[sdiy] [OT] Analog synths with 2 pole filters

Richie Burnett rburnett at richieburnett.co.uk
Thu Feb 21 10:17:26 CET 2013

2-pole or 4-pole can both self oscillate if there is enough loop gain.

The root-locus of the moog 4-pole cascade is an X shape. As the feedback is increased one pair of poles heads left towards the 'more heavily damped' region, and the other pair heads right towards the unstable right side of the s-plane. The filter oscillates when the right pair eventually reaches the y-axis.

In the case of the 2-pole SVF the root-locus is a semi-circle centered on the origin. The poles spread out tracing out a circle as resonance is increased. In the same way as the moog filter, self-oscillation occurs when the pair of poles essentially reaches the y-axis.

The interesting thing is that regardless of whether the filter is 2-pole or moog 4-pole cascade, the highly resonant response is dominated by a single pair of poles that approaches the unstable right half-plane and eventually leads to sustained oscillation.


Sent from my Sony Ericsson Xperia ray

Scott Nordlund <gsn10 at hotmail.com> wrote:

>My understanding is that a 2 pole filter technically shouldn't self-oscillate. Compare the root locus of 2 pole and 4 pole systems. 4 pole filters oscillate when feedback is increased to the point that two of the poles enter the right half plane (which indicates an unstable system). This never happens with 2 pole filters. Of course this assumes linearity. There's inevitably going to be clipping if the resonance is increased enough, but I think it's really only the designs that have a fairly severe nonlinearity that are going to oscillate. So this is limited to the "cheap and crude filter designs that sound all distorted and cool".
>Second, some of the filters shouldn't be included in the list. The "cascade" ones are really 4 pole, tapped at the second stage. It's a 2 pole response with no resonance, but it's still a 4 pole filter. An exception to this is the filter in the Korg Poly 61, which is a sort of 2 pole cascade (which I think doesn't oscillate).
>> Date: Thu, 21 Feb 2013 14:58:05 +0800
>> From: andy at cytomic.com
>> To: synth-diy at dropmix.xs4all.nl
>> Subject: [sdiy] [OT] Analog synths with 2 pole filters
>> This is a repost with [OT] in the subject, sorry for not adding that
>> in the first place.
>> Hi Guys,
>> I'm having a discussion with a friend who said that "most analog
>> synths with 2-pole filters won't self-oscillate". I know quite a few
>> analog synths with 2-pole filters that do self osc, so I thought I'd
>> write up a list to shed more light on the situation.
>> Can you guys please help me fill in the considerable missing gaps by
>> naming some more and/or correcting what I already have? I know that
>> you can make ladder / cascade / svf / sallen key 2-pole analog filters
>> that self oscillate, this is really just seeing how most production
>> synths have been designed.
>> Analog synths with 2 pole filters that self oscillate
>> --
>> Korg MS-10 / MS-20 / Monotron / Monotribe (sallen key)
>> Steiner Parker Synthacon (sallen key)
>> Oxford Synth Company OSCar (svf)
>> Electronic Dream Plant Wasp (svf)
>> Oberheim Xpander (cascade with summed taps)
>> Formanta Polyvoks (svf)
>> Arturia Minibrute (sallen key)
>> Elektron Analog 4 (svf)
>> Dark Energy v2 (svf)
>> Future Retro XS (svf)
>> ARP Odyssey whiteface mk1 (svf)
>> Oberheim OB-Xa (svf)
>> Oberheim OB-8 (cascade-ish - need to adjust res trimmers)
>> Analog synths with 2 pole filters that don't self oscillate
>> --
>> Oberheim SEM (svf)
>> Roland JP6 (svf)
>> Alesis A6 (svf)
>> Prophet 8 / Prophet 08 (cascade)
>> Yamaha CS 80 / CS 40 (svf)
>> Thanks!
>> Andy
>> --
>> cytomic - sound music software
>> _______________________________________________
>> 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

More information about the Synth-diy mailing list