Building a Roll-off Roof Observatory

Introduction

For the first 40 years of my interest in astronomy I was a member of the setup-takedown club. Like most amateurs, if I wanted to observe the night sky I would have to take my equipment outside, set it up, observe with it, then take it down and put it away. For naked-eye and binocular astronomy this is fine: the equipment is lightweight or non-existent, and being out in the open gives better views of the sky than being in an observatory.

However, once a telescope comes on the scene things become more complicated. Most telescopes take at least 15 minutes to set up – more if good polar alignment is required for astrophotography. Ancillary equipment such as eyepieces, barlows, cameras, charts, torches and warm clothing take up extra time to carry outside, so the overhead of effort grows bigger. Now the decision to go out and observe has to include checking on the weather forecast, remembering if you have to be up early next morning, and factoring in the plans of spouse or partner.

How much easier if one has an observatory! The scope is permanently set up ready for action, as well as being polar aligned. Most of the other equipment will be to hand in the observatory, so it’s really just a question of checking the weather, opening the dome or roof and off you go. And since there is little or no take-down time, if the clouds roll in or you have to be up early, you can shut down and be indoors within a couple of minutes.

These thoughts occurred to me regularly during the years when, for one reason or another, I could not realise my dream of having my own observatory. The obstacles in my case included a job where I had to move location regularly, a house in a light-polluted suburb making an observatory pointless, and a lack of time and resources to make such a project happen.

Eventually though, in 2010, I found myself in a position where I could not only afford the time and expense to build an observatory, but also with the opportunity to choose a house with a south-facing garden and reasonably dark sky.

As soon as I had moved in, I began planning my observatory.

Planning

The first decision I had to make was what kind of observatory to build: Dome? Roll-off Roof? Roll-off shed? or something else? I consulted More Small Astronomical Observatories (Patrick Moore – Ed., Springer 2002) which offered a diverse range of designs and gave me plenty to think about.
When it comes down to it, I have to admit that the sight of an observatory dome has always excited me: it’s certainly the classic look. On the other hand a roll-off has the advantage of more rapid cooling as well as an unobstructed view of the sky. It is also more roomy, although many people build a control room adjacent to their dome, which has the advantage that it can be heated. There again, a roll-off is better able to blend in with its surroundings than a dome, looking essentially like a large shed.

Having decided to go for a roll-off I bought a copy of Building a Roll-off Roof Observatory by John Hicks (Springer 2009). I came up with a design that incorporated the features I liked best from among the many variants illustrated in the book.

The next consideration was size: I wanted it big enough to accommodate a 14″ SCT with enough room for a computer desk and equipment storage, but I didn’t want it to totally dominate the garden: so I went for an internal size of 10′ x 8′ – remembering of course that the footprint needs to be twice as long to accommodate the gantry and rails for the rolled-off roof.

I then contacted a builder friend who had done some work for me before, and he provided much-needed expertise and experience – though for him too this was his first observatory project.

Planning the floor slab and pier base

The floor is a concrete slab. It needs to be thick enough to carry the weight of the building and all its contents except the telescope, which has its own separate concrete base. The builder advised that a 4″ thick slab should be sufficient (and so it has proved). The telescope is mounted on a steel pier which is bolted to a 24″-square concrete base that is acoustically insulated from the observatory floor. This prevents vibrations caused by the operator moving around the observatory from being transmitted to the telescope (and potentially ruining any image being exposed at the time).

The pier base has to be considerably deeper than the observatory floor to ensure it is rock-steady with absolutely no possibility of movement. I have seen recommendations for 18″-deep footings but I think this is rather extreme unless you are planning a metre-class scope. I settled on a depth of 8″ for the pier base.

The image shows the concrete shuttering for the floor and pier base. To achieve the acoustic separation the concrete for the pier base was poured first, and once it had set, a layer of polystyrene about 1″ thick was placed around it to create the required gap. The concrete for the main floor slab could then be poured.

The finished concrete

 The image shows the concrete after setting. The polystyrene insulation that acoustically separates the pier base from the floor extends all the way down to the bottom of the concrete. The plastic drain pipes at left are conduits for the electricity supply cables – do not forget to insert these before pouring the concrete!

As can be seen from the image, the pier base is offset from the centre of the observatory towards the north to allow the scope to see more of the sky above the south wall. I was not intending to observe targets in the northern sky: in my location it suffers from light pollution from Warrington, Liverpool and Manchester, and in any case it would be partly obscured by the rolled-off roof.

The separate pier base has worked very well as it is possible to walk around the observatory with an image in progress without any ill effects on the result. On the other hand when I set up a second scope on a tripod on the floor I found that even walking around carefully caused vibrations that ruined the image, so I had to build a delay into my imaging software to allow me time to vacate the observatory before the exposure started. I would definitely recommend taking the trouble to acoustically separate the pier base.

Walls and Roof

The intention was to make a solid structure that would last a minimum of 15 years, so the walls are double-skinned timber: shiplap planks to the exterior and plywood panels for the interior. Within the cavity sit vertical 3″ fence posts for strength, resulting in a total wall thickness of 4″ – necessary in any case to accommodate the width of the roof track.

The pitched roof frame – also of very solid construction to guard against possible wind damage – is timber comprising three A-frames joined by longitudinal purlins. The joists overhang the sides of the building and together with the fascias and soffits effectively ‘box in’ the edges of the roof, leaving about 1½” clearance between the soffits and the top of the walls.

The joists are covered with roofing felt and the exterior covering is steel box-profile roofing sheet coated with green polyester paint. (Although a white finish would heat up less in the sun it would have been a lot more conspicuous and I wanted the observatory to blend in with the surrounding greenery as much as possible: besides, baking hot weather is not so much of an issue in northern England and once the roof is opened the observatory cools down very quickly).

Rails and Wheels

The mechanism for opening the roof comprises six wheels mounted three each side at the base of the A-frames. The wheels are 5″ diameter pulleys that run along rails fashioned from inverted angle-iron welded to a steel track (see left). The rails need a very solid support given the heavy weight of the roof, so the gantry comprises two horizontal 6″x 3″ beams supported on four massive 6″ square posts concreted into the ground. Despite its weight the roof rolls easily under finger pressure.

Fitting Out & Finishing Off

With the building complete attention could turn to fitting out the interior. An electrical supply and lighting were of course required. Take no chances with electricity in an outbuilding: the work should be undertaken by a qualified electrician and the building should have its own consumer unit complete with RCD. I had plenty of 13A sockets fitted, some on the walls for the computer and other ancillary equipment, some floor-mounted for the telescope and accessories. I wanted both white and red dimmable lighting: these were wall-mounted – obviously they could not be fixed to the rolling roof!

The observatory was painted inside and out with exterior woodstain (Cuprinol Garden Shades), and I fitted carpet tiles over the concrete floor, having decided to do away with the original idea of a wooden floor. Initially the roof was designed to be opened by hand, and would be kept closed by a vertically sliding plank secured by nuts and bolts. I also installed a pair of chains to secure the roof beams to hooks in the walls in case of really high winds.

The Scope Installed

Now for the scope. I had chosen a Meade 14″ ACF, this being a modified Schmitt-Cassegrain designed to be coma-free. The optical arrangement combines a spherical primary mirror, a hyperbolic secondary and a corrector lens mounted at the front of the tube that also helps keep dust out. Its f/10 140″ focal length is thereby compressed into a tube only 32″ long, which makes the scope easy to use and to accommodate in an observatory. Even with a dew shield doubling the length of the tube it was still able to slew fully east and west without colliding with the walls.

However, all telescope designs are a compromise to some extent, from astrographs that cannot be used visually, through large refractors that require a lot of room, to Dobsonians that are purely for visual use. I wanted to do astrophotography but I had not made a choice between lunar/planetary and deep space. The Meade design was a compromise that would allow both types of imaging: with a 3556 mm focal length that could be doubled (or more) by a PowerMate it was eminently suitable for planetary imaging, while it could deliver a respectable 20 arcminute fully-illuminated field of view for deep sky work.

I sourced the pier from Astro Engineering: their standard Pro Pier was 41″ tall but my calculations indicated that with the AE MegaWedge I would only need the pier to be 22″. Fortunately AE were happy to supply a cut-down pier of that size. The pier is bolted into the concrete base, the bolts being bonded into the base with epoxy resin.

With the observatory finished I could get down to the fun bits: first light through the scope, polar alignment, adding cameras and filter wheel (a CCD camera for deep sky photography, and a CMOS video camera for lunar and planetary work) and getting to know how to use it all. I added a hydrogen-alpha solar scope and later an 80mm guidescope. I got to grips with all the software needed to make it all work: Maxim DL to control the equipment, FocusMax for automated focusing, planetarium programs to find and frame targets, stacking software for the planetary images and later ACP for observatory automation and PixInsight for advanced image processing. Lots of fun!

Later Modifications

Initially I was content to go out and observe whenever the sky was clear and I was free. But after a while it began to dawn on me that I wasn’t making very efficient use of the observatory: the need to get up for work or school runs prevented me from staying up too late on week-nights, and there were many times when the sky would clear after midnight when I was unable to observe.

I began to look into automating the observatory, and I went as far as motorizing the roof and installing weather sensors, the plan being to use ACP Observatory Control software to take images automatically while I was asleep, and to close the roof if the weather became unsafe. However I was plagued by problems: the Go To performance of the mount was disappointing, as was the tracking and this made it difficult to get autoguiding working reliably – and this is essential for unattended operation. I realised that I would probably need to upgrade my mount to overcome these problems, but a more depressing realisation was that the British weather was unlikely ever to allow me to get the best out of what was becoming an expensive setup.

The problems with the climate in my home location are many-fold: firstly our latitude restricts the targets that can be imaged to those north of the celestial equator, while at the same time it causes us to lose about 10 weeks in the summer when twilight persists all night, writing off most of May, June and July.

We have a large number of cloudy nights due to the endless succession of low-pressure systems that roll in from the Atlantic, sometimes preventing observing for weeks at a time. The climate is also damp so that even when it is clear, it is common for the relative humidity to be over 90% and after about half-an-hour all the surfaces in the observatory become dripping wet. Finally, when it is clear and not too humid the chances are that the wind is gusting, causing autoguiding to fail and ruining exposures.

In the end I took the decision to use the observatory purely for visual observing. I sold the Meade scope and moved my homemade Dobsonian (see the story of that project here) into the observatory, and these days I do my serious astrophotography remotely from an observatory in Spain.