Fault Diagnosis of Brushless DC Motors Using Built-In Hall Sensors

In this paper, a method for fault detection of brushless direct current (BLDC) motors is proposed simply based on the hall sensors that exist in most of the BLDC motors. Two important faults in these motors are bearing faults and demagnetization fault. In the first step, the effects of bearing faults on the BLDC motor speed are studied. Then, using the discrete wavelet transform (DWT) and the continuous wavelet transform (CWT), a residual signal is extracted from the speed signal and analyzed in the time-frequency domain. It will be shown that the residual signal contains the main features of bearing faults such as bearing fault frequencies. In the second step, the effects of the demagnetization fault on the hall sensors outputs are studied. It is shown that the demagnetization fault leads to impulsive oscillations in the duty cycle signal of the hall sensors outputs. Using the kurtosis index of the duty cycle signal, the rotor condition is classified into either a healthy condition or a demagnetized condition. In the third step, using an experimental setup, the efficiency of the proposed methods is demonstrated. The proposed algorithms are simple and only use hall sensors for fault detection and hence lead to a simple and cheaper fault diagnosis system.