Fault detection techniques for induction motors

Motor Current Signature Analysis (MCSA) is being the most widely used method to identify faults in Induction Motors. MCSA focuses its effort in the spectral analysis of the stator current, by analysis of the resulting harmonics when the machine is working under failure state. Motors faults such as broken rotor bars, bearing damage and eccentricity of the rotor axis can be detected. However, the method presents some problems at low speed and low torque, mainly due to the proximity between the frequencies to detect and the small amplitude of the resulting harmonics respectively. In both cases, the problem of frequency accuracy is very tricky since the side band harmonic is closed to the fundamental. Since the accuracy is reverse proportional to the data acquisition time, the spectrum read could be meaningless. Therefore, it is difficult to evaluate the correct magnitude of the side band harmonics even when windowing. This paper proposes to inject into the machine under test an additional voltage at frequency different from the fundamental, and then to study the resulting harmonics around the new frequencies appearing due to composition between injected and main frequencies.