[ExI] pentagon wants orbiting solar power stations

[Jordan Hazen]

On Sun, Oct 14, 2007 at 10:19:00PM +0200, Eugen Leitl wrote:
> I'm using fusion right now. It's called PV and solar.

Have you watched your PV system's activity during sudden changes in
cloudcover?  Mine will drop to 10% of full-sun production as a heavy
cloud rolls in, going from net energy export to import within about
five seconds, and reversing just as quickly as the sun breaks through.

This extreme variability is fine so long as solar amounts to only a
small fraction of total grid generation, as it is now.  Spinning
reserves at nearby power plants instantly make up any sudden
shortfalls, keeping frequency and voltage within spec.  But, there are
limits as to how much non-dispatchable generation existing grids can
accomodate.  Getting beyond about 20% renewables (still room for 5-10x
growth), or trying to use PV & Wind for baseload generation will
require very large-scale energy storage-- something better than pumped
hydro and lead-acid batteries (even today, PbA is still the best
option for off-grid sites, sadly).

At least one US utility has been experimenting with large,
megawatt-sized NaS batteries, deployed to substations for
peak-shaving.  These could form part of the solution, perhaps
supplemented by super-cap banks and intellient load shedding, but a
stable grid will probably always need at least a few large,
traditional baseload plants.  In Europe, France's existing fleet of
nukes might fill this role for a while.  Areas with a lot of existing
hydro (northwestern US & Canada, Quebec...) may be able to go 100%
renewable early on.

> Remember, 1/10000th of terrestrial insolation is enough to
> keep current humanity in business indefinitely.

At 100% collection efficiency?  Or the 14% (minus weather effects,
shading, and conversion losses) of current-tech PV?

-- 
Jordan.