REF Protection and incorrect setup.

[ruggedscot]

Recently discovered an error that has been present for over 5 or 6 years now with a set up. We have identified that we have multiple neutral earth points inadvertently installed on a power distribution system. My question on this involves the CT set up - if we have been running with the neutral earth both before and after the CT set up then will this have had an effect on the relay protection ? I would tend to look towards having the CT and REF systems all checked and injection tested after removing the erroneous N-E link before swinging into service again. Has anyone come across this sort of error before and what was their experiences of this in the long run.

N-E link at the transformer and N-E linked again in the switchboard. Section of busbar linking the transformer with the panelboard.

 

[pwrengrds]

I am guessing that this is a 480/277 system and the CT you are talking about is the one on the main breaker ground fault. If you are then the purpose of the CT is to detect neutral current and subtract it from ground current. If it was shorted then any neutral current would be read as ground current and the system would trip at a lower level of ground current (I trip = I ground + I neutral). Yes it is common to see.

 

[ruggedscot]

Hi

It is a 400 / 230 system - standard uk fare. Building is reasonably new and looks like the paneboard installers didnt talk to the transformer installers and the checker must have had a day off the day they connected it all up....

Didnt think on the neutral / earth making it trip at a lower threshold - thats an interesting angle. Was wondering about the fact that one of the CT's would have been running with in effect a shorted turn and the other three would be paralleled up - would this have had an effect on the CT's probably not but wanted to ask.

 

[ScottyUK]

The REF neutral CT should be on the transformer neutral return prior to it splitting to the neutral bus and the earthing electrode. Its purpose is to detect all current returning to the transformer neutral. If the CT is nearer the transformer than either of the N-E bonds then it will be ok from the perspective of the REF relay although it will still be a poord installation. If there's a bond closer to the transformer than the CT then the scheme will not be balanced and may trip for an out-of-zone fault.

Can you sketch out what you have? It would be easier to work out what the effect would be.


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

If we learn from our mistakes I'm getting a great education!

 

[ruggedscot]

Putting up a sketch now thats getting technical lol....

Transformer star connected three phase out along with neutral running through busbar to the main feeder board. The main feeder board having the incommer ACB. 4 CTs are located on the L1-L3 and N bars before the ACB. 4 CT's are located on these points before the ACB. The neutral earth link is physically between these CT's and the ACB.

Restricted Earth fault - Protects the Busbar and transformer windings from fault. Should any earthfault occur the REF relay should pick that up and trip the HV 'breaker along with the ACB.
Its there to protect the bus and transformer windings from a fault that could take the HV breaker time to pick up on - limit the damage caused and protect the system from the fault.

The issue Im describing is that the system is wired as above except that they have installed a second earth at the tranformer secondary terminations between the starpoint.

 

[rasevskii]

If I understand the actual setup correctly, there is an earth connection on the neutral on both sides of the neutral CT. Therefore this CT is shorted out on the primary side. If this is the usual high impedance REF protection with 4 CT secondaries in parallel connected to one sensitive relay (high impedance), then effectively the relay input is shorted out also, and will not operate for the internal faults that it is supposed to detect.

As Scotty has said, a sketch would be useful.

rasevskii

 

[ruggedscot]

Find attached a quick drawing hope it makes it clearer...

The shorted turn effectively shoring out the other CT's is what I had thought and only wanted to confirm this. Im also dubious about this in operation for any length of time having an effect on the CT's - maybe upset the balance etc.

 

[rasevskii]

Yes that is exactly the problem, as shown in your sketch, the "2nd ground" has to be removed. This means that the protection would not have been working.

Now you should have the REF protection retested by primary injection to verify that all the CTs are correctly polarized. If not, when being taken into service, a false tripping could occur under load or through fault.

rasevskii

 

[Bramb]

Hi everybody,
with that kind of uncorrect scheme, in case of earth fault, the results would be both overfunctioning and underfunctioning of REF protection.

You can have overfunctioning in case of external earth fault: in this case, as the two ground points are connected to the same earth, the total fault current will involve one of the 3 line CT's (the faulted phase one), but just roughly the half of the fault current will involve the neutral CT (restraining the REF), the other half will pass through the "2nd ground" and it will be seen as a fault current by the REF (that will probably trip ).

In case of internal fault you can have underfunctioning as you'll have the same current distribution described above: the half of the total fault current could be insufficient to trip the REF (I don't think this is possible with typical REF settings, but theoretically it can happen).

I don't see any other effect on the CT's functioning.

The fact that plant was in service for years with this error probably means that no fault has never happened or that REF trip is in some way inhibited: please check it!

I hope this should be helpful.
Ciao
Erminio

 

[waross]

You may want to also check the resistance of the grounding electrodes.
Some of the ground fault current will return on the neutral making it less sensitive.
Some of the neutral current will return on ground leading to possible false tripping.
A fault to grounded equipment will follow the equipment grounding conductor to the panel and then split between the neutral and the ground connection.
From there it depends; Is there a solid connection between the two ground points or are they separate electrodes? This will influence the percentage split of the ground fault current between the neutral bus and the ground system.
If you have any large line to neutral loads It may be well to check for inhibited protection operation and/or poor connections in the grounding system.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[ruggedscot]

Cheers for this information -

The earth connection between the two points would be pretty much solid and not through electrode - bit unclear from the drawing. The metal of the busbar would be one path and Im sure that there would be other paths availiable through the installation between the 2nd earth and the starpoint ground. Issues thought about would be the fact that in effect this extra earth would become in effect a parallel neutral causing extraneous return paths between the board and transformer.

The fact that it has an effect on the protection is a worry and will be attended. I had some concerns about the low impedance of the path between the two points making in effect a shorted turn and this having some effect on the action of the CT's. In this sort of situation I feel that it may be best to replace the CT's as a matter of fact to be on the safe side with this?

 

[waross]

This shouldn't damage the CT.
This is not a shorted turn, it is parallel paths, some through the CT and some bypassing the CT. In this case the current is all going in the same direction (assume right to left). If this were a shorted turn the current would go through the CT from right to left and return from left to right.
If another conductor were to be passed through the CT window carrying a current, then transformer action would induce a similar current in the loop. There will be a current induced in the secondary winding that will be proportional to the primary current. This should be within the rating of the CT and should not cause damage.
I would not anticipate damage to the CT unless you have had a sustained fault with currents in excess of the CT rating.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[ScottyUK]

Agree with others that there is no risk to the CT from this mistake. The scheme certainly won't behave as designed, and you've been lucky not to have a mis-operation.

Arranging a suitable primary injection test can be awkward because of the high impedance of the transformer winding - one method I've used successfully on much bigger transformers has been an LV generator with the AVR replaced by a small variable DC power supply to give control of the temrinal voltage. On a small transformer a decent size variac is a possibility because you don't need much primary current to prove the REF scheme is stable. You can steer current through appropriate CTs using the earth switches or you can fit links.


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

If we learn from our mistakes I'm getting a great education!