Confuse the Birds with this Daylight Simulation System

F7WW1DAHQP02H5D.LARGE

Hackers love a challenge. So when [Patrick's] father-in-law asked him to look into a daylight simulation system for his bird breeding cages, [Patrick] quickly discovered the ridiculous prices for commercial systems… so he convinced his father-in-law to let him design and build one instead.

Like any project, [Patrick] quickly listed the requirements of the system before starting anything.

  • It must brighten gradually in the morning
  • Stay constant throughout the day
  • Dim in the evening to a very low intensity

In addition to this, the lighting transition should be smooth, and the lights cannot flicker, as this can be stressful for the birds — oh, and it needs to be reliable.

To build the system, [Patrick] has chosen an Arduino Duemillenova, an RTC, a small LCD screen, some cheap LEDs from eBay and a handful of N-channel NPN transistors to provide the PWM for the LEDs. The entire system cost less than $100 — a much cheaper alternative to commercial systems.

We think the system looks great, but as a big fan of the Hack a Day community, [Patrick] would love to hear your feedback!

37 thoughts on “Confuse the Birds with this Daylight Simulation System

    1. At “An example circuit is shown below” – R3 / 10k is just in parallel with the transistor Collector-Emitter path. This wont keep this pin “low”. A base-current limiting resistor is missing.

  1. “the lights cannot flicker” … “PWM”

    PWM is fast flicker. Birds see things differently and could very well see PWM flicker that humans don’t notice. If flicker is a problem the system should probably use a constant current controller with low current noise.

    1. The controller can do about 4 kHz PWM at 12 bits like the forum thread in the article points out.

      Birds have a critical flicker threshold of about 80 Hz. That’s the fluid transition point where images are percieved as continuous, but as with humans they probably can percieve flicker 4-5 times faster than that up to 500 Hz.

      1. Works OK for things like audio output, but I don’t think that is how LED PWM works. LEDs are (generally) either “off” or “on” and adding a capacitor before it isn’t going to give you an in-between state.

          1. a low-pass would remove the flickering, but he’d need more that just a cap to make a filter he’d R-C or L-C. Add an L C and diode and you basically have a buck converter

      2. I’ve put a couple of sheets of printer paper over a flickering fluorescent lamp and it smoothed it out to a point where I couldn’t see the flicker any more.

    2. Which is why LED was a dumb source. Halogens would be a better choice as they will smooth out anything due to being a very slow acting light source. Plus they would be able to easily get “daylight” bulbs.

      1. Halogen bulbs are just as inefficient as any other incandescent bulb, therefor roughly 90% of the energy provided to the bulb ends up as heat.

        Roast bird anyone?

        1. It is probably warmer where the birds come from ;)
          I’d be more concerned about having to change bulbs every two months and that halogens are not too happy about being dimmed the halogen cycles rely on the
          glass being hot

          1. Halogen doesn’t rely on the glass being hot but on the wire being hot. And dimmed halogen is fine, as long as they are turned up bright once in a while. And here they would be :) I use dimmed halogen a lot and the last longer at 80% then they do at 100% brightness.

            Nice project! I don’t know about what kind of light bird like but they seem to like it. Only weird part in the code are the doubles… Please, don’t use floats when you don’t need them…

  2. This system lacks control of color temperature (CCT), it’s not a daylight simulation, but only a dimmer. Cree for example is now offering modules with sunset dimming – it dims from warm white to reddish white. But it is posible to do this at home – simple use RGB LED controller (or your own construction) with warm white and red LEDs.

      1. And even that won’t be a light that birds confuse for a natural light source as it lacks UV light, which birds can see. To them a RGB (without UV) light will look like just as off-color and tinted as for example those yellow tunnel lights.

  3. ” and it needs to be reliable.”
    “Arduino Duemillenova”
    “some cheap LEDs from eBay”

    Cross the odd one out. Yes – its the first one.

  4. Since the rotation of the Earth does a pretty good job controlling the whole Day/Night thing, what is the purpose of this project? Are they making the bird’s “day” longer? Shorter? And how does the normal house lights figure in? And how does the normal sunlight figure in? Are birds really that picky on when/how they will shag?

    1. - I for instance have kept birds in a room that did not have daylight.
      - Some birds need 14+ hours of light before they enter breeding-mode.
      - I just used TL tubes that also emit UV, and commercial halogen dimmer for sunrise/sunset (the TL tube cut off, and the dimmer started). The halogens were fed from a 12V car battery. No flicker.

  5. Why use transistors? Why not just connect the PWM output to some MeanWell LDD constant current LED drivers, and power the drivers from the power supply. Less parts, cheaper, and still a DIY.

  6. I’d really like to post the details of my similar project online. Is Instructables the best place for this?

    My project’s summary is as follows:
    I have a pet parrot that I’ve had for a couple years now. He’s an awesome Caique… absolutely ridiculous — he knows roll-over and hop on command, and is learning basketball.
    As a pet, you want to keep their hormones down, so keeping their “daylight time” or “awake time” to less than 12 hours/day helps make them think it’s not summer, and not breeding season. So for breeders, you may want days longer than 12 hours, or even adjustable.

    Anyway, I set up my project with a Raspberry Pi running Raspbian, to a couple relay boards (to get the voltages right), to a couple of off-the shelf roller-blinds attached to the ceiling above his cage’s corner of the living room. A cronjob calls a python script and (or webiopi) to raise/lower the curtains for the daylight timing. With the curtains, a CEC signal is sent over HDMI to our TV to give him some entertainment… usually ESPN because he gets excited seeing touchdowns. The curtains are a great substitute for a manual cage-covering dark sheet for a pet, but I think the lighting track is right on for breeders in their own room. A Foscam safely attached to the outside of the cage allows for independent verification of the timed functionality.

    The big advantage to this system is we can give him attention on our schedule, and don’t have to be right there at that exact time in the morning or evening, which works great for his species. Moreover, if someone else is sitting for us or checking in on him over a short weekend, they can make the trip once on their schedule as opposed every morning and night.

    Thoughts and suggestions are welcome!

  7. I see a problem with the concentrated light. LED’s are hard on the eyes when close and not diffused well enough. In that small a space I would have the entire ceiling the diffuser.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s