[time-nuts] telling time without a clock

[Chris Albertson]

On Wed, Jan 25, 2012 at 8:38 PM, Jim Lux <jimlux at earthlink.net> wrote:
> OK.. without getting into celestial navigation, the whole thing of telling
> time with the moon is intriguing.  And with some forethought and data
> available today, we could fairly easily do what folks back in the 18th
> century could not.
>
> Let's say you run a suitable celestial model and identify all the reasonably
> bright and identifiable star that the moon occults in a given day.  The moon
> moves about 1/2-1 degree per hour against the star field, so the question
> is, could you find, say, a star every couple hours.

If you have a telescope and you can measure where it is pointing
relative to the local meridian, then you don't need the moon.  You can
use a fine wre in the optical path an watch for when a star crosses
the wire.  The advantage of this is the telescope does not need a
tracking motorized mount.  It can be fixed to a concrete pier.    Even
a modest scope in the city can see hundreds of stars per hour.

Using the Moon is only useful if you can't measure where the scope is
pointed.  The Moon provides a good, well known reference.  So for a
portable setup it could work best but there is a built-in problem with
the Moon, you may not have good data on the shape of the limb.
Mountain ranges and valleys between peaks are different depending on
your location on Earth.  If you move even a mile your star might hit a
different place.    In fact people have used Lunar occulations to map
the height of lunar mountains.    Another effect is diffraction.  The
stars don't just "wink out" because they do have a finite diameter
People have actually used the moon to measure the diameter of stars by
accuratly measuring the defraction effects.   But the project had
problem because of large boulders and mountains on the moon made it
hard to know the orientation of the "knife edge" and worse, this would
chane if you move just a few feet, some different boulder might be
there.

You should be able to get very good accuracy if you have a stable
local oscillator and make many observations


Another idea that maybe is even better is to use radio observations
with two antenna that have a very long east/west baseline.   You watch
the difference in phase to a distant radio source.   As the phase
different passes zero you know it just went overhead and then the time
would have to equal the R.A. of the radio source.   Problem is the
physical length of the cables you'd need to lay out and the lack of
really bright radio sources.   In theory one could get arbitrary time
accuracy this way.    A few radio source are "easy" to detect with
affordable surplus/ebay equipment.


Chris Albertson
Redondo Beach, California