[Elecraft] Spectrum pic

[Wes Stewart]

Thanks Brian.

I've been reviewing my test setup.  Because of limited USB ports on my normal 
station computer (laptop) I have been using a new (to me) desktop I'm building 
to be the station computer, to run both WSJT-X and my SDR-IQ software 
(SpectraVue) for this investigation.  The sidebands I mentioned are quite 
noticeable but it started to not make sense to me in the way they moved as I 
changed the tone frequency.

To eliminate possible crosstalk between programs I went back to the laptop 
running WSJT and used my other K3 as a manually tuned spectrum analyzer.  The 
sidebands are still there but way down in the mud.  I think I'm hearing what's 
being generated in the SDR-IQ but I have yet to prove that, or understand why 
they might be generated.  It's not overload, the main signal and sidebands 
change amplitude dB per dB in sync with a stepped attenuator on the input.  And 
as I said elsewhere, when I was suffering TX IMD issues this same receiver was 
used to document the problem and when the radio went back to Elecraft for mods, 
they agreed with my numbers.

That said I decided to check calibration on the SDR-IQ while running on the new 
computer.  A lot of the processing is done in the SDR-IQ so I didn't expect an 
issue other than tweaking FFT gain in the SpectraVue program.  I connected the 
XG2 and tuned it in on 14.06 MHz.  The trace was jumping in amplitude about 3-4 
dB at about a 1 Hz rate.  I also run Lady Heather via a serial port to a 
Motorola Oncore timing receiver (1PPS) and I've seen it take over my mouse 
before, making it jump all over the screen.  So I disconnected the serial port 
without effect.  I rebooted and started only SpectraVue and saw the same issue.

One last experiment (for now) was to use a third computer to run the SDR-IQ.  On 
this one everything is perfectly stable.  So I have a baffling problem with one 
computer, unfortunately the one I was using to present earlier data, which are 
of course now suspect.

Nevertheless, the cyclic ALC is still there and the difficulty in setting levels 
is too.

Frankly, I don't know why I'm doing all of this for a mode that I don't even like.

Regards,

Wes


On 8/5/2017 2:47 PM, Brian Hunt wrote:
> Not seeing your -50 dBc side bands. Maybe there is a little blip at -68 dBc. Pic below is WSJT-X FT8 @ 1500 Hz with 100 Hz/division, BW 10 Hz. The old scope is pretty drifty at those numbers. 😄
> I'll add it to my photo site.
>
> 73,
> Brian, K0DTJ
>
>
>