Wednesday, January 30, 2008

GoD dismisses earthquake scenario

The Government of the Day says there is no earthquake risk. AECL states that an earthquake would only cause a small poof of radioactivity that gives you a free Cat Scan. Who can argue with these grand statements?

AECL doesn't publish it's assumptions for the accident scenario, so we can't tell. I suspect it's rather benign, just as assuming the coolant boils away slowly, thus cooling the rods. GoD simply doesn't believe in earthquakes, period!


MarkCh said...

How likely is a severe earthquake in the Chalk River area, in any given year? Of the set of all severe earthquakes, what fraction would be such that the backup power supply would make a real difference? What does "one in a thousand" mean anyway? .001 per day? per month? per year? It makes a big difference.

It's too bad that the reporters and MPs seem to ignorant even to ask sensible questions. It should be simple to make a back-of-the-envelope question to see which option is likelier to cause more deaths. To justify overriding CNSC, the shutdown would have to be hundreds of times more dangerous than continuing to operate.

Harbles said...

And as a follow up .. If the chance is one in 100 per year and it has been 50 years since the last 'event' are the odds not greater?

Harold Asmis said...

The odds always stay constant. As shown in the Parkfield experiment, if an earthquake is 'overdue', it might mean your estimate was wrong!

I consider that area to have a high hazard for an M6.3, much like the earthquake up the river in 1935. That probably puts the odds at 1 in 100 to 1 in 500.

What happens then, depends on the soil conditions. I think it's likely that the coolant dumps, and the rods fail to drop. That's one big cloud of radiation!

MarkCh said...

In the scenario you expect, would the backup pumps make any difference?

Harold Asmis said...

The backup pumps are for the more benign scenario where an earthquake just causes the lost of public grid power. This would be fairly common, less than 1 in 100, caused by an M6 a hundred klicks away.

With this scenario, the rods are dropped and the heavy water just boils away because it isn't going through the rads. AECL is making a big fuss that almost no radiation is leaked here.

I'm basing my scenario on the 50's incident when a hot rod was pulled without cooling. Now imagine a lot of rods!

Also, I hate it when a beleaguered agency suddenly selectively draws from its large store of 'hidden' studies. This always happened to me when I launched an attack on some ridiculous internal department in the old company.

MarkCh said...

Thanks for the info. It looked to me like they were in the process of replacing non-earthquake-resistant backup pumps for earthquake-resistant ones, which seems like a dumb thing to do if the rest of the design is not earthquake-resistant anyway. Did they actually disconnect the old backup pumps without connecting the new ones? OMG if that is the case.

The impression I have got from your blog is that both AECL and CNSC are seriously incompetent over this issue - certainly not a reassuring thought. Are there any engineers there at all?

PS: I graduated UofT EngSci Aero 86

Harbles said...

Also an earthquake is not the only thing that could cause the power to go out, an ice storm or severe winds or tornado etc.

Harold Asmis said...

As I've written before, the CNSC went through a rapid expansion of toadyism, and has little expertise left. AECL has gone the way of the mature techno-bureaucracy, and has few brains left in charge. Who would want to work there?

Anonymous said...

Not me.