Modeling cutter tilt and cutter-spindle stiffness for machine condition monitoring in face milling using high-definition surface metrology

In face milling, the spindle is intentionally tilted to avoid backcutting. The cutter tilt during machining is a combined effect of the intentional initial tilt and cutter-spindle deflection, which varies with cutting load during machining. An accurate estimation of the cutter tilt is critical to surface quality control and machine health monitoring. However, due to the small magnitude, the spindle tilt and deflection are difficult to measure in real time. Conventionally, the cutter tilt can be obtained through in-line sensors mounted on the machine tool but such in-line measurement is greatly influenced by the dynamic machining conditions such as vibration. This paper proposes a method to monitor the spindle setup tilt and deflection using surface data measured by high-definition metrology (HDM). Two parameters are proposed to characterize the cutter tilt, i.e., cutter tilt at idle state (initial cutter tilt) for spindle setup and cutter-spindle stiffness for the cutter-spindle deflection. Cutting force modeling is conducted to estimate these two parameters in conjunction with statistical procedures that fit the model to HDM surface data. The estimated cutter-spindle stiffness variation is also correlated to machine conditions such as a loose or worn bearing for process diagnosis. The method is demonstrated via experimental data from a machining process for automotive engine heads.