Instrumentation

The pendulum is fitted with an infrared optic sensor which is occluded four times per period producing a rectangular wave. The amplitude at the sensor is about 50mm (100mm peak-peak) and the distance from the fulcrum to the sensor is, conveniently, about 1m - hence the amplitude can be measured in mm or mrad interchangeably.

A GPS receiver generates a 1pps square wave. This signal is accurate to 10ns, so they say. The infrared signal and the GPS signal are digitised simultaneously by a datalogger (ADC). Some clever data processing permits the pendulum period and amplitude to be deduced to high accuracy, once every 3 seconds.

A weather station measures temperature, pressure and humidity. Air density is calculated from these three properties according to an accepted formula. The data is logged once every 10 minutes.

All the data is saved on a local PC running OpenSuse (Linux). The data processing is performed in C++. Data is uploaded to the server every 3 hours.

For more details see Instrumentation and data processing, and for even more go to the Theory page.

There is a lot more we'd like to measure:

motion of the tower (perhaps glitches are related to seismic activity);
install variable drag on the pendulum to neutralise sensitvity to air pressure changes;
air currents up the pendulum chamber (air flowing up or down will change the amplitude and going);
measure the actual time displayed on the clock face (we detect only seconds, and can't tell which second we're on)
sound (it'd be nice to hear the ticking realtime and also the bells)
winding alert (it would be good to get a warning before the clock stops, if we forget to wind it)
and of course we need the webcam working...

If anyone wants to help, then please contact the Keeper of the Clock.

Trinity Clock monitoring

Instrumentation