[ntp:questions] Questions about joining pool.ntp.org

Chris Albertson albertson.chris at gmail.com
Tue Aug 30 22:18:11 UTC 2011


On Tue, Aug 30, 2011 at 8:22 AM, unruh <unruh at wormhole.physics.ubc.ca>wrote:

> On 2011-08-30, Chris Albertson <albertson.chris at gmail.com> wrote:
> >> The time stuff I understand, even the radio wave propogation stuff,
> >> though obviously not as well as others here.
> >> I'm not sure what "cable length", "speed of light delay", or "velocity
> >> factor" is all about.
> >>
> >
> > What I was getting at was that that better GPS receivers are so accurate
> > with their pulse per second that you have to carefully account for the
> delay
> > in a short cable.  In other words the delay due to speed of light down a
> > wire that is four feet long is greater then the error in the PPS.   I
>
> Well, no. That is just 4ns, and the jitter in the gps pps of most
> receivers is more than that. (and the ability to get that signal into a
> computer in order to affect the timing of that computer is much much
> longer than that-- of the order of 1us)


Some low cost GPSes really are that good, better than 4nS.  For example this
one sells for about $60
 http://www.synergy-gps.com/images/stories/pdf/m12mt_brochure.pdf

It's overkill for an NTP server  but there are other uses of GPS that can
make use of  nanosecond level error.  Mostly the accurate PPS lets you
calibrate test equipment or a local frequency standard.

Yes you can't get this into a computer without introducing a large error but
people doing work at this level are using specialized hardware.  Maybe with
something like this
http://www.ebay.com/itm/HP-5335A-1-3-GHz-Universal-Counter-Opts-010-030-/120760529105#ht_2039wt_914


Chris Albertson
Redondo Beach, California



More information about the questions mailing list