page 4 of 4

Here we see a close-up of the ZIF-socket board, wired up with some new shielded ribbon cable for the DiskOnChip socket.  This time, I heavily reinforced the connections under the most stress due to the stiffness of the cable with generous amounts of silicone to keep them from moving around, and gave the stuff a good 48 hours to fully cure and set before doing anything else to make sure the wires were good and solidly anchored.  This board is clearly pure, 100% unadulterated "hack": ugly-as-sin point-to-point wiring on Radio Shack perfboard, connections labeled with a Sharpie, wiring tacked down with superglue... but what the hell, it works.

Here's the whole thing going back together again.  Once again, I deliberately made the shielded cable going from the JEDEC socket on the PC104 board to the ZIF socket on top of the unit as short as I could, to minimize interference; the cable length is about 6"-7", just long enough to let you get the case open far enough to reach in and unplug it from the PC104 computer stack so you can get the rest of the unit apart.  I also made very sure to ground the shield-braid!

And here is the unit fully reassembled and back in operation.  The DiskOnChip and Vetra sockets are on the left, and the PIC programmer is on the right.  The PIC programmer was hooked up by the simple expedient of cutting out a slot in the top plastic large enough to push the +18V power leads and the PC/104 computer's parallel-port cable through, then using some rubber spacers and epoxy to secure the board to the plastic case.  (Had I been building it from scratch, I probably would have gone with proper mounting hardware, but in this case I wanted to avoid drilling on an already-assembled circuit board.)  That group of nine pushbuttons in the middle is used to test the Vetra chip after programming; they're wired in such a way as to test all of the row/column combinations used by our application.  That odd-looking LCD display is a surplus item from All Electronics (which they no longer carry, alas) which, according to the spec sheet, is supposed to have some kind of resistive touchscreen built into it... but the sheet doesn't really explain how to make it work, and I couldn't really think of any good way to make use of it for this project anyway, so the touchscreen circuitry isn't actually hooked up to anything.

And that's pretty much it for this project.