[sdiy] Pole Mixing

David Moylan dave at expeditionelectronics.com
Tue Apr 13 03:48:14 CEST 2021

Comments inline.

On 4/12/21 9:22 PM, cheater cheater wrote:
> On Sat, Apr 10, 2021 at 8:18 AM David G Dixon via Synth-diy
> <synth-diy at synth-diy.org> wrote:
>> I did more or less the same thing more than 10 years ago, but in Excel.  In
>> that case, the gain factor of the ABCDE are specified as integer values, and
>> the resonance is altered with a slider.  I found this very useful for
>> finding interesting modes, particularly with the various notches (including
>> the famous double-notch mode, which has the gain factors of 1-4-12-16-8).
> Is that different from 2NT they list in the app note? There's no mode
> with 1-4-12-16-8.

They don't have this one in the app note.  I didn't know it was famous 
:)  I came across it as well poking around in my app.  I've included it 
as "Double Notch C (Dixon)".  You can get all sorts of double notches 
playing with the mix factors.  I figured out that you can spec the 
distance between the notches by starting with pairs

(s^2+1)  and (s^2+1) for the numerator.  Then scale the s terms by 
inverse factors.  So, to get the notches 1 octave up/down from center

((s/2)^2 + 1) and ((2s)^2 + 1)  which comes out to

s^4 + 4.25s^2 + 1

I named that one "Double Notch Octaves" but it has non-integer factors.  
The closest integer one is [ 1 4 10 12 6 ] named "Double Notch A (almost 

>> The gain factors translate directly into resistor values for pole mixing.
>> This was the basis of the Intellijel Polaris.
>> I'd be happy to send the spreadsheet to anybody who wants it.
> Please do!

David Moylan
Expedition Electronics
sonic adventures!

More information about the Synth-diy mailing list