Gina's Observatory Roll-Off-Roof Automation

[Gina]

13 minutes ago, Dr_Ju_ju said:

Just don't wear yourself out, lugging all that heavy gear around....

I won't!  It's only my medium sized bench PSU at just 30v 5A jobby and not heavy - 4-5Kg.

Edited August 1, 2019 by Gina

[Gina]

Test completed and that will probably all I do today regarding physical effort.

Opening roof/window between 1.2A and 2A, closing up to around 3A.  Fully open or closed stall current 4.8A.  Motor drive with chain and pulleys etc. produced a certain of grumbling and groaning but operated fine.  I think the chain may want lubricating - some of it is quite rusty.

Motor current ranges and conclusions :-

1. > 1A - OFF or clutch disengaged
2. 1-4A - Normal operation
3. > 4A - motor stalled - roof fully open or closed or obstructed.

[Gina]

Arduino analog input :- 0-1023 corresponds to 0-5v so 1A = 140mv = 1023x140/5000 = 28 and 4A = 112.

[Gina]

Left to do on this project :-

1. Sort out limit switches
2. Position rain sensor
3. Design circuit board layout for the electronics
4. Design and print case for the electronics and local control push buttons
5. Build electronics unit
6. Sort out how to start the window opening when it's upright and balanced closed

Edited August 1, 2019 by Gina
Forgot the rain sensor

[Gina]

Yep - still lots to do and only 5 months left this year to do it in!  BUT I want it ready long before the end of the year.

[Gina]

This screenshot shows the connections of the dual motor driver when not used as an Arduino Shield.  I only have one motor so will be using the top half only.

M1EN/DIAG and M1PWM will be connected to +5v.  Arduino outputs will connect to M1INA and M1INB to control motor power and direction.  M1CS will go to an analog input to sense motor current.

Here is a diagram of the Arduino Nano pin connections.  Analog pins except for A6 and A7 can be used as digital, adding up to 6 extra digital I/O lines on top of the 12 standard digital pins.

[Gina]

Feeling better this morning so hoping I can co a bit more to this project today.

[Dr_Ju_ju]

Take it easy, don't tire yourself, keep it simple ......

[Gina]

Thank you Julian, I will take it easy.

[Gina]

I shall put the control box on the wall in the warm room.  That will keep it free from the dew that often descends in the scope room.  The RPi will probably go in a separate box high up in the warm room to get the best WiFi signal.  The optocouplers can go on an RPi HAT together with a buck converter to provide the 5v supply for the RPi.  That will keep all RPi stuff separate from the Arduino.

[Gina]

Before deciding on where the rain sensor will go I need to check that the rain here is sufficiently polluted to trigger the rain sensor - I expect it is.

[Gina]

Been out to the observatory and playing with the motorised opening and closing of roof and window.  The chain was tending to ride up on the outer cheek of the small pulley.  Cured with a nail that limited the sideways movement of the chain.  I can replace this with a 3D printed guide for a more free-running chain.

The limit switch for the closed position will need to be quite accurate.  There is very little between not closed enough and motor stalling.  The open position is much less critical - it doesn't matter if the roof is a few millimetres short of fully open.  It does matter if the roof is a few mm from fully closed as the weather strip will have a gap and not work.  One possible solution to this is an extension spring on the end of the chain.

[Gina]

Ordered a box of heavy duty extension springs  Large Assorted Box of Large Expansion / Extension Springs (From 5.5" to 2")  One of these should do the job and the rest will probably come in useful 😀

Due to arrive tomorrow with Amazon Prime.

Edited August 2, 2019 by Gina

[Gina]

Too hot to work in the observatory at present.  From my earlier visit I decided the control box can go high up on the partition wall in the warm room.  In which case one box could contain both local controls and the RPi for remote control.  That would save the interconnecting cable and some work.  Since I plan to have remote control eventually, I might as well allow for it.

[Gina]

These are the digital input lines for the Arduino :-

1. OPEN - Push button
2. CLOSE - Push button
3. ABORT - Push button
4. Rain Sensor (relay)
5. OPEN limit switch
6. CLOSED limit switch
7. RPi Open
8. RPi Close
9. RPi Abort

Digital output lines :-

1. M1INA - Motor control
2. M1INB - Motor control
3. CSD0 - Current Sense D0 to RPi
4. CSD1 - Current Sense D1 to RPi

Analog input :-

1. CS - Current Sense voltage

Total I/O lines = 14.  This is well within the capabilities of the Arduino Nano.

There is another input that might be advisable which I discovered whilst testing the motor drive and that's the ROR lock down latches.  There's also the tower bolts on the window.  I could just provide an OFF switch on the control unit but I couldn't trust myself to switch OFF when the roof is locked down.  "Locked down" is another signal that could be sent to the RPi for display indoors.

[Gina]

Think I've sorted out the window - pulling it open.  Bungee cord from the top of the window to a fence post.  Simple but it works.

[Gina]

If I install the gate hangers as a ring and pin system to prevent the roof lifting off in a storm I may not need the lock-down latches.  The motor drive will prevent the roof opening as long as the clutch is engaged (if not engaged the roof could be blown open!).  I'm going to have to think hard about this. 

First, I thought I could detect if the latches were locked down by checking for an electrical circuit between the two halves but then I realised that the roof opens and the top halves move.  I could run a wire between the two top pieces up under the roof and down the other side but this would only close the circuit if both latches were locked.

Edited August 2, 2019 by Gina

[Gina]

I have a linear actuator that I could use instead of the latches to hold the roof closed.  This could be remote controlled.  It has it's own built-in limit switches.  The ring and pin roof holding system on all four corners in combination with a motorised bolt to hold the roof closed, should be just as strong as the farm trailer latches I'm using ATM.  As for the bolts that hold the window closed, these are probably not necessary - certainly not on the north side as the cord and pulley system holds the window closed.  I could use a solenoid operated latch on the south side.

[Gina]

This is looking like failing the KISS principle.  Getting complicated.  OTOH stalling the motor trying to open the roof when it's locked down is not really recommended!

[Gina]

Tomorrow is another day and looking forward ti receiving the springs to try.  Not doing any more on this job tonight.

[Gina]

Ordered a couple of these limit switches - I like that they're adjustable and waterproof to IP65.

Heschen CNTD Limit switch TZ-8108 Adjustable Roller Lever 10A 250VAC SPDT Momentary for CNC Mill Plasma IP65

[Gina]

The springs have arrived and the biggest one seems fine - I have it fitted.  Here's a photo.

Edited August 3, 2019 by Gina
Photo added

[Gina]

Drilled the pin plates for the roof frame bolts so that the plates can go in the right position and attached them to the roof framework.  They will be held level by wood screws on some of the countersunk holes.  I shall need to shorten the coach bolts to clear the hook plates with a hacksaw.

Edited August 3, 2019 by Gina

[Gina]

I think that will be all I'm doing today.  The limit switches and coloured push buttons are due to arrive tomorrow.  The hook/loop parts of the storm roof lift prevention system have one square hole and two countersunk holes.  I already have coach bolts that fit the square holes and the others will have wood screws to hold the hooks straight.  I shall probably fit those tomorrow too.

[Gina]

This quote shows what I'm talking about though this shows the anti-lift fittings higher up than I want.  With it lower down, the square hole will be in the middle of the top rail of the observatory fixed framework.  The coach bolt will go through that top rail and, with the wood screws, provide a very solid fixing.

On 16/02/2019 at 09:31, Gina said:

Here's a photo of the SE corner roof retention system roughly fastened in position.  This is one possible position, it could go lower if I were to add a block to widen the vertical frame member and move it away from the corner a couple of inches.