A mechanical circus for watch winding

Watch Winder
I never knew much about automatic watches (and I still really don’t know a lot about them!) until a furniture designer came to me with an unusual project for one of his clients. A collector of self-winding (or “automatic”) watches, the client was looking for a way to keep a number of the watches wound and ready to use.

For the uninitiated, an automatic mechanical watch keeps itself alive by virtue of a spinning pendulum, usually taking the form of a fractional sector shaped rotor whose rotational axis coincides with the mechanism used to wind the mainspring. In practice the movement of the wearer’s arm keeps the watch wound. This concept is attributed to either Abraham-Louis Perrelet or Hubert Sarton depending upon what results from recent historical research.

I was given the opportunity to do some research and produce a proposal for the project. After a couple of months of familiarizing myself with the world of mechanical watches, I began to develop some concepts for the project. A couple of reference designs had been supplied but I chose to leave the options wide open. Some early explorations ranged from the simple and mechanical to oddly anthropomorphic. Loosely riffing off of an extant winder style that rotates watches about what would be the wrist axis while simultaneously spinning each of those axes about a common center produced this admittedly strange thought.

Extra helping hands to keep those watched wound
Another early thought was this turntable that picks up a cassette mounted watch and spins it while simultaneously swing it over to the opposite side as the carousel moves the vacant slot 180 degrees to accept it at the end of its winding circuit. The rotation arm then withdraws as the carousel selects another watch for winding.

A watch cassette turntable.
My own fascination with automatons has certainly found traction in responding to this design challenge. Whatever solution was to eventually emerge would likely have some sense of the mechanical circus quality I like in automaton design.

Eventually, inspiration came from the watchmakers’ pride often expressed in a transparent case that allows a full view of the watch mechanism. The sense of mechanical theater provided by this arrangement led to the final design theme. Here is the final proposal presented in scroll form:

Original scroll-form proposal presentation
 A bit more elaboration on the design led to this in simplified view showing only a watch single rotation point at the winding position now located at the top. The general operating principle has a mechanism in the center of the winder whose function is to alternatively rotate a single watch on its own rotation axis and then move another watch into position followed by a repeat of the cycle. It does this by moving the motor driven gear train into an upper engagement to drive the watch rotation followed by moving it to a lower engagement which rotates the watch carrying “ferris ring”. A pair of counterweights (shown missing their hemispherical add-on masses) offset the weight of the gear train mechanism to enable the use of a modest force in moving the mechanism up and down. The design at this stage utilizes a mounting frame similar to those used in a variety of clock designs. The intended material selection is brass for the mechanism and steel for the framework. At this stage in the design, a pair of commercial linear sliders are specified for managing the linear displacement of the mechanism.

Clearly it is possible to maintain automatic watches by simply rotating them on their own axis as is done with many commercially produced winders. The goal in this case is to introduce a bit of theater into the process as an homage to the elaborate mechanism required of  the watches themselves. But at the same time, there is no superfluous motion involved in the winders’ mechanical process.

Clock frame style mounting
A wide variety of conditions must be met in order for this mechanism to function properly. A few of those include:

  • As suggested in the proposal, each individual watch has its own winding preferences so some provision should be made to identify each watch and provide it with it most beneficial winding sequence.
  • Assuming a 24 hr cycle for maintaining all 12 watches, the process is pretty much continuous if modest operation speeds are to be maintained, so the mechanism must be robust to handle a high duty cycle.
  • Some means to track the position of each winding position must be provided.
  • The ideal rotation-mechanism motor is small in size, quiet and powerful enough to perform the tasks.
  • The electrical system should be discrete with respect to the hardware given practical constraints.
  • User interaction with the winder must be simple and easy to perform.
  • An ideal watch carrier exposes the watch design as one would see it on their own wrist.
  • An effective case design will protect the watches from both physical damage and theft.
The design continued to evolve as practical and aesthetic values balanced against each other. The next significant change came in dropping the commercial sliders in favor of a solution more in keeping with the rest of the design.
Winder mechanism revision
This version includes a new roller based vertical translation mechanism and an elaborated support framework. Additional changes in the gear train have also been incorporated. The watch rotation axes can clearly be seen in this view. I begin to sense a balance between mechanical and aesthetic goals at this stage. On to refinements to meet the other goals. This view omits an upper gear train bracket and the rotation drive motor to clarify the mechanicals. The mechanism in this view is in the upper, watch driving position. A pinion is drawn at both positions, the upper, engaged one of which can clearly be seen where the drive motor would mount. The motor chosen for initial evaluation in the rotation mechanism is a Maxon EC-45, flat style, brushless motor with hall sensors. Initial testing (without an encoder) using the hall sensors  for velocity and position reference proved satisfactory enough to continue development. The addition of an encoder would entail additional electronic hardware and wiring that would have to be  aesthetically incorporated into the design. A nice servo drive called EZ-Servohandles the running of the Maxon motor.

Another design challenge is to present the watches on the winder in an unobtrusive fashion. The first draft design for a carrier is intended to be barely visible from the front of the winder while providing a secure placement for the watch and eliminating the possibility of scratching the watch body or band. This design makes a pretty stab at addressing all of these issues.

Prototype watch carrier
In part two of the watch winder project, I’ll go into further detail on the design challenges and solutions proposed for the winder project.


Leave a Reply