[time-nuts] Rb vs.Crystal OCXO
SAIDJACK at aol.com
SAIDJACK at aol.com
Mon Apr 28 14:38:50 EDT 2014
in order not to chase the tempco and aging algorithm (it will reset the
values by itself every 10 minutes while locked to GPS as you have noted) you
may simply turn tempco and aging compensation off manually during your test
This can be done by the serv:tas x, yy... command. The first parameter is
the mode; 0 being both off, 1 being aging compensation only, and 2 being
both aging and temperature compensation.
So the commands would be:
to query the current settings, then issuing the command:
serv: tas 0, xx, xx ...
where xx, xx.. are the same values that were shown with the query command.
To re-enable the aging and tempco compensation when you are done with
your measurement, simply send the
serv: tas 2, xx, xx ...
command using the same settings again as the query command had returned.
Hope that helps,
In a message dated 4/28/2014 05:32:41 Pacific Daylight Time,
hans.holzach at gmail.com writes:
if i understand correctly, while in holdover, activated by
sync:hold:init, temperature and aging compensation are still active
(as it should be). however, to measure drift, i have to somehow
disable these compensations, haven't i?
yesterday, i wrote down the values of temperature and aging
compensation, then set both to 0. in holdover the software does not
seem to recalculate the values, at least not during the hour or two
i had the oscillator unlocked. on uli's Z38XX the EFC curve was
flat, unlike when temperature and aging compensation are active (or
not 0) during holdover. one hour or so after sync:hold:rec:init i
checked the values again, and they were very close to the values
before i reset them to 0.
my question is: would this be the correct procedure to have a truely
unlocked oscillator: start holdover and then set tempco and aging to
There is an easier method that does not jiggle the board mechanically:
Disables the disciplining.
The sync:tint? command can be used to check the drift while in holdover..
time-nuts mailing list -- time-nuts at febo.com
To unsubscribe, go to
and follow the instructions there.
More information about the time-nuts