Abstract

Transcription

1 5 th International & 26 th All India Manufacturing Technology, Design and Research Conference (AIMTDR 2014) December 12 th 14 th, 2014, IIT Guwahati, Assam, India Optimization of Multiple Performance Characteristics of the Electrical Discharge Machining Process on Metal Matrix Composite (Al/5%Ticp) using Grey Relational Analysis V.Chittaranjan Das 1*,N.V.V.S.Sudheer 2 1* R.V.R & J.C.College of Engineering, Guntur , Andhra Pradesh, India vemulapalli.chittaranjandas@gmail.com 2 R.V.R & J.C. College of Engineering, Guntur , Andhra Pradesh, India nvvssudheer@yahoo.co.in Abstract Electrical discharge machining (EDM) is an effective tool in shaping difficult-to-machine metal matrix composites (Al/5%TiCp) to a high degree of accuracy and surface finish. The metal matrix composites used for the tests were Aluminum composites reinforced with 5% of titanium carbide particles (TiCp) produced through the powder metallurgy route. In the present study, an L9 orthogonal array (OA), the process parameters included discharge current, open voltage, pulse ON time and duty cycle with three levels each has been selected. The material removal rate (MRR), tool wear rate (TWR) and surface roughness (SR) wereselected as the evaluation criteria. Optimal combination of process parameters is determined by the grey relational grade (GRG) obtained through GRA for multiple performance characteristics. Analysis of variance for the GRG is also implemented. The optimized process parameters simultaneously leading to higher material removal rate,a lower electrode wear ratio and better surface roughness are then verified through a confirmation experiment. The validation experiments show the machining performance of the material removal rate increases from 2.92 to 3.69 mg/min, the electrode wear ratio decrease from 0.13 to 0.10mg/min and the surface roughness decreases from 2.21 to 1.93µm, respectively. An empirical expressions ofedmed parameters relationship with themrr,twrand SR were developed which are promising in estimating the observed values. Keywords:Electrical discharge machining (EDM), Metal matrix composite, Taguchi Method, Gray relational analysis. 1 Introduction Modern industry demand materials with a specific set of properties for design and manufacturing of a product. The composite materials consisting of two or more physically and chemically distinct phases(meyers et al.(1998)).braziet.al (1997), discussedaluminium alloys reinforced with SiC and TiC particles are often used for various automotive and aerospace applications due to their extreme hardness and temperature resistant properties. However, the full potential of these metal matrix composites was hindered by the high manufacturing cost mainly because of the difficulties in machining such as turning, drilling, sawing, etc. generally results in excessive tool wear due to the very abrasive nature of the material. On-conventional machining process like electro discharge machining (EDM)was increasingly being applied for the machining of particle reinforced metal matrix composites(roux,tle(1993)). A great deal of investigation is needed to optimize the process parameters in the electro discharge machining process. Narender Singh et.al (2004) study the effect of current, pulse on time and flushing pressure on metal removal rate, tool wear rate and surface roughness in the cast Aluminium reinforced with 10% SiC particulates composites. Ramulu et al. (2001) carried out experimental investigations to study the fatigue behaviour of the machined surface on the effect of surface roughness generated by the machining process on mechanical properties of a 15 vol.%sicp/a336 Aluminium metal matrix composite. In the present work, Aluminum metal matrix composites reinforced with 5% percentage of Titanium carbide particles were prepared using the powder metallurgy route. Experiments are planned using Taguchi s L9 orthogonal array.the electrical discharge machining of Al 5%TiCp was done using electrolytic copper and the effects of various parameters, namely, current, Open Voltage, pulse on time and duty cycle on metal removal rate (MRR), Tool Wear Rate(TWR) and Surface Roughness(SR).Optimal machining parameters have been determined by the grey relational grade obtained using the grey relationalanalysisfor multiple

2 Optimization of Multiple Performance Characteristics of the Electrical Discharge Machining Process on Metal Matrix Composite (Al/5%Ticp) using Grey Relational Analysis performance characteristics namely material removal rate, tool wear rateand surface roughness. With the grey relational analysis and analysis of variance (ANOVA) of grey relational grade, the optimal combination of the process parameters has been predicted. Finally, a confirmation test is conducted to validate the optimum process parameters obtained from the analysis of parametric design. Further, the relationship of the EDMed parameters to the MRR, TWRand SR empirical expressions were developed. 2 Experimental and Analytical Procedure The metal matrix composite used for the tests were pure Aluminum(50µm particle size) composites reinforced with 5% percentage of titanium carbide particles (TiCp) produced through the powder metallurgy route. The compositewas compacted to a pressure of 300 MPa and sintered at a temperature of 500 C for two hours in a tube furnace under argon atmosphere. Basic characterization studies such as microstructure of sintered composite was conducted to evaluate the material characteristics. Figure.1 shows the microstructure of sintered composite sample of Aluminium composite 5% TiC particles. A close observation of the micrograph indicates the uniform distribution of titanium carbide particles in the aluminium matrix in Al-5%TiCp which shows slight agglomeration of TiCp particles in the matrix. Figure 1 Microstructure of Al/5% TiCp composite 2.1 Machining parameters selection A series of experiments were performed on an ELECTRONICA-make die-sinking EDM machine (PS leader ZNC). The electrolytic copper of diameter 10 mm and 50mm in height was used as an electrode. Theworkpiece was of of Al 5%TiCp (16X16X3mm). Commercial-grade kerosene was used as the dielectric fluid and the side injection of dielectric fluid was adopted. For the present experimental investigation, three different machining parameters, four factors were identified and their levels were fixed as shown in table 1.The various levels for the individual parameters such as dischargecurrent, open voltage, pulse on time andduty cycle were selected based on previousliterature Chen SL((1999). Table 1 Machining parameters and their levels Symbol Control factors Level Level Level A Current (Amp) B Open voltage (V) C Pulse duration(µs) D Duty factor (%) Machining performance evaluations The machining performance evaluated based on the response variables namely MRR,TWR and surface roughness. The MRR and TWR was calculated based on the weight difference of the workpiece and tool before and after undergoing the EDM process. A highprecision electronic weighing balance Electric Balance, Model: AX 200, Capacity: Max: 200 gms, Readability: 0.1mg, Make: Shimadzu Corporation, Japan was used for this purpose. The surface roughness measurement was then carried out using a Talysurf 10, Rank Taylor Hobson. A traverse length of 5 mm with a cut-off evaluation length of 0.8 mm was selected. The centre line average value of the surface roughness (Ra) is the most widely used surface roughness parameter in industry is selected in this study. For the efficient evaluation of the EDM process, the MRR, TWR and the surface roughness are regarded as "larger-the-better" and "smaller-the-better" characteristics, respectively, in this study. A total of 9 experiments were planned based on the Taguchi model as provided in Table 2.The machining time of each workpiece was 20minutes and each experiment was repeated three times with the average being taken. Three different responses studied were MRR (Metal Removal Rate), TWR (Tool Wear Rate) and SR (Surface roughness) are also given in Table 2. 3 GreyRelational Analysis of the Experimental Data The Taguchi method is a systematic application of design and analysis of experiments to improve product quality. In recent years, the Taguchi method has become a powerful tool for improving productivity during research and development also.antony J (2001) attempted simultaneous optimisation of multiple quality characteristics in manufacturing processes using Taguchi s quality loss function. The use of Taguchi method with the grey relational analysis can greatly 531-2

3 5 th International & 26 th All India Manufacturing Technology, Design and Research Conference (AIMTDR 2014) December 12 th 14 th, 2014, IIT Guwahati, Assam, India simplify the optimization of process parameters for multiple-performance characteristics Deng J (1982). Table 2 Plan of L 9 orthogonal array and results Expt. A B C D MRR TWR SR MRR: (mg/min),twr:mg/min; SR: µm 3.1 Data Pre-processing Data pre-processing is a process of transferring the original sequence to a comparable sequence. The experimental results are normalised in the range between zero and one. Depending on the characteristics of data sequence, there are various methodologies of data pre-processing available for the grey relational analysis Caydaş U et.al, (2008).Material removal rate (MRR) is the dominant phenomenon in EDM which decides the machinability of the material under consideration. For the "larger-the-better"characteristic like MRR, the original sequence can be normalised as follows (1) Where and are the sequence after the data preprocessing and comparability sequence respectively, k=1 for MRR; i=1,2,3...9 for experiments 1 to 9. The tool wear rate and surface roughness are also important measures of EDM performance.fortool wear rate and surface roughness, the smaller-the-better quality characteristic has been used. When the smallerthe-better is a characteristic of the original sequence, then the original sequence should be normalised as follows: (2) Where and are the sequence after the data preprocessing and comparability sequence respectively, k=2,3for TWR, SR respectively ; i=1,2,3...9 for experiments 1 to 9.All the sequences after data preprocessing using Eqs. 1 and 2 are listed in Table 3 Table 3Data processing of each performance characteristic Expt.No. MRR TWR SR Now, (k) is the deviation sequence of the reference and the comparability sequence, i.e. (k) (3) The deviation sequence 01 can be calculated using Eq. 3 as follows (1) 1 1 = (2) 2 2 = (3) 3 3 = So = (0.7187,0.2237,0.5147) Similar calculation was performed for i=1 to 9 and the results of all oi for i=1 9 are listed in Table 4.Investigating the data presented in Table 4, max (k) and min (k) are obtained and are follows (1) = (2) = (3) = 1.00 (1) = (2) = (3) = 0.00 Table 4The deviation Sequences Sequences (1) (2) (3) Expt.no Expt.no Expt.no Expt.no Expt.no Expt.no Expt.no Expt.no Expt.no Evaluation of the grey relational coefficient and the grey relational grade After data pre-processing is carried out, a grey relational coefficient can be calculated with the preprocessed sequence. It expresses the relationship between the ideal and actual normalised experimental results. The grey relational coefficient is defined as followsdeng J (1982)

4 Optimization of Multiple Performance Characteristics of the Electrical Discharge Machining Process on Metal Matrix Composite (Al/5%Ticp) using Grey Relational Analysis (4) 0i (k) is the deviation sequence of the reference sequence andthe comparability sequence, is distinguishing or identification coefficient. If all the parameters are given equal preference, is taken as 0.5. The grey relational coefficient for each experiment of the L 9 orthogonal array can be calculated using Eq. 4 and presented in Table 5.After obtaining the grey relational coefficient, the grey relational grade is computed by averaging the grey relational coefficient corresponding to each performance characteristic. The overall evaluation of the multiple performance characteristics is based on the grey relational grade, that is: ) (5) Where γ i is the grey relational grade for the i th experiment and n is the number of performance characteristics. Table 5 shows the grey relational grade for each experiment using L 9 orthogonal array. The higher grey relational grade represents that the corresponding experimental result is closer to the ideally normalised value. Experiment 3 has the best multiple-performance characteristics among nine experiments because it has the highest grey relational grade as shown in Table 5. It can be seen that in the present study optimisation of the complicated multipleperformance characteristics of EDM has been converted into optimisation of a grey relational grade. Table 5 Grey relational coefficient and grey relational grade GRC GRG Expt. TWR Ra MRR (1) (2) (3) Since the experimental design is orthogonal, it is then possible to separate out the effect of each machining parameter on the grey relational grade at different levels. For example, the mean of the grey relational grade for the discharge current at levels 1, 2 and 3 can be calculated by averaging the grey relational grade for the experiments 1 to3, 4 to 6 and 7 to 9, respectively (Table 6).The mean of the grey relational grade for each level of the other machining parameters, namely, pulseon time, duty cycle and gap voltage can be computed in the same manner. The mean of the grey relational grade for each level of the machining parameters is summarized and shown in the multiresponse performance index Table 6. Table 6 Response table for Grey relational grade Level A B C D * * * * Max-Min Rank * Levels for optimum grey relational grade, Total mean value of the grey relational grade= In addition, the total mean of the grey relational grade for the nine experiments is also calculated and listed in Table 6.Basically, the larger the grey relation grade is, the closer will be the product quality to the ideal value. Thus, larger grey relational grade is desired for optimum performance. Therefore, the optimal parameters setting for better MRR, reducedtoolwear and improved surface quality is (A1B3C2D3) as given in Table 6.Furthermore, ANOVA has been performed on grey relational grade to obtain contribution of each process parameter affecting the two process characteristics jointly.experiment 3 shows the highest grey relational grade, indicating the optimal process parameter set of A1B3C3D3 has the best multiple performance characteristics among the nine experiments. Table 7 ANOVA of the Grey relational grade Process Sum of DOF Parameters Squares Variance Percentage Current (A) Voltage (V) Pulse duration Duty factor Total Table 7 lists the grey relational grade based on the results of ANOVA analysis. It shows that the discharge current is the significant control factor affecting multiple performance characteristics with nearly 50.8% of contribution ratio and the pulse duration has 34.42% contribution

5 5 th International & 26 th All India Manufacturing Technology, Design and Research Conference (AIMTDR 2014) December 12 th 14 th, 2014, IIT Guwahati, Assam, India 4 Confirmation Tests Confirmation test has been carried out to verify the improvement of performance characteristics.optimum parameters are selected for the confirmation test as given in Table 6. The estimated grey relational grade using the optimal level of the machining parameters can be calculated using following equation. 6 whereγ m is the total mean of the grey relational grade, is the mean of the grey relational grade at the optimal level,and q is the number of the machining parameters that significantly affects multiple-performance characteristics. The obtained process parameters, which give higher grey relational grade, are presented in Table 8. The predicted MRR,TWR, surface roughness and grey relational grade for the optimal machining parameters are obtained using Eq. 6 and also presented in Table 8. Table 8 also shows the comparison of experimentally obtained MRR and surface roughness of a trial which gives maximum MRR (trial 3 of the OA) and experimentally obtained MRR and surface roughness at optimum EDM process parameters. It can be seen that the overall performance of EDM process has been improved. Table 8 Comparison between machining performance using the initial and optimal level Optimal Machining Parameters Orthogonal Experime Prediction array nt Setting A1B3C3D3 A1B3C2D3 A1B3C2 D3 MRR TWR Ra (µm) GRG Semi-empirical Expressions The EDMed parameters were correlated to the MRR, EWR and SR using the commercial software SPSS, to get semi-empirical expressions by inputting all the experimental data. 5.1 Model for material removal rate Summary Applying linear regression, which eliminates the insignificant factors one at a time, option of SPSS is used to develop the models. The parameter estimates and analysis of variance are given in table 9 MRR= A B C D(7) The R-Square value of 0.96 indicates that 96% of the variability in material removal rate was explained by the model. It can be observed that the current, voltage, pulse duration and duty factor are effecting Material removal rate. Table 9 Parameter estimates Variable PE SE t Sig Intercept A B C D PE-Parameter estimates;se-standard error Analysis of variance for the model Source df SS MS F-Value Sig Model Error Total SS-Sum of Squares;MS- Mean Square,R-Square Based on the above results, the material removal rate (MRR) modeldevelopled is given byeq Model for tool wear rate Summary Applying linear regression, which eliminates the insignificant factors one at a time, option of SPSS is used to develop the models. The parameter estimates and analysis of variance are given in table 10 Table10Parameter estimates Variable PE SE t Sig Intercept A B C D PE-Parameter estimates;se-standard error Analysis of variance for the model Source df SS MS F-Value Sig Model Error Total SS-Sum of Squares; MS- Mean Square R-Square Based on the above results, the material removal rate (MRR) model developed is given by Eq

6 Optimization of Multiple Performance Characteristics of the Electrical Discharge Machining Process on Metal Matrix Composite (Al/5%Ticp) using Grey Relational Analysis TWR= A B C D (8) The R-Square value of indicates that 97.6% of the variability in tool wear rate was explained by the model. It can be observed that the current, voltage, pulse duration and duty factor are effecting tool wear rate. 5.3 Model for Surface roughness Summary Applying linear regression, which eliminates the insignificant factors one at a time, option of SPSS is used to develop the models. The parameter estimates and analysis of variance are given in table 11 Table 11 Parameter estimates Variable PE SE t Sig Intercept A B C D PE-Parameter estimates;se-standard error Analysis of variance for the model Source df SS MS F-Value Sig Model Error Total SS-Sum of Squares;MS- Mean Square ;R-Square Based on the above results, the material removal rate (MRR) mdoeldevelopled is given by Eq.9 SR= A -001 B C D (9) The R-Square value of 0.749indicates that 74.9% of the variability in surface roughnesss was explained by the model. It can be observed that the current, voltage,pulse duration and duty factor are effecting surface roughness 6 Conclusions An application of the Taguchi method and grey relational analysis to improve the multiple performance characteristics ofmaterial removal rate, the electrode wear rate and surface roughness in the electrical discharge machining of metal matrix composite has been reported in this paper. As a result, this method greatly simplifies the optimization of complicated multiple performance characteristics. The optimal process parameters based on grey relational analysis for the EDM of metal matrix composite include 5 amp discharge current, 200 V open voltage, 200 µs pulse duration and 75% duty factor. The machining performance of the electrode wear ratio increases from 0.13 to 0.10mg/min, the material removal rate increases from 2.92 to 3.69 mg/min and the surface roughness decreases from 2.21 to 1.93µm, respectively. To conclude, as per the findings, GRA, an advanced statistical method of multi-factorial analysis, embodies rich philosophical thought of the unity of opposites, such as continuity and discontinuity, quality and quantity, statics and dynamics, etc. Empirical research on high-tech industries and systems are often constrained, since traditional statistical methods require large sets of data.. On the other hand, semi-empirical expressions have been successfully proposed to calculate the MRR, EWR and SR in the EDM process with several parameters under various machining conditions. References Antony J. (2001), Simultaneous optimisation of multiple quality characteristics in manufacturing processes using Taguchi s quality loss function,international Journal of Advanced ManufacturingTechnology,, Volume 17, pp Brazil, D., Monaghan, J., Aspinwall, D.K., and Ng EG.(1997),Wear characterization of various diamond tooling when single point turning a particle reinforced metal matrix composite,international Proceeding of the IMC-14 conference,pp Caydaş U., Hasçalık A. (2008),Use of the grey relational analysis to determine optimum laser cutting parameters with multi-performance characteristics,optics & Laser Technology, 40:987 Chen S.L., Yan B.H., Huang F.Y. (1999), Influence of kerosene and distilled water as dielectrics on the electric discharge machining characteristics of Ti-6A1 4V,Journal of Material Process Technology,vol. 87, pp Deng J. (1982), Control problems of grey systems,system and Control Letters, 5:288 Meyers, M.A., and Chawla K.K., (1998), Mechanical behaviour of materials. Prentice-Hall, Englewood Cliffs (ISBN: ). NarenderSingh, P., Raghukandan, K., Rathinasabapathi, M. and Pai B.C. (2004),,Electricdischarge machining of Al 10% SiC P as-cast metal matrix composite,journal of Material Processing Technology,pp Ramulu, M., Paul, G. and Patel, J. (2001), EDM surface effects on fatigue strength of 15 vol.% SiCp/Al metal matrix composite material,compos Struct, vol. 54, pp Roux T.Le., Wise M.L.H.,andAspinwall D.K. (1993), The effect of electro discharge machining on the surface integrity of aluminium silicon carbide metal matrix composite,j Process Adv Mater, vol. 3, pp