[sdiy] Digisound SVF no resonance

Andrew Simper andy at cytomic.com
Thu Sep 1 03:40:27 CEST 2016

Hi Nick,

Ok, thanks for pointing that out, the logic remains the same. It
doesn't matter which SVF you use, you don't need to alter the existing
damping gain stage part of the circuit, the point is to add some
positive feedback around the bandpass stage, this pushes the damping
into effectively negative values otherwise it would be only near zero.
If the CEM 3320 has non-inverting inputs and non-inverting buffers
(which looks to be the case because of the way the SVF is wired) then
it's just a matter of placing a largish resistor from bandpass buffer
output to the bandpass input.

Now doing this isn't ideal since it changes the high pass output to
make it less ideal. Having a quick look again at the equation you are
better off adding some positive feedback of the bandpass signal
directly into the high pass summing op-amp as this will preserve the
high pass shape. So looking at the CEM 3320 SVF that would mean taking
then dc blocked bandpass output at the top of C5 (neg side of cap) and
connecting that to the positive input of the IC1 opamp with a 270k or
so resistor.



On 31 August 2016 at 20:39, Nicholas Keller <niroke at tampabay.rr.com> wrote:
> Andy,
>   Ian stated in his initial post that he has the CEM version, 80-7A, not the SSM version 80-7.  Documents for both are included on the page I linked to.  The 7A uses a CEM3335 for the resonance circuit, but it wasn't clear to me where the feedback source was coming from.  The 3320 used for the filter poles does not have VCQ on-chip.
> Nick
>> On Aug 31, 2016, at 8:30 AM, Andrew Simper <andy at cytomic.com> wrote:
>> The standard trick to get more resonance out of an SVF (thanks to
>> Osamu Hoshuyama for pointing this out clearly in his pdfs describing
>> the ARP4023 and OSCar filters!) is to have a large resistor in a
>> positive feedback loop around the bandpass output back to its input
>> (if the buffers of the 2040 are inverting then you can just add this
>> resistor from the bandpass buffer output to the negating bandpass
>> input, start at around 300k and work you way down to taste). You will
>> also probably need to add the regular parallel but in opposite
>> direction diodes from the bandpass output to the high pass op-amp
>> input, which increase damping to maintain resonance stability.
>> Cheers,
>> Andy
>>> On 31 August 2016 at 05:45, ian <ioddity2001 at hotmail.com> wrote:
>>> I have 2 Digisound state variable filters (CEM type) and neither has any
>>> resonance to speak of just a very mild effect. I've read of others having
>>> the same issue so I am wondering if there is a common fault, a design issue
>>> or whether this module is supposed to have any resonance at all. Can anyone
>>> shed any light on this?
>>> _______________________________________________
>>> Synth-diy mailing list
>>> Synth-diy at synth-diy.org
>>> http://synth-diy.org/mailman/listinfo/synth-diy
>> _______________________________________________
>> Synth-diy mailing list
>> Synth-diy at synth-diy.org
>> http://synth-diy.org/mailman/listinfo/synth-diy

More information about the Synth-diy mailing list