Make sure the camera lens has an unobstructed view through the hole drilled into the tin.
With the camera installed - the Raspberry Pi and battery pack can easily be lifted out of the tin to allow of access to the various ports. I find it rests easily on the top of the tin when connecting to a display and keyboard if you need to use the device with a display. SSH works without the need to remove if you have a network connection available.
Using the Python script and setting up Raspbian
Setting up Raspbian to control the camera is pretty easy. Using the raspiLapseCam.py script I opted for an automated startup when the device is booted up. That way there's no need for any SSH commands to be issued and it is simply a case of flicking a switch. To do this log into your box using Terminal on the Pi or log in remotely using Terminal or Putty.
ssh pi@XX.XX.XX.XX(where XX.XX.XX.XX is the IP address of your Raspberry Pi)
First of all, we'll need to get the Python script to control the camera. We'll copy this into the /home directory as it's easy to find and a good starting point, but copy to the place the suits you and your build. This is also a good time to create a folder to write images to, and change the permissions to make it writeable. This setup should work as the script writes to the folder /home/timelapse/ as a default; although you can change this if you wish. Just make sure you update the pathway within the Python script itself.
chmod 777 timelapse
sudo wget https://bitbucket.org/fotosyn/fotosynlabs/raw/9304ee9262638efebdd66d543b844a3075401782/RaspiLapseCam/raspiLapseCam.py
To check everything is installed, simply type
The script should be visible in the directory. Next, we can automate the startup so the camera activates each time it is powered up. To do we need to add a cron job. This will mean on each boot of the Raspberry Pi the script will activate and begin capturing images.
sudo nano crontab -e
At the bottom of the script insert
@reboot python /home/raspiLapseCam.py &
Of course change /home to the correct pathway where you have the script.
Save this script (CTRL + X) and "Y"
Rebooting the device will mean this script executes. On testing I found that it needs a full shutdown before this will reliably work each time. So to reuse the device you'll need to log in, and perform a shutdown with
sudo shutdown "now"
Once you have shut down correctly, you can simply switch on the Raspberry Pi's battery power supply and the script will launch normally when the device boots up. This is easpecially useful if you're deploying the timelapse camera for an extended period outside the range of network connections.
Issues with Crontab
There are some issues with using the above script and restarting the Pi. Dougie Lawson has very kindly posted some advice over on the Raspberry Pi forums on how to use init.d to make things startup more reliably.
A transcript of Dougie's comments below...
chmod 755 raspiLapseCam.py
That means you can execute it by typing raspiLapseCam.py rather than python raspiLapseCam.py, which is a bunch easier.
Then we can use an /etc/init.d script to get it started when the system boots.
Start by copying /etc/init.d/skeleton to /etc/init.d/timelapse. cd /etc/init.d && sudo cp skeleton timelapse
edit /etc/init.d/timelapse (use sudo nano or sudo vi) change lines 23 & 24 to refer to the command /home/pi/raspiLapseCam.py you need to start at boot time (the skeleton has good comments). save it.
sudo chmod 755 /etc/init.d/timelapse # to make it executable
sudo update-rc.d timelapse enable # gets it started at system boot and killed at shutdown
Using the camera
raspiLapseCam.py will create a folder for each timelapse session. You can change where this saves within the .py file but as a default this is in the same location in a subfolder. This way you can easily grab all the images from a session and then sequence them into an animation or video clip.
The effects and adjustments are all still available as it uses raspistill to capture images. Simply edit these parameters within the script if you want to make changes.
Next thing is to use it! There are no rules but as a suggestion best times are at dawn and dusk when the light is at its most changeable and dramatic. Many photographers refer to this time as the 'Golden Hour'
Shooting overnight proved a bit more tricky with everything in darkness. I guess there would be a trade-off between the required increase in exposure time and battery life. Under normal conditions I was getting around 7.5 hours uptime using standard Durcaell AA batteries and a Raspberry Pi Model B.