[Elecraft] APF in 4.16/2.63 works well!

[Paul Christensen]

> The design goal was to duplicate the FT-1000D APF.  The FT-1000D has
> no variable Q function.  What you are proposing is completely different
> than the design parameters ... you want an Autek QF1 not the FT-1000D APF.

I agree with Joe on this. The bandwidth of a CW transmission is narrow and 
does not change.(*)  I think pursuing variable Q would amount to diminishing 
returns over fixed Q and adds an unnecessary variable.   I also like the use 
of the context filtering - a feature not found in the original FT-1000/D 
circuit.

As Joe suggests, an Autek QF1A can be used if one really wants extreme APF 
flexibility, or a single-channel parametric EQ may be a better way to go 
since a PEQ allows the user to alter gain, Q, and frequency.  Either way, 
those devices can be easily inserted into the K3's audio path.

Paul, W9AC

(*) CW bandwidth does change with speed but at normal CW keying rates, 
bandwidth is largely a function of only the waveform slope.  At high 
switching speeds, bandwidth becomes relevant as a function of both waveform 
slope and keying speed.