EDP Sciences logo
Submit your paper
Search
Menu
Home All issues Volume 7 (2020) Manufacturing Rev., 7 (2020) 20 Full HTML
Open Access
Issue
Manufacturing Rev.
Volume 7, 2020
Article Number 20
Number of page(s) 28
DOI https://doi.org/10.1051/mfreview/2020018
Published online 23 June 2020

© S. Naik et al., Published by EDP Sciences 2020

Licence Creative CommonsThis is an Open Access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

1 Introduction

With today's technologies, one of the important challenges for manufacturing industry is to provide workpieces with specified quality characteristics in the required quantity and in the fastest and most cost-effective way possible. Therefore, the improvement of the machining of newer materials and alloys becomes an absolute necessity in manufacturing process. Day-by-day, metal matrix composites (MMC) have increasingly widened their use in manufacturing sectors like aerospace, defense, manufacturing, automobile, electronic, and nuclear industries. These materials are extensively employed in different industrial applications to attain high performances due to their favorable characteristics such as lightweight, more excellent resistance to wear, high specific strength and high-temperature resistance than conventional materials [1]. Lightweight materials are mechanically consistent with lesser manufacturing costs. In the viewpoint of commercial production, the traditional machining techniques are incompetent to machine metal matrix composites for achieving the required accuracy as well as precision, intricate shapes, also time-consuming and sometimes not possible (i.e., extremely difficult to be machined). Such features on a component can be achieved only through the advanced manufacturing process. Recent past, electrical discharge machining (called, EDM) process have made an attention to be an effective technology for machining several hard and difficult-to-cut materials due to the intense heating generated along with localized electric spark that almost produces negligible cutting force including minimum stress on the machined surface for the removal of material, and high flexibility with versatility of production. Hence, the EDM process deals with as a sustainable and reasonable unconventional process for MMCs.

However, due to the complex-dynamic behavior of EDM process and its close connection with various parameters, the achievement of high responsiveness of production is essential from the techno-economical point-of-views. Several factors influencing the cutting behaviour in EDM process are such as dielectric flushing pressure [2,3], electrode materials [4–8], electric spark variables (current, discharge voltage, frequency, pulse duration) [9–11], dielectric fluids [12–15], etc. Moreover, for successful implementation of EDM technology as the substitute of conventional machining of various difficult-to-cut materials can be improved in terms of cutting efficiency, quality, cost, and productivity by considering the most appropriate and optimal process parameters, which mainly consumes precious time and effort. Under such circumstances, the effective utilization of experimental, modeling, and optimization methodology make possible a more considerable improvement in decision-making with a new technological solution that can simultaneously satisfy and control the several distinctive as well as contradictory objectives (multi-objective) in order to make the EDM process an excellent choice for machining of advanced metal matrix composite materials. Several statistical and computational approaches such as RSM [16–21], ANN [22–27] have been applied for predictive modeling and Taguchi method [28–31], GRA [32–36], desirability function approach of RSM [37–40], and PCA [41,42] have been employed for parametric as well as process optimization in electrical discharge machining. Extensive studies have been reported by employing various experimental designs, modeling techniques and optimization approaches in order to assess or investigate the machinability [43–46], to predict the various technological responses, and to control the process parameters in machining of different workpiece materials (AISI D2, D3, D6, MDN 300, AISI 316L, stainless steel, A2 tool steel, grey cast iron, Inconel 600, 601, 625, 825, 718, Ti6Al4V, Ti13Zr13Nb, nickel alloy, Al7075, Al6061, Al6063 alloy, Al-SiC MMC, Si3N4-TiN MMC, Al-Mg2Si, WC, polycrystalline diamond). Table 1 overviews the extensive studies carried out to analyze, predict and control the machining performances in electrical discharge machining of various workpiece materials using different electrode materials under different machining conditions as well as various dielectric mediums.

On the basis of literature overview up to now, it is noticed that no credible studies were performed on machining of Al-22%SiC based MMCs by EDM process. Moreover, under high temperature and sharp cooling conditions, the melting metal particles re-solidify to form a thick recast layer on the surface of the hole. The blockage of the machining gap also leads to an increase in the concentration of contaminants in the working fluid. The poor decontamination of the working fluid may result in extra spark and secondary discharges, and these abnormal discharges cause instable machining, reducing the machining accuracy, surface quality, and machining efficiency. Moreover, because of the poor decontamination of the narrow gap and the high residual Joule heat, decreasing the quality of the machined surface. Hence, the removal of machining by-products is essential for achieving high machining accuracy and surface quality in EDM. Albeit many studies have been carried out in the past and reported in literatures on material removal rate, tool wear rate, and surface roughness as the center of attention towards assessing machining performance: aspects of machined surface morphology and machined hole characteristics have not been emphasized intensively. It was most solicited from an industrial standpoint to process in EDM with the possibility of achieving the good surface quality as well as hole quality. Though the implementation of RSM and desirability function approach was existent in the literatures, until now no systematic study has been reported to predict as well as to control surface quality and dimensional accuracy of the machined hole in the absence of MRR, EWR. Moreover, to the best of the author's knowledge, previous studies did not consider any economic analysis solution enabling cost-effective manufacturing using EDM, which finds the scope for researches from techno-economical point-of-view. In order to fill up the existing research gap, the process work addresses the workpiece surface finish and hole overcut during electrical discharge machining of newly engineered metal matrix composite (MMC) of aluminum reinforced with 22 wt.% of SiC particles with brass electrode. A methodology for predictive modelling and multi-response optimization has been developed by employing response surface methodology (RSM) and desirability function approach (DFA). Additionally, an out of the paradigm investigation of EDM on Al-SiC MMC has been made to analyze how the discharge current affects the hole quality as well as surface finish of the machined component. Finally, a unique economic analysis has been performed: (i) to rationalize the usefulness of EDM process for difficult as well as hard-to-cut materials, and (ii) to confirm the feasibility of the EDM strategy for mass production in industry. Hence, the gap between the literature and the current research is somewhat diminished. Novelty aspect, the present study will be useful as technological guidelines for practical industrial application of EDM process in automotive, aerospace, military, and electricity industries to establish a superior advantage of mechanism from the economical point of view. All of these points bring worthy investigations, contribute to the uniqueness of the current study, and make advancements towards economic manufacturing.

Table 1

Literature overview.

2 Experimental setup and procedure

In this work, conductive silicon carbide particle reinforced aluminum-based metal matrix composite (Al-SiC MMC) of a circular plate having dimensions ϕ65 mm × 5 mm (diameter and thickness, respectively) is selected as workpiece material because of its excellent-unique characteristics (lightweight, strength, hardness, stiffness, wear, and corrosion resistance) as well as due to its economic production and its widespread application in aerospace, automobiles industries. For the preparation of workpiece material, the furnace was preheated at a temperature of 400 °C. Scraps of aluminum of around 1 kg were preheated at 450 °C for 2–3 h, and 3–4 pouches of silicon carbide (having a mesh size of about 300–400 microns) powder of 10 grams each were taken and melted along with the aluminum scraps at a temperature of 800 °C in a crucible. A higher amount of aluminum and silicon carbide was added to the crucible in order to prevent material loss. The molten material was stirred vigorously and was immediately cast into a graphite mold to obtain a tapered-cylindrical ingot of around 70 mm diameter. The diffusion of aluminum silicon carbide occurs in three stages; aluminum-aluminum (Al-Al), aluminum-silicon carbide (Al-SiC), silicon carbide-silicon carbide (SiC-SiC). The grain growth rate of SiC-SiC is much faster than the other two combinations, which, if not controlled might dominate the final grain structure of the material. In order to not let this happen, a small amount of around 0.1% NaCl powder was added just before casting. A tapered cylindrical piece of Al-SiC (22.13% SiC) of fine grain structure was thus obtained. For the present experimental purpose, the ingot was cut by a wire-EDM into six pieces with the piece having an average diameter of 65 mm, and each piece having a thickness of 5 mm. The elemental constituents and particle shape characterization of Al-SiC MMC are identified (refer, Fig. 1) in a scanning electron microscope (SEM) with an embedded energy dispersive X-ray (EDS) analyzer. Commercially available brass rod with diameter of 9 mm has been considered for electrode material in machining. Bio-degradable vegetable oil (with density: 0.917 g/cm3; dynamic viscosity 48.4 g/m‑s; dielectric rigidity: 62 kV/mm; specific heat: 1.67 J/g-K) was employed as dielectric medium.

For performing a series of experiments, a high accuracy computer numerical controlled electrical discharge machine tool (make: ECOWIN, model: MIC 432CS) has been utilized with maximum working current of 60 amp manufactured by Taiwan. During electrical discharge machining of Al-SiC MMC, the measurement of surface finish of machined hole part in term of arithmetical average roughness (Ra) is measured by Surftest SJ-210 Mitutoyo roughness tester. After every successive experimental trial, the overcut (OC) of the machined hole on Al-SiC MMC material is evaluated by employing co-ordinate measuring machine (ZEISS MC850), equipped with a stylus probe accessory. For better understanding of electrical discharge machining process as well as for machinability improvement, a comprehensive investigation is performed on the morphological study of machined surface by employing scanning electron microscope (JEOL JSM-6480LV). A scheme of the experimental setup with methodologies proposed in this work, is pictorially presented in Figure 2.

In this experimental investigation, five machining parameters (discharge current, gap voltage, pulse-on-time, pulse-off-time, flushing pressure) and two major technological performance characteristics (surface roughness, and overcut) of machined part are considered as input process factors and output responses, respectively. The selection of different levels of machining parameters are considered with reference to published research work [37,46] and by inspecting the workpiece for a through-hole of acceptable quality. Moreover, preliminary trails were executed to choose the appropriate range of each input parameters. Table 2 illustrates the detailed input factors with their corresponding levels for the experiment in actual as well as coded values setting. The proposed experimental design involves the variation of five factors (DC, GV, TON, TOFF, FP) at three levels. Machining trials are completely conducted based on Box-Behnken design of experiments associated with forty-six numbers of trial runs. The experimental design layout and results of machining trials are reported in Table 3.

thumbnail Fig. 1

Microstructure of Al-22%SiC MMC: (a) SEM micrograph and (b) EDX analysis.
thumbnail Fig. 2

Schematic of experimental setup with methodologies proposed.
Table 2

Machining process parameters associated with their coded value of levels.

Table 3

Experimental plan layout and results.

3 Results and discussion

3.1 Development of predictive model using response surface methodology

In the present work, Design expert 11.0 is employed to analyze the obtained experimental results of the technological responses of machined part quality (here, surface roughness and overcut) in accordance with Box-Behnken DOEs through RSM. It is used to develop a best fitted empirical model that establishes a correlation between the machining response characteristics (surface finish of the machined component Ra, and overcut of drilled hole OC) with the given machining process parameters (DC, GV, TON, TOFF, FP). Regression equations for each response are presented by,

Ra  =  1.032  +  0.0548DC  +  0.725GV − 0.00254TON − 0.0624TOFF − 3.996FP − 0.002427DC2 − 0.1621GV2 + 0.000001TON2  +  0.000249TOFF2  +  1.875FP2 − 0.0374 DC*GV − 0.000108DC * TON  +   0.002565DC * TOFF  + 0.1312DC*FP − 0.000180GV*TON + 0.00430GV*TOFF + 0.022GV*FP  +  0.000087TON*TOFF  +  0.0051TON*FP + 0.0221TOFF*FP(1)OC = 0.020  +  0.01255DC − 0.0877GV + 0.001718TON + 0.01002TOFF + 0.001FP +  0.001055DC2 +  0.0393GV2 − 0.000001TON2 − 0.000159TOFF2 − 0.272FP2 − 0.01407DC * G V − 0.000069DC * TON  +   0.000286DC*TOFF   + 0.01258DC*FP + 0.000010GV* TON + 0.00197GV*TOFF  + 0.0850GV*FP −  0.000033TON*TOFF  +  0.000548TON*FP − 0.00841TOFF*FP(2)

The results obtained for the surface roughness (Ra) and overcut (OC) from machining experiment were analysed by employing ANOVA. Analysis of variance is a statistical tool, used to illustrate the validation of obtained experimental result. It is also applied to determine the significant effect of selected machining parameters and their interaction effect upon corresponding output responses. The ANOVA table consists of degree of freedom (DoF), sum and mean of squares (SS and MS), factor contribution in total variation (Contr. %), probability (P) and F-value. The statistical tools namely P-value and F-value are employed to determine the statistical significance and adequacy of developed regression model. If the P-value for any input parameter found to be under 0.05 (i.e. for 95% confidence level), then that input parameter may be considered as having statistically significant influence on corresponding output which is desirable [47]. If the calculated F-value is lower than the standardized Fisher's value or P-value is greater than 0.05 for any factor, then that parameter considered as no effect on output. From Table 4a, it is observed that the developed model for surface roughness in electric discharge machining is significant as its P-value is desirable (i.e., under 0.05). After the execution of the analysis, it is also observed that DC, GV, TOFF, FP, DC*DC, FP*FP, DC*GV, DC*TOFF, DC*FP, TON*TOFF and TON*FP are the significant terms affecting Ra. However, among all the significant process parameters, the discharge current with 38.16% contribution has the most significant effect on surface roughness of machined component, as supported and justified by F-statistics (86.06) and P-value (<0.00001). The factors such as, pulse-on-time and interactions (DC*TON, GV*TON, GV*TOFF, GV*FP, TOFF*FP) reflect insignificant impact on Ra, as their contributions are very inconsiderable. In the same context, the ANOVA result of overcut (OC) model is presented in Table 4b, which shows the P-value is desirable (i.e., under 0.05), thereby resulting in excellent significance of regression model. It was observed that (see Tab. 4b) among the several process parameters that have been considered during machining of aluminum-silicon carbide metal matrix composite, the terms DC, GV, TON, DC*DC, TOFF*TOFF, DC*GV, DC*TON, TON*TOFF and TOFF*FP are found to be significant at 95% confidence level as well as are the influencing parameters of OC. The other parameters don't present statistical significance as well as an important role on OC because of their larger P-value, and the calculated F-value is not more than F-value. Considering the criterion of significant level to 0.05 and the insight of the F-value reveals that the individual effect of discharge current (DC) has the dominant contributor on overcut of machined hole, which explains the larger calculated F-values than standardized F-distribution value.

To avoid the misleading conclusion, different diagnostic tests such as adequacy, effectiveness and goodness-of-fit were carried out for developed regression models (Ra and OC). When the regression coefficient (R 2-value) approaches to one, the predicted model effectively fits with the actual data. For surface roughness and overcut models, the calculated R 2-values (0.889 and 0.936, respectively) are very close to unity, which depicts statistical significance as well as goodness-of fit for the proposed model. Moreover, there is a very good degree of resemblance between the experimental and predicted value, as shown in Figure 3. Thus, it is concluded that the proposed model has high effectiveness with good predictability. From the normal probability plot (see Fig. 4), it is noticed that all the terms related with the regression model of Ra and OC are statistically significant as the residuals are approaching to a straight line, which concluded that associated errors were normally distributed. With lower AD-statistic (0.725 for Ra, and 0.435 in case of OC) as well as larger P-value (0.055 for Ra, and 0.288 in case of OC) received from Anderson-Darling test, confirmed that null-hypothesis can't be rejected. Finally, it is concluded that, the proposed predictive models for surface roughness and overcut using RSM is efficient, statistically significant, adequate and also probabilistically validate as it has low probability value (less than 0.05), higher R 2-value and larger AD-test P-value.

Table 4

ANOVA results for predictive models.

thumbnail Fig. 3

Comparison between experimental and predicted values of technological parameters: (a) surface roughness, and (b) overcut.
thumbnail Fig. 4

Normal probability plot for responses (Ra, OC).

3.2 Parametric influence on technological responses

The effect of process parameters (discharge current, gap voltage, pulse-off-time, pulse-on-time, and flushing pressure) on two major technological responses (Ra, OC) of machined part quality are graphically analyzed by three-dimensional (3D) surface plot. Figure 5a illustrates the combined effect of increasing discharge current and gap voltage on surface roughness, Ra. Higher discharge current gives rise to higher discharge spark energy as well as current density. Consequently, increasing MRR results in bigger-deeper carter marks on the machined surface and hence, poor surface finish. Almost the same effect is also observed when spark voltage is raised. Figure 6 shows the SEM micrograph of machined component having poor surface quality that explains the topographical status in terms of crater (due to the existence of different kinds of interaction among the dielectric circulations, debris, and continuous electrical discharge spark), globular modules of debris (attributed to cohesion effect in the machined surface caused by molten material, cold welding effects, and insufficient flushing of dielectric fluid in the gap), micro-voids (due to the gas produced in the discharge process), micro-crack (attributed to the overreach of induced stress over the tensile strength of the workpiece material caused by rapid heating and cooling effect), and micro-pores (due to low fracture toughness and thermal shock resistance of Al-SiC MMC). Moreover, it is observed that, with increased discharge current under machining condition (GV  =  2 V; TON  =  200 μs; TOFF  =  20 μs; FP  =  0.2 kgf/cm2) transferring more thermal energy that develops extremely high temperature at the target surface where the spark strikes cause melting followed by vaporization of MMC material and induces undesirable deep-irregular on workpiece surface, thereby resulting poor surface finish. Figure 7 shows the SEM microphotograph of machined surface under the influence of different discharge currents. It is obvious form these images that, increase of discharge energy aids in the enlarging and deepening of the crater size, which comes with other surface defects such as micro cracks. Therefore, larger discharge energy hinders the surface integrity and to some extent compromises its application capability. Increasing pulse-on-time, at higher pulse-off-time, as shown in Figure 5b, results in increase in surface roughness. This is because the energy available for material removal during a given period is shared less by a large number of sparks; hence the corresponding crater size is increased. Moreover, shorter the pulse-on-time removes very little metal, closer to the accuracy with less thermal damage to the workpiece produces better surface finish, Ra. Apart effect of flushing pressure seems to be insignificant on surface finish of the machined part, as shown in Figure 5c. However, it is advisable to keep suitable flushing pressure during electrical discharge machining to prevent stagnation of dielectric fluid and short circuit.

During EDM, the role of dielectric fluid is to flush out the molten material away from the work surface, which is under melting (or partial melting) condition. While flushing, molten material is transformed into tiny particles (droplets), which are expected to be carried away; thereby, exposing a new layer. But all the particles are not flushed out, and they adhere to the top surface of the machined workpiece due to evaporation of the dielectric fluid. That part of the molten material appearing on the topmost layer is known as re-solidified layer; also called as recast layer [48]. This layer exhibits brittleness to a large extent. Figure 8 depicted the existence of recast layer as noticed from SEM micrographs of machined Al-SiC MMC at parameters setting: GV  =  2 V; TON  =  200 μs; TOFF  = 20 μs; FP  =  0.2 kgf/cm2. It is also observed that the thickness of recast layer had an increasing trend with an increase in the discharge current.

Figures 9a and 9b illustrate the influence of discharge current and pulse-on-time on overcut (OC). The effect of above-mentioned two parameters leads to the increase in discharge energy per every individual spark, as well as cutting time along with cutting velocity, and consequently transfers more thermal energy on workpiece surface that results in more evaporation of material in comparison to sub-surface and in turn it affects increase in dimensional deviation of hole, especially radial overcut caused by side spark erosion [49]. Moreover, it is observed that, with increased discharge current under cutting condition (GV = 1 V; TON  = 200 μs; TOFF  =  20 μs; FP  =  0.4 kgf/cm2) results in increased overcut, presented in Figure 10. In fact, overcut increases with an increase in flushing pressure upto certain point then decrease, as shown in Figure 9c. Similar observation has been reported on overcut in electrical discharge machining of SS316 [50]. The overcut reduces with increase in pressure upto 0.4 kgf/cm2 can simply be illustrated as a consequence of less deposition of debris elements in the machining zone, which develops uniformity of sparking at the periphery of machined surface (i.e. wall-edge at the entrance of machined hole). Nevertheless, more increase in flushing pressure promotes increase in OC which can be explained due to a reduction in flushing effectiveness as the density of the debris particles are too high at certain points within the gap. Figure 11a shows the SEM images of cross-section view for the side wall of machined hole formed by EDM with various discharge current. It is noticed that at lower value of DC, the side wall of machined hole generated on Al-SiC MMC is recommended than that one achieved by machining with a higher discharge current. The reason is that at lower value of DC, the current density is not so much during pulse-on-time, that resulting in less electrode (tool) wear and hence, the improved hole quality. Moreover, it is evidenced that hole taper gradually increases with the rise in discharge current from 5 to 15 amp may be attributed to the fact that thermal energy per unit area is more which might have resulted in high tool wear causing deterioration of tool-tip geometry and thus, resulting geometrical deviation of hole in the form of increased hole taper. Optical micrographs (see Fig. 11b) exhibiting the size of craters formed at various discharge current is evaluated quantitively using a circle to surround the crater marks. It is observed that while executing EDM of Al-SiC MMC, increase in peak discharge current, the crater size increased and resulting in the reduced hole quality in the form of poor surface finish of the hole side wall. Thereby resulting geometric shape deformation of hole as increased overcut.

During electrical discharge machining, intense heat is produced on the workpiece surface because of the spark discharge for which material removal is caused in views of melting followed by evaporation. However, the spark not only melts the workpiece but also melts the tool electrode too. The melting of the tool is called electrode wear. Tool wear rate is considerably affected by specific heat capacity, melting point and thermal conductivity of the electrode material. In addition, dielectric medium circulation flow rate and tool geometry also partly responsible for tool wear. The microscopic view of the bottom surface as well as edge of tool electrode after EDM operation on Al-SiC MMC are presented in Figure 12. Carbon deposition is noticed both at the bottom surface and around the edge of the tool electrode. It is also observed from the optical images that, increase in discharge current results in increased tool wear (obtained at parameters setting: GV  =  1 V; TON  =200 μs; TOFF  =  20 μs; FP  =  0.4 kgf/cm2). This is so because, during electrical discharge machining, heat is mostly transferred to both the tool electrode and work surfaces. Pyrolysis of the dielectric liquid medium produces pyrolytic carbon atoms, which gets deposit on bottom as well as edge of the tool (and at the machined surface as well), developing a blackish layer of carbides. Deposition of hard and low-conductive carbide layer lowers the thermal conductivity of tool electrode as a whole. These issues promote the electrode wear. Particularly, such carbides accumulated at the tool surface serve as a heat conductive barrier. Thus, heat can only be transferred through bulk of the electrode material after this barrier is overcome. This illustrates the mechanism of tool wear, and leads to poor electrode shape retention capability.

thumbnail Fig. 5

Surface plots for illustration of machining parameters effect on surface roughness.
thumbnail Fig. 6

SEM images of poor surface quality of machined component at DC  =  15 amp, GV =  1.5 V, TON  =  200 μs, TOFF =  20 μs, and FP  =  0.6 kgf/cm2.
thumbnail Fig. 7

Influence of discharge current on radial overcut of machined hole at GV  =  2 V; TON  =  200 μs; TOFF  =  20 μs; FP  =  0.2 kgf/cm2.
thumbnail Fig. 8

Cross-sectional SEM images of machined surfaces showing an increase in recast layer thickness with discharge current.
thumbnail Fig. 9

Surface plots for illustration of machining parameters effect on overcut.
thumbnail Fig. 10

Influence of discharge current on radial overcut of machined hole at GV  =  1 V; TON  =  200 μs; TOFF  =  20 μs; FP  =  0.4 kgf/cm2.
thumbnail Fig. 11

Influence of peak discharge current on hole quality formed by EDM at GV  =  1V; TON  =  200 μs; TOFF  =  20 μs; FP  =  0.4 kgf/cm2.
thumbnail Fig. 12

Deposition of carbon layer at the bottom and edge of the tool electrode after EDM on Al-SiC MMC workpiece at GV  =  1 V; TON  =  200 μs; TOFF  =  20 μs; FP  =  0.4 kgf/cm2.

3.3 Optimization using response surface methodology

The present study includes multiple-response optimization based on the desirability function approach of RSM, to keep the surface roughness as well as overcut of machined hole to the minimum. Parameter design is an effective way to improve product quality as well as the process efficiency. Desirability function approach is a statistical based multiple response robust parameter design methodology, employed for solving the multi-objective optimization problems. The approach looks for correct combination of parameter levels that simultaneously takes the responsibility to fulfill the requirements placed on each response. The criterion for achievement of optimization result is evaluated based on overall desirability, which is a weighted geometric mean of respective desirability for the different performance characteristics, which are expressed within the range of 0–1. The response will be completely unaccepted or undesirable if the desirability value approaches to 0. Response will be most desirable or accepted only if the ideal desirability value is near or equal to 1.

For solving the parameter design problems by desirability function approach, the objective function, F(x) is specified as [51];

Overall (i.e. composite) desirability function can be stated as;(3)

Here, DF is the composite desirability function which finds the optimal setting by minimizing the F(x) (i.e. maximizes DF as it is highly desirable for optimization), di is the desirability designated for the ith targeted output, and wi is the weighting of di (considered equally important) in this study.

For a goal to minimization of output, individual desirability can be defined as; (4)

where Li and the Hi are the lowest and largest acceptable value of Y for the ith output response respectively.

The optimal solution was obtained using Design Expert 11 software. A set of thirty-six solutions were obtained, and the solution with the highest desirability (close to 1) was selected, as shown in the ramp function graph in Figure 13a. Once the optimal level of process parameters is selected, the final step to predict and verify the improvement of the performance characteristics using the optimal level of the machining parameters. The point on the graph shows the optimal values for the responses and the height obtained for each point in the graph discloses how much enviable are the optimal values obtained. These values were validated by conducting confirmation experiments. The values for the optimal parameters are derived from the points on the ramp graphs. Figure 13b shows optimization plot based on desirability function approach for technological responses, showing the optimal manufacturing conditions for electrical discharge machining of Al-SiC MMC with discharge current (DC) of 5.12 amp, gap voltage (GV) of 1.95 V, pulse-on-time (TON) of 100.02 μs, pulse-off-time (TOFF) of 12.45 μs, and flushing pressure (FP) of 0.55 kgf/cm2. Finally, the estimated optimal values of pre-cited two technological responses are 0.17032 µm for Ra, and 0.22412 mm in case of OC.

thumbnail Fig. 13

Optimization plot for surface roughness and overcut using desirability function approach.

3.4 Confirmation test

With a view to avoid the misleading conclusion, the optimum machining conditions suggested by desirability function analysis of RSM technique are validated with the results of confirmation test, which could be possible by conducting one additional experiment using the same experimental setup. A comparison between the optimal and experimental values of responses (Ra, OC) under the cutting conditions proposed by RSM is presented in Table 5. The results of RSM approach presents the suitable combination of machining parameters for optimization of surface roughness and overcut because the error percentage is lower (5.8%). Hence, the optimization results obtained by RSM approach is considered for cost analysis.

Table 5

Overview of confirmatory test results.

3.5 Cost analysis

Cost consciousness with respect to machining process is a fundamental venture in efficient manufacturing system. In order to determine the manufacturing cost for a machining operation, important criteria are selected based on convolution of shape, product accuracy and tooling process. Nowadays, with increased pressure on profitability and cost management, manufacturers decided to standardized the total cost in machining operation to ensure consistency and set cost benchmarks for future reference. Because of the large expenditures involved, it is a prime importance to analyze machining operations in order to operate with optimum economic conditions. For components produced by machining, cost estimation is kept minimum by considering the optimum cutting condition and total machining cost per part. This means appropriate selection of machining parameters will definitely affect the production rate as well as manufacturing cost. Considering the optimal cutting conditions suggested by RSM technique, cost analysis is used to perform detailed direct and indirect cost estimation in terms of total machining cost per part in electrical discharge machining process [52], as shown in Table 6. It is noticed that the total machining cost per part in electrical discharge machining is considerably lower around Rs.211.08 at optimum condition. It is interesting to note that the cost estimation of operational activities in EDM process ensures a dramatic gain in productivity and efficiency in finish machining. The cheapest solution to the total machining cost per part, longer tool life with minimized downtime calculations is obtained using brass as electrode material that justifies an economic solution to machining of Al-SiC MMC.

Table 6

Cost estimation in electrical discharge machining of Al-SiC MMC.

4 Conclusions

In the present experimental investigation, predictive modelling and multi-response optimization of EDM process is done for Al-SiC MMC with brass electrode by the combined approach of Box Behnken design (BBD) − analysis of variance (ANOVA), response surface methodology (RSM) and desirability function analysis (DFA) to analyze the surface roughness and overcut. Also, an economic analysis is performed to estimate the total machining cost per component during EDM of Al-SiC MMC. The following conclusions are obtained.

  • The contribution of discharge current is found to be the most responsible factor for the degradation of surface quality as well as finish, and achieved the roughness (Ra) in the range of 0.172–0.781µm.

  • ANOVA analysis followed by the 3D surface effect plot illustrated that, the discharge current followed by pulse-on-time are the influential parameters to control the dimensional deviation of hole diameter, i.e. overcut (OC) produced by electrical discharge machining.

  • Result showed that discharge current has the significant contribution (38.16% for Ra, 37.12% in case of OC) in degradation of surface finish as well as dimensional deviation of hole diameter, especially overcut.

  • Topological as well as morphological status of the machined Al-SiC MMC surface is observed indeed disappointing due to the presence of crater marks, uneven fusion structure, globules of debris, voids and surface microcracks. Recast layer is found at the top of the machined surface. However, intensity of surface irregularities, thickness of recast layer are observed increasing with increase in discharge current.

  • Dielectric cracking also results in deposition of hard carbide layer at the bottom of the tool electrode. Formation of such carbide layer is detrimental since it promotes rapid tool wear hence adversely affecting tool shape retention capability as well as corner size machining accuracy.

  • The predictive models for various technological parameters developed by RSM technique are found to be adequate, statistically significant and probabilistically valid because their higher R 2-value (0.889 and 0.936, respectively for Ra and OC), model probability P-value under 0.05 and larger AD-test P-value (0.055 and 0.288, respectively for Ra and OC).

  • The application of RSM's desirability function analysis for multi-response optimization presented the optimal manufacturing conditions for electrical discharge machining of Al-SiC MMC at discharge current (DC) of 5.12  amp, gap voltage (GV) of 1.95V, pulse-on-time (TON) of 100 μs, pulse-off-time (TOFF) of 12.45 μs, and flushing pressure (FP) of 0.55 kgf/cm2. The estimated optimum value of technological responses is 0.17032µm for Ra, and 0.22412mm in case of OC.

  • At optimal conditions (suggested by RSM technique), the estimated total machining cost per component of only Rs.211.08 in Indian rupees, ensures benefit from economical point of view because of reduced machining time and machine downtime.

  • The proposed multiple techniques demonstrate an effective approach towards improvement in EDM process and it can be implemented in real-time process monitoring, predictive model control and optimization during machining of different workpiece materials as well as in other machining processes via. advances in computer technology.

In terms of future work, this study can be extended to analyze the influence of tool material as well as the type of dielectric fluids to improve the surface quality and reduced overcut. Further investigations can be carried out to assess the effects of EDM process parameters on tool wear rate (TWR) and material removal rate (MRR).

Nomenclature

ф : Diameter in mm

Ra : Arithmetic average surface roughness in μm

R 2 : Coefficient of determination

Tclamp : Workpiece clamping and positioning time in minute

Tp : Electrode positioning and zeroing time in minute

Tedr : Electrode drawing out time in minute

Td : Single hole drilling time at optimum machining conditions in minute

Ttot : Total machining time per component in minute

Celcd : Cost of each electrode in rupees

Cmach/min : Cost of machine working per minute in rupees

Cop/min : Cost of machine operator per minute in rupees

Ctot : Total machining cost per component in rupees

EDM : Electrical discharge machining

Al-SiC : Aluminium-silicon carbide

MMC : Metal matrix composite

DC : Discharge current in ampere

GV : Gap voltage in volt.

TON : Pulse-on-time in μs

TOFF : Pulse-off-time in μs

FP : Flushing pressure in kgf/cm2

OC : Overcut in mm

BBD : Box-Behnken design

RSM : Response surface methodology

ANN : Artificial neural network

GA : Genetic algorithm

GRA : Grey relational analysis

ANOVA : Analysis of variance

SS : Sum of squares

MS : Mean of squares

SEM : Scanning electron microscope

PSO : Particle swarm optimization

References

  1. B. Mohan, A. Rajadurai, K.G. Satyanarayana, Electric discharge machining of Al-SiC metal matrix composites using rotary tube electrode, J. Mater. Process. Technol. 153–154 (2004) 978–985 [CrossRef] [Google Scholar]
  2. D. Kanagarajan, R. Karthikeyan, K. Palanikumar, P. Sivaraj, Influence of process parameters on electric discharge machining of WC/30%Co composites, Proc. Inst. Mech. Eng. B 222 (2008) 807–815 [CrossRef] [Google Scholar]
  3. V. Gohil, Y.M. Puri, Statistical analysis of material removal rate and surface roughness in electrical discharge turning of titanium alloy (Ti-6Al-4V), Proc. Inst. Mech. Eng. B 232 (2016) 1603–1614 [CrossRef] [Google Scholar]
  4. O. Belgassim, A. Abusaada, Investigation of the influence of EDM parameters on the overcut for AISI D3 tool steel, Proc. Inst. Mech. Eng. B 226 (2011) 365–370 [CrossRef] [Google Scholar]
  5. C.P.M. Mohanty, P.S. Satpathy, S. Mahapatra, M.R. Singh, Optimization of cryo-treated EDM variables using TOPSIS-based TLBO algorithm, Sādhanā 43 (2018) 51 [Google Scholar]
  6. C.P.S. Mohanty, S. Mahapatra, M.R. Singh, An intelligent approach to optimize the EDM process parameters using utility concept and QPSO algorithm, Eng. Sci. Technol. 20 (2017) 552–562 [Google Scholar]
  7. S.H. Lee, X.P. Li, Study of the effect of machining parameters on the machining characteristics in electrical discharge machining of tungsten carbide, Journal of Materials Processing Technology 115 (2001) 344–358 [CrossRef] [Google Scholar]
  8. M. Bhaumik, K. Maity, Effect of different tool materials during EDM performance of titanium grade 6 alloy, Eng. Sci. Technol. 21 (2018) 507–516 [Google Scholar]
  9. D. Rahul, K. Mishra, S. Datta, M. Masanta, Effects of tool electrode on EDM performance of Ti-6Al-4V, Silicon 10 (2018) 2263–2277 [CrossRef] [Google Scholar]
  10. M.H. Raza, A. Wasim, M.A. Ali, S. Hussain, M. Jahanzaib, Investigating the effects of different electrodes on Al6061-SiC-7.5 wt% during electric discharge machining, Int. J. Adv. Manufactur. Technol. 99 (2018) 3017–3034 [CrossRef] [Google Scholar]
  11. S. Kumari, S. Datta, M. Masanta, G. Nandi, P.K. Pal, Electro-discharge machining of Inconel 825 super alloy: effects of tool material and dielectric flushing, Silicon 10 (2018) 2079–2099 [CrossRef] [Google Scholar]
  12. M.B.A. Ndaliman, A. Khan, M.Y. Ali, Influence of dielectric fluids on surface properties of electrical discharge machined titanium alloy, Proc. Inst. Mech. Eng. 227 (2013) 1310–1316 [CrossRef] [Google Scholar]
  13. Y. Zhang, Y. Liu, Y. Shen, R. Ji, Z. Li, C. Zheng, Investigation on the influence of the dielectrics on the material removal characteristics of EDM, J. Mater. Process. Technol. 214 (2014) 1052–1061 [CrossRef] [Google Scholar]
  14. P. Sadagopan, B. Mouliprasanth, Investigation on the influence of different types of dielectrics in electrical discharge machining, Int. J. Adv. Manufactur. Technol. 92 (2017) 277–291 [CrossRef] [Google Scholar]
  15. C. Li, X. Xu, Y. Li, H. Tong, S. Ding, Q. Kong, L. Zhao, J. Ding, Effects of dielectric fluids on surface integrity for the recast layer in high speed EDM drilling of nickel alloy, J. Alloys Compd. 783 (2019) 95–102 [CrossRef] [Google Scholar]
  16. K.-T. Chiang, F.-P. Chang, D.-C. Tsai, Modeling and analysis of the rapidly resolidified layer of SG cast iron in the EDM process through the response surface methodology, J. Mater. Process. Technol. 182 (2007) 525–533 [CrossRef] [Google Scholar]
  17. S.S. Habib, Study of the parameters in electrical discharge machining through response surface methodology approach, Appl. Math. Model. 33 (2009) 4397–4407 [CrossRef] [Google Scholar]
  18. S. Prabhu, M. Uma, B.K. Vinayagam, Electrical discharge machining parameters optimization using response surface methodology and fuzzy logic modeling, J. Br. Soc. Mech. Sci. Eng. 36 (2013) 637–652 [CrossRef] [Google Scholar]
  19. M. Hourmand, S. Farahany, A.A.D. Sarhan, M.Y. Noordin, Investigating the electrical discharge machining (EDM) parameter effects on Al-Mg2Si metal matrix composite (MMC) for high material removal rate (MRR) and less EWR–RSM approach, Int. J. Adv. Manufactur. Technol. 77 (2014) 831–838 [CrossRef] [Google Scholar]
  20. R.V.H. Barenji, H. Pourasl, V.M. Khojastehnezhad, Electrical discharge machining of the AISI D6 tool steel: prediction and modeling of the material removal rate and tool wear ratio. Precis. Eng. 45 (2016) 435–444 [CrossRef] [Google Scholar]
  21. A. Soundhar, H.A. Zubar, M.T.B.H. Sultan, J. Kandasamy, Dataset on optimization of EDM machining parameters by using central composite design, Data Brief 23 (2019) 103671 [CrossRef] [Google Scholar]
  22. D. Mandal, S.K. Pal, P. Saha, Modeling of electrical discharge machining process using back propagation neural network and multi-objective optimization using non-dominating sorting genetic algorithm-II, J. Mater. Process. Technol. 186 (2007) 154–162 [CrossRef] [Google Scholar]
  23. A.P.D. Markopoulos, E. Manolakos, N.M. Vaxevanidis, Artificial neural network models for the prediction of surface roughness in electrical discharge machining, J. Intell. Manufactur. 19 (2008) 283–292 [CrossRef] [Google Scholar]
  24. M.K. Pradhan, R. Das, C.K. Biswas, Comparisons of neural network models on surface roughness in electrical discharge machining, Proc. Inst. Mech. Eng. B 223 (2009) 801–808 [CrossRef] [Google Scholar]
  25. S.S. Sidhu, A. Batish, S. Kumar, Neural network-based modeling to predict residual stresses during electric discharge machining of Al/SiC metal matrix composites, Proc. Inst. Mech. Eng. B 227 (2013) 1679–1692 [CrossRef] [Google Scholar]
  26. S. Kumar, A. Batish, R. Singh, T.P. Singh, A hybrid Taguchi-artificial neural network approach to predict surface roughness during electric discharge machining of titanium alloys, J. Mech. Sci. Technol. 28 (2014) 2831–2844 [CrossRef] [Google Scholar]
  27. P. Ong, C.H. Chong, Bin M.Z.W. Rahim, K.C. Lee, K. Sia, M.A.H. Ahmad, Intelligent approach for process modelling and optimization on electrical discharge machining of polycrystalline diamond, J. Intell. Manufactur. (2018) doi:10.1007/s10845-018-1443-6 [Google Scholar]
  28. M. Lin, C. Tsao, C. Hsu, A. Chiou, P. Huang, Y. Lin, Optimization of micro milling electrical discharge machining of Inconel 718 by Grey-Taguchi method, Trans. Nonferrous Metals Soc. China 23 (2013) 661–666 [CrossRef] [Google Scholar]
  29. A.M. Nikalje, A. Kumar, K.V.S. Srinadh, Influence of parameters and optimization of EDM performance measures on MDN 300 steel using Taguchi method, Int. J. Adv. Manufactur. Technol. 69 (2013) 41–49 [CrossRef] [Google Scholar]
  30. L. Tang, Y.F. Guo, Electrical discharge precision machining parameters optimization investigation on S-03 special stainless steel, Int. J. Adv. Manufactur. Technol. 70 (2013) 1369–1376 [CrossRef] [Google Scholar]
  31. V. Gaikwad, V.S. Jatti, Optimization of material removal rate during electrical discharge machining of cryo-treated NiTi alloys using Taguchi's method, J. King Saud Univ. 30 (2018) 266–272 [CrossRef] [Google Scholar]
  32. J.H. Jung, W.T. Kwon, Optimization of EDM process for multiple performance characteristics using Taguchi method and Grey relational analysis, J. Mech. Sci. Technol. 24 (2010) 1083–1090 [CrossRef] [Google Scholar]
  33. S. Singh, Optimization of machining characteristics in electric discharge machining of 6061Al/Al2O3p/20P composites by grey relational analysis, Int. J. Adv. Manufactur. Technol. 63 (2012) 1191–1202 [CrossRef] [Google Scholar]
  34. U.S. Yadav, V. Yadava, Experimental modeling and multiobjective optimization of electrical discharge drilling of aerospace superalloy material, Proc. Inst. Mech. Eng. B 229 (2014) 1764–1780 [CrossRef] [Google Scholar]
  35. R. Khanna, A. Kumar, M.P. Garg, A. Singh, N. Sharma, Multiple performance characteristics optimization for Al 7075 on electric discharge drilling by Taguchi grey relational theory, J. Ind. Eng. Int. 11 (2015) 459–472 [CrossRef] [Google Scholar]
  36. L. Selvarajan, M. Manohar, U.A. Kumar, P. Dhinakaran, Modelling and experimental investigation of process parameters in EDM of Si3N4-TiN composites using GRA-RSM, J. Mech. Sci. Technol. 31 (2017) 111–122 [CrossRef] [Google Scholar]
  37. C.-J. Tzeng, R.-Y. Chen, Optimization of electric discharge machining process using the response surface methodology and genetic algorithm approach, Int. J. Precis. Eng. Manufactur. 14 (2013) 709–717 [CrossRef] [Google Scholar]
  38. S. Gopalakannan, T. Senthilvelan, Application of response surface method on machining of Al-SiC nano-composites, Measurement 46 (2013) 2705–2715 [CrossRef] [Google Scholar]
  39. H.A.M. Hegab, H. Gadallah, A.K. Esawi, Modeling and optimization of Electrical Discharge Machining (EDM) using statistical design, Manufactur. Rev. 2 (2015) 21 [CrossRef] [Google Scholar]
  40. R. Swiercz, D. Oniszczuk-Świercz, T. Chmielewski, Multi-response optimization of electrical discharge machining using the desirability function, Micromachines 10 (2019) 72 [CrossRef] [Google Scholar]
  41. Jagadish, A. Ray, Optimization of process parameters of green electrical discharge machining using principal component analysis (PCA), Int. J. Adv. Manufactur. Technol. 87 (2015) 1299–1311 [CrossRef] [Google Scholar]
  42. S.N. Sahu, N.C. Nayak, Multi-criteria decision making with PCA in EDM of A2 tool steel. Mater. Today 5 (2018) 18641–18648 [Google Scholar]
  43. Y. Yildiz, M.M. Sundaram, K.P. Rajurkar, Statistical analysis and optimization study on the machinability of beryllium-copper alloy in electro discharge machining, Proc. Inst. Mech. Eng. B 226 (2012) 1847–1861 [CrossRef] [Google Scholar]
  44. T. Muthuramalingam, B. Mohan, Influence of discharge current pulse on machinability in electrical discharge machining, Mater. Manufactur. Processes 28 (2013) 375–380 [CrossRef] [Google Scholar]
  45. A. Torres, I. Puertas, C.J. Luis, EDM machinability and surface roughness analysis of INCONEL 600 using graphite electrodes, Int. J. Adv. Manufactur. Technol. 84 (2015) 2671–2688 [Google Scholar]
  46. Rahul, S. Datta, B.B. Biswal, S.S. Mahapatra, Machinability analysis of Inconel 601, 625, 718 and 825 during electro-discharge machining: on evaluation of optimal parameters setting, Measurement 137 (2019) 382–400 [CrossRef] [Google Scholar]
  47. C.P.S. Mohanty, S. Mahapatra, J. Sahu, Parametric optimisation of electrical discharge machining process: a numerical approach, Int. J. Ind. Syst. Eng. 22 (2016) 207–244 [Google Scholar]
  48. J. Sahu, S.S. Mahapatra, C.P. Mohanty, Multi-response optimisation of EDM parameters using data envelopment analysis, Int. J. Product. Qual. Manag. 15 (2015) 309–334 [CrossRef] [Google Scholar]
  49. C.P.S. Mohanty, S. Mahapatra, M.R. Singh, A particle swarm approach for multi-objective optimization of electrical discharge machining process, J. Intell. Manufactur. 27 (2014) 1171–1190 [CrossRef] [Google Scholar]
  50. P. Mukhopadhyay, S. Adhikary, A.K. Samanta, S. Maiti, S. Khan, S. Mudi, External force assisted electro discharge machining of SS 316, Mater. Today 19 (2019) 626–629 [Google Scholar]
  51. N.R. Costa, J. Lourenço, Z.L. Pereira, Desirability function approach: a review and performance evaluation in adverse conditions, Chemometr. Intell. Lab. Syst. 107 (2011) 234–244 [CrossRef] [Google Scholar]
  52. G. Skrabalak, Influence of electrode tool length on the micro EDM drilling performance, Proc. CIRP 68 (2018) 594–598 [CrossRef] [Google Scholar]

Cite this article as: Subhashree Naik, Sudhansu Ranjan Das, Debabrata Dhupal, Analysis, predictive modelling and multi-response optimization in electrical discharge machining of Al-22%SiC metal matrix composite for minimization of surface roughness and hole overcut, Manufacturing Rev. 7, 20 (2020)

All Tables

Table 1

Literature overview.

In the text
Table 2

Machining process parameters associated with their coded value of levels.

In the text
Table 3

Experimental plan layout and results.

In the text
Table 4

ANOVA results for predictive models.

In the text
Table 5

Overview of confirmatory test results.

In the text
Table 6

Cost estimation in electrical discharge machining of Al-SiC MMC.

In the text

All Figures

thumbnail Fig. 1

Microstructure of Al-22%SiC MMC: (a) SEM micrograph and (b) EDX analysis.
In the text
thumbnail Fig. 2

Schematic of experimental setup with methodologies proposed.
In the text
thumbnail Fig. 3

Comparison between experimental and predicted values of technological parameters: (a) surface roughness, and (b) overcut.
In the text
thumbnail Fig. 4

Normal probability plot for responses (Ra, OC).
In the text
thumbnail Fig. 5

Surface plots for illustration of machining parameters effect on surface roughness.
In the text
thumbnail Fig. 6

SEM images of poor surface quality of machined component at DC  =  15 amp, GV =  1.5 V, TON  =  200 μs, TOFF =  20 μs, and FP  =  0.6 kgf/cm2.
In the text
thumbnail Fig. 7

Influence of discharge current on radial overcut of machined hole at GV  =  2 V; TON  =  200 μs; TOFF  =  20 μs; FP  =  0.2 kgf/cm2.
In the text
thumbnail Fig. 8

Cross-sectional SEM images of machined surfaces showing an increase in recast layer thickness with discharge current.
In the text
thumbnail Fig. 9

Surface plots for illustration of machining parameters effect on overcut.
In the text
thumbnail Fig. 10

Influence of discharge current on radial overcut of machined hole at GV  =  1 V; TON  =  200 μs; TOFF  =  20 μs; FP  =  0.4 kgf/cm2.
In the text
thumbnail Fig. 11

Influence of peak discharge current on hole quality formed by EDM at GV  =  1V; TON  =  200 μs; TOFF  =  20 μs; FP  =  0.4 kgf/cm2.
In the text
thumbnail Fig. 12

Deposition of carbon layer at the bottom and edge of the tool electrode after EDM on Al-SiC MMC workpiece at GV  =  1 V; TON  =  200 μs; TOFF  =  20 μs; FP  =  0.4 kgf/cm2.
In the text
thumbnail Fig. 13

Optimization plot for surface roughness and overcut using desirability function approach.
In the text

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.