Tuesday, 23 December 2014

BirdBoxPi2015: Using RaspberryPi model A+

I'm running way behind this year, but I've now started building a RaspberryPi powered nest box for the 2015 season.

The 2014 system will be re-assembled, bolted to the side of a tit box and used again without any real changes.

But the 2015 controller version will be much smaller, and hopefully will fit inside a bird box without the need for its own enclosure.

I've yet to decide whether to build this years version into a tit box (i.e. a bird box with a 25-30mm entrance hole) or fit it inside an open fronted robin/wren box.

This new design is smaller than last seasons because I'm using the new model A+ RaspberryPi.

OK, so there are about a dozen components still to be added to the controller

By building the controller on a small piece of vero-board and fitting it as a daughter-board to the Pi, I've managed to keep the dimensions down to 75mm x 60mm x 40mm.

This includes the wifi dongle which unfortunately sticks out, adding 10mm to the length. I could replace the USB socket for a vertical jobbie (like last year) but I'd then have to cut a slot on the controller board.

By the way, the USB socket on the A+ is still way too tight for my liking. I suggest you slacken the tension clips before trying to force your Edimax into place.

The controller board connects to the Pi via a 26 way long pinned header.

Some of the more useful connections (like +5V) are on the wrong side of the dual header, so I've simply cut off pins 1, 3 & 5 to allow 2, 4 & 6 (the 5V and 0V tracks) to 'run' through without cutting and re-linking the copper tracks. All tracks between the 'odds' & 'evens' pins were cut with a Stanley knife except 1-2, 3-4 & 5-6.

In order to solder the header to the vero-board, I had to allow enough of a gap for the tip of my soldering iron.

However, the header is a good, solid fit to the board, even with 3 of its pins missing.

The Controller Circuit

If I decide to fit this design into an open-fronted robin box, the photo beam trigger circuit isn't going to work. So I plan to try using a PIR sensor mounted inside the box, alongside the camera.

Provisional (still a few things to check)

Also, I won't be able to use bright white l.e.d. lighting because the box is open-fronted, and the chicks will be sitting ducks (er...well, you know what I mean). Therefore I need to use infrared l.e.d. lighting and the PiNOIR camera.

I'm making a few smallish changes to last years design, such as:
  • using a 1N400x to guard against accidental reverse polarity
  • not bothering with a supply fuse
  • using smaller pcb mounted terminal blocks
  • replacing 47uF capacitors with smaller 22uF 16V capacitors
  • increasing the l.e.d. current

The other small problems associated with installing an open-fronted box include location and timing. I need a good, well covered location, and I need to get the box out there during January, because that is when the robins start to look around.

Definitely more to follow...

No comments:

Post a Comment