The astronomers on this list have had no qualms about providing
concrete examples of how their systems work and why they would break
if leap seconds were stopped. Aside from GLONASS, those who wish to
abolish leap seconds have not concreately identified the systems which
don't like leaps. This is an uncomfortable asymmetry.
On Wed 2003-01-29T15:14:14 +0000, Ed Davies hath writ:
> Steve Allen wrote:
> > by a system that runs off a multitasking, 8-bit, 6502 processor. Lick
> Out of curiousity, where do these systems get UTC from? Broadcasts
> like WWV or GPS, via NTP or someplace else? I assume there is a
> local, reasonably accurate, clock which is corrected from the
> external source.
The Lick telescope control systems have a reasonably good crystal
oscillator.
Originally the clocks were set via commands over a home-grown serial
multiplexer interface that used coaxial cable as the interconnect
medium. The commands were issued, nightly as needed, from other 8-bit
microcomputers. Time was set by typing in a target time and hitting
<Return> on the tone of a radio broadcast.
Currently these commands travel over TCP/IP up to an interface board
that talks the old language to the controller. The time setting
commands are issued regularly from a 32-bit Unix box that has NTP.
> Also, do they actually use UTC as an approximation to UT1 or do they
> correct UTC to UT1 but just have some built in assumption that the
> correction will be less than one second? If they make the correction,
> where do they get UT1 - UTC from? Is it entered manually?
There is no distinction between UT1 and UTC in this system, and given
the engineering requirements it would have been a waste even to
consider the difference some 25 years ago.
The telescope for which it was designed is 50 years old. At that era
it was clock driven, and the telescope operator and astronomer both
spent most of their time in the dome, at least one of them peering
through the guide scope while trying not to freeze. This was the era
of photographic emulsions and gentleman astronomers not hurrying the
pointing because the exposure would be 4 hours long and besides he
hadn't got the tobacco in his pipe lit yet. Besides that, the
longitude of the telescope wasn't actually known much better than
a second of time.
When the current system was implemented around 25 years ago the
telescope operator and astronomer had moved into the control room
doing most of their target acquisition and guiding via
image-intensified video. The tons of steel and bearings of the
telescope structure, however, proved to have (modellable) flexure and
(unmodellable) hysteresis effects which made sub-second accuracy not
terribly relevant.
Now we are in the era of CCD astronomy with astronomers unhappy that
the target did not slew right into the center of the autoguider spot,
and there are 50 more objects to acquire tonight (and 100 other
astronomers who are competing for her job if she fails to get tenure).
The new telescope control system system under construction still
presumes that UTC is the available form of time and that UT1 is close
enough to it that it doesn't matter for this old telescope. We
have source code control over this system, so we will be able to
fix it later if necessary.
The automated observing programs to be run on the new telescope that a
contractor will soon build for us do require sub-second UT1. Given
the lack of concrete in this "no leap seconds" proposal we cannot
specify that the contractor provide a means for handling a condition
for which no specification exists. This almost certainly means
calling the contractor back for more work if leap seconds are stopped.
--
Steve Allen UCO/Lick Observatory Santa Cruz, CA 95064
sla_at_ucolick.org Voice: +1 831 459 3046 http://www.ucolick.org/~sla
PGP: 1024/E46978C5 F6 78 D1 10 62 94 8F 2E 49 89 0E FE 26 B4 14 93
Received on Wed Jan 29 2003 - 14:19:49 PST