Re: [rsgb_lf_group] Re: LF: Re: EbNaut test transmission

[Markus Vester <markusvester@aol.com>]

Hi Andy,

 

below are bit-error derived Eb/N0 results of your decodes here. The gap between 17:20 and 20:10 was me taking a break, and 21:20 to 22:20 washed out by local QRM. I'll call it a day now, but SpecLab will save data overnight. 

73 and good night,

Markus (DF6NM)

time  Eb/N0
 UT    dB

 

1600  8.0
1610  9.4
1620  9.6
1630 11.2
1640 10.1
1650  9.4
1700 12.1
1710 11.8

 

2010  8.3
2020  9.7
2030  7.3
(local qrm begins)
2040  2.2
2050  1.6
2100  2.4
2110  2.4
2120 to 2220 nil
(local qrm ends)
2230  2.3
2240  5.4
2250  7.2
2300  7.6
2310  8.8
2320  5.3
2330  7.2

 

-----Ursprüngliche Mitteilung----- 
Von: Andy Talbot <andy.g4jnt@gmail.com>
An: rsgb_lf_group <rsgb_lf_group@yahoogroups.co.uk>; rsgb_lf_group <rsgb_lf_group@blacksheep.org>
Verschickt: Di, 17 Nov 2015 11:50 pm
Betreff: Re: [rsgb_lf_group] Re: LF: Re: EbNaut test transmission

Ok, I'll leave it running overnight.   After  the rain stopped,  antenna resistance fell and power output rose closer to 300 Watts of RF  into the lower Rant, so EIRP is about +20dBm now.

To reiterate for any new readers:

   Frequency 137780.000  +/- 0.2mHz

   Coding 8K19A  

   Symbol period  0.5s   (2 symbols / sec)   BPSK with 10ms Gaussian  shaped phase transition

   Message length is 9 characters consisting of a callsign and the abbreviated name of a well-known cockatiel

PC time is synchronised using rsNTP,  set for periodic updates which it appears to do every 30 minutes.  So far it hasn't made a correction of more than 20ms (ignoring the first one back at 1611z)

The DOCXO (almost a TrOCXO)  appears to have stayed well within 1ppb of its correct frequency all afternoon / evening.   I check by comparing a divided down 1PPS signal against GPS,  looking to see how many microseconds the two respective edges drift apart over a few hours.   If it moves by more than that corresponding to 2 PPB   (7.2us per hour)  a slight tweak to the frequency control pot gets it back again.

For really serious work, the Caesium tube could be turned on, but that requires about 60 to 90 minutes for warm up to reach stable operation.    Something only worth doing for long duration tests at lower rates.  

The last part of the message comes from this chap, who's ghost still haunts my shack ...   http://www.g4jnt.com/basil.htm

Andy  G4JNT

 

On 17 November 2015 at 22:09, Markus Vester markusvester@aol.com [rsgb_lf_group] <rsgb_lf_group@yahoogroups.co.uk> wrote:

 

 
I have set up a dedicated small SpecLab instance for EbNaut data, with spectrogram uploads to

 https://dl.dropboxusercontent.com/u/26404526/EbNaut_spectrogram.png .

The signal is generally weaker now, but still good decodes (2.2 dB Eb/N0 for 21:00 UT). As far as I can see, transmit timing and frequency are perfect.

 

My auxiliary EbNaut utilities 

 http://df6nm.bplaced.net/VLF/fec_tests/df6nm_ebnaut_utilities.zip

include opds-32 SpecLab settings, along with the text file containing info how to get-started. Hope it may be useful...

 

Best 73,

Markus (DF6NM)

 

-----Ursprüngliche Mitteilung-----
Von: Markus Vester <markusvester@aol.com>
An: rsgb_lf_group <rsgb_lf_group@blacksheep.org>
Verschickt: Di, 17 Nov 2015 6:54 pm
Betreff: Re: LF: Re: EbNaut test transmission

Andy,

 

getting even better... 56/704 symbol errors and 11.2 dB Eb/N0 for the 16:30 message.

 

I'm sure that Domenico will be able to pick up the signal in South Italy. Would anyone in TA-land want to attempt a decode tonight?