Return to KLUBNL.PL main page

rsgb_lf_group
[Top] [All Lists]

Re: LF: SL settings for EbNaut FFT recording ?

To: [email protected]
Subject: Re: LF: SL settings for EbNaut FFT recording ?
From: Markus Vester <[email protected]>
Date: Wed, 3 May 2017 13:15:39 -0400
Dkim-signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=mx.aol.com; s=20150623; t=1493831740; bh=iQbdbiVvJcShcMofXoGyEHIxE/cqNiwtWxxWMYTElYg=; h=From:To:Subject:Message-Id:Date:MIME-Version:Content-Type; b=Y9MYKoqVxdgMjnu++7dGHusEh7KGcvjSLu4f5V8upMKopbdwUZRpy7sYeFucezV9x 4JQ5UO6edEUtLc0xof51EnqjHo9HVdlnaEuZ45yB6lFap3QZAPm/ge57AlcEJCvKXI Q7IdoNcPP2EJ097UI4X4kWFoP6PPn23ie0l30mSE=
In-reply-to: <579355A36AEE9D4FA555C45D556003AB8FDD6AEA@servigilant.vigilant.local>
Reply-to: [email protected]
Sender: [email protected]
Hi Luis,

confirmed, running EbNaut v0.7 with 05022225.wav crashes here too.

The reason seems to be the date and time format in the timestamp at the beginning of the 05022225.txt file. The first two entries are
 2017-05-02 22:25:45.9
while they should be
 20170502 222545.9
which is intended to be read as two decimal numbers by ebnaut_ifft3a.

You can correct it manually by editing the txt file and running ifft3a again, which will produce a usable wav file with valid timestamp.

There is a setting in SpecLab's FFT export tab which should define the date and time format as YYYYMMDD hhmmss.s . I may have overlooked that when making the usr configuration file - sorry!

Best regards,
Markus 
 


-----Ursprüngliche Mitteilung-----
Von: VIGILANT Luis Fernández <[email protected]>
An: rsgb_lf_group <[email protected]>
Verschickt: Mi, 3. Mai 2017 14:01
Betreff: RE: LF: SL settings for EbNaut FFT recording ?

Hi Markus
 
Thank you again for the detailed explanation
 
To try a decode from Stefan transmission yesterday at 18:00 I have choosen the file 05022225.txt
This should contain the 7.9h FFT from 17:58 to 03:52
 
The settings on the SR.txt file contains the corrrected SR of the soundcard, decimation, FFT lenght and central frequency
 
This produces the WAV file with ebnaut_ifft3a correctly
But I get Ebnaut crash when opening the file with the decoder. The settings are:
 
Mode:16K21A CRC:16 Symbol:27
Mensagge lenght:5 Start offset: 0.000 (untill checking the real start time by Ebnaut)
List lenght 20000 Freq offset: 0.1  (for 8270.100)
CPUs:1 Phase Step: 30
 
Have tried to short lest lenght to 200 but it crashes anyway. It is Ebnaut_rx V0.6
 
Any idea of what's going on ? Didn't get a problem before with shorter (10 minutes) FFT files
 
I have got an small trace in the spectrogram at 8270.110
The QRG do not match with Stephan's transmission, but here is the picture
 
 
73 de Luis
EA5DOM


De: owner-[email protected] [mailto:[email protected]] En nombre de Markus Vester
Enviado el: martes, 02 de mayo de 2017 23:37
Para: [email protected]
Asunto: Re: LF: SL settings for EbNaut FFT recording ?

Hi Luis,

> BTW. What I'm getting from SL is one FFT file 112Kb size every hour as Markus said
> But how can I merge all the files of a 8h transmission like the one your are running now ?

Each of these files contains the data for approx. 8.9 hours. The filename (and the timestamp, which are the first two numbers in the file) refers to the middle of the recording. For example, a file called "05020430.txt" would be centered on 4:30 UT, and contain all received data from 0.03 to 8:57 UT. The next one would then be "05020530.txt", covering 1:03 to 9:57, etc. So the files are largely overlapping. A short transmission could be contained in more than one file, and a very long one (> 7.9 h) might not be completely covered at all. The easiest way is to select the file whose name is closest to the midtime of the EbNaut transmission.

Don't let yourself be confused by the file timestamps in Windows Explorer, which may be one hour later than the end of the recording. The reason is that SpecLab closes the "old" export file only immediately before the next one will be written. This is also the reason why you cannot delete the latest file while SL is running, but you can copy it to another folder.

Usually the data will have started some time before the announced beginning of an EbNaut transmission. The EbNaut decoder will show the file start time, and you will then have to calculate and enter the start offset.  

Best 73,
Markus   


-----Ursprüngliche Mitteilung-----
Von: VIGILANT Luis Fernández <[email protected]>
An: rsgb_lf_group <[email protected]>
Verschickt: Di, 2. Mai 2017 22:54
Betreff: Re: LF: SL settings for EbNaut FFT recording ?

Hi Stefan

The grabber is now showing full spectrum from 0-24KHz

The Rx is not at Benidorm in the high tower. It is in the roof of a two level house on top of a 5m pole
But clear out from high obstacles and trees
Fortunately no near by AM or LF stations. The strongest station in the first 500KHz is an Algerian LW broadcast
at 252Khz which puts -18dBm here in the central carrier

The pole is a 50cm x 10mm diameter aluminium rod. Will play with the 10MOhm resistor and first fet stage
with filter. I will also test a different location for the probe. The noise at MiniWhip rised 20dB, so must get it
appart to keep the ear at MF/LF. The power supply is a 12V transformer shared with the MiniWhip
Will test another one just for the VLF probe and with better filtering. The GND of the antenna is separated
from mains earth and there is an audio transformer to the PC soundcard

>>It is all not easy! I would say it is harder than EME ;-)

You can say it truly Stephan. It IS much harder than EME :-)
In EME you can run a modest station and the big guns will work you (my case, for example)
In VLF you have to be the big gun and try to target some electrons into modest stations with small antennas
into a noise storm

>>31 mHz? That's quite optimistic! I would rather try 31 uHz.

Sorry, it was a typo. It is 31uHz really. This are the SL settings by Markus

BTW. What I'm getting from SL is one FFT file 112Kb size every hour as Markus said
But how can I merge all the files of a 8h transmission like the one your are running now ?

Your signal in your grabber looks stronger than the previous transmissions

73 de Luis
EA5DOM


De: owner-[email protected] [owner-[email protected]] en nombre de DK7FC [[email protected]]
Enviado: martes, 2 de mayo de 2017 14:55
Para: [email protected]
Asunto: Re: LF: SL settings for EbNaut FFT recording ?

Hi Luis,

Am 02.05.2017 10:49, schrieb VIGILANT Luis Fernández:
I was already running an SL instance showing the first 24KHz in a fast way to get an idea of what's going on with QRM and known signals
It is already included in the grabber at 1 min/pixel. Hour changed to UTC just now
Yes but it starts at 6 kHz, so not the first 24 kHz ;-) From 0...6 kHz there are also informations which are relevant to evaluate the situation of your receiver(site).

 
As you can see, there is a LOT of noise around. What I'm using is your VLF E-Field probe with the first JFet separated in a 50cm aluminium rod probe
Not only the length of the rod, also the height above ground, the effective height, playes an even bigger role. Sourrounded by trees you miss sensitivity, the lower the frequency is. And on the top of a high tower, the signals will be much stronger, with the same rod!
If there is a strong local AM transmitter on MF, you may get some interferences, and so on. Thus it is good to see what happens below 6 kHz too.
When DLF on 153 kHz has been active here, running 500 kW to a 350m high antenna in 40 km distance, i sometimes saw AM components, demodulation stuff on VLF, in one of the first versions. So when the antenna is on a high tower and the signal levels are strong, you may afford a second RC low pass filter in front of the first JFET, giving a cleaner spectrum on the lower VLF. On the other side, with a longer rod (giving more capacity) you may decrease the 10 MOhm resistor. This should reduce the noise of the first JFET a bit.
I just played with all that stuff a week ago when optimising my ULF E field circuit for 970 Hz.

Obviously, no clear trace from your last transmission in the 31mHz spectrogram with this Rx conditions
31 mHz? That's quite optimistic! I would rather try 31 uHz.
Are you using an AF isolation transformer between soundcard and antenna? What type of power supply are you using for the antenna? Is the GND of the antenna separated from the mains earth?

So, 7Km is an outstanding Dx at this moment. But really challenging 
 
I do already have a failed microwave oven to use the transformer. And the audio PA too. So you mean to apply the high voltage to the antenna
without resonating coil, right ?
Yes.
Better start with some measurements in the shack, not on the roof :-)
Measure or estimate the capacity of your antenna. Switch a fixed C of that value parallel to the transformer. Use a sine signal generator connected to the audio PA. Try to find a resonance or check the voltage at 8270 Hz. Maybe a normal mains transformer 12V -> 230V will give better results! Maybe you need to add your own primary winding to the transformer. Try to get 1 kV or so, that should work well. With a mains transformer, 500V may work for a few hours. If you have a toroidal transformer, use that!
You can also switch an external C parallel to the transformer if that helps to rise the voltage on the desired frequency. Or switch it in series to the primary winding. Ah and a resistor, maybe 2 Ohm or so may help to protect the PA from transients and short cuts (resonances).



Nice therm. Don't know if the pipe running down to ground would have any issue with this. At the moment 80w at MF didn't show to be harmful
But there can be some noise at 8KHz and this would really be an annoying side effect if that happens :(
The antenna currents will be very low. There will be no unwanted effects, except maybe some local E field coupling on the roof.

 
The resonating coil for VLF is in the "To-Do" list. Actually found a huge plastic (resin) garden flowerpots which are 60cm diameter by 35cm high
This can be stacked upside down leaving airgap between them for ventilation. Just adding one final on top with no coil serves as rain protection
of the full arrange.
Maybe it's a good idea to start in small steps. With your own monitor you will have some first results. And maybe some of the local EA amateurs can help to reach the next distance step :-)
It is all not easy! I would say it is harder than EME ;-)

73, Stefan

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