Sage Journals: Discover world-class research

Abstract

Nickel-based alloys are finding a wide range of applications due to their superior properties of maintaining hardness at elevated temperatures, low thermal conductivity and resistance to corrosion. These materials are used in aircraft, power-generation turbines, rocket engines, automobiles, nuclear power and chemical processing plants. Machining of such alloys is difficult using conventional processes. Wire-cut electrical discharge machining is one of the advanced machining processes, which can cut any electrically conductive material irrespective of its hardness. One of the major disadvantages of this process is formation of recast layer as it affects the properties of the machined surfaces. In this study, experimental investigation has been carried out to study the effect of wire-cut electrical discharge machining process parameters on micro-hardness, surface roughness and recast layer while machining Inconel-690 material. Interestingly, hardness of the machined surface was found to be lower than that of the bulk material. The micro-hardness and recast layer thickness are inversely related to the variation of process parameters. Recast layer thickness, surface roughness and hardness of the wire-cut electrical discharge machined surfaces of Inconel-690 are found to be in the range of 10–50 µm, 0.276–3.253 µm and 122–171 HV, respectively, for different conditions. The research findings and the data generated for the first time on hardness and recast layer thickness for Inconel-690 will be useful to the industry.

Keywords

Recast layer micro-hardness surface roughness Inconel-690 wire-cut electrical discharge machining

Introduction

Owing to increased usage of nickel-based alloys in various fields such as aerospace, nuclear and automobiles,¹ it is very much necessary to process them so that the parts produced can satisfy the functional requirements. Among nickel-based alloys, Inconel-690 (Table 1), a high chromium nickel-based super alloy, exhibits high hardness at elevated temperature, low thermal conductivity and corrosive resistance. Due to these properties, it is difficult to machine such materials using conventional machining processes. The wire-cut electrical discharge machining (WEDM) process can machine any electrically conductive material irrespective of its hardness² and has the ability to produce accurate, precise, complex and any intricate surfaces.^3,4 In WEDM, there is no direct contact between tool and workpiece and the material is eroded due to thermo electrical principle. When the workpiece approaches the electrode, the gap between workpiece and wire reaches a threshold and a plasma channel is formed thereby generating the sparks. A huge amount of heat is generated due to these sparks and this heat melts and evaporates the workpiece material.

Table 1.

Chemical composition (in %) of Inconel-690.

C	S	Mn	Si	Cr	Cu	Ti	Nb	Fe	Ni
0.017	0.0013	0.21	0.25	29.58	0.01	0.2	0.02	9.05	60

A portion of the melted workpiece material is removed by a dielectric circulation system. The remaining molten material will rapidly re-solidify to form a layer known as the recast layer.^5,6 This recast layer affects the mechanical properties like hardness of the material. The hardness of the WEDMed surfaces will depend on the recast layer thickness (RLT). Recast layer and heat affected zones of electrical discharge machined surfaces were studied by Rajurkar and Pandit⁷ and also developed thermal models to predict the damage layer thickness. Hascalyk and Çaydaş⁸ carried out experimental investigations on the influence of WEDM process parameters on surface integrity of AISI D5 tool steel. They found that the machine surface hardness is harder than the bulk material hardness. Gostimirovic et al.⁹ also found that, compared to bulk material, the machined surface hardness is higher in machining manganese–vanadium tool steel (ASTM A681) with WEDM. Hascalik and Çaydaş¹⁰ performed experiments on titanium alloy with different electrode materials in electrical discharge machining (EDM) and explored the influence of parameters on white layer thickness, roughness and hardness. They found that hardness of machined surfaces is increased due to Ti₂₄C₁₅ carbides formation during machining. Soni and Chakraverti¹¹ investigated experimentally the change in re-solidified layers and micro-hardness (MH) of EDMed surfaces of Die steel. Studies were carried out by Cusanelli et al.¹² on the formation of white layer and hardness of EDMed surfaces of steel. They also studied the white layer in submicron scale for phases present in white layer, micro-cracks, carbon content in the white layer and nano-hardness of the white layer. They found that the hardness of the white layer is more than that of the base material. Iqbal and Khan¹³ carried out experimental investigation on the influence of EDM process parameters on the RLT, micro-cracks and material migration in machining of stainless steel. Along with material removal rate (MRR) and tool wear rate (TWR), recast layer and MH are also studied by Jabbaripour et al.¹⁴ for titanium alloy in EDM process. Experimental investigations on the effect of WEDM parameters on the surface roughness (SR) and MH of high-strength low-alloy steel (HSLA) material were carried out by Khan et al.¹⁵ They also used gray relational analysis to optimize SR and MH simultaneously.

Researchers in the recent past used Taguchi orthogonal experimental design to study the effect of various WEDM process parameters on the responses such as MRR and SR. Regression method^16,17 and neural networks (NN) approach^18–20 were used to develop mathematical models to predict the responses with respect to process parameters and also for further optimization. Some researchers used response surface methodology (RSM) for experimental plans to study the influence of various WEDM process parameters on performance measures such as MRR, SR and Kerf. In studies, most of them used RSM itself to model the responses^21–23 and others used techniques such as back-propagation neural network (BPNN)^24,25 to model the process for comparison purpose.

Majority of the works focussed on modelling and optimization of responses such as MRR, SR and wire wear rate (WWR). Majority of investigations are reported on RLT and MH of machined surfaces of different steels and titanium alloys. Also, most of these works are carried out on die sinking EDM. Furthermore, general observation is that hardness of the machined surfaces is significantly higher than that of the bulk material. Although nickel-based super alloys are increasingly becoming popular in various industrial sectors, comprehensive WEDM studies are not reported on these alloys. This work focuses on the study of RLT and the hardness of the machined surfaces while machining Inconel-690 material. Modelling and prediction of RLT, SR and hardness of WEDMed surfaces are also attempted. The process physics involved in the influence of process parameters on these responses has been thoroughly explained. RSM was used to generate mathematical models for RLT, SR and MH for the first time for Inconel-690 material.

Experimental setup and procedure

Experimental plan is designed using a face cantered central composite design with six centre points of RSM. Design Expert 9.0 software is used for this RSM analysis. A WEDM of Electronica, India, make has been selected as a machine tool. Inconel-690, workpiece material of size 100 mm × 50 mm × 6 mm, is used. Holes of 10 mm size are produced as shown in Figure 1, with zinc-coated brass wire as electrode and deionized water as dielectric fluid. The objective of this work is to study the effect of various process parameters of WEDM such as pulse-on time, pulse-off time, peak current and servo voltage on recast layer, SR and MH of machined surfaces. Trial experiments were conducted to fix the ranges for process parameters as presented in Table 2. MH of the machined surfaces is measured using Economet VH 1 MD, Chennai Metco, India, and the recast layer has been observed using Scanning Electron Microscope of Tescan make VEGA 3 LMU model. A load of 500 g was applied for a dwell time of 10 min to measure the MH. A German make Marsurf M-400 has been used to measure the SR of the machined components.

Figure 1.

Machined circular holes on Inconel-690 workpiece.

Table 2.

Levels of process parameters.

Parameter/level	Pulse-on time (T_on), µs	Pulse-off time (T_off), µs	Peak current (I_p), A	Servo voltage (S_v), V
1	105	50	10	40
2	115	55	11	50
3	125	60	12	60

Results and analysis

Experiments are conducted as per the RSM plan as presented in Table 3 in the coded form along with the measured responses. Analysis of variance (ANOVA) has been conducted for RLT, SR and MH to study the significance of parameters and their interaction effects on the responses. Prediction models are generated using RSM and the adequacies of these models are expressed in terms of R² values. Energy-dispersive X-ray spectroscopy (EDS) analysis is also carried out to know the composition of the machined surfaces.

Table 3.

Experimental plan and results.

S. no.	T_on	T_off	I_p	S_v	RLT (µm)	SR (µm)	MH (HV)
1	2	2	2	2	31.07	0.354	139.7
2	2	3	2	2	27.86	0.357	143
3	3	1	3	3	38.4	2.585	133
4	2	2	1	1	19.14	0.336	155.3
5	1	3	3	3	27.24	0.559	144.5
6	2	2	2	3	27.66	0.373	143.4
7	3	3	1	1	23.13	0.571	159.67
8	2	2	2	2	30.25	0.371	139.9
9	2	2	2	1	29.37	0.348	141
10	1	1	1	1	16.48	0.378	157.1
11	2	2	3	1	41.98	1.842	133
12	3	3	1	3	14.55	0.377	161
13	3	1	1	1	26.66	0.595	145
14	1	3	3	1	33.14	0.669	138.2
15	2	2	2	2	28.68	0.322	142.7
16	1	1	1	3	11.4	0.371	167.4
17	2	1	2	2	31.83	0.407	139.7
18	1	1	3	3	32.14	0.447	139
19	1	3	1	1	12.84	0.332	161
20	1	3	1	3	10.06	0.276	171
21	3	2	2	2	28	0.431	141.2
22	1	2	2	2	26.24	0.317	147.1
23	2	2	2	2	29.29	0.392	142
24	1	1	3	1	36.16	0.761	134
25	3	1	3	1	49.34	3.253	122.5
26	3	3	3	3	37.91	2.786	134
27	2	2	2	2	29.22	0.422	142.1
28	3	3	3	1	47.42	3.012	128.6
29	2	2	2	1	35.5	0.533	137.9
30	3	1	1	3	16.37	0.436	158

RLT: recast layer thickness; SR: surface roughness; MH: micro-hardness.

RLT

When the work material is subjected to very high temperatures in EDM, some portion of the material is evaporated and some amount will be flushed away by the dielectric fluid. However, remaining eroded material will rapidly re-solidify to form a layer known as the recast layer. This RLT affects the mechanical properties like hardness of the material. The recast layers of the machined samples were measured using the inbuilt measuring software of scanning electron microscope (SEM). The RLT is not uniform throughout the periphery of the machined surface. Hence, RLT values at different locations on the periphery of the machined surface are measured and average value has been considered. However, in order to indicate the measurements of RLT, two specimens are chosen, one at lowest RLT corresponding to experiment number 20 (T_on = 105 µs, T_off = 60 µs, I_p = 10 A, S_v = 60 V) and another at highest RLT corresponding to experiment number 25 (T_on = 125 µs, T_off = 50 µs, I_p = 12 A, S_v = 40 V) as shown in Figure 2. The same samples are considered for EDS analysis. A pool of molten metal in the form of debris, which is not flushed away by dielectric fluid, deposits on the machined surface as shown in Figure 3. Micro-cracks and voids can also be observed on the WEDMed surfaces. Figure 3(a) and (c) corresponds to experiment number 25 (T_on = 125 µs, T_off = 50 µs, I_p = 12 A, S_v = 40 V) and Figure 3(b) and (d) corresponds to experiment number 20 (T_on = 105 µs, T_off = 60 µs, I_p = 10 A, S_v = 60 V). A pool of molten metal increases RLT and it cannot be completely eliminated during WEDM process. RLT can be minimized up to certain level by choosing the optimal machining conditions. Also, circulation of dielectric fluid with sufficient pressure will flush most of the molten metal and debris, thereby reducing the RLT. The RLT was observed in the range of 10–50 µm.

Figure 2.

Scanning electron microscope images for RLT at different conditions: (a) T_on = 105 µs, T_off = 60 µs, I_p = 10 A, S_v = 60 V and (b) T_on = 125 µs, T_off = 50 µs, I_p = 12 A, S_v = 40 V.

Figure 3.

Scanning electron microscope (SEM) images of (a) Micro voids/pits, (b) & (d) Pool of debris, and (c) Micro cracks.

It is observed from the ANOVA results of RLT (Table 4 and Figure 4(a)–(f)) that the factors influencing the RLT are pulse-on time, pulse-off time, peak current, servo voltage, interactions of pulse-on time with peak current and pulse-on time with servo voltage. The factors for which the p value is less than 0.05 will influence that model. A mathematical predictive model has been developed using ANOVA. The predictive model in coded form is given by equation (1). For this model, R², adjusted R² and predicted R² values are 98.6%, 97.6% and 94.9%, respectively

\begin{matrix} RLT = 30.08 + 4.23 T_{on} - 1.37 T_{off} + 10.73 I_{p} \\ - 3.61 S_{v} + 0.32 T_{on} T_{off} + 0.9 T_{on} I_{p} \\ - 1.35 T_{on} S_{v} - 0.23 I_{p} S_{v} - 3.39 T_{on}^{2} \\ - 0.67 T_{off}^{2} + 0.047 I_{p}^{2} + 1.07 S_{v}^{2} \end{matrix}

(1)

Table 4.

ANOVA results for recast layer thickness.

Source	SS	DF	MS	F	p
Model	2777.04	12	231.42	99.94484	<0.0001
A-T_on	321.565	1	321.565	138.8762	<0.0001
B-T_off	33.7021	1	33.7021	14.55512	0.0014
C-I_p	2071.53	1	2071.53	894.6466	<0.0001
D-S_v	234.289	1	234.289	101.1839	<0.0001
AB	1.65122	1	1.65122	0.713125	0.4101
AC	13.0682	1	13.0682	5.643858	0.0295
AD	28.9982	1	28.9982	12.52365	0.0025
CD	0.8281	1	0.8281	0.357637	0.5577
A²	29.8336	1	29.8336	12.88441	0.0023
B²	1.15728	1	1.15728	0.499802	0.4892
C²	0.00564	1	0.00564	0.002437	0.9612
D²	2.94788	1	2.94788	1.273119	0.2749
Residual	39.3631	17	2.31548
Lack of fit	35.653	12	2.97108	4.004009	0.0681
Pure error	3.71013	5	0.74203
Cor. total	2816.4	29

Note: SS: sum of squares; DF: degrees of freedom; MS: mean square.

Figure 4.

Effect of process parameters on recast layer thickness: (a) T_on (µs), (b) T_off (µs), (c) I_p (A), (d) S_v (V), (e) T_on (µs) and I_p (A), and (f) T_on (µs) and S_v (V).

Influence of process parameters on RLT is shown in Figure 4(a)–(f). It can be observed from Figure 4(a) that RLT increases as pulse-on time increases. Since pulse-on time corresponds to amount of time allowed to discharge energy in machining process, amount of energy increases with pulse-on time. Higher the energy discharged, higher will be the MRR and cutting speed. As the material removal increases, the dielectric fluid will not be able to flush away the debris effectively at higher cutting speed. Therefore, at high pulse-on time, chances of increasing the RLT on the machined surface will be high. Pulse-off time is the time interval between discharges. Behaviour of this factor is opposite to pulse-on time as shown in Figure 4(b). Peak current is the amount of current applied during machining. The discharge energy is directly proportional to the current. At the high peak current, more energy is discharged and there is continuous spark generation and the cutting speed increases. At higher cutting speed, more amount of material is removed from workpiece as well as electrode and therefore chances of forming the RLT will be high as shown in Figure 4(c). Servo voltage is the reference voltage and is used to control the wire retracts and advances. At lower value of servo voltage, the gap between wire and workpiece becomes narrow allowing more number of sparks per unit time. More amount of heat is generated causing cutting speed to increase, thereby increasing the RLT. In other words, as the servo voltage increases, the cutting speed decreases which causes RLT to decrease as shown in Figure 4(d). The interaction effects on RLT are also shown in Figure 4(e) and (f).

SR

From the ANOVA results of SR (Table 5 and Figure 5(a)–(d)), it is observed that T_on, I_p, S_v and interaction effect of T_on and I_p are influencing the SR. When T_on and I_p increase, the energy applied will also increase and more amount of heat energy will be generated during this period. At high energy levels, load on the wire causes the wire to vibrate, thereby increasing the SR as shown in Figure 5(a) and (b). At higher values of S_v, the mean gap between the wire and workpiece becomes wider which leads to decrease in number of electric sparks, thereby reducing the machining rate and vibrations thereby decreasing the SR as shown in Figure 5(c). R², adjusted R² and predicted R² values were 94.41%, 93.24% and 91.29%, respectively. The mathematical model developed for SR from ANOVA in coded form is given in equation (2)

SR = {(1.62 - 0.24 T_{on} - 0.34 I_{p} + 0.085 S_{v} - 0.11 T_{on} I_{p} - 0.36 I_{p}^{2})}^{- 1 / 2}

(2)

Table 5.

ANOVA results for surface roughness.

Source	SS	DF	MS	F	p
Model	4.326329	5	0.865266	81.02733	<0.0001
A-T_on	1.002378	1	1.002378	93.8671	<0.0001
C-I_p	2.070372	1	2.070372	193.8788	<0.0001
D-S_v	0.129862	1	0.129862	12.16086	0.0019
AC	0.18191	1	0.18191	17.03487	0.0004
C²	0.941807	1	0.941807	88.195	<0.0001
Residual	0.256289	24	0.010679
Lack of fit	0.225651	19	0.011876	1.938213	0.2390
Pure error	0.030637	5	0.006127
Cor. total	4.582618	29

Figure 5.

Effect of process parameters on surface roughness: (a) T_on (µs), (b) I_p (A), (c) S_v (V) and (d) T_on (µs) and I_p (A).

MH

Hardness of Inconel-690 base material is 221 HV. Hardness results of the machined surfaces are presented in Table 3. Interestingly, from the results, it can be observed that the hardness of the machined surfaces is found to be lower than that of the bulk material, that is, zero RLT. This is due to the fact that recast layer is composed of elements such as Cu, Zn, carbides, oxides and salts as observed from the EDS analysis. Therefore, hardness and RLT are inversely related. Although same wire was used in machining of steels in the literature, the carbides and oxides in such case are harder than the bulk material (steel). Furthermore, migration Cu and Zn in machining of steels has negligible influence on the hardness as compared to carbides and oxides. Hardness of the machined surface was found to lie in the range of 122–171 HV due to the variation of the RLT at different conditions.

From the ANOVA (Table 6), it is observed that the factors such as pulse-on time, pulse-off time, peak current and servo voltage influence the hardness of machined surfaces. A prediction model has been developed using ANOVA and is given by equation (3) in coded form. The R², adjusted R² and predicted R² values are 98.37%, 96.86% and 88.95%, respectively, for this model

\begin{matrix} MH = 141.159 - 4.241 T_{on} + 2.51 T_{off} - 12.7 I_{p} \\ + 3.74 S_{v} + 0.47 T_{on} T_{off} - 0.3 T_{on} I_{p} \\ - 0.086 T_{on} S_{v} \\ - 0.52 T_{off} I_{p} - 0.99 T_{off} S_{v} - 0.46 I_{p} S_{v} \\ + 3.06 T_{on}^{2} + 0.26 T_{off}^{2} + 3.06 I_{p}^{2} - 0.44 S_{v}^{2} \end{matrix}

(3)

Table 6.

ANOVA results for micro-hardness.

Source	SS	DF	MS	F	p
Model	3845.28	14	274.662976	64.93432235	<0.0001
A-T_on	323.682	1	323.6816056	76.52303934	<0.0001
B-T_off	113.854	1	113.85405	26.91675337	0.0001
C-I_p	2905	1	2904.998272	686.7838433	<0.0001
D-S_v	251.852	1	251.8516056	59.54138261	<0.0001
AB	3.58156	1	3.58155625	0.846731989	0.3720
AC	1.42206	1	1.42205625	0.336194781	0.5706
AD	0.11731	1	0.11730625	0.027732904	0.8700
BC	4.37856	1	4.37855625	1.035154381	0.3251
BD	15.5433	1	15.54330625	3.674663667	0.0745
CD	3.45031	1	3.45030625	0.81570258	0.3807
A²	24.3159	1	24.31590463	5.748633516	0.0300
B²	0.1799	1	0.179904625	0.04253207	0.8394
C²	24.3159	1	24.31590463	5.748633516	0.0300
D²	0.493632	1	0.493631898	0.11670176	0.7374
Residual	63.44787	15	4.229858202
Lack of fit	55.77454	10	5.577453969	3.634309711	0.0835
Pure error	7.673333	5	1.534666667
Cor. total	3908.73	29

The effect of various process parameters and their interactions on hardness is shown in Figure 6(a)–(d). It can be observed from Figure 6(a) that the hardness decreases as pulse-on time increases. At higher pulse-on time discharge energy is more, which causes the cutting speed to increase. RLT increases with increase in cutting speed, thereby decreasing the hardness of machined surface. The behaviour of pulse-off time is opposite to pulse-on time as shown in Figure 6(b). As the discharge energy is directly proportional to the current, at the high peak current, the cutting speed and melting of metal are more. Therefore, RLT increases, thereby hardness decreases as shown in Figure 6(c). At low value of servo voltage, cutting speed is high due to the reasons explained earlier. As a result, RLT increases, thereby decreasing the hardness as shown in Figure 6(d).

Figure 6.

Effect of process parameters on micro-hardness: (a) T_on (µs) (b) T_off (µs) (c) I_p (A) and (d) S_v (V).

EDS analysis

EDS coupled with SEM is used for compositional analysis and chemical characterization. Surface layers of the WEDMed specimens have been analysed using EDS. Figure 7(a) and (b) shows the EDS spectra of machined surfaces of experiment number 20 (T_on = 105 µs, T_off = 60 µs, I_p = 10 A, S_v = 60 V) and experiment number 25 (T_on = 125 µs, T_off = 50 µs, I_p = 12 A, S_v = 40 V), respectively. Although the bulk material does not have zinc and has very less percentage of copper, the WEDMed surfaces were found to contain these elements in appreciable quantities. The quantities of these elements corresponding to experiment number 20 and 25 are found to be 4.12% Zn and 0.94% Cu and also 8.26% Zn and 1.51% Cu, respectively. These elements are migrated from the wire electrode while machining. Migration of Zn and Cu leads to soften the machined surfaces. Hence, hardness of the machined surface is lesser than that of the bulk material. Hardness of the machined surfaces is presented in Table 3.

Figure 7.

EDS images at different conditions: (a) T_on = 105 µs, T_off = 60 µs, I_p = 10 A, S_v = 60 V and (b) T_on = 125 µs, T_off = 50 µs, I_p = 12 A, S_v = 40 V.

Conclusion

Nickel-based super alloys are gaining importance day by day due to their superior properties like high hardness at elevated temperatures, low thermal conductivity and high resistance to corrosion. Machining of these alloys with conventional machining processes is very difficult due to these properties. Inconel-690, one of the nickel-based alloys, can be machined using WEDM process. A recast layer is formed on the surfaces of these machined components whose hardness is found to be lower than that of the base material. However, the machined components should meet their functional requirements like retaining the hardness at high temperatures. Controlling the recast layer and associated hardness is always a challenging task especially in a machining process like WEDM due to its stochastic nature. In this work, an attempt is made to investigate on the influence of WEDM process parameters on RLT and hardness of the machined surfaces while machining Inconel-690 material. Although formation of recast layer is unavoidable, its thickness can be minimized by proper selection of machining conditions. In order to reduce the rejection rate of the machined components based on their surface quality, SR is also considered as one of the responses in this work. The important findings are listed below:

RLT of machined components increases with increase in pulse-on time and peak current, whereas it is found to decrease with increase in pulse-off time and servo voltage.

Roughness of the machined components is predominantly influenced by pulse-on time, peak current, servo voltage and interaction effect of pulse-on time and peak current.

Hardness of the machined surface is affected by pulse-on time, pulse-off time, peak current and servo voltage.

Hardness of the machined surface was found to be lower than that of the base material due to migration of Zn and Cu from wire electrode and also formation of carbides, oxides and salts while machining.

Micro-cracks and pool of molten metal were also observed on the machined surfaces.

Also, predictive models for RLT, SR and hardness are developed for Inconel-690 material with respect to process parameters for the first time. The validity of these regression models is expressed by their R² values. In the present ANOVA analysis, the R² values for all the responses were found to be greater than 94%. Therefore, the developed regression models exhibit strong correlations. In order to emphasize the validity of the models further, predicted values obtained from the regression models have also been compared against the experimental results. The average deviations were found to be less than 5%. Hence, industry can apply the developed models to predict the responses confidently.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The authors are thankful to DST-SERB of India for the financial assistance to carry out this research work through Project No. SR/FTP/ETA-10/2012.

References

Jain

VK.

Advanced machining processes. New Delhi, India: Allied Publishers Pvt. Limited, 2005.

Kuriakose

Shunmugam

MS.

Characteristics of wire-electro discharge machined Ti6Al4V surface. Mater Lett 2004; 58(17): 2231–2237.

Hewidy

El-Taweel

El-Safty

MF.

Modelling the machining parameters of wire electrical discharge machining of Inconel 601 using RSM. J Mater Process Tech 2005; 169(2): 328–336.

Ramakrishnan

Karunamoorthy

Modeling and multi-response optimization of Inconel 718 on machining of CNC WEDM process. J Mater Process Tech 2008; 207(1): 343–349.

Shahali

Yazdi

MRS

Mohammadi

et al . Optimization of surface roughness and thickness of white layer in wire electrical discharge machining of DIN 1.4542 stainless steel using micro-genetic algorithm and signal to noise ratio techniques. Proc IMechE, Part B: J Engineering Manufacture 2012; 226(5): 803–812.

Goswami

Kumar

Investigation of surface integrity, material removal rate and wire wear ratio for WEDM of Nimonic 80A alloy using GRA and Taguchi method. Eng Sci Technol Int J 2014; 17(4): 173–184.

Rajurkar

Pandit

SM.

Quantitative expressions for some aspects of surface integrity of electro discharge machined components. J Eng Ind 1984; 106(2): 171–177.

Hasçalyk

Çaydaş

Experimental study of wire electrical discharge machining of AISI D5 tool steel. J Mater Process Tech 2004; 148(3): 362–367.

Gostimirovic

Kovac

Sekulic

et al . Influence of discharge energy on machining characteristics in EDM. J Mech Sci Technol 2012; 26(1): 173–179.

10.

Hasçalık

Çaydaş

Electrical discharge machining of titanium alloy (Ti–6Al–4V). Appl Surf Sci 2007; 253(22): 9007–9016.

11.

Soni

Chakraverti

Experimental investigation on migration of material during EDM of die steel (T215 Cr12). J Mater Process Tech 1996; 56(1): 439–451.

12.

Cusanelli

Hessler-Wyser

Bobard

et al . Microstructure at submicron scale of the white layer produced by EDM technique. J Mater Process Tech 2004; 149(1): 289–295.

13.

Iqbal

AKM

Khan

. Influence of process parameters on electrical discharge machined job surface integrity. Am J Eng Appl Sci 2010; 3(2): 396–402.

14.

Jabbaripour

Sadeghi

Faridvand

et al . Investigating the effects of EDM parameters on surface integrity, MRR and TWR in machining of Ti-6Al-4V. Mach Sci Technol 2012; 16(3): 419–444.

15.

Khan

Siddiquee

et al . Investigations on the effect of wire EDM process parameters on surface integrity of HSLA: a multi-performance characteristics optimization. Prod Manuf Res 2014; 2(1): 501–518.

16.

Mahapatra

Patnaik

Optimization of wire electrical discharge machining (WEDM) process parameters using Taguchi method. Int J Adv Manuf Tech 2007; 34(9–10): 911–925.

17.

Sadeghi

Razavi

Esmaeilzadeh

et al . Optimization of cutting conditions in WEDM process using regression modelling and Tabu-search algorithm. Proc IMechE, Part B: J Engineering Manufacture 2011; 225(10): 1825–1834.

18.

Spedding

Wang

ZQ.

Parametric optimization and surface characterization of wire electrical discharge machining process. Precis Eng 1997; 20(1): 5–15.

19.

Sarkar

Mitra

Bhattacharya

Parametric optimisation of wire electrical discharge machining of γ titanium aluminide alloy through an artificial neural network model. Int J Adv Manuf Tech 2006; 27(5–6): 501–508.

20.

Saha

Singha

Pal

SK.

Soft computing models based prediction of cutting speed and surface roughness in wire electro-discharge machining of tungsten carbide cobalt composite. Int J Adv Manuf Tech 2008; 39(1): 74–84.

21.

Datta

Mahapatra

SS.

Modeling, simulation and parametric optimization of wire EDM process using response surface methodology coupled with grey-Taguchi technique. Int J Eng Sci Tech 2010; 2(5): 162–183.

22.

Shandilya

Jain

NK.

Parametric optimization during wire electrical discharge machining using response surface methodology. Procedia Eng 2012; 38: 2371–2377.

23.

Sharma

Raj

Jangra

KK.

Parameter optimization and experimental study on wire electrical discharge machining of porous Ni₄₀Ti₆₀ alloy. Proc IMechE, Part B: J Engineering Manufacture. Epub ahead of print 31 March 2015. DOI: 10.1177/0954405415577710.

24.

Tzeng

Yang

Hsieh

et al . Optimization of wire electrical discharge machining of pure tungsten using neural network and response surface methodology. Proc IMechE, Part B: J Engineering Manufacture 2011; 225: 841–852.

25.

Zhang

Guo

et al . Modeling and optimization of medium-speed WEDM process parameters for machining SKD11. Mater Manuf Process 2013; 28(10): 1124–1132.

Experimental investigations on surface integrity issues of Inconel-690 during wire-cut electrical discharge machining process