Magnetic Levitation

Arduino-based magnetic levitation from my book 15 Dangerously Mad Projects for the Evil Genius (http://www.dangerouslymad.com)

http://www.youtube.com/watch?v=WiUng4zF9_M

Nice but how is arduino incorporated it this?

Is partially explained here http://www.dangerouslymad.com/projects/chapter-13-levitation-machine and in browsing the code. No circuit diagram.

Nice but how is arduino incorporated it this?

The Arduino controls the power to the electromagnet, position is detected with an IR LED / phototransistor pair.

You can't do simple on / off power control when it crosses the beam, you have to take the velocity of the object being levitated into account, or it rapidly becomes unstable and falls off.

The sketch can be downloaded from the page for that project on the website (http://www.dangerouslymad.com)

Have you tried this with a linear hall effect sensor? That seems to be the way to go as you don't need anything around the sides of the device.

Have you tried this with a linear hall effect sensor?

No, I did read up on that as an option but was more confident of getting the optical version working. It probably isn’t any harder, but I had to start somewhere!

I have plans for a version 2 which could use the hall effect sensor, but would have a much bigger electromagnet! I’d like to be able to suspend a set of ‘planet balls’ I have of various sizes - Jupiter’s about 5 inches in dimeter and would look pretty cool.

Si:

Have you tried this with a linear hall effect sensor?

No, I did read up on that as an option but was more confident of getting the optical version working. It probably isn’t any harder, but I had to start somewhere!

I have plans for a version 2 which could use the hall effect sensor, but would have a much bigger electromagnet! I’d like to be able to suspend a set of ‘planet balls’ I have of various sizes - Jupiter’s about 5 inches in dimeter and would look pretty cool.

Nice. You can set them spinning freely too.

I have no idea how to accomplish it, but it would be great if some kind of motor effect could be induced to impart a small spin force to the spheres, without having to set them spinning by hand.

They do spin for quite a while, when set going, but eventually slow down.

Ideas anyone?

It would certainly make a great installation on the ceiling. The ceiling lamp could be disguised as the sun!

Hmm, apart from having a ball which isn't perfectly smooth and blowing air lightly across one side (due to limited resistance it shouldn't take much to set it going), no. Or even a ball which looks smooth, try blowing air over one side (maybe ducted from a fan so you don't turn purple) and see if it gets it moving.

blowing air lightly across one side

That's possible.

I was hoping for an electrical solution.

What would be most impressive (but quite likely impossible) would be 2 axis control of the levitated item. so you could control spin and tilt it remotely.

Maybe one of those tiny toothbrush motors with an eccentric weight powered inductively from the suspension coil?

Or fine control of the coil to drop the item faster than it raises it and have tiny fins. A bit like if you lie on your back in a swimming pool and propel yourself by doing fast stoke one way, then very slow stroke the other.

Maybe one of those tiny toothbrush motors with an eccentric weight powered inductively from the suspension coil?

It would mess up the balance of the ball though and inductive power for a multiple-axis-spinning object would be very tricky.

I kinda wish I had an electromagnet here to try this now (and yes I know I could make one - possibly)

I thought it will just spin freely if you touch it with a gentle pluck. You can't control this degree of freedom with the coil since the coil is symmetric (cross fingers) all way around except for up and down.

Well it will spin freely but as said, will slow down due to the surface not being 'perfectly smooth' and air currents will also affect it.

If you attach a mirror shard on the ball and use a photo gate, you may be able to find its rotational speed thus characterize the slowing down since I expect the ball will make tens of turns before stopping. I am thinking the resistive factor goes with square of the rotational speed, as a result of quadratic air drag, like when the ball is tossed in the air. Would be nice to check that assumption. I am totally making it up with marginal facts. :cold_sweat:

Well as there will then be a greater mass in one place due to the mirror shard, won't that end just want to face downwards all the time so then it won't spin as freely?

mowcius: Well as there will then be a greater mass in one place due to the mirror shard, won't that end just want to face downwards all the time so then it won't spin as freely?

:cold_sweat: :cold_sweat: :cold_sweat: You know I have my theory hat on, right? :zipper_mouth_face:

Use mylar?

I have no idea how to accomplish it, but it would be great if some kind of motor effect could be induced to impart a small spin force to the spheres, without having to set them spinning by hand.

They do spin for quite a while, when set going, but eventually slow down.

Ideas anyone?

I think you pretty much answered yourself there. One possibility could be to glue two (or more, just keep it balanced) small lightweight magnets to the side (or inside) of it, and have two corresponding electromagnets next to them mounted on the suspension assembly (preferably some distance away so as to not disturb the magnetic suspension too much). Keep the magnet and electromagnets N-S direction tangential to the suspension axis, and then you can make it spin. Hmm, not sure if that came out right, but I hope its understandable. The Arduino could control them as well. I’d suggest keeping the rotational speed low.

Well as there will then be a greater mass in one place due to the mirror shard, won't that end just want to face downwards all the time so then it won't spin as freely?

Not in the case of this one - it has a neodinium magnet in the top - gives a much more impressive gap ;)

Over the weekend, I'll do some experiments and see how long it keeps spinning. Its hard to get just the right amount of spin without it falling off, but from memory, from the fastest you can spin it without knocking it off, it will spin for a few minutes.

I think thin skewed longitude lines and a gentle draft from above (or below) might give the desired effect. Assuming the whole thing is fitted to the ceiling.

Since it's in a pretty significant magnetic field, I'm wondering if a coil, of very thin wire, placed off center would work. The magentic field (which is changing pretty often, switching to maintain the suspension) passing through the coil would induce a current and that current flow produces it's own field. That offcenter, small field should impart a force, which being unequal (off center) would cause the ball to spin.

I'm just shooting in the dark for this one.. but is there a reason it won't work? Seems easy enough to try if you are experimenting anyway..

The only problem I can think of is that the force may be either to small to have any real effect, or too large, and cause it to spin much too fast.

Kind of an air-bearinged induction motor?