The Girard-Perregaux Constant Escapement, An Invention That Started At Rolex
Truth is stranger than fiction.
For this week’s installment of A Watch A Week, we have something almost never seen in modern horology, or for that matter, in horology period, even over its many centuries of evolution. The watch we’re looking at this week is from Girard-Perregaux, and it’s the Constant Escapement watch – a watch with a true constant force escapement.
The constant force escapement is probably the rarest type of escapement in watchmaking or clockmaking. You might not think so if you’re thinking of constant force mechanisms in general. There are two basic types of constant force mechanisms. The first is the train remontoir, which is basically a smaller, subsidiary mainspring on the pivot of one of the train wheels, which is periodically wound up by the primary mainspring. The remontoir spring provides a steady flow of energy to the balance which does not change as long as the primary mainspring has enough juice to keep rewinding it (if it rewinds once per second you can drive a deadbeat seconds hand off it, as an added bonus for your efforts). The watch or clock will keep running once the mainspring can’t rewind the remontoir anymore, but torque will start to fall off and isochronism, and precision, will start to suffer. (The spring remontoir was invented by John “Marine Chronometer” Harrison, but the much older gravity remontoir was invented by the Swiss horologist and mathematician Jost Bürgi, at the end of the 16th century. If you haven’t heard of Bürgi there are several reasons he might have escaped your notice, one of which was that he was not very energetic in promoting his own inventions – a contemporary described him as, “an indolent man and very uncommunicative.”)
The other constant force mechanism is the chain and fusée, which consists of a cone with a spiral groove cut into it. A chain running in the groove gradually unwinds from the fusée onto the mainspring barrel and the whole arrangement looks a lot like the gears on a ten-speed bike, and in fact serves the same function – as the mainspring torque runs down, the chain unwinds from a larger and larger diameter section of the spiral, giving the mainspring a progressively greater mechanical advantage as it weakens over the course of its power reserve.
Now the reason that watchmakers bothered with such complicated additions to their already complicated machines, is that for much of the history of watchmaking, precision was very connected to the amount of juice getting to the balance. A balance is supposed to be isochronous – that is, to take the same amount of time for each swing – regardless of amplitude and within reason, a balance with a balance spring will do just that. However, practically, once the amplitude falls below a certain point, isochronism starts to suffer, the balance begins to run erratically on its rate, and it also becomes much more susceptible to positional errors. Since mainsprings deliver a lot more power fully wound than at half wind, or close to the end of the power reserve, it behooved watchmakers to come up with solutions to the problem of, you guessed it, delivering constant force to the escapement.
That’s where the GP Constant Escapement comes in.
In the picture, the balance sits under its balance bridge, and is impulsed by the lever above it which is shaped like an inverted Y with long, curved arms. Running through the tip of each arm is a long, thin, S shaped blade spring. When the balance unlocks the lever, the spring begins to straighten out as the lever pivots. As the lever begins to pivot, the two contra-rotating escape wheels to the right and left unlock, and their curved faces press on the lever’s locking jewels, pivoting the lever further. As the lever reaches the mid-point of its swing, the thin blade spring snaps from one S shaped to its mirror, which in turn, snaps the lever the rest of the way through its rotation and this impulses the balance. One interesting detail is that the escape wheels rotate in opposite directions; in the picture, the escape wheel on the right is pressing against one of the two locking stones on the lever.
The silicon blade is fabricated in one piece with the large oval silicon frame enclosing the balance and escape wheels, and the blade itself is just 14 microns thick.
The action of the escapement is better understood if you can watch it do its thing in slow motion, as in this YouTube video from WristReview:
The Rolex Connection
The story behind the escapement is almost as interesting as the escapement itself. The original idea came from a watchmaker at Rolex named Nicolas Déhon. When this watch first debuted at Baselworld in 2013, GP’s head of research and development explained the invention of the mechanism to me when I wrote it up for Revolution:
“As G-P R&D engineer, Stéphane Oes, explains it, Dehon was commuting by train when he had an epiphany while absentmindedly flicking his ticket between his fingers. A blade spring, kept slightly flexed between two fixed points, could like his train ticket be made to snap back and forth, delivering a predictable pulse of energy each time. If the bit of cardboard between his fingers could be transformed into a source of energy for the balance, it might be the answer to the constant force riddle. The process of development began, and though Dehon and Rolex were able to cobble together a prototype, and even took out a patent for this constant force escapement (granted in 1999) its design was still slightly ahead of its time. The materials needed to build a reliable and effective version were not available.”
One of the prototypes somehow ended up being offered for auction at Antiquorum in May of 2018, where it finally sold for the staggeringly low price, considering it’s about the rarest piece of Rolex ephemera you could ever hope to find, for just CHF 4375. At Rolex, it was called “Projet ELF.”
The relationship between Projet ELF and the GP Constant escapement is immediately obvious, so why didn’t Rolex put the escapement into production? The answer is that it didn’t work reliably and Rolex is all about reliability. The blade spring in particular was made of metal alloy, and did not change configuration reliably at the critical transition point and was also probably susceptible to changes in impulse delivered, depending on temperature changes, which has historically been the obstacle preventing constant force escapements from being more widely produced. George Daniels studied Breguet’s attempts to make constant force escapement clocks, and wrote, in Watchmaking, a couple of apparently contradictory assessments:
“Such escapements do not fulfill their promise, and their rate of timekeeping is not better than a conventional detent escapement.” But, in The Art of Breguet, he commented, “Mechanisms of this type … are extremely delicate and difficult to adjust [but] once properly set up, they are very reliable and keep close time … when properly adjusted and cleaned these escapements are capable of a performance superior to any other balance wheel timekeeper.”
By 2008, however, a new material had started to come into wider use in watchmaking: silicon. The initial research into silicon escapements took place at the Swiss CSEM (Centre Suisse d’Electronique et Microtechnique) lab, as an effort funded by a consortium consisting of Rolex, Patek, the Swatch Group, and which was preceded by Ulysse Nardin’s work on silicon components for the first edition of the Freak. Silicon offered a manufacturing technique capable of producing the tolerances necessary for the Constant Escapement to work, and after several years of intensive research and development, Girard-Perregaux showed a working prototype of the watch.
I have always insisted (sometimes in the face of counterarguments from technical directors from other brands who insist with equal vehemence that their remontoir is a constant force escapement) that a true constant force escapement needs to basically fulfill the definition given by Daniels: “In the constant force escapement, the impulse to the oscillator is delivered by a separate, small detent, either spring or pivoted with spring. After each impulse to the balance the detent releases the escape wheel which re-sets it on its catch ready for the next impulse.” By this definition, a watch with a remontoir or fusée isn’t a constant force escapement, albeit it may have a constant force mechanism – the escapement will generally be a conventional lever and escape wheel.
A Revolution In Waiting
The whole idea of creating a working constant force escapement was the most exciting thing in the world for me ten years ago, and for a lot of the rest of the watch enthusiast press as well. In the comments to the Revolution story, the inventor himself, Nicolas Dehon, commented (pardon the machine translation from his French):
“This project represents 10 years of my professional life, from 1998 to 2008. From this wonderful adventure I will probably have only one regret; that of never again experiencing anything so fascinating through watchmaking. I would like anyone who has followed this development to keep in mind, however, that this is probably only the beginning of a story that I hope will be as long as possible; as long as that stretching from the invention of Thomas Mudge to its probably most successful version: the Pulsomax by Patek Philippe.
Many improvements are still possible on this ‘Constant Escapement’ which I still prefer to call in a more generic way ‘bistable spring escapement.’ It’s up to everyone to rack their brains to improve the system: the games are open!”
As it turns out, the Constant Escapement, as well as other experiments in replacing conventional escapements with flexible, or compliant, silicon components, has not yet managed to substitute wholesale for the standard lever escapement (or its variant, the co-axial). The reasons are pretty obvious – something as absolutely fundamental as an escapement needs to be not only precise, but also, easy to produce at scale without sacrificing precision, and anything that’s going to unseat the existing dominant type of escapement – the lever – is going to have to be, ideally, cheaper to make, require less servicing, and tolerate the sort of bashing around that a lot of mechanical watches are subject to.
On the other hand, there is nothing wrong whatsoever with fascinating experiments that offer a deep thematic connection to some of the most fundamental problems in watchmaking. There are many things in the world, of watches and otherwise, which are impractical but which are no less fascinating – Daniels himself once wrote of the remontoir, “The fact that it is perfectly unnecessary merely adds to its charm.” Thus it is with the Girard-Perregaux Constant Escapement – the action of the escapement is fascinating to watch, it has a fascinating back-story, and it represents a moment of real inspiration in the evolution of horology and micromechanics. Einstein wondered what the world would look like if you rode a beam of light; Nicolas Dehon idly snapped a tram ticket back and forth on his way to work; the rest is history.