CHAPTER 4. OPTIMIZATION OF PROCESS PARAMETER OF TURNING Al-SiC p (10P) MMC USING TAGUCHI METHOD (SINGLE OBJECTIVE)

Transcription

1 55 CHAPTER 4 OPTIMIZATION OF PROCESS PARAMETER OF TURNING Al-SiC p (0P) MMC USING TAGUCHI METHOD (SINGLE OBJECTIVE) 4. INTRODUCTION This chapter presents the Taguchi approach to optimize the process parameters in turning Al-MMC reinforced with 0% volume fraction of silicon carbide (SiC) particulates. The process parameters under consideration are cutting speed, feed rate and the depth of cut. The optimum levels of the process parameters are determined using the procedure explained by Taguchi s concept of quadratic loss function and uses a statistical measure of performance called Signal-to- Noise (S/N) ratio. The predicted optimal value of surface finish and power consumption is confirmed by conducting the confirmation experiment using optimum parameters. 4. OPTIMIZATION STEPS IN TAGUCHI METHOD Taguchi techniques have been used widely in engineering design. The main trust of the Taguchi techniques is the use of parameter design, which is an engineering method for product or process design that focuses on determining the parameter settings producing the best levels of a quality characteristic (performance measure) with minimum variation (Phadke 989; Ross 998). The procedure and the steps to be followed for process parameter optimization in Taguchi method is shown in Figure 4..

2 56 Figure 4. Optimization steps in Taguchi method 4.3 SELECTION OF PROCESS PARAMETERS AND THEIR LEVELS In order to identify the process parameters that affect the quality of the turned parts, an Ishikawa cause-effect diagram was constructed as shown in Figure 4.. The Ishikawa cause-effect diagram depicts that the following process parameters may affect the quality of the turned parts: Cutting parameters: cutting speed, feed rate, depth of cut. Environment parameters: wet, dry. Cutting tool parameters: tool geometry, tool material. Workpiece material: hot worked, cold worked, difficult to machine.

3 57 It has been clearly shown in the literature (Quigley et al 994, Paulo Davim et al 00b, Manna et al 005, Muthukrishnan et al 008b and Seeman et al 00) that machining process parameters such as cutting speed, feed rate and depth of cut significantly influence the process and play a major role in deciding the quality of the surface finish and power consumption. PCD tooling is the only economical tool material for SiC p reinforced MMC and useful lives of between 0 and 50 minutes can be obtained by using speeds in the range m/min (Heath 00). When the cutting speed is decreased to 75 m/min produce high surface roughness, for feed rate of 0.4 mm/rev and depth of cut of mm and the dependence of the depth of cut is not significant (Manna et al 003). Cutting speed zone between 60 m/min to 50m/min is recommended for machining AlSiC MMC, where cutting forces are more or less independent of cutting speed (Manna et al 005). Figure 4. Ishikawa cause-effect diagram of a turning process

4 58 It is therefore, based on the review of literature, the following parameters were identified as potentially important in affecting the quality features of the turned parts under study. cutting speed; feed rate; depth of cut; tool material PCD fine grade inserts; work material Al-SiC(0P) and Al-SiC(5P) MMCs; and Environment dry. Table 4.. The process parameters and their levels (and notations) are listed in Table 4. Process parameters values and their levels Symbol Process parameter Unit Level Level Level 3 A Cutting Speed m/min B Feed rate mm/rev C Depth of cut mm SELECTION OF ORTHOGONAL ARRAY (OA) The selection of orthogonal array (OA) to use predominately depends on these items in order of priority: The number of factors and interactions of interest The number of levels for the factors of interest The desired experimental resolution or cost limitations.

5 59 The orthogonal array (OA) selected should satisfy the following inequality (Ross 998): Degrees of freedom (DOF) of OA Total DOF required for the experiments. As three levels and three factors are taken into consideration, L9 OA is used in this investigation. Only main factor effects are taken into consideration and not the interactions. The degrees of freedom (DoF) for each factor is (No. of levels, i.e. 3-=) and therefore the total DoF required was 8 (xx). Generally the DoF of the OA should be greater than or equal to the total DoF of the factors. As the DoF of L9 OA is 8, it can be suitable for the study. 4.5 EXPERIMENTAL PROCEDURE The experiments for turning operations were conducted according to the Taguchi s L9 orthogonal array (OA), which has 9 rows corresponding to the number of tests with 4 columns at three levels. The first column was assigned to the cutting speed (A), the second column to feed rate (B) and third column to the depth of cut (C). It considers three process parameters to be varied in three discrete levels. Each experimental run were replicated thrice and corresponding response values (results) for surface roughness (Ra) and power consumed (Pc) are listed in Table 4.. The surface roughness (Ra) measurements, in the transverse direction, on the work pieces have been repeated three times and average of three measurements values has been recorded for each replication. The power consumed (Pc) by main spindle was measured using digital watt meter.

6 60 Table 4. Experimental results using L9 OA for Al-SiC P (0P) MMC Exp. Run Process Parameters Surface roughness in microns Power consumption in kilo Watts A B C Rep Rep Rep 3 Rep Rep Rep PERFORMANCE MEASURE Taguchi Signal-to-Noise (S/N) ratio is used as the performance index for the evaluation of responses and to determine the optimal combination of control parameters. The term signal represents the desirable value, i.e., mean for the output characteristics and the noise represents the undesirable value, i.e., the square deviation for the output characteristics. It is a summary statistics and denoted by, and the unit is decibels. Usually, there are three S/N ratios available, depending on the type of characteristic; the smaller-the better (STB), the larger-the better (LTB), and the nominal- the better (NTB). The formula for signal to noise ratios is designed such that the experimentalist can always select the larger factor level settings to optimize the quality characteristics of an experiment. Therefore, the method of calculating the signal to noise ratio depends on whether the quality characteristics has smaller-the-better, larger-the-better or nominal-the- better

7 6 formulation is chosen. In turning, lower surface roughness and power consumption and higher tool life are indications of better performance. Therefore, for obtaining optimum machining performance, the Smaller-the-better S/N ratio is used for both the quality characteristics surface roughness and power consumed. The formula used for calculating the S/N ratio (decibels) of the smaller-the-better quality characteristic is given in Equation (4.). n S / N( ) 0x log y n i i (4.) Where, n - number of replications y i - observed response value The surface roughness (Ra) and power consumed (Pc) of the turning process is analyzed to study the effects of the process parameters. The experimental data are converted into mean and S/N ratio. For example the mean value of Ra for the first experimental run is calculated as the average of three replication results such as.65,.7 and.55 thus gives the mean value of.64. Similarly the mean values for Ra and Pc is calculated for all the experimental runs. The S/N ratio for each experimental run is calculated using the formula given in equation (4.). For example the S/N ratio for Ra in first experimental run was calculated as (-0 log 0 (( )/3)) thus gives a value of db. Similarly the S/N ratio values for Ra and Pc is calculated for all the experimental runs. The calculated mean values and S/N ratio values for each experimental run are tabulated in Table 4.3.

8 6 Table 4.3 Mean value and S/N ratio for Ra and Pc Exp. Run Process Mean value S/N ratio Parameters A B C Ra (µm) Pc (kw) Ra (db) Pc (db) RESULTS AND DISCUSSION Since the experimental design is orthogonal, it is then possible to separate out the effect of each cutting parameter on the performance measures (Mean and S/N ratio) at different levels (Lin et al 00). For example, the average S/N ratio for the cutting speed at levels, and 3 can be calculated by averaging the S/N ratios for the experiments to 3, 4 to 6, and 7 to 9, respectively. For example, the average S/N ratio for cutting speed at level is calculated as the average of S/N ratios such as , -.0 and thus give the value of Similarly average S/N ratio for cutting speed at level is (average of , and -6.04), for cutting speed at level 3 is (average of , -.83 and -.55). The average S/N ratio for each level of the other cutting parameters can be computed in the similar manner. Similarly the main effects for the average mean value for each level of the cutting parameters can be computed. For example, the

9 63 average mean value for cutting speed at level is calculated as the average of.64, 3.55 and 6.34 thus give the value of Similarly average mean value for cutting speed at level is 4.90 (average of.4, 3.84 and 6.3), for cutting speed at level 3 is 3.50 (average of.4, 3.90 and 4.4). The main effects for both mean and S/N ratio values of all levels of the cutting parameters are calculated and listed for surface roughness and power consumption in Tables 4.4 and 4.5 respectively. Irrespective of the objective function whether maximization or minimization, the larger S/N ratio corresponds to the better quality characteristics. Therefore, the optimal level of the process parameters is the level with the highest average S/N ratio. Based on both mean and S/N ratio values, the optimal level setting is A3BC. i.e., the cutting speed is to set at 80 m/min, the feed rate has to be set at 0.mm/rev and the depth of cut is to be 0.6mm based on the experimental results for the objective of minimum surface roughness. Table 4.4 Main effects (Response) table for surface roughness Process Average Mean Average S/N Ratio parameter Level A B C A B C Average value * * * * * * Delta (Max-Min) Rank 3 3 *Optimum levels A 3 B C The main effects plot for surface roughness based on S/N ratio is shown in Figure 4.3. The x-axis represents the levels at which each factor was varied and y-axis shows the resultant change in the S/N ratio. The overall mean S/N ratio has been indicated by a horizontal line. The maximum S/N

10 64 ratio value is observed for the cutting speed at level 3. The decrease in S/N ratio is observed with increase in the level of feed rate. The maximum S/N ratio is observed when feed rate is set at the lowest level of 0.mm/rev and depth of cut at the 0.6mm (level ). Main Effects Plot (data means) for SN ratios -8 A B C Signal-to-noise: Smaller is better 3 Figure 4.3 Main effects plot for surface roughness S/N ratio The main effects, average mean and S/N ratio values of all levels of the cutting parameters are calculated and listed for power consumption in Table 4.5. Based on mean and S/N ratio values for power consumption, the optimal level setting is ABC. i.e., the cutting speed is to set at 75 m/min, the feed rate has to be set at 0.mm/rev and the depth of cut is to be 0.6mm based on the experimental results for the objective of minimum power consumption. The main effects plot based on S/N ratio for power consumption is shown in Figure 4.4. It was observed that the S/N ratio values increases with the decrease in cutting speed. The maximum S/N ratio is found when the cutting speed at the lowest level, i.e. 75 m/min. The S/N ratio decreases with the increase in the level of feed rate. The maximum value of S/N ratio is observed for the depth of cut at level.

11 65 Table 4.5 Main effects (Response) table for power consumption Process parameter Level Average Mean Average S/N Ratio A B C A B C * * * *.959 Average value * * Delta (Max-Min) Rank 3 3 *Optimum levels A B C Main Effects Plot (data means) for SN ratios 6 A B C Signal-to-noise: Smaller is better 3 Figure 4.4 Main effects plot for power consumption S/N ratio 4.8 ANALYSIS OF VARIANCE (ANOVA) The purpose of ANOVA is to find the significant factor statistically. It gives a clear picture of how far the process parameter affects the response and the level of significance of the factor considered (Fisher, 95). The ANOVA table for S/N ratios are calculated and listed in Table 4.6 and Table 4.7 for surface roughness and power consumption respectively. The

12 66 F- test is being carried out for 95% confidence level to study the significances of the process parameter. The high F value indicates that the factor is highly significant in affecting the response of the process. Table 4.6 ANOVA for surface roughness S/N ratio Source DoF SS MS F %P A B C Residual Error Total Table 4.7 ANOVA for power consumption S/N ratio Source DoF SS MS F %P A B C Residual Error Total In the present investigation, for the material Al-SiC p (0P) the feed rate (89.%) is a highly significant factor and plays a major role in affecting the surface roughness of the machined surface. The cutting speed (57.99%) is major influencing factor for the response of power consumption followed by feed rate (3.6%). The effect of depth of cut does not make any impact in the responses.

13 CONFIRMATION RUN The confirmation experiments were carried out by setting the process parameter at optimum levels (A 3 B C ), i.e., the cutting speed, feed rate, and depth of cut were set at 80 m/min, 0. mm/rev and 0.6 mm respectively for surface roughness objective and the average surface roughness value (Ra) was observed as.6 microns. Similarly the confirmation experiments were carried out by setting the process parameter at optimum levels (A B C ), i.e., the cutting speed, feed rate, and depth of cut were set at 75m/min, 0.mm/rev and 0.6mm respectively for power consumption objective and the average value was 0.39 kw. 4.0 CONCLUSIONS In this chapter, the process parameter of turning Al-SiC p (0P) MMC is optimized using Taguchi method. From this investigation, following conclusions are drawn. The L 9 Taguchi orthogonal designed experiments of turning on Al-SiC p (0P) MMC were successfully conducted. The process parameters were optimized to maximize the surface quality and minimize the power consumption for turning. The optimum levels of the cutting speed, feed rate, and depth of cut were found to be 80 m/min, 0. mm/rev and 0.6 mm, respectively (A 3 B C ), for surface roughness objective and 75 m/min, 0. mm/rev and 0.6 mm, respectively (A B C ), for power consumption objective. The feed rate plays a vital role to affect surface roughness, and contributes 89.% to the overall response. The depth of cut does not affect the responses significantly.

14 68 The cutting speed plays a vital role to affect power consumption, and contributes 57.99%, followed by feed rate with 3.6%. This study thus concluded that higher cutting speeds provide for higher surface finish, but will lead to higher power consumption. Thus a compromise must be arrived depending upon the accuracy requirements on the part to be machined and economic requirements. The optimum process parameter evolved with respect to minimization of surface roughness and power consumption leads to two different combinations such as A 3 B C and A B C respectively. This leads to confusion in selection of optimal setting for machining parameters which satisfy more than one quality characteristics. Hence multi-response optimization required to get a single combination of process parameter levels which minimize both surface roughness and power consumption and discussed in the following chapter.