[sdiy] Polymoog resonator question

Donald Tillman don at till.com
Fri Oct 5 04:08:03 CEST 2018


Hey Dave,

The inverter stage in those filters attenuates, reducing the loop gain by a factor of 10, lowering the tuned frequency by a factor of the square root of 10. 

On a more philosophical note, this is because a state variable filter isn't R's and C's; it's two integrators and an inversion in a loop.  It doesn't matter how those parts are implemented, the overall equation is the same.  

In this case, the attenuation in the inverter stage is equivalent to each of the integrators tuned down by sqrt(10).

  -- Don
--
Donald Tillman, Palo Alto, California
http://www.till.com



> On Oct 4, 2018, at 1:00 AM, David G Dixon <dixon at mail.ubc.ca> wrote:
> 
> Hey Team,
> 
> I have been looking at the Polymoog resonator schematic (the one from the
> synthfool site), and I'm a little confused.  It's just three state-variable
> filters, and the RC networks have Rs which range from 2.4k to 12.4k (a 2.4k
> resistor and a 10k pot), and the Cs are 68nF, 12nF, and 2.7nF.
> 
> The stated frequency ranges are 60-300Hz, 300-1500Hz, and 1500-7500Hz.
> However, when I calculate the frequencies one should get from these
> component values (assuming the formula f = 1/(2*pi*RC), I get frequencies
> which are about 3.3 times higher than the targets.  By my calculations, for
> the same resistance ranges, the correct capacitors would be roughly 220nF,
> 47nF, and 8.2nF.
> 
> Is the design wrong, or am I stupid, or am I missing something?  If the
> stated component values are the actual ones used in the Polymoog, then one
> cannot argue with the fact that it sounds fantastic, even if the stated
> frequencies are all wrong.  However, I'd just like to know that I'm not
> smoking crack here.
> 
> Cheers,
> Dave Dixon


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://synth-diy.org/pipermail/synth-diy/attachments/20181004/7e0a9beb/attachment.html>


More information about the Synth-diy mailing list