Machine tool condition monitoring using workpiece surface texture analysis

Tool wear affects the surface roughness dramatically. There is a very close correspondence between the geometrical features imposed on the tool by wea r and microfracture and the geometry imparted by the tool on to the workpie ce surface. Since a machined surface is the negative replica of the shape o f the cutting tool, and reflects the volumetric changes in cutting-edge sha pe, it is more suitable to analyze the machined surface than look at a cert ain portion of the cutting tool. This paper discusses our work that analyze s images of workpiece surfaces that have been subjected to machining operat ions and investigates the correlation between tool wear and quantities char acterizing machined surfaces. Our results clearly indicate that tool condit ion monitoring (the distinction between a sharp, semi-dull, or a dull tool) can be successfully accomplished by analyzing surface image data.