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

SAIDJACK at aol.com SAIDJACK at aol.com
Mon Dec 25 14:20:06 EST 2006

In a message dated 12/25/2006 06:18:57 Pacific Standard Time,  
df6jb at ulrich-bangert.de writes:

In my  own GPSDO i discipline the LO's pps to have a MAXIMUM phase offset
of 500  ms against the receiver's pps. That avoids lots of ambiguity
problems and  leaves the widest phase measurement range to make the loop
even lock when  the OCXO is initially far apart its setpoint.

Otherwise you are correct  that one can optimize for different
parameters. This gives immediate rise  for the following publications:

1) A guide to discipline your frequency  source to a phase source with
minimum phase offset

2) A guide to  disipline  your frequency source to a phase source with
minimum  frequency offset

3) A guide to discipline your frequency source to a  phase source with
best phase stability

1) A guide to discipline your  frequency source to a phase source with
best frequency stability

If  you can deliver or send some links: I will surely devour stuff like

Best Regards
Ulrich Bangert, DF6JB

Hi guys,
this brings us to an interesting discussion: what is more useful:
   * best frequency accuracy
   * or best phase lock (minimum 1PPS phase offset to UTC).
We have one customer at Jackson-Labs telling us most implementations prefer  
frequency accuracy over phase accuracy since these units are often used in the 
 field to drive frequency counters, signal generators, and receivers etc, 
that  don't require tight phase offsets, but rather need a very accurate  
It's a trade off of short-term frequency stability (lower for best phase  
performance) or long term frequency stability, with longer phase drifts  away 
from UTC but better overall frequency accuracy.
I suspect the time nut's would rather have a very accurately locked 1PPS to  
UTC and don't mind larger short term frequency offsets needed to steer the 
1PPS  onto the UTC 1PPS.
When doing frequency measurements, one would prefer to always have the most  
accurate frequency possible.
Two different settings in the loop performance are required here. On our  
Fury GPSDO, we allow the user to set the loop parameters by hand (with some  
suggested defaults) to optimize the unit to be either tightly locked, or  more 
loosely locked to UTC. We also have two OCXO options available to address  the 
problem: a double oven high-stability unit that can be much more  loosely 
coupled, and a more cost-effective single oven that wants to be  tightly controlled 
by the GPS with higher short-term frequency wandering.
On top of the ADEV loop time constants cross-over point discussion  Ulrich 
described, it's also a question of how much UTC (as delivered by the GPS)  phase 
offset these units can achieve in say 1-Sigma measurements (<5ns  and <<15ns 
respectively typically).

More information about the time-nuts mailing list