Repulsive Electromagnetic Levitation by 4 coils

The System contains two main parts:

1. a big ring magnet (underneath or around the four coils, you cannot see it on your picture) and a small strong disk-like magnet. the big one is made out of ferrite, just search for ferrite ring magnet. the smaller magnet is a neodym magnet. Only with these components the system is unstable, and if you try to put the neodym magnet in the center of the ring magnet it will always gets attracted in one direction. this is cause of earnshaws theorem.

so therefore you'll need a second part to get the neodym magnet floating:

2. an adjustable magnetic field controlled by a microcontroler. it contains four coils made out of cupper wire, two hall-effect sensors and some sort of controller which will interpret the datas from the hall sensor and will drive the coils.

The way it works is that the two sensors will measure the (two dimensional) position of the neodym magnet. one sensor will measure the position on the x-axis, the other one of the y-axis. now if the magnet for example moves slightly in the positive x-direction the sensor will measure it, and the controller has to drive the coils so that the magnet will get attracted back to the center. for that there are always two coils connected to each other, but with inverted polarity. so the one coil can push the magnet back to the center, and the coil on the other side can pull the magnet to the center. one hall sensor with two coils is a self contained system, the one system is responsible for the x-axis, the other one for the y-axis as said before. in praxis each sensor has to measure the position 1000 of times per second and the controller has to correct the position every time. if you want to control the coils with a controller and not with an analog circuit you have to consider that the controler is much more slower with calculating the right position than the magnet is moving. therefore you have to implement the controler as a PID-controler, which means it drives the coils mentioning the position of the magnet in the past and the calculated position in the future.

the principle is nearly the same as with the attractive levitation, but instead of just correcting the position of the magnet in one direction you now have two directions..

feel free to share your progress with your project, i made this account to answer your question because i'm interested in this topic too

maxi l.

Hi i would also like to build an electromagnetic levatator . Would it be possible to complete this task without hall sensors?

I think you can create a rotating magnetic field by sequencing electromagnet coils - if you spin the levitating object the rotating field will keep it spinning and the gyroscopic effect will keep the object from flipping over.

Gud After noon sir,

                     Sir i am very interested to do this magentic levtion project,In this project i need circuit digram ,refrence arduino code for this project....Coulde plese send this mail id "eswarnaidu143143@gmail.com".

Here is an article that explains the basics:

https://www.kjmagnetics.com/blog.asp?p=electromagnetic-levitation

Good Day, I would just like to find out if you think this method would work as well if the magnet was in a rectangle formation rather than a solid ring magnet?

Yes many of these devices use 4 permanent magnets plus 4 electromagnets.

Sorry to bump this post--Have you made any progress? I recently decided I wanted to build a slightly larger version of this for my high school Senior Project. I'd be happy to collaborate with you to share ideas.

From my current understanding, you would need the 4 electromagnets in the center repelling the levitated object, and a ring of attracting magnets (permanent, not electromagnets) around the outside. Then, in the very center of the base with all of these magnets, you'd have 3 Ratiometric Hall Effect sensors (not normal Hall Effect sensors) to sense the X, Y, and Z position of the levitated object. This is because Ratiometric Hall Effect sensors detect the strength of magnetic fields, and that somehow can be used to figure out the position. (still trying to figure out how that works.)

Then, you have to build some sort of analog circuit that constantly modulates the 4 electromagnets to balance the levitated object. It's like tug of war but with 4 ropes in a + shape. I don't know how to build such a circuit, but I'm doing research on that too. What are your thoughts?

Really sorry to bump this old post--I'm super curious about this too! I understand all of the technicals behind how the feedback loop is able to modulate the coils. However, I am having trouble understanding how to construct an analog circuit to do that for me! I found a circuit here (MAGLEV MAGNETIC LEVITATION: CIRCUIT SCHEMATIC ) that is for a slightly different project--do you think it would work? What are your thoughts?

This type of circuit is probably more suitable for your project:

https://www.electronoobs.com/eng_circuitos_tut33_2.php