Hello Simon,
The straight vertical lines in the waterfall are harmonics of a stable
clock, perhaps your computer. The "wavy" vertical trace around 7054
looks like the noise from a mains-powered noise source like a
switch-mode power supply or variable-speed motor controller -- the clue
is that it is not stable in frequency. In both of the power-related
sources, the mains voltage is rectified, minimally filtered, then
chopped to produce square waves in the range of 10-20 kHz. For the SMPS,
the waveshape and frequency is more or less constant (but unstable, so
it drifts on startup), while for the motor controller, is varied to
control motor speed. Both of these sources are heard as "growley"
carriers. Both of these sources are normally surrounded by noise, which
is heard as random. In this plot, I see another source at about 7058.
This trash is usually coupled to wiring connected to the source as a
common mode signal, and that wiring acts as a transmitting antenna. It
is not uncommon to hear this stuff several miles from a very strong source.
Gauging by the FT8 signals centered around 7075 kHz, I'd call this QTH
pretty noisy. I don't see anything in the waterfall that looks like CW.
Spectral and waterfall plots like this are an excellent tool for
studying noise. One change I'd suggest though is that the vertical scale
be expanded so that the strongest traces fill the spectrum plot. For
this particular trace, I'd suggest an upper limit of about -100 dBm.
You might find my RFI tutorial useful in thinking about this.
http://k9yc.com/KillingReceiveNoise.pdf Also, NK7Z has a lot of great
RFI stuff on his website, including a large collection of spectral and
waterfall plots of known sources.
73, Jim K9YC
On 12/28/2018 9:22 AM, Simon wrote:
This is an outdoor SOTA receive-only test to evaluate FT8 weak signal
performance.
I have near zero HF experience. Is this considered as 'very good' LOW
noise area? -130dBm noise floor on 7074kHz.
Hope it is not too off topic for the list.
It is a 500 meter hill, protected by country park laws, prohibited
residential building and no people live there.
They also have airport long range radar and HF ship to shore receiving
station in that area (on next nearby hill peaks). May be, the area is
chosen as 'low noise area'?
Test with
SDRPlay RSP1A
Thin Long wire, quarter wavelength at 7MHz, 40m band, just 1 meters
above ground, for simple first test.
Time 1040Z, 1840 local time. Just shortly after Sun set.
Noise floor is -130dBm on SDRUno softwre.
The strongest 7074MHz FT8 signal is -100dBm. I got 3 to 4 decodes per
15 seconds cycle. On WSJT-X Ver 2.0, SNR ranges from +4dB to -20dB,
from station 4000km away from Hong Kong. Including Indonisia,
Australia, New Zealand and Japan.
73
Simon
_______________________________________________
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel
_______________________________________________
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel
