INVESTIGATIONS ON MATERIAL REMOVAL RATE IN WIRE ELECTRICAL DISCHARGE TURNING (WEDT) USING PULSE TRAIN DATA ANALYSIS

Transcription

1 Journal of Mechatronics and Intelligent Manufacturing ISSN: Volume 2, Issue 1/ Nova Science Publishers, Inc. INVESTIGATIONS ON MATERIAL REMOVAL RATE IN WIRE ELECTRICAL DISCHARGE TURNING (WEDT) USING PULSE TRAIN DATA ANALYSIS V. Janardhan and G. L. Samuel Indian Institute of Technology Madras, Adyar, Chennai, Tamil Nadu , India ABSTRACT In the present work, an experimental setup for Wire Electrical Discharge Turning (WEDT) process is developed. The setup consists of a spindle which is designed to meet the accuracy requirements of the process. The pulse train data is analyzed to investigate the influence of various process parameters like pulse off time, rotational speed and flushing pressure on Material Removal Rate (MRR). A data acquisition system for acquiring the pulse train data during WEDT process is developed. A computer based pulse discrimination algorithm is proposed in the present work for offline analysis of the WEDT characteristics. Using the algorithm the pulses are classified into open, normal (spark), arc and short circuit pulses. Based on the pulse classification results, the MRR for WEDT and 2D profile cutting with Wire Electrical Discharge Machining (WEDM) has been compared by varying different parameters. It is observed that the pulse off time is the most significant parameter influencing MRR. From the pulse classification results, it is observed that the arc pulses increased during WEDT compared to WEDM process. Keywords: Wire Electrical Discharge Turning (WEDT), Pulse Train Data Analysis, Material Removal Rate (MRR). 1. INTRODUCTION Electrical discharge machining (EDM) is a nontraditional machining process widely used to produce dies and moulds. Wire Electrical Discharge Machining (WEDM) is a special form of the EDM process in which the tool is a continuously moving conductive wire electrode. Material is eroded from the workpiece by a series of discrete electrical sparks between the workpiece and the wire electrode (tool) in the precense of a thin film of dielectric fluid which Corresponding author: Phone: , Fax: ; samuelgl@iitm.ac.in

2 18 V. Janardhan and G. L. Samuel is continuously fed to the machining zone to flush away the eroded particles. Due to its wide capabilities, it has applications in various fields such as automobile, aerospace, medical and virtually all areas of conductive materials machining (Spedding and Wang, 1997). Turning with WEDM is one of the emerging areas, developed to generate cylindrical form on hard and difficult to machine materials. A rotary axis is added to the WEDM for generating cylindrical and axisymmetric forms. The idea of using WEDM to machine cylindrical parts has been reported by Masuzawa and Tonshoff (1997), to manufacture small-diameter pins and shafts of diameter 5μm. Machining of automobile components like diesel injector plunger with WEDM has been carried out (Qu. et al., 2002). The application of WEDM for Truing of metal bond diamond wheels used for ceramic grinding is reported by Rhoney et al., (2002). Takayuki et al (2004), machined axisymmetric and arbitrary shapes on insulating ceramics. Qu et al., (2002) compared the Material Removal Rate (MRR) in case of WEDM and cylindrical WEDM (called as WEDT in the present work) and derived a mathematical model for material removal rate for cylindrical WEDM. Mohammadi et al., (2008) studied effect of machining parameters like power, time off, open circuit voltage, servo, wire tension, wire speed and rotational speed on MRR in WEDT and the influence of machining parameters on MRR is mathematically modeled using regression analysis method. Haddad and Tehrani (2008) developed a mathematical relation between the MRR and machining parameters using response surface methodology. Qu et al., (2002) reported that the maximum MRR in cylindrical WEDM is greater than WEDM and mentioned that the possible cause for increase in MRR is due to better flushing conditions. Mohammadi et al., (2008) reported that MRR is inversely proportional to the rotational speed and the same is also confirmed by Haddad and Tehrani (2008). It is observed from the literature that the flushing condition and spindle rotational speed affect the MRR. The material removal process of WEDM involves complex and time-varying characteristics. Accurate analysis of the material removal mechanism is difficult. MRR depends mainly on the discharges occurring at the spark gap which are influenced by the machining parameters. The electric spark discharges are random in nature and depends on the debris condition at the spark gap (Kunieda et al., 2005). A clear understanding of the discharges occurring at the spark gap can be possible with pulse train analysis. Pulse train analysis includes classification of discharges occurring at the spark gap. This analysis is generally used in adaptive control of EDM and WEDM to improve the efficiency. To meet this objective various researchers have followed different strategies. The monitoring of EDM and WEDM has been carried out using different sensing parameters like ignition delay time, gap impedance during pulse interval, average feed rate, frequency of occurrence of arcs and occurrence of various pulse types (Snoeys et al., 1980). Dauw et al., (1983) classified the discharge pulses into fifteen types with the help of voltage and time reference levels. A general layout of the pulse analyzing system is mentioned by Jennes et al., (1984). Watanabe et al., (1990) classified the discharge pulses into three categories; normal, deion and arc according to the discharge profile. This method is used for off-line process optimization. Yan and Liao (1996) discriminated the pulses into normal and arc pulses based on ignition delay as a sensing parameter; Liao and Woo (1997) developed a pulse discriminating system to analyze the relationship between the machining settings and pulse trains in WEDM. Tarng et al (1997) classified the pulses in EDM based on the features of the measured gap voltage and gap current by using a fuzzy set

3 Investigations on Material Removal Rate in WEDT 19 theory. Yan and Chien (2007) developed a computer-aided pulse discriminating and control system based on the characteristics of the gap voltage wave form. Classification of different pulse types requires a pulse discriminating system. Pulse discriminating systems are usually designed to improve the efficiency of the WEDM process and to prevent wire breakage by varying the parameters online. Discriminating the pulses based on the voltage waveform data alone requires complicated logic. Pulse discrimination with the help of a computer based algorithm for offline analysis of the process using voltage and current wave form data has not been mentioned in the literature. In the present work a spindle is fabricated to meet the requirements of the Wire Electrical Discharge Turning (WEDT) process. The pulse train data is collected with the help of a pulse train data acquisition system developed. A pulse discrimination algorithm has been proposed based on the voltage, current and time data obtained from data acquisition system. With the pulse discrimination algorithm the pulses are classified into four types; open, normal (spark), arc and short circuit pulses for off line analysis of the WEDT characteristics with various process parameters. In the present work the variation in MRR with pulse of time, flushing pressure and rotational speed for WEDT is investigated using the pulse classification results and compared with the MRR of WEDM. 2. SPINDLE FOR WEDT PROCESS Precision spindle is the key subsystem for WEDT process as it has to provide accurate rotary motion to the workpiece. The components of the subsystem need to have good corrosion resistance as they come in contact with dielectric fluid. It should have good electrical connectivity for the flow of high electrical current from the rotating workpiece to the ground to enable spark to occur. The spindle should have high rotational accuracy, runout less than 10 μm (Qu et al. 2002), to maintain the gap between the workpiece and wire consistently. The spindle should also be capable of holding workpieces of different diameters and should rotate it at different speeds. The spindle consists of components such as DC motor, spindle shaft, bearings, keyless drill chuck, housing and an L-section that acts as a fixture. The speed of the DC motor is controlled with a regulated DC power supply. A double start helical thread is provided on the motor shaft and is mated with a gear. The gear is made up of plastic to provide insulation to the motor. With this gear arrangement, the speed of the workpiece can be varied from 4 to100 rpm. The spindle shaft is machined from SAE 316 steel and it is shown in Figure 1(a). It has provision for attaching drill chuck, gear, locknut and mounting the bearings as shown in Figure 1(a). Two angular contact ball bearings are used for holding the spindle as they are generally preferred for precision machine tool spindles. Housing is made of aluminum and its cross sectional view is shown in Figure 1(b). Spindle shaft and bearings are assembled into the housing and locknuts are used for preventing the motion of the bearings as shown in Figure 1(c). The bearing assembly and motor are fixed to the L-section. An acrylic cover with an oil seal is used for protecting the components of the spindle assembly during dielectric flushing. A schematic diagram and the actual photograph of assembled spindle with various components are shown in Figures 1(d) and 1(e). After the assembly the spindle run out is

4 20 V. Janardhan and G. L. Samuel measured using a capacitance sensor with a resolution of 5 ηm and it is found that the run out for the current set-up is 10 µm. (a) Spindle shaft (b) Cross sectional view of housing (c) Bearing assembly (d) Schematic diagram of spindle assembly (top view)

5 Investigations on Material Removal Rate in WEDT 21 (e) Photograph of assembled spindle for WEDT Figure 1. Assembly of various components of spindle for WEDT. 3. PULSE TRAIN DATA ACQUISITION AND ANALYSIS Pulse train data includes voltage and current waveforms across the spark gap during machining. The variation in the discharges occurring at the spark gap due to the addition of rotary axis is studied using pulse train analysis. Figure 2. Block diagram of pulse train data acquisition system.

6 22 V. Janardhan and G. L. Samuel 3.1. Pulse Train Data Acquisition A block diagram of the pulse train data acquisition system to obtain the spark gap voltage and current waveforms is shown in Figure 2. Current waveform is obtained by measuring the voltage drop across 75mV/75A shunt (resistance 1mΩ). The idea of using electrical shunt for measuring gap current is reported by Murthy and Philip (1987), Gangadhar et al (1992). Cathode ray oscilloscope probes are used to acquire voltage and current waveforms. Pulse train data is collected during the machining for a duration of 10 ms at different parameter settings Pulse Classification Algorithm Voltage and current pulse train data representing different pulse types is shown in Figure 3. The algorithm proposed for classifying the pulses is based on the detection of threshold level crossing points for the gap voltage and current. Time (s) Figure 3. Voltage and current threshold crossing points with different pulse types. Pulses are classified into normal pulse, open circuit pulse, non ignition delay pulses. Non ignition delay pulses are further classified into arc and short pulses. Steps involved in classifying the pulses are given as a flow chart shown in Figure Step 1. Threshold Values For Voltage (V t ) and Current (I t ) and Detection of Crossing Points The voltage, current and time data obtained from data acquisition system is input to algorithm. Threshold values for voltage (V t ) and current (I t ) are fixed as 80V and 20A. Whenever the voltage crosses the value V t, the sampling points prior and after the threshold

7 Investigations on Material Removal Rate in WEDT 23 level are noted as the threshold crossing points. Threshold crossing points are noted in the rising and falling direction indicated by up and down arrows in the Figure 5. Figure 4. Flow chart for proposed algorithm to classify pulses. Figure 5. Continued on next page.

8 24 V. Janardhan and G. L. Samuel Figure 5. Threshold values for voltage (V t ) and current (I t ) and detection of crossing points. The sampling points prior and after the threshold crossing level in the rising direction for the i th voltage pulse are stored as k i, l i. Similarly in the falling direction the values are stored as m i, n i. A similar criterion is followed in finding the current threshold level crossing points. The locations of the current threshold crossing points in the rising direction are stored as o i, p i and that of the falling direction as q i and r i as shown in the current wave form of the Figure 5. Figure 6. Sampling time values for threshold level crossing points Step 2: Sampling Time Values for Threshold Crossing Points The sampling times for voltage (k i, l i, m i, and n i ) and current threshold points (o i ', p' i, q' i, and r' i ) are determined from the data input. Magnitude of the voltage corresponding to the

9 Investigations on Material Removal Rate in WEDT 25 current threshold crossing points o i, p i, q i, and r i is determined using the sampling time and it is represented as o' i, p' i, q' i, and r' i in the voltage wave form as shown in the Figure 6. The sampling time and voltage corresponding to the sampling points o' i, p' i, q' i, and r' i are used in classifying the pulses. It can be noticed from Figure 6 that the sampling times corresponding to o i and o' i and p i and p' i etc, are same. The classification of pulses is based on the time interval between the sampling point n i and o' i. The sampling time corresponding to n i is noted as t v and corresponding to o' i is t c Step 3: Classification of Pulses Based on the Pulse Classification Time Interval (t pc ) The pulse classification time interval (t pc ) is definedd as the time difference between t c and t v v. Pulse classification time interval is determined using the sampling times obtained from step.2. Normal pulses: In case of normal pulse, the current discharge occurs after the ignition delay time. For this case the voltage threshold crossing point (n i ) and corresponding current threshold point (o' i ) occurs within the discharge duration. The time interval t pc should be less than the discharge duration (t f ) as shown in the Figure 7. Figure 7. Normal pulse or spark. Open circuit pulses: In case of open circuit pulse, discharge current does not exist and there will not be current threshold crossing points corresponding to the pulse. For this case the pulse classification time interval t pc is calculated with respect to current threshold crossing point of successive pulse as shown in Figure 8. Time interval between the voltage and current threshold crossing point will be more than the pulse off time (t o ) as shown in the Figure 8. Non Ignition delay pulses: In case of non ignition delay pulse, voltage threshold crossing points does not occur. Theree are two cases in the occurrence of non ignition delay pulses. In the first case, the non ignition delay pulse occurs with an ignitionn delay pulses (open or normal) after it. In this case, pulse classification time interval (t pc ) will become negative, since the occurrence of current threshold crossing point (o' i ) of corresponding non ignition delay pulse is earlier than the voltage threshold crossing point (n i ) as represented in Figure 9(a).

10 26 V. Janardhan and G. L. Samuel Figure 8. Open circuit pulse. (a ) Occurrence of non ignition delay pulse (Case1). (b) ccurrence of nonn ignition delay pulse (Case 2) Figure 9. Classification of non ignition delay pulse.

11 Investigations on Material Removal Rate in WEDT 27 In the second case, there is no ignition delay pulse after the corresponding non ignition delay pulse. In this case the voltage threshold value of the ignition delay pulse prior to the non ignition delay pulse under consideration, is considered for calculating t pc. The time interval will be more than the pulse off time (t o ) as shown in Figure 9 (b). Non ignition delay pulses identified using the above criteria are further classified in to short and arc pulses based peak voltage during the non ignition delay pulses. The threshold voltagefor classifying the different non ignition delay pulses (V a ) is chosen as 20V. If peak voltage is more than V a, it is arc and if it is less then the pulse is considered as short. 4. EXPERIMENTS FOR ANALYSIS OF MATERIAL REMOVAL RATE IN WEDT PROCESS Experiments are conducted to study the effect of process parameters on MRR. The experimental setup developed for the WEDT process is shown in Figure 10. The process parameters such as pulse off time, spindle rotational speed, and flushing pressure are varied to study their effect on MRR in WEDT using pulse train analysis. Each parameter is varied at 3- levels keeping the other two parameters constant. The pulse off time is chosen as 30, 35 and 42 μs. Flushing pressure is varied at three levels and it is mentioned as 1 (lower), 2 (medium), 3 (high). Rotational speeds used are 40, 70 and 100 rpm. The process constants used for conducting the experiments for WEDT are listed in Table 1. The material used for conducting the experiments is brass as it has good electricl conductivity. The diameter of the workpiece used for conducting WEDT experiments is 10mm. Figure 10. Experimental setup of WEDT.

12 28 V. Janardhan and G. L. Samuel Table 1. Process constants used during WEDT Parameter Work piece Material Wire material Depth of cut Diameter of the specimen Wire speed Wire diameter Open gap voltage Specification Brass Brass 0.3 mm 10 mm 4 m/min 0.25 mm 100 V Qu et al (2002) reported that MRR in WEDT is more compared to that of WEDM. Though the tool material, pulse parameters and mechanism of material removal is same for both the configurations a variation in MRR occurred in both the configurations. The variation in MRR can be better understood by pulse train analysis. To study this effect MRR for both the configurations is compared and the results are analyzed with pulse train analysis. 2D profile cutting experiments were conducted on a brass plate of 12 mm thickness. The parameter settings used for conducting the experiments in case of WEDT and WEDM are same. Since the workpiece is fixed to the table in case of WEDM, the experiments conducted considering the variation in rotational speed in WEDT cannot be compared with WEDM for MRR analysis. Considering this, the effect of pulse off time and flushing pressure are studied keeping the rotational speed constant at 40 rpm. 5. RESULTS AND DISCUSSIONS 5.1. Effect of Pulse Off Time on WEDT Performance Pulse train data is collected at different pulse off times 30, 35 and 42 μs for WEDT and WEDM. While varying pulse off time the rotational speed is set at 40 rpm and flushing pressure is set at medium level (level2). The proposed pulse classification algorithm is applied for classifying the pulses. Figure 11 shows the pulse classification results for WEDT and WEDM at 30 μs pulse off time. Arc and spark regions in the pulse train data is represented by a rectangular wave form as shown in Figure 11 (a). The peak portion of the rectangular wave form represents the arc region and valley portion represent the spark region. It was observed that the number of arc pulses in an arc region are more in case WEDT than WEDM as shown in Figures 11 (a) and (b). The possible cause for this is rotation of the workpiece. When discharge occurs, the material is removed from the rotating workpiece and the wire is fed forward during which the unmachined portion of the workpiece faces the wire and the spark gap reduces. This phenomenon increases the formation of arcs. Because of this at a constant pulse off time the

13 Investigations on Material Removal Rate in WEDT 29 arcs in the arc region are more in WEDT than WEDM and the the same was also observed in pulse train data collected at pulse off times 35 and 42 μs as shown in Figures 12 and 13. Number of sparks= 72; Number of arcs= 130; Arc regions= 22. Gap voltge (V) Arc region Time (s) (a) Pulse train data for WEDT Process Spark region Number of sparks=147 ; Number of arcs= 19; Number of short pulses= 5; arc regions=18. Gap voltge (V) Time (s) (b) Pulse train data for WEDM Process Figure 11. Comparison of pulse classification results for WEDT and WEDM at the pulse of time 30 μs. Number of sparks = 80; Number of arcs= 80; Arc regions= 18 Gap voltge (V) Time (s) (a) Pulse train data for WEDT Process Number of sparks= 119; Number of arcs= 24; Number of short pulses= 2; Arc regions=17. Gap voltge (V) Time (s) (b) Pulse train data for WEDM Process Figure 12. Comparison of pulse classification results for WEDT and WEDM at the pulse of time 35 μs.

14 30 V. Janardhan and G. L. Samuel With increase in pulse off time the number of arc regions reduced for both WEDT and WEDM as shown in the Figures 12 and 13. This is due to the increase in the deionization time available for charge carriers formed during discharge with increase in pulse off time. Number of sparks= 61; Number of arcs=74 ; Arc regions= 14. Gap voltge (V) Time (s) (a) Pulse train data for WEDT process Number of sparks=109; Number of arcs=12; Arc regions=10. Gap voltge (V) Time (s) (b) Pulse train data for WEDM process Figure 13. Comparison of pulse classification results for WEDT and WEDM at the pulse of time 42 μs Effect of Pulse Off Time on Material Removal Rate Tables 2 and 3 present the experimental results of pulse classification and material removal rate for WEDT and WEDM. The material removal rate for WEDT is calculated using the Equation below where D 1 is the initial diameter of the workpiece, D 2 is the final diameter of the workpiece and V f is the wire feed rate. The material removal rate for WEDM is calculated by multiplying thickness of the plate, width of cut and the wire feed rate.

15 Investigations on Material Removal Rate in WEDT 31 Table 2. Experimental results to study the effect of machining parameters on MRR and pulse trains in WEDT S.No Rotatonal speed (rpm) flushing level pulse off time (μs) Fluctuaion in Wire feed rate (mm/min) Average voltage (V) Percentage of normal pulses P n (%) Percentage of arc pulses P a (%) Total number of pulses Max. MRR (mm 3 / min) Table 3. Experimental results to study the effect of machining parameters on MRR and pulse trains in WEDM S.No Flushing level Pulse off time (μs) Fluctuation in wire feed rate (mm/min) Fluctuation in average voltage (V) Percentage of normal pulses P n (%) Percentage of arc pulses P a (%) Percentage of short pulses P s (%) Total number of pulses Max. MRR (mm 3 / min) The variation in maximum MRR with pulse off time is shown in Figure 14. MRR with pulse off time is measured keeping the flushing pressure levels constant at 1(lower), 2(medium) and 3(higher) for both WEDT and WEDM. The experimental results for WEDT,

16 32 V. Janardhan and G. L. Samuel shown in Figure 14, were conducted keeping rotational speed constant at 40 rpm. MRR increased with decrease in the pulse off time, however it increases with increase in flushing pressure in WEDT as shown in Figure 14. Figure 14. Effect of pulse off time on MRR in case of WEDT and WEDM. It can be observed from the Table 3 that, the variation in maximum material removal rate with pulse off time at different flushing levels did not change considerably in case of WEDM hence is represented by a dotted line in Figure 14. It is observed from Figure 14 that the maximum MRRR decreased with increase in pulse off time for both WEDT and WEDM. It is due to the decrease in the total number of pulses as indicated in Figure 15(a). It is also observed from the Figure 14 that, for the same pulse off time MRR in case of WEDT is less compared to that of WEDM. This can be attributed to the decrease in percentagee of normal pulses and increase in the percentage of arc pulses in case of WEDT than WEDM as shown in Figures 15(b) and (c). Total number of pulses WEDT 2 WEDT 3 WEDT 1 WEDM 2 WEDM 3 WEDM Pulse off time (μs) (a) Total number of pulses 42

17 Investigations on Material Removal Rate in WEDT 33 Percentage of normal pulses Pulse off time (μm) 1 WEDT 2 WEDT 3 WEDT 1 WEDM 2 WEDM 3 WEDM Percentage of arc pulses (b) Normal pulses Pulse off time (μs) (c) Arc pulses 1 WEDT 2 WEDT 3 WEDT 1 WEDM 2 WEDM 3 WEDM Figure 15. Comparison of pulses at various pulse off times for WEDT and WEDM. Arc pulses reduce the MRR and normal (spark) pulses increase the MRR. (Liao et al (1997), Murthy and Philip (1987), Gangadhar et al., (1992)). It is observed that MRR in case of WEDT is more at higher flushing levels due to the increase in the percentage of normal pulses as shown in Figure 15(b) Effect of Flushing Pressure on Material Removal Rate To study the effect of flushing pressure, experiments were conducted at different pressure levels i.e. level1 (lower), level2 (medium) and level3 (higher). The effect of flushing pressure is studied keeping the pulse off time constant at 30, 35 and 42 μs respectively. The rotational speed is kept constant at 40 rpm during WEDT. A comparison of MRR between WEDT and WEDM is shown in Figure 16. It has been observed that by increasing the flushing rate, MRR has not changed considerably in case of WEDM and increased moderately in WEDT process.

18 34 V. Janardhan and G. L. Samuel Figure 16. Effect of Flushing level on material removal rate in case of WEDT and WEDM. The pulse classification results with flushing levell are shown in Figure 17. Though the total number of pulses decreased with increase in flushing level in WEDT as shown in Figure 17(a), the MRRR increased. This is due to the slight increase in the percentage of normal pulses and decrease in percentage of arc pulses with increase in flushing level in case of WEDT as shown in the Figure 17(b) and (c). The arc pulses reducee with increasee in flushing level due to better debris condition at the spark gap. It is seen from Figure 17(c) that in case of WEDT at pulse off time 30 μs, the percentage of arc pulses increased when the flushing level is increased from level1 to level2. This is due the effect of misalignment in the workpiece with spindle axis, which is confirmed by the severe fluctuation in the average voltage from 20 to V during experiment 4 as shown in Table 2. Misalignment in the workpiece causes severe variation in the spark gap and results in more number of arc pulses when the spark gap reduces. 220 Total number of pulses μs WEDT 35 μs WEDT 42 μs WEDM 30 μs WEDM 35 μs WEDM 42 μs WEDM Flushing level (a) Total number of pulses

19 Investigations on Material Removal Rate in WEDT 35 Percentage of normal pulses Flushing level 30 μs WEDT 35 μs WEDT 42 μs WEDT 30 μs WEDM 35 μs WEDM 42 μs WEDM (b) Normal pulses Percentage of arc pulses Flushing level (c) Arc pulses 30 μs WEDT 35 μs WEDT 42 μs WEDT 30 μs WEDM 35 μs WEDM 42 μs WEDM Figure 17. Comparison of pulse classification results at different flushing pressure levels for WEDT and WEDM. The same experiment is repeated and represented as Experiment 10 in Table 2 in which the voltage fluctuation is 40 to V. MRR is increased from 9.34 to 11.01mm 3 /min due to accurate rotation of the workpiece. This corrected result has been used for further analysis in the following sections Effect of Rotational Speed on Material Removal Rate The rotational speed is varied at 40, 70 and100 rpm keeping the pulse off time constant at 30, 35 and 42 μs. The flushing pressure is kept at level 2 for all these experiments. The variation in MRR with rotational speed is represented graphically in Figure 18. The MRR reduced slightly when the rotational speed is increased as represented in Figure 18, at

20 36 V. Janardhan and G. L. Samuel pulse off times 30 and 42 μs. It is observed that the MRR reduced with increase in rotational speed as the possibility of occurrence of arc pulses is more at lower pulse off time. Figure 18. Effect of rotational speed on material removal rate. The pulse classification results with rotational speed for WEDT are shown in Figure 19. The percentage of normal pulses reduced and arc pulses increased with increase in rotational speed as shown in Figures 19 (a) and (b). As the rotational speed increases circumferential length of the material crossing the sparking region increases but the energy input is same. Consequently, either the volume of the material removed per unit length decreases or the depth of the material removed at the sparking region reduces. If the depth of the material removed is less the gap been the wire and workpiece decreases and the chances of forming arcs increase. The variation in the total number of pulses is as shown in Figure 19 (c). The total number of pulses increased with increase in rotational speed from 40 to 70 rpm at pulse off times 30 and 42 μs; however, the MRR is reduced, as the arc pulses are more when the rotational speed is increased. Percentage of normal pulses μs 35 μs 42 μs Rotational speed (rpm) (a) Normal pulses

21 Investigations on Material Removal Rate in WEDT 37 Percentage of arc pulses Rotational speed (rpm) 30 μs 35 μs 42 μs (b) Arc pulses Total number of pulses μs 35 μs 42 μs Rotational speed (rpm) (c) Total number of pulses Figure 19. Effect of rotational speed on different types discharge pulses in WEDT. CONCLUSIONS In the present work, a precision spindle with a runout of 10µm is fabricated to machine cylindrical components using WEDM. A pulse train data acquisition system for WEDT is developed for acquiring voltage and current waveform data. An algorithm for classifying the open circuit, normal (spark), arc and short circuit pulses has been developed for offline analysis of the WEDT process. Detailed investigations were carried out to study the influence of pulse off time, flushing pressure and rotational speed on MRR during WEDT and the

22 38 V. Janardhan and G. L. Samuel results were compared with WEDM. It is observed that pulse off time is the most significant parameter for increasing the MRR for both WEDT and WEDM. MRR is less in case of WEDT compared to WEDM since the percentage of arc pulses are more and normal pulses are less in WEDT than in WEDM. MRR increased with decrease in the pulse off time. Variation in MRR occurred with constant pulse off time at different flushing levels in WEDM, but it increased with increase in flushing pressure in WEDT. MRR Increased to some extent with increase in flushing pressure in WEDT due to reduction in the percentage of arc pulses and increase in the percentage of normal pulses however it is constant for WEDM. The MRR reduces with increase in rotational speed; the reduction in MRR is more at lower pulse off time since the occurrence of arc pulses is more. From the pulse train data collected for WEDT and WEDM it is observed that number of arcs in the arc regions are more in WEDT than WEDM due to rotation of workpiece. From the investigations it is recommended to have a combination of lower pulse off time, higher flushing level and lower rotational speed to achieve high MRR. From the investigations carried out by the authors, it is noticed that MRR is reduced slightly with increase rotational speed and it has significant influence on occurrence of arc pulses. Occurrence of arc pulses may affect the quality of the machined part. Detailed investigations need to be carried out to verify the influence of rotational speed on surface integrity. REFERENCES Dauw, D.F, Snoeys, R and Dekeyser, W (1983) Advanced pulse discriminating system for EDM process and control, Annals of the CIRP, 32 (2), Gangadhar, A, Shunmugam, M.S and Philip, PK (1992) Pulse train studies in EDM with controlled pulse relaxation, International Journal Machine Tools and Manufacture, 32(5), Haddad, M.J and Tehrani, A.F (2008) Material removal rate (MRR) study in the cylindrical wire electrical discharge turning (CWEDT) process, Journal of Materials Processing Technology, 199, Jennes, M., Snoeys, R and Dekeyser, W (1984) Comparison of various approaches to model the thermal load on the EDM-wire electrode, Annals of the CIRP, 33(1), Kunieda, M., Lauwers, B., Rajurkar, K.P and Schumacher, BM (2005) Advancing EDM through Fundamental Insight into the Process, Annals of the CIRP, 54(2), Liao, Y.S and Woo, J.C (1997) The effects of machining settings on the behavior of pulse trains in the WEDM process, Journal of Materials Processing Technology, 71 (3), Liao, Y.S., Chu, Y.Y and Yan, M.T (1997) Study of wire breaking process and monitoring of WEDM, International Journal Machine Tools and Manufacture, 37(4), Masuzawa, T. and Tonshoff, H.K (1997) Three-Dimensional Micromachining by Machine Tools, Annals of the CIRP, 46, Mohmmadhi, A., Alireza, F.T., Ehsan, E and Davoud, K (2008) Statistical analysis of wire electrical discharge turning on material removal rate, Journal of Material Processing Technology, 205,

23 Investigations on Material Removal Rate in WEDT 39 Murthy, V.S.R and Philip P.K (1987) Pulse Train analysis in ultrasonic assisted EDM, International Journal of Machine Tools and Manufacture, 27(4), Qu, J., Shih, A.J and Scattergood, R.O (2002) Development of the cylindrical wire electrical discharge machining process: Part1: Concept, design, and material removal rate, Journal of Manufacturing Science and Engineering, 124 (3), Rhoney, B.K., Shih, A., Scattergood, R., Akemon, J., Gust, D and Grant, M (2002) Wire electrical discharge machining of metal bond diamond wheels for ceramic grinding, International Journal of Machine Tools and Manufacture, 42, Snoeys, R., Dauw, D and Kruth, J.P ( 1980) Improved adaptive control system for EDM processes, Annals of the CIRP, 29(1), Spedding, T.A and Wang, Z.Q (1997) Study on modelling of wire EDM process, Journal of Materials Processing Technology, 69, Takayuki, T., Yasushi, F., Naotake, M., Nagao, S and Masaaki, O (2004) Machining phenomena in WEDM of insulating ceramics, Journal of Materials Processing Technology, 149, Tarng, Y.S., Tseng, C.M and Chunng, L.K (1997) A fuzzy pulse discriminating system for electrical discharge machining, International journal of machine tools and manufacture, 37(4), Watanabe, H., Wang, T and Suzuki, I (1990) WEDM monitoring with a statistical pulse classification method, Annals of the CIRP, 39 (1), Yan, M.T and Chien, H.T (2007) Monitoring and control of micro Wire-EDM process, International journal of machine tools and manufacture, 47, Yan, M.T and Liao, Y.S (1996) Monitoring and self-learning fuzzy control for wire rupture prevention in wire electrical discharge machining, International Journal of Machine Tools and Manufacture, 36(3),

24 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.