with more effective autonotch filter settings and Wolf's latest
comb-notch option, I decided to have another go at receiving 6470 Hz.
It looks like the railway QRM in the narrowband spectrograms has indeed
gone down significantly. Of course with aggressive filtering and
blanking there's always a risk to kill desired signal as well, throwing
out the baby with the bathwater. As a check, I have added two
artificial carriers at -110 dB target level (same as DL3NDR on 8.27
Stefan, if you like you could perhaps try another long 6.47 kHz carrier
tomorrow or Sunday. If I can see it we mihght be able to compare
daytime propagation patterns between the two frequency bands.