CHAPTER 4 GRINDING FORCE MEASUREMENT

74 CHAPTER 4 GRINDING FORCE MEASUREMENT 4.1 INTRODUCTION It is practically difficult to adequately represent the grinding process by a system of equations based on physical reasoning. The random shapes of the abrasive grains coupled with the complex nature of the process itself, introduce random influences which affect the entire behaviour of the system. Dynamometers are readily available and usable to measure the cutting forces in turning, milling, drilling and surface grinding. There are certain problems which are peculiar to cylindrical grinding operation. Due to rotation of both the wheel and workpiece there are difficulties in physically locating the dynamometer. This study deals with the development of a setup for the measurement of grinding force. The main objective of the work is to measure the grinding force and to evaluate the effect of grinding parameters on the grinding forces. The details of the force measurement and the variation of grinding force with the grinding parameters are discussed in the following section. 4.2 MEASUREMENT OF GRINDING FORCES The workpiece was prepared to the dimensions as shown in Figure 3.1. The selected grinding wheel (A60L5V10) was fixed to grind the workpiece. The normal dead center in the tailstock was replaced with the dead center mounted with the strain gauges. The calibration setup was

75 installed in the machine. The 50mm grinding length is divided into two portions with a length of 25mm each. The electrical output signal obtained by placing various tangential and normal loads as given in Figure 3.9 and Figure 3.10 and the calibration curve was drawn with the output voltage and the load. The measurements were repeated five times and readings were recorded. The experiments were conducted by changing the machining parameters such as work speed, traverse feed and depth of cut. Before every grinding experiment, a dressing was carried out. A single point diamond dresser was used for the dressing of the grinding wheel. Throughout the experiment, the wheel velocity was kept at 33.6 m/sec and Procut oil was used as a coolant. The output signals of the tangential and the normal grinding force were received from the corresponding wheatstone bridge simultaneously and are stored in the computer through Agilent multimeter (34401A and U1233A ). It is designed to receive the signal continuously, from the moment when one third of the grinding wheel is in contact with the workpiece till one-third of wheel is out of contact of the workpiece. The samples were received at a rate of 1 khz for a single pass of the grinding wheel. The stored data s were processed in the MATLAB. The variation of current in the grinding wheel motor during the grinding process was also observed using the current probe (Agilent 1146A). The net grinding power (P) drawn during the grinding process was obtained as discussed in Chapter 3. 4.2.1 Measurement of Tangential Grinding Force The experiments were carried out by varying the grinding parameters, the output signal of the tangential grinding force and the grinding power were recorded. The details of the instruments used to record the grinding output signals are explained in Chapter 3.

76 4.2.1.1 Strain gauge based experimental setup Figure 4.1 presents the typical tangential force signals obtained while grinding a workpiece. It is observed that, in the region A, the magnitude of the tangential force output signal gradually increases, and reaches its maximum value when the wheel is in full contact with the workpiece. In the region B, the magnitude of the signal slowly decreases till the end of full engagement between the wheel and the workpiece. This is due to the fact that, the grinding wheel (load) moves away from the sensor location. After that, since the wheel leaves the workpiece, the magnitude of the output signal decreased rapidly in the region C. The output voltages were taken, when the wheel makes full contact with the workpiece i.e., positions 1 and 2. The force values corresponding to the output voltages were taken from the calibration curve as shown in Figure 3.9. The average of these two values was considered as the grinding force. The Figure 4.1 (a) shows the tangential grinding force signal for a specific grinding condition (work speed-0.36m/s, depth of cut-0.01mm and traverse feed- 0.1m/min). It is observed that, in the region B, the full engagement of the wheel and the workpiece (position 1) was established at the 10 th second and the corresponding output voltage is 0.13V. Full engagement continues for the period of next 15 seconds. During this period, the output voltage decreases slowly because the wheel moves away from the sensor location. At the 25 th second, the output voltage was 0.1V, there after the wheel starts leaving the workpiece. The decrease in the output voltage for the period of 25 th second to 30 th second is due to the fact that, the wheel leaves from the grinding area, resulting in the continuous decrease in the involvement of a number of active abrasive grits. In this case, the calibration curve shows that the tangential force is 19.4N.

77 (a) (b) (c) Figure 4.1 Output signal for tangential force with respect to time.

78 Figure 4.1 (b) shows the tangential force signal for a work speed 0.36m/s, depth of cut 0.03mm and traverse feed 0.2m/min. Since the traverse feed is 0.2m/min, the wheel engages the workpiece for a period of 15 seconds during the feed forward motion in which full engagement was established for a period of 7.5 seconds. During this condition, the tangential force is found to be 27.7N. Figure 4.1 (c) shows the tangential force signal for a work speed 0.36m/s, depth of cut 0.02mm and traverse feed 0.3m/min. Since the traverse feed is 0.3m/min, the wheel engages the workpiece for a period of 10 seconds during the feed forward motion in which full engagement was established for a period of 5 seconds. The calibration curve implies that the tangential force is 33N. 4.2.1.2 Grinding power Figure 4.2 (a) shows a typical net grinding power signal for a work speed of 0.36m/s, depth of cut 0.01mm and traverse feed 0.1m/min. It is observed that, at the beginning of a feed forward motion in the grinding wheel, there is a gradual increment in the power at a very slow rate. The full engagement of the wheel and the workpiece was established at the 10 th sec and continues for the period of next 15 seconds. At region B, the grinding power is almost constant and it is found to be 698Watts. Hence, the tangential grinding force from the power is found to be 20.8 N, considering the wheel velocity as 33.6 m/s. This is due to the fact that the leading edge of the wheel removes uniform portion of the material and the wheel is in full contact with the workpiece. The average of the net power consumed during the region B was used to calculate the tangential grinding force from the equation (3.1). The observed trend is in accordance with the literature presented by Guo et al (2007). In the region C, the grinding power decreases deeply when the wheel leaves the workpiece as shown in Figure 4.2 (a).

79 (a) (b) (c) Figure 4.2 The net grinding power with respect to time.

80 Figure 4.2 (b) shows a typical net grinding power signal for a work speed 0.36m/s, depth of cut 0.03mm and traverse feed 0.2m/min. It is observed that, the full engagement of the wheel and the workpiece was established at the 5 th second and continues for the period of next 7.5 seconds. During this period, the grinding power is almost constant in region B and it is found to be 941Watts. Figure 4.2 (c) shows a typical net grinding power signal for a work speed 0.36m/s, depth of cut 0.02mm and traverse feed 0.3m/min. It is observed that, the full engagement of the wheel and the workpiece was established at 3.33 second and continues for the period of next 5 seconds. During this period, the grinding power is almost constant and it is found to be 1168 Watts. Also, it is observed that, the same trend is followed for all grinding conditions. 4.2.1.3 Comparison of tangential grinding force Experimental Power 40 30 20 19.4 20.8 27.728.4 34.8 33 10 0 1 2 3 Experiment Number Figure 4.3 Comparison of tangential grinding force obtained from experimental setup and the grinding power

81 A comparison of the tangential grinding force measured using the experimental setup and the derived grinding power is shown in Figure 4.3. It is observed that the tangential force measured from the experimental setup and obtained from the grinding power have fair agreement. Hence, the output of the experimental setup was considered for further analysis. 4.2.2 Measurement of Normal Grinding Force Figure 4.4 (a) shows the normal force signal for a work speed 0.36m/s, depth of cut 0.01mm and traverse feed 0.1m/min. It is observed that, the full engagement of the wheel and the workpiece (position 1) was established at the 10 th sec and the corresponding output voltage is 1.724V. Full engagement continues for the period of next 15 seconds. During this period, the output voltage decreases slowly because the wheel moves away from the sensor location. At the 25 th second, the output voltage was 0.839V, there after the wheel starts leaving the workpiece. The decrease in output voltage for the period of 25 th second to 30 th second is due to the fact that, the wheel moves away from the load resulting in the decrease in involvement of a number of active abrasive grits. In this case, the calibration curve shows that the normal force is 63 N. The observed trend is in accordance with the literature presented by Guo et al (2007). Figure 4.4 (b) shows the normal force signal for a work speed 0.36 m/s, depth of cut 0.03mm and traverse feed 0.2m/min. Since the traverse feed is 0.2 m/min, the wheel engages the workpiece for a period of 15 seconds during the feed forward motion in which the full engagement was established for a period of 7.5 seconds. During this condition, the normal force is found to be 78 N. Figure 4.4 (c) shows the normal force signal for a work speed 0.36 m/s, depth of cut 0.02 mm and traverse feed 0.3 m/min. The calibration curve implies that the normal force is 85 N.

82 (a) (b) (c) Figure 4.4 Output signal for normal grinding force with respect to time

83 4.3 THE EFFECT OF GRIDING PARAMETERS ON THE GRINDING FORCES AND SURFACE ROUGHNESS In this study, a total of 27 experiments, each having a differnt combination of the grinding parameters (Work speed-v w, Traverse feed-f and depth of cut-d) was carried out based on full factorial design (3 3 ). Throughout the experiment the wheel speed is kept constant at 33.6 m/sec. The grinding conditions adopted in the present study are shown in Table 4.1. Table 4.1 Grinding conditions and their levels Parameters Levels Low Medium High Work speed, V w (m/s) 0.36 0.47 0.6 Depth of cut, d (mm) 0.01 0.02 0.03 Traverse feed, f (m/min) 0.1 0.2 0.3 The experimental results are tabulated in Table 4.2. The details of the grinding force output voltage, the calibrated force and the surface roughness for each replications are furnished in Table A 1.1 and Table A 1.2. Based on the experimental results, the effect of the cylindrical grinding parameters on the tangential component of grinding force (F t ), normal component of grinding force (F n ) and surface roughness (R a ) were evaluated and the results are presented graphically in Figure 4.5 and 4.7.

84 Table 4.2 Effect of grinding parameters on grinding responses Trial No. Work speed, V w (m/s) Grinding parameters Traverse feed, f (m/min) Depth of cut, d (mm) Tangential force, F t (N) Grinding responses Normal force, F n (N) Surface roughness, R a (µm) 1 0.36 0.1 0.01 19 63 1.17 2 0.36 0.1 0.02 23 67 1.43 3 0.36 0.1 0.03 25 72 1.57 4 0.36 0.2 0.01 23 68 1.57 5 0.36 0.2 0.02 27 73 1.91 6 0.36 0.2 0.03 28 78 1.93 7 0.36 0.3 0.01 25 75 1.8 8 0.36 0.3 0.02 30 78 1.93 9 0.36 0.3 0.03 33 85 2.13 10 0.47 0.1 0.01 24 65 1.4 11 0.47 0.1 0.02 28 74 1.75 12 0.47 0.1 0.03 32 77 1.84 13 0.47 0.2 0.01 26 75 1.7 14 0.47 0.2 0.02 29 77 1.97 15 0.47 0.2 0.03 36 81 2.47 16 0.47 0.3 0.01 30 77 1.8 17 0.47 0.3 0.02 36 93 2.13 18 0.47 0.3 0.03 38 94 2.53 19 0.6 0.1 0.01 26 69 1.57 20 0.6 0.1 0.02 29 76 1.63 21 0.6 0.1 0.03 29 81 2.2 22 0.6 0.2 0.01 27 73 1.63 23 0.6 0.2 0.02 30 86 2.21 24 0.6 0.2 0.03 35 95 2.43 25 0.6 0.3 0.01 32 81 2.38 26 0.6 0.3 0.02 36 93 2.43 27 0.6 0.3 0.03 40 103 2.53

85 Tangential grinding force (F t ) in N 40 30 20 10 V w - 0.36m/s 0 0.01 0.02 0.03 0.04 Dept of Cut (d) in mm (a) Tangential grinding force (F t ) in N 40 30 20 10 0 0.01 0.02 0.03 0.04 Depth of Cut (d) in mm Vw - 0.47 m/s (b) Tangential grinding force (F t ) in N 50 40 30 20 10 0 0.01 0.02 0.03 0.04 Depth of Cut (d) in mm V w - 0.6 m/s (c) f=0.1m/min f=0.2m/min f=0.3m/min Figure 4.5 Variation of tangential grinding forces (F t ) with grinding variables at a constant wheel velocity of 33.6 m/sec.

86 4.3.1 Tangential Grinding Force The measurement of the tangential grinding force is highly essential to analyse the cylindrical grinding performace more effectively. Decrease in the grinding forces results in an improvement in grinding accuracy. The effect of the work speed, traverse feed and depth of cut on tangential force is shown in Figure 4.5. It is observed from the results shown in Figure 4.5 (a), 4.5 (b) and 4.5 (c) that the tangential component of grinding force increases with the increase in the depth of cut. In Figure 4.5 (a), the work speed and the traverse feed are kept at 0.36 m/s and 0.1 m/min respectively, the tangential grinding force increases from 19 N to 25 N with the increase in the depth of cut 0.01 mm to 0.03 mm. The same trend is observed when the work speed and the traverse feed is increased (Hecker 2003 and 2007). When comparing the Figure 4.5 (a) and 4.5 (b), it is observed that the tangential force increases from 19 N to 24 N with an increase in the work speed of 0.36m/s to 0.47m/s, while the traverse feed and depth of cut are kept at 0.1m/min and 0.01mm respectively. The same trend is observed when the traverse feed and depth of cut is increased. It is observed from the Figure 4.5(a) that the tangential force increases from 19 to 25 N when the traversed feed increases from 0.1 m/min to 0.3m/min at 0.01mm depth of cut. Figure 4.5 (a) also implies that the tangential force increases proportionately with the increase in the depth of cut. The same trend is observed in Figure 4.5 (b) and 4.5(c). From Figure 4.5 (a), it is observed that at lower work speed the variation in the tangential force with respect to miminum to maximum of the traverse feed and the depth of cut found to be maximum, i.e., at lower wheel

87 velocity (V w 0.36 m/s) the change in tangential force (F t ) is 73.6% camparing at minimum and maximum of the depth of cut and the traverse feed. At higher work speed (V w 0.36 m/s) the variation found to be minimum (53.8%). From Figure 4.5 (a), 4.5 (b) and 4.5 (c), it is found that there is minmum variation in the tangentail force by varying the traverse feed from minimum to maximum with respect to minimum and maximum conditions in the depth of cut and the work speed. Comparing the traverse feed of 0.1m/min and 0.3 m/min the tangential force variation is only 7.4 % at extreme conditions of the depth of cut and the work speed. It is observed that from Figure 4.5 (a), 4.5 (b) and 4.5 (c) the variation of tangential force at minimum and maximum depth of cut with respect to minimum and maximum conditions in the traverse feed and the work speed found to be 8%. Equivalent chip thickness has been found to be a valuable parameter for easily correlating measured grinding parameters i.e., the grinding force, surface roughness, specific grinding energy, material removal rate (Rowe 2009, De Illio et al 2009). As the depth of cut increases, the removed chip thickness increases which in turn increases the load on each abrasive grain, thereby the tangential force required to remove the material is increased (Saini et al 1985, Fielding and Vickerstaff 1986). Increasing the work speed and traverse feed results in higher average chip thickness and length, and thus an increase of tangential grinding force and grinding power.

88 Normal grinding force (F n ) in N 100 80 60 40 0 0.01 0.02 0.03 0.04 Depth of Cut in mm V w -0.36 m/s (a) Normal grinding force (F n ) in N 100 80 60 40 V w - 0.47 m/s 0 0.01 0.02 0.03 0.04 Depth of Cut in mm (b) Normal grinding force (F n ) in N 120 100 80 60 40 0 0.01 0.02 0.03 0.04 Depth of Cut in mm V w - 0.6 m/s (c) f=0.1m/min f=0.2m/min f=0.3m/min Figure 4.6 Variation of normal grinding force (F n ) with grinding variables at a constant wheel velocity of 33.6 m/sec.

89 4.3.2 Normal Grinding Force The normal grinding force plays an important role in grinding process as it has a strong influence on local contact deflection and the nature of the contact deflection has an important effect on the mechanism of material removal. This means that the normal grinding force is an important quantitative indicator to characterize the mode of material removal in any grinding. It is observed from the results shown in Figure 4.6 (b) that the normal component of grinding force increases with the increase in the depth of cut. The work speed and the traverse feed are kept at 0.47 m/s and at 0.1 m/min respectively, the normal force increases from 65N to 77 N with the increase in the depth of cut from 0.01mm to 0.03mm. The same trend is observed when the work speed and the traverse feed is increased in Figure 4.6 (a) and Figure 4.6 (c). When comparing the Figure 4.6 (a), 4.6 (b) and 4.6 (c), it is observed that the normal force increases from 63 N to 69 N with an increase in the work speed of 0.36 m/s to 0.6 m/s, while the traverse feed and depth of cut are kept at 0.1 m/min and 0.01 mm respectively. At the same time, it is observed that it increases from 72 N to 81N when the depth cut is increased to 0.03 mm. It is observed from the Figure 4.6 (c) that the normal force increases from 69 to 103 N with an increase in the combination of the traversed feed and the depth of cut. The traverse feed has increased from 0.1 m/min to 0.3 m/min and the depth of cut is increased from 0.01 mm to 0.03 mm. In these set of experiment the work speed is kept constant at 0.6 m/s. The same trend was observed in Figure 4.6 (a) and 4.6 (b).

90 From Figure 4.6 it is found increase in dpeth of cut, work speed and traverse feed with resepect to minimum and maximum conditions of other two parameters has a increasing effect on normal force. It is observed that at low work speed, the forces are lower. This is because, at lower work speed i.e., at higher speed ratio, more grains will be involved in removing a given volume of material, and thus the depth of engagement will be low. Hence the normal grinding force is low (Agarwal and Rao 2013). It is observed from the Figure 4.7 that the normal grinding force increased with an increase in depth of cut. The increase in depth of cut causes the maximum undeformed chip thickness to increase. The increase of maximum chip thickness will leads to a higher contact area and hence a higher normal grinding force (Fielding and Vickerstaff 1986). Decreasing the depth of cut and the work speed reduces the chip thickness. Thus, reducing the chip thickness lower the contact area, this in turn, reduces the stress on the abrasive grains and decreases the normal grinding force (Saini 1985). 4.3.3 Surface Roughness The surface finish has been a key issue for the reliable prediction of the grinding performance, and surface roughness is one of the most important parameters in assessing the quality of a ground component. The surface roughness (R a ) of the ground specimens was measured using the Surface roughness testing machine (Surfcoder SE1200). At frequent intervals, the correctness of the measurement was verified using calibration specimen. From Table A 1.2, the range of grinding forces and surface roughness with respect to its mean values in each experimental conditions was found to be less than 5%. The effect of the cylindrical grinding parameters on R a is shown in Figure 4.7.

91 Surface Roughness (R a ) in m 3.00 2.50 2.00 1.50 1.00 0.50 0 0.01 0.02 0.03 0.04 Depth of Cut in mm V w -0.36m/s Surface Roughness (R a ) in µm 3.00 2.50 2.00 1.50 1.00 0.50 (a) V w -0.47m/s 0 0.01 0.02 0.03 0.04 Depth of cut in mm Surface Roughness (R a ) in µm 3.00 2.50 2.00 1.50 1.00 0.50 (b) V w -0.6m/s 0 0.01 0.02 0.03 0.04 Depth of cut in mm (c) f=0.1m/min f=0.2m/min f=0.3m/min Figure 4.7 Variation of surface roughness (R a ) with grinding variables at a constant wheel velocity of 33.6 m/sec.

92 It is observed from the results shown in Figure 4.7 that the values of the surface roughness (R a ) increase with an increase in the work speed (V w ), depth of cut (d) and traverse feed (f). Figure 4.7 (a) shows that the surface roughness increases from 1.17 µm to 2.13 µm, when the depth of cut is increased from 0.01 mm to 0.03 mm and the traverse feed is increased from 0.1 m/min to 0.3 m/min. In these set of experiments, the work speed (V w ) is kept constant at 0.36 m/s. The same trend is observed in Figure 4.7 (b) and 4.7 (c) where the surface roughness increases to 2.53µm when the work speed is increased to 0.6 m/s. The observed trends are in accordance with the literature presented by Hecker and Liang (2003). This is because at lower speed ratios (V s /V w ), that is at constant wheel velocity and higher workpiece velocity, few grains participate in removing a given volume of material, hence the depth of engagements is higher, producing poor surfaces. It is observed from the results shown in Figure 4.7 (a) that the value of surface roughness increases from 1.17 µm to 1.57 µm when the depth of cut is increased from 0.01mm to 0.03mm. In these set of experiment the work speed and the traverse feed are kept constant at 0.36m/s and 0.1m/min respectively. The similar patterns of results were observed when increasing the work speed and the traverse feed. This is due to the fact that increase in the depth of cut causing grain fracture or to pull out there by increasing the grain space and reduces the active number of grains per unit area. This leads to increase in the uncut chip thickness and the cross sectional area of the chip, which gives poor surface roughness. The observed trends are in accordance with the literature presented by Wang et al (2005). It is observed from the results shown in Figure 4.7 (a) that the value of surface roughness increases from 1.17 µm to 2.13 µm with an increase in combination of the depth of cut and the traverse feed. The depth of cut is increased from 0.01 mm to 0.03 mm and the traverse feed is increased from

93 0.1 m/min to 0.3 m/min. In these set of experiment the work speed is kept constant at 0.36 m/s. The same trend is observed in Figure 4.7 (b) and Figure 4.7 (c). Thus it concluded that the increase in the traverse feed gives poor surface finish. This is due to the fact that when the wheel passes the work piece with high feed rates it produces feed marks on the work piece, this leads to the poor surface finish on the work piece. The observed trends are in accordance with the literature presented by Kruszynski and Lajment (2005) and Xu et al (2002). From Figures 4.7 (a), 4.7 (b) and 4.7 (c), the surface roughness values have very significant variation with respect to minimum one parameter compared at minimum and maximum conditions of other two, e.g., at depth of cut 0.01mm surface roughness variation with respect to minimum traverse feed 0.1 m/min and work speed 0.36 m/s, and maximum traverse feed 0.1 m/min and work speed of 0.6 m/s and found to be 103%. In Figures 4.7 (a), 4.7 (b) and 4.7 (c), the minimum value of surface roughness obtained is 1.17 µm at low work speed (V w 0.36 m/s), low depth of cut (d 0.01 mm) and low traverse feed (f 0.1 m/min). The maximum value of surface roughness obtained is 2.67 µm at high work speed (V w - 0.6 m/s), high depth of cut (d- 0.03mm) and high traverse feed (f-0.3 m/min). The results comply with the trends available in the study presented by Kwak et al (2006). 4.4 SUMMARY Grinding was performed by varying the grinding parameters in the cylindrical grinding machine. It was observed that the tangential force measured using the experimental setup matches with the force obtained from the grinding power. The normal force was also measured using the experimental setup. The effect of grinding parameters on the grinding forces and the surface roughness were analyzed.