question about logs of PPS - NTP

This is a discussion on question about logs of PPS - NTP ; hi all, I'm a newbiew on NTP services, but anyway I'm trying to setup a Stratum1 server with the following hardware and software: - GPS Motorola Oncore M12 - Server: - CentOS 3.8 - Linux kernel 2.4.33.3 - ntp 4.2.2p4 ...

+ Reply to Thread
Results 1 to 6 of 6

Thread: question about logs of PPS

  1. question about logs of PPS

    hi all,

    I'm a newbiew on NTP services, but anyway I'm trying to setup a Stratum1
    server with the following hardware and software:

    - GPS Motorola Oncore M12
    - Server:
    - CentOS 3.8
    - Linux kernel 2.4.33.3
    - ntp 4.2.2p4
    - PPSKit 2.1.7

    I have configured the /etc/ntp.conf and /etc/ntp.oncore0 files and all
    seems good but the logs. In the /var/log/messages i see after several
    hours I started the NTPD daemon:

    May 9 13:29:07 khronos sshd(pam_unix)[1052]: session opened for user
    jtbpizac by (uid=0)
    May 9 13:29:07 khronos kernel: hardpps: PPSJITTER: jitter=330839,
    limit=13208
    May 9 13:29:13 khronos kernel: hardpps: new frequency -217491 >> 4 ==
    -13594
    May 9 13:29:13 khronos kernel: hardpps: new pps.freq=-13593 (add 37)
    May 9 13:29:20 khronos su(pam_unix)[1096]: session opened for user root
    by jtbpizac(uid=501)
    May 9 13:29:29 khronos kernel: hardpps: new frequency -217098 >> 4 ==
    -13569
    May 9 13:29:29 khronos kernel: hardpps: new pps.freq=-13568 (add 24)
    May 9 13:29:37 khronos kernel: hardpps: PPSJITTER: jitter=334175,
    limit=10732
    May 9 13:29:45 khronos kernel: hardpps: new frequency -216445 >> 4 ==
    -13528
    May 9 13:29:45 khronos kernel: hardpps: new pps.freq=-13527 (add 40)
    May 9 13:30:07 khronos kernel: hardpps: PPSJITTER: jitter=294411,
    limit=4864
    May 9 13:30:09 khronos kernel: hardpps: PPSJITTER: jitter=302386,
    limit=298056
    May 9 13:30:17 khronos kernel: hardpps: new frequency -434869 >> 5 ==
    -13590
    May 9 13:30:17 khronos kernel: hardpps: new pps.freq=-13589 (add -61)
    May 9 13:30:37 khronos kernel: hardpps: PPSJITTER: jitter=290707,
    limit=6260
    May 9 13:30:39 khronos kernel: hardpps: PPSJITTER: jitter=303013,
    limit=295400
    May 9 13:30:49 khronos kernel: hardpps: new frequency -434859 >> 5 ==
    -13590
    May 9 13:30:49 khronos kernel: hardpps: new pps.freq=-13589 (add 0)
    May 9 13:31:07 khronos kernel: hardpps: PPSJITTER: jitter=327983,
    limit=4232
    May 9 13:31:21 khronos kernel: hardpps: new frequency -433636 >> 5 ==
    -13552
    May 9 13:31:21 khronos kernel: hardpps: new pps.freq=-13551 (add 38)
    May 9 13:31:37 khronos kernel: hardpps: PPSJITTER: jitter=327881,
    limit=3332
    May 9 13:31:53 khronos kernel: hardpps: new frequency -434615 >> 5 ==
    -13582
    May 9 13:31:53 khronos kernel: hardpps: new pps.freq=-13581 (add -30)
    May 9 13:32:07 khronos kernel: hardpps: PPSJITTER: jitter=323754,
    limit=3620
    May 9 13:32:37 khronos kernel: hardpps: PPSJITTER: jitter=322598,
    limit=4844
    May 9 13:32:57 khronos kernel: hardpps: new frequency -870709 >> 6 ==
    -13605
    May 9 13:32:57 khronos kernel: hardpps: new pps.freq=-13604 (add -23)
    May 9 13:33:07 khronos kernel: hardpps: PPSJITTER: jitter=319882,
    limit=5772
    May 9 13:33:37 khronos kernel: hardpps: PPSJITTER: jitter=319118,
    limit=314

    And sometimes it appears:

    May 9 13:27:55 khronos kernel: hardpps: PPSJITTER: jitter=98335,
    limit=14772
    May 9 13:28:07 khronos kernel: hardpps: bad pulse, delta = -794324 ns
    May 9 13:28:08 khronos kernel: hardpps: bad pulse, delta = 769390 ns
    May 9 13:28:09 khronos kernel: hardpps: PPSJITTER: jitter=841424,
    limit=177076

    Is this the normal behavour or are the constants frequency changes a
    signal that something could be misconfigured?

    Thanks a lot,

    Christian

  2. Re: question about logs of PPS

    On 2007-05-09, Christian Pinedo Zamalloa
    wrote:
    > hi all,
    >
    > I'm a newbiew on NTP services, but anyway I'm trying to setup a Stratum1
    > server with the following hardware and software:
    >
    > - GPS Motorola Oncore M12
    > - Server:
    > - CentOS 3.8
    > - Linux kernel 2.4.33.3
    > - ntp 4.2.2p4
    > - PPSKit 2.1.7
    >


    After reading emails of the mailing list and searching the web i decided
    to migrate the ntp server to FreeBSD 6.1. I enabled PPS_SYNC in the
    kernel, compiled the kernel and the world, and configured again the ntp
    daemon. The result is ok, the server is running better and it has
    been announced as a stratum one public server in ntp.isc.org wiki.

    Anyway, i would like to ask to the users of a Motorola M12 receiver what
    should be the accuracy of the time server. Reading the logs and using
    ntpq tool i see that is from several useg to hundred useg. I though that
    it could reach nanoseg. Was I wrong?

    $ ntpdc -c kern
    pll offset: -0.000101122 s
    pll frequency: -21.032 ppm
    maximum error: 0.007368 s
    estimated error: 3e-06 s
    status: 2307 pll ppsfreq ppstime ppssignal ppsjitter nano
    pll time constant: 4
    precision: 1e-09 s
    frequency tolerance: 496 ppm
    pps frequency: -21.032 ppm
    pps stability: 0.645 ppm
    pps jitter: 1.797e-06 s
    calibration interval: 256 s
    calibration cycles: 420
    jitter exceeded: 3717
    stability exceeded: 0
    calibration errors: 27

    Best regards,

    --
    Christian Pinedo Zamalloa

  3. Re: question about logs of PPS

    Christian Pinedo Zamalloa wrote:
    > On 2007-05-09, Christian Pinedo Zamalloa
    > wrote:
    >
    >>hi all,
    >>
    >>I'm a newbiew on NTP services, but anyway I'm trying to setup a Stratum1
    >>server with the following hardware and software:
    >>
    >>- GPS Motorola Oncore M12
    >>- Server:
    >> - CentOS 3.8
    >> - Linux kernel 2.4.33.3
    >> - ntp 4.2.2p4
    >> - PPSKit 2.1.7
    >>

    >
    >
    > After reading emails of the mailing list and searching the web i decided
    > to migrate the ntp server to FreeBSD 6.1. I enabled PPS_SYNC in the
    > kernel, compiled the kernel and the world, and configured again the ntp
    > daemon. The result is ok, the server is running better and it has
    > been announced as a stratum one public server in ntp.isc.org wiki.
    >
    > Anyway, i would like to ask to the users of a Motorola M12 receiver what
    > should be the accuracy of the time server. Reading the logs and using
    > ntpq tool i see that is from several useg to hundred useg. I though that
    > it could reach nanoseg. Was I wrong?



    The Motorola M12+T is specified to deliver its Pulse Per Second output
    with the leading edge within 50 nanoseconds of the top of the second.
    Getting that into your computer is where the problem lies. Things like
    cable length can be compensated for. Interrupt latency is more
    difficult because it's difficult to determine exactly what value you
    need to compensate for. Depending on interrupt priorities and what else
    is going on, the latency may not even be constant.

    I'd say that you are probably within one microsecond of the correct time
    if you are using the Motorola M12+T as a reference clock.


  4. Re: question about logs of PPS

    On 2007-05-15, Richard B. Gilbert wrote:
    >
    > The Motorola M12+T is specified to deliver its Pulse Per Second output
    > with the leading edge within 50 nanoseconds of the top of the second.
    > Getting that into your computer is where the problem lies. Things like
    > cable length can be compensated for. Interrupt latency is more
    > difficult because it's difficult to determine exactly what value you
    > need to compensate for. Depending on interrupt priorities and what else
    > is going on, the latency may not even be constant.
    >
    > I'd say that you are probably within one microsecond of the correct time
    > if you are using the Motorola M12+T as a reference clock.
    >


    So my offset is very bad, i have peaks of milliseconds. I have done
    several graphics with gnuplot and i have put them available at:

    http://bipt106.bi.ehu.es/~jtbpizac/n...s.20070515.png
    http://bipt106.bi.ehu.es/~jtbpizac/n...s.20070516.png

    At 15 i restarted several times the NTP daemon but today (16) i haven't
    touch the server.

    I think that the graphics are very bad and this could be by various
    reasons:

    1. The GPS aerial is in the roof inside a plastic water bottle to protect from
    rain. I have seen that the bottle has too much adhesive tape that could
    difficult the GPS to find satellites. I will remove the excess.

    2. I don't know how my compains obtained the GPS coordinates. I will
    run WinOncore during a week to check the latitude, longitude and height.
    Could a bad value of these parameters produce a unstable GPS clock??

    3. The Motorola M12 board is near other electronic components: wifi
    access point, printer, PC, ... Could these components perturb the clock
    quality??

    Thanks for your comments and if i get a better behavior of the clock
    signal i will post to the group,

    --
    Christian Pinedo Zamalloa

  5. Re: question about logs of PPS

    Christian Pinedo Zamalloa wrote:
    > On 2007-05-15, Richard B. Gilbert wrote:
    >
    >>The Motorola M12+T is specified to deliver its Pulse Per Second output
    >>with the leading edge within 50 nanoseconds of the top of the second.
    >>Getting that into your computer is where the problem lies. Things like
    >>cable length can be compensated for. Interrupt latency is more
    >>difficult because it's difficult to determine exactly what value you
    >>need to compensate for. Depending on interrupt priorities and what else
    >>is going on, the latency may not even be constant.
    >>
    >>I'd say that you are probably within one microsecond of the correct time
    >>if you are using the Motorola M12+T as a reference clock.
    >>

    >
    >
    > So my offset is very bad, i have peaks of milliseconds. I have done
    > several graphics with gnuplot and i have put them available at:
    >
    > http://bipt106.bi.ehu.es/~jtbpizac/n...s.20070515.png
    > http://bipt106.bi.ehu.es/~jtbpizac/n...s.20070516.png
    >
    > At 15 i restarted several times the NTP daemon but today (16) i haven't
    > touch the server.
    >
    > I think that the graphics are very bad and this could be by various
    > reasons:
    >
    > 1. The GPS aerial is in the roof inside a plastic water bottle to protect from
    > rain. I have seen that the bottle has too much adhesive tape that could
    > difficult the GPS to find satellites. I will remove the excess.
    >
    > 2. I don't know how my compains obtained the GPS coordinates. I will
    > run WinOncore during a week to check the latitude, longitude and height.
    > Could a bad value of these parameters produce a unstable GPS clock??


    No. Incorrect values of latitude, longitude, and height can cause the
    clock to be inaccurate but will not affect stability. As long as the
    error in position is small, the error in time will be small. By running
    a 10,000 sample site survey, you should get a position correct to within
    about three hundred feet.

    >
    > 3. The Motorola M12 board is near other electronic components: wifi
    > access point, printer, PC, ... Could these components perturb the clock
    > quality??


    I don't believe so. Mine is within three feet of: two monitors, four
    computers, and a Cisco 1548M switch. It seems to be working just fine.


  6. Re: question about logs of PPS

    Christian Pinedo Zamalloa writes:

    > hi all,
    >
    > I'm a newbiew on NTP services, but anyway I'm trying to setup a Stratum1
    > server with the following hardware and software:
    >
    > - GPS Motorola Oncore M12
    > - Server:
    > - CentOS 3.8
    > - Linux kernel 2.4.33.3
    > - ntp 4.2.2p4
    > - PPSKit 2.1.7
    >
    > I have configured the /etc/ntp.conf and /etc/ntp.oncore0 files and all
    > seems good but the logs. In the /var/log/messages i see after several
    > hours I started the NTPD daemon:
    >
    > May 9 13:29:07 khronos sshd(pam_unix)[1052]: session opened for user
    > jtbpizac by (uid=0)


    These are debugging messages that you can enable or disable, depending on your
    personal taste.

    > May 9 13:29:07 khronos kernel: hardpps: PPSJITTER: jitter=330839,
    > limit=13208


    Jitter exceeded. Use the source, Luke ;-)

    > May 9 13:29:13 khronos kernel: hardpps: new frequency -217491 >> 4 ==
    > -13594
    > May 9 13:29:13 khronos kernel: hardpps: new pps.freq=-13593 (add 37)


    hardpps frequency update by 37 units (to roughly 13.593PPM)

    > May 9 13:29:20 khronos su(pam_unix)[1096]: session opened for user root
    > by jtbpizac(uid=501)
    > May 9 13:29:29 khronos kernel: hardpps: new frequency -217098 >> 4 ==
    > -13569
    > May 9 13:29:29 khronos kernel: hardpps: new pps.freq=-13568 (add 24)
    > May 9 13:29:37 khronos kernel: hardpps: PPSJITTER: jitter=334175,
    > limit=10732
    > May 9 13:29:45 khronos kernel: hardpps: new frequency -216445 >> 4 ==
    > -13528
    > May 9 13:29:45 khronos kernel: hardpps: new pps.freq=-13527 (add 40)
    > May 9 13:30:07 khronos kernel: hardpps: PPSJITTER: jitter=294411,
    > limit=4864
    > May 9 13:30:09 khronos kernel: hardpps: PPSJITTER: jitter=302386,
    > limit=298056
    > May 9 13:30:17 khronos kernel: hardpps: new frequency -434869 >> 5 ==
    > -13590


    shift constant updated from 4 to 5.

    > May 9 13:30:17 khronos kernel: hardpps: new pps.freq=-13589 (add -61)
    > May 9 13:30:37 khronos kernel: hardpps: PPSJITTER: jitter=290707,
    > limit=6260
    > May 9 13:30:39 khronos kernel: hardpps: PPSJITTER: jitter=303013,
    > limit=295400
    > May 9 13:30:49 khronos kernel: hardpps: new frequency -434859 >> 5 ==
    > -13590
    > May 9 13:30:49 khronos kernel: hardpps: new pps.freq=-13589 (add 0)
    > May 9 13:31:07 khronos kernel: hardpps: PPSJITTER: jitter=327983,
    > limit=4232
    > May 9 13:31:21 khronos kernel: hardpps: new frequency -433636 >> 5 ==
    > -13552
    > May 9 13:31:21 khronos kernel: hardpps: new pps.freq=-13551 (add 38)
    > May 9 13:31:37 khronos kernel: hardpps: PPSJITTER: jitter=327881,
    > limit=3332
    > May 9 13:31:53 khronos kernel: hardpps: new frequency -434615 >> 5 ==
    > -13582
    > May 9 13:31:53 khronos kernel: hardpps: new pps.freq=-13581 (add -30)
    > May 9 13:32:07 khronos kernel: hardpps: PPSJITTER: jitter=323754,
    > limit=3620
    > May 9 13:32:37 khronos kernel: hardpps: PPSJITTER: jitter=322598,
    > limit=4844
    > May 9 13:32:57 khronos kernel: hardpps: new frequency -870709 >> 6 ==
    > -13605


    .... to 6

    > May 9 13:32:57 khronos kernel: hardpps: new pps.freq=-13604 (add -23)
    > May 9 13:33:07 khronos kernel: hardpps: PPSJITTER: jitter=319882,
    > limit=5772
    > May 9 13:33:37 khronos kernel: hardpps: PPSJITTER: jitter=319118,
    > limit=314
    >
    > And sometimes it appears:
    >
    > May 9 13:27:55 khronos kernel: hardpps: PPSJITTER: jitter=98335,
    > limit=14772
    > May 9 13:28:07 khronos kernel: hardpps: bad pulse, delta = -794324 ns
    > May 9 13:28:08 khronos kernel: hardpps: bad pulse, delta = 769390 ns


    A noise spike (or kernel time problem)...

    > May 9 13:28:09 khronos kernel: hardpps: PPSJITTER: jitter=841424,
    > limit=177076
    >
    > Is this the normal behavour or are the constants frequency changes a
    > signal that something could be misconfigured?


    What do you guess hardpps does? I thought it would update the frequency drift
    to match the PPS signal ;-)

    If you've seen enough, just disable debugging messages.

    Regards,
    Ulrich

    >
    > Thanks a lot,
    >
    > Christian


+ Reply to Thread