h a l f b a k e r yExpensive, difficult, slightly dangerous, not particularly effective... I'm on a roll.
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
browse anonymously,
or get an account
and write.
register,
|
|
|
A flywheel is a great way of storing energy. One of the big problems with flywheels is the loss of energy due to friction in the bearings. My other flywheel idea had friction which caused 1800W of loss in a flywheel storing 15kWh which is clearly not practical.
So contactless bearings (ie magnetic
bearings) are an obvious choice to eliminate the friction losses.
My design for a flywheel with magnetic bearings is as follows.
Take a big aluminium ring; 1000kg, radius 15m. Place ring in a ring-shaped trench with magnets lining the inside and outside walls of the trench. Once the ring starts spinning, the magnets will cause an eddy current to form in the ring which will cause a magnetic field to repel the magnets by Lenz's law, thus causing the ring to levitate. Loops of wire are placed around the trench so energy can be inputed (thus spinning the ring faster) or outputed (thus slowing the ring down).
The ring could be spun arbitrarily fast (storing an arbitrarily large amount of energy), with the only constraint being the strength of the aluminium ring. I think it could easily store 15kWh (a good amount to comfortably run an off-the-grid house being supplemented by local energy sources: wind, solar etc).
Although energy is no longer lost to friction, instead some energy is lost to electrical resistance. I do not have an intuition (nor a quick calculation) as to how much energy would be lost to electrical resistance.
Lenz's law is hard to visualise, so I have no idea of suitable orientation of magnets, or even whether it could work at all.
(?) These guys know.
http://gltrs.grc.na...TM-2002-211159.html
[2 fries shy of a happy meal, Feb 06 2011]
illustration
http://i.imgur.com/VKiCl.jpg
[xaviergisz, Feb 07 2011]
cross section
http://i.imgur.com/15mEe.jpg
[xaviergisz, Feb 07 2011]
cross section - tilted
http://i.imgur.com/x1Ns0.jpg
[xaviergisz, Feb 07 2011]
Foucault's Flywheel
Foucault_27s_20Flywheel
[xaviergisz, Feb 07 2011]
yet another illustration
http://i.imgur.com/kd5w2.jpg
[xaviergisz, Feb 10 2011]
IEEE article "New concept flywheel-ring flywheel generator"
http://ieeexplore.i...jsp?arnumber=264190
Article available by subscription only unfortunately.
Envisages a ring shaped flywheel 320m in diameter weighing 320,000 tons and storing 5GWh. Wow! [xaviergisz, Mar 30 2011]
Please log in.
If you're not logged in,
you can see what this page
looks like, but you will
not be able to add anything.
Destination URL.
E.g., https://www.coffee.com/
Description (displayed with the short name and URL.)
|
| |
My first thought is that you are expending enough energy to keep the ring suspended against the force of gravity, plus losses. |
|
| |
The will be a definite upper limit to the speed, not 'arbitrary'. Aluminium is not particularly strong. |
|
| |
OK ns, I'll re-phrase that sentence: |
|
| |
The maximum speed of the ring flywheel is constrained by the strength of the aluminium ring. |
|
| |
"Operation was not continued beyond this point because of
... inadequate containment in case of failure." Not
encouraging. |
|
| |
structures suspended by springs, even magnetic springs, have bad tendencies to oscillate. Add lots and lots of energy and the ring will need to be balanced for 3rd order motion. Any distortions in the ring and we have a small disaster on our hands. |
|
| |
I was hoping the large size and relatively low speed (e.g. maximum speed of a few hundred rpm) would help avoid instability problems. |
|
| |
Works out about 200 rpm... doesn't sound too stressful as long as nothing starts wobbling; an aluminium hula-hoop of about 7cm diameter. |
|
| |
Of course you might have to hold the Earth still or put the trench in at an angle. |
|
| |
//Of course you might have to hold the Earth still or put the trench in at an angle.// |
|
| |
I had in mind a spherical shaped trench to allow for the Earth's precession (see illustrations). The red and grey blocks are magnets; red represents north, grey south. |
|
| |
I would've thought that you would just need to have the axis parallel to the earth's axis? I'm not sure something as small (relatively) as your flywheel would care where the sun is. But then I've never had a gyro that would spin for a month or so (a month should be long enough to see yearly effects). I may have to find a flywheel and a motor... |
|
| |
I've just edited the explanation of the last illustration removing reference to tilt according to the season. The precessing would be over the course of the day, not the year. See 'Foucault's Flywheel' for discussion about flywheels and precession. |
|
| |
And, now that I think about it, I agree that spinning the flywheel parallel to the Earth's axis would remove the precession, so my illustrated design is unnecessarily complicated. |
|
| |
Now that I think about it a bit more, I'm not sure whether spinning the flywheel parallel to the Earth's axis would remove the precession. If the flywheel was a perfect gyroscope, then this would be a perfect solution. However, the flywheel is not perfect because its angular momentum will fluctuate. |
|
| |
When the flywheel slows down the influence of gravity will force it towards the horizontal - out of alignment with the Earth's axis. |
|
| |
//the magnets will cause an eddy current to form in the ring which will cause a magnetic field to repel the magnets by Lenz's law// |
|
| |
...And unless your ring is a superconductor, those eddy currents will produce heat in the ring, rapidly draining away the stored energy. |
|
| |
//...And unless your ring is a superconductor, those eddy currents will produce heat in the ring, rapidly draining away the stored energy.// |
|
| |
Yes, but the question is how fast? If the eddy currents create 100W of heat then that is quite managable but 10kW makes this impractical. |
|
| |
I'm also wondering whether a non-conductive ring with a diamagnetic material coating (e.g. bismuth) could be used instead of a metal ring. |
|
| |
Why does it have to be ring shaped? Wouldn't something more like a spoked wheel be better? By distributing the bearing over a large area, you're creating more losses. I would think 1 point in the center would be easier to maintain, and would not be high loss for it to be traditional ball bearing to support the weight with perhaps non-touch magnetic motor at the top just to keep it in balance, prevent oscillating.. in your application of large diameter low rpms, the bearing losses are minimized. |
|
| |
Also the tensile strength (limiting factor) of aluminum blows. Carbon fiber would be way better. |
|
| |
Don't forget to pump out the air to reduce surface friction. |
|
| |
//Why does it have to be ring shaped? Wouldn't something more like a spoked wheel be better?//
To maximise rotational inertia (for a given mass) the mass is concentrated at the rim.
Spreading the load (rather than a centre bearing) gives lower bearing forces -> less friction.
Spokes have high drag (track racing bicycles use disc wheels or a minimal number of very aerodynamic spokes).
So, in a nutshell, a ring is the optimum shape. Traditionally difficult to get the energy out, but with the given electrical system, it is easy.
I agree with the carbon fibre, however. The rpm record is with a thin disc of graphene. |
|
| |
Another thought on this idea: the trench would be 95m in circumference, so would require a huge number of magnets which would be costly. |
|
| |
But it may be possible to use metal plates instead of permanent magnets. |
|
| |
This would require the ring to be initially spun *and* have current 'injected' into it so the ring has an initial magnetic field. As the ring passes the metal plates lining the trench, it creates eddy currents in the metal plates and thus magnetic fields. Essentially the magnetic fields in the ring and the plates become self-sustaining. |
|
| |
Using metal plates instead of permanant magnets may be preferrable because the magnet field produced by the plates would perfectly match up with the magnetic field of the ring perhaps making the levitation more stable. |
|
| |
The disadvantages of using metal plates would be a doubling of losses due to electrical resistance (now in both the metal ring and metal plates). |
|
| |
//huge number of magnets// ... 3 ? |
|
| |
Why has nobody heard of air bearings? |
|
| |
Alright, one final variation on the idea. |
|
| |
The main problem with the idea is the losses due to electrical resistance. Electrical resistance can be minimized if the induced eddy currents in each piece of metal all flow in the same direction. So instead of lining the trench with permanent magnets or metal plates, I think metal rings would be best. Using stationary metal rings allows the eddy currents to form into nice big loops rather than lots of little conflicting eddys as would be the case for metal plates. |
|
| |
In the illustration the direction of the current flow in the stationary metal rings and in the rotating metal ring is represented by the crosses (current going into the screen) and the circles with dots (current coming out of the screen). |
|
| |
The stationary metal rings can also be used both as the means of levitation as well as used to input and output energy to and from the flywheel. Input to and output from the rotating ring is achieved by connecting the rings to either a current source or electrical load respectively. |
|
| |
I like this idea: for retaining a vacuum you can't get much better than a torus. |
|
| |
I finally got around to doing some rough calculations. |
|
| |
If the aluminium ring levitates 1mm from the metal lining the ring and lining will each dissipate 350W in heat. If the ring levitates at a more comfortable 1cm from the metal lining, the ring and lining will each dissipate 3.5kW in heat. |
|
| |
It is interesting to note that the heat dissipated is independent of the speed of the ring. So the more energy the ring stores, the more efficient the flywheel becomes (i.e. the heat dissipation stays the same, but the ratio of stored energy to heat dissipated goes up). So I think this idea might be practical when storing energy in the MWh scale (for example remote communities). |
|
| |
Can't you put it on it's side relative to gravity and use
a permanent magnet to hold it up? |
|
| |
//Can't you put it on it's side relative to gravity and use a permanent magnet to hold it up?// |
|
| |
Probably, but its not going to help with precession. |
|
| |
I just did another rough calculation. Looks like 200rpm is about as fast as it could spin before the aluminium ring would start to stretch and break. So this idea isn't very practical; at least with the dimensions and materials I had in mind. |
|
| |
//not going to help with precession// Making sure the plane of rotation is the same as the plane of rotation of the Earth will. This will result, in almost all places, in a ring that is at an angle, neither flat nor perpendicular to the ground. |
|
| |
//Making sure the plane of rotation is the same as the plane of rotation of the Earth will [overcome problems with precession]// |
|
| |
It just occurred to me that there are *two* types of precession to consider. |
|
| |
The first type of precession is happens when the Earth's rotational axis and the flywheel's axis are not parallel. |
|
| |
The second type of precession is when the flywheel's axis is not parallel with the gravitational field. |
|
| |
Having the flywheel's axis parallel with the Earth's rotational axis overcomes the first type of precession but not the second. Having the flywheel's axis parallel with the gravitational field overcomes the second type of precession but not the first. |
|
| |
It might be possible to arrange the flywheel so the first and second types of precession cancel each other out. |
|
| |
Going back to the practicality of the idea, I think the rotational energy could be increased without affecting the resistive losses or the tensile strength of the ring by simply making it heavier (i.e. making the ring wider or thicker while keeping the radius the same), but I'll have to think about that some more. |
|
| |
//The second type of precession is when the flywheel's axis is not parallel with the gravitational field. // ... eh ? A ring at 45N will be predictably at a 45deg angle to the ground. While its supports have to be perpendicular to the ground, I don't see why its power take-on/offs can't be anywhere on the ring and therefore balanced as needed... if that's what you were getting at. |
|
| |
//design is unnecessarily complicated// |
|
| |
That's never stopped anyone else on here!
I remember reading a book written by an engineer who thought why not sit down and try and work out how a ufo might work. Yes, he thought flywheels, but then again no material strong enough for really, really fast spin..considered the principle again and what is a flywheel..particles moving in a circle, so just use electrons in a toroidal magnetic field. |
|
| |
On second thoughts, you are right FT. The second type of precession I mentioned only occurs when there is distance between the centre of rotation and the support of the flywheel (like in a spinning top) which is not the case in my flywheel or flywheels in general. So a flywheel with axis parallel to the Earth's axis will not precess. |
|
| |
If you are really concerned about off-axis gravitational torque, the obvious solution is to make the ring's circumference equal to that of a line of latitude at your location. |
|
| |
... preferably spun at orbital velocity. |
|
| |
//the obvious solution is to make the ring's circumference equal to that of a line of latitude at your location// |
|
| |
//... preferably spun at orbital velocity.// |
|
| |
A ring around the equator traveling at orbital velocity would be ideal (although completely impractical on Earth). |
|
| |
This way you wouldn't need to worry about the forces to keep it levitating, just a small amount of force to keep it confined in its tunnel. |
|
| |
If you had a wire of 1cm diameter traveling at orbital velocity around the equator you'd have 100MWh. |
|
| |