REF TRIPPINGS

[a430]

HI,everybody.

I have a 90MVA 330/132/13.8KV Yyd transformer that always trips on REF due to an earth fault on the 330KV line,and is very stable during normal loading.
I have sometime ago carried magnetisation test on all CTS and discovered that the ones on the Neutral bushings were of the X- class whereas the external ones are of protection class,and thus have the ones on the bushings replaced with protection types.The CT ratios on primary is 150/1A, and 400/1A on the secondary,and is always the secondary REF relay that flags.Can anyone think of a solution please.

 

[ScottyUK]

What do you mean by 'protection class'? Class X CT's are a protection class CT, just as a 5P20 or whatever is also a protection class CT. Are you saying you have removed the Class X CT's and replaced them with standard types? That sounds like a very strange decision to have made. Is this a Hi-Z REF scheme? Class X CT's are very typical for Hi-Z REF schemes. Plain protection CT's are usually acceptable for more modern schemes using numerical relays.

If the kneepoint voltages are adequately high and the mag current of the CTs is adequately low, check that the neutral CT is not reversed. This fault would give the problem you describe. Look for obvious things like incorrect wiring - and verify that the drawing you are checking against is itself correct! If all the wiring is correct, you can actually prove the relay scheme using a primary current injection test. If the CT's are internally mounted on the bushing turrets it is an awkward test to arrange because the injection source must drive current through the transformer windings.

We've set up primary injection tests on GSU transformers using a diesel generator connected to the LV (delta) side and using the HV (star) earth switches to steer current through the desired path to cause neutral current to circulate. During the test the relay coil current should be zero if the relay is wired correctly. Prove the scheme by deliberately reversing the NCT connections and verifying that relay current circulates when the CT is reversed. You may need some sensitive instruments to measure coil current.

You will need to calculate the generator size based on the impedance of the transformer and available genset. We've sometimes used a distribution transformer to step up the generator output voltage to raise the current to a reasonable level. The generator will not be happy with a huge single phase load - it is important to consider the NPS heating of the rotor.

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

I don't suffer from insanity. I enjoy it...

 

[HiSet]

Hi,
We too had similar problem on couple of our transformer installation which are 110/22kV YnOYn with LV side solidly grounded.I have found that inspite of correct connections, the REF protection scheme can mal-operate due to capacitive discharge current associated with line to ground faults on transmission line connected on EHV side.

 

[a430]

Thanks alot scotty,
Like you rightly said, the scheme is the Hi-z type,and after carrying out a magnetisation test, i discovered that the X-class cts on the nuetral bushings have higher knee points than the out-door cts on both primary and secondary, that was why i replaced it with standard type since there are only two.
Also,the scheme is balanced under normal loading conditions,which means is not a polarity problem.
Some time ago i carreid out stability test on the scheme and the current flowing through the relays is very high for faults withing the zone and very low for out zone faults.

 

[KJvR]

Hi a430,
As you might already known, Class X CTs are prefered for Hi-Imp REF protection on electro-mechanical relays. In previous threads I was advise that normal protection CTs can be used with new numerical relays although I have not done so myself. The difference with the Class X and 5P is the ratio error (Class X < 0.25% and Class 5P < 1%). The relay manufacturer will most probably advise you to use a CT with ratio error <0.25% (Class X) to prevent unbalance during a through fault. However, I just had a case where another consultant used a 5P on a electromechanical feeder diff protection and we designed the protetion on the end of the other end of the feeder. The relay manufacturer specify a ratio error <0.25% but informed us a Class 5P is O.K. depending on your settings. Still a bit unsure about this.

 

[a430]

Hello Hi-set
please how did you solve the problem ? or are you still having the trippings?

 

[ScottyUK]

a430,

Did you stability test definitely force neutral current to flow? As I said, NCT polarity reversal is a pig of a fault to find because the scheme appears stable for balanced loads and out-of-zone phase-phase faults. The problem only shows up with an earth fault.

You would probably find a day or two of consultant time with a protection specialist is a very good investment. You will have the chance to ask a lot of questions and he will be able to check the stability margin of the whole scheme from the mag curves you took plus some additional information which you should have to hand. The costs of his time will almost certainly work out cheaper than replacing the CTs with the wrong type for the relay . Personally, I love spending time with the old-timers who have been doing specialist roles for years: they are usually delighted to share their knowledge with anyone who shows interest, and it is an educational opportunity which is impossible to get in college.

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

I don't suffer from insanity. I enjoy it...

 

[RalphChristie]

Hi a430

I agree with Scotty's comments.

Also,the scheme is balanced under normal loading conditions,which means is not a polarity problem

Under normal conditions there will be no flow of current through the neutral CT - and thus no indication if the scheme behaves correctly. I have found this CT many times the culprit of nuisance trips. If you have access to the primary and secondary connections of the CT, it is fairly easy to test for polarity with common test equipment. Another thing to watch for is test-CTs (additional CT-winding for test purposes) This winding must not be short-circuited but just connected to ground on one point only. If short-circuited it will affect the performance of the protection CT.

transformer that always trips on REF due to an earth fault on the 330KV line

and is always the secondary REF relay that flags

This is strange. I would suspect a primary REF-tripping but you said it is a fault on the secondary side? Is the neutral CTs (secondary and primary) on a common grounding point?

Hi-set: Interesting comment. I have never encountered such behaviour on REF-schemes. Do you have any additional info regarding it? (Settings, problem solving, etc)

Regards
Ralph


[red]Failure seldom stops us, it is the fear for failure that stops us - Jack Lemmon[/red]

Make the best use of Eng-Tips.com
Read the Site Policies at FAQ731-376

 

[a430]

Hello Ralph,

It is really very stange,the earth fault is always on the primary and is the secondary relay that always flags, the Ncts are not on the point,but they are connected to the same station ground,and are very close to each other.

 

[HiSet]

Hi a430
Picking up of Hi impedance based REF protection due to capacitive discharge current although is rare occurrence but never the less is well documented phenomenon.I remember reading about it in a book abount fault calculation by Lacky.We approached this problem by raising REF setting beyond line charging current which can be computed as per following formula (SQ3 VL* W*C)
Regards,

 

[RalphChristie]

a430:

Might there be a situation that a portion of the fault current return through the neutral-earthing point through the secondary-side neutral CT and through the primary-side neutral CTs towards the fault? Check especially if the current might return through the neutral CT from the secondary scheme.

Hi-set:

Thanks for the reference. I am looking for the Fault-Calculation Book from Lackeys for longer than two years without any result. Strangely enough I have found other rare books (Edith Clarke, etc) relative easily.

Regards
Ralph

[red]Failure seldom stops us, it is the fear for failure that stops us - Jack Lemmon[/red]

Make the best use of Eng-Tips.com
Read the Site Policies at FAQ731-376

 

[stevenal]

I suspect the multiple grounds on the CT secondaries is the problem. Fault current in the ground is finding a parallel path through the relay coil. Near is not good enough. Connect together all CT neutrals before attaching to the station ground at a single point. When lifted, a continuity check should find no other ground attachments. This ground point should preferably be located near the relay.

 

[HiSet]

Ralph,
I know some one selling bootleg copy of LACKEY'S book for 100 bucks a piece.