Online chatter detection for single-point diamond turning based on multidimensional cutting force fusion

Abstract

Single-point diamond turning (SPDT) is an advanced ultra-precision processing technology that is widely used in machining optical surfaces and microstructures. However, due to insufficient stiffness or damping of the cutting system and the high strength of the workpiece material, chatter inevitably occurs in the machining process, resulting in inferior workpieces and short tool lives. Thus, chatter detection at the early stage is vital to avoid its adverse effects, and a multidimensional cutting force fusion-based online chatter detection method for SPDT is proposed in this work. According to the characteristics of each component of the cutting force in the chatter SPDT process, the stable frequency component is filtered out, and then an artificial constant offset is added. On this basis, the cutting force-based multidimensional information fusion signal and its derivative are set as the detection variables. The chatter detection indicator is developed based on the sum of squares of the detection variable’s frequency spectrum amplitude, with the chatter warning threshold being adaptively determined by the upper limit of the 95% confidence interval of the same indicator obtained from the measurement noises under air cuts. Experiments conducted under various cutting conditions prove that the proposed method can accurately and timely detect weak chatter, thereby preventing further severe chatter and its detrimental effects on the SPDT system.