OBRABOTKAMETALLOV technology Vol. 26 No. 3 2024 Ta b l e 2 Coded levels and corresponding actual cutting parameters Parameters Levels for alpha value equal to -1.6817 -1 0 +1 +1.6817 Cutting speed (V) (rpm) 100 200 300 400 500 Feed (f) (mm/rev) 0.1 0.15 0.2 0.25 0.3 Number of passes (N) (mm) 0.5 1 1.5 2 2.5 Depth of penetration (mm) 0.5 Average surface roughness values were measured using a Taylor Hobson Talysurf stand-alone surface roughness measuring device on a Surtronic Duo. The surface roughness was measured at three equally spaced points around the circumference of the workpiece to obtain the statistically significant value. Roundness deviation was measured using Bridge type CMM (Make: Zeiss, Model: Contura, Range: 1200×800×800 mm). The geometrical deviations were obtained by measuring the roundness in twelve sections of the calibrated area using a millesimal dial gauge with a measuring range of 12.5 mm, a scale division of 0.001 mm and a maximum permissible error (MPE) of 4 μm. Microhardness was measured by the Vickers microhardness tester using a diamond indenter with an angle of 136o, a load of 100 g, and a dwell time of 20 sec. Results and Discussion In this section, the effect of the roller burnishing process parameters on the process responses is discussed based on the developed regression equations. Curves showing the various responses are plotted by varying one of the input parameters and keeping the other parameters constant to understand the physics of the process and the influence of the cutting parameters on different responses. The contribution of cutting parameters on different responses are also obtained. Finally, a desirability function approach is addressed for optimization of process responses in roller burnishing of Al6061-T6 alloy. Experiments were carried out varying the cutting speed, feed, and the number of passes (Input parameters). The experimental matrix and results of surface roughness, microhardness, and maximum roundness deviation (roundness error) in roller burnishing Al6061-T6 alloy are shown in Table 3. 4 3 = 0, 9734 + 3, 38068 10 2, 7693 + 0, 0563 3, 25 10 - - ⋅ ⋅ - - - Ra V f N Vf 4 6 2 2 2 4,125 10 0, 425 + 2, 6136 10 + 12, 9545 + 0, 02113 ; - - ⋅ ⋅ - - VN fN V f N (1) 119, 534 0, 2611 233, 0681 12, 0056 0, 425 0, 0187 42, 5 - - - - - = + + HV V f N Vf VN fN 4 2 2 2 2, 3636 10 + 604, 5454 + 1, 6363 ; - ⋅ V f N (2) = 9, 525 + 0, 01281 + 157,125 + 3, 3937 + 0, 0925 0, 0203 + 18, 25 + - - Re V f N Vf VN fN 5 2 2 2 + 2,125 10 610 0, 375 . - - - ⋅ V f N (3) The adequacy of the developed equations was checked by Analysis of Variance (ANOVA). R-Squared is a coefficient of multiple determinations, which measures variation proportion in the data points. It is always desirable for the correlation coefficient (R-Squared) to be in the range of -1 to +1. The equation makes sense if the value of R is very close to +1. The Adjusted R-Squared is a measure of the degree of deviation from the mean explained by the model. Predicted R-squared is a measure of how well the model predicts the response value. Adjusted and predicted R-Squared values should differ from each other by approximately 0.20 to ensure a “reasonable agreement”. If they are not, there may be a problem with either the data or
RkJQdWJsaXNoZXIy MTk0ODM1