Manufacture, Faculty of Engineering, University of Malaya, Kuala Lumpur, Malaysia

Transcription

1 Advanced Materials Research Vols (2011) pp (2011) Trans Tech Publications, Switzerland doi: / Optimizing the Machining Parameters in Glass Grinding Operation on the CNC Milling Machine for Best Surface Roughness M.Sayuti 1,a, Ahmed A. D. Sarhan 2,b, M. Hamdi 3,c 1,2,3 Centre Of Advance Manufacturing and Material Processing, Department of Design and Manufacture, Faculty of Engineering, University of Malaya, Kuala Lumpur, Malaysia a mdsayuti@siswa.um.edu.my, b ah_sarhan@um.edu.my, c hamdi@um.edu.my Keywords: CNC Milling, Surface roughness, Glass grinding, Machining parameters, Taguchi Optimization Method Abstract. Glass is one of the most difficult materials to be machined due to its brittle nature and unique structure such that the fracture is often occurred during machining and the surface finish produced is often poor. CNC milling machine is possible to be used with several parameters making the machining process on the glass special compared to other machining process. However, the application of grinding process on the CNC milling machine would be an ideal solution in generating special products with good surface roughness. This paper studies how to optimize the different machining parameters in glass grinding operation on CNC machine seeking for best surface roughness. These parameters include the spindle speed, feed rate, depth of cut, lubrication mode, tool type, tool diameter and tool wear. To optimize these machining parameters in which the most significant parameters affecting the surface roughness can be identified, Taguchi optimization method is used with the orthogonal array of L 8 (2 6 ). However, to obtain the most optimum parameters for best surface roughness, the signal to noise (S/N) response analysis and Pareto analysis of variance (ANOVA) methods are implemented. Finally, the confirmation test is carried out to investigate the improvement of the optimization. The results showed an improvement of 8.91 % in the measured surface roughness. Introduction Glass is an amorphous solid that is a favored material for a lot of reasons. It resists chemical interactions, it is easy to recycle, it does not leach chemicals like plastics do, and it can withstand extremes of heat and cold, although not at the same time. There are many types of glassy materials available in the market, however, soda lime glass is the most prevalent type of glass as it is widely used and easily can be found in the market. Soda lime glass is also plays an essential role in industry such that, the demand for fabrication has been increasing to generate diversified functionalities on many applications [1-2]. Although some manufacturing processes such as chemical etching have been made to fabricate micro patterns on glass, the process takes a long time and is hazardous. Other approaches, therefore, are required to manufacture glass products at a high production rate, higher surface quality and easier operations [3]. Recently, many researchers have studied the machining processes for these important brittle materials [4-6]. Most have tried to machine glass using CNC milling machine and focused on the transition from ductile to brittle mode. According to the research, glass can be machined in a ductile mode when the undeformed chip thickness is less than a micrometer. Machining processes, however, have not yet attained at a high production rate. Therefore, the application of grinding process on the CNC milling machine would make it an extremely efficient process but the poor surface finish produced is a common problem when machining materials harder than 50 HRC [1-3]. 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 the publisher: Trans Tech Publications Ltd, Switzerland, (ID: /10/10,07:37:48)

2 722 Materials Processing Technologies To improve the surface roughness, it requires adopting the right machining parameters. However, due to the large number of the machining parameters related to the glass grinding operation on the CNC milling, the trial and error approach is very time consuming. Hence, a reliable systematic approach to optimize the machining parameters for best surface roughness is thus required [7-8]. Taguchi optimization method is an efficient and effective approach, in which the response parameters that affecting surface roughness can be optimized to identify the most significant response variables with the minimum number of experiments [9]. In this study, the glass grinding process on the CNC milling is carried out; several machining parameters are considered to be significant in affecting surface roughness. These parameters include, the velocity, feed rate, depth of cut, lubrication mode, tool type, and tool diameters (called as control factors here after). The Taguchi orthogonal array of L 8 (2 6 ) is used and the tool wear is considered as a noise factor. The (S/N) response analysis and Pareto analysis of variance (ANOVA) are used to identify the most significant parameters affecting the surface roughness. Finally, confirmation tests are carried out to investigate the improvement of the optimization. The Experimental Set-up, Design and Procedure The experimental set up. The machine used in the study is a vertical-type machining center (Cincinnati Milacron Saber TNC750 VMC). The spindle has constant position preloaded bearings with oil-air lubrication, and the maximum rotational speed is 20,000 min -1. However, to get higher speed to investigate the surface roughness in both normal and high-speed modes, an attachable High-speed spindle (HSS) unit (Model NSK HES510 BT40) with min -1 maximum speed is attached to the machine spindle using a built-in rigid holder to maintain the machining accuracy. The cutting process of a rectangular workpiece of soda lime glass 80 x 50 x 25 mm is selected as a case study. To investigate the effects of the grinding tool type parameter on the glass grinding operation, the diamond and CBN internal grinding tools types are used with different tool diameters. However, to control the cutting temperature and to reducing friction between the tool and workpiece for better surface roughness, the lubricant system is used in the machining process. The Shell Dromus BL lubricant type has been selected since it has a good lubrication quality characteristic and it commonly used in milling. To measure the surface roughness (R a ) of the machined glass workpiece surface, the stylus profilometer (MPi Mahr Perthometer) is used with a cut distance of 5 mm. However, for more stability and to ensure more accurate measurements of surface roughness, a polished granite surface table is used. The measured surface roughness are reported as the average of three measurements taken along the left, middle and right of the machined surface. The experimental design and procedure. The most important stage in the design of an experiment lies in the selection of control factors and identifying the orthogonal array (OA). As many factors as possible should be included, so that it would be possible to identify non-significant variables at the earliest opportunity. In this research work the standardized Taguchi orthogonal array L 8 (2 6 ) is used. This orthogonal array consists of eight experiments with six control factors and two different experimental condition levels for each factor. The factors and levels are specified in Table 1. The eight experiments with the details of the combination of the experimental condition levels for each control factor (A F) are shown in Table 2. The spindle speed has been selected to investigate the surface roughness in both normal and high-speed modes. While, the axial depth of cut and feed rate values are selected based on the capability of the grinding tool to avoid critical damage and also for safety reason. It is however should be noted that the tool wear considered as a noise factors. Its effect on experiment is investigated as well in two different modes. The first mode is new tool referred as P1, while the second mode is also a set of new tool with the same geometric specification but with light tool wear

3 Advanced Materials Research Vols referred as P2. The light tool wear on the inserts is created by lightly grinding the cutting edge with a small abrasive grinder. It is however, very difficult to control the degree of tool wear when grinding the inserts. Hence, a microscope is used to observe and measure the wears on the inserts to control the wear situation of the inserts as similar as possible. Because the difficulty still existed due to the researchers inability to reproduce the exactly same tool wear situation, tool wear is considered a noise factor in this study. The light tool wear here means the existing of some friction marks indicating slight wear. Table 1: The control factors and the experimental condition levels Control Factors Experimental condition (Levels) Level 1 Level 2 A Tool type CBN Diamond B Tool diameter (mm) 3 6 C Lubricant mode Dry Lubricant D Feed rate (mm/min) E Spindle speed (min -1 ) 20,000 35,000 F Depth of cut (mm) Table 2: Standard Orthogonal array L 8 (2 6 ). The eight experiments with the details of the combination of the experimental condition levels Test No. (j) Control factors and levels A B C D E F Results, Analysis and Discussion The experimental results. After the orthogonal array has been identified, the next step in Taguchi optimization method is running the experiment based on that OA. The machined glass surface roughness R a is measured off line after each experiment as a function of the tool wear modes P1 and P2. Each measurement repeated three times and the averages are calculated and summarized in Table 3. The data analysis. The procedures after the experimental runs are analyzing data to optimize the parameters and to identify which process parameters are statistically significant. Analyzing data is conducted using signal to noise (S/N) response analysis and analysis of variance (Pareto ANOVA). (S/N) response analysis. The method for calculating the S/N ratio are classified into three main categories, depending whether the desired quality characteristics is smaller the better, larger the better or nominal the better. In the case of surface roughness, the smaller values are required. The equation for calculating the S/N ratio for smaller the better characteristic (in db) is as the following: 2 ( P1 P2 ) j 1 = -10 log + N (1) n S 2 Where (j) is the test number from 1 to 8 and n is the number of noise factors, in this case n is equal to 2. The S/N values function as a performance measurement to develop processes insensitive to noise factors. The degree of predictable performance of a product or process in the presence of noise factors could be defined from S/N ratios. For each type of the characteristics, the higher the S/N ratio the better is the result. The calculated S/N ratio and TPM values are summarized in Table 3. Where; TPM is the target performance measurement. It is equal to the average of the measured surface

4 724 Materials Processing Technologies roughness under both of P1 and P2 at same level of input parameters. Furthermore, The S/N and TPM response data are calculated and ploted in Fig.1. As for example of S/N or TPM response calculation, A i is the average of all S/N or TPM in Table 3 that has same experimental level (i) under A, shown in Table 2. In this case (i) is equal to 1 or 2. Similarly the S/N, TPM response values are calculated for B i, C i, D i, E i, and F i. The desired smaller the better criteria implies that the lowest surface roughness and TPM would be the ideal result. While, the largest S/N ratio would reflect the best response, which result in the lowest noise influence on the machine set up. This is the criteria employed in this study to determine the optimal machining parameters. As can be seen in Fig.1 and according to the smaller TPM and higher (S/N) ratio base, the lubrication mode is the most significant factor affecting the surface roughness (R a ) followed by depth of cut, cutting speed, feed rate and tool type. However, the tool diameter is found to be insignificant as the slope gradient is very small. Using CBN tool (A1) with highest cutting speed (E2, min -1 ), lowest feed rates (D1, 0.5 mm/min), lowest depth of cut (F1, 0.5 mm), and tool diameter (B1, 3 mm) with the lubrication mode (C2) seems to be the best choice for obtaining the best surface roughness. These values are determined and selected based on the largest (S/N) and lowest (TPM) response in each parameter. Therefore, the optimal parameter is set as A1 B1 C2 D1 E2 F1. At this level, the confirmation test is carried out to validate the finding. Measurements are reported as the average of eight measurements. The surface roughness obtained from this confirmation test is found to be 0.94 µm. The result shows an improvement of 8.91% compared with the lowest value of the surface roughness obtained during the experiments shown in Table 3. TPM (µm) A1 - CBN A2 - Diamond B1-3 mm B2-6 mm C1 - Dry C2 - Lubricant TPM and S/N D1-0.5 mm/min D2-1 mm/min Control Factors E min-1 E min-1 F1-0.5 mm F2-1 mm Fig. 1: Response Graph for TPM and S/N values with different control factors S/N (db) TPM S/N Table 3: The measured surface roughness with the calculated TPM and S/N ratio Measured surface roughness (R a µm) Test No. (j) New tool P1 Light wear P2 Calculated Parameters TPM (µm) S/N (db) Analysis of variance (Pareto ANOVA): an alternative analysis. An alternative way to analyze the data for the optimization process is the analysis of variance using Pareto ANOVA, which is, simplified ANOVA by using Pareto principles. Pareto ANOVA for surface roughness has been constructed in Table 4 by using the S/N response data. The summation of squares of differences (S) for each control factor is calculated such that for example (S A ) can be obtained by the following equation: S A - 2 = ( A1 A2 ) (2) Similarly, S B, S C, S D, S E, and S F are calculated. The contribution ratio for each factor is calculated as the percentage of summation of squares of differences for each factor to the total summation of squares of differences. Pareto diagram is plotted using the contribution ratio and the cumulative contribution. The significant factors are chosen from the left-hand side of the Pareto

5 Advanced Materials Research Vols diagram, which cumulatively contribute up to 99%. It can be seen that the best level of factor combination for best surface roughness are found to be; A1 B1 C2 D1 E2 F1, which is similar to the result obtained from S/N response analysis. Level of input parameters Table 4: Pareto ANOVA analysis for surface roughness S/N response data (db) A i B i C i D i E i F i Tool type Tool diameter Lubrication mode Feed rate Spindle speed Depth of cut Level Level Difference Square of difference S A =1.592 S B = S c =8.095 S D =1.960 S E =2.163 S F =4.436 St=S A +S B +S C +S D +S E +S F Contribution Ratio (%) Pareto ANOVA analysis Contribution ratio (%) C F E D A B Factor Contribution Ratio (%) Cumulative Contribution Discussion In the above case study two techniques of data analysis have been used. Both techniques draw similar conclusions. The Lubrication mode has found to be the most significant effect to produce lower surface roughness (R a ). This could be explained in terms of the cutting temperatures, it is however still high as the friction between cutting tool and workpiece is also high due to extensive contact during machining. The cutting temperature is a key factor, which directly affects tool wear, workpiece surface integrity and machining precision according to the relative motion between the tool and workpiece [6]. The amount of heat generated varies with the type of material being machined and cutting parameters especially cutting speed, which had the most influence on the temperature [6]. Such that at high spindle speed the transition from brittle machining mode to ductile machining mode occurs [4]. Hence, the higher cutting temperature associated with the high spindle speed increases softening of the workpiece material and then reduces the required cutting forces leading to better surface roughness. However, it is believed that the spindle speed should be controlled at an optimum value, as the influence of temperature, especially for low thermal conductive materials such as glass, would significantly affect the chip formation mode, cutting forces, tool life, and surface roughness. Thus the implementation of cutting fluid, which acts as a coolant, is very crucial. In addition, it is not only important as a coolant but also during cutting, the friction coefficient is reduced and the lubrication features enhanced, thus improve the surface roughness of the machined workpiece. The feed rate and the depth of cut are also significant machining parameters affecting surface roughness. Both the feed rate and depth of cut are found to be at lowest level to obtain the best surface roughness. The combination of feed rate and depth of cut determines the undeformed chip section and

6 726 Materials Processing Technologies hence the amount of the cutting forces required to remove a specified volume of material. By decreasing feed rate and depth of cut, less material has to be cut per tooth per revolution, thus energy required is lower. Consequently, this would reduce the cutting forces, which is required for better surface roughness as well as to preserve material properties against residual stress and change in micro hardness at the subsurface. This would explain why feed rate and depth of cut play a key role in obtaining best surface roughness. Conclusion This paper studded the optimization of the machining parameters for best surface roughness in glass grinding operation on the CNC milling machine. To optimize the machining parameters, the Taguchi optimization method is used. The standard L 8 (2 6 ) orthogonal array is used and further analysis conducted using signal to noise ratio (S/N) response analysis and the analysis of variance (Pareto ANOVA) to determine which process parameters are statistically significant. The experimental results have indicated that the lubrication mode is found to be the most significant factor during the glass grinding operation followed by depth of cut, spindle speed, and feed rate. The tool diameter and tool type is found to be the least significant factors. In addition, the noise factor, namely the tool wear, is found to be statistically significant. The optimal parameters for minimum surface roughness is A1 B1 C2 D1 E2 F1 corresponding to a CBN tool (A1) with the 3 mm tool diameter (B1), lubrication mode (C2), 0.5 mm/min feed rate (D1), 35000min -1 cutting speed (E2), and 0.5 mm depth of cut (F1). The surface roughness results achieved in the confirmation test indicated that the optimal parameters obtained through the Taguchi optimization method produced best surface roughness in glass grinding operation and that is achieved with a relatively small number of experimental runs. References [1] F. James Shackelford and H. Robert Doremus, eds. Ceramic and Glass Materials: Structure, Properties and Processing. (2008), Springer. p [2] E.L. Bourhis, Glass: Mechanics and Technology. (2007): Wiley-VCH. [3] R. Doering, and Y. Nishi, Handbook of semiconductor manufacturing technology. (2007): CRC Press [4] Z.W. Zhong, Ductile or partial ductile mode machining of brittle materials. International Journal of Manufacturing Technology, (2003) 21: p [5] T. Shirakashi, and T. Obikawa, Feasibility of gentle mode machining of brittle materials and its condition. Journal of Materials Processing Technology, (2003) 138: p [6] G.Z. Xie, and H. Huang, An experimental investigation of temperature in high speed deep grinding of partially stabilized zirconia. International Journal of Machine Tools and Manufacture, (2008) 48: p [7] G.Z. Xie, H.W. Huang, H. Huang, X.M.Sheng, H.Q.Mi, W. Xiong, Experimental investigations of advanced ceramics in high efficiency deep grinding Chinese Journal of Mechanical Engineering, (2007) 43: p [8] T.W. Hwang, C.J. Evans, and S. Malkin, An investigation of high speed grinding with electroplated diamond wheels Annals of CIRP, (2000) 49 (1): p [9] J.Z. Zhang, J.C. Chen, and E.D. Kirby, Surface roughness optimization in an end-milling operation using the Taguchi design method. Journal of Materials Processing Technology (2007) 184: p

7 Materials Processing Technologies doi: / Optimizing the Machining Parameters in Glass Grinding Operation on the CNC Milling Machine for Best Surface Roughness doi: /