[ntp:questions] Meinberg NTP monitor, silly question

David J Taylor david-taylor at blueyonder.delete-this-bit.and-this-part.co.uk.invalid
Sun Dec 20 07:06:00 UTC 2009


> Hi All..
[]
> Re the status column in the 'NTP Status' tab.
[]
> But what does 'o' mean?

"o" - appears to mean - "synced to the "atom" (PPS) driver", which is 
filtering the clock from the "*" source.

[]
> The "Application Log" shows:-
> Using user-mode PPS timestamp for GPS_NMEA(2)  Not that I think it'll
> make any significant difference, but I thought I'd set all this up so
> that the PPS thing would run in Kernel mode, but obviously not.

Requirements:

- serialpps.sys installed and in use, perhaps Computer Management, Device 
Manager might show what the driver is for that COM port?

- serialpps-ppsapi-provider.dll installed

- environment variable PPSAPI_DLLS set to point to the full path of the 
DLL, in the system environment variables, not the per-user variables.  For 
example:

  PPSAPI_DLLS=c:\serialpps\serialpps-ppsapi-provider.dll

As you say, it doesn't seem to improve the offset jitter a lot, but I know 
that having gone to the bother of installing it, it would be nice to see 
it working!

> What I've got is this in the NTP status tab page...
>
> State Remote Refid Stratum Type When Poll
> Reach Delay Offset Jitter
>
> * 127.127.20.2 GPS 0 Local clock 9 16 377
> 0.000 0.089 0.006
>
> o PPS(2) PPS 0 Local clock 8 16 377
> 0.000 0.124 0.004
[]
> Yes, the mailer wrapped it all, but hope someone can understand it and
> say if it's good/bad or ???

About what I see on my own Windows/GPS/PPS/atom system, so I would say its 
working correctly.


> Preliminary results are a *HUGE* improvement over using Wan based
> servers, but the system is still settling, and the GPS RX is still
> indoors! (high up and seemingly happy, but not as good a sky view as
> it'll be when eventualy sited outside.)
>
> Best Regards.
>
> Dave Baxter.

Here, the GPS 18x LVC working on the top-floor of a building is about as 
good as a GPX 18 LVC on the roof, at least for timekeeping.

73,
David 




More information about the questions mailing list