EP0154889A1 - Constant-power supply means for a stepping motor fed by a variable-tension source - Google Patents

Abstract

The device enables a stepping motor (M) to be supplied at constant power whatever the value of the supply voltage. Using a bridge (R1 to R4), a voltage (UAB) proportional to the supply voltage (U0) is tapped. The signal (T1) required to advance the rotor by one step is split into a large number of sampling periods (Te) of predetermined duration. Each of these periods consists of an elementary pulse followed by a period of interruption. The duration (T0) of the elementary pulse is determined by the time taken by a capacitor (C) to become charged: this will become shorter as the proportional voltage (UAB) applied to terminals of a network (RC) becomes higher. A comparator (16) compares the voltage (Uc) across the terminals of the capacitor with a reference voltage (Ur) so as to determine the duration (T0) of the elementary pulse and to define a chopping rate for the supply signal to the motor (M) which is inversely proportional to the supply voltage (U0). <IMAGE>

Description

The present invention relates to a device for supplying constant energy from a variable voltage source U _o a _stepping motor, in particular for a timepiece, the supply consisting in supplying the motor with pulses of constant duration. T _j , each being chopped according to a dependent hash rate. Dant from the value of the voltage of the source U and constituted by a series of elementary pulses of duration T separated by periods of interruption T _r , the sum T + T defining a predetermined sampling period T _e .

In electronic timepieces that are found today on the market, it is common practice to find a stepper motor to convert the electrical pulses from a quartz time base into mechanical movement for show time. The system is powered by a power source, usually a small battery that will need to be replaced periodically. To save the energy delivered by the battery and therefore make it last as long as possible, adjustment systems have already been proposed which control the duration of the pulse which moves the motor to the load which it must drive; in other words, the pulse is lengthened if the charge increases and it is reduced if the same charge decreases. Such systems are described for example in documents US-A-4,323,834 and US-A-4,346,463.

These systems generally assume that the motor is supplied at constant voltage and do not take into account the differences between the voltage delivered by the battery at the beginning then at the end of its service life. At first glance, this simplification may seem legitimate for a silver battery which has an almost constant discharge characteristic of 1.5 V during its lifetime. However, if the chosen energy source is a silver peroxide battery, the situation presents itself differently since its start-of-life voltage is around 1.8 V and we have to subject it to treatment. special during manufacture which reduces its capacity and which brings its voltage to 1.5 V. So, in this case, we do not make the best use of the characteristics of this type of battery which is normally distinguished by a high capacity given its volume. The situation is even more unfavorable if we consider the newly offered lithium battery on the market which is remarkable in terms of its high reliability and its high energy density, but which shows an extremely unfavorable discharge characteristic in that meaning that its voltage varies, during its lifetime, from 3.6 to 2.4 V approximately.

In the case of the last two types of batteries mentioned above, it will be understood that a reduction in voltage, generally accompanied by an increase in resistance of the battery, will firstly cause unstable operation of the motor and then a total stop and this well before the end of the battery life. To overcome this drawback, the motor could be dimensioned so that it can operate even when the battery supplies its lowest voltage. However, this would result in overconsumption of the engine for a very large part of the life of the battery.

To obviate the drawbacks inherent in the solutions cited, it has already been proposed to supply the motor with attack pulses of constant duration T _i , these pulses being chopped and constituted by a series of elementary pulses of duration T separated by periods interrupt T r, the chopping rate being a function of the value of the voltage present across the battery.

Such an arrangement is described in document GB-A-2 054 916 which recommends supplying the winding of the stepping motor with driving pulses each formed from a series of elementary pulses whose width is determined as a function of the value of the voltage which is supplied by the battery when the latter is connected to the terminals of resistors of known values. According to this arrangement, it is determined, roughly every millisecond, in which range of values this voltage is found and a motor signal form is chosen from among five predetermined signal forms. This is therefore a discontinuous adjustment of the power of the driving pulses as a function of the voltage of the power source, and this results in significant variations in the motor torque which can cause lost steps. In addition, this adjustment being discontinuous, it does not allow effective control of the energy of the driving pulses to the load which the motor must drive if such control is desired.

Another solution for solving the problem posed is proposed in document EP-A1-0 077 293 where the control device makes it possible to adapt the power of each drive pulse substantially continuously to the value of the voltage of the source. power supply. Here, the value of the hash rate is recalibrated periodically depending on the supply voltage, then stored, after which the hash rate is kept at this value until the arrival of a next recalibration signal. However, the proposed system requires the use of a computer and the storage of the calculated difference, which complicates the electronic circuit of the watch. In addition, this system recommends the measurement of the circulating current in the motor winding, which requires the use of a resistor arranged in series with said winding and a switching system which also complicates the motor supply circuit. .

To obviate the drawbacks which have just been mentioned, the present invention provides a device which supplies the stepping motor with constant energy whatever the voltage delivered by the battery and this by means of the means which appear in the claims.

• La figure 1 est un schéma bloc du dispositif selon l'invention;
• La figure 2 est un diagramme montrant comment se présente la tension aux bornes du moteur quand la pile qui l'alimente est au début de sa durée de vie (figure 2a) et quand cette même pile est à la fin de sa durée de vie (figure 2b) lorsque ledit moteur est alimenté selon le dispositif de la figure 1;
• La figure 3 est un schéma détaillé de l'alimentation du moteur suivant le schéma de principe de la figure 1 et selon un mode de réalisation préféré;
• Les figures 4 et 5 sont des diagrammes montrant les signaux en différents points du schéma de la figure 3 dans deux cas de tension de la source d'alimentation;
• La figure 6 est un schéma montrant comment on prélève du diviseur de fréquence les diverses impulsions nécessaires au fonctionnement du dispositif illustré en figure 3;
• La figure 7 est un schéma qui complète le schéma de la figure 3 selon une variante d'exécution de l'invention.

The invention will now be better understood in the light of the following description and for the understanding of which reference will be made, by way of example, to the drawing in which:

• Figure 1 is a block diagram of the device according to the invention;
• FIG. 2 is a diagram showing how the voltage across the terminals of the motor looks when the battery supplying it is at the start of its service life (FIG. 2a) and when this same battery is at the end of its service life ( Figure 2b) when said motor is powered according to the device of Figure 1;
• Figure 3 is a detailed diagram of the motor supply according to the block diagram of Figure 1 and according to a preferred embodiment;
• Figures 4 and 5 are diagrams showing the signals at different points in the diagram of Figure 3 in two cases of voltage of the power source;
• FIG. 6 is a diagram showing how the various pulses necessary for the operation of the device illustrated in FIG. 3 are taken from the frequency divider;
• Figure 7 is a diagram which supplements the diagram of Figure 3 according to an alternative embodiment of the invention.

Figure 1 is a block diagram of the device according to the invention. The motor M is of the stepping type and is supplied by a DC voltage source, a lithium battery for example, delivering a voltage U _o . If, for a predetermined duration T _i , the switch 1 is closed and at the same time the switch 2 is repeatedly pressed, a chopped driving pulse is obtained which will cause the rotor of the motor to advance by one step. As said above, it is a question of supplying the motor with constant energy whatever the value of the battery voltage and to do this we will make the hash rate depend on said voltage. The time during which the switch 1 is closed is divided into a multiplicity of sampling periods T _e fixed in advance, each of these periods consisting of a pulse of duration T _o during which the switch 2 is closed followed by a rest period T _r during which the switch 2 is open. It will be understood that the ratio T _{o /} T _e is precisely the chopping rate which it is a question of controlling as a function of the supply voltage U _o . If U _o is high (start of battery life), T _o will be short and the hash rate low. If, on the contrary, U _o is low (end of the life of the battery), T will have to be lengthened, which will cause a high hash rate.

où E est l'énergie appliquée au moteur et I le courant circulant dans sa bobine. On voit que pour maintenir une énergie E constante il faut s'arranger pour maintenir constante la tension moyenne U_m représentée par l'expression U_oT_o/T_e , étant donné que si la tension moyenne reste constante, le courant I reste également constant.The first of these situations is illustrated in Figure 2a. Here, the battery voltage U is high (for example 3 V) and, while the switch 1 is closed, during the period T _i , the appearance of the voltage U _M at the terminals of the motor has a series of short pulses of duration T _o , the amplitude of which is that of the voltage U _o . The hash rate T _o / T _e imposed by the switch 2 is of the order of 25%. The second of these situations is illustrated in FIG. 2b where the battery voltage U _o is reduced by half (for example 1.5 V). In this case, the voltage U _M is composed of pulses whose amplitude is reduced by half but of duration T _o longer and having a chopping rate of the order of 50%. We see that in. both situations the sampling period T _e is the same. In fact, the period T _e is fixed once and for all by construction and constitutes a predetermined period. In both cases, the average voltage U _m applied to the motor is the same. This can be expressed analytically by the expression:

where E is the energy applied to the motor and I the current flowing in its coil. We see that to maintain a constant energy E it is necessary to manage to maintain constant the average voltage U _m represented by the expression U _o T _o / T _e , given that if the average voltage remains constant, the current I also remains constant.

If we refer again to FIG. 1, we will now explain how we proceed, according to the invention, to deliver a constant energy to the motor regardless of the value of the voltage of the source U. At the terminals of this source, there is first a first block 3 capable of supplying at its output 4 a voltage U _AB strictly proportional to the voltage of the source U _o . The voltage U _AB is then used in a second block 5 which has at its output 6 a first signal U _c , the growth speed of which is all the more rapid the higher the voltage U _AB . This variable signal U _c is repeated during each sampling period T _e . The signal U _c is finally compared in a third block 7 to a reference threshold U _r . Depending on whether the signal U is lower or higher than the reference threshold U, the block 7 produces a different logic state and shows on its output 8 and during each sampling period T _e an elementary pulse of duration T _o which closes the switch 2 the time that said lasts impulse.

FIG. 3 is a detailed diagram of the power supply to the motor and gives a preferred embodiment of the invention based on the block diagram of FIG. 1. The DC voltage source U _o supplies the motor M by the lines 10 and 11. The motor M is for example of the Lavet type at two steps per revolution. To form the bipolar pulses necessary for the advance of this motor, there is, interposed between its terminals 12 and 13 and the power source, a set of four switches 31, 32, 33 and 34 which are constituted by switching transistors controlled by their base. When the transistors 31 and 32 are conductive, the current flows through the motor according to arrow 14 for a period T. and the rotor turns by a first half-turn. After a period of time during which the motor is at rest (1 second for example), the transistors 33 and 34 are driven for the same period T _i , which has the effect of turning the rotor by a second half turn, the motor then being traversed by a current of opposite direction and according to arrow 15. The length of the pulse T _i , the periodicity with which it repeats and the alternating control of the pairs of transistors 31, 32 and 33, 34 (T and T signals) generally come from a common time base which in turn controls a multi-stage frequency divider from which these signals are taken. The production of signals T _i , T and T is known from the state of the art. However, it will be explained later (Figure 6).

To supply the motor with constant energy, the control pulse T _{i will be chopped} according to the device, the principle of which has been explained above.

la tension recueillie aux bornes de la seconde diagonale AB du pont s'écrit :

où il est apparent que U_AB est proportionnelle à la tension de la source U .In particular, the block 3 of FIG. 1 is, according to the preferred embodiment of FIG. 3, in the form of a bridge of resistors R ₁ , R ₂ , R ₃ and R ₄ . The first diagonal of this bridge is connected across the voltage source U _o , i.e. on lines 10 and 11. If the voltage U _A across the resistor R ₂ and the voltage U _B across the resistor R ₄ are written respectively:

the voltage collected across the second diagonal AB of the bridge is written:

where it is apparent that U _AB is proportional to the voltage of the source U.

This voltage U _AB is used to charge a capacitor C through a resistor R arranged in series with said capacitor, the network RC constituting in substance block 5 of FIG. 1. The voltage U across the terminals of capacitor C is written:

This expression (2) shows that the time T _x necessary to reach a determined voltage U _c is shorter the higher the voltage U _AB or, in other words, the higher the voltage U _c across the capacitor a speed of growth all the more rapid as the amplitude of the proportional voltage U _AB is high.

FIG. 3 also shows a comparator 16 which receives on its first input 17 the voltage U _c and on its second input 18 a reference threshold voltage U _r arranged in such a way that U _r <U _AB . The comparator 16 essentially constitutes the block 7 illustrated on the block diagram of FIG. 1. In the example of embodiment shown in the diagram of FIG. 3, as long as the voltage U _c is smaller than the voltage U _r , the output 19 of the comparator 16 is at logic level 1. As soon as U exceeds U, the output 19 goes to logic state 0.

From expression (2) above we can calculate the time T _x = T _o taken by the voltage U _c to reach the reference voltage U _r , that is:

If we now replace U _AB by its value according to expression (1) and we expose R ₁ = R ₂ = R ₃ = kR ₄ , (3) becomes:

It can be seen from expression (4) that the period T _o during which the comparator delivers a logic state 1 is all the greater the lower the voltage U _o of the power source.

FIG. 3 also shows that a switching transistor 20 has been placed at the terminals of the capacitor, the purpose of which is to periodically short-circuit said capacitor. The transistor 20 is controlled on its base by a short duration pulse T _c which makes it conductive, said pulse repeating at the rate T _e . Thus, at periodic intervals T _e , the voltage U _c present at the terminals of the capacitor is reset to zero by short-circuiting it. We know that each interval T _e , or sampling period, is composed of an elementary pulse T _o followed by an interruption period T _r . That said, we can write the following expression which expresses that the duty cycle or hash rate T _{o /} T _e is inversely proportional to the value U _o of the supply voltage:

The output 19 of the comparator 16 is connected to the first input of an AND gate 25, the second input receiving the duration signal T _i . In turn, the output of AND gate 25 is connected simultaneously to each of the first inputs of AND gates 21 and ET 22, the second input of each of said gates receiving signals T and T respectively which aim to change the polarity alternately of the driving pulse received by the coil of the motor M. The diagram also shows that the transistors 32 and 34 are supplied through diodes 24 and 23 respectively, these diodes can moreover be eliminated if one chooses transistors 31 , 34 and 32, 33 which are complementary (P and N).

FIG. 4 is a diagram showing the various signals existing at various points in the diagram of FIG. 3 when the supply voltage U _o is low (end of life of the battery, for example 1.5 V) and FIG. 5 shows these same signals when the voltage U ₀ is high (start of battery life, for example 3 V).

In FIG. 4, there is shown in a) the voltage U _AB at the terminals of the diagonal of the resistance bridge R ₁ , R _{2 '} R ₃ and R ₄ . The voltage U _AB is used to periodically charge the network RC and the voltage U _c across the capacitor appears in b) in FIG. 4. The reference voltage U _r is shown on the same diagram. The voltage U _c is periodically reset to zero by the switching transistor 20 itself controlled by a short duration pulse T _c repeating at regular intervals T _e . This control signal is shown in d).

According to the invention and as seen in e), the signal at the output 19 of the comparator 16 has a logic state 1 as long as U _c <U _r and a logic signal 0 as soon as U _c > U _r and the period d 'sampling T _e then consists of an elementary pulse of duration T followed by an interruption period T. It is understood that the signal illustrated in e) is found at the output of gate 25 as long as the signal T _i appearing in c) is present at the input of said gate 25.

We will now suppose that to progress by one step, the current in the motor must be directed in the direction of arrow 15. At this moment, the polarity pulse of duration T is at 1 while its counterpart T is at 0 The pulses T (FIG. 4e) pass through the AND gate 22, make the transistor 33 conductive and block the transistor 32 via the inverter 24. The same pulses T _o cannot pass through the gate 21 which then has a logic state 0 to his exit. Therefore, the transistor 31 is blocked while the transistor 34 conducts since it receives on its base a signal 1 via the inverter 23. The transistors 33 and 34 leading to each pulse of duration T _o , the current therefore crosses the motor in the direction of the arrow 15. During each pulse of duration T _r (FIG. 4e), there is a signal 0 at the output of gate 25 as well as a signal 0 at the output of each of the gates 21 and 22 It will then be understood that the transistors 31 and 33 are blocked and that the transistors 32 and 34 are conductive: during the period T, the motor coil is short-circuited. Thus, in this case, the motor is supplied by the voltage shown in FIG. 4f, a voltage which is no longer continuous, as would be the case with the control by simple pulse of duration T _i , but chopped at the rate of T _o / T _e and amplitude U _o .

If you are dealing with a two-step motor, you will have to reverse the direction of the current (arrow 14) to take the next step. At this time, the polarity pulse of duration T is at 0 while its counterpart T goes to 1. The same reasoning as that which was made above shows that in this case, it is the transistors 31 and 32 which are conductive while the transistors 33 and 34 are blocked during the periods T _o . On the other hand, during the periods T _r , the motor coil is short-circuited on the transistors 32 and 34 which are conductive.

FIG. 5 shows how the device according to the invention behaves when the supply voltage is doubled. Without going into details, we can see that the voltage U _c is more quickly equal to the reference voltage U, which has the consequence of shortening the supply periods T _o and reducing the chopping rate T _{o /} T _c . At the motor terminals, there are will finally see the power supply shown in FIG. 5f, of doubled amplitude but of application time T _o markedly reduced.

To take a practical example and based on the equations given above, we will assume that it is necessary to supply the motor with constant energy if the supply voltage varies between 3 and 1 volts. By choosing R ₁ = R _{2 =} R ₃ ⁼ 1 MΩ and R ₄ = 250 kn, we find according to equation (1) U _AB = 0.3 U ₀ . By also choosing a reference voltage U _r = 0.1 V, a capacitor C = 0.01 _µ F and a resistance R = 82 kΩ, we can draw up the following table:

According to these data, the sampling period T _e greater than 206 µs, or 244 µs, must be chosen to use a practical value, which can be obtained directly from the frequency divider. We directly derive the hash rate T _o / T _e indicated above.

It should be noted that, in the values given by way of example, the time which it takes to short-circuit the capacitor C has not been considered, time which is seen to appear in FIGS. 4d and 5d and symbolized by the period T _c . To remove this period T _c from the sampling period proper T, we could propose instead of the simple switching transistor 20 (FIG. 3) another arrangement which would make the period T _e independent of the period T _c . The period T _c is chosen here for practical reasons at 30 µs.

It is important to note that the reference voltage U _r must be chosen inter alia according to the characteristics of the engine we're dealing with. It is a rigorously stabilized voltage independent of fluctuations in the power source voltage. A practical example of embodiment of such a generator can be found in the document CH-A-639 810 in FIG. 4e.

The duration T _i of the pulse during which the sampling takes place must be chosen long enough so that in the most unfavorable cases of minimum voltage U _o , the rotor will surely pass its pitch. This duration is of the order of 7.8 ms, duration which can also be taken from the frequency divider available. It will be understood that 32 sampling periods T _e each worth 244 µs may take place during a period T _i of 7.8 ms.

It should be noted that the system proposed in the present invention could be combined with a system for controlling the motor to the load which it must drive. These systems generally lengthen the driving impulse when the load increases and vice versa. This slave pulse could be supplied to the input of gate 25 of FIG. 3 in place of the fixed pulse T _i that we have considered so far.

The diagram in FIG. 6 shows how the pulses T, T, T _i , T _e and T _c necessary to operate the device illustrated in FIG. 3 are produced. All the signals are obtained from a controlled time base 40 by a quartz 41 via a frequency divider 42. The type D flip-flop 43 combines the signals at 16 384 Hz and at 4096 Hz to produce the sampling period T _e at 244 µs, as well as at Within this period, the pulse T _c which short-circuits the capacitor C. Another flip-flop 44 of type D combines the signals at 64 Hz and 1 Hz to produce the pulse T _i at 7.81 ms repeating every second. The output Q of the flip-flop 44 in turn attacks a divider by two 45 which generates the pulses of polarity T and T.

To prevent the proposed device from consuming energy between the supply periods T _i , it can be arranged that it is only supplied during said supply periods. FIG. 7 shows that the resistance bridge R ₁ , R ₃ , R ₂ , R ₄ , of which only the resistors R ₁ and R ₃ have been shown in the drawing, is supplied through a transistor 50 which is only conductive during the periods T _i . It appears from this that the resistance bridge is switched off during the periods separating said periods T _i . Anyway, the resistance values of the bridge will be taken as high as possible.

Claims (7)

1. Device for supplying constant energy from a variable voltage source (U _o ) to a stepping motor, in particular for a timepiece, the supply consisting in supplying the motor with pulses of constant duration T _i , each being chopped according to a chopping rate depending on the value of the source voltage (U _o ) and constituted by a series of elementary pulses of duration T _o separated by periods of interruption T _r , the sum T _o ⁺ T _r defining a predetermined sampling period T _e , characterized in that it comprises first means for taking a voltage (U _AB ) proportional to the voltage of the source (U _o ), second means using this proportional voltage to generate, during each sampling period T _e , a signal (U _c ) having a variable growth rate which is all the more rapid as the amplitude of said proportional voltage (U _AB ) is high and third means for comparing the said signal (U _c ) at a reference threshold (U _r ), the comparison determining a different logic state depending on whether said signal (U) is less than or greater than said reference threshold (U _r ) to generate said pulse of elementary duration T _o and define a ratio T _o / T _e representing said hash rate. 2. Device according to claim 1, characterized in that the first means comprise a resistance bridge (R ₁ , R ₂ , R ₃ , R ₄ ) the first diagonal of which is connected to the terminals of the voltage source (U) and the second diagonal (AB) of which provides a voltage proportional (U _AB ) to the voltage of said source. 3. Device according to claim 1, characterized in that the second means include a resistor (R) and a capacitor (C) arranged in series and supplied by the proportional voltage (U _AB ) to supply the terminals of said capacitor with the signal ( U) having a variable growth rate. 4. Device according to claim 3, characterized in that it further comprises fourth means (20) for short-circuiting the capacitor (C) and thus resetting to zero the signal (U) present at its terminals at the start of Sampling period each ^g e ^T _e. 5. Device according to claim 1, characterized in that the third means comprise a comparator circuit (16) with two inputs, the first receiving the signal (U _c ) at variable growth speed and the second input receiving a voltage fixed to the reference threshold (U _r ). 6. Device according to claim 1, characterized in that the driving pulses of duration T _i and each sampling period T _e are produced from a common time base (40, 41). 7. Device according to claim 1, characterized in that it further comprises an arrangement (50) for limiting the sampling of the proportional voltage (U _AB ) to the times that the driving pulses of duration T _{i last} .

Images