[time-nuts] What's the time Mr Wolf...

Tom Van Baak tvb at LeapSecond.com
Thu Oct 30 14:36:36 UTC 2008


> Hi,
> 
> This may have already been covered in this group so please excuse me
> for not combing the archives. I could Google for this but I'll get a
> quicker answer here and I know it will be the correct one.
> 
> !) Who decides on what is the correct time?

BIPM, the same guys that keeps the meter and the kilogram, etc.

> 2) Is the time standard stored on one master grandfather clock?

No.

> 3) As (2) is unlikely, if a number of clocks are involved, are they in
> different countries?

About 350 clocks in 50 countries, last I heard.

> 4) As (3) is likely, how are they synchronised?

They aren't synchronized in the traditional sense. But they are
slowly steered toward the mean. Each country has their own
official time but all follow the same mean time scale (UTC).

> 5) If this involves communicating with each other, what medium is used
> and how are delays handled (a la NTP pings)?

NTP is not used. Instead the clocks are compared using special
GPS receivers, or dedicated round-trip satellite links. Relativistic
and Sagnac corrections are applied as necessary.

> 6) If all the clocks are synchronised in one place and then
> transported all over the World, how is the Einstein affect of time
> dilation handled due to the clocks changing in flight?

The clocks typically aren't moved; they stay in their laboratories.
But when clocks are transported, yes, relativistic effects are
calculated and corrections applied.

> 
> Now the interesting ones (for me anyway).
> 
> 7) Where are the clock(s) located which provide the time for the GPS satellites?

USNO, in Washington DC.

> 8) If they are actually in the satellites, how are they all kept
> synchronised with each other and the "master grandfather clock" on the
> ground?

Yes, there are clocks in the satellites too. Their phase and rate
can be updated by ground control.

> 9) Are the satellites in geostationary orbit?

No.

> 10) If they are geostationary, or move in relation to the ground, how
> are the affects of the Einstein time dilation handled as they may be
> travelling at a different speed than a point on the surface of the
> Earth?

Geostationary or not, there are satellite relativistic effects which
are taken into account.

> 11) Extrapolating this, a point on the Equator would be moving faster
> that a point at the poles or even Greenwich, for that matter. So would
> a clock at each location move out of synchronisation with each other?

Yes, and this also is taken into account. When you get down
to measuring absolute frequency at 1e-14 and 1e-15 levels one
always takes the local gravitational field into account, which is
mostly a function of altitude, but also latitude.

> Aren't you glad I'm back :) BTW, where is /tvb these days, I haven't
> seen any postings by him while I have catching up?
> 
> 73
> Steve - ZL3TUV
> -- 
> Steve Rooke - ZL3TUV & G8KVD
> Omnium finis imminet

/tvb




More information about the time-nuts mailing list