I seem to remember that the 50/100 Hz sidebands were quoted as -45dB on the
main signal for the growler on 73 kHz. I think we have to be just a little
careful when we interpret displays from sound card FFTs on very big signals.
Despite the bit count I suspect that 45dB is approaching the dynamic range
of the system. There are also effects that can occur which are a function of
the sound card AGC, and the FFT processing, sample rate, windowing,
smoothing, and anti-aliasing. Not to mention intermod effects.

Nevertheless I often see a pair of 100Hz sidebands on the stronger signals
received here on 136kHz. G3KAU, G3XTZ, G3YXM, G3KEV, G3LDO. These are
usually of the order of 30-35dB down on the main signal, but I am not
convinced that this is an accurate view of what is really happening. I
suspect a conventional Spectrum Analyser closer to the Tx would report much
lower levels.

I have noticed that some signals will produce a ghost image that I think is
an intermodulation product of some kind in the wide-band display on FFTDSP4,
but not on the normal display.

Cheers de Alan G3NYK
[email protected]