[time-nuts] fleabay cesium

WB6BNQ wb6bnq at cox.net
Sun May 10 00:54:46 UTC 2009



Lets try to put this into perspective !   Using the specs from the OLD hp 5061A
which says it is intrinsically accurate to 1 part in 10^-11 means the following
assuming it is exactly in error by a stable value of 1 x 10^-11 from the agreed
upon central value :

1 x 10^-11 = 10 picoseconds offset

If all you are doing is trying to keep your clock accurate to the one second
level means it will take the following amount of time to accumulate a one second
error in 10 picosecond increments:

100,000,000,000       seconds or
1,666,666,666.6667   minutes or
27,777,777.7778       hours    or
1,157,407,40741       days     or
3,170.97919838        years

Plus or minus a little here and there as obviously the minutes, hours, days and
years are not even numbers.

So the extremely small drift may not even be noticable depending upon what you
are doing.

Bill....WB6BNQ



Hal Murray wrote:

> > I have a few GPSDO's (Trimble T-Bolt, Z3801A, TrueTime & HB units).
> > How much better can the Cesium be than any of those ??
>
> Here is another viewpoint...  (I don't own a Cesium (but I'm looking) so my
> opinions may not be relevant.)
>
> What are you after?  What is important to you?
>
> Under the covers, a "Cesium" is just a CesiumDO with the PLL parameters tuned
> to take advantage of what the Cesium tube can deliver.  The basic time
> constant is sub-second vs many-seconds for a GPSDO.
>
> If you are nutty enough, the GPSDO numbers don't really match what a Cesium
> can deliver in the medium term time scale, say 1000-10000 seconds.  I'm
> pretty sure I've seen graphs of GPS time shifting slightly from night to day
> due to atmospheric changes.  A Cesium won't do that.
>
> On the other hand, a GPSDO won't have any long term drift.  It's locked to
> the best time we can get.  If you wait long enough a Cesium will drift.  Not
> very fast, but it will drift.
>
> --
> These are my opinions, not necessarily my employer's.  I hate spam.
>
> _______________________________________________
> time-nuts mailing list -- time-nuts at febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.




More information about the time-nuts mailing list