Hello Roger,

To have an accurate reading for the times of sunset and sunrise on an
analemmatic sundial draw arcs through the focus points of the ellipse and a
date point.
You may see this in the attached picture.

Best wishes, Fer.

Fer J. de Vries
[EMAIL PROTECTED]
http://www.iae.nl/users/ferdv/
Eindhoven, Netherlands
lat.  51:30 N      long.  5:30 E

----- Original Message -----
From: "Roger Bailey" <[EMAIL PROTECTED]>
To: "Sundial Mail List" <[email protected]>; "Steve Lelievre"
<[EMAIL PROTECTED]>
Cc: "Mike Deamicis-Roberts" <[EMAIL PROTECTED]>
Sent: Friday, January 11, 2002 5:35 PM
Subject: Seasonal Sunrise Marker


> The original copy of this note did not get to/from the sundial mailing
list.
> I have changed the address on this one and hope for better results. I
> apologize if you happen to receive multiple copies. If you do, keep one
and
> pass the other on to your grandchildren.
>
> The original note follows. Roger Bailey
>
> Hi Steve et al,
>
> You guessed correctly on the sunrise marker. Have a look at the little pdf
> file attached showing the seasonal sunrise marker on Mike
Deamicis-Roberts'
> analemmatic dial.
>
> The seasonal sunrise marker is a point on the east west axis of the
> analemmatic dial that is used in combination with the date line of the
> Zodiac table to show when and where the sun rises throughout the year.
Stand
> on the sunrise marker point and view across date marks on the zodiac to
see
> where the sun will rise on that date. Or stand on the date mark on the
> Zodiac and view past the sunrise marker to see the time of sunrise on that
> date. Use a string from the date through the marker to the hour ellipse to
> convert the dial into a sunrise calculator. What could be easier?
>
> The red line on the sketch shows the summer solstice sunrise at 4:49 AM at
N
> 60.2 East for Mike's latitude of N 36.8. In your mind, rotate the red line
> around the marker point to determine the time and direction of other
sunrise
> dates. For sunsets, use the marker on the east side of the dial. Things
are
> symmetrical. These two marker points provide an excellent new feature for
> analemmatic sundials, the ability to show where and when the sun rises and
> sets and how this changes throughout the year.
>
> This idea started with Mike's "Seasonal Sundial" posting to the sundial
> mailing list last fall. He was looking for a sundial design that would
show
> the cycle of the seasons through the year. I proposed an analemmatic dial
> with distant solstice sunrise markers like a "medicine wheel". This did
not
> suit Mike's topography. He proposed a marker point within the sundial that
> could be used with the date table to show sunrise phenomenon. I had not
> heard of such a point but did the math to reduce the idea to practice. As
> you can see, Mike's brilliant idea works!
>
> Here are the steps to calculate where to put the seasonal markers on any
> analemmatic dial. All you have to do is determine where the red line
crosses
> the axis. This calculation could be done for any date but the error is
least
> if you use the solstice, either the summer or winter (they are
symmetrical).
>
> 1. Calculate the azimuth of the solstice sunrise for your latitude. When
the
> altitude is zero (sunrise), the azimuth (Az) given by Cos (Az) = Sin (Dec)
/
> Cos (Lat). In Mike's case Cos (Az) = Sin 23.44º /Cos 36.8º = .497 so the
> sunrise azimuth, east of north is Az = 60.2º.
>
> 2. Solve the right angle triangle between the two axes and the red line to
> find the marker point on the E/W axis. Start with the zodiac distance on
the
> N/S axis which is size (or the semimajor axis) x Cos Lat x Tan Dec. In
> Mike's case of a 9 meter dial, the semimajor axis is 4.5, so the solstice
> zodiac distance is 4.5 x Cos 36.8º x Tan 23.44º = 1.562 meters. From the
> triangle geometry, the distance to the marker on the E/W axis is 1.562 x
Tan
> (Az) or 2.727 meters.
>
> How accurate is it? The mathematics are not exact as the trig
relationships
> are not the same, but they are pretty close. Both the zodiac distance are
> functions of latitude and declination but not the same functions. The
> declination distance relationship of the zodiac is slightly different for
> the azimuth derivation. From the layout method, the error is zero at the
> solstices. It is also zero at the equinoxes when the sun rises due east.
> There is a sinusoidal periodic error for dates in between. This error
> increases with latitude. In Mike's case the maximum error is only +/-2.3%.
> At my latitude, N51, the maximum error increases to +/-6.5% so the
> relationship is only approximately correct. With sundials we are used to
> this level of accuracy. Corrections for the equation of time and leap
years
> are of similar magnitude.
>
> All these calculations are based on the theoretical sunrise when the
> calculated altitude of the sun is zero. Refraction and semidiameter affect
> the real view. If you have a perfect horizon (ocean view), allow the sun
to
> rise one full diameter, from the horizon to the lower limb to correct for
> semidiameter and average refraction. For other locations, you will have to
> correct for the horizon pollution. In Mike's case there is a devilish
range
> of mountains affecting his horizon by up about 5 degrees. We are working
at
> corrections for this.
>
> My conclusion is that the simple addition of these markers to the design
of
> analemmatic dials adds a lot to their function of demonstrating the cycles
> of the sun with the seasons.
>
> Roger Bailey
> Walking Shadow Designs
> N 51  W 115
>
> -----Original Message-----
> From: Steve Lelievre [mailto:[EMAIL PROTECTED]
> Sent: January 8, 2002 7:03 PM
> To: Roger Bailey
> Subject: Re: Garden/Human Sundial
>
>
> Roger,
>
> What precisely is the "the sunrise seasonal marker proposed by Mike
> Deamicis-Roberts"? I'm guessing it's some sort of mark or curve on the
dial,
> which gives a line from today's place on the date scale to the a place on
> the ellipse showing the corresponding sunrise time, but I've not heard of
it
> before.
>
> Thanks, Steve
>
>
>


Attachment converted: Macintosh HD:analemdial.gif (GIFf/JVWR) (0003A004)

Reply via email to