Electronic transformer protection

[kelvin]

Interested to know if anyone has experienced problems with modern electronic protection relays designed for power transformers. I have recently experienced failure to operate with a close in secondary earth fault on a recently commissioned 66/22kV power transformer. The device appears to have gone to sleep as the event log is competely clear of any events involving the fault. Subsequent primary and secondary tests indicate the relay is now operating perfectly. The manufacturer states no knowledge of problems to date. Leaves me nervous about the possibility of another failure. Reluctant to give manufacturer of relay.
Would appreciate comments

 

[busbar]

Have had multiple problems with two styles of harmonic-restrained transformer-differential relays, but they were false trips.

Hindsight suggestions are: 1.) Had the device been polled for seeming logical “meter data”? and 2.) Were “watchdog” alarm contacts monitored?

You may want to propose to the relay producer sending a new relay and use the packing materials to return the suspect device. Although not an ideal situation, they are likely most qualified to evaluate the device.

Check the IEEE PES-PSRC for Manufacturer’s Service Letter Database Project at

http://www.pes-psrc.org/i/ServiceData.html

or other materials at

http://www.pes-psrc.org/Apublications.html

 

[kelvin]

Unfortunately the relay was not being polled or watchdog contacts monitored. During commissioning integral earths were mistakenly left on at the 22kV incomer. The transformer was livened from the point of supply some 25kM away from the substation. The 66kV transformer protection at the substation failed to operate leaving the back up protection at point of supply to operate instead.

 

[notknown]

I would assume you verified all the settings in the relay. Another issue that may affect why you had no record is that a uP based relay has to be programmed in a manner to tell it what consititutes a fault. A contact can open and close all day long and it may be perfectly normal. Was the fault recording section programmed correctly?

 

[jghrist]

Was the fault in the differential zone or were the overcurrent elements expected to trip? Was there a possibility of CT saturation? Does the relay include oscillography? What logic initiates an event report or oscillography?

 

[kelvin]

Re Watchdog contact. I have since found that the watchdog contact is wired to a spare input on another electronic feeder relay on site. Have now found from the other relays events that the watchdog contact did operate during the fault period. Setings for the relay have been verified as being valid. The fault was not in the differential zone and insufficient current for CT saturation. The relay has oscillography and is initiated with fault current. Unfortunately with this fault there were absolutely no events for this fault period