Frictional behaviour and its influence on quality in centreless grinding

Frictional behaviour and its influence on quality in centreless grinding

147 Wear, 118 (1987) 147 - 160 FICTIONAL BEHAVIOUR AND ITS ~FLUENCE CENTRELESS GRINDING N. G. SUBRAMANYA Department (India) (Received ON QUALITY I...

840KB Sizes 2 Downloads 92 Views

147

Wear, 118 (1987) 147 - 160

FICTIONAL BEHAVIOUR AND ITS ~FLUENCE CENTRELESS GRINDING N. G. SUBRAMANYA

Department (India) (Received

ON QUALITY IN

UDUPA, M. S. SHUNMWGAM and V. RADHAKRISHNAN

of ~ec~~~al July 16, 1986;

~~i~ee~ng,

Indian Institute of Technology, Madras 600 036

revised November 11, 1986; accepted December 17, 1986)

In centreless through feed grinding the quality of the ground parts is influenced by several process variables. Some of these process parameters alter the frictional conditions at the control wheel-workpiece and the grinding wheel-workpiece interfaces which, in turn, influence the axial feed rate and rate of rotation of the workpiece. In addition, the type of cutting fluid plays an important role in de~~~~ the friction at the control wheelworkpiece interface. In this paper, an attempt is made to describe the influence of some of the process parameters on the slip in the axial movement and the change in the rate of rotation of the workpiece. Also the influence of different cutting fluids, which alters the nature of the frictional contact at the control wheel-workpiece interface, on the workpiece movements and quality of the ground parts is described. The experimental work carried out, in connection with this, is included. Further, from the analysis of the experimental results some useful conclusions have been derived which would be helpful in improving the performance of the process.

1. Introduction Centreless grinding is an important and versatile process which is extensively used in the production of precision components. Centreless grinding is not only a finishing technique but is also employed in the formation of symmetrical parts [l]. A variety of items such as needles, shafts, diesel injector valves, piston and firing pins, power steering parts, crank shafts, etc., are economically produced by this method. The quality of centreless-ground components is normally assessed on the basis of their surface finish and roundness error. Some attempts have been made to study the influence of several process parameters, such as workpiece setting height, workplate angle, speed of the control wheel, rate of infeed, etc., on the quality of the ground components [Z - 61. Also a few investigations have been carried out to analyse the process from the stability viewpoint [7 - 91. 0 Elsevier ~quo~~mted

in The Neth~~nds

148

The friction between the workpiece surface and its supports, namely the control wheel, workplate and grinding wheel, greatly influences the quality of the ground parts [lo, 111. Investigations with a metal control wheel, instead of a conventional abrasive rubber-bonded control wheel, have shown improvements in the roundness and surface finish of the ground parts. However, the reduction in the coefficient of friction resulted in the ejection of the workpiece at a larger workpiece setting height [lo]. Hashimato et al. [Xl] investigated the process with a control wheel having a ground surface instead of the surface produced by the conventional single-point diamond dresser. They observed that the trueing of the control wheel by grinding could make its surface very stable and wear resistant [ 111. The workplate material also influences the surface finish of the ground parts. Aluminium, bronze, high speed steel and carbide blades may be used, depending on the degree of surface finish required [12]. The material and finish of the control wheel and workplate alter the frictional conditions at the control wheelworkpiece and workplate-workpiece interfaces. However, the available literature shows that no effort has been made to study the influence of the altered frictional conditions on the workpiece movement, namely the rate of workpiece rotation and its axial feed rate in through feed centreless grinding. Attempts have been made to study the wear characteristics of the control wheel, using rolling-sliding-type test equipment [ 13,141. Investigations into cylindrical grinding with different types of cutting fluids have been carried out to study the effect of altered frictional conditions at the grinding wheel-workpiece interface and its influence on the G ratio, wheel wear, wheel clogging and quality of the ground parts [ 15 - 17 ] . However, the available literature reveals that no attempts has been made to study the effect of altered frictional conditions due to different cutting fluids in centreless grinding. In this connection it is to be mentioned that there is a need for studying the influence of different types of cutting fluids on the quality of ground parts in centreless grinding as the workpiece is friction driven by the control wheel and is supported by the workplate and the control wheel. In centreless through feed grinding it is very difficult to measure the friction at the workplate and the control wheel. As the friction influences the rate of workpiece rotation and the axial feed rate, frictional effects are indirectly studied in this investigation by measuring the variations in the workpiece rotation and slip in the axial feed rate. In this paper an attempt is made to describe the influence of some of the important process parameters on the rate of workpiece rotation and the axial feed rate with different cutting fluids. The effect of these variations in the workpiece movements on the quality of the ground parts, using different cutting fluids, is also described. 2. Principle of operation Centreless grinding differs from cylindrical grinding in that the workpiece does not rotate about a fixed centre. The workpiece is supported by

149 rworkpiece

Fig. 1. Principle of operation.

the workplate and the control wheel (Fig. 1). The workpiece centre is kept above the line joining the centre of the grinding and control wheels. The control wheel also serves as a frictional driving element and governs the movement of the workpiece. In through feed, grinding is carried out by passing the workpiece between the grinding and the control wheels, entering at one end and leaving at the other. The axial feed rate of the workpiece depends on the speed and tilt of the control wheel. The generating mechanism involved in centreless grinding is highly complex and is affected by the geometric and dynamic stability of the workpiece. The stability of the process which affects the quality of the ground parts is greatly influenced by several process parameters such as the workpiece setting height, workplate angle, speed and tilt of the control wheel, depth of cut, dressing conditions of the grinding and control wheels, etc. As the aim of this paper is to describe the effect of different types of cutting fluids on the rate of workpiece rotation, its axial feed rate during grinding and the quality of the ground parts, only those process parameters which significantly influence the frictional conditions at the control wheelworkpiece interface and the grinding wheel-workpiece interface were varied during the investigation. The percentage change in the workpiece rotation in the case of through feed grinding is given by vs =

Nt -N, Nt

x 100

where Nt = D,N,

cos r#l 4,

The percentage change in the axial feed rate of the workpiece

(1)

ft = nD,N, sin 4 From the above equations it is seen that Vs and SL depend on the speed and tilt of the control wheel. In addition, the depth of cut and type of cutting fluid also alter the frictional conditions at the control wheel-workpiece and the grinding wheel-workpiece interfaces. Hence the speed of the control wheel, depth of cut and cutting fluid were selected as variables in this experimental investigation. 3. Experimental set-up The schematic arrangement of the experimental set-up is shown in Fig. 2.. A linear variable differential transducer (LVDT) was used to measure the axial movement of the workpiece as it passed between the grinding and the control wheels in through feed grinding. The LVDT was mounted in a specially made fixture in such a way that the core of the LVDT and the

5 K Fig. 2. Schematic diagram of the experimental set-up: 1, grinding wheel; 2, control wheel; 3, workpiece; 4, reflector; 5, digital tachometer; 6, fixture; 7, linear variable differential transducer; 8, attenuator; 9, carrier frequency amplifier; 10, Oscillofil.

151

cylindrical workpiece were coaxial. The axial movement of the workpiece was transferred to the LVDT from where the signal was taken to a carrier frequency amplifier and fed to an Oscillofil recorder. The workpiece speed was measured with the aid of a digital non-contact-type tachometer. 3.1. Experimental procedure The invest~ations were carried out on a Herminghau~n SR-2G centreless grinder. Some relevant experimental details are given in Appendix A. The process parameters varied during the experiments were as follows. (1) Speed of the control wheel. (2) Depth of cut. A large number of specimens were ground under different combinations of the machining conditions listed above. As all the results cannot be presented here, only a few typical graphs are included in this paper. The rate of workpiece rotation and the axial feed rate were measured as explained earlier. 3.2. Effect of cutting fluids To investigate the effect of different cutting fluids on the axial slip and the rate of rotation of the workpiece due to the altered frictional conditions during grinding, the experiments were carried out under different grinding conditions. They were (a) dry, (b) an emulsion of soluble oil in water (1: 50) as the cutting fluid and (c) oil (Te~us-27) as the cutting fluid. The reason for selecting dry grinding as one of the cutting conditions is mainly to emphasize the effect of dry friction on the axial slip and the variation in the work speed. The surface finish and the out-of-roundness appearance of the ground parts were measured on a Perthometer and a roundness tester respectively,

4. Results and discussion Figure 3 and Fig. 4 show the variation in Vs and the roundness error with the speed of the control wheel and the depth of cut respectively. At lower work speeds, Vs and the roundness error are higher. Very low work speeds result in improper matching of the grinding wheel-workpiece speeds. During grinding, the effects of cutting, ploughing and rubbing are observed [17, 183. At very low work speeds, the ploughing and rubbing action of the gram predominates and, in turn, increases Vs and the roundness error. As the work speed increases, the cutting action becomes effective and ploughmg and rubbing become negligible [ 17 J. The effect of very low speed resulted in tiny spots and a poor finish on the ground components. In addition, as the workpiece is floating in the grinding zone, the rubbing action of the abrasive grains against the work surface tends to drag the workpiece which may affect the work speed. The out~f-ro~dne~ appearance of the ground specimen is more pronounced at higher speeds and larger depths of cut. At higher speeds the

152

153

grinding process tends to become dynamically unstable. The floating nature of the workpiece allows it to vibrate and the workpiece tends to lose contact with the workplate and ride over the control wheel. Also, at higher depths of cut, the normal force and contact pressure at the control wheelworkpiece interface increase and lead to dynamic instability. Such a situation means that metal removal by the abrasive grains is irregular. Owing to friction at the grinding wheel-workpiece interface, the grinding wheel attempts to drag the workpiece which, in turn, increases V, and the roundness error. Workpiece movement, due to instability at higher speeds, increases the axial feed rate and reduces the slip. Also, at larger depths of cut, owing to higher contact pressure, the workpiece is moved more effectively in the axial direction and slip is reduced. Further, any vibration of the workpiece at higher work speeds, or at higher contact pressure due to larger depths of cut, increases the surface roughness of the ground parts (Figs. 5 and 6). Figures 3 - 6 show the effect of water-soluble oil emulsion and oil on Vs, SL, roundness errors and the surface finish in comparison with dry conditions. In dry grinding, the friction at the control wheel-workpiece interface is greater and increases the contact pressure between the control wheel and the workpiece. This reduces the slip in the axial direction and the surface roughness increases. Also, the higher friction between the control wheel and the workpiece reduces the variation in the work speed and makes the process more stable. Therefore the roundness error is reduced. However, it is to be mentioned here that dry grinding results in rapid wear of the abrasive grains and loading of the grinding wheel. The effective self dressing of the wheel cannot be expected. Also, specimens ground at larger depths of cut exhibited several surface defects such as local discoloura‘tion of the surface to give a brownish shade, microcracks, etc. This is due to intense local heating during grinding. With emulsion as the coolant (water-soluble oil) the friction between the control wheel and the workpiece is reduced. This alters the nature of contact at the wheel-workpiece interface and increases the axial slip. Similarly, with oil as the cutting fluid, friction is further reduced and the axial slip increases. Any increase in the axial slip amounts to a reduction in the ,effective feed rate which results in a reduction in the surface roughness. Any reduction in friction at the control wheel-workpiece interface, due to the presence of emulsion or oil, increases the tendency of the workpiece to become unstable. The floating nature of the workpiece permits it to jump freely and vibrate which, in turn, not only alters its speed but also increases the roundness error. Figure 7 shows a few typical polar charts of ground parts under different grinding conditions. In order to examine the influence of both Vs and SL on the out-ofroundness and the surface finish, a regression analysis was carried out. The results of the analysis are

NC

L5 (rpm) 60

16

0.2

0.3

06

Fig. 6. Effect of depth of cut and cutting fluid on SL and R,: - - -, SL; ---,

X,

X,

oil; t = 20 pm.

25

oil; N, = 45 rev min-‘.

R,; 0, dry; 0, emulsion;

t (pm) R,; l, dry; 0, emulsion,

Fig. 5. Effect of cutting fluid and control wheel speed on S, and R, : - - -, S,; -,

30

1.a

2.0

26

26

(a)

(d)

ICI

Fig. 7. Typical Cutting

polar charts of the ground

specimen.

conditions

ht (Pm)

:I& min-‘) (a) (b) (c) (d)

dry emulsion emulsion oil

60 30 60 60

60 20 20 20

2.27 1.0 1.515 2.0

5.6 1.9 4.45 4.54

156 h, = 2.182Vs”.74SLo.25

(dry)

h, = 3.236Vs”.76SL-o.37

(with coolant)

h, = 2.710Vs”~85SL-o~03

(with oil)

R a = 2 -316Vs-0.29SL-0.‘s

(dry)

R

(with coolant)

= () CJ-jOVs~.~~SL-‘.~~ a -

R a = 1 *()24Vs-0.02SL-1.50

(with oil)

(3)

(4)

From the values of the exponents it is found that Vs influences the roundness error to a greater extent. Similarly, the influence of SL on the surface finish is found to be greater than that of Vs. Therefore an attempt has been made to establish the relation between Vs and the roundness error and SL and the surface finish using the following mathematical models: y’ = by”

(5)

y = UO+ 2 UjXi+ iU*iXi2 + C i=l

i=

1

QijXiXj

(f-5)

i
In eqn. (5) y’ represents the roundness error or the surface roughness and y represents Vs or SL, and y is to be derived from eqn. (6). Equation (6) is a second-order model and predicts the independent, interactive and higher order effects of different process parameters on the response. A FORTRAN program was developed to compute the values of the constants of the above equations. In the analysis of eqn. (6) the control wheel speed and the depth of cut were treated as x1 and x2 respectively and the value of n was taken as two. The values of the constants of eqns. (5) and (6) for different grinding conditions have been evaluated and are given in Table 1. To test the adequacy of the models, an analysis of variance was carried out and the computed value of the F ratio was compared with the standard value with the appropriate degrees of freedom. It was found that the models given by eqns. (5) and (6) are adequate. It should be mentioned here that a large number of readings, including those shown in Figs. 5 and 6, are used to arrive at these models. The response curves in Figs. 8 and 9 are valid for the range of values shown. Figures 8 and 9 clearly show the effect of VS on the roundness error and SL on the surface finish under different grinding conditions and are derived from eqns. (5) and (6). From the above analysis it is seen that the presence of oil or an oilbased emulsion reduces the friction at the control wheel-workpiece interface and, in turn, increases the roundness error but improves the surface finish to some extent. Hence, wherever form accuracy of the ground parts is more important, cutting fluid without any oil content could be used. 5. Conclusions (1) Roundness error and the surface finish of the ground parts are influenced by the frictional conditions at the control wheel-workpiece interface.

157 TABLE 1 Values of constants of eqns. (5) and (6) (15 < N, < 60 rev min-‘; Constants

20 < t < 60 pm)

Grinding conditions DJY

Emulsion of

Oil

soluble oil

(1:50) (i) VS and h, b

c 00 a1 a2 011 a22 a12

2.457 0.687 -2.2168 0.1147 -0.0253 -0.0015 0.00011 0.00014

2.550 0.798 -0.5360 0.0143 -0.0180 -0.0005 -0.00009 0.0001

2.618 0.866 -4.7020 0.2406 -0.0387 -0.0028 0.0007 -0.0009

0.808 -0.705 1.4710 0.0205 -0.0054 -0.0003 -0.00015 0.00018

0.914 -1.287 1.2478 0.0525 -0.0074 -0.00065 -0.00002 0.000037

0.993 -1.490 1.9343 0.0227 -0.009 -0.00012 -0.00001 0.00001

(ii) SL and R, b C a0 01 a2 a11 022 012

As it is difficult to measure the friction in centreless grinding, an attempt is made in this paper to study this factor on the basis of the variation in the work speed and the axial slip. (2) Application of oil or an oil-based coolant causes a greater variation in the work speed and increases the axial slip which indicate a reduction in the friction. This reduction in the friction improves the surface finish but increases the roundness error. (3) Wherever the circularity of the ground specimen is crucial, cutting fluids without any oil content, such as water with rust inhibitors, may be used. However, if surface finish is the main requirement oils or soluble oils in water can be used. (4) The variation in the work speed is found to influence the roundness error to a greater extent. Alternatively, the influence of the axial slip on the surface finish is found to be greater than that of Vs. (5) The influence of Vs on the roundness error and SL on the surface finish creates the possibility of monitoring the centreless grinding process using Vs and SL for the roundness error and the surface finish respectively [ 181. This may also find application in technological adaptive control.

158

159

References

6

7 8

9 10 11 12 13 14 15 16 17 18

B. A. Mackenzie, Forming parts by centreless grinding, Mach. Des., 52 (6) (1980) 54 - 55. W. B. Rowe and M. M. Barash, Computer method for investigating the inherent accuracy of centreleas grinding, Znt. J. Mach. Tool Des. Res., 4 (2) (1964) 91 - 116. G. S. Drobashvskii, Optimum settings for a centreless grinder, Much. Tool. USSR, 20 (9) (1973) 20 - 21. W. B. Rowe, Research into mechanics of centreless grinding, Precis. Eng., 1 (2) (1979) 75 - 84. N. G. Subramanya Udupa, M. S. Shunmugam and V. Radhakrishnan, A preliminary investigation on surface fish and roundness error in centreless grinding, Proc. 11 th All India Machine Tool Design and Research Conf., Madras, 1984, pp. 251 - 258. N. G. Subramanya Udupa, M. S. Shunmugam and V. Radhakrishnan, Influence of workpiece position on roundness error and surface finish in centreless grinding, Znt. J. Mach. Tool Manufi, 27 (1) (1987) 77 - 99. M. Miyashita, Diagram for selecting chatter free conditions of centreless grinding, Ann. CZRP, 31 (1) (1982) 221 - 223. M. Frost and P. M. T. Fursdon, Towards the optimum centreless grinding, Proc. 16th Milton C. Shaw Grinding Symp., 1985, American Society of Mechanical Engineers, New York, pp. 313 - 328. A. F. Prokhorov, Optimisation of the configuration of a centreless grinder, Sou. Eng. Res., 2 (6) (1982) 77 - 78. V. L. Ramanov, Metal control wheeIs on centreless grinder, Mach. Tool USSR, 38 (2) (1967) 24 - 25. F. Hashimato, A. Kanai and M. Miashita, High precision trueing method of regulating wheel and effect on grinding accuracy, Ann. CZRP, 32 (1) (1983) 237 - 239. K. S. Micheal, A guide to centreless grinding of carbides, Tool. Prod., (1978) 93 - 95. F. Hashimato and M. Miyashita, Wear characteristics of regulating wheel in centreless grinding, Lect. JSPE, 3 (1981) 548. F. Hashimato and M. Miyashita, Friction characteristics of regulating wheel and its evaluation in centreless grinding, Lect. JSPE, 3 (1979) 169. C. A. Sluhan, Grinding with water miscible grinding fluids, Lubr. Eng., 26 (10) (1970) 352 - 374. L. V. Kudobin, Cutting fluid and its effect on grinding wheel clogging, Mach. Tool. USSR, 40 (9) (1969) 54 - 59. N. Komine, Effects of surrounding environment on wear of grinding wheels, Bull. JSPE, 4 (4) (1970) 115 - 116. N. G. Subramanya Udupa, M. S. Shunmugam and V. Radhakrishnan, Process monitoring in centreless grinding for out-of-roundness and surface finish criteria, Proc. 12th All India &chine Tool Design and Research Conf., 1986, Tata McGraw-Hill, New Delhi, 1987, pp. 316 - 320.

Appendix A TABLE Al Experimental conditions Grinding machine Grinding wheel Grinding wheel speed (rev minControl wheel Work plate angle (deg) Specimen Included tangent angle (deg) Tilt of control wheel (deg)

Herminghausen SR-PG model diameter 300 mm, AGOMV 2000 diameter 200 mm, rubber bonded 30 Solid mild steel pins of diameter 30 mm and length 60 mm 6 1

160

Appendix B: Nomenclature

a,, ai, ati, aij, b, c 4v DC

:b h” ht Nil NC Nt R, SL t

vs a’ P 9

constants diameter of workpiece diameter of control wheel actual axial feed rate of the workpiece theoretical axial feed rate of the workpiece height of workpiece centre roundness error actual workpiece speed speed of control wheel theoretical workpiece speed surface roughness percentage change in the axial feed rate depth of cut percentage variation in the workpiece speed workplate angle included tangent angle tilt of control wheel