The idea is that the spacecraft would not have a propellant tank on board, but instead would have a huge, spherical propellant tender which the spacecraft could climb around in order to turn. This seems like it would be useful, as the propellant wouldn't have to be turned around with the rest of the ship, however, the addition of the legs and handles to the system would increase the weight. Is this a plausible way for a spacefairing species to get around, or would it be overly complex for what it achieves?

  • $\begingroup$ just exactly what's wrong with (& even how is this really any different from) extant attitude control system? $\endgroup$
    – Pelinore
    Commented Apr 23, 2021 at 16:33
  • $\begingroup$ @Pelinore The current system would have to move the propellant $\endgroup$ Commented Apr 23, 2021 at 17:30
  • $\begingroup$ Both systems move one part around another part // so same thing // either way likely uses precisely the same amount of energy // ergo the question // 'what is the point of your way & how is it qualitatively any different?' $\endgroup$
    – Pelinore
    Commented Apr 23, 2021 at 17:37
  • $\begingroup$ Like in Lovers in a Dangerous Spacetime? $\endgroup$
    – IronEagle
    Commented Apr 23, 2021 at 17:38

4 Answers 4


Doable yes. Practical? Probably not.

Assuming you have a spherical propellant tank filled with a fluid type of fuel, you don't actually need to turn the whole contents of the fuel tank, just the fuel tank itself. So, if your main body is only ~5% of your mass and your tank is ~5% of your mass, and your fuel is ~90% of you mass, then on a rigid bodied ship, you only need to turn ~10% of your total mass because your fuel will mostly retain its original form and orientation.

Furthermore, making tracks on which to crawl around your fuel tank adds weight to your ship so any benefit you get from not needing to turn your whole fuel tank is lost in how much extra fuel you need to spend accelerating it. Any fuel spent turning a ship in space is supper negligible compared to what acceleration costs. Turning in space is rare and is done at very slow speeds. Even turning a ship the size of a city is not that big of a deal as long as you are not trying to do it too quickly. So in the end, a simpler lighter ship design will save you more fuel than not having to turn your whole mass.

One case where something similar to this idea might be useful is for a satellite or space station. If your main body is very massive and you want to keep your mass dedicated to propulsion/fuel as small as possible, then having a single RCS thruster that moves to where it is needed could be cheaper than investing in many thrusters or the extra fuel it would take trying to rotate your whole station to use a single main thruster.

  • 1
    $\begingroup$ As soon as any acceleration is applied to the propellent tank the propellent will settle towards the source of the acceleration regardless of the method by which the acceleration is imparted. $\endgroup$
    – Slarty
    Commented Apr 23, 2021 at 17:41

I don't see what's the advantage of this design.

For most of their travel spaceships simply coast without any propulsion, they don't really drive around like cars looking for a parking spot in downtown on a Friday evening. In those times crawling around would not help.

And for those times when a maneuver is needed, the tank will need to be accelerated anyway, saving nothing. Mind that "steering" in space doesn't happen just because the ship points its nose in a certain direction. It needs to exert a force in a certain direction to alter its course.

  • $\begingroup$ The tender doesn't need to rotate with the rest of the ship, which would save energy when the ship has to turn $\endgroup$ Commented Apr 23, 2021 at 15:05
  • $\begingroup$ @IchthysKing Yes it does, in order to point its jet plume in the correct direction. It just moves itself relative to the fuel. Which creates a separate problem of equal-and-opposite reaction too. $\endgroup$
    – Graham
    Commented Apr 24, 2021 at 8:21

If the ship is moving, then the fuel inside the tank is moving as well, with the same speed as the ship so you can't just change it's course without cost. You will probably save a bit of thrust as you will be able to precisely place the thruster on the best angle to achieve the best trajectory to a new course. But what you save is very small compared to the energy you need to overcome the inertia of the moving mass of fuel and ship.

It is hard to tell if what you gain would be enough to justify the additional complexity and weight of the ship. In theory it may be, but I'd guess that even if there will be any gain it will be minimal at best while increased chance of mechanical breakdown would make this design undesirable.


In some cases it can be beneficial, in some, it is not so much if at all

A spherical tank is a good idea on its own, so even without the advantage u ask about u may consider having them, especially for space to space ships - for reasons of best volume to surface ratio, which saves mass.

Being able to change the location of engines on the tank, to some extend may mean being able to detach and quick change the tank at the destination point, or some other stage like strategies may become a possibility, even if it may be too hypothetical, but still

For big ships where u start to hit the limit of steel or other materials, so as when it is important to squeeze a few percent of the mass in cases of interstellar voyages as an example.

If some tank is 10-20km in diameter or a radius then steel can't be considered in the usual way as a strong material and such, and a better way to think about it would be it is a rubber

And the shell not necessarily can be turned around in a reasonable time, less than a month as an example, without rupture of that tank shell. It may or may not be overstretching things, but what if it is 100km or 1000km at some point it will be a nontrivial problem, even if someone imagines km's thick shell and much earlier for thin ones when we slash shell mass to increase payload.

You do not need special rails and what to make the thing work, it basically a body in space u can fly around and dock to a fuel port. The thing probably has to have some cushion solution/base to support spherical shape and all that. So work on it.

It could make an impression it may be a poor man solution, no fusion engines and alike but in fact, it is a must(or one of the viable options) for interstellar and fusion engines, if u talk about 0.1c and higher, even if it is used 4 times during the flight.

Another use case is your Neptunian tankers, where u carry hydrogen for poor belters to make water (yes, the Expanse, I'looking at u, poor thing) or bind oxygen from your metal whatever extraction processes, so as there are other uses.

And then the tank contains your reactive mass, so as it is a thin shell, the less massive it is the merrier. U leave it at the destination, attach an empty one, if required, which is not universally so, and head your engines back.

some tips

U do not need anything special on the surface of the tank to grab, what u need is basically 3 rings of fuel ports, and it will allow u to do any trajectory correction u may need. Spacing of those fuel ports in the worst case 10-30 degrees - so not that much. But more or less knowing which maneuvers and when required it can be less, 2-14 total, as an example.

Inside of that tank - there can be some machines as well which help u to fuel required ports or other needs, so u do not rely on passive means as it is in current rockets, and the internals of the tank may require certain improvements. Meaning those are machines that also change locations, and swim in the tank to do so. It may include reinforcement ribs as an example - u do not waste reactive mass on that, and in some cases, it may be the way to rotate the thing in the way u like for little energy expenses, there are pros and cons so it does not strike out your approach, but maybe a viable option for smaller ships.

U do not need any additional structures in the shell besides fuel ports, to operate it.

It may be a solution up to, including, nanotech and then there is 0 overhead and only benefits in this solution of yours. So better the tech more preferably it may be.


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .