Fault Detection in Small Fan Motors Using MCSA

This paper introduces a new method to detect lubrication faults in small fan brushless DC electric motors by combining stator phase current features. Bearing lubrication faults are generalized-roughness faults, making it difficult to obtain a characteristic frequency in current, especially when compared to single-point bearing faults. Consequently, the paper proposes using a combination of root mean square and crest factor of the phase current as an indicator of lubrication fault. The paper investigates the relationship between the proposed indicator and fault severity in the time domain under different aging hours, lubrication levels, and operating speeds. Furthermore, the authors conduct experiments on bearings with two different sizes to gain insights into the mechanism behind lubrication faults. The results indicate that the dominant factor causing lubrication faults may vary among different bearings. The experimental findings provide evidence supporting the effectiveness of the proposed indicator for detecting bearing lubrication faults in the time domain.