[time-nuts] Re: The SI second and the ease of realization (was: leap seconds finally being retired?)

[Jürgen Appel]

Dear fans of precise time and frequency measurements,

On Friday, 25 November 2022 23:00:12 CET Magnus Danielson via time-nuts wrote:

> > If optical clocks actually shone light, you could point the light
> > from two of them at a screen, and use the resulting pattern of
> > interference fringes to align them to the same exact height :-)
> 
> To within the precision of the realization, yes. 
                accuracy

> It's done with counters
> and not interference patterns, in fact it is done with disimilar
> frequencies. The optical link is some suitable free 1550 nm range
> frequency which is tied to the optical transition usinga stable transfer
> laser and similarly in the other end. This is done using the optical
> combs to jump between frequencies and make comparisons using regular
> frequency counters.
> 
> It's just like the microwave mixing stuff but at higher frequencies.
> 
> Currently the leveling on dm level can be measured.

You would also have to make sure that the air in the lab is _really_ quiet  
where you do the interference fringe experiment:

The gravitational red-shift is about 10^-18/dm. The optical frequencies 
optical clocks run at lie around 450 THz~=4.5*10^-14 Hz. You would have to 
keep the interference fringe drift controlled to better than one per hour to 
get your height measurement right!

Another fun-fact for all time-nerds on the other side of the Atlantic trying 
to do the intercontinental version of this experiment: 
The Doppler-red-shift due to continental drift of ~1cm/year is of the same 
order of magnitude: (1 cm / year)/c = 10^-18 as well.

Cheers,
	Jürgen