Search is on for a cause, and a remedy

Philadelphia Inquirer ^ | Fri, Aug. 15, 2003 | Tom Avril, Inquirer Staff Writer

[Mo1]

Shortly after 4 p.m. Thursday, instruments recorded 300 megawatts of power moving eastward toward New York. Then in a sudden change of direction, 500 megawatts surged westward, Gent said. Wild fluctuations continued for 9 seconds at various points in the Northeast

What would cause this???

[Mo1]

FYI

[cake_crumb]

From what the experts have told us so far, it was lightening, fire, fire and an anomaly at a car manufacturer.

My own is that the delivery mechanism is old.

[Cold Heat]

Likely a provider or multiple provider loss in the west. This bit of info may only be the result of the fault and not an indication of what it could be.

The fault was definitely down line in the west. it could be as small as a 10$ part or timer that failed to disconnect a fault from the grid in time to keep the power plants from going into self protection mode.

Did I make sense?

[cake_crumb]

Better monitors to pinpoint problems aren't going to solve the problem of not having enough towers to deliver energy.

[Mo1]

My own is that the delivery mechanism is old.

Possible .. But another freeper on a different thread made a comment about wasn't all this equipment suppose to have been updated for Y2K

Weren't they are worried about a major blackout back then and what could happen?

[Mo1]

Did I make sense?

*L* .. yes it did, Thanks

[FairOpinion]

"What would cause this??? "


---

Good question. There is something very strange here. I think that those advocating that we just go back to sleep until another and worse thing happens must be out of their minds, like that professor, who was saying that it worked well so far.

The disaster potential is enourmous and we need to everything in our power to find out what happened and make sure that it will never happen again.

[Cold Heat]

It's been my experience that faults generally cause disconnect in about 2-4 seconds. For some reason the safety failed to disconnect the section where the fault was from the grid.

What puzzles me is that this safety stuff is redundant.

In other words a progressively larger section is dropped from the grid. I think this failure will send the engineers into four pot a day overdrive for quite a while.

[FairOpinion]

More puzzling information from another article:

""Our data-gathering network shows that in the first minute -- between 4:09 and 4:10 p.m. Eastern time -- power plants hundreds of miles apart in three different states began tripping offline as the grid became unstable," says Mr. O'Leary. In the second minute the blackout spread much farther, "but it was not a simple domino-like failure moving one area to an adjacent geographic area. The failures hop-scotched all around the Northeast and Midwest regions and into Canada," he says.

"Our unique set of data will help everyone involved find the answer to the most important question -- why the systems that were installed to prevent widespread blackouts didn't function as intended," Mr. O'Leary says. "We are working to increase the availability of our data to maximize the understanding of the source of the blackout event."

http://biz.yahoo.com/prnews/030815/laf061_1.html

[Dolphy]

Did I make sense?

Help me some more. There is a surge (or wasn't it?) eastward and then a surge westward. Was the initial 300 megawatts of power in response to a demand? Was there too little capacity to handle that? And why exactly would it then surge westward? Was this a reaction to the 300 megawatts. Sorry for what may be ridiculously simple questions.

[FairOpinion]

As I am reading articles, I also just found out that 22 nuclear reactors were also shut down.

"He said more than 100 power plants - including 22 nuclear reactors in the United States and Canada - were shut down, and the blackout affected 50 million people in a 9,300-square-mile area from New England to Michigan."

http://www.fortwayne.com/mld/fortwayne/news/local/6541688.htm

[Cold Heat]

The y2K situation only involved processor powered equipment.

Most grid systems have modern monitoring gear, but the working end (The Switches) are mechanical/electric vacuum breakers and transfer switches. The get a signal or lose a signal and they operate. They have built in delay timers and re-connects that are simple and uncomplicated for reliability and usually are 99% reliable. (that is why there is built in redundant for the 1%)

The modern stuff is just a shiny version of the old. Something weird happened and I cannot figure what. My guess is that a big power plant dropped off line to soon. A technician could have mis adjusted a trip or a timer failed perhaps.

But the area where the fault was that started the cascade should have dropped from the grid prior to the power plant having to disconnect.

[cake_crumb]

I know PA managed that, but only because we started working on it long before it became a cacheword of the Clinton Admin. Canada was partnered with PA in the upgrade, which were dubbed the PA2k and CA2k programs respectively. Western New York at least partially upgraded, but all or most of the upgrades were with the computer systems. I've read several times that the delivery system itself has been basically unchanged since the 1950's.

Two years ago, President Bush wanted to upgrade the grid. Which resulted in that stupid Senate non-investigation of Cheny and his choice of energy exect to consult with. The Senate "investigation" died out when media furor turned to other things. Nothing came of it, and no energy policy has been passed.

[Mo1]

The disaster potential is enourmous and we need to everything in our power to find out what happened and make sure that it will never happen again.

I agree ..

I'm not an expert in this area .. but is it normal to have such a westward swing like what was mentioned?

[cake_crumb]

The nuke plants shut down automatically.

[Fellow Traveler]

There is something about this article that does not make sense. They are talking about this event taking one or two minutes to occur. Everything else says the whole thing was over in nine seconds, which is about the time frame I speculated it would take last night. If it really did take a couple of minutes it would indicate a much more serious problem with the grid managers. A nine second long event happens so fast that only automatic responses will occur. If it took two minutes for the blackout to stop it would give operators time to respond.

It may be that most of the blackout happened in the first nine seconds with several systems on the edge struggling to maintain power for a minute or two.

[Mo1]

It's been my experience that faults generally cause disconnect in about 2-4 seconds. For some reason the safety failed to disconnect the section where the fault was from the grid.

Yes, but shouldn't there be more trigger/switches/whatever to have stopped this from being hit buy so many areas? .. They all seemed to have failed till it hit Valley Forge

[FairOpinion]

"The nuke plants shut down automatically. "

--

I know -- but think of the implications, a couple of such mysterious failures could blakcout the entire US, cause all power plants to shut down. I understand it takes days to bring the nuclear power plants back on line.

[Cold Heat]

Most non electricians have problems with the term "surge".

Best thing is to forget the term cuz it is really meaningless.

As I understood the article, the data likely came from a monitoring station at a power tranfer station near NY.

The guy said they were importing power at a 300 million watt rate and the it reversed so they were exporting at 500 million.

All that says is that the problem was not in NY. That about it. The term surge in electric jargon only means someone opened the faucet real fast.