On Sat, 05/01/2002 Dave Bell wrote
> I'm having a hard time visualizing this, this morning! I understand the
> mirror placement and tilt, but am not sure I see how this would yield a
> polar dial, with parallel analemmae. VERY nice idea, if it works, though!
Well, at the moment I do not have installed any drawing program (and
remember
the last mess we got on account of the images!), but the idea is very
simple:
SIMMETRY
Imagine that my leant mirror is so long and wide that it intersects the
ceiling. Now
imagine a little dwarf over the mirror: he sees the ceiling down under
his/her feet
with a slope of lat/2 degs, which added to the lat/2 degs of the mirror add
up to
lat degrees over the floor, and *this* is a polar sundial, QED.
> By the way, I think the best method to draw the meridian line on the
> ceiling of a room is the obvious one: consult in an almanac which is
> the time for noon in this day for your place. Mark the bright dot on
> the ceiling four hours later and four hours before that moment and
> connect both dots: the mediatrix of this line approximates very well
> the meridian line.
>
> Somebody's got a better idea?
>
>Sounds like an excellent suggestion. This would require a very accurately
>levelled mirror, though - perhaps John's mercury dish - or the effect of a
>tilted mirror would introduce a distortion, just as your first paragraph
>suggests...
For that purpose you can perfectly use John's mercury dish and then make
the polar ceiling sundial separately... I'm sure that if you compare the
result with
the one obtained by a GPS compass you could hardly tell the difference.
Anselmo Perez Serrada
[ 41.63 N 4.73 W]
