Robatoy wrote:
....

I keep trying to reconcile the what we know about nuclear powered
satellites and the size of the behemoths we seem bent on building on
the ground. Even the small units that power subs and aircraft
carriers. Why do these power plants always have to be so big and
unwieldy? I don't want to go as far as suggesting 'Neighbourhood Black
Power Boxes' but....(I understand there would be security issues but
that is not why the big nukes are as big as they are.)

Satellites are mostly isotopic (Pu-238) decay heat powered
thermoelectric generators. The US has only had one experimental fission
reactor launched and that was ages ago while the Russians have used
quite a few altho I don't know just how recently.

Hmmm....this seems to be a fairly good article altho I didn't read it
carefully, skimming looks reasonable---

http://www.eoearth.org/article/Nuclear_reactors_for_space

The primary reason for the size differential is the space reactors are
quite low power devices in terms of central generation requirements
(otoo 2-3 to 100-200 kw instead of 1000 MW). Also they don't require
much in the way of shielding onboard as there is no manned payload.
There's sufficient shielding in a commercial design that one can be in
containment but outside the biological shield area during operation even
though that is a rare event not done in normal operation as there is no
need for access there. We did do incore physics tests using
manually-controlled drives to insert probes in the calibration ports of
the fixed incore SPNDs (Rh-emitter self-powered neutron detectors)
during initial physics testing and follow-up at Oconee I to provide
verification data for the physics models and instrumentation to the NRC
for final approval of the design models back in the mid-70s. It was a
100F+/80%RH hellhole in the bottom of the incore termination tank and
miserable suited up but we did it. The thought that there was
2250psi/650F water just on the other side of a 1/2" diameter tube w/
only an end cap and weld was unnerving to say the least...

And, just like the boiler in a 1000 MWe coal-fired unit isn't all that
large, the reactor vessel containing the reactor core itself is only
roughly 25-ft tall (about twice the height of the fuel) and 12-15 ft in
diameter. All the rest is ancillary equipment. The reason containment
buildings are the size they are is that they must be large enough to
allow for adequate maneuverability of equipment inside and have ample
volume such that the design overpressure of a design LOCA is within the
ability of the containment to withstand. Years ago circle-W designed a
set of reactors w/ ice containment (a huge rack lining the upper reaches
of containment w/ blocks of ice and the ancillary ice-making equipment).
This did allow them to reduce the initial capital cost by making the
containment significantly smaller since the ice-melt during LOCA would
quench the steam, thereby holding down the maximum pressure but these
turned out to be high maintenance items and afaik the concept has been
dropped in current generation designs on the docket for licensing now.

--