Part of the fun of this reading-between-the-lines and informed speculation business is seeing how right or wrong one is as hard facts start coming out. One of my speculations, that the HPSI pumps didn’t start was wrong.
The utility has supplied the IAEA with official INES notices of the events and these contain some hard data describing the first few hours of the evolution. As reported here , the HPSI pumps DID start and did operate properly until the water in the torus became too hot.
When the water in the torus approaches the boiling point, a couple of things happen.
- The steam discharge from the HPSI pump is no longer condensed and simply builds up pressure in the torus and primary containment.
- The HPSI pump, which has to have a certain net positive suction head (NPSH) starts to cavitate and lose pumping efficiency. At some point when the water temperature reaches nearly 100 deg C, it simply can’t pump. It’s a multi-thousand HP pump no larger than a compact car and so the specific energy transfer to the water is very high.
A long term total station blackout accompanied by flooding is beyond the design basis of the plant. The design basis assumes a functioning RHR system that removes heat from the torus water. Without the RHR, the torus water simply heated to the boiling point at which point the HPSI and LPSI pumps became ineffective.
Speculation alert: I imagine that if RHR circulation can be restored to the torus, probably by external pumps, the LPSI pumps should restart and resume their feeds to the reactors. This would be MUCH better than actually pumping seawater directly into the reactor vessels. I suspect that this is what they’re doing.
More updates as information becomes available