A number of weeks ago, I got annoyed that one of the three
Wico bat handles I own had apparently gotten the cord badly crimped at the boot. Out of this, not only was the boot broken off, but the joystick would no longer register 'down' due to a broken wire inside the cable... and, well, the rest of the cable at that point could not be trusted.
Being a proud owner of an Atari 7800, with a somewhat dislike for the design of the
Proline controller, I figured it was time to make another controller for it. I've already made one controller, this one a gutted Space Invaders tv game.
I already had some wire I could use from a previous set of projects, but I still had to go to the local Radio Shack (boo!) to get some more parts. Among this was a piece of project perf board, some resistors to match the specs on the
schematics for the 7800 controller pinouts, quick release connectors, and a male 9 pin connector (more on this in a bit). I also grabbed my one remaining untouched Sega Genesis 3-button pad to serve as a donor for the new cable.
My first job was to disassemble the Genesis controller, remove anything else of potential value, and snip the cable off the board. Once it was free, and the wires stripped, I used the male 9 pin to serve as a probe extension while I used my multimeter to verify which color was for which pin. The Genesis controllers don't seem to use the industry standard colors for the inner wires. I then carefully soldered all nine wires onto one side of the perf board, in order.
With that done, I turned my attention to disassembling the Wico. Once I was in, I was surprised to find I wasn't the first to be inside. The Wico has a switch on the base to switch between the stick top and base buttons. Someone prior to me had pulled the disconnects from the buttons, stripped off some insulation in the middle of each wire, twisted the two of them together, and reconnected them as an effort to bypass the switch. However, I had another idea for the switch, so I removed the modification entirely. Removing the switch, which is only friction fit into the base body, I carefully desoldered the entirety of the original wiring installation. In a fair amount of oddity, the switch is actually 'double pole double throw', rather than the 'single pole double throw' one would expect. The switch was made single pole by merely soldering together each 'pair' of conductors. I went ahead and verified that the switch was double throw with the multimeter, and soldered in new wires for the new role of switching the button roles in the 7800 design.
Now I went in to building the circuit. I had the controller cable soldered to the project board to not only provide myself a 'breakout', but also so I wouldn't have to try to shorten the boot on the cable itself. I added in two sets of resistors in series to make up the approximate rating of 520 ohms that the schematic called for, and bridged over to the appropriate second pins for each button. I wired in each side of the 'feed' for the switch to the other end of each bridged circuit to complete the wiring for the two buttons. I also wired in a set of wires for each positive feed for the directions and the buttons (the resistors and other pins for the buttons are actually on the 'ground' side of the selector switch) and crimped on new quick disconnects for those, as well as the two sides of the selector switch.
About this time, I realized I didn't have enough quick disconnects of the right size. I cut some more wire, and, soldering directly to the points on the leaf switches, created a ground loop on the directionals that terminated in a single wire that I then married to the ground 'cluster' on my perf board. I used the multimeter to ensure we had continuity everywhere, as well as to make sure I hadn't shorted together wires on the project board by bridging the gaps with solder, and proceeded to spend several minutes cramming the controller back together, and cursing the tight fit of everything.
I had to open it back up and reclose it when I ran the first test with Xevious, an excellent game for checking a two button stick on the 7800, as I had somehow managed to get the leaf switch for the base button pressed into the closed position the first time. But the second test went off without a hitch. I could fire cannons and bombs seperately, and I could also switch, on the fly, whether I bombed from the button on top of the stick, or on the base. And all four directions were functional again.
All in all, a very successful project, but not one I'll repeat on a good controller quite yet, in part because of the excessively tight fit of the new pieces into the existing controller shell.
