It's lightest just before dark
August 22, 2003 6:52 PM   Subscribe

Utilities apparently should have known that trouble was brewing in Ohio. Kinda makes you wonder if this wasn't just a case of everybody (well, nearly) being asleep at the switch...
posted by clevershark (12 comments total)
The article is SORTA correct. I work at a nuclear power plant and read today a VERY detailed industry report from ITC about what happened and the timeline of events.

First, initial conditions: very hot, mid August. Peak time of afternoon. Davis-Besse nuclear plant on Lake Erie in Ohio was shut down for a maintenance outage. Other coal fired plants offline being worked on.

Basically, when the lines near Cleveland failed, they isolated the transmission lines to protect power plants in Ohio. They then tried to pull power from Michigan and immediately increased the demand in the metro Detroit area by a simple analogy, power which is designed to flow clockwise around Lake Erie began to flow counterclockwise - which very quickly tripped two big power plants in Michigan and took 2000 megawatts off the grid. This caused a sharp undervoltage at about 94% of normal. At one time, one of the main grid lines feeding northern Ohio was SMOKIN' hot...max design was at about 1300 megavolt-amperes during EMERGENCY loads, and this line was carrying over 2000 MVA at one point.

Any time this happens, plant operators get very dicey and start worrying about protecting their billion-dollar investments. Several more power units in Michigan tripped offline to protect themselves, taking another 4000 available megawatts off the grid. When this happened, a SEVERE undervoltage at 89% of normal happened and the catastrophe point was approaching quickly when power was now pulled from Ontario (until this point, it had just been a serious emergency).

At this point, Ontario tried to suck power from New York. the automatic trip units actually worked as designed here, and the grid was isolated from Ontario. However, this created a problem....power was already flowing from New York to Ontario as part of normal operations. Now, New York had TOO much power with nowhere to go, and once again, units began to shut down to protect themselves. This was the straw that broke the camel's back, and the point of no return was now reached.

All in all, I think about 8-9 nuclear plants supplying appx 6500+ megawatts went offline. And, there were another 22 power units (coal fired and nuclear combined, I think? can't remember at this point) that tripped offline.

Oddly enough, for those not in the know, nuclear plants generally MUST have a steady source of supplied offsite power in order to shut down safely. When this power is lost, the units shut down or scram automatically to prevent problems we don't even want to talk about. Emergency diesel generators on site are typically used for very critical operations and shutdown activities only, and cannot sustain the site operations for long under loss of offsite power.

Granted, I wasn't one of the people inconvenienced, but this was an interesting chain of events. I think they estimated that it took appx 39 minutes from the first first event until the time of "point-of-no-return" for a major blackout. Does anyone else here think we're vulnerable to terrorists?
posted by insulglass at 9:34 PM on August 22, 2003

I wish I could mod insulglass up as "informative"
posted by ilsa at 10:06 PM on August 22, 2003

ouch. stavros hits hard. got my sandwich?
posted by quonsar at 11:03 PM on August 22, 2003

wrong thread.
posted by quonsar at 11:06 PM on August 22, 2003

wrong thread, wrong street, wrong lifetime.
posted by xmutex at 12:18 AM on August 23, 2003

I too am a veteran of the US nuclear power generating industry, and the more details I read, the more I believe that for the most part, everything occurred just as it should have. The blackout was minor compared to the potential consequences of failsafes not engaging.

Remember, boys and girls, redundancy is your friend. ;-P
posted by mischief at 2:49 AM on August 23, 2003

Thank you insulglass, that was very informative.

Funny thing is that I was talking with a friend in Cleveland when it happened, and she noted a fluctuation a few minutes prior to actually losing power, which was immediately followed by me losing power here in NYC.
posted by riffola at 8:59 AM on August 23, 2003

So to follow up on what insulglass said...
I work at a non-nuclear (hydro) plant in New York, and we took some shit from a politician because we dropped like a thousand megawatts to keep our selves online. At some point in there, my plant was generating more than half of NY State's power, and had we not dropped that rather large amount of power we would've either a) gone off line completely, or b) damaged our generators.

Without being too specific... we have more than a few generators, powered by water and They work like this:

The water turns the turbine at exactly 120 rpms, I forget the tolerance, but it needs to be really really really close to exactly 120. This is controlled by a gate. This is above the generator, this gate opens and closes, open the gate and more water comes in, close it, and less comes in. More water = more speed.

"But wait" you says to your self "Blake, you just said that it must spin at 120! If you add more water and the speed goes up, then you’re lying!"

"shut up and let me finish" I say...

As the power is drawn out the other end of the generator by the state ISO's (look it up if you want to know what an ISO is, has to do with deregulation) the generator actually slows down. The more power they are pulling out, the slower it turns because the lines are pulling so much power out, and this puts alot more load on the generators, which slows them down. It’s kind of hard to visualize, I know, but it helps if you think of the electricity as something in solid form, rather than just electricity.

Now imagine a generator is pushing out all it can (gates full open, pushing it as hard as possbile) and all of a sudden *blam* all that demand just dries up. The generator was just pushing as hard as it could, and now it's pushing into a wall basically. All that power has no where to go because all those lines just dropped.

So the operators have a few choices, none of them good:
push all that power into no where (drop it)
go offline completely
burn up the generators destroying billions in equipment

So the black out really saved alot of equiptment.

My plant did a good job, we kept the power flowing by dropping what we could at that critical time, and staying on line. Funny thing was, we were getting the blame for a little while by, I think, CNN or some TV station.
posted by Blake at 9:15 AM on August 23, 2003

Oh, and I forgot...
Everyone I know kept emailing, calling, and IMing me asking what was going on, we actually knew little more than what everyone was seeing at the time on TV, so I was telling them this:

When I got hired, I got a tour of the plant. They took me into a room that had a big red button. They said "this is the big red button, never press the red button"
So...around 4 o'clock I was tired, and bored, and I was walking by the red button, and I thought to myself "hmm, I wonder what this does..."
They are coming in tomorrow to install the blue button that will turn everything back on, though I'm not sure I'll be allowed to press it.

posted by Blake at 9:20 AM on August 23, 2003

This is obviously Bush's fault....
posted by WLW at 7:26 PM on August 23, 2003

Confirming followup.
posted by insulglass at 9:02 PM on August 23, 2003

heh! you hit the big red button! you should know better! ;)
kinda like when our hospital execs tell me the network seems faster, asking me what i did.. i tell them i went upstairs and hit the Turbo button :)
posted by shadow45 at 9:59 AM on August 24, 2003

« Older pipes - just the man for the job   |   flying pigs Newer »

This thread has been archived and is closed to new comments