Magnetic Levitation Sculpture
4 Steps
Now available for sale!

If you like it, but don't have the laser access or don't want to do the casting, follow the links at http://www.cadabralabs.com to get a kit.

The Idea:

In this instructable, we build a diamagnetic levitation sculpture, complete with easy adjustments for optimizing the levitation of the small cube magnet.  The instructions are fairly complex, and need a number of tools, but the end result is very nice!  See http://www.youtube.com/watch?v=W_hgNbEqVlE for a video of it in action!

How does it work?

We've all tried to get one magnet just the perfect height above the other so that the magnetic pull matches gravity, and the bottom magnet floats, but while this may be possible in theory, this would only work in a micrometer range of distances.  Even if you had something more steady than your hand holding it, any slight vibration would move it enough to move it out of that range.  So how do we do it here?

The bismuth we're using above and below the bottom magnet is diamagnetic.  This means it very weakly repels any magnetic force, but the stronger that force, the more it repels.  When you get the lifter magnet far enough away that its just barely far enough away from the cube magnet to lift it, the bismuth pushes it down until the cube is far enough away from the bismuth to be affected by it.  At that point, the bismuth below is close enough to the cube magnet to push it back up slightly. All you need to get it floating is find the right balance of the distance between the two bismuth tubes, and the lifter magnet above!  Both of those are adjustable in this project!

Items required:

About 1 square foot of 1/8" MDF or baltic birch plywood.
Access to a laser cutter capable of cutting through 1/8" MDF or plywood; alternatively you could use a jigsaw if you're a very accurate jigsawer.
1/10th lb of bismuth.  You might find this sold as fishing weights or online.
Pan to melt bismuth.
Casting forms to cast the bismuth into nice round shapes.
Sandpaper to smooth out the casts. (Not necessary but recommended)
1 5/16th" elevator bolt.
1 5/16th" nut.
1 5/16th" locking nut
2 6-32x3/4" bolts
1 6-32x1.5" bolt
4 6-32" nuts
1 strong nib lifting magnet.
1 3/16th" strong nib floating magnet

Remove these ads by Signing Up

## Step 1: Step 1: Cast the bismuth

Making the mold:

The mold you use can be made pretty easily with some tin foil. Simply find something with a hole half an inch across, and carefully push the tin foil into the hole, pressing around the inside of the hole to get it as smooth as possible. You can either leave it in the hole (it will get a bit hot) or gently pull it out. It should hold its shape fairly well.

Melting the bismuth:

Bismuth melts at a pretty low temperature, so you can actually melt it on your kitchen stove. Alternatively, you can use a propane torch, but that's overkill.

Casting:

Once the bismuth is melted, carefully pour it into your mold.  Before the bismuth solidifies, stick the 6/32x1.5" bolt halfway in as straight up as posible, being careful not to burn yourself.  It should only take a minute for it to cool down, but you can run water over it to speed up the cooling.  Once its cooled, carefully unscrew the bolt out, and repeat the process with a second mold.

Smoothing:

At this point your bismuth cast is going to look pretty rough.  Use some rough sandpaper to get the big imperfections removed, and then finer sandpaper to remove the smaller ones. With a bit of work, you can make it look really nice. The most important part is making the top (away from the bolt hole) smooth and flat.

Krayzi99 says: Mar 12, 2012. 11:23 PM
wow, i just had a vision, using the same technique as this to make a globe that you can just pull out of the stand... i should attempt it!
seraine says: Dec 31, 2011. 7:43 PM
Are the bolts inside the bismuth magnetic, and if they are, is it a problem?
jfehr67 (author) says: Jan 1, 2012. 8:46 AM
The bolts I used were brass, but you can use any non-magnetic bolts. Magnetic bolts will cause the cube not to levitate.
pooh1485 says: Nov 16, 2011. 6:07 AM
Great design with laser cutter and magnetics, thats creative thinking
stephenfitton says: Sep 11, 2011. 2:56 AM
Great work,by using bismuth you are expressing the relationship between magnetic force and electric force kindred spirits,this is expressed further by the addition of antimony giving you the Peilter effect, where when d.c. current is put accross the junction one side goes hot, the other goes cold . reverse current and the opposite takes place.
Your devices expresses a practical example of the unexplored potential of what electricity,magnetism,gravity have in common. Top Marks
imuhachev says: Sep 4, 2011. 8:37 AM
it very coool
jfehr67 (author) says: Sep 4, 2011. 11:25 AM
A video of it in action? There's one on the very first page. Here it is again:

imuhachev says: Sep 7, 2011. 4:08 AM
thaks :)
mycroft618 says: Sep 5, 2011. 1:15 PM
OK, If you put it in a vacuum, Could you put small White and black flags at each corner to create a radiometer similar to the Glass Child's toy?
splazem says: Sep 4, 2011. 7:15 AM
Holy cow! That's awesome!
jfehr67 (author) says: Sep 4, 2011. 11:27 AM
Thanks! I'm hoping to either win the epilog laser challenge (vote at the top if you like this project!) or buy one if my kickstarter project is successful. (The link is down a couple of comments if you're interested.)
splazem says: Sep 5, 2011. 6:19 AM
Sure!
muhammad.fahrur says: Sep 5, 2011. 1:43 AM

I hope it's can help us knows more about levitation.
bpark1000 says: Sep 5, 2011. 1:15 AM
There is one other component of drag on the rotating magnet: Eddy current drag. The bismuth is conductive. You could reduce this drag by fabricating the bismuth from an array of tiny wires bundled together, but insulated from each other. (Motors have this drag also. Motor iron parts are laminated to reduce the drag.) You can also use pyrolytic graphite instead of bismuth, but the diamagnetic coefficient of graphite is lower than bismuth, so the balance is even more fragile.

One problem with this setup is the extreme temperature sensitivity. You will need to re-adjust the position of the levitation magnet to maintain balance as the temperature changes even if a few degrees. It is a nice demo, but is not practical for any commercial purpose. One dirty trick to simplify the adjustment is to have 2 levitation magnets. One is fixed in the frame, and is set somewhat too far away to levitate. The other is smaller, and is adjusted. The position of adjustment is less fussy, as you are only changing part of the levitating field.

(I assume you are familiar with the Edmund

http://www.scientificsonline.com/diamagnetic-graphite-levitation-kit.html

and the \$12 EnVision Labs kit

http://www.envisionlabs.com/shop/viewitem.php?groupid=3&productid=18

that levitate graphite over a fixed magnet set. This does not need balancing; the diamagnetic force directly levitates the graphite.)
JonnyDude2008 says: Sep 1, 2011. 4:41 AM
You could try putting it in a vacuum so it spins forever
valveman says: Sep 4, 2011. 1:39 PM
There is no such thing as a perfect vacuum so no it can never spin forever. There is always air particles in a vacuum no matter how much vacuum you apply.
Polymorph says: Sep 4, 2011. 8:27 AM
Actually, eddy currents will prevent it from spinning forever. IE, drag from currents induced in it by the spinning within the magnetic field.
jfehr67 (author) says: Sep 4, 2011. 11:31 AM
I read about that somewhere, but isn't that only if the north/south poles are alternating? In this case they're both maintaining their pole direction.

In either case, air resistance places a much bigger part. Try this with a ball magnet with very little air resistance and it spins much longer...
stringstretcher says: Sep 27, 2011. 12:49 PM
try this: drop a small neodyme magnet thru a copper pipe and watch it slowly spin its way through the pipe. this is a demonstration of the drag induced by the Eddy currents.
Polymorph says: Sep 4, 2011. 8:43 PM
No, all eddy currents require is a conductor moving in a magnetic field. Or the magnetic field moving across the conductor.

I do Mr Science demonstrations for my wife's class from time to time. One of my demos is dropping a very strong magnet down a PVC pipe, then down a copper pipe. The magnet takes a lot longer to reach the bottom in the copper pipe.

This is due to eddy currents induced in the copper, which then create their own magnetic field that repels the field that created the eddy currents in the first place. In fact if you look down the copper pipe as the magnet falls, you can see that it floats down the center because it is being repelled from the pipe walls. That is also how maglev trains work, once they are moving more than a few km/hr.

it is pretty cool. Even those tiny supermagnets from Radio Shack will work, but the pipe needs to be not much larger than the magnet for it to be more obvious. Stick a few of those tiny magnets together to make it stronger, you'll also find it lines up vertically, so even if it is too long to fit sideways it stays away from the sides.

Glenn Burrow says: Sep 4, 2011. 2:38 PM
Actually, eddy currents happen any time a magnetic field passes over anything that conducts electricity. It is the main limitation in electric motors, hence they use "electrical steel". It has lower conducting properties plus they usually laminate them in thin sheets of insulating plastics.

Anyhow - eddy current losses are generally greater than drag. But that depends entirely upon the strength of your vacuum and the vessel the contraption is contained in.
jamwaffles says: Sep 4, 2011. 6:09 AM
Really, really good idea, sculpture and Instructable. Great work!
jfehr67 (author) says: Sep 4, 2011. 11:28 AM
Thanks jamwaffles! Took a lot of prototypes to get it to where it is. :)
flyingpuppy says: Aug 31, 2011. 9:01 PM
That's intimidating. But I could see how it could be relaxing to play with this levitation art.
jfehr67 (author) says: Aug 31, 2011. 9:24 PM
The hardest part (assuming you can find a shop to laser cut the parts for you) is the casting, but really, that can be done on a stovetop, and the bismuth doesn't have to be perfect to work, so you could even just use your finger and wrap tin foil around it to make the mold. For my kickstarter project where I'm hoping to make hundreds of the models, I'm getting a bunch of holes laser cut in 3/8th inch metal and using that as my mold, because sponsors tend to want a very clean look.
jfehr67 (author) says: Aug 30, 2011. 10:53 PM
salcera: This is actually part of an 'Open Hardware' kickstarter project, so yes, if you'd rather just get all the parts and assemble it yourself, you can go their and become a backer. Of course, you'll only be charged, and only get the kit if the project gets completely backed.

Here's the link if you're interested: http://www.kickstarter.com/projects/jfehr/magnetic-levitation-sculpture
makendo says: Aug 31, 2011. 2:58 PM
It's beautiful. I'm in! Hope you get funded.
jfehr67 (author) says: Aug 31, 2011. 9:21 PM
Thanks! I appreciate the support. I'm entered in the Epilog laser challenge here too, so I'm hoping one way or the other I can get a laser so I can make and sell these! :)
salcera says: Aug 30, 2011. 9:11 PM
oops i meant COULD get the parts for this???
salcera says: Aug 30, 2011. 9:11 PM
I dont have anything to cut the pattern with but i think this is really cool

jfehr67 can you tell me where i cold get the parts for this?!?!?!?