[Elecraft] Fwd: New Sherwood report

[Kok Chen]

On Dec 3, 2010, at 12/3    6:44 AM, Jan Erik Holm wrote:

> So now we have a 2 ms claim for the K3. I know I measured it to 5 ms  
> but this was something like 2 years ago.
>
> What is it? Can Elecraft please tell.

Quite possibly the waveshaping is done by an FIR window.  So the  
design parameter may not be directly comparable to the risetime of a  
first order IIR.  (I hope Elecraft didn't take a shortcut and used IIR  
waveshaping!)

You can of course use the 10% to 90% points as the equivalent risetime  
even for an FIR waveshaping, but there are subtle difference from the  
"risetime" compared to a typical RC filter, especially when the "tail"  
of the impulse of the FIR is very different from a decaying exponential.

What I'm trying to say is that "2 msec" for one waveshape can produce  
vastly different spectrum from "2 msec" of a different waveshape when  
the standard of measurement is 10% to 90%.  The portions between 0% to  
10% and between 90% to 100% of the leading and trailing edges of a  
pulse are probably more important that what it is doing between 10%  
and 90%, which is usually a smooth function.

If hams apply a frequency domain criterion (like, say, -6 dB and -60  
dB points in the spectrum envelope of a sequence of dits) instead of a  
single number such as "risetime," we probably can more easily decide  
which rigs are problematical in regards to keyclicks.

73
Chen, W7AY