[Elecraft] OK... I give up on what is happening...

[Don Wilhelm]

If I can offer a guess ---

The K2 *did* change from subtractive LO injection below 21 MHz and 
additive on 21 MHz and above.
My guess (and only a guess) is that MixW and Fldigi make the same 
assumptions for the K3 (and that is not correct).
I don't think the K3 is the problem, but you will find the solution in 
the software.  Check the forums for Fldigi and MixW for the 'solution'.

73,
Don W3FPR

On 11/4/2014 10:54 AM, David Cole wrote:
> Hi Ron,
>
> I am not sure it is a serial problem, in fact I would probably bet money
> that it is not...  Here is something new to ponder about this that may
> help you...
>
> If I go to the "CW is LSB" option in MixW, and assert it, all behaves
> fine, (both waterfall and signal track as I turn main tuning dial), on
> 80, through 17 Meters, if on the other hand I then switch to any band 15
> through 10 meters, I get the action where as I adjust the frequency up
> using the tuning dial on the K3, the "dial", (in the waterfall), on
> MixW, tracks fine, (as it does on the lower bands), but the signal moves
> in the opposite direction, so it looks as if the frequency of the signal
> is changing as observed by MixW.
>
> If I then uncheck the "CW is LSB" option, the actions are reversed--
> that is to say, that all works as expected on 21 through 10, and the
> reverse behavior is on 80 through 17 Meters...   This baffles me
> totally...  If it happened with the change boundary at 14 Mhz, (20
> Meters), I would say it is the rig changing sidebands...  But it happens
> on 15 Meters, working just fine on 20...
>
> So my first thought was that it was just the radio changing sidebands,
> so I set the MixW as "CW is LSB", all works on 80-17, I go to 15, see
> the reverse behavior.  I then then hold down ALT for changing to the
> opposite sideband...  The change happens, but the reverse and lack of
> tracking behavior continues, just on the other side of the waterfall.
>
> There are some very odd things about the K3 I still don't
> understand...
>
> If I load up FLDigi, I get the reversed behavior on all bands...
>
> Maybe the K3s are possessed, this baffles me totally...  At the risk of
> sounding frustrated, this did not happen on the Pro III, or PRO II I
> had... The serial port was also able to keep up, where I suspect the K3
> is processor bound and can't keep up with a 38K data stream.
>
>