Dynamic PVDF Sensor Based Monitoring of Single Point Cutting Processes

Abstract

A low-cost, high fidelity measurement system consisting of a thin film Polyvinylidene Fluoride (PVDF) piezoelectric strain rosette and data logging electronics has been designed, fabricated, and evaluated for monitoring the dynamic cutting forces and torque in single-point cutting processes, specifically turning and boring. Physics-based models are used to relate the measured voltage to the process forces and torques. By means of key assumptions about particular strain components, simplified PVDF strain sensor rosettes are developed to isolate the particular strains of interest. Wired and wireless communication methods to transmit the dynamic strains measured by the sensors to a data logging base station are demonstrated. The proposed methods are experimentally validated through comparison with quartz-based piezoelectric cutting force and torque dynamometers. In addition, the performance of several chatter detection algorithms applied to turning force and boring torque data is evaluated with a focus on embedded electronic automation. The dynamic cutting force data is acquired from turning experiments by varying the initial workpiece geometry, while the dynamic torque data is acquired from boring experiments performed on industrial rotor compressor discs. For chatter detection in turning, spectral analysis is demonstrated to be the most robust algorithm and is shown to be capable of detecting dynamic instability before physical damage to the part occurs. For chatter detection in boring, autocorrelation modeling is demonstrated to be the most computationally efficient of the techniques evaluated.