Surface Roughness Estimation from Internal Spindle Torque During Side Milling of Invar 36

Abstract

Surface roughness is a key quality characteristic in machining Invar 36 components used in dimensionally stable applications. This study evaluates the effects of cutting parameters on surface roughness and internal spindle torque during peripheral side milling of annealed Invar 36 and examines the feasibility of estimating roughness from CNC internal data. A Taguchi L9 orthogonal array was used to vary cutting speed, feed per tooth, and radial depth of cut. Surface roughness was measured by three-dimensional optical profilometry, while spindle torque was recorded at an 8 ms sampling interval. The first and final tool paths were analysed separately using Box–Cox-adjusted mean values. Feed per tooth and radial depth of cut were the dominant factors affecting roughness, whereas radial depth of cut had the strongest influence on torque. Within the investigated range of 20–40 m/min, cutting speed produced the smallest S/N response and was not statistically significant in the reduced models. Strong bivariate correlations were obtained (r = 0.938 and 0.932). The fitted linear relationships explained 88.1% and 86.9% of the in-sample variability. The results indicate potential for process-specific surface-roughness screening without a dedicated external torque sensor; however, the fitted equations require independent validation before predictive use.

The authors' abstract, as published at the source. Journal of Manufacturing and Materials Processing, 2026 · DOI ↗