I originally intended to dive right in to Phonetroller II PCB design. I have a few important issues to figure out there; namely, how to make the USB power port swappable (uUSB vs Type C) and adjustable, along with how to generally configure the rest of the chassis. However, after using the existing controller prototype for an hour or so with fresh ALPS switches I feel this build strategy may be premature. The new switches are light-years more comfortable than the tiny thumb-denters I originally installed, but they're a bit too linear for my taste. And a quick survey of the vast array of PCB-mount switches I've purchased or salvaged over the years suggests all tactile switches have some issues; the travel is too short, the force requirement is too high, the actuator is too low, etc. Time to re-examine the prevailing gamepad industry button: the rubber dome.
The last prototyping run focused on remove-ability, compactness, and FR4 construction madness. This one will focus on adapting what I feel is an excellent interface platform to my phone in a similar manner to the previous design. I need to get over the hump of poor usability and actually enjoy playing games on my phone with the damn thing. Normally I'd just fire up the soldering iron ( / glue gun / band saw) and go, but in this case I think a bit of advanced planning will make the result much better.
Uhh, yeah, you get the idea. Something like the 'render' above, but I'll probably be a bit more clever with the layering so the phone's top and bottom bar are under the joysticks and the controls are a closer together. I'll likely shoot for a 16:9 aspect ratio and will try to make the whole thing removable, but no guarantees there -- it may be easier to just build it into a cheap off-the-shelf case for the phone and use a wire to plug in the USB-C port.
The plan is to carefully hot-air off all the components, cut the main board in half, wire up the switch contacts and potentiometers (6! crazy!), cut the case in half, and kludge everything together so it's durable enough to survive in my bag for a month. After a closer look at the controller's design, I think this is a reasonable path despite being fairly destructive.
The PCB isn't FR4; it seems like some cheaper composite. With any luck it will survive the hot air treatment. The wireless SoC looks pretty proprietary; even the SOIC-8 to the left of the main processor is laser-etched 'XBOX'. My guess is that the latter chip is an EEPROM and the former is a slightly customized off-the-shelf microcontroller with an onboard radio, but I'm not interested in playing around with either as even in the event of success that would make this a mostly firmware exercise (far from my strong suit). And besides, I need to cut a section out of the PCB to keep the joysticks as close to the screen as possible. And even without a datasheet I can tell that the microcontroller wouldn't survive a trip through the band saw.
Electronics should be fun. Ripping parts off the boards means I can probably pick up the momentary and resistive contacts wherever is convenient, and the traces are all generous enough that re-using the board won't be nearly as difficult as I've encountered in the past. Stay tuned...