Hi, Newbie question on motors !

Hi,

I'm new here. Hope you're all a nice friendly bunch :)

I don't yet have an Arduino - but I do have a plan. A cunning plan no less.

Well... not that cunning really. I just want to automate my greenhouse ventilation so the heat doesn't kill all my vegetables like it did last year ! I already have one of those cheap wax-cylinder auto-vents for the roof window, but it's not enough in the summer, so I need to be able to have the door open on its own to provide extra ventilation.

My questions are; 1) Do I need a stepper motor or an ordinary motor to open the sliding door ? I want to be able to open & close it in small increments (say 10cm) as needed. Mechanically, I was thinking of mounting the motor to the door frame and running a toothed belt to a pulley on the opposite frame and clamping one side of the belt to the sliding door. The same way a print head works on a dot matrix / inkjet printer. I might need some reducing gears to get the appropriate torque and speed. I was planning on sticking a couple of microswitches at either end of the door's travel, so the Arduino would know when the door was fully closed or open - plus it's a safeguard against the motor trying to continue closing a door that's already closed ! I could use a stepper motor and instruct it to move the exact amount, or a normal motor and just have it running for a brief fixed period each time. Will a normal motor work in reverse ? I suppose the only reason I'm considering a normal motor is that I've presumed they are much cheaper than steppers. What's best here please ?

2) Could I switch the motor with a Mosfet, or would a little relay be more appropriate ?

I've never done anything like this before, but I can program a bit and I'm pretty inventive and practical with mechanics and I know a little electronics. I can see this Arduino thing taking over my life - I'll have automated curtains, a fridge that telephones me when the milk is running low, a gadget to turn on the cooker when I email it, something to feed the dog when he barks a particular sequence... My house is going to be like Wallace & Gromit's in no time :) :)

If the cost is a major concern, DC motors is your choose. A stepper motor is not only more expensive on its own, but also the controller to control it. A DC motor with switches can works just fine, e.g. http://www.youtube.com/watch?v=Z86V_ICUCD4 .

I'd suggest a timed interval with limit switches for fully opened and closed. If you could divide up the total time it takes to drive the door fully open or closed in appropriate intervals. Or you could take it one step further make that total time a variable that the system sets itself by running a calibration cycle once a day or so to make up for variations from wear and tear or motor torque from voltage drop.

I'd also suggest however many steps you create program the last step to fully open or closed run until it hits a limit switch rather than time to ensure fully open and closed just in case things get a little out of whack stepping back and forth earlier in the day.

Thanks for the replies. I've been doing a bit of research and whilst steppers are more expensive than DC motors, they're not that expensive. I could easily afford a stepper and controller. So, I suppose my next question is - which would be best for my application ? Actually, I think I already know... given that I don't really need accuracy in opening the door (just 'another couple of inches' each time is fine), a plain vanilla DC motor will be fine. I like the calibration cycle idea inthebatcave !

I've just ordered an Uno and a breadboard. I've got a funny feeling this could become addictive :) I'm already trying to think of my next project :)

Get hold of a windscreen wiper motor - cheap, powerful and a reasonable speed.

Another common application would be to attach a potentiometer to the vent with a bell crank and linkage, that way as the vent opens and closed the resistance/voltage will change according to position. You wouldn't have to use the limit switches in that situation, but they could still be incorporated into the system as a safety feature.

Thanks for the replies. I think a windscreen wiper motor would be serious overkill - the greenhouse door can be opened with one finger, and a wiper motor would be a bit bulky.

One thing that’s just occurred to me is how am I going to be able to open the door manually if it’s hooked up to a toothed belt & a motor ?
I need to be able to open and close the door manually to get access as well as have the arduino control it when I’m not there. I’ve spotted a geared DC hi-torque motor on ebay that looks ideal - it’s only 3RPM so I wouldn’t need any extra gearing - but… I doubt it would be free-running when not powered (due to all the internal gearing), so with it permanently hooked up to the door via a toothed belt, I wouldn’t be able to just slide the door open myself. I’ve got a couple of stepper motors from some old inkjet printers, and you can slide the unpowered print heads back and fore on those, so they would be free running enough, but I suspect they would be too fast without any reducing gears but impossible to move by hand if I added reducing gears… Hmmm…

This is a sketch of my plan…

What if you used the Arduino to open the door every time? You could have a button on the outside of the door, connected to the Arduino, and when you pushed the button the door opened. Better yet you could use rfid!

http://bildr.org/2011/02/rfid-arduino/

What about two strong magnets as the link between the belt and the door. If the door opens that easy, you should be able to get magnets light enough that they can be pulled apart easily without damaging the mechanism.

That did occur to me ! (opening with a button) However, given that I need the door to open and close very slowly as part of the greenhouse temperature regulation, I don't think I have the patience to stand there and watch it crawl open at a snail's pace ! Maybe there is a halfway speed I could use - something that wouldn't put too much strain on the glass panes in the door plus not keep me waiting too long when I'm trying to get in !

Oooh... another idea... what about dual speed ? I don't know that much about motors or controlling them and my concern is that a motor that is capable of running at a fast speed to open the door when I need to get in won't have the torque needed to open the door when it's running at a crawling speed. I'm guessing I'd have to use PWM to control the motor speed ? Wouldn't that make it slow but really weak ?

I've got a couple of stepper motors from old inkjet printers here.... would they be easier to control ?

P.S. I've just read your newest reply inthebatcave.... I like your thinking ! - very creative ! That would be ideal, but my only concern there is, as you mention, damaging the mechanism. Trying to find magnets that provide just enough grip not to give way under normal operation, yet be weak enough to give way when I pull the door, might be a problem.

I've got it.... a sprung pin. Drill a hole though the top of the door and put a pin that's sprung to the 'inserted' position in there, and have that line up with a hole on a bracket clamped to the toothed belt. I could just pull the pin and open the door.

Ah... getting it to locate, blind, when closing the door is going to be a pain. Back to the drawing board...

You can adjust the magnet pull by just shimming them apart with some thin layers of plastic.

You're a genius ! Magnets it is !!!

Thanks for all your help. :)

I'll be back... !

French spam… unusual !

My plan may be falling apart on the issue of the simplest part of all - the toothed drive belt. I can’t find one !
I need one that is roughly 57" / 1450mm in circumference. I haven’t measured it exactly, but it’s somewhere in this ballpark. Worst case scenario is I have to get two smaller belts and glue them together. Two joins in the belt won’t matter as the belt will never travel more than half of it’s circumference so the joins will never have to go over the pulleys.

I’ve scoured ebay and done a fair bit of googling but I can’t seem to find anything close.

Edit: Maybe I should be looking at chain, rather than belt drive ?

Edit Edit !: Another idea ! (Since I’ve been getting into this stuff, my brain has come alive - it’s never had so much exercise :))… how about I forget the belt or chain idea and just use plain old nylon cord - but - use a weight on the other side to pull the door closed against the motor ?
As long as the motor has enough torque, it can pull the door open and lift the counterweight. I also wouldn’t need the magnet plan then - I could just manually open the door against the counterweight. The only issue there would be the nylon cord (now slack) coming off the motor pulley.

It’s a bit ‘Heath Robinson’ for my liking, but it might work. I’d prefer a more accurate belt / chain drive. It seems a pity to have all those complex electronics ultimately pulling on a bit of string with a water-filled Pepsi bottle dangling on the end. :slight_smile:

After some intensive Googling, I've found this place:- http://www.motionco.co.uk/

It's a veritable Aladdin's Cave of belts, pulleys, drives, wheels etc. !

aneng: My plan may be falling apart on the issue of the simplest part of all - the toothed drive belt. I can't find one !

You need to use "timing belt" as your search term. "T5" is also probably useful (common size of belt, 5mm spacing for teeth).