Investigation of LC-Resonance Driving in Disconnector Bus-Transfer Testing

The ability to transfer load currents between parallel busbars is one of the main requirements for most disconnector switches in gas insulated as well as air insulated substations today. As such, the capability limits and testing methods are internationally standardized in IEC 62271-102. The test circuit dictated in this standard is the Thévenin equivalent of the high current, power frequency approximation of two parallel busbars supplied by an AC current source. This setup inevitably couples different factors influencing the bus-transfer switching process by directly specifying the preswitching current through the disconnector switch and recovery voltage expected across it after successful switching. To facilitate a more detailed and direct investigation of bus-transfer switching processes, a different test method is suggested and consequently implemented as a proof of concept. The suggested method entails the use of an AC current source to drive a current through two parallel impedances representing different sections of a busbar. The disconnector switch under test is placed in series with one of the impedances. In order to analyze the feasibility of such a setup, a proof of concept current source along with two parallel impedances was built. The main design goal for the current source was to provide AC currents over a large range of magnitudes at a constant frequency. In addition, the current provided by the source must not be influenced by the bus-transfer switching process. To fulfill these requirements, an LC-resonance circuit with an adjustable resonance frequency and a substantial initial energy storage capacity was chosen. Its applicability was verified in numerous switching cases of different specifications. The impedances representing substation busbars in the test setup were designed with a main focus on adjustability and precision. The resistance and inductance components were modelled after typical GIS substation values for lengths between 10 m and 600 m. Due to the conductor being comprised of overhead line cable, virtually any length in this range can be represented with minimal effort. Experimental verification has demonstrated that this type of test setup ensures direct comparability between laboratory tests and substation processes due to the ability to directly measure both parallel currents and the recovery voltage occurring at the disconnector after successful switching. This feature can prove beneficial for non-standard testing cases such as failure analysis and substation planning which require precise modelling of substation characteristics as well as sophisticated measurements.