Adding steady motion to a time lapse is not easy


This not-so-simple contraption adds motion to time lapse photography. Not a Rube Goldberg machine by any means, but the design is somewhat convoluted. However, the concept is simple: move a small amount, take a pic, move a small amount, take a pic. A platter spins to pan the camera, and a strip of plywood leans to tilt the camera.


Here’s a time lapse of it in operation (that’s right, a time lapse of a time lapse):


The camera’s shutter release is wired to an Arduino thru a 1N4148 diode.

Shorting the tip and the sleeve releases the shutter


The Arduino controls two stepper motors – one for pan, one for tilt. Each motor is connected to a gear, which is connected to a cylinder (pill container!). Each cylinder has sandpaper on its sides to better grip a length of string. The ‘pan’ string is wrapped around the circumference of the platter, and the ‘tilt’ string is fed around a skateboard wheel to complete its loop. This loop means that the unit can move both directions i.e. up or down.

The stepper motors cannot pull too much, so less friction (via graphite) while panning is necessary. And balancing the tilt arm helps it function with less effort.

A Nintendo D-pad controls the pan (left/right), tilt (up/down), and also the camera’s shutter release (button A). In fact, once the whole contraption is ready to start moving and taking pics, the ‘Start’ button is pressed to initiate the sequence.

NES D-pad controller


Here is an example of what it can capture.


All parameters are adjustable in the code: the exposure time, amount of pan or tilt movement, interval between pictures, number of pictures, etc.