Page 1 of 1

Is there any easy filters with bypass?

Posted: Mon May 18, 2015 5:37 pm
by djbrynte
im looking for a filter with.

bypass
lowpass
bandpass
highpass

and 12 - 24 db to choose

Re: Is there any easy filters with bypass?

Posted: Mon May 18, 2015 7:24 pm
by martinvicanek
What a coincidence: I have just made one for another project. Feel free to use.

Re: Is there any easy filters with bypass?

Posted: Mon May 18, 2015 7:51 pm
by djbrynte
realy nice but why is the lines blue?

Re: Is there any easy filters with bypass?

Posted: Mon May 18, 2015 8:02 pm
by martinvicanek
Because it is a static filter (controlled by knobs, not a control signal), you would not place it in the poly section. With this in mind, I have used SSE for optimization so it won't work in poly.

Re: Is there any easy filters with bypass?

Posted: Thu May 21, 2015 6:15 am
by martinvicanek
Added a dynamic filter version (poly capable).

Re: Is there any easy filters with bypass?

Posted: Thu May 21, 2015 11:34 am
by nix
Thanks mate!
- and for helping DJ B too

So can I just subtract the output of the bandpass from the original signal to get a notch?

Cheers, Nick

Re: Is there any easy filters with bypass?

Posted: Thu May 21, 2015 6:27 pm
by MyCo
nix wrote:So can I just subtract the output of the bandpass from the original signal to get a notch?


No, that would be a band stop filter. It might become a notch, when the band can get to zero, or very close

@Martin, I'm wondering if it is possible to implement filter overdrive into these ZDFs. For regular filters this is done by manipulating the internal delayed samples. Haven't had any luck with that in ZDFs.

Re: Is there any easy filters with bypass?

Posted: Thu May 21, 2015 8:05 pm
by djbrynte
wow very good martin :)

Re: Is there any easy filters with bypass?

Posted: Thu May 21, 2015 8:28 pm
by martinvicanek
Ah, nonlinearities and saturation in filters - a vast topic. If you follow the discussion at KVR it can get really scary how much brain power the guys put into VA filter emulation. I have not taken a deep dive into that matter yet. One thing that seems to be important to people is accurate tuning in self oscillating mode. Filters with an additional unit delay in the feedback path don't provide independent control of resonance frequency and feedback Q, that's one of the reasons why they seem to like ZDF.

My approach was the other way around: start from a recursive siine oscillator where you have complete control over the frequency, then introduce damping or gain. In the latter case, add some saturation mechanism to prevent the oscillator from exploding. For a quadrature oscillator, this can be done without affecting the center frequency. Now you can let the oscillator run freely or drive it with audio input, in which case you obtain a perfectly tunable resonant filter.

I still have the schematic on my HD but it would need a cleanup before posting.