Troubleshooting

Sometimes a mechanism just doesn't want to run or strike correctly - despite pretty much everything looking like it should want to run. On those occasions I set up the mechanism, with its pendulum, on a bench next to my main bench and hook up both a beat amplifier (Time Trax), as well as a beat monitor (Bryan Mumford's Microset) that allows me to plot the beat and/or strikes on my computer monitor, which sits above my main bench. This gallery gives you an idea of how I set all this up.

9 / 24

Microset Program Screen Monitoring only the Tick or the Tock

If you only monitor every other beat (think monitoring just the tick or the tock) you get a pattern like this. Here the tick starts out longer than the tock, then becomes shorter, then longer. This is about the best you can do for an escape wheel that is slightly out of round, or slightly off-center to the arbor. Flip side, if this mechanism was driven with a heavier weight – this pattern would virtually disappear - with the greater overswing that a heavier weight would produce.

Let’s think a little more about what we are seeing. While the pattern above and below the average are not exactly a mirror image of each other, they are, in fact, surprisingly similar. Why would they be similar? Because the long tick (caused by a longer escape-wheel tooth, and not caused by a wider spacing between teeth) will become a long tock when the same tooth comes around to the other pallet. And, in fact, there are instances in the above plot where the pattern above the average looks pretty darned similar to the pattern below the average.

What I am also learning through work like this is that variations in the power supplied to the escape wheel come from not only things like out of round wheels or un-even tooth shapes, but also from where a wheel tooth and pinion leaf are in their interaction (more on this in a subsequent slide). The power going to the escape wheel varies continuously – and, in a mechanism that is running on near-minimal weight, these variations show up in the beat patterns.

The background on this particular beat trace is that I tipped the escape-wheel teeth with the wheel’s pinion chucked up in my lathe. One would hope the pinion was concentric with the arbor/pivots. It wasn’t. Re-tipping with the wheel chucked up on the arbor, removing perhaps a thousandth’s of an inch from the longer escape-wheel teeth, significantly changed this pattern - as see on the next slide.

Screen shot annotated revised