Distribution protection understanding desired 3

[itsmoked]

Last night our power went out. About 30 seconds later it came back on.

Didn't think much about it. Somebody pushed the wrong button, a squirrel committed suicide, meh.

About 30 minutes later the wifey and I are on a walk and we go down our local arterial to the main street through our side of town and there are 6 police cars and and a ladder truck surrounding a car that t-boned an on-the-ground transformer for a business ground floor/apartments top floor at the intersection. Power was out on the main drag but not on our arterial that split off essentially at the scene of the accident. The power on the main drag remained off and hours later the main drag had POCO trucks all over it and guys wandering around looking up and down (3AM).

My question is how did the crushed transformer result in our arterial being without power for merely 30 seconds? I'm coming up blank on any theories.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[davidbeach]

The feeder has instantaneous overcurrent enabled on the first trip. Then a 30 second open interval followed by an automatic reclose. Any fault results in the 30 second interruption you saw. The pole with the transformer was on a tap with a fuse where the tap connects to the main feeder.

In the classic explanation that instantaneous trip of the feeder is a attempt to save the tap fuse (fuse saving scheme); trip before the fuse blows, hope the squirrel or the limb falls free during the interruption, and then all is well on the reclose. What I've been seeing is that often the fuse clears the fault before the breaker opens and the whole feeder is blinked even though the fuse cleared the fault.

I’ll see your silver lining and raise you two black clouds. - Protection Operations

 

[itsmoked]

I learned something new today. Thanks a bunch David.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[waross]

This is old school but may illustrate some of the thinking behind distribution protection.
Most of the distribution outages are short term faults, birds, rodents, falling branches.
The next most common cause of outages was transformers failing and shorting out.
Some of the older electro-mechanical re-closers reclosed 4 or 5 times.
First would be an instantaneous trip followed by a few seconds to allow a rodent, bird or tree limb to drop free.
There may be three re-closes followed by very quick trips.
If this did not clear the circuit, the fault may be a shorted transformer.
There would be a following re-close with a longer time before trip to attempt to clear the fuses on a shorted transformer.
This may be repeated for a total of 5 recloses.
This would leave a circuit in service after the great majority of faults.
This is complimentary to David's information.
Others may share other re-closing patterns.

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

 

[itsmoked]

Bill that's more along the lines of what I've seen before but what I saw this time wasn't adding up without David's explanation.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[davidbeach]

Don’t know about those ampersand guys, but on our system the 30 second open would follow the second trip. Our first trip is followed by a high-speed reclose. Most people, and their clocks, don’t recognize that first open as an actual outage.

Every utility does it slightly differently, our high-speed reclose is far from universal for example. But the general tools are all pretty similar.

I’ll see your silver lining and raise you two black clouds. - Protection Operations

 

[Palletjack]

Another possibility..
A close onto fault self healing scheme.
Car hit transformer, causing the protection to open, and close. Itsmoked is on the load side of the XF. He never saw a reclose.
The self healing was actuated either by a timing scheme or communication scheme.
30 seconds (45?) sounds like a timing scheme.

 

[cranky108]

But... reclosing on an underground transformer? I believe many utilities don't reclose on their underground systems. But maybe a good part of this feeder is overhead.
Fuse saving, does sometimes work, but on smaller fuses, it does not. Just a matter of how fast the fuse really is.
We use fuse blowing, so we can reduce the number of outages, and the number of customers out.
Does not work that well as all the customers see the voltage dip anyway.

What fuse saving really does is to reduce the number of times the utility needs to call out a line-person to replace the fuse, thus saving the utility company money.
I guess that does save the customers in the long run.

 

[davidbeach]

We now have a lot of feeders with automatic retrieval of relay event records. They were highly enlightening.

Our most common substation transformer is 16.8MVA, 115/13kV; within a couple of miles of the substation there's no actual fuse saving, even a 140T fuse will clear faster than the breaker can get open. All of the smaller tap fuses will blow and the smaller they are the further out from the substation they'll blow before the substation breaker can get open. After looking at many event records where the breaker opened to interrupt load current after the fuse had already dealt with the fault current (can't tell the breaker "oh, never mind"), we've started using a delayed (definite time) instantaneous on a lot of feeders that used to be "fuse saving" with a zero delay instantaneous. That way if the fuse is going to blow anyway we let it without tripping the breaker, but if there's a mainline fault it still gets interrupted quickly. With a 6 cycle delay on the instantaneous we've found that we can significantly reduce the number of feeder operations without having to go to more than 20 cycles for a time overcurrent trip.

I’ll see your silver lining and raise you two black clouds. - Protection Operations

 

[Mbrooke]

Its possible there are SCADA controlled recloser loop schemes in place or motorized switches. IF not automatically, its possible someone called the POCO asking for a de-energization of that area. They remotely opened the reclosers around the scene and then closed normally open ties to re-energize loads normally fed by the effected segment.

 

[cranky108]

Not knowing about POCOs system, most of us could not say if they have that level of automation. Most utilities don't have that level, because of the cost, and not that much return for the customer. Granted, for larger load densities it would make since, but most suburb communities it likely would not be.

In the inner city is where it makes since for most automation to restore customers quickly.

For those of us in rural areas, hours are more typical, and in even more rural areas, it can take days.

 

[Palletjack]

We are very rural.
We also have a self healing loop scheme that is timing based rather than comm based. It’s a 22 mile loop. Works with six 651R controls.
Works VERY well, and does almost exactly what the OP described.