[time-nuts] CS reservoir depletion

[Magnus Danielson]

On 14/01/11 03:07, J. Forster wrote:
> Magnus,
>
> That's a very interesting paper. It has some interesting stuff on the
> vacuum techniuque that's needed for Rb or more for Cs.
>
> Do you know if anybody has actually built / is selling a Rb with a diode
> light source, rather than the bulb?

Not to my knowledge. However, they have been hacked for experimental 
purposes:
http://tycho.usno.navy.mil/ptti/ptti2003/poster16.pdf

There is more on the topic, such as:
http://www.pttimeeting.org/archivemeetings/2009papers/poster6.pdf
http://tf.nist.gov/general/pdf/972.pdf
http://doc.rero.ch/lm.php?url=1000,43,4,20101230154956-BF/Affolderbach_Christoph_-_Experimental_Demonstration_of_a_Compact_and_High-Performance_20101230.pdf
http://ixnovi.people.wm.edu/documents/NathanBelcherREUPaper.pdf

A particular trick is to apply the RF onto the driving laser and do away 
with the resonant cavity that way. The "chip scale" cesium clocks is 
doing the same thing. Essentially it is a new type of physical package, 
where the (cesium) beam, (H) active maser bounce box and (Rb) gas cells 
has been the classical precursors. Another strain of development is 
naturally the ion-traps.

There is research in the field, and I am only googling along and reading 
some papers every now and then. But as things progresses, the size of a 
rubidium physical package shrinks a lot when the it comes into 
production. Maybe some of the tiny onces already taken the step by now. 
The OCXO package can become the large part... :)

Cheers,
Magnus