CT saturation curve from secondary injection

[GBell]

I used a Megger CTER-91, followed procedure and produced a nice saturation curve of 1200:5 CT's. With known sceondary voltage injected and known primary amps measured, how do I proceed to interpret the results? What computations do I do and what nameplate values from tested CT's are pertinent? I would greatly appreciate some guidance.

 

[collies99]

Please kindly give us the nameplate spec and a scan of your curve. Thanks.

 

[GBell]

At the moment CT's are on energized 161 KV OCB's. Can you help me with translating the secondary voltage and primary amperage? 1200:5 ratio. With values of 240.4 volts into secondary and 1.0186 amps out of primary, how do I do the math to reconstruct the amount of fault current that would saturate the CT. Assuming the 2 values above represent point of saturation.

 

[oldfieldguy]

I think you need to look at what's going on with your test. With the primary circuit equipment locked out and isolated and the secondary wiring disconnected, you inject voltage into the secondary and measure current in the secondary. As you step the voltage up, secondary current will increase in a basically linear fashion in relation to the injected voltage until you start saturating the core. As you near saturation, a small change in secondary voltage will make a larger change in secondary current.

You can plot this data and compare with published curves for your CT, or you can compare like items. As a rule of thumb, the saturation voltage will be a bit higher than the CT's accuracy class for C-class CT's.

Now, as to what happens when you fault? You can reconnect your secondary wiring and inject an amp at the CT terminals (or the closest terminal block to the CT for normal purposes)and again measure the voltage that your test set takes to push this amount of current. Here, the CT's winding is a high impedance path, so the vast majority of the current will flow through your protection scheme. With this voltage and current you can determine your current loop impedance and calculate your expected terminal voltage at the maximum calculated fault current. If you want to save yourself the calculation, take voltage it takes to push one amp and multiply that times 100. In either case, if the result is less than the saturation voltage you just determined, then your installation should not get into trouble with saturated CTs.

If you have too much circuit impedance, you can easily see what fault level will put the CT into saturation.

Now, since you're dealing with OCB's be sure that you check the saturation on all the CT's. they should be very close for equal ratings. I have seen cases where shifting hardware has shorted the primary path, causing those CTs to drastically drop in saturation due to the shorted primary path. In one case I investigated, this caused a misoperation of a protection scheme on and out-of-zone fault.


old field guy

 

[GBell]

oldfieldguy, thank you very much for your time and response. Is there any values I can compute, numbers I can back into using the 240.4 volts low side injection and 1.0186 amps primary observed? NERC PRC-005-1b requires I verify instrument transformer outputs and correctness of connections to protection system. Can I mathmatically compute anything with the 1200:5 ratio, 240.4 volt secondary injection and 1.0186 primary amps produced (I'm using those 2 values as point of CT saturation) that will help prove out/use as evidence that the CT's will not saturate before generating the 5amps to close relay contacts and trip breaker. I'm an old navy nuke machinist mate struggling with electrical concepts and NERC standards.....Thanks again, oldfieldguy and collies99 for your responses. Am I unrealistically expecting to compute values on the high side from what data I have?

 

[davidbeach]

Nope, no values you can compute. Yep, unrealistically expecting too much. You have to know the burden on the CT, that's all the stuff you had disconnected when doing your test, and you have to know both the magnitude and X/R of the primary fault current. You can't back into a primary magnitude because the X/R of the fault affects saturation. But knowing the nature of the primary fault, it is easy to determine the degree of saturation (or not) that can be expected. Then there's the relay's response to the saturation, and ...

But if you've shown the CT still has its C rating (assuming here), know the burden and the magnitude + X/R of the primary fault current then

(1 + X/R) * I_pu * (|R_B + j * ω * L_B| * 100) / V_R < 20

tells you that you won't have saturation problems.

I_pu is the magnitude of the fault current in per units of the CT rated current.

R_B and L_B are the burden resistance and inductance.

V_R is the C rating of the CT.

It can be a bit above 20 and the relays still won't have any trouble. Let it get into the hundreds and the CT response is really ugly.

 

[FreddyNurk]

There's some information missing from your posts that would indicate what would be required to evaluate saturation, though oldfieldguy has actually outlined the process.

The nameplate of the CT would indicate if the CT is within specification or not, as all that has been presented is the turns ratio, whereas other information (e.g. 5P5F20 or 5P20 1VA for IEC areas) would indicate whether or not the CT is above or below specification.

The other main aspect is what the CT is connected to, as the burden of the CT secondary wiring and the protection device dictates whether or not saturation would occur and at what level.

The connected burden and the voltage for the knee point will give you the maximum amount of current that can flow in the secondary circuit before saturation. Apply the turns ratio to this current and you'll get the maximum amount of primary current that can flow before the CT saturates.

 

[MeinKaun11]

SORRY FOR PITCHING IN GBELL QUESTION:

FOR EXAMPLE IF I REQUIRE A C200 CT BUT IF I BUY c400 , WOULD IT
MAKE ANY DIFFERENCE IN TERMS OF PROTECTION APPLICATION; IS THERE ANY
DISADVANTAGE IN BUYING A BETTER CT?,