Method for verifying the display drive train of a watch or a watch head downstream of its oscillator

The method of continuous acoustic and optical synchronization of a watch's display addresses the uncertainty in kinematic chain verification, achieving precise detection of display defects with reduced measurement uncertainty.

EP4760417A1Pending Publication Date: 2026-06-17OMEGA SA

Patent Information

Authority / Receiving Office
EP · EP
Patent Type
Applications
Current Assignee / Owner
OMEGA SA
Filing Date
2024-12-11
Publication Date
2026-06-17

AI Technical Summary

Technical Problem

Existing watchmaking processes fail to accurately verify the kinematic chain of a watch's display, including the oscillator, mainspring, and gear train, due to uncertainties introduced by gear play, slippage, and transmission errors, which can mask the performance of the best oscillator and lead to significant defects.

Method used

A method involving continuous or pseudo-continuous acoustic and optical measurements over a prolonged period to synchronize the display of a watch, using a combination of acoustic and optical methods to compare the time recorded by the regulator with the display, ensuring accurate synchronization and identifying any display-related issues.

Benefits of technology

Achieves a measurement uncertainty of 5 seconds/day for minute hands and 100 ms/day for second hands, reliably detecting display malfunctions, thereby ensuring high precision and reliability in watch manufacturing.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure IMGAF001_ABST
    Figure IMGAF001_ABST
Patent Text Reader

Abstract

One aspect of the invention relates to a method for verifying the display of a watch head, where the watch head is ensured to run continuously for 24 hours, an initial optical image of the hand is taken at an initial time a with reference to fixed points, a final optical image of the hand is taken after at least 24 hours at a final time measured on a reference clock with reference to the fixed points, the positions of said hand are established at said initial and final times by means of visual recognition, the difference between the reference position with respect to the reference clock and the actual position of the hand is calculated, and the value of the measured state deviation is displayed.
Need to check novelty before this filing date? Find Prior Art

Description

Technical field of the invention

[0001] The invention relates to a method for verifying the kinematic chain of display of a watch or a watch head downstream of its oscillator simultaneously with the tests commonly carried out in watchmaking, on a set of test stations.

[0002] The invention relates to the field of watchmaking operation and status controls. Technological background

[0003] To ensure a certain level of chronometric precision, it is essential to perform rate checks before finishing a watch head or watch. These rate checks focus on the quality of the oscillator in the case of a mechanical watch. However, display accuracy, which relates to condition checking, concerns not only the oscillator (or regulator) but also the mainspring, the display adjustment mechanism, and all the gears, particularly the finishing gear train and the drive system for the indicators, especially the hands. Play in the gears introduces uncertainty into the visual measurement of these indicators or hands. Slippage of the indicators or hands on their pivots is generally much greater than the measurement uncertainty.Finally, a transmission error (number of teeth) is one of the most critical cases, which, in the event of the loss of a tooth, can generate a defect lasting more than 100 seconds per day.

[0004] The performance of the best oscillator can thus be masked by a purely mechanical or kinematic defect: runout, lubrication defect, damaged or missing tooth, gear play, pivot slippage, needle friction, or other.

[0005] It is therefore important to perform a display synchronization check for a watch head or a watch with a mechanical movement. The display check typically performed during time setting does not, by design, affect the transmission of the display during watch operation.

[0006] Since the display kinematic chain is checked independently, condition measurement tests of a watch or watch head can be carried out without the display of the watch or watch head, and therefore only on the regulator. Summary of the invention

[0007] The invention therefore aims to perform a display synchronization check, to avoid the delivery of a timepiece containing one of these defects.

[0008] For this purpose, the invention relates to a method according to claim 1. Brief description of the figures

[0009] The aims, advantages and features of the invention will become clearer upon reading the detailed description that follows, with reference to the attached figures, where: there figure 1illustrates a particular sequence well suited to the implementation of the invention, with operations carried out in parallel over time; the upper line shows winding operations, the next line shows a sequence of operations, in integrated movement within a watch head, these operations being magnetic behavior control, daily precision acoustic control, and water resistance control; the lower line of the figure concerns optical controls; the middle part of the figure details the first of the daily precision acoustic measurements, with a succession of measurements in the different standardized chronometric positions, at two different temperatures; the figure 2 is a diagram showing the timestamp of the ticks and tocks of the movement, with the times of first and last releases, impulse and fall. Detailed description of the invention

[0010] The invention relates to a method for verifying the kinematic chain of display of a watch or a watch head downstream of its oscillator, using a simultaneous measurement of the rate of a watch or a watch head by optical state taking of the hands and continuous or pseudo-continuous acoustic measurement on a set of test stations.

[0011] According to this method, the said watch or watch head is ensured to run continuously for a measurement period TM of at least 24 hours, ideally for several days depending on the desired measurement accuracy, either by providing the watch or watch head with a power source capable of keeping it running for at least the measurement period TM, of at least 24 hours of operation, or by equipping the test station with a winding and / or stirring device with which the watch or watch head is wound or stirred for its resetting for at least this measurement period TM, at least 24 hours, or both by providing the watch or watch head with a power source capable of a certain power reserve and by equipping the said test station with a winding and / or stirring device with which the said watch or watch head is wound or stirred for its resetting for the measurement period TM of at least 24 hours.

[0012] At an initial time t0, at least one initial optical image of the handing of this watch or watch head is taken, with reference to fixed points of the watch or watch head, and / or to fixed points that include a tooling carrying the watch or watch head.

[0013] Immediately after the initial shot, the watch or watch head is placed in a device having a reference clock and a tick and tock detection system, to perform a continuous or pseudo-continuous acoustic measurement of the regulator over the entire time period t.

[0014] A continuous or pseudo-continuous acoustic measurement of the oscillator or regulator is thus carried out comparatively of said watch or watch head between the initial optical shot and the final optical shot, allowing the rate deviation to be determined.

[0015] At the end of time t, after the measurement duration TM, the watch's operation is evaluated acoustically by counting the number of ticks and tocks in relation to the elapsed time.

[0016] We know that many stressors can affect the watch's operation between the two state checks, namely, dynamic movement, static movement, temperature, humidity.

[0017] At the final instant t0+t measured on a reference clock, at least one final optical image of the clock's hand or clock head is taken at fixed points on the clock or clock head, and / or at fixed points on the tooling. The positions of this hand at the initial instant t0 and at the final instant t0+t are established using visual recognition methods installed on the test station. The difference between the reference position relative to the reference clock and the actual position of the hand is calculated, and the measured state error value is displayed.

[0018] Optical measurement therefore makes it possible in particular to ensure chronometric control of the motor + oscillator part and the display part of the watch, whereas acoustic measurement ensures chronometric control only of the motor + oscillator part of the movement.

[0019] To determine the synchronization of the display, we calculate the difference between the rate of the watch or watch head determined by taking the optical state of the display and the rate of the watch determined by counting the ticks and tocks in a continuous or pseudo-continuous acoustic manner. This difference will be close to 0 if the watch is free of display problems, regardless of the performance of its oscillator.

[0020] It is understood that the invention relates to a comparison of the time recorded by the acoustic regulator and the time recorded by the display. The duration of this time only affects the accuracy of the comparative measurement that one wishes to perform.

[0021] In an advantageous implementation of the process, each optical measurement is carried out by several shots, determining the average of the needle positions identified by each shot to reduce the measurement uncertainty.

[0022] More specifically, according to this process, between the initial instant and the final instant, the watch or watch head is subjected to dynamic movements in space, to windings, to waiting in standardized chronometric positions, or any other stress for the regulator and the display mechanism of the watch head or watch according to official standards such as ISO 3159, or internal standards, or others.

[0023] More specifically, the watch or watch head is subjected to movements in space, to perform rate checks in standardized chronometric positions and to determine the rate deviation values ​​in said standardized chronometric positions, and the rate deviation values ​​are compared on the one hand, and the value of the condition deviation on the other hand, with reference to internal specifications, to authorize or prohibit the continuation of the manufacture of the watch or watch head.

[0024] More specifically, intermediate step checks are carried out in the standardized chronometric positions by instantaneous or continuous acoustic measurements, in part or in full.

[0025] More specifically, at least one intermediate instant between this initial instant and this final instant, at least one intermediate check is carried out during which at least one intermediate optical shot of the watch hand or watch head is taken in reference to said fixed points, the positions of the hand are established at this initial instant and at this intermediate instant by means of visual recognition, the difference between, on the one hand, the reference position with respect to said reference clock, and on the other hand, the actual position of the hand is calculated, and the value of the state deviation measured at this intermediate instant is displayed.

[0026] More specifically, between the initial moment and the final moment, the watch or watch head is subjected to a passage in an oven under predefined conditions of duration, temperature and humidity.

[0027] More specifically, the said watch or watch head is enclosed and unblocked in a manipulation box, which is chosen to be transparent, and all the manipulations are carried out between the said initial instant except and the said final instant with the said watch or watch head in the said manipulation box.

[0028] More specifically, at least one initial shot is taken before the watch or watch head is inserted into the manipulation box, and at least one final shot is taken after the watch or watch head is extracted from the manipulation box.

[0029] In a particular mode, this verification process is applied to a mechanical watch.

[0030] It should be noted that acoustic measurements in watchmaking are said to be instantaneous (generally 40 seconds), whereas in the case of the invention, it is a continuous acoustic or pseudo-continuous measurement in order to ensure the counting of each tick of the watch during the entire observation period between the two visual state measurements.

[0031] A particular, but not limiting, implementation of the process according to the invention is described in detail below.

[0032] The evolving requirements for the certification of mechanical movements and watches resistant to magnetic fields of 1.5 T (15,000 G), METAS N001 v 1.2, are compatible with optoacoustic measurement of watches and watch heads. Specifically, a technology based on their placement in acoustic test boxes, described in particular in documents EP3410234, EP3812847, EP22209439.3, CH001445 / 2022, EP22209441.9, and CH001404 / 2022, is used. These universal boxes are compatible with watch heads and offer an accuracy of ±50 ms / day. Often designed for batches of ten or more pieces, these test boxes advantageously contain the watches or watch heads.

[0033] The display synchronization is checked by taking an optical reading of the second and / or minute hands. In a "seconds and minutes" variant, the display synchronization is checked over a period of approximately 6 days (at least 120 hours) in order to minimize measurement uncertainty.

[0034] There figure 1 This illustrates a particular sequence well-suited to the implementation of the invention. The top line shows a series of winding operations R1, ..., Rn, to ensure the maintenance of the mechanism's operation: at the beginning of each cycle described in the METAS N001 requirements, a winding is performed. Below this is a series of cycles that are operations for checking magnetic behavior, daily accuracy, and leak tightness, the order and number of which can be modified without departing from the invention within the limitations of the METAS N001 requirements.

[0035] In preparation for watch head testing according to METAS N001 requirements, the watch head is placed in an acoustic test box as also mentioned above.

[0036] Preferably each acoustic test box contains a batch, for example of ten watch heads; this batch then undergoes together all of these final running and condition checks.

[0037] The following operations are preferably carried out in an automated and robotic manner: each manipulation in space is applied to the acoustic test box, each test under controlled temperature or controlled humidity, or controlled pressure, or other, concerns the entire box.

[0038] The first control operation 10 is a daily precision chronometric control PJ1, with different orientations of the acoustic test box in space. During this test, microphones, including piezoelectric microphones, are used inside the acoustic test box in contact with the watch head to listen for the ticking and isolate the characteristic moments of a mechanical watch oscillator: release, impulse, and fall, common to both Swiss lever and coaxial escapements, even though these moments do not correspond exactly to contact between the same surfaces. This first control operation 10 is carried out according to the requirements of METAS N001 for the details of temperatures and positions.

[0039] The following are: a magnetic field function check operation 20, a second chronometric check operation 30 similar to the first 10 according to the detail above, and according to the METAS N001 requirements for the temperature details, a demagnetization operation 40, a third chronometric check operation 50 similar to the first 30 according to the detail above, and according to the METAS N001 requirements for the temperature details, a fourth chronometric check operation 60 similar to the first 10 according to the detail above, and according to the METAS N001 requirements for the temperature details, a power reserve check operation 70 (cycle 7 of METAS N001), before proceeding with a leak test.

[0040] Alongside the acoustic tests, on the lower line, we see the optical tests advantageously placed during the manipulations of the watch head to confirm or refute the display synchronization.

[0041] Display synchronization is verified by optically scanning the second and minute hands. Optical hand recognition employs a series of continuous measurements over a 60-second observation period, eliminating optical image artifacts, to determine whether a watch head or watch is ahead or behind two states. During this time, each acquisition is time-stamped, and a pattern-matching algorithm searches for the watch head or watch, in terms of position and orientation, and the hands.

[0042] The technology of the acoustic test boxes is compatible with watch heads to measure an accuracy of + / - 50 ms / day or even + / - 10ms per day if a calibration of the reference clock embedded in each box is carried out prior to the measurements.

[0043] The acoustic measurement and processing device, as well as its implementation, are described in detail in sections 0014 to 0021 and Figures 4 to 6 of document EP3812847. In summary, the continuous acoustic measurement system aims to time-stamp the shocks to the movement's oscillator. A control unit processes these time stamps and determines the daily accuracy of the watch heads in the acoustic test box.

[0044] Due to data volume and power consumption (processor processing), the signals from the ten watch heads mounted in the acoustic test chamber are temporally divided into signal trains. The number of activated channels and the length of the signal trains can be adjusted according to the embedded firmware. Each signal train must have a minimum duration of 5 seconds to allow the algorithm to configure itself and be sufficiently accurate in detecting the fall; approximately 2 seconds are used for noise threshold and frequency class detection, and about 3 seconds to accumulate enough data to estimate the impulse and fall by superposition.

[0045] The embedded algorithm then determines the timestamp of the ticks and tocks. At the beginning of each measurement series, the algorithm detects the frequency by classifying it in 0.5 Hz steps between 2 Hz and 5 Hz, then detects the noise level in the quiet zones to determine the threshold for detecting release. The releases of the ticks and tocks are thus identified by crossing this threshold. Since the information is a modulo of the movement frequency, the first and last releases are retained. The fall and the impulse are timestamped by superimposing the signals. These timestamps are therefore only available for the last tick and tock, in order to have a maximum of data for the superposition. figure 2The graph shows the tick timestamp in solid lines and the click timestamp in dashed lines, and only displays the first and last ticks of each measurement train. Each measurement train has a duration TH, for example, 10 seconds, and the measurement trains are separated from each other. For the tick timestamp, t1 is the first release, t2 the last release, t3 the pulse detection, and t4 the fall detection; for the click timestamp, t5 is the first release, t6 the last release, t7 the pulse detection, and t8 the fall detection.

[0046] The acoustic measurement of daytime walking with the acoustic test box is validated by comparative measurements with the optical measurements thus performed. A relatively rapid convergence (after approximately 24 hours) is observed between the acoustic and optical measurements.

[0047] Such a parallel opto-acoustic measurement process over a period of 120 hours makes it possible to achieve a measurement uncertainty of the display synchronization of the order of 5 seconds / day on the minute hand and 100 ms on the second hand, ensuring the reliable detection of possible display malfunctions.

[0048] A typical, non-exhaustive range of operations includes the following sequences: A Place in acoustic measurement box B Reassemble C Wait 24h D Check daily accuracy 01 E Reassemble F Place in standard conditioning G Check operation under magnetic field H Take MS state 0h I Place in acoustic measurement box J Check daily accuracy 02 K Demagnetize L Reassemble M Wait 24h N Check daily accuracy 03 O Reassemble P Wait 24h Q Check daily accuracy 04 R Reassemble PS Wait 48h T Place in standard conditioning U Take MS state 0h V Check RM W Check SA X Check sealing Y Place in storage. In particular, operations A through S, with the exception of operations G and H, are performed at the head of the watch in an acoustic test box, which is subjected to movements imparted by a shaker, as well as programmed passages through an oven. In short, the invention utilizes existing technical packaging: standard-compatible vision and handling equipment for magnetic field testing and acoustic test boxes, already developed, to meet the requirements of METAS N001 and guarantee a perfectly secure inspection cycle without any human intervention that could bias any of the checks.

Claims

1. A method for verifying the kinematic chain of display of a watch or watch head downstream of its oscillator, simultaneously with tests on a set of test stations, wherein a simultaneous measurement of the rate of a watch or watch head is used by optical state taking of the hands and continuous or pseudo-continuous acoustic measurement on a set of test stations, and the said watch or watch head is ensured to run continuously for a measurement period TM of at least 24 hours, either by providing the said watch or watch head with an energy source capable of keeping it in operation for at least the said measurement period TM, or by equipping the said test station with a winding and / or stirring device with which the said watch or watch head is wound or stirred for its resetting during at least the said measurement period TM.

2. Verification method according to claim 1, wherein at an initial time t0 at least one initial optical image of the hand of said watch or watch head is taken with reference to fixed points of said watch or watch head, and / or to fixed points on a tool in which said watch or watch head is enclosed and locked, and after at least said measurement time TM at a final time measured on a reference clock at least one final optical image of the hand of said watch or watch head is taken with reference to fixed points of said watch or watch head, and / or to fixed points on said tool, the positions of said hand are established at said initial time and at said final time by visual recognition means, with which said test station is equipped, the difference between on the one hand the reference position with respect to said reference clock,and on the other hand, the actual position of the switch, and the measured state error value is displayed.

3. Verification method according to claims 1 and 2, wherein a continuous or pseudo-continuous acoustic measurement of the oscillator or regulator is carried out comparatively of said watch or watch head between the initial optical shot and the final optical shot allowing the rate deviation to be determined.

4. Verification method according to claim 3, wherein, after said measurement time TM, the watch's performance is evaluated acoustically by counting the number of ticks and tocks in relation to the elapsed time.

5. A verification method according to any one of claims 1 to 4, wherein, between said initial instant and said final instant, said watch or watch head is subjected to dynamic movements in dynamic space, to perform a winding of said watch or watch head, and / or to perform rate checks in standard chronometric positions.

6. A verification method according to any one of claims 1 to 5, wherein, between the initial instant and the final instant, the watch or watch head is subjected to waiting periods in standardized chronometric positions, and / or to any other stress for the regulator and the display mechanism of the watch head or watch.

7. Verification method according to claim 5 or 6, wherein said watch or watch head is subjected to movements in space, to perform rate checks in standard chronometric positions and to determine the rate deviation values ​​in said standard chronometric positions, and said rate deviation values ​​are compared on the one hand, and said value of the state deviation on the other hand, with reference to internal specifications, to authorize or prohibit the continued manufacture of said watch or watch head.

8. Verification method according to any one of claims 5 to 7, wherein said running checks are carried out in the standardized chronometric positions by acoustic measurements.

9. A verification method according to any one of claims 1 to 8, wherein, between said initial instant and said final instant, said watch or watch head is subjected to a passage in an oven under predefined conditions of duration, temperature and humidity.

10. A verification method according to any one of claims 1 to 9, wherein said watch or watch head is enclosed and unblocked in a manipulation box, which is chosen to be transparent, and all manipulations are carried out between said initial instant except and said final instant with said watch or watch head in said manipulation box.

11. Verification method according to claim 10, wherein said at least one initial shot is taken before the insertion of said watch or watch head into said manipulation box, and said at least one final shot is taken after the extraction of said watch or watch head from said manipulation box.

12. A verification method according to any one of claims 1 to 10, wherein it is applied to a mechanical watch.