There are wonderful oddities of synth creation breeds out there in the wild — strange, one-of-a-kind birds with three wings and forked duck-bills and other oddities. They might not all be practical for more than their creator, but like evolutionary anomalies, some adaptation or design feature might well make it into other productions – all the more reason that open schematics and permissive licenses could benefit the larger ecosystem, the rich, muddy wetland marsh of sounds.

Friend and neighbor Marc Resibois points me this week to the Fatduino. It’s pertinent to our discussion of marriages between DIY synths and the open source Arduino prototyping platform. Specs:

  • Paia Fatman analog synth (described by the creator here as “a digitally-controlled analog monophonic synth kit. It has 2 sawtooth oscillators, a resonant filter, ADSR for amplitude and ASR for filter cutoff.”)
  • Arduino MEGA as controller.
  • Synth + step sequencer + arp + software LFO.

Here’s the idea: the Fatman already has a controller that receives MIDI notes, and uses those as digital control of the analog oscillators fine and coarse tuning (and envelope generators). In this project, the Arduino cleverly substitutes for that controller for easier programmability and inputs.

Upshot: you get arpeggiator and step sequencer modes, and, for those wanting to program, the cozy confines of the Arduino environment.

There’s also a new, software-based, routable LFO.

Full details and schematics and code are available on the site. Bless you, garage Scottish engineering, and Nibbler Nibbles:

This wouldn’t be a bad starting point for other projects interfacing digital control with an analog synth for more convenience. And you get some tasty-sounding results, as in the videos.

On the MeeBlip project, we’ve had a number of conversations about interfacing with the Arduino. They come down to this: the Arduino is most useful in musical purposes as an interface for controls, sequencing, and MIDI. Now, you don’t actually need much in the way of interfacing for most projects. Because the MeeBlip is already a digital project, you don’t really even need a “shield” as such – you can just connect serial or MIDI directly into the synth. By contrast, trying to implement a synth on the Arduino’s fixed hardware and firmware isn’t necessarily satisfying.

But I’ll defend the Arduino for simple sequence timing and control input; on these tasks, it’s pretty straightforward for a hobbyist.

The previous example:
SJS-ONE: Open, Arduino-Based Synth, with Crazy Cases and Web Troubleshooting

I’m interested to see if anyone builds on this kind of project.

  • Marc Resibois

    Funny, I meant to ask you “do you regret sometimes NOT having gone the arduino shield way” because, even if it is not complicated, the fact of having to cope with a new environment and not the standard arduino one would IMHO limit the number of fork and hacks. It’s a lot more easy to pickup and modify.

    Back on the ‘controller’ part, what I would like to see now is a uniform ‘controller’ scheme across all the atmega monosynth – something using SPI for example – that would just be a front end to a collection of ATmega ‘voices’ (MeeBlip, SJS-1, Midivox). It’s main purpose would be to be able to stack arrays of those and transform it into configurable polyphonic synth. You could ‘bundle’ several synths together for added wierdness or make them all monophonic… etc. If it’d be accepted globally across all producers of minisynth it could be quite a thing IMHO.

    I am not referring to uniformizing the voice architecture but more something that would allow to identify voices and their paramters from a central point and control them that way.

    • Peter Kirn

      Well, some separate issues here – I’m not opposing the use of an Arduino!

      The shield can be unnecessary as typically you don’t need to use up all the Arduino’s pins. Avoiding the shield means more flexibility with which Arduino you choose and how you’d set up the housing. For instance, I’d rather just make a single wire connection to a nano.

      And essentially what I’m trying to say here is the “controller” approach for Arduino plays to its strengths better than an audio application would. Once you get to the level of sound and timing, because you most likely need to use *every spare cycle*, you wind up not being able to take advantage of what the Arduino firmware and environment offer.
      Piggy-backing a second piece of hardware atop Arduino doesn’t really mean that you’re using Arduino in your project, in other words, though here – yes, it does work as a controller. In fact, this is the nicest application of Arduino I think I’ve seen yet, apart from some separate MIDI device.

      I think I’m making sense. This might be served by a separate discussion elsewhere; I just wanted to make clear in this context what the Arduino’s role and strength is.

    • Marc Resibois

      Yes, I know I was going on a side discussion… what I meant mostly is that in the light of open source hardware, the strength of putting people in their zone of comfort isn’t to be under-looked. Personally, I noticed I end up picking up more easily arduino based projects (even it their purpose is to be a controller) is simply because I know I won’t have any hurdles to learn a new environment when I decide to put my grain of salt in it.
      The whole array thing is something I should talk to you more in detail about 😉

    • Peter Kirn

      Absolutely — now, that said, the only project I saw really sort of successfully build a *new*, music-ready language for projects was Mididuino by wesen, and its development stalled. But that was nice, seeing the development environment and syntax and toolchain working. Note that that would all be a project in itself. But I share some dissatisfaction with, say, AVR Studio…

    • Sasa Rasa

      – something using SPI for example – that would just be a front end to a collection of ATmega ‘voices’ (MeeBlip, SJS-1, Midivox). It’s main purpose would be to be able to stack arrays of those and transform it into configurable polyphonic synth”

      Ambika from Mutable Instruments does exactly that with its own voice cards. Being an open desing, it could be used as a starting point to set a standard like the one you are imagining

    • Marc Resibois

      I guess it would be good to wait when it’ll come out and see how it could be extended to non-mutable products. It totally makes sense.

  • Redbeard, How to Make Beats

    I’m impressed lol