 Proceedings of Engineering and Technology Innovation , vol. 2, 2016, pp. 23 - 25 23 Experimental Study of Progressive Compression Method of Resin Delivery in Liquid Composite Molding Chih-Yuan Chang * , Wei-Ru Chen Department of Mechanical and Automation Engineering , Kao Yuan University, Kaohsiung, Taiwan. Received 18 March 2016; received in revised form 13 April 2016; accept ed 16 May 2016 Abstract A new technique of resin delivery, which we refer to as the progressive compression method (PCM), has been invented to reduce filling time associated with the vacuum assisted resin transfer mold ing (VA RTM) process. In the method, the bag is divided into several segments. During infusion, a ll seg mented bags are pulled away fro m the prefo rm by the vacuum. Hence, resin is easily infused into the loose preform. Once enough volume of resin is infused, the vacuum within the segmented bags is released in a step-wise manner. The at mospheric pressure of the heated air is progressively applied on the segmented bag that is inflated to compact the wetted preform and drive the resin through the re main ing dry preform. The resin flo w is en- hanced since the dry preform re ma ins loose during the filling process. The research aims to investigate the effect of seven process parame - ters on the PCM co mp lete filling process by applying Taguchi’s method. All chosen factors are designed with two levels. Expe rimental results show that the predicted optimu m settings are higher vacuu m pressure, more co mpression segment, later init iation of the ne xt co mpression segment, higher air te mperature, later introduc- tion of the heated air, lower init ial height of the cavity and less excess infused resin for reducing the filling time. Co mpared with typical VA RTM without flow enhancement, PCM at the optimu m settings reduces the filling time by 72.85%. Ke ywor ds : filling process , Taguchi’s method, vacuum assisted resin transfer molding 1. Introduction Vacuum assisted resin transfer molding (VA RTM) process is a popular method for the manufacture of co mposite components. In a typical VA RTM, the dry preform is placed on the rigid mold and sealed with an elastic bag to create an a irtight mo ld cav ity. The in let line is opened and the resin is driven by atmospheric pressure through the infusion line a fter the vacuum is applied within the mo ld cavity. On ce the preform is comp letely saturated, the removal of the unnecessary resin fro m the wetted pre- form is performed to achieve the uniform thickness of the part. Fig. 1 Schematic diagram for PCM X X X Resin Vacuum Atmosphere (a) (b) (c) (d) Vacuum Vacuum Atmosphere Atmosphere Vacuum *Corresponding aut hor. Email: yuan@cc.kyu.edu.t w Proceedings of Engineering and Technology Innovation , vol. 2, 2016, pp. 23 - 25 24 Copyright © TAETI A key disadvantage of VARTM is a long infusion time because the driving pressure is lo w and the flow resistance offered by the co mpacted preform is high. Recent methods of flow en- hancement have been devised. In the present study, the progressive compression method (PCM), sharing the common characteristic with the articulated resin transfer molding [1] and the vacuum assisted compression resin transfer mo lding [2], is proposed to improve infusion time as shown in Fig. 1. In the process , the s e- quential in flat ion motion of the segmented bag is designed to squeeze out the resin within the wetted preform and drive the resin through the re main ing dry and loose preform. The resin flo w is thus enhanced. In the present study, the effect of seven process parameters, including vacuu m pressure in the mold cavity (factor A), nu mber of the compression segment (factor B), co mpression timing of the ne xt segment (factor C), te mpera - ture of the heated air (factor D), in itiating seg- ment of the heated air (factor E), initia l height of the cavity (factor F) and volu me of the infused resin (factor G), on the PCM comp lete filling process is investigated by applying Taguchi’s method. The typica l VA RTM filling process is also performed for comparison purpose. 2. Method 2.1. Materials In the e xperiments, four layers of bi-directional fiber mat (T GFW-600) we re stacked together to form a preform. The used resin was commerc ial unsaturated polyester resin system (Eterna l Che mica l, 2597PT -6) including 0.5% o f the catalyst. Before resin infusion, resin and hardener (M EKPO) we re mixed well in a we ight ratio 100: 1. The dimen- sions of the rectangular cavity were 280 mm × 150 mm × 10 mm. The cavity thickness could be adjusted by placing an extra plate in the cavity. 2.2. Taguchi’s Method The chosen factors are designed with two levels as shown in Table 1. Thus, the L8(2 7 ) standard matrix can be applicable . Eight ex- periments are performed and the factors corre- sponding to each column are varied a mong the levels within the matrix. Table 1 Chosen factors Fact or Level 1 Level 2 A. vacuum pressure (kP a ) 100 80 B. number of compression seg- ment 2 3 C. compression t iming of t he next segment (sec) 5 10 D. t emperature of the heated air (℃) 20 40 E. init iat ing segment of t he heated air 1 st segment 3 rd segment F. init ial cavit y height (mm) 7 10 G. volume of infused resin (ml) 100 130 For factor B, the position of the compression segment is shown as Fig. 2. Fig. 2 Position of the compression segment The minimizat ion of filling process is the goal of optimization. A signal-to-noise (S/N) ratio depicting the ‘small the better’ character- istic can be calculated using the formula.          n i i y n NS 1 2 10 1 log10/ (1) where yi is the measured property and n corre- sponds to the number of sa mples in each test tria l. The measured property is compiled in decibe l (dB) in the matrix e xperiment. The S/N ratio is also denoted by η. 3. Results and Discussion Performing the technique of the analysis of variance (ANOVA) investigates the relative effect of different factors on the filling t ime as shown in Table 2. The variance ratio, denoted by R, is the ratio of the mean square (MS) due to a factor and the error mean square. That is to say, the larger the value of R, the more important that factor is in influencing the process resp onse η. Therefore, the process variables are ranked in order of importance of vacuum pressure, volume of infused resin, initia l cavity height , tempera - ture of the heated air, co mpression timing of the next segment, nu mber o f co mpression segment and initiat ing segment of the heated air. Re fer- ring to the sum of squares (SS) colu mn , notice that factor A (vacuum p ressure) ma kes the largest contribution to the total sum of squares, name ly, (40.93/57.67) × 100 = 70.97%. Factors air outlet air outlet air inlet Proceedings of Engineering and Technology Innovation , vol. 2, 2016, pp. 23 - 25 25 Copyright © TAETI G (volu me of infused resin), F (in itia l cavity height) and D (temperature of the heated air) ma ke a 20.12% , 6.09% and 2.26% contribution to the total, respectively. Factor E (in itiating segment of the heated air) makes the least co n- tribution to the total, on ly 0.09% . This is because the temperature of the heated air at the outlet is found to be roughly 24℃. Analyses of the ex- perimental η show that the predicted optimu m settings are A1B2C2D2E2F1G1. Table 2 ANOVA Factor Average  by factor level DF SS MS R 1 2 A. -53.8 -58.3 1 40.93 40.93 669.8 B. -56.2 -56.0 1 0.07* 0.07* 1.1 C. -56.2 -55.9 1 0.20 0.20 3.2 D. -56.5 -55.7 1 1.30 1.30 21.4 E. -56.1 -56.0 1 0.05* 0.05* 0.9 F. -55.4 -56.7 1 3.51 3.51 57.5 G. -54.9 -57.3 1 11.60 11.60 189.9 Error 0 0 - T otal 6 57.67 (Error) (2) (0.12) (0.06) * indicates the sum of squares added to estimate the error sum of squares indicated by parenthe- ses. 4. Conclusions A new VARTM process, called progressive compression method, is successfully developed to reduce the filling time . In the present study, the effects of seven process parameters on the filling t ime are investigated by Taguchi’s method. Expe rimental results show that the vacuum pressure, the volume of infused resin and the temperature of the heated air are signif- icant variab les for reducing the filling time , while the in itiat ing timing of the heated air ap- pears to be trivial. References [1] J. H. Choi and C. K. H. Dharan, “Mold fill time and void reduction in resin transfer molding achieved by articulated tooling,” Journal of Composite Materials , vol. 36, no. 19, pp. 2267-2285, October 2002. [2] C. Y. Chang, “Experimental analysis of mold filling in vacuum assisted compression resin transfer molding,” Journal of Reinforced Plastics and Composites, vol. 31, no. 23, pp. 1630-1637, December 2012.