[LEAPSECS] Leap second to be introduced at midnight UTC December 31 this year

[Steve Allen]

On Mon 2016-07-25T10:37:12 +0100, Tony Finch hath writ:
TVB> Right, for LSEM to work, you'd have to extend the DUT1 limit a bit
TVB> beyond 0.9s. Historically, DUT1 has changed from 0.1s to 0.5s to 0.7s to
TVB> 0.9s, so I doubt 1.1s or 2.0s would be the showstopper.

At least we know how to lie to the black-box software in our robotic
telescope so that it could handle that without rejecting the numbers.

> As I understand it the problem with |DUT1| > 0.9s is to do with radio
> signal compatibility, e.g. MSF only allows |DUT1| <= 0.8s +/- 0.05s
>
> http://www.npl.co.uk/upload/pdf/MSF_Time_Date_Code.pdf

WWVx has the same 0.7 s limit in its format.  This is a result of the
way that CCIR Rec. 460 came into being.

The original 1969 draft that became 460 allowed leaps of multiple
seconds.  The 1970 February plenary assembly changed that to leaps of
only one second, but Rec.  460 gave no other detail about how to
implement the scheme.  The 1971 February meeting of CCIR Study Group 7
produced CCIR Report 517 (Question 1/7, Resolution 53) which said
maximum deviation of 0.7 s.  That's what WWVx and CHU implemented
because that's what the BIH had been directed to do.  <sarcasm> and
the BIH always accomplished what it was directed to do, especially
because this directive came not just from an international scientific
union but from an international regulatory agency, so no problem the
BIH will just do it, right?</sarcasm>

Then 1972 came along and BIH found that the reporting of input data to
them and their ability to process those data could not keep the DUT1
within 0.7 s.  This was pointed out at the 1973 IAU GA, and the 1974
revision of CCIR Rec. 460-1 raised the maximum departure to 0.9 s.
Along with that came language that was as close to an apology to the
BIH as I have ever seen in such a document.  Unfortunately the WWVx
and CHU formats were already set with their max of 0.7 s.

Despite the recommendation in 460 that DUT1 should be provided, DCF77
has no means for that, so there is another example of an agency not
bothering to implement the full text of Rec. 460.

Up until 1970 the CCIR recommendations about radio broadcast time
signals had been descriptions of techniques already broadly in use and
tested for interoperability with existing systems.
The one good thing about the ITU-R WRC-15 decision not to decide is
that they did not repeat history by producing another recommendation
with a prescription that had not been tested for interoperability.

--
Steve Allen                    <sla at ucolick.org>              WGS-84 (GPS)
UCO/Lick Observatory--ISB 260  Natural Sciences II, Room 165  Lat  +36.99855
1156 High Street               Voice: +1 831 459 3046         Lng -122.06015
Santa Cruz, CA 95064           http://www.ucolick.org/~sla/   Hgt +250 m