Here's my
initial cut at a mechanical watch FAQ. As you can see, it's largely
a pointer to many of the fine articles contributed to the forum by the
community over the years. That being said, if there are any errors
on this page itself, they are the fault of the compiler (i.e. me) and not
the fault of the other contributors.
Legalese: This FAQ is intended to be informative,
and is not intended to supersede any manufacturer's instructions for service
or operation of watches. No warranty is expressed or implied.
Really, I wouldn't kid you about this.
Please send additions / corrections to ehahn@alum.mit.edu.
Questions answered in this FAQ
Part 0: What is in this FAQ?
Part I: Watch / Movement basics
1.1 What is a mechanical watch?
1.1.1 What is the difference between
a movement, ebauche, and caliber?
1.1.2 What is a "hack" seconds
feature?
1.1.3 What does "17 jewels" mean?
1.1.4 Why do they use synthetic
ruby?
1.1.5 Are more jewels better?
1.1.6 What is shock protection?
1.1.7 What is the "T", "T<25",
and the lower-case Greek sigma on my dial mean?
1.2 What is a MecaQuartz?
1.3 What is an Accutron?
1.4 What do I need to do to keep
a mechanical watch running for a lifetime?
1.5 Why should I get a mechanical
watch when a quartz watch is so much cheaper and more accurate?
2.1 What's the difference between
a "manual" and an "automatic"?
2.2 How does an automatic mechanism
work?
2.2.1 Are the Seiko Kinetic /
Autoquartz therefore automatics?
2.3 What is a watch winder, and
do I need one?
Part II: Brands & Accuracy
3.1. What is the best watch made?
Is it [insert brand here]?
3.2 How accurate can I expect
an [insert brand here] to be?
3.2.1 What does "adjusted" vs.
"unadjusted" vs. "regulated" mean?
Part III: Features
4.1 What is a chronometer?
What is a chronograph?
4.2 What is a Rattrapante?
4.3 What is a Flyback?
4.4 What is a Column-Wheel?
4.4.1 Are Column-Wheel chronographs
better than other types?
4.5 What was the first automatic
chronograph movement?
5.1 What is a complication?
5.2 What's a Reserve de Marche?
5.3 What's a Perpetual Calendar?
5.4 What's a Tourbillon?
5.5 What's a Repeater?
Part IIII: Materials
6.1 Should I be concerned about
radium on a vintage watch dial? How about Tritium?
7.1 What's the difference between
acrylic, mineral, and sapphire crystals?
7.2 How do I remove the scratches
from an acrylic crystal?
8.1 If a watch is advertised as
"18K", what does that mean?
8.2 What do PVD, CVD, or PE-CVD
mean?
8.3 What does Gold-Filled mean?
9.1 What are some other materials
used as watch cases?
9.2 How do I remove scratches
from my watch?
10.1 This FAQ sucked! Where
can I get better answers?
Part 0: What is in this FAQ?
This FAQ is intended to cover the basics of watches
and movements themselves.
It is not a buyer's guide; nor will it tell you where
to get the best deal on a Rado. Furthermore, it does not talk about
many other important issues, such as how the watch industry works, or even
how the TimeZone Forum works. There is a short discussion of watch
brands and accuracy (mainly to dispell the belief that the two are strongly
linked), but otherwise, it does not concentrate on any particular brand
(other than as examples).
While it is based on the contributions of many sources,
some of which are not verified, all errors on this particular page are
mine, and I will correct them if you point them out. I also have
tried to add a bit of spice and atmosphere to the discussion; please let
me know if this is a faux pas.
I plan to upgrade the FAQ from time to time with
illustrations, where appropriate, to better communicate some of the ideas
in here.
PART I: Watch / Movement Basics
1.1 What is a mechanical watch?
A mechanical watch is a device for keeping time, which
uses the energy from a wound spring, and keeps time through the highly
regulated release of that energy through a set of gears (the wheel train)
and an escapement. It differs from the typical quartz watch
in that it uses purely mechanical components to keep time. Mechanical watches
typically can run for about 40 hours on one full winding of the mainspring,
with a few designs available with up to 8 days, or even 10 days, of power
reserve.
A more in-depth technical explanation, with photographs,
of how a watch works can be found in the Horologium article
"The Anchor Escapement". Also, it may
be useful to peruse a diagram
of watch parts residing in the TimeZone archives to familiarize yourself
with the lingo. Finally, the Illustrated
Glossary of Watch Parts is the definitive guide to see schematics of
parts in their natural habitat. (Like animals, they are sometimes
easier or more difficult to see in the wild.)
The basic design of mechanical watches has not changed
very much in the past fifty years. What has changed is the use of
high technology and modern materials in the design and manufacture of watches.
Even with the fusion of CAD/CAM, electrospark erosion in the manufacturing,
and titanium nitride cases; the pinnacle of watchmaking is still an expression
of elegance of design, attention-to-detail in finishing and assembly,
and the art of hand-tweaking movements for optimum performance.
A mechanical watch is an anachronism, it is the ultimate
refinement of "low" technology; collectively they are an obsession shared
by the enthusiasts on TimeZone.
1.1.1 What is the difference between
a movement, an ebauche, and a caliber?
A movement is the completed, finished individual
mechanism contained inside the case of the watch, not including the case
or dial itself, which is responsible for keeping time. An ebauche
is typically understood to mean a "raw" or unassembled, unfinished movement,
including the major structural components (plates, bridges) and sometimes
parts of the wheel train and other moving parts. A caliber
is the collective name given to a series of movements of the same design.
Many watch companies will purchase complete movements
from a major supplier such as ETA or Lemania, engrave their company's name
and other information onto them, and encase it with their own or even contractor-supplied
cases. This practice can be up-front - where the company acknowledges
that the movements are not of their own design or manufacture, or it can
be hidden - where the watch company claims to use "in-house" movements
when the movements are in no way designed or manufactured internally.
Some watch companies will purchase an ebauche from
a major supplier, polish and decorate the parts (i.e. finish the parts),
and assemble it with standard parts to create a higher quality-controlled
movement than the stock ready-made movement.
Other companies purchase ebauches, finish them to
a high standard, modify parts of the movement, and add custom components
like an upgraded escapement assembly - to create what might be called a
custom version of that movement, much like how Carroll Shelby, AMG, or
BMW's "M" division re-engineer existing automobiles to produce something
with higher performance and exclusiveness. Many times, the company
will rename the caliber as its own to reflect the modifications and finishing
of the movement vs. an unmodified stock movement.
There is often a debate on whether a particular company
is being deceptive in renaming a movement based on an outside supplier
as an in-house caliber. This debate is summed up quite nicely in
the article "When
a [Valjoux] 7750 ain't a 7750 any longer" by Time Flies and a host
of regulars.
It should also be noted that the largest supplier
of ebauches in Switzerland, ETA, can provide a wide variety of finishes
on its products, from very raw parts to fully finished movements complete
with Geneva stripes and other decoration. They even have a subsidiary,
Soprod, that can perform custom finishing and even modifications to the
basic movement. Therefore, there is by no means a single level of
quality that one can ascribe to an ETA movement - there are basic versions
all the way up to fully finished ones.
Finally, on TZ it is common to see the word ebauche
used to refer to any third party movement, even if it is completely assembled.
1.1.2 What is a "hack" seconds feature?
This, AFAIK, is a military term referring to watches
that stop the second hand, to allow for more accurate synchronization
between two watches. In the most common type of hacking watch, when
the crown is pulled out to the time-setting position, a lever is moved
which contacts the rim of the balance, thus causing the movement to stop.
John Davis has seen other methods used to stop the movement as well, including
brakes on the third or fourth wheel.
A. Lange & Sohne have produced a watch that stops the balance when the crown is pulled out and automatically
moves the second hand to the "0" position - to help facilitate setting
against an accurate time reference.
1.1.3 What does "17 jewels" mean?
Higher grade watches have traditionally used a jeweled
movements, which means that jewels (originally natural ruby, now synthetic
ruby) were actually used in the movement. These jewels are functional -
they are used as the bearings for the wheel trains and in high wear parts
such as the escape lever and impulse jewel.
A lower-end movement from before 1970 would typically
use 5 or 7 jewels; this end of the market has pretty much been taken over
by quartz. Nowadays, most manual wind watches will have a standard
complement of 17 jewels, which are:
-
1: Impulse jewel (the part of the
balance wheel assembly which receives a kick from the escape lever)
-
2-5: Balance staff pivot bearings
(two pairs - in combinations of one pivot jewel (i.e. jewel with a hole
to receive the axle (pivot) of the wheel) and one cap jewel (i.e. jewel
without a hole outboard of the pivot jewel, to prevent excessive movement
of the balance staff), usually shock protected)
-
6-7: Escape lever pallets (one pair)
-
8-9: Escape lever pivot bearings (one
pair)
-
10-11: Escape wheel pivot bearings
(one pair)
-
12-13: Fourth wheel pivot bearings
(one pair)
-
14-15: Third wheel pivot bearings
(one pair)
-
16-17: Center wheel pivot bearings
(one pair)
Note that automatic winding movements, and movements
with additional functions such as chronographs and calendars, can up the
total number of jewels tremendously. For example, the IWC Il Destriero
Scafusia (claimed to be among the most complicated wristwatches manufactured)
has a total of 76 jewels to accommodate the time, perpetual calendar, rattrapante
chronograph, repeater, and tourbillon functions - and this is a manual
wind watch.
1.1.4 Why do they use synthetic ruby?
Ruby is technically known corundum, and is a crystallized
form of aluminum oxide (Al2O3). In pure form, corundum is white in
color; trace impurities are added to change the color - to red in the case
of rubies. It should also be noted that any other color of corundum
(including clear) is known as sapphire. Ruby is used because it is
an extremely hard and provides a slick surface for the wheel pivots (and
other steel components) to operate on. In a mechanical watch, there
is a constant force applied to the pivot of every wheel in the wheel train,
which is applied by the wound-up mainspring.. (see my
article in TZ Classics on the coefficients of friction for various
materials, if you're really interested...)
Without any jewels, the steel wheel pivots would
very quickly grind away the bridge and plate material until the wheels
came out of alignment, and the movement would crash to a halt. In
the inexpensive watch of yesteryear, the pivot holes may have been provided
with hardened metal bushings.
Ruby is significantly better than steel in handling
the forces involved, wearing long, and providing a nice low friction surface
suitable for both high-load as well as high-speed motion. With modern
production methods, they are cheap (~$0.02 each). And they look nice.
1.1.5 Are more jewels better?
Not necessarily. As noted above, a typical hand-wind
movement today will have only 17 jewels as a full complement. Some really
high-grade or ultra-thin movements will add a few extra jewels to further
protect against any wear, but even these top out at 21-23 jewels.
Only those pieces of the movement which are between
the mainspring and the escape wheel are candidates for jeweling, as these
are the movement parts that experience the large forces or relatively
high speeds of the mainspring or escapement. Other components, such as
the motion works (i.e. hour and minute wheels), calendar mechanisms, and
winding train are not under this constant stress, and thus arguably do
not need jewels.
Automatic winding movements will add about 4-8 jewels
to help most efficiently transfer the relatively small rotor forces into
winding the mainspring. Another factor has to do with how the watch
is constructed - especially for chronograph movements and perpetual calendars.
Some chronograph movements used today (including the ETA 2894-2) are modular
in construction - meaning that a plate containing the chronograph works
is grafted onto a basic timekeeping movement. Since the original timekeeping
movements were not always designed with this in mind, it becomes critical
for the add-on module to add as little "drag" as possible - which may indicate
use of jewels for their low friction properties.
BTW, one will occasionally encounter a quartz movement
with jewels in it - they technically aren't really necessary because a
quartz wheel train is not constantly under stress. There is a discussion
of jeweled quartz movements in the archives.
As a historical note, there was a "jewel
craze" about 50 years ago, where manufacturers, under the belief that the
public thought more was always better, came up with 75 or even 100 jewel
movements. Most of these jewels were not functional in any way, and the
results looked ludicrous to an informed eye.
1.1.6 What is Shock Protection?
As is fairly obvious, a mechanical watch is made up
of numerous tiny parts, many of which are in constant motion. It would
not do for an accidental bump to interfere with the watch's ability to
keep time due to damaging of the balance pivots. So, watchmakers
include "shock-protection" in their watches in the form of a tiny spring that
holds the balance staff jewels in place, instead of being rigidly held.
This gives a slight amount of give - not enough to disrupt the operation
of the watch for more than a moment, but enough to prevent the balance
pivots or cap jewels from damage.
Shock protection is usually only applied to the balance
because the high speeds and regular motion they are designed for - this
kind of design goal leads one to small, extremely hard pivots, with most
of the weight concentrated at the rim of the balance. These factors combine
to make for a lot of broken pivots.
Richard Paige has a short article on the details
of how
a shock proof system works.
1.1.7 What do the "T", "T<25",
and the lower-case Greek sigma on my watch dial mean?
These are optional industry markings, found next to
the inscription "Swiss Made" on some watches, which signify what the markers
on the dial are made from.
The "T" means that tritium (a low-level radioactive
substance) was applied to make the hands and/or markers glow in the dark.
The "T<25" means the same thing, except it spells out that less
than 25 milliCuries of radioactive material is used. See section
6.1 for more information about Tritium.
The lower-case Greek sigma means that the markers
are made of solid gold.
1.2 What is a MecaQuartz?
A movement described as a "mecaquartz" is actually a
quartz movement that contains many mechanical components. An example
is the Jaeger LeCoultre (JLC) Caliber 631, which uses a quartz movement
to drive not only an analog hour/minute/second display, but also mechanically
drives a chronograph function (see 4.1, below). This differs from
a typical quartz chronograph, where the chronograph functions are either
digitally displayed in an LCD window (e.g. Breitling Aerospace), or where
the chronograph hands are individually driven by separate motors (e.g.
Seiko Flight Computer, with four separate motors).
A "mecaquartz" movement as that term is commonly used is not
a quartz movement with a mechanical charging system, such as the Seiko
Kinetic or Swatch Autoquartz (see 2.2.1, below).
1.3 What is an Accutron?
The Accutron is an electrically driven movement developed
in the early 1960's by the Bulova corporation, and was a precursor to the
quartz revolution of the late 60's. Instead of having a mechanical
balance wheel, the Accutron used a mechanically resonating system very much like
a tuning fork to keep a constant vibration rate, and thus time regulation,
for the rest of the mechanical movement.
Accutrons have a distinctive audible hum when operating, and their second hand is driven at such a high frequency
that it truly appears to move continuously, unlike a mechanical watch (which
vibrates at 10 Hz or less) or a modern quartz watch. Accutrons were the
first major advance using electronics in timekeeping technology over mechanical
watches.
Here is an Accutron
homepage, which has a lot of good information about history and theory
of operation. The Horologium contains great dissection of the Bulova
Caliber 214 Accutron movement.
1.4 What do I need to do to keep a
mechanical watch running for a lifetime?
Within reason, a mechanical watch can always be brought
back into good time keeping, and a jeweled movement can last for generations.
However, it is important to periodically service
a watch to ensure that the components are well-lubricated, and that the
mechanism is free from dust, dirt, and moisture. Any water
that gets inside a mechanical watch will wreak havoc with the precision mechanism
inside, especially the anchor escapement and escape wheel which are typically
made of steel.
The typical rule of thumb is to have the water resistance
(i.e., the integrity of the seals in the crown, bezel, and caseback) of
a watch checked every year or so, especially if used for sports or diving.
With the development of modern synthetic lubricants, most manufacturers recommend
a servicing every four or five years.
"How
a Watch is Tested For Water Resistance" by Richard Paige and James
Dowling, and "Cost
of Servicing a Watch" by Walt Odets give much more detail about how
these maintenance steps are carried out, and why servicing, which may appear
to be costly, is in fact a very involved process when done correctly and
worth the investment.
Finally, since the crown is often the only means
by which the watch owner can adjust/abuse the movement inside the case,
Walt Odets has provided Some
Basics In Handling the Crown.
1.5 Why should I get a mechanical
watch when a quartz watch is so much cheaper and more accurate?
Yes, Virginia, a quartz watch is cheaper and more accurate
than a mechanical watch. A good mechanical watch can typically be
made no more accurate than 2-3 seconds per day. Your typical
inexpensive quartz is usually good to 0.5 seconds per day or better.
For an in-depth analysis, Walt Arnstein has written a technical
analysis of quartz vs. mechanical watches.
But mechanical watches are not about achieving the
ultimate in accuracy. Craftsmanship, aesthetics, and tradition are
all part of the allure. Because the wheel train of an analog quartz
watch is not under constant stress from a wound mainspring, it does not
need to be as finely finished, nor does it require painstaking skill and
precision in assembly.
Mechanical watches are good enough for most
people's everyday lives, and they call to our emotional side.
Rather than continue, I'll let Mycroft explain; he
is much more eloquent. "On
Companionship and Soul in Watches"
2.1 What's the difference between
a "manual" and an "automatic"?
An "automatic" wristwatch is a mechanical wristwatch
with a self-winding mechanism. In other words, one does not have
to wind the crown periodically to keep the watch running. A "manual"
or "manual wind" watch must be wound by hand, using the crown, usually
every day, to operate continuously.
If one were going to own only a single watch, and wear
it every day, an automatic would be a good choice, since the watch will
be worn consistently enough to stay wound - the owner would never need
to manually wind the watch, and would only need to adjust the time
to compensate for drift and at changeover to daylight/summer time and back.
(In fact, several early automatic movements dispensed with the crown and
moved the time-setting mechanism onto the back, under the theory that the
mechanism would only be accessed infrequently. This turned out to
be a marketing flop - people liked the look and easy accessibility of the
crown.)
For this reason, most commonly seen watches with
more than a simple date window use automatic movements - this includes
"triple date" calendars, annual calendars, perpetual calendars, and any
of these combined with moonphases (see 5.3 for more information about calendars).
With few exceptions (oddly enough, these seem to be more expensive watches),
most manual wind watches have simpler calendars, although they may include
other complications like chronographs (see section 4.1).
One caveat about automatics - if you have more than
one watch that is worn regularly, the automatic winding advantage is lessened
- the automatic may stop if not worn often enough. With some calendar
mechanisms, this is can be an increased inconvenience when the watch is
reset.
Finally, since frequently worn automatics are usually
at or near a full state of wind most of the time, one may get the impression
that they can be adjusted to be more accurate and consistent over the course
of many days. This, in fact, is not necessarily the case, as a manual-wind
watch that is wound consistently once per day can be tweaked so that the
day to day variation is very small. In short, there is no definite
performance advantage to an automatic - it is mostly a convenience.
Justin Time and Walt Arnstein have supplied some
general advice about how
to start an automatic movement when fully unwound.
2.2 How does an automatic mechanism
work?
All self-winding watches work on the principle of converting
arm motion (kinetic energy) into the winding of the mainspring (potential
energy). Usually, this is performed by a half-disc of metal weighted
at the edge called a rotor, which spins when the wearer's arm is accelerated
unpredictably (that is, when moved normally in the course of everyday
life). This rotary motion is then geared down to wind the central
arbor of the mainspring. Walt Arnstein has written an in-depth article
about the Physics
of the Automatic Watch's Winding System.
All automatic watches have an overwind protection
mechanism of one sort or another, to prevent breaking the mainspring once
fully wound. In a typical system, the mainspring, which is wound
at the central arbor of the barrel, is not rigidly attached to the outside
of the barrel. Instead, there are a series of detents along the outer
edge of the barrel that allow a stiffly constructed mainspring part called
the bridle to slide along when an attempt to overwind is made. On
some watches, a faint click can be heard when this happens, on others,
it cannot be heard. It should be noted that this overwind protection
is critical to avoid damage to the watch, and is reported to be one of
the more tricky things to get right during a watch service because of the
special lubricant needed to ensure proper operation.
The Horologium contains several articles about different
types of automatic winding systems, including a vintage
JLC/Vacheron automatic, a modern production JLC as in used in the International
Watch Company (IWC) Portugieser Automatic (scroll to the bottom of
the article), a modern ETA/Eterna system as used in an $85
Swatch Automatic, and an innovative bi-directional system from the
IWC Caliber
8541 (now used in the new IWC Cal. 5000).
There is also a series of forum posts by several
authors dealing with micro-rotor
watches, as illustrated with photographs in this separate
article by Hans Zbinden.
2.2.1 Are the Seiko Kinetic / Autoquartz
therefore Automatics?
Not quite, but close. The Seiko Kinetic and ETA Autoquartz
movements are quartz movements. However, they use a rotor system
similar to those used by automatics. The difference is that the rotor's
motion is converted to electricity, which is then used to charge a capacitor.
The quartz movement then draws current from the capacitor as if it were
a battery.
It should be noted that the latest generation of
autoquartz movements can store enough power to run the watch for several
months (or even years in the case of the Seiko Auto-Relay); a mechanical
automatic can only store as much power as contained in the mainspring -
which is usually only 40 hours or so for most automatics.
2.3 What is a watch winder, and do
I need one?
As noted in section 2.1, collectors who have more than
one automatic watch may have "difficulty" keeping any one watch going continuously.
This leads to increased inconvenience if calendars and moonphases must
be reset. A solution has been invented - the automatic watch winder.
The idea is quite simple: strap the automatic
watch to a motor, which then moves the watch enough to keep it wound when
not worn on the wrist. That way, one can choose to wear any watch
at any time, and not have to reset the time or calendars. In theory,
this device should be simple and cheap.
In practice, it's not as easy as it sounds.
First, while an automatic watch has an overwind protection mechanism to
avoid damage to the movement, if the overwind mechanism is constantly used
for hours at a time (i.e. though constant turning, rather than the unpredictable
movement of the human wrist), the lubrication of the mechanism is worn
out faster. This means that watch winders must be designed to only
make so many turns per day, and to let the watch sit. Furthermore,
since automatic winding mechanisms vary from watch model to model, the
winder must be designed to have an adjustable number and direction of turns per day.
Finally, mechanical watches are considered luxury items, which means watch
winders are doubly so - due to the small number of people who would actually
want one.
All of this means that manufacturers for the consumer
market must not only make their winders mechanically foolproof to avoid
damaging watches, but they often construct the winder comparably to a jewelry
box. The upshot of all this is that most consumer watch winders
with programmable winding are expensive - often costing several hundred
or thousand dollars.
While cheaper professional models exist, these often
do not come with programmable settings, as the maker assumes that they
would be used in a watchmaker's shop.
Last word - winders are not a necessity, they are
a convenience. You should decide whether they are worth it based
on whether you feel inconvenienced resetting your automatics occasionally,
and whether they are worth it to you as a luxury item.
In any case, Jack Freedman (who sells winders) has
written a well-balanced article on whether
we need watch winders.
PART II: Brands & Accuracy
3.1. Who makes the best watch?
Is it [insert brand here]?
In my opinion, the answer to these questions depend
almost exclusively on what is important to the person asking them.
Consider the following qualities that a watch can have, any of which can
be the most important factor in some people's minds:
-
Accuracy - how well does the watch keep time?
-
Features - can the watch act as a analog or
digital calculator? a calendar which never needs setting? keep important
phone numbers? can be used when diving to hundreds of meters?
double in a pinch as an altimeter, depth gauge, navigation system, and
emergency locator? can it be used as a stopwatch? - the variations
these days are endless!
-
Status - will the other company's negotiator
be unconsciously swayed by the fact that I have a recognizable status symbol
on my wrist? (Note - I believe that this is a legitimate question
for some people!)
-
Ruggedness/Dependability - can I wear this
watch in combat? can I wear it near strong magnetic fields?
will it need a battery replacement at an unacceptable moment?
-
Aesthetics - is this watch a work of art?
does it fit with my personal sense of taste?
-
Craftsmanship - what do the dial, case, and
movement say about the skill and care of the watchmakers who made it?
-
Cost - how cheap can I get it for?
...the list goes on. Ask any number of people,
and they will not only differ on the relative importance of these categories,
but they will also differ on whether a particular watch is satisfactory
in each category!
My own personal opinion - figure out what is important
to you in a watch, and find out whether the watch you're looking for has
it. Don't ask people to make a sweeping judgment about the overall
worthiness of a watch - you'll get way too many conflicting answers.
3.2 How accurate can I expect a [insert
brand here] to be?
This is truly a Frequently Asked Question, in that people
will post that they have just bought a new IWC/Rolex/JLC/Revue Thommen/Omega/etc.,
and it gains/loses X seconds per day - they then ask whether this acceptable
for this brand.
After reading several articles on this subject, I've
come to the following conclusions:
-
Just about any current production watch is capable
of running overall within 1-2 seconds/day - this includes relatively
inexpensive brands like Hamilton and ORIS, as well as high-end brands like
Patek, JLC, and Lange.
-
The stability of a given rate can be quite ephemeral, with the instantaneous
rate even in a single position varying constantly. This
means that the rates for a given watch are constantly drifting - although
for higher grade movements the amount of drift tends to be much smaller.
-
The question of accuracy is more subtle than it appears
on the surface. The reason is that, due to the limits of how
small components can be, forces will act on the components differently
when worn on the wrist vs. when resting in any particular position.
In other words, even a watch which has a daily error rate of 4-5 seconds/day
in any single position can be made to show zero overall drift
though the combination of wearing the watch and letting it sit overnight
in a particular position to compensate for drift obtained during wear.
-
A higher quality watch may or may not be more accurate
than a lower quality watch at any given instant in time, but it has the
potential to be more accurate.
The reason is that a higher quality watch has been adjusted
to minimize the variation in error in several positions, and
has less drift in the rate over time.
-
It is a relatively simple operation for a watchmaker
to adjust a watch that runs consistentlyfast or slow to near zero error. This operation,
which is called regulation - merely increases or decreases the overall
rate without compensating for rates in positions.
Also, the conventional wisdom is that a typical new
watch needs to be run-in - in other words, while sitting unused in the
jeweler's shop, lubricants pool in certain locations. It takes a
couple of months for the lubricants to be properly redistributed, and for
the motion to wear away some of the microscopic imperfections that all
parts have. Bottom line - let a new watch run for about 2 months
before taking it in to be re-regulated.
An in-depth discussion of the above concepts are
contained in a discussion on the "confusing
language of watch "accuracy" by Justin Time.
3.2.1 What does "adjusted" vs. "unadjusted"
vs. "regulated" mean?
These are terms to be used in conjunction with the discussion
in section 3.2.
An "unadjusted" movement is a movement where no attempt
has been made to ensure that the daily error rate in several orientations
(positions) have been minimized across the positions. An "adjusted"
movement, therefore, has had some extra care in ensuring that the variance
in accuracy between several orientations is minimized.
Watches are typically adjusted to 2, 3, 4, 5, and
6 positions. Traditionally, they are ordered as follows:
-
Dial Up
-
Crown Down
-
Dial Down
-
Crown Left
-
Crown Up
-
Crown Right
For example, a watch adjusted to two positions include
positions 1 & 2 from the above list (i.e. face up and crown down).
Similarly, a watch adjusted for 4 positions has been adjusted in positions
1-4, and so on. Watches can also be adjusted for isochronism (i.e.
constant time across varying states of wind) and temperature. Some
manufacturers (Franck Muller, some Patek Phillipe) adjusts to 8 positions
(the above six plus two half-way orientations); some on the forum would
say that this is a bit over-the-top.
There is a 3 part, in-depth technical article on
adjusting, including photos and commentary on the actual adjusting of a
watch, in The Horologium series on "Tweaking
the Mark XII".
A "regulated" movement is a when the overall rate
of the entire movement (either adjusted or not) is brought into correct
absolute timing. Unlike adjustment, this is a simple tweak, which
moves the daily rates of all of the positions up or down without intentionally
changing the relative rates in positions.
PART III: Features
4.1 What is a chronometer? What
is a chronograph?
These two terms are commonly confused among new watch
aficionados. They actually have very little to do with each other.
A chronometer is a watch which has passed a test
given by the Contrôle Officiel Suisse des Chronometres, or COSC.
The COSC is an official Swiss government agency which tests watches to
ensure that they fit within a narrow-but-usually-obtainable window of acceptable
error (i.e., the rate in all positions falls into the range of -4 seconds/day
to +6 seconds/day). While some watch companies tout their products
as having a COSC certificate, it really is not that difficult to pass the
test, and over 95% of the watches submitted pass. Another factor to consider is that the COSC does not
test watches as they are sold in the store, but movements fitted with a
temporary case, dial and hands. In addition, the COSC certificate
cannot say anything about how the movement was handled after testing. For more information on COSC testing
and the meaning of the results reported on the COSC certificate, see Mike Disher's article
"Reading and Understanding a COSC Certificate".
Several watch manufacturers actually put more stringent tests than the
COSC procedures on all of their watch production (the JLC Master Control
1000-Hour series is probably the most well known.) Paul Schliesser
has written an essay on how
COSC testing is performed.
A chronograph is a watch that tells the time of day and also allows the user to time events of short-to-medium durations
(i.e. from a few seconds to a few hours, typically). This is usually
done in a mechanical watch through the central seconds hand, and one or
more subdials (the regular, or continuous seconds, is also located on a
dial). Chronographs are of varying usefulness, and are an interesting
complication to put on a watch, as they often give the watch a sporty image.
In addition to elapsed time, chronographs are often fitted with several
scales designed to measure other things, such as pulse rate or units manufactured
per hour. Mike Margolis has written a short
discourse on chronograph scales.
A typical modern chronograph is operated with two
pushers: one to start and stop the timing, and a second to reset
the hands to zero when the timing is stopped. (In some older chronos
with only one-button for control, the sequence of pushes was start-stop-reset
- there was no provision to continue timing once the mechanism was stopped.)
4.2 What is a Rattrapante?
A "Rattrapante" is a chronograph with an added second
hand, to allow, for example, lap times in a multi-lap event to be read
off without stopping the chronograph. A chronograph is usually started
with one pusher, which starts both second hands moving (one superimposed
over the other). When the operator desires to read an intermediate
time, he/she pushes a second pusher. On the dial, one of the second
hands stops (the "split" hand), while the main second hand continues.
If the second pusher is pressed again, the split seconds hand "catches
up" with the main second hand, and is ready to be used again.
Interestingly, three languages describe this function
in different ways: "Rattrapante" is the French term for "catch up"
(describing the motion of the split hand); the German term for this function
is "Doppelchronograph", or double chronograph; the English term is "Split
Seconds" (describing the appearance of the second hand when the function
is activated).
An example of this function is on the IWC Doppelchronograph
or Breitling Chronoracer (a mecaquartz (!) rattrapante).
4.3 What is a Flyback?
A "Flyback" or "Retour a vol" is a function which allows
the chronograph to be reset to zero without having to stop the chronograph
first (recall that a normal chrono requires one to stop the timing before
resetting). This is useful for aviation, where several segments or
legs of a route are flown for specific periods of time in sequence; the
act of stopping, resetting, and restarting leads to a short delay in the
beginning of timing the next leg. This delay, reportedly, can be
a factor in some precision military operations, for example.
Examples of chronos with flyback include the Blancpain
Flyback Chronograph, and the Breguet Type XX Aeronavale.
4.4 What is a Column-Wheel?
As one can imagine, constructing a chronograph is fraught
with the potential for inadvertently stopping or damaging the entire movement.
For example, if the reset mechanism was somehow activated while the chronograph
was running, large destructive forces would be applied to the entire
movement train, which would at the very least jam the movement, and would
likely destroy several components.
The column wheel was one of the successful early
designs to ensure that none of the above happened. If you can visualize
the top of a castle turret, with tooth-like battlements, this is what a
column wheel looks like. The pushers that control the chronograph
rotate this wheel, and the various parts of the chronograph are controlled
by fingers that fall into and out of the spaces between the teeth.
This ensures that each of the chronograph parts is coordinated properly.
Because producing and finishing
a column wheel is labor-intensive, a simpler, easier method
of producing chronographs was needed to keep this complication from appearing only in very expensive watches.
The most common method is to
use an oscillating pinion to coordinate the chronograph start, stop, and
reset (i.e. a cam shaped device that rotates back and forth as the various
functions are activated.) This allows for reliable operation without
nearly as much fiddling and hand adjustment as a column wheel requires,
and the pinion can be built using pieces of stamped metal.
These days, relatively few column wheel designs are
still being produced - the Zenith El Primero, some Lemania and F. Piguet
movements, and (interestingly) a F. Piguet rattrapante mecaquartz are examples.
The most common movements, such as the Valjoux 7750 and various Lemania
movements (1874, 5100), are not of column wheel design.
Walt Odets dissected a Frederick
Piguet column wheel chronograph and an oscillating pinion Valjoux
7750 in the Horologium.
4.4.1 Are Column-Wheel chronographs
better than other types?
Tough question - much like the question of which watch
is the best.
Certainly column wheels are a traditional method
of coordinating the chronograph components, and tradition counts for a
lot.
On the other hand, the newer, non-column wheel movements
have certainly proved themselves in a variety of demanding situations.
Both the column-wheel and non-column-wheel version of the Omega Speedmaster
were certified for space flight by NASA. And the non-column-wheel
Lemania 5100 has been one of the few mechanical movements to be accepted
as sufficiently rugged by modern military forces.
So - much like the question of which watch is best,
I suggest that one think about which factors are personally most important in
a chronograph, and make the decision from there.
4.5 What was the first automatic chronograph
movement?
The late 60's saw a race among three sets of manufacturers
(two Swiss groups and Seiko in Japan) to create the first automatic winding
chronograph movement. Interestingly enough, they all succeeded within
a few months of each other. This story is documented
by Ignacio.
5.1 What is a complication?
A complication is defined as a wristwatch function beyond
that required for simple hour, minute, and second. More practically,
common additions such as a day/date window are typically excluded from
this definition. This leaves a wide variety of features that can
be either commonly or less frequently found on watches.
Chronographs of all flavors have been dealt with
in Section 4. This section will discuss some of the more commonly
seen (or at least coveted) complications.
5.2 What's a Reserve de Marche?
A Reserve de Marche, or Power Reserve, is an indicator
that displays the approximate number of days or hours left on the current
state of mainspring wind. This typically cannot be done by a simple
gear train, since a watch is wound from the center arbor of the barrel,
and the power is removed from the outer rim of the barrel. Therefore,
a gear train that can act as a differential is required to read out the
difference between the arbor position and the barrel rim position.
This can be a very useful complication, as it lets
one know whether a watch is wound before putting it on.
This is true for automatics (which may be in any state of wind) or long
power reserve watches (such as the Eberhard 8 days, Lange 1, or the new
Patek 5100 ten-day reserve).
5.3 What's a Perpetual Calendar?
How is it different from an Annual Calendar or Triple-Date?
A perpetual calendar is the most developed form of the
simple date window on a typical watch. It keeps track of date, day-of-the-week,
(sometimes weeks), months, year, leap years, and sometimes even centuries.
Because of the relatively complex rules governing the Gregorian calendar, including the
varying lengths of months, and leap years every four years, a typical perpetual calendar has wheels turning from several
times per second (e.g. balance wheel) all the way to once every four years.
Because of the complexity of the Gregorian calendar, some perpetual calendars
will require an experienced watchmaker open the watch to make
an adjustment at AD 2100, or later (assuming that an experienced watchmaker
still exists then).
Some less complex calendars are also available:
-
Semi-perpetual calendars (e.g. the Breitling Montbrilliant
1461), which requires an adjustment on leap year day only.
-
annual calendars (of which the Patek Philippe 5035 is
an outstanding example), which only require a user adjustment once every
February
-
"triple date" calendars, which contain month, day, and
date - but need to be manually advanced at the end of each (short) month
Some would say that the inconvenience in resetting the
date on a true perpetual calendar is the main reason for the existence
of the watch winder industry. :-)
Walt Odets has analyzed the calendar mechanisms in
several
models of the IWC line, and the Patek
Philippe 5035.
5.4 What's a Tourbillon?
A tourbillon is a holdover from the days of the pocket
watch.
As discussed in Section 3.2 on adjusting a watch
for positions, in even the best watches there are small
variations between the different vertical orientations (i.e.
in the crown up, down, left, or right positions). This is largely
due to the combined effect of gravity, the hairspring shape and attachment
point on the balance staff and cock, and the regulator pins.
Unlike a wristwatch, a pocketwatch worn in a vest
will spend the majority of its time in a vertical position. Therefore
18th/19th century watchmaker Abraham Louis Breguet allegedly decided that,
for the absolute best accuracy, some means of balancing out the effects
of gravity in the various orientations was needed. The solution he
devised placed the balance wheel, escape lever, and escape wheel in a cage,
which then rotated as a unit within the movement as a result of the normal
escapement process. In this way, the overall effects of gravity
get balanced out, as the escapement of the movement never spends any significant
time in one vertical position.
However, this solution is very complex, with the
result that a tourbillon has become more a statement about the watchmaker's
skill rather than having any real performance advantage - starting price
on a tourbillon is roughly in the $50,000 range. The performance
advantage is even further nullified by the fact that wristwatches spend
a less predictable amount of time in less predictable positions.
(Fun fact - Lemania makes several tourbillon movements
which are supplied to a number of high-zoot companies, so these companies
can sell tourbillons without having to spend the development time and money.
This list of companies, naturally, is very tightly held.)
5.5 What's a Repeater? How is it different
from a Sonnerie?
A repeater, or "Repetition Minutes" is a variety of
chiming watch. Unlike a striking clock, repeaters do not automatically
strike the hour, quarter, or minutes in passing - they must be activated
by the user, usually through a slide or push button.
Watches which do strike "en passsant" (in passing)
automatically are called Petit or Grande Sonneries.
Repeaters currently come in several varieties, based
on the smallest unit of time which they can indicate:
-
Quarter Repeater: Chimes the hour, followed by
the number of quarter-hour intervals at the current time. (example:
4:21 would have four chimes for the hours, then 1 chime for passing the
first quarter-hour.)
-
Half-quarter Repeater: Chimes the hour, followed by
the number of half-quarter hours (i.e. 7.5 minute intervals) past the hour.
(example: 4:21 would have four chimes for the hours, then two chimes for
two half quarters. Note that it would chime three times once getting past
4:22:30)
-
Five-minute Repeater: Chimes the hour, followed
by the number of five-minute intervals (example: 4:21 would have
four chimes for the hours, then four chimes for passing 20 minutes past
the hour.)
-
Minute Repeaters: Chimes the hour, followed by
quarter-hours, followed by minutes (example: 4:21 would have four
chimes for hours, 1 chime for passing the first quarter-hour, and six chimes
for minutes into the second quarter hour.)
These again are very high-priced items, with one exception
- the Kelek five-minute repeater at $5,300 list. More typically,
they are in the $20,000 and up range.
Part IIII: Materials
6.1 Should I be concerned about radium
on a vintage watch dial? How about Tritium? And what's this
Luminova stuff?
The short answer is, don't worry about it, but don't
eat the dials either.
Radium was used after about 1900 as a means of illuminating
watch hands and markers at night. It was widely used until the 1940's
or so, when the hazards of radiation were (belatedly) understood.
Since them, a less powerful radioactive source, tritium (a kind of hydrogen)
has been used. At the time of this writing (November, 2000), tritium is being
phased out of watch dials, partially due to availability of tritium, and
partially due to the development of non-radioactive luminous compounds like
Luminova which "hold their charge" of light better.
The main victims of radium were the watch dial painters,
who were encouraged to keep a fine point on their paint brushes by licking
the brush end. The wearers of the watch receive only a small dose
of additional radiation per year, much less than the natural background
radiation.
A thorough examination of the issue is contained
in Bruce Lulu's definitive article on luminous
watch hands.
7.1 What's the difference between
acrylic, mineral, and sapphire crystals?
An acrylic watch crystal (i.e. the see-through window
above the dial) is a kind of plastic, which has the advantage of being
resistant to shattering, but can be scratched easily. Fortunately,
scratches can be easily removed from acrylic.
Mineral crystal is a kind of glass, which is more
scratch resistant than acrylic, but not as as good as sapphire. Mineral
crystals are also somewhat better at shatter resistance compared with sapphire.
Unlike acrylic, scratches in mineral glass are more difficult to buff
out; and unlike sapphire, mineral glass will scratch. In my opinion, the mineral
crystal seems to be a poor compromise between the two extremes.
A sapphire crystal is indeed made of synthetic sapphire,
which is a transparent form of corundum, or aluminum oxide (Al2O3).
It is extremely hard (Moh's scale 9), and will resist scratching by most
substances short of diamonds. However, if struck sharply and from
the correct direction, sapphire will shatter. Despite the relatively
large size compared with sapphire gemstones, sapphire crystals aren't very expensive
(~$20). (Another fun fact - the windows built in to grocery
store checkout lines which scan for the bar code on products are often
made of synthetic sapphire - for the same reasons that watch crystals are!)
7.2 How do I remove the scratches
from an acrylic crystal?
Good old-fashioned toothpaste is a good start.
Start by buffing a little bit onto a clean cloth over the area of the scratch.
Rub lightly, and rinse with a very slightly damp cloth. Repeat until
gone.
Note that there are also purpose-made polishes for
this purpose, such as Polywatch or Crystal-Kleer
(Legalese: I cannot be held responsible if
you try this for yourself, and end up ruining your watch. This procedure
is intended for people willing to take responsibility for their own actions
only.)
8.1 If a watch is advertised as "18K",
what does that mean?
The term 18K refers to solid gold. Pure gold is
very soft; gold is made in several "karats", or 1/24th proportions of gold,
to make it harder and stand up to daily wear.
9K = 9/24 purity of gold = 37.5% purity (sometimes
seen in vintage and/or UK market watch cases, along with 10K gold)
14K = 14/24 purity of gold = 58.3% (sometimes marked
583 or 585 in gold hallmarks)
18K = 18/24 purity of gold = 75% (sometimes
marked 750 in gold hallmarks)
24K = 24/24 purity of gold = 100%, which I'm guessing
you'll never see in watch cases.
So, that JLC you were looking at in 18K gold has
a case made of real 75% purity solid gold. No wonder it's so expensive.
:-)
(BTW, bracelets sold with solid gold watches are
often themselves solid gold - which as you might expect adds significantly
to the price.)
8.2 What do PVD, CVD, or PE-CVD mean?
Physical Vapor Deposition is a
method of plating gold or other substances to a thickness of several microns
(1/1000ths of a millimeter) over a base metal surface. Other methods
include Chemical Vapor Deposition (CVD), or Plasma
Enhanced Chemical Vapor Deposition (PE-CVD).
This allows the watch to look like a gold watch, but it won't last for
more than a few years of normal wear. When it does finally show through,
it cannot be repaired. Thanks to Mycroft for his information in How
Hard Can you Get.
8.3 What does Gold-Filled mean? How
about Rolled-Gold Plate?
Gold-filled means gold bars are soldered to
a base metal, then rolled into sheets to give a somewhat higher thickness
of gold than any plating method. The resulting layer of gold is slightly
thicker than a sheet of paper. Gold-filled cases were originally
warranted to last a certain number of years in normal wear - e.g. 25 years,
and are somewhat more durable than modern gold plating.
This practice was common until the 1960's, but has
been largely replaced by plating processes.
Rolled-gold plate was another vintage method of plating
gold, except the gold was rolled out flat before being bonded to the base
metal. This process has been supplanted largely by electrochemical
plating and PVD.
9.1 What are some other materials
used as watch cases?
Aside from the standard gold and stainless steel, high-end
watches are often made of platinum or titanium.
Platinum is a very heavy metal, and gives a shiny
white metal appearance. Unlike gold, pure platinum is fairly hard
and resistant to scratches (similar to hard stainless steel). For
this reason, it is often used in 95% purity (i.e. Pt 950). However,
raw platinum is more expensive to use, not only because of its rarity,
but also because it is used in higher purities and requires more effort
to work into a final shape.
Titanium is a relatively light-weight metal, with
a hardness exceeding that of most steels. It also has a poor heat
transfer capability, which means that it won't carry heat away from your
skin as quickly as steel or aluminum (i.e. it will feel warmer to the touch
- sometimes you'll see claims that titanium "remains at skin temperature"
- this is technically incorrect, any more than a small piece of wood stays
at skin temperature). Several varieties of titanium are available.
Titanium has some interesting mechanical properties: it can "rip"
when cut so it is difficult to machine, and two pieces of titanium pressed
together can "weld" themselves together. This latter property is
why it is important that watches with titanium cases and backs have the
casebacks removed periodically - the threads can actually rip out of the
case if left undisturbed too long.
Other case materials which one will run across occasionally
(especially on the vintage market) are Sterling Silver (92.5% purity silver),
Coin silver (80% purity silver), nickel silver or silveroid (not silver,
but 66% copper, 24% zinc, and 10% nickel), nickel, and nickel plated steel.
Also, more recently, one can even find watch cases made of ceramic (zirconium
oxide, as used recently by IWC) and aluminum (used in conjunction with
other metals, as currently used by Porsche Design and BVLGARI.)
For more information, Mycroft's article How
Hard Can you Get has definitive information comparing many substances
you're likely to encounter in the watch industry. In addition, for
more information on Titanium, see Mycroft's other article on IWC's
Titanium.
9.2 How do I remove scratches from
my watch?
Depends on the finish. Justin Time explains the
process for Stainless
Steel. For gold or platinum, use the appropriate polish from
a jewelry store, keeping in mind that a brushed gold/platinum finish will
be harder to match than a glossy finish (see the stainless steel article
for pointers on matching the pattern.)
A certain member of the forum, whose real name is
Norman, sells a leather patch which people have used to good effect in
removing scratches from watches.
For gold filled or gold plate, I would recommend
that you leave them alone - you don't want to remove any more of the finish
than already exists!
10.1 This FAQ sucked!
Where can I get better answers?
Given this is TimeZone, check out TZ
Classics, the Archives,
the Articles
from the TimeZone Community, Walt Odets' Horologium,
Jack Freedman's Escape
Wheeling, and read the Public
Forum , Advanced
Forum, and Satellite
Forums. Don't be afraid to ask a question, but if you've read
this article, you've hopefully gained a fair amount of knowledge.
If you're really motivated to learn more about the
mechanics of watches, enroll in the TZ
Watch School |