When a launcher's upper stage fails to put a commercial communication satellite in geostationary orbit, it cannot be used as intended. But could it be used in satellite-to-satellite communication instead even if orbits at much lower altitude? For example to enhance the capacity of the Tracking and Data Relay satellite network which is used to communicate with the space station among other things. Or is it impractical to change its pointings multiple times per orbit or do they lose much of their solar power or get damaged from passing through the van Allen Belt?
Possibly, but not likely. Geo satellites are made to work at that constant orbit, and getting it to work without being at the GEO orbit can cause a lot of problems. The radiation isn't the concern, but the various sensors on the spacecraft are meant to work at GEO altitude. Thermal issues are also of concern, as are the batteries.
Some of the tasks could be done, and in fact have been attempted in the past. The most famous example of this was PAS-22, which failed to obtain the correct orbit, was sold to a company by the insurance company, which created a way to swing by the moon to obtain the correct orbit, allowing it to have some service. Other ideas were considered as well, but that made the most sense.
The main reason for sending communication satellites to GEO is that their position is stationary to ground observers.
In case the satellite fails to reach orbit, the ground control usually tries to raise the orbit using on-board fuel, though the life of the satellite may be reduced. For example when PSLV C1 failed to place IRS-ID on orbit, the satellites on-board fuel was used to raise orbit (this was in LEO, though the principle is same).
However, if this is not possible, the mission is usually classified a failure and satellite, in all probability becomes a space junk due to various reasons.
- Satellite to Satellite communication is not possible for practical reasons. It is difficult to change the orientation of the satellite continuously as it uses up on-board fuel, resulting in orbital decay.
- Also, unless the satellite is placed in another stable orbit, tracking is very difficult and communicating becomes very difficult.
- Another issue is that the communications method of the two satellites (or space station for that matter) should be compatible. Satellites are usually designed for communicating with the ground station, not other satellites.
- Even the ground communication is difficult for obvious reasons. A satellite in GEO is expected to be stationary and the ground station is designed with this in mind. In case of Polar satellites, they pass over a set of ground stations at fixed times per day. However, for an orbit somewhere in between, it is difficult to arrange for a set of ground tracking stations to track and communicate with the satellite.
Reduction in orbit altitude doesn't change the solar power available by much. Also, the satellites are in Van Allen belts anyway even inside GEO.
The main problem with failed launches is the failure to get inserted into a stable orbit and ground communication.
In principle a part failed satellite could be repurposed.
The technical obstacles already given are all quite plausible though none are necessarily "game-over" (though recall that the frequency band used by the failed satellite's payload would have to be compatible with the new use which for the satellite-satelite case which could well be the end of the matter).
A further non-technical difficulty is establishing a market for the services against the budget needed for continuing operations. Again, its not an insurmountable challenge.