[Elecraft] K3S audio board for K3 upgrade?

[Eric Swartz - WA6HHQ, Elecraft]

Hi Ron,

That review of the K3 audio by Rob Sherwood was on a very early K3. We quickly 
addressed the high freq harmonic issues (most of which were above 4-5 kHz and 
were more than 50-60 dB down) with a K3 DSP board revision that included a 
change to a beefier low ESR power supply filter choke feeding the audio amp and 
the addition of an audio LPF that rolls off above the DSPs max output of 4 kHz. 
We made even more audio improvements to the new K3S DSP board's audio chain.

On another note, Rob has run spectra on the newer K3S DSP and he has noted it is 
substantially improved over the first K3 he tested almost 7 years ago.

One thought for those using the 2.7 or 2.8 SSB filters. Try reducing the DSP 
bandwidth to 2.4 kHz or so, and shifting the center freq. down 100 Hz.  Many 
other radios have a 2.4 kHz SSB filter, and while it does restrict the audio 
fidelity as compared to the 2.7/2.8 kHz filters, the reduction in the upper 
cutoff to 2.4 kHz reduces the amount of high frequency band noise and 
interference, which may account for what some construe as a 'smoother' sound.

Also note that the stock 2.7 kHz filter has a much slower roll off than the 
optional 8 pole 2.8 kHz filter, and as a result may hear more higher frequency 
RX components.

Lastly, try listening with our new external SP3 speaker for the K3S and K3. It 
has an engineered audio response for a very clean and 'smooth' flat response, 
without major resonances.   :-)

73,

Eric
/elecraft.com/

On 3/21/2016 1:59 PM, Ron Castro wrote:
> The sounds being described are probably not just related to frequency
> response, but to any type of harmonic, intermod or digital distortion, or
> noise that is created in the final IF and analog audio stages of the radio.
> This was referenced by Rob Sherwood in an early review of the K3 and
> demonstrated with an audio spectrum analyzer.  A CW tone with these
> characteristics might be described as "raspy", "brassy", "busy", or with
> some other term audiophiles like to use, and in any case, would likely lead
> to fatigue.
>
> If this is the case, equalization may mask it but won't fix it, and could
> make the problem worse.
>
>            Ron Castro
>             N6IE
>