|Year : 2012 | Volume
| Issue : 2 | Page : 104-112
An Investigation into Dimensional Deviation Induced by Wire Electric Discharge Machining of High temperature Titanium alloy
Mohinder P Garg1, Ajai Jain2, Gian Bhushan2
1 Department of Mechanical Engineering, M. M. Engineering. College, Mullana, India
2 Department of Mechanical Engineering, N. I. T. Kurukshetra, Haryana, India
|Date of Web Publication||4-Aug-2012|
Mohinder P Garg
Department of Mechanical Engineering, M. M. Engineering. College, Mullana
Source of Support: None, Conflict of Interest: None
| Abstract|| |
This article investigates the Wire Electric Discharge Machining of Titanium alloy 6-2-4-2. Six process parameters namely pulse-on time, pulse-off time, peak current, spark gap set voltage, wire feed, and wire tension are taken into account, to study their effect on dimensional deviation. The experiments are conducted using Box-Behnken designs. Empirical relation is developed between the process parameters and dimensional deviation by using regression analysis. Analysis of Variance is carried out to identify the significant process parameters affecting the process. Consequently, the optimal sets of parameters yielding the minimum dimensional deviation are obtained using the desirability approach. The optimal parameter combinations have been verified by conducting confirmation experiments. Results of the confirmation tests show that the developed mathematical models are appropriate for effective machining of Titanium alloy using Wire Electric Discharge Machining.
Keywords: Wire electric discharge machining, dimensional deviation, Box-Behnken design, Titanium alloy 6-2-4-2
|How to cite this article:|
Garg MP, Jain A, Bhushan G. An Investigation into Dimensional Deviation Induced by Wire Electric Discharge Machining of High temperature Titanium alloy. J Eng Technol 2012;2:104-12
|How to cite this URL:|
Garg MP, Jain A, Bhushan G. An Investigation into Dimensional Deviation Induced by Wire Electric Discharge Machining of High temperature Titanium alloy. J Eng Technol [serial online] 2012 [cited 2019 Nov 17];2:104-12. Available from: http://www.onlinejet.net/text.asp?2012/2/2/104/99298
| 1. Introduction|| |
Titanium and its alloys are considered excellent materials for their applications in aerospace and automobile industries because of their improved mechanical and physical properties like higher strength, strength to weight ratio, toughness, corrosion resistance, and oxidation resistance. However, due to these properties, titanium and its alloys are difficult to shape and machine into precise shape and size. Thus, their widespread applications have been hindered by high cost of machining . Conventional machining of Ti and its alloys is difficult due to the following reasons:
- High thermal stresses are induced at the cutting edge due to low heat dissipation by the chips and the workpiece. By the synergetic effect of low thermal conductivity and high thermal capacity, 30% more heat is absorbed by the cutting edge, as compared to machining of ck 45 steel .
- The combination of low Young's modulus with a high yield stress ratio allows only small plastic deformations. The material is elastic and keeps springing back under cutting pressure. At the flank face, this leads to a lower effective clearance angle. Thus friction is enhanced and chatter is supported .
- During machining, due to the large shearing angle, thin chips come in contact with a relatively small area on the tool face and result in high loads per unit area. These high forces, coupled with friction developed by the chips as they pass over the cutting area; result in a greater increase in heat over a much localized portion of the cutting tool. It leads to a shorter tool life. Thus, as cutting speed increases; the tool life drastically decreases .
- During machining Ti 6242 with Tungsten Carbide inserts with a rake angle of 10° deformation of metal increased with increasing speed. The depth of the deformed layer and extent of deformation depends upon cutting parameters such as cutting speed, tool geometry, and so on .
Thus, there is a need to machine Ti and its alloys using nonconventional machining such as Ultrasonic machining, Laser beam machining, Electrical discharge machining (EDM), Wire electrical discharge machining (WEDM), and so on. WEDM is found to be an extremely potential electrothermal process in the field of conductive material machining and is widely used in the manufacture of cam wheels, stators for stepper motors, press tools, dies, and the like. It is a thermoelectric process, which erodes material from the workpiece by a series of discrete sparks occurring between the work and the tool electrode immersed in a dielectric medium. These electrical discharges melt and vaporize a small amount of work material, which is then ejected and flushed away by the dielectric. WEDM provides an effective machining technique that favors the production of parts made from difficult-to-machine materials, with complicated geometries, such as cam wheels, stators for stepper motors, press tools, dies, and so on . In WEDM, due to the stochastic nature of the process, selection of favorable process parameters is required to achieve optimal machining performance in terms of productivity and surface quality. Most of the machine tool manuals provide machining guidelines for conventional materials like steel, brass, and aluminium, but there are no guidelines available for Ti and its alloys. The optimization of process parameters is now dependent on process analysis, to identify the effect of operating variables, for achieving the desired machining characteristics . A number of research studies have been directed toward optimizing Cutting speed, Metal removal rate, and Surface roughness, using WEDM. Shah et al.  investigated seven machining parameters in addition to varying material thickness on machining responses, such as, MRR, kerf, and surface roughness of tungsten carbide samples machined by WEDM. The design of experiments was based on Taguchi orthogonal designs with eight control factors at three levels. The results showed that material thickness has little effect on material removal rate and kerf. Sadeghl et al.  discussed the effects of process parameters on surface roughness and metal removal rate in the WEDM of an AISI D5 steel alloy. Regression was used to model the process and the Tabu search algorithm was opted for optimization. It was found that the discharge current and pulse interval were more influential on MRR and surface roughness than the open circuit voltage. Some research studies have been done, keeping in view the accuracy aspects of the WEDM operation [9,10]. Dimensional deviation, an aspect of accuracy obtained in the WEDM operation, was investigated by Sarkar et al. . The authors developed a second-order mathematical model for surface roughness, dimensional deviation, and cutting speed during the WEDM of γ-TiAl, in the trim cutting operation, using response surface methodology. The residual analysis and experimental results indicated that the proposed models could adequately describe the performance indicators within the limits of the factors that were being investigated. Jangra, Jain, and Grover  presented the optimization of surface roughness and dimensional lag of WEDM, using the Taguchi and Gray Relational Analysis on tool steel. Taguchi's L 18 Orthogonal Array was used to conduct experiments.
Manufacturing of Ti aluminide components is generally carried out using WEDM, and this aerospace alloy is extensively used in the manufacturing of engine ducts, exhaust nozzles, fans, automotive valves, gas turbine blades, engine disks, and so on . No technology tables or charts are available for the WEDM of such important and useful material in the industry. Moreover, the manufacturer's catalog supplied with the WEDM machine does not recommend any parameter setting for the machining of Titanium aluminide (Ti 6-2-4- 2) material. Thus, in the present article an attempt has been made to optimize the process parameters of WEDM, for minimization of the dimensional deviation observed during machining of Ti 6242, for a rough cut, using the Box-Behnken designs and Response Surface Methodology. [Table 1] and [Table 2] show the chemical composition and mechanical properties of Ti 6-2-4-2.
This study has been carried out to develop an empirical relationship between different process parameters, namely, Pulse On Time (TON), Pulse Off Time (TOFF), Peak Current (IP), Spark Gap Set Voltage (SV), Wire Feed (WF), Wire Tension (WT), and output response, that is, dimensional deviation. Dimensional deviation is an important measure of performance, the knowledge of which is very essential to achieve close dimensional control in WEDM. The actual job produced by WEDM is undersized or oversized depending on whether the job is a punch or a die. Determination of dimensional deviation for a particular process parameter combination assists the manufacturing planners in setting a wire offset so that the actual (cut) workpiece dimensions match the part geometry.
| 2. Experiment Set Up and Design|| |
The experiments were performed on a four axis Electronica Sprintcut 734 CNC Wire cut machine manufactured by Electronica India Limited, Pune (India). A rectangular plate of Ti 6-2-4-2, 200 mm × 200 mm × 20.4 mm in size, was taken as the work material. A 10 mm × 10 mm square cut was taken on the workpiece. [Figure 1] illustrates the CNC wire cut Machine used and [Figure 2] shows a view of the machining operation. A diffused brass wire of 0.25 mm diameter (Nikunj HH) was used as tool material and deionized water was used as a dielectric.
[Figure 3] shows the path followed by the wire. The profile of the workpiece to be cut is illustrated in [Figure 3]. The wire enters the workpiece at point O (5, 0). It moves along OABCDO and exits the workpiece from O (5, 0). The CNC code for cutting is generated using the ELAPT software supplied by the manufacturer. It is important to mention that the wire offset is set at zero during machining.
Where (d-a) = width of the cut
d = desired size of workpiece = 10 mm
a = actual size of workpiece obtained after machining
Dimensional deviation is measured using the Mitutoyo digital Vernier Callipers, with a least count of one micron. It is measured at two random places on sides AB, BC, and CD, and the average of these six values represents the dimensional deviation used in the present article.
The present study utilizes six process parameters, namely, TON, TOFF, IP, SV, WF, and WT, each at three levels. These have been decided on the basis of a literature review as well as preliminary study . [Table 3] provides details of these process parameters as well as other fixed parameters. It is important to mention that TON, TOFF, as well as WT values mentioned in the brackets in [Table 3] show the actual settings in micro Siemens (μs) and grams (g), respectively, whereas, the values outside the brackets indicate the machine control panel settings.
The present study utilizes the Box-Behnken experimental design approach as it plans experiments within an identified search space (assuming α=1). In WEDM, loss of productivity occurs due to wire breakage. Thus, if a perliminary study is done before actual experimentation, then the range of parameter conbinations, where wire breakage takes place can be identified and isolated. Moreover, Box-Behnken designs are rotatable or nearly rotatable second-order designs, based on three-level incomplete factorial designs. The output response (y) in the Box-Behnken design can be modeled as given below :
Where x i , x j , x k = input or independent process parameters
β0 , βii, βij =regression coefficients
ε = Random error
As present study considers six control factors at three levels, 54 experiments need to be performed according to the Box-Behnken designs. [Table 4] summarizes various parameter combinations for 54 experiments, as well as run order. It is planned to carry out one replication of each experiment, and thus, a total of 108 experiments are to be conducted.
| 3. Results and Discussion|| |
Experiments are conducted on an Electronica 4 axis Sprintcut-734 CNC Wire Cut machine. Each time an experiment is performed, a particular set of parameter combinations is chosen, and the workpiece is cut as per [Figure 3]. [Table 4] summarizes the results obtained for 54 experiments with one replication. Dimensional deviation values shown in [Table 4] are an average of the values obtained in both runs.
3.1 Effect of process parameters on dimensional deviation
The first step toward identifying the effect of control factors on dimensional deviation is to find goodness of fit of the given data. The quadratic model for dimensional deviation is recommended by the Design expert 6.0 software. [Table 5] shows the analysis of variance (ANOVA) for the quadratic model at 95% confidence level, which is obtained by eliminating the nonsignificant terms by backward elimination.
|Table 5: ANOVA for response surface of the reduced quadratic Model of Dimensional Deviation|
Click here to view
It indicates that the F-value of the model is 39.71 and coresponding P value is less than 0.001, indicating that the quadratic model is significant at the 95% confidence level. Moreover, the lack of fit of 1.67, implies that it is not significant relative to pure error. R 2 of 0.8904 [Table 6] indicates that 89.04% of the variation of dimensional deviation is attributed to the control factors, and only 10.96% of the total variation cannot be explained by the quadratic model. This indicates that the accuracy and general ability of the polynomial model is good. Moreover, the predicted R 2 of 0.8329 is in reasonable agreement with the adjusted R 2 of 0.868, which indicates a high correlation between the observed and predicted values. [Figure 4] shows the normal probability plot of the residuals for dimensional deviation. It shows that most of the residuals are clustered around a straight line, which indicates that errors are normally distributed . It is observed that the regression model is fairly well-fitted with the observed values. As the adequate precision is 25.323, it suggests that the quadratic model can be used to navigate in the design space.
|Figure 4: Normal probability plot of residuals for dimensional deviation|
Click here to view
The following equation (3) represents the relation between the dimensional deviation and control factors, which is obtained by applying the multiple regression technique.
Equation 3 shows that the main effects of TON, TOFF, quadratic effects of TON, IP and SV, and the interaction effect of TON and SV, as well as, TON and TOFF, have significant effects on dimensional deviation, and can be used to predict dimensional deviation within the limits of the control factors. Although [Table 5] shows that Peak current (IP) as a main effect is insignificant, as a quadratic effect it is significant. Hence, it is included in the analysis, to obey the principle of hierarchy.
[Figure 5] shows the interaction effect of TON and TOFF on dimensional deviation. It indicates that with an increase in pulse on time from 112 to 118, when TOFF is at 48, dimensional deviation increases from 180.98 μm to 266.771 μm. With an increase in the pulse off time from 48 to 56, when TON is at 118, the dimensional deviation reduces from 266.7 to 213.3 μm. A higher TON, together with a lower value of TOFF, leads to the release of a large amount of discharge energy on the workpiece surface, resulting in the formation of large craters, which increase the dimensional deviation.
|Figure 5: Interaction effect of Pulse on time and Pulse off time on dimensional deviation|
Click here to view
[Figure 6] shows the interaction effect of TON and SV on dimensional deviation. It indicates that at small values of TON (112) there is a slight decrease in dimensional deviation from 163.7 μm to 159.5 μm, as SV increases from 35 to 55. At a higher value of TON (118), the dimensional deviation decreases from 244.56 μm to 217.09 μm, with an increase in SV from 35 V to 55 V. This is because, a larger value of SV increases the gap between two sparks, thereby producing less number of sparks per unit time, which causes less discharge energy in a given time, which decreases the dimensional deviation due to the formation of shallow craters on the workpiece surface.
|Figure 6: Interaction effect of Pulse on time and Spark gap set voltage on dimensional deviation|
Click here to view
[Figure 7], [Figure 8], [Figure 9] and [Figure 10] show the effect of the pulse on time, pulse off time, spark gap set voltage, and peak current, on dimensional deviation. It is indicated from the figures that the main effects follow the same trend as discussed in the case of interactions. [Table 5] shows that the quadratic effect of the peak current is significant, but its effect is very less, as indicated by a larger P value (0.033), which is closer to 0.05, at which point a factor becomes insignificant.
| 4. Optimization of Process Parameters Using the Desirability Approach|| |
In the desirability function approach, the measured properties of each predicted response are transformed to a dimensionless desirability value, d, where d varies between 0 and 1. The desirability function value 0 suggests that the response is completely unacceptable and d=1 indicates that the response is exactly at the target value. The value of d increases as the desirability of the corresponding response increases . One-sided transformation is used to transform the response into a desirability value. In the present study, the dimensional deviation is smaller the better characteristics. The response is transformed into d i using the equation given below.
And d=1 for y>L
α represents weight, L and U are selected according to the mathematical model in response surface methodology (RSM).
Optimization is carried out in two steps: (1) Obtaining the desirability for the response dimensional deviation, (2) maximization of desirability and identifying the optimal value. In this approach different solutions are obtained. The solution with the highest desirability is preferred. There are 20 solutions generated for getting the true optimal solution, and the best solution is achieved based on the desirability. [Table 7] shows the 20 solutions that are obtained by using the desirability approach. Most of the solutions have a desirability value equal to 1. As the variation of dimensional deviation is very small, that is, 147.2 μm to 151.9 μm among the solutions having desirability equal to 1, any one of the solutions can be considered as an optimal solution.
|Table 7: Optimal solution for dimensional deviations at different parameter combinations|
Click here to view
Contour graphs are plotted for solution No.1. [Figure 11] shows the variation of desirability with respect to variation of pulse on time and pulse off time. It shows that as the value of the pulse on time increases that of the pulse off time decreases, that is, in the lower right hand corner of the figure, the least value of desirability (0.352) is observed. The highest value of desirability that is equal to 1 is obtained at TON=112.61 and TOFF=55.9. [Figure 12] indicates the variation of dimensional deviation with respect to pulse on time and pulse off time, at optimized parameter settings. The trend similar to that observed in the case of [Figure 11] is observed. It shows that larger values of dimensional deviation are obtained when the pulse on time is more and pulse off time is less. The least value of dimensional deviation (148.38 μm) is obtained when TON=112.61 and TOFF=55.9.
|Figure 11: Variation of desirability with variation in pulse on time and pulse off time|
Click here to view
|Figure 12: Variation of dimensional deviation with variation in pulse on time and pulse off time|
Click here to view
Confirmatory experiments are carried out at the parameter settings corresponding to Experiment Nos. 1, 2, and 10 in [Table 7], to check the validity of the optimization results. [Table 8] shows the results of the confirmatory experiments.
It is observed from [Table 8] that the error between the predicted and experimental results is less than 5%, which confirms the excellent reproducibility of the results.
| 5. Conclusions|| |
The present article investigates the effect of process parameters on MRR and the overcut in WEDM of Ti 6-2-4-2 alloy, using Box-Behnken designs. Using the response surface methodology, the empirical model is developed for the variation of dimensional deviation, with different process parameters of WEDM. Consequently, the developed empirical model is utilized for the optimization of dimensional deviation using the desirability approach. The results of optimization are validated by carrying out confirmatory experiments. From the present study, the following inferences can be drawn.
- The established empirical relation has the potential to evaluate the dimensional deviation in the WEDM of the Ti 6-2-4-2 alloy, within the range of the studied parameters, as the adequacy of the model checked is found to be adequate at 95% confidence level.
- The main effects of TON and TOFF; the quadratic effects of TON, IP, and SV; and the interaction effect of TON and SV, as well as TON and TOFF, have significant effects on dimensional deviation.
- TON and TOFF have emerged as the major factors affecting dimensional deviation, as indicated by the larger F values in the ANOVA analysis. When the value of TON is increased or TOFF is decrerased, the dimensional deviation increases, due to the impingement of the increased spark discharge energy on the workpiece.
| References|| |
|1.||D. K. Aspinwall, T. B. Thoe, and M. L. Wise, "Review on Ultrasonic Machining," International Journal of Machine Tools Manufacture, Vol. 38, No. 4, pp. 239-255, 1998. |
|2.||E. Abele, and B. Frolich, "High speed milling of titanium alloys," Journal of advances in production engineering and management, Vol. 3 pp. 131-138, 2008. |
|3.||Donachie, J. Mathew, Titanium: A technical guide, 2 nd ed., ASM International, OH, USA, 2000. |
|4.||S. Jeelani, "Subsurface plastic deformation in machining 6Al-2Sn-4Zr-2Mo Titanium alloy," Wear, Vol. 85, pp. 121-130, 1983. |
|5.||S. Sarkar, S. Mitra, and B. Bhattacharyya, "Parametric analysis and optimization of wire electrical discharge machining of γ-titanium aluminide alloy," Journal of Material Processing Technology, Vol. 159, No. 3, pp. 286-294, 2005. |
|6.||H. Singh, and R. Khanna, "Parametric optimization of cryogenic-treated D-3 for cutting rate in wire electrical discharge machining," J Eng Technol, Vol. 1, pp. 59-64, 2011. |
|7.||A. Shah, N. A. Mufti, D. Rakwal, and E. Bamberg, "Material removal rate, kerf, and surface roughness of tungsten carbide machined with wire electrical discharge machining," International Journal of Materials Engineering and Performance, Vol. 20, No. 1, pp. 71-76, 2010. |
|8.||M. Sadeghi, H. Razavi, A. Esmaeilzadeh, and F. Kolahan, "Optimization of cutting conditions in WEDM process using regression modelling and Tabu search algorithm," Proc I MechE Vol000 Part B: j. Engineering Manufacture, 2011. |
|9.||A. B. Puri, and B. Bhattacharyya, "Modeling and analysis of wire-tool vibration in wire-cut EDM", Journal of Materials Processing Technology, Vol. 141, pp. 295-301, 2003. |
|10.||A. B. Puri, and B. Bhattacharyya, "An analysis and optimization of the geometrical inaccuracy due to wire lag phenomenon in WEDM," International Journal of Machine Tools and Manufacture, Vol. 43, pp. 151-159, 2003. |
|11.||S. Sarkar, M. Sekh, S. Mitra, and B. Bhattacharyya, "Modeling and optimization of wire electrical discharge machining of γ-tial in trim cutting operation," Journal of Material Processing Technology, Vol. 205, pp. 376-387, 2008. |
|12.||K. Jangra, A. Jain, and S. Grover, "Optimization of multiple-machining characteristics in wire electrical discharge machining of punching die using grey relational analysis", Journal of Scientific and Industrial Research, Vol. 69, pp. 606-612, 2010. |
|13.||ASM Material Data Sheet, (Aerospace Specification Metals Inc), 26/12/2008, www.asm.com |
|14.||M. P. Garg, A. Jain, and G. Bhushan, "Investigation of effect of process parameters in wire-EDM of high temperature Titanium alloy 6-2-4-2," ICAME 2011: Proceedings of 5 th International Conference on advances in Mechanical Engg. NIT Surat, India, pp. 626-630, 2011. |
|15.||R. H. Myer, and D. C. Montgomery, "Response surface methodology," New York: John Wiley and Sons, 1995. |
|16.||P. Gupta, R. Khanna, R. D. Gupta, and N. Sharma, "Effect of process parameters on kerf width in WEDM for HSLA using response surface methodology", J Eng Technol, Vol. 2, pp. 1-6. |
|17.||G. Derringer, and R. Suich, "Simultaneous optimization of several response variables," J Qual Technol, Vol. 12, No. 9, pp. 214-219, 1980. |
| Authors|| |
Mohinder P. Garg is working as Assistant Professor in Mechanical Engineering Department at M. M. Engg. College, Mullana (Ambala).He received his B. Tech in Mechanical Engineering from SLIET, Longowal and M.E. in CAD/CAM and Robotics from Thapar Institute of Engg and Technology, Patiala. He is pursuing Ph D at NATIONAL Institute of Technology, Kurukshetra. He is a life member of ISTE. His areas of interest are Non conventional machining processes, Machining of difficult to machine materials, finite element analysis apart from statistical modelling and analysis.
Ajai Jain is working as Associate Professor in Mechanical Engineering Department at National Institute of Technology, Kurukshetra. He received his B. Tech in Mechanical Engineering from Dayalbagh Educational Institute, Agra, M Tech in Mechanical Engineering from IIT, Roorkee and Ph D from Kurukshetra Universty, Kurukshetra. He has published more than 35 research papers in national, international journals and conferences. His areas of interest are advanced machining processes, scheduling and design of manufacturing systems.
Gian Bhushan is working as Associate Professor in Mechanical Engineering Department at National Institute of Technology, Kurukshetra. He received his B. Tech in Aeronautical Engineering from Punjab Engineering College, Chandigarh. He did M Tech in Mechanical Engineering from Punjab Engineering College, Chandigarh and Ph D from Kurukshetra Universty, Kurukshetra. He has published 40 research papers in national, international journals and conferences. His areas of interest are Tribology, Fluid Engineering and CAE.
[Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5], [Figure 6], [Figure 7], [Figure 8], [Figure 9], [Figure 10], [Figure 11], [Figure 12]
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5], [Table 6], [Table 7], [Table 8]