parametric study and optimization of wedm parameters for ck45 steel

www.seipub.org/ijepr International Journal of Engineering Practical Research (IJEPR) Volume 2 Issue 4, November 2013

156

Parametric Study and Optimization of WEDM

Parameters for CK45 Steel Taha A. El‐Taweel*1, Ahmed M. Hewidy2

Production Engineering and Mechanical Design Department

Faculty of Engineering, Menoufia University, Shebin El‐Kom, Egypt

*1 [email protected]

Abstract

CK45 steel is a material that has a wide range of applications

in military and automotive industry. It can be classified as a

difficult‐to‐cut material, not suitable for traditional

machining. The present paper presents a study of the

optimum selection of wire electric discharge machining

(WEDM) conditions for CK45 steel. Feeding speed, duty

factor, water pressure, wire tension and wire speed have

been considered the main factors affecting WEDM

performance criteria. The process performances such as;

material removal rate (MRR), tool wear rate (TWR), and

surface roughness (SR) were evaluated. Response surface

methodology (RSM) was employed to develop experimental

models. The surface composition of the machined surface

was examined through X‐ray diffraction technique. The

results reveal that the feeding speed and duty factor are the

most vital factors controlling the MRR. Water pressure has

been found a sensible effect of MRR especially at a higher

feeding speed. Further, the minimum tool wear rate has

been obtained at the parametric combination of low feeding

speed and low duty factor. It has also been clarified that the

increase of wire tension and wire speed lead to a relatively

insignificant change in surface quality.

Keywords

WEDM, Material Removal Rate; Tool Wear Rate; Surface

Roughness; Response Surface Methodology; Optimization

Introduction

Electrical discharge machining (EDM) is one of the

nontraditional machining processes which has been

widely used to produce dies and molds, as well to

finish parts for aerospace and automotive industry

(Snoeys et al., 1986). This technique was developed in

the late 1940s of which the process is based on

removing material from a part by means of a series of

repeated electrical discharges between a tool called the

electrode and the workpiece in the presence of a

dielectric fluid [(Ho and Newman, 2003), (Bojorquez et

al., 2002)]. Wire electro‐discharge machining (WEDM)

uses the same principles of material removal as EDM,

but using a traveling, small‐diameter wire as the

electrode. The wire travels from a supply spool,

through the workpiece, and onto a take‐up spool.

Hasçalýk et al. (2004) presented an experimental

investigation of the machining characteristics of AISI

D5 tool steel using WEDM process. During their

experiments, parameters such as open circuit voltage,

pulse duration, wire speed and dielectric pressure

were changed to explore their effect on the surface

roughness and metallurgical structure, then using

optical and scanning electron microscopy, surface

roughness and microhardness tests to study the

characteristics of the machined specimens. They found

that the intensity of the process energy does affect the

amount of recast layer and surface roughness as well

as microcracking.

Wire electro‐discharge machining induced a new

phase in Ti–46Al–2Cr intermetallic alloy which has

been identified using X‐ray diffraction technique by

Qin et al, (2003). It was observed that a new phase was

found to be neither TiO2 nor Al2O3, but well consistent

with titanium hydride with the lattice parameter of

4.49–4.50 A˚. Further, the new phase exists in the

WEDM cut alloy surface layer limited to about 70 μm

thickness, much deeper than the WEDM induced

recast layer thickness of 2–3 μm. On processing,

vacuum annealled at 400–600˚C, has been proposed to

remove this hydride induced by WEDM. In addition,

it was shown that the WEDM induced microcracks

extensively happened on the EDM‐cut surface and

penetrated into a matrix of up to 10–30 μm.

Modeling in WEDM helps get a better understanding

of the complex process (Abbas, 2007, Puertas and Luis,

2003, Ho et al, 2004). Semi empirical models of surface

finish, MRR and wire wear [(Tsai and Wang, 2001),

(Poroś and Zaborski, 2009)] of the WEDM process for

various materials have been established by employing

dimensional analysis. Using the design of the

experiment method, the process parameter viz. peak

current, pulse duration, electric polarity and material

International Journal of Engineering Practical Research (IJEPR) Volume 2 Issue 4, November 2013 www.seipub.org/ijepr

157

properties are identified. The final results showed that

the average error between the experiment and

prediction was less than 10% for surface finish model

and less than 20% for MRR.

The modeling was first introduced by Jeswani (1979)

using dimensional analysis to predict the tool wear.

After that, various methods were introduced to

predict the output of EDM process. In addition, the

modeling of the WEDM process by means of

mathematical techniques has also been applied to

effectively relate the large number of process variables

to the different performance of the process (Yahya,

2004). Sharif and Noordin (2006) established an

empirical relation of machining responses using the

RSM technique while machining Ti‐6246 workpiece by

adding SiC powder in the dielectric. The results show

that MRR, tool wear, and surface quality are greatly

influenced by the current, voltage, and pulse on‐time,

and the surface roughness is highly influenced by the

presence of SiC additives.

Chiang et al. (2006) presented an approach for

optimization the WEDM process of Al2O3 particle

reinforced material (6061 alloy) with multiple

performance characteristics based on the grey

relational analysis and Taguchi technique. In their

study, the machining parameters, namely the cutting

radius of workpiece, on‐time, off‐time of discharging,

arc on‐time, arc off‐time of discharging voltage, wire

feed and water flow, are optimized with considerations

of multiple performance, such as MRR and SR.

El‐Taweel (2006) studied the effect and optimization of

machining parameters on the MRR and SR in the

WEDM process of some Al‐Cu‐TiC‐Si composite. The

variation of MRR and SR with machining parameters

was mathematically modeled by using the nonlinear

regression analysis method. The optimal search for

machining parameters for the objective of maximum

MRR and minimum SR was performed. Based on the

experimental confirmation, it was observed that the

material removal rate increased by 1.58 times and

surface roughness was decreased by 1.11 times.

The objective of using response surface methodology

(RSM) is to not only investigate the response over the

entire factor space, but also locate the region of interest

where the response reaches its optimum or near

optimal value. RSM technique has also been applied to

optimize the input of WEDM parameters for Inconel

601 by Hewidy el al. (2005)who reported that the

analysis of response parameters using RSM technique

had the advantage of explaining the effect of each

working parameter on the value of resultant response

parameter.

Derringer and Suich (1980) described a multiple

response method called desirability that is an

attractive method to industry for optimization of

multiple quality characteristic problems, making use

of an objective function, called the desirability function

and transforming an estimated response into a scale

free value called desirability. The desirable ranges are

from zero to one; of which one represents the ideal

case; while zero indicates that one or more responses

are outside their acceptable limits. Composite

desirability is the weighted geometric mean of the

individual desirability for the responses. The factor

settings with maximum total desirability are

considered to be the optimal parameter conditions

[(Castillo, 1996), El‐Taweel, 2009)].

CK45 steel, a material that has a wide range of

applications in military and automotive industries can

be classified as a difficult‐to‐cut material, not suitable

for traditional machining. The published literature has

indicated that few studies have been reported for the

optimization of process parameters in WEDM for

CK45 steel. El‐Taweel (2009) investigated the

correlation of process parameters in die sinking EDM

of CK45 steel with Al–Cu–Si–TiC composite electrode

produced using powder metallurgy (P/M) technique

and evaluated MRR and TWR. It has been found that

such electrodes are more sensitive to peak current and

on‐time than conventional electrodes. To achieve

maximum MRR, minimum TWR and SR, the process

parameters are optimized and experimental

verification were found to be in good agreement.

However, it has been found that there is no available

data of the WEDM process of this material. Therefore,

it is imperative to develop a suitable technology

guideline for optimum WEDM of CK45 steel. So the

aim of the present work is to develop models to

correlate the inter‐relationships of various WEDM

parameters of CK45 steel such as feeding speed (Fs),

duty factor (D), wire tension (T), wire speed (Ws) and

water pressure (P) on MRR, TWR and SR.

Experimental Details

Experimental Procedure and Measurments

In the present study, MRR, TWR and SR have been

considered to evaluate the machining performance.

MRR, TWR and SR are correlated with input

machining parameters such as feeding speed (Fs), duty

factor (D), wire Tension (T), wire Speed (Ws) and

water Pressure (P). The experiments were conducted


158

on a W‐B30J.S CNC wire electrical discharge machine.

Figure 1 shows the used experimental set up attached

with WEDM machine.

FIG. 1 THE USED EXPERIMENTAL SET UP

The wire electrode material used in the WEDM

process was made from Brass‐CuZn37 with diameter

0.25 mm. The workpiece material used in these

experiments was CK45 steel with dimensions 8×8×8

mm3. Table 1 shows the chemical composition of CK45

steel. The selection of this material was made, taking

into account its wide range of applications in military

and automotive industries (El‐Taweel, 2009). It can be

classified as a difficult‐to‐cut material, not suitable for

traditional machining. During machining, de‐ionized

water was circulated as the dielectric fluid around

wire (co‐axial) and with side flushing technique. The

machining time for each experiment was variable and

related to the machining conditions and fixed

parameters which are listed in Table 2. These were

chosen through reviews of experience, literature

surveys [(Abbas, 2007), (Puertas and Luis, 2003), (Ho

et al, 2004), (Tsai and Wang, 2001), (Poroś and

Zaborski, 2009), (Chiang et al. 2006)], and some

preliminary investigations.

TABLE 1 CHEMICAL COMPOSITION FOR CK45 STEEL (WT%)

C Si Mn P S Cr Mo

0.4771 0.2105 0.628 0.0109 0.0111 0.1079 0.0236

Al Co Cu W Sn Fe Ni

0.0073 0.0073 0.1896 <0.005 0.0103 98.23 0.0788

TABLE 2 WEDM MACHINING CONDITIONS

Working condition Value

Workpiece material CK45 steel

Tool (+) Polarity

Wire material Brass CuZn37

Feeding speed 1.7‐ 4.9 mm/min

Pulse‐on time 1‐9 μs

Pulse‐off time 9 μs

Wire tension 800‐2400 gf

Wire speed 5‐15 m/s

Water pressure 1‐3 MPa

Wire diameter 0.25 mm

Peak current Automatically selected

Dielectric fluid De‐ionized water

The specimen and wire were weighted before and

after machining using a digital balance (Sartorius, type

1712, 0.0001 g).

The metal removal rate was specified using the

following equation:

MRR = (Wb‐ Wa)/t (g/min) (1)

where Wb: specimen weight before machining (g); Wa:

specimen weight after machining (g).; t: machining

time (min).

The tool wear rate was evaluated through wire weight

for a length of one meter before and after machining

by using the following equation:

TWR = 60 (Tb‐ Ta)Ws (g/min) (2)

Where Tb: wire weight for one meter before machining

(g/m). Ta: wire weight for one meter after machining

(g/m). Ws: wire speed (m/s).

The roughness average (Ra) for each specimen was

measured by surface tester (Mitutoye, Surftest‐402).

Each surface roughness value was obtained by

averaging four measurements. The instrument is

adjusted to the following conditions during

measurement; measuring length (5mm), measuring

speed (0.5 mm/s) and sampling length (0.8 mm).

The microscopic investigations of the WEDM surface

are investigated by (JEOL JSM‐5200LV) scanning

electron microscope (SEM). X pert pro‐panalytical‐X

ray diffract meter (XRD) is used to investigate the

surface component after WEDM.

Experimental Design

Response surface methodology (RSM) is an interaction

of mathematical and statistical techniques for

modelling and optimizing the response variable

models involving quantitative independent variables

[(Montgomery, 2001), (Antony, 2003)]. In the present

work, the ranges of the feeding speed (Fs), duty factor

(D), wire tension (T), wire speed (Ws) and water

pressure (P) settings have been selected. Actual and

coded values of the input process parameters have

been listed in Table 3. Experiments have been carried

out according to the experimental plan based on

central composite half‐fraction second‐order rotatable

design (CCD) as shown in Table 4. This design

consists of 32 factorial runs, 16 axial runs and the

number of star points is 10. The extra seven points are

added at the center to give roughly equal precision for

Yu within a circle of radius 1. Yu is the uth response in

the experimentation. One important disadvantage of

using second order polynomial in RSM is that the size


159

of experiments becomes too large and analysis

becomes too complicated with more than three X

variables or with more than three levels. Central

composite designs (CCD) are one of those means.

Proceeding a step ahead, central composite rotatable

designs of second order have been found to be the

most efficient tool in RSM to establish the

mathematical relation of the response surface using

the smallest possible number of experiments without

losing its accuracy.

Mathematical Modelling

Through utilizing the design of experiments and

applying regression analysis, the modelling of the

desired response to several independent variables can

be obtained [(Montgomery, 2001), (Antony, 2003)]. If

all variables are assumed to be measurable, the

quadratic response surface model of Yu can be written

as follows:

2

1 1

k k kY b b X b X X b Xu i i ij i ii io jj ii i

(3)

where Yu is the corresponding response function (or

response surface), X1, X2, X3.....Xk are coded values of

the machining process parameters and ε is the fitting

error of the uth observation. In this study, for five

variables under consideration (Fs, D, T, Ws and P), a

second order polynomial regression model, called

quadratic model, is proposed.

TABLE 3 CODED AND ACTUAL VALUES OF INPUT PARAMETERS

Input parameters Symbol Levels

‐2 ‐1 0 +1 +2

Feeding speed (Fs), mm/min X1 1.7 2.5 3.3 4.1 4.9

Duty factor (D) X2 0.1 0.2 0.3 0.4 0.5

Wire tension (T), gf X3 800 1200 1600 2000 2400

Wire speed (Ws), m/s X4 5 7.5 10 12.5 15

Water pressure (P), MPa X5 1 1.5 2 2.5 3

TABLE 4 EXPERIMENTAL DESIGN MATRIX AND RESULTS

Exp. No.

Input process parameters Experimental results

Fs, D T, Ws, P, MRR, g/min TWR, g/min Ra, μm

Coded Actual Coded Actual Coded Actual Coded Actual Coded Actual

1 ‐1 2.5 ‐1 0.2 ‐1 7.5 ‐1 1200 1 2.5 0.0391 0.0260 2.3

2 1 4.1 ‐1 0.2 ‐1 7.5 ‐1 1200 ‐1 1.5 0.0453 0.0440 2.4

3 ‐1 2.5 1 0.4 ‐1 7.5 ‐1 1200 ‐1 .1.5 0.0685 0.0470 2.5

4 1 4.1 1 0.4 ‐1 7.5 ‐1 1200 1 2.5 0.0735 0.0510 2.7

5 ‐1 2.5 ‐1 0.2 1 12.5 ‐1 1200 ‐1 1.5 0.0371 0.0360 2.8

6 1 4.1 ‐1 0.2 1 12.5 ‐1 1200 1 2.5 0.0426 0.0450 2

7 ‐1 2.5 1 0.4 1 12.5 ‐1 1200 1 2.5 0.0698 0.0480 2.5

8 1 4.1 1 0.4 1 12.5 ‐1 1200 ‐1 1.5 0.0739 0.0510 2.6

9 ‐1 2.5 ‐1 0.2 ‐1 7.5 1 2000 ‐1 1.5 0.0347 0.0340 2.7

10 1 4.1 ‐1 0.2 ‐1 7.5 1 2000 1 2.5 0.0650 0.0460 2.8

11 ‐1 2.5 1 0.4 ‐1 7.5 1 2000 1 2.5 0.0601 0.0490 3.3

12 1 4.1 1 0.4 ‐1 7.5 1 2000 ‐1 1.5 0.0755 0.0520 2.5

13 ‐1 2.5 ‐1 0.2 1 12.5 1 2000 1 2.5 0.0374 0.0410 2.7

14 1 4.1 ‐1 0.2 1 12.5 1 2000 ‐1 1.5 0.0569 0.0430 2.8

15 ‐1 2.5 1 0.4 1 12.5 1 2000 ‐1 1.5 0.0474 0.0470 2.6

16 1 4.1 1 0.4 1 12.5 1 2000 1 2.5 0.0943 0.0490 2.5

17 ‐2 1.7 0 0.3 0 10 0 1600 0 2 0.0423 0.0270 1.9

18 2 4.9 0 0.3 0 10 0 1600 0 2 0.0743 0.0460 4.9

19 0 3.3 ‐2 0.1 0 10 0 1600 0 2 0.0284 0.0240 2.5

20 0 3.3 2 0.5 0 10 0 1600 0 2 0.0943 0.0480 2.2

21 0 3.3 0 0.3 ‐2 5 0 1600 0 2 0.0619 0.0410 3.2

22 0 3.3 0 0.3 2 15 0 1600 0 2 0.0701 0.0280 2.2

23 0 3.3 0 0.3 0 10 ‐2 800 0 2 0.0609 0.0310 3.5

24 0 3.3 0 0.3 0 10 2 2400 0 2 0.0758 0.0380 2.1

25 0 3.3 0 0.3 0 10 0 1600 ‐2 1 0.0615 0.0460 3.4

26 0 3.3 0 0.3 0 10 0 1600 2 3 0.0705 0.0250 2.5

27 0 3.3 0 0.3 0 10 0 1600 0 2 0.0659 0.0270 2.4

28 0 3.3 0 0.3 0 10 0 1600 0 2 0.0653 0.0280 2.4

29 0 3.3 0 0.3 0 10 0 1600 0 2 0.0655 0.0270 2.5

30 0 3.3 0 0.3 0 10 0 1600 0 2 0.0639 0.0270 2.4

31 0 3.3 0 0.3 0 10 0 1600 0 2 0.0645 0.0290 2.4

32 0 3.3 0 0.3 0 10 0 1600 0 2 0.0652 0.0270 2.5


160

The coefficient bo is the free term, the coefficients bi are

the linear terms, the coefficients bij are the interaction

terms and the coefficients bii are the quadratic terms.

Using the results presented in Table 4, the full form of

the derived models can be presented. Based on Eq. (3),

the effect of input parameters on values of metal

removal rate, tool wear rate and surface roughness has

been evaluated by computing the values of various

constants in Table 4. The mathematical models of MRR,

TWR, and SR can be expressed in coded form as

follows:

1 2 4

2 25 1 2

MRR 0.0662 0.0082 0.014 0.0025

0.002 0.003 0.0021

X X X

X X X

(4)

21 2 1

2 2 2 22 3 4 5

TWR 0.026 0.004 0.005 0.0035

0.0034 0.003 0.003 0.0033

X X X

X X X X

(5)

1 2 3 5

1 2 1 3 1 5 2 3

2 2 22 5 1 4 5

SR 9.99 1.73 10.963 0.236 4.08

1.77 0.045 0.255 0.4

5.275 0.048 3.196 0.47

X X X X

X X X X X X X X

X X X X X

(6)

TABLE 5 ANOVA FOR MRR

Source SS DF MS F value P.value Prob>F

Model 0.007773 20 0.000389 14.3581 < 0.0001

X1 0.001615 1 0.001615 59.6767 < 0.0001

X2 0.004724 1 0.004724 174.501 < 0.0001

X3 8.28E‐06 1 8.28E‐06 0.30602 0.5912

X4 0.00011 1 0.00011 4.05087 0.0693

X5 0.000153 1 0.000153 5.63409 0.0369

X1X2 6.13E‐06 1 6.13E‐06 0.22629 0.6436

X1X3 2.28E‐05 1 2.28E‐05 0.84231 0.3784

X1X4 0.000521 1 0.000521 19.2462 0.0011

X1X5 1.63E‐06 1 1.63E‐06 0.06005 0.8109

X2X3 2E‐05 1 2E‐05 0.73979 0.4081

X2X4 9.17E‐05 1 9.17E‐05 3.3869 0.0928

X2X5 3.11E‐05 1 3.11E‐05 1.14819 0.3069

X3X4 8.56E‐07 1 8.56E‐07 0.0316 0.8621

X3X5 1.43E‐05 1 1.43E‐05 0.52645 0.4833

X4X5 0.000111 1 0.000111 4.09231 0.0681

X12 0.000231 1 0.000231 8.55102 0.0138

X22 0.000123 1 0.000123 4.53887 0.0565

X32 2.29E‐05 1 2.29E‐05 0.84699 0.3771

X42 2.58E‐06 1 2.58E‐06 0.09532 0.7633

X52 2.29E‐05 1 2.29E‐05 0.84699 0.3771

Residual 0.000298 11 2.71E‐05

Lack of fit 0.000295 6 4.92E‐05 12.43 < 0.0001

Pure error 2.64E‐06 5 5.27E‐07

Core Total 0.008071 31

DF: Degree of freedom; SS: Sum of squares; MS: Mean of squares

The adequacy of the models is checked using the

analysis of variance (ANOVA) technique. As per this

technique, if the calculated F ratios of the developed

models do not exceed the standard tabulated values of

F ratio for desired level of confidence, then the models

are considered to be at the confidence level

(Montgomery, 2001). The ANOVA tables for MRR,

TWR and SR are presented in Tables 5, 6 and 7,

respectively. It can be noted that there are some terms

omitted from the equations. These terms are non‐

significant terms according to the ANOVA of the

models.

TABLE 6 ANOVA FOR TWR

Source SS DF MS F value P.value Prob>F

Model 0.002449 20 0.00012 12.875 <0.0001

X1 0.000345 1 0.00035 8.102 <0.0001

X2 0.000672 1 0.00067 15.78 <0.0001

X3 9.37E‐06 1 9.4E‐06 0.22 0.1281

X4 3.04E‐05 1 3E‐05 0.713 0.0030

X5 7E‐05 1 7E‐05 1.645 0.0091

X1X2 5.26E‐05 1 5.3E‐05 1.234 0.0044

X1X3 2.76E‐05 1 2.8E‐05 0.647 0.1495

X1X4 1.41E‐05 1 1.4E‐05 0.33 0.1026

X1X5 6.25E‐08 1 6.3E‐08 0.001 0.2307

X2X3 2.26E‐05 1 2.3E‐05 0.53 0.1007

X2X4 1.06E‐05 1 1.1E‐05 0.248 0.3279

X2X5 6.25E‐08 1 6.2E‐08 0.001 0.6188

X3X4 1.06E‐05 1 1.1E‐05 0.248 0.3279

X3X5 7.56E‐06 1 7.6E‐06 0.178 0.6188

X4X5 1.81E‐05 1 1.8E‐05 0.424 0.168

X12 0.000362 1 0.00036 8.493 <0.0001

X22 0.000336 1 0.00034 7.899 <0.0001

X32 0.000266 1 0.00027 6.246 <0.0001

X42 0.000266 1 0.00027 6.246 <0.0001

X52 0.000312 1 0.00031 7.326 <0.0001

Residual 0.000468 11 4.3E‐05

Lack of fit 0.000465 6 7.7E‐05 10.7 0.0140

Pure error 3.5E‐06 5 7E‐07

Core Total 0.002918 31

TABLE 7 ANOVA FOR SR

Source SS DF MS F value P. value Prob>F

Model 6.08 20 0.3 18.68 <0.0001

X1 6.08 20 0.3 18.68 <0.0001

X2 0.47 1 0.47 28.9 0.0002

X3 0.09 1 0.09 5.55 0.038

X4 1.17 1 1.17 72.18 <0.0001

X5 0.23 1 0.23 14.19 0.0031

X1X2 0.85 1 0.85 52.44 <0.0001

X1X3 0.32 1 0.32 19.67 0.001

X1X4 0.13 1 0.13 8.06 0.0161

X1X5 0.092 1 0.09 5.65 0.0367

X2X3 0.177 1 0.17 10.22 0.0085

X2X4 0.17 1 0.17 10.33 0.0082

X2X5 0.12 1 0.12 7.31 0.0205

X3X4 1.11 1 1.11 68.4 <0.0001

X3X5 0.02 1 0.02 1.2 0.2959

X4X5 0.022 1 0.022 1.38 0.2644

X12 0.028 1 0.47 28.84 0.0002

X22 0.03 1 0.028 1.75 0.2131

X32 0.09 1 0.03 1.84 0.2021

X42 0.1 1 0.09 5.56 0.038

X52 0.41 1 0.1 6.33 0.0287

Residual 0.18 11 0.41 25.1 0.0004

Lack of fit 0.17 6 0.016

Pure error 0.013 5 0.028 10.25 0.0107

Core Total 6.26 31 2.667 E‐003


161

Results and Discussion

Effect of Machining Parameters on MRR

Metal removal rate in the WEDM process is a vital and

significant factor due to its effect on the industrial

economy. Based on the RSM model, Fig. 2 shows the

effect of feeding speed on MRR at various duty factors,

while keeping the other parameters at centre level.

Figure 2 reflects that MRR increases as the feeding

speed of the wire increases. This result has been

attributed to the increase of the consumed applied

current which leads to the increase in the rate of the

heat energy and, hence, in the rate of melting and

evaporation (El‐Taweel, 2006). Furthermore, Fig. 2

shows that MRR also increases with the increase of the

duty factor. This trend is due to the fact that the

increase of the duty factor means the application of the

same heating temp erature for a long time on the

machined surface. In addition to that, Fig. 2 shows that

the effect of the duty factor is more pronounced than

the feeding speed. This phenomenon reflects that the

increase in peak current, as a function to the increase

of the feeding speed, leads to arcing, which relatively

limits the discharge number and machining efficiency.

1.7

2.5

3.3

4.1

4.9

5.0

7.5

10.0

12.5

15.0

0.034

0.044

0.055

0.066

0.076

MR

R, g

/min

Feeding speed, mm/min Wire speed, m/s

FIG. 2 EFFECT OF FEEDING SPEED ON MRR AT VARIOUS DUTY

FACTORS(T=1600 gf, Ws=10 m/s, P=2 MPa, working cond. Table 2)

Figure 3 shows the effect of feeding speed on MRR at

various wire speeds, and it is reflected that wire speed

generally has almost no effect on the MRR. However,

if the wire speed is reduced to a certain limit (5 m/s),

the cutting process becomes non‐uniform and the wire

electrode breaks down several times because of the

increase of the number of discharge per unit length of

wire. This result has also been reported by Kinoshita

(1976) who deduced that the wire winding speed had

almost no effect on the increase of the feeding rate.


various wire tensions, and that the wire tension has a

limited effect on the MRR value at the center values of

feeding speed and the other machining parameters.

This figure also reveals that the combination of higher

feeding speed and higher wire tension leads to larger

MRR. This phenomenon could be attributed to the

reduction of the wire oscillation which usually

decreases the dynamic conditions and, consequently,

enhances the stability of the process. This explanation

has also been reported by Yan and Pin (2004) during

their attempt to control the wire tension in the wire

EDM process.

1.7

2.5

3.3

4.1

4.9

800

1200

1600

2000

2400

0.021

0.041

0.060

0.080

0.100

MR

R, g

/min

Feeding speed, mm/min Wire tension, gf

FIG. 3 EFFECT OF FEEDING SPEED ON MRR AT VARIOUS WIRE

SPEEDS(D= 0.3, T=1600 gf, P=2 MPa, working cond. Table 2)

1.7

2.5

3.3

4.1

4.9

0.1

0.2

0.3

0.4

0.5

0.005

0.027

0.049

0.072

0.094

MR

R, g

/min

Feeding speed, mm/min Duty Factor

FIG. 4 EFFECT OF FEEDING SPEED ON MRR AT VARIOUS WIRE

TENSIONS(D= 0.3, Ws=10 m/s, P=2 MPa, working cond. Table 2)


various water pressures,as well as the nature of the

variation of the MRR value at different water

pressures ranging from 1 to 3 MPa. A limited effect

has been observed of water pressure at the low

feeding speed. However, this effect was significant

and effective at the higher values of the higher feeding

speeds. This result has been attributed to the fact that

the increase in feeding speed usually leads to an

increase in the rate of heat energy and the rate of

melting and evaporation. So, it is very necessary of the

increase of water pressure to decrease the tendency of

arcing and also to overcome the problem of the


162

increase in the number of gas bubbles and the bigger

volume of the molten metal (Luo, 1999).

1.7

2.5

3.3

4.1

4.9

1.0

1.5

2.0

2.5

3.0

0.039

0.050

0.061

0.072

0.083

MR

R, g

/min

Feeding speed, mm/min Water pressure, MPa

FIG. 5 EFFECT OF FEEDING SPEED ON MRR AT VARIOUS

WATER PRESSURES(D= 0.3, Ws= 10 m/s, T=1600 gf, working cond.

Table 2)

1.7

2.5

3.3

4.1

4.9

0.1

0.2

0.3

0.4

0.5

0.019

0.030

0.040

0.051

0.062

TW

R, g

/min


FIG. 6 EFFECT OF FEEDING SPEED ON TWR AT VARIOUS DUTY

FACTORS(T= 1600 gf, Ws= 10 m/s, P= 2 MPa, working cond. Table 2)

Effect of Machining Parameters on TWR

Tool wear rate in the WEDM process is a vital and

significant factor due to its effect on the industrial

economy of the machining process. It has also been

stated that it is impossible to achieve a higher MRR

and good surface phenomena simultaneously (Luo,

1999). The decrease of TWR is due to the drop of the

produced spark discharge energy at these conditions.

Figure 6 showing the effect of feeding speed on the

TWR at various duty factors indicates that the

minimum TWR has been obtained at the combination

of low feeding speed and low duty factor; while

Figure 7 showing the effect of feeding speed on the

TWR at various wire speeds indicates that wire

tension around 2000 gf and feeding speed about 3

mm/min represent the optimum TWR at certain wire

speed, wire tension and water pressure. The trends of

the above figure were nearly similar to the trend in Fig.

8. Furthermore, the optimum conditions of each figure

have been clarified through the 3‐D graphs. Figure 6

showing the effect of duty factor on the TWR at

various wire speeds, displays that min TWR has been

obtained at the combination of low Duty factor and

moderate values of wire speed. The decrease of TWR

is due to the drop of the length of the spark discharge

energy at these conditions.

1.7

2.5

3.3

4.1

4.9

5.0

7.5

10.0

12.5

15.0

0.025

0.035

0.045

0.055

0.065

TW

R, g

/min


FIG. 7 EFFECT OF FEEDING SPEED ON TWR AT VARIOUS WIRE

SPEEDS(T=1600 gf, D= 0.3, P= 2 MPa, working cond. Table 2)

1.7

2.5

3.3

4.1

4.9

800

1200

1600

2000

2400

0.024

0.032

0.040

0.048

0.056

TW

R, g

/min


FIG. 8 EFFECT OF FEEDING SPEED ON TWR AT VARIOUS WIRE

TENSIONS(D= 0.3, Ws=10 m/s, P= 2 MPa, working cond. Table 2)

1.7

2.5

3.3

4.1

4.9

1.0

1.5

2.0

2.5

3.0

0.024

0.034

0.044

0.054

0.064

TW

R, g

/min


FIG. 9 EFFECT OF FEEDING SPEED ON TWR AT VARIOUS

WATER PRESSURES (D= 0.3,T= 1600 gf, Ws= 10 m/s, working cond.

Table 2)

Figure 7 showing the effect of wire speed on the TWR

at various wire tensions indicates that the min. TWR


163

has been achieved at wire tension of 1600 gf, wire

speed of 12.5 m/s and at water pressure of 2.5 MPa. In

Figs. (7‐9), it has also been observed that the increase

of wire speed, wire tension and water pressure leads

to the reduction of the TWR. This result is due to the

improvement of the flushing conditions and the

stability of the process. After wire speed, wire tension

and water pressure exceeded 10 m/s, 2000 gf and 2.5

MPa, respectively, it has been noticed that TWR starts

to increase, which consequently, implies the presence

of arcing. Obviously, these conditions may lead to

vigorous sparking on the wire surface and the

propagation of wear at the wire surface. It has also

been reported that wire tension and spark pressure are

the two major causes for wire rupture (Luo, 1999).

Effect of Machining Parameters on SR

Surface finish of the machined surface plays an

important role in production, and it becomes more

desirable so as to produce a better surface when

hardened materials are machined, requiring no

subsequent polishing. Surface finish is also important

in case of tools and dies for molding as well as

drawing operations. Surface roughness and

dimensional accuracy of a spark eroded work material

depend on pulse energy, water pressure, wire tension

and wire speed. Furthermore, it could be extended to

include wire polarity, wire material, work piece

material and work piece thickness.

Figures (10‐12) show the relationship between duty

factor and both wire speed and wire tension. The

lowest value of surface roughness has been achieved

at low duty factor, lowest on‐time, lowest pulse

energy, and the higher values of wire tension and wire

speed. The value of surface roughness (Ra) has reached

1.45 μm. It should be noted that wire vibration usually

becomes more serious as wire tension decreases. This

result infers that a fine surface can be achieved by

using stable and large wire tension and higher wire

speed values. This trend is consistent with the same

trend which has been obtained by Liao et al. (1997).

Figure 10 shows that the best parametric combination

to get the optimum surface roughness (less than 2 μm)

has been achieved at low duty factor and low feeding

speed. It should be noted that the reduction of on‐time

decreases the spark energy. It is spark energy that

controls the depth of the discharge pits, which directly

relates to the surface finish. However, the off‐time is

the pause between discharge that allows the debris to

solidify and be flushed away by the dielectric prior to

the next discharge. So, it has been avoided to increase

off‐time that can dramatically increase the cutting

speed, by allowing more productive discharges per

unit of time. Furthermore, reducing off‐time can

overload the wire, causing wire breaks and instability

of the cut by not allowing enough time to evacuate the

debris before the next discharge.

1.7

2.5

3.3

4.1

4.9

0.1

0.2

0.3

0.4

0.5

1.70

2.15

2.60

3.05

3.50

SR

, µ

m


FIG. 10 EFFECT OF FEEDING SPEED ON SR AT A VARIOUS

DUTY FACTORS(T=1600 gf, Ws=10 m/s, P=2 MPa, working cond.

Table 2)

1.7

2.5

3.3

4.1

4.9

5.0

7.5

10.0

12.5

15.0

2.20

2.63

3.05

3.48

3.90

SR

, µ

m


FIG. 11 EFFECT OF FEEDING SPEED ON SR AT VARIOUS WIRE

SPEEDS(T= 1600 gf, D= 0.3, P=2 MPa, Working cond. Table 2)

1.7

2.5

3.3

4.1

4.9

800

1200

1600

2000

2400

2.20

2.55

2.90

3.25

3.60

SR

, µ

m


FIG. 12 EFFECT OF DUTY FACTOR ON SR AT VARIOUS WIRE

TENSIONS(D= 0.3, Ws= 10 m/s, P= 2 MPa, working cond. Table 2)

It is in Figure 13 anticipated that the optimum

relationship between water pressure and feeding speed


164

has been obtained at a water pressure of 2 MPa and

feeding speed about 2.5 mm/min (low pulse energy

and moderate water pressure). This phenomenon

reflects the stability of the WEDM action at these

conditions. The results obtained in the present work

about surface roughness values are similar to, or less

than, those reported by Gokler and Ozanogu (2000).

1.7

2.5

3.3

4.1

4.9

1.0

1.5

2.0

2.5

3.0

2.20

2.68

3.15

3.63

4.10

SR

, µ

m


FIG.13 EFFECT OF DUTY FACTOR ON SR AT VARIOUS WATER

PRESSURES(D= 0.3, T=1600 gf, Ws= 10 m/s, working cond. Table 2)

Investigation of Surface Modification

In the present study, the effects of the EDM process on

the surface generated were studied using SEM

microscope. Figure 14a shows that surfaces produced

at higher feed rates have a characteristically matte

appearance comparing microscopic craters. The craters

are partially superimposed, as pools of molten metal

have been ejected from a series of overlapping craters

as shown in Fig. 14a. Figure 14b shows a SEM

micrograph of EDMed surface at low feeding speed of

1.7 mm/min. It can be observed that the actual eroding

pattern and overlapping of craters in case of 4.9

mm/min (Fig. 14‐b) are much wider and deeper than

that in the case of 1.7 mm/min., as well shows an

overlapping of the re‐solidified layerswhich have been

attributed to the effect of the partially superimposed

adjacent discharges [30]. This is due to the large

amount of heat energy liberated as the feeding speed

increases.

Figure 15 (a, b) shows SEM micrographs comparison

of EDMed surface at low duty cycle of 10% and high

duty cycle of 50%. It can be observed that the surface

is relatively irregular and also shows some turbulence

at a duty cycle of 50%. The eroded surface is nearly

free of pock marks. In fact, bubbles of entrapped gas

come from the gaseous phase which has reached the

surface. Furthermore, due to the effect of the long

pulse duration, the gas bubbles grow larger, causing

the larger bock marks at the surface and great

turbulence beneath it. At higher magnification of X750,

a crack has been observed and propagated for duty

factor equal to 50%. It is also clear that spark crater is

usually associated with fusion and plastic deformation.

a ‐ Fs= 4.9 mm/min

b‐ Fs= 1.7 mm/min

FIG. 14 SEM MICROGRAPHS OF WEDM SURFACE (X750)

(D= 0.3, T= 1600 gf, P= 2 MPa, Ws= 10 m/s, working cond. Table 2)

a ‐ D = 0.1

b ‐ D = 0.5

FIG. 15 SEM MICROGRAPHS OF WEDM SURFACE (X750)(Fs = 3.3

mm/min, T=1600 gf, P= 2 MPa, Ws=10 m/s, working cond. Table 2)

Figure 16 (a, b) shows SEM micrographs of CK45

EDMed surface at lower and upper wire speed. It has

b

a

a

b


165

been observed that the increase of the wire speed

could lead to the increase of pock marks on surfaces

machined at 15 m/s. On the other hand, the EDMed

surface at 5 m/s produced shallow craters and less re‐

solidified material around the crater base.

a ‐ Ws = 5 m/s

b‐ Ws = 15 m/s


(Fs= 3.3 mm/min, D= 0.3, P= 2 MPa, T=1600 gf,working cond. Table 2)

Figure 17 shows a higher magnification for a single

spark crater on surface, to observe more details of the

crater. Tosun et al. (2003) reported that increasing the

wire speed increases both the depth and the diameter

of crater. They have estimated the contribution of this

factor by 4%. It is believed that the cause of this

condition is due to the poor flushing condition of the

water in the gap at the higher speed of the wire.

Furthermore, it could be also related to the instability

of the process at these conditions.

FIG. 17 SEM MICROGRAPH OF WEDM SURFACE (X350)

(Ws=15 m/s, Fs= 3.3 mm/min, D= 0.3, T= 1600 gf, P= 2 MPa, working

cond. Table 2)

Figure 18 (a, b) shows SEM photomicrographs of

EDMed surface in order to assess the effect of water

pressure on the resultant surface appearance. Fusion

and plastic deformation, cracking, solidified deposits,

and micro droplets were seen in the above figures.

However, the intensity of those characteristics

decreases as the water pressure increases. The surface

characteristics show a significant improvement. These

results are consistent with those obtained in MRR and

SR sections. This result can be explained by the cooling

effect of increasing the flushing pressure on the wire

surface. The higher heat transfer rate from the crater

zone and the body results in smaller molten craters.

a ‐ P=1 MPa

b ‐ P= 3 MPa


(Fs=3.3 mm/min, D=0.3, T=1600 gf, Ws=10 m/s, working cond. Table 2)

Recast layer greatly affects fatigue strength and

shortens service life of WEDM‐ed component.

Damaged surface often removed in a post machining

process, greatly increases fabricating time and cost.

The occurrence of the white layer depends on two

main factors, namely, the initial carbon content of the

workpiece and the type of dielectric fluid used. If these

two conditions are met, then the resulting thickness of

the white layer will depend upon the magnitude of the

pulse energy. Also, the hardness is still higher than

that of the base material, because of the fine grain

structure resulting from high cooling and nucleation

rates (Tsai et al., 2003).

Optimization of the Process

In the present study, three responses i.e. MRR, TWR

a

b

b

a


166

and SR have been optimized simultaneously using the

developed models, i.e. Eqs. (4‐6) based on composite

desirability optimization technique [(Castillo et al.,

1996), (Derringer and Suich, 1980), (El‐Taweel, 2009)].

In response to optimization, a measure of how the

solution has satisfied the combined goals for all

responses must be assured. The optimization is

accomplished by:

Obtaining the individual desirability (d) for

each response.

Combining the individual desirabilities to

obtain the composite desirability (D).

Maximizing the composite desirability and

identifying the optimal input variable settings.

If it is desirable to maximize a response, the individual

desirability is calculated as:

di = 0 i < Li

di = (i ‐ Li)/( Ti ‐ Li)ri Li < i < (7)

di = 1 i > Ti

If it is desirable to minimize a response, the individual


di = 0 i > Hi

di = (Hi ‐ i)/(Hi ‐ Ti)ri Ti < i < Hi (8)

di = 1 i < Ti

If it is desirable to target a response, the individual


di = (i ‐ Li))/(Ti ‐ Li)ri Li < i < Ti (9)

di = (Hi ‐ i)/(Hi ‐ Ti)ri Ti < i < Hi (10)

di = 0 i < Li

di = 0 i > Hi

where; i = predicted value of ith response, Ti = target

value for ith response, Li = lowest acceptable value for

ith response, Hi = highest acceptable value for ith

response, di = desirability for ith response, D =

composite desirability, ri = weight of desirability

function of ith response, wi = importance of ith response

and W = wi

If the importance is the same for each response, the

composite desirability is:

D = (d1d2… 　 dn)1/n = (diwi)1/W (11) where n = number of responses. To reflect the possible

difference in the importance of different responses,

where the weight wi satisfies 0< wi <1 and w1+ w2+

w3+ ……+ wn = 1　

The optimality solution is to evaluate the input

parameters in experiment range for maximizing MRR,

minimizing TWR and SR through the MINITAB‐15

software. Minitab determines optimal settings for

input variables by maximizing the composite

desirability. The importance and the weight of

desirability function of for each response is taken as

the same.

The numeric values of the optimum conditions are

presented in Tables 8. Figure 21 represents the

optimized graphs of the three responses (MRR, TWR

and SR) and also the optimization results. The vertical

lines inside the cells represent current optimal

parametric settings, and the horizontal dotted lines

represent the current response values. The value of

composite desirability (D) is taken as 0. 7466.

FIG. 19 SEM MICROGRAPH OF THE RECAST LAYER (X750)

(Ws=15 m/s, Fs= 3.3 mm/min, D= 0.3, T= 1600 gf, P= 2 MPa, working

cond. Table 2)

Position [? Theta] (Copper (Cu))

20 30 40 50 60 70 80 90

Counts

0

50

100

150

Sample No.18

a‐ P = 3 MPa

a

Base metal

Recast layer


167

b‐ P = 1 MPa

FIG. 20 X‐RAY DIFFRACTION ANALYSIS OF THE RECAST LAYER SURFACE

(Fs= 3.3 mm/min, D= 0.3 mm/min, T= 1600 gf, Ws= 10 m/s, working cond. Table 2)

FIG. 21 MULTI‐RESPONSE OPTIMIZATION RESULTS FOR MAXIMUM MRR, MINIMUM TWR AND SR

Confirmation Experiment

Conducting the experiment is a crucial final step of the

optimization results with the purpose to confirm that

the optimum conditions suggested by the optimization

process do indeed give the projected improvement, as

well to validate the accuracy of the proposed RSM

models. Conducting tests with the optimal levels

combination of the cutting parameters concluded

previously entails the confirmation experiments. The

predicted values of the confirmation tests using Eqs.

4–6 and the experimental observations at the optimum

levels of the process parameters are shown in Table 9.

It is observed that the predictions based on the

purposed regression models are quite close to the

experimental observation. The error between the

experimental results with the optimum settings and

the predicted values for MRR, TWR and SR lies within

3%, 6.6%, and 2.6%, respectively. Clearly, this confirms

the excellent reproducibility of the experimental

conclusions.

TABLE 8 CONSTRAINTS OF PARAMETERS AND OPTIMUM VALUES

Parameter Constrains Optimum value

Feeding speed (Fs), mm/min 1.7 ‐ 4.9 3.43

Duty factor (D) 0.1‐ 0.5 0.33

Wire tension (T), gf 800 ‐ 2400 1646.39

Wire speed (Ws), m/s 5 ‐15 11.28

Water pressure (P), MPa 1‐3 2.19

Hi

Lo 0.7466

D

Optimal

Cur

d = 0.77739

Minimum

SR

d = 0.89123

Minimum

TWR

d = 0.47759

Maximum

MRR

y = 2.4452

y = 0.0270

y = 0.0656

1

3

0.1

0.5

1.7

4.9 D P Fs

3.43 0.33 2.19

800

2400T

1646.3

5

15

Ws

11.28

b


168

TABLE 9 PREDICTED AND RESULTS OF THE CONFIRMATION EXPERIMENTS

Response Goal Predicted Experimental Error %

MRR, g/min Maximize 0.072 0.0698 3

TWR, g/min Minimize 0.03 0.032 6.6

SR, μm Minimize 2.30 2.36 2.6

Conclusions

The results concluded from the present work can be

summarized as follows:

1. The WEDM process has proved its adequacy to

machine CK45 steel material under acceptable

metal removal rate and fine surface finish (Ra)

less than 1.5 μm.

2. The metal removal rate generally increases

with the increase of feeding speed value, duty

factor and water pressure. Further, the effect of

wire tension and wire speed on metal removal

rate is limited.

3. Lower tool wear rate has been obtained at the

parametric combination of lower feeding speed

and lower duty factor. It has also been found

that the increase of water pressure, wire

tension and wire speed decreases tool wear.

This trend has been valid up to the generation

of arcing.

4. Fine surface quality has been obtained at lower

feeding speed, lower duty factor and higher

water pressure.

5. Microscopic investigations of the specimens

have revealed that the intensity of pulse energy

is an efficient factor in controlling surface

texture. Plastic deformation, deep craters, and

cracks have been observed at a higher feeding

speed and higher duty factor.

6. X‐ray diffraction technique has explored the

presence of some deposited wire material on a

workpiece surface under certain conditions.

These conditions exist at lower water pressure

equals 1 MPa.

7. An optimization study has also been submitted

to get maximum MRR or minimum TWR or

minimum SR under the different working

conditions.

Finally, the present work facilitates a guideline for

optimum selection of the machining parameters to the

production engineer in WEDM process of CK45 steel.

REFERENCES

Antony J. ʺDesign of experiments for engineers and

scientists.ʺ Linacre House, Jordan Hill, Oxford OX2 8DP,

2003.

Bojorquez B, Marloth R.T, Es‐Said O.S. ʺFormation of a crater

in the workpiece on an electrical discharge machine.ʺ

Engineering Failure Analysis, 9 (2002): 93‐97.

Chiang K.T, Chang F.P, Tsai D.C. ʺModeling and analysis of

the resolidified layer of SG cast iron in the EDM process

through response surface methodology.ʺ J. Mater.

Process. Technol,182 (2007): 525‐533.

Castillo E, Montgomery D, McCarville D. ʺModified

desirability functions for multiple response optimization.ʺ J.

of Quality Technl, 28 (1996): 337‐345.

Derringer G, Suich R. ʺSimultaneous optimization of several

response variables.ʺ J. of Quality Techanl, 12 (1980): 214‐

219.

El‐Taweel T.A. ʺParametric study and optimization of wire

electrical discharge machining of Al‐Cu‐TiC‐Si P/M

composite.ʺ Int. J. Machining and Machinability of

Materials (IJMMM), Vol. 1, No. 4 (2006): 380‐395.

El‐Taweel T. ʺMulti‐response optimization of EDM with Al–

Cu–Si–TiC P/M composite electrode.ʺ Int J Adv Manuf

Technol, 44 (2009): 100‐113.

Gokler M.I, Ozanozgu A.M. ʺExperimental investigation of

effects of cutting parameters on surface roughness in

WEDM process.ʺ Int. J. Machine Tools Manufac, 40

(2000): 1831‐1848.

Hasçalýk A, Caydas V. ʺExperimental study of wire

electrical discharge machining of AISI D5 tool steel.ʺ J.

Mater. Process. Technol, 148 (2004): 362‐367.

Hewidy M.S, El‐Taweel T.A, El‐Safty M.F. ʺModeling the

machining parameters of wire electrical discharge

machining of Inconel 601 using RSM.ʺ J Mater Process.

Technol 169 (2005): 328‐336.

Ho K.H., Newman S.T. ʺState of the art electrical discharge

machining (EDM).ʺ Int. J. Machine Tools Manufac, 43

(2003): 1287‐1300.

Ho K.H, Newman S.T, Rahimifard S, Allen R.D. ʺState of the

art in wire electrical discharge machining (WEDM)ʺ Int. J.

Machine Tools Manufac, 44 (2004): 1247‐1259.

Jeswani M.L. ʺDimensional analysis of tool wear in electrical

discharge machining.ʺ Wear 55 (1979): 153‐161.

Kanlayasiri K., Boonumung S. ʺAn investigation on effects of

wire EDM machining parameters on surface roughness

of newly developed DC 53 die steel.ʺ J. Mater. Processing


169

Technol, 187‐188 (2007): 26‐29.

Kinoshita N, Fukui M, Shichida H, Gamo G. ʺStudy on EDM

with wire electrode; Gap Phenomena.ʺ Annals of CIRP

25/1, (1976).

Ko‐Ta Chiang, Fu‐Ping Chang, ʺOptimization of the WEDM

process of particle‐reinforced material with multiple

performance characteristics using grey relational

analysis.ʺ J. Mater. Processing Technol, Vol. 180, Issues 1‐

3, 2006, 96‐101.

Liao Y.S, , Huang J.T, Sv H.C. ʺA study on the machining

parameters optimization of wire electrical discharge

machining.ʺ J. Mater. Process. Technol, 71(1997): 487‐493.

Luo Y.F. ʺRupture failure and mechanical strength of the

electrode wire used in wire EDM.ʺ J. Mater. Process.

Technol, 94 (1999): 208‐215.

Montgomery D.C. ʺDesign and analysis of experiments.ʺ

Wiley, New York, 2001.

Norliana M. A, Darius G. S, Fuad B. ʺA review on current

research trends in electrical discharge machining

(EDM).ʺ Int. J. Machine Tools Manufac, 47 (2007): 1214‐

1228.

Puertas C.J, Luis. ʺA study on the machining parameters

optimization of electrical discharge machining.ʺ J. Mater.

Process. Technol. 143 (2003): 521‐526.

Puri A.B, Bhattacharyya B. ʺModeling and analysis of white

layer depth in a wire‐cut EDM process through response

surface methodology.ʺ Int J Adv Manuf Technol, 25

(2005): 301‐307.

Qin G.W, Oikawaa K, Smithb G.D.W, Hao S.M. ʺWire

electric discharge machining induced titanium hydride

in Ti–46Al–2Cr alloy.ʺ Intermetallics 11 (2003): 907‐910.

Sharif S, Noordin R.M. ʺMachinability modeling in powder

mixed dielectric EDM of titanium alloy Ti‐6246.ʺ

Proceedings of the First International Conference and

Seventh AUN/SEED‐Net Fieldwise Seminar on

Manufacturing and Material Processing, Kuala Lumpur,

(2006): 133‐138.

Snoeys R, Staelens F, , Dekeyser W. ʺCurrent trends in non‐

conventional material removal processes.” Annals of

CIRP, Vol. 35/2 (1986): 467‐480.

Spedding T, Wang Z. ʺParametric optimization and surface

characterization of wire electrical discharge machining

process.ʺ Precision Engineering, 20 (1997): 5‐15.

Tsai K.M, Wang P.J. ʺSemi‐empirical model of surface finish

on electrical discharge machining.ʺ Int. J. Machine Tools

Manufac, 41 (2001): 1455‐1477.

Yahya A, Manning C.D. ʺDetermination of material removal

rate of an electro‐discharge machine using dimensional

analysis.ʺ Journal of Physics D: Applied Physics 37 (2004):

1467‐1471.

Yan M.T, Pin. ʺAccuracy improvement of wire‐EDM by real

time wire tension control.ʺ, Int. J. Machine Tools

Manufac, (2004): 807‐814.

Taha Ali El‐Taweel

He was born in El‐Gharbia, Egypt in

1956. He received his B.Sc. and the M. Sc.

degrees in in the field of non‐traditional

machining from Menoufiya University,

Egypt in 1979 and 1985 respectively. He

joined the department of production

engineering and mechanical design,

faculty of engineering, Menoufiya University, Egypt in 1979

to 1989 as a demonstrator and assistant lecturer; then he

received his Ph.D. degree in production engineering, from

Gorgian Technical University in 1992, Gorgian Republic,

USSR. From 1992, he joined the department of production

engineering and mechanical design, Menoufiya University,

as an Associate Professor in 1998. He joined as an Professor

in 2008 and head of the department of production

engineering and mechanical design, Menoufiya University

in 2011 till now. He has published many articles in

international journals and many books. His current research

interests are traditional, non‐traditional machining,

optimization teqniques and tribology of composite materials.

parametric study and optimization of wedm parameters for ck45 steel

Documents

wire tension

smalldiameter wire

wire travels

wire speed havebeen

tool wear rate twr

ahigherfeeding speed

asmaterial removal rate

wedm process