h a l f b a k e r y
We don't have enough art & classy shit around here.

meta:

account: browse anonymously, or get an account and write.

 user: pass:
register,

# Centrifugal vacuum balloon

Make a lighter-than-air cylinder by spinning
 (+11, -4) [vote for, against]

Have an stationary outer cylinder, magnetically supported by a spinning inner cylinder, supported by centrifugal force. Increase the diameter until it weighs less than air. Make it long enough to balance any extra weight needed at the ends. There may be something more elegant to do at the ends, but that will do.

This operates a bit like a gas-filled (helium or hot air) balloon, because in those, molecules bounce off the wall, and in this idea the spinning cylinder pushes off the wall.

 — caspian, Dec 21 2007

Smoke ring bubbles Toroidal_20bubbles
This sort of touches on the same ground, if you consider the toroid is spinning on itself. [Ling, Dec 22 2007]

Maglev trains http://en.wikipedia.org/wiki/Maglev_train
Article on maglev trains. The electrodynamic suspension and Inductrack are what I was thinking of [caspian, Dec 28 2007]

_22Prayer_20Wheel_22_20Vacuum_20Blimp [FlyingToaster, Jul 30 2011]

 So instead of using a lighter than air gas and a light flexible envelope, you use a low pressure gas and a rigid envelope. You support the sides of the rigid cylindrical envelope by magnetic repulsion of a smaller cylinder which is held rigid by centrifugal force.

 Hmmm. It does seem halfbaked...and without doing the math I can't say definitively that it won't work, but that is where my money would go. The whole magnetic support thing is too heavy, though I see it is necessary to avoid drag. The ends are and issue, and would have to be rigid enough to support the pressure. I guess you could use a long thin coaxial tubes or some kind of tapering at the ends.

I will Bun(+) you for making me think, but correct the spelling in the title, the sharks are coming.
 — MisterQED, Dec 21 2007

 ahh.....hmm.....umm....

Ignoring the ends for the moment, why not spin a cylinder up 'til you have 14psi centrifugal force then evacuate gas from it ? or is that what you said (and where does the magnetism come in).
 — FlyingToaster, Dec 21 2007

 Are you spinning this with magic?

 Lighter then air is always simple with unlimited energy and weightless engines.

 — mylodon, Dec 21 2007

 // Stationary outer cylinder, magnetically supported by a spinning inner cylinder. // Given just the right set of conditions, you can suspend something inside a magnetic field, and make it stable. You don't get a stable equilibrium for something perched on the outside of a magnetic field, unless it has a magnetic field of its own, in which case this becomes a generator. The drag will be about the same as if it were in physical contact.

 As [FlyingToaster] suggests, a simple spun cylinder would accomplish as much and be lighter.

The bit with the magnetic field is just "If I make it complicated enough, Ma Nature will lose track and I can get away with something." Often suggested, but it does not ever work.
 — lurch, Dec 21 2007

Oh, and be careful that you're not spotted building this..... the term "gas centrifuge" tends to attract large amounts of unwelcome attention and possible pre-emptive strikes.
 — 8th of 7, Dec 21 2007

 As the diameter of the outer tube increases, the magnetic force needed to support it also increases - exponentially (right?). One would presume these greater forces would require a heavier inner cylinder, which would require a bigger outer cylinder.

I could envision doing this with Roman technology. Lodestones would be used, and the inner cylinder spun up to speed using a series of spoked wooden gears, the largest slowly turned about by sweating, loincloth-clad slaves.
 — bungston, Dec 21 2007

The idea of spinning a cylindrical envelope to support the walls against a vacuum is ingenious!
 — MaxwellBuchanan, Dec 21 2007

Yes, I agree. I like new uses for centrifugal force. Croissant for creativity.
 — baconbrain, Dec 21 2007

Has anyone done the calculations? EG, assume the skin is 1mm thick plastic; calculate the radius of cylinder needed to support the mass of the enclosing skin (ignoring the ends, for the moment), then calculate the rate of spin needed to support the skin against 1atm of pressure. Should be easy.
 — MaxwellBuchanan, Dec 21 2007

 OK. Here's the first part. We'll consider a cylindrical balloon, and we're going to ignore the ends (so, this is strictly true only for a cylinder of infinite length - or for a toroidal balloon). Let the skin material have a mass of m grams per square centimetre. Let the length of the balloon be L, and let its radius be R. We want to find R such that the balloon is just neutrally bouyant, assuming it contains a perfect vacuum. OK? So:

 Mass of balloon, M, in grams is given by: M = mL x 2 x pi x R

 The volume of the balloon, V, in cubic centimetres is given by:

 V = L x pi x R" (where I'm using " to mean squared, for ease of notation)

 The density of air at sea-level is 1.2 x 10^-3 g per cubic centimetre. Hence, the bouyancy (B) provided by the balloon is:

 B = 1.2 x 10^-3 x L x pi x R"

 We want to find R such that M = B, in other words:

 mL x 2 x pi x R = 1.2 x 10^-3 x L x pi x R"

 Cancelling and rearranging, we get:

 R = 3.3 x 10^3 x m

 So, let's assume that our skin is made of plastic 1mm thick, giving a mass of roughly 0.1g per cm^2, in which case:

 R = 3.3 x 10^3 x 0.1 or R= 3.3 x 10^2 or about 3.3 metres.

In other words, a balloon about 20ft in diameter (and much much longer) would be neutrally bouyant if it could hold a vacuum.
 — MaxwellBuchanan, Dec 21 2007

 OK, so, next question. How fast does this balloon have to spin, in order to support its skin against a vacuum?

 Consider one square centimetre of the skin, having a mass (m) of 0.1 grams. It experiences a force of almost exactly 1000grams due to atmospheric pressure. This has to be opposed by centrifugal force (C), in grams, given by:

 C = m x w" x R

 where w is the rotation speed in radians per second, and R is the radius in metres (sorry for the change in units) and I'm using " to mean "squared". We know that R is 3.3 metres (see above), and so:

 0.1 x w" x 3.3 = 1000

 so w" = 1000/(0.1 x 3.3) = 3000 so w = 55 radians per second. This is about 8.7 rps or about 521rpm. This is pretty fast, but not unattainable.

One point to bear in mind is that, if the balloon leaks and air gets in, the skin will be ripped apart as if the whole thing had been positively pressurized to 15psi over atmospheric pressure.
 — MaxwellBuchanan, Dec 21 2007

So, in fact, this is sort of feasible. The only killer is the mass of the end-walls. The net force acting on them will want to dimple them inwards in the centre, and they'll need to be quite hefty. If you could make the balloon very, very long and then bend it round into a torus (doughnut) then it would need no ends. It would then rotate by "rolling" about the circular axis of the torus. Of course, the skin would need to stretch and shrink as the torus rotated.
 — MaxwellBuchanan, Dec 21 2007

 [MaxB], that's most impressive. All I'd come up with was the concept of a toroidal balloon.

A toroidal balloon would also have to be spun on the vertical axis through its center, would it not? Otherwise the extra area of the outer half torus (as the walls expand and contract) would cause an inward pressure. I don't know how that motion would complicate things.
 — baconbrain, Dec 21 2007

//A toroidal balloon would also have to be spun on the vertical axis through its center, would it not?// That's something I hadn't figured out, and I'm not sure I can get my head around it. My gut reaction is that there shouldn't be a problem, but you may have spotted something I missed. Can you explain a bit more?
 — MaxwellBuchanan, Dec 21 2007

I'll try to work out coaxial cones, which I think will work.
 — MisterQED, Dec 21 2007

 Have you considered the impact of (a) Coriolis force, (b) gyroscope effect, (c) Magnus effect ?

All these will affect lift and directional control.
 — 8th of 7, Dec 21 2007

//Have you considered the impact of (a) Coriolis force, (b) gyroscope effect, (c) Magnus effect ?//
(a) yes (b) yes (c) yes.
 — MaxwellBuchanan, Dec 21 2007

Jolly good.
 — 8th of 7, Dec 21 2007

 //Can you explain a bit more?//

 I'll try, but I'm not sure if I'm right.

 First, spin up an extremely long cylinder. If you look at it as two long half-cylinders, both have equal area. If you bend it into a torus, by bending toward one of the half-cylinders, that inner half, which is now the inner half of a torus, has to compress. That will reduce its surface area--assuming that it does not wrinkle--if it is elastic. So there will be less area to resist the pressure of the atmosphere.

 Which is the long way of saying that the natural shape of a pressure vessel is a sphere.

The hell with it. Just give the torus an extra few RPM around its circular axis, and everything will be fine.
 — baconbrain, Dec 21 2007

//So there will be less area to resist the pressure of the atmosphere.// Yes, but where the skin is contracted, it will also have more mass per unit area, thereby furnishing additional force.
 — MaxwellBuchanan, Dec 22 2007

Yeah, you are right. I plead cold/flu. I need to go take more meds, or maybe less.
 — baconbrain, Dec 22 2007

 Could you draw a picture of this?

 I'm having a hard time picturing a.... //stationary outer cylinder, magnetically supported by a spinning inner cylinder, supported by centrifugal force.//

 I'm guessing that the outer cylinder (a gas bladder) is pressurized by centrifugal forces created by the inner cylinder (a thin iron shell with paddles), right? And that the outer cylinder is purportedly supported by bouyancy (because it displaces more weight than its volume contains), right?

So.... the question is, "what makes the inner cylinder spin (with a constant power input to overcome the viscoscity of the vaccumed air) and how does it levitate (just saying magnetism is quite vague)?"
 — quantum_flux, Dec 22 2007

 There's a big helicopter underneath. The cylinders sit on top of the blades. The outside cylinder expands like the gown of a princess when she spins around at a ball.

 Except this helicopter's blades are flat underneath, so they deflect no air downwards. They rely entirely the dress expanding outwards to generate vaccuum.

 Helicopter blades rotate at 250-500rpm, much the same number as the one [MaxwellBuchanen] described.

 So when this helicopter spins to normal speed, the vacuum it creates causes the cylinder to reach neutral buoyancy with the air.

 And then -- voila. You have invented a helicopter that weighs the same as a normal helicopter, consumes more energy, and can't fly.

 So I guess the question is, how many pounds per rpm? I'm sure it's a function. My guess is this helicopter will have to have something like 500 x 10^5 rpm's coming off the top end, in order to lift off the ground.

 The gyroscopic forces will not allow it to tilt the propeller to go either forward or backwards, but it can employ more propellers for forward propulsion. Lighter then air is useless for going sideways through air, so they'd probably have to be conventional propellers.

 It would be a very stable flight, but I would worry about potential problems when the outer cylinder goes supersonic and then heads on off to light speed.*

* lower rpms could allowed by adding a lot of extra mass to the outer cylinder. higher rpms would then be required to lift this additional weight. Thus the outer cylinder would approach light speed.
 — mylodon, Dec 22 2007

//Thus the outer cylinder would approach light speed.// May I respectfully refer Mr. Mylodon to the foregoing calculation?
 — MaxwellBuchanan, Dec 22 2007

 I used the results of your calculation.

 I'm pretty sure your calculated 521rpm cylinder could lift a penny, if the penny could somehow power the fairy-silk spun contraption above it. (there would need to be a counter-rotating cylinder or toroid below the penny, however -- otherwise the penny would just spin)

This is just an obfuscated helicopter.
 — mylodon, Dec 22 2007

 // obfuscated helicopter //

No, there is an important difference; the helicopter rises on thrust generated by a downdraught created by a rotor; therefore it is a newtonian "reaction drive". This idea relies on mechanically generated bouyancy for lift.
 — 8th of 7, Dec 22 2007

But wouldn't the force of the particles under the balloon, pushing up, be the same as the generated force under helicopter blades (lifting the same mass). And of course, above is vacuum.
 — mylodon, Dec 22 2007

 For a given lifting weight, yes. But there is an important distinction between lift and buoyancy, as any fule kno.

 Consider a submarine. Once submerged and at neutral buoyancy, it can change its submergence depth by two methods:

 1. By pumping out water ballast, it will reduce its average density below that of the surrounding water; thus, it will rise. In this case, the pressure is exerted over the whole hull area below the centreline.

 2. By use of hydroplanes, it can convert momentum (forward motion) into "lift", thus also causing it to rise; but in this case, the lifting force acts only on the hydroplanes, not on the hull.

 A submarine with negative buoyancy will sink until it reaches an equilibrium point (where the density of the water matches the desnity of the boat), but can maintain level or even rise by using forward propulsion and hydroplanes to bring it up.

The consequence of the two approaches is the same, but the physical mechanism is different.
 — 8th of 7, Dec 22 2007

 I made a wrong assumption in the text of the smoke ring bubbles link, but there is an idea half buried/stated there that uses a spinning toroid bubble film.

The idea of balancing centipetal force against pressure is rather neat, but in the similar way that a spinning toroid bubble film probably wouldn't be so stable, I think any imperfections in caspian's idea would lead to local bulging & failure.
 — Ling, Dec 22 2007

 How did you visualize a helicopter from this idea [mylodon]?

 I'm getting a rotating cylinder within another larger cylinder kind of feel for this idea. Actually, the larger the outer cylinder is, the more total pressure-flux (force) will be on it, which would require structural beams to keep it from imploding assuming there is a vacuous condition inside. I take it that you are trying to cause a dynamic pressure against the skin of the outer cylinder in order to counteract the crushing pressure differential from the outside so that the outer cylinder is supported. Well it is an interesting concept, but it seems that this elaborate method would be awefully inefficient, if not down-Wright impossible.

 But aside from that, you would probably want 2 inner cylinders with innermeshing paddles that are spinning in opposite directions to achieve a net angular momentum of zero as a stability issue. It would be like those things at the car wash that rotate really rapidly in opposite directions and bang up the side of your car with rags, except it would be within a large cylinder, I still don't see how this could fly though.

I guess the next stage would be testing your crazy idea, but what the heck. [+]
 — quantum_flux, Dec 22 2007

Testing? are you insane? We're just at the point where we need to be considering our necessary funding levels!
 — lurch, Dec 22 2007

 Something's missing. Using [MB]'s numbers, I find that the 3.3m cylinder surface is travelling at 3.14 x 6.6m x 8.7 rps = 90.48 m/sec, or about 202 mph = 328 km/hr!

Along with needing unobtanium plastic sheet, the drag will kill it. It could conceivably operate in a vacuum, but that would rather spoil the whole purpose.
 — csea, Dec 22 2007

 Start small. Try using two 64 Oz plastic cups as the outer cylinder and a lightweight motor on each end and go from there. I know this is a fire hazard, but the Wright Brothers did some crazy stuff too. If you turned on the motors you should get some positive pressure in your contraption, so you would want to, somehow, vacuum out the air until the pressure on the inside equalizes with the pressure on the outside. The key is to make the cups, the battery, the inner cylinder paddle contraption, and the motor weigh less than the surrounding air while providing sufficient dynamic pressures to keep the cups from crushing in on themselves. I don't think it will be that easy to make it fly, but it's up to [caspian] to prove me wrong by making it fly through trial and error. Also, you could submerge your cups under water in order to test how much pressure difference it can stand and if it's airtight (but don't get the motors wet). Good luck!

An afterthought: This small of a system would definantly not float in the air, but it sure could float in water. I suppose you'd really have to go larger in order to have it float in air, but it would function as a small scale model for underwater pressure testing purposes in finding out how much dynamic pressure your motors could supply at various RPMs and with various fittings.
 — quantum_flux, Dec 22 2007

 ...actually, why not do something with the ends ? Specifically, make each end of the cylinder into a propeller; the skin is attached to the blades (or more exactly to a ring which is attached to the blades).

Design the propellers such that the rotational speed necessary to make the skin weigh 14psi is the rotational speed necessary for the propellers to evacuate the cylinder.
 — FlyingToaster, Dec 23 2007

 But the air inside the cylinder will be rotating with the cylinder, after just a few minutes, and will be moving right along with the blades.

Have we lost track of the original idea? That was a stationary outer skin that would have no air drag, held by repulsion from centrifugally supported magnets.
 — baconbrain, Dec 23 2007

 // I think any imperfections in caspian's idea would lead to local bulging & failure.// This is a real flaw, I think, and you're right. Any bulge will put that part of the skin further from the axis of rotation, increasing the outward force proportionately, with no corresponding balancing force. Hence, aneurysms will tend to develop unless the skin is quite strong. Fortunately, there's no tendency for the rotational axis to drift off-centre (since centrifugal force is linear with radius), as far as I can tell.

 Nevertheless, the idea does have considerable merit. Above all, it changes the classic "vacuum balloon" problem from one of building an immensely strong vessel in compression into a problem of building a modestly strong vessel in tension, which is very very much easier.

Also, all the numbers get very much nicer as the thing gets bigger. If you wanted to build a very very very large lifting body, this might be the best way to do it.
 — MaxwellBuchanan, Dec 23 2007

 // unless the skin is quite strong //

Carbon fibre composites are extremely strong, light, and exhibit minimal measurabe creep under extreme tension, right up to their failure point.
 — 8th of 7, Dec 24 2007

 You're all barking nutters.

 It's also effing brilliant and I wish I'd come up with it.

 Suspended within a stationary outer cylinder, but does the outer cylinder need to be complete? There will be a resultant significant skin drag, but that could be used as a sort of caterpillar drive - expose that part of the inner cylinder that is going opposite the desired direction of travel, and the skin drag pushes the vacoon in whatever way you want to go.

Assuming the wind doesn't dictate otherwise.
 — elhigh, Dec 24 2007

 I think the suspension is simple enough: the inner cylinder has a few Halbach arrays on it, and the outer cylinder does too. Halbach arrays are passively stable once they lift off, and they lift off at pretty low speeds. So there's your suspension. Add a couple more arrays at the ends to keep the cylinders centered on each other, and that eliminates side-to-side shifting. So now you can place a load on your vacuum balloon without actually touching it. The extra mass of the Halbach array may also permit a slower rotation speed for vacuum inflation.

For a linear balloon I think you'd need something that was pretty rigid along the spin axis. For the torus, I got nothing.
 — elhigh, Dec 24 2007

 // It becomes a self-limiting system, defeating its role as a lifting device. //

But it would be sooo awesome to actually do.
 — elhigh, Dec 24 2007

Cylinders suffer from one major drawback, (in terms of forces) their ends don't justify their mean.
 — 4whom, Dec 24 2007

 [Max], I am so pleased that you posted the physics.

 You all took something completely different away from this idea. I thought that it was the inner magnetic cylinder that spun, and by doing so, induced a magnetic field in the outer cylinder. The two fields repel one another. And the outer cylinder moves away from the spinning inner one.

 If the outer balloon were stationary, one would want large fins on it, since under the influence of the magnetic fileds it will begin to spin too, thus decreasing the motion difference between inner and outer and decreasing the magnetism induced. The idea of keeping the outside still and the inside in motion would be good as regards durability as well.

 With electromagnets on the inside one could alternate the current in the inner cylinder, maintaining the magnetic repulsion but changing direction to prevent spin in any one direction from building up.

 It may be that if the inner cylinder spins very fast, such that the outer cylinder cannot possibly catch up because of air resistance, the inner cylinder could turn the outer cylinder (retaining centrifugal vacuum balloon element) but also contribute magnetic repulsion as well, so the outer balloon does not have to spin as fast. This would be a combination of centrifugal and magnetic evacuation of the outer balloon.

Magnetic repulsion works better the closer together things are. The inner "cylinder" should be able to change in diameter to keep closely apposed to the outer skin. I propose that instead of a cylinder on the inside, the interior driving apparatus be a compressible / expandable metal mesh cage. I have seen desktop toys like this made out of wire. It would also be similar to the metal stents used to treat arteriosclerosis and such. As the exterior balloon expands in diameter, the interior cylindrical cage also expands to keep close to the interior surface of the balloon. Also the cage would be lighter than a solid.
 — bungston, Dec 24 2007

//You all took something completely different away from this idea.// Yes, it's one of those posts that triggers a bunch of related ideas. I never did get the idea of the concentric cylinders, but it turns out that a simple spinning cylinder is close to being a workable idea. I suspect that if you made this thing a few tens of meters in diameter, and a few hundred metres long (comparable to a conventional airship), it might become quite feasible.
 — MaxwellBuchanan, Dec 24 2007

 The thing with the spinning cylinder is how it is kept spinning I guess. If not magnetically, I suppose there could be fore and aft compartments, each with massive vanes to keep them from rotating, one containing the motor and one the far anchor of the axle. It could be powered by one of those Ms Fusion doohickies, or a zed-point jobber from Stargate.

 I was hoping, [max] that you would make with the physics formulas again, but this time using magnets instead of spin.

I
 — bungston, Dec 24 2007

Well ..... three cylinders, the two end ones half the length of the middle one, and contrarotating, will balance the torque.......
 — 8th of 7, Dec 24 2007

An altogether happier solution would be to have two cylinders side by side, with a modest gap betwixt them. The driving mechanism would be a sort of figure-8 affair at each end of the pair of parallel cylinders. The two cylinders would counter-rotate, and this arrangement would give additional stability.
 — MaxwellBuchanan, Dec 24 2007

And then we put the frog into the gap, for acceleration... Oh, no, sorry, that was a different idea.
 — lurch, Dec 25 2007

 The "parallel cylinders" idea uses two axles instead of the single shaft of my single segmented cylinder design, and requires a more complex drive mechanism.

<sniff>
 — 8th of 7, Dec 25 2007

 As MisterQED suggests, I'm not using just a single spinning cylinder because it would have drag against the air. I can't spot any spelling mistakes in my title, which he mentioned. Are the hyphens controversial?

 Both the spinning and repulsion magnetic forces I had thought of as being like maglev trains. I'll link to the wikipedia article. It would need power to keep the inner cylinder spinning against magnetic drag. Stable equilibrium is possible, unlike with unmoving magnets.

 The gap between the inner and outer cylinders is kept as small as possible to increase magnetic forces. The inner cylinder doesn't need to be air-tight since it have vacuum inside and out.

 [quantum_flux] You could picture the outer cylinder as a cylinder made of many rings of maglev rail, but I don't know what maglev tracks look like. I had thought about 10cm thick laminated combination of aluminium (carries induced current) and plastic (insulator). It's not a gas bladder with gas on the inside, it has gas on the outside and vacuum on the inside. If anyone else wants to draw a diagram that would be great. //I'm getting a rotating cylinder within another larger cylinder kind of feel for this idea// Yes, that's what I meant. I didn't imagine any paddles though.

 [elhigh] Yes, Halbach arrays look good.

 I don't have quite enough equations to see how well this would work. Wikipedia's Inductrack page says 200:1 lift-to-drag ratio at 500 km/h, presumably with more lift than weight of maglev components (it's for trains). Also that drag decreases inversely with speed beyond a certain point. Constant power for a given force would make sense, since both correspond with a given current.

 That should be enough to estimate power required per unit of force from air pressure. Not that easily though, or I'd do it now. Then knowing air pressure, you could estimate power required per unit of surface area.

 I have a solution to the cylinder end problem. Make the outer shell a sphere instead. Instead of the inner cylinder, have three sets of spinning rings. One set spins around the Y axis at the right set of speeds to balance half the X and Z axis components of the air pressure on the outside shell. The other two sets spin around the X and Z and each balance half of two axis components of the air pressure. The rings are no longer within the sphere volume, instead, they are embedded in the sphere shell.

I had not considered the possibility that the cylinder could expand. It's simpler if it doesn't.
 — caspian, Dec 28 2007

 You should consider filling the gap between the inner and outer skins with hydrogen or helium. The light, highly mobile gas will reduce drag and turbulence, and even give a bit more lift.... but this will be negilgible compared to the lift of the device itself. Some gas will of course diffuse out (in both directions) but the loss can be managed - it is in conventional airships.

 The idea of a sphere on gimballs is intriguing, but there remains the problem of power transmission through the axles. In many aircraft instruments - admittedly tiny - multi-axis gyros are powered by air pressure (actually, a vacuum pump). Do you really need to spin the sphere in 3 dimensions ?

 If the cylinder is vertical, with the payload/powerplant suspended from the vertical central shaft, this simplifies the design.

<later> There's a film about alien contact which shows a device with a "spinning ring" design.
 — 8th of 7, Dec 28 2007

 Ezekiel 16The appearance of the wheels and their work was like unto the colour of a beryl: and they four had one likeness: and their appearance and their work was as it were a wheel in the middle of a wheel.

 17When they went, they went upon their four sides: and they turned not when they went.

18As for their rings, they were so high that they were dreadful; and their rings were full of eyes round about them four.
 — bungston, Dec 28 2007

Oddly, that occured to us, too. But not as interesting as the book of Exodus ....
 — 8th of 7, Dec 28 2007

23a And yea, even as their wheels were thrust heavenwards with much noise and rushing of wind, there came three wise men. "Verily," quoth the first "the force which is centrifugal doth support the nothingness within." Then spake the second, saying "Thou shalt be smitten in twain, for hath not the Lord said that the force centrifugal is nought but a deception of the devil? The force centripetal is the True Way!" and, so saying, he smote the first wise man in twain, and thus was his prophecy fulfilled. Then spake yet the third of the wise men, crying "Couldst one not fill it also with custard?"
 — MaxwellBuchanan, Dec 28 2007

23b : "And there was much rejoicing."
 — 8th of 7, Dec 28 2007

 I just realised that the amount of structural material needed to support a given force from each end of the cylinder increases in proportion to cylinder length. So my idea of ignoring it by scaling up the cylinder length doesn't work.

 There are a couple of ideas already in the annotations to fix this. One was to loop it around and join the ends making a toroidal. My idea was supporting it with 3 sets of rings spinning around different axes.

A possible issue with making the cylinder into a toroid is the cylinder ends can still kind of compress together by making it into a smaller toroid. This may be fixable with centrifugal support rings around the toroid main axis.
 — caspian, Jul 30 2011

 Everybody keeps mentioning 'magnetic' this and that, which in my experience requires either heavy rare earth magnets or heavy electromagnets (ceramic magnets being shite). Key word being heavy, a word typically anathema to LTA aircraft design.

Am I missing something here, or is everybody else?
 — Alterother, Jul 30 2011

//magnetic// From the looks of things, the magnetic bit is just to keep the outer shell static while the inner shell spins to avoid outside interference. The main idea is centrifugally-induced vacuum which is brilliant. My own take <link>
 — FlyingToaster, Jul 30 2011

 Oh, I agree that it's a work of frigging genius. I bunned the hell out of this sucker. Well, just one bun, but y'all know what I mean.

I just don't have the mathematical chops to figure out for myself whether the magnets will make it too heavy to fly.
 — Alterother, Jul 30 2011

 I can't think of any reason why you couldn't magnetize a sheet of material with North on one side and South on the other; just pile sheets on top of each other and magnetize them all at the same time... no clue how long it would hold its magnetic properties though.

 I just took the magnetic bit as a non-sequitur: you could just as easily make the partition between shells airtight'ish with a bit of overpressure or simply drape the outer shell on a scaffold of some kind.

 However, if you have a heavy inner shell it means you don't have to spin it up as fast.

Calculations are basically the same as any other LTA process: displacement of an amount of air weighting <x>.
 — FlyingToaster, Jul 30 2011

Right, but as with any LTA, the bigger you make it, the tighter that ratio becomes, because you have to have sufficient structural support for this giant empty shape you've built. Even ultra-light space age materials will only take you so far. That's why blimps were/are generally more prevalent than rigid airships (discounting the Hindenburg factor).
 — Alterother, Jul 30 2011

That's why it's being spun: centrifugal force makes the inner shell expand outwards, making a vacuum inside.
 — FlyingToaster, Jul 30 2011

 That makes sense. How does it support its own weight when idle? Is the inner cylinder collapsible?

Sorry if someone already mentioned that and I missed it. I'm missing things right now.
 — Alterother, Jul 31 2011

 I'm assuming it's just fabric and scrunches up when unspun. The problem is that even though radial compression is taken care of by spinning, axial compression isn't, ie: it will remain scrunched up longitudinally without support of some kind.

Or, given the "magnetism" bit, it could be iron foil or something in which case you wouldn't evacuate it until it's spun up, and you'd repressurize before letting it spin back down.
 — FlyingToaster, Jul 31 2011

If you used iron (or "iron-ized") foil, you'd probably have to re-magnetize it fairly frequently. Not a difficult issue to overcome, but an additional ongoing requirement of operation.
 — Alterother, Jul 31 2011

 Could this be another part of the reason a Frisbee flies up? Could it be that the air inside spins and is sent out of the underside, so that there is low pressure which is pushed upwards by the atmosphere?

Perhaps a reason to bring back my (I think deleted) idea for a FrisbeeCopter
 — pashute, Aug 12 2011

I think a frisbee is more a of 'spinning fixed-wing', in that it works just like the wing on an airplane, and the spin simply stabilizes the flight. Somebody help me out here?
 — Alterother, Aug 12 2011

 You have to take the bernoulli effect into account, as in aircraft wings. The drum surface is going to be moving at ridiculous speeds. Also I think that the drum under spin will be stable. It wouldn't be in a vacuum, but air flow I think will damp out instability. maybe. If not, we can structure an outer wall as a foil bearing, much simpler and lighter than magnetic supports. Maybe even doable.

 Aerodynamic effects are to our advantage though. If the drum edge is at the speed of sound, then the effective pressure I think is 0. The thing will then throw itself apart due to centrifugal pseudo- forces.

 The balancing force is actually:

 (1) Patm*(1-(v/c)^2) = rho*v^2/r

 where v is the speed of the drum surface spin, c is the speed of sound, rho is the mass/m^2 of the drum cylinder wall material, Patm is atmospheric pressure, r is the cylinder radius.

 So we can get v from:

 (2) Patm/(rho*c^2/r + Patm) = v^2/c^2

 The mean cylinder density, taking into account just the walls and a contained vacuum is:

 (3) 2*rho/r

 That needs to be below atmospheric density to have lift. In fact: 75% atmospheric for a hot air balloon, 14.7% for helium, and 7.3% for hydrogen, to get a ballpark idea of what we would like. Atmospheric density is 1.225kg/m^3

 For the hell of it, assume a Ti 6Al 4V alloy foil for the cylinder wall, 0.1mm thick. Density is 4470 kg/m^3 for that alloy, 0.447kg/m^2 for a foil.

 That gives 73cm as a breakeven radius. You'll need a radius of 5m to match He density, at 95% the speed of sound or 10.3 rotations/s (620 rpm)

 The other thing that aerodynamics gives is the end plates. Turns out that the pressure acting on a circular end plate with the edge moving at the speed of sound (340m/s) is half the force on a stationary plate. That means that axial support only needs to hold half the force, so it's half the mass needed to hold stationary end plates apart. Also conveniently from an implementation point of view is that the pressure is greatest at the center, decreasing to zero at the edge, so a properly designed axial support truss might be feasible.

 I took a look at Ti 6Al 4V again, and it's maximum buckling compression strength is about 10000 atmospheres. So a rigid support needs a cross section area of 1/20000 of the total end plate area. You can work out an equivalent density from that: 4470kg/m^3/20000 = 0.2235kg/m^3, 18% atmospheric density.

 Unfortunately you have to add the axial support density and cylinder wall density together, so in total we have 18%+14.7% = 32.7% which is still way better than a hot air balloon.

 We can do better. Thinner walls? 0.025mm alloy foil is commercially available. And carbon composites for the axial support would reduce that by about 30%, taking us to maybe 12% atmospheric density for the axial support and 3.7% for the cylinder walls, and spinning a little closer to the speed of sound. About 15.7%, close to helium.

I haven't made any allowance for the end plates themselves at all here.
 — halfbaked_gypsy, Dec 11 2016

I have a good feeling about you, halfbaked_gypsy.
 — bungston, Dec 11 2016

Wow, no doubt. That's some impressive number slingin' there partner.
 — doctorremulac3, Dec 11 2016

Wow, no doubt. That's some impressive number slingin' there partner.
 — doctorremulac3, Dec 11 2016

 Had to fix some errors in the comments above. I just noticed, if you require lift, then there is a minimum edge speed resulting from (3)

 It must spin faster than 261m/s edge speed. Of course less than the speed of sound, 340m/s.

 Between about 75% and 100% the speed of sound.

 Commercial jet aircraft fly at about 75% the speed of sound.

Startup and shutdown I don't see a problem with. At stop, it contains atmosphere, no pressure differential. On startup, we spin it up slowly, while evacuating it at a rate so that the pressure differential always balances the spin pseudo-force at the rim. We can use this same mechanism to adjust lift as desired in flight.
 — halfbaked_gypsy, Dec 12 2016

end caps - concave hemispheres. For axial support, a double layer wall could be used, like a blood-pressure cuff.
 — FlyingToaster, Dec 12 2016

 Hmm. Why a double wall?

 Single wall, using Ti alloy as before, the cross section devoted to support, the drum material itself, needs to be 1/20000 total cross-section area. Means the drum wall must be 1/40000 of radius, 5000mm/40000 = 0.125mm strangely enough. Kind of close to the 0.1mm I pulled out of nowhere.

It wouldn't have to be strictly concave, just a sheet under tension would do. We would use two reasonably weighty rings where those sheets join to the drum, so that centrifugal pseudo-force keeps the sheets under tension, and the net force on the drum foil is strictly up/down, no radial component. Might be tricky to balance precisely.
 — halfbaked_gypsy, Dec 12 2016

 Big problem is that drum wall is thin. It would have a very large length-to-thickness, making it mechanically unstable: it would buckle, unless it was made a lot thicker or suitable bracing was applied at close intervals.

 If it's is aerodynamically stable, maybe the spin and aerodynamics of the thing would supply enough bracing?

Or an outer second wall may be needed for safety anyway, that could be configured as a foil bearing. It wouldn't be load bearing so could be very light, maybe just a chicken wire mesh would do.
 — halfbaked_gypsy, Dec 12 2016

What [bungston] said. I wish I were unlazy enough to help out.
 — Voice, Dec 12 2016

//double wall// Pressurized between, to hold against axial forces... like a long skinny donut. A concave hemispherical cap complements it by putting all the force axially along the edge, as opposed to a flat cap which tries to pull the edge inwards.
 — FlyingToaster, Dec 12 2016

 Sorry, but concave is not relevant. For the mechanics, it is what is happening right on the edge where the end plate joins the drum that is relevant.

 Either the edge ring is rigid, so that any radial force from tension in a thin sheet doesn't carry to the drum (or the entire end plate can be rigid) and/or the edge ring, together with the weight distribution in the end plate, under spin, cancel any residual radial force on the edge of the drum.

 Away from the cylinder edge, what shape the end cap takes is irrelevant.

 Of course the lightest thing in that case is a thin sheet, which will of course look concave under pressure.

 There is still the problem of buckling in a too-thin cylinder wall under compression, just like stomping a soda can.

 If a stationary outer cylinder is configured as a foil bearing, then there can be up to about 100 psi (7 atm) of inward force from the hydrodynamics of how those work, against spin forces pushing out: up to 100 psi of "bracing" force on the cylinder walls to prevent buckling.

 So double wall, yes, but the outer wall is stationary and not hermetic. The inner wall flies or hovers over the outer wall in the way a foil bearing works, same way a HDD head flies over the platter.

In a fluid bearing, the fluid between the drums can be at a higher pressure than the ambient fluid. The speed of sound could be much increased, so the cylinder could spin much faster. The bracing forces could be much increased. We'd probably have to stick to one atmosphere though, to avoid heavy tension belts or mechanical supports.
 — halfbaked_gypsy, Dec 13 2016

 That makes it a different device again now. Three layers.

 Outer wall is at atmospheric pressure.

 Between both walls is a fluid-dynamic gas bearing.

 The inner drum is a sheet under spin, containing a vacuum.

 Ignoring mechanical forces (we don't want to pay the weight penalty) then the outer drum balances pressure against outward force from the inner drum spin.

The inner drum has to balance fluid-dynamic pressures against spin force. So we're limited to one atmosphere of bracing I believe. I think that might dictate the thickness of the drum wall.
 — halfbaked_gypsy, Dec 13 2016

Thing is, a vacuum balloon will only give you a tiny advantage over a helium (or even hydrogen) balloon in terms of lifting capacity, and any such advantage ignores the greater mass of structure needed for a vacuum balloon.
 — MaxwellBuchanan, Dec 13 2016

 We tried it - It created a virus so lethal the subject was dead before he left the table.

Okay, maybe not. Never mind ... mumble ... welcome to the Halfbakery, [halfbaked_gypsy].
 — normzone, Dec 13 2016

Indeed, welcome, [gypsy].
 — MaxwellBuchanan, Dec 13 2016

 hmm...

If you stuck a propeller on each end, spinning with the cylinder, then they could pull the cylinder taut between them, axially.
 — FlyingToaster, Dec 13 2016

 "We tried it - It created a virus so lethal the subject was dead before he left the table."

 Caught one of those the day I started work. 9-5 meme.

Just another brick in the wall now.
 — halfbaked_gypsy, Dec 14 2016

 Ok, think I have it now.

 Outer drum, stationary, foil with lots of pinholes, outer wall of a foil bearing. Under spin, it is under tension: aim for the equivalent of containing 2-15 atmospheres pressure.

 Pressure comes from internal foil drum, containing vacuum, spinning fast enough to generate the 2-15 atmospheres of centripetal force plus one atmosphere.

 That results in a thin cylinder wall, effectively buttressed by 3-16 atm of force every square inch (0.75-4 atm every half inch etc) which can hopefully hold the end caps apart.

 Now we're into truss theory. trusses are spaced as close as reasonable to keep the weight low, but prevent perturbations that make long (thin) supports unstable under compression. Think stomped soda can. If you can prevent the walls from moving in/out, it won't collapse.

For point-wise bracing at regular intervals, the calculations are straightforward, but for a continuous force along the length, I don't know how to do.
 — halfbaked_gypsy, Dec 14 2016

 "Thing is, a vacuum balloon will only give you a tiny advantage over a helium (or even hydrogen) balloon in terms of lifting capacity"

 Hydrogen has 92% the lifting capacity of vacuum. Helium 84%, so vacuum, ignoring containment issues, is only a little better than hydrogen, yes.

 The issues are that everyone is scared of hydrogen since the hindenburg disaster, and helium is expensive and of limited availability, not good if you want to maintain a large rigid airship fleet.

Helium actually costs mass to use. Since it is so expensive, it isn't vented. When you want to reduce lift, you have to compress it. Compressors and pressure cylinders are heavy. I don't think hydrogen airships bother with trying to keep the hydrogen, so no comperssors. Just spare hydrogen cylinders - have to confirm that though.
 — halfbaked_gypsy, Dec 14 2016

 // hindenburg //

 Most of the casualties on the Hindenburg were caused by burning fuel, not the hydrogen.

 Up to that point, hydrogen-filled airships had a good track record, R101 notwithstanding.

 Part of the outcry against Zeppelins was because they were Nazi propaganda tools.

 With the proper precautions, hydrogen airships are actually acceptably safe. Hydrogen disperses very fast in air. Why complain about a hydrogen balloon, but then cheerfully board an aircraft with thousands of litres of AVGAS in its tanks ?

 Don't even mention helicopters …

Oh, and welcome to the 1/2B, [hb_gipsy]. Don't worry about the pedantry, in-jokes, grammatical criticism and generalised abuse. That happens to everyone.
 — 8th of 7, Dec 14 2016

 //Hydrogen has 92% the lifting capacity of vacuum. Helium 84%, so vacuum, ignoring containment issues, is only a little better than hydrogen, yes.//

 You could improve the performance of those lifting gasses by using less of your chosen gas. This has the irritating side-effect of reducing pressure. But you only need the pressure at the edge of the envelope to balance the atmosphere on the other side. Could you just spin the gas into a massive longitudinal stable vortex? Then the pressure acting on the inside of the envelope could be achieved with less gas.

 Could you make do with just air? Start with a full, still envelope. Spin up the gas while pumping air out to maintain 1ATM on each side of the envelope... wait, you don't even need to pump it, just have a leaky envelope and excess pressure will equalize. Then you increase/decrease the speed of the vortex to control lift.

and welcome gypsy.
 — bs0u0155, Dec 15 2016

Helium is monoatomic. Hydrogen is duoatomic, unless you use big lasers to make plasma, then the hydrogen is monoatomic, so that would be even lighter than regular hydrogen yet less awesome than a spinning vacuum balloon
 — beanangel, Dec 15 2016

 I have some reservations. How is this driven? Are the motors constantly on or are the motors spun up while the vacuum is installed just before liftoff?

The extra plumbing details may scuttle this ship.
 — wjt, Dec 16 2016

 Tensairity beam calculations apply. The variation is that with this spinning cylinder case, effectivity the pressure is anisotropic:

 Force per area at the center of the endplates is inward, and force per area is uniformly outward on the drum walls. Straight pressurisation is the same everywhere.

 A very interesting result is: "The buckling load increases with the square root of the pressure and is independent of the length of the compression element." The "pressure" in this case being the effective stiffness of the air bearing carrying the spinning drum.

 Got all the equations I need, just need to make some quick calculations to see if it's physically possible.

 As noted by 8th of 7, Hydrogen is far and away the most practical. I believe it should be possible to come up with an envelope fabric that won't fail catastrophically if the H ignites. Say the envelope is stabbed and the H ignited, ideally the envlope won't suddenly completely disappear leaving you faced with a massive fireball. Something like a glass or ceramic fabric, maybe metallised like a crisp bag to reduce H leakage through it.

 [edit: actually all the containment fabrics I see for H/He are metallised. Reduces H/He diffusion out, just as the crisp bags are intended to reduce oxygen diffusion into the bag. They are metallised with Al and are highly flammable, not a good mix with hydrogen]

 Given that and the expense/unsustainability of using He commercially leaves me mystified why people insist on He over Hydrogen.

I guess there is just no incentive to do the work to make H based airships safe, and prove they're safe.
 — halfbaked_gypsy, Dec 18 2016

 // do the work to make H based airships safe, and prove they're safe //

 It's been done.

 In WW1, special incendiary ammunition had to be developed to down German airships, and even then it took a lot of it to do the job - hundreds of rounds.

 Until the Hindenburg, airships had an enviable reputation for safety. The R101 was effectively on a test flight and is entirely unrepresentative.

Many of the other hull losses were of military ships, often being deliberately operated under stressful conditions where a civil airship would not have gone.
 — 8th of 7, Dec 18 2016

 //The R101 was effectively on a test flight and is entirely unrepresentative. //

 It's worse than that. The R101 was one of two airships built concurrently: one (the R101) was designed and built by the air ministry, and was terrible: overweight, with leaky hydrogen bags, and designed by the same people who had designed the R38 (which broke up in flight, because no structural calculations had been made in designing it).

 The R100 was the R101's competitor, built by Vickers under the supervision of Barnes Wallis and Neville Shute. It was designed by people who knew what they were doing, and exceeded its design specs, and performed more or less flawlessly on its maiden voyage.

 The R101 was lost for many reasons, but mainly because it was a government project run by people who had too much budget and not enough expertise. It crashed not because of fire, but because it was overweight and because of badly-designed gasbags which chafed, tore and leaked, leading to a loss of lift in bad weather.

 After the R101 crashed, the R100 was broken up for scrap, despite its success.

If anyone wants to read a really, really stonkingly good book, they should find a copy of Neville Shute's book "Slide Rule", which describes his role in designing and building the R100.
 — MaxwellBuchanan, Dec 18 2016

 // Barnes Wallis and Neville Shute. It was designed by people who knew what they were doing //

 Once again, we are awed by your capacity for understatement. Barnes Neville Wallis, designer of the Wellington, the "Upkeep" dam-busting bouncing mine, and the Tallboy and Grand Slam "Earthquake" bombs; Neville Shute Norway, a leading light of D.M.W.D. and originator of the Rocket Spear, the Harvey Projector, and numerous other extemporised weapons.

"People who knew what they were doing" doesn't even begin to come close ...
 — 8th of 7, Dec 18 2016

 I find that understatement is, occasionally, somewhat useful.

Anyway, you're just grumpy because you weren't the first to spot the opportunity to mention Wallis and Shute in the same annotation.
 — MaxwellBuchanan, Dec 18 2016

 [annotate]

back: main index