Hi-
Check out this AVR site: http://heim.ifi.uio.no/~haakoh/avr/
David

Gary Fixler wrote:

>Let me start off by saying "this is just for fun."
>
>I whittled away the shell of a very tiny keychain laser and soldered
>in leads so it's PIC-ready. I want to build a small box with
>motor-mounted mirrors in 2 directions to create a mechanical scanner
>that draws a grid on the wall by reflecting the laser beam. I figured
>some kind of light-sensor(s) being tripped each pass could trip a PIC
>interrupt for sync. The problems I ran up against were motor speeds. I
>imagine I'd sync the mirror blocks with a timing belt assembly. They
>can't spin at the same speed, or I'd only be able to draw diagonals -
>it would work out as though I had only one mirror, set at a diagonal.
>
>If the mirrors were triangular blocks, such that every rotation was 3
>mirror passes, and I wanted 30fps, then the Y block could spin at
>10rps, or 60rpm. If I wanted 640x480, just as an example, the X motor
>would need to scan horizontally 480 times for each pass in Y, which
>would be 1440x for every Y block revolution. With the Y block @ 60rpm,
>that would be 86,400rpm for the X block :)
>
>I've tried to think of alternate ways of scanning, including tracing a
>diagonal path that visits every pixel once (even if it skips around a
>bit), but the math on that is a bit above me, and compounds the
>software/video signal issue. I've dumbed down the resolution and
>thought of things like interleaving. Maybe each pass in Y (one third
>of Y block revolution) draws a separate 1/3 of the screen? I've also
>pulled way back on the fps. By the time I was done, I had block rpms
>in the low thousands, with something like 80x60 @ 6fps. I've also
>considered more sides to the mirror blocks, but each new side
>increases the radius, or decreases the side length.
>
>I know there are some great minds on this list, so I thought I'd throw
>the idea into the group and see what kinds of ideas came back. I'm not
>totally sure what speeds PICs can handle, or how fast the interrupt
>routine to flash the laser beam could work, or what kind of sensor
>would pick up the sync signal at that kind of speed (phototransistor,
>CdS cell?). Also, it would be nice if I could alter the brightness of
>each dot to create a 'redscale' image. I tested the laser on 3v with a
>2k pot, and I was able to dim the beam all the way to off. I'll be
>able to run some experiments with flash timings to get values for the
>brightness levels. I'm thinking I'd probably draw lines of on/off
>instead of worrying about every individual pixel when it isn't
>necessary.
>
>Thanks for any suggestions or observations
>-gary
>
>--
>http://www.piclist.com#nomail Going offline? Don't AutoReply us!
>email listserv@mitvma.mit.edu with SET PICList DIGEST in the body
>
>
>

--
http://www.piclist.com hint: To leave the PICList
mailto:piclist-unsubscribe-request@mitvma.mit.edu