Motors are critical components for electric utilities and process industries. A motor failure can result in the shutdown of a generating unit or production line, or require that redundant plant be utilized to circumvent the problem. One major cause of failures is breakdown of the tum insulation leading to puncture of the groundwall. Early detection of intertum shorts during motor operation would eliminate consequential damage to adjacent coils and the stator core reducing repair costs and motor outage time. In addition to the benefits gained from early detection of tum insulation breakdown, signijicant advantages would accrue by locating the faulted coil within the stator winding. Fault location would not only increase the speed of the repair, but would also permit more optimal scheduling of the repair outage. This work was successful in practically implementing a theory to predict changes in the axial kakage flux resulting from stator winding intertum shorts and in developing an algorithm to locate the position of the faulted coil. An experimental setup consisting of a 200 hp motor loaded by a generator was used to validate this theory. Suitable transducers were developed and installed on this motor. Measurement using this experimental configuration clearly validated the theoretical model. On the basis of this experimental work, an instrument to continuously monitor for shorted turns is under development.

No abstract

The use of on-line partial discharge (PD) measurements as an indicator of stator winding insulation condition, is well known. Trending results from on-line PD measurements over time, often allows maintenance personnel to become aware of developing stator insulation problems in motors and generators, and take corrective measures before failure occurs. However, partial discharge measurements on operating rotating machines can be influenced by conditions such as humidity, temperature, terminal voltage, and load. Thus, to provide reliable data which can be trended, it is important to take readings under similar machine operating conditions. In remote plants, or on pump storage units which are frequently cycled, it can be both time consuming and costly to ensure that personnel are available to take PD readings under ideal conditions occur. This paper describes the development of a automatic partial discharge monitoring svstem which is permanently connected to relevant machine sensors and can be configured to collect PD data when suitable conditions are met. The operating principles of the system are described, and some results are presented from data collected to date on 30 hydrogenerators.

No abstract