Doctoral Dissertations

Date of Award

12-2025

Degree Type

Dissertation

Degree Name

Doctor of Philosophy

Major

Mechanical Engineering

Major Professor

Dr Siamak Farhad

Committee Members

Siamak Farhad, Sekhar Rakurty, Bradley Jared, Tony Shi, Michael Gomez

Abstract

This study evaluates and validates a novel chatter detection method in milling operations by periodic sampling of time domain milling signals. The periodic sampling frequency is scanned over a preselected range, and the sampled values of the time domain milling signals are used to calculate a stability metric for each sampling frequency. For stable cutting conditions, the sampled point repeats for each tooth passage, and a local minimum in the stability metric occurs at the tooth passing frequency. For unstable cutting conditions, the sampled point does not repeat and a local minimum in the stability metric occurs at or near the chatter frequency instead of at the tooth passing frequency. It is also observed that at stable cutting conditions, the stability metric values are less than those from unstable cutting conditions, for the same experimental setup.

The location of the local minima in the stability metric with respect to the tooth passing frequency and the chatter frequency, is used to confirm whether the cut is stable or unstable, which validates the novel approach. The stability of the cut is further confirmed by plotting and comparing the locations and values of the stability metric with the frequency content of the milling signals using FFT (fast Fourier transform), by visually inspecting the cut surface of the workpiece for regular or irregular tooth or chatter marks, and by measuring the roughness of the cut surface of the workpiece. Studies are completed using in situ milling signals including displacement, acceleration, and sound.

To evaluate spindle speed selection, unstable cuts are performed, and a new spindle speed is calculated based on the chatter frequency and the number of teeth of the used cutting tool. The new spindle speed and a new feed rate are then applied to achieve a stable cut while maintaining all other cutting parameters and conditions.

This study also introduces a fundamental feasibility study for closed loop chatter detection and spindle speed selection using an algorithm that could be integrated into open-source machine controls to enable automatic identification of both the tooth passing and chatter frequencies using in-built once-per-revolution machine data.

Download
Files over 3MB may be slow to open. For best results, right-click and select "save as..."

Share

COinS