Inexpensive OEM GPS units? - NTP

This is a discussion on Inexpensive OEM GPS units? - NTP ; Can anyone give me some suggestions on inexpensive ( units that support NMEA and 1PPS for use with NTP? Or any non-GPS hardware clock sources that anyone can suggest? I am already aware of the Garmin GPS 18 LVC, which ...

+ Reply to Thread
Page 1 of 2 1 2 LastLast
Results 1 to 20 of 24

Thread: Inexpensive OEM GPS units?

  1. Inexpensive OEM GPS units?

    Can anyone give me some suggestions on inexpensive (< $100) OEM GPS
    units that support NMEA and 1PPS for use with NTP? Or any non-GPS
    hardware clock sources that anyone can suggest?

    I am already aware of the Garmin GPS 18 LVC, which seems the most
    promising, but I'd like to know about other options if they exist.
    Doing random web searching online for a few hours has mostly found lots
    of more expensive alternatives--I'm just trying to find out what is out
    there in the same or better price range.

    Also, is there any place online that keeps an up-to-date list of this
    kind of thing? The NTP wiki seemed like a likely spot, but there wasn't
    much info about actual hardware.

    --
    Wesley J. Landaker
    OpenPGP FP: 4135 2A3B 4726 ACC5 9094 0097 F0A9 8A4C 4CD6 E3D2

  2. Re: Inexpensive OEM GPS units?

    On 2007-11-11, Wesley J. Landaker wrote:

    > Also, is there any place online that keeps an up-to-date list of
    > [inexpensive OEM GPS units] ? The NTP wiki seemed like a likely spot,
    > but there wasn't much info about actual hardware.


    The NTP wiki (i.e. http://support.ntp.org/support) is the logical place
    to start such a list. If for no other reason than the fact that everyone
    is able to contribute their little bit of information.

    Of course this requires that someone or, better yet, a group of people,
    take some interest in doing a bit of work.

    http://support.ntp.org/bin/view/Supp...xpensiveOemGps

    --
    Steve Kostecke
    NTP Public Services Project - http://support.ntp.org/

  3. Re: Inexpensive OEM GPS units?

    Steve Kostecke wrote:
    []
    > The NTP wiki (i.e. http://support.ntp.org/support) is the logical
    > place to start such a list. If for no other reason than the fact that
    > everyone is able to contribute their little bit of information.
    >
    > Of course this requires that someone or, better yet, a group of
    > people, take some interest in doing a bit of work.
    >
    > http://support.ntp.org/bin/view/Supp...xpensiveOemGps


    Steve,

    Thanks for setting that up.

    I added a pointer to my own system using the GPS18 on a spare PC with
    FreeBSD. As I'm a newcomer to FreeBSD it might help those with less
    experience. Could you please check I've edited correctly? I know there
    are more complex systems and perhaps single-board systems which should be
    added as well.

    Cheers,
    David



  4. Re: Inexpensive OEM GPS units?

    Wesley J. Landaker wrote:
    > Can anyone give me some suggestions on inexpensive (< $100) OEM GPS
    > units that support NMEA and 1PPS for use with NTP? Or any non-GPS
    > hardware clock sources that anyone can suggest?
    >
    > I am already aware of the Garmin GPS 18 LVC, which seems the most
    > promising, but I'd like to know about other options if they exist.
    > Doing random web searching online for a few hours has mostly found lots


    > of more expensive alternatives--I'm just trying to find out what is out


    > there in the same or better price range.
    >
    > Also, is there any place online that keeps an up-to-date list of this
    > kind of thing? The NTP wiki seemed like a likely spot, but there wasn't


    > much info about actual hardware.


    Having just considered the different options, for pretty much the same
    reasons, I ended up with plans for Garmin's GPS15L, an external antenna,
    and making a short cable to the main board of the server that is
    supposed to run the sync service. Although it doesn't (as such) has pps,
    but given that my OEM mainboard (AsRock) has a connector for "IR
    Header", I can create a connector that powers the chip from the
    mainboard, and connects the TX/RX wires there. Using standard NMEA
    protocol, and knowing the latency of the port itself, it can give me a
    fairly good, and simple to maintain, clock. It won't be perfect, but it
    won't perfectly break my bank account eighter.

    //Svein

    --
    Svein Skogen | svein@d80.iso100.no
    Solberg ěstli 9 | PGP Key: 0xE5E76831
    2020 Skedsmokorset | svein@jernhuset.no
    Norway | PGP Key: 0xCE96CE13
    ------------------------+-----------------------------
    msn messenger: | Mobile Phone: +47 907 03 575
    svein@jernhuset.no | RIPE handle: SS16503-RIPE

  5. Re: Inexpensive OEM GPS units?

    In article ,
    "Wesley J. Landaker" writes:
    >Can anyone give me some suggestions on inexpensive (< $100) OEM GPS
    >units that support NMEA and 1PPS for use with NTP? Or any non-GPS
    >hardware clock sources that anyone can suggest?
    >
    >I am already aware of the Garmin GPS 18 LVC, which seems the most
    >promising, but I'd like to know about other options if they exist.


    If you find anything better, or even close, please let us know.


    --
    These are my opinions, not necessarily my employer's. I hate spam.


  6. Re: Inexpensive OEM GPS units?

    In article <4736D026.90905@d80.iso100.no>,
    svein@d80.iso100.no (Svein Skogen) wrote:
    > Wesley J. Landaker wrote:
    > > Can anyone give me some suggestions on inexpensive (< $100) OEM GPS
    > > units that support NMEA and 1PPS for use with NTP? Or any non-GPS


    I'm not sure you will get much below USD 100, but there are other brands
    available. I've seen them advertised by small volume suppliers of
    embedded system/microcontroller type hardware. I have the feeling the
    Garmin unit is semi-consumer.

    > mainboard, and connects the TX/RX wires there. Using standard NMEA
    > protocol, and knowing the latency of the port itself, it can give me a


    Port latency is likely to be negligible compared with the NMEA output
    latency in the receiver. It may well be small compared with the jitter
    in the NMEA output. That's why PPS is important.


  7. Re: Inexpensive OEM GPS units?

    Wesley J. Landaker wrote:
    > Can anyone give me some suggestions on inexpensive (< $100) OEM GPS
    > units that support NMEA and 1PPS for use with NTP? Or any non-GPS
    > hardware clock sources that anyone can suggest?
    >
    > I am already aware of the Garmin GPS 18 LVC, which seems the most
    > promising, but I'd like to know about other options if they exist.
    > Doing random web searching online for a few hours has mostly found lots
    > of more expensive alternatives--I'm just trying to find out what is out
    > there in the same or better price range.
    >
    > Also, is there any place online that keeps an up-to-date list of this
    > kind of thing? The NTP wiki seemed like a likely spot, but there wasn't
    > much info about actual hardware.
    >


    There isn't much about actual hardware because there isn't much hardware
    available in that price range. AFAIK, the Garmin unit is the ONLY one
    available new in that price range.

    You can find used Motorola Oncore hardware on e-Bay. Motorola no longer
    makes or supports them. The one's I've seen have all been older 8
    channel units. Motorola sold the business to an outfit called something
    like "SIRF" that haven't heard of since!

    Other than GPS, you can buy or build HF receivers for the time signals
    broadcast by NIST (WWV at 2.5, 5.0, 10.0, 15.0, & 20 MHz) or the LF
    broadcast of WWVB at 60 KHz. CHU in Canada broadcasts a similar HF signal.

    If you are really desperate, you could try to duplicate Terry
    Pratchett's "Glass Clock of Bad Schushein" that synchronizes to the
    "tick of the universe". :-) (See "The Thief of Time" by Terry
    Pratchett; it's a fun read!)





  8. Re: Inexpensive OEM GPS units?

    Steve Kostecke wrote:
    > The NTP wiki (i.e. http://support.ntp.org/support) is the logical
    > place to start such a list. If for no other reason than the fact that
    > everyone is able to contribute their little bit of information.
    >
    > Of course this requires that someone or, better yet, a group of
    > people, take some interest in doing a bit of work.
    >
    > http://support.ntp.org/bin/view/Supp...xpensiveOemGps


    Good idea to start this. If I do end up finding or hearing of some other
    inexpensive alternatives, I'll add them there as well.

    Although from all the other replies (thanks everyone), it sounds like
    maybe there really just aren't tons of other cheap OEM options.

    --
    Wesley J. Landaker
    OpenPGP FP: 4135 2A3B 4726 ACC5 9094 0097 F0A9 8A4C 4CD6 E3D2

  9. Re: Inexpensive OEM GPS units?


    >Although from all the other replies (thanks everyone), it sounds like
    >maybe there really just aren't tons of other cheap OEM options.


    Your original request wanted a PPS signal. If you don't really
    need that, you have a lot more options. There are many low cost
    GPS units with USB connections. They seem to be intended for
    location rather than timing.

    The Garmin 18-USB works well but isn't yet supported by
    a driver in the ntpd source package.

    The Navibe unit doesn't work well (for timing). I'll get a graph
    out there one of these days.

    --
    These are my opinions, not necessarily my employer's. I hate spam.


  10. Re: Inexpensive OEM GPS units?

    It seems that Motorola handed over their Oncore product line to
    http://www.ilotus.com.sg/.
    http://www.synergy-gps.com/ are US agents for this company.

    Paul


  11. Re: Inexpensive OEM GPS units?

    On 2007-11-11, David J Taylor wrote:

    > I added a pointer to my own system using the GPS18 on a spare PC with
    > FreeBSD. As I'm a newcomer to FreeBSD it might help those with less
    > experience.


    Thanks for your contribution. It is much appreciated.

    > I know there are more complex systems and perhaps single-board systems
    > which should be added as well.


    What would be helpful is information about the observed performance of
    these inexpensive OEM GPS units. I've added a plot of my GPS-18LVC
    performance under less than optimum conditions and a snapshot of your
    daily MRTG plot for pixie to
    http://support.ntp.org/Support/Garmi...LVCPerformance

    Would you mind updating that topic so that is shows the version of
    FreeBSD you currently have in use?

    BTW: It would be interesting to run a loopstats file from pixie through
    my plotting tool and get a closer look at your GPS performance.

    --
    Steve Kostecke
    NTP Public Services Project - http://support.ntp.org/

  12. Re: Inexpensive OEM GPS units?

    Steve Kostecke wrote:
    []
    > What would be helpful is information about the observed performance of
    > these inexpensive OEM GPS units. I've added a plot of my GPS-18LVC
    > performance under less than optimum conditions and a snapshot of your
    > daily MRTG plot for pixie to
    > http://support.ntp.org/Support/Garmi...LVCPerformance
    >
    > Would you mind updating that topic so that is shows the version of
    > FreeBSD you currently have in use?
    >
    > BTW: It would be interesting to run a loopstats file from pixie
    > through my plotting tool and get a closer look at your GPS
    > performance.


    Steve,

    I've updated the topic to reflect the BSD version and the current antenna
    status.

    Where would I find the loopstats file in BSD? I'm not even sure that I've
    enabled them. I'm afraid that whatever knowledge I used to have of UNIX
    is now largely gone.

    Thanks,
    David



  13. Re: Inexpensive OEM GPS units?

    On 2007-11-12, David J Taylor wrote:

    > Steve Kostecke wrote:
    >
    >> BTW: It would be interesting to run a loopstats file from pixie
    >> through my plotting tool and get a closer look at your GPS
    >> performance.

    >
    > Where would I find the loopstats file in BSD?


    The OS is relevant only to the extent of the effect it has on the
    file path. The location of your statsdir is defined in ./etc/ntp.conf.

    > I'm not even sure that I've enabled them.


    If you have a block like this in your configuration file, you might see
    the *stats files in your statsdir ...

    | statsdir /var/log/ntpstats/
    | statistics loopstats peerstats clockstats
    | filegen loopstats file loopstats type day enable
    | filegen peerstats file peerstats type day enable
    | filegen clockstats file clockstats type day enable

    Keep in mind that the statsdir must be writable by the UID/GID that ntpd
    is running under.

    If your time server is only polling the GPS, there is probably little
    difference between the entries in the peerstats and loopstats files. If,
    on the other hand, your time server is polling multiple time sources it
    is best to plot the peerstats lines for the GPS.

    --
    Steve Kostecke
    NTP Public Services Project - http://support.ntp.org/

  14. Re: Inexpensive OEM GPS units?

    Steve Kostecke wrote:
    []
    > If your time server is only polling the GPS, there is probably little
    > difference between the entries in the peerstats and loopstats files.
    > If, on the other hand, your time server is polling multiple time
    > sources it is best to plot the peerstats lines for the GPS.


    Steve,

    Thanks for that, but it appears that I'm not collecting those statistics.

    Cheers,
    David



  15. Re: Inexpensive OEM GPS units?

    On 2007-11-13, David J Taylor wrote:

    > Steve Kostecke wrote:
    >
    >> If your time server is only polling the GPS, there is probably little
    >> difference between the entries in the peerstats and loopstats files.
    >> If, on the other hand, your time server is polling multiple time
    >> sources it is best to plot the peerstats lines for the GPS.

    >
    > Thanks for that, but it appears that I'm not collecting those
    > statistics.


    Too bad, that's really the only way to get an accurate view of ntpd
    performance.

    --
    Steve Kostecke
    NTP Public Services Project - http://support.ntp.org/

  16. Re: Inexpensive OEM GPS units?

    Steve Kostecke wrote:
    > On 2007-11-13, David J Taylor wrote:
    >
    >> Steve Kostecke wrote:
    >>
    >>> If your time server is only polling the GPS, there is probably
    >>> little difference between the entries in the peerstats and
    >>> loopstats files. If, on the other hand, your time server is polling
    >>> multiple time sources it is best to plot the peerstats lines for
    >>> the GPS.

    >>
    >> Thanks for that, but it appears that I'm not collecting those
    >> statistics.

    >
    > Too bad, that's really the only way to get an accurate view of ntpd
    > performance.


    Steve,

    If I had more time, I might try and do something, but for now within 20
    microseconds is more than good enough for me!

    Cheers,
    David



  17. Re: Inexpensive OEM GPS units?

    Once upon a time, Steve Kostecke said:
    >What would be helpful is information about the observed performance of
    >these inexpensive OEM GPS units. I've added a plot of my GPS-18LVC
    >performance under less than optimum conditions and a snapshot of your
    >daily MRTG plot for pixie to
    >http://support.ntp.org/Support/Garmi...LVCPerformance
    >
    >Would you mind updating that topic so that is shows the version of
    >FreeBSD you currently have in use?
    >
    >BTW: It would be interesting to run a loopstats file from pixie through
    >my plotting tool and get a closer look at your GPS performance.


    I'm trying to figure out how accurate my setup is, and I've enabled
    statistics gathering. What can I make of the numbers? How is the
    offset arrived at (e.g. how can the offset from "true" be calculated)?

    Is the tool used to generate the big plot in the middle available?

    I'm running an old SVeeSix (NMEA mode) into an FTDI USB to serial
    adapter (I am aware of the additional delays introduced by USB; that's
    part of what I'd be interested in quantifying). I'm using shmpps to get
    my PPS signal signal to ntpd (no Linux kernel PPSKIT which AFAIK doesn't
    do USB to serial anyway).

    --
    Chris Adams
    Systems and Network Administrator - HiWAAY Internet Services
    I don't speak for anybody but myself - that's enough trouble.

  18. Re: Inexpensive OEM GPS units?

    In article <13jkondp53g9sa5@corp.supernews.com>,
    cmadams@hiwaay.net (Chris Adams) wrote:

    > (e.g. how can the offset from "true" be calculated)?


    It can't. If it could, ntpd would compensate for it, with the result
    that it would be reduced to a level where it couldn't be calculated.

    Normally, finding any systematic offset would require the temporary use
    of a time source that was better than your normal time sources. As I think
    you are using USB, which is likely to have a significant impact, you
    would also need to bypass this, either by temporarily feeding the system
    with time through a route with lower latency, or by outputting the
    system's idea of the time through a low latency route and comparing it
    outside of the system.

    The best you can get from the figures is the variability around any
    systematic error.

    This does assume that the ntpd algorithm is optimal.

    --
    Chris Adams
    Systems and Network Administrator - HiWAAY Internet Services
    I don't speak for anybody but myself - that's enough trouble.

  19. Re: Inexpensive OEM GPS units?

    On 2007-11-14, Chris Adams wrote:

    > Once upon a time, Steve Kostecke said:
    > >http://support.ntp.org/Support/Garmi...LVCPerformance

    > Is the tool used to generate the big plot in the middle available?


    The plot is generated with Gnuplot http://www.gnuplot.info/ and my
    (unreleased) Makefile and plot file.

    Assuming that you can live without the nice labels on the X-axis, etc.,
    the plot command you give to Gnuplot is ...

    plot "loopstats.YYYYMMDD" using 2$5*1000) with impulses title "RMS Jitter", \
    "loopstats.YYYYMMDD" using 2$3*1000) with impulses title "Offset", \
    "loopstats.YYYYMMDD" using 2:4 with lines axes x1y2 title "Frequency"

    --
    Steve Kostecke
    NTP Public Services Project - http://support.ntp.org/

  20. Re: Inexpensive OEM GPS units?

    Once upon a time, David Woolley said:
    >Normally, finding any systematic offset would require the temporary use
    >of a time source that was better than your normal time sources. As I think
    >you are using USB, which is likely to have a significant impact, you
    >would also need to bypass this, either by temporarily feeding the system
    >with time through a route with lower latency, or by outputting the
    >system's idea of the time through a low latency route and comparing it
    >outside of the system.


    Okay, that's what I figured.

    I've only got one serial port (and it is a console), so I don't have any
    other way to connect but USB (also, I'm using a USB to serial adapter to
    pull +5V from the USB port). Everyone keeps saying that USB must be
    bad, but I was trying to see just how much impact USB has.

    --
    Chris Adams
    Systems and Network Administrator - HiWAAY Internet Services
    I don't speak for anybody but myself - that's enough trouble.

+ Reply to Thread
Page 1 of 2 1 2 LastLast