experimental investigation of electrochemical machining process using taguchi approach

8/10/2019 Experimental Investigation of Electrochemical Machining Process using Taguchi Approach

1/13

International Journal of Scientific Research in Chemical Engineering, 1(6), pp. 93-105, 2014Available online at http://www.ijsrpub.com/ijsrce

ISSN: 2345-6787; 2014; Author(s) retain the copyright of this articlehttp://dx.doi.org/10.12983/ijsrce-2014-p0093-0105

93

Full Length Research Paper

Experimental Investigation of Electrochemical Machining Process using TaguchiApproach

Sameh S. Habib

Mechanical Engineering Department, Shoubra Faculty of Engineering, Benha University, 108 Shoubra Street, Cairo, EgyptEmail: [email protected]

Received 28 August 2014; Accepted 06 November 2014

Abstract. Electrochemical Machining is one of the major alternatives to conventional methods of machining difficult to cutmaterials and generating complex contours, without inducing residual stress and tool wear. Electrochemical machining processis a metal machining technology based on electrolysis where the product is processed without both contact with the tool andthermal influence. The metal workpiece is partially machined through electricity and chemistry i.e. electrochemical until itreaches the required end shape. The shape accuracy of the end product depends on the size of the gap. In the present study, theinfluences of ECM cutting parameters such as supply voltage, tool feed rate, electrolyte concentration and current, keepingother parameters constant, on the material removal rate and surface roughness were presented. In addition Taguchi approachand analysis of variance (ANOVA) are used to optimize ECM process. Among the four process parameters, supply voltage(46%) influences highly the material removal rate, followed by tool feed rate (19%), current (6%) and the electrolyteconcentration by (3%).The contribution that have significant for surface roughness are current (53%) influences highly,followed by tool feed rate (21%), supply voltage (11.5%) and the electrolyte concentration by (0.2%). A comparative study of

material removal rate and surface roughness mathematically and experimentally basis has been carried out.

Keywords: Electrochemical machining (ECM), material removal rate, surface roughness, Taguchi approach and analysis ofvariance (ANOVA)

1. INTRODUCTION

Recent developments in different methods ofmachining have significantly increased the potentialfor widespread industrial applications of electrochemical machining (ECM) as a non-traditionalmachining process. Although an increase of material

removal arte and a high surface quality has beenachieved in earlier investigations, widespreadindustrial application of electrochemical technologyhas necessitated a better understanding of the effectsof process parameters on material removal rate andsurface quality (Swift and Booker, 1997).

Electro chemical machining processe has someunique advantages over other conventional and non-traditional machining processes but its use requiredrelatively higher initial investment cost, operatingcost, tooling cost, and maintenance costs (McGeough,1998). When using ECM process parametersoptimally, it can significantly reduce the ECMoperating, tooling, and maintenance costs and thus, itwill increase the accuracy of components producedwhich is important in some applications such as

aerospace, space, defense, nuclear areas. Therefore,choice of optimum process parameters is necessary toget the most cost-effective, efficient, and economicutilization of ECM process potentials (Benedict,1987).

Generally the optimization of any process parameters now relies on process analysis to identify

the effect of operating variables on achieving thedesired machining characteristics (Sameh, 2014 andKrishankant et al. 2012). The optimization ofelectrochemical machining process was studied bymany researchers. Senthilkumar et al. (2012), used

Nondominated Sorting Genetic Algorithm-II (NSGA-II) approach to maximize metal removal rate andminimize surface roughness. Rao et al. (2008),

presented a particle swarm optimization algorithm tofind the optimal combination of process parametersfor an electro chemical machining process. Multipleregression model and artificial neural network (ANN)model are developed as efficient approaches todetermine the optimal machining parameters in ECM(Asokan et al., 2008). Acharya et al. (1986), proposedmulti-objective optimization model for the ECM


2/13


3/13


4/13

HabibExperimental Investigation of Electrochemical Machining Process using Taguchi Approach

96

Table 5: Experimental design using L27 orthogonal arrayExp. No.

A B C D MRR S/Nratio

Ra S/Nratio

1 10 0.2 10 20 0.1782 -14.9818 5.982 -15.53692 10 0.2 15 40 0.2975 -10.5303 5.157 -14.24793 10 0.2 20 60 0.3898 -8.1832 4.348 -12.76584 10 0.6 10 40 0.4269 -7.3935 4.932 -13.86055 10 0.6 15 60 0.5776 -4.7675 4.678 -13.40126 10 0.6 20 20 0.5592 -5.0487 4.875 -13.75957 10 1.0 10 60 0.7105 -2.9687 3.624 -11.18388 10 1.0 15 20 0.5822 -4.6986 4.822 -13.66459 10 1.0 20 40 0.6297 -4.0173 4.491 -13.0469

10 20 0.2 10 40 0.5587 -5.0564 5.238 -14.383311 20 0.2 15 60 0.6732 -3.4371 4.371 -12.689512 20 0.2 20 20 0.6665 -3.5240 5.673 -15.076313 20 0.6 10 60 0.6923 -3.1941 3.922 -11.870214 20 0.6 15 20 0.5889 -4.5992 4.7 -13.4420

15 20 0.6 20 40 0.6743 -3.4229 4.422 -12.912416 20 1.0 10 20 0.6822 -3.3218 4.48 -13.025617 20 1.0 15 40 0.7111 -2.9614 4.212 -12.489818 20 1.0 20 60 0.7891 -2.0574 3.82 -11.641319 30 0.2 10 60 0.8051 -1.8830 3.541 -10.982520 30 0.2 15 20 0.7349 -2.6754 4.992 -13.965521 30 0.2 20 40 0.7559 -2.4307 4.621 -13.294722 30 0.6 10 20 0.6733 -3.4358 4.829 -13.677123 30 0.6 15 40 0.7297 -2.7371 4.313 -12.695624 30 0.6 20 60 0.7934 -2.0102 4.029 -12.103925 30 1.0 10 40 0.8289 -1.6300 4.122 -12.302226 30 1.0 15 60 0.9108 -0.8115 3.34 -10.474927 30 1.0 20 20 0.7706 -2.2634 4.624 -13.3004

Table 6: S/N response table for material removal rateParameters S/N ratio (dB) Max. Min. Rank

Level 1 Level 2 Level 3Supply voltage -6.954 -3.508 -2.209 4.746 1Tool feed rate -5.856 -4.068 -2.748 3.108 2

Electrolyte concentration -4.874 -4.135 -3.662 1.212 4Current -4.950 -4.464 -3.257 1.693 3

The mean S/N ratio = -4.22375 dB

Table 7: S/N response table for surface roughnessParameters S/N ratio (dB) Max. Min. Rank

Level 1 Level 2 Level 3

Supply voltage -13.50 -13.06 -12.53 0.96 3Tool feed rate -13.66 -13.08 -12.35 1.31 2

Electrolyte concentration -12.89 -13.01 -13.10 0.12 4Current -13.94 -13.25 -11.90 2.04 1

The mean S/N ratio = -13.429167 dB

2.3. Taguchi experiment: design and analysis

Essentially, traditional experimental design

procedures are too complicated and not easy to use. Alarge number of experimental works have to becarried out when the number of process parametersincreases. Taguchis method of experimental design isone of the widely accepted techniques for offline

quality assurance of products and processes. The stepsapplied for Taguchi optimization in this work are asfollows:

(a) Select noise and control level factors. (b) SelectTaguchi orthogonal array. (c) Analyze results;(Signal-to-noise ratio). (d) Predict optimum

performance. (e) Confirmation experiment. (f)Develop mathematical models by linear regression


5/13

International Journal of Scientific Research in Chemical Engineering, 1(6), pp. 93-105, 2014

97

analysis of the data. (g) Graph computer contour plots between responses different machining parameters.(h) Graph surface plots between responses differentmachining parameters.

In the present study, four process parametersnamely, supply voltage, tool feed rate, electrolyteconcentration and current are considered, although alarge number of factors could be considered forcontrolling the ECM process. Table 3 shows thedesign factors along with their levels. Three levels,

having equal spacing, within the operating range ofthe parameters are selected for each of the factors.

The orthogonal array chosen to set the control parameters and evaluate the process performance isthe L27, which has 27 rows corresponding to thenumber of experiments at three levels. It considersfour control factors, (A, B, C and D) to be varied inthree discrete levels as shown in Table 4. The Taguchianalysis was made using the popular softwarespecifically used for design of experiment applicationsknown as MINITAB 15.

Fig. 1: Principle of an ECM process

302010

-2

-4

-6

1.00.60.2

201510

-2

-4

-6

604020

Supply voltage (V)

ea

ofSN a

os

Tool feed rate (mm/min)

Electrolyte concentration (%) Current

Main ffe cts Plot for SN rat ios

Data Means

Signal-to-noise: Larger is better Fig. 2: Main effects plot for S/N ratios of each factor on material removal rate


6/13


98

3. RESULTS AND DISCUSSIONS

The traditional method of looking into the averages ofresults to determine the desirable factor levels doesnot account the variability of the results within thetrials. In this work the results of the ECM experiments

are investigated by using the signal-to-noise (S/N) andanalysis of variance (ANOVA). The level ofsignificance of influence of a factor or interaction offactors on a particular output response could berevealed by these methods.

Fig. 3: Main effects plot for S/N ratios of each factor on surface roughness

Fig. 4: Interaction plot for material removal rate of SN ratios between supply voltage and(a) tool feed rate, (b) electrolyte concentration and (c) current


7/13


8/13


100

lowest significant. However, the control factors thateffects surface roughness can be ranked as current arethe most significant and electrolyte concentration isthe lowest significant. Main effect plots for signal tonoise ratio of material removal rate and surfaceroughness are plotted with the help of Minitab 15software as shown in Figs 2 and 3. In the main effects

plot, if the line for a particular parameter is nearhorizontal, then the parameter has no significanteffect. On the other hand, a parameter for which theline has the highest inclination will have the mostsignificant effect. It is very much clear from the maineffects plot (Fig. 2) that parameter supply voltage (A)

is the most significant parameter for material removalrate and the parameter current (D) is the mostsignificant parameter for surface roughness, whileelectrolyte concentration (C) has some contributionfor both responses. Thus from the present analysis it isclear that the supply voltage (A) is the mostinfluencing parameter when the designer searches forincreasing material removal rate for the multiple

parameters of electro chemical machining process.However, when the designer searches for improvingsurface finish, current (D) is the most influencing

parameter.

Fig. 5: Interaction plot for surface roughness of SN ratios between supply voltage and(a) tool feed rate, (b) electrolyte concentration and (c) current

Regardless of the category of the performance

characteristics, a greater S/N value corresponds to a better performance. Therefore, the optimal level of themachining parameters is the level with the greatestvalue. Thus, the maximum point on the each graphmeans the optimum condition for each factor affectedmaterial removal rate such as A3 (30V), B3 (1.0mm/min), C3 (20%), D3 (60A) as shown in Fig. 2.The optimal process parameter combination formaximum metal removing rate is found to be athighest level of control parameters supply voltage (A),

pulse off time (B), discharge current (C) and current(D). Figure 3 indicates the optimum condition for

each factor affecting surface roughness is A1 (30V),B1 (1.0 mm/min), C3 (10%), D1 (60A).Interaction plots for material removal rate and

surface roughness of SN ratios for supply voltage with(a) tool feed rate, (b) electrolyte concentration and (c)

current are shown in Figs. 4 and 5 respectively. An

interaction plot is a simple line graph for examininginteractions between variables. The cell means on theresponse variable for each level of one factor are

plotted over all the levels of the second. The resulting profiles are parallel when there is no interaction andnonparallel when interaction is present. It can benoticed from Fig. 4 that extremely there is nointeraction between variables for material removalrate. In addition, supply voltage is best parameterwhen it used with 1.0 mm/min tool feed rate, 15%electrolyte concentration and 60A current. However,there is interaction between supply voltage and

electrolyte concentration for surface roughness asshown in Fig. 5. Supply voltage is best parameterwhen it used with 1.0 mm/min tool feed rate, 10%electrolyte concentration and 60A current.


9/13

Vc f hg..hg International Journal of Scientific Research in Chemical Engineering, 1(6), pp. 93-105, 2014

101

Fig. 6: Contour plot of material removal rate versus (a) supply voltage and tool feed rate(b) electrolyte concentration and current

Fig. 7: Contour plot of surface roughness versus (a) supply voltage and tool feed rate

(b) electrolyte concentration and current

Fig. 8: Surface plot of material removal rate versus supplyvoltage and tool feed rate

Fig. 9: Surface plot of material removal rate versus current andelectrolyte concentration


10/13


102

Fig. 10: Surface plot of surface roughness versus supplyvoltage and tool feed rate Fig. 11: Surface plot of surface roughness versus currentand electrolyte concentration

3.2. Analysis of variance

ANOVA is performed to identify the parameters andinteractions that influence the output variable. Tables8 and 9 show the ANOVA result for the materialremoval rate and surface roughness. The F-ratio,which is used to measure the significance of factor atthe desired significance level, is the ratio betweenvariance due to the effect of a factor and variance due

to error term.From the ANOVA tables it is clearly observed thatthe supply voltage (46%), feed rate of electrode(18.5%), electrolyte concentration (2.8%) and current(5.8%), have significant and physical influence on thematerial removal rate. In addition, it is observed that

the error associated to the ANOVA for materialremoval rate is 2.3%. The contribution that havesignificant for surface roughness are supply voltage(11.5%), feed rate of electrode (21.3%), electrolyteconcentration (0.2%) and current (52.9%)respectively, the error associated with surfaceroughness is 4.9%.

3.3. Mathematical models

To develop a mathematical model of the datacollected for material removal rate and surfaceroughness, the linear regression analysis of the data isdone using the software Minitab 15. The equationsobtained are:

Material removal rate (mm 3/min) = 0.040551 + 0.014728 Supply voltage (V) + 0.216014 Tool feed rate (mm/min)+ 0.005249 Electrolyte concentration (%) + 0.002516 Current (A) (4)

Surface roughness (m) = 6.55086 - 0.02499 Supply voltage (V) - 0.87875 Tool feed rate (mm/min)+ 0.00259 Electrolyte concentration (%) - 0.02601 Current (A) (5)

In general, a model fits the data well if thedifferences between the observed values and themodel's predicted values are small and unbiased. Inthe regression output for Minitab statistical software, you can find S in the Summary of Model section, rightnext to R-squared. Both statistics provide an overallmeasure of how well the model fits the data. S isknown both as the standard error of the regression andas the standard error of the estimate. R-squaredindicates that the model explains most of thevariability of the response data around its mean. Table10 represents the values of S and R-squared for thedeveloped equations 4 and 5.

3.4. Experimental Verification

After performing the statistical analysis on theexperimental data, it has been observed that there isone particular level for each factor for which theresponses are either maximum (in case of materialremoval rate) or minimum (in case of surfaceroughness and gap size). The signal to noise ratio (S/Nratio) of each responses corresponding to each factorlevel also has a maximum and a minimum value. Theoptimal parameter setting have been evaluated fromthe Figs. 2 and 3 for material removal rate and surfaceroughness. The optimal setting comes as shown in
http://www.minitab.com/en-us/products/minitab/http://www.minitab.com/en-us/products/minitab/


11/13


103

table 11. The optimal process parameters that have been identified to yield the best combination of process variables are A3B3C3D3 and A3B3C1D3 formaterial removal rate and surface roughnessrespectively. Using these optimum parameter settings,verification experiments have been carried out and theexperimental results are as shown in Table 12.

3.5. Contour and surface relationships

Figs 6 and 7 show the contour plots between materialremoval rate and surface roughness with controlling

parameters supply voltage, tool feed rate, electrolyteconcentration and current. It can be shown thatmaterial removal rate is higher at the dark area of Fig.6-a, when the supply voltage and tool feed rate values

are higher than 25V and 0.9 mm/min until 30V and1.0 mm/min respectively. However, the surfaceroughness is smaller at the most faint area of Fig. 7-b,when the current ranges from 55A to 60A andelectrolyte concentration value ranges from 10% to13%.

The surface plot between material removal rate and both supply voltage and tool feed rate can be shown inFig. 8. As the supply voltage and tool feed rateincrease, the material removal rate increase. In electrochemical machining, the material removal rate is

proportional to the supply voltage. The tool feed rate

determines the amount of current that can passthrough the work and the tool. As the tool approachesthe work piece the length of the conductive current

path decreases and the magnitude of current increases.This continues until the current is just sufficient toremove the metal at a rate corresponding to the rate oftool advance.

Fig. 9 shows the surface plot between materialremoval rate and both electrolyte concentration andcurrent. It can be shown that there is small effect ofthese parameters on material removal rate. Materialremoval rate increases with increase in current. Sincethe current density is proportional to the concentrationof electrolyte, thus the amount of material removalincreases with the electrolyte concentration.

The surface plot between surface roughness and both supply voltage and tool feed rate can be shown inFig. 10. Good surface roughness can be done whenincrease both supply voltage and tool feed rate. Fig.11 shows the surface plot between surface roughnessand both electrolyte concentration and current. As thecurrent values increases, the surface roughness valuesrelatively decreases. Hence, it was revealed thatirregular removal of material was more likely to occurat high currents values. Electrolyte concentrationseems that has no effect on surface roughness.

4. CONCLUSION

This study investigates electrochemical machining ontool steel SKD11 workpiece using Taguchi approach.The following conclusions are arrived:

(1) Among the four process parameters, supplyvoltage (46%) influences highly the material removalrate response characteristic, followed by tool feed rate(19%), current (6%) and the concentration ofelectrolyte by (3%).

(2) Current (53%) influences highly the surfaceroughness response characteristic, followed by toolfeed rate (21%), supply voltage (11.5%) and theconcentration of electrolyte by (0.2%).

(3) From the S/N curves drawn it is observed thatthe optimum level, of the factors selected, which will

produce maximum material removal rate isA3B3C3D3 and the value obtained is 0.992 mm3/min.In addition, the optimum level for surface roughnessis A3B3C1D3 and the value obtained is 3.218 m.

(4) Mathematical models are developed formaterial removal rate and surface roughness usinglinear regression approach with the help of software

program used in this work.

REFERENCES

Swift KG, Booker JD (1997). Process Selection from

Design to Manufacture. John Wiley & Sons, New York.

McGeough, JA (1998). Advanced methods ofmachining. Chapman and Hall, New York.

Benedict GF (1987). Nontraditional ManufacturingProcesses, Marcel Dekker, New York.

Sameh SH (2014). Parameter optimization ofelectrical discharge machining process by usingTaguchi approach. Journal of Engineering andTechnology Research, 6 (3): 27 42.

Krishankant, Jatin Taneja, Mohit Bector, RajeshKumar (2012). Application of Taguchi Methodfor Optimizing Turning Process by the effectsof Machining Parameters. International Journalof Engineering and Advanced Technology(IJEAT), 2 (1): 263 - 274.

Chinnamuthu Senthilkumar, Gowrishankar Ganesan,Ramanujam Karthikeyan, Optimization of ECMProcess Parameters Using NSGA-II, Journal ofMinerals and Materials Characterization andEngineering, Vol. 11, 2012, pp. 931-937.

Rao RV, Pawar PJ, Shankar R (2008). Multi-objectiveoptimization of electrochemical machining

process parameters using a particle swarmoptimization algorithm. Proceedings of theInstitution of Mechanical Engineers, Part B:Journal of Engineering Manufacture, 222(8):949-958.
http://pib.sagepub.com/search?author1=R+V+Rao&sortspec=date&submit=Submithttp://pib.sagepub.com/search?author1=P+J+Pawar&sortspec=date&submit=Submithttp://pib.sagepub.com/search?author1=R+Shankar&sortspec=date&submit=Submithttp://pib.sagepub.com/search?author1=R+Shankar&sortspec=date&submit=Submithttp://pib.sagepub.com/search?author1=P+J+Pawar&sortspec=date&submit=Submithttp://pib.sagepub.com/search?author1=R+V+Rao&sortspec=date&submit=Submit


12/13


104

Asokan P, Ravi Kumar R, Jeyapaul R, Santhi M(2008). Development of multi-objectiveoptimization models for electrochemicalmachining process. The International Journal ofAdvanced Manufacturing Technology, 39(1-2) :55-63.

Acharya BG, Jain VK, Batra JL (1986). Multi-objective optimization of the ECM process.Precision Engineering , 8 (2) : 88 96.

Chakradhar D, Venu Gopal A (2011). Multi-ObjectiveOptimization of Electrochemical machining ofEN31 steel by Grey Relational Analysis.International Journal of Modeling andOptimization, 1(2): 113 117.

Dharmalingam S, Marimuthu P, Raja K, PandyrajanR, Surendar S (2014). Optimization of Process

Parameters on MRR and Overcut in

Electrochemical Micro Machining on MetalMatrix Composites Using Grey RelationalAnalysis. International Journal of Engineeringand Technology (IJET), 6(2): 519 529.

Glen Stuart Peace (1993). Taguchi Methods: a Hands-on Approach. Addison-Wesley Publishing Co.,Inc.

Goswami R, Chaturvedi V, Chouhan R (2013).Optimization of electrochemical machining

process parameters using Taguchi approach.International Journal of Engineering Scienceand Technology (IJEST), 5(5): 999 1006.

Bisht B, Vimal J, Chaturvedi V (2013). ParametricOptimization of Electrochemical MachiningUsing Signal-To-Noise (S/N) Ratio.International Journal of Modern Engineering

Research (IJMER), 3(4): 1999-2006.
http://link.springer.com/search?facet-author=%22P.+Asokan%22http://link.springer.com/search?facet-author=%22R.+Ravi+Kumar%22http://link.springer.com/search?facet-author=%22R.+Jeyapaul%22http://link.springer.com/search?facet-author=%22M.+Santhi%22http://link.springer.com/journal/170http://link.springer.com/journal/170http://link.springer.com/journal/170/39/1/page/1http://www.sciencedirect.com/science/article/pii/0141635986900917http://www.sciencedirect.com/science/article/pii/0141635986900917http://www.sciencedirect.com/science/article/pii/0141635986900917http://www.sciencedirect.com/science/journal/01416359http://www.sciencedirect.com/science/journal/01416359/8/2http://www.sciencedirect.com/science/journal/01416359/8/2http://www.sciencedirect.com/science/journal/01416359http://www.sciencedirect.com/science/article/pii/0141635986900917http://www.sciencedirect.com/science/article/pii/0141635986900917http://www.sciencedirect.com/science/article/pii/0141635986900917http://link.springer.com/journal/170/39/1/page/1http://link.springer.com/journal/170http://link.springer.com/journal/170http://link.springer.com/search?facet-author=%22M.+Santhi%22http://link.springer.com/search?facet-author=%22R.+Jeyapaul%22http://link.springer.com/search?facet-author=%22R.+Ravi+Kumar%22http://link.springer.com/search?facet-author=%22P.+Asokan%22


13/13


105

Dr. Sameh S. Habib is an Associate Professor of Nontraditional Machining, Department of

Mechanical Engineering, Shoubra Faculty of Engineering, Benha University, Cairo, Egypt. Hereceived his PhD in Mechanical Engineering from Shoubra Faculty of Engineering, BenhaUniversity, Cairo, Egypt, 2003.

experimental investigation of electrochemical machining process using taguchi approach

Documents