[AndyJackson]

The power-grid by definition is a more reliable source of power than any single generator on a particular plant site. Your statement flies in the face of over 40 years of commercial nuclear power licensing and design. Any you are preaching to a degreed nuclear engineer from RPI who has worked in and around commercial nuclear power for more than 25 years -- since 1991 on risk assessment projects

Well, if you read the reports, it was the grid that failed first. Then one of the backup diesel generators. I spend a lot of time watching these kinds of safey analyses. Your "by definition" a priori assumptions are a postiori proven wrong. It isn't a small issue, actually. The reactor plant is designed to be many 9's more reliable than anything else. It is not hard to understand that it turns out to be so.

PS. A lot of folks suggest that had at least one of the Fukushima reactors continued operating, providing power for cooling they would not be where they are today. That does depend upon whether there was actual real safety significant damage to the system before the loss of cooling problems developed. It will be interesting to see where it comes out in the end.

[sefarkas]

The reactor plant is designed to be many 9's more reliable than anything else. It is not hard to understand that it turns out to be so.

The reactor plant is only part of the story at a commercial nuclear power plant. There are numerous single failure points in the combination of systems that convert the steam from the NSSS into MWe on the grid. Furthermore, the "9s" more reliable nature of the NSSS and its ECCS are designed to mitigate accidents and contain radiation. The reliability of individual important pieces of equipment range from one-in-one-hundred for equipment that is operated "manually" to one-in-ten-thousand for equipment that operates passively, e.g., a check-valve. Your "9s" statement is out of context to the issue of a reliable source of AC power.

As for Fukushima, the reason that one plant was not reconfigured to provide AC power to the rest of the units is because

• (1) the load the other plants would provide is too small for the turbine/generator set -- commercial reactors are not sending steam to the turbine until they are at roughly 15% full power because of control issues with the feedwater regulating valves. Unit #6 is rated at 1100MWe -- 15% is 165Mwe. House load at a single shutdown NPP is at most 30Mwe mostly depending on the number of circ water pumps running; none at Fukushima.
• (2) the transformers needed to convert 22kV from Unit #6 into 4160V on the safety buses of the four damaged plants are located outside of the turbine building -- swamped with sea water. Only a journalist would consider what you suggest to be actually practical.
• (3) the grid failed first is inaccurate. The offsite power to the on-site safety buses switch gear at North Anna tripped open. The connection to the power grid was lost. At any utility run power plant, the relays are set to protect the expensive and lengthy time for either GE or Siemens to replace high-voltage equipment. The power grid was supplying AC power to the region during and after the earthquake on Tuesday, 23Aug11. Even at Fukushima, the problem was bring power grid AC to the plant switchyard rather than a wholesale power grid blackout. Loss-of-offsite power is one of the most anticipated events at a nuclear power plant; thus, the heavy emphasis on back-up alternative AC sources typically filled by marine diesel generators on-site.