[sdiy] GEM / GeneralMusic SX-3 service manual

[rsdio at audiobanshee.com]

On second thought, it might not be MIDI unless you have a pair of pins for Rx. MIDI is a current loop, so it’s not referenced to ground, but reacts to the current flowing between two lines. There are only about two opts-isolators that were commonly used for MIDI reception, so those should be easy to spot on the PCB.

It’s still probably asynchronous serial. But if you really only have a single Rx line then it’s probably TTL level signals. TTL serial would be referenced to ground.

It’s interesting that you found two ground lines, because that means they were worried about noise. The shield is no doubt separated from the power ground so that none of the audio picks up the serial data frequencies. But if there isn’t a current loop, then they weren’t worried about the kind of ground loop you might get without MIDI. Since the foot pedal board wouldn’t have any audio signals, there’s probably no need to worry about a ground loop generating hum.

You’ll probably have to follow a few traces to see what’s actually going on with the receive circuits.

If there are ROMs and you know what the main processor is, then it’s remotely possible to figure out what’s expected by disassembling the firmware. That’s the hard way, though. You’d have to follow the traces to the CPU and make note of which Port pins correspond to the serial input, etc.

Brian


On Jul 15, 2018, at 1:11 PM, rsdio at audiobanshee.com wrote:
> Timing is most likely asynchronous serial, since there’s no obvious clock line. It could very well be MIDI. It doesn’t seem to be too old for that. The GEM S2 started in 1992 (?), so the SX-3 should not be any earlier than that.
> 
> I’d recommend wiring up a MIDI setup to that pedal input, send it various MIDI notes, and see what happens. Before you wire it up, measure the impedance to ground (and power) of the Rx channel, just to make sure it won’t damage your MIDI output if it’s shorted to something (i.e. if it’s not actually a MIDI input).
> 
> What’s the exact DIN plug? How many pins? What’s their angle? (there are some variants of DIN that have the same number of pins, but they’re spaced differently).
> 
> Brian
> 
> On Jul 15, 2018, at 12:02 PM, Rutger Vlek <rutgervlek at gmail.com> wrote:
>> Hi Brian,
>> 
>> I know, beautiful thing, a switch matrix. This works differently though. I've been able to identify the pinout of the plug and it carries two types of grounds, one for shield and one for power. It carries a +5v supply line and provides an Rx data input pin. So the original pedal for which this input was designed must carry some active electronics. I haven't yet nailed how they would gave done timing of the Rx channel. Anyone fancy a guess at what would have been in there? 
>> 
>> Rutger 
>> 
>> Op zo 15 jul. 2018 20:34 schreef <rsdio at audiobanshee.com>:
>>> You only need 11 or 12 pins for a matrix to support a 30-note pedal board. How many lines on the DIN plug?
>>> 
>>> Brian
>>> 
>>> On Jul 14, 2018, at 1:25 AM, Rutger Vlek <rutgervlek at gmail.com> wrote:
>>>> Additionally, another question for GEM conaisseurs: there's a pedalboard input (looks like a DIN-variant plug) that seems intended to be used with bass pedals. Anyone know which type of pedals could connect to this? Or what the control protocol is? I'm suspecting it's either a straight diode-matrix-scan input (but don't think there's enough lines to make up the required notes), or communication line to some kind of multiplexer or shift-register internal to the pedals. I have pedals from another brand here, with straight switch-outputs and I'd love to modify it to work with the Gem.