Possible Transient Recovery Voltage Issues 3

[crthompson]

I have recently been introduced to Transient Recovery Voltage issues when a vacuum contactor clears a fault. We had a failure of some lightning arresters in the field and some of the engineers think that it is a TRV issue. I have a basic understanding of what occurs when a vacuum contactor clears a fault and how this causes problems. My question is this: Can a TRV travel from a 4160V secondary on a XFMR to a 995V secondary on the same XFMR? The 4160V secondary has its own lightning arresters at the XFMRs terminals and so does the 13,800V primary. The 4160V secondary has several heavy inductive loads controlled by vacuum contactors. When the fault occured, the 995V circuits were energized to the line side of the MCCBs but they weren't closed. We had 3 lightning arresters (each phase to ground on a high resistance grounded system)these were installed on the line side of the 995V MCCBs within a couple of feet. There is also a reactor on each phase of the 995V circuits between the XFMR and the MCCBs. These arresters failed very violently and caused a subsequent 3 phase to ground fault on the line side of the MCCBs. I would also be grateful if any of you could share any other ideas as to what may have caused this. Thanks in advance.

I am an engineer, so the glass is twice as big as it needs to be.

 

[Skogsgurra]

I would say yes to your first question. The dv/dt from a vacuum contactor is very high, especially if it is clearing a fault and very much so if the load is inductive.

The fast dv/dt couples easily through winding-winding capacitances and can probably find other ways as well.

To me, it sounds like your arresters are not ready to absorb the energy in the surge caused by the cleared fault. Did the designer really calculate with maximum stored magnetic energy during a worst case fault? I think not.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[crthompson]

Thank you Gunnar. I will update when I have more info available.

I am an engineer, so the glass is twice as big as it needs to be.

 

[magoo2]

Are these MOV arresters or older silicon carbide ones? That will impact the energy handling capability.

Also what was the voltage rating of the failed arresters?

 

[crthompson]

As of yet I have been unable to aquire the ratings of the surge arresters. I was assigned damage assesment of the fault and I haven't been able to speak with any of the engineers since I was on site. Even if the 995V surge arresters were inadequate, shouldn't the 4160V arresters have taken care of the surge before it got to the XFMR? If this is the case, would it mean that the 4160V surge arresters were oversized?

I am an engineer, so the glass is twice as big as it needs to be.

 

[waross]

A question Gunnar.
Does the capacitive coupling mean that the voltage on the 995 Volt winding may approach the voltage on the 4160 winding?
Does the transformer ratio become less important than the voltage division of the inter-winding capacitance any any other impedances to ground?
If such is the case then the surge may have been a non-issue on the 13.8kV side.


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

 

[Skogsgurra]

It is difficult to have any opinion on this, Bill. There are so many unknowns. Winding configuration, distance between windings, length of PE conductors and other things. I have had equipment damaged from lightning in a 1600 kVA, 11 kV/690 V transformer where we are absolutely sure that there was no hit on the 690 V side (buried cable from transformer secondary).

There's a difference between lightning and clearing a fault with a vacuum contactor. The lightning is usually less powerful (if it isn't a direct hit) than an overvoltage from a fault. Especially if there are inductive components involved.

I do not think that the arresters on the 13.8 kV side has any influence in this particular case.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[waross]

Thank you Gunnar.
My point is that given the very high dv/dt of the transient there may be little current through the transformer winding and as a result, little transformer action.
I am thinking that the worst case for winding configuration may be when the winding end that takes the worst hit from the surge is closest physically to a winding end serving the 995 Volt loads.
I am thinking that in this case the EMF on the end of the 995 volt winding may approach the voltage of the surge. The actual surge voltage on the 995 Volt terminal would be reduced more or less by the other factors you mentioned,
Other winding configurations may be more conducive to insulation failure to ground of one or both windings.
I agree with you.
I suspect that a surge caused by a very fast dv/dt may cause voltages on the secondary in excess of the greatest voltage possible due to transformer action.
This may explain the 995V Volt arresters failing before the 4160 Volt arresters.

Gunnar; Please, if you think that I am off base then RF my entire post to avoid confusion. If my question/suggestion has merit then please RF this message.

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

 

[prc]

In three winding transformers, surges can be transferred between windings( capacitively plus inductively). It can be 10-50% of the peak of the impinged surge on primary winding.In this particular case the transferred surge ( as % of transferring surge) to 995 V winding from 4160V circuit will be more than that from 13.8 kV as the capacitive coupling between 4160 &995 windings wil be more. The concentric winding arrangent in such transformers normally will be 13kV-4160V-995 V-core.Of course, absolute value of transferred surge will never equal the transferring surge peak, but can be sufficiently high enough to be more than the rated BIL of the winding or terminal bushing or surge arrester.That is the reason 4160V LAs are not failing.Please checkk the rating of 995 V arrester. It has to be rated for the line to line voltgae and not for line to earth voltage.Also some times it will be required to provide LA not between line to earth,but also beteen phases. Please see the IEEE guide for application of LA and/or ABB guide on MV arresters.

 

[waross]

Thanks prc.
Yours
Bill

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

 

[VTer]

The arresters MCOV rating on a resistance grounded system would have to be sized based on the line-to-line voltage not line-to-ground. On a SLG fault the un-faulted phases will see essentially phase to phase voltages.

"Throughout space there is energy. Is this energy static or kinetic! If static our hopes are in vain; if kinetic — and this we know it is, for certain — then it is a mere question of time when men will succeed in attaching their machinery to the very wheelwork of nature". – Nikola Tesla

 

[rocmor]

The "contactors" on the 4160V side - are they indeed contactors or are they circuit breakers used for switching the heavy inductive loads? The reason for asking is that the materials used for the contacts inside the vacuum interrupters are different for (motor) contactor and general circuit breaker applications. VIs used in contactor applications have low chop currents (0.5A) compared to 3 to 3.5A avg for CB applications, the reason being, as you might suspect, to provide a "softer" interruption and less stress on inductive components.