[time-nuts] GPS Locked and Unlocked Performance Comparison
Tom Van Baak
tvb at LeapSecond.com
Tue Feb 12 19:54:46 EST 2008
Hi Said,
Yes, it was quite a bit of work. Thanks for noticing.
My goal is to extend the plots out well past a day but as you
know to get good statistics out there takes a week or two of
data for each run. Or I can switch measurement systems
and try to do several in parallel.
Theoretically just 3 or 4 days of data is sufficient to compute
the ADEV(tau 1 day) but the error bars are too wide that way.
When comparing different GPSDO, one needs to keep the
error bars pretty small in order to be fair. That's why the plot
stops at tau 40k.
Anyway, once I collect enough data, yes, adding a holdover
test is the next thing. Several of these GPSDO have such low
drift as it is that it could take a week to accurately measure it.
/tvb
----- Original Message -----
From: SAIDJACK at aol.com
To: tvb at leapsecond.com ; time-nuts at febo.com
Sent: Tuesday, February 12, 2008 4:13 PM
Subject: Re: [time-nuts] GPS Locked and Unlocked Performance Comparison
Hello Tom,
excellent web-page! That looks like a lot of work was done.
Would you have data for 24 hour hold-over performance (86400s ADEV) for these units?
thanks,
bye,
Said
In a message dated 2/11/2008 21:17:49 Pacific Standard Time, tvb at LeapSecond.com writes:
Sifting through old data, in addition to recent measurements
made in the past few months, I have 4 very interesting plots
of GPSDO performance.
The goal was to see real-life plots of disciplining in action by
contrasting free (unlocked) vs. GPS-locked performance as
compared with my 10 MHz house reference. The four GPSDO
measured so far are: Z3801A, Fury, Thunderbolt, and Miller.
http://www.leapsecond.com/pages/gpsdo/
Comments & questions welcome.
I hope to measure quite a few more this year. Contact me
offline if you want your favorite one included in the list. Note
that I'm not necessarily looking for the best GPSDO; instead,
for this page at least, the more variety the better.
/tvb
More information about the time-nuts
mailing list