Chatter-induced variations in surface roughness parameters and their interactions during internal turning: An experimental statistical approach

Internal turning operations are prone to dynamic instabilities due to the low stiffness of boring systems, often causing substantial deviations between predicted and experimentally obtained surface roughness. This study presents a comprehensive experimental–statistical investigation of surface roughness formation in internal turning of steel bushings, focusing on the role of chatter and its influence beyond conventional Ra-based assessments. A full-factorial experimental design was implemented under dry cutting conditions using three cutting speeds, three feed rates, three depths of cut, and two insert geometries. Surface roughness parameters (Ra, Rq, Rz, and Rt) were measured at multiple circumferential locations in accordance with ISO standards, while vibration and acoustic signals were recorded simultaneously for chatter detection.

Chatter-free and chatter regimes were distinguished using a binary classification approach based on a vibration threshold of 2 m/s² and an acoustic criterion above the ambient noise level (∼70 dB), supported by surface morphologies. Statistical analyses, including Shapiro–Wilk tests, t-tests, factorial ANOVA, and chi-square tests, were used to quantify the effects of cutting parameters and dynamic stability on surface quality. Under chatter-free conditions, cutting speed was the dominant factor governing Ra, with a significant interaction with feed rate. Chatter initiation was primarily driven by cutting speed, while feed rate showed a secondary but statistically significant contribution.

A key finding is that chatter systematically alters surface profile characteristics, as evidenced by significant changes in normalized roughness ratios (Rz/Ra, Rt/Ra, and Rq/Ra), despite increased average roughness. The results indicate that periodic waviness associated with chatter may be attenuated by profile filtering, leading to reductions in roughness ratios. These findings highlight the need to complement Ra with roughness ratios for a more reliable assessment of surface integrity in dynamically unstable internal turning operations.