Return to KLUBNL.PL main page

rsgb_lf_group
[Top] [All Lists]

Re: ULF: New experiment on 970 Hz 3/3 summary

To: [email protected]
Subject: Re: ULF: New experiment on 970 Hz 3/3 summary
From: DK7FC <[email protected]>
Date: Wed, 11 Apr 2018 15:13:24 +0200
Cc: Renato Romero <[email protected]>
Dkim-signature: v=1; a=rsa-sha256; c=relaxed/simple; d=posteo.de; s=2017; t=1523452405; bh=aNekfbjY43AHUXdUt9wzFJi+W9CJ8cB1gyKT+RJ0/SA=; h=Date:From:To:CC:Subject:From; b=n3EZLp5Pfn1DQYNL1OrF1w16csxTPI008HUxfqk7anHXvNpizzR42w7H0BECE3HoM SgehMlc9vTZKeUi2fSLIfjceEk6Hr6eNW+hVOl/R9m2KGNiBCUxhAsCNHErt6FbfAW JN7FFbjkguLTdRso3nXBNu+M/1G0owIEh6tXrhMBtm3q3IBo4kj03/FOD2VFnB76H/ +oXfejzz5TMDAANLuV3H2d0rmH5d6jVomRDcBb/2QzATYzru+fNGWCd36f1BgppfW4 /0L1sb1QtebtvOvKyyS+eJTFTkvJvg+1Gn857R9LAXO/fvIGzZ0Y54JlMjutPQbi2B SPhCba34T/R6w==
In-reply-to: <[email protected]>
References: <[email protected]> <[email protected]> <[email protected]> <[email protected]> <[email protected]> <[email protected]> <[email protected]> <[email protected]> <[email protected]> <[email protected]> <[email protected]> <[email protected]> <[email protected]> <[email protected]> <[email protected]> <[email protected]> <[email protected]> <[email protected]> <[email protected]>
Reply-to: [email protected]
Sender: [email protected]
User-agent: Mozilla/5.0 (Windows; U; Windows NT 6.1; de; rv:1.9.1.8) Gecko/20100227 Thunderbird/3.0.3
Hello ULF friends,

Here is the summary of my recent portable experiment on 970.005 Hz.
The goal of the experiment was to undoubtly detect my own transmission in a distance of 40.5 km. The experiment took about 8 days, starting 30.JAN.2018, 16:30 UTC. For the first 3 days, a carrier was transmitted on 970.005 Hz, for 48 hours, and then 24 hours on 970.0025 Hz. Afterwards, a 2 character EbNaut message was transmitted, taking 24 hours exactly, two times repeated. And finally, a 5 character message was transmitted, also taking 24 hours, two times repeated.

This is the path http://k7fry.com/grid/?qth=JN49BK33UG&from=jn49ik00wd which the 18 nW ERP signal had to cross.

Along with the other experiments i've done during that time, it took nearly two months for me to learn how to operate Paul's vlfrx tools and to do all the post processing.
I had to determine:
-The best noise blanker settings,
-The best filter shape, bandwidth and center frequency
-The optimum hum filter parameter settings
-The phase between E field (whip) and H field (loop)
-The best mixing coefficients (amplitude, phase) between E and H
-The phase difference of the signal between day and night
...all to maximise the SNR of the carrier peak and Eb/N0 of the messages.

Unfortunately the E field was not usable for most of the days. Either something got damaged on the front end JFET or the shielding effect of sourrounding trees has been much higher than expected. The performance of the 84 turn loop antenna having 0.8 m diameter and a low noise preamp was fine, though not optimal since there were AM broadcast signals audible during the sunrise and sunset (a resonance on HF, resulting in an overloaded preamp at times?!).
So all in all there were a lot of performace limiting issues.

Anyway, the two character message was decoded without much tweaking of blanker etc. settings. Based on these signals to analyse, along with the carrier, it was possible to find the best settings to even get a decode of the 5 character message in the end! The carrier result and the 5 character message is H field only!
Here are the results:
Spectrum peak of the 72 hour carrier transmission, combined in one peak: http://www.iup.uni-heidelberg.de/schaefer_vlf/ULF/40km/TR3_Endversion.png
2 character message: http://www.iup.uni-heidelberg.de/schaefer_vlf/ULF/40km/2ch970Hz40km_E+H.png
5 character message: http://www.iup.uni-heidelberg.de/schaefer_vlf/ULF/40km/5ch970Hz40km.png

So let me conclude the experiment was a full success, despite all the trouble and limitations i had. And it means that more distance can be expected if the RX system is working as intended!

This new distance record of 40.5 km on the 309 km band is still in the near field for that wave! See http://www.iup.uni-heidelberg.de/schaefer_vlf/ULF/40km/Nahfeld%C3%BCbergang.png

These are the locations of all experiments done on the 309 km band so far: http://www.iup.uni-heidelberg.de/schaefer_vlf/ULF/40km/970Hz_bisher.png

For the next experiment, i definitively plan to cross the far field border ( > 49.3 km). The idea is to use about 3 dB more power and a working RX system and a better location for the E field :-) When doing it in the summer months, the 3 dB more power will be compensated by QRN i fear.

A photo from the RX site in the snow: http://www.iup.uni-heidelberg.de/schaefer_vlf/ULF/40km/RX_site.jpg
A photo showing the data recorder: http://www.iup.uni-heidelberg.de/schaefer_vlf/ULF/40km/Raspi.jpg
And the batteries afterwards: http://www.iup.uni-heidelberg.de/schaefer_vlf/ULF/40km/Batt.jpg

73, Stefan

PS: Thanks to Paul Nicholson for the support in learning how to use vlfrx tools!





Am 08.02.2018 13:44, schrieb DK7FC:
Hello dear friends of the Ultra Low Frequencies,

Update:
There are a few things to note so far:

I made a mistake in the calculation of the ERP. It is not 10 nW but rather 18 nW.

All transmissions seemed to go out as intended. The stream from the local tree was interrupted almost daily due to lack of solar energy but at least a few hours came through. From these few hours i got decodes of each carrier transmissions. I also got decodes of all EbNaut messages. So things went well from the TX site it seems.

The 36 mA antenna current (12 kV antenna voltage) was running for 8 days without a single interruption. The constructed resonating transformer has done its job flawlessly.

Since i decided to take back the RX equipment on the 7th of February, there was some more time for another EbNaut message:
06.Feb, 17:30 UTC: EbNaut, 16K21A, 1 char, 90 sec, CRC27 (21:12:00)

The experiment stopped at 15:45 UTC. BTW the exact location was JN49BK33UG  That is a distance of 40.5 km to the TX site. But so far there are no decodes or traces, so no success so far.

The next good step is that the whole RX equipment has still been in its place, nothing stolen, nothing damaged. The Raspi was still running after 8 days. The Raspi was supplied by two 12 V / 26 Ah lead gel accus. They were standing in the snow and discharged to 11.8 V! The two accus for the preamps of the loop and vertical were still above 12.3 V. So that was all fine too!

Now the stuff is sorted, cleaned and the accus are charging and warming up again here.

I just started the Raspi and checked the data on the 64 GB USB stick. This looks good and promising, see attachment. And it looks like there were no timing breaks of the GPS module.
For the first time i run the recording using vlfrx tooly by Paul and the octo soundcard. Also it is the first set of data i have available to build a cardioid antenna by software... Exciting stuff.

The recording process by vlfrx tools saves to a new file each day.
Here the files are 7.8 GB large:  2 channels * 24000 samples/(channel*second) * 16 bit/sample * 3600 seconds/hour * 24 hours/day * 1/8 Byte/bit * 1/1024 kByte/Byte * 1/1024 MBity/kByte * 1/1024 GByte/MByte =  7.72 GB/day 

Obviously the recording process was stopped during the last night 03:58 UTC because the USB stick capacity seems to be just 58 GB actually. The recording is started/stopped by switching a switch which is connected to one of the GPIO pins. This is necessary because in the field there is no monitor and no keyboard of course. At start-up, a scripts is executed which checks the logic level of the pin where the switch is connected... During the recording process, a loop is running within this script which periodically checks the remaining available
space of the data carrier. As soon as the remaining space becomes less than 100 kB, the recording process is stopped. This ensures that there is still enough space to save the log files which are written at the end of the recording process. It also ensures that there are no crashes and unwanted data loss, broken files etc...

In a few days i can report about the results, so stay tuned...

73, Stefan




Am 31.01.2018 14:30, schrieb DK7FC:
Hi all,

Since yesterday, ~ 17 UTC, i'm running a new portable experiment on 970 Hz, or the 309 km band. The receiver site is somewhere in the forest in JN49BK, a distance of  40 km to the TX site.

I carried a lot of accus to that location and the USB stick where the data is written on has 64 GB, so i can run the experiment for more than one week! :-)

For the first time i'm using two antennas. An active E field antenna with a 1.5 m long steel wire as the capacitive probe. It is mounted on a 10 m high pole. The E field preamp is supplied by a separate battery and uses an isolation transformer with a low pass filter applied.
The second antenna is my compact loop with 80 cm diameter and 84 turns. It uses a low noise preamp (also supplied by a separate battery) using a LT1028. This loop RX worked surprisingly well on 4470 Hz when beeing in France last September. Hopefully it will pick up the signal.
Unfortunately the E field antenna is not very high above the sourrounding trees, so the sensitivity may be limited. Also the H field antenna showed much more mains hum than expected in this location. So it may be a bit questionable if the experiment will be successful. The long transmissions will allow to stack carrier transmissions and/or EbNaut transmissions though...

Running 36 mA antenna current means an ERP of 10 nW on that frequency.
My transmission plan is:
30.JAN, 17 UTC: Carrier on 970.005 Hz
01.Feb, 17 UTC: Carrier on 970.0025 Hz
02.Feb, 17 UTC: EbNaut, 16K21A, 2 char, 100 sec, CRC22 (24 hours)
03.Feb, 17 UTC: EbNaut, 16K21A, 2 char, 100 sec, CRC22 (24 hours)
04.Feb, 17 UTC: EbNaut, 16K21A, 5 char, 80 sec, CRC17 (23:49:20)
05.Feb, 17 UTC: EbNaut, 16K21A, 5 char, 80 sec, CRC17 (23:49:20)

Not sure if someone else can pick up the signal. Most likely not. But new components to rise the voltage to 20 kV in a next experiment are already in preparation...

73, Stefan


<Prev in Thread] Current Thread [Next in Thread>