[sdiy] SVF & phase

[Don Tillman]

   > Date: Wed, 27 Nov 2002 08:57:44 -0600
   > From: Grant Richter <grichter at asapnet.net>
   > 
   > Mark's question makes me question my assumptions about filters. Maybe
   > someone can clarify.
   > 
   > I am under the impression that a filter without phase change
   > would not be a filter at all?

Pretty much, yeah.  However...

Digital FIR filters can have any phase or amplitude response (wobbling
within a specified accuracy) and are typically built with zero phase
shift.  Of course they have a fixed delay on top of that.

And there are some mechanical operations that work in a manner similar
to digital FIR filters -- someone mentioned tape head azimuth.  There's
also the frequency spectrum of a guitar pickup (see my articles for
details).  And maybe the filtering effect of skin effect on a cable
(okay, that's reaching).

An SVF (State Variable Filter) is a circuit topology that directly
parallels the classic Nth order filter transfer function.  So for an
SVF the phase shift is exactly what the math says it is.  For synths we
typically use a 2nd order SVF which gives us nice control over
Frequency and Q.

Now it's possible to add phase shift stages to compensate for the
phase shifts of the filters.  The result is a sort of hybrid; a larger
Nth order system with a smaller Nth order amplitude response.  But
that's a lot of work for an approximation.

So now... I'm curious what the application is where phase shift is
such a problem.

  -- Don

-- 
Don Tillman
Palo Alto, California, USA
don at till.com
http://www.till.com