[time-nuts] MINOS Status on Measuring Neutrino Velocity

Rick Karlquist richard at karlquist.com
Wed Feb 22 21:43:32 UTC 2012


Marvin Marshak wrote:



> We have just started operating the HP5071A cesium clocks in the last few
> days and are interested in suggestions for optimal operation and
> monitoring of these units.   The HP5071As can be controlled and read out
> via a serial connection, and
> the device has many operating parameters it can report. Which of those
> parameters are most useful for long-term monitoring of the clock
> stability? Are there subtle problems in using these units?

I was on the R&D development team for the 5071A.  To some extent, the
monitoring of parameters was an after the fact add on.  Some of them
are truly useful, but some potentially useful parameters are not
available because it would have been too much trouble to go back
and instrument them, and there are other parameters that are not
especially useful, but were included because we could easily
measure them.  Failure of a parameter is a pretty good indication
that something is broken, but the mere fact of passing all the
parameters is not a sure guarantee that everything is working.
Also, if there is a failure, the parameters are not always useful
in indicating which module is bad.

Having said that, I have not heard of any subtle problems where
a 5071 "worked OK" but was a little off in frequency.  This
kind of situation definitely happened with the 5061A/B, for
example the "top cover effect" and the "silver tarnishing effect".
The 5071A was a ground up design of everything but the CBT
itself, and even that got some improvements.

The best insurance policy is to have an ensemble of 5071A's
with proper comparison facilities so that in the event one
of them goes rogue, it can be "voted out" by the other units.

Rick Karlquist N6RK





More information about the time-nuts mailing list