Thursday, 6 March 2014

How come my 30 hour watch runs for > 40 hours? Or: Management of Isochronism in watches.

Waltham 1899-Vanguard 23J
click on the images for a larger view
This post was inspired by a query from a customer who was surprised that his newly acquired 23J Waltham 1899-Vanguard featured in my post  anatomy of 1899 vanguard was still running 14 hours after he received it after c24 hours in transit.

Firstly what is isochronism in a watch? In brief it is the ability of a watch to run at the same rate as the mainspring unwinds and that is critical for the accurate running of a watch and is a required "adjustment" in a railroad grade watch. So what has this to do with how long a watch runs?

Most pocket watches are designed to run for 30 hours (some are designed to run for 8 days but I am ignoring those for this discussion), to achieve this the mainspring will have to be of sufficient strength to operate the escapement and to have sufficient length and space in the spring barrel for 30 hours operation.

There is then a major problems to overcome, a mainspring of consistent thickness will provide more power when fully wound than when it is almost run down and although careful design of the escapement can make it less susceptible to varying power input it is unlikely to be completely immune so will run fast when fully wound, slower when partially wound and then faster and less reliably when almost wound down due to a weak (short) balance action.

Historically there have been a number of design features built into a watch to overcome these problems including:
  • An English Fusee movement
    showing the chain drive from the
     Fusee (right) with its spiral path for
    the chain, to the mainspring
    barrel (left) 
    The Fusee chain drive which gave a continuously variable gearing and had a high gear at full wind  reducing linearly to a low gear as it ran down. This meant that the torque delivered to the gear train was more consistent. It also reduced the minimum and maximum power delivered by the spring in the same way as...
  • The Geneva stop gear used by the English and Swiss in the late 19th and early 20th centuries but like the Fusee not adopted by the American industry. This limited the winding of the spring from a pre-tensioned state to four complete turns of the mainspring arbore (not the winding gear). This meant that the mainspring was never at maximum or minimum tension and delivered power in a more linier fashion over a narrower range. But it also limited the run time of a watch, particularly one not in tip-top condition.
  • Introduction of more jewelling to reduce friction requiring less mainspring power (see below) and more accurate fitting of the components to reduce friction and other inefficiencies.
  • In later years, as metallurgy and manufacturing processes improved, mainsprings were made of varying thickness through their length to provide more consistent power as they wound down, this technology was not however effective around the turn of the century when the Waltham 1899 & 1908 models were designed.
So why did the Vanguard run for so long?
Waltham 1899-Vanguard 23J
mainspring & barrel assembly

Highly jewelled variants of a movement can run for long periods because of the low friction enabled by so many jewels and in particular the jewelled spring barrel, this means that a weaker spring than the lower jewelled variant is often used (apart from other reasons the movement may otherwise run too hard and break the impulse jewel). A weaker spring is thinner so can be longer and still fit into the usually optimum outer 33% of the diameter of the barrel before winding. So it runs for longer but time keeping will be affected as it runs down.
But as watches are regulated for 24 hours (with a 30 hours maximum) this is one of the ways Isochronism is managed, if the watch runs for say 40 hours then the pressure exerted by the spring will be more consistent over the first 24 hours than if it only ran for 30. 
So a 23 jewel Waltham Vanguard and similar movements in good condition, with a good escapement and a weak main-spring (perhaps 0.15mm thick in the case in point) of the correct length (25.5") will run for considerably longer than a 7 jewel Traveler with its need for more power to overcome friction and a thicker (0.19mm) and shorter (21.3") spring. And because of the extended run time, the Vanguard over the first 24 hours of running, will not suffer the same reduction of power in the later stages than the Taveler and, even before other design factors come into force, will deliver more constant performance.
Length matters!

 

No comments:

Post a Comment