Restricted Earth Fault Protection 3

[RichCarnell]

I am presently testing an REF scheme on a 25MVA TX and was interested to know if anyone could tell me why the current measured by the neutral CT reverses direction when a winding fault occurs. Is it to do with the fault changing the earthed reference point on the winding causing it to act like a center tapped winding?

The TX is a delta/wye 33/11kV with the star point on the secondary solidly earthed.

 

[davidbeach]

What is the pre-fault condition? It should be essentially zero. If so, the direction of a very small number is essentially unknowable.

 

[RichCarnell]

The pre fault condition in the test scenario would be a small amount of zero sequence current (above the required pick up value) flowing through the phase CTs to fulfill the requirements of the element under test. This current is required to determine the direction in the neutral CT. The neutral CT current would be in phase with this until such time as a winding earth fault to earth is detected. The neutral current reverses at this point and if its magnitude is greater than the pick up value, it trips.

 

[IPopov]

Hi, if you have a winding short circuit to the earth this means that you have a short curcuit insite of the zone of REF protection. This current in the CT will be opposite according to the "normal" current when the SC is outsite of the TX. It is better to simulate this current (fron neutral CT) with angle 180 deg. That means you have internal SC and the protection should trip it.

 

[davidbeach]

The neutral current doesn't reverse, it is the terminal residual current that would reverse. For an out of zone fault there would be current up the neutral and out of the terminals. For an in zone fault there would be current up the neutral and in to the terminals.

 

[RichCarnell]

Yes. Thank you for your responses. I understand that the current does change direction. My question is, why

 

[davidbeach]

Current is toward the fault. If the fault is right of the terminal CTs the current will be to the right; if the fault is to the left of the terminal CTs the current will be to the left.

 

[RichCarnell]

So for an out of zone fault the current would flow from the phase terminals of the TX out to the fault, though the fault to earth, then back into the star point via the neutral on the TX. For an in zone fault both currents point in?

 

[davidbeach]

Yep.

 

[RichCarnell]

So why does the current change directions? What is feeding the fault if both the phase and neutral currents are pointing in toward an in zone fault?

 

[davidbeach]

Some remote source is now feeding the internal fault through the terminal CTs. If there is no remote source there may not be any current in through the terminals but there will still be current up through the neutral.

Why do you think that current through a set of CTs wouldn't change depending on which side of the CTs has the fault?

 

[RichCarnell]

Davidbeach thank you for your perseverance. It's not that I don't think that the current would not change dependant on the location of the fault. Is more that I'm trying to work out how the currents flow in a faulted Tx. Please see my attached sketch. I thought that the fault current would have been developed by the faulted winding. What do you think of my faulted current path? The way that I see it is that the failed winding must still generate a voltage, and it's this voltage that sets up the fault current. Your thoughts?

 

[davidbeach]

Your red current is in the wrong direction. Ia may be zero if the isn't a remote source to supply it and if it isn't zero it would be toward the fault. Ib and Ic direction and magnitude depend on many things.

 

[RichCarnell]

So the faulted winding would produce the fault current?

 

[stevenal]

The diagram is fine, as long as you are drawing the negative half cycle. Most of us would draw the arrow the other way to show the current in relation to the A phase voltage.

So turn the arrows around to stay with convention, and then draw it again for an out of zone fault. Also try it with a remote A phase LV source that also feeds both an in zone and out of zone fault.

 

[RichCarnell]

You say that the only time Ia would flow is if there were an external source? (in a faulted winding). That doesnt make sense to me as I would expect that the winding would still produce a voltage of some reduced quantity between the fault to earth and the A phase bushing? Or should I be thinking more along the lines of an ampere turns relationship between the primary and secondary, being that the fault current in the secondary would be so high as to consume all the available power transferred from the primary?

 

[stevenal]

The only time Ia as measured by the a phase CT is if there is an external source feeding your in zone fault. Ia cannot be zero according to Kirchoff, but measured Ia depends on the CT location. The assumption is we are speaking of the quantity seen by the relay. Your diagram, with no external source, shows all the current bypassing the a phase CT.

 

[RichCarnell]

Thank you for your response, although I don't agree with it, reason being that the relay that I am testing (SEL787) requires zero sequence current to flow in the phase current CTs for polarisation. How could this be if it relies on an external source to supply this?

 

[RichCarnell]

If the faulted power TX in question was supplying a delta connected distribution TX, wouldnt the healthy phases of the faulted TX then supply the fault via the delta on the distribution TX? Supplying zero sequence current back into the fault?
I think I have seen the light!

 

[waross]

As shown, the CT is on the grounding jumper to the right of the neutral terminal. The CT will not see any line to neutral load current. The CT will see line to ground current.
For an example, consider the CT to the left of the terminal. The CT will now see all neutral current, both line to neutral and line to ground.
Now consider a variable load connected line to neutral on one phase. The neutral CT (now placed to the left of the terminal) will see this load current. Now increase the load. The current seen by the CT will increase.
Continue to increase the load until the load current reaches fault level. At no point will the current reverse.
Now reconnect the load from line to ground. Apply and increase the load. The results will be the same, no current reversal.
Now consider a winding fault on an unloaded transformer.
If we consider a fault at the 95% point on one winding, then we may simulate this by connecting our test load from a -5% tap to ground.
Same result. No reversal of current.
I accept that the OP saw a reversal of current.
I believe that the question is:
Why was the test set showing a current in reverse of the normal current before the test started??

Concerning delta connections:
If the transforme under test was feeding a wye:delta transformer with a four wire primary connection, it may be possible for the fault current to reverse.

CT position:
If the CT is to the right of the terminal as shown, the CT will see only ground currents.
If the CT is to the left of the terminal, the CT will see both ground and and line to neutral currents.

Another possible cause is if one of the test connections was inadvertently made to the neutral terminal rather than to the proper ground position. Such a connection may allow the fault current to split and part of the current to return via the grounding jumper (and CT) in the reverse direction.

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