Protracted Arcing In Substation 2

[Mbrooke]

Video of what looks like arcing at a Con Edison substation, possibly East 13th st:

https://www.reddit.com/r/SubstationTechnician/comments/h9d07n/no_no_no/

From my vantage point it looks like a series arc on a 69 or 138kv disconnect switch. My question is- why didn't relaying catch it? I'm assuming differential protection is present.

 

[itsmoked]

Love the turquoise color. Don't recall seeing that rich a color before in arcing disasters.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[davidbeach]

There’s not much of anything, protection wise, that would detect and clear a series arc. If it’s in a differential zone the current in and out still balance and the diff won’t trip. You’d have to detect it on negative sequence due to the unbalanced impedance in the zone or on some sort of voltage disturbance. Once it goes to ground or to another phase it’s easy to detect and clear.

 

[sushilksk]

Series arc is not easy to detect by protection system. When arc length is small, change in voltage or current will be small. When arc length is more it will convert into pahse to ground or phase to phase to phase fault. But in most of the cases, damage is not too much, may be fingers of disconnectr.

 

[Mbrooke]

Why wouldn't the energy being dissipated by the arc trigger anything? I know turn-turn faults in a transformer are capable of being detected.

 

[davidbeach]

It’s just current in air rather than current in metal. Until it contacts something else.

 

[Mbrooke]

I have a possible theory... a turn to turn short will cause current to flow in now newly created "third"- and as such more current will entire the primary to keep the core magnetized vs current leaving. So more current in than out when ratio adjusted.


An arc is a series impedance- no different than if load on one phase dropped or was reduced outside the differential zone.

How I think it- makes sense now.

What through me off were guys on another forum saying relaying would have caught it. Hence my double take. I thought I had finally gone mad lol.

 

[cuky2000]

Mbrooke,

Here is additional info hopefully help with your theory on arcing faults.

 

[davidbeach]

Those would be shunt faults, not series faults.

 

[davidbeach]

We had a series arcing event recently. Made great middle of the night video, but no relay on the system took note of it.

 

[stevenal]

Turn to turn faults in a transformer change the turns ratio, so current at the terminals does change. Think turn to turn fault in a reactor instead.

 

[Mbrooke]

@Stevenal- can you further elaborate?

 

[sushilksk]

In case of turn to turn fault, it will act as short circuit winding.

As shown above current I3 will flow in short circuited winding. But this current will not come into calculation of differntial delay. After inter turn fault, due to I3, ratio of I1 and I2 will not match the expected value.

 

[prc]

Stevenal, turn to turn faults will not be like series arcing. True the current in shorted turns can be massive and may melt copper, but the current drawn from line may be only marginally high.Shorted turns will not act as a separate winding but more as a part of auto connected winding. So ampere turns in shorted part will try to match with AT of balanced part of winding. So for detecting turn to turn faults, only sure solution was gas operated relays- Buchholz relays.

That was the situation may be till 20 years back. Now engineers have developed alternate algorithms for detecting inter turn faults in transformers as negative sequence currents, wave form change etc. An Indian professor even wrote a book on this subject !!But still gas operated relay may be the only one relay that acts in such a situation.

 

[stevenal]

prc,

I was trying to make the point that shorted transformer turns were not like a series arc.

But if the shorted turns balance with the ampere turns of the remaining turns in that winding, I guess there will be zero amps in the unfaulted winding (two winding transformer assumed) so the overall ampere turns remain balanced? I don't think so, it will divide up according to ohm's law. Still, you cannot have a turn to turn fault without it affecting the differential current. Whether or not a differential relay will reliably trip for such unbalance is another story. I agree gas pressure relaying or alternate algorithms are needed as well.

 

[itsmoked]

We now have to use arc-detection breakers in our houses. Why isn't there an arc-detect relay that 'listens' for the unique electrical noise generated by them?

Keith Cress
kcress -

http://www.flaminsystems.com

 

[davidbeach]

In your house you have one arc-detecting breaker in the path between the source and the arc. In a networked transmission system you could dozens of relays at dozens of locations that all would all sense fault but none of them could determine if they needed to trip or not. It would make a mess.

 

[itsmoked]

That answers that! Thank David.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[prc]

Stevenal,the AT balance is like this - let faulted turns in primary be fn and current in loop fi. Then (pn-fn)pi=(ft.fi)+(st.si) where st is turns in secondary. Due to turns differences, increase in primary current will be small and hence the difficulty in detection by conventional relays.
Itsmoked- arc detection relay ( is it ELCB?) is new learning for me. Can you give a link for more details.

 

[itsmoked]

Hi prc;

Here's the USA take on it.

Here's a much better European Eaton presentation.
(Sorry for any Youtube ad, yesterday Youtube went ad-insane.)

Here's a guy explaining the six tripping indicators available in the Eatons and demonstrating its trips

Keith Cress
kcress -

http://www.flaminsystems.com