Gearbox fault detection using spectrum analysis of the drive motor current signal

This paper investigates the application of spectrum analysis of the motor current signal to the detection of mechanical faults in a two-stage helical gearbox driven by an 11kW induction motor. The benefits of using the current signal of the drive motor to monitor downstream mechanical components include a non-intrusive approach, potentially applicable remotely from the machine, likely less costly to apply than more conventional approaches like vibration monitoring, and with scope for improved health and safety. Comparison of the spectral content of the motor current signal against the baseline is used for the purposes of detecting and assessing the severity of pinion gear faults in a multi-stage gearbox, and a method is established that quantifies spectral components and provides a basis for assessment of gearbox condition. The spectrum is dominated by the 50Hz mains frequency component in the motor current spectrum and families of sidebands are revealed which correlate with the shaft rotational frequencies (RF) around the gear meshing frequency (GMF). The number and the amplitude of the sidebands rise when a local tooth fault is introduced into the gear and the clear differences can be observed between the faulty and the healthy spectra. The work in this paper then confirms the abilities of motor current signal for fault detection of downstream machines.