International Journal of Engineering Research & Technology (IJERT)

ISSN: 2278-0181
Vol. 4 Issue 09, September-2015

Parametric Optimization of Wire Electrical

Discharge Machining by Taguchi Technique on
Composite Material
Kishore. G. C 1* Aruna Devi. M 2 C. P. S Prakash 3
1 2 3
Post Graduate Student, Assistant Professor, Principal,
Dept. of Mechanical Engineering, Dept. of Mechanical Engineering, Dayananda Sagar College of
Dayananda Sagar College of Dayananda Sagar College of engineering, Bangalore-560078,
engineering, Bangalore-560078, engineering, Bangalore-560078, Karnataka, INDIA.
Karnataka, INDIA. Karnataka, INDIA.

Abstract— In this present study, Al7075+10%Al2O3 metal matrix EDM material removal process were wire is used as cutting
composite (MMC) were fabricated using stir casting method. The tool as shown in Fig-1. The discrete sparking between work
five parameter namely voltage, pulse-on, pulse-off, current, bed piece and wire, erodes the material from the work piece in the
speed were chosen as factors to study the output responses in presence of dielectric fluid which is flushed continuously to
terms of material removal rate (MRR) and surface roughness
working zone and also flush the eroded particles and acts as
(Ra) while machining Al7075+10%Al2O3 metal matrix composite
(MMC) in wire Electrical Discharge Machining (WEDM). coolant. Taguchi orthogonal array is used to conduct the
Experimentation has been carried out using Taguchi’s L18 experiment with less number of experiment and get better
orthogonal array. Evaluation of output responses has been done result. The experimental result is transformed to S/N ratio and
by Signal to Noise (S/N) ratio analysis and to determine the ANOVA are used to determine optimum value and relative
significant effect of each parameter Analysis of Variance contribution of each factor on output responses.
(ANOVA) was carried out. Optimal value of parameters which
maximize material removal rate (MRR) and minimize surface
roughness (Ra) were determined based on experimental result, In
addition mathematical model have developed for output
responses.

Keywords—WEDM, MRR, Ra, Taguchi orthogonal array,

minitab-17 software.
I. INTRODUCTION
A composite is a material which is comprised of two or more
materials. The main part of the composite is called as matrix
material and the material mixed is called as reinforcement.
The composite are mainly classified in to three groups
Polymer matrix composites (PMCs), Metal matrix composites
(MMCs), Ceramic matrix composites (CMCs). To produce
Polymer matrix composites (PMCs), Metal matrix composites
(MMCs), Ceramic matrix composites (CMCs) reinforcement
are added. The advantage of using composite are light weight,
high strength, corrosion resistance, high impact strength,
design flexibility, dimension stability and so on. Aluminium
matrix composite (AMC) exhibit outstanding combination
properties such as high strength, high stiffness, thermal and
electric conductivities from these magnificent property Fig- 1 Wire electric discharge machining (WEDM) model
AMC’s are used in aerospace, automobile, defense and in
many other sectors. These composite can be fabricated by II. EXPERIMENTAL STUDY
using stir casting method. In which reinforcement material can
be mixed with molten metal by mechanical stirring process. A. Material Selection
This method can reduce the final cost and produce large 1) Matrix Material
quantity. Wire electrical discharge machining (WEDM) is a Aluminium Al7075 alloy was considered as the matrix
thermo-electrical process which is classified under non- material based on mechanical properties like machinability,
traditional machining process. WEDM is effectively and fatigue strength, corrosion resistance are excellent compared
successfully used to machine MMC. WEDM is similar to to other Aluminium alloy.

IJERTV4IS090489 www.ijert.org 407

(This work is licensed under a Creative Commons Attribution 4.0 International License.)
 International Journal of Engineering Research & Technology (IJERT)
ISSN: 2278-0181
Vol. 4 Issue 09, September-2015

Table - I Chemical composition of Al7075 selection of process parameter were based on machine
capability. The parameters like Voltage, pulse-on, pulse-off,
Contents Zn Mg Cu Cr Fe Si Mn Ti Al current, bed speed were chosen to carry out the experiments

Composition (%) 6 3 2 0.3 0.6 0.5 0.4 0.3 Bal Table- III Machining parameter used in experiments

2) Reinforcement
Al2O3 (Alumina) of size 50 - 100 microns was used as Level
reinforcing particles with the proportions of 90% and 10%. In parameters
I II III Units
engineering ceramic family Alumina is widely used and it is
cost effective material which have excellent combination of A Voltage 75 100 ---- Volts
properties. With the fine grain alumina has wide range of
application. B Pulse-ON 40 30 20 μ sec
Table - II Chemical composition of AL2O3
C Pulse-OFF 9 12 15 μ sec
Contents SiO2 Fe2O3 TiO2 Na2O AL2O3
D Current 2 4 6 Amps
Composition (%) 0.15 0.05 0.15 0.45 Bal

B. Composite Preparation E Bed speed 50 150 250 μm/sec

The present study was conducted by taking Aluminium
Al7075 as a base matrix and Al2O3 of size 50 - 100 microns Material removal rate (MRR) can be calculated using
was used as reinforcing particles with the proportions of 90% Eq. (1).
and 10% wt. respectively. Aluminium Al7075 alloys were
melted using 6kw melting furnace (silicon element heating) at MRR= (2*Wg+D) * t * L Eq. (1)
a temperature of 7400c for 30 min. Then the mixture was T
stirred using a ceramic coated metallic stirrer rotating at 300 Where,
rpm for 15 minutes and then it was poured in to a metallic Wg = spark gap - “0.02mm”
mould which was preheated. D = Diameter of wire - “0.18mm”
T = Time taken to cut “min”
1) Metallographic Study L = Distance travelled by tool – “60mm”
t = Thickness of work piece – “10mm”

Surface roughness was measured using Mitutoyo Surftest SJ-

210 portable surface measuring unit.

D. Experimental Design
Dr.Genichi Taguchi developed Taguchi method which was
built on traditional concepts of Design of Experiment (DOE).
R.A. Fisher introduced the DOE technique to study the
multiple variables simultaneously. Orthogonal array (OA) is a
specially constructed table based on DOE technique to reduce
the number of experiments. The L18 (2*3) orthogonal array
was chosen to conduct the experiment as shown in Table- IV
Fig- 2 Microstructure
Table- IIV L18 (2*3) orthogonal array
To investigate the distribution of discontinuous reinforcement
matrix in the fabricated specimen, the quality of specimen was L18 (2*3) Orthogonal array
checked by metallographic study using optical microscope
Bed
connected to computer imaging system and scanning electron Exp Voltage Pulse ON Pulse OFF Current
speed
microscope. Microstructure of composite specimen was
No. volts (µs) (µs) Amps (µm/s)
observed at 200X. From the Fig-2 uniformly distributed
reinforcement was revealed. 1 75 40 9 2 50

C. Wire Electrical Discharge Machining (WEDM) 2 75 40 12 4 150

The experiments was conducted by using DK-7732 WEDM
3 75 40 15 6 250
machine tool which is manufactured by CONCORD United
Products Ltd. Based on the thickness, material the operator 4 75 30 9 2 150
can select the input parameter from the manual provided by
the manufacturing company. Molybdenum wire of Diameter 5 75 30 12 4 250
0.18mm was used and demineralized water plus JR3A gel is 6 75 30 15 6 50
used as a dielectric fluid to carry out experiments. The

IJERTV4IS090489 www.ijert.org 408

(This work is licensed under a Creative Commons Attribution 4.0 International License.)
 International Journal of Engineering Research & Technology (IJERT)
ISSN: 2278-0181
Vol. 4 Issue 09, September-2015

7 75 20 9 4 50 12 17.483 24.853 1.595 -4.055

13 28.884 29.213 1.759 -4.905
8 75 20 12 6 150
14 5.880 15.387 1.617 -4.174
9 75 20 15 2 250
15 11.681 21.350 1.768 -4.950
10 100 40 9 6 250 16 17.862 25.039 1.458 -3.275
17 14.270 23.089 1.729 -4.756
11 100 40 12 2 50
18 5.877 15.383 1.501 -3.528
12 100 40 15 4 150

13 100 30 9 4 250 A. Signal to Noise (S/N) ratio

The S/N ratio graph for material removal rate (MRR) is shown
14 100 30 12 6 50 in fig 3. from the graph the optimum value obtained are
15 100 30 15 2 150 Voltage 100 volts, Pulse-ON 40(µs), Pulse-OFF 9(µs),
Current 6 Amps, Bed speed 250 (µm/s). This optimum value
16 100 20 9 6 150 gives maximum MRR. The S/N ratio graph for surface
roughness (Ra) is shown in fig 4. from the graph the optimum
17 100 20 12 2 250
value obtained are Voltage 100 volts, Pulse-ON 40 (µs),
18 100 20 15 4 50 Pulse-OFF 15 (µs), Current 6 Amps, Bed speed 50 (µm/s).
This optimum value gives minimize Ra.
After conducting the experiments results were evaluated by
using, S/N ratio and ANOVA to find the optimal value and
relative parameter influence on output responses i.e. (MRR,
Ra).

The S/N ration is classified as “Larger the better”, “Nominal

the better” and “Smaller the better”. The S/N ratio for MRR
and Ra was calculated by logarithmic transformation function
as shown in Eq. (2). “Larger the better” and Eq. (3) “Smaller
the better” respectively S/N ratio and tabulate in Table- V
MRR = -10 log (∑ (1/y2)/n) Eq. (2)
RA = -10 log (∑y2/n) Eq. (3)

III. RESULT AND DISCUSSION

The analysis of experimental results were carried out by using
minitab-17. The results were transformed to signal to noise
ratio of MRR and Ra for Al7075+10%Al2O3 metal matrix Fig. 3. Signal to noise ratio graph for material removal rate (MRR)
composite (MMC).

Table- V Eexperimental results

L18 (2*3) Orthogonal array

S/N ratio S/N ratio

Exp. No. MRR Ra
MRR Ra
mm3/min (dB) (µm) (dB)
1 5.641 15.027 1.784 -5.028
2 16.500 24.350 1.528 -3.682
3 17.984 25.098 1.572 -3.929
4 13.895 22.857 1.737 -4.796
5 17.671 24.945 1.739 -4.806
6 5.906 15.426 1.387 -2.842
7 5.901 15.418 1.645 -4.323
8 17.886 25.050 1.567 -3.901
Fig. 4. Signal to noise ratio graph for surface roughness (Ra)
9 8.800 18.890 1.749 -4.856
10 30.137 29.582 1.405 -2.954
11 5.911 15.434 1.576 -3.951

IJERTV4IS090489 www.ijert.org 409

(This work is licensed under a Creative Commons Attribution 4.0 International License.)
 International Journal of Engineering Research & Technology (IJERT)
ISSN: 2278-0181
Vol. 4 Issue 09, September-2015

The Response Table for S/N ratio for Material Removal Rate
(MRR) and surface roughness (Ra) is shown in Table-VI and
Table-VII

Table-VI Taguchi Analysis: MRR v/s Voltage, Pulse On, Pulse Off, Current,
Bed Speed
Response Table For Signal To Noise Ratios Larger Is Better
Bed
Level Voltage Pulse-On Pulse-Off Current
Speed
1 20.78 20.48 22.86 19.44 15.35

2 22.15 21.53 21.38 22.36 23.92

3 ------- 22.39 20.17 22.60 25.14

Delta 1.36 1.91 2.69 3.16 9.79

Rank 5 4 3 2 1 Fig. 5 Percentage contributions by process parameters on MRR

Table-VII Taguchi Analysis: Ra v/s Voltage, Pulse-No, Pulse-Off, Current, Table-IX Regression Analysis: Ra v/s Voltage, Pulse-No, Pulse-Off, Current,
Bed Speed Bed Speed
Response Table For Signal To Noise Ratios Smaller Is Better
Analysis of variance
Bed
Level Voltage Pulse-On Pulse-Off Current
Speed Source DF Adj SS Adj MS F-Value P-Value
1 -4.240 -4.107 -4.214 -4.723 -3.974
Regression 5 0.177183 0.035437 4.05 0.022
2 -4.061 -4.412 -4.212 -4.217 -4.110
Voltage 1 0.005000 0.005000 0.57 0.464
3 ------ -3.933 -4.026 -3.512 -4.368
Pulse-On 1 0.002977 0.002977 0.34 0.570
Delta 0.180 0.479 0.187 1.210 0.393
Pulse-Off 1 0.003888 0.003888 0.44 0.517
Rank 5 2 4 1 3
Current 1 0.148964 0.148964 17.04 0.001
B. Analysis Of Variance (ANOVA)
Bed Speed 1 0.016354 0.016354 1.87 0.196
The ANOVA results for Material Removal Rate (MRR) and
surface roughness (Ra) are shown in Table VIII and IX and Error 12 0.104894 0.008741 ----- -------
Fig 5 and 6 shows the Percentage contributions of Material
Removal Rate (MRR) and surface roughness (Ra) Total 17 0.28076 -------- ----- -------

Table-VIII Regression Analysis: MRR v/s Voltage, Pulse On, Pulse Off,
Current, Bed Speed

Analysis of variance

Source DF Adj SS Adj MS F-Value P-Value

Regression 5 860.68 172.14 16.01 0.000

Voltage 1 42.95 42.95 3.99 0.069

Pulse-On 1 44.31 44.31 4.12 0.065

Pulse-Off 1 99.69 99.69 9.27 0.010

Current 1 104.76 104.76 9.74 0.009

Bed Speed 1 568.97 568.97 52.92 0.000

Error 12 129.01 10.75 ----- -------

Total 17 989.70 -------- ----- ------- Fig 6 Percentage contributions by process parameters on Ra

IJERTV4IS090489 www.ijert.org 410

(This work is licensed under a Creative Commons Attribution 4.0 International License.)
 International Journal of Engineering Research & Technology (IJERT)
ISSN: 2278-0181
Vol. 4 Issue 09, September-2015

C. Mathematical Model Table –XI Model summary for Ra

Multiple linear regression (MLR) model is performed by the Model summary

help of MINITAB-17 software. This model is used to predict S R-sq R-sq (adj) R-sq
various performance measures in WEDM process. The (pred)
Equation (4) and (5) shows the mathematical model, Table X 0.0934940 62.81% 47.32% 14.95%
and XI shows the model summary. Experimental value and
predicted value are compared as shown in figure 7 and figure Coefficients
8 of Material Removal Rate (MRR) and surface roughness Term Coef SE Coef T-Value P-Value VIF
(Ra) respectively.
Constant 2.021 0.217 9.32 0.000 -----
Table - X Model summary for MRR
Model summary Voltage -0.00133 0.00176 -0.76 0.464 1.00

S R-sq R-sq R-sq Pulse-On -0.00157 0.00270 -0.58 0.570 1.00

(adj) (pred)
3.27887 86.96% 81.53% 71.74% Pulse-Off -0.00600 0.00900 -0.67 0.517 1.00

Coefficients Current -0.0557 0.0135 -4.13 0.001 1.00

Term Coef SE Coef T-Value P-Value VIF Bed Speed 0.000369 0.000270 1.37 0.196 1.00

Constant -7.50 7.61 -0.99 0.344 -----

Regression Equation
Voltage 0.1236 0.0618 2.00 0.069 1.00
Ra =2.021 - 0.00133 Voltage - 0.00157 Pulse-On –
Pulse-On 0.1922 0.0947 2.03 0.065 1.00 0.00600Pulse-Off - 0.0557 Current +0.000369 Bed Speed.
Eq. (5)
Pulse-Off -0.961 0.316 -3.05 0.010 1.00

Current 1.477 0.473 3.12 0.009 1.00

Bed Speed 0.06886 0.00947 7.27 0.000 1.00

Regression Equation
MRR = -7.50+ 0.1236 Voltage + 0.1922 Pulse-On -
0.961 Pulse-Off + 1.477 Current
+ 0.06886 Bed Speed. Eq. (4)

Fig. 8 comparison of Ra between experimental and predicted

Table –XII Optimum Value

Factor
Response Voltage Pulse- Pulse- Current Bed speed
(volts) on off Amps (µm/s)
Fig. 7 comparison of MRR between experimental and predicted (µs) (µs)
MRR 100 40 9 6 250
(mm3/min)
Ra 100 40 15 6 50
(µm)

IJERTV4IS090489 www.ijert.org 411

(This work is licensed under a Creative Commons Attribution 4.0 International License.)
 International Journal of Engineering Research & Technology (IJERT)
ISSN: 2278-0181
Vol. 4 Issue 09, September-2015

IV. CONCLUSION REFERENCES

Al7075+10%Al2O3 metal matrix composite (MMC) was [1]. S.S.Mahapatra.Amar Patnaik “Optimization of wire electrical
discharge machining (WEDM) process parameters using
fabricated successfully using stir casting process and Taguchi
Taguchi method” International Journal of Advanced
L18 orthogonal array was used to conduct the experiment. manufacturing technology, 2006.
From the S/N ratio and ANOVA analysis the following [2]. Sateesh Kumar Reddy K, Ramesh S, “Parametric Optimization
conclusion were drawn of Wire Electrical Discharge Machining of Composite Material”
International Journal of Advanced Research in Computer
1. Using taguchi method MRRR and Ra were optimized Engineering & Technology Volume 1, Issue 3, May2012
individually. [3]. C.D.Shah, J.R.Mevada, B.C.Khatri, “Optimization of Process
2. Bed speed is the most influential parameter which Parameter of Wire Electrical Discharge Machine by Response
significantly affect the material removal rate (MRR). The Surface Methodology on Inconel-600” International Journal of
Emerging Technology and Advanced Engineering (ISSN 2250-
voltage, pulse-on, pulse-off, current are less influential
2459, ISO 9001:2008 Certified Journal, Volume 3, Issue 4,
parameters. April 2013)
3. According to proposed levels of factors used in this work [4]. Jadi Laxman and Kotakonda Guru Raj “ Optimization of
to maximize MRR can be achieved by selecting Electric Discharge Machining Process Parameters Using
combination of parameters, Voltage 100 (volts), Pulse-on Taguchi Technique” International Journal of Advanced
40 (µs), Pulse-off 9(µs), Current 6 Amps, Bed speed 250 Mechanical Engineering ISSN 2250-3234 Volume 4, Number 7
(µm/s). (2014).
4. Current is the most influential parameter which [5]. Amit Nara and Sahil Takhi “Optimization of Process Parameters
significantly affect the surface roughness (Ra). The in Finishing of Weld Bead Using WEDM” International Journal
of Innovative Research in Science, Engineering and Technology
voltage, pulse-on, pulse-off, Bed speed are less influential
(An ISO 3297: 2007 Certified Organization) Vol. 3, Issue 8,
parameters. August 2014.
5. For achieving minimum surface roughness the optimum [6]. Pujari Srinivasa Rao, Koona Ramji and Beela Satyanarayana “
condition are Voltage 100 (volts), Pulse-on 40 (µs), Experimental Investigation of wire EDM parameter for surface
Pulse-off 9(µs), Current 6 Amps, Bed speed 250 (µm/s). roughness MRR and white layer in machining of aluminium
alloy” International Conference On Advance In Manufacturing
And Materials Engineering AMM 2014.
[7]. Miss. Dipti Ghanashyam Gonjari “Investigation of Effect of
WEDM Process Parameters on Performance Characteristics of
Tool Steel Grade AISI D7” IRACST – Engineering Science and
Technology: An International Journal (ESTIJ), ISSN: 2250-
3498 Vol.4, No.5, October 2014.
[8]. Muthu Kumar V, Suresh Babu A, Venkatasamy R and
Raajenthiren M “Optimization of the WEDM Parameters on
Machining Incoloy800 Super alloy with Multiple Quality
Characteristics” International Journal of Engineering Science
and Technology Vol. 2(6), 2010, 1538-1547.
[9]. Pragya Shandilya, P.K.Jain, N.K.Jain “Parametric Optimization
During Wire Electrical Discharge Machining Using Response
Surface Methodology”
[10]. J.Udaya Prakash, T.V. Moorthy, J.Milton Peter “Experimental
Investigations on Machinability Of Aluminium Alloy (A413)/
Flyash /B4C Hybrid Composites Using Wire EDM”
international conference on design and manufacturing, icon DM
2013.

IJERTV4IS090489 www.ijert.org 412

(This work is licensed under a Creative Commons Attribution 4.0 International License.)