Can you build an object that floats above another object using permanent magnets?

[Towlie]

A while back I was playing with a collection of small, ordinary magnets and marveling at how they attract and repel each other, and I began to wonder about something: With those magnets and some ordinary building materials, such as glue, wooden sticks, and so on, would it be possible to build two objects, such that one of the objects would float in a stable position above the other object without touching anything?

One example that I thought of but never tried is to completely cover the outside of a ball with magnets, all with their north poles facing outward, and line the inside of a bowl with more magnets, all with their north poles facing inward. If the ball was then dropped into the bowl, it seems to me that it should eventually come to rest in the center of the bowl, suspended in mid-air.

Would that work? Is there some other configuration that would work? If not, why not? If so, why do you never see anything like that?

[ben m]

Originally Posted by Towlie

A while back I was playing with a collection of small, ordinary magnets and marveling at how they attract and repel each other, and I began to wonder about something: With those magnets and some ordinary building materials, such as glue, wooden sticks, and so on, would it be possible to build two objects, such that one of the objects would float in a stable position above the other object without touching anything?

One example that I thought of but never tried is to completely cover the outside of a ball with magnets, all with their north poles facing outward, and line the inside of a bowl with more magnets, all with their north poles facing inward. If the ball was then dropped into the bowl, it seems to me that it should eventually come to rest in the center of the bowl, suspended in mid-air.

Would that work? Is there some other configuration that would work? If not, why not? If so, why do you never see anything like that?

1) There is no possible static solution using any combination of permanent magnets, ferromagnets, electrostatic repulsion/attraction, and gravity. This is subject to Earnshaw's Theorem.

2) It is possible to do stable levitation if part of the system is diamagnetic. This is commonly done with bismuth.

3) It is possible to build a stable system in continuous motion, like a Levitron.

4) It is possible to use active feedback to stabilize an Earnshaw-constrained system. (Sense where the levitation object is moving, adjust magnets to move it back, repeat ad infinitum.)

[Zeuzzz]

Yes, the 'levitron' is one example. I've got one.

It has to be spinning to work. I dont think that stationary ones are possible unless you get superconductivity and more complex things involved than normal permanent magnets.

http://en.wikipedia.org/wiki/Levitron

YouTube Video This video is not hosted by the ISF. The ISF can not be held responsible for the suitability or legality of this material. By clicking the link below you agree to view content from an external website.

I AGREE

[shadron]

Magnetic levitation^WP

Superconduction is very good at levitation, in some otherwise unrepeatable ways:

YouTube Video This video is not hosted by the ISF. The ISF can not be held responsible for the suitability or legality of this material. By clicking the link below you agree to view content from an external website.

I AGREE

[shadron]

dup

[Towlie]

So what happens when you drop the ball in the bowl?

[CNY_Dave]

Originally Posted by Towlie

So what happens when you drop the ball in the bowl?

I think your specific configuration would fail for more prosaic reasons- you are attempting to build a ball that would in the aggregate behave like a magnetic monopole.

At a guess there would be a net attraction between the ball and the bowl.


Dave

[BTMO]

Originally Posted by CNY_Dave

I think your specific configuration would fail for more prosaic reasons- you are attempting to build a ball that would in the aggregate behave like a magnetic monopole.

At a guess there would be a net attraction between the ball and the bowl.


Dave

I think that it is an experiment that needs doing!

[jasonpatterson]

Originally Posted by Towlie

So what happens when you drop the ball in the bowl?

It would probably behave much like a single pair of magnets do. You'd be able to get it tantalizingly close to balancing, but then it would turn a bit and attact instead.

[arthwollipot]

http://www.thinkgeek.com/geektoys/cubegoodies/61da/
http://www.thinkgeek.com/geektoys/cubegoodies/5bb6/

[BenBurch]

Shop owned by a JREFer;

http://www.spectrum-scientifics.com/...action&key=225
http://www.spectrum-scientifics.com/...action&key=220
http://www.spectrum-scientifics.com/...action&key=218

[Towlie]

Originally Posted by jasonpatterson

It would probably behave much like a single pair of magnets do. You'd be able to get it tantalizingly close to balancing, but then it would turn a bit and attact instead.

Here. This is a screen capture from a video just recently posted to another thread here.



Now imagine that all of those little mirrors are the north poles of magnets. No matter how the ball turns, you're still looking at the north pole of a magnet. And the inside of the bowl is the same. How could the ball and bowl possibly attract, no matter how the ball turns?

[BenBurch]

Hmmm... My analysis of this is that the ball would have a net zero magnetic field.

The field would be canceled.

A special case of the same problem as the net field inside a charged ball.

Of course it would not be precisely zero as the tiling is not uniform, but it would be close.

Very like a horseshoe magnet with a keeper-bar.

[Gagglegnash]

Hi

Sticking a mess of south-poles to a ball so all the norths are pointed out doesn't make it all-north.

What you have is a fairly complex magnetic field with norths and souths leaking out all over the place in magnetic-field-potential-y hills, valleys, bumps and dips, all of different 'sizes' and strengths because it'd be impossible to tune the thing to get a uniform field.

When you drop it into a similarly, but reversed pole-ed, bowel, it'd probably wiggle around a bit... and probably a very BRIEF bit... until it finds a place where there's a large enough north imbalance getting sufficiently close to a large enough south imbalance for it to clamp down and form a NEW complex magnetic field with the ball with the magnets stuck to the magnets inside the bowl.

Please note all the 'probably's.

[lightfire22000]

I think what you're talking about might require a magnetic monopole if Earnshaw's theorem were to be violated. Of course, if one existed there couldn't be an equilibrium condition. Don't quote me on that though.

[Dilb]

Originally Posted by Towlie

Now imagine that all of those little mirrors are the north poles of magnets. No matter how the ball turns, you're still looking at the north pole of a magnet. And the inside of the bowl is the same. How could the ball and bowl possibly attract, no matter how the ball turns?

The trick is that fields need to spread out. A rough analogy is that a charge, like one pole of a magnet, is sort of constantly producing an incompressible fluid. The south poles on the inside can't spread their field out to the inside, so all their filed spread goes outside. It cancels out the field from the north pole, and so the net effect is the sphere isn't magnetic.

Quote:

I think what you're talking about might require a magnetic monopole if Earnshaw's theorem were to be violated. Of course, if one existed there couldn't be an equilibrium condition. Don't quote me on that though.

Earnshaw's theorem still works even with magnetic monopoles. After all, it already works with electric monopoles and gravity monopoles, so a third type doesn't change it.

[Prometheus]

Originally Posted by arthwollipot

http://www.thinkgeek.com/geektoys/cubegoodies/61da/
http://www.thinkgeek.com/geektoys/cubegoodies/5bb6/

Fun with magnets: http://www.unitednuclear.com/magnets.htm

(check out the Magna View Fluid near the bottom of the page)

[SezMe]

That's an interesting site, Prometheus, but this little line bugged me.

Quote:

For those that find pain relief using magnet therapy (for arthritis, sprained joints, etc.),
our magnetic rings would be ideal for you. Very nice looking on both men & women.

[Prometheus]

Originally Posted by SezMe

That's an interesting site, Prometheus, but this little line bugged me.

[/size][/font]

Yeah, that bugged me too. Especially after all the warnings about taking their magnets seriously, the rings seem like a pretty irresponsible gimmick, too. Then again, a site that sells radioactive isotopes and dangerous chemicals to the general public probably isn't the most likely model of responsible behaviour.

[Dave Rogers]

Putting my pedantry hat on, the claim isn't actually incorrect; the magnetic rings would be ideal for those who found pain relief using magnet therapy, if any such existed. Empty sets are still sets.

Dave

[ben m]

A good way to visualize the sphere setup:

Normally we treat bar magnets as though they were little solenoids, and then we mentally use the Biot-Savart law to find the field. However, you can get exactly the same field behavior by imagining that (a) magnetic monopoles exist and (b) a bar magnet is just a little magnetic dipole with a + monopole at the north pole and a - monopole at the south pole.

OK, now build your sphere out of these monopole-based bar magnets. The nice thing is that now you can solve the magnetic field configuration using Gauss's Law (easy) rather than Biot-Savart or field lines (hard). What does Gauss's Law tell us? Well, we've just built two spheres: first, a uniform sphere of +monopoles (all the south poles). Gauss's Law tells us that the field inside of this sphere is zero, and outside looks like a point charge with a 1/R^2 field. Second, a uniform (slightly larger) sphere of -monopoles (all of the north poles). The field inside this sphere is also zero, outside is also -1/R^2, and please note the sign change. The total field is the sum of the fields of both spheres: zero inside, zero outside because the spheres cancel one another, and nonzero only within the magnets.

[jasonpatterson]

Originally Posted by Towlie

Now imagine that all of those little mirrors are the north poles of magnets. No matter how the ball turns, you're still looking at the north pole of a magnet. And the inside of the bowl is the same. How could the ball and bowl possibly attract, no matter how the ball turns?

I understand what you mean, and it is a bit counterintuitive, but every magnet has two poles. Imagine that the magnetic fields are loops that stretch from the north to south ends of the magnets. Putting them closer together, no matter how tightly, will never be able to remove that looping and so eventually the situation described by Gagglenash would occur. Their behavior would approach that of an unmagnetized object overall as you pushed them sufficiently close, but with a little space between them you should be able to see the sort of behavior described here.

ETA: Figures a simultaneous post would explain that better...

[jasonpatterson]

It occurs to me that there is a bit of qualification that might need to be added to this. Doesn't Earnshaw's thm refer to things that are controlled only by E&M and/or gravity? In this example, we are gluing some magnets to a bowl (which is a molecular scale electrostatic attraction of course) to fix them in place. I thought Earnshaw was basically describing what would happen if we were to allow all of the magnets to move freely. I suppose that we could say that the particles that make up the bowl and glue are always in motion...

If not then I don't see why an electrically charged particle couldn't sit stationary in the center of a fixed position sphere with the same polarity. Similarly I have a set of magnetic rings that rest on a stick. They are of various strengths and so they sit at various separations without moving, though again, their movement is constrained by the stick.

[CNY_Dave]

Originally Posted by Towlie

Here. This is a screen capture from a video just recently posted to another thread here.

http://www.smacaw.com/randi/mirrorball.jpg

Now imagine that all of those little mirrors are the north poles of magnets. No matter how the ball turns, you're still looking at the north pole of a magnet. And the inside of the bowl is the same. How could the ball and bowl possibly attract, no matter how the ball turns?

The field lines from the 'inner' poles wouldn't be contained within the ball, they'd still loop around outside to the 'outer' poles.

Dave

[ben m]

Originally Posted by jasonpatterson

It occurs to me that there is a bit of qualification that might need to be added to this. Doesn't Earnshaw's thm refer to things that are controlled only by E&M and/or gravity? In this example, we are gluing some magnets to a bowl (which is a molecular scale electrostatic attraction of course) to fix them in place.

Earnshaw's tells us that no matter what E&M+gravitational fields you set up, there is no way that field can stably support any configuration of point charges, ferro/para/electromagnets, and masses. Your bowl (held together with glue and other non-EM forces) is setting up a field in its interior. Nothing can hover in that field.

[wackyvorlon]

Pyrolytic graphite works well for levitation above permanent magnets.

[BenBurch]

Originally Posted by ben m

A good way to visualize the sphere setup:

Normally we treat bar magnets as though they were little solenoids, and then we mentally use the Biot-Savart law to find the field. However, you can get exactly the same field behavior by imagining that (a) magnetic monopoles exist and (b) a bar magnet is just a little magnetic dipole with a + monopole at the north pole and a - monopole at the south pole.

OK, now build your sphere out of these monopole-based bar magnets. The nice thing is that now you can solve the magnetic field configuration using Gauss's Law (easy) rather than Biot-Savart or field lines (hard). What does Gauss's Law tell us? Well, we've just built two spheres: first, a uniform sphere of +monopoles (all the south poles). Gauss's Law tells us that the field inside of this sphere is zero, and outside looks like a point charge with a 1/R^2 field. Second, a uniform (slightly larger) sphere of -monopoles (all of the north poles). The field inside this sphere is also zero, outside is also -1/R^2, and please note the sign change. The total field is the sum of the fields of both spheres: zero inside, zero outside because the spheres cancel one another, and nonzero only within the magnets.

I concur.

[MattNelson]

It's technically "pseudo-levitation," as you can learn here:
https://www.kjmagnetics.com/blog.asp...docino-motor-1

...but I'm still calling my invention the world's first levitating platform using permanent magnets. Find 3D printing and build details here:
https://pinshape.com/items/64771-3d-...son-april-2020 -- Ebay link at the bottom of the description!

Photo collection here: https://www.flickr.com/photos/29547300@N03/49795971967 -- since the pinshape site lacks a good photo editor.

[GodMark2]

Originally Posted by MattNelson

https://assets.pinshape.com/uploads/...ing-300081.jpg

It's technically "pseudo-levitation," as you can learn here:
https://www.kjmagnetics.com/blog.asp...docino-motor-1

...but I'm still calling my invention the world's first levitating platform using permanent magnets. Find 3D printing and build details here:
https://pinshape.com/items/64771-3d-...son-april-2020 -- Ebay link at the bottom of the description!

Photo collection here: https://www.flickr.com/photos/29547300@N03/49795971967 -- since the pinshape site lacks a good photo editor.

Congratulations, you have reinvented the Magnetic bearing^WP.

The reason it's not considered 'levitation' is that it requires mechanical constraint between the base and platform. In your case, it's the axle in the middle. This axle restrains any translation of the platform, and also restrains two of the three possible rotation axes.

Still looks cool, though.

[Darat]

How thin can you make that axle?

[Trebuchet]

Seems like with a nice steel axle you could make that with no magnets at all.

[MattNelson]

Originally Posted by Darat

How thin can you make that axle?

In one trial I used a thinner beam (4 mm as opposed to 9). The result was the top level leaning/rotating to one side.

[Crossbow]

Originally Posted by Towlie

A while back I was playing with a collection of small, ordinary magnets and marveling at how they attract and repel each other, and I began to wonder about something: With those magnets and some ordinary building materials, such as glue, wooden sticks, and so on, would it be possible to build two objects, such that one of the objects would float in a stable position above the other object without touching anything?

One example that I thought of but never tried is to completely cover the outside of a ball with magnets, all with their north poles facing outward, and line the inside of a bowl with more magnets, all with their north poles facing inward. If the ball was then dropped into the bowl, it seems to me that it should eventually come to rest in the center of the bowl, suspended in mid-air.

Would that work? Is there some other configuration that would work? If not, why not? If so, why do you never see anything like that?

Sure it is possible to do such a thing, because I have done such a thing.

Many years ago when Radio-Shack used to sell cool electronic stuff for hobbyists and tinkers, such as myself, they sold these 'ring magnets'.

They were about 1.5" in diameter, 1/4" thick with a 3/8" hole in the center.

Anyway, I found if they were placed on a pencil so that they repelled, then the pencil could be turned vertically and several magnets could float as based on their repulsion.

Another magnet thing was to tie one end of a piece of string to a heavy object, then tie the other end of the string to a paperclip. Then, one could position a strong magnet just past the paperclip and the paperclip would stay suspended by the attractive forces produced by the magnetic field.

Are these the sort of things that you are referring to?

[theprestige]

Originally Posted by Crossbow

Sure it is possible to do such a thing, because I have done such a thing.

Many years ago when Radio-Shack used to sell cool electronic stuff for hobbyists and tinkers, such as myself, they sold these 'ring magnets'.

They were about 1.5" in diameter, 1/4" thick with a 3/8" hole in the center.

Anyway, I found if they were placed on a pencil so that they repelled, then the pencil could be turned vertically and several magnets could float as based on their repulsion.

Another magnet thing was to tie one end of a piece of string to a heavy object, then tie the other end of the string to a paperclip. Then, one could position a strong magnet just past the paperclip and the paperclip would stay suspended by the attractive forces produced by the magnetic field.

Are these the sort of things that you are referring to?

Towlie hasn't posted here in almost ten years.

And no, these aren't the sort of things he's referring to.

For one thing, the pencil trick works with permanent magnets because the pencil keeps the rings in place.

And the paperclip trick works because it's not a permanent magnet. Towlie's scenario of permanent magnets, freely floating without constraints other than their own magnetic fields, is prohibited by Ernshaw's theorem. What you're describing are other scenarios, not prohibited by Ernshaw, but also not meeting Towlie's requirements.

[Crossbow]

Originally Posted by theprestige

Towlie hasn't posted here in almost ten years.

And no, these aren't the sort of things he's referring to.

For one thing, the pencil trick works with permanent magnets because the pencil keeps the rings in place.

And the paperclip trick works because it's not a permanent magnet. Towlie's scenario of permanent magnets, freely floating without constraints other than their own magnetic fields, is prohibited by Ernshaw's theorem. What you're describing are other scenarios, not prohibited by Ernshaw, but also not meeting Towlie's requirements.

Thanks much!

I see what you are saying and you are quite correct.

[CORed]

Originally Posted by Crossbow

Thanks much!

I see what you are saying and you are quite correct.

I missed (or completely forgot about) this thread in its first incarnation. It looks like my intuitive response, based on having tried to do it as a kid playing with magnets, that levitation without any mechanical constraints (or some kind of feedback loop) is impossible, is correct.

[MattNelson]

Originally Posted by GodMark2

Congratulations, you have reinvented the Magnetic bearing^WP.

The reason it's not considered 'levitation' is that it requires mechanical constraint between the base and platform. In your case, it's the axle in the middle. This axle restrains any translation of the platform, and also restrains two of the three possible rotation axes.

Still looks cool, though.

Thanks.

You said I reinvented the magnetic bearing, but you left out tons of other magnetic levitation inventions if you are using a spinning bearing to compare.

The "axle" descriptor is inaccurate since there is no spin or wheel. I'm calling it a beam.

Here's a setup using smaller magnets for the 2-pieces, with longer beams half as wide:

[Roboramma]

Originally Posted by ben m

A good way to visualize the sphere setup:

Normally we treat bar magnets as though they were little solenoids, and then we mentally use the Biot-Savart law to find the field. However, you can get exactly the same field behavior by imagining that (a) magnetic monopoles exist and (b) a bar magnet is just a little magnetic dipole with a + monopole at the north pole and a - monopole at the south pole.

OK, now build your sphere out of these monopole-based bar magnets. The nice thing is that now you can solve the magnetic field configuration using Gauss's Law (easy) rather than Biot-Savart or field lines (hard). What does Gauss's Law tell us? Well, we've just built two spheres: first, a uniform sphere of +monopoles (all the south poles). Gauss's Law tells us that the field inside of this sphere is zero, and outside looks like a point charge with a 1/R^2 field. Second, a uniform (slightly larger) sphere of -monopoles (all of the north poles). The field inside this sphere is also zero, outside is also -1/R^2, and please note the sign change. The total field is the sum of the fields of both spheres: zero inside, zero outside because the spheres cancel one another, and nonzero only within the magnets.

Quoting a great response to both highlight it, and also note that I really miss ben m's contributions here.

[wardenclyffe]

Not sure if this qualifies, but it's pretty cool:

https://www.youtube.com/watch?v=NtcQ...re=emb_rel_end

There's a lot of unpacking at the beginning, but the fun stuff starts at about 2:20.

Ward

[MattNelson]

Originally Posted by wardenclyffe

Not sure if this qualifies, but it's pretty cool:

https://www.youtube.com/watch?v=NtcQ...re=emb_rel_end

There's a lot of unpacking at the beginning, but the fun stuff starts at about 2:20.

Ward

Very neat invention, but it doesn't qualify since it's an electromagnet.