The origins of this project were pretty simple… I had an ultrasonic distance sensor, wrote some code to use it, and thought this would be a good way to test it out. I was having fun learning Arduino and had previously experimented with rewiring and controlling clock mechanisms, so why not stick the two together?

The clock control is based on the method shown here, and involves soldering a couple of wires onto the existing circuit so that Arduino can be used to replicate the motor pulses to drive it.

A small green clock circuit board with red and yellow wires soldered onto its contacts, held in a helping hand next to a battery.
Modified clock driver circuit.

For some reason I wanted to be able to run the clock normally too, so didn’t cut any tracks and added extra wires to pass in power to the full circuit (the original battery connections had to be severed for the modifications so I couldn’t use them, but there happened to be some unused power connections on the PCB that could be used instead).

The modified board fitted snugly when returned to the enclosure, and a section at the edge of the plastic case was cut away to provide a small exit hole for the wiring to fit through once the cover was replaced.

The opened clock movement, showing its white plastic gear housing and copper coil, with a bundle of coloured wires leaving through the side of the case.
Wires exiting clock mechanism - rear cover removed.

The distance sensor was connected up and ‘mounted’, dangling below the clock face, pointing out hoping to detect anyone that passed by. Power was supplied by an extra (larger) battery pack fixed to the back.

The performance and behaviour of the clock is not too bad, but could certainly be improved. Power consumption is (unsurprisingly) quite a bit higher than on the unmodified clock, and the second hand can occasionally take a couple of tries to fully tick onto the next position.

I think either: transitioning from one tick-rate to another causes issues with the mechanical movement of the mechanism, or it’s down to timing issues with the motor pulses themselves.

The bottom of a round wall clock with a blue ultrasonic sensor board hanging from its rim by a short ribbon cable.
The watched clock

Making rate changes more slowly could help things stabilise a bit if the mechanism doesn’t like rapidly changing rate, or if pulse timing is a problem it might be better to control an external timer that generates a stable clock pulse directly on the PCB to replace the oscillator that drives the IC on the board.

I might try those improvements later on, but most likely on a different project.