Author Topic: Rotational joint construction on artificial gravity space stations  (Read 18118 times)

Offline Burninate

  • Full Member
  • ****
  • Posts: 1135
  • Liked: 352
  • Likes Given: 73
Conversation forked from this thread
« Last Edit: 03/12/2015 10:00 pm by Burninate »

Offline Burninate

  • Full Member
  • ****
  • Posts: 1135
  • Liked: 352
  • Likes Given: 73
Archived comments on the topic:

* Any actuated joints are prone to failure, and actuated joints or swivels which have contiguous pressure zones, such that a human could climb through them, are probably impossible for more than a small range of motion.
The usual approach to multiple docking ports in a spinning station is to have a de-spun arm connected through the axis of rotation to the hub with the ports coming off of that.
 There was a thread on here a little while ago suggesting that the despin joints would be some sort of insurmountable engineering challenge, which is nonsense.

http://www.artificial-gravity.com/sw/SpinCalc/SpinCalc.htm
The usual approach to multiple docking ports in a spinning station is to have a de-spun arm connected through the axis of rotation to the hub with the ports coming off of that.
 There was a thread on here a little while ago suggesting that the despin joints would be some sort of insurmountable engineering challenge, which is nonsense.

The problem is more that friction on the despin joint, however large or small, requires a continuous expenditure of fuel to desaturate.  Configurations which balance multiple wheels do not, but they vastly increase the amount of compression structural members and hence mass.  Simultaneously, it is extraordinarily difficult to pass anything through that joint while it's rotating other than maybe power via commutators, due to the problem of vacuum sealing.  Attempts to solve the latter problem make the former problem bigger.
The usual approach to multiple docking ports in a spinning station is to have a de-spun arm connected through the axis of rotation to the hub with the ports coming off of that.
 There was a thread on here a little while ago suggesting that the despin joints would be some sort of insurmountable engineering challenge, which is nonsense.

The design challenge certainly isn't insurmountable, but it isn't trivial either. There are alternative designs that avoid the use of pressurized joints entirely, such as for example having a small pressurized module along the center that can be spun up for a pressurized connection with the rotating segment, or spun down to dock with a spacecraft or a non-spinning part of the station.
Rotating pressure joints aren't hard or rare, friction in such joints is low, and vacuum stable lubricants are also not hard.

I won't be replying to further comments on this topic as it's derailing the thread.
« Last Edit: 03/12/2015 10:06 pm by Burninate »

Offline nadreck

I am very interested in this topic, I do feel that in the longer term rotating workspace and habitats will be an important aspect of human activity off Earth. This seems a key aspect to structuring a safe and productive multipurpose station.

It is all well and good to quote those things that made it past your confirmation bias that other people wrote, but this is a discussion board damnit! Let us know what you think! And why!

Offline Coastal Ron

  • Senior Member
  • *****
  • Posts: 4380
  • I live... along the coast
  • Liked: 3029
  • Likes Given: 4047
I've been thinking that the connection to the "floating dock" located off of the center of the spinning station need not be connected all the time, or if it is it only needs to be connected by lightweight arms that are used to keep the dock at a specific location.

I would also imagine that the "floating dock" doesn't even really need to be too close to the spinning station, since mostly what it's needed for is to be a loading and unloading dock for visiting vehicles.  The cargo can be maneuvered to the center of the spinning structure where each load can be grabbed by a counter-rotating set of arms that only need to operate when there is cargo coming or going from the station.  That way the only mass that comes near the spinning station is just cargo, not vehicles.
If we don't continuously lower the cost to access space, how are we ever going to afford to expand humanity out into space?

Offline Alf Fass

  • Full Member
  • ****
  • Posts: 452
  • The Abyss
  • Liked: 89
  • Likes Given: 83
Rotating pressure joints are widespread though out industry, many sizes able to handle various medium (air, oil, milk) and some at high rotation speeds, with low friction and very low leakage, the challenging bit in a space station application would be the need for multiple joints to accommodate lateral movement due to limited imbalance in the stations rotation, I say limited because with or without the despin you don't want off center movement of the hub.
When my information changes, I alter my conclusions. What do you do, sir?
John Maynard Keynes

Offline Paul451

  • Full Member
  • ****
  • Posts: 1573
  • Australia
  • Liked: 789
  • Likes Given: 650
The cargo can be maneuvered to the center of the spinning structure where each load can be grabbed by a counter-rotating set of arms

The arms don't have to counter-rotate. If you spin the cargo to match the station, park on the same axis of rotation, the cargo is effectively stationary WRT the robo-arm. Even if the base of the robo-arm is off-centre (due to mass imbalances in the station), as long as the arm can reach the centre-of-rotation, it can grab the payload. (This is also true even if the grapple-fixture on the cargo (or cargo-module) is not on the line of the axis of rotation. It will still appear stationary WRT to arm.)

The hard part, if the station is off-centre, is what to do with the cargo once it's been grabbed. You still have to get it inside. If the docking hatch is off the axis of rotation, how far can the robo-arm push the cargo into partial gravity before it exceeds its strength? (An alternative is to have an "elevator" for cargo that can move up to any centre-of-rotation, then move down into the gravity areas where the docking-hatch/air-lock is located. That way your docking-module doesn't have to be anywhere near the centre-of-rotation.)

IMO, this simplifies docking and cargo-deliveries enough that it's unlikely that you need a counter-rotating docking module, nor a free-flying equivalent¹ on an early station. And maybe not even on large stations. The areas you will need counter-rotating systems will be the solar arrays, maybe the radiators, and possible the comms (especially for BEO missions that use AG.) Similarly engines. Not just on an NEP/SEP BEO mission, but also for normal orbital maintenance burns for LEO stations.

People typically first think of the docking module when thinking of the need for counter-rotation, but I think it's actually "everything but".

¹ The safety advantage of a free flying docking platform can be retained by having a small robo-armed tug go out and grapple the incoming capsules/etc at a safe distance from the station (even remove unpressurised payload or self-contained pressurised modules while out there), then return to the station and do the spin-alignment docking thing. That way your delivery vehicles don't need any special docking systems beyond a fairly standard grapple fixture.) I suspect the fuel use will be the same as ferrying back and forth between the spin-station and a free-flying facility, while the cost/complexity will be less.

Offline Hanelyp

  • Full Member
  • ***
  • Posts: 369
  • Liked: 65
  • Likes Given: 252
Rotating pressure joints are widespread though out industry, many sizes able to handle various medium (air, oil, milk) and some at high rotation speeds, with low friction and very low leakage
How low is "low leakage" and "low friction"?

Offline Alf Fass

  • Full Member
  • ****
  • Posts: 452
  • The Abyss
  • Liked: 89
  • Likes Given: 83
The ones I'm familiar with are used in dairy sheds, up to 100mm diameter, you can rotate them easily by hand and they have essentially zero leakage (necessary for hygiene reasons). in the application we're talking about a diameter of a meter or so would be necessary, but I just don't see why people think there's some sort of great challenge here.

Here's some swivel joints, diameters up to a meter and working pressures up to 6000psi, working temperature ranges of 300C.

http://nmf-group.com/en/products/swivel-joints/heavy-industry
When my information changes, I alter my conclusions. What do you do, sir?
John Maynard Keynes

Offline Burninate

  • Full Member
  • ****
  • Posts: 1135
  • Liked: 352
  • Likes Given: 73
The ones I'm familiar with are used in dairy sheds, up to 100mm diameter, you can rotate them easily by hand and they have essentially zero leakage (necessary for hygiene reasons). in the application we're talking about a diameter of a meter or so would be necessary, but I just don't see why people think there's some sort of great challenge here.

Here's some swivel joints, diameters up to a meter and working pressures up to 6000psi, working temperature ranges of 300C.

http://nmf-group.com/en/products/swivel-joints/heavy-industry

How do you extrapolate this to space, where lube is a problem because it's exposed to vacuum and exposed to the breathing gas pool, where soft sealing materials & gaskets are a problem because they're exposed to vacuum and/or they are too high-friction, where gaseous nitrox rather than a liquid would be employed for common breathing gas pool, and where leakage must be actually rather than 'near' zero?

We could start with - how is sealing accomplished in docking ports and airlocks today?  How difficult is it, and how much of it is compatible with the door rotating?
« Last Edit: 03/13/2015 10:20 pm by Burninate »

Offline Alf Fass

  • Full Member
  • ****
  • Posts: 452
  • The Abyss
  • Liked: 89
  • Likes Given: 83
Lubricants are only a problem if you're silly enough to use the wrong ones.
http://en.wikipedia.org/wiki/Materials_for_use_in_vacuum#Lubricants.

When I said "near zero", that's what I meant, not "just a little bit".

"We could start with - how is sealing accomplished in docking ports and airlocks today?  How difficult is it, and how much of it is compatible with the door rotating?"

Are you being serious?
When my information changes, I alter my conclusions. What do you do, sir?
John Maynard Keynes

Offline Coastal Ron

  • Senior Member
  • *****
  • Posts: 4380
  • I live... along the coast
  • Liked: 3029
  • Likes Given: 4047
The cargo can be maneuvered to the center of the spinning structure where each load can be grabbed by a counter-rotating set of arms

The arms don't have to counter-rotate. If you spin the cargo to match the station, park on the same axis of rotation, the cargo is effectively stationary WRT the robo-arm. Even if the base of the robo-arm is off-centre (due to mass imbalances in the station), as long as the arm can reach the centre-of-rotation, it can grab the payload. (This is also true even if the grapple-fixture on the cargo (or cargo-module) is not on the line of the axis of rotation. It will still appear stationary WRT to arm.)

Agreed.  My comment was regarding cargo that may not be rotating, but in practice it may be more likely that all cargo will be spun prior to handing off to the station.

Quote
The hard part, if the station is off-centre, is what to do with the cargo once it's been grabbed. You still have to get it inside. If the docking hatch is off the axis of rotation, how far can the robo-arm push the cargo into partial gravity before it exceeds its strength? (An alternative is to have an "elevator" for cargo that can move up to any centre-of-rotation, then move down into the gravity areas where the docking-hatch/air-lock is located. That way your docking-module doesn't have to be anywhere near the centre-of-rotation.)

I while back I spent some time thinking about a rotating station design.  What I came up with was a cable-stayed design (like a bicycle wheel, but more complex), where there would be a core section that pretty much was just an attachment point for the structure.  At the two sides of the core would be the entry points for cargo and people, both of which would immediately be loaded onto elevators to be taken to higher gravity locations on the station.  No doubt the station could rotate off-center to some degree, but you could always compensate by having the arms on the transit point be able to slide to whatever the absolute center of rotation is.

Quote
IMO, this simplifies docking and cargo-deliveries enough that it's unlikely that you need a counter-rotating docking module, nor a free-flying equivalent¹ on an early station.

I would think it makes it even more important to have a free-flying dock, since that becomes the work space for loading and unloading visiting vehicles, and you want dedicated personnel moving cargo towards the rotating station for safety reasons.

Quote
The areas you will need counter-rotating systems will be the solar arrays, maybe the radiators, and possible the comms (especially for BEO missions that use AG.) Similarly engines. Not just on an NEP/SEP BEO mission, but also for normal orbital maintenance burns for LEO stations.

I'm not an engineer, so the design I was working on may be undoable.  But if it worked as I envisioned then you would mount solar panels on the sides of the "wheel" of the station, as well as radiators.  Maneuvering engines are a little trickier, and no doubt it depends on whether the station is in a stable location or if it's in transit somewhere.  However since SEP engines are so small maybe you would just mount a lot of them on the outer ends of the station and use them for spin & de-spin plus location changes.
If we don't continuously lower the cost to access space, how are we ever going to afford to expand humanity out into space?

Offline Burninate

  • Full Member
  • ****
  • Posts: 1135
  • Liked: 352
  • Likes Given: 73
"We could start with - how is sealing accomplished in docking ports and airlocks today?  How difficult is it, and how much of it is compatible with the door rotating?"

Are you being serious?

Yes.  ELI5.  Consider your readers to be ignorant of the status quo here, and explain your map of the factual landscape.  Links are acceptable tools.
« Last Edit: 03/13/2015 11:33 pm by Burninate »

Offline KelvinZero

  • Senior Member
  • *****
  • Posts: 3947
  • Liked: 660
  • Likes Given: 169
Found this with a quick google. Sounds interesting.
http://en.wikipedia.org/wiki/Ferrofluidic_seal

Offline AlanSE

  • Full Member
  • *
  • Posts: 150
  • N Cp ln(T)
    • Gravity Balloon Space Habitats Inside Asteroids
  • Liked: 44
  • Likes Given: 31
I would also imagine that the "floating dock" doesn't even really need to be too close to the spinning station, since mostly what it's needed for is to be a loading and unloading dock for visiting vehicles.  The cargo can be maneuvered to the center of the spinning structure where each load can be grabbed by a counter-rotating set of arms that only need to operate when there is cargo coming or going from the station.  That way the only mass that comes near the spinning station is just cargo, not vehicles.

In this view, the task of docking will be performed by a microgravity station which is, for all practical purposes, an autonomous station. I also find it logical to presume that most atmosphere, electrical, and heat management systems will be independent. There's not good economic motivation for connecting those systems between the rotating station and the non-rotating station.

The Nautilus-X reflects this design philosophy, although it was downplayed. To transfer to/from the centrifuge, you enter into a transfer vessel, and then seal it up. That vessel then spins up and connects to the centrifuge. The transfer vessel will still be on tracks, sure. The obvious implication is that the centrifuge and the ISS have different air. This will result in duplication of systems. The alternative to duplication just isn't very pretty, except for computer systems and possibly electrical supply.

I think we've underestimated the niche for in-atmosphere rotating systems. High RPM systems will be fine if the purpose is to extend the stay of astronauts in space. They'll spend an hour every day on a rotating bike, and then we'll almost certainly see benefits to bone density. We can already extend stays significantly with current exercise research. The initial purpose for artificial gravity will be a quick biological benefit, keeping the rest of activities in microgravity modules.

Even as you increase the scale we're working at, it's obvious that radiation limits stays too. It makes little structural sense to have the shielding rotating along with you. If we are using asteroid resources for shielding at, let's say, an EML-2 station, then you'll go with a stationary shielding envelope with a rotating ring inside of it. So even at this point, a strong decoupling of microgravity and artificial gravity stations makes little sense. Because of that, the in-atmosphere rotation will continue to be an attractive option.

Going even further into the future, and larger in scale, there's no problem with in-atmosphere rotation scaling up. Drag forces will increase, but you can sheathe the rotating portion such that the friction is greatly minimized. All the while, you can keep a nice continuous hallway between the different sections at the hub. Sure, this is more complicated in a naive sense, but it takes care of the rotating joint problem, makes the shielding easier, and presents a lot of safety benefits.

Offline Alf Fass

  • Full Member
  • ****
  • Posts: 452
  • The Abyss
  • Liked: 89
  • Likes Given: 83
It occurs to me that when we develop industrial practices like refining materials from asteroids swivel joints like those being discussed will be used in large numbers, centrifuges are going to be needed in the refining process at many stages.
« Last Edit: 03/16/2015 03:12 pm by Alf Fass »
When my information changes, I alter my conclusions. What do you do, sir?
John Maynard Keynes

Offline Nilof

  • Full Member
  • ****
  • Posts: 914
  • Liked: 376
  • Likes Given: 540
I would also imagine that the "floating dock" doesn't even really need to be too close to the spinning station, since mostly what it's needed for is to be a loading and unloading dock for visiting vehicles.  The cargo can be maneuvered to the center of the spinning structure where each load can be grabbed by a counter-rotating set of arms that only need to operate when there is cargo coming or going from the station.  That way the only mass that comes near the spinning station is just cargo, not vehicles.

In this view, the task of docking will be performed by a microgravity station which is, for all practical purposes, an autonomous station. I also find it logical to presume that most atmosphere, electrical, and heat management systems will be independent. There's not good economic motivation for connecting those systems between the rotating station and the non-rotating station.

The Nautilus-X reflects this design philosophy, although it was downplayed. To transfer to/from the centrifuge, you enter into a transfer vessel, and then seal it up. That vessel then spins up and connects to the centrifuge. The transfer vessel will still be on tracks, sure. The obvious implication is that the centrifuge and the ISS have different air. This will result in duplication of systems. The alternative to duplication just isn't very pretty, except for computer systems and possibly electrical supply.

I think we've underestimated the niche for in-atmosphere rotating systems. High RPM systems will be fine if the purpose is to extend the stay of astronauts in space. They'll spend an hour every day on a rotating bike, and then we'll almost certainly see benefits to bone density. We can already extend stays significantly with current exercise research. The initial purpose for artificial gravity will be a quick biological benefit, keeping the rest of activities in microgravity modules.

Even as you increase the scale we're working at, it's obvious that radiation limits stays too. It makes little structural sense to have the shielding rotating along with you. If we are using asteroid resources for shielding at, let's say, an EML-2 station, then you'll go with a stationary shielding envelope with a rotating ring inside of it. So even at this point, a strong decoupling of microgravity and artificial gravity stations makes little sense. Because of that, the in-atmosphere rotation will continue to be an attractive option.

Going even further into the future, and larger in scale, there's no problem with in-atmosphere rotation scaling up. Drag forces will increase, but you can sheathe the rotating portion such that the friction is greatly minimized. All the while, you can keep a nice continuous hallway between the different sections at the hub. Sure, this is more complicated in a naive sense, but it takes care of the rotating joint problem, makes the shielding easier, and presents a lot of safety benefits.

So to be clear, by an in-atmosphere centrifuge, you mean having a centrifuge moving inside a pressurized module? That is indeed a design that sidesteps many engineering issues, especially considering that all components would be accessible in a shirt-sleeve environment. It could launch as parts inside a supply vehicle and be assembled in orbit so fairing size isn't an issue.
For a variable Isp spacecraft running at constant power and constant acceleration, the mass ratio is linear in delta-v.   Δv = ve0(MR-1). Or equivalently: Δv = vef PMF. Also, this is energy-optimal for a fixed delta-v and mass ratio.

Offline Coastal Ron

  • Senior Member
  • *****
  • Posts: 4380
  • I live... along the coast
  • Liked: 3029
  • Likes Given: 4047
In this view, the task of docking will be performed by a microgravity station which is, for all practical purposes, an autonomous station. I also find it logical to presume that most atmosphere, electrical, and heat management systems will be independent. There's not good economic motivation for connecting those systems between the rotating station and the non-rotating station.

The concept that I was outlining would be for a station (permanent or movable) that would spin and the residents would inhabit the outer regions of the "wheel".  The cargo operations might be autonomous - I hadn't really dealt with that yet, but the trend is to automate more and more, so they likely could be (should eliminate dock strikes).

The idea of creating artificial gravity within a pressure vessel is an interesting one, and it's good to explore what the potential solutions are.
If we don't continuously lower the cost to access space, how are we ever going to afford to expand humanity out into space?

Offline Rich_Zap

  • Member
  • Posts: 3
  • UK
  • Liked: 1
  • Likes Given: 0
I would say the main issue with placing the rotating section within a pressurised environment is the amount of energy you would need to constantly expend. If energy is considered unlimited then it's probably the best solution.

I like the idea of having a sort of rotational lift connecting between the static axle and the rotating section, it would effectively be a self contained pressurised module that would spin up or spin down and dock with either section to provide access.

Offline Hanelyp

  • Full Member
  • ***
  • Posts: 369
  • Liked: 65
  • Likes Given: 252
A centrifuge inside a pressurized module would have aerodynamic drag, and require a pressure shell unsegmented around the perimeter.  Aerodynamic drag is reducible by clean design and reducing centrifuge velocity.  Best case reducible to little more than viscous drag.  But a smaller edge velocity gives a higher rotation rate for the same pseudo-gravity level.

Offline Burninate

  • Full Member
  • ****
  • Posts: 1135
  • Liked: 352
  • Likes Given: 73
A centrifuge inside a pressurized module would have aerodynamic drag, and require a pressure shell unsegmented around the perimeter.  Aerodynamic drag is reducible by clean design and reducing centrifuge velocity.  Best case reducible to little more than viscous drag.  But a smaller edge velocity gives a higher rotation rate for the same pseudo-gravity level.

Yes, putting the rotation inside a static pressure vessel is a different kind of station that doesn't have the same sort of issues, but spends a hell of a lot of mass on the overhead of tracks & propulsion to get the things moving, and it's not functional until it's completely built, and there's no option to seal off segments for holes in the pressure seal.

It might end up being more practical, I can't tell.

I would say the main issue with placing the rotating section within a pressurised environment is the amount of energy you would need to constantly expend. If energy is considered unlimited then it's probably the best solution.

I like the idea of having a sort of rotational lift connecting between the static axle and the rotating section, it would effectively be a self contained pressurised module that would spin up or spin down and dock with either section to provide access.

*Energy* is relatively cheap.  But steady-state torque is not.  Anything that costs a certain amount of RCS mass per hour to correct attitude gets very costly, very fast.  The cheapest way to provide torque in LEO is some kind of gravity gradient scheme, but I'm not sure that's even practical in this context, and it certainly isn't BLEO.
« Last Edit: 03/17/2015 07:33 pm by Burninate »

Tags: