EP3308409A1

EP3308409A1 - Device for controlling a piezoelectric actuator

Info

Publication number: EP3308409A1
Application number: EP16730318.9A
Authority: EP
Inventors: Issa OUATTARA; Jean-Luc Gach; Philippe AMRAM
Original assignee: Aix Marseille Universite; Centre National de la Recherche Scientifique CNRS
Current assignee: Aix Marseille Universite; Centre National de la Recherche Scientifique CNRS
Priority date: 2015-06-12
Filing date: 2016-06-10
Publication date: 2018-04-18
Also published as: CN107924983A; JP2018520514A; WO2016198639A1; CA2989250A1; FR3037454A1; US20180183357A1; US10389277B2; CN107924983B; FR3037454B1; KR20180032556A

Abstract

The present invention relates to a piezoelectric actuator control device able tothat can combine two types of piezoelectric actuator controls, a voltage control and a charge control. Switching between the charge control and the voltage control takes place is accomplished automatically, without adding switching elements, depending on the nature of the voltage applied to the input of the assembly.

Description

PIEZOELECTRIC ACTUATOR CONTROL DEVICE

Field of the Invention The invention lies in the field of piezoelectric actuator controls, and in particular relates to a control circuit which makes it possible to reduce the hysteresis of such actuators.

STATE OF THE ART There are mainly two types of piezoelectric actuators: the direct actuators in which the displacement obtained in the order of one micrometer is equal to the deformation of the piezoelectric material, and the amplified actuators in which a mechanical circuit amplifies the movement. and thus increase the displacement which is of the order of a millimeter. In these devices, an electric field is used to obtain a deformation of the piezoelectric material and to control the displacement. The control of an actuator may be a control voltage or load.

U.S. Patent 4,263,527 to Comstock discloses a control device in charge of a piezoelectric actuator.

U.S.7,015,621 B2 patent of Mizuuchi discloses a device for controlling the voltage of a piezoelectric actuator.

Although being the most common and simple solution to implement, the voltage control has a major disadvantage in that the displacement of the actuator, which is a function of the applied voltage, is not linear. This in particular is due to variations in the physical parameters of the piezoelectric material, creep and hysteresis.

Creep (or "creep" in English) is a slow drift of deformation of the material as a function of time, which results in a slow elongation of the actuator under a constant tension and which leads to the position being set to a few percent per decade of time of the initial displacement instruction.

Hysteresis (or hysteresis) is the property of a system that tends to remain in a certain state when an external cause that has produced the change of state has ceased. The piezoelectric actuators have a hysteresis between the applied voltage and the generated displacement. This non-linearity can vary between 10 and 20% depending on the nature of the material, and the movements of a piezoelectric actuator are not the same depending on whether it is a rising voltage or a decreasing voltage applied to it.

There are solutions to improve the behavior of piezoelectric actuators, in particular to reduce the hysteresis.

The patent application FR2850219 of J. Agnus et al. discloses a combined voltage and load control device for a piezoelectric actuator that provides linear open-loop static behavior, thereby reducing hysteresis. However, the circuit requires the addition of switch and interrupt means to switch from one control mode to another.

There is then the need for a solution that overcomes the disadvantages of known approaches. The present invention meets this need.

SUMMARY OF THE INVENTION An object of the present invention is to provide a system that can combine two types of control of a piezoelectric actuator, a voltage control and a load control, in which the switching between the load control and the control voltage is automatically, depending on the nature of the voltage applied to the input of the assembly. Advantageously, the device of the present invention is simple to implement and inexpensive, using no other external circuit for controlling the piezoelectric actuators. Advantageously, the gain of the device is the same in low and high frequency thanks to the adaptation of the capacitive or resistive bridge. The gain of the amplifier is fixed by the resistances at low frequency or even continuously, while in high frequency, the gain is fixed by the capacitance of the capacitors formed by the piezoelectric actuator and the capacitor referred to as the reference capacitor C _re f . Thus, starting from the inherent capacitance C _pzt of the piezoelectric actuators, given by the manufacturers, and the desired gain, it is very easy to dimension the discrete components of the resistive bridges (R _pzt , Rcref) and capacitive _bridges (C _pzt , C _re f) since the gain A _v is given by the relationship C _ref / Cpzt _pz = R t / Rcref-

Another advantage of the invention is that the hybrid amplifier can operate with any type of piezoelectric actuator for which there is no need to know the equivalent hysteresis model. The hybrid amplifier thus enables any open loop piezoelectric actuator to be controlled because the voltage - displacement characteristic is made substantially linear.

Advantageously, the output voltage of the hybrid amplifier is taken across the piezoelectric actuator and not across the set (C _pzt , R _pzt , Cref, Rcref), in which case the hysteresis is not attenuated .

In a preferred embodiment, the piezoelectric actuator control device of the invention comprises:

an operational amplifier having an inverting input and a non-inverting input, and an output for delivering a voltage Release ;

a capacitive circuit comprising a piezoelectric actuator and a capacitor in series, said capacitive circuit being connected by the free end of the piezoelectric actuator to the output of the operational amplifier and by the free end of the capacitor to a voltage terminal low;

a resistive circuit comprising a first resistor and a second resistor in series, said resistive circuit being connected by the free end of the first resistor to the output of the operational amplifier and by the second resistor to a low voltage terminal; the device being characterized in that the inverting input of the operational amplifier is connected to the common ends of the piezoelectric actuator and the capacitor, and is connected to the common ends of the first and second resistor, the capacitive circuit and the circuit resistive forming an impedance bridge.

Advantageously, the ratio of the impedances of the capacitive circuit is equal to the ratio of the impedances of the resistive circuit. In an operating mode, a control signal at a given frequency applied to the non-inverting input of the operational amplifier provides control of the piezoelectric actuator. In another mode of operation, a continuous control signal applied to the non-inverting input of the operational amplifier makes it possible to obtain a voltage control of the piezoelectric actuator. In one embodiment, the operational amplifier is powered between -60V and + 160V. In an alternative embodiment, the device of the invention comprises unidirectional transils diodes connected to the terminals of the piezoelectric actuator to limit the voltage at said terminals. The voltage across the piezoelectric actuator can be limited to + 150V and -20V, respectively.

In another variant, the device of the invention further comprises a resistor connected to the output of the operational amplifier for limiting the current delivered in the piezoelectric actuator.

The invention also covers an interferometer controlled by a control device as described. Advantageously, the interferometer may be a Fabry-Perot interferometer. Piezoelectric actuators are used in various fields for various purposes such as active vibration control for assistance with the machining of industrial parts, control in microscopes of small displacements for scanning a surface to be probed, control of automobile injectors, the production of fine droplets in printers, the adjustment of the length of a laser cavity in opto-acoustics or the micro-positioning of mirrors in astronomy.

DESCRIPTION OF THE FIGURES Various aspects and advantages of the invention will appear in support of the description of a preferred embodiment of the invention, but without limitation, with reference to the figures below:

Figure 1 illustrates a voltage control device of the prior art; Figure 2 illustrates a control device in charge of the prior art;

FIG. 3 illustrates a hybrid control device according to one embodiment of the invention; FIG. 4 illustrates an exemplary implementation of the device of the invention;

FIG. 5 illustrates a variant of implementation of the device of the invention;

FIG. 6 illustrates another variant of implementation of the device of the invention.

Detailed description of the invention

FIG. 1 illustrates a voltage control device known from the prior art. The principle of the voltage control is to apply a voltage V _s across a piezoelectric actuator (10) through an amplifier (12). The circuit comprises a piezoelectric actuator (10) connected at the output of an operational amplifier (12). The amplifier (12) is mounted in a non-inverting arrangement, and receives at a first input a control voltage V _c . A first resistor Ri (14) is connected between the output of the amplifier (12) and a second input of the amplifier. A second resistor R ₂ (16) is connected between the second input of the amplifier and a low voltage V ₀ . The output voltage V _s of the amplifier (12) is proportional to its control voltage V _c according to the relation: v¾ = i + -J- _e (1)

The gain A _v in voltage of the amplifier (12) is constant and defined by the resistors R1 and R2 (14, 1 6) according to the equation:

Av = ^ = l + ^ - (2) The frequency response of the circuit is given by the following equation:

/ = - - (3)

2 C _pzt V _s where l _s is the output current of the amplifier (1 2) running through the actuator (10) and C _pzt the capacitance value of the actuator. Thus, the frequency f depends on the current I _s and the output voltage V _s . In general, the output current is limited to a value (denoted l | _im ) which sets the value of the maximum frequency.

This type of control has the disadvantage of a fairly high hysteresis, and is not suitable for the control of systems that require micropositionings, such as very precise displacements of mirrors in the field of optics, for example.

FIG. 2 illustrates a load control device known from the prior art. The principle of load control is to apply a constant amount of charge to the electrodes of a piezoelectric actuator in order to obtain a linear behavior between the control and the displacement caused.

The circuit of FIG. 2 comprises a piezoelectric actuator (20) connected between the output UT 'and the inverting input (-) of an operational amplifier (22), the assembly having the configuration of an assembly no inverter. An input voltage V _c is applied to the second non-inverting input (+) of the amplifier. A so-called reference capacitor (24) is connected in series with the piezoelectric actuator (20) to the inverting input. It has a voltage Vc _re f at its terminals. The capacity capacitor C _ref receives a quantity of charge Q _re proportional to the voltage across its terminals.

In a load control circuit, the deformation of the piezoelectric actuator is linear as a function of the electrical load. The skilled person can refer to the work of J. Agnus, "Study, Realization, Characterization and Control of a Piezoelectric Micropince", University of Franche-Comté, 2003 which describes in more detail this relationship and shows that the application a constant amount of charges on the electrodes of a piezoelectric actuator makes it possible to obtain a linear behavior between the control and the displacement caused.

Although this type of circuit makes it possible to significantly reduce the hysteresis, typically by a factor of 10, J. Agnus has demonstrated in the aforementioned article that this arrangement does not allow the long-term maintenance of the initially applied charges. and thus maintaining a constant linear deformation of the piezoelectric actuator. It does not remove creep.

In static or continuous mode, there is saturation of the operational amplifier. Indeed, the frequency of the control signals is zero, and the impedance of the capacitors is infinite, these then behave like open circuits. Under these conditions, the capacitor C _re t (24) and the actuator (20) are in open circuit and the assembly behaves as a voltage comparator, resulting in the saturation of the operational amplifier.

In dynamic mode, the frequency of the output signal across the actuator (20) depends solely on the value of the output current and the input voltage V _c of the amplifier. Referring to FIG. 2, the piezoelectric actuator (20) can be replaced by an equivalent circuit which is a capacitor 'C _pzt ' having an impedance Z _P (p) = 1 / p. C _pzt

Considering also, 'Z _R (p)' the impedance of the reference capacitor (24) C _ref , the operational amplifier (22) being in linear operation, the impedances Z _R (p) and Z _P (p) being in series, they then form a voltage divider. The expression of the input voltage V _c (p) is given by the relation:

The resulting transfer function F (p) is

V _s {p) - Z _R (p) + Z _p (p)

F (p) =

V _e (p) Z _R (p)

By replacing Z _R (p) and Z _P (p) by their respective expression, obtains:

1 1

- + -

1

(6)

Which gives after simplification

C + C

F (p) = ^{ref pzt}

The voltage across the actuator (20) is then

V _pzl (p) = V _s (p) -V _e (p) = V _e (p)

(8)

Thus, in dynamics, the gain of the amplifier in charge therefore depends only on the inherent capacitance value of the piezoelectric actuator and the capacitor in series with the latter. In considering the capacitance of the fixed reference capacitor, that of the actuator varies according to certain parameters including the voltage at its terminals. The expression of the operating frequency of the amplifier is then given by the equation: f = - ÷ * -

Those skilled in the art find that the frequency of the output signal, therefore at the terminals of the piezoelectric actuator depends solely on the value of the output current i _s and the input voltage V _e of the amplifier.

Thus, this type of load-controlled circuit has the disadvantage of having a creep effect and not allowing a continuous control.

FIG. 3 illustrates a hybrid control device according to one embodiment of the invention, combining the two control modes, in load and in voltage.

The device (300) combines a first circuit (302) for load control control with a second circuit (304) for voltage control control, the two circuits being coupled to an operational amplifier (30).

Advantageously, the hybrid assembly of the invention allows switching between the two charge / voltage modes to be done automatically.

The amplifier (30) receives an input voltage V _e on a first non-inverting input (+), and outputs a output voltage V _s at output OUT.

The first circuit (302) comprises a piezoelectric actuator (32) and a capacitor (34) in series. The piezoelectric actuator (32) is connected to the OUT output and to the inverting input of the operational amplifier (30). The capacitor (34) in series is connected to the inverting input of the operational amplifier and to a low voltage V ₀ .

The first circuit is a capacitive circuit comprising a piezoelectric actuator and a capacitor in series, said capacitive circuit being connected by the free end of the piezoelectric actuator to the output of the operational amplifier and by the free end of the capacitor to a capacitor. low voltage terminal.

The second circuit (304) comprises discrete serial components comprised of a first resistor (36) and a second resistor (38) in series. The first resistor (36) is connected to the output UT 'and to the inverting input of the operational amplifier (30). The second resistor (38) is connected to the inverting input of the operational amplifier and to a low voltage V ₀ . The second circuit is a resistive circuit comprising a first resistor and a second resistor in series, said resistive circuit being connected by the free end of the first resistor to the output of the operational amplifier and by the second resistor to a voltage terminal low. The low voltage may preferentially be brought back to the ground.

The inverting output (-) of the amplifier (30) corresponds to a common 'P _c ' connection point between the two capacitive (302) and resistive (304) circuits. The connection point P _c is common between the actuator (32) and the capacitor (34) for the first circuit (302), and between the first resistor (36) and the second resistor (38) for the second circuit (30). 304). The common connection point P _c has a voltage marked 'Vcret'. Thus, the device is characterized in that the inverting input of the operational amplifier is on the one hand connected to the common ends of the piezoelectric actuator and the capacitor, and on the other hand connected to the common ends between the first and the second resistor, the capacitive circuit and the resistive circuit forming an impedance bridge.

The device of the invention thus provides an impedance bridge formed by the capacitors (32, 34) of the first circuit (302) and by the first and second resistors (36, 38) of the second circuit (304). The equilibrium of this bridge is satisfied when: l ^< ^ L = ^R IIL _{(1 0} ) where

C Ώ

^ ref ^ pzt

- Cpzt represents the inherent capacity of the piezoelectric actuator (32);

- Cret represents the capacity of the capacitor (34); - Rpzt and R _re f respectively represent the first and second resistors of the second circuit (304).

Advantageously, the values of the resistors of the resistive bridge (304) are defined to compensate for several effects:

a choice of resistors of high values causes a reduction in the bandwidth of the assembly, and vice versa;

- on a choice of resistors of low values, the control signal must be of high frequency in order to allow the command in load;

a choice of high values of the resistors R _pz t and R _re f of the order of ΜΩ to ΘΩ has the disadvantage that the Discrete resistors are in the same order of magnitude as the values of the leakage resistors of the capacitors, which does not then make it possible to achieve the optimization posed by

R c

R C

the equality w and does not allow to obtain a hybrid control decreasing the hysteresis of the actuators.

In the Laplace domain, the output voltage V _s (p) of the amplifier and the voltage across the piezoelectric actuator Vpzt (p) are expressed respectively by the following equations (1 1) and (12). :

1+ ^{rei pz} '(C _ref + C _pzt ) p

Vs {p) = "> * * ^Kc " ^{+ K} >"* _Ve (p) (1 1)

where the parameters represent:

- 'p' the Laplace variable;

- 'Cpzt' the inherent capacity of the piezoelectric actuator; - 'Rpzt' resistance in parallel with the actuator;

Ccref the capacitance of the reference capacitor;

Rcref resistance in parallel with the reference capacitor.

The value of the cut-off frequency of the device of FIG. 3 can be determined as a function of the maximum output current i _smax delivered by the operational amplifier, of the amplitude A _e of the input voltage Ve (t) considered sinusoidal and capacitance value of the reference capacitor C _ref , according to the following equation: / = (13)

2 * π * Ae * Cc ref

Considering equation (12), one skilled in the art finds that for a given input voltage, when 'p' tends to 0, the gain of the transfer function tends to 'R _pzt / Rcref' and when 'p' goes to infinity, the gain of the transfer function tends to 'Cc _re f / C _pz t'.

From a physical point of view, when the frequency tends to 0, capacitors C _pzt (32) and C _ref (34) are equivalent to open circuits. The output current flows only through the resistors Rp _Zt (36) and 'Rc _re f' (38). In this mode, the device behaves like a voltage control circuit.

When the frequency tends to infinity, the capacitors C _pz t (32) and C _ref (34) are equivalent to short circuits. All the output current flows through these two capacitors. In this mode, the device behaves like a load control circuit. Advantageously, at the extreme frequencies, low and high frequencies, there is no phase difference between the input and output signals. The gain at low frequencies is set by the resistors (36,38) while at high frequencies it is fixed by the capacitors (32,34).

The output current is given by the following equation: iM (14)

The time constant 'τ' of the assembly is: τ - Rpzt ^* C _pz , (15) In order for switching of the load and voltage control modes to be automatic, the low and high frequency voltage gains must be equal.

When this condition of equal voltage gains is fulfilled, the behavior of the device of the invention is as follows:

After applying a control signal at a given frequency, the piezoelectric actuator moves to a certain position with reduced hysteresis. It is the control mode in charge with the capacitive part of the first circuit (302), composed of C _ref and Cp _Zt which is involved.

Then, in steady state, with the application of a continuous signal, the resistive portion of the second circuit (304) composed of R _pzt and R _ref goes into operation. This is the voltage control mode.

The gains in control in load and in voltage being identical, the voltage across the piezoelectric actuator (32) is constant.

Thus, advantageously, the actuator no longer moves, the displacement being maintained stable. There is therefore more creep unlike known load control devices.

The control of piezoelectric actuators for Fabry-Perot interferometers must meet precise positioning requirements. Indeed, these piezoelectric actuators must ensure the displacement and alignment of the interferometer blades for a 0-200μηι stroke with a positioning accuracy of the order of one nanometer, a very high reproducibility of the positioning (very low hysteresis) and a linearity error, as well as very weak drifts (very low creep). The device of the present invention makes it possible to meet these requirements. FIG. 4 illustrates an exemplary implementation of the device of the invention for an application to the control of mobile mirrors in Fabry-Perot interferometers.

The hybrid control device as detailed in FIG. 3, is represented in FIG. 4 by the references (40) for the operational amplifier, (42) for the piezoelectric actuator C _pz t, (44) for the capacitor of FIG. reference C _ref, (46) and (48) respectively to the first and second resistors.

Various conventional application electronic components for those skilled in the art are added around the device of the invention to allow the desired operational operation. Thus, without being limiting, capacitors (C1, C2, C3, C4) may be added to enable decoupling of the supply voltages (+ V _DC , -V _cc ), unidirectional transils diodes (D1, D2) for limiting the voltage across the piezoelectric actuator (42). A resistance R | _im can be added to limit the current delivered in the actuator and a phase compensation capacitor C _cmp . Diodes (D3, D4) for protecting the differential input of the amplifier (40) can be added, as well as transil diodes (D5, D6) to protect the circuit against overvoltages and polarity inversions of the voltages of the amplifier. 'food.

In a preferred implementation, the amplifier may be a high voltage amplifier powered between -60V and + 160V, the voltage across the actuator limited to + 150V and -20V respectively, the current output in the actuator limited to 130mA.

The table below reports the results of the comparison of the hysteresis curves obtained with an operational gain amplifier 20 controlling several piezoelectric actuators, by a conventional voltage control, and by a hybrid control according to the device of the present invention.

The piezoelectric actuators used have the following characteristics:

The comparison in the table below is given for measurements taken on three piezoelectric actuators.

Actuator Hysteresis Hysteresis Factor

Hybrid voltage reduction amplifier control 1 19.01% 0.76% 96.00%

2 19.42% 0.85% 95.62%

3 16.82% 0.23% 98.63%

On the last column, it can be appreciated the significant decrease of the hysteresis whatever the actuator. Indeed, the voltage commands ^(2nd column) resulting in a hysteresis of about 20% of the total amount of displacement of the piezoelectric actuators while the hybrid amplifier ^(3rd column) the same hysteresis is reduced at about 1%.

FIG. 5 illustrates a variant of implementation of the device of the invention in the context of a unipolar control. The hybrid control device as detailed in FIG. 3, is represented in FIG. 5 by the references (50) for the operational amplifier, by (52) for the piezoelectric actuator C _pzt , by (54) for the capacitor reference C _ref, and by (56) and (58) respectively to the first and second resistors. In this arrangement, one terminal of the actuator (52) is connected to ground, the other terminal participating in the midpoint of the impedance bridge.

Various conventional application electronic components for those skilled in the art are added around the device of the invention to allow the desired operational operation. Thus, without this being limiting, unidirectional transils diodes (D3, D4) may be added to limit the voltage across the piezoelectric actuator (52).

In the assembly of Figure 5, an instrumentation amplifier (510) is added, whose purpose is to equalize the reference voltage and the voltage across the assembly (R _ref, C _ref). One input of the instrumentation amplifier receives the output of the operational amplifier, the other input of the instrumentation amplifier being connected to the midpoint of the impedance bridge. The output of the instrumentation amplifier is looped back to the inverting input of the operational amplifier.

By asking

mounting makes it possible to realize an ideal charging amplifier at low frequency.

The operation of the hybrid amplifier is the combination of the two control modes (voltage and load), it is important to identify the frequency range where each mode of operation is dominant. This frequency range is characterized by the central frequency defined by the equation:

For frequencies higher than f _c , it is then the control in charge which is implemented and for frequencies lower than f _c it is the control in tension which is dominant. Such an assembly is particularly suitable for the unipolar control of piezoelectric actuators whose control does not require high voltages, of the order of 10V.

FIG. 6 illustrates a variant of implementation of the device of the invention in the context of a unipolar control for the cases where the control voltage is high, such as the cases of large displacements. For reasons of simplification, all the components of the variant of FIG. 4 are uniquely referenced (400) but are not redescribed. In this variant, a non-inverting amplifier (60) is added upstream of the hybrid amplifier. In this arrangement, the first amplifier (60) serves to amplify the control signal V _in . The gain of this amplifier is then given by the relation: ⁸ .

The second part of the assembly consists of the hybrid amplifier (400) having one of the terminals of the actuator connected to ground.

R C

In such an assembly, equal ^re f ^ is retained, to obtain the same reduction of hysteresis in the circuit of FIG 4. Advantageous applications of this circuit are the adaptive optics applications for controlling deformable mirrors in open loop.

Thus, the present description illustrates various non-limiting implementations of the invention. The examples have been chosen to allow a good understanding of the principles of the invention, but are in no way exhaustive and should allow the skilled person to make modifications and implementation variants while maintaining the same principles.

It is shown that the hybrid amplifier of the invention makes it possible to significantly reduce the hysteresis amplitude of the piezoelectric actuators. The proposed device thus makes it possible to linearize the displacement of an actuator as a function of the voltage applied across its terminals or of the setpoint voltage.

Another advantage of the device of the invention is that it does not require a hysteresis model (Preisach, Maxwell, ...) and can therefore operate with any piezoelectric actuator, within the limits of the electronic assembly. The only parameters to know for its implementation are the inherent capacity of the piezoelectric actuator at rest and its bandwidth, such data being generally available in the catalogs of manufacturers of piezoelectric actuators. The application areas of the hybrid amplifier of the invention may be for example nano and micropositioning, mechatronic applications or the field of virtual reality.

Claims

claims

A piezoelectric actuator control device comprising:

an operational amplifier having an inverting input and a non-inverting input, and an output for outputting an output voltage;

a resistive circuit comprising a first resistor and a second resistor in series, said resistive circuit being connected by the free end of the first resistor to the output of the operational amplifier and by the second resistor to a low voltage terminal;

the device being characterized in that the inverting input of the operational amplifier is connected to the common ends of the piezoelectric actuator and the capacitor, and is connected to the common ends of the first and second resistor, the capacitive circuit and the circuit resistive forming an impedance bridge.

The device according to claim 1 wherein the ratio of the impedances of the capacitive circuit is equal to the ratio of the impedances of the resistive circuit.

The device of claim 1 wherein a control signal at a given frequency applied to the non-input Inverter of the operational amplifier allows obte command in charge of the piezoelectric actuator.

The device according to any one of claims 1 to 3 wherein a continuous control signal applied to the non-inverting input of the operational amplifier provides voltage control of the actuator

piezoelectric.

The device according to any one of claims 1 to 4 wherein the operational amplifier is powered between -60V and + 160V.

The device according to any one of claims 1 to 5 further comprising unidirectional transils diodes connected across the piezoelectric actuator to limit the voltage at said terminals.

The device of claim 6 wherein the voltage across the piezoelectric actuator is limited

respectively at + 150V and -20V.

The device of claim 7 further comprising a resistor connected to the output of the operational amplifier for limiting the current delivered in the piezoelectric actuator.

9. An interferometer controlled by a device according to one

any of claims 1 to 8, the interferometer being to be a Fabry-Perot interferometer.