Acceleration feedback control for enhancing dynamic stiffness of fast tool servo system considering the sensor imperfections

Lorentz type fast tool servo devices have found wide applications in freeform machining but they face problems of insufficient stiffness with large depth of cut. Acceleration feedback control is an alternative way to enhance the dynamic stiffness without the need for a large inertia, which is strictly limited in fast tool servo devices. However, the current knowledge gap in the understanding of the influences of limited sensor bandwidth and sensor noises on positioning performance has impeded the application of acceleration feedback control approach in fast tool servo devices. This paper established an analytical model to reveal, for the first time, how much positioning errors are caused by the added sensor noises and how the acceleration feedback technique changes the closed loop stiffness. The measured positioning error spectrum agrees with the modelled one with different acceleration gains. The stiffness model is verified through frequency response tests. It is found that the dynamic stiffness is significantly improved by 5.6 folds within the acceleration sensor bandwidth, while the stiffness deteriorates at frequencies beyond the bandwidth due to the low-pass characteristics in the acceleration loop. The stiffness analysis results are further verified in the intermittent facing cut experiments. The measured surface form errors can be mapped to the low frequency and high frequency vibrations caused by the cutting forces. The analysis model provides a theoretical basis for adopting acceleration feedback technique, paving the way for its practical implementations in ultra-precision applications.