Can I just get a quick reality check on a project idea? Is this feasible?

[jaytee]

I've had this idea of creating a monster Casio SK sampler/ synth/modular circuit bent monstrosity for a few years now and I am finally moving it out of the concept stage to begin work on it.

One of the most important features is the addition of a filter/filter bank to the Casios. Graham Meredith created a filter bank for his SK-8 and produced instructions on building one, but after two separate attempts, I can't get the durn thing to work, so I'm here looking at other options.

My first thought was to build a different filter than the one Graham was using; I found some decent designs that run off 9v and should work nicely, but my lack of experience troubleshooting circuits and my failure on the previous builds has me worried.

My second thought was to grab some SSM2044 chips and use them as ready-made filter modules. This is a bit more appealing, but the thought of designing a circuit around an existing chip is also a bit daunting. Still, at least these chips are well-documented, as opposed to the free schematics for filter circuits that are floating around.

My third thought (and what brings me to this forum) was that maybe there are other chips around that can natively filter external audio and hey doesn't the SID chip do that and HEY A SID SYNTH AND AN SK-1 IN THE SAME CASE SOUNDS INCREDIBLE.

So basically, I'm hoping that someone can let me know if what I want to do is impossible before I start buying up SIDs and PCBs and go nuts.

I have two possible concepts in mind, one is truly polyphonic and the other is paraphonic.

The first would be to build an MBSID around 4 SID chips (one filter per voice of the Casio SK). I am thinking I could hook up the envelope CVs inside the SK to the MBSID's analog inputs to modulate filter cutoff per note. How does this sound? Do the analog inputs respond quickly enough for this use?

The other, simpler and cheaper idea, would be a paraphonic SID filter, whereby MIDI signals from the SK would trigger filter envelopes programmed into the MBSID patch. This should definitely be possible, right?

I am possibly not explaining myself as well as I hope, so feel free to ask clarifying questions. Right now this is all sort of floating around in my brain, so I'm just as confused as you are, in all likelihood.

Edit: Just to give an idea where I am... Like I said, I am not great at troubleshooting circuits, though I can follow a signal path through one alright. I built my own sammichSID a couple years ago, so while that was a no-fuss intro to MIDIbox, I'm familiar with the platform and have read the MBSID manual probably a dozen times. I don't foresee much of an issue building a modular MBSID (as opposed to possible problems that might arise building standalone filters), but I'm more concerned with integrating external hardware (Casio SK) into the concept.

Basically, the end result will hopefully be an SK-1, an SK-8 and an MBSID, all housed in the same case, with the ability to freely patch the audio path as seen fit; ie, one could play any of the three individually, one could sample the MBSID into an SK, one could filter one SK through the MBSID to be sampled by the second SK.... Something along those lines.

Edited March 11, 2015 by jaytee

[Imp]

AFAIK, SK-1 and SK-8 have neither envelope CVs, nor MIDI-out.  So you'd need to somehow midify the keys first.

Or build an envelope generator and get the SKs to put out GATE signals.

[jaytee]

The SKs have internal envelope CVs (already built and installed a module that modifies these) and MIDI is easy enough to install (Highly Liquid sells boards specifically for this purpose).

ETA: the filters I had planned on installing were designed to respond to the SK's internal CV.

Edited March 12, 2015 by jaytee

[Imp]

Ah, ok.  So It's feasible. 

 

If you still got your sammichSID, you could create a patch with OSCs volume turned down and a sweeped cutoff, then route a SK to it's external input and connect a midi-Keyboard.

By playing simultaneously you can emulate the sound of your planned device and make an educated decision, if this is what you want.

[jaytee]

Yeah, that's a really good idea, actually. No idea why I didn't think of that; the sammichSID would make a perfect prototype and I don't think I'd even have to open up the case to get started!