Current transformer saturation 1

[Gauss2k]

I have a question about current transformers saturation.

Here's the situation. The building has a medium voltage breaker with its protection relays, on the primary side of its main transformer. The current transformers used by the relay have a ratio of 50:5A.

The maximum short-circuit current at this point is about 6kA. If a bolted fault happens, what current will the relay "see"?

I believe that the CTs core will saturate and the relay will see the maximum saturated current (maybe about 1000A (equivalent on the primary side)). Since the instantaneous protection is set at about 500A, there should be no problem with that.

But I have someone who tells me that a CT under heavy saturation could have a secondary current that would "drop", and the relay would then become "blind" to the fault. He doesn't have any source to confirm this...

I agree that if the current really drops under those conditions, it is indeed critical, but I doubt this is possible...

Anyone has ever heard of this CT's secondary current dropping under heavy saturation condition?

 

[jghrist]

The relay would probably see 20 times the rated current (1000A primary) peak, but it would highly distorted. A microprocessor relay will usually use a filter to extract the fundamental component of the measured current and it is quite possible that the fundamental component will not reach the 500A set point.

See Standard CT Accuracy Ratings in Metal-Clad Switchgear by S.E. Zocholl, Schweitzer Engineering Laboratories, Inc. at

http://www.selinc.com/techpprs/6187_StandardCT_SZ_20051020.pdf

for a good explanation of the problem.

 

[davidbeach]

jghrist is correct, the relay will "see" a lower current from a highly saturated CT than the ratio would suggest. Stan's paper provides good information and I believe that paper includes a formula that will help you estimate whether or not the CT will saturate. A good CT will take more current before it saturates than a lesser CT. Unfortunately, your switchgear is more likely to have C50 CTs than it is to have C800 CTs.

 

[dpc]

Another good reference is J.L. Blackburn's Protective Relaying book. It covers the impact of CT saturation on relay performance.

You have to look at in the time domain - because the current is at 60 Hz, the CT is going into and out of saturation as the current changes. Once it saturates, it essentially produces no secondary current - that requires a changing flux and the flux isn't changing much once the core saturates.

If you looked at the secondary current on a scope, you would see, for heavy saturation, a high magnitude blip twice a cycle, then basically zero amps.

The instantaneous element may see enough current prior to saturation to operate. That seems to often be the case. But any inverse time functions could operate much more slowly than expected, or not at all.

 

[scottf]

Also, don't forget that the saturation point of the CT depends on the accuracy rating (as mentioned above) and the actual applied burden.

For instance, a CT rated C100 is guaranteed not to saturate (defined as less than 10% error) up to 20 times rated current with 1ohm burden applied. If the applied burden is 0.5 ohm, then the CT will maintain accuracy up to 40 times rated current. Burden and saturation point effect each other directly...half the burde...twice the saturation point (current).

If you have an electronic relay and a short wiring run...you may get much higher than 20 times rated current.

 

[davidbeach]

Expanding on scottf's post, it is not the current that causes saturation, it is the voltage on the secondary of the CT that causes the saturation.

If you get enough of the waveform on the first cycle of the fault current, you may get operation of an instantaneous element, but once into saturation the current calculated by the relay goes down as the area under the curve of waveform goes down.

Low voltage circuit breaker get around this problem by having a magnetic trip that does not depend on CT performance; even LV breakers with solid state trip units have magnetic instantaneous override elements.

 

[Gauss2k]

Thanks for all the valuable information, i will study it very thoroughly.

I will try to find more information about the CT type in order to predict what would happen during a 6kA fault.

 

[Gauss2k]

I was looking for the book dpc talked about on amazon.com,
Protective Relaying (Power Engineering, 5) by J. Lewis Blackburn:

http://www.amazon.com/Protective-Re...=pd_bbs_1/002-1679553-7031209?ie=UTF8&s=books

I saw another:
Protective Relaying by Walter A. Elmore

http://www.amazon.com/Protective-Re...=pd_bbs_2/002-1679553-7031209?ie=UTF8&s=books

I'm considering to buy one of those books to have a good reference about this subject. Any preference for one of them?

 

[scottf]

Blackburn book is good. I have it now and it was the textbook used at my university for the power system protection class.

 

[davidbeach]

Both are good books, but I'd recommend starting with Blackburn.

 

[ScottyUK]

You might want to look for Areva's Network Protection & Automation Guide. It is available to purchase, but is a free download from their website. It uses IEC terminology but regardless of whether you are in the US or the IEC world it is a great reference for both the relative novice and the experienced engineer.

http://www.areva-td.com/static/html/TDE-AGF_Product-Product_Detail2_1089880316992.html

ftp://ftp.areva-td.com/NPAG/Chap6-78-97.pdf is the chapter dealing with instrument transformers.

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

Sometimes I only open my mouth to swap feet...