For any of those who handle watch repairs…what do you consider an
acceptable timing accuracy you expect from your watchmaker?
Particularly for old mechanical or automatics?
1 min/day? 2? 3?
I do not repair watches, as a long time fine watch aficionado I can
share some thoughts.
The watch accuracy not necessarily depends on a watchmaker but on a
watch itself. This of course should be taken with provision that
watchmaker is properly trained and followed all required steps in
watch maintenance. If the watch is certified chronometer, then it
must conform to the COSC standard. Here is the link for more info
http://www.cosc.ch/chronometre.php?lang=en
If not than it is really difficult to say whether watchmaker did
something wrong or simply the mechanism start showing it age.
Leonid Surpin
Hi Neil,
For any of those who handle watch repairs...what do you consider an acceptable timing accuracy you expect from your watchmaker? Particularly for old mechanical or automatics?
What do you mean by ‘accuracy’? Do you mean regularity of rate in
positions, regularity in wear, ability to show the same time as
another timepiece, ability to compensate for temperature changes?
The short answer is that it depends on the type of watch, its quality
and age… When I restore an 18th century watch I am normally happy if
I get better than 5 minutes per day on timekeeping against a radio
controlled master clock and ± 2 minutes between different static
positions. For a modern watch I would expect to better 30 seconds a
day on rate even on a fairly mediocre watch with maybe ± 10 seconds
maximum difference between positions. However, the age of the watch
and its escapement will have some bearing on the accuracy you can
expect. These rates are also usually only measured in a static
condition, while the watch is standing on a bench, hanging on a hook,
laying on a timing machine etc. and do not reflect the timekeeping in
everyday use where movement of the wrist and knocks and bangs may
affect the timekeeping. Another factor which could come into play is
the ‘settling in’ of the movement again after servicing. When the
lubricant is changed and the mechanical parts are disturbed by
disassembly it takes a little while for the parts to ‘bed in’ again
and so the time keeping can change a little over a matter of days or
even weeks. For this reason it is normal to make the watch run a
little fast to compensate for the loss of initial ‘life’ in the oils
- not such a problem as it used to be with the older lubricants but
still a factor to consider. It is also considered better if a watch
runs fast than if it runs slow - better to be early for an
appointment than late!! Temperature can have a significant effect on
timekeeping, especially in less expensive watches. Ideally, a wrist
watch should be timed at body heat but this is never done. So, also,
lifestyle can have an effect - if a watch is worn regularly it is
more likely to keep time than one that is only used occasionally. The
worst thing one can do with a wrist watch is to take it off at night
and place it on a bedside table - not only does the change in
temperature from 98 on the wrist to 60 on the bedside table change
the accuracy of the timekeeping but also, as the watch cools, the air
inside it contracts and sucks any dust from the tabletop into the
case. This dust then mixes with the oil and grease in the movement
and froms a very effective grinding paste. Every watchmaker is very
familiar with the mass of black gunk which collects on the winding
work close to the winding hole in the case. If you must take your
watch off at night, place it in a ziplock plastic bag - it will
significantly lengthen the life of the watch…
A quartz watch should have a daily error of no more than 10 seconds
but a rate of 2 or 3 seconds a day error shold be easily achievable
depending again on the quality of the watch and whether it has any
means of adjustment available or not - many cheap watches are
non-adjustable…
Best wishes,
Ian
Ian W. Wright
Restorer of fine and complicated watches and chronometers.
Sheffield UK