Dynamic force measurement for milling using a constrained motion dynamometer and time domain signal analysis

The work in this dissertation describes a new milling force measurement method utilizing a constrained-motion dynamometer (CMD), an optical displacement transducer, and processing of a time-domain displacement signal. The research objective is to produce a testing system that accurately measures milling forces using (1) displacement, (2) derived velocity, (3) derived acceleration, and (4) the structural dynamics of the CMD, which is used to measure the displacement of a flexure supported platform due to milling forces. The new knowledge obtained is the ability to derive dynamic force from time-domain displacement and the mass, damping, and stiffness modal parameters for a constrained-motion dynamometer. The outcomes will be new milling force measurement capabilities that: 1) are relevant to industry; 2) provide the foundation for a low-cost, commercially viable dynamometer; and 3) supports Industry 4.0/5.0 efforts.

This approach expands prior low-cost methods based on frequency domain filtering analysis by using time domain displacement signals and modal parameters of the dynamic system [1]. In short, measured displacement and derived velocity and acceleration will be converted to individual force components using the modal mass, damping, and stiffness values and then these force components are summed to predict the milling force. The velocity and acceleration signals will be acquired by deriving the time-based displacement signal.

The new approach will be evaluated with multiple workpiece materials and cutting tools to validate its performance. Results will be compared to identical milling force tests conducted on a commercially available dynamometer system that cost two orders of magnitude more and requires detailed knowledge of digital filtering techniques to make the best use of the data. Additionally, a comparison to the prior frequency domain inverse filtering approach will be compared to determine validity of both low-cost dynamometer approaches.