Using magnets to suspend objects

[jguerra92]

I'm currently an art student and had this idea for a sculpture. The question I have is how would I go about suspending a metallic object in a magnetic field or fields. I understand the idea of it and have seen it used in picture frames and "floating" globes and such, I need to know what kind/what size/how powerful of magnet(s) will objects of various shapes and or masses appear to float.

 

[Drakkith]

Actually, I don't think you can suspend, or levitate, or whatever, an object using only magnetic fields. I think there is an issue with stability. I really don't know for sure, but I remember another thread about suspending something with magnets and the consensus was that it would not work.

 

[wukunlin]

the suspended object would be in unstable equilibrium, an idiot comes along, give it a bump and the whole thing will likely fall apart

 

[jewbinson]

What if the magnet was in the shape of a bowl, for example? Maybe slightly less curved in shape, but yeah.

Wouldn't something like that be stable?

I think the magnetic bowl would have to be glued to the table, or it would have to be heavy so that the magnetic force felt by it from the suspended object does not cause it to move. But I don't see why it's not possible.

If you draw a free force diagram then all that needs to happen is for the force of gravity to balance out the magnetic force.

Is there something I'm missing here?

 

[Cantstandit]

I think you could suspend it in air with electromagnets, very much like that spinning top toy ("inception-like" i saw explanation in Walter Lewin's lectures somewhere) but this would require for you to construct quite sophisticated electronic circuit (if its even possible). I'm quite sure it is not possible with permanent magnets though.

 

[f95toli]

No, it can't be done with ordinary permanent magnets (it is impossible to get it stable, look up Earnshaw's theorem). You would need a rotating magnetic field, which on the other hand shouldn't be impossible to arrange.

 

[Mordred]

I would also think you would need a way to fluctuate the magnetic strength through motion prediction . What I mean by this is any object esting on the table the base of the suspension table is going to have vibrations etc. You can reduce airflow by having the the system in a closed container. Having the rotating magnets can also be used as an offset. However in order to motion predict the suspended item would involve carefully monitoring the suspended objects movements and then sending an offset magnetic field at the correct vector as a correction of position.
The mathematics involved in that are possible but very complex usually involving a process used often in Electronics termed PID.
P= proportional (setpoint)
I = integral (error)
Derivitative (gain) the derivitiave is often described as a kind of look ahead prediction its essentail goal is to prevent the PI combination from overshooting the correct value.

Edit : Forgot to mention in a change in postion along one direction will invariably result in changes in direction of the other direction vectors, To compensate for this you would need a minumum of 3 PID equations. Where the result of one is used to calculate the resultant's of the others. This is referred to as Cascade loop. second order cascade loops are complex enough without adding a 3 or 4 etc, and would give your average electronic engineer nightmares for life.
When researching cascade loops there is lots of misconceptions in how to implement them the biggest misconception is that the inner loop must be smaller than the outer loop by at least half. This is incorrect. Cascade loops can work regardless if the inner loops is smaller, larger or the same size depending on the process model.

To measure the position there is various sensors today that can give you the accuracy needed with the right calculations (ultrasonic distancing sensors is one that would work however you will need 3 of them minimum.)
The return signal is then measured according to the speed of light, however the length of time it takes for the CPU to process that information, The amount of time it takes the information to reach the processor also has to be considered when calculating the PID values. As well as the needed magnetic vectors.

 

[xts]

it is impossible to get it stable, look up Earnshaw's theorem

Are you quite sure? Earnshaw's theorem applies to system made of electric charges (magnetic dipoles) only, but we have a case of mixed magnetic and gravity forces.

Imagine something like that: a horizontal ring (torus), uniformely charged positively, and small ball, also positively charged, levitating above the ring on its axis. That is a stable configuration.

Similarily, you may take round magnet (like in loudspeakers), mounted horizontally S up, and small magnet, S down, attached to a long stiff rod with counterweight on other end. This rod with counterweight prevents the small magnet from flipping upside down.
Such magnet should be levitating on the axis of the ring magnet, on some height above it.

 

[Mordred]

If you decide to carry on with this project send me a PM and I'll direct you to a site where you can get aid in the equations to handle motion prediction and correction. I know an individual on one particular site that would love the mathematics involved here just for the sheer joy of the complexities involved. He is considered one of the top Motion controls specialists in the US and often helps others with complex problems simply for the joy of doing so.

Earnshaw is this case can be ignored there is ways to stabilize the position through motion prediction and PID controls using variable magnets with the above posted process. However it is not easy and extremely complex.

 

[f95toli]

Are you quite sure? Earnshaw's theorem applies to system made of electric charges (magnetic dipoles) only, but we have a case of mixed magnetic and gravity forces.

Yes, I am sure. The magnetic case might even be in the wiki on the theorem.
Where I work we have a nice mechanical model of a spinning field, we use it to explain why you need an oscillating field to trap ions.

Also, the one exception to the theorem is when you have diamagnets, but the only materials diamagnetic "enough" are superconductors (and even then it is very difficult, normal demonstrations of the Meissner effect using high-Tc superconductors relies on flux pinning for stability)

 

[medic_ma]

Hi. I am also trying to suspens an object in the air in an artwork usinf magnets, or electromagenets. I have some basic physics knowledge but don't know if this is a calculation i can do. What are the formulas i would need to calculate the amount of energy needed to suspens a weight of about 1 kg posssibly between 2 horizontal magnets? How would i calculate this?

 

[Inventor 4U2]

Okay, the challenge is on! Let's see what I can do to accomplish, the said "impossible"! Let's see if I can figure out how to engineer, design and perfect a unit using various shaped Rare Earth Magnets, thick copper sheets and rods which will slightly affect the edge of the magnetic Fields enough which may give enough of a slight advantage to allow a steel and/or magnetic object to simply levitate upon it. Maybe, by spinning the single or multiple magnetic fields at the same time with or against each other may cause the effect needed to accomplish this goal,? Has anybody tried these methods yet?

 

[Front Office]

Diamagnetism. Check out http://www.kjmagnetics.com/blog.asp?p=diamagnetic-levitation

 

[Inventor 4U2]

Wow! Its apparent some people have figured out how to make an object levitate but, it apparently requires a magnetic force at each end of the levitated object. Has anybody yet figured out how to make an object levitate using a magnetic field on only one side of the object to be levitated? If nobody has yet figured out how to accomplish that goal than I will attempt to accomplish it, somehow. Then, I would like to produce those units and offer them to the public so that thousands and thousands of people can enjoy observing an object just "floating" upon someone's desk at their home or office. Maybe I can do it without using alien technology from another world! Just kidding but, seriously.

 

[Drakkith]

Instead of trying to figure out the "impossible", why don't you be creative and use already known ways to do this instead? See here: http://en.wikipedia.org/wiki/Magnetic_levitation

I could see some small desktop units that use very little power being able to accomplish what you want by using a rotating field or feedback mechanism or something.

 

[Inventor 4U2]

Thanks for your attention and suggestions of websites to visit, very interesting. But, I would like to buy or invent a magnetize flotation device that sits on the table and causes a small object to float above it one or two inches and which does not need to be spun every so often to keep it floating in the air. So far, I have not seen anything like that on the Internet. I think MAYBE our generation of determined thinkers have a better chance at accomplishing that goal then before. We now have magnets that are 3 to 5 times stronger. We have access to worldwide research reports on magnetic fields and what materials affect it slightly. I believe I have a better chance at designing such a device than I do at designing a self-perpetuating mechanical device, absolutely. That attempt gave me a headache! But, I am not finished with THAT challenge either! I will leave a message on this website with any results I come up with in the coming months. Isn't it really about "Where there's a will there's a way, if it's physically possible" ? Does this make more sense?: "It's not about what's PROBABLE, it's about what's POSSIBLE." No reply necessary.
- Show quoted text -

 

[f95toli]

You can't invent something that is impossible to make...
So no, it is not possible (unless you use a diamagnetic material).

Note that Earnshaw's theorem is essentially a mathematical proof which uses Maxwell's equations as a starting point to show that a stable static configuration is impossible. Hence, unless there is something fundamentally wrong with our understanding of electromagnetics (unlikely), this really IS impossible.

 

[Cantstandit]

Today I stumbled upon a cleverly designed levitating lamp and I immediately thought about this topic. You should watch this video:

http://vimeo.com/28994216

and the related ones. Maybe it will give you some idea how this kind of thing may be done. As you can see, it doesn't use permanent magnets.

 

[Phrak]

Today I stumbled upon a cleverly designed levitating lamp and I immediately thought about this topic. You should watch this video:

http://vimeo.com/28994216

and the related ones. Maybe it will give you some idea how this kind of thing may be done. As you can see, it doesn't use permanent magnets.

Thank you. These guys were driving me crazy with inside-the-box theory.

A microcontroller goes a long way in getting around passive system constraints.

 

[Inventor 4U2]

Here is the message that has just been posted for Inventor 4U2 by "f95toli"
***************
You can't invent something that is impossible to make...
So no, it is not possible (unless you use a diamagnetic material).

Note that Earnshaw's theorem is essentially a mathematical proof which uses Maxwell's equations as a starting point to show that a stable static configuration is impossible. Hence, unless there is something fundamentally wrong with our understanding of electromagnetics (unlikely), this really IS impossible.
***************
Here is my (Inventor 4U2) response to the above reply: The statement " You can't invent something that is impossible to make." is only correct if all options have been considered and tried. I am already thinking about a way which may allow an object to float above 3 magnets by 1 or 2 inches. And, of course it is just a theory at this moment but it seems to me it can be done. And I am not afraid or bothered by sharing it on this website with other thinkers, that is, young inspired thinkers that have not lived long enough to have a defeatist attitude, too often. And here it is: All magnetic fields have a north and south pole. If two S. or N. Poles come together they first repel each other then the magnet on top will immediately flip over and cause one's N. Pole to lock against the other magnet's S. pole. But, if you have at least three magnets rotating in a circular direction against each other underneath the magnet you're trying to levitate or float then, won't the upper magnet not be able to lock polls together with those spinning or rotating magnets below it? And if you think that the floating magnet will just spin as quickly as the lower spinning magnets, you may be correct. So, to solve that problem simply attach at least two small flaps or wings protruding that upper magnet which will create resistance against the air as it tries to rotate and slow it down enough so that it will never rotate at the same RPM as the lower three magnets. This should, per my theory, allow the magnet to levitate one or 2 inches, depending on the magnetic's strength below it. If I am accurate on what I believe will happen, then I just gave away the idea on how to make an object levitate 1 or 2 inches WITHOUT having to use another magnet above the magnet or object you're trying to make levitate or float.