Reaction wheels and general relativity

[k00b]

ohhhh; thanks and thanks - it was OP (not the control system / game); that was the bit going over my head.

can anybody tell me how reaction wheels work IRL - i don't understand how they can orient mass in the vacuum of space ?

[Superfluous J]

31 minutes ago, k00b said:

can anybody tell me how reaction wheels work IRL - i don't understand how they can orient mass in the vacuum of space ?

Get yourself an office chair (or anything that will freely and easily spin around while you're on it) and an electric drill (or anything that can produce rotation). Sit in the chair. Aim the drill upwards and turn it on. Note you start rotating (if the drill is powerful enough to overcome friction). Stop the drill, note you stop rotating. Reverse the drill and turn it on again, and note you start rotating the other way.

Have a friend hold the chair and start the drill. Then have them let go, and you won't be rotating. Then stop the drill, and note you start rotating.

That's how reaction wheels work. except most ships have 3 of them (at least)

Edited October 8, 2019 by 5thHorseman

[Guest]

29 minutes ago, k00b said:

can anybody tell me how reaction wheels work IRL - i don't understand how they can orient mass in the vacuum of space ?

Consider a motor spinning a flywheel. When you spin up the motor, the flywheel starts to rotate, right? Because of the conservation of motion, that momentum has to go somewhere. If your flywheel is on a spacecraft, it goes into the spacecraft, which rotates in the opposite direction. That's your reaction wheel.

Real-life reaction wheels apply much, much, much smaller forces than KSP ones. Also if you keep turning the spacecraft in the same direction, eventually they'll be spinning so fast they risk their bearings. At that point they need to be braked, which of course spins the spacecraft again, and the craft will need to expend reaction mass and use its RCS thrusters to counter it. KSP doesn't model this because it wouldn't be fun.

[AHHans]

1 hour ago, k00b said:

can anybody tell me how reaction wheels work IRL - i don't understand how they can orient mass in the vacuum of space ?

How about Scott Manley on youtube: Reaction Wheels - Things Kerbal Space Program Doesn't Teach

Edit: I can also recommend the other clips in his "Things Kerbal Space Program Doesn't Teach" series.

 

Edited October 8, 2019 by AHHans

[Jake_IV]

3 hours ago, k00b said:

can anybody tell me how reaction wheels work IRL - i don't understand how they can orient mass in the vacuum of space ?

If I remember my satellite control class correctly from a long time ago it's gyroscopes. Reaction wheels are basically gyroscopes. 

[k00b]

thanks all - i didn't understand about "positive and negative momentum" relative to space (...because it's relative to object therein "OOps")

you know they say that "space is expanding"; does this mean that [the mathematic equations behind] einsteins model of spacial relativity are also being "warped" or are they talking about "the space" outside of the solar system does anybody know ?.

 

 

 

 

[AHHans]

11 minutes ago, k00b said:

you know they say that "space is expanding"; does this mean that [the mathematic equations behind] einsteins model of spacial relativity are also being "warped" or are they talking about "the space" outside of the solar system does anybody know ?.

Oh. That's a teeny tiny bit more complicated than reaction wheels or control moment gyros.

The relevant equations are also not special relativity (which doesn't contain gravity!) but general relativity. "Space expanding" means that the distance between objects where said distance is not determined by any of the four forces (the strong and weak nuclear forces, electromagnetism, and gravity) is increasing. E.g. the distances between the molecules and atoms in a piece of solid matter is determined by the electromagnetic force. I don't know how the expansion of the universe would affect the orbits of our / a planetary system if it were in a real(TM) vacuum. But I'm pretty sure that the dense soup of interplanetary space (5 particles per cubic centimeter on average? No wonder that you'll fall into the star sooner or later.) has a stronger effect.

[Guest]

36 minutes ago, k00b said:

you know they say that "space is expanding"; does this mean that [the mathematic equations behind] einsteins model of spacial relativity are also being "warped" or are they talking about "the space" outside of the solar system does anybody know ?.

Maybe there's a physicist in the house who can explain it better, but in my layman's understanding it's about Einstein's model. The observable upshot is that everything in the Universe is receding from everything else, unless it's held together by gravity or some other force (this is observable from spectra of distant objects such as other galaxies),  and if you trace it backwards in time you'll arrive at a point where everything was concentrated in a single point (a singularity), viz. the Big Bang. So it's not stuff in the Universe that's expanding into something, but the Universe – space – itself.

 

Edited October 8, 2019 by Guest

[k00b]

2 hours ago, AHHans said:

Oh. That's a teeny tiny bit more complicated than reaction wheels or control moment gyros.

The relevant equations are also not special relativity (which doesn't contain gravity!) but general relativity. "Space expanding" means that the distance between objects where said distance is not determined by any of the four forces (the strong and weak nuclear forces, electromagnetism, and gravity) is increasing. E.g. the distances between the molecules and atoms in a piece of solid matter is determined by the electromagnetic force.

thankyou - please could i rephrase the question to pertain to general relativity - particularly einsteins model of "the earth / space (whatever...) being suspended in a "Riemannian geometry net" - geometry pertains to measurable points at the end pieces no ?

so if the end pieces points have "moved apart" (re "expansion" of space (in between them...) > then won't that mean that the mathematical equations have been "warped" since formulation ? OR is that why it remains a "theory" ? OR something else ?

thanks.

[Geonovast]

This thread has been split off from an unrelated gameplay question.

Carry on.

[HebaruSan]

1 hour ago, k00b said:

thankyou - please could i rephrase the question to pertain to general relativity - particularly einsteins model of "the earth / space (whatever...) being suspended in a "Riemannian geometry net" - geometry pertains to measurable points at the end pieces no ?

so if the end pieces points have "moved apart" (re "expansion" of space (in between them...) > then won't that mean that the mathematical equations have been "warped" since formulation ?

Mathematical equations are not physical things that exist to be warped, so your question is difficult to interpret meaningfully. But yes, there is a piece of GR that accounts for the universe's expansion:

• https://en.wikipedia.org/wiki/Cosmological_constant

1 hour ago, k00b said:

OR is that why it remains a "theory" ?

No, there is no superior scientific category beyond "theory" into which GR could graduate. In casual usage where "theory" means a guess, it would be more accurate to call GR a "fact" or at least a "best current explanation".

[kerbiloid]

Like a cat rotates the tail to rotate the body in opposite direction when is falling from a bookshelf.

While he/she is falling, it's same zero-G.

[farmerben]

9 hours ago, Brikoleur said:

 

Real-life reaction wheels apply much, much, much smaller forces than KSP ones. Also if you keep turning the spacecraft in the same direction, eventually they'll be spinning so fast they risk their bearings. At that point they need to be braked, which of course spins the spacecraft again, and the craft will need to expend reaction mass and use its RCS thrusters to counter it. KSP doesn't model this because it wouldn't be fun.

 

There is also a mechanism called a magnetorquer this is 3 electromagnets for x,y,z axis.  The spacecraft becomes a magnetic dipole along its choice of axis and in the presence of an external magnetic field it lines up like a compass needle.  This torque can be used to spin down reaction wheels without wasting propellant.  Unwanted vibrations can also be eliminated when simple mechanical dampers are used with magnetorquers.  

The magnetorquer can also change the spacecraft's orbit without using propellant.  But it only works along particular vectors, unlike RCS which works in all directions.  So it is extraordinarily difficult to do rendezvous and docking without propellant but not impossible.  

 

[AHHans]

4 hours ago, Brikoleur said:

Maybe there's a physicist in the house who can explain it better,

I'm not sure that my explanation is better. Right now your explanation seems to be more understandable than mine. (But then, I was never good in theoretical physics! I'm much better in getting useful results out of an actual piece of hardware. )

48 minutes ago, HebaruSan said:

But yes, there is a piece of GR that accounts for the universe's expansion:

• https://en.wikipedia.org/wiki/Cosmological_constant

Well, you can model an expanding universe in GR without the cosmological constant. What it is (or does) is a way of explaining why the expansion of the universe seems to accelerate, same as "vacuum energy" or "dark energy". To a simple experimental physicist like me all three are essentially the same and just a fancy way of saying "We have no clue what is going on. So we give it a fancy name in order to attract more funding."

2 hours ago, k00b said:

geometry pertains to measurable points at the end pieces no ?

No. In this context "geometry" does not mean "I have some actual point here and there" but a mathematical model of what space looks like. In Newtonian physics space is "flat" and can be described with Euclidean geometry. In general relativity space is not flat and you need something like Riemannian geometry to describe it.

One popular example / model is to look at a triangle in 2-dimensional space. In a flat, Euclidean geometry the sum of the inner angles is always 180 degrees. In a curved (e.g. Riemannian) geometry geometry the sum of the inner angles of a triangle is more than 180 degrees if the space inside the triangle has a positive curvature, or less than 180 degrees if the space inside the triangle has a negative curvature. The surface of an inflated balloon for example has a positive curvature, so if you draw a triangle on it(*) and measure the inner angles(**) then you'll see that the sum is larger than 180. If you then inflate the balloon more (i.e. expand your 2d space) then the fundamentals of the space don't change. Also the equations that describe the space don't really change, only some parameters in the equations change. E.g. what is the equation that describes a circle with radius 5? And what is the equation that describes a circle with radius 10? Are those equations significantly different?
(*) I.e. mark three points and connect each pair with the shortest path on the surface of the balloon, which will not look like a straight line when you look at it from the side.
(**) A small enough section of curved space will always be close enough to being flat. So you can use your usual "Euclidean" devices to measure the angles.

2 hours ago, k00b said:

then won't that mean that the mathematical equations have been "warped" since formulation ?

Equations don't get "warped". Except when you are commenting on the legibility of the handwriting of the average physicist.  

[k00b]

1 hour ago, HebaruSan said:

Mathematical equations are not physical things that exist to be warped, so your question is difficult to interpret meaningfully. But yes, there is a piece of GR that accounts for the universe's expansion:

• https://en.wikipedia.org/wiki/Cosmological_constant

Quote

Einstein originally introduced the concept in 1917^[2] to counterbalance the effects of gravity and achieve a static universe

(i.e. "not expanding"), WHICH:

Quote

arises in Albert Einstein's field equations of general relativity

BUT i figured since:

2 hours ago, k00b said:

then won't that mean that the mathematical equations have been "warped" since formulation ?

like you say... / guessing that there will probably be some mathematical constant (ancient greek letter ?) accounting for the "time thereafter" ("space time hodgepodge")

meaning - the only difference between einsteins theories of relativities AND a random person saying "space is spacey" is that albert made spaceflight and kerbals possible ?.

space flight being good for ?

5 hours ago, Brikoleur said:

Maybe there's a physicist in the house who can explain it better, but in my layman's understanding it's about Einstein's model. The observable upshot is that everything in the Universe is receding from everything else, [and if] you trace it backwards in time you'll arrive at a point where everything was concentrated in a single point (a singularity), viz. the Big Bang. So it's not stuff in the Universe that's expanding into something, but the Universe – space – itself.

 

how do they know it is not wobbling ?

10 minutes ago, AHHans said:

Equations don't get "warped". Except when you are commenting on the legibility of the handwriting of the average physicist.  

i was meaning like "pushed to point, where point in time thereafter they don't make sense anymore" BUT my new hypothesis is there is a greek letter floating about...

"guh" maths - you know how newton plagurised robert hooke etc, then albert wrote aload of garbage about space being spacey, and then nowadays they are writing about space being spacey on a quantum level (annoying...).

[AHHans]

26 minutes ago, k00b said:

how do they know it is not wobbling ?

What do you mean with "wobbling"?

27 minutes ago, k00b said:

i was meaning like "pushed to point, where point in time thereafter they don't make sense anymore"

For general relativity that only happens at a singularity: inside a black hole or at the big bang. I.e. we don't know what happens inside a black hole or before the big bang. An expanding universe is no problem for GR, even an universe in which the expansion accelerates. It get pretty boring after a while (when everything moves away from you faster than light), but the equations work fine.

35 minutes ago, k00b said:

space flight being good for ?

Inspiring people to make great computer games!

[magnemoe]

15 hours ago, Brikoleur said:

Consider a motor spinning a flywheel. When you spin up the motor, the flywheel starts to rotate, right? Because of the conservation of motion, that momentum has to go somewhere. If your flywheel is on a spacecraft, it goes into the spacecraft, which rotates in the opposite direction. That's your reaction wheel.

Real-life reaction wheels apply much, much, much smaller forces than KSP ones. Also if you keep turning the spacecraft in the same direction, eventually they'll be spinning so fast they risk their bearings. At that point they need to be braked, which of course spins the spacecraft again, and the craft will need to expend reaction mass and use its RCS thrusters to counter it. KSP doesn't model this because it wouldn't be fun.

This, mind that this is only an problem if you has to compensate all the time like having to counter drag or solar wind. 
If you have an space telescope you can rotate and aim at targets for free. 
You can also set up your craft to have an antenna point toward earth by having it rotate one each orbit like the moon. This only require corrections.
And yes KSP uses strong reaction wheels for gameplay reasons, not fun using 30 minutes rotating an probe. 
Not sure how relevant bleeding of reaction wheels is. How much of an satselite fuel is used for this rater than combat drag and station keeping.

[k00b]

18 hours ago, AHHans said:

What do you mean with "wobbling"?

For general relativity that only happens at a singularity: inside a black hole or at the big bang. I.e. we don't know what happens inside a black hole or before the big bang. An expanding universe is no problem for GR, even an universe in which the expansion accelerates. It get pretty boring after a while (when everything moves away from you faster than light), but the equations work fine.

Inspiring people to make great computer games!

like jelly.... VERY SLOW JELLY > the maths only adds up in one direction, but we don't know if we are riding a wave of a wobble at the moment....

NO - it does NOT happen in a singularity, it happens at a "freeze frame of space time", space time is moving on times time, THUS, if it is wobbling, then it doesn't "work fine" it is "hodgepodge" for the "time being" (until; it wobbles back).

...all of the rare metals, all of the fossil fuel, all of the peoples minds, all of the engery, for 100yrs + and to "to make a computer game" - have you had any apples to the noggin ?

 

[k00b]

15 hours ago, magnemoe said:

This, mind that this is only an problem if you has to compensate all the time like having to counter drag or solar wind. 
If you have an space telescope you can rotate and aim at targets for free.

don't worry; the fresh prince already taught me:

BUT.

seemingly people are taking launch site location, rotation of the earth, "gravity" (read: on newton plagurising hookes work on "electromagnetism" (see: "duality"), then albert hodgepoding, sellotape like theory that "space is composed of rigid matter moving outwards" (and flat earthers get grief for being stupid...).

"let me tell you story all about how, life got flip turned upside down"

[paul23]

^ just what are you saying? Can you rephrase it with less hocus pocus and more direct sentences.

[magnemoe]

7 minutes ago, paul23 said:

^ just what are you saying? Can you rephrase it with less hocus pocus and more direct sentences.

Yes, note that all the self balancing toys like segways uses reaction wheels who is pretty powerful and you can obviously over saturate an segway it will try to move center of mass over center of wheels but leaning forward all the time while braking down an steep hill will probably do it. Don't try this. 

[Snark]

On 10/8/2019 at 1:46 AM, 5thHorseman said:

That's how reaction wheels work. except most ships have 3 of them (at least)

For the benefit of anyone reading this who's not familiar with them:  the reason for having three of them (perpendicular to each other) is that this gives you three axes X, Y, Z for free motion around any axis.

It's a not uncommon configuration to have 4 wheels, arranged like the four vertices of a tetrahedron.  That has the benefit of redundancy:  if any one wheel fails, the remaining three can still rotate the craft.

On 10/8/2019 at 1:47 AM, Brikoleur said:

Real-life reaction wheels apply much, much, much smaller forces than KSP ones. Also if you keep turning the spacecraft in the same direction, eventually they'll be spinning so fast they risk their bearings. At that point they need to be braked, which of course spins the spacecraft again, and the craft will need to expend reaction mass and use its RCS thrusters to counter it.

Well, it only needs to use RCS thrusters to adjust the net angular momentum of the ship, i.e. if the ship actually had an unwanted spin, e.g. due to some unbalanced external force that got it rotating in the first place.

If the ship wasn't initially rotating, and simply uses the reaction wheels to reorient itself, then no RCS is required.  The ship starts rotating when the wheels spin up, and then stops again (perfectly) simply by braking the wheels to a halt when the desired orientation is reached.

On 10/8/2019 at 4:42 AM, Jake_IV said:

If I remember my satellite control class correctly from a long time ago it's gyroscopes. Reaction wheels are basically gyroscopes.

Not really.  A classic reaction wheel doesn't use the "gyroscope effect" (e.g. resistance of a spinning object to changing its axis) at all-- it's a simple matter of "I spin you this way, you spin me the opposite way".

That said, though, there are specific cases of using gyroscopic effect in this way-- basically when you need something really "muscular" that can put out a lot of torque quickly.  This is known as a control moment gyroscope and is used, for example, on the ISS.  In a simpler configuration without the gimbaling hardware, it's referred to as a "momentum wheel".

21 hours ago, magnemoe said:

And yes KSP uses strong reaction wheels for gameplay reasons, not fun using 30 minutes rotating an probe.

That, and the fact that they violate the laws of physics-- it's not just that they're more powerful than a real-life reaction wheel, but that they actually allow violating conservation of angular momentum, which a real-life one obviously does not.  For playability purposes, it's perfectly reasonable.  More discussion here.

4 hours ago, magnemoe said:

Yes, note that all the self balancing toys like segways uses reaction wheels who is pretty powerful and you can obviously over saturate an segway it will try to move center of mass over center of wheels but leaning forward all the time while braking down an steep hill will probably do it. Don't try this. 

My impression is that such devices actually don't use reaction wheels or gyroscopes at all.  They're simply inverted pendulums that use a feedback loop between orientation sensors and wheel motors to keep the wheels under the CoM.

It's basically the same dynamics as a unicycle, but with inhumanly sensitive orientation sensors and inhumanly fast reflexes.

Source:

Quote

The dynamics of the Segway PT are similar to a classic control problem, the inverted pendulum. It uses brushless DC electric motors in each wheel powered by lithium-ion batteries with balance achieved using tilt sensors, and gyroscopic sensors developed by BAE Systems' Advanced Technology Centre.^[27] The wheels are driven forward or backward as needed to return its pitch to upright.

Dunno about other devices beside the Segway itself, but this makes sense to me.  Do you have some source for a device that does use gyroscopes or such to stay upright?  (Other than gyroscopes themselves, that is.)

[GearsNSuch]

So, if I remember correctly (please correct me if I’m wrong, which is probably probable) gyroscopes stay upright by making any mass imbalances even out. When the ring spins, any difference in weight from one side to another is essentially negated because it moves around, tipping the ‘scope the other way before it has time to fall.

So, gyroscopes could not, to my understanding, be used to make things stay upright unless the whole thing spun around. They do make good sensors, though. 

Hopefully that is accurate and intelligible. 
 

(Incorrect, need to retake physics)

Edited October 10, 2019 by GearsNSuch

[Snark]

1 hour ago, GearsNSuch said:

So, if I remember correctly (please correct me if I’m wrong, which is probably probable) gyroscopes stay upright by making any mass imbalances even out. When the ring spins, any difference in weight from one side to another is essentially negated because it moves around, tipping the ‘scope the other way before it has time to fall.

No, that's not the case-- if it were, they'd stand up even when not spinning.

The reason why gyroscopes stand up is that they're designed to have a very large amount of angular momentum.  To "fall over" would require changing that angular momentum to point in a different direction, which requires a lot of torque to do so.  Furthermore, applying a "fall over" torque causes precession, i.e. rotation at a right-angle direction, which further delays the time until they fall over-- precession is what makes them wobble around in a circle.

So, in short:  gyroscopes (or anything with a large amount of angular momentum) resist having their axis of rotation changed, because doing that requires changing the direction of their (large amount of) angular momentum.

[magnemoe]

12 hours ago, Snark said:

My impression is that such devices actually don't use reaction wheels or gyroscopes at all.  They're simply inverted pendulums that use a feedback loop between orientation sensors and wheel motors to keep the wheels under the CoM.

It's basically the same dynamics as a unicycle, but with inhumanly sensitive orientation sensors and inhumanly fast reflexes.

Source:

Dunno about other devices beside the Segway itself, but this makes sense to me.  Do you have some source for a device that does use gyroscopes or such to stay upright?  (Other than gyroscopes themselves, that is.)

Yes you are correct, I assumes they use the unicycle dynamic, this implies it will fail if stuck in one direction like trying to drive up an curbside. 
Think most moving toys use this to like the ball robot from star wars. 

Has seen some moving cubes however who use reaction wheels to generate angular momentum in an direction. Its easy to desaturate by spinning down so slow you are still standing still. 
They had been nice if you could make so small you could use them as dices  
Told on this forum then jumping with motorcycles or even cars they use wheel rotation as reaction wheels to keep orientation