CT Secondary injection 1

[afroeeng]

It is proposed to carry out secondary injection with the relay still connected to the CTs (not using test terminals but injecting directly onto the loop). Has anyone done such a test before, i doubt it is meanful?

 

[electricpete]

I assume you are mapping the saturation curve.

If you separate the loop remotely from the CT for current injection and voltage measurements, then you will have voltage drop that appears in your measurements which is not related to the CT. At first glance you'd think you could subtract out a constant-resistance / linearly-increasing voltage from the result to isolate the CT portion of the voltage, but that doens't account for phase angle. I don't think you'd have much hope of getting a meaningful curve although you could probably tell the knee pretty well and if comparing 3 CT's on similar circuits and they look similar, that's a good sign as well.

Of course it wouldn't hurt anything to inject current remotely if you measure voltage locally at the ct.. then you would get the true ct characteristics and presence of remainder of circuit doesn't affect the measurement.

=====================================
(2B)+(2B)' ?

 

[electricpete]

Then again maybe you are interested in testing the relay.

If your tester can produce the required current (which could be a problem) I personally don't see problem with ct being in the circuit (other than perhaps higher probability of accidentally opening energized ct circuit).

There are probably others that can provide better answers than me.

=====================================
(2B)+(2B)' ?

 

[electricpete]

Actually on a 3-phase circuit if you tested one phase by injecting into ct circuit you may see response on more than one phase. You'd want to step back and look at that.

=====================================
(2B)+(2B)' ?

 

[lz5pl]

I have done such injection several of times, mostly in MV-cubicles and CT's with 1 A secondary. There is no much difference (actually I cannot find any) in the current through relay with connected or disconnected CT secondary. That is because CT secondary winding has much greater impedance than that of the relay.
Of course firstly be sure that CT is not energized primary !

------------------------
It may be like this in theory and practice, but in real life it is completely different.
The favourite sentence of my army sergeant

 

[GTstartup]

Agree 100% with IZ5PL, I also could not find any discernable difference even with a 5A secondary but I suppose it depends on the size of the CT.

 

[electricpete]

Assuming that we are not using ground as return path, then if we are testing from the relay the current will pass through more than more than one ct secondary winding in order to complete a loop, will it not?

I would imagine that would give qualitatively useful results, but not directly comparable to direct testing of a single ct.


=====================================
(2B)+(2B)' ?

 

[ScottyUK]

Just be careful if you have circuit or bar earths in place - they can provide a path for primary current to circulate which does cause incorrect results.


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

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

 

[rasevskii]

Yes if the CT primary circuit is open, as Scotty said, there will hardly be any current through the CT secondary (a few mA), if it is the relay you are testing. But if you actually have a test block in the circuit, why not use it properly and isolate the CT. That is what test blocks are for.

Let us hope that you are not carrying out such tricks on energized circuits...

rasevskii

 

[afroeeng]

Rasevskii, my interest is not the relay but the CT. We changed a transformer from 3 x single phase gen trans bank. The new transformer came with it's own cts(coz the originals could not fit) . we've done mag currents on the cts on their own. Like i said in my first post - i ain't really sure about this test. This test has basically been recommended by some guru as part of commissioning in the new cts - and we are to carry out the test on the affected phase only.

To amke matters worse- we have interposing star-delta ct between the main ct and the relay ( this is for the trans diff protection)
We even have more complicated arrangements on the feeder cts which means during injection there is likelihood of multiple current paths into the other cts .
LZ5PL says he/she has done the test with the cts on before - but is this arrangement a standard test? What do we achieve that can't be accoumplished by perfoming stand alone test of the ct followed by the relay , ir & loop resistance?

 

[rasevskii]

for afroeng:

Well, you don´t achieve anything by leaving the CT connected to the relay if it is the CT itself that you are testing, as said (mag curves, IR, etc.)in fact you cannot possibly do a mag curve if the relay is connected.

Do a primary injection at the CT with all the circuits in service connection, and see that the correct secondary current can be measured at links, test blocks, and the relay itself. If this is a trafo diff protection with aux CTs and something was modified or changed from the original, then a 3-phase injection has to be done onto a short circuit at the other side of the trafo, so that operating currents can be simulated (about 10% of nominal current is enough) using the generator itself (temporary excitation supply regulatable from zero upwards), or a temporary LV generator used to supply the SC current instead.

The above would be tests that were done at the original commissioning only.

In fact, if the CTs were changed as you said, it will likely be that the trafo diff protection will trip due a wrong connection or wrong polarity somewhere. Therefore the above 3-phase test to avoid unpleasant surprises...

How big a unit is this and what sort of relays do you have?

Voltage levels? MVA?

How old is the installation, roughly?

rasevskii

 

[oldfieldguy]

Testing by injecting current at the terminals of the CT may be useful in seeing unexpected burdens or circuit paths. It provides a pretty safe answer to the "Does THIS CT go to THAT input".

That said, I personally would be very reluctant to do this on a CT on a live circuit. There are too many hazards involved.

Also be advised that circuit burden may cause unusual readings. Under conditions where the applied voltage is low in relation to the actual CT saturation voltage, the current flow will be through the current circuit, with the CT essentially acting as a very high impedance.

In other words, an amp applied at the terminals should return an amp at the metering and relaying. However, if circuit burden produces a high voltage (near saturation voltage), a significant portion of current will flow through the CT windings, resulting in part of your injected current NOT showing up at the end devices.

old field guy

 

[Ceast]

Yes, it's a great test. Verification of current circuits, wiring polarity and burden is doable with a test set. We do it all the time for commissioning. Secondary injection with all wiring terminated and the circuit dead. Nothing to do with saturation.