Monday, October 29, 2007

The Case of the Sinking Transformer - Part 1

In all my years as the 'Physics Sherlock' of the big company, I had never run into a case as baffling as the "Leaning Transformer of Pickering".

It all starts with the geology. This was an area which had mixed lake and river deposits. It then got all squished by heavy glaciers, which resulted in a unique mix of hard (over-consolidated) clay, mixed with loose sand lenses. These sand lenses presented the anomaly of being loose, in a matrix of very dense glacial till. The only explanation was that the water in the sand lenses acted as bubbles, which prevented the sand from compacting, as well.

Drilling and excavating in this mess was always a hazard. You pierced one of these sand lenses, and poof! If you were digging, and you hit a sand lens, you would go instantly from a dry hole, to a quicksand mess!

Knowing that this muck could not be excavated, the original designers went for deep piles, which were driven down to 'refusal', which meant that they kept hammering until they didn't go in anymore. Almost everything was put on piles. If they had to dig deep holes for sumps, they installed huge dewatering wells, so they could dig without sinking!

Unfortunately, not everything was put on piles. They decided that if the load was light (such as transformers), then they could get away with just a concrete mat foundation. This was the fateful decision that would come back to haunt them! (not actually the same people, since the 'originals' are mostly dead!).

Thus, we have the main output transformers, which are the huge things that convert turbine power into something suitable to run toasters on. There is only one such transformer per turbine, per reactor, so they are quite important.

(to be continued)

No comments: