Mac Oglesby wrote: >I've noticed that with a slender, pointed gnomon set perpendicularly to the >dial's face, the shadow's tip sometimes isn't very distinct. Probably >that's why such a gnomon might be terminated with a knob or ball. Yes. In preparation for the large Team Disney dial we did a full scale simulation of the shadow casting process, using a variety of gnomon tip shapes to compare, at distances from the gnomon of up to 130 ft, in order to see what the shadows looked like. When the shadow reaches the bottom of the wall of that dial, the angle of incidence on the wall is quite steep and the shadow is considerably elongated. One of the gnomon shapes we tried was a horizontal flat disk with a circular hole in the center, like a donut. I liked this the best, until the shadow reached the bottom of the simulated wall, where it was so elongated that it looked like a couple of dark stripes and your couldn't tell where you were supposed to tell the time. At least it wasn't obvious. This is why the architect decided to go with a simple 2 ft diameter ball as the least ambiguous shadow-casting gnomon shape.
I wrote a report on that full scale simulation and it is available from the Florida Solar Energy Center by writing to the Public Information Office at the address below and asking for "Full Scale Simulation of the Disney Building Sundial", FSEC-PF-139, May 12, 1988. There's another FSEC report you may find interesting that is also available. It is titled "Civil Time, Solar Time, Large Sundials, and the Meaning of Life," FSEC-PF-192-90, reprint of a paper given at the 1990 annual meeting of the American Solar Energy Society which won a special award as being "the most entertaining presentation of the meeting." The abstract starts off as follows: "Human evolution has undoubtedly been influenced by the diurnal variations in daylight and solar radiation availability that are driven by solar movement and climatic changes...." So I think you can see where I was headed in this paper. There is a movement afoot here to get many of our most popular reprints posted on our web site, or at least available for ftp download by a simple point and click. Requests for these might stimulate interest in posting them on the web. The FSEC web site currently has a description of our early interactions with the Disney people as they were trying to decide if a sundial that large was practical. Go to the FSEC web site listed below. It is also linked to the sunpath-designs web site, so either will get you there. Peter Hirtle wrote: > > >>>that's why such a gnomon might be terminated with a knob or ball. > >>> > >>>My questions are, if the gnomon ends with a ball would you measure the > >>>length of that gnomon to the center of the ball or use the overall length? > >>>Would the shadow's tip be easier to determine with one scheme over the > >>>other? > > My thought on this is that the ball should be at least big enough to cast > an umbral shadow. With a smaller gnomon light from one edge of the sun > is able to illuminate where the other edge is blocked. You would measure > to the center of the ball because that is what casts the center of the > shadow. > I did an article on "The Fuzziness of Solar Shadows" in Vol. 1 No. 3, August 1994 of the Compendium, which addresses some of these issues. Of course if you drill a hole through a sphere, the sun will shine through only when it lines up with the hole. This reminds me of the question asked of me by a colleague in Israel. He was working on the design of a lighting system for a highway tunnel and wanted to know on what days and at what times the sun would shine down into the tunnel. I added a section to my program SUNPATH which answers this question rather crudely. Of course the best way to answer it is to draw a circle around the tunnel's direction on a sunpath chart and this will tell you when and on what days the sun will be in or near that direction. The program, thanks to fast computers, runs through all the sunpaths for all the days of the year until it finds a sun direction falling within a circle of user-specified diameter around a selected direction in the sky, specified by the altitude and azimuth angles. It then prints out all the intersections of the rising (or setting) sun with that circle. A 'flaw' is that it doesn't check for the intersections on the other side of the circle, only the first encounter. However, the result is close to what you want, since it tells quickly when and at what times the sun can be expected to enter a cylinder pointed in any arbitrary direction in the sky. Back to the issue of sphere shadows. Yes, you want to make the sphere large enough to cast a clear umbra on all parts of the shadow receiving surface (otherwise known as the dial plate). In practice the elongation of the sphere's shadow into an ellipse is not generally a serious problem, unless the ratio of gnomon height above a planar receiving surface to the farthest most distance of the shadow is quite small. This is why I like to design curved shadow receiving surfaces, so that as the shadow lengthens, it climbs up a slope, all the way to sunrise or sunset, without having to be projected many times the gnomon's height. -- Dr. Ross McCluney, Principal Research Scientist Florida Solar Energy Center, 1679 Clearlake Rd., Cocoa, FL 32922-5703 Voice: 407-638-1414 Fax: 407-638-1439 e-mail: [EMAIL PROTECTED] WEB SITES: Florida Solar Energy Center: http://www.fsec.ucf.edu SunPath Designs: http://www.sunpath-designs.com Introduction to Radiometry and Photometry:http://www.artech-house.com --------------------------------------------------------------
