[time-nuts] hm H Maser

Bruce Griffiths bruce.griffiths at xtra.co.nz
Tue Jan 10 22:43:05 EST 2017

A  spherical cavity resonator with a spherical Rubidium cell configured as an integrating sphere (to enhance the SNR of the optical absorption signal) is a potential option. Its also possible to use the same cell to lock a 795 nm laser to the desired wavelength. Fiber coupling the laser light could also be useful.Note that with an integrating sphere (or any other random scattering process eg scattering from colloidal particles undergoing Brownian motion) laser polarisation isn't preserved which may be convenient.One potential issue with an  integrating sphere is the longevity of the diffusing coating (typically Barium sulphate with an organic binder). Roughening (fine grind  followed by HF etch for stress relief) the outer surface of the cell is also advisable to eliminate light pipe effects in the cell wall.

    On Wednesday, 11 January 2017 3:54 PM, Bruce Griffiths <bruce.griffiths at xtra.co.nz> wrote:

 A goal with ADEV ~ 1E-13/Tau (for Tau <1000sec) may be feasible as its already been done as part of a PhD thesis.Using as large a cavity as possible is probably useful so that a large cell can be employed.What resonant mode is desirable in the cavity?Do we need to avoid field reversal as in the hydrogen maser?

    On Wednesday, 11 January 2017 3:43 PM, Bob Camp <kb8tq at n1k.org> wrote:


I suspect that there are (or will be) some other cheaper / easier ways to do the same thing. The signal to noise requirements 
in the RF chain are sensitive to a couple of things, but not to an absurd level. You do need good close in noise. I would not even bother to 
go for a “final” RF section until the physics stuff had been worked out. Designing today vs designing in a couple
of years will always be the more expensive approach. For the lash up, I might well gut parts out of an existing
cheap Rb simply to get things going …. who knows. Maybe we would need a chain like the one in the paper to figure out 
what is going on.  In four  years take a look at what is on the market and make some decisions about the “final” RF chain. 
Even then you might revisit it several years after that due to cost or performance issues….

This does get back to “state of the art Rb” and what that means. In my suggested case that’s measured in terms of ADEV for Tau = 1 to 1,000,000 
seconds. If you wanted an Rb with (only) state of the art phase noise at 1 MHz offset … that’s a different thing. State of the art for 
power consumption and size is also not what I’m suggesting in this case. Why the choice of spec? … this is TimeNuts. 

For some guidance on what state of the art in Rb’s *is* in this area, check out the many papers on the GPS Rb’s published in in the ION conference proceedings. 
Can a bunch of hackers do quite that well? … likely not. They have been fiddling with that design for many decades. They also have a pretty healthy budget to 
produce each one they build. We certainly can try to get as close as we can.. Testing ours in orbit *might* put a strain on the budget though :)


> On Jan 10, 2017, at 9:15 PM, Bruce Griffiths <bruce.griffiths at xtra.co.nz> wrote:
> A possible RF chain for a Rubidium standard using off the shelf parts plus a couple of custom microwave filters:https://arxiv.org/pdf/1204.4215.pdf
> Bruce 
>    On Wednesday, 11 January 2017 2:10 PM, Bob Camp <kb8tq at n1k.org> wrote:
> Hi



More information about the time-nuts mailing list