Sorry Folks! I see now that Peter had sent a copy to owner-sundial... instead of to just sundial.... Since it had attachments with pictures of the dial, perhaps this is how it should be done?? I'm somewhat new to the list and I don't know. Anyhow, I'm taking a great chance and forwarding Peter's email including the pictures directly to the list. If this is a NO NO NO thing, let me know how to do this properly. I've a number of other drawings/pictures/emails from folk with attachements that I'd like to share.
Enjoy the Light! Edley McKnight [43.126N 123.327W] ------- Forwarded Message Follows ------- Date sent: Mon, 17 Dec 2001 14:27:36 +1030 To: "Edley" <[EMAIL PROTECTED]>, [EMAIL PROTECTED] koeln.de From: Peter Mayer <[EMAIL PROTECTED]> Subject: Re: Polar Cycloid Book Dial Dear Edly, >In the NASS Compendium of December 1998 Page 24, which is part of >Fred Sawyer's article on the Cycloid Polar Dial, it mentions making a >vertical dial facing east or west with the noon line now occurring when >the surface is fully exposed to the sun. The idea of both an east and >west cycloid polar dial mounted together with the flat surface mounted >between them occurred to me. [snip] >What do you think? > About two years ago (c. February 200) I posted pictures of the Dodwell dial in Adelaide which is very similar in conception to the dial you are pondering. (I'm pasting an excerpt from comments then at the bottom of this posting) Dodwell was the South AustralianGovernment Astronomer. The dial dates from 1941 and is located in the Pioneer Womens Garden, just to the north of Government House. I spent a fair amount of time trying to discover what curve Dodwell had used. The nearest I could find was Piet Hein's 'super ellipse'. >Tilted at the proper polar angle it would keep quite good Local >Solar Time. If it were arranged to rotate around the polar axis it >could be set to show standard mean time if set whenever the equation of >time, longitude correction and daylight savings time required it. You may be able to see from the detailed jpg which I'm attaching to this posting that Dodwell incorporated a correction for longitude (Adelaide is 4 degrees west of our peculiar half-hour meridian [142.5 degrees]). As well, his design incorporated an unfolded analemma to incorporate the equation of time for every day of the year. (I've often wondered how he plotted the curves in 1940! Probably by projection) cheers, Peter >The dial is located in Adelaide, in the Pioneer Womens's Memorial, just >to the north of Government House. According to the brass plaque beside >it, the dial was designed by George F. Dodwell, the South Australian >Government Astronomer in 1941 (a position which was I fear abolished many >a moon ago). I'm appending three small JPEGs which should make its >structure and operation transparent. The dial is a form of polar dial, >made to appearances of bronze, engraved or cast. It is about 40 cm (15.5 >in) wide and 42.5 cm (16.75 in) long and about 16 cm (6.25 in) from base >to central ridge. Although the dial has a distinct kinship with >Ferguson's Solar Chronometer which is pictured on p. 192 of Cousins >_Sundials_ (why isn't THAT reprinted by someone??), Dodwell has >introduced several innovations which I think make his design superior. >Ferguson's instrument is essentially a segment (perhaps 10 cm? wide) of a >semi-cylindrical surface (that is: of a cylindrical tube, cut in half >down its axis). At each end of the segment there is a semi-circular metal >frame whose ends are joined at the diameter. (The shape is rather that of >the protractors we used when I studied geometry in school...). The dial >has an adjustment (something Dodwell's doesn't) to permit the plane of >the dial to be aligned parallel to the Earth's polar axis. Instead of the >usual solid gnomon of an ordinary plane polar dial, Ferguson has placed a >rod in the axis of the cylinder which is attached to the end frames. The >rod is notched in the centre to form a nodus. Finally, Ferguson's dial >has two printed cards (or probably two sides of a single card) which have >curved lines for the hour, half and quarter hours which adjust for the >equation of time and successive 'straight' lines for the changing >declination of the Sun. Each card covers half a year from equinox to >equinox. The card can be 'shifted' to the left or the right to compensate >for local latitude. Dodwell's dial is novel in several respects. His >most radical innovation is to cut Ferguson's dial in half (that's how I >imagine it, any way!) and to transpose the two halves. This permits his >dial to operate from earliest sunrise to full sunset. The shadows are >cast by two vertical gnomons 7.5 cm (3 in) high on the outer edges of the >dial. The dial itself is not a quarter of a cylindrical surface. After >MUCH fruitless curve fitting, I've concluded that Dodwell probably used >what Piet Hein later christened a 'superellipse', that is a curve which >uses the formula for a circle but with exponents higher than 2, in this >case, 2.5 (i.e. x ^ 2.5+ y ^ 2.5 = 1). There is a gain and a loss in >this. The dial is much easier to read. But the distances between hours >are unequal and thus this most difficult aspect of construction is made >additionally complicated. I've made a model of the dial using large >diameter PVC pipe and find it generally satisfactory. Dodwell has also >displaced the hour lines to adjust for the 4 degrees that Adelaide is to >the west of its--non-standard--Standard time longitude of 142 degrees 30 >minutes. Dodwell's other innovation is to dispense with the nodus and >unfold the EoT curve in much the fashion that Art Carlson suggested is >possible. The months march in succession down the dial and only the >contructor has to worry about which particular point on the edge of the >gnomon is casting the shadow on any particular day.
