Why did relaying fail to clear this fault? 1

[Mbrooke]

Talk of the year, looks like a bus fault:

https://www.youtube.com/watch?v=Wvz6j_xBzsI

Good birds eye view of the damage:

https://www.youtube.com/watch?v=NgBQ_oCrXkw

But my question is why did this fault persist for a whopping 8 minutes? Why didn't bus-bar differential catch it? Breaker fail if the case? Why didn't zone 2 or zone 3 pickup at the remote ends when all local relaying apparently failed to do so? Has me scratching my head...

 

[HamburgerHelper]

Arcs can have a lot of impedance. The worst types of industrial arcs that I have seen pictures of are those that have so much impedance nothing trips and everything around it just gets burnt up. Not that I think is what happened here. Con Edison had a transformer that blew during hurricane Sandy, too, that lit up the sky.

I would like to know why it took the operators 6 minutes to do anything.

------------------------------------------------------------------------------------------
If you can't explain it to a six year old, you don't understand it yourself.

 

[Mbrooke]

I agree, they have impedance, interestingly though at one point it sounded like the event caused dimming of lights in Queens and outages at LaGuardia airport indicating the event was able to pull down voltage significantly at a very strong source. A fault with such low impedance should have caused relaying to pick up.

From Con Ed's twitter statement, pictures and CBS's chopper footage my best guess is that it looks like a capcacitive coupled voltage transformer failed in some manner on their ring bus, caused an arc between the busbar persisting for several minutes.


Regarding Sandy and East 13th st; flood water disabled protective relaying. A piece of debri hit a busbar section and instead of the fault being cleared and isolated locally to just the section, it persisted causing remote circuit breakers to clear the whole station and in turn cutting power to 6 other substations.

 

[rcw retired EE]

We worked on the 500MW Astoria II power plant that feeds into a ConEd station at the Charles Poletti power plant. When I saw the news, I was concerned that it was my relaying or our station design/build that lit up the sky for that long. The 500MW power plant probably did feed some fault current, it is only a few hundred yards away. I haven't heard if the backup distance and overcurrent relaying actuated anything at the Astoria power plant. Those were set with relatively long coordination time delays, but less than 8 seconds.

ConEd had dual differential relaying on bus and cables at 345kV which should have cleared a bus or cable termination fault quickly. Several 345kV oil filled/cooled cables run from Poletti/Astoria into Manhattan. I couldn't tell if this failure was in one of the cable termination bays that would be protected by the cable differentials.

ConEd protection philosophy in 2010 included three shot reclosing on the 345kV feeders into Manhattan from Poletti & Astoria. When one Manhattan substation failed during Sandy, I thought I could see the flashes from the reclose attempts on the youtube videos. That many reclose attempts at 345kV concerned us, but power to Manhattan was very high priority for ConEd. I don't know if that philosophy is still used.

It would be instructional and interesting to learn what let the fault continue that long.

 

[Mbrooke]

Just a bit of fault current, in I think 2012 they added an auto-transformer and phase angle regulator connecting the 345kv Astoria GIS to Astoria east 138kv ring bus.


Now that you mention it, considering Sandy it would not surprise me if they now completely disable remote backup or have it set so high that anything other then a dead short circuit would trigger it. Had that fault been cleared by remote relaying it would have resulted in de-energizing 16 positions removing significant generation, dropping North Queens substation and radially isolating two others substations possibly dropping load at one or even both of them.


I can understand the re-closing though- some of the busbar and terminations are exposed to open air so it would not be bad idea IMO when pros/cons are weighed.


I doubt it would you your relaying in any case.

Waiting is making me anxious, really educational to learn what caused it.

 

[crshears]

Not often encountered, but complete DC failures have occurred, often with truly catastrophic consequences.

CR

"As iron sharpens iron, so one person sharpens another." [Proverbs 27:17, NIV]

 

[rcw retired EE]

2nd hand information is that the fault was one or more 138kV CCVT's or PLC Coupling Capacitors. Local 138 kV generation tripped off but 345 kV generation was not affected. No details were provided on any protection action, inaction or failure.

This event could be a by-product of a "Keep power on at all reasonable costs" philosophy.

 

[Mbrooke]

@rcwilson: Thank you for the info. Your are indeed right, here is the official explanation by Con Ed and the Governor:

https://www.youtube.com/watch?v=URxWqhMB3ak

Just for the history books, intensity of the arcing as seen across water:

https://www.youtube.com/watch?v=CU_mOuNaaIc

 

[d_k_m]

Few things to note:
Analysing the fault without having right detailed information viz. Sequence Of Events, Disturbance Record Files, Pre fault information etc. will not be right things to do. Let us wait for the report team prepares and then we review the report.

BTW few things I also have in back mind:
1. It was addressed in the video that there was significant fault current; in that case differential should have picked up (above bias current of 0.3% Full Load Current);
2. If the fault was initiated from Voltage Transformer device, whether this equipment is connected in protection Blind Zone;

If anyone of us get the Fault Analysis Report; please share; thanks

 

[rcw retired EE]

Good points DKM. We did learn there were unusual failures/problems in both the primary and backup relaying systems. Murphy's Law at work?

 

[marks1080]

Do you have any feeders or any other elements that sent a 'block' signal to the bus protection? I hate those set-ups. Very prone to misoperate.

 

[dpc]

I'm late to the party, but we shouldn't assume the problem lies with the relays. Possible the relaying operated correctly, but the breaker did not trip for some reason. It happens.

 

[rcw retired EE]

My information is 2nd hand hearsay from a client that was affected by the fault. They had spoken to the relay engineers involved in the preliminary investigation but did not share a lot of details. We'll share more information if it becomes available and is in the public domain.

 

[Mbrooke]

@DPC: I agree, but a substation of this importance would have breaker failure relaying.

 

[marks1080]

Breaker fail has to be initiated by another protection, it's not a primary fault detecting protection.

 

[Mbrooke]

I know, but DPC brought up relaying possibly having operated normally (sensed the fault) but the breaker did not operate when called to. In such a case breaker failure would have been activated by the protection relaying, sensed current still flowing through the stuck breaker and initiated opening of other breakers around the stuck breaker.

 

[Mbrooke]

Update on what happened.

https://newyork.cbslocal.com/2019/02/11/astoria-borealis-city-council-hearing/

Here is Con Edison's testimony, starting 17:45 on the incident. Click on "video" below "agenda" in the top left:

https://legistar.council.nyc.gov/Me...B4-9A8B2AA325A5&Options=ID|Text|&Search=1318#

 

[waross]

Wow! That bylaw amendment proposed by Costa G. Constantinides belongs in/on Project?oard.
Under consideration to replace gas powered plants is zinc batteries.
It's been done. grin
See this delightful spoof posted in the joke thread by John R Baker.

https://vimeo.com/254035263

Unfortunately the Mercedes AA Class seems to be powered by alkaline batteries not zinc batteries.
Alkaline batteries are not included in the proposed bylaw change.
Costa G. Constantinides amy have missed a viable technology here. grin

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

 

[thermionic1]

At about 19:04 of the video something is mentioned about the sensing device (relay) and communication device (unspecified)...Earlier it was mentioned that a voltage monitoring device (VT or CCVT) failed. It was also mentioned that a redundant system was implemented from a different manufacturer.

I would think that at the system voltage (115-138kV), bus differential would be a standard protection scheme. In the EM days, a PVD,CA or MFAC did the trick, feeding a LOR to clear the bus (perhaps backed up by a bus OC, etc). In modern times we use maybe a SEL 587Z and/or 487B or GE B90 or equivalent. Some choose to implement both a High and Low Impedance scheme. What caught my attention was the mention of a communications device malfunction. Was this some kind of 68150 error? I can't imagine how a traditional (EM or IED based) scheme could fail, short of a DC system failure. I see battery chargers that are 30+ years old that have some kind of alarm to SCADA.

I feel for the ConEd guys having to explain that solar, wind or cow poop energy still has to go through a substation. I had to stop watching soon after the discussion went to solar panels and rebates...

 

[Mbrooke]

Might very well be the case. I am not sure what Astoria had, but Con Edison already upgraded East 13th st to IEC61850 in response to Sandy and I know that they are extensively perusing the technology.

https://resources.news.e.abb.com/at..._in_New_York_City_to_improve_power_supply.pdf

https://cdn.selinc.com/assets/Liter...Cannot_AA_20170207_Web2.pdf?v=20180604-215323