Mechanics of Composite Materials, Vol. 47, No. 3, July, 2011 (Russian Original Vol. 47, No. 3, May-June, 2011)
OPTIMIZATION OF THE INJECTION MOLDING PROCESS FOR SHORT-FIBER-REINFORCED COMPOSITES
Ch.-Sh. Chen,1 T.-J. Chen,1 Sh.-Ch. Chen,2,3 and R.-D. Chien4*
Keywords: Taguchi method, fiber orientation, shear layer A fast and effective methodology integrating the finite-element and Taguchi methods is presented to determine the optimal design conditions of the injection molding process for short-fiber-reinforced polycarbonate composites. The finite-element-based flow simulation software, M-flow, was employed to simulate the molding process to obtain the fiber orientation distributions required. The Taguchi optimization technique was used to identify the optimal settings of injection molding parameters to maximize the shear layer thickness. The effects of four main parameters — the filling time, melt temperature, mold temperature, and injection speed — on the fiber orientation or the shear layer thickness were investigated and discussed. It is found that the dominant parameter is the filling time. The best levels of the four parameters to acquire the thickest shear layer are also identified.
1. Introduction The shortcoming of plastic materials is their low strength, which limits their applications as structural parts. In order to extend the application areas of plastic materials, plastic composites are widely developed by adding various reinforcing fibers to a plastic matrix. Fiber-reinforced composite materials mainly consist of high-strength and high-modulus fibers embedded in or bonded to a plastic matrix. Hence, the strength and the modulus-weight ratio of fiber-reinforced composite materials are markedly superior to those of traditional plastic ones. The low density, high strength, and design flexibility of fiber-reinforced composites are the primary reasons for their use in many structural components in the aircraft, automotive, sporting, and other industries.
Department of Mechanical Engineering, Lunghwa University of Science and Technology, Guishan Shiang 33306, Taiwan Department of Mechanical Engineering, Chung Yuan Christian University, Chung-Li 32023, Taiwan 3 R & D Center for Mold and Molding Technology, Chung Yuan Christian University, Chung-Li 32023, Taiwan 4 Department of Mechanical Engineering, Nanya Institute of Technology, Chung-Li 32024, Taiwan * Corresponding author; e-mail:
[email protected]; tel. 886(2) 29679307; fax: 886(2) 29658315 1 2
Russian translation published in Mekhanika Kompozitnykh Materialov, Vol. 47, No. 3, pp. 519-532 , May-June, 2011. Original article submitted October 8, 2010. 0191-5665/11/4703-0359 © 2011 Springer Science+Business Media, Inc.
359
The injection molding process is widely used in manufacturing plastics, because it allows producing complicated parts with a high dimensional accuracy at an effective cost and a relatively short cycle time. The injection molding of fiber-reinforced composites combines the inherent processing advantages of injection molding with the superior performance characteristics of composite materials. Since different injection molding conditions induce different fiber orientations in fiber-reinforced composites [1-2], their characteristics are easily affected by molding parameters. The fiber orientation in short-fiber-reinforced composites during the injection molding has been the subject of several investigations. To simulate the moving boundary of the melt flow and the fiber orientation during filling the mold, Hung and Shen used the finite-element method [3]. The influence of the fountain flow effect and of conditions of the injection molding process on the fiber orientation was examined. Ho and Hwang [4] pointed out that the fiber orientation density in the shear layer affects the strength and stiffness of thermoplastics, and that the wear properties in the melt flow direction are better than in the transverse one. The finite-element method was utilized by Lee et al. [5] to calculate the flow field during the filling stage of the injection molding process. A mold having four cavities with different filling times was produced for the numerical analysis. A good agreement between the experimental and theoretical results was obtained when the compressibility of the polymer melt was allowed for in the numerical simulation. Numerical predictions of fiber orientation during the injection molding of fiber-filled thermoplastics were studied by Greene and Wilkes [6]. The flow was considered as a nonisothermal, free-surface one and described by a new viscosity model dependent upon temperature, pressure, and fiber concentration. Skourlis et al. [7] found that the fiber orientation could vary significantly across the thickness of injection molded parts. The thicknesses of a shear layer with fiber orientation in the flow direction and a core layer with fiber orientation mainly transverse to the flow were measured by using microscopy. Injection mold filling simulations were undertaken to investigate the characteristic fiber orientations in the shear and core layers, as well as the frozen layer thickness at the end of filling and the gapwise gradients of shear rate. The through-thickness fiber orientation distribution of injection-molded polycarbonate plates was studied by Neves et al. [8]. The fiber length distribution was determined by an image analysis. The effect of flow rate and melt temperature on the stiffness and fiber orientation was discussed. Chang et al. [9] used a methodology that combined the grey relational analysis method and a CAE flow simulation software to simulate the injection molding process and to predict the fiber orientation. The grey relational analysis method was used instead of the classical single-parameter method to find an optimal fiber orientation in short-glass-fiber-reinforced polycarbonate composites. A thorough experimental study on the orientation of flax fibers in a plate processed by injection molding was presented by Aurich and Mennig [10]. The composite stiffness was predicted by using a modified classical laminate theory, including unidirectional models and the orientation averaging. A comparison of the measured and calculated elastic moduli in tension showed a good agreement. Kim et al. [11] found that the characteristics of fiber-reinforced plastic composites depend not only on the quantity of fibers, but also on their shape. During the injection molding of composites, the fiber orientation is governed by the flow state. The molded part tends to be anisotropic. It is very important to clarify the relationship between the fiber orientation and injection molding conditions. Han and Im [12] presented a numerical simulation of the flow field and fiber orientation with a special consideration of the fountain flow effect. It is important to predict the nature of the flow field inside the mold and the flow-induced variation of fiber orientation for an effective design of short-fiber-reinforced plastic parts. The justification for using short-glass-fiber-reinforced injection-molded thermoplastic composites was studied by Zainudin et al. [13]. The fiber orientation distribution plays an important role in forming the properties of fiber-reinforced composite parts. Pontes et al. [14] indicated that the mechanical properties of injection-molded parts of glass-reinforced materials are sensitive to the processing. The fiber orientation pattern is complex and varies during molding. Some commercial simulation programs already allow the prediction of the fiber orientation induced during the flow by the associated stress field. The development of fiber orientation structures during injection molding in a model ribbed plate was investigated by Hine and Duckett [15]. Two materials — short-glass-fiber-filled PBT and short-glass-fiber-filled nylon 66 — were used for experiments. It was found that the matrix type was of secondary importance in controlling the fiber orientation. Ahmad [16] demonstrated the application of multiple devices for controlling the fiber orientations in powder-injectionmolded aluminum composites. The investigation has identified the molding composition to achieve the highest level of fiber orientation in a preferred direction, which should result in improved directional mechanical properties. Vincent et al. [17] 360
indicated that the mechanical properties of short-fiber-reinforced thermoplastic injected components depend on the flowinduced fiber orientation. Hence, there is considerable interest in validating and improving models which link the flow field and fiber orientations to mechanical properties. The paper concerns first the observation and quantification of fiber orientation in a rectangular plate with an adjustable thickness, molded with 30 and 50 wt.% short-fiber-reinforced polyarylamide. Kim and Lee [18] indicated that injection-molded products are heterogeneous and anisotropic, which vastly influences their mechanical properties. The fiber orientation state in a fiber-reinforced polymeric composite has also a profound effect on its dynamic qualities. The fiber orientation patterns of short-glass-fiber-reinforced polypropylene developed in the conventional and nonconventional injection molding were studied by Silva et al. [19]. The mold used allowed for a wide variety of operating modes during its filling, which led to a great versatility in obtaining different fiber orientation patterns. It was found that the prediction of variations in fiber orientation was necessary for effectively designing short-fiber-reinforced plastic parts. Kröner [20] presented an injection molding analysis of fiber-reinforced polymers by using the computer codes Moldflow and ABAQUS and showed how a nonlinear orthotropic behavior of the materials can be modeled starting from the generalized Hooke’s law. The effects of material optimization and compounding processes on the properties of injection-molded fiber composites were studied by Sun [21]. The influence of fiber content, coupling agent, fiber geometry, and fiber distribution on their properties was investigated. The characteristics of injection-molded parts are greatly affected by the direction of fibers, so the prediction of fiber orientation in short-fiber-reinforced parts is of great importance for a good structural design. Meanwhile, the fiber orientations are influenced by parameters of the injection molding process. Hence, by controlling the molding conditions, such as the filling time, mold temperature, etc., a higher strength can be achieved in a particular direction. In order to determine the best levels for controlled parameters, Taguchi methods have been widely used to obtain the optimal injection molding conditions [22-28]. However, the investigations of fiber orientation in short-fiber-reinforced injection molded parts are very limited in number. The presented study integrates the finite-element and Taguchi analysis methods to determine the optimal injection molding parameters for short-fiber-reinforced composites. The finite-element-based commercial package M-flow was employed to simulate the injection molding of the composites to obtain the best results for the fiber orientation and the thickness of shear layer. The Taguchi optimization technique was utilized to identify the optimal settings for control factors to acquire the greatest thickness of the shear layer, and the analysis of variance was used to find out the main control factors of the injection molding process. 2. Taguchi Experiments The fiber orientation can vary significantly across the thickness of injection-molded parts. By controlling the movement of melt flow and mold cooling, it should be possible to make the fiber orientation closer to what is required. Basically, an injection-molded product will possess a better strength and stiffness when its fiber orientation density in the shear layer is higher. However, a well-designed injection mold can give defective products because of an improper setting of process parameters. Hence, the present study utilizes the Taguchi method to determine the optimal processing conditions ensuring a fiber orientation parallel to the flow direction to acquire the greatest thickness of the shear layer. The Taguchi method is a robust parameter design technology that is useful for improving the desired properties of materials by studying the key factors controlling their production and by optimizing the procedure to find the best results. Hence, the Taguchi optimization technique will be used to identify the optimal settings of control factors which would maximize the shear layer thickness in injection-molded fiber-reinforced parts with a minimum variation. The Taguchi experimental method is carried out by utilizing an orthogonal array, the signal-to-noise (S/N) ratio, and the analysis of variance (ANOVA). The fractional experimental design of orthogonal arrays allows one to compute the main and interaction effects by using a minimum number of experiments with a permissive reliability. Then, experimental trials based on orthogonal arrays are conducted to yield the best levels of process parameters associated with the greatest thickness of the shear layer. The S/N ratio is used to identify the quality characteristics which optimize process/product performance. More than 60 different S/N ratios have been established and defined by Taguchi for parameter designs in engineering applications. For the present analysis, the 361
Fig. 1. Configuration of cavities and runners. larger-the-better quality characteristic of the S/N ratio is chosen to solve the fiber orientation problem, since the objective is to maximize the shear layer thickness. The optimal process parameters are determined such that the functional characteristic of the product, which is the maximal sensitivity to the S/N ratio, is optimized. To determine the influence of process parameters on the shear layer thickness, the ANOVA can be utilized. The present study uses the statistical method of analysis of variance and the Taguchi experimental method. Based on the Taguchi analysis technique, the procedures for identifying the optimal settings of parameters of the injection molding process for short-glass-fiber-reinforced polycarbonate composites are as follows. 1. Quality characteristics are selected as targets for experiments — suitable characteristics are selected to specify the target value. 2. Effective control factors and their levels are determined, which is the key to a successful use of the Taguchi experiment method. 3. A proper orthogonal array for the control factors is selected. The degrees of freedom for the orthogonal array should be greater than or equal to the degrees of freedom of process parameters. 4. An appropriate formulation of quality characteristics and the S/N ratio are selected. A loss function is defined to calculate the difference between the experimental and desired values of S/N. There are three categories in an analysis of the S/N ratio: lower-the-better, higher-the-better, and nominal-the-better. 5. The loss function is transformed into the S/N ratio according experimental data, which is used to determine the deviation of the quality characteristic from its target value. 6. Auxiliary tables and response diagrams are used to distinguish the influence of quality characteristics with various factors. The significance of the factors is identified, and the data obtained are analyzed using the ANVOA. 3. Experimental Procedures and Results The objective of this study was to present a fast and effective methodology, which integrates the Taguchi and finiteelement methods, to obtain optimal design conditions of the injection molding process for short-fiber-reinforced polycarbonate composites. The Taguchi method was used to identify the optimal settings of control factors which would maximize the shear layer thickness in injection-molded short-fiber-reinforced polycarbonate composites, and the finite-element-based commercial package M-flow was employed to simulate the injection molding for the experimental trials determined by orthogonal arrays to calculate the fiber orientations and the shear layer thickness in the composites. The configuration of cavities and runners in the injection mold is shown in Fig. 1. The polycarbonate reinforced with 30 wt.% short glass fibers was used in this study, whose basic physical properties were as follows: density 0.73817 kg/cm3, specific heat 2420 J/kg °C, thermal conductivity 0.27 W/m °C, and glass-transition temperature 383 K. In general, the quality of injection-molded products are affected by the cooling time, injection pressure, injection speed, injection time, filling time, melt temperature, ejecting pressure, die temperature, die geometry, material properties of the melt, 362
TABLE 1. Control Factors and Their Levels for Experimental Variance Control factor
Labels
Level 1
Level 2
Level 3
Filling time, s
A
1
3
7
Melt temperature, °C
B
270
290
310
Mold temperature, °C
C
65
80
95
Injection speed, %
D
20
60
100
TABLE 2. Experimental Layout of the L9 Orthogonal Array Factors Experiment number
1 2 3 4 5 6 7 8 9
Filling time
Melt temp.
Mold temp.
Injection speed
A
B
C
D
1 1 1 2 2 2 3 3 3
1 2 3 1 2 3 1 2 3
1 2 3 2 3 1 3 1 2
1 2 3 3 1 2 3 2 1
TABLE 3. Fiber Orientation of Short-Fiber-Reinforced Polycarbonate Composites Experiment number
The effective thickness
1
2
3
4
5
6
7
8
9
0
4
7
4
5
5
5
4
5
5
0.123
5
5
5
7
7
8
8
8
9
0.243
5.8
6
8
8
8
45
50
55
60
0.360
8
10
12
75
78
80
78
78
78
0.474
75
80
83
85
85
83
82
85
85
0.583
87
88
88
88
88
87
88
89
90
0.685
90
90
90
93
95
93
100
102
105
0.779
95
97
97
96
104
105
103
107
110
0.864
105
110
110
110
110
110
110
110
110
363
TABLE 4. Effective Thickness of the Shear Layer and the S/N Ratio Experiment number
Factors
The effective thickness
S/N ratio
A
B
C
D
1
1
1
1
1
0.504
17.826
2
1
2
2
2
0.488
19.596
3
1
3
3
3
0.492
18.551
4
2
1
2
3
0.600
20.690
5
2
2
3
1
0.601
20.692
6
2
3
1
2
0.702
21.996
7
3
1
3
3
0.706
20.553
8
3
2
1
2
0.71
22.010
9
3
3
2
1
0.714
22.275
0.613
20.465
Ave:
TABLE 5. Response Table of the S/N Ratio Level
Factors A
B
C
D
1
0.631
0.739
0.775
0.742
2
0.77
0.736
0.737
0.769
3
0.846
0.772
0.736
0.735
Effect
0.215
0.036
0.039
0.034
Rank
1
3
2
4
melt speed, and heat transfer in the flow field. However, as known from [4, 9], the main factors affecting the fiber orientation, which are directly related to the properties of fiber-reinforced parts, are the melt temperature, mold temperature, filling time, and injection speed. Their levels considered to influence the fiber orientation are indicated in Table 1. Each factor was studied at three levels. In the experiment, the melt temperature was set to 270, 290, and 310°C, the mold temperature to 65, 80, and 95°C, the filling time to 1, 3, and 7 s, and the injection speed to 20, 60, and 100%. The number of experimental trials used in the fiber orientation experiment was nine, so that an orthogonal array L9 of factors with three levels was employed. Table 3 presents the experimental layout showing all the control factors and their levels considered, where levels 1, 2, and 3 denote the low, medium, and high values of setting, respectively. For each of the nine experimental trials, the finite-element results for fiber orientations in the short-fiber-reinforced polycarbonate composites, from the center to the surface of cross section of a reinforced polycarbonate plate, are given in Table 3. The cross section can be divided into three distinct layers, and they are easily distinguishable by the fiber orientation relative to the melt flow direction. In the frozen layer near the plate surface, the fibers are oriented randomly, in part perpendicularly to the melt flow direction; in the intermediate shear layer, the fiber orientation is obviously parallel to the flow direction; in the core layer, the fibers are perpendicular to the melt flow direction. It is well known that the optimal fiber orientation is parallel to the melt flow direction, i.e., the fiber orientation is 90° to the injection direction, but with a 10-20° variation due to the fountain-flow effect. The layer with a fiber orientation angle around 90-110° can be considered as a shear layer. For example, 364
0.9
S/N
0.8
0.7
0.6
0.5 A1 A2 A3
B1 B2 B3
C1 C2 C3
D1 D2 D3
Fig. 2. Response diagram of the S/N ratio.
TABLE 6. Analysis of Variance of the Effective Thickness Source of variation
Sum of squares
DOF
Variance
F
Probability
Significant
A B C D
112,785.14 7235.97 1774.25 711.23
2 2 2 2
56,392.57 3617.98 887.13 355.62 Pooled
557.93 35.79 8.78 3.52
12.95 24.56 0.90 3.82
Yes Yes Yes Yes No
Pooled error
14,557.46
2
0
Total
137,064.05
10
Other
Note: at least a 90% confidence in experimental trial 1, the core layer is located between the effective thicknesses 0 and 0.360; the shear layer is between 0.360 and 0.864; the frozen layer is between 0.864 and 1.000. In a similar way, the three layers in the other experimental trials can be identified. It can be observed that the core and frozen layers occupy about 30~50 and 10%, respectively, of the total thickness of the plate. The shear layer thickness evaluated here will be used to compute the S/N ratio, because it is the output performance characteristic most relevant to the problem investigated. The S/N ratio is regarded as the measure of noise variance for each experiment when uncontrolled noise factors are present. Since the purpose of the present study was to maximize the thickness of the shear layer, the S/N ratio related to the larger-the-better quality characteristics was selected. It is generally used for such quality characteristics as the strength and service life of a component. Since the composite with a thicker shear layer will have a greater strength, the calculated S/N ratio related to the shear layer thickness can be used to predict the actual manufacturing conditions appropriately. Table 4 shows the values of S/N and the thicknesses of shear layers corresponding to each trial condition, where the effective thickness of shear layer was transformed to the S/N ratio. The S/N response data and diagram are given in Table 5 and Fig. 2, which were used to determine the optimal control factor settings to maximize the shear layer thickness. Table 5 presents the S/N response values at three levels and the effect of each factor on the S/N ratio. Figure 2 illustrates the effects of control factors by showing S/N values at three settings of levels for each factor. The factor settings corresponding to the greatest S/N ratio should be the optimum process conditions. Using the best levels of control factors should yield the maximum effective thickness of the shear layer. In Table 5, a higher value of 365
TABLE 7. Effect of Various Factors on the Effective Thickness Mold temperature, °C The effective thickness, mm Melt temperature, °C The effective thickness, mm Filling time, s The effective thickness, mm Injection speed, % The effective thickness, mm
65
70
75
80
85
90
95
0.765
0.767
0.765
0.767
0.763
0.786
0.786
260
270
280
290
300
310
0.811
0.826
0.826
0.8389
0.845
0.857
1
2
3
4
5
6
7
0.757
0.784
0.820
0.827
0.838
0.837
0.847
20
30
40
50
60
70
80
90
100
0.718
0.728
0.757
0.757
0.779
0.793
0.781
0.781
0.769
the effect represents a greater influence of the control factor on the effective thickness. Hence, the table shows that the most influential factor is the filling time, followed by the mold temperature and the injection speed. In order to decide which level is better to obtain the maximum thickness of the shear layer, the S/N response values were compared at three setting levels for each factor. According to the S/N response table and diagram, the best combination for the process condition should be A3B3C1D2. The best levels of processing parameters that can maximize the shear layer are 7 s for the filling time, 310°C for the melt temperature, 65°C for the mold temperature, and 60% for the injection speed. In Table 6, the total variance of the effective thickness obtained from all experiments was divided into the sum of squares for each factor. The sum of squares then was divided by the degree of freedom and then converted into the statistic quantity F. The level of confidence was set to 0.05 to determine the significant factor and to improve the accuracy of experiments. From the results, it is seen that the dominant factor for the effective thickness is the filling time. To study the influence of each factor on the effective thickness, the best levels of the optimal control factor settings obtained were used as a start. One factor was changed, while the others remained the same as the best levels for every simulation. In such a way, the influence of each factor on the effective thickness was obtained, as shown in Table 7. It can be observed that the effective thickness increases as the mold temperature exceeds 90°C, the effective thickness grows with melt temperature, and the thickness is smallest at 1 s. The results also show that a longer packing time increases the effective thickness. The thickness increases gradually until the injection speed reaches 70%. When the injection speed exceeds 70%, the effective thickness decreases. 4. Conclusions An effective method based on the finite-element and Taguchi methods and on the analysis of variance has been developed to quantify the fiber orientation distributions in short-fiber-reinforced composites and to determine the optimal design conditions providing the greatest thickness of shear layer in the composites. Three distinct layers — frozen, shear, and core ones — were found to develop under various injection molding conditions. The fiber orientation and layer thickness of injectionmolded parts were sensitive to the injection process. This study demonstrates that the integration of a simulation analysis and the Taguchi method are useful for identifying the optimal settings of process parameters for any injection molding operation. It is clarified that the dominant process parameter of injection molding is the filling time, followed by the mold temperature and the injection speed. It is believed that such an analysis can provide the engineers with an efficient method for identifying the optimal process parameters of injection molding to enhance the mechanical properties of short-fiber-reinforced composites. 366
REFERENCES 1. T. Matsuoka, J. Takabatake, Y. Inoue, and H. Takahashi, “Prediction of fiber orientation in injection molding parts of short-fiber-reinforced thermoplastics,” Polym. Eng. Sci., 30, 957-966 (1990). 2. G. S. Springer and Q. Wang, “The effects of form on the mechanical properties of glass fiber reinforced composites,” J. Rein. Plastics Compos., 13, 448-466 (1994). 3. C. F. Hung and Y. K. Shen, “Numerical simulation of fiber orientation in injection mold filling,” Int. Com. Heat Mass Tran., 22, 791-802 (1995). 4. K. C. Ho and J. R. Hwang, “Impact fatigue of short glass fiber reinforced polycarbonate,” J. Rein. Plastics Compos., 16, 903-905 (1997). 5. S. C. Lee, D. Y. Yang, J. Ko, and J. R. Youn, “Effect of compressibility on flow field and fiber orientation during the filling stage of injection molding,” J. Mater. Process. Tech., 70, 83-92 (1997). 6. J. P. Greene and J. O. Wilkes, “Numerical analysis of injection molding of glass fiber reinforced thermoplastics. Part 2: Fiber orientation,” Polym. Engng. Sci., 37, 1019-1035 (1997). 7. T. P. Skourlis, C. Chassapis, and S. Manoochehri, “Fiber orientation morphological layers in injection molded long fiber reinforced thermoplastics,” J. Therm. Compos. Mater., 10, 453-475 (1997). 8. N. M. Neves, G. Isdell, A. S. Pouzada, and P. C. Powell, “On the effect of the fiber orientation on the flexural stiffness of injection molded short fiber reinforced polycarbonate plates,” Polym. Compos., 19, 640-651 (1998). 9. S. H. Chang, J. R. Hwang, and J. L. Doong, “Optimization of the injection molding process of short glass fiber reinforced polycarbonate composites using grey relational analysis,” J. Mater. Process. Tech., 97, 186-193 (2000). 10. T. Aurich and G. Mennig, “Flow-induced fiber orientation in injection molded flax fiber reinforced polypropylene,” Polym. Compos., 22, 680-689 (2001). 11. E. G. Kim, J. K. Park, and S. H. Jo, “A study on fiber orientation during the injection molding of fiber-reinforced polymeric composites: Comparison between image processing results and numerical simulation,” J. Mater. Proess. Tech., 111, 225-232 (2001). 12. K. H. Han and Y. T. Im, “Numerical simulation of three-dimensional fiber orientation in injection molding including fountain flow effect,” Polym. Compos., 23, 222-238 (2002). 13. E. S. Zainudin, S. M. Sapuan, S. Sulaiman, and M. M. H. M. Ahmad, “Fiber orientation of short fiber reinforced injection molded thermoplastic composites: A review,” J. Inject. Mold. Tech., 6, 1-10 (2002). 14. A. J. Pontes, N. M. Neves, and A. S. Pouzada, “The role of the interaction coefficient in the prediction of the fiber orientation in planar injection moldings,” Polym. Compos., 24, 358-366 (2003). 15. P. J. Hine and R. A. Duckett, “Fiber orientation structures and mechanical properties of injection molded short glass fiber reinforced ribbed plates,” Polym. Compos., 25, 237-254 (2004). 16. F. Ahmad, “Orientation of short fibers in powder injection molded aluminum matrix composites,” J. Mater. Process. Tech., 169, 263-269 (2005). 17. M. Vincent, T. Giroud, A. Clarke, and C. Eberhardt, “Description and modeling of fiber orientation in injection molding of fiber reinforced thermoplastics,” Fluid Phase Equil., 234, 6719-6725 (2005). 18. J. W. Kim and D. G. Lee, “Fiber orientation state depending on the injection mold gate variations during FRP injection molding,” Key Eng. Mater., 321, 938-941 (2006). 19. C. A. Silva, J. C. Viana, H. Van, W. J. Ferrie, and A. M. Cunha, “Fiber orientation in divergent/convergent flows in expansion and compression injection molding,” Polym. Compos., 27, 539-551 (2006). 20. C. Kröner, H. Altenbach, and K. Naumenko, “Coupling of a structural analysis and flow simulation for short-fiberreinforced polymers: property prediction and transfer of results,” Mech. Compos. Mater., 45, 249-256 (2009). 21. Z. Y. Sun, H. S. Han, and G. C. Dai, “Mechanical properties of injection-molded natural fiber-reinforced polypropylene composites: Formulation and compounding processes” J. Rein. Plastics Compos., 29, 637-650 (2010). 22. C. S. Chen, W. S. Cheng, T. S. Wang, and R. D. Chien, “Optimum design of gas-assisted injection molding,” J Rein. Plastics Compos., 24, 1577-1586 (2005). 367
23. H. Oktem, T. Erzurumlu, and I. Uzman, “Application of Taguchi optimization technique in determining plastic injection molding process parameters for a thin-shell part,” Mater. Design, 28, 1271-1278 (2007). 24. T. Erzurumlu and B. Ozcelik, “Minimization of warpage and sink index in injection-molded thermoplastic parts using Taguchi optimization method,” Mater. Design, 27, 853-861 (2006). 25. C. P. Fung and P. C. Kang, “Multi-response optimization in friction properties of PBT composites using Taguchi method and principle component analysis,” J. Mater. Process. Tech., 170, 602-610 (2005). 26. S. H. Tang, Y. J. Tan, S. M. Sapuan, S. Sulaiman, N. Ismail, and R. Samin, “The use of Taguchi method in the design of plastic injection mould for reducing warpage,” J. Mater. Process. Tech., 182, 418-426 (2007). 27. W. S. Cheng, C. S. Chen, S. C. Chen, and R. D. Chien, “Investigation of the effects of injection molding processing parameters on conductive polymeric composites for electromagnetic interference shielding effectiveness,” PolymerPlastics Tech. Eng., 48, 216-220 (2009). 28. M. Altan, “Reducing shrinkage in injection moldings via the Taguchi, ANOVA, and neural network methods,” Mater. Design, 31, 599-604 (2010).
368
