Hmmm

What about the pic measures the time from sun rise to sun set and work
out the difference between day and night.
Then, you still have 12 hours over 180 degrees, say from 6AM to 6PM (or
what ever your average sunrise time is) on  a "day" scale, and 6PM to
6AM on a "night" scale.

At the end of the day, the dial changes direction and runs backwards
through the night, speed based on what was measured, so it will be ready
on the 6AM position the next morning.

Also, by measuring time through the day, any changes due to seasons can
be auto accounted for.
And if it just so happens that it is cloudy that day, the PIC will use
the measurement of the previous day.

Quentin



James Cameron wrote:

>
> The PIC would have to supply the artificial sun.  The scaling would have
> to change to 24 hours for 180 degrees instead of just 12 hours.  I would
> put the sun, a 12V dichroic lamp, on a track in a semicircle around the
> bedside sundial.  It would have to whip back to starting point every
> day; maybe during a favourite television show when nobody is looking.
>
> At least this way the device will work when it is cloudy or rainy.  And
> you won't have to suffer through all these extra calculations for the
> tilt of the planet.
>
> The loss of angular resolution might be a problem, but that could be
> countered with a digital LED display.
>