[LEAPSECS] My FOSDEM slides

[Warner Losh]

> On Mar 6, 2015, at 7:57 PM, Steve Allen <sla at ucolick.org> wrote:
> 
> On Sat 2015-03-07T02:02:12 +0000, Harlan Stenn hath writ:
>> When we get a bit more down the road with NTF's General Timestamp API,
>> I'd appreciate your taking a look at what we're doing and helping out in
>> any way you are up for.  One of the issues that will need more attention
>> is pre-1972 stuff.
> 
> Before 1972 is pretty simple.
> 
> Without some means of comparing an event with a particular radio
> broadcast time signal or a particular astronomical event, everything
> before 1972 is just plain UT or GMT back to 1676, and UT before that.

Before 1972 there were radio signals. LORAN C was operated from the early
60’s onward, and many receivers existed to recover timing data from them.
WWV dates back to the 30’s, with the current Ft Collins transmitter dating from
the 1960’s. It started broadcasting UTC time in December 1968.

> Any subsecond deviations from what those represent (including worries
> about UT0, UT1, UT2) are only available by painstaking inspection (and
> re-interpretation into modern terminology) of the tabulations in BIH
> Bulletin Horaire.

I’m not entirely sure that’s correct. You can recover time to at least tens
of microseconds from WWV or LORAN-C, which is quite a bit better than
you need to notice the difference between UTC and UT1.  With a good cesium
standard, synchronized at NBS, running off battery power for the car ride to the
lab, WWV could keep you on frequency as well, and you could get sub-microsecond
level of performance. I got to hear many tales of the old days when they did
this sort of two-way time exchange before satellites. This allowed the frequency
broadcast by WWV to be within a few parts in e11 (day it is 100x better).

If you read the histories of WWV, they talk about the close proximity to Boulder
enabling this. The reason was that the battery backup would be enough to last
the 45 minutes or so it takes to go from Boulder to Fort Collins…  WWV followed
the frequency offset + phase steps in the rubber leap second era, which was
an attempt to keep UT1 and UTC not only synchronized, but also syntonized. I find
it curious that the practice of adjusting the frequency of the clocks persisted for
5 years after the SI second was defined.

So it isn’t outside the realm of possibilities that you’d have people making measurements
from the late 60’s till 1972 using UTC (and yes, it did exist in a practical form
before 1972, just not in the current form and the common usage often leaves some
ambiguity between the actual, realized form as broadcast by WWV, or the proleptic
form w/o leap seconds). Actual measurements from this time, though, were based
on something approaching the UTC as broadcast by WWV. Not sure how many data
sets from that time survive until today, and how many need to be converted from that
to UT1 or UT2, but evidentially there’s some...

Warner
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 842 bytes
Desc: Message signed with OpenPGP using GPGMail
URL: <https://pairlist6.pair.net/pipermail/leapsecs/attachments/20150306/5c99cc5e/attachment-0001.pgp>