Restricted Earth Fault 2

[a430]

Hi Everyone,

I have a 90MVA 330/132/33KV Transformer that is tripping whenever there is an earth fault on any of the 330KV lines,i carried out magnetisation test on the star point CT and discovered that they were having different charecteristics with the ones on HV side and thus replaced them.Also, i have reduced the sensitivty of REF relay by increasing the plug setting .With all these the transformer is still tripping. Can anyone please help.

 

[RalphChristie]

I suspect a CT with a wrong polarity, most probably the CT in the star-point.

Is this a high impedance scheme?

[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]

Thanks Ralphchristie.

It is a high impedance scheme,and at one time i treid to reverse the polarity during some checks on stability and discovered it to be more stable the way it is at the moment.

 

[RalphChristie]

Was there a time that the scheme behaved correctly?
What type of relay do you use?
Is the scheme incorporated in a differential scheme?

The most common faults I found for incorrect operation of high impedance REF schemes were:

Incorrect settings
Reversed polarity of a CT
Faulty CT

It seems as if your problem is on the HV-side of the transformer. Your three HV-CTs should be in parallel with the CT in the star-point, and they should be in parallel with your relay. First of all, make sure that all these CTs are Class X CTs with a similar rating. Do a ratio test, mag-curve and polarity test on each CT and ensure they are correct. Just a side-note, there is not probably a test-CT incorporated in one of the CT-cores? If there is, do not bridge the test-winding.

If your CTs are correct, the problem is somewhere in the wiring or in the relay. Circuit should be grounded at one point only. Ensure the connections are tight and check especially the test-block connections. The relay will be voltage operated - with a variac you can vary the voltage and ensure it operates at the correct setting.

Hope it helps and good luck.

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

 

[ScottyUK]

I fully agree with Ralph - the most likely cause of the observed mis-operation of the scheme is a problem with the neutral CT. I'll add two more peculiar faults which have plagued my year with REF schemes:

Wrong CT in circuit - easy to do if your transformer has a Class X for the REF and a common CT for the SBEF relay and both CTs are of the same ratio. The scheme becomes unstable on heavy through faults. You will need to carry out a mag curve test to verify this.

Wrong ratio CT - system is marginally stable, depending on the difference in ratio. Primary injection is your only option here if you aren't confident which CT is which.

Do not trust manufacturer's terminal and core markings. Do not trust field wiring either. Assume nothing until you prove it, wire by wire if need be. An OEM terminal marking error caused me a lot of extra work fault-finding one of these schemes because I trusted that it was correct. Never again!

If you have exhausted all the reasonable possibilities from visual inspection, basic electrical tests, mag curves, etc, your best bet may be to carry out primary injection on the scheme. Depending on whether your CTs are internal or external, this could either be a very simple or a fairly awkward test to carry out. Somewhere in Eng-Tips is a post I made ages ago which gives some clues how to carry out such a test. I'll try to find it, or I can give you some hints if you need them.


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

Your body may be a temple. Mine is an amusement park...

 

[khadisa]

ScottyUK, Please give us hints on how to test a bushing CT.

 

[RalphChristie]

What kind of test do you want to do?

 

[a430]

Thanks Ralphchritie

I have earlier on carreid magnetisation test on the cts,and discovered that the bushing one is not the same class with that of the outdoor, and thus shorted it out and mounted one with similar charecteristics,and the relay type is CAG14ZF14A.

The circuit was designed a very long time ago and has no differential scheme on it.In fact there was an auto- transformer on the circuit which developed a fault and was thus shifted, and this present transformer that is not auto but with exactly the same ratio mounted.

 

[ScottyUK]

Exhaust the easy possibilities first! There is a fair bit you can check from the mag curves, knee point voltage, and the resistances of the interconnect wiring before you have to get involved with large-scale testing. You should calculate the worst-case voltage required from the CT - if the knee point of all CTs is high enough, then the mismatch should not be too much of a problem. If the neutral CT can saturate then the scheme will likely be unstable during an earth fault. If all the calculations using the data you collect from measurements - not the nameplate data - say that the scheme is good, then consider testing. Don't forget to compensate measured CT resistances for temperature.


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

Your body may be a temple. Mine is an amusement park...

 

[RalphChristie]

a430:

Good points raised by Scotty.

It difficult to do fault-finding across a forum - it would have been far easier to do the test self or at least could have checked the results from the mentioned tests.

In my opinion - from the facts you've been given:

I have a 90MVA 330/132/33KV Transformer that is tripping whenever there is an earth fault on any of the 330KV lines

Tripping when earth fault occur outside the zone leads me to believe there is a problem somewhere in the neutral CT's circuit. The only time current will possibly flow through the neutral CT is during earth faults. I also make this assumption because the scheme is stable during normal operation. You do not mention through-faults.

I have earlier on carreid magnetisation test on the cts

You do not state what the knee-point voltages obtained from the tests are and what the theoretical voltage (worst case) would be across the CTs / relay during a fault. It is thus difficult to make any comments, but as a rule of thumb the kneepoint voltage of the CTs have to be at least twice the calculated voltage developed across the relay.

and the relay type is CAG14ZF14A.

I am not familiar with this relay, is it a current relay (with an external stabilizing resistor) or a voltage relay? (with an integral resistor) If it is the first, check if the value of the stabilizing resistor is still correct.

[red]do not trust manufacturer's terminal and core markings. Do not trust field wiring either[/red]

Absolutely. Mark this in red.
Star for you Scotty.



Hope it helps

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

 

[ScottyUK]

Ralph,

Thank you for the LPS - I get the impression that we have walked down similar paths in our respective countries, as we seem to share many thoughts and experiences. You last post earned itself an LPS too.

I assume the relay is an Areva MCAG14, or perhaps one of its ancestors, which is a current-operated relay designed for use with an external resistor. I think the 'CAG' part of the designation may have been around a long time, but denotes the same generic relay type.


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

Your body may be a temple. Mine is an amusement park...