Big blackout. What happened? 40

[Skogsgurra]

When I got words about the big outage, I immediatley went to my puter to find out what my engineering friends in the US had to say about it. But no Eng-Tips page available. Of course I can understand that. No power - no Internet.

Power was restored piece by piece and I now find Eng-Tips up and running again. My question is still valid: What happened?

Glad to see you again!

Gunnar Englund, Sweden

 

[lengould]

jstickley: Superb posting. Clear and (sounds) authoritave. The 9th frame carries heavy condemnation of Ontario IMO. ".. stays interconnected and tries to support both Michigan and Ohio for 2 minutes." Clearly an operating error or absence of forethought in the operating rules. Could be called cause of problem for all of Ontario and New York. Definitely needs fixing.

 

[jstickley]

All --

Although the information presented in the ITC analysis seems very clear, the root causes of this event are still under investigation. I would caution everyone to be patient and let NERC (and others) run the course of their investigations, and try to refrain from jumping to conclusions.

The one thing that bothers me most about this entire situation is the amount of finger-pointing that is going on, most all of it based upon very preliminary information. So many different organizations and individuals are trying to use this event as fuel for their own agendas, and almost all of them are doing so without firm information regarding root causes of the event to back their stance. Let's all stand firm, and try to refrain from playing the blame game (note, I too am guilty of this). Answers will be given in due time; if appropriate, blame can then be placed upon those who deserve it.

 

[jbartos]

Suggestion: The following links from above are reproduced:

http://www.itctransco.com/documents/B_8-14-2003_ITC_Analysis.pdf

http://www.spectrum.ieee.org/WEBONLY/special/aug03/black03.html

to focus on their contents addressing the weaknesses in the power distribution and generation hardware. It is clear that the hardware can be designed with redundancy, and with safe loading that the grid would be "selfprotected" or "immune" to more than one fault in a specific location. As the bottom link indicates, a series of adverse events in one narrow location triggered the "grid dominos" to fall. This is the actual vulnerability of the system, which has to be remedied to avoid other blackouts. The built-in robustness in the power transmission and distribution will be a good investment for the future considering that the oil reserves depletion is inevitable in the future time that has already been estimated / projected.

 

[lengould]

jstickley: Agreed. Sorry for stating any conclusion, just running off with an "ass"umption. If makes you feel any better, I'm not involved in industry and have no influence whatever, just one more interested resident of area affected.

 

[SidiropoulosM]

A new interesting article with a lively discussion on the blackout is posted at:

http://www.energypulse.net/centers/article/article_display.cfm?a_id=450

And two more articles and discussions from the same site:

http://www.energypulse.net/centers/article/article_display.cfm?a_id=456

http://www.energypulse.net/centers/article/article_display.cfm?a_id=449

 

[spitfire]

In the reference

http://pro.energycentral.com/professional/news/power/news_article.cfm?id=4087012

there are these two paragraphs at the end:

Even before last week's blackout, federal regulators were keeping close tabs on the way power is managed in the Midwest.

Since May, the Federal Energy Regulatory Commission has been on site monitoring the Midwest Independent Transmission System Operator, which began operating less than two years ago and oversees the Ohio utilities that suffered line failures.

Do any of you attach any significance to the fact that FERC people may have been on site?

 

[jbartos]

Suggestion: It appears that this last blackout will gradually become forgotten just like big New York City blackouts in 1965 and 1977. Visit

http://blackout.gmu.edu/main.html

for more info

 

[lengould]

Thats pretty depressing. Curious. No such thing as web in '66 or '77. Where were these surveys getting their input? Whatever it was, doesn't appear to be available now.

 

[RonShap]

I have a client with a facility on Long Island NY. His building is part of a load curtailment program. When the utility company foresees a load problem (or any problem) on the grid which would ease by reducing load, they contact this building owner to start their standby emergency generators, and have them remove themselves from the grid until notified.
This summer, surprisingly, they had not been contacted as they have been several times per summer in the past, to remove themselves from the grid ....... until 12:30PM on the day of the blackout (4 hours before NYC went down). Somehow, LIPA knew to activate their load curtailment program, 1 1/2 to 2 hours before any major items are documented in Ohio.

 

[SidiropoulosM]

RonShap, this is interesting. It probably indicates a low voltage situation on the LIPA system. Which means heavy loading of transmission lines rather than a generation deficiency. This may explain why the Ohio events caused the cascading which led to the blackout.

It seems odd that after three weeks we still have to speculate. I wonder how long it will take before we know the root cause.

 

[SidiropoulosM]

Regarding the threat of a terrorist attack on the grid:

"Government scientific advisers and officials painted a grim picture Thursday of the consequences of a terror attack on the nation's power grid, saying that any outage that lasted longer than a couple of days would reduce urban centers to chaos and collapse the economy.

"With power out beyond a day or two, both food and water supplies would soon fail. Transportation systems would be at a standstill ... natural gas pressure would decline and some would lose gas altogether -- not good in the winter time ... Communications would be spotty or non-existent. ... All in all, our cities would......"

visit:

http://www.washtimes.com/upi-breaking/20030904-070049-7423r.htm

 

[lengould]

At least I agree with this statement.

quote <a successful assault is &quot;a lot easier than we wish it were.&quot;> Seems they do a lot of wishing.

Quite humerous to read about the PhD student causing the stir. Genius? Sure it wouldn't require Einstein to map utility or fiber grids, just a lot of unimaginative effort. They give PhD's for that sort of thesis? Course it may have included a lot more that referred to in article.

I think they were 'way alarmist in article, though everyone cited and the reporter himself had a clear interest. Something like Y2K computer problem, only ones talking were those with something to gain, including journalists. Of course way out of proportion.

Was just talking with my son about the 1 km asteroid just spotted with 1000 to 1 odds of hitting earth. Were speculating on result. He'd heard that if it hits, will throw enough smoke to darken sun for 9 mos, enough to cause near total crop losses worldwide for at least 1 year. Would be 10 yrs warning, but we agreed that, with reasonable absence of panic very few would need to starve. Reality different though, I suppose.

I can't agree with doom scenario described, however. In 1999 Montreal area was hit with power loss due to ice storm in middle of winter. Most people were without power for about 2 weeks. Large city, subways etc. Much hardship but very little real injury. People very soon figured out how to survive the event, and in doing so proved, I think, they could have done it much longer.

But I may be wrong (again).

 

[jbartos]

Comment: A couple of years ago, a major Diesel genset manufacturer was giving a presentation of its genset products. It was mentioned that 6kW to 10kW gensets were being in high demand by home (house) owners. Apparently, there is some distrust about the electricity supply availability no matter whether the electricity grid is reliable or not. The apartment buildings, except a few, do not have back up gensets, in general; therefore, it is much harder to survive the blackouts in apartment buildings.

 

[jbartos]

Suggestion: Visit

http://www.nautilus.org/energy/grid/papers/bickel.pdf

for:
Figure 4 above shows the desired performance of
a grid in which a 1300 MWe NPP suddenly trips.
In the short term (first 20 – 30 seconds) only the
availability of interchange power prevents the
grid from collapsing. Over the longer term, the
ability to dispatch hydroelectric power recovers
the grid frequency to the nominal 50 Hz.

 

[spitfire]

As a newbie here, I went right away to

http://www.nautilus.org/energy/grid/papers/bickel.pdf

and there is a rather interesting point to be found about nuclear plant standards.

It appears that U.S. plants are predicated on the availability of reliable shared power from the grid, precisely because it is a big grid full of interconnects. The author puts this in contrast with, say, Puerto Rico, where the grid is not as stable. However, the author later says, &quot;overall the trend of U.S. [nuclear power plant] trips has decrease significantly over the ast decade and reliability of onsite power systems has increased. However, as a result of electric utility industry restructuring and the lack of addition of significant new generating plants, capacity margins are less. This brought with it the potential for larger cascading grid events.&quot;

BTW, the author glaringly erred by talking about the &quot;1966&quot; big Blackout on his first page, when it was actually 1965.

 

[lengould]

Star to jbartos for informative article. Bet it can get pretty hairy in the control room of a pair of reactors on a hot August afternoon when the whole state trips out. All eyes on the temp. indicators. Startup the hamsterwheel pumps. Well no, but I stayed in a Days Inn last night.

I still don't quite see why every reactor isn't required to have 250% over-capacity in its condenser installation (capable of sinking total Mwt of the reactor), capable of operating continuously as long as operators want. In a trip scenario, down the turbine / generator and bypass the steam directly to a condenser designed for the purpose. Add a small standby steam turbine / gen. in that circuit which is capable of servicing plant site [ x 2+], leave the reactor running until enough information available to decide if necessary to shut it down. Sure fast-start diesel generators / pumps or eductors could keep such a condenser within sufficient limits to avoid damage to the reactor in all scenarios.

Might at least have saved Darlington from poisoning its reactors, with the week+ delay in restart. I mean, they've got a whole great lake right outside the door. Dont mess up the lake by design, guys, but if it's that or a meltdown, by all means, boil a cup.

Course a bit more complicated with CANDU's than the others since need to also run the moderator cooling pumps etc. Still, should be possible to set up an 8 reactor site with enough redundancy to service all that reliably, even when completely isolated from the grid.

But then what do I know.

 

[lengould]

For a chuckle, try this site. Everyone onto bandwagon.

http://www.evworld.com/databases/storybuilder.cfm?storyid=567

quote< Lovins predicted the Big Blackout was inevitable twenty years ago in his book, Brittle Power. And he believes that until the system is redesigned to be more decentralized, it will continue to be susceptible to failure, both accidental and deliberate>

No argument with much of what Lovins has said over years, and theoretically most of publications make sense. Love the design of the Hypercar! Wish they could get it built (for a price I could afford). At least he declares his shareholdings in the concepts promoted up front.

Anyway.

 

[SidiropoulosM]

Computer and software problems in the control room of Ohio utility FirstEnergy may have been a major cause of the Northeast power blackout on Aug. 14, according to congressional testimony and documents just released.

Visit:

http://www.energycentral.com/sections/news/nw_article.cfm?id=4118550

 

[davefitz]

So, where do we go from here?

1. Can a repeat be avoided if the distributiion system has installed upgraded technology? Can FACTS systems reduce the chances of another blackout?

2. Newer units in developing countries are designed to allow island operation with quick reload back to former loads , by use of turbine bypass systems and upgraded controls. This could have kept the large coal fired and mukes operating ( likely require fast runbacks to 70% - load) , but maybe some areas would have recovered within minutes.

3. UPgraded solid state components at substations has allowed reducing the design margins in transmission systems. Has the reduced margin contributed to the failure?

4. It is my understanding that most transmission lines were licensed to carry a specified nominal max load, but that this is routinely exceeded. Is this practice of &quot;winking&quot; at the max permitted loads part of the &quot;human errors&quot; that represent a systemic error ?

 

[lengould]

The article on the blackout by Thomas Casten at energypulse definitely strikes a note with me.

http://www.energypulse.net/centers/article/article_display.cfm?a_id=457

you may need to register (free) if not already

He presents a case for distributed generation in an interesting manner, but IMHO ignores issues of linesman safety, right of access for less priveleged, long-term fuel sources, etc. etc. These well-known hurdles should have been mentioned.