JPH0352589B2 - - Google Patents

Description

[Detailed description of the invention]

The present invention mainly relates to a control circuit for a method of driving a step motor of an electronic timepiece using low power and preventing detection errors. For step motors that require low power consumption, such as ultra-compact step motors for electronic wristwatches, methods for reducing power consumption include improving the electric-mechanical conversion efficiency of the step motor itself, as well as reducing power consumption during normal operation. A so-called correction drive method has been devised in which when the rotor does not rotate normally for some reason, it is quickly re-driven with higher power than usual. When adopting this correction drive method, the important thing is how to reliably detect rotation or non-rotation of the rotor. Figure 1A is an example of a two-pole step motor that has been used to drive the hands of conventional electronic watches and is also used in the present invention.
is an example of an inverted pulse conventionally used to drive a step motor having this structure. By applying the drive pulse shown in FIG. 1B to the coil 3, the stator 1 is magnetized, and the rotor rotates 180 degrees due to the repulsion and attractive force from the magnetic poles of the rotor 2.
Conventionally, the length of the applied drive pulse has been selected to ensure the output of the motor under all conditions that should be guaranteed for a watch. However,
In this case, there is a calendar load, a large internal resistance of the battery,
It was necessary to include a margin for voltage drop in the final stage, and it was necessary to drive with a pulse width that had a margin. Therefore, by improving this method, the step motor is usually driven with the narrowest possible pulse width, and then a detection circuit is installed to determine whether the rotor has rotated or not. A method has been proposed in which corrective driving is performed with a conventional pulse width only when it is determined that the rotation is occurring. It is difficult to provide a special detection element such as a mechanical contact or a Hall element to detect rotation or non-rotation of the rotor due to the demands for smaller, thinner, and lower cost watches. be. Therefore, a method is used to detect whether the rotor is rotating or not, by taking advantage of the fact that the voltage generated by the vibration of the rotor after applying a drive pulse is different depending on whether the rotor is rotating or not. The present invention aims to improve this correction drive method. FIG. 2 is a circuit example of a step motor driving and detecting section used in the prior art and in the present invention.
This circuit configuration consists of N-channel FET gates (hereinafter abbreviated as N-gate) 4b, 5b and P-channel
FET gate (hereinafter abbreviated as P gate) 4a, 5a
detection resistors 6a, 6b for separating the inputs of the rotor 2 and detecting rotation or non-rotation of the rotor 2, and N gates 7a, 7b for switching these resistors.
It is equipped with FIG. 3 is a time chart in the conventional correction drive system. The voltage across the coil is
In the section shown in FIG. 3a, a current flows like the current path 9 shown in FIG. Next, in the middle of FIG. 3b, the circuit is switched to a closed circuit including the detection resistor 6b like the closed circuit 10 shown in FIG. At this time, a voltage generated by vibration of the rotor 2 after application of the drive pulse is generated at the terminal 8b. If a non-rotation signal is detected in the detection section b, current is applied to the coil 3 again through the current path 9 in FIG. 2 in the section shown in FIG. Performs correction drive of the step motor. Next, the principle of rotor rotation/non-rotation detection will be explained in detail. Figure 4 shows the current waveform when current is passed through the coil 3 of a step motor with a coil resistance of 3KΩ and 10,000 turns. This is a current waveform when the drive pulse length a is 3.9 msec, and shows almost the same waveform regardless of rotation or non-rotation. The section in FIG. 4b is an induced current due to the vibration of the rotor 2 after application of the drive pulse, and the current waveform in this section changes greatly depending on whether the rotor 2 is rotating, not rotating, under no load, or under load. The section of Figure 4b
The waveform b ₁ is the current waveform when the rotor 2 is rotating, and b ₂ is the current waveform when the rotor 2 is not rotating. The drive detection circuit shown in Figure 2 was invented to take out the difference in current due to rotation and non-rotation as a voltage waveform.
Switch the circuit to 0. By doing so, the current generated by the vibration of the rotor 2 flows through the detection resistor 6b, so that a larger voltage waveform appears at the terminal 8b than when no detection resistor is attached. The current flowing in the positive direction in section b is in the opposite direction to the detection resistor 6b in the closed circuit 10 of FIG. 2, and thus appears as a negative voltage. Further, in the OFF state, the N gate 5b has a PN junction between the drain and the P ^- well, and functions as a diode with V _SS as an anode. Therefore, the negative voltage seen from the terminal 8b flows through the N gate 5b, which acts as a diode, and has an impedance similar to that of the closed circuit 11, thereby applying braking to the rotor. The relationship between the function of the rotor 2 and the detection signals will be explained using FIG. 5. FIG. 5 shows the relationship between the stator 1 and the rotor 2, and FIG. 5A shows the rotor 2 in a stationary state. and outer peripheral notches 15a, 15 for integrating the stator.
There is b. However, in the case of a two-piece stator, 15
The stator is separated at portions a and 15b. When the rotor 2 is at rest, the inner notches 16a,
The N and S magnetic poles come to rest at positions approximately 90 degrees from 16b. FIG. 5B is a diagram when a driving pulse is applied to this, and the rotor rotates in the direction of the arrow. Since the driving pulse width is as short as 3.9 msec, the pulse ends when the rotation is almost at the inner notch. When the load is small, the rotor can rotate completely due to its inertia, but when the load is large, it cannot rotate completely and the rotor rotates in the opposite direction, as shown in FIG. 5C. At this time, the magnetic poles of the rotor 2 pass near the outer peripheral notches 15a, 15b, so a large current is generated in the coil. However, at this time, the second
Since it is a closed circuit 10 as shown in the figure, as explained earlier, a negative voltage is generated at the terminal 8b,
A diode forward current flows through the N gate 5b,
Braking is applied to the rotor 2. Therefore, rotor 2
is rapidly decelerated, and thereafter the voltage generated by the vibration of the rotor 2 is small. On the other hand, when the load is small and the rotor 2 rotates, the rotor 2 rotates in the direction of arrow 19 as shown in FIG. 5D. At this time, the magnetic flux generated by the rotor 2 is perpendicular to the outer peripheral notches 15a, 15b, so the induced current is initially small. Furthermore, the magnetic pole has an outer peripheral notch 15a,
When it rotates to around 15b, a large current is generated. At this time, since a negative voltage is generated at the terminal 8b of the closed circuit 10, the rotor is braked due to the diode effect of the N gate 5b. Thereafter, when the rotor rotates considerably beyond the rest position shown in FIG. 5A and returns to the rest position, a voltage is generated at the terminal 8b in FIG. 2 that allows the rotation of the rotor 2 to be detected. The voltage waveform 20 in FIG. 6A is the voltage waveform at the terminal 8b when the rotor 2 described above rotates. The section a is the drive pulse application time, which is 3.9 msec.
The circuit at this time is the current path shown in Figure 2, and V _DD =
It is 1.57V. Section b in Figure 6A is the voltage induced by rotor vibration, and is similar to closed circuit 1 in Figure 2.
This is the voltage waveform when the voltage is 0. The negative voltage is clipped at about -0.5V due to the diode effect of the N gate 5b, and the peak of the positive voltage is 0.4V. On the other hand, waveform 21 is for non-rotation,
The peak of the positive voltage is 0.1V or less, and by distinguishing between these two voltages, it is possible to determine whether the rotor is rotating or not. Furthermore, although the difference between these two voltages is small, it can be easily amplified by the method described below. In the section indicated by b in FIG. 6A, the closed circuit 10 and the closed circuit 11 in FIG. 2 are alternately switched. In the closed circuit 11, N gates 4b and 5b are connected.
Since both ends of the coil 3 are shot with an element having an ON resistance of about 100Ω, the current due to rotor vibration is large. However, when switching to the closed circuit 10, the current momentarily flows through the detection resistor 6b due to the inductance component of the coil 3.
Therefore, a momentary high peak voltage is generated across the detection resistor. When the induced voltage waveform 20 due to the rotor 2 during rotation is alternately switched between the closed circuit 10 and the closed circuit 11 shown in FIG. 2, the voltage waveform at the terminal 8b becomes as shown in FIG. 6B. Voltage waveform 2 at this time
The time axis enlarged waveforms of Nos. 2 and 23 are shown in FIG. 6C. At this time, the peak voltage is delayed by about 30 μsec after switching to loop 10. This is N gate 5
This is because there is a capacitance component between the drain and source of b, which causes a delay in the peak voltage. In recent years, such a method has been proposed, and it has become easier to detect whether the rotor is rotating or not. With this detection method, in addition to the above-mentioned method in which the normal drive pulse width is fixed, in order to further reduce the power consumption of the step motor, we can detect the minimum pulse width that allows rotation of the normal drive pulse width. A driving method has been realized. FIG. 7 is a graph showing the relationship between drive pulse width and torque of a step motor used in an electronic timepiece, in the conventional case and in this embodiment. In the case of fixed pulse drive, the drive pulse width is set at point a in order to guarantee the maximum torque Tgmax of the step motor. The method of performing correction drive is that if the point Tqc is the torque required for calendar feed, the length of the normal drive pulse is a ₁ = 3.4 msec.
The length is set to a ₂ = 3.9 msec.
The reason is that when the rotor cannot rotate completely with the normal drive pulses, an additional correction pulse is added, so if the correction pulse appears too many times, the current consumption of both is added, which causes the current to increase. This is because it is possible that there will be an increase. However, in reality, even with a pulse width of a ₀ =2.4 msec, the rotor rotates when there is no load, so if the rotor can be driven with this pulse width, it is possible to further reduce current consumption. Its operation will be explained with reference to FIG. Normally, the step motor is driven with a pulse width of a ₀ = 2.4 msec, and if the rotor cannot rotate completely with the pulse width of a ₀ due to calendar load, etc.
The detection circuit determines that the rotor is not rotating, and immediately drives it with a corrected drive pulse. The pulse width of this correction drive is generally a pulse width of a=7.8 msec in FIG. 7. The driving pulse width after the next second is a ₁ = slightly longer than a ₀ = 2.4 msec.
A pulse width of 2.9 msec is automatically set as a normal drive pulse, and the drive pulse is applied to the step motor. However, according to the example in Figure 8,
Since the calendar torque Tqc is not reached even when a ₁ =2.9 msec, the rotor stops rotating and is immediately driven with the correction pulse a =7.8 msec. Then, the normal drive pulse after one second automatically becomes a ₂ =3.4 msec. The output pulse in this case is the calendar torque
Since it is larger than Tqc, the step motor is thereafter driven with a pulse width of a ₂ =3.4 msec per second. However, if this continues, the pulse width of a ₂ =3.4 msec will continue even when the calendar load is removed, which is disadvantageous for reducing power consumption. Therefore, by adding a circuit that shortens the drive pulse every N seconds, a ₂ = 3.4 msec is output N times in succession, and once output, the pulse width returns to a ₁ = 2.9 msec. Furthermore, when a ₁ is output N times in succession, it becomes a ₀ . Conversely, when the maximum pulse width of the normal drive pulse is a ₂ = 3.9 msec,
If non-rotation is detected, in the next 1 second after the correction pulse is output, the same a ₃ =
3.9msec is output. In this way, in order to set one of the plurality of normal drive pulses, an up-down counter as shown in FIG. 9 is required. As explained above, it is possible to detect the rotation or non-rotation of the rotor, and also to drive the step motor at the lowest pulse width that can rotate the normal drive pulse width, thereby reducing the power consumption of the step motor. . However, this conventional example has a major drawback. It can be explained by FIG. FIG. 10 shows the relationship between the normal drive pulse width and the detected voltage. When the pulse width exceeds a certain value, the detected voltage decreases rapidly.
This is because normally, as the drive pulse width increases, the rotor is attracted to the magnetic poles and vibrations are damped. Further, the value of this characteristic curve changes depending on the motor. In the past and in the present invention, in order to distinguish between rotation and non-rotation of the rotor,
The detection voltage is set to a certain reference voltage by a comparator.
If it is more than V _COMP , it is judged as rotation, and the detection voltage is
If it is less than V _COMP , it is judged as non-rotating. All of the above controls are performed by an electronic circuit built into the watch. As shown in Figure 10, when V _COMP = 0.8V, when the rotor rotates at a ₃ = 3.9msec, the detected voltage will be approximately 0.9V, and the electronic clock will detect the rotation and to decide. However, the values of this characteristic curve vary depending on the motor, so in other motors, the detected voltage during rotor rotation may be smaller than V _COMP = 0.8V at a ₃ = 3.9msec. . In this case, the electronic circuit of the watch determines that it is not rotating and outputs a correction pulse. The output of the correction pulse does not disturb the movement of the clock's hands. However, a large current flows due to the output of the correction pulse. The detected voltage due to the next rotation will similarly become smaller than V _COMP = 0.8V, and it will be determined that the rotation is non-rotating, and a correction pulse will be output. thus,
Correction pulses are output one after another, and the current for rotating the rotor of the step motor increases rapidly.
Once a conventional rotation detection control circuit enters such a state, it continues to be in that state and cannot escape from that state. Therefore, in the conventional circuit, it is not possible to increase the drive pulse width due to the design. This increases the probability that a correction pulse will be output and increases current consumption. FIG. 11 is a block diagram of an electronic circuit of an electronic timepiece according to the prior art and the present invention. Reference numeral 31 denotes an oscillation section, in which a time reference signal is generated by a crystal oscillator. This signal is frequency-divided by the frequency dividing section 32 to form a plurality of frequency signals. This signal enters the waveform shaping section 33, which converts the step motor drive pulse, timing pulses necessary for detection, etc. into an AND gate.
Synthesize using OR gates, flip-flops, etc. The drive/detection section 35 is composed of the circuit shown in FIG. 2 and a comparator that judges the detection signals outputted to 8a and 8b. Additionally, the step motor drive output is tangential to the step motor 36. The detection signal of the detection section is fed back to the control section 34. FIG. 9 shows an up-down counter for normal drive pulse control included in the control section of FIG. 11. When _S3 becomes a high logic level, such as when the power is turned on, the counters 53 and 51 are initialized, and the output signals α and β become a low logic level. (Hereinafter, a logic level high is indicated as 1, and a logic level low is indicated as φ.) The up signal _S1 and the down drive _S2 are at φ in normal times. As mentioned earlier, when it is determined that there is no rotation, S ₁ becomes 1, and the normal state φ
becomes. By doing this, α becomes 1. S ₁
is 1, and when φ is repeated, α and β become No. as shown in Table 1.
It changes from 1 to No. 4, and when α=1 and β=1, it does not change any further. Also, as mentioned earlier, after a certain period of time, for example 80 seconds, S ₂
becomes 1 and φ, and α and β are No. 4, contrary to the previous
It changes from No. 1 to No. 1, and when α=φ and β=φ, it does not change any further. By setting the pulse width of the normal drive pulse using these α and β signals, it is possible to control the normal drive pulse with rotation and non-rotation as described earlier.

[Table] For the sake of simplicity, this explanation is based on a 2-bit up-down counter, α and β, but in reality, 3 or more bits are possible, and in the case of 3 bits, the normal pulse width of 8 steps is It is possible. Applying the previous explanation to this circuit, α=1, β
This means that the state of =1 continues. In order to eliminate these drawbacks, the present invention makes the change step of the normal drive pulse cyclic, and will be described in detail below with reference to the drawings. FIG. 12 shows an embodiment of the present invention, and the drawing numbers are the same as in FIG. 9. FIG. 12 is a circuit diagram of the up-down counter of the control section in the present invention. When S ₃ becomes 1 and φ, α=φ and β=φ are initialized. When the rotor of the step motor is not rotating, the up pulse _S1 is 1, φ, and α=1,
β=φ. After that, the optimum normal pulse width is selected and the values of α and β are set. In the present invention, in the example explained in the conventional example, for example, V _COMP = 0.8V
So, when the normal drive pulse is a ₃ = 3.9 msec, the detection voltage is V _COMP = even though it is rotating.
Assume that the voltage is about 0.7V, which is less than 0.8V. in this case,
The detection circuit determines that it is not rotating, sets S ₁ to 1, and sets φ. At this time, in the conventional example, counting up was prohibited by the AND gate 41, but in the present invention, the counter is counted up and
α=φ, β=φ. By doing this, it is possible to escape from the undetectable tilt area. After escaping, the optimal normal drive pulse width is selected and stabilized. Regarding the down pulse _S2 , as in the conventional example, when the normal drive pulse width becomes the minimum, the AND gate 54 prohibits the countdown. As described above, by making the normal drive pulse cyclic, the maximum pulse width of the normal drive pulse can be increased in terms of design. If the maximum pulse width of the normal drive pulse is increased,
Even if the rotation detection malfunctions and the rotation is determined to be non-rotation, the situation is immediately escaped and the optimum normal drive pulse width is selected. Therefore, by increasing the maximum pulse width of the normal drive pulse,
It is possible to reduce the probability of correction pulses and reduce the current consumption of the motor. Further, detection failures due to manufacturing variations in the motor can be relieved. Conventionally, the type of drive pulse was normally set depending on the type of motor, but this is not necessary in the present invention because the design range of settings is expanded. Further, the present invention has very great effects, such as being less complicated than conventional circuits and having extremely high safety against erroneous rotation detection. Further, the detection method according to the present invention is applicable not only to the range described above, but also to all methods of detecting the induced voltage of a step motor. Furthermore, in the present invention, the state of the drive energy applied to the step motor is realized by changing the pulse width of the drive pulse, but other methods, such as changing the waveform and duty ratio of the drive pulse, are also possible. The scope of its application is extremely wide, as it can also be achieved by methods such as changing the current value.

[Brief explanation of drawings]

FIG. 1A is a perspective view of a step motor for an electronic timepiece used conventionally and in the present invention. Figure 1B
is a conventional step motor drive pulse waveform diagram. FIG. 2 is a part of a drive and detection circuit diagram of a step motor according to the prior art and the present invention. FIG. 3 is a step motor drive pulse waveform diagram of a conventional visual drive system. Figure 4 shows the current waveform of the step motor and the current waveform induced by rotor vibration when the rotor is rotating and not rotating. FIG. 5A is a positional relationship diagram of the stator and rotor when the rotor is stationary. Figure 5B
FIG. 2 is a diagram showing the rotation direction of the rotor when driving pulses are applied. FIG. 5C is a diagram showing the movement of the rotor when the rotor cannot rotate. FIG. 5D is a diagram showing the movement of the rotor after application of a drive pulse when the rotor rotates. FIG. 6A shows the rotor rotating and non-rotating,
Voltage induced in the detection resistor. FIG. 6B shows the voltage waveform induced in the detection resistor when the rotor is rotating and not rotating and switching between a closed circuit containing high resistance and a closed circuit containing low resistance. FIG. 6C is an enlarged waveform diagram of FIGS. 22 and 23. FIG. 7 is a characteristic diagram showing the relationship between drive pulse width and minute hand torque. FIG. 8 is a drive pulse waveform diagram showing a correction drive method in which the drive pulse width changes according to load fluctuations. FIG. 9 is a circuit diagram of an up-down counter for setting normal drive pulses in a conventional example. FIG. 10 is a diagram showing the relationship between normal drive pulse width and detection voltage. FIG. 11 is a block diagram of an electronic circuit of a conventional electronic timepiece and an electronic timepiece according to the present invention. FIG. 12 is a circuit diagram of the up-down counter of the control section in the present invention. 1... Stator, 2... Rotor, 3... Coil, 10
1... Gate terminal of P gate 4a, 102... Gate terminal of P gate 5a, 103... Gate terminal of N gate 4b, 106... Gate terminal of N gate 7b,
41, 44, 48, 52, 54...AND circuit, 4
2, 45, 49...NOR circuit, 43, 46, 47,
50...NOT circuit, 51, 53...Flip-flop circuit with reset.

Claims (1)

[Claims] 1 At least an oscillation circuit, a frequency dividing circuit, a drive circuit,
In an electronic timepiece comprising a step motor, a rotation detection circuit that detects rotation based on an induced voltage generated in a coil when the rotor of the step motor undergoes damped vibration, and a battery, the frequency dividing circuit and the drive circuit a waveform forming circuit connected between the rotation detection circuit and the waveform forming circuit capable of outputting a plurality of normal drive signal trains having different effective values and at least one correction drive signal having a larger effective value than any of the normal drive signal trains; A normal drive signal with a larger effective value is selected as the next normal drive signal based on the non-rotation detection signal from the rotation detection circuit connected to the formation circuit, and a non-rotation signal with a larger effective value is selected as the next normal drive signal. An analog electronic timepiece characterized by comprising a control circuit for setting a normal drive signal that selects a normal drive signal with a smaller effective value when rotation is detected.