OPTIMIZATION OF MACHINING PARAMETERS IN ELECTRIC DISCHARGE MACHINING USING RESPONSE SURFACE METHODOLOGY

Transcription

1 OPTIMIZATION OF MACHINING PARAMETERS IN ELECTRIC DISCHARGE MACHINING USING RESPONSE SURFACE METHODOLOGY Naresh Kumar Reddy Pallela 1,Mahesh Mallampati 2, Murali Krishna.K 3 Ramakotaiah.M 4 1 Assistant professor, Mechanical Engineering Department, B.V.R.I.T-Narsapur,Telangana, India 2,3,4 Assistant professor, Mechanical Engineering Department, E.V.M-Narsaraopet,A.P., India Abstract The correct selection of manufacturing conditions is one of the most important aspects to take into consideration in the majority of manufacturing processes and, particularly, in processes related to Electrical Discharge Machining (EDM). It is used widely in machining hard metals and alloys in aerospace, automotive and die industries. Optimization is essential to increase the production rate and decrease the machining time. Since the materials, which are processed by EDM and even the process itself are very costly. Response Surface Methodology (RSM) was used to investigate relationships and parametric interactions between the four controllable variables on the material removal rate (). Experiments were conducted on a SS304 work piece with copper electrode and four process variables. The variables used were: duty-cycle, gap voltage, current and quill lift time. The experimentation plan was designed using Box-Behnken Design (BBD) of response surface method. This methodology is very effective and needs less number of experiments for optimization. The response was modeled using a response surface model based on experimental results and optimum conditions were found for machining of SS304. Kerosene has been taken as dielectric fluid. The adequacy of the models on response parameters have been established with analysis of variance (ANOVA). Key words: Electrical Discharge Machining (EDM), Material Removal Rate (), Response Surface Methodology (RSM), ANOVA, Box-Behnken Design (BBD). 1. ELECTRIC DISCHARGE MACHINING 1.1 Introduction of EDM Electric Discharge Machining (EDM) is an electrothermal non-traditional machining process, where electrical energy is used to generate electrical spark and material removal mainly occurs due to thermal energy of the spark. This energy is created between a workpiece and an electrode submerged in a dielectric fluid with a passage of electric current. The work piece and the electrode are separated by a specific small gap called spark gap. Pulsed arc discharges occur in this gap filled with an insulating medium, preferably a dielectric liquid like hydrocarbon oil or de-ionized (demineralized) water. EDM is mainly used to machine difficult-to-machine materials and high strength temperature resistant alloys. EDM can be used to machine difficult geometries in small batches or even on job-shop basis. Work material to be machined by EDM has to be electrically conductive. Schumacher [1] described the technique of material erosion employed in EDM as still arguable. This is because ignition of electrical discharges in a dirty, liquid filled gap, when applying EDM, is mostly interpreted as ion action identical as found by physical research of discharges in air or in vacuum as well as with investigations on the breakthrough strength of insulating hydrocarbon liquids. The typical set up for EDM is shown in Fig.1.1 and working principle of EDM is shown in Fig.1.2 the electrode moves toward the work-piece reducing the spark gap so that the applied voltage is high enough to ionize the dielectric fluid. Short duration discharges are generated in a liquid dielectric gap, which separates electrode and workpiece. The material is removed from tool and workpiece with the erosive effect of the electrical discharges. 1.2 Important parameters of EDM (a) Spark On-time (pulse time or Ton): The duration of time (μs) the current is allowed to flow per cycle. Material removal is directly proportional to the amount of energy applied during this on-time. This energy is really controlled by the peak current and the length of the on-time. (b) Spark Off-time (pause time or Toff ): The duration of time (μs) between the sparks. This time allows the molten material to solidify and to be wash out of the arc gap. This parameter is to affect the speed and the stability of the cut. Thus, if the off-time is too short, it will cause sparks to be unstable. 89

2 (c) Arc gap (or gap): The Arc gap is distance between the electrode and work-piece during the process of EDM. It may be called as spark gap. Spark gap can be maintained by servo system. (d) Discharge current (current Ip): The increase in machining power is obtained by increasing the average machining current. Discharge current is directly proportional to the Material removal rate. (e) Duty cycle (τ): It is the percentage of the on-time relative to the total cycle time. This parameter is calculated by dividing the on-time by the total cycle time. High melting point high melting point leads to less tool wear due to less tool material melting for the same heat load Easy manufacturability Cost cheap The followings are the different electrode materials which are used commonly: i. Graphite ii. Copper iii. Tellurium copper 99% Cu + % tellurium iv. Brass In this experiment copper has been selected as tool material. (f) Voltage (V): The voltage across the gap while current is flowing. The voltage across the electrode/work-piece before current flows is called the open gap voltage. (g) Quill lift time: For proper auto flushing we have to use quill lift time. The following parameters are considered for experimentation: Machining Parameters: 1. Duty cycle (τ) 2. Discharge Current (Ip) 3. Gap Voltage (V) 4. Quill lift time(sec) Response Parameters: 1. Material Removal Rate Constant Parameters: 1. Flushing Pressure 2. Sensitivity 3. Polarity 1.3 Tool material Tool material should be such that it would not undergo much tool wear when it is impinged by positive ions. Thus the localized temperature rise has to be less by tailoring or properly choosing its properties or even when temperature increases, there would be less melting. Further, the tool should be easily workable as intricate shaped geometric features are machined in EDM. The basic characteristics of electrode materials are: High electrical conductivity electrons are cold emitted more easily and there is less bulk electrical heating High thermal conductivity for the same heat load, the local temperature rise would be less due to faster heat conducted to the bulk of the tool and thus less tool wear Higher density for the same heat load and same tool wear by weight there would be less volume removal or tool wear and thus less dimensional loss or inaccuracy 90 Fig -1.1: Electrode used in machining 1.4 Workpiece material There are different types of work-piece materials that can be used in EDM method. Most homogeneous materials used in metalworking can be shaped by the EDM process. Work-pieces of aluminum, brass, and copper should be processed with metallic electrodes of low melting points such as copper or copper-tungsten. Workpieces of carbon and stainless steel that have high melting points should be processed with graphite electrodes. The melting points and specific gravities of the electrode material and of the work-piece should preferably be similar. In this experiment SS304 has been selected as workpiece material. Fig -1.2: SS304 workpiece before and after machining 1.5 Characteristics of EDM: i. The process can be used to machine any work material if it is electrically conductive. ii. Material removal depends on mainly thermal properties of the work material rather than its strength, hardness etc

3 iii. In EDM there is a physical tool and geometry of the tool is the positive impression of the hole or geometric feature machined iv. The tool has to be electrically conductive as well. The tool wear once again depends on the thermal properties of the tool material v. Though the local temperature rise is rather high, still due to very small pulse on time, there is not enough time for the heat to diffuse and thus almost no increase in bulk temperature takes place. Thus the heat affected zone is limited to 2 4 μm of the spark crater. vi. However rapid heating and cooling and local high temperature leads to surface hardening which may be desirable in some applications vii. Though there is a possibility of taper cut and overcut in EDM, they can be controlled and compensated. 1.6 Response surface methodology Response surface methodology (RSM) is a collection of mathematical and statistical techniques that are useful for modeling and analysis of problems in which output or response influenced by several variables and the goal is to find the correlation between the response and the variables. It can be used for optimizing the response. This method has been applied to various experimental analyses in electro discharge machining process in recent times. There are different types of designs in RSM like 2K Design, 3K Factorial Design, Central Composite Design (CCD), Box-Behnken Design etc. The main idea of RSM is to use a set of designed experiments to obtain an optimal response. Box- Behnken Design can be implemented to estimate a second-degree polynomial model. In this work RSM is utilized for determining the relations between the various EDM process parameters with the various machining criteria and exploring the effect of these process parameters on the output response i.e., the material removal rate (). In the general case, the response surface is described by an equation of the form: Where Y is the corresponding response, e.g. the produced by the various process variables of EDM and the Xi (1, 2., K) are coded levels of K quantitative process variables, the terms β0, βi, βii, βij are the second order regression coefficients. The second term under the summation sign of this polynomial equation is attributable to linear effect, whereas the third term corresponds to the higher-order effects; the fourth term of the equation includes the interactive effects of the process parameters. The objective of using RSM is not only to investigate the response over the entire factor space, but also to locate the region of interest where the response reaches its optimum or near value. By studying carefully the response surface model, the combination of factors, which gives the best response, can then be established. The response surface method is sequential process. Experiments were designed on the basis of the experimental design technique that has been proposed by G.E.P. Box and D.W. Behnken [5]. This design is very effective and needs only 27 experiments for optimizing 4 process variables. Fig -1.3: Procedure of Response Surface Methodology 1.7. Box-Behnken Design The Box-Behnken design is an independent quadratic design in that it does not contain an embedded factorial or fractional factorial design [5-6]. In this experimental design the treatment combinations are at the midpoints of edges of the design space and at the center. These designs are rotatable (or nearly rotatable) and require three levels of each factor. The advantages of Box-Behnken design include 1. It uses only three levels for each factor. 2. It is a near-rotatable design. 3. It needs fewer number of experimental runs than does a central composite design when k = 3 or 4; however when k >= 5, this advantage disappears Motivation to present work From the previous discussion it is observed that - The choice of the Process parameters of the EDM process 91

4 92 depends largely on the material combination of electrode and work piece. EDM manufactures supply these parameters for a limited amount of material combinations. When machining other work-piece materials, the user has to develop own technology to get better surface finish and good (Material Removal Rate) with low TWR (tool wear rate). Therefore, it was decided to work on optimization of process parameters of EDM. Also it was found from the literature that response surface method can give results with less no. of experiments & it provides a range of values of parameters. So, it was decided to use RSM for optimization Outline of thesis: This chapter has briefly introduced the topic. The second chapter presents review of literature and compares works of various researchers. The third chapter provides details about the experimental set up and the experiments performed. The fourth chapter presents the results of the experiments and the last chapter lists the conclusions based on the experimental work Objective of the project work: 1. To obtain optimum machining conditions in EDM for better (material removal rate) through response surface methodology. 2. To design matrix of experiments by putting the levels of factors in MINITAB software. 3. To check analysis of variance (ANOVA) for the model. 2. EXPERIMENTAL WORK In this chapter the experimental work which consist information about experimental set up, selection of work-piece, and evaluation of are discussed. 2.1 Experimental set up Experiments were performed on TOOLCRAFT die sinking spark EDM. The machine had a maximum current capacity of 50 A. The machine had the range of 1-10 μs pulse on time settings and 1-10 μs pulse off time settings. Experiments were conducted with positive polarity of electrode. The pulsed discharge current was applied in various steps in positive mode. Fig 2.1 shows the EDM machine which was used for conducting experiments. The EDM consists of following major parts: a. Dielectric reservoir, pump and circulation system. b. Power generator and control unit c. X-Y Table d. The tool holder e. The servo system to feed the tool Dielectric reservoirs pump and circulation system Dielectric reservoirs and pump are used to circulate the EDM oil for every run of the experiment and also used to filter the EDM oil Power generator and control unit The power supply control the amount of energy consumed. Firstly, it has a time control function which controls the length of time that current flows during each pulse; this is called on time. Then it controls the amount of current allowed to flow during each pulse. These pulses are of very short duration and are measured in microseconds. There is a handy rule of thumb to determine the amount of current a particular size of electrode should use: for an efficient removal rate, each square inch of electrode calls for 50 A. Low current level for large electrode will extend overall machine time unnecessarily. Conversely, too heavy a current load can damage the work piece of electrode. The control unit controls all the parameters of the machining by putting the values and maintains the work-piece tool gap X-y table They are used for the movement of the work-piece in X and Y direction. Fig 2.1 shows the EDM machine which was used for conducting experiments with work piece and tool Fig -2.1: Spark EDM The tool holder 2.2 Selection of the work material Fig -2.2: Tool holder The tool holder holds the tool with the process of machining. The tool holder holds the workpiece and tool as shown in Fig The servo system to feed the tool: The servo control unit is provided to maintain the pre determined gap. It senses the gap voltage and compares it with the present value and the different in voltage is then used to control the movement of servo motor to adjust the gap.

5 93 It is capable of machining of hard material component such as heat treated tool steels, composites, super alloys, ceramics, carbides, heat resistant steels etc. The higher carbon grades are typically used for such applications as stamping dies, metal cutting tools, etc. High strength, excellent corrosion resistance and minimized carbon content make Types 304 and 304L Stainless Steels useful for applications where welding is required. Uses include architectural mouldings and trim, kitchen equipment, welded components of chemical, textile, paper, pharmaceutical and chemical industry processing equipment. In this experiment SS304 has been selected as workpiece material. Table -2.1: Work-Piece Composition Component Wt. % Carbon 8 max. Silicon 0.75 max. Manganese 2 max. Sulphur 3 max. Phosphorous 45 max. Nickel 8-1 Chromium Nitrogen 0.10 max. Iron Table -2.2: Mechanical properties of SS304 Density 8x1000 kg/m3 Poisson s Ratio 0.29 Elastic Modulus GPa Shear Modulus 86 GPa Tensile Strength, 505 MPa Ultimate Tensile Strength, Yield 215 MPa Rockwell B (HR B) max 92 Brinell (HB) max 201 Table -2.3: Electrical properties of SS304 Electrical Resistivity 7.2e-005 ohm-cm Magnetic Permeability 08 Table -2.4: Thermal Properties of SS304 CTE, linear 20 C 17.3 µm/m- C CTE, linear 250 C 17.8 µm/m- C CTE, linear 500 C 18.7 µm/m- C Heat Capacity J/g- C Thermal Conductivity 16.2 W/m-K Melting Point C Solidus 1400 C Liquidus 1455 C Flat work-pieces of 20mm X 22mm sides and 10mm thickness were used for experimentation. Copper Electrode of 10mm diameter was used for conducting the experiments. Each work-piece was machined for creating a cavity of 1.5mm. 2.3 Control parameters and their levels Table -2.5: Input factors and their levels FACTOR LOW CENTRE HIGH LEVEL(-1) LEVEL(0) LEVEL(1) Duty cycle(τ) Current(Amp) Gap Voltage(V) Quill lift time (Sec) It was discussed earlier in the section 1.10 that Box- Behnken design uses three levels for each factor, so all the three levels for four parameters are taken by conducting trail experiments on the EDM machine as per the ranges given for it. Table -2.6: Levels of Pulse on time and Pulse off time DUTY FACTOR PULSE ON TIME PULSE OFF TIME EVALUATION OF The is expressed as the ratio of the difference of weight of the work-piece before and after machining to the machining time. The initial weights and final weights of the workpieces were measured using electronic digital balance. The Electronic balance is of SHIMADZU make having an accuracy of 01g Machining time was noted using a Stopwatch. Fig -2.2: Electronic balance is of SHIMADZU make having an accuracy of 01g

6 94 Table -2.7: Matrix of Design of Experiments Run Coded Variable Actual Variable Duty Curre Gap Quill Duty Curre Gap Quill Cycl nt (A) Volta Lift Cycl nt (A) Voltag Lift e ge Time e e (V) Time(s (%) (V) (sec) (%) ec) Based on the above table experiments were conducted for two times to check the consistency and has been evaluated. 3. RESULTS AND DISCUSSION 3.1 Response table The response table for is shown in Table 4.1 along with the input factors. It shows that the maximum occurred when (duty cycle =70%, current = 18Amp, gap voltage = 9V and quill lift time = 0.9sec) Run Duty Curren cycle t (A) (%) Table -3.1: Response Table Gap Voltag e(v) Quill lift Time (sec) 1 st Time nd Time Average

7 Main effect plots for Main effect plot is a plot of means at the various levels of each factor compared to the overall mean. It tells the average change of the output observed during a change from one level of an input to another level. Quill lift time = 0.9 sec Fig -3.3: Main effect plot of Gap voltage Fig -3.4: Main effect plot of Quill lift time Fig -3.1: Main effect plot of Duty cycle. The confirmation experiments were conducted with the above values and the obtained results were: Table -3.2: Confirmation experiment Duty cycle (%) Current (A) Gap voltage (V) Quill lift time (sec) (mg/min) Fig -3.2:Main effect plot of Current The Main effect plots of the mean of each level of the four factors suggests that the best combination is Current = 18 A Duty cycle = 70 % Gap voltage = 9 V From above table it can be noticed that by putting the values proposed by the RSM design, maximum was occurred which shows the reliability of the design. 3.3 Contour plots A contour plot is a two-dimensional graph of three measurement variables: two inputs (x1 and x2) and one response (y), where contour lines connect points on the x1 and x2 plane that have the same value for y. 95

8 Quill lift Time(sec) Gap Voltage(V) Gap Voltage(V) Quill lift Time(sec) Current(A) Quill lift Time(sec) International Journal of Engineering Technology, Management and Applied Sciences Contour Plot of (mg/min) vs Current(A), Duty Cycle(%) (mg/min) < 25 Contour Plot of (mg/min) vs Quill lift Time(sec), Current(A) (mg/min) < 25 Duty Cycle(%) Fig -3.5: Effect of Duty cycle and Current on Contour Plot of (mg/min) vs Gap Voltage(V), Duty Cycle(%) (mg/min) < Current(A) Fig -3.9: Effect of Current and Quill lift Time on Contour Plot of (mg/min) vs Quill lift Time(sec), Gap Voltage(V) (mg/min) < Duty Cycle(%) Fig -3.6: Effect of Duty cycle and Gap voltage on Contour Plot of (mg/min) vs Gap Voltage(V), Current(A) (mg/min) < Current(A) Fig -3.7: Effect of Current and Gap voltage on Contour Plot of (mg/min) vs Quill lift Time(sec), Duty Cycle(%) (mg/min) < Gap Voltage(V) Fig -3.10: Effect of Gap Voltage and Quill lift Time on Fig 3.5 shows the estimated response surface for in relation to the process parameters of current and duty cycle. It can be seen from the figure, the is maximum when current is between 12A to 18A and duty cycle is 70%. In Fig 3.6 we can notice that the response surface with a peak, i.e., when duty cycle =70% and gap voltage = 8V. From Fig 3.7 it can be noted that is increasing when current is increasing from 12A to 18A whereas gap voltage increasing from 7V to 9V. From Fig 3.8 we can notice that maximum occurs when duty cycle=70% and when quill lift time is varying from 0.9 to 0.6sec. Fig 3.9 shows increasing in the region where current is increasing from 12A and quill lift time reducing from 0.9 sec. 3.4 Model analysis of Table -3.3: ANOVA table for (before elimination) Source Sum of Squares df Mean F Square Value p-value Prob > F Duty Cycle(%) Fig -3.8: Effect of Duty cycle and Quill lift Time on Model < 001 Signifi cant A-Duty Cycle 96

9 97 B- Curren t < 001 C- Voltage D-Quill lift time AB AC AD BC BD CD A < B C D Residu al Lack of Signifi Fit cant Pure Error Cor Total The Model F-value of implies the model is significant. There is only a 1% chance that a "Model F-Value" this large could occur due to noise. Values of "Prob > F" less than 500 indicate model terms are significant. In this case A, B, A2, B2 are significant model terms. Values greater than indicatee the model terms are not significant. If there are many insignificant model terms (not counting those required to support hierarchy), model reduction may improve your model. The "Lack of Fit F- value" of implies the Lack of Fit is significant. There is only a 0.12% chancee that a Lack of Fit F- value this large could occur due to noise. Std. Dev R-Squared Adj R- Mean Squared Pred R- C.V. % Squared PRESS Adeq Precision The "Pred R-Squared" of is in reasonable agreement with the "Adj R-Squared"of "Adeq Precision" measures the signal to noise ratio. A ratio grater than 4 is desirable. Your ratio of indicates an adequate signal. This model can be used to navigate the design space. Table -3.4: ANOVA table for (after elimination) Factor Coeff. Estimat e df Std. 95% CI 95% CI VIF Error Low High Intercep t A-Duty Cycle B Current C Voltage D-Quill lift time AB AC AD BC BD CD A B C D Table -3.5: ANOVA table for fitted models Source DF Seq SS Adj. MS F P Regressio n 14 Linear 4 Square Interaction Residual Error Lack-offit Pure Error Total Final Equation in Terms of Coded Factors: = *A *B *C *D *A*B *A*C *A*D *B*C *B*D *C*D *A *B * C *D eq(3)

10 Final Equation in Terms of Actual Factors: = * Duty Cycle * Current * Voltage * Quill lift time * Duty Cycle * Current * Duty Cycle * Voltage * Duty Cycle * Quill lift time * Current * Voltage * Current * Quill lift time * Voltage * Quill lift time * Duty Cycle * Current * Voltage * Quill lift time 2 Fig : Residual Plot for The residual plot for is shown in Fig Normal probability plot indicates the data are normally distributed and the variables are influencing the response. Outliers don t exist in the data, because standardized residues are between -2 and Residuals versus fitted values indicate the variance is constant and a nonlinear relationship exists. 3. Histogram proves the data are not skewed. 4. Residuals versus order of the data indicate that there are systematic effects in the data due to time or data collection order. 4. CONCLUSION The analysis of the experimental observations highlights that the material removal rate in electrical discharge machining are greatly influenced by the various dominant process parameters considered in the present study. The material removal rate increases with an increase of current and gap voltage. However increases if both current and gap voltage increases simultaneously which has been proved in the present study. It is found that three machining parameters (duty cycle, current and gap voltage) and some of their interactions have significant effect on. An attempt has been made to estimate the optimum machining conditions to produce the best possible within the experimental constraints. Optimum machining parameter combinations (duty cycle = 70%, current = 18A, gap voltage = 9V and quill lift time = 0.9 sec) are also tested through confirmation experiments that show good concurrence with prediction of response surface method. 4.1 Scope for future work The mathematical equation which was established in Eq (3) can be checked experimentally with different values of the same parameters which were used in the present study by taking same electrode and work-piece materials for reliability. 5. REFERENCES [1] Schumacher, B.M., 2004 After 60 years of EDM, the discharge process remains still disputed, Journal of Materials Processing Technology, 149, [2] Kuldeep Ojha, K.K.Singh, 2010, Improvement in Sinking Electrical Discharge Machining: A Review, Journal of Minerals & Materials Characterization & Engineering, Vol.9, No 8, pp [3] Kunieda M, Lauwers B, Rajurkar KP, Schumacher BM, 2010, Advancing EDM through fundamental insight into the process. Annals of the CIRP 54(2): [4] Box, G.E.P.; Behnken, D.W.: Three level design for the study of quantitative variables. Technometrics. 2 (1960) [5] Box, G.E.P.; Hunter, I.S.: Multifactor experimental design for explaining response surfaces. Ann. Math. Statist. 28 (1957) [6] Kansal H.K., Singh S, Kumar P, 2005, Parametric Optimization of powder mixed electro discharge machining by response surface methodology, Journal of Materials Processing technology, pp [7] J.L. Lin, C.L.Lin, 2005, The use of grey-fuzzy logic for the optimization of the manufacturing process, Journal of material processing technology, vol.160, pp [8] T.C.Ko, Optimisation of the EDM Process Based on the Orthogonal Array with Fuzzy Logic and Grey Relational Analysis Method, International Journal of Manufacturing Technology, 19: [9] K.S.Wang, Optimization of the electrical discharge machining process based on the Taguchi method with fuzzy logics, Journal of Material processing technology, 102: [10] Yh-Fong Tzeng, 2007,Multi-objective optimisation of high-speed electrical discharge machining process 98

11 99 using a Taguchi fuzzy-based approach, Journal of Materials and Design 28 : [11] Mohan B, 2002, Effect of SiC and rotation of electrode on electric discharge machining of Al-SiC composite, Journal of Material processing technology, 124: [12] Kuntal Maji, D.K.Pratihar, 2010, Modeling of Electrical Discharge Machining Process Using Conventional Regression Analysis and Genetic Algorithms, ASM International. [13] H.C.D.Mohd Radzi, 2009, The Implementation of Taguchi Method on EDM process of Tungsten Carbide, European Journal of Scientific Research, Vol.26 No.4,pp [14] Habib S S, 2009, Study of the parameters in electrical discharge machining through response surface methodology approach, Applied Mathematical Modelling, pp [15] Pradhan M K, Biswas C K, 2008, Modelling of machining parameters for in EDM using response surface methodology, National Conference on Mechanism Science and Technology, Hamirpur. [16] Routara B C, Sahoo P, Bandyopadhyay A, 2007, Application of Response Surface Method for Modelling of Statistical Roughness Parameters on Electric Discharge Machining, International Conference on Mechanical Engineering, Dhaka, Bangladesh [17]U.Natarajan, S.H.Yang, 2010, Multiple- response optimization for micro-end milling process using response surface methodology, International Journal of Manufacturing Technology. [18] Ulas Caydas, 2008, Modeling and analysis of electrode wear and white layer thickness in die-sinking EDM process through response surface methodology, International Journal of Manufacturing Technology, 38: [19] Pradhan M K, Biswas C K, 2009, Modeling and Analysis of process parameters on Surface Roughness in EDM of AISI D2 tool steel by RSM Approach, World Academy of Science Engineering and Technology, vol 57. [20] M.M.Rahman, R.A.Bakar, 2010, Modeling of Material Removal on Machining of Ti-6Al-4V through EDM using Copper Tungsten Electrode and positive polarity, World Academy of Science Engineering and Technology, vol 71. [21] D Kangarajan, R Karthikeyan, 2008, Influence of process parameters on electric discharge machining of WC/30%Co composites, Proceedings of the Institution of Mechanical Engineers, Part B: Journal of Engineering Manufacture. [22] M M Rahman, M M Noor, Mathematical Modeling of Material Removal Rate for Ti-5Al-2.5Sn through EDM process: A Surface Response Method, Advances in Control, Chemical Engineering, Civil Engineering and Mechanical Engineering. [23] B R Sarkar, 2007, Modelling and analysis of EDMED job surface integrity, Journal of Materials Processing technology, 189: [24] T A Taweel, 2009, Multi-response optimization of EDM with Al-Cu-Si-TiC P/M composite electrode, International Journal of Manufacturing Technology, 44: [25] S Banerjee, S Dubey, 2009, Some study on electrical discharge machining of ({WC+TiC+TaC/NbC}-Co}) Cemented carbide, International Journal of Manufacturing Technology, 43: [26] Ko-Ta Chang, 2008, Modeling and analysis of the effects of machining parameters on the performance characteristics in the EDM process of Al2O3 + TiC mixed ceramic, International Journal of Manufacturing Technology, 37: [27] T Erzurumlu, 2007, Comparison of response surface model with neural network in determining the surface quality on moulded parts, Materials and Design [28] R V Rao, P J Pawar, Modelling and optimization of process parameters of wire electrical discharge machining, proceedings of IMechE Vol.223 Part B: Journal of Engineering Manufacture. [29] G.R. Nagpal, (2004), Machine tool engineering, Khana publication, Delhi. [31] McGeough, J. (1988). Advanced Methods of Machining, London. New York: Chapman and Hall, London. [32] Mathews G, (2005), Design of Experiments with Minitab, Pearson Education. [33] Minitab User Manual Release 14 MINITAB Inc, State College, PA, USA, 2003.