Starnamer's Blog

Welcome to Starnamer's Blog! Hope you enjoy your stay!

Thursday, September 08, 2005

Astrophotos using Barn Door Tracker (2nd Version)


Introduction

Sometime after I have completed the 2nd version modifications of my barn door tracker, I finally got out to give it a test. My main subjects were still constellations with 1 series of moon shots. Each shot required minutes of long
exposure, so I was not able to complete the whole roll in a single night and had to do it in 2 separate nights. In between the 2 nights was a very long break, close to a month's time. Guess I was too lazy to finish the roll.

For each object, I had bracketed the shots at different exposure lengths to increase the chances of getting at least one or two good shots in the series. The shots shown here are in the exact chronological order as I have taken them. Sometimes the exposure times might not increase in the correct order due to accidental premature release of the shutter.

After finally finishing roll, I sent it to the photo lab for development directly into digital format. It is much cheaper and more convenient for me that way. The lab scanned the negatives in high resolution with each JPEG file taking up roughly 5MB of disk space. The images shown here are in their original raw versions without any proper post processing using Photoshop.

First, some overall information for the first night.

Date: 14th July 2005
Location: A large open space besides a neighbourhood park, away from flats. However, there were plenty of streetlights nearby which were the biggest potential sources of light pollution.
Camera: Yashica FX-3 Super Manual SLR
Lenses: Yashica 50mm manual focus prime lens F1.9 and Makinon 500mm F8 manual focus mirror lens.
Exposure: All exposures are taken in bulb mode.
Tracking: Manual Barn Door Tracker with version 2 polar alignment system.
Weather: Clear skies with some clouds. It was raining just that afternoon which helped to clear the skies quite a bit.
Other accessories: Cable Release. Tripod. Binoculars. My handy pocket astronomy guidebook with a mini starchart. High-end compass for polar alignment. Screwdriver (in case I need to tighten some screws on the tracker). The only thing I forgot was a torchlight for illumination.


Gibbous Moon

There happened to be a gibbous moon in the sky that night. A perfect chance for me to give a try on shooting the moon. I have seen lots of beautiful moon photos with well-defined craters. I would love to capture that myself. A gibbous moon was also good since it did not produce too much light pollution as compared to a full moon. It was also good since the moonlight provided some illumination for me, my tracker and my watch. That would compensate for the missing torchlight!

Since the exposure lengths were pretty short, I did not use any tracking for the moon shots. I simply took stationary shots of the moon.

Starting time for 1st shot in the series: 10:32pm
Lens: 500mm mirror lens
Aperture: F8

Exposure: 10s


Exposure: 20s



Exposure: 5s


As you can see, on the whole, the moon in all 3 shots were terribly overexposed. It is most probably because I had forgotten that I was using a fast film (ISO 400), combined with a super long focal length of 500mm. I should have shortened the exposure lengths much further, probably to fractions of a second. Looks like I still have a lot to learn about taking moon shots.

If you strain your eyes a bit, you would have noticed some patches of light diagonally above the bright moon. I did some post-processing using Photoshop by turning up the brightness and found the cause of the weird "phenomenon". Below is how the processed image looked like. It turns out that the patches were actually ghost images of the moon. It is actually a type of optical effect which happens when you position a bright object, like the moon, just off the centre of the picture. Due to internal reflections of the light rays within the lense or camera, a "ghost" image of the bright object will be seen diagonally across the centre at the same distance from it. So, to prevent such effects, probably I should have placed the moon directly in the centre.




Sagittarius (The Archerer) and partly Scorpio (The Scorpion)

Starting time for 1st shot in the series: 10:52pm
Lens: 50mm standard lens
Aperture: F2.8

Exposure: 30s
I actually wanted to take a 1 or 2 minute shot using the tracker but I accidentally released the cable release. Anyway, may as well, since there was a passing cloud which spoilt the shot.


Exposure: 2 min



Sagittarius (The Archerer)

Exposure: 4 min


Exposure: 3min 45s
Again, premature stopping due to accidental release of shutter.


Exposure: 8 min



Scorpio (The Scorpion)

Starting time for 1st shot in the series: 11:24pm
Lens: 50mm standard lens
Aperture: F2.8

Exposure: 2 min


Exposure: 4 min


Exposure: 8 min



Cygnus (The Swan) and Lyra (The Harp)

The following shots were taken on another night.

Starting time for 1st shot in the series: 10:00pm
Lens: 50mm standard lens
Aperture: F1.9

Exposure: 13s? (Must be either poor documentation or accident again!)


Exposure: 4 min


Exposure: 8 min



Close-up shots of Antares (Alpha Scorpii)

Starting time for 1st shot in the series: 10:31pm
Lens: 500mm mirror lens
Aperture: F8

I am sure you will be puzzled by what the following photos are showing. They are not flashes of fires. They are close up shots of Antares (Alpha Scorpio, the brightest star in Scorpio) and another nearby star of Scorpio. The photos turned out to be smudges because of the high-power 500mm lens and the instability of the setup.

I accidentally discovered the shakes when I was looking into the camera's viewfinder to ensure that the stars were in view and focus. I noticed that even a minor vibration of the cable release sent the image shaking terribly. Although I knew the photos would most probably turn out bad, I went ahead to try. I was right! Looks like the setup is still not stable enough for high-magnification shots.

Exposure: 2 min


Exposure: 4 min


Exposure: 8 min



Cygnus (The Swan) and Lyra (The Harp) Part II

Lens: 50mm standard lens
Aperture: F1.9

Exposure: 1 min


Exposure: 2 min


Exposure: 4 min



Conclusion

Apart from the failed moon and Antares close-up shots, all other constellation shots turned out to be quite good. In fact, all of them showed little or no star trailing. This was much better than the photos in the first series (see http://astro.jconserv.net/viewtopic.php?t=57 ) where only a handful of shots out of the entire series turned out well. This shows that the 2nd version of the polar alignment system was really better.

The photos also showed less vignetting effects. This might be due to the lighting conditions at that time and probably the use of a faster film (ISO400).

All in all, the shots were great and the polar alignment system was effective. I will proceed to perform post-processing to the shots to enhance them and probably improve the tracker further.


Tuesday, September 06, 2005

Astrophotos using Barn Door Tracker (1st Version)


The Setting

After I completed my 1st version of the barn door tracker, I quickly took it on my first guided astrophotography field trip. I can't really remember how long ago was that. I guess it was a few years back. The location I set up my stuff was an open space near my block. At that time, the open space was still undeveloped and was relatively dark. However, the expressway was just nearby, so the streetlights and lights from nearby blocks became the largest sources of light pollution. Today, it has become a neighbourhood park with plenty of lamps. The pollution is obviously worse.

The general setup was as follows:
Camera: Yashica FX-3 Super (2nd-hand Manual SLR)
Lens: Yashica 50mm 1:1.9 prime lens
Film: Fujifilm Velvia ISO50 slide film
Aperture: F1.9
Exposure Time: Around 2 to 4 mins, bracketed.

As the 50mm lens was not powerful enough to take any deep sky objects, I could only focus on taking constellations. For each constellation, I took 3 bracketed shots. I can't remember the exact exposure times as I did not properly document the whole process. I guess they should be 1 min, 2 min and 4 min. Bracketing helped me to get the correct exposure with at least 1 of the 3 shots. It also helped me to test the accuracy of the tracking.

I used slide films because I was attending a photography field trip for my basic photography course at that time. The instructor wanted us to take night scenes using slide films. I couldn't finish all the shots, so I decided to use the rest for this purpose. My instructor told me that slide films have the best resolution and picture quality given its slow ISO rating and extremely fine grain. The slow ISO rating also cuts down on light pollution from surrounding light. This is because, as my instructor taught, dim light sources (noise and light pollution) are captured more slowly on the slides than strong light sources (the stars and nebulae). This increases the contrast as well.

However, some articles on the Internet argued that a slow film means the tracker must be more accurate and one must take longer exposures. This will cause a side effect called reciprocal effect. On the other hand, for higher ISO films stars are captured faster. So, less time is needed to capture the same amount of details which puts less stress on the tracker. It also prevents the reciprocal effect from affecting the shots. Seemed confusing but anyway, the only way to find out was to try it out for myself.


It was only recently that I decided to send the slides to my favourite photo lab to be scanned and digitised into high resolution JPEG image files for long term storage purposes.
Each file is about 10 to 15MB in size. Then, I applied some simple post-processing techniques I have learnt on the Internet such as sharpening, adjustment of brightness, contrast, setting the black and white points using Adobe Photoshop. There are more advanced techniques like removing vignetting and stacking multiple images which I have yet to master.

Surprisely, the final results were quite ok (at least to myself). Most of the stars in the photos look quite point-like which means the tracker was constructed accurately enough although the bubble mechanism was potentially and terribly inaccurate. The slides showed vivid star colours for some of the shots. Some shots had foggy areas which were caused by light pollution.

In the following sections, I will display and describe a few of the best shots. For the benefit of those who are not familiar with the constellations, I have created corresponding outlined versions of each shots to show the constellation outlines. I have also added their corresponding illustrations to help you visualise how they would look like. Please note that the illustrations are not international standards. There can be more than 1 version of the illustrations.


Auriga (The Charioteer)

This is one of the 2 best constellation shots I have taken in the slide series. The other is Orion. The stars are point-like with no visible trace of star trailing. The colours of the stars after post-processing are quite vivid. Light pollution and vignetting effects are acceptably low. There are supposedly a few Messier objects here but can't seem to see them.



Note: If you have noticed a small purple patch slightly below the centre of the photo, don't mistake it for a nebula or some new astronomical object. Some of the slides including this one had turned slightly mouldy. Guess I didn't store them properly.

Below is the outlined version of the photo.



Below is the illustration obtained from
http://stardate.org/nightsky/constellations/auriga.html.




Orion (The Hunter) and Lepus (The Hare)

This is the favourite constellation which fascinated me when I was young. Thankfully, the stars looked very point-like and colourful. This, together with Auriga, are my best shots in the series. You can also see the Orion Nebula, M42, clearly. Below Orion is part of the constellation, Lepus (The Hare).



Below is the outlined version of the constellation.



Below is the illustration obtained from
http://stardate.org/nightsky/constellations/orion.html, showing a different outline.




Gemini (The Twins)

This photo was the best among the shots of Gemini. You can see obvious star trailing. The tracker must have been pretty misaligned at that time. The vignetting effect was pretty bad in the central zone especially after Photoshop processing. The colours were still ok.



Below is the outlined version.



Below is the illustration obtained from
http://stardate.org/nightsky/constellations/gemini.html.




Canis Major (The Greater Dog)

The brightest star in the photo, which is the dog's neck or head is called "Sirius". In Chinese, we called it "Heavenly Wolf Star". It is the brightest star in the night sky.

If you are imaginative enough, you should be able to make out the body, legs and tail of the dog using the other brighter stars. Tracking wasn't very accurate as you can see small trails forming. If you are observant enough, you will noticed that the stars on the left look more point-like than those on the centre and right. This is probably a case of misalignment of the azimuth (magnetic bearing). Light pollution was acceptable though.



Here's the outlined version of the above photo. This version shows a more clearly defined shape of a dog than other versions. I like this version most.



Here's another shot. This one should have a longer exposure than the previous one as the light pollution and stars are much brighter unless I have used the wrong amount of Photoshop adjustments for these 2 shots.



Here's the outlined version of the above photo. It shows another version of the outline, a much simplistic one. Sufficient to allow one to identify the shape of a dog.



Below is yet another shot of the constellation in portrait mode.



Below is the outlined version.



Below is the illustration obtained from
http://stardate.org/nightsky/constellations/canis_major.html. Yet another outline version.





Taurus (The Bull), Jupiter, Saturn (or Mars?) and Pleiades (The Seven Sisters)

For those who are not familiar with this constellation, the "V" shape group of stars near the centre of the photo to the right of the brightest object is the "head" of the Taurus (The Bull). It is a star cluster called "The Hyades". The brightest yellowish star in the group is Aldebaran (Alpha Tau), the brightest star in Taurus.

The brightest object besides the Hyades is supposedly Jupiter. The other second-brightest object below could be Saturn or even Mars. These 2 must be planets because you will not find any such bright stars in that area of the sky if you refer to a star chart.

The compact group of stars at the right of Jupiter is called "The Pleiades" or some might call them "The Seven Sisters". It's a cluster of young blue stars which you can see using even just your bare eyes. There are of course more than 7 stars in the group.

Note: The big purple patch above the Hyades cluster is a mouldy patch of fungus on the slide film. It is not a deep sky object.



Below is the outlined version. This is my favourite version as it closely resembles the shape of a bull.



Below is another try on Taurus using landscape mode. Pretty bad tracking. The yellowish triangle at the top right hand corner was part of a housing block.



Below is the outlined version.



Below is the illustration obtained from
http://stardate.org/nightsky/constellations/taurus.html, showing another version of the outline.




Crux (The Southern Cross) and Centaurus (The Centaur)

This shot of Crux and Centaurus did not belong to the same slide series as above. It was taken on another night at another location. Can't remember when and where it was taken. However, the settings were pretty much the same except the film used was Fujifilm ISO100 or ISO200 negative film.



The Crux is slightly to the right of the centre. The 2 bright stars to the left are the 2 brightest stars of Centaurus.

As you can see, the tracking was still ok but the light pollution was terrible. Plenty of clouds too. If you noticed carefully, the leftmost star belonging to Centaurus seemed to be more point-like than the 4 stars of Crux, again probably due to misalignment of the azimuth.

Below is the outlined version of the photo.



This webpage, http://www.astrosurf.com/jwisn/centaurus.htm, shows a big illustration of Centaurus and Crux. Too big. So, I decided not to place it here.

Monday, September 05, 2005

My Single-Arm Manually-Driven Barn Door Star Tracker (Scotch Mount)


What's a barn door tracker?


For those who are familiar with it, a barn door tracker is also known as "Scotch Mount" or "Haig Mount". It is a type of mounting which sits betwee
n the camera and its tripod (and probably other stuff) to provide celestial motion guidance or compensation for the camera during long exposures of the night sky.

A person trying to photograph the night sky can simply use a camera mounted on a tripod and start taking long exposures from seconds to minutes long. However, due to the rotation of the Earth, and thus movement of the stars across the sky, an overly-long exposure will cause "star trailing" (ie streaks of trails left by the stars). This can be artistic in a sense but for those who wanted to take sharp star-like photos, this is bad.

The Scotch mount is a home-made tracking mechanism which compensates for the star motion by guiding the camera mounted on it to follow the stars in exactly the same rate. In fact, it moves in the same rate as the Earth's rotation, if precisely constructed, but in the opposite direction.

The Scotch mount is a popular project engaged by many astronomers due to its simplicity in construction (depending on the type you are working on), cost-effectiveness as compared to buying a consumer-grade drive system such as those that come with telescopes, and the shear fun and achievement one can get out of making one.


Types of Barn Door Tracker

There are many variants of the mount. Every astronomer makes his own version of the gadget to suit his needs and that includes me. However, there are a few basic types. The following images are extracted from http://hometown.aol.com/davetrott/, courtesy of Dave Trott.

From this image below, we can see that there are 2 general types of mounts, single-arm and double-arm.



The tracker can also be manually driven or motorised using a simple stepper motor circuit. It is basically a trade-off between simplicity of construction, cost, convenience of use and accuracy.

From the image below (courtesy of Dave Trott), we can see that the tracking errors accumulated by the single-arm version will get pretty bad from 10 to 15 mins onwards, while the double-arm designs remain low in error rate for an hour or so.




Construction Details

The tracker I have built is a single-arm manually driven version customized for places, like Singapore, where the polar stars, like Polaris, are not visible anytime of the year due to low latitudes. The design should work

You can see that my tracker is pretty primitive and simple. If you have seen or constructed more advanced versions, you will know I am not joking. Basically, it consists of 2 wooden boards connected together at 1 end using hinges. On
1 end of the top board is a typical ball-head mount to mount your camera. On the other end is the "drive" mechanism. That's it!




Main Boards

The upper and lower wooden boards came from plywood planks salvaged from trash dumps. ie. they are free! The boards are connected together at one end using simple off-the-shelf cheap hinges that you can easily find in hardware shops.



To allow the tracker to be screwed on top of a tripod, a hole at the bottom of the lower board was drilled and a "Tee-nut" was fixed in. The tee-nut has the same diameter of 3/4" and thread as the tripod screw. Pretty cheap too.










Drive Mechanism

The drive mechanism consists of a handle which you must manually turn clockwise at a rate of 1 revolution per minute. This will cause the distance between the 2 boards to increase. The calculations to derive the various figures are as follows.
Please refer back to Dave Trott's images on the different tracker designs in the beginning of my article for the exact formula.

The Earth takes 24hr or 1440min to complete a single rotation. To be really precise, the Earth actually takes 23hr 56min 4.1s (a sidereal day) or 1436.068333 min to complete the rotation. So, taking a complete rotation to be 360 degrees, in a minute, the Earth would have rotated only about 0.25 degrees. That means, for every minute, my tracker needs to increase the distance between the upper and lower platform to achieve a 0.25 degrees increase in the angle subtended by the 2 platforms at the hinges.

The bolt at the drive mechanism was chosen to have a thread resolution of 1mm (ie the distance between each thread or line on the screw surface is 1mm).
Thus, a single complete rotation of the bolt will move the 2 platforms 1mm apart. A 1-minute rotation rate is chosen out of convenience so that one can easily synchronise the rotation with a minute hand of a watch.

Now we need to derive the distance between the centre of the bolt and the centre of the hinges. This is easily done using simple trigonometry. Dave Trott used "sin" in his illustration. I decided to use a simple tangent formula.

Distance between bolt and hinges = thread resolution / TAN(angle to increase per min)
Distance = 1mm / TAN(0.25) = ~229mm (round off to the nearest mm)

Thus, given these figures, we have achieved a rotation rate pretty close to that of the Earth in the opposite direction.

From the photo below, you can see that the drive mechanism is made up of many parts. A wooden handle is glued to a metallic bolt. The nut of the bolt is glued to the upper board using epoxy resin, a superglue consisting of 2 types of liquids mixed together.



The bolt is also glued to the wooden handle using epoxy resin. The handle is actually a part of a wooden puzzle. Pretty cheap material.



The other end of the bolt is a "cap" (or whatever you would call it) that keeps the mechanism from falling apart as well as preventing scratches to the lower board. It is also glued. The hook at the end of the board is used to hold a heavy object like a small bag. It acts as a counterweight when the camera on the other side is too heavy.



From the following image, you can see the drive mechanism in action after it has been turned many rounds.




Camera Mount

Originally, I wanted to save construction cost, so I tried to build a cheap camera mount myself using plastic pipes, a wooden ball with a hole in the middle (those used in artwork), a bolt and nut, and a "Y-shaped" screw. (see image below).



The pipe on the left is glued to the upper wooden platform. It is used as a base to hold the rest of the mount.



Another pipe sits on top of the base and swivels 360 degrees around it. It also holds the wooden ball above it. A bolt is glued to the wooden ball and has a diameter of 3/4" and a thread suitable for mounting a camera. The ball-and-bolt can move freely as long as the "cap" is not tightened. A "Y-shaped" screw and its nut is used to tighten this pipe to the base.



The "cap" is made up of 2 pipes glued together. A slot is sawed to allow the bolt to be placed horizontally.



On the whole, the home-made mount was stiff while making adjustments. That was also a main source of jerks to the whole setup, which in turn caused the polar alignment to be out. The simple bolt and nut setup couldn't firmly hold the camera in place. Sometimes, the camera would "rotate" due to a shift in the centre of gravity, causing the grip of the nut on the camera to loosen. The epoxy resin used to glue the nut for the Y-screw was weak. Many a times, the nut would "peel off" when I tightened the screw to keep the pipes from shifting
.

With all the hassle and disadvantages of this home-made mount, I decided to upgrade to a real camera mount.
The camera mount can be easily bought from any decent camera shop. It costs around S$20 to S$40. It allows you to position the camera in any direction quickly and smoothly.






Polar Alignment (1st Version)

The hinges of the tracker must be aligned parallelly to the Earth's rotational axis just like a telescope on an equatorial mount. Other trackers use Polaris (the Northern Polar star) but in Singapore we can't see that star since we are too near to the Equator and the star is always on the horizon, so I have to use another method.

I decided to align the tracker in an altazimuth fashion. That is, alignment by azimuth (horizontal directional bearing w.r.t. Magnetic North) and altitude (degree of vertical inclination w.r.t. the horizon according to your latitude). To achieve this, I needed 2 separate scales, one for azimuth and another for altitude. The reference line used for alignment is the line formed by the 2 hinges.

For azimuth alignment, a line parallel to the hinges was drawn in the middle of the upper platform (see images below). A compass will be placed over the line. The entire tracker will be rotated using the tripod's adjustments until the N-S needle of the compass coincides exactly with the line. To make the upper platform open or "rotate" in the opposite direction of Earth's rotation, the compass must point upwards towards north. Actually for very precise alignment, I need to offset the bearing slightly so that I am pointing to the True North and not Magnetic North, but for convenience purposes, I assumed them to be equal. Maybe in future, I will cater for the slight deviation.

I managed to salvage a lousy compass from somewhere in my house for the purpose of azimuth alignment (see image below).





For altitude alignment, to be precise, I actually had to incline the tracker around 1 degrees to the horizon since Singapore is about 1 degree North of the Equator. However, to save cost again, I decided to rely on the "bubble" feature on my tripod. Below are shots of the original tripod I have used which spoilt recently.



It had a small green tube containing a bubble on the head (see image below). To align it horizontally, I needed to adjust the tripod's head until the bubble consistently stayed in the centre. It was a tedious and painstaking way of aligning. I could never get the bubble to stay still in the middle. The adjustment screws and handles of the tripod were also very rough. A slight jerk would send the bubble to jump about. It took me quite long to get a less-than-satisfactory alignment. Moreover, I wasn't able to achieve the 1 degree altitude alignment required.



On the whole, the 1st version alignment design was a terrible hassle. It took me 10 minutes or more to get a rough alignment. I also need to constantly check and realign in between shots. However, to my surprise, some of the photos turned out pretty well.


Polar Alignment (2nd Version)

The first version polar alignment design was a hassle to setup and potentially inaccurate. Moreover, the original tripod was spoilt recently. So, I had to get a new and better one. I tried to look out for a sturdy budget tripod with smooth fine adjustments and the bubble mechanism. However, I couldn't find one. So, I decided to forgo the bubble mechanism and focused on sturdiness and smooth adjustments.

I have also decided to upgrade the old compass. The resolution of the compass reading was coarse and the needle was not stable. That produced poor results sometimes. So I decided to spend S$24 to get a more professional one (see compass below). It has many advantages. The resolution is finer and the rotation of the internal disc is much smoother, which saves me alot of hassle when trying to get a reading. It also has a line on the cap which I can use to align the compass with the line on my tracker. It also has prominent and luminous letters and lines combined with the white background of the disc, which allowed me to easily take readings in a dark place. Finally, I can use this compass for other purposes like trekking.





Without the bubble mechanism in the new tripod, I had to construct my own alignment mechanism. The general idea was to use a protractor, those used for technical drawing in schools, with a simple "pointer". I used easily available and cheap (or free) materials to make it. I shall call it "altitude alignment scale" or "inclination scale". The general idea is that as I tilt the tracker using my tripod, the swinging needle will indicate the angle of inclination from the protractor. Below a shot of the final look.



To start, a special screw was screwed tightly into the lower platform. This will allow more screws to be screwed in.



The special screw is actually a little brass screws used in computers to screw the computer motherboard to the casing. I salvaged many of such screws from old computers mainly because I was also interested in computer DIY assembly. Here are how such screws look like.



The picture of a complete scale component is shown below:



It is made up of 2 main parts. First, a normal plastic protractor which students use to draw geometric figures in schools. A hole is drilled at the centre of the semicircle for the screw and nut.



To help me see the markings more clearly in the dark, I sprayed a light-coloured paint on the reverse side of the protractor. The paint was actually used to paint model vehicles. I once tried to pick up model-making years back but gave up very soon and I was left with a useless can of paint. The worksmanship wasn't too good as there were smudges here and there but it will do.



The second part of the scale was its "needle" or pointer to indicate the degree of inclination.



It was again made up of cheap or free stuff. The needle was made from an ice cream stick which you can easily get from any stationery or arts-and-craft shop. You just need to drill a hole for the screw, saw it to the desired length and file up the sharp pointer end. The screws and nuts are again from old computers. The hole is larger than the screw so that the needle can swing freely.

To ensure that the needle accurately indicates the degree of inclination, it must be accurately made. The hole for the screw must be as close to the centre of gravity of the stick as possible. The same goes for the sharp pointing end. If not, a wrong centre of gravity will cause the needle to tilt, giving rise to false readings. This was my 3rd attempt. Previous 2 attempts were failures due to the lack of worksmanship.



Below is how the whole scale is fitted onto the platform.



The nut keeps the protractor tightly in place. The protractor is aligned horizontally with the lower platform. The screw is not screwed in totally to allow the needle to swing freely as the tracker tilts. The distance between the needle and the protractor is kept small to minimise parallax errors while taking readings.


The new scale was more accurate and easier to use than the bubble. Combined with the smoother adjustments of my new tripod, I was able to meet the 1-degree inclination requirement. The smooth tripod adjustments also reduced jerks and that shortened the time required for me to setup the tracker. Using the screw system, the scale component can be easily detached for storage and transportation. It can also be easily reused in another tracker.


Counter-Weight Hook

Here's a closer look at the counter-weight hook. It is also a cheap hook you can find in normal shops.



As a recap, the 1st version design of the hook placed it at the very end of the upper platform (see image below). Its purpose is to prevent the upper board from flinging open when the camera setup at the other end becomes too heavy.
Anything can be used as a counter-weight such as torch, bags, handphones, etc.



In the 2nd version, the hook was moved to the side to make way for the new inclination scale.




The Complete Setup (1st Version)

Here's a shot of the 1st version tracker with the original tripod.



Here's a shot of the tracker with my Yashica manual SLR, 50mm standard lens and mechanical cable release. The cable release is used to remotely press the shutter button so as to minimise vibrations during the shots.



Here's a sideview of the setup. From this angle, you can see that the handle of the tripod is used to adjust the inclination of the tracker with the help of the green tube.




Other Accessories

During field trips, I need to bring along other accessories. Ideally, I need to an analog watch with a minute hand so that I can synchronise the hand's movement with the turning of the drive handle. In occasions where I only had a digital watch, I had to estimate the amount to turn from the time itself. I also brought along my pocket astronomy guidebook with a mini starchart inside to help me locate constellations and objects. To provide minimal lighting, I brought along a red torch. Red light reduces glare on the eyes. Finally, I would bring along a pair of binoculars at times to help me locate smaller sky objects.


Working on a Dual-Arm Motorised Version

Although the single-arm manual tracker is moderately sufficient for up to 10 minutes of long exposures, manually driving the tracker can be a tiring task. I need to focus on both the watch as well as turning the handle. Most of the time, I also need to keep the cable release pressed. This can be quite a brain taxing task. For shots longer than 10 minutes, this can be an impossible task. So, I have initially planned to work on a dual-arm motorised version to allow me hassle-free accurate long exposures up to 1 hour.

After the initial planning, survey and designing, I bought some materials.

Below is a stepper motor that supposedly turns 48 steps per revolution. By right, I need to achieve the same turning rate as that in the manual version, that is 60 steps per revolution, which is in sync with the 60 seconds of a minute hand. To do so, I need to use a correct combination of gears to step it down to the 60-step turning rate. A controller chip and other peripheral electronic chips are also required to drive the motor. I couldn't find the controller chip in the shops. I also need a 12v DC power supply to drive the motor. This can either be a series of 8 consumer 1.5v batteries or a 12v lead acid battery.







Here are some other small items I have bought or found. The gear was supposedly found somewhere. The battery cable was actually meant for a 9v rectangular battery. The bolt and nut are leftovers from the 1st version. The others are bits and pieces of electronics I have bought such as switch, resistors, transistors, chips, capacitors, etc. I have also bought a circuit board, a soldering iron, a multimeter and a solder sucker.



I would also need to find more wooden planks for the actual construction. I salvaged some of the leftover wood from the 1st version and started working on a few parts. However, due to the lack of important raw materials like the stepper motor controller, I had decided to postpone the work until a later time.


Conclusion

The entire project was quite a success. The completed tracker provided me with the opportunity to take guided astrophotographs for up to 10 minutes long within a tight budget. It also helped me to gain knowledge and experience in astrophotography, electronic circuitries, carpentry and other skills. Although this manual version has some limitations, it will sufficiently fuel my interest in astronomy until I complete my dual-arm motorised version or get a professional equipment. On the whole, it was a fun and learning experience.


References

Finally, some external references. You might like to take a look at them to find one suitable to your needs. Many of them are much better than mine.

http://hometown.aol.com/davetrott/
http://www.astronomyboy.com/barndoor/
http://www.jlc.net/~force5/Astro/ATM/Barndoor/barndoor.html
http://www.skynewsmagazine.com/pages/resources/barndoor.html
http://www.fayettevillenc.com/special/backyard/barn.htm
http://www.tucsonastronomy.org/barndoor.html
http://www.angelfire.com/stars2/bwolfe/BarnDoor.htm