[ExI] solar power satellites

[hkhenson]

At 04:23 PM 5/19/2008, you wrote:
>Thinking about the recent posts on the subject, how small could you 
>make a decent proof of concept? I mean, the figures given on this 
>list are 10,000 metric tonnes for a 4GW satellite beaming microwaves 
>to a rectenna 1km x 1.4km.

Bigger.  5,000,000 kw x 4 sq m/kw is 20,000,000 sq meters.  that's 
more like 5,000 meters in diameter, and that does not count the 
gaussian distribution.  For a 1k transmitting antenna and 2.45 GHz 
the usual size of the ground stations at 45 degree latitude is about 
10 x 14 km.

>Now the ARES V launcher proposed by NASA will lift 130 metric tonnes 
>to LEO. Would it be possible to make a one-hundredth scale satellite 
>and tow it to GEO on an ARES V?

Unfortunately, no.  And the lift from LEO to GEO takes about 70% of 
the mass in LEO to get it up to GEO.  So 130 tons in LEO would only 
get you 39 tons in GEO.  It's a big step, it's as much effort to get 
to lunar orbit as it it to get to GEO.  I have spent the last few 
days immersed in orbital mechanics.  I couldn't find my 50 year old 
physics text books but hhere is a ton of stuff about it on the Wikipedia.

>Coupled with the development of a mobile, easily erected 
>one-hundredth scale rectenna (100m x 140m for 1% area) you would 
>have a proof of concept that could be used to supply 40MW wherever 
>it was needed. Seeing as the most recent official interest in solar 
>power satellites came from a US department of defense study, it 
>might be possible to get this proving mission paid for by the DoD.
>  Does this sound like a worthwhile step in advancing the technology?

Unfortunately no.  A small system can't focus the microwaves into an 
area where they can be picked up in useful amounts.  It's a pure 
optics problem.

Fortunately, this is physics that's been understood for 200 
years.  So there isn't much risk in just doing it.  But unless we can 
come up with a way to get the cost to GEO down to a good deal less 
than $100 a kg, there isn't any point in doing it at all.  It's an 
interesting design to cost problem.

Keith