Relay response with CT Saturation

[dpc]

I'm looking for information on how the instantaneous element on an GE IAC-53 electromechanical relay might be expected to respond to a highly saturated CT secondary current.

My previous understanding was that these induction cup instantaneous units would normally operate in the first 1/2 cycle or so before the CT saturated.

I understand the distorted and reduced secondary current that results from saturation but I'm trying to determine if the saturation could be so severe that the IT element would not operate at all.

If anyone has any practical experience or test data, I'd be interested - thanks.

 

[davidbeach]

There comes a point at which the saturation is bad enough that the element won't work. Unfortunately, if that happens, it happens at the top end of the current range. Higher accuracy CTs help. Sorry, but I don't have any good references to offer.

 

[dpc]

OK, thanks, David. That's actually the answer I'm hoping for since it would explain an apparent breaker failure that doesn't seem to have another explanation at present.

The breaker did not trip on a fault and there are conflicting reports on whether or not the relay target was dropped.

It would be nice simple answer to the problem, but I'm trying to make sure that it isn't too simple.

 

[davidbeach]

dpc, do you know where the fault happened? If so, do you know the expected fault current at that location? What are the CTs, both ratio and accuracy (if IEC I ain't gotta clue, hopefully ANSI) Cxx? How was the relay set? What is the burden on the CT secondary circuit at that anticipated secondary current? You could have a situation where there was enough current for the IT element to pickup, but that current lasted for only a fraction of a millisecond (per cycle) and the relay only chattered but never actually moved far enough to trip.

 

[dpc]

Thanks. There is no information available on the CT class. This is old metal-clad switchgear, so I'm guessing the CTs are C100 at best.

We know the magnitude of the fault current (6000 A)from the upstream digital relays that eventually cleared it (after 20 cycles). The fault was transient, so the location was never determined for sure.

I don't have any doubt that the CT saturated, since it was only a 150:5, and the time element was set on the 0.5 amp tap - this is a burden of over 4 ohms. But the expected response of the old IT unit is still a little unclear to me.

 

[stevenal]

This relates to microprocessor relays, but is still interesting. The switchgear standards need to be revised.

http://www.selinc.com/techpprs/6187.pdf

 

[dpc]

Thanks, stevenal. I did have this paper already - it is interesting. There are a lot of industrial facilities using metal-clad switchgear that probably have terrible issues with CT saturation during faults.

Another common problem in metal-clad is the use of multi-ratio CTs - like a 1200/5 set on the 150/5 tap.

 

[RalphChristie]

dpc

You can do a magnetizing/excitation curve test on the CTs to determine the knee point voltage. (secondary induced voltage against exciting current)

With Vs = IL x (Zct + Zburden + Zlead) you can determine the highest secondary voltage the CT can produce without saturation.
where:
Vs - secondary induced voltage
IL - Maximum secondary current
Zct - internal impedance of CT-winding
Zburden - Connected burden impedance
Zlead - Lead impedance

In your case you'll just have to measure the burden impedances of the instantaneous element, the secondary leads and the CT-winding.

Probably you have done this, but assure that all the other elements in the protective scheme are in a working order. From the most common causes I've found on trip failures (with instantaneous units) are:
Damaged relay contacts
Sticky shunt trip coils
No back-up power



[red]Failure seldom stops us, it is the fear for failure that stops us - Jack Lemmon[/red]

Make the best use of Eng-Tips.com
Read the Site Policies at FAQ731-376

 

[dpc]

Ralph,

The relays were bench-tested after this and worked fine. The breaker was replaced with a spare and it tested OK.

I wish I had more info on the CTs, but this gear is 40 years old and without pulling out the breaker, there is no way to know for sure what is in there. And they are not interested in any more outages at this point.

 

[stevenal]

dpc,

You don't need to pull the CT to perform an excitation test. You just need to get to the secondary leads with the primary open. See IEEE C57.13.1.

 

[dpc]

stevenal,

Agreed. I was thinking about pulling the breaker to physically look at the CTs and perhaps a nameplate. But the test might actually be safer.

This gear will be replaced in another year or two, so probably no one (except me) would be interested in investing in CT testing.

Thanks.

 

[davidbeach]

Ralph & stevenal, thanks for picking this up, good answers.