Document 12914459

advertisement
International Journal of Engineering Trends and Technology (IJETT) – Volume 29 Number 4 - November 2015
Evaluation of Material Removal Rate and Surface Roughness
of Ti6al4v in Wire EDM Process using Response Surface
Methodology
M.Raja Roy#1, N.Ramanaiah2, B.S.K Sundara Siva Rao 3
1#
Sr.Assistant Professor, Department of Mechanical Engineering, ANITS, Visakhapatnam, India
2
Professor, Department of Mechanical Engineering, AUCOE (A) Visakhapatnam, India
3
Former Professor, Department of Mechanical Engineering, AUCOE (A) Visakhapatnam, India
ABSTRACT
The main objective of this work is to
evaluate the Material Removal Rate (MRR) and
surface roughness (Ra) of Ti6Al4V in Wire cut
Electrical Discharge machining process using
Response surface Methodology. In the present work,
Full Factorial Design is considered with three
process parameters: TON, TOFF and IP each to be
varied in three different levels. Data related to
material removal rate (MRR) and surface roughness
(Ra) have been measured for each experimental run.
The variation of output responses with process
parameters were mathematically modelled by using
Response Surface Methodology. Response is
predicted and percentage of error has been
calculated. Results showed that the established
mathematical models can predict the output
responses with reasonable accuracy.
pressure upper and lower flushing nozzles clear out
microscopic debris from the surrounding area of the
wire during the cutting process. The fluid also acts
as a non-conductive barrier, preventing the
formation of electrically conductive channels in the
machining area. When the wire gets close to the part,
the intensity of the electric field overcomes the
barrier and dielectric breakdown occurs, allowing
current to flow between the wire and the work piece,
resulting in an electrical spark.
Keywords — Material Removal Rate, Surface
Roughness, Design of Experiments, Response
Surface Methodology
I. INTRODUCTION
1.1 Wire Electrical Discharge Machining
Wire electrical discharge machining (EDM) is a
non-traditional machining process that uses
electricity to cut any conductive material precisely
and accurately with a thin, electrically charged
copper or brass wire as an electrode. During the wire
EDM process, the wire carries one side of an
electrical charge and the work piece carries the other
side of the charge [1]. When the wire gets close to
the part, the attraction of electrical charges creates a
controlled spark,
melting and vaporizing
microscopic particles of material. The spark also
removes a miniscule chunk of the wire, so after the
wire travels through the work piece one time, the
machine discards the used wire and automatically
advances new wire. The process takes place quickly
and produces thousands of sparks per second—but
the wire never touches the work piece [2]. The spark
erosion occurs along the entire length of the wire
adjacent to the work piece, so the result is a part with
an excellent surface finish and no burrs regardless of
how large or small the cut. Wire EDM machines use
a dielectric solution of de ionized water to
continuously cool and flush the machining area
while EDM is taking place. In many cases the entire
part is submerged in the dielectric fluid, while high-
ISSN: 2231-5381
Fig: 1.1
1.1.1 Major Components
A Wire EDM system is comprised of four
major components.
(1) Computerized Numerical Control: CNC
program is fed to the machine.
(2) Power Supply: It provides energy to the spark.
(3) Mechanical Section: Worktable, work stand,
taper unit, and wire drive mechanism.
(4) Dielectric System: The water reservoir where
filtration,
condition
of
the
water
(resistivity/conductivity) and temperature of
the water is provided and maintained.
1.2 Titanium alloys
Titanium alloys are metals that contain a
mixture of titanium and other chemical elements.
Such
alloys
have
very
high tensile
strength and toughness. They are light in weight,
have extraordinary corrosion resistance and ability to
withstand extreme temperatures. However, the high
cost of raw materials and cost of processing, limits
their use to aircraft, spacecraft and biological
implant[7]. Although "commercially pure" titanium
has acceptable mechanical properties and has been
used for orthopaedic and dental implants, for most
http://www.ijettjournal.org
Page 209
International Journal of Engineering Trends and Technology (IJETT) – Volume 29 Number 4 - November 2015
applications titanium is alloyed with small amounts
of aluminium and vanadium, typically 6% and 4%
respectively, by weight.
1.2.1 Transition Temperature
The crystal structure of titanium at ambient
temperature and pressure is close-packed hexagonal
α phase with a c/a ratio of 1.587. At about 890°C,
the titanium undergoes an allotropic transformation
to a body-centred cubic β phase which remains
stable to the melting temperature.
Some alloying elements raise the alpha-tobeta transition temperature (i.e., alpha stabilizers)
while others lower the transition temperature
(i.e., beta
stabilizers).
Aluminium,gallium, germanium, carbon, oxygen an
d nitrogen are alpha stabilisers. Molybdenum,
vanadium,
tantalum, niobium, manganese, iron, chromium, cob
alt, nickel, copper and silicon are beta stabilizers.
1.2.2 Ti6Al4V or Ti-Grade5
Ti Grade 5, also known as Ti6Al4V or Ti 6-4, is the
most commonly used alloy. It has a chemical
composition of 6% aluminium, 4% vanadium,
0.25% (maximum)iron, 0.2% (maximum) oxygen,
and the remainder titanium. It is significantly
stronger than commercially pure titanium while
having the same stiffness and thermal properties
(excluding thermal conductivity, which is about 60%
lower in Grade 5 Ti than in CP Ti). Among its many
advantages, it is heat treatable. This grade is an
excellent combination of strength, corrosion
resistance and weld ability.
II. LITERATURE REVIEW
Vikram Singh et al [1] devised an
approach to determine machining parameter settings
for WEDM process. Based on the taguchi quality
design and the analysis by Response surface method,
the significant factors affecting the machining
performance such as MRR, surface roughness,
sparking frequency, average gap voltage, and normal
ratio are determined. By means of regression
analysis, mathematical models relating the
machining performance and various machining
parameters are established. Based on the
mathematical models developed, an objective
function under the multi-constraint conditions is
formulated. Experimental results demonstrate that
the machining models are appropriate and the
derived machining parameters satisfy the real
requirements in practice.
Pratik A. Patil & C.A. Waghmare et al [2]
used the Response Surface Methodology approach
for maximizing the material removal rate in wire
ISSN: 2231-5381
electrical discharge machining. The investigated
machining parameters were wire tension, pulse on
time and peak current. Machining was carried on
AISI D2 cold work steel, which is widely used in die
and mould making industries. The experiments were
designed based on response surface design method;
in which central composite design method was
applied for fitting the second order model. After the
experimentation, the effect of the parameters on
MRR was determined by analysis of variance
(ANOVA). Also the interaction of their parameters
was considered for their significance. Regression
analysis was done and a second order mathematical
model was fitted for MRR considering the
parameters and their significant interactions.
R.Pandithurai & I. Ambrose Edward et
al [3] optimized the process parameters for Wire
electro discharge machining (WEDM).WEDM is
extensively used in tool and die industries. Precision
and intricate machining are the strengths. While
machining time and surface quality still remains as
major challenges. The main objective of this study is
to obtain higher material removal rate (MRR) and
lower surface roughness (SR). Ton, T off, Gap
voltage and wire feed rate are the four control
factors taken each at various levels. The genetic
algorithm optimization tool is used to find the
factors level that create a low surface roughness in
WEDM.
S V Subrahmanyam & M. M. M. Sarcar
et al [4] has demonstrated the optimization of Wire
Electrical Discharge Machining process parameters
for the machining of H13 HOT DIE STEEL, with
multiple responses Material Removal Rate (MRR),
surface roughness (Ra) based on the Grey–Taguchi
Method. Taguchi’s L27 (21x38) Orthogonal Array
was used to conduct experiments, which correspond
to randomly chosen different combinations of
process parameter setting, with eight process
parameters: TON, TOFF, IP, SV WF, WT, SF, WP
each to be varied in three different levels. Data
related to the each response viz. material removal
rate (MRR), surface roughness (Ra) have been
measured for each experimental run; The variation
of output responses with process parameters were
mathematically modelled by using non-linear
regression analysis. The models were checked for
their adequacy. Result of confirmation experiments
showed that the established mathematical models
can predict the output responses with reasonable
accuracy.
John O. Rawlings, Sastry G. Pantula &
David A. Dickey et al [5 developed least squares
and related statistical methods without becoming
excessively mathematical. The emphasis is on
regression concepts, rather than on mathematical
proofs. Proofs are given only to develop facility with
matrix algebra and comprehension of mathematical
http://www.ijettjournal.org
Page 210
International Journal of Engineering Trends and Technology (IJETT) – Volume 29 Number 4 - November 2015
relationships. Good students, even though they may
not have strong mathematical backgrounds, quickly
grasp the essential concepts and appreciate the
enhanced understanding. The learning process is
reinforced with continuous use of numerical
examples throughout the text and with several case
studies.
Scott F. Miller, Albert J. Shih & Jun Qu
et al [6] developed a new, advanced engineering
materials and the need for precise and flexible
prototypes and low volume production have made
the wire electrical discharge machining (EDM) an
important manufacturing process to meet such
demands. This research investigates the effect of
spark on-time duration and spark on-time ratio, two
important EDM process parameters, on the material
removal rate (MRR) and surface integrity of four
types of advanced material: porous metal foams,
metal bond diamond grinding wheels, sintered NdFe-B magnets, and carbon–carbon bipolar plates. An
experimental procedure was developed. During the
wire EDM, five types of constraints on the MRR due
to short circuit, wire breakage, machine slide speed
limit, and spark on-time upper and lower limits are
identified. An envelope of feasible EDM process
parameters is generated for each work-material.
Applications of such a process envelope to select
process parameters for maximum MRR and for
machining of micro features are discussed. Results
of Scanning Electron Microscopy (SEM) analysis of
surface integrity are presented.
Fig -3.2 Specimens obtained from Wire EDM
3.2 Design of Experiments
Design of Experiments is an analytical
method commonly used to statistically signify the
relationship between input parameters to output
responses. DOE has wide applications especially in
the field of engineering for the purpose of process
development, optimization and validation tests. DOE
is essentially an experimental based modeling and is
a designed experimental approach which is far
superior to unplanned approach whereby a
systematic way will be used to plan the experiment,
collect the data and analyze the data. In the present
work a mathematical model has been developed by
Response Surface Methodology. Optimization and
Desirability functions obtained helps to optimize
characteristics considered for Maximum Material
Removal Rate and Minimum Surface Roughness.
Process variable for Wire EDM are Considered as
Pulse on Time(T-on), Pulse off Time(T-off) and
Peak Current(IP). Process variable considered based
on the literature[3] and Machine settings are shown
in Table-3.1
III. EXPERIMENTATION
3.2 .1Process Variables
25
6
3.1Preparing the Specimens
Ti6Al4V material was purchased from
SOUTH ASIA Metal Corporation, Mumbai. Raw
material is in the form of a bar of 25.4mm diameter.
Four pieces of 10mm length and 25.4mm diameter
are cut on power saw. Using wire EDM seven
specimens from each block are obtained by varying
the factor considered in DOE. 27 specimens are used
for the present work. Drawing for Wire EDM is
shown in Fig-3.1 and specimens obtained after
machining are shown in Fig-3.2.
Table 3.1: Process variables and their limits
Levels
S.No Parameters Symbol
1
2
3
Pulse-on
1
time (micro
Ton
100 105 110
secs)
Pulse-off
2
time (micro
Toff
45
50
55
secs)
Peak
3
current
Ip
10
11
12
(amp)
3.3 Minitab Software
10
6
Fig 3.1 Drawing for wire EDM
ISSN: 2231-5381
Minitab is a statistics package. It was
developed at the Pennsylvania State University by
researchers Barbara F. Ryan, Thomas A. Ryan, Jr.,
and Brian L. Joiner in 1972. Minitab began as a light
version of MNITAB, a statistical analysis program
by NIST. Minitab is distributed by Minitab Inc, a
privately owned company headquartered in State
College. Pennsylvania, with subsidiaries in Coventry,
http://www.ijettjournal.org
Page 211
International Journal of Engineering Trends and Technology (IJETT) – Volume 29 Number 4 - November 2015
England, Paris, France and Sydney, Australia.
Minitab is often used in conjunction with the
implementation of Six sigma, CMMI and other
statistics-based process improvement methods.
3.4 Full Factorial Method
Experiments have been carried out using
full factorial method. Experimental design which
consists of 27 combinations of Ton, Toff and IP.
According to the design catalogue prepared by
factorial design of experiment has been found
suitable in the present work. It considers three
process parameters to be varied in three discrete
levels. The experimental design has been shown in
Table 3.2 (all factors are in coded form). Factorial
design is used for conducting experiments as it
allows study of interactions between factors.
Interactions are the driving force in many processes.
3.5 Material Removal Rate
The material removal rate of the work piece
is calculated by using the formula
MRR (2Wg D ) t Vc mm3/min
Where: Wg = Spark gap, varies from 0.04mm to
0.06mm,
D = diameter of the wire = 0.25mm
t = Thickness of the work piece in mm
Vc = Cutting speed in mm/min
3.6 Surface Roughness
Surface roughness is measured using
Surface Roughness Tester which directly shows the
reading when placed on the metal surface. It consists
of a stylus which moves on to the surface of the
metal. This directly shows the surface roughness
value in terms of any desired unit. Fig 3.3 shows the
surface roughness tester.
Table 3.2 DOE in Coded form
Expt No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
Ton
1
3
1
3
1
2
3
3
3
1
1
2
1
2
2
3
1
3
3
2
2
2
3
2
1
1
2
Toff
2
1
2
3
1
1
2
3
3
1
1
3
2
2
1
2
3
1
2
1
3
2
1
3
3
3
2
IP
2
3
3
1
3
1
3
3
2
1
2
2
1
1
3
1
1
2
2
2
1
2
1
3
2
3
3
Fig 3.3 Surface Roughness Tester
IV.
RESUTS AND DISCUSSIONS
4.1 Response Surface Methodology
Response surface methodology uses
statistical models, and therefore even the best
statistical model is an approximation to reality. In
practice, both the models and the parameter values
are unknown, and subject to uncertainty on top of
ignorance. An estimated optimum point need not be
optimum in reality, because of the errors of the
estimates and of the inadequacies of the model.
Response surface methodology has an effective
track-record of helping researchers improve products
and services.
4.2 Mathematical model of Response Surface
Methodology
The Response Surface is described by a
second order polynomial equation of the form
Y is the corresponding response
(1,2, . . . , S) are coded levels of S quantitative
process variables,
ISSN: 2231-5381
http://www.ijettjournal.org
Page 212
International Journal of Engineering Trends and Technology (IJETT) – Volume 29 Number 4 - November 2015
The terms are the second order regression
coefficients,
Second term is attributable to linear effect,
Third term corresponds to the higher-order effects,
Fourth term includes the interactive effects,
The last term indicates the experimental error.
4.2.1. Response Surface Methodology in Minitab
In this paper Minitab17 Trial version is
used to obtain the Response Surface Regression
equations, Fitted values and residual for Material
Removal Rate and Surface Roughness.
4.3 Mathematical Relationship between the Input
Parameters and Metal Removal Rate
The
mathematical
relationship
for
correlating the Metal removal rate and the
considered process variables has been obtained as
follows
MRR
=
-54+ 0.99 Ton- 0.41 Toff+ 4.6 IP0.0033 Ton*Ton- 0.0108 Toff*Toff0.066 IP*IP+ 0.0077 Ton*Toff- 0.0565 Ton*IP
+ 0.0520 Toff*IP
4.3.1 Normal Probability Plot for MRR
The normal probability plotin the Fig:4.1
shows a clear pattern indicating that all the factors
and their interaction given in are affecting the MRR.
In addition, the errors are normally distributed and
the regression model is well fitted with the observed
values.
Fig: 4.1 Normal Probability Plot for MRR
Table 3.1 Predicted values and % Error for MRR
S.No
T-On
T-
IP
MRR
MRR
%
1
100
10
12.21
2
105
45
Off
50
11
13.53
12.63
Predict
12.449
3.44
Error
7.99
3
105
55
10
12.54
11.49
8.37
4
100
50
10
12.87
11.9
7.54
5
105
55
12
12.87
11.641
9.55
6
100
50
12
12.54
12.096
3.54
7
100
50
11
13.45
12.064
10.3
8
105
55
10
12.54
11.49
8.37
9
105
45
10
11.81
13.105
10.97
10
110
45
11
14.19
12.754
10.12
11
110
55
12
12.87
11.771
8.54
12
100
55
12
10.56
11.346
7.44
13
100
55
11
10.45
11.054
5.78
14
110
45
12
12.54
11.961
4.62
15
110
55
11
12.54
12.044
3.96
16
110
50
11
11.55
12.669
9.69
17
105
45
12
11.55
12.216
5.77
4.3.3 Main Effect of Input Parameters
18
105
45
11
13.2
12.7265
3.59
19
105
50
10
12.21
12.5675
2.93
20
110
50
11
11.55
12.669
9.69
21
110
50
10
14.52
13.07
9.99
A main effect occurs when the mean
response changes across the levels of a factor. Main
effect plots are used to compare the relative strength
of the effects across factors.
22
110
55
10
11.88
12.185
2.57
23
100
45
12
12.54
12.306
1.87
24
105
50
12
13.53
12.1985
9.84
25
110
45
10
15.475
13.415
13.31
26
100
45
11
14.484
12.534
13.46
27
105
55
11
12.21
11.6315
4.74
ISSN: 2231-5381
4.3.2 Standardized Residual Vs Fitted Value for
MRR
Fig:4.2 indicates that the maximum
variation which shows the high correlation that,
exists between fitted values and observed values.
Fig: 4.2 Residual vs Fitted Values for MRR
http://www.ijettjournal.org
Page 213
International Journal of Engineering Trends and Technology (IJETT) – Volume 29 Number 4 - November 2015
9
105
45
10
3.88
3.6405
6.17
10
110
45
11
4.085
3.90075
4.51
11
110
55
12
4.24
4.45275
5.02
12
100
55
12
3.945
3.72175
5.66
13
100
55
11
3.765
3.17675
15.62
12.4
14
110
45
12
4.205
4.56875
8.65
12.2
15
110
55
11
3.96
3.70275
6.5
16
110
50
11
4.625
4.419
4.45
17
105
45
12
4.485
4.1155
8.24
18
105
45
11
3.345
3.55
6.13
19
105
50
10
3.84
4.12775
7.49
20
110
50
11
4.625
4.419
4.45
21
110
50
10
4.775
4.366
8.57
22
110
55
10
3.57
3.60875
1.09
23
100
45
12
4.31
3.79775
11.89
24
105
50
12
5.235
4.68475
10.51
25
110
45
10
3.635
3.88875
6.98
26
100
45
11
3.205
3.33475
4.05
27
105
55
11
3.33
3.372
1.26
M ain Effects Plot for M RR
Fitted Means
Ton
Toff
IP
13.2
Mean of MRR
13.0
12.8
12.6
12.0
100
105
110 45
50
55 10
11
12
All displayed terms are in the model.
Fig: 4.3 Main Effects plot for MRR
4.3.4 Interaction Effects
Fig: 4.4 Interaction Effects plot for MRR
4.4 Mathematical Relationship between the Input
Parameters and Surface Roughness
The
mathematical
relationship
for
correlating the Metal removal rate and the
considered process variables has been obtained as
follows
Ra = 30.1 - 0.72 Ton + 2.403 Toff - 9.50 IP
+ 0.00271 Ton*Ton - 0.02469 Toff*Toff
+ 0.328 IP*IP - 0.00040 Ton*Toff + 0.0205 Ton*IP
+ 0.0082 Toff*IP
4.4.1 Normal Probability Plot For Ra
The normal probability plot as shown in
Fig:4.5 represents a clear pattern (as the points are
almost in a straight line) indicating that all the
factors and their interaction given in are affecting the
Ra. In addition, the errors are normally distributed
and the regression model is well fitted with the
observed values
Table 3.2 Predicted values and % Error for Ra
S.No
T-On
T-
IP
Ra
Ra
%Error
1
100
Off
45
10
3.56
Predict
3.52775
0.91
2
105
50
11
4.24
4.07825
3.81
3
105
55
10
3.185
3.3805
6.14
4
100
50
10
4.64
4.025
13.25
5
105
55
12
4.08
4.0195
1.48
6
100
50
12
3.69
4.377
18.62
7
100
50
11
3.105
3.873
24.73
8
105
55
10
3.185
3.3805
6.14
ISSN: 2231-5381
Fig: 4.5 Normal Probability Plot for Ra
4.3.2 Standardized Residual Vs Fitted Value for
Surface Roughness
Fig:4.6 indicates that the maximum
variation which shows the high correlation that,
exists between fitted values and observed values.
http://www.ijettjournal.org
Page 214
International Journal of Engineering Trends and Technology (IJETT) – Volume 29 Number 4 - November 2015
Fig: 4.6 Residual vs Fitted Values for Surface
Roughness
Fig: 4.6 Optimisation plot for MRR and Surface
Roughness
4.3.3 Main Effect of Input Parameters
M ain Effects Plot for Ra
Fitted M eans
Ton
Toff
IP
4.75
Mean of Ra
4.50
4.25
4.00
3.75
3.50
100
105
110 45
50
55 10
11
12
All displayed terms are in the model.
Fig: 4.7 Main Effects plot for Surface Roughness
4.3.4 Interaction Effects
The optimization plot as shown in the fig
signifies the affect of each factor (columns) on the
responses or composite desirability (rows). The
vertical red lines on the graph represent the current
factor settings. The numbers displayed at the top of a
column show the current factor level settings (in red).
The horizontal blue lines and numbers represent the
responses for the current factor level. Minitab
calculates the maximum material removal rate and
minimum surface roughness.
From the optimization plot it can be said
that the maximum material removal rate is 13.4545
and the minimum surface roughness is 3.5568
obtained when Ton is 104.6465, Toff is 45.0, IP is
10.3434.
V. CONCLUSION
Fig: 4.4 Interaction Effects plot for Surface
Roughness
4.4 Optimisation Plot:
A Minitab response optimizer tool shows
how different experimental settings affect the
predicted responses for factorial, response surface,
and mixture designs. The optimal solution serves as
the starting point for the plot. This optimization plot
allows to interactively changing the input variable
settings to perform sensitivity analyses and possibly
improve the initial solution.
ISSN: 2231-5381
In this work, two performance parameters
Surface Roughness and Material Removal Rate are
investigated by varying the three Process (machining)
parameters on Ti6Al4V with Brass wire as electrode
in wire electric discharge machine. The performance
parameters included pulse on time (Ton), Pulse off
time (Toff) and Input Voltage (IP). Experiments
were conducted according to Full factorial Design.
The optimum parameters value combination was
found which would yield minimum Surface
Roughness (Ra) and maximum Material Removal
Rate (MRR) .
Regression Equation has been obtained
successfully, to find the Material Removal Rate and
Surface Roughness.
Response optimizer in Minitab software is used
for optimization of Surface
Roughness and
Material Removal Rate.
Experimental values and Predicted values
obtained by Response surface regression are good in
agreement with less than 10% error.
http://www.ijettjournal.org
Page 215
International Journal of Engineering Trends and Technology (IJETT) – Volume 29 Number 4 - November 2015
REFERENCES
[1] Vikram Singh & S.K. Pradhan, Optimization of WEDM
parameters using Taguchi technique and Response Surface
Methodology in machining of AISI D2 Steel, Procedia
Engineering
97
( 2014 )
1597 – 1608,
www.elsevier.com/locate,procedia
[2] Pratik a. Patil & C.A. Waghmare, optimization of process
parameters in wire-edm using Response surface methodology,
proceedings of 10 irf international conference, 01 june-2014, pune,
india, isbn: 978-93-84209-23-0.
[3] R.Pandithurai, I. Ambrose Edward, Optimizing surface
roughness in Wire EDM using machining parameters,
International Journal of Innovative Research in Science,
Engineering and Technology , Volume 3, Special Issue 3, March
2014 2014 IEEE International Conference on Innovations in
Engineering and Technology (ICIET’14)
[4] S V Subrahmanyam, M. M. M. Sarcar, Evaluation of Optimal
Parameters for machining with Wire cut EDM Using GreyTaguchi Method, International Journal of Scientific and Research
Publications, Volume 3, Issue 3, March 2013 ISSN 2250-3153
[5] John O. Rawling, Sastry G. Pantula, David A. Dickey,
Applied Regression Analysis: A Research Tool,Second Edition,
Springer Texts in Statistics
[6] Scott F. Miller, Albert J. Shih, Jun Qu, Investigation of the
spark cycle on material removal rate in wire electrical discharge
machining of advanced materials, International Journal of
Machine Tools & Manufacture 44 (2004) 391–400,
www.elsevier.com/locate/ijmatool
[7] Titanium: A Technical Guide, 2nd Edition (#06112G),
Introduction
to
Selection
of
Titanium
Alloys,
www.asminternational.org
[8] Y.S.Liao, J.T.Huang, and H.C.Su , A study on the machiningparameters optimization of wire electrical discharge machining, J.
Mater. Process. Technol, 71, 487–493(1997).
[9] T.A. Spedding, and Z.O. Wang , Parametric optimization and
surface characterization of wire electrical discharge machining
process. Precision Eng, 20, 5–15 (1997).
[10] Tosun, N., Determination of optimum parameters for multi
performance characteristics in drilling by using grey relational
analysis. Int,.J. Adv. Manuf. Technol., 28, 450–455. 2006
[11] Y.S Liao , J.T.Huang, A study on the machining parameter
optimization of
WEDM, Journal of Material Processing
Technology,71(1997) pp.487-493.
[12] S.S Mohapatra, Amar pattnaik,Optimization of WEDM
process parameters using Taguchi method, International Journal
of Advanced manufacturing Technology (2006).
[13] R. E. Williams and K. P. Rajurkar, Study of wire electrical
discharged machine surface characteristics, Journal of Materials
Processing Technology,28(1991) pp. 127-138.
ISSN: 2231-5381
http://www.ijettjournal.org
Page 216
Download