[time-nuts] Optical transfer of time and frequency

[Paul Boven]

Hi everyone,

On 04/29/2016 03:28 PM, Magnus Danielson wrote:
> Phase/time transfer over fiber is shaping up, but White Rabbit is
> starting to grow up and more reports for long distances is showing up.
> ETFT is one of the placces to check for reports.

And the White Rabbit workshops, with the presentations online:
http://www.ohwr.org/projects/white-rabbit/wiki/Mar2016Meeting

I happen to be working on time transfer via White Rabbit. The White 
Rabbit standard proscribes the use of 1000Base-Bx10 (10km reach 
bi-directional) SFPs, but it turns out to work just fine with longer 
reach SFPs. However, there are several effects that limit the accuracy 
that you can get on longer links:

* Dispersion of the fiber (as the lasers change temperature, their 
wavelength changes, and they experience a slightly different index of 
refraction, hence propagation speed.

* Change in index of refraction in the fiber itself. The propagation 
speed of both the uplink and downlink wavelength change, in absolute 
sense but also their ratio changes. This is something the WR protocol 
can't detect/correct for.

There are several people working on these issues, trying to improve both 
the calibration and stability even further.

Regards, Paul Boven.