[sdiy] Digital 4 Pole Multimode Filter

[Richie Burnett]

If you already know how to implement the 2-pole digital (chamberlain) state variable filter in code, then the easiest route to a 4-pole filter is to cascade two discrete digital SVFs. For the purpose of clarity: pass the audio samples through one digital SVF then pass the output of that filter through another digital SVF.

You get some flexibility in the choice of overall response as each SVF can be LP, BP, HP or notch, and you can independently set the fc and q for each. You can also cascade even more stages for 6-pole, 8-pole, etc, if the desire is there and resources permit.

>From a programming point of view, each SVF has two internal integrator states, and most DSPs have dual accumulators, so it fits nicely to process blocks of audio through one 2-pole SVF at a time keeping the integrator states at maximum-precision in the MAC accumulators. This is harder to do with a moog 4-pole cascade that has 4 internal states. 

-Richie, 

Sent from my Xperia SP on O2

---- Scott Gravenhorst wrote ----

>
>Hi List,
>
>I'm looking for information on digital 4 pole filters.  I need to be able
>to control Q and Fc.  Among the many documents I found with google was this
>document:
>
>http://www.rane.com/pdf/linriley.pdf 
>
>which describes an analog version of a 4th order state variable filter, but
>my math lessons were so many years ago, that it's not entirely helpful.  On
>page 4 the document shows how the resistors are to be given values (right
>side) for a Linkwitz-Riley crossover.  I have a feeling this is close to
>what I want, but 
>
>1) I'm unable to determine how to change the filter so that I can control Q
>and Fc.
>2) I'm unable to convert this to a block diagram representation of the same
>circuit as a digital model.
>3) Information on how to tap the structure for modes other than lowpass
>would be very helpful.
>
>I'm familiar with the 2nd order state variable filter, but expanding that
>by adding two more integrators is not all that simple to my eyes.  I can
>see from the Linkwitz-Riley circuit that each integrator has a negative
>feedback path.
>
>I would also be interested in other 4 pole filter topolgies where access is
>given for Q and Fc control parameters.
>
>I would be extremely grateful for pointers to readings and information
>helpful in this quest.
>
>-- ScottG
>________________________________________________________________________
>-- Scott Gravenhorst
>-- http://scott.joviansynth.com/
>-- When the going gets tough, the tough use the command line.
>-- Matt 21:22
>
>_______________________________________________
>Synth-diy mailing list
>Synth-diy at synth-diy.org
>http://synth-diy.org/mailman/listinfo/synth-diy