[Aavso-sid-list] VLF/SID monitoring

[Jean-Pierre Godet]

On Mon, 4 Aug 2008, Siddhartha Jain wrote:

> ...
> Questions:
> 1. Is there a way to benchmark/test my setup specifically wrt SpecLab
> including talking to a local mentor?
> ...

   Hello friend !

   If you are WV6U, I gess that a local mentoring will be difficult. I am 
using SpecLab, but with a fairly different hardware. Anyway, I give you 
the config used here more than three years for SID minitoring after one 
year of adjustments and tests (only important parameters, you will have 
maybe to adjust them a little bit to agree with your harware sensitivity) 
:

1) SpecLab Configuration and Display Control :
   a) FFT properties, frequ. resolution :
     - Decimate input by (divisor) : 1 ;
     - FFT input size (length) : 8192 ;
     - FFT window function : Hanning ;
     - FFT type : real FFT ;
     - FFT Output Type : Normal ;
     - FFT Output Unit : dB ;
     - FFT Output Internal Average : 1.
   b) I/O :
     - Sample Rate : 48000 ;
     - Input Sample Rate Divisor : 1 ;
     - Bits/sample : 16
   c) Spectrum all is as default, but :
     - One Pixel per FFT bins : Yes ;
     - Waterfall Time Grid : Enabled ; Every 15 min ; Small ticks ;
     - Waterfall Scroll Interval : 2000 (ms) ;
     - Displayed Amplitude Range : -100 ==> -50 dB full scale.

   You have to modify the System Setting too (adjust Timezone)...

2) Watch List and plot Window :
   a) Watch List :
     - for each transmitter monitored, I open a 50 Hz window around the mid 
frequency. Ex : HWU mid freq. is 22.6 kHz, Expression will be : 
(peak_a(22575,22625))+100
Why "+100" ? To play only with positive numbers. It may be 80 for you, or 
another figure, you choose it to have the background noise at 0 dB or a 
little higher.
     - Scale Min : 10 (dB) ;
     - Scale Max : 50 (dB) (to be adjusted to see all the graph in your 
window) ;
   b) Layout : as you want ;
   c) Horizontal :
     - Scroll Rate : 60.0 Sec/Step ;
     - Small markers interv. : 15*60
     - Large makers interv. : 60*60
   d) Channels :
     - style : mixed ;
     - Curve : Average Value.
   e) Memory :
     - Samples : 2500 (the volume of data in memory ==> here 2500 minutes);
     -Channels : 4 (for me, for you the number of stations monitored) ;
   f) Export : as you want. For me : "TAB character as column separator, 
decimal point".

   All the rest is more or less cosmetic. I write my own Waterfall Color 
Palette, something like black ==> blue ==> green  ==> white because I like 
cold colours.
  With this config, you have one acquisition each second, the mean value of 
the last twenty acquisitions is used to built each step of the waterfall 
(scroll interval 20 sec.), the mean value of the last three mean values is 
the figure used to build each step of the plot (plot scroll interval is 
one mn) and to build the data file.

   The Linux utility gnuplot is used to build a graph from the raw data 
file. If you want to have a look to the result of all the stuff, the last 
gnuplot graph (the plot of each previous day) is displayed at :
http://jpmere.online.fr/VLF.htm
   For an exemple of the data structure, the data files of each day of the 
last three years are available at :
http://jpmere.online.fr/incoming/f5yg/
   Looks absolutly like the GOES data files, no ?

   Hope that helps. If you want, I can send a post here with the whole 
config file (you will have to copy it in the root directory of SpecLab, 
then to call it with "Quick Setting") as it is harmless genuine ASCII; to 
make matters easier and to work with the beautiful (?) non-aggressive 
green waterfall of my own ; but only if you, my SID colleagues, are not 
tired of too much chattering about SpecLab.

   Best regards (and more active sun !!!)

   Jean-Pierre

--
Powered by Linux (Slackware 10.0 - kernel 2.4.26)