Building an Observatory

This article is of the construction of a roll off roof design observatory. I will try to discuss and show all the problems that I needed to solve over the course of construction. I started work in October 2008.

Planning and Excavation

Choosing the site for my observatory on my block was relatively easy. I had a flat turn around point which was far enough away from the house to give me a clear view of the south and south east. It is also far enough away from the house to prevent heat bloom issues. I could not really move further away from the house because power lines and a road would interfere with imaging. As you can see from the image below I basically marked out position of the pier support first and the outer corners of the observatory. I am allowed 10sqm before I had to submit plans to council. So I am keeping just below that figure to prevent the hassle of approval. I knew from the outset that there was limestone deposits on my block and that would mean a lot of tough work in the excavation department. Hence the jack hammer in the frame.

The pier support for this observatory was supposed to be 1m x 1m x 1m. My thinking was that 2.4 tonnes of mass would be more than enough to counteract any vibrations from the pier. This would place the center of balance somewhere around 300mm under the surface of the concrete.

We had assumed that there was only one layer of limestone to get through. The first hour was easy, we dug through the first layer of limestone only to discover that there was another layer a little below the first layer. This layer was very thick and very very hard. It is a good idea to have the right tools for the job and a buddy to help you out. A mate of mine helped me out with the excavation. He is seen here working the "pit". The only way we could get anywhere was drilling and chipping the limestone. It took several hours to go down 200mm. In the end I managed 700-800mm in depth and figured this was close enough to having enough mass. The overall dimensions made for a 0.9 cubic meters mass. This would be somewhere around 2.2 tonnes of weight.

Forming up and Concrete

Once the pier pit was as deep as I though prudent I formed the hole up. Try to remember if you are doing this that several things really need to be done to make life easier down the track. Make you form work exactly square. I am not meaning a square, I mean ensure your corners are at 90 degrees. Do this by checking the diagonals. They should be the same distance if your forming is square. The reason for doing this is that all your observatory measurements can come off the edges of the formed concrete.

Also of interest you should try to get the slab as level as possible. If your pier is placed near plumb and level then you will not need to work so hard to get the mount polar aligned down the track. In my case I am planning on placing the mount directly onto the pier without a leveling plate on the top. So everything being level and plumb from the start will make the polar alignment easy. I found that I had to raise one side significantly to gain level, this meant that I had to back fill quite a lot. Most of the big rocks that I had pulled out during the excavation were now used as a dam to prevent he concrete from flowing out during the pour. Some of the dirt that came out of the hole was then placed behind the rocks to provide a bit of stabilizing.

Now for the steel work. Steel is used to reinforce the concrete and prevent it from cracking or fracturing. In this case I used F72 mesh and 15mm re-enforcement bar. I used tie wire to secure it all off. It does not have to look flash just rigid. Ensure that the top of the steel is set well below the finished level of the concrete. The steel as seen below is set down 120mm below finished level.

In the image below you can see the pier hold down bolts and cage construction. I don't have welding skills but working in the construction industry I used a bit of lateral thinking. I am a plasterboard contractor so I used steel stud which I screwed together and fixed into position. I made sure that the hold down bolts were located in the center of the pier holes and then worked from there. Ensuring that all parts were square to the pier. This means that later when I install the pier it will simply slide over the bolts. The image below does not show the correct depth of the cage. It will be in fact a lot deeper into the concrete. However for setting up the cage I thought it best to go as close to the pier as possible to ensure everything was square and fit correctly. Remember you only get shot at this so you want it right from the outset. 

So with the cage built it was time to install this in pier support hole. as you can see from the image below the cage is installed below the top of the reinforcing mesh. To prevent the hold down bolts from moving under the conditions of the pour I fixed guide rails to the form work. As it turned out though I needed to have these guide rails on edge to prevent the timber from sinking. That is exactly what transpired when the concrete was poured. The weight of the concrete around the cage bent the timber downwards. As a further tip I made sure the hold down bolts threads were not exposed to the concrete. I did this by placing extra nuts on the threads. This made for a clean finish after the pour was over.

So when it came to the concrete itself, I wanted something to be maintained at all costs. That being the level of the concrete. For the design of pier I chose it was critical to get my concrete very level. Screeding the concrete level was a bit of a challenge but in the end it was very close to level. This will mean I only needed to shim for any minor level differences in the top plate of the pier. In the image below you can see the finished concrete with the hold down bolts in place.

Next step was to fit the pier onto the pier support. Knowing concrete as I do, it is never level no matter how hard you try or how good a concrete finisher you may be. So to get around this issue I decided to shim the pier so that it is level at the very top. The pier construction would also have defects in it too. Shimming the pier was easy enough. I cute several tin squares roughly 1 inch by 1 inch with a larger piece folded over to prevent the smaller pieces moving all over the place. The image below shows the pier shims up close.

The image below shows the pier in position from a distant vantage point. As you can see the pier is only shimmed on one side. It is now very solid and when you bang your hand on the side of the pier the vibration dies down in under a second. That is by virtue of using just over 2 tonnes of concrete, a very thick walled pier and the gussets. Remember if you are building an observatory this is one of the most important things that you can do during the process. Don't under estimate how important the pier and pier support is in the equation.

Decking and Walls

The next part of the equation was setting out the deck support holes. I had 14 holes drilled by a local chap who was kind enough to work on a Sunday. If you are digging into limestone, take it from me you need heavy machinery. Digging one large hole was bad enough but digging 14 holes that are not real large was going to be more trouble that it is worth.

Once the holes were drilled the first order of business is to get the four corner posts set in a squared position. These posts will determine how well things work later on too, so it is really important for this to be all square. Put one post in first, ensuring it is plumb in two axis. Set the concrete (I used quick set and used two bags and at a depth of 600mm) and then move onto the next corner. Set that one in too. Now the important part is to ensure your next post is square (perpendicular to the first line) to the other two posts. Use Pythagoras theorem to establish that you have a square angle. ie use a2 + b2 = C2. In this case I used 900 + 1200 =1500mm. Once you have this in the final post is just a reflection of the relative distances involved. As you can see below the posts were set in well high enough to allow for some cut off when setting the height of the walls.

Now that the corner posts are in, the next thing that one has to take into consideration is how your deck will be supported. For me I chose to have two central beams supporting the decking and two outer beams taking the load of the walls. All the beams are 190 x 45 treated pine. The outer beams are cantilevered from the central beams by 600mm but supported by the outer posts. With any luck this prevents long term sagging of the outer beams. The image below shows the deck posts in but not cut yet. Two bags of quick set concrete in each hole set to a depth of 500mm for each post. It gives a really solid footing.

When housing out posts, most people make the mistake of housing out the beam to full depth of the beam. This is bad for a number of reasons, but the most important reason is that it weakens the post's strength. So therefore a general rule of thumb is that housing should never be any greater than one third of the posts thickness. So in the case of a 90mm post, 30mm is the maximum that you should house out. The housing is really there to provide a bit of support for the beam and help the bolts take the load. In this case I have housed my beams to 20mm; enough support to prevent shearing off the bolts but not too much to weaken the integrity of the posts. See the image below for a pictorial of what I am talking about. Be mindful of your deck height in relation to your pier height. I have allowed a 900mm high pier once the deck it completed. Take the time to work this out now as it will be too late once you build everything to change your mind.

The next step is to attach all the joists and bearers for the floor. You will want the floor to be very solid and have little if any bounce. The reason for this is that you may be carrying heavy equipment over it and have several of your friends in the observatory at one time or another. For my observatory I placed the main supports and bearers 700mm either side of the pier support center, with the outside of the joists being cantilevered onto some dummy bearers. Structurally this gives a lot of strength to the entire deck. You want to ensure that you triple grip (little metal triangle brackets seen below) each joist onto the bearer after you have nailed them off. This stops rolling of the joists and helps spread the load out.

Once you have it all in place it should look like the image below. You will note that I have inserted some extra joists around the pier so that the decking is supported correctly. This is in spite of the fact that Structaflor can support up to 600mm with 22mm decking. Joists for the deck were 90 x 45 treated pine and the bearers were 190 x 45 treated pine. Using treated pine is best for an exposed under floor, it will not rot readily and of course it prevents termites from causing issues later on.

Once you are satisfied that the deck structure is sound it is time to place your decking down. You could use a variety of materials to deck, either solid timber slats or floor decking is fine. In my case I used a product called Structaflor which is 900mm wide and come in lengths of 3.6 meters. It is water resistant for 3 months of continuous exposure to the elements and it perfect for sticking down a rubber floor mat later. It also has a tongue and groove system for ease of installation. You need to use a maxbond glue which you apply to the joists when dry and then either nail it down or screw it off. It is easy to work with so long as you have a jigsaw and a circular saw. The image below shows the deck completed and ready for installation of the the walls.

The next part of construction involves the walls. I have chosen a wall height of 1.8m which is steel framed using stud and track system from Studco. The studs and track are 1.15BMT with noggin track in the center of the wall. Each stud is screwed off on each side at each junction. This makes for a very rigid wall and of course is impervious to rotting or termite destruction. The image below shows the stud work in place. It took me about 5 hours to erect the walls. You will need a cut off saw and heavy duty screw guns for this work. You should also ensure that the stud and track is well fixed to your posts and deck. I used 12 gauge course threaded screws for the job and I put them in at 400 centers. If you live in a windy location this should be fine for a fixing dimension. You don't want your observatory to blow away. You will also notice that I have installed the door jamb. Use an exterior jamb set which are around 140mm. Ensure that you have it well fixed into position. I plan on using further anchorage from the inside of the jamb set into the wall frame behind. I have temporary installed it with 12 gauge screws, but for security purposes you will want much larger screws placed in behind the door stops. This will prevent most break an enters. Only someone with heavy duty equipment will be able to get in with this system in place.

Now that you have the walls up, you will want to get the external cladding done. This is the most time consuming aspect of the entire project. It takes hours to clad these small walls. The cladding I used was Hardies weather board. It is wood grained 10mm thick and very weather resistant. It is easy enough to cut with a dry saw or even using the score and snap method. You will need to use screws or nails that have a reach of 30mm. The reason being is that each layer up over laps each of the previous layers. To finish off the corners you will need the aluminum corner beads. They are easy to install, just ensure that you leave the stud closest to the corner unscrewed so that you can slip the corner bead into position. Then simply place the holding screw into the top of the corner bead and screw off the board to the corner stud.

To prevent water penetration it is a good idea to use sarking. The type of product you use is largely unimportant, just ensure that you have good coverage. What sarking does (its the blue material seen through the door below) is to act as a vapor seal. It is not insulation. You need sarking in place to keep the insulation in good condition for the next 20 years. The image below shows the walls finished and clad.

Roof Structure

Now that the walls are completed it is time to start thinking about the roll off roof. This has to be constructed first so that you know exactly how the fixed roof will be placed over the top of it. The fixed roof is facing almost due south and that is the direction in which the prevailing weather emanates So when building your own observatory think about the weather, you don't want your observatory to leak.

The roof is constructed from 1.15 BMT steel stud and top hat sections. Each stud is formed to make the truss which has a center pitch of 300mm from the bottom chord. These are placed at 750mm centers and then the top hat sections for the purlins. The purlins are fixed at 400mm centers. You should ensure that you use heavy duty screws with a breaking strain of 600lbs per square inch or more. I found it convenient to build everything in situ rather than building it on the ground and then trying to lift it into place. The image below shows how this is set out.

So now for the slider mechanism. I decided to run the slider mechanism internally within the observatory for security and for looks. It is comprised of a 100mm C purlin with 15mm lips. These are placed lying on their sides and then 5 wheels are placed inside this with the 180 timber beam bolted into position and an axle of M8 threaded bolts. A length of aluminum extrusion is fixed to the bottom of the beam to keep it straight and to fix the wheel studs into. The final element is to fix this slider to roof. Some important lessons I have learnt here is that the slider really needs 5 wheels per each side. The bolts should also be made from hardened steel and finally the connection to the roof should be nice thick plate steel or aluminum extrusion. You also need to ensure that the whole assembly is parallel with each sliding assembly and also with the channels. I worked on a tolerance of 5mm and have found that this should really be around 1-2mm. The image below shows the slider mechanism in place and fixed to the roof.

Once you have it all in place you will need to test it the way it slides out. I found this useful to show where certain defects were in the design and those could be corrected now rather than later. Unfortunately one cannot find all the defects but you get the idea. The image below shows the roof sliding all the way out on the out riggers. The temporary bracing can be seen holding the outriggers in position.