Three-Dimensional Shape Modeling of Diamond Abrasive Grains Measured by a Scanning Laser Microscope

When a scanning laser microscope is used as a tool for measuring the three-dimensional shape of abrasive diamond grains, the result is often distorted due to the presence of noise. After the noise has been removed, a three-dimensional modeling is considered in order to enhance the shape obtained. In the proposed modeling method, the grain body is extracted from the bond surface and horizontally divided into layers. The contour of each layer is modeled by a technique called the cam and follower method, and then the modeled layers are overlapped in order to generate the modeled grain shape. The obtained results are employed for measurements of various shape features of diamond abrasive grains. Introduction The precise measurement of the three-dimensional (3D) shape of diamond abrasive grain is required for various studies on grinding. For example, the grinding wheel wear is attributed to bond fracture, grain fracture, and attritional grain wear [1]. The studies on the last two wear mechanisms require shape measurement of grains under observation. Studies on the effect of grain shape on ground surface integrity [2] and grinding mode [3] also necessitate this measurement. Accurate information on grain shape is also required for a reliable simulation of the grinding process [4]. In order to realize this measurement, various techniques are considered. As for profilometry, 2D (cross-sectional) and 3D data can be obtained [5], but the poor penetration of the stylus tip results in undesirable low accuracy [1]. The scanning electron microscope (SEM) is a powerful measuring tool [6], but the time needed for sample coating process and chamber air pumping is considerable. A system of a laser beam probe with a fixed beam and moving x-y stage is also applicable [7]. This system is similar in principle to profilometry; it employs a laser beam as an optical stylus. Since the problem of stylus tip penetration does not exist in this system, the accuracy is improved, but the time consumed by stage movement is considerable. Compared with the above-mentioned systems, the scanning laser microscope (SLM) is easy and quick to operate and provides good accuracy, but its measurement results are often distorted due to the occurrence of noise. In order to solve this problem, the noise types and the reasons for noise occurrence are investigated [8], and noise is removed by a method developed by the authors. As a subsequent step, this study proposes a way for 3D shape modeling to be applied after noise removal. This approach is observed to smooth the glitches left behind by noise removal and to enhance the obtained shape features in order to achieve increased precision of measurement results. Specimen and Measurement Procedure The measured grains are taken from a metal-bonded diamond grinding wheel (SD#270N100 M) dressed by electro-contact discharge method. The measuring tool is a laser scanning microscope, model 1LM15W, fabricated by Lasertec Corporation. For measurements, the optimal conditions Key Engineering Materials Online: 2003-04-15 ISSN: 1662-9795, Vols. 238-239, pp 131-136 doi:10.4028/www.scientific.net/KEM.238-239.131 © 2003 Trans Tech Publications Ltd, Switzerland All rights reserved. No part of contents of this paper may be reproduced or transmitted in any form or by any means without the written permission of Trans Tech Publications Ltd, www.scientific.net. (Semanticscholar.org-13/03/20,21:50:16) that reduce the occurrence of noise [8] are applied. These conditions are a slow scan rate of 1/16 the normal rate, and a period of 55~80 sec for vertical movement of the stage. A desktop personal computer (Pentium III, 667 MHz) is used for data processing, where the processed image is 320×240 pixels (1 pixel = 0.375 μm) with 0~255 grayscales for the pixel value. The programming language used is Visual Basic 5. Noise Removal and Patching As illustrated in Fig. 1, modeling is the final stage in shape processing, and Fig. 2 presents an example of the steps prior to modeling. The result of grain shape measured using the SLM is shown in Fig. 2(b). There presents an explicit instance of the distortions caused by noise occurrence. In order to remove these distortions, noise is removed as shown in Fig. 2(c), and the openings left behind by noise removal are patched as shown in Fig. 2(d). The obtained result yields a shape with clearer features of the measured grain. However, by comparison with observations via SEM, as shown in Fig. 2(a), the obtained shape shows wavy surfaces with vague edges. Therefore, shape modeling is applied to remove these drawbacks in order to perform quantitative evaluations on the measured shape of the grain that are more precise. Fig. 2 Example illustrating the processing steps prior to modeling (b) 3D view of raw data obtained by SLM (a) Diamond grain observed by SEM (d) After patching (c) After noise removal Fig. 1 Flowchart of processing steps Shape processing Measurement by SLM Noise removal Patching Modeling 132 Advances in Abrasive Technology V