[sdiy] Shruthi 4PM (was re something else...)

Tom Wiltshire tom at electricdruid.net
Mon Dec 2 11:27:38 CET 2013

I don't know whether it's "weird in theory", but there are clear differences in practice. I spent a *long* time fiddling with the SSM2164 SVF design (having started with Roman's posted below), trying different types of diodes (silicon, germanium, LEDs, zeners) and different positions for them (across integrating caps, to ground, res path, main path, etc) and experimenting with altering the current through the filter (reduce 30K Rs to 15K, double caps to compensate). Pretty much everything makes some kind of difference. I was striving for a clean sine wave resonance, and an even resonance response across the full audio range. The final circuit doesn't quite manage that, but it gets close enough for me, and adding more components for a minor extra benefit isn't my style - leave a bit of he character in!

David Dixon and Olivier Gillet were working on the same SVF design at the time, and we shared notes. It was very helpful and productive.


On 2 Dec 2013, at 10:00, Tom Bugs <admin at bugbrand.co.uk> wrote:

> Interested in the placing of the zeners -- I'd previously thought you could put them over either of the integrating caps - but then I tried over the LP caps (had always done over the BP caps) and it didn't work (can't remember what happened). I just tried simulating this now (had a very similar 2164 VCF already drawn - just with invert on the res chain to feed the -ve input of the first opamp) and again the zeners on the LP integrator don't seem to work -- simulation seems to go badbad..
> Weird?!
> Ta, Tom
> On 02/12/2013 10:46, Roman Sowa wrote:
>> hi,
>> I'm not talking about simply shunting the cap with 2 reversed 1n4148, but zener diodes, that exhibit their existence in the circuit at higher amplitudes, simply preventing from overload, and adding some distortion due to softer knee.
>> I may record some samples of the filter and put them at the website, but I assure you it sounds nothing like stomp box.
>> http://sowa.synth.net/modular/m_vcf.gif
>> Roman
>> W dniu 2013-11-30 03:45, Andrew Simper pisze:
>>> Hi Roman,
>>> What I was getting at is that if you want an ssm2164 to have a
>>> non-linearity most similar to that of a OTA then the placement of the
>>> diodes is important:
>>> http://cytomic.com/files/dsp/ssm2164-non-linear-low-pass.png
>>> The input signal is a +- 5 V sawtooth. The blue plots have no
>>> non-linearity, the red is placing the diode clipper where an OTA has
>>> the tanh non-linearity, and the purple plots are placing the
>>> non-linearity to drain the cap and keep its level bounded directly.
>>> The left plot shows v1 v2 v3 which are the outputs, the right shows v4
>>> v5 v6 which are the inputs that the 2164. Clipping the high pass
>>> signal (input - low pass) is good because it preserves the bass of
>>> your signal while still adding drive. Placing the diode clippers over
>>> the cap is similar to putting your whole signal through a stomp box
>>> after it has been low pass filtered.
>>> All the best,
>>> Andy
>>> -- 
>>> cytomic - sound music software
>>> On 28 November 2013 16:35, Roman Sowa <modular at go2.pl> wrote:
>>>> I always do that, by placing diodes across one cap in the filter. Keeps it
>>>> from overload and adds some character.
>>>> Anyway, I encourage you to experiment with nonlinear elements here and
>>>> there. Long time ago I modified Yamaha CS15 with one 1N4148 in the feedback
>>>> if I recall correctly, and it went crazy. The customer lowed it.
>>>> Roman
>>>> W dniu 2013-11-28 03:33, Andrew Simper pisze:
>>>>> Since everything is so nicely buffered in the 4PM would it be possible
>>>>> (for those that know the 2164 chip better than me) that if you wanted
>>>>> more dirt you could insert a pair of diodes at the input of each 2164
>>>>> to distort the high pass signal that is present there?
>>>> _______________________________________________
>>>> Synth-diy mailing list
>>>> Synth-diy at dropmix.xs4all.nl
>>>> http://dropmix.xs4all.nl/mailman/listinfo/synth-diy
>> _______________________________________________
>> Synth-diy mailing list
>> Synth-diy at dropmix.xs4all.nl
>> http://dropmix.xs4all.nl/mailman/listinfo/synth-diy
>> -----
>> No virus found in this message.
>> Checked by AVG - www.avg.com
>> Version: 2014.0.4259 / Virus Database: 3629/6883 - Release Date: 12/01/13
> -- 
> BugBrand LTD
> UK company No. 07199808
> VAT No. GB 988 2629 57
> 1 Ninetree Hill
> United Kingdom
> www.bugbrand.co.uk
> _______________________________________________
> Synth-diy mailing list
> Synth-diy at dropmix.xs4all.nl
> http://dropmix.xs4all.nl/mailman/listinfo/synth-diy

More information about the Synth-diy mailing list