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

Re: [dxatlas] GPS based time server for Faros

Expand Messages
  • Peter
    G Day, My enquiries have led to the following information so far. Feel free to contradict anything written here - I m very low down on the learning curve. A
    Message 1 of 5 , Nov 18, 2009
      G'Day,

      My enquiries have led to the following information so far. Feel free to
      contradict anything written here - I'm very low down on the learning curve.

      A GPS with PPS - plus per second - output is the only type that will
      deliver a satisfactory result, ultimately, to be used by Faros. The GPS
      PPS wire is typically connected to "data carrier detect", (DCD) pin
      (DB-9 pin 1, DB-25 pin 8) - on a serial port, but can also be interfaced
      to the parallel port - ACK pin (DB25 pin 10). The devices are commonly
      interfaced to Linux, although they can be interfaced to Windows.

      These methods of interfacing the GPS PPS signal are special things -
      they take advantage of the standard PC design and make use of the chip
      in the PC interfaces - serial or parallel - to get the signal into the
      PC. They are NOT standard serial or parallel devices. Hence the need for
      custom software drivers to get the "information" from the PPS line of
      the GPS to where it can be useful. It is not as easy as it sounds. To
      quote the NTP material:

      > This driver furnishes an interface for the pulse-per-second (PPS)
      > signal produced by a cesium clock, radio clock or related devices. It
      > can be used to augment the serial timecode generated by a GPS
      > receiver, for example. It can be used to remove accumulated jitter and
      > re-time a secondary server when synchronized to a primary server over
      > a congested, wide-area network and before redistributing the time to
      > local clients. The driver includes extensive signal sanity checks and
      > grooming algorithms. A range gate and frequency discriminator reject
      > noise and signals with incorrect frequency. A multiple-stage median
      > filter rejects jitter due to hardware interrupt and operating system
      > latencies. A trimmed-mean algorithm determines the best time samples.
      > With typical workstations and processing loads, the incidental jitter
      > can be reduced to a few microseconds.
      >

      From that I note - a reasonably stable source of time is needed to get
      started; programs running on the PC can affect time; the GPS clock won't
      always generate a pulse.

      So, it is well beyond the field of us normal amateur radio operators to
      craft our own drivers for the PPS interface. We must use the drivers
      contained in the official, or patched, releases of NTP. Thank goodness
      they are there and we can use them.

      So, without going into the extreme details ...

      One attaches the GPS PPS line to a serial port. One also attaches the
      RS232 serial data lines of the GPS to a serial port of the PC - it can
      be the same one, but a separate port to avoid noise on the CDC line is
      advisable if extreme accuracy is needed - Faros does not need extreme
      accuracy. One installs the operating system (FreeBSD), patching the
      kernel (linux) or serial port drivers (Windows), if necessary. One
      installs the NTP software and provides the necessary configuration
      parameters for the NTP device drivers - a PPS interface driver, and a
      "time source" driver which can be the GPS itself, or another time server
      anywhere. Finally, one tells Faros the IP of our local NTP time server.

      When the power is applied, and it all starts to come alive, time is
      obtained from the "time source". The time will usually be close to the
      correct time, say within hundreds of milliseconds. If the time is within
      a small margin of the PPS "tick" from the GPS PPS line, our NTP software
      installation will sync its time to the "tick" of the GPS clock, and make
      it available to others. Faros will ultimately poll port 123 of the IP
      address of our local server and use the time received in its calculation
      of "Faros time". With luck, raw courage, brute strength, and a bigger
      screw driver - and the fact that each and every beacon uses a very
      similar technique, using a similar PPS GPS, Faros time will be very very
      similar to that used by the NXDF beacons, and Faros will hear each and
      every transmission possible.

      Information can be found at: www.ntp.org - Reference Clock Drivers,
      Driver Types 20 and 22.

      Peter VK4IU


      vk4iu wrote:
      >
      >
      > G'Day,
      >
      > My Internet connection is proving unreliable for Faros time keeping.
      >
      > Does anybody in the group have any practical experience in
      > implementing a local GPS based NTP time server for Faros time keeping?
      >
      > Peter VK4IU
      >
      >


      [Non-text portions of this message have been removed]
    • Dave
      Is anyone in the group using an IC-725 with Faros in Multiband mode? I m having trouble getting radio to change freqs. I m using an serial interface into the
      Message 2 of 5 , May 10, 2010
        Is anyone in the group using an IC-725 with Faros in Multiband mode? I'm
        having trouble getting radio to change freqs. I'm using an serial interface
        into the ACC1 radio jack; the unit is a C-7 interface from xggcomms.com

        Any one having success with IC-725 and Faros in multiband operation?

        Thanks,

        Dave Mynatt
        /====================================/
        http://sciencecenter.no-ip.org
        SID ID #S-0258
        APRS Tier II Server @ http://pueblo.aprs2.net:14501
        10 meter CW Beacon @ 28.2345 Mhz
        RMS Server: 144.950 Mhz // KA0SWT-10
        Echolink Server Node: 473082 @147.480Mhz Simplex KA0SWT-L
        LAT: (38.15.49) 38.26349"N LON: (104.36.48) -104.613297"W Alt: 1460m
        DM78qg // KA0SWT
        /++++++++++++++++++++++++++++++++/?
        No virus found in this outgoing message.
        Checked by AVG - www.avg.com
        Version: 9.0.819 / Virus Database: 271.1.1/2865 - Release Date: 05/10/10
        12:26:00
      • Dave Baxter
        Dave. Hi again. Have you managed to control it from the OmniRig client, not with Faros running, just that client program (that invokes the OmniRig system
        Message 3 of 5 , May 11, 2010
          Dave.

          Hi again.

          Have you managed to control it from the OmniRig client, not with Faros
          running, just that client program (that invokes the OmniRig system
          behind the scenes.)

          I also see there is a IC725 and IC725a in the list of radios OmniRig
          knows about, I guess you've tried both 'drivers'?

          If none of that works, does anything else successfully control the radio
          via that serial adapter, nothing to do with OmniRig or Faros. Ham
          Radio Deluxe (free) for example?

          As it's an Icom that uses a CI-V bus AFIK, is the radio's "address"
          setting correct for OmniRig?

          Looking in the two driver files, I suspect that the 725 file uses
          address 28, and the 725a file uses address 3E, that's about the only
          difference I can see. I don't specifically know the 725 it has to be
          said, and I can't remember much detail about the CI-V bus protocol at
          the moment either..

          As always, one step at a time.

          73.

          Dave 'wbx
        • Dave
          Dave- Got it. The baud rate is supposed to be 1200. That was it; changed baud rate, verified interface plugs, downloaded the OmniRig Client app and it all
          Message 4 of 5 , May 11, 2010
            Dave- Got it. The baud rate is supposed to be 1200. That was it; changed
            baud rate, verified interface plugs, downloaded the OmniRig Client app and
            it all just fell together.

            Working fine now, albeit no signals. I think there's another issue as it
            seems crazy that I see signals when using CW at 6k bandwidth but nothing at
            3k. Weird.

            BUT, working now; switching freqs and doing fine. My web page at:
            http://173.14.27.81/xampp/faros/

            Thanks all for the help,

            Dave Mynatt
            /====================================/
            http://sciencecenter.no-ip.org
            SID ID #S-0258
            APRS Tier II Server @ http://pueblo.aprs2.net:14501
            10 meter CW Beacon @ 28.2345 Mhz
            RMS Server: 144.950 Mhz // KA0SWT-10
            Echolink Server Node: 473082 @147.480Mhz Simplex KA0SWT-L
            LAT: (38.15.49) 38.26349"N LON: (104.36.48) -104.613297"W Alt: 1460m
            DM78qg // KA0SWT
            /++++++++++++++++++++++++++++++++/





            -----Original Message-----
            From: dxatlas_group@yahoogroups.com [mailto:dxatlas_group@yahoogroups.com]
            On Behalf Of Dave Baxter
            Sent: Tuesday, May 11, 2010 07:53
            To: dxatlas_group@yahoogroups.com
            Subject: [dxatlas] Re:FAROS and IC-725 Radio

            Dave.

            Hi again.

            Have you managed to control it from the OmniRig client, not with Faros
            running, just that client program (that invokes the OmniRig system behind
            the scenes.)

            I also see there is a IC725 and IC725a in the list of radios OmniRig knows
            about, I guess you've tried both 'drivers'?

            If none of that works, does anything else successfully control the radio
            via that serial adapter, nothing to do with OmniRig or Faros. Ham
            Radio Deluxe (free) for example?

            As it's an Icom that uses a CI-V bus AFIK, is the radio's "address"
            setting correct for OmniRig?

            Looking in the two driver files, I suspect that the 725 file uses address
            28, and the 725a file uses address 3E, that's about the only difference I
            can see. I don't specifically know the 725 it has to be said, and I can't
            remember much detail about the CI-V bus protocol at the moment either..

            As always, one step at a time.

            73.

            Dave 'wbx


            ------------------------------------

            Yahoo! Groups Links




            No virus found in this incoming message.
            Checked by AVG - www.avg.com
            Version: 9.0.819 / Virus Database: 271.1.1/2866 - Release Date: 05/10/10
            12:26:00

            No virus found in this outgoing message.
            Checked by AVG - www.avg.com
            Version: 9.0.819 / Virus Database: 271.1.1/2867 - Release Date: 05/11/10
            00:26:00
          Your message has been successfully submitted and would be delivered to recipients shortly.