[Elecraft] APF vs 10 Hz DSP, why they don't sound the same

[Bill Tippett]

 > More generally, the APF proponents seem to have the problem that they
treat it almost as magic.  If they could characterise what it does in
terms of filtering, AF gain adjustment, distortion etc. it should be
possible to simulate it in DSP, subject only to having enough processing
power.  In particular, peaking just seems to me to mean turning up the
AF gain in parallel with reducing the bandwidth. If that is all that it
is doing, and if the K3 has soft AF gain controls, a plugin box could
issue both commands in parallel.

         The circuit in question is on page 54 of the FT1000 schematic below:

<http://www.radioamateur.eu/schemi/FT1000_user.pdf>http://www.radioamateur.eu/schemi/FT1000_user.pdf

The APF is around Q3016-2 (including Rs 3062, 3065, 3067 per the 
"un-modification" sheet on N1EU's website below).  The circuit 
schematic above has *not* been modified to the (incorrect) values 
shown in the Yaesu service bulletin.

http://n1eu.com/Mods/apf.pdf

The FT-1000's APF is generally regarded by many as the best APF ever 
done.  It should be possible to plug this into a circuit simulator 
and generate a frequency response plot that Lyle can emulate in the 
DSP.  I have a feeling that this is a very narrow and high-Q response 
that is on the verge of ringing, but it should be possible to 
duplicate the circuit in the DSP.

                                         73,  Bill  W4ZV