How the heck do you have gas-tight bearings between non rotating and rotating sections?

[SomeGuy123]

So there was a proposed "Nautilus X" module that would be added to the international space station so that centrifugal gravity experiments on humans can be performed.  As a side note, this is kind of critical - any proposals to colonize the Moon or Mars need to know if humans can live in reduced gravity long term without going blind or other show stopper problems.  

There's just one teensy problem.  See that spinning ring?  And then that non spinning section that connects to the ISS?  The ring part is going to have a circular "sleeve" that goes over some kind of bearing.  It must turn, so no hermetically sealed connections.

What stops air from just leaking right on out through this sleeve?

 

[Bill Phil]

Non rotating hub?

[Tex_NL]

7 minutes ago, SomeGuy123 said:

...

What stops air from just leaking right on out through this sleeve?

...

And that question is probably the answer why there isn't such a structure on the ISS yet.

But keep in mind the wrist connectors on EVA suits are also rotating bearings. Those too are air tight. Scaling those up might be possible but tolerances will have to be extremely small.

[Frybert]

I remember thinking about this a long time ago, and upon looking it up found the answer. This was a LONG time ago so I don't remember where I found it but basically it was a combination of 14 PSI not actually being that much pressure, a type of hydraulic fluid that would serve as a barrier and just not minding the small amount that would get out.

[cantab]

A little searching around turned up a slightly ... odd ... but nonetheless informative video on the subject:

It focuses mainly on seals that hold liquids, but I expect the same principles will apply to holding a gas pressure difference.

[magnemoe]

6 minutes ago, cantab said:

A little searching around turned up a slightly ... odd ... but nonetheless informative video on the subject:

It focuses mainly on seals that hold liquids, but I expect the same principles will apply to holding a gas pressure difference.

Yes having mechanical seals who handle high pressure steam is far harder, many of the steam turbine shafts are also pretty thick and face far higher loads. 
Worse case use two seals and an vacuum pump 

 

[PB666]

2 hours ago, SomeGuy123 said:

So there was a proposed "Nautilus X" module that would be added to the international space station so that centrifugal gravity experiments on humans can be performed.  As a side note, this is kind of critical - any proposals to colonize the Moon or Mars need to know if humans can live in reduced gravity long term without going blind or other show stopper problems.  

There's just one teensy problem.  See that spinning ring?  And then that non spinning section that connects to the ISS?  The ring part is going to have a circular "sleeve" that goes over some kind of bearing.  It must turn, so no hermetically sealed connections.

What stops air from just leaking right on out through this sleeve?

 

In the lab we use a teflon disk sandwiched between two absolutely flat glass pieces to create a vacuum, under vacuum the disk seal, but the teflon disk has a lubricating feature. It generates a considerable amount of friction. I bet the bearing has a gas recovery system that secondary to the primary seal

BTW I can see they will have a problem with hunchback astronauts soon. Have you ever tried to conduct calisthenics in your attic?

 

Edited January 31, 2016 by PB666

[PB666]

2 hours ago, Tex_NL said:

And that question is probably the answer why there isn't such a structure on the ISS yet.

But keep in mind the wrist connectors on EVA suits are also rotating bearings. Those too are air tight. Scaling those up might be possible but tolerances will have to be extremely small.

But keep in mind that these are not used that much, on the teflon seals in the lab, they eventually wear down, so if the toroid constantly rotates the seals will begin to leak.

 

[K^2]

3 hours ago, SomeGuy123 said:

What stops air from just leaking right on out through this sleeve?

Same problem is found in vacuum pumps, and a whole bunch of other applications where you need moving parts that have to move past each other smoothly, but must not allow gases to leak, and must be vacuum-safe.

The solution is various mineral oils with absurdly low vapor pressures. They form an air-tight seal and lubricate the moving parts. At the same time, they evaporate only very, very slowly even in vacuum.

In some applications, where you need really high quality vacuum, ionic fluids are used instead of regular oils. These are fluids that consist of two kinds of ions with opposite charges. As a result, they have almost zero vapor pressure. If a molecule tries to leave ionic fluid, it's pulled right back in by the resulting electric charge. So they effectively do not evaporate in vacuum. They are, however, more expensive and have narrower temperature range. I think it's going to be simpler and cheaper to just refresh the mineral oil in the seal every once in a while on a space station.

[Rakaydos]

Going a bit more high tech, I remember reading about "plasma walls" that use magnetically contained plasma to block leaking air.

Given a narrow gap of a bearing, do you think this approach would be effective?

[Nibb31]

This is exactly why rotating hubs are a bad idea. It's much easier to simply rotate the entire vehicle.

[magnemoe]

43 minutes ago, K^2 said:

Same problem is found in vacuum pumps, and a whole bunch of other applications where you need moving parts that have to move past each other smoothly, but must not allow gases to leak, and must be vacuum-safe.

The solution is various mineral oils with absurdly low vapor pressures. They form an air-tight seal and lubricate the moving parts. At the same time, they evaporate only very, very slowly even in vacuum.

In some applications, where you need really high quality vacuum, ionic fluids are used instead of regular oils. These are fluids that consist of two kinds of ions with opposite charges. As a result, they have almost zero vapor pressure. If a molecule tries to leave ionic fluid, it's pulled right back in by the resulting electric charge. So they effectively do not evaporate in vacuum. They are, however, more expensive and have narrower temperature range. I think it's going to be simpler and cheaper to just refresh the mineral oil in the seal every once in a while on a space station.

In an vacuum chamber you also don't want traces of oil, its worse than air as you will do experiments who they can easy mess up. 
On an space station leaks is only an issue if it you have to resupply more often as you say. 
One bar is low pressure difference compared to other setting, in others where you pump dangerous stuff leaks is more dangerous 

 

[cantab]

48 minutes ago, Nibb31 said:

This is exactly why rotating hubs are a bad idea. It's much easier to simply rotate the entire vehicle.

Except that has its own practicality issues. Varying gravity and possible Coriolis forces as people move about between the hub and rim could cause issues. To make controlled manoeuvres in a spinning ship is going to be problematic, and to despin and respin it for every manoeuvre will either require a massive gyroscope or be a big waste of propellant. The overall design is constrained in ways that may not be desired. And it's out of the question on a test facility that will be added to the ISS.

If the gas seal truly is an issue, there's a very simple solution: encase the whole thing in an outer, stationary, pressure chamber. This is I believe what was depicted on the Discovery in 2001 - the centrifuge was a ring inside the spherical crew section. A possible drawback of this approach is additional losses from air drag, but I'm inclined to think that needn't be too high.

[fredinno]

5 hours ago, Tex_NL said:

And that question is probably the answer why there isn't such a structure on the ISS yet.

But keep in mind the wrist connectors on EVA suits are also rotating bearings. Those too are air tight. Scaling those up might be possible but tolerances will have to be extremely small.

The better answer is lack of cash, though, really.

[PB666]

1 hour ago, cantab said:

Except that has its own practicality issues. Varying gravity and possible Coriolis forces as people move about between the hub and rim could cause issues. To make controlled manoeuvres in a spinning ship is going to be problematic, and to despin and respin it for every manoeuvre will either require a massive gyroscope or be a big waste of propellant. The overall design is constrained in ways that may not be desired. And it's out of the question on a test facility that will be added to the ISS.

If the gas seal truly is an issue, there's a very simple solution: encase the whole thing in an outer, stationary, pressure chamber. This is I believe what was depicted on the Discovery in 2001 - the centrifuge was a ring inside the spherical crew section. A possible drawback of this approach is additional losses from air drag, but I'm inclined to think that needn't be too high.

Computers can solve this problem these days. Some of the ion thrusters I have seen weight a fraction of a pound such that on a large station they can be located just about anywhere.

The other thing is simply despin the station, make the correction and spin back up. BTW since thrusters don't burn oxygen, the can be set on a part of a ship that does spin, so that for off axis-DV the thruster can simply rotate to were it needs to be on that axis and fire.

[fredinno]

9 minutes ago, PB666 said:

Computers can solve this problem these days. Some of the ion thrusters I have seen weight a fraction of a pound such that on a large station they can be located just about anywhere.

The other thing is simply despin the station, make the correction and spin back up. BTW since thrusters don't burn oxygen, the can be set on a part of a ship that does spin, so that for off axis-DV the thruster can simply rotate to were it needs to be on that axis and fire.

But ion drives increase complexity, power consumption, mass, and cost.

[K^2]

2 hours ago, magnemoe said:

In an vacuum chamber you also don't want traces of oil, its worse than air as you will do experiments who they can easy mess up.

Highly dependent on experiment. There's a bunch of stuff where you really just care about low pressure. But yeah, that's precisely why there's an entire industry of super low vapor pressure lubricants for experiments. Sometimes you really care about the quality of that vacuum. I've never had to work with anything remotely that needy, but I'm told the high end pressure chambers actually have a turbine stage to get rid of these last traces of molecules when gas is basically just in ballistic mode.

[PB666]

3 hours ago, K^2 said:

Same problem is found in vacuum pumps, and a whole bunch of other applications where you need moving parts that have to move past each other smoothly, but must not allow gases to leak, and must be vacuum-safe.

The solution is various mineral oils with absurdly low vapor pressures. They form an air-tight seal and lubricate the moving parts. At the same time, they evaporate only very, very slowly even in vacuum.

In some applications, where you need really high quality vacuum, ionic fluids are used instead of regular oils. These are fluids that consist of two kinds of ions with opposite charges. As a result, they have almost zero vapor pressure. If a molecule tries to leave ionic fluid, it's pulled right back in by the resulting electric charge. So they effectively do not evaporate in vacuum. They are, however, more expensive and have narrower temperature range. I think it's going to be simpler and cheaper to just refresh the mineral oil in the seal every once in a while on a space station.

Having rebuilt quite a few vacuum pumps its a bit of a different situation, Vacuum pumps actually create a dynamic equilibrium, which would not work well except most good lab pumps have dual stage. The vanes on a vacuum pump aren't really designed to capture all gas, just most of it, as the gas finally reaches the second stage it cant get back to the first, and once it is expelled to the second stage the gas is heated, expands floats up and effectively its not capable of reversing. The reason these pumps work like this is that rotary pumps can remove alot of gas before kicking into the second stage, and BTW is you put your hand over the outlet, it will get sprayed with oil, on most of the pumps I used for flash evaporation I had to build a muffler to trap the oil, oil may not be volatile, but when you are through putting couple litters a second it is quite atomizable. If you turn a vacuum pump off without releasing the vacuum, the oil and some gas will go back up the tube, thus the application of energy is essential to keeping the vacuum. In this space application the separation of pressure needs to be more passive, but I suspect they probably have a secondary seal and a pressure monitor with a backup evacuation system (along with pressure doors somewhere close by). Technically speaking the ISS setup does not even need to hold pressure while turning, the could simply pressurize the bearing while stationary transfer the naut and close the pressure doors on both sides, then depressurize the ring seal. That toroid needs a ballast anyway, just put a purificaiton system in there. Non-inertial or not that vessel is not an effective barrier cosmic radiation and x-rays.

We had a vacuum pump that was just barely able to boil water at ambient, it got to the lyophilization limit, when I took it apart I found one vane had a rusted inelastic tensioner spring and the other vane was stuck. Not ideal of course, but just goes to show that a good position dynamic equilibrium (that is turning those vanes rapidly enough). The biggest bane to vacuum pumps is moisture or acids in the vanes, they swell up either freezing the pump or freezing the vanes in the retracted position. The way to prevent this is to every now an then dump a couple cups of vacuum pump oil down the inlet to sort of wash god-only-knows-what through the pump, then turn it off quickly, wait a while and drain the heavies from the bottom of the pan. Things like ether rapidly reduce the viscosity, on the low pressure side they can bind to the oil despite high volatility and vacuum pump is nothing if not viscous.

The vacuum pump on an ultracentrafuge is a diffusion pump, it is almost completely a dynamic process, horribly inefficient but reasonably compact and suitable for circumstances were limited through put of volatile fluids is expected.

[Kerbart]

Probably start with a labyrinth seal to significantly lower the pressure difference, and then use one of the solutions listed above. Perhaps finish it off with some rubber skirt, although I can see how the friction of that would cause problems (you'll have to provide more torque to keep the ring spinning, and that will cause your station to spin very slowly in the opposite direction)

[sgt_flyer]

i don't know if labyrinth seals are adapted for gases liquids sure, but gases will be harder i think (unless you put something that can stay liquid / low viscosity in vacuum within the labyrinth seal, to act as a barrier)

 

 

[mikegarrison]

15 hours ago, sgt_flyer said:

i don't know if labyrinth seals are adapted for gases liquids sure, but gases will be harder i think (unless you put something that can stay liquid / low viscosity in vacuum within the labyrinth seal, to act as a barrier)

They are used in jet engines. There are a lot of seals used in jet engines that keep very high pressure gas from getting past very high speed rotation parts. Many of them are ablative seals, meaning that the rotating part actually cuts its own groove in the non-rotating part, thus making the seal as tight as it can be. They all have some leakage, though.

I don't know what the normal gas leakage rate is from the ISS, but it would probably go up with this. (Of course, it would probably go up with any new module you add on, because nothing is perfect.)

[DerekL1963]

Seals (there are many well proven varieties to choose from) and accepting a small amount of leakage as the price of doing business.

[sgt_flyer]

55 minutes ago, mikegarrison said:

They are used in jet engines. There are a lot of seals used in jet engines that keep very high pressure gas from getting past very high speed rotation parts. Many of them are ablative seals, meaning that the rotating part actually cuts its own groove in the non-rotating part, thus making the seal as tight as it can be. They all have some leakage, though.

I don't know what the normal gas leakage rate is from the ISS, but it would probably go up with this. (Of course, it would probably go up with any new module you add on, because nothing is perfect.)

 

well, here's a paper about iss leakage rates.

http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/20110012997.pdf

though, i doubt jet engine seals can be taken as an example for the problems a centrifuge will face - afterall, if your seal is just to prevent high pressure / high temperature leakage, the resulting microleaks will not be enough to affect the engine.

besides, a jet engine's ablative section of the seals can be replaced on tthe ground. (afterall, jet engines are expensive, so having the capability to extend the life of such engine is useful, and the engine is designed to allpw replacement of such part. - having to change such a seal on a module on a space station, would basically mean having to disconnect the two parts - in space !to replace such an ablative seal (EVAs, canadarm operations and such - dangerous). so using an ablative seal like that would mean giving a maximum lifetime to such a module  (and undocking redocking such a thing would still lead to an increase in leakage rate, even with new seals. (iss modules leakage rate are already 15 times higher in space than they were when tested on ground - the leakage increase is assumed to be because of launch vibrations)

 

 

[magnemoe]

7 hours ago, mikegarrison said:

They are used in jet engines. There are a lot of seals used in jet engines that keep very high pressure gas from getting past very high speed rotation parts. Many of them are ablative seals, meaning that the rotating part actually cuts its own groove in the non-rotating part, thus making the seal as tight as it can be. They all have some leakage, though.

I don't know what the normal gas leakage rate is from the ISS, but it would probably go up with this. (Of course, it would probably go up with any new module you add on, because nothing is perfect.)

Jet engines faces high pressure and high temperature at places. They also has high rotation rate. an labyrinth seal will help cooling the gas if you have cooling in the shaft so it don't burn the bearing. 
An rotating module on the ISS would rotate slowly have one bar over pressure but be large, hollow in center and will face far more vibrations than high speed systems do.
Seal will also be hard to replace. Using two seals and an vacuum pump would solve the problem as I see it. first seal would have most leak as the pressure difference is close to 1 bar, pressure difference in outer seal is millibars. 

[Nibb31]

With the advances in space biology, we don't even know if artificial gravity is even necessary for the foreseeable future. For all we know, we can fight bone loss with medication and exercice. We've proven that people can live on the ISS for over a year. By the time we ever need long duration spaceflight, we might have pushed that limit to two years or more...

But even if it is necessary, why bother when you can rotate the whole spacecraft. You're going to introduce a whole slew of fragile parts that need replacing and maintenance (seals, lubricants, etc...), along with a whole bunch of new failure modes. I know it looks cool (like wings on a spaceship), but rotating segments are just another solution looking for a problem. Keep it simple !

 

Edited February 1, 2016 by Nibb31