[time-nuts] High-end GPSDO's

Attila Kinali attila at kinali.ch
Sat Aug 18 14:53:19 EDT 2018


On Thu, 16 Aug 2018 11:40:57 -0700
Ralph Devoe <rgdevoe at gmail.com> wrote:

>      I've seen several spec sheets on high end GPSDO's that seem to have
> performance approaching a low-noise cesium standard, but only cost $3-6$K
> new. One is the SRS FS740 which appears to combine a GPSDO with the
> interpolator of the SR620 counter. This gets down below 10(-13) in one day
> and drops below 10(-14) for longer times. 

At longer averaging times, GPS derived time is better than any one
frequency standard you can own. Why? Because GPS time is derived
from UTC(USNO) which in turn is steered using Circular-T.

But that alone does not mean anything. Getting to a low uncertainty would be
even possible if you would take your watch and run to Greenwich once a day.
It just would take longer to get down to 1e-13. Just compare the FS740 data
with [1]. While both reach 1e-13 eventually, the 5071 does it at taus
of ~1000s (roughly 20min) while the FS740 does so at ~100ks (roughly 27h).
It looks even worse at low averaging times. The 5071 crosses the 1e-12 line
somewhere around 50-70s, while the FS740 does that well beyond 1000s.

And keep in mind, this all depends on a _very_ good GPS antenna position.
If your antenna is in an urban area, or you have lots of obstructions
of nearby, then the ADEV will detoriate.

Just to avoid confusion: I don't want to bash the FS740, not at all.
It's a very well designed device with lots of ingenious solutions for
small details (see also my quick review of it at [2]), but it's still
just a GPSDO and has to live with the limitations of the GPS/GNSS 
system.

Another side note: The interpolator in the FS740 is simpler than the
one in the SR620 and thus less precise. The obvious reason for this is,
that they do not need the <20ps precision of the SR620 for something
like GPS. Even 100ps is more than good enough.

			Attila Kinali


[1] "Architecture, Design Aspectas and Performance of a New Cesium
Beam Frequency STandard", 
http://leapsecond.com/corby/5071comb.pdf

[2] http://lists.febo.com/pipermail/time-nuts_lists.febo.com/2017-November/089966.html

-- 
<JaberWorky>	The bad part of Zurich is where the degenerates
                throw DARK chocolate at you.



More information about the time-nuts mailing list