[time-nuts] Basic question re adev measurements

John Miles jmiles at pop.net
Mon Mar 26 22:00:15 UTC 2012


It's a bit worse than that.  The DUT and reference phase difference can't be
allowed to exceed 50 ns per trigger interval, in the case where two ~10 MHz
signals are being measured.  If the frequencies disagree enough to create a
phase slope greater than that -- meaning if they are more than 5E-8 * 10E6 =
0.5 Hz apart if you are triggering once per second -- your TI counts will be
aliased.  

You won't be able to unwrap the phase properly if your software isn't aware
that the data is coming from a different Nyquist zone, so to speak.  I'd
expect some odd looking ADEV plots in that case.

-- john

> -----Original Message-----
> From: time-nuts-bounces at febo.com [mailto:time-nuts-
> bounces at febo.com] On Behalf Of Mark Spencer
> Sent: Monday, March 26, 2012 2:08 PM
> To: time-nuts at febo.com
> Subject: [time-nuts] Basic question re adev measurements
> 
> Greetings,  I was reviewing some older adev plots of mine and noticed that
> there may be a correlation between lower adev numbers and lower
> frequency off set between the reference source and the device under test.
> It's my understanding that the adev calculations remove constant frequency
> off sets but I'm wondering in practice this degrades the the measurements.
> 
> It occurs to me that if I am comparing two 10 MHz signals with a TIC that
the
> available dynamic range of each measurements will be 100 ns.   Would a
> constant frequency off set effectively reduce the precision of the
> measurements by eating up some of this dynamic range ?
> 
> To put this in perspective frequency offsets of say one or two parts per
> trillion seem to result in better adev readings than off sets of say ten
or
> more parts per trillion.
> 
> Sorry if I have missed something obvious here.
> 
> Thanks in advance
> Mark Spencer
> 
> Sent from my iPod
> _______________________________________________
> 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