Followup.

> > http://hardware.slashdot.org/story/09/09/01/1656246/Japan-Plans-21B-S...

Sept. 1 (Bloomberg) -- Mitsubishi Electric Corp. and IHI Corp. will
join a 2 trillion yen ($21 billion) Japanese project intending to
build a giant solar-power generator in space within three decades and
beam electricity to earth.

A research group representing 16 companies, including Mitsubishi Heavy
Industries Ltd., will spend four years developing technology to send
electricity without cables in the form of microwaves, according to a
statement on the trade ministry's Web site today.

http://www.bloomberg.com/apps/news?pid=20601101&sid=aJ529lsdk9HI

http://www.bloomberg.com/apps/news?pid=20601080&sid=aF3XI.TvlsJk

I responded on Slashdot. Here is a copy with additions/deletions.

*************

Tekfactory put his finger squarely on the problem. $500/pound is
close enough to $1000/kg and that is ten times to high for space
based solar power to undercut fossil fuels. The Japanese recognize this.

"Transporting panels to the solar station 36,000 kilometers above the
earth's surface will be prohibitively costly, so Japan has to figure
out a way to slash expenses to make the solar station commercially
viable, said Hiroshi Yoshida, Chief Executive Officer of Excalibur
KK, a Tokyo-based space and defense-policy consulting company. "These
expenses need to be lowered to a hundredth of current estimates,"
Yoshida said by phone from Tokyo.

I get the same number close enough. Current price to GEO $20,000/kg;
required for space based solar power to displace fossils by being
substantially less expensive (1-2 cents per kWh) is $100/kg, a factor of 200.

Design to cost. Start with the rocket equation:

Needed 100 t/hr to GEO, $100/kg. Try a two stage to GEO. Requires 14
km/sec, get the first 4 km/sec with a mass ratio 3 hydrogen/oxygen
rocket. Four km/sec is easy to do, ask Elon Musk. To get the
remaining 10 km/sec with a mass ratio 2 means an average exhaust
velocity of 15km/sec.

Because you stage far short of LEO, the second stage must have
relatively high thrust so ion engines won't do. Ablation laser
propulsion (well understood physics) with an average exhaust velocity
of 15 km/sec will provide over a g at 4 GW. The suborbital path keeps
the second stage out of the atmosphere long enough (15 minutes) for
the laser to push the second stage into geosynchronous transfer orbit.

At 4 payloads an hour (working the laser full time), each payload to
GEO needs to be 25 t. So the mass ratio 2 laser stage is 50 t, the
first stage 50 t (16%structure) and 200 t propellant. On takeoff it
masses 300 tons, less than a 747. A large airport handles a lot more
traffic than eight 747 takeoffs and landings an hour.

Hard engineering and not cheap. The laser might eventually cost $40
billion. To get started (to positive cash flow) came out to $60
billion on a first cut proforma analysis.

A UK company, Reaction Engines, has an inordinately clever approach
to boost the effective exhaust velocity so as to put payloads (12
tons) into LEO with hydrogen/oxygen single stage to orbit. What they
are doing is recovering a lot of the energy that goes into liquefying
hydrogen and using that to cool and compress air to rocket chamber
pressures up to 26km and Mach 5+. Google for them. Also see
http://www.theoildrum.com/node/5485 if you want more details.

Added new thoughts from the space elevator conference.

A lot of the mass of a thermal power satellite is heat sink
fluid. That can be made from finely ground rock and a little
gas. Decouples gas pressure from the amount of heat the pseudo fluid
can carry. Seems a shame to be shipping up sacks of cement dust. We
are looking into the payback time for a moving cable space elevator
through L1 to the lunar surface. Existing materials are good enough
for the cable--without taper. 15 MW is enough to lift 33 tons per
hour. Feed lunar dirt through a vibratory ball mill and presto heat
sink fluid.