[time-nuts] Allan Deviation -> more data: GPS 1PPS against OCXO/128
didier at cox.net
Sat Oct 28 09:07:51 EDT 2006
Is there a way to insert comment lines in the data file that will not
My logging program creates a header with useful information to record
how the data was collected, but it messes up Plotter of course.
I have made a run over night of the 10 MHz OCXO divided by 128 against
my Jupiter receiver's 1 PPS output, open loop.
I think the data makes more sense :-) but I need help figuring out if
it's good/useful or not. There are some strange jumps in the phase plot,
which is otherwise very clean.
1PPS was fed to START and 78.125 kHz was fed to STOP. The 5370A was set
to TI, 1 sample, +TI only, and the data was collected at 1 sec
intervals, no averaging. There may have been a few samples missed since
the PC was collecting data based on its own 1 sec clock, which may or
may not have been quite the same thing as the GPS 1 PPS. I am not sure
of the effect of that. I may have missed some samples.
Since it was done open loop, and everything only had an hour or so
warmup before I started collecting data, there is an appreciable drift
initially, and there was a phase jump of 1/78125 (12.8 uS), so I split
the data file in two and analyzed them separately. Both files are on the
web folder below, respectively Counter_TI-10-27-part1.dat and
Counter_TI-10-27-part2.dat respectively, with the related Plotter print
Note: house heating/AC was turned off the entire night, so no strange
thermal effects took place.
The data collection is continuing.
Point your browsers to http://www.ko4bb.com/Test_Equipment/
More information about the time-nuts