# Simple physics question, I think, with regard to swinging your arms to regain your balance

Unlike human arms, cat forelimbs are attached to the shoulder by free-floating clavicle bones, which might make them better at absorbing shocks without damage.

It might also be that a head-down posture allows them to keep sight of their landing spot, for better control.

And if you’re using your breath to get out of the middle of a space capsule, you could just arrange to turn in opposite directions for the inhale and exhale, so the forces are to your benefit both times.

That’s been discussed multiple times on Space Exploration Stack Exchange (where I’m not a member), and the consensus seems to be that it’s not worth it, because inhaling isn’t very directional.

Easily demonstrated by the fact that you can easily blow out a candle but can’t inhale to extinguish a candle.

Back in the days when kid’s toys were made out of metal, they used to make little toy boats called “putt putt” or “pop pop” boats. They had a small chamber with a tube in the middle of the boat, with a small area under the chamber just big enough for a tiny candle. With a little bit of water in the chamber, the water would get heated into steam by the candle, which would blow out through the tube and push the boat forward. With the steam expelled, water would then get sucked back into the tube and into the chamber, where it would be heated into steam, and the cycle would repeat. It’s the same principle, the water is drawn in from all directions but expelled only in a single direction.

I’ve made a simple putt putt boat by just taking a small copper tube and bending it around a broom handle to make a coil “chamber”, then attaching that to a small wooden boat. It doesn’t “pop” the way a tin chamber does but the boat does move.

I had one of those when I was a kid, and I’m under 30 still. I just realized, reading your post, that that’s an external-combustion valveless pulsejet engine. Your DIY one, if it was a pulsejet (i.e. if it didn’t somehow have a different working principle that you didn’t realize) must have pulsed, though I guess not loudly enough to hear. Maybe that’s interesting enough to split into another thread.

To get back on topic (angular momentum in space) a bit, I’ll mention a third kind of device related to reaction wheels and CMGs. This is the momentum wheel. It’s pretty much exactly the same as a reaction wheel, mechanically, but is operated differently. While a reaction wheel is spun slowly and precisely to impart the opposite rotation to rest of the spacecraft, a momentum wheel is spun rapidly and constantly to increase the spacecraft’s overall angular momentum, which increases its passive resistance to being rotated by external forces. This is similar to throwing a flying disc (e.g. Frisbee): if you throw it without any spin, it’ll just flop out of the air; if you spin it, that spin helps it stay stable even as the aerodynamic forces apply a pitching moment to it.

You might think “Let’s have two momentum wheels with parallel axes but opposite directions of rotation, so we don’t get gyroscopic precession effects, so they don’t cause the spacecraft as a whole to spin if their motors simultaneously stop being able to offset bearing friction, and so we can spin them up against each other instead of having to use the RCS thrusters!” That will have those advantages, but it will have the severe disadvantage that they won’t work as momentum wheels anymore. The spacecraft will be just as vulnerable to external torques as it would be with no momentum wheels. Each one’s attempts to stabilize the spacecraft against a given disturbance will just cancel out against the other’s, resulting in potentially large forces in the frame connecting them, but no overall stabilizing effect on the spacecraft. Another way to think about this is to consider that angular momentum is a vector, and the angular momenta of the two opposed momentum wheels cancel out and leave the spacecraft as a whole with no additional angular momentum.

For the same reason, while it’s common to have 3 reaction wheels, one for each axis (or more, for redundancy, because they do fail sometimes), it’s pointless to have more than one momentum wheel (at least, in terms of the physics—available space in the spacecraft may require using multiple small MWs instead of one big one). If you use multiple momentum wheels, even in different axes, their effect will be the same as that of one big momentum wheel whose angular momentum is the vector sum (?) of the small ones’ angular momenta. (Therefore, if their axes aren’t parallel, you get cosine losses, degenerating to the above case where they’re directly opposed and cancel each other out fully, assuming they’re the same size and speed.) So, generally, you’d only use one momentum wheel, and it only stabilizes the spacecraft in two axes (those that are perpendicular to its rotational axis). Sometimes that’s all you need. I suppose, if you have fine control of its rotational speed, you could also use it as a reaction wheel, by superimposing a reaction wheel-style rotation on its momentum wheel-style rotation, but I don’t know if that’s ever been done in practice. Conversely, if your spacecraft has reaction wheels for all three axes, but two of them fail, you might be able to spin the remaining one fast enough (using the RCS thrusters or other open-system actuators such as magnetorquers to hold the spacecraft still while you spin it up) to act as a momentum wheel for the other axes, so you can at least stabilize them but not control them (without using the other actuators).

Probably not interesting to spawn off into another thread, but yes, it pulsed. You could see the pulses in the water but it didn’t pop loudly enough to hear.

If anyone is interested in this in more detail we can split it off into another thread.

I recall in some early SF (I think it was one of the Arcot, Morley, and Wade books) where someone stuck in the middle of a weightless environment took off his shirt and waved it back and forth toward the direction opposite where he wanted to go. This was using it like divers use fins on their feet. Only, of course, much less efficiently since air is much less dense than water.

I don’t think I’ve seen that suggested in any of the discussions I’ve read on solutions to that problem (though some of them did specify that you were totally nude and not holding anything). If you have a shirt or other wavable clothing, it’s a good idea—more sustainable than throwing the clothing, at least.

Neat! If you’re in Low Earth Orbit, you can use Earth’s magnetic field to desaturate your reaction wheels. Either actively, with a magnetorquer (basically just an electromagnet along each axis), or passively (using a permanent magnet and a hysteresis material). But obviously you have to get more creative in deep space.

@engineer_comp_geek : thanks for bringing back memories. I played with that boat for a very very long time. Tried burning candles, then cooking oil lamps : discovered that if there is an optimum amount of water in the wax, it would crackle. Cut and twisted the tubes on one to make it spin…,

This is deep space, unfortunately. I’ve been wondering if magnetorquers would still be viable, just really slow, working against the heliospheric magnetic field. (Well, they’d also lack 3-axis capability, due to not having an ambient field whose direction varies cyclically (as described below), but I’ll have the RWs for that.) I’ve been unable to find any papers or discussion about the possibility of using magnetorquers in deep space, in Google searching a week ago and again today. [Note: “Today” is May 3, because I wrote this post and then got distracted for several days before submitting it.] Everyone just says “they won’t work there” without showing any math. Anyway, if not, vanes should be fine (I say without having done any math).

The bigger difficulty will be thermal design, with the spacecraft being limited (though only by my choice) to 3U CubeSat size and shape, and with a planned aphelion near the orbit of Jupiter. (I like challenges.) The MarCO spacecraft, at 6U each, didn’t survive their aphelion (which I can’t find a number for, though I assume it was lower), but then they weren’t designed to. I’ve been meaning to look up their thermal design for a while.

Back on the general topic of angular momentum and attitude control, I’ve also heard of that magnetorquer alternative you mentioned: a fixed permanent magnet. Advantages I see are that it’s simple and cheap, and it doesn’t require control or moving parts. Disadvantages I see:

• It can’t be used for active attitude control; it only stabilizes the satellite to a predefined orientation (or, if the magnet is strong enough and the diagram and text on the first linked page below are accurate, a predefined rotation over the course of an orbit).
• It can’t be used for 3-axis attitude control like active magnetorquers can (by taking advantage of how Earth’s magnetic field varies in direction over the course of an orbit, though that doesn’t work in equatorial orbits anyway, AFAIK). (However, the description quoted in the first answer on the first page linked below claims that magnetic hysteresis material can “resist changes on the uncontrolled axis as well”; I don’t see how, unless it’s wide enough to have significantly different field strengths going through different regions of it.)
• It can’t actively stabilize the satellite’s orientation, so, as you mentioned, you need something to damp the oscillations and let the spacecraft come to rest with respect to the external magnetic field. You mentioned using a material with high magnetic hysteresis, but my understanding is that you can also use something mechanically dissipative that acts via internal friction (such as a long whip antenna, a closed container of liquid, or a soft material for the magnet mount), or rely on your satellite being big and electrically conductive enough that eddy currents induced by Earth’s magnetic field as the satellite rotates will be sufficient to damp the oscillation.
• When detumbling, it can’t modulate its field strength and direction to be optimal for the instantaneous orientation, so detumbling will take longer even with strong damping (probably not a big deal, unless the orbit is expected to decay very quickly).

Here’s some discussion on flight experience with this technique.

Here’s an idea to gimbalize and motorize the permanent magnet, to get a semi-active actuator that’ll (hopefully) need less power than fully active (electromagnetic) magnetorquers.

The answer there also mentions another disadvantage I didn’t think of: permanent magnets can cause spacecraft in the same dispenser to stick together on deployment and be unusable. (The CubeSat standard does have a limit on external magnetic field strength for this and other reasons, but I don’t know if it’s actually stringent enough to prevent it. It wasn’t prevented in the cited case, obviously, but they might have gotten a waiver for stronger magnetic fields.)

I also know of at least one case where two CubeSats successfully deployed separately, but eventually drifted together and became stuck together while in orbit.

And I’d expect that anything with a mass-to-surface ratio suitable for a solar sail wouldn’t have any problem with thermal dissipation.

Yes. That’s the problem. How do I keep it warm enough to survive when it’s around 5 AU from the Sun? I have some ideas, but I don’t really know the scale of the problem for sure yet, so I don’t know if they’ll be enough, or if they’re overkill.

I thought of this possibility when someone first suggested magnetorquers and did a little research. It turns out that the interplanetary magnetic field is way weaker than the Earth’s. I forget the numbers, but something like 6 orders of magnitude lower. So in theory, it would work, but in practice, probably way too slow for any practical use.

Ah, the opposite of the usual satellite thermal problem. How exactly does cold hurt anything, though, absent an atmosphere? Solid-state electronics get more reliable the colder they get, right up until the (extremely cold) point at which they become superconductive. I suppose the batteries might be an issue?

Probably. I know that that is often what ends the life of Mars missions, when there is not enough power to keep the batteries warm through the cold night, and they don’t wake up in the morning.

I don’t know if that’s because the electrolyte freezes and bursts them or what though.

A link to the paper on the cubesats that Chronos and your link mentioned. We heard the same story during our cubesat development, but in our case all the stabilization development was essentially by rule of thumb. I.e., use these magnets and this material and you’ll be fine. At least in LEO, the stabilization works across a wide range of parameters, and the mass is minimal, so just use what’s known to work.

dtilque mentioned a deep space magnetic field that’s 6 orders of magnitude lower than LEO. That’s not so bad, IMO. The magnets we used were tiny. I guess I’d need to think through the problem more, but isn’t the torque effect going to be related to the volume of the field as well as the strength? It should be possible to make quite a large field.

Except for the sticking together problem, of course. But maybe there’s a way around that. Essentially, have two magnets that before release are arranged oppositely. The field is then small because the field lines just loop over. Then, once deployed, a motor rotates one of them so that the two magnets reinforce each other. It’s passive after the deployment phase. Sorta similar to the gimbaling idea, but the goal here is to have just two configurations, weak and strong.

Magnets aren’t my strong point, though. Could be my intuition is off here.