I think I have always been inspired by the heavens above. My first true exploration into this began when I was young helping my father with the construction of his 6" telescope. He had ground the mirror himself and my contribution was the making of a mount for it. With completion of that telescope came many glorious views. The two things which really fired my imagination were the breathtakingly close views of the cratered moon and especially seeing Saturn with it's multitude of rings. In a way it seemed so foreign and bizarre and yet there it was right above me. These experiences were my first immersion into this wonderful "world".

Since then this interest has become inextricably linked with my interest in horology (clocks). If anything has mass and is in proximity to something else it will orbit. If it orbits it has a period. If it has a period a clock can replicate it. Thus goes the quest of modeling the heavens. It is age old. Geared mechanisms modeling the heavens (such as the Antikythera mechanism) have been built for centuries. I have yet to make anything like an orrery or an astronomical clock but is my greatest hope to do so some day. Below (and in my Wallingford clock section) are some of my first feeble steps toward that end.

Some Star Trail Photography

Here are the results of some ongoing experiments with star trail photography. If you don't know what star trail photography is it is a photographic representation of the Earth spinning. One takes a time exposure of the night sky. Since the Earth will naturally move during that exposure time the effect is that the stars in the night sky will appear as "trails", generally circular about the pole. The effect if captured right can produce some remarkable pictures that truly instill the sense of passing time.

My first attempt into this foray was unsuccessful. I did not get the exposure settings right at all (greatly underexposed) and the result was much like the present background. After doing some further research I have found the following table helpful as a starting point.

Below is my second attempt. In this case I used the table to set the exposure. The exposure started at about 9:30pm and was facing north. The sky was darker than it appears in the photograph. I suspect it was picking up a lot of dusk light as in summer time the sun dips below the northern horizon at a low angle. At the end of this shot when I went out to close the shutter the I noticed the lens was heavily covered in dew. In my opinion the star trails were shorter than I was looking for.

Canon A1 with FD 24mm/f2.8 lens at 400ASA (Kodak Max), Exposure 2hr @ f4.

Given the issue with the dew I made a dew prevention optic heater base on the design of Mark Kaye. I made it to be powered from the cigarette lighter of a vehicle. Below are my 3rd and 4th attempts. In this case in there was a haze on the horizon if I shot in the direction above so I repositioned my set up to shoot facing east. At the time of the shooting the optic heater wasn't complete so I simply wrapped the heater element around the lens, covered it with insulating shop rags and tied it up with a string. The shot on the left was set for 3 hours but at the end of the exposure I found the shutter had already closed as the camera battery had run out. The shot on the right was taken the following night with a fresh battery and it went the full 3 hours. I actually fell asleep waiting for the 3 hours to go by and when I woke up noticed that the moon had risen. This is not the effect I wanted but it is interesting to note that it did not completely ruin the shot like I thought it might. The star trails are getting better but I still think an even longer exposure is needed. Having captured this shot I should note that the camera was unable to take pictures the following day as the new battery had completely drained taking the picture. The optic heater worked flawlessly and the lens was completely dry. As a comparison the tripod was wet with dew droplets.

Left - Canon A1 with FD 24mm/f2.8 lens at 400ASA (Kodak Max), Exposure 3hr? @ f5.6 Right - Canon A1 with FD 24mm/f2.8 lens at 400ASA (Kodak Max), Exposure 3hr @ f5.6.

As of fall 2003 I was in the process of making a 12V to 6V regulator to power the camera through a machined, simulated, battery (basically a plastic slug with brass ends wired to the regulator). As exposures get longer this power source was needed especially as a battery is $12 and I get one picture out of it. I also experimented if a clock timer could be used to shut down the exposure once preset. The plan was to allow me to set up the exposure in the evening and retrieve the camera in the morning without having to stay awake for several hours in complete darkness. The following few paragraphs show my design but I'm sure it could be adapted to many arrangements.

The above pictures show the cigarette lighter powered heater core (left). The heater has yet to be sewn into a lens blanket. The heater core is basically a strip of nichrome wire (covered in yellow heatshrink). The wire is of a length that will cause about 5W dissipation (12Vx12V/5W = 28.8 ohms). The blanket is an exercise wrist weight cut open, stripped of its weights and the heater sewn in serpentine fashion (the middle two pictures above show this in progress). Once the nichrome was sewn in two strips 0.14" thick (each) of neoprene are encapsulated in the heater blanket as insulators. The ends of the nichrome are soldered to the 12V cigarette lighter cord. Lens dew now solved!

The "DarylCell" battery (left) is made of nylon drilled through at 1/8" with a central hole at 90° for wire egress. Each end was then drilled 1/4" dia by 1/4" deep. 1/4" brass rod was cut 0.280" long, slightly rounded at one end and drilled with a #59drill at the other end. A red & black wire was fed into the 90° hole and out each end. Each wire was then soldered to the brass end piece's #59 hole. Then end pieces were then pressed into the nylon until the length matched a real battery. This battery simulator was then connected to a TO-220 linear regulator circuit that receives 12V from the cigarette lighter cord and steps it down to 6V for the "DarylCell" camera battery. The regulator is mounted to a piece of steel shaped to hold the battery in place while letting the "DarylCell" battery wire egress out of the camera to the regulator. Since the camera battery door is left open in use, to allow the battery wire out, I make use of the removable plastic handgrip cap (by Canon - 2nd left picture above) which screws in place to hold and retain everything in place (right picture) i.e. the regulator and plate are retained by Canon's own handgrip piece. Camera power now solved! Not only that but a previous problem (battery drain) has now been turned into an opportunity (automatic shutdown) as you'll see next.

Above is the device (the white thing) which makes the timing easy. It is an Intermatic battery powered, digital, wall switch timer. It is intended to switch AC power ON and OFF at intervals you program (or set) into the clock. The actual switch inside is an isolated form-A relay contact. Since the DC levels are low it works fine in this application. I picked it up at Home Hardware for about $45 and put it into a metal box. At Canadian Tire I bought a cigarette lighter extension cord. I made it so the male end is the input and the female end (for the heater/"DarylCell" arrangement to plug into) is the output. I cut one lead in the middle and spliced in the timer so as to switch that lead ON and OFF. If you have trouble staying awake to stop the exposure this thing is the "cat's ass". Canon's A1 needs the battery to hold the shutter open. Kill power and the exposure is stopped which is exactly what I want here. I also use it to stop the heater as well so as not to drain the car battery too much.

On the left you can see the whole power/heater arrangement installed while on the right the camera is set-up for exposure. The timer (male) is plugged into the cigarette lighter in the car and sits on the front seat. The tripod is set up close to the car and the heater arrangement (male) is plugged into the timer (female) inside the car. To take a picture you first make sure the timer is ON and the shut-off time is programmed in. While it's still light I will compose the direction of the shot with foreground objects and set the focus to infinity, set the camera exposure time to "B" (bulb). When it's dark and you're ready to go you plug in the heater, confirm the camera is powered, do a last verify all things are OK and close the eyepiece shutter to prevent light back feeding in through the eye-piece. When the start time has arrived you use the cable release to lock the shutter open. At this point you can walk away and retrieve everything in the morning if you want to. For those interested the basic schematic is here.

The following shows my next attempt (#5). While I still consider it a failure it is interesting. It also demonstrated to me that all the above equipment works excellently. I brought the above equipment with me on my 2004 vacation up north in the hope of some good photographic opportunities. Unfortunately we were either in towns or it was overcast. The day the picture was taken it had been cloudy all that day so I never thought I would be able to attempt a shot. While sitting at the picnic table by the campfire having an enjoyable discussion with my wife I noticed that all of a sudden the stars were visible. I quickly set up for a long exposure and let it go then went back to the table to continue the discussion. We later packed it in before the shot was completed. The first thing I notice in the exposure (apart from the splotches from my crappy scanner) was it was less clear than I thought. The second is the smaller aperture (f8) seems to make finer (more narrow) star trails however at over 5 hours they are certainly long now. Another thing is while it looked dark at the time there is obviously some settlement across the river which caused some unanticipated light pollution (glow & red lights). This is getting to be a recurrent problem and one I will have to devote more time to avoiding in future. I think in this location however if it wasn't overcast on the horizon the settlement light-glow would have been greatly reduced.The lighting of the tree and the tent by the fire is neat. The campfire was in reality low but looks like a blowing hot foundry fire in the picture yet I'm surprised it didn't over-expose the 5.25 hour shot. Finally my wife and I being ghostly visible in the shot by the picnic table are another interesting indication of the passing of time. For a "what the hell lets give it a try" shot it does have some interesting things in it to learn from .

Canon A1 with FD 24mm/f2.8 lens at 400ASA (Kodak Max), Exposure 5.25hr @ f8.

 

To be continued...

Handy Links

The Ottawa Clear Sky Clock above (Dark is good/White is bad - click blocks for map. Many other good links here!)
Ottawa Light Polution Map (Montreal is the bright spot on the extreme right, Canada/US boarder is the white line)
Complete Sun & Moon Data (Canadians go down to Form B)

Current phase of the Moon updated every 4 hours - but only if you refresh your browser ;-)

 

 

An 18" Brass Horizontal Sundial

Years later after I had left school my mother had expressed a desire for a sundial while we were looking through a Harrowsmith magazine. I decided I would make a pair of them giving her one for Christmas. My plan started off as a simple time scale. It was while researching them that I found out how accurate they can be and how they can show many of the motions of the Earth around the sun. So, as with most of my projects, what started out as a simple time scale soon blossomed into a full featured sundial. I kept learning by adding each bit of "furniture" (a sundial term for features) and understanding its background. By the time the design was completed I had a pretty sound intuitive understanding of the intricacies of the yearly solar movement and how it appeared in the sky.

I did the graphic design on a fridge size computer (this was back in the 80's) which would spit it's 2D output to an HP pen plotter. When the design was complete it would take 45 minutes to come up on the screen! I plotted out the nearly completed design and after some final hand/pen work I sent it out to get positives and negatives made. These were large as this was to be an 18" diameter dial (needed for minute resolution). After obtaining 3/8" thick brass plate for the dials and 1/2" thick brass plate for the gnomens I set about getting the gnomen machined at work (this predated my own metalworking capability but helped create my desire to get into it). The gnomen had to have a 46° angle machined into it. Additionally it was to have a nodus (notch) in the style (top edge). The shadow of this notch would be used to track the position of the Sun on the dial (see the map on the center picture above). After the gnomens were machined I set into etching the design work into the brass.

I had to make a photo-chemical etching studio at home. One first polishes the brass then under red light conditions applies the liquid (honey like) photo resist spreading it evenly over the entire surface. After waiting for it to dry you then give it a gentle bake in the oven. When dry like this it is very light sensitive. At this point you lay on your positive artwork and expose the piece to ultraviolet light, in my case 6 minutes at 2ft. After exposure you develop it by placing the piece in a tub and washing the developer back and forth over it for a predetermined period of time (2 minutes in my case) followed by a rinse. The photo-resist is very soft and susceptible at this point. This point was the source of most of my imperfections as it is difficult to manipulate the sheer mass of the plate through the processes when the resist gets so vulnerable. This also made double sided etching of the gnomen a screaming bitch. After it is developed it is carefully dried by another gentle bake in the oven. At this point there is a plastic like coating over the entire piece except were the artwork had a black line. Where ever there is a black line the resist does not get exposed to UV and hence is washed away in the developing process leaving exposed brass. All that is required now is to put the piece back in a tub and wash an etchant over it (ferric chloride or ammonium persulphate). This will eat into the exposed brass. I went for a deep 0.040" etch. At this point the process gets fun because now you have a plastic coating with etched lines. You simply spray paint the entire piece black then after it's dry you get a flat block with emery paper and begin to sand. Its fun to watch the finished pattern jump out at you as you sand off the painted resist. The paint bonds well to the freshly etched lines and being in a groove is protected from sanding while the painted resist quickly sluffs off. You just progress to finer and finer emery until the brass is shiny and polished. Then you spray a lacquer coat to protect the exposed brass.

The pictures above show the dial. It's nearly 15 years old now so its lost it's original luster but the deep etch has held up remarkably well. It has the following features: minute resolution, noon marks for cites around the world, equation of time correction, lines of declination, latitude marks, date, sign that the sun is in, day length, solar azimuth or compass heading, solar altitude, time and direction of sunrise and sunset, corrections for telling time by the moon, a map projection which, used in conjunction with the nodus on the style, allows the sun's progress throughout the day to be watched.

As I said, the second dial (not shown) was given to my parents as a Christmas gift and to this day it sits in one of the gardens at their home. Being outdoors for 15 years the lacquer has long since disappeared and hence their dial has a darker colour but, given the deep etch, the time is still easily read.

It still amazes me that an immovable lump of metal (albeit with carefully marked lines) can give so much astronomical information.

 

 

A 10" Astrolabe

If you are into astronomy and you don't know how an Astrolabe works you should! It is essentially an analog computer for the heavens and it's effect is absolutely magical once you understand it.

My involvement started one day while I was perusing a book store sale I came across a $2 book about time. It had the most remarkable instrument on the cover, an astrolabe. I bought the book. The more I looked at the cover the more I was driven to understand how this instrument worked. While I could figure out much by looking at it there were a few things which continued to baffle me. Eventually I bought another book, Colin Ronan's "The Practical Astronomer" (an inexpensive but great book), which finally gave me the basic understanding I was looking for. Once I grasped the principles I simply had to make one. I mean if I could make a sundial why couldn't I make one of these. Again the design started off simple but after some discussion on rec.crafts.metalworking I was put onto yet another book, Harold N Saunder's "All The Astrolabes". Here was a book which described nearly everything including many other types of astrolabes other than the planispheric type I had undertaken. While out of print I would highly recommend this book if you want to fully understand astrolabes. The only thing it did not explain, surprisingly, was how to create the date/zodiac translation scales. I reasoned that out on my own.

I started into the design with a newly obtained MAC program called Generic CAD. A rudimentary 2D CAD program. It had one basic flaw. In order to make text follow curves each letter of text had to be placed one by one. It was made more difficult by the fact that the letter of text was within 4 corner squares of an invisible text box. When you printed it out the letter would drop to the bottom of the text box so I had to predict in advance where the letter would be when it printed out. That's one thing when you have under a 100 letters but when you have thousands upon thousands it gets tricky.

I won't get into how an astrolabe works, you can read about that here but suffice it to say armed with the knowledge contained in the Ronan & Saunders books the design blossomed to a great level of complexity. Instead of a single solar hand my design has three. The Sun, Moon and Nodes. The artwork for the back of the matter can be seen in the background of my creative pursuits page but that is only the beginning. There are 18 different latitude plates as well as a horizon plate for determining latitude and of course the movable rete. There is such a job ahead in etching that I have not realized this piece yet. In addition they have discontinued the photo-resist I used to buy. The etching as a result will likely have to be professionally done. The design has been tested, however, by plotting out versions and attaching them to cardboard. Happiness is seeing your design work in Canada and, with a different plate, in St Lucia (what a night sky!!) or having it predict an eclipse and that it would be a grazing one at that!

 

 

A Vertical Declining Dial

This is a dial I started in the winter of 2001/02. I actually started in the fall of 2001 by determining my latitude, longitude and the amount the wall I wanted to mount the sundial to declined toward the west. The first two became trivial with the purchase of a inexpensive GPS unit. The latter took several readings over a week. I used the approach of measuring the shadow cast by a stick perpendicular to a wall at a specific time & date. Once the shadow is measured and the time, date, latitude and longitude are known the angle of the wall from true south can be calculated. I made an Excel spreadsheet program to do the number crunching. In the end I had 3 measurements which provided wall declinations within 0.25° and two of those within 0.01°.

I made another Excel spreadsheet which calculated the hour line angles, the angle of the shadow casting style and it's angle from the vertical noon line. I wanted to add declination lines so I modeled everything on Pro/E (industrial CAD software) but found while it worked it was cumbersome if I wanted an analemma. I then went back to trying to integrate declination information into my Excel sheet. It was while searching for some of the basic trigonometric formula that I bumped into the Shadows program which did nearly everything I was trying to do. The more I played with it the more I was impressed to the point where I abandoned my similar but non-graphic approach and relied on the program to provide the basic outlay for construction. It even factors in the longitudinal distance from your standard meridian. By doing this you have the *full* conversion from local apparent (sundial) time to watch time. The artwork design is a conglomeration of many things I liked about other dials I had seen such as the Sun emblem and the "time ribbon". Others are similar in concept but unique to this dial such as the Ottawa River scene. To my knowledge the end of the declination "web" terminating in a sunset followed by night is a first (some others simply flat-line the web at the sunset point). The idea here was that at a glance you could tell approximately when civil, nautical & astronomical twilight occur which would be useful for star-gazing. The constellation figures on the declination "web" are ala Hevelius.

My plan was to mount the dial below a front bedroom window. Given the existing trim there is a natural place for it below that window so that led to it's basic dimensions of 43.25" x 44". The idea was to make it from some extra 1/2" Baltic birch plywood I obtained when making my CD stands. The problem I had then was to get a 1:1 plot from the Shadows program for the size of dial I wanted. In the end it was Gilmore Reproductions that helped me out with the large plot. Then due to other obligations I set this project aside for a year. This is typical. My projects do tend to leap frog one over the many others before leaping again.

When I got back into it around Christmas 2002 I decided to begin painting the dial The paint I chose was marine paint. Expensive but tough! I wanted this dial to stand up to the elements for a long time. I decided on Hatteras White for the declination "web", red for the line work within the "web", a light blue for the sky background, yellow for the sun, black for the line work, numbers and general tinting while white was chosen for the hour ribbon and for lightening other colours. Finally, for the trees etc in the lower right scene, I already had some Malachy green I had used previously on my Jacob Kayak. The only deviation from marine paint was the sunset, with its multitude of colours, where I used model enamel.

In getting ready to paint the first thing that was discovered was that the centerlines of the Shadows plot didn't match the corresponding centerlines of the dial. After some investigation I found that while all sides of the dial were parallel to each other the dial was slightly trapezoidal (not square). Slight as the problem was it had to be solved as the layout on the dial had to be true if the dial was to work properly. My solution was to use my beam compass and revert to Euclidian geometry. Once the dial was trued the declination web was transferred from the plot to the dial. With that in place I purchased a splendid 3M overhead projector from Advantage Audio Visual Rentals (Thanks Jeff!) and used it to transfer other artwork such as the time ribbon and the Ottawa River scene etc to the dial. After that it was basically many sessions of painting and projecting and more painting. I had never done "artistic" painting before so this was a learning experience.

I decided early on that the gnomen etc was to be made of brass for two main reasons. First I didn't want rust stains developing over the dial and secondly it would be easy to fabricate using my torches & SIL-FOS solder. It would also be easy to machine the sun emblem that I had planned to cast a shadow on the declination "web". I picked up the brass for it at Loucon Metal where I normally pick up most of my steel. While I originally designed a large opaque perforated disc to be the sun emblem I later changed the design to the inverse of that (a radiant cross-haired ring) as I found it had a more realistic and brighter look.

.

to be continued.....