I have a spaceship which is basically a cylinder a little less then a kilometer long. Attached to the ship by pylons connected to a hub is an inhabited ring which spins to produce a gravity-like effect. I want to get people from the hub up to the ship through one of the pylons. "Down" is the outside of the ring and the center of the ship has no gravity. What method of transport would lift them from the ring and by the time they got to the ship, cancel the momentum they would have from the spinning ring?

  • $\begingroup$ Is your question a duplicate of this one? worldbuilding.stackexchange.com/questions/134530/…. Not voting to close yet because maybe you are putting a different spin* on it. *haw haw! $\endgroup$
    – Willk
    Mar 16, 2019 at 16:15
  • $\begingroup$ Thanks. Sorry for the duplicate. $\endgroup$
    – James
    Mar 16, 2019 at 20:46
  • 4
    $\begingroup$ @Willk I'm not convinced this is a duplicate. That other question is asking how to dock a ship on the ring. This question is asking for the best way to get from the ring to the center (or weightless area) of the ship. It's actually a pretty good question since an elevator would need to turn around (or something) to keep you from banging your head when it comes to a stop in a weightless environment (after having endowed the rider with momentum). James? We need details: diameter of the pylons, ring, and ship at least. Are we assuming spin to emulate 1G? Edit your Q with the clarifications. $\endgroup$
    – JBH
    Mar 17, 2019 at 21:05

10 Answers 10


Firstly... make the inner 'hub' have two walls - an outer shell that rotates with the spokes and ring, and an inner wall that is stationary and useful to the people inside. So, the elevator rides down the spokes to the hub, still spinning, and then the elevator car transfers sideways to a track that runs around the central hub... initially moving at the same speed as the rotating spokes and wheel, but gradually slowing down until it's 'stationary' - synced with whatever root motion the inner cavity has. At that point, people can move freely through the hull into the inner cavity in zero g.

  • 2
    $\begingroup$ Or have the vertical lift open into a separate horizontal car that runs on the circular track. That way, you don't need to actually remove the lift from the pylon, or the horizontal car from the ring track. Also, you can easily have several pylons accessing the same ring track, each with its own, independent lift, and only need to coordinate the rendezvous between one lift and the horizontal car at a time. Third variant: You add a small waiting room at the upper end of each pylon, which is a stop to both the pylon's lift, and the horizontal car. With that, the systems are fully independent. $\endgroup$ Mar 18, 2019 at 22:32
  • 1
    $\begingroup$ I got lost in your description. Any chance for a sketch, maybe? $\endgroup$
    – Mołot
    Mar 19, 2019 at 16:06
  • $\begingroup$ When I thought my question was a duplicate and the "original" didn't seem to apply to my problem, I went back and rethought the issue and came up with a solution very similar to this one. I'll go ahead and mark this one as the answer, but I want to thank everyone for your suggestions since they add to my understanding of the described dynamics. Oh, I apologize for not answering sooner but I've been out of town with limited access to the internet. $\endgroup$
    – James
    Mar 21, 2019 at 13:57

I believe the answer is ‘the same way the passangers gained their angular momentum when trqvelling down the pylon.’. They only gain ‘gravity’ if their angular velocity increases as the pylon platform descends from zero g to N g. They’d gain that motion by holding on to hand rails and having their feet in intimate contact with the floor of the uppy-downy ride. And, when they travelled upwards, they’d expend their rotational energy against the floor and railings.

Up near the top its gets dicey since low g means the static coefficient of friction is not realiable — zero times any number is zero. So, sticky floors, like post-its glue, or magnetic boots, or velco, or just hold on with both hands would be sufficent when combined with low acceleration when the uppy-downy bit neared the top low g spot.

  • $\begingroup$ This method applies, and the discussion is the same, regardless if this is a ladder or an elevator. If an elevator, it might be handy to have soft walls for passengers to lean against. Since angular momentum is conserved, when people are brought to the hub the whole ship will spin slightly faster. I don't think this is a problem, but if it is one could have a counter-mass moving in the opposite direction. $\endgroup$
    – cmm
    Mar 20, 2019 at 21:12

The elevator can rotate around an axis parallel to the station axis, allowing the people in it to stay upright during the ride, and the angular momentum transfers itself to the ring automatically when you go upwards.


It is fairly simple if you do not care about using the "pylons" or are okay with them not connecting directly to the ring.

Simply surround the rotating ring with a non-rotating square so that the ring rotates inside the sides of the square at the their centers.

Then have an airlock inside the ring with the pressurized transport "vehicle" and release the vehicle to the square at the center point. The ring will continue rotating but the vehicle will continue along the direct line inside the side of the square. The vehicle would then go into free fall except that it decelerates preferably at a rate that sees it stop before the side of the square ends and it crashes thru the wall into vacuum.

Note that the "square" does not actually need to be one. You can make the sides whatever length gives you acceptable rate of deceleration. It can be oblong or #-shaped. You can even omit most of the sides. The ring will probably rotate fast enough that having one will be enough just like hard drives have been doing just fine with one set of heads. And just like with hard drive heads you will probably want to stack multiple "tracks" and airlocks. And you can have as many airlocks along the ring per track as you want for capacity. Since all the vehicles will decelerate at the same rate they will not collide even if they share the same track.

Anyway once the vehicle reaches the "corner" of the square or end of the deceleration track, you just connect the corner to the center with a pylon and have the vehicle move there. While the deceleration probably is best done with some sort of fairly robust tracks, normal or maglev, after this point the vehicle will be in free fall and assuming the square and pylon are pressurized it can simply fly.

Getting back to the ring just does the same thing in reverse. Fly to a corner, accelerate spinward along a track to match velocity with the ring and time it so that an airlock will catch you.

Technically if the ring rotates in a pressurized shell you might not need the airlocks and simple mechanism to catch and release the vehicles would suffice. But you probably want the airlocks for safety in which case the square will be unpressurized and there will be another set of airlocks at some point of the pylons. I'd guess at both ends of them. In that case some sort of maglev system to handle both acceleration and deceleration and the flight along the pylons would be the simplest solution. Simple rockets as a back up just in case.

EDIT: Added possible space station config to illustrate how it differs from the question. The possible space station


The simplest way is this:

  • the pylons are attached to the outer ring.
  • the pylons are also attached to a central pylon hub.

Now you have the pylon hub rotating (quite slowly) with respect to the rest of the ship, and you only need one airlock connected axially and capable of connecting the two rotating parts. This could be done in a lot of ways, for example with a ferrofluidic seal.

When you go up the pylon, you experience a lateral force - the Coriolis force. You can have the elevator car simply hinged at the "top" or mounted on a rail, and free to rotate. The Coriolis force will be maximum at mid-run, the car will rotate at an angle depending on ascension speed. It will start perpendicular to the ring, then tilt sideways, and then slowly tilt back before entering the hub.

enter image description here

For those interested, here is the mathematical solution to a very similar problem (the "pylon" has a 45° angle here, and gravity creates a lateral component, but simply assuming g=0 makes the two solutions identical).

Rough estimate

We want one G ($9.81 \frac{m}{s^2}$) at the edge of an habitat wheel of radius R. Gravity in a spinning wheel is given by the square of the peripheral linear velocity divided by R; the PLV is the length of the circumference, $2R\pi$, divided by the time taken for one revolution in seconds - which is to say, multiplied by RPM and divided by sixty. So:


which should give approximately $RPM = \frac{30}{\sqrt{R}}$.

For a wheel of radius 50 meters, we need roughly 4.2 RPM.

Supposing the elevator runs at a speed of V = 1 m/s (a run takes one minute, which is reasonable), we need to shed a linear velocity of $\frac{2R\pi(RPM)}{60}$ in a time of R/V, so the average lateral acceleration is $\frac{2\pi(RPM)}{60V}$ and, expressed in G, $V(RPM)(\frac{2\pi}{60\times9.81})$ or approximately:

Average lateral Coriolis acceleration when going down a wheel spoke while the wheel is turning at RPM rotations per minute

$C = V * RPM$ hundredths of a G.

Note that this is independent of the radius, because the longer the radius, the lesser the RPM needed to have one G at the border and the more time we have to shed lateral speed during the elevator run (i.e., the radius is already factored inside the RPM number).

For a V of 1 m/s, lateral speed is on average four hundredths of a G; enough to perceive a little swaying, no more. A tilting elevator might be overkill.

If you shoot down a pressurized pylon at 25 m/s, though, you experience a lateral acceleration of 105 hundredths of a G - in other words, you crash laterally, hard.


Arrange the ship-hub to have a non-rotating pylon with a tube inside it that extended from the hub down into the ring-habitat to within a short distance of the rings floor. Where the pylon passes through the ceiling of the ring it would be attached to a circular band of non-rotating “ceiling”. The rotating and non-rotating parts of the ceiling would be linked via a rotatory seal.

Although the tube and pylon would be non-rotating they would appear to move from within the habitat ring because the habitat ring would itself be rotating. The pylon with the tube would appear to sweep around the surface, a little like looking at the second hand of a clock from inside the clock.

Rails would run around the entire ring on the floor of the habitat directly below the tube pylon’s path and a motorised carriage would be provided to run on these rails. To use the device people would climb into the carriage and it would gently accelerate around the rails against the direction of the ring’s rotation.

When the carriage was directly under the tube it would stop accelerating and remain at constant angular velocity directly below the tube. At this point the occupants would have cancelled all of their angular momentum and would now be weightless. All that would remain would be to jump and they would be propelled along the tube all the way to the hub.

Obviously there are a wide range of variations on a theme here. If the ends of the habitat were to be utilized the rotatory seal issue could be simplified by using an airlock to transfer from the habitat-ring to the pylon/tube attached to the outside of the vessel. This could then be spun down to produce weightlessness and the rotatory seal could be much smaller and located on the axis of the hub.


If time of transportation isn't an issue, then you might be able to use velcro like they do here.

If the time of transportation is important, then my next guess would be something akin to a psuedo-railgun designed elevator where the person indicates where they want to end up, spend half the trip accelerating and the other half decelerating.

Due to the properties of magnets here, it would mean that they're kept in place by non-physical means, and the fluid bits of the body are the only concerns.

Downside is that they would take about the same amount of time to get anywhere inside the ship using this. And that would require your people to be wearing suits that have magnetic properties throughout them. And probably limit how quickly they can move because you don't want to sandblast people's faces off with air. And if it breaks AT ALL, you've got dead people.


Just Angle the Floor!

Or more precisely, rotate the elevator car.

A regular old elevator will have 2 problems: first, as the car rises or falls, changes in angular momentum give the perceived gravity force a horizontal component, as though the floor is sloping! Instead, just rotate the car and that force will feel vertical again.

Second, as the car stops or starts at the central hub, occupants will still have the same momentum/inertia and will launch upward at the ceiling! Instead, just put the floor above them first!

So let's put that all together... Both up and down cars gradually accelerate and then decelerate, as well as gradually rotate in the same direction as the ring spin, turning around until they end upside-down from where they started. For cars heading up to the hub, this gradually adds a soft sideways push from the floor to drain angular momentum, turning into an outward push to gradually settle occupants to a resting stop in microgravity. Cars heading down to the ring will start with the floor in toward the center of the hub, and occupants can just float in there. As the car accelerates, the floor will gently rise to meet them and the occupants will settle onto their feet. Then the push will gradually turn sideways to impart more angular momentum, finally stopping right-side up on the ring.

Easy as that!


Similar to how a transmission in an automobile switches between gears, you can have the elevator (pylon being the elevator shaft and therefore mobile) to/from the hub be a sort of 'clutch' system. The elevator would be an interface between the rotating ring section and the stationary hub, effectively forming mobile spokes. To transfer ring -> hub, the elevator would match velocity with the ring. Once the doors are closed, the pylon would decelerate in such a way as to comfortably come to a stop relative to the hub, in position to open into a hub access door. Transferring hub -> ring would be much the same, with the elevator pylon accelerating to meet the ring's velocity once the people are inside it.

EDIT: I should note that in my mind, the spokes remain equidistant from one another and move in unison, and they are the primary structural component holding the ring in place around the hub. The contact surfaces between the 3 components would be ultra-low friction or easily replaced, and the 3 components would effectively be completely 'detached' from one another, with a sort of rail (like a barn door track) holding their positions relatively constant with respect to one another.


There appear to be two parts to your question:

  1. How to transport someone from the ring to the hub
  2. How to cancel their momentum.

The pylons will need to be fixed at one end (either the ring or the hub).

Pylons Fixed to Hub

In this case since the hub is stationary all the pylons will also be stationary. The ring is just moving round the pylons (possibly using some kind of track).

So in order to cancel the momentum of the person you have to "accelerate" them in the opposite direction to the spin of the ring. However what you are really doing is decelerating them since once you match the velocity of the ring they will be stationary. Ideally you should "stop" them next to the pylon so that they can transfer to the pylon then move toward the center.

The movement of the elevator doesn't need to be complicated, you could use a rack and pinion or just two cables (one to pull it to the center and one to pull it back.

Pylons Fixed to Ring

Ideally you would have two sections of the central hub:

  • A rotating part attached to the pylons.
  • A stationary part.

Since the pylons are now fixed at both ends the elevator could be even simpler (one cable) since there will always be some "gravity" as long as the elevator is slightly offset from the center of rotation.

Once the person has been transported to hub they can be moved to the center of the hub. At that point the only momentum they will have is their rotation, so if you cancel their spin, they will then be "aligned" with the rest of the stationary hub.


I am assuming you don't want the person to have to put on a space suit every time they transfer, so we have to figure out how to transfer people.

In the first case (stationary pylons) you would probably need some kind of "car" that sits on the inner edge of the ring that can run on a track next to the track the pylons run along. Initially the car would be rotating with the ring so the person could use an airlock to enter the car.

The car would then "slow down" so that it was stationary with respect to one of the pylons and the person could transfer through another airlock to the pylon. If the entire pylon was pressurized no further airlocks would be required. If the pylon is not pressurized you would need additional airlocks to enter/exit the elevator.

In the second case (rotating pylons) the entire structure (ring, pylons and central "rotating" part of the hub could all be pressurized avoiding the need for airlocks.

The only problem is transferring the people from the rotating part of the hub to the stationary part of the hub. Probably the easiest way to achieve this would be to have a circular room, in the middle of the hub which can be rotated independently of both the hub and the ring that way you could spin the room to match the ring, let people enter through one airlock.

Then stop the "transfer" room and let them exit into the hub through another airlock into the stationary part of the hub.


You must log in to answer this question.

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