[time-nuts] Maximizing Cs Tube Life

John Miles jmiles at pop.net
Fri Jul 10 05:47:45 UTC 2009


Many standards have a rear-panel jack for a dedicated ion-pump supply, which
can be left powered up 24/7.  If you run the ion pump continuously, but not
the Cs oven, the tube will presumably last forever.

I added a switch to my 5062C that allows me to leave it plugged in with the
ion pump and OCXO running, but little else.  When I want to stabilize the
OCXO, I flip the switch and wait 10-15 minutes while the Cs oven comes up to
its operating temperature.  This seems to be working well so far.

I don't know exactly how long the Cs oven needs to operate before the final
stability specs are reached, but I'd be surprised if it took more than an
hour or two.  Since the OCXO runs all the time, I would think I'd see some
control-voltage drift if the Cs tube took longer than that to warm up, but I
haven't noticed any to date.

This setup is fairly new and I'm still working on characterizing it, so
don't take any of the above as gospel.  Also, most Cs standards are several
times more stable than the 5062C, and they may need (much) more time to
reach their specs.  Waiting a week before doing anything hardcore is
probably smart.

-- john, KE5FX

> Hello Time-nuts,
>
> Just wondering what the group concensus might be concerning maximizing Cs
> tube life.  On one extreme, if a tube is not used at all it will lose its
> vacuum.  That is why Symmetricom instructs that stored tubes be
> powered for
> at least 30 minutes each 6 months.  At the other extreme is powering the
> tube 24/7 which depletes the Cs source.
>
> In the middle is the question of how long does it take to achieve
> satisfactory stability to use the unit to evaluate other devices
> (oscillators - xtal or Rb).
>
> I would like to maximize the tube life and only need to do evaluations
> periodically.  My first inclination is to fire up the cesium a
> week or so in
> advance of use and then shut it down until needed again.
>
> Just wondering what the time-nut collective experience might be?
>
> Regards,
> Skip Withrow
>
>
>




More information about the time-nuts mailing list