[time-nuts] One Kg Quartz Resonator

Magnus Danielson magnus at rubidium.dyndns.org
Fri Jan 25 18:20:02 EST 2013


On 01/24/2013 07:12 PM, Bob Camp wrote:
> Hi
>
> If you take the position that a primary standard is only functional if it's
> under the ideal nominal conditions - you have no primary standards at all.
> They all require corrections of one sort or the other. Having a system with
> no standards is not a system at all...
>
> The practical approach is to define the ideal conditions in a way that you
> can indeed correct back to them. The most common way is to take the
> contribution to zero. There obviously are other approaches. Regardless of
> weather you take it to zero or x.xxx the net result is the same, as long as
> everybody does the same thing.

There is no real "standard" of anything holding up to the ideal "primary 
standard" that some people think of. Rather, one way or another there 
are practical physical limits which comes into play. Defining an "ideal 
case" is needed such that corrections can be made when it is not ideal. 
The 0 K ideal situation is in fact non-reachable by todays knowledge, 
but we can correct for the black body temperature shift. Work on the 
replacement of the NIST F1 will use a cooled tube to lower the black 
body temperature, and that significantly reduces the shifts and 
instabilities.

One has to recall that the selection of Caesium was done with what was 
believed as the best combination of repeatability and least dependence 
of magnetic fields. The magnetic field dependence was in fact better for 
Thallium which also has higher frequency. However, it was believed that 
the higher frequency would cause difficulty in repeatability even if 
it's properties was know better.

With todays knowledge we know that the C-field can be servo-steered, by 
looking at the side-lobes, this removes much of the C-field sensitivity 
issue, even if it still makes sense. Much work has gone into removing or 
compensating many of the effects. As we look forward, it might be that 
rubidium is the next solution, as there is benefits in laser-cooling and 
atom-to-atom interactions in the balls of the fountain. Things is indeed 
moving forward, and modern ion traps as well as optical clocks far 
outshines the capabilities of fountains.

So what's a primary standard, but what we currently think is the best 
technical solution for a reference that has repeatability and stability 
compared to other technical solutions.

Look at how the meter definition has changed over time, and how it now 
has been replaced by being tied to be a derived property of the SI 
second. While the avrogado ball might be a significantly better solution 
than the lump of metal we have now, I think there is difficulties in the 
repeatability which makes the Watts balance a much more attractive 
solution. Then again, maybe there is a benefit in the quartz resonator 
solution, who knows.

Oh, by the way... the SI second as such does not depend on the geoid, 
since if you place to standards at the same acceleration they will 
concurr, within their design limits. However, for the use of dispersed 
standards contributing to the TAI time-scale, the geoid kicks in, as can 
be found in the SI brochure, 8th edition. I knew I saw it there, but now 
I looked it up.

Cheers,
Magnus


More information about the time-nuts mailing list