[time-nuts] Low-long-term-drift clock for board levelintegration?

[SAIDJACK at aol.com]

In a message dated 2/19/2012 21:21:35 Pacific Standard Time, wb6bnq at cox.net 
 writes:

Doing a  few fixes for 30 minutes will, under best conditions, get you 
somewhere on a  circumference around your location with a radius of 15 meters 
(50 feet).   For GPS to get a useful coordinate result with meaningful data 
will take  longer than 30 minutes or so.  Typically, you would want to do a 48 
hour  “survey” of your position to try to achieve a 3 to 5 meter 
resolution.   However, that only gives you an idea of where the GPS antenna  was
Bill, that may have been true for some older commercial GPS receivers, but  
not for the newer high performance receivers.

We did "flight testing" of our FireFly-IIA unit fed from a GPS  simulator, 
and the results are:
 
  better than 0.8 meters horizontal accuarcy rms
  better than 2.1 meters vertical accuracy rms
 
This was then verified in a Turboprop airplane. This was in the USA with  
WAAS being active.
 
30 minutes should be more than enough to get a position with the  above 
accuracy at very high confidence levels.
 
bye,
Said