How reaction actually wheel works?

[Pawelk198604]

To the extent that I understand how the system works RCS that works on the principle of action and reaction, but I wonder as how moving satellites without RCS, only on reaction wheel?

[RainDreamer]

Something about conservation of angular momentum I believe. Someone will be able to explain it much better than me.

[Kerbart]

Wheel starts spinning one way, satellite starts spinning the other way.

The problem is that, if you used a reaction wheel to [I]stop[/I] movement, you have to keep it spinning. Because it is the [I]change[/I] in rotation speed that causes the ship to (start to) rotate.

Which is why once in a while you have to use RCS in real life to “reset” the reaction wheels.

[More Boosters]

Something about the wheel having a native rotation speed which makes the satellite rotate itself when that native rotation speed is changed. Don't quote me on that, I need an ELI5 myself.

[magnemoe]

Then you rotate the wheel the rest of the spaceship will rotate in the other direction and you turn. then you then brake the wheel you will also stop the ships rotation and you end up pointing in an new direction.

you can not move just change the orientation, but you can do this without using fuel.
Having three wheels in three directions you can rotate freely.
This is very nice for space telescopes as you want to look at a lot of different things.

[sal_vager]

Wikipedia explains it fairly well.

[U][URL="https://en.wikipedia.org/wiki/Reaction_wheel"]Reaction wheel.[/URL][/U]

[U][URL="https://en.wikipedia.org/wiki/Control_moment_gyroscope"]Control moment gyroscope.[/URL][/U]
[U][URL="https://en.wikipedia.org/wiki/Angular_momentum#Conservation_of_angular_momentum"]
Conservation of angular momentum.[/URL][/U]

[Hcube]

Basically : m= mass of the reaction wheel. r=ray of the wheel. V=angular velocity.

Angular momentum=(m*r²*V)/2

It's just like p=mv, except here the wheel is a disc and not a point so you have to integrate it, so in the end you divide by 2.

So the angular momentum is conserved, so if a heavy wheel inside turns in a way, the test of the spacecraft will rotate the other way.

[Frozen_Heart]

If its anything like a gyroscope, then magic.

[prophet_01]

I always wondered what happens if you do the following:

You have 2 reaction wheels next to eachother and let them spin up (x-axis), in opposite directions so they cancel each other out relative to the sattelite. The wheels would build up a maximum momentum while the sattelite remains stable, right?
Then you stop one wheel. As a result the sattelite rotates on the x-axis. The second wheel however just continues to spin.
Now you 'rotate' the second wheel by 180° (y-axis) so it ends up with the same orientation as the first wheel. You stop the second wheel and rotate the sattelite further on the x-axis.

To counter the rotation of the second wheel on the y-axis, you could use a dummy weight which rotates 180° the other way around.


Would that work? Or am I missing something? Could that reduce the amount of RCS used on a sattelite?

[Hcube]

[quote name='prophet_01']I always wondered what happens if you do the following:

You have 2 reaction wheels next to eachother and let them spin up (x-axis), in opposite directions so they cancel each other out relative to the sattelite. The wheels would build up a maximum momentum while the sattelite remains stable, right?
Then you stop one wheel. As a result the sattelite rotates on the x-axis. The second wheel however just continues to spin.
Now you 'rotate' the second wheel by 180° (y-axis) so it ends up with the same orientation as the first wheel. You stop the second wheel and rotate the sattelite further on the x-axis.

To counter the rotation of the second wheel on the y-axis, you could use a dummy weight which rotates 180° the other way around.


Would that work? Or am I missing something? Could that reduce the amount of RCS used on a sattelite?[/QUOTE]

I don't understand what you're trying to achieve here... How is your proposal different from a regular reaction wheel ? (I don't really figure out well what you mean ^^ maybe an ugly paint scheme would help :) )

[Gaarst]

[quote name='prophet_01']I always wondered what happens if you do the following:

You have 2 reaction wheels next to eachother and let them spin up (x-axis), in opposite directions so they cancel each other out relative to the sattelite. The wheels would build up a maximum momentum while the sattelite remains stable, right?
Then you stop one wheel. As a result the sattelite rotates on the x-axis. The second wheel however just continues to spin.
Now you 'rotate' the second wheel by 180° (y-axis) so it ends up with the same orientation as the first wheel. You stop the second wheel and rotate the sattelite further on the x-axis.

To counter the rotation of the second wheel on the y-axis, you could use a dummy weight which rotates 180° the other way around.


Would that work? Or am I missing something? Could that reduce the amount of RCS used on a sattelite?[/QUOTE]

I'm not sure I understand either, but because of conservation of angular momentum, if you have no external torque acting on your ship, you're going to end up with the same result, no matter how much you spin your wheels before or the number of times you change their rotation direction.
Only thing that matters is the total angular momentum: as long as your your "final state" has the same angular momentum as the initial one, then it works; otherwise it doesn't.

[Shpaget]

No, that wouldn't work.
You can't cheat the law of conservation of momentum. Precession would screw you over.

[Jovus]

One thing I don't understand still is why we don't use friction to dump angular momentum from our reaction wheels. Yes, conservation of angular momentum, but that only applies to conservative forces. Friction is not a conservative force.

[prophet_01]

Yep. That would definitely violate it. Tbh, I didn't know about conservation of momentum when it comes to rotation. I was only aware of the conservation of linear momentum.
->Yeah I'm a scrub :P
Anyway thx for clearing that up and providing the key phrase to help me fill that particular gap of my knowledge :)

Will still take me some time to figure out the particular mistake with that thought experiment though. Edited November 23, 2015 by prophet_01

[Shpaget]

[quote name='Jovus']One thing I don't understand still is why we don't use friction to dump angular momentum from our reaction wheels. Yes, conservation of angular momentum, but that only applies to conservative forces. Friction is not a conservative force.[/QUOTE]

Propose a model that would produce a net change in angular momentum.

[K^2]

[quote name='llanthas']Hard to describe without graphs, but picture how 2 spinning bicycle wheels are able to stabilize the much-heavier rider above them once they get moving.[/QUOTE]
Angular momentum of bicycle wheels is much too small to contribute significantly to bicycle stability. I know it is an example used frequently, but it is a poor one.

But yeah, reaction wheels have to store a lot of angular momentum to do their job. So precession torque resulting from trying to flip one will be quite great.

[Jovus]

[quote name='Shpaget']Propose a model that would produce a net change in angular momentum.[/QUOTE]

Sure.

A rod is rotating with a certain frequency omega in a vacuum chamber. Air is introduced to the chamber. As a (very!) rough approximation, air resistance is modeled as F = bv + cv^2

For friction specifically; A wheel is rotating with a certain angular momentum L, frequency omega, mass m. Its bearings are frictionless, it's in a vacuum, etc. Then sand gets in the bearings, so they suddenly exert a frictional force modeled by F = kv, where v = omega x r, the radius of the wheel. Or apply brake pads to the outside, sure. Yes, I'm aware the pads will pick up some momentum, but as far as I can see angular momentum is only conserved in this case if you assume the condition of no slippage (which is the angular equivalent of an elastic collision, for these purposes).

As a simple proof that we already know in classical mechanics that angular momentum is not conserved when subjected to nonconservative forces, put a rotating electron in a magnetic field. The angular velocity changes. (In this case it doesn't change in magnitude, but it does change in direction.)

[Hcube]

[quote name='Shpaget']No, that wouldn't work.
You can't cheat the law of conservation of momentum. Precession would screw you over.[/QUOTE]

I don't think precession applies here since you're in zero-g.
You need a force field to create precession...

[Shpaget]

Torque induce precession of a flywheel does not depend on gravity.

[cfds]

[quote name='Jovus']Sure.

A rod is rotating with a certain frequency omega in a vacuum chamber. Air is introduced to the chamber. As a (very!) rough approximation, air resistance is modeled as F = bv + cv^2

For friction specifically; A wheel is rotating with a certain angular momentum L, frequency omega, mass m. Its bearings are frictionless, it's in a vacuum, etc. Then sand gets in the bearings, so they suddenly exert a frictional force modeled by F = kv, where v = omega x r, the radius of the wheel. Or apply brake pads to the outside, sure. Yes, I'm aware the pads will pick up some momentum, but as far as I can see angular momentum is only conserved in this case if you assume the condition of no slippage (which is the angular equivalent of an elastic collision, for these purposes).

As a simple proof that we already know in classical mechanics that angular momentum is not conserved when subjected to nonconservative forces, put a rotating electron in a magnetic field. The angular velocity changes. (In this case it doesn't change in magnitude, but it does change in direction.)[/QUOTE]

Non conservative forces mean that the energy (linear or angular) is not conserved, it does not help with momentum. A totally inelastic collision still preserves (linear) momentum.
Physical background: Conservation of linear or angular momentum is a consequence of invariance under translation or rotation respectively while conservation of energy is a consequence of invariance under shifts in time. Friction does not destroy the invariance under spatial translations and rotations, but that under translations in time.

[Pawelk198604]

[quote name='Kerbart']Wheel starts spinning one way, satellite starts spinning the other way.

The problem is that, if you used a reaction wheel to [I]stop[/I] movement, you have to keep it spinning. Because it is the [I]change[/I] in rotation speed that causes the ship to (start to) rotate.

Which is why once in a while you have to use RCS in real life to “reset” the reaction wheels.[/QUOTE]

Is that way American Explorer 1 satellite spinning before launch?

[gooddog15]

[quote name='Pawelk198604']Is that way American Explorer 1 satellite spinning before launch?[/QUOTE]

Nope. Explorer 1's 2nd, 3rd, and 4th stages were spin stablized, in which their nozzles were pointed slightly to the left or right to induce rotation.

[Redjoker]

Reaction wheels work because angular momentum has to be conserved. You start out with a space craft with nothing rotating so your momentum is zero. Since no external forces, at least in our model, are acting on it, the final momentum also has to be zero. We begin spinning up a disk inside the space craft clockwise, so that it has momentum in the clockwise direction. Since we have clockwise momentum, but our final momentum has to be zero, the rest of the craft will begin to rotate counter-clockwise.

There are a couple of ways you can demonstrate this. First you will need a disk or wheel, for example from a bicycle, with an axle that you can hold on to and a chair or stool that can freely rotate.

1. Sit on the stool and keep your feet on the ground. Hold the bicycle wheel, so that the wheel is parallel with the ground have someone spin it for you as fast as possible. Lift your feet off he ground and then flip the bicycle wheel over. What happens is really cool. Here is a video, but I suggest you try it yourself: [url]https://youtu.be/UZlW1a63KZs?t=50[/url]

2. For this one you will something with a rotating wheel that you can hold while hold the bicycle wheel. Sit in the stool with your feet off the ground. Hold the bicycle wheel in one hand the the object in the other. Press the rotating object against the wheel, so that the wheel will begin to rotate. When I did this in class, we used a drill attached to what appeared to be a hole saw with it's blades ground down. This demonstrates how a reaction wheel works.

[Jovus]

[quote name='cfds']A totally inelastic collision still preserves (linear) momentum.[/QUOTE]

Yes. Sorry, I don't know what I was thinking regarding inelastic collisions.

However, the rest of what you said doesn't make sense. Let's try another way.

Total energy of a system of two particles is the particles' kinetic energy plus their potential energy: E = T + U

Let's restrict motion to a path that doesn't change the potential, which we can do without violence to the idea. Therefore, and deltaE = deltaT

Now we clearly agree that certain non-conservative forces (e.g. friction, B) can take energy out of the system, viz. reduce T.

T = 1/2 (m[SUB]1[/SUB]v[SUB]1[/SUB][SUP]2[/SUP] + m[SUB]2[/SUB]v[SUB]2[/SUB][SUP]2[/SUP]) in our two-particle system. Friction acts to reduce velocities of the objects undergoing friction - in this case, v[SUB]1[/SUB] and v[SUB]2[/SUB]. If both of these velocities are reduced to 0 relative to our origin, then T = 0, which is perfectly possible.

But the total momentum P is defined such that P = m[SUB]1[/SUB]v[SUB]1 [/SUB]+ m[SUB]2[/SUB]v[SUB]2[/SUB]. So how is momentum conserved if both v[SUB]1[/SUB] and v[SUB]2[/SUB] can decrease?

[Shpaget]

But that's outside force.