John, >SOLUTION C: What if I built the gnomon first and use its shadow to tell me >the position of the time lines???? With this method, no calculations, >plotting, protractors or tape measures are needed. Using a shadow >sharpener, the exact position of each timeline could be marked onto the dial >face. Of course, using this method would require the proper EOT, DST, and >longitude corrections. This method would also work well on an irregular >surface. (I think) Marking the time lines would be easier and faster on >those days when EOT=0, right?
I recently completed laying out a 6 metre high x 20 metre diameter horizontal dial which, to add to the problem, had a drainage 'cast' of 60:1 from the root of the style. The geometry of the latter is intriguing and I can send you a jpeg of my analysis. Other problems which arose however dictated a very large precision pivotting laser trigon, the machining of which took me two careful weeks. (jpeg available of the device in use) A chain is as strong as its weakest link however and the promised 'machined edge' of the cast iron gnomon turned out, on delivery to the site, to be left knobbly 'as cast' so all that careful machining of the trigon was to little avail. Even had it marked out the *theoretical* position of the shadow with the greatest precision the result would have been disappointing due to the umbra/penumbra. This is best illustrated by the twin Noon Lines from opposite edges of a 12" wide style which would have been set parallel in theory. Near midsummer at 55° north the shadow is 12" wide at the root of the style but only 10.5" wide at the numerals.(although individuals disagreed on what constituted the precise 'edge' ) The noon lines therefore taper towards the edge of the dial. Although I haven't checked this by direct observation I suggest that the taper will vary with the season because the much shorter shadow in summer will have a broader 'tip' that the 'off-the-dial' winter version. Perhaps the solstices are the best time for laying out such dials to obtain an average position?? Fortunately I was able to 'zero' the p.m.trigon on an observed local noon which solved some of the problems but the a.m. side had to ! be re-set. I would suggest that the ONLY way to set out a large dial with 100% reliability is by direct solar observation of at least alternate hourlines with perhaps a little interpolation in between. Or perhaps a ginormous pair of Dialling Scales :-) The second biggest problem for me was that such work has to be done by contractors working to budgets and deadlines. To them a sundial is just a public ornament to be completed as quickly as possible so that they can send in their account. And the biggest problem???? You've guessed it - NO SUN! 1998 must have had less sunshine in the northern UK summer than for many years. The locals grew used to the wild-eyed 'fool-on-the-hill' shouting imprecations at the sky as that bank of cloud obscured the only glimpse of sun that day after a frantic car journey to the site. And we do this for pleasure......??? Tony Moss
