NEC (zig-zag trsf) temperature trip 4

[RalphChristie]

We have 2 Yd1 transformers in parallel. (20MVA, 66/11kV each) The secondary side of each transformer is earthed with a NEC. (zig-zag transformer, 400A, 10sec) Our problems start with earth-faults on the downstream feeders. Sometimes, when a fault occurs, the NECs trip on top oil temperature. (Earth fault relays don't have time to operate) But, if you touch the NEC, it is still cold.

Now my question is:

Can the mechanical vibration in the NEC during a fault condition operate the temp relay?(Temp-relay is situated on NEC and has mercury contacts) Just have to mention the system is relatively old, the NECs are solidly grounded and not through resistors. We plan to do some earth tests to ensure the ground connection of both NECs are still good, although both show temp flags (trippings)
Or am I missing something regarding oil temperature?

Thanx
Regards


Ralph

 

[jghrist]

I can't say what is causing the temp relay to operate, but a quick solution would be to put a time delay in the circuit.

 

[busbar]

It is hard to imagine an oil-temperature switch outrunning a ground-overcurrent relay. Has the oil-temperature switch been checked for calibration? With a mercury contact, is it mounted level?

What us the routine prefault current on the grounding-transformer neutral with respect to its nameplate rating?

[Also—what does the “C” in “NEC” mean?]

 

[RRaghunath]

RalphChristie,

I think your guess is right. It must be mercury switch operation caused by vibration at the moment of fault.

You said the earth fault current is restricted to 400A and you also said it is solidly grounded. As far as I know, there should be NER in the zig-zag transformer neutral circuit. You may clarify.

When it comes to solution,

1. introduce time delay, say 0.5 sec.

2. Isolate the temperature indicators from the transformer tank surface / vibrations.

You may need both. I guess individual motor feeders have fast acting earth fault protection of their own. So, even if you make the delay 1second in the earth fault protection at the incomer, I think should pose no problem.

Hope it helps.

 

[RalphChristie]

Ok

NEC - Neutral Earth Coupler (or zig-zag transformer)

Like I said, this is a relatively old installation. There is NO resistors on the earth-point, the only thing that limits an earth-fault current, is the impedance of the NEC.
The nameplate impedance on each NEC is 45 ohm/phase. Futhermore, the time-rating is 30 sec and not 10 sec. Sorry for the wrong info.

The temp-relay is mounted level. I have never done calibration of a temp-relay before, but I can try to do something. Maybe to put the probe in oil in an urn and to heat it up, then to check the relay? Personaly I don't think the problem is hot oil, because the NECs are never warm. Mounting the temp-relay next to the NEC can be a solution. If this doesn't work, we'll try a time-delay.

It is difficult to see what is happening, because there is not many earth-faults and we don't know when it occurs. But thanks for the help!

Regards


Ralph

 

[busbar]

Ra;ph, “bench testing” the oil-temp switch temporarily removed from normal service is a way to disqualify it as a possible cause of undesirable outages. When obvious stuff tests OK, it’s time to look for non-obvious causes.

As to effort expended in problem resolution, compare cost of the work to cost and risk of a larger-area-than necessary outage. [Id est, loss of selective coordination.]

 

[stevenal]

One of our servicemen was having touble reading a winding temperature guage due to condensation inside the glass. His solution was to tap on the glass to try to get the water to run down. He decided it wasn't such a good idea after dumping the station. Yes to the vibration theory.

Qualitrol and others have electronic winding temperature indicators that can replace the mechanical dials.

Is your NEC bolted to its foundation?

 

[RalphChristie]

Busbar: Yes, you are quite right. Three factors that prevent me from testing the relay were:
Never tested something like this before (although I'm willing to try and learn)
NEC feels still cold after faults
Plant running

Stevenal: The NEC is bolted onto an installation almost 1.5m high, the installation is bolted to the foundation.

Regards

Ralph

 

[RalphChristie]

Feedback: Problem solved

Problem was the mercury switches in the temp-relays. With the initial setting the angle of the mercury-holder was such that a heavy vibration operated the switch. We increased the setting with 10°C and it never tripped for a earth fault again. We've decided to change the relays with more modern temp-relays.

I've calibrated all the temp-relays with a kettle, trsf oil and a thermostat. By putting the temp-probe with the thermostat in the oil, and heating it up with the kettle, I was able to calibrate the relays. (Compare thermostat-reading with relay-rading)

Practical experience can sometimes be of more use than theoretical knowledge. Sometimes you have to expand your view and look from a different angle to a fault to understand it.

Many thanks for all your inputs

To quote Electripete:
Eng-tips forums: The best place on the web for engineering discussions


Regards

Ralph