Ola the RepolaRap will most likely require an encoder; since it uses a radial positioning system, and the belts use friction only to rotate the platforms, slipping is inevitable; it therefore requires some form of position detection. There will also exist advantage to moving towards servo RepRaps if we can figure out an easy and inexpensive way to create encoders.
I've been struggling with SMD components, soldering them onto PCB stripboard to see if the reflective encoder is going to work. The red LED's I got are *BRIGHT*. It hurts to look at them -- I guess that's what the warning for wearing eye protection on the package was all about. They mention in the data sheet that applications include exterior automotive lighting, so I guess I should have expected that -- I was surprised at the intensity from such tiny devices --3.5mm square, 1.2mm tall; Overall, a great deal for US $0.33 each (I bought 10.) I think 1 per encoder ought to about do it... Seriously.. these things are bright.
I also got two photo transistors stuck down too; I decided as a first pass to use a comparator op-amp, which means two photo transistor per optical track -- 4 total for quadrature, and 6 will be used in the encoders I'm designing for a RepolaRap printer (One track will include digitally encoded absolute position markers.) The two transistor per track allows the detectors to autocorrect for lighting variation, ink reflectivity variation, etc, assuming there is any difference in reflectivity at all between the printed and unprinted spaces on the encoder wheel;
a student project design included this technique, and I thought it had potential in my design. The phototransistors were $0.16 each. I bought 50.
With two photo transistors down, the LED burning brightly, a page with a single black square printed on it, I went about seeing if it can detect the edge. Test was successful. Hurray! I can not determine rate or accuracy yet, as my testing equipment consisted of an analog multi-meter hooked up to the comparitor output. So, now I have to decide whether to design a better circuit and learn how to print PCBs, or figure out a way to get the 6 components to line up correctly and wire up correctly on a stripboard; oh, and figure out how to make a poorman oscilliscope so I can get more accurate information about how the real device performs. I've spent much of this evening learning Kicad, and creating the circuit diagram. The file and image of the circuit is attached.
I welcome any comments or suggestions, either on the circuit itself, or thoughts on stripboard prototype vs. just learning to make a PCB and taking the time to create a circuit design..?
