In my current worldbuilding project, humans find themselves struggling to survive on a cold, dead world with minimal sunlight and the aid of magical machines called "spinners". Spinners are two identical cubes adjoined by a face, always counter-rotating w.r.t one another at a constant rate. No amount of ungodly toque can slow their rate of counter-rotation, betraying unlimited, tappable work.

People use these machines for generating heat and powering machines and vehicles. My question is, what is the best way spinners may be incorporated into the design of a rocket engine? (Couple notes: the fastest spinners rotate at around ~1 Hz which, due to their unwillingness to be slowed by counter-torque, can be geared-up to hundreds, thousands, or hundreds of thousands of Hz by a gearbox; the engine need not necessarily have a thrust-to-weight ratio of 1:1 or greater--the focus is maximum delta-v.)

This sort of perpetual motion won't necessarily get you off the ground and into orbit for free. Rockets will still need exhaust to push on to generate thrust in a vacuum. So, spinners neither eliminate the need for fuel nor the tyranny of the rocket equation.
My knowledge of rocketry is weak, but I imagined a spinner-powered thermal device could be used to heat some propellant, say, liquid hydrogen, up to a few thousand Kelvin like in a typical nuclear thermal rocket, except the nuclear energy doing the heating is replaced by this spinner-powered thermal device. The thermal device could work by forcing a great deal of electricity through many tightly-wound metal coils (the spinners would be used to generate the strong rotating electromagnetic field). The friction in the coils would heat the propellant and so on. However, I don't think there is a metal or metal alloy that doesn't start vaporizing above ~2000 K (probably less than that, actually). At that point, one may as well forget about spinners and use nuclear elements instead.

Is what I suggest feasible? If not, is there another way to feasibly incorporate spinners into rocket technology? (I don't want to continue with the development of my world into the off-world without knowing the applications of spinners in space.)

  • $\begingroup$ Comments are not for extended discussion; this conversation has been moved to chat. $\endgroup$
    – L.Dutch
    May 25, 2021 at 10:49
  • $\begingroup$ How do you harness the energy from (aka "attach gears to") these spinners when they can't be stopped? $\endgroup$
    – RonJohn
    May 25, 2021 at 13:46
  • $\begingroup$ @RonJohn youtu.be/TOsB4Vhpkw0 small rotations can be translated into high rotations. No matter how many gears you continue attach, the counter-torque from all those gears sapping energy from the spinners does not slow down the spinners' rate of rotation. $\endgroup$
    – BMF
    May 25, 2021 at 13:51
  • 1
    $\begingroup$ @RonJohn sounds like a challenge for the engineers to solve. Doesn't strike me as an absolutely crippling problem, though. $\endgroup$
    – BMF
    May 25, 2021 at 14:03
  • 1
    $\begingroup$ @RonJohn Are you honestly arguing that it would be too difficult to attach a gear to a cube rotating at 1Hz? We connect two spinning things all the time via a clutch, but at 1Hz you could probably just attach something by hand $\endgroup$
    – Kevin
    May 25, 2021 at 21:17

6 Answers 6


Electrolysis & Fuel Generation Until You're In Orbit

Not only are you correct about the temperatures at which your heating coils would start to fail, the gearbox needed to generate that kind of power would be prohibitive in terms of of the space required on board the vessel. So...

If you've got life, you've got water (or other materials from which hydrogen and oxygen can be extracted). That gets you into orbit - nice traditional cryogenic launches.

Once you're (most of the way) out of the gravity well, however, that's when your spinners can really shine. You've got unlimited energy, so you want to minimize reaction mass; perfect situation for an ion engine. You could have unlimited burns. It would take a while to get anywhere, but you would have effectively unlimited range.

  • 1
    $\begingroup$ Ah, that hadn't really occured to me. The relatively great energy requirements of water electrolysis is basically eliminated by spinners. The fuel is free, especially on a frozen (water) ice world, just keep the manufacture going. Ion propulsion is probably the best way to use the spinners for efficiency. Great answer $\endgroup$
    – BMF
    May 23, 2021 at 23:37
  • 1
    $\begingroup$ "Unlimited" is difficult, as you still have a finite amount of mass to fire, but the rest is full on the mark. $\endgroup$
    – Trioxidane
    May 24, 2021 at 6:24
  • 4
    $\begingroup$ @Trioxidane you could invest in good ole photon thrusters, but their ISp is even smaller than that of ion engines... and maybe you don't even need to go crazy on the design. Shove the cubes into a generator and let it warm up to nice balmy juuust below the melting temperature, then let it chill in the focus of a parabolic mirror. $\endgroup$ May 24, 2021 at 7:54
  • 1
    $\begingroup$ The nice thing about free energy is that the low ISp of an ion thruster matter less when you take away the incentive to not simply use a honking lot of them. $\endgroup$
    – Matthew
    May 24, 2021 at 16:03
  • 3
    $\begingroup$ Also keep in mind you'll be able to make very efficient and simplified rocket engines compared to today's existing stuff. SpaceX spent a decade developing the Raptor engine with full flow combustion cycle, but with spinners powering the pumps you get some free energy and immensely simplify the material requirements. You won't need a turbine to withstand the oxygen rich high temperature pre-burner gases. And if the spinners are small enough you can make both pumps completely enclosed, eliminating the need to figure out good valves to prevent oxygen & fuel leaks. $\endgroup$
    – csiz
    May 25, 2021 at 0:56

The machines provide the ability to have a reasonably compact source of enormous electrical power. Let's assume this power is "infinite" for practical purposes. In reality it will be limited, but the limitation is with the material science they can screw onto the ends of that cube. With cube size not specified this is hard to estimate.

So.. Ample electrical power. At the very least, some hundreds of megawatt to gigawatt of power.

To get to orbit:
Electrical propulsion. Ingest atmospheric air, superheat it to plasma. Eject the plasma through an acceleration grid. Simple, strong and horrendously inefficient. Fortunately, with "infinite" on one side of the scale, efficiency is a mere footnote.

In space:
If you are in a hurry:
Your fuel is whatever matter you can lay your hands on. Lighter atoms are more useful, hydrogen is ideal. But plain old water will work fine.
Same story as in the air, just superheat the stuff to plasma, then violently shove the plasma to the rear using strong magnetic fields. Something like the VASIMR engine. But again, because we have near infinite power available, we can leverage the reaction for thrust and fuel efficiency, rather than being worried about power as mere mortals have to.

If you are not in a hurry, near a planet with magnetic field, and refuse to spend fuel:
Just hang a long tether overboard, run Thor's own electrical current through it, and use the interaction of that magnetic field with the planetary magnetic field to move around in your orbit. Slow, burn silly amounts of power for very little acceleration, but power is free and it uses absolutely zero fuel.

If you are not in a hurry, and nowhere near a magnetic field, and still refuse to spend fuel:
Use your electrical power to run a photon torch drive. This is the point where you realize that your "infinite" electrical power is anything but infinite, and true infinity would have been useful. Still, a couple of gigawatt of power is enough to power a workable photon drive that will take a light starship to relativistic speeds in years, rather than centuries or millennia.

  • 6
    $\begingroup$ If you have effectively infinite electrical power, water IS hydrogen. $\endgroup$
    – Clearer
    May 24, 2021 at 11:02
  • 6
    $\begingroup$ Worth pointing out that you don't have infinite power, just infinite energy. The torque is infinite, but the materials being used to turn that into useful work are mundane, so the actual power being extracted is finite. This makes a lot of these ideas impractical. $\endgroup$
    – jdunlop
    May 24, 2021 at 22:23
  • 1
    $\begingroup$ @jdunlop So it seems. It's something I totally overlooked when building my world. I'm currently trying to figure out how much torque a socket connecting a gear train to a spinner could reasonably handle before deforming. But, it's not looking promising. $\endgroup$
    – BMF
    May 24, 2021 at 22:39
  • 1
    $\begingroup$ @BMF: Perhaps your spinners could be a different shape, like infinitely strong flat spokes a meter or two long sticking out of a counterrotating axle for the largest ones. (3 or more spokes per wheel, maybe 8, maybe with the number having some significance to their origins, or varying by spinner.) That distributes the load through a much larger volume of mundane material, and over a larger area (since the spokes are wide and flat like cube faces, but the whole face is moving in the direction of motion). $\endgroup$ May 25, 2021 at 1:31
  • 1
    $\begingroup$ @BMF: perhaps you can extract more work from a cube spinner by using it like a camshaft / crankshaft, placed under a pillar to raise it. (Perhaps allowing for the material to abrade away at the slide point.) i.e. using the bouncing of a square wheel with it's non-constant "diameter". $\endgroup$ May 25, 2021 at 1:35

Forget launch rocketry, use an elevator, then use an ion thruster in space

If nothing in the universe can stop a spinner from spinning, then use them to power a space elevator. It might take a day or two to complete the transit, but you're guaranteed to do so regardless the weight of the load (ignoring the mechanical specifications and/or limitations of the car & elevator tower).

Once you're in space, you can use ion thrusters to move around.

  • $\begingroup$ My world is about Mars-mass and size, so space elevators would be most possible, although with natural resources rather scarce on a desolate world my civilization would find it very economically taxing to build one until space infrastructure kicks off. Ion thrusters seem to be the efficient way to make use of the spinners' unique properties. I was actually considering using spinners to power the ion engines of great big reusable hypersonic airships for cargo transfer to low-orbit and back. $\endgroup$
    – BMF
    May 24, 2021 at 2:39
  • 9
    $\begingroup$ This involves inventing more magi-science for the elevator cable; power for space elevators has never been our problem, so it seems odd to say "Hey, with unlimited power, use a space elevator". $\endgroup$ May 24, 2021 at 10:17
  • 1
    $\begingroup$ Even after a few thousand years, it's impossible to believe the resources of a planet are depleted. The easy resources may be depleted (the stuff within 3,000 meters of the surface), but @Clearer's point is 100% dead on. Hook the spinners to diggers and point them straight down. When you're that far underground, what's on the surface is irrelevant. $\endgroup$
    – JBH
    May 24, 2021 at 15:23
  • 3
    $\begingroup$ @Harabeck: Unless the existing materials have improved dramatically since 2012, it seems like it's still a pretty iffy proposition. And as noted, the problem has never been power (they're power saving), so it's weird to go with "the best use for infinite power is something that has never been power constrained in the first place". I'm not arguing whether it's feasible, just whether it matters at all in this context. Granted, the frame challenge is valid (you don't necessarily need rockets), but that applies without infinite power cubes. $\endgroup$ May 24, 2021 at 15:58
  • 1
    $\begingroup$ @JBH Not depleted, just more scarce. Most of the civilization's workforce is devoted to food production. The lack of land to live and grow food on limits the population size. It's not one well-oiled nation either, but several that are often at odds with one another. There comes a point where they've depleted all the obvious veins in their own, most-accessible islands and territories, and begin seeking out the obvious veins on distant, frozen-over islands while also searching for new resources in the homeland. $\endgroup$
    – BMF
    May 24, 2021 at 21:45

Space based jackass merry go round to anywhere

Taking one of these things to low Earth orbit should be relatively cheap. From there you can reach to the stars.

Now have a launcher that is a long cylinder, with such a spinner attached to, and powering a reaction wheel.

You can make the reaction wheel spin arbitrarily fast, and with it the external surface of your launcher. Now all you need is a decoupler on that surface.

Mount your interstellar vessel on that decoupler. Accelerate the launcher to the necessary angular velocity. Activate the decoupler, and the vessel leaves the launcher on a tangential with the same velocity as the launcher's surface's linear velocity.

If you are having trouble visualizing how this contraption would work: the launcher is this merry go round, the reaction wheel is the back wheel of the motorcycle connected to the merry go round's axis, the spinner is the motorcycle's engine and the vessel to be launched is the guy who gets a lesson in cruel physics.

Two guys playing a merry go round too fast, one of them is launched into the air

Wanna escape the Earth? I remember its escape velocity at LEO is below 11 km/s (the escape velocity at sea level), so using that value will do. Wanna escape the sun? You just need to be going at 42.1 km/s when you exit Earth's sphere of influence. If you wish to reach any specific star or even galaxy, you just need to figure out a trajectory and the proper speed - and the spinner + reaction wheel combo will make sure you can reach the latter in finite time. No delta-v budget is too expensive anymore.

If you wish for an orbital injection at the destination, you can make it so that the main vessel also has a spinner and a reaction wheel, and then you can do this mad maneuver in reverse. You will have all the time of the trip to build up angular speed for this secondary launcher.

Edit: As per Harabeck's comment:

SpinLaunch is a real company trying to do something close to this conceptually.

It was founded in 2014, so the concept for the space based jackass merry go round has been a thing for quite a while now.

Also in space you don't need a perpetual motion machine - sunlight is practically unlimited, and you can keep converting solar energy into reaction wheel love as long as the materials involved can take the stress.

  • 3
    $\begingroup$ This reminds me of a world building project I did a long time ago. A group needed to reach a certain star system before another--beat them to the punch. They instructed autonomous assemblers to garner and refine space resources to build a thousand-kilometer-long guass gun delivery system. A gun that shoots a backwards gun that shoots the payload. A series of telescoping coilguns that accelerate the next series onward into high fractions of light-speed, and then back down again to a relative stop at the destination. $\endgroup$
    – BMF
    May 24, 2021 at 3:43
  • 4
    $\begingroup$ A 100 meter radius cylinder spinning with an angular speed of 42 km/s experiences a centrifugal acceleration of almost 18 million gees. Wow! $\endgroup$
    – BMF
    May 24, 2021 at 3:48
  • 11
    $\begingroup$ @bmf well, you can't make an omelet without liquefying a few astronauts. $\endgroup$ May 24, 2021 at 3:51
  • 6
    $\begingroup$ @TheSquare-CubeLaw - at eighteen million gees, it's less "liquefying" and more "compressed to a laminate only a few molecules thick". $\endgroup$
    – jdunlop
    May 24, 2021 at 6:41
  • 5
    $\begingroup$ @JoeBloggs if you launch the payload at twice the orbital velocity, the launcher won't deorbit, only change inclination by 180 degrees. Making it function more than once is a problem for the engineers. $\endgroup$ May 24, 2021 at 8:07

For getting into orbit, you could use the spinners fairly easily to power a Lofstrom loop. The classic issue here is largely "you really want several nuclear reactors to power the thing" (which has political issues among other things), which you sidestep fairly easily. (Why nuclear? Because it requires a fair bit of power, and rather reliable power at that. Both of which you have here.)

(There are ways to deal with a power failure on a Lofstrom loop, mainly involving stealing momentum from the cable itself to keep the containment going, but it's, uh, hair-raising at best.)

(For reference: a Lofstrom Loop (or Launch Loop) is effectively just a really large loop of cable, spinning quickly enough that its own inertia holds it up against gravity. To launch, just steal momentum from the cable.)

Once you're in orbit, a couple of possibilities spring to mind.

One obvious approach is an EDT (electrodynamic tether). An electrodynamic tether is essentially just using the magnetic field of whatever you're near as the magnet for a generator/motor. Normally these are largely used "in reverse", to provide power at the expense of decaying your orbit, but you can use this as a motor too (taking power to raise your orbit, essentially).

They do have a bunch of downsides (power-hungry, relatively low acceleration, can interfere with nearby craft, etc.), but their main drawback is simply the power use - and you've got power in spades. And they don't use reaction mass.

One more classical approach here is an ion drive. Normally an ion drive has a relatively low effective Isp, mainly due to the mass of the engine's power supply - but you have power to burn. So you can run at very high power consumptions (which boosts efficiency). Still a very low power-to-weight ratio however.

A photonic drive sounds great initially. Honestly however, this likely isn't worth it. Even assuming "infinite" power, we don't have any good way to produce anywhere near the photon density required to make it worthwhile.

(As an aside, I'd expect the early space program to focus far more heavily on magnetohydrodynamic drives for airplanes than we did. They have largely been discounted for us due to a lack of a decent power supply... but given these you could probably build a fairly decent spaceplane.)

(As another aside: I actually think that it'd be rather difficult to use these. The required torque (and hence, mass) to get a decent amount of power from a 60RPM rotation is not exactly small. 1 horsepower would require ~550 pound-feet of torque (for comparison, a bicycle is, what, 50-100 pound-feet of torque?). Someone who is more patient than I could probably work out the maximum effective amount of power you can get out of one of these based on material limits at the attachment point... but I expect it to be surprisingly small.)

  • $\begingroup$ Is it possible to reduce the amount of torque at the attachment point by some clever arrangement of mechanics? Could using multiple spinners and multiple attachments for the same high-rpm output (by some clever arrangement) mitigate the issue? $\endgroup$
    – BMF
    May 24, 2021 at 21:35
  • 2
    $\begingroup$ @BMF - multiple spinners stacked would help. It would require a fair few spinners though (each one "adds" 60RPM to the final output). Also I'm pretty sure I'm off by a factor of pi (or 2pi?) on at least some of these numbers. $\endgroup$
    – TLW
    May 25, 2021 at 1:56
  • 2
    $\begingroup$ @BMF the fundamental issue here is power = torque * rotation rate = force * radius * rotation rate... and radius and rotation rate are both fixed and both fairly small. Ultimately the power is transmitted through the surface of a, say 4-inch sphere, and hence is limited by, among other things, the shear stress of the surface of said sphere. Which because radius and rotation rate are limited limits your overall power output. You'd do much better if it was e.g. 2 4-foot bars rotating against each other (counter-rotating-propellor-like). $\endgroup$
    – TLW
    May 25, 2021 at 2:03

This sort of perpetual motion won't necessarily get you off the ground and into orbit for free.

Actually, it can (maybe), if your engineering is good enough. Leaving aside PcMan's answer...

We know how to produce thrust by pushing air around. Getting to orbit isn't a matter of getting high enough (well, unless you want to get really, really high; so high that, without a reaction-less drive, this approach simply isn't practical thanks to the tyranny of the rocket equation) but of going fast enough. So... do this in the atmosphere so that you can use the planet's air as reaction mass rather than having to haul your own. If necessary, produce downward thrust to keep yourself in the atmosphere until you've got to your desired speed. You might still need to bring along some delta-V for orbital corrections (though, see other answers), but not nearly what you would need to just get to orbital velocity.

The trouble comes when you need to balance the required speed against certain unfortunate effects of atmospheric heating, which means you probably need to do this at a sufficient altitude that the air is thin enough for this to be manageable. Thus, you probably are looking at something like a ramjet engine. Also, really, really good heat shielding.

  • $\begingroup$ Actually I had been leaning towards this sort of approach when I first started this project. At first, I wasn't sure if you could actually build an hypersonic engine that, well, didn't oxidize fuel. Supersonic engines work by accepting and compressing air through an intake and exploding it with fuel, exhausting it out the back faster than it came. I wasn't sure if mere propellers could actually move air supersonically, but from what I've read I think the answer is yes. $\endgroup$
    – BMF
    May 24, 2021 at 16:41
  • $\begingroup$ I was considering asking a question in the future about a spinner-driven hypersonic low-orbit cargo transfer vehicle, whether you could build such a thing (given the fantasy infinite spinning cubes) with known materials, but I haven't nailed down all the details I need to properly define the constraints, like, for instance, the gearbox that gears-up the spinner rotation--its mass and construction could be the limiting factor in whether something like that could ever get off the ground. $\endgroup$
    – BMF
    May 24, 2021 at 16:47
  • 1
    $\begingroup$ It's also worth noting that my planet is Mars-mass and size with a human-congruous atmosphere (1 bar at the surface). The scale height is much greater and the edge of space starts at around ~160 km up. With a lower orbital velocity than Earth, it may be easier for hypersonic aircraft to achieve orbit (with some help from traditional rockets). $\endgroup$
    – BMF
    May 24, 2021 at 16:50
  • $\begingroup$ @BMF: With lower gravity, atmosphere can more easily escape into space. For consistency, you're going to need some explanation of why that hasn't happened (yet?) on your planet, with a much greater height of air on a lower-grav world. I'm not an expert on the mechanics of that, but it's something that sounds made-up and not fully physically plausible to me, without circumstances that explain it (like geoengineering via spinners to split water into H2 and O2 on an incredibly vast scale, to keep making more atmosphere) $\endgroup$ May 25, 2021 at 4:28
  • $\begingroup$ @PeterCordes Oh, it's complicated (but not really). It's really not implausible for a Mars analog to have a thick atmosphere, even across geological timescales. There are natural mechanisms to protect and replenish on that scale. Luckily, I don't have to go about convincing the reader that, because this particular world was artificially manufactured for humans by aliens. It's not supposed to be perfect. It's got near-Mars gravity and only a dithering of land, the rest is deep sea. To explain why would be to go a bit off the rails and get into the lore of my project. $\endgroup$
    – BMF
    May 25, 2021 at 4:37

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .