International Communications in Heat and Mass Transfer 37 (2010) 13591365

Contents lists available at ScienceDirect

International Communications in Heat and Mass Transfer

j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / i c h m t

Inuence of injection parameters and mold materials on mechanical properties of

ABS in plastic injection molding
Babur Ozcelik a,, Alper Ozbay a, Erhan Demirbas b
a
b

Department of Mechanical Engineering, Gebze Institute of Technology 41400 Gebze-Kocaeli/Turkey

Department of Chemistry, Gebze Institute of Technology, 41400 Gebze-Kocaeli/Turkey

a r t i c l e

i n f o

Available online 1 August 2010

Keywords:
Plastic injection molding
ABS
Mechanical properties
DOE
Taguchi optimization method
Regression analysis

a b s t r a c t
This study optimized effect of injection parameters such as melt temperature, packing pressure, cooling time
and injection pressure on the mechanical properties of AcrylonitrileButadieneStyrene (ABS) moldings.
Mold materials having two different thermal conductivities, 191 W/mK for aluminum 2000 series and 50 W/
mK for AISI 1020 at 25 C were selected to use in experimental studies. Taguchi's L9(34) orthogonal array
design was employed for the experimental plan. Mechanical properties of ABS specimens such as elasticity
module, tensile strength and tensile strain at yield, tensile strain at break, exural modules and izod impact
strength (notched) were measured by using some test methods. Signal to noise ratio for mechanical
properties of ABS using the Taguchi method was calculated and effect of the parameters on mechanical
properties was determined using the analysis of variance. Linear mechanical models were also created by
using regression analysis.
2010 Elsevier Ltd. All rights reserved.

1. Introduction
Mold materials with the different thermal properties affect the
mechanical properties of the plastic parts in injection molding.
Aluminum (Al) mold shows advantages in terms of weight, heat
transfer and low production cost as compared to steel materials.
Aluminum molds are used to produce between 5,000 and 50,000
plastic parts [1].
Researches on the mechanical properties of injection molded of
ber reinforced thermoplastics and injection molding parameters
are realized. Sadabadi and Ghasemi investigated the effects of the
injection molding process parameters including injection ow rate,
mold wall temperature, packing pressure using short ber reinforced
polystyrene composites which could affect ber orientation and
tensile modulus of injection molded parts [2]. Yang studied tribological behaviors and mechanical properties of polycarbonate reinforced with 20% short glass ber and 6% polytetrauoroethylene using
Taguchi's orthogonal arrays and analysis of variance (ANOVA) under
different conditions of injection molding such as lling time, melt and
mold temperatures and packing pressure [3]. Yang also examined the
mechanical properties and tribological behaviors mainly ultimate
stress and surface roughness using grey relational analysis [4].

Communicated by W.J. Minkowycz.

Corresponding author.
E-mail address: ozcelik@gyte.edu.tr (B. Ozcelik).
0735-1933/$ see front matter 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.icheatmasstransfer.2010.07.001

Guerrica-Echevarria et al. determined the mechanical properties of

injection molded parts using injection rate, melt temperature and
screw rotation rate during plasticization [5]. Kenig et al. analyzed the
effect of process parameters namely, mold and melt temperatures,
packing pressure, injection speed and cooling time for tensile modulus using both multivariate regression analysis and articial neural
network [6].
Bociaga examined mechanical and thermal properties of HDPE
moldings for mold temperature and injection velocity. He noticed that
higher tensile strength, yield stress, tensile modulus and hardness
were obtained when the mold temperature was increased [7].
Nagaoka et al. [8] studied tensile and three-point exural tests from
the sandwich and normal injection moldings for polypropylene
material. Shie [9] investigated optimization of injection molding
process for mechanical properties of polypropylene with regression
neural network. The effect of the process parameters on warpage [10]
and sink marks [11] in plastic injection molding and on drill bit [12] in
drilling were studied with the Taguchi method.
In this study, changes in mechanical properties of Acrylonitrile
ButadieneStyrene (ABS) material were optimized with injection
parameters for two mold materials (aluminum 2000 series and AISI
1020). Taguchi's L9(34) orthogonal array design was used for experimental plan. Signal to noise ratio (S/N) for mechanical properties of
each mold material was determined and effect of injection parameters
on mechanical properties was carried out using analysis of variance
(ANOVA). Linear mechanical models were also obtained from regression analysis.

1360

B. Ozcelik et al. / International Communications in Heat and Mass Transfer 37 (2010) 13591365

Table 1
An orthogonal array L9(34) of Taguchi.
A

Melt temperature
(C)

Packing pressure
(MPa)

Cooling time
(sn)

Injection pressure
(MPa)

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
2
3
1

Fig. 2. Specimen dimensions for three-point exural test (ISO 178).

Table 2
The process parameters and levels.
Process parameters
Melt temperature (C)
Packing pressure (MPa)
Cooling time (s)
Injection pressure (MPa)

A
B
C
D

Level 1

Level 2

Level 3

200
28
16
36

240
34
19
43

280
39
22
50

Fig. 3. Specimen dimensions for izod impact test (ISO 180).

parameters on mechanical properties. In this way, optimal levels of

the process parameters can be estimated.

3. Experimental
Table 3
Physical and mechanical properties of ABS.

3.1. Taguchi's orthogonal arrays

Physical properties
Melt ow index

ASTM D1238

200 C/5 kg

1.8 g/10 min

Mechanical properties
Tensile strength
Flexural strength
Flexural modules
Izod impact strength

ASTM D638
ASTM D790
ASTM D790
ASTM D256

5 mm/min
2.8 mm/min
2.8 mm/min
1/4 in.

47 MPa
66 MPa
2200 MPa
235 J/m

2. Denition of the Taguchi method

The Taguchi method developed by Taguchi consists of three stages
which are system, parameters, and tolerance designs, respectively
[13]. The system design involves the application of scientic and
engineering knowledge required in manufacturing of a product. The
parameter design is employed to nd optimal process values for
improving of the quality characteristics. The tolerance design is used
for determining and analyzing of the tolerances in optimal settings
recommended by the parameter design. By applying the Taguchi
method based on orthogonal arrays, time and cost required for
conducting of the experiments can be reduced. Taguchi recommends
the use of the S/N ratio for determination of the quality characteristics
implemented in engineering design problems. The S/N ratio characteristics with signed-target type can be divided into three stages:
the smaller is the better, the nominal is the best, and the larger is
the better [13]. In this study, the larger is better approach was used
for maximizing mechanical properties of the product. In addition to
the S/N ratio, ANOVA was employed to obtain for effect of the process

The effects of several process parameters based on Taguchi's

orthogonal design could be determined effectively from matrix experiments [13]. Minitab 14 software was used for statistical calculations [14]. Tests were organized in using Taguchi's L9(34) orthogonal
array (Table 1). An experimental plan for four parameters with three
levels was organized by the Taguchi method (Table 2). Melt temperature taken from the ABS data sheet was set to 200 C for level 1
and 280 C for level 3, and level 2 was set to 240 C which was an
average value of these two levels. Values of packing pressure, cooling
time and injection pressure for level 1 was obtained from Mold Flow
Inside (MPI 6.1) software [15] and parameters for levels 2 and 3 were
selected as 1.2 and 1.4 times of level 1, respectively [16].

3.2. Material
ABS (Strarex ABS-SD-0150) compound was used for this study.
The properties of ABS compound were shown in Table 3.

3.3. Injection molding machine

A double cavity of the mold was manufactured in the CNC machine
according to ASTM D638 standards. A test part shown in Fig. 1 was
injected by a plastic injection machine (MIR, Turkey) which has a
clamping force of 637 kN and an injection pressure of 1480 bar,
respectively.

Fig. 1. Specimen dimensions for tensile strength (ASTM D638).

B. Ozcelik et al. / International Communications in Heat and Mass Transfer 37 (2010) 13591365

1361

Table 4
Experimental and S/N ratio results of mechanical properties for AISI 1020.
Exp. no

1
2
3
4
5
6
7
8
9

Elasticity module
(MPa)

Tensile strength at yield

(MPa)

Tensile strain at yield

(mm/mm)

Tensile strain at break

(mm/mm)

Flexural modules
(MPa)

Izod impact strength

(notched) (J)

Exp. result

S/N

Exp. result

S/N

Exp. result

S/N

Exp. result

S/N

Exp. result

S/N

Exp. result

S/N

2782
2679
2621
2458
2502
2368
1870
2217
2128

68.89
68.56
68.37
67.81
67.96
67.49
65.44
66.92
66.56

39.09
40.00
39.24
39.64
38.12
38.06
36.75
36.37
36.48

31.84
32.04
31.87
31.96
31.62
31.61
31.31
31.21
31.24

0.02963
0.03132
0.03012
0.03155
0.03171
0.03160
0.03167
0.03037
0.03110

30.56
30.08
30.42
30.02
29.98
30.00
29.99
30.35
30.14

0.03016
0.04258
0.05846
0.09336
0.06136
0.07799
0.31779
0.06472
0.06875

30.41
27.42
24.66
20.60
24.24
22.16
9.96
23.78
23.25

1823
1885
2098
2158
1903
1970
2122
2375
1844

65.22
65.50
66.44
66.68
65.59
65.89
66.53
67.51
65.32

0.564
0.571
0.593
0.641
0.586
0.601
0.720
0.710
0.592

4.97
4.87
4.54
3.86
4.64
4.42
2.85
2.97
4.55

injection parameters on mechanical properties of the product. The

formula of S/N ratio was shown in Eq. (1)

3.4. Experimental testing methods and equipment

A suitable dimension was cut from Fig. 1 to carry out three-point
exural and impact tests. Three-point exural and impact tests were
performed with ISO 178 and ISO 180 standards. The specimens for the
tensile and three-point exural test were shown in Figs. 1 and 2. The
tensile and three-point exural test speeds were found as 10 mm/min
and 2 mm/min (ASTM D638). 10 J of the impact hammer was used in
the impact test and specimen used for the test was shown in Fig. 3.
Tensile strength, three-points exural and impact test were
determined from INSTRON 4411, INSTRON 5560 and Ceast 6545,
respectively. The tests were repeated three times and the mean values
were presented.

"
S = N = 10 log10

1 n 1

n i = 1 y2i

#
1

where S is the standard deviation, yi is the measured experimental

results and n explain to the number of samples in each test trial [17].
Result from measurements of elasticity module, tensile strength
and tensile strain at yield, tensile strain at break, exural modules and
izod impact strength were presented for mechanical properties of the
product (Tables 4 and 5) which determined the optimal levels of four
process parameters for both mold materials [16].

4. Results and discussion

4.2. Analysis of the mechanical results by the thermal conductivity
4.1. Analysis of the S/N ratio
Kovacs and Bercsey examined the inuence of the mold thermal
conductivity on the warpage of the mold. In their study, the typical
mold thermal conductivity using conventional steel tool was between

The test results were evaluated in terms of signal/noise (S/N) ratio.

The S/N was calculated by larger is better for determining effect of

Table 5
Experimental and S/N ratio results of mechanical properties for Al mold.
Exp. No

1
2
3
4
5
6
7
8
9

Elasticity module
(MPa)

Tensile strength at yield

(MPa)

Tensile strain at yield

(mm/mm)

Tensile strain at break

(mm/mm)

Flexural modules
(MPa)

Izod impact strength

(notched) (J)

Exp. result

S/N

Exp. result

S/N

Exp. result

S/N

Exp. result

S/N

Exp. result

S/N

Exp. result

S/N

2746
2595
2548
2620
2772
2564
2411
2831
2621

68.77
68.28
68.12
68.36
68.85
68.19
67.64
69.04
68.37

34.38
41.02
41.03
40.42
40.27
40.08
39.59
38.99
38.58

30.73
32.26
32.26
32.13
32.10
32.21
31.95
31.82
31.73

0.03445
0.03434
0.03380
0.03094
0.03088
0.03203
0.03235
0.03187
0.03132

29.26
29.28
29.42
30.19
30.21
29.89
29.80
29.93
30.08

0.15599
0.17153
0.19370
0.07041
0.07538
0.18574
0.09773
0.09520
0.20044

16.14
15.31
14.26
23.05
22.45
14.62
20.12
20.43
13.96

2252
2276
2273
2141
2154
2160
1984
1926
1935

67.05
67.14
67.13
66.61
66.67
66.69
65.95
65.69
65.73

0.683
0.640
0.698
0.677
0.654
0.651
0.643
0.643
0.659

3.31
3.88
3.12
3.39
3.69
3.73
3.84
3.84
3.62

Table 6
ANOVA results for elasticity module for AISI 1020 mold.
Average S/N

Melt temperature (C) (A)

Packing pressure (MPa) (B)
Cooling time (sn) (C)
Injection pressure (MPa) (D)
Error
Total

67.55
Degree of
freedom (DOF)

Average S/N values

Level 1

Level 2

Level 3

2
2
2
2
1
9

68.60
67.38
67.76
67.80

67.76
67.81
67.64
67.16

66.30
67.47
67.26
67.70

Sum of
square (s)

Variance
(V)

P (%)

8.13
0.31
0.42
0.71
0.00
9.58

4.07
0.16
0.21
0.36
0.00

84.90
3.28
4.37
7.45
0.00
100.00

1362

B. Ozcelik et al. / International Communications in Heat and Mass Transfer 37 (2010) 13591365

Table 7
ANOVA results for tensile strength at yield for AISI 1020 mold.
Average S/N

Melt temperature (C)

Packing pressure (MPa)
Cooling time (sn)
Injection pressure (MPa)
Error
Total

31.64
Degree of
freedom (DOF)

Average S/N values

Level 1

Level 2

Level 3

2
2
2
2
1
9

31.92
31.70
31.56
31.57

31.73
31.63
31.75
31.65

31.25
31.58
31.60
31.68

Sum of
square (s)

Variance
(V)

P (%)

0.71
0.02
0.06
0.02
0.00
0.81

0.35
0.01
0.03
0.01
0.00

86.78
3.06
7.53
2.64
0.00
100.00

Sum of
square (s)

Variance
(V)

P (%)

0.19
0.01
0.08
0.10
0.00
0.38

0.10
0.00
0.04
0.05
0.00

50.05
1.51
22.08
26.35
0.00
100.00

Sum of
square (s)

Variance
(V)

P (%)

110.05
35.68
53.95
56.31
0.00
255.99

55.02
17.84
26.98
28.15
0.00

42.99
13.94
21.08
22.00
0.00
100.00

Sum of
square (s)

Variance
(V)

P (%)

0.82
0.18
0.26
3.44
0.00
4.69

0.41
0.09
0.13
1.72
0.00

17.41
3.76
5.57
73.26
0.00
100.00

Sum of
square (s)

Variance
(V)

P (%)

2.73
0.55
0.28
1.39
0.00
4.96

1.37
0.28
0.14
0.69
0.00

55.13
11.19
5.73
27.95
0.00
100.00

Table 8
ANOVA results for tensile strain at yield for AISI 1020 mold.
Average S/N

Melt temperature (C)

Packing pressure (MPa)
Cooling time (sn)
Injection pressure (MPa)
Error
Total

30.17
Degree of
freedom (DOF)

Average S/N values

Level 1

Level 2

Level 3

2
2
2
2
1
9

30.36
30.19
30.31
30.23

30.00
30.14
30.08
30.03

30.16
30.19
30.13
30.27

Table 9
ANOVA results for tensile strain at break for AISI 1020 mold.
Average S/N

Melt temperature (C)

Packing pressure (MPa)
Cooling time (sn)
Injection pressure (MPa)
Error
Total

22.94
Degree of
freedom (DOF)

Average S/N values

Level 1

Level 2

Level 3

2
2
2
2
1
9

27.50
20.32
25.45
25.97

22.33
25.15
23.76
19.84

19.00
23.36
19.62
23.01

Table 10
ANOVA results for exural module for AISI 1020 mold.
Average S/N

Melt temperature (C)

Packing pressure (MPa)
Cooling time (sn)
Injection pressure (MPa)
Error
Total

66.08
Degree of
freedom (DOF)

Average S/N values

Level 1

Level 2

Level 3

2
2
2
2
1
9

65.72
66.14
66.21
65.37

66.05
66.20
65.84
65.98

66.46
65.88
66.19
66.88

Table 11
ANOVA results for izod impact limit for AISI 1020 mold.
Average S/N

Melt temperature (C)

Packing pressure (MPa)
Cooling time (sn)
Injection pressure (MPa)
Error
Total

4.19
Degree of
freedom (DOF)

Average S/N values

Level 1

Level 2

Level 3

2
2
2
2
1
9

4.79
3.90
4.13
4.72

4.31
4.16
4.43
4.05

3.46
4.50
4.01
3.79

B. Ozcelik et al. / International Communications in Heat and Mass Transfer 37 (2010) 13591365

1363

4.3. ANOVA results

Table 12
The highest S/N values for AISI 1020 mold.
Table No.

Highest S/N values

Table
Table
Table
Table
Table
Table

A1
A1
A2
A3
A3
A3

6
7
8
9
10
11

B2
B1
B2
B1
B2
B1

C1
C2
C2
C3
C1
C3

The inuence of process parameters on the mechanical properties

of ABS material was analyzed by ANOVA which determined how
many parameters to be affected by the mechanical properties. The
percentage contribution of variance was calculated by the following
equations [17]

D1
D3
D2
D2
D3
D3

SST = i m

i=1

25 and 80 W/mK [18]. The selective laser sintered tool insert's thermal
conductivity was less than 15 [W/mK] and the unlled epoxy resins'
thermal conductivity was around 0.5 [W/mK], which cause increased
warpage of the part [18]. Silva et al. examined that the characterization of the performance of the products produced by RIM (Reaction
Injection Molding) was done using molds in several materials
(aluminum, silicone and resin with graphite) and polyurethane
(PUR) [1]. Thermal conductivity values of the mold materials were
0.2 W/mK for lab 850, 0.2 W/mK for prolab 65, 100 W/mK for resin
with graphite and 162 W/mK for alumec 89, respectively. They found
that the tensile strength at yield decreased with the increasing of
thermal conductivity of the mold materials and the exural modules
decreased for high values of thermal conductivity of the mold material
[1]. Nagaoka et al. [8] stated that the molding conditions such as
injection speed, cylinder temperature, and mold temperature conferred on the mechanical properties of the sandwich moldings for PP
material. Tensile strength increased with increasing mold temperature. When the core material cylinder temperature was set to 230 C,
an increase in tensile strength with increasing mold temperature was
observed. However, at 270 C core cylinder temperature, the opposite
tendency had been observed whereby tensile strength decreased with
increasing mold temperature.
Values of elasticity module and tensile strength at yield for Al mold
were higher than that of steel mold in this study when melt
temperature and cooling time were high as shown in Tables 4 and 5.
There was hardly any difference observed for values of tensile strain at
yield and at break. Values of exural modules and izod impact strength
were found to be higher for Al mold when melt temperature and
cooling time were low. Results from this study and literature works
showed that mechanical properties of products from molds were not
changed proportionally as thermal conductivities of mold temperature
or mold material. There was a big difference in thermal conductivities of
mold materials but changes in mechanical properties of injected
products from mold materials were hardly noticeable. Thermal
conductivity of steel mold material used in this study was 4 times
higher than that of aluminum mold material. There was a 1020%
change in mechanical properties when molding materials had different
thermal conductivities. Nagaoka et al. [8] had obtained similar results in
their studies. Silva et al. [1] explained that there was some noticeable
changes occurred in mechanical properties at high values of thermal
conductivities when there was a big difference in values of thermal
conductivities of mold materials.

Total sum of squared deviations SST

i = S/N ratio,
m = average S/N ratio, and
n = number of test.
SST is also expressed in Eq. (3)

SST = SSM SSA + SSB + SSC: + SSE

where SSE is the sum of squared error. Eq. (4) is illustrated for SSM
i
kA h
2
SSM SSB ; SSC ;:::: = nAi *Ai m

i=1

where
kA
nAi
Ai

number of level for factor A

experiment number in i level for factor A
S/N ratio in i level for factor A

4.3.1. ANOVA results for steel and aluminum mold materials

Results from ANOVA were given for steel and aluminum mold
materials in Tables 611 and 1318. In these tables, F value was not
compared with F-test table because calculated error was too small and
F value was innite. Results from the parameters (elasticity module,
tensile strength and tensile strain at yield, tensile strain at break,
exural modules and izod impact strength (notched) limit) for steel
aluminum mold materials were illustrated in Tables 612 and in
Tables 1319, respectively. F0.05;2;9 was 4.26. F ratio was accurately
computed corresponding 95% condence level in calculation of
process parameters. It can be calculated from the ratio of the mean
sum of squared deviations. P value refers to the signicance level. It
was seen from Tables 69 and 11 that the most important parameter
for elasticity module, tensile strength and tensile strain at yield,
tensile strain at break, was melt temperature and its percentage was
obtained for the above parameters by 84.90%, 86.78%, 50.05%, and
42.99%, respectively. The most important parameter affecting exural
module in Table 10 was injection pressure by 73.26% [16]. In case of
aluminum mold material, percentage of injection pressure was 44.21%
for elasticity module, 35.32% for tensile strength at yield and 36.93 for
izod impact limit (Tables 13, 14 and 18), and percentages of melt

Table 13
ANOVA results for elasticity module for Al mold.
Average S/N

Melt temperature (C) (A)

Packing pressure (MPa) (B)
Cooling time (sn) (C)
Injection pressure (MPa) (D)
Error
Total

68.40
Degree of
freedom (DOF)

Average S/N values

Level 1

Level 2

Level 3

2
2
2
2
1
9

68.39
68.26
68.66
68.67

68.47
68.72
68.34
68.04

68.35
68.22
68.21
68.51

Sum of
square (s)

Variance
(V)

P (%)

0.02
0.47
0.33
0.65
0.00
1.46

0.01
0.23
0.16
0.32
0.00

1.40
31.88
22.50
44.21
0.00
100.00

1364

B. Ozcelik et al. / International Communications in Heat and Mass Transfer 37 (2010) 13591365

Table 14
ANOVA results for tensile strength at yield for Al mold.
Average S/N

Melt temperature (C) (A)

Packing pressure (MPa) (B)
Cooling time (sn) (C)
Injection pressure (MPa) (D)
Error
Total

31.89
Degree of
freedom (DOF)

Average S/N values

Level 1

Level 2

Level 3

2
2
2
2
1
9

31.75
31.60
31.53
31.52

32.10
32.06
32.04
32.09

31.83
32.02
32.10
32.07

Sum of
square (s)

Variance
(V)

P (%)

0.20
0.38
0.58
0.63
0.00
1.80

0.10
0.19
0.29
0.32
0.00

11.00
21.15
32.53
35.32
0.00
100.00

Sum of
square (s)

Variance
(V)

P (%)

1.01
0.01
0.04
0.07
0.00
1.13

0.50
0.00
0.02
0.04
0.00

89.39
0.53
3.65
6.43
0.00
100.00

Sum of
square (s)

Variance
(V)

P (%)

35.25
56.77
6.13
10.04
0.00
108.19

17.62
28.39
3.06
5.02
0.00

32.58
52.48
5.66
9.28
0.00
100.00

Sum of
square (s)

Variance
(V)

P (%)

2.68
0.00
0.02
0.03
0.00
2.73

1.34
0.00
0.01
0.01
0.00

98.29
0.08
0.69
0.95
0.00
100.00

Sum of
square (s)

Variance
(V)

P (%)

0.16
0.18
0.01
0.21
0.00
0.57

0.08
0.09
0.01
0.10
0.00

28.74
32.00
2.33
36.93
0.00
100.00

Table 15
ANOVA results for tensile strain at yield for Al mold.
Average S/N

Melt temperature (C) (A)

Packing pressure (MPa) (B)
Cooling time (sn) (C)
Injection pressure (MPa) (D)
Error
Total

29.78
Degree of
freedom (DOF)

Average S/N values

Level 1

Level 2

Level 3

2
2
2
2
1
9

29.32
29.75
29.69
29.85

30.09
29.81
29.85
29.66

29.94
29.80
29.81
29.85

Table 16
ANOVA results for tensile strain at break for Al mold.
Average S/N

17.82
Degree of
freedom (DOF)

Melt temperature (C) (A)

Packing pressure (MPa) (B)
Cooling time (sn) (C)
Injection pressure (MPa) (D)
Error
Total

2
2
2
2
1
9

Average S/N values

Level 1

Level 2

Level 3

15.24
19.79
17.06
17.52

20.04
19.40
17.44
16.71

18.20
14.28
18.97
19.24

Table 17
ANOVA results for exural module for Al mold.
Average S/N

Melt temperature (C) (A)

Packing pressure (MPa) (B)
Cooling time (sn) (C)
Injection pressure (MPa) (D)
Error
Total

66.52
Degree of
freedom (DOF)

Average S/N values

Level 1

Level 2

Level 3

2
2
2
2
1
9

67.11
66.54
66.48
66.48

66.66
66.50
66.50
66.60

65.79
66.52
66.58
66.48

Table 18
ANOVA results for izod impact limit for Al mold.
Average S/N

Melt temperature (C) (A)

Packing pressure (MPa) (B)
Cooling time (sn) (C)
Injection pressure (MPa) (D)
Error
Total

3.60
Degree of
freedom (DOF)

Average S/N values

Level 1

Level 2

Level 3

2
2
2
2
1
9

3.44
3.51
3.63
3.54

3.60
3.80
3.63
3.81

3.77
3.49
3.55
3.45

B. Ozcelik et al. / International Communications in Heat and Mass Transfer 37 (2010) 13591365
Table 19
The highest S/N values for Al mold.

Table 21
Linear model between parameters and mechanical properties.

Tables No.

Highest S/N values

Table
Table
Table
Table
Table
Table

A2
A2
A2
A2
A1
A3

13
14
15
16
17
18

B2
B2
B2
B1
B1
B2

C1
C3
C2
C3
C3
C1

D1
D2
D3
D3
D2
D2

temperature were 89.39% for tensile strain at yield and 98.29% for
exural module (Tables 15 and 17), respectively. In Table 16 the most
important parameter affecting tensile strain at break was packing
pressure by 52.48% [16].
The highest S/N values in Tables 611 showed the most suitable
injection parameters for each parameter were listed in Table 12.
4.4. Regression analysis results for steel and aluminum mold materials
Regression analysis was a statistical tool for the investigation of
relationships between variables. A linear model between injection
parameters and mechanical properties were created by using
MINITAB. The model result was best explained by values of regression
coefcient, r2, close to 1. The results were given in Table 20 for steel
and Table 21 for aluminum mold materials. The elasticity module,
tensile strength at yield, exural module and izod impact strength
for steel and exural module for aluminum mold materials had linear
relationships between the injection parameters whereas other mechanical properties resulted in non linear relationships.
5. Conclusions
In this study, changing of mechanical properties of ABS material
was optimized by ANOVA and regression analysis with respect to
injection parameters and two mold materials. The most important
parameter affecting the elasticity module, tensile strength and tensile
strain at yield, tensile strain at break was melt temperature and its
effect was determined for steel as 84.90%, 86.78%, 50.05% and 42.99%,
respectively. The other parameter affected by exural module
(73.26%) was injection pressure.
In case of aluminum mold material, percentages of injection
pressure were found as 44.21% for elasticity module, 35.32% for tensile
strength at yield and 36.93% for izod impact limit, and percentages of
melt temperature were 89.39% for tensile strain at yield and 98.29%
for exural module, respectively. The most important parameter
affecting tensile strain at break was packing pressure by 52.48%.
The elasticity module, tensile strength at yield, exural module
and izod impact strength for steel and exural module for aluminum
mold materials gave linear relationships (based on values of r2) with
injection parameters whereas other mechanical properties resulted in
non linear relationships.

Table 20
Linear model between parameters and mechanical properties.
Mechanical properties

The equations obtained

from Minitab

R-sq
(%)

R-sq (Adj)
(%)

Elasticity module

70.4 1.15 A + 0.047

B 0.255 C 0.052 D
32.3 0.332 A 0.0650
B + 0.0233 C + 0.0567 D
64.1 + 0.367 A 0.130
B 0.010 C + 0.750 D
5.95 + 0.668 A 0.305
B + 0.055 C + 0.465 D

87.40

74.80

87.40

74.80

91.80

83.60

91.60

83.20

Tensile strength at yield

Flexural module
Izod impact strength

1365

Mechanical properties

The equations obtained

from Minitab

R-sq
(%)

R-sq (Adj)
(%)

Flexural module

3013 3.98 A 0.65

B + 4.06 C 0.05 D

96.60

93.20

Values of elasticity module and tensile stress at yield for Al mold

were higher than that of steel mold when melt temperature and
cooling time were high as shown in Tables 4 and 5. There was hardly
any difference observed for values of tensile strain at yield and at
break. Value of exural modules and izod impact strength were found
to be higher for Al mold when melt temperature and cooling time
were low.
Acknowledgements
The authors would like to thank to Gebze Institute of Technology
for nancial support of the project, 2007-A17. The authors also thank
to Dr. T. Sinmazcelik from Kocaeli University, Galsan Co. and College
of PAGEV in Turkey for the experimental equipments.
References
[1] M. Silva, A. Mateus, P. Bartolo, A.S. Pouzada, A.J. Pontes, The effect of mould
materials in the performance of products moulded by RIM, IV International
Material Symposium, Porto, Portugal, 14 April 2007.
[2] H. Sadabadi, M. Ghasemi, Effects of some injection molding process parameters on
ber orientation tensor of short glass ber polystyrene composites (SGF/PS),
Journal of Reinforced Plastics and Composites 26 (17) (2007) 17291741.
[3] Y.K. Yang, Optimization of injectionmolding process of short glass ber and
polytetrauoroethylene reinforced polycarbonate composites via design of experiments method: a case study, Materials and Manufacturing Processes 21 (8) (2006)
915921.
[4] Y.K. Yang, Optimization of injectionmolding process for mechanical and tribological
properties of short glass ber and polytetrauoroethylene reinforced polycarbonate
composites with grey relational analysis: a case study, Polymer-Plastics Technology
and Engineering 45 (7) (2006) 769777.
[5] G. Guerrica-Echevarria, J.I. Eguiazabal, J. Nazabal, Inuence of the preparation method
on the mechanical properties of a thermotropic liquid crystalline copolyester, Polymer
Testing 20 (4) (2001) 403408.
[6] S. Kenig, A. Ben-David, M. Omer, A. Sadeh, Control of properties in injection
molding by neural networks, Engineering Applications of Articial Intelligence 14
(6) (2001) 819823.
[7] E. Bociaga, The effect of mold temperature and injection velocity on selected
properties of polyethylene moldings, Polimery 45 (1112) (2000) 830836.
[8] T. Nagaoka, U.S. Ishiaku, T. Tomari, H. Hamada, S. Takashima, Effect of molding
parameters on the properties of PP/PP sandwich injection moldings, Polymer
Testing 24 (8) (2005) 10621070.
[9] J.-R. Shie, Optimization of injectionmolding process for mechanical properties of
polypropylene components via a generalized regression neural network, Polymers for Advanced Technologies 19 (1) (2008) 7383.
[10] S.H. Tang, Y.J. Tan, S.M. Sapuan, S. Sulaiman, N. Ismail, R. Samin, The use of Taguchi
method in the design of plastic injection mould for reducing warpage, Journal of
Materials Processing Technology 182 (2007) 418426.
[11] S.-J. Liu, C.H. Lin, Y.C. Wu, Minimizing the sink marks in injectionmolded
thermoplastics, Advances in Polymer Technology 20 (3) (2001) 202215.
[12] M. Savakan, Y. Taptk, M. rgen, Deney tasarm yntemi ile matkap ularnda
performans Optimizasyonu, itdergisi/d, mhendislik, 3 (6) (2004) 117128
(in Turkish).
[13] G. Taguchi, Introduction to Quality Engineering, McGraw-Hill, New York, 1990.
[14] Minitab 14 User Manual.
[15] MoldFlow Plastic Insight, Release 6.1, MPI 6.1, 2001.
[16] A. Ozbay, The investigation of the mechanical properties in the plastic products
produced using different mold and polymer materials, MSc Thesis, Gebze Institute
of Technology, 2008 (in Turkish).
[17] P.J. Ross, Taguchi Techniques for Quality Engineering, second ed.McGraw-Hill,
New York, 1996.
[18] J.G. Kovacs, T. Bercsey, Inuence of mold properties on the quality of injection
molded parts, Periodica Polytechnica Ser. Mech. Eng. 49 (2) (2005) 115122.