# Satellite salvaging: safely de-orbit and retrieve spacecraft or other objects

(This question is not about getting scrap metal out of orbit and recycling it: please see the second and third sections of the question text.)

Consider an object which was not designed for, or has lost its capacity to perform, atmospheric reentry (for instance: a space station or part of it, a spacecraft with failed propulsion), in orbit around a planet. Suppose now that the residents of the planet’s surface decide that it is desirable to have the object be brought, (mostly) intact, to the surface.

What would seem like a reasonable way to achieve this?

-

The relevant setting is near-future-ish, but post-cataclysm: the predecessors of the surface dwellers had the capability to deploy extensive infrastructure in space, and performed interplanetary travel. The technology presently available to the recovering civilisation on the surface is recognisable as modern day, maybe plus a couple of pieces of handwavium gadgetry. The surface dwellers are not capable of reproducing the Magical Space Drive™ that enabled commonplace spaceflight for the predecessors, but they would know enough, for instance, to repurpose a spacecraft engine retrieved from orbit to be used as a power source.

The constraint on technology level is important, since it would not make sense if the surface dwellers could, at a lower cost, instead manufacture things on the surface that serve the same purpose as the salvage. This rules out sending a re-entry-capable cargo ship up, loading the stuff in, then flying it back down. However, it is also required that they be able to send things to a sufficient altitude at least momentarily, for them to be able to carry out the salvage operation in the first place.

The closest thing I could come up with was for someone to launch themselves into orbit, bringing with them thruster modules (perhaps plain chemical rockets) that would then be attached to the target, and be used to decelerate it and commit it to deorbit. However, I know of no robust way of modifying arbitrary satellites to enable them to survive re-entry.

Additional information that might be relevant:

• The operation is routine, but does not need to be indefinitely sustainable. If necessary, the surface dwellers might make use of technological relics found on the surface: for example, they expend one Magical Space Drive to send a crew up, expecting them being able to bring back at least two Magical Space Drives. Alternatively, they might make use of any infrastructure put in place by the predecessors that has survived the cataclysm and the subsequent lack of maintenance and remains sufficiently operable.

• It is preferred for the planet itself to be as similar to Earth as possible, but if the bottleneck to this question turns out to be planetary mass, atmospheric composition, or something along those lines, those parameters may be changed.

-

I’m aware of several questions dealing with related subjects, but have been unable to piece from this information a complete solution to the present question. Links are left here for reference, and for the purpose of highlighting the differences of this question from similar ones.

• How to efficiently deorbit space junk has some creative solutions for de-orbiting without requiring much futuristic technology; however, the question deals only with the matter of removing space junk from orbit, allowing it to burn in the descent. The present question requires that the de-orbited object to be returned to the surface mostly intact.

• Would ablation be an effective way to redirect objects in space? examines an alternative way to impart delta-V, but similarly does not treat the issue of re-entry.

• How can I catch an asteroid asks the same question, except aiming to capture asteroids instead of artificial satellites. Several solutions involve destroying the asteroid beforehand for ease of transport, which is undesirable here where the objects of interest might be delicate technological artifacts. Other answers suggest use of a space elevator, which would be unavailable to the technology level required by this question. The option of using an elevator built by the predecessors leaves the questions of maintenance and operation of a space elevator without the capability of building one.

• How to Effectively Collect and Recycle Space Junk deals with the collection and recycling of objects in orbit, but is performed near a body without an atmosphere and explicitly forbids de-orbiting. The level of technology considered is also far beyond the constraints of this question.

There is also an xkcd post that I’m sure is bound to be brought up at some point.

• Possible duplicate of How to Effectively Collect and Recycle Space Junk? – sphennings Dec 9 '17 at 20:31
• @sphennings From the other question "It is also important for me to avoid deorbiting." This question is explicitly about de-orbiting, as seen in the title. Therefore, not a duplicate. – kingledion Dec 9 '17 at 21:17
• @sphennings the post has now been edited to explicitly specify why several similar questions do not provide adequate answers to this one. – Lok Dec 10 '17 at 2:34
• The real issue there is no system they could build that would not use more resources than they could recover. You're are talking about a massive expenditure of fuel and resources to retrieve a few kilograms of aluminum. That same effort put into deeper or low purity mining would give far more return. You may want to rethink why they are doing this. – John Dec 10 '17 at 14:52
• @John see the part in the question text about the Magic Space Drives, and about repurposing retrieved technology. They can't manufacture handwavium engines, but know how to use them if they could get their hands on one of the many relics lying in orbit. This is why the question places emphasis on retrieving satellites in one piece, instead of as chunks of aluminium. – Lok Dec 10 '17 at 16:42

Astronautical engineer here, I'll try to take a shot at this.

As you should probably know from the related questions you linked, capturing and de-orbiting uncontrollable derelict satellites is extremely difficult and costly, and has been a subject of research and speculation since the late 1960s. To get them down to the surface in one piece, you have to contend with the following factors:

• Acquisition of the target to be salvaged, which requires the ability to match and change orbits and grapple with the target
• The intense heat of reentry, which is due to the fact that orbital speed is roughly 17,500 mph, whereas subsonic flight in atmosphere is between 680 and 750 mph, depending on altitude
• The dynamic forces of reentry, including aerodynamic buffeting and deceleration loads
• Vehicle control during reentry, which is essential to both hitting your target landing zone and preventing the vehicle from tumbling out of control and breaking apart under the dynamic forces mentioned above
• Landing site selection; landing in the ocean is easier but requires extensive support equipment and may damage your vehicle and salvage, landing on land requires much finer control and a much slower final speed

Capture of orbital debris is a hideously complex science, and whole PhD papers have been written on the subject. But in simple terms, you need a way to catch the thing you want to salvage, stabilize it in some way so you can grab it, and then secure it for reentry. So your vehicle needs a supply of fuel and a full set of maneuvering thrusters to be able to change its orbit to that of the target and, once the target is captured, move back to the return orbit (which may or may not be the same as the launch orbit). Grappling is a real problem, as derelict satellites might be spinning out of control at relatively high rates. Your capture vehicle will probably need some kind of sensors to determine the axis of rotation, and then will need to maneuver along that axis to match the spin and grab it with a net or robotic manipulator. But then you have to slow the thing down, without hitting it or losing control yourself, which is no simple matter. Your maneuvering thrusters and/or reaction wheels will need very fine control, and your guidance system will need to be fairly clever. This isn't something that a human could do intuitively with a reasonable degree of safety.

During reentry, the vehicle is traveling so fast that the friction between its skin and the air dissociates the air molecules, forming a hot plasma. This both disrupts radio communications with the vehicle and imparts tremendous heat. However, handling the heat of reentry is actually not as big a deal as it first seems. The most common method is to use an ablative heat shield, typically ceramic or some Inconel alloy, to protect your vehicle. These are usually blunt by necessity for stability purposes, and can be very large (e.g. the Space Shuttle's entire belly was one giant heat shield). However, these things are difficult to make and, as you might imagine, there is no margin for quality errors. You can potentially reduce the amount of heat generated on reentry by using a hypersonic parachute or retropropulsion to kill most of your speed before you hit thick atmosphere, but these technologies don't completely eliminate the problem.

The act of using air resistance or retropropulsion to slow yourself down means you're in for a bumpy ride. You need to bleed off roughly 17,000 mph of speed in as short a time as possible (to avoid burning up), so you need to decelerate at between 1 and 5 times the acceleration of gravity, depending on your trajectory. Reentry also imparts a lot of buffet to the vehicle, which means very high-intensity vibrations. Most spacecraft are only able to handle high accelerations and vibrations in particular directions and in particular configurations -for example, with the solar arrays and antennas stowed in the folded position. You'll either have to return the vehicle to its launch configuration, or provide a means of cushioning it against vibration and acceleration.

Your descent vehicle will need a particular set of aerodynamic characteristics to remain stable during reentry. This could take the form of aerodynamic control surfaces like the grid fins on the Falcon 9, be derived purely from the shape of your vehicle, as with the Apollo capsules, or involve compressed gas thrusters. You also need a guidance system that can operate in isolation from ground control or any visual input; typically an inertial system using gyroscopes and/or accelerometers.

Once you're subsonic, parachutes are the simplest method for slowing down to the point of impact. If you're aiming for a water landing, you need some floatation devices to keep from sinking, but otherwise, you're done. If you want to land on solid ground, however, you'll need to deploy some kind of landing legs or gear (and may have to jettison your heat shield to do so) and you may need larger parachutes or some low-thrust retropropulsion to give a softer landing.

Given the above considerations, and the fact that this reentry vehicle has to be reusable, and the point is to recover random satellites for recycling, here are some suggestions that would make the job much easier for your salvagers:

• Consider making the salvage ship stay in orbit, with only the cargo vehicle and crew having to endure launch and reentry forces. This will make launches cheaper and reduce the wear and tear on your salvage ship (and the complexity of landing all that extra mass).
• Make sure the salvage ship has powerful thrusters and reaction wheels, plenty of fuel for big orbit changes, and a smart guidance system.
• Chop the satellite being recovered into small, manageable pieces that can be efficiently packed together. This way you won't have to worry about odd shapes, deployed features like antennas or solar panels, and irregular centers of mass.
• Enclose the salvaged satellite bits in a capsule or cargo hold of your reentry vehicle. This will allow you to control the reentry vehicle shape and center of mass, rather than having to accommodate oddly-shaped salvage, and it'll protect the salvage from the hot plasma during reentry.
• Astronautical engineer here Nice! Not often someone in the actual field is around to answer questions like this! – Draco18s Dec 11 '17 at 0:56