O ring seals (not the kind that catch tossed hoops) need grease to work with a vacuum. both need replacement, the grease frequently.
Good point, but there you can spread your seal along a considerable length of the propeller shaft. Unless you make you seal really wide, your rotating seal on a spacecraft item (unless you’re sealing a similar rotating shaft) will be a very different beast.
Some. I’ve built and used a lot of Wilson seals, in which a cylindrical object (often a glass or metal tube) is sealed by an O-ring that gets compressed so that it “flattens out” and gets pushed against the cylinder:
http://icecube.berkeley.edu/~bramall/laserhowtohtml/laser_project/wilson_seal.html
The O-ring is, in vacuum seals, generally sealed with a THIN layer of silicone vacuum grease. The cylinder in the resulting seal can rotate. If you want to feel better, you can use multiple seals.
it will hold an atmosphere of pressure, easily, but this isn’t something you use in high vacuum situations, and it’s not used for a constantly-rotating shaft. I suspect it would quickly enough wear away. For heavy-duty constant-use rotation, I’m sure you’d want something different.
The suggestion about a submarine propeller is a good one. But I have no idea what kind of packing they use, or how much resistant it offers to rotation. For a rotating seal on a spacecraft I’d think you’d want to minimize resistance, or else lose rotation. The submarine propeller has the very considerable torque of the sub motors behind it.
I like the idea of separating the two pieces - you have a section that does not rotate, with the hub, locks on and people go inside. The doors on both sides are sealed, the connection is broken; then the piece rotates until it is in sync with the rotating part, and lines up with that. They lock together, the doors open, and ta-da! into the rotating part.
Sort of like an elevator - maybe, even, you could make the transfer container be the elevator up and down the spoke too.
I think losing rotation is not that big of a problem. Electric energy is not so hard to get in space, you can use motor to spin it (yeah, I know motor would cause both non-rotating and rotating part to spin, and that you need rocket engine (and expensive fuel) to stop one in the beginning, but later friction would cause non-rotating part to slowly spin in direction of centrifuge and electric motor would be enough to reverse it).
Actually, electric energy is only easy to get in space in the form of solar energy (and maybe esoteric things like harvesting the solar wind). The problem is that you’re not connected to anything. You can’t take advantage of gravity differentials (like waterfalls) or temperature differentials, and you’ve got no good heat sinks. When spacecraft have to use large amounts of power, they use atomic batteries, and there was talk of nukes to power the SDI units. You sure can’t go about burning anything, because that uses up oxygen. Conservation of momentum being what it is, as you properly observe, spinning one thing one way results in the other part spinning the other way. This is useful on satellites, where you can use a small rapidly spinning wheel to change attitude without using propellant. But you need free rotation and low friction to make this efficient. (Heinlein had his spacecraft using the same dodge, putting the energy of rotation into storage as electricity).
If you have a lot of friction, though, you lose rotation without anything useful to show for it – it goes into heat. to maintain rotation, you have to burn propellant. Even if you store up solar energy and use wheels to move one part of your structure against the other, you still want a non-rotating part, so you need external thrusters to keep that part steady. (or else counter-rotate two parts of a shop and leave the center stationary. But now you need to maintain two sets of seals.)
On balance, it’d be easier to maintain “rotation discipline” and try to keep spinning things spinning with as little rotational friction as possible, I’d think. Give me my smoother rotation inside a non-rotating seal. I’ll have to make up the frictional losses from my bearings enough as it is.
You very likely want a non-rotating hub for mlutiple reasons; for one, it is much easier for spacecraft to dock with a non-rotating interface. Having to match rotation brings along an entirely new set of dynamic parameters which have to be matched in addition to the orbital elements of the station, which will be compounded by any tendancy of liquid propellants or cargo onboard the spacecraft to contribute slosh dynamics, notwithstanding the disorientation and potential nausea passengers may experience. For another, a non-rotating hub is the best place to mount communications and observation systems. While it is certainly possible to have communications systems which automatically point toward a given target, this also requires a lot of tolerance or a receiver with very high gain. For S-band, X-band, or laser communications necessary for high bandwith communications it is highly advantageous to have the transmitter mounted on a non-rotating structure that can maintain accurate pointing. A third reason is that, for the foreseeable future, solar electric power will likely be the primary power source for space stations, and a large station would require a very large solar array which could be optimally pointed, and which you would not want to design to have to withstand centripetal and centrifugal forces. This would necessicate either a connection to a non-rotating part. (In theory this could also be done by having a separate ‘solar farm’ constellation of solar satellites beaming power to the station via microwave, but then you’d still have to have a microwave receiver which you would like to remain fixed.) If you have a stationkeeping propulsion system it would also be best to be mounted on a fixed hub to simplify dymics.
The question the o.p. asks is not trivial, and there currently aren’t really seals and bearings designed for this type of application. Underwater seals for propeller shafts and similarly seals in turbopumps might seem to be a useable starting point at first blush, but in both cases you have a powered shaft which is (in theory) almost completely rigid, so that both friction losses and deflection can generally be discounted (if correctly designed), and the seal only has to worry about intrusion from the axial direction while the bearings only act in essentially the radial direction. For a hub of a space station, where you would want to minimize drag and bear some amount of loading in all directions, you would need a very different configurtion, which would likely have the bearings at the outside ends of the hub, designed to take both axial and moment loads, and then a face seal inside to maintain pressure. You would not use an o-ring for a rotating face seal, and in fact you’d probably want to have a pressurized fluid seal so that wear would not be an issue, with one or more shear layers (either highly pressurized inert gas or some kind of slip plate) to minimize losses due to shear friction. The best selection for bearings would likely be electromagnetic bearings, which would not require lubrication or expereience wear.
Stranger
Actually, yeah you would lost some energy as heat, but you would also cause non-rotating part to spin in direction of centrifuge, so motor would suffice to reverse it. Think about this from this way: from the centrifuge point of view, a non-rotating part is the part that rotates - there is friction on both parts of the bearing or in other words friction works on one body and second body experiences same force but with reverse direction.
It’s not the rotation I’m objuecting to, but how much energy is lost to friction and heating things up. It takes a fair amount of energy to rotate a mass with a large moment of inertia – and any portion of a spacecraft you could live in or perform experiments in would have a large moment of inertia. If you’re holding the seal tightly closed, my guess is that you;ll have a lot of friction, which will go into heat and be unrecoverable (and will heat your ship). To re-spin, you apply force with your motors. That could well be a LOT of force you’re going to have to supply, and solar power may not be up to it. If you use some other method of generating the power, you’ll be generating more heat.
angular momentum of everything being conserved means that, yes, you ought to be able to reach the same situation of rotating part + non-rotating hub. but it’ll require inputting as much energy as was lost to friction 9-plus what is lost to friction during the recovery).
as said, they sell rotational attitude adjustment devices for satellites – I’ve seen the websites – but they use relatively small, low-friction wheels. If you’ve got friction slowing down your spin there’s going to be a price to pay.