Horological Meandering

nickd
1591

• Show profile
• Show other posts

External influences

I think it's partly down to the magnitude of the influence of external forces on the system.  

Compare the watch in question with an uber-precision clock in a clock vault:

The clock is designed to run in an environment where as many external influences as possible are removed, nullified or taken into account in the design.  Previous experience, experiment and research mean that the vast majority of systemic factors are known and quantifiable.  What are left are new systemic factors (eg inherent in a new design) and external factors.   The watch is running in an environment where there are many external factors (shock, vibration, temperature, pressure, position...).  The components and forces involved in the clock are relatively massive compared to the external forces acting on them (vibrations in a clock vault are small compared to a pendulum weighing several kilos, but the jolts from playing tennis or mountaining biking are enormous compared to the inertia of the balance or the locking forces).  So, assuming the design is good,  the clock has to cope with a very small number of low-magnitude external influences. But the watch has to cope with a large number of high-magnitude external factors.

The operation of the clock always takes place in the same orientation, so any unlocking, for example, is always operating against the same forces acting in the same direction.  This means that the actions can be designed to operate with the lowest forces and smallest travel consistent with good operation.  The operation of the watch can take place in any orientation, so the design has to take into account a huge variation in the forces and direction of operation.  Given the very small components, the manufacturing tolerances are relatively large compared to the component sizes and acting distances, which adds another variable.

Because the clock is inherently stable compared to the magnitude of the random or non-random external forces it is easy to detect the error and trace it back to its root cause.  In a watch that is worn normally, there are an enormous number of high-magnitude forces involved and it's difficult to trace any given effect back to its root cause (is it fast because of the jolting or the cold or whatever?).

So, the watch that's running +/-0s/d on the wrist has found an equilibrium where the adjustments/regulation balance out the average of the external influences.  On the winder, these factors aren't present, so it goes to hell in a bucket (or vice versa).

I think the tourbillon discussion has become futile as it doesn't take into account the major impact of the magnitude and variability of the systemic errors in the old pocketwatches Breguet targeted, and similarly doesn't take into account the very different situation in a modern wristwatch.

nick