1. Introduction
Economic globalization constantly pushes the manufacturing sector to advanced and innovative transformations through groundbreaking and state-of-the-art technologies. Hence, to remain competitive in today’s marketplace, mechanical components must be manufactured with high accuracy and reliability and in the shortest possible time. Materials science has developed advanced engineering materials such as super alloys, composites, and ceramics. These materials are difficult to machine and often impede smooth machining with traditional machining processes such as turning, milling, drilling, and grinding. As a result, electrochemical machining, ultrasonic machining, wire cutting, wire electrical discharge machining (WEDM), and other non-traditional methods of machining are used on difficult-to-machine materials.
Wire electric discharge machining (WEDM) is a fast-growing and non-conventional advanced manufacturing process used for high-strength materials. Every hard material which is difficult to cut with other conventional methods can be sliced using WEDM [
1]. To meet the rapidly increasing need for materials with unique properties in sophisticated professional programs such as aeronautical and medical devices, WEDM is the best option [
2]. The WEDM process extracts particles from the surface through thermal erosion [
3]. In order to use WEDM, the workpiece and tool must be electrically conductive [
4,
5]. The wire acts as a cathode, and the work material acts as an anode. There is no physical contact between the workpiece and the wire, thereby allowing sensitive, breakable objects to be fabricated without risking damage [
6,
7]. Mechanical stresses are also reduced since the tool and workpiece do not interact during machining [
8,
9]. An electric shock is further created in discharge energy, and electricity spikes at limited intervals are supplied. A release is normally started when the electric field is improved [
10,
11]. When the spark comes in contact with the dielectric fluid (present between the workpiece and tool), it ionizes and allows the flow of current between the tool and the workpiece, forming an ionization gradient [
12]. This then results in a rapid increase in the temperature of the metal, i.e., from 8000 °C to 12,000 °C (occasionally much higher), causing the surface particles of the workpiece to melt rapidly. The gap between the tool and the workpiece is precisely acclimatized to ensure that ionization occurs. If the gap between the workpiece and the tool is not correctly maintained, there will be no spark, leading to no cutting. Therefore, a constant gap of 0.5 mm is maintained between the workpiece and the tool to produce a spark [
13,
14]. Moreover, dielectric fluid is mostly used in WEDM with sufficiently high dielectric resistance to not only avoid a quick breakdown electrically but also to ionize when electrons collide with its molecules. Furthermore, during sparking, the dielectric fluid should be thermally resistant as well. In such a case, deionized water is used as a dielectric fluid for most of the WEDM processes [
15,
16]. In order to measure WEDM’s performance, the literature shows the material removal rate (MRR) and surface roughness [
17,
18] as the key output factors. Similarly, the popular input factors for WEDM are current, voltage, pulse-on time, pulse-off time, wire tension, and wire speed [
19]. Over the years, a lot of interest has been invested by researchers in improving the MRR for faster manufacturing. Here, several variables affect the MRR, such as current, voltage, and pulse-on-time. Current is usually determined during the ‘on-time’ of each pulse while the pulse-on-time is determined as the time during which the current is permitted to flow in each cycle [
20,
21]. Similarly, voltage influences the MRR as well and can be calculated from the spark zone average power during machining. Voltage also influences the amount of overcutting and the gap in the spark [
22,
23,
24,
25,
26]. Therefore, in fabricating carbon–carbon alloys, the best WEDM machine setup is for input parameters that involve current, voltage, and pulse-on time, which further also impact the electrode wear rate.
To achieve optimum machining in WEDM, it is critical to choose appropriate machining parameters. These parameters are usually chosen via experience, and it may not guarantee optimum or near-optimal machining performance for WEDM. Gupta et al. [
27] studied the effects of WEDM machining input parameters such as servo voltage, wire-feed speed, and wire tension, on surface roughness and cutting speed of titanium (Ti-6Al-4V) alloy using response surface methodology (RSM) and analysis of variance (ANOVA). Moreover, the effects of WEDM machining input parameters on the MRR, surface roughness, gap voltage, gap current, and cutting rate for AISI D2 steel were analyzed by Singh et al. [
28]. Taguchi L27 orthogonal array (OA) was also used along with RSM and ANOVA. Furthermore, the effects of WEDM input parameters on the MRR for hot die steel AISI H-11 were explained by Singh et al. [
29], wherein input parameters such as pulse-on time, pulse-off time, gap voltage, peak current, wire feed, and wire tension, were studied. The one variable at a time (OVAT) approach showed that pulse-on time was directly proportional to the MRR. Selvakumar et al. [
30] studied the effects of WEDM input parameters on cutting speed, surface roughness, and the taper error for AISI D3 tool steel. Taguchi-based grey relational analysis (GRA) was used along with the Taguchi L
9 OA for the analysis. In another work, the effects of pulse-on time, pulse-off time, wire feed, and wire tension on the MRR and surface roughness of tungsten carbide were studied by Masooth and Arunnath [
31]. Again, Taguchi L
9 OA, along with ANOVA, was used. Moreover, the effects of WEDM input parameters (pulse-on time, pulse-off time, wire feed, flushing pressure, spark voltage, and wire tension) on the surface roughness and the MRR for high-strength armor steel were studied by Bobbili et al. [
32] using Taguchi’s L27 OA. Mohamed and Lenin [
33] also studied the effects of WEDM input parameters on the machining time for aluminum 6082 T6 alloy using pulse-on time, pulse-off time, and current as the input parameters. Satyanarayana [
34] examined the same for Inconel 600. Deshmukh et al. [
35] explained the effects of WEDM input parameters on the surface roughness and kerf width of AISI 4140 using pulse-on time, pulse-off time, servo voltage, and wire feed as input parameters. Taguchi’s L
9 OA was used along with GRA, ANOVA, and regression analysis to analyze the interactions and main effects. For DC53 die steel, Nawaz et al. [
36] examined the effects of WEDM input parameters on the MRR, kerf width, and surface roughness. Similarly, Gavisiddesha et al. [
37] studied the effects of WEDM input parameters on surface roughness of composite material (AL6061/SICP) using pulse-on time, pulse-off time, and current as input parameters. Again, Taguchi’s L
9 OA was used along with ANOVA. Input parameters such as pulse-on time, pulse-off time, servo voltage, peak current, wire tension, and water pressure were further used to study the effects of WEDM input parameters on the surface roughness of Vanadis-4E (powder metallurgical cold worked tool steel) by Sudhakara and Prasanthi [
38] using design of experiments (DOE). Khan et al. [
39] examined the effects of WEDM input parameters on stainless steel’s surface roughness and kerf width (SS 304) using Taguchi’s DOE with L
9 OA, GRA, and ANOVA. Introducing spark gap as an output parameter in addition to the MRR and surface roughness, Rajyalakshmi and Ramaiah [
40] explained the effects of WEDM input parameters (pulse-on time, pulse-off time, voltage, flushing pressure, wire feed rate, wire tension, spark gap, and servo feed) on Inconel 825. Taguchi’s DOE with L36 OA was used along with GRA and ANOVA. Lastly, Lingadurai et al. [
41] studied the effects of WEDM input parameters on the metal removal rate, kerf width, and surface roughness of stainless steel AISI grade-304. Again, Taguchi’s DOE with L18 OA was used along with ANOVA.
In light of the reviewed state-of-the-art, it is evident that optimizing the input process parameters of WEDM is one of the most important design objectives for achieving a higher MRR. Taguchi DOE and ANOVA are the most effective techniques for determining the optimal settings of process variables and their corresponding interaction effects for a given target. Moreover, in previous approaches, the majority of optimization has been conducted on different alloys and super alloys by using different optimization techniques such as optimization through mathematical modeling, full factorial design, optimization through RSM, and the finite element method. Compared to these approaches, Taguchi’s DOE is not only straightforward, efficient, and trustworthy for decreasing costs and enhancing quality, but it also significantly decreases the number of trials.
Consequently, in this work, the parametric optimization of machining process parameters of WEDM is undertaken with the help of Taguchi DOE and ANOVA, considering AISI 1045 medium carbon steel to increase the MRR. The remainder of the paper is structured as follows:
Section 2 discusses the adopted methodology in detail along with experimentation and results;
Section 3 discusses the results; and
Section 4 concludes the paper with the optimal input parameter settings along with validation.
3. Results and Discussion
This research performed nine experiments (using L
9 OA) with different input controllable factors on an AISI 1045 medium carbon-steel-based timing chain sprocket. The Taguchi DOE was used to find the best input factor settings to obtain the maximum MRR. Thereafter, the S/N ratio and ANOVA were applied. The results from the S/N ratio are shown in
Table 10, and their ranking is shown in
Table 11.
Table 11 clearly shows that the current ranks first, indicating that the current has the greatest influence on the MRR. The results from the ANOVA are also listed in
Table 12 for the MRR vs. current, voltage, and pulse-on time. It is evident from
Table 12 that the MRR vs. the current shows that the concerned p-value is 0.026, thereby implying that the current has a statistically significant impact on the MRR.
Moreover, the R2 value for the ANOVA model for the MRR was 70.23% for the effect of current. For the other two process parameters (voltage and pulse-on time), the R2 values were not significantly higher.
Figure 7 and
Figure 8 show the main effect plot for means and S/N ratios. Thus, a combined analysis of
Table 11 and
Figure 7 and
Figure 8 shows that I = 16-amp, V = 50-volt, and pulse-on time = 100 µs were observed as the optimum parameters to reach the maximum MRR for the timing chain sprocket. A confirmation run was also conducted using the optimal set, which was approved by the automobile manufacturing industry.
The results of the study showed that the current had the greatest influence as an input parameter on the MRR for AISI 1045 medium carbon steel, followed by pulse-on time and voltage. As the current was increased, the MRR also increased. However, when the voltage was increased, the MRR decreased. The impact of increased current for increased MRR in the case of AISI 1045 medium carbon steel was also supported by Mohammadumar et al. [
42] and Patel et al. [
47]. This is because the voltage determines the width of the spark gap between the leading edge of the electrode and the workpiece. Higher voltage settings increase the gap, due to which the number of sparks decreases, and the machining rate slows down. Similarly, as the pulse-on time increased, the MRR decreased. Material removal is directly proportional to the amount of energy applied during the pulse-on time. This energy is controlled by the peak current and the length of the pulse-on time. However, the extreme pulse duration can be counterproductive. When the optimum pulse duration for each tool and workpiece material combination was exceeded, the MRR started to decrease. Long pulse durations can also restrict the electrodes from machining.