[time-nuts] 1 pps Accuracy in two locations

[Magnus Danielson]

Hi,

On 2020-01-15 23:34, Attila Kinali wrote:
> On Wed, 4 Dec 2019 09:40:34 -0000
> <martyn at ptsyst.com> wrote:
>
>> I'm always being asked to provide equipment that can produce two 1 pps
>> outputs aligned to each other to within a few ps.
>>
>> These two 1 pps pulses are not in the same location and could be 100 metres
>> to a few km away.
> As others have written, getting down to a few ps is not feasible, at least
> not with the amount of money your customers are likely willing to pay.
> To get down to these levels you will need to pull fibres from one location
> to another and using special circuitry to activly compensate variation
> in length due to temperature changes and vibration, even for burried fibres.
> Just to put into perspective what your customers are asking for: in 1ps
> light travels 300µm in vacuum/air or ~150µm in fibre/coax.

Let me correct that a little.

For fibre the relative dielectrics of the silica glass is just about
2.25 giving the index just about 1.5, which then gives the 300 um / 1.5
to about 200 um. I am known to indicate the length of 1 ns in fibre
betwen my index finger and thumb, roughly 2 dm, giving the delay for 1 m
to be about 5 ns, letting the round-trip-time for 1 m be 10 ns which is
a very handy number for rule of thumb conversions for fibre. If you look
in more detail, the actual property depend on the wavelength being used
and the temperature of the fibre, as this changes the actual delay.
While first degree compensation is trivial in two-way systems, you end
up having calibration issues.

Coax is less easy. If you have the normal RG58 crap, it aligns to about
the same numbers as fiber, as the dielectrics is about the same.
However, for more phase-stable cables with lower dielectric loss one
simply has less dielectrics to start with, such as foam or other form of
support for center conductor. That gives the relative dielectric go
towards 1 and thus the velocity factor with that. It's much more a "it
depends".

Other than that, I agree with the general analysis of Attila, it is
close to my experience, and I've been working on these things
commercially for over 10 years now. If you want to know how things works
(or rather not work) in a telecom, it is even more painful than this.

So end conclusion being, if you required precision of 1 ps from a timing
system, you are likely going to have one very expensive system and it
will be a pain to operate, it may be worth considering if you are doing
it the right way. I've seen requirements in the 10s of ps for a fixed
system setup, but that is while challenging kind of doable, but then
that requires quite a bit of additional control loops and knowing what
one does.

Cheers,
Magnus