Automatic chicken door time input (low power, simple)

[NivagSwerdna]

I haven't completely read the thread but I do notice...

Accepted you don't have the geometry for a sliding door... however you seem to assume this needs lots of power... my Aluminium sliding door opens and closes every day and I only have to change the 4xAA batteries every two years or so.

If you want a RTC.... A DS3231 module is hard to beat value wise... I actually buy them just for the parts which I transfer to my own PCBs. The 1Hz output can be used to wake from sleep if required.  Like any clock there is an error... but it is small.... however you open up the can of worms that is setting the clock.

[Amper]

Oh, im sure sliding doors take less power but its just mechanically not possible in the location im in. Also as i said the mechanical part is not a problem at all, its working perfectly fine with very little power. one AA battery contains roughly 10kJ and assuming ill use 4 and it takes 100J a day with my existing door a set will last around one year ignoring self discharge and disadvantages at low/ high temperatures. Though thats exactly what i said in my original post. My problem is not a power source, that was never an issue, just everyone thought it was. Also setting and operating the rtc never was an issue to me, just look at my picture of the data logger at the first page, i have done all of this before. My issue is only how to do manual setting of the opening and closing times without damaging my power consumption or making a lot of effort in programming easy to use UI with a lot of changes in power management to get my standby current below 10uA again.

[spec]

Hi Amper

What fun this thread is. We used to have four chickens and good layers they were. They were very much creatures of habit and would start getting anxious when the sun started going down and go into their coop for the night. And at sun-up they would be fussing to be free in the garden, especially as we spoilt them to hell with little treats. They were battery hens rather than mains and when their feathers grew, I taught two of them to fly a bit.

I too have thought about a solution to Mr *&%#! fox and the best low-tech approach that I came up with, is to have a medium size, but aggressive dog living with the chickens to look after them. You would not need to worry about locking doors or anything else then.

An alternative approach is to have the chickens unlock the door themselves- don't laugh...yet

So how does the chicken operated lock work:

Chickens are not the brightest of birds, but they are not totally dumb either, especially when food is concerned. So what I table is two pads, one just outside the door and one just inside the door, which the chickens have to peck a number of times before the door opens. Of course, a new chicken would not know what to do, but it would not take long for them to learn, especially if their food is in the coop. I was going to try this approach out, but our chickens left for a posher neighborhood before I got around to it. But I did try out a few similar tasks for them to carry out before they got a treat, and they managed without any problem. They weren't very good with simultaneous equations though.

The actual keypad could be made so that only a chickens beak could operate it, so that any smart-arse fox, even if he knew the code, would not be able to operate the lock with his paw or nose, but I think it is unlikely that a fox would figure out how to operate the lock anyway.

I will get my coat now.

[NivagSwerdna]

In my other thread I devised, aided and abetted by Ian M et al., a device that extended my door gadget to allow manual override of the door opening times... it has two buttons, UP and DOWN (with LEDs in the buttons),,, allows.. UP (until next close), UP(always), DOWN(until next open), DOWN(always) by either a short or long press (and the ability to cancel a currently working override).  Seems that such a combo could be extended to allow an UP to set the opening time everyday and the DOWN to set the closing time every day.  So the user just needs to be present once at OPEN time and once at CLOSE time to set the system.

https://www.eevblog.com/forum/projects/4-aa-cells-direct-to-pic/msg1219663/#msg1219663

[Amper]

Thats actually a really sweet idea! Also it will work very nicely when some day the battery in the rtc fails because it wont care about the absolute time. ill consider this! Also it would make it really easy to write every "correction command" to an sd card so after a few months or even a year i could get a graph and know exactly which times will be suitable for closing.

The chicken training method sounds fun as well, just im afraid they will sit at the door all day expecting food   Also sadly we have a bunch of mice and sparrows round, they will likely learn it as well and roam the place at night.

EDIT:
Also a litte OT: There is a children's book in wich chicken are solen by a fox so they build a chicken from a pepper filled balloon attached to a bunch of fireworks:

[Ian.M]

Yes, NivagSwerdna's learning by doing idea is probably the simplest.  Just add a button to store the last  action as the new opening or closing time so you can still use the other buttons for manual override as required.

It will be a PITA setting the clock in the first place, which you need to do if you want auto-adjustment to track sunrise/sunset so you dont have to tweak its hours so often.  Either make the RTC battery backed and removable so you can preset it before you plug the module in, or connect a laptop or tablet with a terminal program to the Arduino, and set it that way.

[Amper]

I just do that by putting the example sketch on it wich sets it to the current time when it was compiled. Thats not accurate to the second but enough to me. Downside is i need to redo it when i replace the lithium battery but maybe for that purpose i can add a jumper that when set will put time and date to a known value so i have so reference. It should last around 10a, so its not something to worry about for now i guess.

[nsrmagazin]

This became another thread with unnecessary complications.

While you are on it, why don't you put a wifi module so you can check remotely and a chicken counter too.....

Just put a solar cell or a light sensor to a comparator with a potentiometer on the other pin, and use your own system that you have now.

[Amper]

Please go somewhere else for trolling, i had enough of it in this thread.

[nsrmagazin]

Why am I trolling?

[Ian.M]

You appear to be trolling because the O.P. has already considered and rejected wireless connectivity, tagging, counting and weighing chickens etc.  If you are not trollng, then either you are too lazy to read the whole topic before commenting, or you lack reading comprehension and short term memory! 

One point that has not been considered by those who advocate a simple light sensor, is: Adult humans in developed countries  typically set their daily routine by the clock.   Its undesirable to open the coop too early before  humans are up and about, even if its well after sunrise due to the risk of an over-bold fox!

[Amper]

My apologies, i misread your post after what happened here before...

Your idea with the comparator and timer is nice and light sensing is still in the race. Though since i need the microcontroller anyways for coordinating pumps and valves i wont have to build it the old school way. Just throwing in one or two lines of code will be faster to me than building a new pcb, programming i can do in my college dorm or in the train and then just upload it when i arrive home.

Using the solar cell as light sensor will propably be done as well since im interested in the power production anyways for other projects, so i will have a shunt in there and measure over the day.

[Doctorandus_P]

Well, you said you were asking for random ideas in your first post
And how would anyone know to search for "RTC" if they did not read the post history?

[Amper]

Well yes, i guess i asked like that. I dont expect anyone to read everything, thats obviously to much but even in the very first post i mention that im using an rtc for example. It just seems like people only read the title and then post what ever without considering at all that they are the 5th person by now suggesting the exact same thing that is also specifically excluded in my question.

Im still very happy for all your suggestions just at some point i got a bit frustrated...

[soldar]

Search youtube for "chicken training"and you will be amazed how chickens can be trained to do a lot of things, including come home when called.

[soldar]

You could combine the door opening and the lock actuators into one. Have it actuate on the door along a slot and a spring. In the drawing it is locked. Then the actuator starts moving back and unlocks. Finally it reaches the end of the slot and starts opening the door. For closing it is the same in reverse.

[Amper]

If i understand it correctly opening would work but on closing the latch would hit the ring because there is nothing to compress the spring. Its possible of cause to get rid of one cylinder by nice mechanisms but i really didnt have to because by connecting them in the right way and adding a pneumatic momentary switch i still only have one air line for opening and one for closing. The hardware was leftover scrap anyways and its nice not to have it sit in a box unused until some day ill throw it away.

[Ian.M]

No, I think Soldar's drawing shows a bracket on the door retaining the spring.  Therefore the spring force is available to close the door.  However if the hinges get even a little stiff, the bolt will protrude before the door closes, stopping it locking.  Also it needs a hinge joint where the piston rod connects to the bolt or it will jam on opening. 

As you've got a well proven mechanism and plenty of spare parts, much as it hurts my parsimonious nature, I am forced to agree that changing it is not worthwhile.

When you do get done with this, please update the topic with schematic and code.  Even if others will have to mod the design for different door openers, its still valuable.

[Amper]

I will!

[soldar]

No, I think Soldar's drawing shows a bracket on the door retaining the spring.  Therefore the spring force is available to close the door. 

Yes, the bracket with the slot is attached to the door.

  However if the hinges get even a little stiff, the bolt will protrude before the door closes, stopping it locking. 

You can use a spring as stiff as you want so this should not be a problem. Plus hinges with a drop of oil should go smoothly for years. you could substitute a slip latch which can be closed without being pulled back. A more likely scenario is a stick or a pebble prevents closing.

As you've got a well proven mechanism and plenty of spare parts, much as it hurts my parsimonious nature, I am forced to agree that changing it is not worthwhile.

Well, yes, from the beginning this was a general discussion thread on the topic as it was clear the OP was satisfied with what he already had. So I think the discussion might be of interest to others pursuing similar solutions.

[Amper]

Yep, suggestions and general ideas are welcome if not repeated a ton of times

With the slip latch i agree it would work but with a simple spring loaded slider thespring has to hold the latch closed, so when closing the bold is already extended and will hit the hole from the outside. When the spring is loaded the other way pushing the latch open it will operate nicely closing and opening but when air pressure is relieved it will push the latch open.