Gear fault diagnosis by motor current analysis: application to industrial cases

Abstract

The use of Motor Current Signature Analysis (MCSA) for motor fault detection such as broken rotor bar is now well established, however the detection of mechanical faults related to the driven system is a more challenging task. Recently there has been a growing interest for the detection of gear faults by MCSA. Even if well adopted, the use of traditional vibration measurements can present several drawbacks: technical difficulties of access to the machine in some cases, influence of the transmission path and sensitivity to the sensor position. The motor supply current, being an image of the load torque, seems to be a relevant tool for the detection of local tooth faults, which will induce sharp variations of the mesh stiffness and consequently of the instantaneous torque. The use of one-phase stator current (1I) may be sufficient for some applications in case of a low rotation speed of the faulty gear. However in other cases the three-phase stator currents (3I) may be necessary in order to compute the Park’s vector components, whose advantage is to bypass the demodulation step and the Bedrosian’s conditions. The 1I current amplitude modulation function or the Park’s vector modulus may also be averaged on the gear rotation period in order to enhance the local tooth faults. Advantages and drawbacks of these techniques are presented and discussed on a few industrial cases.