Gina's Observatory Roll-Off-Roof Automation

[Gina]

Having checked that it fits perfectly I have now dismantled the window and applied Ronseal Total Wood Preserver to all the timber parts starting by soaking the ends by immersion for at least ten minutes and then applying to the rest with a paint brush.  Now drying.

[Gina]

While we've had a spell of sunshine and reasonably warm, I've been re-doing the weather proofing around the roll off roof.  Turned cold now so packed it in for today.  I'll take and post some pics later.

[Gina]

Re-assembled the east wall flap/window, took a couple of photos and re-fitted it to the observatory   So it's now ready for sorting out the rope and pulleys to control it and fit the weather proofing.  I have also been sorting out the weather proofing of the north side of the roof.  Quite a wind this afternoon and it kept blowing the roof closed   The wind has died this evening and it's clear   I was expecting to have to wedge the roof open for imaging but with no wind it isn't necessary

[Gina]

Draft brush strip added to window frame to keep the rain out   Roof is now fully weatherproofed with the window replacing the heavy wooden wall flap.  Window currently held closed with the rope with a loop hooked over the roof clamp down buckle.  Now going to produce pulleys.

[Gina]

Now have one of the pulleys made and fitted   Some photos... 

Note :- The galvanised iron pole (weather station) will be moved out of the view when I get round to it.  Also the apple trees will be pruned   The heavy wooden wall flap is still attached and will be removed later once I'm satisfied that the new window is totally satisfactory - it isn't in the way.

Open

Closed

This shows the south side of the window when closed, window frame fitting against the roof frame.

[Gina]

???? :confused: :confused:

[Gina]

Raining now

[Gina]

Been thinking about open and closing the window from the rolling roof.  A lever operated by a roller on the roof carriage would pull the rope to close the window, as shown in the sketch below.  The problem is that with this arrangement, quite considerable force is required to pull the window from the horizontal (open) position.  Far more than that required to open or close the roof.  Initially the force on the lever is small with the rope and lever starting in line but increases radiply as the lever is pushed.

I think the electric motor that will open or close the roof may be quite strong enough but I shall be putting more thought into operating the window.

Edited March 28, 2015 by Gina

[Pan Euro]

Have you tried moving the attachment point from the middle of the window frame to near the top it may make it easier to move. Just a thought.

[Gina]

The sketch above is not quite accurate - the pulley is actually at the top of the north wall and that is as high as it can go.  See post #255 above which shows the actual photos.  The attachment to the window frame matches this.  With the pulley in this position, nothing is gained by moving the attachment point of the rope onto the frame.  Thank you for the suggestion

[Gina]

I have come up with another idea.  This is based on making use of the full range of the roof matched to the range of the window movement to give maximum mechanical advantage and reduce the force on the roof.  It is based purely on rope and pulleys.  The rope from the window frame goes over the top pulley and down nearly to the floor where it's attached to the frame.  Pull is applied about half way up which minimises the tension required to lift the window from fully open.  A pulley is attached to the middle to provide a 2:1 purchase with the rope from the roof carriage.  I checked how far the middle point of the down rope had to move to close the window and I found this to be about half the roof movement so concluded that I could use a 2:1 purchase in addition.

[Gina]

Another idea for opening and closing the wall flap window would be to use bicycle chain driven by a windscreen wiper motor with a sprocket on the output shaft.  The motor could be driven by the second motor control H bridge on my Arduino motor shield.  Or I could use a drum on the wiper shaft and wind up the rope - less accurate but not a problem with limit switches on the window frame.  The wall flap/window only needs opening when the object being imaged is low in the east - in other cases it can remain closed.

Edited April 7, 2015 by Gina

[Starlight 1]

Still keeping a eye on your post ,was thinking of doing the same . So I an just leting you do the hard work in working it all out.

[Gina]

Not posted in this thread for some time as I've been putting more emphasis on other projects but minor things have been going on such as printing brackets for the chain ends and thinking about such things as limit switches.

Regarding limit switches, three types come to mind :-

1. So called micro switches with extension levers - difficult to seal against the damp.
2. Reed relays operated by magnets on the ROR carriage.
3. Hall effect devices again operated by magnets on the ROR carriage.

I've ruled out option 1. so considering the other two, either of which are fine for use with the Arduino.

I shall also want a sensor on the mount to ensure that it's parked in the Safe position before moving the roof, though this only applies to the MN190 as shorter scopes are well clear of the roof in all positions.

[Gina]

With the lovely sunny weather today I hopped over the electric fences into my neighbour's field and took some more photos of the south side and ends of my observatory, showing the new east wall flap/window.  It's now ready for connecting up the motor for remote/automatic control.

Edited May 16, 2015 by Gina

[Gina]

With going over to Raspberry Pi running Linux (Ubuntu Mate or Raspbian) with the KStars/Ekos/INDI control and capture system for astro imaging, I'm looking into also controlling the roll off roof and fold down flap/window remotely using the same system.  I won't actually be doing anything to the hardware until the weather improves but nothing wrong with planning and designing the system in readiness.

There is already an INDI driver for roll off roof control in the Dome group which seems to cover most of what I want and I'm already modifying drivers for my astro imaging rigs so I plan to edit the source code to cater for my own requirements.

My observatory presents some extra problems that a standard ROR (if there is such a thing) don't have.  Due to the exposed location of my observatory it has proved necessary to provide substantial lock-down latches for the ROR and it wouldn't be practical to automate these.  Other ideas for stopping the roof lifting off in a storm would require considerable modifications to the structure.  I will need to detect if these are in the lock-down position and prevent the remote control trying to open the roof.

Another problem is the fold-down window which I was unable to mechanically connect to the roof roll-off (like others have done) so this will probably need a separate motor/actuator.  Again, the opening mechanism is backed up with bolts to fix it shut against storm force winds - it's amazing how wind can suck things open as well as blow things around.  Clearly the opening mechanism can't work if the window is bolted shut.

I don't see any particular reason why I can't arrange sensors on all the latches and bolts that can feed into the roof control system but I will need to look into controlling the window opening as it currently needs a push to open it due to the brush type draught excluder used to keep the weather out.  So the window will need pushing open as well as pulling closed.  ATM I'm using cord to pull the window closed and hold it just below horizontal in the open position.

Edited February 26, 2017 by Gina

[Gina]

If anyone is thinking "why don't you just not try remotely opening the roof when it's locked down", then the answer is that I can do all sorts of silly things that I shouldn't!!  I definitely need anti-stupidity devices!

Also, I know that closing roof and window remotely doesn't make them storm-proof but I wouldn't disable the storm-proofing if there was any chance of a storm developing.  However, as a corollary to the above, I could forget to lock down before the next storm arrived.  I could tie in the observatory control system with my automatic weather station to sound an alarm if the wind speed exceeded a certain value and the storm-proofing was disabled.  I plan to add a raindrop detector to tell INDI system to close the roof and window.

[Gina]

I've been thinking about the flap/window.  The only sealing brush actually holding the window shut is the top one that brushes against the top of the roll-off-roof and once the roof is open slightly it ceases to hold the window shut.  So as long as the roof opening motor is strong enough to pull the roof open against this resistance (and the other causes of resistance), the window should open easily.  Possibly a weight on a lever on the axle, or a spring, could pull or push the window open against the cord.

I've also been looking again at ways of allowing the window to open as the roof rolls back and the "crank lever" system shown above might work and is relatively simple.

[Gina]

Sunshine and showers today and relatively mild so I've been out to the observatory and checked up on the roof and window.  The weather strip on the top of the window doesn't add any significant resistance to opening the roof - I can push it open with one finger   The window is balanced closed until at about 5° open and by 10° will open on its own just held by the cord.  A spring or weight could easily start the window opening.

The length of cord between open and closed window is about 2ft so this is the distance the end of the crank must move to open the window (or close it).  The uprights of the observatory framework are also about 2ft apart.  I can now look into the geometry of the crank and how it can be implemented.

[Gina]

Here is a SketchUp model of the window crank to scale, showing partly open and closed positions of the roll-off-roof.  The black disk is a wheel which acts as a roller on the longer lever arm.  Green is the cord pulley.

Edited February 27, 2017 by Gina

[Gina]

Looked at using the NoIR RPi camera to remotely watch the imaging rig on the mount to make sure it was pointing roughly where expected and found that the INDI driver indi_v4l2_ccd works with the RPi camera so I can use this with the KStars/Ekos client.  No doubt combined with the ROR control.  But I could set up an RPi with camera any time now to check what's happening now that I have my widefield rig mounted on the EQ8

[Gina]

Been looking at the fixings for the chain ends onto the roll-off-roof.  The "pull it open" end is relatively easy, a block of wood with the chain screwed onto the bottom.  The block would have its bottom side level with the pen mark on the draught excluder.  It would stick out just a bit further than the aluminium channel that holds the window panel.  The chain would then line up with the top of the large pulley.

The "pull it closed" end is probably a bit more tricky as it has to clear the large pulley on the dividing wall and the chain from the bottom of that to the drive motor whilst not being too far off straight to the smaller pulley.  A block of wood might do it - I need it check it out more closely.  Here's a rather poor photo - I'll take a better one when weather permits and maybe from another angle as well.

 

Edited March 2, 2017 by Gina

[Gina]

Since I seem to be getting back into imaging I think there's a very good chance of resurrecting this project.  The main outstanding problem is the fold down window flap but if rain arrives it's usually from the SW and the window faces east so wouldn't matter if it were left open.  The storm latches and bolts can't really be automated so there would be a manual "pre-opening" anyway.

[JamesF]

Have you seen the animation of one of Olly's observatories with a fold-down flap, Gina?  It has two "outriggers" that press against extensions to the roof frame such that the flap is lowers as the roof rolls away and lifts back up as it is closed.

James

[Gina]

Yes, I have.  I may be able to do something like that.