Hi Jim,
Am 20.12.2018 23:26, schrieb [email protected]:
Hello
Stefan,
Very
nice signal at 80 Hz (seen in your grabber plot).
Your
DFCW-5400 plot seems to indicate about 25dB-30dB SNR during a quiet
period around 02:30 on 27 November; does 25-30dB SNR in the quietest
period 27 Nov – 29 Nov sound approximately correct (in other words am I
interpreting the plot correctly)?
Yes you do. But the noisy periods were not from QRN but from wind! At
least, when i listen to the stream it sounds like wind which is causing
E field noise. There must have been some charge somewhere, maybe
charged small rain drops. The antenna is grounded by a 100 MOhm
resistor.
Is
5400 is equal to or similar to the number of seconds of integration?
There is a feedback loop of the soundcard output to the soundcard input
which is there to correct eventual phase glitches. Thus, SpecLabs
digimode terminal can't be used to transmit DFCW. That means, i have to
write a command line in the scheduled actions for each start and stop
(and QSY) of a dash. 5400 is just the length in seconds of each dash.
This is 90 minutes and so it is easy to calculate the start and stop
times.
It
is exciting to read that you are headed toward 23 Hz.
Oh, just a few minutes ago i managed to produce 4 kV rms at 23 Hz. Not
as easy as i expected.
I
had guessed that natural noise at 23 Hz might be ~ 20 dB higher than at
9 kHz, and that an impulse blanker (lightning noise blanker) might not
work so well at 27 Hz because the long tails of the sferics overlap.
Natural signals (“noise”) at 1 Hz and below can have long spatial
correlation distances (many hundreds of km) which would enable noise
reduction by subtraction of digitized signal from a distant receiver
(much farther from the transmitter), from the digitized signal of the
local receiver. I don’t know if the spatial correlation lengths are
anywhere near as long at 23 Hz as at 1 Hz (23 Hz has a greater ratio of
sferic noise to magnetotelluric noise), but still wonder if you could
subtract raw data from another receiver perhaps 170 - 230 km away to
cancel 23 Hz noise captured by your local receiver?
This is possible in vlf-rx tools and there is a receiver in that
distance but i never used the possibility so far.
More soon.
73, Stefan
73,
Jim
AA5BW
SLF,
For the completion of this thread, here is the reprocessed spectrogram
of a DFCW-5400 message at 80 Hz (attachment).
It is just a small distance but anyway challenging, on the TX side as
well as on the RX side.
Usually i want to detect a carrier, then an EbNaut message and finally
a DFCW message when visiting one of these really low bands.
Next step is ELF at 23 Hz but this is even more challenging. Thus there
is some delay...
73, Stefan
Am 27.11.2018 09:25, schrieb DK7FC:
Hi SLF,
Yesterday, a 45 character EbNaut was transmitted at 80.005 Hz. It was
received on my tree in 3.5 km distance. After the improvement of the RX
antenna it looks like i gained about 2.5 dB SNR. It was an easy decode,
the message could have been transmitted 4 times faster.
Capture attached :-)
73, Stefan
PS: With that SNR, a 5 character message could be transferred out to 7
km distance in less than 1 day.
Am 23.11.2018 10:21, schrieb DK7FC:
Dear SLF friends,
A new step forward towards DC: Since yesterday 22:30 UTC i'm TXing a
carrier at 80.005 Hz. That's the 3750 km band, where
the far field begins at a distance of 598 km.
The antenna voltage is just 5 kV and i'm getting 1.2 mA antenna
current, giving an ERP of 20 fW (2E-14 W).
The voltage is even higher than in my recent experiment on 270 Hz,
anyway the S/N is lower, indicating that my RX lacks of sensitivity.
Indeed this experiment helps to estimate the lack of sensitivity. Looks
like i'm missing about 10 dB on that frequency. Anyway the signal is
making the path of 3.5 km to my tree grabber, as usual shown at http://www.iup.uni-heidelberg.de/schaefer_vlf/DK7FC_VLF_Grabber2.html.
Not a big distance but still well beyond most garden fences!
The signal leaves a barely visible trace in the ZEVS window which is
running at 3.8 mHz FFT bin witdh. Unfortunately ZEVS is off the air
since a week or so. actually i wanted to transmit side by side with
ZEVS, of course a bit deeper even.
>From a quiet period at night, i calculated the SNR of a 1 hour carrier
period using vlfrx-tools: carrier S/N: 16.83 dB in 277.8 uHz,
-18.73 dB in 1Hz
For the crazy homebrewers i'd like to share that website where i
ordered the HV-transformer, http://www.hosin.de/Specials/HV-Trafo/hv-trafo.html
A nightmare for all safety-fetishists! For the completeness it must be
mentioned together with that page: http://kurts-werkstatt.de/hv-trafo.htm
73, Stefan
|
|
