Gina's Observatory Roll-Off-Roof Automation

[perfrej]

Just got it. The largest is SEK 1862 plus tax and freight, the middle one SEK 1215 plus tax and freight. The freight to Olly's last time around was SEK 280 plus tax.

/p

[Gina]

Just got it. The largest is SEK 1862 plus tax and freight, the middle one SEK 1215 plus tax and freight. The freight to Olly's last time around was SEK 280 plus tax.

/p

Thank you Per   Somewhere around £150 by my calculations depending on difference in freight between France and UK.  Not too bad for the right item for the job

Edited January 9, 2015 by Gina

[Gina]

I now have the motor to drive the chain for the ROR.  I plan to use this with the motor in the case with chain guide.  Here's a photo showing the drive sprocket and chain guide.

And here is a diagram of how I think I can arrange the chain drive for the roof - chain shown in red.

[redtail]

I now have the motor to drive the chain for the ROR.  I plan to use this with the motor in the case with chain guide.  Here's a photo showing the drive sprocket and chain guide.

Motor 01.jpg

And here is a diagram of how I think I can arrange the chain drive for the roof - chain shown in red.

ROR Automation Chain 01.JPG

I have the same make of motor.I bought the control unit and motor unit for £50 second hand but unused with no chain or rack.

I removed the motor and limit switches to use with a rack i got laser cut at work and fabricated side wall plates each side. should be finished by sunday .

The model of motor and remote control unit is a duo 650l

[perfrej]

That looks right, Gina. Just lose all the electronics and drive the motor directly.

/per

[Gina]

Thanks Per   Not sure yet whether I shall use relays or an H bridge of power MOSFETs - I rather like the idea of MOSFETs and I already have some.

[perfrej]

Gina,

You should use a REAL H-bridge for several reasons:

Protected by diodes

Reverse polarity protection with external FET

Thermal shutdown

Protection against accidental ON on both sides

PWM input pin

Logic levels

30A capacity, about 8A continuous without cooling

Low Ron (18 mohm)

Current sense output

Suggested product is the VNH-5019, available from Pololu as either a dual chip Arduino shield or a single chip breakout board:

https://www.pololu.com/product/1451

https://www.pololu.com/product/2507

This is the one I have on the bench right now and it is dynamite!

/per

[Gina]

The Arduino shield sounds a very good idea for a couple of reasons   As you say great protection and also very easy to implement - quick and simple.  There's even an Arduino library to make things easier   I presume you've checked the current requirement of the motor - the ammeters I have only go up to 10A and I gues the current may exceed that.  With the shield there are two drivers which it says can be connected to one motor to double the current capacity.

I've ordered the Arduino shield from uk.rs-online.com, should be here Monday.

Edited January 23, 2015 by Gina

[perfrej]

OK, first of all, get the breakout. It is the same thing only you have to solder the pins to the Arduino. It is also much cheaper. The software is the same and it is configurable for which pin is connected to what on the breakout. Second, don't use the software supplied as it is a poorly written library. Better to just check what it does and mimic the code in a better way. It is dead simple to write the software for it.

Our motor uses only 1.5A rolling our 4x5m roof that weighs several hundred pounds. Initial current at startup rarely exceeds 4A, and when the snow is on the roof and it moves with varying friction we do not usually exceed 2.5A. In other words, you are looking at reasonably low currents in your setup. In Provence, which is not H-bridge controlled, I measure the current five times a second, and if ten successive readings exceed 4A I stop it. It has never stopped due to this in actual operation, but if you put your shoulder against it while closing it will trip after about two seconds. That way I can take the varying currents, but a fault in the rails or a boar sleeping on the wall will trip it and save the motor. Let's call it "soft fuse".

In order to cope with initial currents without issues I suggest you use a 13.8V power supply and a 12V 20Ah (or more) battery in parallel. That will absorb all current spikes nicely which a 10A supply may not do. This is especially important if you run other stuff off the same supply. It is the principle employed in Provence (a 45 Ah car battery) and it works very nicely. I have a voltage drop at the computer of about 0.3V with roof running, making it 13.3 or so. I run everything off of that supply so I am battery backed and will pack up and close after 5 minutes of power gone (implementing that right now).

Now, back to current consumption... You are used to a winch motor and they are beasts with power to spare. they are also inefficient and may very well use 5A or more just turning the motor. I am running my tests on a windshield wiper motor, which is very similar, albeit a bit more power hungry, to the garage door motor. You will find that using properly balanced motor solutions will keep you well down in terms of current consumption, something that is generally very good.

So, promise me you will never attempt to use a 1.5 HP winch again! (except for towing cars out of ditches)

All the best,

Per

[perfrej]

I have the same make of motor.I bought the control unit and motor unit for £50 second hand but unused with no chain or rack.

I removed the motor and limit switches to use with a rack i got laser cut at work and fabricated side wall plates each side. should be finished by sunday .

The model of motor and remote control unit is a duo 650l

Gina's motor is a Marathon 800N from Sommer/Aperto. I guess motor wise it is the bigger brother of your's. There is also the 1100N which we use in Provence.

I just found UK prices for the full units: http://www.sommer.eu/mm/mm004/PL_Preisliste_UK_201103.pdf

From that:

I do not know the drive method of the Duo, but Marathon is straigh, standard bicycle chain which is convenient.

All the best,

Per

[redtail]

The duo uses the same method.

[perfrej]

Excellent! That means parts available

/per

[Gina]

OK, first of all, get the breakout. It is the same thing only you have to solder the pins to the Arduino. It is also much cheaper. The software is the same and it is configurable for which pin is connected to what on the breakout. Second, don't use the software supplied as it is a poorly written library. Better to just check what it does and mimic the code in a better way. It is dead simple to write the software for it.

Our motor uses only 1.5A rolling our 4x5m roof that weighs several hundred pounds. Initial current at startup rarely exceeds 4A, and when the snow is on the roof and it moves with varying friction we do not usually exceed 2.5A. In other words, you are looking at reasonably low currents in your setup. In Provence, which is not H-bridge controlled, I measure the current five times a second, and if ten successive readings exceed 4A I stop it. It has never stopped due to this in actual operation, but if you put your shoulder against it while closing it will trip after about two seconds. That way I can take the varying currents, but a fault in the rails or a boar sleeping on the wall will trip it and save the motor. Let's call it "soft fuse".

In order to cope with initial currents without issues I suggest you use a 13.8V power supply and a 12V 20Ah (or more) battery in parallel. That will absorb all current spikes nicely which a 10A supply may not do. This is especially important if you run other stuff off the same supply. It is the principle employed in Provence (a 45 Ah car battery) and it works very nicely. I have a voltage drop at the computer of about 0.3V with roof running, making it 13.3 or so. I run everything off of that supply so I am battery backed and will pack up and close after 5 minutes of power gone (implementing that right now).

Now, back to current consumption... You are used to a winch motor and they are beasts with power to spare. they are also inefficient and may very well use 5A or more just turning the motor. I am running my tests on a windshield wiper motor, which is very similar, albeit a bit more power hungry, to the garage door motor. You will find that using properly balanced motor solutions will keep you well down in terms of current consumption, something that is generally very good.

So, promise me you will never attempt to use a 1.5 HP winch again! (except for towing cars out of ditches)

All the best,

Per

I ordered the shield rather than the braekout board but I have the east end lifting flap to drive as well as the main roof so the second H bridge could come in handy.  I will probably need to source a 12v DC motor gearbox for that.  I may get the breakout to control a motor gearbox for my widefield mini obsy roof.  I promise you I will NOT be using a winch for any of this

That is a nice low current requirement for the motor.  I already have a 13.8v PSU and 45AH car battery providing my main power supply with the same again powering the EQ8 mount with that battery beside the pier.  I'm planning to use my main supply for the roof motor as the roof will not be powered at the same time as the imaging rig and the location inside the warm room next to the partition wall, will be good for the roof motor.

[Gina]

I have been out in the observatory today looking at implementing the ROR motor drive and also looking at the present fold down flap on the eastern wall.  Currently the latter is a heavy wooden framework with shiplap cladding. I do NOT propose to motorise this - it's pretty heavy.  I am going to replace it with a much lighter polycarbonate roofing sheet 10mm dual wall 610mm x 2m.  Not only is this much lighter but will provide a window to let in more light when I'm working in the scope room in the daytime with the roof closed - saving on lighting

The present flap opens inwards which limits what I can have on the mount somewhat   So the choices are folding outwards or sliding up and down in the style of a sash window.  After close inspection and consideration of how it could be implemented the sash window approach won.  I guess the weight could be counter balanced rather like a sash window.  Then I just need to sort out a drive mechanism   Timing belts, timing pulleys and stepper motors would be one option, I guess.

[Gina]

Now back to the main roof automation.

Referring to the latest design for powering the ROR with chain drive as reproduced in the diagram below, the large chain pulley will need to go outside the observatory just behind the dividing wall and will be covered with a 3D printed plastic cover to keep the weather off and provide sealing to keep rain out of the scope room.

This photo (with the roof open) shows where the large pulley will be mounted.  The green arrow points to the framework of the dividing wall and the red arrow a piece of weatherproofing board that will need notching out to take the pulley.

This photo (with the roof closed) shows the outside looking towards the partition wall and west end of the ROR.  The arrow points to where the large pulley will stick outside of the main construction and would be out in the weather.

This photo shows turning the groove in the acrylic blank to form a pulley.  This was originally designed for rope and will be turned to a square section groove for the chain.

Finally, here is a sketch of the cover design for printing in ABS plastic.

[perfrej]

Just let the roof push a lever. I designed one of those approximately and Olly realized with some changes of his. It works very well and is 5m long and 50cm high. Doing it that way eliminates any login in the open and close sequences as well, which is nice.

/per

[Gina]

Just let the roof push a lever. I designed one of those approximately and Olly realized with some changes of his. It works very well and is 5m long and 50cm high. Doing it that way eliminates any login in the open and close sequences as well, which is nice.

/per

Yes, I've been giving this more thought and will be having another look to see if I can make the wall flap rotate outwards as there are definite advantages to that.  I have some aluminium tube of 1" and 1.25" long enough to go right across the width of the observatory that I could use as an axle.  I'm thinking I can modify the original framework and use that but without the heavy hardwood shiplap cladding.

Do you have a link to your design or Olly's mods, please?

Edited January 24, 2015 by Gina

[perfrej]

This was my conceptual of it and Olly just changed some details regarding angles.

Obviously, the roof has not been cut out to make use of the flap in this image...

In real life, this is what it turned out as:

/per

[Gina]

Thank you Per   My design will have to vary a bit from that but I get the idea

[Gina]

Been out in (and outside) the observatory and done some work towards the motor roof drive.  Cut out the slot for the big chain pulley and casing, and drilled the hole for the mounting bolt in the timber upright.

Before

After

[Gina]

One possible idea for closing the wall flap using a roller on an arm on the flap axle and a "pusher" piece on the main roof.

[Gina]

I'm going to leave the wall flap for the moment and concentrate on the main roof.  Most rain comes from the south west and the wall flap faces east with about 6" overhang on the roof including the guttering so rain only comes in if the wind direction is east of north or south.  I don't mean I'm not going to drive the wall flap, just that I shall take more time thinking about it and maybe do it when the weather is better (though there were several hours of warm enough sunshine yesterday).

Meanwhile, any other ideas are always welcome - or comments etc.

Edited January 25, 2015 by Gina

[Gina]

Just for fun I've been playing with SketchUp again and have come up with a new wall flap drive design that gives the full 180 degrees of rotation from the up (closed) to the down (open) position.  A wooden rack and pinion arrangement   This does have the advantage that the wall flap could be counter balanced as the rack and pinion drives in both directions.

Closed

Not quite fully open.

In fact the full pinion isn't needed as 4 of the 9 teeth shown are never used.  Mind you, I think the full one looks nicer

Edited January 25, 2015 by Gina

[Horwig]

I love it, with that mechanism in mind, why not add a drawbridge as well  :grin:    

H

[perfrej]

Haha! Well, our flap rests nicely horizontally on three stands. Now why would you need to fold all the way down? Another type of flap to use would be the one that Peter Shah uses on his roll-off. It is a wonderful design!

/per