- Email: [email protected]
An investigation on solid lubricant moulded grinding wheels
International Journal of Machine Tools & Manufacture 43 (2003) 965–972
An investigation on solid lubricant moulded grinding wheels S. Shaji ∗, V. Radhakrishnan
1
Manufacturing Engineering Section, Department of Mechanical Engineering, Indian Institute of Technology-Madras, Chennai 600 036, India Received 12 July 2002; received in revised form 24 February 2003; accepted 27 February 2003
Abstract The conventional flood coolants employed in grinding suffer many limitations in performing their functions. They cannot be recommended in the light of ecological and economic manufacture. Application of solid lubricant in grinding has proved to be a feasible alternative to the fluid coolants, if it could be applied in a proper way. Towards finding out an improved method of application of solid lubricant, attempts on development of solid lubricant moulded grinding wheels with various bonding and lubricants have been reported here. Such wheels with resin bonding were successfully made and improved process results were obtained. But the wheel wear depended on the type of the lubricant used. 2003 Elsevier Science Ltd. All rights reserved. Keywords: Grinding; Lubrication; Coolant; Graphite; Calcium fluoride; Solid lubricant
1. Introduction The functions of cutting fluid in grinding are critical in ensuring product quality, because of the intense heat generation and the consequent risk of thermal damage associated with the process [1,2]. Liquid coolants, applied in flood form, are the conventional choice to achieve cooling and lubrication in grinding. Many limitations have been observed in this approach, though it is a means for effective removal of grinding swarf and wheel cleaning. The ‘air barrier’ inhibits the flow of cutting fluids into the grinding zone [2]. Even if a little quantity reaches there, ‘film boiling’ is another limitation to the heat control [3]. When the grinding zone temperature exceeds a certain limit, lubricity of the fluid is degraded [4]. Grinding fluids also incur a significant share of the total manufacturing cost [5,6]. Moreover, the use of it is not advisable as it can have negative impact on workers’ health [7,8]. All these factors prompt investigations on the development of new abrasive technologies with fluid minimization or elimination strategies by substituting the functions, which have to be ∗
Corresponding author. Present address: Government Engineering College, Trivandrum, Kerala 695 035, India. E-mail address: [email protected] (S. Shaji). 1 Present address: School of Mechanical and Production Engineering, Nanyang Technological University, Singapore 639 798.
normally met by them, through some other means. Use of biodegradable and cryo-coolants and the concept of minimum quantity lubrication (MQL) are the alternative approaches in this direction [2,9]. If friction at the wheel–workpiece interaction can be minimized, by providing effective lubrication, the heat intensity responsible for thermal damage can be reduced to some extent. Advancement in modern tribology has identified many solid lubricants, which can sustain and provide lubricity over a wider range of temperatures [10,11]. If suitable solid lubricant can be successfully applied to the grinding zone in a proper way, it should yield better process results than conventional grinding. These authors have investigated the feasibility of application of graphite as lubricant in surface grinding [12]. In this preliminary study, graphite was applied in a suitable paste form to the working surface of the wheel, with a special attachment. The effective role of graphite as lubricant was evident from the improvement of process results related to frictional factors. But the wheel clogging in the absence of proper flushing and non-uniformity of lubricant admission were found to be a major hindrance in obtaining more desirable results. If lubricant can be applied in a more refined and defined way, just sufficient for effective lubrication, improved process results could be expected. An ideal way of application of solid lubricant in grinding is to have a built-in supply of lubricant from the
0890-6955/03/$ - see front matter 2003 Elsevier Science Ltd. All rights reserved. doi:10.1016/S0890-6955(03)00064-6
966
S. Shaji, V. Radhakrishnan / International Journal of Machine Tools & Manufacture 43 (2003) 965–972
wheel itself. This paper deals with investigations on solid lubricant moulded grinding wheels by including lubricant in the wheel structure during the moulding stage. Trials were done to develop wheels of vitrified and resin bonding with various compositions of solid lubricants. Besides graphite, calcium fluoride (CaF2), which is also a commonly used high temperature solid lubricant, has been tried [10,11].
2. Development of wheels 2.1. Trials with vitrified bonded wheels The majority of grinding wheels used in general engineering industry are of vitrified bond. Vitrified bond is made of clay or feldspar which fuses at high temperature to form a glass like structure. Vitrified bonds are strong, rigid, impart good finish and are insensitive to water or other grinding fluids. For making the wheel, the ingredients such as abrasive grains, bonding material, porosity media, water etc. are uniformly mixed and properly moulded. These green wheels are baked at very high temperature for a long duration, followed by slow cooling [13–16]. Several trials were done to make solid lubricant moulded vitrified bonded wheels with A60K5 wheel composition by mixing solid lubricant powder along with the usual ingredients, during the moulding stage. Trials were done to make graphite and CaF2 moulded wheels separately, with lubricant composition of 5% and 10% by weight, but failed due to various reasons. The vitrifying temperature is of the order of 1400 °C. Graphite starts to oxidise at about 550 °C. If the firing is done in the presence of oxygen, graphite will decompose loosing its lubricative property and the resulting products would affect the bonding strength of the wheel. As a suitable reduction furnace was not available, attempts to make the graphite moulded wheels were done by embedding the lubricant moulded green wheel in graphite mass itself with sufficient packing thickness such that the chances of oxidation would not affect the wheel. But the wheels produced were multi-coloured, showing the occurrence of oxidation. These wheels could not withstand the speed testing at the standard conditions. CaF2 moulded wheels were also made in the same way, but they also failed in the speed test. At high temperature, CaF2 might react with the silicate ingredients, producing products which weakened the bond structure. 2.2. Trials with resin bonded wheels Resin bonded wheels are manufactured in a similar manner to the vitrified wheels, but the bonding medium is thermosetting synthetic resin and the baking temperature is of the order of 150–200 °C only. These wheels
are soft structured and are used for high speed grinding, rough grinding, cutting-off operations etc. The wheel wear rate is high and the wheel softens when exposed to grinding fluids [13–16]. The low baking temperature involved in resin bonding prompted the trials with solid lubricants included in the wheel structure during moulding. An A60L5BM4 (250–76.2–25 mm size) wheel was taken as the standard wheel. This wheel consists of ingredients of 91% abrasive grain and 9% bonding material by weight. Three graphite moulded wheels were made by adding graphite in 3%, 6% and 9% by weight to the standard wheel ingredients. Similarly, three CaF2 moulded wheels were also made by adding CaF2 in 5%, 10% and 15% by weight to the standard wheel. The possibility of weakening the bond strength constrained the inclusion of graphite in lower proportion. However, CaF2 increases adhesiveness and enhances the bonding strength at higher temperature [17]. All wheels withstood speed testing at standard conditions.
3. Analysis of critical process parameters In any machining set-up, a systematic analysis of process parameters is essential to find out their relative influence on the direct process outputs or on the secondary variables which have direct impact on the process results. Taguchi’s method of experimental design has become a powerful tool for serving this purpose and thereby optimising the process outputs [18,19]. Grinding is a complex manufacturing process, with a large number of interacting variables, and the process outputs depend on the wheel characteristic, workpiece characteristic, machine characteristic and the operating conditions. In a given grinding set-up of wheel, workpiece and machine, the relative influence of process parameters on process outputs or on other secondary variables has been analysed here, based on Taguchi’s approach, in order to proceed with further experimentation. Taking normal force, tangential force and surface roughness as quality characteristics and speed, feed, infeed and mode of dressing as factors of study on three levels, Taguchi’s L9 orthogonal array was followed for the analysis. Experiments were conducted in a horizontal surface grinding machine with bearing steel (En-31) as workpiece and employing the standard wheel A60L5BM4. The relative influence of various factors was determined through the ANOVA based on S/N data. Infeed was found to be the most influential factor on both normal and tangential forces (about 85%). As the forces developed in the process directly depend upon the undeformed chip thickness, the prominent influence of the infeed on forces is understandable [20,21]. In the case of surface roughness, though the mode of dressing was found to have a prominent influence, its relative contribution was substantially less (about 40%), when
S. Shaji, V. Radhakrishnan / International Journal of Machine Tools & Manufacture 43 (2003) 965–972
compared to that with vitrified wheels (about 75%) [22]. The soft structure of resin bonding is sensitive to thermal loading. Wheel wear is dominated by thermal softening of bonds and their failure, rather than grain fracture and subsequent bond fracture due to mechanical stress in the case of vitrified wheels [13–15]. As the wear rate is higher in the case of resin wheels, the mode of dressing will not have much influence on the process results, except at the initial stage of cutting.
4. Experimentation Based on the preliminary findings on the influence of process parameters, experiments for the performance analysis of the newly developed wheels were done in one dressing mode only, varying the infeed in steps. The experimental conditions are shown in Table 1. The factors selected for the study are force components, force ratio, specific energy, surface roughness, wheel wear and intermittent forces in continuous test. For force measurements a three- component quartz dynamometer (Kistler—type 9257B) with charge amplifier (Kistler— type 5019B) and a two-channel oscilloscope connected directly to a PC were employed. The forces reported are those when the process was in stable state with almost steady pulses. Surface roughness (Ra) of the ground part was taken using a Talysurf.
5. Results and discussion The comparative performance analyses of the solid lubricant moulded wheels with the standard wheel were done separately for graphite and CaF2 moulded wheels. Figs. 1 and 2 show the comparisons of normal and tangential force components obtained (per unit width of the wheel) for different wheels, when both types of workpieces were ground. Grinding forces are important parameters by which performance of any grinding process
can be evaluated. These forces may consist of components due to primary ploughing, secondary ploughing, shearing, micro fracturing, wheel loading and friction between the grain flats, the bonding material, redeposited debris etc. and the workpiece [23]. The relative contributions of these elements depend upon the type of work material, abrasive characteristics, wheel surface condition, process parameters and the presence of grinding fluid. In all the cases under study, force components increased with infeed as expected. This agrees with the various physical and empirical models reported which mention that force components are directly proportional to the mean undeformed chip thickness, which in turn is directly proportional to the infeed [20,21]. The productive components of the forces, which result in material removal such as shearing, micro fracturing and secondary ploughing etc. are non-frictional in nature and are more or less proportional to the infeed. The nonproductive components of forces, which are mostly frictional in nature, are almost independent of infeed [23]. Normal force or radial force component influences all the parameters pertaining to the performance of a grinding wheel. This force presses the abrasive grains into the work surface and penetrates the surface, leading to material removal. At higher infeeds, the normal force increases causing more material removal. This force is also responsible for wheel wear, as when the thrust on the wheel exceeds certain limits, the grain fracture and bond fracture will occur. It also affects the surface roughness and the geometrical and dimensional accuracy of the finished component. Wheel clogging by grinding swarf also causes an increase in the normal force, by hindering the effective penetration of grains into the work surface. The normal forces obtained were lower with the lubricant moulded wheels compared to the standard wheel, in all the cases. The normal force components due to the frictional effects are reduced due to the effective lubrication by lubricant moulded wheels. In the earlier study on the external application of lubricant with vitrified wheels, the normal forces obtained in some
Table 1 Experimental conditions Machine Wheel
Workpiece material Cutting speed Feed Infeed Method of grinding Dressing conditions
967
BLOHM Horizontal spindle surface grinding machine, 6.5 kW 1. Standard: A60L5BM4, CUMI make 2. Graphite moulded wheels with 3%, 6% and 9% graphite by weight included in the wheel 3. CaF2 moulded wheels with 5%, 10% and 15% CaF2 by weight included in the wheel (All wheels are of 250–76.2–25 mm size) 1. Medium carbon steel (En-2), HRC 22 2. Bearing steel (En-31), HRC 60 42 m/s 10 m/min 10–80 µm in steps of 10 µm Dry With single point diamond dresser, 1 carat, at wheel speed of 2500 rpm in dry condition with 20 µm dressing depth, 苲150 mm/min cross feed and without any spark out pass
968
S. Shaji, V. Radhakrishnan / International Journal of Machine Tools & Manufacture 43 (2003) 965–972
Fig. 1.
Fig. 2.
Comparison of normal forces (per unit width of the wheel) in the case of lubricant moulded wheels with different workpieces.
Comparison of tangential forces (per unit width of the wheel) in the case of lubricant moulded wheels with different workpieces.
S. Shaji, V. Radhakrishnan / International Journal of Machine Tools & Manufacture 43 (2003) 965–972
cases were higher than those in normal grinding, due to the non-uniform supply of lubricant and increased chances of wheel clogging [12]. But here, an efficient way of lubricant supply, which minimized wheel clogging by grinding swarf, might reduce the normal force to some extent. As the wear rate is higher with resin wheels, fresh grains might also more frequently come into contact with the workpiece on subsequent passes, preventing the increase of forces due to grain dullness. The tangential force component determines the power requirement in the process. The intensity of heat generation depends upon this force and it is primarily important as far as the grinding temperatures and surface integrity aspects of the products are concerned. The lubricant effectiveness in minimizing the frictional effects at the wheel–workpiece interaction in the case of lubricant moulded wheels is evident from the reduced tangential forces compared to the standard wheel. But higher quantities of lubricant inclusion do not reduce forces significantly compared with the lowest quantity. This might be due to a possible saturation in lubricant effectiveness. The ratio of tangential force to normal force (Ft/Fn) is termed grinding coefficient [24] or grinding force ratio [20,21]. In grinding, the normal force is approximately double the tangential force, unlike in other machining
969
processes like turning and milling [16,24]. This is due to grinding process characteristics such as very large negative rake angles of the grit, excessive rubbing action and adverse chip accommodation space which leads to wheel loading. This force ratio, though it does not give the actual coefficient of friction (m), is a similar term like m and gives an indication of the frictional effects in the grinding zone [24]. Fig. 3 shows the comparison of grinding force ratio for different lubricant moulded wheels. This ratio is generally lower in all the cases of lubricant moulded wheels, compared to the standard wheel, substantiating the effective lubrication by solid lubricant in those cases. This ratio has been reported as differing considerably depending upon the material of the workpiece and indicates the workpiece behaviour in the process [20,21]. The present results show that for the same material the grinding environment providing effective lubrication changes this ratio remarkably. This may be attributed to the behaviour of the material in the presence of solid lubricant. Fig. 4 shows the comparison of specific energy, which is the energy required per unit volume of material removal, obtained with various types of wheels. The ‘size effect’, which is the drastic increase of specific energy at comparatively lower infeeds could be clearly observed in all the cases [14,16]. The energy require-
Fig. 3. Comparison of grinding force ratio in the case of lubricant moulded wheels (different workpieces).
970
S. Shaji, V. Radhakrishnan / International Journal of Machine Tools & Manufacture 43 (2003) 965–972
Fig. 4.
Comparison of specific energy in the case of lubricant moulded wheels (different workpieces).
ment in material removal can be classified as due to the sliding, ploughing and chip formation components. The size effect in grinding is due to the variation of relative amount of these components in contributing towards the unit material removal. Sliding can be due to elastic contact between the abrasive grit and workpiece and rubbing of wear flat of grains, bonding material, grinding debris etc. with the workpiece. Ploughing causes metal to be displaced sideways around the grain rather than be removed as chip. These non-productive energy components depend mostly upon the ductility and hardness of the workpiece, sharpness of the grains and the local lubrication [23]. At relatively smaller infeeds, domination of sliding and ploughing components with relatively less chip formation calls for higher specific energy [14]. The percentage contribution of these components in the total specific grinding energy at higher infeeds is reduced, because these factors of energy are more or less independent of infeed after certain limit. In the case of lubricant moulded wheels, local lubrication by solid lubricants reduces the sliding and primary ploughing component of forces to some extent and results in reduced specific energy, especially at lower infeeds. At higher material removal or infeeds the specific energy converges to more or less the same value (~15 J/mm3). This agrees with Malkin and Anderson’s [25] postulation that at higher infeeds the ploughing component of grind-
ing energy approaches zero and in the limit, the specific cutting energy (which is the sum of ploughing energy and chip formation energy) corresponds to the specific energy required for chip formation. Chip formation energy is assumed to be constant for a particular material. For ferrous material it is 13.8 J/mm3, irrespective of the hardness of the material [14]. Fig. 5 shows the comparisons of workpiece surface roughness. Surface topography of the wheel determines the nature of the tribological interaction between the wheel and workpiece. Workpiece surface roughness is the minute scratches formed by the interacting wheel surface under the dynamic conditions of the process. The surface roughness scatter is not wider for various types of wheels. This is due to the fact that for a particular wheel, the surface condition of the wheel is more or less the same for all conditions of infeed, due to the increased wear of the resin bonding. The wheel lost its dressed surface condition immediately after the initial passes. But, in the case of vitrified wheels, it can maintain the surface condition of the wheel for a longer duration as its wear rate is less [12]. Fig. 6 shows wheel wear for different wheels. The wear of graphite-moulded wheels was higher, but that of CaF2-molded wheels was lower, when compared to the standard wheel. The presence of graphite may weaken the bonding, leading to higher wear. It is notice-
S. Shaji, V. Radhakrishnan / International Journal of Machine Tools & Manufacture 43 (2003) 965–972
Fig. 5.
Fig. 6.
971
Comparison of workpiece surface roughness obtained under different grinding conditions.
Wheel wear obtained with different wheels. Fig. 7. Intermittent forces obtained in continuous grinding test.
able that the wear is very high with 9% graphite inclusion. The difference in trend with CaF2 might be due to the fusion bonding nature of CaF2 [17]. It is also noticeable that wear with 15% CaF2 inclusion is more than that with 10% CaF2 inclusion. This might be due to a saturation in the bonding strength due to the presence of CaF2 at some limit and any further increase of lubricant inclusion would be weakening the bond. This is also on par with the saturation in the lubricant effectiveness observed earlier. Fig. 7 shows the force pattern obtained from intermittent measurements during continuous grinding test. The test was done with bearing steel workpiece, giving 10
µm infeed for every upgrinding pass. Forces were measured in every 25th pass. The force values are more or less the same throughout the test. This is in contrast to the case with vitrified wheels, where forces increased to certain extent in continuous test [24]. As the wheel wear rate is higher with resin wheels, during continuous grinding, fresh grains might frequently come into interaction. But with vitrified wheels, where the wheel wear is dominated by the mechanical fracture, load on the wheel increases till the critical force required for the friability of grains occurs.
972
S. Shaji, V. Radhakrishnan / International Journal of Machine Tools & Manufacture 43 (2003) 965–972
6. Conclusion As an alternative way to extract the advantages of solid lubricant application in grinding throughout the usage of the wheel, solid lubricant moulded resin wheels were developed with graphite and CaF2 separately, by including solid lubricants during the moulding stage of the wheel. Trials to make such wheels with vitrified bonding failed due to higher vitrifying temperature and non-availability of a good reduction furnace for firing. The effectiveness of lubricants was evident from the improved process results related to friction. The wheel wear depended on the type of the lubricant used. It was higher with graphite and lower with CaF2. Saturation in lubricant effectiveness could be observed on increase of the quantity of lubricant.
References [1] R. Snoeys, K.U. Leuven, M. Maris, N.F. Wo, J. Peters, Thermally induced damage in grinding, Annals of the CIRP 27 (1) (1978) 571–581. [2] E. Brinksmeier, C. Heinzel, M. Wittmann, Friction, cooling and lubrication in grinding, Annals of the CIRP 48 (2) (1999) 581– 598. [3] T.D. Howes, Assessment of the cooling and lubricative property of grinding fluids, Annals of the CIRP 39 (1) (1990) 313–315. [4] G.S. Cholakov, T.L. Guest, G.W. Rowe, Lubricating properties of grinding fluids: comparison of fluids in surface grinding experiments, Lubrication Engineering 48 (2) (1992) 155–163. [5] E. Brinksmeier, A. Walter, R. Janssen, P. Diersen, Aspects of cooling lubrication reduction in machining advanced materials, Proc. Instn Mech. Engrs, Part B, Journal of Engineering Manufacture 213 (1999) 769–778. [6] F. Klocke, G. Eisenblaetter, Dry cutting, Annals of the CIRP 46 (2) (1997) 519–526. [7] T.D. Howes, H.K. Toenshoff, W. Heuer, Environmental aspects of grinding fluids, Annals of CIRP 40 (2) (1991) 623–629.
[8] G. Byrne, E. Scholta, Environmentally clean machining processes—a strategic approach, Annals of the CIRP 42 (1) (1993) 471–474. [9] I. Inasaki, H.K. Toenshoff, T.D. Howes, Abrasive machining in the future, Annals of the CIRP 42 (2) (1993) 723–732. [10] B. Bhushan, Modern Tribology Handbook, CRC Press LLC, New York, 2001. [11] G.W. Stachowiak, A.W. Batchelor, Engineering Tribology, Butterworth-Heinenann, Woburn, 2001. [12] S. Shaji, V. Radhakrishnan, An investigation on surface grinding using graphite as lubricant, International Journal of Machine Tools and Manufacture 42 (2002) 733–740. [13] R.I. King, R.S. Hahn, Hand Book of Modern Grinding Technology, Chapman and Hall, New York, 1986. [14] S. Malkin, Grinding Technology: Theory and Applications of Machining with Abrasives, Ellis Horwood Ltd, UK, 1989. [15] S.C. Salmon, Modern Grinding Process Technology, McGrawHill, New York, 1992. [16] M.C. Shaw, Principles of Abrasive Processing, Oxford University Press Inc., New York, 1996. [17] A.R. Lansdown, High Temperature Lubrication, Mechanical Engineering Publication Ltd, London, 1994. [18] M.S. Phadke, Quality Engineering Using Robust Design, Prentice Hall, Englewood cliffs, NJ, 1989. [19] P.J. Ross, Taguchi Techniques for Quality Engineering, McGrawHill, New York, 1996. [20] H.K. Tonshoff, J. Peters, I. Inasaki, T. Paul, Modeling and simulation of grinding process, Annals of the CIRP 41 (2) (1992) 677–688. [21] E. Brinksmeier, H.K. Tonshoff, I. Inasaki, J. Peddinghans, Basic parameters in grinding, Annals of the CIRP 42 (2) (1993) 795– 799. [22] S. Shaji, V. Radhakrishnan, Analysis of process parameters in surface grinding using graphite as lubricant based on the Taguchi’s method, Journal of Material Processing Technology, 2003 (in press). [23] S. Paul, A.B. Chattopadhayay, The effect of cryogenic cooling on grinding forces, International Journal of Machine Tools and Manufacture 36 (1) (1996) 63–72. [24] E.C. Marshal, M.C. Shaw, Forces in dry surface grinding, Transactions of the ASME 74 (1952) 51–59. [25] S. Malkin, R.B. Anderson, Thermal aspects of grinding, Part 1— Energy partition, Transactions of the ASME, Journal of Engineering for Industry 98 (1974) 1177–1183.
An investigation on solid lubricant moulded grinding wheels S. Shaji ∗, V. Radhakrishnan
1
Manufacturing Engineering Section, Department of Mechanical Engineering, Indian Institute of Technology-Madras, Chennai 600 036, India Received 12 July 2002; received in revised form 24 February 2003; accepted 27 February 2003
Abstract The conventional flood coolants employed in grinding suffer many limitations in performing their functions. They cannot be recommended in the light of ecological and economic manufacture. Application of solid lubricant in grinding has proved to be a feasible alternative to the fluid coolants, if it could be applied in a proper way. Towards finding out an improved method of application of solid lubricant, attempts on development of solid lubricant moulded grinding wheels with various bonding and lubricants have been reported here. Such wheels with resin bonding were successfully made and improved process results were obtained. But the wheel wear depended on the type of the lubricant used. 2003 Elsevier Science Ltd. All rights reserved. Keywords: Grinding; Lubrication; Coolant; Graphite; Calcium fluoride; Solid lubricant
1. Introduction The functions of cutting fluid in grinding are critical in ensuring product quality, because of the intense heat generation and the consequent risk of thermal damage associated with the process [1,2]. Liquid coolants, applied in flood form, are the conventional choice to achieve cooling and lubrication in grinding. Many limitations have been observed in this approach, though it is a means for effective removal of grinding swarf and wheel cleaning. The ‘air barrier’ inhibits the flow of cutting fluids into the grinding zone [2]. Even if a little quantity reaches there, ‘film boiling’ is another limitation to the heat control [3]. When the grinding zone temperature exceeds a certain limit, lubricity of the fluid is degraded [4]. Grinding fluids also incur a significant share of the total manufacturing cost [5,6]. Moreover, the use of it is not advisable as it can have negative impact on workers’ health [7,8]. All these factors prompt investigations on the development of new abrasive technologies with fluid minimization or elimination strategies by substituting the functions, which have to be ∗
Corresponding author. Present address: Government Engineering College, Trivandrum, Kerala 695 035, India. E-mail address: [email protected] (S. Shaji). 1 Present address: School of Mechanical and Production Engineering, Nanyang Technological University, Singapore 639 798.
normally met by them, through some other means. Use of biodegradable and cryo-coolants and the concept of minimum quantity lubrication (MQL) are the alternative approaches in this direction [2,9]. If friction at the wheel–workpiece interaction can be minimized, by providing effective lubrication, the heat intensity responsible for thermal damage can be reduced to some extent. Advancement in modern tribology has identified many solid lubricants, which can sustain and provide lubricity over a wider range of temperatures [10,11]. If suitable solid lubricant can be successfully applied to the grinding zone in a proper way, it should yield better process results than conventional grinding. These authors have investigated the feasibility of application of graphite as lubricant in surface grinding [12]. In this preliminary study, graphite was applied in a suitable paste form to the working surface of the wheel, with a special attachment. The effective role of graphite as lubricant was evident from the improvement of process results related to frictional factors. But the wheel clogging in the absence of proper flushing and non-uniformity of lubricant admission were found to be a major hindrance in obtaining more desirable results. If lubricant can be applied in a more refined and defined way, just sufficient for effective lubrication, improved process results could be expected. An ideal way of application of solid lubricant in grinding is to have a built-in supply of lubricant from the
0890-6955/03/$ - see front matter 2003 Elsevier Science Ltd. All rights reserved. doi:10.1016/S0890-6955(03)00064-6
966
S. Shaji, V. Radhakrishnan / International Journal of Machine Tools & Manufacture 43 (2003) 965–972
wheel itself. This paper deals with investigations on solid lubricant moulded grinding wheels by including lubricant in the wheel structure during the moulding stage. Trials were done to develop wheels of vitrified and resin bonding with various compositions of solid lubricants. Besides graphite, calcium fluoride (CaF2), which is also a commonly used high temperature solid lubricant, has been tried [10,11].
2. Development of wheels 2.1. Trials with vitrified bonded wheels The majority of grinding wheels used in general engineering industry are of vitrified bond. Vitrified bond is made of clay or feldspar which fuses at high temperature to form a glass like structure. Vitrified bonds are strong, rigid, impart good finish and are insensitive to water or other grinding fluids. For making the wheel, the ingredients such as abrasive grains, bonding material, porosity media, water etc. are uniformly mixed and properly moulded. These green wheels are baked at very high temperature for a long duration, followed by slow cooling [13–16]. Several trials were done to make solid lubricant moulded vitrified bonded wheels with A60K5 wheel composition by mixing solid lubricant powder along with the usual ingredients, during the moulding stage. Trials were done to make graphite and CaF2 moulded wheels separately, with lubricant composition of 5% and 10% by weight, but failed due to various reasons. The vitrifying temperature is of the order of 1400 °C. Graphite starts to oxidise at about 550 °C. If the firing is done in the presence of oxygen, graphite will decompose loosing its lubricative property and the resulting products would affect the bonding strength of the wheel. As a suitable reduction furnace was not available, attempts to make the graphite moulded wheels were done by embedding the lubricant moulded green wheel in graphite mass itself with sufficient packing thickness such that the chances of oxidation would not affect the wheel. But the wheels produced were multi-coloured, showing the occurrence of oxidation. These wheels could not withstand the speed testing at the standard conditions. CaF2 moulded wheels were also made in the same way, but they also failed in the speed test. At high temperature, CaF2 might react with the silicate ingredients, producing products which weakened the bond structure. 2.2. Trials with resin bonded wheels Resin bonded wheels are manufactured in a similar manner to the vitrified wheels, but the bonding medium is thermosetting synthetic resin and the baking temperature is of the order of 150–200 °C only. These wheels
are soft structured and are used for high speed grinding, rough grinding, cutting-off operations etc. The wheel wear rate is high and the wheel softens when exposed to grinding fluids [13–16]. The low baking temperature involved in resin bonding prompted the trials with solid lubricants included in the wheel structure during moulding. An A60L5BM4 (250–76.2–25 mm size) wheel was taken as the standard wheel. This wheel consists of ingredients of 91% abrasive grain and 9% bonding material by weight. Three graphite moulded wheels were made by adding graphite in 3%, 6% and 9% by weight to the standard wheel ingredients. Similarly, three CaF2 moulded wheels were also made by adding CaF2 in 5%, 10% and 15% by weight to the standard wheel. The possibility of weakening the bond strength constrained the inclusion of graphite in lower proportion. However, CaF2 increases adhesiveness and enhances the bonding strength at higher temperature [17]. All wheels withstood speed testing at standard conditions.
3. Analysis of critical process parameters In any machining set-up, a systematic analysis of process parameters is essential to find out their relative influence on the direct process outputs or on the secondary variables which have direct impact on the process results. Taguchi’s method of experimental design has become a powerful tool for serving this purpose and thereby optimising the process outputs [18,19]. Grinding is a complex manufacturing process, with a large number of interacting variables, and the process outputs depend on the wheel characteristic, workpiece characteristic, machine characteristic and the operating conditions. In a given grinding set-up of wheel, workpiece and machine, the relative influence of process parameters on process outputs or on other secondary variables has been analysed here, based on Taguchi’s approach, in order to proceed with further experimentation. Taking normal force, tangential force and surface roughness as quality characteristics and speed, feed, infeed and mode of dressing as factors of study on three levels, Taguchi’s L9 orthogonal array was followed for the analysis. Experiments were conducted in a horizontal surface grinding machine with bearing steel (En-31) as workpiece and employing the standard wheel A60L5BM4. The relative influence of various factors was determined through the ANOVA based on S/N data. Infeed was found to be the most influential factor on both normal and tangential forces (about 85%). As the forces developed in the process directly depend upon the undeformed chip thickness, the prominent influence of the infeed on forces is understandable [20,21]. In the case of surface roughness, though the mode of dressing was found to have a prominent influence, its relative contribution was substantially less (about 40%), when
S. Shaji, V. Radhakrishnan / International Journal of Machine Tools & Manufacture 43 (2003) 965–972
compared to that with vitrified wheels (about 75%) [22]. The soft structure of resin bonding is sensitive to thermal loading. Wheel wear is dominated by thermal softening of bonds and their failure, rather than grain fracture and subsequent bond fracture due to mechanical stress in the case of vitrified wheels [13–15]. As the wear rate is higher in the case of resin wheels, the mode of dressing will not have much influence on the process results, except at the initial stage of cutting.
4. Experimentation Based on the preliminary findings on the influence of process parameters, experiments for the performance analysis of the newly developed wheels were done in one dressing mode only, varying the infeed in steps. The experimental conditions are shown in Table 1. The factors selected for the study are force components, force ratio, specific energy, surface roughness, wheel wear and intermittent forces in continuous test. For force measurements a three- component quartz dynamometer (Kistler—type 9257B) with charge amplifier (Kistler— type 5019B) and a two-channel oscilloscope connected directly to a PC were employed. The forces reported are those when the process was in stable state with almost steady pulses. Surface roughness (Ra) of the ground part was taken using a Talysurf.
5. Results and discussion The comparative performance analyses of the solid lubricant moulded wheels with the standard wheel were done separately for graphite and CaF2 moulded wheels. Figs. 1 and 2 show the comparisons of normal and tangential force components obtained (per unit width of the wheel) for different wheels, when both types of workpieces were ground. Grinding forces are important parameters by which performance of any grinding process
can be evaluated. These forces may consist of components due to primary ploughing, secondary ploughing, shearing, micro fracturing, wheel loading and friction between the grain flats, the bonding material, redeposited debris etc. and the workpiece [23]. The relative contributions of these elements depend upon the type of work material, abrasive characteristics, wheel surface condition, process parameters and the presence of grinding fluid. In all the cases under study, force components increased with infeed as expected. This agrees with the various physical and empirical models reported which mention that force components are directly proportional to the mean undeformed chip thickness, which in turn is directly proportional to the infeed [20,21]. The productive components of the forces, which result in material removal such as shearing, micro fracturing and secondary ploughing etc. are non-frictional in nature and are more or less proportional to the infeed. The nonproductive components of forces, which are mostly frictional in nature, are almost independent of infeed [23]. Normal force or radial force component influences all the parameters pertaining to the performance of a grinding wheel. This force presses the abrasive grains into the work surface and penetrates the surface, leading to material removal. At higher infeeds, the normal force increases causing more material removal. This force is also responsible for wheel wear, as when the thrust on the wheel exceeds certain limits, the grain fracture and bond fracture will occur. It also affects the surface roughness and the geometrical and dimensional accuracy of the finished component. Wheel clogging by grinding swarf also causes an increase in the normal force, by hindering the effective penetration of grains into the work surface. The normal forces obtained were lower with the lubricant moulded wheels compared to the standard wheel, in all the cases. The normal force components due to the frictional effects are reduced due to the effective lubrication by lubricant moulded wheels. In the earlier study on the external application of lubricant with vitrified wheels, the normal forces obtained in some
Table 1 Experimental conditions Machine Wheel
Workpiece material Cutting speed Feed Infeed Method of grinding Dressing conditions
967
BLOHM Horizontal spindle surface grinding machine, 6.5 kW 1. Standard: A60L5BM4, CUMI make 2. Graphite moulded wheels with 3%, 6% and 9% graphite by weight included in the wheel 3. CaF2 moulded wheels with 5%, 10% and 15% CaF2 by weight included in the wheel (All wheels are of 250–76.2–25 mm size) 1. Medium carbon steel (En-2), HRC 22 2. Bearing steel (En-31), HRC 60 42 m/s 10 m/min 10–80 µm in steps of 10 µm Dry With single point diamond dresser, 1 carat, at wheel speed of 2500 rpm in dry condition with 20 µm dressing depth, 苲150 mm/min cross feed and without any spark out pass
968
S. Shaji, V. Radhakrishnan / International Journal of Machine Tools & Manufacture 43 (2003) 965–972
Fig. 1.
Fig. 2.
Comparison of normal forces (per unit width of the wheel) in the case of lubricant moulded wheels with different workpieces.
Comparison of tangential forces (per unit width of the wheel) in the case of lubricant moulded wheels with different workpieces.
S. Shaji, V. Radhakrishnan / International Journal of Machine Tools & Manufacture 43 (2003) 965–972
cases were higher than those in normal grinding, due to the non-uniform supply of lubricant and increased chances of wheel clogging [12]. But here, an efficient way of lubricant supply, which minimized wheel clogging by grinding swarf, might reduce the normal force to some extent. As the wear rate is higher with resin wheels, fresh grains might also more frequently come into contact with the workpiece on subsequent passes, preventing the increase of forces due to grain dullness. The tangential force component determines the power requirement in the process. The intensity of heat generation depends upon this force and it is primarily important as far as the grinding temperatures and surface integrity aspects of the products are concerned. The lubricant effectiveness in minimizing the frictional effects at the wheel–workpiece interaction in the case of lubricant moulded wheels is evident from the reduced tangential forces compared to the standard wheel. But higher quantities of lubricant inclusion do not reduce forces significantly compared with the lowest quantity. This might be due to a possible saturation in lubricant effectiveness. The ratio of tangential force to normal force (Ft/Fn) is termed grinding coefficient [24] or grinding force ratio [20,21]. In grinding, the normal force is approximately double the tangential force, unlike in other machining
969
processes like turning and milling [16,24]. This is due to grinding process characteristics such as very large negative rake angles of the grit, excessive rubbing action and adverse chip accommodation space which leads to wheel loading. This force ratio, though it does not give the actual coefficient of friction (m), is a similar term like m and gives an indication of the frictional effects in the grinding zone [24]. Fig. 3 shows the comparison of grinding force ratio for different lubricant moulded wheels. This ratio is generally lower in all the cases of lubricant moulded wheels, compared to the standard wheel, substantiating the effective lubrication by solid lubricant in those cases. This ratio has been reported as differing considerably depending upon the material of the workpiece and indicates the workpiece behaviour in the process [20,21]. The present results show that for the same material the grinding environment providing effective lubrication changes this ratio remarkably. This may be attributed to the behaviour of the material in the presence of solid lubricant. Fig. 4 shows the comparison of specific energy, which is the energy required per unit volume of material removal, obtained with various types of wheels. The ‘size effect’, which is the drastic increase of specific energy at comparatively lower infeeds could be clearly observed in all the cases [14,16]. The energy require-
Fig. 3. Comparison of grinding force ratio in the case of lubricant moulded wheels (different workpieces).
970
S. Shaji, V. Radhakrishnan / International Journal of Machine Tools & Manufacture 43 (2003) 965–972
Fig. 4.
Comparison of specific energy in the case of lubricant moulded wheels (different workpieces).
ment in material removal can be classified as due to the sliding, ploughing and chip formation components. The size effect in grinding is due to the variation of relative amount of these components in contributing towards the unit material removal. Sliding can be due to elastic contact between the abrasive grit and workpiece and rubbing of wear flat of grains, bonding material, grinding debris etc. with the workpiece. Ploughing causes metal to be displaced sideways around the grain rather than be removed as chip. These non-productive energy components depend mostly upon the ductility and hardness of the workpiece, sharpness of the grains and the local lubrication [23]. At relatively smaller infeeds, domination of sliding and ploughing components with relatively less chip formation calls for higher specific energy [14]. The percentage contribution of these components in the total specific grinding energy at higher infeeds is reduced, because these factors of energy are more or less independent of infeed after certain limit. In the case of lubricant moulded wheels, local lubrication by solid lubricants reduces the sliding and primary ploughing component of forces to some extent and results in reduced specific energy, especially at lower infeeds. At higher material removal or infeeds the specific energy converges to more or less the same value (~15 J/mm3). This agrees with Malkin and Anderson’s [25] postulation that at higher infeeds the ploughing component of grind-
ing energy approaches zero and in the limit, the specific cutting energy (which is the sum of ploughing energy and chip formation energy) corresponds to the specific energy required for chip formation. Chip formation energy is assumed to be constant for a particular material. For ferrous material it is 13.8 J/mm3, irrespective of the hardness of the material [14]. Fig. 5 shows the comparisons of workpiece surface roughness. Surface topography of the wheel determines the nature of the tribological interaction between the wheel and workpiece. Workpiece surface roughness is the minute scratches formed by the interacting wheel surface under the dynamic conditions of the process. The surface roughness scatter is not wider for various types of wheels. This is due to the fact that for a particular wheel, the surface condition of the wheel is more or less the same for all conditions of infeed, due to the increased wear of the resin bonding. The wheel lost its dressed surface condition immediately after the initial passes. But, in the case of vitrified wheels, it can maintain the surface condition of the wheel for a longer duration as its wear rate is less [12]. Fig. 6 shows wheel wear for different wheels. The wear of graphite-moulded wheels was higher, but that of CaF2-molded wheels was lower, when compared to the standard wheel. The presence of graphite may weaken the bonding, leading to higher wear. It is notice-
S. Shaji, V. Radhakrishnan / International Journal of Machine Tools & Manufacture 43 (2003) 965–972
Fig. 5.
Fig. 6.
971
Comparison of workpiece surface roughness obtained under different grinding conditions.
Wheel wear obtained with different wheels. Fig. 7. Intermittent forces obtained in continuous grinding test.
able that the wear is very high with 9% graphite inclusion. The difference in trend with CaF2 might be due to the fusion bonding nature of CaF2 [17]. It is also noticeable that wear with 15% CaF2 inclusion is more than that with 10% CaF2 inclusion. This might be due to a saturation in the bonding strength due to the presence of CaF2 at some limit and any further increase of lubricant inclusion would be weakening the bond. This is also on par with the saturation in the lubricant effectiveness observed earlier. Fig. 7 shows the force pattern obtained from intermittent measurements during continuous grinding test. The test was done with bearing steel workpiece, giving 10
µm infeed for every upgrinding pass. Forces were measured in every 25th pass. The force values are more or less the same throughout the test. This is in contrast to the case with vitrified wheels, where forces increased to certain extent in continuous test [24]. As the wheel wear rate is higher with resin wheels, during continuous grinding, fresh grains might frequently come into interaction. But with vitrified wheels, where the wheel wear is dominated by the mechanical fracture, load on the wheel increases till the critical force required for the friability of grains occurs.
972
S. Shaji, V. Radhakrishnan / International Journal of Machine Tools & Manufacture 43 (2003) 965–972
6. Conclusion As an alternative way to extract the advantages of solid lubricant application in grinding throughout the usage of the wheel, solid lubricant moulded resin wheels were developed with graphite and CaF2 separately, by including solid lubricants during the moulding stage of the wheel. Trials to make such wheels with vitrified bonding failed due to higher vitrifying temperature and non-availability of a good reduction furnace for firing. The effectiveness of lubricants was evident from the improved process results related to friction. The wheel wear depended on the type of the lubricant used. It was higher with graphite and lower with CaF2. Saturation in lubricant effectiveness could be observed on increase of the quantity of lubricant.
References [1] R. Snoeys, K.U. Leuven, M. Maris, N.F. Wo, J. Peters, Thermally induced damage in grinding, Annals of the CIRP 27 (1) (1978) 571–581. [2] E. Brinksmeier, C. Heinzel, M. Wittmann, Friction, cooling and lubrication in grinding, Annals of the CIRP 48 (2) (1999) 581– 598. [3] T.D. Howes, Assessment of the cooling and lubricative property of grinding fluids, Annals of the CIRP 39 (1) (1990) 313–315. [4] G.S. Cholakov, T.L. Guest, G.W. Rowe, Lubricating properties of grinding fluids: comparison of fluids in surface grinding experiments, Lubrication Engineering 48 (2) (1992) 155–163. [5] E. Brinksmeier, A. Walter, R. Janssen, P. Diersen, Aspects of cooling lubrication reduction in machining advanced materials, Proc. Instn Mech. Engrs, Part B, Journal of Engineering Manufacture 213 (1999) 769–778. [6] F. Klocke, G. Eisenblaetter, Dry cutting, Annals of the CIRP 46 (2) (1997) 519–526. [7] T.D. Howes, H.K. Toenshoff, W. Heuer, Environmental aspects of grinding fluids, Annals of CIRP 40 (2) (1991) 623–629.
[8] G. Byrne, E. Scholta, Environmentally clean machining processes—a strategic approach, Annals of the CIRP 42 (1) (1993) 471–474. [9] I. Inasaki, H.K. Toenshoff, T.D. Howes, Abrasive machining in the future, Annals of the CIRP 42 (2) (1993) 723–732. [10] B. Bhushan, Modern Tribology Handbook, CRC Press LLC, New York, 2001. [11] G.W. Stachowiak, A.W. Batchelor, Engineering Tribology, Butterworth-Heinenann, Woburn, 2001. [12] S. Shaji, V. Radhakrishnan, An investigation on surface grinding using graphite as lubricant, International Journal of Machine Tools and Manufacture 42 (2002) 733–740. [13] R.I. King, R.S. Hahn, Hand Book of Modern Grinding Technology, Chapman and Hall, New York, 1986. [14] S. Malkin, Grinding Technology: Theory and Applications of Machining with Abrasives, Ellis Horwood Ltd, UK, 1989. [15] S.C. Salmon, Modern Grinding Process Technology, McGrawHill, New York, 1992. [16] M.C. Shaw, Principles of Abrasive Processing, Oxford University Press Inc., New York, 1996. [17] A.R. Lansdown, High Temperature Lubrication, Mechanical Engineering Publication Ltd, London, 1994. [18] M.S. Phadke, Quality Engineering Using Robust Design, Prentice Hall, Englewood cliffs, NJ, 1989. [19] P.J. Ross, Taguchi Techniques for Quality Engineering, McGrawHill, New York, 1996. [20] H.K. Tonshoff, J. Peters, I. Inasaki, T. Paul, Modeling and simulation of grinding process, Annals of the CIRP 41 (2) (1992) 677–688. [21] E. Brinksmeier, H.K. Tonshoff, I. Inasaki, J. Peddinghans, Basic parameters in grinding, Annals of the CIRP 42 (2) (1993) 795– 799. [22] S. Shaji, V. Radhakrishnan, Analysis of process parameters in surface grinding using graphite as lubricant based on the Taguchi’s method, Journal of Material Processing Technology, 2003 (in press). [23] S. Paul, A.B. Chattopadhayay, The effect of cryogenic cooling on grinding forces, International Journal of Machine Tools and Manufacture 36 (1) (1996) 63–72. [24] E.C. Marshal, M.C. Shaw, Forces in dry surface grinding, Transactions of the ASME 74 (1952) 51–59. [25] S. Malkin, R.B. Anderson, Thermal aspects of grinding, Part 1— Energy partition, Transactions of the ASME, Journal of Engineering for Industry 98 (1974) 1177–1183.