Electronic Levitron...How the heck does this thing work?

[Pirate88179]

I haven't read the patent yet but here's what I think.

The black disc contains a big ring magnet polarized on the faces, and a couple smaller magnets around the edge of the ring, but facing sideways.  The base contains a big coil facing up/down, and a few smaller coils around the edges. The big coil is switched in polarity, rather then being just turned off and on as in my levitation system. The opposite switching polarities of the big coil act on both poles of the ring magnet to "lock" it into position, whereas in the usual system like mine, the levitated object is not locked but rather balanced between gravity pulling it down, and magnetism pulling it up. It's always falling slightly when the coil is off and rising slightly when the coil is on. But in the Levitron, gravity doesn't play a role, the thing is first attracted-repelled by one polarity of the coil acting against the faces of the ring magnet and then attracted-repelled oppositely by the other polarity of the coil. So it's "locked" in position. The rotation and perhaps lateral stabilization come from the side-facing magnets in the black disc, the rotation coming from being slowly attracted or repelled by the smaller coils around the outside of the big coil in the base, in ordinary "pulse motor" fashion. This locking in position, regardless of direction of gravity, can be seen in my demonstration of what I call the "Nikolayev trailer hitch", where a ring magnet and a cylinder magnet are kept in that relationship, and the cylinder is prevented from flipping over by being inside a central tube. In the Levitron the black levitating disc doesn't flip over because the polarity of the coil is flipping over instead.  The spacing is probably maintained by photodetectors and UV or IR LEDs, and the visible LEDs are there to tell the operator when the disk is in the correct position for the sensing system to take over. 
That's my guess.

Now I suppose I should read the patent.

(There is another really cool levitation system that works by having most of the weight of the levitated object taken up by a strong Permanent Magnet in the top of the assembly, and the electromagnet coil is wrapped around the PM. The strength of the overall field of the PM/coil is regulated in the usual way by sensing shadow  edge or by Hall sensor, and it takes very little current in the coil to maintain the levitation. If the power goes off, the levitated object doesn't drop like it does in the usual system like mine, but rather is pulled UP to the permanent magnet in the top part of the device. It also will "self start" from this position when the power is turned on, it doesn't need any intervention from the operator to get things in the right position. This system is much more efficient than the usual one since it takes so little current to regulate the field of the PM that is supporting the weight of the object.)


ETA: now I've read the patent. I almost guessed it! It looks like sort of a combination of what I guessed and the other system mentioned above, partially permanent magnets and partially coils modulating the field of the PMs.But I had the base and disc PM configurations reversed, apparently, from what's in the patent.

Good find Bill, thanks!

TK:

That was a very well educated guess on your part.  The patent did not seem to me to be very clear but, the info is there and, I suppose they don't want to make it easy for folks to make on their own.  What is doing the polarity switching?  Is it a mosfet or something?  It seems that it would have to handle a lot of power.  As a matter of fact, one of the things I found was a review from a guy that said his device worked great for 3 days after he bought it...then...it died.  Maybe the switching circuit can not handle the power very well?

Also, I did not know that swapping the polarity of the input to an electromag changed the poles of the magnet.  It makes perfect sense, I just never knew that.  So, it ignores gravity and repels/attracts in a very fast cycle and the permanent mags are there to keep the disk centered.  I wonder if someone made one of these using very powerful neos and, boosted the electromag output if it would hover a lot higher?  I would not even mind a larger base with a very good heat sink and possibly even a cooling fan if needed.

No matter what, I have to say that this is a very clever application of known effects and they deserve a lot of kudos for inventing it.  I love this type of thing.

Bill

PS  Fatbird:  Thanks for the PDF.

[TinselKoala]

Yes, there are few strong permanent magnets and with sensor usage it just balances the object all the time with weak electromagnets.

It does more than "just balance" because it supports the levitated disc even when the base is turned on its side. It actually locks the disc in position, up down and sideways.
It's a really neat use of magnetic fields. I hope Bill buys one and tears it apart! 

[Pirate88179]

It's a really neat use of magnetic fields. I hope Bill buys one and tears it apart!

TK:

Well, I was hoping that you would buy one and tear it apart, ha ha.  Hey, maybe we can start a crowd funding campaign in order to raise the $100 needed to purchase one?  If we accidentally get $200,000, we can move to Morocco.

Bill

[TinselKoala]

TK:

That was a very well educated guess on your part.  The patent did not seem to me to be very clear but, the info is there and, I suppose they don't want to make it easy for folks to make on their own.  What is doing the polarity switching?  Is it a mosfet or something?  It seems that it would have to handle a lot of power.  As a matter of fact, one of the things I found was a review from a guy that said his device worked great for 3 days after he bought it...then...it died.  Maybe the switching circuit can not handle the power very well?

Yes probably it died from heat buildup. The circuits given in the patent aren't complete of course, they just show the critical parts. The triangle symbols are op-amp comparators that will flip output polarity based on what happens at their two inputs. They could be high-power opamps that drive the electromagnets directly, or they could drive  a current amplifier stage, like an H-bridge of bipolars or mosfets, that then drive the coils. It would be fun to examine an actual conventional schematic instead of the patent-type sketches of the circuitry.

Also, I did not know that swapping the polarity of the input to an electromag changed the poles of the magnet.  It makes perfect sense, I just never knew that.  So, it ignores gravity and repels/attracts in a very fast cycle and the permanent mags are there to keep the disk centered.  I wonder if someone made one of these using very powerful neos and, boosted the electromag output if it would hover a lot higher?  I would not even mind a larger base with a very good heat sink and possibly even a cooling fan if needed.

Yes, it's got me thinking too. I imagine that it wouldn't really be too hard to build something similar, now that we have the "hints" in the patent. I might even have some magnets that would be appropriate, like a big ring magnet,  and another thin flat-plate thing that is amazingly strong, originally meant for attaching a tracking device underneath a car body. I dunno if I have the patience to make a functional version but I might do some "proof of concept" experimenting later on.

No matter what, I have to say that this is a very clever application of known effects and they deserve a lot of kudos for inventing it.  I love this type of thing.

Bill

PS  Fatbird:  Thanks for the PDF.

Dittoes. I'd like to see how fast the spinning could be made to go; it's regulated to be slow in the device as we see it, I think, since it's designed to display stuff rather than do crazy fast spinning.

[TinselKoala]

TK:

Well, I was hoping that you would buy one and tear it apart, ha ha.  Hey, maybe we can start a crowd funding campaign in order to raise the $100 needed to purchase one?  If we accidentally get $200,000, we can move to Morocco.

Bill