[hpsdr] ADC Overload, a conclusion.

[Jeff Cook]

Using KISS Konsole (thanks George) as a spectrum analyser I was able to 
see a large peak very much down at the LF end of things.  Going over to 
the narrow band view and tuning, I found the peak to be at 0.7033 MHz 
with diminishing harmonics at either side.  This peak was at some -18 to 
-16 dB's and was static in amplitude and width, therefore not a 
broadcast station but something much closer to home.  This ghastly RF 
noise was being generated by my main switch mode power supply which 
supplies 29 amps at 12 V.  Switching this off and reverting to my linear 
25 amp power supply removed the spike, further tuning revealed a another 
series of spikes centred on 0.4525 MHz only half the size but 
nonetheless there.  These were being generated by another switch mode 
power supply, a smaller one with 3 different voltage outputs on it used 
to power my Trimble Thunderbolt GPS disciplined clock, the 10 MHz output 
of which I use to synchronise Excalibur.

This has completely solved the ADC overload problem I was having on 
receive, however I'm still grappling with a transmit ADC problem.

Thanks to all that helped to point me in the right direction.  I'm now 
going to do a carte blanche capacitor replacement in my two switch mode 
power supplies, or, I may just chuck them in the bin.

Jeff Cook, G0AFQ.