MEMS loudspeaker with closed control system

Abstract

MEMS loudspeaker (1) for generating sound waves in the audible wavelength spectrum with a support element (9), a membrane (2) that can be deflected relative to the supporting element (9) along a z-axis, at least one piezoelectric actuator (7) supported on the support element (9) for deflecting the membrane (2), which is connected to the membrane (2) via a lifting structure (6), and an electronic control unit (11) for controlling the actuator (7), characterized by that the MEMS loudspeaker (1) includes at least one position sensor (19) by means of which a sensor signal (37) dependent on the membrane deflection can be provided to the control unit (11), that the position sensor (19) is at least partially integrated into the actuator (7), so that the position of the membrane (2) along the z-axis can be indirectly detected with it, that a connecting element (22) connects a free end of the position sensor (19) to the hub structure (6) and that the control unit (11) is designed in such a way, that it takes into account the elastic vibration properties of the connecting element (22) and that it performs a functional self-test procedure, a distortion reduction procedure, a damage protection procedure and / or a compensation procedure for changes in behavior, analyzing the sensor signal (37) and / or comparing it with a reference signal (8) stored in the control unit (11) and controlling the actuator (7) in a controlled manner, taking into account the analysis result (35) and / or comparison result (34).

Description

The present invention relates to a MEMS loudspeaker for generating sound waves in the audible wavelength spectrum, having a support element, in particular a membrane support frame, a membrane which can be deflected with respect to the support element along a z-axis, at least one piezoelectric actuator which is supported on the support element for deflecting the membrane, and an electronic control unit for actuating the actuator.The term MEMS stands for microelectromechanical systems. Such systems are installed in particular in electronic devices that offer only little installation space. The performance of known MEMS loudspeakers is largely dependent on ideal environmental conditions. Even small blows or other environmental influences, such as temperature or air pressure, can have a disadvantageous effect on the performance of the system. Moreover, aging effects can negatively affect the performance of such MEMS speakers. However, the present requirements for such MEMS loudspeakers require an at least constant sound quality in the long term even when external influences change.US 2014 / 0177881 A1 discloses a membrane device. This comprises a support, a membrane suspended on the support and a first actuator which is in contact with the membrane for applying a force. A second actuator, also in contact with the diaphragm, serves to apply a further force. Means for determining the position of the membrane relative to the support are also present. Control means of the first and second actuators apply a deformation signal to one of the actuators. A control signal is applied to the other actuator to control the displacement of the diaphragm. The application of the control signal is based on the position of the membrane.A MEMS loudspeaker is known from DE 10 2014 106 753 A1. This serves to generate sound waves in the audible wavelength spectrum. It comprises a carrier substrate having a substrate cavity and two substrate openings on opposite sides. A piezoelectric actuator structure is arranged in the region of a substrate opening and is connected to the edge region of the carrier substrate. A membrane anchored in its edge region can be set into oscillation by the actuator structure. In addition, a coupling element is present, which connects the actuator structure to the membrane and enables oscillations with respect to the carrier substrate.It is an object of the present invention to provide a MEMS loudspeaker with which the disadvantages known from the prior art can be eliminated, in particular to provide a MEMS loudspeaker which has a longer lifetime, an improved performance and / or an improved sound quality compared to those known from the prior art.The object is achieved by a MEMS loudspeaker having the features of independent claim 1.A MEMS loudspeaker for generating sound waves in the audible wavelength spectrum is proposed. The MEMS loudspeaker has a supporting element and a membrane which can be deflected along a z-axis with respect to the supporting element. The support element can be a support frame. The membrane preferably spans a cutout of the supporting frame, so that the membrane can be deflected in the z direction with respect to the supporting frame in order to generate sound waves. Furthermore, the MEMS loudspeaker comprises at least one piezoelectric actuator, which is supported on the supporting element, for deflecting the diaphragm. The piezoelectric actuator is preferably designed as a edged filer or cantilever. Accordingly, one of its two ends is firmly fastened to the support element, whereas the free end is deflectable in the z direction. With the other end freely movable relative to the support element, the actuator is indirectly connected to the membrane via a support structure, so that when the actuator is deflected, the membrane is simultaneously deflected in the z-axis. The piezoelectric actuator preferably comprises at least one piezoelectric layer.In addition, the MEMS loudspeaker comprises an electronic control unit for driving the actuator. The MEMS loudspeaker furthermore has at least one position sensor. With this, the position or deflection of the membrane along the Z axis can be indirectly detected. The position sensor is preferably a piezoelectric, a piezoresistive and / or a capacitive sensor. By means of the position sensor, a sensor signal dependent on the diaphragm deflection can be provided to the control unit.The control unit is designed to carry out a self-function test. The control unit is also designed to reduce distortions. In addition, the control unit is designed to protect the moving components of the MEMS loudspeaker from overloading and consequently damage. Furthermore, the control unit is designed to be able to compensate for changes in the behavior of the MEMS loudspeaker caused by aging effects and / or external influences. For this purpose, the control unit is designed such that the sensor signal can be analyzed and / or adjusted with a reference signal stored in the control unit. By analyzing the actual sensor signal, properties of the MEMS loudspeaker, such as its natural frequency, can be determined, for example. Furthermore, for example, a non-linear oscillation of the sensor signal can be identified.By comparing the sensor signal with the reference signal, changes in the behavior of the MEMS loudspeaker can be determined when a deviation is determined. Such changes in behavior can be caused, for example, by external influences, in particular temperature and / or pressure changes. Likewise, however, aging effects, in particular of the piezoelectric layer of the actuator, can also cause a change in the behavior. Thus, the piezoelectric coefficient may be degenerate by external influences or aging effects, thereby reducing piezoelectric performance. As a result, at a reference voltage, the piezoelectric actuator may produce an ever-decreasing force over time and, as a result, an ever-decreasing stroke of the diaphragm.Furthermore, however, nonlinear oscillations of the MEMS loudspeaker can also be identified by such a matching, which oscillations cause distortion of the output sound. However, overloads which can lead to damage to the moving components of the MEMS loudspeaker can also be identified by such an analysis and / or such a matching.In order to be able to avoid damage to the MEMS loudspeaker and to be able to ensure the sound quality at a constantly high level independently of external influences and aging effects, the control unit is furthermore designed in such a way that the actuator can be controlled in a controlled manner taking into account the analysis result and / or the comparison result. Accordingly, the actuator can be operated with a higher electrical voltage for the following, for example, when a reduction in power caused by aging or by external influences is detected. In this case, the piezoelectric actuator is accordingly operated with a higher voltage compensating for the reduction in power. Likewise, however, the piezoelectric actuator can also be operated with a reduced voltage in the event of a detected overload and / or distortion. In summary, it can therefore be stated that the MEMS loudspeaker can be operated with a control unit embodied according to the preceding description with a long-lived sound quality and / or with a constantly high sound quality over a longer period of time.It is advantageous if at least one signal value can be analyzed and / or adjusted with a reference value by means of the control unit. Additionally or alternatively, it is further advantageous if the temporal signal profile of the sensor signal can be analyzed and / or matched to a reference signal profile. This can preferably take place within a time window and / or a defined frequency range. Thus, it is particularly advantageous to analyze at least one individual signal value and / or to compare it with a reference value in order to be able to carry out a functional test, to be able to adjust a change in behavior and / or to be able to identify an overload. Furthermore, a non-linear oscillation can be established in particular by the analysis and / or the matching of a signal profile for distortion reduction.For the self-function test and / or for adjusting a change in behavior, it is advantageous if the control unit is designed such that an actual standard behavior of the MEMS loudspeaker can be determined by analyzing the actual sensor signal transmitted from the position sensor to the control unit. This determined actual standard behavior can then be stored in a memory of the control unit as a reference standard behavior, in particular during the first startup of the MEMS loudspeaker. This actual standard behavior established in a first step then forms the new reference signal or the new target standard behavior, by means of which a new change in behavior of the MEMS loudspeaker can be established by a renewed adjustment.For the self-function test, it is furthermore advantageous if a setpoint sensor signal dependent on a reference actuator signal is stored in the control unit. This setpoint sensor signal accordingly represents the setpoint standard behavior of the MEMS loudspeaker.For the self-function test, it is furthermore advantageous if the control unit is designed such that the actuator can be controlled with the reference actuator signal, which causes or generates a desired deflection of the diaphragm. After this activation with the reference actuator signal, it is advantageous if the actual sensor signal dependent on the actual deflection can be matched to the desired sensor signal. It is furthermore advantageous if the control unit is able to determine a deviation of the actual sensor signal from the desired sensor signal during this comparison. Additionally or alternatively, it is advantageous if the control unit outputs an error message in the event of a detected deviation and / or reports it to a higher-level control unit. As a result, the MEMS loudspeaker can preferably check independently whether it is operating properly. If this is not the case, it is advantageous if the control unit is designed in such a way that it compensates for this change in behavior when the actuator is activated by adapting the electronic activation.Furthermore, it is advantageous if the control unit is designed in such a way that it is able to carry out the self-function test automatically in the event of an event detected by sensors. Such an event is in particular an identified system start, a sensorially detected external impact effect or a sensorially detected change in the behavior of the MEMS loudspeaker.In the self-function test, it is advantageous if the control unit is designed in such a way that it can differentiate at least between two types of faults, in particular actuator damage, membrane damage and / or a blockage of the sound outlet opening. This allows a targeted countermeasure to be initiated in order to be able to at least partially compensate for the identified error.Advantageous embodiments of the control unit for distortion reduction are provided below. Accordingly, it is advantageous if the control unit for distortion reduction is designed in such a way that a non-linear oscillation of the MEMS loudspeaker can be determined by the analysis of the actual sensor signal. For this purpose, the temporal signal profile of the sensor signal is preferably analyzed. By a non-linear oscillation identified in this way, distortions can be detected.For the purpose of reducing distortion, it is advantageous if the control unit is designed in such a way that the actuator can be actuated with a compensation actuator signal which at least partially reduces the determined non-linear oscillation. This allows the sound quality of the MEMS loudspeaker to be improved.To achieve damage protection, the control unit is preferably designed according to the following description. Accordingly, it is advantageous if the control unit is designed such that a maximum deflection of the membrane can be determined by the analysis of the actual sensor signal, in particular the height of at least one signal value. As a result, an already occurring or imminent overloading of the moving components of the MEMS loudspeaker can be reliably identified.For damage protection, it is furthermore advantageous if a threshold value with a maximum permissible threshold value level is stored in the control unit. The threshold value or its threshold value height defines the maximum permissible deflection along the z-axis, up to which the MEMS loudspeaker can be operated at least without damage.In addition, it is advantageous if the control unit is designed to protect the MEMS loudspeaker from damage in such a way that, in order to detect an overload, the actual sensor signal, in particular at least one signal single value, which is dependent on a first deflection, can be matched to the maximum permissible threshold value. It is furthermore advantageous if the control unit is designed to actuate the actuator with an, in particular reduced, actuator signal which generates a second deflection reduced for the first deflection when the threshold value is identifiedly exceeded. This advantageously makes it possible to prevent the moving components of the MEMS loudspeaker from being deflected to such an extent that they are damaged.The following specifications of the control unit are advantageous in particular for damage protection and / or for compensating changes in the behavior of the MEMS loudspeaker. Accordingly, it is advantageous if the control unit is designed for this purpose in such a way that an actual natural frequency of the MEMS loudspeaker can be determined by the analysis of the actual sensor signal. The actual natural frequency represents a property representing the current behavior of the MEMS loudspeaker. Furthermore, knowledge about the actual natural frequency is necessary in order to be able to avoid damage to the MEMS loudspeaker in the region of the natural frequency in the case of a control which is too long in time.It is advantageous if an analysis signal, in particular a noise signal, is stored in the control unit for determining the actual natural frequency. The analysis signal preferably extends within a frequency range below the resonant frequency of the MEMS loudspeaker.In this regard, it is furthermore advantageous if the control unit for determining the actual natural frequency is designed such that the actuator can be controlled with the analysis signal and / or a maximum identifying the actual natural frequency can be determined within the frequency range. As a result, the actual natural frequency of the MEMS loudspeaker can be determined quickly and with simple means.For damage protection and / or for compensating changes in behavior, it is advantageous if a reference natural frequency is stored in the control unit.In addition, it is advantageous in this regard if the control unit is designed for damage protection and / or for compensating changes in behavior in such a way that the ascertained actual natural frequency of the MEMS loudspeaker can be matched to the stored reference natural frequency.Additionally or alternatively, it is further advantageous if, in the event of a deviation of the actual natural frequency from the reference natural frequency, the determined actual natural frequency can be stored as a new reference natural frequency.In this regard, it is advantageous if the control unit is designed for damage protection and / or for compensating changes in behavior in such a way that the actuator is actuated taking into account the new reference natural frequency. Accordingly, the newly determined and stored reference natural frequency can now be taken into account in order to actuate the MEMS loudspeaker in the region of the new natural frequency in a time-reduced manner in order to avoid overloading. Furthermore, the new stored reference natural frequency serves to be able to identify further changes in behavior over time.Furthermore, a working method for a MEMS loudspeaker according to the preceding description is proposed, wherein the aforementioned features, in particular with respect to the functioning of the control unit, can be present individually or in any combination. Furthermore, in particular, a working method for self-function testing, for distortion reduction, for damage protection and / or for compensating changes in behavior is proposed, wherein at least one of these methods is designed according to the preceding description. In this case, the method features mentioned can likewise be present individually or in any combination.Further advantages of the invention are described in the following exemplary embodiments. It shows: FIG. 1 is a perspective sectional view of a MEMS loudspeaker, according to an embodiment of the present invention, FIG. 2 shows a schematic plan view of an exemplary embodiment of a piezoelectric actuator with an integrated position sensor, FIG. 3 shows a schematic plan view of a second exemplary embodiment of a piezoelectric actuator with an integrated position sensor, FIG. 4 shows a schematic side view of the second exemplary embodiment of a piezoelectric actuator with an integrated position sensor, FIG. 5 shows a schematic plan view of a third exemplary embodiment of a piezoelectric actuator with piezoresistive position sensor, FIG. 6 shows a schematic plan view of a fourth exemplary embodiment of a piezoelectric actuator with capacitive position sensor, and FIG. 7 shows a schematic illustration of the information foot between the control unit, actuator and position sensor, illustrating the operation of the MEMS loudspeaker.In the following description of the figures, relative terms, such as above, below, above, above, below, left, right, vertical and horizontal, are used to define the relationships between the various elements, with reference to the position of the objects represented in each of the figures. It is understood that these terms may change if they deviate from the position of the devices and / or elements shown in the figures. Accordingly, for example, in the case of an inverted orientation of the devices and / or elements illustrated with reference to the figures, a feature specified as above in the following description of the figures would now be arranged below. The relative terms used are thus used merely to simplify the description of the relative relationships between the individual devices and / or elements described below.FIG. 1 shows a first exemplary embodiment of a MEMS sound transducer, which is designed as a MEMS loudspeaker 1.The MEMS loudspeaker 1 is designed to generate sound waves in the audible wavelength spectrum. For this purpose, the MEMS loudspeaker 1 has a membrane 2 and a membrane carrier 3. The membrane 2 is connected in its edge region 4 to the membrane carrier 3 and is capable of vibrating along a z-axis with respect to the membrane carrier 3. The z-axis extends substantially perpendicular to the membrane 2.In addition to the diaphragm 2, the MEMS loudspeaker 1 has a lifting structure 6, which is coupled to the diaphragm 2, and at least one piezoelectric actuator 7. The actuator 7 is connected to the membrane 2 movable in the z direction via a lifting structure 6. The membrane carrier 3 is arranged on a carrier substrate 9 of the piezoelectric actuator 7. The piezoelectric actuator 7 is arranged below the membrane 2 and / or substantially parallel thereto. The piezoelectric actuator 7 is designed to cause a uni- or bi-directional lifting movement of the lifting structure 6 in order to deflect the diaphragm 2. It interacts with the diaphragm 2 in order to convert electrical signals into acoustically perceptible sound waves. The piezoelectric actuator 7 or the piezoelectric actuator structure is arranged on a side of the carrier substrate 9 facing away from the membrane 2.Furthermore, the MEMS loudspeaker 1 comprises a printed circuit board 10. The control unit 11 is thus completely encapsulated. In addition to the control unit 11, further passive components 12, such as electrical resistors and / or I / O contacts, can also be embedded in the printed circuit board 10 and / or arranged on it. The MEMS loudspeaker 1 and in particular the piezoelectric actuator 7 are connected to the control unit 11 by electrical contacts not shown in any further detail in the figures. The MEMS loudspeaker 1 can thus be controlled or operated via the control unit 11 so that the diaphragm 2 is set into oscillation with respect to the diaphragm carrier 3 for generating sound energy by the piezoelectric actuator 7. The piezoelectric actuator 7 is designed as a cantilever arm 13. Accordingly, it has a fixedly clamped end and a freely oscillating end.The MEMS loudspeaker 1 is arranged in a housing 14 according to FIG. 1. The housing 14 comprises an upper housing part 15 and a lower housing part 16. This is arranged laterally on an outer surface of the MEMS loudspeaker 1. The housing 14 offers in particular additional protection for the membrane 2, since it covers it from the environment.The MEMS loudspeaker 1 has at least one position sensor 19. The position sensor 19 is designed to provide the electronic control unit 11 with a sensor signal dependent on the diaphragm deflection. The control unit 11 is designed to actuate the actuator 7 in a regulated manner on the basis of the sensor signal. For this purpose, the position sensor 19 can be a piezoelectric, a piezoresistive and / or a capacitive sensor. The position sensor 19 is at least partially integrated in the actuator 7, in particular in the cantilever 13.In the exemplary embodiment shown in FIG. 1, the position sensor 19 can be designed according to the following exemplary embodiments shown in the figures. Accordingly, the position sensor 19 and the piezoelectric actuator 7 according to the exemplary embodiment shown in FIG. 2 are formed by a common piezoelectric layer 41 (cf. FIG. 2 ). The piezoelectric layer 41 is formed of lead zirconate titanate (PZT). At least one region is a sensor region 20, by means of which two actuator regions 21 are arranged spaced apart from one another. The sensor and actuator regions 20, 21 are electrically insulated from each other. Since the requirements for sensor system and actuator system can differ, a combination of different piezoelectric materials with different properties is also conceivable. In this case, the sensor region 20 can be formed from PZT and the actuator region 21 from aluminum nitride (AlN).According to FIG. 2, the sensor region 20 is arranged between the two actuator regions 21 and extends symmetrically in the cantilever longitudinal direction. The actuator regions 21 are completely separated from one another by the sensor region 20. The sensor region 20 and the actuator region 21 have the same length in the cantilever longitudinal direction. The two actuator regions 21 are larger in area than the sensor region 20.In the event of a deflection of the diaphragm 1 via the actuator 7, its position or deflection in the z direction is now detected by means of the position sensor 19. In this case, the voltage generated via the piezoelectric effect, which is approximately proportional to the deflection of the lifting structure 6, is tapped via the actuator electrodes and evaluated accordingly. The control unit 11 determines the actual position or actual deflection of the diaphragm 2 on the basis of this detected input signal. The connecting element 22 connects a free end of the position sensor 19 to the lifting structure 6. Depending on this detected actual position of the diaphragm 2, the control unit 11 determines a desired desired position of the diaphragm and / or an electronic output signal depending thereon. The output signal is transmitted to the actuator 7, which deflects the membrane 2 accordingly. During and / or at the end of the deflection movement, the actual position of the diaphragm 2 is detected again via the position sensor 19 and, if appropriate, is adapted again to the ambient conditions according to the preceding description.FIG. 2 shows, as already mentioned above, a schematic plan view of a first exemplary embodiment of a piezoelectric actuator 7 with integrated position sensor 19. Both regions 20, 21 are formed from PZT. However, other piezoelectric materials are also conceivable. It would also be conceivable for a large part to be used for the actuator system and only a small region to be used for the sensor. The sensor region 20 is electrically insulated from the actuator regions 21. In order to prevent an unwanted tilting of the lifting structure 6 due to an asymmetric drive, the actuator and sensor regions 21, 20 should each be arranged opposite one another in pairs.FIGS. 3 and 4 each show a schematic view of a second exemplary embodiment of the piezoelectric actuator 7 with position sensor 19. The piezoelectric actuator 7 is formed by a second piezoelectric layer 24, in particular made of PZT. The two layers are electrically insulated from one another and are arranged one above the other with respect to the z-axis.The first piezoelectric layer 23 is divided into a plurality of sensor regions 20 according to FIG. 3. The sensor regions 20 are separated from one another and / or electrically insulated. In the exemplary embodiment shown in FIG. 3, three sensor regions 20 are formed, which are arranged spaced apart from one another in the transverse cantilever direction. This is effected in particular equidistantly. The second piezoelectric layer 24 has an actuator region 21 extending over the cantilever arm 13. This actuator region 21 extends over the whole surface of the cantilever arm 13, at least in a plan view. However, it is also conceivable for the sensor region 20 not to extend over the entire cantilever longitudinal direction, but only over a part thereof. In this case, the difference to the cantilever length would be formed by a further actuator region, not shown.As shown in FIG. 4, both piezoelectric layers 23, 24 form a stack supported by a support layer 25. The support layer 25 is connected to the circuit board 10. In the exemplary embodiment shown, the first piezoelectric layer 23, which forms the position sensor 19, is arranged above the second piezoelectric layer 24, in particular the actuator 7. However, the first piezoelectric layer 23 could also be arranged under the piezoelectric actuator 7.FIG. 5 shows a schematic plan view of a third exemplary embodiment of a piezoelectric actuator 7 with an integrated position sensor 19. The power line 26 is formed in the support layer 25 of the piezoelectric actuator 7 by an ion implantation method. The power line 26 extends from a first electrical contact 27 to a second electrical contact 28. The power line 26 is U-shaped and has a first longitudinal part 30 and a second longitudinal part 31. The first longitudinal part 30 extends from the first electrical contact 27 in the cantilever longitudinal direction into the cantilever 13. The second longitudinal part 31 extends from a transverse part 32 in the cantilever longitudinal direction out of the cantilever 13 to the second electrical contact 28, wherein the transverse part 32 extends in the cantilever transverse direction. In the present case, four such electrical resistors 33 are formed in the manner just described. The resistors 33 are different from one another and are connected to the control unit 11 in such a way that a Wheatstone bridge is formed.The current lines 26, in particular the resistors 33, react to deformations which arise as a result of the pressure change which results from the diaphragm deflection. The resistors 33 react to this with a change in resistance, which is detected and evaluated by the control unit 11.FIG. 6 shows a schematic plan view of a fourth exemplary embodiment of the piezoelectric actuator 7 with integrated capacitive position sensor 19. Each extension is movable in the z-direction. In the exemplary embodiment shown, the recesses are arranged on a frame 36 and the extensions on the cantilever arm 13. The cantilever arm 13 is likewise deflectable in the z direction. The frame 36, on the other hand, is stationary and is preferably formed by the carrier substrate 9. However, it is also conceivable for the recesses in the cantilever arm 13 and the extensions to be formed on the frame 36. The recess has two inner surfaces, wherein at least one of the inner surfaces is designed as a first measuring electrode. The extension is either designed as a second measuring electrode or as a dielectric. In this way, an electric capacitor is formed.As a result of excitation of the membrane 2 by the actuator 7, the extensions on the cantilever arm 13 are likewise deflected. The distance of the individual extensions from the respective corresponding recess then increases. Consequently, the distance of the two measuring electrodes or the distance of the first measuring electrode from the dielectric also increases. Since the capacitance is determined by just this distance, the control unit 11 detects a change in capacitance as a result of the deflection. Depending on this capacitive sensor signal, the actuator 7 can be controlled in a controlled manner in order to control the diaphragm 2 in a controlled manner (cf. also FIG. 1 in this regard).FIG. 7 schematically shows the mode of operation of the MEMS sound transducer, which is designed as a MEMS loudspeaker 1 according to the preceding description. The MEMS loudspeaker comprises the diaphragm 2 which is mounted in the support element or diaphragm carrier 3 such that it can be deflected in the z direction. Furthermore, the MEMS loudspeaker comprises the piezoelectric actuator 7 and the position sensor 19, which are in turn electrically connected to the control unit 11.The control unit 11 is designed to carry out a functional test, to reduce distortions, to avoid damage to the MEMS loudspeaker and / or to detect and compensate changes in the behavior of the MEMS loudspeaker. For this purpose, the control unit 11 is designed in such a way that it is able to analyze the actual sensor signal 37 and / or to compare it with a reference signal 8 stored in the control unit 11. Furthermore, the control unit 11 is designed such that the actuator 7 can be controlled in a controlled manner taking into account an analysis result 35 that can be determined by the analysis of the actual sensor signal 37 and / or comparison result.For the self-function test, the MEMS loudspeaker can accordingly actuate the actuator 7 with a reference actuator signal 48. A setpoint sensor signal 42 is stored as reference signal 8 in the memory unit of the control unit 11 and represents a setpoint deflection of the diaphragm 2 that is dependent on the reference actuator signal 48. By controlling the diaphragm 2 by means of the actuator 7 with the reference actuator signal 48, the latter is deflected. The deflection is detected by the position sensor 19, which transmits an actual sensor signal 37 to the control unit 11. The actual sensor signal represents the actual standard behavior 38 and / or serves to determine the actual standard behavior 38 of the MEMS loudspeaker. Within the scope of an actual / setpoint value adjustment, the control unit 11 compares the actual sensor signal 37 with the stored setpoint sensor signal 42. Otherwise, the control unit 11 detects an erroneous operation. The control unit 11 can perform a functional self-test described above automatically, in particular upon system start-up or after external interference effects.Furthermore, the control unit 11 is designed in such a way that it is able to analyze the incoming actual sensor signal 37. As analysis results 35, the control unit 11 can determine, in particular, a maximum deflection 44, an actual natural frequency 46 and / or a non-linear oscillation 43 of the MEMS loudspeaker. Furthermore, the control unit 11 has stored, in particular in a memory, a threshold value 45 and / or a reference natural frequency 47.Within the scope of a distortion reduction, the control unit 11 can accordingly recognize a non-linear oscillation 43 of the diaphragm 2 by analyzing the actual sensor signal 37. Such non-linear oscillations are an indication of distortion of the output sound. During the analysis and / or during the comparison of the actual sensor signal 37, the control unit 11 uses in particular at least one individual signal value and / or the temporal signal profile, in particular within a frequency range, of the sensor signal.In order to be able to compensate for this identified distortion, the control unit 11 is furthermore designed in such a way that it is able to actuate the actuator 7 with a compensation actuator signal 49, which at least partially reduces the determined non-linear oscillation 43. As a result, the diaphragm 2 can be returned to a non-distorting deflection region.In order to be able to avoid damage to the diaphragm 2, the actuator 7 and the position sensor 19 due to an overload, the control unit 11 is furthermore designed in such a way that it can first identify a corresponding overload by analyzing the actual sensor signal 37 and by adjusting it with a threshold value 45. Accordingly, the threshold value 45 defines a maximum permissible threshold value height up to which the MEMS loudspeaker can be operated at least without damage.By comparing the maximum deflection 44 and the threshold value 45, the control unit 11 can determine whether an overload is imminent or has already occurred. Shortly before the maximum deflection 44 exceeds the threshold value 45 or immediately after it has exceeded it, the control unit 11 controls the actuator 7 with a reduced actuator signal 39. This reduced actuator signal 39 compared to the previous actuator signal accordingly brings about a second deflection reduced with respect to the first deflection of the diaphragm 2. This advantageously makes it possible to avoid overloading of the moving MEMS components.Furthermore, for damage protection, knowledge of the actual natural frequency 46 of the MEMS loudspeaker is advantageous. The same applies to the compensation of changes in the behavior of the MEMS loudspeaker.In order to be able to determine the actual natural frequency 46 analytically, an analysis signal, which is not shown in detail here, is stored in the memory unit of the control unit 11. This is preferably a noise signal which extends in a frequency range below the resonant frequency of the MEMS loudspeaker. The actuator 7 is activated with this analysis signal. The actual sensor signal 37 fed back by the position sensor 19 is analyzed by the control unit 11 in order to be able to ascertain a maximum within the frequency range. This maximum now identifies the actual natural frequency 46 of the MEMS loudspeaker. The control unit 11 is now designed in such a way that it is able to ascertain a change in the natural frequency by comparing the analytically determined actual natural frequency 46 with the stored reference natural frequency 47.In the case of a detected deviation, it is advantageous for damage protection if the new natural frequency, i.e. the actual natural frequency 46, is stored as a new reference natural frequency 47 in the memory of the control unit 11. In order to be able to avoid damage to the MEMS loudspeaker, this must namely only be excited over a limited period of time in the range of the natural frequency. When further controlling the diaphragm 2, the control unit 11 now takes into account the new reference natural frequency 47 and accordingly controls the diaphragm 2 in a time-controlled manner in this frequency range.Furthermore, however, such a change in the natural frequency of the MEMS loudspeaker can also be a consequence of external influences, in particular pressure or temperature changes, or else of aging effects. For compensating such changes in the behavior of the MEMS loudspeaker, it is accordingly likewise advantageous if, in the case of an analytically determined natural frequency change, the actual natural frequency 46 is stored as a new reference natural frequency 47 and is subsequently taken into account in the control of the actuator 7. For this purpose, the control unit 11 controls the actuator 7 with a corresponding compensation actuator signal 49.The present invention is not limited to the illustrated and described embodiments. Modifications within the scope of the patent claims are possible as well as a combination of the features, even if these are shown and described in different exemplary embodiments.List of reference characters1 MEMS loudspeaker 2 membrane 3 membrane carrier 4 edge region 5 amplification element 6 lifting structure 7 actuator 8 stored reference signals 9 carrier substrate 10 printed circuit board 11 control unit 12 passive additional components 13 cantilever arm 14 housing 15 upper housing part 16 lower housing part 17 sound duct 18 acoustic inlet / outlet opening 19 position sensor 20 sensor region 21 actuator region 22 connecting element 23 first piezoelectric layer 24 second piezoelectric layer 25 carrier layer 26 current line 27 first electrical contact 29 firmly clamped end 30 first longitudinal part 31 second longitudinal part 32 transverse part 33 resistors 34 comparison result 35 analysis result 36 frame 37 actual sensor signal 38 actual standard behavior 39 reduced actuator signal 40 ASIC 41 common piezoelectric layer 42 target sensor signal 43 nonlinear oscillation 44 maximum deflection 45 threshold value 46 actual natural frequency 47 reference natural frequency 48 reference actuator signal 49 compensation actuator signal

Claims

MEMS loudspeaker (1) for generating sound waves in the audible wavelength spectrum, having a supporting element (9), a membrane (2) which can be deflected with respect to the supporting element (9) along a z-axis, at least one piezoelectric actuator (7) which is supported on the supporting element (9) for deflecting the membrane (2) and is connected to the membrane (2) via a lifting structure (6), and an electronic control unit (11) for actuating the actuator (7), characterized in that the MEMS loudspeaker (1) comprises at least one position sensor (19), by means of which a sensor signal (37) which is dependent on the membrane deflection can be provided to the control unit (11), in that the position sensor (19) is at least partially integrated into the actuator (7) such that the position of the membrane (2) along the z-axis can be indirectly detected therewith, a connecting element (22) connecting a free end of the position sensor (19) to the lifting structure (6), and the control unit (11) being designed such that it takes into account elastic vibration properties of the connecting element (22) and that it carries out a function self-test method, a distortion reduction method, a damage protection method and / or a compensation method of changes in behavior, wherein it analyzes the sensor signal (37) and / or balances it with a reference signal (8) stored in the control unit (11) and actuates the actuator (7) in a controlled manner taking into account the analysis result (35) and / or comparison result (34).MEMS loudspeaker according to the preceding claim, characterized in that at least one signal value and / or the temporal signal profile of the sensor signal (37) can be analyzed and / or adjusted, in particular within a frequency range.MEMS loudspeaker according to one or more of the preceding claims, characterized in that the control unit for the self-function test is designed in such a way that an actual standard behavior (38) of the MEMS loudspeaker can be determined by the analysis of the actual sensor signal (37).MEMS loudspeaker according to one or more of the preceding claims, characterized in that a setpoint sensor signal (42), which is dependent on a reference actuator signal (48) and which reproduces the setpoint standard behavior of the MEMS loudspeaker, is stored in the control unit (11) for the self-function test.MEMS loudspeaker according to one or more of the preceding claims, characterized in that the control unit (11) for the self-function test of the MEMS loudspeaker is designed in such a way that the actuator (7) can be controlled with the reference actuator signal (48) generating a desired deflection, the actual sensor signal (37) dependent on the actual deflection can be matched with the desired sensor signal (42), a deviation of the actual sensor signal (37) from the desired sensor signal (42) can be determined and / or an error message can be output in the case of a determined deviation.MEMS loudspeaker according to one or more of the preceding claims, characterized in that the control unit (11) for distortion reduction is designed in such a way that a non-linear oscillation (43) of the MEMS loudspeaker can be determined by the analysis of the actual sensor signal (37).MEMS loudspeaker according to the preceding claim, characterized in that the control unit (11) for distortion reduction of the MEMS loudspeaker is designed in such a way that the actuator (7) can be actuated with a compensation actuator signal (49) which at least partially reduces the determined non-linear oscillation (43).MEMS loudspeaker according to one or more of the preceding claims, characterized in that the control unit (11) for damage protection of the MEMS loudspeaker is designed in such a way that a maximum deflection (44) of the MEMS loudspeaker can be determined by the analysis of the actual sensor signal (37).MEMS loudspeaker according to one or more of the preceding claims, characterized in that a threshold value (45), in particular with a maximum permissible threshold value height up to which the MEMS loudspeaker can be operated at least without damage, is stored in the control unit (11) for damage protection.MEMS loudspeaker according to the preceding claim, characterized in that the control unit (11) for damage protection of the MEMS loudspeaker is designed in such a way that, in order to detect an overload, the actual sensor signal (37) dependent on a first deflection can be matched to the maximum permissible threshold value (45), and / or in that the actuator (7) can be actuated with a reduced actuator signal (39) when the threshold value (45) is exceeded, which generates a second deflection reduced in relation to the first deflection.MEMS loudspeaker according to one or more of the preceding claims, characterized in that the control unit (11) is designed for damage protection and / or for compensating changes in the behavior of the MEMS loudspeaker in such a way that an actual natural frequency (46) of the MEMS loudspeaker can be determined by the analysis of the actual sensor signal (37).MEMS loudspeaker according to the preceding claim, characterized in that the control unit (11) is designed to determine the actual natural frequency (46) in such a way that the actuator (7) can be controlled with an analysis actuator signal (50) and / or a maximum identifying the actual natural frequency (46) can be determined within the frequency range.MEMS loudspeaker according to one or more of the preceding claims, characterized in that a reference natural frequency (47) is stored in the control unit (11) for damage protection and / or for compensating changes in behaviour.MEMS loudspeaker according to one or more of Claims 11 to 13, characterized in that the control unit (11) is designed for damage protection and / or for compensating changes in behaviour in such a way that the determined actual natural frequency (46) of the MEMS loudspeaker can be matched to the stored reference natural frequency (47) and / or, in the event of a deviation, the determined actual natural frequency (46) can be stored as a new reference natural frequency (47).MEMS loudspeaker according to the preceding claim, characterized in that the control unit (11) is designed for damage protection and / or for compensating changes in behavior in such a way that the actuator (7) is controlled taking into account the new reference natural frequency (47), wherein this is carried out with a compensation actuator signal (49) for compensating the detected change in behavior.

Images