Author Options:

can i make a magnetic POV? Answered

i made a small scetch (in under 1minute) of what i mean, its really raw, just to get the idea.

my idea is to make a small box, with a moving POV in it, wich moves only from the left to the right, and back.
this uses a small POV system, 2 elektromagnets, some way to make the leds go in the right direction, some iron/steel, and something to controll the magnets.

i wanted to know, is this possible, and if so, what magnets, controlling device do i need and how do i make the leds go in the right direction. i've considered the tracks from a CD player, but i think theres a cheaper/better way to acieve this...

if i get enough info to make one, il make a full instructable of it :)



Best Answer 8 years ago

Have you thought of using a piston attached to a motor? The speed may not be uniform, but you should be able to account for this in software. And position can be determined by switches at the end of the piston-head's path.

english is not my mother language, and i have no idea what a piston is...

A "piston" is (for example) the thing in a car engine that goes up and down.  More generally, it is a flat plate or cylinder put inside of a tube.  High pressure on one side and low pressure on the other side makes it move within the tube.  Usually the piston has a rod or "stick" attached to it, either to push it in the tube, or to get work done from the pressure pushing the piston.

I like the piston idea a lot.  If it's a stepper motor, determining position will be super easy as you would only need to count steps.  Furthermore, the software for the display would be easy to manage since a piston follows a sinusoidal path; therefore, the refresh rate of the display could be decreased at the crest and trough with quick math.

You want to set the magnets up to repel, right?  My suspicion is that the system is going to equilibrate very quickly, with the display stationary at the midpoint. 

The situation is equivalent to replacing the "mysterious" magnets with two springs.  It's true that if you pull the display off to one side, it'll get pulled back past the midpoint to the other side, and then oscillate back and forth. 

The trouble is, friction on the guides will dissipate some of that kinetic energy, so the display won't reach as far to the second side as it started initially.  That damping will continue on every oscillation until the thing wobbles about at the center, then stops.

It's that damped oscillation, of course, which requires real POV devices like this to either use a driving motor, or a pendulum (which has much less friction, since the contact area is tiny).

It would be much more conceivable and controllable with ELECTROmagnets.

i already thought i misspelled that, in my language it is "elektromagneet" and i thought it was something like that :D

Your language is fine. I was adding the "ELECTRO" to supplement what kelseymh was saying about magnets, to clarify that regular ones won't work as well as electromagnets would.

In the English language, "electro" is the way that root is spelled.

i thought i'd set the two magnets in opposite direction, one positive and one negative, and by the time it gets to one end, i change both polaritys, but i didnt really thought about how i'd do that...

Ah, switching polarities on the magnets synchronized with the motion.  That's getting very close to being an electric motor :-)

What you want is a solenoid. You can find one that will work well for this project in an old door chime. Here's how a door chime works.

The bottom end of a rod is a metal which is attracted to magnets and the top end is not attracted to magnets. The rod goes through a coil of wire (electromagnet), and gravity makes the metal part sit below the coil when the coil is not on. When power is applied to the coil the rod shoots up because the metal at the bottom is pulled to the center of the magnet. The rod hits a bell making the chime. When the power goes off the rod falls back down.

Far too difficult. Magnetism works on an inverse-square law, which means that if the thing is central and you move it half-way to one magnet you only need 1/4 the power there and 4x the power in the other to balance. Go closer and you need a really feeble magnet close and a really pumped-up one distant. The control-circuit would be a nightmare...
Do it with a motor or similar.


what kind of motor? i have  no idea how to make it go from left to right and back then...

A modification would be to add a spring either side, these would counteract the magnetic attraction. However it's only going to work over a short distance. Kelsey's suggestion of a cam is good, so long as you don't want it to move too fast. Do see the RELATED instructables over on the right of this page.


You use a regular motor, but instead of wheels, the motor is connected to a cam (that's the word in English).  The cam has connection off to one side.  The result, with two loosely connected rods, is that you can convert circular to linear motion.  Look up "cam" in Wikipedia for more details.

As said below, magnets aren't the best way to go here.  Leaving the maths out, they go nothing . . . nothing . . . nothing . . . nothing . . . WHACK!
I'd be tempted to use a high-quality servo with a bit of a linkage, but vibration is going to be a major factor with a reciprocating array whichever way you do it.

Don't forget that one vital thing about a POV display is that the electronics need to know where the light bar is at all times so it can show the correct display for that point in time.  It would be very hard to do this with any system where you cannot accurately control the speed profile.

the magnets are't very reliable<br /> <br /> <br /> the closer magnets get to each other, the more fore the pull with it it's 2m away, it will have 4 times less power then if it were at 1m<br />

i never thought of that...
then again, i made this up in 5 minute :D
then a motor should be better, just need something that will make it go from right to left and back instead of making circles...

Are you sure its only a square law ? My memory may be failing, but I thought it was 4<br />

Dipoles fall off like 1/r3, not 1/r2; there's also the cos(theta) effect, but normally you orient them pole-to-pole.

What's a POV in this context ? 

Persistence Of Vision display.  Like those funky clocks that are just a couple of LEDs on a stick, but looks like the digital time is floating in space above your desk.