[time-nuts] Cesium vs H Maser clocks

[Magnus Danielson]

Mike S skrev:
> At 08:20 PM 11/28/2008, Tom Van Baak wrote...
>> If you really get into the details of the physics, remember that no
>> commercial or laboratory Cs clock actually resonates at precisely
>> 9 192 631 770.000 Hz. There are corrections for magnetic fields,
>> velocity of atoms, temperature, cavity design, even for gravity; a
>> whole bunch of interesting effects.
> 
> Isn't the temperature the _only_ thing to correct for?
> 
> The definition of the second is "...the duration of 9 192 631 770 
> periods of the radiation corresponding to the transition between the 
> two hyperfine levels of the ground state of the cesium 133 atom." (and 
> affirmed by the CIPM in 1997 that this definition refers to a cesium 
> atom in its ground state at a temperature of 0 K)
> 
> That other factors can change the relative frequency of different Cs 
> clocks is a problem with the definition, not an indication that any 
> particular one is better than another. If a magnetic field changes the 
> relative frequency, but that isn't reflected in the definition, is it 
> not the definition which is faulty, and not the timepiece? The second 
> is imprecise in this regard. 

Far from. What all Cesium clocks do is to de-tune with the C-field. This 
is not a bad thing since you may hit a numerically more suitable number 
that way such that the generated 5 MHz or 10 MHz has a neat relationship 
to the de-tune frequency. Modern CPU controlled cesiums contains a loop 
in which the nearby peaks is also monitored, as they spread out from the 
central peak very rapidly (square of C-field). This spreading you want 
anyway, since then you can observe the stable transition free from the 
more sensitive transitions. By monitoring the side-bands distance the C 
field can be locked up and remain at the stable value throughout the 
clocks life. This avoids a classic long term drift component and is part 
of a battery of improvements to achieve better frequency error. It also 
actively cancels whatever magnetic field that did get through the 
magnetic shielding.

Temperature and gravitational potential are among those that needs 
compensation.

The temperature of the gas stream will provide a continous set of 
doppler frequencies.

Shifts in phase of the microwave interrigation also cause problems, and 
it can be off to start with and shift with temperature of assembly.

There are many error sources. The simple story is that nobody get it 
"right". We have a common goal to approximate towards.

Cheers,
Magnus