2009 Third Asia International Conference on Modelling & Simulation
Optimization of Warpage Defect in Injection Moulding Process using ABS
Material
A. H. Ahmad1* Z. Leman2, M. A. Azmir1, K. F. Muhamad1, W.S.W. Harun1, A. Juliawati1, A.B.S.
Alias1
1
Faculty of Mechanical Engineering, Universiti Malaysia Pahang, Lebuhraya Tun Razak, 26300
Kuantan, Pahang, Malaysia.
2
Department of Mechanical and Manufacturing Engineering, Universiti Putra Malaysia, 43400
UPM, Serdang Malaysia.
Email: asnul@ump.edu.my1, zleman@eng.upm.edu.my2
Abstract
Plastic injection moulding process produces
various defects such as warpage, sink marks, weld
lines and shrinkage. The purpose of present paper is
to analyze the warpage defect on Acrylonitrile
Butadiene Styrene (ABS) for selected part using FEA
simulation. The approach was based on Taguchi’s
Method and Analysis of Variance (ANOVA) to
optimize the processing parameters namely packing
pressure, mould temperature, melt temperature and
packing time for effective process. It was found that
the optimum parameters for ABS material are packing
pressure at 375 MPa, mould temperature at 40 0C,
melt temperature at 200 0C and packing time at 1 s.
Melt temperature was found to be the most significant
factor followed by packing time and mould
temperature. Meanwhile, packing pressure was
insignificant factor contributing to the warpage in
present study.
1. Introduction
One of the plastic processing techniques is using
injection moulding machine. The injection moulding
process actually is the most practical and cost effective
to produce plastic products [1]. Plastic injection
moulding defects such as warpage, sink marks, weld
lines and shrinkage are the common defects occur in
the plastics injected parts.
The main cause of warpage is commonly known
as the variation in shrinkage towards injection process
of thin-shell plastic parts [2]. Thick sections cool
slower than thin sections. The thin section first
solidifies, and the thick section is still not fully
solidified. As the thick section cools, it shrinks and the
material for the shrinkage comes only from the
unsolidified areas, which are connected, to the already
solidified thin section. In practice, the dimensions,
potential for warpage and internal stress level for a
plastics part will influenced by a variety of material,
978-0-7695-3648-4/09 $25.00 © 2009 IEEE 470
DOI 10.1109/AMS.2009.120
part geometry, tooling and processing related factors
[3].
Parts with thick wall are most difficult to cool and
pack. Thicker sections take longer to cool and required
additional packing [1] .When parts have both thick
and thin sections, gating into the thick section is
preferred because it enables packing of the thick
section, even if the thinner sections lead to shrinkage
related internal stresses in the wall thickness regions.
The shrinkage part easily transform into warpage if the
parameter settings are not well control [4].
The warpage and shrinkage is understood as the
process of non-uniform (heterogeneous) change of the
geometrical dimensions of products in time resulting
in a change (distortion) of their original form [5].
However, the researchers are using different parameter
settings according to the part and raw material
available thus it will contribute to the different results
compare to each other.
The scope of this study is focusing on the
simulation of warpage defect on the raw materials
including determination of the effective parameters
that contribute to the defect, the selection of the
orthogonal arrays (OAs) and determination of the
optimum parameter. The selection of the orthogonal
arrays (OAs) depends on the level and parameter
involved thus the 3 levels and 4 parameters were
chosen.
The chosen parameters and level influenced the
type of orthogonal arrays and the Taguchi L18
orthogonal arrays were used. Finally, the optimum
parameters were determined by exploiting S/N ratio
and ANOVA.
2. Experiment Set Up
2.1 Tools and Material
A series of simulations were conducted using
Moldflow 2007 software to observe the warpage
defect on the part. The simulation processes start with
the development of the Computer Aided Design
(CAD) data using Solidworks software. The selected 2.3 Simulation Approach
part for this study is identification card holder where it
was chosen because it could easily deform the Finite Element (FE) analysis of the part is
warpage effect. The Acrylonitrile Butadiene Styrene performed using the commercial MoldFlow plastic
(ABS) material was used for the experiment. The insight software (MPI) to carry out all the injection
material properties for ABS are shown in Table 1 [6]. moulding experiment. This software uses the finite
elements and finite difference methods to calculate a
Table 1 series of mathematical functions representing the
Material Properties for ABS moulding process. This simulation provide
Material Structure Amorphous Melt information such as distribution and variation of
Temperature (0C) 200~240 temperature, pressure, flow rate, skin property,
Mould Temperature(0C) 40~80 molecule orientation, shear stress and shear rate of the
Max Shear Stress (MPa) 0.3 material in filling, packing and cooling stages. The
Max Shear Rate (1/s) 50000 results together with the moulding condition can be
Melt Density (g/cm3) 0.94752 useful in product and mould design, where it may
Solid Density (g/cm3) 1.0432 include the optimal gate location and runner size,
Elastic Modulus (MPa) 2240 prediction of weld line location, shrinkage and
warpage. In addition the moulding conditions of the
2.2 Selection of the Parameters injection process can be optimized. The simulation
started with the conversion process from CAD data
The selection of the parameters involved in this into CAE data. The development of the CAD data
experiment based on the literature studies that have before that used Solidworks software was transferred
been made before. Based on the literature, a lot of into the Moldflow plastics insight software. Moldlfow
parameters influence the warpage defect such as filling plastics insight software has the capability to convert
speed, melt temperature, packing time, cooling time, directly the CAD data from Solidworks software.
injection time, injection pressure, packing pressure
and mould temperature. Huang et al., [7] claims that, 3. Results and Discussions
the most effective factors contributes warpage are
ranks as follows, packing pressure, mould 3.1 S/N ratio Approach
temperature, melt temperature and packing time. The
parameters and their levels are shown in Table 2. The S/N ratio approach measure the quality
To determine the best set of parameter among the characteristic deviating from the desired values. The
effective by reducing the number of experiments, the S/N ratio approach using the average values to convert
Taguchi method has been chosen [8]. Hence, the the experimental results into the value for the
selection of the factors that will affect warpage, evaluation characteristic in the optimum parameter
selection of the factor levels and selection of analysis. The S/N ratios can be defined as [8]:
orthogonal array (OA) based on Taguchi method
needed. From the number of factors and levels that η = -10 log (M.S.D.) (1)
have been selected, the L18 was chosen as an OA
where M.S.D. is the mean-square deviation for the
because it suitable for 4 factors with 3 levels. The L18
output characteristics.
OA has 18 numbers of trial runs. The L18 OA are
The S/N ratio characteristic can be divided into
presented in Table 3.
three stages: the nominal-the better, the smaller the
better and the higher the better when the quality
Table 2 characteristics are continuous for engineering analysis
The parameter for 3 levels of selected factors
[1]. Since the objective of this study is to reduce the
Factors Level 1 Level 2 Level 3
shrinkage problem through optimum parameters in
Packing
injection moulding process, the smaller-the better
Pressure, A (MPA) 300 375 450
quality characteristic was employed in this study. The
Mould
M.S.D. for the smaller the better quality characteristic
Temp., B (0C) 40 60 80
can be expressed by [8]:
Melt
Temp., C (0C) 200 220 240 M.S.D. = 1/N (Yi2) (2)
Packing
Time, D (s) 0.6 0.8 1.0 where Yi is the value of the warpage for the ith test. N
is the total number of data points. Meanwhile to

471
Table 3:
The shrinkage and S/N ratio values
_____________________________________________________________________________________________
Trial no. Packing Mould Melt Packing Warpage(mm) S/N (dBi)
Pressure, Temp., Temp., Time
A (MPa) B(0C) C(0C) D(s)
1 300 40 200 0.6 0.4679 6.597
2 300 60 220 0.8 0.4036 7.881
3 300 80 240 1.0 0.4469 6.996
4 375 40 200 0.8 0.3136 10.072
5 375 60 220 1.0 0.3755 8.508
6 375 80 240 0.6 0.4908 6.182
7 450 40 220 0.6 0.4046 7.859
8 450 60 240 0.8 0.4382 7.167
9 450 80 200 1.0 0.3608 8.855
10 300 40 240 1.0 0.3969 8.026
11 300 60 200 0.6 0.3823 8.352
12 300 80 220 0.8 0.4328 7.274
13 375 40 220 1.0 0.3455 9.231
14 375 60 240 0.6 0.4646 6.658
15 375 80 200 0.8 0.3862 8.264
16 450 40 240 0.8 0.4211 7.512
17 450 60 200 1.0 0.3205 9.883
18 450 80 220 0.6 0.4521 6.895

calculate minimum warpage defect, it can be value of each factor. Thus, the result obtained are
expressed by [1]: Packing Pressure A, level 2(A2), Mould Temperature
B, level 1(B1), Melt Temperature C, level 1(C1) and
W1op1+W2op2+W3op3+W4op4 – 3 x (Y) (3) Packing Time D, level 3(D3).

where Wop is the optimum warpage value for its level.
Y is the total defect for warpage in the cycle.
The warpage response diagram for each parameter
at level 1(minimum), level 2(medium) and level
3(maximum) were created by utilizing the S/N ratio
values. The values obtained by this process are
recorded in form of the graph in Figure 1:
3.2 Analysis of Variance (ANOVA)
In order to determine the significant factors that
contribute to the warpage, the result have been
analyzed using Analysis of Variance (ANOVA). The
ANOVA concept involving the relative percentage
contribution among the factor is determined by
comparing their relative variance. The ANOVA will
compute the quantities such as degree of freedom (f),
sum of squares (S), variance (V), F-ratio (F), and
percentage contribution (P). The results can be
summarized in Table 4.

Table 4
ANOVA table

Source f S(10-3) V(10-3) F P(%)

P.Pressure,A 2 2.30 1.15 2.02 5.38
Md.Temp,B 2 4.70 2.35 4.12 10.98
Mt. Temp,C 2 15.30 7.65 13.42 35.75
P.Time,D 2 14.80 7.40 12.98 35.58
Figure 1: S/N response diagram Error 10 5.70 0.57 12.31
Total 18 42.80 100
The best set of combination parameters can be
determined by selecting the level with the highest

472
 Because on the F.10(2,10)=2.9245 [8], it is
obvious to conclude that mould temperature(B), melt
temperature(C) and packing time(D) significantly
affect the warpage defect with 90% confident
intervals. The percentage weight for each factor shows
the influence factor that contributes to the warpage
defect. The melt temperature contributes the most
which is 35.75% meanwhile for the packing time, it
contribute 35.58% which is the second factor that
ranks as the influence factor for warpage defect. The
mould temperature contributed 10.98% and lastly
packing pressure just contributed 5.38% and it can be
said that the packing pressure is not the significant
factor for the warpage defect in this study. The error
values are quite high which is contributed 12.31%
where definitely influenced the warpage values in this
study.
3.3 Effect of Processing Factors
The effects of processing factors for all
parameters are shown in Figure 2. The trend of the
warpage defect for each parameter is similar to each
other except packing pressure and packing time. When
the melt temperature and mould temperature
parameter settings increased, the warpage defect will
significantly increase. Meanwhile for the packing
pressure and packing time, the warpage decreased
when the parameter settings increased.
(a) (b)
(c) (d)
Figure 2: Processing factors for (a) packing pressure,
(b) mould temperature, (c) melt temperature and (d)
packing time
473
Based on Huang [2], the warpage defect will
increase significantly when the packing pressure was
set more than 85% of the injection pressure value, and
it refers to 425MPa for packing pressure value in this
study. It shows that if the packing pressure value is set
beyond 425MPa the warpage value will increase and
to get smaller warpage value the packing pressure
need to be set not exceed than 425MPa.
Meanwhile, the warpage values will increase
significantly when the packing time was set on the
range 1-2 seconds [2]. The maximum packing time for
this study was set 1 second and the simulation not
exceed beyond that limit. The small warpage values
occurred when the packing time increased not exceed
more than 1 second thus it give the bright
understanding that to get smaller warpage defect, the
packing time setting need to be set at 1 second.
The other parameters involved, mould
temperature and melt temperature; the warpage defect
will increase constantly if the parameter settings for
these two parameters increase [2]. The conclusion for
these two parameters is they needs to be set at the
lowest level value to get the smaller warpage defect.
3.4 Verification Test
By using the optimum parameters, the minimum
warpage was estimated based on the equation 3. The
calculations for minimum warpage are shown as
follows:
Minimum warpage
=A2 + B1 + C1 + D3 – 3 x (Y)
=0.3910+0.3816+0.3709+0.3704 – 3(0.4058)
=1.5139 – 1.2174
=0.2965mm
Meanwhile for the minimum warpage from simulation
is 0.2780mm. Thus the error between simulation and
calculated value is 6.65%.
4. Conclusion
In this study, the warpage for ABS material show
the different values when the parameter settings were
changed. The used of Taguchi’s OAs simplified the
experiment runs and ANOVA shows the influenced
factor that contributed to the warpage defect. The
optimum parameters for the ABS material are packing
pressure at level 2(375MPa), mould temperature at
level 1(400C), melt temperature at level 1(2000C) and
packing time at level 3(1s). However among these
factors, packing pressure is not a significant factor.
Meanwhile for the processing factors, it shows the
warpage value for mould temperature and melt
temperature was increased if the parameter settings for
its factor increased but in versus warpage values for
packing time and packing pressure decrease if the
parameter settings increased.

5. References

[1]Hasan Oktem, Tuncay Erzurumlu, and Ibrahim

Uzman, 2007. Application of Taguchi optimization
technique in determining plastic injection molding
process parameters for a thin-shell part. Materials and
Design 28, 2007 1271–1278.

[2]Ming-Chih Huang, Ching-Chih Tai, 2001. The

effective factors in the warpage problem of an
injection-molded part with a thin shell feature.
Materials Processing Technology 110(2001) 1-9.

[3]Robert A. Malloy, 1995, Injection Moulding

Processes 3rd Edition, Canada, Prentice Hall.

[4]S.H. Tang , Y.J. Tan, S.M. Sapuan, S. Sulaiman, N.

Ismail, and R. Samin, 2007. The use of Taguchi
method in the design of plastic injection mould for
reducing warpage. Journal of Materials Processing
Technology 182, 2007 418–426.

[5]Hossein Hosseini, Boris Vasilich Berdyshev, Arjom

and Mehrabani-Zeinabad, 2006, A solution for
warpage in polymeric products by plug-assisted
thermoforming. European Polymer Journal 42, 2006
1836-1843.

[6]Moldflow2007.User manual Accessed on 13

February 2008.

[7]Chiaming Yen, J.C. Lin, Wujeng Li, and M.F.

Huang, 2006. An adductive neural network approach
to the design of runner dimensions for the
minimization of warpage in injection mouldings.
Journal of Materials Processing Technology 174 ,
2006 22–28.

[8]Ranjit K. Roy, 1990. A primer on the taguchi

method.Van Nostrand Reinhold, New York, USA.

474