Distributed parameter circuit model for shaft voltage prediction in induction motors fed by PWM based AC drives

Abstract

The development of high frequency, pulse width modulation (PWM) based adjustable speed drives (ASDs) has increased the efficiency, performance and controllability in induction motor applications. However, the high switching frequencies and the faster switching times of IGBTs (the device of choice in these ASDs) introduce disadvantages like overvoltages at the motor terminals when long cables are used between the drive and the motor. Another industry-wide concern is the generation of rotor shaft voltage and the resulting bearing current. The grease film in a bearing acts as a capacitor that charges due to the transitions in the common mode voltage imposed at the motor terminals by the drive. The breakdown of the film causes a spike of current to flow that can damage the bearing and reduce life. A significant amount of effort has been directed at understanding the shaft voltage phenomenon and the associated bearing current. This study attempts to develop distributed circuit models to predict the level of the shaft voltage. The circuit models can then be used to predict the shaft voltage levels at different installations, using simulation software like PSpice. Circuit models for two specific motors are developed. The predicted shaft voltage is very close to the actual voltage levels seen on the shaft when the motors are operated by ASDs.