melittophily
Joined: Jun 18, 2015 Posts: 5 Location: us south
|
Posted: Fri Jun 19, 2015 5:56 am Post subject:
Joystick -> CV Inputs modification |
|
|
While I'm still early in my DIY development, I've gotten a burning intention to modify my Yamaha TG-33 to have 2 switchable jacks to stand in for vector/detune joystick. And it actually looks legit and easy. I have seen forum posts of people mentioning it for the Kawai K1 but only in theory for the TG-33.
Here is the joystick assembly from the service manual.
Pretty clear: a +5v supply splits and goes to two 10k pot voltage dividers and outputs to V-X and V-Y respectively, headed for analog input pins on the CPU. I wouldn't normally modify a synth like this with something going direct to the CPU, but after staring at this part of the service manual and the CPU pin layout and the other analog inputs for a while, I couldn't fight it (tho the other analog inputs are just setup for backup battery monitoring, bummer). How fundamentally different could it be from regularly sending some wild voltages into expression pedal inputs?
I realize the safe road here would be to use two 10k optoisolator/vactrols on the stock 5V into V-X and V-Y. But then we have issues with voltage vs. current control, and after having seen and read about some brilliant expression pedal input audio rate FM success over the years on digital FX and synth units that would be impossible with vactrols, and getting sad at my pedals that fart out quickly after the LFO reaches audio rate, I find myself concluding that as long as I keep it under 5V (and it passes through the stock 1k resistor+0.01uf(?)cap-to-ground-on-the way there) and either way stays well below 1MHZ it has to be worth a shot. Right? Right?
This surprisingly isn't a Yamaha proprietary CPU, but two data PDFs still didn't tell me anything about analog input frequency ranges or anything else of use to my eyes, I suppose because this issue is partly software-dependent. It's a Hitachi 16 bit 20mhz CPU with 7 analog inputs and 10-bit A/D conversion. I'm not sure if there's a "half the system clock speed" guideline for this sort of thing as I know it's a complex 74-pin chip and with no clue about the OS/etc I accept that it could crash with modulation as low as 80hz for all we know. But I've jerked the hell out of the joystick more than a few times in drone jams and it responds well, so maybe we're alright.
I'm sure it will be quantized to some extent (maybe even 0-127 natively) but that's okay because in theory it should still run faster than MIDI. I've never heard any stepping from the joystick except with MIDI loop issues. Direct connection would also obviously reduce component count, troubleshooting steps, space issues, etc. over a vactrol solution I guess I'm asking you to talk me out of it if there's a real case for problems here. I don't know anything about the impedance on this circuit so I don't think I can calculate the cutoff.of the R-C filter going into the CPU inputs, but if it's there I'm guessing it should kill any frequencies that would hurt it.
Secondly, I have to figure out how to handle voltage protection. For CV sources I currently only have a Microbrute + Silent Way through AC Encoder and will probably be adding a Werkstatt soon, so it usually stays around around +5.5V tops and I will have some passive attenuation boxes/cables soon as well, but I still want to protect my cherished synth from freak accidents. I assume a diode to block neg voltages would be a good idea (seriously doubt it will just happen to respond musically to one instead of crashing the software), but what about overvoltage? Simple attenuation will reduce the available control range, so should I look to a zener diode from the inputs that will short everything above 5V to ground?
Finally, and maybe this is where I look really dumb, but I'm unsure about the cleanest way to wire up switching on this. The scheme is:
x jack vs x pot to V-X [to CPU]
y jack vs y pot to V-Y [to CPU]
I take it I would break the signal coming from the middle pin of the joystick pot and wire that into one side of the switch on a NC switching 3.5mm jack, with the other side of the switch merged to the tip signal and going back to V-X/Y on the joystick assembly -> CPU.
Now, when thinking about this, it occurred to me that with something plugged in, it would leave the joystick discharging +0-5V to a shared ground and I'm wondering if that might cause problems. Should I use a jack that leaves the switch floating if something is plugged in? Maybe I'm overthinking it.
Any help is appreciated and any progress will be shared for others.
Thanks! |
|