Re: VLF: Carrier detection DL > VK7 on 17.4701 kHz

[Markus Vester <markusvester@aol.com>]

Hi Stefan, Edgar,

>> boundary of the garden fence
Well the signal did end up in a garden fence. Only this time the garden is on the other side of Earth ;-)

Absolutely impressive and truly groundbreaking work!

Best 73,
Markus (DF6NM)

-----Ursprüngliche Mitteilung-----
Von: DK7FC <selberdenken@posteo.de>
An: rsgb_lf_group <rsgb_lf_group@blacksheep.org>
Verschickt: Fr, 6. Apr 2018 14:07
Betreff: VLF: Carrier detection DL > VK7 on 17.4701 kHz

Hi VLF,

A few weeks ago i had the crazy idea to try an attempt to be detected on
17.4701 kHz in Tasmania / VK7.

Edgar J.T. is the one on the other side. A few years ago he managed to
detect my DFCW-180 signal on 136 kHz at two different locations. He is
currently watching ZEVS as well as other special signals on the VLF.

Using an omidirectional whip antenna on his side we attempted to detect
a pure carrier transmission that was transmitted daily from my site in
JN49IK00WD, beginning at March, 15th.
Each day i transmitted from 16...23 UTC with about 1 A antenna current
on my INV-L antenna wire 30m above the ground. My average ERP estimate
is 2.5 mW on that frequency (average estimate, not average ERP ;-) ).

At the beginning there were frequency stability problems on the RX site
but since the 21st of March the system was running well and stable. So
we could observe a growing SNR during certain times within the stacked
files.

We ended up stacking 13 daily transmissions into one file to analyse. I
build a simple script in vlfrx tools to produce a set of data which was
plotted in Excel. Here is the latest plot:
http://www.iup.uni-heidelberg.de/schaefer_vlf/VLF/DL_VK7_17470.png

The curves show the SNR of the carrier for different start times and
durations. The best SNR (12.70 dB in my stack) is obtained when starting
at 19:20 UTC and watching the signal in 93 uHz, i.e. 3 hours.

After reaching about 12 dB SNR it became clear that this will become a
successful detection. So i informed Paul Nicholson to evaluate the files
and asked for a critical comment regarding the significance. He used his
new fast PC to run a number of permutations of the files to produce an
even better SNR, by dropping a few days which do not contribute
positively to the stack. In the end we ended up at > 14 dB SNR and found
the strongest peak of the file(s) exactly on the expected frequency.

I leave it to Paul to present the best spectrum peak and comment further
on the statistical significance of the signal.

BTW this is the path: http://k7fry.com/grid/?qth=QE37PD&from=JN49IK00WD

My thanks and congratulations go to Edgar J.T. for the successful
carrier detection and the permanent available RX system, as well as for
daily providing of the txt files containing the FFT data. Oh and not to
forget about the patience :-)
Also thanks to Markus and Paul for the discussion and support in the
background. And not to forget about DL4YHF for developing SpecLab, which
is a part of most of these achievements.

The boundary of the garden fence has been pushed a little bit more these
days!

73, Stefan