This text is from an article I wrote for the MBSI Journal in the Spring
of 1993.  If someone will tell me how to do it, I'll send in the
illustrations (black and white line drawings) as well.

The fusee was invented to solve a problem that made time-keeping
notoriously inaccurate in the earliest days of clocks (and watches),
and was the third major invention that attempted to better control the
release of power from a spring that was to run a clock.

The earliest existing spring driven clock is dated circa 1450 and is in
the Victoria and Albert museum in London.  In those early days (15th
century) most clocks were weight-driven, the invention of the spring-
driven clock probably came about to make the clock more suitable for
domestic use.

The problem with the early spring driven clock was that the springs,
made early-on of hammered brass, and later of hardened and tempered
steel, could not be made uniform and therefore could not give off their
power in a uniform manner.  This was hardly an appropriate state of
affairs for running a clock.

When the spring was fully wound it was very powerful and drove the
mechanism quickly; when almost unwound it was weak and drove the
mechanism slowly.  In the middle of the run was the most steady power
and this was the reason for the development of a mechanism called the
'stopwork' that limited the coiling and uncoiling of the spring to the
middle of the range.  We call that stopwork the 'Geneva lock', I suppose
after the town in Switzerland in which the present form of stopwork used
on music boxes was first designed.

But even the stopwork could not properly provide uniformity of power.
Soon someone invented a variable brake called the Stackfreed.  (In figure
2 you can see a schematic of how the stackfreed works.)  Basically, it is
a spring loaded cam which gradually reduces its braking effect on the
main drive wheel of the mechanism as the spring unwinds.

The stackfreed was not very effective and did not last long.  It was
superseded, early in the fifteenth century, by an ingenious mechanism
called the fusee.  (It was probably invented by Leonardo Da Vinci -- an
illustration of a fusee is found in one of his notebooks dated a little
before the end of the 15th century.)  This device is so simple and yet so
effective that it is still being used today in certain special clocks,
especially mechanical marine chronometers.

According to Eric Bruton, 'fusee' probably comes from the Latin 'fusata',
meaning a spindle wound with thread.  (In figure 1 we see the details of
a typical fusee movement.)  The spring barrel provides the power, the
chain or gut line transmits this power to the cone-shaped fusee and the
fusee transmits its variable power to whatever it is supposed to drive.
Of course, the fusee and the spring barrel can be set horizontally as
well and that is how they are set in fusee watches and in marine
chronometers -- and also in the little musical movement we've been
discussing.

If you think of the fusee as an infinite series of levers (you can think
of it also as infinitely variable gearing or as a stepless automatic
transmission in an automobile) you can see that when the spring barrel is
fully wound it acts on the small part of the fusee with relatively little
leverage.  As the spring unwinds and reduces in power it acts on the ever
increasing diameter of the fusee, in effect, using a bigger and bigger
lever to run the mechanism.

This tends to even out the power provided by the spring and, with a
carefully calculated and fitted fusee the power applied to the clock's
gear train will be almost perfectly uniform.  In the early days springs
and fusees were made for each other, i.e. matched, so that power
transmission would be as perfectly uniform as possible.

The earliest music boxes, especially the early cartel boxes, used fusees,
perhaps to keep the power applied to the pin barrel more uniform and
perhaps because through the concept of 'cottage industries', the drive
mechanisms of the early boxes were made by clock and watch-making
families.  Fusees continued to be used until about 1815 and perhaps
longer as makers used up old stocks of parts but then were replaced by
the familiar 'going barrel' which, with its integral gearing, drives the
pin barrel directly.

Fusees were used in almost all of the early non-weight-driven mechanical
organs.

Frank Metzger