A moderately sensitive phototransistor should do the job. Of course you will have to bias it close to its threshold and amplify the result. It's also a good idea to have a pretty strong light source. Most phototransistors respond best to infrared (750-910 nm is a common range, IIRC). I don't know how you would configure a voice-coil mechanism, or even what sort of mechanism you have in mind. I have some old voice-coils somewhere that weigh quite a bit (2 or 3 kg) and require a large current to drive them. OTOH, the head-stack actuator from a laptop (2-1/2 inch) drive would not be very heavy. If both your photodiode (source) and your phototransistor were attached to "see" through the film I mentioned, you would not have to make a slit, per-se, as there are actual assemblies that you can obtain that have both the transmitter and receiver in the assembly and are intended for use in such "choppers." All you'd have to do in such a case is to devise a circuit that would correctly interpret the signal at the sensor and drive the counter in your MCU. That would tell you where you are.


It would appear that you intend to drive the laser output in both horizontal and vertical planes. This becomes both more complex and more massive. Each voice-coil actuator, e.g. the head stack assembly from a fixed disk, would have to carry a mirror, one to effect the vertical deflection and one to effect the horizontal.



It's a good thing to remember that the maximal step rate for those older floppy disk drives was typically 3 ms per step. Now, with careful velocity profiling, one could get longer excursions to operate with much shorter times per step, but the motors are often not that well specified that one can just proceed to do that.

If you've ever looked at the assembly commonly used in retail stores to scan bar codes, you're aware that it's possible to use a constant-velocity motor spinning a multi-faceted mirror and simply modulate the laser light on and off. Then you'd probably only need one voice-coil and mirror, and only a fast-spinning motor with a multi-faceted mirror to draw in the same way that a CRT draws images using a raster imaging scheme. In fact, you could use two such mirrors, one sweeping in the vertical direction, slowly, and one sweeping in the horizontal, much more quickly. Then all you have to do is to create a circuit that modulates the laser on and off based on the raster image. I haven't thought about this much, but it would appear that it's possible. You just need one color, so that's one bit plane, and if you scan at typical TV rates, you should be able to handle it, provided you have the means to gate your laser on and off at a frequency greater than 10-12 MHz. That would enable you to put pictures or text on the wall if you wish. You could, of course simply limit your display to various sorts of vectors if that's what you want.

RE