In the last few decades, non-traditional machining made the machining
process easier than the traditional machining method. Electric discharge
machining (EDM) is one of the most prominent methods of non-traditional
machining processes. By the use of EDM, a complex profile and high hardness
materials can be easily machined, which cannot easily be machined by the
traditional machining method. EDM is widely used by the industries. This
paper investigates an experiment with the cryogenically treated copper
electrode and an ordinary copper electrode with various input parameters
like the electrode rotation, gap voltage and discharge current for an EN24 (a high-strength and wear-resistant steel)
material. An experiment was performed with electric discharge machining.
Designs of an experiment are carried out using the Taguchi approach. An
orthogonal L16 array prepared and used the different combination of the
three input parameters (current, electrode rotation and gap voltage) to find
an optimum value of the factors. The output factors are the overcut (OC),
the tool wear rate (TWR) and surface roughness (Ra). The optimal level and importance
levels of each of these parameters are obtained statically using an analysis-of-variance (ANOVA) table through the analysis of the

The electrical discharge machining (EDM) is widely used to machine a very hard surface material at a low cost and with fewer hardness tools (Huang et al., 2003). This machining operation was improved and upgraded with time (Habib and Sameh, 2009). In the 1930s, EDM came into existence for the first time for a machining purpose, but due to overheating, lower material removal rate (MRR) and lower quality of the machined surface, it could not be used on a large scale. So to overcome this problem, the researchers worked continuously after the introduction of the EDM to improve the quality of the machining operation, improving surface finish, the material removal rate, etc. In the EDM process, removal of metal takes place due to the erosion carried by the sparks occurring between the tool and work piece (Srivastava and Panday, 2011). At present, EDM is the most widely used technique for high-precision machining of all types of conductive metals, irrespective of hardness. It is also used in the automobile industry, in aerospace and in the farm industry (Shazarel et al., 2009). The initial cost of EDM machining is high, but with the selection of the optimal parameter levels, its wastage and operating cost decrease with quality improvements (Kapoor et al., 2012). The EDM process continuously analysed with different levels of parameters to improve the quality of machining output. In this present research, machine factors like the overcut, TWR and Ra are taken as machining output (response) parameters. So for the better dimensioning and quality, TWR, Ra and overcut size should be at a minimum (B. R. Kumar et al., 2014).

Cryogenic treatment and tempering (CTT; Singh and Singh, 2011).

Kumar and M (2014) performed an experiment on a cryogenic cooled electrode and found a more beneficial effect on the machining operation than the conventional electrode. Gill and Singh (2010) investigated the effect of deep cryogenic treatment on the Ti 6246 alloy for a machining purpose. The investigators found a higher material removal rate and lower tool wear rate. Srivastava and Panday (2013) conducted an experiment with the process parameters like the pulse in time, current, duty cycle and voltage for response factors like the tool wear rate, material removal rate and surface roughness with a cryogenically treated electrode. They revealed that the current, pulse in time and duty cycle have a significant effect on the tool wear and material removal rate. In this research paper, the cryogenic treatment of an electrode is examined. Yildiz and Nalbant (2008) made an improvement in tool life by using the cryogenic process in the cutting operation. Abdulkareem et al. (2009) reveal that surface roughness is reduced by using a cryogenically cooled electrode during machining. Kumar and M (2015) carried out an investigation with machining parameters including the pulse in time, gap voltage and current for output responses comprised of the material removal rate, electrode wear rate, and electrode temperature for conventional and cryogenic EDM. The pulse in time has the most significant factor, and moreover, electrode wear was reduced to 18 % by the cryogenic electrode.

EDM is applicable to electrically conductive materials. In this paper, the research is conducted on a EN24 steel work piece. This material is used for variety of parts, dyes, gears, shafts and moulds. EN24 steel is widely used in power and transmission due to its high tensile strength. In this experiment a 10 mm diameter of the copper electrode is used. The weight of the electrode and work piece was measured with the help of a weighting machine. Moreover thickness and the diameter of the work piece and tool are measured with a digital micrometer.

Chemical composition of material.

The work piece and electrode were submerged in dielectric fluid, which is shown in
Fig. 2. The dielectric fluid is the insulator of electricity. Therefore this
insulation property of the dielectric produces a required limit of the potential
difference between the tool and work piece that is necessary for spark
generation (Zhiguang et al., 2015). The gap difference between the work piece and
electrode was maintained with the help of the servo mechanism. This mechanism
maintains a gap of about 0.03 mm. In this mechanism, operation length and the diameter of
the work piece are respectively 28 and 15 mm. In this machining, the dielectric
fluid is kerosene oil (Gill and Singh, 2010). The electrode material used in EDM
is the copper electrode, having 100 HB (HB – Brinell hardness unit) hardness and a electrical resistivity
of 9

Chemical composition of electrode.

The D.C (direct current) supply of a high ampere current at low voltage is given to the electrode. This permits the spark between the electrode and the work piece through the dielectric fluid (Singh and Singh, 2011). This spark generates high heat between the electrode and work piece. This heat is sufficient for eroding or melting the material from the work piece. The molten particles are flushed with the flow of a dielectric fluid. This flow is circulated with the help of a pump. In this experiment both electrodes connected with negative polarity and the work piece in positive polarity, as shown in Fig. 2.

Overcut (Habib and Samesh, 2009).

The Taguchi technique is very helpful for optimizing the results. It is mostly used by manufacturing engineers, scientists and quality-assurance experts. The main target of the Taguchi method is to set up an experiment and to develop superior performances. Classical experimental design methods are too complex and not easy to use. Furthermore, a large number of experiments have to be carried out as the number of the process parameters increases. Instead of testing all possible combinations like the factorial design, the Taguchi method is applied. So to solve this important task, the Taguchi method uses a special design of the orthogonal array to study the entire parameter space with only a small number of experiments. An experimental design scheme of statistical experiments that uses orthogonal arrays, however, entails the following considerations and consequences.

The orthogonal array focusses only on a main effect design. An orthogonal array has been used by researchers for the last few years to determine the minimum number of experiments. The selection of orthogonal array is one of the important sets for performing the experiments to determine the optimum level for each parameter.

Analysis of variance (ANOVA) is selected for choosing new parameter
values to optimize the performance characteristic at 95 %. The

The experimental results are then transformed into a signal-to-noise (

The objective of this research is to minimize the tool wear rate, surface
roughness and overcut. In this research paper

Cryogenical treatment of the electrode is performed in a cryogenic chamber, which is shown in Fig. 1. In this operation liquid nitrogen gas is used to
perform the cryogenic treatment. The cryogenic treatment (CT) is a slow
cooling process, with a cool-down rate of approximately 2–3

Untreated and treated electrode microstructure.

Therefore, to control on this problem, the tool is tempered after cryogenic treatments, which is shown in Fig. 1.

Input factor levels and values.

Cryogenically treated (CT) electrode microstructure.

Cryogenically treated with tempering (CTT) of the electrode microstructure.

The tempering process helps in reducing the stress and brittleness. Moreover it transforms the retained austenite into martensite with the formation of a fine and uniform chromium carbide microstructure. The untreated, cryogenically treated and deep cryogenically treated with tempering (CTT) microstructure analyses are shown in Figs. 3, 4 and 5 respectively. The microstructure in CTT is more uniform, small in size and symmetrical than the cryogenically treated and the non-cryogenic electrode. In Fig. 5 microstructures are more uniform than in Figs. 3 and 4. The precipitation of the fine carbide particle structure of carbide shown in Fig. 6b is more fine and uniform than in Fig. 6a. So this fine carbide improves toughness, fatigue resistance, hardness, wear resistance, surface smoothness and dimensional stability. Therefore CTT is more significant than CT.

Scanning electron microscope (SEM) images of

In these cases (deep cryogenically treated and untreated), the same level and parameters are used. During the operation in EDM, some material was removed from the tool and as well as from the work piece, but the material removal rate from the work piece is very high compared to the tool wear rate because the work piece is connected with positive polarity, and therefore a large amount of heat is generated on the work piece during sparking (Kumar et al., 2012; Kalsi et al., 2010). The Taguchi design is used to perform the experiment, and afterwards ANOVA is used for analysis. In this research, L16 is used for the experiment. There are three input factors (current, electrode rotation and voltage) with four levels that are given in Table 3. In the literature there is much less work done with electrode rotation with deep cryogenic treatment. There is no research in the literature with these levels that has been selected for a research paper on EN24 steel. Due to this, these levels are selected for traditional and non-traditional EDM.

Orthogonal array with performance for simple (non-cryogenic) electrode.

Quality is an important factor that customers use to evaluate a product or
service. So the new quality-control and improvement programmes have to make
their products more acceptable to the customers. On the other hand, a customer
evaluates a product performance based on a number of diverse qualitative
characteristics. To improve the rational decision-making, the evaluations of
various attributes should be combined to give a composite index. Such a
composite index is known as the utility of a product. The sum of utilities of
each quality attribute represents the overall utility of a product. It is
difficult to obtain the best combination of process parameters when there
are multiple responses. If

In EDM, the cavity is produced with the help of a spark that is generated
between the electrode and work piece. Therefore the overcut is the difference between
the size of the cavity on the work piece and the diametrical size of the
electrode:

Surface roughness is measured with the help of a portable Profilometer Talysurf. Tables 4 and 5 represent the values of the overcut, tool wear rate and surface roughness for conventional EDM and the deep cryogenic electrode and cryogenically treated EDM (CCT) respectively. For the tool wear rate,

Orthogonal array with performance for deep cryogenic electrode.

Tables 6 and 7 represent the ANOVA for the

Analysis of variance (ANOVA) for

Analysis of variance (ANOVA) for

Tables 8 and 9 represent the ranked value of an input variable like the current, rotation and voltage for an output variable or response variable for the untreated electrode and deep cryogenically treated electrode. In this table the current has a high ranked value (high impact) for the tool wear rate and surface roughness, but in the case of the overcut, the current has much less impact.

Mean response table for untreated electrode.

Mean response table for the deep cryogenically treated with tempering (CTT) electrode.

OC

TWR

The graphs are made with the help of Minitab software. These graphs highlight the process parameters and different optimal levels for machining.

Current and voltage relation of OC for the non-cryogenic tool.

Rotation and voltage relation of OC for the non-cryogenic tool.

Ra

OC

These optimal parameters are the current (130), rotation of electrode (1100) and voltage (165 V) for an ordinary electrode (non-cryogenically treated electrode). These levels give an optimal value of the overcut (rough cut) that is shown in Fig. 7.

TWR

Ra

Rotation and voltage relation of OC

Current and voltage relation of OC

This figure helps in selecting the optimal levels (having a small value of the

In this section, the optimal values of the response characteristic (overcut, tool wear rate and surface roughness) and their levels have been predicted for the untreated and deep cryogenically treated electrode.

The 95 % confidence intervals of confirmation experiments and the sample group
(CI

The confidence intervals of confirmation experiments are represented by

After completing the above experiment, the next step is to find the significant parameters and their importance and a validation test for traditional EDM by using optimal levels.

In the case of untreated electrode for overcut (OC), find the predicted
values at the significant condition. The selected values of significant factors
like current (A1), electrode rotation (B1) and voltage (C3) are shown
in Fig. 7 and Table 6. The estimated mean of the response characteristic
can be determined as

The confidence interval for the sample group (CI

By putting these values in Eq. (1), CI

The confidence interval for the entire group (CI

From the CI

The optimal values of process variables are as follows:

The first level of the current is AI

The first level of the electrode rotation is BI

The third level of the voltage is CIII

The optimal value of TWR is predicted at significant variables. The
estimated mean of the response characteristic can be determined as

so CI

The predicted range for sample size is

The predicted range for entire group is

The optimal values of process variables are as follows:

The first level of the current is AII

The first level of the electrode rotation is BIV

The third level of the voltage is CII

The optimal value of Ra is predicted at significant variables. The estimated
mean of the response characteristic can be determined as

so CI

The predicted range for the sample size is

The predicted range for entire group is

The optimal values of process variables are as follows:

The first level of the current is AI

The first level of the electrode rotation is BI

The third level of the voltage is CI

Similarly the next step is to find the significant parameters and their importance and the validation test for non-traditional EDM by using the same optimal levels that are used in traditional EDM.

In the case of OC

The confidence interval for entire group (CI

Similarly, put the values of

The predicted range for entire group is

The equations show that the mean values lie within limits. Therefore the both tests are validated. The optimal level values of factors are the same as in the overcut of traditional EDM.

The optimal value of TWR

The predicted range for sample size is

The predicted range for entire group is

The optimal value of Ra

So CI

The predicted range for sample size is

The predicted range for entire group is

From this, CI

For the overcut (OC), it is 1.112

For the tool wear rate (TWR), it is 4.031

For surface roughness (Ra), it is 5.75

For overcut (OC

For the tool wear rate (TWR

For surface roughness (Ra

To evaluate the adequacy of the proposed approach and statistical analysis, a set of verifications is conducted on predicted values. The confirmation test is carried out to check the reproducibility of predicted results. The predicted results came from Eqs. (9) to (14).

Conformation table.

By using optimal parameters in the above equation, predicted OC and OC

Figure 17a and b show SEM images of the work piece surface by using the surface roughness optimal values (AI, BI and CI) of factors for both traditional and non-traditional EDM respectively. The significant variables labels are a current of 130 A, an electrode rotation of 1100 rpm (rpm – rotations per minute) and voltage of 90 V. Based on the recast layer, globules and cracks are observed. The cracks are formed due to thermal stress that is generated during machining. The Fig. 17a notation image shows the work piece surface that is used for the untreated electrode, and Fig. 17b shows the work piece surface that is used for the deep cryogenically treated electrode.

Percentage difference between traditional and non-traditional EDM.

Response factor chart.

Figure 17 shows that the globule size and crack size are smaller in
non-traditional than traditional or conventional EDM. Sizes of globules and
cracks affect the surface roughness. If the size of globules and cracks is smaller, then the surface roughness is also smaller. So the cryogenically treated
electrode has a more significant resultant than the untreated electrode
(conventional EDM). Figure 18 and Table 10 show the percentage difference
between the traditional and non-traditional EDM. The experiment performed by
taking the same levels for the untreated and cryogenically treated electrode
found that the overcut, tool wear rate and surface roughness were reduced by
9.12 %, 13.22 % and 15.75 % respectively in non-traditional EDM compared to
conventional EDM. Figure 19 shows which response factor is more
affected if the experiment is conducted with the deep cryogenically treated
electrode. The percentages affecting the response factors are 24 %,
35 % and 41 % for the overcut, tool wear rate and surface roughness
respectively. From the experiment that is represented in the form of a chart in
Figs. 18 and 19, surface roughness is found to have reduced more than the
other response factors. The percentage difference that can be calculated is given
below:

Tool life is longer and no special tooling is required, so the tool cost decreases. Non-traditional EDM eliminates secondary finishing operations because the surface roughness is less than traditional EDM. A thin wall can be machined due to a smaller overcut in the deep cryogenically treated electrode than in a normal electrode. Surface finish of product is good, and wastage of raw material is less in the deep cryogenic treatment than in traditional EDM.

The present experimental work is concerned with determining the optimum setting of EDM to find the optimal overcut, TWR and Ra. L16 was used for performing the experiment.

The voltage and electrode rotation have the maximum effect on the radial overcut (OC), also known as ROC, compared to the pulse current in the case of a non-cryogenic electrode. In the case of surface roughness for an untreated electrode, the pulse current has a higher ranked effect than rotation and voltage. Like in the tool wear rate, the pulse current has dominating effects.

Optimum parameters (for both cases) of the overcut are as follows: a current of 130 A, a rotation speed of 1100 rpm and a voltage of 165 V. The best fits for the tool wear rate are a current of 210 A, an electrode rotation speed of 2100 rpm, and a voltage of 120 V. In cases of surface roughness, 130 A, 1100 rpm and 90 V are the best levels with respect to the current, rotation speed and voltage.

The pulse current and rotational speed are the most significant factors for the tool wear rate and surface roughness in the case of the deep cryogenically treated electrode, but in the case of the deep cryogenically treated electrode, for the overcut, only electrode rotational speed has a larger effect.

The validation and confirmation test confirmed that the adequacy of a law of additively is justified. The deep cryogenically treated electrode has a smaller value of the overcut than the non-cryogenically electrode. So the deep cryogenically treated electrode (non-traditional) has larger effects on the response factors than the traditional electrode; 9.12 % less overcut was found in this experiment. In other words, approximately 9 % less overcut was produced in the cryogenically tempering-treated tool compared to the non-cryogenically treated tool. In the case of the tool wear rate, the wear rate is reduced by 13.22 % with cryogenic treatment. On the other hand, it was observed in the experiment that the surface roughness was reduced by 15.75 % (approximately 15 % to 16 %) by using the deep cryogenically treated electrode instead of the untreated electrode. So in the end, it is found that machining has positive results after the cryogenic tempering treatment of the electrode.

Deep cryogenic treatment improves the thermal conductivity of the tool and reshapes carbide into a uniform and homogenous structure more frequently than the non-cryogenic tool. Due to this, the power consumption and wear losses are reduced. Therefore, the deep cryogenic tool is used as a priority base more than the non-cryogenic tool where a high-speed-cutting operation was performed.

The deep cryogenic treatment improves tool life, so it reduces investment for the purchase of new cutting tool; moreover less wear and tear decreased the reshaping and regrinding of a tool, so it reduced the labour cost and ideal time of machining to replace the tool. The deep cryogenically treated electrode improves the quality and production rate more than the non-cryogenic electrode.

The experiment between the deep cryogenic and untreated electrode represented non-traditional EDM being more effective and efficient than the traditional EDM process.

Data used in this article are available upon request.

GSG and DPD were responsible for problem finding, conceiving and designing the analysis, collecting the data, contributing data or analysis tools, conducting the experiment and receiving a fruitful result, performing the analysis, and writing the paper.

The authors declare that they have no conflict of interest.

Gurdev Singh Grewal would like to express his deep gratitude to his father S. Sukhdev Singh Grewal for guidance and financial support.

This paper was edited by Xichun Luo and reviewed by Senthil Kumaran Selvaraj and one anonymous referee.