[ntp:questions] best gps receiver for time synchronization

Dave Hart davehart at gmail.com
Mon Mar 9 21:40:33 UTC 2009

On Mar 9, 8:02 pm, jack <j.jack.w... at gmail.com> wrote:
> I am trying to sync my Windows box to an external GPS source. I
> currently have BU353, whose output is not very periodic. I read up on
> ntpd implementation that uses PPS signal but I don't even have an RS
> 232 port on my computer.

Google tells me that's a USB GPS mouse.

> My questions:
> 1) what's the best GPS antenna (and protocol) in terms of
> consistencies in its output?
> 2) what kind of accuracy can i expect?

As Richard noted, USB connected GPSes perform particularly poorly with
ntp.  Sadly, you can probably get better time from the internet than
from a local USB GPS refclock.  A 232-connected GPS without PPS would
be better, particularly if configured with a sentence whose length
doesn't vary like $GPGLL (use "mode 7" on server line to
enable all 3 sentences and configure the GPS to send only $GPGLL).
I'm guessing you could get sub-millisecond accuracy with that route,
after calibration.  For PPS, the Garmin GPS 18x LVC is an affordable
option, though it "only" claims one microsecond accuracy of the PPS
signal, not tens of nanoseconds like timing receivers.  With ntpd on
windows you can't sync close enough for that to matter so far.  With
PPS, you should be able to stay within +/- 150 usec of UTC over 24
hours on Windows in my recent experience.  Short-term jitter reported
by ntpq -c peers is lower, 5-10 usec.  Your mileage may vary,  If the
temperature is consistent day and night, for example, ntp will be able
to keep offsets lower than if not.  Systems with 10 msec clock period
are currently worse off, but 15..6 msec is much more common.

Dave Hart

More information about the questions mailing list