Grinding Wheel Condition Prediction and Improement Ping Zhang, Michele H. Miller Michigan Technological Uniersity, Houghton, MI Introduction Grinding is regarded as a good way to do ceramics machining due to its multiple abrasie cutting particles. While a single point tool may dull and require re-sharpening, a grinding wheel distributes wear oer many cutters and is capable o replenishing itsel. Howeer, the quality o the inished workpiece greatly depends on the constantly changing condition o the grinding wheel. Indiidual grits dull, and some racture or pull out o the binder. Other grits at dierent locations with dierent shapes and sizes replace the lost grits as cutting grits. Chips accumulate in the spaces between grits thus remoing clearance or subsequent chips. Chip accumulation (or loading) is particularly problematic with the ine grit wheels used in ine grinding o ceramics. Wheel loading occurs when workpiece chips either adhere to the grits or embed in the space between grits. [1] Wheel loading aects the eiciency o the grinding process as well as the lietime o the grinding wheel in ine grinding. [2] With wheel loading, orces and temperatures can increase to leels that cause poor surace inish, reduced material remoal rate, and high wheel wear rate. To maintain the grinding wheel condition with little chip loading, dressing is usually used. Traditionally, the grinding operation is interrupted to dress grinding wheels using methods such as stick or wire brush dressing. Seeral in-process dressing methods hae been deeloped such as mechanical in-process dressing, electrolytic in-process dressing, electro-contact dischae dressing and waterjet in-process dressing. These oer the adantages o consistent wheel condition oer time as well as uninterrupted operation. In order to make better use o the arious dressing methods, our research seeks to understand the actors leading to wheel condition deterioration. We ocus speciically on chip loading and study how actors such as chip size, wheel characteristics, work material, temperature, and coolant inluence the rate o chip accumulation. Computer Modeling The literature indicates that the main cause o chip (or wheel) loading is adhesion between the actie grits and the workpiece material deormed in the orm o chips. [2] Chips accumulate in the
spaces between grits. Thus the relationship between chip olume and olume between grits is likely to inluence the tendency or wheel loading. These quantities were inestigated using a kinematic based simulation. [] The wheel/work interace or the surace grinding geometry shown in Fig. 1 was modeled. grinding wheel z y workpiece Figure 1: Surace grinding geometry Based on the grinding conditions and wheel characteristics, the simulation program calculates chip olumes. The olume between grits is assumed as the binder olume per grit. It can be calculated rom the grit size and abrasie olume raction. Approximating the grits as spheres with radius r g, the grit olume is: 4 grit = πr g (1) The abrasie olume raction o the grinding wheel is deined as: Soling or the olume between grits: Combining (1) and (), grit = (2) + grit = = grit 1 1 () 4 π r g (4) Fig. 2 shows the relationship o chip olume and olume between grits or grit radii ranging rom 25 µm to 175 µm and olume ractions ranging rom 12.5% to 7.5%. Though not shown in the igure, both olumes increase with increasing grit size and decreasing olume raction. The igure indicates that the olume between grits grows more quickly than the olume o chips as the grit size increases and/or olume raction decreases. A higher ratio o / chip would reduce the tendency or chip accumulation. This result seems to alidate the common obseration that ine grit wheels experience more chip loading than coarse grit wheels.
(um ) 2.0E+08 1.8E+08 1.6E+08 1.4E+08 1.2E+08 1.0E+08 8.0E+07 6.0E+07 4.0E+07 2.0E+07 0.0E+00 0.0E+00 5.0E+05 1.0E+06 1.5E+06 2.0E+06 chip (um ) Wheel rpm=2907 Table speed=100 mm/s Depth o cut=25 µm Workpiece width=100 µm Grit radius=25-175 µm olume raction=12.5-7.5% Fig 2: Relationship o chip olume to olume between grits Equation (4) describes the eect o grit radius and olume raction on olume between grits. In summary, the olume between grits is proportional to r g and approximately proportional to -1. We must use the simulation to determine the eect o grit radius and olume raction on chip olume. Fig. shows chip olume results as a unction o olume raction or 4 dierent grit sizes. Fitting trend lines to these points indicates that 0.8 (5) chip Fig. 4 shows chip olume results as a unction o grit radius or 4 dierent olume ractions. Ater adding trend lines or these points, we get the ollowing equation: 2.17 chip r g (6) The ratio o olume between grits to chip olume is thus approximately: chip 1 0.2 0.8 = (7) 0.8 2.17 This ratio can be increased (and the tendency or wheel loading reduced) by increasing grit size or decreasing olume raction. The ratio is much more sensitie to changes in grit size than to changes in olume raction.
chip s Concentration Grit radius s chip chip(um ) 10000000 1000000 100000 10000 =25 =65 =115 =175 chip(um ) 10000000 1000000 100000 10000 =0.125 =0.2 =0.25 =0.75 1000 0.1 1 Concentration 1000 10 100 1000 Grit Radius(um) Fig : Relationship between chip olume and wheel concentration Fig 4: Relationship between chip olume and grit radius Experimental Method We hae begun experiments to test our conclusions about the eects o grit size and concentration on wheel loading. Other researchers hae used chemical detection, calorimetry, spectroscopy, eddy current sensing, magnetization, radiotracing, and x-ray luorescence to ealuate wheel loading. We hae adopted a simple approach or quantiying wheel loading that uses microscope images and image analysis sotware. [4] Beore the experiment, we set the workpiece, a truing wheel, and an optical microscope on the worktable o the grinding machine. Ater regular interals we take seeral microscope images at dierent locations on the grinding wheel. Figure 5 shows a sample image o an aluminum oxide wheel ater grinding steel. We then use Scion Image sotware to conert the image to black and white (shown in Figure 6) and to compute the percentage o black area. This percentage is our measurement o wheel loading. Next, we aerage the alues rom the dierent locations. Fig 5: Microscope image o the grinding wheel Fig 6: Black and white image
In our irst experiments, we used this method to study the loading o aluminum oxide wheels grinding steel. Next, we plan to use the microscope and image analysis sotware to study the loading condition o diamond wheels grinding ceramic workpieces. Conclusion Our computer modeling results alidate the common obseration that ine grit wheels experience more chip loading than coarse grit wheels. We also get equations to predict chip olume based on wheel concentration and grit radius. Our current plans are to analyze wheel loading using a microscope and image analysis sotware. The inal goal or this research is to study how chip size, wheel characteristics, work material, temperature, and coolant aect the rate o chip accumulation, and then use these results to improe dressing techniques. Reerences 1. Nagaraj, A. P. and A. K. Chattopadhyay, On Some Aspects o Wheel Loading, Wear, ol. 15, pp. 41-52, 1989. 2. Sriastaa, A.K., K. Sriram and G.K. Lal, A Simple Analysis or Ealuating Grinding Wheel Loading, Int. J. Mach. Tool Des. Res. ol. 28, No. 2, pp. 181-190, 1988.. Miller, M. H. and K. S. Moon, The Eect o Workpiece Modulation on Grinding Kinematics, Proc. o the ASPE Spring Topical Meeting, pp. 70-75, 1996. 4. Sriastaa, A. K., K. Sriram and G. K. Lal, A New Technique or Ealuating Wheel Loading, Int. J. Mach. Tool Des. Res., ol. 25, No. 1, pp. -8, 1985.