[time-nuts] TIC resolution impact on GPSDO's performance

[Poul-Henning Kamp]

In message <45906C9A.806 at xtra.co.nz>, Dr Bruce Griffiths writes:

>Poul-Henning
>
>You are using a different meaning for coherence than the standard one.
>The standard meaning is something like:
>Two signals or waves are said to be */coherent/ *if their behaviour at 
>various times/places is linked in a deterministic way.

The term coherence originates in optics and was briefly used in
radiotelegraphy and is no almost exclusively used in the context
of lasers and waveguides.

When two signals are coherent, you can predict one from the other.

We are dealing with three signals here:

	GPS_PPS:	the 1 Hz signal as the Oncore wants to do it.

	HW_PPS:		the 1 Hz signal as the Oncore can do it.

	NEG_SAW:	(GPS_PPS - HW_PPS)

None of the above three signals are coherent, you need two out of
three to predict the last one.

And since the first one is a virtual signal, not available except
as the sum of the other two, talking about coherence is especially
pointless.

I think what you are trying to express is that the frequency from
the internal Xtal (at times) is in an overtone of the 1Hz PPS, which
gives rise to hanging bridges.

And to conclude this by now surely far too tiresome debate before
both of us gets booted from the list:

The field in the oncores serial data-stream called negative sawtooth
correction must be taken into account to get optimum performance
out of the receiver.  Ignoring it in toto or simply averaging it
will generally not do for OCXO disciplining.

-- 
Poul-Henning Kamp       | UNIX since Zilog Zeus 3.20
phk at FreeBSD.ORG         | TCP/IP since RFC 956
FreeBSD committer       | BSD since 4.3-tahoe    
Never attribute to malice what can adequately be explained by incompetence.