John Carmichael wrote: > Hi Thibaud: > > you suggested the following following solution to the Azimuthal gnomon > height problem: > > >How about making concentric rings for each declination, the innermost for > >+23.44 and the outermost for -23.44 > > I thought about this ring order also and have not rejected it. If you look > at Mayall's azimuthal dial on page 181, you will see that he does something > similar to what you propose, by combining two months in each ring. (In his > version, however, he places the summer months on the outside instead of the > inside and he doesn't correct for the EOT or longitude). > > Someone wrote you back and said that your idea wouldn't work because it > would produce the traditional analemma shaped hour lines. But I'm not so > sure he was correct, because that kind of dial requires a nodus and an > azimuthal dial does not. > > John C.
That someone was I. My point was that for a given declination your design gives one ring. That ring is shared by two dates, because the same declination occurs twice a year. (For instance, the 0 degree ring is shared by March 21st and September 23rd.) The ring must be marked with the EoT-adjusted hours for both dates. Thus each ring will have two sets of hour marks. The proposed arrangement - from +23.4 to -23.4 degrees, mimics the effect of a nodus, which casts a shorter shadow for high declinations (well, here in the northern hemisphere it does). In fact, the rings' radii could be chosen to give exactly the same effect at, say, noon. However they are chosen, the hour marks will form a figure-of-eight. It is only by forcing each ring to be used on only one date, or only on dates when the EoT is the same, that you can achieve the goal of having one set of hour lines on each ring. I hope that's clearer. Chris Lusby Taylor Newbury, England 51.3N 1.4W
