Practical Rolloff Roof Observatory in Michigan USA

Dennis Allen

My family owns property up in west-central Michigan. This is an area known for its relatively dark skies. It's a place I go to hunt, fish, and enjoy the occasional clear night. Early this spring, I was treated to a whole flock of clear nights. One problem: too much snow on the ground. There was simply no place to set up my telescope.

So this year, I vowed to build an observatory.

My original idea was to create a peaked roll-off roof. This building would have a 12 ft (3.6 m) square

Figure 1.1 Dennis Allen's roll-off roof observatory.

Figure 1.1 Dennis Allen's roll-off roof observatory.

wood floor and 4 ft (1.2 m) walls. Wide enough to leave plenty of room for my 13.1 in (333 mm) reflector. Whenever I got a bigger telescope, something requiring more stability, I could always pour a small concrete pad. I wanted something simple, practical, and durable. But I didn't want to spend years planning and months building.

I kept my design simple: a one-piece roof, rolling to the north. Three inch (75 mm) caster wheels would extend down from each truss and would ride on aluminum channel. To keep the roof light, I'd use corrugated sheet metal. The south wall would have a standard 3ftx7ft (910mmx2130mm) door, cut off at the 4 ft (1220 mm) mark. The upper 3 ft (910 mm) section would hang from the southern gable.

Step one was to build a scale model. Most people do not know what a roll-off looks like. A one-inch-to-the-foot (1:12) scale model helps illustrate your intentions (see Figure 1.2). You can obtain materials to make the model from any model airplane shop.

As it happens, my father is a carpenter. I told him my plans and showed him my model. I kinda knew he'd help! He quickly drew up a list of materials. To keep snow off the roof, he suggested a 6/12 pitch roof. To maintain head clearance, he suggested using church trusses. With 12 ft (3.7 m) church trusses, the bottom 2 in x 4 in (50 mm x 100 mm) doesn't go straight across. Instead, two 6 ft (1.8 m) horizontal 2in x 4 in (50 mm x 100 mm) pieces connect to a vertical 2 ft 6 in (0.8 m) 2inx4in (50 mm x 100 mm), creating an interior 3/12 pitch.

Figure 1.2 Scale model of the prospective observatory.

As soon as the snow melted, I contracted a bulldozer to clear and level the top of my hill. My dad ordered the trusses, custom made, from the local lumber company. One regular 12 ft (3.7 m) truss (for the northern gable), and three of the 12 ft (3.7 m) church trusses. Meanwhile, I ordered four 16 ft (4.9m) sections of 1|inx|in (44mmx19mm) aluminum channel from a local sheet metal shop.

By the time I was ready to build, several people told me a small concrete truck could make it up the hill. I always wanted a concrete floor. Concrete makes for a solid foundation, and is less expensive than treated wood. With a concrete floor, my building could house a bigger telescope. To house an 8 ft (2.5 m) long telescope, for example, I'd simply locate its base a few feet north of center. I had considered the thermal problem of concrete. But this is a roll-off, after all. Once opened, the heat should dissipate quickly.

There was one drawback, however. A concrete floor meant a permanent structure. Such a structure would require a special building permit from the local township board. I would have to hire a surveyor to obtain the exact location of the structure. Finally, I would be required to withdraw that location from the Commercial Forest Act of Michigan.

While acquiring the permits, I decided to upgrade my design. I opted for a 12 ft x 14 ft (3.6m x 4.3 m) building with 5 ft (1.5 m) walls. I would have liked a 14 ft (4.3 m) square building, but I already had the 12 ft trusses. These trusses were designed for 4 ft (1.2 m) centers. So my dad made a fifth truss, using the other trusses as a pattern, to give me 3 ft 4 in (1 m) centers.

By the time I got my permits it was almost the end of June. But with help from my dad and brothers, I knew it wouldn't take long to build. In fact, it didn't take an hour and we already had the forms in the ground. Once the forms were down, I had the local cement company bring in three cubic yards of concrete. We went with a 4 in (100 mm) thick floor, 10 in (300 mm) edges. We used 5-gallon (23-litre) buckets, open at both ends, as forms for the outside rail posts. The whole process took only a half day. There was plenty of leftover concrete, though no extra forms. We should have poured an outside viewing pad, something you might want to keep in mind if you decide to pour concrete.

Actual construction started a couple of days later. On the first day of construction, we threw up the walls and the rails (see Figure 1.3). The walls were built out of simple 2 in x 4 in (50mm x 100mm), 16in (410 mm) centers. We used treated pieces of 4 in x 4in for the top of the walls, the bottom of the trusses, and the outside rails. To connect the rails to the walls, each piece of 4 in x 4 in (100 mm x 100 mm) had a 2 in (50 mm) square notch at the end.

On the second day of construction, we put up the plywood. Originally I thought of using cheap particle wood (chipboard), covered with vinyl siding. My dad, however, talked me into using fake rough-cut 7/16 in (10 mm) plywood. This material looks like rough-cut pieces of 2inx8in (50 mm x 200 mm). As it turns out, this material is stronger than particle wood and already had a gray primer coat.

We brought 13 sheets of plywood. The sheets were cut with a 2 in (50 mm) overhang on the bottom and a 6 in (150 mm) overhang on top. The top overhang turned out to be a blessing. It would end up overlapping the 4inx 4in (100mm x 100mm) roof beams, covering the caster wheels completely, thus keeping the elements out. As a bonus, this top overhang would serve to keep the roof rolling in a straight line.

We brought a full-size door, cut at the 5 ft (1.5 m) mark. So to finish the day, we hung the bottom section. We made this section of door swing to the outside, thus preventing people from kicking it down. If you hang a door this way, however, remember to use special outdoor hinges.

On the third day of construction, the roof went up. We mounted the ten 3 in (75 mm) caster wheels on the two 14 ft (3.6 m) pieces of 4inx4in (100 mm x 100 mm). The caster wheels were spaced so that each wheel would rest under a truss. The channel was used to make sure the wheels were lined up correctly. This channel was already counter-tapped, so we quickly screwed it onto the rails.

One suggestion: keep your location in mind. Apart from a portable generator, we had no electricity. So try to have as much of your material prepared off-site as possible.

The two 14 ft (3.6 m) pieces of 4 in x 4 in (100 mm x 100 mm) were dropped into each channel and the trusses placed on top. We used 14 ft (3.6 m) pieces of 2inx4in (50 mm x 100 mm) to connect the trusses. After some adjustments to the trusses and caster wheels, we could roll the roof back and forth.

Originally, we ordered 16 ft (4.9 m) pieces of 2 in x 4 in (50 mm' 100 mm) to mount the corrugated sheet metal. We didn't stop, however, just because we were stuck with 14 ft (3.6 m) pieces. To get our north-south overhangs, we simply used scrap pieces of 2 in x 4in (50mmx 100 mm). This added a little weight to the roof but hey, if you stop construction for every minor inconvenience, you'll never get any work done, will you?

For the roof, we used eleven panels of 8 ft (2.4 m) White McElroy. These sheet metal panels went up in only a couple hours. We did have to cut one end-piece. For that, however, a roofing knife did the trick. Simply run a straight edge with the knife and flex the sections until they split. But whatever you do, be careful: all the panels have a smooth edge, but the cut pieces are razor sharp!

Here's another useful tip. When you install your panels, do both sides at the same time. Each time you have enough panels, put a section of cap on. When we installed our panels, we left the cap to last, which wasn't easy. Being the lightest in weight, I had to perform a high-wire act just to get the caps nailed down.

On the fourth day of construction, we worked on the gables. We were running short of plywood, and had to buy three more sheets. Which, as it turned out, was about how many sheets worth of scrap we had left over!

The northern gable was easy. The fake rough-cut plywood was measured and cut to butt right up to the corrugated sheet metal. It was notched out for the 2 in x 4 in (50 mm x 100 mm) slacks. To keep out the elements, we left a few inches of overhang on the bottom of the gable.

The southern gable was a different story. I wanted 3 ft x 7 ft (910 mm x 2130 mm) of clearance for the door. To achieve that, we couldn't place a piece of 4inx4in (100 mm x 100 mm) across the threshold. The fake rough-cut strengthened the walls considerably, but the southern wall was still the weakest. So for more strength, I decided to add tables to each corner on the southern wall (see Figure 1.4a).

For the upper section of door, we built a 2 in x 4in (50 mm x 100 mm) frame. For strength, we used a couple of 2 in x 4 in (50 mm x 100 mm) struts to connect the lower gable corners to the next adjoining truss. We placed our hinges at the top of the upper door, so that it would swing inward. When I want to move the roof, I simply prop the upper door with an extra piece of plywood (see Figure 1.4b). To lock the upper door, I mounted I-bolts and drilled two holes into the 2 in x 4 in (50 mm x 100 mm) frame.

To roll the roof off, there couldn't be any plywood overhang on the southern gable. So we used 1 in x 6in (25mmx 150 mm) trim, nailed to the southern wall, to cover the crack. We also used this material around each section of door (see Figure 1.5).

To keep the roof from blowing off, I installed chain binders to each corner of the building. These chain

Figure 1.4 The interior of the southern wall. a Tables added to the corners for strength. Chain binder also visible, left. b The upper section of the door propped open.

Figure 1.4 The interior of the southern wall. a Tables added to the corners for strength. Chain binder also visible, left. b The upper section of the door propped open.

Figure 1.5 1 in x 6in trim added to the southern wall and door frame.

Figure 1.6 Side view of the observatory and external rails.

binders hook to big eye-screws, which are screwed into the roof's 4 in x 4 in (100 mm x 100 mm) pieces.

And that's it! Since then, most of the work has been minor. For security, I installed a latch guard on the bottom door and a 12 ft (3.7 m) cattle gate at the bottom of the hill. They may not stop anybody from breaking in, but they should make people think twice.

I added 40 in (1m) strips of 4 in (100 mm) square foam between the trusses and the pieces of 4 in x 4in (100 mm x 100 mm). They keep the elements out, as well as animals and insects. This last month, we've had lots of rain in Michigan. The building, however, has remained bone-dry.

Figure 1.6 Side view of the observatory and external rails.

Figure 1.7 External rails with the roof partly rolled.

As an added touch, I installed a 12 ft x 14 ft (3.7m x 4.3 m) piece of outdoor carpeting. The carpet helps protect your telescope from the corrosive effects of concrete, and saves that occasionally dropped eyepiece!

Were there mistakes? Most certainly. When the cement truck left, he had to dump the extra concrete. As I said, that concrete could have been used for another viewing pad.

We could have reduced the weight of the roof if we had single 16 ft (4.9 m) strips of 2 in x 4 in (50 mm x 100 mm). In fact, we could probably have gotten away with 16 ft (4.9 m) 2 in x 2 in (50 mm x 50 mm) strips (although the structure has to be within the local building code).

If I had to do it over, I'd have used 4 in (100 mm) caster wheels instead of 3 in (75 mm) wheels. The 3 in (75 mm) wheels have already developed a fine film of rubber, probably due to wear and tear; and at some point I may end up replacing them.

But there were pleasant surprises. The plywood overhangs cover the caster wheels rather well, and made building the roof easier. In addition, I don't have to insert foam strips between the caster wheels to keep the weather out.

The church trusses make the inside look like a cathedral (see Figure 1.8). Had I known I'd have that much head room, I'd have stuck with 4 ft (1.2 m) walls.

Figure 1.7 External rails with the roof partly rolled.

Figure 1.8 The church trusses supporting the roof.

Figure 1.8 The church trusses supporting the roof.

I was a little worried about the channel. The caster wheels are 1jin (38 mm) wide, while the channel is less than 1| in (44 mm) wide at the ID. I figured for sure the wheels were going to bind. As it turns out, however, the tight channel keeps the roof running in a straight line (Figure 1.9), and there is no need for side casters.

At first the roof was very hard to roll. I was already thinking I might have to rig up a block-and-tackle system, but as time went on, the rolling became easier. The plywood overhang tends to swell, so I've been inserting wooden shims to keep it peeled back. Applying silicone spray to the caster wheels also helps reduce friction.

Figure 1.9 The observatory with the roof fully rolled back.

Figure 1.9 The observatory with the roof fully rolled back.

Was this article helpful?

0 0

Post a comment