How hard would it be to make a VCO board?

So how hard would it be to make a VCO board as a replacement for the control board? are the limitations:

a) size
b) complexity
c) willingness
d) all of the above

I think i remember Pichinettes saying something about the shruthi not adhering to the 1v/octave standard, which would make it a lot harder… :confused:

No, the 1V/octave is not a problem here – it would be if the goal was to put the VCO on the filter board and drive it from one of the CVs from the existing digital board, which is not what is suggest here.

The problems are:

  • Size. You’ll need a microcontroller (maybe a 328p since you won’t have all those wavetables) to decode the MIDI, a good precision DAC (at least 12 bits), and a bunch of chips for the expo conversion, the VCO core and the waveshapping to derive the other waveforms. Look at examples of board sizes on Google image. It’ll already eat at least half of the control board real estate. And then you might want to add another one. And then you realize that you have not a square inch left to add things like pots, switches, etc.
  • Willingness. Given that it would lead to a much inferior product in terms of synthesis options, with a limited user interface (room left for just one switch and one encoder?), this is not where I want to go.
1 Like

I was going to reply when I saw this last night, but it wasn’t for me to speak for Olivier.
But that confirms what I thought too… You just wouldn’t be able to fit much analogue synth parts/features on a board that size. Whatever you could get on there would result in a pretty feeble synth, which probably wouldn’t even be as good/useful as a Monotron.
Which reminds me… You’d be able to do more with SMT parts (Like the monotron). But DIY’ers generally don’t want to mess with SMT parts (Me included)

that’s just a question of flux, honestly. i just soldered a few smt parts, easier every time.
my attempt to an answer would have been try to control the pot’s resistance on the sidekick via vactrol, there you’d have some kind of voltage control. but i guess that was not the idea.

OT, but it’d be worth reopening the SMT discussion. This week I have done my first two SMT soldering jobs, one for a SD card holder (memory cards talk is in the air :D) the other for… repairing stuff in my colleague’s car key. Was not as hard as I thought, so I’ll probably design, route and order an SMT toy for me to practice more. Not because I want to do SMT kits, but because I’d like some of the new things I’m working on to be manufactured for a larger public.

So… Todd gave a clue… Olivier fuels the clue… And we’re all still clueless! haha
Don’t make me straddle both forums reading through tons more posts trying to seek out more clues dammit! Just spit it out!

Back to SMT… I had proper SMT training some years ago for my job. So i got pretty good at it. But you do need more appropriate tools for it.
I don’t mind working with it now for the occasional repair or modification. But I would not like to hand assemble entire SMT boards… Even small Shruthi sized ones. I even chickened out of soldering the SMT USB chips in my x0xb0xes! Still, I am marginally more sane, and slightly less cross eyed as a result.

i should add: ICs are quite easy to handle, as you can hold them down while applying the first two blobs. but last week i had my first smt capacitor experience. the first two were really a pain until i figured out how to do it. could be i have to learn it again next time :smiley:
i guess it could be quite annoying to hand assemble a pure smt kit.
another question is sending audio through smt parts… there are different opinions about that afaik, just like with tubes and transistors.
i’d be interested in your opinion about that, since even Buchla’s new 200E systems are all smt (again afaik)

I’m certainly not an “audio quality wizard” guy. I make circuits that work, with an increasing amount of care and common sense (still learning…), it’s up to you people to judge quality and have theories about why it sounds or doesn’t sound good :slight_smile: However, according to the “whatever sounds like what was on the radio the day you had your first kiss” principle, I understand why some people think SMT should sound bad.

back to the topic… would an adaptor board that fits the shuthis filter board needs to the usual 1v/oct level be more useful as there are already myriads of vcos in the world?

you mean something like a Buchlidian Voltage Processor?

But there surely could be a situation where there didnt need to be encoders/pots mounted on the board with something like the Shruthi programmer there to cover controls?

You still have oscillators, lfo’s and some kind of digital interfacing needed to make the boards work together (oh, and for midi), and somehow make it all fit onto a board that likely doesn’t have space for that… Plus considering the limited number of programmers there are likely to be, you still couldn’t really justify developing the even lower number of analogue boards to go with it.

If its something you are determined to do, you might be better off with a modular oscillator module and lfo etc in a micro modular case (Doepfer make them) Then wire that to a regular Shruthi’s audio input to do some filtering?

Just my 2 ohms worth!

well whether you have buttons/ encoders or a programmer instead, both imply the presence of a structure that controls all. and if you translate the load of f unctions that the firmware provides into discrete (analog) circuits, it would surely look very different. so to hook to the filter just a vco instead of the dcos, and with the modulations controlling the analog part, is easy to implement, by feeding it in the input.
but i’m not sure about what exactly we’re talkking here actually.

If you want an analogue monosynth, why not just buy an SH-101, or a Waldorf Pulse, or build or buy some analogue modular components. The whole point of the Shruthi-1 is that it’s a hybrid analogue/digital synth, combining the best of both worlds. If you want a 100% analogue machine, buy one, or design one from scratch. That’s my advice, anyway, for what it’s worth.


One problem with the “let’s just stick the programmer” solution is the slow refresh rate (the reason why there are only 4 pots on the Shruthi), though it would be vastly improved since the main MCU wouldn’t have to do any audio synthesis, just generating the control signals. Wouldn’t it be fun to have an analogue synthesizer, but with digital envelopes and LFOs? This doesn’t ring true to me.