|
THe Observatory Uplift>This is my observatory build record I thought I'd share since there are always some things which are novel and worth repeating elsewhere. I had built a telescope OTA that turned out to be too big for my mount and dome together - I could get it on the mount but couldn't move it without snagging the dome. So I made the decision to sell the dome and make or buy a new one. I always prefer to make things but one came up on Ebay that was worth biddng for at the right price and, having fought off the competition, I made a trip down to Wiltshire to pick up a 2.7m second hand Polaris flass-fibre dome. The dome came down really easily into its component parts and into the truck. Cleaning up the parts from the silicon rubber used to seal it was another matter. I probably spent at least two afternoons using a pressure washer and eventually the flat blade from a plane as an effective means of removing the silicon. The dome then spent about two months sitting down the side alley in parts while I re-worked the plot. Being larger than the previous one in diameter, the concrete block the pier had sat on needed to be moved further in from the corner of the garden, to the new centre of the dome. Moreover, the dome walls of the Polaris dome were half a metre shorter than the previous one. So I had to do something - either slice some height off my pier or raise the floor. THe pier is a 12" tube of steel and I like it being tall to avoid the cricked neck syndrome. Also I didn't want to start slicing the end off or trying to cut a traight section out of the middle and re-welding. Quotes from companies were coming in at silly prices. So I raised the floor. After careful measurement and calculation, with some modelling of telescope size and dome clearances thrown in, I reckoned I needed to raise the floor by about 80 cm in the worst part of a slightly slooping garden. This was to be achieved by sinking 4 4x4" uprights in a square and laying decking on it. First I had to clear the existing concrete to allow a new pier block to be poured. Out came the SDS drill and chisels and an hour later, with the use of a post hole digger, I have a 3' deep hole, 12" round with a 8" deep by 30" square flat slab excavated. The flat slab sides are framed with wood to hold the concrete when opoured and some rebar sunk in the middle to reinforce the long stem. This is a smaller concrete pour than the last one. One big one had been enough to show me that its not really necessary to go massive as long as you go deep and stable. I recall 9 bags of ballast and 2 of cement and some postcrete I had lying around went into the hole, along with some brick ends etc to make up the space. The cenrew was marked by a bolt in the cement which was removed later. This bolt set into a hole in the base of the pier to allow easy marking of the base holes for drilling. The holes were drilled 25mm diameter by 6" deep for 20mm studding on a 18" square pattern. The alignment of the square around the centre mark was determined by sitting the pier on the central stud and using a piece of studding bolted into the centre of the top, the length of the stud was adjusted by moving the bolts until Polaris just crested the top of the stud when viewed from the south reference mark on the top plate. The pier was then placed on top with the studs bolted tight in it at full length while the mortar around the studs set. The pier was used to ensure that the bolts were all parallel as they set. Finally the pier was set to working height and accurately levelled. The four uprights were concreted in place on a square with 3m sides. Four 8" by 2" joists were bolted to these uprights using 9" bolts and fixing spikes, 2 bots to each end, one joist to a side. Between these joists were fixed more joists using joist hangers on a 18" center spacing, except to allow enough space for the foot of the pier to be lifted through the floor, so there was a 20" spacing there between joist sides. I had been recommended to use joist ties rather than noggins. These went in really quickly and easily and stiffened the floor up significantly. The rest of the effort was just laying the 38mm redwod decking and screwing in 100s of screws. I eventually succumbed and bought myself a new drill to do the job with. The dome went on next, we had play with the foot ring, the wall ring and the topping ring, trying to work out the relevant positions until we twigged they were all related to the door position. Once we had that, the rest just fell into place. Thousands of bolts later, the component parts for the dome were all in place ready for sealing. The foot ring screwed to the deck, the bolts all tightened using the drill to torque and sealing mastic used to close the gaps for waterproofing. Finally the foot ring was sealed on the outside using window frame low modulus rubber. It flowed easily into the decking ruts and gaps and seems to have taken well, sealing out the dew that rolls off the roof. Issues to face have been: The original concrete was set in the ground where a compost heap had been. It all dried ut and contracted 8 inches. Hence the ground level was 8" lower than expected once uncovered. So I had to lower the entire decking by 8" or the height of a joist, or build up the concrete 8" higher than expected. I chose to lower the joists since at that point I hadnt put anything more than the 4 side joists in. For power I put in a small frame to mount an external waterproof socket pair. This has an eight-way individually switched distribution board plugged in to power the computer, screen and ethernet-over-power adapter. I also cut a new hole in the pier at the top - so there is now a pair of 4" holes, front and back and a new one at the decking height. This all helps with cable management inside the pier. The internal lighting of the dome I copied from elsewhere - a 8m strip of red leds was fitted around the dome under the top ring, controlled using a Ebay PWM controller unit. At the minimum setting it is still a touch bright so needs a limiting resistor in series. I also built a curved desk which hangs off the top ring bolts and carries a monitor stand pillar. This shelf supports another longer shelf below, where the keyboard, mouse and light controller sit. It all works rather well actually. If anything it needs to be bigger, with holes for eyepieces in and enough space for a log book. The final twiddles include:
Continuing work going forward:Motorise the dome for rotation. I tried using a simple small motor for driving the dome. The probelm wth this version of ppulsar ome is that the inside surface is not flat but has flanges where dome shell parts meet and are bolted together and there is very little space to get under the lip to drive the bottom or inside edges. Hence I ended up looking for a small but powerful motor that could be pressed against the inside edge of the lip and some sort of driving wheel that wouldn't sli[ tooo much. I resorted to right-angle worm driven motors for the output gearing and small size which drive via BaneBot wheel pair. You can get the banebots wheels in differt surface hardnesses and diameters. The 68mm diameter seemed best fit for size vs space. The motors are mounted onto horizontally sprung bogies using bolt shafts, springs and linear bearings. These are then just pressed into place betwwen the dome support wheels and tend not to move. The existing i2C dual motor driver compoents were sued to drive them. In the end to motor turned out to not be enough, so I expanded it to four, equi-spaced around the dome. That takes out stickiness and dome departure from circular. Writing the dome driver in C# under ASCOM turned out to be easy to do and hard to debug. I chose to make it asynchronous to the calls ( i.e. return to the user before completion) but timeouts and wireless radio services to the dome radio receiver caused occasional drop-outs and the dome to be sent off spinning via a mis-read. the code is available on GitHub here Motorise the roof shutter.The shutter on this dome is an up-and-over two-part shutter which slides on teflon sliders, the lower part sliding up under the upper but is quite heavy. My idea was to use 1mm wire rope running on pulleys to the top and returning via cable guides to the motor winch pulley, so the winch can behave like a sailing winch driving an infinite rope and pull the shutter in either direction using a single motor, single wire and single pulley wheel. In this way the bottom shutter is pulled up to the top, by the winch hauling on the cable and collecting wire from the cable guide which emerges at a idler pulley at the top. The lower shutter collects the top shutter on the way up until a switch is depressed and shuts off the motor. The motor is reversed to pull it back down again. The wire is laid down over pulley wheels on the way down until finally the spike on the lower shutter engages the solenoid latch and switch combination to turn off power and lock the shutter in place. Illuminate the steps.Its dark there - I have bought a PIR sensor to detect people approaching and will use this to turn on leds under the steps for footing . Since my wife also painted the door as a Tardis I thought I would also add a sound player with Tardis noise playing when triggered. Might be a bit much after midnight. Add an external scope mount on one of the uprights.So I can do visual while photographing in the dome. |
|