CH712940A2 - Method and system for determining certain parameters of a watch - Google Patents

Abstract

The present invention relates to a system (1) equipped with a non-acoustic sensor (3) for determining the gait, the moment in time (t MAX) where the amplitude of an oscillator of a watch movement is maximum , the amplitude and the reference of a movement (10) of a mechanical watch. The system uses methods for processing and analyzing signals measured by the sensor during the passage of a member making a periodic movement, such as the hand of the watch, the oscillator (15) or the exhaust. Due to the low cost of the sensor, preferably an optical sensor, and other components of the device, the latter is advantageous over the technologies disclosed in the state of the art.

Description

Description: [0001] The present invention relates to a method for determining parameters related to the regulating member of a watch, such as the running of the watch, the amplitude of the oscillator, the period and / or the frequency of this. last, and / or the benchmark, among others. The invention also relates to an apparatus for implementing the method according to the invention.

State of the art and problems at the origin of the invention [0002] In mechanical watchmaking such as wristwatches, the period of an oscillator is used to define a time base. An indication of the precision of a watch is the daytime running, or the error, expressed in seconds, taken after an interval of precisely 24 hours. The march is positive for an advance and negative for a delay. One element responsible for the error of the time base lies in the anisochronism of the oscillator, that is to say, the fact that the period is dependent on the amplitude of the oscillator. In order to be able to comprehensively analyze the regulating organ of a mechanical watch, it is therefore necessary to determine the relationship between the period and the amplitude. Aside from daytime walking, another important parameter is the benchmark. This last parameter, measured in milliseconds, quantifies the degree of asymmetry of the oscillation with respect to the exhaust.

Various devices and methods designed to determine the above parameters are known in the art. For example, document EP 2 881 809, filed in the name of the company Witschi Electronic AG, discloses a method for measuring the frequency and the amplitude of the oscillations of a mechanical oscillator, comprising, inter alia, the step to command a camera to acquire images of the oscillator when the latter is at minimum and / or maximum elongations, and to determine the amplitude from these images. A process using a camera is also disclosed in the publication by Meissner et al, "A new device for measuring the regulating organ for the mechanical watch", Proceedings of the International Congress of Chronometry, in Colombier on September 26-27, 2007.

In devices that contain a camera, walking is generally determined by taking images at two moments spaced in time to capture two moments and by comparing the position of the second hand of these two moments (for example:

Qualimatest, COSC, and US patent document 9,348,317). The techniques by shooting cameras work on several dimensions (2D and more), which requires further processing and therefore more computing resources. The costs for processing information are therefore higher in the case of cameras. The accuracy of the measurement is related to the resolution of the camera with the optics. The high resolution camera is impractical at high frequency (i.e. at 20 kHz), due to the excessive computing resource or storage requirements.

On the market, there is for example the product WisioScope S, designed to measure mechanical movements acoustically and optically. This device is equipped with a laser, a microphone and a camera associated with the microphone, in order to determine the angle of lift. Similarly, the SMEV Laser Velocimeter product has a laser and microphone.

The above documents and apparatus are characterized by the presence of relatively complex and / or expensive equipment, such as a laser or a camera. On the other hand, the methods for determining the amplitude or other parameters generally take into account the data received by the microphone and thus require an interpretation of the acoustic signal in parallel with the interpretation of the signal from the optical system. Overall, the prices for these devices are high. A disadvantage of using the acoustic signal to determine, for example, the benchmark, is that these methods are generally adapted to a particular exhaust system, generally to the Swiss anchor type exhaust, since the noises measured depend on the type of the exhaust. Given these drawbacks, an objective of the present invention lies in the implementation of a simple device, using simple and inexpensive components, to carry out the measurements of the regulating member of a watch while freeing itself the type of exhaust.

Other disadvantages of existing measurement systems are listed in the above-mentioned publication by Meissner et al, 2007.

Document CH 691 992 discloses a method for determining the lifting angle, that is the angle that the balance traveled between two impacts of its exhaust. The device disclosed in this document includes an electrooptical unit as well as an electroacoustic unit comprising a microphone. In order to measure the lift angle, this process measures the time between two shocks produced by the escapement of the movement. The present invention does not seek to determine the lifting angle but to determine the gait, the amplitude and the reference point on the basis of optical measurements only, in the absence of an electroacoustic unit and without knowing the lifting angle and without using this parameter. This would make it possible to determine the amplitude independently of the type of exhaust, since the noise measured as described in CH691992 is characteristic of an exhaust of the Swiss anchor type.

In summary, the present invention seeks to implement a device for measuring the main parameters of the regulating organ of a watch is the gait, the amplitude and the reference on the basis of optical measurements only, without need acoustic information from, for example, exhaust. Preferably, the present invention seeks to implement an apparatus whose components are simple, without the need for a camera or a laser, using for example a simple optical proximity sensor.

CH 712 940 A2 [0010] The Swiss patent application CH 706 642 A1 relates to an optoelectronic instrument for characterizing in real time the movement of the mobile elements of a caliber of mechanical watch. The device disclosed in this patent includes an array of optical sensors and at least one lens. The camera is used to acquire images. Overall, the device is complex. This device does not determine the benchmark of a mechanical watch. In view of this document, it is an objective of the present invention to implement a device of a simpler construction, less costly, and also making it possible to determine the reference point.

Summary of the invention In one aspect, the present invention relates to a system for determining one or more parameters chosen from: (i) the running of a timepiece, (ii) the amplitude of the oscillator of the movement of a mechanical timepiece, (iii) the moment in time (t _M Ax) when the amplitude of an oscillator of a movement of the mechanical timepiece is maximum, and / or ( iv) the reference of a mechanical timepiece. In one embodiment, the system makes it possible to determine all of these parameters.

In one aspect, the present invention also relates to one or more methods for determining one or more parameters chosen from: the running of a watch, the amplitude of the oscillator of the movement of a mechanical timepiece, the moment in time (ϊμαχ) when the amplitude of an oscillator of a movement of the mechanical timepiece is maximum, and the benchmark of a mechanical timepiece.

In one aspect, the invention relates to a system comprising: an apparatus comprising: a non-acoustic sensor arranged to be able to distinguish disturbances during the passage of at least one distinctive element of an object of said watch, said object performing a periodic movement and being chosen from the hand of a watch, the oscillator of a watch, the escapement and / or part of one of the aforementioned objects; a support, arranged to temporarily place a watch or watch movement relating to said sensor in order to allow the measurement of the aforementioned disturbances; a microcontroller, arranged to be able to receive signals from said sensor, said microcontroller comprising or being associated with a time base; and at least one computer code configured to be able to implement the method according to the invention.

In one aspect, the invention relates to a method for determining the running of a watch, the method comprising the steps of providing a non-acoustic sensor arranged to be able to distinguish disturbances during the passage of a distinctive element of a watch object, said object performing a periodic movement and being chosen from the hand of a watch, the oscillator of a watch and / or part of one of the two aforementioned objects; measuring a first real signal by said sensor during the passage of said distinctive element; creating on the basis of an actual measured signal a theoretical signal at a later defined time, said theoretical signal substantially defining a signal which had occurred, or which would occur later, if the running of the watch were zero; measuring at least a second real signal by said sensor during the subsequent passage of said distinctive element; determining the difference in time between said second real signal and said theoretical signal; and, determining the gait by reducing said difference to an interval of 24 hours.

In one aspect, the invention relates to a method for determining a moment in time (ϊμαχ) where the amplitude of an oscillator of a watch movement is maximum, the method comprising the steps of meter available d 'A non-acoustic sensor arranged to be able to distinguish disturbances during the passage of at least two distinctive elements of said oscillator during the oscillation of the latter; measuring successive signals during said oscillation; digital processing of the measured signals, so as to obtain separate successive signals having a distinctly lower or greater amplitude than the signal obtained outside said passages; optionally: determining the frequency and / or the effective period of said oscillator, preferably using the step determined according to the method of the invention; extracting a sequence of signals obtained during a duration equal to or longer than said period; determining at least one center of symmetry in said sequence; determine t _M Ax as the moment in time corresponding to said at least one center of symmetry.

In one aspect, the invention relates to a method for determining the amplitude of an oscillator of a wristwatch on the basis of non-acoustic measurements only, the method comprising the steps of providing a sensor non-acoustic arranged to be able to distinguish disturbances during the passage of at least two distinctive elements of said oscillator during the oscillation of the latter; measuring successive disturbances from at least two distinctive elements of said oscillator; digital processing of the disturbances measured during the passage of said at least two distinctive elements, so as to obtain a distinct signal for each passage of one of said at least two distinctive elements; determining time values L and t ₂ by assigning a time in time to two distinct signals associated with the successive passage of a first and a second of said at least two distinctive elements; determine the amplitude (A _M ax) of the oscillator by using the properties of the signal at times ti and t ₂ and by solving the function of the sinusoidal movement according to formula (V):

A (t) = Amax * sin (ω * t + φ) (V) in which ω = 2 * π * f, f being the frequency measured by the oscillator; φ is the phase shift.

CH 712 940 A2 In one aspect, the invention relates to a method for determining the benchmark of a watch movement, the method comprising the steps of providing a non-acoustic sensor arranged to be able to distinguish disturbances during the change of position of an element of the movement escapement from a first stop position to a second stop position, said change taking place due to the passage of the oscillator at the point of zero elongation; acquiring successive signals by said sensor of said element of the exhaust; digital processing of the measured signals, so as to obtain a processed signal containing distinct successive signals which can be associated with said two positions of said element of the exhaust; determining time durations d-ι and d ₂ as the durations of two separate successive signals;

determine the motion frame as (d ₂ -di) / 2 or (d- | -d ₂ ) / 2.

In one aspect, the system of the invention preferably operates through a preferably non-acoustic sensor. The sensor is preferably a one-dimensional sensor. The sensor preferably provides a single signal and / or a value at a time in time. This sensor allows and / or is preferably limited to carrying out one-dimensional measurements. Preferably, this sensor is not a camera and not a network of sensors.

In one embodiment, the sensor can be chosen from an optical sensor, a capacitive sensor, and an electromagnetic sensor, preferably an optical sensor chosen from sensors arranged to pick up a light reflected or interrupted by a distinctive element of the movement.

In one aspect, the system sensor is arranged to be able to detect disturbances during the passage of at least one distinctive element of an object performing a periodic movement, such as a needle, the oscillator or the escapement or part of the escapement of said timepiece. Based on the processing and analysis of signals and using appropriate algorithms, the present invention makes it possible to determine several key parameters of the movement of a timepiece. The sensor is preferably a sensor having a comparatively low cost, which makes the system of the invention particularly advantageous and competitive.

In one aspect, the invention relates to the use of a one-dimensional sensor and / or a sensor preferably providing a single signal and / or a single value in a moment in time in a system and / or apparatus for determining the running of a watch, the amplitude of the oscillator, the moment in time (t _M Ax) when the amplitude of an oscillator of a movement of the mechanical timepiece is maximum, and the mark of a mechanical timepiece.

Other aspects of the invention and preferred embodiments are defined in the appended claims and in the description below.

Description of the Drawings Other aspects, characteristics, properties and advantages of the present invention will appear more clearly on reading the detailed description of preferred embodiments which follows, made with reference to the appended drawings given by way of examples non-limiting and in which:

Fig. 1 schematically shows an embodiment of the system of the invention.

Fig. 2 illustrates the processing and analysis of signals in accordance with a method for determining the running of a watch according to an embodiment of the invention.

Fig. 3

Fig. 4A to 4B

Fig. 5

Fig. 6

Fig. 7 illustrates signals detected by the sensor and processed in order to determine t _M Ax of the oscillator during oscillation in accordance with an embodiment of the invention (part A) the sinusoidal movement of the oscillator (part A) .

illustrate the determination of t _M Ax according to an embodiment of the invention, by the definition of a supposed center of symmetry (A, B) and by the improvement of the supposed center of symmetry using the correlation (C, D ).

shows a signal detected on the exhaust, the signal being able to be used to determine the reference in accordance with an embodiment of the invention.

shows a desired and / or ideal theoretical signal, created on the basis of a signal as shown in fig. 5, to determine the benchmark in accordance with an embodiment of the invention.

illustrates the results of a correlation for determining the moments of change of position of the exhaust, in accordance with an embodiment of the invention.

Detailed description of the preferred embodiments The present invention relates to a system and methods for determining one or more parameters of a timepiece, such as the running of a watch, the amplitude of the oscillator of the movement of a mechanical timepiece, the moment in time (t _M Ax) when the amplitude of an oscillator of a movement of the timepiece is

CH 712 940 A2 maximum, and the reference of a timepiece. In a preferred embodiment, the timepiece is provided with a mechanical movement. Preferably, the timepiece is a wristwatch with mechanical movement.

Figure 1 schematically shows an embodiment of the system 1 of the invention. The system includes at least one device 2 and at least one computer code 7 for determining the aforementioned characteristics using, for example, algorithms as will be described in more detail below. Preferably, the device 2 comprises a support 4 arranged to temporarily place a watch or a watch movement 10. The support 4 preferably makes it possible to position the watch / the movement relating to a sensor 3 in order to allow the measurement of signals as described below. For example, the support may include a vice and / or screws, or any other assembly for attaching and / or immobilizing the watch and / or the movement at least temporarily, during the time taken for the implementation of one or all methods of the invention.

In one embodiment, the device 2 comprises a microcontroller 5 arranged to be able to receive signals from said sensor 3. The microcontroller comprises or is associated with a time base 8. The time base is preferably more precise than the timepiece whose progress is to be determined. In a preferred embodiment, the time base 8 itself has a step less than or equal to (<) 1 second per day (s / y), preferably <0.7 s / d, more preferably "0.5 s / d , for example <0.1 seconds per day.

The computer code 7 can be produced in the form of software, for example. In order to be able to operate, the system of the invention preferably uses a data processing unit 6, capable of providing for executing the computer code 7. The computer code 7 is preferably configured to be operated on the processing unit data 6, the code being configured so as to implement at least one and preferably all the methods according to the invention. The data processing unit 6 can be integrated into the device 2.

Alternatively, the device 2 has an interface 9 allowing the device to be connected to a separate data processing unit 6.

The data processing unit typically comprises a processor, one or more memories making it possible to store data at least temporarily, such as a RAM memory, a motherboard, supports, an operating system, for example .

Preferably, the unit 6 includes a display 11, for example a screen, making it possible to display the results of the methods of the invention.

It appears from the above, that the data processing unit 6 may or may not be part of the system 1 of the invention. In one embodiment, the data processing unit is integrated into the system of the invention, for example it is located inside the device 2. For example, the unit 6 is integrated into the microcontroller 5.

In a preferred embodiment, the data processing unit 6 is not part of the system 1, but is preferably required for the operation of the system. Unit 6 can be a computer, desktop computer, laptop, laptop, smartphone, smartwatch and / or tablet, for example. The unit 6 is preferably connected by wire or wirelessly to the device 2 or capable of being connected to the latter. In one embodiment, the device 2 comprises an interface 9, capable of establishing a connection with the unit 6, for example, via an interface 9 'of the latter. The connection can be wired or wireless, for example.

The fact of making use of an external unit 6 makes it possible to make the system 2 particularly economical, advantageously making use of a computer which a user of the system already has, the computers being omnipresent and in particular forming part of life. professional of a watchmaker, for example. In one embodiment, the invention thus makes it possible to present a system comprising an apparatus 2 consisting essentially of a measurement and / or management and / or signal transmission unit, and of the computer code 7 supplied, for example as than software.

In the embodiment shown in FIG. 1, the device 2 comprises a transmitter 12, for example a transmitter of visible or non-visible light such as ultraviolet or infrared lights, such as an LED lamp. In this embodiment, the sensor 3 is a sensor capable of detecting the light reflected by the watch or its movement 10, for example by the oscillator 15 of the movement 10. It should be emphasized at this stage that the movement 10 is shown at purely illustrative title in fig. 1, because in one embodiment, the apparatus of the invention is capable of determining the running of a watch on the basis of disturbances detected by a sensor during the passage of a hand of a watch, or the mark based on the movements of an exhaust element.

In the embodiment shown, the transmitter 12 and / or the sensor 3 are connected to the device 2, preferably by means of a support structure 13. In one embodiment, the position and / or the orientation of the sensor 3 can be adjusted, in order to optimize the capture of signals from the watch and / or the movement 10.

On the other hand, as mentioned above, the sensor 3 can be any sensor arranged to be able to distinguish disturbances during the passage of at least one distinctive element of an object of said timepiece, for example of the movement of a watch, said object performing a periodic movement and being chosen from the needle, the oscillator of a watch, and the escapement of a timepiece. Apart from an optical sensor, the sensor 3 can be a capacitive sensor or an electromagnetic sensor, for example. In a preferred embodiment, the sensor 3 is a proximity sensor.

CH 712 940 A2

The sensor 3 is preferably a one-dimensional sensor, making it possible to produce a value on a time signal. Preferably, the system of the invention uses only one such sensor, preferably a single sensor overall.

By way of illustration, the oscillator 15 of the movement 10 shown in FIG. 1 comprises a balance spring with a balance comprising three arms, 16, 16 ′ and 16.

The distinctive element of the object performing a periodic movement is preferably a part of the object making it possible to detect a distinctive signal disturbance. For example, said distinctive element is an element which, when it passes through the measuring field of the sensor, generates a signal disturbance which makes it possible, by means of signal processing, to obtain a distinct signal having a distinct amplitude lower or higher than the signal obtained outside said passage. In general, said distinctive element is at least a part generating an interruption or disturbance of a signal, for example an optical signal, following a change in geometry, size, type of surface or color of the material under the measuring range of the sensor. Examples of distinctive elements are the second hand of a watch, an arm of a balance-spring, the escapement, or part of one of these. In one embodiment, said distinctive element is the object performing a periodic movement.

In one embodiment, the system 1 is a system for implementing at least one of the methods of the invention, preferably all of the methods. Instead of system 1 as shown in fig. 1, the invention encompasses any other system making it possible to implement one or more methods of the invention. On the other hand, the invention also allows the methods of the invention to be implemented with other devices or systems.

In several embodiments, the methods of the invention use the processing of signals acquired by the sensor, the analysis, the management and / or the manipulation of these signals and preferably one or more algorithms and / or functions to determine one or more parameters of the timepiece.

Signal processing methods are in principle known and the present invention is not limited to a particular type of processing. As an example, mention is made of the treatments resulting in an increase ("enhancement") of the differences / contrasts in the signal, a frequency and / or time filtering in order to remove the unwanted noise.

In some embodiments, the invention relates to the use of cross-correlation and / or autocorrelation techniques in order to determine a parameter and / or to improve the accuracy of the value of a parameter. Preferably, the use uses the (cross) correlation in one dimension, ie on a time signal. The term correlation in this description preferably refers to mathematical correlation. To the knowledge of the inventors, the state of the art does not disclose the use of techniques of this type of correlation for the determination of parameters in mechanical watchmaking.

In one embodiment the system of the invention and / or the software 7 is configured to use the correlation of signals, for example autocorrelation, to determine a parameter or to improve the accuracy of a parameter chosen from : the walk, the amplitude of the oscillator, the moment in time (ϊμαχ) when the amplitude of an oscillator is maximum, the benchmark of a mechanical timepiece, and a combination of the above parameters.

In several embodiments of the methods of the invention, the correlation between a generated signal and a distinct real signal is used. For example, to determine a time difference, correlation can be used. To determine the moment t _M Ax of the oscillator, a correlation can be used to assist in the identification of centers of symmetry in the signal. To determine the benchmark, the correlation between one, preferably two generated signals and an actual signal can be used to determine when an element of the escapement changes position.

In one embodiment, the invention relates to a method for determining the progress of a timepiece. Walking is generally the watch's precision error, expressed in seconds, reduced to an interval of precisely 24 hours. The running value is generally negative if the watch is delayed and positive if the watch advances. The international standard ISO 3159: 2009 (F), incorporated by reference in the present description, defines the term "stopwatch" and allows the reader to know more about the daytime walking in particular.

The system of the invention is preferably configured to determine said step. In one embodiment, walking is daytime walking. The invention makes it possible to determine the gait on the basis of measurements carried out for less than 24 hours, more than 24 hours or exactly 24 hours. By way of example, as will appear on reading the description below, the invention makes it possible to determine the rate of operation on the basis of measurements carried out for one minute or more, for example in the case where the moving object is the second hand of a watch.

The method for determining the progress of a watch preferably includes the provision of a sensor arranged to be able to distinguish disturbances during the passage of a distinctive element of an object of a watch, for example the hand of a watch, the oscillator of a watch, or part of one of the two aforementioned objects. The invention also covers the possibility that the periodic movement of an element of the escapement is used to determine the running, or any other element of a timepiece movement performing a periodic movement.

The methods of the invention preferably include the acquisition of signals by said sensor. The sensor can preferably be chosen from the aforementioned one-dimensional sensors. Preferably, the method for determining the gait comprises the step of measuring a first real signal by said sensor 3 during the passage of said distinctive element and / or during the passage of said object.

CH 712 940 A2 [0049] FIG. 2 shows an actual signal 21 detected by the sensor. It can be the signal as detected, but preferably it is a processed signal.

The method preferably comprises the step of creating, on the basis of a real measured signal 21, a theoretical signal 21 'at a later defined time 27, said theoretical signal 21' substantially defining a signal which was produced, or would occur later, if the watch's running was zero.

In fig. 2, the reference number 26 indicates the time in time of the signal 21. In this case, the maximum of the peak 21 is chosen to determine the particular time in the time of the signal 21. By analogy, the moments 27 and 28 indicate the moments of peaks 21 'and 22, respectively.

Preferably, the method of the invention comprises the step of measuring at least one second real signal 22 by said sensor during the subsequent passage of said distinctive element. Said second real signal 22 can be a signal as detected, but preferably it is a processed signal.

In the embodiment shown in FIG. 2, the second signal 22 is associated with the moment 28 in time. The time 27 of the theoretical signal 21 'corresponds to the time when the second signal 22 would be expected if the running of the watch was zero.

Preferably, the method of the invention comprises the step of determining the difference in time between said second real signal 22 and said theoretical signal 21 '. In fig. 2, this difference corresponds to duration 25, ie duration 24 minus duration 23. As the person skilled in the art will understand, in the case of FIG. 2, the watch from which the actual signals 21 and 22 were acquired is delayed, since the second signal 22 appeared after the theoretical signal 21 ′, that is to say later than expected.

Preferably, the method of the invention comprises the step of determining the gait by reducing said difference to an interval of 24 hours. In this step, it is preferably a question of transposing the difference 25 to a duration of 24 hours. If, for example, the duration 23 is exactly 24 hours, the difference 25 corresponds to walking. To make another example, if the duration 23 was one minute (60 seconds), the difference 25 (in seconds) should be multiplied by 1440 (60x 24) to determine the gait.

In one embodiment, said subsequent defined moment 27 is the moment, counting from moment 26 of said first real signal 21, being at a theoretical or ideal period T _TH e of said moving object or approximately at a multiple of this ideal period T _TH e · If at each repetition the object completed its movement exactly at the end of an ideal period T _TH e, the running of the watch would be zero. In the embodiment shown in FIG. 2, the duration 23 between moments 26 and 27 can be a period T _T he, a multiple of the latter or can be chosen arbitrarily. In the latter case, the subsequent defined moment 27 does not necessarily correspond to a period or a multiple of the period. The walk can be determined if the duration 23 is known (defined by the system) and the durations 24 and 25 (fig. 2) are determined during the implementation of the method, even if the duration 23 is not a period or a multiple of the latter. To improve the accuracy and / or reduce the calculation time, the moment 27 is preferably chosen close to the expected signal, which is the case if one chooses a period or a multiple of the latter (from moment 26 ) to choose the moment 27 of the theoretical signal 21 '. On the other hand, the more distant the first real signal 21 and the second signal 22, the more precise the determination of the gait will be.

In one embodiment, said step of determining said time difference comprises: performing a correlation between said second real signal 22 and said theoretical signal 21 '; and, defining said difference 25 as the travel time between said two signals where the correlation between said two signals is maximum. Preferably, the correlation is a cross-correlation and / or a one-dimensional correlation, on a time signal. Preferably, a correlation is made between the theoretical signal 21 'and the second real signal 22 at different times. In other words, the similarity between a theoretical signal and the second real signal is determined as a function of the time separating the signals. The phase shift or deviation time allowing a maximum correlation between the theoretical signal and said second real signal is preferably chosen as said difference between the two signals.

Instead of correlating said second real signal 22 and said theoretical signal 21 ', the latter can also be made with the first real signal 21, and the difference can then be determined with the reference to the second signal real 22. In other words, the terms "first" and "second" do not necessarily indicate the order in time of the succession of signals. According to the invention, preferably at least two real signals and at least one theoretical signal are used, and the order and / or the sequence of the signals can preferably be chosen so as to make the method more effective, but preferably does not constitute a limiting element of the concept of the invention.

In one embodiment, said step of determining said time difference comprises: moving in time at least one of the two signals chosen from said subsequent real signal 22, said theoretical signal 21 ', and both, one relative to each other so as to approximate said signals; defining said difference as the travel time between said two signals where the correlation between said two signals is maximum.

Said correlation preferably comprises a comparison by correlation of two signals in one dimension (1D), for example, the light intensity detected as a function of time.

In one embodiment of the method of the invention, the rate M (in seconds) is determined on the basis of the formula (I):

CH 712 940 A2

Μ = (Tthe-Îmes) * (86 400 / T _T he) (I) t _M Es is the time in seconds between the first real signal and the second real signal,

T _T he is the time in seconds between the first real signal and the theoretical signal, and, (Tthe- tMEs) being said difference.

The number 86,400 corresponds to the number of seconds per day (60 x 60 x 24). In the example shown in fig. 2, tMEs, © he, and (îthe- îmes), correspond to the reference numbers 24, is 23 and 25, respectively.

In one embodiment of the method for determining the gait, the object performing a periodic movement is a watch hand or the oscillator of a mechanical timepiece. The invention therefore makes it possible to determine the gait by receiving signals from one of the two aforementioned objects. In one embodiment, the object is the second hand and / or the minute hand, preferably the second hand.

In one embodiment, said object is the second hand and the step M (seconds per day) and determined on the basis of the formula (la):

M = ((k * 60) - t _M Es) * 86 400 / (k * 60) (la) in which k represents the number of minutes between the first real signal and said theoretical signal, and, t _M Es and is the time (in seconds) between the first real signal and the second real signal.

The formula (la) is a special case of the formula (I), in which Fhe = k * 60 seconds. Preferably, k is a natural whole number.

It is possible to use the step M to determine the frequency and / or the period of a system performing a periodic movement, for example of the oscillator of a mechanical timepiece. Formula (I) can also be written as follows (Ib):

M = (T _T he-T _M es) * (86,400 / T _TH e) (Ib) in which T _TH e and T _M es are the theoretical period and the measured period (actual or real) of an oscillation. If the durations î _M es and î _M es are determined, for example according to the method of the invention, we can deduce î _M es and thus the frequency (Îmes = 1 / T _M es) of the oscillator, T _T he being known.

As those skilled in the art will understand, the system and / or the method of the invention make it possible to determine instantaneous walking (M | _NS t) and / or daytime walking (M _D |). The path can thus be determined, at the user's choice, in an observation interval of less than 24 hours (M | _NST ), for example in a few seconds when the object under observation is the oscillator and in a few minutes when the object under observation is the second hand.

It may be noted that, when the object performing a periodic movement is a watch hand, the system and the method of the invention make it possible to determine the running of a non-mechanical watch, for example of an electric watch.

In one embodiment of the method for determining the walking, the object performing a periodic movement is the oscillator, and the walking is determined using signals resulting from the movement of the oscillator, for example, of a balance-spring oscillator 15 (fig. 1). In this case, the distinctive element can be a 16-16 arm or a balance weight. In this case also, the method makes it possible to determine the frequency of the oscillator.

In one embodiment, the invention relates to a method for determining î _M ax of an oscillator, for example of a balance spring, ie a moment in time when the amplitude of an oscillator a watch movement is maximum. The moment î _M ax also corresponds to the moment when the oscillator turns back and / or changes direction.

The method for determining î _M ax preferably comprises the provision of a non-acoustic sensor 3 arranged to be able to distinguish disturbances during the passage of at least two distinctive elements of said oscillator during the oscillation of this latest. For example, the two distinctive elements are two different arms of the balance spring, or two different weights, or parts of these.

The method preferably comprises a step of acquiring signals while the oscillator is in oscillation, preferably the measurement of successive signals.

The method preferably comprises digital processing of the measured signals, so as to obtain separate successive signals having a distinctly lower or greater amplitude than the signal obtained outside said passages. Exemplary methods of signal processing have been cited above. In a preferred embodiment, the measured signals are processed so as to obtain distinct signals each having a point of maximum amplitude, ie a "peak" proper, as illustrated in FIG. 3. In order to avoid any misunderstanding, it is clarified that the moment

CH 712 940 A2 in the time when the amplitude of a signal is maximum, used to determine the moment of the signal concerned, should not be confused with the amplitude of the oscillator which the method seeks to determine according to an aspect of l invention, discussed in more detail later below.

The method for determining î _M ax preferably comprises the step of extracting a sequence of signals obtained for a duration equal to or longer than said period, preferably equal to or greater than a period and a half, for example during two periods.

In order to be able to extract a sequence acquired during a period expressed relative to the period of the oscillation, it may be necessary to know the frequency F _M es and / or the period T _M es of the oscillation. Instead of F _M es- Instead of F _M es, H is also possible to implement the method for determining î _M ax using the theoretical frequency of the oscillator.

According to one embodiment, the method for determining î _M ax, comprises determining the frequency and / or the effective period of said oscillator, preferably using the step determined according to the method of the invention.

Walking can be determined as described above using instant walking (for example, the formula (lb) above). Preferably, the walking is determined based on the disturbances created by distinctive elements of the oscillator. The method for determining _M ax may include the step of determining the frequency, or the value of the frequency may be taken from subsequent measurements, for example in the context of determining the running of the same watch. As indicated, it would also be possible, but less advantageous, to refer to the theoretical frequency. Preferably, the method for determining max comprises determining the frequency of the oscillator.

The method for determining î _M ax preferably comprises a step of determining at least one center of symmetry in said sequence. The symmetries in said sequence can be determined by techniques of analysis, processing and / or manipulation of signals.

Once the centers of symmetry in the distinct signals have been determined, the method of the invention preferably determines î _M ax as the moment in time corresponding to said at least one center of symmetry.

The method for determining îmax is illustrated in FIG. 3A. To illustrate the oscillatory movement, fig. 3B shows a sinusoidal function. Following the acquisition of successive signals and the processing of these, one obtains for example a succession 30 of distinctive signals and / or peaks 31 to 38 (FIG. 3A). The peaks 31-38 are the consequence of the passage of two distinctive elements of the oscillator during the oscillation of the latter, for example during the passage of two arms 16, 16 '(fig. 1) of a balance spring , said passage having been detected by the sensor 3. In the example shown in FIG. 3A, the third arm 16 does not pass through the detection field of the sensor 3, and this arm 16 does not cause a disturbance of the signal received.

Having knowledge of the frequency, for example after having determined it in accordance with the invention, an extract 40 of distinct signals can be chosen arbitrarily for the rest of the method. The extract contains a sequence of successive distinct signals obtained during a duration equal to or longer than said period. The place of the extract in the signal being arbitrary, the duration covered by the extract is chosen deliberately. In fig. 3, the duration 42 corresponds to a period of the oscillation and the extract 40 covers a duration 41 of a period and a half. It may be preferable to extract a longer duration, for example between 1.5 and 2 periods, in order to be sure that at least two points of symmetry are in the extract. In one embodiment, the extract 40 is chosen so as to contain at least two points î _M ax. In one embodiment, the extract comprises 1 to 2 periods, preferably 1.1 to 1.9 periods, more preferably 1.2 to 1.8 periods.

Following the analysis of the separate signals 32-37 of the extract 40, centers of symmetry 45 and 44 'are identified. The fact that these two centers are spaced apart by a half-period 43 indicates that the centers of symmetries have been correctly identified. One of the two centers of symmetries 45, 44 ′ is considered a moment î _M ax and the other is also a î _M ax, that is to say a moment when the amplitude has a maximum negative value. In the case of fig. 3, points 44, 44 'are moments t _M Ax- Point 45 is also considered to be a moment î _M ax for the present description, even if it is distinguished from moments 44 and 44' in that one could also consider a moment t _M iN- Points 44, 44 'and 45 define centers of symmetry.

From the points î _M ax (and / or ϊ _Μ ιν) it is also possible to find the point t ₀ , ie the point of zero elongation (rest position). This point 46 is located exactly in the middle between two centers of symmetry 44 and 45, or 45, 44 '. As will be described below below, depending on the mathematical approach chosen to determine the amplitude of the oscillator, it is possible to use îmax and / or t ₀ , and the present invention also makes it possible to determine the amplitude without know neither îmax, nor t ₀ .

In one embodiment, the step of determining at least one center of symmetry in said sequence comprises the steps of determining a plurality of centers of symmetry assumed between pairs of signals distinct from said sequence of signals; determining a measure of symmetry for each of the assumed centers of symmetry; and, determining t _M Ax as the center of symmetry for which said measure of symmetry indicates the most likely symmetry.

With reference to FIG. 3, the method preferably comprises the step of creating or assuming centers of symmetries between pairs of peaks, for example between peaks 32 and 33, 33 and 34, 34 and 35, etc. and, assuming it

CH 712 940 A2 could be a true center of symmetry, determine a measure of symmetry for each supposed center of symmetry. Preferably, all the peaks of the extract are combined in pairs and an assumed center of symmetry is determined for each pair of peaks of the extract. In an extract containing n peaks, this generally results in a combination of two centers of supposed symmetry. For each center of symmetry, a measure of symmetry is determined, and the center whose measurement indicates the most probable symmetry is considered to be the true center of symmetry, ie a moment t _M Ax. FIG. 4A shows a possibility for determining a supposed center of symmetry 56 between two distinct signals 33, 34 of the extract 40. By way of illustration, in FIGS. 4A-4D, the two peaks come from the passage of two distinct elements of the oscillator, which is why there is no t _M Ax between the two peaks 33 and 34, ie the supposed center of symmetry 56 is not a true center of symmetry. In fig. 4A, the supposed center of symmetry 56 is positioned in the middle between the points 53, 54 on the time axis, these halves indicating the moments when signals 33 and 34, respectively, have their maximum.

In one embodiment, the step of determining at least one center of symmetry in said sequence comprises the steps: determining a supposed center of symmetry between two distinct signals of said sequence 40 of signals, a first distinct signal and a second separate signal; generating a theoretical signal 33 'by inverting one of the two distinct signals and reproducing and / or reflecting it on the other side of the supposed center of symmetry 56; and; determining a measure of symmetry as a function of a similarity between said theoretical signal and the other of said two distinct signals.

FIG. 4B shows the two processed signals 33 and 34, as well as the theoretical signal 33 ', which is a reflection of the signal 33, the line 56 indicating the supposed center of symmetry serving as the axis of symmetry. The step of determining a measure of symmetry preferably includes comparing the distinct signal 34 with the theoretical signal 33 'and calculating a value of symmetry which expresses the degree of similarity between the two signals. This measure of symmetry can take into account several factors, for example the point in time of the maximum amplitude of the two signals, the area of the area under the curve, the inclinations on either side of the maxima of the two signals, for example. In the case shown in fig. 4B, this measure of symmetry gives a result which indicates that the symmetry at point 56 is comparatively unlikely, because in this case, the center of symmetry assumes 56 is not a true center of symmetry and / or not a point t _M Ax The same method, applied to signals 33 and 36 in fig. 3, gives a measure of symmetry whose value indicates a comparatively more likely point 45 symmetry.

Preferably, the method uses the means of correlation to determine î _M ax, preferably to determine and / or improve said measure of symmetry. Preferably, the correlation is a cross-correlation and / or a one-dimensional correlation, on a time signal. In one embodiment, the step of determining at least one center of symmetry in said comprises the step of performing a correlation between said theoretical signal 33 'and the other of said two distinct signals 34; and determining or, if necessary, modifying the position of a supposed center of symmetry so that said correlation is maximum.

FIGS. 4C and 4D illustrate the correlation used in this case to improve the supposed center of symmetry 56 so that said symmetry measurement indicates a value showing a center of symmetry 56 'more likely than the initial supposed center of symmetry 56. In this embodiment, the theoretical signal 33 'is moved until the symmetry measurement indicates a maximum correspondence between the signal 34 and the signal 33'. In fig. 4C, the signal 33 ′ has been phase shifted at the position indicated by the signal 33. In this embodiment, the measurement of similarity between the two signals 34 and 33 is higher than the measurement obtained by comparing the signals 34 and 33 ', the latter being the result of reflection at the supposed center of symmetry 56. Because of this result, the supposed center of symmetry 56 is moved to position 56' as illustrated in fig. 4D, this last position indicating a center of symmetry supposed to be improved is a center of symmetry supposed more probable than the center of symmetry assumed initial 56.

During the final determination of the centers of symmetry, apart from the measurement of similarity of the peaks on either side of an assumed center of symmetry, the method can take account of the value of the measurement of symmetry determined at half a period in front and / or behind of a supposed center of symmetry, because the real moments t _M Ax (or t _M Ax and î _M in) are spaced by half a period. If a higher measure of symmetry is also found half a period ahead and / or behind a supposed center of symmetry, this increases the probability that it is a true center of symmetry and / or moment t _M Ax- In one embodiment, the method of the invention provides for weighting the maximum correlation value by the presence or not of another point of symmetry at a half-period forward and / or backward of this point. This is preferably done once the set of signals distinct from the extract have been considered in pairs.

In one embodiment, the present invention relates to a method for determining the amplitude of an oscillator of a timepiece with mechanical movement on the basis of non-acoustic measurements, preferably on the basis of measurements non-acoustic only.

The method preferably makes it possible to determine the amplitude without using signals from an acoustic sensor and / or without analysis of the shocks coming from the exhaust. Surprisingly, the present invention operates without it being necessary to determine the value of t ₀ on the basis of acoustic information originating from the movement, in particular from its escapement. The devices of the state of the art use, unlike the present invention, generally an acoustic sensor and / or acoustic information in order to be able to determine the moment when the oscillator passes through its rest position.

CH 712 940 A2 In one embodiment, the method includes the provision of a non-acoustic sensor arranged to be able to distinguish disturbances during the passage of at least two distinctive elements of said oscillator during oscillation of the last. The sensor can be chosen from the sensors specified above. Preferably, the same sensor 3 is used. Preferably, at least two distinctive elements of said oscillator are observed, for example at least two arms 16, 16 ′ of a spiral balance 15 (fig. 1).

In one embodiment, the method comprises the steps of measuring successive disturbances originating from at least two distinctive elements of said oscillator and the digital processing of the disturbances measured during the passage of the at least two distinctive elements, so as to obtain a separate signal for each passage of one of said at least two distinctive elements. These steps can in principle be carried out as described above with respect to the determination of t _M Ax- For example, in FIG. 3A, the distinct signals 33 and 34 are the result of the passage of two different distinct elements, for example the two arms 16, 16 '.

In one embodiment, the method of the invention comprises the step of determining the times q and t ₂ by assigning a moment in time to two distinct signals associated with the successive passage of a first and a second of said at least two distinctive elements. Basically, these moments in time of ti and t ₂ generally not in relation to the oscillatory movement, but, for example, determined with respect to the start of the measurement or with respect to an arbitrary reference moment generated by the system. Taking fig. 3A, a moment in time L can be attributed to signal 33 and another moment in time t ₂ is attributed to signal 34. In one embodiment, the moments q and t ₂ are attributed to the moments when the amplitude signals 33 and 34 is maximum. In one embodiment, the invention can also take account of a moment t _M Ax, determined according to the method described above, to determine the moments L and t ₂ . For example, ti and t ₂ can be chosen so as to be distinct from the moment t _M Ax, and / or can be defined by taking account of the symmetrical patterns in the signal.

In one embodiment, the method of the invention comprises the step of determining the amplitude (A _M ax) of the oscillator by using the properties of the signal at times L and t ₂ and by solving the function sinusoidal movement according to formula (V):

A (t) = A _M ax * sin (ω * t + φ) (V) in which ω = 2 * π * f, f being the measured frequency the oscillator and φ is the phase shift.

In formula (V), ω is known, as the frequency of the oscillator was determined as described above, for example.

The inventors of the present invention have developed several approaches which make it possible to solve the formula (V). On the basis of the information available to the inventors, the state of the art does not disclose a way to resolve the formula (V) for determining the amplitude A _M ax on the basis of a signal from a sensor not -acoustic. In the prior art, an acoustic sensor is generally used to detect the moment t ₀ and thus determine the moments L and t ₂ with respect to t ₀ . This would determine the phase shift φ. It should be noted that when acquiring signals as shown in fig. 3A, one cannot know the moment when the pendulum passes through its rest position.

In one embodiment of the determination of the amplitude (A _M ax)> the function according to the formula (V) is solved using the following equations of motion (VJ to (XI):

φ = π / 2 - ω * îmax (VI)

A (t ₂ ) -A (ti) = known, i.e. the angle between the two distinctive elements (VII)

A (t2) - A (ti) = Amax [sin (ω * t2 + φ) - sin (ω * ti + φ)] (Vili)

We use the trigonometric identity according to which:

sin (A) -sin (B) = 2 * cos [(A + B) / 2] * sin [(A-B) / 2] (IX) therefore:

A (t2) - A (ti) = 2 * Amax * cos [(to * (t2 + ti) + 2 * <p) / 2] * sin [(œ * (t2-ti)) / 2] (X ) [0101] It is advisable to replace φ according to (VJ in the formulas (X) in order to solve the system and to determine Amax [0102] Preferably, in the abovementioned embodiment, the function according to the formula (V) is resolved without go through the point of zero amplitude (t ₀ ). According to this approach, we do not give a concrete value at each of the moments t ₀ , L and t ₂ , but can solve the formula (V) using the information of relative times (L - îmax; t ₂ - îmax), which can be obtained without determining t ₀ .

CH 712 940 A2 [0103] In another embodiment of the determination of the amplitude (A _M ax), the function according to the formula (V) is solved by the steps of determining the moment t ₀ , ie the moment when the amplitude A (t) is zero, by defining t ₀ , as the time lying in the middle of two points îmax determined in accordance with the method of the invention for determining t _M Ax; and to deduce the moments ti, t ₂ and / or φ from the moment of t ₀ , and determine A _M ax without îmax preferably using equation (X). In this embodiment, the value of t ₀ is determined, for example as described below, preferably without resorting to acoustic information originating from the exhaust. The values of L, t ₂ les can then be expressed with respect to t ₀ (in relation to the oscillatory movement) and / or φ can be determined. To make a concrete example and without wishing to be limited, this embodiment would make it possible to set tb to 0 (zero) and to determine values of L, and t ₂ with respect to a moment 0.

In yet another embodiment of the determination of the amplitude (A _M ax), the function according to formula (V) is solved using the following functional relationships:

- the derivative of the equation (V) for t = L, corresponding to the speed (VJ of the first distinct element at (XX) moment L, V-ι = cos (co * Li + φ)

- the derivative of equation (V) for t = t ₂ , corresponding to the speed (V ₂ ) of the second distinct element at (XXI) moment t ₂ , V ₂ = cos (co * t ₂ + φ) the method comprising:

- determine the ratio Rv between the width of the signals of L and t ₂ , Rv = width of the signal 2 / width of the signal 1, and assume the correspondence between the width of the signals and the speeds of the at least two distinctive elements, so that

V ₂ / Vi = (signal width 1 / signal width 2) = Rv (XXII)

- using the value for Rv obtained from the actual distinct signal (signal width 1 / signal width 2), solve the system of equations (XX), (XXI) and (XXII); and, deduce A _M ax, for example using the value φ obtained in the previous step.

With regard to the formula (XIX), it should be noted that the speeds V ₂ , V-ι are inversely proportional to the width of their respective distinct signal. In summary, this embodiment exploits the information contained in the width of a signal associated with the passage of an arm 16-16 '. A wider signal / peak indicates that the separate element passes more slowly under the sensor, staying longer in the detection range of the latter, which is why the peak resulting from the passage is wider and / or larger. In the case of this embodiment, it is preferable that the at least two distinct elements are two elements having the same dimensions, preferably the same width, which is the case for example if the at least two distinct elements are arms separate 16-16 'from the oscillator.

In one aspect, the present invention relates to a method for determining the mark of a mechanical movement of a timepiece. The coordinate system is a measure of geometric alignment between the zero-stretch position of the balance wheel and the balance-anchor axis. Due to this alignment, the rotary oscillation is not entirely symmetrical around the point of rest, that is to say, the pendulum oscillates further in one direction than in the opposite. The benchmark is generally measured in milliseconds (ms). The benchmark can, for example be expressed by the following formula:

Mark = (t _E1 - t _E2 ) / 2 (XXX) in which t _E1 in is the time of oscillation on the one hand from the rest position and t _E1 is the time of oscillation on the other hand from the rest position.

In one embodiment, the method for determining the benchmark includes the provision of a non-acoustic sensor arranged to be able to distinguish disturbances during the change of position of an element of the exhaust of the movement of a first stop position to a second stop position. The sensor can be chosen from the same sensors described above, for example the sensor 3 shown in FIG. 1.

In the case of determining the benchmark, the sensor is arranged and / or adjusted to detect a disturbance when the position of an element of the movement exhaust is changed from a first stop position to a second stop position. Typically, the exhaust moves back and forth between two stop positions. The change between the two positions generally takes place when the oscillator goes to the point of zero elongation. Preferably, measuring the benchmark by non-acoustic means requires access to the exhaust and is generally not possible only by observing the pendulum. This applies in particular if an optical sensor is used, in accordance with a preferred embodiment of the method for determining the mark.

CH 712 940 A2 [0109] For example, in the case of an anchor type escapement, the movement of the anchor pivoting is limited by two stops which define, together with the pivot axis and the geometry of the anchor, the two stop positions.

In one embodiment of the method for determining the mark, the entire movement back and forth from the exhaust is captured by the sensor, that is to say occurs under the detection field of the sensor . Preferably, we do not observe the passage only of a separate element at current rates, but we preferably receive signals representing the whole of the duration in which an element distinct from the exhaust is in the two positions. This generally does not require that the same object be entirely under the sensor at all times, but that at least one separate movable element describing the same movement as the exhaust is preferably permanently visible. This is generally the case. For example, most movements have holes used to observe the movement of the anchor and / or part of it. It is also possible to observe elements of the escapement through the movement.

In one embodiment of the method for determining the benchmark, said escapement is chosen from: an anchor escapement, preferably of the "Swiss anchor" type, and a coaxial escapement, as described, for example, in EP 0 018 796. In this type of exhaust, the anchor is replaced by a kind of lever. More generally, the invention can be applied to any exhaust having two separate positions, which alternate during the passage of the oscillator at the point of zero elongation.

In one embodiment of the method for determining the coordinate system, said element of the exhaust is one of the vanes of the exhaust, for example of the anchor and / or of the coaxial exhaust. The paddles are preferably anchor stones that interact with the escape wheel. There are usually two paddles in an anchor escapement, for example.

In one embodiment, the method for determining the benchmark comprises the step of acquiring successive signals by said sensor of said element of the exhaust. These successive signals are obtained when the oscillator is oscillating.

[0114] FIG. 5 shows disturbances and / or raw signals captured by an optical sensor positioned above a pallet of the exhaust. In these signals, one can clearly distinguish the two plates 61, 62, corresponding to the two positions of the observed element. The reference number 63 indicates the transition between the two positions. Fig. 5 also shows signals 64 linked to the passage of the arms of the balance wheel. This indicates that the sensor also detects the passage of the arms, in addition to the positions of the exhaust. It would be possible to combine the measures described above, for example with respect to the determination of t _M Ax, with the measures of the reference frame. At this stage, the present description is limited to the determination of the benchmark on the basis of the acquired signals.

In one embodiment, the method for determining the benchmark comprises the digital processing of the measured signals, so as to obtain a processed signal containing distinct successive signals which can be associated with said two positions of said element of the escapement.

[0116] FIG. 6 shows a possible representation of the ideal theoretical signal corresponding to these transitions. This signal represents what we are looking for in order to be able to determine L and t ₂ in the case of the reference frame.

In one embodiment, the method for determining the benchmark comprises a step of determining time durations d! and d ₂ as the durations of two separate successive signals. In fig. 6, the durations L and t ₂ correspond to the audited durations d-ι and d ₂ . The method thus makes it possible to determine the reference frame of the movement as being (d ₂ -di) / 2 or (di-d ₂ ) / 2 [0118] It should be noted that, according to a preferred embodiment, the durations of the transitions between the two positions, a transition being indicated by the reference number 63 in FIG. 5, are ignored. This is possible because the transitions are assumed to be symmetrical by their time, or when the difference between the two times is negligible.

According to one embodiment, the method for determining the benchmark comprises the step of averaging individual benchmark measurements. For example, the time d-ι can be determined by averaging several durations d-ι, d ₂ times representing the time during which the distinct element is in the first position, and / or by averaging d ₂ analogically. Or the benchmarks determined successively may be average. Generally, the arithmetic means is used.

In one embodiment, said step of determining time durations d! and d ₂ of two separate successive signals comprises the steps of performing a correlation to increase the precision of the durations di and d ₂ and / or to increase the precision of the moments in time Τ _Ί , T ₂ , T ₃ , T _N , changes between the two positions.

As can be seen in FIG. 5, determining the durations di, d ₂ etc. is susceptible to noise. The disturbances measured by the sensor may even be less clear than those shown in fig. 5. To determine the benchmark more precisely, the present invention contemplates the use of the correlation to determine the durations di and d ₂ . Preferably, the correlation is a cross-correlation and / or a one-dimensional correlation, on a time signal.

In one embodiment, the method for determining the benchmark comprises: generating two partial signals, a first partial signal and a second partial signal, like parts of a processed signal; said first partial signal (SP) being generated in the image of a signal part covering at least part of the transition of said element from a first

CH 712 940 A2 at a second position; and, said second partial signal (SN) being generated in the image of a signal part covering at least part of the transition of said element from a second to a first position; correlate each of the partial signals with the signal, preferably with the original signal after filtering, and determine the moments T, T ₂ , T ₃ , T _N , such as the moments when the correlations are at their maximum and / or minimum.

It is clear that the word "transition" in the expression "said first partial signal (SP) being generated in the image of a signal part covering at least part of the transition of said element from a first at a second position ”refers to the signal measured during the transition between the two positions. The signal part can be generated on the basis of a processed signal and / or on the basis of a theoretical signal.

The generated partial signal can also be considered as a theoretical partial signal, since it is generated on the basis of a theoretical ideal signal, like the one shown in fig. 5.

In one embodiment, said partial signal generated comprises a part of a low half-wave followed by a part of a high half-wave, or vice versa.

Preferably, said signal part covers the entire transition of said element from a first to a second position.

In one embodiment, said partial signal generated comprises a segment representing the change between the two positions of the exhaust, said segment being vertical or inclined by ± 60 ° or less, preferably not more than ± 45 ° and more preferably ± 20 ° or less from the vertical. For example, said segment can be tilted> 0 to 60 ° from the vertical, in one as in the other direction. This segment preferably represents the processed signal indicating the change and / or transition between the two positions of the exhaust. The segment is preferably a segment of a straight line.

In a preferred embodiment, the two partial signals (SP), (SN) are symmetrical with respect to each other, and or at least said aforementioned segment is symmetrical in the two partial signals. For example, if the aforementioned segment of one of the two signals is inclined by, for example, + 45 °, with respect to the vertical, the segment of the other of the two signals is preferably inclined in the opposite direction, -45 ° according to this example.

In one embodiment, said partial signal generated is chosen from signals having the general appearance according to (a) and / or (b):

(a) (b) With respect to the above embodiment, the signal (a) can be considered as said first partial signal (SP) and / or the signal (b) as said second partial signal (SN) ), or the opposite. In this embodiment, the segment representing the change between the two positions of the exhaust is the vertical segment.

The horizontal elements and / or segments in a partial signal (a) or (b) correspond to part of the low half-wave and part of the high half-wave, or the reverse, for example of the theoretical signal shown in fig. 5.

In one embodiment, each of the two partial signals (SP), (SN) comprises two essentially horizontal elements and / or segments.

[0133] FIG. 7 shows the correlations of the partial signals (SN, SP) with the original filtered signal. The maximums of the correlation indicate the moments of the change in position, and therefore the moments Tî, T ₂ , T ₃ . For example, T-ι indicates the time when the duration dì (L) in fig. 6, and T ₂ the moment of the start of the duration dz (T ₂ ), and T ₃ the end of the duration d ₂ , and so on.

From what proceeds, it appears that the present invention makes it possible to determine the characteristics of the running of a watch, the moment in time (t _M Ax) where the amplitude of an oscillator of a movement of watch is maximum, the amplitude and the reference of a movement of a mechanical watch. The system uses methods of processing and analyzing signals measured by the one-dimensional sensor during the passage of an element performing a periodic movement, such as the watch hand, the oscillator or the escapement.

Claims (15)

claims 1. A system (1) for determining one or more parameters of a watch movement, said parameter being chosen from: (i) the running of a watch; (ii) the amplitude of the oscillator of the movement; (iii) the movement marker; and (iv) the moment t _M Ax of said oscillator, ie the moment in time when the amplitude of an oscillator of a watch movement is maximum for the current oscillation, said system (1) comprising: CH 712 940 A2 - an apparatus (2) comprising: - a non-acoustic sensor (3) arranged to be able to distinguish disturbances during the passage of at least one distinctive element of an object of said watch, said object performing a periodic movement and being chosen from the hand of a watch , the oscillator of a watch, the escapement and or part of one of the aforementioned objects; - a support (4), arranged to temporarily place a watch or a watch movement relating to said sensor (3) in order to allow the measurement of the above-mentioned disturbances; - a microcontroller (5), arranged to be able to receive signals from said sensor (3), said microcontroller comprising or being associated with a time base (8); and - at least one computer code (7) configured to be able to implement the method according to any one of claims from 3 to 17. 2. The system (1) of claim 1, in which said non-acoustic sensor (3) is a one-dimensional sensor, chosen from an optical sensor, a capacitive sensor, and an electromagnetic sensor, preferably an optical sensor chosen from sensors arranged to pick up a light reflected or interrupted by said distinctive element. 3. A method for determining the performance of a watch, the method comprising the steps of: - the provision of a non-acoustic sensor (3) arranged to be able to distinguish disturbances during the passage of a distinctive element of an object of a watch, said object performing a periodic movement and being chosen from among watch hand, the oscillator of a watch and / or part of one of the two aforementioned objects; - measuring a first real signal (21) by said sensor (3) during the passage of said distinctive element; - creating on the basis of a real measured signal a theoretical signal (21 ') at a later defined time, said theoretical signal (21') substantially defining a signal which had occurred, or which would occur later, if walking of the watch was void; - measuring at least a second real signal (22) by said sensor during the subsequent passage of said distinctive element; - determine the difference in time (25) between said second real signal (22) and said theoretical signal (21 '); and, - determine the gait by reducing said difference to an interval of 24 hours. 4. The method according to claim 3, characterized in that said step of determining said time difference comprises: - performing a correlation between said second real signal (22) and said theoretical signal (21 '); and, - Define said difference (25) as the travel time between said two signals where the correlation between said two signals is maximum. 5. The method according to any one of claims 3 to 5, characterized in that said object is the hand of a watch. 6. A method for determining a moment in time (t _M Ax) when the amplitude of an oscillator of a watch movement is maximum, the method comprising the steps of: - the provision of a non-acoustic sensor (3) arranged to be able to distinguish disturbances during the passage of at least two distinctive elements of said oscillator during the oscillation of the latter; - measure successive signals during said oscillation; - digital processing of the measured signals, so as to obtain separate successive signals having a distinctly lower or greater amplitude than the signal obtained outside said passages; - optionally: determine the frequency and / or the effective period of said oscillator, preferably using the rate determined according to the method one of claims 3 and 4; - extract a sequence of signals obtained during a duration equal to or longer than said period; - determining at least one center of symmetry in said sequence; - determine t _M Ax as the moment in time corresponding to said at least one center of symmetry. 7. The method according to claim 6, in which the step of determining at least one center of symmetry in said sequence comprises the steps of: - determining a supposed center of symmetry (56) between two distinct signals (33) of said sequence of signals (40), a first distinct signal and a second distinct signal (33, 34); - generate a theoretical signal (33 ') by inverting one of the two distinct signals (33) and reproducing and / or reflecting it on the other side of the supposed center of symmetry (56); - determining a measure of symmetry as a function of a similarity between said theoretical signal (33 ') and the other of said separate signals (34); - perform a correlation between said theoretical signal (33 ') and the other of said two separate signals (34); and, - determine or, if necessary, modify, the position of a supposed center of symmetry so that said correlation is maximum. 8. A method for determining the amplitude of an oscillator of a wristwatch on the basis of non-acoustic measurements only, the method comprising the steps of: CH 712 940 A2 - the provision of a non-acoustic sensor (3) arranged to be able to distinguish disturbances during the passage of at least two distinctive elements of said oscillator during the oscillation of the latter; - measure successive disturbances from at least two distinctive elements of said oscillator; - digital processing of the disturbances measured during the passage of said at least two distinctive elements, so as to obtain a separate signal for each passage of one of said at least two distinctive elements; - Determine time values t-ι and t ₂ by assigning a time in time to two distinct signals associated with the successive passage of a first and a second of said at least two distinctive elements; - determine the amplitude (Amax) of the oscillator using the properties of the signal at times t-ι and t ₂ and by solving the function of the sinusoidal movement according to formula (V): A (t) = Amax * sin (co * t + φ) (V) in which ω = 2 * π * f, f being the frequency measured by the oscillator; φ is the phase shift. 9. The method according to claim 10, in which the function according to formula (V) is solved using the following equations of motion: φ = π / 2 - ω * îmax (VI) A (t2) -A (ti) = known, i.e. the angle between the two discrete elements (VII) A (t2) - A (ti) = Amax [sin (ω * t2 + φ) - sin (ω * ti + φ)] (Vili) trigonometric identity according to which sin (A) -sin (B) = 2 * cos [(A + B) / 2] * sin [(AB) / 2] (IX) therefore: A (t ₂ ) - A (ti) = 2 * Amax * cos [(Cû * (t2 + ti) + 2 * <p) / 2] * sin [(cû * (t ₂ -ti)) / 2] (X) replace φ according to (VI) in (X) in order to solve the system. 10. The method according to claim 8, in which the function according to formula (V) is solved by the steps of: - determine the moment fc, that is to say the moment when the amplitude A (t) is zero, by defining t ₀ , as the time being in the middle of two points t _M Ax determined in accordance with any one of claims 8 to 11 ; - deduce the moments L, t ₂ and / or φ from the moment of t ₀ , and determine A _M ax preferably without t _M Ax, preferably using equation (X). 11. The method according to claim 8, in which the function according to formula (V) is solved using the following functional relationships: - the derivative of equation (V) for t = q, corresponding to the speed (VJ of the first distinct element at (XX) moment ν _Ί = cos (co * ti + φ) - the derivative of equation (V) for t = t ₂ , corresponding to the speed (V ₂ ) of the second distinct element at (XXI) moment V ₂ = cos (co * t ₂ + φ) the method comprising: - determine the ratio Rv between the width of the signals of L and t ₂ , Rv = width of the signal 2 / width of the signal 1, and assume the correspondence between the width of the signals and the speeds of the at least two distinctive elements, so that V ₂ A /! = (signal width 1 / signal width 2) = Rv (XXII) - using the value for Rv obtained from the actual distinct signal (signal width 1 / signal width 2), solve the system of equations (XX), (XXI) and (XXII); - deduce A _M ax, for example using the value φ obtained in the previous step. 12. A method for determining the benchmark of a watch movement, the method comprising the steps of: - the provision of a non-acoustic sensor (3) arranged to be able to distinguish disturbances during the change of position of an element of the escapement of the movement from a first stop position to a second position stopping, said change taking place due to the passage of the oscillator at the point of zero elongation; - acquire successive signals by said sensor of said element of the exhaust; CH 712 940 A2 - digital processing of the measured signals, so as to obtain a processed signal containing distinct successive signals which can be associated with said two positions of said element of the exhaust; - determining time durations d-ι and d ₂ as the durations of two separate successive signals; - determine the movement reference as (d ₂ -di) / 2 or (d- | -d ₂ ) / 2. 13. The method according to claim 12, comprising: - generate two partial signals, a first partial signal and a second partial signal, like parts of a processed signal; - said first partial signal (SP) being generated in the image of a signal part covering at least part of the transition of said element from a first to a second position; and, - said second partial signal (SN) being generated in the image of a signal part covering at least part of the transition of said element from a second to a first position; - perform a correlation of each of the partial signals with the original filtered signal and determine the moments TT, T ₂ , T ₃ , T _N , as the moments when the correlations are at their maximum and / or minimum. 14. The method according to any one of claims 12 and 13, in which said partial signal generated comprises a segment representing the change between the two positions of the exhaust, said segment being vertical or inclined by ± 60 ° relative to the vertical. 15. The method according to any one of claims 12 to 14, in which said partial generated signal is chosen from signals having the general appearance according to (a) and / or (b): (a) (b) CH 712 940 A2