Laboratory investigation into the radio interference produced by artificially polluted 22 KV distribution line insulators

Cunliffe, C. J

Laboratory investigation into the radio interference produced by artificially polluted 22 KV distribution line insulators

Thesis / Dissertation

1990

Publisher

University of Cape Town

Department

Department of Electrical Engineering

Faculty

Faculty of Engineering and the Built Environment

Abstract

Although it is known that polluted power line insulators can, under certain conditions, cause interference with radio communication networks, there is little useful practical data as to the extent of the problem. This is mainly due to the difficulties associated with quantifying and measuring this interference. There are various test procedures to measure radio noise characteristics of hv insulators, set out in recommendations published by the International Electrotechnical Commission (IEC). These however, only apply to clean and dry conditions. The IEC has also recommended methods for the laboratory simulation of natural pollution conditions in insulator flashover tests. This project involved combining the two techniques into a laboratory procedure for measuring RN produced by artificially polluted insulators. The investigation was concentrated on 22 kV line insulation and RN in AM frequency bands. Initial tests were done to develop a repeatable pollution simulation method based on the IEC 507 clean fog artificial pollution test. This method was then combined with the IEC 437 RN measurement test circuit to form the basis of two different types of test procedure. The first test procedure was a prolonged or constant voltage method. It consisted of monitoring the insulator RN with time while it was subjected to a pre-determined pollution condition and a constant voltage for the test duration. This is representative of the field situation but makes it a very time consuming process to obtain interference values over a range of test voltages. The second, accelerated type, procedure was to record RN as the voltage was varied within any one test run. This is not representative of a practical case where large voltage variations are rare, but it did have time saving implications. The accelerated procedure generally gave lower RIV values than those indicated by the prolonged procedure at the same voltage and pollution severity. Nonetheless, it was concluded that, by paying particular attention to wetting rates and voltage step durations, the accelerated procedure could also be further developed into a standard polluted insulator RN test. Two different insulator types - a porcelain line post insulator and a glass disc insulator - were used in 153 clean fog test runs. The results showed that clean, or polluted but dry insulators, operating at practically used specific creepages, produce insignificant levels of interference. Only at impractically low specific creepage were unacceptably high ranges of RN emitted. When the insulators were polluted and dampened with the clean fog, the RN increased rapidly with voltages above 1 kV. For the line post insulator the curve reached a knee point at about 8 kV, beyond which very little or no RIV increase with voltage was observed up to 18 kV. For the glass disc units, although the slope of the curve decreased at the higher voltages the same knee point was not so evident. For both insulators, except at very low voltages, little variation in RN level with pollution severity at fixed voltage was observed. These results are in general agreement with the conclusions of previous work. It would be useful to further refine the test methods, both to develop a RN measurement standard and to obtain more data pertaining to specific insulator types. A standard procedure would also permit investigating the correlation of both polluted insulator leakage current and the practical nuisance value of the RN, with laboratory measured RIV values.