Hi guys, well recently I’ve been looking for an inexspensive optical encoder for our bot. After some searching I found a pretty nice one and wanted to share it with you guys. It is the ep4 from US digital. What is nice is that you can specify the CPR from 100 to 360 free of charge as well as specify some other options. We decided to go with the low 100 resolution since we don’t need it to be very high and to minimize the interrupts to the PIC. Also an important this is that it is cheap at a price of about $20 each! We only need one,but bought two! Be sure to buy a cable with the molex connector unless you have these lying around. If you don’t need the quadrature output and dont need to know the direction of rotation you can just use either singal lines A or B to give you a squarewave pulse train.
Thanks Windell. This is the same encoder that the Andymark super-shifters use (ep4 or e4p 250-250).
They work very well, however, you must set them up correctly.
To install a shaft on them, you have to carefully remove the plastic cover. Then, place a piece of paper over top of the two mounting screws. Finally, push the encoder disk down onto the shaft until it’s snug against the paper, and remove the paper.
If you do not do it this way, you risk pushing the encoder ring onto the mounting screws and a single turn of the encoder will ruin it. If you put it frrther away from the optical sensor, you will get flaky results, as several other teams have noticed with the AndyMark shifters.
Tom, have you verified that one paper thickness (about 0.004") is the correct spacing between the mounting screws and the encoder disk for proper operation?
I’m asking because this advice seems to conflict with the information given on the manufacturer’s web site(http://www.usdigital.com/products/e4p/index.shtml) under the heading “Accessories/Spacer Tool”, which indicates a nominal gap between codewheel and encoder base of 0.070" with +/-0.030" tolerance.
When one of the encoders are purchased they come with an encoder spacer so I think everything will be alright for us. Thanks for bringing my attention to this otherwise I probably would’ve not paid much attention to this!
We gapped ours per the information I referenced above. It they work OK that way, we’ll leave them alone. If not then we’ll try a thinner gap using card stock, as you did.
You know, its kinda funny… we pressed our encoder wheel in too far MULTIPLE times (so the screw has ‘ruined’ it), and it still works! like all 250 counts still happen… its pretty magical.
i guess we have electronics that live past their death? our IR board was the same way.
If you’re going to use open-frame encoders… make sure that you can hold the very tight tolerances and low runouts specified in the encoder datasheets, and that you’ll be using them in a relatively clean environment (or have taken provisions to keep junk out of them).
Just speaking from experience. :]
The 4 series of miniature encoers are quite nice, I agree… but I suggest that you use instead the S4-type sealed and bearing supported encoder. Personally, I think the extra 10 bucks to go from an open-frame encoder to a sealed encoder is worth it as far as peace-of-mind goes.
