International Review of Mechanical Engineering (I.RE.M.E.), Vol. 6, N.

7
ISSN 1970 - 8734 November 2012
Manuscript received and revised October 2012, accepted November 2012 Copyright 2012 Praise Worthy Prize S.r.l. - All rights reserved
1434
Study of Machining Parameters of Wire Electrical Discharge
of Aluminum Matrix Composites (AMCs) with Taguchi Method


M. S. Hussin, M. F. M. A. Hamzas, Hasnul Hadi, A. B. Sanuddin, Z. A. Zailani



Abstract Wire Electro-Discharge Machining (WEDM) is a special form of electrical discharge
machining that uses a small diameter wire as the electrode to cut a narrow keft in the work. The
objective of this study is to determine the optimal machining parameters to achieve the best
performance of Material Removal Rate (MRR) and surface roughness (SR) while cutting the newly
advance material, Aluminum Matrix Composite with 6% fiber (AMCs) by using WEDM. Taguchi
method and the orthogonal array of L-9 (34) has been use in this research to design the
experiment. The parameters chosen are Pulse on Time (TON), Pulse off Time (TOFF), Servo
Voltage (SV), and Wire Speed (WS). These parameters have been use to design 9 different setting
of machining to determine the best combination of parameter. The effect of these machining
parameters to MRR and SR was analyzed with Mean Square Deviation (MSD), Signal-to-Noise
(S/N) ratio and Average Factor Effect (calculation), and Design Expert (DX7) software. Analysis
of Variance (ANOVA) is performed to determine the most influenced or significant parameter. In
this study, the results on the material removal rate and surface roughness have been discussed.
Confirmation test must be made to confirm the accuracy of the value predicted. Copyright 2012
Praise Worthy PrizeS.r.l. - All rights reserved.

Keywords: WEDM, Material Removal Rate, Surface Roughness, Taguchi Method, Optimization


I. Introduction
In WEDM process, which is similar to contour cutting
with a band saw, a slowly moving wire travels along a
prescribed path, cutting the work-piece [1]. Besides,
intricate and complex shapes in various conductive
materials can be cut with WEDM Technology. The most
important performance measures in WEDM include
material removal rate (MRR), surface roughness, and
surface microstructure. To obtain the optimal
performance in WEDM, it is important to set proper
machining parameters.
Aluminum matrix composites (AMCs) refer to a
newly advanced material which is light weight, high
performance, good corrosion resistance and high
damping capacity. Properties of AMCs can be tailored to
the demands of different industrial applications by
suitable combinations of matrix, reinforcement and
processing route.
In the last few years, AMCs has been utilized in high-
tech structural and functional applications including
aerospace, defense, automotive, and thermal management
areas, as well as in sports and recreation [2].
Taguchi method is a powerful experimental design
tool, uses simple, effective, and systematic approach for
deriving of the optimal machining parameters [3].
Further, this approach reduces drastically the number of
experiments which required modeling the response
functions and efficiently reduces the effect of the source
of variation.
Taguchi method is based upon the technique of matrix
experiments and the experimental matrices are special
orthogonal arrays, which allow the simultaneous effect of
several design parameters to be studied efficiently. The
purpose of conducting an orthogonal experiment is to
determine the optimal level for each design parameter
and to establish the relative significance of the individual
parameters in terms of their main effects on the
responses.
II. Literature Review
In WEDM, the wire is continuously advanced between
a supply spool and a take-up spool to present a fresh
electrode of constant diameter to the work and this helps
to maintain a constant keft width during cutting [4]. The
electrical discharges actually cut the workpiece with
submerged in a tank of dielectric fluid, typically de-
ionized water. The dielectric fluid serves some purposes
in WEDM include flushing the cutting debris away from
the cutting zone while flushing is an important factor in
determining the maximum feed rate for a given material
thickness. Dielectric fluid also provides a cooling
medium to minimize the heat affected zone, thereby
preventing potential damage to the work-piece, and acts
as an insulator until the potential is sufficiently high.
There is no direct contact between the workpiece and
the wire electrode during the cutting process. If observe
the WEDM process under a microscope, could be
Copyright 20
discover that
metal to be
remove smal
be moved th
WEDM is a
discharges be
[4]. Wire ED
Fig. 1. Schem
II.1
Pulse on
discharge en
cause of th
increasing th
roughness in
with increasi
of current
parameter ha
as increase t
also increase
the pulse on
were as exp
desirable for
The incre
decrease of
decreases w
voltage param
given to th
associated w
using a suita
chips is nece
the breaking
The most
are material r
keft width.
adjusted to a
finish and cu
process, mat
responses tha
the efficienc
can affect th
M. S.
012 Praise Worthy
t the wire itse
cut and the
ll amounts of
hrough the wo
achieved by
etween the ele
DM cutting pro
matic representati
1. WEDM M
time (TON)
nergy. The ma
he increasing
he value of pu
ncrease is cle
ing discharge
and pulse o
as a direct effe
the pulse on
es [5]. Pulse of
time and its e
pected [6]. T
machining of
ases of pulse
servo voltag
when the pul
meter is the s
he gap and
with the greate
able feeding ra
essary to preve
of the wire [7
II.2. WE
important per
removal rate (
WEDM m
achieve a bett
utting width
terial removal
at need to be
cy and effecti
he machining
Hussin, M. F
y Prize S.r.l. - All
elf actually d
electrical di
f material and
orkpiece. The
thermal ener
ectrode wire a
ocess is illustr


ion of Wire EDM
Machining Par
) is capable
achining spee
of the disc
lse on time (T
arly follows
e energy as a
on time. The
ect on the mat
time the mate
ff time was no
ffects on the m
oo low pulse
f MMCs due t
e off time (TO
ge. The mate
se off time
setting of supp
the higher
er energy in
ate of the wir
ent the second
7].
EDM Respons
rformance me
(MRR), surfac
machining par
ter metal rem
simultaneousl
l rate (MRR)
considered on
iveness of th
output either
F. M. A. Hamz
l rights reserved
does not touch
ischarges actu
allow the wi
e cutting actio
rgy from ele
and the workp
rated in Fig. 1
M cutting process
rameters
to influence
ed becomes f
charge energy
TON). The sur
the trend den
result of incr
e pulse time
erial removal
erial removal
ot as significa
machining pro
e off time is
to wire breaka
OFF) result in
erial removal
increased. S
ply voltage w
servo voltag
the gap. Bes
re to remove b
dary discharge
es
easures in WE
ce finish (SF)
rameters can
moval rate, sur
ly [5]. In WE
) is the impo
n how to mea
he machine. M
acceptable or

zas, Hasnul Ha

1435
h the
ually
re to
on in
ectric
piece
.
[5]
e the
faster
y by
rface
noted
rease
e on
rate,
rate
ant as
ocess
s not
ages.
n the
rate
Servo
which
ge is
sides,
brass
e and
EDM
, and
n be
rface
EDM
ortant
asure
MRR
r not
to th
accu
diam
(deb
Gre
of t
can
resi
be
surf
grea
ener
by r
the
T
thro
cont
in th
exp
of
char
wid
a p
arra
exp
T
dep
T
T
T
l
F
L
L
L
L
f
Q
resu
fact
resp
is di
Q
e
Q
r
Q
d
c
A
proc
and
adi, A. B. San
Internation
he application
uracy of the m
meter of the w
bris in the ga
atest surface
he WEDM m
improve the
stance of the
the most sig
face finish. H
ater surface ro
rgy in the gap
reducing both
surface rough
I
Taguchi is an
ough design, in
trol of critical
he process or
erimental proc
machining
racteristics (c
dth) and to ide
particular pe
ays (OA) a
eriments to stu
The selection
ends on the ite
The number o
The number o
The desired
limitations [8]
Following are
L-4 (23) 4 exp
L-8 (27) 8 exp
L-9 (34) 9 exp
L-12 (211) 1
factors
Quality Chara
ult of the exp
tors in the
ponse and it is
ivided as follo
QC=B: bigge
evaluation wil
QC=S: smalle
results are alw
QC=N: nomin
desired. The
called the targ
II.4. A
ANOVA is
cedure in the
helps to deter
uddin, Z. A. Z
nal Review of Mec
n. Kerf width d
machined par
wire in the gap
ap), the spark
finish is the m
machining pro
e corrosion, f
workpiece. P
gnificant mac
High level of
oughness and
p at increased
h pulses durat
hness can be im
II.3. Taguch
approach that
n this case thr
l variables tha
r product qua
cedure is requ
parameters
utting rate, su
entify the opti
erformance c
are usually
udy the effect
of which O
ems in order o
of factors and
of levels for th
d experimen
]
some commo
periments, 2 le
periments, 2 le
periments, 3 le
2 experiment
acteristic (QC
periment. QC
experiment t
s used to repr
ows:
er is better.
ll be preferred
er is better. S
ways preferred
nal is the best
fixed level o
get or nominal
Analysis of Va
a powerful
social science
rmine the imp
Zailani
chanical Enginee
determines the
rt and it depe
p, the material
k gap and w
most desired
cess. Fine qu
fatigue streng
Pulse on time
chining param
f pulse on tim
d this attribute
pulse on time
tion and disch
mproved.
hi Method
t attempts to e
rough the iden
at cause devia
ality. A speci
uired to evalua
on the
urface roughn
imal paramete
characteristic.
employed i
t of several co
OA to use p
of priority as f
interactions o
he factors of in
ntal resolutio
on arrays:
evels and 3 de
evels and 7 de
evels and 4 de
ts, 2 levels an
C) represents
C will base o
to achieve t
resent the obj
The larger m
d over smaller
Smaller magn
d over the othe
. The fixed va
of achieveme
l value.
ariance (AVON
and commo
es and it is us
portant variatio
ering, Vol. 6, N. 7
e dimensional
ending on the
l removal rate
wire vibration.
characteristic
uality surfaces
gth, and wear
was found to
meter on the
me results in
ed to the high
e [6]. Besides,
harge current,
ensure quality
ntification and
tions to occur
ally designed
ate the effects
performance
ness, and kerf
er settings for
Orthogonal
in industrial
ontrol factors.
predominantly
follows:
of interest
nterest
on or cost
esign factors
esign factors
esign factors
nd 11 design
the objective
on the design
the optimum
ectives and it
magnitude of
r ones.
nitudes of the
ers.
alue is always
ent desired is
NA)
on statistical
sed to analyze
on of factors.
7
l
e
e
.
c
s
r
o
e
n
h
,
,
y
d
r
d
s
e
f
r
l
l

y
t
n
e
n
m
t
f
e
s
s
l
e

M. S. Hussin, M. F. M. A. Hamzas, Hasnul Hadi, A. B. Sanuddin, Z. A. Zailani
Copyright 2012 Praise Worthy Prize S.r.l. - All rights reserved International Review of Mechanical Engineering, Vol. 6, N. 7
1436
In statistics, ANOVA is a collection of statistical
models, and their associated procedures, in which the
observed variance is partitioned into components due to
different explanatory variables. ANOVA decomposes the
variability in the response variable amongst the different
factors. Depending upon the type of analysis, it may be
important to determine:
Which factors have a significant effect on the
response, and/or
How much of the variability in the response variable
is attributable to each factor.
III. Methodology
In general, there are five general steps that had been
set to make sure the utilization of DOE tools can be held
efficiently. The five general steps are as below:
Plan the experiment
Describe the experiment
Conducting the experiment
Analyze the data from the experiment
Confirmation of experiment
Planning the experiment is the stage that clarifies the
objective and design factors; determine the number of
factor and level, and the description of the experiment.
Using Taguchis parameter design, significant
machining parameters affecting the performance
measures are identified as pulse on time (T
ON
), pulse off
time (T
OFF
), servo voltage (SV), and wire speed (WS).
Input machining parameters such as pulse on time (T
ON
),
pulse off time (T
OFF
), servo voltage (SV), and wire speed
(WS) used in this research are shown in Table I. Each
factor is investigated at three levels to determine the
optimum settings for the WEDM process.
These machining parameters and their levels were
chosen based on the literature review. Taguchi method is
based on orthogonal arrays and the type of array used in
this experimental research is L-9 (34).

TABLE I
DESIGN FACTORS WITH PARAMETERS USE IN EXPERIMENT
LEVEL OF DESIGN FACTOR
Design Factors Unit
Level
1 2 3
Pulse On Time
(T
ON
)
s 9 10 11
Pulse Off Time
(T
OFF
)
s 20 30 40
Servo Voltage
(SV)
V 40 50 60
Wire Speed (WS) mm/s 160 170 180

The overall experiment will arrange by the Taguchi
method (orthogonal array) by referring the Appendix C
after all values of the design factor are determined. The
layout underlines how the experiment will carry on and
all the selected machining parameters that involve in the
experiment. There are 9 experiments that arrange in rows
and 4 design factors in arranging in columns.
The table was indicated the balance of 4 design factors
that had a chance to affect the responses.
There are four important steps in conducting the
experiments, which are prepared of AMCs with 6 Vol %
fiber workpiece, measure the materials volume after the
machining, cutting the material into a desired dimension
based on parameters that had been set with WEDM
machine and the surface roughness analysis[9]. AMCs is
prepared by using Powder Metallurgy (PM) technology.
The conventional PM sequence consists of three steps
include blending and mixing of powders, compaction, in
which the powders are pressed into the desired part
shape, and sintering, which involves heating to a
temperature below the melting point to cause solid-state
bonding of the particles and strengthening of the part.
The material volume has been measured after the
machining is to calculate the material removal rate
(MRR) for the material during the WEDM process. The
following formula is indicated how to calculate MRR:

HRR =
Iolumc o motcriol rcmo:cJ (cm)
Hocbining timc (min)
(1)

In this research, the workpiece was machined by using
a Sodick AQ327L WEDM machine with different
parameters setting to obtain the optimal machining
parameters of best machining performance. Another
experiment variable that was measured is surface
roughness (Ra). The surface roughness analysis is done
by using equipment Mitutoyo CS-3100.
The relationship between control factors and
responses like MRR and surface roughness (Ra) were
established by Mean Square Deviation (MSD), Signal-to-
Noise (S/N) ratio, Average Factor Effect, and Analysis of
Variance (ANOVA), resulting in a valid mathematical
model in this research. Besides, the best combination of
parameters to obtain the best performance in MRR and
surface roughness (Ra) is determined by Design Expert
software. The following equations are used in result
calculation:

HS =
l
l
l
l[
1
HRR
u

2
+ [
1
HRR
b

2
+
+[
1
HRR
c

2
1
1
1
1
n
(C = B)
(2)

HS =
_
(SR
u
)
2
+ (SR
b
)
2
+
+(SR
c
)
2
_
n
(C = S)
(3)

SN = -1u log
10
(HS) (4)

The confirmation experiment is the final step in the
design of experiment process and it purposely to validate
the conclusions drawn during the analysis phase.
The confirmation experiment is performed by
conducting a test with a combination of the factors and
levels which previously evaluated. The optimal
machining parameters was confirmed and compared with
the actual value after the experiment had conducted.
Copyright 20
The perce
%. The expe
the error perc
of the experi
the improvem
IV
Material R
(Ra) result w
Mean Squar
ratio, Averag
(ANOVA).
(bigger is b
roughness, M
will be chose
The arithm
will direct
machine. MS
used to anal
ANOVA is
parameter.
Design Ex
analysis. Gr
were created
of factors ov
predicted and
best MRR an
IV.1. Sta
The Mean
and Signal-to
determine th
Each factor
All the value
the table and
Besides, g
value will be
in the Table I
QC=B
Fac
T
O
T
O
SV
W
Factors
T
ON
(s)
T
OFF
(s
SV (V)
WS (mm/

M. S.
012 Praise Worthy
entage of error
eriment needs
centage excee
iment. The fin
ment of the pe
V. Result
Removal Rate
will use the sam
re Deviation
ge Factor Eff
For the MR
better) will be
Mean Square
en.
metic mean v
take from
SD, S/N ratio,
lyze the pred
used to de
xpert (DX7) so
raph and cha
d based on the
ver responses
d actual optim
nd Ra will be c
atistical Analys
n Square Dev
o-Noise (S/N)
e average effe
is divided int
e for the avera
d the peak valu
grand average
e calculated. A
II.
TA
MRR AVER
Average
tor Level 1
ON
-9.8123
OFF
-5.9368
V -3.3666
WS -8.5820
Gran
Optimum
Max Va
TA
THE BEST COMB
s Level
) Level
s) Level
Level
/s) Level
Hussin, M. F
y Prize S.r.l. - All
rs should not
s to be redone
eds 10 % as th
nal step is to
erformance ch
t and Discu
e (MRR) and
me method of
(MSD), Sign
fect, and Ana
RR, Mean S
e chosen wh
Deviation (
value (Ra) of
the surface
, and Average
dicted optimu
etermine the
oftware will b
art regarding
e result analy
will be discu
mization param
compared and
sis of Materia
(MRR)
viation, MSD
) ratio will be
ect of factors
to 3 levels fo
age effect of
ue of the facto
e, optimum S/
All the value c

ABLE II
RAGE FACTORS EF
e Factors Effect
Level 2
-6.7678
-8.8623
-8.2818 -
-8.1892
nd Average
S/N Value (dB)
alue (cm
3
/min)

ABLE III
BINATION PARAM
l Para
2
1
1
3
F. M. A. Hamz
l rights reserved
be higher tha
e all over aga
his leads to inv
predict and v
haracteristic.
ussion
surface rough
f analysis whi
nal-to-Noise (
alysis of Vari
Square Devia
hereas for sur
smaller is be
surface rough
roughness t
e Factor Effec
um parameter
most influe
be used to adv
factors influ
ysis. The beha
ussed. At last
meter to mee
d discussed.
al Removal Ra
(bigger is be
e calculated. T
at different le
or the calcula
factors is fille
or will be selec
/N value and
calculated wil
FFECT
Level3 P
-8.0106 -6.
-9.7915 -5.
-12.9422 -3.
-7.8194 -7.
-8.
0.7
1.
METER
ameter Setting
10
20
40
180

zas, Hasnul Ha

1437
an 10
ain if
valid
verify
hness
ch is
(S/N)
iance
ation
rface
etter)
hness
tester
ct are
and
nced
vance
uence
avior
t, the
et the
ate
etter)
Then,
evels.
ation.
ed in
cted.
max
ll fill
Peak
7678
9368
3666
8194
1969
7001
.084
A
are
F
T
O
T
O
SV
W
E
T
IV
A
show
In
wor
mag
to g
adi, A. B. San
Internation
Analysis of M
shown as belo
Fig. 2. M
Fig. 3.
Fig. 4. T
actor f
ON
(A) 2
OFF
(B) 2
V (C) 2
WS (D) 2
Error 0
Total 8
V.2. Statistic
Analysis of R
wn as in Figs.
n this study
rkpieces usin
gnification of
give a uniform
uddin, Z. A. Z
nal Review of Mec
MRR by Desig
ow (Figs. 2, 3,

Model graph for p


. Model graph of


The optimumvalu

TABLE
ANOVA TABLE
S
173.6097
197.1085
645.6124
55.8123
0
cal Analysis of
Ra by Design
. 6, 7, 8 and in
y, photograp
ng the SEM
500, 2000, an
m comparison b
Zailani
chanical Enginee
gn Expert (D
, 4 and Table
pulse off time (T
O
servo voltage (SV
ue of predicted M
E IV
E FOR MRR
V F
86.8049 0
98.5543 0
322.8062 0
27.9061 0
0 0
f Surface Roug
n Expert DX7
n Table VII.
phs were ta
operated a
nd 5000 times
between speci
ering, Vol. 6, N. 7
DX7) software
IV).

OFF
)

V)
MRR
F PC %
0 16.19
0 18.38
0 60.22
0 5.21
0
100
ghness (Ra)
7 Software is
aken of the
at 20kV and
s were chosen
imens.
7
e

s
e
d
n
Copyright 20
The two
worst profile

SURFAC
QC=B
Fact
T
ON
T
OF
SV
WS
Fact
T
ON
(
T
OFF
(
SV (
WS (m
Fig
F
M. S.
012 Praise Worthy
profiles whic
e are shown as
TA
CE ROUGHNESS (R
Average
tor Level 1
N
-11.0455
FF
-12.7041
V -11.5972
S -12.1421
Grand
OptimumS
Max V
TA
THE BEST COMB
ors Leve
(s) Level
(s) Level
(V) Level
mm/s) Level
Fig. 5. Surface r
g. 6. Model graph
Fig. 7. Model gra
Hussin, M. F
y Prize S.r.l. - All
ch consist of
s below.
ABLE V
RA) AVERAGE FAC
e Factors Effect
Level 2
-12.5446
-11.8006
-11.9260
-12.6773
d Average
S/N Value (dB)
Value (m)

ABLE VI
BINATION PARAM
el Param
l 1
l 2
l 1
l 3


roughness test lay


h for pulse on tim


aph for wire speed
F. M. A. Hamz
l rights reserved
the best and
CTORS EFFECT
Level 3 P
-13.2739 -11.
-12.3593 -11.
-13.3409 -11.
-12.0446 -12.
-12.
-9.6
3.0
METER
meter Setting
9
30
40
180


yout

me (TON)

d (WS)

zas, Hasnul Ha

1438
d the
eak
.0455
.8006
.5972
.0446
.2880
6239
0283
F
T
O
T
O
S
W
E
T
F
was
void
9 de
F
surf
sma
frac
Ma
Ma
A
the
are
2
(WS
0.99
8.07
and
ANO
adi, A. B. San
Internation
Fig. 8.
ANOVA T
Factor f
ON
(A) 2
OFF
(B) 2
V (C) 2
WS (D) 2
Error 0
Total 8
From Fig. 9, c
s obtained by
d formation, d
esigned machi
From observat
face was obta
all craters, v
cture occurred
agnification of
500X
Fi
agnification of
500X
Fig.
After the calcu
confirmation
Pulse On Tim
20s, Servo V
S) 180mm
971cm3/min.
71%. The per
software a
OVA table fo
uddin, Z. A. Z
nal Review of Mec

The optimumval

TABLE
TABLE FOR SURF
S
136.2896
37.4766
111.0162
31.6988
0

can be seen th
cutting 9 as t
dimple and cl
ining.
tion, Fig. 10 s
ained by cutt
void formatio
on the cutting

f Magnifica
2000

g. 9. The best pro

f Magnifica
2000

. 10. The worst pr

ulation, the pe
test. The be
me (T
ON
) 10
Voltage (SV)
m/s. The MRR
The percenta
rcentage of e
nalysis resul
or MRR, can b
Zailani
chanical Enginee
lue of predicted R
VII
FACE ROUGHNESS
V
68.1448
18.7383
55.5081
15.8494
0
hat a fine mac
the surface ha
leavage fractu
show that a po
ing 8 as ther
on, dimple, a
g surface

ation of
0X
M
ofile, Cutting 9

ation of
0X
Ma
rofile, Cutting 8
eak values ha
st combinatio
0s, Pulse Of
40V, and
R of confirm
age of error
error for conf
lt is 8.652%
be seen that th
ering, Vol. 6, N. 7
Ra)
S (RA)
F PC %
0 43.06
0 11.84
0 35.08
0 10.02
0
100
chined surface
ad the least of
ure among the
oor machined
re are lots of
and cleavage
Magnification
of 5000X
agnification of
5000X
ad used to run
on parameters
ff Time (T
OFF
)
d Wire Speed
matory test is
calculated is
firmation test
%. From the
he percentage
7

e
f
e
d
f
e
f
n
s
)
d
s
s
t
e
e

M. S. Hussin, M. F. M. A. Hamzas, Hasnul Hadi, A. B. Sanuddin, Z. A. Zailani
Copyright 2012 Praise Worthy Prize S.r.l. - All rights reserved International Review of Mechanical Engineering, Vol. 6, N. 7
1439
contribution of parameters are Pulse on Time (T
ON
)
16.19%, Pulse off Time (T
OFF
) 18.38%, Servo Voltage
(SV) 60.22%, and Wire Speed (WS) 5.21%.
Though the ANOVA analysis result can be concluded
that servo voltage is the most influenced or significant
factor to obtain the best performance of MRR.
After the calculation, the peak values had used to run
the confirmation test. The best combination parameters
are Pulse On Time (T
ON
) 9s, Pulse Off Time (T
OFF
)
30s, Servo Voltage (SV) 40V, and Wire Speed
(WS) 180mm/s. The Ra of confirmatory test is
2.919m.
The percentage of error calculated is 3.609%.
The percentage of error for confirmation test and
software analysis result is 3.669%.
From the ANOVA table for SR, can be seen that the
percentage contribution of parameters is Pulse on Time
(T
ON
) 43.06%, Pulse off Time (T
OFF
) 11.84%,
Servo Voltage (SV) 35.08%, and Wire Speed (WS)
10.02%.
Though the ANOVA analysis result can be concluded
that pulse on time is the most influenced or significant
factor to obtain the best performance of surface
roughness.
V. Conclusion and Recommendation
This study has presented an investigation at the
optimization and the effect of machining parameters on
material removal rate (MRR) and surface roughness (SR)
in WEDM operations. The level of importance of the
machining parameters MRR and SR is determined by
using ANOVA. Based on ANOVA table, the highly
significant parameter on both the MRR and SR was found
as servo voltage (SV) and pulse on time (T
ON
)
respectively.
The optimum combination of parameter for the
maximum MRR and minimum SR was obtained by using
the analysis of mean square deviation (MSD), and signal-
to-noise (S/N) ratio. The confirmation tests indicated that
it is possible to increase MRR and decrease SR (Ra)
significantly by using the proposed statistical technique.
The experimental results confirmed the validity of the
used Taguchi method for enhancing the machining
performance and optimizing the machining parameters in
WEDM operations.
The optimum performance of MRR will be obtained
by the setting of Pulse on Time (T
ON
) - 10s, Pulse off
Time (T
OFF
) - 20s, Servo Voltage (SV) - 40V, and Wire
Speed (WS) - 180mm/s. The result of the confirmation
test is 0.9971cm3/min and the MRR value obtained from
the calculation is 1.084cm3/min. The percentage of error
is 8.017%. The optimum performance of surface
roughness will be obtained by the setting of Pulse on
Time (T
ON
) - 9s, Pulse off Time (T
OFF
) - 30s, Servo
Voltage (SV) - 40V, and Wire Speed (WS) - 180mm/s.
The result of conformation test is 2.919m and the
result of calculation is 3.0283m. The percentage of
error is 3.609%. There are some recommendations for
further improvement as below:
There are a lot of WEDM parameters can affect the
MRR and surface roughness such as Peak Current,
Voltage, and Water Pressure. These parameters can
be used for future research.
Using different type of analysis method such as 2-
Level Factorial Design to design experiments and
compare the results with Taguchi Method.
Using different parameters to find the maximization
of MRR and minimization of surface roughness.
Using different type of Design of Experiment such
as Response Surface Methodology (RSM) to conduct
the experiment.
The Design of Experiment layout based on the
Taguchi method in this study can be used to research
other similar newly advanced materials by using
another type of machining such as Plasma Arc
Cutting.
Acknowledgements
This work was supported by Fundamental Research
Grant (FRGS) 900300319.
References
[1] Serope Kalpakjian and Steven Schmid (2006). Manufacturing
Engineering and Technology, Fifth Edition, Pearson Education
South Asia Pte Ltd.
[2] M K SURAPPA (2003). Aluminum matrix composites:
Challenges and opportunities, Sadhana, Vol. 28, Parts 1 & 2, pp.
319334.
[3] S. S., MahapatraandAmar Patnaik (2006). Optimization of wire
electrical discharge machining (WEDM) processes parameters
using Taguchi method, Int J Adv Manuf Technol.
[4] Mikell, P. Groovier (2007). Fundamentals of Modern
Manufacturing: Materials, Processes, and Systems. Third Edition,
J ohn Wiley & Sons (Asia) Pte Ltd.
[5] H. Singh and R. Garg (2009). Effects of process parameters on
material removal rate in WEDM, J ournal of Achievements in
Materials and Manufacturing Engineering, Volume 32 Issue 1.
[6] Nilesh, G. Patil and Brahmankar, P. K. (2010). Some studies into
wire electro-discharge machining of alumina particulate-
reinforced aluminum matrix composites, Int J Adv Manuf
Technol, 48:537555.
[7] C.T. Yang, S.L. Song, B.H. Yan and F.Y. Huang (2006).
Improving machining performance of wire electrochemical
discharge machining by adding SiC abrasive to electrolyte,
International J ournal of Machine Tools & Manufacture , 46 2044
2050.
[8] Phillip J . Ross (1996). Taguchi techniques for quality
engineering, 2nd edition. McGraw-Hill, New.
[9] Hasnul Hadi, Liyana Tajul, M. F. M. A. Hamzas, M.S. Hussin, Z.
A. Zailani, M. H. M Radzi. Parameter Optimization For J IS S45C
Steel Turning Process Based on Taguchi Method, International
Review of Mechanical Engineering (IREME), Vol. 5 N. 2, pp.
315-320.
[10] Z. A. Zailani, M. S Hussin, M. F. M. A Hamzas, M. J Hasnul, A.
B. Sanuddin; The Correlation between Surface Finish in Milling
Process Involving Solid Lubricant, International Review of
Mechanical Engineering (IREME) 6 (6) , pp. 1262-1267.
[11] Hadi, H., Tajul, L., Hussin, M.S., Zailani, Z.A.,Hamzas,
M.F.M.A., Bahari, M.S.,2012;Study on ability of palm oil as
viscosity index improvement (VII) additive in thermal stability
(base mineral oil - Cooking palmoil); International Review of
Mechanical Engineering (IREME), 6 (3) , pp. 456-461.
Copyright 20


Integration Syst

Analysis and Fu


M. S.
012 Praise Worthy
Authors
Mohd S
Manufact
Malaysia
and Engi
Malaysia
modeling
engineeri

Mohd F
his BSc d
Universit
in Mech
from Un
His mai
problems

Hasnul H
his BEn
Islamic
Masters
Engineer
Malaysia
Manufact
Energy
tem.
Asmawi
March 1
Mechanic
Malaysia
on He co
Mechanic
Malaysia
is related
unctionally Grade
Zailani Z
degree i
Universit
Masters
Universit
The resea
and manu

Hussin, M. F
y Prize S.r.l. - All
informati
abri Hussin rec
turing Engineer
a Perlis and Maste
ineering Design
a, Malaysia .The
g and control, pr
ing simulation for
Fidzwan Md Am
degree in Mechan
ti Malaysia Perli
hanical Enginee
niversiti Tun Hu
in research are
s and engineering
Hadi Mohamma
ng in Mechatron
University Mal
degree in M
ing from Unive
a . His main
turing System,
Conversion
B. Sanuddin w
984. Holding a
cal Engineering
a (UPM) and gra
ntinued his study
cal Engineering
a (UPM) in 2008
d to Mechanica
ed Materials.
Zainal Abidin
in Manufacturin
ti Malaysia Per
degree in Manuf
ti Putra Malaysia
arch interest is in
ufacturing system
F. M. A. Hamz
l rights reserved
on
ceived his B. E
ring from Univ
er degree in Innov
formUniversiti
research interest
roduct innovation
r automotive safe
min Hamzas rec
nical Engineering
s and Masters d
ering (Manufactu
ussein Onn, Mal
ea is in sched
g simulation
ad Jaafar rec
nics fromInterna
laysia, Malaysia
Manufacturing Sy
ersiti Putra Mal
research area
Renewable En
and Mecha
was born on 18
Bachelors degr
fromUniversiti
aduated in 2007.
y (master of scien
at Universiti
8. His research in
al Engineering,
received his B.
ng Engineering
rlis, , Malaysia
facturing System
a, Selangor, Mal
machining techn
m.

zas, Hasnul Ha

1440
Eng in
versiti
vation
Putra
t is in
n and
ety.
ceived
g from
degree
uring)
laysia.
duling
ceived
ational
a and
ystem
laysia,
is in
nergy,
atronic
8th of
ree in
Putra
Later
nce) in
Putra
nterest
Stress
. Eng
from
a and
mfrom
laysia.
nology
adi, A. B. San
Internation
uddin, Z. A. Z
nal Review of Mec
Zailani
chanical Enginee ering, Vol. 6, N. 77
Copyright of International Review of Mechanical Engineering is the property of Praise Worthy Prize S.r.L. and
its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's
express written permission. However, users may print, download, or email articles for individual use.