Cutting force monitoring in the endmilling operation for chatter detection

Abstract

The chatter phenomenon shows the dynamic characteristics of tool vibration in the endmilling process and will change according to the irregular dynamic characteristics of that tool vibration. This chatter produces an adverse effect on tool life, machining integrity, surface quality of the workpiece, and other geometric accuracy. Chatter behaviour in endmilling is a complex, non-linear phenomenon, which is very difficult to detect and diagnose. It is therefore necessary to suggest a new method for analysing chatter mechanics. This paper presents a new method for the detection of chatter in the endmilling operation based on the wavelet transform. This wavelet transform method provides various ways to determine chatter characteristics. The fundamental coefficient property of the wavelet transform is reviewed. The reliability of the wavelet transform method is verified by comparing the spectra using the fast Fourier transform (FFT). The behaviour of the detail coefficients obtained by wavelet transform reveals the possibility to detect and analyse chatter and other malfunction states using tool dynamometer cutting force. Because wavelets are closely related to filter, the method presented in this paper can be applied to other real-time cutting force monitoring and analysis in a range of endmilling processes.

Advanced monitoring of machining operations

“…Kwak and Song [90] develop a method based on AE signals to recognise chatter vibration in grinding. Yoon and Chin [165] apply wavelet transform of cutting force signals for real-time detection of chatter in endmilling operations. Griffin and Chen [166] propose a multiple classification of AE signals to obtain signatures for both chatter and burn phenomena in grinding.…”

Characterization of machining chattering dynamics: An R/S scaling analysis approach

“…The process was determined stable if the 1/rev sampled position approached a steady constant value, otherwise the process was unstable. Chatter frequencies were also detected using wavelets [31,32]. In turn, a more sensitive detection was reached with wavelets because transient and nonlinear signals can be identified; even though interpretation of results depends on several detail coefficients.…”

A new method for the prediction of chatter stability lobes based on dynamic cutting force simulation model and support vector machine

“…Signal acquisition is usually achieved by the sensor. There are many kinds of sensors that can be used to acquire vibration signal in the machining operation, such as accelerometer [27], dynamometer [28], acoustic emission [29], displacement [30], and electrical power sensors [31]. In this study, the cutting force signals in the X and Y directions are selected to identify chatter, and are measured by a Kistler dynamometer 9257B mounted between the workpiece and the workbench.…”