Space Exploration Stack Exchange is a question and answer site for spacecraft operators, scientists, engineers, and enthusiasts. It only takes a minute to sign up.
Sign up to join this community
DARPA has proposed orbital maintenance and salvage robots. I had gathered a list of links that was lost when DARPA rearranged their website. But I was able to find this page: Phoenix Makes Strides in Orbital Robotics and Satellite Architecture Research.
Suppose someone successfully designs an orbital maintenance and salvage robot that can rendezvous with many satellites.
What parts of a dead satellite can be salvaged?
It really depends on what stage the satellite is in.
If the satellite still functions at a reduced capacity, then likely everything except the batteries, and fuel, is salvageable. Some of the transmitters, instruments, etc could also work, although it really depends on why the satellite failed.
If the satellite isn't functioning, then some components may be working. These components would need to be tested. The heater may not be functioning, which might make the components less likely to work after being salvaged. In such a case, the fuel, solar panels, housing, and possibly others might be salvageable, but it would be more difficult. More likely everything except the listed items would best be used as scrap material, and without manufacturing on orbit would be difficult.
Solar panels degrade at a rate of about 2-3% per year. After 20 years, they will be about half of what they were when made. There might be some use in collecting them, but overall they will have issues due to radiation, thermal, and debris degradation.
The bottom line is, the best thing is to repair working satellites, as opposed to putting two or more together. You might get some commonality for satellites of the same design, but even then, it might be difficult. I'd say the best bet is components of satellites of the same design, fuel, power generation, possibly batteries, and scrap metal is all that would really be reusable in a salvage situation for now.
A few things I come to think of:
Heat radiators for cooling a power source.
Physical structure needed in-space but not during launch. Maybe for example such needed to spin a spacecraft for simulated gravity. (Once that kind of stuff has been standardized)
A memory chip with petabytes of science and engineering data stored on it, to be collected and soft landed on Earth. I think this is very doable, data has low mass and travels lightly through space. Even in "cartridges".
Equipment, such as a large radio parabola and sensitive instruments. Like taking the optical mirror off an old space telescope and putting it on a new Kind of a Hubble servicing mission++.
Espionage! Since it is about DARPA, I suspect that they want to bring parts of enemy spy satellites and such back to their laboratories on Earth. To assess and reverse engineer whatever the foe has done.
Finally, Apollo 12 brought back parts of a previous robotic Surveyor lander to Earth. Just to examine how it survived up there.
Versions of that above could maybe be useful to do in-space instead of being brought back to Earth as has yet been the prefered way to do it AFAIK.
Even if everything is dead and cannot be repaired or salvaged.
A momentum exchange tether could be used to throw it into the atmosphere and applying momentum to something useful.
Mass is the costly thing to accelerate to orbital velocity, so it could easily be cheaper to put up a droid and some new parts to inhabit the existing shell. This is pretty much the hermit crab approach.
Consider the size of historic satellites; an iridium satellite is 689 kg of stuff - that is a lot of mass. A small satellite could make use of the basic shell of a lump of matter like that for protection. Historically, small satellites have been cheap to launch because they can be loaded into spare capacity on existing launches, whereas something like an iridium satellite would bear the main cost of a launch.
It would be likely that a lot of subsystems would still work; there might be some propulsion still available, gyros and so on which are vital for certain types of mission.
A particularly ruthless scavenger could simply cut up the existing satellite body and use it as a reaction mass; some solar panels, capacitors and some sputtering could provide a usable railgun and by alternating shots toward Earth and away it could propel itself without adding to the debris cloud.
As the other question responses have hinted, it depends upon what you mean by "can be salvaged". This field, robotic intervention, is a mixture of bright ideas, immature technology and poor but growing cost effectiveness.
The technically "easiest" salvage and most readily exploitable commercially would be the whole satellite. This idea has been around for a decade or more for the limited situation of a satellite that has residual function in terms of power, command access etc but which has, or will soon, run out of propellant.
The first step in salvage is rendezvous and docking. If the salvager has attitude control authority over the target then it can point it towards the Earth or move it to another location to give it a new lease of life. This has already been considered to be the end goal in a number of studies.
The more adventurous steps of salvage, which has included repair, refueling, or now in the DARPA Phoenix project, dismantling for re-use elsewhere will all need some additional robotic capability over and above the initial rendezous and docking.
Thus all the ideas suggested in other responses and comments could be workable, just depending on the level of robotic sophistication and price, both unknown at the moment. I've heard the antenna removal option specifically in conjunction with the DARPA programme recently and suspect it relates to reuse of large mesh antennas though I haven't yet found any references to flesh it out.
