Int J Adv Manuf Technol (2007) 33: 1122–1127

DOI 10.1007/s00170-006-0550-2

ORIGINA L ARTI CLE

P. Vijian . V. P. Arunachalam

Optimization of squeeze casting process parameters

using Taguchi analysis

Received: 7 March 2005 / Accepted: 10 January 2006 / Published online: 19 April 2006
# Springer-Verlag London Limited 2006

Abstract Squeeze casting is a hybrid metal forming studies concentrated on aluminum and magnesium alloys for
process combining features of both casting and forging in light weight components to enhance fuel efficiency. Squeeze
one operation. This paper reports a research in which an casting accounts for 15 to 40% improvement of the
attempt was made to prepare solid cylindrical components mechanical properties than gravity die casting [7].
of LM24 aluminum alloy through squeeze casting. The Lynch et al. have carried out investigations with squeeze
primary objective was to investigate the effect of process casting of aluminum and presented various aspects of the
parameters on the mechanical properties exhibited by the process [8]. Yue studied the process parameter of squeeze
castings produced though squeeze casting process. A set of casting and reported that too high pouring temperature and
trials were conducted based on parameters settings sug- pressures were undesirable as these will lead to poor tensile
gested in Taguchi’s offline quality control concept. The properties [9]. Iwata et al. performed research to investigate
experimental results indicate that the squeeze pressure and the relationship between shrinkage and pressure transfer time
the die-preheating temperature were the parameters making in the squeeze casting of Al-Si alloys [10]. Basu et al.
significant contribution toward improvement in mechani- reported the mechanical properties of squeeze cast compo-
cal properties of squeeze cast LM24 aluminum alloy. nents revealing that the strength and ductility combination
observed is better than that obtained in the case of permanent
Keywords Hardness . Squeeze casting . die casting [11]. Wu et. al investigated the optimal set of
Taguchi method . Tensile strength parameters using Taguchi method for pressure die casting of
Mg alloys components [12]. It is reported that squeeze cast
LM25 component exhibits desirable structural properties and
1 Introduction is superior to those produced in permanent moulds [13].
Though several research works applying Taguchi meth-
The automotive and aerospace industries have been driving ods on die cast components have been reported in
research communities to search for materials with improved literature, it appears that very limited works have been
mechanical properties and new casting processes [1]. carried out for squeeze cast components. On considering
Conventional casting process cannot produce parts as strong the importance of aluminum alloys, the main objective of
as forged parts. The major drawback of casting processes is the research was to apply Taguchi method to find the
the formation of casting defects such as porosity, segrega- optimal set of control parameters for squeeze casting of
tion, hot tears etc. [2–4]. A distinctly different approach to LM24 aluminum alloy (Al 88%, Si 8%, Cu 3%, Fe 1%).
component making is possible with squeeze casting, an
emerging metal forming process. The major advantage of
this process is its ability to eliminate microporosity to 2 Squeeze casting – process outline
provide isotropic properties and minimize component
machining [5, 6]. This process has been applied on a wide Squeeze casting process, is based on the pressurized
range of metals, ranging from low melting alloys of lead and solidification of the molten metal in re-usable dies, and
zinc to high melting alloys of iron and nickel. Most of the involves the following steps [13]:
P. Vijian (*) . V. P. Arunachalam 1. Preheating of the die and the punch.
Department of Mechanical Engineering, 2. Pouring molten metal into the die cavity.
Government College of Technology,
641 013 Coimbatore, India 3. Application of squeeze pressure and allowing for
e-mail: [email protected] solidification.
Tel.: +91-0422-2432221 4. Ejection of solidified casting.
 1123
Table 1 Control factors and levels Table 3 S/N ratio for tensile strength and hardness
Factor Control factor Level 1 Level 2 Level 3 Exp. A B E C S/N Ratio for tensile S/N Ratio for
notation no strength(dB) hardness(dB)

A Squeeze pressure (MPa ) 35 70 106 1 1 1 1 1 47.4928 38.7887

B Die preheating temperature 34 150 300 2 1 2 2 2 47.7112 38.8880
(°C) 3 1 3 3 3 46.9260 38.1518
C Duration of pressure 15 30 45 4 2 1 2 3 48.1290 39.2742
application (s) 5 2 2 3 1 48.5285 39.6440
6 2 3 1 2 48.0271 39.1745
7 3 1 3 2 48.6244 39.7299
The parameters that affect squeeze casting have been 8 3 2 1 3 48.2275 39.3636
identified as composition of the casting alloy, level of 9 3 3 2 1 47.9220 39.0849
applied pressure, die preheating temperature, pouring
temperature, die coat material (lubricant), melt superheat, Exp. no–Experiment number
A–Squeeze pressure, MPa
duration of pressure application, punch temperature, delay B–Die pre-heat temperature, °C
time to achieve maximum pressure [3]. The combination of C–Duration of pressure application, seconds
high pressures and metal die leads to a high heat transfer E–Error
coefficient that in turn leads to alterations in microstruc-
ture. Such a high heat transfer rate results in a high level of
nucleation and rate of growth, which leads to finer grain concept has been applied for carrying out robust design of
size, close dendrite arm spacing and smaller constituent processes and products and solving several complex
particles. As the solidification under load reduces shrink- problems in manufacturing industry [15]. Further, this
age and porosity, the microstructure becomes isotropic in technique determines the most influential parameter on the
nature. In contrast to other casting methods, faster output response for the significant improvement in the
solidification yields a very fine microstructure resulting overall performance.
in improved mechanical properties, significantly exceeding During trials, as similar to the experimentation reported
that of conventional castings. Further, excellent material [7], the pouring temperature was approximately maintained
saving is achieved through this method, as no running and at 750°C. Graphite was used as the lubricant. In order to
feeding systems are required [5]. observe the influencing degree of process parameters in
squeeze casting, three parameters namely squeeze pressure,
die preheating temperature and duration of pressure
3 Taguchi method application each at three levels were considered and are
listed in Table 1. Maintaining these processing parameters
Taguchi method is an efficient problem-solving tool, which as constants enabled us to study the effect of squeeze
can upgrade/improve the performance of the product, pressure, die preheat temperature, and duration of pressure
process, design, and system with a significant slash in application. The total degrees of freedom for three
experimental time and cost [14]. This method combining parameters in each of three levels were 6. A three level
the experimental design theory and quality loss function L9 34 orthogonal array with nine experimental runs was
selected (degrees of freedom = 9-1 = 8).
Table 2 Experimental observations
Exp. no A B C Tensile strength MPa Hardness BHN
R1 R2 R1 R2 Table 4 Pareto ANOVA for three level factors

1 35 34 15 234 240 86 88 Factors A B C E Total

2 35 150 30 245 241 89 87
Sum at factor levels 1 ΣA1 ΣB1 ΣC1 ΣE1 T
3 35 300 45 224 220 78 84
2 ΣA2 ΣB2 ΣC2 ΣE2
4 70 34 45 258 252 91 93
3 ΣA3 ΣB3 ΣC3 ΣE3
5 70 150 15 270 264 95 97
Sum of squares of differences SA SB SC SE ST
6 70 300 30 254 250 93 89
Degrees of freedom 2 2 2 2 8
7 106 34 30 274 266 99 95
(Contribution ratio)/100 SA/ST SB/ST SC/ST SE /ST 1
8 106 150 45 263 253 91 95 P P P
9 106 300 15 246 252 90 90 T ¼ PA1 þ PA2 þ AP 3 P P P
SA ¼ðP A1
P A2 Þ þ ðP A1
P A3 Þ2 þ ðP A2
P A3 Þ2
2
Exp. no – Experiment number SB ¼ðP B1
P B2 Þ þ ðP B1
P B3 Þ þ ðP B2
P B3 Þ2
2 2
A – Squeeze pressure, MPa
B – Die pre-heat temperature, °C SC ¼ðP C1
P C2 Þ2 þ ðP C1
P C3 Þ2 þ ðP C2
P C3 Þ2
C – Duration of pressure application, sec SE ¼ð E1
E2 Þ2 þ ð E1
E3 Þ2 þ ð E2
E3 Þ2
R1 and R2, – Replication 1 and 2 of trail casting ST ¼SA þ SB þ SC þ SE
1124
Table 5 Pareto ANOVA for tensile strength
Factors A B C E Total

Sum at factor levels 1 142.1300 144.2462 143.9433 143.7474 431.5885

2 144.6846 144.4672 144.3627 143.7622
3 144.7739 142.8751 143.2825 144.0789

Sum of squares of differences 13.5242 4.4636 1.7794 0.2104 19.9776

Degrees of freedom 2 2 2 2 8
Contribution ratio 67.70 22.34 8.91 1.05 100
Cumulative contribution ratio 67.70 90.04 98.95 100
Optimum level A3 B2 C2

Table 6 Pareto ANOVA for hardness

Factors A B C E Total

Sum at factor levels 1 115.8285 117.7928 117.5176 117.3268 352.0996

2 118.0927 117.8956 117.7924 117.2471
3 118.1784 116.4112 116.7896 117.5257

Sum of squares of differences 10.6559 4.1228 1.6111 0.1236 16.5134

Degrees of freedom 2 2 2 2 8
Contribution ratio 64.53 24.97 9.76 0.74 100
Cumulative contribution ratio 64.53 89.5 99.26 100
Optimum level A3 B2 C2

4 Experimental procedure resulting in minimization of quality characteristic variation

due to uncontrollable parameter [16]. The tensile strength
A 40 tonne Universal testing machine was modified to and hardness were considered the quality characteristics
apply pressure during solidification of the aluminum alloy. with the concept of the “larger the better”. The S/N ratio
The LM24 aluminum alloy was melted in an electrical used for this type of response is given by:
resistance furnace and the die was preheated using a !
ceramic electric heater. The experimental set-up is shown 1X n
1
in Fig. 1. Two trial castings were made as per the data sheet S=N ðdBÞ ¼
10 log10 ;
n i¼1 Yi2
of L9 (34) orthogonal array. Tensile strength and hardness
specimens were machined from these castings and the
obtained values are tabulated in Table 2.

5 Results and discussion 80

Percentage Contribution of Tensile Strength

Taguchi method stresses the importance of studying the 70

67.7

response variation using the signal-to-noise (S/N) ratio,

60

50

40

30
22.34

20

8.91
10
1.05

0
A B C D
Control Factors

Fig. 1 Experimental set-up Fig. 2 Pareto diagram of tensile strength

 1125
70
Table 8 ANOVA for hardness
Percentage Contribution of Hardness

6 4 .5 3

60 Sources of Sum of squares Degrees of Mean Variance

variation of differences freedom square
50

A 10.6559 2 5.3280 86.21

40
B 4.1228 2 2.0644 33.40
30 2 4 .9 7
C 1.6111 2 0.8056 13.03
E 0.1236 2 0.0618
20 Total 8
9 .7 6
10
0 .7 4

0
temperature (B). Hence in order to study the parameter
A B C D
significance, ANOVA were performed for tensile strength
Control Factors
and hardness. The details are shown in Tables 7 and 8,
Fig. 3 Pareto diagram of hardness respectively.
The response curves of tensile strength and hardness are
shown in Figs. 4 and 5, which depict the pictorial view of
where Yi is the response value for a trial condition repeated variation of each factor and describe what the effect on the
n times. system performance could be, and when a parameter shifts
The squeeze cast process parameters, namely squeeze from one level to another.
pressure (A), die preheating temperature (B) and duration As this work focused on multi response variables in tensile
of pressure application (C) were assigned to the 1st, 2nd, strength and hardness, it became essential to construct the
and 4th columns of L9 34 array, respectively. The 3rd overall summary Table 9 to arrive at optimum condition.
column was assigned as error (E). The S/N ratios were Based on the highest values of the S/N ratio levels for the
computed for tensile strength and hardness in each of the significant factors A and B, the overall optimum condition
nine trial conditions and their values are given in Table 3. thus obtained were A3 and B2. (Squeeze pressure of 106 MPa
Computation scheme of Pareto ANOVA for three level and die preheating temperature of 150°C).
factors is shown in Table 4. In order to study the
contribution ratio of the process parameters, Pareto
ANOVA was performed for tensile strength and hardness. 6 Estimation of predicted mean
The details are given in Tables 5 and 6, respectively.
In both cases, it was found that squeeze pressure, At the optimum setting condition, predicted mean S/N ratio μ
contributes a larger impact on mechanical properties of of tensile strength and hardness were estimated as follows:
squeeze cast LM24 aluminum alloys. This is depicted in
Figs. 2 and 3. S=N Tensile strength μ¼A3 þB2
T ¼ 48:2580 þ 48:1557
47:9843
Further, Pareto ANOVA was used to determine the
¼ 48:46dB:
optimum level of process parameters. These levels are a
squeeze pressure of 106 MPa (A3), die preheating temper-
ature of 150°C (B2), and pressure application duration of 30 s S=N Hardness; μ¼A3 þB2
T ¼ 39:3928 þ 39:2985
39:1221
(C2). These levels were found to improve tensile strength and ¼ 39:57dB:
hardness. It must be noted that the above combination of
factor levels A3, B2, C2 are not among the nine combinations
tested for the experimentation. This was expected because of
the multifactor nature of the experimental design employed. 48.4
(9 from 33=27 possible combinations) [16]. 48.3
From the contribution ratio of Pareto ANOVA, Tables 5 A3
48.2 A2
and 6, the significant factors which affect the tensile strength B2
C2
Average of S/N ratio

and hardness are squeeze pressure (A) and die-preheating 48.1

B1
48.0 C1
47.9
Table 7 ANOVA for tensile strength
47.8
Sources of Sum of squares Degrees of Mean Variance C3
47.7
variation of differences freedom square B3
47.6
A 13.5242 2 6.7621 64.28 47.5
B 4.4636 2 2.2318 21.21 47.4 A1
C 1.7794 2 0.8897 8.46 47.3
E 0.2104 2 0.1052 Control factors
Total 8
Fig. 4 Response curve of tensile strength
1126
39.5 7 Confidence interval
39.4 A3
A2
39.3 B2 C2
The confidence interval of the predicted estimations was
B1
computed as enumerated below:
Average f S/N ratio

39.2
C1 rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
39.1 1
Confidence Interval ¼  Fðv1 ;v2 Þ  Ve 
39.0 neff
38.9
C3
38.8 B3
38.7
A1
38.6
Control factors

Fig. 5 Response curve of hardness

where, Ā3 and B̄2 are the average values of S/N ratio of the
factor levels A and B obtained from Tables 5 and 6,
respectively.

Pooled variation of non
significant sources

Ve ¼ :
Pooled degrees of freedom of non
significant sources

Number of experiments
neff ¼ :
1 þ ðTotal degrees of freedom associated with the items used
estimateÞ

At 99.95% significance, the confidence interval, optimum settings of the squeeze cast parameters were
S/N Tensile strength was found to be 48.46 ± 0.62 dB = found to be 266, 273 MPa and 96, 99 BHN respectively
47.84 to 49.08 dB. and their S/N ratios were 48.60 dB and 39.77 dB,
At 99.95% confidence interval, the confidence interval, respectively. These values were found to be within the
S/N Hardness was found to be 39.57 ± 0.58 dB = 38.39 to confidence interval limits.
40.15 dB.

9 Conclusion
8 Confirmation test
Squeeze cast technology has the potential to play an
Two confirmation tests were conducted at the optimum important role in the near future for improving the quality
settings as determined through the experimentation. The of the engineering components. In this context, this paper
values of tensile strength and hardness obtained at the has reported a research in which Taguchi’s off-line quality
control method was applied to determine the optimal
Table 9 Optimum condition summary process parameters which maximize the mechanical
properties of squeeze cast LM24 aluminum alloy. For
F Level S/N ratio Tensile S/N ratio Optimum Result
this purpose, concepts like orthogonal array, S/N ratio and
strength (dB) hardness (dB)
ANOVA were employed. After determining the optimum
A 1 142.1300 115.8285 A3 process parameters, two confirmation experiments were
2 144.6846 118.0927 conducted. In light of our analysis, the following conclu-
3 144.7739 118.1784 * sions were drawn:
B 1 144.2462 117.7928 B2 The optimum level of process parameters to obtain good
2 144.4672 117.8956 * mechanical properties of squeeze cast components of
3 142.8751 116.4112
LM24 aluminum alloy are a squeeze pressure of 106 MPa
and die preheating temperature of 150°C.
F – Factors
 1127

From the Pareto analysis, it was evident that the squeeze 5. Brown JR, Barlow J et al (1994) Second report of Institute
pressure is a major contributing factor for improving tensile working group T20 casting process. Foundryman 87:386–390
6. Morton JR, Barlow J (1994) Squeeze casting from a theory to
strength and hardness. profit and a future. Foundryman 87:23–28
The constant pouring temperature and same lubricant 7. Yong MS, Clegg AJ (1999) Evaluation of squeeze cast
(graphite) were applied throughout the conduct of experi- magnesium alloy and composite. Foundryman, pp 71–75
ments. As both these parameters are important and 8. Lynch RF, Olley R et al (1975) Squeeze casting of aluminum.
AFS Trans 122:569–576
significant enough to determine the quality of castings, 9. Yue TM (1997) Squeeze casting of high strength aluminum
future studies accompanied by the conduct of a greater wrought alloy AA7010. J Mater Process Technol 66:179–185
number of trial testings are required in this direction to 10. Iwata Y, Sugiyama Y, Iwahori H, Awano Y (2000) Transferred
improve the accuracy of the results reported in this paper. pressure and shrinkage cavity of aluminum alloys on squeeze
Fig. 1 castings. J Japan Foundry Eng Soc 72:263–267
11. Basu K, Solonki RS, Muneshwar VS (1989) Squeeze casting of
aluminum and its alloys. Indian Foundry October 25–34
Acknowledgements The authors are thankful to the learned 12. Wu DH, Chang MS (2004) Use of Taguchi method to develop a
referees whose constructive comments and suggestions have been robust design for the magnesium alloy die casting process.
used to refine the previous version of this paper. Mater Sci Eng A 379:366–371
13. Padmaprabha Seshan S (2005) Structure and properties of
squeeze cast LM25 alloy. Foundry May/June 31–34
14. Avani Gandhi MS (2003) Problem solving using Taguchi DOE
References techniques. Ind Eng J XXXII:16–25
15. Taso CC, Hocheng H (2004) Taguchi analysis of delamination
1. Vijian P, Arunachalam VP (2005) Experimental study of associated with various drill bits in drilling of composite
squeeze casting of gunmetal. J Mater Process Technol 170: material. Int J Mach Tools Manuf 44:1085–1090
32–36 16. Anastasiov KS (2002) Optimization of the aluminium die
2. Yue TM, Chadwick GA (1996) Squeeze casting of the light casting, process based on the Taguchi method. Proc Ins Mech
alloys and their composites. J Mater Process Technol 58: Eng 216:969–977
302–307 17. Sung H Park (1996) Robust design and analysis quality
3. Ghomashchi MR, Vikhrov A (2000) Squeeze casting:an engineering. Chapman and Hall, London
overview. J Mater Process Technol 101:1–9 18. Ross PJ (1996) Taguchi techniques for quality engineering.
4. Kleiner K et al (2002) Microstructure and mechanical proper- McGraw-Hill, Singapore
ties of squeeze cast and semi-solid cast Mg-Al alloys. J Lht Met
2:277–280