Loading ...
Sorry, an error occurred while loading the content.
 

Re: [SpectrumLabUsers] LR Lightning Radar DSP and GPS improvement.

Expand Messages
  • wolf_dl4yhf
    ... Yes. This has nothing to do with the final accuracy. It s just a pre-requisite for the algorithm to find the UTC second to which a 1-second pulse belongs;
    Message 1 of 11 , Jul 20, 2013
      Am 20.07.2013 15:22, schrieb Gerry G.:
       
      Hi Wolf,

      I see all your saying.  I think we must be talking about different things on the 500 Usec.  I was only referring to this sentence in your manual where its talking about using the GPS without NMEA.

         Quote:  " Method B is only possible if the PC's system time is accurately set (difference between the PC's system time and the GPS second way below 500 milliseconds) !" I think that just means the digital clock readout stuff. Not the true clock phase.

      Yes. This has nothing to do with the final accuracy. It's just a pre-requisite for the algorithm to find the UTC second to which a 1-second pulse belongs; if there is no additional NMEA input from the GPS into the soundcard.

      We are using Frank Kooimans program which uses no GPS data at all or sync except from the cpu clock and sound card clock.  What he calls Real time and non real time (ie sound card).  He then phases the stations together if they are not too far apart drift and frequency wise.  

      I take it from your discussion that this LRA effort that is using the parallel port input for GPS then cant work?  Due to all the things you mentioned in the last discussion.   Or maybe again we're talking about two different things?
      Yes possibly. I don't see how the LRA hardware can actually remove the sampling errors from a soundcard. Its job is to minimize errors in the PC's system time (at least I think that's what it is supposed to do).

       Here is a link to the LRA sync screen where it shows 0 Usec GPS PPS Vs. 'processor' clock error.  If I understand it right this means the station cpu clock and GPS are now in sync with 0 difference according to that screen anyway.. This sync method is using the port method that doesn't work, so between the LRA sync and what your telling me I'm becoming a little confused to say the least.  It would seem that either SL could do the same as LRA is doing on the port, or LRA is not doing what it shows?
      SL could do, with an awful lot of work, the same job (synchronizing the PC's system time) but I still don't see the point how this could be used to provide a highly accurate timestamp for a block of samples *FROM THE SOUNDCARD*.

      Ok, time to go outside and do something else now.
      Have a nice weekend,
         Wolf.

    • Graham Mcleod
      Hi Wolf, Er, the speed of light is 300,000 Km/sec, not 300 Km/sec. But this * 500 us IS 150 Km, OK. Gra. Hi Wolf, Er, the speed of light is 300,000 Km/sec,
      Message 2 of 11 , Jul 21, 2013
        Hi Wolf,
        Er, the speed of light is 300,000 Km/sec, not 300 Km/sec.
        But this * 500 us IS 150 Km, OK. Gra.
      Your message has been successfully submitted and would be delivered to recipients shortly.