The space train concept involves a ship moving between 2 bodies such as earth/mars with a constant speed and without slowing down. As it flies by each body it orbits around it and continues to the next body, in a constant loop.

Assuming this were the case how would such a ship release and pickup cargo?

I imagine it would release cargo as it orbited one body, but that cargo would still have momentum and would have to do many orbits around the body to slow down before it could connect/be captured by an orbital station or space elevator.

What about picking up cargo?

Assuming the cargo was in orbit at the precise location the train was flying, using magnetic connectors timed perfectly what would happen to the momentum of the ship if a large mass that was moving slowly compared to the ship was connect to the ship that is moving at high speed?

  • 6
    $\begingroup$ This type of orbit is called a 'cycler'. See e.g. en.wikipedia.org/wiki/Mars_cycler $\endgroup$
    – Hobbes
    Commented Jun 25, 2018 at 12:07
  • 6
    $\begingroup$ It's not really possibly to maintain a 'constant speed' when falling and rising in gravity wells - such as when approaching or departing a planet. See this question for more information about cyclers. Also, any cargo you release without propulsion would essentially continue along the space train's original trajectory $\endgroup$
    – Jack
    Commented Jun 25, 2018 at 12:11
  • 6
    $\begingroup$ You should read Seveneves by Neal Stephenson. It posits some very interesting space-travel-related superstructures that I think would be very relevant to this question. $\endgroup$
    – Tin Wizard
    Commented Jun 25, 2018 at 17:30

3 Answers 3


This whole concept breaks at the $^2$ of $E = {1 \over 2} m v^2$.

You must realize what speeds are involved, and what energies appear as result.

1kg of TNT has a little over 4 megajoules of energy - as chemical energy of its explosion.

1kg of pretty damn much any material in Low Earth orbit has 32 megajoules of energy.

That means a kilogram block of, say, wood, hitting a thick surface at LEO's 8km/s will cause a crash that will behave as an explosion 8 times stronger than a kilogram of TNT detonated at rest on that surface.

So, picking payloads like that, dropping them off, and using the raw kinetic energy of the train in that process just won't work. You park your payload in LEO (traveling at 8km/s), then the train arrives at 11.3km/s from Mars, and tries to grab the payload, with 3.3km/s speed difference. 3.3km/s speed difference is 6 megajoules/kg. Instead of grab-and-accelerate you have a crash of energy equivalent of roughly 1.5 times the payload weight in TNT. Nothing will survive that.

What you CAN do though, is to launch a nice, comfortable ship with life support systems, artificial gravity, plenty of storage space, good shielding, generally a comfy long-term 'space hotel.' Put it in the cyclic orbit. Then, when it comes, launch a supply ship - just a dumb tube stuffed to the brim with food and necessities; and a passenger capsule - a small, rugged, cramped thing that has enough oxygen and food for a couple days - then using the rocket engines match trajectory, dock them to the cycler, and board it. Use the supplies, facilities, have the crew live in relative comfort on the way to Mars, and then have them board a reentry vehicle, another rugged capsule, that will be able to land on Mars. Cycler goes back to Earth, the capsule brakes against Mars atmosphere and lands at a base. Possibly a different crew launches in an ascent vehicle to board the cycler on the way back to Earth.

The cycler would never enter orbit of Mars or Earth - it would only do fly-bys. It would hardly accelerate or decelerate. It would just serve as a comfortable mobile base. No energy savings on accelerating/decelerating the payload, but instead of a cramped capsule to Mars, you get a space hotel to Mars.

  • $\begingroup$ I've reformatted the question only by adding line breaks; the OP's question was already well-structured. It might be helpful to adjust this a little bit to match the OP's structure. $\endgroup$
    – uhoh
    Commented Jun 25, 2018 at 14:49
  • $\begingroup$ Nice answer, seems pretty feasible too. Maybe in the future, the transportation to Mars will not really be a spaceship, but more of a modular space station built piece by piece as the demand grows. $\endgroup$
    – JohnEye
    Commented Jun 25, 2018 at 17:19
  • 7
    $\begingroup$ The key to this is that the design constraints for atmosphere-capable ships are entirely different than those for the "long haul" between planetary orbits. The Cycler should in fact be large enough to have multiple greenhouse modules, any one of which can fully recycle all human waste back into food, clean water, and O2. It needs to be capable of sustaining life indefinitely without any resupply. $\endgroup$ Commented Jun 25, 2018 at 18:27
  • 1
    $\begingroup$ One thing to note is that cycler transfers take much longer than conventional (Hohmann) transfer orbits. $\endgroup$
    – Skyler
    Commented Jun 25, 2018 at 18:40
  • $\begingroup$ This is what I immediately thought--except instead of sporadic trips by different people, it would probably be permanently operated by a team (or the team would only change every few trips), and the docking modules would only need to carry cargo. $\endgroup$
    – user26326
    Commented Jun 25, 2018 at 19:33

Please note, that this answer is currently pure science fiction, and we do not yet have the necessary technology. Therefore this is more of a far future approach.

As SF, wrote the difference in velocity is the problem. My suggestion would be to use the released payload for accelerating the new payload.

This would require two identical structures in earth and mars orbit. The structure would "catch" released payloads (possibly by means of electromagnetic deceleration), buffer the released energy and / or transmit the energy to the new payload for acceleration. This process could not be loss-free, so at some point some additional energy has to be added.

However, this system would require the released and new payload to be roughly the same weight.

Using the numbers provided by SF and allowing a temporarily acceptable acceleration/deceleration of ~2g (non-human payload could possibly bear more) the structure would need to be 600km long (300km for deceleration and 300km for acceleration).

If the deceleration and acceleration tube are build end-to-end the accelerated payload would have roughly the same position relative to the train, as the decelerated one had.

Another possible problem would be the trains schedule, since the relative position of earth and mars shift within the solar system, a regular schedule might be hard to maintain.


This is not going to be feasible any time soon, however it could theoretically be done with momentum exchange tethers. Your ship will have to consist of at least two parts: one heavy “counterweight” and one “satellite” part, connected by a long, strong tether so the parts can orbit around their common center of mass (similarly to a planet-moon system, except the force is not gravity but mechanical tension). If you choose the angular velocity in a suitable way, then the velocity of the satellite will cancel out the relative velocity between Earth and the transit orbit, i.e. if you attach yourself at the right time you need at least somewhat less $\Delta v$ than you would for getting to the transit velocity right away. You should probably take care to always drop the same mass you attach, simultaneously, so the transfer doesn't throw off the ship's trajectory.

As a bonus, you would have artificial gravity due to the centrifugal force.

As a penalty, you would have a huge amount of artificial gravity and centrifugal force that the tethers need to withstand. And you will have very little time for performing a rendezvous that's demanding in both precision and mechanical force.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.