Observatory Automation

[martin_h]

Obs Automation....

After reading many postings on this forum I decided to automate my obs, partly because it would be an interesting project and partly because the running rails on my roll off roof needed some attention after many years of service.

I decided to go for a garage door opener driven system as this looked to be the easyest system to implement as all the needed kit was in one package. I also decided, while I was about it, to make the roof lighter to provide less work for the motor/gearbox.

Like others I went for the sliding gate track and wheels as these seem to be robust and low in friction so 3 x 3mtrs of track and 4 wheels were duly orderd along with a few sheets of lightweight  corrugated bitumen roofing and a few lengths of timber...just in case!

Day one consisted of de-mounting scopes and cameras and packing them away and covering the mount with bin liners because knowing my luck as soon as the roof comes off it will rain. Now, my original running rails had developed a twist - part of the reason for the refit, after removing them it became obvious why they had twisted.....2 of the screws holding it to the obs wall had sheared!!! So this was to be a timely project. New rail supports were soon installed and fixed this time with 2 x 10mm diameter coach screws. (they aint going anywhere this time!) The metal track was loosely layed on the supports until the final position was determined.

The roof itself was soon sorted, the original 3/4" osb and felt removed, a couple of support rails installed and the bitumen sheets fixed in place. after proping up the roof, the rails holding my original running wheels were removed and the new rollers fixed in place, again with coach screws, those rollers are no small thing! The whole lot was lowered onto the rails and run back and forth to settle the rails into the correct position then it was fixed down.

Now for the automation bit..........

These garage door openers are designed to hang from a ceiling and run back on its track, pulling a garage door below it up and over a pivot, my track was going to have to be mounted "upside down" and pull a roof mounted above it, which was a fairly easy task and just involved making a support for the motor end and a new link to attach to the roof.

I had decided, in the planning stages, to drive the roof motor directly and not via its own control board and to that end the circuit board was soon removed, it was handy that it was connected with spade push on connectors, so now I have a motor/gearbox sprocket and chain and need some way to drive it all. And after many internet searches decided on the LesveDome driver,  which http://www.dppobservatory.net/ROR/ROR.php is mainly for driving an observatory dome. but can also be used for a roll off roof by using the shutter control part. Using the supplied circuit diagram a relay board was constructed, the only change made was a 24V supply was used for the motor and 12V for the control side of things, 2 micro switches act as limit switches to control power to the motor when fully open or closed, also even tho the roof is still heavy there is a small amount of run on when the motor is stopped I can use this by using the original magnetic reed switches that came with the opener set a few mm past the limit switches to report back to the circuit board that the roof is actually open or closed, they also cut the power to the motor and so have the added bonus of a backstop should one of the limit switches fail.

Testing was a bit of a nervous time........will it start, and more importantly will it stop?

Well that was a mixed bag because it only ran in one direction! why oh why did I mount all the circuitry in a box on the wall, under the worktop....out of the way?? So after an hour with the circuit tester on my hands and knees under the bench it was found to be a faulty relay, mechanically it was working, electrically it wasn't! I am so pleased I went for the slightly more expensive option of plug in relays! A quick trip to Maplins the next morning and all was sorted!

To be continued..........

[martin_h]

Obs automation pt 2

Ok so now I have a nice new roof that will open and close at the touch of a button or the click of a mouse (her indoors was less than impressed), what now?

Remember in Pt1 I mentioned the possibility of rain when the roof was off? Well..........  Not quite when the roof was off but the obs did fill with rain. Now, the channel that the opener chain runs in has to sit partly outside and I knew at some stage I would have to cover it to protect the chain and pulley from the elements, but I hadn't thought about it acting as a gutter!!! As I said in Pt1 the channel is upside down and the end outside is blocked off with the pully connection so the channel filled with rain and just ran out of the open end, which was inside the obs, it must have hosed out of the end because its about 8" away from the end wall but there was a huge wet patch on the wall and the floor was soaked, good job I used the bin liners over the mount! Luckily the next few days were sunny and warm and the floor soon dried out with no damage done.

The cover over the track had to be a moveable one because the running gear needs to move along the length of it, so I came up with a length of gutter inverted over it and fixed to the moving roof so that its moves over the track as the roof opens, the gutter protrudes from the roof when fully opened, but that's in the dark so no one can see it!

On order was an auto focuser (SharpSky Pro), 7 position 2" filterwheel (Brightstar), cloud watcher weather station (AAG CloudWatcher) and....a new scope! (TS Imaging Star 71) Very soon I had all the bits to hand, apart for the scope!! I had to wait 3 weeks for the scope to arrive and all the time there were sunny days and clear nights passing by and there is only so much you can do running simulations.

Finally came the time to assemble all the bits - adaptor nightmare! 48mm on the scope, 52mm on the filterwheel and T2 on the camera, and after a false(expensive) start got it all cobbled together and mounted in the observatory. I have decided to use Sequence Generator Pro for the control software mainly through cost, but if that doesn't work out I will have to go the Maxim/CCD Commander route, so to hedge my bets a bit I purchased Pinpoint for plate solving as this will work with both packages.

So the first clear night came and it was time to test out all this wonderfull technology, well it fell at the first hurdle....I couldn't get it to focus! and after 2 hours of getting more and more frustrated, abandoned automation and went old school with manual focus and set the timer on my phone when a filter change was needed. I really wanted to test the scope for colour correction and sharpness so didn't want to waste a clear night faffing about with software/hardware problems. I am pleased to say the scope passed with flying colours, no more blue bloat on stars!

So on to the second clear night with another mixed bag of results, still no auto focus but it did find its target, plate solve and centre, and image through 4 filters with no intervention from me.

Night 3....final chance!!

I managed to crack the auto focus problem - user input failure! In setting the step size for focusing I was inputting tens and twenties but not getting a "V" curve only a straight line, so in desperation I whacked the step size up to 100........bingo! The motor/gearbox on these focuser's must have a very fine output. Set step size to 100 and a perfect "V" curve first time, so now on to a full automated run.

I picked the same target as the night before, so I could use a 'solved' image at a template, I had set up the sequence during the day so when it was dark enough I hit the button and stepped back and let it rip.

The sequence ran - slew to position, plate solve and centre until within 5 pixels of set target, set 1st filter (LUM) and run for 20 subs at 3 mins exposure, change filter (RED), refocus and image for 10 subs at 3 mins, change filter etc......now during all of this at about 2.30 am! there was a meridian flip coming up and I had set it to auto flip, now all the time this was going on I was chatting to a couple of astronomers on line who had kindly stayed up to hold my hand through all of this and offer advice...one of whom was taking bets on the flip, cheeky so and so!

So, the flip came, a sync point was issued, the guiding stopped and the mount traversed to its new position on the other side, a plate solve was completed, again within 5 pixels of position, Dec guiding reversed and guiding started and filter changed, re focus and on with the imaging, all without a hitch.

Things to do now.

I need to integrate this with the roof/weather monitor, havn't plucked up the courage to do this yet. also I need to set the auto focus to run with temperature changes, as I noticed the focus slip over the course of the night, but all in all I was well chuffed with the result, now I have to repeat it!

[lunator]

Martin

Nice work, I am starting to think about am Obs, having an automated roof does have an appeal.

Cheers

Ian

[Gina]

Brilliant   I hope to get back to finishing off my automation soon - not very much to do to get the roof operating by electric motor   I need some fine weather though, particularly for testing.  I just can't seem to get that song "It might as well rain until September" out of my mind 'coz it looks like it might do just that