Contraptor

So if you watch this super cool video it would explain :)
http://www.flickr.com/photos/31697990@N00/4170462649/in/set-72157622690487871/
But there are 4 wires. Ground, VCC, and then your 2 digital read signals. I will post some more info when I create it. For now it was more to see if it could be done. I have been a bit side tracked on some real life "work".

So until then I think the video will explain it. If not let me know.

Riley

Sweet. So would you need to make a strip of magnets along the axes and setup the sensor on the linear motion parts?

Or would this be attached to the motor and monitor the internal rotation?

Basically magnetic strip along the axes

I have another magnetic encoder I ordered to do some stuff with rotation. But this one uses magnetic strips (multipole). Here is an attempt at me creating my own multipole strip very cheaply with really strong mini magents.

Writeup Here

However I have not tested to see how high res of pulses this provides as it was late when I finished. You can see the manufacturer's high res magnetic strip i got from there on this link:
http://blog.synthetos.com/initial-linear-encoder-results/

I will post more when I get some initial results.

Ril3y

Let me bounce an idea off of you.
Could you gut an optical (or laser) mouse for its optical engine, and use that for tracking on the linear motion components? I'm sure it wouldn't be too hard to interface the mouse component with an arduino.
I vaguely remember something I read a few years ago about using optical mouse engines on a roomba or some robot to track its own movement.

The engine would have to work on aluminum though. You'd probably have to put a strip of tape or something on a strategic position for each axis. Probably would work.

So yes we could do this however you should think about what we are trying to do and on what application. If we are trying to ensure that steppers are stepping accurately by relying on the linear optical encoders we run the risk of one major flaw in this design. We are cutting material with a dremmel, router, etc throwing bits, dust and god knows what else all over the place. Including right into the linear optical sensor :)

If we rely on magnetic fields for accuracy then we dont care about dust and debris.
However I have ordered a few linear optical encoders to mess with. For non-cnc applications.. perhaps on additive fabbing. IE makerbot style.

Ril3y

good point. optical wouldn't be as robust