Control room extension.


CCD imaging requires more equipment than visual observing and even "simple" astrophotography, and much of it isn't designed to be exposed to the damp, nighttime air. While I never had any problems caused by this, I was always worried about it. By the time I had 3 computers and several power supplies in the observatory I was quite concerned and thought I needed somewhere more protected to keep the delicate bits.

Technology had moved on, however, and a control room might not be necessary. Running the CCD remotely (via something like VNC) might be good enough. In the end I decided that I still wanted somewhere dry to keep the computers and that a control room would not be a detrimental step, so I did both -installed a network and added a control room.

Network

I could install the network (even temporarily) and try it out before I built the control room and then decide whether I wanted the control room or not. Installing the network was simple - deciding which network wasn't. In the end I opted for cheapness and simplicity (usually my guiding rules) and installed an old-fashioned coax 10MBPS ethernet. The cable was cheap (cost me nothing for 200 metres) and the network adaptors were $5 each. It requires no hub, no repeaters and is adequately fast.

After some playing with the network cable with it just laying on top of the ground I decided that the network was definitely worthwhile and I should go ahead with that phase without delay. It takes a bit over 4 seconds to transfer a single ST-8 FITS image (3MB) from the observatory to the house computer. Compared to the effort I used to go through this is a breeze, and trivial relative to the 60 seconds to read the CCD out in the first place (I don't have the USB upgrade). I also decided to go ahead with the control room as the extra room it would generate would be desired, as well as keeping the computers out of the weather.

The coax was put into a 20mm conduit (along with a couple of extra multi-core cables to be used for an intercom and possible remote control) and buried. There is about 90-metres of cable between the study in my house and the control room outlet, and another 30-metres between the control room and dome outlet. There's a few metres of cable inside the house and control room and another 10-metres or so inside the dome. A coax network can cope with 176-metres and I'm within that distance and it works fine. Had I opted for 100MBPS CAT5 then it would have been considerably more expensive and difficult (although I'm told that it can be done in a single length of CAT5 between 2 computers only without interference problems).

The network works fine. I can set the system going from the control room then retreat to the house to monitor progress via VNC. Even better, I can then go to the dome and do some real observing while the CCD is busy doing its thing and keep an eye on it from there.

(As an aside, I tried TightVNC and found that it was unreliable at updating the remote screen from menu button-clicks. It seemed no faster than VNC so I continue to use it.)

Control Room

The design of the control room was dictated by the initial design of the roll-off roof. Had they been designed as one then it might have been different, but as it turned out (more by luck) the extension looks and acts as if it were intended to be this way.

After much thinking I decided that the control room would be a garden shed just the same as the original skyshed - this would stop it looking too out of place. Careful measurement showed that the extra 25mm gap in the roll-off roof height would be able to just fit over the top of the standard shed roof line with them both having the same floor level.

To complicate matters, the dome was north-east of the skyshed and the (almost invisible) guy-wires bracing the end of the roof rails were a severe hazard to visitors in the dark as they were in a direct line walking to the dome. It would be most sensible to have the control room under the rolled-off section but the height of the walls and other constraints precluded this, so the direction of roll of the roof had to be swapped. This was the first job done.

With an angle-grinder I cut the rails at the door end of the skyshed. I then separated the pieces joining the two sides. Holes were dug for the uprights (including their concrete foundations) and the straight sections, along with their uprights, were picked up and placed in their new holes (on the opposite side so that the rails lines up). This was easier said than done as the concrete footings were not light. But a bit of brute force and ignorance, a wheelbarrow (and my neighbour Rob for one that was too heavy for me) had it done quickly enough. It was then a matter of lining them up carefully, levelling them (which required some extensions added to one side as the ground slopes down that way) and then welding the rails together again. I then cleaned them up and painted them - a job I never got around to doing the first time. Some small shrubs were planted near the guy-wire to make them more noticeable.

The roof now rolled off the other way so I was part way there. Next was to prepare the area for the new shed. It was while doing this that I had a major stroke of luck - I found a shed of exactly the right make for free. All I had to do was to dismantle it and cart it away. My son Alexander and I managed this in a few hours, although things were slowed down a little when it came to the base frame which had been buried in the concrete floor by the inept original owners. The slight damage it sustained while trying to extracate it contributes now to the door not opening quite so smoothly as it should.

With the expensive part now solved (the price of a new shed was now near $600!) I could get on with the rest of the task. The foundations were dug and leveled, the waterproofing pastic layed in place and mesh suspended in just the right way. Then the concrete truck arrived. Here is a view showing the truck ready to unload outside the shed (you can see the new position of the roof tracks out towards the west) - but the driver was over inspecting the dome (he had delivered the concrete for the base of that and was interested in how it all turned out). While here you can see me tramping down the concrete after he had gone.

The cement was allowed to set for a few weeks, then the shed was erected. It took me the best part of a day with the help of my (sometimes helpful) children. This is a view of the completed shed from the south-west showing how well the new shed blends in with the old. There is a thin, dark line just visible which is the gap between the two. This was subsequently covered over with a section cut from one of the wall panels which wasn't used. You can see the overlap of the roof over the new shed here. The slightly higher roll-off roof is on the left, just overlapping the lower new roof. They blend in perfectly. The covering section is also in place here.

The shed was bolted down in place and the finishing jobs were done. The short gap between the two sheds is sealed on the outside, but there is still a gap on the inside when the roof is rolled back. I cut pieces of thin galvanised steel sheet (remnants of the dome skin) and screwed them between the tops of the sheds to seal this part. Finally, I had to cut a hole in one corner between the two sheds to allow cables to be passed through. This is 100mm diameter and I pushed a short section of 100mm plastic sewer pipe (unused) through the hole (large enough to take the biggest connector I would expect to use, and then a bit bigger). A rolled-up piece of foam is inserted as a seal once the cables are in place. Mains power was brought through a separate conduit and sockets and lights (both red and white) were wired in place.

I managed to find some old carpet and underlay to line the floor, then moved the equipment in. Here is the view from the outside looking in. You can see where what would be the back of the shed is only partially complete, allowing access to the old doors. This shed is not quite as large as the skyshed, but is plenty big enough at 2·9 × 2·7 metres.

How does it work? Brilliantly! Remote observing from the next room is very convenient. You can have as much light as you want, safe in the knowledge that you won't ruin your exposures, and you stay warmer by simply not radiating your heat away into space or wafted away in even the slightest breeze. The equipment stays dry (as do you on humid nights!) and it's a safe place to stow unused equipment. I thoroughly recommend a control room next to the telescope - even if you do end up running your exposures from your house.

Naturally, there are still jobs to complete - the main one being to insulate it and perhaps add an air conditioner/heater. Always something to do...


Update 2007/12:

Last summer I painted the outside of the skyshed in white "solarguard" paint. I'd been intending to do this ever since I built the original observatory as it should provide some control over the heat of the sun - but just never got around to it. I was astounded at just how much difference it made. Inside it now remains at the ambient air temperature instead of considerably hotter. Well worth doing and I'm annoyed that I waited for so long before doing it.

I also added a small air conditioner for when I'm working down there during summer days and want it even cooler. I've added more desks for more computers (I now run a twin-head computer so that there's enough screen space for both the Paramount and CCD software). The plethora of cables for motor focusers etc. and the new desk being on the other side to the original cable hole meant I had to make a new cable hole. This one is only 50mm diameter, but as connectors have shrunk from the old clunky parallel port 25-pin to smaller 9-pin serial and even smaller USB ones it's quite adequate.


home back to roll-off roof back to ATM on to remote control


Page last updated 2007/12/12
Steven Lee