There is an interesting paper Published on the Entech site. These are the people that created the fresnel lenses of the solar array on the Deep Space 1.
http://www.entechsolar.com/
"1,000W/Kg Solar Concentrator Arrays for Far-Term space missions"
The paper argues that 500W/Kg will be available by 2011 and 1000W/Kg by 2023.
I find the conclusions conservative, backed by both extrapolation of current trends, and technology that is under development. It assumes, as I do not, that solar arrays will be automatically deployed. My feeling is that robotic, or even manual assembly at the ISS is inevitable by then for deep space missions, and probably even commercial satellites.
What changes when solar collectors reach 500W/Kg? Well, obviously satellites become a lot more powerful, which is going to be great for HD TV and general communications, but there is more. If you take it in conjunction with progress in Ion Drives, lots more. Look here:
http://www.spacetransportation.com/a...df/9a_katz.pdf
Lets do some maths.
Assume a 1000kw array with a mass of 2 tons. Assume a 1000Kw array of ion motors with exhaust velocity of 80km/S, efficiency of 80%, and mass of 1 Ton (figures broadly compatible with the results of the Nexis program). Assume 1 ton for xenon tanks, structure ect. These are conservative assumptions.
The complete structure has a mass of 4 Tons, with a thrust of 10 Newtons. 10N is the equivalent of 1 mm/S2 for a total mass of 10 tons (the additional mass payload and fuel).
1 mm/S2 is turbocharged. It's 1km/s in only 278 hours (half that in near earth orbit due to time coasting in the earth's shadow). A "space tug" would move an assembled satellite from the ISS into synchronous orbit in about 2 months, and return in half that time for the next delivery. Xenon fuel for the round trip would be only about 10% of the satellites mass. To collect for servicing and return will take about 4 months, and 20% of the satellites mass in fuel.
I predict that by 2020 almost all commercial satellites will be assembled, serviced and repaired at the ISS. The advantages are overpowering:
1) Currently Solar Arrays and Antenna must deploy automatically If they were "flat packed" to the ISS and assembled manually, or roboticaly, they would be more reliable, lighter, and easier to develop.
2) Much of the cost of a satellite is design. A DIY kit of modules for assembly at the space station avoids this. The modules are simply clipped onto a standard lightweight frame that never has to handle gravity.
3) There is a good chance that a satellite component will fail during the high-G boost into space. But now this becomes a minor problem. If the assembled kit fails to work, the offending module is replaced from stock, repaired, or a replacement ordered.
4) It's a lot cheaper to boost payloads into low, loosely controlled orbits, than the high synchronous orbits needed by most satellites With the high escape velocity of ion engines, using a space tug to collect the deliveries from low orbit to the ISS, and then deliver the assembled satellite to synchronous orbit is far more fuel efficient.
With such a large commercial interest in earth orbit solar cells for powering satellites, NASA should consider using them for powering missions to mars, and even further afield. It's just so much cheaper to use technology that is going to be developed anyway, than develop nuclear generators that can only be used, at most, in a handful of missions. The inner planets are firmly solar powered territory. The outer planets should only be visited once a generation, at most. It takes at least that long to get funding, design, build, launch, wait for the ship to get to its destination, and finally do the observations. There is no hope of using the same power plant for a follow up mission. The technology would be obsolescent, and the parts unobtainable.
For the planned JIMO miss
