Sometimes we hear from people who tell us that MicroSet readings are unstable. A new customer may say "I've been working on clocks for 15 years, and I KNOW this clock is not that erratic." Actually, the clock is almost certainly as erratic as MicroSet shows it to be. The following graphs will help you understand why.
The graphs which follow were prepared with the MicroSet Windows Interface Software. You don't need to use this software to use MicroSet, but the computer makes it much easier to understand the changing rates of your clocks and watches. Though these examples use a clock movement, the same issues are true with watches.
The first thing you have to understand about clocks (especially) and watches (to a lesser extent) is that the escape wheels exhibit large fluctuations of rate. They're often not perfectly round and, with age or damage, the spacing between individual teeth becomes inconsistent. This is not something you could easily see without MicroSet, but it has a very large impact on the readings you take with any electronic timer. The examples that follow use a new Hermle model 771-000. I use this model because it has no strike train to interfere with measurements.
The graph below shows this Hermle movement measured with the MicroSet acoustic sensor. MicroSet was configured to read every other beat. This means that each reading is the measurement of the same pallet striking each escape wheel tooth one after another. This shows how consistent (or inconsistent) the tooth spacing is. The scale of this graph is located in the lower left corner. It says "Each grid line = 12.4 mins/day". This means that each horizontal line in the graph represents a rate change of 12.4 minutes per day. The data values in this graph reach the top and nearly to the bottom, covering at leat eight grid lines. Eight times 12.4 is 99.2. This means that this clock changes rate by 99 minutes a day with every rotation of the escape wheel.
You may find this hard to believe, but it is true. Without an accurate timer and software to record every other beat, you would never know. Notice that, in each cycle of the graph, the exact same pattern is repeated (some lines stick up higher than others). This is because the spacing of the escape wheel teeth is slightly irregular. The pattern you see is the pattern of tooth spacing. The escape wheel rotated six times in this graph.
This Hermle escape wheel has very consistent tooth spacing but is mounted with considerable eccentricity. If you think escape wheel flaws are unique to modern Hermle movements, click on the links below to see graphs of several other escape wheels: