On Sat, 7 Jun 2003, [ISO-8859-1] Anselmo P?rez Serrada wrote:

> I think I can help you on this. They use a very elegant method to draw a 
> sundial based on geometrical
> affinity that traces back to our High School days:
> 
> 1. Draw two concentrical circles : one of radius r and the other one of 
> radius r*sin(Lat)
>  2. Now draw a sheaf of  24 equispaced lines from its center as if it 
> were an equatorial dial.
>  3. These lines intersect the circles at points I' and I'',  II' and 
> II'', and so on up to XXIV' and XXIV''.
>  4.  Now trace horizontal lines from the inner points and vertical lines 
> from the outer points. Let's call
> I the point where the lines from I' and I'' intersect, II the point for 
> II' and II'', and so on.
>  5. If we connect these points we just have the analemmatic ellipse, 
> right? Well, but if we trace lines
> from the center to these points we get a horizontal dial for that 
> latitude. Isn't that nice?

Very well described, Anselmo!

Everything agrees perfectly with the original article, except you have one
circle of radius 1 and the other of 1*sin(lat). Now, if I am looking at
this correctly, that would make the second circle smaller than the first
(and also avoid the singularityat 0 degrees!) However, is the ratio of the
two circles' size the same in both constructions? I'll have to draw them
up, and scale one to match the other...

Dave
37.28N 121.97W


-

Reply via email to