CT TESTING 1

[cranky108]

Is there a good method of testing CT's in bus differential circuits without taking an outage?

The ones with signific current on them is eazy, but what of the others?

 

[cranky108]

Ok I don't want to preform an accurate test, I just want to know that the thing is in the ballpark (working or failed).

It looks like there isen't a better way than the clamp-on measurment, and verify with another CT on the same breaker.

 

[stevenal]

Is this a system that has been fully commissioned, has been in service, seen through faults, and is now subject to NERC 005? So far, NERC has not specified the testing interval, or the extent of testing needed. Do to the risk inherent in this type of testing, perhaps you might want to decide that a periodic visual inspection is all that is needed. This decision must be justified, perhaps by utility experience. I know of no standards that address periodic testing of CTs. We passed audit using this type of approach.

 

[cranky108]

Yes we have several installations that have been in service for several years. The question has been how do other companies test these.

The other thought we had is that we verify that they have not misoperated on other system faults. Take a SOE log and calculate an equivlent short circuit study to show it would have seen current.

 

[oldfieldguy]

Back into the fray...

The big electrical utility for whom I used to work tested CT's periofically. I worked as a contractor for another larger utility conglomerate that did periodic testing of CT's.

One rationale for testing of CT's mounted externally to GCB's was that one manufacturer had problems with sealing the CT enclosures against water intrusion. Water in the CT steel corroded the steel and negated the effects of inter-plate insulation. This caused electrical paths to form in the steel which resulted in lower saturation voltages.

The second problem area was with CT's inside OCB's. OCB action causes a considerable mechanical movement. The jarring has caused CT's to shift on their mountings. I have found cases where this shifting caused the CT mounting plate to contact the bushing, resulting in a shorted turn in the CT parallel with the primary current path. This caused a severe drop in saturation voltage and is the cause of a differential operation on an out-of-zone fault.

In both situations the CT's accurately reflect the current passing through them. The problem is the change in saturation voltage.

That being said, a two-prong periodic testing approach might be prudent: Compare secondary currents with other metering to prove ratio and phase angle, and when the circuit is in an off-line condition, perform saturation tests.

Yes, it's a low-probability situation, but much of periodic testing is.

And all this verbiage assumes (!) that the circuitry ahs gone through proper commissioning tests and has been in service with no abnormal indications.

I won't tell you about finding a set of GE PVD relays with the trip lead lifted and taped because "they just rip at the most inopportune times"...

old field guy

 

[cranky108]

That is one nice thing about the 587Z relays is that a second setting can be applied at a much lower setting to provide an alert to some problems.

 

[DiscoP]

Not sure if this is acceptable for your requirements, but we periodically test the BBP CTs by ramping up voltage on the secondary terminals and checking the current.
This is done when each feeder is out for maintenance.

 

[DTR2011]

OK -

This middle aged field guy has an "in service test" for proving CT's and associated wiring for a Hi-Z Diff scheme.

I only recommend that this be performed by experienced and qualified field personnel who are good with hand tools.

**********************************************************

NO I am NOT CRAZY, I know what I am doing.

**********************************************************

I am assuming you have back up OC protection and some flexibility in switching. I would recommend setting the back up OC more sensitive. Any other overlapping Zone protection could also be set more sensitive.

The 87B trips from the lock out (86) relay will have to be disabled for the test. As a matter of fact, the audible sound of the 86 will speed things up. I would certainly verify that there are absolutely no 'sneaky circuits' in the 86 tripping scheme. It wouldn't hurt to verify that there are no DC grounds prior to commencing with the test.

#1 - 87Z theory - assuming that all secondary wiring and the CT ratios / polarity are correct, there will be no diff current at the 87 relay terminals, under load. The only way to prove this scheme (the anti test) is to create a false diff current / voltage and let the relay trip the lock out (86) device. Utilize load switching if you have a by-pass / maintenance breaker for double bus arrangements. This

***********************************************************

NO I am NOT CRAZY - I know what I am doing

***********************************************************

The idea of the test is to open circuit the CT at each summation point. Yes, I said open circuit the CT.

Use insulated needle nose pliers and a holding screwdriver. It only takes a second. Hold the lug / wire with one hand (in insulated pliers) and the holding screwdriver with the other - lift the wire just a little bit. You will expect to
see arcing and hear the 86 roll. The wire only has to be lifted just enough for an arc to develop. For those who are interested, there is a time constant for a dangerous voltage to develop. For the 30 cycles that you are lifting and the relay is operating only a few hundred volts will develop. I would guess that David Beech could fill us in on the exact formula.

For those still concerned about personal safety, the MOV's (or metrosils) will generally clamp at ~600V or so, which I believe will satisfy OSHA (or other local safety bodies) if you are wearing Class 0 hot gloves and properly rated insulated tools.

That proves the rest of the CT's in the scheme. Repeat for every CT that is included in the summation of the 87Z scheme. You may need a helper and two way radios if the CT's are paralleled in a marshaling cabinet in the switchyard, which is the proper location for this scheme. Have the helper confirm the trip and reset the 87 and 86 relays after each CT is tested.

In my experience, it takes longer to explain what you are doing to those who do not understand the 87Z scheme than to perform all the tests in the scheme.

************************************************************

Just keep explaining that you are NOT CRAZY and understand what you are doing.

************************************************************

The primary injection scheme proves restraint for through fault conditions, by making an open circuit, sufficient voltage and current is developed to test the MOV's, the relay 87H and 87L element and the trip to the 86 device.

***********************************************************

This was a 'patented trick' I learned from another OFG some years ago. I am passing this trick on as I enjoy the quality of the discussions on this board with some obviously sharp engineers and field technicians.

The first time I did it, my heart was pounding and I was sweating like a pig.

I commissioned 2 8 breaker ring buses with 587Z and MFAC relays using this method in about 4 hours some years ago.


************************************************************

All the best!

 

[electricpete]

I am sure you are experienced and know what you're doing. But you're right... it sounds crazy.

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

 

[scottf]

smallgreek-

As someone who works for a CT manufacturer, I would say that testing scheme is dangerous.

Also, how exactly does the test prove the CTs to be in proper condition?

 

[GTstartup]

Besides it doesn't prove anything more than if you would just short the CT

 

[cranky108]

Shorting a CT on a high impedance bus differential is the same as taking the differential out of service. And in fact measuring a voltage on the differential circuit is a way to prove there isen't a ground connection.

 

[GTstartup]

I think if you shorted only one CT you would see that it's not the same as taking the relay out of service

 

[stevenal]

High Z bus differential has CTs paralleled. Short one, and you've shorted all on the same phase. No voltage can develop at the relay, so relay is out of service.

 

[ScottyUK]

Many of our relaying designs have the Metrosil close up to the relay. If you O/C the CT then the Metrosil doesn't clamp the voltage. A few years ago I inadvertently caused this condition on a very large Class X CT on a generator LV side diff scheme. I've described the circumstances in more detail on here before but suffice it to say that it was a brown trouser moment. There was a fair bit of damage to the lug and the terminals - hardly what you want during an online test.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[NAZ55]

Cranky,

It seems like you have answered your own question here

The most efficient way I have seen is to just use a clamp on ammeter.

If not available, then after disabling the diff trip take one CT offline at a time (short circuiting secondary side) and reading the high side and low side currents from the relay. Subtract one from another and the difference should be what your CT would put out.

 

[NAZ55]

Please ignore my second suggestion. Didn't notice that you were talking about bus diff. However electronic relays should provide you current on each line and the sum of all should still give you the number you are looking for.

 

[cranky108]

We sort of decided on visual review of all CT's in the bus diff. that we can't verify with a clamp-on.

These are mostly future and moble tie breakers (Yes we have breakers for future equipment. Makes good spares also).

Thanks very much.

 

[laketrash]

We @ TVA simply measure current and phase angle at relay with one of the Bus breakers current diconnected (Pull PK inside breaker cabinent). The measured current should be close to the current expected from that breaker except with opposite polarity. This is only true however when all Bus contributing currents are connected to Diff. circuit and the measured current is close to 0. Do this check first. You can do this procedure on all breakers (one at a time)to check C.T. integrity and it's associated wiring.

 

[tommom]

Just take a voltage across each input of the bus differential relay (if it's for example a GE SVD bus relay). Any voltage substantially higher than zero (if your bus has tested out) will tell you if you have a failing CT - just maybe not which one.