I have been in the habit of going for a walk in the mornings to clear my head and this was no exception. The one exceptional event was it's trash day and I was out very early. I came upon one house with what appeared to be a paper feeder for a copier (or at least most of such a beast.) I walked past it but then went back and took a closer look at the motors (which appear to be pretty beefy) and that it had a bunch of smaller springs (that I can use as tensioners.) I walked it back home and then went on with other things
Later on, I got a chance to test the motors. Actually, I just drilled a hole in a piece of dowel to use as a roller and bolted the motor in place — I had drilled holes in the the motor holder for what appeared to be two "standard" sizes. Well, this one is as big a motor as will fit and I brought it to the big attic fan and gave it a whirl. It generated about 25 volts open at that wind speed (a lot) and output 7.5 volts into 51 ohms: about 0.15 amps or 1.1 watts. I installed a larger spindle and got 8.8 volts (0.17A) into it or 1.5 watts. I found that with just a diode, I could charge the battery at 0.12 amps out. That's not bad, but it could be better. I decided that if I also throw on the solar panel I have (supposedly 1.5 watts) and hook it all up, I might be able to get to a point where the whole system is pretty close to self-sufficiency.
If the draw during the day is 0.2A and the draw at night is 0.4A, I might assume an average of 0.3 amps all day. If the generator can average 0.1 amps (very optimistic) and the solar panel 75mA for half the day (35mA average, also pretty optimistic) that's 0.135A average in, reducing my overall average load to 0.165A: the 7 amp-hour battery would then last about 42 hours! That's way better than the 23 hours I can get at 0.3 amps.
1,898 total views
Ok, so now it's 2 days left. It took a while, but I got the program working using the built-in timers and interrupts. Basically I created a main program polling loop that waits for the "fast" clock (which is programmable through macros in Hertz based in the assembly language program) to run "fast" events (like light animation) and a slow clock derived from that for pushing buttons on the MP3 player. For some reason the internal clock runs at 3x the crystal frequency. Who knew?
I also got to wire up all the connections for the socketed chips: I got the latches in place on the board and wired everything up: I'm going to try using the master synchronous serial port (MSSP) to run the latches. I've got three inputs set up that drive PNP transistors with a common emitter to +5V and the collector tied to the anodes of the respective colors.
1,744 total views