Method for determining a reference signed value of a clock oscillator

Abstract

A method for determining a signed reference value of a watch oscillator (150), such as a balance spring oscillator cooperating with an escapement (10, 100), comprising at least the following steps: - setting the oscillator (150) into oscillation, each oscillation comprising an input alternation and an output alternation, a plurality of sounds being emitted, - performing acoustic detection of the plurality of sounds emitted during the oscillations, - assigning the sounds of the plurality of sounds to the input alternation or the output alternation, - deducing an oscillation amplitude for the input alternation and for the output alternation, based on the assigned sounds, - comparing the oscillation amplitudes for the input alternation and for the output alternation, and deducing the signed reference value.

Description

[0001] The present invention relates generally to the field of watchmaking, more particularly to the adjustment of watch oscillators.

[0002] Methods are known for measuring the reference value of a mechanical watch movement, such as that disclosed in document FR2780169. Document WO2023 / 066614 discloses a method for determining a reference value and a method for adjusting a reference value. Document FR2767205 discloses the measurement of the amplitude of a balance wheel of a mechanical watch.

[0003] However, the methods of the prior art have the disadvantages of being complex to implement and of being sensitive to disturbances, in particular to external light disturbances.

[0004] The present invention aims to propose a new method for facilitating the adjustment of watch oscillators, and in particular the oscillator reference mark.

[0005] Thus, the invention relates, in a first aspect, to a method for determining a signed reference value of a watch oscillator, such as a spiral balance oscillator, cooperating with an escapement, the method comprising at least the following steps: setting the oscillator in oscillating motion relative to a frame of the watch movement, each oscillation comprising an input alternation and an output alternation, a plurality of sounds being emitted during the oscillations of the oscillator, performing acoustic detection of the plurality of sounds emitted during the oscillations, assigning (or identifying) the sounds of the plurality of sounds to the input alternation or to the output alternation for at least one oscillation, deriving an oscillation amplitude for the input alternation and for the output alternation, on the basis of the assigned sounds, comparing the oscillation amplitudes for the input alternation and for the output alternation, and deducing the signed reference value, preferably independently of a position of the oscillator.

[0006] This makes it possible to propose a new method to facilitate the adjustment of watch oscillators, and in particular the oscillator marker.

[0007] If the watch is not at the reference point, the oscillator's alternations will not be equal, particularly at low amplitude, disrupting the isochronism and thus the precision of the watch.

[0008] This is particularly useful regardless of the amplitude. At low amplitude, this allows for self-starting on the one hand, and to have a greater power reserve. At high amplitude, the impact will be on the stability of the gait, by limiting the oscillation amplitude for example.

[0009] Since the value of the reference point is generally a time expressed in [ms], it is necessary to compare the duration of the input and output alternations to obtain the value of the reference point. The sign of the reference point is defined by comparing the relative durations. The amplitudes of the oscillations constitute a subsequent result (i.e. a consequence of the relative durations of the alternations).

[0010] The lack of reference varies the time between the tick noise sequence (consisting of the three sounds called "DIB" or "DIB": respectively first sound, second sound, third sound as detailed below in relation to the general theory of watchmaking concerning oscillators and escapements, and in particular Swiss lever escapements) and the tock noise sequence (consisting of the three sounds called "DIB" or "DIB").

[0011] Advantageously, the plurality of sounds correspond to shocks of the exhaust, preferably releases and / or impulses, and sequences of shocks allow the attribution of the sounds to the input alternation or to the output alternation.

[0012] This allows the inputs and outputs to be identified. If the escapement is known, only the relative difference between the two durations called "DI" (i.e. between the first sound and the second sound) is necessary. If the escapement is unknown, a time difference between the DIs is measured. If the escapement is known, the release duration is sufficient; if the escapement is unknown, it is necessary to know the DI and the measurement of the DI phase difference.

[0013] Advantageously, the step of assigning the sounds emitted comprises a step of separating the sounds emitted according to the direction of rotation of the oscillator.

[0014] Advantageously, the step of assigning the emitted sounds comprises a step consisting of comparing a duration between a first sound and a second sound (DI) for the input alternation with a duration between a first sound and a second sound (DI) for the output alternation.

[0015] This makes it possible to propose a method for determining a signed reference value in the case of a previously known escapement.

[0016] Furthermore, this makes it possible to offer an efficient process, in particular because the DI corresponds to the noise of the impulse and therefore to the contact between the pallet and the end of a tooth of the anchor, so if the anchor is out of round (or has an out of roundness), the DI can directly reflect this.

[0017] Advantageously, the step of assigning the emitted sounds includes a step of detecting an out-of-roundness of an escape wheel of the escapement.

[0018] This allows us to account for a circularity defect in the escape wheel linked to its manufacture and / or assembly. This may be a structural defect. The out-of-roundness directly induces a sinusoidal shape on the analysis of DI times.

[0019] Advantageously, the out-of-roundness of the escape wheel is detected over a plurality of oscillations, such as a plurality of sounds emitted during a complete revolution of the escape wheel or a plurality of sounds emitted during a rotation of the escape wheel over at least three teeth of a plurality of teeth of the escape wheel, preferably over at least five teeth of the plurality of teeth of the escape wheel, preferably over at least half of the plurality of teeth of the escape wheel, preferably over the entire plurality of teeth of the escape wheel, and preferably over at least several revolutions of the escape wheel.

[0020] Advantageously, the deduction of the signed reference value is carried out in a single position (or inclination) of the oscillator, during one or more oscillations. In other words, it is a watch position or inclination, such as 3 o'clock, 6 o'clock, 9 o'clock, 12 o'clock, dial up, dial down. It is the entire oscillator, the watch movement (or watch mechanism) or the watch, which is positioned in the watch position or inclination, which is a classic control position.

[0021] Advantageously, a phase shift (phase advance or phase delay) or a difference between the oscillation amplitude for the input alternation and the oscillation amplitude for the output alternation makes it possible to determine the sign and / or the absolute value of the signed reference value. Note that the oscillation amplitude is a necessary information in the approaches to calculating the reference. It is identical, parallel or complementary to the duration / amplitude (DIB or DIB) and is an alternative that gives the same information to construct the first column of the table below (i.e. we know where the input is vs. the output). The DIB depends on the design of the escapement (asymmetry) while the offset of the oscillation amplitude depends on an assembly or manufacturing defect.

[0022] Advantageously, the method comprises the following step: - assigning the sounds of the plurality of sounds to the input alternation or to the output alternation for a plurality of oscillations.

[0023] This allows values ​​to be smoothed or averaged over a plurality of oscillations.

[0024] The invention relates, in a second aspect, to a method for adjusting the mark of an oscillator for a watch, comprising a method for determining the signed mark value according to the first aspect, and a step of adjusting the oscillator according to the determined signed mark value.

[0025] Advantageously, the step of adjusting the oscillator reference mark comprises an indexing adjustment step, preferably by adjusting an indexer or an indexing mechanism, a step of adjusting a mobile stud holder, a step of adjusting the collet and / or a step of adjusting a double chainring. These different possibilities are called "indexing adjustments", or simply adjustment. This involves rotating the position of the dead center (of the balance) by a certain angle relative to the line of the centers of the two mobiles. The dead center is the equilibrium position of the balance, it is the neutral point which separates two alternations (the entry and the exit).

[0026] The invention relates, in a third aspect, to a method for instructing adjustment of an oscillator for a watch, comprising a method for determining the signed reference value according to the first aspect, and a step of instructing an operator or a machine to adjust the oscillator according to the determined signed reference value. The machine can perform the adjustment automatically or semi-automatically.

[0027] Other characteristics and advantages of the present invention will appear more clearly on reading the detailed description which follows, of embodiments of the invention given by way of non-limiting example and illustrated by the appended drawings, in which: there Figure 1 represents an escape wheel cooperating with an anchor fitted with pallets and an oscillator for a watch, the Figure 2represents an audible signal characteristic of the operation of the watch oscillator cooperating with the escape wheel, the Figure 3 represents an oscillation amplitude of the oscillator for watch as a function of time, the Figure 4 represents a characteristic oscillation time of the watch oscillator as a function of a tooth position on the escape wheel cooperating with the oscillator, over a complete revolution of the escape wheel,

[0028] There Figure 1 represents oscillator 150 for watch cooperating with an escape wheel 10 and an anchor 100 provided with pallets, the escape wheel 10 and the anchor 100 forming an escapement 10, 100.

[0029] The escape wheel 10 cooperates with the anchor 100 in a conventional manner for an escapement preferably with a Swiss anchor. In conventional operation, the escape wheel 10 tends to want to rotate because it is engaged with a barrel of the watch movement (not shown) and the anchor 100 prevents it from rotating, or allows it to rotate at a certain frequency, and transmits energy to the oscillator of the watch movement 150 to maintain it.

[0030] The anchor 100 comprises a fork 110 in a conventional manner, in order to allow cooperation with an axis of the balance of the oscillator 150.

[0031] The anchor comprises two pallets 101, 102, for example made of ruby, silicon or diamond: these are the input pallet 101 and the output pallet 102. The anchor may be a single piece or comprise assembled pallets. In the latter case, the pallet 101 is housed in an input pallet housing 101a of the anchor 100 and the pallet 102 is housed in an output pallet housing 102a of the anchor 100. The anchor 100 is rotatable about an axis of rotation of an anchor rod 103 of the anchor 100.

[0032] The escape wheel 10 comprises a plurality of teeth 11 and is rotatable about an axis of rotation of a wheel shaft 12. The escape wheel may be, for example, made of steel, NiP, silicon, oxidized silicon or coated with a tribological layer such as, for example, diamond.

[0033] A contact 15 is present when the contacting tooth 13, among the teeth 11 of the escape wheel 10, cooperates with one of the pallets 101, 102. In the example of the Figure 1 , the escape wheel has twenty teeth 11, including one tooth in contact 13 with the input pallet 101 or the output pallet 102, depending on whether we consider the Figure 1 left or right. Contact 15 can take place on an area or a surface.

[0034] To the left of the Figure 1 is shown the escape wheel 10 and the anchor 100, in a position in which the contacting tooth 13 (among the teeth 11) touches the input pallet 101. To the right of the Figure 1 the escape wheel 10 and the anchor 100 are shown, in a position in which the contacting tooth 13 (among the teeth 11) touches the output pallet 102.

[0035] The process is thus detailed with reference to the Figures 1 and 2 .

[0036] For each passage of a tooth in contact 13 with one of the pallets 101, 102, a sound (or several sounds) is emitted, which may correspond at least in part to the so-called “TIC” sounds or to the so-called “TAC” sounds, conventionally. Thus, as discussed above, we can define by “TIC” the sound linked to the cooperation between a tooth of the escape wheel with the rest plane of a first pallet, and by “TAC” the sound linked to the cooperation between a tooth of the escape wheel with the rest plane of a second pallet, that is to say that the “TIC” and the “TAC” are sounds characteristic of the impact between the rest plane of a first and respectively of a second pallet. We can consider the "TIC" as the entry sound (i.e. of the entry pallet on the tooth) and the "TAC" as the exit sound of the anchor (i.e. of the exit pallet on the tooth) for said tooth of the escape wheel.In other words, it is neither the same tooth nor the same pallet for the TIC and the TAC: there are generally three teeth apart between the TIC and the TAC, and the TIC comes from the input pallet while the TAC comes from the output pallet. The number of three teeth is a non-limiting example for the Swiss anchor. It is the number of teeth embraced between the two pallets. It is possible to use a synchronous analysis to allow each tooth of the wheel to be numbered and to know which tooth cooperates with which pallet at each moment. Synchronous analysis or the use of a modulo (equal to the number of teeth of the escape wheel) allows the teeth to be numbered in the repetition of the sequences. Synchronous analysis is a known technique for extracting low amplitude, but narrowband signals drowned in broadband ambient noise, in particular by multiplying the signal by a sinusoidal signal with a frequency close to the average frequency to be detected.It can also be defined as coherent demodulation or frequency change performed on an intermediate frequency carrier. It can be done using a synchronous detection amplifier.

[0037] Acoustic detection of the assembly in operation can be provided by an acoustic sensor, by multimode piezoacoustic detection or even more advantageously by acoustic and ultrasonic stethoscopic detection or even one or more ultrasonic microphones. Particularly near-field acoustic and ultrasonic stethoscopic detection, which has very high qualities of bandwidth, spectral flatness, low noise and immunity to parasitic disturbances, makes it possible to propose a determination method. This is also applicable to components made of silicon, steel or metal alloy, amorphous or not, crystalline materials,and generally any material used for watch components. The piezoacoustic sensor preferably has a good signal-to-noise ratio over a wide frequency band. It is also possible to provide multi-mode operation with this piezoacoustic sensor in order to obtain a high signal-to-noise ratio over a wide frequency range via the natural vibration modes. This has the advantage of being able to go to high frequency, improve the speed of rise of the sound signal and obtain better time accuracy. Measurement via direct contact (solid-borne transmission, for example by a crown) is possible. It is also possible to cut low frequencies in order to isolate ambient noise. MEMS (microelectromechanical system) technology makes it possible to cover a wide bandwidth (for example 100Hz - 80kHz),while allowing a near-field contactless measurement (stethoscope type). The various sensors mentioned above allow finer, more faithful listening, and therefore greater recognition capacity by obtaining a greater quantity of information, by extending the band towards high frequencies which contain useful acoustic energy, by obtaining information less disturbed by the acoustic (and vibratory) environment, by the fact of the rapid dissipation of short wavelengths in air and in solids, by obtaining information less disturbed by the connection and electrical environment, by digitizing the signals as close as possible to the transduction source, by obtaining very low self-noise by multiplying paired transducers, by the possibility of implementing coherent reconstruction, time shift, convolution, filtering and focusing techniques,that is to say more precisely all the methods of signal processing and acoustic antenna algorithms by microphone arrays and applicable in near field. In addition, by the use of miniature transducers of known sensitivity and factory matched in amplitude and phase, and for their entire lifetime, the technology does not require initial or periodic calibration.

[0038] There Figure 2 represents an audible signal 200 which is characteristic of the operation of the oscillator 150 for watch cooperating with the escape wheel 10 and the anchor 100 provided with pallets.

[0039] The sound signal is subject to the acoustic detection of the operation of the escapement, i.e. of the escape wheel 10 for a given tooth 11, 13 with the anchor 100 and the oscillator 150. The acoustic detection can be carried out using an acoustic sensor, for example a piezo-acoustic sensor or a microphone, in particular ultrasonic.

[0040] It is considered that the detected sound signal 200 comprises a first sound 201, a second sound 202 and a possible third sound 203. The third sound 203 is advantageously and in particular used to ensure the presence and coherence of the first sound 201 and the second sound 202. In other words, the shape and amplitude of the third sound 203 makes it possible to confirm that the first sound 201 and the second sound 202 are indeed the sounds that one hopes to detect. It is further possible to carry out this confirmation by analyzing an envelope of the sounds 201, 202 or by considering their amplitude or any other parameter, such as a distance from another sound (for example, parasitic) not represented. Furthermore, it is possible to use the third sound 203 for measuring the amplitude of the pendulum, as will also be discussed below.Generally speaking, it is considered that the first sound 201 occurs at the first shock, when the oscillator 150 plate pin touches the entry of the fork of the anchor 100. The second sound 202 occurs when a tooth of the escape wheel 10 falls onto the impulse plane of the pallet and the fork of the anchor 100 catches up with the oscillator 150 plate pin. The third sound 203, is generally the loudest, and occurs when a tooth of the escape wheel falls onto the rest plane of the pallet (fall) and the anchor rod presses against the limiting pin (lost path).

[0041] The elapsed time 204 between the first two sounds 201, 202 of the escapement is also called DI, and may be a function of the penetration distances of the pallets 101, 102. The elapsed time 204 may be considered from peak to peak or from an average (or median or characteristic) value of the first sound 201 and the second sound 202. The elapsed time 204 may furthermore be a function of the operating amplitude of the balance of the oscillator 150 coupled to the escapement.

[0042] The elapsed time 205 between the first sound 201 and the third sound 203 is also called DB.

[0043] The method according to the present invention makes it possible to determine the signed reference value of the oscillator 150 directly on the assembled watch movement, without visual access to the components. In particular, it is necessary to hear the noise of the escapement and clearly distinguish the three shocks of the escapement (it is necessary to isolate and not take into account the parasitic noises).

[0044] In a particular embodiment, the measurement can be carried out by analyzing the acoustic signal coupled with the optical signal of the balance, in particular to measure the actual operating amplitude of the balance.

[0045] Furthermore, it is possible to know the amplitude either by acoustic detection (as for example with the third sound as discussed above), or by optical detection, or any combination of these.

[0046] An advantage of this measurement is that it is carried out in masked time during the time control. This measurement strategy is useful when diagnosing parts already assembled to provide a means of verification.

[0047] The advantage of this measurement is to ensure that under dynamic conditions, the exhaust operates within its tolerance interval (IT). This allows for an additional layer of quality control at no extra cost.

[0048] There Figure 3 represents an oscillation amplitude of the oscillator 150 for watch as a function of time. This is the operating amplitude with an offset mark. The indication "E" on the Figure 3 corresponds to the input sequence, the indication "S" corresponds to the output sequence, and the indication "D" corresponds to the marker shift line. The times t1 and t2 are mentioned below.

[0049] Oscillator 150 conventionally comprises a balance wheel and a hairspring. The balance wheel is mounted on a balance shaft and cooperates with the escapement, in particular at the fork 110 of the anchor 100, in the case of a Swiss anchor escapement.

[0050] The oscillator oscillates around its equilibrium point, with a certain amplitude on either side. This is generally not perfectly centered (i.e. a reference point defect): it is offset to one side or the other (i.e. to the input side or the output side). This is due to the assembly of the elements together and their mechanical tolerances. One way to identify the direction of the reference point defect is to rely on the position of the input and output of the escapement. The input alternation is the alternation during which the input pallet transmits the impulse. The output alternation is the alternation during which the output pallet transmits the impulse.

[0051] If, for example, the reference frame is offset to the input side, the amplitudes of the alternations on the input side will be smaller than the amplitudes of the alternations on the output side. The duration of the alternations on the input side will also be shorter than the alternations on the output side.

[0052] This is represented in the Figure 3 by a line of the reference shifted towards the top of the figure above the zero amplitude, and "positive" amplitudes (i.e. above zero) on the input side smaller (in absolute value) and "negative" amplitudes on the output side larger (in absolute value). In other words, the amplitude of the oscillator during the time t1 indicated in Figure 3 is smaller than the amplitude during time t2 of the Figure 3 .

[0053] The measured reference point is defined by R[s] = (t1 - t2) / 2.

[0054] The interest for the user is to be able to convert information provided in [s] or [ms] into [°] or [degrees] to identify the adjustment (or adjustment) necessary to correct the reference mark. We thus have R[°] = A10 * sin(ω * R[s]), with A0 being the value of the instantaneous amplitude of the oscillator. A0 is typically between 100° and 320°, preferably between 200° and 300°. w is the natural pulsation of the oscillator, which is equal to 2 * pi * f, f being the frequency typically between 2 and 15 Hz.

[0055] For an oscillator oscillating at 4hz with a measured reference error of 5 ms, and operating at 300° amplitude, the reference is shifted by 3.8°. The amplitude of the alternations on the first side is 296.2° and on the second side 303.8°. The correction (using an indexing adjustment means) must therefore be 3.8° in a direction yet to be identified.

[0056] The next step is to identify the direction of correction.

[0057] The exhaust inlet and outlet produce a characteristic noise sequence as detailed above. The relative durations of the impact sequences allow us to differentiate which exhaust phase has taken place.

[0058] There are several differentiation methods.

[0059] The first method of differentiation is as follows.

[0060] For movements whose escapement is known, it is possible to proceed in this way. Typically for certain escapements, the clearance time at the entry DI is shorter than the clearance time DI at the exit, this is a consequence of the construction choices (as is for example explained in the book "the theory of watchmaking" by Charles-André Reymondin, George Monnier, Didier Jeanneret and Umberto Pelaratti, notably on page 122) and which is concretely reflected in observables such as the different DI / DB ratio. It is possible to associate a shock sequence with one of the sides of the escapement.

[0061] In reference to the Figure 3 , the pendulum moves from the input side to the output side, the alternation duration being greater in this direction, the amplitude is greater and the reference mark is shifted towards the input side.

[0062] There Figure 4represents a characteristic oscillation time of the oscillator 150 for watch as a function of a tooth position on the escape wheel 10 cooperating with the oscillator 150, over a complete revolution of the escape wheel 10. The abscissa corresponds to the number of teeth of the escape wheel and the ordinates to a time in seconds.

[0063] The second method of differentiation is the following in relation to the Figure 4 .

[0064] For unknown escapements or for escapements for which the DI / DB entry and exit signature is unknown, the entry or exit can be identified via the out-of-roundness. It is also possible to use the observation of the out-of-roundness to identify the escapement phase in progress. A phase shift of the DI times between the tick and the tock can be observed on the revolution of the escape wheel. The same tooth passing successively through the entry then the exit after two or three teeth, the observation of the phase shift makes it possible to associate the shock sequence with one of the sides of the escapement. The phase shift is thus visible in Figure 4 between tooth 5 and tooth 8. The upper curve corresponds to a regression of the DI durations for the output and the lower curve corresponds to a regression of the DI durations for the input. Figure 4 thus represents the set of DI durations for entry and exit, over one revolution of the escape wheel 10.

[0065] There Figure 4thus represents a curve for each DI of the TIC and TAC still undetermined on which is the inlet or the outlet, regardless of the construction. The exhaust inlet will necessarily be the phase advance curve and by knowing the inlet and the outlet, it is possible to determine the direction of rotation for the adjustment of the mark.

[0066] From the moment the entry and exit are identified by tick or tock according to the first or second method, the mark can be signed. If we detect which escapement phase has just occurred, we can determine the direction of rotation of the balance and associate the duration of alternation with a certain direction (association of tick / tock with entry / exit). The table below illustrates the case of an escape wheel rotating clockwise seen from the back. Note that this is the same convention as for the WO document mentioned above. Table 1 Shock sequence heard previously Direction of movement of the pendulum Relative duration of the alternation following the shock sequence Interpretation Entrance The pendulum moves from the input side to the output side Bigger The marker is offset from the entry side. You must move the skewer or the movable piton holder to the exit side, or make an equivalent indexing adjustment. positive sense Negative marker to move in the positive direction Smaller The marker is offset to the exit side. You must move the skewer or the movable piton holder to the entry side, or make an equivalent indexing adjustment. Positive marker to move in the negative direction Exit The pendulum moves from the output side to the input side Bigger The marker is offset to the exit side. The skewer or the movable piton holder must be moved to the entry side or an equivalent indexing adjustment must be made. Positive marker to be moved in the negative direction Negative meaning Smaller The marker is offset from the entry side. You must move the skewer or the movable piton holder to the exit side, or make an equivalent indexing adjustment. Negative marker to move in the positive direction

[0067] One of the advantages of the present invention is to allow the implementation of the method in a single watch position or inclination (3 o'clock, 6 o'clock, 9 o'clock, 12 o'clock, with high dial or low dial), thanks to the recognition of the current phase (tick or tock, entry or exit). This is not possible with the prior art, in particular because it is not possible with the prior art to reconstruct the first column of the table above.

[0068] It will be understood that various modifications and / or improvements obvious to those skilled in the art can be made to the different embodiments of the invention described in the present description, such as an indexing adjustment by angular displacement of the chainring pin or the ferrule, for example.

[0069] In particular, reference is made to the possibility of having another method for assigning sounds to input or output.

[0070] It is possible to combine the embodiments as much as possible or necessary.

Claims

1. Method for determining a signed reference value of an oscillator (150) for a watch, such as a spiral balance oscillator, cooperating with an escapement (10, 100), the method comprising at least the following steps: - setting the oscillator (150) in oscillating motion relative to a frame of the watch movement, each oscillation comprising an input alternation and an output alternation, a plurality of sounds (201, 202, 203) being emitted during the oscillations of the oscillator (150), - carrying out an acoustic detection of the plurality of sounds emitted (201, 202, 203) during the oscillations, - assigning the sounds of the plurality of sounds (201, 202, 203) to the input alternation or to the output alternation for at least one oscillation, - deducing an oscillation amplitude for the input alternation and for the output alternation, based on the assigned sounds (201, 202, 203),- compare the oscillation amplitudes for the input alternation and for the output alternation, and deduce the signed reference value, preferably independently of a position of the oscillator (150)., 2. Method for determining the signed reference value according to claim 1, wherein the plurality of sounds (201, 202, 203) correspond to shocks of the escapement (10, 100), preferably releases and / or impulses, and wherein sequences of shocks allow the attribution of the sounds (201, 202, 203) to the input alternation or to the output alternation.

3. Method for determining the signed reference value according to one of the preceding claims, in which the step of assigning the emitted sounds (201, 202, 203) comprises a step of separating the emitted sounds (201, 202, 203) according to the direction of rotation of the oscillator (150).

4. Method for determining the signed reference value according to one of the preceding claims, in which the step of assigning the emitted sounds (201, 202, 203) comprises a step of comparing a duration between a first sound (201) and a second sound (202) for the input alternation with a duration between a first sound (201) and a second sound (202) for the output alternation.

5. Method for determining the signed reference value according to one of the preceding claims, in which the step of attributing the emitted sounds (201, 202, 203) comprises a step of detecting an out-of-roundness of an escape wheel of the escapement (10).

6. Method for determining the signed reference value according to the preceding claim, in which the out-of-roundness of the escape wheel (10) is detected on a plurality of oscillations, such as a plurality of sounds emitted (201, 202, 203) during a complete revolution of the escape wheel (10) or a plurality of sounds emitted (201, 202, 203) during a rotation of the escape wheel (10) on at least three teeth of a plurality of teeth of the escape wheel, preferably on the entire plurality of teeth of the escape wheel (10).

7. Method for determining the signed reference value according to one of the preceding claims, in which the deduction of the signed reference value is carried out in a single position of the oscillator (150).

8. Method for determining the signed reference value according to one of the preceding claims, in which a phase shift or a difference between the oscillation amplitude for the input alternation and the oscillation amplitude for the output alternation makes it possible to determine the sign and / or the absolute value of the signed reference value.

9. Method for adjusting an oscillator (150) for a watch, comprising a method for determining the signed reference value according to one of the preceding claims, and a step of adjusting the oscillator (150) according to the determined signed reference value.

10. Method for adjusting a watch oscillator according to the preceding claim, in which the step of adjusting the oscillator (150) comprises an indexing adjustment step, preferably by adjusting an indexer, a step of adjusting a mobile stud holder or a step of adjusting a double plate.

11. Method for instructing the adjustment of an oscillator (150) for a watch, comprising a method for determining the signed reference value according to one of claims 1 to 8, and a step of instructing an operator or a machine to adjust the oscillator (150) according to the determined signed reference value.

Images