Re: [Phys-l] Nuclear Power and the Grid

[John Denker <jsd@av8n.com>]

On 09/10/2011 05:25 PM, Spinozalens@aol.com wrote:

> To try 
> to  keep the plant running  under a blackout condition is crazy.

That wasn't what I was suggesting.

I was talking about something along the lines of the reactor core
isolation cooling system (RCIC), which does useful things with the
residual decay heat.  If it is so crazy, why did GE provide such a 
system in the Mark I reactor design, as used at Fukushima and many 
other places?
  http://www.google.com/search?q=ge+"mark+i"+rcic

> This is not right.  Station blackouts are classified as one of  the most 
> likely events.  Nuclear plants are designed to handle a station  blackout and 
> have done so numerous times without incident.

Perhaps I wasn't clear about the terminology.  Please do not confuse
the following:
 -- A "station blackout" is where the station itself, including the 
  control room and the coolant pumping systems, are without power.  
  This is not common, and the US NRC and others have known for years
  that many scenarios involving a prolonged station blackout end in
  disaster.
 -- A regional blackout is a much less dire situation.  It means that
  the plant cannot obtain power from the grid, but the control room and
  the pumps et cetera can operate from the on-site diesel generators.
  We agree this has happened numerous times without leading to disaster.
  OTOH this is *not* what the NRC calls a station blackout, and not
  what I was talking about.

>  The situation is  Japan was 
> out side the design analysis and its very justifiably to  criticize this 
> failing, that's never supposed to happen. Also the early GE  containments  has 
> long been known to be problematic, that's another very  justifiable criticism. 
> But your general statement is just plain wrong.

My general statement was that the plants weren't very well designed.
You agree that the design is in several ways "problematic" but disagree
with the general statement?   What's wrong with this picture?  Even if
I'm wrong about the number of MW required to operate the station, I stand
by the general statement.  These plants weren't all that well designed.

=============

As another bit of evidence supporting that general statement, consider all
the upgrades that the US NRC has required Mark I plants to install over the
years, at a cost of billions of dollars per plant.  Note that "regulatory
capture" (where the industry controls the regulators and not vice versa) is
even more of a problem in Japan than in the US, and the Japanese regulators
did not require the same level of upgrades, leaving it to the discretion of
the operating companies, since "they were in the best position to know" what
needed doing.  At least one of these upgrades, namely hardening the vent
stack, by all accounts would have prevented at least one of the explosions
at Fukushima.  This should be an obvious lesson for those who are demanding
"deregulation" in the US ... as if the Deepwater Horizon were not lesson
enough.

As yet another example supporting my general statement:  There are conflicting
stories about this, but the impression I get is that at Fukushima, after the
diesel generators failed, operators ran the coolant pumps from batteries for
several hours.  It was only after the batteries were exhausted that cooling
stopped and losses of many trillion yen were incurred.

At this point the question arises, what about the RCIC?  The story is that
without battery power, they were unable to control the valves to run the
RCIC.  If any part of this story is true, this is some verrry badly designed 
hardware *and* some verrry badly designed operating procedures.  It boggles 
the mind.

 -- The RCIC hardware should have been fail-safe.  It should have been
  operable, somehow, without battery power.

 -- Hours earlier, when it was obvious that it would be days (not hours) 
  before anyone would be able to bring in help from outside, the operators
  should have started husbanding battery power.  They should have switched 
  off the electrically driven pumps and used the battery power to control 
  the RCIC.  Controlling a valve takes a lot less power than driving a pump.
  I don't know for sure, but it seems plausible that this would have increased 
  the battery lifetime by orders of magnitude.  The hypothesis is that they 
  could have operated the RCIC for days, buying them enough time to bring in
  pumps and generators from outside, time to figure out what was going on,
  time to let the residual decay heat in the core taper off, et cetera.
  This could have led to a very different outcome.  

  Note that hooking up external power is a lot easier if you do it /before/ 
  the plant explodes and spreads high level radioactivity all over the site.

Even if we don't understand the details, we know that the RCIC was intended 
to be a last-ditch cooling system, but turned out to be not operable under 
last-ditch conditions.  This is not something to be proud of.