[time-nuts] Time Dilation tinkering

[Attila Kinali]

On Tue, 21 Mar 2017 13:38:51 +1100
Hugh Blemings <hugh at blemings.org> wrote:

> This got me to wondering if a Rubidium based standard might do the trick 
> - the Efratom SLCR-101s seem readily available for ~USD$200 mark.

As TvB wrote, a single one will not do the trick. You will need 
a stability 1e-14 @1d. IIRC most Rb standards floor out at 1e-12 to 1e-13
somewhere between 1k and 100k seconds. Even the Super-5065 has a floor
of about 3-4e-14 (unless our friends here improved on this already).

There will be a few things that you will need to do, if you want to go
with Rubidiums:
1) Stabilize or compensate for environmental effects (temperature, air pressure)
2) Build ensembles of Rb clocks.

The former is to prevent the ADEV from "turning upwards" again and
keep the Rb's as stable as possible as long as possible. I have no
experience how much this helps, but I suspect, with the correct modeling
that you could get the ADEV to be flat up to a day, maybe even two.
You will have to include aging and drift of the cell itself in the model.
This also means that you will have your Rbs running for a few weeks/months
prior to any measurement so that the aging settles down a bit.

The ensemble is to get the ADEV down. Averaging of clocks with the same
noise decreases the ADEV by the square root of the number of clocks used.
If you are close to the ADEV you need, the averaging will help you getting
it further down. Unfortunately, getting from 1e-13 down to 1e-14 is
impractical as you'd need 100 clocks to average over.... But a factor
of 2 to 3 should be possible.


Alternatively, you can ask Oscilloquartz whether they let you borrow
two of their new OSA3300 for a publicity stunt :-)
	
			Attila Kinali
-- 
It is upon moral qualities that a society is ultimately founded. All 
the prosperity and technological sophistication in the world is of no 
use without that foundation.
                 -- Miss Matheson, The Diamond Age, Neil Stephenson