In-process monitoring and empirical modeling of the tool wear in turning of aluminum alloys using thermoelectric signals

Abstract

Tool wear is an important criterion for economic machining. It does not only determine the tool life, but also affects the surface layer properties of metallic components and thus their performance. In addition to the machining parameters, there are numerous other influencing factors like batch variations, workpiece inhomogeneities, process vibrations, or the formation of built-up edges that can impact the wear progression. Consequently, it is difficult to predict the tool wear analytically. Therefore, the determination of the tool condition during machining is of great importance. The tool-workpiece thermocouple enables the in-process measurement of a thermoelectric voltage and current during machining generated by the Seebeck effect at the interface between two electrically conductive materials. Thus, the thermoelectric signals are sensitive for the changing contact interface between the tool and the specimen. To allow for an in-process characterization of the tool wear, finding an empirical relationship between the flank wear land width of cemented carbide indexable inserts and the thermoelectric signals in turning is aspired. To vary the flank wear land width in a reproducible manner, tools with a flank face chamfer resulting in a clearance angle of 0° and five different flank face land widths (80 µm – 280 µm) are used. These geometrical properties are measured with a 3D laser scanning microscope. Afterwards, cylindrical specimens of the aluminum alloy EN AW-2017 are machined by turning applying the modified tools and cutting speeds in the range of 300 m/min to 550 m/min. The depth of cut (0.4 mm) and the feed (0.04 mm) are kept constant. Besides the tool-workpiece thermocouple, the components of the resultant force are measured by a dynamometer. The geometrical properties of the machined surfaces are characterized using a stylus measurement instrument.The experimental investigations show an increase of the temperature and the components of the resultant forces with increasing flank wear land width, which impacts the geometrical properties of the surface. An empirical regression model derived from the in-process measurement results depicts the relationship between the thermoelectric signals, the force components, and the changing area of the tool which is in contact with the specimen. Consequently, the combined in-process measurement of the process forces and thermoelectric signals allows for an in-process determination of the tool wear progression and therefore enables the prevention of a strong impairment of the surface-layer properties by a timely tool change or a controlled adjustment of the machining parameters.