Reliable Airgap Search Coil Based Detection of Induction Motor Rotor Faults Under False Negative Motor Current Signature Analysis Indications

Abstract

Rotor cage damage in induction motors must be detected and repaired to prevent costly in-service failures that can occur due to broken cage fragments or arcing. Motor current signature analysis (MCSA) has become the most widespread test in the field for detecting these faults as it can provide remote, online monitoring using current sensors available in the motor control center. Although MCSA has been successful in detecting broken rotor bars, many cases of false indications have been reported in the field. False negative indications, where MCSA fails to detect the fault due to load variation, low slip operation, error in speed estimate, and nonadjacent broken bars, is considered serious as it can lead to unexpected motor failure. This has triggered research on alternative monitoring techniques for reliable detection of rotor faults with flux monitoring being one of them, as it can be implemented with low cost search coils. In this article, a new test method based on time-frequency analysis of the airgap flux is proposed and verified. It is shown that the second rotor rotational frequency (2x) upper sideband can provide reliable detection of rotor faults. Experimental test results are given under load variation, low slip, and nonadjacent broken bar conditions to support the claims on the reliability of the proposed method for cases where MCSA produces false negative indications.