Ductor Test Problem/Question 2

[davidbeach]

This is a problem that has been related to me and I'd like any input people with experience with such tests could provide. On a 161kV ring bus a breaker was out of service and being ductor tested. During testing of the first two phases no problems were encountered. On the third phase, a transformer differential relay picked up and tripped. Apparently CTs on each of two breakers on the ring are paralleled into the primary side of the differential relay, and the transformer was being fed through the other breaker. Has anybody seen this happen before during this kind of testing? Any theories as to why it might happen?

 

[electricpete]

I would look carefully at the location of the CT and see if there is possibility that you sent test current through the primary of the CT. If so, that of course can lead to a differential trip.

A sneaky path for current flow can involve loops to ground. For instance personnel grounds on both sides of CT can create a loop path which can allow current flow through the primary of CT.

Bushing CT's or freestanding CT's?

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[rcw retired EE]

David - The Ductor circulates a DC current between its probes through the component under test and measures the voltage drop (millivolts) to get resistance,(usually in milliohms). Ductors use 10 Amps or 100 Amps as test current.

The easy way to Ductor a circuit breaker is to go from bushing to bushing, which puts current through the complete assembly including any bushing CT's. Flipping the Ductor switch to get a reading turns on the DC current. Most instruments have a polarity reversal switch that allows nulling out external influences by reversing the current flow and averaging the two readings. That action might mimic an AC current to the relay.

Any on and off DC current through the CT will show up as a pulse on the output to the relay.

Must have been some sensitive CT and relay setting.

 

[davidbeach]

Thanks electripete & rcwilson. My guess is that the ductor test would have been bushing to bushing, through the primary of CTs in the transformer differential protection. The reversal of the DC could be the critical item. Like many tests, they probably did the third phase faster than the first two phases and the speed of reversal was high enough that the diff relay responded. It was a solid state diff relay, so shorting out the CTs would have disabled transformer protection. It would have been necessary to short the CTs and then disconnect them, taking them out of the parallel loop.

 

[RalphChristie]

Ductor test: For a moment I thought it is a new kind of test I'm not familiar with.

A dc milli-ohm resistance measuring test, normally used to measure circuit-breaker and tap-changer contact resistance. The test is carried out by injecting a high current and measuring the voltage across the device. (Ductor is a proprietary name. Other makes of test set are also used.)

David, I think rcwilson's comment and your summary is spot on. The only way how you can "create" a secondary output from a dc-source on the primary side is either by:
Pulsing/flicking the source.
Polarity reversal of the source.

Star for rcwilson

Regards
Ralph

 

[davidbeach]

One of my coworkers has pointed out that the ductor test probably saturates the CT that is seeing the DC current, so that current from the other CT in parallel has a path through the saturated CT rather than having all of its current go through the relay. Any thoughts on this possibility?

Any thoughts on the impact of the residual magnetization of the CTs following the tests? Is it possible that a CT on on of the first two phases tested was left very near saturation and was driven into saturation by secondary current coincidentally with the third phase testing? Any thoughts on what to do about the residual magnetism at the end of the test other than repeatedly injecting DC with alternating polarities and decreasing magnitudes?

We're trying to find out more about when the trip happened in the testing process (we're the manufacturer of the diff relay) to better understand what happened. The test set being used was from Programma, is anybody familiar with that piece of equipment?

 

[subtech]

Davidbeach
Some micro-ohmmeters don't "ramp up" their current output at the beginning of a test. If you are testing at 100 amps, its an instant on that sometimes contains a healthy "spike" at the onset of current flow. This fast changing spike has probably caused more trips than most techs are willing to admit. Relatively recently, Vangard Instruments (California, USA) re-designed their line of micro-ohmmeters after an embarassing moment in Asia that involved an incident similar to the one you described. Among those who do testing regularly, the situation you decribed above really isn't all that rare.
Where I am currently employed, we use the new Vangard micro-ohmmeter and have yet to have any "unplanned outages".
I'll step out on a limb here and say that if you were to analyze the output of your "ductor" you might just see a fairly healthy spike on the leading edge of the current rise waveform. Depending on the diff. scheme, the settings, and the relay(s) used, your "event" may very well be repeatable.
After having been involved in a similar situation some years back, I think I can say without fear of retribution, the silence after such a "discovery" is deafening.
Regards
Mike

 

[davidbeach]

Thanks Mike. In this case, as I understand it, it was a utility testing their own equipment. Our involvement is that of the manufacturer of the diff relay. It sounds to me like this kind of testing is not compatible, in terms of protection system security, with diff relays that use paralleled CTs. It the transformer diff was provided with a numeric relay where each CT connected to its own input on the relay, it would have been possible to short out the CTs on the breaker under test without degrading the protection on the other feed to the transformer.