1. Limited time only! Sign up for a free 30min personal tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

B Gyroscopes on spaceships

  1. Dec 20, 2017 #1

    DaveC426913

    User Avatar
    Gold Member

    I'm not looking for fiction; I'm looking for physics.

    Assumptions:
    - An engine efficient enough that you can afford to thrust for very long periods - maybe fusion? (OK, that's fiction)
    - A lot of maneuvering (eg. you could thrust until turnaround point and then decelerate)
    - With a lot of maneuvering and a single main engine it would be would be better to rotate the ship with gyros rather than retros.

    What would the properties of the gyro(s) need to be?

    Does it need to be at the centre of mass?
    Can you have more than one?
    Can multiple gyros be away from the centre of mass?
    Clearly, the more spherical the ship, the better use it could make of the momentum, so an ideal ship might be as spherical as practical. (just a supposition)
    Is there a ratio of how large and how fast the gyro must be to turn the ship efficiently?
    Would the gyro spin down quickly under all that applied torque? (You could always bleed off some thrust to power a generator that would keep it spun up.)
     
  2. jcsd
  3. Dec 20, 2017 #2

    FactChecker

    User Avatar
    Science Advisor
    Gold Member
    2017 Award

  4. Dec 21, 2017 #3

    DaveC426913

    User Avatar
    Gold Member

    Fascinating. Thank you! That answers several of my questions - and raises a few more.
    I assume current gyros, such as those on the ISS provide a gradual, gentle rotation. I'm imagining a spaceship that can/needs to turn fast, like to do evasive maneuvers (which is why I envision a nearly spherical ship a la ballet pirouette analogy).

    The reason I wanted to understand if the gyro(s) need to be at the CoM, is because it would be better to put the pilot there, so that vertigo is minimized when under max rotation.

    How big a gyro - as a ratio of ship mass - would one need, to spin the ship, say, 90 degrees in one second? Obviously that depends on how much momentum the gyro could store (which be limited in top speed, because of materials). I guess such a set of gyros (sound like you'd need at least 4) would take up a fair chunk of ship's volume/mass.
     
  5. Dec 21, 2017 #4

    jbriggs444

    User Avatar
    Science Advisor

    Why should it matter where the gyros are? The ship rotates around the center of mass regardless of where the gyros are placed.
     
  6. Dec 21, 2017 #5

    FactChecker

    User Avatar
    Science Advisor
    Gold Member
    2017 Award

    I doubt that gyros will help in evasive maneuvering in outer space. They will rotate but not redirect motion. So the spaceship will just wiggle around without evading anything. It's not like an airplane, where rolling redirects the lift force vector and changes the direction of motion. Now, if you are talking about turning the ship to point a rocket nozzle or a gun, that is something gyros might help with. But there are probably better ways to accomplish that. During significant maneuvering, gyros have the problem of reduced control authority due to saturation.
     
    Last edited: Dec 21, 2017
  7. Dec 21, 2017 #6

    mfb

    User Avatar
    2017 Award

    Staff: Mentor

    If you want fast changes in thrust direction, rotate your engine, not the whole spacecraft. Rotating the whole spacecraft structure by 90 degrees in one second is unfeasible for any realistic spacecraft large enough for humans. I don't see any scenario where this would be useful either.

    Back of the envelope estimate: Dragon 2 has a dry mass of 6.5 tons, CST-100 Starliner seems to have a bit more but it is hard to find numbers, Orion will have a dry mass of 9.3 tons. These are basically the minimal viable capsules to support humans in space for a week, or a smaller crew for a few weeks if absolutely necessary. They don't have a big engine. Let's say your spacecraft has a mass of 10 tons (excluding gyros) at typically 2 meters away from the center of mass for a given axis of rotation. That gives it a moment of inertia of 40,000 kg m2. For full three axis control, you need three gyros. Let's make them huge at 1 ton each as ring with 4 meter diameter (basically filling a whole wall of the capsule). That gives them 4,000 kg m2, or 1/12 of the remaining spacecraft which is now at 48,000 kg m2 if we rotate one gyro and the other two are dead mass.

    The minimal torque approach for a 90 degree = pi/2 spacecraft rotation in one second is a constant angular acceleration of 2pi/s2 for half a second and deceleration of the same amount afterwards. That needs a torque of 48,000 kg m2 * 2pi/s2 = 300,000 Nm, a peak differential angular velocity of 13 pi/s and a peak power of 12 MW. Going from 0 to 12 MW in half a second, then from 12 MW to -12 MW in "zero time" and back to 0 in half a second will need a really unrealistic motor, a huge battery array and various other oversized components. You would probably need most of the spacecraft mass just for motors and powering infrastructure.

    Making your spacecraft larger just makes it worse. Scaling up all masses by a constant factor is fine, but scaling up the size makes the required torque and power increase with the square of the length, with nothing to compensate it.
     
  8. Dec 22, 2017 #7

    DaveC426913

    User Avatar
    Gold Member

    Which is why you have the main engine burning continuously.

    Yeah, I read about saturation. Fascinating. It's why you'd need at least 4 gyros, like the ISS has. That side-steps the saturation problem:(https://en.wikipedia.org/wiki/Control_moment_gyroscope#International_Space_Station)
     
    Last edited: Dec 22, 2017
  9. Dec 22, 2017 #8

    DaveC426913

    User Avatar
    Gold Member

    Rotating the engine while thrusting doesn't in-and-of-itself move the craft from its trajectory. Mostly what it does is spin the craft. You still have to return the engine back to straight so that the thrust is now oblique to the initial trajectory. Then you can move off-course.

    Whereas, if you rotate the whole craft, you accomplish both steps in one, because the thrust is aligned with the CoM so it immediately contributes to oblique acceleration.

    OK, I guess I'm getting a bit hair-splitting now in terms of usefulness of the idea. Considering what you wrote below, I can see the benefits of gyro-rotation disappearing before my eyes:


     
    Last edited: Dec 22, 2017
  10. Dec 22, 2017 #9

    DaveC426913

    User Avatar
    Gold Member

    Yes. That's what I wanted to determine. I wasn't sure if placing the gyro(s) off-centre from the CoM would affect their usefulness at these kinds of power. It'd be fine for a slow rotation, but I was thinking a high rotation might essentially waste some of their angular momentum, as well as place undue lateral stress on components.
     
  11. Dec 22, 2017 #10

    jbriggs444

    User Avatar
    Science Advisor

    To the extent that the gyros are more dense than the rest of the ship, moving them away from the center of gravity will tend to increase the moment of inertia of the vessel. The angular momentum delivered is fixed regardless.
     
  12. Dec 22, 2017 #11

    FactChecker

    User Avatar
    Science Advisor
    Gold Member
    2017 Award

    Sorry. Right. I forgot about that from your original post.
    I don't know how that works. I don't think it side-steps the problem. The article says that the ISS uses its high gravity-gradient torque to desaturate the CMGs. That may take a long time. I am not sure how well the small, mild maneuvers of the ISS would compare with "evasive" maneuvers.
     
  13. Dec 22, 2017 #12

    mfb

    User Avatar
    2017 Award

    Staff: Mentor

    It does both. For the center of mass motion it doesn't matter where your engine is, only the thrust direction matters. If you rotate the engine at its current location (instead of rotating it around the spacecraft, what I was imagining) then the engine won't be able to keep firing in the same direction quickly based on the rotation of the spacecraft, but that is a different issue.
     
  14. Dec 22, 2017 #13

    DaveC426913

    User Avatar
    Gold Member

    Extrapolating further into the realm of fictional tactics, I guess what you might do is "burn" all your ... uh ... 1/saturation ... in the emergency evasion, then you'd have to desatch them once the danger has passed.
     
    Last edited: Dec 22, 2017
  15. Dec 23, 2017 #14

    Filip Larsen

    User Avatar
    Gold Member

    If you are talking about evasive manoeuvrers (i.e. change of trajectory) then I assume fairly high thrust acceleration is available. If so, then there should also be a fair bit of torque available to change orientation via "traditional" thrust vectoring and it may be useful for you to relate performance of CMG with that.

    If the spacecraft already can accelerate linearly with acceleration ##a## and has moment of inertia ##I = k m r^2## for some value of ##k## then via thrust vectoring I get (on my supply of envelope backsides) that the spacecraft also has access to a maximum angular acceleration of $$\dot{\omega}_m = \frac{a}{k r}.$$ This can be compared to the required average angular acceleration ##\dot{\omega}_r## needed to change the orientation of the spacecraft through the angle ##\Delta\psi## in the time ##\Delta t## of $$\dot{\omega}_r = 4\frac{\Delta\psi}{\Delta t^2}.$$

    For instance, to change orientation 90 degree in 1 second of a spacecraft with ##k = \frac{1}{2}## then it needs ##\dot{\omega} \approx 2\pi## which for a 3 g thrust is possible up until around 9 m radius. For a 10 g thrust that would increase maximum radius to around 30 m.

    [Edit: corrected wrong power for radius r. I blame the poor state of my envelope backsides.]
     
    Last edited: Dec 23, 2017
  16. Dec 26, 2017 #15
    In terms of evasive maneuvers, you would do better with a ship that blows itself apart with the pieces still connected by ropes, and then reassembles by pulling on the ropes.

    Niket Patwardhan
     
  17. Jan 2, 2018 #16

    DaveC426913

    User Avatar
    Gold Member

    You;re supposed to add a :biggrin: after a post like that.
     
  18. Feb 22, 2018 #17
    Was interested in the bit above about rotating the engine to spin the craft. If for example there were three engines in the 'rear' of the craft to provide thrust and only one of the outside engines was thrusting would that spin the craft. I'm more thinking of moving in the direction of travel in 'free' space rather than in orbit.
     
  19. Feb 22, 2018 #18

    Bandersnatch

    User Avatar
    Science Advisor
    Gold Member

    Yes, it would. But it would also move forward. So it'd go in circles rather than just spin.
     
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted