I always wanted to have a network camera that provides a good image quality but does not cost a fortune. Modern network cameras that have a resolution of 4K cost upwards of 4000$. There are some cheap cameras that actually cost less, but their image quality is most often not that great.
A couple months ago I made a great deal on some second-hand Nikon Coolpix L31 cameras. I paid 15$ for two pieces of them, because one had a crack in the display and the other one had a broken battery lid.
My idea was to connect the cameras to a Raspberry PI and use gphoto2 to take pictures and send them back to a sever over the network. The only thing I needed was a outdoor weather-proof enclosure. I found a suitable one at a local hardware store.
The enclosure I used is actually the body of an halogen spotlight used on construction sites. It is perfectly suited for this purpose as it is IP44 protected and has a glass front through which the camera can take pictures without being exposed to the elements. But the best thing is the price as it costs only 12$ which is an absolutely amazing price for such an enclosure.
Before I go into details how I built the camera, here are some sample shots of it during the night and in daylight.
If you pay close attention to the picture during the night you might notice the orange reflection on the top. This is because I forgot to cover a LED of the Raspberry Pi with black electrical tape. I will fix this soon 🙂
When I started to build this I first removed the guts of the lamp and installed a 3D printed mount for the camera. I designed the mount in OpenScad and you can find my the files for the print
on my Thingiverse account here (actually I currently am not able to verify the mail address of my Thingiverse account as I only get an HTTP 404 error). The slots fit perfectly for a camera screw.
I used the following camera screws: Link to Amazon
The camera holder is attached by the two screws at the bottom.
To mount the Raspberry PI I used nylon screws with a nut used as a spacer.
In the back of the enclosure a plate is mounted and on that plate the Raspberry PI is screwed on.
I had to cut away some of the plastic of the micro USB connector because it was too long and touched the case.
After the PI was mounted I inserted the camera screw into the camera holder.
Then the camera was screwed on and the network cable was fed through the hole and stripped.
As the camera needs a 3 volt supply, I build one with an LM350 which is quiet overkill. It does not need the current supply capability of 3 amperes. In this picture I had not added the capacitors which should be used with the linear regulator. On the bottom is a DCDC converter from ebay.
The network cable is also used for power according to the power over ethernet “proprietary” standard that uses the four unused wires of an ethernet cord and provides power and 100 Mbit connectivity.
You can find the appropriate wiring here.
The four power wires were attached to the DC DC converter which was set to provide 5 volts for the PI. I attached the USB cord of the PI to the output of the DC DC converter.
From the 5 volt output two wires were attached to the input of the linear regulator. The output of this regulator is attached to wires that are soldered on to the battery tabs of the camera.
In this picture you can also see the USB cable used to connect the camera and the PI.
The finished assembly with the voltage regulators on the right.
After completing the assembly of the camera enclosure I attached the ethernet cord to a power over ethernet switch and tested the connection to the PI.
The power over ethernet switch is a modified cheap ethernet switch to which I added a DC DC converter and connected the appropriate wires for power between the ethernet plugs.
Now the camera is operating for some weeks and I have not seen any problems due to rain. The only thing to pay attention to is to cover every LED of the Raspberry Pi so that your night pictures do not show ugly reflections from the glass.
In part 2 of this post I will provide a small description howto compile gphoto2 and install the necessary scripts on the PI to regularly take pictures.
You can follow me on Twitter so that you do not miss the second part.