Two machines reading the same watch disagreed by about 30 degrees of amplitude because their default lift angles differed. Both were correct. The number published without its angle was not.
Two machines, two defaults, one watch
issue ISSUE 01written 2026-09-10responsible Escapement Papers (a named bench, not a person)
tape card: the two readings
| Machine | Default angle | Amplitude reported |
|---|---|---|
| Machine A | 52 degrees | 282 degrees |
| Machine B | 44 degrees | about 312 degrees |
| After setting both to the movement's 48 degrees | 48 degrees | about 297 degrees on both |
What happened
Both machines were reading a movement whose published lift angle is 48 degrees. Machine A defaulted to 52, machine B to 44. Neither operator had changed the setting, because on both machines the setting is buried and neither manual makes a point of it. The rate readings agreed to within a second per day, which is exactly why nobody noticed.
| Reading | Sensitive to lift angle | Why |
|---|---|---|
| Rate | Almost not at all | It compares the beat to a reference frequency |
| Amplitude | Directly | It is computed from the lift angle |
| Beat error | Slightly | The two beats are compared to each other |
| Everything published without the angle | Completely | There is no way to correct it later |
Why this is the bench's fourth cell
Every page on this site carries a four-cell card, and the fourth cell is the lift angle. It is not a reading; it is the disclosure that makes the other three usable. A rate, an amplitude and a beat error with no angle are three numbers about an unknown watch.
The bench holds that a published amplitude without its lift angle should be treated as a claim rather than a measurement, and that asking for the angle is the cheapest way to test whether the person quoting it knows what they measured.
What to do about it
- Set the lift angle before the first reading, and write it next to every number.
- Do not trust a default. Check it on the machine you are using.
- When comparing two readings, compare the angles first and the amplitudes second.
References
- Ref 1Insight: Regulating a Mechanical Watch Movement, SJX Watchessupports the regulation and timing context the disagreement arose in
- Ref 2Brouillage, Horopedia – Watchmaking Encyclopediasupports the measurement-interference concept behind two machines disagreeing
- Ref 3Jewel: Setting Ruby Bearings in Watch Movements, Fondation de la Haute Horlogeriesupports the jewelled bearings that the impulse and the angle depend on
One link, one document, titled as it titles itself. No home pages, no search results.
Set this page down Set this page down when you know your machine's default and whether it is right.