Prototype - Fixing design flaws in Boss RC-5 Looper

Hey all,
Just wanted to organize my thoughts and materials in some kind of public space where other people can weigh in if they so choose. I think I can maybe do this solo but if anyone has insights I certainly appreciate them.

I recently purchased a Boss RC-5 looper pedal for my midi controlled FX station. It boosts full midi integration.

After farting around for a couple hours unsuccessfully, I learned that the extensive midi implementation chart does NOT include any way to engage overdub mode through midi. Overdubbing can only be used by tapping the pedal itself, or by plugging in one of Boss’ remote aux switch thingies and assigning overdub to that control.

This absolutely massive oversight could be easily rectified with a couple lines of code and pushed into a firmware update. People are asking for it in forums all over the internet, so I have to assume boss intentionally hamstrung the unit to upsell the RC-10. From the bottom of my heart, screw you, Boss.

I plan on fixing this design “oversight” by making a very small midi device that lives in the battery compartment and shorts the remote aux switch pins to ground whenever a defined midi message is received.

My current plan is to see if the (now empty) 9v battery harness is receiving 9v from the power supply. If so, I can grab 9v off it, regulate it down to 5V with some kind of L7805 regulator, and power an ATTiny85 with it. The plan would then be to connect two of the ATTiny’s pins to some 2N3904 transistors as described in this thread:

The transistors are connected internally to the Aux input pins on the RC-5. I’ll probably need to drill a hole in the enclosure to get wires from the aux jack to the battery compartment.

With the the microcontroller powered, and able to programmatically pulse the aux switch inputs, we now need to get a Midi signal to the attiny85. I don’t believe Midi is well supported on the ATTiny85, but it is technically possible. Sources:

I could easily grab a midi signal from the RC-5’s built in midi through, and pipe it over to the ATTiny in the normal way, through an optical isolator. However, if possible I’d like to reduce the latency introduced by the Midi-through and instead grab a copy of the midi signal at the input. Based on my reading, it seems like it might be possible to passively split the midi input signal and have it drive BOTH optical isolators (on the RC-5 and the parasitic ATTiny device). It my also be possible for me to passively tap the signal AFTER the RC-5’s optical isolator and go straight to the ATTiny, bypassing any optical isolator on the Tiny. I’m still not positive exactly how I want to go about this, and I’ll need to experiment.

Anyway, if anyone actually made it through that trainwreck and has any thoughts, feel free to spam them at me. No obligations though, just wanted to organize my thoughts.

I also like that this device could essentially function as a midi controllable tap tempo mod for pedals that have tap tempo inputs like the JHZ Unicorn V2 to name a random example. Perhaps one could even integrate the Tiny straight into a cable and run it off the 5v signal found in the midi reference voltage, as Mitxela does here: MIDI on the ATtiny - mitxela.com
I considered using that approach, but I am a little worried it would cause noise issues. I have no idea if running the Tiny off the same 9v supply as the RC-5 will introduce noise. I guess time will tell.

EDIT: Bashed this together real quick to get my bearings. Haven’t built and tested yet but I think this might do the trick:
MORE EDIT: That draft layout was very wrong, replacing it with a new one:

3 Likes

Interesting stuff.
I have a boss rc30 and despite it’s having both midi and usb support it is not linked into my studio or patch bay. However, as I’m in a rebuild phase in the studio, I decided to work the looper with it’s expression and other midi data into my permanent rig.
I hadn’t known about the lack of midi loop control but I can see how an implementation, like you are describing could create some odd timing issues.
Could the Attiny or other wee MCU not just listen for a message and then actuate the pedal switch?

Keep us posted

I may have been a little vague, but essentially that it what I plan to do - listen for a certain midi message and then send a pulse to the switch’s pins on the pedal when received.

2 Likes

Farbide’s post has me realizing I can also pulse the main switch, not just the Aux switches. Also noticed a mistake in my layout and some missing components that some schematics suggest using… heres V2:
EDIT REMOVED DEFUNCT DIAGRAM

1 Like

That should be fine, and you may well be able to find 5V somewhere on the board that you can use to power the ATTiny, too, so probably no need for the extra regulator. Just check that your transistor switch provides the same functionality as the hardware switch – you can make a switch that is held low and switches high, or vice versa, depending on whether you use NPN or PNP transistors, and what you do with your MCU logic.

1 Like

After poking around in the pedal a bit, I think I can access the jack pins but the main board is incredibly dense and tiny SMD parts, I think I might have a tough time finding solder points behind the opto isolator. You raise a good point about checking around in solder-able points for a 5v source though…

2 Likes

Built out the stripboard layout and for the life of me I can’t get the optical isolator to output a low voltage even when I manually connect the data and voltage reference lines to 5V, and move the data over to GND. Banging my head against this all day today. I’ve built midi circuits before… wtf am I doing wrong lol.
Can’t even seem to get it to work on a breadboard isolated from everything else. What a head scratcher.

EDIT:
There’s one mystery solved - I damaged not one, but two of my opto-isolators. Putting a fresh one into the breadboard yielded the expected result. I think perhaps the issue is that my power regulator isn’t regulating all the way down to 5volts, and is outputting some 6.5 volts or so, which might have been enough to damage the isolator. Now I gotta figure out how to modify the circuit to fix the regulator problem.

Managed to kludge the operating voltage down to an acceptable range and the optical isolator and microcontroller seem to be working - I can get the controller to recognize when the data input of the opto isolator is manually grounded with an alligator clip, but I can’t seem to get the isolator to drive low from a midi signal. >.<
the battle continues.

This project is now done with really legit code and my first custom PCB. Works like a dream. I am in the process of organizing everything into a nice open source release, I will post a link to my github with all the materials as soon as I can.

2 Likes