[ntp:questions] Accuracy of GPS satellite signal ?

Koos van den Hout koos at cs.uu.nl
Thu Feb 22 14:23:40 UTC 2007

Rob Kimberley <time.bandit at btinternet.com> wrote:

> 3) Oscillator - a good quality OCXO or most likely a Rubidium are needed, to 
> allow decent holdover in the unlikely event of loss of GPS. For your 500nS 
> mission statement, my personal recommendation would be Rubidium, as an OCXO 
> would quickly pass the 500nS figure if the unit had to flywheel.

I'd like to add:

An environment with a stable temperature. We recently had airco problems
with the server room where our ntp stuff is located. After some changes in
the airco settings the room temperature graph started to look like a nice
sawtooth. NTP accuracy graphs started to look like that same sawtooth.

The airco has been re-educated and the ntp servers will move to a rack
together where they can keep eachother at a reasonable temperature and
out of the airco winds.


Koos van den Hout, herding Suns and networks as koos at cs.uu.nl
+31-30-2534104              PGP keyid 0x27513781
http://idefix.net/~koos/            Use PGP when possible!
Visit my site about books with reviews http://www.virtualbookcase.com/

More information about the questions mailing list