velocities and at last in acceleration analysis. nonmenclature y is lateral displacement of wheelset ψ is yaw angle (angle of attack) fy is lateral force tψ is yaw torque mw is mass of wheelset iw is moment of inertia for wheelset f11 is longitudinal creep co efficient f22 is lateral creep coefficient issn: 2180-1053 vol. 7 no. 1 january june 2015 correlation of lateral and yaw analysis responses to tracking of linearlized rail wheelset model 41 references polach, o. (2005). influence of wheel/rail contact geometry on the behaviour of the railway vehicle at stability limit. enoc 2005, eindhoven, netherlands, 7(12), 2203-2210. lee, s.y., cheng, y.c. (2005). hunting stability analysis of high-speed railway vehicle trucks on tangent tracks. journal of sound and vibration, 282, 881-898. anant, m., mehdi, a. (2004). nonlinear investigation of the effect of primary suspension on the hunting stability of a rail wheelset. asme rail transportation division, 27, 53-61. baldovin d., baldovin s. (2012). the lateral stability of a bogie with independently rotating wheelsets. the annual simposium of the inst. of solid mechanics, sisom. yabuno, h., okamoto, t., aoshima, n. (2002). effect of lateral linear stiffness on nonlinear characteristics of hunting motion of a railway wheelset. meccanica, 37, 555-568. dukkipatti, r.v., narayana s, s. (2002). lateral stability and steady state curving performance of unconventional rail tracks. jsme international journal, series c, 45(1). c snen (2006). railway applications testing for the acceptance of running characteristics of railway vehicles testing of running behavior and stationary tests (in czech). czech institute for normalisation. zelenka, j., mich´alek, t. (2010). a new method of the assessment of rail vehicles guiding behavior in small-radius curves. international journal of applied mechanics and engineering 15(2), 511–519. soomro. za, i. hussain, chowdhary. bs, (2014). creep forces analysis at wheelrail contact patch to identify adhesion level to control slip on railway track. new horizons, journal of ieeep, 14-17. soomro. za, (2015). step response and estimation of lateral and yaw motion disturbance of rail wheel set. journal of engineering and technology 5(1), 1-8 issn: 2180-1053 vol. 3 no. 2 july-december 2011 application of lattice boltzmann method in predicting flow of shear driven cavities 55 application of lattice boltzmann method in predicting flow of shear driven cavities fudhail abdul munir1, nor azwadi che sidik2, mohd irwan mohd azmi1 , mohd rody mohamad zin1 1faculty of mechanical engineering, universiti teknikal malaysia melaka, hang tuah jaya, 76100, durian tunggal, melaka, malaysia 2universiti teknologi malaysia,86100,skudai,johor, malaysia email: 1fudhail@utem.edu.my abstract in this paper, prediction of fluid flow in shear driven cavities is presented. lattice boltzmann method is used as the alternative to conventional computational fluid dynamics. the geometry of shear driven cavities as well as the reynolds numbers is varied. the simulation is conducted for three types of shear driven cavities which are square cavity and triangular cavities. the obtained streamline patterns and the centre of vortex for each type is in excellent agreement with benchmark results. it is also found that the streamline patterns is significantly affected by the geometrical shape of cavities. keywords: lattice boltzmann method, computational fluid dynamics, shear driven cavites, streamline patterns. 58 application of lattice boltzmann method in predicting flow of shear driven cavities fudhail abdul munir1, nor azwadi che sidik2, mohd irwan mohd azmi1 mohd rody mohamad zin1 1faculty of mechanical engineering, universiti teknikal malaysia melaka, hang tuah jaya, 76100, durian tunggal, melaka, malaysia 2universiti teknologi malaysia,86100,skudai,johor, malaysia email: 1fudhail@utem.edu.my abstract in this paper, prediction of fluid flow in shear driven cavities is presented. lattice boltzmann method is used as the alternative to conventional computational fluid dynamics. the geometry of shear driven cavities as well as the reynolds numbers is varied. the simulation is conducted for three types of shear driven cavities which are square cavity and triangular cavities. the obtained streamline patterns and the centre of vortex for each type is in excellent agreement with benchmark results. it is also found that the streamline patterns is significantly affected by the geometrical shape of cavities. keywords: lattice boltzmann method, computational fluid dynamics, shear driven cavites, streamline patterns. 1.0 introduction recently, due to rapidly increasing computational power, computational methods have become the essential tools to conduct researches in various engineering fields. in parallel to the development of high speed digital computer, computational fluid dynamics (cfd) has become the new third approach apart from theory and experiment in the philosophical study and development of the whole discipline of fluid dynamics (anderson, 1995). solving the famous navier-stokes equation would require the knowledge of cfd since the non-linearity and complexity of the equation making it that there is currently no analytical solution to these equations except for a small number of special cases (sidik, 2007). a few examples of numerical methods were introduced by experts in cfd field in order to solve the navier-stokes equation numerically. the methods are like finite difference method, finite element method and finite volume method. the lattice boltzmann method (lbm) has become considerably alternative method to solve fluid flow (munir et al , 2011). the way lbm works is by predicting the evolution of particle distribution function and calculates the macroscopic variables by taking moment to the distribution function. issn: 2180-1053 vol. 3 no. 2 july-december 2011 journal of mechanical engineering and technology 56 59 the basic idea of boltzmann work is that a gas is composed of interacting particles that can be explained by classical mechanics. the mechanics can be very simple where it contains streaming in space and billiard-like collisions interactions (sidik, 2007). the starting point in lbm scheme is by tracking the evolution of the single-particle distribution. the concept of particle distribution has already well developed in the field of statistical mechanics while discussing the kinetics theory of gases and liquids. the definition implies that the probable number of molecules in a certain volume at certain time made from a huge number of particles in a system that travel freely, without collisions, for distances (mean free path) long compared to their sizes. once the distribution functions are obtained, the hydrodynamics equation can be derived. there are many advantages of lbm as compared to the conventional computational fluid dynamics. one of the main merits of lbm is that it has been proven successfully able to solve compressible navier-stokes equations (malapinas et al., 2010). apart from that, the algorithm of lbm can be easily re-worked to enable it to be applied on more complex simulation components (mohd irwan et al., 2010). 1.1 mesoscale lattice boltzmann model ludwig boltzmann (1844-1906) introduced a transport equation based on statistical mechanics describing the evolution of gas particle in a system as;          c f a x f c t f (1) where  f ,  c,  a and c stand for density distribution function, mesoscopic speed, acceleration due to external force and collision function respectively. if there is no external force, eq. (1) is no more than a hyperbolic wave equation with source term given as       x f c t f (2) any solution of the boltzmann equation, eq. (2), requires an expression for the collision operator . if the collision is to conserve mass, momentum and energy, it is required that 0 1 2             dc c c (3) however, the expression for  is too complex to be solved. even if we only consider twobody collision, the collision integral term needs to consider the scattering angle of the binary collision, the speed and direction before and after the collision, etc. any replacement of collision must satisfy the conservation law as expressed in eq. (3). the idea behind this replacement is that large amount of detail of two-body interaction is not likely to influence significantly the values of many experimental measured quantities (succi, 2001). there are a few version of collision operator published in the literature. however, the most well accepted version due to its simplicity and efficiency is the bhatnagar gross crook 58 application of lattice boltzmann method in predicting flow of shear driven cavities fudhail abdul munir1, nor azwadi che sidik2, mohd irwan mohd azmi1 mohd rody mohamad zin1 1faculty of mechanical engineering, universiti teknikal malaysia melaka, hang tuah jaya, 76100, durian tunggal, melaka, malaysia 2universiti teknologi malaysia,86100,skudai,johor, malaysia email: 1fudhail@utem.edu.my abstract in this paper, prediction of fluid flow in shear driven cavities is presented. lattice boltzmann method is used as the alternative to conventional computational fluid dynamics. the geometry of shear driven cavities as well as the reynolds numbers is varied. the simulation is conducted for three types of shear driven cavities which are square cavity and triangular cavities. the obtained streamline patterns and the centre of vortex for each type is in excellent agreement with benchmark results. it is also found that the streamline patterns is significantly affected by the geometrical shape of cavities. keywords: lattice boltzmann method, computational fluid dynamics, shear driven cavites, streamline patterns. 1.0 introduction recently, due to rapidly increasing computational power, computational methods have become the essential tools to conduct researches in various engineering fields. in parallel to the development of high speed digital computer, computational fluid dynamics (cfd) has become the new third approach apart from theory and experiment in the philosophical study and development of the whole discipline of fluid dynamics (anderson, 1995). solving the famous navier-stokes equation would require the knowledge of cfd since the non-linearity and complexity of the equation making it that there is currently no analytical solution to these equations except for a small number of special cases (sidik, 2007). a few examples of numerical methods were introduced by experts in cfd field in order to solve the navier-stokes equation numerically. the methods are like finite difference method, finite element method and finite volume method. the lattice boltzmann method (lbm) has become considerably alternative method to solve fluid flow (munir et al , 2011). the way lbm works is by predicting the evolution of particle distribution function and calculates the macroscopic variables by taking moment to the distribution function. issn: 2180-1053 vol. 3 no. 2 july-december 2011 application of lattice boltzmann method in predicting flow of shear driven cavities 57 60 collision model with a single relaxation time (bhatnagar et al., 1954). the equation that represents this model is given by ;  eqff   (4) where eqf is the equilibrium distribution function and  is the time to reach equilibrium condition during collision process and is often called the relaxation time. eq. (4) also describes that 1/  of non-equilibrium distribution relaxes to equilibrium state within time  on every collision process. substituting eq. (4) into eq. (2) yield f eqff x f c t f         (5) the equation (5) above is known as boltzmann bhatnagargross-krook(bgk) equation. eq. (5) describes two main processes at mesoscale level. the left hand side refers to the propagation of distribution function to the next node in the direction of its probable velocity, and the right hand side represents the collision of the particle distribution functions. in lattice boltzmann formulation, magnitude of c is set up so that in each time step t, every distribution function propagates in a distance of lattice nodes spacing  x. this will ensure that distribution function arrives exactly at the lattice nodes after t and collides simultaneously. in order to apply eq. (5) into the digital computer, the mesoscopic velocity space has to be discretised. this can be done by discretising the physical space into uniform lattice nodes. every node in the network is then connected with its neighbours through a number of lattice velocities to be determined through the model chosen. the general form of the lattice velocity model is expressed as dnqm where d represents spatial dimension and q is the number of connection (lattice velocity) at every node. there are many lattice velocity models published in the literature, however, the most well used due to its simplicity is d2q9. 1.2 the lattice boltzmann equation descretization the boltzmann equation with bgk collision model is as below: f eqff x f c t f         (5) where eq. (5) is well-known as the bgk boltzmann equation as stated in previous sub section. the maxwell-boltzmann equilibrium distribution function is defined as (liboff, 1990)                 rtrt f d eq 2 exp 2 1 22 uc   (6) the bgk lattice boltzmann equation can be derived by further discretise eq. (5) using an euler time step in conjunction with an upwind spatial discretization and then setting the grid spacing divided by the time step equal to the velocity;  issn: 2180-1053 vol. 3 no. 2 july-december 2011 journal of mechanical engineering and technology 58 61         f eqffttfttf t tfttf         x xxx c xx ,,,, (7)         f eqff t ttftttf t tfttf         c xcx c xx ,,,, (8) as a result:              f eqff ttftttf  ,, xcx (9) the equation above has a simple physical interpretation in which the collision term is evaluated locally and there is only one streaming step operation per lattice velocity. this stream and collide particle interpretation is a result of the fully lagrangian character of the equation for which the lattice spacing is the distance traveled by the particles during a time step ( sterling, 1996). although first order discretizations have been used, the lattice boltzmann method is second order in both space and time when contributions that result from discretization error are taken to represent physics (reider et al., 1995). the macroscopic variables such as the density,  and flow velocity, u can be evaluated as the moment to the distribution function as follow  eqff or   cc dffd eq (10) ucc  eqff or ucccc   dffd eq (11) 1.3 prediction of flow for shear driven cavities by using lbm scheme over the years, fluid flow behaviors inside lid driven cavities have drawn many interested researchers and scientists. examples of the applications of lid driven cavities are in material processing, dynamics of lakes, metal casting, galvanizing and etc. two dimensional lbm simulation has been done successfully by houat &youcefi in 2011numerous studies have been carried out on flow patterns inside a cavity. excellent reviews on lid driven square cavity were done by (ghia et al. , 1982), (erturk et al., 2005) and (erturk et al., 2007). erturk et al. has successfully conducted simulation of flows inside triangular cavities. however, all these researchers conducted the fluid flow simulation by solving the navierstokes equations.in addition to that, numerical simulations of fluid flow in square cavity by using lbm have been done by (hou et al., 1995). however, the reynolds number had been used is only up to 7000. apart from the square cavity, simulations of triangular cavity up to 500 by using lbm has been shown successfully by (duan et al., 2007). 62 2.0 method of solution to solve flow in shear driven cavities in this section, the details of methodology in simulating fluid flow inside shear driven cavities are presented. 2.1 simulation of flow for shear driven square cavity the lid driven cavity flow is a flow inside a cavity where the top wall slides to the right at a constant speed of u while the other three walls are made stationary. this type of flow has been used as a benchmark problem for many numerical methods due to its simple geometry but complicated flow behaviors. the geometry of the square cavity for this problem is shown in figure 1. figure 1 geometry of shear driven square cavity lbm is applied to this lid driven cavity flow of height l. the reynolds number (re) was varied from 100 to 10000. table 1 shows the grid size used for the corresponding reynolds numbers. table 1 grid size for each reynolds number for lid driven square cavity flow reynold s number grid size 100 400 x 400 400 400 x 400 1000 400 x 400 3200 400 x 400 5000 400 x 400 7500 1000 400 x 400 400 x 400 for triangular cavity case, three types of the triangular cavity geometry is selected for this problem. figure 2 shows the geometry of the triangles. issn: 2180-1053 vol. 3 no. 2 july-december 2011 application of lattice boltzmann method in predicting flow of shear driven cavities 59 62 2.0 method of solution to solve flow in shear driven cavities in this section, the details of methodology in simulating fluid flow inside shear driven cavities are presented. 2.1 simulation of flow for shear driven square cavity the lid driven cavity flow is a flow inside a cavity where the top wall slides to the right at a constant speed of u while the other three walls are made stationary. this type of flow has been used as a benchmark problem for many numerical methods due to its simple geometry but complicated flow behaviors. the geometry of the square cavity for this problem is shown in figure 1. figure 1 geometry of shear driven square cavity lbm is applied to this lid driven cavity flow of height l. the reynolds number (re) was varied from 100 to 10000. table 1 shows the grid size used for the corresponding reynolds numbers. table 1 grid size for each reynolds number for lid driven square cavity flow reynold s number grid size 100 400 x 400 400 400 x 400 1000 400 x 400 3200 400 x 400 5000 400 x 400 7500 1000 400 x 400 400 x 400 for triangular cavity case, three types of the triangular cavity geometry is selected for this problem. figure 2 shows the geometry of the triangles. 63 figure 2 geometry of triangular cavities used the grid size used for triangular cavity with 90° at top right corner is shown in table 2 below. table 2. grid size for each reynolds number for lid driven triangular cavity flow for triangular cavity’ type a’ reynold s number grid size 100 300 x 300 500 300 x 300 1000 300 x 300 1500 300 x 300 2000 300 x 300 2500 300 x 300 in addition to that, the grid size used for triangular type ‘b’is shown in table 3 below. table 3. grid size for each reynolds number for lid driven triangular cavity flow for triangular cavity’ type b’ reynold s number grid size 100 300 x 300 500 300 x 300 1000 300 x 300 1500 300 x 300 2000 300 x 300 2500 300 x 300 (a)isosceles right triangle with 90 º at top right corner(type a) (b)isosceles right triangle with 90º at top left corner (type b) (c)isosceles right triangle with 90º at corner angle (type c) issn: 2180-1053 vol. 3 no. 2 july-december 2011 journal of mechanical engineering and technology 60 63 figure 2 geometry of triangular cavities used the grid size used for triangular cavity with 90° at top right corner is shown in table 2 below. table 2. grid size for each reynolds number for lid driven triangular cavity flow for triangular cavity’ type a’ reynold s number grid size 100 300 x 300 500 300 x 300 1000 300 x 300 1500 300 x 300 2000 300 x 300 2500 300 x 300 in addition to that, the grid size used for triangular type ‘b’is shown in table 3 below. table 3. grid size for each reynolds number for lid driven triangular cavity flow for triangular cavity’ type b’ reynold s number grid size 100 300 x 300 500 300 x 300 1000 300 x 300 1500 300 x 300 2000 300 x 300 2500 300 x 300 (a)isosceles right triangle with 90 º at top right corner(type a) (b)isosceles right triangle with 90º at top left corner (type b) (c)isosceles right triangle with 90º at corner angle (type c) 64 table 4 depicts the corresponding grid size with respect to reynolds number. table 4. grid size for each reynolds number for lid driven triangular cavity flow for triangular cavity type ‘ c ‘ reynold s number grid size 100 400 x 200 400 400 x 200 700 400 x 200 1000 400 x 200 3000 400 x 200 5000 400 x 200 7000 400 x 200 10000 400 x 200 for each case, velocity, u of 0.1 lu/s is applied on top side of the triangular cavities. the simulation was done by using fortran 90 language. the flowchart of the programming implementation is depicted in figure 3. figure 3 flow chart of the execution of the programming 63 figure 2 geometry of triangular cavities used the grid size used for triangular cavity with 90° at top right corner is shown in table 2 below. table 2. grid size for each reynolds number for lid driven triangular cavity flow for triangular cavity’ type a’ reynold s number grid size 100 300 x 300 500 300 x 300 1000 300 x 300 1500 300 x 300 2000 300 x 300 2500 300 x 300 in addition to that, the grid size used for triangular type ‘b’is shown in table 3 below. table 3. grid size for each reynolds number for lid driven triangular cavity flow for triangular cavity’ type b’ reynold s number grid size 100 300 x 300 500 300 x 300 1000 300 x 300 1500 300 x 300 2000 300 x 300 2500 300 x 300 (a)isosceles right triangle with 90 º at top right corner(type a) (b)isosceles right triangle with 90º at top left corner (type b) (c)isosceles right triangle with 90º at corner angle (type c) table 2 issn: 2180-1053 vol. 3 no. 2 july-december 2011 application of lattice boltzmann method in predicting flow of shear driven cavities 61 64 table 4 depicts the corresponding grid size with respect to reynolds number. table 4. grid size for each reynolds number for lid driven triangular cavity flow for triangular cavity type ‘ c ‘ reynold s number grid size 100 400 x 200 400 400 x 200 700 400 x 200 1000 400 x 200 3000 400 x 200 5000 400 x 200 7000 400 x 200 10000 400 x 200 for each case, velocity, u of 0.1 lu/s is applied on top side of the triangular cavities. the simulation was done by using fortran 90 language. the flowchart of the programming implementation is depicted in figure 3. figure 3 flow chart of the execution of the programming issn: 2180-1053 vol. 3 no. 2 july-december 2011 journal of mechanical engineering and technology 62 65 3.0 simulation results 3.1 shear driven square cavity figure 4 (a)-(h) shown below depicts the corresponding streamline contours for lid driven cavity square cavity. (a) re= 100 (b) re=400 (a) re=1000 (d) re=3200 (e) re=5000 (f) re= 7500 (g) re= 10000 (h) re= 12500 figure 4 streamline patterns for lid driven square cavity by using lbm scheme from the figure 4 shown above, it can be deduced the number of secondary vortex increases when the reynolds number is increased. for instance, when reynolds number applied is 400, the first secondary vortex appears in the streamline patterns. the second secondary vortex appeared when reynolds number is increased to 1000 as shown in figure 4 (b). the maximum number of secondary vortex appeared in the streamline contours for this type of problem is three. 66 next, the location of the primary vortex for every reynolds number was also calculated and is shown in table 5 below. table 5. location of the centre of the primary vortex for lid driven square cavity. reynolds number (re) obtained results reference benchmark (ghia et al.,1982) reference benchmark (hou, et al., 1995) 100 (0.6200,0.7400) (0.6172, 0.7344) (0.6196,0.737 3) 400 (0.5600,0.6000) (0.5547,0.605 5) (0.5608,0.607 8) 1000 (0.5300,0.5650) (0.5313,0.562 5) (0.5333,0.564 7) 3200 (0.5200, 0.5400) (0.5165,0.546 9) na 5000 (0.5150,0.5350) (0.5117,0.535 2) (0.5176,0.537 3) 7500 (0.5150,0.5235) (0.5117,0.532 2) (0.5176,0.533 3) 10000 (0.5133,0.5283) (0.5117,0.533 3) na from the results presented in figure 4 (a) to (h) and also table 5, it is proven that the lbm is able to produce an excellent agreement with the results predicted by conventional numerical methods. they are apparent that the flow structures are in good agreement with the results published in the literature by previous researchers. 3.2 isosceles triangular type ‘a’ figure 5 (a) to (f) show the streamline patterns of flow inside isosceles triangle cavity with 90° at top right corner. (a) re=100 (b) re=500 (c) re=1000 issn: 2180-1053 vol. 3 no. 2 july-december 2011 application of lattice boltzmann method in predicting flow of shear driven cavities 63 66 next, the location of the primary vortex for every reynolds number was also calculated and is shown in table 5 below. table 5. location of the centre of the primary vortex for lid driven square cavity. reynolds number (re) obtained results reference benchmark (ghia et al.,1982) reference benchmark (hou, et al., 1995) 100 (0.6200,0.7400) (0.6172, 0.7344) (0.6196,0.737 3) 400 (0.5600,0.6000) (0.5547,0.605 5) (0.5608,0.607 8) 1000 (0.5300,0.5650) (0.5313,0.562 5) (0.5333,0.564 7) 3200 (0.5200, 0.5400) (0.5165,0.546 9) na 5000 (0.5150,0.5350) (0.5117,0.535 2) (0.5176,0.537 3) 7500 (0.5150,0.5235) (0.5117,0.532 2) (0.5176,0.533 3) 10000 (0.5133,0.5283) (0.5117,0.533 3) na from the results presented in figure 4 (a) to (h) and also table 5, it is proven that the lbm is able to produce an excellent agreement with the results predicted by conventional numerical methods. they are apparent that the flow structures are in good agreement with the results published in the literature by previous researchers. 3.2 isosceles triangular type ‘a’ figure 5 (a) to (f) show the streamline patterns of flow inside isosceles triangle cavity with 90° at top right corner. (a) re=100 (b) re=500 (c) re=1000 67 (d) re=1500 (e) re=2000 (e) re=2500 figure 5 streamline patterns for isosceles triangular type ‘a’ from the figures shown above, there are two significant features revealed by the streamline contours. the first feature is that the number of vortices is increased when the reynolds (re) numbers are increased. as we can see in figure 5 (c), the number of vortex is increased from previous which are two to three when the re number is 1000. furthermore, the second significant feature is that the centre of the primary vortex moves downstream to the right as reynolds number is increased. for instance, figure 5(a) depicts the centre of the primary vortex being located at 4/5 of the bottom vertex. however, this centre moves downward to 3/5 of the bottom vertex as the reynolds number increases. besides that, the primary vortex moves to downstream to the left as the value of the reynolds number is increased. the location of the primary vortex for respective reynolds number is shown in figure 6 below. figure 6 effect of the reynolds number to location of the centre of the primary vortex for isosceles triangular type ‘a’ is shown in figure below. it is noticeable that the centre of the primary vortex moves downward to the left as the reynolds number is increased. apart from the plotted location of the primary vortex, the coordinate of the primary vortex is also compared with the existing benchmarks. the results is presented in table 6 below. 66 next, the location of the primary vortex for every reynolds number was also calculated and is shown in table 5 below. table 5. location of the centre of the primary vortex for lid driven square cavity. reynolds number (re) obtained results reference benchmark (ghia et al.,1982) reference benchmark (hou, et al., 1995) 100 (0.6200,0.7400) (0.6172, 0.7344) (0.6196,0.737 3) 400 (0.5600,0.6000) (0.5547,0.605 5) (0.5608,0.607 8) 1000 (0.5300,0.5650) (0.5313,0.562 5) (0.5333,0.564 7) 3200 (0.5200, 0.5400) (0.5165,0.546 9) na 5000 (0.5150,0.5350) (0.5117,0.535 2) (0.5176,0.537 3) 7500 (0.5150,0.5235) (0.5117,0.532 2) (0.5176,0.533 3) 10000 (0.5133,0.5283) (0.5117,0.533 3) na from the results presented in figure 4 (a) to (h) and also table 5, it is proven that the lbm is able to produce an excellent agreement with the results predicted by conventional numerical methods. they are apparent that the flow structures are in good agreement with the results published in the literature by previous researchers. 3.2 isosceles triangular type ‘a’ figure 5 (a) to (f) show the streamline patterns of flow inside isosceles triangle cavity with 90° at top right corner. (a) re=100 (b) re=500 (c) re=1000 66 next, the location of the primary vortex for every reynolds number was also calculated and is shown in table 5 below. table 5. location of the centre of the primary vortex for lid driven square cavity. reynolds number (re) obtained results reference benchmark (ghia et al.,1982) reference benchmark (hou, et al., 1995) 100 (0.6200,0.7400) (0.6172, 0.7344) (0.6196,0.737 3) 400 (0.5600,0.6000) (0.5547,0.605 5) (0.5608,0.607 8) 1000 (0.5300,0.5650) (0.5313,0.562 5) (0.5333,0.564 7) 3200 (0.5200, 0.5400) (0.5165,0.546 9) na 5000 (0.5150,0.5350) (0.5117,0.535 2) (0.5176,0.537 3) 7500 (0.5150,0.5235) (0.5117,0.532 2) (0.5176,0.533 3) 10000 (0.5133,0.5283) (0.5117,0.533 3) na from the results presented in figure 4 (a) to (h) and also table 5, it is proven that the lbm is able to produce an excellent agreement with the results predicted by conventional numerical methods. they are apparent that the flow structures are in good agreement with the results published in the literature by previous researchers. 3.2 isosceles triangular type ‘a’ figure 5 (a) to (f) show the streamline patterns of flow inside isosceles triangle cavity with 90° at top right corner. (a) re=100 (b) re=500 (c) re=1000 67 (d) re=1500 (e) re=2000 (e) re=2500 figure 5 streamline patterns for isosceles triangular type ‘a’ from the figures shown above, there are two significant features revealed by the streamline contours. the first feature is that the number of vortices is increased when the reynolds (re) numbers are increased. as we can see in figure 5 (c), the number of vortex is increased from previous which are two to three when the re number is 1000. furthermore, the second significant feature is that the centre of the primary vortex moves downstream to the right as reynolds number is increased. for instance, figure 5(a) depicts the centre of the primary vortex being located at 4/5 of the bottom vertex. however, this centre moves downward to 3/5 of the bottom vertex as the reynolds number increases. besides that, the primary vortex moves to downstream to the left as the value of the reynolds number is increased. the location of the primary vortex for respective reynolds number is shown in figure 6 below. figure 6 effect of the reynolds number to location of the centre of the primary vortex for isosceles triangular type ‘a’ is shown in figure below. it is noticeable that the centre of the primary vortex moves downward to the left as the reynolds number is increased. apart from the plotted location of the primary vortex, the coordinate of the primary vortex is also compared with the existing benchmarks. the results is presented in table 6 below. issn: 2180-1053 vol. 3 no. 2 july-december 2011 journal of mechanical engineering and technology 64 67 (d) re=1500 (e) re=2000 (e) re=2500 figure 5 streamline patterns for isosceles triangular type ‘a’ from the figures shown above, there are two significant features revealed by the streamline contours. the first feature is that the number of vortices is increased when the reynolds (re) numbers are increased. as we can see in figure 5 (c), the number of vortex is increased from previous which are two to three when the re number is 1000. furthermore, the second significant feature is that the centre of the primary vortex moves downstream to the right as reynolds number is increased. for instance, figure 5(a) depicts the centre of the primary vortex being located at 4/5 of the bottom vertex. however, this centre moves downward to 3/5 of the bottom vertex as the reynolds number increases. besides that, the primary vortex moves to downstream to the left as the value of the reynolds number is increased. the location of the primary vortex for respective reynolds number is shown in figure 6 below. figure 6 effect of the reynolds number to location of the centre of the primary vortex for isosceles triangular type ‘a’ is shown in figure below. it is noticeable that the centre of the primary vortex moves downward to the left as the reynolds number is increased. apart from the plotted location of the primary vortex, the coordinate of the primary vortex is also compared with the existing benchmarks. the results is presented in table 6 below. issn: 2180-1053 vol. 3 no. 2 july-december 2011 application of lattice boltzmann method in predicting flow of shear driven cavities 65 69 (d) re=1500 (e) re=2000 (f) re= 2500 figure 7 streamline pattern for isosceles triangular type ‘b’ the location of the primary vortex is presented in the next table 7. table 7. location of the centre of the primary vortex for isosceles triangular cavity type ‘b’ reynolds number reference (erturk & gokcol ,2007) obtained results by using lbm scheme 100 (0.4473,0.851 6) (0.4450,0.850 0) 500 (0.5469,0.8496) (0.5550,0.850 0) 1000 (0.6094,0.8691) (0.6050,0.865 0) 1500 (0.6582,0.8848) (0.6567,0.883 3) 2000 (0.6953,0.8965) (0.6900,0.893 3) 2500 (0.7227,0.9043) (0.7167,0.903 3) figure 8 depicts the effect of the reynolds number to location of the centre of the primary vortex. as indicated in the figure, the primary vortex moved upward to the right as the reynolds number increases. this behaviour is further validated in table 6 above which present the coordinate of the primary vortex with respect to the reynolds number. 68 table 6. location of the centre of the primary vortex for isosceles triangular type ‘a’ reynolds number reference (erturk & gokcol ,2007) obtained results by using lbm scheme 100 (0.7090,0.8320) (0.7100,0.830 0) 500 (0.7070,0.7676) (0.7100,0.765 0) 1000 (0.6992,0.7559) (0.7000,0.755 0) 1500 na (0.7000,0.7467) 2000 na (0.7000,0.7467) 2500 (0.6973,0.7441) (0.7000,0.743 3) from table 6 above, the results obtained is in good coherent as compared to the results done by previous researchers. 3.3 isosceles triangular type ‘b’ the results in term of streamline patterns for isosceles triangular type ‘b’ is shown in figure 7 (a)-(f) below. as shown in the figure, it is noticeable that the secondary vortex becomes bigger as reynolds number increases. the second significant feature of the results obtained is the additional number of secondary vortex when reynolds number is higher. (a) re=100 (b) re= 500 (c) re=1000 issn: 2180-1053 vol. 3 no. 2 july-december 2011 journal of mechanical engineering and technology 66 69 (d) re=1500 (e) re=2000 (f) re= 2500 figure 7 streamline pattern for isosceles triangular type ‘b’ the location of the primary vortex is presented in the next table 7. table 7. location of the centre of the primary vortex for isosceles triangular cavity type ‘b’ reynolds number reference (erturk & gokcol ,2007) obtained results by using lbm scheme 100 (0.4473,0.851 6) (0.4450,0.850 0) 500 (0.5469,0.8496) (0.5550,0.850 0) 1000 (0.6094,0.8691) (0.6050,0.865 0) 1500 (0.6582,0.8848) (0.6567,0.883 3) 2000 (0.6953,0.8965) (0.6900,0.893 3) 2500 (0.7227,0.9043) (0.7167,0.903 3) figure 8 depicts the effect of the reynolds number to location of the centre of the primary vortex. as indicated in the figure, the primary vortex moved upward to the right as the reynolds number increases. this behaviour is further validated in table 6 above which present the coordinate of the primary vortex with respect to the reynolds number. 70 figure 8 effect of the reynolds number to the location of the centre of the primary vortex for isosceles triangular type ‘b’ 3.4 isosceles triangular type ‘c’ the results in term of streamline patterns for isosceles triangular type ‘c’ is presented in figure 9 (a) –(f) below. (a) re= 100 (b) re= 400 (c) re=700 (d) re=1000 (e) re=3000 (f) re=5000 (g) re= 7000 (h) re=10000 figure 9 streamline pattern for isosceles triangular type ‘c’ issn: 2180-1053 vol. 3 no. 2 july-december 2011 application of lattice boltzmann method in predicting flow of shear driven cavities 67 71 in figure 9 shown above, the flow contours (streamline patterns) with different reynolds numbers are presented. the flow patterns reveal two significant features. firstly, as re numbers are increased, the primary vortex (eddy) moves downstream to the right. for an instance, at re=100, the location of primary vortex is roughly at 4/5 from the bottom vertex as shown in figure 9 (a). apart from the secondary vortex, no other vortex is visible for re=100. however, for re=400, there is secondary vortex located near the stagnant corner of the triangle as shown in figure 9 (b). the shape of this secondary vortex becomes larger as reynolds numbers is further increased as shown in figure 9 (d) to figure 9 (h). the second significant feature is the number of vortices in the cavity which is increased as the re number is increased. figure 9 (e) shows that the third secondary vortex appears for re=3000, located about 3/5 from the bottom corner of the cavity. the primary vortex moves further upstream to the left before splitting into another secondary vortex when re=5000, as shown in figure 9 (f). for re=7000 and re=10000, the numbers of vortex in the cavity are five and six respectively. table 8 location of the centre of the primary vortex for isosceles triangular type ‘c’ reynolds number results obtained by lbm 100 (0.5450,0.7600) 400 (0.6100,0.7500) 700 (0.5950,0.7150) 1000 (0.5875,0.7150) 3000 (0.7400,0.8200) 5000 (0.7350,0.8100) 7000 (0.7583,0.6200) 10000 (0.4117,0.5375) 70 figure 8 effect of the reynolds number to the location of the centre of the primary vortex for isosceles triangular type ‘b’ 3.4 isosceles triangular type ‘c’ the results in term of streamline patterns for isosceles triangular type ‘c’ is presented in figure 9 (a) –(f) below. (a) re= 100 (b) re= 400 (c) re=700 (d) re=1000 (e) re=3000 (f) re=5000 (g) re= 7000 (h) re=10000 figure 9 streamline pattern for isosceles triangular type ‘c’ issn: 2180-1053 vol. 3 no. 2 july-december 2011 journal of mechanical engineering and technology 68 71 in figure 9 shown above, the flow contours (streamline patterns) with different reynolds numbers are presented. the flow patterns reveal two significant features. firstly, as re numbers are increased, the primary vortex (eddy) moves downstream to the right. for an instance, at re=100, the location of primary vortex is roughly at 4/5 from the bottom vertex as shown in figure 9 (a). apart from the secondary vortex, no other vortex is visible for re=100. however, for re=400, there is secondary vortex located near the stagnant corner of the triangle as shown in figure 9 (b). the shape of this secondary vortex becomes larger as reynolds numbers is further increased as shown in figure 9 (d) to figure 9 (h). the second significant feature is the number of vortices in the cavity which is increased as the re number is increased. figure 9 (e) shows that the third secondary vortex appears for re=3000, located about 3/5 from the bottom corner of the cavity. the primary vortex moves further upstream to the left before splitting into another secondary vortex when re=5000, as shown in figure 9 (f). for re=7000 and re=10000, the numbers of vortex in the cavity are five and six respectively. table 8 location of the centre of the primary vortex for isosceles triangular type ‘c’ reynolds number results obtained by lbm 100 (0.5450,0.7600) 400 (0.6100,0.7500) 700 (0.5950,0.7150) 1000 (0.5875,0.7150) 3000 (0.7400,0.8200) 5000 (0.7350,0.8100) 7000 (0.7583,0.6200) 10000 (0.4117,0.5375) 72 figure 10 effect of the reynolds number to location of the centre of the primary vortex for isosceles triangular type ‘c’ from the figure 10 above, initially the primary vortex moves to upstream to the right as reynolds number increases. however, at re=3000 onwards, the primary vortex moves downstream to the left. this profile is significantly different that profile for the other two type of triangular cavities shown in previous sections. 4.0 conclusion among the microscopic models existing in the literature, lbm, the model developed from continous boltzmann equation, has evolved into a powerful tool for modelling complex flow since it was first appeared in 1980s. although the approach is based on the microscopic interactions, all macroscopic continuum equations such as the navier-stokes equation can be derived and recovered. fluid flow behaviours in shear driven cavities have been demonstrated by using lattice boltzmann scheme successfully. it was found that, the present approach correctly predicted the flow feature for different reynolds numbers and yield excellent agreement with the results from previous works. the streamline contours or patterns are in good agreement with ghia et al., and erturk et. al. apart from that, it is found that the streamline patterns are heavily affected by the reynolds number and also the geometry of the cavity. however there are few demerits of lbm. when reynolds number is large, the relaxation parameter in the lbm approaches to the stability margin if the number of mesh points is not very large. there are few solutions have been proposed (he et al., 1996). however, a novel solution to this problem is still required. there is also not sufficient evidence to show that the lbm can be applied to aerodynamic turbulent flows. at present time, one of the weaknesses of lbm for computational fluid dynamics (cfd) is the lack of turbulence issn: 2180-1053 vol. 3 no. 2 july-december 2011 application of lattice boltzmann method in predicting flow of shear driven cavities 69 72 figure 10 effect of the reynolds number to location of the centre of the primary vortex for isosceles triangular type ‘c’ from the figure 10 above, initially the primary vortex moves to upstream to the right as reynolds number increases. however, at re=3000 onwards, the primary vortex moves downstream to the left. this profile is significantly different that profile for the other two type of triangular cavities shown in previous sections. 4.0 conclusion among the microscopic models existing in the literature, lbm, the model developed from continous boltzmann equation, has evolved into a powerful tool for modelling complex flow since it was first appeared in 1980s. although the approach is based on the microscopic interactions, all macroscopic continuum equations such as the navier-stokes equation can be derived and recovered. fluid flow behaviours in shear driven cavities have been demonstrated by using lattice boltzmann scheme successfully. it was found that, the present approach correctly predicted the flow feature for different reynolds numbers and yield excellent agreement with the results from previous works. the streamline contours or patterns are in good agreement with ghia et al., and erturk et. al. apart from that, it is found that the streamline patterns are heavily affected by the reynolds number and also the geometry of the cavity. however there are few demerits of lbm. when reynolds number is large, the relaxation parameter in the lbm approaches to the stability margin if the number of mesh points is not very large. there are few solutions have been proposed (he et al., 1996). however, a novel solution to this problem is still required. there is also not sufficient evidence to show that the lbm can be applied to aerodynamic turbulent flows. at present time, one of the weaknesses of lbm for computational fluid dynamics (cfd) is the lack of turbulence 73 modelling. the application of lbm to turbulent flows at high reynolds number remains as an area of future development. 5.0 acknowledgment the authors would like to acknowledge universiti teknikal malaysia melaka and universiti teknologi malaysia for supporting this work through research grant. 6.0 references albensoeder, s. & kuhlmann, h.c.(2005).accurate three-dimensional lid-driven cavity flow. journal of computational physics, vol. 206, 536-558. issn 0021-9991 anderson, d.j.( 1995), computational fluid dynamics, mcgraw-hill, singapore. bhatnagar, p. l., gross, e. p. and krook,m.(1954). a model for collision processes in gasses. 1. small amplitude processes in charged and neutral one component. system physics review, vol. 94, 511–525. cheng m. & hung k.c. (2006).vortex structure of steady flow in a rectangular cavity. computers & fluids, vol. 35, n. 1, pp. 1046-1062. issn 0045-7930. duan, y.l. & liu,r.x. (2007). lattice boltzmann simulations of triangular cavity flow and free-surface problems. journal of hydrodynamics, vol.19, n.2, pp. 127-134.issn 1001 6058. erturk e. & dusun, b. (2007). numerical solutions of 2-d steady incompressible in a driven skewed cavity. journal of applied mathematics and mechanics. vol. 87, pp 377-392. issn 0021-8928. erturk, e. & gokcol o.(2007). fine grid numerical solutions of triangular cavity flow. the european physical journal applied physics, vol.38, pp. 97-105, issn 1286-0042. erturk e. ; corke t.c. & gokcol, o. (2005).numerical solutions of 2-d steady incompressible driven cavity flow at high reynolds number, international journal for numerical methods in fluids, vol.48, pp. 747-774. issn 1097-0363. francois, c. (1989). derivation of slip boundary conditions for the navier-stokes system from the boltzmann equation. journal of statistical physics. vol. 54. issn 1572-9613. frish, u. ; hasslacher, b. & pomeau, y.(1986). lattice gas automata for the navierstokes equation. physics review letter. vol. 56, pp. 1505-1508. issn 1079-7114. ghia, u. ; ghia, k.n. & c.y. chin, (1982). high re solutions for incompressible flow using the navier-stokes equations and a multigrid method, journal of computational physics, vol. 48, pp. 387-411. issn 0021-9991 gladrow ,d.g. (2000). lattice gas cellular automata and lattice boltzmann models, an introduction. springer. new york. he, x. & doolen, g. (1997). lattice boltzmann method on curvelinear coordinates system: flow around a circular cylinder. journal of computational physics, vol 134,pp. 306-315. issn 0021-9991 he, x. and luo, l.s.(1997). lattice boltzmann model for the incompressible navier stokes equations. journal of statistical physics. vol. 88, pp. 927-944. issn 1572-9613. hou, s., q., zou, s., chen and, doolen, g.(1995). simulation of cavity flow by the lattice boltzmann method. journal of computational physics. vol. 118, 329-347. issn 00219991. issn: 2180-1053 vol. 3 no. 2 july-december 2011 journal of mechanical engineering and technology 70 73 modelling. the application of lbm to turbulent flows at high reynolds number remains as an area of future development. 5.0 acknowledgment the authors would like to acknowledge universiti teknikal malaysia melaka and universiti teknologi malaysia for supporting this work through research grant. 6.0 references albensoeder, s. & kuhlmann, h.c.(2005).accurate three-dimensional lid-driven cavity flow. journal of computational physics, vol. 206, 536-558. issn 0021-9991 anderson, d.j.( 1995), computational fluid dynamics, mcgraw-hill, singapore. bhatnagar, p. l., gross, e. p. and krook,m.(1954). a model for collision processes in gasses. 1. small amplitude processes in charged and neutral one component. system physics review, vol. 94, 511–525. cheng m. & hung k.c. (2006).vortex structure of steady flow in a rectangular cavity. computers & fluids, vol. 35, n. 1, pp. 1046-1062. issn 0045-7930. duan, y.l. & liu,r.x. (2007). lattice boltzmann simulations of triangular cavity flow and free-surface problems. journal of hydrodynamics, vol.19, n.2, pp. 127-134.issn 1001 6058. erturk e. & dusun, b. (2007). numerical solutions of 2-d steady incompressible in a driven skewed cavity. journal of applied mathematics and mechanics. vol. 87, pp 377-392. issn 0021-8928. erturk, e. & gokcol o.(2007). fine grid numerical solutions of triangular cavity flow. the european physical journal applied physics, vol.38, pp. 97-105, issn 1286-0042. erturk e. ; corke t.c. & gokcol, o. (2005).numerical solutions of 2-d steady incompressible driven cavity flow at high reynolds number, international journal for numerical methods in fluids, vol.48, pp. 747-774. issn 1097-0363. francois, c. (1989). derivation of slip boundary conditions for the navier-stokes system from the boltzmann equation. journal of statistical physics. vol. 54. issn 1572-9613. frish, u. ; hasslacher, b. & pomeau, y.(1986). lattice gas automata for the navierstokes equation. physics review letter. vol. 56, pp. 1505-1508. issn 1079-7114. ghia, u. ; ghia, k.n. & c.y. chin, (1982). high re solutions for incompressible flow using the navier-stokes equations and a multigrid method, journal of computational physics, vol. 48, pp. 387-411. issn 0021-9991 gladrow ,d.g. (2000). lattice gas cellular automata and lattice boltzmann models, an introduction. springer. new york. he, x. & doolen, g. (1997). lattice boltzmann method on curvelinear coordinates system: flow around a circular cylinder. journal of computational physics, vol 134,pp. 306-315. issn 0021-9991 he, x. and luo, l.s.(1997). lattice boltzmann model for the incompressible navier stokes equations. journal of statistical physics. vol. 88, pp. 927-944. issn 1572-9613. hou, s., q., zou, s., chen and, doolen, g.(1995). simulation of cavity flow by the lattice boltzmann method. journal of computational physics. vol. 118, 329-347. issn 00219991. 74 houat, s. & youcefi, a. (2011). two dimensional simulation of incompressible fluid flow using lattice boltzmann method, international review of mechanical engineering, vol .1,n. 3, pp. 286 – 292. issn 1970-8734. kampen, v.n. g. (1987). chapman-enskog as an application of the method for eliminating fast variables. journal of statistical physics. vol. 46, nos. ¾. issn 1572-9613. liboff, r.l.(1990). kinetic theory. prentice-hall, englewood cliff new jersey. malapinas, o. ; fieter, n. & m. deville, lattice boltzmann method for the simulation of viscoelastic fluid flows, journal of non-newtonian fluid mechanics, vol. 165, pp. 16371653, 2010. issn 0377-0257. mohd irwan, m.a. ; fudhail, a. m. ; nor azwadi, c. s. & masoud, g. (2010). numerical investigation of incompressible fluid flow through porous media in a lid driven square cavity, american journal of applied sciences, vol. 7, n. 10, pp. 1341-1344.issn 15543641. munir, f.a. ; sidik,n.a.c . & ibrahim,n.i.n.(2011). numerical simulation of natural convection in an inclined square cavity, journal of applied sciences, vol. 11, n2, pp.373-378. issn 1812-5654. reider, m.b. & sterling, j.d. (1995). accuracy of discrete velocity bgk models for the simulation of the incompressible navier-stokes equation. computer and fluids,vol 24, pp. 459-467. issn 0045-7930. sidik, n.a.c, the development of new thermal lattice boltzmann models for the simulation of thermal fluid flow problems. ph.d. dissertation, faculty of science & technology, keo univ., japan, 2007. sterling, j.d. (1996). stability analysis of lattice boltzmann methods. journal of computational physics, vol 123,n.1,pp. 196-206. issn 0021-9991. succi, s.(2001). the lattice boltzmann equation for fluid dynamics and beyond. oxford science publications, new york. sukop, m.c. and thorne, d.t. (2006). lattice boltzmann modeling : an introduction for geoscientists and engineers. springer, new york. 06(67-70).pdf microsoft word 5393-14495-1-rv-3.doc 1 hygrothermal effect on mwcnt-filled epoxy electrically conductive adhesives s. h. s. m. fadzullah1*, m. m. nasaruddin2, g. omar3, z. mustafa4, m. b. ramli5, m. z. akop6 and i. s. mohamad7 1,2,3,5,6,7 faculty of mechanical engineering, universiti teknikal malaysia melaka, hang tuah jaya, 76100 durian tunggal, melaka, malaysia 1,2,3,5,6,7 centre for advanced research on energy, universiti teknikal malaysia melaka, hang tuah jaya, 76100 durian tunggal, melaka, malaysia 4 advanced manufacturing centre, faculty of mechanical engineering, universiti teknikal malaysia melaka, hang tuah jaya, 76100 durian tunggal, melaka, malaysia abstract to-date, limited studies are found in the literature on the reliability performance of electrically conductive adhesive (eca) using multiwalled carbon nanotube (mwcnt) fillers. hence, this study aims to provide an understanding on the performance of the eca with the objectives (i) to study the electrical conductivity and (ii) joint strength of eca with varying conductive filler’s aspect ratio and environmental conditions. here, epoxy with mwcnt aspect ratio of 55.5 and 1666.5 were subjected to 85°c and 85% rh for up to 96 hours. the test specimens were prepared in accordance with astm f390-11 using a four-point probe for electrical conductivity measurement while the lap shear test was conducted with reference to astm d1002-10 using a universal testing machine. for the thermal aging study, the eca samples were conditioned in a humidity chamber at 85 °c and 85 % of relative humidity to assess the reliability performance of the eca. overall, it was found that eca filled with higher aspect ratio of mwcnt exhibit better electrical and mechanical stability when subjected to hygrothermal aging. moreover, the presence of moisture attack has yield in an increase in the electrical conductivity of the eca with thermal aging period. meanwhile, lap shear test results revealed a contradicting trend. regardless of the amount of mwcnt filler loading, voids are created in the epoxy matrix of the eca, which results in a decrease in the shear strength of the eca, when the samples were subjected to thermal aging. keywords: eca; mwcnt; aspect ratio; hygrothermal aging; mechanical strength 1.0 introduction electrically conductive adhesives (ecas) are the alternative for conventional interconnect materials (pb/sn and sn/ag/cu) in electronic packaging due to it toxic-free material and also low in processing temperature (lee et al., 2005; tan et al., 2006). eca is predominantly made up of two main materials; a matrix and a filler. the polymer matrix material function is to provide the mechanical properties of the eca. meanwhile, the role of a conductive filler is to enable the adhesive to inherit its ability in conducting electricity. in addition, properties of the filler such as types, shapes and sizes will remarkably affect the performance of the eca (lu et al., 2002). besides, the _______________________________________ *corresponding author. email: hajar@utem.edu.my issn 2180-1053 vol. 12 no. 2 30 jun 31 dec 2020 journal of mechanical engineering and technology 2 eca’s performance is also affected by environmental conditions (cui et al., 2013). therefore, it is essentially important to study the effect of environmental condition on the eca since interconnect materials are often exposed to various surrounding setting during its actual applications. hence, the focus of this study is to investigate the effect of hygrothermal aging on the electrical performance of eca filled with different aspect ratio (size) of mwcnt (in terms of the mwcnt length). 2.0 experimental procedure eca was prepared by mixing the epoxy with hardener by 30% to the weight of the epoxy and manually blended until the solutions are homogenized within one minute of manual mixing time. the conductive filler, that is the mwcnt is then inserted into the suspension and further mixed for an additional 5 minutes. the filler loading used for this present study are 7.0 wt.%, 8.5 wt.% and 10.0 wt.% for both aspect ratios of mwcnt; low aspect ratio (l-mwcnt) and high aspect ratio (h-mwcnt). description of the aspect ratios are displayed in table 1. table 1. details dimension of mwcnt d mwcnt outer diameter, od (nm) length, l (µm) aspect ratio (l/od) min. max. min. max. min. max. avg. l-mwcnt 110 170 5 9 29 82 55.5 h-mwcnt 10 30 10 30 333 3000 1666.5 d by referring to astm f390 as a guideline for the electrical resistivity measurement, the mixed adhesive was printed on 3 mm-thick acrylic substrates, as shown in figure 1. the adhesive was cured at temperature of 100 °c for 30 minutes in a readily heated curing oven. the cured sample was cooled for 24 hours and further measured in terms of its electrical resistivity by using a jandel four-point probe. the samples were then subjected to hygrothermal aging for up to 96 hours at 85 °c and 85 % rh in a memmert humidity chamber model hcp 108 figure 1. schematic diagram of printed eca arrangement furthermore, astm d1002 is reffered as the guidlines to conduct the single-lap bonded joint test for the adhesive in order to clarify the joint strength of the eca. the sample for this kind of test is assembled as presented in figure 2 and conducted by using universal testing machine model hengzhun hz-1003. according to the standard, the test speed is fixed at 1.3 mm/min and with five times repitition for each of test parameter. issn 2180-1053 vol. 12 no. 2 30 jun 31 dec 2020 3 figure 2. sample assembly for single-lap bonded joint test 3.0 results and discussion figure 3 illustrates the experimental results following electrical resistivity measurements, with respect to different filler loading, at room temperature condition, for each aspect ratio used in this study. it is apparent that the electrical resistivity for both types of eca exhibit the same trend, that is a decreased in electrical resistivity with an increasing filler loading. such observation suggests that by increasing the amount of mwcnt filler loading will lead in an improved electrical conductivity of the adhesive. furthermore, with higher amount of the mwcnt, the tendency of the mwcnt to be in contact with each other will also increase. this can improve the conductive path away of the electron in between the adhesive as well as providing better electrical conductivity for the eca. nonetheless, visual observation during the experiment showed that at much higher filler loading, the rheological characteristic of the adhesive is disturbed since the adhesive would become too viscous (potschke et al., 2002) and lead to some difficulties during the printing process. in addition, figure 3 also depicts that eca with higher mwcnt aspect ratio yield in better electrical conductivity by having lower electrical resistivity of the eca, which is 0.54 ω.cm compared to 0.88 ω.cm at 8.5 wt.% filler loading. this is because mwcnt with higher aspect ratio (longer in length) could more easily form contact with other mwcnt, due to its flexibility plus high aspect ratio of the filler itself (geng et al., 2008). this occurrence will build the conductive path away and allow high percentage of electron to flow through the adhesive. the arrangement of the mwcnt in the eca are captured by using sem and the micrographs are as displayed in figure 4. figure 3. volume resistivity against filler loading for l-mwcnt/epoxy and h-mwcnt/epoxy 4 (a) (b) (c) (d) figure 4. sem micrographs showing the cross-sectional view of eca with 10.05 wt.% filler loading for (a) & (b) a-mwcnt/epoxy andd (c) & (d) b-mwcnt/epoxy, at 5000x and 10000x magnification respectively. followed by hygrothermal aging, the results in figure 5 show that the volume resistivity shift of the eca decrease with higher mwcnt filler loading. this signifies that the destruction of the eca’s electrical conductivity can be delimited by adding up the volume of mwcnt inside the eca. this occurrence could possibly be due to the barrier effect provided by the mwcnt itself (sima, 2015). as its amount is increase, the gap between each of the mwcnt become smaller, hence hinder the movement of water molecule inside the conductive adhesive. another important finding to note from this experiment is that, in order to make the barrier effect become prominently effective, high aspect ratio of mwcnts can be used as the conductive filler for the eca. figure 5 shows that the volume resistivity shift for low aspect ratio of mwcnt/epoxy is 2.23 ω.cm while the resistivity shift value recorded for the eca with high aspect ratio is only 0.04 ω.cm, at 10 wt.%. due to the flexibility and high aspect ratio of the mwcnt, it has a tendency to agglomerate with each other (geng et al., 2008) which provide the closer gap consequently enhance the barrier effect towards the water movement. issn 2180-1053 vol. 12 no. 2 30 jun 31 dec 2020 5 figure 5. volume resistivity shift after aged in 85 °c and 85 %rh for 96 hours figure 6 suggest that in normal condition, lower aspect ratio of mwcnt (amwcnt/epoxy) filled eca exhibit better shear strength compared to eca filled with higher aspect of mwcnt (b-mwcnt/epoxy). the results are also supported by surface failure of the eca as displayed in figure 7. in eca failure analysis, there are two types of failure mode that can be obtain which are adhesive and cohesive failure modes, (subramaniam et al., 2016) and adhesive failure is the type of failure that need to be avoided in conductive adhesive field since it indicates the weak performance of the eca. as shown in figure 7(b), the failure type for b-mwcnt/epoxy are adhesivecohesive failure and significantly shifted to adhesive failure at 10.0 wt.%. in comparison with a-mwcnt/epoxy, the failure types shown are constant at cohesiveadhesive failure. this is because, higher aspect ratio of mwcnt (b-mwcnt) have larger surface area which will causing more contact to the epoxy resin in the adhesive system. therefore, there will be less amount of epoxy resin left to make contact or adhere with the surface of the substrate hence resulting weak adhesion between them. this problem will become more severe as the filler loading is increase which means more mwcnt surface are exposed to be clinged with the epoxy resin. this explain the significantly shifted failure mode from adhesive-cohesive failure to adhesive failure of b-mwcnt/epoxy. figure 6. shear strength of eca at normal and elevated temperature and humidity issn 2180-1053 vol. 12 no. 2 30 jun 31 dec 2020 6 (a) (b) figure 7. surface failure of the eca subjected to normal temperature and relative humidity, showing (a) adhesive-cohesive failure and in (b) adhesive failure however, following hygrothermal aging at temperature and humidity of 85 °c/85 % rh for 96 hours, the eca filled with both type of mwcnt aspect ratio showed a similar mode of failure, that is the adhesive failure, as shown in figure 8 (a) and (b). this indicates that the moisture attack is prominently affecting the adhesive strength. what can be highlighted from this outcome is that, although bmwcnt/epoxy produce relatively poorer shear strength following hygrothermal aging, the strength shift is much lower than the case for a/mwcnt/epoxy as displayed in figure 9. this outcome suggests that higher aspect ratio of mwcnt is effective as barrier effect to hinder the movement of water molecule from diffused inside the eca system. due to high flexibility characteristic from high aspect ratio of mwcnt, it leads to higher tortuosity in the diffusion path of the eca system. this event will intensely resist the penetrating of moisture from getting inside the eca. figure 8. surface failure for ecas at 85 °c temperature and 85 % relative humidity issn 2180-1053 vol. 12 no. 2 30 jun 31 dec 2020 7 figure 9. shear strength shift for a-mwcnt/epoxy and b-mwcnt/epoxy ecas apart from the aspect ratio of mwcnt, another factor that is helping the sustainability performance of an eca under hygrothermal aging is the amount of filler loading used in the system. based on figure 9 above, it can be concluded that as the filler loading is increase the shear strength shift will also decrease. this occurrence flashes that mwcnt can be used to improve the stability performance of the eca under high humidity and temperature surrounding conditions. this is because, by having higher crowd of mwcnt, it will fill up the empty spaces inside the eca and between the filler itself. there will be less distance between the mwcnt which consequently become a wholesome blockade towards water molecule. as a result, the water will be having difficulties in diffusing into the eca systems. however, too much filler loading in an eca system will degrade the adhesion performance of eca as exhibited in figure 6 where the shear strength is the lowest at the highest filler loading for both cases of aspect ratio. therefore, right amount of filler loading need to be used in order to produce good balance between mechanical bonding and durability under high temperature and humidity condition. 4.0 conclusion the experimental result shows that eca with high mwcnt filler loading and aspect ratio exhibit greater stability in resisting the hygrothermal aging with relatively lower electrical resistivity degradation shift in comparison to those of lower aspect ratio. besides that, the same result also displayed in term of mechanical aspect where the higher aspect ratio of mwcnt gives lower lap shear strength shift compared to eca filled with lower aspect ratio of mwcnt. beside aspect ratio of the filler, amount of the filler also plays an important role in order to improve the durability performance of eca under extreme condition. however, too high amount of mwcnt will not also going to produce a good overall performance of an eca. therefore, right amount of filler loading need to be used in order to produce good balance between mechanical bonding and durability under high temperature and humidity condition. issn 2180-1053 vol. 12 no. 2 30 jun 31 dec 2020 8 5.0 acknowledgement the authors are grateful to the faculty of mechanical engineering, universiti teknikal malaysia melaka (utem), ministry of higher education malaysia (mohe) and jabil circuit sdn bhd. for the technical and financial supports under pjp/2016/fkm/hi1/s01464 and gluar/jabil/2016/fkm-care/i00016 grants. special thanks to zamalah scheme under utem for funding the msc research degree for one of the authors. 6.0 references cui, h., li, d., fan, q. & lai, h. (2013). electrical and mechanical properties of electrically conductive adhesives from epoxy, micro-silver flakes, and nanohexagonal boron nitride particles after humid and thermal aging. international journal of adhesive and adhesion, 44, 232–236. geng, y., liu, m. y., li, j., shi, x. m. & kim, j. k. (2008). effects of surfactant treatment on mechanical and electrical properties of cnt/epoxy nanocomposites, composite part a: applied science and manufacturing, 39(12), 1876–1883. lee, h. h., sen chou, k., & shih, z.w. (2011). effect of nano-sized silver particles on the resistivity of polymeric conductive adhesives. international journal of adhesive and adhesion, 25(5), 437–441. lu, d., luo, s. & wong, c. p. (2002). conductive polymer composites. encyclopedia of polymer science and technology. georgia institute of technology, 652–697. potschke, p., fornes, t. d. & paul, d. r. (2002). rheological behavior of multiwalled carbon nanotube/polycarbonate composites. polymer, 43(11), 3247–3255. sima nabavizadeh, r. (2015). the effect of moisture on electrical resistivity of mwcnt reinforced epoxy nanocomposites. (unpublished master’s thesis). concordia university, montreal, quebec, canada. subramaniam, a. s., tey, j. n., zhang, l., ng, b. h., roy, s., wei, j. & hu, x. m., (2016). synergistic bond strengthening in epoxy adhesives using polydopamine/mwcnt hybrids. polymer, 82, 285-294. tan, f., qiao, x., chen, j. & wang, h. (2006). effects of coupling agents on the properties of epoxy-based electrically conductive adhesives. international journal of adhesive and adhesion, 26(6), 406–413. issn 2180-1053 vol. 12 no. 2 30 jun 31 dec 2020 issn: 2180-1053 vol. 10 no.2 june – december 2018 15 global uncertainty and sensitivity analysis of a reduced chemical kinetic mechanism of a gasoline, n-butanol blend in a high pressure rapid compression machine edirin agbro * school of chemical and process engineering, university of leeds, leeds, ls2 9jt, united kingdom abstract a detailed evaluation of a recently developed combined n-butanol/toluene reference fuel (trf) reduced chemical kinetic mechanism describing the low temperature oxidation of n-butanol, gasoline and a gasoline/n-butanol blend was performed using both global uncertainty and sensitivity methods with ignition delays as the predicted output for the temperature range 678 858 k, and an equivalence ratio of 1 at 20 bar. the results obtained when incorporating the effects of uncertainties in forward rate constants in the simulations, showed that uncertainties in predicting key target quantities for the various fuels studied are currently large but driven by few reactions. global sensitivity analysis of the mechanism based on predicted ignition delays of stoichiometric trf mixtures, showed the toluene + oh route = phenol + ch3 to be among the most dominant pathways in terms of the predicted output uncertainties but an update on the mechanism based on data from a recent study led to the toluene + oh hydrogen abstraction reaction becoming the most dominant reaction as expected. for the trf/n-butanol blend, hydrogen abstraction reactions by oh from n-butanol appear to be key in predicting the effect of blending. uncertainties in the temperature dependence of relative abstraction rates from the α and γ sites may still be present within current mechanisms, and in particular may affect the ability of the mechanisms to capture the low temperature delay times for n-butanol. further studies of the product channels for n-butanol + oh for temperatures of relevance to combustion applications could help to improve current mechanisms. at higher temperatures, the reactions of ho2 and that of formaldehyde with oh also became critical and attempts to reduce uncertainties in the temperature dependent rates of these reactions would be useful. keywords: n-butanol, ignition delay, blending, global sensitivity, uncertainty quantification 1.0 introduction in order to continue to use liquid fuels at lower emission levels, modern combustion devices need to become significantly more efficient. bio-derived alcohols such as methanol, ethanol and butanol are currently being projected as suitable blends for fossil * corresponding author e-mail: edirigbo@yahoo.com journal of mechanical engineering and technology 16 issn: 2180-1053 vol. 10 no.2 june – december 2018 derived fuels in order to reduce their overall carbon footprint (agarwal, 2007). the similarity of their physical and chemical properties to those of fossil-derived fuels make them compatible with modern engines, particularly when used as blends (sarathy et al., 2014, szwaja and naber, 2010). ethanol has been used extensively and can be used at low blending ratios with gasoline without requiring engine modifications. however, there is presently some support for biobutanol (n-butanol or 1-butanol) as a potential replacement for ethanol in spark ignition (si) and compression ignition (ci) engines due to its numerous similarities with gasoline (table 1) and advantages over ethanol. due to its higher energy density, butanol offers better fuel economy when blended with gasoline compared to ethanol. with many properties (i.e. lower heating value and stoichiometric air-fuel ratio) that are more similar to gasoline than ethanol, butanol can be blended with gasoline at higher concentrations without the need for engine retrofitting or modification (wigg, 2011). in one of the studies reported in the literature (dernotte et al., 2009) up to 80 % of butanol by volume was blended with gasoline. other advantages of butanol over ethanol include its tolerance for water contamination in gasoline and less tendency to corrosion allowing it to be transported with existing distribution fuel pipelines. while renewable bio-derived liquid fuels and their blends with conventional fuels (i.e. n-butanol blended with gasoline) are a promising option for achieving a lower carbon footprint, a wider penetration and sensible use of these fuels in internal combustion engines requires first and foremost, an in-depth understanding of the performance of the fuel blends under a wide range of operating conditions. achieving this using an experimental approach for a range of fuel blends is currently quite challenging due to the cost involved, hence the need for a computer approach. computer simulation and analysis provides the ability to relatively solve the complex problems related to these new and completely different fuels cheaply and quickly without having to go into the rigors of very expensive and time consuming experimental testing (baulch, 1997). where experimental measurements are difficult or impossible, the wide range of data provided through computer modelling can also be effectively utilised for the design, testing and control of new and conventional combustion technologies required to use alternative fuels optimally. however, a successful application of computational strategies depends on the availability of reliable and detailed well validated chemical kinetic mechanisms of the various fuels/fuel blends as input in computer simulations for characterization of the engine combustion processes. gasoline’s complexity makes it practically impossible to model its chemistry exactly, so an appropriate 3-component toluene reference fuel (trf) surrogate comprised of toluene, n-heptane and iso-octane, formulated in (agbro et al., 2017), is used to represent gasoline in this work. the detailed blended chemical kinetic model of nbutanol and trf, developed in (agbro et al., 2017) was evaluated using linear sensitivity method employing the brute force approach. here, global uncertainty and sensitivity methods described fully in (tomlin, 2013, tomlin, 2006) is employed to provide further insight into the underlying chemistry mainly influencing the observed ignition delay behavior of the gasoline/butanol blends. while the linear sensitivity approach serves to highlight the important reactions driving the influence of n-butanol on ignition delay times when blended with gasoline at low temperatures, the global uncertainty and sensitivity analysis is carried out here to explore non-linear effects across the entire range of the input parameter space and the impact of the inherent uncertainties in the combined gasoline and n-butanol scheme on the predicted ignition global uncertainty and sensitivity analysis of a reduced chemical kinetic mechanism of a gasoline, n-butanol blend in a high pressure rapid compression machine issn: 2180-1053 vol. 10 no.2 june – december 2018 17 delay times of n-butanol, trf and trf-n-butanol blend. this is aimed at providing useful information for kinetic studies that will improve model robustness. sensitivity indices calculated within the global analysis, based on the application of a hdmr metamodel(ziehn and tomlin, 2009, tomlin and turanyi, 2013) further helps to appropriately identify the key reaction rates that mostly influence (or determine) the predicted target uncertainties and this is quite useful where a nonlinear relationship exist between the sampled rates and predicted ignition delays within particular region of the input space. the global approach also allows us to understand how the interaction between various parameters in the kinetic model affect the predicted target output. such information is critical to gain better insight into the complex chemistry behind the autoignition process for improved quantification of the chemical kinetic model. table 1: properties of gasoline, n-butanol, ethanol and methanol (wigg, 2011) fuel gasoline regular (pon 87) n-butanol ethanol methanol chemical formula ch1.87 c4h9oh c2h5oh ch3oh specific gravity 0.7430 0.8097 0.7894 0.7913 lower heating value (mj/kg) 42.9 32.01 26.83 20.08 stoichiometric air-fuel ratio (kgair/kgfuel) 14.51 11.12 8.94 6.43 energy density (mj/l) 31.9 25.9 21.2 15.9 latent heat of vaporisation (at boiling point) (kj/kg) 349 584 838 1098 octane number (ron+mon)/2 87 86 100 99 2.0 methodology 2.1 chemical kinetic scheme while a few detailed and reduced mechanisms of gasoline oxidation involving primary reference fuels (prfs), toluene reference fuels (trfs) and more complex surrogates currently exist in the literature (mehl et al., 2011, tanaka et al., 2003, glaude et al., 2002, westbrook et al., 1988, andrae et al., 2007, andrae, 2008, naik et al., 2005), the only combined oxidation mechanism for gasoline (toluene, n-heptane, iso-octane mixture)/n-butanol blends available at the time of this study was the detailed scheme presented in (agbro et al., 2017). for the purpose of this study, a reduced version of the trf/n-butanol blended mechanism, developed from the detailed scheme for use in the context of simulating autoignition and knock in the engine was adopted. the detailed scheme contains 1944 species and 8231 elementary reactions while the reduced scheme employed here is comprised of 527 species and 2644 reaction steps. more information on the detailed trf/n-butanol blended mechanism can be found in (agbro et al., 2017) while information on the reduced scheme can be found in (agbro, 2017).the reduced trf/n-butanol kinetic scheme, originally in chemkin format, was first converted to cantera input format (.cti file including the thermodynamic data) using the cantera 2.1.2 ck2ti.py subroutine before it was used in the simulations. journal of mechanical engineering and technology 18 issn: 2180-1053 vol. 10 no.2 june – december 2018 2.2 simulations and uncertainty/sensitivity analysis ignition delay times measured in the leeds rcm and presented in (agbro et al., 2017), were simulated in this work using the open source cantera software toolbox (version 2.1.2) (goodwin, 2013) by running homogeneous variable volume history simulations accounting for heat loss in the experiments. the volume profiles of the rcm employed in the simulations were determined from the measured pressure trace of the non-reactive experiment using isentropic core relations and a temperature-dependent mixture specific heat ratio (weber and sung, 2013). a screening process utilizing local sensitivity method and based on the brute-force method, was first applied to the n-butanol/trf kinetic scheme in order to reduce the number of input parameters involved in the global uncertainty/sensitivity analysis since only a few key reactions are likely to greatly influence the accuracy of the predicted targets. the screening technique and the results of this local approach are presented in (agbro et al., 2017). in the work of agbro (agbro et al., 2017), bruteforce sensitivity analyses were conducted at 20 bar and various temperature conditions using the closed homogeneous batch reactor module in chemkin pro (reaction design, 2011) and constant volume simulations. a total of 32 reactions (see appendix a) were captured in the linear sensitivity analyses reported in (agbro et al., 2017) and these set of reactions are here further analyzed using global uncertainty and sensitivity methods. the global sampling technique described in detail in (tomlin, 2013), was applied in the simulations in order to quantify the error bars of the ignition delays predicted by the trf/n-butanol scheme while incorporating the uncertainties of the input rate parameters in the simulations. uncertainty factors obtained from either both published evaluations (baulch et al., 2005, baulch et al., 1994, baulch et al., 1992, tsang, 1992, tsang and hampson, 1986) and experimental data or from estimates made in the absence of sufficient data were assigned to the 32 most important reactions screened out across the three fuel mixture using the brute-force method. an uncertainty factor of 10 was assigned to the reaction rates in the cases where there were no data on the uncertainty range of the reaction rate. the list of the uncertainty factors assigned to the set of reactions considered in the global analysis of the trf/n-butanol mechanism can be found in appendix a. in addition, a variance-based global sensitivity analysis using hdmr (ziehn and tomlin, 2009) is carried out to understand and rank the parameters responsible for the predicted uncertainties. global sensitivity plots representing the first-order and second-order response between sampled input rates and predicted output are presented and discussed in the result section to explore and demonstrate how the choice of a parameter in the scheme impacts on the predicted ignition delay uncertainties. 3.0 results and discussions 3.1 global uncertainty and sensitivity analysis based on predicted trf ignition delays figure 1 presents the uncertainty plot for predicted trf ignition delays at ϕ = 1 and temperature range of 679 858 k using the blend mechanism while accounting for the effect of uncertainties in the input rate parameters. the uncertainty factors adopted in the uncertainty analysis of the trf/n-butanol blended mechanism are given in global uncertainty and sensitivity analysis of a reduced chemical kinetic mechanism of a gasoline, n-butanol blend in a high pressure rapid compression machine issn: 2180-1053 vol. 10 no.2 june – december 2018 19 appendix a. in figure 1, the boxes represent 25 th and 75 th percentiles while whiskers represent 5th and 95 th percentiles. the blue dashed line represents model simulation with nominal parameter values while the large crosses and horizontal lines represent the mean and median of the predicted output from the 256 simulations respectively. figure 1 shows that the error bars currently existing within the trf system are quite large rising above an order of magnitude in the negative temperature coefficient (ntc) region where the model performance is weakest. however, the experimental data points overlap fairly well with the predicted error bars indicating that reasonable values of uncertainty factors have been adopted for the key rates in the blend scheme. this also indicate that the model is reasonably sound in terms of its structure or mechanistic framework despite the parametric deficiency. within the ntc region, the measured ignition delays are in closest agreement with the 25 th percentile of the predicted distribution suggesting that some key input parameters would need to be fairly close to the limit of their input uncertainty range in order to improve the level of agreement of the model with experimental data. figure 1. comparison of predicted trf ignition delays with experimental data (red line) obtained in agbro et al. figure 2 highlights the first–order global sensitivity indices computed for ignition delay times using the variance based hdmr method for three representative temperature conditions at ϕ = 1 and p = 20 bar. this approach provides a ranking of each input parameter in terms of their contribution to the overall output variance. figure 2 shows that at the lower temperature (i.e. 679 k), a total of seven reactions involving fuel + oh contribute to over 80% of the predicted error bars. the most dominant reaction at lower temperatures is that of oh + toluene expressed as the reverse (ch3 + c6h5oh = c6h5ch3 + oh) with its contribution being about 30 % of the overall predicted uncertainties. this is somewhat surprising since a recent theoretical study by seta et al. (seta et al., 2006) suggested this to be significantly slower than the hydrogen abstraction route via oh. further investigation performed in this study to understand why the h abstraction is not the dominant route is presented in section 3.4. hydrogen abstraction reactions by oh from the α, β and γ sites of iso-octane and n-heptane were also found to play a significant role in agreement with the local sensitivity study presented in (agbro et al., 2017). at higher temperatures, the contribution from the 1 10 100 1000/t (1/k) o v e ra ll ig n it io n d e la y ( m s ) 1.20 1.25 1.30 1.35 1.40 1.45 = 1, t = 679 858 k temperature (k) 840 820 800 780 760 740 720 700 680 journal of mechanical engineering and technology 20 issn: 2180-1053 vol. 10 no.2 june – december 2018 reaction ch3 + c6h5oh = c6h5ch3 + oh diminishes considerably (disappearing at t = 858 k) with the h abstraction reaction from the γ site for iso-octane via oh becoming far more dominant. the main first-order global sensitivities shown in figure 2 indicates that the alkyl + ho2 reactions for toluene are also quite important for the predicted trf ignition delays at high temperatures. also, for toluene a growing importance is observed for the isomerisation reaction from ro2 to qooh as the temperature increases. the white portion in figure 2 represents the contribution from reactions that are not displayed in the legend or the combined effect from higher order terms. figure 2. main first-order sensitivity indices for simulated ignition delays of trf at ϕ =1 and p = 20 bar with respect to the key reaction rates at selected temperatures and pressures. the analysis of the first-order component functions plots further helps to explore the shape of the relationship between the input parameters and the target output. the component function plots shown in figure 3 highlights the individual response of the predicted targets to changes in the a-factor for these reactions. the data points in these figures represent the individual responses from the quasi random sample whereas the line (component function) illustrates the individual effect of the chosen parameter. in each case shown, the middle point on the x-axis (0.5) represents the current nominal value of the a-factor used in the model. the first-order component plots (figure 3) show that at t = 679 k, a nonlinear relationship exists between the target output and input rates across a large portion of the input space for all three most important reactions dominating the predicted uncertainties. decreasing the rate of the phenol route (ch3 + c6h5oh = c6h5ch3 + oh) (figure 3a) would likely improve the agreement with the experimental data at low temperature due to the attendant increase in reactivity while reducing the rate of the h abstraction reaction for iso-octane from the γ site (figure 3c), would have no significant effect on the predicted uncertainties as the effect saturates in the lower part of the input space. on the other hand, increasing the rate of the abstraction reaction from the α site for iso-octane (figure 3b) could potentially lead to an increase in reactivity of the trf system at low temperature and better agreement with experiment but this is still dependent on the influence of second-order and higher order interactions. one interesting thing we observe in figure 3b is that the influence of the uncertainties from all other reactions reduces considerably in the upper part of the input range as shown by the scatter which narrows down in this region and this would global uncertainty and sensitivity analysis of a reduced chemical kinetic mechanism of a gasoline, n-butanol blend in a high pressure rapid compression machine issn: 2180-1053 vol. 10 no.2 june – december 2018 21 suggest that some reasonable level of constraint is provided by the ignition delay measurements on this iso-octane h abstraction rate by oh. (a) (b) (c) figure 3. hdmr component functions (solid line) of simulated trf ignition delays shown on-top of the scatter. p = 20 bar, ϕ = 1, t = 679 k. sensitivity with respect to (a) ch3 + c6h5oh = c6h5ch3 + oh (b) ic8h18 + oh = ac8h17 + h2o (c) ic8h18 + oh = cc8h17 + h2o. journal of mechanical engineering and technology 22 issn: 2180-1053 vol. 10 no.2 june – december 2018 figure 4. hdmr component functions (solid line) of simulated trf ignition delays shown on-top of the scatter. p = 20 bar, ϕ = 1, t = 761 k. sensitivity with respect to ic8h18 + oh = cc8h17 + h2o. within the ntc region, specifically at t = 761k, the iso-octane h abstraction reaction by oh from the γ site, dominates the predicted uncertainties (figure 2) with the reactions ch3 + c6h5oh = c6h5ch3 + oh, ic8h18 + oh = ac8h17 + h2o and nc7h16 + oh = c7h15-2 + h2o also contributing to a smaller degree. however, looking at the functional relationship between the rate of this reaction ic8h18 + oh = cc8h17 + h2o and the predicted ignition delays (figure 4), no significant constraint is provided by the measured delays on this rate in the lower region of the input space where better agreement may be obtained as the slope of the first-order response is very close to zero in that region. on the other hand, a plot of the predicted log ignition delay against the scaled ratio of the log reaction rates for the iso-octane h abstraction reactions by oh from the α and γ site results in an almost linear relationship as shown in figure 5. the computed sensitivity index of this branching fraction for iso-octane is 0.622 which is about three times the value of sensitivity for the individual reactions. again, similar to what was observed for the n-butanol + oh system (agbro and tomlin, 2017), this demonstrates the importance of the relative rates of the hydrogen abstraction reactions of iso-octane from the different sites that lead to chain branching compared to the competing reaction channels that lead to chain propagation or termination, on the accurate prediction of the ignition delay times of trf in the rcm. therefore better constraint is provided by the measured ignition delay data on the branching ratio for isooctane than on the individual abstraction rates via oh from the and γ site. global uncertainty and sensitivity analysis of a reduced chemical kinetic mechanism of a gasoline, n-butanol blend in a high pressure rapid compression machine issn: 2180-1053 vol. 10 no.2 june – december 2018 23 figure 5. scatter plot and hdmr component function for predicted log (ignition delay) of trf against the scaled branching ratio for the two iso-octane main h abstraction reactions t = 761 k, ϕ = 1, p = 20 bar. 3.2 global uncertainty and sensitivity analysis of predicted ignition delays based on the influence of n-butanol blending on gasoline figure 6 and figure 7 show the calculated error bars for the simulated ignition delay times of trf/n-butanol and neat n-butanol using the combined trf and n-butanol scheme adopted in this study. in figure 6 and figure 7, the boxes represent 25 th and 75 th percentiles while whiskers represent 5th and 95 th percentiles. the large crosses and horizontal lines represent the mean and median of the predicted output from the 256 simulations respectively. looking at figure 6, we see that the predicted uncertainties for the trf/n-butanol mixtures are largest (i.e. above an order of magnitude) in the temperature region 761 – 834 k where the discrepancy between the model’s prediction and measured data is most pronounced. however, the experimental data falls well within the median (50 th percentile) of the predicted ignition delay distribution. in contrast to the experimental data, at the lowest temperatures, the simulated ignition delay profiles for n-butanol fall close to the outliers far away from the median of the distribution. in the uncertainty analysis, at very low temperatures, certain combinations of the sampled input rates resulted in extremely long ignition delay times and such results were therefore truncated in order to reduce the required computational time. this explains why the simulated delay times at the nominal rate (blue line) are now shifted closer to the outliers of the distribution rather than the median of the distribution. this explanation is also true for the predicted trf/n-butanol distribution shown in figure 6 but in this case the effect is less pronounced compared to that of pure n-butanol due to the lower predicted ignition delay times of the blend. for n-butanol, the predicted uncertainties (figure 7) are the largest and are over two orders of magnitude in the low temperature region where the models agreement with the measured data is also worse. the discussion in the next section is centred on the global hdmr analysis carried out in order to highlight the journal of mechanical engineering and technology 24 issn: 2180-1053 vol. 10 no.2 june – december 2018 most important reactions influencing the predicted n-butanol and trf/n-butanol output distribution. figure 6. comparison of predicted trf/n-butanol ignition delays (blue) with experimental data (red) obtained in this study. figure 7. comparison of predicted n-butanol ignition delays (blue) with experimental data obtained (red) in this study. figure 8 shows the first-order sensitivity indices calculated in the hdmr analysis for predicted n-butanol + trf ignition delay times. at lower temperatures, the n-butanol + oh hydrogen abstraction reaction from the α site is found to be the most dominant reaction in terms of its contribution to the predicted uncertainties. other key reactions contributing to the predicted uncertainties include c4h8oh-1 + o2 = c4h8oh-1o2, ic8h18 + oh = ac8h17 + h2o and c6h5oh+ ch3 = c6h5ch3 + oh. as the temperature is increased to 858 k, the relative dominance of the n-butanol + oh abstraction reaction 1 10 100 o v e ra ll i g n it io n d e la y ( m s ) 1.20 1.25 1.30 1.35 1.40 1.45 = 1, t = 679 858 k 1000/t (1/k) 840 820 800 780 760 740 720 700 680 temperature (k) 1 10 100 1000 o v e ra ll i g n it io n d e la y ( m s ) 1.20 1.25 1.30 1.35 1.40 1.45 = 1, t = 679 858 k 1000/t (1/k) 840 820 800 780 760 740 720 700 680 temperature (k) global uncertainty and sensitivity analysis of a reduced chemical kinetic mechanism of a gasoline, n-butanol blend in a high pressure rapid compression machine issn: 2180-1053 vol. 10 no.2 june – december 2018 25 from the α site becomes significantly smaller while that of abstraction from the γ site conversely increases with abstraction from the γ site dominating the predicted uncertainties at t = 858 k. the trend observed within the global sensitivity framework for the trf/n-butanol system is similar to that obtained using the local sensitivity approach except that in the local sensitivity analysis, at t = 858 k, the reaction involving hydroperoxyl, leading to the formation of h2o2 (ho2 + ho2 = h2o2 + o2) was slightly more dominant compared to the n-butanol + oh abstraction reaction from the γ site. for the n-butanol system, the results of the hdmr analysis (figure 9) show that at the lower temperature (i.e. t = 679 k), the chain branching pathway (alpha-hydroxybutyl + o2) leading to the formation of the peroxy radical (ro2) (γ -c4h8oh-1 + o2 = c4h8oh1o2) is the most dominant reaction, being responsible for over 20 % of the predicted uncertainties. this was not the case for the local sensitivity analysis of n-butanol where the hydrogen abstraction from the γ site of the n-butanol + oh channel dominated the uncertainties in the predicted ignition delay times. in the n-butanol system, a smaller fraction of the overall uncertainties (about 10 % and 12 %) is also accounted for by the α and γ branching fractions of n-butanol + oh respectively. the slight difference between the most dominant reaction channel obtained in the local sensitivity analysis and that captured in the global sensitivity analysis can be attributed to the impact of the input uncertainty range adopted for the chain branching pathway relative to that of the h abstraction reaction from n-butanol by oh (see appendix a for table of uncertainty range). the impact of the chain branching reaction however diminishes with increases in temperature while the contribution from n-butanol + oh abstraction reaction from the γ site, becomes more significant similar to the result obtained for the local sensitivity analysis. at high temperature, the h abstraction reaction from n-butanol by ho2 leading to the formation of c4h8oh-1 and h2o2 is shown to be equally as important as the abstraction reaction from the γ site. figure 8. main first-order sensitivity indices for simulated trf/n-butanol ignition delays with respect to reaction rates at selected temperatures and pressures. the shading for each reaction is shown in the legend. p = 20 bar, ϕ = 1. journal of mechanical engineering and technology 26 issn: 2180-1053 vol. 10 no.2 june – december 2018 figure 9. main first-order sensitivity indices for simulated n-butanol ignition delays with respect to reaction rates at selected temperatures and pressures. the shading for each reaction is shown in the legend. p = 20 bar, ϕ = 1. based on the computed sensitivity indices from the hdmr analysis (figure 8) it is clear that the branching fractions of n-butanol + oh (α and γ site) with global sensitivity indices of 0.234 and 0.142 respectively are important for the trf/n-butanol system at 761 k as they account for about 40 % of the predicted output uncertainties in this region where the highest discrepancy occurred. figure 10 presents the first-order component plots for these two abstraction reactions at t = 761 k with the scatter in the figure representing the impact of the uncertainties in the other parameters within the mechanism. the overall response of these two parameters to the predicted delays is nonlinear and the overall slopes are opposite to one another. while a reasonable level of constraint is provided in the individual rate of the two abstraction reactions by the measured data as indicated by the computed sensitivities, none of them solely dominates the predicted output uncertainties meaning that different combinations of these two rates could lead to different levels of improvement in terms of the agreement with the experimental data. the high temperature component plot for the trf/n-butanol system (figure 11) shows that a decrease in the γ abstraction rate of n-butanol + oh could potentially also lead to improvement in the model’s prediction at high temperatures. global uncertainty and sensitivity analysis of a reduced chemical kinetic mechanism of a gasoline, n-butanol blend in a high pressure rapid compression machine issn: 2180-1053 vol. 10 no.2 june – december 2018 27 (a) (b) figure 10. component function for trf/n-butanol mixture at t = 761 k with respect to (a) n-c4h9oh + oh = c4h8oh-1 + h2o (b) n-c4h9oh + oh = c4h8oh-3 + h2o. figure 11. component function for trf/n-butanol mixture at t = 858 k with respect to n-c4h9oh + oh = c4h8oh-3+ h2o. figure 12. component function for n-butanol mixture at 679 k with respect to c4h8oh1 + o2 = c4h8oh-1o2. journal of mechanical engineering and technology 28 issn: 2180-1053 vol. 10 no.2 june – december 2018 figure 12 shows that the predicted n-butanol ignition delays are well correlated to the o2 addition pathway and a large increase in this rate could potentially lead to a considerable decrease in the predicted n-butanol delays at lower temperatures. the large scatter however indicates that other reaction pathways such as the competing termination step leading to the formation of ho2 (c4h8oh-1 + o2 = n-c3h7cho + ho2), could become more significant as the rate of this reaction is increased. 3.4 analysis of toluene + oh system 3.4.1 comparison of arrhenius parameters the results of the local sensitivity analysis reported in (agbro et al., 2017) and the global sensitivity analysis described in section 3.3 for predicted ignition delay times for trf using the combined trf/n-butanol mechanism, showed a strong sensitivity to the reaction toluene + oh = phenol + ch3 rather than the hydrogen abstraction channels by oh (toluene + oh = c6h4ch3 + h2o). this was however not expected as a recent study by seta et al. (seta et al., 2006) on the reaction of oh radicals with benzene and toluene suggested that the hydrogen abstraction route (toluene + oh = c6h4ch3 + h2o) is significantly faster than the toluene + oh route leading to the formation of phenol. figure 13 shows arrhenius plots in which the temperature dependence of the forward rates of the toluene + oh = c6h4ch3 + h2o and toluene + oh = phenol + ch3 reaction pathways obtained from the study of seta (seta et al., 2006), are compared. from figure 13, it is clear the oh abstraction routes could be over ten times faster than the phenol route across the temperature range. figure 13. comparison of the forward rates of toluene h abstraction route (toluene + oh) and the phenol route from a recent study of seta (seta et al., 2006). in order to understand why the h abstraction channel is not the dominant route, a critical investigation of the sources of the data for the current parametrisation of the two toluene + oh routes in the available version of the llnl trf mechanism was therefore carried out. it was found in the course of the investigation that the current parametrisation of the h abstraction route (toluene + oh = c6h4ch3 + h2o) in the llnl trf mechanism, is based on the recent data from the theoretical study of seta. the h abstraction reactions in the llnl scheme were updated from the paper of seta 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1e9 1e10 1e11 1e12 1e13 lo g k 1000/t (1/k) toluene + oh = c 6 h 4 ch 3 + h 2 o (seta) toluene + oh = ch 3 + phenol (seta) global uncertainty and sensitivity analysis of a reduced chemical kinetic mechanism of a gasoline, n-butanol blend in a high pressure rapid compression machine issn: 2180-1053 vol. 10 no.2 june – december 2018 29 (seta et al., 2006), but for some reason which is not clear as at the time of this study, it appears that the toluene + oh channel leading to the formation of phenol (toluene + oh = phenol + ch3) was not updated from the same source. in the update of mehl (mehl et al., 2011), all attacks on the toluene ring by oh including the reaction toluene + oh = c6h4ch3 + h2o in the mechanism were taken to be the same with the ones estimated by seta (seta et al., 2006) for benzene. in order to test the impact of the differences between the rate parameterisation of the reversed form of the phenol route (phenol + ch3 = toluene + oh) which is currently in the trf/n-butanol mechanism and that derived from the study of seta (seta et al., 2006), on the predicted ignition delays, a new set of reaction rates was first of all computed for the reversed form of the aforementioned reaction using the forward rate data obtained from the paper of seta (seta et al., 2006). the method employed for the computation of the reversed reaction rates is described briefly in the following section. 3.4.2 calculation of reversed rate based on data of seta in the cantera chemical kinetic tool (version 2.1.2) (goodwin, 2013), the temperature dependence of the forward rate constants 𝑘𝑓 follows the arrhenius expression given by: 𝑘𝑓 = 𝐴𝑓𝑇 𝑛𝑓exp ⁡(−𝐸𝑓 𝑅𝑇)⁄ (1) where𝐴𝑓 is the 𝐴 -factor (pre-exponential factor), 𝑛𝑓 is the temperature exponent, 𝐸𝑓 is the activation energy, t is the absolute temperature and r is the universal gas constant. the equilibrium constant 𝑘𝑒𝑞relates the forward rate to the reversed rate and can be calculated from standard thermodynamic properties using the relationship: 𝑘𝑒𝑞 = exp⁡( 𝛥𝑆0 𝑅 ⁄ ) ∗ exp(−𝛥𝐻 0 𝑅𝑇 ⁄ ) (2) where 𝛥𝑆0 and 𝛥𝐻0 are respectively the standard molar entropy and enthalpy changes of the reaction computed from the respective standard molar entropies 𝑆0 and enthalpies 𝐻0of the species taking part in the reaction and r is the gas constant. also, the equilibrium constant is given by, 𝑘𝑓 𝑘𝑟 =⁡𝑘𝑒𝑞 (3) by using equation 2 and 3, the reversed rates of any reaction can be calculated if the forward rates are known. the temperature-dependent reversed rates for the phenol route were determined using the value of the forward rates of the reaction given in the paper of seta (seta et al., 2006) alongside the equilibrium rate constants estimated using the nasa polynomials in the thermodynamic data of the trf/n-butanol mechanism for the involved species. in the thermodynamic data seven polynomial coefficients are specified for the low temperature range typically from 300 k to 1000 k and another seven for the high temperature range usually from above 1000 k up to 5000 k. the nasa polynomials for standard molar heat capacity at constant pressure 𝐶𝑝 𝜃 , enthalpy 𝐻𝜃, and entropy 𝑆𝜃,⁡take the form: journal of mechanical engineering and technology 30 issn: 2180-1053 vol. 10 no.2 june – december 2018 𝐶𝑝 𝜃 �̅� = 𝑎1 + 𝑎2𝑇 + 𝑎3𝑇 2+⁡𝑎4𝑇 3 +⁡𝑎5𝑇 4 (4) 𝐻𝜃 �̅�𝑇 = 𝑎1 + 𝑎2 2 𝑇 + 𝑎3 3 𝑇2 +⁡ 𝑎4 4 𝑇3 +⁡ 𝑎5 5 𝑇4 +⁡ 𝑎6 𝑇 (5) 𝑆𝜃 �̅� = 𝑎1 ln𝑇 + 𝑎2𝑇 + 𝑎3 2 𝑇2 +⁡ 𝑎4 3 𝑇3 +⁡ 𝑎5 4 𝑇4 +⁡𝑎1 (6) where t is temperature in kelvin, �̅� is the universal gas constant in kj/kmol and the 𝑎𝑛 parameters are the nasa polynomial coefficients. table 2 gives the values of the equilibrium constant and reversed rates calculated across the temperature range 700 -1900 k using equations (1-6). table 2. calculated equilibrium constant and reversed rates temperature (k) 𝑘𝑓(t) c 𝑘𝑒𝑞(𝑇) 𝑘𝑟(𝑇) 700 1.24 x 10 10 6.94 x 10 2 1.79 x 10 7 800 2.43 x 10 10 3.53 x 10 2 6.90 x 10 7 900 4.28 x 10 10 2.09 x 10 2 2.05 x 10 8 1000 6.95 x 10 10 1.38 x 10 2 5.02 x 10 8 1100 1.06 x 10 11 9.85 x 10 1 1.08 x 10 9 1200 1.54 x 10 11 7.42 x 10 1 2.07 x 10 9 1300 2.15 x 10 11 5.83 x 10 1 3.69 x 10 9 1400 2.91 x 10 11 4.73 x 10 1 6.16 x 10 9 1500 3.84 x 10 11 3.95 x 10 1 9.72 x 10 9 1600 4.95 x 10 11 3.37 x 10 1 1.47 x 10 10 1700 6.26 x 10 11 2.93 x 10 1 2.14 x 10 10 1800 7.78 x 10 11 2.57 x 10 1 3.02 x 10 10 1900 9.53 x 10 11 2.27 x 10 1 4.20 x 10 10 c values obtained from the paper of seta (seta et al., 2006) the associated reversed rate parameters required in the cantera input file for the simulations, such as the temperature exponent n, frequency factor a and activation energy e were further estimated using a least square fit to the reversed rate data. as presented in figure 14, a comparison of the rates of the reversed form of the phenol route (toluene + oh = phenol + ch3) captured in the llnl mechanism with those estimated from the data of seta shows a significant difference in their temperature dependence. although both rate constant parameterisation are closely matched at high temperature, the disparity is quite large at lower temperatures. global uncertainty and sensitivity analysis of a reduced chemical kinetic mechanism of a gasoline, n-butanol blend in a high pressure rapid compression machine issn: 2180-1053 vol. 10 no.2 june – december 2018 31 figure 14. comparison of the reversed rates of the phenol route (toluene + oh = phenol + ch3) captured in the llnl mechanism with those estimated from the data of seta (seta et al., 2006). 3.4.3 impact of update on reaction mechanism based on new data the rate of the phenol route in the mechanism was finally updated to that in the paper of seta and variable volume ignition delay simulations were repeated based on the new set of data. figure 15. ignition delay simulations showing how the updated mechanism compares with original llnl data, trf mixtures at p = 20 bar, ϕ = 1. figure 15 shows the result of the predicted trf ignition delays based on the updated mechanism. interestingly, as shown in figure 15, the updated mechanism gives a better agreement with the experimentally measured ignition delays of trf at p =20 bar under stoichiometric conditions. also, we see that the ntc region is now predicted to a higher 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1e7 1e8 1e9 1e10 1e11 l o g k 1000/t (1/k) toluene + oh = ch 3 + phenol (seta) toluene + oh = ch 3 + phenol (mech) 1.2 1.3 1.4 1.5 5 10 15 20 25 30 35 40 p= 20 bar, = 1 gasoline trf variable volume variable volume (updated) ig n it io n d e la y ( m s ) 1000/t(1/k) temperature (k) 860 840 820 800 780 760 740 720 700 680 660 journal of mechanical engineering and technology 32 issn: 2180-1053 vol. 10 no.2 june – december 2018 level of accuracy and this is important for accurate prediction of autoignition and knock in practical engines. the update also leads to a reasonable improvement in the predicted ignition delays of the trf, n-butanol blend (figure 16) mainly within the lower to intermediate temperature region. figure 16. ignition delay simulations showing how the updated mechanism compares with original llnl data, trf, n-butanol blend at p = 20 bar, ϕ = 1. furthermore, local sensitivity analysis was repeated for the trf mixture using the updated scheme to see if the importance of this channel will now be replaced by the h abstraction route. the result of local sensitivity analysis based on the updated mechanism is presented in figure 17 for fifteen (15) of the most sensitive reactions. as expected the toluene + oh hydrogen abstraction route is now captured as one of the most important (dominant) reactions for low temperature ignition delay prediction of trf mixtures while the phenol route is shown to be relatively unimportant as it is not among the set of reactions identified in the sensitivity analysis. interestingly, this is in agreement with the analysis of the component plot presented in figure 3a (section 3.1) where the sensitivity of the phenol route given by the gradient of the curve is shown to be quite low at the lower end of the adopted input range. it is also worth pointing out that based on the update, the iso-octane chemistry, specifically the iso-octane + oh hydrogen abstraction reaction from the γ site (figure 17) now dominates the predicted ignition delays of trf. also the alkyl + ho2 route for toluene which was prominent at higher temperatures in the local sensitivity result based on the original trf/n-butanol mechanism has now disappeared. 1.2 1.3 1.4 1.5 1 10 100 gasoline + 20% n-butanol trf + 20% n-butanol variable volume variable volume (updated) ig n it io n d e la y ( m s ) 1000/t (1/k) 860 840 820 800 780 760 740 720 700 680 660 p= 20 bar, = 1 temperature (k) global uncertainty and sensitivity analysis of a reduced chemical kinetic mechanism of a gasoline, n-butanol blend in a high pressure rapid compression machine issn: 2180-1053 vol. 10 no.2 june – december 2018 33 figure 17. brute-force local sensitivity result based on updated mechanism for trf mixtures at p = 20 bar, ϕ = 1. 4.0 conclusions a recently developed reduced chemical kinetic mechanism describing the low temperature oxidation of n-butanol, gasoline and a gasoline/n-butanol blend was investigated using both global uncertainty and sensitivity methods with ignition delays as the predicted output for the temperature range 678 858 k, and an equivalence ratio of 1 at 20 bar. the work highlights and elucidates on the most important input parameters influencing the predictive output uncertainties in the chemical kinetic models when incorporating the effects of uncertainties in forward rate constants within a global sampling approach. for trf, a total of seven reactions involving fuel + oh were identified as contributing to over 80% of the predicted error bars. the dominant reaction at lower temperatures is that of oh + toluene expressed as the reverse (ch3 + c6h5oh = c6h5ch3 + oh) but an update on the mechanism based on recent data from the study of seta resulted in the toluene + oh channel becoming the most dominant reaction as expected. at higher temperatures, the contribution from the reaction ch3 + c6h5oh = c6h5ch3 + oh diminishes considerably (disappearing at 858 k) while the h abstraction reaction from the γ site via oh for iso-octane becomes far more dominant. the work showed that the hydrogen abstraction reactions by oh from n-butanol are the most important reactions in predicting the effect of n-butanol blending on gasoline particularly at the low temperature but these rates are still currently not well known and hence the large discrepancies currently existing in the models prediction in the low temperature region. for predicted n-butanol ignition delay times, the chain branching pathway (α-hydroxybutyl + o2) leading to the formation of the peroxy radical (ro2) (αc4h8oh-1 + o2 = c4h8oh-1o2) is the most dominant, being responsible for over 20 percent of the predicted uncertainties. for both the n-butanol and trf/n-butanol system, the contribution from n-butanol + oh abstraction reaction from the γ site, is the most significant at higher temperatures (i.e. 858k). i-c 8 h 18 +oh=cc 8 h 17 +h 2 o c 6 h 5 ch 3 +oh =c 6 h 5 ch 2 +h 2 o i-c 8 h 18 +oh=ac 8 h 17 +h 2 o i-c 8 h 18 +oh=bc 8 h 17 +h 2 o ho 2 +ho 2 =h 2 o 2 +o 2 n-c 7 h 16 +oh=c 7 h 15 -2+h 2 o ch 2 o+oh=hco+h 2 o n-c 7 h 16 +oh=c 7 h 15 -3+h 2 o i-c 8 h 18 +oh=dc 8 h 17 +h 2 o dc 8 h 17 o 2 =dc 8 h 16 ooh-c dc 8 h 17 o 2 =ac 8 h 16 ooh-b dc 8 h 17 o 2 =dc 8 h 16 ooh-b h 2 o 2 (+m)=oh+oh(+m) n-c 7 h 16 +oh=c 7 h 15 -1+h 2 o ch 3 o 2 +ho 2 =ch 3 o 2 h+o 2 -1.0 -0.5 0.0 0.5 1.0 normalized local sensitivity index 858k 761k 679k journal of mechanical engineering and technology 34 issn: 2180-1053 vol. 10 no.2 june – december 2018 the global sensitivity plots representing the first-order and second-order response between sampled input rates and predicted output were also discussed to explore and illustrate how the choice of a parameterization in the scheme impacts on the predicted output uncertainties. first-order functional plots for trf indicate that modifications to the rate of fuel + oh for toluene and that of h abstraction for iso-octane from the γ site are unlikely to improve the level of agreement with the experimental data at lower temperatures, but increasing the rate of the abstraction reaction from the α site for isooctane could lead to a decrease in reactivity and better agreement. this is however, dependent on the influence of second-order and higher-order interactions. within the ntc region where the iso-octane h abstraction reaction by oh from the γ site dominates, better constraint is provided by the measured ignition delay data on the rate of the branching ratio for iso-octane than on the overall or individual abstraction rate for the α and γ site. for trf + n-butanol, the overall response for the two most dominant nbutanol + oh abstraction rates (α and γ site) to the predicted delays is nonlinear and opposite to one another. while a reasonable level of constraint is provided in the individual rate of these two abstraction reactions by the measured data, as indicated by the computed sensitivities, none of them solely dominants the predicted output uncertainties. for predicted n-butanol + trf ignition delay times, the n-butanol + oh hydrogen abstraction reaction from the α site is found to be the most dominant in terms of its contribution to the predicted uncertainties, despite the low blending ratio of butanol at 20%, while. 5.0 acknowledgements the author would like to thank the tertiary education trust fund (tetfund), nigeria, for providing funding for this research. my sincere appreciation also goes to professor alison tomlin for her guidance and most importantly for the helpful discussions. 6.0 references agarwal, a. k. 2007. biofuels (alcohols and biodiesel) applications as fuels for internal combustion engines. progress in energy and combustion science, 33, 233-271. agbro, e. 2017. experimental and chemical kinetic modelling study on the combustion of alternative fuels in fundamental systems and practical engines. phd, the university of leeds. agbro, e. & tomlin, a. s. 2017. low temperature oxidation of n-butanol: key uncertainties and constraints in kinetics. fuel, 207, 776-789. agbro, e., tomlin, a. s., lawes, m., park, s. & sarathy, s. m. 2017. the influence of n-butanol blending on the ignition delay times of gasoline and its surrogate at high pressures. fuel, 187, 211-219. andrae, j. c. g. 2008. development of a detailed kinetic model for gasoline surrogate fuels. fuel, 87, 2013-2022. global uncertainty and sensitivity analysis of a reduced chemical kinetic mechanism of a gasoline, n-butanol blend in a high pressure rapid compression machine issn: 2180-1053 vol. 10 no.2 june – december 2018 35 andrae, j. c. g., björnbom, p., cracknell, r. f. & kalghatgi, g. t. 2007. autoignition of toluene reference fuels at high pressures modeled with detailed chemical kinetics. combustion and flame, 149, 2-24. baulch, d. l. 1997. kinetic databases. in: pilling, m. j. (ed.) comprehensive chemical kinetics: low temperature combustion and a uto-ignition. elservier. baulch, d. l., bowman, c. t., cobos, c. j., cox, r. a., just, t. h., kerr, j. a., pilling, m. j., stocker, d., troe, j., tsang, w., walker, r. w. & warnatz, j. 2005. evaluated kinetic data for combustion modeling: supplement ii. journal of physical and chemical reference data, 34, 757-1397. baulch, d. l., cobos, c. j., cox, r. a., esser, c., frank, p., just, t., kerr, j. a., pilling, m. j., troe, j., walker, r. w. & warnatz, j. 1992. evaluated kinetic data for combustion modelling. journal of physical and chemical reference data, 21, 411-734. baulch, d. l., cobos, c. j., cox, r. a., frank, j. h., hayman, g., just, t. h., kerr, j. a., murrels, t., pilling, m. j.; troe, j., walker, b. f. & warnatz, j. 1994. summary table of evaluated kinetic data for combustion modeling: supplement 1. combustion and flame, 59-79. dernotte, j., mounaim-rousselle, c., halter, f. & seers, p. 2009. evaluation of butanol–gasoline blends in a port fuel-injection, spark-ignition engine. oil & gas science and technology–revue de l’institut français du pétrole, 65, 345351. glaude, p. a., conraud, v., fournet, r., battin-leclerc, f., côme, g. m., scacchi, g., dagaut, p. & cathonnet, m. 2002. modeling the oxidation of mixtures of primary reference automobile fuels. energy & fuels, 16, 1186-1195. goodwin, d. m., n; moffat, h; speth, r 2013. ca ntera : an object-oriented software toolit for chemical kinetics, thermodynamics, and transport processes, https://code.google.com/p/cantera. mehl, m., pitz, w. j., westbrook, c. k. & curran, h. j. 2011. kinetic modeling of gasoline surrogate components and mixtures under engine conditions. proceedings of the combustion institute, 33, 193-200. naik, c. v., pitz, w. j., westbrook, c. k., sjöberg, m., dec, j. e., orme, j., curran, h. j. & simmie, j. m. 2005. detailed chemical kinetic modeling of surrogate fuels for gasoline and application to an hcci engine. sae international. reaction design 2011. chemkin-pro, san diego. sarathy, s. m., oßwald, p., hansen, n. & kohse-höinghaus, k. 2014. alcohol combustion chemistry. progress in energy and combustion science, 44, 40-102. seta, t., nakajima, m. & miyoshi, a. 2006. high-temperature reactions of oh radicals with benzene and toluene. the journal of physical chemistry a , 110, 50815090. journal of mechanical engineering and technology 36 issn: 2180-1053 vol. 10 no.2 june – december 2018 szwaja, s. & naber, j. d. 2010. combustion of n-butanol in a spark-ignition ic engine. fuel, 89, 1573-1582. tanaka, s., ayala, f. & keck, j. c. 2003. a reduced chemical kinetic model for hcci combustion of primary reference fuels in a rapid compression machine. combustion and flame, 133, 467-481. tomlin, a. s. 2006. the use of global uncertainty methods for the evaluation of combustion mechanisms. reliability engineering & system safety, 91, 12191231. tomlin, a. s. 2013. the role of sensitivity and uncertainty analysis in combustion modelling. proceedings of the combustion institute, 34, 159-176. tomlin, a. s. & turanyi, t. 2013. investigation and improvement of mechanism using sensitivity analysis and optimization. in: battin-leclerc, f., simmie, j. m. & blurock, e. (eds.) cleaner combustion: developing detailed chemical kinetic models. london: springer-verlag. tsang, w. 1992. chemical kinetic data base for propellant combustion. ii. reactions involving cn, nco, and hnco. journal of physical and chemical reference data, 21, 753-791. tsang, w. & hampson, r. f. 1986. chemical kinetic data base for combustion chemistry. part i. methane and related compounds. journal of physical and chemical reference data, 15, 1087-1279. weber, b. w. & sung, c.-j. 2013. comparative autoignition trends in butanol isomers at elevated pressure. energy & fuels, 27, 1688-1698. westbrook, c. k., warnatz, j. & pitz, w. j. 1988. a detailed chemical kinetic reaction mechanism for the oxidation of iso-octane and n-heptane over an extended temperature range and its application to analysis of engine knock. symposium (international) on combustion, 22, 893-901. wigg, b. 2011. a study on emission of butanol using a spark ignition engine and their reduction using electrostatically a ssisted injection. phd, university of ilinois, urbana. ziehn, t. & tomlin, a. 2009. gui–hdmr–a software tool for global sensitivity analysis of complex models. environmental modelling & software, 24, 775785. global uncertainty and sensitivity analysis of a reduced chemical kinetic mechanism of a gasoline, n-butanol blend in a high pressure rapid compression machine issn: 2180-1053 vol. 10 no.2 june – december 2018 37 appendix a reactions selected from local sensitivity analysis of trf/n-butanol blended mechanism and assigned input uncertainty factors reaction gi k max k min source of uncertainty information ho2 + ho2 = h2o2 + o2 1.41 (baulch et al., 2005) h2o2 (+ m) = oh + oh (+m) ( k0,k∞) 3.16 (baulch et al., 2005) h2o2 + oh = h2o + ho2 1.58 (tsang, 1992) ch2o + oh = hco + h2o 2.24 (baulch et al., 2005) ch3o2 + ho2 = ch3o2h + o2 5.0 estimated nc3h7o2 = c3h6ooh1-3 10.0 estimated nc4ket13 = ch3cho + ch2cho + oh 10.0 estimated tc4h9o2 = ic4h8 + ho2 10.0 estimated ic8h18 + oh = ac8h17 + h2o 7.94 estimated ic8h18 + oh = bc8h17 + h2o 3.98 estimated ic8h18 + oh = cc8h17 + h2o 7.94 estimated ac8h17+ o2 = ac8h17o2 10.0 estimated dc8h17o2 = dc8h16 ooh-b 10.0 estimated dc8h17o2 = dc8h16 ooh-c 10.0 estimated nc7h16 + oh = c7h15-1 + + h2o 10.0 estimated nc7h16 + oh = c7h15-2 + + h2o 10.0 estimated nc7h16 + oh = c7h15-3 + + h2o 10.0 estimated c7h15o2-2 = c7h14 ooh2-4 10.0 estimated c6h5oh+ ch3 = c6h5ch3 + oh 10.0 estimated c6h5ch3 + ho2 = = c6h5ch2 j + h2o2 3.16 estimated c6h5ch2j + ho2 = c6h5ch2oj + oh 7.94 estimated c4h9oh +oh= c4h8oh-1 + h2o 10.0 estimated c4h9oh +oh= c4h8oh-3 + h2o 10.0 estimated c4h9oh +oh= c4h8oh-4 + h2o 10.0 estimated c4h9oh + ho2 = c4h8oh-1 + h2o2 10.0 estimated c4h8oh-1 + o2 = c3h7cho + ho2 10.0 estimated c4h8oh-1 + o2 = c4h8oh-1o2 10.0 estimated c4h8oh-1o2 = c4h7oh-1ooh-3 10.0 estimated c4h8oh-3o2 = c4h7oh-3ooh-1 10.0 estimated c4h8oh-1o2 = c4h7oh1-1 + ho2 10.0 estimated c4h7oh-3ooh-1 + o2 = nc4ket13 + ho2 10.0 estimated c4h7oh-3ooh-1 + o2 = c4h7oh3ooh-1o2 10.0 estimated journal of mechanical engineering and technology 38 issn: 2180-1053 vol. 10 no.2 june – december 2018 issn: 2180-1053 vol. 2 no. 1 january-june 2010 image classification of temperature distribution using fourier series strategy 11 image classification of temperature distribution using fourier series strategy m.a.salim1, a. r. m. rosdzimin2, k.osman3, a.noordin4, m.a.mohd rosli5 1,5faculty of mechanical engineering, universiti teknikal malaysia melaka, karung berkunci 1752, pejabat pos durian tunggal, 76109 durian tunggal, melaka, malaysia. 2department of mechanical, faculty of engineering, universiti pertahanan nasional malaysia, kem sungai besi, 57000 kuala lumpur, malaysia. 3faculty of mechanical engineering, universiti teknologi malaysia, 81310 skudai, johor bahru, johor, malaysia. 4faculty of electrical engineering, universiti teknikal malaysia melaka, karung berkunci 1752, pejabat pos durian tunggal, 76109 durian tunggal, melaka, malaysia. email: 1azli@utem.edu.my, 2rosdzimin@upnm.edu.my, 3kahar@fkm.utm.my, 4aminurrashid@ utem.edu.my, 5afzanizam@utem.edu.my abstract this paper presents the analysis of temperature distribution by using fourier series strategy. using this strategy, two analyses have been made where an even and odd number is applied in the series. in even number, the trivial solution occur where there are no solution been made, but in odd number the solution is succeed. by using the equation made by odd number, the equation has been analysis using matlab-programming. in this analysis, the contour mapping of temperature distribution is shown clearly. based on the mapping, it can be concluded that the temperature distribution happens because of the adiabatic phenomenon of the material properties itself. keywords: fourier series, isotherms, adiabatic, temperature distribution. 1.0 introduction it is not easy to give an exact definition of temperature because issn: 2180-1053 vol. 2 no. 1 january-june 2010 journal of mechanical engineering and technology 12 majority of human around the world classify temperature to only familiar with hotness or coldness. in physiological ambiance, the level of temperature can be measured as a cold, freezing cold, warm, hot and red hot. for the basic example, metal will feel much colder than the wood when both materials are at the same temperature and same place. according to this example, several materials properties can be changed regarding the changes of the temperature around the material itself. the material properties can be changed with temperature in a repeatable and predictable way according the basis of accurate temperature measurement. in a common example, a cup of hot tea present on the table ultimately cools off and cold drink sooner or later warms up. in this phenomenon, a body of hot tea and cold drink are brought into contact with another body with different temperature and heat transfer phenomenon is happen where the hot temperature transfer it temperature to the low temperature areas. this phenomenon will continue until both area reach with same temperature. when both areas are in a same temperature, these two areas are said in a thermal equilibrium condition. zeroth law of thermodynamics explains that if two bodies in thermal equilibrium with a third body, they are also in thermal equilibrium for each others. this law had been formulated by r.h.fowler in 1931. in this study, the authors have studied the mediums’ properties according to the heat transfer effect in several materials. therefore, the knowledge on medium’s value of temperature at all point is necessary. heat transfer analysis basically plays a central role in the design of chemical processes and in the development of process system. in order to make accurate analysis to the heat transfer problem, parameters such as the roll speed, thermal conductivity, rate of cold air, thickness and temperature surface is needed. heat transfer problem can occur by three mechanisms and there are conduction, convection and radiation. conduction mechanism is a collision of molecules causes the thermal energy to be transferred from one molecule to other one molecule. in this heat transfer process, the very energetic molecules will lose their energy while the lower energy molecules will get the more energy. in convection mechanism, it only occurs when the energy in macroscopic flow in fluid was associated with a parcel. then, the fluid was converted to another region of space. in this case, it also can be called as an unsteady state behaviour. the radiation mechanism happens when the molecular vibrate and give an electromagnet radiation which certain amount to the other molecular. the radiation behaviour transmits the energy throughout the space issn: 2180-1053 vol. 2 no. 1 january-june 2010 image classification of temperature distribution using fourier series strategy 13 and vacuum containing. these three mechanisms have a potential to generate the instantaneous values of temperature at all points of the medium of interest and also called as a temperature distribution or field. the unsteady temperature distribution appears when medium’s temperature not only varies from point to point, but also depends on time. in time domain, the temperature at a various points in a medium can be changed and then the internal energy of the molecular also changed. a steady state temperature distribution occurs when a temperature at a given point never varies with time and this type is called as a space coordinates only. the temperature distribution by governing equations also called a three dimensional. this equation of temperature is describes as a function of three space coordinates, temperature, these two areas are said in a thermal equilibrium condition. zeroth law of thermodynamics explains that if two bodies in thermal equilibrium with a third body, they are also in thermal equilibrium for each others. this law had been formulated by r.h.fowler in 1931. in this study, the authors have studied the mediums’ properties according to the heat transfer effect in several materials. therefore, the knowledge on medium’s value of temperature at all point is necessary. heat transfer analysis basically plays a central role in the design of chemical processes and in the development of process system. in order to make accurate analysis to the heat transfer problem, parameters such as the roll speed, thermal conductivity, rate of cold air, thickness and temperature surface is needed. heat transfer problem can occur by three mechanisms and there are conduction, convection and radiation. conduction mechanism is a collision of molecules causes the thermal energy to be transferred from one molecule to other one molecule. in this heat transfer process, the very energetic molecules will lose their energy while the lower energy molecules will get the more energy. in convection mechanism, it only occurs when the energy in macroscopic flow in fluid was associated with a parcel. then, the fluid was converted to another region of space. in this case, it also can be called as an unsteady state behaviour. the radiation mechanism happens when the molecular vibrate and give an electromagnet radiation which certain amount to the other molecular. the radiation behaviour transmits the energy throughout the space and vacuum containing. these three mechanisms have a potential to generate the instantaneous values of temperature at all points of the medium of interest and also called as a temperature distribution or field. the unsteady temperature distribution appears when medium’s temperature not only varies from point to point, but also depends on time. in time domain, the temperature at a various points in a medium can be changed and then the internal energy of the molecular also changed. a steady state temperature distribution occurs when a temperature at a given point never varies with time and this type is called as a space coordinates only. the temperature distribution by governing equations also called a three dimensional. this equation of temperature is describes as a function of three space coordinates, . therefore, if the points of a medium with equal temperatures are connected, the resulting surfaces are called isothermal surfaces. this intersection of isothermal surfaces with a plane yields a family of isotherms on the place surface. important to note that the two isothermal surfaces never cut each other since there are no part of the medium can have two different temperatures at the same time, respectively (kreysig, 2006), (howard et al., 2003). fourier series of all dimension is a general type of summation process under which the convergence or non-convergence of the corresponding partial sums at a given point depend only on the behaviour of the function at given point, and that continuity of the function at the point is sufficient for convergence (bochner, 1935). fourier series has been used for solving many heat transfer problems. maria and power (2000) was developed an efficient bem scheme for the numerical solution of two-dimensional heat problems. the double fourier series was rewritten using green function that obtained by the images method. the double fourier series is use in the domain integral of the integral representation formula to transform such integral into equivalent surface integrals. maksimovich and tsybul''skii (2004) determined nonstationary nonaxisymmetric temperature fields in bodies of revolution appearing on heating by internal heat sources through and due to convective heat exchange with an external medium. the solution of the problem is represented in the form of a fourier series in an angular coordinate with coefficients being determined by a method of boundary elements. . therefore, if the points of a medium with equal temperatures are connected, the resulting surfaces are called isothermal surfaces. this intersection of isothermal surfaces with a plane yields a family of isotherms on the place surface. important to note that the two isothermal surfaces never cut each other since there are no part of the medium can have two different temperatures at the same time, respectively (kreysig, 2006), (howard et al., 2003). fourier series of all dimension is a general type of summation process under which the convergence or non-convergence of the corresponding partial sums at a given point depend only on the behaviour of the function at given point, and that continuity of the function at the point is sufficient for convergence (bochner, 1935). fourier series has been used for solving many heat transfer problems. maria and power (2000) was developed an efficient bem scheme for the numerical solution of two-dimensional heat problems. the double fourier series was rewritten using green function that obtained by the images method. the double fourier series is use in the domain integral of the integral representation formula to transform such integral into equivalent surface integrals. maksimovich and tsybul’’skii (2004) determined nonstationary nonaxisymmetric temperature fields in bodies of revolution appearing on heating by internal heat sources through and due to convective heat exchange with an external medium. the solution of the problem is represented in the form of a fourier series in an angular coordinate with coefficients being determined by a method of boundary elements. issn: 2180-1053 vol. 2 no. 1 january-june 2010 journal of mechanical engineering and technology 14 2.0 fourier series strategy fourier series is a one method to analyze the periodic phenomenon and they occur quite frequently in engineering and elsewhere such as in rotating of machines, alternating electric currents or the motions of planets. fourier series is also called as a trigonometric series and it represent in periodic functions. the function on trigonometric can be defined from 2.0 fourier series strategy fourier series is a one method to analyze the periodic phenomenon and they occur quite frequently in engineering and elsewhere such as in rotating of machines, alternating electric currents or the motions of planets. fourier series is also called as a trigonometric series and it represent in periodic functions. the function on trigonometric can be defined from , and has a period of of if for all values of x. according to the definition of the periodic, most practical situations such as a function can be expressed as a complex fourier series like (kreysig, 2006), (black, 2008): the number is called as complex coefficients periodic functions. it can be compute by the integration as: this fourier series also can be written as sine and cosine function. the new equation of this fourier series represents as a: by the denoting the equation: yields: this series is called a fourier sine-cosine expansion. for this case, for all values of j, which it is, implies that must be real and then: suppose that, fourier series expansion for a more general function of and having a period of p instead of . then, the new function can be introduced by: and has a period of 2.0 fourier series strategy fourier series is a one method to analyze the periodic phenomenon and they occur quite frequently in engineering and elsewhere such as in rotating of machines, alternating electric currents or the motions of planets. fourier series is also called as a trigonometric series and it represent in periodic functions. the function on trigonometric can be defined from , and has a period of of if for all values of x. according to the definition of the periodic, most practical situations such as a function can be expressed as a complex fourier series like (kreysig, 2006), (black, 2008): the number is called as complex coefficients periodic functions. it can be compute by the integration as: this fourier series also can be written as sine and cosine function. the new equation of this fourier series represents as a: by the denoting the equation: yields: this series is called a fourier sine-cosine expansion. for this case, for all values of j, which it is, implies that must be real and then: suppose that, fourier series expansion for a more general function of and having a period of p instead of . then, the new function can be introduced by: for all values of x. according to the definition of the periodic, most practical situations such as a function can be expressed as a complex fourier series like (kreysig, 2006), (black, 2008): 2.0 fourier series strategy fourier series is a one method to analyze the periodic phenomenon and they occur quite frequently in engineering and elsewhere such as in rotating of machines, alternating electric currents or the motions of planets. fourier series is also called as a trigonometric series and it represent in periodic functions. the function on trigonometric can be defined from , and has a period of of if for all values of x. according to the definition of the periodic, most practical situations such as a function can be expressed as a complex fourier series like (kreysig, 2006), (black, 2008): the number is called as complex coefficients periodic functions. it can be compute by the integration as: this fourier series also can be written as sine and cosine function. the new equation of this fourier series represents as a: by the denoting the equation: yields: this series is called a fourier sine-cosine expansion. for this case, for all values of j, which it is, implies that must be real and then: suppose that, fourier series expansion for a more general function of and having a period of p instead of . then, the new function can be introduced by: the number is called as complex coefficients periodic functions. it can be compute by the integration as: 2.0 fourier series strategy fourier series is a one method to analyze the periodic phenomenon and they occur quite frequently in engineering and elsewhere such as in rotating of machines, alternating electric currents or the motions of planets. fourier series is also called as a trigonometric series and it represent in periodic functions. the function on trigonometric can be defined from , and has a period of of if for all values of x. according to the definition of the periodic, most practical situations such as a function can be expressed as a complex fourier series like (kreysig, 2006), (black, 2008): the number is called as complex coefficients periodic functions. it can be compute by the integration as: this fourier series also can be written as sine and cosine function. the new equation of this fourier series represents as a: by the denoting the equation: yields: this series is called a fourier sine-cosine expansion. for this case, for all values of j, which it is, implies that must be real and then: suppose that, fourier series expansion for a more general function of and having a period of p instead of . then, the new function can be introduced by: this fourier series also can be written as sine and cosine function. the new equation of this fourier series represents as a: 2.0 fourier series strategy fourier series is a one method to analyze the periodic phenomenon and they occur quite frequently in engineering and elsewhere such as in rotating of machines, alternating electric currents or the motions of planets. fourier series is also called as a trigonometric series and it represent in periodic functions. the function on trigonometric can be defined from , and has a period of of if for all values of x. according to the definition of the periodic, most practical situations such as a function can be expressed as a complex fourier series like (kreysig, 2006), (black, 2008): the number is called as complex coefficients periodic functions. it can be compute by the integration as: this fourier series also can be written as sine and cosine function. the new equation of this fourier series represents as a: by the denoting the equation: yields: this series is called a fourier sine-cosine expansion. for this case, for all values of j, which it is, implies that must be real and then: suppose that, fourier series expansion for a more general function of and having a period of p instead of . then, the new function can be introduced by: by the denoting the equation: 2.0 fourier series strategy fourier series is a one method to analyze the periodic phenomenon and they occur quite frequently in engineering and elsewhere such as in rotating of machines, alternating electric currents or the motions of planets. fourier series is also called as a trigonometric series and it represent in periodic functions. the function on trigonometric can be defined from , and has a period of of if for all values of x. according to the definition of the periodic, most practical situations such as a function can be expressed as a complex fourier series like (kreysig, 2006), (black, 2008): the number is called as complex coefficients periodic functions. it can be compute by the integration as: this fourier series also can be written as sine and cosine function. the new equation of this fourier series represents as a: by the denoting the equation: yields: this series is called a fourier sine-cosine expansion. for this case, for all values of j, which it is, implies that must be real and then: suppose that, fourier series expansion for a more general function of and having a period of p instead of . then, the new function can be introduced by: yields: 2.0 fourier series strategy fourier series is a one method to analyze the periodic phenomenon and they occur quite frequently in engineering and elsewhere such as in rotating of machines, alternating electric currents or the motions of planets. fourier series is also called as a trigonometric series and it represent in periodic functions. the function on trigonometric can be defined from , and has a period of of if for all values of x. according to the definition of the periodic, most practical situations such as a function can be expressed as a complex fourier series like (kreysig, 2006), (black, 2008): the number is called as complex coefficients periodic functions. it can be compute by the integration as: this fourier series also can be written as sine and cosine function. the new equation of this fourier series represents as a: by the denoting the equation: yields: this series is called a fourier sine-cosine expansion. for this case, for all values of j, which it is, implies that must be real and then: suppose that, fourier series expansion for a more general function of and having a period of p instead of . then, the new function can be introduced by: this series is called a fourier sine-cosine expansion. for this case, 2.0 fourier series strategy fourier series is a one method to analyze the periodic phenomenon and they occur quite frequently in engineering and elsewhere such as in rotating of machines, alternating electric currents or the motions of planets. fourier series is also called as a trigonometric series and it represent in periodic functions. the function on trigonometric can be defined from , and has a period of of if for all values of x. according to the definition of the periodic, most practical situations such as a function can be expressed as a complex fourier series like (kreysig, 2006), (black, 2008): the number is called as complex coefficients periodic functions. it can be compute by the integration as: this fourier series also can be written as sine and cosine function. the new equation of this fourier series represents as a: by the denoting the equation: yields: this series is called a fourier sine-cosine expansion. for this case, for all values of j, which it is, implies that must be real and then: suppose that, fourier series expansion for a more general function of and having a period of p instead of . then, the new function can be introduced by: for all values of j, which it is, implies that must be real and then: issn: 2180-1053 vol. 2 no. 1 january-june 2010 image classification of temperature distribution using fourier series strategy 15 2.0 fourier series strategy fourier series is a one method to analyze the periodic phenomenon and they occur quite frequently in engineering and elsewhere such as in rotating of machines, alternating electric currents or the motions of planets. fourier series is also called as a trigonometric series and it represent in periodic functions. the function on trigonometric can be defined from , and has a period of of if for all values of x. according to the definition of the periodic, most practical situations such as a function can be expressed as a complex fourier series like (kreysig, 2006), (black, 2008): the number is called as complex coefficients periodic functions. it can be compute by the integration as: this fourier series also can be written as sine and cosine function. the new equation of this fourier series represents as a: by the denoting the equation: yields: this series is called a fourier sine-cosine expansion. for this case, for all values of j, which it is, implies that must be real and then: suppose that, fourier series expansion for a more general function of and having a period of p instead of . then, the new function can be introduced by: suppose that, fourier series expansion for a more general function of and having a period of p instead of . then, the new function can be introduced by: this has a period of . this function of can be rewrite to a new equation and representing as: whereas: therefore the equation can be express into: sometimes, there occur functions that can be expanding as a series of a sine term only or as a series of cosine terms only. in a case, if the function is originally defined for therefore, it can be making as a this is a series where it involves only a sine terms. similarly that, if this term consists only cosine term and finally, the equation can be represented as: this has a period of . this function of can be rewrite to a new equation and representing as: this has a period of . this function of can be rewrite to a new equation and representing as: whereas: therefore the equation can be express into: sometimes, there occur functions that can be expanding as a series of a sine term only or as a series of cosine terms only. in a case, if the function is originally defined for therefore, it can be making as a this is a series where it involves only a sine terms. similarly that, if this term consists only cosine term and finally, the equation can be represented as: where as: this has a period of . this function of can be rewrite to a new equation and representing as: whereas: therefore the equation can be express into: sometimes, there occur functions that can be expanding as a series of a sine term only or as a series of cosine terms only. in a case, if the function is originally defined for therefore, it can be making as a this is a series where it involves only a sine terms. similarly that, if this term consists only cosine term and finally, the equation can be represented as: therefore the equation can be express into: this has a period of . this function of can be rewrite to a new equation and representing as: whereas: therefore the equation can be express into: sometimes, there occur functions that can be expanding as a series of a sine term only or as a series of cosine terms only. in a case, if the function is originally defined for therefore, it can be making as a this is a series where it involves only a sine terms. similarly that, if this term consists only cosine term and finally, the equation can be represented as: sometimes, there occur functions that can be expanding as a series of a sine term only or as a series of cosine terms only. in a case, if the function is originally defined for this has a period of . this function of can be rewrite to a new equation and representing as: whereas: therefore the equation can be express into: sometimes, there occur functions that can be expanding as a series of a sine term only or as a series of cosine terms only. in a case, if the function is originally defined for therefore, it can be making as a this is a series where it involves only a sine terms. similarly that, if this term consists only cosine term and finally, the equation can be represented as: therefore, it can be making as a this has a period of . this function of can be rewrite to a new equation and representing as: whereas: therefore the equation can be express into: sometimes, there occur functions that can be expanding as a series of a sine term only or as a series of cosine terms only. in a case, if the function is originally defined for therefore, it can be making as a this is a series where it involves only a sine terms. similarly that, if this term consists only cosine term and finally, the equation can be represented as: this is a series where it involves only a sine terms. similarly that, if this has a period of . this function of can be rewrite to a new equation and representing as: whereas: therefore the equation can be express into: sometimes, there occur functions that can be expanding as a series of a sine term only or as a series of cosine terms only. in a case, if the function is originally defined for therefore, it can be making as a this is a series where it involves only a sine terms. similarly that, if this term consists only cosine term and finally, the equation can be represented as: issn: 2180-1053 vol. 2 no. 1 january-june 2010 journal of mechanical engineering and technology 16 this term consists only cosine term and finally, the equation can be represented as: this has a period of . this function of can be rewrite to a new equation and representing as: whereas: therefore the equation can be express into: sometimes, there occur functions that can be expanding as a series of a sine term only or as a series of cosine terms only. in a case, if the function is originally defined for therefore, it can be making as a this is a series where it involves only a sine terms. similarly that, if this term consists only cosine term and finally, the equation can be represented as: whereas, this equation can be modifying as: whereas, this equation can be modifying as: fourier series is a function of approximation using a finite number of terms and then, the resulting function may oscillate in regions where the actual function is discontinues or in other hand it is called as changes rapidly. this undesirable behaviour can be reduced by using a smoothing procedure described by lanezos where it use this fourier series and close it related to a function of and it is defined by a local averaging process according to equation 16 below. the averaging interval of should be a small fraction of the period of . then, rewrite the equation of with . the functions of and are identical as . hence, also match exactly at any point of x, where varies linearly between importantly, this agreed closely with for a small value of but at the same time in a fourier series which converges more rapidly than the series for . last but not least, it can be defined as: where and for . then, evidently this fourier series of the coefficients of are easily obtained from the function of . when this series for converges slowly, using the same number of the terms in the series for often gives an approximation preferable to that provided by the series of . finally this process is called as smoothing. fourier series is a function of approximation using a finite number of terms and then, the resulting function may oscillate in regions where the actual function is discontinues or in other hand it is called as changes rapidly. this undesirable behaviour can be reduced by using a smoothing procedure described by lanezos where it use this fourier series and close it related to a function of and it is defined by a local averaging process according to equation 16 below. whereas, this equation can be modifying as: fourier series is a function of approximation using a finite number of terms and then, the resulting function may oscillate in regions where the actual function is discontinues or in other hand it is called as changes rapidly. this undesirable behaviour can be reduced by using a smoothing procedure described by lanezos where it use this fourier series and close it related to a function of and it is defined by a local averaging process according to equation 16 below. the averaging interval of should be a small fraction of the period of . then, rewrite the equation of with . the functions of and are identical as . hence, also match exactly at any point of x, where varies linearly between importantly, this agreed closely with for a small value of but at the same time in a fourier series which converges more rapidly than the series for . last but not least, it can be defined as: where and for . then, evidently this fourier series of the coefficients of are easily obtained from the function of . when this series for converges slowly, using the same number of the terms in the series for often gives an approximation preferable to that provided by the series of . finally this process is called as smoothing. the averaging interval of should be a small fraction of the period of . then, rewrite the equation of with . the functions of and are identical as . hence, also match exactly at any point of x, where varies linearly between whereas, this equation can be modifying as: fourier series is a function of approximation using a finite number of terms and then, the resulting function may oscillate in regions where the actual function is discontinues or in other hand it is called as changes rapidly. this undesirable behaviour can be reduced by using a smoothing procedure described by lanezos where it use this fourier series and close it related to a function of and it is defined by a local averaging process according to equation 16 below. the averaging interval of should be a small fraction of the period of . then, rewrite the equation of with . the functions of and are identical as . hence, also match exactly at any point of x, where varies linearly between importantly, this agreed closely with for a small value of but at the same time in a fourier series which converges more rapidly than the series for . last but not least, it can be defined as: where and for . then, evidently this fourier series of the coefficients of are easily obtained from the function of . when this series for converges slowly, using the same number of the terms in the series for often gives an approximation preferable to that provided by the series of . finally this process is called as smoothing. importantly, this agreed closely with for a small value of but at the same time in a fourier series which converges more rapidly than the series for . last but not least, it can be defined as: whereas, this equation can be modifying as: fourier series is a function of approximation using a finite number of terms and then, the resulting function may oscillate in regions where the actual function is discontinues or in other hand it is called as changes rapidly. this undesirable behaviour can be reduced by using a smoothing procedure described by lanezos where it use this fourier series and close it related to a function of and it is defined by a local averaging process according to equation 16 below. the averaging interval of should be a small fraction of the period of . then, rewrite the equation of with . the functions of and are identical as . hence, also match exactly at any point of x, where varies linearly between importantly, this agreed closely with for a small value of but at the same time in a fourier series which converges more rapidly than the series for . last but not least, it can be defined as: where and for . then, evidently this fourier series of the coefficients of are easily obtained from the function of . when this series for converges slowly, using the same number of the terms in the series for often gives an approximation preferable to that provided by the series of . finally this process is called as smoothing. issn: 2180-1053 vol. 2 no. 1 january-june 2010 image classification of temperature distribution using fourier series strategy 17 where whereas, this equation can be modifying as: fourier series is a function of approximation using a finite number of terms and then, the resulting function may oscillate in regions where the actual function is discontinues or in other hand it is called as changes rapidly. this undesirable behaviour can be reduced by using a smoothing procedure described by lanezos where it use this fourier series and close it related to a function of and it is defined by a local averaging process according to equation 16 below. the averaging interval of should be a small fraction of the period of . then, rewrite the equation of with . the functions of and are identical as . hence, also match exactly at any point of x, where varies linearly between importantly, this agreed closely with for a small value of but at the same time in a fourier series which converges more rapidly than the series for . last but not least, it can be defined as: where and for . then, evidently this fourier series of the coefficients of are easily obtained from the function of . when this series for converges slowly, using the same number of the terms in the series for often gives an approximation preferable to that provided by the series of . finally this process is called as smoothing. and whereas, this equation can be modifying as: fourier series is a function of approximation using a finite number of terms and then, the resulting function may oscillate in regions where the actual function is discontinues or in other hand it is called as changes rapidly. this undesirable behaviour can be reduced by using a smoothing procedure described by lanezos where it use this fourier series and close it related to a function of and it is defined by a local averaging process according to equation 16 below. the averaging interval of should be a small fraction of the period of . then, rewrite the equation of with . the functions of and are identical as . hence, also match exactly at any point of x, where varies linearly between importantly, this agreed closely with for a small value of but at the same time in a fourier series which converges more rapidly than the series for . last but not least, it can be defined as: where and for . then, evidently this fourier series of the coefficients of are easily obtained from the function of . when this series for converges slowly, using the same number of the terms in the series for often gives an approximation preferable to that provided by the series of . finally this process is called as smoothing. then, evidently this fourier series of the coefficients of are easily obtained from the function of . when this series for converges slowly, using the same number of the terms in the series for often gives an approximation preferable to that provided by the series of . finally this process is called as smoothing. 3.0 fourier series analysis of thin square metal plate t 0 0 0 y = π x = 0 y = 0 x = π figure 1 plate a thin square metal plate has three sides that are held at temperature 0. the other side (the top) is fixed at a temperature above 0. using the boundary conditions: 3.0 fourier series analysis of thin square metal plate figure 1 plate a thin square metal plate has three sides that are held at temperature 0. the other side (the top) is fixed at a temperature above 0. using the boundary conditions: according to the laplace equation, it can be derived as a: let: since, where, t 0 0 0 y = π x = 0 y = 0 x = π according to the laplace equation, it can be derived as a: 3.0 fourier series analysis of thin square metal plate figure 1 plate a thin square metal plate has three sides that are held at temperature 0. the other side (the top) is fixed at a temperature above 0. using the boundary conditions: according to the laplace equation, it can be derived as a: let: since, where, t 0 0 0 y = π x = 0 y = 0 x = π issn: 2180-1053 vol. 2 no. 1 january-june 2010 journal of mechanical engineering and technology 18 let: 3.0 fourier series analysis of thin square metal plate figure 1 plate a thin square metal plate has three sides that are held at temperature 0. the other side (the top) is fixed at a temperature above 0. using the boundary conditions: according to the laplace equation, it can be derived as a: let: since, where, t 0 0 0 y = π x = 0 y = 0 x = π since, where, from equation 20, it can be rewritten as: from equation 20, it can be rewritten as: re-arrange the equation and substitute equation 21 and 22 into . the new equation can be written as: from equation 23, it needs to be divided with fg. therefore, the equation can be expressed as: based on equation 24, equating the equation with where it is a constant value. re-written of equation 24 is: hence, and so, from equation 20, it can be rewritten as: re-arrange the equation and substitute equation 21 and 22 into . the new equation can be written as: from equation 23, it needs to be divided with fg. therefore, the equation can be expressed as: based on equation 24, equating the equation with where it is a constant value. re-written of equation 24 is: hence, and so, re-arrange the equation and substitute equation 21 and 22 into from equation 20, it can be rewritten as: re-arrange the equation and substitute equation 21 and 22 into . the new equation can be written as: from equation 23, it needs to be divided with fg. therefore, the equation can be expressed as: based on equation 24, equating the equation with where it is a constant value. re-written of equation 24 is: hence, and so, the new equation can be written as: from equation 20, it can be rewritten as: re-arrange the equation and substitute equation 21 and 22 into . the new equation can be written as: from equation 23, it needs to be divided with fg. therefore, the equation can be expressed as: based on equation 24, equating the equation with where it is a constant value. re-written of equation 24 is: hence, and so, from equation 20, it can be rewritten as: re-arrange the equation and substitute equation 21 and 22 into . the new equation can be written as: from equation 23, it needs to be divided with fg. therefore, the equation can be expressed as: based on equation 24, equating the equation with where it is a constant value. re-written of equation 24 is: hence, and so, from equation 23, it needs to be divided with fg. therefore, the equation can be expressed as: from equation 20, it can be rewritten as: re-arrange the equation and substitute equation 21 and 22 into . the new equation can be written as: from equation 23, it needs to be divided with fg. therefore, the equation can be expressed as: based on equation 24, equating the equation with where it is a constant value. re-written of equation 24 is: hence, and so, based on equation 24, equating the equation with where it is a constant value. re-written of equation 24 is: from equation 20, it can be rewritten as: re-arrange the equation and substitute equation 21 and 22 into . the new equation can be written as: from equation 23, it needs to be divided with fg. therefore, the equation can be expressed as: based on equation 24, equating the equation with where it is a constant value. re-written of equation 24 is: hence, and so, issn: 2180-1053 vol. 2 no. 1 january-june 2010 image classification of temperature distribution using fourier series strategy 19 hence, from equation 20, it can be rewritten as: re-arrange the equation and substitute equation 21 and 22 into . the new equation can be written as: from equation 23, it needs to be divided with fg. therefore, the equation can be expressed as: based on equation 24, equating the equation with where it is a constant value. re-written of equation 24 is: hence, and so, and so, from equation 20, it can be rewritten as: re-arrange the equation and substitute equation 21 and 22 into . the new equation can be written as: from equation 23, it needs to be divided with fg. therefore, the equation can be expressed as: based on equation 24, equating the equation with where it is a constant value. re-written of equation 24 is: hence, and so, the left and the right side of the boundary conditions can be implied as: it can give as a: hence that, the ordinary differential equations (odes) for g with can represent as a: then, from equation 28, differentiate respect to y by putting boundary condition obtained a = b. therefore from equation 28: from equation 29, using euler’s formula in trigonometry, the equation can be expressed as: multiply both f and g, represent the equation as a: the boundary condition of then is applied at the top side of the thin plate which subsequently the equation 31 can be rewritten again as: the left and the right side of the boundary conditions can be implied as: it can give as a: the left and the right side of the boundary conditions can be implied as: it can give as a: hence that, the ordinary differential equations (odes) for g with can represent as a: then, from equation 28, differentiate respect to y by putting boundary condition obtained a = b. therefore from equation 28: from equation 29, using euler’s formula in trigonometry, the equation can be expressed as: multiply both f and g, represent the equation as a: the boundary condition of then is applied at the top side of the thin plate which subsequently the equation 31 can be rewritten again as: hence that, the left and the right side of the boundary conditions can be implied as: it can give as a: hence that, the ordinary differential equations (odes) for g with can represent as a: then, from equation 28, differentiate respect to y by putting boundary condition obtained a = b. therefore from equation 28: from equation 29, using euler’s formula in trigonometry, the equation can be expressed as: multiply both f and g, represent the equation as a: the boundary condition of then is applied at the top side of the thin plate which subsequently the equation 31 can be rewritten again as: the ordinary differential equations (odes) for g with can represent as a: the left and the right side of the boundary conditions can be implied as: it can give as a: hence that, the ordinary differential equations (odes) for g with can represent as a: then, from equation 28, differentiate respect to y by putting boundary condition obtained a = b. therefore from equation 28: from equation 29, using euler’s formula in trigonometry, the equation can be expressed as: multiply both f and g, represent the equation as a: the boundary condition of then is applied at the top side of the thin plate which subsequently the equation 31 can be rewritten again as: issn: 2180-1053 vol. 2 no. 1 january-june 2010 journal of mechanical engineering and technology 20 then, the left and the right side of the boundary conditions can be implied as: it can give as a: hence that, the ordinary differential equations (odes) for g with can represent as a: then, from equation 28, differentiate respect to y by putting boundary condition obtained a = b. therefore from equation 28: from equation 29, using euler’s formula in trigonometry, the equation can be expressed as: multiply both f and g, represent the equation as a: the boundary condition of then is applied at the top side of the thin plate which subsequently the equation 31 can be rewritten again as: from equation 28, differentiate respect to y by putting boundary condition the left and the right side of the boundary conditions can be implied as: it can give as a: hence that, the ordinary differential equations (odes) for g with can represent as a: then, from equation 28, differentiate respect to y by putting boundary condition obtained a = b. therefore from equation 28: from equation 29, using euler’s formula in trigonometry, the equation can be expressed as: multiply both f and g, represent the equation as a: the boundary condition of then is applied at the top side of the thin plate which subsequently the equation 31 can be rewritten again as: obtained a = b. therefore from equation 28: the left and the right side of the boundary conditions can be implied as: it can give as a: hence that, the ordinary differential equations (odes) for g with can represent as a: then, from equation 28, differentiate respect to y by putting boundary condition obtained a = b. therefore from equation 28: from equation 29, using euler’s formula in trigonometry, the equation can be expressed as: multiply both f and g, represent the equation as a: the boundary condition of then is applied at the top side of the thin plate which subsequently the equation 31 can be rewritten again as: from equation 29, using euler’s formula in trigonometry, the equation can be expressed as: the left and the right side of the boundary conditions can be implied as: it can give as a: hence that, the ordinary differential equations (odes) for g with can represent as a: then, from equation 28, differentiate respect to y by putting boundary condition obtained a = b. therefore from equation 28: from equation 29, using euler’s formula in trigonometry, the equation can be expressed as: multiply both f and g, represent the equation as a: the boundary condition of then is applied at the top side of the thin plate which subsequently the equation 31 can be rewritten again as: multiply both f and g, represent the equation as a: the left and the right side of the boundary conditions can be implied as: it can give as a: hence that, the ordinary differential equations (odes) for g with can represent as a: then, from equation 28, differentiate respect to y by putting boundary condition obtained a = b. therefore from equation 28: from equation 29, using euler’s formula in trigonometry, the equation can be expressed as: multiply both f and g, represent the equation as a: the boundary condition of then is applied at the top side of the thin plate which subsequently the equation 31 can be rewritten again as: the boundary condition of then is applied at the top side of the thin plate which subsequently the equation 31 can be rewritten again as: the final equation can represent as a: then, if n is even number, n = 2, 4, 6, and 8… equation 36 will become as: if n is odd number, n = 1, 3, 5, 7 … therefore, from equation 37 and above, the general equation of the fourier series of this problem can be representing as a: the final equation can represent as a: then, if n is even number, n = 2, 4, 6, and 8… equation 36 will become as: if n is odd number, n = 1, 3, 5, 7 … therefore, from equation 37 and above, the general equation of the fourier series of this problem can be representing as a: the final equation can represent as a: the final equation can represent as a: then, if n is even number, n = 2, 4, 6, and 8… equation 36 will become as: if n is odd number, n = 1, 3, 5, 7 … therefore, from equation 37 and above, the general equation of the fourier series of this problem can be representing as a: the final equation can represent as a: then, if n is even number, n = 2, 4, 6, and 8… equation 36 will become as: if n is odd number, n = 1, 3, 5, 7 … therefore, from equation 37 and above, the general equation of the fourier series of this problem can be representing as a: issn: 2180-1053 vol. 2 no. 1 january-june 2010 image classification of temperature distribution using fourier series strategy 21 then, the final equation can represent as a: then, if n is even number, n = 2, 4, 6, and 8… equation 36 will become as: if n is odd number, n = 1, 3, 5, 7 … therefore, from equation 37 and above, the general equation of the fourier series of this problem can be representing as a: if n is even number, n = 2, 4, 6, and 8… equation 36 will become as: the final equation can represent as a: then, if n is even number, n = 2, 4, 6, and 8… equation 36 will become as: if n is odd number, n = 1, 3, 5, 7 … therefore, from equation 37 and above, the general equation of the fourier series of this problem can be representing as a: if n is odd number, n = 1, 3, 5, 7 … the final equation can represent as a: then, if n is even number, n = 2, 4, 6, and 8… equation 36 will become as: if n is odd number, n = 1, 3, 5, 7 … therefore, from equation 37 and above, the general equation of the fourier series of this problem can be representing as a: therefore, the final equation can represent as a: then, if n is even number, n = 2, 4, 6, and 8… equation 36 will become as: if n is odd number, n = 1, 3, 5, 7 … therefore, from equation 37 and above, the general equation of the fourier series of this problem can be representing as a: the final equation can represent as a: then, if n is even number, n = 2, 4, 6, and 8… equation 36 will become as: if n is odd number, n = 1, 3, 5, 7 … therefore, from equation 37 and above, the general equation of the fourier series of this problem can be representing as a: from equation 37 and above, the general equation of the fourier series of this problem can be representing as a: the final equation can represent as a: then, if n is even number, n = 2, 4, 6, and 8… equation 36 will become as: if n is odd number, n = 1, 3, 5, 7 … therefore, from equation 37 and above, the general equation of the fourier series of this problem can be representing as a: when n = 1 issn: 2180-1053 vol. 2 no. 1 january-june 2010 journal of mechanical engineering and technology 22 4.0 result and discussion un(x , y )=-2t/pi * s um [ (c os nx -1)s in nx s inh ny / n s inh npi ] x y 0 0. 5 1 1. 5 2 2. 5 3 0 0. 5 1 1. 5 2 2. 5 3 0 10 20 30 40 50 60 70 80 90 100 110 figure 2 temperature contour of thin square metal plate x y 0 0. 5 1 1. 5 2 2. 5 3 0 0. 5 1 1. 5 2 2. 5 3 figure 3 isotherms line of thin square metal plate 0 0. 5 1 1. 5 2 2. 5 3 0 0. 5 1 1. 5 2 2. 5 3 x y 20 40 60 80 100 120 figure 4 two dimensional of thin square metal plate issn: 2180-1053 vol. 2 no. 1 january-june 2010 image classification of temperature distribution using fourier series strategy 23 0 1 2 3 4 0 1 2 3 4 0 0. 2 0. 4 0. 6 0. 8 1 xy u (x ,y ) 0. 1 0. 2 0. 3 0. 4 0. 5 0. 6 0. 7 0. 8 0. 9 figure 5 three dimensional of thin square metal plate figure 2 shows the temperature contour of thin square metal plate. it can be seen that heat is transfer from high temperature to low temperature. the direction of heat transfer is due to temperature setting at the boundary condition. left and right edges are set as cold (t=0) and the bottom edge set as an adiabatic boundary condition and the top edge is set at temperature t. isotherm lines was plotted in figure 3, it shows isotherm move out from the bottom edge. this is due the adiabatic boundary condition. at the left and right edges of thin plate, the isotherm lines keep parallel with these edges. this occurs due to left and right edges of thin plate was set as a cold (t=0). figure 4 and figure 5 show the two and three dimensional of temperature distribution of thin square metal plate. figure 5 clearly shows the steady state heat transfer conditions. at steady state condition, heat transfer from the top edge of thin square metal plate is equal to heat transfer out from bottom edge of thin square metal plate. 5.0 conclusion fourier series analysis is also known as a trigonometric series analysis. this series has a potential to solve the analytical solution on temperature distribution of plate. temperature distribution can be distributed independently according to the adiabatic phenomenon. this phenomenon is based on the time domain. issn: 2180-1053 vol. 2 no. 1 january-june 2010 journal of mechanical engineering and technology 24 6.0 acknowledgement the author would like to express appreciation and gratitude to faculty of mechanical engineering, universiti teknikal malaysia melaka and universiti teknologi malaysia for the technical supports. 7.0 references erwin kreysig. 2006. advanced engineering mathematics. 9th edition. wiley international edition. howard b. wilson, louis h. turcotte and david halpen. 2003. advanced mathematics and mechanics and application using matlab. 3rd edition. chapman and hall. maksimovich, v. n and tsybul’’skii, o. a. 2004. application of fourierseries methods and integral equations for solving nonstationary nonaxisymmetric heat conduction problems for bodies of revolution journal of engineering physics and thermophysics volume 68, number 6 / november, 1995 pp828-834 maria t. ibanez and h. power, 2000. an efficient direct bem numerical scheme for heat transfer problems using fourier series, international journal of numerical methods for heat & fluid flow vol. 10/7 issn 0961-5539 pp 687 720 s. bochner, 1935. summation of derived fourier series. an application to fourier expansions on compact lie groups. annals of mathematics. vol 37, no. 2 taddeus h. black. 2008. derivation of applied mathematics. revised. debian project. buckling load analysis of cracked curved beams using differential quadrature element method m. zare1, a. asnafi2*1 1department of mechanical engineering, shahid chamran university, ahvaz, 61355, iran. 2hydro-aeronautical research center, shiraz university, shiraz, 71348-51154, iran. abstract this paper investigated the buckling load of a cracked curved beam subjected to external excitations considering the effects of shear deformations and geometric nonlinearity due to large deformations. the governing nonlinear equations of motion were derived. the analysis in stationary case was developed for each half of the beam and then the differential quadrature element method (dqem) was used to discretize and solve the problem. the resulting nonlinear system of equation was analyzed using continuity conditions between the beam segments and an arc length strategy. to verify the validity of the proposed method, the beam was modeled using the finite element method. the agreement between the results showed the accuracy of the proposed method in prediction the buckling load of the beam. finally, the effect of crack parameters (depth and location) on the buckling load was investigated. as the results showed, the crack depth and buckling load related conversely. furthermore, the closer the crack to the midpoint, the less load was required to make the beam undergo buckling. keywords: differential quadrature element method (dqem), curved beam vibration, buckling load, arc length method, non-linear analysis. 1.0 introduction curved beams have been applied in a variety of industries to be used as mechanical devices, building arches and so on. obviously, over time, such structures can be damaged and one of the most prevalent damages is crack. moreover, the stability of structures is of high significance; so, it is a great idea to have a deeper insight into the stability analysis of cracked structural elements. much research has been done about the stability of beams as well as their damage phenomena. bradford et al. (bradford, 2002) studied the in-plane elastic stability of arches under a central concentrated load analytically. the stability of a cracked beam exited by a follower load was done by wang (wang, 2004). in that research, the critical load was obtained on the basis of the variation of resonant frequencies. in-soo et.al (son, 2007) studied the natural frequencies of a cracked beam exposed to a follower force. in their study the buckling load of the beam was also calculated. the static stability analysis of a uniform column with multiple cracks has been done by caddemi et.al (caddemi, *corresponding author. email: asnafi@shirazu.ac.ir mailto:email@address.ac.ir 2013). in that research the buckling modes as well as the corresponding buckling loads were presented. in another study done by nikpour (nikpour, 1990), the buckling phenomenon of a beam with an edged-notched composite was analyzed. in the mentioned study, the local compliance due to the crack was considered as a function of the crack-tip stress intensity factors and the elastic properties of material. the buckling and post buckling analysis of curved beams under distributed, concentrated and thermal loads were done by eslami (eslami, 2018). in that study, different boundary conditions and loads were studied and then the pre and post buckling loads of rings were obtained and discussed. attard (attard, 1986) presented two new finite element formulations for obtaining the lateral buckling load of elastic beams under static loads. torabi et al. (torabi, 2014) studied the free vibration of a timoshenko beam with multiple cracks using differential quadrature element method (dqem). in that study, they revealed that how crack parameters influenced the natural frequencies. in this study the buckling load of a general cracked curved beam with a radial concentrated force at middle point of the beam is investigated. the problem is solved considering the static analysis based on the differential quadrature element and arc length methods and finally, a formula is proposed for the buckling load with respect the radial displacement of the mid-point of the beam. also, the effect of crack parameters, depth and location, on the buckling load is studied. 2.0 equation of motion of the cracked curved beam the equations of motion for a curved beam's post-buckled state, taking into account the effects of shear deformation and rotary inertia, as well as, the extension of the neutral axis, can be written as (nikpour, 1990): 1 ( ) q n au r s s           (1) 1 ( ) n q aw r s s          (2) ( 1) m q n n q a s kag ea         (3) (( 1) cos( ) ( )sin( ) 1) n u w w u ea s r s r             (4) ( ( 1)sin( ) ( ) cos( )) q u w w u kag s r s r             (5) m ei s    (6) where dot means the derivative with respect to time. as shown in fig. 1, which represents an element of a curved beam, w, u and φ denote the radial and tangential displacements and the angle of rotation. m, n and q represent the bending moment, normal and shear distributed forces respectively. moreover, a, i, γ, g, e and k are the structural properties of the beam, which denote area and moment of inertia of the cross section, mass density per unit volume of the beam material, shear and young's modulus, and shear factor of the cross section, respectively. in this figure β is the angular location of the crack. figure 1. load and displacement components of a curved beam element (karaagac, 2011) in order to include the effect of crack, a rotational spring with stiffness (k0) is assumed as (cerri, 2008): 3 3 0 1 1 , ( ) , ( ) 2 ( ) / 12 12 d d cd ei ei k ei e b h h ei e bh ei ei h        (7) the equations of motion are solved by dqe method which is a new numerical method for rapidly solving linear and nonlinear differential equations (appendix a). at equilibrium state, the terms containing time evolutions in eqs. (1-6) are eliminated. then, by applying dq discretization to the equations of motion at an interior node mi, in an element i, the following equilibrium equations are obtained: ( ) ( )1 ( ) ( ) 0 i i i i i e ei m m em q n r s s        (8) ( ) ( )1 ( ) ( ) 0 i i i i i e ei m m em n q r s s        (9) ( ) ( ) ( ) ( ) ( ) ( 1) 0 i i i i i i i i e e ei im m m e em m m q n n q s kag ea       (10) ( ) ( ) ( ) ( ) ( ) (( 1) cos(( ) ) ( )sin(( ) ) 1) 0 i i i i i i i i i i i i e e e e ei im m m m m e em m n u w w u ea s r s r              (11) ( ) ( ) ( ) ( ) ( ) ( ( 1)sin(( ) ) ( ) cos(( ) )) 0 i i i i i i i i i i i i e e e e ei im m m m m e em m q u w w u kag s r s r              (12) ( ) ( ) 0 i i i ei i m em m ei s     (13) now, the continuity conditions are applied to each interface of the discretized segments of the beam. the continuity conditions make some relations between the radial and tangential displacements, the angular rotation, the normal and shear forces and the bending moment of adjacent elements. the radial and tangential displacements and the angular rotation continuity conditions at the inter-element boundary of two adjacent elements i and i + 1, except for the crown of the beam and the cracked section, are expressed as: 1 1 1 1 1 1 , ,i i i i i i i i i n n n w w u u         (14) in addition, the normal and shear forces and the bending moment continuity conditions at the inter-element boundary of two adjacent elements i and i + 1 can be expressed, respectively, as: 1 1 1 1 1 1 1 11 1 1 1 1 1 (( 1) cos( ) ( )sin( ) 1) (( 1) cos( ) ( )sin( ) 1) i i i i i i i i i i i i i in n n n n n i i i i i i i i e a e a u w w u s r s r u w w u s r s r                                (15) 1 1 1 1 1 1 1 1 11 1 1 1 1 1 ( ( 1)sin( ) ( ) cos( )) ( ( 1)sin( ) ( ) cos( )) i i i i i i i i i i i i i i in n n n n n i i i i i i i i i u w w u s r s r u w w u k a g s r s k a r g                                 (16) 1 1 1 1 1 i i i i i i i in i i n e i e i s s           (17) the continuity conditions at the crown of the arch in the equilibrium state are: 1 1 1 1 1 0 1 0 1 1 1 1, , 2 , , ,i i i i i i i i i i i n n n i ii i n n u u m i s u u w w w w s s s                       (18) where, w0 is the radial displacement of the middle point of the beam. also, the continuity conditions at the cracked section is: 1 1 1 1 1 0 1 , , ( )i i i i i i i i i cn n n w w u u k m         (19) where mc is the bending moment at the cracked section. finally, the boundary conditions for a beam clamped at both ends are: 1 1 1 1 1 1 0 0 0 w u   (20) 0 0 0i i i m m m n n n w u      (21) to compute the buckling load, the radial displacement of the crown of the beam (w0) is used as the input of the arc length strategy to obtain the concentrated load in each half of the arch. in this manner, by investigation the normal force n, shear force q, and the angular rotation  at the arch crown, the concentrated load in each half of the beam is calculated as (zhu, 2014): now, by applying the continuity conditions at the crown point of the beam, the value of buckling force is obtained as: 3.0 results and discussion without any loss of generality, using eqs. (22 & 23), the magnitude of the concentrated load versus the radial displacement of the middle point (w0), for a clamped-clamped beam with the properties of table.1 was obtained and plotted in fig. 2. it is to be noted that to ensure about the accuracy of the proposed method, a finite element simulation was also done and the results obtained throughout above methods were compared. 1 1 1 1 1 2 1 1 1 1 sin( ) cos( ), sin( ) cos( ) i i i i i i i i n n n n f n q f n q            (22) 1 2 , 2 c m p f f i   (22) table 1. mechanical properties of the curved beam property notation value radius of the beam axis r 83 (cm) opening angle of the beam θ 40 (deg) height of the cross section h 0.5 (cm) base of the cross section b 2 (cm) young’s modulus e 11 (gpa) poisson’s ratio v 0.3 density γ 7800 (kg/m3) a good agreement between the results obtained by our model and those obtained through fe modeling (ansys) was seen (see fig. 2) which confirmed the accuracy of the proposed method. figure 2. variation of the concentrated load versus the crown radial displacement note also that in the fe simulation, the 3-node element beam189 was used to mesh and analyze the nonlinear buckling behavior of the beam. in the analysis, 100 elements were used to achieve more accurate results. see the finite element model along with the structured meshes in fig.3. 0 100 200 300 0 0.02 0.04 0.06 0.08 0.1 p (n ) w0(m) fem dqem (a) (b) figure 3. a: the finite element model of the beam. b: assumed element for meshing the model 3.1 the effect of crack depth on the buckling load in fig. 4, the changing of the buckling load for the beam introduced in table. 1 by a variation in the crack depth was shown. in this figure the relative location of the crack (the proportion of the angular location of the crack to the opening angle of the beam) is 0.17. note that in this figure the relative depth is defined as the crack depth to the height of the cross section. figure 4. changing of the buckling load versus the midpoint displacement for different crack depths 0 100 200 300 0 0.02 0.04 0.06 0.08 p (n ) w0 (m) relative depth=.1 relative depth=.3 relative depth=.5 as the figure shows, increasing the crack depth make a decrease in the buckling load which is due to the decrease in the flexural stiffness of the beam. 3.2 the effect of crack location on the buckling load in fig.5, it is shown how the buckling load changes with the variation of the crack location for the aforementioned beam with relative crack depth of 0.5. it represents that as the crack location gets closer to the middle point of the beam, the buckling load decreases. this is mainly because the bending moment increases by moving toward the midpoint of the beam. also, the crack effect is related to the bending moment as eq.(19) shows; so, the buckling load decreases. figure 5. changing of the buckling load versus the midpoint displacement for different crack locations 4.0 conclusion the differential quadrature element method along with an arc length strategy were used to obtain the buckling load of a cracked curved beam in this article. using the equation of motion in stationary case, and applying the continuity conditions between adjacent segments of the beam, a formula for buckling load was proposed. the proposed method was firstly validated by a finite element simulation. after that, the effects of the crack depth and location on the buckling load were studied. it was shown that the buckling load became smaller as the depth of the crack increased or the crack location approached to the beam crown. 5.0 conflict of interest 0 100 200 300 0 0.02 0.04 0.06 0.08 0.1 p (n ) w0 (m) relative location=0.25 relative location=.083 relative location=0.42 we declare that we have no conflict of interest. 6.0 references attard, m. m. (1986). lateral buckling analysis of beams by the fem. computers & structures, 23(2), 217-231. bradford, m. a. (2002). in-plane elastic stability of arches under a central concentrated load. journal of engineering mechanics, 128(7), 710-719. caddemi, s. c. (2013). the influence of multiple cracks on tensile and compressive buckling of shear deformable beams. international journal of solids and structures, 50(20-21), 3166-3183. cerri, m. n. (2008). vibration and damage detection in undamaged and cracked circular arches: experimental and analytical results. journal of sound and vibration, 314(1-2), 83-94. eslami, m. r. (2018). buckling and post-buckling of curved beams and rings. in buckling and postbuckling of beams, plates, and shells (pp. 11-188). springer. karaagac, c. o. (2011). crack effects on the in-plane static and dynamic stabilities of a curved beam with an edge crack. journal of sound and vibration, 330(8), 17181736. nikpour, k. (1990). buckling of cracked composite columns. international journal of solids and structures, 26(12), 1371-1386. son, i. s. (2007). stability analysis of cracked cantilever beam with tip mass and follower force. transactions of the korean society for noise and vibration engineering, 17(7), 605-610. torabi, k. a. (2014). a dqem for transverse vibration analysis of multiple cracked nonuniform timoshenko beams with general boundary conditions. computers & mathematics with applications, 6(3), 527-541. wang, q. (2004). a comprehensive stability analysis of a cracked beam subjected to follower compression. international journal of solids and structures, 41(18-19), 4875-4888. zhu, j. a. (2014). in-plane nonlinear buckling of circular arches including shear deformations. archive of applied mechanics, 84(12), 1841-1860. appendix a. differential quadrature element method the dqem is a new numerical method for rapidly solving linear and nonlinear differential equations. the dqem is based on the dq method, an approximate method for expressing partial derivatives of a function at a point located in the domain of the function, as the weighted linear sum of the values of the variable function at all the defined precision points in the derivation direction. eq. (a.1) is the mathematical representation of the dq expansion: where f is the function, n is the number of precision points, xi is the precision associated with the i-th point of the function domain, and represents the weighting coefficients used for finding the first derivative of the function at the i-th precision point of the function domain represented as xi above. in the dqem, the studied structure is divided into several elements. then, the continuity conditions are applied on the inter-element boundary of two adjacent elements and the boundary conditions of the beam as well as the governing equations on each element, using the differential quadrature method. according to eq. (a.1), two important factors in the dq method include: calculation of the dq weighting coefficients, and selecting the precision points. the lagrangian functions were used to compute the weighted coefficients, and the gauss–lobatto chebyshev polynomial was used for selecting the precision points. (1) 1 | , 1, ,3, , 2 i n x x ij j j df c f i j n dx      (a.1) preparation of papers in a two column model paper format journal of mechanical engineering and technology *corresponding author. email: shamanuar@utem.edu.my issn 2180-1053 vol. 12 no.1 june-december 2020 53 development of a practical motorcycle towing device n. a. a. nik hak1, s. a. shamsudin*2, m. a, abdullah2, s. zakaria3, a. rivai4 1 skid systems engineering sdn. bhd., no.21, jalan utama 2/18, taman perindustrian puchong utama, 47100 puchong, selangor darul ehsan, malaysia. 2 centre for advanced research on energy (care), fakulti kejuruteraan mekanikal (faculty of mechanical engineering), universiti teknikal malaysia melaka, 75450 ayer keroh, melaka, malaysia. 3 centre for robotics and industrial automation (ceria), fakulti teknologi kejuruteraan elektical dan elektronik, universiti teknikal malaysia melaka, 75450 ayer keroh, melaka, malaysia. 4 sekolah tinggi teknologi bandung, jl. soekarno hatta no.378 bandung 40235, indonesia. abstract motorcycle to motorcycle towing device is not readily available in the malaysian market. most of the motorcycle towing service uses a pick-up truck to pick up the motorcycle. such towing or on-site service fee can be too expensive anyways. in addition, some of the towing process is done only using a foot to propel the towed motorcycle. the aim of this project to produce a product that can be commercialize and solve the problem. in this project, a prototype is produced after the fabrication process. the prototype seemed to function. the length of the sling successfully retracted for storing, thus the sling retractor is working accordingly. keywords: motorcycle, towing, safety, design 1.0 introduction towing is an act of pulling or dragging a driven object that fastened behind another driver object and by coupling these two objects together it will keep these objects together while in motion. source of towing can be range from the biggest aircraft to a cow. these may be coupled by a chain, rope, bar, hitch, three-point fifth wheel, coupling, drawbar, tow bar or other means of keeping the objects in motion. the first vehicle that use the application of towing is tow truck which is invented by german automotive pioneer gottlieb daimler who also the inventor of the first world gas-powered motorcycle. towing varies widely in scale and type, on land, water and in the air. the most common form of towing is the transport of disabled vehicle by a tow truck. other than that, are motorcycle-trailer combination and cargo or leisure vehicles coupled via trailer-hitches to smaller trucks and bike as shown in figure 1. journal of mechanical engineering and technology issn 2180-1053 vol. 12 no.1 june-december 2020 54 nowadays, the application of towing is used in many types of field and industry especially vehicle recovery industry. vehicle recovery is towing assistance given to any disabled or broken-down vehicle to place of interest with the help of recovery vehicle like a tow truck. recovery operators are the people who undertake the recovery service. early motorists were often capable of carrying out minor repairs themselves but as automobiles became more complicated, this became more difficult to carry out successfully. thus, towing service is needed to transport the broken vehicle to auto repair shop. this project was ventured to redesign the device that used for towing between two motorcycles. this project will involve redesigning, analyzing and testing process for the improved design of the device for towing motorcycle which is required to fulfil the objective of this project. this device may help users to tow motorcycle by using another motorcycle especially in emergency. as the title is development of a towing device between two motorcycles, the challenge is to redesign and produce an improve device with better mechanism and portability. this mechanism would help people especially for motorcycle user that has the probability to face towing situation of their motorcycle. commonly, motorcyclist use the vehicle recovery service that use powered vehicle such as truck or trailer to tow their motorcycle which will burden them with the expensive service charge of the vehicle recovery service. other than that, there is other method for towing motorcycle, but it is not suitable to perform and can cause harm not only to motorcyclist but also to other road users. having finished product testing and analyze process, conclusions and recommendations can be made. the conclusion of this project is based on research objectives. it is considered successful if all objectives are achieved. the proposal would involve improvements that can be made for the studies that have been done. figure 1. trailer-hitcher for motorcycles (tarakaner, 2014) a towing apparatus for towing a two-wheeled vehicle, for example, a bike behind another vehicle, for example, a car. the apparatus incorporates a wheel bearer distinctly appended to the back of the towing vehicle by methods for a couple of keyed extending sleeves, one journal of mechanical engineering and technology issn 2180-1053 vol. 12 no.1 june-december 2020 55 of which stays joined to the towing vehicle. the bearer incorporates a support into which the significant controlling wheel of the towed vehicle rides. this apparatus needs two persons to connect the apparatus with the motorcycle. this apparatus came with optional cradles. there are single-cradle, two-cradle and also three (fred, 1969). motorcyclist always have problem towing their motorcycle if their vehicle is broken down by the roadside or on the highway. they either call the highway operator, recovery operator or tow their motorcycle with the help of another motorcyclist. the problem when calling the highway operator or the recovery operator is an expensive service charge. most of them will avoid calling those operators. they rather ask help from another motorcyclist to tow their motorcycle, but it may harm the motorcyclist and the other road users. it is harmful because the towing process involve physical help from human by using the human leg to push the broken motorcycle with the present of the motorcyclist controlling their driven motorcycle as shown in figure 2. this push is extremely dangerous as it neglects the safety of both motorcyclists. using a motorcycle as towing vehicle with proper attachment as a towing device can help to improve the method of towing motorcycle which is more affordable and efficient to the road user. (a) (b) journal of mechanical engineering and technology issn 2180-1053 vol. 12 no.1 june-december 2020 56 figure 2. using a foot in pushing another bike (a) in malaysia and (b) vietnam. a motorcycle towing device incorporating a stage part with a front end for connection to the trailer hitch of a land vehicle. an upright part is oppositely joined to the backside of the stage part. a lift is specifically raised and brought down upon the upright part by methods for a jackscrew situated on the stage part. the lift has a head tube and a couple of turn orientation situated at the best and base of the head tube. the head tube conveys a rotate part. the rotate part has a carriage plate and a couple of turn arms expanding forwardly from the best and base thereof for significant engagement with the rotate direction. a couple of wheel engagement arms, for supporting a motorcycle wheel, expand rearwardly from the carriage plate (cataldo, 2001). this device consists of five main parts which are ball joint, universal coupling, connecting plate, connecting shaft and body frame. the body frame is to connect with both towing and towed motorcycle. the ball joint and universal coupling is to ensure the connection is flexible. the problem with this device is the size is very big and it is limited to only 110cc type of motorcycles only as shown in figure 3 (hasif, 2015). figure 3. towing device between two motorcycles (hasif, 2015) a quick detachable bumper hitch for towing a motorcycle that is manually operated to raise and bring down the front wheel of the motorcycle to keep it above the road surface on which the raise wheel of the cruiser is being bolstered. this device is particularly directed to a sturdy, lightweight bumper hitch for towing motorcycle. a hitch support portion which is detachably mounted on the vehicle's rear bumper and a rearwardly extending tow-bar portion detachably secured in wheel straddling relationship to the depending sides of the motorcycle's steering fork (hancock, 1978). the towing sling is one of the compact towing gadgets. this device is an easy to use towing link. it is an emergency towing device for a motorcycle or other vehicle for moving or towing the motorcycle from a one place to another. the compact normal for this gadget is reasonable for crisis reason while doing motorcycle activities, for example, cruiser escort or rally following exercises (gertler, 1975). journal of mechanical engineering and technology issn 2180-1053 vol. 12 no.1 june-december 2020 57 mild steel is a type of carbon steel with a low measure of carbon – it is also known as "low carbon steel." the measure of carbon ordinarily found in mild steel is 0.05% to 0.25% by weight, while higher carbon steels are regularly portrayed as having a carbon content from 0.30% to 2.0%. if any more carbon than that is included, the steel would be considered as cast iron. less carbon means that mild steel is typically more ductile, machinable, and weldable than high carbon steel. an advantage of the high carbon content which is championed by numerous fabricators is ductility, which makes mild steel greatly simple to cut, penetrate and weld to suit any extend. not just a perfect decision of material for building move enclosures and edges, it is additionally hard sufficiently wearing to go about as an extremely viable edging material (marc, 2013). the absence of alloying components, for example, those found in stainless steels implies that the iron in mild steel is liable to oxidation (rust) if not legitimately covered. be that as it may, the immaterial measure of alloying components likewise causes mild steel to be generally reasonable when contrasted and different steels. it is the affordability, weldability, and machinability that make it such a popular choice of steel for consumers. nylon is a sleek thermoplastic that is high strength, very durable and it is also elastic. since it is a plastic, it is exceptionally impervious from such common nasties as molds, bugs, and growths. it is also waterproof and quick drying because water particles cannot easily penetrate the outer surface. 2.0 method the conceptual designs that have been draw will be pick and go a thorough inspecting process. at this stage a portion of the ideas will be exhibited and will be assessed in view of the criteria previously moving to the following procedure. the preliminary design will be choosing from the concepts using pugh concept selection method. the chosen design which called the preliminary idea will be proceed to the following stage which is drawing the model by utilizing catia. analysis will be made based on the drawing. if the result of the analysis is not as expected it turn to be it will go back to conceptual design and if succeed a model of prototype will be made. testing of model is should have been led to ensure the model working admirably and safe to utilize. finally, a final modification will be made based on the testing on the prototype model result. the final design will be fabricated, analyzed and tested. all the findings and results of this project will be hold forth in the discussion part. change that can be made on the towing device is expressed with the purpose of further research. conceptual design is the principal phase of the product design process, where illustrations and different outlines or models are utilized. this gives some initial ideas and sketches for easy communication with other people involved in this project that are based of many disciplines in engineering as explained in many sources like (myszka, 2012), (mills, 2016), (hanson, 2017), (hunt, 1978), (ham, 1958) and (dicker jr. et al., 2016). there are a couple of things to be considered to design a product for example conceptual design, morphological chart, concept selection and preliminary design. journal of mechanical engineering and technology issn 2180-1053 vol. 12 no.1 june-december 2020 58 a morphological chart is a visual approach to catch the vital item usefulness and investigate elective means and mixes of accomplishing that usefulness. for every component of item work, there might be various conceivable arrangements. the chart empowers these answers for be communicated and gives a structure to thinking about option blends. this can empower the early thought of the item 'design' through the age and thought of various mixes of 'sub-arrangements' that have not already been distinguished. utilized fittingly, it can energize a client driven way to deal with the age of potential arrangement. there are three stages of this method. right off the bat is the list the portrayal of the subfunction of the towing device that needs to be accomplished. the second step is to produce however many ideas as could be expected under the circumstances for each subwork. in conclusion consolidate new ideas to the idea of people who meet every practical prerequisite. morphological chart for towing gadget between two bikes is appeared underneath in table 1. there are two sub-function leaned to produce the possibilities of concept design. there are couple of discretionary ideas that identified with the sub-function that can be utilized as a part of request to get the best conceptual design. table 1. morphological chart for towing device (nik hak, 2018) sub-function concept a b c connection between motorcycles rigid shaft ear hook sling tow joints coupling hook shackles concept selection is the way toward assessing concepts as for client needs and other criteria, looking at the relative qualities and shortcomings of the concepts, and choosing at least one concepts for advance research, testing, or improvement. in this stage, the assessed idea from morphological chart with the best characteristic will be pick. to make the selection easier, the concept design is visualized by sketching the concepts in order to give a clearer picture of the concept design. in view of the morphological chart, there are three ideas that can be produced. all the idea is the mix of the characteristic from morphological chart. toward the finish of this stage, the best idea determination will be pick before continuing to detail design. table 2 explains the general characteristics of the product that would form the product specifications. detailed design is where the design is refined, and plans, specifications and estimates are made. detailed design can incorporate yields, for example, 2d and 3d models, cost develop gauges, acquisition designs and so on. this stage is the place the full cost of the venture is distinguished by and large. detailed design is such a crucial need to makers that it exists at the convergence of numerous item improvement forms. detail design is the last design activity to finish design process before advancement starts. journal of mechanical engineering and technology issn 2180-1053 vol. 12 no.1 june-december 2020 59 at the point when a project undertaking fabrication process, detail drawing which incorporates all the data about item or model is required as reference when led fabrication process. other than that, detail drawing is required for drawing documentation for this project to give the clear picture of this product. typically, the illustrations that are required in a project are isometric views, orthographic projections and solid models. table 2. characteristics between conceptual designs (nik hak, 2018) characteristics design 1 design 2 design 3 performance this design has a flexible joint and highly portable. quick releasing technology to connect with the towed motorcycle front tire bolt. it has a very high tensile strength and a very lightweight device. economy low manufacturing and material cost high manufacturing and low material cost low manufacturing and material cost target production cost rm 70 rm 100 rm60 service life 1 year 1 year 3 years size width: 0.4m height: 0.2m length: 1.2m width: 0.4m height: 0.2m length: 1.2m width: 0.04m thickness: 0.005m length: 1.2m weight 5 kg 5 kg 2 kg material mild steel mild steel nylon ergonomics only compatible with certain motorcycle only compatible with certain motorcycle compatible with most motorcycle safety a meter of distance between the motorcycle a meter of distance between the motorcycle a meter of distance between the motorcycle design time 0.5 weeks 1 weeks 0.5 weeks appearance simple design complex design simple design journal of mechanical engineering and technology issn 2180-1053 vol. 12 no.1 june-december 2020 60 there are numerous techniques for producing drawing has revolt throughout the years. prior to the innovation were developed, specialized illustration is utilized by engineers to produce drawing for some reasons. the aptitude of specialized illustration is profoundly required as the illustration is physically draft by the engineer and designer. as of late, pcs for all intents and purposes kill the specialized illustration as there are various pc helped outline programming, for example, autocad, catia, solidworks and others. for the detail design of the towing device, catia software is utilized to create the illustration that required for this project. the drawing incorporates all parts of the model drawing and assembly drawing of the model. catia (an acronym of computer aided three-dimensional interactive application) is a multi-platform software suite for computer-aided design (cad), computer-aided manufacturing (cam), computer-aided engineering (cae), product lifecycle management (plm) and 3d, created by the frenchbased corporation dassault systèmes. catia is the main item advancement answer for all assembling associations, from oems, through their supply chains, to little free makers. the scope of catia abilities enables it to be connected in a wide assortment of businesses, for example, aviation, car, modern hardware, electrical, gadgets, shipbuilding, plant outline, and purchaser merchandise, including plan for such different items as gems and apparel. catia is used to create 3d drawing for this project. catia empower the production of 3d sections where it conquers the answer for shape, styling, expelling, alter and numerous different functions. strong demonstrating with this product gives clearer photo of the first idea that has been chosen. strong works additionally empower to produce orthographic projection from strong displaying that will give diverse perspective of the of the illustration protest, for example, top plane, side plane and front plane. 4.0 results and discussion through the use of screening and scoring selection processes, the third conceptual design was chosen as preliminary design and the design was modified to make it safer to use. this conceptual drawing is then drawn using computer aided design software. the design that was generated from the cad software consist of assembly of the device and exploded view of the device. figure 4 displays the assembly drawing of the improved conceptual design that have been generated using catia software. the general dimensioning of the device also has been specified. the part drawings are assembled in the software to form the drawing. the exploded view will graphically explain what the parts in the device are. next, figure 5 shows the exploded view of the device. the parts are labeled and named in the bill of materials table. this drawing also provide checklist for the fabrication process. table 3 further describes each part in this towing system. journal of mechanical engineering and technology issn 2180-1053 vol. 12 no.1 june-december 2020 61 figure 4. assembly drawing for motorcycle towing device figure 5. exploded view of the device journal of mechanical engineering and technology issn 2180-1053 vol. 12 no.1 june-december 2020 62 table 3. parts and description parts descriptions case • to protect the chassis and the moving parts of the device • aesthetic value chassis • as the body frame of the device • to hold the rod in place sling • as the connection between the towed motorcycle to the device • maximum length for the sling is 0.65m nylon sling tow • to connect the device with the towing bike • 0.3m in length female buckle • as a locking mechanism to lock the male buckle male buckle • to lock into the female buckle and secure the connection while towing process torsional spring • to help the process of retraction • it stores potential energy • connected with the rod rod • to hold and to store the sling after the design is chosen and generated in the software. an experiment has been conducted to determine the towing force between the motorcycle from stationary to move on normal road. elevated roads up to 25 slope, single person towing (a rider on the troubled motorcycle) and also with a pillion rider. the motorcycles that have been used for this experiment were yamaha lc 135 for the towing and honda wave for the towed bike, with the motorcycle wet weights of 110 kg and 105 kg, respectively. the data has been recorded. the rod which is the crucial part in securing the webbing with the chassis was run through generative structural analysis using catia using the data in the experiment. the pull force needed are reported in table 4 below. the experiment was conducted to determine the towing force. spring balance was used as a tool to measure the force taken by the towed bike to start moving. the spring balance was tied in between the sling connection. after that, the towing motorcycle will make sure the sling is tense between the bike. the towing motorcycle then will move forward as it slowly pulling the towed bike, the reading is then recorded. each type of test ran five times to make sure less risk of error. table 4. experiment pull force results test average towing force (n) single person passenger (pillion rider) normal road 38 42 25-degree elevated road 40 45 generative structural analysis is an analysis that provided in catia software. this analysis will simulate the force that acted on the part and generate translational journal of mechanical engineering and technology issn 2180-1053 vol. 12 no.1 june-december 2020 63 displacement and von misses stress. the force that acted on the rod is 25 n and the force that acted on each rod openings on the chassis is 12.5 n. figure 6 show the distribution of stress of some parts under the load. the displacement is caused by the pulling sling that attached to the rod during the towing process. the maximum displacement (not shown) of the rod under the load is 0.000118 mm and for the chassis is 4.1677 mm. von mises stress result also produced in this structural analysis. stress is produced on the rod when the load is pulled along the rod and also the stress affecting the chassis. the maximum stress is marked with red and the minimum is blue. the maximum von misses stress on the rod under the load is 579660 pa and the minimum is 20335.6 pa and the maximum stress at the rod openings is 267099 pa then for the minimum is 6.487 mpa. these values are relatively low and further in-depth tests or simulation would be done to verify them. figure 6. von mises stress distribution on (a) rod pin and (b) chassis. 5.0 on-road testing the prototype is tested on real situation at a maintained speed of between 30 to 40 km/h. (a) (b) journal of mechanical engineering and technology issn 2180-1053 vol. 12 no.1 june-december 2020 64 two motorcycles are needed for this test, the towed motorcycle is honda wave 125 and towing motorcycle is yamaha lc 135. the prototype is assembled and ready for the test. the test was observed, and the important data has been collected. the towing process was successful, and the prototype worked accordingly. furthermore, sling retractor did pull the sling into the chassis to keep the connection away from the front wheel of towed motorcycle. thus, the safety of both bikers and bike are well kept. the result of this project has shown that the conceptual design has been successfully generated using cad software, catia. besides that, analysis also shown a good result in proceeding to fabrication process. the factor of safety is more than 1, thus making the prototype safe and could be fabricate. in addition, fabrication process with the guidance of the technical drawing was completed. plus, a case that made from galvanized steel 0.5 mm was made to protect both sling retractors. moreover, the prototype keeps the connection away from the risk of tangling and successfully towed the motorcycle, but it tends to jerk the towed motorcycle when the webbing is retract then suddenly pulled to its maximum length. this jerking will affect the sling retractor rod integrity and if the jerking is too high, it might cause the safety of the towed biker. figure 7 show the application of the towing device during the test. figure 7. application of the device on both vehicles for road test. (a) (b) (c) journal of mechanical engineering and technology issn 2180-1053 vol. 12 no.1 june-december 2020 65 6.0 conclusion a study was done to redesign a towing device for use between two motorcycles. some concepts were generated by following morphological chart, then the developed concept was selected using a derivative of pugh concept selection method. then, data and result are achieved. in addition, this section helped me in improving the selected concept and will also help improve the design as the design description without disobeying the engineering characteristic before it proceeds to detailed design. next, the detailed design is generated using catia. finally, the prototype is tested in the real situation. the prototype solved the problems of previous device which is being rigid and big in size. 7.0 acknowledgement the authors like to thank the universiti teknikal malaysia melaka (utem) and the centre for advanced research on energy (care) in supporting this work. 8.0 references cataldo, r. w. (2001). motorcycle towing device, u.s patent 6,244,813, issued june 12, 2001. uspto, alexandria, virginia, usa. dicker jr., j. j., pennock, g. r., & shigley, j. e. (2016). theory of machines and mechanisms (5th ed.). new york: oxford. tarakaner. (2014). stainless steel caravan calypso towed by honda st 1100 pan european in spain. accessed from https://commons.wikimedia.org/wiki/file:motorcycle_caravan_trailer.jpg gertler, b. h. (1975). vehicle towing cable apparatus, u.s patent, us3893709a issued july 1975. uspto, alexandria, virginia, usa. ham, c. w., crank e. j. and rogers, w. l. (1958). mechanics of machinery, mcgrawhill, 1958. hancock, c. l. (1978). bumper hitch for towing motorcycles, u.s patent 4,111,449, issued september 5, 1978. uspto, alexandria, virginia, usa. hasif, z. (2015). towing device between two motorcycles, issued september 2015 from diploma project report, faculty of mechanical engineering, universiti teknikal malaysia melaka, malaysia. hunt, k. h. (1978). kinematic geometry of mechanisms, oxford university press, new york, 1978. hanson, p. (2017). why you need to know “how much does a motorcycle weigh?” issued on february 2017 from http://letsridemotorbike.com/how-much-does-amotorcycle-weigh/ journal of mechanical engineering and technology issn 2180-1053 vol. 12 no.1 june-december 2020 66 marc. (2013). the strength of mild steel, issued on august 2013 from https://www.austenknapman.co.uk/blog/material-information/the-strength-ofmild-steel/ mills, b. r. (2016). “what is webbing?” issued on august 2016 from https://www.ballyribbon.com/what-is-webbing. myszka, d. h. (2012). machines & mechanisms: applied kinematic analysis, 4th ed., pearson, boston, ma. nik hak, n. a. a. (2018). redesign of motorcycle towing device for commercialization, beng final year project report, faculty of mechanical engineering, universiti teknikal malaysia melaka, malaysia. issn: 2180-1053 vol. 7 no. 2 july december 2015 the exergy efficiency and pumping power of nanofluid through a helically coiled tube heat exchanger under turbulent flow 75 the exergy efficiency and pumping power of nanofluid through a helically coiled tube heat exchanger under turbulent flow alireza falahat1*, mohsen shabani2, mohsen maleki3 1 department of mechanics, islamic azad university, mahshahr, iran 2production technology research institute (acecr), ahvaz, iran 3energy management, imam petrochemical company, mahshahr, iran abstract this paper theoretically examines the effects of water-al2o3 nanofluid on exergy destruction, exergy efficiency and pumping power in the helically coiled tube heat exchanger under turbulent flow and subjected constant wall condition. the effects of the nanoparticles volume concentration, nanoparticle dimensions, reynolds number, curvature ratio and dimensionless inlet temperature considered to be the main parameters in this study. it is found that when the reynolds number increases, dimensionless total exergy destruction decreases. it is observed that by increasing the nanoparticles volume concentration from 2% to 6%, the dimensionless thermal exergy destruction reduces by 3.64% to 20.21 % compared to pure water. also, it is seen that when nanoparticles dimensions increases, the exergy efficiency increases and pumping power decreases. finally, the exergy efficiency increases with increasing of curvature ratio and pumping power decreases with increasing of curvature ratio. keywords: helically coiled tube, exergy efficiency, second law analysis, turbulent flow, nanofluid. 1.0 introduction in industry and engineering applications, helically coiled heat exchangers are effective equipment, since it is used in industrial fields including power generation, food processing, petrochemical industry, hvac and refrigeration. (chingulpitak and wongwises, 2011),(zhao et al, 2011). helically coiled heat exchangers increase the heat transfer surface area per unit volume and enhance the heat transfer coefficient of the flow inside the tube without turbulence or additional heat transfer surface area. the centrifugal forces in the coiled tube induce a secondary flow pattern consisting of two vortices perpendicular to the axial flow direction is set up, and heat transport will occur not *correspondimg author e -mail: a.falahat@mhriau.ac issn: 2180-1053 vol. 7 no. 2 july december 2015 journal of mechanical engineering and technology 76 only by diffusion in the radial direction but also by convection. the contribution of this secondary convective transport dominates the overall process and enhances the rate of heat transfer per unit length of tube compared to a straight tube of equal length (prabhanjan et al, 2002). the common fluids such as water, ethylene glycol and oil have poor heat transfer performance in heat exchanger (godson et.al, 2010). if metallic nanoparticles with high thermal conductivity use in typically fluids, heat transfer will increase. enhancement of heat transfer using nanoparticles suspended in a base fluid has been studied widely in recent years (xie et al, 2002), (buongjorno, 2006), (bianco et al, 2009), (shafahi et al, 2010). xuan and li (2003) experimentally studied the flow and heat transfer characteristics for cu-water based nanofluids through a plain tube with a constant heat flux boundary condition. they found that the nanofluids give substantial enhancement of heat transfer rate compared to pure water. a number of researchers have investigated the performance of heat exchangers (mohammed et al, 2011),(ahmed et al, 2011),(lotfi et al, 2012),(raja et al, 2012) using nanofluids. naraki et al. (2013) investigated experimentally the heat transfer coefficient of cuo/water nanofluids in a car radiator under laminar flow regimes. they found that the overall heat transfer coefficient with nanofluid is more than the pure water and increases with increasing the nanoparticles volume fraction from 0 to 0.4%. peyghambarzadeh et al. (2011) experimentally studied the heat transfer of water-al2o3 and eg-al2o3 nanofluid in car radiator under different flow rate (2–6 liter per minute). their results reveal that the heat transfer increases enhancement about 40% compared to the base fluids. mukesh kumar et al. (2013),(2014) investigated experimentally the heat transfer and friction factor of a shell and helically coiled tube heat exchanger using water-al2o3 nanofluid under laminar and turbulent flow regimes. for laminar flow regime, they found that the overall heat transfer coefficient, inner heat transfer coefficient and experimental inner nusselt number are 24%, 25% and 28%, respectively. their results for turbulent flow regime indicated the nusselt number for coiled tube of 0.1%, 0.4% and 0.8% nanofluids increase 28%, 36% and 56%, respectively higher than base fluid. they found that wateral2o3 nanofluid was negligible pressure drop. exergy destruction or entropy generation minimization is a useful tool for evaluating the irreversibilities associated in process or device (bejan et al. . falahat and vosough [24] computed entropy generation in a coiled tube under constant heat flux for both laminar and turbulent regimes using alumina–water nanofluids. they found that by adding 1% volume fraction of nanoparticles to the base fluid, issn: 2180-1053 vol. 7 no. 2 july december 2015 the exergy efficiency and pumping power of nanofluid through a helically coiled tube heat exchanger under turbulent flow 77 entropy generation decreases about 3% in laminar flow. also, they obtained an optimal reynolds number for the turbulent flow for which the entropy generation was minimized. ko and ting [25] have applied this concept to find the most appropriate flow conditions of a fully developed, laminar forced convection flow through a helical coil tube for which entropy generation is minimized. shokouhmand and salimpour [26,27] studied deals with entropy generation analysis of fully developed laminar forced convection in a helical tube with uniform wall temperature. the second law of thermodynamic analysis of a helical coil heat exchanger using three different types of nanofluids is investigated analytically with khairule et al. [28]. they found that, the cuo/water is best nanofluid when compared with al2o3/water and zno/water, because the enhancement of heat transfer and entropy generation reduction in this type were obtained about 7.14% and 6.14% respectively. to the best of authors’ knowledge, the exergy analysis and pumping power of nanofluid in helically coiled tube heat exchanger under turbulent flow regime are not considered up to now. the main aims of this work are to investigate the exergy destruction or entropy generation and pumping power inside a helically coiled tube heat exchanger, subjected to constant wall temperature using nanofluid with turbulent flow regime. the effects of reynolds number, nanoparticles volume concentration, nanoparticles dimension, coil-to-tube radius ratio and dimension inlet temperature on exergy efficiency (second law efficiency) and pumping power are investigated. 2.0 methodology 2.1 physical model a typical helically coiled tube heat exchanger has been shown in figure 1. in this figure, d is inner diameter of the tube and d is curvature diameter of the coil, and h is the coil pitch. the curvature ratio, δ, is defined as the coil-to-tube radius ratio, d / d. the characteristics parameter and working conditions are shown in table1. the other three important dimensionless parameters namely, reynolds number (re), nusselt number (nu) , and dean number (dn) are defined as follow. 3 dimension, coil-to-tube radius ratio and dimension inlet temperature on exergy efficiency (second law efficiency) and pumping power are investigated. 2.0 methodology 2.1 physical model a typical helically coiled tube heat exchanger has been shown in figure 1. in this figure, d is inner diameter of the tube and d is curvature diameter of the coil, and h is the coil pitch. the curvature ratio,  , is defined as the coil-to-tube radius ratio, dd / . the characteristics parameter and working conditions are shown in table1. the other three important dimensionless parameters namely, reynolds number (re) , nusselt number )(nu , and dean number )(dn are defined as follow. 5.0 re,,re        d d dn k dh nu du   (1) where, u and h are average velocity and convective heat transfer coefficient respectively. figure 1 geometry configuration of a helically coiled tube heat exchanger. table 1. characteristic and working condition of helically coiled tube heat exchanger characteristic/working conditions numerical values coil diameter, )(md 0.12 coil length, )(ml 0.9 curvature ratio,  0.03, 0.06, 0.12 wall temperature, )(ktw 360 dimensionless temperature,  0.05, 0.1, 0.15 reference dead temperature, )(kto 298 issn: 2180-1053 vol. 7 no. 2 july december 2015 journal of mechanical engineering and technology 78 where, u and h are average velocity and convective heat transfer coefficient respectively. 3 dimension, coil-to-tube radius ratio and dimension inlet temperature on exergy efficiency (second law efficiency) and pumping power are investigated. 2.0 methodology 2.1 physical model a typical helically coiled tube heat exchanger has been shown in figure 1. in this figure, d is inner diameter of the tube and d is curvature diameter of the coil, and h is the coil pitch. the curvature ratio,  , is defined as the coil-to-tube radius ratio, dd / . the characteristics parameter and working conditions are shown in table1. the other three important dimensionless parameters namely, reynolds number (re) , nusselt number )(nu , and dean number )(dn are defined as follow. 5.0 re,,re        d d dn k dh nu du   (1) where, u and h are average velocity and convective heat transfer coefficient respectively. figure 1 geometry configuration of a helically coiled tube heat exchanger. table 1. characteristic and working condition of helically coiled tube heat exchanger characteristic/working conditions numerical values coil diameter, )(md 0.12 coil length, )(ml 0.9 curvature ratio,  0.03, 0.06, 0.12 wall temperature, )(ktw 360 dimensionless temperature,  0.05, 0.1, 0.15 reference dead temperature, )(kto 298 figure 1 geometry configuration of a helically coiled tube heat exchanger. table 1. characteristic and working condition of helically coiled tube heat exchanger 3 dimension, coil-to-tube radius ratio and dimension inlet temperature on exergy efficiency (second law efficiency) and pumping power are investigated. 2.0 methodology 2.1 physical model a typical helically coiled tube heat exchanger has been shown in figure 1. in this figure, d is inner diameter of the tube and d is curvature diameter of the coil, and h is the coil pitch. the curvature ratio,  , is defined as the coil-to-tube radius ratio, dd / . the characteristics parameter and working conditions are shown in table1. the other three important dimensionless parameters namely, reynolds number (re) , nusselt number )(nu , and dean number )(dn are defined as follow. 5.0 re,,re        d d dn k dh nu du   (1) where, u and h are average velocity and convective heat transfer coefficient respectively. figure 1 geometry configuration of a helically coiled tube heat exchanger. table 1. characteristic and working condition of helically coiled tube heat exchanger characteristic/working conditions numerical values coil diameter, )(md 0.12 coil length, )(ml 0.9 curvature ratio,  0.03, 0.06, 0.12 wall temperature, )(ktw 360 dimensionless temperature,  0.05, 0.1, 0.15 reference dead temperature, )(kto 298 4 nanoparticle volume fraction,  0-6% nanoparticle dimensions, )(nmdnp 30, 50, 70 reynolds number, re 20000-140000 2.2 thermo-physical properties of nanofluid the nanofluid in the channel is newtonian and assumed that the fluid phase and nanoparticles are in the thermal equilibrium state and they flow with the same velocity. the thermophysical properties of pure water, al2o3 nanoparticles which are density, heat capacity, effective dynamic viscosity and effective thermal conductivity are given in table 2. the density and heat capacity of the nanofluid can be defined as (corcione,2010) pbfnf   )1( (2) the effective dynamic viscosity of the nanofluid is given as (ko and ting(2005); 03.13.0)/(8.341 1     bfnpbf eff dd (3) where npd is the nanoparticle diameter, bfd is the diameter of molecule of a base fluid, and it is defined as:          bf bf n m d  6 (4) n is the avogadro number 12310022.6  mol . the thermal conductivity of the nanofluid due to the brownian motion is given as [30] ),(105 )()2( )(2)2( 4       tf d t c kkkk kkk kk pp pbf pbfbfp kbfbfp bfeff b f p              (5) where 07304.1)100(4407.8   (6) )1091123.3100669.3())(10917.3108217.2(),( 3232    reft t tf (7) 2.2 thermo-physical properties of nanofluid the nanofluid in the channel is newtonian and assumed that the fluid phase and nanoparticles are in the thermal equilibrium state and they flow with the same velocity. the thermophysical properties of pure water, al2o3 nanoparticles which are density, heat capacity, effective dynamic viscosity and effective thermal conductivity are given in table 2. the density and heat capacity of the nanofluid can be defined as (corcione,2010) 4 nanoparticle volume fraction,  0-6% nanoparticle dimensions, )(nmdnp 30, 50, 70 reynolds number, re 20000-140000 2.2 thermo-physical properties of nanofluid the nanofluid in the channel is newtonian and assumed that the fluid phase and nanoparticles are in the thermal equilibrium state and they flow with the same velocity. the thermophysical properties of pure water, al2o3 nanoparticles which are density, heat capacity, effective dynamic viscosity and effective thermal conductivity are given in table 2. the density and heat capacity of the nanofluid can be defined as (corcione,2010) pbfnf   )1( (2) the effective dynamic viscosity of the nanofluid is given as (ko and ting(2005); 03.13.0)/(8.341 1     bfnpbf eff dd (3) where npd is the nanoparticle diameter, bfd is the diameter of molecule of a base fluid, and it is defined as:          bf bf n m d  6 (4) n is the avogadro number 12310022.6  mol . the thermal conductivity of the nanofluid due to the brownian motion is given as [30] ),(105 )()2( )(2)2( 4       tf d t c kkkk kkk kk pp pbf pbfbfp kbfbfp bfeff b f p              (5) where 07304.1)100(4407.8   (6) )1091123.3100669.3())(10917.3108217.2(),( 3232    reft t tf (7) the effective dynamic viscosity of the nanofluid is given as (ko and ting(2005); issn: 2180-1053 vol. 7 no. 2 july december 2015 the exergy efficiency and pumping power of nanofluid through a helically coiled tube heat exchanger under turbulent flow 79 4 nanoparticle volume fraction,  0-6% nanoparticle dimensions, )(nmdnp 30, 50, 70 reynolds number, re 20000-140000 2.2 thermo-physical properties of nanofluid the nanofluid in the channel is newtonian and assumed that the fluid phase and nanoparticles are in the thermal equilibrium state and they flow with the same velocity. the thermophysical properties of pure water, al2o3 nanoparticles which are density, heat capacity, effective dynamic viscosity and effective thermal conductivity are given in table 2. the density and heat capacity of the nanofluid can be defined as (corcione,2010) pbfnf   )1( (2) the effective dynamic viscosity of the nanofluid is given as (ko and ting(2005); 03.13.0)/(8.341 1     bfnpbf eff dd (3) where npd is the nanoparticle diameter, bfd is the diameter of molecule of a base fluid, and it is defined as:          bf bf n m d  6 (4) n is the avogadro number 12310022.6  mol . the thermal conductivity of the nanofluid due to the brownian motion is given as [30] ),(105 )()2( )(2)2( 4       tf d t c kkkk kkk kk pp pbf pbfbfp kbfbfp bfeff b f p              (5) where 07304.1)100(4407.8   (6) )1091123.3100669.3())(10917.3108217.2(),( 3232    reft t tf (7) where dnp is the nanoparticle diameter, dbf is the diameter of molecule of a base fluid, and it is defined as: 4 nanoparticle volume fraction,  0-6% nanoparticle dimensions, )(nmdnp 30, 50, 70 reynolds number, re 20000-140000 2.2 thermo-physical properties of nanofluid the nanofluid in the channel is newtonian and assumed that the fluid phase and nanoparticles are in the thermal equilibrium state and they flow with the same velocity. the thermophysical properties of pure water, al2o3 nanoparticles which are density, heat capacity, effective dynamic viscosity and effective thermal conductivity are given in table 2. the density and heat capacity of the nanofluid can be defined as (corcione,2010) pbfnf   )1( (2) the effective dynamic viscosity of the nanofluid is given as (ko and ting(2005); 03.13.0)/(8.341 1     bfnpbf eff dd (3) where npd is the nanoparticle diameter, bfd is the diameter of molecule of a base fluid, and it is defined as:          bf bf n m d  6 (4) n is the avogadro number 12310022.6  mol . the thermal conductivity of the nanofluid due to the brownian motion is given as [30] ),(105 )()2( )(2)2( 4       tf d t c kkkk kkk kk pp pbf pbfbfp kbfbfp bfeff b f p              (5) where 07304.1)100(4407.8   (6) )1091123.3100669.3())(10917.3108217.2(),( 3232    reft t tf (7) n is the avogadro number 6.022x1023 mol-1. the thermal conductivity of the nanofluid due to the brownian motion is given as [30] 4 nanoparticle volume fraction,  0-6% nanoparticle dimensions, )(nmdnp 30, 50, 70 reynolds number, re 20000-140000 2.2 thermo-physical properties of nanofluid the nanofluid in the channel is newtonian and assumed that the fluid phase and nanoparticles are in the thermal equilibrium state and they flow with the same velocity. the thermophysical properties of pure water, al2o3 nanoparticles which are density, heat capacity, effective dynamic viscosity and effective thermal conductivity are given in table 2. the density and heat capacity of the nanofluid can be defined as (corcione,2010) pbfnf   )1( (2) the effective dynamic viscosity of the nanofluid is given as (ko and ting(2005); 03.13.0)/(8.341 1     bfnpbf eff dd (3) where npd is the nanoparticle diameter, bfd is the diameter of molecule of a base fluid, and it is defined as:          bf bf n m d  6 (4) n is the avogadro number 12310022.6  mol . the thermal conductivity of the nanofluid due to the brownian motion is given as [30] ),(105 )()2( )(2)2( 4       tf d t c kkkk kkk kk pp pbf pbfbfp kbfbfp bfeff b f p              (5) where 07304.1)100(4407.8   (6) )1091123.3100669.3())(10917.3108217.2(),( 3232    reft t tf (7) where 4 nanoparticle volume fraction,  0-6% nanoparticle dimensions, )(nmdnp 30, 50, 70 reynolds number, re 20000-140000 2.2 thermo-physical properties of nanofluid the nanofluid in the channel is newtonian and assumed that the fluid phase and nanoparticles are in the thermal equilibrium state and they flow with the same velocity. the thermophysical properties of pure water, al2o3 nanoparticles which are density, heat capacity, effective dynamic viscosity and effective thermal conductivity are given in table 2. the density and heat capacity of the nanofluid can be defined as (corcione,2010) pbfnf   )1( (2) the effective dynamic viscosity of the nanofluid is given as (ko and ting(2005); 03.13.0)/(8.341 1     bfnpbf eff dd (3) where npd is the nanoparticle diameter, bfd is the diameter of molecule of a base fluid, and it is defined as:          bf bf n m d  6 (4) n is the avogadro number 12310022.6  mol . the thermal conductivity of the nanofluid due to the brownian motion is given as [30] ),(105 )()2( )(2)2( 4       tf d t c kkkk kkk kk pp pbf pbfbfp kbfbfp bfeff b f p              (5) where 07304.1)100(4407.8   (6) )1091123.3100669.3())(10917.3108217.2(),( 3232    reft t tf (7) 4 nanoparticle volume fraction,  0-6% nanoparticle dimensions, )(nmdnp 30, 50, 70 reynolds number, re 20000-140000 2.2 thermo-physical properties of nanofluid the nanofluid in the channel is newtonian and assumed that the fluid phase and nanoparticles are in the thermal equilibrium state and they flow with the same velocity. the thermophysical properties of pure water, al2o3 nanoparticles which are density, heat capacity, effective dynamic viscosity and effective thermal conductivity are given in table 2. the density and heat capacity of the nanofluid can be defined as (corcione,2010) pbfnf   )1( (2) the effective dynamic viscosity of the nanofluid is given as (ko and ting(2005); 03.13.0)/(8.341 1     bfnpbf eff dd (3) where npd is the nanoparticle diameter, bfd is the diameter of molecule of a base fluid, and it is defined as:          bf bf n m d  6 (4) n is the avogadro number 12310022.6  mol . the thermal conductivity of the nanofluid due to the brownian motion is given as [30] ),(105 )()2( )(2)2( 4       tf d t c kkkk kkk kk pp pbf pbfbfp kbfbfp bfeff b f p              (5) where 07304.1)100(4407.8   (6) )1091123.3100669.3())(10917.3108217.2(),( 3232    reft t tf (7) where t is the fluid temperature, tref is the reference temperature and equals 293°k and к is the boltzman constant. table 2. physical properties of water and al2o3 nanoparticle. property water a12o3 cp (j / kg k) 4179 765 ρ (kg / m3) 997.1 3600 µ (kg / m.s) 0.001 _ _ _ k (w / mk) 0.613 36 issn: 2180-1053 vol. 7 no. 2 july december 2015 journal of mechanical engineering and technology 80 2.3 exergy analysis exergy is defined as the maximum amount of work that can be obtained by a system or a flow in which complete equilibrium with a reference environment is attained. the general exergy balance in steady state flow can be written as follows [31]: 5 where t is the fluid temperature, reft is the reference temperature and equals ko293 and  is the boltzman constant. table 2. physical properties of water and al2o3 nanoparticle. 2.3 exergy analysis exergy is defined as the maximum amount of work that can be obtained by a system or a flow in which complete equilibrium with a reference environment is attained. the general exergy balance in steady state flow can be written as follows [31]: desoutin xexexe   (9) or desoutmassinmassworkheat exexexexex  ,, (10) the rate form of general exergy balance can be expressed by equation 11: desoutoutininw exmmwqt t        01 (11) where t is the average temperature of the fluid inside the coiled tube, estimated as [32]         in out inout t t tt t ln (12) outt computed from energy balance for a control volume:  nfpnfinwwout cmldhtttt ,/)(exp)(  (13) or 5 where t is the fluid temperature, reft is the reference temperature and equals ko293 and  is the boltzman constant. table 2. physical properties of water and al2o3 nanoparticle. 2.3 exergy analysis exergy is defined as the maximum amount of work that can be obtained by a system or a flow in which complete equilibrium with a reference environment is attained. the general exergy balance in steady state flow can be written as follows [31]: desoutin xexexe   (9) or desoutmassinmassworkheat exexexexex  ,, (10) the rate form of general exergy balance can be expressed by equation 11: desoutoutininw exmmwqt t        01 (11) where t is the average temperature of the fluid inside the coiled tube, estimated as [32]         in out inout t t tt t ln (12) outt computed from energy balance for a control volume:  nfpnfinwwout cmldhtttt ,/)(exp)(  (13) the rate form of general exergy balance can be expressed by equation 11: 5 where t is the fluid temperature, reft is the reference temperature and equals ko293 and  is the boltzman constant. table 2. physical properties of water and al2o3 nanoparticle. 2.3 exergy analysis exergy is defined as the maximum amount of work that can be obtained by a system or a flow in which complete equilibrium with a reference environment is attained. the general exergy balance in steady state flow can be written as follows [31]: desoutin xexexe   (9) or desoutmassinmassworkheat exexexexex  ,, (10) the rate form of general exergy balance can be expressed by equation 11: desoutoutininw exmmwqt t        01 (11) where t is the average temperature of the fluid inside the coiled tube, estimated as [32]         in out inout t t tt t ln (12) outt computed from energy balance for a control volume:  nfpnfinwwout cmldhtttt ,/)(exp)(  (13) where t is the average temperature of the fluid inside the coiled tube, estimated as [32] 5 where t is the fluid temperature, reft is the reference temperature and equals ko293 and  is the boltzman constant. table 2. physical properties of water and al2o3 nanoparticle. 2.3 exergy analysis exergy is defined as the maximum amount of work that can be obtained by a system or a flow in which complete equilibrium with a reference environment is attained. the general exergy balance in steady state flow can be written as follows [31]: desoutin xexexe   (9) or desoutmassinmassworkheat exexexexex  ,, (10) the rate form of general exergy balance can be expressed by equation 11: desoutoutininw exmmwqt t        01 (11) where t is the average temperature of the fluid inside the coiled tube, estimated as [32]         in out inout t t tt t ln (12) outt computed from energy balance for a control volume:  nfpnfinwwout cmldhtttt ,/)(exp)(  (13) tout computed from energy balance for a control volume: 5 where t is the fluid temperature, reft is the reference temperature and equals ko293 and  is the boltzman constant. table 2. physical properties of water and al2o3 nanoparticle. 2.3 exergy analysis exergy is defined as the maximum amount of work that can be obtained by a system or a flow in which complete equilibrium with a reference environment is attained. the general exergy balance in steady state flow can be written as follows [31]: desoutin xexexe   (9) or desoutmassinmassworkheat exexexexex  ,, (10) the rate form of general exergy balance can be expressed by equation 11: desoutoutininw exmmwqt t        01 (11) where t is the average temperature of the fluid inside the coiled tube, estimated as [32]         in out inout t t tt t ln (12) outt computed from energy balance for a control volume:  nfpnfinwwout cmldhtttt ,/)(exp)(  (13) the flow exergy is computed as: 6 the flow exergy is computed as: )()( 000 ssthh  (14) where h , s and subscript zero are respectively enthalpy, entropy and properties at the restricted dead state )( 00 pandt . the entropy and enthalpy deviations and heat transfer rate of nanofluid in the helically coiled heat exchanger can be obtained as: )ln()ln(, in out in out nfpinout p p r t t csss  (15)  inoutnfpinout ttchhh  , (16)  inoutnfpw ttcmq  . (17) if equations (12) to (17) is replaced in equation (11), it may be rewritten as:                   0,0 0, ln tc p t t t tt tcmex nfpnfin outinout nfpdes  (18) where d ulf p nf 2 2  (19) dimensionless form of equation 18 can be expressed by equation 20: pdestdes nfpnfin outinout des eetc p t t t tt e                           ,, 0,0 ln  (20) the exergy efficiency or second law efficiency is computed as dipippo(2004); )1()(ln)( ln)( 11 0 , 0 ,00, ,0, t t ttc p t t ctttc p t t ctttc ex ex inoutnfp nf in nfpinnfp nfin out nfpinoutnfp in des                      (21) where h, s and subscript zero are respectively enthalpy, entropy and properties at the restricted dead state (t0 and p0. the entropy and enthalpy deviations and heat transfer rate of nanofluid in the helically coiled heat exchanger can be obtained as: 6 the flow exergy is computed as: )()( 000 ssthh  (14) where h , s and subscript zero are respectively enthalpy, entropy and properties at the restricted dead state )( 00 pandt . the entropy and enthalpy deviations and heat transfer rate of nanofluid in the helically coiled heat exchanger can be obtained as: )ln()ln(, in out in out nfpinout p p r t t csss  (15)  inoutnfpinout ttchhh  , (16)  inoutnfpw ttcmq  . (17) if equations (12) to (17) is replaced in equation (11), it may be rewritten as:                   0,0 0, ln tc p t t t tt tcmex nfpnfin outinout nfpdes  (18) where d ulf p nf 2 2  (19) dimensionless form of equation 18 can be expressed by equation 20: pdestdes nfpnfin outinout des eetc p t t t tt e                           ,, 0,0 ln  (20) the exergy efficiency or second law efficiency is computed as dipippo(2004); )1()(ln)( ln)( 11 0 , 0 ,00, ,0, t t ttc p t t ctttc p t t ctttc ex ex inoutnfp nf in nfpinnfp nfin out nfpinoutnfp in des                      (21) issn: 2180-1053 vol. 7 no. 2 july december 2015 the exergy efficiency and pumping power of nanofluid through a helically coiled tube heat exchanger under turbulent flow 81 6 the flow exergy is computed as: )()( 000 ssthh  (14) where h , s and subscript zero are respectively enthalpy, entropy and properties at the restricted dead state )( 00 pandt . the entropy and enthalpy deviations and heat transfer rate of nanofluid in the helically coiled heat exchanger can be obtained as: )ln()ln(, in out in out nfpinout p p r t t csss  (15)  inoutnfpinout ttchhh  , (16)  inoutnfpw ttcmq  . (17) if equations (12) to (17) is replaced in equation (11), it may be rewritten as:                   0,0 0, ln tc p t t t tt tcmex nfpnfin outinout nfpdes  (18) where d ulf p nf 2 2  (19) dimensionless form of equation 18 can be expressed by equation 20: pdestdes nfpnfin outinout des eetc p t t t tt e                           ,, 0,0 ln  (20) the exergy efficiency or second law efficiency is computed as dipippo(2004); )1()(ln)( ln)( 11 0 , 0 ,00, ,0, t t ttc p t t ctttc p t t ctttc ex ex inoutnfp nf in nfpinnfp nfin out nfpinoutnfp in des                      (21) if equations (12) to (17) is replaced in equation (11), it may be rewritten as: 6 the flow exergy is computed as: )()( 000 ssthh  (14) where h , s and subscript zero are respectively enthalpy, entropy and properties at the restricted dead state )( 00 pandt . the entropy and enthalpy deviations and heat transfer rate of nanofluid in the helically coiled heat exchanger can be obtained as: )ln()ln(, in out in out nfpinout p p r t t csss  (15)  inoutnfpinout ttchhh  , (16)  inoutnfpw ttcmq  . (17) if equations (12) to (17) is replaced in equation (11), it may be rewritten as:                   0,0 0, ln tc p t t t tt tcmex nfpnfin outinout nfpdes  (18) where d ulf p nf 2 2  (19) dimensionless form of equation 18 can be expressed by equation 20: pdestdes nfpnfin outinout des eetc p t t t tt e                           ,, 0,0 ln  (20) the exergy efficiency or second law efficiency is computed as dipippo(2004); )1()(ln)( ln)( 11 0 , 0 ,00, ,0, t t ttc p t t ctttc p t t ctttc ex ex inoutnfp nf in nfpinnfp nfin out nfpinoutnfp in des                      (21) where 6 the flow exergy is computed as: )()( 000 ssthh  (14) where h , s and subscript zero are respectively enthalpy, entropy and properties at the restricted dead state )( 00 pandt . the entropy and enthalpy deviations and heat transfer rate of nanofluid in the helically coiled heat exchanger can be obtained as: )ln()ln(, in out in out nfpinout p p r t t csss  (15)  inoutnfpinout ttchhh  , (16)  inoutnfpw ttcmq  . (17) if equations (12) to (17) is replaced in equation (11), it may be rewritten as:                   0,0 0, ln tc p t t t tt tcmex nfpnfin outinout nfpdes  (18) where d ulf p nf 2 2  (19) dimensionless form of equation 18 can be expressed by equation 20: pdestdes nfpnfin outinout des eetc p t t t tt e                           ,, 0,0 ln  (20) the exergy efficiency or second law efficiency is computed as dipippo(2004); )1()(ln)( ln)( 11 0 , 0 ,00, ,0, t t ttc p t t ctttc p t t ctttc ex ex inoutnfp nf in nfpinnfp nfin out nfpinoutnfp in des                      (21) dimensionless form of equation 18 can be expressed by equation 20: 6 the flow exergy is computed as: )()( 000 ssthh  (14) where h , s and subscript zero are respectively enthalpy, entropy and properties at the restricted dead state )( 00 pandt . the entropy and enthalpy deviations and heat transfer rate of nanofluid in the helically coiled heat exchanger can be obtained as: )ln()ln(, in out in out nfpinout p p r t t csss  (15)  inoutnfpinout ttchhh  , (16)  inoutnfpw ttcmq  . (17) if equations (12) to (17) is replaced in equation (11), it may be rewritten as:                   0,0 0, ln tc p t t t tt tcmex nfpnfin outinout nfpdes  (18) where d ulf p nf 2 2  (19) dimensionless form of equation 18 can be expressed by equation 20: pdestdes nfpnfin outinout des eetc p t t t tt e                           ,, 0,0 ln  (20) the exergy efficiency or second law efficiency is computed as dipippo(2004); )1()(ln)( ln)( 11 0 , 0 ,00, ,0, t t ttc p t t ctttc p t t ctttc ex ex inoutnfp nf in nfpinnfp nfin out nfpinoutnfp in des                      (21) the exergy efficiency or second law efficiency is computed as dipippo(2004); 6 the flow exergy is computed as: )()( 000 ssthh  (14) where h , s and subscript zero are respectively enthalpy, entropy and properties at the restricted dead state )( 00 pandt . the entropy and enthalpy deviations and heat transfer rate of nanofluid in the helically coiled heat exchanger can be obtained as: )ln()ln(, in out in out nfpinout p p r t t csss  (15)  inoutnfpinout ttchhh  , (16)  inoutnfpw ttcmq  . (17) if equations (12) to (17) is replaced in equation (11), it may be rewritten as:                   0,0 0, ln tc p t t t tt tcmex nfpnfin outinout nfpdes  (18) where d ulf p nf 2 2  (19) dimensionless form of equation 18 can be expressed by equation 20: pdestdes nfpnfin outinout des eetc p t t t tt e                           ,, 0,0 ln  (20) the exergy efficiency or second law efficiency is computed as dipippo(2004); )1()(ln)( ln)( 11 0 , 0 ,00, ,0, t t ttc p t t ctttc p t t ctttc ex ex inoutnfp nf in nfpinnfp nfin out nfpinoutnfp in des                      (21) the nusselt number and friction in coiled tube heat exchanger are calculated using the same correlations as described below (kakac and liu); 7 the nusselt number and friction in coiled tube heat exchanger are calculated using the same correlations as described below (kakac and liu);   4.08.08.0 pr)(re024.0)1(6.31 nu (22)   1047700rere0084.0 22.025.0    andforf (23) another important parameter for performance of heat exchanger is pumping power (pp), which can be expressed by equation 24: p m pp nf  )(  (24) the pressure drop can be expressed by equation 19 and the mass flow rate can be expressed by equation 25: nfdm   re 4  (25) 3.0 results and discussion the results of this work are presented in two sections. in the first section, the exergy analysis of nanofluid and in the second section, the pumping power are discussed. 3.1 exergy analysis results figure 2(a), (b) and (c) illustrates the dimensionless total, thermal and frictional exergy destruction rate as a function of reynolds number for different nanoparticles volume concentration. parameters that are fixed constant in this results include nmdnp 30 , 06.0 and 05.0 .the dimensionless thermal exergy destruction decreases with increasing reynolds number and nanoparticles volume concentration. this is because a higher nanoparticles volume concentration enhances nusselt number and increasing thermal conductivity of nanofluid. by increasing the nanoparticles volume concentration from 2% to 6%, the dimensionless thermal exergy destruction reduces by 3.45% to 19.29% for low reynolds number (re=40000) and 3.64% to 20.21 % for high reynolds number (re=120000) compared to pure water. it is seen from figure 2(b), the dimensionless frictional exergy destruction increased with increasing reynolds number and nanoparticles volume concentration. when nanoparticles volume concentration increases, the viscosity increased and causing a increasing in frictional losses. this, consequently, results in increasing the dimensionless frictional exergy destruction. figure 2(c) shows that the dimensionless frictional exergy destruction has a minor effect on dimensionless total exergy destruction because the value of frictional exergy destruction is too small for all nanoparticles volume concentration. also this figure issn: 2180-1053 vol. 7 no. 2 july december 2015 journal of mechanical engineering and technology 82 another important parameter for performance of heat exchanger is pumping power (pp), which can be expressed by equation 24: 7 the nusselt number and friction in coiled tube heat exchanger are calculated using the same correlations as described below (kakac and liu);   4.08.08.0 pr)(re024.0)1(6.31 nu (22)   1047700rere0084.0 22.025.0    andforf (23) another important parameter for performance of heat exchanger is pumping power (pp), which can be expressed by equation 24: p m pp nf  )(  (24) the pressure drop can be expressed by equation 19 and the mass flow rate can be expressed by equation 25: nfdm   re 4  (25) 3.0 results and discussion the results of this work are presented in two sections. in the first section, the exergy analysis of nanofluid and in the second section, the pumping power are discussed. 3.1 exergy analysis results figure 2(a), (b) and (c) illustrates the dimensionless total, thermal and frictional exergy destruction rate as a function of reynolds number for different nanoparticles volume concentration. parameters that are fixed constant in this results include nmdnp 30 , 06.0 and 05.0 .the dimensionless thermal exergy destruction decreases with increasing reynolds number and nanoparticles volume concentration. this is because a higher nanoparticles volume concentration enhances nusselt number and increasing thermal conductivity of nanofluid. by increasing the nanoparticles volume concentration from 2% to 6%, the dimensionless thermal exergy destruction reduces by 3.45% to 19.29% for low reynolds number (re=40000) and 3.64% to 20.21 % for high reynolds number (re=120000) compared to pure water. it is seen from figure 2(b), the dimensionless frictional exergy destruction increased with increasing reynolds number and nanoparticles volume concentration. when nanoparticles volume concentration increases, the viscosity increased and causing a increasing in frictional losses. this, consequently, results in increasing the dimensionless frictional exergy destruction. figure 2(c) shows that the dimensionless frictional exergy destruction has a minor effect on dimensionless total exergy destruction because the value of frictional exergy destruction is too small for all nanoparticles volume concentration. also this figure the pressure drop can be expressed by equation 19 and the mass flow rate can be expressed by equation 25: 7 the nusselt number and friction in coiled tube heat exchanger are calculated using the same correlations as described below (kakac and liu);   4.08.08.0 pr)(re024.0)1(6.31 nu (22)   1047700rere0084.0 22.025.0    andforf (23) another important parameter for performance of heat exchanger is pumping power (pp), which can be expressed by equation 24: p m pp nf  )(  (24) the pressure drop can be expressed by equation 19 and the mass flow rate can be expressed by equation 25: nfdm   re 4  (25) 3.0 results and discussion the results of this work are presented in two sections. in the first section, the exergy analysis of nanofluid and in the second section, the pumping power are discussed. 3.1 exergy analysis results figure 2(a), (b) and (c) illustrates the dimensionless total, thermal and frictional exergy destruction rate as a function of reynolds number for different nanoparticles volume concentration. parameters that are fixed constant in this results include nmdnp 30 , 06.0 and 05.0 .the dimensionless thermal exergy destruction decreases with increasing reynolds number and nanoparticles volume concentration. this is because a higher nanoparticles volume concentration enhances nusselt number and increasing thermal conductivity of nanofluid. by increasing the nanoparticles volume concentration from 2% to 6%, the dimensionless thermal exergy destruction reduces by 3.45% to 19.29% for low reynolds number (re=40000) and 3.64% to 20.21 % for high reynolds number (re=120000) compared to pure water. it is seen from figure 2(b), the dimensionless frictional exergy destruction increased with increasing reynolds number and nanoparticles volume concentration. when nanoparticles volume concentration increases, the viscosity increased and causing a increasing in frictional losses. this, consequently, results in increasing the dimensionless frictional exergy destruction. figure 2(c) shows that the dimensionless frictional exergy destruction has a minor effect on dimensionless total exergy destruction because the value of frictional exergy destruction is too small for all nanoparticles volume concentration. also this figure 3.0 results and discussion the results of this work are presented in two sections. in the first section, the exergy analysis of nanofluid and in the second section, the pumping power are discussed. 3.1 exergy analysis results figure 2(a), (b) and (c) illustrates the dimensionless total, thermal and frictional exergy destruction rate as a function of reynolds number for different nanoparticles volume concentration. parameters that are fixed constant in this results include dnp=30nm, δ = 0.06 and θ = 0.05. the dimensionless thermal exergy destruction decreases with increasing reynolds number and nanoparticles volume concentration. this is because a higher nanoparticles volume concentration enhances nusselt number and increasing thermal conductivity of nanofluid. by increasing the nanoparticles volume concentration from 2% to 6%, the dimensionless thermal exergy destruction reduces by 3.45% to 19.29% for low reynolds number (re=40000) and 3.64% to 20.21 % for high reynolds number (re=120000) compared to pure water. it is seen from figure 2(b), the dimensionless frictional exergy destruction increased with increasing reynolds number and nanoparticles volume concentration. when nanoparticles volume concentration increases, the viscosity increased and causing a increasing in frictional losses. this, consequently, results in increasing the dimensionless frictional exergy destruction. figure 2(c) shows that the dimensionless frictional exergy destruction has a minor effect on dimensionless total exergy destruction because the value of frictional exergy destruction is too small for all nanoparticles volume concentration. also this figure indicates that the behavior of dimensionless total exergy destruction is similar to dimensionless thermal exergy destruction. issn: 2180-1053 vol. 7 no. 2 july december 2015 the exergy efficiency and pumping power of nanofluid through a helically coiled tube heat exchanger under turbulent flow 83 8 indicates that the behavior of dimensionless total exergy destruction is similar to dimensionless thermal exergy destruction. (a) (b) (c) figure 2. variation of dimensionless exergy destruction with reynolds number for several nanoparticles volume concentration (a) thermal,(b) frictional and (c) total figure 2. variation of dimensionless exergy destruction with reynolds number for several nanoparticles volume concentration (a) thermal,(b) frictional and (c) total issn: 2180-1053 vol. 7 no. 2 july december 2015 journal of mechanical engineering and technology 84 figure 3 shows the exergy efficiency as a function of reynolds number for different nanoparticles volume concentration. parameters that are fixed constant in this results similar figure 2. it can be seen that the exergy efficiency increases with increasing reynolds number and nanoparticles volume concentration. this enhancement is because of the fact that, by increasing nanoparticles volume concentration, heat transfer increases. also, the curvature effect lead to forms the secondary flow in helically coiled tube [35] and this phenomena enhances heat transfer an increases exergy efficiency. for water-al2o3 nanofluid and nanoparticles volume concentration 2%, 4% and 6%, the enhancement of exergy efficiency is about 1.2%, 3.23% and 6.7% respectively when compared with pure water. for every nanoparticles volume concentration, when reynolds number increase from 20000 to 140000, the exergy efficiency increases about 10% and no exist optimal reynolds number for minimizing the exergy efficiency. 9 figure 3 shows the exergy efficiency as a function of reynolds number for different nanoparticles volume concentration. parameters that are fixed constant in this results similar figure 2. it can be seen that the exergy efficiency increases with increasing reynolds number and nanoparticles volume concentration. this enhancement is because of the fact that, by increasing nanoparticles volume concentration, heat transfer increases. also, the curvature effect lead to forms the secondary flow in helically coiled tube [35] and this phenomena enhances heat transfer an increases exergy efficiency. for water-al2o3 nanofluid and nanoparticles volume concentration 2%, 4% and 6%, the enhancement of exergy efficiency is about 1.2%, 3.23% and 6.7% respectively when compared with pure water. for every nanoparticles volume concentration, when reynolds number increase from 20000 to 140000, the exergy efficiency increases about 10% and no exist optimal reynolds number for minimizing the exergy efficiency. figure 3. variation of exergy efficiency with reynolds number for several nanoparticles volume concentration figure 4. variation of exergy efficiency with nanoparticles volume concentration for several nanoparticles dimension figure 3. variation of exergy efficiency with reynolds number for several nanoparticles volume concentration 9 figure 3 shows the exergy efficiency as a function of reynolds number for different nanoparticles volume concentration. parameters that are fixed constant in this results similar figure 2. it can be seen that the exergy efficiency increases with increasing reynolds number and nanoparticles volume concentration. this enhancement is because of the fact that, by increasing nanoparticles volume concentration, heat transfer increases. also, the curvature effect lead to forms the secondary flow in helically coiled tube [35] and this phenomena enhances heat transfer an increases exergy efficiency. for water-al2o3 nanofluid and nanoparticles volume concentration 2%, 4% and 6%, the enhancement of exergy efficiency is about 1.2%, 3.23% and 6.7% respectively when compared with pure water. for every nanoparticles volume concentration, when reynolds number increase from 20000 to 140000, the exergy efficiency increases about 10% and no exist optimal reynolds number for minimizing the exergy efficiency. figure 3. variation of exergy efficiency with reynolds number for several nanoparticles volume concentration figure 4. variation of exergy efficiency with nanoparticles volume concentration for several nanoparticles dimension figure 4. variation of exergy efficiency with nanoparticles volume concentration for several nanoparticles dimension issn: 2180-1053 vol. 7 no. 2 july december 2015 the exergy efficiency and pumping power of nanofluid through a helically coiled tube heat exchanger under turbulent flow 85 the influence of nanoparticles dimension on exergy efficiency is shown in figure 4. parameters that are fixed constant in this results include re = 100000, δ = 0.06 and θ = 0.05. it can be seen that the exergy efficiency increases with increasing nanoparticles volume concentration and decrease with increasing nanoparticles dimension. this is because of the fact that smaller nanoparticles dimensions has a higher surface of interaction with the base fluid and increases heat transfer as well as the increase thermal conductivity and decreasing the dimensionless thermal exergy destruction, also smaller nanoparticles dimensions increases viscosity of nanofluid strongly and increasing the dimensionless frictional exergy destruction, but the dimensionless frictional exergy destruction has a minor effect. therefore exergy efficiency increases. one of important parameter in helically coiled tube heat exchanger is a curvature ratio. figure 5 illustrates the effect of δ on exergy efficiency for several nanoparticles volume concentration. parameters that are fixed constant in this results include re = 100000, dnp = 30nm and θ = 0.05. it can be seen that exergy efficiency increases with the increase of δ for pure water and increases by increasing the nanoparticles volume concentration. for all nanoparticles volume concentration, when δ increases from 0.03 to 0.12, the exergy efficiency increases approximately 33%. 10 the influence of nanoparticles dimension on exergy efficiency is shown in figure 4. parameters that are fixed constant in this results include 100000re  , 06.0 and 05.0 . it can be seen that the exergy efficiency increases with increasing nanoparticles volume concentration and decrease with increasing nanoparticles dimension. this is because of the fact that smaller nanoparticles dimensions has a higher surface of interaction with the base fluid and increases heat transfer as well as the increase thermal conductivity and decreasing the dimensionless thermal exergy destruction, also smaller nanoparticles dimensions increases viscosity of nanofluid strongly and increasing the dimensionless frictional exergy destruction, but the dimensionless frictional exergy destruction has a minor effect. therefore exergy efficiency increases. one of important parameter in helically coiled tube heat exchanger is a curvature ratio. figure 5 illustrates the effect of  on exergy efficiency for several nanoparticles volume concentration. parameters that are fixed constant in this results include 100000re  , nmdnp 30 and 05.0 . it can be seen that exergy efficiency increases with the increase of  for pure water and increases by increasing the nanoparticles volume concentration. for all nanoparticles volume concentration, when  increases from 0.03 to 0.12, the exergy efficiency increases approximately 33%. figure 5. variation of exergy efficiency with  for several nanoparticles volume concentration figure 6 shows the exergy efficiency as a function of dimensionless temperature for different nanoparticles volume concentration. it can be seen that exergy efficiency decreases with increasing  , because when dimensionless temperature increases, the temperature difference between the wall of tube and the average nanofluid temperature decreases and heat transfer reduces. figure 5. variation of exergy efficiency with δ for several nanoparticles volume concentration figure 6 shows the exergy efficiency as a function of dimensionless temperature for different nanoparticles volume concentration. it can be seen that exergy efficiency decreases with increasing θ, because when dimensionless temperature increases, the temperature difference between the wall of tube and the average nanofluid temperature decreases and heat transfer reduces. issn: 2180-1053 vol. 7 no. 2 july december 2015 journal of mechanical engineering and technology 86 11 figure 6. variation of exergy efficiency with  for several nanoparticles volume concentration 3.2 pumping power results figure 7 illustrates the effect of nanoparticles volume concentration on the pumping power. it can be seen that, when reynolds number increases, the pumping power increases, because for fluid flowing at higher reynolds number, more pumping power is required but for flows with a very small reynolds number, pumping power trends to zero. pumping power increases with increasing nanoparticles volume concentration for fixed reynolds number because by adding nanoparticles to increase in fluid friction and pressure drop. this results are matched with falahat and vosough (2012) and kabeel et al. (2013). figure 7. variation of pumping power with reynolds number for several nanoparticles volume concentration figs. 8 shows the results of pumping power at various nanoparticles volume concentration and nanoparticles dimension. it is observed from the figure that the pumping power increases with increasing nanoparticles volume concentration and figure 6. variation of exergy efficiency with θ for several nanoparticles volume concentration 3.2 pumping power results figure 7 illustrates the effect of nanoparticles volume concentration on the pumping power. it can be seen that, when reynolds number increases, the pumping power increases, because for fluid flowing at higher reynolds number, more pumping power is required but for flows with a very small reynolds number, pumping power trends to zero. pumping power increases with increasing nanoparticles volume concentration for fixed reynolds number because by adding nanoparticles to increase in fluid friction and pressure drop. this results are matched with falahat and vosough (2012) and kabeel et al. (2013). 11 figure 6. variation of exergy efficiency with  for several nanoparticles volume concentration 3.2 pumping power results figure 7 illustrates the effect of nanoparticles volume concentration on the pumping power. it can be seen that, when reynolds number increases, the pumping power increases, because for fluid flowing at higher reynolds number, more pumping power is required but for flows with a very small reynolds number, pumping power trends to zero. pumping power increases with increasing nanoparticles volume concentration for fixed reynolds number because by adding nanoparticles to increase in fluid friction and pressure drop. this results are matched with falahat and vosough (2012) and kabeel et al. (2013). figure 7. variation of pumping power with reynolds number for several nanoparticles volume concentration figs. 8 shows the results of pumping power at various nanoparticles volume concentration and nanoparticles dimension. it is observed from the figure that the pumping power increases with increasing nanoparticles volume concentration and figure 7. variation of pumping power with reynolds number for several nanoparticles volume concentration figs. 8 shows the results of pumping power at various nanoparticles volume concentration and nanoparticles dimension. it is observed from the figure that the pumping power increases with increasing nanoparticles volume concentration and decreases with increasing issn: 2180-1053 vol. 7 no. 2 july december 2015 the exergy efficiency and pumping power of nanofluid through a helically coiled tube heat exchanger under turbulent flow 87 nanoparticles dimension. according to this figure, when nanoparticles dimension increases from 30 to 70 nm, the maximum pumping power decreasing is about 47% at 6% nanoparticles volume concentration. 12 decreases with increasing nanoparticles dimension. according to this figure, when nanoparticles dimension increases from 30 to 70 nm, the maximum pumping power decreasing is about 47% at 6% nanoparticles volume concentration. figure 8. variation of pumping power with nanoparticles volume concentration for several nanoparticles dimension figure 9 shows the pumping power as a function of curvature ratio for different nanoparticles volume concentration. the figure shows that the pumping power decreases with increasing curvature ratio and increases when nanoparticles volume concentration increases and. this is because of the fact that, when coiled diameter is constant, by increasing curvature ratio, diameter of tube increases and pressure drop is reduced. also, we can be seen that, in larger curvature ratio, increasing of nanoparticles volume concentration has a minor effect on pumping power and this effect is significant on smaller curvature ratio. figure 9. variation of pumping power with  for several nanoparticles volume concentration figure 8. variation of pumping power with nanoparticles volume concentration for several nanoparticles dimension figure 9 shows the pumping power as a function of curvature ratio for different nanoparticles volume concentration. the figure shows that the pumping power decreases with increasing curvature ratio and increases when nanoparticles volume concentration increases and. this is because of the fact that, when coiled diameter is constant, by increasing curvature ratio, diameter of tube increases and pressure drop is reduced. also, we can be seen that, in larger curvature ratio, increasing of nanoparticles volume concentration has a minor effect on pumping power and this effect is significant on smaller curvature ratio. 12 decreases with increasing nanoparticles dimension. according to this figure, when nanoparticles dimension increases from 30 to 70 nm, the maximum pumping power decreasing is about 47% at 6% nanoparticles volume concentration. figure 8. variation of pumping power with nanoparticles volume concentration for several nanoparticles dimension figure 9 shows the pumping power as a function of curvature ratio for different nanoparticles volume concentration. the figure shows that the pumping power decreases with increasing curvature ratio and increases when nanoparticles volume concentration increases and. this is because of the fact that, when coiled diameter is constant, by increasing curvature ratio, diameter of tube increases and pressure drop is reduced. also, we can be seen that, in larger curvature ratio, increasing of nanoparticles volume concentration has a minor effect on pumping power and this effect is significant on smaller curvature ratio. figure 9. variation of pumping power with  for several nanoparticles volume concentration figure 9. variation of pumping power with δ for several nanoparticles volume concentration issn: 2180-1053 vol. 7 no. 2 july december 2015 journal of mechanical engineering and technology 88 4.0 conclusions in the present paper, an analytical analysis was carried out to study the effects of nanoparticles volume concentration, nanoparticles dimension, reynolds number, curvature ratio and dimensionless inlet temperature of water-al2o3 nanofluid on the dimensionless exergy destruction and exergy efficiency and pumping power in a helically coiled tube heat exchanger under turbulent flow regime. the results of this study show that: • with increasing the reynolds number, the dimensionless thermal exergy destruction decreases and the dimensionless frictional exergy destruction increases but the dimensionless frictional exergy destruction has a negligible effect on dimensionless total exergy destruction. therefore trends of dimensionless total exergy destruction is similar to dimensionless thermal exergy destruction. • it is observed that with increasing nanoparticles volume concentration led to decreasing on dimensionless thermal exergy destruction. for example, by adding 6% nanoparticles volume concentration, dimensionless thermal exergy destruction decreases about 20% compared to pure water. • the exergy efficiency increases with increasing reynolds number and nanoparticles volume concentration. • when nanoparticles dimensions increases, the exergy efficiency increases and pumping power decreases. this is because of the fact that smaller nanoparticles dimensions increases heat transfer as well as the increase thermal conductivity and decreasing the dimensionless thermal exergy destruction, increases viscosity of nanofluid that led to increasing pumping power. • the exergy efficiency increases with increasing of curvature ratio and pumping power decreases with increasing of curvature ratio. for example, when δ increases from 0.03 to 0.12, the exergy efficiency increases approximately 33%. • the exergy efficiency decreases with increasing dimensionless inlet temperature. acknowledgements financial support for this research was provided by the mahshahr branch, islamic azad university, mahshahr, iran with title: "exergy criteria for performance investigation helically coiled tube heat exchanger using nanofluid". issn: 2180-1053 vol. 7 no. 2 july december 2015 the exergy efficiency and pumping power of nanofluid through a helically coiled tube heat exchanger under turbulent flow 89 nomenclature 14 nomenclature pc specific heat, kkgkj / d coil diameter, m d tube diameter, m dn dean number npd particle size, nm ex exergy rate, w f friction factor h heat transfer coefficient, kmw 2/ k thermal conductivity of the fluid, kmw / l length of coiled tube, m m mass flow rate, skg / n avogadro number nu nusselt number p pressure, pa pr prandtl number wq heat transfer rate, w re reynolds number s specific entropy, kkgkj / t temperature, k greek symbols  viscosity of the fluid, spa .  density, 3/ mkg  nanoparticles volume fraction  coil-to-tube ratio  exergy efficiency  dimensionless temperature, w inw t tt  subscripts bf base fluid in inlet out outlet nf nanofluid p particles o dead state w wall issn: 2180-1053 vol. 7 no. 2 july december 2015 journal of mechanical engineering and technology 90 references ahmed m., shuaib n., yusoff m. and al-falahi a. (2011). numerical investigations of flow and heat transfer enhancement in a corrugated channel using nanofluid”. international communications in heat and mass transfer, 38(10), 1368-1375. bejan a. (1988). advanced engineering thermodynamics. john wiley and sons inc. new york. bejan a. (1996). entropy generation minimization: the new thermodynamics of finite-size devices and finite-time processes. journal of applied physics, 7, 1191-1218. bianco v., manca o. and nardini, s. (2013). second law analysis of al2o3water nanofluid turbulent forced convection in a circular cross section tube with constant wall temperature”. advances in mechanical engineering, 6, 1-12. bianco v., chiacchio f., manca o. and nardini s. (2009). numerical investigation of nanofluids forced convection in circular tubes. applied thermal engineering, 29(17), 3632-3642. buongiorno j. (2006). convective transport in nanofluids. journal of heat transfer, 128(3), 240-250. chingulpitak, s., and wongwises s. (2011). a comparison of flow characteristics of refrigerants flowing through adiabatic straight and helical capillary tubes. international communications in heat and mass transfer, 38(3), 398-404. corcione m. (2010). heat transfer features of buoyancy driven nanofluids inside rectangular enclosures differentially heated at the side walls. international journal of thermal science, 49,1536-1546. dipippo, r. (2004). second law assessment of binary plants generating power from low-temperature geothermal fluids. international journal of heat and mass transfer, 33,5655-86. dravid a. n., smith k. a., merrill e. w. and brain p. l. t. (1971). effect of secondary fluid motion on laminar flow heat transfer in helically coiled tubes. aiche journal,17, 1114-1122. ebru k. a. (2006). evaluation of heat transfer and exergy loss in a concentric double pipe exchanger equipped with helical wires. energy conversion and management,47,3473–3486. falahat a. and vosough a. (2012). effect of nanofluid on entropy generation and pumping power in coiled tube. journal of thermophysics and heat transfer, 26(1), 141-146. ghasemi b. and aminossadati s. (2010). brownian motion of nanoparticles in a triangular enclosure with natural convection. international journal of thermal science, 49(6), 931-940. issn: 2180-1053 vol. 7 no. 2 july december 2015 the exergy efficiency and pumping power of nanofluid through a helically coiled tube heat exchanger under turbulent flow 91 godson, l. raja, b. lal d.m. and wongwises. s. (2010). “enhancement of heat transfer using nanofluids an overview”. renewable and sustainable energy review, 14, 629-641. gupta a. and das s.k. (2007). second law analysis of cross flow heat exchanger in the presence of axial dispersion in one fluid. energy, 32(5), 664-672. guo j.f, cheng l. and xu m.t. (2010). multi-objective optimization of heat exchanger design by entropy generation minimization. asme journal of heat transfer, 132(081801),1-8. kabeel, a.e, el maaty, a. and el samadony y., (2013). the effect of using nanoparticles on corrugated plate heat exchanger performance. applied thermal engineering, 52(1). 221-229. kakaç s, and liu h.t. (2002). heat exchangers selecting, rating, and thermal design. 2nd edition. boca raton: crc press. khairul m.a., saidur m.m., rahman r., alim a., hossain a. and abdin, z. (2013). heat transfer and thermodynamic analyses of a helically coiled heat exchanger using different types of nanofluids. international journal of heat and mass transfer, 67. 398-403. kotcioglu i., caliskan s., cansiz a., baskaya s., (2010). second law analysis and heat transfer in a cross-flow heat exchanger with a new winglettype vortex generator”. energy, 35. 3686-3695. ko t. h. and ting k. (2005). entropy generation and thermodynamic optimization of fully developed laminar convection in a helical coil. international communications in heat and mass transfer, 32(1-2), 214-232. lotfi r, rashidi a.m. and amrollahi, a (2012). experimental study on the heat transfer enhancement of mwnt-water nanofluid in a shell and tube heat exchanger”. international communications in heat and mass transfer, 39(1), 108-111. mohammed h., bhaskaran g., shuaib n. and abu-mulaweh h.i., (2011) influence of nanofluidson parallel flow square microchannel heat exchanger performance. international communications in heat and mass transfer, 38(1), 1-9. mukesh kumar p. c., kumar j. and suresh s., (2013). experimental investigation on convective heat transfer and friction factor in a helically coiled tube with al2o3/water nanofluid. journal of mechanical science and technology, 27(1), 239-245. mukesh kumar, p. c., kumar j., tamilarasan r., sendhil nathan s. and suresh, s. (2014). heat transfer enhancement and pressure drop analysis in a helically coiled tube using al2o3 / water nanofluid. journal of mechanical science and technology, 28(5), 1841-1847. issn: 2180-1053 vol. 7 no. 2 july december 2015 journal of mechanical engineering and technology 92 naraki m., peyghambarzadeh s.m.,. hashemabadi s.h, vermahmoudi y., (2013) parametric study of overall heat transfer coefficient of cuo/ water nanofluids in a car radiator. international journal of thermal sciences, 66, 82-90. peyghambarzadeh s.m., hashemabadi s.h., hoseini s.m., seifi and jamnani m., (2011). experimental study of heat transfer enhancement using water/ethylene glycol based nanofluids as a new coolant for car radiators. international communications in heat and mass transfer, 38, 1283-1290. prabhanjan, d.g. raghavan, g.s.v. and rennie. t.j. (2002). comparison of heat transfer rates between a straight tube heat exchanger and a helically coiled heat exchanger. international communications in heat and mass transfer, 29(2), 185-191. raja m., arunachalam r.m. suresh s. (2012) experimental studies on heat transfer of alumina/water nanofluid in a shell and tube heat exchanger with wire coil insert. international journal of mechanical and materials engineering, 7(1), 16-23. satapathy a.k. (2009). thermodynamic optimization of a coiled tube heat exchanger under constant wall heat flux condition. energy, 34(9), 1122-1126. shafahi m., bianco v., vafai k., and manca o. (2010). thermal performance of flat-shaped heat pipes using nanofluids. international journal of heat and mass transfer, 53,(7-8), 1438-1445. shokouhmand h. and salimpour m. r., (2007). entropy generation analysis of fully developed laminar forced convection in a helical tube with uniform wall temperature. heat and mass transfer, 44(2), 213-220. shokouhmand h. and salimpour m. r., (2007). optimal reynolds number of laminar forced convection in a helical tube subjected to uniform wall temperature. international communications in heat and mass transfer, 34(6), 753-761. zhao, z., wang, x., che d. and cao, z. (2011). numerical studies on flow and heat transfer in membrane helical-coil heat exchanger and membrane serpentine-tube heat exchanger. international communications in heat and mass transfer, 38(9), 1189-1194. xie, h. wang, j. xi t., liu y. and ai f.(2002). dependence of the thermal conductivityof nanoparticle-fluid mixture on the base fluid. journal of materials science letters, 21(19), 1469-1471. xuan y., li, q. (2003). investigation on convective heat transfer and flow features of nanofluids. journal of heat transfer, asme 125(1), 125-151. effect of load and temperature on friction using banana peel blended with paraffin oil under high loading capacity n. a. b. masripan1,2* h. s. o. al-nasrawi1, z. b. a. rashid1, g. omar1,2,3 , m.a. salim1,2,3, m.r. mansor1,2, a. m. saad1,2, a.h. nurfaizey1,2 and i. s. b. mohamed1,2 1faculty of mechanical engineering, universiti teknikal malaysia melaka 2centre for advanced research on energy (care), universiti teknikal malaysia, melaka 3advanced manufacturing center (amc), universiti teknikal malaysia, melaka abstract increased severity in operating conditions coupled with the environmental and toxicity issues related with using conventional lubricants. in addition, high price of fossil fuels has led to exploration of new kind natural additives as bio-lubricant. banana peel as agricultural wastes are potential to be developed as bio-oils that to replace the petroleum products, due to their environmentally friendly characteristics, biodegradable, nontoxic and renewable. the purpose of this study is to produce lubricant oil from banana peel (bp) as bio additives in paraffin oil, as well as to determine their physical and tribological properties as bio-lubricant under severe operation conditions to identify their ability for lubricants. tribological performance of banana peel (bp) as a biolubricant was tested using four-ball test machined under extreme pressure conditions, according to astm d 2783-03. experimental results showed significant improvement in overall performance with increased bp content compared with paraffin oil (po) through coefficient of friction parameter (cof) at 100 ˚c, lower value of cof which 0.086 for 50 %bp followed by 20% bp, 5% bp and 100 %po at values 0.089, 0.456 and 0.595 respectively. as results, banana peel as extreme pressure and anti-wear additives has been proven itself able for use in lubrication applications for gear and engine oils. . keywords: bio-lubricant, banana peel, high loading capacity, temperature, friction. __________________________________________________________ *corresponding author, email : norazmmi@utem.edu.my journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 2 june – december 2019 24 mailto:norazmmi@utem.edu.my 1. introduction tribology can be defined as the science and technology of interacting surface in relative motion which are present in various machined elements (nosonovsky & bhushan 2010; banjac et al. 2014). in almost every aspect of our daily lives, some appearances of tribology such as sliding, brushing, gripping, holding, machinery works, friction between skin and clothes, movement of artificial hip joints etc (mattei et al. 2011; gaikwad et al. 2013). friction is the force that resisting the relative motion of solid surface, fluid layers and material elements sliding against each other. there are many types of friction like, lubricated friction, fluid friction and dry friction. an important consequence of many types of friction is wear which lead to decline in performance and/or damaged to components. wear can be defined as undesired removal of material due to mechanical action (berman et al. 2013a; golshokouh et al. 2014). the rough surface, deep valley of asperities that formed helped to create an oil reservoir of the lubricant and prevented metal to metal contact (azmi et al. 2015). lubrication is the process or technique employed to reduced friction between two surface and wear of one of them or both. most of friction and wear are created during start-up and shout down of engines, whereas boundary lubrication (bl) occur at low speeds (tuszynski 2006; berman et al. 2013b; dou et al. 2016). the major reasons of using lubricants in engines are to control friction properties, reduce wear, and improve the efficiency. other reasons are for cooling, sealing, load balancing, cleaning and rust prevention (linke-diesinger 2008). engine oils consist of the base oil and additives. mineral based oil is used in most application to increase effectiveness in lubrication of various industrial parts fixed and mobile. although this oil is very useful, it is also an environmental hazard, poses damage on human, high price and is nonrenewable source (pettersson et al. 2008; s. syahrullail, nuraliza, et al. 2013). vegetable oils are known as renewable resources, environmentally friendly, non-toxic fluids, and are readily biodegradable (adhvaryu et al. 2004; tiong et al. 2012; shahabuddin, masjuki & kalam 2013; shahabuddin, masjuki, kalam, et al. 2013) . the bio-based lubricant is promising to protect the surfaces from wear and damage in comparison with the mineral oil due to lower value of dynamic pressure (nazri et al. 2013). in recent years, great development in engines and requirement on load carrying capacity of new and environmentally friend source (agricultural waste) especially at severe operating conditions. vegetable oils as additive have several properties which can achieve this purpose comparable to mineral oils, such as high lubricity, low volatility, high viscosity index, environmental friendly, more biodegradability, low coefficient of friction (cof) and low wear scar (waara 2006; alves et al. 2013; quinchia et al. 2014). low oxidation stability is one of the major factors hampering industry acceptance of vegetable oil-based lubricants(erhan et al. 2006; fox & stachowiak 2007; mustafa et al. 2015). banana skin been often referred as slipping tool by literatures. based on mabuchi et al. 2012, friction under banana skin was measured on flat panel common floor material during the sliding motion of shoes sole. the frictional coefficient was about 0.07 and this much lower than value on common materials and similar on well lubricated journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 2 june – december 2019 25 surfaces (mabuchi et al. 2012). bananas are one of the most popular fruits on the world market. it is well known that fruits contain various antioxidants (someya et al. 2002; gonzalez-montelongo et al. 2010; baskar et al. 2011; espinosa & santacruz 2017). the dispersion of banana peel in paraffin is stable and smooth without any sedimentation problem. moreover, oil shows good and promising tribological characteristic of lubricant (hamid et al. 2015). in this study, banana peel (bp) had been investigated as an additive in lubrication system. this is a novel attempt to use banana peel in lubrication system. hence, it is important and necessary to evaluate the characteristics of bp as lubricant additive to show their effect of load and temperature on friction performance to test their validity in industry application. 2. experimental set up 2.1 material preparation cavendish banana belongs to musaceae family under subgroup of triploid authentication, authorization and accounting (aaa) cultivar group. they include commercially important cultivars like the ‘dwarf cavendish’ and the ‘grand nain’ (hallam 2003).cavendish banana skin or banana peel (bp) which is pericarp (outside skin) had been chosen as natural additives in paraffin oil. paraffin oil as based-oil has been mixed with banana peels because of simple structure, unique tribological behavior and flexible for use under different percentage in preparation of lubrication samples by using ultrasonic homogenizer. it is also chemically composed of saturated hydrocarbons, consisting of a mixture of hydrocarbons chiefly of the alkene series and it is one of the higher members of alkane series which largely constitute the commercial form of this substance (zhou et al. 2001). ball bearings are common in mechanical studies, because they are widely used in automotive industry. the rolling balls have a much lower coefficient of friction as compared if two flat surface sliding against each other. however, ball bearings tend to have lower load capacity for their size than other kinds rolling element bearing due to the smaller contact area between the balls with inner and outer barriers (taher 2011). 2.2 lubricant composition there were four types of lubricant samples which are state in table 1 below. table 1: composition of lubricant samples lubricant samples composition of lubricant sample sample a 100% pure paraffin oil sample b paraffin oil +5% banana peel journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 2 june – december 2019 26 sample c paraffin oil +20% banana peel sample d paraffin oil +50% banana peel ultrasonic homogenizer was used to mix the paraffin oil with banana peel in one hour. the mixing of paraffin oil and banana peel was refined by using the clothe filter. preparation of lubrication samples was determined by using equation 1. volume percentage was referred after solution was made by mixing two liquids. total volume for each lubricant sample fixed to 100 ml that contained of banana peel and paraffin oil. c% v/v = (1) where, c% v/v is volume/volume percentage, (v solution) is volume of solution and (v substance) is the volume of substance. 2.3 friction test three design parameters were performed which are percentage of lubricant, temperature and load. the four sample of lubricant are test under the temperature of 27, 80 and 100 ̊c which the load gradually increased until obtaining load capacity. test temperature were selected for imitation of real engine temperature which begin at 27 ̊c and 80 ̊c is ideal temperature during engine running. friction test were carried out according to standard test methods for measurement of coefficient of friction (cof) and extreme pressure (ep) properties of lubricants until obtaining welding point on four-ball testing, according to astm d 2783-03 (astm 2003). the test has been conducted for 30 minutes on four samples and the loads on ball bearings were in the range of 500 n to 1750 n. figure 1: schematic diagram of four ball tester 2.3 coefficient of friction calculation the four ball machine was purposed for the first time by boerlage. the four ball apparatus can be used to test load carrying capacity, welding point, coefficient of journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 2 june – december 2019 27 friction and anti-wear properties of lubricating oil under standard operating conditions. table 2 shows the specification for using four ball test machines. table 2: specifications for using four ball test machines parameter unit tr 30l load newton (n) max. 10,000 spindle speed rpm 1000-3000 power v/hz/va 380/50/30/2000 scar range micron 100-4000 temperature (˚c) ambient to 100 ˚c drive motor kw 1.5 suitable weight was selected for end of lever arm and weights to give required test load until obtaining weld point. values coefficient of friction (cof) was investigated through a series of friction test. coefficient of friction of sample was investigated through a series of 4 ball friction tests and calculated by frictional torque, spring constant and applied load as equation 2, similar to calculation as used by (ing et al. 2012)(s. syahrullail, kamitani, et al. 2013) (zulkifli et al. 2014). (2) where,  is coefficient of friction, t is frictional torque in kg.mm, r is distance from the center of the contact surface on the lower balls to the axis of rotation, which is 3.67 mm and w is applied load in kg. 3. results and discussion 3.1 analysis of coefficient of friction, cof coefficient of friction had been reduced significantly by dispersing different concentration of banana peel compare to paraffin oil at all temperatures. this because of banana peel made both “ball bearing effect” and ‘polishing effect”, and consequently smoothing the rough friction contact surfaces. mabuchi et al. (2012) (mabuchi et al. 2012) estimated that polysaccharide follicular gel played the dominate role in lubricating effect of banana skin after the crush. some authors reported that banana contain a good source of natural oxidations, such as vitamin c, vitamin e, carotene, dopamine, flavonoids (kanazawa & sakakibara 2000). figure 2 show, at 27 ˚c and 700 n, values of coefficient of friction for lubricants 100% paraffin oil, 5% banana peel, 20% banana peel were 0.118, 0.086 and 0.098 until welding points of steel balls at 1100, 1500 and 1650 n with values of 0.605, 0.162 and 0.11 respectively. journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 2 june – december 2019 28 400 600 800 1000 1200 1400 1600 1800 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 c o e ff ic ie n t o f fr ic ti o n , c o f applied load, w (n) 100% po po +5% bp po +20% bp po +50% bp figure 2: effect of applied load, n on coefficient of friction, cof at 27 ˚c figure 3 shows paraffin oil at 80 ˚c, there was a significant increase in cof value from starting to welding point 0.18 to 0.538 for loads 500 n and 1000 n respectively. others lubricant has increasing superiority compared to at 27 ˚c. there clearest superiority was by bio-lubricants, where trend of the curves was decreasing, except sample 5% bp after 1000 n. values of cof for lubricants 5% bp, 20% bp and 50% bp at start and welding point were 0.131, 0.128, 0.149 and 0.152, 0.114, 0.094 respectively. on top of this, behavior of these samples was more stable in comparison with temperature at 27 ˚c, which shown excellent anti-friction performance and high load carrying capacity. the higher load differentiated the behavior of tested additive compositions at elevated temperatures. 500 600 700 800 900 1000 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 c o e ff ic ie n t o f fr ic ti o n , c o f applied load, w (n) 100% po po +5% bp po +20% bp po +50% bp figure 3: effect of applied load, n on coefficient of friction, cof at 80 ˚c journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 2 june – december 2019 29 figure 4 shows lubricants 5% bp, 20% bp, 50% bp at having lower values of cof compared with values at 27 ˚c and 80 ˚c. more severe conditions influenced the cof intensification in case po because the surface protection was much worse in comparison with bio-lubricant. therefore, po obtained highest values for cof up to welding point because of the breakdown of the lubricant film at this load. the curves for other lubricants decreased except for sample 5% bp at weld point 1475 n. figure 4 below show at 100 ˚c, values of cof for all lubricants of 100% po, 5% bp, 20% bp and 50% bp at start and welding point were 0.12, 0.121, 0.121, 0.13 and 0.595, 0.456, 0.089 and 0.086 respectively. 400 600 800 1000 1200 1400 1600 1800 0.1 0.2 0.3 0.4 0.5 0.6 0.7 c o e ff ic ie n t o f fr ic ti o n , c o f applied load, w (n) 100% po po +5% bp po +20% bp po +50% bp figure 4: effect of applied load, n on coefficient of friction, cof at 100 ˚c figure 1, 2, 3 and 4 shows that there was no clear relation between the load and cof. this behavior has been reported by many authors such as jiyuan, quan and lixia (2001) (jiyuan et al. 2001), rastogi and yadav (2003) (rastogi & yadav 2003), syahrullail et al. (2013) (s syahrullail et al. 2013), zulkifli et al. (2013) and hamid et al. (2015). these figures show that better friction resistance was at 50% follow by 20% and 5% of bp respectively with lower values of cof compared with po in all cases. as a result, it can be said that the friction resistance and stability increased with increasing load and temperature until welding point, which largely depends percentage of bp mixture. a similar effect was observed by adhvaryu et al. (2006), where there was sharp decrease in cof with increasing natural additive concentration like cottonseed, canola and olive oils in hexadecane (99% + anhydrous). a similar result was found by adhvaryu, erhan and perez (2004) (adhvaryu et al. 2004) on soybean oil and their derivatives in hexadecane. journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 2 june – december 2019 30 3.2 effect of temperature on coefficient of friction the results of variation value of cof with applied loads for various lubricants under different temperature shown in figure 5. in case of 20% bp at all temperatures, similar behavior was observed by this lubrication with its counterpart of 50% bp with results more superior as reflected in the shape of chart. in the case of 50% bp, progressive reduction of cof at all temperature was observed. anti-friction action of additive can be explained by the formation of the protect film (hamid et al., 2015). 50% bp showed a better result of friction reduction and test state was in boundary lubrication. besides, tests at lower temperature could not supply reliable information about oxidative stability and sometimes cannot evaluate the presence of protective film or anti-oxidation compound (gonzaga & pasquini 2006). at elevated temperature, a film was formed on the metal surface during thermal decomposition, containing effective compounds, resulting in the friction reduction thus cof reduction (hamid et al., 2015). hence, for bio-lubricant, mechanism of friction reduction could be achieved by increasing the content of banana peel in paraffin oil, especially at 100 ˚c, for lubricant 20% bp and 50% bp. 0 20 40 60 80 100 120 0.0 0.1 0.2 0.3 0.4 0.5 0.6 c o e ff ic ie n t o f fr ic ti o n , c o f temperature, t (°c) 100% po po +5% bp po +20% bp po +50% bp figure 5: effect of temperature (˚c) on value of coefficient of friction, cof weld points were selected for comparison between lubricants used at temperature of 27, 80, and 100 ˚c. moreover, metal debris moved between the balls. hence, the surface became highly worn with increased temperature. this was due to the fact at elevated temperature, the lubricant film formed by po tended to be less stable or was more likely to break down. related results were found by ing et al. (2012a)(ing et al. 2012) . moreover, metal to metal contact between balls would increase the frictional resistance. as a result, cof increase compared to bio-lubricant at region welding. journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 2 june – december 2019 31 4. conclusion the coefficient of friction was found to decrease with increase of banana peel content. the behavior of the lubricant under extreme pressure conditions became better with increase of the banana peel content. the value of coefficient of friction at 100 ˚c was 0.086, 0.089, 0.046, 0.595 was refer to 50% bp, 20% bp, 5% bp and 100% po respectively. banana peel as natural additives has ability to improve physical and tribological properties of paraffin oil. acknowledgements the author would like to thank ministry of higher education malaysia and universiti teknikal malaysia melaka, utem for funding this research under fundamental research grant scheme frgs/1/2015/tk10/fkm/02/f00275. references adhvaryu, a., erhan, s.z. & perez, j.m., 2004. tribological studies of thermally and chemically modified vegetable oils for use as environmentally friendly lubricants. wear, 257(3–4), pp.359–367. alves, s.m., barros, b.s., trajano, m.f., ribeiro, k.s.b., moura, e.., 2013. tribological behavior of vegetable oil-based lubricants with nanoparticles of oxides in boundary lubrication conditions. tribology international, 65, pp.28–36. astm, 2003. astm d 2783-03: standard test method for measurement of extremepressure properties of lubricating fluids ( four-ball method ), west conshohocken: astm. azmi, m.i.r, tee b.t, masripan n.a.b, chong c.t, 2015. preliminary study of friction and wear on natural oil-based lubricants. , (november), pp.220–221. banjac, m., vencl, a. & otovic, s., 2014. friction and wear processes – thermodynamic tribology in industry. tribology in industry vol.36, no. 4 (2014) 341-347, 36(4), pp.341–347. ramakrishnan, b., selvaraj s., babu s., radhakrishnan s., radhakrishnan n., palanisamy p., 2011. antioxidant potential of peel extracts of banana varieties. food and nutrition sciences, 2(10), pp.1128–1133. berman, d., erdemir, a. & sumant, a. v., 2013a. few layer graphene to reduce wear and friction on sliding steel surfaces. carbon, 54, pp.454–459. berman, d., erdemir, a. & sumant, a. v., 2013b. reduced wear and friction enabled by graphene layers on sliding steel surfaces in dry nitrogen. carbon, 59, pp.167– 175. xuan d., andrew r. k., xingliang h., hee d. j., qian w., yip-w. c., and jiaxing h., journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 2 june – december 2019 32 2016. self-dispersed crumpled graphene balls in oil for friction and wear reduction. proceedings of the national academy of sciences, 113(6), pp.1528–1533. erhan, s.z., sharma, b.k. & perez, j.m., 2006. oxidation and low temperature stability of vegetable oil-based lubricants. industrial crops and products, 24(3), pp.292– 299. espinosa, a. & santacruz, s., 2017. phenolic compounds from the peel of musa cavendish , musa acuminata and musa cavandanaish compuestos fenólicos a partir de la corteza de musa cavendish , musa acuminata y musa cavandanaish. revista politécnica, 38(2), pp.1–5. fox, n.j. & stachowiak, g.w., 2007. vegetable oil-based lubricants-a review of oxidation. tribology international, 40(7), pp.1035–1046. gaikwad, a., coe, p. & ghalme, s.g., 2013. effect of lubricant viscosity and surface roughness on coefficient of friction in rolling contact tribology in industry. , 35(4). golshokouh, i., syahrullail, s. & ani, f.n., 2014. influence of normal load and temperature on tribological properties of jatropha oil. jurnal teknologi, 71(2), pp.145–150. gonzaga, f.b. & pasquini, c., 2006. a new method for determination of the oxidative stability of edible oils at frying temperatures using near infrared emission spectroscopy. analytica chimica acta, 570(1), pp.129–135. gonzalez-montelongo, r., gloria lobo, m. & gonzalez, m., 2010. antioxidant activity in banana peel extracts: testing extraction conditions and related bioactive compounds. food chemistry, 119(3), pp.1030–1039. hallam, d., 2003. the world banana economy, 1985-2002. bananas and plantains, p.98. hamid a.h, masripan n.a.b, basiron j., mustafa m.m.b, hasan r., abdollah m.f.b, ismail r. , 2015. effect of banana peels as an additive on the tribological proporties of paraffin oil. jurnal teknologi, 77(21), pp.73–77. ing, t.c. et al., 2012. the effect of temperature on the tribological behavior of rbd palm stearin. tribology transactions, 55(5). jiyuan, c. a i., quan, g. a n. & lixia, d. a i., 2001. investigation on tribology behavior of lubricants using the coefficient of friction test method. , 44(august). kanazawa, k. & sakakibara, h., 2000. high content of dopamine, a strong antioxidant, in cavendish banana. journal of agricultural and food chemistry, 48(3). linke-diesinger, a., 2008. system of commercial turbofan engine 1st ed., new delhi, india: springer-verlag berlin heidelberg. mabuchi, k. et al., 2012. frictional coefficient under banana skin. tribology online, 7(3), pp.147–151. journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 2 june – december 2019 33 mattei l., di puccio f., piccigallo b., ciulli e., 2011. lubrication and wear modelling of artificial hip joints: a review. tribology international, 44(5), pp.532–549. mustafa m.m.b., masripan n.a.b., abdollah m.f.b, basiron j., 2015. preliminary study on tribological properties of banana peel broth as additive in paraffin oil. , (march), pp.51–52. nazri z.h., rody m.z.m., abdollah m.f.b., rafeq s.a., amiruddin h., noreffendy t., masripan n.a.b., 2013. elastohydrodynamics lubrication for bio-based lubricants in elliptical conjunction. procedia engineering, 68, pp.123–129. available at: nosonovsky, m. & bhushan, b., 2010. green tribology: principles, research areas and challenges. philosophical transactions. series a, mathematical, physical, and engineering sciences, 368, pp.4677–4694. pettersson, a., elisabet, k. & minami, i., 2008. additives for environmentally adapted lubricants – friction and wear protection. tribology online, 3(3), pp.163–167. quinchia l.a., delgado m.a., reddyhoff t., gallegosa c., spikes h.a., 2014. tribological studies of potential vegetable oil-based lubricants containing environmentally friendly viscosity modifiers. tribology international, 69, pp.110– 117. rastogi, r.b. & yadav, m., 2003. suspension of molybdenum-sulphur complexes in paraffin oil as extreme pressure lubricants. tribology international, 36(7). shahabuddin, m., masjuki, h.h., kalam, m.a., et al., 2013. comparative tribological investigation of bio-lubricant formulated from a non-edible oil source (jatropha oil). industrial crops and products, 47, pp.323–330. shahabuddin, m., masjuki, h.h. & kalam, m.a., 2013. experimental investigation into tribological characteristics of biolubricant formulated from jatropha oil. procedia engineering, 56, pp.597–606. someya, s., yoshiki, y. & okubo, k., 2002. antioxidant compounds from bananas (musa cavendish). food chemistry, 79(3), pp.351–354. syahrullail s., wira j.y., wan nik w.b., fawwaz w.n., 2013. friction characteristics of rbd palm olein using four-ball tribotester. , 315, pp.936–940. syahrullail, s., nuraliza, n., et al., 2013. wear characteristic of palm olein as lubricant in different rotating speed. procedia engineering, 68, pp.158–165. syahrullail, s., kamitani, s. & shakirin, a., 2013. performance of vegetable oil as lubricant in extreme pressure condition. procedia engineering, 68, pp.172–177. taher, m., 2011. tribological performance of novel boron dithiocarbamate lubricant additives. lulea university of technology. tiong, c.i., 2012. tribological evaluation of refined, bleached and deodorized palm stearin using four-ball tribotester with different normal loads. journal of zhejiang university science a, 13(8), pp.633–640. journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 2 june – december 2019 34 tuszynski, w., 2006. an effect of lubricating additives on tribochemical phenomena in a rolling steel four-ball contact. tribology letters, 24(3), pp.207–215. waara, p., 2006. lubricants influence on wear in sharp rail curves. lulea, sweden: lulea university of technology. zhou, j.., 2001. study on an antiwear and extreme pressure additive of surface coated laf3 nanoparticles in liquid paraffin. wear, 249(5–6), pp.333–337. zulkifli, n.w.m., 2014. the effect of palm oil trimethylolpropane ester on extreme pressure lubrication. proceedings of the institution of mechanical engineers, part j: journal of engineering tribology, 228(2), pp.160–169. journal of mechanical engineering and technology issn 2180-1053 vol. 11 no. 2 june – december 2019 35 issn: 2180-1053 vol. 7 no. 1 january june 2015 investigating the surface elasticity and tension effects on critical buckling behaviour of nanotubes based on differential transformation method 11 investigating the surface elasticity and tension effects on critical buckling behaviour of nanotubes based on differential transformation method f. ebrahimi1*, m. boreiry1, g. shaghaghi1 1department of mechanical engineering, imam khomeini international university qazvin, iran abstract by considering the coupled effects of surface and nonlocal elasticity theory, the critical buckling load response of silicon/aluminium nanotubes is investigated in this paper. the nonlocal eringen theory takes into account the effect of small scale size while the gurtin-murdoch model is used to incorporate the surface effects. governing equations are derived through hamilton’s principle. the differential transformation method (dtm) as an efficient and accurate numerical tool is employed to solve the governing equations of nanotubes subjected to different boundary conditions. the output results are compared favourably with available published works. the detailed mathematical derivations are presented and numerical investigations are performed while the emphasis is placed on investigating the effect of the nonlocal parameter, surface effect, aspect ratio, mode number and beam size on critical buckling loads of the nanotube in detail. the results show that increasing the nonlocal parameter increase the buckling ratio of the nanotubes. keywords: critical buckling behaviour, nonlocal elasticity theory, surface elasticity and tension effects, differential transformation method 1.0 introduction in order to study the mechanical behaviours of nanostructures, the surface effects and nonlocal elasticity theory are two important fields which are investigated by researchers separately, or simultaneously. the surface of a solid is a region with small thickness which has different properties from the bulk. if the surface energy-to-bulk energy ratio is large, for example in the case of nanostructures, the surface effects cannot be ignored (he et. al, 2004). on the other hand, the nonlocal elasticity theory which is initiated in the paper of eringen * corresponding author email: febrahimy@eng.ikiu.ac.ir issn: 2180-1053 vol. 7 no. 1 january june 2015 journal of mechanical engineering and technology 12 (1983) expresses that the stress at a point is a function of strains at all points in the continuum. to account for the effect of surfaces/interfaces on mechanical deformation, the surface elasticity theory is presented by modelling the surface as a two dimensional membrane adhering to the underlying bulk material without slipping (gurtin and murdoch 1975),(gurtin et al, 1998). there are many studies related to the wave propagation, static, buckling and free linear and nonlinear vibration analysis of nanobeams and carbon nanotubes based on different beam theories gurtin et al (1998) established the theory of surface elasticity to explain various size-dependent phenomena at the nanoscale, and the predictions fit well with atomistic simulations and experimental measurements. wang and feng (2007) analysed the surface effects on the axial buckling of nanowires. by using the surface cauchy–born model (park, 2009) analysed the size-dependent effect of the residual surface stress on the resonant frequencies of silicon nanowires under finite deformation. hosseini et al. (2013) studied the surface and nonlocal effects on free vibration of nanobeam based on both timoshenko and euler-bernoulli beam theory (ebt) for different boundary conditions. in similar work, malekzadeh et al. (2013) studied surface and nonlocal effect on free nonlinear vibration of non-uniform nanobeams based on ebt and timoshenko beam theory. they expressed that the influence of surface and nonlocal effects depends on the boundary conditions of the nanobeam. also, eltaher et al. (2013) studied the coupling effects of nonlocal and surface energy on free vibration of nanobeam based on ebt for simply supported nanobeam. they showed that the surface effects depend on the size and the material of the nanobeam by calculating natural frequencies for two different materials. recently, (ansari and sahmani, 2011) studied bending behaviour and buckling of nanobeams including surface stress corresponding to different beam theories without consideration of nonlocality effect. moreover, the governing motion equations are often solved by analytical method hosseini et al. (2013) or finite element methods (eltaher et al, 2013) or generalized differential quadrature (gdq) method (ansari and sahmani, 2011) and other solutions which need high cpu time to solve. dtm is also used to find the exact solution of both linear and nonlinear equations and even partial differential equations with high precision and also is simpler in compare with other methods. although this method comes from taylor series expansion, but dtm is different from the traditional high order taylor’s series method. because the traditional taylor expansion requires symbolic competition of the necessary derivatives of the data functions and issn: 2180-1053 vol. 7 no. 1 january june 2015 investigating the surface elasticity and tension effects on critical buckling behaviour of nanotubes based on differential transformation method 13 thus is computationally taken long time for large orders, while dtm takes less time to solve polynomial series. in other words, by applying dtm, governing equations for different boundary conditions reduces to algebraic equations, and finally all the calculations turn into simple iterative process. also as seen in the literature, dtm has been used for solving a vast range of problems in different fields of engineering. to the best knowledge of the authors, no research effort has been devoted so far to find the solution of critical buckling load of nanotubes considering both surface and small scale effects employing differential transformation method. motivated by this fact, in this study, differential transformation method is applied in analysing the surface effects, including surface elasticity and stress, on critical buckling load of nanotubes, made of aluminium and silicon, using nonlocal elasticity theory. hamilton’s principle is employed to derive the governing equation and corresponding boundary conditions. dtm is then used to obtain the critical buckling load of nanotubes with various boundary conditions. to this end, the output results are compared favourably with those published works and influences of the surface effect, nonlocal parameter and size of nanotube on the critical buckling load are investigated. 2.0 theory and formulation nonlocal constitutive relation for euler-bernoulli beam is given as (eringen, 1983): 3 2.0 theory and formulation nonlocal constitutive relation for euler-bernoulli beam is given as (eringen, 1983): 2 2 xx xx xxex         (1) where xx and xx are the nonlocal stress and strain, respectively. the distributed transverse loading induced by the residual surface tension is (farshi et al , 2010): 2 0 2 w q q h x     (2) and h is the surface effect constant is given by: 2 oh d (3) effective flexural rigidity, *ei for nanotube is given by: * 4 4 3 31 ( ) ( ) 4 s o i o iei e r r e r r     (4) the general differential equation of euler-bernoulli beam based on nonlocal continuum model and surface effect are derived using the principle of hamilton and expressed by (reddy, (2007): * 2 2 2 2 2 2 2 2 2 2 2 2 4 0 02 2 2 4 2 22 [ ] w w w w ei n a i x x x x x t t x w w w n a i x x t t w q h q x w h xx                                                             (5) the coordinate system for nanotube is shown in figure 1. figure 1. geometry of nanotube with length l , inner and outer radii ri and ro 3.0 differential transformation method differential transformation method is one of the useful techniques to solve the differential equations with small calculation errors and ability to solve nonlinear equations with boundary conditions value problems. abdel-halim hassan (2002) applied the dtm on eigenvalues and normalized eigenfunctions. where 3 2.0 theory and formulation nonlocal constitutive relation for euler-bernoulli beam is given as (eringen, 1983): 2 2 xx xx xxex         (1) where xx and xx are the nonlocal stress and strain, respectively. the distributed transverse loading induced by the residual surface tension is (farshi et al , 2010): 2 0 2 w q q h x     (2) and h is the surface effect constant is given by: 2 oh d (3) effective flexural rigidity, *ei for nanotube is given by: * 4 4 3 31 ( ) ( ) 4 s o i o iei e r r e r r     (4) the general differential equation of euler-bernoulli beam based on nonlocal continuum model and surface effect are derived using the principle of hamilton and expressed by (reddy, (2007): * 2 2 2 2 2 2 2 2 2 2 2 2 4 0 02 2 2 4 2 22 [ ] w w w w ei n a i x x x x x t t x w w w n a i x x t t w q h q x w h xx                                                             (5) the coordinate system for nanotube is shown in figure 1. figure 1. geometry of nanotube with length l , inner and outer radii ri and ro 3.0 differential transformation method differential transformation method is one of the useful techniques to solve the differential equations with small calculation errors and ability to solve nonlinear equations with boundary conditions value problems. abdel-halim hassan (2002) applied the dtm on eigenvalues and normalized eigenfunctions. are the nonlocal stress and strain, respectively. the distributed transverse loading induced by the residual surface tension is (farshi et al , 2010): 3 2.0 theory and formulation nonlocal constitutive relation for euler-bernoulli beam is given as (eringen, 1983): 2 2 xx xx xxex         (1) where xx and xx are the nonlocal stress and strain, respectively. the distributed transverse loading induced by the residual surface tension is (farshi et al , 2010): 2 0 2 w q q h x     (2) and h is the surface effect constant is given by: 2 oh d (3) effective flexural rigidity, *ei for nanotube is given by: * 4 4 3 31 ( ) ( ) 4 s o i o iei e r r e r r     (4) the general differential equation of euler-bernoulli beam based on nonlocal continuum model and surface effect are derived using the principle of hamilton and expressed by (reddy, (2007): * 2 2 2 2 2 2 2 2 2 2 2 2 4 0 02 2 2 4 2 22 [ ] w w w w ei n a i x x x x x t t x w w w n a i x x t t w q h q x w h xx                                                             (5) the coordinate system for nanotube is shown in figure 1. figure 1. geometry of nanotube with length l , inner and outer radii ri and ro 3.0 differential transformation method differential transformation method is one of the useful techniques to solve the differential equations with small calculation errors and ability to solve nonlinear equations with boundary conditions value problems. abdel-halim hassan (2002) applied the dtm on eigenvalues and normalized eigenfunctions. effective flexural rigidity, 3 2.0 theory and formulation nonlocal constitutive relation for euler-bernoulli beam is given as (eringen, 1983): 2 2 xx xx xxex         (1) where xx and xx are the nonlocal stress and strain, respectively. the distributed transverse loading induced by the residual surface tension is (farshi et al , 2010): 2 0 2 w q q h x     (2) and h is the surface effect constant is given by: 2 oh d (3) effective flexural rigidity, *ei for nanotube is given by: * 4 4 3 31 ( ) ( ) 4 s o i o iei e r r e r r     (4) the general differential equation of euler-bernoulli beam based on nonlocal continuum model and surface effect are derived using the principle of hamilton and expressed by (reddy, (2007): * 2 2 2 2 2 2 2 2 2 2 2 2 4 0 02 2 2 4 2 22 [ ] w w w w ei n a i x x x x x t t x w w w n a i x x t t w q h q x w h xx                                                             (5) the coordinate system for nanotube is shown in figure 1. figure 1. geometry of nanotube with length l , inner and outer radii ri and ro 3.0 differential transformation method differential transformation method is one of the useful techniques to solve the differential equations with small calculation errors and ability to solve nonlinear equations with boundary conditions value problems. abdel-halim hassan (2002) applied the dtm on eigenvalues and normalized eigenfunctions. for nanotube is given by: 3 2.0 theory and formulation nonlocal constitutive relation for euler-bernoulli beam is given as (eringen, 1983): 2 2 xx xx xxex         (1) where xx and xx are the nonlocal stress and strain, respectively. the distributed transverse loading induced by the residual surface tension is (farshi et al , 2010): 2 0 2 w q q h x     (2) and h is the surface effect constant is given by: 2 oh d (3) effective flexural rigidity, *ei for nanotube is given by: * 4 4 3 31 ( ) ( ) 4 s o i o iei e r r e r r     (4) the general differential equation of euler-bernoulli beam based on nonlocal continuum model and surface effect are derived using the principle of hamilton and expressed by (reddy, (2007): * 2 2 2 2 2 2 2 2 2 2 2 2 4 0 02 2 2 4 2 22 [ ] w w w w ei n a i x x x x x t t x w w w n a i x x t t w q h q x w h xx                                                             (5) the coordinate system for nanotube is shown in figure 1. figure 1. geometry of nanotube with length l , inner and outer radii ri and ro 3.0 differential transformation method differential transformation method is one of the useful techniques to solve the differential equations with small calculation errors and ability to solve nonlinear equations with boundary conditions value problems. abdel-halim hassan (2002) applied the dtm on eigenvalues and normalized eigenfunctions. issn: 2180-1053 vol. 7 no. 1 january june 2015 journal of mechanical engineering and technology 14 the general differential equation of euler-bernoulli beam based on nonlocal continuum model and surface effect are derived using the principle of hamilton and expressed by (reddy, (2007): 3 2.0 theory and formulation nonlocal constitutive relation for euler-bernoulli beam is given as (eringen, 1983): 2 2 xx xx xxex         (1) where xx and xx are the nonlocal stress and strain, respectively. the distributed transverse loading induced by the residual surface tension is (farshi et al , 2010): 2 0 2 w q q h x     (2) and h is the surface effect constant is given by: 2 oh d (3) effective flexural rigidity, *ei for nanotube is given by: * 4 4 3 31 ( ) ( ) 4 s o i o iei e r r e r r     (4) the general differential equation of euler-bernoulli beam based on nonlocal continuum model and surface effect are derived using the principle of hamilton and expressed by (reddy, (2007): * 2 2 2 2 2 2 2 2 2 2 2 2 4 0 02 2 2 4 2 22 [ ] w w w w ei n a i x x x x x t t x w w w n a i x x t t w q h q x w h xx                                                             (5) the coordinate system for nanotube is shown in figure 1. figure 1. geometry of nanotube with length l , inner and outer radii ri and ro 3.0 differential transformation method differential transformation method is one of the useful techniques to solve the differential equations with small calculation errors and ability to solve nonlinear equations with boundary conditions value problems. abdel-halim hassan (2002) applied the dtm on eigenvalues and normalized eigenfunctions. the coordinate system for nanotube is shown in figure 1. 3 2.0 theory and formulation nonlocal constitutive relation for euler-bernoulli beam is given as (eringen, 1983): 2 2 xx xx xxex         (1) where xx and xx are the nonlocal stress and strain, respectively. the distributed transverse loading induced by the residual surface tension is (farshi et al , 2010): 2 0 2 w q q h x     (2) and h is the surface effect constant is given by: 2 oh d (3) effective flexural rigidity, *ei for nanotube is given by: * 4 4 3 31 ( ) ( ) 4 s o i o iei e r r e r r     (4) the general differential equation of euler-bernoulli beam based on nonlocal continuum model and surface effect are derived using the principle of hamilton and expressed by (reddy, (2007): * 2 2 2 2 2 2 2 2 2 2 2 2 4 0 02 2 2 4 2 22 [ ] w w w w ei n a i x x x x x t t x w w w n a i x x t t w q h q x w h xx                                                             (5) the coordinate system for nanotube is shown in figure 1. figure 1. geometry of nanotube with length l , inner and outer radii ri and ro 3.0 differential transformation method differential transformation method is one of the useful techniques to solve the differential equations with small calculation errors and ability to solve nonlinear equations with boundary conditions value problems. abdel-halim hassan (2002) applied the dtm on eigenvalues and normalized eigenfunctions. figure 1. geometry of nanotube with length l , inner and outer radii ri and ro 3.0 differential transformation method differential transformation method is one of the useful techniques to solve the differential equations with small calculation errors and ability to solve nonlinear equations with boundary conditions value problems. abdel-halim hassan (2002) applied the dtm on eigenvalues and normalized eigenfunctions. table 1. some of the transformation rules of the one-dimensional dtm 4 table 1. some of the transformation rules of the one-dimensional dtm original function transformed function     ( )f x g x h x      ( )f k g k h k    ( )f x g x   ( )f k g k     ( )f x g x h x     0 ( ) k l f k g k l h l      ( ) n n d g x f x dx     ! ( ) ! k n f k g k n k      nf x x     1 0 k n f k k n k n        table 2. transformed boundary conditions (b.c.) based on dtm x=0 x=l original b.c. transformed b.c. original b.c. transformed b.c. f (0) 0 f[0] 0 f ( ) 0l  0 [ ] 0 k f k    df (0) 0 dx  df[0] 0 dx  df ( ) 0 dx l  0 [ ] 0 k k f k    2 2 (0) 0 dx d f  2 2 [0] 0 dx d f  2 2 ( ) 0 dx d f l    0 1 [ ] 0 k k k f k     3 3 (0) 0 dx d f  3 3 [0] 0 dx d f  3 3 ( ) 0 dx d f l     0 1 2 [ ] 0 k k k k f k      in this method, certain transformation rules are applied to both the governing differential equations of motion and the boundary conditions of the system in order to transform them into a set of algebraic equations as presented in table 1 and table 2. the solution of these algebraic equations gives the desired results of the problem. the basic definitions and the application procedure of this method can be introduced as follows. the transformation equation of function ( )f x can be defined as (chen et al 2004),   0 1 ( ) ( ) ! k x xk d f x f k k dx   (6) where ( )f x is the original function and [ ]f k is the transformed function. the inverse transformation is defined as:    0 0 ( ) k k f x x x f k     (7) combining equations (6) and (7) one obtains:     0 0 0 ( ) ( ) ! kk x xk k d f xx x f x k dx       (8) in actual application, the function f(x) is expressed by a finite series and equation (8) can be written as follows: issn: 2180-1053 vol. 7 no. 1 january june 2015 investigating the surface elasticity and tension effects on critical buckling behaviour of nanotubes based on differential transformation method 15 table 2. transformed boundary conditions (b.c.) based on dtm 4 table 1. some of the transformation rules of the one-dimensional dtm original function transformed function     ( )f x g x h x      ( )f k g k h k    ( )f x g x   ( )f k g k     ( )f x g x h x     0 ( ) k l f k g k l h l      ( ) n n d g x f x dx     ! ( ) ! k n f k g k n k      nf x x     1 0 k n f k k n k n        table 2. transformed boundary conditions (b.c.) based on dtm x=0 x=l original b.c. transformed b.c. original b.c. transformed b.c. f (0) 0 f[0] 0 f ( ) 0l  0 [ ] 0 k f k    df (0) 0 dx  df[0] 0 dx  df ( ) 0 dx l  0 [ ] 0 k k f k    2 2 (0) 0 dx d f  2 2 [0] 0 dx d f  2 2 ( ) 0 dx d f l    0 1 [ ] 0 k k k f k     3 3 (0) 0 dx d f  3 3 [0] 0 dx d f  3 3 ( ) 0 dx d f l     0 1 2 [ ] 0 k k k k f k      in this method, certain transformation rules are applied to both the governing differential equations of motion and the boundary conditions of the system in order to transform them into a set of algebraic equations as presented in table 1 and table 2. the solution of these algebraic equations gives the desired results of the problem. the basic definitions and the application procedure of this method can be introduced as follows. the transformation equation of function ( )f x can be defined as (chen et al 2004),   0 1 ( ) ( ) ! k x xk d f x f k k dx   (6) where ( )f x is the original function and [ ]f k is the transformed function. the inverse transformation is defined as:    0 0 ( ) k k f x x x f k     (7) combining equations (6) and (7) one obtains:     0 0 0 ( ) ( ) ! kk x xk k d f xx x f x k dx       (8) in actual application, the function f(x) is expressed by a finite series and equation (8) can be written as follows: in this method, certain transformation rules are applied to both the governing differential equations of motion and the boundary conditions of the system in order to transform them into a set of algebraic equations as presented in table 1 and table 2. the solution of these algebraic equations gives the desired results of the problem. the basic definitions and the application procedure of this method can be introduced as follows. the transformation equation of function can be defined as (chen et al 2004), 4 table 1. some of the transformation rules of the one-dimensional dtm original function transformed function     ( )f x g x h x      ( )f k g k h k    ( )f x g x   ( )f k g k     ( )f x g x h x     0 ( ) k l f k g k l h l      ( ) n n d g x f x dx     ! ( ) ! k n f k g k n k      nf x x     1 0 k n f k k n k n        table 2. transformed boundary conditions (b.c.) based on dtm x=0 x=l original b.c. transformed b.c. original b.c. transformed b.c. f (0) 0 f[0] 0 f ( ) 0l  0 [ ] 0 k f k    df (0) 0 dx  df[0] 0 dx  df ( ) 0 dx l  0 [ ] 0 k k f k    2 2 (0) 0 dx d f  2 2 [0] 0 dx d f  2 2 ( ) 0 dx d f l    0 1 [ ] 0 k k k f k     3 3 (0) 0 dx d f  3 3 [0] 0 dx d f  3 3 ( ) 0 dx d f l     0 1 2 [ ] 0 k k k k f k      in this method, certain transformation rules are applied to both the governing differential equations of motion and the boundary conditions of the system in order to transform them into a set of algebraic equations as presented in table 1 and table 2. the solution of these algebraic equations gives the desired results of the problem. the basic definitions and the application procedure of this method can be introduced as follows. the transformation equation of function ( )f x can be defined as (chen et al 2004),   0 1 ( ) ( ) ! k x xk d f x f k k dx   (6) where ( )f x is the original function and [ ]f k is the transformed function. the inverse transformation is defined as:    0 0 ( ) k k f x x x f k     (7) combining equations (6) and (7) one obtains:     0 0 0 ( ) ( ) ! kk x xk k d f xx x f x k dx       (8) in actual application, the function f(x) is expressed by a finite series and equation (8) can be written as follows: in actual application, the function f(x) is expressed by a finite series and equation (8) can be written as follows: 5     0 0 0 ( ) ( ) ! kk x x n k k d f xx x f x k dx     (9) which implies that the term in relation (9) is negligible:     0 0 1 ( ) ( ) ! kk x xk k n d f xx x f x k dx        (10) 3.1 implementation of differential transform method while solving the equation (5) authors preferred dtm approach which avoids solving complicated transcendental algebraic equations for general boundary conditions. in order to derive differential form of equation (5) we refer table 1 and the following expression is written as:  * ( 4)! ( 2)!( ) [ 4] [ 2] 0 ! ! k k ei h n w k n h w k k k            (11) and the various boundary condition for nanotube by using table 2 can be expressed as:  simply supported–simply supported:    0 0 , w 2 0w   (12) 0 0 [ ] 0 , ( 1) [ ] 0 k k w k k k w k          clamped–clamped:    w 0 0 , w 1 0  (13) 0 0 [ ] 0 , [ ] 0 k k w k k w k         clamped–simply supported:    w 0 0 , w 1 0  (14) 0 0 [ ] 0 , ( 1) [ ] 0 k k w k k k w k         by using equation(11) and with the transformed boundary conditions one arrives at the following eigenvalue problem:  11 12 21 22 ( ) ( ) 0 ( ) ( ) a n a n c a n a n        (15) where c correspond to the missing boundary conditions at x=0. for the non-trivial solutions of equation(15), it is necessary that the determinant of the coefficient matrix is equal to zero: 11 12 21 22 ( ) ( ) 0 ( ) ( ) a n a n a n a n  (16) issn: 2180-1053 vol. 7 no. 1 january june 2015 journal of mechanical engineering and technology 16 3.1 implementation of differential transform method while solving the equation (5) authors preferred dtm approach which avoids solving complicated transcendental algebraic equations for general boundary conditions. in order to derive differential form of equation (5) we refer table 1 and the following expression is written as: 5     0 0 0 ( ) ( ) ! kk x x n k k d f xx x f x k dx     (9) which implies that the term in relation (9) is negligible:     0 0 1 ( ) ( ) ! kk x xk k n d f xx x f x k dx        (10) 3.1 implementation of differential transform method while solving the equation (5) authors preferred dtm approach which avoids solving complicated transcendental algebraic equations for general boundary conditions. in order to derive differential form of equation (5) we refer table 1 and the following expression is written as:  * ( 4)! ( 2)!( ) [ 4] [ 2] 0 ! ! k k ei h n w k n h w k k k            (11) and the various boundary condition for nanotube by using table 2 can be expressed as:  simply supported–simply supported:    0 0 , w 2 0w   (12) 0 0 [ ] 0 , ( 1) [ ] 0 k k w k k k w k          clamped–clamped:    w 0 0 , w 1 0  (13) 0 0 [ ] 0 , [ ] 0 k k w k k w k         clamped–simply supported:    w 0 0 , w 1 0  (14) 0 0 [ ] 0 , ( 1) [ ] 0 k k w k k k w k         by using equation(11) and with the transformed boundary conditions one arrives at the following eigenvalue problem:  11 12 21 22 ( ) ( ) 0 ( ) ( ) a n a n c a n a n        (15) where c correspond to the missing boundary conditions at x=0. for the non-trivial solutions of equation(15), it is necessary that the determinant of the coefficient matrix is equal to zero: 11 12 21 22 ( ) ( ) 0 ( ) ( ) a n a n a n a n  (16) and the various boundary condition for nanotube by using table 2 can be expressed as: • simply supported–simply supported: 5     0 0 0 ( ) ( ) ! kk x x n k k d f xx x f x k dx     (9) which implies that the term in relation (9) is negligible:     0 0 1 ( ) ( ) ! kk x xk k n d f xx x f x k dx        (10) 3.1 implementation of differential transform method while solving the equation (5) authors preferred dtm approach which avoids solving complicated transcendental algebraic equations for general boundary conditions. in order to derive differential form of equation (5) we refer table 1 and the following expression is written as:  * ( 4)! ( 2)!( ) [ 4] [ 2] 0 ! ! k k ei h n w k n h w k k k            (11) and the various boundary condition for nanotube by using table 2 can be expressed as:  simply supported–simply supported:    0 0 , w 2 0w   (12) 0 0 [ ] 0 , ( 1) [ ] 0 k k w k k k w k          clamped–clamped:    w 0 0 , w 1 0  (13) 0 0 [ ] 0 , [ ] 0 k k w k k w k         clamped–simply supported:    w 0 0 , w 1 0  (14) 0 0 [ ] 0 , ( 1) [ ] 0 k k w k k k w k         by using equation(11) and with the transformed boundary conditions one arrives at the following eigenvalue problem:  11 12 21 22 ( ) ( ) 0 ( ) ( ) a n a n c a n a n        (15) where c correspond to the missing boundary conditions at x=0. for the non-trivial solutions of equation(15), it is necessary that the determinant of the coefficient matrix is equal to zero: 11 12 21 22 ( ) ( ) 0 ( ) ( ) a n a n a n a n  (16) • clamped–clamped: 5     0 0 0 ( ) ( ) ! kk x x n k k d f xx x f x k dx     (9) which implies that the term in relation (9) is negligible:     0 0 1 ( ) ( ) ! kk x xk k n d f xx x f x k dx        (10) 3.1 implementation of differential transform method while solving the equation (5) authors preferred dtm approach which avoids solving complicated transcendental algebraic equations for general boundary conditions. in order to derive differential form of equation (5) we refer table 1 and the following expression is written as:  * ( 4)! ( 2)!( ) [ 4] [ 2] 0 ! ! k k ei h n w k n h w k k k            (11) and the various boundary condition for nanotube by using table 2 can be expressed as:  simply supported–simply supported:    0 0 , w 2 0w   (12) 0 0 [ ] 0 , ( 1) [ ] 0 k k w k k k w k          clamped–clamped:    w 0 0 , w 1 0  (13) 0 0 [ ] 0 , [ ] 0 k k w k k w k         clamped–simply supported:    w 0 0 , w 1 0  (14) 0 0 [ ] 0 , ( 1) [ ] 0 k k w k k k w k         by using equation(11) and with the transformed boundary conditions one arrives at the following eigenvalue problem:  11 12 21 22 ( ) ( ) 0 ( ) ( ) a n a n c a n a n        (15) where c correspond to the missing boundary conditions at x=0. for the non-trivial solutions of equation(15), it is necessary that the determinant of the coefficient matrix is equal to zero: 11 12 21 22 ( ) ( ) 0 ( ) ( ) a n a n a n a n  (16) • clamped–simply supported: 5     0 0 0 ( ) ( ) ! kk x x n k k d f xx x f x k dx     (9) which implies that the term in relation (9) is negligible:     0 0 1 ( ) ( ) ! kk x xk k n d f xx x f x k dx        (10) 3.1 implementation of differential transform method while solving the equation (5) authors preferred dtm approach which avoids solving complicated transcendental algebraic equations for general boundary conditions. in order to derive differential form of equation (5) we refer table 1 and the following expression is written as:  * ( 4)! ( 2)!( ) [ 4] [ 2] 0 ! ! k k ei h n w k n h w k k k            (11) and the various boundary condition for nanotube by using table 2 can be expressed as:  simply supported–simply supported:    0 0 , w 2 0w   (12) 0 0 [ ] 0 , ( 1) [ ] 0 k k w k k k w k          clamped–clamped:    w 0 0 , w 1 0  (13) 0 0 [ ] 0 , [ ] 0 k k w k k w k         clamped–simply supported:    w 0 0 , w 1 0  (14) 0 0 [ ] 0 , ( 1) [ ] 0 k k w k k k w k         by using equation(11) and with the transformed boundary conditions one arrives at the following eigenvalue problem:  11 12 21 22 ( ) ( ) 0 ( ) ( ) a n a n c a n a n        (15) where c correspond to the missing boundary conditions at x=0. for the non-trivial solutions of equation(15), it is necessary that the determinant of the coefficient matrix is equal to zero: 11 12 21 22 ( ) ( ) 0 ( ) ( ) a n a n a n a n  (16) by using equation(11) and with the transformed boundary conditions one arrives at the following eigenvalue problem: 5     0 0 0 ( ) ( ) ! kk x x n k k d f xx x f x k dx     (9) which implies that the term in relation (9) is negligible:     0 0 1 ( ) ( ) ! kk x xk k n d f xx x f x k dx        (10) 3.1 implementation of differential transform method while solving the equation (5) authors preferred dtm approach which avoids solving complicated transcendental algebraic equations for general boundary conditions. in order to derive differential form of equation (5) we refer table 1 and the following expression is written as:  * ( 4)! ( 2)!( ) [ 4] [ 2] 0 ! ! k k ei h n w k n h w k k k            (11) and the various boundary condition for nanotube by using table 2 can be expressed as:  simply supported–simply supported:    0 0 , w 2 0w   (12) 0 0 [ ] 0 , ( 1) [ ] 0 k k w k k k w k          clamped–clamped:    w 0 0 , w 1 0  (13) 0 0 [ ] 0 , [ ] 0 k k w k k w k         clamped–simply supported:    w 0 0 , w 1 0  (14) 0 0 [ ] 0 , ( 1) [ ] 0 k k w k k k w k         by using equation(11) and with the transformed boundary conditions one arrives at the following eigenvalue problem:  11 12 21 22 ( ) ( ) 0 ( ) ( ) a n a n c a n a n        (15) where c correspond to the missing boundary conditions at x=0. for the non-trivial solutions of equation(15), it is necessary that the determinant of the coefficient matrix is equal to zero: 11 12 21 22 ( ) ( ) 0 ( ) ( ) a n a n a n a n  (16) where correspond to the missing boundary conditions at x=0. for the non-trivial solutions of equation(15), it is necessary that the determinant of the coefficient matrix is equal to zero: issn: 2180-1053 vol. 7 no. 1 january june 2015 investigating the surface elasticity and tension effects on critical buckling behaviour of nanotubes based on differential transformation method 17 5     0 0 0 ( ) ( ) ! kk x x n k k d f xx x f x k dx     (9) which implies that the term in relation (9) is negligible:     0 0 1 ( ) ( ) ! kk x xk k n d f xx x f x k dx        (10) 3.1 implementation of differential transform method while solving the equation (5) authors preferred dtm approach which avoids solving complicated transcendental algebraic equations for general boundary conditions. in order to derive differential form of equation (5) we refer table 1 and the following expression is written as:  * ( 4)! ( 2)!( ) [ 4] [ 2] 0 ! ! k k ei h n w k n h w k k k            (11) and the various boundary condition for nanotube by using table 2 can be expressed as:  simply supported–simply supported:    0 0 , w 2 0w   (12) 0 0 [ ] 0 , ( 1) [ ] 0 k k w k k k w k          clamped–clamped:    w 0 0 , w 1 0  (13) 0 0 [ ] 0 , [ ] 0 k k w k k w k         clamped–simply supported:    w 0 0 , w 1 0  (14) 0 0 [ ] 0 , ( 1) [ ] 0 k k w k k k w k         by using equation(11) and with the transformed boundary conditions one arrives at the following eigenvalue problem:  11 12 21 22 ( ) ( ) 0 ( ) ( ) a n a n c a n a n        (15) where c correspond to the missing boundary conditions at x=0. for the non-trivial solutions of equation(15), it is necessary that the determinant of the coefficient matrix is equal to zero: 11 12 21 22 ( ) ( ) 0 ( ) ( ) a n a n a n a n  (16) solution of equation (16) is simply a polynomial root finding problem. many techniques such as newton’s method, laguerre’s method, etc. can be used to find the roots of this equation. 4.0 results and discussions a nanotube with circular cross section and two different materials, aluminium and silicon, are considered. the elastic bulk and surface properties of aluminium with crystallographic direction of [1 1 1] and silicon with crystallographic direction of [1 0 0] are tabulated in table 3. the nondimensional buckling load is defined as: 6 solution of equation (16) is simply a polynomial root finding problem. many techniques such as newton’s method, laguerre’s method, etc. can be used to find the roots of this equation. 4.0 results and discussions a nanotube with circular cross section and two different materials, aluminium and silicon, are considered. the elastic bulk and surface properties of aluminium with crystallographic direction of [1 1 1] and silicon with crystallographic direction of [1 0 0] are tabulated in table 3. the nondimensional buckling load is defined as: table 4 presents the first critical buckling load where the length-to-thickness ratio is 10 while varying the nonlocal parameter. as can be noted, the obtained results are in good agreement with those of reddy (2007) and even more conservative than those presented by thai (2012). the critical buckling load decreases as the nonlocal parameter increases. this emphasizes the significance of the nonlocal effect on the buckling response of beams. table 3. material properties of al and si mater ial e(gpa) ρ(kg/m3)  es(n/m) τo(n/m) al 70 2700 0.3 5.1882 0.9108 si 210 2370 0.24 -10.6543 0.6048 table 4. the nondimensional buckling load for simply supported beam l/h µ thai[19] reddy[14] present 10 0 9.8696 9.8696 9.86960440 1 8.9830 8.9830 8.98301623 2 8.2426 8.2426 8.24258361 3 7.6149 7.6149 7.61491765 4 7.0761 7.0761 7.07607999 the effects of nonlocal effect (ne), nonlocal parameter and nonlocal surface effect (nse) on the first three nondimensional buckling loads with different boundary conditions are presented in table 5. it should be noted that 0  corresponds to local beam theory. from obtained results, it can be deduced that, when the nonlocal parameter increases, the buckling load decrease. figure 2,3 and 4 depict the variation of the normalized nondimensional buckling load versus the nanotube length for three boundary conditions, i.e. simply–simply (s–s), clamped–simply (c–s) and clamped– clamped (c–c). it can be noted, buckling load ratio is decreased with increasing in the beam size. but, the increasing in nonlocality parameter leads to increase the buckling ratio for the same beam size.  2 *cr ln n ei (17) table 4 presents the first critical buckling load where the length-tothickness ratio is 10 while varying the nonlocal parameter. as can be noted, the obtained results are in good agreement with those of reddy (2007) and even more conservative than those presented by thai (2012). the critical buckling load decreases as the nonlocal parameter increases. this emphasizes the significance of the nonlocal effect on the buckling response of beams. table 3. material properties of al and si 6 solution of equation (16) is simply a polynomial root finding problem. many techniques such as newton’s method, laguerre’s method, etc. can be used to find the roots of this equation. 4.0 results and discussions a nanotube with circular cross section and two different materials, aluminium and silicon, are considered. the elastic bulk and surface properties of aluminium with crystallographic direction of [1 1 1] and silicon with crystallographic direction of [1 0 0] are tabulated in table 3. the nondimensional buckling load is defined as: table 4 presents the first critical buckling load where the length-to-thickness ratio is 10 while varying the nonlocal parameter. as can be noted, the obtained results are in good agreement with those of reddy (2007) and even more conservative than those presented by thai (2012). the critical buckling load decreases as the nonlocal parameter increases. this emphasizes the significance of the nonlocal effect on the buckling response of beams. table 3. material properties of al and si mater ial e(gpa) ρ(kg/m3)  es(n/m) τo(n/m) al 70 2700 0.3 5.1882 0.9108 si 210 2370 0.24 -10.6543 0.6048 table 4. the nondimensional buckling load for simply supported beam l/h µ thai[19] reddy[14] present 10 0 9.8696 9.8696 9.86960440 1 8.9830 8.9830 8.98301623 2 8.2426 8.2426 8.24258361 3 7.6149 7.6149 7.61491765 4 7.0761 7.0761 7.07607999 the effects of nonlocal effect (ne), nonlocal parameter and nonlocal surface effect (nse) on the first three nondimensional buckling loads with different boundary conditions are presented in table 5. it should be noted that 0  corresponds to local beam theory. from obtained results, it can be deduced that, when the nonlocal parameter increases, the buckling load decrease. figure 2,3 and 4 depict the variation of the normalized nondimensional buckling load versus the nanotube length for three boundary conditions, i.e. simply–simply (s–s), clamped–simply (c–s) and clamped– clamped (c–c). it can be noted, buckling load ratio is decreased with increasing in the beam size. but, the increasing in nonlocality parameter leads to increase the buckling ratio for the same beam size.  2 *cr ln n ei (17) table 4. the nondimensional buckling load for simply supported beam 6 solution of equation (16) is simply a polynomial root finding problem. many techniques such as newton’s method, laguerre’s method, etc. can be used to find the roots of this equation. 4.0 results and discussions a nanotube with circular cross section and two different materials, aluminium and silicon, are considered. the elastic bulk and surface properties of aluminium with crystallographic direction of [1 1 1] and silicon with crystallographic direction of [1 0 0] are tabulated in table 3. the nondimensional buckling load is defined as: table 4 presents the first critical buckling load where the length-to-thickness ratio is 10 while varying the nonlocal parameter. as can be noted, the obtained results are in good agreement with those of reddy (2007) and even more conservative than those presented by thai (2012). the critical buckling load decreases as the nonlocal parameter increases. this emphasizes the significance of the nonlocal effect on the buckling response of beams. table 3. material properties of al and si mater ial e(gpa) ρ(kg/m3)  es(n/m) τo(n/m) al 70 2700 0.3 5.1882 0.9108 si 210 2370 0.24 -10.6543 0.6048 table 4. the nondimensional buckling load for simply supported beam l/h µ thai[19] reddy[14] present 10 0 9.8696 9.8696 9.86960440 1 8.9830 8.9830 8.98301623 2 8.2426 8.2426 8.24258361 3 7.6149 7.6149 7.61491765 4 7.0761 7.0761 7.07607999 the effects of nonlocal effect (ne), nonlocal parameter and nonlocal surface effect (nse) on the first three nondimensional buckling loads with different boundary conditions are presented in table 5. it should be noted that 0  corresponds to local beam theory. from obtained results, it can be deduced that, when the nonlocal parameter increases, the buckling load decrease. figure 2,3 and 4 depict the variation of the normalized nondimensional buckling load versus the nanotube length for three boundary conditions, i.e. simply–simply (s–s), clamped–simply (c–s) and clamped– clamped (c–c). it can be noted, buckling load ratio is decreased with increasing in the beam size. but, the increasing in nonlocality parameter leads to increase the buckling ratio for the same beam size.  2 *cr ln n ei (17) issn: 2180-1053 vol. 7 no. 1 january june 2015 journal of mechanical engineering and technology 18 the effects of nonlocal effect (ne), nonlocal parameter and nonlocal surface effect (nse) on the first three nondimensional buckling loads with different boundary conditions are presented in table 5. it should be noted that 6 solution of equation (16) is simply a polynomial root finding problem. many techniques such as newton’s method, laguerre’s method, etc. can be used to find the roots of this equation. 4.0 results and discussions a nanotube with circular cross section and two different materials, aluminium and silicon, are considered. the elastic bulk and surface properties of aluminium with crystallographic direction of [1 1 1] and silicon with crystallographic direction of [1 0 0] are tabulated in table 3. the nondimensional buckling load is defined as: table 4 presents the first critical buckling load where the length-to-thickness ratio is 10 while varying the nonlocal parameter. as can be noted, the obtained results are in good agreement with those of reddy (2007) and even more conservative than those presented by thai (2012). the critical buckling load decreases as the nonlocal parameter increases. this emphasizes the significance of the nonlocal effect on the buckling response of beams. table 3. material properties of al and si mater ial e(gpa) ρ(kg/m3)  es(n/m) τo(n/m) al 70 2700 0.3 5.1882 0.9108 si 210 2370 0.24 -10.6543 0.6048 table 4. the nondimensional buckling load for simply supported beam l/h µ thai[19] reddy[14] present 10 0 9.8696 9.8696 9.86960440 1 8.9830 8.9830 8.98301623 2 8.2426 8.2426 8.24258361 3 7.6149 7.6149 7.61491765 4 7.0761 7.0761 7.07607999 the effects of nonlocal effect (ne), nonlocal parameter and nonlocal surface effect (nse) on the first three nondimensional buckling loads with different boundary conditions are presented in table 5. it should be noted that 0  corresponds to local beam theory. from obtained results, it can be deduced that, when the nonlocal parameter increases, the buckling load decrease. figure 2,3 and 4 depict the variation of the normalized nondimensional buckling load versus the nanotube length for three boundary conditions, i.e. simply–simply (s–s), clamped–simply (c–s) and clamped– clamped (c–c). it can be noted, buckling load ratio is decreased with increasing in the beam size. but, the increasing in nonlocality parameter leads to increase the buckling ratio for the same beam size.  2 *cr ln n ei (17) corresponds to local beam theory. from obtained results, it can be deduced that, when the nonlocal parameter increases, the buckling load decrease. figure 2,3 and 4 depict the variation of the normalized nondimensional buckling load versus the nanotube length for three boundary conditions, i.e. simply–simply (s–s), clamped– simply (c–s) and clamped–clamped (c–c). it can be noted, buckling load ratio is decreased with increasing in the beam size. but, the increasing in nonlocality parameter leads to increase the buckling ratio for the same beam size. 7 figure 2. buckling load ratio of the nanotube with various length and nonlocal parameters (s-s) figure 3. buckling load ratio of the nanotube with various length and nonlocal parameters(c-c) figure 2. buckling load ratio of the nanotube with various length and nonlocal parameters (s-s) 7 figure 2. buckling load ratio of the nanotube with various length and nonlocal parameters (s-s) figure 3. buckling load ratio of the nanotube with various length and nonlocal parameters(c-c) figure 3. buckling load ratio of the nanotube with various length and nonlocal parameters(c-c) issn: 2180-1053 vol. 7 no. 1 january june 2015 investigating the surface elasticity and tension effects on critical buckling behaviour of nanotubes based on differential transformation method 19 8 figure 4. buckling load ratio of the nanotube with various length and nonlocal parameters(c-s) table 5. the first three critical buckling loads versus nonlocal parameters for nanotubes with various boundary conditions µ in s-s ne nse(al) nse(si) 0 i=1 9.8696 42.9089 34.2707 i=2 39.4784 72.5177 63.7895 i=3 88.8264 121.8660 113.2280 2 i=1 8.2425 41.2819 32.6437 i=2 22.4976 55.0996 46.4614 i=3 31.9919 65.0312 56.3930 4 i=1 7.0760 40.1154 31.4772 i=2 15.3068 48.3461 39.7079 i=3 19.5092 52.5485 43.9103 µ in s-s ne nse(al) nse(si) 0 i=1 20.1907 53.2300 44.5918 i=2 59.6759 92.7188 84.0806 i=3 118.9000 151.9390 143.3010 2 i=1 14.3828 47.3828 38.7839 i=2 27.2063 60.2456 51.6074 i=3 35.1838 68.2376 59.5994 4 i=1 11.1697 44.2090 35.5708 i=2 17.6192 50.6585 42.0203 figure 4. buckling load ratio of the nanotube with various length and nonlocal parameters(c-s) table 5. the first three critical buckling loads versus nonlocal parameters for nanotubes with various boundary conditions 8 figure 4. buckling load ratio of the nanotube with various length and nonlocal parameters(c-s) table 5. the first three critical buckling loads versus nonlocal parameters for nanotubes with various boundary conditions µ in s-s ne nse(al) nse(si) 0 i=1 9.8696 42.9089 34.2707 i=2 39.4784 72.5177 63.7895 i=3 88.8264 121.8660 113.2280 2 i=1 8.2425 41.2819 32.6437 i=2 22.4976 55.0996 46.4614 i=3 31.9919 65.0312 56.3930 4 i=1 7.0760 40.1154 31.4772 i=2 15.3068 48.3461 39.7079 i=3 19.5092 52.5485 43.9103 µ in s-s ne nse(al) nse(si) 0 i=1 20.1907 53.2300 44.5918 i=2 59.6759 92.7188 84.0806 i=3 118.9000 151.9390 143.3010 2 i=1 14.3828 47.3828 38.7839 i=2 27.2063 60.2456 51.6074 i=3 35.1838 68.2376 59.5994 4 i=1 11.1697 44.2090 35.5708 i=2 17.6192 50.6585 42.0203 issn: 2180-1053 vol. 7 no. 1 january june 2015 journal of mechanical engineering and technology 20 9 i=3 20.6567 53.6960 45.0578 µ in c-s ne nse(al) nse(si) 0 i=1 39.4784 72.5177 63.8795 i=2 80.7629 113.8020 105.1640 i=3 157.914 190.9530 182.3150 2 i=1 22.0603 55.0996 46.4614 i=2 30.8814 63.9207 55.2825 i=3 37.9758 71.0151 62.3769 4 i=1 15.3068 48.3461 39.7079 i=2 19.0906 52.1298 43.4917 i=3 21.5831 54.6224 45.9842 5.0 conclusions in the present study, the buckling behaviour of nanotubes including the effect of surface stress was predicted via linear partial differential equations of motion and related boundary conditions were derived. the nanotubes are considered to be made of al with positive surface elasticity and si with negative surface elasticity. afterward, the differential transformation method as an efficient and accurate numerical tool was applied to solve the linear equations of nanotubes subjected to different boundary conditions. the good agreement between the results of this article and those available in literature validated the presented approach. numerical results demonstrate that the small scale effects play an important role on the buckling behaviour of the nanotube. also, it is observed that increasing the nonlocal parameter increased the buckling ratio of the nanotubes. nomenclatures *ei effective flexural rigidity q distributed transverse loading n critical buckling load a cross sectional area i mass moment of inertia se surface elasticity modulus e elasticity modulus greek symbols  nonlocal parameter  mass density 0 surface tension  poisson’s ratio 5.0 conclusions in the present study, the buckling behaviour of nanotubes including the effect of surface stress was predicted via linear partial differential equations of motion and related boundary conditions were derived. the nanotubes are considered to be made of al with positive surface elasticity and si with negative surface elasticity. afterward, the differential transformation method as an efficient and accurate numerical tool was applied to solve the linear equations of nanotubes subjected to different boundary conditions. the good agreement between the results of this article and those available in literature validated the presented approach. numerical results demonstrate that the small scale effects play an important role on the buckling behaviour of the nanotube. also, it is observed that increasing the nonlocal parameter increased the buckling ratio of the nanotubes. nomenclatures 9 i=3 20.6567 53.6960 45.0578 µ in c-s ne nse(al) nse(si) 0 i=1 39.4784 72.5177 63.8795 i=2 80.7629 113.8020 105.1640 i=3 157.914 190.9530 182.3150 2 i=1 22.0603 55.0996 46.4614 i=2 30.8814 63.9207 55.2825 i=3 37.9758 71.0151 62.3769 4 i=1 15.3068 48.3461 39.7079 i=2 19.0906 52.1298 43.4917 i=3 21.5831 54.6224 45.9842 5.0 conclusions in the present study, the buckling behaviour of nanotubes including the effect of surface stress was predicted via linear partial differential equations of motion and related boundary conditions were derived. the nanotubes are considered to be made of al with positive surface elasticity and si with negative surface elasticity. afterward, the differential transformation method as an efficient and accurate numerical tool was applied to solve the linear equations of nanotubes subjected to different boundary conditions. the good agreement between the results of this article and those available in literature validated the presented approach. numerical results demonstrate that the small scale effects play an important role on the buckling behaviour of the nanotube. also, it is observed that increasing the nonlocal parameter increased the buckling ratio of the nanotubes. nomenclatures *ei effective flexural rigidity q distributed transverse loading n critical buckling load a cross sectional area i mass moment of inertia se surface elasticity modulus e elasticity modulus greek symbols  nonlocal parameter  mass density 0 surface tension  poisson’s ratio issn: 2180-1053 vol. 7 no. 1 january june 2015 investigating the surface elasticity and tension effects on critical buckling behaviour of nanotubes based on differential transformation method 21 references he, l. h., c. w. lim, and b. s. wu. (2004). a continuum model for sizedependent deformation of elastic films of nano-scale thickness. international journal of solids and structures 41(3), 847-857. eringen, a.c. (1983). on differential equations of nonlocal elasticity and solutions of screw dislocation and surface waves. journal of applied physics 54(9), 4703-4710. gurtin, morton e., and a. ian murdoch. (1975). a continuum theory of elastic material surfaces. archive for rational mechanics and analysis 57(4), 291-323. gurtin, m. e., j. weissmüller, and f. larche. (1998). a general theory of curved deformable interfaces in solids at equilibrium. philosophical magazine a 78(5), 1093-1109. asgharifard sharabiani, pouya, and mohammad reza haeri yazdi. (2013). nonlinear free vibrations of functionally graded nanobeams with surface effects. composites part b: engineering 45(1), 581-586. wang, g. f., x. q. feng, and s. w. yu. (2007). surface buckling of a bending microbeam due to surface elasticity. epl (europhysics letters) 77(4) assadi, abbas, and behrooz farshi. (2011)”size-dependent longitudinal and transverse wave propagation in embedded nanotubes with consideration of surface effects.” acta mechanica 222(12), 27-39. park, harold s. (2009). quantifying the size-dependent effect of the residual surface stress on the resonant frequencies of silicon nanowires if finite deformation kinematics are considered. nanotechnology 20(11) hosseini–hashemi, sh, m. fakher, and r. nazemnezhad. (2013). surface effects on free vibration analysis of nanobeams using nonlocal elasticity: a comparison between euler-bernoulli and timoshenko. journal of solid mechanics 5(3) 290-304. eltaher, m. a., f. f. mahmoud, a. e. assie, and e. i. meletis. (2013). coupling effects of nonlocal and surface energy on vibration analysis of nanobeams. applied mathematics and computation 224, 760-774. malekzadeh, parviz, and mohamad shojaee. (2013). surface and nonlocal effects on the nonlinear free vibration of non-uniform nanobeams. composites part b: engineering 52, 84-92. ansari, r., and s. sahmani. (2011). bending behavior and buckling of nanobeams including surface stress effects corresponding to different beam theories. international journal of engineering science 49(11), 12441255. issn: 2180-1053 vol. 7 no. 1 january june 2015 journal of mechanical engineering and technology 22 farshi, behrooz, abbas assadi, and ali alinia-ziazi. (2010). frequency analysis of nanotubes with consideration of surface effects. applied physics letters 96(9). reddy, j. n. (2007). nonlocal theories for bending, buckling and vibration of beams. international journal of engineering science 45(2), 288-307. abdel-halim hassan, i. h. (2002). on solving some eigenvalue problems by using a differential transformation. applied mathematics and computation 127(1), 1-22. chen, cha’o-kuang, and shin-ping ju. (2004). application of differential transformation to transient advective–dispersive transport equation. applied mathematics and computation 155(1), 25-38. ogata, shigenobu, ju li, and sidney yip. (2002). ideal pure shear strength of aluminum and copper. science 298(5594), 807-811. zhu, richard, et al. (2006). atomistic calculation of elastic moduli in strained silicon. semiconductor science and technology 21(7). thai, huu-tai. (2012). a nonlocal beam theory for bending, buckling, and vibration of nanobeams. international journal of engineering science 52, 56-64. preparation of papers in a two column model paper format issn: 2180-1053 vol. 8 no.1 january – june 2016 1 a study on the tensile test properties of medium carbon steel specimens under specific manufacturing conditions m. riaz 1* , n. atiqah 2 1,2 department of mechanical and manufacturing engineering, universiti kuala lumpur malaysia france institute, jalan teras jernang, section 14, bangi, 43650 selangor, malaysia abstract the experimental process is a fundamental technique used to determine the nature and behavior of many materials under study. in order to evaluate the fundamental properties of many engineering materials the use of mechanical testing techniques frequently play a crucial role. the development of new materials and the control of material quality are essential in the process of design and in their usage for industrial applications and construction. in this experiment, there were 2 sets (5 samples each) of medium type steel specimens s45c of 0.45% carbon content that were rigorously prepared according to astm standards under specific varied parameters. these parameters include the machining cutter speeds on the lathe machine and the mediums selected and temperatures set for the quenching process on the steel specimens in order to affect their overall microstructure. the specimens were then subjected to aggregate mechanical loading using a conventional tensile testing machine. the consequential effect of material structure metamorphosis under the selected quenching techniques was ultimately not part of the study as it involves microscopic analysis of grain boundaries and high-end precision equipment would be necessary to do any significant material analysis. the results of the experiment suggest that the macroscopic (not microscopic) effect of the lathe machining speeds do not significantly affect the tensile material strength of the s45 medium carbon steel specimen. thus, the mediums and temperatures selected for the quenching process on the specimens had a marginal but significant effect on the elevated levels of tensile mechanical strength and strain of medium type carbon steel. keywords: tensile test; medium carbon steel; manufacturing conditions; stress; strain 1.0 introduction the importance of tensile testing in relation to mechanical material properties is fundamental and rudimentary for most applications. the nature of experimentation would hinge on details such as the material used, the size of the tensile specimen, the type of cutting tools used, the machining cutter speed, the type of milling machine and the medium of quenching employed on the specimen. * corresponding author e-mail: mriaz08@gmail.com journal of mechanical engineering and technology 2 issn: 2180-1053 vol. 8 no.1 january – june 2016 1.1 medium carbon steel s45c in most industrial manufacturing applications, plain carbon steels are widely used because of their low cost and ease of fabrication (smith & hashemi, 2006). they are classified on the basis of their carbon content as their major alloying element is carbon. the carbon content of high carbon steels normally range above 0.65% (see figure 1). medium carbon steels can be heat treated to have a good balance of ductility and strength. hardness and other mechanical properties of plain carbon steels typically increase when the content of carbon dissolved in the austenite phase prior to quenching during hardening and heat treatment (anand b. deshpande, 2010) is transformed from austenite into martensite (feng & tahir, 2008), (grishin & huryukin, 1996; kulikov, 1997; tolstousov & bannykh, 1999). figure 1. chemical composition of carbon steel (s45c) 1.2 tensile testing the tensile properties of a material are indicative of how the material will react while being subjected to forces in tension. the tensile test is a fundamental mechanical test where a carefully prepared specimen is loaded in a controlled environment or mechanism while measuring the applied load and the elongation of the specimen over some predetermined distance. among the more common parameters available for measurement during tensile testing include the modulus of elasticity, elastic limit, elongation, proportional limit, reduction in cross-sectional area, tensile strength, yield point, and yield strength. a study on the tensile test properties of medium carbon steel specimens under specific manufacturing conditions issn: 2180-1053 vol. 8 no.1 january – june 2016 3 a tensile specimen is a standardized sample of cross-section with two shoulders and a gauge (section) in between them that are larger so that it can readily gripped, whereas the gauge section has a smaller cross-section so that the deformation and failure can occur in that predetermined area. tensile tests are performed for several reasons. the results of tensile tests can be used in the selection of materials for engineering applications and design, to enhance quality, or even to predict the behavior of a material under specific forms of loading other than uniaxial tension. the strength of a material is often of primary concern, measured in terms of either the stress necessary to cause appreciable plastic deformation or the maximum quantum of stress that the material can withstand. also of interest is the material’s ductility, which is a measure of how much it can be deformed before it fractures. ductility is directly incorporated into design and material specification to ensure quality and toughness. the fracture of a material measures how much of a load a material can take before it breaks when it is in the process of being stretched for low ductility in a tensile test and is often accompanied by low resistance (yunkai lu, 2002). in figure 2 the graph shows the strain versus stress relationship under tensile loading of a typical medium carbon steel specimen at normal length before tensile test until the specimen fractures or breaks. figure 2. strain versus stress under tensile testing 1.3 tensile testing machine and test specimen a testing machine (see figure 3) also known as a universal tester, materials testing machine or materials test frame, is frequently used to test the tensile test and compressive strength of materials. it is named after the fact that it can perform many journal of mechanical engineering and technology 4 issn: 2180-1053 vol. 8 no.1 january – june 2016 standard tensile and compression tests on materials, components, and structures. the set-up and usage of the machine is often detailed in a procedure that outlines sample preparation, fixturing, gauge length (the length which is under study or observation), analysis, etc. the specimen is placed in the machine between the grips and an extensometer (if required) can automatically record the change in gauge length during the test. if an extensometer is not fitted, the machine itself can record the displacement between its cross heads on which the specimen is held. however, this method not only records the change in length of the specimen but also all other extending or resulting elastic components of the testing machine and its drive systems, including any slipping of the specimen in the grips. once the machine is started it begins to apply an increasing amount of load on the specimen. throughout the tests the control system and its associated software record the load and extension (or compression) of the specimen. machines of similar type and function range from very small table top systems to ones that can enforce loading capacity in the range of 53,000 kn. . figure 3. tensile testing machine a study on the tensile test properties of medium carbon steel specimens under specific manufacturing conditions issn: 2180-1053 vol. 8 no.1 january – june 2016 5 in figure 4, a cross section of a test sample specimen is shown, emphasizing the limited flexibility of the machine to test objects that do not follow standardized dimensions. the alphanumeric designated dimensions can be found in the respective machine’s standardized manuals for sample preparation. figure 4. test specimen 2.0 methodology in the experiment, the tensile test specimens are initially fabricated using the ramo 33 conventional lathe machine and later tested using the vew 2302 universal tensile test machine. the specimens were fabricated using the astm e8 standard for metallic materials. arrays of specific parameters were carefully selected for mediums of quenching, heat treatments and tensile testing. the drawing of the tensile test specimens was prepared using the catia® (v5r16) software to avoid any occurrences of error during the machining process and to adopt distinct levels of accuracy for the specified length and diameter of 200 mm and 20mm respectively. the shape shown in figure 5 is commonly referred to as the rod/dumbbell type shaped specimen. figure 5. catia ® (v5r16) software journal of mechanical engineering and technology 6 issn: 2180-1053 vol. 8 no.1 january – june 2016 the material selected for the fabrication of the 10 sample test specimens was medium type carbon steel s45 (refer to figure 1 for composition range) since it is widely used and available in industry. the lathe process (see figure 6) were run at a constant cutting speed of 430 rpm and 860 rpm respectively using the widia uncoated micro grain tungsten carbide cutting tool fastened at a 55 degrees fixed angle of insert. figure 6. lathe process (ramo 33) the following aggregated heat treatment process on the samples between the ranges of 200°c to 600°c controls the heating and cooling of the material in order to alter its mechanical properties without physically altering its original dimensions. subsequently, the specimens were quenched for 1 hour in alternating mediums of water and isorapid oil in order to convert a variety of present microstructures from soft and ductile spheroidite to hard and brittle martensite (tensi et al.,1995). the samples were finally subjected to different axial tensions and loads until failure using the vew 2302 (see figure 7) yielding a variety of mechanical properties from elongation and crosssectional areas to the determination of values for young’s modulus, yield strength and strain hardening characteristics. figure 7. tensile test machine (vew 2302) a study on the tensile test properties of medium carbon steel specimens under specific manufacturing conditions issn: 2180-1053 vol. 8 no.1 january – june 2016 7 3.0 results and discussions the results of the tensile tests performed on the rod/dumbbell type shaped specimens on medium carbon steel material and quenched using water as a medium is shown in table 1 below. table 1. results for stress, strain, and percentage elongation (430 rpm & 860 rpm / water quenching) specimen no. machining speed (rpm) temp ( 0 c) stress (mpa) strain (%) elongation (%) 1 430 & 860 200 637 31.050 10.364 2 430 & 860 300 596 28.850 9.2285 3 430 & 860 400 599 28.525 9.8005 4 430 & 860 500 673 27.220 8.9875 5 430 & 860 600 657 30.100 10.467 it is important to note that the lathe machining speeds were alternated between 430 rpm and 860 rpm respectively for each of the 2 sets (5 samples per set) to investigate for the occurrences of data variations between the observed parameters of maximum loading (before failure) for stress, strain, and percentage elongation so as to warrant any documentation. however, there were inconsequential differences to the recorded data when machining speeds were indeed varied and were therefore not reflected in the data shown in table 1 (see also table 2). that allowed for the blanket hypothesis that varied lathe machining speed has no adverse or significant effect on the samples used for tensile testing in the experiment. therefore, focus would primarily be on the varied temperatures and mediums used in the heat treatment process that did in fact yield significant variations for further deliberations. in figure 8 it can be seen that as temperature increases between 200 to 300°c, the stress values indicates an inverse relationship for the material. the drop in stress was 41% while marginally increasing by 0.05% until 400°c. following that, the increase in stress was 12.4% to 500°c, and reduced linearly by 2.4% to 600°c until fracture. these linear relationship variations may give rise to the assumption that the effect of quenching in water only allows for stress to increase significantly from 400 to 500°c, and thereby reducing once reaching that limit as referenced in the sampled specimen material. journal of mechanical engineering and technology 8 issn: 2180-1053 vol. 8 no.1 january – june 2016 figure 8. stress versus temperature in figure 9, the strain follows a steady linear value decline of 3.83% from 200 500 ° c and in stark contrast increases linearly by 2.88% from 500 600°c. thus, the temperature increase from 200°c onwards reduces the stain in the material for an aggregated 300°c block increase only. figure 9. strain versus temperature a jagged linear relationship is shown in figure 10 for the percentage elongation experienced by the specimen sample material after quenching indicating a disproportionate relationship of linear and inversely linear correlations for aggregated temperature increases of 100°c. for example, from 200-300°c, there is a 1.14% decrease that is approximately mirrored by the quantum of 0.813% in the 400-500°c range. however, a linear increase of 0.57% and 1.48% can be seen in the 300-400°c and 500-600°c range. for the 400-600°c range, the averaged fluctuation is 1.15%. therefore, a conclusive linear correlation may not necessarily exist in terms of the recorded percentage elongation data for the material. a study on the tensile test properties of medium carbon steel specimens under specific manufacturing conditions issn: 2180-1053 vol. 8 no.1 january – june 2016 9 figure 10. percent elongation versus temperature the results of the tensile tests performed on the specimens on medium carbon steel material and quenched using isorapid oil as a medium for quenching is in table 2. table 2. results for stress, strain, and percentage elongation (430 rpm & 860 rpm / oil quenching) specimen no. machining speed (rpm) temp ( 0 c) stress (mpa) strain (%) elongation (%) 1 430 & 860 200 653 27.890 8.993 2 430 & 860 300 654 28.150 9.728 3 430 & 860 400 639 29.650 10.305 4 430 & 860 500 524 19.880 7.562 5 430 & 860 600 634 29.120 10.333 in figure 11, the almost horizontal correlation between stress and temperature increase due to quenching in isorapid oil gives the hypothetical assumption that the medium used does not adversely influence the mechanical material properties of the sample specimen. the only noticeable decrease in stress was at 500°c, which may also not be significant should the sample frequency be increased in later trials. figure 11. stress versus temperature journal of mechanical engineering and technology 10 issn: 2180-1053 vol. 8 no.1 january – june 2016 in figure 12, a similar reoccurrence of a horizontal relationship between strain and temperature can be noticed. as with the relationship with stress, the increase in strain due to temperature change after isorapid oil quenching yields stain values in the range of 27.5 – 29.5 % only for a 400°c block of temperature fluctuation. as such, it may not be premature to assume that the selected medium of quenching (isorapid oil) does not adversely influence the mechanical material properties of the sample specimen. once again, the only noticeable decrease in strain was at 500°c (as mentioned earlier, may also not be significant should the sample frequency be increased in later trials). figure 12. strain versus temperature in figure 13, a minor fluctuation in percentage elongation can be seen for quenching in isorapid oil over the 200 600°c temperature range. in relative comparison to figure 9 (quenching in water), the jagged linear relationship is only apparent at 400°c. the relationship remains mainly horizontal in nature from 200 400°c with a 1.31% fluctuation, and the averaged higher jagged increase (and decrease) of approximately 2.75% is significantly higher than the 1.15% average fluctuation noticed for water. figure 13. percent elongation versus temperature a study on the tensile test properties of medium carbon steel specimens under specific manufacturing conditions issn: 2180-1053 vol. 8 no.1 january – june 2016 11 4.0 conclusions in the experiment, a study into the factors affecting the tensile material strength of specimens fabricated under specific lathe machining conditions was successfully undertaken and completed. the results indicate that a change in machining speed (430 to 860 rpm) does not influence the mechanical properties of medium carbon steel in any way since no significant incremental changes were noticed in any of the data recorded. however, noticeable fluctuation in stress, strain and percentage elongation of the material specimen were apparent over the 200 600°c temperature range in both water and isorapid oil that were selected as the mediums of quenching after the heat treatment process. however, a more substantial change in these measured mechanical properties was recorded for the sample material quenched in water as compared to isorapid oil. that leads to the conclusion that heat treatment and quenching in water can be more effective in material hardening and thus significantly affecting its mechanical properties for future analysis and application. references alaneme, k., ranganathan, s., & mojisola, t. (2010). mechanical behaviour of duplex phase structures in a medium carbon low alloy steel. journal of minerals and materials characterization and engineering, 09(07), 621-633. daramola, o., adewuyi, b., & oladele, i. (2010). effects of heat treatment on the mechanical properties of rolled medium carbon steel. journal of minerals and materials characterization and engineering, 09(08), 693-708. hashemi, p., moztarzadeh, f., & hashemi, t. (2002). anodization of electroformed nickel for the manufacture of compact disc stampers. journal of materials science letters, 21(1), 37-40. lu, y. (2002). mechanical properties of random discontinuous fiber composites manufactured from wetlay process (unpublished master's thesis). odusote, j. k., ajiboye, t. k., & rabiu, a. b. (2012). evaluation of mechanical properties of medium carbon steel quenched in water and oil. journal of minerals and materials characterization and engineering, 11(09), 859-862. sales, w. f., diniz, a. e., & machado, á. r. (2001). application of cutting fluids in machining processes. journal of the brazilian society of mechanical sciences, 23(2). senthilkumar, t., & ajiboye, t. k. (2012). effect of heat treatment processes on the mechanical properties of medium carbon steel. journal of minerals and materials characterization and engineering, 11(02), 143-152. journal of mechanical engineering and technology 12 issn: 2180-1053 vol. 8 no.1 january – june 2016 sverdlin, a. v., totten, g. e., bates, c., & jarvis, l. m. (1996). use of the quenching factor for predicting the properties of polymer quenching media. metal science and heat treatment, 38(6), 248-251. tensi, h. (2010). wetting kinematics. quenching theory and technology, second edition. boca raton: crc press. totten, g. e., webster, g. m., & bates, c. e. (1998). cooling curve and quench factor characterization of 2024 and 7075 aluminum bar stock quenched in type 1 polymer quenchants. heat trans res, 29(1-3), 163-175. microsoft word 04_shukri et al .docx journal of mechanical engineering and technology `_______________________________________ *corresponding author. email: mshukriy@utem.edu.my issn 2180-1053 vol. 12 no. 2 30 jun – 31 dec 2020 stiffness of thin-walled chassis reinforced by vertical l-gusset plate under bending and torsional loads m.s. yob1,2, r. junaidi2, n.a. mat tahir2 , o. kurdi 3,4 , m.j. abd latif1,2 1advanced manufacturing centre (amc), universiti teknikal malaysia melaka, hang tuah jaya, 76100, durian tunggal, melaka, malaysia 2faculty of mechanical engineering, universiti teknikal malaysia melaka, hang tuah jaya, 76100, durian tunggal, melaka, malaysia 3departement of mechanical engineering, diponegoro university, semarang, indonesia 4national center of sustainable transportation technology, indonesia abstract: it has been a great concern to increase the efficiency of a vehicle in the automotive industry. typical energy breakdown of the energy consumption reveals that around 22% of the total fuel energy were used, where the rest were loss to the environment. huge reserved of fuel can be made when a fraction of the loss can be minimized. thin-walled chassis is one of the suggested methods in order to reduce weight of a vehicle. however, it was also debated that the beam or shell model might lose its ability to withstand forces acting on it when the critical load reach the highest point. thus, this study intended to analyze the effect of gusset plate reinforcement on chassis towards bending and torsional load. from the study, it was found that the bending stiffness of the chassis reduced about 222.45% and increase about 84.75% for torsional stiffness. hence, it can be concluded that gusset plate reinforcement designed and placed in this study is a good alternative to improve torsional stiffness and not the bending stiffness of thin-walled chassis. keywords: thin-walled chassis; thin-walled beam; bending; torsion; 46 journal of mechanical engineering and technology issn 2180-1053 vol.12 no.2 30 jun – 31 dec 2020 1.0 i n t ro d uc t i o n it is well known that weight is an important feature in designing and fabricating a vehicle in order to improve the performance in automotive industry. the performance that often discussed in passenger car are speed, acceleration, and fuel consumption. there were plenty of research has been conducted in order to increase the performance of vehicle including modification of engine and body design, additional of component to increase the effectiveness of combustion, introduction of new type of lubricant, and reduction of vehicle’s weight [1–6]. despite those improvements, it has to be highlighted that altering or modification of a structure must not compromise the strength and stiffness of a vehicle since these parameters has huge effect on the safety and ride comfort for drivers. this issue has been highlighted by other researchers where simulation software are the simplest possible method to alter the weight of a vehicle structure without jeopardizing the safety of the vehicle [7, 8]. thin-walled chassis is one of the suggested methods in order to reduce weight of a vehicle. however, it was also debated that the beam or shell model might lose its ability to withstand forces acting on it when the critical load reach the highest point [9], [10]. it was also point out that high stress could cause failure on the thin-walled chassis. apart from that, it was also highlighted that acceptable bending and torsional rigidity are also needed for a better handling feature [8, 11–13]. chassis stiffness generally indicates the resistance to bending or flexing meanwhile torsional stiffness indicates the resistance to twisting. solid structure has high resistance to bending and torsion meanwhile thin-walled structure are not so great in resisting these forces. when subjected to compression in flexural bending, axial compression, shear or bending, thin elements may buckle locally at stress level lower that the yield point of steel. thin-walled column are often presents failure such as euler buckling and localize buckling [8]. hence, in order to retain the low weight and not compromising the buckling resistance, this study attempt to introduce additional gusset plate as additional member in order to improve the stiffness of thin-walled chassis for bending and torsional loads. 2.0 methodology this study is focusing on the experimental work to analyze the performance of gusset plate as reinforcement element on thin-walled chassis structure. the 47 journal of mechanical engineering and technology issn 2180-1053 vol.12 no.2 30 jun – 31 dec 2020 test was carried on 2 loading conditions in order to evaluate the strength and stiffness of thin-walled chassis (with and without gusset plate). the relation between stress, stiffness and load for both models and loading condition will be obtained from the test finding. 2.1 preparation of thin-walled chassis for testing in this study, thin-walled beams were used to replicate the thin-walled chassis by using mig technique. for reinforced thin-walled chassis, thin plate was used as gusset plate to reinforce the joint areas. the detail of beam and plate used were shown in table 1 meanwhile the design for both thin-walled chassis (with gusset and without) are shown in figure 1. table 1: detail of beam and plate used for chassis fabrication beam gusset plate materials astm a36 structural steels ultimate strength 400 – 500 mpa yield strength 250 mpa modulus of elasticity 200 gpa bulk modulus 140 gpa thickness 1.2 mm size 50 x 50 mm 100 x 50 mm figure 1: detail dimension for (a) thin-walled chassis and (b) thin-walled chassis with gusset plate 2.2 experimental setup 48 journal of mechanical engineering and technology issn 2180-1053 vol.12 no.2 30 jun – 31 dec 2020 the test was set up in order to test the beam based on bending and torsional test. the experimental setup is shown in figure 2. both with and without gusset plate will be tested at the same condition as shown in figure 3. for bending test, load will be placed on the centre of the chassis meanwhile for the torsion test, the load will be placed at the 1 edge of the chassis. in both conditions, the load will be controlled by a hydraulic jack that was connected with load cell and data logger to record the force exerted. figure 2: experimental setup for the loading condition of the chassis figure 3: bending and torsional loading condition 49 journal of mechanical engineering and technology issn 2180-1053 vol.12 no.2 30 jun – 31 dec 2020 3.0 results and discussion the deflection of the chassis was recorded each time the load increased. the data collected were then plotted into force against displacement as shown in figure 4 and 5. then, best fit lines are plotted for both with or without gusset reinforcement are compared in order to obtain the percentage difference. the differences are shown in table 2. figure 4: graph plotted for bending test figure 5: graph plotted for torsional test 50 journal of mechanical engineering and technology issn 2180-1053 vol.12 no.2 30 jun – 31 dec 2020 table 2: differences between slopes for bending and torsion test type of test differences (%) bending -22.45 torsion 84.78 from figure 4, it can be seen that chassis with no gusset reinforcement has lower displacement compared to chassis with gusset reinforcement. meanwhile, figure 5 shows that chassis with gusset reinforcement was indeed has lower displacement compared with chassis without gusset. when comparing the differences based on the best fit line slopes, it can be said that additional of gusset reinforcement on chassis structure has weaken the structure by 22.45 % (by 0.22 times) in withstanding the bending but increase the resistance in torsion by 84.78% (0.85 times). here, it can be seen that the additional of gusset plate are only effective in withstanding the torsional force and not the bending force. it was believed that the position of the placement of the gusset plate plays an important role in determining the stress distribution along the structure. this was in agreement with other research where the performance of the strengthening scheme depends upon the size of the core material, the presence of bond between the steel plate and core material, and the existence of extra plates for double angle members [14]. it was also discussed and suggested by other researcher that the reinforcement could be in a more 3-dimension structure with combinations of two or more materials [15–19]. 4.0 conclusion in this research work, the effect reinforcement of gusset plate to increase the strength of stiffness of thin-walled chassis under bending and torsional loads were identified. from this study, it was found that the reinforcement of gusset plate weakens the chassis structure by 22.45% for withstanding bending load and increase the integrity by 84.78% for withstanding the torsional load. it was suggested that in the near future, this study can be improved by variates the type and location of gusset placement. 51 journal of mechanical engineering and technology issn 2180-1053 vol.12 no.2 30 jun – 31 dec 2020 acknowledgements the authors would like to thank applied mechanical design laboratory utem for the support throughout this study. the authors gratefully acknowledge utem for the consent granted to access and use the pictures and all journals for this work. references [1] k. t. chau and c. c. chan, “emerging energy-efficient technologies for hybrid electric vehicles,” proc. ieee, vol. 95, no. 4, pp. 821–835, 2007. [2] l. hou, y. lei, and y. fu, “effects of lightweight gear blank on noise , vibration and harshness for electric drive system in electric vehicles,” proc imeche part k j. multi-body dyn., pp. 1–18, 2020. [3] s. kamiya and t. desaki, “technical trend of friction reduction in engine bearings,” tribol. online, vol. 12, no. 3, pp. 89–93, 2017. [4] n. a. mat tahir, m. f. bin abdollah, n. tamaldin, h. amiruddin, and m. r. mohamad zin, “the tribological potential of graphene growth from solid waste,” prog. ind. ecol. an int. j., vol. 13, no. 4, pp. 401–413, 2019. [5] n. a. mat tahir, m. f. b. abdollah, r. hasan, and h. amiruddin, “the effect of sliding distance at different temperatures on the tribological properties of a palm kernel activated carbon-epoxy composite,” tribol. int., vol. 94, pp. 352–359, 2016. [6] m. i. muazzam, f. a. munir, a. roseli, m. s. yob, and r. jumaidin, “numerical simulation of spark ignition engine with pre combustion chamber (pcc),” j. adv. res. fluid mech. therm. sci., vol. 52, no. 2, pp. 198–204, 2018. [7] d. satrijo, o. kurdi, i. haryanto, m. s. yob, n. riyantiarno, and i. taufiqurrahman, “rollover performance analysis of electric bus superstructure frame with alternative material using finite element method,” aip conf. proc., vol. 2217, 2020. [8] m. s. yob, s. mansor, and r. sulaiman, “individual stiffness of 3d space frame thin walled structural joint considering local buckling effect,” appl. mech. mater., vol. 554, pp. 411– 415, 2014. [9] f. m. aydın korucuk, m. maali, m. kılıç, and a. c. aydın, “experimental analysis of the effect of dent variation on the buckling capacity of thin-walled cylindrical shells,” thinwalled struct., vol. 143, pp. 1–11, 2019. [10] m. s. ismail, o. ifayefunmi, and s. h. s. m. fadzullah, “buckling of imperfect cylindercone-cylinder transition under axial compression,” thin-walled struct., vol. 144, 2019. 52 journal of mechanical engineering and technology issn 2180-1053 vol.12 no.2 30 jun – 31 dec 2020 [11] f. duddeck, s. hunkeler, p. lozano, e. wehrle, and d. zeng, “topology optimization for crashworthiness of thin-walled structures under axial impact using hybrid cellular automata,” struct. multidiscip. optim., vol. 54, no. 3, pp. 415–428, 2016. [12] z. you xie, “the reinforcement optimization of thin-walled square tubes for bending crashworthiness,” int. j. crashworthiness, vol. 0, no. 0, pp. 1–7, 2020. [13] g. zhu, z. wang, x. huo, a. cheng, g. li, and c. zhou, “experimental and numerical investigation into axial compressive behaviour of thin-walled structures filled with foams and composite skeleton,” int. j. mech. sci., vol. 122, pp. 104–119, 2017. [14] s. el-tawil and e. ekiz, “inhibiting steel brace buckling using carbon fiber-reinforced polymers: large-scale test,” j. struct. eng., pp. 530–538, 2009. [15] h. nikkhah, v. crupi, and a. baroutaji, “crashworthiness analysis of bio-inspired thinwalled tubes based on morpho wings microstructures,” mech. based des. struct. mach., vol. 0, no. 0, pp. 1–18, 2020. [16] s. verbruggen, d. g. aggelis, t. tysmans, and j. wastiels, “bending of beams externally reinforced with trc and cfrp monitored by dic and ae,” compos. struct., vol. 112, no. 1, pp. 113–121, 2014. [17] j. wang and l. chen, “experimental investigation of extended end plate joints to concretefilled steel tubular columns,” j. constr. steel res., vol. 79, pp. 56–70, 2012. [18] j. wang and s. guo, “structural performance of blind bolted end plate joints to concretefilled thin-walled steel tubular columns,” thin-walled struct., vol. 60, pp. 54–68, 2012. [19] s. tokgoz and c. dundar, “experimental study on steel tubular columns in-filled with plain and steel fiber reinforced concrete,” thin-walled struct., vol. 48, no. 6, pp. 414–422, 2010. 53 issn 2180-1053 e-issn 2289-8123 vol.14 no.2 31 design and construction of baby rocking and monitoring system m. jidda1, a. m. maleka1, m. a. ibrahim1, a. g. ibrahim1 1department of mechatronics and system engineering, abubakar tafawa balewa university, pmb 0248, bauchi, nigeria. corresponding author’s email: 1alifamaleka@gmail.com article history: received 15 april 2021; revised 1 august 2022; accepted 5 december 2022 abstract: this paper presents the design and construction of a system for a busy parent that will monitor the condition of the infant whether crying or not as well as pacify the baby by rocking and regulating the nearby temperature. the system transmits the baby’s cry in real time to a receiving device which is with the parent, a wooden cradle incorporated with a dc motor pacifies the baby when a sound signal is received. a temperature sensor is also used to monitor the nearby temperature and send the signal to microcontroller which actuate the fan on/off as programed. the system could transmit the baby’s cry in real-time up to 40 meters away from the receiving device. the system also oscillates at 15o when the system is loaded up to 12kg and the load has no effect on the frequency of oscillation. keywords: mechanical rocking; wireless transmission; dc motor control 1.0 i n t ro d uc t i o n during the early stages, infants need proper rest and sleep for growth and development. hence, it is the responsibility of the parents/guardian to provide the necessary care and attention to the infant. but with the modern lifestyle, parents are busy and have a lot of work with little time to provide for their little ones [1]. child care is of most extreme significance for a parent. the present quick paced world makes it hard for parent to continuously look after their kid. after long working hours, it is hard for parent to constantly watch out their kid. keeping an eye on child or employing caretaker is an expensive [2]. it may be expensive for the household to afford a nanny and there is a need to develop a low cost system that will help the parents to take care of their baby and reduce their stress in parenting of infant as most of the once availably are imported, expensive and are not designed to journal of mechanical engineering and technology (jmet) 32 issn 2180-1053 e-issn 2289-8123 vol.14 no.2 meet our local demand based on our environment and tradition. the baby rocking and monitoring system is design to detect and pick up infant’s cry which might results due to loneliness of being alone, swing in order to pacify the baby and make him/her quiet or fall asleep as it is naturally proven and observed, transmit the sound of the infant to a monitoring/receiving device that will be with the parent in real time which might hint the parent to feed the baby in case he/she continue crying even when the cradle is swinging and lastly regulate nearby temperature if it becomes of higher degree which can precipitate the baby to cry. rachnapalaskar et al (2016) designed an automatic baby cradle system which was a reliable and efficient baby monitoring system. the system uses sensors for monitoring vital parameters such as detecting baby cry, environmental temperature and mattress moisture. a dc motor was used for cradle movement, cloud was used to store data and a sound buzzer was used for alarm all controlled by a single arduino mega microcontroller [3]. similarly, srivastara et al (2019) proposed a smart cradle system for child monitoring using internet of things (iot) which utilizes the cloud server for monitoring the child. pir, noise and dht sensors were used to detect child’s movement, child’s cry and temperature of the environment. all the data obtained were sent to mobile application using the using iot [2]. likewise, ashok et al (2019) designed an infant cradle monitoring system using iot for monitoring purpose and sensors such as mic, dht, for voice and temperature detection respectively. a dc motor was used to swing the cradle while an apr module was used for recording and playing of songs [1]. srikanth et al (2018) designed and implemented a smart cradle. the system was designed using raspberry pie 3, wet sensor, pir sensor, sound sensor, dc motor, sms module and camera. the system encloses camera monitoring, automatic swinging of cradle when the baby cries, sensing the wetness of baby’s bed and monitoring baby’s presence in the cradle all of which were sent to parent via sms [4]. shastry et al (2017) developed and infant safety smart cradle that uses instrumental sensors such as sound sensor and moisture sensor for detecting sound and wetness of baby diaper. the cradle swings automatically when the baby’s cry was detected, moisture was accurately sensed and data were sent via sms to parent. a live design and construction of baby rocking and monitoring system issn 2180-1053 e-issn 2289-8123 vol.14 no.2 33 streaming was achieved using webcam [5]. joshi and mehetre (2015) described the advancement of baby cradle over period. they observed that baby can get better attention with advanced cradle. although automated cradle cannot handle crying baby at all times as the reason might be hunger [6]. kadu et al (2014) designed an automatic cradle that has inbuilt wet sensor which alarm if the baby wets. the cradle was made to swing using dc motor and a shaft controlled by a microcontroller with three seconds rotation in both clockwise and counter clockwise direction [7]. sambhar & tadwalkar (2017) proposed an automatic cradle movement system having a microphone that detect the baby’s cry and sends it to microcontroller. the microcontroller output is sent to drivers that drives a dc motor and makes the cradle swing. message is sent via gsm to parent when baby’s cry persists and vibration is used to wake up baby [8]. nathan et al (2018) did a survey on digital age smarter cradle system that gives baby comfort and makes sure the baby experiences a sound sleep. the wetness of the baby is detected and the condition is sent via an android application. the baby’s cry is detected through a microphone and exclusive video of the baby is recorded and saved through cloud computing. temperature was measured and the baby’s weight is also known on regular basis [9]. 2.0 methodology 2.1 software development for this project, c programming language is used to write the program and was written on arduino platform. the preparation symbol “start” begins the programming process. the system is then initialized, connecting the various units of the circuit. after this, the sound module detects available sound if any from the baby and then convert it to signal input to the microcontroller which in turn transmit it via the nrf24l01 transmitter. the digital output of the sound module is fed into the motor driver that in turn drives the cradle motor. the program code is attached at the end of the project; figure 2.2 is the flow chat. journal of mechanical engineering and technology (jmet) 34 issn 2180-1053 e-issn 2289-8123 vol.14 no.2 figure 2.1 system block diagram figure 2.2: system flowchart 2.2 hardware development the hardware development is divided into different units which includes 1. power supply/charging unit 2. transmitting unit 3. receiving unit design and construction of baby rocking and monitoring system issn 2180-1053 e-issn 2289-8123 vol.14 no.2 35 4. temperature regulation and display unit 5. mechanical rocking unit 2.2.1 power supply/charging unit this unit consist of a transformer and a bridge rectification section as well as a charging circuit. 2.2.2 rectification in this project, a center-tapped transformer and two diodes were used for full wave rectification. in using the center-tap (c) as a common, the voltage a and b is 180 degrees out of phase. when a is positive, d1 will be forward biased and conduct, while b will be negative thus reverse-biasing d2, while is non-conductive. on the negative half cycle in relation to a when d1 doesn't conduct, d2 will conduct. figure 2.3: charging circuit diagram figure 2.4: rectification journal of mechanical engineering and technology (jmet) 36 issn 2180-1053 e-issn 2289-8123 vol.14 no.2 parameters: transformer type: single phase step down transformer input voltage 220/230v, 50hz output voltage 2x 15v ac current rating output current = 2a filtering: capacitors are used to serve as the filter components of the circuit that will hold the peakto-peak ripples at approximately 5% of the peak voltage. therefore, using the values obtained, it is calculated as follows: 𝐼𝑚𝑖𝑛 = 𝐼𝑑.𝑐 = 500 𝑚𝐴 (𝑚𝑒𝑎𝑠𝑢𝑟𝑒𝑑 𝑎𝑛𝑑 𝑎𝑝𝑝𝑟𝑜𝑥) root mean square voltage 𝑉𝑟𝑚𝑠 = √2 𝑥 𝑉𝑚 (1) and max. voltage 𝑉𝑚 = 24 𝑉𝑚 substituting 𝑉𝑚 into eq. 1 we have 𝑉𝑟𝑚𝑠 = √2 𝑥 𝑉𝑚 = √2 𝑥 24 𝑉𝑟𝑚𝑠 = 33.94𝑉 taking a 5% ripple factor of the peak voltage 𝑉𝑟 = 𝑟𝑖𝑝𝑝𝑙𝑒 𝑓𝑎𝑐𝑡𝑜𝑟 𝑥 𝑉𝑚 (2) 𝑉𝑟 = 𝑅𝐹 𝑥 𝑉𝑚 = 5 100 𝑥 33.94 = 0.05 𝑥 33.94 𝑉𝑟 = 1.697𝑉 using the relationship below to obtain the value of the capacitor 𝐶1. 𝐶1 = 𝐼𝑑𝑐 4√3𝑓𝑉𝑟 (3) where 𝐶1 is the filtering capacitor, f is the frequency of the input supply and 𝐼𝑑𝑐 is the current taken by the load. so: 𝐶1 = 500 𝑥 10−3 4√3 𝑥 50 𝑥 1.697 design and construction of baby rocking and monitoring system issn 2180-1053 e-issn 2289-8123 vol.14 no.2 37 𝐶1 = 850.5 𝜇𝑓 using the idea of 20-30% addition of the calculated value to have a very close standard value obtainable at the market, in this project design, a 20% addition is considered. 𝐶1 𝑎𝑑𝑑𝑖𝑡𝑖𝑜𝑛 = 850.5 𝑥 10 −6 𝑥 20% 𝐶1 𝑎𝑑𝑑𝑖𝑡𝑖𝑜𝑛 = 850.5 𝑥 10 −6 𝑥 20 100 𝐶1 = 170.1 𝜇𝑓 in view of the above, a filtering capacitor of 1000 𝜇𝑓 was used. 2.2.3 regulation and charging circuit taken r1= 200 ω, rs=1 ω, vo = 13v 𝑉𝑜𝑢𝑡 = 1.25𝑣 𝑥 𝑅2 𝑅1+1 (4) 13 = 1.25 𝑥 𝑅2 200 + 1 𝑅2 = 2090.4 ≈ 2 kω a 10 kω variable resistor is used in place of r2 for the purpose of varying the output voltage. 2.2.4 transmitting unit the transmitter (nrf24l01) and the big sound module (sound sensor) are connected to a nano board. 2.2.5 receiving unit the receiver (nrf24l01) uses power supplied from lm117t 3.3 v regulator. the receiver is connected directly to a nano arduino board and the speaker which is powered by 12v battery is connected to a digital amplifier that amplifies the signal using 12 v. 2.2.6 temperature regulation and display unit journal of mechanical engineering and technology (jmet) 38 issn 2180-1053 e-issn 2289-8123 vol.14 no.2 this unit is made up of a 16x2 lcd, a dht11 sensor, and a 12v dc fan. the dht 11 sensor measures the surrounding temperature and send it to the microcontroller, if the temperature rises above 300c the microcontroller actuates the fan via a relay. the display unit shows the data readings from the dht11 sensor and is connected as shown in figure 2.5 below. both the lcd and dht are powered by the atmega328 ic. the atmega is powered by 12v system power via lm 7805 (5v) regulator and is connected according to the data sheet as follows; figure 2.5 temperature regulation and display unit c1 = 22pf, c2 = 22pf, x1 = 16 mhz, r1 = 1kohm. the fan is a 12v dc motor, it is switched by a 12v relay which is powered by 12v battery. the relay receives command signal from the atmega and then actuates the fan on or off. 2.2.7 mechanical rocking unit the mechanical rocking unit is made up of: 1. driver/ switching unit 2. dc motor unit 3. wooden cradle and support 2.2.8 driver/switching unit a driver/switch is built with resistors, transistors and are incorporated design and construction of baby rocking and monitoring system issn 2180-1053 e-issn 2289-8123 vol.14 no.2 39 with relay to aid its functions. the sound detected by the sound module is converted into signals that drives a dc motor which in turn operate the wooden cradle. before the motor is operated, the bits of sound sent out by the module signal passes through a monostable multivibrator that control the relay switching afterwards. the setup is as shown below figure 2.6. monostable and relay driver circuit a 555 timer is used for the multivibrator design. from the 555 timer data sheet. the monostable circuit configuration is obtained as shown in figure 2.6. the followings are obtained in the data sheet 𝐶1 = 1000𝜇𝑓 𝑎𝑛𝑑 𝑅2 = 1𝑘 𝑎𝑛𝑑 𝑅4 = 1𝑘 𝑚𝑜𝑛𝑜𝑠𝑡𝑎𝑏𝑙𝑒 𝑜𝑢𝑡𝑝𝑢𝑡 𝑝𝑢𝑙𝑠𝑒 = 1.1 𝑥 𝑅1 𝑥 𝐶2 (5) the design specification is to have 11 seconds pulse width. therefore, taken r1 = 1000 kω, 11 = 1.1 𝑥 100000 𝑥 𝐶2 𝐶2 = 100𝜇𝑓 the transistor q1 is used for switching purpose and a 2n222 transistor is used. the output of the 555timer is fed to the transistor q2 also 2n2222 via a base resistor r5 of 1kohm value for switching purpose. the transistor switches the relay that drives the dc motor. coil current can be on or off so relays have two switch positions and they are double throw (changeover) switches as shown in figure 2.7 the relay used is 12v. journal of mechanical engineering and technology (jmet) 40 issn 2180-1053 e-issn 2289-8123 vol.14 no.2 figure 2.7 relay symbol 2.2.9 dc motor/wooden cradle dc motor: in this project, we used permanent magnet dc motor, this motor converts electric energy into mechanical energy. its action is based on the principle that when a current-carrying conductor is placed in a magnetic field, it experiences a mechanical force whose direction is given by fleming’s left-hand rule. the schematic representation of a dc motor is shown below: figure 2.8: dc motor symbol for this project, a dc motor is used to swing the wooden cradle back and forth. the motor has the following specification: breaking torque=28n.m motor speed; the low and high speed of the motor is 35rpm and 50rpm respectively. working torque = 6n.m working current = 4.5a working voltage = 12v as such, the motor can withstand the load and work effectively. design and construction of baby rocking and monitoring system issn 2180-1053 e-issn 2289-8123 vol.14 no.2 41 2.2.10 wooden cradle the wooden cradle is designed to cater for a 6 month old baby. plate 1, 2 and 3 shows the solidworks representation of the wooden cradle while figure 2.9 shows the completely coupled wooden cradle. (a) (b) (c) (d) figure 2.9 (a) plate 1 (b) plate 2 (c) plate 3 (d) completed coupled wooden cradle the design as follows: average weight of 6month old baby = 7kg target weight = 10kg average length of 6 month old baby = 45.5 cm inner part: 50 × 20 × 30 cm outer part: 85 × 75 × 40cm using: max. stress = (max. moment × c)/(moment of inertia) (6) journal of mechanical engineering and technology (jmet) 42 issn 2180-1053 e-issn 2289-8123 vol.14 no.2 where max. moment = (pl) / 4 (7) max. moment = 9.81 𝑥 10 𝑥 0.5 4 = 12.262 𝑁𝑚 moment of inertia = (bh3) / 3 (8) moment of inertia = 0.2 𝑥 0.653 3 = 0.0183 𝑚4 𝛿𝑚𝑎𝑥= 12.262 𝑥 0.25 0.0183 = 167.5 𝑁𝑚−3 the steel of 2 cm diameter with a tensile strength 841 mpa was used which will withstand the stress developed. 3.0 results and discussion this chapter deals with the description of tests performed on the various sections of the overall system and their corresponding result as well as the result of the whole system. in order to verify the correct functionality of the system, each component had to be tested individually. to achieve the effective testing of these components and the entire system, the following were used:  digital multimeter  bread board  proteus simulation software 3.1 testing and results different types of testing were carried out in implementing this project, the following tables contained various results obtained from the device for audio transmission range, loading test, sound detection, power supply regulators and temperature sensing. design and construction of baby rocking and monitoring system issn 2180-1053 e-issn 2289-8123 vol.14 no.2 43 table 3.1: transmission range s/n distances (meter) output signal (based on audible pitch) 1 4 on 2 8 on 3 12 on 4 16 on 5 20 on 6 24 on 7 28 on 8 32 on 9 26 on 10 40 on 11 44 off table 3.2 sound detection s/n distance (centimeters) output (based on audible pitch) 1 5 on 2 10 on 3 15 on 4 20 on 5 25 on 6 30 off table 3.3 cradle oscillation when loaded s/n weight (kg) angle (oc) frequency (hz) 1 1 15 1.18 2 2 15 1.18 3 5 15 1.18 4 7 15 1.18 5 10 15 1.18 6 12 15 1.18 journal of mechanical engineering and technology (jmet) 44 issn 2180-1053 e-issn 2289-8123 vol.14 no.2 table 3.4 power supply supply theoretical voltage (v) measured voltage (v) lm 7812 12.0 11.7 regulator lm117 3.3 3.29 regulator lm7805 5.0 4.91 table 3.5 battery charging time s/n charging time difference charging voltage difference 1 53min 0.43v 2 5min 0.01v 3 17min 0.07v 4 45min 0.22v average 30min 0.1825v 3.2 discussion of result the result and performance of the system depicts a reasonable achievement in the design approach of the baby rocking and monitoring system. so many challenges were encountered and most were resolved before this level of success could be achieved. the real time transmission of audio signal makes the concept of the project feasible. the system was able to transmit the sound signal to a distance of 40 meters as shown in table 3.1, the received signal strength reduces upon increasing the distance of transmission until it finally diminishes to zero at 44 meters. suitable equipment needed to measure the sound frequency are not available and there for the result signifies availability or not of the sound i.e. “on or off”. the transmission range of 20 meters from the design has there for been achieved. sound signals are picked up from about 30cm from the sound sensor as shown in table 3.2. it is worthy to note that the closer the sound generator is to the sound transducer, the higher the output pitch heard at the receiving end. this is due to the specifications of the sound module used but the output is still appreciable. design and construction of baby rocking and monitoring system issn 2180-1053 e-issn 2289-8123 vol.14 no.2 45 speed and frequency and angle of oscillation of the cradle remain constant when loaded with different weight values as shown in table 3.3. this is expected as the dc motor used is of high torque and hence the designed objective of 10kg is achieved. however, the swinging speed can be considered to be 50% of the expected outcome this is due to the speed of the dc motor used. when the device is turned on it will be noticed that the cradle swings for some time even without the audio input, this is due to the mono stable multi vibrator driver resetting condition and it is expected that the user keeps his baby only after this few seconds initial swinging in the cradle. the battery charging time is nonlinear as shown in table 3.5 above and a charging time of 0.365v/hr. was achieved. with this rate, it will approximately take 36hrs to charge a 12v battery which is not the case in reality since the higher the voltage value of battery under charging, the lower it draws current and hence the slower it gets charged and vice versa. this is because at high voltage values, the potential difference between the charger and the battery becomes smaller. 4.0 conclusion an attempt was made in this paper to improve on the limitations of the existing versions of automatic cradle systems. the following conclusions can be made based on the result obtained. the device is safe to work with loading range of 0-10 kg baby weight with transmission range of 0-30 meters in open space and 020meters with barrier. due to the limited range of sound detection of 30cm as discussed above, the baby should be placed in the cradle with his/her head within the stipulated range acknowledgment the authors gratefully acknowledged department of mechatronics and system engineering, abubakar tafawa balewa university bauchi for the supports and throughout the research. journal of mechanical engineering and technology (jmet) 46 issn 2180-1053 e-issn 2289-8123 vol.14 no.2 references [1] m a, s., u a, n. k., & ashok, n.," infant cradle monitoring system using iot", ijarcce, vol. 8, no. 4, pp. 15–20, 2019. [2] srivastava, a., yashaswini, b. e., jagnani, a., & sindhu,"smart cradle system for child monitoring using iot" , international journal of advance in scientific research and engineering, vol. 3, no. 9, pp. 2764–2768, 2019. [3] rachanapalaskar, akshada wagh, shweta pandey, a. t, "automatic baby cradle and monitoring for infant care", international journal of advanced research in science and engineering, vol. 5, no. 5, pp. 55–60, 2016. [4] srikanth, s., ramya, p., satheesh, m., philip, g. t., & vineetha," smart baby cradle system", international journal of advance in scientific research and engineering ,, vol. 4, no. 3, pp. 51–60, 2018. [5] shastry, s. p., harshitha, s., vamsi, r., lokende, v., & bhanumathi, k. s," instrumentation sensors based infant safety smart cradle", international journal of advanced research in science and engineering, vol. 4, no. 6, pp. 2–6, 2017. [6] joshi, m. p., & mehetre, "a survey on advancement of baby cradle" , international journal of science and research, vol. 6, no.7, pp. 1466–1469, 2017. [7] kadu, a. b., dhoble, p. c., ghate, j. a., bhure, n. b., & jhunankar, v. a, "design, fabrication and analysis of automated cradle", int. j. mech. eng. & rob. res, vol. 3, no. 2, pp. 380–383, 2014. [8] sambhar, v. k., & tadwalkar,"automatic cradle system for infant care," international journal of advanced research in science and engineering, vol. 6, no. 4, pp. 21–24, 2017. [9] nathan, s. s., kumar, s., & kanmani, m, "survey on digital age smarter cradle system for enhanced parenting," international journal of applied engineering research, vol.13, no 10, pp. 8187-8193, 2018. issn: 2180-1053 vol. 8 no.1 january – june 2016 13 inspection interval determination for mechanical/service systems using an integrated promethee method and delay time model ikuobase emovon  school of marine science and technology, newcastle university, newcastle upon tyne, ne1 7ru, uk abstract optimum inspection maintenance decision problem is a multi-criteria problem which many researchers have viewed as a single criterion problem such as mainly using downtime or cost, as the basis for selecting interval for the task. however, a combination of a number of criteria can yield a more appropriate interval for inspection maintenance task for mechanical/service system. this paper proposes an integrated promethee technique and delay time concept for implementing optimum inspection interval for mechanical/service systems based on combination of conflicting decision criteria. while the delay time concept is used to model decision criteria, the promethee method is used to aggregate decision criteria and ranking of alternative inspection interval. the promethee technique had been enhanced in this paper by incorporating utility function concept into it, in order to embed maintenance practitioners risk perception into the decision making process. the applicability and suitability of this methodology is demonstrated with two case studies. keywords: promethee technique; inspection intervals; delay time model; mechanical/service system 1.0 introduction british standard define maintenance as (bs 1993) “the combination of all technical and administrative actions, intended to retain an item in, or restore it to a state in which it can perform a required action”. maintenance of mechanical/service system with so many components is still a challenge across the globe, as the cost vary from 20 to 30 percent of the overall cost of its operation. however having a sound and effective maintenance system in place will help reduce cost of maintenance without compromising system reliability and availability. one of the greatest challenge of mechanical/service system maintenance is the determination of the interval for performing inspection. the purpose of carrying out inspection activities on mechanical/service systems is to establish their true condition and in the course of performing these activities, if a defect is found, a repair or replacement task is schedule and carried out to prevent the equipment from further deterioration. however failure to perform inspections task, defects may go unnoticed which can result in catastrophic system failure that may have irreversible impact on the company. even if inspection tasks are carried, defects can still  corresponding author e-mail: i.emovon@newcastle.ac.uk mailto:i.emovon@newcastle.ac.uk journal of mechanical engineering and technology 14 issn: 2180-1053 vol. 8 no.1 january – june 2016 occur between successive intervals if they are not properly timed. it is then obvious that the subject of inspection interval determination problem is critical, justifiable and worthy of investigation as it is central to the effective operation of mechanical/service system. traditionally, maintenance practitioners rely on experience or original equipment manufacturers’ recommendation in determining appropriate time interval for carrying out inspection and the result is far from being optimal (christer et al., 1997). one of the most reliable technique applied in recent time is the delay time concept and it was introduced by christer (1982). the delay time is the time between when a defect become noticeable and when the actual failure of the system occur. pillay et al. (2001) used the expected downtime model based on the delay time concept to determine optimum inspection intervals for fishing vessel equipment items. the inspection plan was developed with the purpose of reducing vessel downtime as a result of machinery failure that could occur between discharge ports. in order to determine the suitability of the approach, the winch system of the fishing vessel was used as case study. the case study results showed that an inspection period of 12 hours was appropriate for the system. arthur (2005) applied the delay time model to determine an optimum inspection interval for condition monitoring of an offshore oil and gas water injection pumping system. the approaches used in the literature based on the delay time concept utilises a single decision criteria in the determination of appropriate interval for performing inspection for mechanical/service systems. however a more appropriate interval can be determine by applying multi-criteria based approaches. to achieve this aim, promethee method, a multi-criteria decision making (mcdm) tool is integrated with the delay time model in order to formulate a more efficient tool for inspection interval determination for application to mechanical/service systems. in this paper delay time concept is applied in modelling of two decision criteria; cost and downtime while the promethee method is used to aggregate decision criteria such that, multiple criteria can be used simultaneously in the ranking of alternative inspection interval. however to make the promethee method more robust, utility function concept is integrated in order for risk perception of the maintenance practitioners to be included in the decision making process. the rest of the paper is organized as follows: the proposed methodology is presented in section 2. this is then followed with two case studies to illustrate the applicability of proposed method in section 3. finally the conclusion is presented in section 4. 2.0 methodology 2.1 delay time concept inspection task can only be beneficial if there is a sufficient period between the time that the defect is observed and the actual time of failure of the equipment. as previously stated the time interval between when a defect becomes identifiable and the actual time of failure is referred to as the delay time (h). figure 1 is used to illustrate the delay time concept. inspection interval determination for mechanical/service systems using an integrated promethee method and delay time model issn: 2180-1053 vol. 8 no.1 january – june 2016 15 figure 1. delay time concept showing a defect’s initial points and failure points figure 1 shows multiple points of failure, both initial and actual points where failure occurs and also two different inspection plans for a mechanical system. it is obvious from the figure that if the inspection of the system is performed at an interval of b a lot of failures will happen in the system since most of the defects would have resulted in actual failure. alternatively inspection plan a would result in detecting virtually all of the defects before the actual failure of the system could occur. the key to achieving maximum success in mitigating catastrophic failure of mechanical/service systems is to have a proper understanding of the delay time (h) of the system such that maintenance can be performed within this period. based on christer and waller (1984), “a defect occurring within a period of (0, t) in a system has a delay time, h , and h has a probability density function of f(h). if failure of the system occurs at a period (0, t-h) the maintenance (repair or replacement) carried out is referred to as breakdown maintenance otherwise the maintenance is inspection maintenance. for the system, if all possible values of, h, are added up, according to christer and waller (1984), the probability of a defect occurring as a breakdown failure is”: 𝐵(𝑇) = ∫ 𝑇 − ℎ 𝑇 𝑇 0 𝑓(ℎ)𝑑ℎ (1) the above equation was established based on the following assumptions: (1) inspection is performed at regular intervals (2) defects discovered during inspection are repaired (3) perfect inspection meaning all defects are discovered during inspection (4) arrival rate of defects is constant however it is worth noting that some of these assumptions may not be realistic in practical situations. for example, it may not be possible to identify all defects during inspection as some defects could be hidden although the system performance journal of mechanical engineering and technology 16 issn: 2180-1053 vol. 8 no.1 january – june 2016 degradation may have started during inspection. some of these assumptions are made to ease the modelling of the system and for ease of computation of the models. since in this paper weibull distribution is assumed, probability density function of the delay time f(h) is evaluated as: 𝑓(ℎ) = 𝛼 𝛽 ( ℎ 𝛽 ) 𝛼−1 𝑒𝑥𝑝 [− ( ℎ 𝛽 ) 𝛼 ] (2) where 𝛼 and 𝛽 represents shape parameter and scale parameter respectively. 2.2 decision criteria modelling for this paper two decision criteria; cost and downtime were chosen based on which optimum interval for mechanical/service system is determined. the two decision criteria had been modelled based on the delay time concept (christer & waller, 1984) and are discussed as follows: the downtime criteria which is the expected downtime per unit time d(t) to be suffered when operating an inspection time interval, t, is presented as: 𝐷(𝑇) = 𝜑 + 𝑘𝑟 𝑇𝐵(𝑇)𝑑𝑎 𝑇 + 𝜑 (3) where t = inspection time interval 𝜑 = downtime as a result of inspection 𝑑𝑎 = average downtime due to breakdown repair ℎ = delay time 𝑘𝑟 = arrival rate of defects per unit time the cost criteria which is the expected cost per unit time c(t) of inspection time interval t is presented as: 𝐶(𝑇) = [𝑘𝑟 𝑇{𝐶𝑏 𝐵(𝑇) + 𝐶𝑖𝑖 [1 − 𝐵(𝑇)]} + 𝐼𝑐 ] 𝑇 + 𝜑 (4) where 𝐶𝑏 = breakdown repair cost 𝐶𝑖𝑖 = inspection repair cost 𝐼𝑐 = inspection cost inspection interval determination for mechanical/service systems using an integrated promethee method and delay time model issn: 2180-1053 vol. 8 no.1 january – june 2016 17 2.3 promethee method promethee a multi-criteria decision making method is an acronym for preference ranking organisation method for enrichment evaluations, developed by brans, first presented in 1982 (brans, 1986) and further extended by brans and vincke (brans & vincke, 1985). it is one of the outranking technique for solving multi-criteria decision problem. there are seven variant of the promethee method (behzadian et al., 2010) but promethee ii is the most popular of all the versions and it’s fundamental to the implementation of the other versions. the technique have been applied successfully in solving multi-criteria problem such as material selection problem and maintenance strategy selection problem (emovon et al., 2015b). the basic steps of the promethee method can be defined as follows: (1) determination of a decision matrix: consider a multi-criteria problem with, n number of alternatives i.e. a1, a2,…, an and m number of decision criteria i.e. b1, b2,…, bm upon which the alternatives are evaluated. an example of such problem is the decision matrix in table 1. table 1. decision matrix alternatives (ai) decision criteria (bj) c(t) d(t) a1 x11 x12 a2 x21 x22 a3 x31 x32 an xn1 xn2 (2) determination of utility functions: the maintenance practitioners’ behaviour with respect to risk is put into consideration through utility function, instead of analysing variables in table 1 directly into the promethee model. the risk perceptions of the maintenance practitioners are of three categories which are incorporated into the utility function and these are; risk prone, risk neutral and risk averse. according to ferreira et al. (2009) the maintenance practitioners are risk neutral as regards to cost criterion and as such a linear function is applicable while for the downtime criterion, the maintenance practitioners are risk prone and a negative exponential function is utilised. the utility function for c(t) and d(t) were presented as follows: 𝑢(𝐶(𝑇)) = 𝐶(𝑇) − max 𝐶(𝑇) min (𝐶(𝑇) − max 𝐶(𝑇) (5) journal of mechanical engineering and technology 18 issn: 2180-1053 vol. 8 no.1 january – june 2016 𝑢(𝐷(𝑇)) = 𝐶(𝑇) − min 𝐷(𝑇) max 𝐷(𝑇) − min 𝐷(𝑇) ln (0.01) (6) where max 𝐶(𝑇) represents the maximum value of elements, xij, for cost criterion, min 𝐶(𝑇) represents the minimum value of element, xij, for cost criterion, max 𝐷(𝑇) represents the maximum value of element, xij, for downtime criterion, min 𝐷(𝑇) represents the minimum value of element, xij, for downtime criterion. the results are then use to form a utility function decision matrix as shown in table 2. table 2. utility function alternatives (ai) decision criteria (bj) u(c(t)) u(d(t)) a1 r11 r12 a2 r21 r22 a3 r31 r32 an rn1 rn2 (3) definition of preference function comparison of alternatives a and b for each criterion are performed based on preference function which transform the difference between the alternatives into a value ranging from 0 to 1. the preference of alternative a over b for each criterion is represented as: 𝑃𝑗 (𝑎, 𝑏) = 𝐹𝑗 { 𝑓𝑗 (𝑎) − 𝑓𝑗 (𝑏)} (7) where 𝐹𝑗 is the function of the deviation (d) between alternative a and b. there are six types of preference function and are presented in table 3. (2) determination of numerical weights of criteria: this is a measure of the relative importance of each criterion. there different methods for evaluating decision criteria weight such as analytic hierarchy process (ahp), entropy method and variance method. in this paper ahp method was chosen because of its capability to incorporate both quantitative and qualitative information. the normalisation of the weight is carried out as follows: ∑ 𝑤𝑗 𝑚 𝑗 = 1 (8) where 𝑤𝑗 is the weights of criteria 𝑗 inspection interval determination for mechanical/service systems using an integrated promethee method and delay time model issn: 2180-1053 vol. 8 no.1 january – june 2016 19 table 3. preference functions, adapted from (figueira et al., 2005) journal of mechanical engineering and technology 20 issn: 2180-1053 vol. 8 no.1 january – june 2016 (5) evaluation of the overall preference index of a over b, 𝜋(𝑎, 𝑏): the weighted average of all the preference functions pj (a, b) for all criteria is mathematically defined as follows: 𝜋(𝑎, 𝑏) = ∑ 𝑤𝑗 𝑚 𝑗=1 𝑃𝑗 (𝑎, 𝑏) (9) the net flow 𝜙 is then determined, which is the measure of the performance of the alternatives. the net flow which is the difference between the positive flow ∅+ and the negative flow ∅−, is computed as follows: 𝜙(𝑎) = ∅+(𝑎) − ∅−(𝑎) (10) where ∅+(𝑎) = 1 𝑚 − 1 ∑ 𝜋 𝑏≠𝑎 (𝑎, 𝑏) (11) ∅−(𝑎) = 1 𝑚 − 1 ∑ 𝜋 𝑏≠𝑎 (𝑏, 𝑎) (12) the alternatives (inspection intervals) are ranked on the basis of the net flow and the higher the value the better the alternative. the steps for the proposed methodology for optimum inspection interval determination are as follows: 1. 2. 3. decision maker determination of both alternatives inspection interval and decision criteria 4. modelling of decision criteria based on delay time concept 5. determination of the parameters of the decision criteria (c(t) and d(t)) such as delay time distribution and associated parameters, cost of inspection, cost of breakdown repair and cost of inspection repair 6. evaluation of d(t) and c(t) for every alternative inspection interval t 7. evaluation of weights of d(t) and c(t) using ahp 8. ranking of alternative inspection interval using promethee 3.0 case study 3.1 case study 1: marine diesel engine-sea water cooling pump to illustrate the suitability of the proposed integrated promethee method and the delay time model, the sea water cooling pump is used. the sea water pump is one of the equipment item of the central cooling system of the marine diesel engine and it has been inspection interval determination for mechanical/service systems using an integrated promethee method and delay time model issn: 2180-1053 vol. 8 no.1 january – june 2016 21 established that scheduled inspection is the most appropriate maintenance strategy use to mitigate it failure (emovon et al., 2015b). the data used as input into the delay time model were obtained from logged records, expert’s opinion, ongoing phd research and from the work of cunningham et al. (2011). the data obtained from these sources are: breakdown repair cost (cb) = £52,500 inspection repair cost (cii) = £10,500 inspection cost (ic) = £210 shape (𝛼) = 10 scale (𝛽) = 5 downtime due to inspection = 12.5 minutes downtime due to breakdown repair = 168 hours arrival rate of defects = 1277 per 10 6 hour the possible intervals of inspection of the equipment also need to be determined and were obtained with the aid of an expert with several years of marine diesel engine maintenance experience. the possible inspection intervals arrived at are 1 hour to 28 hours in steps of 1 hour. 3.1.1 data analysis the above data were used as input into equation 3 and 4 to determine cost and downtime for different inspection intervals. the results obtained for downtime and cost are presented in figures 2 and 3 respectively. figure 2. inspection interval and cost effect 0 5 10 15 20 25 30 40 60 80 100 120 140 160 180 200 alternative inspection intervals (t(hrs)) c (t ) £ journal of mechanical engineering and technology 22 issn: 2180-1053 vol. 8 no.1 january – june 2016 figure 3. inspection interval and downtime effect from the results of the two decision criteria; cost and downtime in figures 2 and 3 respectively, it is obvious that there is conflict among them. for example the optimum solution for the cost criteria is inspection interval of 9 hours while that of downtime is 7 hours. to determine the most appropriate inspection interval, promethee method is utilised in this study. the first step in obtaining a solution using the promethee method is to form a decision matrix. although there are a total of 28 alternative inspection interval as in 1 hour to 28 hours in a step of an hour only the first 10 alternatives are considered since the optimum solution for both cost and downtime lies within this range. the decision matrix formed from the two decision criteria results are shown in table 4. table 4. decision matrix of the sea water cooling pump alternative inspection intervals (hrs) c(t) £ d(t) hrs 1 184.8950 0.1724 2 107.2421 0.0943 3 78.0135 0.0650 4 62.7330 0.0498 5 53.4584 0.0411 6 47.4421 0.0362 7 43.5487 0.0345 8 41.2698 0.0358 9 40.3398 0.0403 10 40.5326 0.0477 since the promethee technique does not determine decision criteria weights, ahp method is utilised to determine the weights. the criteria weights obtained for criteria c(t) and d(t) using ahp method are 0.35 and 0.65 respectively. the utility function values of c(t) and d(t) are then determined using equations 5 and 6 respectively and the results are presented in table 5. 0 5 10 15 20 25 30 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16 0.18 alternative inspection interval (t(hrs)) d (t )h rs inspection interval determination for mechanical/service systems using an integrated promethee method and delay time model issn: 2180-1053 vol. 8 no.1 january – june 2016 23 apart from weight determination of decision criteria, there is also the need to established preference function for each decision criteria. type five and type three preference function in table 3 was chosen for c(t) and d(t) respectively. the net flow of each inspection interval is then evaluated using equation 10 which is the difference between the positive flow and the negative flow and results are presented in table 6. table 5. utility function of the sea water cooling pump alternative inspection intervals (hrs) u (c(t)) u (d(t)) 1 0.0000 0.0100 2 0.5372 0.1357 3 0.7394 0.3611 4 0.8451 0.5999 5 0.9092 0.8022 6 0.9509 0.9448 7 0.9778 1.0000 8 0.9936 0.9575 9 1.0000 0.8239 10 0.9987 0.6435 table 6. promethee performance index for sea water pump inspection interval alternative inspection interval (hrs) ∅− ∅+ 𝜙 rank 1 0.9580 0.0000 -0.9580 10 2 0.0895 0.0692 -0.0204 9 3 0.0216 0.1151 0,0935 8 4 0.0063 0.1209 0,1146 7 5 0.0016 0.1273 0.1257 5 6 0.0002 0.1326 0.1325 3 7 0.0000 0.1350 0.1350 1 8 0.0001 0.1332 0.1331 2 9 0.0013 0.1280 0.1268 4 10 0.0049 0.1222 0.1173 6 from table 6 the optimum inspection interval for the sea water pump is 7 hours. this is closely followed with inspection interval of 8 and 6 hours respectively. the worst solution is the inspection interval of 1 hour. 3.1.2 comparison of proposed ranking tools with vikor method the ranking tool used in this paper is the promethee technique and in order to validate the method for application in prioritising alternative inspection intervals another mcdm tool, vikor, was used in solving the sea water pump inspection journal of mechanical engineering and technology 24 issn: 2180-1053 vol. 8 no.1 january – june 2016 decision problem. although vikor method had not been previously applied in solving inspection interval decision problem but has been successfully use in addressing problems such as risk prioritisation, material selection and selection of outsourcing providers (liou and chuang, 2010, emovon et al., 2015a, anojkumar et al., 2014). the ranks of the ten alternative inspection intervals obtained using promethee and vikor methods are presented in table 7. table 7. comparison of ranking methods alternative inspection interval (hrs) promethee vikor 1 10 10 2 9 9 3 8 8 4 7 7 5 5 5 6 3 3 7 1 1 8 2 2 9 4 4 10 6 6 table 7 showed that both techniques produces the same ranking for the 10 alternative inspection intervals and invariably the same optimum solution for the sea water pump. it is evident that both ranking techniques can individually be use in the ranking of alternative inspection intervals. the result has validated the proposed promethee technique as a tool for solving inspection interval decision problem. 3.2 case study 2: gearbox maintenance decision problem to further illustrate the applicability of the proposed methodology, a case study of a gear box system of an automobile company based in hong kong is applied. the data for the investigation of the optimum inspection interval for the gear boxes of the company is taken from the work of leung and kit-leung (1996). the data are as follows: breakdown repair cost (cb) = $23,000 inspection repair cost (cii) = $18,400 inspection cost (ic) = $1,920 shape (𝛼) = 1.34 scale (𝛽) = 46.9 downtime due to inspection = 0.0625 downtime due to breakdown repair = 5 days arrival rate of defects = 0.839 gearbox per day inspection interval determination for mechanical/service systems using an integrated promethee method and delay time model issn: 2180-1053 vol. 8 no.1 january – june 2016 25 leung and kit-leung (1996) utilise only cost model, c(t), in the determination of optimum inspection interval for the gear box system but in this paper both cost model, c(t), and downtime model, d(t), are simultaneously use to obtain optimum solution. 3.2.1 data analysis applying the maintenance decision data into equation 3 and 4 results of the cost and downtime for various possible inspection intervals were obtained and are presented in figures 4 and 5 respectively. figure 4. inspection interval and cost effect figure 5. inspection interval and downtime effect from the results of the two decision criteria; cost and downtime in figures 4 and 5 respectively it is obvious that there is conflict among them. for example the optimum solution for the cost criteria is inspection interval of 6 days while that of downtime is 2 days. to reach a compromise solution mcdm tool is utilised in this study. although there are a total of 28 alternative inspection interval as in 1 day to 28 days in a step of a day only the first 10 alternatives are considered since the optimum solution for the cost 0 5 10 15 20 25 30 1.56 1.57 1.58 1.59 1.6 1.61 1.62 1.63 1.64 x 10 4 inspection interval (t(days)) c t (£ ) 0 5 10 15 20 25 30 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 inspection intervals (t(days)) d t (h rs ) journal of mechanical engineering and technology 26 issn: 2180-1053 vol. 8 no.1 january – june 2016 and downtime lies within this range. in order to apply the mcdm tool, there is need to form a decision criteria. the decision matrix formed from the results of the two decision criteria are shown in table 8. the criteria weights obtained for criteria c(t) and d(t) using ahp method are 0.80 and 0.20. the utility function values of c(t) and d(t) are determined using equation 5 and 6 respectively and the results obtained are presented in table 9. type six and type three preference function in table 3 are utilise for c(t) and d(t) respectively. the net flow is then evaluated using equation 10 and results are presented in table 10. table 8. decision matrix for the gearbox system alternative inspection intervals (days) c(t) £ d(t) days 1 16345.49 0.0685 2 15923.93 0.0555 3 15789.74 0.0642 4 15731.99 0.0798 5 15706.13 0.0991 6 15696.87 0.1209 7 15697.52 0.1444 8 15704.62 0.1692 9 15716.23 0.1952 10 15731.13 0.2221 table 9. normalised decision matrix for the gearbox system alternative inspection interval u(c(t)) u (d(t)) 1 0.0000 0.6981 2 0.6499 1.0000 3 0.8568 0.7862 4 0.9459 0.5108 5 0.9857 0.2996 6 1.0000 0.1640 7 0.9990 0.0857 8 0.9881 0.0432 9 0.9702 0.0210 10 0.9472 0.0100 table 10. promethee performance index of gear box inspection interval alternative inspection intervals (days) ∅− ∅+ 𝜙 rank 1 0.0380 0.0028 -0.0352 10 2 0.0039 0.0068 0.0029 7 3 0.0007 0.0071 0.0064 1 4 0.0008 0.0066 0.0058 2 5 0.0013 0.0062 0.0049 3 inspection interval determination for mechanical/service systems using an integrated promethee method and delay time model issn: 2180-1053 vol. 8 no.1 january – june 2016 27 6 0.0018 0.0059 0.0041 4 7 0.0022 0.0057 0.0035 5 8 0.0024 0.0054 0.0030 6 9 0.0025 0.0051 0.0026 8 10 0.0027 0.0048 0.0021 9 the result in table 10 reveals that the optimum inspection interval for the gear box is 3 days having rank first among the 10 alternative inspection intervals. the worst solution is the inspection interval of 1 day since it occupies the last position. leung and kitleung (1996) obtained 6 days as the optimum solution while considering only cost without putting into consideration downtime effect. downtime is an important criteria that should be considered in addressing problem of inspection interval especially in service industries were plant system downtime may result to company reputation being damage irreversibly. the 3 days optimum solution obtained using the proposed method will be more ideal as it will result to lower downtime for the system while still maintaining an optimum cost. 3.2.2 comparison of proposed ranking tools with vikor method the ranking of the 10 alternative inspection intervals for the gear box system using promethee method is compared with vikor method and results are presented table 11. table 11. comparison of ranking methods alternative inspection interval (hrs) promethee vikor 1 10 10 2 7 9 3 1 2 4 2 1 5 3 3 6 4 4 7 5 5 8 6 6 9 8 7 10 9 8 from table 11, both promethee and vikor methods produces almost completely the same ranking for the 10 alternative inspection interval for gear box system. the spearman rank correlation between both methods were evaluated to further establish the relationship between them. the spearman rank correlation coefficient of 0.950 was obtained and this again shows that the two technique are strongly correlated. this has further validated the proposed promethee techniques as a viable tool for ranking of alternative inspection interval. generally the advantage of using mcdm tool for journal of mechanical engineering and technology 28 issn: 2180-1053 vol. 8 no.1 january – june 2016 ranking of alternative inspection intervals is that more than one decision criteria can be applied simultaneously in arriving at optimum solution, instead of utilising a single criteria for a multi-criteria problem which is the current practice in most shipping industry. the various mcdm tools has one limitation or the other, their individual use will depend on the maintenance practitioners’ and/or analysts’ choice which may be guided by ease of implementation and suitability (løken, 2007). the promethee technique was chosen as a ranking tool in this paper mainly because of the availability of software that will aid maintenance practitioners in solving inspection decision problem with much ease. 4.0 conclusions one of the popularly used maintenance strategy is the scheduled inspection. however the major challenge with the approach is the determination of the optimum interval for performing the task. in addressing this problem, promethee and ahp methods were integrated and then combine with delay time model such that an optimum inspection interval for any mechanical/service system can be determined based on multiple criteria as oppose to single criteria currently being applied by most industrial maintenance practitioners. the promethee technique had been enhanced in this paper by incorporating utility function concept such that the risk perception of maintenance practitioners can be embedded in the decision making process. the approach have been demonstrated with two case studies; the sea water pump of a central cooling system of a marine diesel engine and a gear box system of an automobile company and the results revealed that the technique is capable of addressing the inspection interval problem of any mechanical/service system that requires maintenance. further work can be done by including more decision criteria such as safety and availability in the decision making process. 5.0 acknowledgements the author would like to appreciate the federal university of petroleum resources, effurun, nigeria for providing the fund for this research through the tetfund academic staff training intervention fund. my appreciation also goes to dr rose a. norman and dr alan j. murphy for improving my skills with respect to conducting research and technical writing. finally, my appreciation goes to the school of marine science and technology, newcastle university, united kingdom for providing an enabling environment for conducting the research. 6.0 references anojkumar, l., ilangkumaran, m. & sasirekha, v. (2014). comparative analysis of mcdm methods for pipe material selection in sugar industry. expert systems with applications, 41, 2964-2980. arthur, n. (2005). optimization of vibration analysis inspection intervals for an offshore oil and gas water injection pumping system. proceedings of the institution of mechanical engineers, part e: journal of process mechanical engineering, 219, 251-259. inspection interval determination for mechanical/service systems using an integrated promethee method and delay time model issn: 2180-1053 vol. 8 no.1 january – june 2016 29 behzadian, m., kazemzadeh, r. b., albadvi, a. & aghdasi, m. (2010). promethee: a comprehensive literature review on methodologies and applications. european journal of operational research, 200, 198-215. brans, j.-p. (1986) l'élaboration d'instruments d'aide à la décision [the development of the decision support tools], raymond nadeau et maurice landry, les presses del l' université laval, québec. brans, j.-p. & vincke, p. (1985). a preference ranking organisation method: (the promethee method for multiple criteria decision-making). management science, 31, 647-656. christer, a. h. (1982). modelling inspection policies for building maintenance. journal of the operational research society, 723-732. christer, a. h. & waller, w. m. (1984). delay time models of industrial inspection maintenance problems. journal of the operational research society, 401-406. christer, a. h., wang, w., sharp, j. m. & baker, r. d. (1997). stochastic maintenance modelling of high-tech steel production plant. stochastic modelling in innovative manufacturing. springer, 445, 196-214. cunningham, a., wang, w., zio, e., wall, a., allanson, d. & wang, j. (2011). application of delay-time analysis via monte carlo simulation. journal of marine engineering & technology, 10, 57-72. emovon, i., norman, r. a., j, m. a. & pazouki, k. (2015a). an integrated multicriteria decision making methodology using compromise solution methods for prioritising risk of marine machinery systems. ocean engineering, 105, 92-103. emovon, i., norman, r. a. & murphy, a. j. (2015b). hybrid mcdm based methodology for selecting the optimum maintenance strategy for ship machinery systems. journal of intelligent manufacturing. retrieved from http://dx.doi.org/10.1007/s10845-015-1133-6 ferreira, r. j. p., de almeida, a. t. & cavalcante, c. a. v. (2009). a multi-criteria decision model to determine inspection intervals of condition monitoring based on delay time analysis. reliability engineering & system safety, 94, 905-912. figueira, j., greco, s. & ehrgott, m. (2005). multiple criteria decision analysis: state of the art surveys, springer science & business media. leung, f. & kit-leung, m. (1996). using delay-time analysis to study the maintenance problem of gearboxes. international journal of operations & production management, 16, 98-105. liou, j. j. h. & chuang, y.-t. (2010). developing a hybrid multi-criteria model for selection of outsourcing providers. expert systems with applications, 37, 37553761. løken, e. 2007. use of multicriteria decision analysis methods for energy planning problems. renewable and sustainable energy reviews, 11, 1584-1595. pillay, a., wang, j. & wall, a. (2001). optimal inspection period for fishing vessel equipment: a cost and downtime model using delay time analysis. marine technology, 38, 122-129. preparation of papers in a two column model paper format journal of mechanical engineering and technology *corresponding author. email: shamanuar@utem.edu.my issn 2180-1053 vol. 12 no.1 jun-december 2020 67 development of planar, shape-changing rigid body segmentation process for general design profiles s. a. shamsudin*1,3, z. zainal2,3, m. n. sudin1,3, h. a. al-issa4 1 centre for advanced research on energy (care), universiti teknikal malaysia melaka, 75450 ayer keroh, melaka, malaysia 2 centre for robotics and industrial automation (ceria), universiti teknikal malaysia melaka, 76100 durian tunggal, melaka, malaysia 3 fakulti kejuruteraan mekanikal (faculty of mechanical engineering), universiti teknikal malaysia melaka, 75450 ayer keroh, melaka, malaysia 4 department electrical and electronics engineering, albalqa applied university, assalt, jordan abstract this work describes the early segmentation results in the progress of a mechanism design process to produce simple planar machines that could approximate a shape change defined by a set of curves with significant differences in arc length. the design profiles vary from one another by a combination of rigid-body displacement and shape change that includes significant differences in arc length. where previous rigid-body shape-change work focused on mechanisms composed of rigid links and revolute joints to approximate curves of roughly equal arc length, this work introduces prismatic joints into the mechanisms in order to produce the different desired arc lengths. the first step is to convert the design profiles into piecewise linear curves, referred to as target profiles. the piecewise linear representation that proves most useful has points identified along the curve at roughly equal distances. the second step is to compare segments of the target profiles seeking those that are best approximated by a common rigid body and those that share curvature similarities allowing for the introduction of a prismatic joint. in the end, implementing the procedure in matlab could create a chain of rigid bodies that are joined by revolute and prismatic joints. the chain can closely estimate the shape of a set of design profiles. keywords: shape-change, mechanisms, kinematic synthesis, planar 1.0 introduction some machines like aircraft wings, for instance, rely on their ability to fluctuate between specific shapes in a pre-determined way. for instance, evaluate the usefulness of diverse wing airfoils for cruising as opposed to active dog-fighting situations. many military agencies are always exploring for a better technology that will enable wings to actively shift shapes to achieve wide variety of flight characteristics as well as surface control, which might be impossible with the state-of-the-art wings (weishaar, 2006). the rewards from such a shape-changing mechanism that can morph among specified profiles and then * corresponding author. email: shamanuar@utem.edu.my journal of mechanical engineering and technology issn 2180-1053 vol. 12 no.1 june-december 2020 68 hold-on to the shapes is noteworthy. meanwhile, others like (lu & kota, 2003) applied some optimization algorithms for discrete topology in compliant mechanisms for shifting the shapes of parabola antennas. many methods can be used to achieve a morphing capacity such as compliant mechanisms and memory alloy materials as described in literature like (trease, moon & kota, 2005), (lateş, căşvean & moica, 2017) and (rubbert et al., 2017). besides these, there are also works with shape-memory alloys that can be actuated by electrical signals or heat such as by (lobo, almeida & guerreiro, 2015), (mohd jani et al., 2013), and (nespoli et al., 2010). concerns with these methods include practical cost and limited displacement size. shape-changing rigid body mechanisms have been proposed as an alternate to the above technologies to achieve the range of displacement needed from a rigid-body linkage mechanism with a well-established set of mechanical design principles (korte, 2006). consequently, the rigid-body shape-changing mechanisms use the concept of breaking up the curves into segments. each segment is optimized in shape and length so that it can best approximate the same portion on each target profile. figure 1 depicts the outcome of such segmentation process. the shape of each segment is basically the mean shape of that portion. researchers in (murray, schmiedeler & korte, 2008) suggested adding up binary links to each segment in order to connect the segment to fixed pivots while achieving lower degree-of-freedom when the system change shapes. the presently available synthesis methods for designing shape-changing rigid-body mechanisms better address problems with profiles of roughly the same arc length. this is a serious limitation of the current methodology. hence, this paper includes developments to the theory that introduce prismatic joints into the chain of bodies used to approximate design profiles. this paper presents the segmentation methodology for profiles with significant differences in arc length that are expected to include prismatic joints in their mechanized form. first, design profiles are converted into target profiles. then, the curvatures of the profiles are compared to allow for the introduction of a prismatic joint. finally, the remaining segments of the design curves are approximated with rigid bodies connected by revolute joints. 2.0 method a design profile is a curve defined by (murray, schmiedeler & korte, 2008) such that an ordered set of points on the curve and the arc length between any two such points can be determined. in earlier rigid-body shape-change work, design profiles were converted into piecewise linear target profiles where each target profile contained the same number of points. when the curves were assumed to be of roughly equal arc length, this distribution of points resulted in curves where each piecewise linear segment is roughly equal in length. given that the curves may now possess large differences in arc length, representation of different design profiles by the same number of points could produce individual linear segments of considerably different length. the length of segment in a piecewise linear curve is journal of mechanical engineering and technology issn 2180-1053 vol. 12 no.1 june-december 2020 69 si = √(xi+1 − xi) 2 + (yi+1 − yi) 2, i = 1, … n − 1 (1) note that i is the point number on the profile curve. the total length of the curve is then lj = ∑ si n−1 i=1 (2) note that j is the curve number on the profile set. given that we seek to represent a curve by desired segments of length sd, mj = 𝐿𝑗 sd⁄ (3) hence, we get the first guess at the number of points that needs to be on the curve as shown in equation (4). the mj number of segments is actually rounded down to make errors on the side too short. nj = maj + 1 (4) equation (2) shows that as all the segment length si where i = 2, 3, 4… n, the last point on the curve. however, in equations (3) and (4), the new number of segments mj and new number of points nj are determined. these numbers must be integer. the next step is to distribute the n points (which are nj of that profile curve) equidistantly along the curve. then, this new attribute of the curve is reviewed. manipulating equation (3), check the segment length. basically, this is refinement step aimed at getting the segment length as close as possible to the desired segment length. the result is the actual segment length as shown in equation (5). saj = 𝐿𝑗 𝑚𝑎𝑗⁄ (5) actually, to get the actual segment length as close as possible to the desired segment length, we increase or decrease the number of points n by 1 as we compare saj and sd. figure 1(a) shows the part in the software being developed, that manipulate the segment length in all the curves in concern. on the other hand, in figure 1(b), the points have been redistributed by segment length. the desired segment length here is 3 units. as a result, the number of points also changes in both the curves in order to get as close as possible to that desired value. 3.0 curvature calculations with the 3 data points, the curvature of the middle point (jth point) is calculated. this will mean the first data point and the last one cannot have their curvature calculated as such since these points are not in the middle of the first or the last 3 points. however, the first point will assume the curvature of the second point the last point (nth point) will assume that which is before it i.e. (n-1) th point. it is known that the curve can be approximated by circles of various radii. the radius of each circle is conversely proportional to its curvature. thus, the curvature is journal of mechanical engineering and technology issn 2180-1053 vol. 12 no.1 june-december 2020 70  = 1 𝑟⁄ (6) and the center point for the circle is (a, b). figure 1. (a) this shows a continuous design profile. (b) this shows a target profile curve with 9 pieces at a certain average piece length shown with two points per piece. the smaller mean piece length of the target profiles seems to approximate well the shape of the design profiles. the curvature will also have sign that is positive or negative. the polarity is determined by considering the vectors formed by the 3 points considered at a time. take points 1, 2, and 3 to be represented by {x1, y1}, {x2, y2}, and {x3, y3} respectively. then, take the two direction vectors p1 and p2 as 𝐏𝟏 = { x2 − x1 y2 − y1 } = { dx1 dy1 } 𝐏𝟐 = { x3 − x2 y3 − y2 } = { dx2 dy2 } hence the cross product becomes 𝐂𝐏 = 𝐏𝟏 × 𝐏𝟐 = (dx1dy2 − dy1dx2)𝒌 (7) if the magnitude cp = |cp| is negative, the curvature  also becomes negative. another term that can be derived is the angle  between the vectors p1 and p2. next, the dot product must be calculated as shown by equation (8). dp = 𝐏𝟏 ∙ 𝐏𝟐 = (dx1dx2 + dy1dy2) (8) (a) (b) a piece a point journal of mechanical engineering and technology issn 2180-1053 vol. 12 no.1 june-december 2020 71 since p1 × p2 = |p1||p2| sin θ and p1 ∙ p2 = |p1||p2| cos θ, it follows that θ = tan−1(|𝐏𝟏 × 𝐏𝟐| (p1 ∙ 𝐏𝟐)⁄ ) (9) the curvature distribution in its original form possesses lots of spikes especially as the curvature is changing signs. the conditions of the spikes depend on how smooth each curve is. to assist on selecting the points for each prismatic joint, the curvature plot might need to be smoothened. however, doing so will also change the corresponding profiles. the user then will have to decide whether the resulting changes to the profiles are acceptable or not. murray, schmiedeler, and korte (2008) described vividly of a concept to rotate and translate a curve to another curve provided they both have the same number of points. shamsudin and murray (2013) even expanded the idea to also include a scaling factor. however, in the rigid body shape-changing case, the scaling is seldom used. if there is a set of p design profiles, let the jth target profile be defined by zji = { xj yj}t, i = 1, … , n. 𝐙𝐣𝐢 = 𝐀𝐳𝐣𝐢 + 𝐝 (10) where the rotational matrix and translational vector are defined as 𝐀 = [ cos θ − sin θ sin θ cos θ ] (11) and 𝐝 = { d1 d2 } (12) equation (10) can be manipulated to achieve optimum conditions that result in 𝐝 = 1 n (zkt − azjt ) (13) where it is defined that 𝐳𝐣𝐓 = ∑ zji = { xjt yjt } tn i=1 (14) the rotation angle θ for 𝐀 can be calculated as 𝑁𝑢𝑚 = (1 𝑛⁄ )(xkt yjt − xjt ykt ) − ∑ (xkt yjt − xjt ykt ) n i=1 (15a) 𝐷𝑒𝑛 = ∑ (𝑥𝑗𝑖 𝑥𝑘𝑖 − 𝑦𝑗𝑖 𝑦𝑘𝑖 ) 𝑛 𝑖=1 − (1 𝑛⁄ )(𝑥𝑗𝑇 𝑥𝑘𝑇 − 𝑦𝑗𝑇 𝑦𝑘𝑇 ) (15b) θ = tan−1(𝑁𝑢𝑚 𝐷𝑒𝑛⁄ ) (15c) now, with θ and 𝐝 known, the profile particular j profile can be shifted to the k profile. journal of mechanical engineering and technology issn 2180-1053 vol. 12 no.1 june-december 2020 72 figure 2. the mean profile is shown here in thick lines brought back in optimized position to each design profile. mean segment was explained in detail by (korte, 2006) and the concept is shown in figure 2, where a mean profile tries to approximate the three different shapes. corresponding section of profiles that have same number of points can be shifted together using the minimized distance algorithm and then the mean profile of the section can be generated by zmi = 1 p (z1i − ∑ zji p j=2 ) (16) where i = 1, … , n. all profiles that share the same number of points can be rotated and translated to a fixed profile, say target profile 1. then from each point, using equation (16), the mean segment or profile can be generated. figure 2 also shows that the one mean profile is optimally positioned back to the 3 profiles. furthermore, the concept of error e is also useful where the maximum value of the point-to-point distances is calculated as the mean profile is placed back to the design profiles. this error reading can be used for further manipulations. 4.0 curvature manipulation oftentimes, the curvature distributions from the curves are filled with spikes. direct calculation of a curvature from 3 points at a time may probably lead to that. however, the curvature values can be manipulated with a scheme, and the resulting new curvature can be used to regenerate the geometric curve it now represents. should the regenerated curve be still close to the original profile curve, then data from the curvature distribution the mean profile is placed back to approximate the 3 shapes. journal of mechanical engineering and technology issn 2180-1053 vol. 12 no.1 june-december 2020 73 can be used for further operations. the scheme that is suggested here is shown in equations (17) through (19) ̃1 = (2 3⁄ )1 + (1 3⁄ )2 (17) and for middle points, the new curvatures become ̃i = (1 4⁄ )i−1 + (1 2⁄ )i + (1 4⁄ )i+1 (18) where i = 2,3,4, …, n-1, then ̃n = (2 3⁄ )n + (1 3⁄ )n−1 (19) the simple example shown in figures 3 and 4 is meant to show that the smoothing of the curvature plot enables us to see the trend of the plot better. the plot has less spikes and their magnitudes are somewhat reduced anyway. figure 3. this plot shows the distribution of original curvature data for each profile curve. profiles 1 and 2 correspond to data 1 and data 2, respectively. point numbers c u rv at u re journal of mechanical engineering and technology issn 2180-1053 vol. 12 no.1 june-december 2020 74 figure 4. this shows curvature distribution that was smoothened 25 times in order to better see the trend of the distribution. again, profiles 1 and 2 correspond to data 1 and data 2, respectively. the application of the smoothing techniques is shown to the same example can be shown as a function of how many times the curvature data is smoothened. figure 5 below shows the changes in the generated target profiles as compared to the original ones as the curvature distribution changes. the target profiles match the shapes of the design profiles almost perfectly as evident in figure 6. figure 5. the curvature distribution after smoothing. point numbers c u rv at u re point numbers c u rv at u re journal of mechanical engineering and technology issn 2180-1053 vol. 12 no.1 june-december 2020 75 figure 6. the curves plotted after 5 times of smoothing processing. the user of this system has the ability to choose how much compromise is acceptable based on the differences between profiles that are generated from curvature data and the original profiles. a smoother distribution of curvature can definitely assist in selecting the regions suitable for prismatic joints. if figure 6(b) is selected as the curvature plot, then in the next process, it will be easier for the user of the system to select the start and end points where a prismatic link will be placed. the regenerated target profiles were made from information of the curvature. to do this, we refer again to the equation used to find the curvature. however, now the curvature ̃ is already known from equation (17) through equation (19) while the unknowns are the center point (a, b) and the third point (x3, y3). the first part is to solve for the center point (a, b) by knowing (x1, y1) and (x2, y2). (a2 + b2) − 2ax − 2by = (r2 − x2 − y2) (20) this equation (20) can be expanded in matrix form as [ 1 −2x1 1 −2x2 ] { a a2 + b2 } + { 2y1 2y2 } = { r2 − x1 2 − y1 2 r2 − x2 2 − y2 2 } (21) now, having center (a, b) from equation (21), one can solve for (x3, y3). having these coordinates, the profile could then be redrawn from a fixed frame. once this is done, the method of shifted profiles is applied so that the regenerated profile is shifted back to the original target profile. one of the ways to find (x3, y3) is by using the circle equation (22) shown below. 2ax + 2by + (r2 − a2 − b2) = (x2 + y2) (22) profile 1 profile 2 journal of mechanical engineering and technology issn 2180-1053 vol. 12 no.1 june-december 2020 76 this can also be shown in a matrix form as [ 1 2x1 2y1 1 2x2 2y2 1 2x3 2y3 ] { r2 − a2 − b2 a b } = { x1 2 + y1 2 x2 2 + y2 2 x3 2 + y3 2 } (23) equation (23) solves for the curvature where curvature is 𝜅 = 1/𝑟 as in equation (6). however, if the distance between points 1 and 2 is not the same as the distance between points 2 and 3, then when we use data points 1 and 2 as well as the signed value of the curvature of point 2, we can have a choice of 4 possible points for (x3, y3). nevertheless, since the construction of the target profiles uses finite numbers of roughly equal length segments, then 2 of these possible points coincide with point 1. the other 2 possibilities will have either positive or negative curvature value. the algorithm created should compare these values to match the input curvature value used. this could be from the original data or after the curvature was smoothened. figure 7 shows the detection of (x3, y3) when the segment lengths are different and another when they are about the same. normally, the algorithm created would choose the correct point 3. thus, it is a reliable way to find a unique point 3 in the geometry. figure 7. the process of creating the constant curvature segment from (a) through (d). journal of mechanical engineering and technology issn 2180-1053 vol. 12 no.1 june-december 2020 77 5.0 selection of constant curvature links with prismatic joints in matlab, the selection process involves looking at the plot of curvature and point numbers as in figure 8(a) for example. the less clutter the data distribution gets, the easier it is to see which curvature points could fall under the same band. the band of curvature translate to the range that the points will have rather similar average radius, so that a single radius prismatic link can connect them. the selection can be done for one prismatic joint or link at a time. within the closed curvature band, the start point and the stop point of the prismatic link are selected for each profile. after this is done, the plot of the profiles will be updated with all curves beginning at the stop points in the previous selection. next, if there are points that fall in another band, the same selection process is taken. figure 8(b) illustrates this concept. the aim is mainly to get as many links with prismatic joint as possible. however, each prismatic set is to be selected from a band of curvatures that are common throughout all curves involved. for each link with prismatic joint, the range of point numbers for each target profile can be determined by using the crosshair selection process. from this information the curvature values pertaining to those points, a mean curvature can be calculated. using the relationship in equation (6), the mean radius of the constant curvature link with prismatic joint can be found. the prismatic link, whether it is curved or straight, must basically have the same mean radius to operate. rm = (1 𝐶⁄ ) ∑ ((1 nj⁄ ) ∑ 1 ij⁄ nj i=1 )cj=1 (24) c here is the number of profiles in the synthesis. by knowing this mean radius from equation (24), next is basically forming an arc with the center at (0,0). the polarity of rm is important here since it determines the shape of the arc, either concave up or concave down. a negative rm will start the arc at (-rm, 0) and move clockwise to create concave down shape. on the other hand, a positive rm will start the arc at (rm, 0) and move clockwise to create concave up shape. each small segment of the arc formed is equal to the size of a mean piecewise linear segment sm that make up the target profiles. hence, the formation of the arc occurs at specific delta angle until the range of point numbers selected for a particular prismatic link is covered for that profile. equation (25) shows how to get the angle. δ = cos−1[(sm 2 − 2rm 2 ) (−2rm 2 )⁄ ] (25) the step angle δ does not have polarity. the different length arcs are created one for each profile since they have different number of points. then methods described in section 3 are used to optimally place these arcs back into position. figure 9 shows the concept for one prismatic link that works on 2 profiles of different length extensions. journal of mechanical engineering and technology issn 2180-1053 vol. 12 no.1 june-december 2020 78 figure 8. (a) the curvature band identified for the use of prismatic joint and constant curvature segments. (b) curvature band left that is suitable for a mean segment. profiles 1 and 2 relate to data 1 and data2, respectively. (a) (b) curvature band possible end point for profile 2 selected end point for profile 1 common start points journal of mechanical engineering and technology issn 2180-1053 vol. 12 no.1 june-december 2020 79 figure 9. the length-changing link is the same but in two positions. the next part is filling up the gaps on the profiles that are not specified for prismatic links. each portion of the gaps must have the same number of points. they each will be filled with one or more rigid bodies until the gaps are filled. the number of rigid bodies used is determined by an error limit set in the program. the motive is to have the rigid bodies to approximate the shape of the profile section that they want to fill. one would see that the errors after the links or members are connected via revolute joints seem to be higher than they were at segmentation part. the final position of each member is not at optimized minimum distance anymore. instead, starting from link number 2, each link has to start where the previous link ended. they are connected there. next, the end of the link points towards the point on each target profile where the link approximates the curve. therefore, the error of this final position can become significantly higher than before the links are hooked up. real world applications include a hull design that would need to change shape in order to obtain the required drag for a craft to move in a fluid. this is depicted by the cross-section of the machine in figure 10 that may change shape and thus achieving different arc length. shamsudin in (shamsudin, 2013) showcased an example of shape-changing slat for a 30p30n airfoil wing. researchers in (ismail, shamsudin & sudin, 2015) and (shamsudin & ismail, 2018) also touched on this novel design of the wing and slat profiles. the paper also suggested some of the benefits especially in terms of lowering noise level. if the airfoil is used underwater, ultrasound can be used to measure fluid velocity around the profiles as in (daosaeng & thong-un, 2019). figure 11 displays a more recent design for the same slat in (ismail, shamsudin & sudin, 2015) where the mechanisms can be simplified further. some analytical methods in (myszka, 2009) may help in locating the fixed pivot points in a four-bar mechanism using three-position synthesis. there are various methods – mostly graphical – for this synthesis as explained in enough detail in (dicker jr., pennock & shigley, 2016), (myszka, 2015) and (waldron, kinzel & agrawal, 2016). the same constant curvature segment profile 1 profile 2 journal of mechanical engineering and technology issn 2180-1053 vol. 12 no.1 june-december 2020 80 figure 10. the aircraft wing slat changes between stowed, midway position, and fully deployed as it changes from (a) to (c). (a) (b) (c) journal of mechanical engineering and technology issn 2180-1053 vol. 12 no.1 june-december 2020 81 6.0 conclusion the algorithm being developed here is very promising in approximating the target profiles with links that consist of rigid bodies and prismatic links. the number of rigid bodies that fill the gaps that is not covered by prismatic links can be optimized by using the distances of the mean segment profiles to the target section profiles. the largest distance becomes the error associated with the mean profile. while this error is less than a specified maximum error, such as 0.1, the size of the mean segment profile is increased until its error is just below the maximum error. then another mean segment is created following the same algorithm until the same-number-of-points section is covered. the overall error will increase though, as the bodies are linked up together with revolute joints. 7.0 acknowledgement the authors like to thank the universiti teknikal malaysia melaka (utem) and the centre for advanced research on energy (care) in supporting this work. the first author is highly indebted to professors andrew p. murray and david h. myszka at the university of dayton in ohio, usa, who were instrumental in the early versions of this manuscript. this article is also based primarily on the dissertation in (shamsudin, 2013). 8.0 references daosaeng, j., & thong-un, n. (2019). a study of flowrate calculation using esprit technique for ultrasonic velocity profiles. engineering journal, 23(2). dicker jr., j. j., pennock, g. r., & shigley, j. e. (2016). theory of machines and mechanisms (5th ed.). new york: oxford. ismail, m. h., shamsudin, s. a., & sudin, m. n. (2015). design and analysis of rigidbody shape-change mechanism for aircraft wings. jurnal teknologi, 77(21), 1-7. korte, b. (2006). the application of rigid-body kinematics to shape-changing mechanism design. msc thesis, the ohio state university, ohio, usa. lateş, d., căşvean, m., & moica, s. (2017). fabrication methods of compliant mechanisms. procedia engineering, 181, 221 – 225. lobo, p. s., almeida, j., & guerreiro, l. (2015). shape memory alloys behaviour: a review. procedia engineering, 114, 776 – 783. lu, k. j. & kota, s. (2003). design of compliant mechanisms for morphing structural shapes. journal of intelligent material systems and structures, 14, 379 – 391. mohd jani, j., leary, m., subic, a., & gibson, m. a. (2013). a review of shape memory alloy research, applications and opportunities. materials and design, 56, 1078 – 1113. journal of mechanical engineering and technology issn 2180-1053 vol. 12 no.1 june-december 2020 82 murray, a. p., schmiedeler, j. p., & korte, b. (2008). kinematic synthesis of planar, shape-changing rigid-body mechanisms. asme journal of mechanical design.130(3), 032302:1–10. myszka, d. h. (2009). kinematic synthesis and analysis techniques to improve planar rigid-body guidance. ph.d. dissertation, mechanical & aerospace engineering, the university of dayton, ohio, usa. myszka, d. h. (2015). machines and mechanisms: applied kinematic analysis (4th ed.). new jersey: pearson. nespoli, a., besseghini, s., pittaccio, s., villa, e., & viscuso, s. (2010). the high potential of shape memory alloys in developing miniature mechanical devices: a review on shape memory alloy mini-actuators. sensors and actuators a, 158, 149–160. rubbert, l., charpentier, i. henein, s., & renaud, p. (2017). higher-order continuation method for the rigid-body kinematic design of compliant mechanisms. precision engineering, 50, 455–466. shamsudin s. a., & murray, a.p. (2013). a closed-form solution for the similarity transformation parameters of two planar point sets. journal of mechanical engineering and technology (jmet), 5(1), 59 – 68. shamsudin, s. a., murray, a. p., myszka, d. h., & schmiedeler, j. p. (2013). kinematic synthesis of planar, shape-changing, rigid body mechanisms for design profiles with significant differences in arc length. mechanism and machine theory, 70, 425–440. shamsudin, s. a. (2013). kinematic synthesis of planar, shape-changing rigid body mechanisms for design profiles with significant differences in arc length. ph.d. dissertation, the university of dayton, ohio, usa. shamsudin s. a., & ismail, m. h. (2018). the applications of shape-changing rigid body mechanisms in arts and engineering. journal of advanced manufacturing technology (jamt), 12-1(2), 301-312. trease, b. p., moon, y. m., & kota, s. (2005). design of large-displacement compliant joint. transaction of the asme, 127, 788 – 798. waldron, k. j., kinzel, g. l., & agrawal, s. k. (2016). kinematics, dynamics, and design of machinery. sussex, uk: john wiley. weishaar, t. a. (2006). morphing aircraft technology – new shapes for aircraft design. in multifunctional structures/integration of sensors and antennas meeting proceedings. rto-mp-avt-141, neuilly-sur-seine, france. microsoft word 6075-17214-1-ce.doc journal of mechanical engineering and technology _______________________________________ *corresponding author. email: mshukriy@utem.edu.my issn 2180-1053 vol. 13 no. 1 june – december 2021 design and development of low cost bending machine m. s. yob 1,2, m. a. ahmad sedek 3 , n. a. mat tahir 2 , o. kurdi 4,5 , m. j. abd latif1,2 1advanced manufacturing centre (amc), universiti teknikal malaysia melaka, hang tuah jaya, 76100, durian tunggal, melaka, malaysia 2faculty of mechanical engineering, universiti teknikal malaysia melaka, hang tuah jaya, 76100, durian tunggal, melaka, malaysia 3faculty of department of mechanical engineering, faculty of engineering, university of selangor, bestari jaya, selangor, malaysia 3departement of mechanical engineering, diponegoro university, semarang, indonesia 4national center of sustainable transportation technology, indonesia abstract the bending process has been a core step in fabricating and manufacturing products nowadays. as the industries are rapidly growing, the demand for machinery is also increasing including the bending machine. however, for personal and light use, a commercial bending machine is relatively expensive and bulky. thus, this study intended to design, fabricate, and analyse a lowcost manually operated bending machine for light use. the bending machine was design based on the intended function and the ergonomics to users mainly malaysian. the bending machine was first designed and then fabricated using mild steel as its primary materials due to its high hardness and ease of welding. the bending machine was then tested to pressed two different sizes and two different aluminium sheet thicknesses. then, finite element analysis was conducted on the bending machine's component which is the bender plate and bending base to find the allowable maximum stress and deformation. the findings show that the bending machine is able to bend both large and small aluminium plate thickness 1.0 mm and below without facing any part deformation or failures. keywords: bending; deformation; stress; ansys; finite element analysis; 1.0 introduction the bending of sheet metal is a common and vital process in the manufacturing industry. sheet metal bending is the plastic deformation of the work over an axis, creating a change in its geometry. similar to other metal forming processes, bending changes the shape of the workpiece, while the volume of material will remain the same (hagenah et al., 2019; kulkarni et al., 2015). 1 journal of mechanical engineering and technology issn 2180-1053 in some cases, bending may produce a small change in sheet thickness (hanoof, vishwanth, sureshkumar, & saravanan, 2014). however, for most operations, bending will produce essentially no change in the sheet metal thickness. in addition to creating a desired geometric form, bending is also used to impart strength and stiffness to sheet metal, to change a part's moment of inertia, for the cosmetic appearance and to eliminate sharp edges (eltantawie, 2013; hanoof et al., 2014; kulkarni et al., 2015). figure 1 shows the basic principle of the bending machine. most of the bending machine in the market are generated and powered by the hydraulic or pneumatic system it was very expensive and difficult to operate. these conventional machines are rather expensive and difficult to move. figure 1: basic principle of bending machine simple work involving a typical plate bending process can be conducted using a simpler version of the bending machine. with a simpler version, costs such as electrical, motor, hydraulic, and components can be reduced. when designing and fabricating components and machines, there were many aspects to be considered, such as selecting materials, design dimensions, stress and pressure distributions, and ergonomics. it has been discussed by other previous research where the components of the bending machine are susceptible to the tremendous amount of stress as it is working with deforming another material's properties (engel, sara, & hassan, 2017; yob, mansor, & sulaiman, 2013). for a shaft rotary bending machine, it was reported that the shaft is the most components that fail after repeatable overload works. shafts mostly work under the influence of fluctuated loads or combined torsion and bending loads. if a shaft supports a static load, the bending stresses are fully reversed and the torsion is steady (bello, 2013; engel et al., 2017; hanoof et al., 2014; kadam & deshpande, 2015). rather than conducting a destructive test or reverse engineer of failures, failure analysis is a very simple and cost-effective tool that has been applied widely by the industry sector to develop or improve the product design. there was plenty of research that has been conducted in determining the failure of components/products by using this vol. 13 no. 1 june – december 2021 2 journal of mechanical engineering and technology issn 2180-1053 technique (engel et al., 2017; gandhi, gajjar, & raval, 2008; helguero, ramírez, & amaya, 2019; kane, mishra, & dutta, 2016; yob et al., 2013). to determine the failure modes, analytical, experimental, and computational modelling software analysis methods can be used (gandhi et al., 2008; mat tahir et al., 2017; yob et al., 2013; yob, mansor, & sulaiman, 2014). this analysis method requires complete information about the component geometry, material, load condition, work environment, and work constraints (yob et al., 2013, 2014). this study aimed to design, fabricate, and analyse the stress distribution of a manuallypowered bending machine for light purposes in order to provide alternatives to the heavy and expensive conventional bending machine. 2.0 methodology the bending machine was designed and fabricated by following the flowchart in figure 2. the process was first started with the conceptualisation of the design of the bending machine. then, selection of materials to be used before manufacture the bending machine. the components were then painted before assembled. lastly, the complete bending machine undergoes testing where if the machine or components fails, the process will be back to the redesign phase. figure 2: design process of low-cost manually operated bending machine since this bending machine will be manually operated by operators, one of the machine design concerns is the bending machine's height. an unsuitable or inappropriate high of machinery will lead to an unpleasant experience by the operator. this bending machine was designed so that the lever can be moved by a hand while the sheet will be placed at the waist area height. in this case, the bending machine's height must in the range of the waist height and elbow height in the standing position (as shown in figure 3). according to the anthropometric data for malay male and female, aged 18 to 24 years, y vol. 13 no. 1 june – december 2021 3 journal of mechanical engineering and technology issn 2180-1053 the elbow height in standing position is 769 mm female 5th percentile as the referent for designing the bending machine (karmegam et al., 2011). figure 3: proposed height of bending machine according to anthropometric data 2.1 fabrication of bending machine the bending machine was designed by following these three crucial points; easy operation bending machine, small workpiece, and maximum bending of 130°. the material used for the bending machine was mild steel due to its high hardness and eased of being welded. the bending machine was designed to have an 'a-frame base with two levers welded to the bending place at the left and right operating area. the bending machine is fabricated as shown in figure 4. figure 4: image of designed bending machine (a) drawing and (b) fabricated vol. 13 no. 1 june – december 2021 4 journal of mechanical engineering and technology issn 2180-1053 2.1 testing the test was conducted using four aluminum sheets where each of them is different in width and thickness (2 different thickness; 2 different widths). the test series conducted was shown in table 1. this test was conducted in order to determine the force needed to push the bender bar according to the metal sheet thickness and the bending angle wanted. table 1: series of test conducted thickness (mm) dimension (mm) angle(o) 0.6 170.0 x 75.0 45 90 130 1.0 45 90 130 0.6 815.0 x 95.0 45 90 130 1.0 45 90 130 for the simulation studies, a computational modelling analysis named finite element analysis was used in order to determine the stress distributions and deformations of the bending machine components. there were two conditions set for the analysis by involving two parts of the bending machine which are; bender plate and bending base. both analyses for big and small plates were set at 50n distributed load as shown in figure 5. figure 5: boundary load condition for (a) small plate and (b) big plate 3.0 findings a von-mises stress analysis result can be considered a method for engineers to design and get information about the design. the design will fail if the maximum value of von -mises stress induced in the material is more than the material's strength. figure 6 and 7 shows the finite element analysis for the stress and deformation of the bending machine components respectively meanwhile table 2 shows the tabulated data. vol. 13 no. 1 june – december 2021 5 journal of mechanical engineering and technology issn 2180-1053 figure 6: stress analysis for (a) bender plate for a large sample, (b) bender plate for a small sample, (c) bending base for a large sample, and (d) bending base for small sample vol. 13 no. 1 june – december 2021 6 journal of mechanical engineering and technology issn 2180-1053 figure 7: deflection analysis for (a) bender plate for a large sample, (b) bender plate for a small sample, (c) bending base for a large sample, and (d) bending base for small sample table 2: stress and deformation analysis for bender and base part allowable stress (mpa) dimension (mm) result stress (mpa) maximum deflection (mm) bender plate 240 170 (small) 0.782 0.0177 815 (large) 2.510 0.0176 bending base 170 (small) 1.730 0.0025 815 (large) 1.370 0.0150 figures 6 and 7 show that the analysis shows the high stress area with red color while no stress or '0 stress' are coloured with blue. from the images, it can be analysed that the high stress area is the area where the components meet directly with the pressed plate. through the obtained results, it can be seen that the designed bending machine able to bend both small and large aluminum plates (thickness of 1.0 mm and less). the analysis also showed that the bending machine might function in good condition, with no wear and tear caused by other deformation issues. based on the tabulated result, it can be seen that a large sheet produces more stress on the bender plate. however, the stresses on the bending base are almost the same. the maximum stress obtains only 2.51 mpa at the bender plate. for the deflection, the maximum deflections are almost the same for vol. 13 no. 1 june – december 2021 7 journal of mechanical engineering and technology issn 2180-1053 both small and large sheets for the bender plate. however, the deflection at the bending base shows a small sheet produces a lot of lesser deflection than the big sheet. 4.0 conclusion it can be concluded that the designed bending machine is sturdy and able to perform its intended purposes. the bending machine shows no sign or hint of failures on both stress and deflection analysis when the constraints are applied for both conditions (bending small and big aluminum plate). by varying the thickness, the aluminum sheet tested (1.0 mm) while the stress and bending analysis shows that the components could withstand the applied force with only small deformations. the designed and the fabricated bending machine performs well in the testing. however, it was believed that the bending machine could perform better. thus, in the future, it is recommended that the test parameters be expanded by increasing the thickness and varying the materials of the plate intended to bend. this variety of parameters provided versatility to the bending machine and is believed to have the potentials to be commercialised. 5.0 acknowledgements the authors would like to thank applied mechanical design laboratory utem and advanced manufacturing centre (amc) for the support throughout this study. the authors gratefully acknowledge universiti teknikal malaysia melaka (utem) for the consent granted to access and use the pictures and all journals for this work. 10.0 references bello, r. s. (2013). development and evaluation of metal rolling machine for smallscale manufacturers. agricultural engineering international: cigr journal, 15(3), 80–85. eltantawie, m. a. e. (2013). design, manufacture and simulate a hydraulic bending press. international journal of mechanical engineering and robotics research, 2(1), 1–9. engel, b., sara, s., & hassan, a.m. (2017). failure analysis and fatigue life estimation of a shaft of a rotary draw bending machine. international scholarly and scientific research & innovation, 11(11), 1785–1790. gandhi, a. h., gajjar, h. v, & raval, h. k. (2008). mathematical modelling and finite element simulation of pre-bending stage of three-roller plate bending process. proceedings of the 2008 international manufacturing science and engineering conference, 1–9. hagenah, h., schulte, r., vogel, m., hermann, j., scharrer, h., lechner, m., & vol. 13 no. 1 june – december 2021 8 journal of mechanical engineering and technology issn 2180-1053 merklein, m. (2019). 4.0 in metal forming questions and challenges. procedia cirp, 79, 649–654. hanoof, i. m., vishwanth, s. r., sureshkumar, p., & saravanan, n. (2014). design and fabrication of hydraulic rod bending machine. international journal of research in science, engineering and technology, 3(2), 237–241. helguero, c. g., ramírez, e. a., & amaya, j. l. (2019). engineering interface inside the operative room: design and simulation of a fracture-plate bending machine. procedia cirp, 79, 655–660. kadam, s., & deshpande, g. (2015). a review on design analysis and optimisation of centrifugal casting machine shaft. international journal of engineering research and general science, 5(1), 1–5. kane, s. n., mishra, a., & dutta, a. k. (2016). stress reduction of pickup truck chassis using finite element method. the 3rd international conference on mathematics, science and education, 755(1), 4–9. karmegam, k., sapuan, s. m., ismail, m. y., ismail, n., shamsul bahri, m. t., shuib, s., … hanapi, m. j. (2011). anthropometric study among adults of different ethnicity in malaysia. international journal of physical sciences, 6(4), 777–788. kulkarni, a., pawar, m., yadav, p., patil, a., & jagtap, s. (2015). sheet-metal bending machine. international journal of innovations innengineering research and technology (ijiert), 2(3), 297–334. mat tahir, n. a., abdollah, m. f. b., hasan, r., amiruddin, h., & abdullah, m. i. h. c. (2017). statistical models for predicting wear and friction coefficient of palm kernel activated carbon-epoxy composite using the anova. industrial lubrication and tribology, 69(5). yob, m. s., mansor, s., & sulaiman, r. (2013). finite element modelling to predict equivalent stiffness of 3d space frame structural joint using circular beam element. applied mechanics and materials, 431, 104–109. yob, m. s., mansor, s., & sulaiman, r. (2014). individual stiffness of 3d space frame thin walled structural joint considering local buckling effect. applied mechanics and materials, 554, 411–415. vol. 13 no. 1 june – december 2021 9 microsoft word 03_6113-17277-1-ed.docx journal of mechanical engineering and technology issn 2180-1053 vol. 13 no. 2 december 2021 measurement of lithium transference number in pmma solid polymer electrolytes doped with micron-sized fillers r. w. eric koh1*, c. c. sun 1, y. l. yap1, p. l. cheang1, a. h. you1 1 faculty of engineering and technology, multimedia university, jalan ayer keroh lama, bukit beruang, 75450, melaka, malaysia, email: abstract pmma solid polymer electrolytes (spes) are much safer than gel polymer electrolytes (gpes) due to their better mechanical and thermal stabilities. in this study, pmma-licf3so3-ec, pmma-licf3so3-ec-al2o3 (≤10µm), and pmma-licf3so3-ec-sio2 (≤10µm) were prepared using solution cast method, their ionic conductivity and lithium transference number was investigated using electrochemical impedance spectroscopy (eis) and bruce-vincent method, respectively. the experimental result shows that pmma polymer electrolytes doped with sio2 (≤10µm) exhibits the highest ionic conductivity of 2.35×10-4 s/cm and lithium transference of 0.263 at room temperature. linear sweep voltammetry (lsv) and cyclic voltammetry (cv) analysis also shows that pmma spes incorporated with sio2 (≤10µm) fillers can achieve electrochemical stability up to 3.2v, exhibits excellent reversibility, and good discharging performance. keywords: ionic conductivity; solid polymer electrolyte; transference number; inorganic fillers. 1.0 introduction increasing demand and advancement in electronic devices over the years had driven forward the development of high-capacity rechargeable batteries and fuel cells. since the first industrial implementation by sony corporation in the 1990s, lithium-ion (li-ion) battery has achieved tremendous technological advancement (osinska, 2009). their contributions, such as larger capacity, higher application voltage, longer lifespan, and lower weight, had made them an ideal power source for many applications. lithium-ion batteries can be used in electric vehicles, portable power sources, and space exploration. an electrolyte is the heart of a lithium battery, and it acts as a medium for the charge to transfer between electrodes. however, the liquid and gel electrolyte system in lithium batteries (lib) has potential safety issues and poor electrochemical stabilities, which makes them not safe for certain applications (yao, 2019). on the other hand, solid polymer electrolytes (spes) are a safer alternative. spes have better electrochemical stability, environmentally friendly, and safer for consumers (song, 2015; lim, 2018; pitawala, 2007). despite the potential advantages, spes usage in batteries is hindered due to their high degree of crystallinity and high interfacial resistance at ambient temperature, leading to low ionic conductivity and lithium transference number. pmma spes were a semi-amorphous polymer with improved mechanical and thermal properties over gpes but can still exhibit respectable ionic conductivity and lithium *corresponding author email:1142701349@student.mmu.edu.my 30 journal of mechanical engineering and technology issn 2180-1053 vol. 13 no. 2 december 2021 transference number at room temperature (sun, 2019; faridi, 2018). however, most of the studies to date were focused on pmma gpes, and there are not many studies related to pmma spes and their lithium transference number. faridi et al. investigated the electrochemical properties in pmma-liclo4 based gpes for different salt concentrations (faridi, 2018). they achieved an ionic conductivity and lithium transference number of 12.1×10-3 s/cm-1 and 0.42 respectively for 10% of pmma and 0.75m of liclo4. they explained that the improvement in lithium transference number is associated with the higher polymer content. higher pmma concentration improves the degree of dissociation of lithium salts in the polymer complexes, frees up li+ ions for ion conduction. musil et al. investigated the lithium transference number using the bruce-vincent method in pmma-lipf6-ec: dmc gpes with various salt concentrations (musil, 2014). they obtained a lithium transference number of 0.24 and 0.69 for 0.005mol and 1mol of lipf6 salt concentration. hosseinioun et al. also investigated the usage of crosslinked pmma gpes in li-ion batteries (hosseinioun, 2019). they reported that crosslinking increases the degree of dissociation of lithium salt in gpes. the free anions or degree of dissociation of lithium salts are higher at 83% in pmma gpes than 71% of liquid electrolytes, which means pmma gpes have better ionic mobility than liquid electrolytes. this is evidenced by the obtained lithium transference number of 0.34 for pmma gpes and 0.24 for liquid electrolytes. pmma gpes are capable of exhibiting high lithium transference number and ionic conductivity at room temperature, but they had poorer mechanical stability and thermal stability compared to spes. this makes spes better canditate for devices that are operating in high-temperature environments and require strict safety standards. thus, in this study, pmma in solid form, a safer alternative, was studied. the ionic conductivity and lithium transference number in inorganic filler enhanced pmma spes were investigated and compared to results from several literature. multiple studies have shown that pmma spes can be manufactured into solid electrolytes used in supercapacitors and li-ion batteries requiring higher safety standards (lim, 2018; kurapati, 2019; zakariya, 2020). 2.0 material preparation and characterization 2.1 sample preparation in this study, spe films were prepared using solution cast method. poly(methylmethacrylate) (pmma), ethylene carbonate (ec), lithium triflate (licf3so3), silicon dioxide (sio2), and alumina oxide (al2o3) in the experiment were obtained from sigma aldrich. in a typical synthesis, pmma, ec, and licf3so3 were dissolved in tetrahydrofuran (thf) and stirred at room temperature for 24 hours until a homogenous solution is obtained. the homogenous solution was poured into a petri dish and kept in a desiccator until dry electrolytes film of pmma-licf3so3-ec (spe1) was obtained. similar steps were used to synthesis pmma-licf3so3-ec-al2o3 (spe2) and pmmalicf3so3-ec-sio2 (spe3). the prepared sample composition and sample thickness are tabulated in table 1. 31 journal of mechanical engineering and technology issn 2180-1053 vol. 13 no. 2 december 2021 table 1. composition ratio and thickness of pmma spe film prepared id weight percentage wt% pmma ec licf3so3 al2o3 sio2 spe1 55 18 25 spe2 55 18 25 2 (≤10µm) spe3 55 18 25 2 (≤10µm) 2.2 electrochemical impedance spectroscopy (eis) electrochemical impedance spectroscopy (eis) was carried out using gamry reference 600 series potentiostat for frequency ranging from 0.1hz to 1mhz at room temperature with the sample sandwiched between two spring-loaded stainless-steel electrodes. the ionic conductivity is calculated by using the following equation: 𝜎 = 𝑡 𝑅!𝐴 (1) , where t (cm) is the thickness of spe film, rb is the bulk resistance of electrolyte obtained from the nyquist plot, and a (cm2) is the contact area between the film and the electrode, which equates to 2.56 cm2 in this work. the results of pmma spe ionic conductivity were tabulated in table 2. 2.3 transference number characterisation the lithium transference number was defined as the number of moles of lithium-ion transferred for one faraday of charge transferred. ideally, the lithium transference number should be close to one in a high conductivity lithium polymer battery, and the lithium transference number, t+ is calculated as: 𝑡" = 𝐼# 𝐼$ (2) where io and is represent the initial and steady-state cell currents, respectively, however, the existing system is more complicated, with minute traces of contaminant of oxidation on the electrodes that can alter the experiment results. hence bruce et al. [12] introduced a correction factor to characterize the system before and after polarization, and the equation was rewritten as: 𝑡" = 𝐼#(𝛥𝑉 − 𝐼$𝑅$) 𝐼$(𝛥𝑉 − 𝐼#𝑅#) (3) ∆v is the polarization voltage of 20mv, ro and rs are the bulk resistance of electrolytes film taken into account interfacial resistance before and after the polarization, respectively (bruce, 1987). thus, the electrolyte film was characterized using eis before and after dc polarization to determine the initial resistance, ro, and the steady-state resistance, rs of the sample (pozyczka, 2017). 32 journal of mechanical engineering and technology issn 2180-1053 vol. 13 no. 2 december 2021 the bruce-vincent method is a reliable technique employed by multiple researchers to obtain lithium transference numbers in the polymer electrolytes system (pozyczka, 2017; yang, 2019; xiao, 2018). high interfacial resistance was always the factor in limiting the performance of spes. multiple researchers have reported that interfacial resistance was caused by the deposition of the oxide layer resulting from the diffusion of non-lithium ions during the chemical reaction. besides, poor contact between the electrode and electrolytes interface, originating from the two different materials mismatched lattice and the volume changes occurring during cycling also increase the interfacial resistance (ding, 2020; jiang, 2019). 2.4 voltammetry and discharging test linear sweep voltammetry (lsv) was used to investigate the electrochemical stability window of spes. the potential window, representing the sample stability voltage, is a critical parameter that has to be evaluated to determine the application voltage used in electrochemical devices (aziz, 2019). the voltage was swept from 0v (versus stainless steel electrode) at a scan rate of 5mvs-1 until the breakdown of electrolyte films (sharp increment of current) to determine the electrochemical stability voltage of the samples. then, five cycles of cyclic voltammetry (cv) were performed on the spe film from 0v to 3v at a scan rate of 20 mvs-1, in which the spe film was sandwiched between two stainless steel electrodes. lastly, the spe film was then assembled to evaluate its discharge performance. the cell was discharged with a current of 0.01ma at room temperature from 2v to 0.1v using stainless steel electrodes. all the measurements were carried out using a gamry reference 600 series potentiostat. 3.0 results and discussion figure 1. nyquist plot of pmma-licf3so3-ec (square), pmma-licf3so3-ec-al2o3 (triangle), and pmma-licf3so3-ec-sio2 (circle). 33 journal of mechanical engineering and technology issn 2180-1053 vol. 13 no. 2 december 2021 table 2. ionic conductivity of pmma-based solid polymer electrolytes. sample id sample composition filler filler size ionic conductivity (s/cm) spe1 pmma-licf3so3-ec 1.22 × 10-5 spe2 pmma-licf3so3-ec-al2o3 al2o3 10 μm 1.83×10-4 spe3 pmma-licf3so3-ec-sio2 sio2 10 μm 2.35×10-4 figure 1. shows the nyquist plot for spe1, spe2, and spe3, respectively (see figure 1). the appearance of semicircle for spe1 suggests the presence of bulk capacitance in the system. simultaneously, the linear line adjacent to the semicircle indicates the diffusion process resulting in electrode polarization effect (saikia, 2008; sivakumar, 2015). furthermore, no semicircle is observed for spe2, and spe3 suggests only resistive components prevail. the value of bulk resistance was determined at the interception point between the nyquist curve and the x-axis, which corresponds to 964 ω, and the obtained ionic conductivity is shown in table 2. ionic conductivity calculated for spe1 was 1.22×10-5 s/cm. in comparison, pal et al. obtained ionic conductivity of 3.52×10-5 s/cm for plasticized pmma-liclo4 spes (pal, 2018). kurapati et al. also reported ionic conductivity values of 8.21×10-5 s/cm for her study on pmma-ch3cooli spes (kurapati, 2019). the obtained ionic conductivity for spe1 was too low for electrochemical device applications (chauvin, 2006). the low ionic conductivity of spes is the attribute of their crystalline nature, where the ionic mobility is low. besides that, most of the li+ ions already formed polymer-salt complexes with the polymer, reducing the fraction of ions available for ionic conduction. improvement in ionic conductivity was seen for spe2 and spe3 when inorganic fillers al2o3 and sio2 of (≤10µm) were added, yielding an ionic conductivity of 1.83×10-4 s/cm and 2.35×10-4 s/cm, respectively. chew et al. reported similar findings where they also observed improvement in ionic conductivity when they added al2o3 into pmma spes (chew, 2011). they obtained an ionic conductivity of 2.05×10-4 s/cm for pmma spes with al2o3 fillers, an improvement from 1.36×10-5 s/cm without inorganic fillers. marcinek et al. explained that the improvement in ionic conductivity was associated with the increases in conduction pathways for ionic conduction provided by inorganic filler due to their larger surface area (marcinek, 2000). besides that, the presence of inorganic particles in the polymer matrix also lowers the fraction of polymer-salt complexes, which increases the fraction of li+ ions free for ionic conduction. saikia et al. (saikia, 2008) reported that the glass transition temperature (tg) and crystallinity of p(vdf-hfp)-pc-liclo4 decreases from -98 ̊ c to -104.6 ̊ c with the addition of 4 wt% sio2 aerogel particles. in addition, pitawala also reported decreases in tg from -44 ̊ c to -49 ̊ c when they added 10 wt% of al2o3 fillers into (peo)9litf furthermore, dissanayake et al. also reported that the presence of inorganic fillers promotes amorphous region in the polymer matrix by lowering the transition temperature (dissanayake, 2003). but they also informed that higher filler concentration above its optimal level would impose geometrical constriction, lowering the ion mobility. they reported 15 wt% of inorganic filler provided the maximum enhancement. furthermore, 34 journal of mechanical engineering and technology issn 2180-1053 vol. 13 no. 2 december 2021 another study conducted by yang et al. (yang, 2010) show that decreases in conductivity with increasing filler content beyond 2.5 wt% attributed to an aggregation of fillers, strongly impeding polymer chain movement. figure 2. dc polarization curve obtained via chronoamperometry for pmmalicf3so3-ec (spe1) samples. (inset nyquist curve of spe1 before and after dc polarization) table 3. transference number of pmma-based polymer electrolytes. sample id io (μa) is (μa) ro (ω) rs(ω) t+ spe1 0.617 0.055 964.1 1159 0.088 spe2 0.834 0.180 69.42 75.31 0.215 spe3 1.010 0.266 51.63 81.25 0.263 figure 2. shows the current versus time plot obtained using the dc polarization technique, the resistance before and after polarization, and the obtained lithium transference number was tabulated in table 3. the measured initial and steady-state currents were used to calculate the lithium transference number using eqn (3) and included in table 3. the lithium transference number obtained for spe1 is 0.088, which is slightly lower than 0.16 reported by xiao et al. for p(vdf-hfp)-lipf6-ec polymer electrolytes (xiao, 2018). the low lithium transference number reported for spe1 is associated with insufficient free li+ ions available for ionic conduction due to the formation of polymer-salt complexes. however, with micron-sized al2o3 and sio2 fillers in the polymer matrix, the lithium transference number increases up to 0.215 and 0.263 for spe2 and spe3, respectively. xiao et al. reported similar findings in their studies where the incorporation of nano-sized inorganic fillers enhances the lithium transference number (xiao, 2018). they obtained a lithium transference number of 0.28 to 0.41 at room temperature for p(vdf-hfp) polymer electrolytes with 5-15 wt% of nano pmma-zro2 particles, which is an improvement from 0.16 without nano pmma-zro2 particles. the improvement in lithium transference number is associated with the distribution of inorganic filler in the polymer matrix. 35 journal of mechanical engineering and technology issn 2180-1053 vol. 13 no. 2 december 2021 firstly, the presence of inorganic fillers reduces the fraction of polymer-salt complexes as inorganic fillers such as al2o3 and sio2 fillers also forms polymer-ceramic complexes with the polymer host. this means that there is more li+ ions will be available for ionic conduction. besides that, inorganic fillers also promote amorphous region in the polymer matrix as they reduced the degree of crystallinity of the polymer chain by lowering the glass transition temperature, tg, which in turn enhances the ionic mobility, thus leading to improved lithium transference number. table 4. lithium transference number obtained in this work compared to other researchers. composition type reference t+ pmma-licf3so3-ec solid this work 0.088 pmma-licf3so3-ec-al2o3 (≤ 10µm) solid this work 0.215 pmma-licf3so3-ec-sio2 (≤ 10µm) solid this work 0.263 pmma-liclo4-ec gel faridi [7] 0.39-0.42 pmma-lipf6-ec gel musil [8] 0.24-0.69 pmma-liclo4-pc-ec gel appetecchi [26] 0.40 pmma-libob-ec gel hosseinioun [9] 0.34 peo-licf3so3ec-sio2 (≤ 12nm) gel wang [27] 0.54 peo-libf4-ec-sio2 (≤ 12nm) gel liu [28] 0.34-0.56 p(vdf-hfp)-lipf6-ec-pmma-zro2 solid xiao [15] 0.28 table 4. shows the comparison for the lithium transference number obtained in this work to other literature. from the comparison, we can see that pmma spes still have a lower lithium transference number compared to gpes, as reported by (faridi, 2018; hosseinioun, 2019; liu, 2004). this is acceptable because gpes were much amorphous compared to spes due to their higher liquid solution. however, the higher liquid solution in gpes results in a mechanically soft system and lower thermal stability. although pmma spes incorporated with inorganic fillers has lower lithium transference number compared to gpes, but the presence of fillers in the polymer matrix served as a backbone for pmma spes that ensure better mechanical and thermal stability compared to gpes. liang et al. show that the dispersion of nano-al2o3 fillers into peo-pmma-litfsi polymer matrix enhances the tensile strength of spes up to 3.26 mpa compared to 2.78mpa for spes without inorganic filler (liang, 2015). nevertheless, the lithium transference number in our study is comparable to the work from xiao et al.. they reported lithium transference number ranging from 0.28 to 0.41 for p(vdf-hfp) polymer electrolytes doped with nano pmma-zro2 particles, close to 0.263 obtained for pmma-licf3so3-ec-sio2 in our study (xiao, 2018). the lithium transference number in their study is higher than ours because they were using 5%-15% of nano pmma-zro2 particles while we only added 2% of sio2 fillers. furthermore, the nano pmma-zro2 particles also have a greater effective surface area than micron-sized sio2 fillers, which also contributed to the higher lithium transference number obtained in their studies. figure 3. shows the lsv plot of the samples. the voltages were swept from 0v (versus stainless steel electrode) at a scan rate of 5 mvs-1 until a large increment of current was observed (see figure 3). the large increment in current observed indicates the breakdown 36 journal of mechanical engineering and technology issn 2180-1053 vol. 13 no. 2 december 2021 of the electrolyte films. the stability voltage can be determined as the intersection of the extrapolated linear current in the high voltage region with the voltage axis. figure 3. lsv curves with a scan rate of 5mvs-1 for pmma-licf3so3-ec (solid line), (b) pmma-licf3so3-ec-al2o3 (dash dotted line), and pmma-licf3so3-ec-sio2 (dash line). lsv results revealed that the potential window of the blend is all above 3.0v; spe1 (3.07v), spe2 (3.13v), and spe3 (3.2v). there was a notable improvement in the spe voltage stability window containing inorganic filler compared to the polymer salts system. in comparison, chandra et al. obtained an electrochemical stability voltage of 2.69v for pvc-pmma-licl-tio2 polymer blend, an improvement from 1.69v without tio2 nanoparticles (chandra, 2017). dhatarwal et al. also reported stability of 3.0v for peo-pmma polymer blend doped with various nanoparticles (dhatarwal, 2018). 37 journal of mechanical engineering and technology issn 2180-1053 vol. 13 no. 2 december 2021 figure 4. five cycles of cyclic voltammetry curve versus stainless steel electrodes with a scan rate of 20mvs-1 for (a) pmma-licf3so3-ec, (b) pmma-licf3so3-ec-al2o3 and (c) pmma-licf3so3-ec-sio2 the cv curve of the spes was shown in figure 4. (see figure 4). the absence of redox peak voltages and the overlapping of the subsequent sweeps indicates that the charging and discharging reaction at the interface between the electrolytes film and stainless-steel electrode is fully reversible for all the spe films tested (bandarayake, 2015). jinisha et al. reported cv curve of a similar shape for peo/pvp-lino3 spes in their studies (jinisha, 2017). they explained that the missing oxidation-reduction peaks in cv curves suggested that the spe film has great reversibility within the voltage range. furthermore, bandarayake et al. also reported that cv curves with subsequent overlapping sweeps show that the spe has good cycling capability and a longer lifetime. their samples can retain 96% of its original specific capacity after 20 cycles (bandarayake, 2016). figure 5. discharge curve with a discharge current of 0.01ma for pmma-licf3so3ec (black line), pmma-licf3so3-ec-al2o3 (red line) and pmma-licf3so3-ecsio2 (blue line) versus stainless steel electrodes furthermore, pmma-licf3so3-ec-sio2 that were doped with sio2 fillers samples has the highest energy density as evidenced by the discharging time as presented in figure 5. (see figure 5). it can be seen that pmma-licf3so3-ec-sio2 takes approximately 16% and 190% longer time to discharge compared to pmma-licf3so3-ec-al2o3 and pmma-licf3so3-ec, respectively. 4.0 conclusions this work provided the ionic conductivity, lithium transference number, and electrochemical performance of pmma spes. sio2 fillers enhanced pmma spes can exhibit ionic conductivity of 2.35×10-4 s/cm and lithium transference number of 0.263 at room temperature. it is suggested that inorganic fillers promote amorphous region in pmma spes, and the greater effective area of the fillers also provides pathways for ionic conduction. besides, the presence of inorganic fillers reduces the fraction of polymer-salt complexes, freeing more li+ ions for ionic conduction. the electrochemical studies also show that inorganic filler enhanced pmma spes has great cycling capability and 38 journal of mechanical engineering and technology issn 2180-1053 vol. 13 no. 2 december 2021 reversibility within a voltage range of 3.0v. inorganic filler enhanced spes may not exhibit superb ionic conductivity and lithium transference number like gpes, but they had better mechanical and thermal properties than gpes, which made spes much suitable for lower energy density but higher safety applications. 5.0 acknowledgement this work is supported by frgs/2019, ministry of higher education, malaysia. 6.0 references appetecchi, g. b., croce, f., & scrosati, b. (1997). high-performance electrolyte membranes for plastic lithium batteries. j. power sources, 66(1–2), 77–82. aziz, s., abdulwahid, r., & hamsan, m. (2019). proton conducting chitosan-based polymer blend electrolytes with high electrochemical stability. molecules, vol. 24, pp. 1–15, 2019. bandara, l. r. a. k., dissanayake, m. a. k. l., & mellander, b. (1998). ionic conductivity of plasticized (peo)-licf3so3 electrolytes. electrochimica acta., vol. 43, 1447-1451. bandaranayake, c. m., weerasinghe, w. a. d. s. s., vidanapathirana, k. p. (2016). a cyclic voltammetry study of a gel polymer electrolyte based redox-capacitor. sri lankan j. phys., 16(1), 19-27. chandra, m. v. l., karthikeyan, s., & selvasekarapandian, s. (2017). study of pvacpmma-licl polymer blend electrolyte and the effect of plasticizer ethylene carbonate and nanofiller titania on pvac-pmma-licl polymer blend electrolyte. j. polym. eng., 37(6), 617–631. chauvin, c., alloin, f., judeinstein, p., & foscallo, d. (2006). electrochemical and nmr characterizations of mixed polymer electrolytes based on oligoether sulfate and imide salts. electrochim. acta, 52(3), 1240–1246. chew, k. w., & tan, k. w. (2011). the effects of ceramic fillers on pmma-based polymer electrolyte salted with lithium triflate, licf3so3. int. j. electrochem. sci., 6(11), 5792–5801. dhatarwal, p., choudhary, s., & sengwa, r. j. (2018). electrochemical performance of li+ ion conducting solid polymer electrolytes based on peo–pmma blend matrix incorporated with various inorganic nanoparticles for the lithium ion batteries, compos. commun., 10, 11–17. ding, z., li, j., & an, c. (2020). review—interfaces: key issue to be solved for all solid-state lithium battery technologies. j. electrochem. soc., 167(7), 070541. 39 journal of mechanical engineering and technology issn 2180-1053 vol. 13 no. 2 december 2021 dissanayake, m. a. k. l., jayathilaka, p. a. r. d., bokalawala, r. s. p. & albinsson, i. (2003). effect of concentration and grain size of alumina filler on the ionic conductivity enhancement of the (peo)9licf3so3:al2o3 composite polymer electrolyte. j. power sources, 119–121, 409–414. evans, j., vincent, c. a., & bruce, p. g. (1987). electrochemical measurement of transference numbers in polymer electrolytes. polymer (guildf)., 28(13), 2324– 2328. faridi, m., naji, l., kazemifard, s., & pourali, n. (2018). electrochemical investigation of gel polymer electrolytes based on poly (methyl methacrylate) and dimethylacetamide for application in li-ion batteries. chem. pap., 72(9), 2289– 2300. hosseinioun, a., nürnberg, p., schönhoff, m., & diddens, d. (2019). improved lithium ion dynamics in crosslinked pmma gel polymer electrolyte. rsc adv., 9(47), 27574–27582. jiang, z., han, q., wang, s., & wang, h. (2019). reducing the interfacial resistance in all-solid-state lithium batteries based on oxide ceramic electrolytes. chemelectrochem, 6(12), 2970–2983. jinisha, b., manoj, m., & pradeep, p. (2017). development of a novel type of solid polymer electrolyte for solid state lithium battery applications based on lithium enriched poly (ethylene oxide) (peo)/poly (vinyl pyrrolidone) (pvp) blend polymer. electrochim. acta, 235, 210–222. kurapati, s., gunturi, s. s., nadella, k. j., & erothu, h. (2019). novel solid polymer electrolyte based on pmma:ch3cooli effect of salt concentration on optical and conductivity studies. polym. bull., 76(10), 5463–5481. liang, b., tang, s., jiang, q., & chen, c. (2015). preparation and characterization of peo-pmma polymer composite electrolytes doped with nano-al2o3. electrochim. acta, 169, 334–341. lim, y. s., jung, h. a., & hwang, h. (2018). fabrication of peo-pmma-liclo4-based solid polymer electrolytes containing silica aerogel particles for all-solid-state lithium batteries. energies, 11(10), 2559. liu, y., lee, j. y., & hong, l. (2004). in situ preparation of poly (ethylene oxide)-sio2 composite polymer electrolytes. j. power sources, 129(2), 303–311. marcinek, m., bac, a., & lipka, p. (2000). effect of filler surface group on ionic interactions in peg-liclo4-al2o3 composite polyether electrolytes. j. phys. chem. b, 104(47), 11088–11093. musil, m., & vondrak, j. (2014). transference number measurements on gel polymer electrolytes for lithium-ion batteries. ecs transactions, 63(1), 315-319. 40 journal of mechanical engineering and technology issn 2180-1053 vol. 13 no. 2 december 2021 osińska, m., walkowiak, m., & zalewska, a. (2009). study of the role of ceramic filler in composite gel electrolytes based on microporous polymer membranes. j. memb. sci., vol. 326, no. 2, pp. 582–588. pal, p., & ghosh, a. (2018). investigation of ionic conductivity and relaxation in plasticized pmma-liclo4 solid polymer electrolytes. solid state ionics, 319, 117–124. pitawala, h. m. j. c., dissanayake, m. a. k. l. & seneviratne, v. a., (2007). combined effect of al2o3 nano-fillers and ec plasticizer on ionic conductivity enhancement in the solid polymer electrolyte (peo)9litf. solid state ionics, 178(13–14), 885–888. pożyczka, k., marzantowicz, m., dygas, j. r. & krok, f. (2017). ionic conductivity and lithium transference number of poly (ethylene oxide): litfsi system. electrochim. acta, 227, 127–135. saikia, d., chen, y. t., li, y. k., & lin, s. i. (2008). investigation of ionic conductivity of composite gel polymer electrolyte membranes based on p(vdf-hfp), liclo4 and silica aerogel for lithium-ion battery. desalination, 234(1–3), 24– 32. sivakumar, p., & gunasekaran, m. (2015). highly porous polymer electrolytes based on p(vdf-hfp)/ pema with propylene carbonate/diethyl carbonate for lithium battery applications. int. j. energy power eng. int. j. energy power eng. spec. issue energy syst. dev., 4(5), 17–21. song, c., xu, c., & chen, y. (2015). enhanced thermal and electrochemical properties of pvdf-hfp/pmma polymer electrolyte by tio2 nanoparticles. solid state ionics, 282, 31–36. sun, c. c., you, a. h., & teo, l. l. (2019). characterizations of pmma-based polymer electrolyte membranes with al2o3. j. polym. eng., 39(7), 612–619. wang, w. & alexandridis, p. (2016). composite polymer electrolytes: nanoparticles affect structure and properties. polymers (basel)., 8(11), 387. xiao, w., wang, z., & zhang, y. (2017). enhanced performance of p(vdf-hfp)-based composite polymer electrolytes doped with organic-inorganic hybrid particles pmma-zro2 for lithium ion batteries. j. power sources, 382, 128–134. yang, j., wang, x., zhang, g., ma, a., & chen, w. (2019). high-performance solid composite polymer electrolyte for all solid-state lithium battery through facile microstructure regulation. front. chem., 7, 1–11. yao, p., yu, h., & ding, z. (2009). review on polymer-based composite electrolytes for lithium batteries. front. chem., 7, 1–17. 41 journal of mechanical engineering and technology issn 2180-1053 vol. 13 no. 2 december 2021 zakariya’u, i., gultekin, b., singh, v. (2020). electrochemical double-layer supercapacitor using poly(methyl methacrylate) solid polymer electrolyte. high perform. polym. 32(2), 201–207. 42 issn 2180-1053 e-issn 2289-8123 vol.14 no.2 17 artificial neural network prediction of performance characteristics of biofuel produced from sweet potato (ipomoea batata) y.k. abubakar1, b. bongfa1, m. shaibu1, o.g. onomen1 and u.j. tokula1 1department of mechanical engineering, federal polytechnic p.m.b 1037, idah, kogi state, nigeria. corresponding author’s email: 1babayas8069@gmail.com article history: received 25 may 2022; revised 01 november 2022; accepted 1 december 2022 abstract: fossil fuel depletion and the harm it causes to the environment has led to the development of alternative fuels. in this research, biofuel (ethanol) was produced and characterized from sweat potatoes. blends of premium motor spirit with 0% (e0), 2% (e2), 4% (e4), and 10% (e10) of the produced biofuel at various percentages were separately used to power a fourstroke, single-cylinder si engine on an engine test bed, and data of the engine performance brake power, brake torque, brake mean effective pressure (bmep), and the exhaust gas temperature reported in each test. the results of the physicochemical analysis revealed that the physical state of the biofuel is colorless, the viscosity at 300c, density, calorific value, and ph level are 0.9834 mpa.s, 0.85 g/cm3,19 kj/kg, and 1.82, respectively. it was observed that an increase in ethanol in the blend increases the performance of the engine, although the bmep at e0 gave the highest value of 0.3 bar compared to other blends. an artificial neural network (ann) model for predicting engine performance characteristics was developed, trained, validated, and tested using the reported data. the result of the ann model revealed that the levenberg-marquardt training algorithm (lmta) with 10 hidden layer neurons offers the best fit for the features for both training, validation, testing, and overall. with the r for training equal 1, validation equal to 0.99468, testing equal to 0.90103, and overall r equal to 0.93842 as compared to the rest in terms of the number of neurons and training algorithms. keywords: biofuel, characterization, engine performance, artificial neural network, algorithm journal of mechanical engineering and technology (jmet) 18 issn 2180-1053 e-issn 2289-8123 vol.14 no.2 1.0 i n t ro d uc t i o n energy is an important commodity that contributes to the standard of living and economic growth of humanity. between 2010 and 2030, global basic energy demand is expected to rise by 1.6% per year [1]. a larger percentage of the primary energy consumed is gotten from fossil resources, essentially coal (29%), crude oil (35%), and natural gas (24%), while nuclear and renewable resources constitute about 7% and 5% of world energy consumption, respectively [1]. fossil fuels are thus the largest source of energy, accounting for 88% of the total global energy consumption. meanwhile, fossil fuels are being consumed rapidly. in 2013, it was projected that a peak in the world’s crude oil production would occur between 2015 and 2030 based on the then production rates [2]. furthermore, internal combustion (ic) engines are the primary source of poisonous gas emissions that have negative effects on the environment and human health, such as acid rain, greenhouse effect, global warming, unprecedented flooding, and other negative effects. the search for alternative fuels is on the rise for possible emission reduction, reducing fuel prices, providing clean energy, improving fuel availability, and reducing reliance on fossil fuels [3,4]. a great deal of biomass has been considered for producing sustainable alternative fuels. however, the most combative issue with the production of this biomass is the use of agricultural land for biomass production [2]. this means that lands that were intended to produce foods to meet consumption requirements are now being used for biomass production. meanwhile, sweet potatoes (ipomoea batata) are grown in nigeria, especially in kogi, benue, plateau (bui), taraba (mambilla plateau), and borno in commercial quantities [5]. nigeria is among the largest producers of sweet potatoes in sub-saharan africa, with yearly production estimated at 4.03 million tons per year [6]. also, sweet potatoes can do well both in tropical and temperate regions, whereas irish potatoes grow well only in temperate regions [5]. consequently, undertaking experiments with vehicle engines and determining fuels’ multiple parameters requires substantial amounts of fuel, which can be a challenge from new sources [2]. moreso, fuel evaluation requires sophisticated equipment and expert personnel, which can be costly. for this reason, ann that can predict from small data sets was employed. artificial neural network prediction of performance characteristics of biofuel produced from sweet potato (ipomoea patata) issn 2180-1053 e-issn 2289-8123 vol.14 no.2 19 2.0 materials the following materials were employed for this study: i. the biomass used for this research work is sweet potato, which was sourced from a local market (e.g. a general market), in idah area of kogi state. ii. a weighing balance was used for weighing the starch extracted from the biomass in the laboratory and the reagents used. iii. a grader was used to grade the biomass (sweet potato) into smaller sizes. iv. a conical flask/beaker was used to prepare the solution and collect the biofuel during the distillation process. v. a thermometer was used for taking temperature readings during the production of biofuel. vi. reagents: reagent such as tetraososulphate (vi) acid (h2so4) was used to hydrolyse the compound (the starch), alpha-amylase (dehydrogenase enzyme) was used for breaking down starches in grains into fermentation sugars, glucoamylase (amyloglucosidax) was used for breaking down starches into glucose, and saccharomyces levadura yeast was used to convert the sugar after liquefaction and saccharification process into biofuel. 3.0 method 3.1 biofuel production from sweet potatoes the experimental procedure for the production of biofuel from sweet potatoes is shown in figure 1. the biofuel production procedure was adopted from salelign and duraisamy in the year 2021 [7]. in this work, 400 g of sweet potato starch was mixed with 1600 ml of distilled water (1:4). the mixture was stirred properly and heated at a temperature of 800c for the gelatinization process. during the liquefaction process, tetraososulphate (vi) acid (h2so4) was added to achieve the ph level. the sample was allowed to cool down to a temperature of 650c, then 100 g of the alpha-amylase was mixed into 250 ml of distilled water and poured into the sample to liquefy the heated sample slowly. consequently, the sample was allowed to cool down to a temperature of 300c through natural convection. after that, 10 g of the glucoamylase was mixed with 20 ml of distilled water and poured into the sample. the sample was further cooled, and 40 g of saccharomyces cerevisiae yeast was mixed with 100 ml of distilled water and poured into the sample. the sample was kept for 72 hours at room temperature in a closed vessel for the fermentation process. the fermentation process lasted 72 hours. the biofuel crude was transferred to the distillation apparatus for the distillation process where the biofuel was collected. journal of mechanical engineering and technology (jmet) 20 issn 2180-1053 e-issn 2289-8123 vol.14 no.2 figure 1: experimental procedure for biofuel production from sweet potatoes 3.2 physicochemical characterization of the biofuel produced physicochemical characterization of the biofuel produced was achieved through the american society for testing and material (astm) standard test procedure. astm d3588, d1298-99, d445, and d6423 were used to determine the heating value, density, viscosity, and ph value of the fuel produced. 3.3 engine performance evaluation of blends of the biofuel produced the engine performance was carried out using a tq200 small engine testbed with specifications shown in table 1. the experimental setup consists of four strokes, and a single-cylinder carburetor si engine coupled with a dynamometer for load control. the instrumentation unit is connected to a computer system using the versatile data acquisition system (vdas) software for taking readings. torque, engine speed, brake power, brake mean effective pressure (bmep), and exhaust gas temperature of the blends were measured. the experimental setup for the evaluation of engine performance is shown in figure 2. during the performance evaluation test, the biofuel was blended with premium motor spirit (pms) obtained from a reliable filling station in idah area of kogi state, nigeria. the blends were code named e2, e4, e6, e8, and e10, respectively. where e2 represents 2% of the biofuel produced, mixed with 98% of the pms, and so on. the performance evaluation was carried out for each blend at a constant speed of 3,000 rpm, and the results are presented in the result and discussion sections artificial neural network prediction of performance characteristics of biofuel produced from sweet potato (ipomoea patata) issn 2180-1053 e-issn 2289-8123 vol.14 no.2 21 table 1: td200 small petrol engine testbed (tq brand) specifications engine type 4-stroke, single cylinder continuous rated power 2.6 kw at 3000 rpm 2.9 kw at 3600 rpm bore/stroke/crank radius 67mm/49mm/24.5mm displacement vol. 172cc compression ratio 8.5:1 engine cool/fuel water cool/gasoline (petrol) ignition system electric maximum dynamometer rating 7.5 kw at 7000 rpm figure 2: experimental set up for engine performance test 3.4 engine performance characteristics prediction of the biofuel using artificial neural network (ann) due to the sophisticated nature of the equipment used for fuel testing, and the scarcity of expert personnel needed in fuel evaluation, and also, high cost of undertaking experiments that relate to fuel produced from new sources, an ann model was developed to help in the rapid evaluation of new fuel that might be produced from sources similar to the source used in this study. an artificial neural network is a collection of interconnected neurons grouped into a network [8]. the network consists of the input layer, hidden layer, and an output layer. the input layer accepts data (features) and communicates the data with the hidden layer. the hidden layer is where the computation and activation of neurons occurs. the output layer gives the predicted results. activation of the neurons depends on the type of activation function (such as sigmoid, relu, threshold, hyperbolic tangent, etc.) used. however, irrespective of the activation function, weight journal of mechanical engineering and technology (jmet) 22 issn 2180-1053 e-issn 2289-8123 vol.14 no.2 parameterization is what is taking place in the hidden layer, which is then summed up and caused to activate the neurons. for example, assuming there are 𝑥1, 𝑥2, 𝑥3, … , 𝑥𝑛 number of inputs, and 𝑤1, 𝑤, 𝑤3, … , 𝑤𝑛 number of weight, respectively, the sum of the products of the inputs and weight (𝑖 = 1 to 𝑖 = 𝑛) as shown in equation 1, produced the activation, a, of the neuron [8]. 𝑎 = ∑ 𝑥𝑖 𝑤𝑖 𝑛 𝑖=1 (1) for this work, the ann model as in figure 3 accepts the engine speed and fuel-blend ratio as the inputs to the network. the torque, brake mean effective pressure (bmep), brake power (bp), and exhaust gas temperature (egt) of the blends gotten from the testbed were entered into the network as the target outputs all photographs and figures should have good resolution, and contrast quality. at least 300 dpi is applied for the resolution. the levenberg-marquardt training algorithm (lmta) and the bayesian regularization algorithm (bra) were used for training the network. the training was done for three sets of hidden layer neurons the first, second, and third sets consist of 10, 15, and 7 neurons, respectively. the data was divided into three random sets: training, testing, and validation data set which are 70%, 15%, and 15%, respectively. figure 3: artificial neural network for predicting engine performance characteristics of biofuel artificial neural network prediction of performance characteristics of biofuel produced from sweet potato (ipomoea patata) issn 2180-1053 e-issn 2289-8123 vol.14 no.2 23 4.0 results and discussion 4.1 physicochemical properties of the biofuel produced the physicochemical properties of the characterized biofuel are summarized in table 2. the biofuel is colorless and has an affinity for water and is readily miscible with pms. table 2 shows that the biofuel has a density of 0.85 g/cm3 and the viscosity of the biofuel produced was 0.9834 mpa.s at a temperature of 300c. biofuel and pms have very close specific gravity, making them miscible to form a homogeneous substance [4]. the result also shows that the ph (1.82) level of the produced biofuel is very low, which is responsible for a very high acidic content compared to the ph level of pms and a standard biofuel. table 2: physicochemical properties of the biofuel from sweet potatoes s/n parameter unit result 1 physical state colorless 2 density g/cm3 0.85 3 viscosity at 300c mpa.s 0.9834 5 calorific value kj/kg 19 6 ph 1.82 4.2 engine performance evaluation 4.2.1 torque the result of the engine performance test conducted shows that as the percentage of the biofuel increases in the blend, the torque equally increases as shown in figure 4, with the highest value recorded by e10. this corresponding increase in torque may be due to the low heating value, high density, and higher latent heat of evaporation of the biofuel compared to that of base gasoline [9]. the produced biofuel, therefore, exhibits the potential to increase the ability of an engine to perform work, which agrees with the work of [4]. 4.2.1 exhaust gas temperature (egt) the behaviour of egt with a change in biofuel blends is shown in figure 5. during the performance test, it was observed that as the biofuel percentage increased in the blend, there was an increase in the exhaust gas temperature at a constant speed of 3000 rpm. meanwhile, the increase in the egt between e0 and e2 is twice as high as that of e2 to e10. this behaviour shows that the biofuel produced moderates egt as the blend increases. figure 5 reveals that e10 has a maximum value of egt, which is 4810c. the scope of this work does not cover the emission journal of mechanical engineering and technology (jmet) 24 issn 2180-1053 e-issn 2289-8123 vol.14 no.2 analysis of the produced biofuel. however, according to [10] an increase in egt at a high speed of around 6000 rpm causes about three times increase in nitrogen oxide (nox) emission. and basically, egt is an indication of the air-to-fuel mixture of an engine. figure 4: the engine torque (n/m) with ethanol blend (%) figure 5: egt (0c) of the biofuel blend (%) 4.2.3 brake power (bp) the brake power is the power available at the crankshaft. in the case of the ic engine, it is the output power. from the engine performance test, the result shows that the biofuel blends increase the brake power, respectively, at a constant speed of 3000 rpm, as shown in figure 6. therefore, using a fuel blend with biofuel is useful to improve the engine power output. artificial neural network prediction of performance characteristics of biofuel produced from sweet potato (ipomoea patata) issn 2180-1053 e-issn 2289-8123 vol.14 no.2 25 figure 6: engine performance on brake power (w) with biofuel blend (%) 4.2.4 brake mean effective pressure (bmep) the brake mean effective pressure is a calculation of the engine cylinder pressure that would give the measured brake power. it is an indication of engine efficiency regardless of capacity or engine speed. from the engine performance test, it was also observed that the brake mean effective pressure (bmep) of the pure pms has the highest value (0.3 bar) in comparison to the ethanol blend of e2, e4, e6, e8, and e10, respectively, at a constant speed of 3000 rpm as shown in figure 7. therefore, the pms gave the higher engine efficiency, meanwhile, e4 indicates a competitive engine efficiency having a bmep value of 0.28%. figure 7: the bmep (bar) of an engine with different ethanol blend (%) journal of mechanical engineering and technology (jmet) 26 issn 2180-1053 e-issn 2289-8123 vol.14 no.2 4.3 artificial neural network model of the engine performance characteristics the neural network model revealed that the lmta training algorithm with 10 hidden layer neurons as in figure 8 offers the best fit for the features for both training, validation, and testing as compared to fifteen and seven-layer neurons as is figure 9 and figure 10, respectively. with the, r, value for training equal to 1, validation equal to 0.99468, testing equal to 0.90103, and overall, r, equal to 0.93842 as compared to the rest in terms of the number of neurons and training algorithms. figure 9 to figure 10 are the regression plots for the 15 neurons and 7 neurons optimized with lmta. figure 8: regression for lmta ten (10) hidden neurons figure 9: regression for lmta fifteen (15) hidden neurons artificial neural network prediction of performance characteristics of biofuel produced from sweet potato (ipomoea patata) issn 2180-1053 e-issn 2289-8123 vol.14 no.2 27 figure 10: regression for lmta seven (7) hidden neurons 5.0 conclusion biofuel has been produced from sweet potato and characterized. the results showed that it has a density of 0.85 g/cm3, viscosity of 0.9834 mpa.s, calorific value of 19 kj/kg, and a ph level of 1.82. the torque, brake mean effective pressure (bmep), brake power (bp), and exhaust gas temperature (egt). engine performance evaluation on separate mixtures of commercial gasoline fuel and the produced biofuel revealed that there is about 20% increase in brake power (bp) as the ethanol content increases in the blend. the petrol e0 happens to have the highest brake mean effective pressure (bmep) compared to the blends. the torque of the engine increases as the blend increases as a result of high density and low heating value. the exhaust gas temperature (egt) increases as the blend of ethanol increases as a result of improved combustion, and the speed was constant throughout the experimental process at 3000 rpm. consequently, for the ann model, the lmta training algorithm with 10 hidden layer neurons offers the best fit for the features, having a r value for training equal to 1, validation equal to 0.99468, testing equal to 0.90103, and overall r equal to 0.93842. journal of mechanical engineering and technology (jmet) 28 issn 2180-1053 e-issn 2289-8123 vol.14 no.2 6.0 recommendation the biofuel produced from sweet potato (ipomoea batata) as an alternative fuel for internal combustion engines has shown positive engine performance characteristics when blended with pms. however, there is a need for emission analysis to be carried out to check the impact of its use on the environment. references [1] bp energy outlook 2030, 2012 ed., energy outlook., 2012, 230. [2] i.m. jahirul, j.r. brown, w. senadeera, m.o. hara and z.d. ristovski, “the use of artificial neural networks for identifying sustainable biodiesel feedstocks”, energies, vol. 6, issue 8, pp. 3764-3806, 2013. [3] m.s. koc, y.t. topgul and h.s. yucesu, “the effects of ethanol-unleaded gasoline blends on engine performance and exhaust emissions in a spark-ignition engine”, renewable energy, vol. 34, no. 10, pp. 2102-2106, 2009. [4] a.s. olawore, w.i. oseni, k.o. oladosu and e. fadele, “performance evaluation of a single cylinder spark ignition engine fuelled by mixing ethanol and gasoline”, journal of applied sciences and environmental management, vol. 25, no. 6, pp. 1-6, 2021. [5] e. johnson (2018). how to start sweet potato farming in nigeria: complete guide. [online: agro and business blog]. available https://www.enibest.com.ng/all-posts/agriculture/sweet-potatofarming/#google_vignette [6] m.e. ejechi, i.o. ode and e. sugh, “empirical analysis of production behaviour among small-scale”, nigeria agricultural journal, vol. 51, no. 1, pp. 17-21, 2020. [7] k. salelign and r. duraisamy, “heliyon sugar and ethanol production potential of sweet potato (ipomoea batatas) as an alternative energy feedstock: processing and physicochemical characterizations”, heliyon, vol. 7, no. 11, 2021. [8] k. gurney, an introduction to neural networks. taylor & francis, 2004. [9] k. kapil and a. nayyar, “effects of ethanol gasoline blends on performance and emissions of gasoline engines”, international research journal of engineering and technology, vol. 04, no. 1, pp. 1092-1096, 2017. https://www.enibest.com.ng/all-posts/agriculture/sweet-potato-farming/#google_vignette https://www.enibest.com.ng/all-posts/agriculture/sweet-potato-farming/#google_vignette artificial neural network prediction of performance characteristics of biofuel produced from sweet potato (ipomoea patata) issn 2180-1053 e-issn 2289-8123 vol.14 no.2 29 [10] j. zareei, a. rohani, f. mazari and m. vladimirovna, “numerical investigation of the effect of two-step injection (direct and port injection) of hydrogen blending and natural gas on engine performance and exhaust gas emissions”, energy, vol. 231, 2021. journal of mechanical engineering and technology *corresponding author. email: ddanardono@staff.uns.ac.id issn 2180-1053 vol. 14 no. 1 july – december 2022 analysis of addition the number of half circle type slot on performance characteristics of disc conductor eddy current brake alfian jihan saputra1, dominicus danardono dwi prija tjahjana1,2*, muhammad nizam2,3,4, mufti reza aulia putra1 1 mechanical engineering department, faculty of engineering, universitas sebelas maret, jl. ir. sutami 36a, surakarta 57126, indonesia 2 national center of sustainable transportation technology (ncstt) itb, bandung 40132, indonesia 3 electrical engineering department, faculty of engineering, universitas sebelas maret, jl. ir. sutami 36a, surakarta 57126, indonesia 4 lithium battery research and technology centre, universitas sebelas maret, jl. slamet riyadi 435, surakarta 57146, indonesia abstract brake is a vital component of a vehicle, particularly for motor vehicles. one of the braking used the principle of eddy current brake by utilizing electromagnetic. eddy current brake is a braking technology without direct contact by utilizing eddy currents. eddy current brake performance can be influenced by several factors, one of them is the surface shape of the disc conductor. using finite element simulation, this research examines the impact of increasing the number of slot half-circles on the performance of the eddy current brake with the number of slot changes. variations number of slots that used are 6, 8, 10, and 12 slots. the result of this study obtained the best braking torque value in the variation with the number of 10 slots at a rotational speed of 450 rpm with a 15,930 nm torque value. the addition slots of the number of half-circle types have a less significant effect on the torque from the simulation. keywords: eddy current brake, finite element method, half-circle slotted 1.0 introduction vehicles currently use a lot of conventional brakes to slow down the speed and the most common types of brakes are drum brakes and disc brakes. in the operation of both types of brakes, the friction concept is applied. friction that occur will change from kinetic energy to heat energy when the brake operating (günay et al., 2020). the vehicle's speed will be reduced by friction until it comes to a halt. however, the friction between the two objects will almost definitely produce an increase in temperature surrounding the brakes (gerdes & hedrick, 1999). to solve the problems developed technology on braking. using an eddy current brake (ecb) is one of them (mufti reza aulia putra et al., 2020). eddy current brakes are 1 mailto:ddanardono@staff.uns.ac.id issn 2180-1053 vol. 14 no. 1 july – december 2022 intended to be used in place of konventional brakes (cho, liu, lee, et al., 2017). karakoc also used the fem method to analyze the dispersion of eddy currents. in his research, it was found that ecb braking can be used to replace conventional brakes (karakoc et al., 2016). eddy current braking is a sort of electric braking that relies on eddy currents for braking (lequesne, 1997). the ecb braking system is frictionless braking which has advantages such as not producing wear, making little noise, and having a fast response (cho, liu, ahn, et al., 2017). ecb is now being developed for light vehicles, such as motorcycles, in addition to being employed in large vehicles (sinmaz et al., 2016). in the development of ecb braking, robert, et al. analyze the surface design of disk conductors by adding slots. the slot can be defined as a hollow shape on the surface of the disc that serves to improve braking performance. the addition of the number of slots in this study affects the resulting braking torque. the results indicated that the brake with a slotted disk rotor had 1.1–1.2 times the torque of the brake with a plain rotor (robert, 2017). the form of the conductor slot's surface is next investigated by prayoga et al. compared to other shapes, the conductor disc with a semicircular shape has a maximum torque value (prayoga et al., 2019). in this study, research will be carried out with variations in the number of slots on a semicircular type of conductor plate. the studies in the research is how the number of slots on a semicircular type conductor plate impacts the ecb's braking torque when using a single magnet. 2.0 methodology 2.1 governing equation the resulting torque can be influenced by the size of the magnetic field that arises (cho, liu, lee, et al., 2017). the amount of torque on braking can be calculated through the following equation (rodrigues et al., 2016): 𝑇 = 1 2 × 𝜎𝛿𝜔𝜋𝑟2𝑚2𝐵𝑧 2 [1 − { ( 𝑟 𝑎 ) 2 (1−( 𝑚 𝑎 ) 2 ) 2}] (1) where 𝜎 is the electrical conductivity of the disc (s/m), 𝛿 = disc thickness (mm), 𝜔 = angular velocity (rad/s), r = electromagnetic radius (mm), m = distance between disc axis and magnetic axis position (mm), 𝐵𝑧 . = magnetic flux density (t), and 𝑎 = disc radius (mm). in equation 2.8 it can be seen that there are several factors that affect the braking torque on the eddy current brake. the influencing factors include the choice of material on the disc conductor, rotational speed, surface area of the electromagnet, magnetic flux density, disc thickness, and disc radius. the area of the electromagnet can also be interpreted as the area of the active region. the active region is the area on the conductor that is perpendicular to the magnetic face, while the passive region is the area that is affected by the magnetic field around the magnetic face. the size of the active region is influenced by the dimensions of the magnetic face and the surface area of the conductor perpendicular to the magnetic face. the ecb works according to faraday's law. when a conductor cuts the magnetic force line, it will produce a loop whose magnitude is proportional to the magnetic field and velocity of the conductor (satya et al., 2021). faraday's law equation can be seen in equation (2). according to lenz's law, the rotating loop will generate a new magnetic 2 issn 2180-1053 vol. 14 no. 1 july – december 2022 field. as a result, the magnet will provide a drag force on the moving conductor (satya et al., 2021). lenz's law equation can be seen in equation (3). conductivity is also important in ecb design because the eddy current is proportional to the eddy current conductivity shown in equation (4). the interaction between eddy currents and magnetic flux density produces a braking force as in equation (5) so that the braking torque equation can be obtained in equation (6). ∇ × �⃗� = − 𝜕�⃗� 𝜕𝑡 (2) 𝑒 = (𝑣 . 𝐵). 𝑙 (3) 𝐽 = 𝜎. �⃗� (4) 𝐹 = 𝐽 × �⃗� (5) �⃗� = 𝐹 . 𝑟 (6) where �⃗� = magnetic field intensity (n/c), �⃗� = magnetic induction (t), 𝑒 = induced emf (v), 𝑣 = velocity (m/s), 𝑙 = conductor length (m), 𝐽 = eddy current density (a/m2), 𝜎 = conductivity of the material (s/m), 𝐹 = braking force (n), �⃗� = torque (nm), and r = radius (m). 2.2 developed model the finite element method (fem) is used in this study's simulation process. fusion 360 software was used for modeling, and ansys electronics 2018 was used for simulation. simulation data is displayed in the form of a curve that represents the rotational speed and braking torque generated for each modification. this simulation uses parameters that affect the performance of the ecb, namely by varying the effect of increasing the number of half-circle slots and comparing the data between all simulation results. variations in the number of half-circle slots used in this study are 6, 8, 10, and 12 slots. as for variations in rotational speed include 150, 300, 450, 600, and 750 rpm. the speed variation is also taken from low to high speed in order to know the difference in its effect on braking performance. then the variation in the number of slots of the halfcircle type is carried out in order to find out about the difference in the number of slots that will affect the performance of the eddy current brake which will be better. the variation in the number of slots was chosen based on the fact that the more slots with the half-circle type, the better the braking process performance (razavi & lampérth, 2006). the difference in distance between slot variations with a difference of 2 slots is used to find out more details and specifically regarding changes in the addition of braking torque generated in the simulation. the speed variation used refers to the average speed of the vehicle in general. figure 1 shows the design that will be used in the simulation. (a) (b) figure 1. (a) 3d model of a unipolar axial ecb, (b) eddy current design variable (waloyo et al., 2020). 3 issn 2180-1053 vol. 14 no. 1 july – december 2022 table 1 lists the parameters that were employed in this study's simulation design. tabel 1. design parameter of ecb (mufti reza aulia putra et al., 2020). variable value unit current (i) 20 a count of coil (n) 360 length of pole shoe (a) 30 mm wide of pole shoe (b) 12.5 mm total length of winding core (l) 248 mm distance from center to center of pole shoe (r) 83.5 mm air gap 0.5 mm thick of disc (d) 5 mm radius disc brake (r) 120 mm relative permeability aluminium (μal) 1.26 x 10 -6 h· m-1 relative permeability iron (μfe) 6.3×10 -3 h· m-1 aluminium conductivity (σ) 36.9 × 106 s· m-1 disc modeling consists of 3 layers, namely disc groove layer, mid layer, and flat layer. where the groove layer is a layer that is given a half-circle slot, while the flat layer is a layer that is not given a slot. the material used on the disc consists of 2 materials, namely iron material for the mid section and aluminum material for the flat disc section and disc groove (ubaidillah et al., 2020). the disc parts used can be seen in figure 2a and the dimensions of the half-circle slots are shown in figure 2b, where a = thickness, b = slot width, and c = slot depth. (a) (b) figure 2. (a) ecb disc section, (b) aluminum disc slot dimensions. the meshing used in this study uses a meshing size of 5 mm and the meshing shape is used with a tetrahedral model (m. r.a. putra et al., 2019). figure 3 shows an image of the meshing process in the simulation. the materials used in this simulation include aluminum, iron, copper, and air (vacuum). the value of the material properties of these materials is shown in table 2. figure 3. the meshing process in the ecb simulation 4 issn 2180-1053 vol. 14 no. 1 july – december 2022 tabel 2. material properties materials relative permeability (μr) bulk conductivity (s/m) mass density (kg/m 3 ) aluminium 1.000021 3.8 x 107 2689 iron 4000 1.03 x 107 7870 copper 0.999991 5.8 x 107 8933 vacuum 1 0 0 3.0 result and discussion 3.1 verification data validation was carried out by re-modelling in accordance with the modeling and simulation of previous research. the validation was carried out by comparing the data to the previous research conducted by putra (m. r.a. putra et al., 2019). the goal of this simulation is to determine the braking torque and initial flux. modeling is done by creating a 3d design using fusion 360 software. the modeling that will be used for data validation is designed as closely as possible to the design to be compared, so that it can produce an assessment similar to the previous design. validation will hold the simulation's parameters accountability. table 3 shows a table of data validation against previous studies. table 3. table of data validation against previous research rotating speed (rpm) braking torque (nm) deviation putra’s research (m. r.a. putra et al., 2019) new research (%) |%| 150 5.086 4.847 4.7 4.7 300 8.300 7.937 4.37 4.37 450 9.429 9.120 3.28 3.28 600 9.383 9.129 2.71 2.71 750 8.828 8.542 3.24 3.24 average 3.66 table 3 shows that the average deviation in this study is 3.66%. this shows that the results of the comparison of data between previous studies and new studies have a deviation of less than 5%. it can be said that the modeling and simulation results are valid. 3.2 braking torque the simulation results for each variation are shown in table 4. table 4. ecb simulation result data no rotating speed (rpm) braking torque (nm) 6 slots 8 slots 10 slots 12 slots 1 150 8.86 8.99 9.18 9.02 2 300 14.09 14.27 14.51 14.26 3 450 15.52 15.71 15.93 15.72 4 600 14.93 15.12 15.30 15.07 5 750 13.66 13.70 13.80 13.63 5 issn 2180-1053 vol. 14 no. 1 july – december 2022 it can be seen that the largest braking torque in each variation of the number of slots is found at a rotational speed of 450 rpm and the smallest braking torque in each variation is found at a rotational speed of 150 rpm. the highest braking torque value produced is the number of slots 10 with a speed of 450 rpm of 15.93 nm. after the speed passes 450 rpm, the resulting data will tend to decrease as the rotational speed increases. this is in accordance with the nature and characteristics of the disc conductor used from nonferrous material, namely aluminum which causes the emergence of critical speed during the braking process (kou et al., 2014). figure 4 shows the relationship between braking torque and rotational speed. the graph shows that each variation in the number of slots has almost the same graphic pattern. figure 4 shows that when the number of slots increases from 6 to 8 to 10, the brake torque value increases. however, when 12 slots are added to the conductor plate, the value of brake torque produced in 12 slots is less than the value in 10 slots. the chart tends to show a downward trend after crossing the 450-rpm mark. this occurs as a result of the skin effect (waloyo et al., 2020). skin effect is an alternating electric current that appears in the disc conductor so that the current density is formed on the surface of the conductor and will decrease as it gets deeper in the conductor (taghizadeh kakhki et al., 2016). with the skin effect on the disc conductor, it will greatly affect the braking torque at high speeds (waloyo et al., 2020). the addition of a semicircular type conductor disc slot on the ecb braking system has only a small impact on the braking performance of the ecb. figure 4. graph of the relationship between braking torque and rotational speed for each variation calculation of braking is done by braking force when the vehicle slows down from a certain speed. the calculation is only to prove the characteristics of the modeling process used, namely regarding the effect of adding a conductor disk slot. at the next stage the braking process can be analyzed with several additional parameters such as rolling resistance, etc. 6 issn 2180-1053 vol. 14 no. 1 july – december 2022 3.3 comparation of magnetic flux on variation of slots the spread of magnetic flux in each slot variation is shown in figure 5 below. (a) (b) (c) (d) figure 5. distribution of magnetic flux in variations of (a) 6 slots, (b) 8 slots, (c) 10 slots, and (d) 12 slots. the magnetic flux distribution in figure 5 in each variation has a magnetic flux distribution that is almost the same or uniform. this makes the addition of the number of half-circle type slots have a less significant effect. the value of the torque produced will also have almost the same value. in addition, the distribution of the magnetic flux is all concentrated in the conductors and cores of the ecb. this phenomenon can be seen in the image in red which indicates that the magnetic field strength is highest in the area closer to the core that carries the electric current. 3.4 comparation of magnetic flux on rotating speed the spread of magnetic flux at each rotational speed variation is shown in figure 6 below. (a) (b) 7 issn 2180-1053 vol. 14 no. 1 july – december 2022 (c) (d) (e) figure 6. spread of magnetic flux at variations in rotational speed (a) 150 rpm, (b) 300 rpm, (c) 450 rpm, (d) 600 rpm, and (e) 750 rpm. in figure 6 it can be seen that the higher the rotational speed, the greater the spread of the magnetic flux on the disc conductor. in addition, the magnetic field at low speeds is more concentrated to the magnetic source (core) and will spread as the speed increases. the magnetic field strength (red color) will disappear with increasing speed. the phenomenon that occurs is related to the skin effect. the skin effect has a significant impact on braking torque and disc rotation speed (waloyo et al., 2020). skin effects appear at high rotational speeds which will cause a decrease in the value of the resulting torque. at low speeds, it is seen that the spread of magnetic flux is still around the center of the magnetic field, while at high speeds the distribution of the magnetic field will be further away from the center of the magnetic field and the resulting area will be larger. 4.0 conclusion from the research that has been done, it can be concluded that the addition of the number of half-circle type slots with slot variations of 6, 8, 10, and 12 slots does not have a significant effect on the torque generated from the simulation. the trend of the graph of adding slots to the conductor has almost the same trend. the highest braking torque value is produced by a variation of 10 slots with a torque value of 15.93 at a speed of 450 rpm. the spread of magnetic flux in the variation of the number of slots has almost the same area shape. it is concluded that the distribution of magnetic flux with variations in the number of slots does not have much effect on ecb braking. in variations in rotational speed, the spread of magnetic flux from low to high-speed changes the shape of the area. the higher the rotational speed, the wider the magnetic flux distribution because the magnetic field strength will spread and be influenced by the skin effect that appears at high speeds. 8 issn 2180-1053 vol. 14 no. 1 july – december 2022 5.0 acknowledgment thanks to uns for providing funding and laboratory facilities for ecb research. thank you to ice-seam 2021 for accepting and presenting the results of this research work. 6.0 references cho, s., liu, h. c., ahn, h., lee, j., & lee, h. w. (2017). eddy current brake with a two-layer structure: calculation and characterization of braking performance. ieee transactions on magnetics, 53(11). https://doi.org/10.1109/tmag.2017.2707555 cho, s., liu, h. c., lee, j., lee, c. m., go, s. c., ham, s. h., woo, j. h., & lee, h. w. (2017). design and analysis of the eddy current brake with the winding change. journal of magnetics, 22(1), 23–28. https://doi.org/10.4283/jmag.2017.22.1.023 gerdes, j. c., & hedrick, j. k. (1999). brake system modeling for simulation and control. journal of dynamic systems, measurement and control, transactions of the asme, 121(3), 296–503. https://doi.org/10.1115/1.2802501 günay, m., korkmaz, m. e., & özmen, r. (2020). an investigation on braking systems used in railway vehicles. engineering science and technology, an international journal, 23(2), 421–431. https://doi.org/10.1016/j.jestch.2020.01.009 karakoc, k., suleman, a., & park, e. j. (2016). analytical modeling of eddy current brakes with the application of time varying magnetic fields. applied mathematical modelling, 40(2), 1168–1179. https://doi.org/10.1016/j.apm.2015.07.006 kou, b., jin, y., zhang, h., zhang, l., & zhang, h. (2014). modeling and analysis of force characteristics for hybrid excitation linear eddy current brake. ieee transactions on magnetics, 50(11). https://doi.org/10.1109/tmag.2014.2323334 lequesne, b. (1997). eddy-current machines with permanent magnets and solid rotors. ieee transactions on industry applications, 33(5), 1289–1294. https://doi.org/10.1109/28.633808 prayoga, a. r., ubaidillah, nizam, m., & waloyo, h. t. (2019). the influence of aluminum conductor shape modification on eddy-current brake using finite element method. icevt 2019 proceeding: 6th international conference on electric vehicular technology 2019, 146–150. https://doi.org/10.1109/icevt48285.2019.8994005 putra, m. r.a., nizam, m., tjahjana, d. d. d. p., & waloyo, h. t. (2019). the effect of air gap on braking performance of eddy current brakes on electric vehicle braking system. icevt 2019 proceeding: 6th international conference on electric vehicular technology 2019, 355–358. https://doi.org/10.1109/icevt48285.2019.8993987 9 issn 2180-1053 vol. 14 no. 1 july – december 2022 putra, mufti reza aulia, nizam, m., tjahjana, d. d. d. p., aziz, m., & prabowo, a. r. (2020). application of multiple unipolar axial eddy current brakes for lightweight electric vehicle braking. applied sciences (switzerland), 10(13). https://doi.org/10.3390/app10134659 razavi, h. k., & lampérth, m. u. (2006). eddy-current coupling with slotted conductor disk. ieee transactions on magnetics, 42(3), 405–410. https://doi.org/10.1109/tmag.2005.862762 robert, r. s. (2017). 2d model of axial-flux eddy current brakes with slotted conductive disk rotor. 2017 international siberian conference on control and communications, sibcon 2017 proceedings, 0–5. https://doi.org/10.1109/sibcon.2017.7998501 rodrigues, o., taskar, o., sawardekar, s., clemente, h., & dalvi, g. (2016). design & fabrication of eddy current braking system. international research journal of engineering and technology, 03(04), 809–815. satya, p. s., chandra, p. g. s., & raghavendra, m. b. (2021). a study about eddy current brakes. ijtiir, 50–55. sinmaz, a., gulbahce, m. o., & kocabas, d. a. (2016). design and finite element analysis of a radial-flux salient-pole eddy current brake. eleco 2015 9th international conference on electrical and electronics engineering, 1, 590–594. https://doi.org/10.1109/eleco.2015.7394612 taghizadeh kakhki, m., cros, j., & viarouge, p. (2016). new approach for accurate prediction of eddy current losses in laminated material in the presence of skin effect with 2-d fea. ieee transactions on magnetics, 52(3). https://doi.org/10.1109/tmag.2015.2481924 ubaidillah, u., suwolo, s., prayoga, a. r., nizam, m., & waloyo, h. t. (2020). magnetic flux distribution and braking torque of a grooved eddy current brake. 2020 2nd international conference on computer and information sciences, iccis 2020. https://doi.org/10.1109/iccis49240.2020.9257655 waloyo, h. t., ubaidillah, u., tjahjana, d. d. d. p., nizam, m., & aziz, m. (2020). a novel approach on the unipolar axial type eddy current brake model considering the skin effect. energies, 13(7), 1–15. https://doi.org/10.3390/en13071561 10 preparation of papers in a two column model paper format issn: 2180-1053 vol. 8 no.1 january – june 2016 31 analysis of a house with mud layer in roof for summers r. k. pal 1* 1 department of mechanical engineering panjab university ssg regional centre, hoshiarpur (pb.) abstract the brick & cement-concrete houses are not comfortable to live in the extreme weather conditions in many parts of india. so an enormous amount of energy is needed for heating and cooling. here an effort is made to discover the outcome of using mud layer in the roof of a house on indoor room air temperature in summer season. it was found that a lesser room air temperature existed in case of a brick & cement-concrete house with mud layer in the roof as compared to a brick & cement-concrete house without a layer of mud in the roof in may to september. a maximum temperature difference of 1.04°c, 0.97 °c, 0.78°c, 0.70°c and 0.72°c is achievable in a brick & cement-concrete house with mud layer in the roof as compared to a brick & cement-concrete house without a layer of mud in the roof in may to september respectively. energy and money are also saved by using houses with a layer of mud in the roof. energy savings are of the order of 506 units of electricity and money savings have a value of rs. 2528 in summer seasons from may to september for houses with layer of mud in the roof as compared to houses without layer of mud in the roof. therefore the houses with a layer of mud in the roof are slightly more suitable for living in comparison to houses without layer of mud in the house in summer season in addition to savings in terms of energy and money. keywords: mud layer house; room air temperature; solar irradiation; energy savings. 1.0 introduction the brick and cement-concrete houses are not comfortable to live in the extreme weather conditions in many parts of india. the summers in northern parts of india are very hot and winters are very cold. therefore the indoor air temperature rises above comfort level in summers and falls below the comfort level in winters. the indoor air needs to be conditioned in order to make it comfortable for living. so a huge amount of energy is needed for heating and cooling (pal, 2012 & harvey, 2009). rapid urbanization is another factor which is increasing the energy expenditure. urban population is increasing at a fast rate in india (wbcsd, 2009 & pal, 2015). the overall population in india is also increasing (pal, 2015). these two factors are too causing the consumption of energy to increase. the use of energy efficient buildings can reduce this energy expenditure. these energy efficient buildings are also environment friendly and help in reducing pollution (jadhav, 2007). the amount of cooling load depends on the cooling degree days. *corresponding author e-mail: ravinder_75@yahoo.com journal of mechanical engineering and technology 32 issn: 2180-1053 vol. 8 no.1 january – june 2016 in northern india the cooling degree days are usually highest for the month of may and reduce gradually to the month of september. cooling load can be reduced with a large mass in the walls and the roof and this can save energy expenditure by reducing the indoor temperature variation (eben, 1990). the roof is main source of cooling or heating load. therefore the amount of energy expenditure can be reduced by using a layer of insulating material in the roof. thermal insulation of the buildings can also be done by applying appropriate methods of insulation along with insulating building materials (naseer, 2013). due to the insulating properties of the mud, the overall heat transfer coefficient for mud walls is lower as compared to red brick walls (“handbook of functional requirements of buildings”, 1987, p. 37). therefore a layer of mud in the roof of a brick and cement-concrete house can help to reduce the cooling load. also rammed soil naturally controls the relative humidity of inside air in house and thereby improves air quality (uthaipattrakul, 2004). these methods can be used for reducing the heating and cooling necessity as it will keep the indoor air temperature within comfort level. the effect of mud roof needs to be evaluated for savings in energy consumption for creating comfort conditions. the present effort is to find out the effect of mud layer in roof of a house on the thermal performance of the house in summer season. parameters like thermal conductivity of building materials like plaster, brick, cement concrete mixture and mud & solar irradiation etc. were taken from the available literature. the parameters like indoor air temperature, indoor relative humidity, outdoor air temperature, outdoor relative humidity were either computed or noted down. 2.0 materials and methods the types of houses considered for the study are a brick and cement-concrete house with layer of mud in roof (figure 1) and a brick & cement-concrete house (figure 2). figure 1. house without mud layer in roof analysis of a house with mud layer in roof for summers issn: 2180-1053 vol. 8 no.1 january – june 2016 33 figure 2. house with mud layer in roof the exterior area exposed to the solar radiation is equal for both the houses. however lesser amount of heat will be transferred to the indoor air in case of mud layer house. the thermal resistance for brick cement-concrete and the house with mud layer is 0.010 °c/w and 0.065 °c/w respectively. the daily average temperature and solar irradiation was taken from the available literature and the room air temperature was computed. following formulae (equations 1-9) were used for calculations: the sol-air temperature, tsa (chel & tiwari, 2009) is given as; oo t hsa h r h i tt   (1) the total heat gain by house, qgain is given as; windowwallroofgain qqqq  (2) the total heat loss from the house, qloss is given as; ventiloss qq  (3) the heat gain from the roof, qroof is given as; )(** rasrroofroofroof ttauq  (4) the heat gain from the walls, qwall is given as; )(** raswallwallwallwall ttauq  (5) the heat gain from the windows, qwindow (chel & tiwari, 2009) is given as; )(**** raswindowwindowwindowtwindowwindow ttauiaq   (6) the heat loss due to ventilation, qventi (chel & tiwari, 2009) is given as; journal of mechanical engineering and technology 34 issn: 2180-1053 vol. 8 no.1 january – june 2016 3600 )( arararaa venti ttncv q    (7) heat balance for the house, (chel & tiwari, 2009) is given as;  )(** rasrroofroof ra rara ttau dt dt cm  twindowraswallwallwall iattau **)(**  3600 )( )(** arararaa raswindowwindowwindow ttncv ttau    (8) energy saving potential of the house with mud layer in roof, es (pal, 2015) is given as; 3600 )(* = wmlmlrara s ttcm e  (9) 3.0 results and discussions the room air temperature for brick & cement-concrete houses with and without mud layer is compared in the figure 3 to figure 7 for a very hot day in may to september respectively. room air temperature was lower in case of house with a layer of mud in the roof as compared to a house without mud layer in the roof in the month of may to september. the lesser value of indoor room air temperature in a house with a layer of mud in the roof is because the mud has more heat capacity and lower thermal conductivity as compared to a house without layer of mud in the roof. the room air temperature is lower in case a house with mud layer as compared to a house without a mud layer throughout the day and night due to the reason explained. the maximum value of difference in room air temperature (figure 3 to figure 7) in a house with a mud layer in roof is 1.04°c, 0.97 °c, 0.78°c, 0.70°c, 0.72°c as compared to a house without mud layer in roof in may to september respectively. the difference in temperature for the two houses decreases from may to august due to fall in the outside temperature or in other words fall in cooling degree days . but the difference in temperature for two types of houses increases slightly in september as compared to that in august due to slight increase in outside temperature in september. energy and money are saved by using houses with a layer of mud in the roof. energy savings (figure 8) are of the order of 119, 112, 99, 87, 89 units of electricity in the month of may to september respectively. a total of 506 units of electricity can be saved for the summer season per house. money savings (figure 9) have a value of rs. 594, 561, 494, 435, 443 in may to september respectively for a house with layer of mud in the roof as compared to a house without layer of mud in the roof. a total of rs. 2528 can be saved for the whole summer in case of a house with mud layer in the roof. analysis of a house with mud layer in roof for summers issn: 2180-1053 vol. 8 no.1 january – june 2016 35 figure 3. room air temperature variation inside house in may figure 4. room air temperature variation inside house in june journal of mechanical engineering and technology 36 issn: 2180-1053 vol. 8 no.1 january – june 2016 figure 5. room air temperature variation inside house in july figure 6. room air temperature variation inside house in august analysis of a house with mud layer in roof for summers issn: 2180-1053 vol. 8 no.1 january – june 2016 37 figure 7. room air temperature variation inside house in september figure 8. energy savings for the mud layer house in summers journal of mechanical engineering and technology 38 issn: 2180-1053 vol. 8 no.1 january – june 2016 figure 9. money savings for the mud layer house in summers 4.0 conclusions lower value of room air temperature exists in case of a brick and cement-concrete house with a layer of mud in the roof as compared to a brick & cement-concrete house without a layer of mud in the roof in may to september. a maximum difference of 1.04°c, 0.97 °c, 0.78°c, 0.70°c and 0.72°c in temperature exists in a house with a layer of mud in roof as compared to a house without a layer of mud in the roof. energy and money are saved by using houses with a layer of mud in the roof. energy savings are of the order of 119, 112, 99, 87, 89 units of electricity in the month of may to september respectively. money savings have a value of rs. 594, 561, 494, 435, 443 in may to september respectively. total energy savings are of the order of 506 units of electricity and money savings have a value of rs. 2528 for summer season from may to september for houses with layer of mud in the roof as compared to houses without layer of mud in the roof. therefore the houses with a layer of mud in the roof are slightly more suitable for living in comparison to houses without layer of mud in the house in summers in addition to savings in terms of energy and money. nomenclature tsa = sol-air temperature (°c). tsr, tswall, tswindow =sol-air temperature for roof, wall and window respectively (°c). th = current outside dry bulb temperature (°c). tra = current room air temperature (°c). tml, twml = current room air temperature for house with and without mud layer respectively (°c). α = surface absorptance for solar radiation. it = total incident solar load (w/m 2 ). analysis of a house with mud layer in roof for summers issn: 2180-1053 vol. 8 no.1 january – june 2016 39 δr = difference of longwave radiation incident on the surface from the sky and surroundings and the radiation emitted by a black body at outdoor air temperature (w/m 2 ). ho, hi = film co-efficient over the building and for indoor air respectively (w/m 2 -k). εδr/ho = longwave radiation factor. uroof,uwall, uwindow = overall heat transfer coefficient for roof, walls and window respectively (w/m 2 -k). aroof, awall,awindow = area of the roof, walls and window respectively (m 2 ). k1,k2,k3, k4 = thermal conductivity of cement, cement-concrete, brick and mud layer (w/m-k). δ1,δ2,δ3, δ4 = thickness of cement, concrete, brick and mud layer respectively (m). qroof,qwall,qwindow = heat gain through roof, walls and windows respectively (kj/s). qventi = heat loss due to ventilation (kj/s). qgain, qloss = total heat gain and heat loss (kj/s). ρa, vra, cra, mra = density (kg/m 3 ), volume (m 3 ) and specific heat (kj/kg-k) and mass (kg) of room air respectively. es = energy saving potential of the house with mud layer in roof. references handbook of functional requirements of buildings (other than industrial building). sp: 41(s&t) (1987). new delhi: bureau of indian standard. p. 37. chel arvind & tiwari, g.n. (2009). performance evaluation and life cycle cost analysis of earth to air heat exchanger integrated with adobe building for new delhi composite climate”. energy and buildings, 41, 56–66. eben, s.m.a. (1990). adobe as a thermal regulating material. solar wind technology, 7, 407–416. harvey, f. (2009, april). efforts increase to improve sustainability. energy efficient buildings, financial times. 1-3. jadhav, r. (2007). green architecture in india: combining modern technology with traditional methods. un chronicle, 154 (2), 66-71. naseer, m. a. (2013). energy efficient building design: revisiting traditional architecture. the asian conference on sustainability, energy & the environment, official conference proceedings, osaka, japan, 2013 (pp. 470-482). pal, r. k. (2012). analysis of geothermal cooling system for buildings. international journal of engineering sciences & research technology, 1(10), 569-572. journal of mechanical engineering and technology 40 issn: 2180-1053 vol. 8 no.1 january – june 2016 pal, r. k. (2015). thermal performance of mud houses. research journal of engineering and technology, 6(4), 439-442. uthaipattrakul, dh. (2004). mud-house construction technique. building the house with mud. suan-ngarn-mena press, bangkok, 27-50. wbcsd (2009). energy efficiency in buildings: transforming the market. a report by world business council for sustainable development. issn: 2180-1053 e-issn: 2289-8123 vol.14 no.1 15 investigation on close-loop water-cooled photovoltaic module: effect of water volume on the temperature profile and performance enhancement of the module c. puganesa1* 1 department of mechanical engineering, college of engineering, universiti malaysia pahang, 26300 gambang, pahang, malaysia *corresponding’s author email: puganesachandran@gmail.com article history: received 3 july 2021; revised 18 october 2021; accepted 10 january 2022 abstract: the volume of water in the water reservoir is one of the most important parameters involved in closed-loop water-cooled photovoltaic (pv) systems. however, there are no studies reported on this parameter. therefore, in this paper, pv modules with different effective surface areas are cooled using different sets of water volumes in a closed-loop system to study the temperature reduction and performance enhancement of water-cooled pv modules. the experimental result indicated that for the 250w water-cooled module, different sets of water volumes have a significant effect on the surface temperature reduction of pv module and its performance compared to the 10-w and 30-w water-cooled modules. hence, the preliminary data indicates that the volume of water affects the inlet temperature of the water, which subsequently affects the temperature reduction and performance enhancement of the water-cooled module in the closed-loop system. besides, the inlet temperature of the water is also affected by the effective surface area of the water-cooled module. keywords: water-cooled photovoltaic module, temperature reduction, performance 1.0 introduction renewable energy has been holding its ground as the best alternative to non-renewable energy sources such as natural gas, coal, and fossil fuels, which make up the non-renewable energy group. in this context, solar radiation from the sun plays a strong role in clean energy compared to non-renewable sources. this elemental source of energy has made it viable for the discovery of photovoltaic modules that convert solar energy to electrical energy. in addition to that, pv technology has increased significantly, with an installed capacity of 97 gw in 2019 alone [1]. albeit pv technology has grown by leaps and bounds over the decades, the improvement of efficiency of the pv modules is still an expanding aspect. this is due to the fact that one of the major drawbacks of this pv module is its low conversion efficiency, which is mainly contributed by the rise in temperature of the pv module that results in overheating [2]. this is because only 20% of the solar radiation reaching the panel is converted into electrical energy; the remaining 80% is converted into waste heat energy by the pv module, which causes a rise in cell temperature with respect to solar radiation intensity [3]. the power output from the pv module is a function of temperature, in which the current across the solar cells increases slightly with increasing temperature, but the voltage across the solar cells decreases significantly. consequently, the conversion efficiency decreases [4]. journal of mechanical engineering and technology (jmet) 16 issn: 2180-1053 e-issn: 2289-8123 vol.14 no.1 hence, to overcome this drawback, many researchers have integrated water-cooling systems into closed-loop systems to increase the performance of pv modules while also minimising water consumption. for example, mah et al. [5] conducted an outdoor experiment to investigate the effect of different water flow rates on the performance of closed-loop water-cooled pv modules under malaysian conditions. it was found that at an optimum flow rate of 6 l/min, the performance of the pv module increased by 15% at peak irradiation. besides, basrawi et al. [6] compared the performance of half-cooled and full-surface water-cooled pv modules in a closedloop system. the temperatures of the half-surface water-cooled and full-surface water-cooled modules were found to be 22.05% and 51.04% lower, respectively, than the uncooled module. reflecting an increase of 6.10% and 13.50% in power output for half-surface and full-surface cooled pv, respectively. on top of that, it was found that the temperature of the water increased by around 6.9 °c at the end of the experiment. yong et al. [7] integrated an automated pv water spraying technique into the closed-loop system. the author found that the efficiency of pv module was enhanced by 22.14% compared to that of the uncooled module. notably, the temperature of the water in the tank increases over time. in addition, hosseini et al. [8] integrated a heat exchanger as a part of the water-cooled pv module in a closed-loop system. it was found that the inlet temperature of the water was lower than the outlet water temperature with the presence of a heat exchanger. as a result, the efficiency increased in the range of 815% compared to that of the uncooled module. similarly, yang et al. [9] integrated a geothermal heat soil exchanger to cool a 0.56 m2 area of pv module in a closed loop system at a lower inlet water temperature as well as to maintain the temperature of the water. it was found that the inlet water temperature in the closed loop system was around 29 °c throughout the cooling process. in other words, the inlet temperature of the water was found to be lower than without a ground soil exchanger, and as a result, the performance of water-cooled pv integrated with a soil heat exchanger was better than without a soil exchanger and an uncooled module. on the other hand, odeh and behnia [10] cooled the front surface of a pv module and compared the performance by using underground water and direct tank water as a water source. when compared to direct tank water, it was discovered that using underground water to cool the module achieves the best performance. although extensive research has been conducted on improving performance in water-cooled pv systems, to the best of the author's knowledge, the effect of water volume in the closed-loop watercooled pv module has been overlooked. the volume of water in the reservoir is an important parameter. this is because as the water cools the front surface module, the temperature of the water will increase significantly if the volume of water is insufficient to compensate for the rise in water temperature. ultimately, the reduction in pv module temperature and the increase in performance of water-cooled pv will be compromised as less heat will be taken away from the module surface at a higher inlet water temperature [11]. therefore, the objective of this study is to study the effect of different sets of water volumes in the closed loop system on the water-cooled pv temperature reduction and performance enhancement with different effective surface areas. 2.0 methodology in this experimental work, three pv modules with different effective areas were tested under indoor conditions with the aid of a solar simulator. four halogen lamps rated at 150 w were used to simulate the solar irradiation for the 30 w pv module, whereas six and 19 halogen lamps were used for the 30 w and 250 w pv modules, respectively. the technical details of the pv modules are presented in table 1 and figure 1. investigation on close-loopwater cooled photovoltaic module: effect of water volume on the temperature profile and performance enhancement of the module issn: 2180-1053 e-issn: 2289-8123 vol.14 no.1 17 table 1: technical specification of pv modules. model solar tif-stf-010p6 venus solar kl-30w-36p mys-60p/b3/cf-250 pmax [w] 10 30 250 voc ([v] 21.42 21.6 38.056 isc [a] 0.66 1.8 9.044 vmpp [v] 17.28 17.28 29.607 impp [a] 0.58 1.73 8.563 no.of cells 6 36 60 effective panel area[m2] 0.068 0.173 1.118 figure 2 shows the experimental setup for a 10 w, 30 w, and 250 w water-cooled pv module. this study only looked at the effect of different water volume sets on pv module surface temperature reduction and performance enhancement in a closed-loop system. a 10w submersible pump was used in this experiment to lift the water from the water reservoir and force out the water through the sprinkler onto the pv module's front surface. subsequently, the water is collected underneath the pv module in the water tank after it flows across the pv module surface. besides, a thin piece of perspex was attached to the side of the module to prevent the water from spraying out of the pv module. figure 1: schematic layout of the experimental setup. journal of mechanical engineering and technology (jmet) 18 issn: 2180-1053 e-issn: 2289-8123 vol.14 no.1 table 2 depicts the testing methodology used in this research work for 10, 30, and 250 w pv modules. the 10 w pv module and the 30 w module were tested under 1000 w/m2; however, due to the insufficient number of halogen lamps in the 250 w pv module to simulate 1000 w/m2, the average solar irradiation falling on the pv module was only 455 w/m2; as a result, the pv module produces power lower than 250 w for cooled modules and uncooled modules. despite the limitation, both cooled and uncooled modules were tested under the same value of solar irradiance; therefore, the results can be compared as both have a fair comparison. furthermore, the flow rates are chosen for 10 w, 30 w, and 250 w are 80, 120, and 400 l/hr, respectively. this is because, at a lower flow rate than the value stated in table 2, the pv module was not fully established with the water layer, especially at the bottom edge. it is worth remarking that the pv module has to be fully covered with water to produce a significant reduction in temperature. table 2:testing condition of the pv pv module power [w] irradiance[w/m2] water flow rate[l/hr] volume of water [litres] 10 1000 80 10,15,20,25 30 1000 120 15,20,25,30 250 455 400 20,40,60,80 figure 3 shows the measurement setup for all of the testing conditions. the measured parameters in this study are load current, load voltage, the surface temperature of the pv module, water temperature in the tank, solar irradiance, and flow rate. the load current and load voltage were measured by connecting the rheostat of 200 w 4r7j, which has a range of resistance between 1ω and 200ω, to the pv modules. the values of voltage and current were recorded using a fluke 317 clamp metre and a lutron cm-9930 clamp meter, as shown in figure 5. the temperature of the pv module and the temperature of the water were measured using k-type thermocouples. the value of temperature was recorded using a lutron btm-4208sd 12-channel temperature data figure 2:experimental setup;(a) 10 w water-cooled module, (b) 30 w water-cooled module, (c) 250 w water-cooled pv, (d) water cooling system. investigation on close-loopwater cooled photovoltaic module: effect of water volume on the temperature profile and performance enhancement of the module issn: 2180-1053 e-issn: 2289-8123 vol.14 no.1 19 logger. all sets of experiments were allowed to run for 30 minutes because, within that time, the pv panel had reached equilibrium. the flow was measured using a yf-s201 flow metre that was connected to an arduino. the flow rate data was sent from the arduino to the laptop. on top of that, the water flow rate was controlled using a water valve. figure 3: measurement setup for all testing conditions the power yield by the uncooled module and the cooled module was calculated using (1). in addition to that (2) was used to calculate the conversion efficiency of the pv module. 𝑃𝑃 = 𝑉𝑉𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙 × 𝐼𝐼𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙 (1) where p is the power output for the pv panel measured in watts [w], v is the measured voltage measured in volts [v], and i is the current measured in ampere [a]. 𝜂𝜂 = 𝑃𝑃 𝐺𝐺 × 𝐴𝐴 × 100% (2) where p is the power output calculated in watts [w], v is the instantaneous measured voltage [v], i is the instantaneous measured current [a], η is the module efficiency [%], a is the effective module area [m2], and g is the irradiance value [w/m2]. 3.0 results and discussion 3.1.1 relationship between different sets of water volume and the temperature of water figure 4 shows the temperature profile for different sets of water volumes used to cool the 10 w, 30 w, and 250 w modules. the colours differentiate the temperature of water for different sets of water volumes. the dotted line, dashed line, and normal line represent the temperatures of the water volumes used to cool the 10 w, 30 w, and 250 w modules, respectively. the initial water temperature for all sets of water volumes was around 28°c–28.2°c. it can be seen that, for 10 w and 30 w water-cooled pv, there was no significant rise in water temperature from the initial temperature for all sets of volumes. journal of mechanical engineering and technology (jmet) 20 issn: 2180-1053 e-issn: 2289-8123 vol.14 no.1 figure 4: temperature profile for different sets of water volume however, when the volume of water is increased, the change in water temperature takes a longer time due to the inverse relationship between the volume of water and the rise in water temperature. besides, for a 250 w water-cooled module, the rise in water temperature from the initial water temperature was found to be significant for all sets of volumes, where the increment in water temperature decreased significantly from 36% to 25% when the volume of water in the tank was increased from 20 litres to 80 litres. hence, the trend of the graph indicates that the increase in water temperature is influenced by the volume of water in the reservoir and the effective surface area of water-cooled pv. in other words, the rate of change in water temperature decreases with increasing volume; nevertheless, the rate of change in water temperature increases when the effective surface area of water-cooled pv is greater. 3.1.2 effect of different sets of water volume on the pv modules surface temperature reduction the effect of different water volumes on the average temperature of the water and the average surface temperature of modules in a closed-loop system is shown in figure 5. it can be seen that the uncooled module has the highest temperature in all cases compared to the water-cooled module, where the average temperature of the cooled module at different sets of water volume is suppressed significantly in the range of 53.54-56.81%, 54.46-52.85%, and 49.60-53.47% compared to the uncooled module of 10 w, 30 w, and 250 w, respectively. besides, the decrement in pv module temperature is highly influenced by the inlet water temperature. for 10 w and 30 w water-cooled pv modules, the increase in water temperature for all sets of water volume from the initial temperature was found to be insignificant due to the smaller surface area of the module, where the heating of water takes a very long period of time, hence the cooling effect of water on the module surface is not compromised. as a result, there were no significant changes in the surface temperature of the pv module when it was cooled using different sets of water volumes. in the case of 250 w, the average increment in temperature of the water has dropped significantly from 25.23%to 11.82% when 80 litres of water are being used to cool the surface of the module compared to 20 litres of water. as a result, the module's average surface temperature was the lowest, at 31.31oc, when compared to other sets of water volumes. in other words, the interpretation of this data indicates that as the temperature of the water increases, the cooling effect of the water on the module decreases as less heat is being absorbed by the water from the module surface. besides, this data also indicates that there is no need for a heat exchanger to cool 25 30 35 40 0 3 6 9 12 15 18 21 24 27 30 t em pe ra tu re [ o c ] time [min] water temperature for 10 litres (10 w cooled module) water temperature for 15 litres (10 w cooled module) water temperature for 20 litres (10 w cooled module) water temperature for 25 litres (10 w cooled module) water temperature for 15 litres (30 w cooled module) water temperature for 20 litres (30 w cooled module) water temperature for 25 litres (30 w cooled module) water temperature for 30 litres (30 w cooled module) water temperature for 20 litres (250 w cooled module) water temperature for 40 litres (250 w cooled module) water temperature for 60 litres (250 w cooled module) water temperature for 80 litres (250 w cooled module) investigation on close-loopwater cooled photovoltaic module: effect of water volume on the temperature profile and performance enhancement of the module issn: 2180-1053 e-issn: 2289-8123 vol.14 no.1 21 the inlet water temperature, but instead a sufficient amount of water according to the effective surface area of water-cooled pv could still maintain the temperature of the water around 29oc as reported in [9], where the cooling effect of the water on the pv module will not be compromised significantly. it should be noted that, under outdoor conditions, there should be enough water in the reservoir based on the effective surface area of a water-cooled pv module to compensate for the rise in water temperature and achieve a better cooling effect. figure 5: effects of different sets of water volume on the average temperature of water and average temperature of cooled pv module. 3.1.3 effect of different sets of water volume on the performance of the pv modules figure 6 shows the effect of different water volumes on the average performance of the uncooled and water-cooled modules. the performance of a pv module is a function of its surface temperature. it can be seen that the uncooled module in all cases has the lowest power output and efficiency yield compared to the water-cooled module. for 10 w, 30 w, and 250 w, the average increase in power output is in the range of 27.42–27.68%, 28.27–29.05%, and 33.39–35.32%, respectively, in comparison to an uncooled module. as a result, the electrical efficiency of cooled modules increased in the range of 27.42–27.08%, 28.26–29.03%, and 33.83–35.85% for 10 w, 30 w, and 250 w, respectively. besides, it can be seen that the increase in performance of the 10 w and 30 w water-cooled modules was found to be insignificant when compared relative to other sets of water volume, as there were no significant changes in the surface temperature of the watercooled modules. on the other hand, the effect of water volume on the water-cooled 250 w module temperature reduction was found to be significant, where the power gain by the water-cooled pv module increased by around 1.52% when it was cooled using 80 litres of water in a closed loop system compared to that of 20 litres of water. in short, the increasing trend of the performance of water-cooled pv shows that the performance of a 250-watt water-cooled system can be further increased if the volume of water is increased by more than 80 litres, as the rise in water temperature could be further suppressed at a greater volume. ultimately, a better cooling effect could be achieved. 28 .0 5 27 .9 27 .7 8 27 .6 1 31 .3 2 31 .1 8 31 .1 1 30 .9 9 35 .0 8 34 .3 9 32 .5 7 31 .3 17 0. 99 31 .5 6 31 .5 2 30 .9 30 .6 6 64 .6 3 30 .7 2 30 .6 5 30 .5 9 30 .4 7 7 2. 57 36 .5 7 35 .8 7 34 .4 8 33 .7 7 0 10 20 30 40 50 60 70 80 u nc oo le d m od ul e co ol ed m od ul e( 10 lit re s) co ol ed m od ul e (1 5 lit re s) co ol ed m od ul e (2 0 lit re s) co ol ed m od ul e (2 5 lit re s) u nc oo le d m od ul e co ol ed m od ul e (1 5 lit re s) co ol ed m od ul e (2 0 lit re s) co ol ed m od ul e (2 5 lit re s) co ol ed m od ul e (3 0 lit re s) u nc oo le d m od ul e co ol ed m od ul e (2 0 lit re s) co ol ed m od ul e (4 0 lit re s) co ol ed m od ul e (6 0 lit re s) co ol ed m od ul e (8 0 lit re s) 10 w 30 w 250 w te m pe ra tu re [℃ ] average temperature water [℃] average temperature of pv module [℃] journal of mechanical engineering and technology (jmet) 22 issn: 2180-1053 e-issn: 2289-8123 vol.14 no.1 figure 6: effects of different water volume on the average performance of water-cooled pv module. 4.0 conclusion in the present experimental study, photovoltaic modules having different effective surface area was cooled using different sets of water volume in close-loop system to study the temperature reduction and performance enhancement of water-cooled pv module. the highlights of this study are summarized as follows: • the increase in water temperature is influenced by the volume of water in the water reservoir and the effective surface area of water-cooled pv. • different sets of water volume have significant effect on the surface temperature reduction of 250 w water cooled pv module and its performance compared to the 10-w and 30-w watercooled modules. • the average temperature of cooled module at different sets of water volume is suppressed significantly in range of 53.54-56.81%, 54.46-52.85 % and 49.60-53.47 % compared to uncooled module of 10 w ,30 w and 250 w respectively. • the increase in performance of 10 w and 30 w water-cooled module was found to insignificant when it is compared relative to other sets of water volume as there were no significant changes in the surface temperature of water-cooled module due to the smaller surface area of the module, in such the heating of water takes a very long duration in water reservoir, hence the cooling effect of water for all sets of volume on the pv module surface is not compromised. • the effect of water volume on the water cooled 250 w module temperature reduction were found to be significant as a result the power gain by the water-cooled pv module increased around 1.52 % when it is cooled using 80 litres of water in close loop system compared to that of 20 litres of water. 7. 66 9. 76 9. 78 9. 78 9 .7 9 20 .3 8 26 .1 4 26 .2 2 26 .2 6 26 .3 44 .4 2 5 9. 21 59 .3 1 59 .3 7 60 .1 1 11 .2 7 14 .3 6 14 .3 8 14 .3 9 14 .3 9 11 .7 8 15 .1 1 15 .1 6 15 .1 8 15 .2 3. 96 5 .3 5. 3 5. 31 5. 38 0 2 4 6 8 10 12 14 16 5 15 25 35 45 55 65 u nc oo le d m od ul e co ol ed m od ul e (1 0l itr es ) co ol ed m od ul e (1 5 lit re s) co ol ed m od ul e (2 0 lit re s) co ol ed m od ul e (2 5 lit re s) u nc oo le d m od ul e co ol ed m od ul e (1 5 lit re s) co ol ed m od ul e (2 0 lit re s) co ol ed m od ul e (2 5 lit re s) co ol ed m od ul e (3 0 lit re s) u nc oo le d m od ul e co ol ed m od ul e (2 0 lit re s) co ol ed m od ul e (4 0 lit re s) co ol ed m od ul e (6 0 lit re s) co ol ed m od ul e (8 0 lit re s) 10 w 30 w 250 w ef fic ie nc y [% ] po w er [w ] average power [w] average efficiency [%] investigation on close-loopwater cooled photovoltaic module: effect of water volume on the temperature profile and performance enhancement of the module issn: 2180-1053 e-issn: 2289-8123 vol.14 no.1 23 • greater the effective surface area of water-cooled module in close loop system, greater the volume of water need in tank is required to suppresses the rise in water temperature to maintain the cooling effect. this suggest there is a unique relationship between volume of water in tank and effective surface area of pv module. thus, further investigation is needed on this relation. acknowledgements the financial support offered by the universiti malaysia pahang under rdu 180313 are gratefully acknowledged. references [1] i.r.e agency. (2020). renewable power generation costs in 2019. available: https://www.irena.org/publications/2020/jun/renewable-power-costs-in-2019 [2] s. dubey, j.n. sarvaiya, b. seshadri. "temperature dependent photovoltaic (pv) efficiency and its effect on pv production in the world a review," energy procedia, vol. 33, pp. 311–321, 2013. [3] j. zhang, h. zhai, z. wu, y. wang, h. xie, m. zhang. "enhanced performance of photovoltaic– thermoelectric coupling devices with thermal interface materials," energy reports, vol. 6, pp. 116– 122, 2020. [4] p.k. dash, n.c. gupta. "effect of temperature on power output from different commercially available photovoltaic modules," j. eng. res. appl, 2015. [5] c.y. mah, b.h. lim, c.w. wong, m.h. tan, k.k. chong, a.c. lai. "investigating the performance improvement of a photovoltaic system in a tropical climate using water cooling method," energy procedia, vol. 159, pp. 78–83, 2019. [6] f. basrawi, y.c. leon, t.k. ibrahim, m.h. yusof, a.a. razak, s.a. sulaiman, t.yamada. "experimental analysis on the effect of area of surface cooling for a water-cooled photovoltaic," in matec web conf., 2018, vol. 225, pp. 1–6. [7] e.e.d. yong, s.m.r. fulge, j.e.r lisaca, d.m. olarte, j.m. rosario. "development of a water-based pv cooling system," in 2018 ieee 10th int. conf. humanoid, nanotechnology, inf. technol. commun. control. environ. manag. hnicem 2018 , 2019, pp. 1–7. [8] r. hosseini, n. hosseini, h. khorasanizadeh. "an experimental study of combining a photovoltaic system with a heating system," presented at proc. world renew. energy congr. – sweden, linköping, sweden 57, 2011. [9] l.h. yang, j.d. liang, c.y. tseng, s.l. chen. "improvements on the efficiency of the photovoltaic panel by integrating a spray cooling system with shallow geothermal energy heat exchanger," in conference: ises eurosun 2018 conference – 12th international conference on solar energy for buildings and industry, 2019, pp. 1–11. [10] s. odeh, m. behnia. "improving photovoltaic module efficiency using water cooling," heat transf. eng., vol. 30, pp. 499–505, 2009. [11] l.w. zhe, m.i. yusoff, a.a razak, m.i. misrun, s. ibrahim, m.i. fahmi, a.s. rosmi. "effect of water cooling temperature on photovoltaic panel performance by using computational fluid dynamics (cfd)," j. adv. res. fluid mech. therm. sci. , vol. 56, pp. 133–146, 2019. c. puganesa1* abstract: the volume of water in the water reservoir is one of the most important parameters involved in closed-loop water-cooled photovoltaic (pv) systems. however, there are no studies reported on this parameter. therefore, in this paper, pv modules... keywords: water-cooled photovoltaic module, temperature reduction, performance references microsoft word 03_jmet-template2020-tire_model_rev3-forjmetonline.docx journal of mechanical engineering and technology _______________________________________ *corresponding author. email: mohdazman@utem.edu.my issn 2180-1053 vol. 13 no. 1 june – december 2021 25 tire model verification and performance comparison using double lane change test mohd azman abdullah1*, mohd hanif harun1, fauzi ahmad1, amrik singh phuman singh1, ahmed esmael mohan2 1 centre for advanced research on energy (care), fakulti kejuruteraan mekanikal (faculty of mechanical engineering), universiti teknikal malaysia melaka, 75450 ayer keroh, melaka, malaysia 2 al-furat al-awsat technical university, technical college al-mussaib, 54003 babylon, iraq abstract three tire models, namely dugoff, calspan, and magic formula are used in this paper. the models are developed based on their equations in matlab/simulink and verified using carsim software through the standard double lane change (dlc) test. the comparison of their performances is carried out on three different vehicles namely the sedan, the sports car, and the sport utility vehicle (suv) and at three different speeds. further analyses are performed on the lateral and longitudinal tire forces performances of the vehicles at different speeds on the dlc test. it can be observed that the dlc test is best carried out at low speed and with a less heavy vehicle. the tire models can be used for future analysis of vehicle lateral and longitudinal dynamics. keywords: double lane change, tire models, vehicle dynamics, tire dynamics, tire model verification. 1.0 introduction the vehicle dynamic can be modelled and simulated for the purpose of study and analysis. vehicle dynamics are concerned with the movements of vehicles, which including acceleration, braking, ride, and cornering. the dynamic behaviour, which is determined by the forces imposed on the vehicle from the tires, gravity, and aerodynamics also can observe through simulation [1-17]. these methods are used in analyzing and studying the dynamics performances of vehicles due to the constraints such as cost, time, and safety of other approaches. the vehicle dynamics system interaction consists of the input by visual (from the driver camera), ground elevations and surface irregularities which act on the tires and aerodynamic loads which act on the body of the vehicle. the outputs evaluation of the vehicle is measured in terms of performance, handling, and ride [1842]. the vehicle and its components are studied to determine what forces will be produced by each of these sources at a particular manoeuvre and how the vehicle will respond to these forces. forces and moments from the road act on each tire of the vehicle and highly affected the vehicle’s dynamic. the tire deforms due to the vertical load on it and makes contact with the roads over a non-zero footprint area, which is called as contact patch. forces and moments from the road act on each tire of the vehicle and highly influence the dynamics of the vehicle. the forces acting on the tire are assumed to be at the centre of the contact patch and these forces can be composed along 3 axes (figure 1), which are: x-axis: longitudinal tire force, fx, y-axis: lateral tire force, fy, and z-axis: normal to journal of mechanical engineering and technology issn 2180-1053 vol. 13 no. 1 june – december 2021 26 vertical tire force, fz. besides that, the moments acting on the tire can be decomposed to 3 axes too, which are: x-axis: overturning moment, mx, y-axis: rolling resistance moment, my, and z-axis: aligning moment, mz. longitudinal tire force at small slip ratio is in the forward direction in case of a driving wheel and proportional to the slip ratio, which means, longitudinal tire force at small slip ratio will increase as the slip ratio is increased. the slip ratio is the difference between theoretically calculated forward speed based on angular speed of the rim and rolling radius, and actual speed of the vehicle, expressed as a percentage of the latter. longitudinal tire force generated by each tire depends on slip ratio, vertical forces on the tire, and friction coefficient of the road surface. the lateral tire force at a small slip angle is proportional to the slip angle. the lateral tire force at a small slip angle is increases as the slip angle is increased. the slip angle of the tire is an angle between the direction of heading and direction of travel of the wheel. in this study, the tire dynamic equations for the sedan car, sports car and suv are modeled using magic, calspan and dugoff tire formulae in matlab/simulink and simulated to study their performance. the models are verified through double lane change (dlc) test using commercial software carsim. figure 1: tire forces and moments in the magic tire model, the lateral tire force derivation requires tire slip angle and the longitudinal tire force derivation requires longitudinal tire slip ratio [43-51]. the lateral tire force, fy is calculated by, journal of mechanical engineering and technology issn 2180-1053 vol. 13 no. 1 june – december 2021 27 �� = � sin� tan� ����� + �� (1) where, � = � ��� + ���� (1) = 1.30 (2) � = � + ∆� (3) � = �1 − ���� + �� � + �� tan� ���� + �� � (4) and, �� = �� !��� + � �� �" (5) the vertical tire force, fz in eqn(2) is the input from the simulation results in the unit of kn. the values of b1 and ∆b1 in eqn(3) are calculated by, � = �#$ � (6) ∆� = −� �|"|� (7) the values of e and sh in eqn(4) are calculated by, � = �& ��� + �' �� + �( (8) �� = �)" (9) the tire slip angle, α in eqn(4) is the input from the simulation results in the unit of degree. the camber angle, γ in eqn(5) is in the input in the unit of degree. the value bcd in eqn(6) is calculated by, �#$ = �* sin��+ tan� ��, �� � (10) the longitudinal tire force, fx is calculated by, �= � sin� tan� ����� (11) where c = 1.65 and d is the same as in eqn(2). the value b and φ are calculated by, � = �#$ � (12) � = �1 − ��. + �� tan� ��.� (13) journal of mechanical engineering and technology issn 2180-1053 vol. 13 no. 1 june – december 2021 28 the value bcd in eqn(12) is calculated by, �#$ = �* ��� + �+��/0123 (14) the longitudinal tire slip ratio, σ in eqn(13) is the input from the simulation results. the value e in eqn(13) is calculated by, � = �& ��� + �' �� + �( (15) the constants a1, a2, a3, a4, a5, a6, a7, a8, a9, a10, a11, a12 and a13 are tabulated in table 1 and 2 [45, 48, 49, 51]. table 1: constants a1 to a8 force a1 a2 a3 a4 a5 a6 a7 a8 fy -22.1 1011 1078 1.82 0.208 0.000 -0.354 0.707 fx -21.3 1144 49.6 226 0.069 -0.006 0.056 0.486 table 2: constants a9 to a13 force a9 a10 a11 a12 a13 fy 0.028 0.000 14.8 0.022 0.000 in calspan tire model, the lateral and longitudinal tire forces are derived from the combination of tire slip conditions [52-60]. the lateral and longitudinal tire forces, fy and fx are calculated by, �� = 4�.� 5 tan �6 5� tan� � + 789� .-� :�� (16) �= 4�.� 789 .-6 5� tan� � + 789� .-� :�� (17) where, 4�.� = �;:�� = .* + � .� + <4>? . .* + *.� + +. + 1 (18) 789 = 78 + @ 5 − 78 absin� � + .-� cos� � (19) : = :e + <1 − fg bsin� � + .-� cos� �? (20) journal of mechanical engineering and technology issn 2180-1053 vol. 13 no. 1 june – december 2021 29 the lateral and longitudinal tire stiffnesses, cα and cσx, in eqn(16) and eqn(19) are calculated by, 5 = 2@�ie a� jke + k �� − k k� ���l (21) 78 = 2@�ie a� �� � � �m⁄ � (22) the values of tire slip angle, α, vertical tire force, fz, and longitudinal slip ratio, σx are the inputs from the simulation results. the combined slip, σ in eqn(18) is calculated by, . = >�i�8:e �� p 5� tan� � + 78� j .-1 − .l � (23) the original contact length, apo, in eqn(21) and eqn(22) is calculated by, �ie = 0.0768b�� �stuv @ui + 5a (24) the value of tire contact patch, ap, in eqn(23) is calculated by, �i = �ie j1 − f0 �-�� l (25) all the constants in eqn(18) and eqn(20) to (25) are tabulated in table 3 [99]. table 3: constant for calspan tire model parameters rwd radial rwd bias ply fwd radial fwd radial tire designation 155sr13 p155/80d13 p185/70 r13 p185/70 r13 thread width, tw 6 6 7.3 7.3 tire pressure, tp 24 24 24 24 fzt 810 900 980 980 c1 1.0 0.535 1.0 1.0 c2 0.34 1.05 0.34 0.34 c3 0.57 1.15 0.57 0.57 c4 0.32 0.8 0.32 0.32 a0 914.02 1817 1068 1068 a1 12.9 7.48 11.3 11.3 a2 2028.24 2455 2442.73 2442.73 a3 1.19 1.857 0.31 0.31 k4 0.05 0.2 0.05 0.05 cs/fz 18.7 15.22 17.91 17.91 μo 0.85 0.85 0.85 0.85 journal of mechanical engineering and technology issn 2180-1053 vol. 13 no. 1 june – december 2021 30 in dugoff tire model, the lateral and longitudinal tire forces are derived from lateral and longitudinal tire stiffness values [61-73]. the lateral and longitudinal tire forces, fy and fx are calculated by, �� = 5 j tan �1 + .l 4��� (26) �= 7 j .-1 + .l 4��� (27) where, � = :�� �1 + .�2@ b� 7 .�� + � 5 tan ���x a (28) 4��� = y�2 − ��� [4 � < 1 1 [4 � ≥ 1 (29) the tire slip angle, α, tire longitudinal slip ratio, σx and tire vertical force, fz are the inputs from simulation results. the constants in eqn(26) to (28) are tabulated in table 4 [61, 62, 64, 65, 66, 67, 70]. table 4: dugoff tire constants parameters value cα -1.56 x 105 cσ 2.37 x 105 μ 0.99 2.0 methodology three different vehicles namely the sedan car, the sports car and the sport utility vehicle (suv) are used in the simulation of dlc (figure 2). the test is carried out at 3 different speeds of 80, 100, and 120 km/h. figure 3 shows the standard dimension of the dlc route [74-88]. the simulation of dlc produces the vertical tire force, fz, longitudinal tire force, fx, lateral tire force, fy, tire slip angle, α, and tire slip ratio, σ. the fz, α and σ are used as the inputs for the tire models. the outputs of the tire models, fx and fy are compared with the fx and fy from the simulation. figures 4 and 5 show the simulink subsystem models of fy and fx developed from eqn(1) and eqn(11) respectively. figure 6 shows the simulink subsystem model of calspan tire model for fy and fx developed from eqn(16) and eqn(17). figure 7 shows the simulink subsystem model of dugoff tire model for fy and fx developed from eqn(26) and eqn(27). the ‘miew’ in figure 7 is the constant µ in table 4. journal of mechanical engineering and technology issn 2180-1053 vol. 13 no. 1 june – december 2021 31 (a) (b) (c) figure 2: dlc simulation in carsim software (a) sedan car, (b) sports car and (c) suv journal of mechanical engineering and technology issn 2180-1053 vol. 13 no. 1 june – december 2021 32 figure 3: standard dlc procedure route (iso 3888 part 2) figure 4: magic tire simulink model to find fy journal of mechanical engineering and technology issn 2180-1053 vol. 13 no. 1 june – december 2021 33 figure 5: magic tire simulink model to find fx figure 6: calspan tire simulink model to find fy and fx journal of mechanical engineering and technology issn 2180-1053 vol. 13 no. 1 june – december 2021 34 figure 7: dugoff tire simulink model to find fy and fx the verification of the models is carried through the root mean square (rms) value of the difference between the tire forces from the model to the tire forces from the simulation. the equation of the rms is shown in eqn(30) with n is the number of data. the lower the value of rms, the better the tire model performance. figure 8 shows the comparison flow of the tire forces. the vertical tire force, fz, tire slip angle, α, and tire slip ratio, σ, from the results of the simulation are used as the inputs for the model. the longitudinal and vertical tire forces from both simulation (xsim,i) and model (xmodel,i) results are compared using rms values [89-98]. ^_� = 6 ̀ ∑ @bcde,d − beeghi,d a�d̀j (30) figure 8: comparison between simulation and model journal of mechanical engineering and technology issn 2180-1053 vol. 13 no. 1 june – december 2021 35 3.0 results and discussion figure 9 to 17 show the longitudinal and lateral tire forces from simulation and models for the sports car, sedan, and suv at speeds of 80, 100, and 120 km/h dlc test. it can be observed that the shape and pattern of the graphs from the models are almost identical to the results from the simulation. at 80 km/h (figure 9), the longitudinal and lateral tire forces from the calspan tire model are about the same as the result from simulation for the sports car. however, only longitudinal tire forces of the calspan tire model are about the same as in the simulation for dlc test at 100 and 120 km/h presented in figure 10 and 11 respectively. the best graph fitting for longitudinal tire forces is from the magic tire model at these speeds. as for the sedan type of car, at all speeds, the best tire model is calspan where all the longitudinal and lateral tire forces are very nearly alike as the simulation as shown in figures 12 to 14. the tire models however perform randomly for suv (figure 15 to 17). at 80 and 100 km/h, longitudinal tire forces from calspan tire models are the best. but at 120 km/h, longitudinal tire force from the magic tire model is the finest. the magic tire model also contributes in producing the unsurpassed lateral tire force for the suv at 80 km/h (figure 15). at 100 and 120 km/h (figure 16 and 17), the best tire model matching the simulation results for lateral tire forces is the dugoff tire model. journal of mechanical engineering and technology issn 2180-1053 vol. 13 no. 1 june – december 2021 36 (a) (b) figure 9: comparison of (a) fx and (b) fy for sports car at 80 km/h journal of mechanical engineering and technology issn 2180-1053 vol. 13 no. 1 june – december 2021 37 (a) (b) figure 10: comparison of (a) fx and (b) fy for sports car at 100 km/h f o rc e , n journal of mechanical engineering and technology issn 2180-1053 vol. 13 no. 1 june – december 2021 38 (a) (b) figure 11: comparison of (a) fx and (b) fy for sports car at 120 km/h journal of mechanical engineering and technology issn 2180-1053 vol. 13 no. 1 june – december 2021 39 (a) (b) figure 12: comparison of (a) fx and (b) fy for sedan car at 80 km/h f o rc e , n journal of mechanical engineering and technology issn 2180-1053 vol. 13 no. 1 june – december 2021 40 (a) (b) figure 13: comparison of (a) fx and (b) fy for sedan car at 100 km/h journal of mechanical engineering and technology issn 2180-1053 vol. 13 no. 1 june – december 2021 41 (a) (b) figure 14: comparison of (a) fx and (b) fy for sedan car at 120 km/h f o rc e , n journal of mechanical engineering and technology issn 2180-1053 vol. 13 no. 1 june – december 2021 42 (a) (b) figure 15: comparison of (a) fx and (b) fy for suv at 80 km/h f o rc e , n journal of mechanical engineering and technology issn 2180-1053 vol. 13 no. 1 june – december 2021 43 (a) (b) figure 16: comparison of (a) fx and (b) fy for suv at 100 km/h f o rc e , n journal of mechanical engineering and technology issn 2180-1053 vol. 13 no. 1 june – december 2021 44 (a) (b) figure 17: comparison of (a) fx and (b) fy for suv at 120 km/h journal of mechanical engineering and technology issn 2180-1053 vol. 13 no. 1 june – december 2021 45 in order to verify the models, the rms values are generated. the lower the value of rms, the better the performance of the models. table 5 tabulated the rms results comparison among the tire forces from the models to the simulation. from the table, at all speeds, using the magic tire model, the best vehicle performance is a sports car for both longitudinal and lateral tire forces. using the calspan tire model, the best vehicle is the sports car for longitudinal tire force and the sedan car for lateral tire force. as for the dugoff tire model, the suv is the best for longitudinal force and the sedan car is the best for lateral force. with average rms of 30.03 and 80.31 for longitudinal and lateral tire forces respectively, the best speed for dlc test is at the lowest speed of 80 km/h. comparing all tire models at all speeds, the sports car and sedan car are good using the calspan tire model for both fx and fy. the suv however is good using the calspan tire model for fx and magic tire model for fy. among all tire models, it is observed that the calspan tire model is the best in mimicking the result from the simulation. table 5: rms values tire model speed (km/h) 80 100 120 rms value vehicle type fx fy fx fy fx fy magic sport car 9.79 46.77 17.11 49.38 45.29 72.37 sedan 27.73 41.80 42.27 41.12 78.99 121.90 suv 33.63 3.21 63.07 428.30 123.80 2163.00 calspan sport car 4.51 21.46 11.04 213.70 20.21 141.00 sedan 11.87 15.71 20.27 36.63 33.23 12.61 suv 15.31 31.19 28.38 905.20 125.60 3640.00 dugoff sport car 59.20 273.60 126.10 856.90 269.50 158.40 sedan 74.41 268.70 136.80 54.40 205.50 534.60 suv 33.84 20.32 47.58 286.40 124.70 1919.00 figures 18 to 20 show the simulation results from carsim software of fx and fy for all vehicles at different speeds. the longitudinal forces are different among all vehicles. journal of mechanical engineering and technology issn 2180-1053 vol. 13 no. 1 june – december 2021 46 (a) (b) figure 18: comparison of (a) fx and (b) fy for all vehicles at 80 km/h f o rc e , n journal of mechanical engineering and technology issn 2180-1053 vol. 13 no. 1 june – december 2021 47 (a) (b) figure 19: comparison of (a) fx and (b) fy fy for all vehicles at 100 km/h journal of mechanical engineering and technology issn 2180-1053 vol. 13 no. 1 june – december 2021 48 (a) (b) figure 20: comparison of (a) fx and (b) fy for all vehicles at 120 km/h f o rc e , n journal of mechanical engineering and technology issn 2180-1053 vol. 13 no. 1 june – december 2021 49 figures 21 to 23 show the comparison of the y-coordinate of all vehicles to the targeted route. at 80 and 100 km/h, the y-coordinates of suv are quite offset compared to the other vehicles. this is merely due to the lateral acceleration and lateral force. since suv is the heaviest vehicle, the lateral force is the highest. at zone a when the suv arrives, due to the momentum, the vehicle tends to go further, however, the counter-steering wheel action returns the suv to the targeted path. the path of the sports car is the best matching the targeted paths. this is proven by the rms values of 0021 and 0.0927 (table 6). at 120 km/h, the best vehicle is the sedan since its rms value is 0.8711. according to the table, the best performance for dlc is at the lowest speed of 80 km/h since its rms is the lowest in average for all vehicles. this is logically right. at low speed, the lateral acceleration is low. therefore, it produces low lateral force and low lateral motion, and eventually, the vehicle can follow the targeted path. in terms of the weight of the vehicle, the lightest vehicle performed the best in dlc test. low weight means low lateral force at easier for the vehicle to counter lateral motion to follow the targeted path. additionally, the lowest cog of the vehicle is the best is following the targeted path. this is due to the roll moment. at low cog, the roll moment is low. therefore, low moment compensation is needed to follow the path. figure 21: y coordinate of all vehicles compared to target at 80 km/h tabel 6: rms values comparison between the vehicle y-coordinate and target speed (km/h) 80 100 120 vehicle weight (kg) cog (m) rms sport car 1360 0.375 0.0021 0.0927 0.8818 sedan 1650 0.530 0.0026 0.0946 0.8711 suv 2257 0.781 0.0507 0.8651 2.0110 a journal of mechanical engineering and technology issn 2180-1053 vol. 13 no. 1 june – december 2021 50 figure 22: y coordinate of all vehicles compared to target at 100 km/h figure 23: y coordinate of all vehicles compared to target at 120 km/h journal of mechanical engineering and technology issn 2180-1053 vol. 13 no. 1 june – december 2021 51 4.0 conclusion the common models in analyzing the tire behaviours in terms of longitudinal and lateral forces have been studied. the simulink models of the magic, calspan, and dugoff tire models have been developed. the models have been verified by the simulation results of three different vehicles at three different speeds through dlc test. it can be concluded that the best dlc test is performed at low speed and the best tire model is the calspan tire model. in term of vehicle, the lower the cog and weight, the better the performance of the vehicle. the simulink models of the tires can be used for further analysis on longitudinal and lateral dynamic control of a vehicle. based on the results and observations, with the same tire size, the performance of different vehicles can be studied and analyzed. 5.0 acknowledgement the authors are gratefully and acknowledge the support received from the centre for advanced research on energy (care), universiti teknikal malaysia melaka, malacca, malaysia, al-furat al-awsat technical university, technical college al-mussaib, babylon, iraq, as well as the financial support provided by the short term research grant, grant no. pjp/2020/fkm/pp/s01784 and fundamental research grant scheme (frgs), grant no.: frgs/1/2020/fkm-care/f00439. 6.0 references [1] benekohal, r. f., & treiterer, j. (1988). carsim: car-following model for simulation of traffic in normal and stop-and-go conditions. transportation research record, 1194, 99-111. [2] dupuy, s., egges, a., legendre, v., & nugues, p. (2001). generating a 3d simulation of a car accident from a written description in natural language: the carsim system. arxiv preprint cs/0105023. [3] kinjawadekar, t., dixit, n., heydinger, g. j., guenther, d. a., & salaani, m. k. (2009). vehicle dynamics modeling and validation of the 2003 ford expedition with esc using carsim (no. 2009-01-0452). sae technical paper. [4] åkerberg, o., svensson, h., schulz, b., & nugues, p. (2003). carsim: an automatic 3d text-to-scene conversion system applied to road accident reports. in demonstrations. [5] johansson, r., williams, d., berglund, a., & nugues, p. (2004, july). carsim: a system to visualize written road accident reports as animated 3d scenes. in proceedings of the 2nd workshop on text meaning and interpretation (pp. 57-64). [6] li, y., deng, h., xu, x., & wang, w. (2018). modelling and testing of in-wheel motor drive intelligent electric vehicles based on co-simulation with carsim/simulink. iet intelligent transport systems, 13(1), 115-123. [7] kinjawadekar, t. (2009). model-based design of an electronic stability control system for passenger cars using carsim and matlab-simulink (doctoral dissertation, the ohio state university). journal of mechanical engineering and technology issn 2180-1053 vol. 13 no. 1 june – december 2021 52 [8] etienne, l., lúa, c. a., di gennaro, s., & barbot, j. p. (2020). a super-twisting controller for active control of ground vehicles with lateral tire-road friction estimation and carsim validation. international journal of control, automation and systems, 18(5), 1177-1189. [9] konghui, g., hao, f., haitao, d., dang, l., & lingge, j. (2008). development of controller for vehicle stability system based on carsim rt [j]. automobile technology, 3. [10] khalili, e., ghaisari, j., & danesh, m. (2017, november). control and analysis of the vehicle motion using sliding mode controller and carsim software. in 2017 5th international conference on control, instrumentation, and automation (iccia) (pp. 15). ieee. [11] zhu, m. t., shao, c. z., & wang, g. l. (2010). research on road model rebuilding and vehicle simulation of ride comfort based on carsim software [j]. machinery design & manufacture, 10, 78-80. [12] egges, a., nijholt, a., & nugues, p. (2001). generating a 3d simulation of a car accident from a formal description: the carsim system. [13] zhao, s., & zhu, l. (2018). cruise control system based on joint simulation of carsim and simulink. open access library journal, 5(7), 1-8. [14] liu, j., zhang, l., xiao, s., & xin, x. (2014, july). development of virtual drive hils system based on vr and carsim. in proceedings of the 33rd chinese control conference (pp. 6441-6444). ieee. [15] ji, f. z., zhou, x. x., & zhu, w. b. (2014). coordinate control of electro-hydraulic hybrid brake of electric vehicles based on carsim. in applied mechanics and materials (vol. 490, pp. 1120-1125). trans tech publications ltd. [16] dumitriu, d. n., chiroiu, v., & munteanu, l. (2015). car vertical dynamics simulations using both an in-house 7 dof model simulator and carsim commercial software. upb scientific bulletin, series d: mechanical engineering, 77(1), 77-84. [17] sun, y. k., & fan, x. b. (2017). research on the application of carsim in vehicle simulation and development. international journal, 4. [18] abdullah, m. a., jamil, j. f., & salim, m. a. (2015, november). dynamic performances analysis of a real vehicle driving. in iop conference series: materials science and engineering (vol. 100, no. 1, p. 012017). iop publishing. [19] abdullah, m. a., & rahim, m. a. (2016). driving behaviour analysis of young vehicle drivers. proceedings of mechanical engineering research day, 2016, 19-20. journal of mechanical engineering and technology issn 2180-1053 vol. 13 no. 1 june – december 2021 53 [20] abdullah, m. a., jamil, j. f., & mohan, a. e. (2016). vehicle dynamics modeling & simulation. centre for advanced research on energy (care), faculty of mechanical engineering, universiti teknikal malaysia melaka, 2016. [21] abdullah, m. a., ibrahim, m., & abdul rahim, m. a. h. (2017). experimental and analysis of vehicle dynamics performance based on driving behavior. journal of mechanical engineering (jmeche), (1), 193-206. [22] abdullah, m. a., jamil, j. f., yamin, a. m., nuri, n. m., & hassan, m. z. (2015). vehicle dynamics. teaching and learning series, faculty of mechanical engineering, module, 10. [23] abdullah, m. a., salim, m. a., & nasir, m. m. (2014). dynamics performances of malaysian passenger vehicle. arpn journal of engineering and applied sciences 10 (17), 7759-7763. [24] zhang, h., zhang, x., & wang, j. (2014). robust gain-scheduling energy-to-peak control of vehicle lateral dynamics stabilisation. vehicle system dynamics, 52(3), 309340. [25] sierra, c., tseng, e., jain, a., & peng, h. (2006). cornering stiffness estimation based on vehicle lateral dynamics. vehicle system dynamics, 44(sup1), 24-38. [26] zhang, h., & wang, j. (2015). vehicle lateral dynamics control through afs/dyc and robust gain-scheduling approach. ieee transactions on vehicular technology, 65(1), 489-494. [27] oudghiri, m., chadli, m., & el hajjaji, a. (2008). robust observer-based faulttolerant control for vehicle lateral dynamics. international journal of vehicle design, 48(3-4), 173-189. [28] du, h., zhang, n., & naghdy, f. (2011). velocity-dependent robust control for improving vehicle lateral dynamics. transportation research part c: emerging technologies, 19(3), 454-468. [29] du, h., zhang, n., & dong, g. (2010). stabilizing vehicle lateral dynamics with considerations of parameter uncertainties and control saturation through robust yaw control. ieee transactions on vehicular technology, 59(5), 2593-2597. [30] mammar, s., glaser, s., & netto, m. (2006, june). vehicle lateral dynamics estimation using unknown input proportional-integral observers. in 2006 american control conference (pp. 6-pp). ieee. [31] el hajjaji, a., chadli, m., oudghiri, m., & pages, o. (2006, june). observer-based robust fuzzy control for vehicle lateral dynamics. in 2006 american control conference (pp. 6-pp). ieee. journal of mechanical engineering and technology issn 2180-1053 vol. 13 no. 1 june – december 2021 54 [32] liaw, d. c., chiang, h. h., & lee, t. t. (2007). elucidating vehicle lateral dynamics using a bifurcation analysis. ieee transactions on intelligent transportation systems, 8(2), 195-207. [33] huang, y., liang, w., & chen, y. (2021). stability regions of vehicle lateral dynamics: estimation and analysis. journal of dynamic systems, measurement, and control, 143(5). [34] sakthivel, r., mohanapriya, s., ahn, c. k., & selvaraj, p. (2018). state estimation and dissipative-based control design for vehicle lateral dynamics with probabilistic faults. ieee transactions on industrial electronics, 65(9), 7193-7201. [35] su, j., & chen, w. h. (2016, march). fault diagnosis for vehicle lateral dynamics with robust threshold. in 2016 ieee international conference on industrial technology (icit) (pp. 1777-1782). ieee. [36] su, j., & chen, w. h. (2016, march). fault diagnosis for vehicle lateral dynamics with robust threshold. in 2016 ieee international conference on industrial technology (icit) (pp. 1777-1782). ieee. [37] jin, x., yin, g., li, y., & li, j. (2016). stabilizing vehicle lateral dynamics with considerations of state delay of afs for electric vehicles via robust gain‐scheduling control. asian journal of control, 18(1), 89-97. [38] jin, x., yin, g., li, y., & li, j. (2016). stabilizing vehicle lateral dynamics with considerations of state delay of afs for electric vehicles via robust gain‐scheduling control. asian journal of control, 18(1), 89-97. [39] farrelly, j., & wellstead, p. (1996, september). estimation of vehicle lateral velocity. in proceeding of the 1996 ieee international conference on control applications ieee international conference on control applications held together with ieee international symposium on intelligent contro (pp. 552-557). ieee. [40] peng, h. (1993). vehicle lateral control for highway automation. [41] mohanapriya, s., sakthivel, r., & almakhles, d. j. (2020). repetitive control design for vehicle lateral dynamics with state-delay. iet control theory & applications, 14(12), 1619-1627. [42] varrier, s., koenig, d., & martinez, j. j. (2012, december). robust fault detection for vehicle lateral dynamics. in 2012 ieee 51st ieee conference on decision and control (cdc) (pp. 4366-4371). ieee. [43] kuiper, e. v. o. j., & van oosten, j. j. m. (2007). the pac2002 advanced handling tire model. vehicle system dynamics, 45(s1), 153-167. [44] tezuka, y., ishii, h., & kiyota, s. (2001). application of the magic formula tire model to motorcycle maneuverability analysis. jsae review, 22(3), 305-310. journal of mechanical engineering and technology issn 2180-1053 vol. 13 no. 1 june – december 2021 55 [45] mashadi, b., mousavi, h., & montazeri, m. (2015). obtaining relations between the magic formula coefficients and tire physical properties. international journal of automotive engineering, 1, 911-922. [46] mizuno, m. (2003). development of tire side force model based on magic formula with the influence of tire surface temperature. r&d review of toyota crdl, 38(4), 1722. [47] cabrera, j. a., castillo, j. j., pérez, j., velasco, j. m., guerra, a. j., & hernández, p. (2018). a procedure for determining tire-road friction characteristics using a modification of the magic formula based on experimental results. sensors, 18(3), 896. [48] pacejka, h. b., & bakker, e. (1992). the magic formula tyre model. vehicle system dynamics, 21(s1), 1-18. [49] boyle, s. (2019, december). pacejka magic formula tire model parser. in 2019 international conference on computational science and computational intelligence (csci) (pp. 517-518). ieee. [50] li, b., yang, x., & yang, j. (2014). tire model application and parameter identification-a literature review. sae international journal of passenger carsmechanical systems, 7(2014-01-0872), 231-243. [51] guang-sheng, r. e. n. (2001). optimization of curve fitting used in development of magic formula tire model [j]. journal of chongqing university (natural science edition), 3. [52] salaani, m. k., heydinger, g. j., & grygier, p. a. (2006). measurement and modeling of tire forces on a low coefficient surface. sae transactions, 392-399. [53] allen, r. w., magdaleno, r. e., rosenthal, t. j., klyde, d. h., & hogue, j. r. (1995). tire modeling requirements for vehicle dynamics simulation. sae transactions, 484-504. [54] kasprzak, e. m., & gentz, d. (2006). the formula sae tire test consortium-tire testing and data handling (no. 2006-01-3606). sae technical paper. [55] kasprzak, e. m., lewis, k. e., & milliken, d. l. (2006). inflation pressure effects in the nondimensional tire model. sae transactions, 1781-1792. [56] sayers, m. w., & han, d. (1996). a generic multibody vehicle model for simulating handling and braking. vehicle system dynamics, 25(s1), 599-613. [57] sadeghi, s., & ahmadian, m. t. (2001). tire modeling with nonlinear behavior for vehicle dynamic studies. journal of mechanical engineering and technology issn 2180-1053 vol. 13 no. 1 june – december 2021 56 [58] nasir, m. z. m., hudha, k., amir, m. z., & kadir, f. a. a. (2012). modelling, simulation and validation of 9 dof vehicles model for automatic steering system. in applied mechanics and materials (vol. 165, pp. 192-196). trans tech publications ltd. [59] allen, r. w., rosenthal, t. j., & chrstos, j. p. (1997). a vehicle dynamics tire model for both pavement and off-road conditions (no. 970559). sae technical paper. [60] bergman, w., & clemett, h. r. (1975). tire cornering properties. tire science and technology, 3(3), 135-163. [61] ding, n., & taheri, s. (2010). a modified dugoff tire model for combined-slip forces. tire science and technology, 38(3), 228-244. [62] chen, l., bian, m., luo, y., & li, k. (2013, july). maximum tire road friction estimation based on modified dugoff tire model. in 2013 international conference on mechanical and automation engineering (pp. 56-61). ieee. [63] bian, m., chen, l., luo, y., & li, k. (2014). a dynamic model for tire/road friction estimation under combined longitudinal/lateral slip situation (no. 2014-01-0123). sae technical paper. [64] song, s., chun, m. c. k., huissoon, j., & waslander, s. l. (2014, june). pneumatic trail based slip angle observer with dugoff tire model. in 2014 ieee intelligent vehicles symposium proceedings (pp. 1127-1132). ieee. [65] chen, l., bian, m., luo, y., & li, k. (2015). estimation of road-tire friction with unscented kalman filter and mse-weighted fusion based on a modified dugoff tire model (no. 2015-01-1601). sae technical paper. [66] he, r., jimenez, e., savitski, d., sandu, c., & ivanov, v. (2016). investigating the parameterization of dugoff tire model using experimental tire-ice data. sae international journal of passenger cars-mechanical systems, 10(2016-01-8039), 83-92. [67] dugoff, h., fancher, p. s., & segel, l. (1970). an analysis of tire traction properties and their influence on vehicle dynamic performance. sae transactions, 1219-1243. [68] zhou, l., & zhang, x. w. (2012). simulation of vehicle dynamics in tire blow-out process based on dugoff tire model [j]. computer simulation, 6. [69] villagra, j., d’andréa-novel, b., fliess, m., & mounier, h. (2011). a diagnosisbased approach for tire–road forces and maximum friction estimation. control engineering practice, 19(2), 174-184. [70] bhoraskar, a., & sakthivel, p. (2017, january). a review and a comparison of dugoff and modified dugoff formula with magic formula. in 2017 international conference on nascent technologies in engineering (icnte) (pp. 1-4). ieee. journal of mechanical engineering and technology issn 2180-1053 vol. 13 no. 1 june – december 2021 57 [71] kissai, m., monsuez, b., tapus, a., & martinez, d. (2017, september). a new linear tire model with varying parameters. in 2017 2nd ieee international conference on intelligent transportation engineering (icite) (pp. 108-115). ieee. [72] han, k. s., lee, e., & choi, s. (2015, october). estimation of the maximum lateral tire-road friction coefficient using the 6-dof sensor. in 2015 15th international conference on control, automation and systems (iccas) (pp. 1734-1738). ieee. [73] jin, x., yin, g., & lin, y. (2014). interacting multiple model filter-based estimation of lateral tire-road forces for electric vehicles (no. 2014-01-2321). sae technical paper. [74] iso 3888-2:2011, passenger cars test track for a severe lane-change manoeuvre part 2: obstacle avoidance. [75] peng, y., & yang, x. (2012). comparison of various double-lane change manoeuvre specifications. vehicle system dynamics, 50(7), 1157-1171. [76] lee, j., & chang, h. j. (2018). explicit model predictive control for linear timevariant systems with application to double-lane-change maneuver. plos one, 13(12), e0208071. [77] katzourakis, d., de winter, j. c., de groot, s., & happee, r. (2012). driving simulator parameterization using double-lane change steering metrics as recorded on five modern cars. simulation modelling practice and theory, 26, 96-112. [78] el hajjaji, a., & ouladsine, m. (2001, september). modeling human vehicle driving by fuzzy logic for standardized iso double lane change maneuver. in proceedings 10th ieee international workshop on robot and human interactive communication. roman 2001 (cat. no. 01th8591) (pp. 499-503). ieee. [79] arefnezhad, s., ghaffari, a., khodayari, a., & nosoudi, s. (2018). modeling of double lane change maneuver of vehicles. international journal of automotive technology, 19(2), 271-279. [80] hatipoglu, c., ozguner, u., & redmill, k. a. (2003). automated lane change controller design. ieee transactions on intelligent transportation systems, 4(1), 13-22. [81] huang, c., naghdy, f., & du, h. (2016, november). model predictive controlbased lane change control system for an autonomous vehicle. in 2016 ieee region 10 conference (tencon) (pp. 3349-3354). ieee. [82] kutluay, e., & winner, h. (2012, december). assessment methodology for validation of vehicle dynamics simulations using double lane change maneuver. in proceedings of the 2012 winter simulation conference (wsc) (pp. 1-12). ieee. [83] naude, a. f., & steyn, j. l. (1993). objective evaluation of the simulated handling characteristics of a vehicle in a double lane change manoeuvre (no. 930826). sae technical paper. journal of mechanical engineering and technology issn 2180-1053 vol. 13 no. 1 june – december 2021 58 [84] sledge jr, n. h., & marshek, k. m. (1997). comparison of ideal vehicle lanechange trajectories. sae transactions, 2004-2027. [85] hess, d., & sattel, t. (2011, october). double-lane change optimization for a stochastic vehicle model subject to collision probability constraints. in 2011 14th international ieee conference on intelligent transportation systems (itsc) (pp. 206211). ieee. [86] yang, x., & gander, j. (2011). driver's preview strategy and its impact on nato double lane change maneuver. sae international journal of materials and manufacturing, 4(1), 1025-1035. [87] forkenbrock, g. j., garrott, w. r., heitz, m., & o'harra, b. c. (2003). an experimental examination of double lane change maneuvers that may induce on-road, untripped, light vehicle rollover. sae transactions, 1128-1144. [88] angelis, s., tidlund, m., leledakis, a., lidberg, m., nybacka, m., & katzourakis, d. (2014). optimal steering for double-lane change entry speed maximization. in acev'14 international symposium on advanced vehicle control, 22-26 september 2014, tokyo, japan. society of automotive engineers. [89] willmott, c. j., & matsuura, k. (2005). advantages of the mean absolute error (mae) over the root mean square error (rmse) in assessing average model performance. climate research, 30(1), 79-82. [90] chai, t., & draxler, r. r. (2014). root mean square error (rmse) or mean absolute error (mae). geoscientific model development discussions, 7(1), 1525-1534. [91] chai, t., & draxler, r. r. (2014). root mean square error (rmse) or mean absolute error (mae)?–arguments against avoiding rmse in the literature. geoscientific model development, 7(3), 1247-1250. [92] maiorov, v. n., & crippen, g. m. (1994). significance of root-mean-square deviation in comparing three-dimensional structures of globular proteins. journal of molecular biology, 235(2), 625-634. [93] salmon, t. o., & van de pol, c. (2006). normal-eye zernike coefficients and rootmean-square wavefront errors. journal of cataract & refractive surgery, 32(12), 20642074. [94] applegate, r. a., ballentine, c., gross, h., sarver, e. j., & sarver, c. a. (2003). visual acuity as a function of zernike mode and level of root mean square error. optometry and vision science, 80(2), 97-105. [95] huffman, g. j. (1997). estimates of root-mean-square random error for finite samples of estimated precipitation. journal of applied meteorology, 36(9), 1191-1201. journal of mechanical engineering and technology issn 2180-1053 vol. 13 no. 1 june – december 2021 59 [96] hespanha, j. p. (2003). root-mean-square gains of switched linear systems. ieee transactions on automatic control, 48(11), 2040-2045. [97] distefano, c., liu, j., jiang, n., & shi, d. (2018). examination of the weighted root mean square residual: evidence for trustworthiness?. structural equation modeling: a multidisciplinary journal, 25(3), 453-466. [98] o’donovan, t. s., & murray, d. b. (2007). jet impingement heat transfer–part i: mean and root-mean-square heat transfer and velocity distributions. international journal of heat and mass transfer, 50(17-18), 3291-3301. [99] szostack, h. t., allen, r. w., & rosenthal, t. j. (1988). analytical modeling of driver response in crash avoidance maneuvering. volume 2: an interactive tire model for driver/vehicle simulation (no. sti-tr-1227-1-v2). issn 2180-1053 e-issn 2289-8123 vol.15 no.1 25 3d modelling of a reconditioned piston of a single-cylinder four-stroke diesel engine by using solid works software f.o isaac1, o. obodeh2 and o. ighodalo2 1faculty of engineering, department of mechanical engineering, edo state university uzairue. 2edo state, nigeria, 2faculty of engineering and technology, department of mechanical engineering, ambrose alli university, ekpoma. edo state, nigeria corresponding author’s email: isaac.oamen@edouniversity.edu.ng article history: received 11 february 2023; revised 03 march 2023; accepted 28 march 2023 abstract: this paper gives the possibility of modelling a reconditioned piston of a single-cylinder four-stroke diesel engine using the zs1115nm diesel engine specifications. due to the upsurge of counterfeit spare parts in the market, meeting the original equipment manufacturer (oem) standards requires a reconditioning process. the reconditioned piston is a thermal barrier coated one with a ceramic material that enables it to withstand high gas combustion temperatures without cracking. a piston converts thermal energy to mechanical energy in an internal combustion engine (ice). the methodology includes sizing and modelling of the conventional piston, topcoat and bond-coat layers and finally assembling them to get a reconditioned piston using solidworks computer-aided design (cad) software. the material chosen for the piston is an aluminum alloy designated as a2618, due majorly to its high coefficient of thermal expansion (cte) which enables the piston to withstand high thermal stress without cracking or failing. the ceramic material chosen is a 7.5% yttria-stabilized zirconia which is the topcoat with low thermal conductivity and a high coefficient of thermal expansion (cte) on a bond-coat metallic material called nickel chromium aluminum cobalt yttria which are applied by plasma sprayed method on the crown of the substrate. the chosen thickness from the literature of the topcoat layer is 0.35 mm and that of the bond-coat layer is 0.15 mm. also, from the literature, the major reason for the thermal barriercoating (tbc) of a diesel engine piston crown using a ceramic material was to improve its performance. keywords: 3d-modelling; original equipment manufacturer; solidworks software; 7.5% yttriastabilized zirconia; nickel chromium aluminum cobalt yttria 1.0 introduction the thermal barrier-coatings (tbcs) are advanced ceramic materials applied on metallic surfaces of aero-engine, turbine and spark and compression-ignition engine parts (cylinder liner, cylinder head, valves, piston crown, etc.), which work at very high temperatures [1]. coatings help to insulate metallic parts from heavy and excessive heat loads using thermally insulating materials which withstand reasonable temperature difference between the combustion chamber and coating surfaces. this results to high operating temperatures on the metallic or component surfaces. coatings also reduce the problems of oxidation and thermal fatigue in order to extend the life span of the machine components. modern coating systems behave as barriers to heat transfer through metallic surfaces so as to protect engine parts from oxidation and hot corrosion [2,3]. the piston in an internal combustion engine (ice) is a round piece of metal that converts the rotary motion of the crank-shaft into a reciprocating motion in the cylinder and exerts a force on the air-fuel mixture contained in the cylinder [4]. piston has journal of mechanical engineering and technology (jmet) 26 issn 2180-1053 e-issn 2289-8123 vol.15 no.1 compression and oil control rings preventing oil from entering the combustion-chamber including the fuel air from mixing with the oil [5]. most fitted pistons in engine cylinders have piston rings [6]. two or more compression rings are acting as seals or barriers between the piston and cylinder-wall. there are also one or two oil control-rings below the compression-rings (figure 1). the piston head may be flat, bulged or otherwise shaped. pistons which are either forged or cast have rounded shapes [5]. the preferred common materials for petrol and diesel engine pistons are aluminium alloys because they possess high thermal conductivity, low density, simple machinability, highreliability, simple fabrication processes and very good recycling-characteristics [7]. figure 1: the different parts or elements of the piston the single-component coating has not satisfied some multifunctional requirements of some engine parts. as a result, a complex thermal-barrier-coating structure was introduced. research from the 1970s focused on a preferred coating system that comprises three separate layers on the substrate to achieve long-term improvement in the control of oxidation and corrosion at high temperatures [8,9]. adnan et al. [13] conducted a test on a single-cylinder, indirect injection ricardo e6-ms/128/76 type diesel engine. they coated the cylinder head, valves and piston with mgo–zro2 layer having 0.35 mm thickness on a nicral bond-coat layer also having 0.15 mm thickness. they discovered that in low-heat-rejection (lhr) diesel-engine the nox emissions were reduced by about 40% and the brake specific fuel consumption (bsfc) also reduced by about 6% compared to the conventional engines when injection timing was retarded to 3400 crank-angles before top dead centre (tdc) to that of a conventional engine. ekrem et al. [14] compared a conventional engine with a lhr engine. they used mgzro3 as a coating material for the diesel piston and cazro3 for the cylinder head and valves. the piston was coated with mgzro3 having a thickness of 350 μm on a nicral bond-coatlayer with 150 μm thickness. the results obtained showed that the combustion gas temperature for the lhr engine was increased by approximately 65 0c while the bsfc and particulate emissions were reduced by about 6% and 40%, respectively as compared to a conventional engine. rohini and prema [11] reviewed thermal barrier-coating (tbc) on the same diesel engine performance to improve thermal efficiency by reducing 3d modelling of a reconditioned piston of a singlecylinder four-stroke diesel engine by using solid works software issn 2180-1053 e-issn 2289-8123 vol.15 no.1 27 specific fuel consumption (sfc) and exhaust emissions [15,16]. they were able to make a comparison between a standard diesel engine and a low-heat-rejection (lhr) engine. experimental results from various researchers show improvement in efficiency and rate of specific fuel consumption. navin et al. [16] analyzed the performance and emission of a thermal barrier-coated engine by using palm oil biodiesel and diesel as fuels. they prepared tbc using a series of a mixture consisting of different blend ratios of yttriastabilized zirconia (y2o3.zro2) and aluminium oxide-silicon oxide (al2o3-sio2) via plasma spray coating method. their experimental results revealed the mixtures of tbc with 60% y2o3.zro2 + 40% al2o3-sio2 had excellent nitrogen oxide (no), carbon monoxide (co), carbon dioxide (co2), and unburned hydrocarbon (ubhc) reductions when compared with other blend-coated pistons [12,17-18]. plasma spray-coating system is the process whereby a powder feedstock is injected into a high-temperature plasma-jet where finely divided metallic and non-metallic materials are deposited in a molten or semi-molten state on a prepared substrate [19,20]. it is used as an effective and economical method for producing ceramic-coatings on metallicsubstrates and production of bulk-powders from spheroidization [21]. the plasma spray-coating system is shown in figure 2 while the spraying-gun system is displayed in figure 3. the system consists of a power unit, gas supply unit, spraying-gun, powdersupply unit, cooling-system and control unit [20,22]. figure 2: plasma spray-coating system the plasma spray-coating is the most widely accepted method of coating [20,23]. figure 3 shows some coated piston tops. journal of mechanical engineering and technology (jmet) 28 issn 2180-1053 e-issn 2289-8123 vol.15 no.1 (a) (b) figure 3: ceramic coated piston tops 2.0 methodology 2.1 materials the three (3) materials used for the 3d modelling were a metal substrate called the aluminium alloy piston a2618, a metallic bond-coat of thickness 0.15 mm called the nickel chromium aluminium cobalt yttria (nicralcoy) with a chemical composition of bal ni, 17.5% cr, 5.5% al, 2.5% co, 0.5% y2o3; and ceramic topcoat also of thickness 0.35 mm called the yttria stabilized zirconia (7.5% y2o3-zro2) [22]. this paper is part of our phd work [22]. 2.2 methods 2.2.1 the design of the piston elements figure 4 shows the cross-sectional view of conventional or uncoated piston [22]. figure 4: cross-section of the model conventional piston 3d modelling of a reconditioned piston of a singlecylinder four-stroke diesel engine by using solid works software issn 2180-1053 e-issn 2289-8123 vol.15 no.1 29 2.2.1.1 design of the thickness of the piston head or crown, tc according to grashoff’s formula, the piston head thickness tc is given by, 𝑡𝐶 = √ 3𝑝𝑚𝑎𝑥𝐷 2 16𝜎𝑦 (1) where 𝑝𝑚𝑎𝑥 is the maximum gas pressure (pa), d is the cylinder bore or outside diameter of the piston (m), 𝜎𝑦 is the permissible or yield tensile stress (strength) for the piston material (pa). 2.2.1.2 the number of piston rings from figure 2, we have total number of rings = 4 (number of compression rings = 3 and number of oil ring = 1) 2.2.1.3 design of the radial thickness of the piston ring, t1 consider eq. (2) for the design of piston ring radial thickness. 𝑡1 = 𝐷√ 3𝑝𝑤 𝜎𝑝 (2) where 𝑝𝑤 is an allowable radial pressure of the gas on the cylinder wall taken as 0.025 mpa, σp is permissible bending or tensile stress for cast iron rings which is 84 mpa. 2.2.1.4 design of the axial thickness of piston ring, t2 also, consider eq. (3) for the design of piston ring axial thickness. 𝑡2 = 𝐷 10𝑛𝑅 or = 0.7t1 (3) where nr = number of rings taken as 4. 2.2.1.5 determining the length of the piston pin in the connecting rod bushing, 𝒍𝟏 eq. (4) gives the length of the piston pin. 𝑙1 = 0.45𝐷 (4) 2.2.1.6 design of the width of the piston top land h1 h1= 1.2 tc (5) 2.2.1.7 design of the width of other piston ring lands h2 h2 = 0.75t2 (6) journal of mechanical engineering and technology (jmet) 30 issn 2180-1053 e-issn 2289-8123 vol.15 no.1 2.2.1.8 determining the piston barrel piston barrel thickness t3 at the top end is; t3 = 0.03d + b1 + 4.5 (7) b1 = t1 + 0.4 (8) where b1 = radial depth of the piston ring groove (mm). the piston barrel thickness t4 at the open end is: t4 = 0.25 t3 (9) 2.2.1.9 design of the length of the piston and piston skirt length of the piston skirt, ls = 0.6 d (10) total length of piston l = length of the piston skirt + length of the ring section + top land = ls + (4 t2 + 3 h2) + h1 (11) the length of the piston usually varies from d and 1.5d. 2.2.1.10 design of the diameter of the piston boss and pin outside diameter d0 of piston pin: 𝑑0 = 0.3𝐷 (12) piston boss diameter d = 1.5 d0 (13) although, d0 is given in the owner’s manual. the value is 36 mm. the inside diameter d1 of the piston pin: d1 = 0.6 d0 (14) 2.2.1.11 design of the centre of the pin the centre of the pin is 0.02d to 0.04d above the centre of the skirt. centre of pin = 0.04d + 0.5 ls (15) the specifications for designing and modelling the conventional piston of the dieselengine with the help of solidworks software were that of the zs1115nm singlecylinder, inline and four-stroke direct injection diesel engine manufactured by changchai company ltd, china. the engine specifications are given in table 1 [24]. equations (1) through (15) can only be used in designing the piston if the maximum gas pressure 𝑝𝑚𝑎𝑥 is known. the maximum gas pressure from our phd work was 10.2 x 10 6 n/m2 or 10.2 n/mm2 [22]. the yield tensile strength of the material used for the piston, 3d modelling of a reconditioned piston of a singlecylinder four-stroke diesel engine by using solid works software issn 2180-1053 e-issn 2289-8123 vol.15 no.1 31 t  = 372 n/mm2 [22]. table 2 summarizes the sizes obtained from the design of the piston elements. table 1: engine specification item specification engine model zs1115nm type single cylinder, four stroke, horizontal type, direct injection cylinder bore (d) (mm) 115 piston stroke (𝐿𝑆) (mm) 115 piston displacement (vs) (litre) 1.19 compression ratio (c.r) 17:1 rated output/brake power (b.p)(kw) 15.7 rated speed (n)(rev/min) 2200 brake specific fuel consumption (bsfc) (g/kwh) ≤ 244.8 specific lube oil consumption (g/kwh) ≤ 2.04 lubricating method single circuit cooling method water cooled, evaporative cooling system radiator, natural convection starting method electric starting or hand cranking fuel injection pressure (mpa) 18.13 ± 0.49 net weight (kg) 205 overall dimension (l x w x h) (mm) 965 x 457 x 713 mean piston speed (cm) (m/s) 8.433 fuel injection timing before tdc 220 fuel type diesel chemical formula c14.4h24.9 connecting rod length (𝐿𝑅 ) (mm) 258.5 intake valve closes after tdc 380 table 2: size of piston parameters s/n piston element size obtained (mm) 1 cylinder bore, 𝐷 115 2 thickness of the piston head, 𝑡𝐶 8.25 3 radial thickness of the piston ring, 𝑡1 3.436 4 axial thickness of the piston ring, 𝑡2 2.405 5 no of the piston rings, 𝑛𝑅 4 6 width of the top land, ℎ1 9.9 7 width of the ring land, ℎ2 1.80375 8 radial depth of the piston ring groove, 𝑏1 3.836 9 thickness of the piston barrel at the top end, 𝑡3 11.786 10 thickness of the piston barrel at the open end, 𝑡4 2.9465 11 piston pin diameter, 𝑑0 34.5 12 diameter of the piston boss, 𝑑 51.75 13 length of skirt, 𝑙𝑠 69 14 total length of the piston, 𝐿 93.93 15 centre of pin above the centre of the skirt 39.10 16 inside diameter of the piston pin, 𝑑1 20.7 17 length of the piston pin in the connecting rod bushing, 𝑙1 51.75 journal of mechanical engineering and technology (jmet) 32 issn 2180-1053 e-issn 2289-8123 vol.15 no.1 2.3 solidworks modelling of the conventional piston the sizes obtained from the design of the piston elements in table 4 were used in modelling the conventional piston in solidworks cad software [23,25]. figures 5 and 6 show the views of the modelled conventional piston. figure 5: the isometric view of the 3d modelled conventional piston figure 6: the sectional view of the modelled conventional piston 3d modelling of a reconditioned piston of a singlecylinder four-stroke diesel engine by using solid works software issn 2180-1053 e-issn 2289-8123 vol.15 no.1 33 2.4 solidworks modelling of the bond-coat layer the bond-coat layer of 0.15 mm thick was modelled in solidworks software. see figure 7. figure 7: the modelled isometric view of the bond-coat layer of 0.15 thickness 2.5 solidworks modelling of the topcoat layer the topcoat layer of 0.35 mm thick was modelled in solidworks software. see figure 8. figure 8: the modelled isometric view of the topcoat layer of 0.35 thickness journal of mechanical engineering and technology (jmet) 34 issn 2180-1053 e-issn 2289-8123 vol.15 no.1 2.6 solidworks assembling of the conventional piston, bond and topcoats layers the assembling of the piston, bond-coat and topcoat layers was also carried out using the solidworks software. figures 9 and 10 show the views of the modelled reconditioned piston. figure 9: the assembly view of the reconditioned piston with bond and topcoat layers figure 10: the exploded view of the reconditioned piston with bond and topcoat layers 3d modelling of a reconditioned piston of a singlecylinder four-stroke diesel engine by using solid works software issn 2180-1053 e-issn 2289-8123 vol.15 no.1 35 3.0 results and discussion figures 6 to 11 show the results of modelling conventional and reconditioned pistons of the zs1115nm single-cylinder, inline and four-stroke direct injection diesel engine using solidworks 2013 cad. this modelling provides the next stage involved in the reconditioning or coating of diesel engine pistons for improved performance. findings from literature have it that reconditioned or thermal barrier coated pistons with a ceramic material that have a very low thermal conductivity give higher piston surface temperature and brake thermal efficiency, reduced brake specific fuel consumption and emissions than the conventional ones. 4.0 conclusions due to the upsurge of counterfeit spare parts in the market, meeting the original equipment manufacturer (oem) standards requires a reconditioning process. having modelled the thermal barrier-coated piston of a single-cylinder, inline and four-stroke direct injection diesel engine using solidworks 2013 cad, it could be concluded that with given engine specification suitable materials for designing and modelling a reconditioned piston of diesel engine are chosen and the model reconditioned. acknowledgements we appreciate god almighty and families for supporting us during this work. references [1] b. ekrem, “thermal analysis of functionally graded coating alsi alloy and steel pistons”, journal of energy conversion and management, vol. 202, no. 16, pp. 325-336, 2008. [2] j.i. ramos, internal combustion engine modelling, 2nd edition. usa: taylor and francis, 1989. [3] j.b. heywood, internal combustion engine fundamentals, international edition. usa: mcgraw-hill book company, 1988. [4] j.a. dolan, motor vehicle technology and practical work, reprinted edition. ibadan: heinemann educational books ltd, 1991. [5] m.d. röhrle, pistons for internal combustion engines – fundamentals of piston technology, 2nd edition. germany: verlag moderneindustrie, 1995. [6] g.s. shirisha and d.k. sravani, “thermal analysis of ic engine piston using finite element method”, international journal of mechanical engineering and computer applications, vol. 4, no. 1, pp. 200-209, 2016. [7] s. lokesh, s.r. suneer, h. taufeeque and k. upendra, “finite element analysis of piston in ansys”, international journal of modern trends in engineering and research (ijmter), vol. 2, no. 4, pp. 619-626, 2015. journal of mechanical engineering and technology (jmet) 36 issn 2180-1053 e-issn 2289-8123 vol.15 no.1 [8] j.r. davis, handbook of thermal spray technology. usa: asm international, 2004. [9] t. anders, thermal barrier-coatings for diesel engines, 1st edition. sweden: kth royal institute of technology, 2017. [10] p.m. pierz, “thermal barrier-coating development for diesel engine aluminium pistons”, surface and coatings technology, vol. 61, no. 1, pp. 60-66, 1993. [11] a. rohini and s. prema, “a review on thermal barrier-coating for diesel engine and its characteristics studies”, international conference on thermo-fluids and energy systems, vol. 1473, pp. 1-11, 2020. [12] f.a. ansari and y. dhanajay, “review paper on simulation, analysis and validation on thermal barrier coated piston of diesel engine”, international research journal of modernization in engineering technology and science, vol. 4, no. 2, pp. 260-270, 2022. [13] p. adnan, y. halit, h. can and k. ahmet, “the effects of injection timing on nox emissions of a low heat rejection indirect diesel injection engine”, journal of applied thermal engineering, vol. 25, pp. 3042–3052, 2005. [14] b. ekrem, e. tahsin and c. muhammet, “effects of thermal barrier-coating on gas emissions and performance of a lhr engine with different injection timings and valve adjustments”, journal of energy conversion and management, vol. 47, pp. 298–1310, 2006. [15] b. ekrem and c. muhammet, “thermal analysis of a ceramic coating diesel engine piston using 3-d finite element method”, journal of surface and coatings technology, vol. 202, pp. 398–402, 2007. [16] r. navin, a.k. mohammad, v. mahendra and h.t. yew, “effect of thermal barriercoating on the performance and emissions of diesel engine operated with conventional diesel and palm oil biodiesel”, coatings, vol. 11, no. 692, pp. 1-14, 2021. [17] k. masera and a.k. hossain, “combustion characteristics of cottonseed biodiesel and chicken fat biodiesel mixture in a multi-cylinder compression ignition engine”, sae tech vol. 4, pp. 1-14, 2019. [18] k.a. khor, y. murakosh, m. takahashi, and t. sano, “plasma spraying of titanium aluminide coatings: process parameters and microstructure”, journal of materials processing technology, vol. 48, pp. 413–419, 1995. [19] f.o. isaac, “a review of coating methods and their applications in compression and spark-ignition engines for enhanced performance”, fuoye journal of engineering and technology, vol. 7, no. 2, pp. 217-221, 2022. [20] s. malmberg and j. heberlein, “effect of plasma spray operating conditions on plasma jet characteristics and coating properties”, journal of thermal spray technology, vol. 2, no. 4, pp. 339–344, 1993. [21] s. kuldeep and o.p. jakhar, “the behaviour of temperature on insulated (mgzro3) diesel engine piston with ansys”, international journal of emerging technology and advanced engineering, vol. 4, no. 8, pp. 692-695, 2014. 3d modelling of a reconditioned piston of a singlecylinder four-stroke diesel engine by using solid works software issn 2180-1053 e-issn 2289-8123 vol.15 no.1 37 [22] f.o. isaac, “investigation of the effectiveness of thermal barrier coating in reconditioning diesel engine piston crown,” ph.d. dissertation, department of mechanical engineering, ambrose alli university, ekpoma, edo state, 2023. [23] f.o. isaac and l. abu, (2022). “modelling of a conventional piston of a single-cylinder four-stroke diesel engine by using solidworks”, fuoye journal of engineering and technology, vol. 7, no. 1. pp. 65-68, 2022. [24] zs1115nm, single cylinder diesel engine owner’s manual. changchai company ltd, china, 2015. [25] b.d. james, engineering design and graphics with solidworks, 1st edition. boston: pearson, 2016. issn:2180-1053 e-issn:2289-8123 vol.14 no.1 25 cfd study of a micro air vehicle (mav) l.y. shern1, f. a. z. mohd sa’at1,2,3 and m. a. abdul wahap2,4 1faculty of mechanical engineering,universiti teknikal malaysia melaka, hang tuah jaya, 76100 duriantunggal, melaka, malaysia. 2centre for advanced research on energy, universiti teknikal malaysia melaka, hang tuah jaya, 76100 durian tunggal, melaka, malaysia. 3centre of excellence geopolymer and green technology, universiti malaysia perlis, 01000 kangar, perlis, malaysia. 4faculty of mechanical and manufacturing engineering technology, universiti teknikal malaysia melaka, hang tuah jaya, 76100 durian tunggal, melaka, malaysia. corresponding author’s email: fatimah@utem.edu.my article history: received 3 august 2021; revised 1 october 2022; accepted 11 january 2022 abstract: micro air vehicle (mav) is a new type of aircraft technology that is maturing day by day and has recently reached unprecedented levels of growth. mav is small in size and provides enormous potential in many applications, both for military and civilian use. there are three types of mav, namely rotary wing, flapping wing and fixed wing. due to their small size, mav faces difficulty in flying properly due to atmospheric perturbations. this study aims to model a suitable fixed-wing mav using computational fluid dynamics (cfd) to investigate the lift coefficient, drag coefficient and lift-todrag ratio when mav is used in perturbed flow conditions. when there is wind disturbance, the simulation results show that the lift and drag coefficient for several angles of attack changes. however, the lift-to-drag ratio seems unaffected. results showed that mav is best operated at an 8° angle of attack as it provides the maximum lift-to-drag ratio for situations without and with the presence of wind disturbances. the fluid dynamics behavior of flow around mav is also discussed accordingly. even though mav is small in size, it is found that vortex or vorticity flow also exists in mav, especially at a high degree angle of attack. keywords: micro air vehicle (mav); computational fluid dynamics (cfd); fixedwing; wind disturbances journal of mechanical engineering and technology (jmet) 26 issn:2180-1053 e-issn:2289-8123 vol.14 no.1 1.0 i n t ro d uc t i o n micro air vehicles (mav) are a relatively new technology and are currently undergoing rapid growth of research developments [1]. research related to mav had been carried out in the countries such as the united states, japan and china as early as the 19th century. according to michelson, an american nonprofit institution, rand corporation, which helps in improving policy and making decisions through research and analysis, founded the mav feasibility study in 1994 and concluded that mav had great potential for military applications [2]. mav were first defined by the us defense advanced research projects agency (darpa) in 1997 as unmanned aircraft that are less than 15 centimeters or 6 inches in any dimension [3]. nowadays, the target dimension and development of insect-sized aircraft are reportedly expected in the near future. however, according to hassanalian and abdelkefi, they state that the dimension for mav can also be considered to be within 15 cm to 1 m [4]. besides, aboelezz et al. also mentioned that mavs should have weight in the range between 50 g to 2 kg and a flight endurance of 20 minutes should be expected [5]. also, the materials used to build mav should be as light as possible to reduce the burden of its weight [6]. mav are also categorised as a small kind of unmanned air vehicles (uav) which are used for surveillance, reconnaissance, armed attacking, search and rescue operations, as well as for transportation and scientific research [7]. mav can be remotely or autonomously controlled without a human operator on board [8]. basically, mav are smaller in size and weight compared to uav. because of their smaller size, the probability of mav being intercepted by radar is low, and therefore, they are manufactured for several missions. they can reach a maximum travel speed of 20 m/s and is suitable for day and night usage [9]. thus, they are very suitable for military surveillance applications and image recording. besides, mav also produces lower noise when functioning and have lower production cost compared to the uav. in addition, mav can be operated at reynolds numbers of up to 200000, depending on the size and types of mav [10]. mav has the potential to be used in urban applications to monitor traffic flow and mapping areas. mav can also be used to observe the weather condition and provide real-time tracking of the current cfd study of a micro air vehicle (mav) issn 2180-1053 e-issn 2289-8123 vol.14 no.1 27 location with the installation of gradient sensors and flight control feedback, which provides weather updates from time to time to the community [11]. nowadays, mavs are developed with great improvements in designs with advanced features of computer-aided technology, power supply with better battery technology, and visual communications with better transmitters and receivers [12]. there are various types of mav, such as rotary wings, flapping wings and fixed wings. they are all now in existence, and each of them contains specific capabilities and limitations. rotary wing mavs, as shown in figure 1, are basically functionally similar to the concept of helicopters; the lift and thrust are generated by the spinning of rotor blades. the surface area of rotors would determine the magnitude of the supplied aerodynamic forces. when the rotors are spinning in opposite directions in a balanced manner, the rotary wing mav can be easily stabilized, and the rotation of downwash air will be minimized [13]. thus, the movement can be easily controlled and piloted by the users. in addition, the generation of both lift and thrust could also be increased by using multiple sets of rotors, such as quadrotors. figure 1: sample of rotary wing mav [13] the flapping wing mavs, which are also known as the biomimetic mavs, are bioinspired with the ideas from the flapping wings of insects and birds. a sample of this mav is shown in figure 2. flapping wing mav is relatively smaller in size compared to the others, and it is more suitable for indoor applications due to its ability to fly through a narrow gap. it is more complex and complicated to be built as it consists of lightweight structures and small-scale electronic devices. lift and thrust are achieved by flapping the wings, and the journal of mechanical engineering and technology (jmet) 28 issn:2180-1053 e-issn:2289-8123 vol.14 no.1 corresponding flapping frequency depends on the surface area of the wings [13]. figure 2: sample of flapping wing mav [13] figure 3 shows another type of mav, known as a fixed-wing mav, which may not be very suitable for indoor usage. it is more suitable for outdoor surveillance missions since they have higher payload and endurance capabilities compared to rotary and flapping wing mavs of equal size [13]. a propeller-driven electrical motor is normally used in fixed-wing mav to produce thrust. lift is generated by air flowing over the non-moving wings of airfoil cross sections. fixed-wing mavs have difficulties achieving good performances at low-speed flights as their wings are associated with stringent dimension constraints requiring high cruise speeds. figure 3: sample of fixed-wing mav [13] normally, all of the mavs are equipped with inertial measurement unit (imu) to estimate the angular velocities and accelerations of the vehicles in vertical, longitudinal and lateral axis [14]. in order to improve the flight performance of mav, bowles et al. and patel et al. cfd study of a micro air vehicle (mav) issn 2180-1053 e-issn 2289-8123 vol.14 no.1 29 stated that pressure sensors could be implemented to predict the leading edge flow separation and also trigger plasma flow actuators for flow and attitude control of mav [15, 16]. the small size and low operating speed of mav lead to unique aerodynamic conditions [17]. according to kunz, the most common problem that is faced by the mav, is its difficulty in flying stably at low reynolds numbers due to its small size [18]. previous research indicated that there would be an increase in maximum lift coefficient with a decrease in reynolds number. however, as the reynolds number decreases, the lift-to-drag, l/d ratio also decreases. hence more power is required to operate the flight. therefore, flight at these reynolds numbers is much less efficient than at higher reynolds numbers. it is important to operate the airfoil at its maximum lift-todrag ratio for its optimum performance [19]. besides the limitation of the supplied power, mav also faces technological and manufacturing challenges due to its small feature. flow at low reynolds numbers is dominated by viscosity, and as the reynolds number is reduced, the effects of increasing boundary layer thickness become more significant [20]. this leads to the effect of higher drag conditions. thus, low reynolds numbers affect the aerodynamic efficiency and the propulsion efficiency dramatically. these problems cause difficulty in flying the mavs properly as they have to face sensitivity issues due to atmospheric perturbations [21]. therefore, the lift coefficient, drag coefficient and the fluid dynamics around the mav are important aspects to be investigated so that the mav can function under optimized conditions. 2.0 methodology in this study, catia v5 software will be used to draw the mav model, while ansys version 16.1 cfd fluent software will be used to carry out all relevant simulations. previous results from the other researchers will be used as a reference to validate the simulation models. validation is important as it would affect the accuracy of results for further study and investigation of fluid dynamics behavior on mav under different cases, situations or conditions. based on the number of journal of mechanical engineering and technology (jmet) 30 issn:2180-1053 e-issn:2289-8123 vol.14 no.1 simulations required for the results, a sufficient and optimum volume of a computational domain that satisfied the trailing vortex flow and turbulence establishment is considered. the streamwise length of mav, l (250 mm), is used as a measurement for the computational domain volume. the dimensions of the computational domain are identified as 1l (250 mm) before the leading edge and 3l (750 mm) after the trailing edge, whereas from the top, bottom and sides of the wing are 1.5l (375 mm). figure 4 shows the mav allocated inside the computational domain. the bottom picture of figure 4 shows the 3d view of the domain with inlet and outlet boundaries. figure 4: the side view of the computational domain with dimensions (top) and the three-dimension view of the domain (bottom) the dimensions of the model and the types of the airfoil are obtained from nal’s black kite mav by sankaranarayanan et al. [22] as cited in ramprasadh and devanandh [23]. the profile of the selig 4083 airfoil is chosen for the wing that is investigated in the 3d model. based on the uiuc airfoil coordinates database, selig 4083 airfoil has a maximum thickness of 8% at 22.5% chord and a maximum chamber of 3.4% at 33.5% chord. a 3d cad model of the low aspect ratio (lar) wing for the fixed-wing mav is modeled using catia v5. it is a modified inverse zimmerman (miz) planform with an aspect ratio of 1.45, wingspan of 300 mm, mean aerodynamic chord (mac) of 209 375mm 375mm 250mm 750mm mav inlet outlet mav cfd study of a micro air vehicle (mav) issn 2180-1053 e-issn 2289-8123 vol.14 no.1 31 mm and the center of the wing consists of a chord length of 250 mm. figure 5 shows the cad model of the wing from various views. figure 5: cad model of wing ansys cfd fluent is used to simulate the models. the solver is set as a pressure-based type with absolute velocity formulation and a steadystate simulation. a realizable k-epsilon turbulence model with standard wall functions was employed. the properties of the fluid are treated as constant, and the values are shown in table 1. table 1: fluid properties properties value density (kg/m3) 1.225 specific heat (j/kg.k) 1006.43 thermal conductivity (w/m.k) 0.242 viscosity (kg/m.s) 1.7894e-5 in boundary conditions, the velocity specification method of flow for both inlet and outlet is defined as normal to the boundary, with gauge pressures equal to zero. besides, the walls of mav and domain should be defined as stationary walls with the no-slip condition [24]. table 2 summarised the baseline conditions that were solved for the current simulation. table 2: flow parameters thermodynamic parameters velocity parameters turbulence parameters temperature: 293.20 k pressure: 101325 pa velocity in: 1. x-direction: 12 m/s 2. y-direction: 0 m/s 3. z-direction: 0 m/s turbulence intensity: 0.10% length: 1.00e-4 m for the solution methods, the simple algorithm was used for the pressure-velocity coupling when solving the navier-stokes equations. journal of mechanical engineering and technology (jmet) 32 issn:2180-1053 e-issn:2289-8123 vol.14 no.1 green-gauss node based gradient and second-order pressure were selected for the spatial discretization scheme. the turbulent kinetic energy, turbulent dissipation rate, momentum and energy equations are discretized by the second-order upwind method. a grid-independent test was carried out to ensure that the simulation result was independent of the gird size. the test was done by monitoring the resulting lift coefficients from models with several mesh elements. the test was carried out for a 24° angle of attack to observe the stability of simulated lift coefficient results. the grid test results are shown in table 3 and figure 6. with reference to table 3 and figure 6, the lift coefficient decreases with the number of elements initially until it reaches the grid size of 732207 elements. after that, the lift coefficient result is almost constant. hence, the model, which consists of 732207 elements, is selected for use in this current study as it provides stable results with the least computational time. table 3: grid test on lift coefficient at 24° angle of attack number of elements 488265 522277 606548 732207 771964 777731 781601 lift coefficient, cl 1.138 1.134 1.130 1.125 1.125 1.125 1.126 figure 6: lift coefficient grid test at 24° angle of attack cfd study of a micro air vehicle (mav) issn 2180-1053 e-issn 2289-8123 vol.14 no.1 33 3.0 results and discussions 3.1 validation for the purpose of model validation, the model was set up following the parameter used by ramprasadh and devanandh [23]. the data and results for lift coefficient, cl at different angles of attack are shown in table 4 and figure 7, respectively. from table 4 and figure 7, when the angle of attack increases, the lift coefficient also increases. this indicates that a higher amount of lift force will be generated at a higher degree angle of attack. the simulation results on the lift coefficient for this mav model at 0°, 8°, 16°, and 24° are 0.110, 0.423, 0.760 and 1.125, respectively. the average percentage of error between results from the current model and the experimental and simulation models of ramprasadh and devanandh [23] is 10.30% and 11.85%, respectively. the deviation is bigger as the reynolds number is bigger. however, the general trend of change of lift coefficient with reynolds number is the same as predicted by ramprasadh and devanandh [23]. table 4: lift coefficient at various angles of attack lift coefficient, cl percentage error (%) angle of attack (°) nal experimental data [23] cfd fluent data results [23] simulation results experiment fluent 0 0.097 0.090 0.110 13.40 22.22 8 0.425 0.420 0.423 0.47 0.71 16 0.900 0.830 0.760 15.56 8.43 24 1.275 1.340 1.125 11.76 16.04 average percentage error (%) 10.30 11.85 journal of mechanical engineering and technology (jmet) 34 issn:2180-1053 e-issn:2289-8123 vol.14 no.1 figure 7: lift coefficient at a different angle of attack 3.2 fluid dynamic behavior other than the lift coefficient, the fluid dynamics behavior of flow around mav at a different angle of attack can also be further investigated through the contours and streamline plots. for the convenience of presentation, the plots are presented in tables. the contours of velocity and pressure for the mav model at its middle cross section for each respective angle of attack are shown in table 5. according to bernoulli’s principle, pressure and velocity are inversely proportional. the greater the velocity, the lower the pressure. referring to the contours, the velocity of fluid flow is exactly as predicted, as it is greater at the low-pressure region. as expected, the velocity of flow across the airfoil at 0° angle of the attack shows that the flow is not severely affected by the presence of the airfoil, and the velocity of flow around the airfoil is slightly lower compared to the surrounding due to the viscous effect near the surface. as the angle of attack increases, the velocity of air increases especially within the region near the leading edge. these velocity changes are the main factors that caused the pressure difference around the model. at a 24° angle of attack, the leading edge experiences a maximum velocity field with a value of around 22 m/s. hence, the velocity of airflow within that particular region is relatively big compared to other cases of angles of attack. the ratio of lift-to-drag, l/d, for the case of 8° angle of attack, was recorded to be the best for this design of mav with a value of 6.934. the velocity that is flowing through the upper path from cfd study of a micro air vehicle (mav) issn 2180-1053 e-issn 2289-8123 vol.14 no.1 35 the leading edge is around 16.50 m/s, and the pressure at the leading edge is about 82.424 pa. the pressure contours in table 5 show that there is a maximum pressure distribution at the leading edge of mav model at 0° angle of attack. when the angle of attack increases, the pressure distribution at the lower path of mav also increases. thus, it causes a greater pressure difference between the lower path and the upper path of mav. the differential pressure distributed around the airfoil causes lift force to be generated. hence, the lift force generated and lift coefficient is greater at a higher degree angle of attack as the pressure at both paths is gradually changing. besides lift, the drag force and drag coefficient also increase with the increase of the angle of attack, as there is an increment in the frontal area for the airfoil, which restricts the airflow. the path of air that is flowing across the mav model could also be traced through the streamlines. the results of streamlines at different angles of attack are shown in table 6. based on the results, the streamline from 0° angle of attack is considered as an attached flow, and the velocity is relatively constant throughout the domain. as the angle of attack increases, there is a separation of flow, and the separation of flow is slowly becoming more obvious towards the 24° angle of attack. however, the wake region is still not significant. besides, the streamlines, after passing through the model, do not behave in a parallel flow manner. the velocity of the streamline is relatively similar to the velocity contour as discussed earlier. the maximum velocity at the leading edge is about 22 m/s at a 24° angle of attack. in order to see the flow behaviour on the third dimension, a plane was created around the middle section of mav in the yz plane, as shown in figure 8, to observe the existence of vortex or vorticity of flow at that dimension. the behavior of the flow of air can be represented by the flow vector. again, the plots are presented in the form of a table for ease of data presentation and analysis. based on the results shown in table 7, the flow vector becomes more significant and forms a more obvious circular vector as the angle of attack increases. this indicates that vortex or vorticity of flow exists in the third dimension of the flow around mav, especially at a high degree angle of attack. the flow journal of mechanical engineering and technology (jmet) 36 issn:2180-1053 e-issn:2289-8123 vol.14 no.1 vector shows that the maximum magnitude is around 16.50 m/s at a 24° angle of attack for the selected area of analysis. hence, a higher angle of attack will lead to a greater vortex or vorticity problem, which might influence the performance of mav. therefore, the angle of attack is an important factor that should be taken into consideration for design purposes. figure 8: side view of mav at 24° angle of attack 3.3 the impact of wind disturbances considering that the mav is operating under an actual condition where the cruise takes place in uncertain flow conditions and with the presence of wind disturbances, additional investigation is carried out to determine whether the impact of wind would affect the cl, cd or the overall performance of this mav. typically, there are 3 cases that were investigated. first, wind disturbance of 2 m/s in both x and z directions, respectively, where x direction is the direction that is parallel to the flow, whereas z direction is from the side. secondly, the wind velocity in the x direction is amplified to 5.5 m/s, and then lastly, the condition with a wind of 3 m/s coming from the z direction was also solved. these conditions were identified based on findings from literature surveys. the conditions were stated as maximum potential disturbances which will cause the mav to be dynamically unstable [5]. table 8 and figure 9 show the resulting lift coefficient of mav under the three different situations with 4 cases of the angle of attack. table 5: velocity and pressure contour at a different angle of attack angle of attack (°) velocity contour pressure contour cfd study of a micro air vehicle (mav) issn 2180-1053 e-issn 2289-8123 vol.14 no.1 37 0 8 16 24 table 6: streamlines at a different angle of attack angle of attack (°) streamline journal of mechanical engineering and technology (jmet) 38 issn:2180-1053 e-issn:2289-8123 vol.14 no.1 0 8 16 24 table 7: flow vector at a different angle of attack angle of attack (°) flow vector cfd study of a micro air vehicle (mav) issn 2180-1053 e-issn 2289-8123 vol.14 no.1 39 0 8 16 24 table 8: lift coefficient in a different situation lift coefficient, cl angle of without wind wind in (x,z), wind in x, wind in z, journal of mechanical engineering and technology (jmet) 40 issn:2180-1053 e-issn:2289-8123 vol.14 no.1 attack (°) disturbance 2 m/s 5.5 m/s 3 m/s 0 0.110 0.110 0.168 0.086 8 0.423 0.420 0.424 0.204 16 0.760 0.750 1.166 0.731 24 1.125 1.108 1.129 0.539 figure 9: lift coefficient under different situations results in table 8 and figure 9 shows that when there is wind disturbances of 2 m/s in both x and z directions, there will be no significant effect on the cl of mav from 0° to 24° angle of attack compared to the situation that is without any wind disturbances which was discussed earlier. however, when the wind is only coming from x direction with a higher velocity of 5.5 m/s, the cl of mav at 16° angle of attack increases significantly to 1.129 (maximum), and it reaches a stall condition for 24° angle of attack as the cl dropped. thus, the 16° angle of attack reflects that it is the critical angle of attack for that case. the cl will decrease once the critical angle of attack is exceeded [25]. for the wind disturbance from z direction with a velocity of 3 m/s, the cl at 8° and 24° angles of attack are relatively lower compared to that of without wind disturbance, but the cl does not affect much at 0° and 16° angle of attack. besides, the mav also experienced a stalling condition at a 24° angle of attack when there was a wind of 3 m/s from the z direction. hence, the wind in different cases might be affecting the cl of mav for several angles of attack. from table 9 and figure 10, it seems that the wind disturbances in both x and z directions with 2 m/s strength are still not having a big impact on the cd of mav for all the investigated angles of attack. but when cfd study of a micro air vehicle (mav) issn 2180-1053 e-issn 2289-8123 vol.14 no.1 41 the wind is 5.5 m/s in x direction, the cd at 16° is higher in comparison to all other cases. for the case of wind in the z direction with 3 m/s, the cd at 8° slightly dropped, and the cd at 24° is extremely lower compared to the situation without wind disturbances. these show that the influence of wind disturbances would also affect the cd significantly depending on the situation. table 9: drag coefficient in a different situation drag coefficient, cd angle of attack (°) without wind disturbance wind in (x,z), 2 m/s wind in x, 5.5 m/s wind in z, 3 m/s 0 0.023 0.023 0.034 0.019 8 0.061 0.061 0.059 0.030 16 0.156 0.154 0.236 0.152 24 0.342 0.336 0.342 0.165 figure 10: drag coefficient under different situations table 10 and figure 11 show the performance of mav based on the liftto-drag, l/d ratio. it can be seen that the l/d ratios are relatively similar for all of the investigated cases at every angle of attack. the l/d ratio is just slightly higher at 8° when 5.5 m/s wind is imposed from the x direction. since the trend of the graph for the l/d ratio for all situations is almost the same, it can be concluded that the l/d ratio is not much affected by the presence of wind disturbances. it seems that, although the cl and cd were shown earlier to be affecting the mav performance, the l/d ratio is not. the overall performance of mav is still the best at an 8° angle of attack as the l/d ratio is highest between 0° to 24° angle of attack. journal of mechanical engineering and technology (jmet) 42 issn:2180-1053 e-issn:2289-8123 vol.14 no.1 table 10: lift-to-drag ratio in a different situation lift-to-drag ratio, l/d angle of attack (°) without wind disturbance wind in (x,z), 2 m/s wind in x, 5.5 m/s wind in z, 3 m/s 0 4.783 4.783 4.941 4.526 8 6.934 6.885 7.186 6.800 16 4.865 4.870 4.941 4.809 24 3.296 3.298 3.301 3.267 figure 11: lift-to-drag ratio under different situations 4.0 c o n c l u s i o n a computational fluid dynamics, cfd, study of the impact of flow disturbances on the performance of a fixed-wing mav is reported. the model was first validated with a benchmarked case. the results from the validated cfd models show that the best angle of attack for this mav model is 8° as it obtained the maximum l/d ratio compared to 0°, 16° and 24° angles of attack. besides, as mav is smaller in size compared to other aircraft, such as uav, it may have sensitivity issues when operating in an actual situation. thus, the parametric investigation is carried out to determine whether flow disturbances are giving an impact on the lift coefficient, drag coefficient and lift-to-drag ratio for this mav model when it is operating under conditions with wind disturbances. the simulation results, it shows that the lift coefficient and drag coefficient might be influenced by the wind disturbances on some angle of attack and could cause potential stall condition. however, the lift-to-drag ratio, which determines the performance of mav, is not affected significantly. in conclusion, this mav model is found to be able to operate well even with disturbances (within the limitation of investigated cfd study of a micro air vehicle (mav) issn 2180-1053 e-issn 2289-8123 vol.14 no.1 43 conditions) and had the best performance at an 8° angle of attack with the consideration of wind disturbances. acknowledgments the authors would like to thank universiti teknikal malaysia melaka for providing the facilities for conducting the research works. references [1] d.m. atwater, “the commercial global drone market”, emerging opportunities for social and environmental uses of uavs, pp. 18, 2015. [2] r.c. michelson, “very small flying machines”, yearbook of science and technology, mcgraw-hill, new york, pp. 341-344, 2006. [3] l. petricca, p. ohlckers, and c. grinde, “microand nano-air vehicles: state of the art”, international journal of aerospace engineering, pp. 1-17, 2011. [4] m. hassanalian, and a. abdelkefi, “classifications, applications and design challenges of drones: a review”, progress in aerospace science, pp. 99–131, 2017. [5] a. aboelezz, m. hassanalian, a. desoki, b. elhadidi, and g. elbayoumi, “design, experimental investigation and nonlinear flight dynamics with atmospheric disturbances of a fixed wing micro air vehicle”, aerospace science and technology, vol. 97, 2020. [6] m. hossain, f. hasan, a. seraz, and s. rajib, “development of design and manufacturing of a fixed wing radio controlled micro air vehicle (mav)”, mist journal: galaxy (dhaka), vol.3, 2011. [7] d. hodgkinson, and r. johnston, “aviation law and drones”, unmanned aircraft and the future of aviation, routledge, 2018 [8] a. tahir, j. boling, m. haghbayan, h.t. toivonen, and j. plosila, “swarms of unmanned aerial vehicles a survey”, journal of industrial information integration, vol. 16, 2019. [9] d.f. kurtulus, “unsteady aerodynamics of a pitching naca 0012 airfoil at low reynolds number”, international journal of micro air vehicles, vol. 11, 2019. [10] t.j. mueller, and g.e. torres, “aerodynamics of low aspect ratio wings at low reynolds numbers with applications”, micro air vehicle design and optimization, pp. 39-72, 2001. journal of mechanical engineering and technology (jmet) 44 issn:2180-1053 e-issn:2289-8123 vol.14 no.1 [11] a. panta, a. mohamed, m. marino, s. watkins, and a. fisher, “unconventional control solutions for small fixed wing unmanned aircraft”, progress in aerospace sciences, vol. 102, pp. 122-135, 2018. [12] b. bataille, d. poinsot, c. thipyopas, and j.m. moschetta, “fixed-wing micro air vehicles with hovering capabilities”, platform innovations and system integration for unmanned air, land and sea vehicles (avtsci joint symposium), vol. 38, pp. 1-16, 2007. [13] t.a. ward, c.j. fearday, s. erfan, and s. norhayati, “a bibliometric review of progress in micro air vehicle research”, international journal of micro air vehicles, pp. 1-20, 2017. [14] g.f. emilio, o. alberto, b.p. francisco, and p.c. joan, “a mosaicing approach for vessel visual inspection using a micro aerial vehicle”, ieee/rsj international conference on intelligent robots and systems (iros), 2015. [15] p. bowles, t. corke, and e. matlis, “stall detection on a leading-edge plasma actuated pitching airfoil utilizing onboard measurement”, 47th aiaa aerospace sciences meeting including the new horizons forum and aerospace exposition, 2009. [16] m.p. patel, z.h. sowle, t.c. corke, and c. he, “autonomous sensing and control of wing stall using a smart plasma slat”, jaircr 2007, pp. 27-44, 2007. [17] x.q. zhang, and l. tian, “three-dimensional simulation of micro air vehicles with low aspect ratio wings”, key engineering materials, vol. 339, pp. 377-381, 2007. [18] p.j. kunz, “aerodynamics and design for ultra low reynolds number flight”, phd thesis. stanford: stanford university, 2003. [19] m. hassanalian, h. khaki, and m. khosrawi, “a new method for design of fixed wing micro air vehicle”, institution of mechanical engineers, part g: journal of aerospace engineering, vol. 229, pp. 837–850, 2014. [20] f. hsiao, c. lin, y. liu, d. wang, c. hsu, and c. chiang, “thickness effect on low-aspect-ratio wing aerodynamic characteristics at a low reynolds number”, journal of mechanics, vol. 24, no. 3, pp. 223-228, 2008. [21] a. mohamed, k. massey, s. watkins, and r. clothier, “the attitude control of fixed-wing mavs in turbulent environments”, progress in aerospace sciences, vol. 66, pp. 37-48, 2014. [22] s. sankaranarayanan, a. roshan, and c. suraj, technology driven cfd study of a micro air vehicle (mav) issn 2180-1053 e-issn 2289-8123 vol.14 no.1 45 programme for the development of a fixed wing micro air vehicle at naldriven, 2008. [23] c. ramprasadh, and v. devanandh, “a cfd study on leading edge wing surface modification of a low aspect ratio flying wing to improve lift performance”, international journal of micro air vehicles, vol. 7, no.3, pp. 361-373, 2015. [24] t. flint, m. jermy, t. new, and w. ho, “computational study of a pitching bio-inspired corrugated airfoil”, international journal of heat and fluid flow, vol. 65, pp. 328-341, 2017. [25] w. shyy, y. lian. j. tang, h. liu, p. trizila, b. stanford, l. bernal, c. cesnik, and p. ifju, “computational aerodynamics of low reynolds number plunging, pitching and flexible wings for mav applications”, acta mech sin, vol. 24, pp. 351–373, 2008. 4faculty of mechanical and manufacturing engineering technology, universiti teknikal malaysia melaka, hang tuah jaya, 76100 durian tunggal, melaka, malaysia. 1.0 introduction 2.0 methodology 3.0 results and discussions 3.1 validation 3.2 fluid dynamic behavior 3.3 the impact of wind disturbances 4.0 concl u s ion acknowledgments references preparation of papers in a two column model paper format issn 2180-1053 e-issn 2289-8123 vol.14 no.2 1 corroded gas pipeline remaining life under variable operating pressure m.h. belkacemi1, d. benzerga2*, a.choutier3, a. haddi4 1,2,3lscmi, mechanical department, university of sciences and technology of oran, , b.p. 1505, 31000 oran, algeria 4 university of artois, ea 4515, laboratoire de génie civil et géo-environnement, béthune f62400, france *corresponding’s author email: djeb_benz@yahoo.fr article history: received 18 january 2022; revised 18 february 2023; accepted 06 march 2023 abstract: gas pipelines are subjected to mechanical and chemical stresses which lead to failures of various types such as corrosion, cracking, deformation and rupture. corrosion damage to pipelines has become a growing concern in the gas industry. corrosion defects in the form of pitting caused by the corrosion phenomenon cause high concentrations of stresses and plastic strains thus reducing the strength of the pipe by threatening its structural integrity. indeed, the internal operating pressure is variable and can generate the phenomenon of fatigue, which is dangerous, given its insidious nature, causing damage to the corroded zone for stress levels well below the yield stress of the material. the standards used in the framework of the rehabilitation of corroded pipes allow the determination of their burst pressure but not their remaining life. to address this issue, we have developed a model based upon damage mechanics to predict the remaining life of a pipe in the presence of an external corrosion defect. keywords : corrosion, gas pipeline, remaining life, variable operating pressure, fatigue, damage mechanics, finite element. 1.0 introduction pipeline transportation is of global importance in the oil and gas industries. it is an evolving system around the world representing appropriate solutions for reliable transport. indeed, over the past 50 years, pipelines have become the cheapest and safest way to transport large amounts of energy and over long distances [1]. today, under strong economic pressure, the operating life of structures is often extended, under service conditions which can sometimes be more severe than those foreseen in the design. gas pipelines are exposed to mechanical and chemical stresses leading to failures of various types such as corrosion, cracking, deformation, and rupture. the deterioration of underground pipelines by the phenomenon of corrosion has become a growing concern in the gas and environmental sectors. european gas pipeline incident group has published that among 1060 cases of rupture in pipelines, 15.3% were caused by the phenomenon corrosion [2]. the losses of metal in the form of pitting caused by the phenomenon of corrosion, cause concentrations of stresses and significant plastic deformations in the vicinity of the corrosion defects thus reducing the strength of the pipe by threatening its structural integrity and causing its rupture [3-5]. the rupture of the corroded pipe also depends on the variation of the service pressure of the gas which causes fatigue of the structure [6-7]. the stress and strain field as well as the burst journal of mechanical engineering and technology (jmet) 2 issn 2180-1053 e-issn 2289-8123 vol.14 no.2 pressure closely depend on the geometry of the corrosion defect [8]. burst is assumed to occur when the operating pressure is greater than the pressure predicted by the various standards used for the evaluation of the corroded pipe burst pressure [9-10]. the periodic inspections by intelligent tool not only allow the localization of corrosion defects but also their dimensions, their nature and the severity of their danger [11]. managers in gas industry are faced with a major problem given the large number of accidents occurring in gas and oil pipeline networks and the huge rehabilitation budgets that result [12]. the pipes affected by corrosion must be renovated by recalculating the maximum allowable operating pressure (maop) according to the geometry of the corrosion defect or by replacing the sections affected by corrosion or by installing new lines to meet demand new consumers or to increase the reliability of the pipeline system [13]. short-term increases in demand and / or reductions in supply may cause variations in operating pressure, especially during winter. these pressure variations may cause a phenomenon of fatigue which can be dangerous in the vicinity of corrosion defects because of its insidious nature. the current assessment procedures used in the gas pipeline industry for the rehabilitation of corroded pipes (b31g, modified b31g, restrengh, dnv, shell, pcorr, fitnet ffs and the approaches of choi and cronin) allow the determination of burst pressure [14-15]. however, the disadvantage of these standards is that they can't predict the remaining life of corroded pipe when the operating pressure is variable. the main motivation for this work is to respond to the concern of algerian gas pipeline industry leaders to determine the remaining life of a corroded pipeline subjected to variable operating pressure. the study and the analysis of the damage in the vicinity of a corrosion defect of an x65 steel pipe, were carried out with the aim to develop a numerical simulation tool allowing the determination of the remaining life of a superficially corroded pipe. 2.0 modelling manuscript should content a title, list of authors, abstract, body, conclusions, acknowledgement (if necessary) and references. in this work, we consider that the damage is very localized in the vicinity of the corrosion defect where the plastic deformations are important so that the damage of the material occurs only in a small volume. this is due to the high sensitivity of damage to stress concentrations at the macro scale. this allows us to consider that the effect of damage on the state of stress and strain only occurs in very small damaged areas. the coupling between the damage and the strains can be neglected in the whole of the pipe except in the corrosion defect where the damage develops [16-17]. the locally coupled analysis is well suited for fatigue damage cases since the behavior of the corroded pipe remains elastic everywhere except in the vicinity of the corrosion defect [18]. the critical zone where the equivalent stress is maximum (figure 1) constitutes the bridge between computation by finite elements and the post processor.      ***** mmmaxm   the elastoplasticity calculation by the ansys code allows the determination of the strain and stress fields then the local analysis will deal with the elastoplastic constitutive laws coupled with the damage at the level of the critical zone only. corroded gas pipeline remaining life under variable operating pressure issn 2180-1053 e-issn 2289-8123 vol.14 no.1 3 taking into account the symmetry of the corroded pipe, a quarter of the pipe was modeled using code ansys [19] with a parabolic defect taking into account its geometry and the boundary conditions. the stress-number of elements curve allows the best meshing choice of the structure [20]. figure 1: locally coupled approach of the damage initiation. a finite element analysis (fea) was performed using ansys simulation software for the determination of the critical zone (m*) in the vicinity of the corrosion defect where damage of the material develops giving rise to microcracks whose propagation will lead to failure then the bursting of the pipe [21]. a post processor based upon the damage mechanics using coupled damage strain constitutive equations and introducing a variable of continuous isotropic damage [22-23], allows the computation of the conditions of crack initiation from the history of the strain components of the critical zone (m*) taken as the output of the finite elements. s s d d    ; 10  d (1) the introduction of the effective stress notion allows to: d1 σ σ  ~ (2)         2 2 1* ** 2131 3 2                 eq h v veq r rwith supm (3) rv: is the triaxiality function which depends on the triaxiality coefficient σh /σeq. in the majority of the cases, this criterion is satisfied in the zones with high concentration of stresses with a high value of coefficient of triaxiality σh / σeq. journal of mechanical engineering and technology (jmet) 4 issn 2180-1053 e-issn 2289-8123 vol.14 no.2 the resolution of the constitutive equations below in incremental form by the numerical method of newton, allow the determination of the evolution of the damage of the material in the critical zone within the corrosion defect [24]: pppr e e eee if es ffifp dede v eq eq d ijp ij ij kkije ij p ij e ijij d 0 2 2 0 2 3 11 1 ~                (4) where 0~  seq f  is the plastic yield function, s is the threshold of plasticity (the condition sequ   deviates any plastic strain and ensures a pure elastic strain) and p is the accumulated plastic strain. the method used for solving the above constitutive equations is integration schemes such as the radial return method [25]. it is assumed in a first that all variables of the model are known at the initial time (tn) and that the behavior is purely elastic.  p n tr   21~ (5) λ and μ are the lamé coefficients and 1 is the identity tensor of order 2. all other "plastic" variables are equal to their values at time (tn). if this "elastic predictor" satisfies the condition of the load function f ≤ 0, the assumption is then valid, and the calculation procedure for this time increment is completed. in the contrary case f > 0, this elastic state is "corrected" to find the plastic solution. the method developed above has been implemented in the ansys commercial code and the post-processor. it will use as data, the parameters of the material and the components of the total deformations. as result, it will give, a function of the internal pressure in the corroded pipe, the damage value, the accumulated plastic strain and the stress components at each step, until initiation of macroscopic cracks in the critical zone. this will permit to determine the remaining life that the corroded pipe can withstand as a function of the variation of inner pressure. 3.0 numerical results the study was carried out on pipes of external diameter of 40'' and a thickness of 12.7 mm in steel x65 whose mechanical properties are given in the table 1[26]. corroded gas pipeline remaining life under variable operating pressure issn 2180-1053 e-issn 2289-8123 vol.14 no.1 5 table 1: mechanical properties of x65 steel young’s modulus poisson’s ratio yield strength tensile strength 210.7 (gpa) 0.3 464.5 (mpa) 563.8 (mpa) for numerical simulation allowing the determination of the remaining life of corroded pipes. each pipe with its corresponding corrosion defect (extension, width and depth) was subjected to variable operating pressure as function of time. table 2 below shows the corroded pipes selected for numerical simulation. table 2: various parameters of the corrosion defect (extension, width and depth representing a percentage of the pipe wall thickness) the internal pressure of the gas fluctuates as a function of the gas demand (see figure 2), therefore undergoes variations which cause the phenomenon of fatigue. fatigue is an insidious phenomenon due to its hidden nature which can cause fractures for stress levels below the yield strength. during service the pipe can be damaged this is explained by the fact that the phenomenon of fatigue is a phenomenon characterized by a strong micro plasticity in the vicinity of micro defects (micro voids, inclusions, precipitates) that are potential sources of damage [15]. figure 2: variation in service pressure as function of time defect extension in mm defect width in mm c=5 c=10 c=15 c=20 c=25 20 20% 20% 20% 20% 20% 40 30% 30% 30% 30% 30% 60 40% 40% 40% 40% 40% 80 50% 50% 50% 50% 50% 100 60% 60% 60% 60% 60% journal of mechanical engineering and technology (jmet) 6 issn 2180-1053 e-issn 2289-8123 vol.14 no.2 the method developed above gives as function of the internal operating pressure, the value of the damage, and the accumulated plastic strain and stress components at each instant until a macroscopic crack initiation in the corrosion defect. this allows determining the maximum remaining life that a corroded pipe could support before rupture. the maximum remaining life is the number of years corresponding to the critical value of the damage dc (corresponding to crack initiation). in figure 3, histogram shows the damage as a function of years’ number for retained corrosion defects. it is noted that the damage is initiated in the material for a number of years equal to y0 (microscopic crack initiation) and the microcrack propagates until a macroscopic crack is obtained for a number of cycles equal to yr. figure 3 also highlights the corrosion defect geometry effect (extension, width and depth) on the corroded pipe remaining life. the calculation of the remaining life of corroded pipes is summarized in the table 3. table 3: the remaining life of corroded pipes in cycles, days and years. corrosion defect cycles days years width, extension, depth number of cycles at crack initiation (n0) number of cycles at rupture (nr) number of days at crack initiation (d0) number of days at rupture (dr) number of years at crack initiation of (y0) number of days at rupture (yr) def 1 5 20 20% 1,10e+02 9,30e+02 1,10e+03 9,30e+03 3,01e+00 2,55e+01 def 2 5 40 30% 9,70e+01 5,59e+02 9,70e+02 5,59e+03 2,66e+00 1,53e+01 def 3 5 60 40% 8,53e+01 4,98e+02 8,53e+02 4,98e+03 2,34e+00 1,36e+01 def 4 5 80 50% 8,23e+01 3,61e+02 8,23e+02 3,61e+03 2,25e+00 9,89e+00 def 5 5 100 60% 5,56e+01 2,67e+02 5,56e+02 2,67e+03 1,52e+00 7,32e+00 def 6 10 20 20% 1,32e+02 9,12e+02 1,32e+03 9,12e+03 3,62e+00 2,50e+01 def 7 10 40 30% 1,01e+02 5,41e+02 1,01e+03 5,41e+03 2,77e+00 1,48e+01 def 8 10 60 40% 7,98e+01 4,89e+02 7,98e+02 4,89e+03 2,19e+00 1,34e+01 def 9 10 80 50% 7,33e+01 3,76e+02 7,33e+02 3,76e+03 2,01e+00 1,03e+01 def 10 10 100 60% 6,74e+01 3,21e+02 6,74e+02 3,21e+03 1,85e+00 8,79e+00 def 11 15 20 20% 1,23e+02 8,95e+02 1,23e+03 8,95e+03 3,37e+00 2,45e+01 def 12 15 40 30% 1,01e+02 6,32e+02 1,01e+03 6,32e+03 2,77e+00 1,73e+01 def 13 15 60 40% 8,97e+01 5,29e+02 8,97e+02 5,29e+03 2,46e+00 1,45e+01 def 14 15 80 50% 7,46e+01 4,16e+02 7,46e+02 4,16e+03 2,04e+00 1,14e+01 def 15 15 100 60% 6,66e+01 3,49e+02 6,66e+02 3,49e+03 1,82e+00 9,56e+00 def 16 20 20 20% 9,70e+01 8,14e+02 9,70e+02 8,14e+03 2,66e+00 2,23e+01 def 17 20 40 30% 8,80e+01 6,45e+02 8,80e+02 6,45e+03 2,41e+00 1,77e+01 def 18 20 60 40% 7,90e+01 5,75e+02 7,90e+02 5,75e+03 2,16e+00 1,58e+01 def 19 20 80 50% 7,21e+01 4,68e+02 7,21e+02 4,68e+03 1,98e+00 1,28e+01 def 20 20 100 60% 6,37e+01 3,65e+02 6,37e+02 3,65e+03 1,75e+00 1,00e+01 def 21 25 20 20% 8,90e+01 7,97e+02 8,90e+02 7,97e+03 2,44e+00 2,18e+01 def 22 25 40 30% 7,82e+01 6,25e+02 7,82e+02 6,25e+03 2,14e+00 1,71e+01 def 23 25 60 40% 6,98e+01 4,85e+02 6,98e+02 4,85e+03 1,91e+00 1,33e+01 def 24 25 80 50% 6,28e+01 3,55e+02 6,28e+02 3,55e+03 1,72e+00 9,73e+00 def 25 25 100 60% 5,35e+01 2,98e+02 5,35e+02 2,98e+03 1,47e+00 8,16e+00 corroded gas pipeline remaining life under variable operating pressure issn 2180-1053 e-issn 2289-8123 vol.14 no.1 7 figure 3: remaining life as function of corrosion defect geometry the histogram groups the remaining life results obtained in number of years for each corrosion defect. investigated pipeline has an initial of years y0 whose damage appeared (crack initiation). on the other hand, the pipeline considered can go up to a yr number of years (the maximum remaining life) related to the damage causing a rupture. it can be seen that the life y0 at crack initiation or yr at failure decreases as the extension and the depth of the defect increase while keeping a constant width. it is also noted that the remaining life decrease, when the width of the defect becomes important, this can be explained by the increase in the loss of metal and the increase in the volume of the corrosion defect. the remaining life of a corroded pipe as a function of the geometry of the corrosion defect represents a three-dimensional surface, figure 4. the maximum life that a pipe with a certain geometry of the corrosion defect could withstand is located in the zone safety limited by the 3d surface corresponding to the rupture of the corroded zone. the three-dimensional surface can be interpreted differently by determining, for a given residual life, the size of the critical defect. the zone located under the three-dimensional surface is the range of variation of the remaining life not to be exceeded for a critical size of the defect. indeed, for a corrosion defect of width c, depth t and extension l, the life of which is below the curve, this defect does not cause the rupture of the pipe. in figure 4, we have represented only the two cases where the width of the defect has the smallest value c = 5 mm and the largest value c = 25 mm to highlight the effect of the width on the remaining service life. journal of mechanical engineering and technology (jmet) 8 issn 2180-1053 e-issn 2289-8123 vol.14 no.2 remaining life in years for width c = 5mm remaining life in years for width c = 25mm figure 4: three-dimensional representation of the remaining 4.0 conclusion in this study, an approach for predicting the remaining life and assessing the integrity of corroded gas pipelines subjected to cyclic pressure was investigated. cyclic pressure has a greater influence on the life of corroded gas pipelines. the proposed approach uses locally coupled analysis which is most appropriate for fatigue failure cases when the structure remains elastic everywhere except in the critical point. the determination of the critical point is the bridge between computation fem and the post processor. this method is much simpler and saves a lot of computer time compared to the fully coupled analysis which takes into account the coupling between damage and strain in the whole structure. the variation in the corrosion defect geometry was considered to calculate the cyclic damage at the critical point. due to the presence of non-linearity in the formulas, an iterative method was used to calculate the remaining life. the described method provides a basis for determining the priority of inspecting corroded pipelines and, ultimately, for developing a renewal strategy. it is a real decision-making tool for those in charge who manage the transport of gas by pipelines. references [1] a. g. rafael, s.s. mauricio, b. a. emilio, s. franck, and m. felipe, “reliability assessments of corroded pipelines based on internal pressure”, engineering failure analysis, vol. 98, pp. 190-214, 2009. [2] zhao, y., mingliang, l., min s. and kingie, j. “research on leakage detection and analysis of leakage point in the gas pipeline system”, open journal of safety science and technology ,vol.11, no.3, 2011. [3] zhong-ying h., xiao-guang h, ’’stress corrosion behavior of x80 pipeline steel in the natural seawater with different dissolved oxygen contents’’, frattura ed integrità strutturale vol. 50, pp. 21-28, 2019. [4] ankang c., nian-zhong c. “corrosion fatigue crack growth modelling for subsea pipeline steels”, ocean engineering, vol. 142, pp.10–19, 2017. https://www.sciencedirect.com/science/journal/13506307/98/supp/c https://www.scirp.org/(s(351jmbntvnsjt1aadkposzje))/journal/paperinformation.aspx?paperid=16527 https://www.scirp.org/(s(351jmbntvnsjt1aadkposzje))/journal/paperinformation.aspx?paperid=16527 https://www.scirp.org/(s(351jmbntvnsjt1aadkposzje))/journal/home.aspx?journalid=600 https://www.scirp.org/(s(351jmbntvnsjt1aadkposzje))/journal/home.aspx?journalid=600 https://www.scirp.org/(s(351jmbntvnsjt1aadkposzje))/journal/home.aspx?issueid=1317 corroded gas pipeline remaining life under variable operating pressure issn 2180-1053 e-issn 2289-8123 vol.14 no.1 9 [5] dmytrakh i.m., leshchak r.l., syrotyuk a.m. “influence of sodium nitrite concentration in aqueous corrosion solution on fatigue crack growth in carbon pipeline steel”, international journal of fatigue, 128, 105192, 2019. [6] chinedu i. o., brian b., ian j. d. “pipeline failures in corrosive environments – a conceptual analysis of trends and effects”, engineering failure analysis, vol. 53, pp.36–58, 2015. [7] mansor n.i.i., abdullah s., ariffin a.k., syarif j. “a review of the fatigue failure mechanism of metallic materials under a corroded environment”, engineering failure analysis, vol. 42, pp. 353–365, 2014. [8] xu, l.y., cheng. y.f. “reliability and failure pressure prediction of various grades of pipeline steel in the presence of corrosion defects and pre-strain”, international journal of pressure vessels and piping, vol. 89, pp. 75-84, 2012. [9] ehsan a., rouzbeh a., vikram g., jonathan b., christopher c., nima k., genserik r. “developing a dynamic model for pitting and corrosion-fatigue damage of subsea pipelines”, ocean engineering, vol. 150, pp. 391–396, 2018. [10] mechab b., medjahdi m., salem m., serier b. “probabilistic elastic-plastic fracture mechanics analysis of propagation of cracks in pipes under internal pressure”, frattura ed integrità strutturale, vol. 54, pp. 202-210, 2020. [11] budhe s., banea m.d., de barros s. “prediction of the burst pressure for defective pipelines using different semi-empirical models”, frattura ed integrità strutturale, vol. 52, pp. 137-147, 2020. [12] zelmati, d., bouledrouab, o., hafsid, z., djukice, m. b. “probabilistic analysis of corroded pipeline under localized corrosion defects based on the intelligent inspection tool”, engineering failure analysis, vol. 115, 2020. [13] tee, k. f., wordu, a. h. “burst strength analysis of pressurized steel pipelines with corrosion and gouge defects”, engineering failure analysis, vol. 108, 2020. [14] zelmati, d., ghelloudj, o., amirat, a. “reliability estimation of pressurized api 5l x70 pipeline steel under longitudinal elliptical corrosion defect”, the international journal of advanced manufacturing technology, vol. 90, pp. 2777–2783, 2017. [15] berrekia, h., benzerga, d., haddi, a. “behavior and damage of a pipe in the presence of a corrosion defect depth of 10% of its thickness and highlighting the weaknesses of the asme/b31g”, frattura ed integrità strutturale, vol. 49, pp. 643654, 2019. [16] lemaitre, j., benallal, a., marquis, d. “lifetime prediction of structures in anisothermal viscoplasticity coupled to damage”, nuclear engineering and design , vol. 133, no. 3, pp. 345-360, 1992. https://www.sciencedirect.com/science/journal/03080161 https://www.sciencedirect.com/science/journal/03080161 https://www.sciencedirect.com/science/journal/03080161/89/supp/c https://www.sciencedirect.com/science/journal/13506307 https://www.sciencedirect.com/science/journal/13506307/115/supp/c https://www.sciencedirect.com/science/journal/13506307 https://www.sciencedirect.com/science/journal/13506307/108/supp/c https://link.springer.com/article/javascript:; https://link.springer.com/article/javascript:; https://link.springer.com/journal/170 https://link.springer.com/journal/170 https://www.sciencedirect.com/science/journal/00295493 https://www.sciencedirect.com/science/journal/00295493/133/3 journal of mechanical engineering and technology (jmet) 10 issn 2180-1053 e-issn 2289-8123 vol.14 no.2 [17] lemaitre, j. micro-mechanics of crack initiation. in: knauss w.g., rosakis a.j. (eds) non-linear fracture. springer, dordrech, 1990. [18] lemaitre, j. a course on damage mechanics. berlin; new york: springer, 1996. [19] regard, a., berrekia, h., benzerga, d., haddi, a. “repair and rehabilitation of corroded hdpe100 pipe using a new hybrid composite”, frattura ed integrità strutturale, vol. 56, pp. 115-122, 2021. [20] benzerga, d. burst pressure estimation of corroded pipeline using damage mechanics. [2015] mmssd, isbn 978-3319-14531-0. doi: 10. 1007/978-3-319-14532-7 springer. [21] chouiter, a., benzerga, d., haddi, a. and tamine, t. “prediction of cycle life of expansion bellows for fixed tube sheet heat exchanger”, frattura ed integrità strutturale, vol. 47, pp.30-38, 2019. [22] lemaitre, j., and chaboche, j. l., mécanique des matériaux solides. dunod, paris, 1985. [23] benzerga, d., haddi, a., lavie, a. “effect of non-polluting and renewable load on delamination of a composite biomechanical material”, journal of mechanical engineering and technology, vol. 6, no. 1, 2014. [24] benallal, a., billardon, r., and doghri, i. “an integration algorithm and the corresponding consistent tangent operator for fully coupled elastoplastic and damage equations”, international journal for numerical methods in biomedical engineering (e4), pp. 731–740, 1998. [25] silva, m., taylor, w. “análise elastoplastica de estruturas metálicas utilizando algorítmos de retorno radial”, revista internacional de métodos numéricos para cálculo y diseño en ingeniería, vol. 20, no.3, 2004. [26] antónio.a. f, abílio m. p., renato n.j. “monotonic and ultra-low-cycle fatigue behaviour of pipeline steels experimental and numerical approaches”, pp. 36-120, 2018. preparation of papers in a two column model paper format issn 2180-1053 e-issn 2289-8123 vol.15 no.1 11 development and evaluation of asbestos-free brake pads produced from costus afer waste and local gum arabic l.c ngwaba1 and d.o. aikhuele2 1,2, department of mechanical engineering university of port harcourt, east west road, port harcourt, nigeria corresponding’s author email: loveth_ngwaba@uniport.edu.ng article history: received 18 january 2022; revised 18 february 2023; accepted 06 march 2023 abstract: this study dealt with the development and evaluation of a new asbestos-free noncarcinogenic brake pads with costus afer waste particle as the base material. three sets of brake pads with different sieve sizes (90, 100 and 200 µm) were developed, through compression molding from a 55% costus afer waste particles, 22% local gum arabic as binder, 5% of rubber seed husk and 5% of walnut shell as fillers, 10% iron filling as frictional addictive, 1% of carbon black as friction modifier, 1% of cobalt nepthanate as catalyst and 1% of methyl ethyl ketone as accelerator. physico-mechanical test was carried out on the costus afer waste-based brake pad with a brinell hardness test value of 103hb, compressive strength of 115mpa and density of 1.3 g/cm3 which is far better than the commercially available brake pad upon comparison. these properties were found to increase with a decrease in particle sizes while the water absorption (1.2%), oil absorption (0.55%), wear rate(1.8mg/m), and flame resistance (charred with 20% ash) increased with increasing particle sizes due to enhanced porosity. the coefficient of friction is approximately 0.3 µm which is within the acceptable standard and the element analysis based on energydispersive x-ray fluorescence for the developed brake pad was carried out. the developed costus afer waste-based brake pad especially with the grain size of 90µm, shows a better wear performance and other properties as compared to the control commercial brake pad. the study can conclude therefore that the costus afer waste-based brake pad stands as a better replacement for the existing commercial asbestos-based brake pads. keywords : non-carcinogenic brake pads; costus afer waste; compression molding; grain size; bush cane 1.0 introduction in recent times, several research efforts in the automobile industry are being channeled to the production of organic brake pads to replace the currently used asbestos-based brake pads that has been found to contain several negative components that affects human health and the environment. asbestos, which has some good engineering properties (e.g., quality strength, wear, and heat resistance) that has made them very suitable for inclusion in brake liners and as filler material up till 1989 [1], are being withdrawn gradually from almost all its applications where there is possibility of man consuming or inhaling its dust, due to its carcinogenic nature. to mitigate the health risk associated with the use of asbestos material for brake pad design, it is necessary therefore, to use ecofriendly materials or organic materials for their development. among the ecofriendly materials that has found application in literature include cashew nut shells and palm kernel shells (pks) [2], coal ash and palm kernel fiber [3], banana peels [4], coconut shells [5], snail shell [6], periwinkle shell [7], canarium sweinfurthii shell [8], cocoa beans shell and maize husk [9], hazelnut powder mailto:loveth_ngwaba@uniport.edu.ng journal of mechanical engineering and technology (jmet) 12 issn 2180-1053 e-issn 2289-8123 vol.15 no.1 [10], sawdust [11], sugar cane bagasse [12], palm kernel shell and cow bone [13], palm ash [14], bamboo fiber [15], walnut shell powder [16], coconut shell powder and palm kernel shell [17], miscanthus [18] , recycled clay tiles [19], groundnut shell [20], and kenaf fiber [21]. in other application of ecofriendly and organic materials for brake pad design and development, ossia and big-alabo, [24] investigated the development and characterization of green brake pads using waste shells of giant african snails (achatina achatina l) and resin. their result showed that an increase in the sample particle size leads to a decrease in density, brinell hardness, and compressive strength of the snail shell (ss) brake pads. this followed a negative index power-law model after the order of the hall-petch equation, whereas the absorption features increased with an increase in particle size, and the model followed a positive index power law because of the pores in the matrix. their result also shows that the snail shell-based brake pad has a better frictional grip at the rubbing interfaces than the commercial brake pad sample. also, the density, brinell hardness, and compressive strength of the snail shell brake pad were better or superior to the commercial sample used while the wear rates of the commercial brake pad were superior to that of the developed brake pad, and they suggested that, the result can be improved by adding more iron fillings to improve the thermal conductivity and the wear features. resins which is very toxic was used by the author for the brake pad design and has been utilized by several researchers as binders in the development of brake pads with either organic waste [2,6,10-11,22-24] or inorganic waste [25]. hence, in this paper the authors seek to replace the toxic resins binder with a new ecofriendly and easily available material and to address the wear performance issues. a locally sourced material, gum arabic has been proposed as a replacement for the resins, while costus afer waste as a replacement for the asbestos material. the remaining part of the paper is organized as follows, in section 2, the methodology used in the design and development of the brake pad is presented. this is followed closely by the results and discussion in section 3, while some concluding remarks are presented in section 4. 2.0 methodology 2.1 materials the reinforcing fibers or base material for the brake pads development is costus afer waste (bush cane) sourced from obingwa in osisioma ngwa local government area of abia state, nigeria. figure1, shows a typical picture of costus afer leaf and costus afer waste. the following components were used in formulating the caw-brake pads matrix; costus afer waste (fiber) as the base material (55%), rubber seed husk and walnut shell (fillers, 5% each), local gum arabic as binder (22%), iron filling as frictional addictive (10%), carbon black as friction modifier (1%), cobalt nepthanate as catalyst (1%) and methyl ethyl ketone as accelerator (1%). other components and equipment used include the following: sae 40 oil (engine oil), distilled water, brake pad mold fabricated from carbon steel plate (grade: a36), vernier caliper, weighing balance, hardness tester, crushing machine, sieves of different sizes, milling machine, electric oven, universal wear machine and compression machine. development and evaluation of asbestos -free brake pads produced from costus afer waste and local gum arabic issn 2180-1053 e-issn 2289-8123 vol.15 no.1 13 figure 1: diagram of (a) costus afer leaf, (a) costus afer waste and (c) grounded costus afer waste 2.2 methods the method adopted in the development of this caw brake pad and its performance evaluation is similar to the one presented by ossia and alabo, [6], where they develop an asbestos-free and non-carcinogenic brake pads from waste shells of giant african snails. in this paper however, a different set of material types has been used. the materials and their composition are discussed in table 1, while the steps required for development and production of the brake pad are presented in figure 2. table 1: material compositions item material composition (wt%) 1 costus afer waste (fiber) 55.0 2 rubber seed husk (filler) 5.0 3 walnut shell (filler) 5.0 4 local gum arabic (binder) 22.0 5 iron filling (frictional addictive) 10.0 6 carbon black (friction modifier) 1.0 7 cobalt nepthanate (catalyst) 1.0 8 methyl ethyl ketone (accelerator) 1.0 (a) (b) (c) journal of mechanical engineering and technology (jmet) 14 issn 2180-1053 e-issn 2289-8123 vol.15 no.1 figure 2: the process of development of the caw –brake pad. 2.3 performance evaluation in the testing and evaluation of the feasibility and usability of the newly developed caw-brake pad, the following physico-mechanical test has been carried out for its performance evaluation. development and evaluation of asbestos -free brake pads produced from costus afer waste and local gum arabic issn 2180-1053 e-issn 2289-8123 vol.15 no.1 15 2.3.1 absorption test the water and oil absorption test helps to determine the effect of the absorbed water and oil in its dimensions. the specimen is oven-dried at 25 °c for 3 hours and its initial weight is measured by weighing balance. subsequently, the dimensions (thickness) of the specimen were measured using a vernier caliper, after twenty-four hours of submersion in water and engine oil (sea 40) at 30oc the specimen was weighed after the excess water and oil had drained off. the percentage absorption of caw-brake pad samples in oil and water was determined using equation (1) [22]. absorption (%) = 𝑊𝑓−𝑊𝑏 𝑊𝑏 × 100 1 (1) where 𝑊𝑓 is the weight of the sample after immersion into a given absorbent media, and 𝑊𝑏 is the weight of the sample before immersion into a given absorbent media. 2.3.2 density test the density measurement on the caw-brake pad sample was carried out using archimedes principles as per equation (2). the buoyant force on a submerged object is equal to the weight of the fluid displaced. this principle is useful for determining the volume and therefore the density of an irregularly shaped object by measuring its mass in air and its effective mass when submerged in water. this effective mass underwater was its actual mass minus the mass of the fluid displaced. the difference between the real and the effective mass, therefore, gives the mass of water displaced and allows the calculation of the volume of the irregular-shaped object. the mass divided by the volume thus gives a measure of the average density of the sample. hence, determining the density of the caw-brake pad samples [26]. density (ρ) = mass(m) volume (v) = m0 m0−m1 g/cm3 (2) where m0 is the mass of sample in air, and m1 is the mass of the sample in water. other test types carried out on the newly developed brake pad include, hardness test, (as per astm e10) compressive strength test (as per bs en 12390 part 9), wear test (as per astm g99 and sae 1661.), flame resistance test and energy-dispersive x-ray fluorescence (edxrf) test. 3.0 results and discussion as stated above, the physico-mechanical test carried out for the performance evaluation of the newly developed caw-based brake pad include, oil and water absorption test, density test, and mechanical property test that include brinell hardness test, compressive strength, flame resistant test, and wear test. these performance evaluation tests have been discussed and compared with existing brake pad in this section. results from the water and oil absorption test for the caw-based brake pad shows that the brake pad samples increased with an increasing grain size and vice versa, this results which are compare with sample from a commercial brake pad (cbp) in the market are journal of mechanical engineering and technology (jmet) 16 issn 2180-1053 e-issn 2289-8123 vol.15 no.1 shown in figure 3. this can be attributed to the increase in pores as the sieve size increases. furthermore, the improved bonding between the smaller grain (90µm) sizes and the binder could be responsible for this. for the oil absorption test for the cawbased brake pad, the samples results can be attributed to improved bonding between the smaller grain size and the binder. the sample size90µm showed a better property and will do better where there is oil leakage from the hydraulic system. these results are in agreement with the findings of earlier researchers as shown in ibahadode and dagwa, yawas et al., akincioglu et al. and aigbodion et al work [2, 7, 10 and 12]. for instance, yawas et al. [7] studied the microstructures of brake pads made of periwinkle shell show that the surface morphology of the brake pad became increasingly homogenous with the decrease in the sieve sizes of periwinkle. this further explains the increase in the pores as the costus afer particle sizes increases. this means that as the sieve size increases, this will lead to increase in the grain size of the costus afar particles and this increases the pores. the increase in porosity with grain size causes the absorption to increase. also the interfacial bonding between costus afer particles and the binder also affects the porosity of the samples, proper bonding will be achieved when decreasing the sieve size from 200 to 90 µm. figure 3: comparison of water and oil absorption of the brake pads produced with varying sample sizes of the costus afer waste with cbp. results from the density test for the caw-based brake pad samples shows that the samples progressively increased with decreasing particle size of the caw from 200 µm to 90 µm and this have been shown in figure 4. the decrease in density can be attributed to the increase in pore size due to increased particle size. the least sample size90 µm has the highest density (1.3 g/cm3) which is due to close packing of caw particle with binder and other filler materials creating more homogeneity in the entire phase of the composite body. other researchers had earlier reported similar results [6-7, 12]. conversely, the study by olabisi et al. [29] reported an irregular change in brake pad density of based-brake pad with increase in the percentages of base materials (cocoa bean shell, maize husk and palm kernel shell). according to edokpia et al. [26] the increase in the density of brake pad is as a result of the increased packing of the filler particles forming more homogeneity in the entire phase of the brake pad composite body. in this regard, a certain composition of the composite will give the blend with the most acceptable density, but this can only be 1.5 2.7 4.5 0.9 0.6 1 1.3 0.3 0.0 1.0 2.0 3.0 4.0 5.0 90 100 200 cbp % a b so rp ti o n caw-sample sizes absorption in water (%) absorption in oil (%) development and evaluation of asbestos -free brake pads produced from costus afer waste and local gum arabic issn 2180-1053 e-issn 2289-8123 vol.15 no.1 17 achieved by optimization of the constitutes. however, the densities of all caw-based brake pad samples were lower than the commercial brake pad (cbp) density (1.89 g/cm3), this makes the caw-based brake pad lighter and contributes to mass reduction of the automotive braking assembly. figure 4: comparison of density of brake pads produced with varying sample sizes of the costus afer waste with cbp. the hardness test value of the caw-based sample brake pad varies with increase in grain size as shown in figure 5. the 90 µm sample which has the highest hardness value with 103 hb, is higher than the hardness test value of the cbp sample which is 101 hb, as well as that of palm kernel shell which is 92 hb and bagasse which is 100.5 hb [12]. however, it is less than the value obtained by yawas et al., [7] for periwinkles shellbased brake pad which is 116.7 hb, the hardness value however is decreases sharply with an increase in the grain size. the hardness value increase of the caw-based sample brake pad with the 90 µm sample was as a result of increase in surface area of the caw-particles which is due to increase in its bonding ability with the binder. the hardness value compares reasonably with results from previous studies presented in yawas et al., aigbodion et al. and ossia et al. work [7, 12, 24]. figure 5: comparison of hardness value of various sample sizes of the costus afer waste particlesbased brake pads with cbp. 1.3 1.1 0.9 1.89 0 0.3 0.6 0.9 1.2 1.5 1.8 2.1 90 100 200 cbp a v g . d e n si ty ( g /c m 3 ) caw-sample sizes (µm) 103 101 95 101 90 92 94 96 98 100 102 104 90 100 200 cbp h a rd n e ss v a lu e ( h b ) caw-sample sizes (µm) journal of mechanical engineering and technology (jmet) 18 issn 2180-1053 e-issn 2289-8123 vol.15 no.1 results for the compressive strength test increases with decrease in the caw-sample grain size as shown in figure 6. the caw-sample size of 90 µm had the highest compressive strength of 115 mpa which was greater than that of the control commercial brake pad (110 mpa). the decrease in compressive strength of the caw-samples as the grain sizes increases can be attributed to decrease in surface area of the caw particles in the binder. therefore, the compressive strength of the caw-brake pad samples increases as the cawparticles decreases. similar result was obtained by olabisi et al. [29], here, they noted that increase in the percentage by weight of the epoxy resin decreased the compressive strength of the brake pad. their study opined that the decrease in the compressive strength of the brake pad was as result of interference of particle mobility or deformability of the matrix. this interference according to ademoh and olabisi. [9] and adeyemi et al. [27] was created through the physical interaction and immobilization of the binder by the presence of mechanical restraints, thereby reducing the strength of the brake pad. brakes are exposed to continuous compressive force during braking and this result shows that sample size 90 µm will do well under such condition. this result agrees with other researchers [7,12,24] finding using different materials. results of the wear test shows that, there are decrease in wear rate as the caw-based brake pad sample grain size decreases and this is shown in figure 7. the sample size with 200 µm has been found to have the highest wear rate with 6.47 mg/m, while the sample size with 90 µm has least rate with 1.8 mg/m. these wear rate however are far better as compared to the control cbp that is 3.8 mg/m. this could be attributed however, to the closer packing which has the chances of influencing a stronger bonding of the caw-particles with the binder and other components. this may also be due to the high values of both the hardness number and compressive strength [1]. the frictional coefficient were determined from the wear test carried out. figure 8 shows the frictional coefficient of the caw-samples which increases with an increasing caw-particle sizes. these results corroborate with findings from other researchers as shown in lawal et al. and aigbodion et al. work [11-12]. figure 6: comparison of the compressive strength of brake pad produced varying sizes of costus afer waste with cbp. 115 105 91 110 0 20 40 60 80 100 120 140 90 100 200 cbp c o m p re ss iv e s tr e n g th ( m p a ) caw-sample sizes (µm) development and evaluation of asbestos -free brake pads produced from costus afer waste and local gum arabic issn 2180-1053 e-issn 2289-8123 vol.15 no.1 19 figure 7: comparison of wear rate of brake pads produced with various sample sizes of the costus afer waste particles with commercial brake pads cbp. figure 8: comparison of frictional coefficient of brake pads produced with various sample sizes of the costus afer waste particles with commercial brake pads cbp. the flame resistance of the caw-brake pad samples was tested by placing the sample on wire gauze positioned directly on the blue flame of a bunsen burner. the caw-brake pad sample weight before and after were measured after 10 minutes, and this was used to determine the percentage of flame resistance. results from the flame resistance test for the caw-based brake pad samples shows an increase as the caw-sample grain size increases. this result has been depicted in figure 9 which also compare the results from cbp sample. this can be attributed to the increase in pores as the caw-sample grain sizes increases. sample size 90 µm charred at 20% after 10 minutes. this result also corroborates with previous research by aigbodion et al., [12] and lawal et al. [11]. 1.8 1.89 6.47 3.8 0 2 4 6 8 90 100 200 cbp w e a r r a te (m g /m ) caw-sample sizes (µm) 0.28 0.3 0.52 0.3 0 0.1 0.2 0.3 0.4 0.5 0.6 90 100 200 cbp f ri c ti o n a l c o e ff ic ie n t (µ m ) caw-sample sizes (µm) journal of mechanical engineering and technology (jmet) 20 issn 2180-1053 e-issn 2289-8123 vol.15 no.1 20 30 42 9 0 10 20 30 40 50 90 100 200 cbp f la m e r e si st a n c e ( % ) caw-sample sizes (µm) figure 9: comparison of flame resistance of brake pads produced with various sample sizes of the costus afer waste particles with commercial brake pads cbp. edxrf analysis was carried out using an oxford x-supreme 8000 to determine the chemical composition of the costus afer waste and brake pad. the test was carried out under controlled temperature and humidity conditions of 22.9oc and 52%rh respectively. the result revealed that caw mainly contains semi-metals and non-metals, that is: silicon, manganese, iron, cobalt, nickel, copper, zinc, lead, phosphorus, molybdenum, magnesium and titanium. these elements are equally found in asbestos which suggested that caw can be used as a replacement of asbestos for brake pads. the chemical composition of caw was comprised of mainly iron (fe) (52.78 %), magnesium (mg) (23.22 %) and zinc (zn) (9.95 %). the chemical composition of caw compares reasonably with other agricultural materials previously used for non-asbestos brake pad production [30]. from table 2, it can be observed that, the values for the properties such as compressive strength, hardness, wear rate, coefficient of friction, flame resistance, density and absorption obtained from caw particle size of 90 µm compares favorably with values reported by aigbodio et al. [12] and to replace commercial brake pad with compressive strength -110 mpa, hardness values101, wear rate-3,8 mg/m, coefficient of friction0.3 µm, density-1.89 g/cm3, flame resistance9% of charred ash, absorption in water 0.9 % and absorption in oil -0.3%. the best values obtained using caw particle size of 90 µm are: compressive strength -115 mpa, hardness values103, wear rate-1.8 mg/m, coefficient of friction0.28 µm, density-1.3 g/cm3. it is obvious that these value compare favorably, hence suggesting the promising prospect of replacing asbestos based brake pad with caw based brake pad. the finished caw-based brake pad with 90 µm particle size is shown in figure 10. table 2: summary of result findings compared with existing ones features palm based (aigbodio et al., 2010). bagasse based (aigbodio et al., 2010). asbestos – based (commercial) costus afer waste based (proposed material) compressive strength (mpa) 103.50 105.60 110.00 115.00 brinell hardness value 92.00 100.50 101.00 103.00 specific gravity (g/cm3) 1.65 1.43 1.89 1.30 average wear (mg/m) 4.40 4.20 3.80 1.80 coefficient of friction (µm) 0.42 0.30 0.28 flame resistance after 10 minutes (%) charred with 46% ash charred with 34% ash charred with 9% ash charred with 20% ash development and evaluation of asbestos -free brake pads produced from costus afer waste and local gum arabic issn 2180-1053 e-issn 2289-8123 vol.15 no.1 21 thickness swell in water after 24hrs (%) 5.03 3.48 0.90 1.20 thickness swell in oil (sae 40) after 24hrs (%) 0.44 1.11 0.30 0.55 figure 10: finished caw-based brake pad with 90 µm particle size 4.0 conclusion eco-friendly asbestos-free brake pad from caw base material and local gum arabic binder has been developed by compression molding and curing at 120 0c for 8 hours. the caw-based brake pads with different particle grain sizes (90, 100 and 200 µm) were subjected to several physico-mechanical and tribological performance tests and the results compared with the commercial brake pad used as control. the compressive strength (115mpa), hardness (103) and density (1.3g/cm3) of the developed caw-based brake pad samples were increasing with a decrease in the particle sizes. but the water (1.2%) and oil absorption (0.55%), wear rate (1.8 mg/m) and flame resistance (charred with 20% ash) increased with increasing cawparticle sizes. the cwa brake pad showed a better wear performance than commercial brake pad. the coefficient of friction is approximately 0.3(µm). the caw-brake pad sample with grain size of 90 µm showed the best properties in the entire test performed and the results obtained compared favorably with that of the commercial brake pad, hence the research objective is achieved. also, the element analysis based on energy-dispersive x-ray fluorescence for the caw and brake pad was carried out. the developed brake pad performed well when used in a toyota camry (2006 model). hence, the result of this work showed that caw and local gum arabic can be used as replacements for asbestos and resin binders respectively in the manufacture of brake pads. acknowledgements thanks to engr. dr. sadiq sius lawal, an inventor, and lecturer, mechanical engineering department, federal university of technology minna, niger state, nigeria and engr ekeoma emmanuel for their great support. thanks to the management of everpraise technologies limited port harcourt for financial support and for allowing me access to their machines in the course of this work. also, thanks to the management of doxa int’l engineering services limited for their financial support. omojuanfo victor, journal of mechan ical engineering and te chnology (jmet) 22 issn 2180-1053 e-issn 2289-8123 vol.15 no.1 prayer o. and faith ogbonnaya are appreciated for helping in carrying out some specific tests in the laboratory. references [1] b. dan-asabe, p.b. madakson, j. manji, a.e. anglais, and f. french. “material selection and production of a cold-worked composite brake pad”,world of engineering and pure and applied sci, vol. 2, no .3 , pp 92–97, 2012. [2] a.o. a ibhadode, and i.m dagwa, “development of asbestos-free friction lining material from palm kernel shell”., journal of brazilian society of mechanical sciences and engineering, vol. 30, no.2, pp 166–173, 2008. [3] c.h. achebe, j.l chukwuneke, f.a anene, and c.m. ewulonu, “a retrofit for asbestosbased brake pad employing palm kernel fiber as the base filler material”. journal of materials design and applications, vol. 233, no. 9, pp 1906–1913, 2019. [4] u. d idris, and v.s aigbodion, “eco-friendly asbestos free brake-pad: using banana peels”. journal of king saud university engineering sciences, vol. 27, pp 185-192, 2015. [5] a. kholil, s.t dwiyati, j.p siregar, riyadi, and sulaiman. “development brake pad from composites of coconut fiber, wood powder and cow bone for electric motorcycle”. international journal of scientific and technology research, vol. 9, no. 2, pp 2938–2942, 2020. [6] c. v ossia and a. big-alabo, “development and characterization of green automotive brake pads from waste shells of giant african snail (achatina achatina l.)”, the international journal of advanced manufacturing technology, vol. 114, pp 2887-2897, 2021. [7] d. s yawas, s.y aku, and s.g amaren , “morphology and properties of periwinkle shell asbestos-free brake pad” , journal of king saud university – engineering sciences, vol. 28, no. 1, pp 103–109, 2016. [8] o. usman, “tribological properties of canarium schweinfurthii shells as frictional material for automotive brake system”, journal of science technology and education, vol. 8, no. 4, pp284–294, 2020. [9] n.a. ademoh, and a.i olabisi, “development and evaluation of maize husks (asbestosfree) based brake pad”, industrial engineering letters, vol. 5, no 2, pp 67–80, 2015. [10] g. akıncıoğlu, s. akıncıoğlu, h. öktem, and i. uygur, “experimental investigation on the friction characteristics of hazelnut powder reinforced brake pad” reports in mechanical engineering, vol. 2, no.1, pp 23–30, 2021. [11] s. s lawal, n.a ademoh, k.c bala, and a.s abdulrahman, “reviews in automobile brake pads production and prospects of agro base composites of cashew nut shells and nigerian gum arabic binder”, covenant journal of engineering technology, vol. 3, no. 2, 2019. [12] v. s aigbodion, u. akadike, s.b hassan, f. asuke, and j.o agunsoye , “development of asbestos free brake pad using bagasse” ,tribology in industry, vol. 32, no.1, pp 12–18, 2010. development and evaluation of asbestos -free brake pads produ ced fr om cos tus afer waste and l ocal gu m arabic issn 2180-1053 e-issn 2289-8123 vol.15 no.1 23 [13] a. mayowa, o.k abubakre, s.a lawal, and r. abdulkabir, “experimental investigation of palm kernel shell and cow bone reinforced polymer composites for brake pad production”, international journal of chemistry and material research, vol. 3, no. 2, pp 27– 40, 2015. [14] c.m ruzaidi, h. kamarudin, j.b shamsul, a.m. bakri, and a. alida, “morphology and wear properties of palm ash and pcb waste brake pad”, international conference on asia agriculture and animal, 13, 145–149, 2011. [15] y. ma, s. shen, j. tong, w. ye, y. yang, and j zhou, “effects of bamboo fibers on friction performance of friction materials”, journal of thermoplastic composite materials, vol. 26, no. 6, pp 845–859, 2013. [16] s. qi, z. fu, z., r. yun, s. jiang, x. zheng, y. lu, v. matejka, j. kukutschova, v. peknikova, and m. prikasky. (2014). “effects of walnut shells on friction and wear performance of eco-friendly brake friction composites”, proceedings of the institution of mechanical engineers, part j: journal of engineering tribology, vol. 228, no. 5, pp 511–520, 2014. [17] d. egeonu, c. oluah, and p.n okolo, “production of eco-friendly brake pad using raw materials sourced locally in nsukka”, journal of energy technologies and policy, vol., 5, pp 1–8, 2015. [18] m. unaldi and r. kus, “the effect of the brake pad components to some physical properties of the ecological brake pad samples”. iop conference series: materials science and engineering, 2017. [19] o.a. johnson, a. b sefiu, b. adeola, a.o kayode, and a. isaac, “recycled ceramic tile composite for automobile applications, a comparative study with nissan jeep cherokee brake pad”, engineering and applied science research, vol. 45, no.3, pp 180-187, 2018. [20] w. c solomon, m.t lilly, and j.i. sodiki, “production of asbestos-free brake pad using groundnut shell as filler material”. international journal of science and engineering invention, vol. 4, no. 12, pp 21–27, 2018. [21] z. u elakhame, y.l. shuaib-babata, s.o jimoh, l.k. bankole and i. o ambali,“ production and characterization of asbestos free brake pads from kenaf fiber composite”. adeleke university journal of engineering and technology, vol. 3, no. 1, pp 69–78, 2020. [22] c, v ossia, a. big-alabo, e.o ekpruke, “effect of grain size on the physicomechanical properties”. advances in manufacturing science and technology, vol. 44, no. 4, pp 135–144, 2020. [23] j. abutu, s.a lawal, m.b ndaliman, r.a lafia-araga, o. adedipe, and i.a. choudhury “production and characterization of brake pad developed from coconut shell reinforcement material using central composite design”, applied sciences, vol. 1, no.1, 2019. [24] m. gürü, m. atar, and r. yildirim, “production of polymer matrix composite particleboard from walnut shell and improvement of its requirements”, materials and design, vol. 29, no. 1, pp 284–287, 2008. journal of mechanical engineering and technology (jmet) 24 issn 2180-1053 e-issn 2289-8123 vol.15 no.1 [25] e.a.m said, n. elzayady, r.i. el-soeudy, and a.b omar, “manufacturing and development of low-cost asbestos-free brake pad composite material”, journal of the egyptian society of tribology, vol. 16, no.1, pp 55–67, 2019. [26] r. o edokpia, v.s aigbodion, o.b obiorah, c.u atuanya, “evaluation of the properties of ecofriendly brake pad using egg shell particles–gum arabic”. results in physics, 2014. [27] i. o adeyemi, a. a nuhu, and t.e boye, “development of asbestos-free automotive brake pad using ternary agro-waste fillers”, journal of multidisciplinary engineering science and technology, vol 3, no. 7, pp 2458-9403, 2016. [28] s. s lawal, k.c bala, a.t. alegbede, “development and production of brake pad from sawdust composite”, leonardo journal of sciences, vol. 30, pp 47–56, 2017. [29] a. i olabisi, a.n adam, and o.m okechukwu, “development and assessment of composite brake pad using pulverized cocoa beans shells filler”, international journal of materials science and applications, vol 5, pp 66-78, 2016. [30] k. ikpambese, d. gundu, and l. tuleun, “evaluation of palm kernel fibers (pkfs) for production of asbestos-free automotive brake pads”, journal of king saud universityengineering sciences, vol. 28, pp 110-118, 2016. issn: 2180-1053 vol. 10 no.1 january – june 2018 1 entropy generation of pseudo-plastic nonnewtonian nanofluids in circular duct under constant wall temperature a. falahat1, m. shabani2*, m. r. saffarian3 1department of mechanical engineering, shahid chamran university of ahvaz, iran 2production technology research institute (acecr), ahvaz, iran 3department of mechanical engineering, shahid chamran university of ahvaz, iran abstract in this paper the second law analysis of thermodynamic irreversibilities in pseudo-plastic non-newtonian nanofluids through a circular duct under uniform wall temperature thermal boundary have been carried out for laminar flow condition. this nanofluid consists of sodium carboxymethyl cellulose (cmc)–water and two different types of nanoparticles; namely, cuo and al2o3. entropy generation is obtained for various power law number, various volume concentration of nanoparticles, various dimensionless temperature and various reynolds number. it is found that with the decreasing power law number and duct length values, total entropy generation at fixed reynolds number decreases and with increasing wall temperature values, total entropy generation increases, also entropy generation decreases with increasing volume concentration of nanoparticles. keywords: entropy generation; non –newtonian fluid; power law number; laminar flow. 1.0 introduction improvement of convective heat transfer is very important for many thermo-fluid systems. the heat convection can passively be enhanced by fluid thermo physical properties. one way of improving the thermal conductivities of fluids is to suspend small solid particles in the fluid. pak and cho (1998) presented an experimental investigation of the convective turbulent heat transfer characteristics of al2o3 nanofluids. the heat transfer for the nanofluids increases with the increase of volume concentration and reynolds number. masuda et al. (1993) showed that the viscosity and the thermal conductivity of liquids are changed by dispersing very-fine particles of some nanoparticles like al2o3, sio2 and tio2 . *corresponding author e-mail: m-shabani@phdstu.scu.ac.ir mailto:m-shabani@phdstu.scu.ac.ir journal of mechanical engineering and technology 2 issn: 2180-1053 vol. 10 no.1 january – june 2018 das et al. (2003) have investigated the increase of thermal conductivity with temperature for wateral2o3 and water-cuo nanofluids by the temperature oscillation technique. entropy generation or exergy destruction is very important for design of thermo-fluid devises and for optimization, entropy generation must be decreased. for minimizing the entropy generation inside a duct has been extensively studied (bejan, 1996, 1972, 1996a, 1996b). oztop et al. (2009) have investigated the entropy generation in rectangular ducts with semicircular ends cross section with two boundary conditions: constant wall temperature and constant wall heat flux. ozotop et al. (2009) investigated the entropy generation in for hexagonal duct ducts with constant heat flux boundary condition. also, entropy generation in ducts with various cross sectional geometries under constant wall heat flux and laminar flow investigated by sahin, (1996, 1998a, 1998b). falahat and vosough (2012) computed entropy generation in a coiled tube under constant heat flux for both laminar and turbulent regimes using alumina–water nanofluids. they found that by adding 1% volume fraction of nanoparticles to the base fluid, entropy generation decreases about 3% in laminar flow. also, they obtained an optimal reynolds number for the turbulent flow for which the entropy generation was minimized. falahat (2011) made a study on entropy generation in s confocal elliptical ducts under constant heat flux. moghaddami et al. (2011) obtained optimum reynolds number which minimized entropy generation for water–al2o3 and ethylene glycol al2o3 nanofluids using a circular tube under constant heat flux. the main aims of this work to investigate a second law analysis for forced convection of non-newtonian nanofluids in circular cross section duct with constant wall temperature boundary condition. this base fluid is cmc–water with two different types of nanoparticles: cuo and al2o3.the effects of power-law nanofluids viscosity, reynolds number, nanoparticles volume fraction, dimensionless temperature and length of pipe on entropy generation are investigated. 2.0 methodology 2.1 physical model and thermo physical properties of non-newtonian nanofluids a geometrical configuration of the present problem has been shown in figure 1. the geometries consist of circular duct with constant wall temperature. the flow in this work is considered laminar, steady, fully developed and incompressible. figure 1. geometrical configuration entropy generation of pseudo-plastic non-newtonian nanofluids in circular duct under constant wall temperature issn: 2180-1053 vol. 10 no.1 january – june 2018 3 the nanofluid in this channel is non-newtonian and assumed that the fluid phase and nanoparticles are in the thermal equilibrium state and they flow with the same velocity. the cmc–water with low concentration (0.1–0.4%) is used as a base fluid of the nanofluid. the viscous properties of the cmc–water are given in table 1. jin et al. (2000) have shown that the thermo physical properties of the cmc–water (<6%) is similar to water. n is the power-law number of the non-newtonian base fluid. for newtonian fluid, n equals 1, n < 1 is descriptive of the pseudo-plastic fluid while n > 1 describes the dilatant fluid. table 1. viscous properties of cmc–water (jin et al., 2009) physical property n m cmc-water (0.0%) 1.00 0.000855 cmc-water (0.1%) 0.91 0.006319 cmc-water (0.2%) 0.85 0.017540 cmc-water (0.3%) 0.81 0.0313603 cmc-water (0.4%) 0.76 0.0785254 thermo physical properties of the nanofluid are obtained from the flowing relations is available in the literature, as discussed by khanafer et al. (2003). (1 ) nf bf p        (1) ( ) (1 )( ) ( )c c c p nf p bf p p        (2) 2 2 ( ) 2 ( ) k k k k k nf p bf bf p k k k k k bf p bf bf p          (3) 2.5 (1 ) bf nf      (4) table 2. thermophysical properties of pure fluid and nanoparticles (santra et al, 2008; raptis et al, 2004) physical properties cmc-water (0.00.4%) al2o3 cuo cp(j/kg k) 4179 765 535.6 𝜌 (kg/m3) 997.1 3970 6500 k (w/m k) 0.613 40 20 2.2 mathematical modeling on the basis of average heat transfer and fluid friction, the equation of entropy generation rate is presented by sahin (1998) as follows: journal of mechanical engineering and technology 4 issn: 2180-1053 vol. 10 no.1 january – june 2018 (5)                                      1 ln 8 )1( 1 .1 ln .4 .4 .4 l l l nst nst nst pgen e st ecf e e cgs where, the non-dimensional entropy generation number ns can be defined as (6) p gengen cg s tq s ns       / in above equations some dimensionless parameters can be defined as (7) puc h st   (8) )( 2 iwp ttc u ec   (9) t t w i t w    (10) l n l d  for power-law model, average velocity, friction factor, reynolds number (coulson, and richardson, 1999) and nusselt number (chhabra and richardson, 2008) are defined as (11)                       22 . 413 1 2 d d uf m d n n u n  (12) re 64 f (13) 2 re 3 11 8 4 n n u d n nn m n          (14) 1 1 33 1 31.75 4 g cn pnf nu n k l nf              3.0 results and discussions the effect of the power-law number, volume concentration of nanoparticles, reynolds number and length of duct for different nanofluids on the dimensionless entropy generation are investigated in circular duct under constant wall temperature. the surface temperature of duct is 350k. the present results was also validated against the results of sahin, 2008. figure 2 shows the total dimensionless entropy generation of water with respect to reynolds number. it entropy generation of pseudo-plastic non-newtonian nanofluids in circular duct under constant wall temperature issn: 2180-1053 vol. 10 no.1 january – june 2018 5 can be seen from the comparison that both solutions are in a good agreement with each other. two reasons for the discrepancies are due to different thermo physical properties and different nusselt number. sahin (2008) used nusselt number for tube under constant wall temperature (nu=3.66) but the present study utilized the equation (14). figure 2. comparing the present results with the results of sahin 2008 (n=1, θ=0.01 and φ=0) the effect of power-law number and volume concentration of al2o3 nanoparticles on dimensionless entropy generation have been shown in figure 3. it can be seen that dimensionless entropy generation decreases with decrease of power-law number in fixed volume concentration of nanoparticles and also it decreases with increase of volume concentration of nanoparticles. figure 3. the effect of n and volume concentration of al2o3 on dimensionless entropy generation (θ=0.08, re=500) journal of mechanical engineering and technology 6 issn: 2180-1053 vol. 10 no.1 january – june 2018 figure 4 shows the effect of dimensionless temperature and volume concentration of nanoparticles on dimensionless entropy generation. as the dimensionless temperature increases, the dimensionless entropy generation increases for each volume concentration of nanoparticles. also, entropy generation decrease with increase of volume concentration of nanoparticles for each dimensionless temperature. figure 4. the effect of dimensionless temperature and volume concentration of al2o3 on dimensionless entropy generation (n=0.85, re=500) figure 5 shows the effect of reynolds number and nanoparticles volume fraction on dimensionless entropy generation. it can be seen that dimensionless entropy generation decreases with the increase of reynolds number for each nanoparticles volume fraction. figure 5. the effect of reynolds number and volume concentration of al2o3 on dimensionless entropy generation (n=0.85, θ=0.08) entropy generation of pseudo-plastic non-newtonian nanofluids in circular duct under constant wall temperature issn: 2180-1053 vol. 10 no.1 january – june 2018 7 figure 6 shows the effect of different nanoparticles types (al2o3 and cuo) and volume concentration of nanoparticles on dimensionless entropy generation. when volume concentration of nanoparticles is increased, the dimensionless entropy generation decreases in two nanoparticles types. the cuo /cmc-water nanofluid with higher volume fraction of nanoparticles and higher volume of cmc may be a good choice as a working fluid because of their dimensionless entropy generation rate is lower than the al2o3/cmc-water nanofluids. figure 6. the effect of power law number and volume concentration of al2o3 and cuo on dimensionless entropy generation (re=500, θ=0.08) figure 7 shows the effect of length of duct and nanoparticles types on dimensionless entropy generation for fixed volume concentration of nanoparticles. it can be seen that dimensionless entropy generation increases with the increase of duct length for each type of nanoparticles, because by increasing of length of duct thermal irreversibility increases. journal of mechanical engineering and technology 8 issn: 2180-1053 vol. 10 no.1 january – june 2018 figure 7. the effect of length of duct and nanoparticles types on dimensionless entropy generation (n=0.85, θ=0.08, φ=2% and re=500) 4.0 conclusions in this study second law analysis of laminar flow of pseudo-plastic non-newtonian nanofluids has been obtained for circular duct under uniform wall temperature thermal boundary. some conclusions can be given as follows:  dimensionless entropy generation decreases with increasing of volume concentration of nanoparticles and reynolds number.  as the power-law number decreased, dimensionless entropy generation decreases for each volume concentration of nanoparticles.  dimensionless entropy generation increases with the increase of dimensionless temperature and increase of duct length for each type of nanoparticles.  dimensionless entropy generation of cuo /cmc-water nanofluids is lower than the al2o3/cmc-water nanofluids.  dimensionless entropy generation increases with the increase of duct length for each type of nanoparticles. entropy generation of pseudo-plastic non-newtonian nanofluids in circular duct under constant wall temperature issn: 2180-1053 vol. 10 no.1 january – june 2018 9 5.0 references bejan, a. (1979). a study of entropy generation in fundamental convective heat transfer. journal of heat transfer, 101(4), 718-725. bejan, a. (1982). entropy generation through heat and fluid flow. new york, wiley. bejan, a. (1996a). entropy generation minimization. boca raton, fl, crc press. bejan, a. (1996b). entropy generation minimization: the new thermodynamics of finite size devices and finite-time processes. journal of applied physics, 1191-1218. coulson, j.m. and richardson, j.f. (1999). chemical engineering (6th edition). oxford: butterworth–heinemann. chhabra, r.p. and richardson. j.f. (2008). non-newtonian flow in the process industries fundamentals and engineering applications. chemical engineering (2nd edition). das, sk., putra, n., thiesen, p. and roetzel. w. (2003). temperature dependence of thermal conductivity enhancement for nanofluids. journal of heat transfer, 567–574. dagtekin, i.,ozotop, h.f. and sahin, a.z. (2005). an analysis of entropy generation through a circular duct with different shaped longitudinal fins for laminar flow. international communications in heat and mass transfer, 171–181. falahat, a.r. (2011). entropy generation analysis of fully developed laminar forced convection in a confocal elliptical duct with uniform wall heat flux. indian journal of science and technology, 1649-1653. falahat, a.r. and vosough, a. (2012). effect of nanofluid on entropy generation and pumping power in coiled tube. j thermophys. heat transfer, 26 (1), 141–146. jin dx, wu yh and zou jt. (2000). studies on heat transfer to pseudo plastic fluid in an agitated tank with helical ribbon impeller. petro-chemical equipment. 29 (2): 7–9. khanafer, k., vafai, k. and lightstone, m. (2003). buoyancy-driven heat transfer enhancement in a two-dimensional enclosure utilizing nanofluids. international journal of heat and mass transfer, 3639-3653. masuda, h., ebata, a., teramae, k. and hishiunma, n. (1993). alteration of thermal conductivity and viscosity of liquid by dispersed ultra-fine particles (dispersion of al2o3, sio2, and tio2 ultra-fine particles). netsu bussei, 227–233. moghaddami, m., mohammadzade, a. and esfehani, s.a.v. (2011). second law analysis of nanofluid flow. energy conversion and management, 1397–1405. journal of mechanical engineering and technology 10 issn: 2180-1053 vol. 10 no.1 january – june 2018 ozotop, h.f., dagtekin, i. and sahin. a.z. (2009). second law analysis of fully developed laminar flow for rectangular ducts with semicircular ends. international communications in heat and mass transfer, 725–730. pak, b.c. and cho. i.y. (1998). hydrodynamic and heat transfer study of dispersed fluids with sub-micron metallic oxide particles. experimental heat transfer, 151-170. raptis, a., perdikis, c. and takhar. h.s. (2004). effect of thermal radiation on mhd flow. application mathematical computation, 645–649. sahin, a.z. (1996). thermodynamics of laminar viscous flow through a duct subjected to constant heat flux. energy, 1179–1187. sahin, a.z. (1998a). irreversibilities in various duct geometries with constant wall heat flux and laminar flow. energy, 465–473. sahin, a.z. (1998b). a second law comparison for optimum shape of duct subjected to constant wall temperature and laminar flow. journal heat and mass transfer, 425–430. santra, a.k. sen, s. and chakraborty, n. (2008). study of heat transfer augmentation in adifferentially heated square cavity using copper-water nanofluid. international journal thermal science, 1113–1122. nomenclature cp specific heat , kj/kg k greek symbols d diameter, m  viscosity of the fluid, pa.s f friction factor  density, 3 / mkg g mass flow rate, kg/s  nanoparticles volume fraction h heat transfer coefficient, w/m 2 k  dimensionless temperature k thermal conductivity of the fluid, w/m k subscripts l length of coiled tube, m bf base fluid m power law consistency nf nanofluid n power law index p particles nu nusselt number w wall nl dimensionless length pr prandtl number re reynolds number s specific entropy, kj/kg k t temperature, k x local position along the flow direction, m microsoft word 01_6129-17431-1-ed.docx journal of mechanical and technology issn 2180-1053 vol. 13 no. 2 december 2021 developments of cargo loss-mitigating strategies: a review k. i. kamarumtham1*, a. s. a. kader1 1), 2) department of mechanical engineering, university of technology malaysia 81310, utm skudai, johor, malaysia abstract covid-19 has stalled the growth of international seaborne trade to its nadir since the world’s financial crisis of 2008-2009. in light of this, shipping companies, more than ever, have to seek beyond conventional wisdom in devising their master plans to tolerate the adverse implications levied by the pandemic. in this paper, several advancements in the maritime industry, shipspecific and non-ship-specific, formulated by the scientific community over the last two decades are presented. for ship-specific advancements, the guidelines for establishing effective monitoring systems for container ships, optimizing ship-to-ship transfer operations between liquefied natural gas vessels and floating storage and regasification units, enhancing ethylene gassing-up operation for liquefied petroleum gas vessels, and minimizing volatile organic compounds emission of crude oil tankers are presented. for non-ship-specific advancements, the technologies of data-driven analytics, independent automotive damage appraisers intelligent cargo systems, and blockchain technology are presented. all these advancements can be leveraged by shipping companies to maximise their profits, particularly during the challenging epoch, by keeping their cargo loss at a minimal level. keywords: covid-19, cargo loss, blockchain technology, data-driven analytics, iada intelligent cargo system 1.0 introduction global pandemic covid-19 has caused the slowdown in the world economy and trade, which subsequently stalled the growth of international maritime trade to its lowest point since the financial crisis of 2008-2009 (unctad, 2020). moreover, the pandemic has encouraged shipping companies to foster technological solutions and keep abreast of the most recent advancements in the market as one of the strategies to tolerate the impacts of the pandemic since it has been demonstrated that shipping companies with swifter technological uptake are better at dealing with the disruptions engendered by the pandemic (unctad, 2020). this paper will focus on the major advancements that the scientific community has contributed over the last two decades to help shipping companies in minimizing cargo loss and subsequently maximising their profits. by and large, there are two forms of cargo loss; 1) cargo damage, and 2) cargo theft (wu et al., 2017). the former form of cargo loss is particularly evident for container ships carrying perishable goods (e.g., fruits, vegetables, meat, fish, etc.). the south african fruit industry accounts for 50 per cent of the country’s agricultural export (barrientos, 2012). moreover, the south african fruit industry generates approximately $1,02 billion per annum (economic research division, 2010). however, a massive portion of this *corresponding author email:khairul.kamarumtham@gmail.com 1 journal of mechanical and technology revenue and commodities are lost due to quality deprivation of these products before they reach their customers (emenike et al., 2016). from this, it is fair to conclude that other nations may confront the same circumstances. this conclusion is supported by the fact that roughly 35 per cent of perishable goods, namely fruits and vegetables, are lost in the cold chain logistics (vega, 2008). on top of that, $30 to $50 billion worth of cargo is stolen per annum (xu et al., 2018). from these two statistics, it is manifest that shipping companies must be equipped with effective cargo loss-mitigating strategies to further improve their profits. cargo loss may occur due to a number of factors, for instance, it may be due to shipping companies not being equipped with efficient monitoring systems (emenike et al., 2016). moreover, cargo loss may arise due to a poor security system, which consequently leads to cargo theft (xu et al., 2018). last but not least, cargo loss may take place owing to improper cargo loading and unloading procedures (shigunov et al., 2015). these three examples are just a few factors that may cause cargo loss. from the given examples, it is clear that the solutions to this problem must be sought from various fields of studies. in the next section, this paper will introduce the methodology utilized in providing shipping companies with multi-faceted solutions. subsequently, the paper will highlight a number of ship-specific strategies that shipping companies can implement to minimise their cargo loss. afterwards, this paper will outline several non-ship-specific strategies that shipping companies can foster to further reduce their cargo loss. ship-specific strategies are strategies that are exclusive to certain types of ships, whereas non-shipspecific strategies are strategies that can be adopted irrespective of ship types. finally, a conclusion will be drawn as the final chapter of this paper. 2.0 methodology in this paper, the major contributions formulated by the scientific community over the last two decades in developing effective strategies to reduce cargo loss will be discussed. to that end, the commonly recognized and accepted prisma (preferred reporting items for systematic reviews and meta-analyses) was adopted in searching the relevant scientific publications that will fulfil this paper’s objective (moher et al., 2009). in order to offer shipping companies with multi-faceted solutions having the capacity to comprehensively neutralize the cargo loss problem, the germane scientific publications were sought from a wide range of scientific fields (e.g., marine engineering, electrical and electronic engineering, environmental engineering, management, etc.). inasmuch as the objective of this paper is to provide shipping companies with the advancements that were undertaken over the last 20 years, the search span was set from the year 2000 – 2021. all publications before 2000 were excluded from the search. the following terms were used to systematically extract the relevant literature from two major scientific databases; 1) scopus and 2) web of science. other platforms (e.g., springerlink journal, sciencedirect journal, etc.) were also utilized to obtain additional references. “cargo loss*” and (ship* or vessel* or marine*) issn 2180-1053 vol. 13 no. 2 december 2021 2 journal of mechanical and technology the search included journal articles, review papers, and conference papers that were published in english. other document types were excluded (e.g., reports, case studies, letters, etc.). articles written in languages other than english were also excluded (e.g., chinese, russian, japanese, etc.) at this stage, 76 studies were extracted from the databases. subsequently, the studies were screened to remove duplicates. consequently, 11 studies were excluded from the search result. next, the abstracts of the studies were thoroughly screened for their relevance. accordingly, 43 studies were removed from the search result. the exclusion criterion at this stage was as follows. articles not related to strategies to reduce cargo loss in a maritime domain (e.g., cargo loss mitigating strategies for other forms of transportations; strategies for dealing with different problems such as ship collision, grounding, and so forth; causes of cargo loss without providing any remedies for the problem; etc.) after excluding studies fulfilling this exclusion criterion, the full texts of the remaining 22 studies were perused, and as a result, 14 studies were removed based on the following criteria. on closer perusal, the study did not address the strategies to reduce cargo loss in a maritime domain at all. two studies were removed due to this reason. the study proposes a solution that can only be effectively implemented by the relevant international organizations (e.g., proposal to imo to develop ship-specific operational guidance). two studies were removed due to this reason. no full text was available for closer perusal, neither from the publisher, research institution, or researcher’s personal web pages, namely researchgate.com and academia.edu. 10 studies were removed due to this reason. having completed this process, eight studies were left for discussion. figure 1 illustrates the whole process in a flowchart. issn 2180-1053 vol. 13 no. 2 december 2021 3 journal of mechanical and technology figure 1. prisma flowchart of research strategy 3.0 ship-specific strategies 3.1 container ship as hitherto mentioned, roughly 35 per cent of perishable goods are lost in the cold chain logistics. this problem is partially due to the lack of cost-effective monitoring systems of the good’s temperature (emenike et al., 2016). in this subsection, a solution to this problem will be discussed. i. rfid temperature sensing and predictive modelling the need for an improved strategy for effective in-transit cold chain management has motivated emenike et al. (2016) to undertake a series of experiments to compile a data set that represents cold chain operations in southern africa. subsequently, emenike et al. (2016) utilized the compiled data set to train neural network models that can predict the following two parameters; 1) current in-cargo temperatures based on the periphery’s temperature, and 2) future in-cargo temperatures based on the current temperatures. the latter parameter is particularly important in enabling proactive prevention of circumstances that may lead to cargo loss. the study has shown promising results in terms of its prediction accuracy; however, as emenike et al. (2016) have already mentioned in their study, more focus should be given to simplifying the system’s implementation method. for instance, the system may use wireless sensors in lieu of conventional sensors. this will allow this strategy to be all the more practical to be adopted by shipping companies. issn 2180-1053 vol. 13 no. 2 december 2021 4 journal of mechanical and technology 3.2 gas carrier 3.2.1 lng vessel the demand for floating storage and regasification units (frsu) has been rapidly increasing since it was first inaugurated in 2005 (wood & kulitsa, 2018). fsru capacity in operation had soared from approximately 35 million tonnes per annum (mtpa) to 795 mtpa in january 2017 (igu, 2017; wood & kulitsa, 2018). in february 2020, the capacity reached 826 mtpa (igu, 2020). in line with this, a significant amount of effort has been dedicated in diversifying fsru capabilities, ranging from comparatively cheap conversion lng vessels to larger-capacity, newly-built vessels with higher-tank-pressure ratings (wood & kulitsa, 2018). given this, fsru has opened several new markets to lng trade (wood & kulitsa, 2018). however, notwithstanding the impressive track records, a majority of the fsru fleet is run based on operating practices developed for the lower-tank pressure ratings of lng vessels which lead to operating inefficiencies (wood & kulitsa, 2018). these inefficiencies, if not treated, will lead to increments in onboard gas consumption and atmospheric emissions that consequently engender appreciable cargo loss (wood & kulitsa, 2018). fortunately, a study conducted by wood and kulitsa (2018) has outlined several cost-effective strategies, primarily interconnected with ship-to-ship (sts) transfer procedures and lng cargo rollover-mitigation measures, that can be fostered by shipping companies to improve their lng trade. moreover, the operation efficiency of fsru can be enhanced by exploiting the standard use of the recondensing equipment (kulitsa & wood, 2018; wood & kulitsa, 2018). in this subsection, all these strategies will be discussed. it is noteworthy that all these strategies do not require any capital investments or major modifications. i. sts transfer procedures wood and kulitsa (2018) have proposed 12 straightforward measures to minimize gas consumption in the gas combustion units (gcu) during sts transfers. these measures can be implemented individually; however, higher efficiency and lower cargo consumption can be achieved if implemented collectively (wood & kulitsa, 2018). these measures are as follows. a) ensure the lng cargo to be transferred by the lng carrier is as cold as possible immediately before sts transfer operations. b) ensure “lowest operating tank pressure” on fsru (or receiving lng carrier) at the beginning of sts transfer operations. c) avoid running safety equipment, namely gcu or steam dump, for the purpose of precooling the fsru tank’s heel cargo before sts transfer operations. d) ensure optimum coldest state of the receiving-tank structures around the vapour space volume is attained before commencing sts transfer operations. issn 2180-1053 vol. 13 no. 2 december 2021 5 journal of mechanical and technology e) minimize the increments of tank pressure on the receiving side during the sts transfer-lines-cool-down stage by injecting gas vapour produced during the linescool-down phase directly into the heel lng cargo in lieu of the vapour space of the fsru tank. on the whole, this measure is only effective for lng carriers to fsru and lng carriers to floating storage units (fsu) sts transfer operations. f) utilize continuous top spraying at a minimum rate to ensure the receiving tank’s vapour space and the surrounding tank structure to be as cold as possible throughout the sts transfer operations. g) minimize the use of gcu or steam dump equipment by utilizing the upper operating limits of the particular tank designs of the receiving vessel as a pressure control reference in identifying the ideal time to start the equipment (i.e., gcu or steam dump equipment). h) minimize the generation of boil-off gas (bog) on the discharging lng carrier and consequently boost vapour return flow from the receiving vessel by maintaining the tank pressure of the discharging vessel close to the pressure recorded during the initial cargo survey (i.e., from the custody transfer measuring system (ctms) on that particular vessel) throughout the sts transfer operations, immediately after commencing the transfer. i) optimize the combined application of measures 7 and 8 by carefully coordinating the fsru and lng carrier to gain more savings. this combination of these measures is known as the “tandem-pressure approach”, and it has the capacity to significantly increase efficiency. j) avoid overheating the lng by excessive recirculation of the regasification feed pumps within the fsru tanks. k) partially remove ballast water from ballast tanks that are close to the receiving vessel’s lng tanks to a certain level where the cargo tank’s outer structure is not wetted by the ballast. this measure enables natural heat ingress to be minimised; however, it is limited by the ship’s strength, stability, and safety limit allowances. l) as sts transfers are nearly completed, let the expected short-lived pressure increments take place in any of the receiving vessel’s tanks, up to the upper-operatingpressure limit of the tank. maintain pressure at that point, by briefly running the gcu/steam dump or related equipment. this temporary tank-pressure increment is due to the reduced vapour return flow towards the discharging lng carrier and it is not necessary to operate the gcu excessively to decrease the tank pressure at this stage since it will naturally decline once the sts transfer operation ends. ii. lng rollover-mitigation measures wood and kulitsa (2018) have proposed several measures to improve the rollover management of fsru. the measures can be divided into three categories; 1) preventative measures, 2) pro-active mitigation measures, and 3) reactive mitigation measures. these three measures are as follows. issn 2180-1053 vol. 13 no. 2 december 2021 6 journal of mechanical and technology issn 2180-1053 vol. 13 no. 2 december 2021 a) preventative measures in this category, several measures are proposed by wood and kulitsa (2018) in preventing lng stratification on fsru. the measures are as follows. • ensure one or more of the fsru’s tanks are in a near-empty state before sts transfer operations. the low lng heel cargo in this tank will likely cause the stratification to be eliminated by itself during the operation. however, in the case of stratification that is already developed, its impacts can be minimized by top spraying. • load lng into the fsru’s tank with the lowest cargo heel, at initially highest flow rates, during the initial stage of the sts transfer to boost mixing. this will slightly reduce the thickness of the top lng layer, or may permanently eliminate stratification upon its development. • allocate certain fsru tanks as “pump-out” tanks. upon completion of sts transfer operations, the bottom-lng layer in these tanks is subsequently removed before the development of rollover begins to occur. “pump-out” tanks are chosen since their bottom layers are normally far smaller than the bottom layers of other tanks. these tanks are mainly used to draw in the interim lng withdrawal for regasification feed. b) pro-active mitigation measures in this category, several measures are proposed by wood and kulitsa (2018) as proactive rollover mitigation measures for fsru. these measures are as follows. • when the sts transfer operations are initiated, those tanks on the fsru with the smallest lng heel cargo must be top sprayed at maximum pump rates. this means pumping lng from the new bottom lng layer, introduced by the sts transfer operation, and spraying it on the top lng layer from above, via the vapour space in the tank. this top spraying will force the densities of both lng layers to converge all the more rapidly; hence, shortening the roll-over “incubation” period. the shorter the rollover “incubation” period the “weaker” the future rollover is likely to be. two additional benefits of top spraying are as follows; 1) it slightly lessens the total volume of the bottom lng layer, which also weakens the rollover; and 2) it keeps the tank structure around the vapour space cold; hence any additional cold vapour exuded during the rollover is protected from being heat up and later expand that ultimately leads to a further increment in tank pressure. • by and large, it is better to avoid top spraying the tanks designated as “pump-out” tanks, as it would be counter-productive to the purposes of extending the rollover “incubation” period. nonetheless, minimal top spraying may be required, in some cases, for the purpose of keeping a tank’s vapour-space structure cold. this is particularly necessary when it becomes clear that the intended pump-out of the entire bottom layer is not going to be achieved and rollover becomes unavoidable. in such cases, minimal top spraying is recommended, but only when the pre-roll-over signs are conspicuous. 7 journal of mechanical and technology • sts transfer operations are often conducted at rates of 5000 to 6000 m3/h (in some cases, 8000 to 9000 m3/h), whilst lng carrier and fsru cargo lines are devised to cope with flow rates of up to 12,000 m3/h lng flow. this spare cargo line should be utilized whenever possible, to redistribute lng between tanks and afterwards produce a larger number of tanks with small top lng layers. • excessive recirculation of the bottom-layer lng should be avoided to prevent unnecessary heat induction from the pumps. this is relevant to all tanks either they are designed for top spraying or pump out. this measure is particularly important to lessen bog consumption throughout the fsru operations. • in the case of fsru with membrane tanks, minimize the contact between ballast water and the outer cargo tank structure to ensure the fsru’s trim, stability and hull strength are in good condition. this measure also reduces heat ingress to the bottomlng layer. moreover, this measure also contributes to small decrements in gas consumption during sts transfer operations. c) reactive mitigation measures in this category, several measures are proposed by wood and kulitsa (2018) as reactive rollover mitigation measures for fsru. these measures are as follows. • in the case of fsru with marvs-250-mbarg-rated tanks, it is recommended to run the safety equipment which burns a certain amount of gas for a short time, immediately before rollover and during onset as a precautionary measure. note that this should be only conducted when tank pressure is expected to reach the upperoperating-pressure limit. in short, the use of gcu or steam dump equipment should be limited to its essential use only to avoid unnecessary waste of the valuable cargo. iii. non-standard use of an fsru’s recondenser most modern fsru are equipped with recondensers that feed condensed bog to the regasification units and to a stream of lng extracted from the cargo tanks (kulitsa & wood, 2018; wood & kulitsa, 2018). the utilization of the recondensers during the regasification operations reduces gas loss by inhibiting excess bog to be consumed in gcu, steam dumps, flares, etc (kulitsa & wood, 2018; wood & kulitsa, 2018). lng gas loss can be further reduced if recondensers are also used in the fsru’s recirculation mode (kulitsa & wood). however, the use of recondensers in recirculation mode is still far from common (kulitsa & wood, 2018; wood & kulitsa, 2018). simply put, it can be concluded that the practice of using recondensers in recirculation mode is still “nonstandard”. in recirculation mode, lng is transferred from the lng tanks to recondensing equipment, and back again to the lng tanks. once the regasification process is halted, not much bog is required by the fsru engine room; thus, the vessel has to handle this excess. by condensing the bog to lng and returning it to the cargo tanks, the significant volume reduction involved has caused the lng tank pressure to rise at a slower rate due to the rising saturated vapour pressure (svp) of the lng in the tank as warmer lng is sent back to the tank. since the lng is distributed through several cargo tanks on an fsru, issn 2180-1053 vol. 13 no. 2 december 2021 8 journal of mechanical and technology the recondenser recirculation process can be executed in cycles by switching the feed lng and returning lng periodically from one bank to another. this measure helps to effectively redistribute the lng returning in the tanks in such a way that the tank pressure increment rate can be slowed down and concomitantly prolong the period without recourse to the gcu. this measure, ultimately, will lead to appreciable decrements in gas loss and degree of emission and simultaneously improve fsru’s efficiencies, particularly during periods of low or no gas send out from the fsru, provided that the recondensers are run within the safe operating limits. again, note that there are no additional operating costs inasmuch as this measure consumes far less energy than sending the lng back in the tanks via heat. moreover, as recondensers becomes simpler to be installed, it is fair to conclude that older fsru built without recondensers can have them retrofitted easily, and the non-standard use of recondensers outlined by kulitsa and wood (2018), sooner or later, will be acknowledged by shipping companies as a common practice. in conclusion, kulitsa and wood (2018) have outlined comprehensive guidelines in improving the efficiencies of fsru and reducing cargo loss. however, the studies should be supported by experimental data to gain more recognition from shipping companies. 3.2.2 lpg vessel ethylene (c2h4) is a colourless, flammable gas (national library of medicine, 2021). ethylene is one of the indispensable elements of the petrochemical industry and it is primarily used to formulate plastics, polyethylene, chlorostyrene derivatives, ethanol, and higher aliphatic alcohols (mcguire, 2000; schaller, 2012). recently, the demand for ethylene has soared particularly in china, the middle east, and the far east, which has engendered an unprecedented rise in the demand for the maritime service to deliver ethylene, principally to the hitherto mentioned regions (wieczorek, 2020). typically, specifically built lpg vessels are used to transport ethylene (wieczorek, 2020). in running such vessels, two of the most important operations are inerting and gassing-up operations (wieczorek, 2020). on one hand, the inerting operation involves forming an inert atmosphere in cargo tanks to inhibit the formation of an explosive mixture between oxygen and ethylene (wieczorek, 2020). on the other hand, the gassing-up operation involves removing an inert gas, namely nitrogen, by using ethylene vapour to prevent cargo contamination (mcguire, 2000). gassing-up operations are particularly challenging due to the following three reasons. i. both gases, ethylene and nitrogen, have very similar densities at certain temperatures (nanowski, 2016). ii. ethylene is one of the most expensive cargoes to be carried by marine transports (schaller, 2012). iii. there are no clear guidelines for gassing-up operations on the ethylene carriers (wieczorek, 2020). significant amounts of ethylene cargoes are often lost during the gassing-up operations, resulting in considerable financial loss for shipowners (wieczorek, 2020). approximately $40,000 worth of ethylene is lost during the operations (wieczorek, 2020). in view of this, issn 2180-1053 vol. 13 no. 2 december 2021 9 journal of mechanical and technology wieczorek (2020) conducted a study to optimize gassing-up operations. in this subsection, the important findings of the study will be highlighted. initially, wieczorek (2020) developed a theoretical mathematical model to simulate gassing-up operations. from the model, wieczorek (2020) proposed that gassing-up operations should be carried out with the minimum nitrogen and ethylene pressures in tanks. this will allow the gases (i.e., nitrogen and ethylene) to be separated and consequently enable the gassing-up operations to be undertaken in the shortest time with the smallest loss of ethylene cargo. subsequently, an experimental study was conducted by wieczorek (2020) to verify the accuracy of the findings. the following gassing-up guidelines were applied to m/v neptune, an ethylene carrier with four cargo tanks. the outcome of the novel procedure was then compared with other ethylene carriers (i.e., m/v saturn and m/v orion) that administer the conventional procedure. i. tanks must be gassed-up in pairs, in cascade. ii. after the first tank of each cascade is gassed-up, it must be isolated from the other tank in the cascade, and cargo cooling must be commenced. iii. cold ethylene vapour must be leaded, and parallel gassing-up must be started in the other two tanks that were not gassed-up during the cascade process. iv. the pressures of the tanks must be maintained to their minimum possible values. v. the pressures of tanks in a cascade must be similar. the ethylene cargo loss during the gassing-up operation of m/v neptune was then respectively compared with a three-cargo tank carrier and a two-cargo tank carrier, m/v saturn and m/v orion. note that the guidelines listed above were not applied to these two ethylene carriers since they act as control variables in this experiment. figure 2. comparison of ethylene cargo loss during gassing-up operations (wieczorek, 2020) 0 5 10 15 20 25 30 35 40 45 50 m/v saturn m/v orion m/v neptune loss during cooling the cargo [t] loss during gassing-up operation [t] issn 2180-1053 vol. 13 no. 2 december 2021 10 journal of mechanical and technology issn 2180-1053 vol. 13 no. 2 december 2021 figure 2 demonstrates the outcome of the experiment. as shown in figure 2, m/v neptune experiences approximately 43 tons loss of ethylene during the gassing-up operation, the highest loss as opposed to m/v orion and m/v saturn. however, unlike m/v orion and m/v saturn, m/v neptune experiences no cargo loss during the cargo cooling process. moreover, no cargo compressor stopped working due to extremely high pressure on the second stage of discharge (wieczorek, 2020). in conclusion, more studies are required to realize the conclusion obtained from the theoretical mathematical model established by wieczorek (2020); nonetheless, this study has provided substantial insight in developing effective guidelines to reduce cargo loss during gassing-up operations of ethylene carriers. 3.3 tanker 3.3.1 crude oil tanker cargo loss in crude oil tankers is mainly due to the following events; 1) ship collision, 2) grounding, and 3) emission of volatile organic compound (voc) (husain et al., 2003). on one hand, cargo loss due to collision and grounding has been estimated at approximately 25,000 tons per year, with occasional disastrous losses (husain et al., 2003). on the other hand, cargo loss due to voc emission has been estimated at roughly 1.6 million tons (gunner, 1999). currently, crude oil tankers are required to ensure the ullage space to be suitably inerted and the ullage pressure must be maintained at a positive value. concomitantly, the ullage pressure must always be below a specified value. these requirements are made to reduce cargo loss due to voc emission. however, the requirements are not effective enough to ensure ullage mixture to reach its equilibrium state throughout a tanker’s sailing period. consequently, voc emission continues to occur, particularly during transit, due to frequent venting and inerting. in light of this, husain et al. (2003) conducted a study to develop a strategy to minimize cargo loss due to voc emission. the study was divided into two parts; 1) analysis of crude oil under negative pressure, and 2) minimization of voc emission by negative pressure. i. analysis of crude oil under negative pressure initially, husain et al. (2003) conducted theoretical and experimental analyses to identify the vaporization properties of representative crude oil at below atmospheric pressures and at moderately elevated temperatures (i.e., typical temperatures of crude oil tankers). for experimental analysis, husain et al. (2003) selected three crude oils; 1) 12 api, 2) 30 api and 3) 37 api. note that api stands for american petroleum institute, a commonly used index to measure the density of crude oil or refined products (api, 2021; mckinsey & company 2021). each of the crude oils was then undergone a series of gas evolution tests at pressures of 0 psig, -2 psig, -3 psig and -5 psig at temperatures of 67 f, 80 f, and 110 f. finally, the compositions of the liberated gas were analysed. figure 3 and table 1 show the schematic of equipment used in the experimental analysis and results of the evolution tests, respectively. 11 journal of mechanical and technology issn 2180-1053 vol. 13 no. 2 december 2021 figure 3. equipment arrangement for experimental analysis (husain et al., 2003) table 1. results of gas evolution tests (husain et al., 2003) the following conclusions can be made from the result shown in table 1. • api gravity of crude oil for heavy oils, vaporization can be considered insignificant. • temperature temperature increments boost the vaporization of intermediate components. • pressure as pressure decreases below atmospheric pressure, the gas composition becomes denser with intermediate components. ii. minimization of voc emission by negative pressure the data obtained from the theoretical and experimental studies were then utilized by hashim et al. (2003) to develop an automated closed-loop control system that can limit in-transit voc emissions by ensuring ullage pressure is maintained at a constant value. the system enables the ullage gases to be circulated in a closed-loop arrangement at subatmospheric pressure through a blower and seawater heat exchanger. given this, the ullage is maintained at a constant value despite the perturbations generated by mass and heat exchange during a voyage. most importantly, the proposed model is inexpensive and can be easily installed on tankers. 0.0013 0.0500 0.0744 0.1234 0.0092 0.0568 0.0807 0.1288 0.0245 0.0703 0.0933 0.1398 0.0009 0.0327 0.0507 0.0956 0.0167 0.0535 0.0756 0.1371 0.0868 0.1626 0.2240 0.5298 0.0007 0.0300 0.0470 0.0906 0.0139 0.0479 0.0687 0.1284 0.0683 0.1329 0.1837 0.4093 80 110 67 80 110 -3 psig -5psig mole fraction 12 api 30 api 37 api 67 80 110 67 f crude oil 0 psig -2 psig pressure temperature 12 journal of mechanical and technology issn 2180-1053 vol. 13 no. 2 december 2021 in conclusion, hashim et al. (2003) have provided comprehensive discussion, theoretically and experimentally, on the importance of containing ullage pressure at a constant value. moreover, hashim et al. (2003) have proposed an automated closed-loop control system that can limit in-transit voc emission. needless to say, the proposed system has to be supported with experimental data to make certain of its effectiveness in limiting voc emission. all things considered; it is fair to conclude that the study has projected a promising future for crude oil tankers. 4.0 non-ship-specific strategies 4.1 data-driven analytics cargo loss incidents not only impose financial losses to shipping companies but also disrupt the entirety of logistics systems. in addition to financial losses, cargo loss may also cause shipping companies to be susceptible to increased insurance costs, loss of business market opportunities, and degradation of companies’ reputation (burges, 2012). given this, it is evident that cargo loss in logistics systems requires immediate attention. notwithstanding this, in the field of logistic risk management, the number of studies dedicated to solving this issue, particularly via data-driven analytics, is rather nominal. in consequence, wu et al. (2017) conducted a study to develop a framework of business analytics to enable shipping companies to exploit their data to determine events that may trigger cargo loss incidents, and subsequently formulate strategies and distribute resources to prevent cargo loss in their logistics systems. the framework of business analytics is divided into three parts; 1) descriptive analytics, 2) predictive analytics, and 3) prescriptive analytics. shipping companies can implement the proposed business analytics to exploit their logistics data to identify causal factors of cargo loss in their logistics systems, whilst companies without logistics data can integrate the proposed cargo loss mitigation measures into their logistic systems. figure 4 shows the frameworks of the proposed business analytics. figure 4. the proposed business analytics for cargo loss severity (wu et al., 2017) 13 journal of mechanical and technology issn 2180-1053 vol. 13 no. 2 december 2021 the applications of data-driven analytics can also be observed in other domains od the maritime industry that can be clustered into three groups; 1) anomaly detection methods, 2) ais data analytics, and 3) simulation of maritime traffic data (munim et al., 2020; sidibe & shu, 2017; wang et al., 2019; yang et al., 2019). 4.2 intelligent cargo security system as previously mentioned, at least $30 billion worth of cargo is stolen every year. current measures implemented by governments to prevent cargo theft, notwithstanding being strict, are highly burdensome (xu et al., 2018). the united states department of homeland security (dhs) necessitates the scanning of 100 per cent of maritime cargo entering the united states from 2012, covering roughly 11.6 million containers per year (mcneil & zuckerman, 2010). besides, current measures enforced by governments to make cargo systems safer are insufficient to prevent illegal practices such as tampering, load theft, terrorism, and unauthorized access, etc., from occurring (dinesh et al., 2017). given this, dinesh et al. (2017) and xu et al. (2018) conducted studies to develop intelligent cargo security systems. these systems will be discussed in this subsection. 4.2.1 iada intelligent cargo system iada stands for “independent automotive damage appraisers”. iada acts as the final authorization centre for any type of equipment and facility to be hosted on the cargo by inspecting and customizing the cargo with the necessary components to fulfil the guidelines imposed by governments (jijesh, 2016; wang, 2008; zhou, 2015). the primary objective of the system is to make certain the safety of the onboard crew members and relevant shipping operations. to further enhance the performance of the system, dinesh et al. (2017) incorporated fingerprint and gps modules into the system as an attempt to develop a completely standalone framework of the system. the fingerprint module acts at the primary level of authentication to the system. whilst, the gps module provides uninterrupted access and location services for the system. the proposed model is claimed to enhance current iada systems by supplementing six major benefits; 1) saving of labours, 2) easy coding and maintenance, 3) cost-effective, 4) higher consistency and quality, 5) higher accuracy, and 6) higher safety rigidity (dinesh et al., 2017). figure 5 shows the architecture and the sequence diagram of the proposed system. figure 5. system architecture and sequence diagram of the proposed system (dinesh, 2017) 14 journal of mechanical and technology issn 2180-1053 vol. 13 no. 2 december 2021 4.2.2 blockchain technology blockchain technology, which was initially used in a crypto-currency, bitcoin, as a decentralized bookkeeping system to prevent double-spending, was leveraged by xu et al. (2018) to enhance the current cargo security system (nakamoto, 2009; xu et al., 2018). by and large, blockchain has three important features; 1) public accessibility: all data stored with blockchain is available to the public. 2) immutability: it is extremely difficult, if not impossible, to modify, alter, or delete information that has been fed to the blockchain when the security assumption is fulfilled. 3) resilience: each participant of the system possesses a complete copy of the blockchain and no single point of failure can impinge the availability of the stored data. simply put, blockchain can provide a unified, immutable, and resilient information management portal for all maritime transportation participants and ultimately improve the transparency of the cargo flow, accelerate the inspection process and minimize fraud (xu et al., 2018). the proposed scheme developed by xu et al. (2018) to embed blockchain technology into the maritime cargo management system will be discussed in this subsection. i. participants screening and digital identity generation the first step to secure the maritime supply chains, particularly for border-crossing cargo flows, is to screen involved participants and generate digital identities for them. an exclusive identity management system is developed to fulfil this objective, which acts as an extended public key infrastructure. a party that is involved in maritime transportation will possess a public/private key pair as its digital identity. if the party is a company, every employee involved in the cargo handling operations should possess his/her own digital identity. fortunately, several hardware technologies have been developed to improve maritime supply chain transparency, such as smart containers, rfid tags, tracking devices, etc (carn, 2011; shi et al., 2011; talukder, 2007). ii. checking digital identity subsequently, to make certain that a participant can only use his/her own digital identity, the system will execute the following tasks. • ensure both public and private keys are matched. • ensure the public key is valid and contains all the necessary information. • ensure the user is the real owner of the keys. • record all the activities and save them in the blockchain. as discussed above, it is evident that blockchain technologies can be used by shipping companies to further solidify the measures committed by the governments and concomitantly streamline the cargo inspection processes. however, there are several challenges that maritime communities have to solve to fully exploit the expected benefits of blockchain technologies (munim et al., 2021). these challenges are summarised in table 2. fortunately, a growing body of studies have been dedicated to resolving these challenges and all the major resolutions to these challenges are reviewed in detail by munim et al. (2021). 15 journal of mechanical and technology issn 2180-1053 vol. 13 no. 2 december 2021 table 2. challenges in the implementation of blockchain technologies challenges source lack of authority for standardization jovic et al. (2020), segers et al. (2019) interoperability and lack of scale allen et al. (2019), irannezhad (2020), pranav et al. (2020), shi & wang (2018), todd (2019) antitrust law and commercial privacy jovic et al. (2019), todd (2019) environmental concern jovic et al. (2020) dispute resolution perkusic et al. (2019), todd (2019) data tampering and hacking dutta et al. (2020), kermani et al. (2020), pranav et al. (2020), nyugen et al. (2020), greiman (2019) 5.0 conclusion covid-19 has left shipping companies with no other options but to leverage recent technologies to endure the impacts subjected by the pandemic. in this paper, several advancements contributed by the scientific community over the last two decades have been discussed to assist shipping companies to minimize their cargo loss and simultaneously maximise their profits. cargo loss can occur due to several factors such as inefficient cargo loading and unloading operations, substandard security systems, and poor goods monitoring systems, just to name but a few. for lng vessels, sts operations can be improved by implementing the guidelines outlined by wood and kulitsa (2017). for lpg vessels carrying ethylene, shipping companies can exploit the guidelines provided by wieczorek (2020) to optimize the gassing-up operations. moreover, crude oil tankers can reduce cargo loss by adopting closed-loop control systems to minimize voc emissions. furthermore, shipping companies can utilize iada systems and blockchain technologies, as respectively proposed by dinesh et al. (2017) and xu et al. (2018) to improve their cargo security systems. also, for container ships carrying perishable goods, a monitoring system proposed by emenike et al. (2016) can be adopted to reduce cargo damage. finally, shipping companies can exploit their logistics data, as recommended by wu et al. (2017), to determine the factors that cause cargo loss and subsequently develop specific strategies and allocate necessary resources to prevent cargo loss. 6.0 list of references allen, d., berg, c., davidson, s., novak, m., & potts, j. (2019). international policy coordination for blockchain supply chains. asia & the pacific policy studies, 6(3), 367-380. https://doi.org/10.1002/app5.281 american petroleum institute (api). (2021). oil categories. retrieved 19th june from https://www.api.org/products-and-services/engine-oil/eolcs-categories-andclassifications/oil-categories burges, d. (2012). cargo theft, loss prevention, and supply chain security. butterworth-heinemann ltd. 16 journal of mechanical and technology issn 2180-1053 vol. 13 no. 2 december 2021 carn, j. (2011). smart container management: creating value from real-time container security device data technologies for homeland security (hst), 2011 ieee international conference, dinesh, p. p., prabhakar, m., murthy, m. v., jijesh, j. j. (2017). iada intelligent cargo system with integrated fingerprint module and gps modules 2nd ieee international conference on recent trends in electronics, information and communication technology, proceedings, dutta, p., choi, t., somani, s., & butala, r. (2020). blockchain technology in supply chain operations: applications, challenges and research opportunities. transportation research part e: logistics and transportation review, 142, 102067. https://doi.org/10.1016/j.tre.2020.102067 emenike, c. c., van eyk, n. p., and hoffman, a. j. (2016). improving cold chain logistics through rfid temperature sensing and predictive modelling 2016 ieee 19th international conference on intelligent transportation systems (itsc), https://ieeexplore.ieee.org/document/7795932/ greiman, v. (2021). navigating the cyber sea: dangerous atolls ahead. in proceedings of the iccws 2019 14th international conference on cyber warfare and security: iccws 2019. stellenbosch, south africa. husain, m., hunter, h., altshuller, d., shtepani, e., luckhardt, s. (2003). crude oil un der negative pressures and hydrocarbons emission containment [conference p aper]. transactions society of naval architects and marine engineers, 111, 58 4-607. http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.538.2947&rep =rep1&type=pdf international gas union (igu). (2017). 2017 world lng report, igu. international gas union (igu). (2020). 2020 world lng report. igu. irannezhad, e. (2020). is blockchain a solution for logistics and freight transportation pr oblems? transportation research procedia, 48, 290-306. https://doi.org/10.101 6/j.trpro.2020.08.023 jijesh, j. j., suraj, s., bolla, d. r., sridhar, n. k., dinesh, d. p. (2016, 6th-8th october). a method for the personal safety in real scenario international conference on computation system and information technology for sustainable solutions (csitss), jović, m., filipović, m., tijan, e., & jardas, m. (2019). a review of blockchain technology implementation in shipping industry. pomorstvo, 33(2), 140-148. https://doi.org/10.31217/p.33.2.3 jović, m., tijan, e., žgaljić, d., & aksentijević, s. (2020). improving maritime transpo rt sustainability using blockchain-based information exchange. sustainabilit y, 12(21), 8866. https://doi.org/10.3390/su12218866 17 journal of mechanical and technology issn 2180-1053 vol. 13 no. 2 december 2021 kermani, m., parise, g., shirdare, e., & martirano, l. (2020). transactive energy solution in a port’s microgrid based on blockchain technology. in proceedings of the 2020 ieee international conference on environment and electrical engineering and 2020 ieee industrial and commercial power systems europe (eeeic/i&cps europe). new york, usa. kulitsa, m., and wood, d. a. (2018). enhanced application for fsru recondensing equipment during periods of low or no gas send out to minimize lng cargo losses. petroleum, 4(4), 365-374. https://doi.org/10.1016/j.petlm.2018.01.002 mcguire, g., white, b., (2000). liquefied gas handling principles on ships and in terminals (third ed.). witherby & co ltd. mckinsey & company. (2021). api gravity. retrieved 19th june from https://www.mck inseyenergyinsights.com/resources/refinery-reference-desk/api-gravity/ mcneill, j. b., zucerman, j. (2010). the cargo-screening clog: why the maritime mandate needs to be re-examined. t. h. foundation. munim, z., duru, o., & hirata, e. (2021). rise, fall, and recovery of blockchains in th e maritime technology space. journal of marine science and engineering, 9 (3), 266. https://doi.org/10.3390/jmse9030266 moher, d., liberati, a., tetzlaff, j., altman, d. g., and group, p. (2009, jul 21). prefer red reporting items for systematic reviews and meta-analyses: the prisma state ment. plos medicine, 6(7), e1000097. https://doi.org/10.1371/journal.pmed.100 0097 munim, z., dushenko, m., jimenez, v., shakil, m., & imset, m. (2020). big data and artificial intelligence in the maritime industry: a bibliometric review and future research directions. maritime policy & management, 47(5), 577-597. https://doi.org/10.1080/03088839.2020.1788731 nakamoto, s. (2009). bitcoin: a peer-to-peer electronic cash system,”. https://www.res earchgate.net/publication/228640975_bitcoin_a_peer-to-peer_electronic_cash _system nanowski, d. (2016). the influence of incondensible gases on the refrigeration capacity of the reliquefaction plant during ethylene carriage by sea. journal of kones, 23(3), 359-364. national library of medicine. (2021). ethylene. retrieved 19th june from https://pubch em.ncbi.nlm.nih.gov/compound/ethylene nguyen, s., chen, p., & du, y. (2020). risk identification and modeling for blockchain -enabled container shipping. international journal of physical distribution & l ogistics management, 51(2), 126-148. https://doi.org/10.1108/ijpdlm-01-2020-0 036 18 journal of mechanical and technology issn 2180-1053 vol. 13 no. 2 december 2021 perkušić, m., jozipović, š., & piplica, d. (2020). need for legal regulation of blockchain and smart contracts in the shipping industry. transactions on maritime science, 9(2). https://doi.org/10.7225/toms.v09.n02.019 pranav, p., saikiran, a., mukul, m., ravishankar, b., & shailaja, v. (2021). critical analysis of international shipments within mainstream blockchain framework using industrial engineering techniques. in proceedings of the 2020 international conference on mainstreaming block chain implementation (icombi). bengaluru, india. schaller, g. e., (2012, feb 20). ethylene and the regulation of plant development. bmc biology, 10, 9. https://doi.org/http://doi.org/10.1186/1741-7007-10-9 segers, l., ubacht, j., tan, y., & rukanova, b. (2019). the use of a blockchain-based smart import declaration to reduce the need for manual cross-validation by customs authorities. in proceedings of the 20th annual international conference on digital government research. dubai, uae. shigunov, v., moctar, o. e., and rathje, h. (2015). operational guidance for prevention of cargo loss and damage on container ships. ship technology research, 57(1), 8-25. https://doi.org/10.1179/str.2010.57.1.002 shigunov, v., rathje, h., and moctar, b. e. (2015). towards safer container shipping. ship technology research, 60(1), 34-40. https://doi.org/10.1179/str.2013.60. shi, h., & wang, x. (2021). research on the development path of blockchain in shipping industry. in proceedings of the asia-pacific conference on intelligent medical 2018 & international conference on transportation and traffic engineering 2018. beijing, china. shi, x., tao, d., vob, s. (2011). rfid technology and its application to port-based container logistics. journal of organizational computing and electronic commerce, 21, 332-347. sidibé, a., & shu, g. (2017). study of automatic anomalous behaviour detection techniques for maritime vessels. journal of navigation, 70(4), 847-858. https://doi.org/10.1017/s0373463317000066 talukder, n., ahamed, s. i., and abid, r. m. "smart tracker: light weight infrastructu re-less assets tracking solution for ubiquitous computing environment," 2007 fourth annual international conference on mobile and ubiquitous systems: net working & services (mobiquitous), 2007, pp. 1-8, doi: 10.1109/mobiq.2007.4 451037. todd, p. (2019). electronic bills of lading, blockchains and smart contracts. internation al journal of law and information technology, 27(4), 339-371. https://doi.org/ 10.1093/ijlit/eaaa002 19 journal of mechanical and technology issn 2180-1053 vol. 13 no. 2 december 2021 wang, k., liang, m., yan, l., liu, j., & liu, r. (2019). maritime traffic data visualization: a brief review. in ieee 4th international conference on big data analytics (icbda). suzhou, china. wang, y., potter. a., (2008, 16-18 december 2007). the application of real time trac king technologies in freight transport 2007 third international ieee conferen ce on signal-image technologies and internet-based system, shanghai, china. wieczorek, a. (2020). an experimental ethylene carrier gassing-up operation. scientific journals of the maritime university of szczecin-zeszyty naukowe akademii morskiej w szczecinie, 62(134), 43-48. https://doi.org/10.17402/418 wood, d. a., & kulitsa, m. (2018). a review: optimizing performance of floating storage and regasification units (fsru) by applying advanced lng tank pressure management strategies. international journal of energy research, 42(4), 1391-1418. https://doi.org/10.1002/er.3883 wu, p.j., chen, m. c., and tsau, c. k. (2017). the data-driven analytics for investigating cargo loss in logistics systems. international journal of physical distribution & logistics management, 47(1), 68-83. https://doi.org/10.1108/ijpdlm-02-20160061 xu, l., chin, l., gao, z., chang, y., iakovou, e., shi, w. (2018). binding the physical and cyber worlds: a blockchain approach for cargo supply chain security enhancement 2018 ieee international symposium on technologies for homeland security, hst, yang, d., wu, l., wang, s., jia, h., & li, k. (2019). how big data enriches maritime re search – a critical review of automatic identification system (ais) data applicat ions. transport reviews, 39(6), 755-773. https://doi.org/10.1080/01441647.201 9.1649315 zhou, l., lou, c., chen, y., xia, y. (2015). multi-agent-based smart cargo tracking system (international journal of high performance computing and networking) 20 issn: 2180-1053 vol. 7 no. 2 july december 2015 analitical method to calculate the unknown geometry of cylindrical gears 57 analitical method to calculate the unknown geometry of cylindrical gears g. gonzález rey1*, a. garcía toll2 1 universidad de tecnológica de aguascalientes, blvd. juan pablo ii, no. 1302, fracc. exhacienda, la cantera, ags. c.p. 20200. méxico 2, instituto superior politécnico josé antonio echeverría, cujae. marianao 15. la habana. cuba abstract a procedure of reverse engineering to determine the basic geometry for manufacturing of external parallel-axis cylindrical involute gears by means of workshop measurement tools is presented. this procedure proposes a practical method to obtain the fundamental gear parameters in order to have a reference for calculating the load capacity of cylindrical gears or when a “copy” of an external parallel-axis cylindrical involute gear is necessary for recreating other new gear according to iso standards by generation cutting. keywords: cylindrical gear, unknown geometry, inverse engineering 1.0 introduction gear engineering requires of professional skills in several action fields like design, production, operation, maintenance, repair and recycled. generally, the main action fields are established by the industry profile. industries and companies with actions in the maintenance and repair of gear usually ask for professionals with skill in the recovery of these elements. in general, the repair of gear implies bigger challenge to the gear engineers, because the problems and solutions involve already manufactured gears whose geometry is generally unknown. in this situation, the engineer needs to know the previous basic geometry of the gears in order to have a reference for the recovering or remanufacturing. actually, there are a wide variety of cnc generative gear testers and coordinate measuring machines (cmm) destined for inspection and corresponding author e-mail: gonzalo.gonzalez@utags.edu.mx issn: 2180-1053 vol. 7 no. 2 july december 2015 journal of mechanical engineering and technology 58 control of spur and helical gears with fully automatic measuring cycle and extremely short measuring times combined with high measuring accuracies. in this advanced gear measuring machines, the profile of the tooth can be checked and compared with a flank topography reference and by means of a trial and error procedure it is possible to obtain an approximate geometry of the analyzed gears (kumar, 2014). moreover, some advanced measurement machine have incorporated special program for measuring gears with unknown parameters and determining some important data of the gear basic geometry (grimsley, 2003). unfortunately, the price lists of these machines are very high, somewhere in the $300,000 to $500,000 range, and not often accessible to the company or factory involves with gear remanufacturing. concerning with this situation, gear specialists (gonzález rey, 1999), (innocenti, 2007), (belarifi et al., (2008), (schultz, 2010) involved with recreating replacement gears are considered alternative procedures to determine the unknown gear geometry using more simple measurement tools. consequently, this paper presents a method of reverse engineering to determine the unknown gear geometry in order to have a reference for the design or manufacturing. this method, based on author’s experiences in the analysis, recovery and conversion of helical and spur gears, proposes a practical procedure with results not too exact, but practically acceptable, to obtain the fundamental parameters by means of conventional measurement tools. this method is useful for the recreating of new external parallelaxis cylindrical involute gears according to iso standards by a generation cutting process. 2.0 basic gear data to determine the unknown external cylindrical involute gear geometry it is known that the question of what data is required to specify an external parallel-axis cylindrical involute gear can be answered perfectly by means of the theory associated with the involute helicoids surface of the flank of a helical gear (maag, 1990). in this case, it is necessary to know number of teeth, tip diameter, root diameter, base diameter, base helix angle and base tooth thickness. the three first data can be determined easily by measurement but the data associated with the base cylinder can be determined only by special gear measuring equipment. thus, where only a sample of a gear but not complete gear data is available initially, the specification for generating the gear can be calculated. main formulas involved with the theory of the involute helicoids surface of the flank of a helical gear are summarized issn: 2180-1053 vol. 7 no. 2 july december 2015 analitical method to calculate the unknown geometry of cylindrical gears 59 below (equations (1) –(8)). some of them are fundamentals in the determination of the gear geometry that fulfills the data requested as reference for the design or manufacturing. 2 involves with gear remanufacturing. concerning with this situation, gear specialists (gonzález rey, 1999), (innocenti, 2007), (belarifi et al., (2008), (schultz, 2010) involved with recreating replacement gears are considered alternative procedures to determine the unknown gear geometry using more simple measurement tools. consequently, this paper presents a method of reverse engineering to determine the unknown gear geometry in order to have a reference for the design or manufacturing. this method, based on author's experiences in the analysis, recovery and conversion of helical and spur gears, proposes a practical procedure with results not too exact, but practically acceptable, to obtain the fundamental parameters by means of conventional measurement tools. this method is useful for the recreating of new external parallelaxis cylindrical involute gears according to iso standards by a generation cutting process. 2.0 basic gear data to determine the unknown external cylindrical involute gear geometry it is known that the question of what data is required to specify an external parallel-axis cylindrical involute gear can be answered perfectly by means of the theory associated with the involute helicoids surface of the flank of a helical gear (maag, 1990). in this case, it is necessary to know number of teeth, tip diameter, root diameter, base diameter, base helix angle and base tooth thickness. the three first data can be determined easily by measurement but the data associated with the base cylinder can be determined only by special gear measuring equipment. thus, where only a sample of a gear but not complete gear data is available initially, the specification for generating the gear can be calculated. main formulas involved with the theory of the involute helicoids surface of the flank of a helical gear are summarized below (equations (1) –(8)). some of them are fundamentals in the determination of the gear geometry that fulfills the data requested as reference for the design or manufacturing.  xchmzmd af  **2cos  (1) mcdad fwa  *22 1,22,1 (2) tb zm d   cos cos    (3)            cos tan tan 1t (4)          tan2 2 xmsn (5)          t n bn invmz s mzs cos (6) 3                    cos tan tan 1 zm db b (7)  cos mp bn (8) where: moreover, standards (norma nc 02-04-04, 1978; iso standard 1340, 1976; agma standard 910-c90; 1990) with guidelines about the complete information to be given to the manufacturer in order to obtain the gear required give you an idea about the proper data to be placed on drawings of gears for general or special purposes. the mentioned information includes details of the gear body, the mounting design, facewidth, and fundamental gear data for manufacturing, inspection and reference. usually, the gear data can be efficiently and consistently specified on the gear drawing in a standardized block format. figure 1 shows the typical gear data block and information required on drawings for standard helical gears according to cuban standard nc 02-04-04:1998. figure 1. typical data for gear drawings to be given by the gear designer for the gear manufacturer, according to nc 02-04-04: 78 sn : normal tooth thickness on reference cylinder (mm) sbn : normal base tooth thickness (mm) pbn : normal base pitch (mm)  : pressure angle () t : transverse pressure angle () ha* : factor of addendum c* : factor of radial clearance z : number of teeth m: normal module (mm) x: addendum modification coefficient : helix angle at a reference diameter () da: tip diameter (mm) df: root diameter (mm) aw: centre distance (mm) db: base diameter (mm) b: base helix angle () where: z : number of teeth sn : normal tooth thickness on reference m: normal module (mm) cylinder (mm) x: addendum modification coefficient sbn : normal base tooth thickness (mm) β: helix angle at a reference diameter (°) pbn : normal base pitch (mm) da: tip diameter (mm) α : pressure angle (°) df: root diameter (mm) αt : transverse pressure angle (°) aw: centre distance (mm) ha* : factor of addendum db: base diameter (mm) c* : factor of radial clearance βb: base helix angle (°) moreover, standards (norma nc 02-04-04, 1978; iso standard 1340, 1976; agma standard 910-c90; 1990) with guidelines about the complete information to be given to the manufacturer in order to obtain the gear required give you an idea about the proper data to be placed on drawings of gears for general or special purposes. the mentioned information includes details of the gear body, the mounting design, facewidth, and fundamental gear data for manufacturing, inspection and reference. usually, the gear data can be efficiently and consistently specified on the gear drawing in a standardized block format. figure issn: 2180-1053 vol. 7 no. 2 july december 2015 journal of mechanical engineering and technology 60 1 shows the typical gear data block and information required on drawings for standard helical gears according to cuban standard nc 02-04-04:1998. 3                    cos tan tan 1 zm db b (7)  cos mp bn (8) where: moreover, standards (norma nc 02-04-04, 1978; iso standard 1340, 1976; agma standard 910-c90; 1990) with guidelines about the complete information to be given to the manufacturer in order to obtain the gear required give you an idea about the proper data to be placed on drawings of gears for general or special purposes. the mentioned information includes details of the gear body, the mounting design, facewidth, and fundamental gear data for manufacturing, inspection and reference. usually, the gear data can be efficiently and consistently specified on the gear drawing in a standardized block format. figure 1 shows the typical gear data block and information required on drawings for standard helical gears according to cuban standard nc 02-04-04:1998. figure 1. typical data for gear drawings to be given by the gear designer for the gear manufacturer, according to nc 02-04-04: 78 sn : normal tooth thickness on reference cylinder (mm) sbn : normal base tooth thickness (mm) pbn : normal base pitch (mm)  : pressure angle () t : transverse pressure angle () ha* : factor of addendum c* : factor of radial clearance z : number of teeth m: normal module (mm) x: addendum modification coefficient : helix angle at a reference diameter () da: tip diameter (mm) df: root diameter (mm) aw: centre distance (mm) db: base diameter (mm) b: base helix angle () figure 1. typical data for gear drawings to be given by the gear designer for the gear manufacturer, according to nc 02-04-04: 78 3.0 initial data and measurements in the proposed procedure, to calculate the fundamental gear tooth data of an external parallel-axis cylindrical involute gear, it is necessary to know the following parameters: • number of teeth on pinion and gear (z1, z2) • tip diameters on pinion and gear (da1, da2) in mm • facewidth on pinion and gear (b1, b2) in mm • base tangent length spanned in k teeth on pinion and gear (wk1, wk2) in mm • number of teeth spanned for the base tangent length on pinion and gear (k1, k2) • tooth depth on pinion and gear (h1, h2) in mm • centre distance (aw) in mm • helix angle at tip diameter (βa) in degree number of teeth (z): special care should be had counting the quantity of teeth in the gears. it is a good practice to make some mark with chalk in the tooth where the count begins to assure that the number of teeth was correctly determined. an incorrect specification of the number of teeth on gears will be catastrophic in the next calculation. issn: 2180-1053 vol. 7 no. 2 july december 2015 analitical method to calculate the unknown geometry of cylindrical gears 61 tip diameters (da): a conventional vernier caliper of suitable size can be use to determine the distance between the two outer extremities of external gear teeth in position diametrically opposed. the measure will always be more accurate in gears with even number of teeth, but it is also practically applicable in gears with odd quantity of teeth, always better in gears with large number of teeth. facewidth (b): it is the width over the toothed part of a gear, measured along a generator of the reference cylinder. the measurement can be made using a vernier caliper, although it can be enough a simple rule with precision of millimetres. base tangent length (wk): the measurement is made over a group of teeth using a conventional vernier calliper or plate micrometer. for a good results is required that the controlled flanks are perfectly clean and without appreciable wear. moreover, the calliper jaws must penetrate sufficiently into two tooth spaces to make tangent contact with the tooth surfaces without interfering with the teeth adjoining the span measurement. thus, the measurement of the distance between two parallel planes tangent to the outer flanks of a number of consecutive teeth, along a line tangent to the base cylinder, is taken. in case of not considering space between non-working flanks of the mating gears when the working flanks are in contact (zero backlash), the distance measured is equal to the normal thickness of one tooth at the base cylinder sbn plus the product of the number of teeth spanned less one (k -1) and the normal base pith pbn , see equation (9). suffixes k1 (for pinion) and k2 (for gear) after the letter w specifies the number of teeth between the flanks measured. figures 2 and 3 illustrate the span measurement applied to spur and helical gears. 4 3.0 initial data and measurements in the proposed procedure, to calculate the fundamental gear tooth data of an external parallel-axis cylindrical involute gear, it is necessary to know the following parameters:  number of teeth on pinion and gear (z1, z2)  tip diameters on pinion and gear (da1, da2) in mm  facewidth on pinion and gear (b1, b2) in mm  base tangent length spanned in k teeth on pinion and gear (wk1, wk2) in mm  number of teeth spanned for the base tangent length on pinion and gear (k1, k2)  tooth depth on pinion and gear (h1, h2) in mm  centre distance (aw) in mm  helix angle at tip diameter (a) in degree number of teeth (z): special care should be had counting the quantity of teeth in the gears. it is a good practice to make some mark with chalk in the tooth where the count begins to assure that the number of teeth was correctly determined. an incorrect specification of the number of teeth on gears will be catastrophic in the next calculation. tip diameters (da): a conventional vernier caliper of suitable size can be use to determine the distance between the two outer extremities of external gear teeth in position diametrically opposed. the measure will always be more accurate in gears with even number of teeth, but it is also practically applicable in gears with odd quantity of teeth, always better in gears with large number of teeth. facewidth (b): it is the width over the toothed part of a gear, measured along a generator of the reference cylinder. the measurement can be made using a vernier caliper, although it can be enough a simple rule with precision of millimetres. base tangent length (wk): the measurement is made over a group of teeth using a conventional vernier calliper or plate micrometer. for a good results is required that the controlled flanks are perfectly clean and without appreciable wear. moreover, the calliper jaws must penetrate sufficiently into two tooth spaces to make tangent contact with the tooth surfaces without interfering with the teeth adjoining the span measurement. thus, the measurement of the distance between two parallel planes tangent to the outer flanks of a number of consecutive teeth, along a line tangent to the base cylinder, is taken. in case of not considering space between non-working flanks of the mating gears when the working flanks are in contact (zero backlash), the distance measured is equal to the normal thickness of one tooth at the base cylinder sbn plus the product of the number of teeth spanned less one (k -1) and the normal base pith pbn , see equation (9). suffixes k1 (for pinion) and k2 (for gear) after the letter w specifies the number of teeth between the flanks measured. figures 2 and 3 illustrate the span measurement applied to spur and helical gears. )1(  kpsw bnbntk (9) 5 figure 2. measurement of base tangent length over 3 teeth in spur gear figure 3. measurement of base tangent length over 3 teeth in helical gear (top view) on external parallel-axis cylindrical involute gear, the actual base tangent lengths (wk1 and wk2) are less than the theoretical dimensions for zero backlash by the necessary amount of the normal backlash allowance, but this fact doesn't affect the practical results because standard values of gear backlash (iso/tr 10064-2, 1996) are relatively small (not bigger than 3 or 7 % of module) for industrial drives with typical commercial manufacturing tolerances. in gear with profile or helix modifications, the span measurement should be carried out on the un-modified part of the tooth flank. moreover, in some case, span measurement cannot be applied when a combination of high helix angle and narrow facewidth prevent the caliper from spanning a sufficient number of teeth, see equation (10). in this situation should be considered other alternative procedures to determine the unknown gear geometry using conventional measurement tools (regalado, 2000) or exhaustive search method with a trial and error procedure to obtain an approximate geometry of the analyzed gears. bkwb sin015,1min  (10) where: bmin: minimum value for facewidth in mm. there is an additional value of 1,5% to make an stable span measurement. figure 2. measurement of base tangent length over 3 teeth in spur gear issn: 2180-1053 vol. 7 no. 2 july december 2015 journal of mechanical engineering and technology 62 5 figure 2. measurement of base tangent length over 3 teeth in spur gear figure 3. measurement of base tangent length over 3 teeth in helical gear (top view) on external parallel-axis cylindrical involute gear, the actual base tangent lengths (wk1 and wk2) are less than the theoretical dimensions for zero backlash by the necessary amount of the normal backlash allowance, but this fact doesn't affect the practical results because standard values of gear backlash (iso/tr 10064-2, 1996) are relatively small (not bigger than 3 or 7 % of module) for industrial drives with typical commercial manufacturing tolerances. in gear with profile or helix modifications, the span measurement should be carried out on the un-modified part of the tooth flank. moreover, in some case, span measurement cannot be applied when a combination of high helix angle and narrow facewidth prevent the caliper from spanning a sufficient number of teeth, see equation (10). in this situation should be considered other alternative procedures to determine the unknown gear geometry using conventional measurement tools (regalado, 2000) or exhaustive search method with a trial and error procedure to obtain an approximate geometry of the analyzed gears. bkwb sin015,1min  (10) where: bmin: minimum value for facewidth in mm. there is an additional value of 1,5% to make an stable span measurement. figure 3. measurement of base tangent length over 3 teeth in helical gear (top view) on external parallel-axis cylindrical involute gear, the actual base tangent lengths (wk1 and wk2) are less than the theoretical dimensions for zero backlash by the necessary amount of the normal backlash allowance, but this fact doesn’t affect the practical results because standard values of gear backlash (iso/tr 10064-2, 1996) are relatively small (not bigger than 3 or 7 % of module) for industrial drives with typical commercial manufacturing tolerances. in gear with profile or helix modifications, the span measurement should be carried out on the un-modified part of the tooth flank. moreover, in some case, span measurement cannot be applied when a combination of high helix angle and narrow facewidth prevent the caliper from spanning a sufficient number of teeth, see equation (10). in this situation should be considered other alternative procedures to determine the unknown gear geometry using conventional measurement tools (regalado, 2000) or exhaustive search method with a trial and error procedure to obtain an approximate geometry of the analyzed gears. 5 figure 2. measurement of base tangent length over 3 teeth in spur gear figure 3. measurement of base tangent length over 3 teeth in helical gear (top view) on external parallel-axis cylindrical involute gear, the actual base tangent lengths (wk1 and wk2) are less than the theoretical dimensions for zero backlash by the necessary amount of the normal backlash allowance, but this fact doesn't affect the practical results because standard values of gear backlash (iso/tr 10064-2, 1996) are relatively small (not bigger than 3 or 7 % of module) for industrial drives with typical commercial manufacturing tolerances. in gear with profile or helix modifications, the span measurement should be carried out on the un-modified part of the tooth flank. moreover, in some case, span measurement cannot be applied when a combination of high helix angle and narrow facewidth prevent the caliper from spanning a sufficient number of teeth, see equation (10). in this situation should be considered other alternative procedures to determine the unknown gear geometry using conventional measurement tools (regalado, 2000) or exhaustive search method with a trial and error procedure to obtain an approximate geometry of the analyzed gears. bkwb sin015,1min  (10) where: bmin: minimum value for facewidth in mm. there is an additional value of 1,5% to make an stable span measurement. where: bmin: minimum value for facewidth in mm. there is an additional value of 1,5% to make an stable span measurement. issn: 2180-1053 vol. 7 no. 2 july december 2015 analitical method to calculate the unknown geometry of cylindrical gears 63 number of teeth spanned for the base tangent length (k): in case of gears with tooth data specified, the number of teeth spanned for the base tangent length can be calculated (maag, 1990), but for gears with unknown geometry, the number of teeth between the measuring surfaces can be established so that the points of contact with vernier caliper or plate micrometer are roughly at mid tooth height. the number of teeth to be spanned will be larger for gears with larger numbers of teeth and for gears with higher helix angle. recommendations on table 1, based on author’s experiences and calculation of the base tangent length, can be used as guideline values of number of teeth for span measurement. for more detailed information about values of the number of teeth spanned for the base tangent length from the helix angle, the number of teeth, pressure angle and the addendum modification coefficient can be obtained in maag gear book. table 1. guideline for the number of teeth spanned for the base tangent length 6 number of teeth spanned for the base tangent length (k): in case of gears with tooth data specified, the number of teeth spanned for the base tangent length can be calculated (maag, 1990), but for gears with unknown geometry, the number of teeth between the measuring surfaces can be established so that the points of contact with vernier caliper or plate micrometer are roughly at mid tooth height. the number of teeth to be spanned will be larger for gears with larger numbers of teeth and for gears with higher helix angle. recommendations on table 1, based on author's experiences and calculation of the base tangent length, can be used as guideline values of number of teeth for span measurement. for more detailed information about values of the number of teeth spanned for the base tangent length from the helix angle, the number of teeth, pressure angle and the addendum modification coefficient can be obtained in maag gear book. table 1. guideline for the number of teeth spanned for the base tangent length helix angle at a tip diameter (a) number of teeth between the measuring surfaces so that the points of contact are roughly at mid tooth height (k). 2 3 4 5 6 7 8 9 10 11 12 13 number of teeth (z) 0 11 18 18 28 28 36 36 44 44 55 55 65 65 75 75 85 85 95 95 100 100 110 110 120 20 10 16 16 24 24 30 30 42 42 48 48 55 55 65 65 75 75 85 85 95 95 100 100 110 40 6 9 9 14 14 18 18 24 24 28 28 32 32 36 36 42 42 46 46 50 50 54 54 60 tooth depth (h): this magnitude is usually specified as the radial distance between the tip and root diameters. tooth depth may be measured by means of a gear tooth vernier caliper or in tooth space using a simple vernier caliper with blade for depth measurements. the calliper blade must penetrate sufficiently and to make contact with the surface at the bottom of a tooth space without interfering with adjacent teeth flanks. centre distance (aw): involute gears can operate correctly with small change of centre distance according with the proper tolerances for deviations, but assembled gears with incorrect operating centre distance will not operating properly, for that reason, the centre distance should be determined with a good precision. this magnitude is accepted as the shortest distance between the axes of a gear pair and this is also the distance between the axes of shafts that are carrying the gears. a common method to determine the gear centre distance is the measurement in parallel planes of the center holes distance located in their functional shafts, but taking into account the accuracy of cylindrical bearing seatings on shafts and in housing bores, a more satisfactory method is consider the nominal centre distance as the sum of the housing bores radii (or outer radii of bearings) plus the distance between them. figure 4 and equation (11) show this idea. usually, speed reducers and enclosed gear units boxes have specified the nominal centre distance based on series of preferred numbers (is0 standard 3, 1973) and checking it may provide clues to nominal value of the centre distance. trraw  21 (11) tooth depth (h): this magnitude is usually specified as the radial distance between the tip and root diameters. tooth depth may be measured by means of a gear tooth vernier caliper or in tooth space using a simple vernier caliper with blade for depth measurements. the calliper blade must penetrate sufficiently and to make contact with the surface at the bottom of a tooth space without interfering with adjacent teeth flanks. centre distance (aw): involute gears can operate correctly with small change of centre distance according with the proper tolerances for deviations, but assembled gears with incorrect operating centre distance will not operating properly, for that reason, the centre distance should be determined with a good precision. this magnitude is accepted as the shortest distance between the axes of a gear pair and this is also the distance between the axes of shafts that are carrying the gears. issn: 2180-1053 vol. 7 no. 2 july december 2015 journal of mechanical engineering and technology 64 a common method to determine the gear centre distance is the measurement in parallel planes of the center holes distance located in their functional shafts, but taking into account the accuracy of cylindrical bearing seatings on shafts and in housing bores, a more satisfactory method is consider the nominal centre distance as the sum of the housing bores radii (or outer radii of bearings) plus the distance between them. figure 4 and equation (11) show this idea. usually, speed reducers and enclosed gear units boxes have specified the nominal centre distance based on series of preferred numbers (is0 standard 3, 1973) and checking it may provide clues to nominal value of the centre distance. 6 number of teeth spanned for the base tangent length (k): in case of gears with tooth data specified, the number of teeth spanned for the base tangent length can be calculated (maag, 1990), but for gears with unknown geometry, the number of teeth between the measuring surfaces can be established so that the points of contact with vernier caliper or plate micrometer are roughly at mid tooth height. the number of teeth to be spanned will be larger for gears with larger numbers of teeth and for gears with higher helix angle. recommendations on table 1, based on author's experiences and calculation of the base tangent length, can be used as guideline values of number of teeth for span measurement. for more detailed information about values of the number of teeth spanned for the base tangent length from the helix angle, the number of teeth, pressure angle and the addendum modification coefficient can be obtained in maag gear book. table 1. guideline for the number of teeth spanned for the base tangent length helix angle at a tip diameter (a) number of teeth between the measuring surfaces so that the points of contact are roughly at mid tooth height (k). 2 3 4 5 6 7 8 9 10 11 12 13 number of teeth (z) 0 11 18 18 28 28 36 36 44 44 55 55 65 65 75 75 85 85 95 95 100 100 110 110 120 20 10 16 16 24 24 30 30 42 42 48 48 55 55 65 65 75 75 85 85 95 95 100 100 110 40 6 9 9 14 14 18 18 24 24 28 28 32 32 36 36 42 42 46 46 50 50 54 54 60 tooth depth (h): this magnitude is usually specified as the radial distance between the tip and root diameters. tooth depth may be measured by means of a gear tooth vernier caliper or in tooth space using a simple vernier caliper with blade for depth measurements. the calliper blade must penetrate sufficiently and to make contact with the surface at the bottom of a tooth space without interfering with adjacent teeth flanks. centre distance (aw): involute gears can operate correctly with small change of centre distance according with the proper tolerances for deviations, but assembled gears with incorrect operating centre distance will not operating properly, for that reason, the centre distance should be determined with a good precision. this magnitude is accepted as the shortest distance between the axes of a gear pair and this is also the distance between the axes of shafts that are carrying the gears. a common method to determine the gear centre distance is the measurement in parallel planes of the center holes distance located in their functional shafts, but taking into account the accuracy of cylindrical bearing seatings on shafts and in housing bores, a more satisfactory method is consider the nominal centre distance as the sum of the housing bores radii (or outer radii of bearings) plus the distance between them. figure 4 and equation (11) show this idea. usually, speed reducers and enclosed gear units boxes have specified the nominal centre distance based on series of preferred numbers (is0 standard 3, 1973) and checking it may provide clues to nominal value of the centre distance. trraw  21 (11) 7 figure 4. parameters for calculation of centre distance (aw) by means of center holes distance or bearing housing bores radii (r1 + r2) plus the distance (t) between them helix angle at tip diameter (a): for spur gears  = a =0º, because the helix is a straight line parallel to its rotating axis, but in case of helical gears the measuring of the helix angle at reference diameter is one of the most difficult of specifying and should be done with an special helix angle tester. when a helix angle measuring is not possible with these special equipments, the helix angle at reference diameter can get by a simple method based in the approximate measured of the helix angle at tip diameter (a) with results not too exact, but practically acceptable. for this, it is necessary apply a marking compound to the tip surface of external gear teeth and roll the helical gear in straight line on a white paper to collect their generated trace (see figure 5). figure 5. measuring of outer helix angle a by their generated trace figure 4. parameters for calculation of centre distance (aw) by means of center holes distance or bearing housing bores radii (r1 + r2) plus the distance (t) between them helix angle at tip diameter (βa): for spur gears β = βa =0°, because the helix is a straight line parallel to its rotating axis, but in case of helical gears the measuring of the helix angle at reference diameter is one of the most difficult of specifying and should be done with an special helix angle tester. when a helix angle measuring is not possible with these special equipments, the helix angle at reference diameter can get by a simple method based in the approximate measured of the helix angle at tip diameter (βa) with results not too exact, but practically acceptable. for this, it is necessary apply a marking compound to the tip surface of external gear teeth and roll the helical gear in straight line on a white paper to collect their generated trace (see figure 5). issn: 2180-1053 vol. 7 no. 2 july december 2015 analitical method to calculate the unknown geometry of cylindrical gears 65 7 figure 4. parameters for calculation of centre distance (aw) by means of center holes distance or bearing housing bores radii (r1 + r2) plus the distance (t) between them helix angle at tip diameter (a): for spur gears  = a =0º, because the helix is a straight line parallel to its rotating axis, but in case of helical gears the measuring of the helix angle at reference diameter is one of the most difficult of specifying and should be done with an special helix angle tester. when a helix angle measuring is not possible with these special equipments, the helix angle at reference diameter can get by a simple method based in the approximate measured of the helix angle at tip diameter (a) with results not too exact, but practically acceptable. for this, it is necessary apply a marking compound to the tip surface of external gear teeth and roll the helical gear in straight line on a white paper to collect their generated trace (see figure 5). figure 5. measuring of outer helix angle a by their generated trace figure 5. measuring of outer helix angle βa by their generated trace 4.0 determination of the unknown gear geometry the output results of the unknown gear have strong relation with the measured values and depending of uncertainty of the measuring and including all manufacturing errors, wear and deformation on flanks in the gear itself. it is important understand this concept because modules, pressure angles, helix angles, addendum modification coefficient and other gear geometry features are given at calculated values and they are not necessarily the values used in the initial manufacturing of the gears, but they are very useful as reference to establish the fundamental parameters for reproduction of new gears or evaluation of the load capacity of gears. with the initial data and measurements above mentioned, fundamental gear geometry parameters according to iso standards can be obtained applying the following calculations. normal module (m) the module m in the normal section of the gear is the same module m of the standard basic rack tooth profile (is0 standard 53, 1998) and is defined as the quotient of the pitch “p” (distance measured over the reference circle from a point on one tooth to the corresponding point on the adjacent tooth of the gear), expressed in millimetres, to the number π. 8 4.0 determination of the unknown gear geometry the output results of the unknown gear have strong relation with the measured values and depending of uncertainty of the measuring and including all manufacturing errors, wear and deformation on flanks in the gear itself. it is important understand this concept because modules, pressure angles, helix angles, addendum modification coefficient and other gear geometry features are given at calculated values and they are not necessarily the values used in the initial manufacturing of the gears, but they are very useful as reference to establish the fundamental parameters for reproduction of new gears or evaluation of the load capacity of gears. with the initial data and measurements above mentioned, fundamental gear geometry parameters according to iso standards can be obtained applying the following calculations. normal module (m) the module m in the normal section of the gear is the same module m of the standard basic rack tooth profile (is0 standard 53, 1998) and is defined as the quotient of the pitch “p” (distance measured over the reference circle from a point on one tooth to the corresponding point on the adjacent tooth of the gear), expressed in millimetres, to the number . m p m  (12) the module is a commonly referenced gear parameter in the iso gear system and very important to defined the size of gear tooth. the module cannot be measured directly from a gear; yet, it is a common referenced value. tooling for commercially available cylindrical gears are stocked in standardized modules (iso standard 54, 1996) (ansi/agma 1102-a03, 2003). generally, when gear generation has ended a perfect engagement between gear and its generating hob occurs. thus, the normal module in the unknown gear geometry may be determined by a simple search of gear generating hob with known module which has a perfect mating with the analyzed gear, but this procedure requires of a complete set of generating hob to give solutions and it is not economically desirable, especially when the measurement has to be done in the field. moreover, the normal module could be determined using a more practical procedure based on the difference between values of base tangent lengths over a consecutive number of teeth spanned and their relations with the normal base pitch. once the base tangent lengths have been measured, the value for reference of the normal module may be calculated applying equations (13) and (14) for pinion and gear respectively. since the values m1 and m2 need not be exactly precise can be taken for calculation propose a value of   20º.  cos 111 1    kk ww m (13)  cos 122 2    kk ww m (14) the module is a commonly referenced gear parameter in the iso gear system and very important to defined the size of gear tooth. the module cannot be measured directly from a gear; yet, it is a common referenced value. tooling for commercially available cylindrical gears are stocked in standardized modules (iso standard 54, 1996) (ansi/ agma 1102-a03, 2003). generally, when gear generation has ended a perfect engagement between gear and its generating hob occurs. issn: 2180-1053 vol. 7 no. 2 july december 2015 journal of mechanical engineering and technology 66 thus, the normal module in the unknown gear geometry may be determined by a simple search of gear generating hob with known module which has a perfect mating with the analyzed gear, but this procedure requires of a complete set of generating hob to give solutions and it is not economically desirable, especially when the measurement has to be done in the field. moreover, the normal module could be determined using a more practical procedure based on the difference between values of base tangent lengths over a consecutive number of teeth spanned and their relations with the normal base pitch. once the base tangent lengths have been measured, the value for reference of the normal module may be calculated applying equations (13) and (14) for pinion and gear respectively. since the values m1 and m2 need not be exactly precise can be taken for calculation propose a value of α ≅ 20°. 8 4.0 determination of the unknown gear geometry the output results of the unknown gear have strong relation with the measured values and depending of uncertainty of the measuring and including all manufacturing errors, wear and deformation on flanks in the gear itself. it is important understand this concept because modules, pressure angles, helix angles, addendum modification coefficient and other gear geometry features are given at calculated values and they are not necessarily the values used in the initial manufacturing of the gears, but they are very useful as reference to establish the fundamental parameters for reproduction of new gears or evaluation of the load capacity of gears. with the initial data and measurements above mentioned, fundamental gear geometry parameters according to iso standards can be obtained applying the following calculations. normal module (m) the module m in the normal section of the gear is the same module m of the standard basic rack tooth profile (is0 standard 53, 1998) and is defined as the quotient of the pitch “p” (distance measured over the reference circle from a point on one tooth to the corresponding point on the adjacent tooth of the gear), expressed in millimetres, to the number . m p m  (12) the module is a commonly referenced gear parameter in the iso gear system and very important to defined the size of gear tooth. the module cannot be measured directly from a gear; yet, it is a common referenced value. tooling for commercially available cylindrical gears are stocked in standardized modules (iso standard 54, 1996) (ansi/agma 1102-a03, 2003). generally, when gear generation has ended a perfect engagement between gear and its generating hob occurs. thus, the normal module in the unknown gear geometry may be determined by a simple search of gear generating hob with known module which has a perfect mating with the analyzed gear, but this procedure requires of a complete set of generating hob to give solutions and it is not economically desirable, especially when the measurement has to be done in the field. moreover, the normal module could be determined using a more practical procedure based on the difference between values of base tangent lengths over a consecutive number of teeth spanned and their relations with the normal base pitch. once the base tangent lengths have been measured, the value for reference of the normal module may be calculated applying equations (13) and (14) for pinion and gear respectively. since the values m1 and m2 need not be exactly precise can be taken for calculation propose a value of   20º.  cos 111 1    kk ww m (13)  cos 122 2    kk ww m (14) although mating gears can have different base tangent lengths and number of teeth, mating gears must have the same module and pressure angle, for that reason the correct normal module for gear m should be established equal to the nearest standardized module to the values m1 and m2. table 2 can be used as guideline for values of standardized normal modules. table 2. standardized normal modules of cylindrical gears for general and heavy engineering 9 although mating gears can have different base tangent lengths and number of teeth, mating gears must have the same module and pressure angle, for that reason the correct normal module for gear m should be established equal to the nearest standardized module to the values m1 and m2. table 2 can be used as guideline for values of standardized normal modules. table 2. standardized normal modules of cylindrical gears for general and heavy engineering series i 1 1,25 1,5 2 2,5 3 4 5 6 ii 1,125 1,375 1,75 2,25 2,75 3,5 4,5 5,5 (6,5) series i 8 10 12 16 20 25 32 40 50 ii 7 9 11 14 18 22 28 36 45 note: preference should be given to the use of the normal modules as given in series i. the module 6,5 of series ii should be avoided. these normal modules are not necessary applicable to gears used in the automotive field. helix angle at reference diameter (). as it is known, in spur gears the helix angle at reference diameter is  = 0º. in case of helical gears the helix angle at reference diameter can be calculated based in the measured of the helix angle at tip diameter (a) as follows:          a a d zm sen   tan1 (15) nominal pressure angle (). it is an important characteristic of the standard basic rack tooth profile for cylindrical involute gears cutting by generating tool and constitutes a geometrical reference for involute gears in order to fix the sizes and profiles of their teeth. in general, gears are generating with a cutter normal profile angle chosen from the range between 14.5º and 25º. standard values for nominal pressure angle are 14.5º, 17.5º, 20º, 22.5º, and 25º. some gear manufacturers use non-standard cutter profile angles to accomplish specific design goals, in these case this method of reverse engineering can give some idea for recreating other new gear with standardized values of pressure angle. taking into account the sum of theoretical base tangent lengths of both toothed wheels (wtk = wtk1 + wtk2) the nominal pressure angle can be estimated. by means of mathematical processing of the equations (6), (8), (9) and (16) for pinion and gear is possible the determination of equation (17). in particular, equation (17) is relevant because the numerical values obtained are derived directly from the basic gear data specified previously and can be used as important factor in the decision making task.    2121 tan2 zz invinv xxx twt       (16)       cos1 212121  minvzzkkwww twtktktk (17) with note: preference should be given to the use of the normal modules as given in series i. the module 6,5 of series ii should be avoided. these normal modules are not necessary applicable to gears used in the automotive field. helix angle at reference diameter (β). as it is known, in spur gears the helix angle at reference diameter is β= 0°. in case of helical gears the helix angle at reference diameter can be calculated based in the measured of the helix angle at tip diameter (βa) as follows: issn: 2180-1053 vol. 7 no. 2 july december 2015 analitical method to calculate the unknown geometry of cylindrical gears 67 9 although mating gears can have different base tangent lengths and number of teeth, mating gears must have the same module and pressure angle, for that reason the correct normal module for gear m should be established equal to the nearest standardized module to the values m1 and m2. table 2 can be used as guideline for values of standardized normal modules. table 2. standardized normal modules of cylindrical gears for general and heavy engineering series i 1 1,25 1,5 2 2,5 3 4 5 6 ii 1,125 1,375 1,75 2,25 2,75 3,5 4,5 5,5 (6,5) series i 8 10 12 16 20 25 32 40 50 ii 7 9 11 14 18 22 28 36 45 note: preference should be given to the use of the normal modules as given in series i. the module 6,5 of series ii should be avoided. these normal modules are not necessary applicable to gears used in the automotive field. helix angle at reference diameter (). as it is known, in spur gears the helix angle at reference diameter is  = 0º. in case of helical gears the helix angle at reference diameter can be calculated based in the measured of the helix angle at tip diameter (a) as follows:          a a d zm sen   tan1 (15) nominal pressure angle (). it is an important characteristic of the standard basic rack tooth profile for cylindrical involute gears cutting by generating tool and constitutes a geometrical reference for involute gears in order to fix the sizes and profiles of their teeth. in general, gears are generating with a cutter normal profile angle chosen from the range between 14.5º and 25º. standard values for nominal pressure angle are 14.5º, 17.5º, 20º, 22.5º, and 25º. some gear manufacturers use non-standard cutter profile angles to accomplish specific design goals, in these case this method of reverse engineering can give some idea for recreating other new gear with standardized values of pressure angle. taking into account the sum of theoretical base tangent lengths of both toothed wheels (wtk = wtk1 + wtk2) the nominal pressure angle can be estimated. by means of mathematical processing of the equations (6), (8), (9) and (16) for pinion and gear is possible the determination of equation (17). in particular, equation (17) is relevant because the numerical values obtained are derived directly from the basic gear data specified previously and can be used as important factor in the decision making task.    2121 tan2 zz invinv xxx twt       (16)       cos1 212121  minvzzkkwww twtktktk (17) with nominal pressure angle (α). it is an important characteristic of the standard basic rack tooth profile for cylindrical involute gears cutting by generating tool and constitutes a geometrical reference for involute gears in order to fix the sizes and profiles of their teeth. in general, gears are generating with a cutter normal profile angle chosen from the range between 14.5° and 25°. standard values for nominal pressure angle are 14.5°, 17.5°, 20°, 22.5°, and 25°. some gear manufacturers use non-standard cutter profile angles to accomplish specific design goals, in these case this method of reverse engineering can give some idea for recreating other new gear with standardized values of pressure angle. taking into account the sum of theoretical base tangent lengths of both toothed wheels (∑wtk = wtk1 + wtk2) the nominal pressure angle can be estimated. by means of mathematical processing of the equations (6), (8), (9) and (16) for pinion and gear is possible the determination of equation (17). in particular, equation (17) is relevant because the numerical values obtained are derived directly from the basic gear data specified previously and can be used as important factor in the decision making task. 9 although mating gears can have different base tangent lengths and number of teeth, mating gears must have the same module and pressure angle, for that reason the correct normal module for gear m should be established equal to the nearest standardized module to the values m1 and m2. table 2 can be used as guideline for values of standardized normal modules. table 2. standardized normal modules of cylindrical gears for general and heavy engineering series i 1 1,25 1,5 2 2,5 3 4 5 6 ii 1,125 1,375 1,75 2,25 2,75 3,5 4,5 5,5 (6,5) series i 8 10 12 16 20 25 32 40 50 ii 7 9 11 14 18 22 28 36 45 note: preference should be given to the use of the normal modules as given in series i. the module 6,5 of series ii should be avoided. these normal modules are not necessary applicable to gears used in the automotive field. helix angle at reference diameter (). as it is known, in spur gears the helix angle at reference diameter is  = 0º. in case of helical gears the helix angle at reference diameter can be calculated based in the measured of the helix angle at tip diameter (a) as follows:          a a d zm sen   tan1 (15) nominal pressure angle (). it is an important characteristic of the standard basic rack tooth profile for cylindrical involute gears cutting by generating tool and constitutes a geometrical reference for involute gears in order to fix the sizes and profiles of their teeth. in general, gears are generating with a cutter normal profile angle chosen from the range between 14.5º and 25º. standard values for nominal pressure angle are 14.5º, 17.5º, 20º, 22.5º, and 25º. some gear manufacturers use non-standard cutter profile angles to accomplish specific design goals, in these case this method of reverse engineering can give some idea for recreating other new gear with standardized values of pressure angle. taking into account the sum of theoretical base tangent lengths of both toothed wheels (wtk = wtk1 + wtk2) the nominal pressure angle can be estimated. by means of mathematical processing of the equations (6), (8), (9) and (16) for pinion and gear is possible the determination of equation (17). in particular, equation (17) is relevant because the numerical values obtained are derived directly from the basic gear data specified previously and can be used as important factor in the decision making task.    2121 tan2 zz invinv xxx twt       (16)       cos1 212121  minvzzkkwww twtktktk (17) with with 10            cos tan tan 1t                cos2 cos cos 211 w t tw a zzm wtwtwtinv   )tan( where: wtk : sum of theoretical base tangent lengths of mating pinion and gear. wt: pressure angle at the pitch cylinder. t: transverse pressure angle to determine the nominal pressure angle in the unknown gear should be compared the sum of the theoretical base tangent lengths (wtk = wtk1 + wtk2) with the result of the sum of the measured base tangent lengths (wk = wk1 + wk2). thus the nominal pressure angle  must be estimated equal to the nearest standard value of pressure angle with smaller difference between the sum of the theoretical (wtk) and measured (wk) base tangent lengths of both gears. the starting value in the search should be 20°, since the majority of cutting tools use that angle conforming to world-wide acceptance. smaller pressure angles can be analyzed for case of gears with higher transverse contact ratios when lower noise levels are desirable, in this circumstances these gears usually have high numbers of teeth and lightly loaded. higher pressure angles are sometimes preferred for gears with lower numbers of teeth and heavily loaded when tooth bending strength is required. table 3 shows a sample of how to determine a nominal pressure angle. table 3. sample of the procedure to determine the standardized pressure angle by means of difference between the sum of the theoretical (wtk) and measured (wk) base tangent lengths of both gears basic gear data pinion (1): number of teeth z1 = 16 number of teeth between measured flanks k1 = 2 actual base tangent length (average) w2 = 13,88 mm wheel (2): number of teeth z2 = 83 number of teeth between measured flanks k2 = 10 actual base tangent length (average) w10 = 87,48 mm gear: m = 3 mm  = 8,11 aw = 150 mm nominal pressure angle () 14,5º 17,5º pressure angle at the pitch cylinder (tw) 14,64º 17,66º inv tw 0,00571 0,01015 sum of the theoretical base 102,01 101,75 where: ∑wtk : sum of theoretical base tangent lengths of mating pinion and gear. αwt : pressure angle at the pitch cylinder. αt : transverse pressure angle issn: 2180-1053 vol. 7 no. 2 july december 2015 journal of mechanical engineering and technology 68 to determine the nominal pressure angle in the unknown gear should be compared the sum of the theoretical base tangent lengths (∑wtk = wtk1 + wtk2) with the result of the sum of the measured base tangent lengths (∑wk = wk1 + wk2). thus the nominal pressure angle α must be estimated equal to the nearest standard value of pressure angle with smaller difference between the sum of the theoretical (∑wtk) and measured (∑wk) base tangent lengths of both gears. the starting value in the search should be 20°, since the majority of cutting tools use that angle conforming to world-wide acceptance. smaller pressure angles can be analyzed for case of gears with higher transverse contact ratios when lower noise levels are desirable, in this circumstances these gears usually have high numbers of teeth and lightly loaded. higher pressure angles are sometimes preferred for gears with lower numbers of teeth and heavily loaded when tooth bending strength is required. table 3 shows a sample of how to determine a nominal pressure angle. table 3. sample of the procedure to determine the standardized pressure angle by means of difference between the sum of the theoretical (∑wtk) and measured (∑wk) base tangent lengths of both gears 10            cos tan tan 1t                cos2 cos cos 211 w t tw a zzm wtwtwtinv   )tan( where: wtk : sum of theoretical base tangent lengths of mating pinion and gear. wt: pressure angle at the pitch cylinder. t: transverse pressure angle to determine the nominal pressure angle in the unknown gear should be compared the sum of the theoretical base tangent lengths (wtk = wtk1 + wtk2) with the result of the sum of the measured base tangent lengths (wk = wk1 + wk2). thus the nominal pressure angle  must be estimated equal to the nearest standard value of pressure angle with smaller difference between the sum of the theoretical (wtk) and measured (wk) base tangent lengths of both gears. the starting value in the search should be 20°, since the majority of cutting tools use that angle conforming to world-wide acceptance. smaller pressure angles can be analyzed for case of gears with higher transverse contact ratios when lower noise levels are desirable, in this circumstances these gears usually have high numbers of teeth and lightly loaded. higher pressure angles are sometimes preferred for gears with lower numbers of teeth and heavily loaded when tooth bending strength is required. table 3 shows a sample of how to determine a nominal pressure angle. table 3. sample of the procedure to determine the standardized pressure angle by means of difference between the sum of the theoretical (wtk) and measured (wk) base tangent lengths of both gears basic gear data pinion (1): number of teeth z1 = 16 number of teeth between measured flanks k1 = 2 actual base tangent length (average) w2 = 13,88 mm wheel (2): number of teeth z2 = 83 number of teeth between measured flanks k2 = 10 actual base tangent length (average) w10 = 87,48 mm gear: m = 3 mm  = 8,11 aw = 150 mm nominal pressure angle () 14,5º 17,5º pressure angle at the pitch cylinder (tw) 14,64º 17,66º inv tw 0,00571 0,01015 sum of the theoretical base 102,01 101,75 11 tangent lengths (wtk = wtk1 + wtk2); mm sum of the actual base tangent lengths (wk= wk1 + wk2); mm 101,36 difference between theoretical (wtk) and measured (wk); mm 0,65 0,39 estimated value of standardized pressure angle  = 20º addendum modification coefficient (x1, x2) the profile shift is the amount that is added to, or subtracted from, the gear teeth addendum to enhance the operational performance of the gear mating or meet fixed design criteria. for specialists involved with gear design based on iso standards, it’s very familiar that the datum line of the basic rack profile need not necessary be tangent to the reference diameter on gear, thus the tooth profile and his shape can be modified by shifting the datum line from the tangential position (gonzález rey, g. et al, 2006). the main parameter to evaluate the addendum modification is the addendum modification coefficient x, also know by american as profile shift factor or rack shift coefficient. the addendum modification coefficients for pinion (x1) and gear (x2) can be estimated by equations (18) and (19) obtained by consideration of normal backlash and mathematical processing of the equations (6), (8), (9) and (15)               t bnk invzk m jw x   11 1 1 5,0costan2 1 (18)     1212 tan2 xzz invinv x twt       (19) where: jbn = normal backlash (mm). normal backlash is the shortest distance between non-working flanks of two gears when the working flanks are in contact. some backlash should be present in all gear meshes. it is required to assure that the non-driving flanks of the teeth do not make contact. backlash in a given mesh varies during operation as a result of changes in speed, temperature and load. the amount of backlash required depends on the size of the gears, their accuracy, mounting and the application. for purpose of this procedure, normal backlash is preferable measured with feeler gauges when gears are mounted in the housing under static conditions. when normal backlash can not be measured can be used table 4 as guideline of values of minimum backlash (iso/tr 10064-2, 1996) recommended for industrial drives with ferrous gears in ferrous housings, working at pitchline speeds less than 15 m/s, with typical commercial manufacturing tolerances for housings, shafts and bearings. issn: 2180-1053 vol. 7 no. 2 july december 2015 analitical method to calculate the unknown geometry of cylindrical gears 69 addendum modification coefficient (x1, x2) the profile shift is the amount that is added to, or subtracted from, the gear teeth addendum to enhance the operational performance of the gear mating or meet fixed design criteria. for specialists involved with gear design based on iso standards, it’s very familiar that the datum line of the basic rack profile need not necessary be tangent to the reference diameter on gear, thus the tooth profile and his shape can be modified by shifting the datum line from the tangential position (gonzález rey, g. et al, 2006). the main parameter to evaluate the addendum modification is the addendum modification coefficient x, also know by american as profile shift factor or rack shift coefficient. the addendum modification coefficients for pinion (x1) and gear (x2) can be estimated by equations (18) and (19) obtained by consideration of normal backlash and mathematical processing of the equations (6), (8), (9) and (15) 11 tangent lengths (wtk = wtk1 + wtk2); mm sum of the actual base tangent lengths (wk= wk1 + wk2); mm 101,36 difference between theoretical (wtk) and measured (wk); mm 0,65 0,39 estimated value of standardized pressure angle  = 20º addendum modification coefficient (x1, x2) the profile shift is the amount that is added to, or subtracted from, the gear teeth addendum to enhance the operational performance of the gear mating or meet fixed design criteria. for specialists involved with gear design based on iso standards, it’s very familiar that the datum line of the basic rack profile need not necessary be tangent to the reference diameter on gear, thus the tooth profile and his shape can be modified by shifting the datum line from the tangential position (gonzález rey, g. et al, 2006). the main parameter to evaluate the addendum modification is the addendum modification coefficient x, also know by american as profile shift factor or rack shift coefficient. the addendum modification coefficients for pinion (x1) and gear (x2) can be estimated by equations (18) and (19) obtained by consideration of normal backlash and mathematical processing of the equations (6), (8), (9) and (15)               t bnk invzk m jw x   11 1 1 5,0costan2 1 (18)     1212 tan2 xzz invinv x twt       (19) where: jbn = normal backlash (mm). normal backlash is the shortest distance between non-working flanks of two gears when the working flanks are in contact. some backlash should be present in all gear meshes. it is required to assure that the non-driving flanks of the teeth do not make contact. backlash in a given mesh varies during operation as a result of changes in speed, temperature and load. the amount of backlash required depends on the size of the gears, their accuracy, mounting and the application. for purpose of this procedure, normal backlash is preferable measured with feeler gauges when gears are mounted in the housing under static conditions. when normal backlash can not be measured can be used table 4 as guideline of values of minimum backlash (iso/tr 10064-2, 1996) recommended for industrial drives with ferrous gears in ferrous housings, working at pitchline speeds less than 15 m/s, with typical commercial manufacturing tolerances for housings, shafts and bearings. where: jbn = normal backlash (mm). normal backlash is the shortest distance between non-working flanks of two gears when the working flanks are in contact. some backlash should be present in all gear meshes. it is required to assure that the non-driving flanks of the teeth do not make contact. backlash in a given mesh varies during operation as a result of changes in speed, temperature and load. the amount of backlash required depends on the size of the gears, their accuracy, mounting and the application. for purpose of this procedure, normal backlash is preferable measured with feeler gauges when gears are mounted in the housing under static conditions. when normal backlash can not be measured can be used table 4 as guideline of values of minimum backlash (iso/tr 100642, 1996) recommended for industrial drives with ferrous gears in ferrous housings, working at pitchline speeds less than 15 m/s, with typical commercial manufacturing tolerances for housings, shafts and bearings. issn: 2180-1053 vol. 7 no. 2 july december 2015 journal of mechanical engineering and technology 70 table 4. recommended values (in mm) for minimum backlash jbn 12 table 4. recommended values (in mm) for minimum backlash jbn normal module (m); mm centre distance (aw); mm 50 100 200 400 800 1,5 0,09 0,11 2 0,10 0,12 0,15 3 0,12 0,14 0,17 0,24 5 0,18 0,21 0,28 8 0,24 0,27 0,34 0,47 12 0,35 0,42 0,55 equations (20) and (21), derived from the equation (5), give a possible cross-check for the estimated values of addendum modification coefficients if the normal tooth thicknesses on reference cylinder for pinion and gear (sn1 and sn2) are known.   tan2 2 1 1    m s x n (20)   tan2 2 2 2    m s x n (21) factor of radial clearance (c*) and factor of addendum (ha*) shape and geometrical parameters of the basic rack tooth profile for involute gears are setting by special standards (see table 5) in corresponding with the rack shaped tool (such as hobs or rack type cutters) used in the cutting of gear by means of generation methods. the dimensions of the standard basic rack tooth profile give information about standardized values of radial clearance and addendum as a multiple of the normal module. table 5. some standard values of basic rack tooth profile parameters.  ha * c* f * standard 20,0 1,00 0,25 0,380 iso 53:1998 20,0 1,00 0,25 0,300 20,0 1,00 0,25 0,250 20,0 1,00 0,40 0,39 20,0 1,00 0,25 0,300 agma 201.02-68 20,0 1,00 0,25 0,350 20,0 0,8 0,20 0,3 25,0 1,00 0,25 0,300 25,0 1,00 0,25 0,350 14,5 1,00 0,157 20,0 1,00 0,25 0,375 jis b 1701-72 20,0 1,00 0,25 0,400 gost 13755-68 equations (20) and (21), derived from the equation (5), give a possible cross-check for the estimated values of addendum modification coefficients if the normal tooth thicknesses on reference cylinder for pinion and gear (sn1 and sn2) are known. 12 table 4. recommended values (in mm) for minimum backlash jbn normal module (m); mm centre distance (aw); mm 50 100 200 400 800 1,5 0,09 0,11 2 0,10 0,12 0,15 3 0,12 0,14 0,17 0,24 5 0,18 0,21 0,28 8 0,24 0,27 0,34 0,47 12 0,35 0,42 0,55 equations (20) and (21), derived from the equation (5), give a possible cross-check for the estimated values of addendum modification coefficients if the normal tooth thicknesses on reference cylinder for pinion and gear (sn1 and sn2) are known.   tan2 2 1 1    m s x n (20)   tan2 2 2 2    m s x n (21) factor of radial clearance (c*) and factor of addendum (ha*) shape and geometrical parameters of the basic rack tooth profile for involute gears are setting by special standards (see table 5) in corresponding with the rack shaped tool (such as hobs or rack type cutters) used in the cutting of gear by means of generation methods. the dimensions of the standard basic rack tooth profile give information about standardized values of radial clearance and addendum as a multiple of the normal module. table 5. some standard values of basic rack tooth profile parameters.  ha * c* f * standard 20,0 1,00 0,25 0,380 iso 53:1998 20,0 1,00 0,25 0,300 20,0 1,00 0,25 0,250 20,0 1,00 0,40 0,39 20,0 1,00 0,25 0,300 agma 201.02-68 20,0 1,00 0,25 0,350 20,0 0,8 0,20 0,3 25,0 1,00 0,25 0,300 25,0 1,00 0,25 0,350 14,5 1,00 0,157 20,0 1,00 0,25 0,375 jis b 1701-72 20,0 1,00 0,25 0,400 gost 13755-68 factor of radial clearance (c*) and factor of addendum (ha*) shape and geometrical parameters of the basic rack tooth profile for involute gears are setting by special standards (see table 5) in corresponding with the rack shaped tool (such as hobs or rack type cutters) used in the cutting of gear by means of generation methods. the dimensions of the standard basic rack tooth profile give information about standardized values of radial clearance and addendum as a multiple of the normal module. table 5. some standard values of basic rack tooth profile parameters. 12 table 4. recommended values (in mm) for minimum backlash jbn normal module (m); mm centre distance (aw); mm 50 100 200 400 800 1,5 0,09 0,11 2 0,10 0,12 0,15 3 0,12 0,14 0,17 0,24 5 0,18 0,21 0,28 8 0,24 0,27 0,34 0,47 12 0,35 0,42 0,55 equations (20) and (21), derived from the equation (5), give a possible cross-check for the estimated values of addendum modification coefficients if the normal tooth thicknesses on reference cylinder for pinion and gear (sn1 and sn2) are known.   tan2 2 1 1    m s x n (20)   tan2 2 2 2    m s x n (21) factor of radial clearance (c*) and factor of addendum (ha*) shape and geometrical parameters of the basic rack tooth profile for involute gears are setting by special standards (see table 5) in corresponding with the rack shaped tool (such as hobs or rack type cutters) used in the cutting of gear by means of generation methods. the dimensions of the standard basic rack tooth profile give information about standardized values of radial clearance and addendum as a multiple of the normal module. table 5. some standard values of basic rack tooth profile parameters.  ha * c* f * standard 20,0 1,00 0,25 0,380 iso 53:1998 20,0 1,00 0,25 0,300 20,0 1,00 0,25 0,250 20,0 1,00 0,40 0,39 20,0 1,00 0,25 0,300 agma 201.02-68 20,0 1,00 0,25 0,350 20,0 0,8 0,20 0,3 25,0 1,00 0,25 0,300 25,0 1,00 0,25 0,350 14,5 1,00 0,157 20,0 1,00 0,25 0,375 jis b 1701-72 20,0 1,00 0,25 0,400 gost 13755-68 issn: 2180-1053 vol. 7 no. 2 july december 2015 analitical method to calculate the unknown geometry of cylindrical gears 71 the factor of radial clearance is the distance, along the line of centres, between the root surface of a gear and the tip surface of its mating gear given in relation to normal module. radial clearance is the same between the root surface and the tip surface for pinion and gear with the same tooth depth (see figure 6). 13 the factor of radial clearance is the distance, along the line of centres, between the root surface of a gear and the tip surface of its mating gear given in relation to normal module. radial clearance is the same between the root surface and the tip surface for pinion and gear with the same tooth depth (see figure 6). figure 6. identification of radial clearances between the teeth of gear meshes equations (22) and (23) can be used to determine the factor of radial clearances. for purpose of this procedure, radial clearances are preferable measured with gauges when gears are mounted in the housing under static conditions.    m hdadaa m c c w 22111* 5,0   (22)    m hdadaa m c c w 12122* 5,0   (23) equations (24) and (25), derived directly from the basic gear are given to estimate values of factor of addendum. 1 1 2 1 * 2 cos2 x m zmada h w a            (24) 2 2 1 2 * 2 cos2 x m zmada h w a            (25) since the majority of cutting tools use values of ha* = 1 and c* = 0,25, conforming to world-wide acceptance, these values should be analysed firstly in the searching. it is possible to found other non-standard cutter to accomplish specific purpose as ha* = 0,75 for stub gears or ha* = 1,25 for gears with deep teeth. in case of non-standard system of basic rack tooth profile, equations (22) to (25) can give some idea for recreating other new gear with standardized values. figure 6. identification of radial clearances between the teeth of gear meshes equations (22) and (23) can be used to determine the factor of radial clearances. for purpose of this procedure, radial clearances are preferable measured with gauges when gears are mounted in the housing under static conditions. 13 the factor of radial clearance is the distance, along the line of centres, between the root surface of a gear and the tip surface of its mating gear given in relation to normal module. radial clearance is the same between the root surface and the tip surface for pinion and gear with the same tooth depth (see figure 6). figure 6. identification of radial clearances between the teeth of gear meshes equations (22) and (23) can be used to determine the factor of radial clearances. for purpose of this procedure, radial clearances are preferable measured with gauges when gears are mounted in the housing under static conditions.    m hdadaa m c c w 22111* 5,0   (22)    m hdadaa m c c w 12122* 5,0   (23) equations (24) and (25), derived directly from the basic gear are given to estimate values of factor of addendum. 1 1 2 1 * 2 cos2 x m zmada h w a            (24) 2 2 1 2 * 2 cos2 x m zmada h w a            (25) since the majority of cutting tools use values of ha* = 1 and c* = 0,25, conforming to world-wide acceptance, these values should be analysed firstly in the searching. it is possible to found other non-standard cutter to accomplish specific purpose as ha* = 0,75 for stub gears or ha* = 1,25 for gears with deep teeth. in case of non-standard system of basic rack tooth profile, equations (22) to (25) can give some idea for recreating other new gear with standardized values. equations (24) and (25), derived directly from the basic gear are given to estimate values of factor of addendum. 13 the factor of radial clearance is the distance, along the line of centres, between the root surface of a gear and the tip surface of its mating gear given in relation to normal module. radial clearance is the same between the root surface and the tip surface for pinion and gear with the same tooth depth (see figure 6). figure 6. identification of radial clearances between the teeth of gear meshes equations (22) and (23) can be used to determine the factor of radial clearances. for purpose of this procedure, radial clearances are preferable measured with gauges when gears are mounted in the housing under static conditions.    m hdadaa m c c w 22111* 5,0   (22)    m hdadaa m c c w 12122* 5,0   (23) equations (24) and (25), derived directly from the basic gear are given to estimate values of factor of addendum. 1 1 2 1 * 2 cos2 x m zmada h w a            (24) 2 2 1 2 * 2 cos2 x m zmada h w a            (25) since the majority of cutting tools use values of ha* = 1 and c* = 0,25, conforming to world-wide acceptance, these values should be analysed firstly in the searching. it is possible to found other non-standard cutter to accomplish specific purpose as ha* = 0,75 for stub gears or ha* = 1,25 for gears with deep teeth. in case of non-standard system of basic rack tooth profile, equations (22) to (25) can give some idea for recreating other new gear with standardized values. since the majority of cutting tools use values of ha* = 1 and c* = 0,25, conforming to world-wide acceptance, these values should be analysed firstly in the searching. it is possible to found other non-standard cutter to accomplish specific purpose as ha* = 0,75 for stub gears or ha* = 1,25 for gears with deep teeth. in case of non-standard system of basic rack tooth profile, equations (22) to (25) can give some idea for recreating other new gear with standardized values. issn: 2180-1053 vol. 7 no. 2 july december 2015 journal of mechanical engineering and technology 72 5.0 conclusions the theory of the involute surface of the flank of a cylindrical gear can give information about basic gear tooth data needed to determine the unknown gear geometry. based in the mentioned theory, a procedure of reverse engineering to determine the basic geometry of external parallel-axis cylindrical involute gears has been presented. the proposed method can be used as an alternative procedure to determine the unknown gear geometry using conventional measurement tools. it is important to highlight that all results, more or less accurate, represent the estimated values of the gear mating, depending on the uncertainty of the measurement and including all manufacturing errors in the gear itself. this is an important concept because modules, pressure angles, helix angles, addendum modification coefficients and other gear geometry features determined using this method are given at estimated values and they are not necessarily the values used in the initial manufacturing of the gears, but they are very useful as reference to establish the fundamental parameters for the evaluation of the load capacity of cylindrical gear or the reproduction of a new gear pair. the method, based on author’s experiences in the analysis, recovery and conversion of helical and spur gears according to iso, agma and non-standard gear systems, proposes a practical method with results not too exact, but practically acceptable, to obtain by calculating and conventional measurement tools the fundamental parameters needed for the reproduction of a new cylindrical gears according to iso standards. references agma standard 910-c90. (1990). formats for fine-pitch gear specification data. virginia.usa: american gear manufacturers association, ansi/agma 1102-a03 (2003). tolerance specification for gear hobs. virginia. usa: american gear manufacturers association, belarifi, f., bayraktar, e., benamar, a.(2008).the reverse engineering to optimise the dimensional conical spur gear by cad. journal of achievements in materials and manufacturing engineering, 31(2), 429-433. gonzález rey, g. (1999). procedimiento para la obtención de los parámetros geométricos básicos de un engranaje cónico de dientes rectos. ingeniería mecánica, 2 (1), 23-31. issn: 2180-1053 vol. 7 no. 2 july december 2015 analitical method to calculate the unknown geometry of cylindrical gears 73 gonzález rey, g., frechilla fernández, p. and garcía martin, r. (2006). cilindrical gear conversions: iso to agma. gear solutions, march 2006, 22-29. grimsley, p. (2003). software solutions for unknown gear. gear solutions. june 2003, 16-23. innocenti, c. (2007). simple techniques for measuring the base helix angle of involute gears. proceedings of the 12th iftomm world congress, besançon, france, june 18-21, 2007 (pp. 406-412). is0 standard 3. (1973). preferred numbers. series of preferred numbers. iso. genève 20, switzerland. iso standard 1340 (1976). cylindrical gears. information to be given to the manufacturer by the purchaser in order to obtain the gear required. iso. genève 20, switzerland. iso standard 54. (1996). cylindrical gears for general and heavy engineering. modules. iso. genève 20, switzerland. iso/tr 10064-2. (1996). cylindrical gears. code of inspection practice. part 2: inspection related to radial composite deviations, runout, tooth thickness and backlash. iso. genève 20, switzerland. is0 standard 53. (1998). cylindrical gears for general and heavy engineering. standard basic rack tooth profile. iso. genève 20, switzerland. kumar, a., jain, p.k., and pathak, p.m. (2014). machine element reconstruction using integrated reverse engineering and rapid prototyping approach. proceedings of the 26th all india manufacturing technology, design and research conference (aimtdr 2014). iit guwahati, assam, india. december 12th–14th, 2014 (pp. 123-1, 123-5) maag (1990). maag gear book. zurich-switzerland: maag gear company ltd. norma nc 02-04-04 (1978). reglas para elaborar los planos de trabajo de las ruedas dentadas cilíndricas. la habana. cuba. schultz, c. d. (2010). reverse engineering. proceedings of the agma fall technical meeting. agma technical paper 10ftm09. milwaukee, wisconsin, usa. _________________________________________ *corresponding author e-mail: pathiyasseril@yahoo.com issn: 2180-1053 vol. 8 no. 2 july – december 2016 81 optimization of resistance spot welding process parameters using moora approach p.sreeraj1* 1laxmi department of mechanical engineering, younus college of engineering and technology, kollam, kerala, india691010. abstract efforts optimization of resistance spot welding (rsw) process parameters was carried out to obtain optimal parametric combination to yield favorable weld nugget diameter, heat affected zone (haz) and breaking load in aisi 316 l austenitic stainless steel plates. taguchi’s l16 orthogonal array (oa) design and signaltonoise ratio (s/n ratio) have been used in this study. weld nugget diameter, heat affected zone (haz) and breaking load are selected as objective functions. in this case the multi objective optimization on the basis of ratio analysis (moora) is applied to solve this multi objective, problem. moora in combination with standard deviation (sdv) was used for optimization process. standard deviation (sdv) was used to determine the weights that were used for normalizing the responses obtained from the experimental results. it was found that welding current of 14 ka, welding time 14 cycle, electrode force 200kgf and holding time 10 cycle produced the weldment with the best mechanical properties. this method can be used successfully in other welding applications. keywords: rsw; sdv; orthogonal array; moora; haz 1.0 introduction resistance spot welding (rsw) is a multi factor, multi objective metal joining process, in which several process control parameters interact in a complicated manner and influence quality of weld. in most resistance spot welding (rsw) the weld quality is judged by nugget size, heat affected zone (haz) and joint strength. so it is important to select the welding process parameters to get the desired quality of the weld. usually, the selection of the desired process parameters is selected by trial and error. this is time consuming costly and may not be accurate. this does not ensure optimum weld nugget and other properties to ensure a proper weld. in order to overcome this problem various optimization techniques are used so that a perfect relationship between input and output variables can be developed using mathematical relationship so that desired output can be predicted. there are many research work done in modelling and process optimization in rsw and other welding process like gas metal arc welding (gmaw) flux cored arc welding (fcaw) and tungsten inert gas welding (tig). thakur and nandedkar presented a systematic approach to determine effect of process parameters on tensile shear strength of journal of mechanical engineering and technology 82 issn: 2180-1053 vol. 8 no. 2 july– december 2016 resistance weld joining of austenitic stainless steel aisi 304 using taguchi method (thakur & nandedkar, 2010). joseph, william and odinikuku (2015) optimized gas metal arc welding parameters using moora approach. norasiah, yupiter, manurung and hafidzi (2012) optimized resistance spot welding parameters towards development of weld nugget zone and heat affected zone (haz) using multi objective taguchi method (mtm). in this study taguchi method coupled with sdv-moora method was used to optimize the welding process parameters used for resistance spot welding on aisi 316 l austenitic steel plates. sdv was standard deviation method used for determining the weight attached to each mechanical property. the traditional taguchi method cannot solve multi-objective optimization problems. in order to overcome this difficulty, the taguchi method coupled with moora analysis used to solve the optimization problem in this study. 2.0 moora method since standard deviation is applied to this study for unbiased allocation of weights. the importance of weights in solving multi criteria decision making (mcdm) cannot be over emphasized .to determine the standard deviation the range standardization wad done using equation (1) to transform different scales and units among various criteria in to common measurable units in order to compute weights. 𝑋𝑖𝑗 𝑖 = 𝑋𝑖𝑗−𝑚𝑖𝑛𝑋𝑖𝑗 𝑚𝑎𝑥𝑋𝑖𝑗−𝑚𝑖𝑛𝑋𝑖𝑗 (1) where max xij , min xij are the maximum and minimum values of criterion (j) respectively. the standard deviation is calculated for every criterion using equation (2). sdvj =√ 1 𝑚 ∑ (𝑋𝑖𝑗 − 𝑋𝑗 𝑖̅̅ ̅)𝑚𝑖=1 2 (2) where 𝑋𝑖𝑗 𝑖̅̅ ̅̅ is the mean of jth criterion after normalization and j=1, 2 ...n .after calculating for sdv for all criteria the next step is to determine the weights wj of criteria considered using equation (3). wj = 𝑆𝐷𝑉𝑗 ∑ 𝑆𝐷𝑉𝑗 𝑛 𝑗=1 (3) where i=1...m; j=1 ...n. the multi objective optimization on the basis of moora method starts with a decision matrix as shown in equation (4): optimization of resistance spot welding process parameters using moora approach issn: 2180-1053 vol. 8 no. 2 july – december 2016 83 d = 𝐴1 𝐴2 𝐴3 ⋮ ⋮ 𝐴𝑛 1 2 3 11 12 13 1 21 22 23 2 31 32 33 3 1 2 3 n n n n m m m mn c c c c x x x x x x x x x x x x x x x x                    (4) step 1: compute the normalized decision matrix by vector method defined by equation (5) 𝑋𝑖𝑗 𝑖 = 𝑋𝑖𝑗 √∑ 𝑋𝑖𝑗 2𝑚 𝑖=1 (5) where i=1.....m; j=1...m step 2; calculate the composite score as expressed in equation (6) 𝑍𝑖 =∑ 𝑋𝑖𝑗 𝑖𝑏 𝑗=1 ∑ 𝑋𝑖𝑗 𝑖𝑛 𝑗=𝑏+1 ; where i=1 ...m (6) where ∑ 𝑋𝑖𝑗 𝑖𝑏 𝑗=1 and ∑ 𝑋𝑖𝑗 𝑖𝑛 𝑗=𝑏+1 are the benefit and non benefit criteria respectively .if there are some attributes more important than others, the composite score becomes as expressed in equation (7). 𝑍𝑖 = ∑ 𝑤𝑗 𝑏 𝑗=1 𝑋𝑖𝑗 𝑖 -∑ 𝑤𝑗 𝑛 𝑗=𝑏+1 𝑋𝑛 𝑖 i=1...m (7) where, wj is the weight of the jth criterion. step 3: rank the alternatives in descending order. 3.0 experimentation the sheets were cut parallel to the rolling direction. the dimension of austenitic stainless steel plate of grade aisi 316 l sheet are 140 mm length (l), 40 mm width (w) and 1 mm thick (t) shown in figure 1. overlap is equal to width of the sheet as per aws standard. sheet surfaces were chemically cleaned by acetone before resistance spot welding to eliminate surface contamination. the properties of base metal are shown in table 1. figure 1 shows kirperker rsw welding machine and fig. 2 shows sample specimen. journal of mechanical engineering and technology 84 issn: 2180-1053 vol. 8 no. 2 july– december 2016 figure 1. kirperker rsw welding machine table 1 chemical composition of base metal elements, weight % material c si mn p s al cr mo ni 316 l 0.030 0.75 2 0.045 0.03 0.1 figure 2. dimension of specimen 4.0 plan of investigation the research work is carried out in the following steps (tarng & yang, 1998). 1. identifying the quality characteristics and process parameters to be evaluated. 2. determine the number of levels for the process parameters and possible interactions between process parameters. 3. select appropriate orthogonal array and assign process parameters to the orthogonal array. 4. conduct experiment as per arrangement of orthogonal array. optimization of resistance spot welding process parameters using moora approach issn: 2180-1053 vol. 8 no. 2 july – december 2016 85 5. define problem. 6. selection of alternatives. 7. selection of the criteria describing alternatives. 8. determination of criteria values. 9. normalization of matrix. 10. determination of complex rationality. 11. ranking alternatives. 4.1 identification of factors and responses the weld nugget size, haz and breaking load has a significant effect on quality of resistance spot welding. the properties of the welding is the significantly influenced by diameter of weld nugget obtained. hence control of nugget diameter is important in resistance spot welding where a low diameter is highly desirable. the chosen factors have been selected on the basis to get minimal weld nugget diameter, low haz and higher breaking load. these are current, hold time; weld time and electrode force. the responses chosen were weld nugget diameter, haz and breaking load. the responses were chosen based on the impact of parameters on final composite model (gunaraj & murugan, 1999). 4.2 finding the limits of process variables working ranges of all selected factors are fixed by conducting trial run. this was carried out by varying one of factors while keeping the rest of them as constant values. working range of each process parameters was decided upon by inspecting the smooth appearance without any visible defects. the chosen level of the parameters with their units and notation are given in table 2. table 2. welding parameters and their levels parameters factor levels unit notation 1 2 3 4 welding current ka i 8 10 12 14 welding time cycle t 10 12 14 16 electrode force kgf f 180 200 220 240 holding time cycle c 10 20 30 40 journal of mechanical engineering and technology 86 issn: 2180-1053 vol. 8 no. 2 july– december 2016 4.3 development of orthogonal array design matrix chosen to conduct the experiments was taguchi’s orthogonal design. the design matrix comprises of l16 orthogonal array. sixteen experimental trails were conducted that make the estimation of nugget diameter, haz and breaking load (vermal et al., 2014). figure 3. scanned specimens table 3. design matrix trial number design matrix i t f c 1 1 1 1 1 2 1 2 2 2 3 1 3 3 3 4 1 4 4 4 5 2 1 2 3 6 2 2 1 4 7 2 3 4 1 8 2 4 3 2 9 3 1 3 4 10 3 2 4 3 11 3 3 1 2 12 3 4 2 1 13 4 1 4 2 14 4 2 3 1 15 4 3 2 4 16 4 4 1 3 optimization of resistance spot welding process parameters using moora approach issn: 2180-1053 vol. 8 no. 2 july – december 2016 87 i welding current; t welding time; f – electrode force; c – hold time 4.4 conducting experiments as per orthogonal array in this work sixteen experimental run were allowed as per orthogonal array correspond to each treatment combination of parameters on weld nugget diameter, haz and breaking load as shown table 3 at random. at each run settings for all parameters were disturbed and reset for next deposit. this is very essential to introduce variability caused by errors in experimental set up. figure 4. welded specimen 4.5 recording of responses for measuring the weld nugget diameter, toolmakers microscope is used. for conducting tensile test specimens were prepared as per asi 40 and specimen figure is shown in fig 2. the tensile test is conducted in a utm at younus college of engineering technology, kollam, kerala india. the observed values are shown in table 4. the tensile-shear test is the most widely used test for evaluating the spot weld mechanical behaviours in static condition. peak load, obtained from the tensile-shear load displacement curve, describes mechanical behaviour of spot welds. figure 3 shows scanned specimen and fig. 4 shows welded specimen. journal of mechanical engineering and technology 88 issn: 2180-1053 vol. 8 no. 2 july– december 2016 table 4. design matrix and observed values of weld nugget diameter, haz and max breaking load table 5. weights assigned to criteria property sdvj wj weld nugget diameter(mm) 0.28739 0.199576 max breaking load in kn 0.56175 0.390104 haz (mm) 0.59278 0.411653 trial no. design matrix bead parameters i t f c weld nugget diameter(mm) max breaking load in kn haz (mm) 1 1 1 1 1 7.306 18.81 1.072 2 1 2 2 2 8.243 19.54 0.8734 3 1 3 3 3 7.731 20.67 1.125 4 1 4 4 4 8.925 21.93 0.9238 5 2 1 2 3 8.792 18.44 0.8475 6 2 2 1 4 8.415 19.77 1.2581 7 2 3 4 1 6.777 19.18 0.8945 8 2 4 3 2 8.614 20.59 0.9765 9 3 1 3 4 8.908 21.53 1.1498 10 3 2 4 3 7.371 19.39 0.805 11 3 3 1 2 8.087 18.43 1.1689 12 3 4 2 1 8.112 20.52 0.986 13 4 1 4 2 9.125 19.42 1.0255 14 4 2 3 1 8.753 17.56 1.072 15 4 3 2 4 8.971 20.69 0.8734 16 4 4 1 3 8.807 19.24 1.125 optimization of resistance spot welding process parameters using moora approach issn: 2180-1053 vol. 8 no. 2 july – december 2016 89 table 6. the square value of xij table 7. normalized weld parameters bead parameters weld nugget diameter(mm) max breaking load in kn haz(mm) 1 53.37764 353.8161 1.787569 2 67.94705 381.8116 1.505529 3 59.76836 427.2489 0.880219 4 79.65563 480.9249 1.149184 5 77.29926 340.0336 0.762828 6 70.81223 390.8529 1.265625 7 45.92773 367.8724 0.853406 8 74.201 423.9481 0.718256 9 79.35246 463.5409 1.582816 10 54.33164 375.9721 0.80013 11 65.39957 339.6649 0.953552 12 65.80454 421.0704 1.32204 13 83.26563 377.1364 0.648025 14 76.61501 308.3536 1.366327 15 80.47884 428.0761 0.972196 16 77.56325 370.1776 1.05165 ∑ 𝑿𝒊𝒋 𝟐 𝒏 𝒊=𝟏 1111.8 9468.941 17.6193 √∑ 𝑋𝑖𝑗 2 𝑛 𝑖=1 33.343 97.308 4.1975 bead parameters weld nugget diameter max breaking load haz 1 0.219116 0.193304 0.318523 2 0.247218 0.200806 0.292317 3 0.231863 0.212418 0.223514 4 0.267672 0.225367 0.25539 5 0.263684 0.189501 0.208076 6 0.252377 0.203169 0.268017 7 0.203251 0.197106 0.220083 8 0.258345 0.211596 0.201906 journal of mechanical engineering and technology 90 issn: 2180-1053 vol. 8 no. 2 july– december 2016 table 8. clustered weld properties according to criteria 9 0.267163 0.221256 0.299726 10 0.221066 0.199264 0.213103 11 0.24254 0.189399 0.232638 12 0.243289 0.210877 0.273925 13 0.273671 0.199572 0.191781 14 0.262514 0.180458 0.278475 15 0.269052 0.212624 0.234902 16 0.264133 0.197723 0.244312 weights wj 0.333844 0.66155 0.411653 numbers (maximum) (minimum) (minimum) max breaking load in kn weld nugget diameter(mm) haz (mm) 1 0.12788 0.073159 0.131121 2 0.132843 0.082542 0.120333 3 0.140525 0.077415 0.09201 4 0.149092 0.089371 0.105132 5 0.125364 0.08804 0.085655 6 0.134406 0.084265 0.11033 7 0.130395 0.067862 0.090598 8 0.139981 0.086257 0.083115 9 0.146372 0.089201 0.123383 10 0.131823 0.07381 0.087725 1 0.125297 0.08098 0.095766 12 0.139506 0.08123 0.112762 13 0.132027 0.091374 0.078947 14 0.119382 0.087649 0.114635 15 0.140661 0.089832 0.096698 16 0.130804 0.08819 0.100572 optimization of resistance spot welding process parameters using moora approach issn: 2180-1053 vol. 8 no. 2 july – december 2016 91 table 9 ranking step 5.0 result analysis in this study the weight allocation for each output parameters, that is, the weld mechanical properties were determined. in determining the weights the range of standardized decision matrix is determined using equation (1). table 5 shows allocated weight. by applying the equation (5) table 6 and table 7 created. next step is to multiply the allocated weights to the values in table 7.this leads to the creation of table 8.the last step is to sum the parameters comparing higher the better and smaller the better values and table 9 is created and then parameters are ranked. rank number one determines the optimized condition. the nugget diameter considered in this study range from 6.7 mm to 9.2 mm. applying moora method the selected parameters produced a weld with nugget diameter 7.7 mm. breaking load considered in this study is within the range of 17.5 kn to 22 kn .by applying the moora method penetration is found to be 20.6 kn. haz considered in this study range from 1.2 mm to 0.8 mm. applying moora method the selected parameters produced a weld with haz 1.072 mm. bead parameters no ∑ 𝒎𝒂𝒙 ∑ 𝒎𝒊𝒏 ∑ 𝒎𝒂𝒙 -∑ 𝒎𝒊𝒏 rank 1 0.12788 0.20428 -0.0764 16 2 0.132843 0.202875 -0.07003 15 3 0.140525 0.169425 -0.0289 3 4 0.149092 0.194503 -0.04541 9 5 0.125364 0.173695 -0.04833 8 6 0.134406 0.194595 -0.06019 13 7 0.130395 0.15846 -0.02807 2 8 0.139981 0.169372 -0.02939 4 9 0.146372 0.212584 -0.06621 14 10 0.131823 0.161535 -0.02971 5 11 0.125297 0.176746 -0.05145 10 12 0.139506 0.193992 -0.05449 11 13 0.132027 0.170321 -0.03829 6 14 0.119382 0.202284 -0.0829 1 15 0.140661 0.18653 -0.04587 7 16 0.130804 0.188762 -0.05796 12 journal of mechanical engineering and technology 92 issn: 2180-1053 vol. 8 no. 2 july– december 2016 figure 5. weld structure of optimized model for this study weld sample 14 produced optimum weld. from table 3 it was found that welding current of 14 ka, welding time 14 cycle, electrode force 200kgf and holding time 10 cycle produced the weld with the best mechanical properties. i4t2f3c1 is the optimum process parameters obtained from this study.fig 5 represents the optimized condition. 6.0 conclusions in this study, a detailed methodology of moora technique has been presented for evaluating the nugget diameter, maximum breaking load and, haz and parametric combinations in resistance spot welding process. for achieving optimal parametric combination to get minimum nugget diameter, minimum haz and maximum breaking load of the weldment produced by resistance spot welding a multi objective optimization process is used. taguchi method coupled with moora analysis is very popular and efficient method for optimization that can be performed with limited number of runs. however standard deviation was used to determine the weights allocated to each value of mechanical property utilized in the course of running moora process. it is here by concluded that moora method has successfully optimized the process parameters considered in this study and microstructure of the optimized weldment agree that optimization result produced confirm the quality of the weldment. optimization of resistance spot welding process parameters using moora approach issn: 2180-1053 vol. 8 no. 2 july – december 2016 93 acknowledgements authors sincerely acknowledge the help and facilities extended to them by the department of mechanical engineering, younus college of engineering and technology, kollam, india. references gunaraj, v. & murugan, n. (1999). prediction and comparison of the area of the heat effected zone for the bead on plate and bead on joint in saw of pipes, journal of material processing technology, 95, 246 261. joseph achebo & william ejenavi odinikuku (2015) optimization of gas metal arc welding process parameters using standard deviation (sdv) and multiobjective optimization on the basis of ratio analysis , journal of minerals and materials characterization and engineering, 3, 298-308 norasiah m., yupiter hp m. & muhammed h. (2012). optimization and modelling of spot welding parameters with simultaneous response consideration using multi objective taguchi method and utility concept. journal of mechanical science and technology, 26 (8), 2365 2370. tarng, y.s & yang, w.h. (1998) optimization of weld bead geometry in gas tungsten arc welding by the taguchi method, international journal of advanced manufacturing technology, 14, 549-554. thakur, a.g & nandedkar, v.m. (2010) application of taguchi method to determine resistece spot welding conditions of austenitic stainless steel aisi 304, journal of scientific and industrial research , 69, 680-683. verma1 a.b., ghunage s.u., ahuja b.b. (2014). resistance welding of austenitic stainless steels (aisi 304 with aisi 316) 3, the 5th international & 26th all india manufacturing technology, design and research conference (aimtdr 2014) december 12th -14th, 2014, iit guwahati, assam, india. journal of mechanical engineering and technology 94 issn: 2180-1053 vol. 8 no. 2 july– december 2016 issn: 2180-1053 vol. 7 no. 1 january june 2015 propagation of stress wave in a functionally graded nano-bar based on modified couple stress theory 43 propagation of stress wave in a functionally graded nano-bar based on modified couple stress theory m. a. khorshidi1, m. shariati1* 1department of mechanical engineering, ferdowsi university of mashhad, mashhad, iran abstract in this paper, propagation of a one-dimensional elastic stress wave in a functionally graded (fg) nano-bar is analysed based on the modified couple stress theory. it is assumed that the material properties of fg bar are distributed as an exponential function along the axial direction. the two main advantages of the modified couple stress theory over the classical couple stress theory are the inclusion of a symmetric couple stress tensor and the involvement of only one material length scale parameter. according to the modified couple stress theory, only one material length scale parameter is used to describe the size effect in nano-bar. also, the shear stress components come from the lateral inertia effect are considered in the elastic strain energy relation. then, the governing equations are derived using hamilton’s principle and are generally solved. finally, effects of length scale parameter, material inhomogeneity constant and poisson’s ratio on stress wave propagation velocity and harmonic behavior of stress wave are evaluated and can be observed that using the classical continuum theory leads to considerable errors in analysis of stress wave propagation. keywords: nano-bar, modified couple stress theory, stress wave propagation, impact mechanic, functionally graded material 1.0 introduction analysis of the stress wave propagation is necessary to study structures subjected to the impact loading. therefore, the preliminary assumptions does not govern to these problems. several basic studies are accomplished on impact mechanics problems (fowles & williams, 1970; jones, 1989; stronge, 2000; qiao et al., 2008). however, the stress wave and generally imact problems are very important and applicable, but there are no enough studies and researchs in available about them. the one-dimensional bars are most common structure to analyse the stress wave propagation, which the stress wave propagates along the * corresponding author email: mshariati44@um.ac.ir issn: 2180-1053 vol. 7 no. 1 january june 2015 journal of mechanical engineering and technology 44 axial direction of them. (anderson, 2006) obtained the longitudinal stress wave propagation of an elastic bar by using higher order rod approximations. (shen & yin, 2014) presented the dynamic analysis of stress waves generated by impacts on non-uniform rod structures. (kaishin & bin, 2001) studied the dynamic behavior of a layered orthotropic bar with rectangular cross section due to impact torque. also, (shariat et al., 2010)studied on other geometry for impact analysis. they analysed the stress wave in thick-walled fg cylinder with temperature-dependent material properties. two main approaches usually use to analyse the longitudinal wave in bars. the first of these is called to be bernoulli-euler rod theory (elementary wave theory). this theory assumes that deformation occurs only in the longitudinal direction and that deformed planes remain orthogonal to the deformed bar axis. the second approach is known as love rod theory (love, 1944). in this thoery, addition to the assumptions of the elementary wave theory, it is assumed that the plane cross sections can expand or contract in their own planes. the love rod theory has more accuracy than bernoulli-euler rod theory, so, this theory is employed to describe the lateral inertia effects in the present study. when dimension of the structures becomes very small, accuracy of classical continuum theory is decreased. consequently, we should utilize especial theories (nonlocal theory, couple stress theory, surface effect theory) to model the small scale structures, mathematically. modified couple stress theory proposed by (yang et al., 2002) is one of these theories, which developed over the classical couple stress theory (mindlin, 1964). the modified couple stress theory is a quick and simple to mathematical modelling because makes use of only one material parameter to capture the size effect. also, this theory includes a symmetric couple stress tensor. several studies based on modified couple stress theory in the contexts of mechanical engineering reveal the exactness and capability of this theory (shaat et al., 2012; ke & wang, 2011; salamat-talab et al., 2012; thai & choi, 2013). since small scale (micro or nano) bars can be useful and applicable in small scale devices and systems such as biosensors, atomic force microscopes (afm), mems, and nems. but, study on stress wave propagation of nanostructures is rarely found. (guven, 2011, 2012, 2014) presented some solutions for propagation of stress wave in small scale bars under different situations and methods. issn: 2180-1053 vol. 7 no. 1 january june 2015 propagation of stress wave in a functionally graded nano-bar based on modified couple stress theory 45 this paper presents a modified couple stress based analysis for propagation of stress wave in longitudinally fg nano-bars using love rod theory and hamilton’s principle. the shear stress components are considered in total strain energy relation. finally, an explicit solution is obtained for the fg nano-bar, and effects of material length scale parameter, material inhomogeneity constant and poisson’s ratio on velocity of sress wave propagation and behavior of generated stress wave are evaluated. 2.0 computational method 2.1 functionally graded materials consider a solid bar with uniform cross section and area of a and length of l (see figure 1), which material properties such as young’s modulus and density vary on the basis of an exponential function along the axial (longitudinal) direction. 2 two main approaches usually use to analyse the longitudinal wave in bars. the first of these is called to be bernoulli-euler rod theory (elementary wave theory). this theory assumes that deformation occurs only in the longitudinal direction and that deformed planes remain orthogonal to the deformed bar axis. the second approach is known as love rod theory (love, 1944). in this thoery, addition to the assumptions of the elementary wave theory, it is assumed that the plane cross sections can expand or contract in their own planes. the love rod theory has more accuracy than bernoullieuler rod theory, so, this theory is employed to describe the lateral inertia effects in the present study. when dimension of the structures becomes very small, accuracy of classical continuum theory is decreased. consequently, we should utilize especial theories (nonlocal theory, couple stress theory, surface effect theory) to model the small scale structures, mathematically. modified couple stress theory proposed by (yang et al., 2002) is one of these theories, which developed over the classical couple stress theory (mindlin, 1964). the modified couple stress theory is a quick and simple to mathematical modelling because makes use of only one material parameter to capture the size effect. also, this theory includes a symmetric couple stress tensor. several studies based on modified couple stress theory in the contexts of mechanical engineering reveal the exactness and capability of this theory (shaat et al., 2012; ke & wang, 2011; salamat-talab et al., 2012; thai & choi, 2013). since small scale (micro or nano) bars can be useful and applicable in small scale devices and systems such as biosensors, atomic force microscopes (afm), mems, and nems. but, study on stress wave propagation of nanostructures is rarely found. (guven, 2011, 2012, 2014) presented some solutions for propagation of stress wave in small scale bars under different situations and methods. this paper presents a modified couple stress based analysis for propagation of stress wave in longitudinally fg nano-bars using love rod theory and hamilton's principle. the shear stress components are considered in total strain energy relation. finally, an explicit solution is obtained for the fg nano-bar, and effects of material length scale parameter, material inhomogeneity constant and poisson's ratio on velocity of sress wave propagation and behavior of generated stress wave are evaluated. 2.0 computational method 2.1 functionally graded materials consider a solid bar with uniform cross section and area of a and length of l (see figure 1), which material properties such as young's modulus and density vary on the basis of an exponential function along the axial (longitudinal) direction. (1) (2) where and are respectively young's modulus and density of the bar at the initial point of the bar (x=0). also, is material inhomogeneity constant. where 2 two main approaches usually use to analyse the longitudinal wave in bars. the first of these is called to be bernoulli-euler rod theory (elementary wave theory). this theory assumes that deformation occurs only in the longitudinal direction and that deformed planes remain orthogonal to the deformed bar axis. the second approach is known as love rod theory (love, 1944). in this thoery, addition to the assumptions of the elementary wave theory, it is assumed that the plane cross sections can expand or contract in their own planes. the love rod theory has more accuracy than bernoullieuler rod theory, so, this theory is employed to describe the lateral inertia effects in the present study. when dimension of the structures becomes very small, accuracy of classical continuum theory is decreased. consequently, we should utilize especial theories (nonlocal theory, couple stress theory, surface effect theory) to model the small scale structures, mathematically. modified couple stress theory proposed by (yang et al., 2002) is one of these theories, which developed over the classical couple stress theory (mindlin, 1964). the modified couple stress theory is a quick and simple to mathematical modelling because makes use of only one material parameter to capture the size effect. also, this theory includes a symmetric couple stress tensor. several studies based on modified couple stress theory in the contexts of mechanical engineering reveal the exactness and capability of this theory (shaat et al., 2012; ke & wang, 2011; salamat-talab et al., 2012; thai & choi, 2013). since small scale (micro or nano) bars can be useful and applicable in small scale devices and systems such as biosensors, atomic force microscopes (afm), mems, and nems. but, study on stress wave propagation of nanostructures is rarely found. (guven, 2011, 2012, 2014) presented some solutions for propagation of stress wave in small scale bars under different situations and methods. this paper presents a modified couple stress based analysis for propagation of stress wave in longitudinally fg nano-bars using love rod theory and hamilton's principle. the shear stress components are considered in total strain energy relation. finally, an explicit solution is obtained for the fg nano-bar, and effects of material length scale parameter, material inhomogeneity constant and poisson's ratio on velocity of sress wave propagation and behavior of generated stress wave are evaluated. 2.0 computational method 2.1 functionally graded materials consider a solid bar with uniform cross section and area of a and length of l (see figure 1), which material properties such as young's modulus and density vary on the basis of an exponential function along the axial (longitudinal) direction. (1) (2) where and are respectively young's modulus and density of the bar at the initial point of the bar (x=0). also, is material inhomogeneity constant. are respectively young’s modulus and density of the bar at the initial point of the bar 2 two main approaches usually use to analyse the longitudinal wave in bars. the first of these is called to be bernoulli-euler rod theory (elementary wave theory). this theory assumes that deformation occurs only in the longitudinal direction and that deformed planes remain orthogonal to the deformed bar axis. the second approach is known as love rod theory (love, 1944). in this thoery, addition to the assumptions of the elementary wave theory, it is assumed that the plane cross sections can expand or contract in their own planes. the love rod theory has more accuracy than bernoullieuler rod theory, so, this theory is employed to describe the lateral inertia effects in the present study. when dimension of the structures becomes very small, accuracy of classical continuum theory is decreased. consequently, we should utilize especial theories (nonlocal theory, couple stress theory, surface effect theory) to model the small scale structures, mathematically. modified couple stress theory proposed by (yang et al., 2002) is one of these theories, which developed over the classical couple stress theory (mindlin, 1964). the modified couple stress theory is a quick and simple to mathematical modelling because makes use of only one material parameter to capture the size effect. also, this theory includes a symmetric couple stress tensor. several studies based on modified couple stress theory in the contexts of mechanical engineering reveal the exactness and capability of this theory (shaat et al., 2012; ke & wang, 2011; salamat-talab et al., 2012; thai & choi, 2013). since small scale (micro or nano) bars can be useful and applicable in small scale devices and systems such as biosensors, atomic force microscopes (afm), mems, and nems. but, study on stress wave propagation of nanostructures is rarely found. (guven, 2011, 2012, 2014) presented some solutions for propagation of stress wave in small scale bars under different situations and methods. this paper presents a modified couple stress based analysis for propagation of stress wave in longitudinally fg nano-bars using love rod theory and hamilton's principle. the shear stress components are considered in total strain energy relation. finally, an explicit solution is obtained for the fg nano-bar, and effects of material length scale parameter, material inhomogeneity constant and poisson's ratio on velocity of sress wave propagation and behavior of generated stress wave are evaluated. 2.0 computational method 2.1 functionally graded materials consider a solid bar with uniform cross section and area of a and length of l (see figure 1), which material properties such as young's modulus and density vary on the basis of an exponential function along the axial (longitudinal) direction. (1) (2) where and are respectively young's modulus and density of the bar at the initial point of the bar (x=0). also, is material inhomogeneity constant. is material inhomogeneity constant. 3 figure 1. shematics of geometry of coordinate system. in many studies in the different contexts of solid mechanics, the radial distribution of the material properties is used. since in this paper, analysis of one-dimensional stress wave propagation along the axial direction of an elastic bar is considered, so, it is preferred that a longitudinal exponential function is used to describe the material distribution of the bar. 2.2 love rod theory as mentioned, according to love rod theory, although the rod cross-sections remain plane after deformation, but the plane cross-sections can expand or contract in their own planes. therefore, the following displacement field is assumed as: , , (3) where u, v and w are respectively the x-, yand z-components of the displacement on a point (x, y, z) on a bar cross-section. also, v is poisson's ratio. according to equation (3), the non-zero components of the strain and the stress are expressed as follow: , , , , (4) , , (5) figure 1. shematics of geometry of coordinate system. 3 figure 1. shematics of geometry of coordinate system. in many studies in the different contexts of solid mechanics, the radial distribution of the material properties is used. since in this paper, analysis of one-dimensional stress wave propagation along the axial direction of an elastic bar is considered, so, it is preferred that a longitudinal exponential function is used to describe the material distribution of the bar. 2.2 love rod theory as mentioned, according to love rod theory, although the rod cross-sections remain plane after deformation, but the plane cross-sections can expand or contract in their own planes. therefore, the following displacement field is assumed as: , , (3) where u, v and w are respectively the x-, yand z-components of the displacement on a point (x, y, z) on a bar cross-section. also, v is poisson's ratio. according to equation (3), the non-zero components of the strain and the stress are expressed as follow: , , , , (4) , , (5) issn: 2180-1053 vol. 7 no. 1 january june 2015 journal of mechanical engineering and technology 46 where u, v and w are respectively the x-, yand zcomponents of the displacement on a point (x, y, z) on a bar cross-section. also, v is poisson’s ratio. according to equation (3), the non-zero components of the strain and the stress are expressed as follow: 3 figure 1. shematics of geometry of coordinate system. in many studies in the different contexts of solid mechanics, the radial distribution of the material properties is used. since in this paper, analysis of one-dimensional stress wave propagation along the axial direction of an elastic bar is considered, so, it is preferred that a longitudinal exponential function is used to describe the material distribution of the bar. 2.2 love rod theory as mentioned, according to love rod theory, although the rod cross-sections remain plane after deformation, but the plane cross-sections can expand or contract in their own planes. therefore, the following displacement field is assumed as: , , (3) where u, v and w are respectively the x-, yand z-components of the displacement on a point (x, y, z) on a bar cross-section. also, v is poisson's ratio. according to equation (3), the non-zero components of the strain and the stress are expressed as follow: , , , , (4) , , (5) 4 , and are respectively x-, yand z-components of the normal strain. and are the shear components of the strain tensor. is the normal stress along the xdirection. also, and are the shear stresses due to the lateral inertia effect. 2.3 modified couple stress theory according to the modified couple stress theory (yang et al., 2002), the total elastic strain energy u into a region with a volume element dv, expresses as follow: (6) where , , m and are cauchy stress tensor, classical strain tensor, deviatoric part of the couple stress tensor and symmetric curvature tensor, respectively. m and are defined as (7) (8) is the rotation vector and defines as (9) where u is the displacement vector, which the parameters described in equation (3) are the components of this vector. also, is the material length scale parameter, which is mathematically the square of the ratio of the modulus of curvature to the modulus of shear and is physically regarded as material property measuring the effect of couple stress (mindlin, 1963; park & gao, 2006). since, is a function of the material, so, must be varied along the axial direction similar to young's modulus and density. but for simplicity case and parametric study, similar to several studies accomplished on fg nanostructures (reedy, 2011; jung et al., 2014), this parameter assumes constant. substituting equation (3) into equation (9), the non-zero components of rotation vector are obtained as: , (10) and by substituting equation (10) into equation (7), we have: , (11) are respectively x-, yand zcomponents of the normal strain. 4 , and are respectively x-, yand z-components of the normal strain. and are the shear components of the strain tensor. is the normal stress along the xdirection. also, and are the shear stresses due to the lateral inertia effect. 2.3 modified couple stress theory according to the modified couple stress theory (yang et al., 2002), the total elastic strain energy u into a region with a volume element dv, expresses as follow: (6) where , , m and are cauchy stress tensor, classical strain tensor, deviatoric part of the couple stress tensor and symmetric curvature tensor, respectively. m and are defined as (7) (8) is the rotation vector and defines as (9) where u is the displacement vector, which the parameters described in equation (3) are the components of this vector. also, is the material length scale parameter, which is mathematically the square of the ratio of the modulus of curvature to the modulus of shear and is physically regarded as material property measuring the effect of couple stress (mindlin, 1963; park & gao, 2006). since, is a function of the material, so, must be varied along the axial direction similar to young's modulus and density. but for simplicity case and parametric study, similar to several studies accomplished on fg nanostructures (reedy, 2011; jung et al., 2014), this parameter assumes constant. substituting equation (3) into equation (9), the non-zero components of rotation vector are obtained as: , (10) and by substituting equation (10) into equation (7), we have: , (11) and 4 , and are respectively x-, yand z-components of the normal strain. and are the shear components of the strain tensor. is the normal stress along the xdirection. also, and are the shear stresses due to the lateral inertia effect. 2.3 modified couple stress theory according to the modified couple stress theory (yang et al., 2002), the total elastic strain energy u into a region with a volume element dv, expresses as follow: (6) where , , m and are cauchy stress tensor, classical strain tensor, deviatoric part of the couple stress tensor and symmetric curvature tensor, respectively. m and are defined as (7) (8) is the rotation vector and defines as (9) where u is the displacement vector, which the parameters described in equation (3) are the components of this vector. also, is the material length scale parameter, which is mathematically the square of the ratio of the modulus of curvature to the modulus of shear and is physically regarded as material property measuring the effect of couple stress (mindlin, 1963; park & gao, 2006). since, is a function of the material, so, must be varied along the axial direction similar to young's modulus and density. but for simplicity case and parametric study, similar to several studies accomplished on fg nanostructures (reedy, 2011; jung et al., 2014), this parameter assumes constant. substituting equation (3) into equation (9), the non-zero components of rotation vector are obtained as: , (10) and by substituting equation (10) into equation (7), we have: , (11) are the shear components of the strain tensor. 4 , and are respectively x-, yand z-components of the normal strain. and are the shear components of the strain tensor. is the normal stress along the xdirection. also, and are the shear stresses due to the lateral inertia effect. 2.3 modified couple stress theory according to the modified couple stress theory (yang et al., 2002), the total elastic strain energy u into a region with a volume element dv, expresses as follow: (6) where , , m and are cauchy stress tensor, classical strain tensor, deviatoric part of the couple stress tensor and symmetric curvature tensor, respectively. m and are defined as (7) (8) is the rotation vector and defines as (9) where u is the displacement vector, which the parameters described in equation (3) are the components of this vector. also, is the material length scale parameter, which is mathematically the square of the ratio of the modulus of curvature to the modulus of shear and is physically regarded as material property measuring the effect of couple stress (mindlin, 1963; park & gao, 2006). since, is a function of the material, so, must be varied along the axial direction similar to young's modulus and density. but for simplicity case and parametric study, similar to several studies accomplished on fg nanostructures (reedy, 2011; jung et al., 2014), this parameter assumes constant. substituting equation (3) into equation (9), the non-zero components of rotation vector are obtained as: , (10) and by substituting equation (10) into equation (7), we have: , (11) is the normal stress along the x-direction. also, 4 , and are respectively x-, yand z-components of the normal strain. and are the shear components of the strain tensor. is the normal stress along the xdirection. also, and are the shear stresses due to the lateral inertia effect. 2.3 modified couple stress theory according to the modified couple stress theory (yang et al., 2002), the total elastic strain energy u into a region with a volume element dv, expresses as follow: (6) where , , m and are cauchy stress tensor, classical strain tensor, deviatoric part of the couple stress tensor and symmetric curvature tensor, respectively. m and are defined as (7) (8) is the rotation vector and defines as (9) where u is the displacement vector, which the parameters described in equation (3) are the components of this vector. also, is the material length scale parameter, which is mathematically the square of the ratio of the modulus of curvature to the modulus of shear and is physically regarded as material property measuring the effect of couple stress (mindlin, 1963; park & gao, 2006). since, is a function of the material, so, must be varied along the axial direction similar to young's modulus and density. but for simplicity case and parametric study, similar to several studies accomplished on fg nanostructures (reedy, 2011; jung et al., 2014), this parameter assumes constant. substituting equation (3) into equation (9), the non-zero components of rotation vector are obtained as: , (10) and by substituting equation (10) into equation (7), we have: , (11) are the shear stresses due to the lateral inertia effect. 2.3 modified couple stress theory according to the modified couple stress theory (yang et al., 2002), the total elastic strain energy u into a region with a volume element dv, expresses as follow: 4 , and are respectively x-, yand z-components of the normal strain. and are the shear components of the strain tensor. is the normal stress along the xdirection. also, and are the shear stresses due to the lateral inertia effect. 2.3 modified couple stress theory according to the modified couple stress theory (yang et al., 2002), the total elastic strain energy u into a region with a volume element dv, expresses as follow: (6) where , , m and are cauchy stress tensor, classical strain tensor, deviatoric part of the couple stress tensor and symmetric curvature tensor, respectively. m and are defined as (7) (8) is the rotation vector and defines as (9) where u is the displacement vector, which the parameters described in equation (3) are the components of this vector. also, is the material length scale parameter, which is mathematically the square of the ratio of the modulus of curvature to the modulus of shear and is physically regarded as material property measuring the effect of couple stress (mindlin, 1963; park & gao, 2006). since, is a function of the material, so, must be varied along the axial direction similar to young's modulus and density. but for simplicity case and parametric study, similar to several studies accomplished on fg nanostructures (reedy, 2011; jung et al., 2014), this parameter assumes constant. substituting equation (3) into equation (9), the non-zero components of rotation vector are obtained as: , (10) and by substituting equation (10) into equation (7), we have: , (11) where 4 , and are respectively x-, yand z-components of the normal strain. and are the shear components of the strain tensor. is the normal stress along the xdirection. also, and are the shear stresses due to the lateral inertia effect. 2.3 modified couple stress theory according to the modified couple stress theory (yang et al., 2002), the total elastic strain energy u into a region with a volume element dv, expresses as follow: (6) where , , m and are cauchy stress tensor, classical strain tensor, deviatoric part of the couple stress tensor and symmetric curvature tensor, respectively. m and are defined as (7) (8) is the rotation vector and defines as (9) where u is the displacement vector, which the parameters described in equation (3) are the components of this vector. also, is the material length scale parameter, which is mathematically the square of the ratio of the modulus of curvature to the modulus of shear and is physically regarded as material property measuring the effect of couple stress (mindlin, 1963; park & gao, 2006). since, is a function of the material, so, must be varied along the axial direction similar to young's modulus and density. but for simplicity case and parametric study, similar to several studies accomplished on fg nanostructures (reedy, 2011; jung et al., 2014), this parameter assumes constant. substituting equation (3) into equation (9), the non-zero components of rotation vector are obtained as: , (10) and by substituting equation (10) into equation (7), we have: , (11) are cauchy stress tensor, classical strain tensor, deviatoric part of the couple stress tensor and symmetric curvature tensor, respectively. m and 4 , and are respectively x-, yand z-components of the normal strain. and are the shear components of the strain tensor. is the normal stress along the xdirection. also, and are the shear stresses due to the lateral inertia effect. 2.3 modified couple stress theory according to the modified couple stress theory (yang et al., 2002), the total elastic strain energy u into a region with a volume element dv, expresses as follow: (6) where , , m and are cauchy stress tensor, classical strain tensor, deviatoric part of the couple stress tensor and symmetric curvature tensor, respectively. m and are defined as (7) (8) is the rotation vector and defines as (9) where u is the displacement vector, which the parameters described in equation (3) are the components of this vector. also, is the material length scale parameter, which is mathematically the square of the ratio of the modulus of curvature to the modulus of shear and is physically regarded as material property measuring the effect of couple stress (mindlin, 1963; park & gao, 2006). since, is a function of the material, so, must be varied along the axial direction similar to young's modulus and density. but for simplicity case and parametric study, similar to several studies accomplished on fg nanostructures (reedy, 2011; jung et al., 2014), this parameter assumes constant. substituting equation (3) into equation (9), the non-zero components of rotation vector are obtained as: , (10) and by substituting equation (10) into equation (7), we have: , (11) are defined as 4 , and are respectively x-, yand z-components of the normal strain. and are the shear components of the strain tensor. is the normal stress along the xdirection. also, and are the shear stresses due to the lateral inertia effect. 2.3 modified couple stress theory according to the modified couple stress theory (yang et al., 2002), the total elastic strain energy u into a region with a volume element dv, expresses as follow: (6) where , , m and are cauchy stress tensor, classical strain tensor, deviatoric part of the couple stress tensor and symmetric curvature tensor, respectively. m and are defined as (7) (8) is the rotation vector and defines as (9) where u is the displacement vector, which the parameters described in equation (3) are the components of this vector. also, is the material length scale parameter, which is mathematically the square of the ratio of the modulus of curvature to the modulus of shear and is physically regarded as material property measuring the effect of couple stress (mindlin, 1963; park & gao, 2006). since, is a function of the material, so, must be varied along the axial direction similar to young's modulus and density. but for simplicity case and parametric study, similar to several studies accomplished on fg nanostructures (reedy, 2011; jung et al., 2014), this parameter assumes constant. substituting equation (3) into equation (9), the non-zero components of rotation vector are obtained as: , (10) and by substituting equation (10) into equation (7), we have: , (11) 4 , and are respectively x-, yand z-components of the normal strain. and are the shear components of the strain tensor. is the normal stress along the xdirection. also, and are the shear stresses due to the lateral inertia effect. 2.3 modified couple stress theory according to the modified couple stress theory (yang et al., 2002), the total elastic strain energy u into a region with a volume element dv, expresses as follow: (6) where , , m and are cauchy stress tensor, classical strain tensor, deviatoric part of the couple stress tensor and symmetric curvature tensor, respectively. m and are defined as (7) (8) is the rotation vector and defines as (9) where u is the displacement vector, which the parameters described in equation (3) are the components of this vector. also, is the material length scale parameter, which is mathematically the square of the ratio of the modulus of curvature to the modulus of shear and is physically regarded as material property measuring the effect of couple stress (mindlin, 1963; park & gao, 2006). since, is a function of the material, so, must be varied along the axial direction similar to young's modulus and density. but for simplicity case and parametric study, similar to several studies accomplished on fg nanostructures (reedy, 2011; jung et al., 2014), this parameter assumes constant. substituting equation (3) into equation (9), the non-zero components of rotation vector are obtained as: , (10) and by substituting equation (10) into equation (7), we have: , (11) issn: 2180-1053 vol. 7 no. 1 january june 2015 propagation of stress wave in a functionally graded nano-bar based on modified couple stress theory 47 4 , and are respectively x-, yand z-components of the normal strain. and are the shear components of the strain tensor. is the normal stress along the xdirection. also, and are the shear stresses due to the lateral inertia effect. 2.3 modified couple stress theory according to the modified couple stress theory (yang et al., 2002), the total elastic strain energy u into a region with a volume element dv, expresses as follow: (6) where , , m and are cauchy stress tensor, classical strain tensor, deviatoric part of the couple stress tensor and symmetric curvature tensor, respectively. m and are defined as (7) (8) is the rotation vector and defines as (9) where u is the displacement vector, which the parameters described in equation (3) are the components of this vector. also, is the material length scale parameter, which is mathematically the square of the ratio of the modulus of curvature to the modulus of shear and is physically regarded as material property measuring the effect of couple stress (mindlin, 1963; park & gao, 2006). since, is a function of the material, so, must be varied along the axial direction similar to young's modulus and density. but for simplicity case and parametric study, similar to several studies accomplished on fg nanostructures (reedy, 2011; jung et al., 2014), this parameter assumes constant. substituting equation (3) into equation (9), the non-zero components of rotation vector are obtained as: , (10) and by substituting equation (10) into equation (7), we have: , (11) where u is the displacement vector, which the parameters described in equation (3) are the components of this vector. also, is the material length scale parameter, which is mathematically the square of the ratio of the modulus of curvature to the modulus of shear and is physically regarded as material property measuring the effect of couple stress (mindlin, 1963; park & gao, 2006). since, 4 , and are respectively x-, yand z-components of the normal strain. and are the shear components of the strain tensor. is the normal stress along the xdirection. also, and are the shear stresses due to the lateral inertia effect. 2.3 modified couple stress theory according to the modified couple stress theory (yang et al., 2002), the total elastic strain energy u into a region with a volume element dv, expresses as follow: (6) where , , m and are cauchy stress tensor, classical strain tensor, deviatoric part of the couple stress tensor and symmetric curvature tensor, respectively. m and are defined as (7) (8) is the rotation vector and defines as (9) where u is the displacement vector, which the parameters described in equation (3) are the components of this vector. also, is the material length scale parameter, which is mathematically the square of the ratio of the modulus of curvature to the modulus of shear and is physically regarded as material property measuring the effect of couple stress (mindlin, 1963; park & gao, 2006). since, is a function of the material, so, must be varied along the axial direction similar to young's modulus and density. but for simplicity case and parametric study, similar to several studies accomplished on fg nanostructures (reedy, 2011; jung et al., 2014), this parameter assumes constant. substituting equation (3) into equation (9), the non-zero components of rotation vector are obtained as: , (10) and by substituting equation (10) into equation (7), we have: , (11) is a function of the material, so, must be varied along the axial direction similar to young’s modulus and density. but for simplicity case and parametric study, similar to several studies accomplished on fg nanostructures (reedy, 2011; jung et al., 2014), this parameter assumes constant. substituting equation (3) into equation (9), the non-zero components of rotation vector are obtained as: 4 , and are respectively x-, yand z-components of the normal strain. and are the shear components of the strain tensor. is the normal stress along the xdirection. also, and are the shear stresses due to the lateral inertia effect. 2.3 modified couple stress theory according to the modified couple stress theory (yang et al., 2002), the total elastic strain energy u into a region with a volume element dv, expresses as follow: (6) where , , m and are cauchy stress tensor, classical strain tensor, deviatoric part of the couple stress tensor and symmetric curvature tensor, respectively. m and are defined as (7) (8) is the rotation vector and defines as (9) where u is the displacement vector, which the parameters described in equation (3) are the components of this vector. also, is the material length scale parameter, which is mathematically the square of the ratio of the modulus of curvature to the modulus of shear and is physically regarded as material property measuring the effect of couple stress (mindlin, 1963; park & gao, 2006). since, is a function of the material, so, must be varied along the axial direction similar to young's modulus and density. but for simplicity case and parametric study, similar to several studies accomplished on fg nanostructures (reedy, 2011; jung et al., 2014), this parameter assumes constant. substituting equation (3) into equation (9), the non-zero components of rotation vector are obtained as: , (10) and by substituting equation (10) into equation (7), we have: , (11) and by substituting equation (10) into equation (7), we have: 4 , and are respectively x-, yand z-components of the normal strain. and are the shear components of the strain tensor. is the normal stress along the xdirection. also, and are the shear stresses due to the lateral inertia effect. 2.3 modified couple stress theory according to the modified couple stress theory (yang et al., 2002), the total elastic strain energy u into a region with a volume element dv, expresses as follow: (6) where , , m and are cauchy stress tensor, classical strain tensor, deviatoric part of the couple stress tensor and symmetric curvature tensor, respectively. m and are defined as (7) (8) is the rotation vector and defines as (9) where u is the displacement vector, which the parameters described in equation (3) are the components of this vector. also, is the material length scale parameter, which is mathematically the square of the ratio of the modulus of curvature to the modulus of shear and is physically regarded as material property measuring the effect of couple stress (mindlin, 1963; park & gao, 2006). since, is a function of the material, so, must be varied along the axial direction similar to young's modulus and density. but for simplicity case and parametric study, similar to several studies accomplished on fg nanostructures (reedy, 2011; jung et al., 2014), this parameter assumes constant. substituting equation (3) into equation (9), the non-zero components of rotation vector are obtained as: , (10) and by substituting equation (10) into equation (7), we have: , (11) also from eqs. (8) and (11), the non-zero components of tensor m obtain as: 5 also from eqs. (8) and (11), the non-zero components of tensor m obtain as: , (12) 2.4 equation of motion in this study, the equation of motion is derived using hamilton's principle. first, the virtual strain energy and the virtual kinetic energy are obtained as: (13) (14) now, by using hamilton's principle as , where is variation symbol. finally, the equation of motion is derived as follow: (15) in this analysis, a harmonic longitudinal wave propagating along the axial direction is considered, which can be expressed in the complex form as: (16) where k, c and are the wave number, the mean velocity of wave propagation in an fg nano-bar and the wave amplitude, respectively. similar to what was mentioned for the material length scale parameter , the velocity of wave propagation in an fg nano-bar must be varied as a function of x-component (preferably exponentially), but this velocity is assumed to be constant and term of the mean velocity in the bar is used for it. substituting eqs. (1), (2) and (16) into equation (15), the equation of motion achieves as: (17) by a direct solution, we have: (18) 2.4 equation of motion in this study, the equation of motion is derived using hamilton’s principle. first, the virtual strain energy and the virtual kinetic energy are obtained as: 5 also from eqs. (8) and (11), the non-zero components of tensor m obtain as: , (12) 2.4 equation of motion in this study, the equation of motion is derived using hamilton's principle. first, the virtual strain energy and the virtual kinetic energy are obtained as: (13) (14) now, by using hamilton's principle as , where is variation symbol. finally, the equation of motion is derived as follow: (15) in this analysis, a harmonic longitudinal wave propagating along the axial direction is considered, which can be expressed in the complex form as: (16) where k, c and are the wave number, the mean velocity of wave propagation in an fg nano-bar and the wave amplitude, respectively. similar to what was mentioned for the material length scale parameter , the velocity of wave propagation in an fg nano-bar must be varied as a function of x-component (preferably exponentially), but this velocity is assumed to be constant and term of the mean velocity in the bar is used for it. substituting eqs. (1), (2) and (16) into equation (15), the equation of motion achieves as: (17) by a direct solution, we have: (18) now, by using hamilton’s principle as 5 also from eqs. (8) and (11), the non-zero components of tensor m obtain as: , (12) 2.4 equation of motion in this study, the equation of motion is derived using hamilton's principle. first, the virtual strain energy and the virtual kinetic energy are obtained as: (13) (14) now, by using hamilton's principle as , where is variation symbol. finally, the equation of motion is derived as follow: (15) in this analysis, a harmonic longitudinal wave propagating along the axial direction is considered, which can be expressed in the complex form as: (16) where k, c and are the wave number, the mean velocity of wave propagation in an fg nano-bar and the wave amplitude, respectively. similar to what was mentioned for the material length scale parameter , the velocity of wave propagation in an fg nano-bar must be varied as a function of x-component (preferably exponentially), but this velocity is assumed to be constant and term of the mean velocity in the bar is used for it. substituting eqs. (1), (2) and (16) into equation (15), the equation of motion achieves as: (17) by a direct solution, we have: (18) , where is variation symbol. finally, the equation of motion is derived as follow: issn: 2180-1053 vol. 7 no. 1 january june 2015 journal of mechanical engineering and technology 48 5 also from eqs. (8) and (11), the non-zero components of tensor m obtain as: , (12) 2.4 equation of motion in this study, the equation of motion is derived using hamilton's principle. first, the virtual strain energy and the virtual kinetic energy are obtained as: (13) (14) now, by using hamilton's principle as , where is variation symbol. finally, the equation of motion is derived as follow: (15) in this analysis, a harmonic longitudinal wave propagating along the axial direction is considered, which can be expressed in the complex form as: (16) where k, c and are the wave number, the mean velocity of wave propagation in an fg nano-bar and the wave amplitude, respectively. similar to what was mentioned for the material length scale parameter , the velocity of wave propagation in an fg nano-bar must be varied as a function of x-component (preferably exponentially), but this velocity is assumed to be constant and term of the mean velocity in the bar is used for it. substituting eqs. (1), (2) and (16) into equation (15), the equation of motion achieves as: (17) by a direct solution, we have: (18) in this analysis, a harmonic longitudinal wave propagating along the axial direction is considered, which can be expressed in the complex form as: 5 also from eqs. (8) and (11), the non-zero components of tensor m obtain as: , (12) 2.4 equation of motion in this study, the equation of motion is derived using hamilton's principle. first, the virtual strain energy and the virtual kinetic energy are obtained as: (13) (14) now, by using hamilton's principle as , where is variation symbol. finally, the equation of motion is derived as follow: (15) in this analysis, a harmonic longitudinal wave propagating along the axial direction is considered, which can be expressed in the complex form as: (16) where k, c and are the wave number, the mean velocity of wave propagation in an fg nano-bar and the wave amplitude, respectively. similar to what was mentioned for the material length scale parameter , the velocity of wave propagation in an fg nano-bar must be varied as a function of x-component (preferably exponentially), but this velocity is assumed to be constant and term of the mean velocity in the bar is used for it. substituting eqs. (1), (2) and (16) into equation (15), the equation of motion achieves as: (17) by a direct solution, we have: (18) where k, c and are the wave number, the mean velocity of wave propagation in an fg nano-bar and the wave amplitude, respectively. similar to what was mentioned for the material length scale parameter 5 also from eqs. (8) and (11), the non-zero components of tensor m obtain as: , (12) 2.4 equation of motion in this study, the equation of motion is derived using hamilton's principle. first, the virtual strain energy and the virtual kinetic energy are obtained as: (13) (14) now, by using hamilton's principle as , where is variation symbol. finally, the equation of motion is derived as follow: (15) in this analysis, a harmonic longitudinal wave propagating along the axial direction is considered, which can be expressed in the complex form as: (16) where k, c and are the wave number, the mean velocity of wave propagation in an fg nano-bar and the wave amplitude, respectively. similar to what was mentioned for the material length scale parameter , the velocity of wave propagation in an fg nano-bar must be varied as a function of x-component (preferably exponentially), but this velocity is assumed to be constant and term of the mean velocity in the bar is used for it. substituting eqs. (1), (2) and (16) into equation (15), the equation of motion achieves as: (17) by a direct solution, we have: (18) , the velocity of wave propagation in an fg nano-bar must be varied as a function of x-component (preferably exponentially), but this velocity is assumed to be constant and term of the mean velocity in the bar is used for it. substituting eqs. (1), (2) and (16) into equation (15), the equation of motion achieves as: 5 also from eqs. (8) and (11), the non-zero components of tensor m obtain as: , (12) 2.4 equation of motion in this study, the equation of motion is derived using hamilton's principle. first, the virtual strain energy and the virtual kinetic energy are obtained as: (13) (14) now, by using hamilton's principle as , where is variation symbol. finally, the equation of motion is derived as follow: (15) in this analysis, a harmonic longitudinal wave propagating along the axial direction is considered, which can be expressed in the complex form as: (16) where k, c and are the wave number, the mean velocity of wave propagation in an fg nano-bar and the wave amplitude, respectively. similar to what was mentioned for the material length scale parameter , the velocity of wave propagation in an fg nano-bar must be varied as a function of x-component (preferably exponentially), but this velocity is assumed to be constant and term of the mean velocity in the bar is used for it. substituting eqs. (1), (2) and (16) into equation (15), the equation of motion achieves as: (17) by a direct solution, we have: (18) by a direct solution, we have: 5 also from eqs. (8) and (11), the non-zero components of tensor m obtain as: , (12) 2.4 equation of motion in this study, the equation of motion is derived using hamilton's principle. first, the virtual strain energy and the virtual kinetic energy are obtained as: (13) (14) now, by using hamilton's principle as , where is variation symbol. finally, the equation of motion is derived as follow: (15) in this analysis, a harmonic longitudinal wave propagating along the axial direction is considered, which can be expressed in the complex form as: (16) where k, c and are the wave number, the mean velocity of wave propagation in an fg nano-bar and the wave amplitude, respectively. similar to what was mentioned for the material length scale parameter , the velocity of wave propagation in an fg nano-bar must be varied as a function of x-component (preferably exponentially), but this velocity is assumed to be constant and term of the mean velocity in the bar is used for it. substituting eqs. (1), (2) and (16) into equation (15), the equation of motion achieves as: (17) by a direct solution, we have: (18) where 6 where states the gyration radius and i is the polar moment of inertia with respect to the z-axis. thus, for the circular cross section, we have . equation (18) presents the mean velocity of longitudinal stress wave propagation for an fg nano-bar by consideration of poisson's effect. now, this general relation can be derived for some particular cases. for example, when the nano-bar made of a homogeneous material with constant young's modulus and constant density ( ), we have: (19) to obtain the mean velocity of stress wave propagation based on classical theory, it is enough that the material length scale parameter comes from the modified couple stress theory sets to zero ( ). so, we have: (20) by disregard the poisson's effect (v=0), equation (18) rewrites as follow: (21) where is the velocity of stress wave propagation in a simple bernoulli-euler bar. 3.0 results and discussions in this paper, a general solution for different cross sections is done. this section presents numerical results of the stress wave propagation in an fg nano-bar made of circular cross section with radius a=0.34 nm. effects of size, heterogeneity of material and poisson's ratio on the velocity and behavior of the stress wave are evaluated. figure 2 shows the non-dimensional mean velocity of stress wave propagation versus the wave number with different material length scale parameters, where is non-dimensional wave number. in this figure, the size effect is clearly shown and it is observed that by increasing the material parameter at a given radius, the mean velocity of stress wave propagation is increased. this exposes the size-dependent behavior of nano-bars subjected to excitation of the harmonic stress wave. in this figure expresses the non-dimensional mean velocity of stress wave propagation based on the classical theory. as can be seen, the classical theory has considerable errors to estimate the velocity of stress wave propagation and this theory can be useful for macro scale structures. states the gyration radius and i is the polar moment of inertia with respect to the z-axis. thus, for the circular cross section, we have 6 where states the gyration radius and i is the polar moment of inertia with respect to the z-axis. thus, for the circular cross section, we have . equation (18) presents the mean velocity of longitudinal stress wave propagation for an fg nano-bar by consideration of poisson's effect. now, this general relation can be derived for some particular cases. for example, when the nano-bar made of a homogeneous material with constant young's modulus and constant density ( ), we have: (19) to obtain the mean velocity of stress wave propagation based on classical theory, it is enough that the material length scale parameter comes from the modified couple stress theory sets to zero ( ). so, we have: (20) by disregard the poisson's effect (v=0), equation (18) rewrites as follow: (21) where is the velocity of stress wave propagation in a simple bernoulli-euler bar. 3.0 results and discussions in this paper, a general solution for different cross sections is done. this section presents numerical results of the stress wave propagation in an fg nano-bar made of circular cross section with radius a=0.34 nm. effects of size, heterogeneity of material and poisson's ratio on the velocity and behavior of the stress wave are evaluated. figure 2 shows the non-dimensional mean velocity of stress wave propagation versus the wave number with different material length scale parameters, where is non-dimensional wave number. in this figure, the size effect is clearly shown and it is observed that by increasing the material parameter at a given radius, the mean velocity of stress wave propagation is increased. this exposes the size-dependent behavior of nano-bars subjected to excitation of the harmonic stress wave. in this figure expresses the non-dimensional mean velocity of stress wave propagation based on the classical theory. as can be seen, the classical theory has considerable errors to estimate the velocity of stress wave propagation and this theory can be useful for macro scale structures. . equation (18) presents the mean velocity of longitudinal stress wave propagation for an fg nano-bar by consideration of poisson’s effect. now, this general relation can be derived for some particular cases. for example, when the nano-bar made of a homogeneous material with constant young’s modulus 6 where states the gyration radius and i is the polar moment of inertia with respect to the z-axis. thus, for the circular cross section, we have . equation (18) presents the mean velocity of longitudinal stress wave propagation for an fg nano-bar by consideration of poisson's effect. now, this general relation can be derived for some particular cases. for example, when the nano-bar made of a homogeneous material with constant young's modulus and constant density ( ), we have: (19) to obtain the mean velocity of stress wave propagation based on classical theory, it is enough that the material length scale parameter comes from the modified couple stress theory sets to zero ( ). so, we have: (20) by disregard the poisson's effect (v=0), equation (18) rewrites as follow: (21) where is the velocity of stress wave propagation in a simple bernoulli-euler bar. 3.0 results and discussions in this paper, a general solution for different cross sections is done. this section presents numerical results of the stress wave propagation in an fg nano-bar made of circular cross section with radius a=0.34 nm. effects of size, heterogeneity of material and poisson's ratio on the velocity and behavior of the stress wave are evaluated. figure 2 shows the non-dimensional mean velocity of stress wave propagation versus the wave number with different material length scale parameters, where is non-dimensional wave number. in this figure, the size effect is clearly shown and it is observed that by increasing the material parameter at a given radius, the mean velocity of stress wave propagation is increased. this exposes the size-dependent behavior of nano-bars subjected to excitation of the harmonic stress wave. in this figure expresses the non-dimensional mean velocity of stress wave propagation based on the classical theory. as can be seen, the classical theory has considerable errors to estimate the velocity of stress wave propagation and this theory can be useful for macro scale structures. and constant density 6 where states the gyration radius and i is the polar moment of inertia with respect to the z-axis. thus, for the circular cross section, we have . equation (18) presents the mean velocity of longitudinal stress wave propagation for an fg nano-bar by consideration of poisson's effect. now, this general relation can be derived for some particular cases. for example, when the nano-bar made of a homogeneous material with constant young's modulus and constant density ( ), we have: (19) to obtain the mean velocity of stress wave propagation based on classical theory, it is enough that the material length scale parameter comes from the modified couple stress theory sets to zero ( ). so, we have: (20) by disregard the poisson's effect (v=0), equation (18) rewrites as follow: (21) where is the velocity of stress wave propagation in a simple bernoulli-euler bar. 3.0 results and discussions in this paper, a general solution for different cross sections is done. this section presents numerical results of the stress wave propagation in an fg nano-bar made of circular cross section with radius a=0.34 nm. effects of size, heterogeneity of material and poisson's ratio on the velocity and behavior of the stress wave are evaluated. figure 2 shows the non-dimensional mean velocity of stress wave propagation versus the wave number with different material length scale parameters, where is non-dimensional wave number. in this figure, the size effect is clearly shown and it is observed that by increasing the material parameter at a given radius, the mean velocity of stress wave propagation is increased. this exposes the size-dependent behavior of nano-bars subjected to excitation of the harmonic stress wave. in this figure expresses the non-dimensional mean velocity of stress wave propagation based on the classical theory. as can be seen, the classical theory has considerable errors to estimate the velocity of stress wave propagation and this theory can be useful for macro scale structures. , we have: issn: 2180-1053 vol. 7 no. 1 january june 2015 propagation of stress wave in a functionally graded nano-bar based on modified couple stress theory 49 6 where states the gyration radius and i is the polar moment of inertia with respect to the z-axis. thus, for the circular cross section, we have . equation (18) presents the mean velocity of longitudinal stress wave propagation for an fg nano-bar by consideration of poisson's effect. now, this general relation can be derived for some particular cases. for example, when the nano-bar made of a homogeneous material with constant young's modulus and constant density ( ), we have: (19) to obtain the mean velocity of stress wave propagation based on classical theory, it is enough that the material length scale parameter comes from the modified couple stress theory sets to zero ( ). so, we have: (20) by disregard the poisson's effect (v=0), equation (18) rewrites as follow: (21) where is the velocity of stress wave propagation in a simple bernoulli-euler bar. 3.0 results and discussions in this paper, a general solution for different cross sections is done. this section presents numerical results of the stress wave propagation in an fg nano-bar made of circular cross section with radius a=0.34 nm. effects of size, heterogeneity of material and poisson's ratio on the velocity and behavior of the stress wave are evaluated. figure 2 shows the non-dimensional mean velocity of stress wave propagation versus the wave number with different material length scale parameters, where is non-dimensional wave number. in this figure, the size effect is clearly shown and it is observed that by increasing the material parameter at a given radius, the mean velocity of stress wave propagation is increased. this exposes the size-dependent behavior of nano-bars subjected to excitation of the harmonic stress wave. in this figure expresses the non-dimensional mean velocity of stress wave propagation based on the classical theory. as can be seen, the classical theory has considerable errors to estimate the velocity of stress wave propagation and this theory can be useful for macro scale structures. to obtain the mean velocity of stress wave propagation based on classical theory, it is enough that the material length scale parameter comes from the modified couple stress theory sets to zero 6 where states the gyration radius and i is the polar moment of inertia with respect to the z-axis. thus, for the circular cross section, we have . equation (18) presents the mean velocity of longitudinal stress wave propagation for an fg nano-bar by consideration of poisson's effect. now, this general relation can be derived for some particular cases. for example, when the nano-bar made of a homogeneous material with constant young's modulus and constant density ( ), we have: (19) to obtain the mean velocity of stress wave propagation based on classical theory, it is enough that the material length scale parameter comes from the modified couple stress theory sets to zero ( ). so, we have: (20) by disregard the poisson's effect (v=0), equation (18) rewrites as follow: (21) where is the velocity of stress wave propagation in a simple bernoulli-euler bar. 3.0 results and discussions in this paper, a general solution for different cross sections is done. this section presents numerical results of the stress wave propagation in an fg nano-bar made of circular cross section with radius a=0.34 nm. effects of size, heterogeneity of material and poisson's ratio on the velocity and behavior of the stress wave are evaluated. figure 2 shows the non-dimensional mean velocity of stress wave propagation versus the wave number with different material length scale parameters, where is non-dimensional wave number. in this figure, the size effect is clearly shown and it is observed that by increasing the material parameter at a given radius, the mean velocity of stress wave propagation is increased. this exposes the size-dependent behavior of nano-bars subjected to excitation of the harmonic stress wave. in this figure expresses the non-dimensional mean velocity of stress wave propagation based on the classical theory. as can be seen, the classical theory has considerable errors to estimate the velocity of stress wave propagation and this theory can be useful for macro scale structures. . so, we have: 6 where states the gyration radius and i is the polar moment of inertia with respect to the z-axis. thus, for the circular cross section, we have . equation (18) presents the mean velocity of longitudinal stress wave propagation for an fg nano-bar by consideration of poisson's effect. now, this general relation can be derived for some particular cases. for example, when the nano-bar made of a homogeneous material with constant young's modulus and constant density ( ), we have: (19) to obtain the mean velocity of stress wave propagation based on classical theory, it is enough that the material length scale parameter comes from the modified couple stress theory sets to zero ( ). so, we have: (20) by disregard the poisson's effect (v=0), equation (18) rewrites as follow: (21) where is the velocity of stress wave propagation in a simple bernoulli-euler bar. 3.0 results and discussions in this paper, a general solution for different cross sections is done. this section presents numerical results of the stress wave propagation in an fg nano-bar made of circular cross section with radius a=0.34 nm. effects of size, heterogeneity of material and poisson's ratio on the velocity and behavior of the stress wave are evaluated. figure 2 shows the non-dimensional mean velocity of stress wave propagation versus the wave number with different material length scale parameters, where is non-dimensional wave number. in this figure, the size effect is clearly shown and it is observed that by increasing the material parameter at a given radius, the mean velocity of stress wave propagation is increased. this exposes the size-dependent behavior of nano-bars subjected to excitation of the harmonic stress wave. in this figure expresses the non-dimensional mean velocity of stress wave propagation based on the classical theory. as can be seen, the classical theory has considerable errors to estimate the velocity of stress wave propagation and this theory can be useful for macro scale structures. by disregard the poisson’s effect (v=0), equation (18) rewrites as follow: 6 where states the gyration radius and i is the polar moment of inertia with respect to the z-axis. thus, for the circular cross section, we have . equation (18) presents the mean velocity of longitudinal stress wave propagation for an fg nano-bar by consideration of poisson's effect. now, this general relation can be derived for some particular cases. for example, when the nano-bar made of a homogeneous material with constant young's modulus and constant density ( ), we have: (19) to obtain the mean velocity of stress wave propagation based on classical theory, it is enough that the material length scale parameter comes from the modified couple stress theory sets to zero ( ). so, we have: (20) by disregard the poisson's effect (v=0), equation (18) rewrites as follow: (21) where is the velocity of stress wave propagation in a simple bernoulli-euler bar. 3.0 results and discussions in this paper, a general solution for different cross sections is done. this section presents numerical results of the stress wave propagation in an fg nano-bar made of circular cross section with radius a=0.34 nm. effects of size, heterogeneity of material and poisson's ratio on the velocity and behavior of the stress wave are evaluated. figure 2 shows the non-dimensional mean velocity of stress wave propagation versus the wave number with different material length scale parameters, where is non-dimensional wave number. in this figure, the size effect is clearly shown and it is observed that by increasing the material parameter at a given radius, the mean velocity of stress wave propagation is increased. this exposes the size-dependent behavior of nano-bars subjected to excitation of the harmonic stress wave. in this figure expresses the non-dimensional mean velocity of stress wave propagation based on the classical theory. as can be seen, the classical theory has considerable errors to estimate the velocity of stress wave propagation and this theory can be useful for macro scale structures. where 6 where states the gyration radius and i is the polar moment of inertia with respect to the z-axis. thus, for the circular cross section, we have . equation (18) presents the mean velocity of longitudinal stress wave propagation for an fg nano-bar by consideration of poisson's effect. now, this general relation can be derived for some particular cases. for example, when the nano-bar made of a homogeneous material with constant young's modulus and constant density ( ), we have: (19) to obtain the mean velocity of stress wave propagation based on classical theory, it is enough that the material length scale parameter comes from the modified couple stress theory sets to zero ( ). so, we have: (20) by disregard the poisson's effect (v=0), equation (18) rewrites as follow: (21) where is the velocity of stress wave propagation in a simple bernoulli-euler bar. 3.0 results and discussions in this paper, a general solution for different cross sections is done. this section presents numerical results of the stress wave propagation in an fg nano-bar made of circular cross section with radius a=0.34 nm. effects of size, heterogeneity of material and poisson's ratio on the velocity and behavior of the stress wave are evaluated. figure 2 shows the non-dimensional mean velocity of stress wave propagation versus the wave number with different material length scale parameters, where is non-dimensional wave number. in this figure, the size effect is clearly shown and it is observed that by increasing the material parameter at a given radius, the mean velocity of stress wave propagation is increased. this exposes the size-dependent behavior of nano-bars subjected to excitation of the harmonic stress wave. in this figure expresses the non-dimensional mean velocity of stress wave propagation based on the classical theory. as can be seen, the classical theory has considerable errors to estimate the velocity of stress wave propagation and this theory can be useful for macro scale structures. is the velocity of stress wave propagation in a simple bernoullieuler bar. 3.0 results and discussions in this paper, a general solution for different cross sections is done. this section presents numerical results of the stress wave propagation in an fg nano-bar made of circular cross section with radius a=0.34 nm. effects of size, heterogeneity of material and poisson’s ratio on the velocity and behavior of the stress wave are evaluated. figure 2 shows the non-dimensional mean velocity of stress wave propagation versus the wave number with different material length scale parameters, where 6 where states the gyration radius and i is the polar moment of inertia with respect to the z-axis. thus, for the circular cross section, we have . equation (18) presents the mean velocity of longitudinal stress wave propagation for an fg nano-bar by consideration of poisson's effect. now, this general relation can be derived for some particular cases. for example, when the nano-bar made of a homogeneous material with constant young's modulus and constant density ( ), we have: (19) to obtain the mean velocity of stress wave propagation based on classical theory, it is enough that the material length scale parameter comes from the modified couple stress theory sets to zero ( ). so, we have: (20) by disregard the poisson's effect (v=0), equation (18) rewrites as follow: (21) where is the velocity of stress wave propagation in a simple bernoulli-euler bar. 3.0 results and discussions in this paper, a general solution for different cross sections is done. this section presents numerical results of the stress wave propagation in an fg nano-bar made of circular cross section with radius a=0.34 nm. effects of size, heterogeneity of material and poisson's ratio on the velocity and behavior of the stress wave are evaluated. figure 2 shows the non-dimensional mean velocity of stress wave propagation versus the wave number with different material length scale parameters, where is non-dimensional wave number. in this figure, the size effect is clearly shown and it is observed that by increasing the material parameter at a given radius, the mean velocity of stress wave propagation is increased. this exposes the size-dependent behavior of nano-bars subjected to excitation of the harmonic stress wave. in this figure expresses the non-dimensional mean velocity of stress wave propagation based on the classical theory. as can be seen, the classical theory has considerable errors to estimate the velocity of stress wave propagation and this theory can be useful for macro scale structures. is non-dimensional wave number. in this figure, the size effect is clearly shown and it is observed that by increasing the material parameter at a given radius, the mean velocity of stress wave propagation is increased. this exposes the size-dependent behavior of nano-bars subjected to excitation of the harmonic stress wave. 6 where states the gyration radius and i is the polar moment of inertia with respect to the z-axis. thus, for the circular cross section, we have . equation (18) presents the mean velocity of longitudinal stress wave propagation for an fg nano-bar by consideration of poisson's effect. now, this general relation can be derived for some particular cases. for example, when the nano-bar made of a homogeneous material with constant young's modulus and constant density ( ), we have: (19) to obtain the mean velocity of stress wave propagation based on classical theory, it is enough that the material length scale parameter comes from the modified couple stress theory sets to zero ( ). so, we have: (20) by disregard the poisson's effect (v=0), equation (18) rewrites as follow: (21) where is the velocity of stress wave propagation in a simple bernoulli-euler bar. 3.0 results and discussions in this paper, a general solution for different cross sections is done. this section presents numerical results of the stress wave propagation in an fg nano-bar made of circular cross section with radius a=0.34 nm. effects of size, heterogeneity of material and poisson's ratio on the velocity and behavior of the stress wave are evaluated. figure 2 shows the non-dimensional mean velocity of stress wave propagation versus the wave number with different material length scale parameters, where is non-dimensional wave number. in this figure, the size effect is clearly shown and it is observed that by increasing the material parameter at a given radius, the mean velocity of stress wave propagation is increased. this exposes the size-dependent behavior of nano-bars subjected to excitation of the harmonic stress wave. in this figure expresses the non-dimensional mean velocity of stress wave propagation based on the classical theory. as can be seen, the classical theory has considerable errors to estimate the velocity of stress wave propagation and this theory can be useful for macro scale structures. in this figure expresses the non-dimensional mean velocity of stress wave propagation based on the classical theory. as can be seen, the classical theory has considerable errors to estimate the velocity of stress wave propagation and this theory can be useful for macro scale structures. issn: 2180-1053 vol. 7 no. 1 january june 2015 journal of mechanical engineering and technology 50 7 figure 2. effect of material parameter on velocity of stress wave propagation with and v=0.25. figure 3. effect of the material inhomogeneity constant on velocity of stress wave propagation and v=0.25. figure 2. effect of material parameter on velocity of stress wave propagation with 7 figure 2. effect of material parameter on velocity of stress wave propagation with and v=0.25. figure 3. effect of the material inhomogeneity constant on velocity of stress wave propagation and v=0.25. and v=0.25. 7 figure 2. effect of material parameter on velocity of stress wave propagation with and v=0.25. figure 3. effect of the material inhomogeneity constant on velocity of stress wave propagation and v=0.25. figure 3. effect of the material inhomogeneity constant on velocity of stress wave propagation 7 figure 2. effect of material parameter on velocity of stress wave propagation with and v=0.25. figure 3. effect of the material inhomogeneity constant on velocity of stress wave propagation and v=0.25. and v=0.25. figure 3 illustrates the effect of the material inhomogeneity constant 8 figure 3 illustrates the effect of the material inhomogeneity constant on velocity of stress wave propagation. this figure shows that increasing the material inhomogeneity constant leads to decreasing the mean velocity of stress wave propagation. in fact, the velocity of stress wave propagation is averagely reduced when the heterogeneity of material increases. poisson's effect on velocity of stress wave propagation expresses in figure 4. for small non-dimensional wave number (approximately less than 3), the velocity of stress wave propagation is decreased by increasing poisson's ratio, while for larger non-dimensional wave numbers, the velocity of stress wave propagation is increased by increasing poisson's ratio. also, when the lateral effect is neglected (v=0), the velocity of stress wave propagation becomes equal to a constant value (velocity of stress wave propagation in a homogeneous bernoulli-euler bar). as can be seen in figs. 2-4, for large non-dimensional wave numbers , the velocity of stress wave propagation is increased by increasing , and increasing of for small non-dimensional wave numbers leads to decreasing the velocity of stress wave propagation. according to equation (5), the stress wave made in the nano-bar obtains as , where . variations of real part of the non-dimensional stress wave against non-dimensional wave number with different material length scale parameters under , v=0.25, x=10a and t=0.1s are shown in figure 5. in this figure, the stress wave behavior is completely harmonic except for very small values of . this is because of the fact that when the wave number tends to zero then the incoming wave loses its harmonic vitality and becomes a constant wave (equation (16)). moreover, by increasing , the wave length of stress wave is decreased because of the wave number introduced in equation (16) relates with inverse of the incoming wave length. also, the size effect on stress wave is studied and it is observed that by increasing the material parameter , the stress wave propagated in nano-bar starts its harmonic behavior earlier and leads to increasing of stress wave intensity. similar to what was mentioned for figure 5, the material inhomogeneity constant and poisson's ratio have similar effect on harmonic behavior of the stress wave (figures 6 and 7). maximum shear stress wave made in nano-bar with circular cross section is as (equation 5), where ( ). it should be noted that for circular cross section, we have: . the harmonic behavior of non-dimensional shear stress wave against non-dimensional wave number is shown in figure 8. by increasing , intensity and amplitude of the shear stress increases. this is because of the fact that the shear stress made in nano-bar is caused by lateral inertia, therefore, this is dependent on radius of bar. consequently, by increasing at a given wave number, the radius of bar increases. so, , by increasing , amplitude of the shear stress wave increases. because the behavior of the shear stress wave versus the material parameter, material inhomogeneity constant and poisson's ratio is similar to axial stress wave, evaluation of theses behaviors are not considered. on velocity of stress wave propagation. this figure shows that increasing the material inhomogeneity constant leads to decreasing the mean velocity of stress wave propagation. in fact, the velocity of stress wave propagation is averagely reduced when the heterogeneity of material increases. poisson’s effect on velocity of stress wave propagation expresses in figure 4. for small non-dimensional wave number (approximately less issn: 2180-1053 vol. 7 no. 1 january june 2015 propagation of stress wave in a functionally graded nano-bar based on modified couple stress theory 51 than 3), the velocity of stress wave propagation is decreased by increasing poisson’s ratio, while for larger non-dimensional wave numbers, the velocity of stress wave propagation is increased by increasing poisson’s ratio. also, when the lateral effect is neglected (v=0), the velocity of stress wave propagation becomes equal to a constant value (velocity of stress wave propagation in a homogeneous bernoulli-euler bar). as can be seen in figs. 2-4, for large non-dimensional wave numbers 8 figure 3 illustrates the effect of the material inhomogeneity constant on velocity of stress wave propagation. this figure shows that increasing the material inhomogeneity constant leads to decreasing the mean velocity of stress wave propagation. in fact, the velocity of stress wave propagation is averagely reduced when the heterogeneity of material increases. poisson's effect on velocity of stress wave propagation expresses in figure 4. for small non-dimensional wave number (approximately less than 3), the velocity of stress wave propagation is decreased by increasing poisson's ratio, while for larger non-dimensional wave numbers, the velocity of stress wave propagation is increased by increasing poisson's ratio. also, when the lateral effect is neglected (v=0), the velocity of stress wave propagation becomes equal to a constant value (velocity of stress wave propagation in a homogeneous bernoulli-euler bar). as can be seen in figs. 2-4, for large non-dimensional wave numbers , the velocity of stress wave propagation is increased by increasing , and increasing of for small non-dimensional wave numbers leads to decreasing the velocity of stress wave propagation. according to equation (5), the stress wave made in the nano-bar obtains as , where . variations of real part of the non-dimensional stress wave against non-dimensional wave number with different material length scale parameters under , v=0.25, x=10a and t=0.1s are shown in figure 5. in this figure, the stress wave behavior is completely harmonic except for very small values of . this is because of the fact that when the wave number tends to zero then the incoming wave loses its harmonic vitality and becomes a constant wave (equation (16)). moreover, by increasing , the wave length of stress wave is decreased because of the wave number introduced in equation (16) relates with inverse of the incoming wave length. also, the size effect on stress wave is studied and it is observed that by increasing the material parameter , the stress wave propagated in nano-bar starts its harmonic behavior earlier and leads to increasing of stress wave intensity. similar to what was mentioned for figure 5, the material inhomogeneity constant and poisson's ratio have similar effect on harmonic behavior of the stress wave (figures 6 and 7). maximum shear stress wave made in nano-bar with circular cross section is as (equation 5), where ( ). it should be noted that for circular cross section, we have: . the harmonic behavior of non-dimensional shear stress wave against non-dimensional wave number is shown in figure 8. by increasing , intensity and amplitude of the shear stress increases. this is because of the fact that the shear stress made in nano-bar is caused by lateral inertia, therefore, this is dependent on radius of bar. consequently, by increasing at a given wave number, the radius of bar increases. so, , by increasing , amplitude of the shear stress wave increases. because the behavior of the shear stress wave versus the material parameter, material inhomogeneity constant and poisson's ratio is similar to axial stress wave, evaluation of theses behaviors are not considered. , the velocity of stress wave propagation is increased by increasing 8 figure 3 illustrates the effect of the material inhomogeneity constant on velocity of stress wave propagation. this figure shows that increasing the material inhomogeneity constant leads to decreasing the mean velocity of stress wave propagation. in fact, the velocity of stress wave propagation is averagely reduced when the heterogeneity of material increases. poisson's effect on velocity of stress wave propagation expresses in figure 4. for small non-dimensional wave number (approximately less than 3), the velocity of stress wave propagation is decreased by increasing poisson's ratio, while for larger non-dimensional wave numbers, the velocity of stress wave propagation is increased by increasing poisson's ratio. also, when the lateral effect is neglected (v=0), the velocity of stress wave propagation becomes equal to a constant value (velocity of stress wave propagation in a homogeneous bernoulli-euler bar). as can be seen in figs. 2-4, for large non-dimensional wave numbers , the velocity of stress wave propagation is increased by increasing , and increasing of for small non-dimensional wave numbers leads to decreasing the velocity of stress wave propagation. according to equation (5), the stress wave made in the nano-bar obtains as , where . variations of real part of the non-dimensional stress wave against non-dimensional wave number with different material length scale parameters under , v=0.25, x=10a and t=0.1s are shown in figure 5. in this figure, the stress wave behavior is completely harmonic except for very small values of . this is because of the fact that when the wave number tends to zero then the incoming wave loses its harmonic vitality and becomes a constant wave (equation (16)). moreover, by increasing , the wave length of stress wave is decreased because of the wave number introduced in equation (16) relates with inverse of the incoming wave length. also, the size effect on stress wave is studied and it is observed that by increasing the material parameter , the stress wave propagated in nano-bar starts its harmonic behavior earlier and leads to increasing of stress wave intensity. similar to what was mentioned for figure 5, the material inhomogeneity constant and poisson's ratio have similar effect on harmonic behavior of the stress wave (figures 6 and 7). maximum shear stress wave made in nano-bar with circular cross section is as (equation 5), where ( ). it should be noted that for circular cross section, we have: . the harmonic behavior of non-dimensional shear stress wave against non-dimensional wave number is shown in figure 8. by increasing , intensity and amplitude of the shear stress increases. this is because of the fact that the shear stress made in nano-bar is caused by lateral inertia, therefore, this is dependent on radius of bar. consequently, by increasing at a given wave number, the radius of bar increases. so, , by increasing , amplitude of the shear stress wave increases. because the behavior of the shear stress wave versus the material parameter, material inhomogeneity constant and poisson's ratio is similar to axial stress wave, evaluation of theses behaviors are not considered. , and increasing of 8 figure 3 illustrates the effect of the material inhomogeneity constant on velocity of stress wave propagation. this figure shows that increasing the material inhomogeneity constant leads to decreasing the mean velocity of stress wave propagation. in fact, the velocity of stress wave propagation is averagely reduced when the heterogeneity of material increases. poisson's effect on velocity of stress wave propagation expresses in figure 4. for small non-dimensional wave number (approximately less than 3), the velocity of stress wave propagation is decreased by increasing poisson's ratio, while for larger non-dimensional wave numbers, the velocity of stress wave propagation is increased by increasing poisson's ratio. also, when the lateral effect is neglected (v=0), the velocity of stress wave propagation becomes equal to a constant value (velocity of stress wave propagation in a homogeneous bernoulli-euler bar). as can be seen in figs. 2-4, for large non-dimensional wave numbers , the velocity of stress wave propagation is increased by increasing , and increasing of for small non-dimensional wave numbers leads to decreasing the velocity of stress wave propagation. according to equation (5), the stress wave made in the nano-bar obtains as , where . variations of real part of the non-dimensional stress wave against non-dimensional wave number with different material length scale parameters under , v=0.25, x=10a and t=0.1s are shown in figure 5. in this figure, the stress wave behavior is completely harmonic except for very small values of . this is because of the fact that when the wave number tends to zero then the incoming wave loses its harmonic vitality and becomes a constant wave (equation (16)). moreover, by increasing , the wave length of stress wave is decreased because of the wave number introduced in equation (16) relates with inverse of the incoming wave length. also, the size effect on stress wave is studied and it is observed that by increasing the material parameter , the stress wave propagated in nano-bar starts its harmonic behavior earlier and leads to increasing of stress wave intensity. similar to what was mentioned for figure 5, the material inhomogeneity constant and poisson's ratio have similar effect on harmonic behavior of the stress wave (figures 6 and 7). maximum shear stress wave made in nano-bar with circular cross section is as (equation 5), where ( ). it should be noted that for circular cross section, we have: . the harmonic behavior of non-dimensional shear stress wave against non-dimensional wave number is shown in figure 8. by increasing , intensity and amplitude of the shear stress increases. this is because of the fact that the shear stress made in nano-bar is caused by lateral inertia, therefore, this is dependent on radius of bar. consequently, by increasing at a given wave number, the radius of bar increases. so, , by increasing , amplitude of the shear stress wave increases. because the behavior of the shear stress wave versus the material parameter, material inhomogeneity constant and poisson's ratio is similar to axial stress wave, evaluation of theses behaviors are not considered. for small non-dimensional wave numbers leads to decreasing the velocity of stress wave propagation. according to equation (5), the stress wave made in the nano-bar obtains as 8 figure 3 illustrates the effect of the material inhomogeneity constant on velocity of stress wave propagation. this figure shows that increasing the material inhomogeneity constant leads to decreasing the mean velocity of stress wave propagation. in fact, the velocity of stress wave propagation is averagely reduced when the heterogeneity of material increases. poisson's effect on velocity of stress wave propagation expresses in figure 4. for small non-dimensional wave number (approximately less than 3), the velocity of stress wave propagation is decreased by increasing poisson's ratio, while for larger non-dimensional wave numbers, the velocity of stress wave propagation is increased by increasing poisson's ratio. also, when the lateral effect is neglected (v=0), the velocity of stress wave propagation becomes equal to a constant value (velocity of stress wave propagation in a homogeneous bernoulli-euler bar). as can be seen in figs. 2-4, for large non-dimensional wave numbers , the velocity of stress wave propagation is increased by increasing , and increasing of for small non-dimensional wave numbers leads to decreasing the velocity of stress wave propagation. according to equation (5), the stress wave made in the nano-bar obtains as , where . variations of real part of the non-dimensional stress wave against non-dimensional wave number with different material length scale parameters under , v=0.25, x=10a and t=0.1s are shown in figure 5. in this figure, the stress wave behavior is completely harmonic except for very small values of . this is because of the fact that when the wave number tends to zero then the incoming wave loses its harmonic vitality and becomes a constant wave (equation (16)). moreover, by increasing , the wave length of stress wave is decreased because of the wave number introduced in equation (16) relates with inverse of the incoming wave length. also, the size effect on stress wave is studied and it is observed that by increasing the material parameter , the stress wave propagated in nano-bar starts its harmonic behavior earlier and leads to increasing of stress wave intensity. similar to what was mentioned for figure 5, the material inhomogeneity constant and poisson's ratio have similar effect on harmonic behavior of the stress wave (figures 6 and 7). maximum shear stress wave made in nano-bar with circular cross section is as (equation 5), where ( ). it should be noted that for circular cross section, we have: . the harmonic behavior of non-dimensional shear stress wave against non-dimensional wave number is shown in figure 8. by increasing , intensity and amplitude of the shear stress increases. this is because of the fact that the shear stress made in nano-bar is caused by lateral inertia, therefore, this is dependent on radius of bar. consequently, by increasing at a given wave number, the radius of bar increases. so, , by increasing , amplitude of the shear stress wave increases. because the behavior of the shear stress wave versus the material parameter, material inhomogeneity constant and poisson's ratio is similar to axial stress wave, evaluation of theses behaviors are not considered. . variations of real part of the non-dimensional stress wave against non-dimensional wave number with different material length scale parameters under 8 figure 3 illustrates the effect of the material inhomogeneity constant on velocity of stress wave propagation. this figure shows that increasing the material inhomogeneity constant leads to decreasing the mean velocity of stress wave propagation. in fact, the velocity of stress wave propagation is averagely reduced when the heterogeneity of material increases. poisson's effect on velocity of stress wave propagation expresses in figure 4. for small non-dimensional wave number (approximately less than 3), the velocity of stress wave propagation is decreased by increasing poisson's ratio, while for larger non-dimensional wave numbers, the velocity of stress wave propagation is increased by increasing poisson's ratio. also, when the lateral effect is neglected (v=0), the velocity of stress wave propagation becomes equal to a constant value (velocity of stress wave propagation in a homogeneous bernoulli-euler bar). as can be seen in figs. 2-4, for large non-dimensional wave numbers , the velocity of stress wave propagation is increased by increasing , and increasing of for small non-dimensional wave numbers leads to decreasing the velocity of stress wave propagation. according to equation (5), the stress wave made in the nano-bar obtains as , where . variations of real part of the non-dimensional stress wave against non-dimensional wave number with different material length scale parameters under , v=0.25, x=10a and t=0.1s are shown in figure 5. in this figure, the stress wave behavior is completely harmonic except for very small values of . this is because of the fact that when the wave number tends to zero then the incoming wave loses its harmonic vitality and becomes a constant wave (equation (16)). moreover, by increasing , the wave length of stress wave is decreased because of the wave number introduced in equation (16) relates with inverse of the incoming wave length. also, the size effect on stress wave is studied and it is observed that by increasing the material parameter , the stress wave propagated in nano-bar starts its harmonic behavior earlier and leads to increasing of stress wave intensity. similar to what was mentioned for figure 5, the material inhomogeneity constant and poisson's ratio have similar effect on harmonic behavior of the stress wave (figures 6 and 7). maximum shear stress wave made in nano-bar with circular cross section is as (equation 5), where ( ). it should be noted that for circular cross section, we have: . the harmonic behavior of non-dimensional shear stress wave against non-dimensional wave number is shown in figure 8. by increasing , intensity and amplitude of the shear stress increases. this is because of the fact that the shear stress made in nano-bar is caused by lateral inertia, therefore, this is dependent on radius of bar. consequently, by increasing at a given wave number, the radius of bar increases. so, , by increasing , amplitude of the shear stress wave increases. because the behavior of the shear stress wave versus the material parameter, material inhomogeneity constant and poisson's ratio is similar to axial stress wave, evaluation of theses behaviors are not considered. are shown in figure 5. in this figure, the stress wave behavior is completely harmonic except for very small values of 8 figure 3 illustrates the effect of the material inhomogeneity constant on velocity of stress wave propagation. this figure shows that increasing the material inhomogeneity constant leads to decreasing the mean velocity of stress wave propagation. in fact, the velocity of stress wave propagation is averagely reduced when the heterogeneity of material increases. poisson's effect on velocity of stress wave propagation expresses in figure 4. for small non-dimensional wave number (approximately less than 3), the velocity of stress wave propagation is decreased by increasing poisson's ratio, while for larger non-dimensional wave numbers, the velocity of stress wave propagation is increased by increasing poisson's ratio. also, when the lateral effect is neglected (v=0), the velocity of stress wave propagation becomes equal to a constant value (velocity of stress wave propagation in a homogeneous bernoulli-euler bar). as can be seen in figs. 2-4, for large non-dimensional wave numbers , the velocity of stress wave propagation is increased by increasing , and increasing of for small non-dimensional wave numbers leads to decreasing the velocity of stress wave propagation. according to equation (5), the stress wave made in the nano-bar obtains as , where . variations of real part of the non-dimensional stress wave against non-dimensional wave number with different material length scale parameters under , v=0.25, x=10a and t=0.1s are shown in figure 5. in this figure, the stress wave behavior is completely harmonic except for very small values of . this is because of the fact that when the wave number tends to zero then the incoming wave loses its harmonic vitality and becomes a constant wave (equation (16)). moreover, by increasing , the wave length of stress wave is decreased because of the wave number introduced in equation (16) relates with inverse of the incoming wave length. also, the size effect on stress wave is studied and it is observed that by increasing the material parameter , the stress wave propagated in nano-bar starts its harmonic behavior earlier and leads to increasing of stress wave intensity. similar to what was mentioned for figure 5, the material inhomogeneity constant and poisson's ratio have similar effect on harmonic behavior of the stress wave (figures 6 and 7). maximum shear stress wave made in nano-bar with circular cross section is as (equation 5), where ( ). it should be noted that for circular cross section, we have: . the harmonic behavior of non-dimensional shear stress wave against non-dimensional wave number is shown in figure 8. by increasing , intensity and amplitude of the shear stress increases. this is because of the fact that the shear stress made in nano-bar is caused by lateral inertia, therefore, this is dependent on radius of bar. consequently, by increasing at a given wave number, the radius of bar increases. so, , by increasing , amplitude of the shear stress wave increases. because the behavior of the shear stress wave versus the material parameter, material inhomogeneity constant and poisson's ratio is similar to axial stress wave, evaluation of theses behaviors are not considered. . this is because of the fact that when the wave number tends to zero then the incoming wave loses its harmonic vitality and becomes a constant wave (equation (16)). moreover, by increasing 8 figure 3 illustrates the effect of the material inhomogeneity constant on velocity of stress wave propagation. this figure shows that increasing the material inhomogeneity constant leads to decreasing the mean velocity of stress wave propagation. in fact, the velocity of stress wave propagation is averagely reduced when the heterogeneity of material increases. poisson's effect on velocity of stress wave propagation expresses in figure 4. for small non-dimensional wave number (approximately less than 3), the velocity of stress wave propagation is decreased by increasing poisson's ratio, while for larger non-dimensional wave numbers, the velocity of stress wave propagation is increased by increasing poisson's ratio. also, when the lateral effect is neglected (v=0), the velocity of stress wave propagation becomes equal to a constant value (velocity of stress wave propagation in a homogeneous bernoulli-euler bar). as can be seen in figs. 2-4, for large non-dimensional wave numbers , the velocity of stress wave propagation is increased by increasing , and increasing of for small non-dimensional wave numbers leads to decreasing the velocity of stress wave propagation. according to equation (5), the stress wave made in the nano-bar obtains as , where . variations of real part of the non-dimensional stress wave against non-dimensional wave number with different material length scale parameters under , v=0.25, x=10a and t=0.1s are shown in figure 5. in this figure, the stress wave behavior is completely harmonic except for very small values of . this is because of the fact that when the wave number tends to zero then the incoming wave loses its harmonic vitality and becomes a constant wave (equation (16)). moreover, by increasing , the wave length of stress wave is decreased because of the wave number introduced in equation (16) relates with inverse of the incoming wave length. also, the size effect on stress wave is studied and it is observed that by increasing the material parameter , the stress wave propagated in nano-bar starts its harmonic behavior earlier and leads to increasing of stress wave intensity. similar to what was mentioned for figure 5, the material inhomogeneity constant and poisson's ratio have similar effect on harmonic behavior of the stress wave (figures 6 and 7). maximum shear stress wave made in nano-bar with circular cross section is as (equation 5), where ( ). it should be noted that for circular cross section, we have: . the harmonic behavior of non-dimensional shear stress wave against non-dimensional wave number is shown in figure 8. by increasing , intensity and amplitude of the shear stress increases. this is because of the fact that the shear stress made in nano-bar is caused by lateral inertia, therefore, this is dependent on radius of bar. consequently, by increasing at a given wave number, the radius of bar increases. so, , by increasing , amplitude of the shear stress wave increases. because the behavior of the shear stress wave versus the material parameter, material inhomogeneity constant and poisson's ratio is similar to axial stress wave, evaluation of theses behaviors are not considered. , the wave length of stress wave is decreased because of the wave number introduced in equation (16) relates with inverse of the incoming wave length. also, the size effect on stress wave is studied and it is observed that by increasing the material parameter 8 figure 3 illustrates the effect of the material inhomogeneity constant on velocity of stress wave propagation. this figure shows that increasing the material inhomogeneity constant leads to decreasing the mean velocity of stress wave propagation. in fact, the velocity of stress wave propagation is averagely reduced when the heterogeneity of material increases. poisson's effect on velocity of stress wave propagation expresses in figure 4. for small non-dimensional wave number (approximately less than 3), the velocity of stress wave propagation is decreased by increasing poisson's ratio, while for larger non-dimensional wave numbers, the velocity of stress wave propagation is increased by increasing poisson's ratio. also, when the lateral effect is neglected (v=0), the velocity of stress wave propagation becomes equal to a constant value (velocity of stress wave propagation in a homogeneous bernoulli-euler bar). as can be seen in figs. 2-4, for large non-dimensional wave numbers , the velocity of stress wave propagation is increased by increasing , and increasing of for small non-dimensional wave numbers leads to decreasing the velocity of stress wave propagation. according to equation (5), the stress wave made in the nano-bar obtains as , where . variations of real part of the non-dimensional stress wave against non-dimensional wave number with different material length scale parameters under , v=0.25, x=10a and t=0.1s are shown in figure 5. in this figure, the stress wave behavior is completely harmonic except for very small values of . this is because of the fact that when the wave number tends to zero then the incoming wave loses its harmonic vitality and becomes a constant wave (equation (16)). moreover, by increasing , the wave length of stress wave is decreased because of the wave number introduced in equation (16) relates with inverse of the incoming wave length. also, the size effect on stress wave is studied and it is observed that by increasing the material parameter , the stress wave propagated in nano-bar starts its harmonic behavior earlier and leads to increasing of stress wave intensity. similar to what was mentioned for figure 5, the material inhomogeneity constant and poisson's ratio have similar effect on harmonic behavior of the stress wave (figures 6 and 7). maximum shear stress wave made in nano-bar with circular cross section is as (equation 5), where ( ). it should be noted that for circular cross section, we have: . the harmonic behavior of non-dimensional shear stress wave against non-dimensional wave number is shown in figure 8. by increasing , intensity and amplitude of the shear stress increases. this is because of the fact that the shear stress made in nano-bar is caused by lateral inertia, therefore, this is dependent on radius of bar. consequently, by increasing at a given wave number, the radius of bar increases. so, , by increasing , amplitude of the shear stress wave increases. because the behavior of the shear stress wave versus the material parameter, material inhomogeneity constant and poisson's ratio is similar to axial stress wave, evaluation of theses behaviors are not considered. , the stress wave propagated in nano-bar starts its harmonic behavior earlier and leads to increasing of stress wave intensity. similar to what was mentioned for figure 5, the material inhomogeneity constant and poisson’s ratio have similar effect on harmonic behavior of the stress wave (figures 6 and 7). maximum shear stress wave made in nano-bar with circular cross section is as 8 figure 3 illustrates the effect of the material inhomogeneity constant on velocity of stress wave propagation. this figure shows that increasing the material inhomogeneity constant leads to decreasing the mean velocity of stress wave propagation. in fact, the velocity of stress wave propagation is averagely reduced when the heterogeneity of material increases. poisson's effect on velocity of stress wave propagation expresses in figure 4. for small non-dimensional wave number (approximately less than 3), the velocity of stress wave propagation is decreased by increasing poisson's ratio, while for larger non-dimensional wave numbers, the velocity of stress wave propagation is increased by increasing poisson's ratio. also, when the lateral effect is neglected (v=0), the velocity of stress wave propagation becomes equal to a constant value (velocity of stress wave propagation in a homogeneous bernoulli-euler bar). as can be seen in figs. 2-4, for large non-dimensional wave numbers , the velocity of stress wave propagation is increased by increasing , and increasing of for small non-dimensional wave numbers leads to decreasing the velocity of stress wave propagation. according to equation (5), the stress wave made in the nano-bar obtains as , where . variations of real part of the non-dimensional stress wave against non-dimensional wave number with different material length scale parameters under , v=0.25, x=10a and t=0.1s are shown in figure 5. in this figure, the stress wave behavior is completely harmonic except for very small values of . this is because of the fact that when the wave number tends to zero then the incoming wave loses its harmonic vitality and becomes a constant wave (equation (16)). moreover, by increasing , the wave length of stress wave is decreased because of the wave number introduced in equation (16) relates with inverse of the incoming wave length. also, the size effect on stress wave is studied and it is observed that by increasing the material parameter , the stress wave propagated in nano-bar starts its harmonic behavior earlier and leads to increasing of stress wave intensity. similar to what was mentioned for figure 5, the material inhomogeneity constant and poisson's ratio have similar effect on harmonic behavior of the stress wave (figures 6 and 7). maximum shear stress wave made in nano-bar with circular cross section is as (equation 5), where ( ). it should be noted that for circular cross section, we have: . the harmonic behavior of non-dimensional shear stress wave against non-dimensional wave number is shown in figure 8. by increasing , intensity and amplitude of the shear stress increases. this is because of the fact that the shear stress made in nano-bar is caused by lateral inertia, therefore, this is dependent on radius of bar. consequently, by increasing at a given wave number, the radius of bar increases. so, , by increasing , amplitude of the shear stress wave increases. because the behavior of the shear stress wave versus the material parameter, material inhomogeneity constant and poisson's ratio is similar to axial stress wave, evaluation of theses behaviors are not considered. (equation 5), where 8 figure 3 illustrates the effect of the material inhomogeneity constant on velocity of stress wave propagation. this figure shows that increasing the material inhomogeneity constant leads to decreasing the mean velocity of stress wave propagation. in fact, the velocity of stress wave propagation is averagely reduced when the heterogeneity of material increases. poisson's effect on velocity of stress wave propagation expresses in figure 4. for small non-dimensional wave number (approximately less than 3), the velocity of stress wave propagation is decreased by increasing poisson's ratio, while for larger non-dimensional wave numbers, the velocity of stress wave propagation is increased by increasing poisson's ratio. also, when the lateral effect is neglected (v=0), the velocity of stress wave propagation becomes equal to a constant value (velocity of stress wave propagation in a homogeneous bernoulli-euler bar). as can be seen in figs. 2-4, for large non-dimensional wave numbers , the velocity of stress wave propagation is increased by increasing , and increasing of for small non-dimensional wave numbers leads to decreasing the velocity of stress wave propagation. according to equation (5), the stress wave made in the nano-bar obtains as , where . variations of real part of the non-dimensional stress wave against non-dimensional wave number with different material length scale parameters under , v=0.25, x=10a and t=0.1s are shown in figure 5. in this figure, the stress wave behavior is completely harmonic except for very small values of . this is because of the fact that when the wave number tends to zero then the incoming wave loses its harmonic vitality and becomes a constant wave (equation (16)). moreover, by increasing , the wave length of stress wave is decreased because of the wave number introduced in equation (16) relates with inverse of the incoming wave length. also, the size effect on stress wave is studied and it is observed that by increasing the material parameter , the stress wave propagated in nano-bar starts its harmonic behavior earlier and leads to increasing of stress wave intensity. similar to what was mentioned for figure 5, the material inhomogeneity constant and poisson's ratio have similar effect on harmonic behavior of the stress wave (figures 6 and 7). maximum shear stress wave made in nano-bar with circular cross section is as (equation 5), where ( ). it should be noted that for circular cross section, we have: . the harmonic behavior of non-dimensional shear stress wave against non-dimensional wave number is shown in figure 8. by increasing , intensity and amplitude of the shear stress increases. this is because of the fact that the shear stress made in nano-bar is caused by lateral inertia, therefore, this is dependent on radius of bar. consequently, by increasing at a given wave number, the radius of bar increases. so, , by increasing , amplitude of the shear stress wave increases. because the behavior of the shear stress wave versus the material parameter, material inhomogeneity constant and poisson's ratio is similar to axial stress wave, evaluation of theses behaviors are not considered. it should be noted that for circular cross section, we have: 8 figure 3 illustrates the effect of the material inhomogeneity constant on velocity of stress wave propagation. this figure shows that increasing the material inhomogeneity constant leads to decreasing the mean velocity of stress wave propagation. in fact, the velocity of stress wave propagation is averagely reduced when the heterogeneity of material increases. poisson's effect on velocity of stress wave propagation expresses in figure 4. for small non-dimensional wave number (approximately less than 3), the velocity of stress wave propagation is decreased by increasing poisson's ratio, while for larger non-dimensional wave numbers, the velocity of stress wave propagation is increased by increasing poisson's ratio. also, when the lateral effect is neglected (v=0), the velocity of stress wave propagation becomes equal to a constant value (velocity of stress wave propagation in a homogeneous bernoulli-euler bar). as can be seen in figs. 2-4, for large non-dimensional wave numbers , the velocity of stress wave propagation is increased by increasing , and increasing of for small non-dimensional wave numbers leads to decreasing the velocity of stress wave propagation. according to equation (5), the stress wave made in the nano-bar obtains as , where . variations of real part of the non-dimensional stress wave against non-dimensional wave number with different material length scale parameters under , v=0.25, x=10a and t=0.1s are shown in figure 5. in this figure, the stress wave behavior is completely harmonic except for very small values of . this is because of the fact that when the wave number tends to zero then the incoming wave loses its harmonic vitality and becomes a constant wave (equation (16)). moreover, by increasing , the wave length of stress wave is decreased because of the wave number introduced in equation (16) relates with inverse of the incoming wave length. also, the size effect on stress wave is studied and it is observed that by increasing the material parameter , the stress wave propagated in nano-bar starts its harmonic behavior earlier and leads to increasing of stress wave intensity. similar to what was mentioned for figure 5, the material inhomogeneity constant and poisson's ratio have similar effect on harmonic behavior of the stress wave (figures 6 and 7). maximum shear stress wave made in nano-bar with circular cross section is as (equation 5), where ( ). it should be noted that for circular cross section, we have: . the harmonic behavior of non-dimensional shear stress wave against non-dimensional wave number is shown in figure 8. by increasing , intensity and amplitude of the shear stress increases. this is because of the fact that the shear stress made in nano-bar is caused by lateral inertia, therefore, this is dependent on radius of bar. consequently, by increasing at a given wave number, the radius of bar increases. so, , by increasing , amplitude of the shear stress wave increases. because the behavior of the shear stress wave versus the material parameter, material inhomogeneity constant and poisson's ratio is similar to axial stress wave, evaluation of theses behaviors are not considered. . the harmonic behavior of non-dimensional shear stress wave against non-dimensional wave number is shown in figure 8. by increasing 8 figure 3 illustrates the effect of the material inhomogeneity constant on velocity of stress wave propagation. this figure shows that increasing the material inhomogeneity constant leads to decreasing the mean velocity of stress wave propagation. in fact, the velocity of stress wave propagation is averagely reduced when the heterogeneity of material increases. poisson's effect on velocity of stress wave propagation expresses in figure 4. for small non-dimensional wave number (approximately less than 3), the velocity of stress wave propagation is decreased by increasing poisson's ratio, while for larger non-dimensional wave numbers, the velocity of stress wave propagation is increased by increasing poisson's ratio. also, when the lateral effect is neglected (v=0), the velocity of stress wave propagation becomes equal to a constant value (velocity of stress wave propagation in a homogeneous bernoulli-euler bar). as can be seen in figs. 2-4, for large non-dimensional wave numbers , the velocity of stress wave propagation is increased by increasing , and increasing of for small non-dimensional wave numbers leads to decreasing the velocity of stress wave propagation. according to equation (5), the stress wave made in the nano-bar obtains as , where . variations of real part of the non-dimensional stress wave against non-dimensional wave number with different material length scale parameters under , v=0.25, x=10a and t=0.1s are shown in figure 5. in this figure, the stress wave behavior is completely harmonic except for very small values of . this is because of the fact that when the wave number tends to zero then the incoming wave loses its harmonic vitality and becomes a constant wave (equation (16)). moreover, by increasing , the wave length of stress wave is decreased because of the wave number introduced in equation (16) relates with inverse of the incoming wave length. also, the size effect on stress wave is studied and it is observed that by increasing the material parameter , the stress wave propagated in nano-bar starts its harmonic behavior earlier and leads to increasing of stress wave intensity. similar to what was mentioned for figure 5, the material inhomogeneity constant and poisson's ratio have similar effect on harmonic behavior of the stress wave (figures 6 and 7). maximum shear stress wave made in nano-bar with circular cross section is as (equation 5), where ( ). it should be noted that for circular cross section, we have: . the harmonic behavior of non-dimensional shear stress wave against non-dimensional wave number is shown in figure 8. by increasing , intensity and amplitude of the shear stress increases. this is because of the fact that the shear stress made in nano-bar is caused by lateral inertia, therefore, this is dependent on radius of bar. consequently, by increasing at a given wave number, the radius of bar increases. so, , by increasing , amplitude of the shear stress wave increases. because the behavior of the shear stress wave versus the material parameter, material inhomogeneity constant and poisson's ratio is similar to axial stress wave, evaluation of theses behaviors are not considered. , intensity and amplitude of the shear stress increases. this is because of the fact that the shear stress made in nano-bar is caused by lateral inertia, therefore, this is dependent on radius of bar. consequently, by increasing 8 figure 3 illustrates the effect of the material inhomogeneity constant on velocity of stress wave propagation. this figure shows that increasing the material inhomogeneity constant leads to decreasing the mean velocity of stress wave propagation. in fact, the velocity of stress wave propagation is averagely reduced when the heterogeneity of material increases. poisson's effect on velocity of stress wave propagation expresses in figure 4. for small non-dimensional wave number (approximately less than 3), the velocity of stress wave propagation is decreased by increasing poisson's ratio, while for larger non-dimensional wave numbers, the velocity of stress wave propagation is increased by increasing poisson's ratio. also, when the lateral effect is neglected (v=0), the velocity of stress wave propagation becomes equal to a constant value (velocity of stress wave propagation in a homogeneous bernoulli-euler bar). as can be seen in figs. 2-4, for large non-dimensional wave numbers , the velocity of stress wave propagation is increased by increasing , and increasing of for small non-dimensional wave numbers leads to decreasing the velocity of stress wave propagation. according to equation (5), the stress wave made in the nano-bar obtains as , where . variations of real part of the non-dimensional stress wave against non-dimensional wave number with different material length scale parameters under , v=0.25, x=10a and t=0.1s are shown in figure 5. in this figure, the stress wave behavior is completely harmonic except for very small values of . this is because of the fact that when the wave number tends to zero then the incoming wave loses its harmonic vitality and becomes a constant wave (equation (16)). moreover, by increasing , the wave length of stress wave is decreased because of the wave number introduced in equation (16) relates with inverse of the incoming wave length. also, the size effect on stress wave is studied and it is observed that by increasing the material parameter , the stress wave propagated in nano-bar starts its harmonic behavior earlier and leads to increasing of stress wave intensity. similar to what was mentioned for figure 5, the material inhomogeneity constant and poisson's ratio have similar effect on harmonic behavior of the stress wave (figures 6 and 7). maximum shear stress wave made in nano-bar with circular cross section is as (equation 5), where ( ). it should be noted that for circular cross section, we have: . the harmonic behavior of non-dimensional shear stress wave against non-dimensional wave number is shown in figure 8. by increasing , intensity and amplitude of the shear stress increases. this is because of the fact that the shear stress made in nano-bar is caused by lateral inertia, therefore, this is dependent on radius of bar. consequently, by increasing at a given wave number, the radius of bar increases. so, , by increasing , amplitude of the shear stress wave increases. because the behavior of the shear stress wave versus the material parameter, material inhomogeneity constant and poisson's ratio is similar to axial stress wave, evaluation of theses behaviors are not considered. at a given wave number, the radius of bar increases. so, by increasing 8 figure 3 illustrates the effect of the material inhomogeneity constant on velocity of stress wave propagation. this figure shows that increasing the material inhomogeneity constant leads to decreasing the mean velocity of stress wave propagation. in fact, the velocity of stress wave propagation is averagely reduced when the heterogeneity of material increases. poisson's effect on velocity of stress wave propagation expresses in figure 4. for small non-dimensional wave number (approximately less than 3), the velocity of stress wave propagation is decreased by increasing poisson's ratio, while for larger non-dimensional wave numbers, the velocity of stress wave propagation is increased by increasing poisson's ratio. also, when the lateral effect is neglected (v=0), the velocity of stress wave propagation becomes equal to a constant value (velocity of stress wave propagation in a homogeneous bernoulli-euler bar). as can be seen in figs. 2-4, for large non-dimensional wave numbers , the velocity of stress wave propagation is increased by increasing , and increasing of for small non-dimensional wave numbers leads to decreasing the velocity of stress wave propagation. according to equation (5), the stress wave made in the nano-bar obtains as , where . variations of real part of the non-dimensional stress wave against non-dimensional wave number with different material length scale parameters under , v=0.25, x=10a and t=0.1s are shown in figure 5. in this figure, the stress wave behavior is completely harmonic except for very small values of . this is because of the fact that when the wave number tends to zero then the incoming wave loses its harmonic vitality and becomes a constant wave (equation (16)). moreover, by increasing , the wave length of stress wave is decreased because of the wave number introduced in equation (16) relates with inverse of the incoming wave length. also, the size effect on stress wave is studied and it is observed that by increasing the material parameter , the stress wave propagated in nano-bar starts its harmonic behavior earlier and leads to increasing of stress wave intensity. similar to what was mentioned for figure 5, the material inhomogeneity constant and poisson's ratio have similar effect on harmonic behavior of the stress wave (figures 6 and 7). maximum shear stress wave made in nano-bar with circular cross section is as (equation 5), where ( ). it should be noted that for circular cross section, we have: . the harmonic behavior of non-dimensional shear stress wave against non-dimensional wave number is shown in figure 8. by increasing , intensity and amplitude of the shear stress increases. this is because of the fact that the shear stress made in nano-bar is caused by lateral inertia, therefore, this is dependent on radius of bar. consequently, by increasing at a given wave number, the radius of bar increases. so, , by increasing , amplitude of the shear stress wave increases. because the behavior of the shear stress wave versus the material parameter, material inhomogeneity constant and poisson's ratio is similar to axial stress wave, evaluation of theses behaviors are not considered. , amplitude of the shear stress wave increases. because the behavior of the shear stress wave versus the material parameter, material inhomogeneity constant and poisson’s ratio is similar to axial stress wave, evaluation of theses behaviors are not considered. issn: 2180-1053 vol. 7 no. 1 january june 2015 journal of mechanical engineering and technology 52 9 figure 4. poisson's effect on velocity of stress wave propagation with and . figure 5. behavior of non-dimensional axial stress wave versus non-dimensional wave number with different material length scale parameter under , v=0.25, x=10a and t=0.1s. figure 4. poisson’s effect on velocity of stress wave propagation with 9 figure 4. poisson's effect on velocity of stress wave propagation with and . figure 5. behavior of non-dimensional axial stress wave versus non-dimensional wave number with different material length scale parameter under , v=0.25, x=10a and t=0.1s. . 9 figure 4. poisson's effect on velocity of stress wave propagation with and . figure 5. behavior of non-dimensional axial stress wave versus non-dimensional wave number with different material length scale parameter under , v=0.25, x=10a and t=0.1s. figure 5. behavior of non-dimensional axial stress wave versus non-dimensional wave number with different material length scale parameter under 9 figure 4. poisson's effect on velocity of stress wave propagation with and . figure 5. behavior of non-dimensional axial stress wave versus non-dimensional wave number with different material length scale parameter under , v=0.25, x=10a and t=0.1s. . issn: 2180-1053 vol. 7 no. 1 january june 2015 propagation of stress wave in a functionally graded nano-bar based on modified couple stress theory 53 10 figure 6. behavior of non-dimensional axial stress wave versus non-dimensional wave number with different material inhomogeneity constant under , v=0.25, x=10a and t=0.1s. figure 7. behavior of non-dimensional axial stress wave versus non-dimensional wave number with different poisson's ratio under , , x=10a and t=0.1s. figure 6. behavior of non-dimensional axial stress wave versus nondimensional wave number with different material inhomogeneity constant under 10 figure 6. behavior of non-dimensional axial stress wave versus non-dimensional wave number with different material inhomogeneity constant under , v=0.25, x=10a and t=0.1s. figure 7. behavior of non-dimensional axial stress wave versus non-dimensional wave number with different poisson's ratio under , , x=10a and t=0.1s. . 10 figure 6. behavior of non-dimensional axial stress wave versus non-dimensional wave number with different material inhomogeneity constant under , v=0.25, x=10a and t=0.1s. figure 7. behavior of non-dimensional axial stress wave versus non-dimensional wave number with different poisson's ratio under , , x=10a and t=0.1s. figure 7. behavior of non-dimensional axial stress wave versus non-dimensional wave number with different poisson’s ratio under 10 figure 6. behavior of non-dimensional axial stress wave versus non-dimensional wave number with different material inhomogeneity constant under , v=0.25, x=10a and t=0.1s. figure 7. behavior of non-dimensional axial stress wave versus non-dimensional wave number with different poisson's ratio under , , x=10a and t=0.1s. issn: 2180-1053 vol. 7 no. 1 january june 2015 journal of mechanical engineering and technology 54 11 figure 8. behavior of non-dimensional shear stress wave versus non-dimensional wave number with different material length scale parameter under , , v=0.25, x=10a and t=0.1s. 4.0 conclusions propagation of longitudinal stress wave in an fg nano-bar which graded longitudinally is studied in this paper. the equation of motion is derived using modified couple stress theory, hamilton's principle and love rod theory. the velocity of stress wave propagation of the nano-bar is obtained as a function of poisson's ratio, material length scale parameter and material inhomogeneity constant by a direct solution of the equation of motion. the following results are concluded from analysis of the stress wave by the mentioned parameters. behavior of the stress wave propagation of the nano-bar is a size-dependent behavior and this dependency exposes using the material length scale parameter . the numerical results show that by increasing the , the velocity and intensity of the stress wave are increased. moreover, neglecting of material length scale parameter (use of classical theory, ) leads to considerable errors. thereupon, the inability of the classical theory to analyse the micro/nanostructures is confirmed. the non-dimensional stress wave against the non-dimensional wave number behaves harmoniously and by increasing non-dimensional wave number the wave length of the stress wave is decreased. also, when tend to zero, the stress wave loses its harmonic behavior and consequently the stress wave becomes constant. by variation of the material inhomogeneity constant in graded structures can be derived the velocity of the wave and behavior of the stress wave. the results show that the graded materials have a less velocity than homogeneous materials ( ). also, by figure 8. behavior of non-dimensional shear stress wave versus non-dimensional wave number with different material length scale parameter under 11 figure 8. behavior of non-dimensional shear stress wave versus non-dimensional wave number with different material length scale parameter under , , v=0.25, x=10a and t=0.1s. 4.0 conclusions propagation of longitudinal stress wave in an fg nano-bar which graded longitudinally is studied in this paper. the equation of motion is derived using modified couple stress theory, hamilton's principle and love rod theory. the velocity of stress wave propagation of the nano-bar is obtained as a function of poisson's ratio, material length scale parameter and material inhomogeneity constant by a direct solution of the equation of motion. the following results are concluded from analysis of the stress wave by the mentioned parameters. behavior of the stress wave propagation of the nano-bar is a size-dependent behavior and this dependency exposes using the material length scale parameter . the numerical results show that by increasing the , the velocity and intensity of the stress wave are increased. moreover, neglecting of material length scale parameter (use of classical theory, ) leads to considerable errors. thereupon, the inability of the classical theory to analyse the micro/nanostructures is confirmed. the non-dimensional stress wave against the non-dimensional wave number behaves harmoniously and by increasing non-dimensional wave number the wave length of the stress wave is decreased. also, when tend to zero, the stress wave loses its harmonic behavior and consequently the stress wave becomes constant. by variation of the material inhomogeneity constant in graded structures can be derived the velocity of the wave and behavior of the stress wave. the results show that the graded materials have a less velocity than homogeneous materials ( ). also, by . 4.0 conclusions propagation of longitudinal stress wave in an fg nano-bar which graded longitudinally is studied in this paper. the equation of motion is derived using modified couple stress theory, hamilton’s principle and love rod theory. the velocity of stress wave propagation of the nanobar is obtained as a function of poisson’s ratio, material length scale parameter and material inhomogeneity constant by a direct solution of the equation of motion. the following results are concluded from analysis of the stress wave by the mentioned parameters. behavior of the stress wave propagation of the nano-bar is a size-dependent behavior and this dependency exposes using the material length scale parameter 11 figure 8. behavior of non-dimensional shear stress wave versus non-dimensional wave number with different material length scale parameter under , , v=0.25, x=10a and t=0.1s. 4.0 conclusions propagation of longitudinal stress wave in an fg nano-bar which graded longitudinally is studied in this paper. the equation of motion is derived using modified couple stress theory, hamilton's principle and love rod theory. the velocity of stress wave propagation of the nano-bar is obtained as a function of poisson's ratio, material length scale parameter and material inhomogeneity constant by a direct solution of the equation of motion. the following results are concluded from analysis of the stress wave by the mentioned parameters. behavior of the stress wave propagation of the nano-bar is a size-dependent behavior and this dependency exposes using the material length scale parameter . the numerical results show that by increasing the , the velocity and intensity of the stress wave are increased. moreover, neglecting of material length scale parameter (use of classical theory, ) leads to considerable errors. thereupon, the inability of the classical theory to analyse the micro/nanostructures is confirmed. the non-dimensional stress wave against the non-dimensional wave number behaves harmoniously and by increasing non-dimensional wave number the wave length of the stress wave is decreased. also, when tend to zero, the stress wave loses its harmonic behavior and consequently the stress wave becomes constant. by variation of the material inhomogeneity constant in graded structures can be derived the velocity of the wave and behavior of the stress wave. the results show that the graded materials have a less velocity than homogeneous materials ( ). also, by . the numerical results show that by increasing the 11 figure 8. behavior of non-dimensional shear stress wave versus non-dimensional wave number with different material length scale parameter under , , v=0.25, x=10a and t=0.1s. 4.0 conclusions propagation of longitudinal stress wave in an fg nano-bar which graded longitudinally is studied in this paper. the equation of motion is derived using modified couple stress theory, hamilton's principle and love rod theory. the velocity of stress wave propagation of the nano-bar is obtained as a function of poisson's ratio, material length scale parameter and material inhomogeneity constant by a direct solution of the equation of motion. the following results are concluded from analysis of the stress wave by the mentioned parameters. behavior of the stress wave propagation of the nano-bar is a size-dependent behavior and this dependency exposes using the material length scale parameter . the numerical results show that by increasing the , the velocity and intensity of the stress wave are increased. moreover, neglecting of material length scale parameter (use of classical theory, ) leads to considerable errors. thereupon, the inability of the classical theory to analyse the micro/nanostructures is confirmed. the non-dimensional stress wave against the non-dimensional wave number behaves harmoniously and by increasing non-dimensional wave number the wave length of the stress wave is decreased. also, when tend to zero, the stress wave loses its harmonic behavior and consequently the stress wave becomes constant. by variation of the material inhomogeneity constant in graded structures can be derived the velocity of the wave and behavior of the stress wave. the results show that the graded materials have a less velocity than homogeneous materials ( ). also, by , the velocity and intensity of the stress wave are increased. moreover, neglecting of material length scale parameter (use of classical theory, 11 figure 8. behavior of non-dimensional shear stress wave versus non-dimensional wave number with different material length scale parameter under , , v=0.25, x=10a and t=0.1s. 4.0 conclusions propagation of longitudinal stress wave in an fg nano-bar which graded longitudinally is studied in this paper. the equation of motion is derived using modified couple stress theory, hamilton's principle and love rod theory. the velocity of stress wave propagation of the nano-bar is obtained as a function of poisson's ratio, material length scale parameter and material inhomogeneity constant by a direct solution of the equation of motion. the following results are concluded from analysis of the stress wave by the mentioned parameters. behavior of the stress wave propagation of the nano-bar is a size-dependent behavior and this dependency exposes using the material length scale parameter . the numerical results show that by increasing the , the velocity and intensity of the stress wave are increased. moreover, neglecting of material length scale parameter (use of classical theory, ) leads to considerable errors. thereupon, the inability of the classical theory to analyse the micro/nanostructures is confirmed. the non-dimensional stress wave against the non-dimensional wave number behaves harmoniously and by increasing non-dimensional wave number the wave length of the stress wave is decreased. also, when tend to zero, the stress wave loses its harmonic behavior and consequently the stress wave becomes constant. by variation of the material inhomogeneity constant in graded structures can be derived the velocity of the wave and behavior of the stress wave. the results show that the graded materials have a less velocity than homogeneous materials ( ). also, by ) leads to considerable errors. thereupon, the inability of the classical theory to analyse the micro/nanostructures is confirmed. the non-dimensional stress wave against the non-dimensional wave number behaves harmoniously and by increasing non-dimensional wave number 8 figure 3 illustrates the effect of the material inhomogeneity constant on velocity of stress wave propagation. this figure shows that increasing the material inhomogeneity constant leads to decreasing the mean velocity of stress wave propagation. in fact, the velocity of stress wave propagation is averagely reduced when the heterogeneity of material increases. poisson's effect on velocity of stress wave propagation expresses in figure 4. for small non-dimensional wave number (approximately less than 3), the velocity of stress wave propagation is decreased by increasing poisson's ratio, while for larger non-dimensional wave numbers, the velocity of stress wave propagation is increased by increasing poisson's ratio. also, when the lateral effect is neglected (v=0), the velocity of stress wave propagation becomes equal to a constant value (velocity of stress wave propagation in a homogeneous bernoulli-euler bar). as can be seen in figs. 2-4, for large non-dimensional wave numbers , the velocity of stress wave propagation is increased by increasing , and increasing of for small non-dimensional wave numbers leads to decreasing the velocity of stress wave propagation. according to equation (5), the stress wave made in the nano-bar obtains as , where . variations of real part of the non-dimensional stress wave against non-dimensional wave number with different material length scale parameters under , v=0.25, x=10a and t=0.1s are shown in figure 5. in this figure, the stress wave behavior is completely harmonic except for very small values of . this is because of the fact that when the wave number tends to zero then the incoming wave loses its harmonic vitality and becomes a constant wave (equation (16)). moreover, by increasing , the wave length of stress wave is decreased because of the wave number introduced in equation (16) relates with inverse of the incoming wave length. also, the size effect on stress wave is studied and it is observed that by increasing the material parameter , the stress wave propagated in nano-bar starts its harmonic behavior earlier and leads to increasing of stress wave intensity. similar to what was mentioned for figure 5, the material inhomogeneity constant and poisson's ratio have similar effect on harmonic behavior of the stress wave (figures 6 and 7). maximum shear stress wave made in nano-bar with circular cross section is as (equation 5), where ( ). it should be noted that for circular cross section, we have: . the harmonic behavior of non-dimensional shear stress wave against non-dimensional wave number is shown in figure 8. by increasing , intensity and amplitude of the shear stress increases. this is because of the fact that the shear stress made in nano-bar is caused by lateral inertia, therefore, this is dependent on radius of bar. consequently, by increasing at a given wave number, the radius of bar increases. so, , by increasing , amplitude of the shear stress wave increases. because the behavior of the shear stress wave versus the material parameter, material inhomogeneity constant and poisson's ratio is similar to axial stress wave, evaluation of theses behaviors are not considered. the wave length of the stress wave is decreased. also, when 8 figure 3 illustrates the effect of the material inhomogeneity constant on velocity of stress wave propagation. this figure shows that increasing the material inhomogeneity constant leads to decreasing the mean velocity of stress wave propagation. in fact, the velocity of stress wave propagation is averagely reduced when the heterogeneity of material increases. poisson's effect on velocity of stress wave propagation expresses in figure 4. for small non-dimensional wave number (approximately less than 3), the velocity of stress wave propagation is decreased by increasing poisson's ratio, while for larger non-dimensional wave numbers, the velocity of stress wave propagation is increased by increasing poisson's ratio. also, when the lateral effect is neglected (v=0), the velocity of stress wave propagation becomes equal to a constant value (velocity of stress wave propagation in a homogeneous bernoulli-euler bar). as can be seen in figs. 2-4, for large non-dimensional wave numbers , the velocity of stress wave propagation is increased by increasing , and increasing of for small non-dimensional wave numbers leads to decreasing the velocity of stress wave propagation. according to equation (5), the stress wave made in the nano-bar obtains as , where . variations of real part of the non-dimensional stress wave against non-dimensional wave number with different material length scale parameters under , v=0.25, x=10a and t=0.1s are shown in figure 5. in this figure, the stress wave behavior is completely harmonic except for very small values of . this is because of the fact that when the wave number tends to zero then the incoming wave loses its harmonic vitality and becomes a constant wave (equation (16)). moreover, by increasing , the wave length of stress wave is decreased because of the wave number introduced in equation (16) relates with inverse of the incoming wave length. also, the size effect on stress wave is studied and it is observed that by increasing the material parameter , the stress wave propagated in nano-bar starts its harmonic behavior earlier and leads to increasing of stress wave intensity. similar to what was mentioned for figure 5, the material inhomogeneity constant and poisson's ratio have similar effect on harmonic behavior of the stress wave (figures 6 and 7). maximum shear stress wave made in nano-bar with circular cross section is as (equation 5), where ( ). it should be noted that for circular cross section, we have: . the harmonic behavior of non-dimensional shear stress wave against non-dimensional wave number is shown in figure 8. by increasing , intensity and amplitude of the shear stress increases. this is because of the fact that the shear stress made in nano-bar is caused by lateral inertia, therefore, this is dependent on radius of bar. consequently, by increasing at a given wave number, the radius of bar increases. so, , by increasing , amplitude of the shear stress wave increases. because the behavior of the shear stress wave versus the material parameter, material inhomogeneity constant and poisson's ratio is similar to axial stress wave, evaluation of theses behaviors are not considered. tend to zero, the stress wave loses its harmonic behavior and consequently the stress wave becomes constant. issn: 2180-1053 vol. 7 no. 1 january june 2015 propagation of stress wave in a functionally graded nano-bar based on modified couple stress theory 55 by variation of the material inhomogeneity constant 11 figure 8. behavior of non-dimensional shear stress wave versus non-dimensional wave number with different material length scale parameter under , , v=0.25, x=10a and t=0.1s. 4.0 conclusions propagation of longitudinal stress wave in an fg nano-bar which graded longitudinally is studied in this paper. the equation of motion is derived using modified couple stress theory, hamilton's principle and love rod theory. the velocity of stress wave propagation of the nano-bar is obtained as a function of poisson's ratio, material length scale parameter and material inhomogeneity constant by a direct solution of the equation of motion. the following results are concluded from analysis of the stress wave by the mentioned parameters. behavior of the stress wave propagation of the nano-bar is a size-dependent behavior and this dependency exposes using the material length scale parameter . the numerical results show that by increasing the , the velocity and intensity of the stress wave are increased. moreover, neglecting of material length scale parameter (use of classical theory, ) leads to considerable errors. thereupon, the inability of the classical theory to analyse the micro/nanostructures is confirmed. the non-dimensional stress wave against the non-dimensional wave number behaves harmoniously and by increasing non-dimensional wave number the wave length of the stress wave is decreased. also, when tend to zero, the stress wave loses its harmonic behavior and consequently the stress wave becomes constant. by variation of the material inhomogeneity constant in graded structures can be derived the velocity of the wave and behavior of the stress wave. the results show that the graded materials have a less velocity than homogeneous materials ( ). also, by in graded structures can be derived the velocity of the wave and behavior of the stress wave. the results show that the graded materials have a less velocity than homogeneous materials 11 figure 8. behavior of non-dimensional shear stress wave versus non-dimensional wave number with different material length scale parameter under , , v=0.25, x=10a and t=0.1s. 4.0 conclusions propagation of longitudinal stress wave in an fg nano-bar which graded longitudinally is studied in this paper. the equation of motion is derived using modified couple stress theory, hamilton's principle and love rod theory. the velocity of stress wave propagation of the nano-bar is obtained as a function of poisson's ratio, material length scale parameter and material inhomogeneity constant by a direct solution of the equation of motion. the following results are concluded from analysis of the stress wave by the mentioned parameters. behavior of the stress wave propagation of the nano-bar is a size-dependent behavior and this dependency exposes using the material length scale parameter . the numerical results show that by increasing the , the velocity and intensity of the stress wave are increased. moreover, neglecting of material length scale parameter (use of classical theory, ) leads to considerable errors. thereupon, the inability of the classical theory to analyse the micro/nanostructures is confirmed. the non-dimensional stress wave against the non-dimensional wave number behaves harmoniously and by increasing non-dimensional wave number the wave length of the stress wave is decreased. also, when tend to zero, the stress wave loses its harmonic behavior and consequently the stress wave becomes constant. by variation of the material inhomogeneity constant in graded structures can be derived the velocity of the wave and behavior of the stress wave. the results show that the graded materials have a less velocity than homogeneous materials ( ). also, by . also, by increasing , velocity of the stress wave propagation is decreased, but the harmonic behavior of the stress wave occurs earlier. the results show that neglecting the lateral effct (v=0) leads to make the considerable error in the impact behavior of structures. for small 8 figure 3 illustrates the effect of the material inhomogeneity constant on velocity of stress wave propagation. this figure shows that increasing the material inhomogeneity constant leads to decreasing the mean velocity of stress wave propagation. in fact, the velocity of stress wave propagation is averagely reduced when the heterogeneity of material increases. poisson's effect on velocity of stress wave propagation expresses in figure 4. for small non-dimensional wave number (approximately less than 3), the velocity of stress wave propagation is decreased by increasing poisson's ratio, while for larger non-dimensional wave numbers, the velocity of stress wave propagation is increased by increasing poisson's ratio. also, when the lateral effect is neglected (v=0), the velocity of stress wave propagation becomes equal to a constant value (velocity of stress wave propagation in a homogeneous bernoulli-euler bar). as can be seen in figs. 2-4, for large non-dimensional wave numbers , the velocity of stress wave propagation is increased by increasing , and increasing of for small non-dimensional wave numbers leads to decreasing the velocity of stress wave propagation. according to equation (5), the stress wave made in the nano-bar obtains as , where . variations of real part of the non-dimensional stress wave against non-dimensional wave number with different material length scale parameters under , v=0.25, x=10a and t=0.1s are shown in figure 5. in this figure, the stress wave behavior is completely harmonic except for very small values of . this is because of the fact that when the wave number tends to zero then the incoming wave loses its harmonic vitality and becomes a constant wave (equation (16)). moreover, by increasing , the wave length of stress wave is decreased because of the wave number introduced in equation (16) relates with inverse of the incoming wave length. also, the size effect on stress wave is studied and it is observed that by increasing the material parameter , the stress wave propagated in nano-bar starts its harmonic behavior earlier and leads to increasing of stress wave intensity. similar to what was mentioned for figure 5, the material inhomogeneity constant and poisson's ratio have similar effect on harmonic behavior of the stress wave (figures 6 and 7). maximum shear stress wave made in nano-bar with circular cross section is as (equation 5), where ( ). it should be noted that for circular cross section, we have: . the harmonic behavior of non-dimensional shear stress wave against non-dimensional wave number is shown in figure 8. by increasing , intensity and amplitude of the shear stress increases. this is because of the fact that the shear stress made in nano-bar is caused by lateral inertia, therefore, this is dependent on radius of bar. consequently, by increasing at a given wave number, the radius of bar increases. so, , by increasing , amplitude of the shear stress wave increases. because the behavior of the shear stress wave versus the material parameter, material inhomogeneity constant and poisson's ratio is similar to axial stress wave, evaluation of theses behaviors are not considered. , increasing the poisson’s ratio estimates less velocity for the stress wave; and for large 8 figure 3 illustrates the effect of the material inhomogeneity constant on velocity of stress wave propagation. this figure shows that increasing the material inhomogeneity constant leads to decreasing the mean velocity of stress wave propagation. in fact, the velocity of stress wave propagation is averagely reduced when the heterogeneity of material increases. poisson's effect on velocity of stress wave propagation expresses in figure 4. for small non-dimensional wave number (approximately less than 3), the velocity of stress wave propagation is decreased by increasing poisson's ratio, while for larger non-dimensional wave numbers, the velocity of stress wave propagation is increased by increasing poisson's ratio. also, when the lateral effect is neglected (v=0), the velocity of stress wave propagation becomes equal to a constant value (velocity of stress wave propagation in a homogeneous bernoulli-euler bar). as can be seen in figs. 2-4, for large non-dimensional wave numbers , the velocity of stress wave propagation is increased by increasing , and increasing of for small non-dimensional wave numbers leads to decreasing the velocity of stress wave propagation. according to equation (5), the stress wave made in the nano-bar obtains as , where . variations of real part of the non-dimensional stress wave against non-dimensional wave number with different material length scale parameters under , v=0.25, x=10a and t=0.1s are shown in figure 5. in this figure, the stress wave behavior is completely harmonic except for very small values of . this is because of the fact that when the wave number tends to zero then the incoming wave loses its harmonic vitality and becomes a constant wave (equation (16)). moreover, by increasing , the wave length of stress wave is decreased because of the wave number introduced in equation (16) relates with inverse of the incoming wave length. also, the size effect on stress wave is studied and it is observed that by increasing the material parameter , the stress wave propagated in nano-bar starts its harmonic behavior earlier and leads to increasing of stress wave intensity. similar to what was mentioned for figure 5, the material inhomogeneity constant and poisson's ratio have similar effect on harmonic behavior of the stress wave (figures 6 and 7). maximum shear stress wave made in nano-bar with circular cross section is as (equation 5), where ( ). it should be noted that for circular cross section, we have: . the harmonic behavior of non-dimensional shear stress wave against non-dimensional wave number is shown in figure 8. by increasing , intensity and amplitude of the shear stress increases. this is because of the fact that the shear stress made in nano-bar is caused by lateral inertia, therefore, this is dependent on radius of bar. consequently, by increasing at a given wave number, the radius of bar increases. so, , by increasing , amplitude of the shear stress wave increases. because the behavior of the shear stress wave versus the material parameter, material inhomogeneity constant and poisson's ratio is similar to axial stress wave, evaluation of theses behaviors are not considered. , increasing the piosson’s ratio leads to increase the velocity of stress wave propagation. also, increasing the poisson’s ratio leads to the generated stress wave arrives to its harmonic behavior earlier. 5.0 references anderson, s. p. (2006). higher-order rod approximations for the propagation of longitudinal stress waves in elastic bars. journal of sound and vibration, 290, 290-308. fowles, r., & williams, r. f. (1970). plane stress wave propagation in solids. journal of applied physics, 41, 360-363. güven, u. (2011). the investigation of the nonlocal longitudinal stress waves with modified couple stress theory. acta mechanica, 221, 321-325. güven, u. (2012). a more general investigation for the longitudinal stress waves in microrods with initial stress. acta mechanica, 223, 2065–2074. güven, u. (2014). a generalized nonlocal elasticity solution for the propagation of longitudinal stress waves in bars. european journal of mechanics a/ solids, 45,75-79. jones, n. (1989). stractural impact (1st ed.). london, uk: cambridge university. jung, w. y., han, s. ch., & park, w. t. (2014). a modified couple stress theory for buckling analysis of s-fgm nanoplates embedded in pasternak elastic medium. composites part-b engineering, 60, 746-756. kaishin, l., & bin, l. (2001). a numerical solution of torsional stress wave propagation in layered orthotropic bar of rectangular cross-section. international journal of solids and structures, 38, 8929-8940. ke, l. l., & wang, y. sh. (2011). size effect on dynamic stability of functionally graded microbeams based on a modified couple stress theory. composites structures, 93, 342-350. love, a. e. h. (1944). a treatise on the mathematical theory of elasticity. new york, usa: dover publications. issn: 2180-1053 vol. 7 no. 1 january june 2015 journal of mechanical engineering and technology 56 mindlin, r. d. (1963). influence of couple-stresses on stress concentrations. experimental mechanics, 3, 1–7. mindlin, r. d. (1964). microstructure in linear elasticity. archive for rational mechanics and analysis, 16, 51–78. park, s. k., & gao, x. l. (2006). bernoulli–euler beam model based on a modified couple stress theory. journal of micromechanics and microengineering, 16, 2355–2359. qiao, p., yang, m., & bobaru, f. (2008). impact mechanics and high-energy absorbing materials: review. journal of aerospace engineering, 21(4), 235-248. reddy, j. n. (2011). microstructure-dependant couple stress theories of functionally graded beams. journal of the mechanics and physics of solids, 59, 2382–2399. salamat-talab, m., nateghi, a., & torabi, j. (2012). static and dynamic analysis of third-order shear deformation fg micro beam based on modified couple stress theory. international journal of mechanical sciences, 57, 6373. shaat, m., mahmoud, f. f., alshorbagy, a. e., alieldin, s. s., & meletis, e. i. (2012). size-dependent analysis of functionally graded ultra-thin films. structural engineering mechanics, 44(4), 431-448. shariat, m., khaghani, m., & lavasani, s. m. h. (2010). nonlinear thermoelasticity, vibration, and stress wave propagation analyses of thick fgm cylinders with temperature-dependant material properties. european journal of mechanics a/solids, 29, 378-391. shen, y., & yin, x. (2014). dynamic substructure analysis of stress waves generated by impacts on non-uniform rod structures. mechanism and machine theory, 74, 154-172. stronge, w. j. (2000). impact mechanics (1st ed.). london, uk: cambridge university. thai, h. t., & choi, d. h. (2013). size-dependent functionally graded kirchhoff and mindlin plate models based on a modified couple stress theory. composites structures, 95, 142–153. yang, f., chong, a. m., lam, d. c. c., & tong, p. (2002). couple stress based strain gradient theory of elasticity. international journal of solids and structures, 39, 2731-2743. preparation of papers in a two column model paper format issn: 2180-1053 vol. 8 no.1 january – june 2016 77 oscillatory flow across plates with different shape of edges c. weiyang 1 and f. a. z. mohd saat 1,2* 1 faculty of mechanical engineering, universiti teknikal malaysia melaka, hang tuah jaya, 76100 durian tunggal, melaka, malaysia. 2 centre for advanced research on energy, universiti teknikal malaysia melaka, hang tuah jaya, 76100 durian tunggal, melaka, malaysia. abstract oscillatory flow is the type of flow found in the greener thermoacoustic based technologies. understanding the behavior of the less understood oscillatory flow of this kind is one of the key feature for the success of the system. heat exchanger is one of the important part of the system. in this study, oscillatory flow across pile of hot and cold parallel-plates heat exchanger with three different shape of edges (i.e. rectangular, round and triangular shape of edges) were investigated. a suitable computational model was created in ansys. the results were compared to theoretical predictions and a good match was found. the study shows that the shape of the edge affects the flow and heat transfer of the system. a triangle-shaped edges with shorter length provides the higher heat transfer between plates and the oscillating fluid compared to plates with round and square edges. the results indicated that the entrance effect could be the reason for the change of heat transfer performance as the shape of edge changes. keywords: oscillatory flow; thermoacoustics; heat exchanger; cfd. 1.0 introduction oscillatory flow is a cyclic flow found in engineering applications such as reactor and thermoacoustic systems. thermoacoustic has been introduced into the industries for centuries. it can convert heat energy to work using thermoacoustic principles. thermoacoustic, as the name goes, uses the combination of thermodynamics and acoustic as the working principle. it involves transfer of heat and also acoustical wave’s movement. besides, thermoacoustic also involves density and pressure variations in the production of energy process. nicholas rott is the pioneer in deriving the correct equations for motion, pressure and time-averaged in energy transport in a channel with small, sinusoidal oscillations and with a temperature gradient (swift, 2001). this eases the trouble of acoustic study on oscillations encountered in engines and refrigerators. stirling engine century ago compromises a lot moving parts. in 1969, william beale realized that under proper circumstances, forces on connecting rods will be small resulting in free-piston. *corresponding author e-mail: fatimah@utem.edu.my journal of mechanical engineering and technology 78 issn: 2180-1053 vol. 8 no.1 january – june 2016 peter ceperly, after realizing that time phasing between pressure and velocity in the thermodynamic elements of stirling engine is the same as in acoustic travelling wave, suggested on removing every moving part except for the working gas itself. not long after that, the los alamos group started their research and development of standingwave thermoacoustic engines and refrigerators with different time phasing from ceperly’s idea and stirling engine (swift, 2001). a thermoacoustic device can work in two ways. one is to produce work using heat and is mainly called as prime mover. another way is to create heat by using work and is commonly called as heat pump. thermoacoustic devices have gained more attention mainly due to its independence on moving parts and hence, more efficient. furthermore it can be powered easily with sources such as solar or waste heat and the working medium is of environmental-friendly type. last but not least, the cost of fabrication of such device is low and yet it is reliable (piccolo, 2011). these are just few of the reasons why thermoacoustic devices are favorable nowadays. the oscillatory flow conditions modelled in the current study will mimic the conditions found in thermoacoustic systems. the main working medium behind thermoacoustics is a type of flow called oscillatory flow. this flow is formed from sound waves with amplitudes high enough to transfer heat from one place to another. on the other hand, sufficient high temperature gradient can also create sound waves of reasonably high amplitudes. this principle plays an important role because the oscillatory flow will move back and forth expanding and contracting in order to do work. 2.0 literature review over the years, numerous researches had been carried out in the field of thermoacoustics. the type of fluid flow involved in thermoacoustic device is oscillatory flow where the fluid flow travels back and forth. oscillatory flow can enhance heat transfer of a system. this has been proved experimentally and theoretically by volk (2006). this feature is preferable in application such as a prime mover because heat is required to be dispersed as fast as possible in order to sustain the effectiveness of the prime mover. heat transfer at the parallel-plates heat exchanger of a thermoacoustic system is very important as it will affect performance and efficiency of the device. a simple explanation about devices using thermoacoustics principle may be explained with the aid of figure 1. generally the thermoacoustic effects occur within an area inside the device where structures shown in figure 1 are placed. the structures, in general, contain a pile of solid structure known as ‘stack’. this ‘stack’ is sandwiched between a pair of heat exchangers. thermoacoustic effects occur when the oscillatory flow inside the device interacts with the ‘stack’. depending on the source of energy, the interaction between the flowing fluid and the solid surface of the stack may produce either cooling effect or power (swift, 2001). oscillatory flow across plates with different shape of edges issn: 2180-1053 vol. 8 no.1 january – june 2016 79 figure 1. illustration on position of heat exchangers and stack the heat exchangers at the ends of ‘stack’ are responsible to effectively remove heat from the system and provide cooling capacity to the refrigerated space that is attached to the system. on the other hand, if the heat exchangers provide a high enough temperature gradient to the fluid, such that allows the fluid particle to excite, power will be produced. the energy produced may then be harnessed for other useful application (swift, 2001). the challenge in commercializing the thermoacoustic technologies lies, among others, on understanding the behavior of the flow and heat transfer phenomena inside the system. current analytical solution used in designing the thermoacoustic system is based on a one-dimensional linear model. however, in practical system, the flow may consist of irregularities such as natural convection (mohd saat et al., 2012), streaming and vorticity (mao et al., 2008). it is pertinent that these effects are investigated so that a proper understanding may be gained. this involves the fundamental knowledge of oscillatory flow. the study on heat transfer phenomenon in a heat exchanger across oscillatory flow is important. there are many types of heat exchanger and the parallelplate heat exchanger is one of them. as the geometry of the parallel-plates structure is changed, the flow properties near the plates will also change (irwan shah et al., 2011). this may somehow affect the interactions between the oscillatory flow and the solid surfaces. the changing of geometry and the channel dimensions may also create disturbances or other effects on the flow. the usual shape of the plates is rectangle with sharp edges but what changes may be observed if the shape of the plates is changed for example, to rectangle with blunt edges or to other shape? whenever there is flow passing a solid body, a flow pattern such as vortices will be generated. vortex is a mass of fluid or air moving in circular motion due to pressure changes. the center of vortex will commonly cause suction to the area around the vortex. von karman is a type of vortex shedding where detached pairs of vortex appear at the back of a bluff body alternately. alternating vortex shedding has been the main cause for the failure of numerous designs. in a study regarding the vortex shedding flow patterns, it is said that oscillatory flows past bluff bodies usually gives a more complicated flow pattern than the von karman type of shedding in steady flows. journal of mechanical engineering and technology 80 issn: 2180-1053 vol. 8 no.1 january – june 2016 this is due to the oscillation of the flow that will cause impingement of vertices on the body and the interaction with vertices that are generated by the flow when the direction of flow is reversed (shi et al., 2010). in another research, it has been shown that the flow pattern of oscillatory flow is more complex compared to those of steady flow. therefore, the knowledge on the flow pattern around the structure of parallel plates is essential so that researchers can have clearer idea of the thermoacoustic effect at the structure such as stack or heat exchanger (mao & jaworski, 2010). however, if the geometry of parallel plates and channel dimensions are changed, it may cause changes in the flow behaviors such as the boundary layer, viscous and thermal penetration depth and also the occurrence of vortex shedding. boundary layer, a region of flow around the surface of parallel plates that may encounter viscous force, may bring disturbance to the heat transfer process between the flow and the solid boundaries. difference in the gap dimension will cause the flow behaviors that are affected by thermal and viscous penetration depth to be disrupted. if the gap is too large, the thermal interaction between the flow and the solid boundaries may be too weak. in a study of geometrical optimization of thermoacoustic heat engines, it is shown by simulation that by decreasing the stack spacing, all the gas parcels are confined within the thermal boundary layer. this allows the gas parcels to interact with the stack and eventually increases the performance of the heat engine. however, if the stack spacing is decreased, the effect of viscous forces becomes greater causing rise in the viscous losses. unfortunately, this will lower the performance of the heat engine. therefore, the stack spacing needs to be maintained at an optimum level where the thermal effect is good whereas the viscous effect is not too strong (ibrahim et al., 2011). apart from that, the occurrence of vortex shedding at region around the parallel-plates has gained attention from researchers too. few studies have been carried out to determine the effect of geometry on the formation of vortices around the stack in thermoacoustic devices. the application of oscillatory flow as working medium does not lighten the burden of researchers. instead, the oscillatory flow along with the changes in cross section make the flow structures at the end of the stack even more complex (mao et al., 2008). a deeper understanding of the flow throughout the internal structures of the system is necessary in order to design a high performance thermoacoustic heat engine. as the flow moves to the end of a rectangular plate with sharp edges, flow separation can be observed (mao et al., 2008). when the flow is out of the channel, it may generate vortex-like wake which will complicate the flow pattern. generation of vortices at the entrance and exit of the channel will cause the flow energy to dissipate into heat and therefore reducing the performance of the thermoacoustic device. in this study, a flow pass a pile of parallel-plates acting as heat exchangers will be modelled. the fluid will flow back and forth in an oscillatory manner. there will be formation of various flow patterns when a fluid flows past the parallel-plates. besides, there is also heat transfer between the two entities since the parallel-plates will be acting as heat exchangers. the temperature difference around the plates region will cause the occurrence of heat transfer between the fluid and the plates due to the temperature gradient. the focus of this investigation is to study how the changes in the plate geometry (shape of edges of the parallel-plate) may disturb or affect the flow and heat transfer across a parallel-plate heat exchanger. oscillatory flow across plates with different shape of edges issn: 2180-1053 vol. 8 no.1 january – june 2016 81 3.0 computational modelling the domain used in this study is a simplified model for parallel-plates heat exchangers. the dimensions of the domain are set in accordance to mohd saat and jaworski (2013) in order to facilitate the validation of the model. the length of the horizontal walls is set to be 600 mm while the height is set to be 132 mm. the total length of the parallel-plate structure is 70 mm. hot heat exchanger is represented by half of the length of the parallel-plate, that is 35 mm. the other half of the parallel-plate represents cold heat exchanger. the parallel-plates have thickness of 3.2 mm with 6 mm wide gap between the plates. the pile of parallel-plates is located at 265 mm from the inlet and 23 mm from the bottom wall. there are a total of 10 plates in parallel arrangement. the side of parallel-plates in red represents the hot heat exchanger while the side in blue represents cold heat exchanger. figure 2. illustration of the computational domain and its dimensions the model was solved using a pressure-based solver. a transient solver was selected to model the cyclic nature of the oscillatory flow. the effect of gravity was also turned on to correctly model natural convection due to the presence of the hot plate. taking upwards direction as positive, acceleration in y-direction was set to be -9.81 m/s 2 which is the gravity acceleration. heat transfer was modelled using two-dimensional energy equation with the consideration of viscous dissipation. this is to ensure that heat transfer process within the plates and nearby areas are modelled correctly. for the pressure-velocity coupling, simple algorithm was chosen. nitrogen gas was selected as the working medium of the domain. the gas was modelled as compressible flow. the thermal conductivity of the gas was set as temperature-dependent following the equation proposed by abramenko et al., (1992). the boundary conditions of the computational domain were calculated using lossless equation as shown in equations (1) and (2) (mohd saat & jaworski, 2013). p1 = pa cos(kax1) cos(2πft) (1) m1 = (pa/c) sin(kax2) cos(2πft+ө) (2) the term pa is pressure at pressure antinode, ka is wave number and c is the speed of sound with a value of 353 m/s. the value of x1 is 4.23 m and x2 is 4.83 m. journal of mechanical engineering and technology 82 issn: 2180-1053 vol. 8 no.1 january – june 2016 the flow amplitude modelled in this study is for the drive ratio of 0.3%. the drive ratio is defined as the ratio between pressure at antinode and the mean pressure. the mean pressure was set as atmospheric pressure. the wave number, ka, was calculated as 2/, where =c/f is the wavelength. the frequency, f, of the flow was set to be 13.1 hz. the phase, , between outlet mass flux, m1, and the inlet pressure was set to 90 so that the flow is in standing wave mode. the simulation was run for a minimum of 40 cycles so that the simulation reached a steady oscillatory flow condition. figure 3 shows the sketch of the parallel-plate heat exchanger used in this study. the original model has a rectangular shape of edge. case 1 has a round shape of edge, while cases 2 and 3 have a rectangular shape of edges. however the rectangular shape of edges in cases 2 and 3 are different in sizes. case 3 has the longest horizontal distance because the distance from the original edge to the tip of triangle is two times of that for case 1 and case 2. note that the total length of all the four cases investigated are slightly different. however, the total area of hot and cold plates (for the calculation of heat) four all the four cases are the same. the mesh numbers and sizes for these three cases were adjusted so that suitable meshing can be acquired. the configurations for solver settings for these three cases are the same as the configurations used for the original model. figure 3. geometry of simulation models with different geometries 4.0 model verification the model was verified by comparing the axial velocity changes over time at point ‘m’ (please refer to figure 2 for the location of ‘m’) between the results from simulation and the results from theoretical calculation using linear thermoacoustic model as reported in swift (2001). the velocity magnitudes of this oscillatory flow were recorded for 20 phases within one flow cycle. figure 4 shows that the velocity increases from phases 1 to 5 and then decreases as it flows until phase 10. the fluid starts reversing after phase 10 with an increase of reverse velocity magnitude until phase 15. after that, the velocity of the reverse flow decreases and the flow will repeat with a new cycle after phase 20. as shown in figure 4, a good match was found between the results and the theoretical predictions. oscillatory flow across plates with different shape of edges issn: 2180-1053 vol. 8 no.1 january – june 2016 83 a closer look will reveal that the simulation results differ slightly from theory particularly between phases 4 to 10. the maximum percentage of error between the theory and results from simulation was determined to be 2.48% which is still acceptable. thus, this simulation model was considered acceptable and can be used for simulation of the other models with different plate geometries. figure 4. changes of x-velocity of point ‘m’ at different phases figure 5 shows the grid sensitivity test for this study where the simulation was run using another model with higher number of meshes. it can be seen that with higher number of meshes, deviation occurs between phases 4 to 12. the greatest deviation occurs at phase 9 with a percentage error of 8 percent. this indicates that with different number of meshes, the results obtained from this study may slightly differ. however, the model with the most appropriate number of meshes was already selected for this study so that the curve obtained will be similar to that predicted by theoretical calculation. figure 5. grid sensitivity test journal of mechanical engineering and technology 84 issn: 2180-1053 vol. 8 no.1 january – june 2016 5.0 results and discussions 5.1 comparison of vorticity contours near the edges figure 6 shows that the vortex pair formed at the end of plate for cases 2 and 3 are more symmetry compared to the other two cases. this indicates that flow is less disturbed at the end of plate when the edge is made in triangular shape. the existence of two layers of vorticity adjacent to the wall within the channel indicates that rotational flow occurs near the plate. this is a typical phenomenon for oscillatory flow where fluid flows back and forth in a cyclic manner. at phase 10, the flow is about to reverse. this explains the existence of the two layers of vorticity with different signs at locations near the wall. in overall, the strength of the second layer of vorticity away from the wall is stronger for cases 1, 2 and 3. the flow was able to enter the channel smoothly for the three different shape of edges for cases 1, 2 and 3. thus, the stream of vortices for all the models other than the original model flew smoothly into the channel at phase 10. however for the original model, the second layer of vorticity is weak. this indicates that the flow was slightly disturbed hence showing signs of vortices discontinuities when the flow was heading into the channel from the right side. figure 6. vorticity contours near the edges of all the models at phase 10 unfortunately, for 0.3 percent drive ratio the flow tend to be laminar. hence the differences in shape of vortices at the edges between the models cannot be seen clearly. if the models were to be simulated with turbulent flow, such as at higher drive ratio, the differences could be clearer and easier to be observed. oscillatory flow across plates with different shape of edges issn: 2180-1053 vol. 8 no.1 january – june 2016 85 5.2 comparison of average total surface heat flux between all cases the effect of edge shape on the heat transfer performance of the parallel-plate heat exchanger is examined by looking at the total surface heat flux gained at the heat exchanger’s plate. heat flux at the surface is defined as: dy dt kq  (3) the terms q, k, t and y are the heat flux, thermal conductivity of the gas medium, temperature and vertical distance from the surface of the plate, respectively. the total surface heat flux for all cases presented in figure 7 was calculated based on an areaweighted average of the heat which was also averaged over one flow cycle. figure 7. average total surface heat flux at hot heat exchangers for all cases it is clearly shown in figure 7 that case 2 has the highest average total surface heat flux over time at the hot heat exchangers surfaces compared to the other 3 cases. with reference to figure 3, case 2 was defined as the parallel-plate structure with a triangular shape of edge. case 2 recorded a value of 12.6274 w/m 2 while the lowest value of average total surface heat flux recorded is from case 1 with only 10.8627 w/m 2 . the original model and case 3 have almost the same value with original model being slightly higher than case 3. the heat energy transfer rate at hot heat exchangers surfaces for case 2 has shown huge increment after the alteration in the geometries of the edge of the parallel-plates. this is probably due to the change of flow structure within the layer which enhances the heat transfer at the plate. journal of mechanical engineering and technology 86 issn: 2180-1053 vol. 8 no.1 january – june 2016 5.3 comparison of total surface heat flux between all cases detail analysis could be done by looking at the local heat flux values at several locations from the entrance of the left end of the parallel-plate structures as illustrated in figure 8. as shown in figure 8, the location for points ‘a’, ‘b’ and ‘c’ are 270 mm, 280 mm and 290 mm respectively from the left end of the computational domain. figure 8. location of point ‘a’, ‘b’ and ‘c’ note that the illustration in figure 8 is represented using an enlarged view of just one pair of parallel-plate structures. the total number of plates in the real domain was as reported in figure 2. the locations of points ‘a’, ‘b’ and ‘c’ are the same for all the models and the points are located on the hot heat exchanger side. figure 9 shows the variation of local heat flux measured at location ‘a’ over twenty phases within one flow cycle. location ‘a’ is located near the left entrance and far from the joint between the hot and cold plates. case 2 which has a triangle-shaped edges with shorter edge length has the highest local surface heat flux at point ‘a’ for most of the phases. this could be related to the change of entry length due to the change of edge’s shape. mohammed and salman (2007) proposed in their experimental study that as the entrance section length becomes longer, heat transfer decreases. this is probably caused by the resistance exerted from the flow as the entry length increases. based on this argument, case 2 with the highest value of local surface heat flux seems to have the shortest entry length presumably due to the less flow disturbance at the edge. as mentioned earlier, flow across original model seems disrupted particularly at the edges because of the rectangular shape of the edges. the triangle shape of edge has smaller form drag compared to rectangle shape of edge. hence, the triangle shape of edge of case 2 provides a smoother path for the fluid to oscillate. as a result, a better heat transfer through point ‘a’ is possible. however, if the triangle edge is made longer (case 3) the heat flux drops. the results shown in figure 9 also shows that the heat flux values for round edge (case 1) are also consistently smaller than case 2 and original model. the reason for this is not clear but could be related to the effect of the viscous and thermal boundary layers which may change as the fluid flow through the edge with different shape. furthermore, the performance of heat transfer can also be greatly affected by vortices (shi et.al., 2010). a deeper study in this area may be needed to help understand this phenomena. oscillatory flow across plates with different shape of edges issn: 2180-1053 vol. 8 no.1 january – june 2016 87 figure 9. local surface heat flux at point ‘a’ for all the cases figure 10 shows the local heat flux at location ‘b’. location ‘b’ is located somewhere midway between the left entrance and the joint between the cold and hot plates. it is noteworthy that the temperature gradient at the joint between the cold and hot plates is very high. since point ‘b’ is located midway between the left end and the joint, the thermal and viscous layer at this point are expected to be less affected by the entrance effect and the effect of temperature gradient at the joint. this effect could be seen in the original model, case 2 and case 3 which have recorded only slight fluctuations of local surface heat flux throughout the phases. however, case 1 showed significant fluctuations of values for local surface heat flux at point ‘b’. case 1 which has roundshaped edges also has the highest value of local surface heat flux at almost all phases. this indicates that the viscous and thermal layer of case 1 are not as steady as the other three cases even at locations away from the entrance and the joint. figure 10. local surface heat flux at point ‘b’ for all the cases journal of mechanical engineering and technology 88 issn: 2180-1053 vol. 8 no.1 january – june 2016 figure 11. local surface heat flux at point ‘c’ for all the cases figure 11 shows the local heat flux at location ‘c’. location ‘c’ is located near the joint where temperature different between the hot and cold plates is high. according to figure 11, both case 1 and case 2 showed similar results. the curves for both cases are almost similar for all the phases. local surface heat flux at point ‘c’ for case 3 also showed minor fluctuations throughout the phases. this fluctuations are very small. the original model, however, showed huge fluctuation of values of local surface heat flux through the entire phases. the curve for original model showed uphill and downhill trend several times. this indicates that the original model felt the effect of the temperature gradient the most. the rectangular-shaped edge alters the condition of flow at this point so that the temperature gradient at the joint has the greatest influence on heat transfer at point ‘c’. the results presented in figures 10 to 11 indicate that the high value of total heat flux for case 2, as presented in figure 7 may be due to the influence of entrance length. this is based on the high value of heat flux for case 2 at location ‘a’ as presented in figure 9. other fluctuations of local heat flux value at points ‘b’ and ‘c’ seem to have minor effect on the total heat flux. 5.4 comparison of average totral surface heat flux between all cases velocity profile near the entrance at point ‘a’ (refer figure 8 for point ‘a’) is compared between all the cases. a vertical line was created from point ‘a’ to the bottom surface of the parallel-plates above point ‘a’. the line connects the two parallel-plates and axial velocity data can be extracted through the line. the data were extracted at phase 5 where the fluid flows forward at maximum velocity and phase 15 where the maximum velocity of the reverse flow is achieved. the size of gap between the parallel plates is 6 mm. therefore the value of 0.006 m in figure 12a and figure 12b is the distance from point ‘a’ to the bottom surface of the parallel-plates heat exchanger above it. oscillatory flow across plates with different shape of edges issn: 2180-1053 vol. 8 no.1 january – june 2016 89 (a) (b) figure 12. velocity profile for all the cases at location ‘a’ for (a) phase 5 and (b) phase 15 figure 12 shows the velocity profiles for all the cases at phases 5 and 15 when the flow is at the highest value of magnitude during the first half and second half of the cycles, respectively. the velocity profile at phases 5 and 15 for original case is symmetry. they have a form closest to the fully developed flow profile. all other cases presented asymmetry velocity profiles between the two stages of the flow cycle. at phase 5, the velocity boundary layers of the other 3 cases seem not yet reaching the fully developed profile. these behaviors may be related to the entrance effect. however when the flow reversed its direction at phase 15, the velocity profile for most of the cases are similar and the profiles are more fully-developed like. when the flow reversed its direction, it entered the channel from the cold heat exchanger sides. during this part of a flow cycle, the flow have already travelled for a certain distance before it reach point ‘a’. thus, the velocity boundary layer for all the cases have developed fully forming the fullydeveloped shape of velocity profile as seen in figure 12 (b). 6.0 conclusions the flow and heat transfer of the validated original model were analyzed and compared to the results of the other three cases with different shape of plate edges. at the low drive ratio investigated in this study (0.3% drive ratio), the vortex structures at the end of the plate are slightly different from one case to another as the shape of edge changes. the difference is not so big due to the laminar feature of the low drive ratio investigated. as drive ratio increases the vortex pattern is expected to be more complicated. future work should look into this matter closely. it is expected that the edge shape will give a more significant impact on the vortex pattern when fluid oscillates at higher drive ratio. the results also showed that the heat transfer performance may be different if the shape of edges are different. this study suggested that a triangle-shaped edges with shorter length provides the higher heat transfer between plates and the oscillating fluid. the results of the local heat transfer investigation and velocity profiles shown at point ‘a’ suggest that the increase of heat may be related to the entrance length which was altered due to the change of shape of edge. however, deeper investigations are needed so that a better understanding could be journal of mechanical engineering and technology 90 issn: 2180-1053 vol. 8 no.1 january – june 2016 gained about the viscous and thermal boundary layers influence on the flow and heat transfer of an oscillatory flow across the parallel-plate structures. acknowledgements the authors would like to thank universiti teknikal malaysia melaka (utem) and ministry of education (moe) malaysia for supporting this research activities. this research work is part of the works funded by moe under a grant frgs/1/2015/tk03/utem/03/3. references abramenko t. n., aleinikova v. i., golovicher l. e., & kuz’mina n. e. (1992). generalization of experimental data on thermal conductivity of nitrogen, oxygen, and air at atmospheric pressure. j. eng. thermophys, 63, 892-897. ibrahim, a. h., arafa, n. m., & khalil, e. e. (2011, january). geometrical optimization of thermoacoustic heat engines. mechanical power engineering. irwan shah, a., normah, m., jamaluddin, m., aminullah, a., & dairobi, g. (2011). stack geometry effects on flow pattern with particle image velocimetry (piv). jurnal mekanikal 33, 82–88/. mao, x., & jaworski, a. j. (2010). oscillatory flow at the end of parallel-plate stacks: phenomenological and similarity analysis. fluid dynamics research, 42(5), 055504. mao, x., yu, z., jaworski, a. j., & marx, d. (2008). piv studies of coherent structures generated at the end of a stack of parallel plates in a standing wave acoustic field. experiments in fluids, 45(5), 833–846. mohammed & salman (2007). the effects of different entrance sections length and heating on free and forced convective heat transfer inside a horizontal circular tube. international communications in heat and mass transfer, 34(6), 769-784. mohd saat, f. a. z., & jaworski, a. j. (2013, july). oscillatory flow and heat transfer within parallel-plate heat exchangers of thermoacoustic systems. international conference of mechanical engineers, london, united kingdom. mohd saat, f.a.z., jaworski a.j., mao x., & yu z. (2012, july). cfd modelling of flow and heat transfer within the parallel-plate heat exchanger in standing wave thermoacoustic system. in proceedings of the 19th international congress on sound and vibration, vilnius, lithuania. oscillatory flow across plates with different shape of edges issn: 2180-1053 vol. 8 no.1 january – june 2016 91 piccolo, a. (2011). numerical computation for parallel plate thermoacoustic heat exchangers in standing wave oscillatory flow. international journal of heat and mass transfer, 54 (21-22), 4518–4530. shi, l., yu, z., & jaworski, a.j. (2010). vortex shedding flow patterns and their transitions in oscillatory flows past parallel-plate thermoacoustic stacks. experimental thermal and fluid sciences, 34(7), 954-965. swift, g. (2001). thermoacoustics: a unifying perspective for some engines and refrigerators. (fifth draf). los alamos national laboratory. volk, j. (2006). enhancing heat transport through oscillatory flows. phd thesis, university of florida. preparation of papers in a two column model paper format _________________________________________ *corresponding author e-mail: esalimipour@qiet.ac.ir issn: 2180-1053 vol. 8 no.1 january – june 2016 41 stall flutter control of a wing section by leading edge modifications e. salimipour 1* , m. saei moghaddam 2 , sh. yazdani 3 1,3 department of mechanical engineering, quchan university of advanced technology, quchan, iran 2 department of chemical engineering, quchan university of advanced technology, quchan, iran abstract a solution procedure is described for determining the two-dimensional and twodegrees of freedom flutter characteristics for wings at large angles of attack. this procedure requires a simultaneous integration in time of the solid and fluid equations of motion. the fluid relations of motion are the unsteady, compressible navier-stokes equations, solved implicitly by second-order roe’s approximation scheme in a moving coordinate system. the solid equations of motion were integrated in time by use of fourth-order runge-kutta method. in this paper, the stall flutter of a rectangular wing with section of naca 0012 is studied. therefore, the aeroelastic responses for the system were calculated by applying mode shapes for vibrating wing. then the obtained responses resulted from several changes in leading edge shape of wing are compared. results showed that these different leading edge shapes cause the changes on oscillating parameters of the system. in these changes, applying a camber with 25 o angle had the best result in this study. keywords: stall flutter; navier-stokes equations; leading edge shape; aeroelastic responses; mode shapes. 1.0 introduction aeroelasticity is defined as the interaction of aerodynamics, elasticity and dynamics. classical theories of aeroelasticity assume that the aerodynamic and structural forces are linear. for many decades, the classical approach has been successful in providing approximate estimates of aircraft response to gusts, turbulence and external excitation. the flutter boundaries are often quite accurately predicted when compared to flight test results. on the other hand, these classical methods are unable to capture phenomena arising from structural and aerodynamic nonlinearities. aerodynamic nonlinearities are often encountered at transonic speeds or high angles of attack where flow separation occurs (ghadiri & razi, 2007). flutter phenomenon is a kind of dynamic instability that results from the interaction of inertia, mailto:esalimipour@qiet.ac.ir journal of mechanical engineering and technology 42 issn: 2180-1053 vol. 8 no.1 january – june 2016 elastic and aerodynamic forces and causes the vibration of the wing to diverge. at transonic and supersonic regime, the shock wave incidence affects the oscillating parameters of the system, because this event can alter generalized forces, which operate on the system and at high angles of attack the flutter encounters a phenomenon which is called dynamic stall. dynamic stall is a phenomenon caused by vortex shedding on the surface of oscillating airfoils at high angles of attack. this causes a huge decrease in lift and increase in drag force and pitching moment too. if, during part or all of the time of oscillation, the flow was separated, then the flutter phenomenon exhibits some different characteristics and is called stall flutter (fung, 2007). in past decades existing analytical tools for the prediction of stall flutter treated this problem as a fluid-structural interaction problem. since the unsteady aerodynamics associated with this phenomenon is complex, researchers have in the past relied on experimental data for airfoil static and dynamic characteristics. these data are usually synthesized from a series of analytical expressions for different parts of the dynamic stall loop and used in a flutter analysis (jiunn-chi, kaza & sankar, 1987). the fluid-structural equations can be integrated in time simultaneously. the aerodynamic loads determined from the integration of the unsteady, compressible navier-stokes equations, drive the structural dynamics equations. the results of structural deformations alter the aerodynamic loads. under certain conditions, the coupling between the aerodynamic loads and the structural motion by imposing small disturbances on the wing angle of attack cause the aeroelastic responses grow rapidly in a divergent oscillatory fashion, and finally, flutter occurs. there are many studies of stall flutter at high reynolds numbers starting from about 1950. maybe, the most detailed of the former investigations are those of halfman, johnson & haley (1951) and rainey (1958). in these studies, the wing was placed at a mean angle of attack and then forced to vibrate, either in pitch or heave. one of the first stall flutter analyses using numerical solution of navier-stokes equations carried out by jiunn-chi et al. (1987). they studied naca 0012 airfoil flutter by simultaneous integration in time of the solid and fluid equations of motion at high angles of attack. price & keleris (1955) investigated naca 0012 airfoil flutter by applying nonlinear effects of aerodynamic loads at stall angle of attack by using semi-experimental methods. more recent studies such as razak, andrianne & dimitriadis (2011) have performed piv 1 measurements for an elastically mounted wing undergoing stall flutter. the wing was free to move in pitch and heave, but they observed that in stall flutter the pitching mode was predominant. an analysis based on modified leishman-beddoes model at low mach number was carried out by song, qinghua, chenglin & xianping (2011). the main modifications for l-b model included a new dynamic stall criterion and revisions of normal force and pitching moment coefficient. bhat & govardhan (2013) experimentally studied the stall flutter boundaries of a naca 0012 airfoil at low reynolds numbers by measuring the forces and flow fields around the airfoil when it is forced to oscillate. these measurements indicated that for large mean angles of attack of the airfoil, there is positive energy transfer to the airfoil over a range of reduced frequencies, indicating that there is a possibility of airfoil excitation or stall flutter. sun, haghighat, liu & bai (2015) developed a nonlinear time-domain aeroservoelastic model to study stall flutter and design flutter suppression control systems. a review 1 particle image velocimetry stall flutter control of a wing section by leading edge modifications issn: 2180-1053 vol. 8 no.1 january – june 2016 43 research for nonlinear flutter wind tunnel test and numerical analysis of folding fins with freeplay nonlinearities was performed by ning, nan, xin & wei (2016). the scope of the present work is to find a way for delaying of stall flutter effects by limiting the vibration amplitudes of the wing. one of the ways is the eliminating of stall by moving forward the leading edge separation point. by modifying the leading edge shape, we can alter the position of flow separation on the airfoil surface. in this study, raynolds average navier-stokes (rans) equations with two-dof structural dynamics model of wing are solved simultaneously in a moving computational grid by writing a computer code. the rans equations are applied on a full unsteady turbulent compressible flow. for this purpose, the turbulent shear stresses were modeled by using an eddy viscosity concept. in this study, the spalart-allmaras one-equation model is used because of its adequate accuracy for flows with adverse pressure gradient (blazek, 2001). generally, the solution procedure is that the aerodynamic loads obtained from the solution of the rans equations were placed into the solid equations of motion in each time step. then, the aeroelastic responses and the new velocities of moving grid will be achieved. these values will be used in the fluid flow equations for new time step. since there is a need to carefully validate any aerodynamics model prior to its application to flutter calculations, a number of code validation studies are first presented, including some cases where deep dynamic stall and stall flutter occurs. following the code validation, the studies on leading edge modifications are presented. 2.0 matematical and numerical formulation 2.1 flow-field equations in this section, according to jiunn-chi et al. (1987), the numerical procedure used to compute the unsteady viscous flow is briefly described. in the (x,y,t) coordinate system, the two-dimensional, unsteady navier-stokes equations may be written as v v f gu f g t x y x y              (1) where u is conservative variables vector, f and g are convective flux vectors and fv and gv denote viscous term vectors, which may be written as     2 2 0 0 , , , , xx xy v v yx yy xx xy yx yy u v u p uvu u f g f g uv v pv u v u vu e p v e pe                                                                          (2) e is the total energy per unit volume and is defined as journal of mechanical engineering and technology 44 issn: 2180-1053 vol. 8 no.1 january – june 2016 2 2 1 2 p u v e            (3) the shear stress components are expressed as   4 2 3 3 xx t u v x y              (4)   4 2 3 3 yy t v u y x              (5)  xy yx t u v y x                (6) where μ and μt represent the dynamic viscosity and turbulent addy viscosity, respectively. all the calculations were performed in a transformed coordinate system (ξ,η,τ), which is linked to the physical plane (x,y,t) according to the following relationship:    , , ; , , ;x y t x y t t       (7) the jacobian of transformation j is given by 1 j x y y x       (8) and the metrics of transformation are given by    ; ; ; ; ;x x t tz z t t t ty x y x x y x y x y x y j j j j j j                         (9) ξt and ηt include the grid points velocities in the coordinate system. in the (ξ,η,τ) coordinate system at non-dimensional form, the two-dimensional, unsteady navier-stokes equations may be written as * ** * * re v v f gmu f g                       (10) where m∞ and re denote the free-stream mach number and reynolds number, respectively. the quantities f * , g * , fν * and gν * are given by stall flutter control of a wing section by leading edge modifications issn: 2180-1053 vol. 8 no.1 january – june 2016 45         * * * * * / ; / ; / ; / ; / x y t x y t v x v y v v x v y v u u j f f g u j g f g u j f f g j g f g j                      (11) above equations are discretized using an implicit, finite-volume scheme based on roe’s second-order approximate riemann solver (blazek, 2001). 2.2 dynamic equations of wing resisting elastic forces are developed that are proportional to the airfoil torsional and translational displacement. in figure 1, a sketch of the two-dof rectangular cantilever wing is shown. note that the pitching axis may offset from the center of mass of the wing, leading to a coupling between the pitching and heaving degrees of freedom. the governing equations of the two-dof system motion are (fung, 1945): 2 2 4 2 2 4h h h h m s c ei l t t t y               (12) 2 2 2 .2 2 2 e a h i s c gj m t t t y                  (13) where y is the coordinate axis along the wing, h is the wing vertical displacement, α the angular displacement (angle of attack), iα the wing section moment of inertia (per unit span) about the pitching axis, m and sα=mxα the mass and static moment per unit span, and ch and cα the structural damping for the pitching and plunging motion coefficients, respectively. the ei and gj are the torsional and bending stiffnesses, respectively. the sα depends on the offset between the wing pitching axis and the wing center of gravity. the lift force (l) and pitching moment (me.a) are calculated by following relationships:     dspl yyyxyx   (14)    dlxxm aeae .. (15) where ds is the surface element. p, τxy and τyy are calculated by solving the unsteady, compressible navier-stokes equation (10). figure 1. schematic diagram of the two-dof system journal of mechanical engineering and technology 46 issn: 2180-1053 vol. 8 no.1 january – june 2016 the partial differential equations (12) and (13) are solved based on galerkin’s method. for this purpose, the h and α can be decomposed to product of two independent variables as follows:   1( )h h t  (16)   1( )t    (17) more details of above method are found in (hodges & pierce, 2002). φ and ψ are the torsional and bending mode shapes for vibrating wing and can be written as (marzocca, librescu & silva, 2002); 1 1 1 1 1 1 1 1 1 sinh sin (cos( ) cosh( ) sinh( ) sin( ) cosh cos k                        (18) 2 2 sin( )k    (19) where k1 and k2 are constant coefficients and, β1=0.5969π, β2=0.5π and η=y/l. by replacing the equations (16) and (17) into the equations (12) and (13) the nondimensional forms of wing equations of motion can be written as below: 2 3 6 3 1 5 1 1 3 12 3 2 h h a c a l a h a x h a h b b                           (20) 2 24 7 . 4 1 5 1 1 4 14 4 2 e a a c a m a r a x h a r b b                       (21) where: 1 1 2 1 4 2 2 2 3 4 ; ; ; ; ; h ih a ei a gj m h t r b a ml a i l b mb                  (22) the symbol apostrophe used in equations (20) and (21) represents derivative with respect to τ and the coefficient a1 to a7 are constant values that obtained from vibrating mode shapes where are determined in appendix. in order to reach aeroelastic responses, first the lift force and pitching moment will be obtained at each time step by solving the navier-stokes equations, then by placing them into the equations (20) and (21) the new values of h1 and α1 will be calculated. consequently, the flow conditions will be determined for next time step. we can see that the variables fo equations (12) and (13) will be updated at each time step. thus, these equations are fully unsteady. figure 2 shows a flowchart of this solution procedure. in the present work, the equations (20) and (21) era integrated in time using a fourth-order runge-kutta method and the aeroelastic responses are calculated at the end of the wing where y=l (or η=1). stall flutter control of a wing section by leading edge modifications issn: 2180-1053 vol. 8 no.1 january – june 2016 47 figure 2. solution procedure flowchart 3.0 results and discussions 3.1 code validation study before the application of this computer code in flutter applications, the flow solver was extensively calibrated for a number of test cases. in this work, only a few of these studies are reported. a 575×81 c-type grid with excellent orthogonality was used in this study which is shown in figure 3. figure 3. view of c-type grid used in flow computations in figure 4, the surface pressure distribution for a naca 0012 airfoil at 11.0264 deg angle of attack is shown. the free-stream mach number and reynolds number were solve rans equations and obtain l and me.a put l and me.a in structural equations solve structural equations and obtain h, α, add α to wing angle of attack obtain computational grid velocities using go to new time step journal of mechanical engineering and technology 48 issn: 2180-1053 vol. 8 no.1 january – june 2016 0.301 and 3×10 6 , respectively. for comparison, experimental data obtained by charles, acquilla & william (1987) are also shown. the present results and the experiments are seen to be in good agreement everywhere. figure 4. comparisons of theory versus experiment for the surface distribution over a naca 0012 airfoil at 11.0264 deg angle of attack (m∞ = 0.3, re = 3×10 6 ) in order to illustrate the capability of the navier-stokes solver to obtain time-accurate results in highly separated flow, the lift, drag, and moment hysteresis loops are shown and compared with experiments (mcalister, carr & mccorosky, 1982) in figure 5 for a naca 0012 airfoil oscillating in pitch. the mean angle and amplitude of oscillation were 14.91 and 9.88 deg, respectively. the reduced frequency of oscillation, normalized with respect to the half-chord, was k=0.151. the free-stream mach number and reynolds number were 0.283 and 3.45×10 6 , respectively. it is seen that the theory correctly predicts the near linear increase in lift during the upstroke, the dynamic stall causing rapid variations in lift, drag, and moment alike, and the poststall recovery phase of the flow during the downstroke. the fact that the flow solver is able to capture much of the dynamic stall flow features increases the confidence in the capability of this code to handle stall flutter problems. it must be mentioned that use of wall function approach for turbulence modeling could not be accurately reliable in this study. because the law of wall doesn’t always hold for flow near solid boundaries, most notably for separated flows (wilcox, 1994). experiment (charles et al., 1987) present study stall flutter control of a wing section by leading edge modifications issn: 2180-1053 vol. 8 no.1 january – june 2016 49 figure 5. comparisons of theory and experiments for the unsteady airloads on a naca 0012 airfoil experiencing dynamic stall (m∞ = 0.283, re = 3.45×10 6 , k = 0.151). as a final test of the above solver's ability to handle aeroelastic responses manner, the stall flutter calculations by use of the navier-stokes/structural dynamics solver explained above were considered. in this test case, the starting point was the steady viscous flow over a naca 0012 airfoil at 0.3 mach number and 9×10 6 reynolds number at 15 o angle of attack. at this angle of attack, the airfoil is on the verge of stall. the airfoil was given a small amplitude perturbation in its angle of attack and the subsequent motion was obtained. the pitch and heave responses are plotted as a function of non-dimensional time and are compared with obtained results performed by jiunn-ch et al. (1987) as shown in figure 6. very good agreement is observed between the two solvers. it was found that the airfoil returned to steady state following a period of damped oscillations. the non-dimensional aeroelastic parameters are assigned as below: 0 0 0.2, 0.25, 100, 51.5, 0.5, 15 , 0.25 h x r h b              figure 6. time response of a two-degree of freedom solid-fluid system experiencing stall flutter (naca 0012 airfoil, m∞ = 0.3, re = 9×10 6 , initial angle of attack 15 o ) 3.2 flutter calculations we use the current unsteady solver in the mentioned fluid-structure interaction method for the two-dimensional naca 0012 wing section. this model simulates the bending and torsional motion of the wing. it consists of two degrees of freedom, heaving and pitching, for a naca 0012 wing section. figure 7 shows the variation of flutter speed jiunn-chi et al. (1987) present experiment (mcaliste et al., 1987) present study journal of mechanical engineering and technology 50 issn: 2180-1053 vol. 8 no.1 january – june 2016 index versus the freestream mach number (m∞) for inviscid flow. flutter speed index includes the parameters which are affected on flutter incidence and is defined as (kirshman & liu, 2006)     f f f u v b      (23) where (u∞)f and (ωα)f are the values leading to happening flutter. the non-dimensional aeroelastic parameters are assigned as 0 0 0.5; 0.25; 26.35; 0.5; 1 ; 0.1 h x r h b           it is seen that by increasing the freestream mach number, the flutter speed is reduced. also, at transonic speeds the curve slop is increased and clearly, the system stability will be lower than subsonic one. figure 7. variation of flutter speed index versus mach number 3.3 leading edge modifications in this section, we attempt to find a way to limit the stall flutter phenomenon. the most effective factor in this kind of flutter is dynamic stall incidence, which alters the aerodynamic loads. dynamic stall is a phenomenon caused by vortex shedding on the surface of oscillating wings at high angles of attack. this causes a huge decrease in lift and increase in drag force and pitching moment. therefore, by eliminating of dynamic stall, we can suppress the stall flutter. one of the most effective portions of the wings on flow separation is the leading edge. by modifying the leading edge shape, we can alter the position of flow separation on the wing surface. we applied several modifications on the leading edge shape of the naca 0012 wing section and compared the aeroelastic stall flutter control of a wing section by leading edge modifications issn: 2180-1053 vol. 8 no.1 january – june 2016 51 responses, lift and pitching moment coefficients with the standard case. the following values are selected for the non-dimensional aeroelastic parameters: 6 0 0 0.5; 0.25; 100; 51.5; 0.5 0.4; re 2.5 10 ; 15 ; 0.25 h x r m h                     in the first case, we increased the sharpness of the leading edge and then solved the aeroelastic equations. in figure 8, the results are compared to the standard leading edge shape. the vibration amplitude has been reduced in heave motion; although the pitch amplitude is rising in compare to the standard case responses. therefore, this change could not be suitable. figure 8. comparisons of sharped versus standard leading edge for aeroelastic responses the second case involves a comparison of blunted versus standard naca 0012 leading edge for the aeroelastic responses with same parameters as shown in figure 9. we can see that by increasing the leading edge thickness, the vibrations amplitudes have been descended. the reason is that the leading of the relocation of separation onset toward journal of mechanical engineering and technology 52 issn: 2180-1053 vol. 8 no.1 january – june 2016 the trailing edge as shown in figure 10. by moving forward the separation onset point, the vortex growth opportunity will be reduced. also, the separation in blunted leading edge occurs at the higher angle of attack than the standard one; consequently, the stall will exist in smaller regions. figure 9. comparisons of blunted versus standard leading edge for aeroelastic responses 0.07 c 0.14 c α=21.5 ᵒ α=24.5 ᵒ stall flutter control of a wing section by leading edge modifications issn: 2180-1053 vol. 8 no.1 january – june 2016 53 figure 10. comparisons of blunted versus standard leading edge for separation onset figure 11. comparisons of small cambered versus standard leading edge for aeroelastic responses for the next case, a small camber in leading edge is applied (carr, mcalister & mccorosky, 1977). in figure 11, the results with same parameters are compared with the standard leading edge shape. it is seen that the amplitudes are decreased. because, in this case, the stall depth has been reduced. at the last case, we applied a camber with 25 o angle on the leading edge where in figure 12, the results with same parameters are compared with the standard leading edge shape. by observing the results, the reduction of amplitudes is determined. figures 13 and 14 show the flow field for some positions of the wings in a period of vibrations in two cases of standard and 25 o cambered leading edge. it can be seen flow separation condition changes by cambering of leading edge angle of attack as positive and negative, changes in each period on these vibrations. since in two last cases downward cambering of leading edge tend to delay in flow separation on positive angles of attack and vice versa. since the initial angle of attack is 15 o , there is not any separation at the beginning unless angle of attack was negative, flow separation had occurred slightly; however, when it is positive again, separation disappeared. it can be observed that there is not any deep stall in the flow field until the end of vibrations. journal of mechanical engineering and technology 54 issn: 2180-1053 vol. 8 no.1 january – june 2016 figure 12. comparisons of 25 o cambered versus standard leading edge for aeroelastic responses. figure 13. flow field for standard naca 0012. stall flutter control of a wing section by leading edge modifications issn: 2180-1053 vol. 8 no.1 january – june 2016 55 figure 14. flow field for 25 o cambered naca 0012 leading edge conclusions the turbulent compressible flows were solved in the time domain. with existence of dynamic stall, the amplitude and frequency of responses has been affected. it was observed that by producing a camber on the leading edge the flutter onset had been delayed. because the stall depth was reduced at lift force and pitching moment. specially applying a camber with 25 o angle had the best result in this study. it should be noted that the changes of leading edge could affect in flutter decreasing greatly, but in some cases (for example: cambering over of leading edge) these changes could eliminate aerodynamic properties of wing. appendix a by replacing the equations (16) and (17) into the equations (12) and (13), based on galerkin's method, the space dependent terms will appear to integral relations and respect to the wing geometry, will have constant values where form coefficients a1 to a7. these values have been obtained as   2 2 1 2 1 12 0 22.94429 d a d k d              (a.1)   2 21 2 2 2 0 8 d a d k d               (a.2)    21 2 3 1 0 1.85598a d k    (a.3)    2 21 2 4 0 2 k a d    (a.4)     1 5 1 2 0 0.92348a d k k       (a.5)   1 6 1 0 1.06667a d k     (a.6)   1 7 2 0 0.63662a d k    (a.7) journal of mechanical engineering and technology 56 issn: 2180-1053 vol. 8 no.1 january – june 2016 nomenclature b semichord length rα radius of gyration about elastic axis c airfoil chord length xα static unbalance in the structural model ch damping coefficients in heave vf flutter speed index cα damping coefficients in pitch u, v cartesian velocity components cd drag coefficient p pressure cd drag coefficient t time cl lift coefficient ρ density cm moment coefficient τ non-dimensional time cp pressure coefficient e total energy per unit volume dof degree of freedom k reduced frequency based on semichord length, ωf b/ u∞ h airfoil vertical displacement φ, ψ torsional and bending mode shapes iα moment of inertia α angle of attack j jacobian of transformation η non-dimensional span length of wing l wing width ξ,η transformed coordinates l lift force ωf frequency of vibration m∞ freestream mach number ωh natural frequency in plunging me.a pitching moment around elastic axis ωα natural frequency in pitching u∞ freestream velocity μ airfoil-to-air-mass ratio e.a elastic axis τxx,yy,xy shear stresses c.g center of gravity references ghadiri, b., & razi, m. (2007). limit cycle oscillations of rectangular cantilever wings containing cubic nonlinearity in an incompressible flow. journal of fluids and structures, 23, 665-680. fung, y. c. (1945). an introduction to the theory of aeroelasticity. newyork: dover publication inc. jiunn-chi, w., kaza k. r. v. & sankar, l. n. (1987). technique for the prediction of airfoil flutter characteristics in separated flow. journal of aircraft, 26(2), 168177. halfman, r. l., johnson, h. c. & haley, s. m. (1951). evaluation of high angle-ofattack aerodynamic derivative data and stall-futter prediction techniques. naca tn 2533. rainey, a. g. (1958). measurement of aerodynamic forces for various mean angles of attack on an airfoil oscillating in pitch and on two finite span wings oscillating in bending with emphasis on damping in the stall. naca report 1305. stall flutter control of a wing section by leading edge modifications issn: 2180-1053 vol. 8 no.1 january – june 2016 57 price, s. j., keleris & j. p. (1955). non-linear dynamics of an airfoil forced to oscillate in dynamic stall. journal of sound and vibration, 193, 823-864. razak, n. a., andrianne, t. & dimitriadis, g. (2011). flutter and stall flutter of rectangular wing in a wind tunnel. aiaa journal, 49(10), 2258–2271. song, sh., qinghua, z., chenglin, zh. & xianping, n. (2011). airfoil aeroelastic flutter analysis based on modified leishman-beddoes model at low mach number. chinese journal of aeronautics, 24, 550-557. bhat, s. s., govardhan & r. n. (2013). stall flutter of naca 0012 airfoil at low reynolds numbers. journal of fluids and structures, 41, 166-174. sun, zh., haghighat, s., liu, h. & bai, j. (2015). time-domain modeling and control of a wing-section stall flutter. journal of sound and vibration, 340, 221–238. ning, y., nan, w., xin, zh. & wei, l. (2016). nonlinear flutter wind tunnel test and numerical analysis of folding fins with freeplay nonlinearities. chinese journal of aeronautics, 29(1), 144-159. blazek, j. (2001). computational fluid dynamics: principles and applications. amsterdam, london, new york: elsevier. hodges, d. h. & pierce, g. a. (2002). introduction to structural dynamics and aeroelasticity. cambridge university press. marzocca, p., librescu, l. & silva, w. a. (2002). aeroelastic response and flutter of swept aircraft wings. aiaa journal, 40, 801–812. charles, l. l., acquilla. s. h. & william, g. j. (1987). pressure distributions from high reynolds number transonic tests of an naca 0012 airfoil in the langley 0.3meter transonic cryogenic tunnel. nasa langley research center 15. mcalister, k., carr, l. w. & mccorosky, w. j. (1982). experimental study of dynamic stall on advanced airfoils. nasa tm, 84245-tr, 82-a-8. wilcox, d. c. (1994). turbulence modeling for cfd. california: dcw industries inc. kirshman, d. j. & liu, f. (2006). flutter prediction by an euler method on non-moving cartesian grids with gridless boundary conditions. computers & fluids, 35, 571–586. carr, l. w., mcalister, k. w. & mccorosky, l. j. (1977). analysis of the development of dynamic stall based on oscillating airfoils. nasa tn, d-8382. paper5-le-v9n1_pages-87-102 issn: 2180-1053 vol. 9 no.1 january – june 2017 87 numerical analysis of the effectiveness of brake insulator in decreasing the brake squeal noise m. a. abdullah1,2*, e. abdul rahim2, a. r. abu bakar3, m. z. akop1,2 1 centre for advanced research on energy, universiti teknikal malaysia melaka, hang tuah jaya,76100 durian tunggal, melaka, malaysia. 2 faculty of mechanical engineering, universiti teknikal malaysia melaka, hang tuah jaya,76100 durian tunggal, melaka, malaysia. 3 department of automotive engineering, faculty of mechanical engineering, universiti teknologi malaysia, 81310 utm skudai, johor, malaysia. abstract brake functions when two different materials are in contact to reduce a motion. due to surface irregularity, this contact at high revolution and contact force produces irritating noise called brake squeal. this paper presents the study of introducing brake insulator into the brake assembly in order to reduce the noise. different configurations of insulators are used in the finite element analysis (fea). the effectiveness of the brake insulator is analyzed using different type of materials. the finite element model of the brake is developed based on actual drum brake dimensions. fea is used for modal analysis to predict the modal frequencies and mode shapes. various friction coefficients, wheel speeds and brake forces are considered in the analysis. the squeal is shown by positive real part of the baseline graph. the accompanied slip rate in the baseline model of the insulator increases the brake squeal noise significantly. keywords: brake insulator; brake squeal; finite element analysis; modal analysis; brake insulator damping 1.0 introduction brake is one of the important components in transportation. in advancement of vehicle, the improvement of brake has focused on increasing brake power and reliability. however, purification of vehicle acoustic and comfort has greatly increased the benefaction of brake noise to this aesthetic and environmental concern. brake noise is an irritant to the users. most of them believed that brake noise is symptomatic of a defective brake and this problem will lead them to claim a warranty from the company that produced the vehicles although the brake functioning as well as it had been designed. brake noise or generally called brake squealing has no precise definition. thus, in brake part design and manufacture, noise generation and suppression have become conspicuous consideration. despite, as noted by previous researchers (abendroth & wernitz, 2000), (hamid et al., 2014), many makers of materials for brake pads spend up to 50 % of their engineering budgets on noise, vibration and harshness *corresponding author e-mail: mohdazman@utem.edu.my journal of mechanical engineering and technology 88 issn: 2180-1053 vol. 9 no.1 january – june 2017 issue. there are few terminologies for brake noise such as squeal, groan, chatter, judder, moan, hum and squeak. however, the terminology that often be used is squeal. the phenomenon of squeal is probably the most common, disturbing to users and environment, and its cost the manufacturer in term of warranty. there are no precise definitions for brake squeal, but it is frequently agreed that squeal occurred at frequency above 1000 hz (lazim et al., 2014). brake squeal is a phenomenon of dynamic instability that occurs at one or more of the natural frequencies of the brake system. the research on this field had started since 1950’s where the busses was banned from using the road because producing a squealing or noisy sound from the brake (crolla & lang, 1991). there are a few approaches used in predicting the probability of the squeal occurrence which are theoretical, experimental and finite element (fe) approaches. besides, there are several methods also proposed to suppress or reduce the squeal occurrence which are; structural modifications, active control and adding damper (kinkaid et al., 2003). from the three methods, adding damper is the most efficient method and it may be applied by changing the material with high damping material or by adding insulator to the pad or shoe, which depend on what type of brake that will be used. drum brake produces significant amount of noise compared to disc brake. motorcycle brake which is using the drum brake type is used for this study. brake is one of the most important things that need to be considered when producing a vehicle. the squealing sound that produced from the brake not only contributed to the noise pollution, but also make the users are not comfortably used the vehicle. they thought that the brake might be broken down and the vehicle are not safely be used which will lead them to claim a warranty from the company that produced the vehicle. frictional contact is one of the important sources of squeal noise in braking systems. a merging of two real modes in an unstable complex one as the friction coefficient increases describing the general scenario for squeal occurrence. this phenomena is called modal coalescence, which the example can be seen in review of (giannini, 2008). many researchers have done the experimental works on laboratory or industrial brake setups. for example, nakata et al. (2001) shows that squeal involved a mode of the pads coupled with a mode of the disc. even if this phenomena is likely to be known, numerical modal analyses remain insufficiently predictive to improve the design of brake systems (hervé et al., 2008). despite, in fem models contrary to lumped models, important phenomena such as gyroscopic effects, circulatory forces and damping behavior are generally counted (flint et al., 2010). all those phenomena will result in stabilize or destabilize the system because they affect the modal eigenvalue. among the methods, increasing the system damping has been shown to be very effective to control squeal noise (kappagantu, 2008). a common way to do that is by applying the shims on the pad back plates (liu et al., 2007). in the previous research, great efficiency inside a vehicle subjected to road tests (festjens et al., 2012). several brake systems components have a high damping behavior such as brake shims. because of their glue and rubber layers those components have viscoelastic behavior (triches et al., 2004). the use of damping material glued on the back-plate of the pads may reduce squeal propensity (fritz et al., numerical analysis of the effectiveness of brake insulator in decreasing the brake squeal noise issn: 2180-1053 vol. 9 no.1 january – june 2017 89 2007). while the general idea among the researchers is that the damping increase the stability of the system, junior et al., (2008) in his paper shows that, if damping is equally distributed on the two modes involved in the mode coupling phenomenon, the only effect is a shift of the curves towards the negative real parts. if damping is spread non-equally over the two modes, a shifting and a smoothing effect can be seen on coalescence curves. if the ratio in damping between the two modes is sufficient, an increase in damping tends to make the brake unstable for a lower value of the friction coefficient (kappagantu & denys, 2008). according to nakata et al., (2001) the insulator is a laminated plate attached to the pad back plate, and its application leads to an increase in the pad stiffness. however, the pad stiffness is very important to determine the stability characteristics of the brake system. therefore, the effect of the additional damping provided by the insulator must be greater that the effect of the stiffness modification. this condition, however, is a variable that depends on the brake system analyzed. insulators usually are a multi-layer construction of steel, visco-elastic and rubber materials that are positioned between the backing plate and piston fingers of the braking system. as the insulator vibrates along with the backing plate, the different layers tend to bend and unbend resulting in the shearing and extension/contraction of the viscoelastic material. this causes vibration energy to dissipate in the form of heat (flint et al., 2010). there are a few squeal mechanisms to describe the squeal generation. the most common mechanisms are friction characteristic, sprag-slip model and modal frictioninduced vibration (mode coupling). friction-induced vibration is the most popular mechanism and widely used among the researchers to predict the squeal generation. this mechanism assumed that the coefficient of friction varied when the normal force of the friction is varied due to geometric coupling, then under some condition a pair of natural frequency may merge then squeal will be occurs. as it’s mentioned before, the squeal is excited more at low than at high speeds and squeals occur only over limited ranges of pressures and are most common at temperatures below 100 oc. from the literature also, it is clear that there are three methods to investigate the squeal generation which are experimental methods, analytical methods or numerical methods. experimental methods reflect real situation of the system but it takes time and somehow expensive. analytical methods does not representing an actual system and is limited to one particular frequency of squeal. numerical methods (finite element methods) provide a reasonable results and similar to the experimental results with short time. the complex eigenvalue analysis noted, most common method to evaluate system stability. that attributed to, the complex eigenvalue provide fast and reliable results for the brake system as compared with the other methods (transient solution, routh criterion). the information about the stability of the system facilitates solve the problem with a suitable solution. in this paper, in order to reduce the brake noise, several type of brake insulators are introduced in the assembly of drum brake with different thickness. using fea software abaqus, the effectiveness of these insulators in reducing the noise is numerically analyzed. journal of mechanical engineering and technology 90 issn: 2180-1053 vol. 9 no.1 january – june 2017 2.0 methodology the squeal that generate attributed to the system instability. in the previous section, several methods for analyzing the stability of the system have been introduced. however, in this section will focus more on the complex eigenvalue analysis method. complex eigenvalue analysis can determine which system vibration modes are unstable. addition of damping material in the brake system will be used to stabilize unstable mode (triches et al., 2004). the fe software used to predict the squeal generations of a drum brake system by using complex eigenvalue analysis is abaqus. a 3d solid model is generated based on an actual rear drum brake assembly. previous design and analysis using commercial software are practiced (abdullah et al., 2014), (abdullah et al., 2013a), (abdullah et al., 2013b), (abdullah et al., 2012). the drum brake assembly is taken from yamaha lc. the brake assembly consists of a drum, brake shoes, back plate, beam, springs and braking lever. however in this project, only related parts are modelled for stability analysis, i.e., the drum and the shoes. the drum is the most important part in the brake system. the real part has two bearings and bays in between, while it’s simplified in the fe model and built as one part. shoe in the real system consist of two parts the shoe body and the friction material and they attached together by adhesive. in the fe model it’s also consisted of two parts attached together. there are two springs in the brake system they will not modelled, just their effect will add between the two shoes. the stiffness used is 10 kn/m for one side, and 20 kn/m for another side following the design in (hamid et al., 2014), (lazim et al., 2014). table 1 shows the fe model parts with their element specifications and mechanical properties. table 1. fe model parts with their specifications component drum shoe & lining type of element tetrahedral c3d4 tetrahedral c3d4 no. of elements 93499 44106 no. of nodes 14796 2208 density (ρ) kg/m3 3400 2700 , 2100 young’s modulus (e) gpa 70 71 , 8 poisson’s ratio (v) 0.3 0.3 , 0.3 numerical analysis of the effectiveness of brake insulator in decreasing the brake squeal noise issn: 2180-1053 vol. 9 no.1 january – june 2017 91 the fe modal analysis can be done by run one step in abaqus with free-free boundary condition setting the interested frequency range (1 to 10 khz). there are several factors determine the natural frequency and the mode shape of the fe model, changing these factors may help to determine the squeal occurrences in the fe model result. first is the geometry, since the geometry was built based on the real component it will be not that suitable to change the geometry as a way to reduce the error. next is the material properties that have been assigned to the fe model developed. there are three parameters we are dealing with, which are the modulus of elasticity, density and poisson’s ratio. tuning these three parameters is the method used in this project to validate the model, this method called model updating. in order to determine the stability of the drum brake system complex eigenvalue available in abaqus software will be used. the positive real parts of the complex eigenvalues demonstrate the degree of instability of the drum brake system which means probability of squeal occurrence. the stability of the drum brake system will determine without insulator first to check the unstable frequencies, and then an insulator will attach to the shoe to evaluate the effectiveness of the insulator on the stability of the brake system. in order to observe the complex eigenvalue analysis using abaqus, four main steps are recommended (festjens et al., 2012). they are given as follows: i. nonlinear static analysis for applying brake-line pressure. ii. nonlinear static analysis to impose rotational speed on the drum. iii. normal mode analysis to extract natural frequency of undamped system. iv. complex eigenvalue analysis that incorporates the effect of friction coupling. as it is mentioned in the literature the complex eigenvalue problem can be given in the following form: (1) where: m, c and k are mass, damping and stiffness matrices, respectively. for underdamped systems the eigenvalues always occur in complex conjugate pairs. for a particular mode the eigenvalue pair is: (2) where σi is the real part and wi is the imaginary part for the ith mode. the motion for each mode can be described in terms of the complex conjugate eigenvalue and eigenvector: (3) thus, σi and wi are the damping coefficient and damped natural frequency describing damped sinusoidal motion. if the damping coefficient is negative, decaying oscillations typical of a stable system result. however a positive damping coefficient causes the amplitude of oscillations to increase with time. therefore the system is not stable when the damping coefficient is positive. by examining the real part of the system eigenvalues the modes that are unstable and likely to produce squeal are revealed (du et al., 2015b). journal of mechanical engineering and technology 92 issn: 2180-1053 vol. 9 no.1 january – june 2017 2.1 insulator configuration in reducing the squeal occurrences, insulator will be used in stability analysis. the insulator which consists of different material layer will be attached to the inner side of the shoe. in this project, four pair of sample insulators with different material will be tested. table 2 shows the proposed configurations of the insulators with their description, where the yellow color refers to the polymer material and the red color refers to the iron material. table 2. different insulators configurations sample no. description configuration sample 1 1 frame rubber (yellow), 0.5 mm thickness & 1 steel plate (red), 0.5 mm thickness. sample 2 1 frame abs (yellow), 0.5 mm thickness & 1 steel plate (red), 0.5 mm thickness. numerical analysis of the effectiveness of brake insulator in decreasing the brake squeal noise issn: 2180-1053 vol. 9 no.1 january – june 2017 93 sample 3 1 rubber plate (yellow), 0.5 mm thickness & 1 aluminum plate (red), 0.5 mm thickness. sample 4 1 frame abs (yellow), 0.5 mm thickness & 1 aluminum plate (red), 0.5 mm thickness sample 5 1 rubber plate (yellow), 0.5 mm thickness & 1 composite plate (red), 0.5 mm thickness journal of mechanical engineering and technology 94 issn: 2180-1053 vol. 9 no.1 january – june 2017 sample 6 1 frame abs (yellow), 0.5 mm thickness & 1 composite plate (red), 0.5 mm thickness complex eigenvalue analysis can be performed to get the unstable frequencies of the system without insulator then test the system with the insulator. the stability analysis will perform with different coefficients of friction for the baseline. the positive real parts of the complex eigenvalues indicate the degree of instability of the drum brake system where it indicates the probability of squeal occurrence. brake is one of the most important things that need to be considered when producing a vehicle. the squealing sound that produced from the brake not only contributed to the noise pollution, but also make the users are not comfortably used the vehicle. they thought that the brake might be broken down and the vehicle are not safely be used which will lead them to claim a warranty from the company that produced the vehicle. besides, the development to suppress and eliminate the squeal produced in the brake has been the target of many researchers for recent years. however, there is no fully solution for this problem and the squeal only can be suppressed but cannot be eliminated. theoretical and finite element (fe) approaches are the methods that will be used to determine the squeal noise. 2.2 finite element (fe) model of a baseline drum brake unit. a 3d (cad) model of a drum brake system shall be developed that based on a real brake system. this cad model will be used in the finite element (fe) software called abaqus to create a baseline brake fe model. the component and assembled fe model will be analyzed using modal analysis to predict its dynamic properties such as modal frequencies and modes shapes. 2.3 stability analysis of the baseline fe model. stability analysis shall be performed on the fe model in order to predict squeal noise using complex eigenvalue analysis within 1 to 10 khz frequency range. various friction coefficients, wheel speeds and cable forces should be considered in the stability analysis. the positive real parts of the eigenvalue indicate degrees of squeal noise in the brake unit. both squeal frequencies and modes of the brake shoes will provide useful information to establish fe model of the insulator and to determine appropriate amount of damping for the insulator to prevent squeal occurrences. numerical analysis of the effectiveness of brake insulator in decreasing the brake squeal noise issn: 2180-1053 vol. 9 no.1 january – june 2017 95 2.4 finite element (fe) model of brake insulators the insulator fe model shall be developed based on different design configuration such shapes, number of layers and thickness of layer. damping properties of the insulator can be estimated using rayleigh damping, structural damping and loss factor. 3.0 results and discussions the result displayed the squeal occurrences after a few parameters was set up such as coefficient of friction, μ, the sliding velocity between the drum and the brake shoe, ν and the frequencies of interest (1 khz to 10 khz). the result will also lead in determining which coefficient of friction that influencing more squeal to occurred. from the result collected, the model which has the highest number of squeal occurred will be suppressed by using brake insulator. 3.1 stability analysis result figure 1 shows the predicted result of the squeal occurrence of the fe model of the drum brake system. the result of the project will be compared between the baseline models which have a static coefficient of friction without slip rate and another one is baseline models with the presence of slip rate. the complex eigenvalue analysis performed by setting a set of boundary condition, where the frequency of interest is between 1 to 10 khz, the rotational speed of drum is set to be 40 rad/s, the displacement of the cam for brake applying is set to be 4mm and the coefficient of friction, μ is varied from 0.3 to 0.5. based from figure 1, is clearly can be seen that baseline model without slip rate that contains coefficient of friction, μ=0.30 have no squeal generation. however, the squeal is detected when the coefficient of friction, μ is 0.4 and 0.5. there is one squeal detected for μ=0.4 at frequency 5677.7 hz and μ=0.5 at frequency 5659.9 hz. previously in literature review, it is state that this phenomenon occurred is due to the magnitude of stiffness matrix with increasing friction coefficient. the addition of the friction coupling forces at the friction interface result in the stiffness matrix for the system containing unsymmetrical off-diagonal coupling term. from this stability point of view, this coupling is considered as the reason of brake squeal occurred (hervé et al., 2015). baseline model with slip rate produced more squeal occurrence compared to baseline model without slip rate. there are 7 unstable mode detected for μ=0.3 and μ=0.4, and 6 unstable mode detected for μ=0.5. previously, it is state that this phenomenon occurred due to the change of the friction force direction. when the velocity of vibration changed the direction alternately, the friction force will also change the direction accordingly if the magnitude of the sliding velocity is less than the magnitude of the vibration velocity. in this situation, a repeating sequence of friction force in opposite direction is produced to maintain the vibration in undamped condition. sliding velocity with friction coefficient decreased critically when the vibration velocity decrease. the system will be stable when the vibration velocity decrease with the increasing of friction coefficient since higher friction force provides higher vibration resistance to the system. figure 2 shows relationship between frequency and the squeal generation occurred for baseline model which contains coefficient of friction, μ and slip rate. journal of mechanical engineering and technology 96 issn: 2180-1053 vol. 9 no.1 january – june 2017 figure 1. comparison baseline models with and without slip rate. figure 2. relationship between frequency and brake squeal generation. based from the above figure, is clearly can be seen that baseline model which contains slip rate and friction coefficient of 0.3 and 0.4 produced more positive real part compared to the other model. thus, baseline model which consists of slip rate and friction coefficient of 0.3 or 0.4 will be models that need to be suppress. in this case, the model chose is baseline model with friction coefficient of 0.4. the squealing generated based on the friction coefficient, μ and the existence of vibration velocity has been numerical analysis of the effectiveness of brake insulator in decreasing the brake squeal noise issn: 2180-1053 vol. 9 no.1 january – june 2017 97 discussed. from the discussion, the baseline model for friction coefficient of 0.4 has been chose to be suppressed. however, another tested is conducted by using the same model to determine the brake squeal occurrence causes by the difference operating condition; which is different value of velocity. the friction coefficient, μ of 0.4 produced the most unstable mode. by using the same model, the velocity of the system is changed. this tested is conducted to observe the effectiveness of velocity in producing brake squeal. the rotational speed used are 30 rad/s, 40 rad/s and 100 rad/s. figure 3 show the result of the test. the result shows that the baseline model with rotational speed of 30 rad/s, 40 rad/s, and 100 rad/s have same number of unstable mode. however, the baseline model with rotational speed of 30 rad/s recorded the highest brake squeal generated at frequency of 4741.8 hz. thus, the baseline model with μ=0.4, and rotational speed of 30 rad/s will be choose as the model that need to be suppressed. 3.2 brake insulator result the system can be stabilized by increasing the damping (du et al., 2015a). there is a few common methods used to increase the damping of the system; first by choosing highly damped material for the drum brake system or the second, adding a damping material to the drum brake system. four insulator configurations have been tested against squeal generation, the test conducted under friction coefficient of 0.4 and the speed of 30 rad/s. the purpose is to obtain the effectiveness of these insulators to suppress the squeal. however, in this case, the test will be focuses more on applying different types of material on brake insulator, but not the damping properties. the materials that have been tested are rubber, acrylonitrile butadiene styrene (abs), steel, aluminum and composite. figure 4 show the stability analysis of drum brake system for insulator with different types of iron material and abs. based from figure 4, it shows that one squeal occurred for all three pairs of material at the frequency range between 6 khz to 7 khz. previously on the data shows on figure 3, there are at least six squeal occurred at the frequency range between 2 khz to 10 khz. however, all three materials in figure 4 succeeded in suppressing the squeal occurred and only one squeal is detected. journal of mechanical engineering and technology 98 issn: 2180-1053 vol. 9 no.1 january – june 2017 figure 3. relationship of frequency and brake squeal occurrence for different speed. figure 4. stability analysis of drum brake system for insulator with different types of iron material and abs. based from figure 5, it shows that a few squeal occurred for all three pairs of material. steel and rubber recorded the highest squeal occurred compared to the others pair. however, all three materials in figure 5 succeeded in suppressing the squeal occurred and only one squeal is detected for pairs of rubber with composite and aluminum. numerical analysis of the effectiveness of brake insulator in decreasing the brake squeal noise issn: 2180-1053 vol. 9 no.1 january – june 2017 99 figure 5. stability analysis of drum brake system for insulator with different types of iron material and rubber. figure 4 and figure 5, both show that the material used successfully reduce the squeal generation. however, compared to figure 4 and figure 5 displayed more precise result. this can be prove from the data collected from abaqus software where the pairs of iron material with rubber read more data compared to the pairs of iron material with abs. table 3 shows the example of squeal occurred for rotational speed of 30 rad/s and coefficient of friction of 0.4. table 3. data collected for ν=30 rad/s and μ=0.4. mode number real part frequency (hz) 1 4.1933 2212.2 2 402.44 2266.4 3 959.27 3153.3 4 844.7 4102.8 5 1475.7 4741.8 6 1181.6 5568.5 7 94.606 7213.7 8 404.7 7622.3 journal of mechanical engineering and technology 100 issn: 2180-1053 vol. 9 no.1 january – june 2017 the brake insulators that have been used should reduce the squeal generated near the frequencies values as shown in table 4. example, let’s have a look at data collected for aluminum pair with abs and rubber at frequencies range of 2 khz to 3 khz. table 4. data collected for pair of aluminum and abs mode number real part frequency (hz) 1 0 2109.9 2 0 2221 3 0 2318.7 4 0 2620.9 5 0 2696.1 6 0 2814 4.0 conclusions the complex eigenvalue available in abaqus software utilized to predict the stability of the drum brake system of the motorcycle, the stability analysis performed with different coefficients of friction to study the influencing of the friction coefficient. the worst condition was associated with µ= 0.4 and v= 30 rad/s, where there were eight unstable frequencies detected. apply an insulator is the proposed method used to reduce the unstable frequencies. the insulator consists of multi layers of rubber and iron plate attached together, then they glued to the inner side of the shoes body. six insulator with different pairs of iron material and polymer have been tested, and all of them could reduce the unstable frequencies. however, the best pair of insulator is the pairs that have the least value of the positive real part in the complex eigenvalue analysis. a pairs of composite body and aluminum with rubber recorded the least value of positive real part. thus, a pairs of composite body and aluminum with rubber is the best material to suppress the squeal generated. acknowledgements the authors gratefully acknowledged the advanced vehicle technology (active) research group of centre for advanced research on energy (care), the financial support from universiti teknikal malaysia melaka and the ministry of education, malaysia. numerical analysis of the effectiveness of brake insulator in decreasing the brake squeal noise issn: 2180-1053 vol. 9 no.1 january – june 2017 101 references abendroth, h., & wernitz, b. (2000). the integrated test concept: dyno-vehicle, performance-noise (no. 2000-01-2774). sae technical paper. hamid, m. a., shamsudin, n. i., lazim, a. m., & bakar, a. a. (2014). effect of brake pad design on friction and wear with hard particle present. jurnal teknologi, 71(2), 135-138. lazim, a. r. m., bakar, a. r. a., & kchaou, m. (2014). the study of disc brake noise on three different types of friction materials. applied mechanics and materials, 663, 113. crolla, d. a., & lang, a. m. (1991). paper vii (i) brake noise and vibration-the state of the art. tribology series, 18, 165-174. kinkaid, n. m., o'reilly, o. m., & papadopoulos, p. (2003). automotive disc brake squeal. journal of sound and vibration, 267(1), 105-166. giannini, o., & massi, f. (2008). characterization of the high-frequency squeal on a laboratory brake setup. journal of sound and vibration, 310(1), 394-408. nakata, h., kobayashi, k., kajita, m., & chung, c. j. (2001). a new analysis approach for motorcycle brake squeal noise and its adaptation (no. 2001-01-1850). sae technical paper. hervé, b., sinou, j. j., mahé, h., & jezequel, l. (2008). analysis of squeal noise and mode coupling instabilities including damping and gyroscopic effects. european journal of mechanics-a/solids, 27(2), 141-160. flint, j., chinnasamy, a., & stikvoort, a. (2010). new method to identify dynamic normal stiffness and damping of shims for cae modeling (no. 2010-01-1711). sae technical paper. kappagantu, r. v. (2008). vibro-impact rotor dampers for brake squeal attenuationtowards an insulator free design to quell squeal. sae international journal of passenger cars-mechanical systems, 1(2008-01-2549), 1188-1193. liu, p., zheng, h., cai, c., wang, y. y., lu, c., ang, k. h., & liu, g. r. (2007). analysis of disc brake squeal using the complex eigenvalue method. applied acoustics, 68(6), 603-615. festjens, h., gaël, c., franck, r., jean-luc, d., & remy, l. (2012). effectiveness of multilayer viscoelastic insulators to prevent occurrences of brake squeal: a numerical study. applied acoustics, 73(11), 1121-1128. triches jr, m., gerges, s. n. y., & jordan, r. (2004). reduction of squeal noise from disc brake systems using constrained layer damping. journal of the brazilian society of mechanical sciences and engineering, 26(3), 340-348. journal of mechanical engineering and technology 102 issn: 2180-1053 vol. 9 no.1 january – june 2017 fritz, g., sinou, j. j., duffal, j. m., & jézéquel, l. (2007). effects of damping on brake squeal coalescence patterns–application on a finite element model. mechanics research communications, 34(2), 181-190. junior, m. t., gerges, s. n., & jordan, r. (2008). analysis of brake squeal noise using the finite element method: a parametric study. applied acoustics, 69(2), 147-162. kappagantu, r. v., & denys, e. (2008). geometric tuning of insulators for brake squeal attenuation (no. 2008-01-2546). sae technical paper. abdullah, m. a., tamaldin, n., ramli, f. r., sudin, m. n., & mu’in, m. a. m. (2014). design and development of low cost all terrain vehicle (atv). in applied mechanics and materials (vol. 663, pp. 517-521). trans tech publications. abdullah, m. a., mohamad, a. h., & ramli, f. r. (2013a). design, analysis and fabrication of fixed-base driving simulator frame. journal of engineering and technology (jet), 4(2), 85-102. abdullah, m. a., mansur, m. r., tamaldin, n., & thanaraj, k. (2013b). development of formula varsity race car chassis. in iop conference series: materials science and engineering (vol. 50, no. 1, p. 012001). iop publishing. abdullah, m. a., mansor, m. r., mohd tahir, m., kudus, a., ikhwan, s., hassan, m. z., & ngadiman, m. n. (2012). design, analysis and fabrication of chassis frame for utem formula varsitytm race car. preparation of papers in a two column model paper format issn: 2180-1053 vol. 10 no.1 january – june 2018 11 a comprehensive study on the efficiency of three different types of the variable metric method in determining the unknown rocket inner heat flux h. khoshkam 1*, m. beyrami 2, k. javaherdeh3 1,3iran faculty of engineering, department of mechanical engineering, university of guilin 2department of mechanical engineering, asadabad branch,islamic azad university, asadabad abstract in this paper, the unknown heat flux is estimated with davidon-fletcherpowell (dfp), broydon–fletcher–goldfarb–shanno (bfgs) and symmetric rank-one (sr1) version of variable metric method (vmm). the numerical techniques used in this study solved the inverse problems with various boundary and environmental conditions so efficiently. the results shows the sensitivity of measurement errors and different parameter including changes of slope and angle which can be functions of an unknown parameter. further, the speed of convergence is assessed and the convergence behavior is found. the accuracy of results show that this study is a powerful reference for comparing results obtained based on the three proposed techniques. the solution procedure introduced a general fast method which can be used for the inverse heat conduction problem in rocket nuzzle and same heat conduction, radiation and convection problems. keywords: bfgs; sr1; dfp; rocket nozzle; heat conduction; vmm method; inverse problem 1.0 introduction the nozzle plays an important role in the rocket motor, as the equipment of converting energy and producing thrust force for rocket, it converts the thermal energy of gases into the kinetic energy. during the motor operation process, the nozzle must endure the impact of jet flow with high temperature and pressure. the high temperature from the inner contour of the nozzle conducted into its shell leads to the increase and decrease of the erosion of nozzle throat insert and the material strength of the nozzle shell, respectively. additionally, it enlarges the throat radius which makes thrust descend, and also reduces the nozzle thrust efficiency. under these situations, the nozzle design obviously affects the motor performance. evaluation of rocket nozzle safety and its reliability can be assessed through numerical analysis of heat transfer and wall temperature. in order to meet the requirements in resisting the nozzle shell temperature and the jet flow, the throatinsert materials need to be inserted in the nozzle inner counter to form protection. there are three major material requirements, which are the throat-insert, thermal liner and insulator materials (tsung-chien chen & chiun-chien liu, 2008). * corresponding author e-mail: hamed.khoshkam@gmail.com journal of mechanical engineering and technology 12 issn: 2180-1053 vol. 10 no.1 january – june 2018 the nozzle throat which consists of an expensive super alloy, will directly affect the nozzle efficiency and is very effective for reduction of the running costs of a power generation plant. accordingly, it is very important for the life assessment of the nozzle to predict the operating conditions and to establish a basis for the criteria of repair. therefore, the heat conduction problem design in the nozzle and the method to choose moderate throat-insert materials are of the essential importance indeed (h.n. wang &j.h. wang, 2006). as a result, the estimation of heat flux in the rocket nozzle throat in high temperature environment is a crucial building block for assessing the safety and reliability of the nozzle. the inverse heat conduction problem is concerned with different parameters. among these parameters, one can refer to the thermal conductivity, the volumetric heat capacity, the initial condition, the boundary conditions, and the heat sources from knowledge of the temperature or heat flux measurements taken at the interior point of the solid or on its back surface (g. stolz, jr., 1960). solution methods of the inverse heat transfer problems (ihtp) for heat flux estimation can generally be classified into two categories: sequential methods and whole domain methods. both of these two categories involve minimization of a sum of squares of errors function defined on the basis of the difference between measured and calculated temperatures. in the sequential function (sfs) method (j. v. beck, b. blackwell and c.r. st. clair, 1985), which is the most noted algorithm of the first group, unknown heat fluxes are estimated in a consecutive manner. that is, the algorithm is based on marching in time and determining the unknown heat flux in current time step using future data by setting the derivative of the error function with respect to unknown heat fluxes equal to zero. in the whole domain methods, in which all of the known heat fluxes are estimated simultaneously, the minimization of the sum of squares of error function is achieved by the iterative minimization (optimization) techniques (j.g. bauzin & n. laraqi, 2004). one of these techniques is called the variable metric method (vmm). the vmm has superior characteristics as compared to the conjugate gradient method. the vmm is a powerful technique in the context of nonlinear optimization problems. this method has been utilized in the solution of inverse problems (m. prud'homme & s. jasmin, 2003). in this paper, a comprehensive discussion on dfp, bfgs and sr1 is presented for estimating the unknown boundary heat flux based on the boundary temperature measurements history that is measured at outside the body. furthermore, three examples are employed to demonstrate and discuss results of the three version of vmm in detail in the following sections (h.khoshkam & m.alizadeh, 2011). 2.0 the direct problem the specimen is a nozzle throat (see figure 1) this slab originally has a uniformly distributed temperature. a heat flux q (t) is applied to x = l at a specific time (t > 0), convective heat flux at x = 0 with constant heat transfer coefficients at the constant temperature. the following hypotheses have been taken into account: 1thermo-physical properties are assumed to be constant. 2heat transfer is one-dimensional. 3heat transfer coefficients are constant. 4radiation is not important. a comprehensive study on the efficiency of three different types of the variable metric method in determining the unknown rocket inner heat flux issn: 2180-1053 vol. 10 no.1 january – june 2018 13 under these conditions, the heat transfer process in the specimen can be described by the following system of equations: 𝑘 𝜕2𝑇 𝜕𝑥2 = 𝜌𝑐 𝜕𝑇 𝜕𝑡 0 ≤ 𝑥 ≤ 1 𝑎𝑛𝑑 𝑡 ≥ 0 (1) 𝑇(𝑥, 𝑡) = 𝑇0 0 ≤ 𝑥 ≤ 1 𝑎𝑛𝑑 𝑡 = 0 (2) −𝑘 𝜕𝑇(𝑥,𝑡) 𝜕𝑥 = ℎ(𝑇 − 𝑇∞) 𝑥 = 0 𝑎𝑛𝑑 𝑡 > 0 (3) 𝑘 𝜕𝑇(𝑥,𝑡) 𝜕𝑥 = 𝑞(𝑥, 𝑡) 𝑥 = 1 𝑎𝑛𝑑 𝑡 > 0 (4) here k, ρ and c are the thermal conductivity, density, heat and capacity, respectively. the governing equation is parabolic and the solution for the above heat conduction problem is solved by using finite volume method (s.v. patankar, 1980). 3.0 simulated inexact measurement the measured temperature data must contain measurement errors. a normally distributed uncorrelated error with zero mean and constant standard deviation are considered, in order to compare the results for situations involving random measurement errors can be expressed as: y=y_exact+ωσ (5) where y_exact and y in equation(5) are the solution of the direct problem with an exact boundary heat flux q(l,t) and the measured temperature, respectively. furthermore, ω is the random variable with normal distribution, zero mean and unitary standard deviation and for the 99% confidence bound we have (ch. h. h. & h. h. wu, 2006) -2.576 < ω < 2.576 figure 1. the one-dimensional rocket nuzzle throat geometry journal of mechanical engineering and technology 14 issn: 2180-1053 vol. 10 no.1 january – june 2018 3.1 inverse problem in this inverse problem the heat flux q (t) is unknown; and the unknown heat flux find by the vmm method stated below the boundary heat flux at x = l is regarded as being unknown, but everything else in equations (1)–(4) is known. in addition, temperature readings at x = 0 are considered available. the temperature reading taken by sensors at x = 0 be denoted by y (0, t), it is noted that the measured temperature y (0, t) contain measurement errors. with the above mentioned measured temperature data y (0, t), the method estimate the unknown boundary heat flux q (l, t) in such a way that the following functional is minimized: 𝑓 = ∑ ∑ [𝑌(𝑥𝑘, 𝑡𝑚) − 𝑇(𝑥𝑘, 𝑡𝑚, 𝑞�̃�)] 2𝑀 𝑚=1 𝑘 𝑘=1 (3) in the above definition, k is total number of sensors, y is the measured temperature at sensor location of x_k, and t is the calculated temperature utilizing the direct heat conduction model (equation 1) based on a given or assumed vector for q⃗. 4.1. variable metric method (vmm) the variable metric method (vmm), belong to the gradient optimization techniques. it can minimize the object function through an iterative procedure. the variable metric method is very stable and continues to progress towards the minimum even when dealing with highly distorted and eccentric functions. the steepest descent method, the conjugate gradient method, the newton method and the variable metric method (vmm), all belong to the gradient based class of unconstrained optimization techniques. however, vmm has superior characteristics in relation to the others (f. kowsary, a. behbahaninia, a. pourshaghaghy, 2006). the variable metric method is very stable and continues to progress towards the minimum even when dealing with highly distorted and eccentric functions. zhang et al. demonstrated mathematically that for a strictly convex quadratic objective function, the generated iterative sequence of vmm converges to the unique solution of the problem globally and super linearly (z.z. zhang, d.h. cao, j.p. zeng, 2004). the iterative procedure for the vmm can be summarized as follows: step 1: find the pulse sensitivity coefficients for each components of �⃗� by solving equations (6) – (9) in the entire time domain. step 2: at the start an initial guess for �⃗� and with a m× m positive definite symmetric matrix h1 (m is total number of unknowns). h1 is taken as the identity matrix i. set iteration number as i=1. step 3: compute the gradient of the objective function,∇f⃗⃗⃗⃗ i;(defineat below)at the base point 𝑞𝑖⃗⃗⃗ ⃗.; and define search direction as: 𝑆𝑖 = −𝐻∇f ⃗⃗⃗⃗ i step 4: normalize 𝑆𝑖 by its magnitude: 𝑆𝑖 = 𝑆𝑖/‖𝑆𝑖‖ step 5: compute the optimal step length λ𝑖 ∗in the direction 𝑆𝑖 and achieve to the next�⃗�𝑖 using: �⃗�𝑖+1 = �⃗�𝑖 + λ𝑖 ∗𝑆𝑖 a comprehensive study on the efficiency of three different types of the variable metric method in determining the unknown rocket inner heat flux issn: 2180-1053 vol. 10 no.1 january – june 2018 15 step 6: test the new �⃗�𝑖+1 for optimality. if �⃗�𝑖+1is optimal, terminate the iteration process. otherwise, go to step (7). step 7: update the ((h)) matrix step 8: set the new iteration number i=i+1, and go to step 3 ''m'' is total number of time steps that cover the entire time domain and the unknown heat flux q (t) is discretized into m time components. all the components are gathered inside a vector �⃗� as: �⃗� = [�⃗�(𝑡1), �⃗�(𝑡2), … , �⃗�(𝑡𝑀)] (4) for kth sensors, sensitivity coefficient of measured temperature is obtained with respect to each𝑞�̃�: x (𝑥𝑘, 𝑡𝑚, 𝑞�̃�) = 𝑑𝑇(𝑥𝑘,𝑡𝑚) 𝑑𝑞�̃� for �̃�=1, 2,..., m (5) 𝜕2𝑋 𝜕𝑥2 = 1 𝛼 𝜕𝑋 𝜕𝑥 (6) 𝜕𝑋 𝜕𝑥 = −ℎ𝑋 (7) 𝑘 𝜕𝑋 𝜕𝑥 = { 1 𝑖𝑓𝑡𝑚−1 ≤ 𝑡 < 𝑡𝑚 0 𝑜𝑡ℎ𝑒𝑟 𝑡 (8) x (𝑥𝑖, 𝑡 ≤ 𝑡𝑚−1) =0 for �̃�=1, 2, ..., m (9) the above equation should also be solved separately for every m time in order to compute x(𝑥𝑘, 𝑡𝑚, 𝑞�̃�). the gradient of objective function which must be minimized used in vmm has the form of: ∇⃗⃗⃗𝑓𝑛×1 = [ 𝜕𝑓 𝜕𝑞1 , 𝜕𝑓 𝜕𝑞2 , … , 𝜕𝑓 𝜕𝑞𝑛 ]𝑇 (10) and from the equation (10) we can write ∇f⃗⃗⃗⃗ ias: ∇⃗⃗⃗𝑓𝑛×1 = [ −2 ∑ ∑ ([𝑌(𝑥𝑘, 𝑡𝑚) − 𝑇(𝑥𝑘, 𝑡𝑚, 𝑞�̃�)x(𝑥𝑘, 𝑡𝑚, 𝑞1)] 𝑀 𝑚=1 𝑘 𝑘=1 −2 ∑ ∑ ([𝑌(𝑥𝑘, 𝑡𝑚) − 𝑇(𝑥𝑘, 𝑡𝑚, 𝑞�̃�)x(𝑥𝑘, 𝑡𝑚, 𝑞2)] 𝑀 𝑚=1 𝑘 𝑘=1 . . . −2 ∑ ∑ ([𝑌(𝑥𝑘, 𝑡𝑚) − 𝑇(𝑥𝑘, 𝑡𝑚, 𝑞�̃�)x(𝑥𝑘, 𝑡𝑚, 𝑞𝑀)] 𝑀 𝑚=1 𝑘 𝑘=1 ] (11) journal of mechanical engineering and technology 16 issn: 2180-1053 vol. 10 no.1 january – june 2018 in step (5) of vmm, the optimal step size (λ𝑖 ∗) in the direction of 𝑆𝑖 is a value of λ𝑖 ∗that minimizes 𝑓(�⃗�𝑖 + λ𝑖 ∗𝑆𝑖) with respect to λ𝑖 ∗ i.e., λ𝑖 ∗ is the root of the following equation: 𝑑𝑓(�⃗⃗�𝑖+λ𝑖 ∗𝑆𝑖) 𝑑λ𝑖 ∗ = 0 (12) for the stopping criteria (step 8); in this work‖f‖ ≤ 𝜀 is used as the stopping criteria,in the case of non-noisy data, 𝜀 is an arbitrary small number (in this work 𝜀 = 0.001). but in the case of noisy data, 𝜀 should be chosen based on the iterative regularization method in order to reduce sensitivity of the solution to the random noise errors. the main idea in the iterative regularization is to stop the iterative procedure close but not exactly at the optimum point. then, it will tend to regularize the solution and to damp out the destructive effects of random noises in data. 𝜀 = 𝑘 × 𝑀 × 𝐴2 (13) but in this work we use 𝜀 = 3 for noisy data because it has better results.different version of vmm has a different way to update the h (step7). symmetric rank-one (sr1) update by: 𝐻𝑖+1 = 𝐻𝑖 + (1 − 𝑄𝑖 𝑇𝐻𝑖𝑄𝑖 𝑄𝑖 𝑇(λ𝑖 ∗𝑆𝑖) ) −1 ∗ 1 𝑄𝑖 𝑇(λ𝑖 ∗𝑆𝑖) (λ𝑖 ∗𝑆𝑖 − 𝐻𝑖𝑄𝑖)(λ𝑖 ∗𝑆𝑖 − 𝐻𝑖𝑄𝑖) t (14) davidon-fletcher-powell (dfp) : 𝐻𝑖+1 = 𝐻𝑖 + λ𝑖 ∗ 𝑆𝑖𝑆𝑖 𝑇 𝑆𝑖 𝑇𝑄𝑖 − (𝐻𝑖𝑄𝑖)(𝐻𝑖𝑄𝑖) 𝑇 𝑄𝑖 𝑇𝐻𝑖𝑄𝑖 (15) broydon–fletcher–goldfarb–shanno (bfgs): 𝐻𝑖+1 = 𝐻𝑖 + λ𝑖 ∗ 𝑆𝑖𝑆𝑖 𝑇 𝑆𝑖 𝑇𝑄𝑖 − (𝐻𝑖𝑄𝑖)(𝐻𝑖𝑄𝑖) 𝑇 𝑄𝑖 𝑇𝐻𝑖𝑄𝑖 +…….. ………+ (𝑄𝑖 𝑇𝐻𝑖𝑄𝑖)*( 𝑆𝑖 𝑆𝑖 𝑇𝑄𝑖 − 𝐻𝑖𝑄𝑖 (𝑄𝑖 𝑇𝐻𝑖𝑄𝑖) ) ( 𝑆𝑖 𝑆𝑖 𝑇𝑄𝑖 − 𝐻𝑖𝑄𝑖 (𝑄𝑖 𝑇𝐻𝑖𝑄𝑖) ) 𝑇 (16) a comprehensive study on the efficiency of three different types of the variable metric method in determining the unknown rocket inner heat flux issn: 2180-1053 vol. 10 no.1 january – june 2018 17 5.0 results and discussions our simulations define from eqs. (1) – (4) that estimates the strength of the boundary heat flux. to illustrate the accuracy of the bfgs, sr1 and dfp in predicting boundary heat flux q (l, t) with the present inverse analysis, three different boundary heat flux functions over temporal domain; namely, a third degree polynomial function, triangular function and a step function are adopted to illustrate the numerical modeling. the exact temperature and the heat flux used in the following examples are selected so that these functions can satisfy eqs. (1) – (4). q (l,t)initial = 0 the following computational parameters are chosen for the numerical experiments: t0=25 °c, t∞ =25°c, l =0.01 m, k =138 w/ (m.k) , α=5.369×10-5 m2 /s, h=5000 w/ (m2 k) here α is the thermal diffusivity of the material. besides, the space and time increments used in numerical calculations are taken as ∆x =0.00001m (i.e.1000 grid points in space) and ∆t =1s (i.e.30 grid points for 𝑡𝑓= 30s). we now present below tree numerical test cases in determining q (l,t) by the inverse analysis using the different version of the vmm. numerical test-case 1: the unknown transient boundary heat flux q (l,t) is assumed applied at x = l in the following form: q(l, t) = −252.5t3 + 1.547 × 104t2 − 3.475 × 105t + 3.26 × 106 (17) the relative root mean square error (erms) for the estimated q (l, t) is defined as: 𝑒𝑅𝑀𝑆 = √ 1 𝑁 ∑ [𝑞(𝑙,𝑡)−�̅�(𝑙,𝑡)]2𝑁𝐼=1 √ 1 𝑁 ∑ [𝑞(𝑙,𝑡)]2𝑁𝐼=1 × 100% (18) where i and n represent the index of discrete time and total number of measurements, respectively, while �̅�(𝑙, 𝐼) denote the estimated values of heat flux. journal of mechanical engineering and technology 18 issn: 2180-1053 vol. 10 no.1 january – june 2018 table 1.root mean square error and convergence criteria for estimating heat flux to the third degree polynomial. figure 2. the estimation of the third degree polynomial heat flux with 𝜎 = 0 number of iterations rms e run time (s) ||∇𝑓|| at final dpf o 0 c.  26 9.70 11.07 3.91*10 -6 sr1 o 0 c.  22 9.71 10.81 3.68*10 -6 bfgs o 0 c.  23 9.69 10.56 3.21*10 -6 dpf o 3 c.  13 11.75 8.62 3.34*10 -4 sr1 o 3 c.  12 11.86 8.21 2.28*10 -4 bfgs o 3 c.  12 11.25 8.18 3.12*10 -4 dpf o 10 c.  17 39.81 9.35 6.34*10 -4 sr1 o 10 c.  14 45.88 8.32 7.59*10 -4 bfgs o 10 c.  14 31.02 8.94 4.35*10 -4 a comprehensive study on the efficiency of three different types of the variable metric method in determining the unknown rocket inner heat flux issn: 2180-1053 vol. 10 no.1 january – june 2018 19 figure 3. the estimation of the third degree polynomial heat flux with 𝜎 = 3 figure 4. the estimation of the third degree polynomial heat flux with 𝜎 = 10 journal of mechanical engineering and technology 20 issn: 2180-1053 vol. 10 no.1 january – june 2018 table 2. root mean square error and convergence criteria for estimating triangular heat flux figure 2 to figure 4 and table 2 show that in the third degree polynomial heat flux, if the measurement error for the temperatures, measured by sensor, is σ = 0 ◦c, each of the three version of vmm converge very rapidly and exact enough to the real heat flux and have a same root mean square error, but bfgs and sr1 are faster than dfp. in σ =3 dfp and bfgs have a good root mean square error, however, sr1 and bfgs converge faster than dfp. in large measurement errors bfgs have faster and more accuracy answers than two other version. numerical test-case 2: the unknown boundary heat flux q(l,t) is assumed applied at x= l in the following form:   6 6 6 6 0 0 t 1 10 t-10 q l, t 1 t 16 5 -10 t+29 10 16 t 30 5                (17) number of iterations rms e run time (s) ||𝛁𝒇|| at final dpf o 0 c.  24 12.56 10.88 3.26*10 -6 sr1 o 0 c.  20 12.55 10.25 3.62*10 -6 bfgs o 0 c.  20 12.55 10.25 4.83*10 -6 dpf o 3 c.  9 12.67 7.45 3.26*10 -4 sr1 o 3 c.  10 14.18 8.10 3.33*10 -4 bfgs o 3 c.  10 12.80 7.78 3.67*10 -4 dpf o 10 c.  15 30.17 11.35 7.19*10 -4 sr1 o 10 c.  14 38.09 11.32 8.91*10 -4 bfgs o 10 c.  14 36.63 9.15 6.27*10 -4 a comprehensive study on the efficiency of three different types of the variable metric method in determining the unknown rocket inner heat flux issn: 2180-1053 vol. 10 no.1 january – june 2018 21 figure 5. the estimation of the triangular heat flux with 𝜎 = 0 figure 6. the estimation of the triangular heat flux with 𝜎 = 3 journal of mechanical engineering and technology 22 issn: 2180-1053 vol. 10 no.1 january – june 2018 figure 7. the estimation of the triangular heat flux with 𝜎 = 10 numerical test-case 3: the unknown boundary heat flux q (l , t ) is assumed applied at x = l in the following form: 𝑞(𝑙, 𝑡) = { 0 0 ≤ 𝑡 ≤ 16 , 3.6 × 106 16 < 𝑡 ≤ 30 } (17) table 3. root mean square error and convergence criteria for estimating step heat flux number of iterations rms e run time (s) ||𝛁𝒇|| at final dpf o 0 c.  33 0.25 12.07 2.10*10 -6 sr1 o 0 c.  28 0.20 10.72 4.01*10 -6 bfgs o 0 c.  28 0.18 10.92 2.01*10 -6 dpf o 3 c.  17 7.77 9.16 3.68*10 -6 sr1 o 3 c.  16 8.45 8.67 2.09*10 -4 bfgs o 3 c.  16 6.47 8.42 3.39*10 -4 dpf o 10 c.  17 21.54 8.62 7.19*10 -4 sr1 o 10 c.  16 25.55 8.72 3.41*10 -4 bfgs o 10 c.  16 29.27 8.91 4.41*10 -4 a comprehensive study on the efficiency of three different types of the variable metric method in determining the unknown rocket inner heat flux issn: 2180-1053 vol. 10 no.1 january – june 2018 23 figure 8. the estimation of the step heat flux with 𝜎 = 0 figure 9. the estimation of the step heat flux with 𝜎 = 3 journal of mechanical engineering and technology 24 issn: 2180-1053 vol. 10 no.1 january – june 2018 figure 10. the estimation of the step heat flux with 𝜎 = 10 in this numerical test-case results are same to triangular heat flux and show that if the function of unknown heat flux is change suddenly, the dfp is better method for estimation of unknown heat flux than other two kinds. for utilize these inverse methods for design and analyze of nozzle we must show that the estimated flux by bfgs or dfp lead to same temperature distribution to real flux. figure 11. the exact and estimate temperature history at x=l with 𝜎 = 3 (the triangular heat flux). a comprehensive study on the efficiency of three different types of the variable metric method in determining the unknown rocket inner heat flux issn: 2180-1053 vol. 10 no.1 january – june 2018 25 figure 12. the exact and estimate temperature history at x= l with 𝜎 = 3 (the third degree polynomial heat flux). figure 11 and 12 show that, the estimate temperature history (the temperature that calculate with estimate heat flux) at x = l , have a good accuracy and could be used in engineering design for rocket nozzle. figure 13. trend for reduction of ‖𝛻𝑓⃗⃗⃗⃗⃗⃗ ‖after each iteration of vmm cycle with 𝜎 = 0 (the third degree polynomial heat flux). 10 15 20 25 30 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 x 10 -3 iteration ||g er ad f || bfgs  =0 sr1  =0 dfp  =0 journal of mechanical engineering and technology 26 issn: 2180-1053 vol. 10 no.1 january – june 2018 figure 14. trend for reduction of ‖𝛻𝑓⃗⃗⃗⃗⃗⃗ ‖after each iteration of vmm cycle with 𝜎 = 3 (the third degree polynomial heat flux). figure 15. trend for reduction of ‖𝛻𝑓⃗⃗⃗⃗⃗⃗ ‖after each iteration of vmm cycle with 𝜎 = 0 (the third degree polynomial heat flux). 5 6 7 8 9 10 11 12 13 14 0 1 2 3 4 5 6 x 10 -3 iteration ||g er ad f || bfgs  =3 sr1  =3 dfp  =3 2 4 6 8 10 12 14 16 18 0 0.002 0.004 0.006 0.008 0.01 0.012 0.014 0.016 0.018 0.02 iteration ||g er ad f || bfgs  =10 sr1  =10 dfp  =10 a comprehensive study on the efficiency of three different types of the variable metric method in determining the unknown rocket inner heat flux issn: 2180-1053 vol. 10 no.1 january – june 2018 27 figure 16. trend for reduction of ‖𝛻𝑓⃗⃗⃗⃗⃗⃗ ‖ after each iteration of vmm cycle with 𝜎 = 0 (the step heat flux). figure 17. trend for reduction of ‖𝛻𝑓⃗⃗⃗⃗⃗⃗ ‖ after each iteration of vmm cycle with 𝜎 = 3 (the step heat flux) 5 10 15 20 25 30 35 0 0.001 0.002 0.003 0.004 0.005 0.006 0.007 0.008 0.009 0.01 iteration ||g er ad f || bfgs  =0 sr1  =0 dfp  =0 2 4 6 8 10 12 14 16 18 0 0.005 0.01 0.015 0.02 0.025 0.03 iteration ||g er ad f || bfgs  =3 dfp  =3 sr1  =3 journal of mechanical engineering and technology 28 issn: 2180-1053 vol. 10 no.1 january – june 2018 figure 18. trend for reduction of ‖𝛻𝑓⃗⃗⃗⃗⃗⃗ ‖ after each iteration of vmm cycle with 𝜎 = 10 (the step heat flux) trend for reduction of ‖𝛻𝑓⃗⃗⃗⃗⃗⃗ ‖ after each iteration can be seen in figure 13 to figure 18. this clearly shows bfgs converge with less iteration than dfp and almost sr1, the total time for estimate the third degree polynomial, the triangular and the step heat flux by the bfgs is 83.6s and by sr1 is 84.63s and dfp is 88.54s, total times show that the sr1 rate for convergence is better than other version of the vmm. total e_rms for estimate the third degree polynomial, the triangular and the step heat flux with σ = 0 , σ = 3 , σ =10 by the bfgs is 149.86 and the sr1 is 165.86 and finally by the dfp is 146.22. 6.0 conclusions inverse heat conduction problem algorithms based on bfgs, sr1 and dfp were formulated in this paper. the various version of vmm was successfully applied for the solution of the inverse heat conduction problem in determining the unknown transient boundary heat flux by utilizing simulated temperature obtained from the boundary with measurement error. from the numerical test cases in this study it is concluded that the inverse solution obtained by using the technique of bfgs is best method for estimation of unknown function with uniform change, and for the function with sudden changes, the dfp has better convergence criteria than other two kinds. 6 8 10 12 14 16 18 0 0.002 0.004 0.006 0.008 0.01 0.012 iteration ||g er ad f || bfgs  =10 sr1  =10 dfp  =10 a comprehensive study on the efficiency of three different types of the variable metric method in determining the unknown rocket inner heat flux issn: 2180-1053 vol. 10 no.1 january – june 2018 29 7.0 references abboudi, s. and artioukhine, a. (2002). two dimensional computational estima-tion of transient boundary conditions for a flat specimen. in: proceedings of the fourth international conference on inverse problems engineering: theory and practice, june 13–18, rio, brasil, asme 2003. bae, j.h. hyun, j.m. and kwak, h.s. (2004) mixed convection from a multiblock heater in a channel with imposed thermal modulation, numerical heat transfer part a, 45, 329–345. bauzin, j.g. and laraqi, n. (2004) simultaneous estimation of frictional heat flux and two thermal contact parameters for sliding contacts, numerical heat transfer 45 (4) 313–328. beck, j. v. blackwell b. and st. clair, c.r. (1985). inverse heat conduction-ill posed problem, wiley: new york. ch. h. h. and wu, h. h. (2006). an inverse hyperbolic .heat conduction problem in estimating surface heat flux by the conjugate gradient method, journal of physics d: applied physics,39, 4087-4096. daun kj, morton dp, howell jr. (2003). geometric optimization of radiant enclosures containing specular surfaces. asme journal of heat transfer, 125, 845–851. franc-a fr., howell j., ezekoye oa., morales jc. (2002). inverse design of thermal systems. in: hartnett jp, irvine tf,(ed.). advances in heat transfer, (vol. 36. pp.1-110), new york: elsevier. hong, y. k. and baek, s.w. (2006). inverse analysis for estimating the unsteady inlet temperature distribution for two-phase laminar flow in a channel, int. j.heat mass transfer , 49, 1137–1147. huang c h and chen (2000) a three-dimensional inverse forced convection problem in estimating surface heat flux by conjugate gradient method int. j. heat mass transfer 43 3171–81. junxiang shi, jianhua wang, (2009) inverse problem of transpiration cooling for estimating wall heat flux by ltne model and cgm method, international journal of heat and mass transfer, 52, 2714–2720 khoshkam, h. and alizadeh m., (2011) inverse problem of rocket nozzle throat for estimating inner wall heat flux by broydon–fletcher–goldfarb–shanno & conjugate gradient method, international review of mechanical engineering (i.re.m.e), 5(5) kowsary, f. behbahaninia, a. pourshaghaghy, a. (2006). transient heat flux function estimation utilizing the variable metric method, international communications in heat and mass transfer, 33, 800–810. journal of mechanical engineering and technology 30 issn: 2180-1053 vol. 10 no.1 january – june 2018 linhua l., heping t. and qizheng y. (1999). inverse radiation problem of temperature filed in three dimensional rectangular furnaces. international communications in heat and mass transfer; 26, 239–48. luk �̌� an l. and spedicato e. (2000). variable metric methods for unconstrained optimization and nonlinear least squares. journal of computational and applied mathematics, 124, 61–95. patankar. s.v. (1980). numerical heat transfer and fluid flow. new york: mcgraw hill pourshaghagh, a. kowsary, f. and behbahaninia, a. (2007) comparison of four different versions of the variable metric method for solving inverse heat conduction problems. heat mass transfer 43, 285–294 prud'homme, m. and jasmin, s. (2003). determination of a heat source in porous medium with convective mass diffusion by an inverse method. international journal of heat and mass transfer 46 2065–2075. rao s.s. (1995). optimization; theory and applications. new delhi: new age international (p) limited publishers. sparrow, e. m., haji-sheikh, a. and lundgren, t.s. (1964). the inverse problem in transient heat conduction, asme journal of applied mechanics, 86, 369-375. stolz, g. jr., (1960) numerical solutions to an inverse problem of heat condu-ction for simpl shapes, asme journal of heat transfer, 82, 20-26. tikhonov a.n. and arsenin, v.y. (1977). solutions of ill-posed problems, winston: washington. tsung-chien chen, chiun-chien liu, (2008). inverse estimation of heat flux and temperature on nozzle throat-insert inner contour, international journal of heat and mass transfer, 51, 3571–3581. wang, h.n. and wang, j.h. (2006). a numerical investigation of ablation and transpiration cooling using the local thermal non-equilibrium model [r], in: proceeding of the 42nd aiaa/asme/sae/asee joint propulsion conference & exhibit, sacramento, california, july 9–12, aiaa-5264. wang, j.h., wang, h.n., sun, j.g. and wang, j. (2007). numerical simulation of control ablation by transpiration cooling, heat mass transfer, 43, 471–478. zhang, z.z., cao, d.h. and zeng, j.p. (2004). property of a class of variable metric methods. applied mathematics letters 17 , 437–442. http://en.wikipedia.org/w/index.php?title=numerical_heat_transfer_and_fluid_flow&action=edit&redlink=1 issn: 2180-1053 vol. 8 no. 2 july – december 2016 63 a non-dimensional study on both analytic and numeric thermo-elastic behavior of functionally graded thick-walled cylinders under a combination of thermo-mechanical loads h. asgharzadeh shirazi1, m. abedi2, a. asnafi13*, m. salimi1 1school of mechanical engineering, iran university of science and technology, narmak, 16846-13114, tehran 2school of mechanical engineering, shiraz university, shiraz 71348-13668, iran 3hydro-aeronautical research center, shiraz university, shiraz, 71348-13668, iran. abstract efforts using non-dimensional parameters, the governing equations of homogeneous and heterogeneous cylinders made of functionally graded material (fgm) were derived under a combination of thermo-mechanical loads. the equations were solved analytically and numerically in a severe temperature and pressure gradient environment. the radial and circumferential stresses together with the radial displacement of fgm cylinder were analytically evaluated and then compared to conventional homogeneous ones. besides, in order to assess the accuracy of derived equations, a numerical solution (ns) was performed using finite element method. it was shown that the numerical solution was in accordance to the analytical solution (as). the results of present work show that the use of fgm can optimize the thermo-elastic performance of the cylinders which are exposed to the joint mechanical and thermal loads. keywords: thick-walled cylinder; fgm heterogeneous materials; finite element method; 1.0 introduction metals and many composites with high levels of strength to weight ratio, have been successfully employed in the fields of mechanical engineering; however, they do not demonstrate good performance in the environments with high temperature conditions. on the other hand, materials such as ceramics show excellent performance in hightemperature environments while, they are sometimes unreliable in terms of strength and stiffness. in order to encompass these conversely goals in some applications, the functionally graded materials (fgms) have been introduced. first time, they were proposed in the mid-1980s in japan as a thermal coating for some engineering applications (mahamood, akinlabi, shukla & pityana, 2012). japanese researchers used this developed composite as a thermal barrier across a 10 mm thickness with two different temperatures of 1000 and 2000k on both sides of the area (mahamood et al., 2012). generally, fgms are heterogeneous composites whose properties such as modulus of elasticity, thermal conductivity and mass density, change gradually from one side to the other side of the material domain. these materials are well-known for * corresponding author email: asnafi@shirazu.ac.ir journal of mechanical engineering and technology 64 issn: 2180-1053 vol. 8 no. 2 july– december 2016 their ability in isolating or separating two different environments with diverged designed goals. these materials can support high temperatures and extreme temperature gradients as they do for mechanical loads. as practical examples, the internal parts of combustion engines, turbines and power plants are under high temperature and severe temperature gradient jointly. it is worth noting that the low thermal conductivity and low coefficient of thermal expansion of fgm materials allow them to accept high temperatures and severe temperature gradients (azadi, 2009; azadi & shariyat, 2010; damircheli & azadi, 2011). in recent years, many works have been conducted on the behavior of functionally graded materials (tuntucu & ozturk, 2001). in 2001, tutuncu and ozturk (2001) used the infinitesimal theory of elasticity to obtain closed-form solutions for both stress and displacement in functionally graded cylindrical and spherical vessels subjected to internal pressure. in 2008, argeso and eraslan (2008) assessed an estimation of the thermo-elastic response of cylinders and tubes, using temperature-dependent physical properties. using exact closed-form solutions based on plane elasticity theory, nejad, abedi, lotfian and ghannad (2012) studied both the stress and displacement profiles in a thick spherical shell made of functionally graded materials with exponential-varying elasticity modulus under uniform pressure. they assumed plane strain condition and considered a fixed value for the poisson’s ratio (nejat et al.,2012). in 2012, bayat, ghannad and torabi (2012) studied a functionally graded thick-walled hollow sphere on the assumption of one-dimensional steady temperature distribution. they supposed the thermal and mechanical properties of sphere to be varied exponentially in the radial direction (bayat et al., 2012). using non-dimensional parameters, this research attempts to develop thermo-elastic analysis of thick-walled structures like cylinder and disk made of functionally graded materials. therefore, the aim of this paper is to improve the thick-walled cylinder behaviors by proposing dimensionless governing equations of homogeneous and heterogeneous fgm cylinder under thermo-mechanical loads. the results were also compared to other works and discussed. 2.0 fundamental equations in the plane elasticity theory, it is assumed that the cross-section plane which is perpendicular to the central axis of the cylinder will remain planar and perpendicular to the central axis after applying pressure and deformation. furthermore, it is supposed that the radial deformations along the perimeter remain fixed but vary in radial direction. theoritically, radial deformations dependent only on the radius ur(r). in order to solve the problem, a hollow cylinder of non-homogeneous fgm material with inner and outer radii of ri and ro under uniform internal and external pressures of pi and po and the internal and external temperature surfaces of ti and to, is considered. figure 1 shows a section of the assumed cylinder under combined mechanical and thermal loading. a non-dimensional study on both analytic and numeric thermo-elastic behavior of functionally graded thick-walled cylinders under a combination of thermo-mechanical loads issn: 2180-1053 vol. 8 no. 2 july – december 2016 65 (a) (b) figure 1. (a) cross-section of a thick-walled cylinder with internal radius “ri” and external radius “ro”, (b) finite element mesh region. journal of mechanical engineering and technology 66 issn: 2180-1053 vol. 8 no. 2 july– december 2016 due to axisymmetric condition, material properties, loads and boundary conditions and according to plane elastic theory, the shear stresses and strains become all zero. thus, the normal strains are: (1)   1 r r z t e          (2)   1 r z t e           (3)   1 z z r t e          where e is the elasticity modulus, ν is poisson’s ratio, α is coefficient of thermal expansion and t is the temperature gradient. moreover, σr, σθ and σz indicate normal stresses and εr, εθ and εz denote normal strains in r, θ and z directions, respectively. for plane strain condition in cylinders, normal strain in z direction turns into zero i.e. εz=0; so eq. (3) develops into: (4)  z r e t       by substituting eq. (4) into eqs. (1) and (2), we have: (5)        1 1 1 1 1 r r t e                   (6)        1 1 1 1 1 r t e                    rewriting eqs. (5) and (6) for stresses results in: (7)       1 1 1 2 1 r r e t                   (8)       1 1 1 2 1 r e t                    now, eqs. (7) and (8) can be rearranged to the following general form: (9) 1 2 1 2 2 1 1 2 2 2 r r a a a a e a a a a t                         in which a1 and a2 relate to the poisson’s ratio as: (10)       1 2 1 1 2 1 1 2 1 a a                 a non-dimensional study on both analytic and numeric thermo-elastic behavior of functionally graded thick-walled cylinders under a combination of thermo-mechanical loads issn: 2180-1053 vol. 8 no. 2 july – december 2016 67 in axisymmetric condition, all displacements become zero except for radial displacement (ur). the radial displacement, of course, is a function of polar radius r only; therefore, strain-displacement relations in axisymmetric condition in cylindrical coordinate will be: (11) ir r i ir i pdu du dr e dr pu u r e r             where ei is the elasticity modulus of internal surface. now, the dimensionless radius r and the dimensionless radial displacement u can be defined as: (12) i r r r  (13) i r i i e u u r p  substituting eq. (11) into eq. (9) yields: (14) 1 2 1 2 2 1 1 2 2 2 r du dr a a a a u e a a a a r t                                where: (15)  1 2 1 2 , , , , i i i i i i i i a a e te t e t p e t a p             in eq. (15), αi is the thermal expansion coefficient of cylinder internal surface. neglecting the body forces, the equilibrium equation in axisymmetric conditions and in cylindrical coordinates will be: (16) 0rr d dr r       it can also be expressed as: (17) 0 ; r r irr i pd dr r p                    replacing eq. (14) in eq. (17) gives up: journal of mechanical engineering and technology 68 issn: 2180-1053 vol. 8 no. 2 july– december 2016 (18)   2 2 2 2 1 1 1 1 1 ad u rde du rde u f r dr r edr dr r a edr               where (19)    d e t f r e dr   eq. (18) is a bessel-like second order differential equation with general solution as in (kreyszig, 2010): (20)        1 2u r c g r c h r i r   where c1 and c2 are coefficients that can be calculated easily from the boundary conditions. h(r) and g(r) are general solutions; while i(r) is the particular solution of the differential equation. the particular solution of the differential equation can be calculated as (kreyszig, 2010): (21)                   f r h r f r g r i r g r dr h r dr w r w r     where: (22)          dh r dg r w r g r h r dr dr   to solve eq. (18), a particular relation must be firstly attributed to the mechanical properties. power function is the most common function for cylindrical hollow tubes. for constant poisson’s ratio (ν), the elasticity modulus (e), the thermal expansion coefficient (α) and thermal conductivity (k) can be considered as: (23)   1 n i e r e r (24)   2 n i r r  (25)   3 n i k r k r where ki is the thermal conductivity on the inner surface of the cylinder and n1, n2 and n3 are material parameters. the homogeneous part of the eq. (18) can be also obtained just by applying eqs. (23) to (25): (26)   2 2 2 1 12 1 1 1 0 ad u du r n r n u dr dr a           eq. (26) demonstrates an euler-cauchy equation. assuming u=rm, the characteristic equation can be achieved as: (27) 2 2 1 1 1 1 0 a m n m n a          the roots of eq. (27) are: a non-dimensional study on both analytic and numeric thermo-elastic behavior of functionally graded thick-walled cylinders under a combination of thermo-mechanical loads issn: 2180-1053 vol. 8 no. 2 july – december 2016 69 (28) 21 2 1,2 1 1 1 4 ; 4 2 n a m n n a         given the limitations of poisson's ratio (0<υ<0.5), the value of δ is always positive. as a result, m1 and m2 always have distinct real values. thus, the general solutions of g(r) and h(r) are: (29)     1 2 m m g r r h r r     substituting eq. (29) into eq. (22) results in: (30)     1 2 1 2 1 m m w r m m r     additionally, by substituting eqs. (29) and (30) in eq. (21), one can reach such the following relation: (31)      1 1 2 2 1 1 2 1 1 m m m m i r r r f r dr r r f r dr m m            in order to calculate the particular solution stated in the above equation, the term f(r) should be expressed explicitly; however, in order to calculate f(r), the relation of temperature gradient t should be obtained in explicit form and in terms of polar radius r. 2.1 axisymmetric heat transfer equation the steady state heat transfer in cylindrical coordinate under axisymmetric condition can be obtained using the following famous ode (rohsenow & warren,1998): (32) 0 d dt kr dr dr       with reference to eq. (25), the above equation can be expressed as: (33) 3 1 0 nd dt r dr dr       relative to the value of ni, two cases of ni=0 and ni≠0 must be studied and analyzed. 2.1.1 case 1 ni≠0 in this case, the solution of the differential equation becomes: (34) 31 2 n t d r d    where d1 and d2 are constants that can be calculated from the following boundary conditions: journal of mechanical engineering and technology 70 issn: 2180-1053 vol. 8 no. 2 july– december 2016 (35)         * 1 1 i i o o t rt r r t t r r t t r k t              note also that: (36) * o i o i r k r t t t        by applying the boundary conditions, one can attain such the following relations: (37) 3 3 3 * 1 * 2 1 1 1 n n n t d k t k d k              thus, in this case, f(r) in eq. (18) can be obtained by substituting eqs. (23), (24) and (37) into eq. (19): (38)      2 3 2 1 1 1 2 3 1 1 2 2 n n n f r n n n d r n n d r           consequently, the particular solution i(r), will be accomplished by substituting eq. (37) into eq. (31) as: (39)   2 3 2 1 1 3 4 n n n i r c r c r      where: (40)           1 2 3 1 3 1 2 3 2 2 3 1 2 2 4 1 2 2 2 1 1 1 1 n n n d c m n n m n n n n d c m n m n                        now, the complete solution for u(r) is the sum of homogenous and particular solutions, i.e. (41)   2 31 2 2 1 1 1 2 3 4 n nm m n u r c r c r c r c r        finally, by substituting eqs. (23), (24), (34) and (40) into eq. (14), the resulting radial and circumferential stress expressions are: (42) 1 2 31 1 2 1 1 2 1 1 11 12 13 14 n n nm n m n n n r q r q r q r q r           (43) 1 2 31 1 2 1 1 2 1 1 21 22 23 24 n n nm n m n n n q r q r q r q r             where: a non-dimensional study on both analytic and numeric thermo-elastic behavior of functionally graded thick-walled cylinders under a combination of thermo-mechanical loads issn: 2180-1053 vol. 8 no. 2 july – december 2016 71 (44)         21 1 1 2 1 22 1 2 2 2 23 2 3 2 3 1 3 1 1 24 2 2 4 1 4 1 2 1 1 q a m a c q a m a c q n n a c a c a d q n a c a c a d                        11 1 1 2 1 12 2 1 2 2 13 2 3 1 3 2 3 1 1 14 2 1 4 2 4 1 2 1 1 q m a a c q m a a c q n n a c a c a d q n a c a c a d                in eqs. (41), (42) and (43) the constants c1 and c2 are still unknown. as mentioned previously, these constants must be calculated from mechanical boundary conditions i.e. (45)         * 1 1 rr i i r o o r rr r p r r p r k p                    in which: (46) * o i p p p  by applying the boundary conditions of eq. (45), constants c1 and c2 are obtained as: (47)   1 3 4 1 32 1 2 1 2 1 1 1 2 1 1 1 2 1 1 1 2 1 1 1 1 * 13 141 1 1 1 1 1 1 1 1 2 n n n n nm n m n m n m n m n m n m n m n m n k k k k k p q q k k k k k k c m a a                                 (48)   1 3 4 1 31 1 1 1 1 1 2 1 1 1 2 1 1 1 2 1 1 1 1 1 1 * 13 141 1 1 1 1 1 2 2 1 2 n n n n nm n m n m n m n m n m n m n m n m n k k k k k p q q k k k k k k c m a a                                 it is obvious that the calculation of the constants c1 and c2, will subsequently results in the non-dimensional radial displacement u together with the non-dimensional radial and circumferential stresses σr and σθ. 2.1.2 case 2 ni =0 with the same procedure, the solution of the differential equation becomes: (49)  1 2lnt d r d  where d1 and d2 are the constants that must be calculated from the thermal boundary conditions expressed in eq. (35): (50)   * 1 2 1 ln 1 t d k d       thus, in this case by substituting eqs. (23), (24) and (50) into eq. (19), f(r) turns into: (51)        2 2 1 1 1 2 1 1 1 2 2 ln n n f r n n d r r d n n d r            consequently, the particular solution i(r) will be realized by substituting eq. (51) in eq. (31): journal of mechanical engineering and technology 72 issn: 2180-1053 vol. 8 no. 2 july– december 2016 (52)    2 2 1 1 3 4 ln n n i r c r r c r     where: (53)                  1 2 1 3 1 2 2 2 1 1 2 2 1 2 1 2 1 4 2 2 1 2 2 2 1 2 2 2 1 1 2 2 1 1 1 1 n n d c m n m n d n n d n n n n d c m n m n m n m n                                  the complete solution for u(r), which is the sum of homogenous and particular solutions, develops into: (54)    1 2 2 2 1 1 1 2 3 4 ln m m n n u r c r c r c r r c r       at last, by substituting eqs. (23), (24), (49) and (54) into eq. (14), the resulting radial and circumferential stress expressions are: (55)  1 1 2 1 1 2 1 2 1 1 11 12 13 14 ln m n m n n n n n r q r q r q r r q r           (56)  1 1 2 1 1 2 1 2 1 1 21 22 23 24 ln m n m n n n n n q r q r q r r q r             where: (57)         21 2 1 1 1 22 2 2 1 2 23 2 2 3 1 3 1 1 24 2 3 2 2 4 1 4 1 2 1 1 q a m a c q a m a c q n a c a c a d q a c n a c a c a d                        11 1 1 2 1 12 1 2 2 2 13 2 1 3 2 3 1 1 14 1 3 2 1 4 2 4 1 2 1 1 q a m a c q a m a c q n a c a c a d q a c n a c a c a d                the constants c1 and c2, in this case, are gained by applying mechanical boundary conditions expressed in eq. (45), i.e. (58)     1 3 1 32 1 2 1 1 1 2 1 1 1 2 1 1 1 2 1 1 1 * 13 141 1 1 1 1 1 1 1 1 2 ln n n n nm n m n m n m n m n m n m n m n k k k k k p q q k k k k k k c a m a                              (59)     1 3 1 31 1 1 1 2 1 1 1 2 1 1 1 2 1 1 1 1 1 * 13 141 1 1 1 1 1 2 1 2 2 ln n n n nm n m n m n m n m n m n m n m n k k k k k p q q k k k k k k c a m a                              a non-dimensional study on both analytic and numeric thermo-elastic behavior of functionally graded thick-walled cylinders under a combination of thermo-mechanical loads issn: 2180-1053 vol. 8 no. 2 july – december 2016 73 3.0 results and discussions in previous section, the governing equations of fgm thick-walled cylinder under mechanical and thermal loads were successfully derived using plane elasticity theory. in this section, a practical case study is investigated and the results are compared to that one obtained from elastic cylinders. without any loss of generality, the inside and outside temperatures and relative pressures are assumed to be 25 ̊c, 300 ̊c, 80 mpa and zero, respectively. the elasticity modulus, poisson’s ratio, the coefficients of thermal expansion and heat conduction at the inner surface of the cylinder were selected as ei = 200 gpa, ν=0.3, αi = 17.5×10 -6 /◦c and ki=15 w/mk, respectively. due to use of r as dimensionless radius, the derived equations are independent on the inner (ri) and outer (ro) radii; however, the radii values of ri=40 mm and ro=60 mm were chosen to analyze the numerical solution. to better investigation, the results have been presented for various values of fgm parameter in the range of -2≤n≤2. in order to perform numerical analysis, a geometry sample was modeled using finite element method for a comparative study. the finite element (fe) model was constructed using comsol multiphysics® software. the outputs of this simulation have been utilized to compare the thermo-elastic results obtained from both analytical and numerical solutions for the functionally graded thick-walled cylinder under a combination of mechanical and thermal loads. figure 2 shows the radial distribution of elasticity modulus e versus dimensionless radius r for various values of n. as can be observed from this figure, elasticity modulus is constant at inner surface for different values of n, while it increases at outer surface when n enhances from n=-2 to n=2. on the other hand, with attention. figure 2. radial distribution of elasticity modulus. 0 0.5 1 1.5 2 2.5 1 1.1 1.2 1.3 1.4 1.5 e / e i r n=-2 n=-1 n=0 n=1 n=2 journal of mechanical engineering and technology 74 issn: 2180-1053 vol. 8 no. 2 july– december 2016 to the slope of elasticity modulus curves, it is seen that the absolute value of curve slopes for n>0, are greater than the case of n<0. figure 3 indicates the radial distribution of temperature in homogeneous (n=0) and heterogeneous (n≠0) cylinders in present work. as shown in this figure, the temperature reduces by increasing the value of n from n=-2 to n=2. moreover, this figure reveals that the temperature decreases with an increase in radius. figure 3. distribution of the normalized radial temperature under the thermal loading. figure 4 demonstrates the dimensionless radial stress distribution 𝜎𝑟 versus dimensionless radius r in response of both analytical (as) and numerical (ns) solutions. according to this figure, the magnitude of radial stress decreases/increases for n<0 / n>0, respectively. therefore, this decrease and increase in the radial stress depends on |𝑛|. based on this figure, for non-dimensional radius r<1.1, the stress values for all amount of n are relatively close to each other; while, significant differences in stress values are seen for r>1.1. the dimensionless circumferential stress 𝜎𝜃 versus dimensionless radius r for homogeneous (n=0) and heterogeneous (n≠0) cylinders is plotted in figure 5. according to this figure, circumferential stress for more negative values of n in internal, central and 0.4 0.6 0.8 1 1 1.1 1.2 1.3 1.4 1.5 t / t i r as, n=-2 as, n=-1 as, n=0 as, n=1 as, n=2 ns, n=-2 ns, n=-1 ns, n=0 ns, n=1 ns, n=2 a non-dimensional study on both analytic and numeric thermo-elastic behavior of functionally graded thick-walled cylinders under a combination of thermo-mechanical loads issn: 2180-1053 vol. 8 no. 2 july – december 2016 75 figure 4. normalized radial stress of homogeneous (n=0) and fgm heterogeneous (n≠0) cylinders under the combined thermo-mechanical loads. figure 5. normalized circumferential stress of homogeneous (n=0) and fgm heterogeneous (n≠0) cylinders under the combined thermo-mechanical loads. -1.8 -1.6 -1.4 -1.2 -1 -0.8 -0.6 -0.4 -0.2 0 1 1.1 1.2 1.3 1.4 1.5 σ r / p i r as, n=-2 as, n=-1 as, n=0 as, n=1 as, n=2 ns, n=-2 ns, n=-1 ns, n=0 ns, n=1 ns, n=2 -10 -5 0 5 10 15 20 1 1.1 1.2 1.3 1.4 1.5 σ θ / p i r as, n=-2 as, n=-1 as, n=0 as, n=1 as, n=2 ns, n=-2 ns, n=-1 ns, n=0 ns, n=1 ns, n=2 journal of mechanical engineering and technology 76 issn: 2180-1053 vol. 8 no. 2 july– december 2016 external regions of cylinder is less than, equal to and higher than the corresponding homogeneous (n=0) cylinder respectively and vice versa for more positive values of n. in other words, with reference to this figure, it can be found that the curves of circumferential stress meet and cross each other in the range of 1.3≤r≤1.4. figures 4 and 5 can illustrate the important role of radial thickness in terms of radial and circumferential stresses in fgm thick-walled cylinder. figure 6 explains the distribution of normalized radial displacement versus nondimensional radius r in homogeneous (n=0) and heterogeneous (n≠0) cylinders. as can be observed from this figure, the displacement of heterogeneous (n≠0) cylinder is lower than homogeneous (n=0) one for negative values of n and would be vice versa for n>0. this ratio, of course, is almost constant along the wall and the amount of differences depend on the magnitude of |𝑛|. figure 6. normalized radial displacement of homogeneous (n=0) and fgm heterogeneous (n≠0) cylinders under the combined thermo-mechanical loads overall, according to the all above-mentioned results, it can be concluded that the radial and circumferential stresses and the radial displacement of fgm heterogeneous (n≠0) cylinder increase/decrease relative to the sign of n (positive or negative) so that whatever |𝑛| grows to be larger, more changes can be seen. therefore, based on the need to decrease or increase in stress and displacement of fgm heterogeneous (n≠0), we can use the positive or negative n parameter in our design. moreover, it can be found from figures 4, 5 and 6 that there are good agreements between analytical (as) and numerical (ns) solutions. these agreements give acceptable approvals on the governing equations obtained in this present study. the fem contours of stresses and 10 11 12 13 14 15 16 1 1.1 1.2 1.3 1.4 1.5 u r as, n=-2 as, n=-1 as, n=0 as, n=1 as, n=2 ns, n=-2 ns, n=-1 ns, n=0 ns, n=1 ns, n=2 a non-dimensional study on both analytic and numeric thermo-elastic behavior of functionally graded thick-walled cylinders under a combination of thermo-mechanical loads issn: 2180-1053 vol. 8 no. 2 july – december 2016 77 displacements of both homogeneous (n=0) and fgm heterogeneous (n≠0) cylinders for all values of n are presented in figure 7. (a) (b) (c) figure 7. distribution of (a) radial stress, (b) circumferential stress and (c) radial displacement of homogeneous (n=0) and fgm heterogeneous (n≠0) cylinders under the combined thermo-mechanical loads (row no.1 to row no.5 are respectively related to: n=-2, -1, 0, 1, 2). journal of mechanical engineering and technology 78 issn: 2180-1053 vol. 8 no. 2 july– december 2016 4.0 conclusions in this research, both the analytic and numeric solutions of homogeneous and fgm heterogeneous thick-walled cylinders were successfully performed under a combination of mechanical and thermal loads. the numerical results indicated that the governing equations, obtained in present work, are acceptable in order to analyze the homogeneous and fgm heterogeneous thick-walled cylinders under joint mechanical and thermal loads. they let somebody see and predict the optimum state of problem in terms of stress and displacement based on desired design requirements. it was found that the material parameter has great effects on the stress and displacement of fgm heterogeneous thick-walled cylinders. in other words, from a design point of view, they would be useful parameters as they can be tailored to specific applications in order to control the stress and displacement of thick-walled fgm cylinder. references argeso, h. a. & eraslan, n. (2008). on the use of temperature-dependent physical properties in thermo mechanical calculations for solid and hollow cylinders, international journal of thermal sciences, 47, 136–146. azadi m., & shariyat, m. (2010). nonlinear transient transfinite element thermal analysis of thick-walled fgm cylinders with temperature-dependent material properties, meccanica, 45(3), 305-318. azadi, m. & azadi m., (2009). nonlinear transient heat transfer and thermoelastic analysis of thick-walled fgm cylinder with temperature-dependent material properties using hermitian transfinite element, journal of mechanical science and technology, 23(10), 2635-2644. bayat y., ghannad m., & h. torabi. (2012). analytical and numerical analysis for the fgm thick sphere under combined pressure and temperature loading, archive of applied mechanics, 82(2), 229-242. damircheli m., & azadi m. (2011). temperature and thickness effects on thermal and mechanical stresses of rotating fg-disks, journal of mechanical science and technology, 25(3), 827-836. kreyszig e. (2010). advanced engineering mathematics, john wiley & sons. mahamood r.m., akinlabi t.a., shukla m., & pityana s. (2012). functionally graded material: an overview, proceedings of the world congress on engineering (wce) 3, 83-86. nejad m.z., abedi m., lotfian m.h., & ghannad m. (2012). an exact solution for stresses and displacements of pressurized fgm thick-walled spherical shells a non-dimensional study on both analytic and numeric thermo-elastic behavior of functionally graded thick-walled cylinders under a combination of thermo-mechanical loads issn: 2180-1053 vol. 8 no. 2 july – december 2016 79 with exponential-varying properties, journal of mechanical science and technology, 26(12), 4081-4087. rohsenow, warren m. (1998). handbook of heat transfer. vol. 3. new york: mcgraw-hill. tutuncu, n. & ozturk, m. (2001). exact solution for stresses in functionally graded pressure vessels, composites part b: engineering, 32 (8), 683-686. journal of mechanical engineering and technology 80 issn: 2180-1053 vol. 8 no. 2 july– december 2016 issn: 2180-1053 vol. 7 no. 2 july december 2015 development of an unmanned underwater remotely operated crawler (roc) for monitoring application 41 development of an unmanned underwater remotely operated crawler (roc) for monitoring application m. s. m. aras1*, iktisyam zainal1, s.s. abdullah2, a.m. kassim1 and h.i. jaafar1 1faculty of electrical engineering, universiti teknikal malaysia melaka, hang tuah jaya, 76100 durian tunggal, melaka, malaysia, 2department of electronic system engineering, malaysia-japan international institute of technology, universiti teknologi malaysia kl campus, jalan sultan yahya petra, 54100 kuala lumpur, malaysia abstract underwater vehicles are a type of vehicle that a type of vehicles that able to explore the underwater world. remotely operated crawler (roc) is one of the unmanned underwater vehicle (uuv) that can be categorized in remotely operated vehicle (rov) class. the specialty of roc allows for underwater intervention by staying a direct contact with the seabed. the common issues face for the crawlers are the underwater pressure, maneuverability, power and control. besides that, the surface of the seabed become one of the problems in that restrict on roc maneuverability. designing a roc that can crawl in any surface conditions is one of the issues emerged in this project. this project is about developing the roc in order to fulfil a specific mission involving certain tasks. roc lend themselves to long-term work and offer a very stable platform for manipulating objects and taking measurements better than other rov. development the roc based on wheel mechanism that allows the roc moves with direct contact with the seabed without any glitch and have an ability to operate in any condition of the underwater environment. the wheel mechanism is adapted based on the tanks which is the chain type wheels. the performance of the roc will be verified based on experiments conducted on the cluttered condition either on the surface or underwater. the operation of roc can achieve excellent performance with an unexpected level of environmental condition. keywords: remotely operated crawler, wheel mechanism, chain type wheels. 1.0 introduction all the exploration of the oil and gas industries is not concentrated at on the land, but also in the offshore and deep sea as more oil wells found. corresponding author e-mail: shahrieel@utem.edu.my issn: 2180-1053 vol. 7 no. 2 july december 2015 journal of mechanical engineering and technology 42 thus, as offshore explorations have increased the risk taken by human to drill petroleum. there are many cases regarding on the drilling, pipelines, transportation and storage accidents (www.nationalgeographic. com). even though there are safety measures performed, yet accident can happen anytime without notice. underwater pipelines have a total length of kilometers. they carry oil, gas, condensate, and their mixtures. pipelines are among the main factors of environmental risk during offshore oil developments, along with tanker transportation and drilling operation. the causes of pipeline damage can be range from material defects and pipe corrosion to ground erosion, tectonic movements at the bottom of the sea and encountering ship anchors and bottom trawls (www.eia.gov). statistical data show that the average probability of accidents occurring on the underwater main pipelines of north america and western europe is 9.3x10-4 and 6.4x10-4, respectively (www.eia.gov),(www.offshore-environment.com),(hyakudome,2011). the main causes of these accidents are material and welding defects just like what happened in russia offshore project sakhalin-1, in the year of 1994 and cause a huge impact to the arctic ecosystems as the pipeline collapse (wood et. al, 2013), (moonesun et. al, 2012),(welling and edwards, 2005). modern technology of pipeline construction and exploitation have been introduced. for example, the usage of rov and roc in construction the underwater pipeline connections. underwater technology research group (uterg) from the faculty of electrical engineering, universiti teknikal malaysia melaka has developed the rov (mohd aras et. al, 2013a), (mohd aras et. al, 2013b), (mohd aras et. al, 2013c), (ali et. al, 2013), (mohd aras et. al, 2013d). the model of rov obtained from the system identification technique can be referred to (mohd aras et. al, 2013e). this technology eliminates the risk taken by divers to dive into the deep and cold water condition. the roc used in pipeline inspections and even constructions on the seabed along with other types of rov and reduce human intervention doing the welding and inspection process. thus, this kind of incident motivates to study on the roc design requirement to fulfil underwater inspections based on the project’s scope and later there will be innovations in the development of roc that help to build offshore facilities. one of the roc design for the task of pipeline construction is the subsea crawler as shown in figure 1 owned by ihc marine and mineral projects, south africa. the crawler is owned by qinetiq north america as shown in figure 2 which mainly use for explosive ordnance disposal (eod) hull inspection. scopes for this project are limited into few aspects. first, the crawler will have two degrees of freedom (dof) for the maneuverability. then, upon completion, the crawler will be tested on the hard surface underwater bed. the motions of the controller will be designed as forward, reverse, left and right movement. the design issn: 2180-1053 vol. 7 no. 2 july december 2015 development of an unmanned underwater remotely operated crawler (roc) for monitoring application 43 specifications are based on the scope drafted which are the operation depth is more or less than 50 meters. the control range of the crawler are strictly depends on the length of the connection cord and the pressure to withstand is about more than 5 bars. the crawler must be water and shock resistance and durability in term of maneuverability and movement, either on the land or underwater. 2 pipelines of north america and western europe is 9.3x10-4 and 6.4x10-4, respectively (www.eia.gov),(www.offshore-environment.com),(hyakudome,2011). the main causes of these accidents are material and welding defects just like what happened in russia offshore project sakhalin-1, in the year of 1994 and cause a huge impact to the arctic ecosystems as the pipeline collapse (wood et. al, 2013), (moonesun et. al, 2012),(welling and edwards, 2005). modern technology of pipeline construction and exploitation have been introduced. for example, the usage of rov and roc in construction the underwater pipeline connections. underwater technology research group (uterg) from the faculty of electrical engineering, universiti teknikal malaysia melaka has developed the rov (mohd aras et. al, 2013a), (mohd aras et. al, 2013b), (mohd aras et. al, 2013c), (ali et. al, 2013), (mohd aras et. al, 2013d). the model of rov obtained from the system identification technique can be referred to (mohd aras et. al, 2013e). this technology eliminates the risk taken by divers to dive into the deep and cold water condition. the roc used in pipeline inspections and even constructions on the seabed along with other types of rov and reduce human intervention doing the welding and inspection process. thus, this kind of incident motivates to study on the roc design requirement to fulfil underwater inspections based on the project’s scope and later there will be innovations in the development of roc that help to build offshore facilities. one of the roc design for the task of pipeline construction is the subsea crawler as shown in figure 1 owned by ihc marine and mineral projects, south africa. the crawler is owned by qinetiq north america as shown in figure 2 which mainly use for explosive ordnance disposal (eod) hull inspection. scopes for this project are limited into few aspects. first, the crawler will have two degrees of freedom (dof) for the maneuverability. then, upon completion, the crawler will be tested on the hard surface underwater bed. the motions of the controller will be designed as forward, reverse, left and right movement. the design specifications are based on the scope drafted which are the operation depth is more or less than 50 meters. the control range of the crawler are strictly depends on the length of the connection cord and the pressure to withstand is about more than 5 bars. the crawler must be water and shock resistance and durability in term of maneuverability and movement, either on the land or underwater. figure 1. subsea crawler for oil and gas pipeline constructions. (c.r. deepak) figure 1. subsea crawler for oil and gas pipeline constructions. (c.r. deepak) figure 2. hull crawler by qinetiq north america. (jansen, g.,2013) 2.0 methodology for the development and modelling of unmanned underwater remotely operated crawler (roc) for monitoring application, it all starts with a project plan. every detail of the project must be pointed out in term of designs, costs, materials selections, components selections and prototype testing and assembly process. for this project, it is divided into two parts; software and hardware. the first objective which is to design the remotely operated crawler. the design uses the cad software (solidwork) and simulations need to be done for different designs and selected with the best design as shown in figure 3. all designs have a different chassis design while the wheels, movement mechanisms and controller remain the same. then, development of the prototype in terms of hardware. figure 3. assembly design of the crawler. by determining the objectives, research can be done by reviewing journals, conference papers and other research. from the literature review, current problems can be identified and proposed solutions can be made. from the analysis, then came up the solutions which in terms of conceptual design first then goes into detailed designs. the figure 2. hull crawler by qinetiq north america. (jansen, g.,2013) 2.0 methodology for the development and modelling of unmanned underwater remotely operated crawler (roc) for monitoring application, it all starts with a project plan. every detail of the project must be pointed out in term of designs, costs, materials selections, components selections and prototype testing and assembly process. for this project, it is divided into two parts; software and hardware. the first objective which is to design the remotely operated crawler. the design uses the cad software (solidwork) and simulations need to be done for different designs and selected with the best design as shown in figure 3. all issn: 2180-1053 vol. 7 no. 2 july december 2015 journal of mechanical engineering and technology 44 designs have a different chassis design while the wheels, movement mechanisms and controller remain the same. then, development of the prototype in terms of hardware. figure 2. hull crawler by qinetiq north america. (jansen, g.,2013) 2.0 methodology for the development and modelling of unmanned underwater remotely operated crawler (roc) for monitoring application, it all starts with a project plan. every detail of the project must be pointed out in term of designs, costs, materials selections, components selections and prototype testing and assembly process. for this project, it is divided into two parts; software and hardware. the first objective which is to design the remotely operated crawler. the design uses the cad software (solidwork) and simulations need to be done for different designs and selected with the best design as shown in figure 3. all designs have a different chassis design while the wheels, movement mechanisms and controller remain the same. then, development of the prototype in terms of hardware. figure 3. assembly design of the crawler. by determining the objectives, research can be done by reviewing journals, conference papers and other research. from the literature review, current problems can be identified and proposed solutions can be made. from the analysis, then came up the solutions which in terms of conceptual design first then goes into detailed designs. the figure 3. assembly design of the crawler. by determining the objectives, research can be done by reviewing journals, conference papers and other research. from the literature review, current problems can be identified and proposed solutions can be made. from the analysis, then came up the solutions which in terms of conceptual design first then goes into detailed designs. the best designs that fulfil every requirement should be chosen to solve current problems. each design will be simulated in order to identify design’s weakness and strength. then, the development of prototype can be done. the prototype must be tested in the lab and even possible field test. troubleshooting the prototype will help in determining the error or problems and improves the prototype design. figure 3 shows the 3d assembly drawing of the crawler using solidworks. the idea of designing such crawler came out from the mechanism of a tank. with this type of wheels, the crawler can crawl on any surface of the terrain. this will help improving the maneuverability of the crawler. the wheels used sprocket instead of belting and gear. this will reduce the cost in the fabrication process since sprocket is a standard part and available in the market. a little adjustment needed so that the sprocket will fit with the crawler. figure 4 shows the process flowchart of this project. issn: 2180-1053 vol. 7 no. 2 july december 2015 development of an unmanned underwater remotely operated crawler (roc) for monitoring application 45 4 best designs that fulfil every requirement should be chosen to solve current problems. each design will be simulated in order to identify design’s weakness and strength. then, the development of prototype can be done. the prototype must be tested in the lab and even possible field test. troubleshooting the prototype will help in determining the error or problems and improves the prototype design. figure 3 shows the 3d assembly drawing of the crawler using solidworks. the idea of designing such crawler came out from the mechanism of a tank. with this type of wheels, the crawler can crawl on any surface of the terrain. this will help improving the maneuverability of the crawler. the wheels used sprocket instead of belting and gear. this will reduce the cost in the fabrication process since sprocket is a standard part and available in the market. a little adjustment needed so that the sprocket will fit with the crawler. figure 4 shows the process flowchart of this project. figure 4. process flowchart. 3.0 results and discussion the remotely operated underwater crawler is an unmanned type vehicle that works underwater seabed. the designs of the crawler are shown in table 1. all the designs have the same dimensions, type of wheel used, motor, gears configuration and control. figure 4. process flowchart. 3.0 results and discussion the remotely operated underwater crawler is an unmanned type vehicle that works underwater seabed. the designs of the crawler are shown in table 1. all the designs have the same dimensions, type of wheel used, motor, gears configuration and control. table 1. the specification of the roc. table 1. the specification of the roc. items dimensions length height of the chassis width height chassis to the ground type of wheels gear ratio motor type material weight : 450 mm : 100 mm : 297.6 mm : 30mm : track or chain type wheels : 1:1 (use sprocket and chain) : dc geared motor : stainless steel : 9.8 kg + 7 kg weighter figure 5. the prototype of roc inside view. the chassis of the roc is made of stainless steel as shown in figure 5. the other components of the roc are made of steel and also aluminum. the shaft for motors are made of aluminum and the wheels are steel. in order to avoid corrosion, all parts made of steel will be painted later. inside each sprocket, waterproof bearings are fixed inside the brackets. brackets will also prevent any water from getting through the chassis as shown in figure 6. issn: 2180-1053 vol. 7 no. 2 july december 2015 journal of mechanical engineering and technology 46 table 1. the specification of the roc. items dimensions length height of the chassis width height chassis to the ground type of wheels gear ratio motor type material weight : 450 mm : 100 mm : 297.6 mm : 30mm : track or chain type wheels : 1:1 (use sprocket and chain) : dc geared motor : stainless steel : 9.8 kg + 7 kg weighter figure 5. the prototype of roc inside view. the chassis of the roc is made of stainless steel as shown in figure 5. the other components of the roc are made of steel and also aluminum. the shaft for motors are made of aluminum and the wheels are steel. in order to avoid corrosion, all parts made of steel will be painted later. inside each sprocket, waterproof bearings are fixed inside the brackets. brackets will also prevent any water from getting through the chassis as shown in figure 6. figure 5. the prototype of roc inside view. the chassis of the roc is made of stainless steel as shown in figure 5. the other components of the roc are made of steel and also aluminum. the shaft for motors are made of aluminum and the wheels are steel. in order to avoid corrosion, all parts made of steel will be painted later. inside each sprocket, waterproof bearings are fixed inside the brackets. brackets will also prevent any water from getting through the chassis as shown in figure 6. 6 figure 6. the brackets hold bearings and waterproofing the chassis. 3.1 the control box the control box is actually a box containing all circuits, battery and ps2 controller as shown in figure 7 and figure 8. the box will protect the circuits and other electronics components from shock and provide an exclusive design. the cables for the crawler can be stored inside this box. control box is placed on the land while the crawler working underwater. figure 7. the control box. figure 8. the circuit of the controller. in this test, the crawler is sealed and all the connections are completely attached. later, the crawler was dipped into a water tank as in figure 9. before any further steps continue, motors are removed and the inside part of the chassis is clean and dry as shown in figure 9. after a few minutes dipped, the crawler is retrieved back and the chassis is opened. if there is no water or contamination inside the crawler, thus, it is figure 6. the brackets hold bearings and waterproofing the chassis. issn: 2180-1053 vol. 7 no. 2 july december 2015 development of an unmanned underwater remotely operated crawler (roc) for monitoring application 47 3.1 the control box the control box is actually a box containing all circuits, battery and ps2 controller as shown in figure 7 and figure 8. the box will protect the circuits and other electronics components from shock and provide an exclusive design. the cables for the crawler can be stored inside this box. control box is placed on the land while the crawler working underwater. 6 figure 6. the brackets hold bearings and waterproofing the chassis. 3.1 the control box the control box is actually a box containing all circuits, battery and ps2 controller as shown in figure 7 and figure 8. the box will protect the circuits and other electronics components from shock and provide an exclusive design. the cables for the crawler can be stored inside this box. control box is placed on the land while the crawler working underwater. figure 7. the control box. figure 8. the circuit of the controller. in this test, the crawler is sealed and all the connections are completely attached. later, the crawler was dipped into a water tank as in figure 9. before any further steps continue, motors are removed and the inside part of the chassis is clean and dry as shown in figure 9. after a few minutes dipped, the crawler is retrieved back and the chassis is opened. if there is no water or contamination inside the crawler, thus, it is figure 7. the control box. figure 8. the circuit of the controller. in this test, the crawler is sealed and all the connections are completely attached. later, the crawler was dipped into a water tank as in figure 9. before any further steps continue, motors are removed and the inside part of the chassis is clean and dry as shown in figure 9. after a few minutes dipped, the crawler is retrieved back and the chassis is opened. if there is no water or contamination inside the crawler, thus, it is concluded that the crawler is waterproof. the result obtained is, the crawler is waterproof and good to go for underwater operations. concluded that the crawler is waterproof. the result obtained is, the crawler is waterproof and good to go for underwater operations. figure 9. chassis is submersed to identify any leakage. the waterproof test is done twice in order to confirm there is no leakage. since the first test is failed, second test is done. the first test shows a leakage to the body due to improper sealed. the water still can get through the body via the joint of the cover with the chassis. in the second test indicates there is no leakage since more proper sealant is applied. 3.2 control and maneuverability test (field test) this experiment is about testing the ability of the crawler to operate in any terrain. first, the crawler is tested on the land. there are three surfaces that been chosen for the crawler to operate which is on hard surface (cement), dirt and on the grass as shown in table 2. time taken for the crawler to complete 3 m distance is recorded as follows: table 2. the time taken for the crawler to crawl in a distance of 3 m. surfaces time taken (s) mean time taken (s) test 1 test 2 test 3 test 4 test 5 cement 56.23 56.35 56.18 56.44 56.87 56.41 dirt 57.67 57.73 57.46 57.80 58.45 57.82 grass 59.34 59.56 59.89 60.45 60.13 59.87 figure 9. chassis is submersed to identify any leakage. issn: 2180-1053 vol. 7 no. 2 july december 2015 journal of mechanical engineering and technology 48 the waterproof test is done twice in order to confirm there is no leakage. since the first test is failed, second test is done. the first test shows a leakage to the body due to improper sealed. the water still can get through the body via the joint of the cover with the chassis. in the second test indicates there is no leakage since more proper sealant is applied. 3.2 control and maneuverability test (field test) this experiment is about testing the ability of the crawler to operate in any terrain. first, the crawler is tested on the land. there are three surfaces that been chosen for the crawler to operate which is on hard surface (cement), dirt and on the grass as shown in table 2. time taken for the crawler to complete 3 m distance is recorded as follows: table 2. the time taken for the crawler to crawl in a distance of 3 m. concluded that the crawler is waterproof. the result obtained is, the crawler is waterproof and good to go for underwater operations. figure 9. chassis is submersed to identify any leakage. the waterproof test is done twice in order to confirm there is no leakage. since the first test is failed, second test is done. the first test shows a leakage to the body due to improper sealed. the water still can get through the body via the joint of the cover with the chassis. in the second test indicates there is no leakage since more proper sealant is applied. 3.2 control and maneuverability test (field test) this experiment is about testing the ability of the crawler to operate in any terrain. first, the crawler is tested on the land. there are three surfaces that been chosen for the crawler to operate which is on hard surface (cement), dirt and on the grass as shown in table 2. time taken for the crawler to complete 3 m distance is recorded as follows: table 2. the time taken for the crawler to crawl in a distance of 3 m. surfaces time taken (s) mean time taken (s) test 1 test 2 test 3 test 4 test 5 cement 56.23 56.35 56.18 56.44 56.87 56.41 dirt 57.67 57.73 57.46 57.80 58.45 57.82 grass 59.34 59.56 59.89 60.45 60.13 59.87 8 figure 10. comparison chart between surfaces against the time taken. figure 10 indicates that, the crawler moves slower on the grassy terrain compare to other conditions which is dirt and cemented surfaces. this is because, the crawler exerted more friction on the base of the crawler with the grass. besides that, grassy surface provides more uneven surface. it is a bumpy ride as we can describe. the cemented surface gives no friction to the base of the crawler. the only frictions come from the wheels spike to the surface. there is a difference in time taken for each test even though tested at the same terrain. this is because other external factor such as, power supplied by the battery is decreasing, surface interventions and the way the crawler has been controlled. thus, if this test is done underwater, the time taken will be much higher due to water resistance and the surfaces of the terrain. obstacles test is done to measure how height and identify the limit of the crawler. the first two tests are carried out on the land and the last test is in the tank filled with water. the time taken for the crawler to climb the obstacles of each height is recorded. wooden planks are used for this test. each plank is 0.5 cm thick. the maximum height to crawler can climb is 9.5 cm. the results as follows in table 3 and plotted in graph as shown in figure 11. table 3. table for the crawler to climb the wooden plank. test height (cm) remarks time taken (s) descriptions 1 0.5 x 1.35 2 1.0 x 2.80 3 1.5 x 4.20 4 2.0 x 5.43 5 2.5 x 6.23 6 3.0 x 7.12 7 3.5 x 8.34 8 4.0 x 9.51 9 4.5 x 10.43 10 5.0 x 11.35 11 5.5 x 13.54 12 6.0 x 15.76 13 6.5 x 16.48 14 7.0 x 18.02 slightly stuck figure 10. comparison chart between surfaces against the time taken. figure 10 indicates that, the crawler moves slower on the grassy terrain compare to other conditions which is dirt and cemented surfaces. this is because, the crawler exerted more friction on the base of the crawler with the grass. besides that, grassy surface provides more uneven surface. it is a bumpy ride as we can describe. the cemented surface gives no friction to the base of the crawler. the only frictions come from the wheels spike to the surface. there is a difference in time taken for each test even though tested at the same terrain. this is because other external factor such as, power supplied by the battery is decreasing, issn: 2180-1053 vol. 7 no. 2 july december 2015 development of an unmanned underwater remotely operated crawler (roc) for monitoring application 49 surface interventions and the way the crawler has been controlled. thus, if this test is done underwater, the time taken will be much higher due to water resistance and the surfaces of the terrain. obstacles test is done to measure how height and identify the limit of the crawler. the first two tests are carried out on the land and the last test is in the tank filled with water. the time taken for the crawler to climb the obstacles of each height is recorded. wooden planks are used for this test. each plank is 0.5 cm thick. the maximum height to crawler can climb is 9.5 cm. the results as follows in table 3 and plotted in graph as shown in figure 11. table 3. table for the crawler to climb the wooden plank. 8 figure 10. comparison chart between surfaces against the time taken. figure 10 indicates that, the crawler moves slower on the grassy terrain compare to other conditions which is dirt and cemented surfaces. this is because, the crawler exerted more friction on the base of the crawler with the grass. besides that, grassy surface provides more uneven surface. it is a bumpy ride as we can describe. the cemented surface gives no friction to the base of the crawler. the only frictions come from the wheels spike to the surface. there is a difference in time taken for each test even though tested at the same terrain. this is because other external factor such as, power supplied by the battery is decreasing, surface interventions and the way the crawler has been controlled. thus, if this test is done underwater, the time taken will be much higher due to water resistance and the surfaces of the terrain. obstacles test is done to measure how height and identify the limit of the crawler. the first two tests are carried out on the land and the last test is in the tank filled with water. the time taken for the crawler to climb the obstacles of each height is recorded. wooden planks are used for this test. each plank is 0.5 cm thick. the maximum height to crawler can climb is 9.5 cm. the results as follows in table 3 and plotted in graph as shown in figure 11. table 3. table for the crawler to climb the wooden plank. test height (cm) remarks time taken (s) descriptions 1 0.5 x 1.35 2 1.0 x 2.80 3 1.5 x 4.20 4 2.0 x 5.43 5 2.5 x 6.23 6 3.0 x 7.12 7 3.5 x 8.34 8 4.0 x 9.51 9 4.5 x 10.43 10 5.0 x 11.35 11 5.5 x 13.54 12 6.0 x 15.76 13 6.5 x 16.48 14 7.0 x 18.02 slightly stuck 15 7.5 x 19.79 slightly stuck 16 8.0 x 21.78 slightly stuck 17 8.5 x 23.89 slightly stuck 18 9.0 x 25.87 slightly stuck 19 9.5 x 27.78 stuck but can climb 20 10.0 o the base stuck to the obstacles 21 10.5 o the base stuck to the obstacles figure 11. the ability of the crawler to climb chart. the time taken for the crawler to climb the obstacles of the height of 9.5 is recorded. wooden platform are used for this test. the results as follows in table 4. table 4. table for the crawler to crawl over wooden platform. test height (cm) remarks time taken (s) descriptions 1 9.5 x 25.28 able to climb 2 9.5 x 24.79 able to climb 3 9.5 x 25.44 able to climb 4 9.5 x 25.56 able to climb issn: 2180-1053 vol. 7 no. 2 july december 2015 journal of mechanical engineering and technology 50 15 7.5 x 19.79 slightly stuck 16 8.0 x 21.78 slightly stuck 17 8.5 x 23.89 slightly stuck 18 9.0 x 25.87 slightly stuck 19 9.5 x 27.78 stuck but can climb 20 10.0 o the base stuck to the obstacles 21 10.5 o the base stuck to the obstacles figure 11. the ability of the crawler to climb chart. the time taken for the crawler to climb the obstacles of the height of 9.5 is recorded. wooden platform are used for this test. the results as follows in table 4. table 4. table for the crawler to crawl over wooden platform. test height (cm) remarks time taken (s) descriptions 1 9.5 x 25.28 able to climb 2 9.5 x 24.79 able to climb 3 9.5 x 25.44 able to climb 4 9.5 x 25.56 able to climb figure 11. the ability of the crawler to climb chart. the time taken for the crawler to climb the obstacles of the height of 9.5 is recorded. wooden platform are used for this test. the results as follows in table 4. table 4. table for the crawler to crawl over wooden platform. 15 7.5 x 19.79 slightly stuck 16 8.0 x 21.78 slightly stuck 17 8.5 x 23.89 slightly stuck 18 9.0 x 25.87 slightly stuck 19 9.5 x 27.78 stuck but can climb 20 10.0 o the base stuck to the obstacles 21 10.5 o the base stuck to the obstacles figure 11. the ability of the crawler to climb chart. the time taken for the crawler to climb the obstacles of the height of 9.5 is recorded. wooden platform are used for this test. the results as follows in table 4. table 4. table for the crawler to crawl over wooden platform. test height (cm) remarks time taken (s) descriptions 1 9.5 x 25.28 able to climb 2 9.5 x 24.79 able to climb 3 9.5 x 25.44 able to climb 4 9.5 x 25.56 able to climb 10 figure 12. the crawler climbed the 9.5 cm obstacles. the crawler can climb up to 9.5 cm obstacles and it is the maximum height it can climb as shown in figure 12. this is because the bottom base of the crawler stuck to the edge of the obstacles. spikes on chain help the crawler to have greater tractions and pull the crawler up. this condition can be overcome by having larger sprocket that tied to the chain or weld longer spike to the chain. 3.3 control and maneuverability test (underwater field test) the first objective is to determine the time taken for the crawler to crawl underwater with the distance covered for 1m. this test is done in water tank with the depth of 0.9m. the following table 5 is the result for the test. as we can evaluate, the time taken is high for the crawler to crawl and slow. this is because of the design of the wheels which the chain type wheels, weight of the crawler and also resistance as shown in figure 13. table 5. 1m underwater test. test time taken (s) 1 16.23 2 16.31 3 16.28 4 16.34 average 16.29 figure 12. the crawler climbed the 9.5 cm obstacles. issn: 2180-1053 vol. 7 no. 2 july december 2015 development of an unmanned underwater remotely operated crawler (roc) for monitoring application 51 the crawler can climb up to 9.5 cm obstacles and it is the maximum height it can climb as shown in figure 12. this is because the bottom base of the crawler stuck to the edge of the obstacles. spikes on chain help the crawler to have greater tractions and pull the crawler up. this condition can be overcome by having larger sprocket that tied to the chain or weld longer spike to the chain. 3.3 control and maneuverability test (underwater field test) the first objective is to determine the time taken for the crawler to crawl underwater with the distance covered for 1m. this test is done in water tank with the depth of 0.9m. the following table 5 is the result for the test. as we can evaluate, the time taken is high for the crawler to crawl and slow. this is because of the design of the wheels which the chain type wheels, weight of the crawler and also resistance as shown in figure 13. table 5. 1m underwater test. 10 figure 12. the crawler climbed the 9.5 cm obstacles. the crawler can climb up to 9.5 cm obstacles and it is the maximum height it can climb as shown in figure 12. this is because the bottom base of the crawler stuck to the edge of the obstacles. spikes on chain help the crawler to have greater tractions and pull the crawler up. this condition can be overcome by having larger sprocket that tied to the chain or weld longer spike to the chain. 3.3 control and maneuverability test (underwater field test) the first objective is to determine the time taken for the crawler to crawl underwater with the distance covered for 1m. this test is done in water tank with the depth of 0.9m. the following table 5 is the result for the test. as we can evaluate, the time taken is high for the crawler to crawl and slow. this is because of the design of the wheels which the chain type wheels, weight of the crawler and also resistance as shown in figure 13. table 5. 1m underwater test. test time taken (s) 1 16.23 2 16.31 3 16.28 4 16.34 average 16.29 figure 13. underwater time result. the last obstacles test is carried out underwater. iron column and brick are used for this test. the crawler is submerged and controlled to climb obstacles in the tank which places in a line with the distance of 1m. time taken for the crawler to climb the obstacles is recorded in table 6. all the obstacles can be climbed by the crawler. the hardest obstacles for the crawler to climb is the brick. the results as follows: table 6. table for the crawler to crawl over underwater obstacles. test remarks time taken (s) descriptions 1 x 34.67 able to climb 2 x 42.34 able to climb 3 x 45.44 able to climb 4 x 44.65 able to climb 4.0 conclusions designing an unmanned underwater vehicle (uuv) gives a lot challenge. the first objective of this project is; to design an unmanned underwater remotely operated crawler (roc) using cad. solidwork is used as the software and platform in designing the crawler. several simulation test is done using the application available in the software which is the simuationxpress. based on the application, the chassis design of the crawler is tested with force of 10n and a pressure of 50000 to imitate the condition of 50m underwater environment and above. the test included the stress, displacement, deformation and factor of safety test. all simulations shows that the design of the chassis plays an important role for the crawler to withstand the underwater environment. as a conclusion, the simulation test help in decision making process. it provide details about the material used, sustainability and simulation when the design is tested in real situation. every details must be precise since the roc will operate underwater. from the design process to fabrication, the roc is inspected and developed properly. the figure 13. underwater time result. issn: 2180-1053 vol. 7 no. 2 july december 2015 journal of mechanical engineering and technology 52 the last obstacles test is carried out underwater. iron column and brick are used for this test. the crawler is submerged and controlled to climb obstacles in the tank which places in a line with the distance of 1m. time taken for the crawler to climb the obstacles is recorded in table 6. all the obstacles can be climbed by the crawler. the hardest obstacles for the crawler to climb is the brick. the results as follows: table 6. table for the crawler to crawl over underwater obstacles. figure 13. underwater time result. the last obstacles test is carried out underwater. iron column and brick are used for this test. the crawler is submerged and controlled to climb obstacles in the tank which places in a line with the distance of 1m. time taken for the crawler to climb the obstacles is recorded in table 6. all the obstacles can be climbed by the crawler. the hardest obstacles for the crawler to climb is the brick. the results as follows: table 6. table for the crawler to crawl over underwater obstacles. test remarks time taken (s) descriptions 1 x 34.67 able to climb 2 x 42.34 able to climb 3 x 45.44 able to climb 4 x 44.65 able to climb 4.0 conclusions designing an unmanned underwater vehicle (uuv) gives a lot challenge. the first objective of this project is; to design an unmanned underwater remotely operated crawler (roc) using cad. solidwork is used as the software and platform in designing the crawler. several simulation test is done using the application available in the software which is the simuationxpress. based on the application, the chassis design of the crawler is tested with force of 10n and a pressure of 50000 to imitate the condition of 50m underwater environment and above. the test included the stress, displacement, deformation and factor of safety test. all simulations shows that the design of the chassis plays an important role for the crawler to withstand the underwater environment. as a conclusion, the simulation test help in decision making process. it provide details about the material used, sustainability and simulation when the design is tested in real situation. every details must be precise since the roc will operate underwater. from the design process to fabrication, the roc is inspected and developed properly. the 4.0 conclusions designing an unmanned underwater vehicle (uuv) gives a lot challenge. the first objective of this project is; to design an unmanned underwater remotely operated crawler (roc) using cad. solidwork is used as the software and platform in designing the crawler. several simulation test is done using the application available in the software which is the simuationxpress. based on the application, the chassis design of the crawler is tested with force of 10n and a pressure of 50000 n/m2 to imitate the condition of 50m underwater environment and above. the test included the stress, displacement, deformation and factor of safety test. all simulations shows that the design of the chassis plays an important role for the crawler to withstand the underwater environment. as a conclusion, the simulation test help in decision making process. it provide details about the material used, sustainability and simulation when the design is tested in real situation. every details must be precise since the roc will operate underwater. from the design process to fabrication, the roc is inspected and developed properly. the fabrication process that has been done to develop the roc are bending, welding and modify the available components to suit the application of the crawler. sprocket for example is available in the market. but for it can be used in the roc, some adjustment have been done to them so that it can fit to the shaft that linked with motors. motors selection also important. the weight of the crawler is determined and suitable motor is chosen. the torque of the motor is 1960 nm which can carry the weight of the crawler. waterproofing the crawler also gave a challenge. the body of the crawler is sealed with sealant, chassis is welded perfectly, and components are issn: 2180-1053 vol. 7 no. 2 july december 2015 development of an unmanned underwater remotely operated crawler (roc) for monitoring application 53 designed to fit the chassis so that the body of the roc is waterproof. this is important to protect motors inside it. the total weight of the crawler once it completed is 9.8 kg. after the completion of the roc, analysis the maneuverability of the roc underwater and on land are tested. tests are set up to identify the limits and capability of the crawler to operate. in this test, obstacles set up for the roc to climb. this test is carried out on land and also underwater. the roc is capable of climbing an obstacles of the maximum height of 9.5 cm. this is because of the design of the chassis and wheels. other test is carried out is buoyancy test. this test is crucial since crawler need to sink since it will operate on the seabed not floating. from the test, weight of 7kg need to be added to the crawler. less than 7kg will cause the crawler to have a slightly positive buoyancy. the roc operates as expected by theory even though there is unexpected problems emerged. one of the problems is the body of the roc is hollow. hence, there is air pocket inside it. in theory, 10 kg is quite heavy and the crawler will sink but it won’t. weight need to be added so that the roc will sink to the bottom. the design of the roc is based on tank and have a slot modular design. more components can be added and the design can be improvised for future work. acknowledgements we wish to express our gratitude to honorable university, universiti teknikal malaysia melaka (utem) especially for underwater technology research group (uterg), centre of research and innovation management (crim) and to the faculty of electrical engineering from utem to give the financial as well as moral support for complete this project successfully. issn: 2180-1053 vol. 7 no. 2 july december 2015 journal of mechanical engineering and technology 54 references ali, fara ashikin and abdul azis, fadilah and mohd aras, mohd shahrieel and muhammad nur , othman and shahrum shah, abdullah (2013) design a magnetic contactless thruster of unmanned underwater vehicle. international review of mechanical engineering, 7(7). 1413-1420. c.r. deepak, m. a. shajahan, m. a. atmanand, k. annamalai, r. jeyamani, m. ravindran, e. schulte, r. handschuh, j.panthel, h. grebe, w.schwarz. developmental tests on the underwater mining system using flexible riser concept. isope-oms-01-mt-02. how much oil is consumed in the united states? (2014, may 13). retrieved october 29, 2014, from http://www.eia.gov/tools/faqs/faq. cfm?id=33&t=6 hyakudome, t. (2011). design of autonomous underwater vehicle. international journal of advanced robotic systems, 8(1), 131-139. jansen, g. (2013). by land, sea and air, unmanned vehicles focus on new ways to conduct pipeline inspections. auvsi, 14-17. m. moonesun, m. javadi, p. charmdooz. (2012, dec, 5). evaluation of submarine model test in towing tank and comparison with cfd and experimental formulas for fully submerged resistance. international conference on underwater system technology: theory and applications 2012 (usys’12), shah alam, malaysia. m welling, d., & b. edwards, d. (2005). multiple autonomous underwater crawler control for mine reacquisition, (imece2005-81716). mohd aras, mohd shahrieel and jaafar, hazriq izzuan and anuar , mohamed kassim (2013a) tuning process of single input fuzzy logic controller based on linear control surface approximation method for depth control of underwater remotely operated vehicle. journal of engineering and applied sciences, 8(6). 208-214. mohd aras, mohd shahrieel and abdul rahman, ahmad fadzli nizam (2013b) analysis of an improved single input fuzzy logic controller designed for depth control using microbox 2000/2000c interfacing. international review of automatic control, 6(6). 728-733. mohd aras, mohd shahrieel and mohd shah, hairol nizam and ab rashid, mohd zamzuri (2013c) robust control of adaptive single input fuzzy logic controller for unmanned underwater vehicle. journal of theoretical and applied information technology, 57 (3). 372-379. mohd aras, mohd shahrieel and jaafar, hazriq izzuan and razilah , abdul rahim and ahmad , arfah (2013d) a comparison study between two algorithms particle swarm optimization for depth control of underwater remotely operated vehicle. international review on modelling & simulations, 6 (5). 1-10. issn: 2180-1053 vol. 7 no. 2 july december 2015 development of an unmanned underwater remotely operated crawler (roc) for monitoring application 55 mohd aras, mohd shahrieel and ab rashid, mohd zamzuri and azhan , ab. rahman (2013e) development and modeling of unmanned underwater remotely operated vehicle using system identification for depth control. journal of theoretical and applied information technology, 56 (1). 136-145. oil and gas accidents during the offshore exploration and production (oil and gas accidents during the offshore exploration and production) retrieved october 29, 2014, from http://www.offshore-environment. com/accidents.html short-term energy and winter fuels outlook. (2014, october 7). retrieved october 29, 2014, from http://www.eia.gov/forecasts/steo/report/ global_oil.cfm underwater exploration. retrieved october 22, 2014, from http://www. nationalgeographic.com/125/timelines/underwater-exploration/ underwater exploration history, oceanography, instrumentation, diving tools and techniques, deep-sea submersible vessels, key findings in underwater exploration deep-sea pioneers (jrank articles). retrieved october 22, 2014, from http://science.jrank.org/pages/7100/ underwater-exploration.html u.s. energy information administration eia independent statistics and analysis (short-term energy outlook) retrieved october 29, 2014, from http://www.eia.gov/forecasts/steo/report/global_oil.cfm wood, s., harris, w., ismail, t., malone, j., nanney, m., ojeda, j., vandedrinck, s. (2013). hybrid robot crawler / flyer for use in underwater archaeology. 1-11. paper4-le-v9n1_pages-67-86 issn: 2180-1053 vol. 9 no.1 january – june 2017 67 an integrated fuzzy analytical hierarchical process and fuzzy grey relational analytical model with vikor for maintenance system appraisal d.e. ighravwe1,2 and s.a. oke1* 1department of mechanical engineering, university of lagos, lagos, nigeria 2department of mechanical engineering, ladoke akintola university of technology, ogbomoso, nigeria abstract the motivation for this research lies in the understanding that the evaluation of a maintenance department for a manufacturing organization strongly depends on a wide range of uncertainties and vague parameters. consequently, utilising intuition may not be technically correct and downplays on the supposed results for the right management decisions on maintenance. the need for a new method to correct this anomaly is very much pressing to enhance the performance of maintenance systems. in this paper, the fusion of fuzzy analytical hierarchy process with fuzzy grey relational analysis as well as vikor is presented. a measuring instrument, questionnaire, for evaluating the performance of maintenance systems was developed and administered in four companies. using the pair-wise comparisons of criteria relevant to systems reliability, profitability, lead-time, system safety, production cost and manufacturing goals, the crisp values for the major components were generated. computation of the grey relational grade, best and worst values, utility regret measure and vikor index, and finally the ranking of the maintenance system were made. the approach is feasible in maintenance system evaluation. the unique and innovative approach that established a link between maintenance system’s goals and variables when dealing with maintenance system appraisal is the main novelty of the work. an additional novelty not reported earlier in literature is the consideration of human attributes and environments in an integrated manner. this study contributes a significant approach for correctly evaluating the technical aspects of the maintenance system. keywords: maintenance performance criteria; membership functions; fuzzy grey relational analysis; vikor; fuzzy analytical hierarchy process 1.0 introduction the maintenance system in a manufacturing organisation is top among the value-adding units of the industrial enterprise. the function is responsible for making the repaired equipment and facilities safe and ensuring minimum breakdowns of the same. over the past several years, the maintenance function has been evaluated for its quality of service in terms of its output and the progress of the function determined by comparing the * corresponding author e-mail: sa_oke@yahoo.com journal of mechanical engineering and technology 68 issn: 2180-1053 vol. 9 no.1 january – june 2017 outputs with the input measures of labour, material, equipment hours, capital and energy (baluch et al., 2010). the traditional approach to evaluating maintenance department hugely leans on a variety of parameters but tracking the environmental, economic, financial, machine performance and human attribute consequences in analysing maintenance system is often a great challenge (plantweb, 2003; simoes et al., 2011; muchiri et al., 2011). yet maintenance practices cannot improve without the adequate coverage of these main parametric determinants. novel and insightful theories and applications that would aid sound decision making in maintenance are a necessary requirement for progress. it is however unfortunate that through a detailed literature review and an understanding of the accomplishments of scholars in the maintenance performance appraisal area, no work seems to have engaged, in a very detailed manner, the engineering concepts as well as technology while also considering the managerial, economic and environmental perspectives in which the industry thrives. as a response to this literature gap and challenge, the present paper focuses on the analysis and modelling of the important maintenance parameters in the evaluation of the maintenance system in a manufacturing system. it is interesting to identify the engineering aspects of the system from perspectives of measures of availability, meantime-to-failure, mean-time-to-restore, mean-downtime as well as the overall equipment effectiveness. so, the engineering aspects referred to as the machine performance indicators in this work have been taken as an important component in the modelling and analysis of maintenance performance. as advocated earlier, if we are to consider the technological aspect, then the examination of factors such as vibration control, temperature control and lighting are of principal importance. recall that it was mentioned earlier that a good evaluation system must also contain the managerial factors. such a consideration is expected to have factors reflecting labour-management relations, communication and cooperation among others. economic aspects include costs of spare parts, training, bonuses, worker’s salaries and compensation. the environmental perspective which reflects sustainable practices include noise control and cleanliness. from the above analysis, it has been established that practical and sound decision based on managerial, economic, engineering, technology, managerial and environmental perspectives is a must towards attaining a strong theoretical base that works in practice.the objective of the current paper is to propose a conceptual framework for maintenance systems appraisal based on maintenance environments (physical and organisational), machine performance indicator, maintenance cost and human attributes. the proposed framework is an integrated fuzzy analytical hierarchy process (fahp), fuzzy grey relational analysis (fgra) and vikor approach. fahp is employed to evaluate the weights for the above mentioned five maintenance criteria for the evaluation process based on manufacturing system’s goals. each criterion of the principal components is aggregated into a single performance index using the fgra approach. the ranking of maintenance systems is based on the vikor technique. in the remaining parts of this paper, the literature review is elaborated in the second section. in the third section, the methodological aspect of the work is discussed. the fourth section showcases the application of the model, carried out in four companies. the first company produces sheets, coils and circles and it is known as a rolling mill. the second company manufactures agricultural sacks (sack manufacturing) while the third company produces household utensils (hollowware manufacturing). the fourth case study is engaged in the production of noodles (food company). this section also contains a discussion of research results while the fifth section presented the conclusion of the study. an integrated fuzzy analytical hierarchical process and fuzzy grey relational analytical model with vikor for maintenance system appraisal issn: 2180-1053 vol. 9 no.1 january – june 2017 69 2.0 literature review several theories have been advanced in literature to explain maintenance performance characteristics. certainly, the literature has covered the different types of such theories; the review undertaken here will only focus on three main themes that repeatedly occur throughout the review literature. the themes are namely, the measurement of maintenance profitability, the improvement of maintenance profitability, the need to measure maintenance productivity and the importance of maintenance quality and its associated parameters. although the literature explains the above themes in a diversity of contexts, the current research mainly direct attention to their applications in manufacturing systems. furthermore, a wide range of authorities has contributed to the development of maintenance performance literature and the coverage of literature is intensive. however, the direction of focus of the current study shall be on those that consistently contribute to literature for the past several years. the review of literature in the current study is approached first by identifying the major theories in research on maintenance profitability, maintenance productivity, maintenance quality and generally on maintenance performance. the next stage of research brought out notable contributors in the field, who works were inspirational to the development of the field. then the major theories in the field are reviewed. the final phase of the literature review identified the gaps in the literature relevant to maintenance performance. in literature, a number of studies on the performance of maintenance systems have been made. these investigations are further broken down into more specific issues but treating the various criteria of performance as individual topics of interest. recall that performance has been noted to contain criteria such as productivity, profitability, innovation, quality and quality of working life. out of all these criteria, significant reporting could only be found for maintenance productivity, maintenance profitability and maintenance quality while reports on maintenance innovation and maintenance quality of working life are almost non-existent. in maintenance profitability, the common themes of research are that (i) profitability can be measured; and (ii) can be improved (oke, 2005; maletic et al. 2014). oke (2005) contributed a mathematical approach to measuring maintenance profitability. the author proved the utility of the approach in a case study. it was argued that a change in the perception of the maintenance function from a cost centre to a profit centre, wherein profit could be made by the function was the arguement. maletic et al. (2014) detailed out the function of maintenance with respect to enhancing company’s profitability using empirical data from a textile mill. it was concluded that practices in maintenance associated with condition-based maintenance method had the greatest potential for improvement. oke et al. (2008) viewed maintenance from a value-adding perspective, using the concept of charging the services by maintenance to production in prices. a mathematical framework that describes maintenance profitability while considering inflation was contributed by the authors. from the three studies reviewed above, oke (2005), oke et al. (2008) and maletic et al. (2014), the two themes of the drive for the measurement and improvement of performance were adopted in the current study. a group of maintenance performance appraisal studies focused on maintenance productivity evaluation, wherein the output to input ratio from the conceptual perspective was taken into account. the main themes of the studies are that (i) maintenance productivity could be associated with safety, quality and reliability; (ii) maintenance productivity can be improved. researchers strongly believed in the linkages of safety, productivity, maintenance, quality and reliability as evident in the journal of mechanical engineering and technology 70 issn: 2180-1053 vol. 9 no.1 january – june 2017 study by narayan (2012) in which the associations among safety, quality, reliability and productivity were established. the conclusion was that integrating the technological as well as the behavioural aspects of humans presents a holistic viewpoint of maintenance. further in an associative effort, khan and darrab (2010) related productivity with quality as well as maintenance. it was concluded that the developed approach predicted the most acceptable productivity outcomes in association of maintenance with quality indices from practical data. elangovan et al. (2007) established a linkage between quality and productivity enhancement of maintenance executive decisions. the conclusion from the report was that it is feasible to link quality and productivity in maintenance using data collected from practical experience. this is however consistent in view with that of earlier researches on concept integration in maintenance. still on the association of maintenance with other concepts, abdul-raouf (2004) related productivity and safety maintenance, claiming that they enhance maintenance in terms of performance. they outlined the tasks that aided the elimination of accidents as well as removing potential interruption causes. raouf’s (1994) contribution is similar to the theme in current literature on profitability, whereby productivity was argued as a candidate for improvement. now, drawing from the themes of researchers’ arguments on productivity, we add the idea of integrating issues and not treating measures in compartments with each item being accounted for in stand-alone perspectives. rather, a holistic approach has been adopted in the current paper. it is worth noting that majority of appraisal studies on maintenance are captioned under the general term of maintenance performance instead of maintenance productivity, maintenance profitability and maintenance quality investigations. so, the next set of review relates to maintenance performance appraisal studies (de groote, 1995; parida and kumar, 2006). in this literature review, the question answered here relates to what has been documented in the maintenance performance field. drawing from the works of major authors that have contributed in a significant manner to developing the maintenance performance measurement field, kumar and co-workers, labib and co-researchers, parida and co-workers as well as pintellon and co-researchers may be mentioned. most attention has been directed to strategic issues, tools, models and the diverse applications such as mining and railway infrastructure. arising from the literature analysis is the gap that no reported studies have been documented in the nigerian environment. there have not been comprehensive reports in any form, worldwide on the applications of maintenance performance to rolling mills. the cases of bag manufacturing, household utensils and food products are missing. the springboard for the current research on maintenance performance is the performance measurement field, which majorly hinged on the criteria of productivity, profitability, innovation, quality and quality of working life. as interest in maintenance performance sprang up, researchers began to engage in the adoption of criteria to the maintenance field. the major theories in this area of research are related to productivity theory, theory concerning profitability and the quality theory. the general theory concerning performance is also well-documented in literature. associated in the performance theories are the theories on indicators, multicriteria, measurement and performance (parida and kumar, 2004; parida et al., 2005.; kumar and parida, 2006; parida, 2007. ahren and parida, 2009; parida and uday, 2009). however, most of these theories have not matured to incorporate artificial intelligence models. in maintenance performance models, the use of ahp, fuzzy logic and grey relational analysis in an integrated form has not been reported. an integrated fuzzy analytical hierarchical process and fuzzy grey relational analytical model with vikor for maintenance system appraisal issn: 2180-1053 vol. 9 no.1 january – june 2017 71 furthermore, muchiri et al. (2011) developed a framework for evaluating the performance of a maintenance system. their study reported that performance gaps in maintenance system could be identified based on information on maintenance cost and machine performance indicators. apart from machines and maintenance work attributes, sondalini (2016) suggested that human-factors should be considered when evaluating maintenance systems. they also reported that the desire of setting too high maintenance performance indicator should be avoided by decision makers. wu et al. (2012) proposed the use of fuzzy multi-criteria decision-making process for maintenance workforce performance analysis. they considered professionalism, teamwork, discipline and innovation as criteria for workforce evaluation. a case study of the proposed approach which integrates fuzzy analysis hierarchy process and vikor was used to demonstrate the applicable of the approach in an aircraft maintenance system. from the above highlighted issues, it becomes apparent that developing an appropriate maintenance system appraisal strongly hinges on the layout of a suitable measurement scheme, the development of a system with system enhancement in mind and a system that could be audited, taking into consideration system flexibility that permits quantifiable inputs and outputs of the system. in addition, despite the large volume of literature on maintenance performance evaluation (muchiri et al., 2011), the use of fuzzy logic in capturing vagueness of maintenance parameters has been sparsely reported in literature. in addition, sparse information has been documented on vikor (vlse kriterijumska optimizacija i kompromisno resenje) approach to maintenance system appraisal. the need to address this important knowledge gaps serves as the motivation for the current study. addressing this gap has implications for management decision making as proper evaluation of system is made and actions carried out will have direct and long-lasting impact on organisational survival. in view of the aforementioned issues, the maintenance system appraisal developed in this work has been made in the perspective of literature support, by considering all the issues raised as themes as well as appropriately filling the gap identified in the current paper. thus, the work is strongly oriented at applying the integrated fuzzy analytical hierarchy process and fuzzy grey relational analytical scheme while solidifying the integration with the vikor concept. in verifying the feasibility of the developed model, a questionnaire-oriented feedback method was employed and analyzed in four companies operating in the nigerian industrial environment. in addition, based on the above related works, the issues of human-factors and the appreciation of vagueness in maintenance performance indicators have been downplayed by researchers and industrial practitioners. also, most studies on the development maintenance performance framework do not consider the physical and organisational environments. furthermore, the use of fgra for maintenance performance indictors’ aggregation has not been reported in literature to the best of our understanding. consequently, the current study has considered these parameters in presenting the proposed framework. 3.0 methodology the identification of best practice in maintenance system provides means for performance gaps analysis. in order to identify performance gaps in a maintenance system, there is the need for maintenance system appraisal. based on the information journal of mechanical engineering and technology 72 issn: 2180-1053 vol. 9 no.1 january – june 2017 obtained from literature, this study proposes a conceptual framework for maintenance systems appraisal (table 1). the framework is based on the integration of fuzzy logic, ahp, gra and vikor (figure 1). brief descriptions on how each of the above mentioned tools in the proposed conceptual framework is presented as follows: table 1: factors and criteria for maintenance system appraisal criteria principal components physical environment (c1) noise control (x11) vibration control (x12) temperature control (x13) lighting (x14) cleanliness (x15) organisation’s environment (c2) cooperation (x21) communication (x22) labour-management relationships (x23) promotion rate (x24) retrenchment rate (x25) machine performance (c3) overall equipment effectiveness (x31) mean-downtime (x32) mean-time-to-restore (x33) mean-time-to-failure (x34) availability (x35) human attributes (c4) stress (x41) fatigue (x42) team work (x43) workers’ agility (x44) turnover rate (x45) responsiveness (x46) work pressure (x47) maintenance cost (c5) bonuses (x51) workers’ salaries (x52) compensation (x53) training (x54) spare parts (x55) in order to have a clear understanding of the methodology adopted in this paper, an outline for the proposed framework is presented as follows: step 0: decision-makers size given that the proposed framework is a multi-decision making framework, the number of decision-makers makers for its implementation is first determined. this serves as the initialisation of the proposed framework. an integrated fuzzy analytical hierarchical process and fuzzy grey relational analytical model with vikor for maintenance system appraisal issn: 2180-1053 vol. 9 no.1 january – june 2017 73 figure 1. a conceptual framework for maintenance system appraisal step 1: selection of manufacturing goals information on manufacturing goals may be different from one maintenance system to another. it is the responsibility of the decision-makers to select the most suitable manufacturing goals for their evaluation process. step 2: selection of maintenance system appraisal criteria the number of maintenance system appraisal criteria that will be used for maintenance system appraisal is dependent on the decision-makers. also, the number of principal components for a selected maintenance system appraisal criterion is a function of the decision-makers judgements. step 3: evaluation of maintenance system appraisal criteria weight in order to make the proposed model an easy-to-apply tool, the evaluation of the maintenance system appraisal criteria weights are expressed using linguistic terms. the information obtained is processed using a fahp. step 4: evalution of the impact of principal components on maintenance system since some of the principal components values can only be expressed using linguistic terms, the impact of principal components on maintenance system are evaluated using linguistic terms. step 5: aggregation of impact of principal components values the aggregate of impact of principal components values is carried out using fgra approach. this approach provides a means of using the desired direction of a principal component (cost-based or benefit-based criterion). the results from frga provide insights to the ranking of maintenance systems. select maintenance criteria for maintenance systems appraisal determine the weight of each criterion using fahp select principal components which constitute a maintenance criterion determine the crisp value of each factor using fuzzy logic approach rank the maintenance systems using vikor select manufacturing system goals combine the principal components using fgra technique journal of mechanical engineering and technology 74 issn: 2180-1053 vol. 9 no.1 january – june 2017 step 6: aggregation frga results the results from frga are aggregated using vikor. the outputs from vikor are used to determine the best ranked maintenance system using three criteria (utility, regret measure and vikor index). brief descriptions on how each of the above mentioned tools in the proposed conceptual framework is presented as follows: fuzzy-ahp fuzzy-ahp is a modified version of ahp for systems where information is presented in linguistic terms (saaty, 1990; chang, 1996). the weights for maintenance criteria are evaluated with respect to five manufacturing system goals. the manufacturing system goals are system reliability (g1), profitability (g2), production lead-time (g3), system safety (g4) and production cost (g5). the weights for the maintenance criteria determined based on a fahp approach (chang, 1996). in order to convert responses from decision makers into crisp values, triangular membership function is considered (table 2). table 2. linguistic variables and triangular membership fuzzy conversation scale linguistic variables triangular membership fuzzy conversation scale triangular membership fuzzy reciprocal scale just equal (1,1,1) (1,1,1) equally important (1/2,1,3/2) (2/3,1,2) weakly more important (1,3/2,2) (1/2,2/3,1) moderately more important (3/2,2,5/2) (2/5,1/2,2/3) strongly more important (2,5/2,3) (1/3,2/5,1/2) extremely more important (5/2,3,7/2) (2/7,1/3,2/5) the conversion of the triangular membership functions in table 2 for a multiresponses analysis into crisp values is achieved using equations (1) and (2).   1 2 3 1 1 1 1 2 3 , , , , k k k k k k k k k a a a a a a k       (1) 1 2 3 4 6 a a a a    (2) where k represents decision-maker. after the conversion of the fuzzy values for the principal components, standard ahp approach of weights determination is then applied. information on how to apply standard ahp is contained in saaty (1980). furthermore, the mathematics of ahp can be avoided using commercial software. an integrated fuzzy analytical hierarchical process and fuzzy grey relational analytical model with vikor for maintenance system appraisal issn: 2180-1053 vol. 9 no.1 january – june 2017 75 fuzzy-gra gra is a tool for aggregating the components of a factor into single value. this study considered fgra as a means for aggregating the principal components that constitute a maintenance criterion into a single-index because the responses from decision-makers are in linguistic terms. the responses from decision-makers linguistic terms are analysed using trapezoidal membership functions (equation 3, figure 2 and table 3). the aggregated value of trapezoidal membership function for multi-responses is obtained based on (equations 4 to 7).  1 2 3 4, , ,ij ij ij ij ijx x x x x (3)  1 1minij ijkx x (4) 2 2 1 1 k ij ijk k x x k    (5) 3 3 1 1 k ij ijk k x x k    (6)  4 4minij ijkx x (7) 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 vp p f g vg 1 0 μ figure 2. membership functions for gra analysis journal of mechanical engineering and technology 76 issn: 2180-1053 vol. 9 no.1 january – june 2017 table 3. linguistic variables and corresponding fuzzy number for gra analysis linguistic variables abbreviations fuzzy number very poor or very low vp (0.0, 0.1, 0.2, 0.3) poor or low p (0.2, 0.3, 0.4, 0.5) fair or moderate f (0.4, 0.5, 0.6, 0.7) good or high g (0.6, 0.7, 0.8, 0.9) very good or very high vg (0.8, 0.9, 1.0, 1.0) the normalisation of the principal components that are considered for the physical environment criterion is based on the higher-the-better criterion (equation 8). during organisation criterion normalisation, cooperation, communication, labour-management relationships and promotion rate are normalised based on a higher-the-better criterion. a lower-the-better criterion is used for retrenchment rate normalisation (equation 9). apart from mean-time-to-restore and mean downtime which is normalised using lowerthe-better criterion, other principal components for machine performance criterion are normalised based on higher-the-better criterion. stress, fatigue, turnover rate and work pressure are normalised based on lower-the-better criterion. higher-the-better criterion is used for responsiveness, workers’ agility and teamwork normalisation. the normalisation scheme for maintenance cost factors is based on lower-the-better criterion. 1 2 3 4 4 4 4 4 , , , , ij ij ij ij ij i ij ij ij ij x x x x c b x x x x           (8) 1 2 3 4 1 1 1 1 , , , , ij ij ij ij ij i ij ij ij ij x x x x c c x x x x           (9) where ij is the normalised values for factor i belonging to criterion j, ic represents factor i, a centroid scheme defuzzification scheme is used in this study (opricovic and tzeng, 2004). in girubha and vinodh (2012) study, centroid scheme defuzzification was expressed as equation (10).     2 2 4 3 2 1 4 3 2 1 4 3 2 1 1 1 3 3 ij ij ij ij ij ij ij ij ij ij ij ij ij x x x x x x x x x x x x x          (10) where xij represents the crisp value of factor i for maintenance criterion j. after the normalisation of the maintenance criteria, the next stage of fgra implementation is the determination of grey relation coefficient (hasani et al., 2012). the grey relation coefficient for a maintenance criterion is obtained using equation (11). an integrated fuzzy analytical hierarchical process and fuzzy grey relational analytical model with vikor for maintenance system appraisal issn: 2180-1053 vol. 9 no.1 january – june 2017 77     max maxmin ,       k k io i (11) min min min o i j i k x x         (12) max max max o i j i k x x         (13) where  kx o  represents the reference sequence,  kx i  represents the comparative sequence, and  is called identification coefficient and its values lies between (0,1). the grey relational grade for each maintenance criterion for a maintenance system is obtained using equation (14).   1 1 m sj i i f k m     (14) 4.0 vikor vikor methodology is based on the analysis of alternatives with respect to measures of closeness-to-ideal alternative under conflicting criteria. the multi-criteria measure is used for compromised ranking (opricovic and tzeng, 2004). vikor is used to rank the different maintenance systems using the results obtained from the fgra (equation 14). the implementation of vikor involves five basic steps. these steps are discussed as follows (opricovic and tzeng, 2004; wang and pang, 2011): step 1: evaluation of the worst and best maintenance criterion. the worst maintenance criterion is the minimum value maintenance criterion among the maintenance systems (equation 15), while the best maintenance criterion is the best value maintenance criterion among the maintenance systems (equation 16).  minj sjf f   (15)  maxj sjf f   (16) where fsj represents the gra value for criterion j obtained from maintenance system s, j f  represents maximum value for criterion j, and jf  represents minimum value for criterion j. step 2: computation of utility (s) and regret measure (r) for each alternative. the value of s is expressed as equation (17), while r value is expressed as equation (18).   1 n j j sj s j j j w f f s f f         (17) journal of mechanical engineering and technology 78 issn: 2180-1053 vol. 9 no.1 january – june 2017   max j j sj j j j j w f f r f f            (18) step 3: determination of vikor index for each maintenance system. the value of vikor index for maintenance system is based on utility and regret measure values as well as weight (v). the vikor index for a maintenance system is expressed as equation (19).     1s s s s s s s s s v s s v r r q s s r r              (19) where t r  represents min( s r ), t r  represents max( s r ), s s  represents max( s s ), and s s  represents min( s s ). step 4: ranking and selection of the maintenance systems using the values obtained from equations (17) to (19). the best maintenance system is the maintenance system with the lowest value for s, r and q. step 5: generation of compromise solution using the vikor indices that are obtained from equation (20). the conditions for compromise solution generation are given as follows: cc1: acceptable advantage     1 1 q a q a t     (20) where arepresents the second-ranked alternative based on vikor indices. cc2: acceptable stability the best alternative must also be the best alternative based on either utility or/and regret measure. when any of the above conditions is violated, a compromise solution is generated as follows: i. alternative a and a when the acceptable stability is violated. ii. alternative , m a a a  when the acceptable advantage is violated. alternative m a is determined based on equation (21).     1 1 m q a q a t    (21) an integrated fuzzy analytical hierarchical process and fuzzy grey relational analytical model with vikor for maintenance system appraisal issn: 2180-1053 vol. 9 no.1 january – june 2017 79 5.0 case study and discussion of results as earlier outlined in this work, a robust literature exists on performance appraisal concerning the maintenance system but the use of non-traditional optimization tools and methodologies involving the fusion of fuzzy logic with saaty’s ahp prioritization scheme as well as the fuzzfied gra have not been experimented with industrial data. there have not been any robust efforts and results from individual application perspective, to validate its worthiness. consequently, fuzzified ahp, fuzzified gra and vikor were subjected to industrial and practical analysis using the developed framework and the outcome of this research exercise are reported in the current section. the proposed conceptual framework was applied in four manufacturing systems. the first manufacturing system (s1) specialised in the production of packaged fast foods, while the second (s2) and third (s3) manufacturing systems specialised in the production of metallic products for household utensils and industrial purposes. the last manufacturing system (s4) specialises in the production of packaging materials for industrial and domestic purposes. information used for the implementation of the proposed conceptual framework was obtained using questionnaires. interviews were conducted with two main decision-makers in each of the maintenance systems. the participants were asked to give answers to five categories of questions. the categories were: (i) physical environment; (ii) organisation’s environment; (iii) machine performance; (iv) human attributes; and (v) maintenance cost. for each maintenance system, three decision-makers from a maintenance department were considered as respondents (maintenance manager and supervisors). during the computation of the importance of the maintenance criteria, it was observed that the importance of each of the maintenance criterion varies with respect to a selected manufacturing system goal (table 4 to 9). for instance, the most important criterion with respect to any selected manufacturing system goal varies from goal to goal (tables 4 to 9). in terms of system reliability, the most important criterion was human attributes. this was followed by machine performance criterion. the least important criterion under system reliability goal was organisational environment (table 4). table 4. pair-wise comparisons of maintenance criteria with respect to system reliability criteria c1 c2 c3 c4 c5 priorities c1 1.0000 3.9028 1.5522 0.9446 1.4591 0.1160 c2 1.1819 1.0000 1.4633 0.8155 1.0730 0.0044 c3 2.9583 3.6042 1.0000 0.9954 4.1250 0.2550 c4 4.5833 0.8155 4.6875 1.0000 1.0863 0.4403 c5 3.4167 1.0730 1.4113 3.9583 1.0000 0.1843 during the consideration of profitability goal, maintenance cost was the most important criterion. the importance of maintenance cost was slightly greater than machine performance importance (table 5). furthermore, the physical environment of the manufacturing companies was the least important criterion under profitability goal. however, physical environment was identified as the most important criterion under production lead-time goal. this was followed by organisational environment criterion (table 6). there was slight difference between maintenance cost and machine journal of mechanical engineering and technology 80 issn: 2180-1053 vol. 9 no.1 january – june 2017 performance criteria under production lead-time goal. in addition, human attributes criterion was the least important criterion under production lead-time goal (table 6). table 5. pair-wise comparisons of maintenance criteria with respect to profitability criteria c1 c2 c3 c4 c5 priorities c1 1.0000 3.7500 1.0099 0.9690 1.2383 0.0646 c2 1.0583 1.0000 1.3163 4.2500 0.8446 0.1057 c3 4.0833 4.3125 1.0000 4.3750 1.2571 0.3164 c4 4.4583 1.1871 1.4321 1.0000 2.0161 0.1333 c5 4.3125 4.6042 4.0625 3.3411 1.0000 0.3799 table 6. pair-wise comparisons of maintenance criteria with respect to production lead-time criteria c1 c2 c3 c4 c5 priorities c1 1.0000 1.8722 1.7472 4.1119 4.0000 0.2618 c2 3.7994 1.0000 3.7500 1.3472 3.3514 0.2489 c3 3.7889 1.2792 1.0000 3.7264 1.9716 0.1937 c4 1.5813 3.7889 1.6452 1.0000 0.6369 0.1024 c5 1.3472 1.6792 3.5494 3.9583 1.0000 0.1932 the results for the importance of the criteria under system safety showed that organisational environment was the most important criterion. this was followed by physical environment criterion (table 7). there was a slight difference between the physical environment and human attributes importance (table 7). machine performance was the least important criterion under a system safety goal (table 7). in term of production cost goal, maintenance cost was the most importance criterion (table 8). the importance of human attributes criterion under production cost was ranked second. human attributes criterion importance was slightly more than that of machine performance (table 8). there was a slight difference between the importance values of organisation and physical environments under production cost goal (table 8). table 7. pair-wise comparisons of maintenance criteria with respect to system safety criteria c1 c2 c3 c4 c5 priorities c1 1.0000 3.7028 3.2056 2.1107 4.5000 0.2710 c2 1.5482 1.0000 4.1250 0.9718 4.7500 0.3016 c3 1.8536 1.0946 1.0000 1.9925 2.9000 0.0377 c4 3.6869 4.5833 2.9222 1.0000 1.3927 0.2632 c5 1.0135 0.9690 1.9091 4.2361 1.0000 0.1265 table 8. pair-wise comparisons of maintenance criteria with respect to production cost criteria c1 c2 c3 c4 c5 priorities c1 1.0000 1.5480 4.1250 1.1530 0.9216 0.1425 c2 3.0625 1.0000 1.0821 3.7500 1.5127 0.1399 c3 1.3766 3.8333 1.0000 1.6744 3.8514 0.2139 c4 4.1875 1.2196 3.5389 1.0000 1.0863 0.2237 c5 3.4583 3.7111 1.6821 3.9583 1.0000 0.2801 an integrated fuzzy analytical hierarchical process and fuzzy grey relational analytical model with vikor for maintenance system appraisal issn: 2180-1053 vol. 9 no.1 january – june 2017 81 from the pair-wise comparison of the manufacturing goals, the most important goal was g2 (profitability), while g3 (production lead-time) was the least important goal. the difference between the importance of system safety and production cost was close. this study ranked system safety as second, while production cost was ranked third. table 9. pair-wise comparisons of manufacturing goals goals g1 g2 g3 g4 g5 priorities g1 1.0000 4.1250 1.5036 1.1504 1.7988 0.1570 g2 1.3044 1.0000 3.6250 4.3542 1.3516 0.2750 g3 3.3333 1.3688 1.0000 3.2292 1.8639 0.1269 g4 4.3125 0.8897 1.2833 1.0000 3.0778 0.2243 g5 3.4139 3.8333 2.8250 1.7183 1.0000 0.2168 based on the information in tables 4 to 9, the weight for the criteria were determined (table 10). the most important criterion for the manufacturing system goals was maintenance cost (c5). this was followed by human attributes criterion (c4), which had a weight value that was closed to maintenance cost. physical environment criterion (c1) was the least important criterion (table 10). there exists a slight difference between the organisation environment and machine performance criteria (table 10). the grey relational coefficients for the different principal components were generated by converting linguistic values that were obtained from the different maintenance systems into crisp values (table 11). table 10. criteria weights based on fahp cig1 cig2 cig3 cig4 cig5 total weights c1 0.0182 0.0178 0.0332 0.0608 0.0309 1.1738 0.1677 c2 0.0007 0.0291 0.0316 0.0676 0.0303 1.2348 0.1764 c3 0.0400 0.0870 0.0246 0.0085 0.0464 1.3501 0.1929 c4 0.0691 0.0367 0.0130 0.0590 0.0485 1.6135 0.2305 c5 0.0289 0.1045 0.0245 0.0284 0.0607 1.6278 0.2325 table 11. crisp values for the principal components s1 s2 s3 s4 x11 0.4352 0.8333 0.6574 0.6574 x11 0.6574 0.8333 0.6574 0.8333 x11 0.2130 0.8796 0.6574 0.4815 x11 0.7037 0.8796 0.8333 0.8796 x11 0.7917 0.9222 0.4333 0.7917 x21 0.6574 0.8796 0.6574 0.8796 x22 0.6574 0.8796 0.6574 0.8796 x23 0.2130 0.8796 0.6111 0.7037 x24 0.4667 0.4667 0.7000 0.7833 x25 0.9048 0.7857 0.5595 0.5595 x31 0.8452 1.0714 0.8452 1.0714 journal of mechanical engineering and technology 82 issn: 2180-1053 vol. 9 no.1 january – june 2017 x32 0.6574 0.8333 0.4352 0.6574 x33 0.6574 0.8796 0.6574 0.8796 x34 0.3056 0.6746 0.6574 0.8333 x35 0.6574 0.8796 0.8333 0.8796 x41 0.2738 0.7857 0.8452 0.5595 x42 0.3333 0.7857 0.8452 0.7857 x43 0.7857 1.0714 0.9048 1.1310 x44 1.1833 1.2667 0.8667 1.8444 x45 1.0714 1.0714 0.8452 1.3175 x46 0.8452 1.3175 0.8452 1.3175 x47 0.5595 0.9048 0.8452 1.1310 x51 0.3000 0.4667 0.7000 0.3833 x52 0.4667 0.7833 0.7833 0.3833 x53 0.4667 0.7833 0.4667 0.3833 x54 0.7857 0.5000 0.2143 0.5000 x55 0.5595 0.9048 0.5000 0.8452 based on the results in table 12, s1 was the worst ranked maintenance system for all the maintenance system appraisal criteria. in terms of physical and organisation environments criteria, s2 was the best ranked maintenance system, while s4 was the best ranked maintenance system in terms of machine, human attributes and maintenance cost criteria (table 12). from the perspective of maintenance system-wise, the highest grey relational grades for all the maintenance system was human attribute criterion. furthermore, maintenance cost grey relational grade was the lowest for s1, s2 and s3. the lowest grey relational grade for s4 was physical environment criterion. table 12. grey relational grade goals s1 s2 s3 s4 c1 0.5602 0.8696 0.6478 0.7287 c2 0.5798 0.7783 0.6371 0.7612 c3 0.6246 0.8677 0.6857 0.8643 c4 0.7218 1.0290 0.8568 1.1552 c5 0.5157 0.6876 0.5329 0.8452 in order to apply vikor technique, the grey relational grades for each of the maintenance systems were computed (table 12). the results obtained showed that maintenance system s2 had the maximum values for criteria c1 to c3, while maintenance system s4 had the maximum values for criteria c4 and c5. the minimum values for criteria c1 to c5 were obtained from maintenance system s1 (table 13). table 13. calculated best and worst values c1 c2 c3 c4 c5 i f  0.8696 0.7783 0.8677 1.1552 0.8452 i f  0.5602 0.5798 0.6246 0.7218 0.5157 an integrated fuzzy analytical hierarchical process and fuzzy grey relational analytical model with vikor for maintenance system appraisal issn: 2180-1053 vol. 9 no.1 january – june 2017 83 the values for s, r and q were generated using the information in tables 12 and 13 using equations (17) to (19). from the information in table 14, the acceptable advantage and stability were checked and it was observed that they were satisfied (table 15). it could be deduced that the best maintenance system was s4. based on the results in tables 12 and 13, it is obvious that the use of single performance index to appraise maintenance system is not as reliable as a multi-criteria approach. table 14. utility, regret measure and vikor index s1 s2 s3 s4 s 0.9999 0.1840 0.7731 0.0899 r 0.2470 0.1181 0.2341 0.0733 q (v = 0.2) 1.0000 0.1344 0.7858 0.0000 q (v = 0.5) 1.0000 0.1809 0.8383 0.0000 q (v = 0.8) 1.0000 0.2273 0.8908 0.0000 the vikor results showed there is consistency in the utility, regret measure and vikor index results (table 15). from table 15, the best maintenance system was s4, while s1 was the least ranked maintenance system. the results obtained from this study can be used to benchmark maintenance system. this will reveal best practices that can be used to improve manufacturing goals. furthermore, the proposed model can be used to carry out internal benchmarking process. this could be factory-wise or maintenance section-wise. it will require minor adjustments of the proposed framework, by changing maintenance system with factory or maintenance section. table 15. ranking of maintenance systems 1 2 3 4 s s4 s2 s3 s1 r s4 s2 s3 s1 q s4 s2 s3 s1 from the foregoing, the concept framework has the capacity to generate ranks for maintenance systems. the principal components that were considered for each of the maintenance criterion could be either increase or decrease to suite a maintenance system of interest. for instance, the proposed framework can be applied to service systems. to achieve this, redefinition of the terms used in the proposed framework are required. one of the limitations of the proposed framework is that it relies mainly on subjective responses from decision makers. this implies that biasness of decision makers may affect the outcome of the proposed model. this may be experienced when the model is used to evaluate maintenance sections in a maintenance department. the contributions of this study are as follows: it introduces the concept of environment, human and machine criteria under a single framework for maintenance system appraisal. the use of vikor approach for maintenance system appraisal has been introduced. journal of mechanical engineering and technology 84 issn: 2180-1053 vol. 9 no.1 january – june 2017 6.0 conclusions this study presents a conceptual framework, based on fuzzy analytical hierarchy process (fahp), fuzzy grey relational analysis (fgra) and vikor, for maintenance systems appraisal. the proposed framework has addressed three problems: (i) determination of weights for maintenance criteria under fuzzy environments using fahp; (ii) aggregation of maintenance criterion principal components when dealing with fuzzy environment using fgra; and (iii) appraisal of maintenance systems using vikor technique. this study has shown that the appraisal of maintenance systems using multicriteria is a more robust means for maintenance activities analysis when compared with single factor performance indicators. the proposed framework applicability was verified using information obtained from four manufacturing systems. the results obtained showed that the proposed framework is a veritable tool for maintenance system appraisal. in addition, the results from the proposed framework have shown that it has the capacity to drive the quest for improved performance of manufacturing systems maintenance departments. furthermore, there is the need to perform system stability analysis prior to data collection during the proposed framework application. a study which considers the classification of maintenance environment using expert systems could be pursued as a further study. prioritisation of manufacturing system goals from maintenance perspective could be considered as a future study. a future study which considered the application of proposed framework for ranking maintenance policy in manufacturing system could be pursued. acknowledgements the authors are grateful to the editor and reviewers and all who filled the questionnaires. references ahren, t. & parida, a. (2009). maintenance performance indicators (mpi) for benchmarking the railway infrastructure. a case study, benchmarking: an international journal, 16(2), 247-258. baluch, n., abdullah, c.s.b. & mohtar, s.b. (2010). maintenance management performance: an overview towards evaluating malaysian palm oil mill. the asian journal of technology management, 3(1), 1-4. chang, d.y. (1996). application of extend analysis method on fuzzy ahp. european journal of operational research, 96, 343-350. de groote, p. (1995). maintenance performance analysis: a practical approach, journal of quality in maintenance engineering, 1(2), 4-24. elangovan, k., selladurai, v., devadasan, s.r., goyal, s.k. & muthu, s. (2007). quality and productivity improvement of executive decisions in maintenance engineering: an ess-based approach, international journal of productivity and quality management, 2(1), 112-139. an integrated fuzzy analytical hierarchical process and fuzzy grey relational analytical model with vikor for maintenance system appraisal issn: 2180-1053 vol. 9 no.1 january – june 2017 85 girubha, r.j. & vinodh, s. (2012). application of fuzzy vikor and environmental impact analysis for material selection of an automotive component. materials and design, 37, 478-486. hasani, h., tabatabaei, s.a. & amiri, g. (2012). grey relational analysis to determine the optimum process parameters for open-end spinning yarns. journal of engineered fibers and fabrics, 7(2), 81-86. khan, m.r.r. & darrab, i.a. (2010). development of analytical relation between maintenance, quality and productivity, journal of quality in maintenance engineering, 16(4), 341-353. kumar, u. & parida, a. (2006). maintenance performance measurement: the need of the hour for the mechanized mining industry, proceedings of the 1st asian mining congress, 16-18 jan 2006, kolkatha, india. maletic, d., maletic, m., al-najjar, b. & gomiscek, b. (2014). the role of maintenance in improving company’s competitiveness and profitability: a case study in a textile company, journal of manufacturing technology management, 25(4), 441-456. muchiri, p., pintelon, l., gelders, l. & martin, h. (2011). development of maintenance function performance measurement framework and indicators. international journal of production economics, 131(1), 295-302. oke, s.a. (2005). an analytical model for the optimization of maintenance profitability, international journal of productivity and performance management, 54(2), 113-136. oke, s.a. oyedokun, o.i, akanbi, o.g. & oyawale f.a. (2008). an inflation-based maintenance profitability model, international journal of productivity and quality management, 3(3), 325-339. opricovic, s. & tzeng, g.h. (2004). compromise solution by mcdm methods: a comparative analysis of vikor and topsis. european journal of operations research, 156, 445-455. parida, a. & kumar, u. (2004). managing information is key to maintenance effectiveness, e-proceedings of the intelligent maintenance system’s (ims), july 15-17, arles, france. parida, a., chattopadhyay, g. & kumar, u. (2005). multi-criteria maintenance performance measurement: a conceptual model. proceedings of the 18th international congress of condition monitoring and diagnostic engineering management, 31 august – 2 september, cranfield, uk, pp. 349-356. parida, a. & kumar, u. (2006). maintenance performance measurement (mpm): issues and challenges, journal of quality in maintenance engineering, 12(3), 239-251. journal of mechanical engineering and technology 86 issn: 2180-1053 vol. 9 no.1 january – june 2017 parida, a. (2007). study and analysis of maintenance performance indicators (mpis) for lkab: a case study, journal of quality in maintenance engineering, 13(4), 325-327. parida, a. & uday, k. (2009). maintenance performance measurement: methods, tools and application, maintworld, 1(1), 50-53. raouf, a. (1994). improving capital productivity through maintenance, international journal of operations & production management, 14(7), 44-52. wang, c-h. & pang, c-t. (2011). using vikor method for evaluating service quality of online auction under fuzzy environment. international journal of computer science engineering and technology, 1(6), pp. 307-314. plantweb (2003).white paper: reducing operations and maintenance costs. www.emersonprocess.com (accessed june 20, 2016). saaty, t.l. (1990). how to make a decision: the analytic hierarchy process. european journal of operational research, 48, 9-26. saaty, t.l. (1980). the analytical hierarchical process. mcgraw-hill, new york. simoes, j.m., gomes, c.f. & yasin, m.m. (2011). a literature review of maintenance performance measurement: a conceptual framework and directions for future research, journal of quality in maintenance engineering, 17(2), 116-137. sondalini, m. (2016). useful key performance indicators for maintenance. www.lifetime-reliability.com (accessed june 20, 2016). wu, h-y., chen, j-k. & chen, i-s. (2012). performance evaluation of aircraft maintenance staff using a fuzzy mcdm approach. international journal of innovative computing, information and control, 8(6), 39193937. issn: 2180-1053 vol. 7 no. 2 july december 2015 hydromagnetic short bearings 19 hydromagnetic short bearings g. m. deheri1, r. m. patel2* and p. a. vadher3 1department of mathematics, sardar patel university, vallabh vidyanagar, 388 120, gujarat state, india. 2department of mathematics, gujarat arts and science college, ahmedabad, 380 006 gujarat state, india. 3department of physics, government science college,gandhinagar, 382 016 gujarat state, india. abstract this article deals with the performance of a hydromagnetic short porous bearing. an electrically conducting lubricant in the presence of a transverse magnetic field has been taken into consideration while the plates are electrically conducting. the related reynolds’ equation governing the fluid film pressure is solved under suitable boundary conditions to get the pressure distribution leading to the computation of load carrying capacity. the results presented in graphical form establish that the bearing system registers an improved performance due to hydromagnetization. besides, the load carrying capacity increases considerably with respect to the conductivity. it is revealed that the negative effect of porosity and the ratio of breadth to height can be neutralized up to a considerable extent by the positive effect of hydromagnetization suitably choosing the plate conductivity and the aspect ratio. it is found that the hydromagnetization presents the friction at both the plates to be equal. keywords: hydromagnetic lubrication,short bearing, reynolds’ equation, load carrying capacity, friction 1.0 introduction (pinkus and sternlicht, 1961) laid down the classical analysis of the hydrodynamic lubrication of slider bearings. subsequently, in this direction significant amount of works were done by several investigators (lord rayleigh, 1918), (archibald, 1950), (charnes and saibel, 1952), (cameron, 1966), (gross et al., 1980), (hamrock, 1994), (basu et al., 2005), (majmudar, 2008). equally important are the contributions of (bagci and singh, 1983), (osterle et al., 1958), (patel and gupta, 1983) and (abramovitz, 1955) concerning the performance of hydrodynamic slider bearing. (mc. allister et al. 1980) discussed the design of optimum * corresponding author email: jrmpatel@rediffmail.com issn: 2180-1053 vol. 7 no. 2 july december 2015 journal of mechanical engineering and technology 20 one dimensional slider bearing in terms of the load carrying capacity. an approximate analytic solution for performance characteristics of a porous metal bearing was proposed for the first time by (morgan and cameron, 1957). the exact solution of this problem was obtained by (rouleau, 1963). (prakash and vij, 1973) investigated the hydrodynamic lubrication of a plane slider bearing resorting to several geometries. it is a well known fact that if the liquid metals such as mercury and sodium are pumped or held between moving surfaces of a bearing, larger loads can be supported by employing a strong magnetic field. the application of a large magnetic field results in electromagnetic pressurization as the liquid metals are large electrical conductors. this aspect of study was explored by (elco and huges, 1962), (kuzma, 1964) and (kuzma et al. 1964). from these investigations, it becomes clear that it is possible to increase the load carrying capacity by the utilization of electromagnetic force, thereby overcoming the defects associated with the lubricant at higher temperature and hence alleviating the drawback of low viscosity. the load carrying capacity can be made to register high increase by taking recourse to super conducting magnets while little amount of power is required to provide the magnetic field. a good deal of research has been done regarding the theoretical and experimental studies on the hydromagnetic lubrication of porous as well as plane metal bearings (snyder, 1962), (shukla, 1963), (patel and hingu, 1978). (shukla and prasad, 1965) analyzed the performance of hydromagnetic squeeze films between two conducting non-porous surfaces and discussed the effect of conductivities on the behavior of squeeze film. (sinha and gupta, 1974) investigated the hydromagnetic effect on the behavior of squeeze film between porous annular plates. (patel and gupta, 1979) deployed morgan – cameron approximations simplifying the analysis for hydromagnetic squeeze films between parallel plates for a number of geometrical shapes. (prajapati, 1995) also studied the behavior of magnetic fluid based porous squeeze film between plates of various geometries. for a short bearing (patel et al. 2010) observed that the magnetic fluid resulted in a marginally improved performance. here it has been sought to analyze the performance of a hydromagnetic short bearing. 2.0 analysis the geometrical configuration of the bearing which is infinite in z-direction is presented in figure 1. issn: 2180-1053 vol. 7 no. 2 july december 2015 hydromagnetic short bearings 21 2 employing a strong magnetic field. the application of a large magnetic field results in electromagnetic pressurization as the liquid metals are large electrical conductors. this aspect of study was explored by (elco and huges, 1962), (kuzma, 1964) and (kuzma et al. 1964). from these investigations, it becomes clear that it is possible to increase the load carrying capacity by the utilization of electromagnetic force, thereby overcoming the defects associated with the lubricant at higher temperature and hence alleviating the drawback of low viscosity. the load carrying capacity can be made to register high increase by taking recourse to super conducting magnets while little amount of power is required to provide the magnetic field. a good deal of research has been done regarding the theoretical and experimental studies on the hydromagnetic lubrication of porous as well as plane metal bearings (snyder, 1962), (shukla, 1963), (patel and hingu, 1978). (shukla and prasad, 1965) analyzed the performance of hydromagnetic squeeze films between two conducting non-porous surfaces and discussed the effect of conductivities on the behavior of squeeze film. (sinha and gupta, 1974) investigated the hydromagnetic effect on the behavior of squeeze film between porous annular plates. (patel and gupta, 1979) deployed morgan – cameron approximations simplifying the analysis for hydromagnetic squeeze films between parallel plates for a number of geometrical shapes. (prajapati, 1995) also studied the behavior of magnetic fluid based porous squeeze film between plates of various geometries. for a short bearing (patel et al. 2010) observed that the magnetic fluid resulted in a marginally improved performance. here it has been sought to analyze the performance of a hydromagnetic short bearing. 2.0 analysis the geometrical configuration of the bearing which is infinite in z-direction is presented in figure 1. figure 1. geometrical configuration of the bearing system in the x-direction the slider moves with the uniform velocity u. l is the length of the bearing and the breadth b lies in the z-direction, wherein, b << l. the pressure gradient p / z is much larger than the pressure gradient p / x as the dimension b is figure 1. geometrical configuration of the bearing system in the x-direction the slider moves with the uniform velocity u. l is the length of the bearing and the breadth b lies in the z-direction, wherein, b << l. the pressure gradient ∂p / ∂z is much larger than the pressure gradient ∂p / ∂x as the dimension b is very small. therefore, ∂p / ∂x can be neglected. the lubricant film is considered to be isoviscous, incompressible and the flow is laminar. under the usual assumptions of hydromagnetic lubrication the modified reynolds’ equation governing the lubricant film pressure is obtained as (patel and deheri, 2004), (vadher et al., 2008), (patel et al., 2010). 3 very small. therefore, p / x can be neglected. the lubricant film is considered to be isoviscous, incompressible and the flow is laminar. under the usual assumptions of hydromagnetic lubrication the modified reynolds’ equation governing the lubricant film pressure is obtained as (patel and deheri, 2004), (vadher et al., 2008), (patel et al., 2010). d(-c) μ h 3 dx dh 6u 2dz p2d   (1) where              l x 1m1 2 hh c=             2c ψ m/2m/2t anh 3m 2 d=                   m/2 m/2t anh 10 1 10 solving this equation with the associated boundary conditions, p = 0 at z = (b/2) and 0 dz dp  at z = 0 (2) one gets the expression for pressure distribution as, d(c) μ 3h l 2z 4 2b 2 mh3u p            (3) where 2h 2h1hm   in view of the following non-dimensional quantities, 2μub p3 2 h p  b z z l x x  where 3 very small. therefore, p / x can be neglected. the lubricant film is considered to be isoviscous, incompressible and the flow is laminar. under the usual assumptions of hydromagnetic lubrication the modified reynolds’ equation governing the lubricant film pressure is obtained as (patel and deheri, 2004), (vadher et al., 2008), (patel et al., 2010). d(-c) μ h 3 dx dh 6u 2dz p2d   (1) where              l x 1m1 2 hh c=             2c ψ m/2m/2t anh 3m 2 d=                   m/2 m/2t anh 10 1 10 solving this equation with the associated boundary conditions, p = 0 at z = (b/2) and 0 dz dp  at z = 0 (2) one gets the expression for pressure distribution as, d(c) μ 3h l 2z 4 2b 2 mh3u p            (3) where 2h 2h1hm   in view of the following non-dimensional quantities, 2μub p3 2 h p  b z z l x x  solving this equation with the associated boundary conditions, 3 very small. therefore, p / x can be neglected. the lubricant film is considered to be isoviscous, incompressible and the flow is laminar. under the usual assumptions of hydromagnetic lubrication the modified reynolds’ equation governing the lubricant film pressure is obtained as (patel and deheri, 2004), (vadher et al., 2008), (patel et al., 2010). d(-c) μ h 3 dx dh 6u 2dz p2d   (1) where              l x 1m1 2 hh c=             2c ψ m/2m/2t anh 3m 2 d=                   m/2 m/2t anh 10 1 10 solving this equation with the associated boundary conditions, p = 0 at z = (b/2) and 0 dz dp  at z = 0 (2) one gets the expression for pressure distribution as, d(c) μ 3h l 2z 4 2b 2 mh3u p            (3) where 2h 2h1hm   in view of the following non-dimensional quantities, 2μub p3 2 h p  b z z l x x  issn: 2180-1053 vol. 7 no. 2 july december 2015 journal of mechanical engineering and technology 22 and 3 very small. therefore, p / x can be neglected. the lubricant film is considered to be isoviscous, incompressible and the flow is laminar. under the usual assumptions of hydromagnetic lubrication the modified reynolds’ equation governing the lubricant film pressure is obtained as (patel and deheri, 2004), (vadher et al., 2008), (patel et al., 2010). d(-c) μ h 3 dx dh 6u 2dz p2d   (1) where              l x 1m1 2 hh c=             2c ψ m/2m/2t anh 3m 2 d=                   m/2 m/2t anh 10 1 10 solving this equation with the associated boundary conditions, p = 0 at z = (b/2) and 0 dz dp  at z = 0 (2) one gets the expression for pressure distribution as, d(c) μ 3h l 2z 4 2b 2 mh3u p            (3) where 2h 2h1hm   in view of the following non-dimensional quantities, 2μub p3 2 h p  b z z l x x  one gets the expression for pressure distribution as, 3 very small. therefore, p / x can be neglected. the lubricant film is considered to be isoviscous, incompressible and the flow is laminar. under the usual assumptions of hydromagnetic lubrication the modified reynolds’ equation governing the lubricant film pressure is obtained as (patel and deheri, 2004), (vadher et al., 2008), (patel et al., 2010). d(-c) μ h 3 dx dh 6u 2dz p2d   (1) where              l x 1m1 2 hh c=             2c ψ m/2m/2t anh 3m 2 d=                   m/2 m/2t anh 10 1 10 solving this equation with the associated boundary conditions, p = 0 at z = (b/2) and 0 dz dp  at z = 0 (2) one gets the expression for pressure distribution as, d(c) μ 3h l 2z 4 2b 2 mh3u p            (3) where 2h 2h1hm   in view of the following non-dimensional quantities, 2μub p3 2 h p  b z z l x x  where 3 very small. therefore, p / x can be neglected. the lubricant film is considered to be isoviscous, incompressible and the flow is laminar. under the usual assumptions of hydromagnetic lubrication the modified reynolds’ equation governing the lubricant film pressure is obtained as (patel and deheri, 2004), (vadher et al., 2008), (patel et al., 2010). d(-c) μ h 3 dx dh 6u 2dz p2d   (1) where              l x 1m1 2 hh c=             2c ψ m/2m/2t anh 3m 2 d=                   m/2 m/2t anh 10 1 10 solving this equation with the associated boundary conditions, p = 0 at z = (b/2) and 0 dz dp  at z = 0 (2) one gets the expression for pressure distribution as, d(c) μ 3h l 2z 4 2b 2 mh3u p            (3) where 2h 2h1hm   in view of the following non-dimensional quantities, 2μub p3 2 h p  b z z l x x  in view of the following non-dimensional quantities, 3 very small. therefore, p / x can be neglected. the lubricant film is considered to be isoviscous, incompressible and the flow is laminar. under the usual assumptions of hydromagnetic lubrication the modified reynolds’ equation governing the lubricant film pressure is obtained as (patel and deheri, 2004), (vadher et al., 2008), (patel et al., 2010). d(-c) μ h 3 dx dh 6u 2dz p2d   (1) where              l x 1m1 2 hh c=             2c ψ m/2m/2t anh 3m 2 d=                   m/2 m/2t anh 10 1 10 solving this equation with the associated boundary conditions, p = 0 at z = (b/2) and 0 dz dp  at z = 0 (2) one gets the expression for pressure distribution as, d(c) μ 3h l 2z 4 2b 2 mh3u p            (3) where 2h 2h1hm   in view of the following non-dimensional quantities, 2μub p3 2 h p  b z z l x x  the distribution of pressure in dimensionless form can be obtained as, 4 the distribution of pressure in dimensionless form can be obtained as, 3edc 2 4 1 l 2 3mh p          z (4) where e=   x1m1  hence the dimensionless load carrying capacity is found to be, 4μub w32hw     2 1 2 1 1 0 dzdxz)p(x, dc)( 2 h 4b 21)(m 2)m(m     (5) at the lower plane of the moving plate the frictional forcef per unit width is derived as,  1/2 1/2dzτf (6) where τ μu 2hτ        is dimensionless shearing stress. while h μu 2 h y dz dp τ        (7) simplification of equation (7) leads to, e 1 2 1 ye 2 dz dp τ        h b (8) where h y y  for the moving plate (y = 0) the dimensionless shearing stress takes the form, e 1 2edc)( 2 h l 3mz τ    (9) where 4 the distribution of pressure in dimensionless form can be obtained as, 3edc 2 4 1 l 2 3mh p          z (4) where e=   x1m1  hence the dimensionless load carrying capacity is found to be, 4μub w32hw     2 1 2 1 1 0 dzdxz)p(x, dc)( 2 h 4b 21)(m 2)m(m     (5) at the lower plane of the moving plate the frictional forcef per unit width is derived as,  1/2 1/2dzτf (6) where τ μu 2hτ        is dimensionless shearing stress. while h μu 2 h y dz dp τ        (7) simplification of equation (7) leads to, e 1 2 1 ye 2 dz dp τ        h b (8) where h y y  for the moving plate (y = 0) the dimensionless shearing stress takes the form, e 1 2edc)( 2 h l 3mz τ    (9) hence the dimensionless load carrying capacity is found to be, 4 the distribution of pressure in dimensionless form can be obtained as, 3edc 2 4 1 l 2 3mh p          z (4) where e=   x1m1  hence the dimensionless load carrying capacity is found to be, 4μub w32hw     2 1 2 1 1 0 dzdxz)p(x, dc)( 2 h 4b 21)(m 2)m(m     (5) at the lower plane of the moving plate the frictional forcef per unit width is derived as,  1/2 1/2dzτf (6) where τ μu 2hτ        is dimensionless shearing stress. while h μu 2 h y dz dp τ        (7) simplification of equation (7) leads to, e 1 2 1 ye 2 dz dp τ        h b (8) where h y y  for the moving plate (y = 0) the dimensionless shearing stress takes the form, e 1 2edc)( 2 h l 3mz τ    (9) at the lower plane of the moving plate the frictional force 4 the distribution of pressure in dimensionless form can be obtained as, 3edc 2 4 1 l 2 3mh p          z (4) where e=   x1m1  hence the dimensionless load carrying capacity is found to be, 4μub w32hw     2 1 2 1 1 0 dzdxz)p(x, dc)( 2 h 4b 21)(m 2)m(m     (5) at the lower plane of the moving plate the frictional forcef per unit width is derived as,  1/2 1/2dzτf (6) where τ μu 2hτ        is dimensionless shearing stress. while h μu 2 h y dz dp τ        (7) simplification of equation (7) leads to, e 1 2 1 ye 2 dz dp τ        h b (8) where h y y  for the moving plate (y = 0) the dimensionless shearing stress takes the form, e 1 2edc)( 2 h l 3mz τ    (9) per unit width is derived as, 4 the distribution of pressure in dimensionless form can be obtained as, 3edc 2 4 1 l 2 3mh p          z (4) where e=   x1m1  hence the dimensionless load carrying capacity is found to be, 4μub w32hw     2 1 2 1 1 0 dzdxz)p(x, dc)( 2 h 4b 21)(m 2)m(m     (5) at the lower plane of the moving plate the frictional forcef per unit width is derived as,  1/2 1/2dzτf (6) where τ μu 2hτ        is dimensionless shearing stress. while h μu 2 h y dz dp τ        (7) simplification of equation (7) leads to, e 1 2 1 ye 2 dz dp τ        h b (8) where h y y  for the moving plate (y = 0) the dimensionless shearing stress takes the form, e 1 2edc)( 2 h l 3mz τ    (9) issn: 2180-1053 vol. 7 no. 2 july december 2015 hydromagnetic short bearings 23 where 4 the distribution of pressure in dimensionless form can be obtained as, 3edc 2 4 1 l 2 3mh p          z (4) where e=   x1m1  hence the dimensionless load carrying capacity is found to be, 4μub w32hw     2 1 2 1 1 0 dzdxz)p(x, dc)( 2 h 4b 21)(m 2)m(m     (5) at the lower plane of the moving plate the frictional forcef per unit width is derived as,  1/2 1/2dzτf (6) where τ μu 2hτ        is dimensionless shearing stress. while h μu 2 h y dz dp τ        (7) simplification of equation (7) leads to, e 1 2 1 ye 2 dz dp τ        h b (8) where h y y  for the moving plate (y = 0) the dimensionless shearing stress takes the form, e 1 2edc)( 2 h l 3mz τ    (9) while 4 the distribution of pressure in dimensionless form can be obtained as, 3edc 2 4 1 l 2 3mh p          z (4) where e=   x1m1  hence the dimensionless load carrying capacity is found to be, 4μub w32hw     2 1 2 1 1 0 dzdxz)p(x, dc)( 2 h 4b 21)(m 2)m(m     (5) at the lower plane of the moving plate the frictional forcef per unit width is derived as,  1/2 1/2dzτf (6) where τ μu 2hτ        is dimensionless shearing stress. while h μu 2 h y dz dp τ        (7) simplification of equation (7) leads to, e 1 2 1 ye 2 dz dp τ        h b (8) where h y y  for the moving plate (y = 0) the dimensionless shearing stress takes the form, e 1 2edc)( 2 h l 3mz τ    (9) simplification of equation (7) leads to, 4 the distribution of pressure in dimensionless form can be obtained as, 3edc 2 4 1 l 2 3mh p          z (4) where e=   x1m1  hence the dimensionless load carrying capacity is found to be, 4μub w32hw     2 1 2 1 1 0 dzdxz)p(x, dc)( 2 h 4b 21)(m 2)m(m     (5) at the lower plane of the moving plate the frictional forcef per unit width is derived as,  1/2 1/2dzτf (6) where τ μu 2hτ        is dimensionless shearing stress. while h μu 2 h y dz dp τ        (7) simplification of equation (7) leads to, e 1 2 1 ye 2 dz dp τ        h b (8) where h y y  for the moving plate (y = 0) the dimensionless shearing stress takes the form, e 1 2edc)( 2 h l 3mz τ    (9) where 4 the distribution of pressure in dimensionless form can be obtained as, 3edc 2 4 1 l 2 3mh p          z (4) where e=   x1m1  hence the dimensionless load carrying capacity is found to be, 4μub w32hw     2 1 2 1 1 0 dzdxz)p(x, dc)( 2 h 4b 21)(m 2)m(m     (5) at the lower plane of the moving plate the frictional forcef per unit width is derived as,  1/2 1/2dzτf (6) where τ μu 2hτ        is dimensionless shearing stress. while h μu 2 h y dz dp τ        (7) simplification of equation (7) leads to, e 1 2 1 ye 2 dz dp τ        h b (8) where h y y  for the moving plate (y = 0) the dimensionless shearing stress takes the form, e 1 2edc)( 2 h l 3mz τ    (9) for the moving plate (y = 0) the dimensionless shearing stress takes the form, 4 the distribution of pressure in dimensionless form can be obtained as, 3edc 2 4 1 l 2 3mh p          z (4) where e=   x1m1  hence the dimensionless load carrying capacity is found to be, 4μub w32hw     2 1 2 1 1 0 dzdxz)p(x, dc)( 2 h 4b 21)(m 2)m(m     (5) at the lower plane of the moving plate the frictional forcef per unit width is derived as,  1/2 1/2dzτf (6) where τ μu 2hτ        is dimensionless shearing stress. while h μu 2 h y dz dp τ        (7) simplification of equation (7) leads to, e 1 2 1 ye 2 dz dp τ        h b (8) where h y y  for the moving plate (y = 0) the dimensionless shearing stress takes the form, e 1 2edc)( 2 h l 3mz τ    (9) therefore, the frictional force in non-dimensional form is given by, 5 therefore, the frictional force in non-dimensional form is given by, e 1 0f  (10) in addition, at the fixed plate (y = 1) one finds that, e 1 2edc)( 2 h l 3mz τ    (11) finally, the frictional force in dimensionless form is obtained as, e 1 f1  (12) 3.0 results and discussions it is clearly seen from equations (4) and (5) that the non-dimensional pressure and load carrying capacity are dependent on various parameters such as magnetization m, porosity , conductivity 0 + 1, aspect ratio m and ratios l/h2 and b/h2. however, the equations (10) and (12) suggest that the friction depends on the aspect ratio m and obviously x = x/l. it is manifest that the friction is independent of hydromagnetization m. taking the conductivity 0 + 1 to be zero in the limiting case of m  0; the present analysis turns in essentially, the discussions of (basu et al., 2005) in the absence of porosity. it is noticed that conductivity 0 + 1 increases the load carrying capacity for fixed values of magnetization m, porosity , aspect ratio m and the ratio b/h2. in addition, the distribution of load carrying capacity comes through the factor,                   110 m/2 m/2t anh 10 for large values of m this approaches to,           1 10 10 as tanh(m/2)  1. it is observed that as conductivity 0 + 1 increases the load carrying capacity increases. here it is pertinent to see that the bearing can support a load even when there is no flow. lastly, a comparison of this investigation with the discussion of (patel and deheri, 2004) reveals that the load carrying capacity is comparatively reduced here. probably, this is due to the fringing phenomena which occur when the plates are electrically conducting. (13) (14) in addition, at the fixed plate (y = 1) one finds that, 5 therefore, the frictional force in non-dimensional form is given by, e 1 0f  (10) in addition, at the fixed plate (y = 1) one finds that, e 1 2edc)( 2 h l 3mz τ    (11) finally, the frictional force in dimensionless form is obtained as, e 1 f1  (12) 3.0 results and discussions it is clearly seen from equations (4) and (5) that the non-dimensional pressure and load carrying capacity are dependent on various parameters such as magnetization m, porosity , conductivity 0 + 1, aspect ratio m and ratios l/h2 and b/h2. however, the equations (10) and (12) suggest that the friction depends on the aspect ratio m and obviously x = x/l. it is manifest that the friction is independent of hydromagnetization m. taking the conductivity 0 + 1 to be zero in the limiting case of m  0; the present analysis turns in essentially, the discussions of (basu et al., 2005) in the absence of porosity. it is noticed that conductivity 0 + 1 increases the load carrying capacity for fixed values of magnetization m, porosity , aspect ratio m and the ratio b/h2. in addition, the distribution of load carrying capacity comes through the factor,                   110 m/2 m/2t anh 10 for large values of m this approaches to,           1 10 10 as tanh(m/2)  1. it is observed that as conductivity 0 + 1 increases the load carrying capacity increases. here it is pertinent to see that the bearing can support a load even when there is no flow. lastly, a comparison of this investigation with the discussion of (patel and deheri, 2004) reveals that the load carrying capacity is comparatively reduced here. probably, this is due to the fringing phenomena which occur when the plates are electrically conducting. (13) (14) finally, the frictional force in dimensionless form is obtained as, 5 therefore, the frictional force in non-dimensional form is given by, e 1 0f  (10) in addition, at the fixed plate (y = 1) one finds that, e 1 2edc)( 2 h l 3mz τ    (11) finally, the frictional force in dimensionless form is obtained as, e 1 f1  (12) 3.0 results and discussions it is clearly seen from equations (4) and (5) that the non-dimensional pressure and load carrying capacity are dependent on various parameters such as magnetization m, porosity , conductivity 0 + 1, aspect ratio m and ratios l/h2 and b/h2. however, the equations (10) and (12) suggest that the friction depends on the aspect ratio m and obviously x = x/l. it is manifest that the friction is independent of hydromagnetization m. taking the conductivity 0 + 1 to be zero in the limiting case of m  0; the present analysis turns in essentially, the discussions of (basu et al., 2005) in the absence of porosity. it is noticed that conductivity 0 + 1 increases the load carrying capacity for fixed values of magnetization m, porosity , aspect ratio m and the ratio b/h2. in addition, the distribution of load carrying capacity comes through the factor,                   110 m/2 m/2t anh 10 for large values of m this approaches to,           1 10 10 as tanh(m/2)  1. it is observed that as conductivity 0 + 1 increases the load carrying capacity increases. here it is pertinent to see that the bearing can support a load even when there is no flow. lastly, a comparison of this investigation with the discussion of (patel and deheri, 2004) reveals that the load carrying capacity is comparatively reduced here. probably, this is due to the fringing phenomena which occur when the plates are electrically conducting. (13) (14) 3.0 results and discussions it is clearly seen from equations (4) and (5) that the non-dimensional pressure and load carrying capacity are dependent on various parameters such as magnetization m, porosity ψ, conductivity ϕ0 + ϕ1, aspect ratio m and ratios l/h2 and b/h2. however, the equations (10) and (12) suggest that the friction depends on the aspect ratio m and issn: 2180-1053 vol. 7 no. 2 july december 2015 journal of mechanical engineering and technology 24 obviously x = x/l. it is manifest that the friction is independent of hydromagnetization m. taking the conductivity ϕ0 + ϕ1 to be zero in the limiting case of m → 0; the present analysis turns in essentially, the discussions of (basu et al., 2005) in the absence of porosity. it is noticed that conductivity ϕ0 + ϕ1 increases the load carrying capacity for fixed values of magnetization m, porosity ψ, aspect ratio m and the ratio b/h2. in addition, the distribution of load carrying capacity comes through the factor, 5 therefore, the frictional force in non-dimensional form is given by, e 1 0f  (10) in addition, at the fixed plate (y = 1) one finds that, e 1 2edc)( 2 h l 3mz τ    (11) finally, the frictional force in dimensionless form is obtained as, e 1 f1  (12) 3.0 results and discussions it is clearly seen from equations (4) and (5) that the non-dimensional pressure and load carrying capacity are dependent on various parameters such as magnetization m, porosity , conductivity 0 + 1, aspect ratio m and ratios l/h2 and b/h2. however, the equations (10) and (12) suggest that the friction depends on the aspect ratio m and obviously x = x/l. it is manifest that the friction is independent of hydromagnetization m. taking the conductivity 0 + 1 to be zero in the limiting case of m  0; the present analysis turns in essentially, the discussions of (basu et al., 2005) in the absence of porosity. it is noticed that conductivity 0 + 1 increases the load carrying capacity for fixed values of magnetization m, porosity , aspect ratio m and the ratio b/h2. in addition, the distribution of load carrying capacity comes through the factor,                   110 m/2 m/2t anh 10 for large values of m this approaches to,           1 10 10 as tanh(m/2)  1. it is observed that as conductivity 0 + 1 increases the load carrying capacity increases. here it is pertinent to see that the bearing can support a load even when there is no flow. lastly, a comparison of this investigation with the discussion of (patel and deheri, 2004) reveals that the load carrying capacity is comparatively reduced here. probably, this is due to the fringing phenomena which occur when the plates are electrically conducting. (13) (14) for large values of m this approaches to, 5 therefore, the frictional force in non-dimensional form is given by, e 1 0f  (10) in addition, at the fixed plate (y = 1) one finds that, e 1 2edc)( 2 h l 3mz τ    (11) finally, the frictional force in dimensionless form is obtained as, e 1 f1  (12) 3.0 results and discussions it is clearly seen from equations (4) and (5) that the non-dimensional pressure and load carrying capacity are dependent on various parameters such as magnetization m, porosity , conductivity 0 + 1, aspect ratio m and ratios l/h2 and b/h2. however, the equations (10) and (12) suggest that the friction depends on the aspect ratio m and obviously x = x/l. it is manifest that the friction is independent of hydromagnetization m. taking the conductivity 0 + 1 to be zero in the limiting case of m  0; the present analysis turns in essentially, the discussions of (basu et al., 2005) in the absence of porosity. it is noticed that conductivity 0 + 1 increases the load carrying capacity for fixed values of magnetization m, porosity , aspect ratio m and the ratio b/h2. in addition, the distribution of load carrying capacity comes through the factor,                   110 m/2 m/2t anh 10 for large values of m this approaches to,           1 10 10 as tanh(m/2)  1. it is observed that as conductivity 0 + 1 increases the load carrying capacity increases. here it is pertinent to see that the bearing can support a load even when there is no flow. lastly, a comparison of this investigation with the discussion of (patel and deheri, 2004) reveals that the load carrying capacity is comparatively reduced here. probably, this is due to the fringing phenomena which occur when the plates are electrically conducting. (13) (14) as tanh(m/2) → 1. it is observed that as conductivity ϕ0 + ϕ1 increases the load carrying capacity increases. here it is pertinent to see that the bearing can support a load even when there is no flow. lastly, a comparison of this investigation with the discussion of (patel and deheri, 2004) reveals that the load carrying capacity is comparatively reduced here. probably, this is due to the fringing phenomena which occur when the plates are electrically conducting. 6 0.30 0.80 1.30 1.80 2.30 2.80 4.00 6.00 8.00 10.00 12.00 l oa d m 010 011 012 013 014 figure 2. variation of load carrying capacity with respect to m and 0+1. 0.30 0.80 1.30 1.80 2.30 2.80 4.00 6.00 8.00 10.00 12.00 l oa d m m=0.25 m=0.50 m=0.75 m=1.00 m=1.25 figure 3. variation of load carrying capacity with respect to m and m. figure 2. variation of load carrying capacity with respect to m and ϕ0+ϕ1. issn: 2180-1053 vol. 7 no. 2 july december 2015 hydromagnetic short bearings 25 6 0.30 0.80 1.30 1.80 2.30 2.80 4.00 6.00 8.00 10.00 12.00 l oa d m 010 011 012 013 014 figure 2. variation of load carrying capacity with respect to m and 0+1. 0.30 0.80 1.30 1.80 2.30 2.80 4.00 6.00 8.00 10.00 12.00 l oa d m m=0.25 m=0.50 m=0.75 m=1.00 m=1.25 figure 3. variation of load carrying capacity with respect to m and m. figure 3. variation of load carrying capacity with respect to m and m. 7 0.40 1.40 2.40 3.40 4.40 5.40 6.40 7.40 4.00 6.00 8.00 10.00 12.00 l oa d m b/h2=10 b/h2=20 b/h2=30 b/h2=40 b/h2=50 figure 4. variation of load carrying capacity with respect to m and b/h2. 0.10 0.60 1.10 1.60 2.10 2.60 3.10 3.60 4.00 6.00 8.00 10.00 12.00 l oa d m 0 0.01 0.05 0.1 0.5 figure 5. variation of load carrying capacity with respect to m and . figures 2 to 5 depict the variation of load carrying capacity with respect to the magnetization parameter m for various values of conductivity 0 + 1, aspect ratio m, the ratio b/h2 and porosity . it is noticed that the load carrying capacity gets increased with increasing values of magnetization parameter m. it is also seen that the conductivity 0 + 1 has an important role in improving the performance of the bearing system. the porosity  has a sharp adverse effect on the performance of the bearing system. figure 3 indicates that the load carrying capacity increases substantially with the increasing values of the aspect ratio m. besides, the load carrying capacity decreases with the increasing values of the ratio b/h2. in addition, the combined effect of magnetization m and the ratio b/h2 is more sharp as compared to the other combinations. figure 4. variation of load carrying capacity with respect to m and b/h2. 7 0.40 1.40 2.40 3.40 4.40 5.40 6.40 7.40 4.00 6.00 8.00 10.00 12.00 l oa d m b/h2=10 b/h2=20 b/h2=30 b/h2=40 b/h2=50 figure 4. variation of load carrying capacity with respect to m and b/h2. 0.10 0.60 1.10 1.60 2.10 2.60 3.10 3.60 4.00 6.00 8.00 10.00 12.00 l oa d m 0 0.01 0.05 0.1 0.5 figure 5. variation of load carrying capacity with respect to m and . figures 2 to 5 depict the variation of load carrying capacity with respect to the magnetization parameter m for various values of conductivity 0 + 1, aspect ratio m, the ratio b/h2 and porosity . it is noticed that the load carrying capacity gets increased with increasing values of magnetization parameter m. it is also seen that the conductivity 0 + 1 has an important role in improving the performance of the bearing system. the porosity  has a sharp adverse effect on the performance of the bearing system. figure 3 indicates that the load carrying capacity increases substantially with the increasing values of the aspect ratio m. besides, the load carrying capacity decreases with the increasing values of the ratio b/h2. in addition, the combined effect of magnetization m and the ratio b/h2 is more sharp as compared to the other combinations. figure 5. variation of load carrying capacity with respect to m and ψ. issn: 2180-1053 vol. 7 no. 2 july december 2015 journal of mechanical engineering and technology 26 figures 2 to 5 depict the variation of load carrying capacity with respect to the magnetization parameter m for various values of conductivity ϕ0 + ϕ1, aspect ratio m, the ratio b/h2 and porosity ψ. it is noticed that the load carrying capacity gets increased with increasing values of magnetization parameter m. it is also seen that the conductivity ϕ0+ϕ1 has an important role in improving the performance of the bearing system. the porosity ψ has a sharp adverse effect on the performance of the bearing system. figure 3 indicates that the load carrying capacity increases substantially with the increasing values of the aspect ratio m. besides, the load carrying capacity decreases with the increasing values of the ratio b/h2. in addition, the combined effect of magnetization m and the ratio b/h2 is more sharp as compared to the other combinations. 8 0.30 0.80 1.30 1.80 2.30 2.80 0.00 1.00 2.00 3.00 4.00 l oa d 01 m=0.25 m=0.50 m=0.75 m=1.00 m=1.25 figure 6. variation of load carrying capacity with respect to 0+1 and m. 0.30 1.30 2.30 3.30 4.30 5.30 6.30 0.00 1.00 2.00 3.00 4.00 l oa d 01 b/h2=10 b/h2=20 b/h2=30 b/h2=40 b/h2=50 figure 7. variation of load carrying capacity with respect to 0+1 and b/h2. figure 6. variation of load carrying capacity with respect to ϕ0+ϕ1 and m. 8 0.30 0.80 1.30 1.80 2.30 2.80 0.00 1.00 2.00 3.00 4.00 l oa d 01 m=0.25 m=0.50 m=0.75 m=1.00 m=1.25 figure 6. variation of load carrying capacity with respect to 0+1 and m. 0.30 1.30 2.30 3.30 4.30 5.30 6.30 0.00 1.00 2.00 3.00 4.00 l oa d 01 b/h2=10 b/h2=20 b/h2=30 b/h2=40 b/h2=50 figure 7. variation of load carrying capacity with respect to 0+1 and b/h2. figure 7. variation of load carrying capacity with respect to ϕ0+ϕ1 and b/h2. issn: 2180-1053 vol. 7 no. 2 july december 2015 hydromagnetic short bearings 27 9 0.05 0.55 1.05 1.55 2.05 2.55 0.00 1.00 2.00 3.00 4.00 l oa d 01 0 0.01 0.05 0.1 0.5 figure 8. variation of load carrying capacity with respect to 0+1 and . 0.60 1.60 2.60 3.60 4.60 5.60 6.60 0.25 0.45 0.65 0.85 1.05 1.25 l oa d m b/h2=10 b/h2=20 b/h2=30 b/h2=40 b/h2=50 figure 9. variation of load carrying capacity with respect to m and b/h2. figure 8. variation of load carrying capacity with respect to ϕ0+ϕ1 and ψ. 9 0.05 0.55 1.05 1.55 2.05 2.55 0.00 1.00 2.00 3.00 4.00 l oa d 01 0 0.01 0.05 0.1 0.5 figure 8. variation of load carrying capacity with respect to 0+1 and . 0.60 1.60 2.60 3.60 4.60 5.60 6.60 0.25 0.45 0.65 0.85 1.05 1.25 l oa d m b/h2=10 b/h2=20 b/h2=30 b/h2=40 b/h2=50 figure 9. variation of load carrying capacity with respect to m and b/h2. figure 9. variation of load carrying capacity with respect to m and b/h2. 10 0.10 0.60 1.10 1.60 2.10 2.60 3.10 0.25 0.45 0.65 0.85 1.05 1.25 l oa d m 0 0.01 0.05 0.1 0.5 figure 10. variation of load carrying capacity with respect to m and . 0.10 1.10 2.10 3.10 4.10 5.10 6.10 7.10 10.00 20.00 30.00 40.00 50.00 l oa d b/h2 0 0.01 0.05 0.1 0.5 figure 11. variation of load carrying capacity with respect to b/h2 and . the effect of conductivity 0 + 1 on the load carrying capacity is presented in figures 6 to 8. these figures suggest that the conductivity 0 + 1 increases the load carrying capacity and this increase is more for smaller values of aspect ratio m, the ratio b/h2 and porosity . however, the load carrying capacity is substantially more in the case of the ratio b/h2. figures 9 and 10 deal with the effect of aspect ratio m on the variation of load carrying capacity. it is manifest that aspect ratio m increases the load carrying capacity and this increase is more in the case of the ratio b/h2. lastly, figure 11 says that the combined effect of the ratio b/h2 and porosity  is significantly adverse as the load carrying capacity is more decreased at the initial stages. a close scrutiny of these figures reveals that the negative effect of porosity  and the ratio b/h2 can be compensated up to certain extent by the positive effect of figure 10. variation of load carrying capacity with respect to m and ψ. issn: 2180-1053 vol. 7 no. 2 july december 2015 journal of mechanical engineering and technology 28 10 0.10 0.60 1.10 1.60 2.10 2.60 3.10 0.25 0.45 0.65 0.85 1.05 1.25 l oa d m 0 0.01 0.05 0.1 0.5 figure 10. variation of load carrying capacity with respect to m and . 0.10 1.10 2.10 3.10 4.10 5.10 6.10 7.10 10.00 20.00 30.00 40.00 50.00 l oa d b/h2 0 0.01 0.05 0.1 0.5 figure 11. variation of load carrying capacity with respect to b/h2 and . the effect of conductivity 0 + 1 on the load carrying capacity is presented in figures 6 to 8. these figures suggest that the conductivity 0 + 1 increases the load carrying capacity and this increase is more for smaller values of aspect ratio m, the ratio b/h2 and porosity . however, the load carrying capacity is substantially more in the case of the ratio b/h2. figures 9 and 10 deal with the effect of aspect ratio m on the variation of load carrying capacity. it is manifest that aspect ratio m increases the load carrying capacity and this increase is more in the case of the ratio b/h2. lastly, figure 11 says that the combined effect of the ratio b/h2 and porosity  is significantly adverse as the load carrying capacity is more decreased at the initial stages. a close scrutiny of these figures reveals that the negative effect of porosity  and the ratio b/h2 can be compensated up to certain extent by the positive effect of figure 11. variation of load carrying capacity with respect to b/h2 and ψ. the effect of conductivity ϕ0+ϕ1 on the load carrying capacity is presented in figures 6 to 8. these figures suggest that the conductivity ϕ0+ϕ1 increases the load carrying capacity and this increase is more for smaller values of aspect ratio m, the ratio b/h2 and porosity ψ. however, the load carrying capacity is substantially more in the case of the ratio b/h2. figures 9 and 10 deal with the effect of aspect ratio m on the variation of load carrying capacity. it is manifest that aspect ratio m increases the load carrying capacity and this increase is more in the case of the ratio b/h2. lastly, figure 11 says that the combined effect of the ratio b/h2 and porosity ψ is significantly adverse as the load carrying capacity is more decreased at the initial stages. a close scrutiny of these figures reveals that the negative effect of porosity ψ and the ratio b/h2 can be compensated up to certain extent by the positive effect of magnetization m and conductivity ϕ0+ϕ1 by choosing suitably the aspect ratio m. it is found that the increased load carrying capacity due to the conductivity ϕ0+ϕ1 gets further increased due to hydromagnetization. this is crucial for overcoming the negative effect of the ratio b/h2 and porosity ψ. a comparison of this investigation with the study of (patel et al., 2010) tends to suggest that the overall performance is relatively better here. as can be seen from equations (10) and (12) for friction at the both plates, the friction remains unaltered. issn: 2180-1053 vol. 7 no. 2 july december 2015 hydromagnetic short bearings 29 4.0 conclusions it is concluded that the effect of hydromagnetization is comparatively sharp unlike some of the previous studies (vadher et al., 2008), (patel et al., (2010). the analysis incorporated here modifies and extends the earlier analysis concerning the performance of a magnetic fluid based squeeze film in a short bearing and also presents at least an additional degree of freedom to compensate the adverse effect of porosity. furthermore, this investigation offers some scopes for the extension of the life period of the bearing system through the observations that the bearing with a magnetic fluid can support a load even when there is no flow unlike the case of a conventional lubricant. nomenclature 11 magnetization m and conductivity 0 + 1 by choosing suitably the aspect ratio m. it is found that the increased load carrying capacity due to the conductivity 0 + 1 gets further increased due to hydromagnetization. this is crucial for overcoming the negative effect of the ratio b/h2 and porosity . a comparison of this investigation with the study of (patel et al., 2010) tends to suggest that the overall performance is relatively better here. as can be seen from equations (10) and (12) for friction at the both plates, the friction remains unaltered. 4.0 conclusions it is concluded that the effect of hydromagnetization is comparatively sharp unlike some of the previous studies (vadher et al., 2008), (patel et al., (2010). the analysis incorporated here modifies and extends the earlier analysis concerning the performance of a magnetic fluid based squeeze film in a short bearing and also presents at least an additional degree of freedom to compensate the adverse effect of porosity. furthermore, this investigation offers some scopes for the extension of the life period of the bearing system through the observations that the bearing with a magnetic fluid can support a load even when there is no flow unlike the case of a conventional lubricant. nomenclature h fluid film thickness at any point (mm) h1 maximum film thickness (mm) h2 minimum film thickness (mm) b breadth of the bearing (mm) l length of the bearing (mm) m aspect ratio u uniform velocity in x – direction p lubricant pressure (n/mm2) p dimensionless pressure w load carrying capacity (n) w non-dimensional load carrying capacity  lubricant viscosity (n.s/mm2)  shear stress (n/mm2) τ dimensionless shear stress f frictional force (n/mm2) f dimensionless frictional force 0f dimensionless frictional force (at moving plate) 1f dimensionless frictional force (at fixed plate) s electrical conductivity of the lubricant m = 1/2 0 μ s hb       = hartmann number k permeability (col2kgm/s2) h0 thickness of the porous facing m* porosity of the porous matrix 12 references abramovitz s. (1955). theory for a slider bearing with a convex pad surface; side flow neglected. journal of franklin institute, 259(3). 221-233. archibald f. r. (1950). a simple hydrodynamic thrust bearing. transactions of asme, 72.393. bagci c. and singh a. p. (1983). hydrodynamic lubrication of finite slider bearing: effect of one dimensional film shape and their computer aided optimum designs. journal of lubrication technology. asme, 105.48-66. basu s. k.. sengupta s. n. and ahuja b. b. (2005). fundamentals of tribology. new delhi: prentice-hall of india private limited. cameron a. (1966). the principles of lubrication. london: longmans. charnes a. and saibel e. (1952). on the solution of the reynolds’ equation for slider bearing lubricationpart 1. transactions of asme. 74.867. elco r.a. and huges w.f. (1962). magnetohydrodynamic pressurization in liquid metal lubrication wear,5. 198-207. gross w.a., matsch lee a., castelli v., eshel a., vohr j. h. and wildmann m. (1980). fluid film lubrication. new york: a wiley-interscience publication. john wiley and sons. hamrock b. j. (1994). fundamentals of fluid film lubrication. new york: mcgrawhill. inc.  = 3 0 h h*m = porosity b0 uniform transverse magnetic field applied between the plates. c2 = *mh km 1 2 2  ' 0h surface width of the lower plate (m) ' 1h surface width of the upper plate (m) s0 electrical conductivity of lower surface (mho) s1 electrical conductivity of upper surface (mho) 0(h) = sh hs '00 = electrical permeability of the lower surface 1(h) = sh hs '11 = electrical permeability of the upper surface issn: 2180-1053 vol. 7 no. 2 july december 2015 journal of mechanical engineering and technology 30 12 references abramovitz s. (1955). theory for a slider bearing with a convex pad surface; side flow neglected. journal of franklin institute, 259(3). 221-233. archibald f. r. (1950). a simple hydrodynamic thrust bearing. transactions of asme, 72.393. bagci c. and singh a. p. (1983). hydrodynamic lubrication of finite slider bearing: effect of one dimensional film shape and their computer aided optimum designs. journal of lubrication technology. asme, 105.48-66. basu s. k.. sengupta s. n. and ahuja b. b. (2005). fundamentals of tribology. new delhi: prentice-hall of india private limited. cameron a. (1966). the principles of lubrication. london: longmans. charnes a. and saibel e. (1952). on the solution of the reynolds’ equation for slider bearing lubricationpart 1. transactions of asme. 74.867. elco r.a. and huges w.f. (1962). magnetohydrodynamic pressurization in liquid metal lubrication wear,5. 198-207. gross w.a., matsch lee a., castelli v., eshel a., vohr j. h. and wildmann m. (1980). fluid film lubrication. new york: a wiley-interscience publication. john wiley and sons. hamrock b. j. (1994). fundamentals of fluid film lubrication. new york: mcgrawhill. inc.  = 3 0 h h*m = porosity b0 uniform transverse magnetic field applied between the plates. c2 = *mh km 1 2 2  ' 0h surface width of the lower plate (m) ' 1h surface width of the upper plate (m) s0 electrical conductivity of lower surface (mho) s1 electrical conductivity of upper surface (mho) 0(h) = sh hs '00 = electrical permeability of the lower surface 1(h) = sh hs '11 = electrical permeability of the upper surface references abramovitz s. (1955). theory for a slider bearing with a convex pad surface; side flow neglected. journal of franklin institute, 259(3). 221-233. archibald f. r. (1950). a simple hydrodynamic thrust bearing. transactions of asme, 72.393. bagci c. and singh a. p. (1983). hydrodynamic lubrication of finite slider bearing: effect of one dimensional film shape and their computer aided optimum designs. journal of lubrication technology. asme, 105.48-66. basu s. k.. sengupta s. n. and ahuja b. b. (2005). fundamentals of tribology. new delhi: prentice-hall of india private limited. cameron a. (1966). the principles of lubrication. london: longmans. charnes a. and saibel e. (1952). on the solution of the reynolds’ equation for slider bearing lubricationpart 1. transactions of asme. 74.867. elco r.a. and huges w.f. (1962). magnetohydrodynamic pressurization in liquid metal lubrication. wear,5. 198-207. gross w.a., matsch lee a., castelli v., eshel a., vohr j. h. and wildmann m. (1980). fluid film lubrication. new york: a wiley-interscience publication. john wiley and sons. hamrock b. j. (1994). fundamentals of fluid film lubrication. new york: mcgraw-hill. inc. kuzma d.c. (1964). magnetohydrodynamic squeeze films. journal of basic engineering. asme, 86. 441-444. kuzma d.c., maki e.r. and donnelly r.j. (1964). the magnetohydrodynamic squeeze films. journal of fluid mechanics, 19. 395-400. issn: 2180-1053 vol. 7 no. 2 july december 2015 hydromagnetic short bearings 31 lord rayleigh. (1918) notes on the theory of lubrication. philosophical magazine. 35.1-12. mcallister g. t.. rohde s. m. and mcallister m. w. (1980). a note on the optimum design of slider bearing. journal of lubrication technology. asme, 102(1). morgan v. t. and cameron a. (1957). mechanism of lubrication in porous metal bearings. proceeding of conference on the lubrication and wear. imeche. london. 89.151-157. osterle f. and saibel e. (1958). the effect of bearing deformation in slider bearing lubrication. journal of lubrication technology. asme. 213-216. patel k. c. and gupta j. l. (1983). hydrodynamic lubrication of a porous slider bearing with slip velocity. wear, 85. 309-317. patel k. c. and gupta j. l. (1979). behavior of hydromagnetic squeeze film between porous plates. wear, 56 . 327-339. patel r. m.. deheri g. m. and vadher p. a. (2010). performance of a magnetic fluid based short bearing. acta polytechnica hungarica journal of alied sciences. 7(3).63-78. patel r.m. and deheri g.m. (2004). magnetic fluid based squeeze film behaviour between annular plates and surface roughness effect. aimeta international tribology conference. rome. italy. 631-638. patel k.c. and hingu j.v.(1978). hydromagnetic squeeze film behaviour in porous circular disks. wear, 49. 239-246. pinkus o. and sternlicht b. (1961). theory of hydrodynamic lubrication. new york : mcgraw-hill. prajapati b.l. (1995). on certain theoretical studies in hydrodynamics and electro magnetohydrodynamic lubrication. doctor of philosophy thesis. sardar patel university. india. prakash j. and vij s. k. (1973). hydrodynamic lubrication of porous slider. journal of mechanical engineering and science. 15.232-234. rouleau w. t. (1963). hydrodynamic lubrication of narrow press-fitted porous metal bearings. asme journal of lubrication technology. 85.123. snyder w.t. (1962). the magnetohydrodynamic slider bearing. journal of basic engineering. asme, 84. 197-204. shukla j.b. (1963). principles of hydromagnetic lubrication. journal of the physical society of japan. 18(7). 1086-1088. shukla j.b. and prasad r. (1965). hydromagnetic squeeze films between two conducting surfaces. journal of basic engineering. asme, 87. 818-822. issn: 2180-1053 vol. 7 no. 2 july december 2015 journal of mechanical engineering and technology 32 sinha p.c. and gupta j.l. (1974). hydromagnetic squeeze films between porous annular disks. journal of mathematical and physical science, 8.413-422. vadher p. a., vinodkumar p. c., deheri g. m. and patel r. m. (2008). a study on the behavior of hydromagnetic squeeze films between two conducting rough porous annular plates. proceedings of the pakistan academy of sciences. 45(2).81-95. issn: 2180-1053 vol. 8 no.2 july – december 2016 1 elements of maintenance systems and tools for implementation within the framework of reliability centred maintenancea review ikuobase emovon1, rosemary a. norman2 and alan j. murphy2 1 department of marine engineering, federal university of petroleum resources, effurun, nigeria 2 school of marine science and technology, newcastle university, newcastle upon tyne, ne1 7ru, uk abstract for plant systems to remain reliable and safe they must be effectively maintained through a sound maintenance management system. the three major elements of maintenance management systems are; risk assessment, maintenance strategy selection and maintenance task interval determination. the implementation of these elements will generally determine the level of plant system safety and reliability. reliability centred maintenance (rcm) is one method that can be used to optimise maintenance management systems. this paper discusses the three major elements of a maintenance system, tools utilised within the framework of rcm for performing these tasks and some of the limitations of the various tools. each of the three elements of the maintenance management system has been considered in turn. the information will equip maintenance practitioners with basic knowledge of tools for maintenance optimisation and stimulate researchers with respect to developing alternative tools for application to plant systems for improved safety and reliability. the research findings revealed that there is a need for researchers to develop alternative tools within the framework of rcm which are efficient in terms of processing and avoid the limitations of existing methodologies in order to have a safer and more reliable plant system. keywords: plant systems; reliability centred maintenance; risk assessment; maintenance strategy selection. 1.0 introduction dhillon (2002) defined maintenance as the combination of activities undertaken to restore a component or machine to a state in which it can continue to perform its designated functions. maintenance usually involves repair in the event of a failure (a corrective action) or a preventive action. on the other hand the british standard defines maintenance as (bs 1993) “the combination of all technical and administrative actions, *corresponding author e-mail: ikuoy2k@yahoo.com journal of mechanical engineering and technology 2 issn: 2180-1053 vol. 8 no.2 july – december 2016 intended to retain an item in, or restore it to, a state in which it can perform a required action”. the costs incurred in this are normally a major percentage of the total operating cost in most industries including the maritime sector. vavra (2007) reported that wasted energy as a result of poorly maintained compressed air systems collectively cost us industry up to $3.2 billion annually. this can be attributed to the general perception in the past that maintenance is an evil that plant managers cannot do without and that it is impossible to minimise maintenance cost (mobley, 2004). however in order to minimize cost of plant system maintenance without compromising the system safety and reliability there is a need to have an effective maintenance system in place. there are three major elements that make up a maintenance system; risk assessment, maintenance strategy selection and maintenance task interval determination. these elements must be performed optimally in the maintenance management of a plant system in order to have a safe and reliable system at reasonable cost. different maintenance methodologies have been applied in optimizing these elements of maintenance. one of the most notable is reliability centered maintenance (rcm). within this maintenance framework, different tools/methods are used to perform these three elements of a maintenance system. the paper discusses an overview the rcm methodology, tools utilised within the framework in carrying out the three major elements of a maintenance system, advances and associated limitations. the rest of the paper is organized as follows: an overview of rcm is presented in section 2. this is then followed by a discussion of the three elements of a maintenance system in turn in sections 3, 4 and 5. finally the conclusion is presented in section 6. 2.0 rcm overview moubray (1991) defined rcm as “a process used to determine what must be done to ensure that any physical asset continues to function in order to fulfil its intended functions in its present operating context.” from this definition it is obvious that rcm focuses, not on the system hardware itself rather, on the system function. maintenance practitioners are faced with challenges with respect to maintaining their asset and some of these challenges are; difficulty in selecting the most appropriate maintenance strategy for each equipment item, difficulty in prioritizing the risk of component failure modes of the system, difficulty in ascertaining the most cost effective approach and difficulty in getting the best support from the workforce. all of these challenges are addressed by the rcm frame-work in a systematic manner. in fact, moubray (1991) categorically stated that no maintenance technique has proven to be more successful in preserving the function of a system than rcm. the development of the rcm technique can be traced to the aviation industry where the maintenance steering groups (msg) formed within the industry developed a maintenance methodology which was reported in three documents referred to as msg1 msg2 and msg3, released in the years 1968, 1970 and 1980 respectively (dhillon, 2002). this technique evolved into classical rcm which has since been embraced by industries ranging from manufacturing to the marine sectors and has proven to be successful in all these industries. elements of maintenance system and tools for implementation within the framework of reliability centred maintenancea review issn: 2180-1053 vol. 8 no.2 july – december 2016 3 the first step to the successful implementation of the rcm technique is to ask seven basic questions about the asset that the methodology is intended to be applied on. these seven questions are as follows, (moubray, 1991): 1) what are the intended functions and performance standards of the asset or machinery in its current operating situation? 2) how does it fail to fulfil these intended functions? 3) what are the causes of each failure? 4) what are the corresponding consequences? 5) in what way does each failure matter? 6) what task should be performed in order to avert each failure? 7) what should be done if no preventive task is found to be applicable? the basic steps of the rcm analysis are reviewed as follows (rausand & vatn, 1998, dhillon, 2002, selvik & aven, 2011): (1) preparatory stage: rcm is generally performed by a team and, as such, the first step in the rcm analysis is to set up the rcm team. the team should consist of experts with adequate knowledge of the system to be investigated. generally the team should have a minimum of one person each from the maintenance and the operational units. the team should also have a member with a vast knowledge of the rcm methodology and who could serve as the facilitator. the rcm team have the responsibility for determining; the scope of the study, the level of the assembly to be investigated (i.e. plant, system, sub-system) and the equipment or asset to be investigated. they also have the responsibility, among others, of data gathering for the analysis. (2) maintenance significant items (msi) identification: failure mode and effect analysis (fmea) is generally applied here in determining the maintenance significant items. fmea methodology is discussed in detail in sections 3.1.1 and 3.1.2 below. these items are then used to populate the rcm decision diagram in order to determine the most appropriate maintenance task. for a very simple system, msi can easily be identified without resorting to any specialized tools. for the non-msi items, the items are generally allowed to fail before repair or replacement can be implemented. however for the msi items, preventive maintenance tasks are usually more appropriate but in some cases they are allowed to fail before repair or replacement activities are performed and these are dependent on the msi item failure characteristics, the impact of the failure and maintenance costs. (3) maintenance strategy classification: the maintenance strategy for addressing crucial failure modes of the critical components of an asset have been classified in different ways. rausand and vatn (1998) considered five distinct maintenance strategies namely continuous predictive maintenance, scheduled predictive maintenance, scheduled overhaul, scheduled replacement and scheduled function testing for preventing or mitigating the effects of failures. dhillon (2002) presented the following four maintenance strategies for use in the rcm methodology; reactive maintenance, preventive maintenance, predictive testing and inspection and proactive maintenance. nevertheless both the five maintenance strategy types considered by rausand and vatn (1998) and the four maintenance strategies considered by dhillon (2002) fall within the journal of mechanical engineering and technology 4 issn: 2180-1053 vol. 8 no.2 july – december 2016 three basic main maintenance strategies: corrective maintenance, preventive maintenance and condition/predictive maintenance. (4) maintenance task selection: here the rcm logic is designed and applied in selecting the appropriate maintenance task for the crucial failure mode of each of the critical components of the asset. the rcm logic is expressed in decision diagram form which, through a series of yes and no questions, enables the rcm facilitator to arrive at an optimal maintenance strategy for the particular failure mode/component in question. there are various versions of the decision rcm logic tree and a sample is shown in figure 1. however all of the versions are based on the basic decision criteria of the rcm for selecting the maintenance task which are; cost effectiveness, applicability and failure characteristics. the term applicability, with respect to selecting the maintenance task, means a maintenance preventive task that is capable of mitigating or preventing failure and in the case of a potentially hidden failure is capable of discovering it. the term cost-effectiveness is a decision criterion for determining the maintenance task, from different alternatives, that is the most cost effective. if there is no applicable preventive maintenance task available, then the only alternative is to select run– to– failure. in the case of cost effectiveness; the cost of the applicable preventive maintenance task to mitigate or prevent failure must be greater than the aggregate cost related to the failure itself, otherwise run–to–failure will be more appropriate except with a safety-related issue or a failure situation where redesign may be compulsory. figure 1. a sample of rcm logic adapted from (rausand and vatn, 1998) elements of maintenance system and tools for implementation within the framework of reliability centred maintenancea review issn: 2180-1053 vol. 8 no.2 july – december 2016 5 (5) maintenance planning: here the optimal intervals are determined for the preventive maintenance tasks assigned for the crucial failure modes of the critical components of the asset. some of the failure modes are assigned scheduled predictive maintenance and some scheduled overhaul, etc. using the rcm logic. the process of determining the interval for a preventive maintenance task is, in many instances, very challenging and, in general, mathematical models are applied in obtaining these intervals. however in some cases mathematical models are not applied and preventive maintenance task intervals are not optimized but are determined based on experts’ opinions, operational experience and manufacturers’ recommendation. it is worth mentioning that in the traditional rcm process there is no provision for tools for use in the determination of preventive maintenance task intervals. (6) implementation and update: here the managerial procedures, with respect to how the results of the rcm analysis that is performed by the rcm team are applied, is described. this step includes among others; communication of the rcm analysis results from the rcm team to the management, results documentation and undertaking updating from time to time which is generally subject to availability of new, relevant data. from the rcm discussion it can be seen that there are three key elements of maintenance that the methodology is used to optimize; (1) risk assessment, (2) maintenance strategy selection from different alternatives, and (3) maintenance task interval determination. the three elements of a maintenance system and the advances of the tools utilised for performing them within the frame work of rcm together with the limitations of the tools, are discussed next. 3.0 risk assessment the american bureau of shipping (2000) defined risk as the product of the probability of the occurrence of a failure and consequence of the failure. while risk assessment, according to cross and ballesio (2003), is defined as being a systematic method that combines diverse aspects of design and operation in assessing risk. arendt (1990) described risk assessment as activities involving hazard identification, chances of the occurrence of failure estimation and the consequences of the failure estimation. with the advent of risk-based inspection and maintenance in the 1990s in conjuction with reliability maintenance, risk assessment has gained popularity in the maintenance world and it is worth noting that risk assessment is clearly the most critical phase of risk-based maintenance since maintenance decisions to be taken will be based on the assessed level of risk (arunraj & maiti, 2007). risk assessment is also a very important aspect of reliability-centred maintenance (rcm) though rcm is mainly intended for preserving the reliability of plant equipment and systems. the risk assessment in the rcm process involves assessing the risk of failure of equipment items and, based on the assessed risk, an appropriate maintenance strategy will be recommended. however the acceptable level of risk must be defined, possibly through a retrospective study of earlier successful items etc. journal of mechanical engineering and technology 6 issn: 2180-1053 vol. 8 no.2 july – december 2016 some of the factors that affect the quality of the output from a risk analysis are; data sources, human factors, methods and tools for performing the analysis itself, and the ability and experience of the analyst. 3.1 risk assessment tool an analyst has the option of choosing from a variety of tools for performing risk analysis in each of the three major phases of risk assessment; hazard identification, risk estimation and risk evaluation. the commonly used tool/method within the framework of rcm for prioritising risk is fmea. 3.1.1 fmea siddiqui and ben-daya (2009) defined failure mode and effect analysis (fmea) “as a systematic failure analysis technique that is used to identify the failure modes, their causes and consequently their fallouts on the system function”. the methodology was developed by the united states army in 1947 and in the 1970s industries such as the automotive, aerospace and manufacturing embraced the use of the technique in the maintenance of their asset (scipioni et al., 2002). nowadays fmea is a popular risk assessment tool in the production of hardware such as mechanical and electronic components (scipioni et al., 2002). the technique has also become a popular tool for performing risk assessment for ship systems. when fmea is combined with criticality analysis (ca) it is referred to as failure mode effect and criticality analysis (fmeca) and its essence is to rank the impact of each of the failure modes for the various components that make up the entire system (headquarters department of the army, 2006, sachdeva et al., 2009a). according to ben-daya (2009) fmea basically performs three functions. these are: (1) to identify and recognize potential failures including their causes and effects. (2) to evaluate and prioritize identified failure modes. (3) to identify and suggest actions to either eliminate or reduce the chance of the potential failures from occurring. the technique can be applied to any well-defined system but it is best suited to the risk assessment of mechanical and electrical systems (e.g. fire suppression systems, propulsion systems) and the approach can either be quantitative or qualitative, (american bureau of shipping, 2000, headquarters department of the army, 2006). the availability or non-availability of failure data will determine, to a large extent, the approach that is used in carrying out fmea risk assessment. a quantitative approach is used when variables such as failure rate (λi), failure mode ratios (αi ), failure effect probability (βi ) and its operating time (t) are known and are used to generate the criticality number (cn) which can then be used to rank the ith failure mode (headquarters department of the army, 2006, braglia, 2000). this can be represented mathematically as: cni = αi x βi x λi x t (1) elements of maintenance system and tools for implementation within the framework of reliability centred maintenancea review issn: 2180-1053 vol. 8 no.2 july – december 2016 7 in applying the qualitative method each failure mode is rated or ranked by developing a risk priority number (rpn) which is computed by multiplying the severity rating (s) by both the occurrence probability (o) and the detection rating (d): rpn = s x o x d (2) qualitative terms are used to determine these three parameters, usually on a numerical scale of 1 to 10 having been determined based on collective expert opinion (sachdeva et al., 2009b, siddiqui & ben-daya, 2009, ling et al., 2012, kahrobaee & asgarpoor, 2011, zammori & gabbrielli, 2012, braglia, 2000). typically values are assigned to o, s and d by a team of experts using an ordinal scale, an example of which is shown in table 1. in performing fmea for any assets a series of steps are followed and are represented diagrammatically in figure 2. 3.1.2 fmea based on mcdm technique multi-criteria decision making (mcdm) tools have been applied in literature to an extent in addressing some of the challenges of the conventional fmea. this is due to their ability to judge different alternatives based on certain decision criteria. braglia (2000) utilised the analytic hierarchy process (ahp) technique in aggregating the decision criteria (o, s and d) in the conventional fmea system together with an economic cost criterion in prioritising possible causes of failure of a refrigerator manufacturing plant. the decision problem was structured in a three-level hierarchy, with the top level signifying the goal, the intermediate level signifying the four decision criteria; o, s, d and economic cost and the bottom level signifying the alternative causes of failure of the plant. pairwise comparison judgements were obtained and evaluated to produce weights of decision criteria and local priorities of possible causes of failure for every decision criterion. the decision criteria weights were then synthesized with the local priorities of causes of failure to produce overall weights of the possible causes of failure. carmignani (2009) used the braglia (2000) methodology in prioritising the risk of failure modes of an electro-injector, a fuel system component. the author however developed a new mathematical model for evaluating the economic cost criterion. however the use of ahp has been criticised due to its use of an unbalanced scale of judgement and its inadequacy in addressing risk criteria rating that may be uncertain and imprecise in the pairwise comparison process (deng, 1999, ilangkumaran & kumanan, 2009). additionally, the use of ahp methodology limits risk prioritisation to the use of a maximum of 15 decision criteria in order to reduce the number of pairwise comparison judgements and evaluation complexity (vidal et al., 2011). journal of mechanical engineering and technology 8 issn: 2180-1053 vol. 8 no.2 july – december 2016 table 1. ratings for occurrence (o), severity (s) and detectability (d) in a marine engine system, adapted from (yang et al., 2011, pillay & wang, 2003, cicek & celik, 2013, emovon et al., 2015) issn: 2180-1053 vol. 8 no.2 july – december 2016 9 figure 2. fmea methodology, adapted from (cicek & celik, 2013, emovon et al., 2015) journal of mechanical engineering and technology 10 issn: 2180-1053 vol. 8 no.2 july– december 2016 maheswaran and loganathan (2013) proposed an integrated ahp and preference ranking organisation method for enrichment evaluation (promethee) as an alternative to rpn in the traditional fmea system for prioritising failure modes of a boiler system in the tyre manufacturing industry. the ahp was applied in determining weights of decision criteria while promethee was used in the ranking of risk of failure modes of the system. other authors have also used promethee for prioritising risk of failure modes. ayadi et al., (2013) applied promethee for prioritising risk of failure modes of a gas treatment plant. moreira et al., (2009) utilised promethee in the ranking of failure modes of equipment items of a power transformer. the main limitation of the promethee technique is that it results in poor structuring of decision problems and when more than seven decision criteria are used, it becomes difficult to have a clear view of the problem thereby making the evaluation process very complex (macharis et al., 2004). seyed-hosseini et al., (2006) proposed decision making trial and evaluation laboratory (dematel) as an alternative tool to the rpn of the conventional fmea for prioritisation of risk of failure modes. with this approach alternative failure modes are prioritised based on severity of effect and direct/indirect relationships between them. however the major demerit of the approach is that a lot of computational effort is required. furthermore, the technique cannot address the limitations of the conventional rpn method in systems where each cause of failure is linked to a single failure mode and for such systems the results obtained by both methods are the same (shaghaghi & rezaie, 2012). sachdeva et al., (2009b) proposed the technique for order preference by similarity to ideal solution (topsis) as an alternative to the rpn of the conventional fmea for risk prioritisation. the author applied six decision criteria of o, d, maintainability, spare parts availability, economic safety and economic cost upon which the risk of failure modes were ranked. a case study of the digester of a paper manufacturing plant in india was used to demonstrate the applicability of the method. braglia et al., (2003) used topsis under a fuzzy environment for risk prioritisation of a foaming machine of a refrigerator production line. emovon et al., (2014) proposed a technique referred to as avtopsis for risk prioritisation of marine machinery systems. the approach is based on a combination of an averaging technique with topsis. the technique allows the use of imprecise information from experts in the decision making process and that is made possible with the averaging technique which aggregates the information and the result is used as input to the topsis methodology which executes the ranking of the failure modes of the system. the technique was demonstrated with a case study of the marine diesel engine of a ship. the topsis technique basic concept is that the best alternative is the one closest to the positive ideal solution and farthest from the negative ideal solution. one major limitation of the technique is the lack of measure of the relative distance between positive ideal and negative ideal in the evaluation process which seems to negatively affect the outcome. in a similar study emovon et al., (2015) proposed an integrated vikor and compromise programming (cp) method with averaging technique as an alternative to the standard rpn calculation of the fmea system for prioritising risk of failure mode of marine machinery systems. while the averaging technique was use for imprecise data aggregation, the cp and vikor methods were used for ranking of risk of failure modes. elements of maintenance systems and tools for implementation within the framework of reliability centred maintenancea review issn: 2180-1053 vol. 8 no.2 july – december 2016 11 4.0 maintenance strategy selection one of the main challenges of maintenance management is the selection of the appropriate maintenance strategy for each equipment item in the system because not all maintenance strategies are applicable and cost effective for different components. hence choosing the right maintenance strategy for the system will help maintain a balance between the system availability and cost of performing the maintenance. however when choosing the type of maintenance strategy for plant systems, several conflicting decision criteria must be taken into consideration such as cost, reliability, availability and safety. these make maintenance strategy selection analysis critical and complex and the investigation fundamental and justifiable (bevilacqua & braglia, 2000). despite the significance of the subject, only a few studies have dealt with the maintainance selection policy problem (bertolini & bevilacqua, 2006). there are different maintenance strategies that can be used for mitigating the different failure modes of a plant system. generally there are three types of maintenance strategy that are available for maintenance practitioners to choose from. the three maintenance strategies and a review of the methods utilised for the selection of the optimum strategy for each of the different component/failure mode of the system are discussed next. 4.1 maintenance strategies according to pintelon et al., (2006) a maintenance strategy is generally viewed from the perspective of maintenance policies such as breakdown maintenance, preventive maintenance and predictive maintenance and sometimes rcm or tpm. it is worth noting that the maintenance strategy is one of the most influential factor affecting the effectiveness of a maintenance system (stanojevic et al., 2000, stanojevic et al., 2004) and the process of estimating the optimal combination of maintenance strategies for different plant system equipment items is a very hard and complex task as the maintenance program must combine both technical and management requirements (sachdeva et al., 2009b, bertolini and bevilacqua, 2006, bevilacqua et al., 2000). the selections usually require a vast amount of information relating to the following decision criteria (bertolini & bevilacqua, 2006): manpower utilization, cost and budget constraints, safety factors, environment factors and mean time between failure (mtbf) for each piece of equipment. 4.1.1 run-to-failure the rationale of the run-to-failure management approach is simple and straightforward. when an equipment item fails it is fixed. that is to say equipment is allowed to fail before any maintenance (repair or replacement) is carried out and, as such, resources are not deployed until equipment breaks down. it is, in fact, a no-maintenance approach to maintenance management of an asset and it is generally the least cost effective technique of maintenance management, since the maintenance costs are higher and plant availability is lower. in fact maintenance cost analysis revealed that repair carried out in reactive mode is nearly three times higher in cost than that carried out in preventative mode (mobley, 2001). journal of mechanical engineering and technology 12 issn: 2180-1053 vol. 8 no.2 july– december 2016 this type of maintenance is usually effective for non-critical and low cost components and equipment in a system (pride, 2008). for the plant manager to know that a component is non-critical, criticality analysis is carried out and, based on the result, an appropriate maintenance strategy is recommended for the plant equipment. 4.1.2 preventive maintenance preventive maintenance (pm) is defined as maintenance actions performed on plant systems at a definite interval with the aim of preventing wear and functional degradation, extending the useful life and mitigating the risk of catastrophic failure (sullivan et al., 2004) and it concerns itself with such activities as the replacement and renewal of components, inspections, testing and checking of working parts during their operation (ebrahimipour et al., 2015). in utilising this approach for maintenance management, equipment repairs or replacement are performed at pre-established intervals. the length of the intervals is usually based on equipment items’ industrial average-life such as mean time between failures (mtbf). however some plant managers rely on machine or component manufacturer’s recommendation to schedule preventive maintenance activities. for the shipping industry, imo in 1993 set the foundation for preventive maintenance implementation by releasing the international safety management (ism) code (imo 1993). chapter 10 of it clearly states the procedure and the duties of the shipping industry for preventive maintenance implementation in such a way that international regulations are adhered to strictly. the major merit of pm is its ability to increase the average life of equipment items and to reduce the risk of catastrophic failure (sullivan et al., 2004). however despite the numerous benefits of pm, the major limitation is that it often results in unnecessary repair or replacement. another limitation is the difficulty in evaluating the optimum interval of performing the maintenance task as this may take years of data collection and analysis (chen, 1997). the time based preventive maintenance approach can further be divided into two categories as follows: (1) scheduled overhaul: in this type of time based preventive maintenance, equipment overhaul or repair is performed on a definite time interval. the strategy is suitable to equipment with identifiable age when failure rate function rapidly increases and large units of the equipment can survive to that age. furthermore, where reworking can restore the equipment to its normal operating condition (rausand, 1998). (2) scheduled replacement: the application of this type of time based preventive maintenance approach, requires an equipment item or a unit of it being replaced at a specific time interval. this strategy is generally best for equipment that is exposed to critical failure and where the majority of the items of the equipment must survive to the minimum replacement time. additionally, the equipment failure type must be of prime economic consequences (rausand, 1998). elements of maintenance systems and tools for implementation within the framework of reliability centred maintenancea review issn: 2180-1053 vol. 8 no.2 july – december 2016 13 4.1.3 condition based maintenance this refers to the maintenance strategy in which the condition of an equipment item is monitored in order to detect potential failure and recommend appropriate corrective action. basically there are two types of condition based maintenance (cbm); the continuous on-condition task and the scheduled on-condition task (rausand and vatn, 1998). the continuous on-condition task is the approach where equipment item condition is monitored uninterruptedly using diagnostics devices. the main shortcoming of this approach is that it is expensive (jardine et al., 2006). the scheduled on-condition task is a cbm strategy in which the condition of an equipment item is monitored at regular time intervals with the objective of detecting potential failure (rausand and vatn, 1998). the check carried out on an equipment item is implemented by maintenance practitioners or operators with or without the use of diagnostic tools. this approach is nowadays more commonly used by most industries than the continuous oncondition task because it’s less expensive and yet effective. however the main challenge of the scheduled on-condition task is the difficulty in determining the appropriate interval for carrying out the task (jardine et al., 2006). in designing a condition monitoring program for ship systems the general procedure to follow has been put in place by bsi/iso 17359 (2003). the standard includes procedures for equipment auditing, criticality assessment and overview of the condition monitoring procedure and the determination of the maintenance action to be used. the technique for scheduling maintenance tasks is the major difference between time based preventive maintenance and condition based maintenance. while the time based preventive maintenance activity is scheduled based on average life evaluated using historical data of the particular piece of equipment, the condition based maintenance activity is scheduled in response to a performance degradation observed from diagnostic device readings and/or human sensing which deviate from standard equipment operating conditions (noemi & william, 1994). 4.2 maintenance strategy selection methods the reliability centered maintenance (rcm) technique has been used extensively for maintenance strategy selection (bevilacqua & braglia, 2000, mohan et al., 2004). “rcm represents a method for preserving functional integrity and it is designed to minimise overall maintenance costs by balancing the higher cost of corrective maintenance against the cost of preventive maintenance” (crocker & kumar, 2000). within the rcm framework the rcm logic diagram is the tool used for selecting the most appropriate maintenance strategy for different failure modes of a system (conachey, 2005, american bureau of shipping, 2004). however the use of rcm is a very time consuming exercise (waeyenbergh & pintelon, 2004) and this may be attributed to the excessive time that may elapsed for decision makers or maintenance practitioners to reach a consensus decision on every failure mode. furthermore, the use of the rcm logic tree does not allow for ranking of maintenance strategy alternatives such that the optimum solution can easily be determined. journal of mechanical engineering and technology 14 issn: 2180-1053 vol. 8 no.2 july– december 2016 the use of different multi-criteria decision making (mcdm) tools such as ahp, analytical network process (anp) and topsis has been reported in literature for addressing maintenance strategy selection problems for various industries. these techniques have either been applied individually or in combination with one another or they have been integrated with other tools such as fuzzy set theory and mathematical programming. bevilacqua and braglia (2000) used ahp in conjunction with failure mode effect and criticality analysis (fmeca) principles to select the optimum maintenance strategy for each equipment item of an integrated gasification and combined cycle plant. the five possible maintenance strategy alternatives considered were; preventive, predictive, condition-based, corrective and opportunistic maintenance. goossens and basten (2015) used ahp in the selection of the optimum maintenance strategy for naval ship systems. the authors involved three different groups in the ranking of three maintenance strategies; corrective, time/use-based maintenance and condition based maintenance based on some decision criteria. the different groups within the maritime industry from which pairwise comparison judgements were obtained for the prioritisation of maintenance strategy alternatives were; shipbuilders, the owners/operators and the original equipment manufacturers (oem). the authors structured the decision problem in five levels. the first level (top) representing the goal (best maintenance strategy for naval ship), the second, third and fourth levels, representing decision criteria, consisted of two, eight and 29 decision criteria respectively while the fifth level (bottom) representing the three maintenance strategy alternatives. the optimum maintenance strategy as determined based on data from shipbuilder, owner/operator and the oem was condition based maintenance. resobowo et al., (2014) presented the ahp technique in the ranking of the factors that affect military ship maintenance management. the factors the authors considered are; cost, availability, reliability, safety, human resource, operations, types of ship and ship characteristics. these factors were prioritised based on three decision criteria consisting of planned maintenance, preventive maintenance and routine maintenance. from the analysis, human resource was considered as the most important factor that affects military ship maintenance management. other examples of the use of ahp for maintenance strategy selection are: triantaphyllou et al. (1997) presented an ahp technique for the selection of an optimum maintenance strategy taking into consideration four maintenance decision criteria, nyström and söderholm (2010) proposed a procedure based on ahp for prioritising diverse maintenance strategies in railway infrastructure, and labib et al. (1998) developed a methodology based on ahp for selecting the optimum maintenance strategy for an integrated manufacturing system. the limitations of ahp in addressing multiple criteria decision problems have been described in section 3.1.2. bertolini and bevilacqua (2006) presented a model which combines ahp with the goal programming (gp) technique for the selection of maintenance strategies for centrifugal pumps in an oil refinery. the methodology takes into consideration decision criteria such as account budget and number of man-hour constraints in selecting the optimum strategy from among three alternative maintenance strategies (corrective, preventive and predictive). the authors concluded that the methodology proved to be a viable tool for minimization of maintenance cost (bertolini & bevilacqua, 2006). in a similar study, arunraj and maiti (2010) used ahp and gp methods for the selection of the optimum elements of maintenance systems and tools for implementation within the framework of reliability centred maintenancea review issn: 2180-1053 vol. 8 no.2 july – december 2016 15 maintenance strategy for a benzene extraction unit within a chemical plant. the decision criteria, on the basis of which optimum maintenance strategy was selected, are equipment failure risk and the cost of performing maintenance. the authors used ahp to determine decision criteria weights and the gp considered the assigned weights to rank the alternative maintenance strategies (corrective, time based, condition based and shutdown maintenance). the major improvement to the work of bertolini and bevilacqua (2006) was the use of the fussell-vesely (f-v) importance measure by the authors to calculate the different equipment items risk contributions to the plant. the introduction of goal programming increases the computation complexity of the decision making process as the decision makers or maintenance practitioners will require knowledge of programming. zaim et al., (2012) reported the use of an hybrid mcdm approach based on the integration of ahp and anp techniques for selection of an optimum maintenance strategy for a newspaper printing facility located in turkey. from the comparative study, the two techniques yielded almost the same results. the use of mcdm within the fuzzy environment has also been reported in literature for addressing maintenance strategy decision problem. lazakis et al., (2012) presented a methodology based on a combination of fuzzy set theory and topsis for the selection of the optimum maintenance strategy for a diesel generator in a cruise ship. the author ranked three maintenance strategy alternatives; corrective, preventive and predictive maintenance based on eight decision criteria; maintenance cost, efficiency/effectiveness, system reliability, management commitment, crew training, company investment, spare parts inventories and operation loss. condition based maintenance (cbm) was considered as the optimum maintenance strategy for the cruise ship diesel generator from the analysis. in an attempt to improve the fuzzy-topsis methodology, lazakis and olcer (2015) integrated ahp into it. the use of ahp was for the determination of decision criteria weights. the result of the ahp-fuzzy-topsis yielded preventive maintenance as the best strategy for the ship diesel generator maintenance. al-najjar and alsyouf (2003) presented integrated fuzzy logic and ahp methods for solving pump station maintenance strategy selection decision problem. wang et al., (2007) also used an integrated fuzzy logic and ahp technique to select optimal maintenance strategies for different equipment items in a manufacturing firm. however some doubts remain with regard to the practical use of the fuzzy approach because of the difficulty in testing and developing extensive sets of fuzzy rules (zammori and gabbrielli, 2012, braglia, 2000). 5.0 maintenance interval determination after determining the type of maintenance strategy for each of the failure modes/components of an asset or plant system, the next assignment is to determine the interval for carrying out the maintenance task. this process is an essential phase of rcm. different maintenance strategies have been discussed earlier for preventing or mitigating the effects of failure and these strategies are; corrective maintenance, scheduled overhaul, scheduled replacement, scheduled on-condition task (inspection) and continuous on-condition task. for the preventative maintenance approaches, different models have been developed by researchers for determining the interval for performing them and they have been applied in different fields with variations to suit specific industrial needs. however the basic principle for the determination of the journal of mechanical engineering and technology 16 issn: 2180-1053 vol. 8 no.2 july– december 2016 interval is to have a balance between the cost of achieving the highest reliability and the cost of unexpected failure. in the subsequent sections the different models that have been developed by different researchers for determining intervals for (1) scheduled replacement and (2) scheduled on-condition task (inspection) are discussed. 5.1 scheduled replacement interval determination preventive maintenance involves repair or replacement activities being performed at regular intervals and, as such, scheduled replacement is one of the strategies used within the framework to recover the functions of an equipment item. bahrami-g et al., (2000) defined scheduled replacement as a practice that involves making decisions, on the optimal time to replace an equipment item with respect to certain criteria with the aim of reducing or eliminating a sudden breakdown. optimization techniques are used to define the appropriate interval for the replacement of an equipment item in order to have a balance between availability of the equipment item and the cost of the related maintenance. to justify the use of the technique, two conditions must be met. the conditions are: (1) the value of weibull shape parameter β of the equipment statistical variability must be greater than 1 and (2) the cost of performing a replacement task as a result of failure must be greater than the cost of replacement under preventative mode. it therefore means that data on the failures of the equipment and related cost information are essential in order to ascertain whether or not there is the need for scheduled replacement to be carried out. this information is also required as an input into the replacement model in order to determine the optimum interval for the task. once it is ascertained that scheduled replacement is the optimum option for performing the recovery or sustainment of items of equipment, the most appropriate interval is then determined. in the determination of the optimum interval for performing scheduled replacement tasks, two models are prominent and these are; the age replacement model (arm) and the block replacement model (brm) (aven & jensen, 1999). in the application of the arm, an equipment item is replaced at a predetermined age (tp) or at failure. the implication is that if failure occurs before, tp, replacement will be performed at failure otherwise replacement is carried out at a predetermined age. additionally if an equipment item is replaced as a result of system failure, the replacement equipment is assumed to be as good as new and as such the maintenance practitioner would have to wait for another tp to elapse before performing the next replacement. the universal arm mathematical model, which is generally used for determining the appropriate time interval (tp) for scheduled replacement, is the one that was proposed by barlow and hunter (1960) and it is represented as follows: 𝐶(𝑡𝑝) = 𝐶𝑎 (1 − 𝑅(𝑡𝑝)) + 𝐶𝑏 𝑅(𝑡𝑝) ∫ 𝑡𝑓(𝑡)𝑑𝑡 𝑡𝑝 0 (3) where: 𝐶(𝑡𝑝) is the cost function per unit time 𝐶𝑎 is the cost of unit failure replacement 𝐶𝑏 is the cost of unit scheduled replacement elements of maintenance systems and tools for implementation within the framework of reliability centred maintenancea review issn: 2180-1053 vol. 8 no.2 july – december 2016 17 𝑡𝑝 is the given scheduled replacement interval and 𝑓(𝑡) is the probability density function 𝑅(𝑡𝑝) is the reliability function the essence of this age replacement model is to evaluate cost of equipment replacement for different values of ‘𝑡𝑝’. the value of 𝑡𝑝with the lowest cost is then chosen as the optimum replacement interval. hence the main purpose of this model is to minimise the cost of replacement of equipment. for the block replacement model however, equipment/components are replaced at constant time intervals and in the case of failure before the constant time interval has elapsed the equipment/components are replaced and will be replaced again once the constant time interval is attained. this type of replacement model can then result in unnecessary replacement of equipment/components. hence the generally accepted perception is that the arm is more cost effective than the brm. nevertheless the brm can be applied for less expensive equipment items whose replacement can be carried out in a group. the only advantage of brm over arm is that brm is easier to apply and manage since replacement is performed at regular intervals as opposed to arm where the maintenance practitioner would have to consider the time for replacement at failure before knowing the exact date that the next preventative replacement will be performed. the general brm mathematical model is the one developed by barlow and hunter (1960) represented as follows (ahmad & kamaruddin, 2012): 𝐶(𝑡𝑝) = 𝐶𝑏 + 𝐶𝑎 . 𝑁(𝑡𝑝) 𝑡𝑝 (4) where 𝑁(𝑡𝑝) is the number of failures expected in an interval of 0 to 𝑡𝑝. as in the case of arm, the main purpose of this model is to obtain an optimum replacement interval at the least cost. these models (arm and brm) and variations have been applied in solving replacement problems for both single unit and multi-unit systems in different industries. 5.1.1 arm and brm applications and improvement huang et al., (1995) proposed a standard solution for the arm developed by barlow and hunter (1960). the standard solution was organised in the form of tables and charts for ease of use. another novel idea for the solution is the reduction of input parameters by applying a cost ratio (ratio of 𝐶𝑎 to 𝐶𝑏) in place of failure replacement cost (𝐶𝑎 ) and preventative replacement cost ( 𝐶𝑏 ). to demonstrate the suitability of the approach, various hypothetical examples were used. cheng and tsao (2010) applied the huang et al., (1995) standard solution to obtain optimum replacement intervals for a rolling stock component. das and acharya (2004) presented two alternative techniques based on arm for optimum replacement of equipment items which indicated signs of performance degradation but operated in that state for some random time before failure. since the equipment item the authors investigated had a delay time which is the time between the point of equipment item failure initiation and the point at which the journal of mechanical engineering and technology 18 issn: 2180-1053 vol. 8 no.2 july– december 2016 equipment item eventually failed, the concept was taken into consideration in the development of two replacement methods. the first technique recommends that replacement of equipment items whether at failure or in preventative mode is performed after a fixed time during its delay time. the second technique, which extend the first policy to opportunistic age replacement, recommends that a failing equipment item should be replaced at the next available maintenance opportunity. according the authors, the two policies although designed for a single unit system are capable of addressing a multi-unit system when there is difficulty in tracking the whole life of each individual equipment item. jiang et al., (2006) examined the connection between the preventative effect associated with various replacement intervals and equivalent cost savings. the replacement models that the authors studied were arm and brm. the result obtained from the study shows that reasonable cost savings can be made if the equipment item is replaced when it has reached satisfactory age. the authors also opined that the often increasing failure rate of equipment or components does not necessarily translate to representing ‘satisfactory age’ and this has to be determined by the maintenance practitioners based on the maintenance goal. ahmad et al., (2011) used the basic arm developed by hunter and barlow in evaluating the optimum replacement interval for a production machine. the significant feature of the approach was the consideration of the covariate effect on the life of the machine, in arriving at the optimum solution. the authors compared the result they obtained with the result of the existing model which did not consider covariate effect. from the comparative analysis, the replacement interval with covariate effect and the existing replacement interval without the covariate effect were at variance. while the replacement interval with covariate effect produced a 21 day interval for replacement of the production machine, the replacement interval without the covariate effect produced a 35 day interval for the replacement of the production machine. bahrami-g et al., (2000) proposed a novel model for the scheduled replacement of an equipment item with an increasing failure rate. the technique proposed is a simplified version of the brm developed by hunter and barlow. to demonstrate the applicability of the technique a case study of an equipment item whose failure rate followed a normal distribution was applied. the result the authors obtained from the model was almost exactly the same as the result from the method of hunter and barlow. they concluded that the proposed model will support maintenance practitioners to easily define optimum replacement intervals for plant systems for better productivity and cost minimisation. from the above discussion, the majority of the methods for defining optimum replacement intervals for most systems, published in the literature are based on a single criterion. furthermore, a number of them are too abstract requiring a high level of programming, mathematical and statistical skills which can limit their use in real life situations (vatn et al., 1996, duarte et al., 2006, huang et al., 1995). additionally, approaches based on a single criterion are neither reliable nor flexible for appropriate interval selection decision making (gopalaswamy et al., 1993). elements of maintenance systems and tools for implementation within the framework of reliability centred maintenancea review issn: 2180-1053 vol. 8 no.2 july – december 2016 19 the essence of undertaking preventive maintenance is to reduce the chances of failure of plant equipment such that plant reliability and availability is optimised. the reliability of a system is dependent on the reliability of the individual components/equipment items that collectively make up the system and in order to achieve this aim, a suitable preventive maintenance and inspection programme should be in place (duarte et al., 2006). a multi-criteria decision making method which combines numerous decision criteria may be more appropriate for solving a preventive replacement interval selection decision problem that involves a number of multiple conflicting decision criteria. 5.2 inspection interval determination one of the strategies used in condition based maintenance for monitoring system performance degradation is the scheduled on-condition task and the inspection is carried out on plant systems with the aim of detecting potential failure and eliminating the failure to prevent further system degradation. the inspection task is performed on equipment items by maintenance practitioners or operators, basically with the use of handheld diagnostic tools and human intelligence. this technique nowadays is more commonly used by most industries for monitoring the condition of plant systems because it is less expensive than online monitoring or the continuous on-condition task. however the main challenge of the scheduled on-condition task is the difficult in determining the appropriate interval for performing the inspection task. this is generally due to the possibility of failure occurring between inspections if the interval is not properly timed (jardine et al., 2006) which may result in irreversible damage to the image of the company. this makes the subject of inspection interval determination important and worthy of investigation. traditionally, maintenance practitioners determined appropriate inspection intervals for their systems by relying merely on experience and/ or on the equipment manufacturers’ recommendation and in most cases the results are far from being optimal (christer et al., 1997). the inspection task, as an alternative maintenance approach for equipment item maintenance, can only be beneficial if there is a sufficient period between the time that a potential defect is observed and the actual time of failure of the equipment. hence the time that elapses between point of failure initiation, p, and the point of failure, f, is vital in estimating the appropriate inspection interval. the time that elapses between points p and f is referred to as the p-f interval (tpf) within the rcm frame work and is illustrated in figure 3. journal of mechanical engineering and technology 20 issn: 2180-1053 vol. 8 no.2 july– december 2016 figure 3. p-f interval (rausand, 1998) in rcm, the p-f interval principle is applied in determining the frequency of the condition monitoring of equipment and it was suggested that an inspection interval (t) be set at t ≤ tpf/2 (arthur, 2005). the author however stated that one major challenge of the use of the p-f approach is that there is usually no data to evaluate the p-f interval (tpf) and in most cases the evaluation is based on experts’ opinion. moubray (1991), on the other hand, suggested five ways of determining the inspection interval based on p-f but the author concluded that: “it is either impossible, impractical or too expensive to try to determine p-f intervals on an empirical basis”. apart from this approach that is used in the conventional rcm, other approaches have been described in the literature for determining inspection intervals. in the majority of these methods, cost optimization is the main decision criterion for determining the inspection interval. christer et al., (1997) proposed the delay time (dt) concept and this concept has been applied by many researchers in the modelling of the problem of inspection intervals. this approach has surpassed alternative techniques developed by other researchers for enhancing inspection intervals under different situations (wang et al., 2010). the dt concept and its application in the modelling of inspection programmes is discussed next. 5.2.1 inspection interval determination based on delay time in the delay time concept the failure processes of plant systems are divided into two phases; the first phase is the time period from when the plant system is new, to the time that it starts displaying signs of performance degradation. the second phase commences when the system starts showing some sign of degradation and runs until the system finally fails. elements of maintenance systems and tools for implementation within the framework of reliability centred maintenancea review issn: 2180-1053 vol. 8 no.2 july – december 2016 21 the time that elapsed between when the plant system initially shows signs of degradation and when it eventually fails is denoted as the delay time of the system. the delay time concept is essentially the same as the p-f interval principle described previously. the main difference between the two concepts is in the mathematical model used in determining the optimum solution for the inspection interval decision problem. the delay time concept is illustrated in figure 4. figure 4. the delay time concept in figure 4, hf denotes the delay time; pf denotes the time of plant system performance degradation initiation and, f, denotes the time that the plant system fails. the best time to carry out inspection tasks is during the delay time of the plant system. the delay time concept was introduced by christer (1982) and has been applied by many researchers for the determination of optimum maintenance inspection intervals for different industrial systems. christer (1982) applied the delay time concept in the development of a cost model for building inspection maintenance. the model was utilised in determining an appropriate inspection maintenance plan for a complex building as an alternative maintenance strategy to the reactive approach. the following assumptions were made; (1) the cost function varied within the delay time period and (2) inspection is perfect. in determining the probability density function of the delay time a subjective method was proposed. on that basis the author suggested that information such as time of failure initiation and delay time of system parts should be obtained based on experts’ (that is engineers and inspectors) estimates. a questionnaire developed for obtaining information from experts asked questions such as: (1) for how long has it been since the fault was first observed (=hla)? (2) if repair or replacement is not performed, what duration of time could the fault stay before parts may or will eventually fail (=hml)? the delay time is then evaluated for each fault as, hf = hla+hml. the distribution function f(hf) is therefore then obtained by observing a sufficient number of faults or defects. christer and waller (1984a) developed two models based on the delay time concept for inspection interval determination for a complex industrial system. the two models; cost and downtime, were firstly developed with the assumption that inspection is perfect and secondly with the assumption that inspection is imperfect. the suitability of the models journal of mechanical engineering and technology 22 issn: 2180-1053 vol. 8 no.2 july– december 2016 was demonstrated with some numerical examples. the limitation of the work is that only a single criterion is used to determine inspection interval. christer and waller (1984b) then presented an inspection interval determination technique based on a combination of a delay time model and a snapshot model. the proposed technique was used to evaluate downtime consequences for every inspection interval such that the interval with the lowest downtime is selected for the system. to demonstrate the applicability and suitability of the method, a case study of a canningline plant in a production company was used and data for the analysis was obtained subjectively through the administering of questionnaires. the method again is limited to the use of a single criterion in the determination of inspection interval. wang (1997) proposed a unique delay time methodology for determining optimum inspection interval for use in the face of insufficient data either in quantity or quality. this was achieved by developing a new technique for estimating delay time distribution from a combination of experts’ judgements rather than using actual failure data. the proposed methodology was demonstrated through two case studies. from the results of the analysis it was concluded that the technique produces similar results to the existing method that uses actual failure data. in a similar study, wang and jia (2007) developed a method based on a combination of an empirical bayesian-based technique with a delay time concept for determining the optimum inspection interval for an industrial boiler. the introduction of the empirical bayesian model was to make it possible for the proposed technique to utilise both subjective and objective data. however only a single criterion was used to determine the inspection interval. arthur (2005) presented a delay time model for the determination of the optimum inspection interval for condition monitoring of an offshore oil and gas water injection pumping system. the purpose of the study was to produce a more cost effective inspection plan for the system than the current inspection regime of a one month cycle. from the comparative analysis it was revealed that the proposed delay time model produced an inspection interval of 5 months against the current interval of 1 month with annual cost savings of £21,000. tang et al., (2014) proposed a model based on the delay time concept for inspection interval determination for a system subjected to wear whilst taking into consideration the wearing characteristics of the system. a blowout preventer core and a filter element of an oil and gas drilling system were used to demonstrate the suitability of the proposed model. for the delay time concept based model, parameters were obtained from failure and maintenance data relevant to both components. pillay et al., (2001) used an expected downtime model based on the delay time concept for determination of optimum inspection interval for a fishing vessel. the technique was applied with the objective of reducing vessel downtime due to machinery failure that could occur between discharge ports. the suitability of the approach was demonstrated with a case study of the winch system. reliability data was gathered to complement the with experts’ opinions and used as input into the proposed model. the result of the analysis shows that an inspection period of 12 hours was optimum for the system. in the elements of maintenance systems and tools for implementation within the framework of reliability centred maintenancea review issn: 2180-1053 vol. 8 no.2 july – december 2016 23 authors’ approach, only a single criterion was utilised in the determination of inspection interval. the main highlights of this review paper are presented in table 2. table 2. summary of review rcm major elements tools users/authors merits demerits risk assessment fmea cicek and celik (2013) computationally easy limited to use of only three criteria, allow only use of precise data, result may not be reliable risk assessment ahp braglia (2000), carmignani (2009) allows use of both quantitative and qualitative data unbalanced scale, limited to use of precise information risk assessment promethee maheswara and loganathan (2013), ayadi et al., (2013), moreira et al., (2009) allows use of multiple criteria challenges of determining preference function for each criterion, poor problem structuring risk assessment dematel seyedhosseini et al., (2006) failure mode severity effects & relationship between them are considered requires a lot of computational effort risk assessment topsis sachdeva et al., (2009) allows use of multiple criteria risk assessment fuzzy topsis braglia et al., (2003) allows the use of multiple criteria computational complexity due to fuzzy logic risk assessment avtopsis emovon et al., (2014) allow both use of precise and imprecise information, allows the use of more than three criteria risk assessment vikor/cp emovon et al., (2015) use of more than three criteria journal of mechanical engineering and technology 24 issn: 2180-1053 vol. 8 no.2 july– december 2016 maintenance strategy selection rcm logic tree crocker and kumar (2000), conachey (2005) time consuming, does not allowing ranking of alternatives maintenance strategy selection ahp braglia (2000), goosen and basten (2015), resobowo et al., (2014), triantaphyllou et al., (1997), nystrom ad soderholm (2010) allows use of both quantitative and qualitative data unbalanced scale, limited to use of precise information maintenance strategy selection ahp-gp bertolini and bevilacqua (2006), aruraj and maiti (2010) allows use of multiple criteria requires high level of programming skills, computational complexity maintenance strategy selection ahp-anp zaim et al., (2012) the anp allows interrelationship between criteria to be considered computational complexity due to anp maintenance strategy selection fuzzy topsis lazaklis et al., (2012) criteria weights methods determination not considered computational complexity due to fuzzy logic maintenance strategy selection ahp-fuzzy topsis lazaklis and olcer (2015) allows use of both quantitative and qualitative data, criteria weights not assumed. computational complexity due to fuzzy logic maintenance strategy selection fuzzy-ahp najjar and alsyouf (2003), wang et al., (2007) allows use of both quantitative and qualitative data computational complexity due to fuzzy logic, scheduled replacement interval determination arm huang et al., (1995), barlow and hunter (1960), cheng and more cost effective than brm. only a single criterion is considered, more costly to implement elements of maintenance systems and tools for implementation within the framework of reliability centred maintenancea review issn: 2180-1053 vol. 8 no.2 july – december 2016 25 tsao (2010), jiang et al., (2006), ahmad et al., (2011) than brm scheduled replacement determination brm barlow and hunter (1960), bahrami-g et al., (2000), jiang et al., (2006) easier to determine and implement only a single criterion is considered, approach is not cost effective inspection interval determination p-f interval principle rausand (1998) impractical to determine p-f interval, not possible to consider multiple criteria simultaneously inspection interval determination dtm christer (1982), christer and waller (1984a), christer and waller (1984b), christer et al., (1997), wang and jia (2007),wang et al., (2010), arthur (2015), tang et al., (2014), pillay (2001) result more reliable than of the p-f approach not possible to consider multiple criteria simultaneously from table 2 it is obvious that the approaches used by the different authors in solving maintenance problems within the framework of rcm all have one limitation or another. hence there is a need for researchers to develop alternative approaches that avoid the limitations of the current approaches. for example, approaches used in the determination of scheduled replacement intervals and inspection interval determination mainly use single criteria however in real-world situations multiple-criteria are generally involved in the decision making process. these criteria are in conflict with one another in most cases and in such circumstances, the use of mcdm tools such as maut or promethee may become imperative for simultaneously prioritising maintenance interval alternatives. journal of mechanical engineering and technology 26 issn: 2180-1053 vol. 8 no.2 july– december 2016 6.0 conclusions in this paper a thorough literature survey was conducted with respect to providing relevant information pertaining to the need for researchers to develop more efficient tools within an rcm framework for application to plant systems in order to make the systems safer and more reliable. to achieve this aim, the three major elements of maintenance systems; risk assessment, maintenance strategy selection and maintenance interval determination were discussed in detail and for the risk assessment a particular focus was on fmea, its variants and their corresponding limitations. for the maintenance strategy selection, the three types of maintenance strategies; corrective maintenance, preventive maintenance and condition based maintenance were presented. a survey of methods used by previous researchers for the selection of the appropriate maintenance techniques was considered together with associated limitations. for the maintenance interval determination, the discussion was centred on scheduled replacement and scheduled inspection types of maintenance with respect to current approaches and limitations of these approaches. from the review it was obvious that some of the tools and the variants utilised within the framework of rcm for the optimisation of the three main elements of maintenance systems have limitations and there is a need for researchers to develop alternative approaches that avoid such limitations. references ahmad, r. & kamaruddin, s. (2012). an overview of time-based and condition-based maintenance in industrial application. computers & industrial engineering, 63, 135-149. ahmad, r., kamaruddin, s., azid, i. & almanar, i. (2011). maintenance management decision model for preventive maintenance strategy on production equipment. journal industrial. engineering. inteter, 7, 22-34. al-najjar, b. & alsyouf, i. (2003). selecting the most efficient maintenance approach using fuzzy multiple criteria decision making. international journal of production economics, 84, 85-100. american bureau of shipping. (2000). guidance notes on risk assessment applications for the marine and offshore oil and gas industries. [online]. houston, usa: american bureau of shipping. american bureau of shipping. (2004). guidance note on reliability-centered maintence. houston, usa: american bureau of shipping. arendt, j. s. (1990). using quantitative risk assessment in the chemical process industry. reliability engineering and system safety, 29, 133-149. arthur, n. (2005). optimization of vibration analysis inspection intervals for an offshore oil and gas water injection pumping system. proceedings of the elements of maintenance systems and tools for implementation within the framework of reliability centred maintenancea review issn: 2180-1053 vol. 8 no.2 july – december 2016 27 institution of mechanical engineers, part e: journal of process mechanical engineering, 219, 251-259. arunraj, n. s. & maiti, j. (2007). risk-based maintenance-techniques and applications. journal of hazardous materials, 142, 653-661. arunraj, n. s. & maiti, j. (2010). risk-based maintenance policy selection using ahp and goal programming. safety science, 48, 238-247. aven, t. & jensen, u. (1999). stochastic models in reliability, springer. ayadi, d., azzabi, l., kobi, a., bachar, k. & robledo, c. (2013). a multicriteria failure mode and effects analysis approach for optimizing human safety. ieee reliability and maintainability symposium (rams), january 18-31 2013 orlando fl. 1-9. bahrami-g, k., price, j. w. h. & mathew, j. (2000). the constant-interval replacement model for preventive maintenance: a new perspective. international journal of quality & reliability management, 17, 822-838. barlow, r. & hunter, l. (1960). optimum preventive maintenance policies. operations research, 8, 90-100. ben-daya, m. (2009). failure mode and effect analysis. handbook of maintenance management and engineering. first ed.: springer. bertolini, m. & bevilacqua, m. (2006). a combined goal programming ahp approach to maintenance selection problem. reliability engineering and system safety, 91, 839-848. bevilacqua, m. & braglia, m. (2000). analytic hierarchy process applied to maintenance strategy selection. reliability engineering and system safety, 70, 71-83. bevilacqua, m., braglia, m. & gabbrielli, r. (2000). monte carlo simulation approach for a modified fmeca in a power plant. quality and reliability engineering international, 16, 313-324. braglia, m. (2000). mafma: multi-attribute failure mode analysis. international journal of quality and reliability management, 17, 1017-1034. braglia, m., frosolini, m. & montanari, r. (2003). fuzzy topsis approach for failure mode, effects and criticality analysis. quality and reliability engineering international, 19, 425-443. carmignani, g. (2009). an integrated structural framework to cost-based fmeca: the priority-cost fmeca. reliability engineering and system safety, 94, 861-871. journal of mechanical engineering and technology 28 issn: 2180-1053 vol. 8 no.2 july– december 2016 chen, f. (1997). issues in the continuous improvement process for preventive maintenance: observations from honda, nippondenso and toyota. production and inventory management journal, 38, 13-16. cheng, y.-h. & tsao, h.-l. (2010). rolling stock maintenance strategy selection, spares parts’ estimation, and replacements’ interval calculation. international journal of production economics, 128, 404-412. christer, a. h. (1982). modelling inspection policies for building maintenance. journal of the operational research society, 723-732. christer, a. h. & waller, w. m. (1984a). delay time models of industrial inspection maintenance problems. journal of the operational research society, 401-406. christer, a. h. & waller, w. m. (1984b). reducing production downtime using delaytime analysis. journal of the operational research society, 499-512. christer, a. h., wang, w., sharp, j. m. & baker, r. d. (1997). stochastic maintenance modelling of high-tech steel production plant. stochastic modelling in innovative manufacturing. springer. cicek, k. & celik, m. (2013). application of failure modes and effects analysis to main engine crankcase explosion failure on-board ship. safety science, 51, 6-10. conachey, r. m. (2005). development of machinery survey requirement based on reliability-centerd maintenance. abs technical papers. crocker, j. & kumar, u. d. (2000). age-related maintenance versus reliability centred maintenance: a case study on aero-engines. reliability engineering and system safety, 67, 113-118. cross , b. r. & ballesio, j. e. (2003). a quantitative risk assessment model for oil tankers. abs technical paper. america bureau of shipping. das, a. n. & acharya, d. (2004). age replacement of components during ifr delay time. reliability, ieee transactions on, 53, 306-312. deng, h. (1999). multicriteria analysis with fuzzy pairwise comparison. international journal of approximate reasoning, 21, 215-231. dhillon, b. s. (2002). engineering maintenance: a modern approach. florida: crc press. duarte, j. a. c., craveiro, j. c. t. a. & trigo, t. p. (2006). optimization of the preventive maintenance plan of a series components system. international journal of pressure vessels and piping, 83, 244-248. elements of maintenance systems and tools for implementation within the framework of reliability centred maintenancea review issn: 2180-1053 vol. 8 no.2 july – december 2016 29 ebrahimipour, v., najjarbashi, a. & sheikhalishahi, m. (2015). multi-objective modeling for preventive maintenance scheduling in a multiple production line. journal of intelligent manufacturing, 26, 111-122. emovon, i., norman, r. a., j, m. a. & pazouki, k. (2015). an integrated multicriteria decision making methodology using compromise solution methods for prioritising risk of marine machinery systems. ocean engineering, 105, 92-103. emovon, i., norman, r. a. & murphy, a. j. (2014) a new tool for prioritising the risk of failure modes for marine machinery systems. proceedings of the 33rd international conference on ocean, offshore and arctic engineering omae14, june 8 13 2014 california united states,. americal society of mechanical engineers. goossens, a. j. m. & basten, r. j. i. (2015). exploring maintenance policy selection using the analytic hierarchy process; an application for naval ships. reliability engineering and system safety, 142, 31-41. gopalaswamy, v., rice, j. a. & miller, f. g. (1993). transit vehicle component maintenance policy via multiple criteria decision making methods. journal of the operational research society, 37-50. headquarters department of the army (2006). failures modes, effects and criticality analysis (fmeca) for command, control, communications, computer, intelligence, surveillance, and reconnaissance (c4isr) facilities. washington, dc. huang, j., miller, c. r. & okogbaa, o. g. (1995). optimal preventive-replacement intervals for the weibull life distribution: solutions and applications. reliability and maintainability symposium, 1995. proceedings., annual, 1995. ieee, 370377. ilangkumaran, m. & kumanan, s. (2009). selection of maintenance policy for textile industry using hybrid multi-criteria decision making approach. journal of manufacturing technology management, 20, 1009-1022. jardine, a. k. s., lin, d. & banjevic, d. (2006). a review on machinery diagnostics and prognostics implementing condition-based maintenance. mechanical systems and signal processing, 20, 1483-1510. jiang, r., ji, p. & tsang, a. h. c. (2006). preventive effect of optimal replacement policies. journal of quality in maintenance engineering, 12, 267-274. kahrobaee, s. & asgarpoor, s. risk-based failure mode and effect analysis for wind turbines (rb-fmea). 2011. journal of mechanical engineering and technology 30 issn: 2180-1053 vol. 8 no.2 july– december 2016 labib, a. w., o'connor, r. f. & williams, g. b. (1998). an effective maintenance system using the analytic hierarchy process. integrated manufacturing systems, 9, 87-98. lazakis, i. & olcer, a. i. (2015). selection of the best maintenance approach in the maritime industry under fuzzy multiple attribute group decision-making environment. proc imeche part m: journal of engineering for maritime environment, 1-13. lazakis, i., turan, o. & olcer, a. i. (2012). determination of the optimum ship maintenance strategy through multi attribute decision making. 11th international marine design conference 2012 glasgow, scotland, uk. 473-487. ling, d., huang, h. z., song, w., liu, y. & zuo, m. j. (2012) design fmea for a diesel engine using two risk priority numbers. 2012. macharis, c., springael, j., de brucker, k. & verbeke, a. (2004). promethee and ahp: the design of operational synergies in multicriteria analysis.: strengthening promethee with ideas of ahp. european journal of operational research, 153, 307-317. maheswaran, k. & loganathan, t. (2013). a novel approach for prioritisation of failure modes in fmea using mcdm. international journal of engineering research and application, 3, 733-739. mobley, r. k. (2001). plant engineers handbook. jkb. mobley, r. k. (2004). maintenance fundamental usa, butterworth-heinemann. mohan, m., gandhi, o. p. & agrawal, v. p. (2004). maintenance strategy for a coalbased steam power plant equipment: a graph theoretic approach. proceedings of the institution of mechanical engineers, part a: journal of power and energy, 218, 619-636. moreira, m. p., dupont, c. j. & vellasco, m. m. b. r. (2009) promethee and fuzzy promethee multicriteria methods for ranking equipment failure modes. intelligent system applications to power systems, 2009. isap'09. 15th international conference on, 2009. ieee, 1-6. moubray, j. (1991). reliability-centred maintenance buttenvorth. heinemann oxford. noemi, m. p. & william, l. 1994. maintenance scheduling issues, results and research needs. international journal of operations and production management, 14, 4769. nyström, b. & söderholm, p. (2010). selection of maintenance actions using the analytic hierarchy process (ahp): decision-making in railway infrastructure. structure and infrastructure engineering, 6, 467-479. elements of maintenance systems and tools for implementation within the framework of reliability centred maintenancea review issn: 2180-1053 vol. 8 no.2 july – december 2016 31 pillay, a. & wang, j. (2003). modified failure mode and effects analysis using approximate reasoning. reliability engineering and system safety, 79, 69-85. pillay, a., wang, j. & wall, a. (2001). optimal inspection period for fishing vessel equipment: a cost and downtime model using delay time analysis. marine technology, 38, 122-129. pintelon, l., pinjala, s. k. & vereecke, a. (2006). evaluating the effectiveness of maintenance strategies. journal of quality in maintenance engineering, 12, 720. pride, a. 2008. reliability centered maintenance. whole building design guide [online]. rausand, m. (1998). reliability centered maintenance. reliability engineering & system safety, 60, 121-132. rausand, m. & vatn, j. (1998). reliability centered maintenance. in: soares, c. g. (ed.) risk and reliability in marine technology. balkema holland. resobowo, d. s., buda, k. a. & dinariyana, a. a. b. (2014). using sensitivity analysis for selecting of ship maintenance variables for improving reliability of military ship. academic research international, 5, 127. sachdeva, a., kumar, d. & kumar, p. (2009a). multi-factor mode criticality analysis using topsis. journal of industrial engineering international, 5, 9. sachdeva, a., kumar, p. & kumar, d. (2009). maintenance criticality analysis using topsis. 2009 ieee international conference on industrial engineering and engineering management, 2009b hong kong. 199-203. scipioni, a., saccarola, g., centazzo, a. & arena, f. (2002). fmea methodology design, implementation and integration with haccp system in a food company. food control, 13, 495-501. selvik, j. t. & aven, t. (2011). a framework for reliability and risk centered maintenance. reliability engineering &amp; system safety, 96, 324-331. seyed-hosseini, s. m., safaei, n. & asgharpour, m. j. (2006). reprioritization of failures in a system failure mode and effects analysis by decision making trial and evaluation laboratory technique. reliability engineering & system safety, 91, 872-881. shaghaghi, m. & rezaie, k. (2012). failure mode and effects analysis using generalized mixture operators. journal of optimization in industrial engineering, 11, 1-10. journal of mechanical engineering and technology 32 issn: 2180-1053 vol. 8 no.2 july– december 2016 siddiqui, a. w. & ben-daya, m. (2009). reliability centered maintenance. handbook of maintenance management. springer. stanojevic, p., masonic, v. & bovid, v. (2000). maintenance systems organisation structure design methodology based on modelling and simulation. esm'2000, simulation congress, 2000 gent. stanojevic, p., miskovic, v., bukvic, v. & aleksic, m. (2004). multilevel maintenance systems influence factors analysis. belgrade: military-technical gazette. sullivan, g. p., pugh, r., melendez, a. p. & hunt, w. d. (2004). operations and maintenance best practices: a guide to achieving operational efficiency. us: us department of energy. tang, y., jing, j. j., yang, y. & xie, c. (2014). parameter estimation of a delay time model of wearing parts based on objective data. mathematical problems in engineering. triantaphyllou, e., kovalerchuk, b., mann jr, l. & knapp, g. m. (1997). determining the most important criteria in maintenance decision making. journal of quality in maintenance engineering, 3, 16-28. vatn, j., hokstad, p. & bodsberg, l. (1996). an overall model for maintenance optimization. reliability engineering & system safety, 51, 241-257. vavra, b. (2007). expert lay out a case for roi of maintenance. . plant engineering, 61, 1-12. vidal, l.-a., marle, f. & bocquet, j.-c. (2011). using a delphi process and the analytic hierarchy process (ahp) to evaluate the complexity of projects. expert systems with applications, 38, 5388-5405. waeyenbergh, g. & pintelon, l. 2004. maintenance concept development: a case study. international journal of production economics, 89, 395-405. wang, l., chu, j. & wu, j. (2007). selection of optimum maintenance strategies based on a fuzzy analytic hierarchy process. international journal of production economics, 107, 151-163. wang, w. 1997. subjective estimation of the delay time distribution in maintenance modelling. european journal of operational research, 99, 516-529. wang, w., banjevic, d. & pecht, m. (2010). a multi-component and multi-failure mode inspection model based on the delay time concept. reliability engineering and system safety, 95, 912-920. elements of maintenance systems and tools for implementation within the framework of reliability centred maintenancea review issn: 2180-1053 vol. 8 no.2 july – december 2016 33 wang, w. & jia, x. (2007). an empirical bayesian based approach to delay time inspection model parameters estimation using both subjective and objective data. quality and reliability engineering international, 23, 95-105. yang, j., huang, h.-z., he, l.-p., zhu, s.-p. & wen, d. (2011). risk evaluation in failure mode and effects analysis of aircraft turbine rotor blades using dempster–shafer evidence theory under uncertainty. engineering failure analysis, 18, 2084-2092. zaim, s., turkyílmaz, a., acar, m. f., al-turki, u. & demirel, o. f. (2012). maintenance strategy selection using ahp and anp algorithms: a case study. journal of quality in maintenance engineering, 18, 16-29. zammori, f. & gabbrielli, r. (2012). anp/rpn: a multi criteria evaluation of the risk priority number. quality and reliability engineering international, 28, 85-104. journal of mechanical engineering and technology 34 issn: 2180-1053 vol. 8 no.2 july– december 2016