rotation of the escape wheel. The lock face is the portion of the pallet that stops an escape tooth. There are also lift angles on the ends of the pallets (the lift faces) that drive the pendulum sufficiently to keep the clock running, and are subject to wear (as are the lock faces). My first goal was to measure the lift angles. To do so, I measured the pallets from the center of the pivot to the
1.5degrees is most desireable for such clocks. To draw a
drop, which is always adjusted before the exit drop. We found the entrance drop to be too large (due to the fact that material was removed in the polishing process), so we put the pallets in a vice and heated them gently while squeezing, being careful not to break them. This achieved the desired effect of decreasing the entrance drop. Having done that, we next checked exit drop, which is adjusted by changing the distance from the
pallets to the escape wheel instead of opening or closing the pallets themselves. After both sides had equal drop and sufficient lock (to ensure the wheel didn’t slip past or hit the lift face), it was time to adjust beat rate and time keeping!
Beat and Rate Adjustments – Nuts and Knobs
With the movement ticking, the time had come to check the performance of the clock. First, however, I had to set it up properly on the movement stand and adjust it to keep time. The first thing I adjusted was the beat, or the consistency of the
arc. To do this, I loosened the screw where the leader attaches to the pallet arbor, which is the part of the clock that connects the pendulum to the escape pallets, and rotated it slightly so that it drove the escape pallets the same distance on each side of the
pendulum’s swing. To make sure the beat was correct, I used a timing machine (also used to measure the rate). This machine
picks up the sound made by the clock and measures how much time passes between the beats. Then it calculates the difference. After getting the beat nearly perfect,
11