[sdiy] Clock-controlled filters?

Scott Nordlund gsn10 at hotmail.com
Mon Oct 22 22:50:56 CEST 2012

> From: cheater00 at gmail.com
> Date: Fri, 19 Oct 2012 21:11:02 +0200
> To: pete.hartman at gmail.com
> CC: synth-diy at dropmix.xs4all.nl
> Subject: Re: [sdiy] Clock-controlled filters?


> Richard, yeah, I did recollect that they used switched-cap filters but
> I decided to ask because you never know. I don't want to use a
> fixed-freq DAC for the exact reason that it's fixed frequency. Instead
> I would like to output this from a variably-clocked DAC. The idea
> which I'm playing with is to build a sampler which does not use
> digital interpolation. I think it might be possible to build a FM'able
> clock for some sort of uC that can accept variable clock rate. There
> are many arm-based offerings that can do that. For the clock, perhaps
> one of the many pulse-train-output products out there, similar to what
> David was looking at but with higher frequency interval (it'd still
> need to track about 10 octaves). Horowitz-Hill lists some things (some
> have since become obsolete but got replaced with newer ones). I guess
> I'd just output the waveform via an R-2R DAC and buffer and put it
> through the imaging filter. In this case the interpolation is purely
> analog (i.e. the imaging filter). I do believe the original e-mu
> sampler used a variable rate DAC approach, but I wonder how they did
> the imaging filter. In specific I wonder if it was a switched-cap or a
> VCF.

Look at the service manuals for the Kurzweil 250 and Akai S900. They do exactly what you want (variable rate DAC and switched capacitor filter). Anyway, variable rate isn't really variable unless it's clocked from a VCO. If you're not doing something tricky (using analog electronics to generate a cleaner clock signal, which I think no one has done in any musical application), it's equivalent to a fixed rate system that uses the master clock frequency as the sample rate.  The "variable clock" that the DAC sees is going to have the same jitter as an ordinary variable increment phase accumulator. You can get the same result in software by simply oversampling. I was thinking along the same lines as you, but unless you're going to do something that's absolutely overkill (and aside from price and complexity, overkill approaches have their own problems), I think it's not worth it. I'm getting better results in software by constraining frequencies to subharmonics of the sample rate (i.e. periods are always integer numbers of samples). I'll have to write more about that later.

More information about the Synth-diy mailing list