THE TOURBILLON

The following article has been prepared for Europa Star by an expert in the field, Jean-Claude Nicolet, master watch-craftsman and professor in La Chaux-de-Fonds. His study of the Tourbillon is certainly food for thought, and may solicit some controversy among our readers. It is not Jean-Claude Nicolet’s intention to contest the exceptional, professional qualities it takes to construct a hand-crafted Tourbillon. Mr. Nicolet wanted to focus on certain inappropriate qualities which are often attributed to it.

This said, it is important to note that there exists “tourbillons” and “Tourbillons”. In other words, this important watch-making technique may vary in quality. There are a number of differences between a hand-crafted cage, and certain industrially fabricated creations. We look forward to your comments on this important matter.

Tourbillon mounted on a piece with a diameter of 66 mm. A very fine work started by Robert Gafner and finished by his successor Paul Vuilleumier, both teachers at the Watchmaking School of La Chaux-de-Fonds. The piece is fitted with an ingenious independent seconds system imagined by Rober Gafner.

History

Of all the prestigious mechanisms invented by watch-makers over the centuries, the tourbillon enjoys a very special reputation. There are two essential reasons for this: The invention belongs, unquestionably to the most famous watchmaker of all times, Abraham-Louis Breguet, who has rightly been called the “watch-maker of kings and the king of watch-makers”. The second reason is because the tourbillon is a mechanism which, even in its most conventional version, is extremely hard to manufacture. At a time when all tools were still somewhat primitive, the production of a tourbillon watch was a certificate of competence far more appreciated than the diplomas awarded by the best watchmaking schools.

Over the generations of watch-makers, since Breguet, there were probably fewer than 250 able to execute such a masterpiece. Richard Meis, in his book “The Tourbillon”, a work that can serve as historical reference, quotes 120 names. On the 3rd Nivose of year 9 in the Napoleonic calendar (23rd December 1800), Breguet applied to the French Minister of the Interior for a patent to protect the tourbillon. But the invention goes still further back. Sir David Salomons, Breguet’s biographer situates it in the year 1795.

Since Breguet, several great watchmakers, especially those who had dedicated themselves to the research of precision, have created tourbillon movements. The tourbillon regulator is supposed to improve watch precision in a decisive fashion. We’ll see more about it further on.

Definition of the Tourbillon Regulator

What may be referred to as a tourbillon regulator, or more simply a tourbillon, is a mechanical device consisting essentially of a rotating cage carrying at least the balance and the escapement.

The cage is driven in rotation by the gear-train. It turns co-axially to a toothed crown fixed on to the plate. This crown serves to maintain the oscillator by means of a pinion carried by the cage. In conventional tourbillons, this pinion is that of the escape wheel, but this is not obligatory.

In a variant of tourbillon called “carousel”, the cage is driven by the medium pinion, while the third wheel drives the second pinion which is co-axial to the carousel cage. There is no fixed crown in this system, but the angular speed of the second wheel is increased by that of the cage carrying it. This must be taken into account in calculations. We owe the invention of the carousel, in 1895, to the Dane, Bonniksen (1859-1935).

The existence or not of a crown wheel fixed on the plate makes it possible to distinguish the tourbillon from the carousel.

Use of the Tourbillon

In the tourbillon as in the carousel, every time the balance receives an impulse, the cage undergoes a slight rotating movement with all the elements it contains. While the watch is immobile, the oscillator on the other hand changes its position continuously. In the vertical positions of the watch, exclusively, where the residual defect of equilibrium of the organ causes perturbations, this constant rotation of the oscillator keeps up a continual agitation of the rating results. Successive day by day observations in different vertical positions, do not reveal marked differences of rate. One deduces that the tourbillon compensates the effects of gravity as Breguet had anticipated.

A Practical Example

A practical example will help to explain this phenomenon. Suppose we observe four times in succession a good pocket watch at 24 hour intervals in four vertical positions.

With a watch without tourbillon, the readings will be as follows:

The greatest difference between the average daily rate and one of the rates is 5 (difference of 0 to +5 or of 0 to -5). The same watch fitted with the same oscillator but equipped with a tourbillon will give the following results in the same conditions.

The average daily rate is equal to zero as in the first case but the difference corresponding to a position change is nil. The watch seems markedly better. Is it really the case?

Was Breguet right?

In his application for a patent, the famous watchmaker mentions several advantages of the tourbillon, which he summarizes below:

“By means of this invention, I have succeeded in canceling through compensation the anomalies caused by the different positions of the centers of gravity of the regulator movements, to distribute frictions on all parts of the circumference of this regulator’s pivots and of the holes in which these pivots move, so that the oiling of the friction surfaces is always even, in spite of oil coagulation, and finally to cancel many other causes of error influencing movement accuracy, which could only be attained until now by constant trial and error and often even without any certainty of success.” Afterwards, Breguet mentions only the compensation of the effects of gravity due to changes of position.

He is careful not to say that the tourbillon does not correct the rate variations in the “pendant up” and “dial up” positions. This is how timers define the rate variations between the two horizontal positions and the three vertical positions usually controlled in the observatories.

Finding the Right Position

As a result, a watch laid flat at night and always placed in the same vertical position during the day does not provide better results when fitted with a tourbillon regulator. It’s more the contrary that is true, because in a vertical position the tourbillon watch gives an average rate while the watch without a tourbillon gives only the rating result corresponding to the position it occupies. It can easily be arranged that one of these vertical positions - as chosen - gives much the same result as the horizontal positions. The possible advantage of the tourbillon then shows only if the watch changes occasionally its vertical position during the day. If this change of position is very frequent, the “mixing” of positions occurs of itself.

In view of this, it is certain that, when worn, a wristwatch with a tourbillon is not more accurate than a conventional watch.

Why is there so much admiration for the tourbillon?

If what you have just read is true, it is difficult to understand the veneration in which the tourbillon has been held by watchmakers for some two centuries. There are several reasons for this. First of all, the unquestioned and deserved prestige of Breguet, the inventor.

Secondly, all the tourbillon watches presented in the observatories were manufactured by watchmakers or renowned companies. Thirdly, the manual production of a tourbillon watch can be carried out only by particularly skilled watchmakers and its successful execution is an unquestionable certificate of mastery.

Finally, the superiority of the tourbillon becomes evident only in observatory controls where the pieces are tested for 25 days in five successive positions, 2 horizontal and 3 vertical ones. The watches are treated with the greatest care. The length of observation between two reports is exactly 24 hours. It’s the only way to act in order to show the variations due to position changes but these conditions are far from being the same as when the watch is worn. Tested in this manner, tourbillon watches benefit from a marked advantage. They possess a little mischievous imp - the tourbillon - who is always modifying the vertical positions so that all the efforts of the testers, to show the position differences, are thwarted. The tourbillon does not correct position variations; it only prevents their being detected in the usual testing conditions. But when placed on the vibrograph, the tourbillon reveals its weaknesses immediately.

Comparing two Options

In order to make a valid comparison between the two types of watches, those without a tourbillon should be placed in a vertical position, in a rotating cage with a rotating speed of about 1 r.p.m. or else one should give up observing the vertical positions. Under these conditions, the advantage would go unquestionably to the watches without tourbillon. We’ll see why later.

The admiration given to the tourbillon is of the same nature as that given to a clever conjuror who causes an elephant to vanish! What we think we see is a mere optical illusion.

From Doubt to Certainty

The first doubts regarding the efficiency of the tourbillon were felt quite early on but the inventor’s halo prevented them from being expressed. If the tourbillon had an effect comparable to that of the hairspring applied to the balance, the difficulty of its design would not have seemed so great. The small number of tourbillons made is another factor that reveals the system’s lack of efficacy. In a remarkable study that was published in issue number 22 of the “Revue de Association française des amateurs d’horlogerie ancienne” Messrs Droz and Florès make a few apt criticisms that we are reproducing with authorization.

A Subject of Controversy

Claudius Saunier, editor of the “Revue Chronométrique” for almost 60 years (1855-1914), seems almost to ignore the system. There are practically no articles devoted to it. A laconic remark at the bottom of a page says: “The tourbillon is no good, Breguet recognized the fact...”That is an extreme opinion. In the Revue Chronométrique, the tourbillon by Benoît is presented as “a modern tourbillon”. In a short article, Maillard Salins, shop manager of the Leroy firm, expresses the greatest doubts on the efficacy of the tourbillon and carousel systems.

In “Pendule, Spiral, Diapason”, a book published in 1920, H. Bouasse, professor at the Science Faculty of Toulouse, France, writes in tome 11: “In order to cancel the centering errors of the balance, someone had the ingenious idea of making it turn: this resulted in the tourbillon escapement, but its high price and problematic advantages reduced it to being a historical curiosity.... Experience shows that in the end, the use of this”perfected“system involves considerable costs. It is ingenious but far too complicated.”

Other authors, such as Marius Favre in the “Journal Suisse d’Horlogerie”, issue No 3/1905, analyses some results obtained in the observatories and gives the preference to fixed escapements. His analysis is particularly apt.

Defining Criteria

The table below presents the respective rankings of 706 observatory pieces of which 174, that is 25%, are fitted with a tourbillon. This 25% obtained 59% of the first places. At priori, these results speak in favor of the tourbillon. However, if one remembers that one of the criteria among the most difficult to satisfy - the one concerning positions - does not enter into consideration in the assessment of the tourbillons, one is somewhat disappointed with these results. The other criteria need to be unfavorable to them for tourbillons not to obtain systematically the first places. A detailed examination, criterion by criterion, confirms this hypothesis.

Why are tourbillons disappointing?

As we have seen, the tourbillon does not correct anything. It only prevents the detection of an error that still subsists. It does not even produce a contrary defect able partly to cancel the first one, as does for instance the recoil escapement in clocks. This escapement partly corrects a pendulum’s error of isochronism.

The tourbillon is in fact an additional mechanism that consumes energy without producing anything except misinformation. It is a parasitical mechanism. The energy it consumes is taken from the reserve destined to the regulator. As a result, the balance with less energy will have reduced advantages. All watches established to function in both manners, with or without tourbillon, always have smaller balances of less good performance in tourbillon version.

Energy distribution is carried out during the rotation of the escapement wheel. It is at that moment too that the tourbillon cage turns. It consumes a great deal of energy to start moving and hardly is it launched than it stops suddenly against the escapement pinion whose wheel has just come to rest. All the energy drawn by the cage is absorbed by the shock without any advantage. The light escapement wheel has given back 25% of its energy to the balance which has done nothing but shake up all the regulating system. This is what is described as the “ingenious invention of Breguet”. The latter luckily has other titles of glory.

The fineness of the tourbillon cage makes its execution very difficult. This cage is the work of one of the three great cage manufacturers of the end of last century. Messrs Albert Pellaton, from Le Locle, Roulet and Grether from the Ponts-de-Martel.

Timing and Balance

The faster the tourbillon cage turns, the more spectacular its operation and the more it consumes energy at a loss. The tourbillon would be less harmful if it turned more slowly because it would consume less energy and its effect of garbling information would be identical, as also its uselessness. Timers have always sensed the disrupting effect of the tourbillon. Why did they build cages as fine as cobwebs? It was to reduce to the maximum their moment of inertia and not to force the admiration of connoisseurs. Why did they seek to balance the filled cages, if not to correct their imbalance, which would constantly change the amplitude of the balance and show up the errors of isochrononism.

Other inconvenience: by integrating all the vertical positions, the tourbillons make the timing in the “plat pendu” (“pendant up” and “dial up”) positions more difficult. This is felt more particularly when the pieces are small. On the whole tourbillon wristwatches time less well than conventional wristwatches of the same size. In fact, all traditional wristwatches are fitted with a free tourbillon deriving its energy from the wearer and not from the spring. When they are worn, these watches occupy an infinite variety of positions and their rating results give an average of this constant motion.

Eliminating the Gravity

Gravity is one of the main causes of rate variations in watches. By creating the tourbillon, Breguet thought he was eliminating its effects. It was an error on his part. He succeeded only in masking them a little. If he had found a way of eliminating the force of gravity on watches, he would have realized an improvement comparable to the invention of the balance-spring. Unfortunately, there is no way of suppressing gravity on earth, except by putting watches into artificial satellites. This has been done, since astronauts wear watches. It seems however that nobody has thought of comparing the rate of these watches on earth and in a weightless atmosphere. We propose this very interesting experiment to the watch and clock industry.

After having observed them on earth according to specifications to be defined, a few watches of different qualities would be entrusted to some astronauts. The same observations to be carried out again in a weightless atmosphere and comparisons made. An interesting evaluation could then be obtained on the effects of gravity.

One can be pretty sure that the watches would have a better comportment in a weightless atmosphere. All errors of equilibrium would disappear as by enchantment as would the sagging of the balance spring and the friction’s due to the weight of the mobiles. It can even be surmised that the rate variations would be more marked with the least good watches and that, in a weightless atmosphere, they would behave like chronometers.