Re: Avoiding the Leap Second

Chris L Peterson wrote:
On Sat, 02 Jun 2007 20:51:18 -0700, Quadibloc <jsavard@xxxxxxxxx> wrote:
..
Divide the year into ten parts of 37 and 36 days in alternation. Start
from March 1 to keep things simple in leap years.
..
For the first 33 1/3 days of each of those parts, sweep increments of
100 milliseconds "under the rug" by adding one millisecond to the last
second of each eight-hour period. This would allow a time scale to be
kept within 0.1 seconds of mean solar time, and it would also mean
that, most of the time, a time signal would consist of a steady stream
of SI seconds; the long seconds would come at predictable intervals.
..
I think it's easier just to divide the year up into an even number of
uniform length seconds for civil time. The thing is, I can't think of
any reason that civil time needs to be particularly accurate. We can
just agree that "accurate" civil time is within 1/10 second or so of the
correct time (as derived from the atomic standard), and set our clocks
now and then to keep it that way. No leap seconds necessary.
..
Well, I couldn't think of reasons why it had to be particularly
accurate. But it turns out that the WWV signal includes a field called
DUT, giving UT1-UTC to an accuracy of 1/10 of a second.

That's because before we used leap seconds, the time scale was kept
within 1/10 of mean solar time, using a frequency offset to change the
average length of the second plus occasional corrections that involved
adding, or subtracting, 1/10 of a second - 100 milliseconds - from a
second.

The frequency offset meant that it was awkward to use the standard
time signals for frequency measurements, and the corrections of 100
milliseconds came too frequently. So that's why the old system was
considered to be 'unworkable'.

Because there are applications that need a very precise standard of
time, tied to the atomic second, I was trying to figure out: how could
the "official" time used by those who need such precision be changed
so that the rest of us - who do just set our clocks now and then to
keep them in step with the official time - could get into step, within
1/10 of a second, or 1/100 of a second, or whatever might be required,
just by setting our clocks now and then?

Of course, "the rest of us" needs some qualification. Those of us who
set our clocks so that we will get up in the morning at the right time
never had needed to worry about leap seconds.

But even people who are trying to maintain critical time
synchronization over computer networks aren't using rubidium
oscillators - with a very few exceptions. At best, they're using
quartz crystals in thermally controlled ovens. And even those are only
accurate to 2/3 of a second per year.

If the official time has leap seconds *in* it, then setting their
clocks by WWV every now and then won't eliminate the need to handle
the leap second when it comes. But if we can spread the leap second
out in a more even way over the official time, then periodic
resetting of one's clock to keep it in step will also take care of the
discrepancy between the SI second and the earth's rotation.

At least I've thought of a way to do it a bit more simply. Let's
decide on 100ms corrections on a *monthly* basis, and add in the extra
milliseconds at the end of every six hours for the first 25 days of
the month. That way there's no need to work from a new calendar of 10
months that alternate between 37 and 36 days in length!

John Savard

.

Relevant Pages

  • Re: Avoiding the Leap Second
    ... 100 milliseconds "under the rug" by adding one millisecond to the last ... correct time, and set our clocks ... No leap seconds necessary. ... average length of the second plus occasional corrections that involved ...
    (sci.astro.amateur)
  • Re: Avoiding the Leap Second
    ... For the first 33 1/3 days of each of those parts, sweep increments of ... 100 milliseconds "under the rug" by adding one millisecond to the last ... any reason that civil time needs to be particularly accurate. ... No leap seconds necessary. ...
    (sci.astro.amateur)
  • Re: The Leap Millisecond
    ... from March 1 to keep things simple in leap years. ... we'd have an *enormous* number of leap milliseconds. ... For human-precision purposes, a leap second doesn't matter, ...
    (sci.astro.amateur)
  • Re: Date Difference - Qsn
    ... Hal Rosser wrote: ... if you know how many milliseconds there are between two dates, ... not a leap year ... (ObJavaQuestion: ...
    (comp.lang.java.programmer)
  • Take an extra second to reflect on 2005
    ... Yep, it's a leap second moment, one of those rare occasions when clocks around ... Greenwich, England, where, by the way, the extra second is formally marked. ... Greenwich Observatory. ...
    (sci.geo.geology)