[time-nuts] Re: Phase coherence with 2x GPSDO

[Tom Van Baak]

Hi Krishna,

 > Hello Tom,
 > Yes, the GPSDOs are working well. However, when I use each as a reference
 > to a separate radio, I find there is a slow phase change over time 
between
 > said radios. I imagine this is expected since there will always be some
 > error between two discrete oscillators.

This is typical; it's not just the oscillators, but also the GPS 
receiver, s/w algorithms, board level components, and environmental 
conditions. How much phase change do you see? And equally important, 
over what span of time?

 > However, I am hoping to use the PPS
 > and FEE metadata to compute what the phase *should* have been in
 > post-processing. So far, it is not working out for me. I am wondering if
 > that is even possible or if my math is just wrong.

Sorry. The TINT and FEE values are more like phase and frequency error 
estimates, not offsets that you can use to improve the output of the 
GPSDO. I mean, if it were that easy then the GPSDO would just do it for 
you. So think of them as a rough attempt at "error bars". To give you an 
idea of the actual measured performance of a GPSDO like the Fury, read 
this test report from 2007. Your Fury may differ from mine so take this 
all with a grain of salt.

http://leapsecond.com/pages/fury/

There are also phase noise measurements, in case that is of interest:

http://leapsecond.com/pages/fury/phase.htm

It would help to know your expectation or requirement for phase 
coherence. In a sense all GPS/1PPS signals are phase coherent because 
they are all synchronized to UTC. So usually the question is not if they 
are coherent but how close. And not just how close, but how close for 
how long.

 > Where does the PPS offset come from?

For GPSDO that report that sort of thing, it's the time interval between 
the GPS-output 1PPS and the OCXO-derived 1PPS. GPS tends to vary a lot 
from a few ns to a much as tens of ns, but this averages out very well 
over time. The OCXO hardly varies at all short-term but drifts and 
wanders over the long-term. The purpose of a GPSDO is to meld the 
stability (but inaccuracy) of quartz and the accuracy (but instability) 
of GPS.

 > Isn't it from the positioning error?

No. In fact many GPSDO run in so-called zero-D mode where the antenna 
position is hardwired into the receiver.

 > Typical GPS receivers have 1-3 m of positioning error which should give
 > you +- 10 ns. Why is this a "dream" performance?
 > It should be expected from any modern GPS receiver.

I don't see where Bob used the word "dream" or "10 ns". Your 
measurements and my plots linked above show the Fury can do ±10 ns. If 
you are expecting 3-4 degrees at 150 MHz then that's ±50 ps so a Fury 
GPSDO is 200× away from your goal.

/tvb