CH714484A2 - Timepiece comprising a mechanical oscillator associated with a control system. - Google Patents

Abstract

The invention relates to a timepiece (2) comprising a mechanical movement (4) with a mechanical oscillator and an electronic device for regulating the average frequency of this mechanical oscillator. It comprises an electromagnetic transducer and an electrical converter that includes a primary storage unit for supplying the control circuit. The electromagnetic transducer is arranged to provide a voltage signal having first voltage lobes in first half-cycles and second voltage lobes in second half-cycles of oscillations of the mechanical oscillator. The regulating device comprises a charge pump arranged to transfer electric charges from the primary storage unit to a secondary storage unit, these electric charges being picked up selectively according to a time drift detected in the operation of the mechanical oscillator relative to an auxiliary oscillator (32), in particular quartz.

Description

Description

Technical Field [0001] The present invention relates to a timepiece comprising a mechanical oscillator associated with a system for regulating its average frequency. The regulation is of the electronic type, that is to say that the control system comprises an electronic circuit connected to an auxiliary oscillator which is arranged to provide a high precision electric clock signal. The control system is arranged to correct any temporal drift of the mechanical oscillator relative to the auxiliary oscillator.

In particular, the mechanical oscillator comprises a mechanical resonator formed by a sprung balance and a maintenance device formed by a conventional exhaust, for example Swiss anchor. The auxiliary oscillator is formed in particular by a quartz resonator or by a resonator integrated in the electronic control circuit.

BACKGROUND [0003] Movements forming timepieces as defined in the field of the invention have been proposed in some prior documents. Patent CH 597 636, published in 1977, proposes such a movement with reference to FIG. 3. The movement is equipped with a sprung balance resonator and a conventional maintenance device comprising an anchor and an escape wheel in kinematic connection with a spring-loaded barrel. This watch movement comprises a system for regulating the frequency of the mechanical oscillator. This control system comprises an electronic circuit and an electromagnetic assembly formed of a flat coil, arranged on a support under the beam of the balance, and two magnets mounted on the balance and arranged close to each other so as to both pass over the coil when the oscillator is on.

The electronic circuit comprises a time base comprising a quartz resonator and for generating a reference frequency signal FR, this reference frequency being compared with the frequency FG of the mechanical oscillator. The detection of the frequency FG is carried out via the electrical signals generated in the coil by the pair of magnets. The control circuit is arranged to be able momentarily to generate a braking torque via a magnet-coil magnetic coupling and a switchable load connected to the coil. Document CH 597 636 gives the following teaching: "The resonator thus formed must have a variable oscillation frequency depending on the amplitude on either side of the frequency FR (isochronism defect)". It is therefore taught that a variation of the oscillation frequency of a non-isochronous resonator is obtained by varying its amplitude of oscillation. An analogy is made between the amplitude of oscillation of a resonator and the angular velocity of a generator comprising a rotor provided with magnets and arranged in a cog of the watch movement to regulate its operation. As a braking torque decreases the rotational speed of such a generator and thus its rotation frequency, it is here only considered to be able to reduce the oscillation frequency of a necessarily non-isochronous resonator by applying a torque braking decreasing its amplitude of oscillation.

To perform an electronic control of the frequency of the generator, respectively the mechanical oscillator, it is provided in one embodiment that the charge is formed by a switchable rectifier via a transistor that recharges a storage capacity during braking pulses, for recovering the electrical energy in order to supply the electronic circuit. The constant teaching given in document CH 597 636 is as follows: When FG> FR the transistor is conductive; a power Pa is then taken from the generator / oscillator. When FG <FR, the transistor is non-conductive; no more energy is drawn from the generator / oscillator. In other words, it regulates only when the frequency of the generator / oscillator is greater than the reference frequency FR. This regulation consists of braking the generator / oscillator in order to reduce its frequency FG. Thus, in the case of the mechanical oscillator, those skilled in the art understand that regulation is possible only when the mainspring is heavily armed and the free oscillation frequency (natural frequency) of the oscillator mechanical is greater than the reference frequency FR, as a result of a desired isochronism defect of the selected mechanical oscillator. There is therefore a double problem, namely the mechanical oscillator is selected for what is normally a defect in a mechanical movement and the electronic control is functional only when the natural frequency of this oscillator is greater than a nominal frequency.

The patent application EP 1 521 142 also deals with the electronic control of a sprung balance. The control system proposed in this document is similar in its general operation to that of patent CH 597 636.

The patent application EP 1 241 538 teaches that the moment of braking of the mechanical oscillator, during an alternation of any oscillation of the latter, allows either to decrease the value of the oscillation period. in progress, to increase it. To do this, there is provided an electromagnetic magnet-coil assembly and a control circuit which is arranged to make conductive or not the coils during certain time intervals. In general, a braking of the mechanical oscillator, by the generation of an electric power in the coils during a magnet-coil coupling, during a period of oscillation generates an increase in the corresponding period when this braking occurs before the passage of the mechanical resonator by its neutral point (rest position), or a decrease in the corresponding period when the braking occurs after the passage of the mechanical resonator by its neutral point.

Regarding the implementation of an electronic control taking advantage of the above-mentioned finding, the document EP 1 241 538 proposes two embodiments. In these two embodiments, there is provided a piezoelectric system associated with the exhaust to detect a tilting of its anchor in each oscillation period. With such a detection system, it is intended, on the one hand, to compare the oscillation period with a reference period, defined by a quartz oscillator, to determine whether the timepiece's running has a advance or delay and, secondly, to determine in alternate alternation the passage of the mechanical oscillator by its neutral point. In the first embodiment, according to whether the time drift corresponds to an advance or a delay, provision is made to make the coils conductive for a certain time interval respectively before or after the passage through the neutral position of the mechanical oscillator in an alternation . In other words, it is provided here to short-circuit the coils before or after the passage through the neutral position, depending on whether the regulation requires respectively an increase or a decrease in the oscillation period.

In the second embodiment, it is intended to supply the control system by periodically taking energy from the mechanical oscillator via the electromagnetic assembly. For this purpose, the coils are connected to a rectifier which is arranged to recharge a capacitor (storage capacity), which serves as a power source for the electronic circuit. The electromagnetic assembly is that given in FIGS. 2 and 4 of the document and the electronic circuit is shown schematically in FIG. 5 of this document. The only indications given for the operation of the control system are as follows: 1) the coils are made conductive during constant time intervals which are centered on respective passages of the mechanical resonator (balance spring) by its neutral position (middle position alternations); 2) during these time intervals an induced current is rectified and stored in the capacitor; and 3) during said time intervals, the period of oscillation of the balance spring can be effectively controlled by adjusting the value of the power generated by the induced current, without further details given.

It may be thought that the choice of conduction intervals of the coils centered on the neutral positions of the mechanical resonator is intended not to induce parasitic temporal drift in the mechanical oscillator by taking energy from the latter for power the electronic circuit. By making the coils conductive for the same duration before and after the passage through the neutral position, the author may think to balance the effect of a braking preceding such a passage by the neutral position with the effect of braking according to this passage so as not to change the oscillation period in the absence of a correction signal of the control circuit involved following the measurement of a time drift. It is highly doubtful that this can be achieved with the disclosed electromagnetic assembly and a conventional rectifier connected to a storage capacitor. First, the recharge of this storage capacity depends on its initial voltage at the beginning of a given time interval. Then, the induced voltage and the current induced in the coils vary in intensity with the angular speed of the sprung balance, this intensity decreasing when moving away from a neutral position where the angular velocity is maximum. The disclosed electromagnetic assembly makes it possible to determine the shape of the induced voltage / induced current signal. Although the angular position of the magnets relative to the coils for the neutral position (rest position) is not given and we can not deduce a teaching on the phase of the signal, it can be deduced that the recharge of the capacitance of Storage will normally take place for the most part before passing through the neutral position. Thus, it results in a braking which is not symmetrical relative to the neutral position and a parasitic delay in the timepiece market. Finally, as for the adjustment of the power induced during the time intervals provided to regulate the running of the timepiece, nothing is indicated. One does not understand how such an adjustment is made, no teaching being given on this subject. SUMMARY OF THE INVENTION [0011] A general objective, within the framework of the development that led to the present invention, was to produce a timepiece, comprising a mechanical movement with a mechanical oscillator and an electronic control system of this oscillator. mechanical, for which it is not necessary to initially displace the mechanical oscillator so that it advances, so as to have a timepiece that has the precision of an auxiliary electronic oscillator (in particular equipped with a resonator to quartz) when the control system is functional and, in the opposite case, the accuracy of the mechanical oscillator corresponding to its best setting. In other words, it seeks to add an electronic control to a mechanical movement also set as precisely as possible so that it remains functional, with the best possible operation, when the electronic control is not active.

The present invention has the primary objective of providing a timepiece of the type described above and which is capable of correcting a delay or an advance in the time drift of the mechanical oscillator while allowing to effectively provide a self -supply of the regulation system.

A particular objective is to provide such a timepiece that is capable, for a defined electromagnetic assembly, of providing a continuous or almost continuous power supply voltage that remains above a supply voltage which is sufficient to supply the regulation device, and this independently of the regulation of the average frequency of the mechanical oscillator, in particular of the electrical energy generated by the regulation, and thus also in the absence of correction of a drift temporal (in which case it remains weak or even nil).

Another particular objective is to ensure the self-supply of the control system without inducing a parasitic time drift, in particular in the absence of a correction of a time drift, or at least so that such a possible parasitic temporal drift remains minimal and negligible.

Another objective is to use the electrical regulation energy to supply an auxiliary function and thus an auxiliary load, effectively accumulating this electrical energy without inducing instability in the operation of the control device or disturbance of the regulation.

For this purpose, the present invention relates to a timepiece, comprising: - a mechanism, in particular a mechanism for indicating the time, - a mechanical resonator capable of oscillating around a neutral position corresponding to its state of minimum mechanical potential energy, each oscillation of the mechanical resonator defining a period of oscillation and having two successive alternations each between two extreme positions which define the amplitude of oscillation of the mechanical resonator, each alternation having a passage of the mechanical resonator by its neutral position at a median instant and consisting of a first half-alternation between an initial moment of this alternation and its median instant and a second half-alternation between this median moment and a final moment of this alternation, a mechanical resonator maintenance device forming with this mechanical resonator a bone mechanical oscillator which defines the rate of operation of said mechanism, - an electromechanical transducer arranged to be able to convert the mechanical power of the mechanical oscillator into electrical power when the mechanical resonator oscillates with an amplitude within a useful operating range, this transducer electromagnetic unit being formed by an electromagnetic assembly comprising at least one coil, mounted on one of the mechanical assembly consisting of the mechanical resonator and its support, and at least one magnet mounted on the other element of this mechanical assembly, the assembly electromagnetic device being arranged to be able to provide a voltage signal induced between the two output terminals of the electromechanical transducer at least when the mechanical resonator oscillates with an amplitude within the useful operating range, - an electrical converter connected to both terminals of the lot ie electromechanical transducer so as to be able to receive an electromechanical transducer induced electrical current, the electrical converter comprising a primary storage unit arranged to accumulate electrical energy supplied by the electromechanical transducer, the electromechanical transducer and the electrical converter forming together a braking device of the mechanical resonator, - a device for regulating the frequency of the mechanical oscillator, this regulating device comprising an auxiliary oscillator and a measuring device arranged to be able to detect an eventual time drift of the mechanical oscillator relatively to the auxiliary oscillator, the regulating device being arranged to be able to determine if the measured time drift corresponds to at least some advance.

The timepiece according to the invention is characterized in that: - the control device is arranged to also be able to determine if the measured time drift corresponds to at least a certain delay, - the braking device is arranged to whereby, in each oscillation period of the mechanical resonator when the oscillation amplitude thereof is within the useful operating range, the induced voltage signal has at least a first voltage lobe which is at least substantially active in a first half-cycle and capable of generating in this first half-wave a first induced current pulse to recharge the primary storage unit after a charge of an electric charge thereof and at least one second lobe of voltage intervening at least for the most part in a second half-cycle and likely to generate in this second half-alternation u a second induced current pulse for recharging the primary storage unit after taking an electrical charge therefrom, the induced voltage signal thus having a plurality of such first voltage lobes and a plurality of such second lobes the regulating device comprises a charge pump device arranged to be able to transfer on command a certain electric charge of the primary storage unit to a secondary storage unit; the regulating device further comprises a control logic circuit which receives as input a measurement signal supplied by the measuring device and which is arranged to be able to activate the charge pump device so that it performs, when the measured time drift corresponds to said at least some advance, a transfer of a first electrical charge of the primary storage unit in the unit of secondary accumulation so that a recharge of the primary storage unit, following this transfer of the first electric charge, is generated for the most part by at least a first voltage lobe among said plurality of first voltage lobes, the a control logic circuit being further arranged to enable the charge pump device to perform, when the measured time drift corresponds to the at least some delay, a transfer of a second electrical charge of the charging unit. primary accumulation in the secondary storage unit so that a recharge of the primary storage unit, following this transfer of the second electric charge, is generated for the most part by at least one second voltage lobe among said plurality of second voltage lobes.

By "voltage lobe" is meant a voltage pulse which is situated entirely above or entirely below a zero value (defining a zero voltage), that is to say a variation of voltage in a certain time interval with either a positive voltage whose positive value goes up then down, or a negative voltage whose negative value goes down then up.

The transfer of a first electrical charge in a first time zone as defined is intended to increase the recharge of the power capacity when the appearance of a first voltage lobe following this transfer, relative to the case. where no transfer would take place. This increase of the recharge means a greater mechanical energy taken by the mechanical oscillator by the braking system and thus a higher braking of this mechanical oscillator. As will be explained later, a braking in a first half-wave before the passage of the mechanical resonator by its neutral position generates a negative time phase in the oscillation of the resonator, and thus the duration of the alternation in question is increased . Thus, the instantaneous frequency of the mechanical oscillator is temporarily decreased and a certain delay in the operation of the mechanism which at least partially corrects the advance detected by the measuring device results. Likewise, the transfer of a second electric charge in a second time zone as defined is intended to increase the recharging of the supply capacitance when a second voltage lobe appears according to this sampling, relative to the case where no picking would take place. As will be understood later, this generates a positive temporal phase shift in the oscillation of the resonator, and thus the duration of the alternation in question is decreased. Thus, the instantaneous frequency of the mechanical oscillator is increased momentarily and this results in a certain advance in the operation of the mechanism which at least partially corrects the delay detected by the measuring device.

In a main embodiment, the timepiece comprises a main load connected or likely to be regularly connected to the electrical converter to be powered by the primary storage unit, the main load including the device of regulation.

In an advantageous embodiment, the timepiece comprises an auxiliary load connected to or capable of being connected intermittently to the secondary accumulation unit so as to be powered by the secondary accumulation unit.

In a preferred embodiment, the charge pump device is arranged to form a voltage booster which is arranged so that an auxiliary supply voltage across the secondary storage unit is greater than one. main supply voltage across the primary storage unit.

In a particular embodiment, the control device comprises at least one dissipative circuit for dissipating electrical energy accumulated in the primary storage unit, at least one switch associated with the dissipative circuit to be able to momentarily connect this dissipative circuit to the primary storage unit and a measuring circuit arranged to detect whether the voltage across the secondary storage unit is greater than a first voltage limit or the filling level of the storage unit. secondary accumulation is greater than a first filling limit. Then, the control logic circuit is arranged so that, when the voltage at the terminals of the secondary storage unit is greater than the first voltage or filling limit, momentarily connect the at least one dissipative circuit to the primary accumulation unit so as to perform, when the measured time drift corresponds to said at least some advance, a first dissipative discharge of the primary accumulation unit so that a recharge thereof, following this first discharge, is generated mainly by at least a first voltage lobe among said plurality of first voltage lobes, and so to perform, when the measured time drift corresponds to said at least a certain delay, a second discharge of the primary accumulation unit so that a recharge of it, following this second discharge, is generated in major part e by at least one second voltage lobe among said plurality of second voltage lobes.

In a particular variant of the advantageous embodiment mentioned above, the timepiece further comprises a measuring circuit arranged to detect whether the voltage across the secondary storage unit is less than a second limit of voltage (less than the first voltage limit mentioned above) or if the fill level of the secondary storage unit is less than a second fill limit (less than the first fill limit mentioned above). Then, the control logic circuit is arranged so that when the voltage at the terminals of the secondary storage unit is lower than the second voltage or filling limit and when the measured time drift is between said at least delay and said at least some advance, activate the charge pump device so that it transfers a third electrical charge from the primary accumulation unit to the secondary accumulation unit, whereby a recharge of the primary storage unit following this transfer of a third electrical charge is generated for the most part by at least a first voltage lobe among said plurality of first voltage lobes, and a transfer of a fourth electrical charge of the primary storage unit in the secondary storage unit, so that a recharge of the ac unit primary cumulation following this transfer of a fourth electrical charge is generated for the most part by at least one second voltage lobe among said plurality of second voltage lobes, the fourth electric charge being substantially equal to the third electric charge.

BRIEF DESCRIPTION OF THE DRAWINGS The invention will be described hereinafter in more detail with the aid of appended drawings, given by way of non-limiting examples, in which: FIG. 1 is a general view from above of a first embodiment of a timepiece according to the invention, FIG. 2 is a partial and enlarged view of the timepiece of FIG. 1, showing the tromagnetic electromagnetic assembly forming an electromagnetic transducer of a control system incorporated in this timepiece, FIG. 3 represents, for an electromagnetic assembly given in FIGS. 4A to 4C which corresponds to the first embodiment, the voltage induced in the coil of this electromagnetic assembly when the sprung balance oscillates and the application of a first braking pulse in a certain alternation before the sprung balance passes by its neutral position, as well as the angular speed of the beam and its angular position in a time interval in which the first braking pulse occurs, FIGS. 4A to 4C show, for the electromagnetic transducer considered in FIG. 3, the pendulum at three particular instants of an alternation of the mechanical oscillator during which the first braking pulse is provided, FIG. 5 is a figure similar to that of FIG. 3 with the application of a second braking pulse in a certain alternation after the sprung balance has passed through its neutral position, FIGS. 6A to 6C show the pendulum at three particular moments of an alternation of the mechanical oscillator during which the second braking pulse is provided, FIG. 7 shows the circuit diagram of an electric converter and a device for regulating the mechanical oscillator provided in the first embodiment of a timepiece, FIG. 8 shows the electronic circuit of a charge pump variant forming the control device shown in FIG. 7, FIG. 9 is a flowchart of a mode of regulating the running of the timepiece according to the first embodiment, FIGS. 10A to 10C represent various electrical signals involved in the electrical diagram of FIG. 7, FIG. 11 is a partial view of a second embodiment of a timepiece according to the invention, showing the particular arrangement of its electromagnetic transducer, FIG. 12 shows the electrical diagram of the electric converter and the mechanical oscillator control device as arranged in the second embodiment of a timepiece according to the invention, FIG. 13 is a flowchart of a mode of regulation of the walking of the timepiece according to the second embodiment, FIG. 14 shows various electrical signals involved in the electrical diagram of FIG. 12 in the case of a correction of an advance observed in the measured time drift, FIG. 15 shows various electrical signals involved in the electrical diagram of FIG. 12 in the case of a correction of a delay observed in the measured time drift, FIG. 16 is a partial view of a third embodiment of a timepiece according to the invention, showing the particular arrangement of its electromagnetic transducer, FIG. 17 shows the electrical diagram of the electric converter and the mechanical oscillator control device as arranged in the third embodiment of a timepiece according to the invention, FIG. 18 shows the electronic circuit of a charge pump variant forming the voltage booster of the control device shown in FIG. 17, fig. 19 is a flowchart of a mode of regulation of the running of the timepiece according to the third embodiment, FIGS. 20A to 20C represent various electrical signals involved in the electrical diagram of FIG. 17, and FIGS. 21 and 22 show an advantageous embodiment of a mechanical resonator associated with an electromagnetic appearance of the timepiece according to the invention.

Detailed Description of the Invention [0026] Referring to FIGS. 1 and 2, a timepiece object of the present invention will be described below. Fig. 1 is a partial plan view of a timepiece 2 comprising a mechanical movement 4, equipped with a mechanical resonator 6, and a control system 8. The maintenance means 10 of the mechanical resonator are conventional. They include a barrel 12 with a motor spring, an exhaust 14 formed of an escape wheel and a pallet anchor, and an intermediate gear 16 kinematically connecting the barrel to the escape wheel. The resonator 6 comprises a rocker 18 and a conventional coil spring, the rocker being pivotally mounted about an axis of rotation 20 between a plate and a bridge. The mechanical resonator 6 and the maintenance means 10 (also called excitation means) together form a mechanical oscillator. It will be noted that, in general, only the escapement as maintenance means / excitation means of this mechanical oscillator, the energy source and an intermediate gear train are included in the definition of a mechanical clock oscillator. being considered separately. The sprung balance oscillates about the axis 20 when it receives mechanical impulses from the exhaust whose escape wheel is driven by the barrel. The wheel 16 is part of a mechanism of the watch movement whose running is clocked by the mechanical oscillator. This mechanism comprises, in addition to the wheel 16, other mobiles and analog indicators (not shown) kinematically connected to the wheel 16, the displacement of these analog indicators being paced by the mechanical oscillator. Various mechanisms known to those skilled in the art can be provided.

FIG. 2 is a partial view of FIG. 1, in horizontal section at the rocker 18, showing a magnet 22 and a coil 28 forming an electromagnetic assembly 27 according to the invention. The coil 28 is preferably of the wafer type (disk shape having a relatively small thickness). It is arranged on the turntable of the watch movement and conventionally comprises two connection ends E1 and E2. In general, the electromagnetic assembly comprises at least one coil and a magnetic structure formed of at least one magnet generating a magnetic flux, in the direction of a general plane of the coil, which passes therethrough when the The mechanical resonator oscillates with an amplitude within a useful operating range. In the example shown, the rocker 18 carries, preferably in an area located near its outer diameter defined by its serge, the bipolar magnet 22 which has an axially oriented axis of magnetization. It should be noted that it is preferable to confine the magnetic flux of the magnet or of the magnets carried by the balance beam by means of a shield formed by parts of the balance, in particular by magnetic parts arranged on both sides of the beam. magnet in the axial direction so that the coil is partially located between these two magnetic parts.

The rocker arm 18 defines a half-axis 24, from its axis of rotation 20 and perpendicular to the latter which passes in the center of the magnet 22. When the balance-spring is in its rest position, the half-axis 24 defines a neutral position (angular rest position of the sprung balance corresponding to a zero angle) around which the sprung balance can oscillate at a certain frequency, in particular at a free frequency FO corresponding to the natural oscillation frequency of the mechanical oscillator, ie not subjected to external force torques (other than the periodically supplied via the exhaust). In fig. 2, the mechanical resonator 6 (shown without its spiral which is located above the section plane) is shown in its neutral position, corresponding to its minimum potential mechanical energy state. Note that, in the neutral position, the half-axis 24 defines a reference half-axis 48 which is angularly offset by an angle θ relative to the fixed half-axis 50 which perpendicularly intercepts the axis of rotation 20 and the central axis of the coil 28. In other words, in projection in the general plane of the balance, the center of the coil 28 has an angular offset θ relative to the reference half-axis 48. In FIG. 2, this angular offset is 120 ° in absolute value. Preferably, the angular offset θ is between 30 ° and 120 ° in absolute value.

Each oscillation of the mechanical resonator defines a period of oscillation and has a first alternation followed by a second alternation each between two extreme positions defining the oscillation amplitude of the mechanical resonator (note that we consider here the oscillating resonator and therefore the mechanical oscillator as a whole, the oscillation amplitude of the balance-spring being defined among other things by the maintenance means). Each alternation has a passage of the mechanical resonator by its neutral position at a median time and a certain duration between an initial moment and a final instant which are respectively defined by the two extreme positions occupied by the mechanical resonator respectively at the beginning and at the end of this alternation. Each alternation thus consists of a first half-cycle ending at said median instant and a second half-wave alternating at this median instant.

The system 8 for regulating the frequency of the mechanical oscillator comprises an electronic circuit 30 and an auxiliary oscillator 32, this auxiliary oscillator comprising a clock circuit and for example a quartz resonator connected to this clock circuit . Note that in a variant, the auxiliary oscillator is integrated at least partially in the electronic circuit. The control system further comprises the electromagnetic assembly 27 described above, namely the coil 28 which is electrically connected to the electronic circuit 30 and the bipolar magnet 22 mounted on the balance. Advantageously, the various elements of the regulation system 8, with the exception of the magnet, are arranged on a support 34 with which they form a module independent of the watch movement. Thus, this module can be assembled or associated with the mechanical movement 4 that when they are mounted in a watch case. In particular, as shown in FIG. 1, the aforementioned module is attached to a casing ring 36 which surrounds the watch movement. It is understood that the control module can be associated with the watch movement once the latter fully assembled and adjusted, the assembly and disassembly of this module can occur without having to intervene on the mechanical movement itself.

[0031] Referring to FIGS. 3 to 6C, we first describe the physical phenomenon on which the regulation principle implemented in the timepiece according to the invention is based. We consider here a timepiece similar to that of fig. 1. The mechanical resonator 40, of which only the rocker 42 has been shown in FIGS. 4A-4C and 6A-6C, carries a single bipolar magnet 44 whose magnetization axis is substantially parallel to the axis of rotation 20 of the balance, that is to say with an axial orientation. In this case, the half-axis considered 46 of the mechanical resonator 40 passes through the center of rotation 20 and the center of the magnet 44. In the example discussed here, the angle θ between the reference half-axis 48 and the half-axis 50 has a value of about 90 °. The two half-axes 48 and 50 are fixed relative to the watch movement, whereas the half-axis 46 oscillates with the balance and gives the angular position β of the magnet mounted on this balance relative to the reference half-axis, the latter defining the zero angular position for the mechanical resonator. More generally, the angular offset θ is such that an induced voltage signal generated in the coil at the passage of the magnet facing said coil is located, during a first alternation of any oscillation, before the passage of the mid-axis median by the reference half-axis (thus in a first half-cycle) and, during a second alternation of any oscillation, after the passage of this median half-axis by the reference half-axis (so in a second half-alternation).

FIG. 3 shows four graphs. The first graph gives the voltage in the coil 28 as a function of time when the resonator 40 oscillates, that is to say when the mechanical oscillator is activated. The second graph indicates the time tP1 at which a braking pulse is applied to the resonator 40 to make a correction in the operation of the mechanism clocked by the mechanical oscillator. The instant of the application of a pulse of rectangular shape (that is to say of a binary signal) is considered here as the temporal position of the middle of this pulse. There is a variation of the oscillation period during which the braking pulse occurs and thus a point variation of the frequency of the mechanical oscillator. In fact, as seen in the last two graphs of FIG. 3, which respectively show the angular velocity (values in radians per second: [rad / s]) and the angular position (values in radian: [rad]) of the pendulum over time, the temporal variation relates to the only alternation during from which intervenes the braking pulse. It will be noted that each oscillation has two successive alternations which are defined in the present text as the two half-periods during which the rocker is respectively subjected to an oscillation movement in one direction and then an oscillation movement in the other direction. In other words, as already explained, an alternation corresponds to a rocking of the rocker in one direction or the other direction between its two extreme positions defining the amplitude of oscillation.

By braking pulse, it includes an application, substantially during a limited period of time, a certain force torque to the mechanical resonator for braking, that is to say a torque force that s opposes the oscillation motion of this mechanical resonator. In general, the braking torque can be of various natures, in particular magnetic, electrostatic or mechanical. In the embodiment described, the braking torque is obtained by the magnet-coil coupling and therefore corresponds to a magnetic braking torque exerted on the magnet 44 via the coil 28 which is controlled by a control device. Such braking pulses may for example be generated by momentarily short-circuiting the coil. This action is recognizable on the graph of the coil voltage in the time zone during which the braking pulse is applied, this time zone being provided at the occurrence of a voltage pulse induced in the coil by the passage of the magnet. It is obviously in this time zone that the magnet-coil coupling allows a contactless action via a magnetic torque on the magnet fixed to the balance. It is observed that the coil voltage goes down to zero during a short-circuit braking pulse (the voltage induced in the coil 28 by the magnet 44 being shown in broken lines in the aforementioned time zone). Note that the short-circuit braking pulses shown in FIGS. 3 and 5 are mentioned here in the context of the explanations given, because the present invention provides a recovery of the braking energy to supply in particular the control device.

In figs. 3 and 5, the oscillation period TO corresponds to a "free" oscillation (that is to say without application of regulation pulses) of the mechanical oscillator. Each of the two alternations of an oscillation period has a duration TO / 2 without disturbance or external stress (in particular by a regulation pulse). The time t = 0 marks the beginning of a first alternation. It will be noted that the "free" frequency FO of the mechanical oscillator is here approximately equal to four Hertz (FO = 4 Hz), so that the period TO = approximately 250 ms.

[0035] With reference to FIGS. 3 and 4A-4C, we will describe the behavior of the mechanical oscillator in a first case. After a first period TO begins a new period T1, respectively a new alternation A1 during which occurs a braking pulse P1. At the initial moment you begin the alternation A1, the resonator 40 then being in the state of FIG. 4A where the magnet 44 occupies an angular position β corresponding to an extreme position (maximum positive angular position Am). Then comes the braking pulse P1 at time tP1 which is located before the median time tNi at which the resonator passes through its neutral position, FIGS. 4b, 4C representing the resonator respectively at the two successive instants tP1 and tN1. Finally the alternation A1 ends at the final time tF1.

In the first case, the braking pulse is generated between the beginning of an alternation and the passage of the resonator by its neutral position, that is to say in a first half-wave of this alternation. As expected, the angular velocity in absolute value decreases at the moment of the braking pulse P1. This induces a negative temporal phase shift TC1 in the oscillation of the resonator, as shown by the two graphs of the angular velocity and the angular position in FIG. 3, a delay relative to the undisturbed theoretical signal (shown in broken lines). Thus, the duration of the alternation A1 is increased by a time interval Τ0Ί. The oscillation period T1, comprising the alternation A1, is therefore extended relative to the value TO. This causes a specific decrease in the frequency of the mechanical oscillator and a momentary slowing of the operation of the associated mechanism.

[0037] Referring to FIGS. 5 and 6A-6C, we will describe the behavior of the mechanical oscillator in a second case. The graphs of fig. 5 represent the temporal evolution of the same variables as in fig. 3. After a first period TO begins a new period T2, respectively an alternation A2 during which a braking pulse P2 occurs. At the initial time t02 begins the alternation A2, the resonator 40 then being in an extreme position (maximum negative angular position not shown). After a quarter period (TO / 4) corresponding to a first half-half cycle, the resonator reaches its neutral position at the median time tN2 (configuration shown in Fig. 6A). Then comes the braking pulse P2 at the instant tP2 which is located after the median moment tN2 at which the resonator passes through its neutral position in the alternation A2, that is to say in a second half-cycle of this alternation. Finally, this alternation ends at the final time tF2 at which the resonator again occupies an extreme position (maximum positive angular position). Figs. 6B and 6C represent the resonator respectively at the two successive instants tN2 and tP2. It will be noted in particular that the configuration of FIG. 6A differs from the configuration of FIG. 4C by the inverse directions of the respective oscillation movements. Indeed, in fig. 4C, the pendulum rotates in a clockwise direction when it passes through the neutral position in the alternation A1, whereas in FIG. 6A this balance rotates counterclockwise when passing through the neutral position in alternation A2.

In the second case considered, the braking pulse is generated, alternately, between the median instant at which the resonator passes through its neutral position and the final instant at which this alternation ends. As expected, the angular speed in absolute value decreases at the moment of the braking pulse P2. Remarkably, the braking pulse here induces a positive phase shift TC2 in the oscillator period of the resonator, as shown by the two graphs of the angular velocity and the angular position in FIG. 5, an advance relative to the undisturbed theoretical signal (shown in broken lines). Thus, the duration of the alternation A2 is reduced by the time interval TC2. The oscillation period T2 comprising the alternation A2 is therefore shorter than the value TO. This therefore generates a "punctual" increase in the instantaneous frequency of the mechanical oscillator and a momentary acceleration of the operation of the associated mechanism.

With reference to FIGS. 1 and 2 already described and in FIGS. 7 to 10C, will be described hereinafter a first embodiment of a timepiece according to the invention. This timepiece 2 comprises: a mechanism 12, 16 (shown partially); a mechanical resonator 6 (spiral balance) capable of oscillating around a neutral position 48 corresponding to its state of minimal mechanical potential energy each alternation of successive oscillations having a passage of the mechanical resonator by its neutral position at a median instant and consisting of a first half-cycle ending at its median instant and a second half-cycle beginning at its median instant, - A maintenance device 14 of the mechanical resonator forming with this mechanical resonator a mechanical oscillator that speeds the operation of the mechanism, - an electromechanical transducer arranged to be able to convert the mechanical power of the mechanical oscillator into electrical power when the mechanical resonator 6 oscillates with an amplitude within a useful operating range, this tra an electromagnetic transducer being formed by an electromagnetic assembly 27 comprising a coil 28 (the only element of the electromagnetic assembly shown schematically in FIG. 7), mounted on the support (in particular the plate of the movement 4) of the mechanical resonator, and a magnet 22 mounted on this mechanical resonator, the electromagnetic assembly 27 being arranged so as to be able to provide an induced voltage signal U, ( t) (Fig. 10A), between the two output terminals E1 and E2 of the electromechanical transducer when the mechanical resonator oscillates with an amplitude within the useful operating range, - an electrical converter 56 connected to the two output terminals of the electromechanical transducer in order to be able to receive from this electromechanical transducer an induced electric current IREC (FIG 10B), this electrical converter comprising a supply capacitor Cal arranged to accumulate electrical energy supplied by the electromechanical transducer, this electromechanical transducer and the converter together forming a braking device of the mechanical resonator, - a control device 52 of the frequency of the mechanical oscillator, this control device comprising an auxiliary oscillator 58 &amp; CLK and a measuring device arranged to be able to measure an eventual time drift of the mechanical oscillator relative to the auxiliary oscillator, the control device being arranged to be able to determine if the measured time drift corresponds to at least some advance or to least a certain delay.

Preferably, the electromagnetic assembly 27 also partly forms the measuring device. This measuring device furthermore comprises a bidirectional counter CB and a comparator 64 (of the Schmidt trigger type). The comparator receives at one input the induced voltage signal U, (t) and at the other input a threshold voltage signal Uth whose value is positive in the example given. Since the induced voltage signal U, (t) has two positive lobes in each oscillation period of the resonator 6 (FIG 10A) exceeding the value Uth, the comparator outputs a signal Comp "having two pulses S1 and S2 (FIG. Fig. 10C) per oscillation period. This "Comp" signal is supplied on the one hand to a control logic circuit 62 and on the other hand to a flip-flop 66 which inhibits every other pulse so as to provide a single pulse per oscillation period to a first input " UP "bidirectional counter CB. The bidirectional counter includes a second "Down" input which receives a clock signal ShOr at a nominal frequency / frequency for the oscillation frequency, this clock signal being derived from the auxiliary oscillator which provides a digital reference signal defining a reference frequency. The auxiliary oscillator comprises a clock circuit CLK for exciting the crystal resonator 58 and for providing in return the reference signal which is composed of a succession of pulses respectively corresponding to the oscillation periods of the quartz resonator.

The clock circuit provides its reference signal to a divider DIVI &amp; DIV2 which divides the number of pulses in this reference signal by the ratio between the nominal period of the mechanical oscillator and the nominal reference period of the auxiliary oscillator. The divider thus provides a clock signal Shor defining a setpoint frequency (for example 4 Hz) and having a pulse per set period (for example 250 msec) on the counter CB. Thus, the state of the counter CB determines the advance (if the number is positive) or the delay (if the number is negative) accumulated over time by the mechanical oscillator relative to the auxiliary oscillator with a resolution corresponding substantially to a set period. The state of the counter is supplied to a control logic circuit 62 which is arranged to determine whether this state corresponds to at least some advance (CB> N1, where N1 is a natural number) or at least some delay (CB < - N2, N2 being a natural number).

The electric converter 56 comprises an electrical energy storage circuit D1 &amp; CAl which is arranged, in the variant described, to be able to recharge the power supply CAL only with a positive voltage input of the electric converter, that is to say only with a positive induced voltage supplied by the coil 28. This feed capacity forms here alone a primary accumulation unit. When recharging the supply capacity, the amount of electrical energy supplied by the braking device to this power supply capacity increases as the level of voltage of this supply capacity is low. A main load is connected or capable of being regularly connected to the electric converter 56 and powered by the supply capacity which provides the main supply voltage UAL (t), shown in FIG. 10A, between the two power supply terminals VDD and VSs, this main load comprising in particular the control circuit 54.

The timepiece 2 is remarkable in that the control circuit 54 of the regulating device comprises a charge pump 60 arranged to be able to transfer on command a certain electrical charge of the power supply CAL in a unit. secondary accumulation formed here of CAux capacity. This CAux capacity is provided as a secondary power source for an auxiliary load, for example a light diode, an RFID circuit, a temperature sensor, or another electronic unit that can be incorporated into the timepiece according to the invention. For this purpose, the capacitance CAux has at its two terminals respectively a lower potential VL and a higher potential VH defining an auxiliary supply voltage. An alternative embodiment of such a charge pump is shown in FIG. 8. This is a simple form of charge pump that only transfers charges without voltage increase so that in this case the auxiliary supply voltage is lower than the main supply voltage supplied by the electric converter 56. It will be noted that this is a special case which is not preferred. Other load pump variants known to those skilled in the art can be provided, especially those which have a voltage booster function. Such a variant will be described later in the third embodiment. The charge pump 60 comprises an input switch Sw1 and an output switch Sw2 with a transfer capacitance CTr- The switches Sw1 and Sw2 are controlled by the control logic circuit 62 according to a control method (FIG. in the first embodiment of the timepiece according to the invention and which will be described later.

In figs. 10A and 10B, the induced voltage signal Ui (t) corresponds to that generated by the electromagnetic assembly 27 associated with the mechanical resonator 6 when the latter resonates in a useful operating range. On the time axis [t] are indicated the median moments TNn, n = 0, 1, 2, ..., corresponding to the successive passages of the mechanical resonator by its neutral position, as well as the instants TMn, n = 0, 1,2, ..., corresponding to the successive passages of the mechanical resonator alternately by its two extreme positions where its angular velocity is zero and the direction of its sway is reversed. According to the invention, the braking device 27 &amp; 56 is arranged so that, in each oscillation period of the mechanical resonator 6 at least when the oscillation amplitude of this mechanical resonator is in the useful operating range, the induced voltage signal Ui (t) presents a first voltage lobe LU-i intervening in a first half-wave DA11, DA1P and a second voltage lobe LU2 intervening in a second half-wave DA21, DA2P. The induced voltage signal thus has alternately a succession of first voltage lobes LUi and second voltage lobes LU2. Each first voltage lobe LUi has a first maximum value UM-i at a first instant L of the corresponding first half-cycle and each second voltage lobe LU2 has a second maximum value UM2 at a second instant h of the second half-cycle. corresponding.

The first and second voltage lobes define, on the one hand, first time zones ZT1 each located before the first moment b of a first different voltage lobe and after the second instant t2 of the second voltage lobe preceding this second time lobe. first voltage lobe and, secondly, the second time zones ZT2 each located before the second time t2 of a second different voltage lobe and after the first moment b of the first voltage lobe preceding this second voltage lobe. The first voltage lobes LLb generate pulses S1 in the "Comp" signal at the output of the comparator 64, whereas the second voltage lobes LU2 generate pulses S2 in this "Comp" signal (FIG. In the variant shown in FIG. 10A, the lobes considered for the generation of signals S1 and S2 are the positive voltage lobes because the threshold voltage Uth has been chosen positive. In a variant which will not be described in more detail later, it is possible to choose a negative threshold voltage supplied to the "+" input of the comparator 64 (the induced voltage signal then being supplied at its "-" input). and it is then the negative voltage lobes that generate the signals S1 and S2.

Next, the braking device is arranged so that, at least when no time drift is detected by the measuring device and at least when said main load, connected to terminals Vss and VDd, consumes continuously or in a quasi-continuous manner of the electrical energy accumulated in the supply capacitor Cal (during a normal operation phase of the timepiece, as shown in FIG 10A where the supply voltage UAL (t) has a certain negative slope in the absence of correction of the operation of the mechanical oscillator), the first voltage lobes LLb and the second voltage lobes LU2 alternately generate induced current pulses P1 and P2 (FIG. the feeding capacity. It will be noted that the electrical converter 56 comprises a diode D1 arranged so that only the positive voltage lobes are capable of reloading the capacitance CAL. However, in a variant that will not be described in more detail later, the electrical converter may have a diode arranged to define a single-wave rectifier so that it is then the negative voltage lobes that are likely to recharge the CAL capacity. In this case, it is thus the negative voltage lobes that generate the induced current pulses and are considered to determine the sampling time zones of a certain electric charge as a function of the measured time drift, as explained below. It should be noted that in another variant, the converter may comprise a full-wave converter. In this case, there is obtained at each passage of the magnet 22 in front of the coil 28 a first pair of consecutive first voltage lobes or a second pair of two consecutive second voltage lobes all having substantially the same amplitude. Duplicates of the first second voltage lobes described above are thus obtained. This particular case should be treated in connection with the above discussion by taking the first and second pairs of voltage lobes instead of the first and second voltage lobes, and taking to determine the first and second ZT time zones. 1 and ZT2 the times b and t2 of the two adjacent lobes from two successive pairs.

The charge pump 60 is arranged to be able to take a certain electrical charge from the power supply capacitance CAL and transfer it into the auxiliary capacitance CAux, so as to momentarily reduce the voltage level UAL (t) of the CAL power capacity. As soon as the supply capacity has been sufficiently charged to be able to supply the control circuit 54, the control logic circuit 62 receives as input a measurement signal supplied by the measuring device, namely the bidirectional counter CB. This control logic circuit is arranged to activate the charge pump 60 so that, when the measured time drift corresponds to at least some advance (CB> N1), a sampling of a first electrical charge of the capacitance power supply CAL in a first time zone ZT 1 and a transfer of this first load in the auxiliary capacity which forms a secondary power source. This results in a drop in the voltage UAL (t). Similarly, the control logic circuit is arranged to activate the charge pump 60 so that, when the measured time drift corresponds to at least a certain delay (CB <- N2), a charge of a second charge electrical power supply capacity Cal in a second time zone ZT2, to lower the voltage UAL (t), and a transfer of this second electric charge in the auxiliary capacity.

The control method implemented in the first embodiment of the invention is given in flow chart form in FIG. 9. After initialization of the control circuit at "POR", the counter CB is reset. Next, the detection of a rising edge of a pulse S1 or S2 supplied by the comparator 64 in the "Comp" signal (see Fig. 10C) that it transmits to the control logic circuit 62 is then initiated. the time counter CT. Next, the next rising edge is expected to be detected in the "Comp" signal (the second rising edge of a pulse S2 or S1).

As soon as the above-mentioned second rising edge is detected in the signal "Comp", the logic circuit 62 transfers the state / value of the time counter CT into a register and compares this value with a differentiation value Tdiff which is selected lower. at a first time interval between a first pulse S1 and a second pulse S2 and greater than a second time interval between a second pulse S2 and a first pulse S1. As soon as the state of the time counter CT is transferred into the register, this time counter is reset and a timer associated with the logic circuit 62 is triggered to measure a certain delay whose value TC1 or TD1 is selected according to the result of the comparison. the value of the CT counter with the value Tdiff. In the first embodiment, the regulating device therefore comprises a detection device, arranged to be able to detect the successive appearance alternately of first voltage lobes and second voltage lobes, and a time counter CT associated with the control logic circuit. 62 to allow the latter to distinguish a first time interval, separating a first voltage lobe from a second voltage lobe that follows, and a second time interval separating a second voltage lobe from a first voltage lobe which follows, the first and second time intervals being different due to the arrangement of the electromagnetic assembly.

The arrangement of the electromagnetic assembly is provided here such that the curve of the induced voltage signal Ui (t) has two voltage lobes LU2 and LU-i, with the same maximum amplitude (UM2 = UM-i). , which occur in a second half-wave and in the first half-wave next, but no voltage lobe of substantially the same amplitude is generated in the following two half cycles. The curve of the induced voltage signal Ui (t) shown in FIG. 10A follows from the electromagnetic assembly 27 described above. In the first embodiment, the coil 28 has at its center an angular offset 9 relative to the reference half-axis 48 (FIG 2, angular position of the magnet 22 when the mechanical resonator 6 is in its rest position) so as to generate in each oscillation period of the mechanical resonator, in said useful operating range, only two voltage lobes of the same polarity and of substantially the same maximum amplitude which intervene in two consecutive half-cycles and which respectively form one of said second semiconductors voltage lobes and one of said first voltage lobes. Preferably, the angular offset θ is between 30 ° and 120 ° in absolute value.

In the aforementioned comparison between the value of the time counter CT and the differentiation value Tdiff, the timer associated with the logic circuit waits either a delay TCi when the value of the time counter CT is greater than the differentiation value Tdiff, or a delay TD1 when the value of the time counter CT is smaller than the differentiation value Tdiff. In the first case, the comparison makes it possible to know that the detected pulse is a pulse S2 generated by a second voltage lobe LU2 and the delay TCi is chosen so that it ends in a first time zone ZT1 following this second voltage lobe. . In the second case, the comparison makes it possible to know that the detected pulse is a pulse S1 generated by a first voltage lobe LUi and the delay TDi is chosen so that it ends in a second time zone ZT2 following this first voltage lobe. . In general, the regulating device comprises a timer associated with the control logic circuit to allow the latter to activate, if necessary, the charge pump device after a first determined time since the detection of a second lobe of voltage, this first delay being selected so that it ends in a first time zone, or after a second determined time since the detection of a first voltage lobe, this second delay being selected so that it ends in a second time zone.

In the first case mentioned above, when the delay TCi is reached, it is detected whether the counter CB, indicating an eventual time drift of the mechanical oscillator, has a value greater than a given natural number N1 (number positive or equal to zero). If this is the case, the mechanical oscillator has an advance relative to the auxiliary oscillator. To correct such an advance, it is provided according to the invention to transfer a first electrical charge of the supply capacity in the auxiliary capacity at the end of the aforementioned delay TCi and therefore in the corresponding first time zone ZT1. The resulting decrease of the supply voltage UAi_ (t) (indicated by the reference PC-ι in FIG 10A) generates, during the appearance of the first voltage lobe following the above-mentioned transfer, a current pulse. induced P1pc having an amplitude greater than that of a pulse P1 which would occur in the absence of activation of the charge pump. This increase in the induced current in the coil 28 means a greater mechanical energy taken from the mechanical oscillator by the braking device in a first half-wave DA1P. As already stated, a braking in a first half-wave generates a negative phase shift in the oscillation of the mechanical resonator 6, and thus the duration of the half-wave in question is increased. By the more intense braking carried out in the first half-alternation DA1P, the momentary frequency of the mechanical oscillator is momentarily diminished and this results in a certain delay in the running of the mechanism that it is clocking, which corrects at least partially the advance detected by the measuring device.

In the second case mentioned above, when the delay TD1 is reached, it is detected whether the counter CB has a value lower than a given negative number - N2, N2 being a natural number. If this is the case, the mechanical oscillator has a delay relative to the auxiliary oscillator. To correct such a delay, it is provided according to the invention to transfer a second electrical load of the supply capacity in the auxiliary capacity at the end of the aforementioned delay TDi and therefore in the corresponding second time zone ZT2. The resulting decrease of the supply voltage UAi_ (t) (indicated by the reference PC2 in Fig. 10A) generates, during the appearance of the second voltage lobe following the above-mentioned transfer, an induced current pulse P2pc. having an amplitude greater than that of the pulse P2 which would occur in the absence of regulation. This increase in the induced current in the coil 28 means a greater mechanical energy taken from the mechanical oscillator by the braking device in a second half-wave DA2P. As already stated, a braking in a second half-wave generates a positive phase shift in the oscillation of the mechanical resonator, and thus the duration of the half-wave in question is decreased. By the more intense braking carried out in the second half-alternation DA2P, momentary increase of the instantaneous frequency of the mechanical oscillator is obtained and it results a certain advance in the running of the mechanism that it runs, which corrects at least partially the delay detected by the measuring device.

A sampling of an electrical charge in a first time zone ZT1 at the end of the delay TCi, indicated by the reference PCi which points a downward step in the supply voltage UalÎO, therefore generates an induced current pulse P1pc d higher amplitude in a first half-alternation DA1P of an alternation A2, this first half-alternation having a duration greater than those of the second half-alternations DA1 ° and DA11 which respectively correspond to a half-half cycle during which no pulse induced current is generated and at a half-wave in which a compensation pulse P1 of the power consumption of the main load occurs. A sampling of an electric charge in a second time zone ZT2 to Latin of the delay TD1, indicated by the reference PC2 which points a downward step in the supply voltage UAL (t), thus generates an induced current pulse P2pc of greater amplitude in a second half-alternation DA2P of an alternation A1, this second half-alternation having a shorter duration than the second half-alternations DA2 ° and DA21 which respectively correspond to a half-half cycle in which no current pulse induced is generated and a half-wave in which a compensation pulse P2 of the power consumption of the main load occurs.

With the aid of FIGS. 11 to 15, there will be described hereinafter a second embodiment of a timepiece according to the invention.

FIG. 11 is similar to FIG. 2, but for an electromagnetic assembly 29 forming the electromagnetic transducer of a timepiece according to the second embodiment. It shows the mechanical resonator 6a in horizontal section at its balance 18a, this mechanical resonator being incorporated in a watch movement, similar to that of FIG. 1, in place of the resonator 6 shown in this FIG. 1. The references already described will not be described again here. In general, there is provided an electromagnetic assembly which comprises at least the coil 28 and a magnetic structure formed of at least one magnet and having at least one pair of magnetic poles of opposite polarities, each generating a magnetic flux in the direction of the magnet. a general plane of the coil, this pair of magnetic poles being arranged so that, when the mechanical resonator 6a oscillates with an amplitude within a useful operating range, their respective magnetic fluxes pass through the coil with a time shift but with at least partly a simultaneity of the incoming magnetic flux and the outgoing magnetic flux, so as to form a central voltage lobe having a peak value which is maximum.

In the advantageous variant of FIG. 11, the balance 18a carries a pair of bipolar magnets 22 and 23 having axially oriented magnetization axes with opposite polarities. This pair of magnets and the coil 28 form the electromagnetic assembly 29 which is part of the control system. The magnets are arranged close to one another, at a distance allowing an addition of their respective interactions with the coil 28 as regards the voltage induced therein (more precisely for the generation of central voltage lobes) . In a variant not shown, it is possible to arrange a single bipolar magnet with its axis of magnetization parallel to the plane of the balance and oriented tangentially to a geometric circle centered on the axis of rotation 20. The voltage signal induced in the coil may present substantially the same profile as for the pair of magnets described above, but with a lower amplitude since only a portion of the magnetic flux of the magnet passes through the coil. Magnetic flux conduction elements may be associated with the single magnet to direct its magnetic flux substantially towards the general plane of the coil.

The balance 18a defines a half-axis 26, from its axis of rotation 20 and perpendicular to the latter, which passes in the middle of the pair of magnets. When the sprung balance is in its rest position, the half-axis 26 defines a neutral position around which the sprung balance can oscillate. The mechanical resonator 6a is shown in its neutral position in FIG. 11 and its half-axis 26 defines a reference half-axis 48 which is angularly offset by an angle θ relative to the fixed half-axis 50 which intercepts the axis of rotation 20 and the central axis of the coil 28. From preferably, the angular offset θ is between 30 ° and 120 ° in absolute value.

In the variant shown in FIGS. 14 and 15, the induced voltage signal Ui (t) generated by the electromechanical assembly 29 has, in each oscillation period of the mechanical oscillator, a first central voltage lobe LUC-i (also called first voltage lobe). ) having a maximum negative voltage UM-i and a second voltage lobe LUC2 (also called second voltage lobe) having a maximum positive voltage UM2. Due to the angular offset 6 of the coil relative to the reference half-axis 48, a second voltage lobe and a first voltage lobe occur respectively in a second half-wave of an alternation AOi, A1i, ... (AN- ι, N being a natural number) and in a first half-cycle of the following alternation A02, A12, ... (AN2, N being a natural number) of each oscillation period. In another variant, the polarities of the voltage lobes are opposite, that is to say that the first voltage lobes have a positive voltage while the second voltage lobes have a negative voltage. It will be noted that a simple inversion of the terminals E1 and E2 of the coil 28 or, in an equivalent manner, of the winding direction of the wire forming this coil generates a change of polarity for the induced voltage so that such an inversion makes it possible to switch from one variant to another.

Preferably, the electromagnetic assembly 29 also forms part of the measuring device, as in the first embodiment. The part of the electrical diagram of fig. 12 relating to the device for measuring an eventual time drift of the mechanical oscillator will not be described again in detail. It will be noted that the comparator 64 delivers a signal "Comp", represented in FIG. 14, which has a pulse S2 per oscillation period. Thus, this signal can be directly supplied to the bidirectional counter CB.

In FIG. 12, the electric converter 57 comprises a first circuit D1 &amp; C1 of electrical energy accumulation which is arranged to be able to recharge a first power supply C1 of the primary storage unit only with a positive voltage input of the electric converter and a second circuit D2 &amp; C2 electrical energy accumulation which is arranged to be able to recharge a second power supply C2 of the primary accumulation unit only with a negative voltage input of the electric converter. When charging, the amount of electrical energy selectively supplied by the braking device to the first power supply capacity and the second power supply capacitance is greater the greater the voltage level in absolute value of this power supply. first feed capacity, respectively of this second feed capacity is low.

A main load is connected or may be regularly connected to the output of the electrical converter 57 and powered by the primary power supply unit which supplies the supply voltages VDD and VSs- This main load includes in particular the circuit of Regulation 55. Preferably, the first and second supply capacitors have substantially the same capacitance value.

The control circuit 55 of the regulating device 53 comprises a charge pump device 61 formed by two charge pumps PC1 and PC2, advantageously identical, which are arranged to transfer on command electric charges respectively of the first capacitor. C1 power supply and the second power supply C2 in the auxiliary capacity CAux. As in the first embodiment, this auxiliary capacitance forms a secondary storage unit which provides an auxiliary supply voltage between its two terminals VL and VH. Both charge pumps PC1 and PC2 are controlled by the control logic 62a. An alternative embodiment of a charge pump that can form each of the two charge pumps has already been described with reference to FIG. 8. In a main variant, the two charge pumps are replaced by one and the same charge pump which then comprises switches controlled by the control circuit 62a so as to be able to transfer electric charges into the auxiliary capacity by selectively collecting these electric charges in the first capacitor C1 and in the second capacitor C2 according to the desired correction, as will be described later in the description of the control method implemented in the control circuit 62a in the context of the second embodiment. In the variant described, the control circuit 55 further comprises two dissipative circuits each formed of a resistor and a switch Sw3, respectively Sw4. These two dissipative circuits comprise a certain resistance and are respectively arranged in parallel with the two capacitors C1 and C2, between these and the two charge pumps PC1 and PC2.

In figs. 14 and 15 are also shown the positive voltage VCi at the upper terminal (defining VDd) of the supply capacitance C1 and the negative voltage VC2 at the lower terminal (defining VSs) of the supply capacitance C2 (the zero voltage being that of the end E1 of the coil connected between the two capacitors arranged in series). The supply voltage VAL available is therefore given by VC1-VC2, ie the addition of the respective voltages of the first and second capacitors C1 and C2. In the preferred variant described here, a main load is arranged at the output of the electrical converter. It comprises in particular the control circuit 55 which is powered by the first and second supply capacitors arranged in series and delivering the supply voltage VAL. The voltage lobes LUCi and LUC2 which respectively have the maximum negative induced voltage UM-i (in absolute value) and the maximum positive induced voltage UM2 serve to recharge the capacitors C2 and C1, respectively. Thus, apart from short recharging periods of either of the two power supply capacities, there is a certain gradual decrease (in absolute value) of the voltages VC 1 and VC 2 over time.

In the first oscillation period TO during which no regulation event occurs, an induced current pulse I12 reloads the capacitor C1 in a second half-cycle and an induced current pulse Il-ι recharges the C2 capacity in a first half-wave. These induced current pulses correspond to electrical powers generated by the electromechanical transducer in the electromagnetic assembly 29 and absorbed by the electric converter 57. These electrical powers thus correspond to mechanical powers provided by the mechanical oscillator. They are converted by the electric converter and consumed by the main load associated with it. Thus each current pulse induces In-, and IN2, N = 1, 2, ..., supplied by the electromechanical transducer to the electrical converter corresponds to a braking pulse and therefore to a certain momentary braking torque applied to the oscillator. mechanical. According to the physical phenomenon described above with reference to FIGS. 3 to 6, the induced current pulses IN2, each occurring in a second half-cycle, cause a decrease in the duration of the alternations during which they occur, and therefore an increase in the instantaneous frequency of the mechanical oscillator, while the induced current pulses IN-i, each occurring in a first half-cycle, generate an increase in the duration of the alternations during which they occur, and therefore a decrease in the instantaneous frequency of the mechanical oscillator.

In a period of operation during which no regulation event and no particular behavior resulting from such a regulation event occurs, that is to say in a period corresponding to normal operation without regulation, therefore, the situation represented in the first oscillation period in FIGS. 14 and 15 concerning the voltages VC1 and VC2 and the charging pulses of the capacitors C1 and C2 respectively generated by the induced current pulses I12 and Ι1Ί, namely a balanced situation in which a first electrical energy absorbed by the electric converter generally in the first two half-cycles of each oscillation period is substantially identical to a second electrical energy absorbed by the electric converter generally in the two second half-cycles of this oscillation period. Thus, the positive temporal phase shift which occurs globally in the two second half-cycles is compensated by the negative temporal phase shift which occurs globally in the first two half-cycles of each oscillation period. In the particular case shown in FIGS. 14 and 15, the positive phase shift that occurs in the first alternation A01 is compensated by the negative phase shift that occurs in the second alternation A02 of the corresponding oscillation period. It is therefore understood that, although the duration of the first half cycle is different from that of the second half cycle, their sum is equal to a natural oscillation period TO of the mechanical oscillator not subject to a regulation action.

The control method implemented in the control logic 62a of the charge pump device 61 is given by the flowchart of FIG. 13. After having initialized the regulation circuit at "POR" and in particular the bidirectional counter CB, a certain delay is expected, that is to say a certain time interval, for example a period TO or several periods TO, and the control circuit 62a determines whether at least some advance (CB> N1) has occurred in the running of the timepiece. If this is the case, in the present variant, the control circuit is arranged so that the control circuit can detect if the voltage VCa across the auxiliary capacitor is greater than a voltage threshold Vth, which corresponds to a a certain voltage for which the auxiliary capacity is filled to a level such that the charge pumps can no longer transfer significant electrical charges from either of the capacitors C1 and C2 into the auxiliary capacity. In this case, to effect a correction of the detected advance, the switch Sw2 is closed for a short time interval At to generate a certain discharge of the capacitor C2 through the corresponding dissipative circuit, indicated by the step DC2 (which is descending in absolute value because the voltage of the capacitance C2 decreases) in the voltage VC2 in FIG. 14.

If the voltage VCA is equal to or lower than the voltage threshold Vth, then the control circuit activates the charging pump PC2 to make a transfer of a first electric charge of the second power supply C2 in the auxiliary capacity CAux · This regulation action also results in a decrease in the voltage VC2 indicated by the DC2 downward movement. This reduction of the voltage VC2 generates, at least in a period of oscillation following such a transfer, an increase in the recharge of the second capacitor C2 relative to the hypothetical case where such a transfer of the first electric charge would not take place. The reduction of the voltage VC2 operated by the control circuit in the alternating A1Ί generates, when the next voltage lobe LUC-i appears in the next alternation A12, an induced current pulse Ι2Ί whose amplitude (value of peak voltage) is greater than that of the previous Ι1Ί. Since this current pulse induces Ι2Ί occurs in a first half-wave, like all the induced current pulses that recharge the capacitor C2, a decrease in the voltage of this capacitor C2 always generates at least one regulating pulse that generates a negative phase shift in the oscillation of the mechanical resonator and thus which momentarily decreases the oscillation frequency to at least partially correct the advance detected in the running of the timepiece (positive time drift). It will be noted that the pulses I12 and I22 have an amplitude, in absolute value, substantially equal to that of the pulse 11Ί, these pulses each corresponding to an induced current pulse generated by the single consumption of the main charge. This is therefore standard / nominal charging pulses.

If no advance is detected in the running of the timepiece, then the control circuit determines whether at least a certain delay (CB <- N2) has intervened in the running of this timepiece. If this is the case, the control circuit detects whether the voltage VCA across the auxiliary capacitor is greater than the voltage threshold Vth. In this case, to effect a correction of the detected delay, the switch Sw1 is closed during a short time interval At to generate a certain discharge of the capacitor C1 through the corresponding dissipative circuit, indicated by the DCi step (which is downward in absolute value because the voltage of the capacitance C2 decreases) in the voltage VC2 in FIG. 15. If the voltage VCA is equal to or lower than the voltage threshold Vth, then the control circuit activates the charge pump PC1 to perform a transfer of a second electric charge of the first supply capacitor C1 into the Auxiliary capacity CAux · This regulation action also results in a decrease in the voltage ν0Ί indicated by the DCi step. This reduction of the voltage VCi generates, at least in a period of oscillation following such a transfer, an increase in the recharge of the second capacitor C1 relative to the hypothetical case where such a transfer of the second electric charge would not take place. The reduction of the voltage VCi operated by the control circuit in the alternation Α1Ί generates, when the next LUC2 voltage lobe occurs, in the same alternation, an induced current pulse I32 whose amplitude is greater than that of the previous one. I12. Since this current pulse induces I32 occurs in a second half-wave, like all induced current pulses that recharge the capacitor C1, a decrease in the voltage of this capacitor C1 always generates at least one control pulse that generates a positive phase shift in the oscillation of the mechanical resonator and therefore temporarily increases the oscillation frequency to at least partially correct the delay detected in the running of the timepiece (negative time drift). The next pulse 131 again substantially shows a standard / nominal amplitude.

The second embodiment has an important advantage in that the selective charging of an electrical charge in the capacitor C1 or C2 as a function of a time drift detected in the running of the timepiece can intervene in all the time since the first lobes of tension, which intervene only in first half-alternations, have the same first polarity whereas the second lobes of tension, which intervene only in second half-alternations, have a same second polarity opposite to the first polarity and in that the capacitors C1 and C2 can be recharged respectively by induced voltages of opposite polarities. It is therefore sufficient for the control logic circuit to know which polarity, first or second, is capable of reloading which capacitance, C1 or C2, to selectively perform a sampling of a certain electrical charge in one or the other of these two capacitances according to the nature of a detected time drift, advance or delay, by a transfer of this certain electric charge in the auxiliary capacitance or by its dissipation through one of the two dissipative circuits provided if the auxiliary capacitance is full. In a variant, however, a timer is provided which determines a certain delay following the appearance of a pulse S2 in the signal "Comp" to perform selective sampling of an electric charge.

In an advantageous variant, for transferring a first or second electric charge, the number of cycles for transferring lower electric charges by a charge pump is increased when the voltage VCA across the auxiliary capacitor increases, so as to collect a substantially constant electrical charge of the capacitors C1 and C2 by sequence of the preceding regulation. In another variant where the number of transfer cycles of lower electric charges is expected to be constant, the increase of the voltage VCA generally causes a decrease in the first or second electrical charge taken and therefore a smaller correction by regulation sequence. However, insofar as the control system is configured to be able to easily correct drifts in a standard drift range for the watch movement in question, a decrease in the value of the first and second electric charges per control sequence, for a drift given time, will result in an increase of control sequences per unit of time. The above remarks concern conventional capacitors and also super-capacitors whose voltage-electric charge characteristic curve is substantially linear. On the other hand, it is also possible to provide as secondary storage unit an electrical capacitor whose voltage varies little, beyond a certain minimum charge level, depending on the accumulated electric charge. In this case, the electric charges transferred by the charge pump (s) are substantially constant regardless of the charge level of this secondary storage unit. In such a case, the regulation method described above may vary as regards the decision to transfer a certain electric charge in the secondary storage unit or consume this electric charge in the dissipative circuit provided. The regulating device will generally comprise means for determining the level of filling of the secondary accumulation unit.

With the aid of FIGS. 16 to 19 and 20A to 20C, will be described hereinafter a third embodiment of a timepiece according to the invention. The watch movement of this timepiece differs from that shown in FIG. 1 essentially by the configuration of the balance 18b, forming the mechanical resonator 6b, which here carries two pairs of bipolar magnets 82 and 84. The lessons already given that intervene again here will not be exposed in detail. What makes remarkable this third embodiment with respect to the first embodiment resides in particular in the choice of the electromagnetic assembly 86 forming the electromagnetic transducer and the electrical converter 76 associated therewith. The electromagnetic assembly comprises two pairs 82 and 84 of bipolar magnets 90 and 91, respectively 92 and 93, which are mounted on a rocker arm 18b of the mechanical resonator 6b and which have respective magnetization axes that are parallel to the axis of the magnet. rotation 20 of the balance, and a coil 28 which is secured to the support of the mechanical resonator.

Each of the two pairs 82, 84 of magnets, with its two bipolar magnets having respective opposite polarities, is similar to the pair of magnets 22, 23 of the electromagnetic assembly of the second embodiment and their interaction with the coil 28 is identical. Each pair of bipolar magnets defines a median half-axis 24a, 24b starting from the axis of rotation 20 of the beam and passing through the middle of the pair of bipolar magnets considered. Each half-center axis defines a respective reference half-axis 48a, 48b when the resonator 6a is at rest and thus in its neutral position, as shown in FIG. 16. The coil 28 has at its center a first angular offset θ relative to the first reference half-axis 48a and a second angular offset - θ (same absolute value as the first angular offset, but opposite mathematical sign) relative to the second half. reference axis 48a, so as to generate in each alternation of the mechanical resonator, in a useful operating range, two central voltage lobes LUCi and LUC2 having opposite polarities (negative and positive) and substantially the same amplitude UM-i, UM2 in absolute value and forming respectively a first voltage lobe and a second voltage lobe (Fig. 20A).

As in the second embodiment, the first and second voltage lobes LUC-i and LUC2 occur respectively in first half-cycles and second half-cycles. Preferably, to balance the balance 18a, the first and second angular phase shifts have an absolute value of 90 ° (variant shown in Fig. 16). The two pairs of magnets 82 and 84 are arranged so that the polarities of the magnets of one pair are symmetrical to the polarities of the magnets of the other pair relative to a plane passing through the center of the coil and comprising the axis of rotation 20 (this plane comprising the half-axis 50 passing through the center of the coil and perpendicularly intercepting the axis of rotation 20). It will be noted that the variant of the third embodiment described with reference to the figures is an improved variant. In another variant which will not be described in more detail later, there is provided a single pair of magnets having an angular offset of between 30 ° and 120 ° (in absolute value). This other variant comprises a control circuit without the flip-flop 66. The control method remains similar and the skilled person will be able to adapt it to this particular variant.

The induced voltage signal U i (t), shown in FIG. 20A alternatively has LUC-i voltage lobes having a negative voltage and LUC2 voltage lobes having a positive voltage. The electric converter 76 comprises a full-wave rectifier 78 formed by a bridge of four diodes well known to those skilled in the art. Thus, at the output of the rectifier 78, the first voltage lobes are rectified, which is shown in FIG. 20A by the lobes in broken lines. As in the first embodiment, in the absence of activation of the charge pump 60b, the first and second voltage lobes LUC 1 and LUC 2 alternately recharge the supply capacitance Cal, which in particular supplies the control circuit 74. Since there are two pairs of magnets, each alternation has a first voltage lobe in a first half-cycle and a second voltage lobe in a second half-wave. As the "Comp" signal has two pulses per oscillation period, there is provided a flip-flop 66 upstream of the bidirectional counter CB so as to inhibit every other pulse in the signal supplied to this counter. The variant shown in FIGS. 20A and 20C provides a positive Uth threshold voltage while the first voltage lobes are negative. The threshold voltage can be chosen positive or negative. These choices determine the instants at which the pulses S2 or S1 (see FIG. 1C) intervene in the "Comp" signal supplied by the comparator 64. Thus, the regulating device comprises a detection device which is arranged to be able to detect the appearance successive first voltage lobes or second voltage lobes. Note that it is also possible to detect alternately these first and second voltage lobes using two comparators having respectively a positive voltage threshold and a negative voltage threshold. Those skilled in the art will be able to adapt the control method implemented in the control logic circuit 62b accordingly, in particular for determining the delays TC2 and TD2.

The charge pump device is formed of a charge pump 60b which defines a voltage booster and which is arranged between the supply capacitance Cal (primary storage unit) and an electrical capacitor (unit of charge). secondary accumulation) so that electric charges can be transferred from the primary storage unit to the secondary storage unit. The charge pump 60b quadruple the main supply voltage Ual delivered by the primary power supply so that the voltage of the auxiliary power supply VCA of the electric capacitor can be higher, in particular twice the voltage Ual- The construction and operation such a voltage booster are well known to those skilled in the art. The electrical diagram of a variant is given in fig. 18. It includes four transfer capabilities Cyr, two input switches Sw1, six switches 82, three switches 84 and two output switches Sw2. To take a certain electrical charge from the capacitor Cal, the switches Sw1 and 82 are closed while the switches Sw2 and 84 are open (the capacitors CTr are then arranged in parallel). To then charge the Caco electric capacitor, the switches Sw1 and 82 are open while the switches Sw2 and 84 are closed (the capacitors CTr are then arranged in series).

Although the primary storage unit of this third embodiment is identical to that of the first embodiment with a single capacitance Cal which receives all of the induced currents supplied by the electromagnetic transducer, the fact that the electromagnetic assembly 86 is arranged in a manner similar to that of the second embodiment, with the first voltage lobes and the second voltage lobes having opposite polarities, allows the comparator 64 to directly detect either the first voltage lobes or the second voltage lobes. voltage lobes (as shown in Fig. 20A). It is therefore not necessary here to differentiate in the pulses provided by the comparator those which correspond to the first lobes of those which correspond to the second lobes, which is why there is no time counter CT, but only one timer associated with the control logic circuit, which can be integrated within this logic circuit, to measure two delays TC2 and TD2. In fig. 20C, it is observed that the "Comp" signal has only pulses S2 which each correspond to the appearance of a second LUC2 voltage lobe.

FIG. 19 is a flowchart of the control method implemented in the control logic 62b of the third embodiment. We will no longer describe in detail all the elements, all the electrical signals and the consequences of the various events that occur, because they derive from the explanations already given previously, and the results are easily understood in the light of these explanations.

When the regulating device is put into operation, the regulation circuit 74 is initialized to "POR", in particular the bidirectional counter CB. The logic circuit then waits for the appearance of a pulse S2, namely in particular its rising edge in the signal "Comp". The detection of this rising edge triggers the timer which measures a first time interval TC2 whose duration is chosen so that its end occurs in a first temporal zone ZT 1 situated temporally between a second voltage lobe LUC2 and a first voltage lobe LUC. -i, especially between the moment t2 and the moment b where these two lobes respectively present their maximum values UM2 and UM-i (FIG 20A). In parallel, the logic circuit detects whether the value of the bidirectional counter CB is greater than a natural number N1 to determine whether there is an advance in the operation of the mechanism in question. If so, the control circuit waits for the end of the delay TC2 and, equivalent to the control method of the second embodiment, determines whether the electric capacitor CAcc is full (i.e., detects whether its level of accumulation of electrical charges is above a certain limit). If the electric capacitor CAcc is full, it discharges the supply capacitance CAl of a first electrical charge by closing the switch Sw5 of the dissipative circuit comprising a certain resistance and provided in parallel with the charge pump for a certain period of time At (Fig. 17). In the opposite case, it performs a transfer of a first electrical charge of the capacitor Cal into the electric capacitor CAcc in a first time zone ZT1. A sampling of a first electric charge generates a downward movement PC-ι in the supply voltage UAl (1) and the next induced current pulse P1pc, which occurs in a first half-cycle, then has an amplitude greater than that of a pulse P1 in the absence of previous sampling of an electric charge (see right part of Figs 20A to Fig. 20C), so that the mechanical oscillator then undergoes a higher braking in the first half-cycle considered.

If the counter CB has a value equal to or less than the natural number N1, then the logic circuit waits for a second delay TD2, directly following the first delay TC2, to come to an end (FIG 20C). To do this, at the end of a first time interval TC2, the timer begins to measure a second time interval TD2. This second delay TD2 is chosen so that its end occurs in a second time zone ZT2 located between a first voltage lobe LUCi and a second voltage lobe LUC2. In parallel, the logic circuit detects whether the value of the bidirectional counter CB is smaller than a number - N2, N2 being a natural number, to determine if there is a delay in the operation of the mechanism in question. If this is the case, the control circuit waits for the end of the delay TC2 + TD2 and determines whether the electric capacitor CAcc is full. Depending on whether the capacitor is full or not, the control circuit then operates in a manner similar to that described above in the case of the detection of an advance. The taking of a second electric charge in the capacitance Cal generates a downward movement PC2 in the supply voltage UalW and the induced current pulse P2pc, which intervenes in a second half-wave, then has an amplitude greater than that a pulse P2 in the absence of previous sampling of an electric charge (see left part of Figs 20A to Fig. 20C), so that the mechanical oscillator then undergoes a higher braking in the second half-wave considered .

In conclusion, as in the first embodiment, a delay or advance observed in the operation of the mechanism in question is corrected by the temporally selective sampling of an electrical charge in the capacitance Cal forming primary storage unit of the regulating device.

The control method of the third embodiment further comprises an improvement in connection with the fact that the secondary accumulation unit continuously or intermittently feeds an auxiliary load by delivering an auxiliary supply voltage VCa to this auxiliary charge. Indeed, the auxiliary load is preferably associated with a useful auxiliary function of the timepiece, so it is desirable to be able to supply the auxiliary load. For this purpose, as indicated in the flowchart of FIG. 19, if the counter CB has a value equal to or greater than the number - N2 and a value equal to or smaller than the number N1, then the control circuit 62b determines by appropriate means whether the capacitor is empty or not. By "vacuum", it is understood that the level of accumulation of electric charges in the capacitor Cacc is below a given lower limit and therefore to a situation no longer possible to ensure proper supply of the auxiliary function (diode light, RFID circuit, temperature measurement, North indication (compass function), etc.). Such a situation therefore occurs in the case where no temporal drift, generating a correction of the instantaneous frequency of the mechanical oscillator according to the invention, is detected. If this situation occurs and the Caco electric capacitor is empty (in other words not sufficiently recharged), then the control circuit performs a recharging operation of the electric capacitor by taking a first load in a first time zone ZT1 and a second one. electric charge, substantially of the same value as the first electric charge, in a second time zone ZT2. These two events generate phase shifts in the oscillation of the mechanical resonator which compensate each other, so that a double electric charge is transferred from the primary storage unit to the secondary storage unit without generating a time drift in the walking of the timepiece. Once the control sequence is complete, the control logic circuit waits for detection of the rising edge of the next pulse S2 to perform the next control sequence.

As already mentioned, the transfer of a first electric charge, respectively a second electric charge can be performed by a plurality of cycles of transfer of lower electrical charges by the charge pump in the same control sequence, in particularly in the same time zone ZT 1, respectively ZT2. In a variant, the control logic circuit is arranged in such a way that, when the measured time drift corresponds to at least some advance, a plurality of electrical charge takings respectively in a plurality of first time zones during an same sequence of regulation. Similarly, when the measured time drift corresponds to at least a certain delay, a plurality of electrical charge takings respectively in a plurality of second time zones are made.

In figs. 21 and 22 is shown an advantageous embodiment of a mechanical oscillator 106 incorporated in a movement according to the invention. The resonator 106 is formed by a rocker 18c which comprises two plates of ferromagnetic material 112 and 114. The upper plate 112 bears on the side of its underside the two bipolar magnets 22 and 23. This upper plate also serves to close the upper lines of fields of the two magnets. The lower plate 114 serves to internally close the field lines of the two magnets. The two plates of the balance thus axially form a magnetic shielding for the two magnets so that their respective magnetic fields remain substantially confined in a volume located between the respective outer surfaces of these two plates. The coil 28 is arranged partially between the two plates which are fixedly mounted on a cylindrical piece 116 of non-magnetic material, this piece being fixedly mounted on a shaft 118 of the balance. In a variant, the piece 116 may be made of steel and thus conduct a magnetic field, which may be an advantage in a variant provided with a single bipolar magnet, having its magnetic axis oriented axially, on one of the two trays or on each two trays. In the latter case, if the cylindrical connecting piece is non-magnetic, then at least one plate may have a ferromagnetic portion that approaches the other or the key to close the field lines of each magnet through the two trays and thus allow the coil or coils to be traversed axially by substantially the entire magnetic field produced by each magnet when the pendulum oscillates. It will also be noted that the trays can be made only partially by a high magnetic permeability material which forms two parts located respectively above and below the magnet or, where appropriate, magnets, these two parts being arranged in in order to let the coil or, if necessary, the coils of the control system between them when the pendulum oscillates.

The resonator 106 also comprises a spiral spring 110, one end of which is conventionally fixed to the shaft 118. It will be noted that the spiral spring is preferably made of non-magnetic material, for example silicon, or paramagnetic material. In fig. 22 is also shown an escape mechanism formed of a pin arranged on a small plate integral with the balance shaft, an anchor 120 and an escape wheel 122 (shown partially). Under the upper plate, opposite the magnets 22 and 23, there is provided a balancing mass 124 of the balance. Other means to perform a fine adjustment of the inertia and balance of the balance can also be provided. Note that in a variant, magnets are also carried by the lower plate. Such magnets are preferably arranged in front of the magnets carried by the upper plate.

Thus, in the context of the advantageous variant described above, the balance generally comprises a magnetic structure which is arranged to define a magnetic shield for the magnet or the magnets carried by the pendulum while facilitating the magnetic coupling of this magnet or magnets with the coil or the coils provided (s).

Claims (22)

claims 1. Timepiece (2), comprising: - a mechanism, - a mechanical resonator (6; 6a; 6b) capable of oscillating about a neutral position corresponding to its minimum mechanical potential energy state, each oscillation mechanical resonator defining an oscillation period and having two successive alternations each between two extreme positions which define the oscillation amplitude of the mechanical resonator, each alternation having a passage of the mechanical resonator by its neutral position at a median instant (TNi, i = 1,2,3 ...) and consisting of a first half-alternation (DA1) between an initial moment of this alternation and its median instant and a second half-alternation (DA2) between this median instant and a final moment of this alternation, - a maintenance device (14) of the mechanical resonator forming with this mechanical resonator a mechanical oscillator which defines the rate of the said electromechanical transducer arranged to be able to convert the mechanical power of the mechanical oscillator into electrical power when the mechanical resonator oscillates with an amplitude included in a useful operating range, this electromagnetic transducer being formed by an electromagnetic assembly ( 27; 29; 86) comprising at least one coil (28), mounted on one of the mechanical assembly consisting of the mechanical resonator and its support, and at least one magnet (22; 90) mounted on the other element of this mechanical assembly, the electromagnetic assembly being arranged to be able to provide an induced voltage signal U i (t) between the two output terminals (E1, E2) of the electromechanical transducer at least when the mechanical resonator oscillates with an amplitude in the range of useful operation, - an electrical converter (56; 57; 76) connected to the two output terminals of the electromechanical transducer so as to be able to receive from this electromechanical transducer an induced electric current (Irec), this electrical converter comprising a primary accumulation unit (Cal; C1, C2) arranged to accumulate electrical energy supplied by the electromechanical transducer, this electromechanical transducer and the electric converter together forming a braking device of the mechanical resonator; - a regulating device (52; 53; 72) of the frequency of the mechanical oscillator, this regulating device comprising an auxiliary oscillator (58) and a measuring device (64, 66, CB) arranged to be able to detect an eventual time drift of the mechanical oscillator relative to the auxiliary oscillator, the regulating device being arranged to be able to determine if the measured time drift corresponds to at least some advance; the timepiece being characterized in that the control device (52; 53; 72) is arranged to also be able to determine if the measured time drift corresponds to at least a certain delay; in that the braking device is arranged so that, in each oscillation period of the mechanical resonator when the oscillation amplitude thereof is within said useful operating range, the induced voltage signal has at least one first voltage lobe (LUi, LUC-i) intervening at least in a major part in a first half-wave (DA1) and capable of generating in this first half-wave a first induced current pulse (P1; Ιη-ι, n = 1.2, 3) for recharging the primary accumulation unit after a certain amount of electric charge has been taken out of it and at least one second voltage lobe (LU2, LUC2) is involved at least in a major part in a second half-wave (DA2) and capable of generating in this second half-wave a second pulse of induced current (P2; ln2, n = 1, 2, 3) to recharge the primary accumulation unit after a certain collection of a the electrical charge thereof, the induced voltage signal thus having a plurality of such first voltage lobes and a plurality of such second voltage lobes; in that the regulating device comprises a charge pump device (60; 61; 60b) arranged to be able to transfer a certain electrical charge of the primary storage unit (Cal! C1, C2) to a control unit in a control unit. secondary accumulation (CAux, CAcc); and in that the regulating device further comprises a control logic circuit (62; 62a; 62b) which receives as input a measurement signal provided by the measuring device and which is arranged to be able to activate the charge pump device. so that, when the measured time drift corresponds to said at least some advance, a transfer of a first electric charge of the primary accumulation unit into the secondary accumulation unit so that a recharge of the primary storage unit, following this transfer of the first electrical charge, is generated mainly by at least a first voltage lobe among said plurality of first voltage lobes, the control logic being furthermore arranged to enable the charge pump device to perform, when the measured time drift corresponds to the at least one delay, a transfer of a second electric charge of the primary storage unit in the secondary storage unit so that a recharge of the primary storage unit, following this transfer of the second electric charge, is generated for the major part by at least one second voltage lobe among said plurality of second voltage lobes. 2. Timepiece according to claim 1, characterized in that it comprises a main charge (54; 55; 74) connected to or likely to be regularly connected to the electrical converter to be powered by the primary accumulation unit. , the main load comprising the regulating device. 3. Timepiece according to claim 2, characterized in that it comprises an auxiliary charge connected to or capable of being connected intermittently to the secondary accumulation unit so as to be powered by this accumulation unit. secondary. 4. Timepiece according to claim 3, characterized in that the charge pump device (60b) is arranged to form a voltage booster which is arranged for an auxiliary supply voltage across the terminals of the charging device. secondary storage unit (Cacc) is greater than a main supply voltage across the primary storage unit. 5. Timepiece according to any one of claims 2 to 4, characterized in that the primary storage unit is formed by a supply capacity (CAL) capable of being recharged by each first voltage lobe of said plurality of first voltage lobes and said plurality of second voltage lobes after charging of an electrical load in that power supply; in that each first voltage lobe has, in absolute value, a first maximum value (UM-i) at a first instant (t-ι) of the corresponding first half-cycle and each second voltage lobe present, in absolute value a second maximum value (UM2) at a second instant (t2) of the corresponding second half-cycle, the first and second voltage lobes defining, on the one hand, first time zones (ZT1) each located before said first instant of a first different voltage lobe and after the second instant of the second voltage lobe preceding this first voltage lobe and secondly of the second time zones (ZT2) each located before said second instant of a second lobe of voltage different and after the first moment of the first voltage lobe preceding this second voltage lobe; and in that said transfer of a first electrical charge comprises taking this first electrical charge of the supply capacity in a first time zone from among said first time zones (ZT1) and said transfer of a second electric charge comprises a taking a second electrical charge of the supply capacity in a second time zone from among said second time zones (ZT2). 6. Timepiece according to claim 5, characterized in that the regulating device (52; 72) further comprises a timer associated with the control logic circuit (62; 62b) to enable the latter to activate the where appropriate, the charge pump (60; 60b) after a given first delay (Tel; Tc2) from the detection of a first voltage lobe or a second voltage lobe or after a given second delay (TD1; TC2 + TD2) since the detection of a first voltage lobe or a second voltage lobe. Timepiece according to Claim 5 or 6, characterized in that the charge pump device consists of a charge pump (60; 60b), this charge pump and the control logic circuit being arranged whereby the taking of said first electric charge and the taking of said second electric charge are each carried out in several cycles of transfer of a lower electrical charge between the supply capacitance (Cal) and the secondary storage unit (CAux). CAcc) by the charge pump. 8. Timepiece according to any one of claims 5 to 7, characterized in that the control logic (60; 60b) is arranged to be able to perform, when the measured time drift corresponds to said at least some advance or at least a given advance greater than the latter, a plurality of transfers of first electrical charges respectively during a plurality of first time zones and so as to be able to perform, when the measured time drift corresponds to said at least a certain delay or at least one delay greater than the latter, a plurality of samples of second electrical charges respectively during a plurality of second time zones. 9. Timepiece according to any one of claims 5 to 8, characterized in that the electromagnetic assembly (26) comprises a bipolar magnet (22), mounted on a rocker (18) of the mechanical resonator (6) and having a magnetization axis in a geometric plane comprising the axis of rotation of the balance, and a coil (28) which is integral with the support of the mechanical resonator and arranged to be traversed by the magnetic flux of the bipolar magnet, a half -axial axis (24) starting from the axis of rotation (20) of the beam and passing through said axis of axial magnetization defining a reference half-axis (48) when the resonator is at rest and thus in its neutral position; and in that said coil has at its center an angular offset (θ) relative to the reference half-axis and said bipolar magnet is arranged on the balance so that the only coupling between this bipolar magnet and the coil can be generated in each period oscillation of the mechanical resonator, in said useful operating range, two voltage lobes (LU-i, LU2) of the same polarity which respectively form one of said first voltage lobes and one of said second voltage lobes. 10. Timepiece according to claim 9, characterized in that said angular offset is between 30 ° and 120 ° in absolute value. Timepiece according to claim 9 or 10, characterized in that the regulating device comprises a detection device (64), arranged to be able to detect alternately the successive appearance of said first voltage lobes (LU1) and of said second voltage lobes (LU2), and a time counter (CT) associated with the control logic circuit (62) to enable the latter to distinguish a first time interval, separating a first voltage lobe from a second lobe of voltage thereafter, and a second time interval separating a second voltage lobe from a first voltage lobe that follows, the first and second time intervals being different due to the arrangement of said electromagnetic assembly. Timepiece according to any one of claims 5 to 8, characterized in that the primary accumulation unit comprises a first feed capacity (C2) and a second feed capacity (C1), both arranged to be able to supply said main load; in that the electromagnetic transducer (6a, 29) is arranged such that the plurality of first voltage lobes (LUC-i) each have a first polarity and the plurality of second voltage lobes (LUC-2) each have a second opposite polarity at the first polarity; in that the electrical converter (57) is formed by a first electrical energy storage circuit (D2, C2) which comprises the first power supply capacitance and which is arranged to be able to recharge this first power supply capacity only with a voltage having the first polarity at the input of the electrical converter and a second electrical energy storage circuit (D1, C1) which includes the second power supply capacitance and which is arranged to be able to recharge this second power supply capacity only with a voltage having the second polarity at the input of the electrical converter, the amount of electrical energy supplied during a refill by the braking device to the first feed capacity, respectively to the second feed capacity being all the greater than the voltage level of this first power supply, respectively of this second feeding capacity is low; and in that the regulating device is arranged such that said transfer of said first electric charge consists of a transfer of this first electric charge of the first power supply capacity to the secondary storage unit and said transfer of said second electrical charge electric charge is a transfer of this second electric charge from the second power supply to the secondary storage unit. 13. Timepiece according to claim 12, characterized in that the first and second feed capacities (C2, C1) have substantially the same capacity value and are arranged to jointly feed said main load. Timepiece according to claim 12 or 13, characterized in that the first and second power supply capacitors are arranged to deliver a supply voltage corresponding to the addition of the respective voltages (VC1, - VC2). of these first and second feed capacities. 15. Timepiece according to any one of claims 5 to 8 and 12 to 14, characterized in that the electromagnetic assembly (86) comprises a pair of bipolar magnets (22, 23; 82) mounted on a balance (18a). 18b) of the mechanical resonator (6a; 6b) and having two respective magnetization axes which are parallel to a geometric plane including the axis of rotation (20) of the balance with respective opposite polarities, and a coil (28) which is secured to the support of the mechanical resonator, the two bipolar magnets (22, 23; 90, 91) of said pair being arranged on the balance so that their respective magnetic fluxes pass through the coil with a time shift but with in part a simultaneity of the incoming magnetic flux and the outgoing magnetic flux such that an induced voltage pulse generated between the two ends (E1, E2) of the coil at the passage of the pair of magnets opposite said coil a central lobe (LUC-i, LUC2) having a maximum amplitude resulting from simultaneous coupling of the two magnets of the pair of magnets with the coil; in that a middle half-axis (26; 24a) extending from the axis of rotation of the beam and passing through the middle of the pair of bipolar magnets defines a reference half-axis (48; 48a) when the resonator is at rest and thus in its neutral position, the coil having at its center an angular offset (θ) relative to the reference half-axis so as to generate in each oscillation period of the mechanical resonator, in said useful operating range, two central voltage lobes (LUCI, LUC2) having opposite polarities and respectively forming said first voltage lobe and said second voltage lobe. 16. Timepiece according to claim 15, characterized in that said angular offset is between 30 ° and 120 ° in absolute value. 17. Timepiece according to claim 15 or 16, characterized in that the regulating device (53; 72) comprises at least one detection device (64) arranged to be able to detect the successive appearance of first voltage lobes. (LUC!) And / or second voltage lobes (LUC2). Timepiece according to any one of the preceding claims, characterized in that the regulating device (53; 72) comprises at least one dissipative circuit for dissipating electrical energy accumulated in the primary accumulation unit, at least one switch (Sw3, Sw4; Sw5) associated with the dissipative circuit for momentarily connecting this dissipative circuit to the primary storage unit and a measuring circuit arranged to detect if the voltage across the secondary storage unit is greater than a first voltage limit or if the fill level of the secondary storage unit is greater than a first fill limit; and in that the control logic circuit is further arranged such that, when the voltage across the secondary storage unit is greater than or equal to the first voltage or load limit, momentarily connect the at least one a dissipative circuit to the primary storage unit so as to perform, when the measured time drift corresponds to said at least some advance, a first dissipative discharge of the primary storage unit so that a recharge of that after this first discharge, it is generated for the most part by at least one first voltage lobe among said plurality of first voltage lobes, and so to perform, when the measured time drift corresponds to said at least a certain delay, a second discharge of the primary accumulation unit so that a recharge thereof, following this second discharge, is gen most of which is provided by at least one second voltage lobe among said plurality of second voltage lobes. 19. Timepiece according to any one of the preceding claims, characterized in that it comprises a measuring circuit arranged to detect whether the voltage across the secondary storage unit is less than a second voltage limit or if the fill level of the secondary storage unit is less than a second fill limit; and in that the control logic circuit is arranged so that when the voltage across the secondary storage unit is lower than the second voltage or fill limit and the measured time drift is between at least some delay and at least some advance, activating the charge pump device to effect a transfer of a third electrical charge from the primary accumulation unit to the secondary accumulation unit , so that a recharge of the primary storage unit following this transfer of a third electric charge is generated for the most part by at least a first voltage lobe among said plurality of first voltage lobes, and a transfer a fourth electrical charge of the primary storage unit in the secondary storage unit, so that a recharge of the primary accumulation result of this transfer of a fourth electric charge is generated mainly by at least one second voltage lobe of said plurality of second voltage lobes, the fourth electrical load is substantially equal to the third electric charge. 20. Timepiece according to any one of the preceding claims, characterized in that the secondary accumulation unit is formed by a super capacitance or an electrical capacitor. 21. Timepiece according to any one of the preceding claims, characterized in that the mechanical resonator comprises a sprung balance; and in that said maintenance device comprises an escapement (14) kinematically connected to a barrel (12) provided with a mainspring, the escapement being capable of supplying the balance spring with a mechanical torque of maintenance of its oscillations. 22. Timepiece according to any one of the preceding claims, characterized in that said electromagnetic assembly (26; 86) also partially forms the measuring device.