Sensor fusion based tool-workpiece contact detection in micro-milling

Abstract

Information about distance between tool and workpiece is crucial for proper machining process. In micro-milling, due to very small tool dimensions, observation of the tool position with regard to the workpiece is possible only with a microscope. Exploiting of different signals obtained during micromilling process in establishing the tool-workpiece contact moment is an interesting and efficient way of improving accuracy of this activity. Data from sensors contain material that could be an large source of diagnostic information, which might be helpful from different points of view. This paper presents a method for detecting a contact between a tool and a workpiece based on a fusion of the data representing the cutting forces and the accelerations. A digital signal processing algorithm introduced in this research, basing on the Fast Fourier Transform (FFT) and the Root Mean Square (RMS) calculations give positive results in field of the tool-workpiece contact detection and stands an alternative for commonly used techniques exploiting microscopes with large magnification. The method is designed to work in real-time in order to be used in practical industrial applications and satisfy demands regarding the automatic tool-workpiece contact detection.