Primary Current Injection Test on 67 Relay

[GRAEE]

We have a 67 relay (directional relay) and a power circuit breaker with 3-CTs connected in wye-grounded of 500:5 Amps tap used. Kindly see attached diagram for details.

We injected current at the primary side of the breaker specifically at phase B where the phases A & B CTs were shorted. We assumed that only phase B will have a reading even the other phases are shorted. But, when we injected current and reached to 300 A, we noticed that phases A & C registered about 67 A.

We believe that all CTs should not be shorted. It was just an experiment out of curiosity. Does anybody give an explanation or illustration how this happened...and help me establish an equation to describe this scenario?

 

[Arevaengineer]

In my opinion all three CTs must be shorted.

 

[ProtectionEngineer]

GRAEE,
What is your reading on the 67N relay?

 

[rasevskii]

You have to leave all 3 CTs unshorted for this test. The high impedance of the unshorted CTs will prevent any spill current flowing via the other phase CTs. This is assuming that the primary side is actually open also.

regards, rasevskii

 

[marks1080]

I agree with Arevaengineer. This is especially important when injecting currents. Small mistake could lead to a big problem.

 

[jghrist]

I agree with resevskii. If you short all CTs, then how do you get any secondary current through the relay element being tested? If you short the ØA and ØC CTs, then the ØB current will split among the three relay elements.

What are you using for polarizing during the test?

 

[Ohmly]

When you inject with this connection, the current has a few different options, once it leaves the Bph CT secondary winding.

Go through the shorts, then back through the A & C phase elements, or go up through the IN element.

They will then summate and go back up through IB and back to the B phase CT.

Do you have a measurement for In, also do you have phase angles ?

I would think the arrows through Ia and Ic would be in the opposite direction - the phase angle measurements would prove it.

 

[GRAEE]

I agree with you guys that it should not be left shorted. And expected spill current will flow through the shorted CTs. If you have experienced this kind of experimentation, can you express this mathematically or vectorially?

The readings i got from the display for Ia & Ic were about 67 Amps while Ib = 300 Amps. I was not able to capture the In. Likewise, with the phase angles. My apology for lack of data.

For your information, the relay has a multi-functions and we disabled the 67 & other functions and tested only the 51 (TOC).

 

[Ohmly]

Attachis a a drawing of what I believe is happening.

With the shorts applied, there are a number of paths for the current to flow, but it will all end up going back through the IB element.

If is 67 Amps in A and C, the remaining 166 Amps is going through IN

Without the shorts all seconardy current goes through IN and IB only, they will both read 300 A.

The shorts have no effect on the current in IB, but I believe the shorts are not required. If that was what was happening in real life, no current in A or C would appear on the secondary circuit.

There is no safety issue, the overcurrent relay elements are the "shorts".

 

[alehman]

I think rasevskii and ghrist have it. Yes, IB is the same either way. Without the shorts, IA and IC will be small because the open CT impedances would be high compared to the relay coils.

With the shorts, and if the relay coils have equal impedances, IA=IC=-1, IN=1.

Alan
“The engineer's first problem in any design situation is to discover what the problem really is.” Unk.

 

[slavag]

Hi.
As pointed above, you don't need short "unused" current terminals, not on the relay and not on the CT.

Please take in account Q's of jghrist :
"What are you using for polarizing during the test? "

You can't test polarization with such test, on this moment you test only wiring.

Good Luck.
Slava.

In the dynamic test, with real load, you must test with shorting of two phases and disconnection of one voltage phase.
For example current A-C shorted, voltage B disconnected.
Of cource if you use internal calculated brken delta voltage.

 

[GRAEE]

slavag,

I understand about the polarizing of the relay that should be used. But what we did was a current injection without the polarizing or reference voltage from the PTs. Its like we're testing the overcurrent functionality since the relay has a multi-functional features where 67 and other parameters are disabled. Meaning to say, regardless of current direction, the non-directional overcurrent relay element will be triggered.

regards,

 

[marks1080]

I am sorry, I didn't understand what you were doing. Normally if I was testing a relay I would short the CT links and I would break into the relay circuit via flexitest switches and inject there. You may not be using these switches and need to inject through the CT links.