Magnetic Levitation System

[thecritic]

I am using this circuit to levitate small objects by using an electromagnet.
**broken link removed**

When the Object comes in the path of IR beam, the output from the IR doide will be around 10 volts and it will be around 0.1 Volts when not obstructed.

Everything seems fine, except for 1 thing, the system don't work, its highly unstable. Basically objects just bounce up and down before being thrown away.

I hope someone would be able to help.

 

[#12]

The circuit looks good. I think the problem is in the physical world. Stable levitation can be tricky. The shape of the magnetic field, the aproach path of the object, if the levitation field keeps the object in the IR sensor range.

 

[thecritic]

In some similar woriking circuit around the net, I see Power transistor being used instead of the Mosfet. Since the Mosfet is Either off Or on, but transistor can be inbetween, delivering say Half power, I am guessing Replacing Mosfet with transistor might work, what do you think?

As for the problem with physics, I don't think so. In videos of some levitation I seen, the object is kept stable right from the start. But in my case, it just starts wobbling around and then shoots out, even I carefully keep it in the right place. May be, my electromagnet isn't good?

And one thing; To connect the IR led and IR diode to the circuit from the wooden structure, I have used good length of wire (About 50 cm). Do you think this can cause problem?

 

[shortbus=]

In some similar woriking circuit around the net, I see Power transistor being used instead of the Mosfet. Since the Mosfet is Either off Or on, but transistor can be inbetween, delivering say Half power, I am guessing Replacing Mosfet with transistor might work, what do you think?

As for the problem with physics, I don't think so. In videos of some levitation I seen, the object is kept stable right from the start. But in my case, it just starts wobbling around and then shoots out, even I carefully keep it in the right place. May be, my electromagnet isn't good?

Don't believe every thing you see done on the Internet ! Some/most of the stuff is faked, unless you can verify where its coming from. Like a well know laboratory, school or company.

Don't you need two magnetic fields to make something levitate? Like a maglev train or monorail, they have one pole in the track and the other in the vehicle. The way your doing it is more like a coil gun.

 

[#12]

Easy part first...50cm of wire could cause problems if it is too thin. It depends on how many amps the coil uses what size wire to use.

Second part: The amplifiers are in charge of whether the transistor goes full on, and it looks like they would full on either a MOSFET or a BJT. MOSFET can be in between! but I don't think that's the problem because you said you had .1V to 10v at the infra-red diode. That is enough to make the amplifiers turn any transistor full on.

 

[crutschow]

You have an unstable feedback loop, which is no surprise. Any negative feedback loop can be unstable if not properly compensated. You might try reducing the gain of the circuit and/or adding a capacitor across the 220k resistor on the output op amp.

 

[thecritic]

Thanks for the pointer crutschow. Reducing the Gain worked. But still the object drops off after about 10-15 seconds. I hope, I will be able to improve it.

 

[4pyros]

Reducing the Gain worked. But still the object drops off after about 10-15

that may be your coil. Andy

 

[DerStrom8]

Apologies. My computer posted this twice. See below.
Der Strom

 

[DerStrom8]

I built an electromagnetic levitator a while back and had the same problem. There were a few things I experimented with, and I eventually got the object to hover (very stable) for over a minute. One, I used a TIP3055 transistor to control the current to the coil. Two, I experimented with the location of the IR emitter and detector (their proximity to the coil, etc). Three, I practiced setting the object perfectly into the beam without jerking it when I let go. These three things I tried are what allowed it to remain levitated for over a minute. I'm sure that if I continued experimenting, and had a better coil, it would be possible to keep the object hovering indefinitely.
Just a few ideas that you may want to try.
Good luck!
Der Strom

 

[thecritic]

I improved the levitator by attaching a small powerful magnet to the object to be levitated. The equilibrium position is then set (by adjusting IR sensor/emitter height) near to but little below the magnetic-gravitational neutral point of the magnetic object and the electromagnet core. In this way, the coil needs only tiny amount of power, (in my case less than quarter watts), to maintain the equilibrium position. It don't need to hold and support the object, that is done by the magnet.
**broken link removed**

But still I can't get the object suspended indefinitely. By experimenting with various frequency, various configuration of IR-emitter receiver (such as inclined orientation, too close, too far etc), I could get as much as may be an hour. But, in any case, the object slowly and slowly accumulate lateral motions and which finally turns violent and the object either drop off or go-snap-attach with the electromagnet core.
May be some of the genius out there can find out way to damp out the lateral motions before it turns violent.

 

[shortbus=]

As I said in post#4, you need two magnetic fields. I think you'll find that for what your trying to do you also need to have two IR systems also. An analogy would be a window comparator. You need to control the upper and lower magnetic field separately to keep the object stable in the fields.

Kind of like if the top pulls up too much it shuts down and the bottom field increases, and vie-versa.

This site has some magnetism experiments that might help; **broken link removed**

 

[thecritic]

In post#4, you said we need two electromagnets even for the levitation and you were skeptical about me successing with my plan.

 

[blueroomelectronics]

I don't see the need for two fields, all the designs I've seen use one. Might want some light shields around the optics.

 

[DerStrom8]

I don't see the need for two fields, all the designs I've seen use one. Might want some light shields around the optics.

When I built mine, I drilled two 5mm holes in the upright pieces of wood so that I could place the IR emitter and the IR receiver directly across from each other. When they were inserted into the holes (from the outside), it acted both as a light shield and as a holder for the LED and receiver. It seemed to work fairly well.
Der Strom

 

[shortbus=]

In post#4, you said we need two electromagnets even for the levitation and you were skeptical about me successing with my plan.

If you don't have two fields all you can do is attract. you need two fields to levitate, although its really not levitating. Its just balancing the two fields to keep the object suspended.

Didn't your experiment start to work better when you added the magnet to your bolt head?

As far as electromagnets that is what you were asking about. You can use three ring magnets and a non-magnetic rod to levitate. place the magnets on the rod with the poles N-N-S-S and the center one will float.

A Maglev vehicle uses Two electromagnet field to levitate, one in the track, one in the train.

A super conductor can levitate with one field.

What I was saying about the youtube videos, is that many are fake. They try to make people think that the person with the video is doing something no one else can do.

 

[shortbus=]

Here is a link that tells how to do it; Google Translate

 

[DerStrom8]

If you don't have two fields all you can do is attract. you need two fields to levitate

One field can work fine (though perhaps not necessarily as well) simply because gravity is also acting on the object to be levitated. Two fields, I admit, may keep the object more stable, but a single field can work as well.
Der Strom

 

[shortbus=]

One field can work fine (though perhaps not necessarily as well) simply because gravity is also acting on the object to be levitated. Two fields, I admit, may keep the object more stable, but a single field can work as well.
Der Strom

For a single electromagnet(above object) the two sensors would work better. The top one to shut off the magnet when it gets close, and a bottom one to turn it back on when it gets too far away.

But the desk displays/curios, like the floating globes, have a permanent magnet inside the globe to make them work. Which there again is two fields.

 

[atferrari]

One field is enough.

I don't see the need for two fields, all the designs I've seen use one.

Me too; all designs I've seen, after googling to death use just one field.

You are neglecting something that made apples famous: gravity.