Space station artificial gravity - how to spin up to speed?

[Al_]

If a space station has artificial gravity created by spinning, how can it best be spun up to speed? Little attitude rockets could do it, but they would use up fuel, and limit your ability to change the spin rate in future. What if you had an external wheel that you spin up very fast in the opposite direction? If you did this, would it cancel out the gyroscope effect because the total angular momentum of the whole would be zero, and make it easier to orient the station with respect to the Sun, spacecraft , etc.?

 

[Drakkith]

What if you had an external wheel that you spin up very fast in the opposite direction? If you did this, would it cancel out the gyroscope effect because the total angular momentum of the whole would be zero, and make it easier to orient the station with respect to the Sun, spacecraft , etc.?

Sure. This is known as a reaction wheel (though it isn't "external" since it would be inside the station). You might be able to spin one part of the station in one direction, and another part of the station in the opposite direction, with both parts exerting torque on each other to spin up to speed.

 

[Al_]

Thanks. I am thinking that the wheel could be mounted outside the station on an axle.
Maybe it could contain some kind of payload or machinery, to avoid having to carry extra mass.
But it would be difficult to connect to stuff spinning round that fast.
Maybe magnetic bearings could be used to eliminate vibration?

 

[Drakkith]

Thanks. I am thinking that the wheel could be mounted outside the station on an axle.

You can mount it inside or outside. It doesn't really matter.

Maybe it could contain some kind of payload or machinery, to avoid having to carry extra mass.

Sure. You can make the reaction wheel a large module or room that is spun up in the opposite direction as the rest of the station. This is similar to what I said above with having separate parts of the station spinning in different directions.

But it would be difficult to connect to stuff spinning round that fast.
Maybe magnetic bearings could be used to eliminate vibration?

I don't know what kind of bearings they would use in space. Magnetic bearings would certainly be a possibility.

 

[sophiecentaur]

You can mount it inside or outside. It doesn't really matter.

I'm assuming that the reaction wheel diameter would have to be smaller than the ship of the suspension would get in the way of 'navigation' around the ship.
also, it would be a much more stable arrangement if the CMs of the annular station and the reaction wheel were both in the same plane and on the common axis. But would it be a problem to have the wheel at the centre of the station? If it were to be an embarrassment (vibration and safety issues), it could always be moved away using its one on-board motors.
I was thinking about angular momentum conservation and the big ratio between the main station diameter and that of the reaction wheel. The mass of the reaction wheel could be considered as 'lost' to the system in the same way that rocket propellant mass needs to be lost. The rotation rate would need to faster than the ship's rotation by a factor of the ratio of the Moments of Inertia of the two. Double whammy here because the inertia wheel has small diameter and it's a lot lower mass. What sort of angular velocity could be considered? It wouldn't be too hard to do some rough sums and compare the results with just having tangential thrusters. Of course, if you kept the reaction wheel on board, you could always slow the station down again when you wanted to, for free, by locking the two together again. (Regenerative braking even!)

 

[mfb]

A counter-rotating wheel would need a lot of mass. If it is dead mass, make it as large as possible to reduce its mass. That might limit spacecraft docking to one side. If it is a symmetric space station, with the two parts rotating in opposite directions, all those problems do not arise. Moving between the space stations would need some rotating air-tight connector, but you want such a connector for spacecraft docking anyway.

 

[sophiecentaur]

A counter-rotating wheel would need a lot of mass. If it is dead mass, make it as large as possible to reduce its mass. That might limit spacecraft docking to one side. If it is a symmetric space station, with the two parts rotating in opposite directions, all those problems do not arise. Moving between the space stations would need some rotating air-tight connector, but you want such a connector for spacecraft docking anyway.

What would be the advantage of this method, even if it could be made to work? The mass would need to be transported into orbit and even that would involve an Energy Premium.
The tangential force needed could be very small if the spin up time could be made a matter of days or weeks. That would open up the possibility of Ion or Photon drive, with a PV (Solar) energy source.

 

[anorlunda]

For energy purposes, I like @Drakkith 's suggestion best. Spin parts of the station in different directions. No additional mass for a wheel.

I guess the disadvantage of that would be safety with objects or people crossing from one half to the other half. It would also be difficult to share plumbing and wiring between the parts.

 

[sophiecentaur]

For energy purposes, I like @Drakkith 's suggestion best. Spin parts of the station in different directions. No additional mass for a wheel.

I guess the disadvantage of that would be safety with objects or people crossing from one half to the other half. It would also be difficult to share plumbing and wiring between the parts.

That would be a better idea than having a reaction wheel. The energy to bring a large mass all that way would surely be much more than the rotational energy the mechanism would provide.
But quite high speeds would be involved. For 1g at the periphery and 100m radius, the speed would be over 30m/s (relative 60m/s). That could involve an exciting transition, going from one arm to the other - not something to do every day, even for a hard space crew member.
But the spin up energy would not be (particularly) high so where's the return for such inconvenience?

 

[anorlunda]

so where's the return for such inconvenience?

I thing when you consider all factors in addition to energy plus operational flexibility, the "old fashioned" thruster rockets are best.

 

[mfb]

What would be the advantage of this method, even if it could be made to work? The mass would need to be transported into orbit and even that would involve an Energy Premium.

Which mass? If it is a second part of the space station, there is no additional mass.

Spinning it up with thrust would be my suggestion as well, but that's not what OP was asking about.

That could involve an exciting transition, going from one arm to the other - not something to do every day, even for a hard space crew member.

Huh? You would cross at the center, where the two parts are connected. Relative velocity: Negligible.

Spinning a space station up to 30 m/s (velocity at the rim) needs ~1% of the space station mass as fuel for chemical rockets - affordable. Ion thrusters using 0.1% of the station mass would need 450 kJ/kg of station mass. At a power consumption of 0.1W/kg (ISS-like), it would need 2 months at 100% efficiency, and several months at a more realistic efficiency.

 

[sophiecentaur]

Which mass?

The mass of a reaction wheel.

Spinning it up with thrust would be my suggestion as well, but that's not what OP was asking about.

We have discussed what was suggested and, as we seem to have rejected it, we are discussing alternatives. Better than just saying 'no can do', I should have thought

Huh? You would cross at the center, where the two parts are connected. Relative velocity: Negligible.

Yes that makes sense. I was sort of assuming that the hub would contain the reactor and would best be avoided. But the docking of supply craft would have to be at the centre in any case. It would, of course, mean a spot of zero gravity during the transfer but I guess you could get used to anything like that pretty soon.

At a power consumption of 0.1W/kg (ISS-like), it would need 2 months at 100% efficiency, and several months at a more realistic efficiency.

Are you ignoring the fact that the rotational energy has to be supplied from some source, however you do the job? We'd need to discuss the relative efficiency of the methods and I was assuming that time would not be a high priority. Once the main masses in the station were in place, the spin up could start. Components would be 'lowered' from the central dock. Alternatively, the docking could be on the periphery - a sort of tangential approach, which could be done easily with computerised navigation. The heavy bits could be 'bringing their own' momentum with them to the periphery. (I'd need to think that through as there would be the question of balancing the supplied loads etc). Quite a sexy idea though, don't you think?

 

[mfb]

Are you ignoring the fact that the rotational energy has to be supplied from some source, however you do the job?

I am not ignoring it. I tried to find some realistic estimate for the power such a station would have, and used the ISS power density. It does not matter where the energy comes from - solar power, nuclear power, and even every hypothetical future power source will all lead to the same results. If you make the power source appropriate for the station size, you can spin it up with 0.1% of the station mass within a few months. Take more propellant and you can do it faster.

Docking at the periphery would be like docking a rocket (coming from below) to a ceiling on Earth. Possible, but unnecessarily complicated and risky if you can dock at a zero-g section.

Humans can easily handle short periods of zero-g. They will have that in the approaching spacecraft as well. And machines are not a problem either. If they have to run in zero-g, they need special designs, otherwise just take care of all liquids and don't switch them on before they have their proper acceleration again.

 

[sophiecentaur]

I am not ignoring it. I tried to find some realistic estimate for the power such a station would have, and used the ISS power density. It does not matter where the energy comes from - solar power, nuclear power, and even every hypothetical future power source will all lead to the same results. If you make the power source appropriate for the station size, you can spin it up with 0.1% of the station mass within a few months. Take more propellant and you can do it faster.

Docking at the periphery would be like docking a rocket (coming from below) to a ceiling on Earth. Possible, but unnecessarily complicated and risky if you can dock at a zero-g section.

Humans can easily handle short periods of zero-g. They will have that in the approaching spacecraft anyway. And machines can handle it anyway.

That's when you have to eject a mass. Two rotating halves avoid the need to consume mass but it may not be very relevant as it's only a one-off cost.

You are spoiling my fun now. But would it be that risky? It could be a fail safe situation - perhaps safer, actually than aiming for the middle of the wheel. Complexity is a relative thing when you have the use of machine intelligence. The ship and wheel would always need a near-zero relative velocity for any docking and it would even be possible to use a tether (just thought of that) to grab the craft as it coasts slowly past. I have done the equivalent many times, picking up a mooring with wind and tide to deal with. Once you are hooked on, you can do things at your leisure. When you think about landing on an aircraft carrier in bad weather, the space docking is a relative piece of cake, I would reckon.

 

[mfb]

Reaching a precise velocity is easier if you don't have your rocket running at full thrust. Tethers would still need some connection procedure at low relative velocities.

Docking with high thrust also makes every previous mass saving effort irrelevant. A single second of 1g thrust needs ~0.3% of the spacecraft mass. And you cannot dock in a single second. The most fuel-efficient procedure would start with a high-speed (30m/s) approach towards the space station - which is dangerous already, get your course wrong and you crash into the station. Afterwards, fire your main thrusters and use the RCS system to fly in a curve together with the station until you are docked. Your spacecraft suddenly needs a strong engine, something spacecraft s rarely need. It also means docking is limited to a single place - opposite of the main engine. The space station would need additional structural strength at the connection points, would need some movable masses to keep the center of gravity balanced, and the connection points would be under strong tension. The spacecraft would have to be designed to handle tension (instead of just compression, as normally).
No. Just no.

 

[sophiecentaur]

Reaching a precise velocity is easier if you don't have your rocket running at full thrust. Tethers would still need some connection procedure at low relative velocities.

Docking with high thrust also makes every previous mass saving effort irrelevant. A single second of 1g thrust needs ~0.3% of the spacecraft mass. And you cannot dock in a single second. The most fuel-efficient procedure would start with a high-speed (30m/s) approach towards the space station - which is dangerous already, get your course wrong and you crash into the station. Afterwards, fire your main thrusters and use the RCS system to fly in a curve together with the station until you are docked. Your spacecraft suddenly needs a strong engine, something spacecraft s rarely need. It also means docking is limited to a single place - opposite of the main engine. The space station would need additional structural strength at the connection points, would need some movable masses to keep the center of gravity balanced, and the connection points would be under strong tension. The spacecraft would have to be designed to handle tension (instead of just compression, as normally).
No. Just no.

I think you are misunderstanding my idea here. The visiting craft would not necessarily be contributing to the station's existing angular momentum; no acceleration of the wheel would be involved. By docking at the periphery, it would avoid the need for the station to supply the added mass with any angular momentum. Even with a zero velocity docking, the system wins out compared with docking at the centre and then being taken out to the periphery, which would require angular momentum to be supplied (to keep the rotation rate constant)
I appreciate that some structural strength would be needed in order to deal with the balance problem. But the balance problem would apply to the delivered payload, wherever it landed. Two docking platforms, one on each side of the station could take care of imbalance. It wouldn't be too much of a stretch to arrange for the arriving craft to have two halves which could dock, one at a time awn the two opposite sides of the station - just waiting for a half revolution so they both supply the same momentum contribution. Balance would be a major consideration throughout the operation of the station, particularly during construction, so I can't see that a relative light ship arriving would be a major problem. Would a metre or so, of wobble, be a serious issue if a ship with 1% if the total mass were to land on the edge? I don't think so. I was suggesting that every consitiuent part of the station could be delivered to the station in this manner. The station could expand for the initial framwork construction phase.

 

[mfb]

The visiting craft would not necessarily be contributing to the station's existing angular momentum

I never suggested that. I just outlined the easiest possible procedure to dock at the rotating periphery. The spacecraft has to co-rotate, and as long as it is not docked that involves firing thrusters.

If you leave with the same mass as you join the spacecraft , docking at the center doesn't change the rotation speed. Once the station is built and spun up, there is nothing that would constantly increase the mass of the station.

Delivered payload is lighter than a spacecraft .

If you absolutely have to add significant mass with each arriving spacecraft , let them also add some fuel to the station that is used to keep it spinning. That will need less fuel than docking to the moving outside.

It wouldn't be too much of a stretch to arrange for the arriving craft to have two halves which could dock

That sounds really complicated. And I still don't see any advantage.

 

[sophiecentaur]

I still don't see any advantage.

Yep, I can tell.
Let me justify the tangential docking.
It is energy neutral. The relative speed (relative to the periphery) can be near zero as the ship is approaching the CM of the station. No 'boost' needed, to catch up the rim. Navigation can bring you to the ship at any velocity you want. When the ship leaves, it is as if it had never stopped off; it leaves at the tangential velocity and just carries on forward. The approach timing would need to be good but I really can't see that being a problem in the future (or even now). All that video game palaver of docking would not be needed. As with all other space manoeuvres, it's actually more straightforward than where wind a waves are present. The tangential speed is actually negligible compared with the orbital speed so you have pretty well exactly the same navigational issues in order to dock anywhere on the station. You are probably right that the energy involved is low enough for the fuel issue to be insignificant, though.
If all stores, construction materials and visiting crew are to be brought to the hub, first, everything would need to be carried on an almost superfluous full capacity lift arrangement. Goods and staff arriving at the periphery would be carried round on rails (needed with both systems).
The ring structure of the station would already be strong as it has to cope with things weighing what they do on Earth. But a resilient capture / coupling would be perfect feasible for coping with minor speed adjustments on contact. I am assuming that the station would be far bigger than ISS (that's obvious if we're talking artificial gravity) so things would be very different. It is worth while exploring all possibilities for a revolutionary (no pun intended) system.

 

[anorlunda]

I tried to find some realistic estimate for the power such a station would have, and used the ISS power density. It does not matter where the energy comes from

Thinking back to the OP, I think what he had in mind was spinning up using a replaceable power source (like solar) as opposed to using irreplaceable reaction mass in a rocket. In other words, not all power sources are equal in features other than energy.

Rockets do use mass that needs an outside source to replace, but they are very versatile. Consider the case when a rotating station developed a wobble. That would be hard to correct with a torque at the central hub, whereas rockets could be aimed in any direction to correct almost anything.

 

[sophiecentaur]

Thinking back to the OP, I think what he had in mind was spinning up using a replaceable power source (like solar) as opposed to using irreplaceable reaction mass in a rocket. In other words, not all power sources are equal in features other than energy.

Rockets do use mass that needs an outside source to replace, but they are very versatile. Consider the case when a rotating station developed a wobble. That would be hard to correct with a torque at the central hub, whereas rockets could be aimed in any direction to correct almost anything.

Just like in the tyre shop!
Absolutely

 

[mfb]

Energy neutral with respect to what? And what about the energy used by the spacecraft ?

The relative speed (relative to the periphery) can be near zero as the ship is approaching the CM of the station.

How would the ship approach the center of mass if it docks at the outside?

Navigation can bring you to the ship at any velocity you want.

Only thrust can do that.

The approach timing would need to be good but I really can't see that being a problem in the future (or even now).

Docking to the ISS takes something like an hour - the approaching spacecraft approaches the ISS slowly, and always in a way that a sudden failure of a booster doesn't make the spacecraft crash into the ISS. An hour for the last 100 meters is centimeters per second. You want relative velocities a factor 1000 times faster, with basically instantaneous docking times, establishing a solid connection in fractions of a second? Good luck. The orbital speed does not matter, relative speeds and accelerations between spacecraft do.

Thinking back to the OP, I think what he had in mind was spinning up using a replaceable power source (like solar) as opposed to using irreplaceable reaction mass in a rocket. In other words, not all power sources are equal in features other than energy.

It doesn't make sense to compare an electricity source to propellant. What are you going to do with electricity? Accelerate propellant. That's the ion thruster case I calculated. You can also go for a photon rocket, but that is ridiculously ineffective, especially if you want a final speed of 30m/s.
The chemical rocket doesn't need an electricity source, but leads to slower exhaust velocities.

 

[sophiecentaur]

You want relative velocities a factor 1000 times faster,

Not at all. The relative velocity of the ship and the rim is nearly zero. The trick is to rendezvous the ship with the right section of the rim. They will both be traveling at the same velocity. For a large structure like an annular station the problems and techniques would be totally different from those in the ISS.
You surely can't be claiming that the only way to dock in space is to use twenty year old technology.

The orbital speed does not matter, relative speeds and accelerations between spacecraft do.

It's the relative velocity between the ship and the docking bay that counts. The only difference is the short time window but there is no reason why it would be as short as you imply or as difficult to make a suitable tethering arrangement. I really can't think that you have understood what I am describing. You have referred to boosters and huge velocity differences when neither are involved. Imagine riding a motorcycle towards a merry-go-round and holding hands with a passenger on one of the outer horses. Your speeds are identical for some while and you make contact when the velocities are both the same. No short term speed adjustment needed, Furthermore, no need to avoid a bit of speed and positional when the docking is side by side and not head-on. It could involve an 'arrestor wire' as on an aircraft carrier but the speeds involved would be so much lower.
The reason that the ISS docking procedure is so 'careful' is that the method has no slop and no fail safe. I wonder how long ago that basic system was developed and what the level of available control system complexity was available.
As a matter of fact, this docking time could be as long as you wanted - just waiting until the intercept time and place were planned right as the ship approaches the station. In fact, the whole of the periphery of the station could have a rail running round it which could be grasped (at very low relative speed) as the ship comes into nudging contact. The ship could then proceed to the docking port, moving along the rail. (like an upside down aircraft runway) Someone earlier used the description 'flying up to land on the ceiling' - precisely.

 

[mfb]

The relative velocity of the ship and the rim is nearly zero.

The relative velocity between spacecraft and center of mass of the space station is large. You'll fly past the space station at high speed, and you either dock quickly, or fire your engines for a while to follow its circular motion, or you won't dock. The slow approach currently used in spaceflight will not work if you dock to anything not in inertial motion.

You surely can't be claiming that the only way to dock in space is to use twenty year old technology.

I don't say that, but to me your approach sounds massively impractical, and without any advantage. Why risk an impact with a billion dollar space station at every docking attempt if you don't have to?

Imagine riding a motorcycle towards a merry-go-round and holding hands with a passenger on one of the outer horses. Your speeds are identical for some while and you make contact when the velocities are both the same.

Yes. You have a single second for that, if we make a docking mechanism with a meter tolerance. And if your spacecraft , approaching the station at 30 m/s, is off course by more than this meter, it might crash into the rim (slowly, fine, but even slow collisions can be bad). If it is off by multiple meters (remember the failing thruster scenario?), it can crash into the space station elsewhere and will do so really hard.

A rail would make slow docking possible. But then you lose any difference to docking at the center, apart from the need for an additional rail: you speed up the spacecraft by braking, slowing the station in the process. You could also install a rail going from the center to the rim. You can also put that rail in the interior. And there we have the "dock at the center" scenario.

 

[sophiecentaur]

either dock quickly,

In a short window in time- agreed. But that goes for most other space manoeuvres. With the exception of the existing docking procedures, all other space manoeuvres are 'dynamic' and time dependent (orbit change, takeoff and landing). You are approaching this massive station tangentially. The 'rail' could even be suspended on long legs and would be the only part of the station at risk. There is a strong parallel with an airport, with a runway and a terminal building. Arrivals and departures and a 'parking area' are all in the model.

or fire your engines for a while to follow its circular motion,

Yes. A radial thrust could do the job of an arrestor wire but it wouldn't be good value, I think. The force needed would be about equal to the Weight of the ship - much higher than normal manoeuvring thrust forces.

(remember the failing thruster scenario?)

What would failing thrusters do to this procedure? The ship would run past the station at 30m/s (which happens to be the landing speed of a jet airliner) but the velocity, relative to the nearest parts of the station (potential collision speed) would be much much lower. If you are comparing this with a landing on the hub then 'failing thrusters' could result in a collision with the central part of the station and no hope of a brushing encounter.

: you speed up the spacecraft by braking,

This, I don't follow. What braking is required? The ship latches on to the station at zero relative velocity. The centripetal force from the rail will constrain the ship to rotate (orbit). The relative momentum of the ship will be shared between it and the station so the station will speed up by 30 m/(M+m). When the ship takes off, its relative speed will be the same (30m/s) and the momentum of the station is back to what it was before the visit (except for the momentum of the payload, which the station will keep. The ship separates from the station without the use of thrusters.

You could also install a rail going from the center to the rim.

Have you thought this one through? If you have a radial approach, do you aim inwards or outwards? What possible advantage would that have? The rail would be traveling normal to the arrival direction. There are no advantages over a conventional docking at the centre and the list of possible disaster scenarios is even longer than your list for my proposal.

 

[A.T.]

I don't say that, but to me your approach sounds massively impractical, and without any advantage.

Not sure if it was mentioned: Docking at the rim would shift the centre of mass more, and thus create more wobble.

 

[sophiecentaur]

Not sure if it was mentioned: Docking at the rim would shift the centre of mass more, and thus create more wobble.

Yes. That's true but so will re arranging the cargo around the station (particularly) during construction. So some sort of shifting counterweight would be required in any case. But the wobble would not exactly be teeth jarring with a 20s rotate period. (For the 100m radius I have been working to).
Speaking of rotation, I wonder what an acceptable rotation rate would actually be for the crew. If they are looking out at the stars etc., would they cope with the constant drift. Would it actually be necessary to go for a much bigger radius?

 

[jbriggs444]

If they are looking out at the stars etc., would they cope with the constant drift. Would it actually be necessary to go for a much bigger radius?

Unless you are building a toy space station, putting windows in the walls is just silly. Increasing the radius because you put in windows is doubly so.

 

[Al_]

There are some good ideas being spun up here!
I'm tending toward the reaction wheel idea, thinking that it could be relatively low mass, very strong, rotating very fast, using solar power to accelerate. (Splitting the station into two sounds like an awkward arrangement. It might make the station's internal space less useful?)
I thought of another thing - spinning fast, it holds a great deal of energy. So why not have two wheels? Then, you can store energy in them, as well as spinning the station.
AND - here's a real advantage - you could halt the station's spin, transferrring all the kinetic energy to the wheels, then quickly spin it back up again using the stored energy, without having to wait for the solar power to provide the energy.
Using wheels to store energy is better than batteries cus they last indefinitely and can have very low internal loss during storage.

 

[Al_]

Or - if the space station is disc shaped, and has shielding made from something like Lunar dirt, then can that be mounted on a framework or cage, with the actual station inside it, and rotated in the opposite direction?

 

[mfb]

With the exception of the existing docking procedures, all other space manoeuvres are 'dynamic' and time dependent (orbit change, takeoff and landing).

Docking is also the only case where you are approaching a multi-billion dollar object where people are killed if you hit it.

What would failing thrusters do to this procedure? The ship would run past the station at 30m/s

Or crash into it, depending on its course.

If you are comparing this with a landing on the hub then 'failing thrusters' could result in a collision with the central part of the station and no hope of a brushing encounter.

With landing at the hub, you always have sufficient time to react: if thrust in some direction fails, you can turn the spacecraft and use different thrusters, for example.

What braking is required?

Spacecraft at rest relative to the center of mass attaches to rails (I think that was the point of the rail system?). Now the spacecraft is moving at 30m/s relative to the station rim and has to brake (rotating coordinate system)/accelerate (inertial coordinates). If you don't use the rails as rails there is no point in having rails.

There are no advantages over a conventional docking at the centre

Yes, that was my point. But there is no disaster scenario. The spacecraft makes a slow, careful approach, and docks at the center. You could move the spacecraft outwards but that doesn't make sense. It would still be better than docking at the rim - the same final result but without the risk.

A reaction wheel would need at least ~2% of the station mass at the same radius, about 1% at twice its radius: Its rim speed cannot exceed ~2km/s. Slower speeds are advisable, the 2 km/s are for the best disks we can manufacture today.