I'm looking for a way to create a non-rotating persistence of vision device. I have all the electronics set up but I'm stumped with the mechanical design. I tried these two designs:enter image description here
But these didn't work so well. Everything shakes violently and it doesn't go nearly as fast as I need (about 20 swipes per second) Any ideas on how I can build this?

  • $\begingroup$ Take a look at the details of an automobile's motor, especially at the crank and the plunger. en.wikipedia.org/wiki/… the image below the diagram. $\endgroup$
    – ott--
    Commented Jul 1, 2013 at 10:55
  • $\begingroup$ @ott-- this is very similar to the drawing on the left. The problem is that it's very difficult to balance this to run fast. $\endgroup$
    – shoosh
    Commented Jul 1, 2013 at 11:03
  • $\begingroup$ de.wikipedia.org/wiki/Hubkolbenmotor This is a better image (the page is in german, but it's only the image). It's clearly to see why the piston cannot shake around. $\endgroup$
    – ott--
    Commented Jul 1, 2013 at 11:05
  • $\begingroup$ @ott-- I mean the whole thing shakes since the rotation is not balanced $\endgroup$
    – shoosh
    Commented Jul 1, 2013 at 11:19
  • $\begingroup$ You need a more solid base plate of a certain weight. Try 10 kg first. $\endgroup$
    – ott--
    Commented Jul 1, 2013 at 11:24

2 Answers 2


I'm looking for a way to create a non-rotating persistence of vision device.

It seems to me that the two designs that you propose yourself, at least rely on rotation.

I would suggest getting rid of rotation entirely. In particular since you say you have problems swinging these arms around.

One way to go about this, is by using magnets, springs and electro magnets. Actually, instead of me explaining this, go look at a loud speaker unit, and apply that same principle to your challenge. You want it to move further though, so use some soft springs to suspend your moving component between, so you can get some good range on it.

          <fixed electro magnet>               L
<spring>--<free moving magnet>--<electronics>--E--<spring>
          <fixed electro magnet>               D

The middle line can move from side to side, the electromagnet is curled around the sliding assembly

You would likely need to drive the electro magnet in both directions (reversing the poles) to get some good speed too.

The benefit of this, is that the suspended electronics, can be really light, so you dont need some really heavy weights to keep the vibrations down. And the two springs of either end, could carry the power to the moving electronics.

EDIT: I got inspired by rocketnagnet's graphics, so i stole some of it, to clarify just a little bit, just to give a hint as to how my ascii graphics was intended. vertical led strip vibrating horizontally along the axis of the solenoid


I think Zuu's answer was very good, but I'd like to add some more detail to it.

Persistence of Vision oscillator

The idea of using a spring is very good. From the start this creates the oscillating movement you want. I would suggest using a steel leaf spring. This is good for two reasons:

  • It restricts the movement to just left-right.
  • It allows you to add solenoids to drive the movement.

Firstly, set up your spring system. The PCB with LEDs is mounted somehow on the end of the leaf spring. As well as the LEDs, the PCB should also have an accelerometer to measure the movement of the PCB.

Pull back the PCB with LEDs and let it oscillate. Look at the output of the accelerometer to measure the frequency of oscillation. Adjust the length of the spring until the oscillation frequency is about 20 swipes per second.

You're almost there. Now you need to drive the oscillation using the solenoids. Use the accelerometer to synchronise energising the solenoids.


The reason this is an improvement over Zuzu's answer is that, while solenoids are great for short sharp movements, they are terrible for long smooth movements. The travel you get from them will be too short to make an interesting display.

What you need is some mechanism that allows the solenoids to work over a short distance, while the LEDs move a large distance. Hence allowing the stick to bend at the leaf spring. Whether you consider this to be a 'rotating' mechanism depends on whether you consider bending to be a type of rotating.

  • $\begingroup$ If your intention here was to "add more detail to Zuu's answer", then i think my answer was misunderstood. The image you show is very similar to the OPs first image, of a bar that rotates around its bottom end. My answer does not have anything rotating about anything. its basically a piston that vibrates back and forth. $\endgroup$
    – Zuu
    Commented Feb 1, 2014 at 17:10
  • $\begingroup$ I borrowed some of your graphics to give a better hint as to what my idea could look like. $\endgroup$
    – Zuu
    Commented Feb 1, 2014 at 17:39

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.