Bearing inner race fault detection and size estimation using the variable reluctance sensor

Abstract

The variable reluctance sensor (VRS) has been widely used in the speed measurement system in the fields of automobile, helicopter, and ship, because of its high robustness. In recent years, most studies have verified the effectiveness of the application in rotor torsional vibration measurement and blade monitoring by the VRS. However, there is less research devoted to the VRS in bearing fault diagnosis, and its influencing mechanism between bearing faults and signal characteristics is not clear. In this paper, the raw signal of the VRS is used for bearing diagnosis. The mechanism of bearing partial fault signal response is analyzed by establishing the bearing inner ring fault model and the sensor signal response model. The characteristics of the VRS signal are given and verified by experiments. The results show that the gearwheel tooth-pass frequency is modulated by the bearing fault characteristic frequency, and there is an obvious double peak phenomenon in the high-frequency part. Moreover, the fault data with widths of 1 mm and 2 mm are analyzed by using the interval time of double peak. It is found that the fault sizes estimated are close to the actual defect size, and have small standard deviations. The original signal of the VRS has great potential in bearing diagnosis.