International Journal of Engineering Materials and Manufacture (2018) 3(1) 55-62 https://doi.org/10.26776/ijemm.03.01.2018.07 M. Y. Ali , A. Sabur, A. Banu, M. A. Maleque, and E. Y. T. Adesta Department of Manufacturing and Materials Engineering International Islamic University Malaysia PO Box 10, 50728 Kuala Lumpur, Malaysia E-mail: eadesta@iium.edu.my Reference: Ali, M. Y., A. Sabur, A. Banu, Maleque, M. A., and Adesta, E. Y. T. (2018). Micro Electro Discharge Machining of Nonconductive Ceramic. International Journal of Engineering Materials and Manufacture, 3(1), 55-62. * This article was presented in 2nd World Symposium on Mechanical and Control Engineering, Guilin, China, 2-3 Nov. 2017. Micro Electro Discharge Machining of Nonconductive Ceramic * Mohammad Yeakub Ali, Abdus Sabur, Asfana Banu, Md. Abdul Maleque and Erry Y. T. Adesta Received: 19 March 2018 Accepted: 26 March 2018 Published: 30 March 2018 Publisher: Deer Hill Publications © 2018 The Author(s) Creative Commons: CC BY 4.0 ABSTRACT The micro-electro discharge machining (micro-EDM) models established for single spark erosion are not applicable for nonconductive ceramics because of random spalling. Moreover, it is difficult to create single spark on a nonconductive ceramic workpiece when the spark is initiated by the assisting electrode. In this paper, process development for nonconductive zirconium oxide (ZrO2) is discussed. It is shown that the charging and discharging duration depend on the capacitance and resistances of the circuit. The number of sparks per unit time is estimated from the single spark duration derived from heat transfer fundamentals. The model showed that both the capacitance and voltage are significant process parameters for material removal rate (MRR). However, capacitance was found to be the dominating parameter over voltage. As in case of higher capacitances, the creation of a conductive carbonic layer on the machined surface was not stable; the effective window of machining 101 - 103 pF capacitance and 80 - 100 V gap voltage or 10 - 470 pF capacitance and 80 - 110 V gap voltage. This fact was confirmed EDX analysis where the presence of high carbon content was evident. Conversely, the spark was found to be inconsistent using parameters beyond these ranges and consequently insignificant MRR. Nevertheless, the effective numbers of sparks per second were close to the predicted numbers when machining conductive copper material. In addition, higher percentage of ineffective pulses was observed during the machining which eventually reduced the MRR. In case of validation, average deviations between the predicted and experimental values were found to be around 10%. Keywords: Micro-EDM, Nonconductive ceramics, assisting electrode, MRR 1 INTRODUCTION Advancement in technology has opened the windows of some special materials to meet particular requirements by means of miniaturization. Nonconductive ceramics are considered one of the most promising materials that can fulfil the demands of multifarious miniaturized applications in microfluidics, reactors, electromechanical generators and for medical purposes. However, these materials are difficult to machine by conventional machining techniques due to their high hardness and brittleness. Nonconventional machining processes are regarded as promising for the structuring or shaping of nonconductive ceramics [1]. Micro-electro discharge machining (micro-EDM) is a nonconventional, noncontact and effective process for structuring conductive, hard and brittle materials. Most of the advanced ceramics are electrically nonconductive; therefore, micro-EDM cannot be used directly. In recent years, an advanced technique has been introduced to utilize micro-EDM for nonconductive ceramics in which an electrically conductive metallic layer, referred as the ‘assisting electrode’ (AE) is applied on the workpiece surface. Using the AE and controlling the micro-EDM process parameters such as capacitance, voltage and polarity, nonconductive ceramics (ZrO2, Al2O3, Si3N4 and SiC) have been successfully structured. 1.1 Nonconductive Ceramics Nonconductive ceramics are also known as engineering, advanced, technical, or insulating ceramics. Their electrical conductivity is less than 0.1 Scm-1 [2-3]. Due to the excellent chemical and physical properties, nonconductive ceramics are now being used in fabrication of domestic, industrial and building products. Specific applications include cutting Micro Electro Discharge Machining of Nonconductive Ceramic 56 tools, self-lubricating bearings, turbine blades, internal combustion engines, heat exchangers, ballistic armour, ceramic composite automotive brakes, diesel particulate filters, piezo-ceramic sensors and a wide variety of prosthetic products. Currently, nonconductive ceramic materials are used in the biomedical field to fabricate femoral heads and acetabular cups for total hip replacement, dental implants and restorations, bone fillers and scaffolds for tissue engineering. Nano-structured alumina (Al2O3) and zirconia (ZrO2) based ceramics and composites or non-oxide ceramics could be potential candidate material for biomedical applications in the near future [4]. 1.2 Micro-electro Discharge Machining Micro-electro discharge machining (Micro-EDM) is an electro-thermal material removal process which is used for manufacturing three-dimensional micro-components. In this machining process, a series of electrical sparks or discharges occurs rapidly in a short span of time within a constant spark gap between the tool electrode and the workpiece. It is a complex process in which thermodynamic, hydrodynamic, electromagnetic and electrodynamic processes are involved. Any material having a minimal electrical conductivity of 0.1 Scm-1 can be machined by micro- EDM irrespective of its hardness and brittleness [3]. In micro-EDM, thermal energy evolved from the electrical sparks flows into the electrode, workpiece and dielectric fluid. A fraction of the heat is conducted through the workpiece and an enormous temperature rise causes melting and vaporization of a small amount of material from the workpiece [5]. In this way a micro-crater is formed in every effective spark. Thus, theoretical models of material removal rate (MRR) and average surface roughness (Ra) in micro-EDM of conductive materials have been developed based on electrical circuit and heat conduction theories considering the material removal by melting and vaporization [6-8]. However, the micro-EDM process can also be used for micro-structuring of electrically nonconductive ceramics applying AE [3]. In micro-EDM of nonconductive ceramics, material is removed by melting, vaporization and spalling [9]. In micro-EDM of nonconductive ceramics, after the removal of the initially applied assisting electrode, a layer of pyrolytic carbon (PyC) would be deposited on the workpiece surface, disassociating the carbonic dielectric in the absence of oxygen at the higher temperature. The new layer would have the threshold of electrical conductivity for continuation of the machining, resulting in material removal by melting, vaporization and spalling. Since the melting, vaporization and spalling is the effect of energy and energy is the function of capacitance and voltage in a RC pulse circuit, it is postulated that MRR would have a functional relationship with capacitance and voltage. The models of micro-EDM for nonconductive ceramics can certainly contribute towards making the process user-friendly and helping machining parameters selection for desired outputs. In micro-EDM, the discharge duration and supplied energy are much lower than EDM, with discharge duration limited to around 100 ns and the amount of supplied energy being about 10 µJ. Figure 1: Schematic diagram of relaxation type pulse generator in EDM [12] Figure 2: Schematic diagram of transistor type pulse generator in EDM [12] Ali, et al. (2018): International Journal of Engineering Materials and Manufacture, 3(1), 55-62 57 However, the power density is higher in micro-EDM [10-11]. To create the pulses, relaxation or transistor type generators can be used. Schematic diagrams of relaxation and transistor type pulse generators are shown in Figure 1 and Figure 2 respectively. In a relaxation type generator, resistance and capacitance are connected to the circuit in which the capacitor charging and discharging occur alternatively. Using this circuit, discharge can be created with a high peak current for a short duration. Current and gap voltage are controlled at a predefined level throughout the pulse on-time. In a transistor type pulse generator, a series of resistance and transistors are connected in series as a switching device. This circuit can provide longer discharge duration with a larger peak current. The transistor type circuit has a longer ignition delay time. Generator selection in micro-EDM depends on the desired machining outputs. The transistor type generator is capable of creating a higher energy with a larger peak current and longer discharge duration. On the other hand, the RC pulse generator can be controlled precisely to generate very low energy with a shorter discharge duration [12]. Therefore, the RC pulse generator is used as the pulse creator in micro-EDM to produce highly precise microstructures [13]. A multi-purpose miniature machine tool (DT-110, Mikrotools Inc., Singapore) has been manufactured for micromachining with high precision in which a RC pulse circuit is used [14-15]. As a noncontact machining process, micro-EDM is capable of machining any material which is electrically conductive, irrespective of its hardness, shape and strength [16]. Micro-EDM is useful in producing three-dimensional micro-structures with a higher MRR, a machining accuracy of 1-5 µm and an aspect ratio of more than 20 [17]. A minimum feature of 10 µm is possible in fabricating precise micro-hole, micro-mechanical parts, milling tools and complex micro-structures by micro-EDM [18]. 1.3 Assisting Electrode AE is an additional conductive layer on the workpiece that is applied to initiate sparks during machining or structuring of nonconductive ceramic materials [19]. There are two methods of applying AE. In the first method, AE is adhered onto the workpiece to start machining. Once the AE layer finishes, carbon molecule cracked from carbonic dielectric is deposited on the machined surface and a conductive thin layer is generated which assists in creating a further spark. The adhesive copper or aluminium foil, coated graphite, silver varnish, sputtered workpiece material with copper, silver or gold are the main materials used as AE. A schematic diagram of micro-EDM of nonconductive ceramics using fixed AE is shown in Figure 3. In the second method, conductive metallic AE foil or sheet is continuously fed into the machining zone by a servo mechanism as shown in Figure 4. Electro discharge (ED) milling of Al2O3 has been successfully accomplished using the continuously feeding AE method and it is considered a suitable process for large area machining [20]. In this process, a steel wheel is used as a tool electrode mounted on a rotary spindle driven by an AC motor. The tool electrode and the AE are connected to the positive and negative poles of the pulse generator respectively. This paper presents the research development in micro-EDM of nonconductive ceramic materials. Micromachining techniques were discussed briefly to justify the selection of micro-EDM for this study. Basic principles, process parameters and characteristics of micro-EDM were highlighted. Assisting electrode techniques and material removal mechanisms in micro-EDM of nonconductive ceramic materials were discussed extensively. Special focus was given in exploring the theoretical modelling in micro-EDM of nonconductive ceramic materials. From the review, it appeared that the most of existing micro-EDM theoretical models are not applicable for nonconductive ceramic materials which results in the need to develop new models for process characteristics. Figure 3: Schematic diagram of micro-EDM of nonconductive ceramics with fixed AE [2] Micro Electro Discharge Machining of Nonconductive Ceramic 58 Figure 4: ED milling of nonconductive ceramics [21] 2 EFFECT OF DIFFERENT PULSES ON MRR Pulses were recorded during micro-EDM of ZrO2 and five categories of pulses have been identified which are as follows: 2.1 Normal (effective) Pulse: Normal pulses are produced when the capacitor is fully charged before the spark and fully discharged during the sparks. One of the images of effective pulse is shown in Figure 5. Normal pulses are effective in material removal because of the maximum energy supply by the circuit. Normal pulses were observed during the machining of Cu AE and during the initial stage of ceramic removal. According to Figure 5, The spark had occurred after the completion of charging, without any delay. The discharge shape was almost vertical, which indicates that there was no resistance between the tool and workpiece. The calculated spark number is 909090 for R = 103 Ω and C = 220 pF. From Figure 5, it is obvious that the effective sparks recorded by oscilloscope to be 902000 which agrees with the calculated value. Figure 5: Normal pulse at C = 220 pF and V = 110 V 2.2 Pulse During Cutting Ceramics One of the images of ceramic pulses is shown in Figure 6. Once the copper AE was finished, sparks began to occur between PyC AE and the tool electrode, removing ceramic material. At the beginning of ceramic machining, the normal shape of voltage signal was observed in the charging stage. As the machining progressed, pulses began to Ali, et al. (2018): International Journal of Engineering Materials and Manufacture, 3(1), 55-62 59 remain constant at peak for long periods which indicate that the capacitor was fully charged but was waiting to release energy. It can be assumed that the conductive PyC layer had yet to attain the sufficient thickness and conductivity for the sparks. Once the conditions were fulfilled, the sparks occurred. From Figure 6, it can be concluded that the ceramic pulses have a longer pulse-off time than normal pulses. 2.3 Effective Arc Pulse Some pulses were observed in which the capacitor was not fully discharged and the spark had occurred at higher voltages than zero, with lower currents. These are known as effective arc pulses. One of the images of effective arcs is given in Figure 7. Effective arcs have a smaller current than normal or ceramic discharges and remove a little amount of material. Effective arcs occurred due to the presence of a small amount of debris in the gap. Fewer electrons can reach the anode because of insufficient energy [20-21]. Two types of effective arc pulses were observed. In the first type of effective arcs, pulse-off time was similar to the normal pulse. In the second category of effective arcs, pulse- off time was longer which is similar to ceramic pulse. Figure 6: Pulses during cutting ceramics at C = 100 pF and V = 110 V Figure 7: Effective arc discharge at C = 100 pF and V = 100 V Longer pulse-off Shorter pulse-off Charging and discharging at higher voltage A: Longer pulse-off time Micro Electro Discharge Machining of Nonconductive Ceramic 60 Figure Error! No text of specified style in document.8: Immature discharges at C = 100 pF and V = 100 V 2.4 Immature Pulse During the deep machining of nonconductive ceramics, several pulses were observed in which sparks occurred after the capacitor voltage had reached a peak value. But charging started at a much higher voltage. These are identified as immature pulses. Immature pulses are produced due to the presence of a huge amount of debris inside the cavity and it creates a very small amount of energy. One of the images of immature pulses is given in Figure 8. Most of the immature pulses do not follow the charging and discharging characteristics of the RC circuit. In many cases, they are mixed with normal and ceramic pulses and produce complex pulse shapes [22]. 2.5 Short Circuit The short circuit occurred frequently in micro-EDM of nonconductive ceramics mainly due to two reasons: i) The presence of debris which creates the channel of current flow without the sparks; ii) Tool movement towards the workpiece without creation of a sufficiently conductive carbonic layer. Table Error! No text of specified style in document.1 Experimental parameters Conditions Values Workpiece ZrO2 Tool electrode Cylindrical copper rod Tool electrode diameter (mm) 1 Tool electrode polarity -ve Assisting electrode Adhesive copper foil Assisting electrode thickness (µm) 60 Dielectric Kerosene Pulse generator RC Resistance (kΩ) 1 Variables Factors Parameters Levels I II III A Capacitance, C (pF) 1000 100 10 B Voltage, V (V) 100 90 80 C Speed, n (rpm) 350 300 250 D Feed rate, f (µm/s) 2 1.6 1.2 Ali, et al. (2018): International Journal of Engineering Materials and Manufacture, 3(1), 55-62 61 2.6 Experimental Design The design of experiments (DOE) is the technique of defining and investigating all possible conditions in an experiment involving multiple factors. A scientific approach was used for planning and conducting the experiments and for analysing the data efficiently. The DOE was used to: i) Reduce the number of trials significantly. ii) Identify important decision variables which control and improve the characteristics of the product or the process. iii) Find out the optimal setting of the parameters. In this study, the Taguchi method was used for DOE and subsequent analysis of the results. It requires a minimum of one run per condition of the experiment. But one run does not represent the range of possible variability in the results. Repetition allows determination of variance index called signal to noise ratio. The greater this value, the smaller the product variance around the target. The parameters and their levels for the MRR were selected based on an experimental study as presented in Table 1. In the experimental investigations for the modelling of MRR, it was observed that the speed and feed rate are not significant parameters in micro-EDM of ZrO2. To apply the models in micro-EDM of nonconductive ZrO2 ceramic, parameters were found out by solving the developed models based on the expected MRRc. Using the parameter values, micro-channels and micro cavities were machined on nonconductive ZrO2 ceramic material. Expected MRRc were compared with the experimental values. 3 CONCLUSIONS In this research, theoretical models of the material removal rate (MRR) in micro-EDM of nonconductive ZrO2 ceramic have been developed and validated. The research findings have been summarized as follows. 1. Process parameters for effective Micro-EDM of ZrO2 have been identified by experimental study. Capacitance ranges of 101 - 103 pF and gap voltages of 80 - 100 V or capacitance ranges of 10 - 470 pF and gap voltages of 80 - 110 V have been identified as effective electrical parameters for micro-EDM of ZrO2 with a –ve copper tool electrode and adhesive copper foil assisting electrode in kerosene dielectric fluid. 2. Models have been developed for RC pulse micro-EDM based on the single spark erosion using electro- thermal theories. Multiplying correction factor (ε) derived from experimental investigations have been introduced to adjust the MRR in micro-EDM of nonconductive ZrO2. The theoretical MRR model has been expressed by following equation.       += mv s c H k H kkCVN MRR 21 2 2ρ ε ( )20022.000009.0043.0 VC ×+×+−=ε )(5 1 21 RRC N s + = where ε is an experimental correction factor which has been determined in terms of the process parameters of capacitance and voltage and is expressed by following equation. Ns is the number of sparks per second which is estimated from the RC pulse charging and discharging duration and is expressed by: 3. It was found that the experimental MRRc is very low compared to the theoretical values due to the longer pulse-off time in micro-EDM of nonconductive ceramics and for the effect of various ineffective pulses such as delayed discharge, short circuit, arching and immature discharges. In modelling, the formation energy of PyC and the machining energy of PyC were not quantified. Therefore, in addition to spalling effect, the correction factor, ε also includes the effect of formation-dissociation of PyC and the effect of ineffective pulses. 4. From the experimental investigation, it was found that capacitance is the most significant parameter for the creation of a conductive PyC layer in micro-EDM of nonconductive ZrO2 ceramic. However, the MRR is controlled by voltage as a whole. 5. The MRR in micro-EDM of ZrO2 was found to be very low compared to other hard materials. Because of electrical nonconductivity, a higher number of ineffective pulses were produced, which eventually lowered the MRR in ZrO2 as compared to conductive materials. 6. Experiments were conducted to validate the MRR in micro-EDM of conductive copper materials. The effective number of sparks were recorded by oscilloscope at the initial stage of machining about 902000, Micro Electro Discharge Machining of Nonconductive Ceramic 62 which agrees with the calculated spark numbers of 909090 (at R = 103 Ω and C = 220 pF). It has been also observed that the number of effective pulses decreased with the increase of depth of cut and about 50% of pulses were found to be ineffective in micro-EDM of conductive materials. 7. Micro-channels have been created on ZrO2 with selected parameters for the expected MRRc values as one of the applications of micro-EDM models. 8. It was shown that the MRRc in micro-EDM of ZrO2 were found to be in agreement with the calculated expected values by an average error of 8.86 %. REFERENCES 1. Ji, R., Liu, Y., Zhang, Y., Wang, F., Chen, Z., & Dong, X. (2011). Study on single-discharge machining characteristics of non-conductive engineering ceramics in emulsion with high open voltage and large capacitor. Proceedings of the Institution of Mechanical Engineers, Part B: Journal of Engineering Manufacture, 225(10), 1888-1898. 2. Hösel, T., Cvancara, P., Ganz, T., Müller, C., & Reinecke, H. (2011). Characterisation of high aspect ratio non- conductive ceramic microstructures made by spark erosion. Microsystem Technologies, 17(2), 313-318. 3. Hösel, T., Müller, C., & Reinecke, H. (2011). Spark erosive structuring of electrically nonconductive zirconia with an assisting electrode. CIRP Journal of Manufacturing Science and Technology, 4(4), 357-361. 4. Chevalier, J., & Gremillard, L. (2009). Ceramics for medical applications: A picture for the next 20 years. Journal of the European Ceramic Society, 29(7), 1245-1255. 5. Yoo, B. H., Min, B.-K., & Lee, S. J. (2010). Analysis of the machining characteristics of EDM as functions of the mobilities of electrons and ions. International Journal of Precision Engineering and Manufacturing, 11(4), 629- 632. 6. Dhanik, S., & Joshi, S. S. (2005). Modeling of a single resistance capacitance pulse discharge in micro-electro discharge machining. Journal of Manufacturing Science and Engineering, 127(4), 759-767. 7. Salonitis, K., Stournaras, A., Stavropoulos, P., & Chryssolouris, G. (2009). Thermal modeling of the material removal rate and surface roughness for die-sinking EDM. The International Journal of Advanced Manufacturing Technology, 40(3-4), 316-323. 8. Zahiruddin, M., & Kunieda, M. (2012). Comparison of energy and removal efficiencies between micro and macro EDM. CIRP Annals-Manufacturing Technology, 61(1), 187-190. 9. Liu, Y., Yu, L., Xu, Y., Ji, R., & Li, Q. (2009). Numerical simulation of single pulse discharge machining insulating Al2O3 ceramic. Proceedings of the Institution of Mechanical Engineers, Part B: Journal of Engineering Manufacture, 223(1), 55-62. 10. Schubert, A., Zeidler, H., Wolf, N. & Hackert, M. (2011). Micro electro discharge machining of electrically nonconductive ceramics. In AIP Conference Proceedings, 1353, 1303. 11. Wong, Y., Rahman, M., Lim, H., Han, H., & Ravi, N. (2003). Investigation of micro-EDM material removal characteristics using single RC-pulse discharges. Journal of Materials Processing Technology, 140(1), 303-307. 12. Kunieda, M., Lauwers, B., Rajurkar, K. P., & Schumacher, B. M. (2005). Advancing EDM through fundamental insight into the process. CIRP Annals-Manufacturing Technology, 54(2), 64-87. 13. Shin, H. S., Park, M. S., & Chu, C. N. (2011). Machining characteristics of micro EDM in water using high frequency bipolar pulse. International Journal of Precision Engineering and Manufacturing, 12(2), 195-201. 14. Lim, H., Wong, Y., Rahman, M., & Edwin Lee, M. (2003). A study on the machining of high-aspect ratio micro- structures using micro-EDM. Journal of Materials Processing Technology, 140(1), 318-325. 15. Jahan, M., Wong, Y., & Rahman, M. (2009). A study on the fine-finish die-sinking micro-EDM of tungsten carbide using different electrode materials. Journal of Materials Processing Technology, 209(8), 3956-3967. 16. Kumar, S., Singh, R., Singh, T., & Sethi, B. (2009). Surface modification by electrical discharge machining: A review. Journal of Materials Processing Technology, 209(8), 3675-3687. 17. Liow, J. (2009). Mechanical micromachining: A sustainable micro-device manufacturing approach? Journal of Cleaner Production, 17(7), 662-667. 18. Yan, M.-T., & Lin, S.-S. (2011). Process planning and electrode wear compensation for 3D micro-EDM. The International Journal of Advanced Manufacturing Technology, 53(1-4), 209-219. 19. Mohri, N., Fukuzawa, Y., Tani, T., Saito, N., & Furutani, K. (1996). Assisting electrode method for machining insulating ceramics. CIRP Annals-Manufacturing Technology, 45(1), 201-204. 20. Liu, Y., Li, X., Ji, R., Yu, L., Zhang, H., & Li, Q. (2008). Effect of technological parameter on the process performance for electric discharge milling of insulating Al2O3 ceramic. Journal of Materials Processing Technology, 208(1), 245-250. 21. Liu, Y., Ji, R., Li, X., Yu, L., Zhang, H., & Li, Q. (2008). Effect of machining fluid on the process performance of electric discharge milling of insulating Al2O3 ceramic. International Journal of Machine Tools and Manufacture, 48(9), 1030-1035. 22. Sabur, A., Moudood, A., Ali, M. Y., & Maleque, M. A. (2014). Effect of micro-EDM parameters on material removal rate of nonconductive ZrO2 ceramic. Applied Mechanics and Materials, 465, 1329-1333.