JPH07131087A - Piezoelectric element drive device - Google Patents

Abstract

PURPOSE:To reduce the loss due to the discharge of electric charge of a piezoelectric element by comparing a displacement command value with the voltage of a first charging/discharging element and then controlling the direction of electric charge transfer between first and second charge/discharge elements according to polarity and the amount of transferred electric charge according to the level. CONSTITUTION:A displacement command value or a corresponding voltage is compared with the actual value of a first charge/discharge element 6b, the direction of electric charge transfer between first and second charge/discharge elements is selected according to the polarity of operation result, and then the amount of transferred electric charge is controlled by either commutation circuit 10 according to the level. When the displacement actual value of the piezoelectric element 6b is smaller than the displacement command value, electric charge is commutated from a charge/discharge element 6a to the piezoelectric element 6b for expanding the piezoelectric element 6b. When it is larger, electric charge is commutated from the piezoelectric element 6b to the charge/discharge element 6a to shrink the piezoelectric element 6b. Even if, for example, a power supply voltage and load fluctuate and the amount of displacement fluctuates, the displacement actual value of the piezoelectric element is finely subjected to feedback control and is precisely matched to the displacement command value, thus achieving drive with a small power.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a piezoelectric element driving device. More specifically, the present invention relates to a piezoelectric element driving device that drives a piezoelectric element with low power.

[0002]

Background Art and Problems There are cases where expansion and compression of a piezoelectric element are alternately repeated, or a plurality of piezoelectric elements are alternately driven. In such a case, however, conventionally,
The piezoelectric element is charged and driven by the driving power supply, and after the work is completed, the piezoelectric element is discharged and contracted as before.When the piezoelectric element is driven again, the piezoelectric element is charged again from the driving power supply. Was there.

In order to reduce waste of power consumption due to repeated charging / discharging of the piezoelectric element and to reduce the size of the piezoelectric element driving apparatus, the applicant of the present invention has found that the piezoelectric element drive device is arranged between two piezoelectric elements (or between the piezoelectric elements). It was considered that the electric charge for driving the piezoelectric element can be repeatedly used by commuting the electric charge between the capacitors. This piezoelectric element drive device
It has been filed as Japanese Patent Application No. 4-292209.

Since this piezoelectric element driving apparatus employs an impulse resonance type driving method, it is effective in the case where a momentary displacement amount and displacement speed are required such as in a press machine, but it is effective for disturbance. If there is a load fluctuation such as, it is not possible to maintain a constant displacement.

Therefore, in applications where it is necessary to maintain a constant displacement with respect to load fluctuations such as disturbances, it is possible to transfer charges mutually between the piezoelectric element and the reference capacitor, and at the same time from the auxiliary charging power source to the reference capacitor. Auxiliary charging can be performed, and the voltage of the reference capacitor after charge transfer and the voltage of the auxiliary charging power source are precisely matched by the auxiliary charging to adjust the voltage, and the charge is transferred from the reference capacitor controlled to a constant voltage to the piezoelectric element. A method of causing the piezoelectric element to flow can be considered.

However, in this method, since the auxiliary charge is used, it is impossible to make a delicate response of the displacement amount of the piezoelectric element to a load change. It is difficult to reduce the voltage, and if a circuit for discharging is added, power will be consumed and the energy saving effect will be reduced.

[0007]

SUMMARY OF THE INVENTION The present invention has been made in view of the drawbacks of the above-mentioned conventional examples, and an object thereof is to feed back the voltage or displacement amount of a piezoelectric element,
In an application where the value is held for a certain period, it is possible to drive the piezoelectric element with low power consumption without losing electric charge as much as possible by discharge.

[0008]

A piezoelectric element driving apparatus according to the present invention comprises first and second charging / discharging elements in which at least a first charging / discharging element is a piezoelectric element, and first to second charging / discharging elements. A commutation circuit for commutating charges to the charge / discharge element of
A commutation circuit for commutating charges from the second charging / discharging element to the first charging / discharging element, a charging circuit for charging the second charging / discharging element from the driving source, a displacement command value or a displacement command. A means for comparing and calculating the voltage corresponding to the value and the actual value of the first charge / discharge element; and a means for selecting the charge transfer direction between the first and second charge / discharge elements according to the polarity of the calculation result. ,
And a means for controlling the transfer charge amount by any one of the commutation circuits according to the magnitude of the calculation result.

Further, in the above piezoelectric element driving device,
Means may be provided for determining whether charging of the second charge / discharge element from the charging circuit is necessary.

As the piezoelectric element driving device, means for intermittently generating the first signal, means for delaying the first signal for a certain period of time to generate the second signal, and the commutation circuit Means for generating a third signal by detecting that the current value of the commutation current flowing through the capacitor reaches a constant value, and a transfer from the second charge / discharge element to the first charge / discharge element by the first signal. Flow operation is started, and the second or third signal
And a means for stopping the commutation operation from the charging / discharging element, and charging the second charging / discharging element when the third signal is not generated at least between the first signal and the second signal. It may be provided with a means for making it.

In the piezoelectric element driving device, the commutation circuit diverts charges from the second charging / discharging element to the first charging / discharging element by a plurality of commutation operations. The cycle of the commutation operation may be gradually changed according to the result of comparison between the command value and the actual value of the first charge / discharge element evaluated by the calculation means.

Further, in the above piezoelectric element driving device, the second charge / discharge element may be a capacitor or a piezoelectric element.

[0013]

In the piezoelectric element driving device of the present invention, the displacement command value or the voltage corresponding to the displacement command value is compared with the actual value of the first charge / discharge element, and the polarity of the calculation result is used. A means for selecting a charge transfer direction between the first and second charge / discharge elements according to the above, and a means for controlling the transfer charge amount by any one of the commutation circuits according to the size of the calculation result. Therefore, when the actual displacement value of the first charging / discharging element is smaller than the displacement command value,
The charge can be transferred to the charging / discharging element of No. 1 to expand the first charging / discharging element, and conversely, if the actual displacement value of the first charging / discharging element is larger than the displacement command value, The charge can be transferred from the charge / discharge element to the second charge / discharge element to contract the first charge / discharge element. Therefore, even if there is a change in the power supply voltage, a change in the load, a change in the displacement amount due to disturbance, etc., the actual displacement value of the first charging / discharging element can be finely feedback-controlled to precisely match the displacement command value. .

Further, since the charging circuit for charging the second charging / discharging element is provided, the commutation current to the first charging / discharging element does not become insufficient, and the second charging / discharging element does not. First
It is possible to secure necessary charges for charging the charging / discharging element.

Further, if the piezoelectric element driving device is provided with a means for judging the necessity of charging the second charging / discharging element from the charging circuit, when the charge amount of the second charging / discharging element is insufficient. Can quickly charge the second charging / discharging element from the charging circuit, and the commutation current from the second charging / discharging element to the first charging / discharging element does not become insufficient.

Further, in the piezoelectric element drive device in which the charges are communicated from the second charge / discharge element to the first charge / discharge element by a plurality of commutation operations, the commutation operation is performed once. It is not necessary to fully charge the first charging / discharging element, and the first charging / discharging element is divided into several times until the command value is reached.
Since a charge can be transferred to the second charging / discharging element, a DC power supply having a voltage lower than the driving voltage of the first charging / discharging element is used for the second charging / discharging element or a DC power supply for auxiliary charging the element. be able to. Therefore, a small charging circuit or a DC power supply can be used, and the piezoelectric element driving device can be made small and lightweight.

Further, means for intermittently generating the first signal, and means for detecting that the current value of the commutation current flowing in the commutation circuit has reached a constant value and generating the third signal. Means for starting the commutation operation from the second charge / discharge element to the first charge / discharge element by the first signal and stopping the commutation operation from the second sub-discharge element by the third signal. In such a case, since the peak value of the commutation current can be made constant, the utilization factor of the switching element used in the commutation circuit can be improved. Moreover, when the second signal is generated with a certain time delay from the first pulse signal and the third signal is not generated, the commutation operation is stopped by the second signal, and the second charge / discharge element is generated. When the peak value of the commutation current drops, the second charging / discharging element should be immediately auxiliary charged to return the peak value of the commutation current to a constant value again. You can

Furthermore, when commutating to the first charging / discharging element by a plurality of commutation operations, the commutation is performed according to the result of comparison between the actual displacement value of the first charging / discharging element and its displacement command value. If the cycle of the operation is gradually changed and, for example, the cycle of the commutation operation is inversely proportional to the difference between the actual displacement value and the displacement command value, even if the difference between the actual displacement value and the displacement command value is large, the first The charge / discharge element can be quickly displaced to the command value.

[0019]

[First Embodiment] FIG. 1 is a main circuit connection diagram showing a piezoelectric element driving device A according to an embodiment of the present invention. In this piezoelectric element driving device A, a series connection circuit in which the cathode of the diode 9a and the drain of the MOSFET 8a are connected, the free wheeling diode 5a, and the reference capacitor 6a.
And 3 are connected in parallel so as to connect the anode of the diode 9a, the cathode of the free wheeling diode 5a, and the positive electrode side of the reference capacitor 6a (however, when the reference capacitor 6a has no polarity, the polarity can be ignored). It is connected. Similarly, the cathode of the diode 9b and the MOSFE
The series connection circuit in which the drain of T8b is connected, the free wheeling diode 5b, and the piezoelectric element 6b are respectively an anode of the diode 9b, a cathode of the free wheeling diode 5b, and
Three parallel connections are made so as to connect to the positive electrode side of the piezoelectric element 6b. These two 3 parallel connection circuits are MOSFET
The sources 8a and 8b are connected in series so as to connect the sources to each other. An arm composed of a thyristor 3a and a commutation diode 4a, which are connected in anti-parallel so as to connect the anode of the thyristor 3a and the cathode of the commutation diode 4a, has the cathode side of the thyristor 3a connected to the anode of the diode 9a. Then, the thyristor 3b and the commutation diode 4b are connected in anti-parallel so as to connect the anode of the thyristor 3b and the cathode of the commutation diode 4b.
The arm made of is connected to the anode of the diode 9b at the cathode side of the thyristor 3b. Furthermore, these two antiparallel connection arms are commutating diodes 4a,
The cathodes of 4b are connected by the inductor 2 to form a closed circuit, and the closed circuit forms the commutation circuit 10.

3 parallel connection circuits Connection points of the same, ie, M
The connection points between the sources of the OSFETs 8a and 8b are connected to the positive electrode side of the reference capacitor 6a via the DC power supply 1, the charging current limiting resistor 11, the protective fuse 12, and the charging thyristor 7, and the closed charging circuit is thereby formed. 13 are configured.

The current flowing in the commutation circuit 10 can be detected by a current detector 14 such as an instrument current transformer (CT) provided between the inductor 2 and the thyristor 3b. The current value detected by 14 is output to the control circuit 16 through the current detection / comparison circuit 15.

Further, the gates of the thyristors 3a and 3b and the charging thyristor 7 are controlled by the control circuit 16 through the thyristor gate drive circuit 17 in the turn-on timing. Similarly, the gates of the MOSFETs 8 a and 8 b are on / off controlled by the control circuit 16 via the MOSFET gate drive circuit 18.

FIG. 2 is a circuit diagram showing the configuration of the control circuit 16, and FIGS. 3 (a) to 3 (n) are actual displacement values L of the piezoelectric element 6b.
6 is a waveform diagram showing operation waveforms of respective parts of the control circuit 16 when is smaller than a displacement command value Li. From the triangular wave generator 25, a triangular wave carrier signal (voltage signal) Sa as shown in FIG. 3A is continuously output. The displacement command value Li input from the input device or the like and the actual displacement value L of the piezoelectric element 6b detected by the displacement detector 21 are compared and calculated at the addition point 22, and then input to the gain adjuster 23. . From the gain adjuster 23, the deviation (Li−−) between the displacement command value Li of the piezoelectric element 6b and the actual displacement value L
Voltage signal proportional to L) (hereinafter referred to as deviation signal) So
Is output (therefore, the actual displacement value L and the displacement command value Li may be replaced with the actual voltage value Vp and the voltage command value Vpi of the piezoelectric element 6b). Zero cross comparator 2
The output of the triangular wave generator 25 is connected to the first input of 6, and the zero potential (0V) 27 is connected to the second input of the zero-cross comparator 26. In the zero cross comparator 26, the carrier signal Sa and the zero potential 27
3B is generated, a reference pulse signal Sb as shown in FIG. 3B is generated, and the reference pulse signal Sb (duty ratio 50 %) Is output. The output of the gain adjuster 23 is connected to the first input of the comparator 24, and the output of the triangular wave generator 25 is connected to the second input of the comparator 24. The comparator 24 compares the carrier signal Sa with the deviation signal So, and as shown in FIG.
The pulse signal Sc is output from the comparator 24 so that it becomes H level in the interval of carrier signal Sa> deviation signal So.

The pulse signal Sc output from the comparator 24 is a rising trigger type and falling trigger type monostable multivibrator 2 respectively.
The input terminals 8 and 29 are connected to the first inputs of the NOT operation element 30 and the AND operation element 32 which are connected to the first input of the AND operation element 31. Further, the outputs of the monostable multivibrators 28 and 29 are connected to the second input of the AND operation element 33 and the first input of the AND operation element 35, respectively. Reference pulse signal S output from the zero-cross comparator 26
b is a rising trigger type and a falling trigger type monostable multivibrator 36, 37.
Is input to the AND operation element 35 and is connected to the second input of the AND operation element 35.
3 to the first input. Further, the outputs of the monostable multivibrators 36 and 37 are connected to the second inputs of the AND operation elements 31 and 32, respectively. The output of the AND operation element 31 is input to the charge pulse control circuit 38 as the charge start trigger signal Pcs, and the output of the AND operation element 33 is input to the charge pulse control circuit 38. At the same time, the output of the AND operation element 31 is connected to the thyristor gate drive circuit 17 and outputs a trigger signal to the thyristor 3b. AND
The output of the arithmetic element 32 is the discharge start trigger signal Pds,
The output of the AND operation element 35 is input to the discharge pulse control circuit 39 as the discharge end trigger signal Pde. At the same time, the output of the AND operation element 32 is connected to the thyristor gate drive circuit 17 and outputs a trigger signal to the thyristor 3a.

Therefore, the monostable multivibrator 36 outputs a trigger pulse which is the basis of the charging start trigger signal Pcs at the rising of the reference pulse signal Sb, and the monostable multivibrator 28 ends the charging at the rising of the pulse signal Sc. A trigger pulse that is the source of the trigger signal Pce is output, and a trigger pulse that is the source of the discharge start trigger signal Pds is output from the monostable multivibrator 37 when the reference pulse signal Sb falls, and the monostable multivibrator 29 is used. Outputs a trigger pulse which is a source of the discharge end trigger signal Pde when the pulse signal Sc falls.
The logic circuit 58 including the AND operation elements 31, 32, 33, 35 and the NOT operation elements 30, 34 has a deviation (Li-
It works differently when L)> 0 and when deviation (Li-L) <0. When the deviation (Li-L)> 0, as shown in FIGS. 3B and 3C, the pulse width of the pulse signal Sc is narrower than the pulse width of the reference pulse signal Sb and within the section of the reference pulse signal Sb. Therefore, the trigger pulse that is the source of the discharge start trigger signal Pds and the trigger pulse that is the source of the discharge end trigger signal Pde output from the monostable multivibrator 37, 29 are the logic circuit 58.
Is cut and is not output from the logic circuit 58. However, when signals are output from the monostable multivibrators 36 and 28, the AND operation element 3 of the logic circuit 58 is
The charging start trigger signal Pcs and the charging end trigger signal Pce are output from 1 and 33 to the charging pulse control circuit 38, respectively (FIGS. 3D and 3E). Also, the AND operation element 3
When the charge start trigger signal Pcs is output from 1, the thyristor 3b is turned on. Conversely, deviation (Li-L)
When <0 (see FIGS. 5B and 5C), the pulse width of the pulse signal Sc is wider than the pulse width of the reference pulse signal Sb and includes the section of the reference pulse signal Sb. The trigger pulse that is the source of the charge start trigger signal Pcs and the trigger pulse that is the source of the charge end trigger signal Pce output from the multivibrators 36 and 28 are cut by the logic circuit 58 and are not output from the logic circuit 58. However, the monostable multivibrator 37,
When a signal is output from 29, the AND operation elements 32 and 35 of the logic circuit 58 to the discharge pulse control circuit 39 respectively, the discharge start trigger signal Pds and the discharge end trigger signal Pde.
Is output. When the discharge start trigger signal Pds is output from the AND operation element 32, the thyristor 3a is turned on.

The charging pulse control circuit 38 comprises an OR operation element 40 and a JK flip-flop 41. The charge start trigger signal Pcs from the AND operation element 31 and the charge end trigger signal Pce from the AND operation element 33 are input to the OR operation element 40, respectively.
Is the clock input CK of JK flip-flop 41
It is connected to the. Both the J and K inputs of the JK flip-flop 41 are pulled up to the voltage Vcc, the set input S is pulled down to the ground potential, and the reset input R
The output of the power-on reset circuit 44 is connected to the. The Q output of the JK flip-flop 41 (the output of the charging pulse control circuit 38) is connected to the MOSFET gate drive circuit 18 and outputs an on / off control signal to the MOSFET 8b. Then, in the charge pulse control circuit 38, the JK flip-flop 41 is operated by the charge start trigger signal Pcs or the charge end trigger signal Pce.
The Q output of is alternately inverted. Therefore, the deviation (Li
In the case of −L)> 0, when the charge start trigger signal Pcs is input to the charge pulse control circuit 38, the MOSFET 8b is controlled to be turned on in synchronization with the turn-on of the thyristor 3b, and the charge pulse control circuit 38 is also turned on. When the charge end trigger signal Pcs is input to the MOSFET 8, the MOSFET 8b is controlled to be off.

Similarly, the discharge pulse control circuit 39 uses the OR
It is composed of an arithmetic element 42 and a JK flip-flop 43.
The discharge start trigger signal Pds from the AND operation element 32 and the discharge end trigger signal Pde from the AND operation element 35 are input to the OR operation element 42, and the output of the OR operation element 42 is the clock input of the JK flip-flop 43. It is connected to CK. JK flip-flop 43
The J input and the K input are pulled up to the voltage Vcc, the set input S is pulled down to the ground potential, and the reset input R is connected to the output of the power-on reset circuit 44. Further, the Q output of the JK flip-flop 43 (output of the discharge pulse control circuit 39) is connected to the MOSFET gate drive circuit 18, and is turned on to the MOSFET 8a.
Outputs an off control signal. Then, in the discharge pulse control circuit 39, the JK flip-flop 4 is operated by the discharge start trigger signal Pds or the discharge end trigger signal Pde.
The Q output of 3 alternately inverts. Therefore, the deviation (L
When i−L) <0, when the discharge start trigger signal Pds is input to the discharge pulse control circuit 39, the thyristor 3a
The MOSFET 8a is controlled to be turned on in synchronism with turn-on, and the MOSFET 8a is controlled to be turned off when the discharge end trigger signal Pds is input to the charge pulse control circuit 39.

The power-on reset circuit 44 includes a resistor 45.
And the series connection circuit of the capacitor 46 with the voltage V on the resistor 45 side.
It is configured by pulling up to cc, pulling down to the ground potential on the side of the capacitor 46, and connecting the connection point of the resistor 45 and the capacitor 46 to the two inputs of the Schmitt trigger type NAND operation element 47 at the same time. 47
Output as the output of the reset circuit 46 when the power is turned on.
It is connected to each reset input R of the K flip-flops 41 and 43.

Thus, the charge pulse control from the power-on reset circuit 44 is performed until the voltage of the capacitor 46 rises to the threshold level of the Schmitt trigger type NAND operation element 47 when the power of the piezoelectric element driving device A is turned on. A reset signal is output to each reset input R of the circuit 38 and the discharge pulse control circuit 39, and the outputs of the charge pulse control circuit 38 and the discharge pulse control circuit 39 are reset to the L level.

The output of the reference capacitor voltage detection circuit 48 for detecting the voltage of the reference capacitor 6a and the output of the reference voltage setting device 49 are connected to the inputs of the comparator 50, respectively, and further, the monostable multivibrator is used. The output of 37 and the output of the comparator 50 are connected to the first and second inputs of the AND operation element 51. The output of the AND operation element 51 is input to the rising trigger type monostable multivibrator 52, and the monostable multivibrator 52 is output.
Is output to the comb-teeth pulse generator 53, and the output of the comb-teeth pulse generator 53 is output to the charging thyristor 7 as a trigger signal via the thyristor gate drive circuit 17.
This comb-tooth pulse generator 53 is a Schmitt trigger type N
One input of the AND operation element 54 is the comb-tooth pulse generator 5
3 is an external input terminal to which the output of the monostable multivibrator 52 is connected. Also, NAN
A resistor 55 is provided between the output of the D arithmetic element 54 and the other input.
And the other input of the NAND operation element 54 is grounded via the capacitor 56. further,
The output of the NAND operation element 54 is output to the charging thyristor 7 through the NOT operation element 57.

Then, the reference capacitor voltage detection circuit 4
When the output voltage Vc of No. 8 is smaller than the threshold voltage Vs set by the reference voltage setting unit 49, the comparator 50 outputs an H level signal. Therefore,
When the auxiliary charging trigger signal (= discharge start trigger signal) Pds is output from the monostable multivibrator 37, this auxiliary charging trigger signal Pds is input to the monostable multivibrator 52, and the trigger signal is output from the monostable multivibrator 52. To be done. When the trigger signal is output from the monostable multivibrator 52, the comb-tooth pulse generator 53 outputs the comb-tooth pulse, and the charging thyristor 7 of the charging circuit 13 is turned on.

Next, the operation of the piezoelectric element driving device A will be described with reference to FIGS. 3 (a) to 3 (n). FIG. 3A shows a triangular wave carrier signal Sa output from the triangular wave generator 25 and a displacement command value Li output from the gain adjuster 23.
And a deviation signal So proportional to the deviation (Li-L) of the actual displacement value L. 3B shows the reference pulse signal Sb output from the zero-cross comparator 26, and FIG. 3C shows the pulse signal Sc output from the comparator 24. 3D and 3E show a charge start trigger signal Pcs and a charge end trigger signal Pce for causing the output of the charge pulse control circuit 38 to toggle. Figure 3 (f) shows MOSF
The on / off operation of the ET 8b is shown, and FIG. 3 (g) shows the on / off state of the thyristor 3b. FIG. 3H shows an auxiliary charging trigger signal (discharge start trigger signal) Pds output from the monostable multivibrator 37, and FIG.
(I) shows the output of the comparator 50 that determines whether auxiliary charging is necessary. 3 (j) and (k) show the discharge current Ic of the reference capacitor 6a and the voltage Vc of the reference capacitor 62,
3 (l) and (m) show the charging current Ip of the piezoelectric element 6b and the voltage Vp of the piezoelectric element 6b. FIG. 3 (n) shows the commutation current I ₁ flowing through the inductor 2. Below, FIG.
According to (n) to (n), the operation when the electric charge is transferred to the piezoelectric element 6b to charge the piezoelectric element 6b will be described.

FIG. 3 shows a case where the actual displacement value L of the piezoelectric element 6b is smaller than the displacement command value (target value) Li and the piezoelectric element 6b needs to be supplemented with electric charges.
As shown in (a), the deviation signal So corresponding to the deviation (Li-L)> 0 is located on the positive value side, and the reference pulse signal Sb is output from the comparator 24 as shown in FIGS. A pulse signal Sc having a width narrower than that of the pulse signal Sc is output. The charge start trigger signal Pcs is output at the rising edge of the reference pulse signal Sb as shown in FIG. 3D, the charge end trigger signal Pce is output at the rising edge of the pulse signal Sc as shown in FIG. 3E, and As shown in FIG. 3 (f), M
Since the OSFET 8b is turned on by the charge start trigger signal Pcs and turned off by the charge end trigger signal Pce, as can be seen from FIGS. 3 (a) to 3 (f), the MOSFET is
The on-time of 8b is proportional to the deviation (Li-L). Also,
When the charge start trigger signal Pcs is output, the thyristor 3b is turned on as shown in FIG.

When the charging start trigger signal Pcs is output from the monostable multivibrator 36 as described above, the MOSFET 8b is turned on and the thyristor 3b is turned on, so that the FIGS. As shown in (a), the reference capacitor 6a is discharged and the reference capacitor 6a → commutation diode 4a → inductor 2 → thyristor 3b → diode 9b → MOS.
The commutation circuit 10 is formed along the path from the FET 8b to the reference capacitor 6a.
The commutation current I ₁ is commutated to. Next, when the charging end trigger signal Pce is output from the monostable multivibrator 28 and the MOSFET 8b is turned off, the induced electromotive force of the inductor 2 causes the induced electromotive force in FIG.
As shown in (b), inductor 2 → thyristor 3b → piezoelectric element 6b → reflux diode 5a → commutation diode 4a
→ A commutation current I ₁ flows through the commutation circuit 10 through the path of the inductor 2, the piezoelectric element 6b is charged as shown in FIG. 3 (m), and the piezoelectric element 6b starts displacement. Then, this commutation current I
_{When 1} becomes 0, the thyristor 3b turns off and the charging operation ends.

Here, the deviation signal So is limited to a value slightly lower than the peak value of the carrier signal Sa, and when the deviation (Li-L) is maximum, the deviation signal So is set.
Is located slightly below the peak value of the carrier signal Sa. Therefore, the commutation time t1 of the operation of FIG. 4A is smaller than 1/4 cycle of the cycle Ta of the carrier signal Sa,
When t1 ≦ (Ta / 4) is satisfied and the capacitance of the reference capacitor 6a is Cs and the inductance of the inductor 2 is L, the maximum value (t1) max of the commutation time t1 is given by the following equation. Can be set as: That is, (t1) max = (π / 2) (LCs) ^1/2 ≈Ta / 4 (1) This means that the piezoelectric element 6b is driven to a predetermined maximum voltage ((1) by one commutation operation from the reference capacitor 6a. The maximum extension amount) is reached. Note that FIG.
The commutation time t2 = (t3) max (see equation (2) described later) in the operation of (b) does not need to match the commutation time (t1) max. Then, if the setting is made according to the equation (1), the commutation current I ₁ when the commutation time t1 is the maximum value (t1) max≈Ta / 4
Reaches a peak value and the deviation (Li-L) is maximum (for example, when the initial displacement of the piezoelectric element 6b is 0, the displacement command value Li
Is the maximum value), all charges in the reference capacitor 6a can be commutated to the piezoelectric element 6b by one commutation operation, and the piezoelectric element 6b can be displaced to the maximum displacement amount equal to the displacement command value Li. Further, the time t1 at which the electric charge of the reference capacitor 6a is discharged is equal to the ON time of the MOSFET 8b and is proportional to the deviation (Li-L) [that is,
The pulse width of the pulse signal output from the charging pulse control circuit 38 is pulse-width modulated so as to be proportional to the deviation (Li-L)]. Therefore, the charge amount corresponding to the deviation (Li-L) is used as the reference capacitor 6a. To the piezoelectric element 6b, the piezoelectric element 6b can be displaced by an amount substantially equal to the deviation (Li-L), and a displacement amount substantially equal to the displacement command value Li can be obtained.

The section T ₁ in FIG. 3 is a case where the initial displacement of the piezoelectric element 6b is small, and in this case, as shown in FIG.
As shown in (n), a large commutation current I ₁ flows in the commutation circuit 10 (inductor 2), causing the piezoelectric element 6b to have a displacement command value Li.
Is displaced up to. If the actual displacement value L of the piezoelectric element 6b does not reach the displacement command value Li due to the first commutation operation, or if the actual displacement value L of the piezoelectric element 6b decreases due to discharge of the piezoelectric element 6b, the interval in FIG. T ₂
As shown in, the commutation operation is performed for a short commutation time t1 according to the small deviation (Li-L), and the actual displacement value L of the piezoelectric element 6b approaches the displacement command value Li. Thus
The actual displacement value L becomes equal to the displacement command value Li and the deviation (Li
When −L) = 0, the commutation operation is not performed and the displacement of the piezoelectric element 6b becomes stable as shown in the section T ₃ in FIG.

Further, as shown in FIG. 4A, when the reference capacitor 6a is discharged and the voltage Vc of the reference capacitor 6a becomes lower than the threshold voltage Vs set by the reference voltage setting device 49, the output of the comparator 50 becomes H. The level becomes [Fig. 3 (i)], and the auxiliary charge trigger signal (discharge start trigger signal) P from the monostable multivibrator 37.
When ds is output [Fig. 3 (h)], the charging thyristor 7
4 is turned on, the reference capacitor 6a is supplementarily charged from the DC power supply 1 as shown in FIG. 4C, and when the voltage Vc of the reference capacitor 6a becomes equal to the voltage V _DC of the DC power supply 1, the charging thyristor 7 is turned off and the reference Capacitor 6a
Returns to the original voltage. However, since the auxiliary charging is considered only when the voltage Vc of the reference capacitor 6a drops to the threshold voltage Vs or less due to the AND condition of the AND operation element 51, a voltage sufficient for the next commutation operation is applied to the reference capacitor 6a. If it remains, auxiliary charging is not performed.

Next, the actual displacement value L of the piezoelectric element 6b is larger than the displacement command value (target value) Li and the deviation (Li-L) <
The case where the discharge from the piezoelectric element 6b is required when the value is 0 will be described. FIG. 5 is a diagram showing operation waveforms of each part in the case of deviation (Li-L) <0. FIG. 5A shows a triangular wave carrier signal Sa output from the triangular wave generator 25.
And the deviation signal So of the displacement command value Li and the actual displacement value L output from the gain adjuster 23. 5B shows the reference pulse signal Sb output from the zero-cross comparator 26, and FIG. 5C shows the pulse signal Sc output from the comparator 24. 5D and 5E show a discharge start trigger signal Pds and a discharge end trigger signal Pde for inverting the output of the discharge pulse control circuit 39. Figure 5
5F shows the on / off operation of the MOSFET 8a, and FIG. 5G shows the on / off state of the thyristor 3a. 5 (h) (i) show the discharge current Ip of the piezoelectric element 6b and the voltage Vp of the piezoelectric element 6b, and FIG. 5 (j) (k) show the charging current Ic of the reference capacitor 6a and the voltage Vc of the reference capacitor 6a. Show. FIG. 5 (l) shows the commutation current I ₂ flowing through the inductor 2. In the following, the operation when the charges are transferred from the piezoelectric element 6b and discharged will be described with reference to FIGS.

FIG. 5 shows a case where the actual displacement value L of the piezoelectric element 6b is smaller than the displacement command value (target value) Li and the electric charge is discharged from the piezoelectric element 6b. Therefore, as shown in FIG. The deviation signal So corresponding to (Li-L) <0 is located on the negative value side, and as shown in FIGS. 5B and 5C, a pulse signal wider than the reference pulse signal Sb is output from the comparator 24. Sc is output. The discharge start trigger signal Pds is output when the reference pulse signal Sb falls as shown in FIG. 5D, and the discharge end trigger signal Pde is shown in FIG.
As shown in FIG. 5E, the pulse signal Sc is output at the falling edge, and as shown in FIG. 5F, the MOSFET 8a is turned on by the discharge start trigger signal Pds and turned off by the discharge end trigger signal Pde. (A) ~
As can be seen from (f), the on-time of the MOSFET 8a is proportional to the absolute value of the deviation (Li-L). When the discharge start trigger signal Pds is output, the thyristor 3a is turned on as shown in FIG.

When the discharge start trigger signal Pds is output from the monostable multivibrator 37 as described above, the MOSFET 8a is turned on and the thyristor 3a is turned on, so that FIGS. As shown in (a), the electric charge of the piezoelectric element 6b is discharged and the piezoelectric element 6b → the commutation diode 4b → the inductor 2 →
Thyristor 3a → diode 9a → MOSFET 8a →
The commutation current I ₂ is commutated to the commutation circuit 10 through the path of the piezoelectric element 6b, and the piezoelectric element 6b contracts. Then, the discharge end trigger signal P is output from the monostable multivibrator 29.
When de is output and the MOSFET 8a is turned off, the induced electromotive force of the inductor 2 causes the inductor 2 → thyristor 3a → as shown in FIGS. 5 (j) (l) and 6 (b).
The commutation current I ₂ flows through the commutation circuit 10 along the path of the reference capacitor 6a → the freewheeling diode 5b → the commutation diode 4b → the inductor 2, and as shown in FIG.
The electric charge is regenerated. When the commutation current I ₂ becomes 0, the thyristor 3a is turned off and the regenerative operation is completed.

The deviation signal So is limited to a value slightly higher than the bottom value of the carrier signal Sa. When the deviation (Li-L) is the minimum, the deviation signal So is set.
Is located slightly above the bottom value of the carrier signal Sa. Therefore, the commutation time t3 of the operation of FIG. 6A is smaller than ¼ cycle of the cycle Ta of the carrier signal Sa, and t3 ≦
(Ta / 4), and when the capacitance of the piezoelectric element 6b is Cp and the inductance of the inductor 2 is L, the maximum value (t3) max of the commutation time t3 is given by the following equation. Can be set. That is, (t3) max = (π / 2) (LCp) ^1/2 ≉Ta / 4 (2) This is because the piezoelectric element 6b has the original minimum voltage (minimum) due to one regenerative operation from the piezoelectric element 6b. It is possible to reach the extension amount). The commutation time t4 = (t1) max in the operation of FIG. 6 (b) is determined by the commutation time (t3) ma.
It does not have to match x. Then, (t3) max is
It is set as in equation (2), and the commutation time t3 is the maximum value (t3) m.
When ax≈Ta / 4, the commutation current I ₂ reaches the peak value, and even when the absolute value of the deviation (Li−L) is the maximum, the total charge of the piezoelectric element 6b is converted by one commutation operation to the reference capacitor 6a.
And the piezoelectric element 6b can be displaced to a minimum displacement amount equal to the displacement command value Li. Further, the time t3 when the charge of the piezoelectric element 6b is discharged is equal to the on-time of the MOSFET 8a and is proportional to the absolute value of the deviation (Li-L). Therefore, the charge amount corresponding to the deviation (Li-L) is set to the piezoelectric element. 6b to the reference capacitor 6a so that the piezoelectric element 6
b can be returned to the original state.

A section T ₄ in FIG. 5 is a case where the initial displacement of the piezoelectric element 6b is very large. In this case, as shown in FIG. 5 (n), a large commutation circuit 10 (inductor 2) is transferred. A current I ₂ flows, and the piezoelectric element 6b is displaced to the displacement command value Li. If the displacement actual value L of the piezoelectric element 6b does not reach the displacement command value Li is the commutation operation of the first, as shown in section T ₅ in FIG. 5, a small absolute value of the deviation (Li-L) The commutation operation is performed for a short commutation time t3 in accordance with the above, and the actual displacement value L of the piezoelectric element 6b is brought close to the displacement command value Li. In this way, the actual displacement value L becomes equal to the displacement command value Li and the deviation (Li-
L) = becomes zero, the regenerative operation is no longer performed as shown in section T ₆ in FIG. 5, the displacement of the piezoelectric element 6b is stabilized.

Although the current detector 14 and the current detection / comparison circuit 15 are not used in the above-described operation of the piezoelectric element driving apparatus A, they can be used for failure diagnosis of the control circuit 16 (also. , And also when using the control circuit 16 of FIG. 12 or 16.) That is, the charge end trigger signal Pce and the discharge end trigger signal Pde
When the MOSFETs 8b and 8a are not turned off by the above or when a short-circuit failure occurs in the piezoelectric element 6b, the commutation currents I ₁ and I ₂ reach a peak value due to resonance and then start to fall beyond the peak value. By detecting the state of changes in the commutation currents I ₁ and I ₂ together with the voltage of the piezoelectric element 6b, the control circuit 16 can make a self-judgment at the time of failure. If the control circuit 16 can control the voltage V _DC of the DC power source 1 as shown in FIG. 1, the MOSFET 8b is turned on in the initial state (state where the voltage Vp of the piezoelectric element 6b is 0). Time t1
There will be a maximum, since the commutation current I ₁ from the reference capacitor 6a to the inductor 2 reaches a peak value, to detect the peak value of the commutating current I ₁ by the current detector 14 or the like by using this time, the The voltage V _DC of the DC power supply 1 can be adjusted from the control circuit 16 so that the peak value becomes a predetermined magnitude.

[Second Embodiment] FIG. 7 shows a reference capacitor 6a.
Reference capacitor from the commutation current I ₁ to commutation to the inductor 2, the commutation current I ₁ to commutation from the inductor 2 to the piezoelectric element 6b, the commutation current I ₂ is commutated from the piezoelectric element 6b to the inductor 2, the inductor 2 Commutation current I ₂ commutating to 6a
And these commutation times (t1) max, t2, (t3) max,
It shows t4. However, the current commutated from the reference capacitor 6a to the piezoelectric element 6b is positive, and the current commutated from the piezoelectric element 6b to the reference capacitor 6a is negative. Generally, if the operating voltage (Vp) max of the piezoelectric element 6b is at the same level as the voltage _VDC of the DC power supply 1, the electrostatic capacitance Cp of the piezoelectric element 6b and the electrostatic capacitance Cs of the reference capacitor 6a are approximately the same. However, if the operating voltage (Vp) m of the piezoelectric element 6b is
Assuming that ax is n times the voltage V _DC of the DC power supply 1, in order to drive the piezoelectric element 6b having the electrostatic capacitance Cp, the reference capacitor 6a having the electrostatic capacitance Cs = n ² × Cp which is n ² times the Cp. is necessary. This is derived from the following relationship representing the electrostatic energy stored in the piezoelectric element 6b and the reference capacitor 6a. (CsV _DC ² ) / 2 = [Cp (Vp) max ² ] / 2 Therefore, if Cs> Cp, the relationship of commutation time is shown in FIG. 7 from equations (1) and (2). As shown, (t1) max = t4> t2 = (t3) max (3). Under this condition, the piezoelectric element driving device A is shown in FIGS.
However, since the capacitance Cs of the reference capacitor 6a increases, the commutation time (t1) max
If [= t4] becomes longer than a quarter cycle Ta / 4 of the carrier signal Sa and the triangular wave of the carrier signal Sa is a symmetrical waveform, as shown in FIG. 8, the transfer from the reference capacitor 6a to the piezoelectric element 6b is performed. flow period Tcp is (t1) commutation current I ₁ shorter than max is the maximum value (I ₁₎ does not reach the max. In order to shorten the commutation time (t1) max, the inductance L of the inductor 2 may be reduced. However, when the inductance L is reduced, the commutation time (t3) max is also shortened, so that the piezoelectric element 6b to the reference capacitor 6a are reduced. On the contrary, in the commutation period Tpc to (3), a limit circuit for limiting the commutation time t3 to (t3) max is required.

However, such a defect is limited to the case where the charge of the reference capacitor 6a needs to be commutated to the piezoelectric element 6b by one commutation operation to fully charge the piezoelectric element 6b. When the final displacement command value Li of the piezoelectric element 6b is reached by repeating the commutation operation, such a problem does not pose a problem. Further, even when the piezoelectric element 6b is charged by a plurality of commutation operations, the commutation periods Tcp, Tpc << t1, t2 of the carrier signal Sa.
If the setting is made, even if the number of commutation operations is increased, the time to reach the displacement command value Li can be shortened because the frequency is high,
The operating speed can be increased.

FIG. 9 is a circuit diagram showing a piezoelectric element driving apparatus B according to another embodiment of the present invention. From the above consideration, the commutation time from the piezoelectric element 6b to the reference capacitor 6a ( t3) max and the commutation time (t1) max from the reference capacitor 6a to the piezoelectric element 6b are made to coincide with each other. In this piezoelectric element driving device B, 2
Two commutation circuits 10 using two inductors 2a and 2b
a and 10b, the anode of the thyristor 3a is connected to one end of the inductor 2a without being connected to the cathode of the commutation diode 4a, and the cathode of the commutation diode 4b is not connected to the anode of the thyristor 3b. 2a connected to the other end of one of the commutation circuits 1
0a is configured. The cathode of the commutation diode 4a is connected to one end of the inductor 2b, and the anode of the thyristor 3b is connected to the other end of the inductor 2b to form the other commutation circuit 10b. Further, the current detector 14 is installed so as to detect both the commutation currents I ₁ and I ₂ in both directions passing through the inductors 2a and 2b.

In such a circuit configuration, when the electric charge is transferred from the reference capacitor 6a to the piezoelectric element 6b, the commutation current I ₁ is _first turned on by turning on the MOSFET 8b and turning on the thyristor 3b. It flows in the path of diode 4a → inductor 2b → thyristor 3b → diode 9b → MOSFET 8b → reference capacitor 6a. At this time, assuming that the inductance of the inductor 2b is Lb, the commutation time (t1) max is (t1) max = (π / 2) (LbCs) ^1/2 (4). Then, when the MOSFET 8b is turned off, the commutation current I1 flows in the path of the inductor 2b → thyristor 3b → piezoelectric element 6b → reflux diode 5a → commutation diode 4a → inductor 2b. Commutation time t2 at this time
Becomes t2 = (π / 2) (LbCp) ^1/2 (5). Further, when the charge is transferred from the piezoelectric element 6b to the reference capacitor 6a, first, the commutation current I ₂ is turned on by turning on the MOSFET 8a and turning on the thyristor 3a.
Is piezoelectric element 6b → commutation diode 4b → inductor 2a
→ Thyristor 3a → Diode 9a → MOSFET 8a
→ Flows in the path of the piezoelectric element 6b. At this time, assuming that the inductance of the inductor 2a is La, the commutation time (t3) max is (t3) max = (π / 2) (La
Cp) ^1/2 (6) Then, when the MOSFET 8a is turned off, the commutation current I ₂ is changed to the inductor 2a → thyristor 3a → reference capacitor 6a → reflux diode 5b → commutation diode 4
It flows in the path of b → inductor 2a. The commutation time t4 at this time is t4 = (π / 2) (LaCs) ^1/2 (7).

Therefore, in order to satisfy (t1) max = (t3) max, it is necessary to establish the relationship of LbCs = LaCp (8). For example, the inductance Lb of the inductor 2b is Lb = La. Cs). Here, by rewriting equations (4) to (7) using equation (8), (t1) max = (π / 2) (LaCs) ^1/2 (9) t2 = (π / 2) ( LaCp) ^1/2 = (t1) max · (Cp / Cs) ^1/2 (10) (t3) max = (π / 2) (LaCs) ^1/2 (11) t4 = (π / 2) Cs (La / Cp) ^1/2 = (t1) max (Cs / Cp) ^1/2 (12) Further, here, the driving voltage (V
p) max is three times the voltage V _DC of the DC power supply 1, that is, Cs =
Assuming 9 Cp, equations (9) to (12) are as follows.
(T1) max = (π / 2) (LaCs) ^1/2
(13) t2 = (t1) max / 3 (14) (t3) max = (t1) max (15) t4 = 3 (t1) max (16) Alternatively, if the voltage ratio is expressed as Cs / Cp = n ² , the formulas (9) to (12) are expressed as (t1) max = (π / 2) (LaCs) ^1/2 (17) t2 = (t1) max / n ... a (18) (t3) max = (t1) max ... (19) t4 = n · (t1) max ... (20) Lb = n 2 · La ... (21).

As described above, in the piezoelectric element driving apparatus B as shown in FIG. 9, if the circuit constants are determined so as to satisfy the expression (8) or the expression (21), the commutation time in the commutation period Tcp ( t
1) max and the commutation time (t3) max in the commutation period Tpc can be equalized to (t1) max = (t3) max≈Ta / 4 (22), and the voltage V _{DC of the} DC power supply 1 Is the piezoelectric element 6
Even when the driving voltage (Vp) max of b is 1 / n, the piezoelectric element 6b can be fully charged by one commutation operation.

[Third Embodiment] On the other hand, the peak value Ip of the commutation current is generally Ip = V (C / L, where V is the voltage on the discharge side, C is the capacitance of the commutation circuit, and L is the inductance. ) ^1/2 , therefore, in the piezoelectric element driving device B as shown in FIG. 9, if the circuit constants are determined so as to satisfy Cs = n ² Cp and the equation (21), the commutation currents I ₁ , I ₂ Peak value (I ₁ ) m
ax, (I ₂ ) max is (I ₁ ) max = V _DC (Cs / La) ^1/2 (23) (I ₂ ) max = n · V _DC (Cp / Lb) ^1/2 = n · V _DC [(Cs / n ² ) / (n ² La)] ^1/2 = (V _DC / n) (Cs / La) ^1/2 = (I ₁ ) max / n (24) Therefore, from the piezoelectric element 6b to the reference capacitor 6a
The peak value (I ₂ ) max of the commutation current I ₂ to the piezoelectric element 6b decreases in inverse proportion to the voltage ratio n with respect to the peak value (I ₁ ) max of the commutation current I ₁ from the reference capacitor 6a to the piezoelectric element 6b. To do. Therefore, in the piezoelectric element driving device B as shown in FIG. 9, the peak values (I ₁ ) max and (I ₂ ) max of the commutation currents flowing in the commutation circuits 10b and 10a during charging and discharging may be significantly different. Yes, thyristors 3a, 3b and MOSFETs 8a, 8b
The current allowable amount of the switching element such as is a problem. In other words, in order to increase the allowable peak current, it is necessary to use a large-capacity switching element, or to increase the capacity by serially connecting switching elements in parallel, which reduces the utilization efficiency of the switching element. There is a possibility that a problem such as an increase in size of the element driving device B may occur.

In order to solve such a problem, FIG.
1 is used as the control circuit 16 used in the piezoelectric element driving apparatus B as shown in FIG.
The piezoelectric element driving device B can be controlled as shown in 1 (a) to (o). The displacement command value Li input from the input device or the like and the actual displacement value L of the piezoelectric element 6b detected by the displacement detector 21 are compared and calculated at the addition point 22, and then input to the gain adjuster 23. . Gain adjuster 23
A deviation signal So proportional to the deviation (Li-L) between the displacement command value Li of the piezoelectric element 6b and the actual displacement value L is output from (FIG. 11 (a)). The deviation signal So output from the gain adjuster 23 is input to the first input of the first comparator 61 and the first input of the second comparator 62, respectively. First and second reference voltage setting device (variable resistor) 6
3, 64 are connected in series, and the first reference voltage setter 63
Is maintained at the positive voltage + V, and the second reference voltage setting device 64
Is maintained at a negative voltage -V, and both reference voltage setters 63, 6 are
The connection point of 4 is grounded to 0 potential. From the first reference voltage setting device 63 and the second reference voltage setting device 64, the threshold voltage V _S1 (0 <which determines the dead band width V _S2 ≦ So ≦ V _S1
V _S1 ≤ + V) and V _S2 (-V ≤V _S2 <0) are output (FIG. 11A), and the output V _S1 of the first reference voltage setting unit 63 is the output of the first comparator 61. 2 is connected to the second input, and the output V _S2 of the second reference voltage setter 64 is connected to the second input of the second comparator 62. Thus, from the first comparator 61, H-level signal is output when the deviation signal So.> threshold voltage V _S1, L-level signal when the deviation signal So. ≦ threshold voltage V _S1 Is output (FIG. 11B). The second comparator 62 outputs an H level signal when the deviation signal So <V _S2 and an L level signal when the deviation signal So ≧ V _S2 (FIG. 11C). The output of the first comparator 61, the first input of the AND operation element 65,
It is connected to the input of a falling trigger type monostable multivibrator 66, and the output of the monostable multivibrator 66 is connected to a first input of an AND operation element 68 through a NOT operation element 67. Similarly, the output of the second comparator 62 is
It is connected to the second input of the AND operation element 68 and the input of the falling trigger type monostable multivibrator 69, and the output of the monostable multivibrator 69 passes through the NOT operation element 70 to form an AN.
It is connected to the second input of the D arithmetic element 65. Then, the output of the first comparator 61 is directly output from the AND operation element 65 except when the output of the second comparator 62 falls and the trigger signal is output from the monostable multivibrator 69. Similarly, the output of the second comparator 62 is directly output from the AND operation element 68 except when the output of the first comparator 61 falls and the trigger signal is output from the monostable multivibrator 66.

The output of the synchronous pulse oscillator 75 is connected to the respective inputs of the charge pulse generator 76 and the discharge pulse generator 77, and the charge reference pulse signal Pc and the discharge reference pulse signal Pd are output from both pulse generators 76 and 77. Are output in synchronization with each other (FIGS. 11D and 11E). However, the charge reference pulse signal Pc and the discharge reference pulse signal Pd output from the charge pulse generator 76 and the discharge pulse generator 77 are converted into pulse signals having an appropriate duty ratio by the pulse generators 76 and 77, respectively. There is. The output of the charging pulse generator 76 is input to each of the rising trigger type and falling trigger type monostable multivibrators 78 and 79. The output of the discharge pulse generator 77 is input to the rising trigger type monostable multivibrator 80.

The output of the AND gate 71 is connected to the thyristor gate drive circuit 17 to turn on the thyristor 3b, and the output of the AND gate 72 is connected to the MOSFET gate drive circuit 18 to turn on the MOSFET 8b. It is designed to be turned off, and the first input of the AND gate 71 and the AND gate 7
The output of the AND operation element 65 is connected to the second input of No. 2. Also, Monostable Multivibrator 7
The output of 8 is connected to the second input of the AND gate 71,
The output of charging pulse generator 76 is directly connected to the first input of AND gate 72. When the deviation signal So is larger than the threshold voltage Vs1, the AND operation element 65 outputs an H level signal to the AND gates 71 and 72,
Since the charging reference pulse signal Pc is output from the charging pulse generator 76 to the AND gate 72, the AND gate 72
The same signal as the charging reference pulse signal Pc is sent from the MOSFET 8b to the gate of the MOSFET 8b, and the MOSFET 8b is on / off controlled in synchronization with the charging reference pulse signal Pc (FIG. 1).
1 (h)). Also, from the AND gate 71 to the thyristor 3
A trigger signal (Fig. 11 (f)) is output to b in synchronization with the rising of the charging reference pulse signal Pc, and the thyristor 3
b turns on (FIG. 11 (j)).

The output of the AND gate 73 is connected to the thyristor gate drive circuit 17 to turn on the thyristor 3a, and the output of the AND gate 74 is connected to the MOSFET gate drive circuit 18 to turn on the MOSFET 8a. The output of the AND operation element 68 is connected to the respective first inputs of the AND gate 73 and the AND gate 74.
The output of the monostable multivibrator 80 is connected to the second input of the AND gate 73, and the output of the discharge pulse generator 77 is directly connected to the second input of the AND gate 74. When the deviation signal So is lower than the threshold voltage V _S2 , the AND operation element 68 outputs an H level signal to the AND gates 73 and 74, and the discharge pulse generator 77 outputs the discharge reference pulse signal Pd to the AND gate 74. Is output, the AND gate 74 outputs MOSF
The same signal as the discharge reference pulse signal Pd is sent to the gate of the ET 8a, and the MOSFET 8a receives the discharge reference pulse signal Pd.
On / off control is performed in synchronization with d (FIG. 11 (i)).
Further, a trigger signal (FIG. 11 (g)) is output from the AND gate 73 to the thyristor 3a in synchronization with the rising of the discharge reference pulse signal Pd, and the thyristor 3a is turned on (FIG. 11 (k)).

The output Vc of the reference capacitor voltage detection circuit 48 for detecting the voltage of the reference capacitor 6a is connected to the first input of the comparator 50, and the output of the reference voltage setter 49 is connected to the second input of the comparator 50. ing.
The output of the comparator 50 and the output of the monostable multivibrator 79 are input to the AND operation element 51, and the output of the AND operation element 51 is the comb-tooth pulse generator 53.
Is input to the thyristor gate driving circuit 17, and the output of the comb-tooth pulse generator 53 is turned on to control the charging thyristor 7.

Thus, the reference capacitor voltage detection circuit 4
When the output voltage Vc of 8 becomes lower than the threshold voltage Vs of the reference voltage setting unit 49, the comb-tooth pulse generator 53 outputs the comb-tooth pulse to the charging thyristor 7 in synchronization with the falling operation of the charging reference pulse signal Pc. , The charging thyristor 7 is turned on.

Next, referring to FIGS. 11 (a) to 11 (o), FIG.
The operation of the control circuit 16 of 0 will be described. Here, FIG.
(L) is the discharge current Ic of the reference capacitor 6a, (m) is the voltage Vc of the reference capacitor 6a, (n) is the charging current Ip of the piezoelectric element 6b, and (o) is the piezoelectric element 6 in FIG.
The voltage Vp of b is shown. First, when the actual displacement value L of the piezoelectric element 6b is smaller than the displacement command value Li and the deviation voltage So is larger than the threshold voltage V _S1 , the first comparator 61 as shown in FIG. 11B. Output and the output of the AND operation element 65 become H level, and the control circuit 16 enters the charging mode. In the section in the charging mode, the charging pulse generator 76 and the monostable multivibrator 7
8 and AND gates 71, 72 charge control circuit 8
1 becomes the operating state, and the MOSFET 8b is on / off controlled in synchronization with the charging reference pulse signal Pc output from the charging pulse generator 76 as shown in FIG.
1 (h)) and the trigger signal as shown in FIG. 11 (f) output from the monostable multivibrator 78 in synchronization with the rising of the charging reference pulse signal Pc turns on the thyristor 3b (FIG. 11 (j)). . As a result, as shown in FIGS. 11 (l) (n), the discharge from the reference capacitor 6a to the inductor 2b and the inductor 2
Charging from b to the piezoelectric element 6b is alternately repeated. At the same time, the reference capacitor 6a is discharged and the reference capacitor 6a
When the voltage Vc of the reference voltage setting unit 49 becomes lower than the threshold voltage Vs of the reference voltage setting unit 49, the output of the comparator 50 becomes the H level, so that the output of the AND operation element 51 becomes the H level at the fall of the charging reference pulse signal Pc, and the comb A comb-tooth signal is output from the tooth pulse generator 53 to the charging thyristor 7 via the thyristor gate drive circuit 17, and the DC power source 1 charges the reference capacitor 6a from the charging circuit 13 (FIG. 11 (m)). As a result, the piezoelectric element 6b
The actual displacement value L of is gradually increased (FIG. 11 (o)),
When the displacement command value Li is gradually approached and the deviation signal So ≦ V _S1 is satisfied, the charging operation ends. On the other hand, when the deviation signal So> V _S1 , the output of the second comparator 62 is at the L level, so the discharge pulse generator 77, the monostable multivibrator 80, and the AND gates 73 and 74.
The discharge control circuit 82 consisting of does not operate.

As described above, the commutation operation from the reference capacitor 6a to the piezoelectric element 6b is performed several times in synchronization with the charge reference pulse signal Pc, and the charge reference pulse generated by the charge pulse generator 76 is generated. Since the period and pulse width of the signal Pc are constant, the peak value (I ₁ ) max of the commutation current I ₁ transferred from the reference capacitor 6a to the piezoelectric element 6b.
And the peak value (I ₁ ) max of the commutation current I ₁ is always constant as long as the voltage Vc of the reference capacitor 6a is constant. Therefore, the charge transferred by one commutation operation becomes constant, and the necessary amount of charge is replenished to the piezoelectric element 6b by controlling the number of commutation operations.

The reference capacitor voltage detection circuit 48,
The auxiliary charging control circuit 83 including the reference voltage setter 49, the comparator 50, and the AND operation element 51 performs auxiliary charging from the DC power supply 1 to the reference capacitor 6a when the voltage Vc of the reference capacitor 6a drops below a certain value Vs. However, the auxiliary charging control circuit 83 is omitted, and the output of the monostable multivibrator 79 is directly input to the comb-tooth pulse generator 53 so that the reference capacitor 6a is auxiliary charged at each commutation operation. May be.

On the contrary, the actual displacement value L of the piezoelectric element 6b is larger than the displacement command value Li, and the deviation voltage So is the threshold voltage V.
When it is smaller than _S2 , as shown in FIG. 11C, the output of the second comparator 62 and the output of the AND operation element 68 become H level, and the control circuit 16 enters the discharge mode. In the section in the discharge mode, the discharge control circuit 8
2 becomes an operating state, and the MOSFET 8a is on / off controlled in synchronization with the discharge reference pulse signal Pd output from the discharge pulse generator 77 as shown in FIG.
1 (i)) and the trigger signal as shown in FIG. 11 (g) output from the monostable multivibrator 80 in synchronization with the rising of the discharge reference pulse signal Pd turns on the thyristor 3a (FIG. 11 (k)). . As a result, as shown in FIGS. 11 (l) and (n), the piezoelectric element 6b
To the inductance 2a and the charging from the inductor 2a to the reference capacitor 6a are alternately repeated. As a result, the actual displacement value L of the piezoelectric element 6b becomes gradually smaller (FIG. 11 (o)), gradually approaches the displacement command value Li, and the discharge operation ends when the deviation signal So ≧ Vs2. on the other hand,
When the deviation signal So <V _S2 , the first comparator 6
Since the output of 1 is at the L level, the charge control circuit 8
1 does not work. In the discharging operation, the voltage Vc of the reference capacitor 6a rises, but the voltage Vc of the reference capacitor 6a tends to discharge due to the influence of the impedance of the reference capacitor voltage detection circuit 48. The auxiliary charging is performed by the charging circuit 13.

In this way, the commutation operation from the piezoelectric element 6b to the reference capacitor 6a is also performed several times in synchronization with the discharge reference pulse signal Pd, and the discharge reference pulse generated by the discharge pulse generator 77 is generated. Since the period and pulse width of the signal Pd are constant, the peak value (I ₂ ) max of the commutation current I ₂ transferred from the piezoelectric element 6b to the reference capacitor 6a.
Can be made small and can be kept constant. Therefore, even in this case, the charge transferred by one commutation operation is constant, but the amount of charge transferred to the reference capacitor 6a is controlled by the number of commutation operations.

Since the charge pulse generator 76 and the discharge pulse generator 77 can separately set the pulse width of the charge reference pulse signal Pc and the discharge reference pulse signal Pd, these pulse widths can be set appropriately. By setting the peak value (I ₁ ) ma of the commutation current I ₁ during the charging operation.
It is possible to reduce x and the peak value (I ₂ ) max of the commutation current I ₂ during the discharge operation and adjust them so that they are substantially at the same level. Therefore, the magnetic circuit by the inductors 2a and 2b can be downsized, and the thyristors 3a and 3b and the MOSF can be reduced.
The utilization rate of the ETs 8a and 8b can be improved, and the piezoelectric element driving device B can be prevented from increasing in size.

Further, in the control circuit 16, the first
The dead band V _S2 ≤So ≤V _S1 is set by the dead band setting circuit 84 configured by the first and second reference voltage setters 63 and 64 and the first and second comparators 61 and 62, and therefore, it is shown in FIG. As described above, even if the level of the deviation signal So passes through the dead zone, the charging operation is not immediately switched to the discharging operation, or the discharging operation is not immediately switched to the charging operation. When the control accuracy of the piezoelectric element 6b is increased, it is preferable that the width of the dead zone is made small, but by setting the dead zone, it is possible to prevent the problem of the timing of switching the charge / discharge operation. When the actual displacement value L of the piezoelectric element 6b substantially matches the displacement command value Li and the deviation (Li-L) ≈0, it is possible to prevent repetitive chattering of charge / discharge due to noise or the like.

In the control circuit of FIG. 10, the monostable multivibrators 66, 69 and the NOT operation elements 67, 70 are used.
The reason why the charge / discharge timing control circuit 85 including the AND operation elements 65 and 68 is provided is as follows.
When the first comparator 61 falls from the H level (charge mode) to the L level, the output of the second comparator 62 may rise to the H level at the same time depending on the timing. In such a case, since there is a possibility that the charging mode is immediately switched to the discharging mode, the charging / discharging timing control circuit 85 shifts the timing for a certain period of time when switching from the charging mode to the discharging mode to prevent a malfunction due to the immediate switching. There is. On the contrary, the charging / discharging timing control circuit 85 can prevent the trouble caused by the immediate switching from the discharging mode to the charging mode. However, when the dead zone setting circuit 84 is provided as in the control circuit of FIG. 10, the charge / discharge timing control circuit 85 is not necessary, or the dead zone setting circuit 84 is not necessary.
Alternatively, one of the charge and discharge timing control circuit 85 may be used alternatively. Further, in the control circuit 16 of FIG. 10, since the rising timings of the charge pulse generator 76 and the discharge pulse generator 77 are synchronized by the synchronous pulse oscillator 75, the need for the charge / discharge timing control circuit 85 becomes lower. There is.

[Fourth Embodiment] FIG. 12 is a circuit diagram showing the configuration of still another control circuit 16 used in the piezoelectric element driving apparatus B. 13 (a) to 13 (l) are diagrams showing the operation waveforms of the respective parts of the control circuit 16 in the above when the charge is transferred from the reference capacitor 6a to the piezoelectric element 6b, and FIG. Signal So, FIG. 13 (b) is the output of the AND operation element 65, FIG. 13 (c) is the synchronization pulse signal P1,
13 (d) shows the limit pulse signal P2, FIG. 13 (e).
Is the current arrival pulse signal P3, and FIG.
On / off operation of T8b, FIG. 13 (g) shows thyristor 3
b on / off state, FIG. 13 (h) shows the reference capacitor 6
1a is the discharge current Ic, FIG. 13 (i) is the charge current Ip of the piezoelectric element 6b, and FIG. 13 (j) is the auxiliary charge command signal P4.
3 (k) is the voltage Vc of the reference capacitor 6a, FIG.
(L) is the voltage Vp of the piezoelectric element 6b. In the control circuit 16 of FIG. 10, since the commutation period from the reference capacitor 6a to the inductor 2b during charging is constant, the inductor 2
The peak value (I ₁ ) max of the commutation current I ₁ flowing in b depends on the voltage V _DC of the DC power supply 1. To avoid this,
The voltage V _DC of the DC power supply 1 is controlled to be a constant voltage, or the peak values (I ₁ ) max and (I ₂ ) max of the commutation currents I ₁ and I ₂ during charging and discharging are set to It can be controlled so as to be constant, or either method can be taken. 12
The control circuit 16 is the commutation current I ₁ by the latter method, I ₂ of the peak value _{(I 1) max, (I} 2) max voltage V of the DC power supply 1 of
_It is designed to be maintained at a constant value without depending on _DC .

Below, the control circuit 16 of FIG.
A description will be given with reference to (a) to (l). Displacement command value L
i and the actual displacement value L detected by the displacement detector 21 are compared at an addition point 22 and then calculated.
Entered in. The deviation signal So output from the gain adjuster 23 is input to the dead zone setting circuit 84, and the two outputs of the dead zone setting circuit 84 are connected to the two inputs of the charge / discharge timing control circuit 85, respectively. Two ANs configuring the output part of the charge / discharge timing control circuit 85
Of the D operation elements 65 and 68, the output of the AND operation element 65 for charge mode is the AND for controlling the thyristor 3b.
It is connected to the first input of the gate 71 and the second input of the AND gate 72 which controls the MOSFET 8b. Further, the output of the AND operation element 65 is also connected to the first input of the AND gate 91 which controls the charging thyristor 7. The output of the AND operation element 68 for the discharge mode is the AND gate 73 for controlling the thyristor 3a and the MOSFET.
8a is connected to each first input of an AND gate 74.

Therefore, the dead zone setting circuit 84 sets a dead zone in the range of V _S2 ≤So ≤V _S1 as shown in FIG. 13A, and the actual displacement value L of the piezoelectric element 6b is the displacement command value Li. When the displacement signal So> V _S1 is smaller than the above, the output of the AND operation element 65 becomes the H level and the output of the AND operation element 68 becomes the L level as shown in FIG. AND gate 73 for discharging,
Output of 74 is fixed at L level and AND for charging
A control signal can be output from the gates 71 and 72. When the displacement signal So <V _S2 , the AND operation element 6
The output of No. 5 is at the L level, the output of the AND operation element 68 is at the H level, and the discharge mode is set, and the outputs of the AND gates 71 and 72 for charging are fixed at the L level and controlled by the AND gates 73 and 74 for discharging. A signal can be output.

The output of the synchronous pulse oscillator 75a is connected to the respective first inputs of the charge pulse generator 76a and the discharge pulse generator 77a. From the synchronous pulse oscillator 75a, a synchronous pulse signal P1 for every constant period as shown in FIG. 13C is output. The charge pulse generator 76a and the discharge pulse generator 77a both rise to H level when the trigger signal (that is, the synchronization pulse signal P1) is input to the first input, and the trigger signal is input to the second input. Then it falls to L level. Charging pulse generator 76
The output of a is the input of the rising trigger type monostable multivibrator 78 and the first of the AND gate 72.
Of the AND gate 71, and the output of the monostable multivibrator 78 is connected to the second input of the AND gate 71. Therefore, when the output of the charging / discharging timing control circuit 85 is in the charging mode, when the synchronous pulse signal P1 is output from the synchronous pulse oscillator 75a and the output of the charge pulse generator 76a rises, the MOSFET is
8b is turned on, the thyristor 3b is turned on, and the commutation operation from the reference capacitor 6a to the inductor 2b is started. The output of the discharge pulse generator 77a is connected to the input of the rising trigger type monostable multivibrator 80 and the second input of the AND gate 74, and the output of the monostable multivibrator 80 is connected to the second input of the AND gate 73. It is connected to the. Therefore, when the output of the charge / discharge timing control circuit 85 is in the discharge mode, the synchronous pulse signal P1 is output from the synchronous pulse oscillator 75a to generate the discharge pulse generator 77.
When the output of a rises, the MOSFET 8a is turned on, the thyristor 3b is turned on, and the commutation operation from the piezoelectric element 6b to the inductor 2a is started.

The limit pulse generator 75b outputs the limit pulse signal P2 as shown in FIG. 13 (d) in the same cycle with a delay of a fixed time td from the sync pulse signal P1 of the sync pulse oscillator 75a. The output of the current detection / comparison circuit 15 connected to the current detector 14 is connected to the first input of the comparator 92 via the absolute value circuit 15a, and the output of the peak current setting device 93 is the second input of the comparator 92. It is connected to the input, and the output of the comparator 92 is input to the monostable multivibrator 94. The absolute value circuit 15a detects the commutation current I ₁ , which is detected by the current detector 14 regardless of the flowing directions of the commutation currents I ₁ and I ₂ .
This is for outputting the absolute value of the value of I ₂ to the comparator 92. Therefore, the commutation circuits 10a and 10b (or,
The commutation currents I ₁ and I ₂ flowing through the inductors 2a and 2b) are set in the peak current setting device 93 by the set peak current I.
When it becomes equal to m, the comparator 92 rises, and the monostable multivibrator 94 is operated.
A current arrival pulse signal P3 such as Both the output of the limit pulse generator 75b and the output of the monostable multivibrator 94 are input to the OR operation element 95, and the output of the OR operation element 95 is the second input of each of the charge pulse generator 76a and the discharge pulse generator 77a. It is connected to the. Therefore, in the case where electric charges are commutated from the reference capacitor 6a to the piezoelectric element 6b, the commutation current I ₁
When the current reaches the set peak current Im, the monostable multivibrator 94 outputs the current arrival pulse signal P3, the output of the charging pulse generator 76a falls to the L level, the MOSFET 8b turns off, and the commutation current I ₁
Is transferred from the inductor 2b to the piezoelectric element 6b. Similarly, when the charge is commutated from the piezoelectric element 6b to the reference capacitor 6a, the commutation current I ₂ is the set peak current Im.
The moment you reach the limit, Monostable Multivibrator 94
The current arrival pulse signal P3 is output from the discharge pulse generator 77a, the output of the discharge pulse generator 77a falls to the L level, and the MOSFET 8
a is turned off, and the commutation current I ₂ is transferred from the inductor 2a to the reference capacitor 6a. These are normal operating conditions, and in this case, the peak value of the commutation current I ₁ or I ₂ is a constant value (set peak current Im) set by the peak current setting device 93. Alternatively, the limit pulse generator 7 can be operated without outputting the current arrival pulse signal P3.
Even when the limit pulse signal P2 is output from 5b,
The output of the charge pulse generator 76a or the discharge pulse generator 77a falls to the L level, and the MOSFET 8b or 8a is turned off. This is, for example, when the voltages of the reference capacitor 6a, the piezoelectric element 6b, and the DC power supply 1 are lowered, and in that case, the peak value of the commutation current I ₁ or I ₂ is smaller than the set peak current Im.

Reference numeral 96 is an auxiliary charging control circuit. Limit pulse generator 75b and AND operation element 65 in charge mode
The output of the AND operation element 97 connected to the input side and the power-on initial charging circuit 98 are connected to each input of the OR operation element 99. The power-on initial charging circuit 98 includes a resistor 100 and a capacitor 101 connected in series to form a resistor 10
It is pulled up to the voltage Vcc on the 0 side and pulled down to the ground potential on the capacitor 101 side, and the resistor 100 and the capacitor 1 are connected.
The connection point of 01 is the Schmitt trigger type NOT arithmetic element 10
It is configured by connecting to 2 and has a Schmitt trigger type NO.
The output of the T arithmetic element 102 is connected to the OR arithmetic element 99. Furthermore, the output of the OR operation element 99 is a rising trigger type monostable multivibrator 10
3 connected, monostable multivibrator 10
3 is connected to the comb-tooth pulse generator 53, and the output of the comb-tooth pulse generator 53 is connected to the AND gate 91 for controlling the turn-on of the charging thyristor 7.

When the piezoelectric element driving apparatus B is turned on, the voltage applied to the resistor 100 rises to Vcc, and the output of the initial charging circuit 98 at the time of turning on the power becomes H level. Auxiliary charge command signal P4 to the device 53
Is output, the charging thyristor 7 is turned on, and the DC capacitor 1 initially charges the reference capacitor 6a. Also,
In the charge mode in which the output of the AND operation element 65 is at the H level, the auxiliary charge command signal P4 is output to the comb-tooth pulse generator 53 in synchronization with the limit pulse signal P2 output from the limit pulse generator 75b, The charging thyristor 7 is turned on and the DC capacitor 1 charges the reference capacitor 6a.

Next, referring to FIGS. 13 (a) to 13 (l), FIG.
The operation of the second control circuit 16 in the charging mode will be described. The charging mode in which the deviation signal So corresponding to the deviation (Li-L) is larger than the dead zone threshold V _S1 (FIG. 13A).
In (b), the MOSFET 8b is synchronized with the sync pulse signal P1 output from the sync pulse oscillator 75a.
Is turned on and the thyristor 3b is turned on (FIGS. 13 (c) (f) (g)), and the commutation current I ₁ flows from the reference capacitor 6a to the inductor 2b (FIG. 13).
(H)). When this commutation current I ₁ reaches the set peak current Im and the current arrival pulse signal P3 is output from the monostable multivibrator 94, the MOSFET 8b is turned off (FIGS. 13 (e) and (f)), and the inductor 2b to the piezoelectric element. The charges are transferred to 6a (FIG. 13 (i)), and the piezoelectric element 6b is displaced. When the commutation current I ₁ does not reach the set peak current Im, the MOSFET 8b is turned off in accordance with the limit pulse signal P2 (FIG. 13 (d)).
(F)) Electric charges are transferred from the inductor 2b to the piezoelectric element 6b. Therefore, the piezoelectric element 6b has the set peak current I
The reference capacitor 6a is charged by a plurality of commutation operations having a peak value of m or less. Further, although the voltage Vc of the reference capacitor 6a decreases due to each commutation operation, the charging thyristor 7 is turned on in synchronization with the limit pulse signal P2 to charge the reference capacitor 6a every time (FIG. 13 (j) (k). )), The commutation current I ₁ is prevented from gradually decreasing. In this way, when the actual displacement value L of the piezoelectric element 6b reaches Li-V _S1 and the deviation signal So reaches the dead zone, the charging mode is released (FIG. 13).
(A) (l)).

As described above, in this embodiment, the piezoelectric element 6b is charged and discharged by a plurality of commutation operations, and the peak values of the commutation currents I ₁ and I ₂ during charging and discharging are constant values (= It is controlled so that it becomes the set peak current Im) set by the peak current setting device. According to this method, the commutation current I
_Since the peak values of ₁ and I ₂ can be controlled to be substantially constant regardless of the magnitude of the voltage V _DC of the DC power supply 1, and the repetition frequency is also constant, the MOSFETs 8a and 8b, the thyristor 3a,
The utilization efficiency of the switching elements such as 3b is improved, and the thermal design of the piezoelectric element driving device B is facilitated.

[Fifth Embodiment] The control circuit 1 shown in FIG.
6, the value of the set peak current Im may be different between the charging operation and the discharging operation (the same applies to the control circuit 16 after FIG. 16). FIG. 14 is a partial circuit diagram showing a part of the control circuit 16 therefor. That is, in the control circuit of FIG. 14, the peak current setter 93a set to the set peak current Imc is connected to the second input of the comparator 92 via the analog switch 93c, and the set peak current Imd (≠ Imc)
The peak current setting device 93b set to is connected to the second input of the comparator 92 via the analog switch 93d. Switching signals are input to the switching terminal of the analog switch 93c directly and to the switching terminal of the analog switch 93d via the inverter 93e. When the switching signal is H, the analog switch 93c is turned on and the analog switch 9c is turned on.
3d is turned off, and conversely, when the switching signal is L, the analog switch 93c is turned off and the analog switch 93d is turned on. Therefore, for example, in the charging operation, the switching signal is set to H and Imc is used as the value of the set peak current, and in the discharging operation, the switching signal is set to L and Imd is used as the value of the set peak current. it can.

[Sixth Embodiment] FIG. 15 is a partial circuit diagram showing a control circuit 16 having another structure in which different set peak current values are used during the charging operation and the discharging operation. In this control circuit 16, the output of the current detection / comparison circuit 15 connected to the current detector 14 is the absolute value circuit 15a.
Are connected to the respective first inputs of the two comparators 92a and 92b via. Further, the peak current setter 93a set to the set peak current Imc is the comparator 92a.
Is connected to the second input of the comparator 92b, and the peak current setter 93b set to the set peak current Imd is connected to the second input of the comparator 92b. Further, the comparator 92
a is connected to the rising trigger type monostable multivibrator 94a, and the output of the monostable multivibrator 94a is connected to the second input of the charging pulse generator 76a of the charging control circuit 81 via the NOR operation element 94a. . Similarly, the comparator 92
b is connected to the rising trigger type monostable multivibrator 94b, and the output of the monostable multivibrator 94b is connected to the second input of the discharge pulse generator 77a of the discharge control circuit 82 via the NOR operation element 94b. .

Therefore, during the charging operation, the signal resulting from the comparison between the commutation current I ₁ and the set peak current Imc is input to the charging control circuit 81 (active state) and the discharging control circuit 82 (inactive state) to discharge the current. During operation, a signal obtained as a result of comparing the commutation current I ₂ and the set peak current Imd is the charge control circuit 8
1 (inactive state) and the discharge control circuit 82 (active state).

[Seventh Embodiment] FIG. 16 is a circuit diagram showing still another control device 16 according to the present invention. This control circuit 16 is different from the control circuit 16 of FIG.
Only the configuration of 11 is different. The auxiliary charge control circuit 111 in this embodiment is the RS flip-flop circuit 1
The NOR operation element 113 is connected to the reset input R of 12, and the NAND operation element 114 is connected to the set input S. Further, the output of the RS flip-flop circuit 112 is connected to the second input of the AND operation element 97,
The output of the AND operation element 65 that outputs a signal in the charging mode is connected to the first input of the ND operation element 97.
The output of the AND operation element 97 and the output of the power-on initial charging circuit 98 are connected to each input of the OR operation element 99, and the output of the OR operation element 99 is connected to the rising trigger type monostable multivibrator 103. The output of the monostable multivibrator 103 is connected to the comb-tooth pulse generator 53, and the output of the comb-tooth pulse generator 53 is connected to the AND gate 91 for controlling the turn-on of the charging thyristor 7. Also, NOR
A synchronous pulse oscillator 75 is connected to the first input of the arithmetic element 113.
The output of a is input, and the output of the power-on initial charging circuit 98 is input to the second input of the NOR operation element 113 via the inverter 102b. The output of the charging pulse generator 76a and the output of the limit pulse generator 75b are connected to two inputs of the NAND operation element 114.

The auxiliary charge control circuit 111 is in the charge mode with the H level signal output from the AND operation element 65, and operates only when the reference capacitor 6a charges the piezoelectric element 6b. , Does not operate in the discharge mode in which charges are transferred from the piezoelectric element 6b to the reference capacitor 6a. Further, when the power is turned on and when the synchronous pulse signal P1 is output from the synchronous pulse oscillator 75a, the L level signal is input to the reset input R of the RS flip-flop circuit 112, and the auxiliary charge control circuit 1
11 is reset. On the other hand, the charging pulse generator 76a
When the limit pulse generator 75b outputs the limit pulse signal P2 while the charge pulse signal Pc is being output from the auxiliary charge control circuit, the L level signal is input to the set input S of the RS flip-flop circuit 112. 111 is set, the charging thyristor 7 is turned on, and the reference capacitor 6a is auxiliary charged from the DC power supply 1. However, since the output of the charging pulse generator 76a falls to the L level due to the limit pulse signal P2, the operation timing in this case is delicate. To stabilize this, the charging pulse generator 76a uses the second input. The L level may be set in synchronization with the fall of the trigger signal applied to the. On the contrary, when the output of the charging pulse generator 76a is at the L level, even if the limit pulse signal P2 is output from the limit pulse generator 75b, the L is input to the set input S of the RS flip-flop circuit 112.
Since the level signal is not output, the auxiliary charging operation is not performed.

FIG. 17 is a diagram showing operation waveforms in the charging mode of this embodiment. In this embodiment, the commutation current I
_The electric charge is commutated from the reference capacitor 6a to the piezoelectric element 6b by a plurality of commutation operations in which the peak value of ₁ is equal to the set peak current Im, but unlike the embodiment of FIG. Only when the peak value of the flowing current I ₁ does not reach the set peak current Im, the auxiliary charging command signal P4 is output to the comb-tooth pulse generator 53 to perform the auxiliary charging operation. That is, as can be seen from FIGS. 17D and 17F, the peak value of the commutation current I ₁ is the set peak current Im.
When the limit pulse signal P2 is output, the output of the charging pulse generator 76a is at the L level, but when the peak value of the commutation current I ₁ does not reach the set peak current Im. Since the output of the charging pulse generator 76a is at the H level when the limit pulse signal P2 is output,
The charge pulse generator 76 is generated by the NAND arithmetic element 114.
The output of a and the limit pulse generator 75b are ANDed to control the auxiliary charging control circuit 111. When the auxiliary charging is performed, the commutation operation can be performed again with a peak value equal to the set peak current Im, and the deviation signal So
When reaches the dead zone, the charging mode ends.

As described above, in this embodiment, the piezoelectric element 6b is charged and discharged by a plurality of commutation operations, and the peak value of the commutation current I ₁ during charging and discharging is a constant value (= peak current setting). The control is performed so that the set peak current Im) set by the device is obtained. However, before the auxiliary charging is done,
It is possible that the commutation current I ₁ does not reach the set peak current Im within the set period. According to this method
The peak value of the commutation current I ₁ can be controlled to be almost constant regardless of the magnitude of the voltage V _DC of the DC power supply 1, and the repetition frequency is also constant, so that the utilization efficiency of the switching element is improved and the piezoelectric element driving device is improved. The heat design of B etc. becomes easy.

[Eighth Embodiment] FIG. 18 is a circuit diagram showing a control circuit 16 according to still another embodiment of the present invention.
(A)-(m) is a figure which shows the operation waveform of each part.
In this piezoelectric element driving device B, the gain adjuster 23
The deviation signal So output from the VCO (voltage controlled oscillator circuit) 121 is input to the VCO 121, and the frequency of the pulse signal output from the VCO 121 increases as the magnitude (absolute value) of the deviation signal So increases. It decreases as the signal So decreases. The output of the VCO 121 is connected to the synchronous pulse oscillator 75a, and the output of the synchronous pulse oscillator 75a is also output to the limit pulse generator 75b together with the charge pulse generator 76a and the discharge pulse generator 77a. From the synchronous pulse oscillator 75a, the VCO
The trigger signal is output in synchronization with the rising edge of the pulse signal 121, and the limit pulse generator 75b outputs the trigger signal with a delay of a predetermined time td from the trigger signal of the synchronous pulse oscillator 75a. For other circuit configurations,
This is similar to the embodiment of FIG.

However, in this embodiment, the deviation signal So
Is large and the actual displacement value L is far from the displacement command value Li, the VCO 121 outputs a high frequency signal and the frequency of the synchronization pulse signal P1 also increases, so that the charge mode commutation is performed. The operation cycle becomes shorter,
The piezoelectric element 6b is rapidly charged from the reference capacitor 6a. Further, when the deviation (Li-L) approaches 0, the period of commutation operation becomes longer and the piezoelectric element 6b is slowly charged, and the actual displacement L of the piezoelectric element 6b becomes the displacement command value Li. Approach slowly. The same applies to the case of the discharge mode. By providing the gain adjuster 23 and the like with the period adjusting function of the commutation operation in this way, it is possible to provide the response characteristic that the actual displacement value L of the piezoelectric element 6b quickly approaches the displacement command value Li. In the above description, the case of commutation from the reference capacitor 6a to the piezoelectric element 6b has been described, but also in the case of commutation from the piezoelectric element 6b to the reference capacitor 6a, the commutation operation is performed according to the deviation (Li-L). Of course, the cycle of can be changed.

Incidentally, FIG. 10, FIG. 12, FIG. 14, FIG.
Needless to say, the control circuit as shown in FIGS. 16 and 18 can be used not only in the piezoelectric element driving device B as shown in FIG. 9 but also in the piezoelectric element driving device A as shown in FIG. In particular, as in the control circuits of FIGS. 12, 14, 15, 16, and 18, when the set peak current values can be individually set and controlled during charging and discharging, respectively, FIG. As shown in the main circuit of the piezoelectric element driving device, an inductor is
It is not necessary to divide the commutation circuit into two parts, and the commutation circuit may be configured with only one inductance as in the piezoelectric element driving device of FIG.

Further, as an application of each piezoelectric element driving device, it is used as a drive system for controlling a hydraulic system for adjusting the spring force in a suspension system of an automobile, or an injection for fuel injection. Can also be used as a drive system for driving.

[0085]

According to the piezoelectric element driving device of the present invention, when the actual displacement value of the first charging / discharging element is smaller than the displacement command value, the second charging / discharging element changes to the first charging / discharging element. The charge can be diverted to extend the first charging / discharging element. Conversely, if the actual displacement value of the first charging / discharging element is larger than the displacement command value, the Since charges can be transferred to the second charging / discharging element to contract the first charging / discharging element, even if the fluctuation of the power supply voltage or the fluctuation of the displacement amount due to disturbance occurs, It is possible to precisely match the displacement command value with feedback control of the actual displacement value.

Moreover, an unnecessary portion of the charge of the first charge / discharge element is commutated to the second charge / discharge element, and the first charge / discharge element is again charged to the first charge / discharge element.
Since it can be commutated to the charging / discharging element, the electric charge can be reused and the driving power of the piezoelectric element driving device can be remarkably reduced. Moreover, the power supply does not require a voltage regulation function.

Further, since the charging circuit for charging the second charging / discharging element is provided, the commutation current to the first charging / discharging element does not become insufficient, and the second charging / discharging element does not First
It is possible to secure necessary charges for charging the charging / discharging element.

Further, if the piezoelectric element driving device is provided with a means for judging the necessity of charge auxiliary charging from the charging circuit to the second charge / discharge element, the charge amount of the second charge / discharge element is insufficient. In this case, the second charging / discharging element can be quickly and auxiliary charged from the charging circuit, so that the voltage due to the commutation loss caused by the resistance component of the inductor or the forward voltage drop of the various switching elements of the commutation circuit is caused. You can replenish the descent.

Further, in the piezoelectric element driving device in which the charges are communicated from the second charge / discharge element to the first charge / discharge element by a plurality of commutation operations, the commutation operation is performed once. It is not necessary to fully charge the first charging / discharging element, and the first charging / discharging element is divided into several times until the command value is reached.
Since the charge can be transferred to the second charge / discharge element, a power supply for auxiliary charging the second charge / discharge element or the element must have a voltage lower than the drive voltage of the first charge / discharge element. You can Therefore, a small charging circuit or a driving power source can be used, and the piezoelectric element driving device can be reduced in size and weight.

For example, there are electromagnetic actuators and piezoelectric elements used in vehicle-mounted fuel injection devices, but in automobiles, it is required to reduce their power consumption as the number of electrical components increases. Therefore, by using the present invention, the limited capacity of the battery can be saved and controlled, and this requirement can be met. Also,
Similarly, when driving the piezoelectric element for active suspension, power consumption can be reduced.

Furthermore, if the peak value of the commutation current flowing in the commutation circuit is made constant, the utilization factor of the switching elements used in the commutation circuit can be improved. Moreover, when the commutation operation does not reach a certain peak value, the commutation operation is stopped for a certain period of time after the start of commutation to perform auxiliary charging, so the peak value of the commutation current decreases. In this case, the second charging / discharging element can be immediately auxiliary charged to return the peak value of the commutation current to a constant value again.

Furthermore, when commutating to the first charge / discharge element by a plurality of commutation operations, the commutation operation is performed according to the difference between the actual displacement value of the first charge / discharge element and the displacement command value. If the cycle of the commutation operation is inversely proportional to the difference between the actual displacement value and the displacement command value, for example, even if the difference between the actual displacement value and the displacement command value is large, The discharge element can be promptly displaced to the command value.

[Brief description of drawings]

FIG. 1 is a circuit diagram showing a piezoelectric element driving device according to a first embodiment of the present invention.

FIG. 2 is a circuit diagram showing a configuration of a control circuit in the above piezoelectric element driving device.

FIG. 3 is a diagram showing operation waveforms of respective parts of the piezoelectric element driving device of the above when the actual displacement value of the piezoelectric element is smaller than the displacement command value.

4A and 4B are diagrams showing a commutation operation during charging of a piezoelectric element, and FIG. 4C is a diagram showing an auxiliary charging operation.

FIG. 5 is a diagram showing operation waveforms of respective parts of the piezoelectric element driving device of the above when the actual displacement value of the piezoelectric element is larger than the displacement command value.

6A and 6B are diagrams showing a commutation operation during discharge of a piezoelectric element.

FIG. 7 is a diagram showing a change in commutation current flowing in the above commutation circuit.

FIG. 8 is a diagram showing a relationship between a carrier signal and a commutation current.

FIG. 9 is a circuit diagram showing a piezoelectric element driving device according to a second embodiment of the present invention.

FIG. 10 is a circuit diagram showing a control circuit used in a piezoelectric element driving device according to a third embodiment of the present invention.

FIG. 11 is a diagram showing operation waveforms of respective parts for explaining the operation of the control circuit of the above.

FIG. 12 is a circuit diagram showing a control circuit used in a piezoelectric element driving device according to a fourth example of the present invention.

FIG. 13 is a diagram showing operation waveforms of respective parts for explaining the operation of the control circuit of the above.

FIG. 14 is a partial circuit diagram showing a part of a control circuit according to a fifth embodiment of the present invention.

FIG. 15 is a partial circuit diagram showing a part of a control circuit according to a sixth embodiment of the present invention.

FIG. 16 is a circuit diagram showing a control circuit used in a piezoelectric element driving device according to a seventh embodiment of the present invention.

FIG. 17 is a diagram showing operation waveforms of respective parts for explaining the operation of the above control circuit.

FIG. 18 is a circuit diagram showing a control circuit used in the piezoelectric element driving device according to the eighth embodiment of the present invention.

FIG. 19 is a diagram showing operation waveforms of various parts for explaining the operation of the control circuit of the above.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 DC power source 2, 2a, 2b Inductor 6a Reference capacitor 6b Piezoelectric element 10, 10a, 10b Commutation circuit 13 Charging circuit 16 Control circuit

Claims (6)

[Claims] 1. A first and second charging / discharging element in which at least a first charging / discharging element is a piezoelectric element, and a transfer for transferring charges from the first charging / discharging element to the second charging / discharging element. Flow circuit, a commutation circuit for commutating charges from the second charge / discharge element to the first charge / discharge element, a charging circuit for charging the second charge / discharge element from the drive source, and a displacement command Means for comparing and calculating the voltage corresponding to the value or the displacement command value and the actual value of the first charge / discharge element, and the charge transfer direction between the first and second charge / discharge elements according to the polarity of the calculation result. A piezoelectric element driving device comprising: a selecting unit; and a unit that controls a transfer charge amount by any one of the commutation circuits according to the magnitude of the calculation result. 2. The piezoelectric element drive device according to claim 1, further comprising means for determining whether or not charging from the charging circuit to the second charge / discharge element is necessary. 3. A means for intermittently generating a first signal, a means for delaying the first signal by a predetermined time to generate a second signal, and a commutation current flowing through the commutation circuit. A means for generating a third signal by detecting that the current value has reached a constant value, and a commutation operation from the second charging / discharging element to the first charging / discharging element is started by the first signal, Means for stopping the commutation operation from the second charge / discharge element by the second or third signal, and when at least the third signal does not occur between the first signal and the second signal, 3. The piezoelectric element drive device according to claim 1, further comprising a means for causing the second charge / discharge element to be charged. 4. The commutation circuit diverts charges from the second charging / discharging element to the first charging / discharging element by a plurality of commutation operations, wherein the commutation circuit evaluates the electric charges. 4. The piezoelectric element driving device according to claim 1, wherein the cycle of the commutation operation is gradually changed according to the result of comparison between the command value and the actual value of one charging / discharging element. . 5. The piezoelectric element driving device according to claim 1, wherein the second charging / discharging element comprises a capacitor. 6. The piezoelectric element driving device according to claim 1, 2, 3 or 4, wherein the second charge / discharge element is a piezoelectric element.