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

Varying time divisions in SL

Expand Messages
  • Chris Dillon
    Hi - I wonder if someone could help with this. In using SL to monitor the VLF spectrum I ve noticed that occasionally the dotted lines showing the hour
    Message 1 of 4 , Aug 9, 2012
    • 0 Attachment
      Hi - I wonder if someone could help with this. In using SL to monitor
      the VLF spectrum I've noticed that occasionally the dotted lines showing
      the hour divisions on the waterfall display are closer together than at
      other times. So, for example, in the attached jpeg, which shows the
      current Russian ZEVS transmissions on 82Hz, the dotted lines for 1300
      and 1400 UTC are closer than the others. It's as though SL has stopped
      for a few minutes and then resumed because part of the signal trace
      sometimes also seems to be missing. I've set up SL so that it is locked
      to GQD on 22.1kHz and set the primary time source to use the
      audio-sampling clock.
      I can see the same effect at the same times in monitoring the wideband
      VLF spectrum from 0-24kHz (e.g. see http;//g3wcd.bplaced.net). This
      happens about once every 24 hours (is this a clue?) - mostly, however,
      the time divisions are equally spaced.

      I'm using XP on a 1.8Ghz machine. Is this a well-known effect and is
      there anything I can do to avoid it? - or is it an artefact of the
      system timing? Would be grateful for any advice.

      Thanks

      Chris G3WCD
    • wolf_dl4yhf
      Hi Chris, An occasional glitch of the timescale may be caused by this: Even if the timing is based on the audio sampling clock, SL compares the time based on
      Message 2 of 4 , Aug 9, 2012
      • 0 Attachment
        Hi Chris,

        An occasional glitch of the timescale may be caused by this:
        Even if the timing is based on the audio sampling clock, SL compares the time based on the audio-sample counter with the PC's system time. If a certain difference is exceeded, it may 'warp' to the system time.
        But if, as you wrote, this happens every 24 hours, it's suspicious and the reason may be different. Do you have an internet-based time keeping service (like Dimension 4) running on your system which corrects the system time only once per day ?
        Also, are there any clues in SL's debugging display (in the menu: "View/Windows" .. "Error History") ?

        All the best, and congratulations for your excellent ZEVS reception,
           Wolf .



        Am 09.08.2012 18:10, schrieb Chris Dillon:
         

        Hi - I wonder if someone could help with this. In using SL to monitor
        the VLF spectrum I've noticed that occasionally the dotted lines showing
        the hour divisions on the waterfall display are closer together than at
        other times. So, for example, in the attached jpeg, which shows the
        current Russian ZEVS transmissions on 82Hz, the dotted lines for 1300
        and 1400 UTC are closer than the others. It's as though SL has stopped
        for a few minutes and then resumed because part of the signal trace
        sometimes also seems to be missing. I've set up SL so that it is locked
        to GQD on 22.1kHz and set the primary time source to use the
        audio-sampling clock.
        I can see the same effect at the same times in monitoring the wideband
        VLF spectrum from 0-24kHz (e.g. see http;//g3wcd.bplaced.net). This
        happens about once every 24 hours (is this a clue?) - mostly, however,
        the time divisions are equally spaced.

        I'm using XP on a 1.8Ghz machine. Is this a well-known effect and is
        there anything I can do to avoid it? - or is it an artefact of the
        system timing? Would be grateful for any advice.

        Thanks

        Chris G3WCD


      • Chris Dillon
        Hi Wolf - many thanks indeed for your comments. I am getting the occasional error in the debugging display of the form: Audio-sample based time (A) deviates
        Message 3 of 4 , Aug 10, 2012
        • 0 Attachment
          Hi Wolf - many thanks indeed for your comments. I am getting the occasional error in the debugging display of the form:

          "Audio-sample based time (A) deviates from the 'system time' (B)! A-B =-2.376547 sec"

          but the times the error messages occur seem not to correlate with the time slips on the waterfall display, which anyway seem to be several minutes long rather than a few seconds.I do keep the system synchronized with an internet time server (I use ntp2.npl.co.uk) but this corrects hourly. I'll turn this off and see if it makes a difference. I did experience an SL time slippage last night beteeen 2200 and 2300 UTC but nothing since.

          I'm currently running two instances of SL and save the display images to my Dropbox folder every few minutes to upload to my VLF grabber. Since I'd set SL to periodically save the images every 5 mins I did wonder whether there was an occasional resource conflict if both instances saved at the just same time. I've changed the interval times but it seems not to make a difference.

          I'll keep a check on it to see whether anything comes to light.

          Thanks again Wolf - SL is an excellent piece of software whose features I'm still exploring.

          73 Chris

          On 09/08/2012 19:20, wolf_dl4yhf wrote:
           

          Hi Chris,

          An occasional glitch of the timescale may be caused by this:
          Even if the timing is based on the audio sampling clock, SL compares the time based on the audio-sample counter with the PC's system time. If a certain difference is exceeded, it may 'warp' to the system time.
          But if, as you wrote, this happens every 24 hours, it's suspicious and the reason may be different. Do you have an internet-based time keeping service (like Dimension 4) running on your system which corrects the system time only once per day ?
          Also, are there any clues in SL's debugging display (in the menu: "View/Windows" .. "Error History") ?

          All the best, and congratulations for your excellent ZEVS reception,
             Wolf .



          Am 09.08.2012 18:10, schrieb Chris Dillon:
           

          Hi - I wonder if someone could help with this. In using SL to monitor
          the VLF spectrum I've noticed that occasionally the dotted lines showing
          the hour divisions on the waterfall display are closer together than at
          other times. So, for example, in the attached jpeg, which shows the
          current Russian ZEVS transmissions on 82Hz, the dotted lines for 1300
          and 1400 UTC are closer than the others. It's as though SL has stopped
          for a few minutes and then resumed because part of the signal trace
          sometimes also seems to be missing. I've set up SL so that it is locked
          to GQD on 22.1kHz and set the primary time source to use the
          audio-sampling clock.
          I can see the same effect at the same times in monitoring the wideband
          VLF spectrum from 0-24kHz (e.g. see http;//g3wcd.bplaced.net). This
          happens about once every 24 hours (is this a clue?) - mostly, however,
          the time divisions are equally spaced.

          I'm using XP on a 1.8Ghz machine. Is this a well-known effect and is
          there anything I can do to avoid it? - or is it an artefact of the
          system timing? Would be grateful for any advice.

          Thanks

          Chris G3WCD



        • Chris Dillon
          Hi Wolf - just for the record I ve tracked this problem down to my antivirus software. I m running Sophos EndPoint Security and Control V10. When this
          Message 4 of 4 , Aug 13, 2012
          • 0 Attachment
            Hi Wolf - just for the record I've tracked this problem down to my antivirus software. I'm running Sophos EndPoint Security and Control V10. When this downloads and installs updates, which can take some time, CPU usage rises to almost 100% and any other processes are slowed to a crawl. Apparently this is a known issue. There is 100% correlation between the updating events and the reduction of the timeline spacing in the SL display.

            73 Chris


            On 10/08/2012 11:13, Chris Dillon wrote:
             

            Hi Wolf - many thanks indeed for your comments. I am getting the occasional error in the debugging display of the form:

            "Audio-sample based time (A) deviates from the 'system time' (B)! A-B =-2.376547 sec"

            but the times the error messages occur seem not to correlate with the time slips on the waterfall display, which anyway seem to be several minutes long rather than a few seconds.I do keep the system synchronized with an internet time server (I use ntp2.npl.co.uk) but this corrects hourly. I'll turn this off and see if it makes a difference. I did experience an SL time slippage last night beteeen 2200 and 2300 UTC but nothing since.

            I'm currently running two instances of SL and save the display images to my Dropbox folder every few minutes to upload to my VLF grabber. Since I'd set SL to periodically save the images every 5 mins I did wonder whether there was an occasional resource conflict if both instances saved at the just same time. I've changed the interval times but it seems not to make a difference.

            I'll keep a check on it to see whether anything comes to light.

            Thanks again Wolf - SL is an excellent piece of software whose features I'm still exploring.

            73 Chris

            On 09/08/2012 19:20, wolf_dl4yhf wrote:
             

            Hi Chris,

            An occasional glitch of the timescale may be caused by this:
            Even if the timing is based on the audio sampling clock, SL compares the time based on the audio-sample counter with the PC's system time. If a certain difference is exceeded, it may 'warp' to the system time.
            But if, as you wrote, this happens every 24 hours, it's suspicious and the reason may be different. Do you have an internet-based time keeping service (like Dimension 4) running on your system which corrects the system time only once per day ?
            Also, are there any clues in SL's debugging display (in the menu: "View/Windows" .. "Error History") ?

            All the best, and congratulations for your excellent ZEVS reception,
               Wolf .



            Am 09.08.2012 18:10, schrieb Chris Dillon:
             

            Hi - I wonder if someone could help with this. In using SL to monitor
            the VLF spectrum I've noticed that occasionally the dotted lines showing
            the hour divisions on the waterfall display are closer together than at
            other times. So, for example, in the attached jpeg, which shows the
            current Russian ZEVS transmissions on 82Hz, the dotted lines for 1300
            and 1400 UTC are closer than the others. It's as though SL has stopped
            for a few minutes and then resumed because part of the signal trace
            sometimes also seems to be missing. I've set up SL so that it is locked
            to GQD on 22.1kHz and set the primary time source to use the
            audio-sampling clock.
            I can see the same effect at the same times in monitoring the wideband
            VLF spectrum from 0-24kHz (e.g. see http;//g3wcd.bplaced.net). This
            happens about once every 24 hours (is this a clue?) - mostly, however,
            the time divisions are equally spaced.

            I'm using XP on a 1.8Ghz machine. Is this a well-known effect and is
            there anything I can do to avoid it? - or is it an artefact of the
            system timing? Would be grateful for any advice.

            Thanks

            Chris G3WCD




          Your message has been successfully submitted and would be delivered to recipients shortly.