[time-nuts] Pressure related rubidium oscillator aging

Bob Camp kb8tq at n1k.org
Tue Aug 14 14:26:55 EDT 2018


Aging in the short term tends to be dominated by equilibrium issues. Some of them can have “many months” sort of time constants . That makes any study like this pretty hard to do. 

Worst case is when a short term effect has the opposite sign from the long term. You go through a really flat region .... for a while ...


Sent from my iPhone

> On Aug 14, 2018, at 12:53 PM, Skip Withrow <skip.withrow at gmail.com> wrote:
> Hello Time-Nuts,
> For the past several months I have been investigating the change in
> aging rate of a rubidium oscillator with change in pressure.  This has
> been done with an operating oscillator in a temperature controlled
> vacuum chamber.
> Obviously, the frequency changes (a lot) with pressure changes due to
> 'oil canning' of the rubidium cell.  What I have been interested in is
> the behavior of the change in frequency (aging) with pressure. If an
> inflection point can be found or the aging minimized there is
> potential to have an oscillator that performs very well.
> The only 'change in aging' mechanism that I have run across is helium
> permeation of the rubidium cell.  FEI has a paper published when the
> oscillators for the first GPS satellites were being developed.
> Temex/Spectratime is the only company that I have seen that addresses
> this in their specifications.
> So far, I have been operating in the 10-50 Torr range and have seen a
> very definite trend on aging with pressure.  But there is much more
> research to do.  My big question is - what pressure related mechanism
> might affect aging?  I would think that helium permeation would have a
> somewhat longish time constant.
> I'm wondering if anyone from the group has some experience in this area?
> Thanks in advance for any help.
> Regards,
> Skip Withrow
> _______________________________________________
> time-nuts mailing list -- time-nuts at lists.febo.com
> To unsubscribe, go to http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
> and follow the instructions there.

More information about the time-nuts mailing list