EP2447387B1 - Barrel spring of a timepiece - Google Patents

Description

Technical area

The present invention relates to the field of mechanical watchmaking and relates to the field of barrel springs used particularly in watches.

State of the art

In a mechanical watch movement, the energy required for its operation is generally stored in a spring, housed in a barrel and, in fact, called the mainspring. The force supplied is distributed by the escapement and regulated by an oscillator, which is generally a balance-spring.

The force to be transmitted by the spring is determined by the characteristics of the movement from which the spring is sized. The diameter of the barrel conditions the number of turns that can comprise the spring, this number of turn being the essential parameter determining the power reserve of the movement, that is to say the maximum duration during which the barrel can operate the movement in correct conditions.

At present, the barrel springs are made almost exclusively in Nivaflex®. Indeed, a barrel spring must meet several fundamental qualities. It must be stainless, non-magnetic, indefatigable, and have a very high coefficient of elasticity. At the moment, Nivaflex® has all these qualities and offers remarkable performances. We note that they occupy a quasi-monopoly position in the market.

From a practical point of view, if a Nivaflex® barrel spring is used in an existing movement, the dimension available for the barrel and the force that the spring must provide are determined, there is no possibility to modify the mainspring spring to significantly improve the power reserve of the movement, since the number of winding tower without resizing the place occupied by the barrel and thus change the movement.

The present invention aims to allow a significant improvement in the power reserve of a movement, without changing the dimensions of the barrel, nor the elements of the regulating organ.

Disclosure of the invention

To achieve these objects, the invention relates to a timepiece cylinder spring as defined in the claims. The invention also relates to a barrel comprising such a spring and a timepiece equipped with such a barrel.

Brief description of the drawings

• les figures 1 et 2 sont des graphiques illustrant les avantages de l'invention dans le cadre de son application à un ressort de barillet, et
• les figures 3a et 3b montrent un autre exemple comparatif d'un avantage de l'invention dans cette même application.

Other details of the invention will emerge more clearly on reading the description which follows, made with reference to the appended drawing in which:

• the figures 1 and 2 are graphs illustrating the advantages of the invention in the context of its application to a mainspring, and
• the Figures 3a and 3b show another comparative example of an advantage of the invention in this same application.

Mode (s) of realization of the invention

• carbone : de 0.1 à 1%,
• manganèse : de 5 à 25%,
• chrome : de 16 à 20%,
• azote : de 0.1 à 5%, mais au plus dans une proportion correspondant à la limite de sa solubilité dans l'alliage,
• niobium : ≤0.25%,
• molybdène : de 2.5 à 4.2%,
• fer : le solde.

The present invention is based on the use for the production of a clock spring of a timepiece, watch, pendulum or the like, of a metal alloy of the following mass composition:

• carbon: from 0.1 to 1%,
• manganese: from 5 to 25%,
• chrome: from 16 to 20%,
• nitrogen: from 0.1 to 5%, but at most in a proportion corresponding to the limit of its solubility in the alloy,
• niobium: ≤0.25%,
• molybdenum: 2.5 to 4.2%,
• iron: the balance.

This kind of alloy is known as steel 1.4452. It is generally used in medical applications, for prostheses, and even in the watch industry for clothing parts, such as bracelets or watch cases. Indeed, this alloy does not contain nickel and is, in fact, well tolerated in allergies. To the knowledge of the applicant, this alloy is not used for its elasticity qualities.

However, as will be shown below, the Applicant has noted the extreme interest in using this alloy to make watch cylinder springs with remarkable properties.

As mentioned above, the alloy comprises at least iron, in a proportion generally greater than 50% by weight, but without this threshold being mandatory. The alloy also comprises, at least nitrogen, in a proportion ranging from 0.1% to 5%, but at most in a proportion corresponding to the limit of its solubility in the alloy. Depending on the other metals composing the alloy, this solubility limit may vary, so that a numerical value is not relevant.

By way of non-limiting illustration, the alloy may contain less than 0.15% of carbon. It may also contain, more particularly, between 0.75% and 1% of nitrogen. In addition, the alloy may also contain between 12 and 16% manganese.

In order to underline the advantages of the alloy proposed for use in a timepiece cylinder spring, we will now show its performance with respect to the material currently considered as the best, the Nivaflex®.

Thus, to test in a comparative manner the performance of a barrel spring made in such an alloy, barrel springs were made to the dimensions of a standard barrel spring, in Nivaflex®. In the examples below, the springs tested have the following dimensions: 1.18x0.115x455 (mm). For the sake of clarity, it will be specified that these dimensions are, respectively, the height, the thickness and the length of the leaf spring.

For the manufacturing, the first tests were carried out using a manufacturing method usually used with Nivaflex®, this in order to be able to make a rigorous comparison. This method being known, it will not be described in detail.

To make a Nivaflex® spring with the required dimensions, one must have a 0.55mm diameter wire. To achieve this dimension, an annealed blank wire with a diameter of 1.1 mm is drawn, which corresponds to an optimal work hardening rate of 65% to 75% and a tensile strength of 2000 MPa at 2200 MPa. Optimal hardening means hardening to obtain an optimum in the ratio between the elasticity and fragility of the material. Thus, with Nivaflex®, it is known that the drawing must not be maximum, because the material becomes too fragile.

For 1.4452 steel, the optimum work hardening rate is greater than 98%. However, to date, the steel as used in the context of the invention is available on the market in the form of 1.29mm diameter wire, which corresponds to a work hardening rate of 82%. The following tests have therefore been carried out using such a wire, which points to possible additional improvements starting from a wire which has been hardened.

The Nivaflex® and 1.4452 steel wires are then laminated and thermally treated at a temperature of 380 ° for 4 hours.

The springs obtained, therefore having the same dimensions and having been obtained according to the same method for the Nivaflex® spring and the 1.4452 steel spring, have the following results: Nivaflex ^® Steel 1.4452 wire Springs 1.18 x 0.115 x 455 treated 380 ° C wire Springs 1.18 x 0.115 x 455 treated 380 ° C ∅0.55mm M0.5 M4.8 ∅0.55mm M0.5 M4.8 Rm 2018MPa at 6.3% 7.88mNm 6.35mNm Rm 2480MPa to 5.9% 8.25mNm (+ 5%) 6.88mNm (+ 8%)

In this table, Rm means the tensile strength of the spring, the force applied to break it. The value A is the relative elongation of the spring during this rupture. The values M0.5 and M4.8 are the torques provided by the cylinder integrating said spring, respectively when, after having been fully armed, the springs are discharged by 0.5 or 4.8 rounds of barrel. Naturally, when it comes to watchmaking, we want the torque provided by the spring is as constant as possible during disarming of the spring, and that the values M0.5 and M4.8 are as close as possible to one another.

We can already see a clear improvement in the mechanical characteristics of an alloy spring according to the invention compared to a Nivaflex® spring. It can be noted that the break is at a comparable elongation, but that the 1.4452 steel spring withstands much more traction (+ 22.9%). In addition, for a spring of the same dimensions, the transmitted torque is larger (+ 5% or + 8%, respectively at M0.5 and M4.8), especially with better stability during disarming.

In addition, as mentioned above, the alloy used to make a spring according to the invention can be further drawn. It was thus drawn to a diameter of 0.18mm. The theoretical values which would be obtained with a cylinder spring of the required dimensions, obtained from a 0.18mm wire drawn, were calculated. Steel 1.4452 Theoretical properties calculated for treatment at 380 ° C Wire ∅0.18mm M0.5 M4.8 Rm 3010MPa to 5.9% 8.73mNm (+ 11%) 7.44mNm (+ 17%)

It can be seen that a 1.4452 steel spring with these dimensions makes it possible to obtain a 49% greater breaking strength than the Nivaflex® reference spring. It will be recalled that a drawn Nivaflex® spring would give even worse results than the reference spring. In addition, the torque supplied (theoretical values, calculated by modeling) is even better than during tests obtained with a 0.55mm diameter wire, with greater stability during disarming.

In addition, this wire still has an elongation at break of 5.9%, so it is likely that it can be drawn further, thus further improving its breaking strength.

The figure 1 represents the curves of applied tensile force (in MPa) as a function of relative elongation. The sudden drop in force applied corresponds to the rupture of the blade. Curve a corresponds to a 0.55mm diameter wire Nivaflex® (75% work-hardening), the curve b corresponds to a 1.4452 steel wire of 0.55mm diameter (82% of hardening), and curve c corresponds to a steel wire 1.4452 of diameter 0.18mm (98% of work hardening).

With the springs obtained, it was also carried out fatigue tests. To do this, the spring is armed and disarmed successively between 10% and 90% of its power reserve, until the spring break. The following results were obtained. Nivaflex M0.5 to 8.40mNm Steel 1.4452 M0.5 to 8.25mNm Shells ∅2.40mm 1269 cycles 2047 cycles (+ 61%)

One can thus see a very important improvement of the resistance to the fatigue of the springs with the repetition of the armings and disarmages.

Shells tests were carried out to improve this point. The shelling consists in producing the shell, that is to say the inner end of the mainspring, bent in the form of a ring, which is intended to be attached to the barrel shaft. Today, all literature and experience impose a limit on the ratio between the diameter of the barrel shaft and the thickness of the blade forming the spring. This ratio is in principle greater than or equal to 20. In other words, the ring formed by the shell must not be too small relative to the thickness of the leaf spring. The current limit is based on the ductility of Nivaflex®, since the quasi-exclusivity of the barrel springs is made in this material. Thanks to better ductility, 1.4452 steel makes it possible to reduce this ratio.

Tests were carried out with a ratio of 12 times, between the diameter of the barrel shaft and the thickness of the blade forming the spring. In the case tested, the reduction of this ratio is concretely translated by a reduction in the diameter of the barrel shaft from 2.40mm to 1.35mm.

The Figures 3a and 3b show respectively Nivaflex® and 1.4452 steel springs with a shell made with a ratio of 12 between the diameter of the barrel shaft on which the spring is to be mounted (this diameter corresponds substantially to the inside diameter of the ring formed by the shell), and the thickness of the leaf spring. We can see on the figure 3a that the ductility of Nivaflex® does not allow to obtain a perfectly circular shell to this dimension. The elliptical shape promotes the rupture of the blade at the level of the shell. On the contrary, figure 3b illustrates the possibility of obtaining a shell having a satisfactory circular shape. Thus, despite the relationship between the diameter of the shaft and the thickness of the spring blade, the risk of breakage is not increased.

Additional tests were performed. The figure 2 illustrates torque measurements provided at M0.5 and M4.8. They were made with springs obtained by heat treatments at different temperatures. The curves a and b , obtained with Nivaflex® respectively at M0.5 and M4.8 pairs, are relatively steeper than the curves c and d obtained with steel 1.4452, respectively at pairs M0.5 and M4. .8.The decrease in the M0.5-M4.8 deviation is 5% compared to Nivaflex®, resulting in a 25% slope differential. Tests carried out at different temperatures show that, beyond a certain area shown in the graph, the springs obtained are unsatisfactory, either because they are too brittle or because they are not elastic enough. It can be seen that 1.4452 steel offers a much larger working temperature range than Nivaflex®. In addition, 1.4452 steel can be worked at lower temperatures, which reduces the energy consumed for its treatment. In addition, the temperature range and the slope of the curves obtained with 1.4452 steel allow a simpler definition of the desired torque. Indeed, we see that, for the same tolerance, the control of the temperature is less strict to obtain a precise torque.

Concretely, all these improvements give the possibility, for a barrel of 11.50mm of diameter with a blade of 1.18x0.115x455mm, according to the example tested, to go from a power reserve of 48 hours for a given movement, to a reserve of 66.3 hours for the same movement, an increase of 38%.

Thus, the tests carried out show a very clear advantage in making steel barrel springs 1.4452. The test values given above are only non-limiting illustrations. Thus, thanks to the elastic qualities of the material, which are added to its qualities of resistance to oxidation and amagnetism, the improvement made to the barrel springs is notable. By replacing barrels with Nivaflex® springs in existing movements, by barrels with steel springs 1.4452 with a shaft adapted to the new geometry of the shell, we can expect an increase in the power reserve of more than 30%, which is considerable, especially since only the change of the barrel is to be envisaged, without any other modification on the movement, since this improvement is done with constant torque and with constant external cylinder dimension. One could also consider keeping a power reserve similar to that obtained with the springs of the state of the art, but by decreasing the size of the barrel.

It can be added that the interest aroused by these improvements is further reinforced by the ease of implementation of this alloy, which allows optimization of manufacturing costs.

Claims (8)

1. A timepiece barrel spring made from a metal alloy, characterized in that the metal alloy has the following weight composition:

   - carbon: from 0.1 to 1 %,

   - manganese: from 5 to 25%,

   - chromium: from 16 to 20%,

   - nitrogen: from 0.1 to 5%, but at most in a proportion corresponding to the limit of its solubility in the metal alloy,

   - niobium: ≤ 0.25%,

   - molybdenum: from 2.5 to 4.2%,

   - iron: the rest.
2. The barrel spring according to claim 1, characterized in that the alloy contains less than 0.15% carbon.
3. The barrel spring according to one of the preceding claims, characterized in that the alloy contains between 0.75% and 1% nitrogen.
4. The barrel spring according to one of claims 1 to 3, characterized in that the alloy contains between 12 and 16% manganese.
5. The barrel spring according to one of the preceding claims, characterized in that it has an inner eye diameter / blade thickness ratio of less than 20.
6. The barrel spring according to claim 5, characterized in that it has an inner eye diameter / blade thickness ratio of less than 12.
7. A barrel provided with a spring according to one of claims 1 to 6.
8. A timepiece provided with a barrel according to claim 7.

Images