PSI - Issue 10

N.M. Vaxevanidis et al. / Procedia Structural Integrity 10 (2018) 333–341 N.M. Vaxevanidis et al. / Structural Integrity Procedia 00 (2018) 000 – 000

338

6

For constant rotational speed 1600 rpm, Fc reduces when using the lowest level of depth of cut as it is expected. Higher values for feed tend to increase main cutting force Fc except from using a=1.0 mm where both the second and the third level of feed maintain the same value of Fc for that depth of cut. Fc reaches its highest value under a =1.5 mm and f =0.33 mm/rev and its lowest value under a =0.5 mm and f =0.1 mm/rev (Fig.4b). For the same constant rotational speed of n =1600 rpm Ra increases towards higher values of depth of cut as well as higher values of feed rate with quite similar trend (Fig.5b). Rt exhibits its lowest value for the second level of cutting depth a =1.0 mm under f =0.1 mm/rev (Fig.6b). However, changing from f =0.18 mm/rev to f =0.33 mm/rev an exponential increase is observed for Rt . By comparing the trends of Rt with regard to the constant rotational speeds of n =800 and n =1600 rpm, similar effects are shown for a =0.5 mm and a =1.0 mm whilst quite the opposite prevails for a =1.5 mm. Rt parameter reaches its highest value for a=0.5 mm and f=0.33 mm/rev.

(a)

(b) Fig. 6. Effect of feed rate on the maximum surface roughness, Rt : (a) n=800rpm; (b) n=1600 rpm.

3.2. Statistical analysis

The influence of cutting conditions upon the mean surface roughness, Ra , the maximum surface roughness, Rt and the main cutting force Fc , were justified by constructing the main effects plots. MINITAB ® 17 software was employed to perform the analysis and to obtain the necessary outputs for results interpretation. According to ANOVA and the main effects plots (Fig.7a), the depth of cut has considerable effect on main cutting force followed by feed rate and their interaction ( f*a ); see Fig.7b. Note that the combination of feed and cutting depth determines the undeformed chip section, thus; the amount of power consumption required for removing a specified material volume which is directly related to cutting force. From Fig.7c it is evident that, as far as average roughness ( Ra ) is concerned, feed rate followed by rotational speed dominate against depth of cut which does not seem to affect significantly Ra . The product n*f yields noticeable influence in terms of Ra (Fig.7d). Rt is primarily affected by rotational speed (Fig.7e) whereas the product of rotational speed and feed rate, n*f comes as the second influential attribute according to corresponding ANOVA analysis (Fig.7f). Feed rate comes as the second stand-alone influential parameter on Rt . To guarantee the generation of highly reliable regression models for predicting the objectives examined, full quadratic regression was employed to further examine the significance of higher order interactions. Note, that Tables summarizing the ANOVA results for Fc, Ra and Rt are not included due to restriction in space. To achieve a desired surface finish as well as good machinability in terms of cutting forces, capable predictive models are necessary for stable machining. Based on the experimental results and the number of performance re sponses, several regression approaches are available so as to select one and generate mathematical relations correlating the independent variables to responses. In this work, second-order mathematical relations were created to predict the variation among parameters and to examine their efficient applicability in turning of brass. All prediction models were generated by statistically processing the experimental results by considering a 95% confidence interval. Regression was 4. Prediction models