Air pulse generator having common mode movement and differential mode movement

Abstract

To provide an air pulse generator for improving defects of a conventional technology.SOLUTION: An air pulse generator has a membrane structure including a pair of flaps. The membrane structure is actuated so as to perform the common mode movement which forms an amplitude modulated ultrasonic air pressure change having an ultrasonic carrier frequency. The membrane structure is actuated so as to perform the differential mode movement which forms an aperture at a rate synchronized with the ultrasonic carrier frequency. The air pulse generator generates a plurality of air pulses in accordance with the amplitude modulated ultrasonic air pressure change.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to an air pulse generator, and more particularly, to an air pulse generator having common mode movement and operating mode movement.

Background Art

[0002] Speaker drivers and back enclosures are two major design challenges in the speaker industry. In conventional speakers, for example, it is not easy to cover the entire audio frequency band from 20 Hz to 20 kHz. In order to generate sounds with high fidelity and sufficiently high sound pressure levels (SPL), both the radiating / moving surface and the volume / size of the back enclosure of a conventional speaker need to be sufficiently large.

[0003] Therefore, how to design a small sound generating device while overcoming the design challenges faced by conventional speakers has become an important objective in this field.

Summary of the Invention

Problems to be Solved by the Invention

[0004] The main object of the present invention is to provide an air pulse generator that improves the drawbacks of the prior art.

Means for Solving the Problems

[0005] In an embodiment of the present disclosure, there is provided an air pulse generator, having a membrane structure including a flap pair, wherein the membrane structure is operated to perform a common mode movement in which an amplitude-modulated ultrasonic air pressure change having an ultrasonic carrier frequency is formed, and the membrane structure is operated to perform a differential mode movement in which an opening is formed at a speed synchronized with the ultrasonic carrier frequency. There is provided an air pulse generator that generates a plurality of air pulses in accordance with the amplitude-modulated ultrasonic air pressure change.

[0006] These and other objects of the present invention will be apparent to those skilled in the art from a reading of the following detailed description of the preferred embodiments shown in the various drawings and the drawings.

Brief Description of the Drawings

[0007]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Figure 15

Figure 16

Figure 17

Figure 18

Figure 19

Figure 20

Figure 21

Figure 22

Figure 23

Figure 24

Figure 25

Figure 26

Figure 27

Figure 28

Figure 29

Figure 30

Figure 31

Figure 32

Figure 33

Figure 34

DETAILED DESCRIPTION OF THE INVENTION

[0008] A basic aspect of the present invention relates to an air pulse generator, and more particularly to an air pulse generator comprising modulation means and demodulation means, said modulation means generating an ultrasonic air pressure wave / variation (UAW) having a frequency f UC and the amplitude of the UAW is modulated according to an input audio signal S IN which is an electrical (analog or digital) representation of the audio signal SS. Next, this amplitude-modulated ultrasonic air pressure wave / variation (AMUAW) is synchronously demodulated by said demodulation means, and the spectral components embedded in the AMUAW are shifted by ±n f UC (n is a positive integer). Here, n is a positive integer. As a result of this synchronous demodulation, the spectral components of the AMUAW corresponding to the audio signal SS are partially shifted to the baseband, and as a result, an audible audio signal SS is reproduced. Here, the amplitude-modulated ultrasonic air pressure wave / variation AMUAW corresponds to a carrier component having an ultrasonic carrier frequency f UC and the modulation component corresponds to the input audio signal S IN .

[0009] FIG. 1 shows a schematic diagram of an air pulse generation (APG) device 100 according to an embodiment of the present invention. The device 100 may be applied as an acoustic generation device that generates an acoustic sound in response to an input (audio) signal S IN but is not limited thereto.

[0010] Device 100 has a device layer 12 and a chamber defining layer 11. The device layer 12 has walls 124L, 124R and support structures 123R, 123L that support a thin film layer etched up to flaps 101, 103, 105, and 107. In one embodiment, the device layer 12 may be manufactured by a MEMS (Micro-Electro-Mechanical System) manufacturing process using a Si substrate with a thickness of 250 - 500 μm etched to form, for example, 123L / R and 124R / L. In one embodiment, on this Si substrate, a thin layer typically 3 - 6 μm thick, composed of a silicon-on-insulator SOI or a POLY-on-insulator POI layer, is etched, and flaps 101, 103, 105 and 107 are formed.

[0011] The chamber defining layer (which may also be referred to / named as the "cap" structure) 11 has a pair of chamber side walls 110R, 110L and a chamber ceiling 117. In one embodiment, the chamber defining layer (or cap structure) 11 may be manufactured using MEMS manufacturing techniques. A resonance chamber 115 is formed between this chamber defining layer 11 and the device layer 12.

[0012] In other words, device 100 may be regarded as having a membrane structure 10 and a cap structure 11, with a chamber 115 formed therebetween. The membrane structure 10 can be seen as including a modulation part 104 and a demodulation part 102. The modulation part 104 having (modulation) flaps 105 and 107 is configured to form and operate ultrasonic air / acoustic waves within chamber 115, and the air / acoustic waves can be seen as a kind of air pressure change that varies in both time and space. In one embodiment, the ultrasonic air / acoustic wave or air pressure change may be an amplitude DSB - SC (Double Sideband Suppressed Carrier) modulated air / acoustic wave having an ultrasonic carrier frequency f UC which may be, for example, in the range from 160 kHz to 192 kHz, which is significantly greater than the maximum frequency of human audible sound. UC

[0013] ​Hereinafter, the terms air wave and acoustic wave are used interchangeably.

[0014] (Demodulating) Demodulation unit 102 having flaps 101 and 103 is configured to operate in synchronization with modulation unit 102, and shifts the spectral components of the DSB-SC modulated acoustic wave generated by modulation unit 104 by ±n×f UC only, where n is a positive integer, and generates a plurality of air pulses toward the surroundings according to the ultrasonic air wave in chamber 115. The baseband frequency components of the plurality of air pulses (generated by demodulation unit 102 according to the ultrasonic air wave in chamber 115) are the input (audio) signal S IN or the input (audio) signal S IN corresponds to / is related to. The low-frequency components of the plurality of air pulses represent the frequency components of the plurality of air pulses within the audible spectrum (e.g., less than 20 or 30 kHz). In the present application, the baseband may usually be referred to as the audible spectrum, but is not limited thereto.

[0015] In other words, in the sound generation application, modulation unit 104 may be operated to form an air wave modulated according to input audio signal S IN , and demodulation unit 102 operates in synchronization with modulation unit 104 to generate a plurality of air pulses having the input audio signal S IN as (or corresponding to / related to the input audio signal S IN ) its low-frequency components. Usually, in a sound generation application where f UC ≧96kHz≒5×20kHz, such that f UC is much higher than the highest audible frequency of humans, due to the natural / environmental low-pass filtering effect on the plurality of air pulses (physical environment such as walls, floors, ceilings, furniture, or high propagation loss such as sound waves, and caused by the human auditory system such as the external ear canal, eardrum, malleus, incus, stapes), what the listener perceives is only the audible sound or music represented by the input audio signal S IN .

[0016] Exemplarily, FIG. 34 conceptually / schematically shows the effect of the modulation (demodulation) operation by showing the frequency spectra of the signals before and after the modulation (demodulation) operation. In FIG. 34, the modulation operation generates an amplitude-modulated ultrasonic acoustic / air wave UAW, and an input audio signal S which is an electrical (analog or digital) representation of the acoustic signal SS IN has a spectrum shown as W(f), and the spectrum of S IN / SS is represented as S(f) in FIG. 34. The synchronous demodulation operation that generates an ultrasonic pulse array UPA (including multiple pulses) having a spectrum shown as Z(f) can be regarded as shifting (including steps) the spectral components of the ultrasonic acoustic / air wave UAW by ±n×f UC (n is an integer), and the spectral components of the ultrasonic air wave UAW corresponding to the acoustic signal SS are partially carried to the baseband. Therefore, as can be seen from Z(f), the baseband components of the ultrasonic pulse array UPA are significant compared to the amplitude-modulated UAW W(f). The ultrasonic pulse array UPA propagates towards the surroundings. Due to the natural / physical environment and the inherent low-pass filtering effect of the human auditory system, the resulting spectrum Y(f) corresponding to the audio signal SS can be reproduced.

[0017] Different from the conventional DSB-SC amplitude modulation using a sine wave carrier, it is noted that W(f) has components in the harmonics of ±3×f UC , ±5×f UC , and higher-order f UC (not shown in FIG. 34). This is because the carrier wave of the modulation of the present invention is not purely a sine wave.

[0018] Referring to FIG. 1 again, as an embodiment of the synchronous demodulation operation, the demodulation unit 102 may be activated, and the aperture 112 may be formed at the time and position corresponding to / matched with the peak of the modulated air wave. In other words, when the modulated air wave reaches the peak at the position of the aperture 112, the demodulation unit 102 may be activated so that the aperture 112 also reaches the peak.

[0019] In the embodiment shown in FIG. 1, the demodulation unit 102 forms an opening 112 at the central position between the side walls 110L and 110R, which has a surface-to-surface or 111L-to-111R, and (substantially) λ UC apart only, and the tips of the flaps 101 and 103 are (substantially) λ UC / 2 away from the side walls 111L and 111R, or from the side wall surfaces 111L and 111R. Here, λ UC represents the wavelength corresponding to the ultrasonic carrier frequency f UC , that is, λ UC = C / f UC , where C is the speed of sound.

[0020] In one embodiment, the demodulation unit 102 may be operated to form the opening 112 at a valve opening speed synchronized with the ultrasonic carrier frequency f UC . In the present invention, the valve opening speed being synchronized with the ultrasonic carrier frequency f UC usually means that the valve opening speed is a value obtained by multiplying the ultrasonic carrier frequency f UC by a rational number, that is, f UC × (N / M). Here, N and M are integers. In one embodiment, the valve opening speed (of the opening 112) may be the ultrasonic carrier frequency f UC . For example, the valve / opening 112 can be opened every operating period T CY , where the operating period T CY is the reciprocal of the ultrasonic carrier frequency f UC , that is, T CY = 1 / f UC .

[0021] Also in the present invention, the modulation (demodulation) part 102 / 104 is used to indicate the modulation (demodulation) flap pair. Further, the demodulation unit (or flap pair) 102 that forms the opening 112 may be regarded as a virtual valve and perform an opening / closing motion (periodically) according to a specific valve / demodulation drive signal to form the opening 112.

[0022] In one embodiment, the modulation unit 104 may substantially generate a mode 2 (or second harmonic) resonance (or standing wave) within the resonance chamber 115, such as the pressure profile P104 and the air flow profile U104 shown in FIG. 1. In this regard, the distance between the side wall surfaces 111L and 111R is substantially determined by the full wavelength λ UC corresponding to the ultrasonic carrier frequency f UC , that is, W115≒λ UC =C / f UC . Further, in the embodiment shown in FIG. 1, the free ends of the modulation flaps 105 / 107 are arranged by the side walls 110L / 110R.

[0023] It should be noted that intermodulation (or cross-coupling) may occur between the modulation that generates the modulated air wave and the demodulation that forms the aperture 112, which may deteriorate the resulting sound quality. To improve the sound quality, it is desirable to minimize the intermodulation (or cross-coupling). To achieve this (i.e., to minimize the cross-coupling between modulation and demodulation), the modulation flaps 105 and 107 are driven to have a common-mode movement, and the demodulation flaps 101 and 103 are driven to have a differential-mode movement. That the modulation flaps 105 and 107 have a common-mode movement means that the flaps 105 and 107 are actuated / driven simultaneously and move in the same direction. That the demodulation flaps 101 and 103 have a differential-mode movement means that the flaps 101 and 103 are actuated simultaneously and move in opposite directions. Further, in one embodiment, the flaps 101 and 103 may be actuated to move in opposite directions with (substantially) the same displacement / magnitude.

[0024] The demodulation unit 102 may substantially generate a mode 1 (or first harmonic) resonance (or standing wave) within the resonance chamber 115 as the pressure profile P102 and the air flow profile U102 formed by the demodulation unit 102 shown in FIG. 1. Therefore, the demodulation unit 102 has a valve operation / drive frequency f D_V such as W115≒λ D_VIt operates in accordance with (valve / demodulation - corresponding to the drive signal). Here, λ D_V = C / f D_V where the valve operation / drive frequency is half of the ultrasonic carrier frequency Fuc, i.e., f D_V = fuc / 2.

[0025] Common - mode movement and differential - mode movement can be driven by the modulation (demodulation) drive signal. FIG. 2 shows the waveforms of the demodulation drive signals S101, S103, and the modulation drive signal SM. The modulation drive signal SM is used to drive the modulation flaps 105 and 107. The demodulation drive signals (or valve drive signals) S101, S103 are used to drive the demodulation flaps 101, 103 respectively.

[0026] In one embodiment, the modulation drive signal SM can be regarded as a pulse - amplitude - modulation (PAM) signal modulated in accordance with the input audio signal S IN . Further, different from the conventional PAM signal, the polarity of the signal SM (with respect to the constant voltage) is toggled within one operating cycle (period) T CY . Generally, the modulation drive signal SM includes pulses having an alternating polarity (with respect to the constant voltage), and the envelope / amplitude of the pulses is (substantially) the same as or proportional / corresponding to the AC (alternating - current) component of the input audio signal S IN . In other words, the modulation drive signal SM can be regarded as having a pulse - amplitude - modulation signal or having PAM - modulated pulses with alternating polarities with respect to the constant voltage. In the embodiment shown in FIG. 2, the toggling speed of the modulation drive signal SM is 2×fuc, which means that the polarity of the pulses in the modulation drive signal SM alternates / toggles twice within one operating cycle T CY .

[0027] The demodulation drive signals S101 and S103 include two drive pulses with equal amplitudes and opposite polarities (with respect to a constant / average voltage). In other words, at a specific time, if a given S101 includes a first pulse with a first polarity (with respect to a constant / average voltage) and S103 includes a second pulse with a second polarity (with respect to a constant / average voltage), the first polarity is opposite to the second polarity. As shown in Figure 2, the toggling speed of the demodulation drive signal S101 / S103 is fuc, which means that the polarity of the pulses within the demodulation drive signal S101 / S103 alternates / toggles once within one operating cycle T CY This means that within one operating cycle T, the polarity of the pulses within the demodulation drive signal S101 / S103 alternates / toggles once. Therefore, the toggling speed of the modulation drive signal (SM) is twice that of the demodulation drive signal S101 / S103.

[0028] The slopes (and associated shaded regions) of S101 / S103 are a schematic representation of energy reuse during transitions between voltage levels. It should be noted that the transition periods of signals S101 and S103 overlap. Energy reuse may be achieved by using the characteristics of an LC oscillator when most of the piezoelectric actuators of flaps 101 / R are capacitive loads. Details of the energy reuse concept may be referred to U.S. Patent No. 11,057,692, which is incorporated herein by reference. It should be noted that the piezoelectric actuator is provided as one embodiment and is not limited thereto.

[0029] To emphasize that the flap pair 102 is driven differentially, the signals S101 and S103 may be represented as -SV and +SV, which indicates that although the drive signals for this pair have the same waveform, their polarities are different. In the description, as shown in Figure 2, -SV corresponds to S101 and +SV corresponds to S103, but it is not limited thereto. In one embodiment, S101 may be +SV and S103 may be -SV.

[0030] In another embodiment, there is a DC bias voltage V BIAS and the drive signal S101 = V BIAS-SV, S103 = V BIAS In the situation like +SV, V BIAS may not be equal to 103. Such a deformed form needs to be regarded as within the scope of the present disclosure.

[0031] Furthermore, FIG. 2 shows the difference in the toggling speed between the modulation drive signal SM and the demodulation drive signal ±SV. The relative phase delay between the modulation drive signal SM and the demodulation drive signal ±SV means timing alignment, which may be adjusted according to actual requirements.

[0032] In one embodiment, the drive circuit that generates the signals SM and ±SV may have a sub-circuit configured to generate a (relative) delay between the modulation drive signal SM and the demodulation drive signal ±SV. The details of the sub-circuit that generates the delay are not limited. Known techniques can be incorporated into the sub-circuit. As long as the sub-circuit can generate a delay to meet the timing alignment requirements (detailed later), the requirements of the present invention are met, so this is within the scope of the present invention.

[0033] It is noted that the tips of the flaps 101 and 103 are substantially at the same position (the central position between the side walls 111L and 111R), and at that position, they receive substantially the same air pressure. Also, the flaps 101 and 103 move differentially. Therefore, the movement of the tips of the flaps 101 and 103 has a common-mode rejection behavior similar to that known in the field of analog differential OP amplifier circuits, which means that the displacement difference of the tips of the demodulation flaps 101 and 103, or |d 101 -d 103 | is hardly affected by the air pressure formed by the modulation flaps 105 and 107.

[0034] Common-mode rejection or modulator-demodulator separation can be explained with reference to FIG. 3. FIG. 3 shows the simulation results generated from the equivalent circuit model of the device 100. Curve d 101 and d 103represent the movement / displacement of the tips of flaps 101 and 103 respectively. As can be seen from FIG. 3, d 101 and d 103 vary quite significantly due to the sound pressure generated by the modulating flaps 105 / 107 (P104), but the differential movement represented by the curve shown by d 101 -d 103 in FIG. 3 remains (substantially) constant. That is, the width / gap of the valve opening 112 remains constant even when the modulating portion 104 is operating. In other words, the movement of the modulator has little effect on the function and characteristics of the demodulator, which is what "modulator-to-demodulator isolation" means.

[0035] On the other hand, regarding demodulator-to-modulator isolation, since flaps 101 / 103 generate a first harmonic resonance or standing wave in chamber 115, as can be seen from FIG. 1, the pressures applied to flaps 105 and flap 107 by P102 are substantially the same magnitude but have opposite polarities, causing changes of the same magnitude but opposite polarities in the movement of flaps 105 and flap 107 (by P102). This generates two ultrasonic waves that change with the same magnitude but opposite polarities (one is 105 and the other is 107). When these two ultrasonic waves propagate to the position above the valve opening 112 (indicated by the dotted line region shown in FIG. 1), they combine into one pressure. Since this "confluence" location occurs at an equal distance from the tips of 105 and 107 and at the center of device 100 along the X-axis or X-direction, the changes induced by P102 cancel / compensate each other, generating a net residual that is largely free from interference with the operation of the demodulator / virtual valve.

[0036] Exemplarily, in FIG. 4, S INis a 10-tone equal-amplitude test signal (within 650~22 kHz and having equal logarithmic scale intervals), and the simulated frequency response of the SPL (sound pressure level) measured 1 meter away from the device 100 is plotted under the condition that the equivalent circuit simulation model of the device 100 is used. In the current simulation, the ultrasonic carrier frequency is set to fuc = 192 kHz, and the valve operating frequency is f D_V = f UC / 2 = 96 kHz is set.

[0037] The demodulator-to-modulator separation can be explained by the absence of irrelevant spectral components near about 96 kHz (indicated by the block arrow in Fig. 4). This indicates a high level of separation.

[0038] As a result, the interference of the movements of these two flap pairs (101 / 103 vs. 105 / 107) is minimized through the orthogonality / arrangement of the differential mode (on the demodulator) with respect to the common mode (on the modulator).

[0039] Also, the ratio of the time the valve remains open, i.e., the duty factor, is an important factor affecting the output of the device 100. Increasing the amplitudes of the drive voltages S101 and S103 increases the amplitudes of the movements of the flaps 101 and 103, which increases the maximum opening width of the valve opening 112. By increasing the drive voltage, the duty factor of the valve opening also increases. That is, the drive voltages S101 and S103 can determine the duty factor of the valve opening 112 and the maximum opening width / gap of the valve opening 112.

[0040] As in the example shown in FIG. 5 generated from one of the aforementioned equivalent circuit simulation models, as the duty factor of the valve opening approaches 50%, each period of the valve opening, shown as a curve labeled V(open)>0, overlaps with the same half period of the ultrasonic standing wave amplitude-modulated at a position above the valve opening 112 (shown by the dotted area in FIG. 1). By synchronizing the opening and closing of the valve opening 112 with the standing wave in the chamber shown as the curve labeled V(p_vlv) in FIG. 5 and adjusting the timing, a well-formed output pressure pulse shown as the curve labeled V(ep_vlv) is generated.

[0041] In FIG. 5, the curve labeled V(d2)-V(d3) represents the difference in displacement between the flaps 101 and 103, i.e., d101 - d103, and the curve labeled V(open) represents the degree of opening of the virtual valve 112. When |V(d2)-V(d3)|>TH, V(open)>0, where TH is a threshold defined by parameters such as the thickness of the flaps 101 and 103, the width of the slit between the flaps 101 and 103, and the boundary layer thickness. The preferably shaped V(ep_vlv), unlike the highly symmetric V(p_vlv), may represent that the pulse shown by V(ep_vlv) is highly asymmetric. The asymmetry of the output pressure pulse indicates the low-frequency component (i.e., the frequency component in the audible band) of the air pulse generated by the air pulse generator, or simply the APG device, which is a desirable feature for the APG device. The higher the asymmetry, the stronger the baseband frequency component of the air pulse. The inset of FIG. 5 is shown in FIG. 6, which shows the asymmetry of V(ep_vlv) corresponding to the envelope of the 1.68 kHz baseband audio signal. In the present invention, the opening (112) is opened / formed or in an open state when the difference in displacement between the flap (101) and the flap (103) is greater than the threshold, for example, when |V(d2)-V(d3)|>TH, and is closed or in a closed state otherwise.

[0042] Furthermore, the maximum output occurs when the duty factor of the valve opening defined as |V(d2)-V(d3)|>TH is 50% or slightly greater, for example, in the range of 55 - 60%, but is not limited thereto. However, when the duty factor of the valve opening is sufficiently higher than 50%, such as 80 - 85%, more than half of the half cycle of the ultrasonic standing wave in the chamber passes through the valve, and the portions of the standing waves having different polarities cancel each other out, resulting in a lower net SPL output from the device 100. Therefore, it is generally desirable to maintain the duty factor of the valve opening close to 50%, typically in the range of 50% - 70% (a duty factor in the range between 45% and 70% is within the scope of the present invention).

[0043] In addition to the duty factor, to ensure modulator - demodulator separation, the resonance frequency f of the demodulation flaps 101 / 103 R_V is proposed to deviate sufficiently from the ultrasonic carrier frequency fuc, which is another design factor.

[0044] Under the constraint of a valve - opening duty factor equal to 50%, for any given thickness of the flaps 101 / 103, the higher the resonance - to - drive ratio (f R_V :f D_V or f R_V / f D_V ), the wider the valve can be opened, as can be observed from the equivalent - circuit simulation model. The output of the device 100 is positively related to the maximum width of the valve opening, and thus, it is desirable to make the resonance - to - drive ratio greater than 1.

[0045] However, when f R_V is within the range of f UC ±max(f SOUND ), the flaps 101 / 103 begin to resonate with the AM ultrasonic standing wave, and a part of the ultrasonic energy is converted into the common - mode deformation of the flaps 101 / 103. Here, max(f SOUND ) is the input audio signal S INmay represent the maximum frequency. Such a common mode deformation of the flap 101 / R changes the volume above the flaps 101 / 103, resulting in pressure fluctuations in the chamber 105 near the valve opening 112 over the affected frequency range and a decrease in the SPL output.

[0046] To avoid frequency response fluctuations induced by valve resonance, it is preferable to design the flaps 101 / 103 to have a resonance frequency outside the range of (f UC ±max(f SOUND ))×M. Here, M is a safety margin to cover factors such as manufacturing errors, temperature, height, etc., but is not limited thereto. As a rule of thumb, usually, f R_V ≦(f UC -20 kHz)×0.9 as in the case where f UC is significantly lower than, or f R_V ≧(f UC +20 kHz)×1.1 as in the case where f UC is significantly higher than f R_V is desirable. It should be noted that 20 kHz is often accepted as the highest audible frequency of humans, so 20 kHz is used. In applications such as HD / Hi-Res audio, 30 kHz or 40 kHz may be adopted as max(f SOUND ), and the aforementioned formula can be modified accordingly.

[0047] Also, it is assumed that w(t) and z(t) represent functions of time for amplitude-modulated ultrasonic acoustic / air waves UAW and ultrasonic pulse arrays UPA (including multiple pulses). Since the opening 112 is periodically formed at the opening ratio of the ultrasonic carrier frequency f UC , it is shown as r(t) and can be expressed as r(t)=z(t) / w(t). The function of the ratio of z(t) to w(t) can be the ultrasonic carrier frequency f UChas an aperture ratio and is periodic. In other words, z(t) can be regarded as the multiplication of w(t) and r(t) in the time domain, that is, z(t)=r(t)·w(t), and the synchronous demodulation operation performed on the UAW can be regarded as the multiplication of w(t) and r(t) in the time domain. This means that Z(f) can be regarded as the convolution of W(f) and R(f) in the frequency domain, that is, Z(f)=R(f)*W(f), where * represents the convolution operator, and the synchronous demodulation operation performed on the UAW can be regarded as the convolution of W(f) and R(f) in the frequency domain. When r(t) is periodic in the time domain with a rate of frequency f UC in the frequency domain, R(f) is discrete, and the frequency / spectral components of R(f) are spaced equidistantly by f UC It should be noted. Therefore, the convolution of W(f) with R(f), or the synchronous demodulation operation, has / includes steps of shifting W(f) (or the spectral components of the UAW) by ±n×f UC (where n is an integer). Here, r(t) / w(t) / z(t) and R(f) / W(f) / Z(f) form a Fourier transform pair.

[0048] FIG. 7 is a schematic diagram of an APG device 200 according to an embodiment of the present invention. The device 200 is similar to the device 100, and thus the same reference numerals are used. Different from the device 100, the device 200 further has an enclosure structure 14. A chamber 125 is formed between the enclosure structure 14 and the cap structure 11. The vents 113L / R are formed in the ceiling 117 located at λ UC / 4 from the side walls 111L / R, respectively, on the nodes of the ultrasonic standing pressure wave P104 as shown by the lines 135 / 137.

[0049] The purpose of vents 113L / R in FIG. 7 is to allow the air flow generated during the demodulation operation (as indicated by the curved double-headed arrows of the two dashed lines between 112L / R and 113L / R) to vent from chamber 115, thereby minimizing the difference between the average pressure inside chamber 115 and the average pressure of the surrounding outside. The function of chamber 125 is to interfere with the spectral components carried into chamber 125 by the air flow and prevent these air flows from forming additional audible acoustic signals. By placing vents 113L / R on the nodes of the steady pressure wave, the spectral components surrounding f UC are prevented from exiting chamber 115, an UPA (ultrasonic pulse array) is formed by demodulation, and the desired APPS (air pressure pulse speaker) effect is generated.

[0050] In the present invention, an APG device having an APPS effect generally means that the baseband frequency components (especially the frequency components in the audible band) embedded in the air pulses output by the APG device at the ultrasonic carrier frequency are not only observable but also have corresponding intensities. In an APG device that generates an APPS effect, the spectrum of the electrical input signal S IN will be acoustically reproduced within the baseband of the audible spectrum (low frequency compared to the carrier frequency) through the generation of a plurality of air pulses by the APG device, which is suitable for use in sound generation applications. The intensity of the baseband generated through the APPS effect is related to the amount or degree of asymmetry of the air pulses generated by the APG device. Note that the asymmetry will be described later.

[0051] The support structures 123L and 123R of the device 100 or 200 have walls that are parallel (with respect to the X-axis) and straight, and it is noted that the space / channel between 123L and 123R functions as an acoustic outlet. Simulation results using FEM (finite element method) show that when the frequency exceeds 350 kHz, a transverse standing wave along the X direction begins to form between the walls of 123L / 123R, and the output begins to self-annihilate. Such a transverse resonance-induced self-invalidation phenomenon reduces the energy transfer ratio across the height (Z direction) of the 123L-123R walls.

[0052] To avoid this problem, a horn-shaped outlet is proposed. For example, FIG. 8 is a schematic view of a part of the APG device 300 according to an embodiment of the present invention. Similar to the device 100, the device 300 has flaps 101 and 103, which are respectively fixed on the support structures 123L” and 123R” and are configured to form an opening 112 that generates a plurality of air pulses directed towards the surrounding environment through the outlet 320. Different from the support structures 123L and 123R of the device 100 having straight and parallel walls, the walls of the support structures 123L” and 123R” of the device 300 are oblique and have an angle θ that is non-perpendicular to the X-axis or the X direction. A horn-shaped outlet 320 is formed. The non-perpendicular angle θ may be designed according to actual requirements. In one embodiment, the non-perpendicular angle θ may be 54.7°, but is not limited thereto. In the present invention, a horn-shaped outlet generally refers to an outlet in which the outlet dimensions or tunnel dimensions gradually widen from the membrane structure towards the surrounding environment.

[0053] Figures 9 and 10 respectively show the frequency responses of the energy transfer ratios of apparatuses 100 and 300 for eight different displacements of flaps 101 and 103. Here, Dvv = k means that the displacement at the tip of each flap is k μM, and a differential motion of 2k μM occurs. Figures 9 and 10 are simulated using FEM. By comparing Figure 9 and Figure 10, apparatus 100 generates an energy transfer ratio that rises as the frequency exceeds 170 kHz, with several jumps and dips and starts to roll off beyond 170 kHz. On the other hand, apparatus 300 maintains an upward trend beyond about 120 kHz and generates an energy transfer ratio with a smoother frequency response for frequencies beyond 170 kHz. This means that the frequency response (beyond 170 kHz) of the energy transfer ratio of apparatus 300 is much smoother than that of apparatus 100, and this is beneficial for APG apparatuses operating at ultrasonic pulse rates (i.e., ultrasonic carrier frequencies f UC ) and their higher-order harmonics (e.g., n×f UC ). Furthermore, apparatus 300 generates an energy transfer ratio that is about five times higher than that generated by apparatus 100. Therefore, from Figures 9 and 10, it can be explained that a horn-shaped outlet brings a better energy transfer ratio for APG apparatuses.

[0054] Figure 11 shows an embodiment of a two-step etching / fabrication method for etching walls at two different angles. First, the 123R” / 123L” walls are etched at a taper angle (as shown in Figure 11(b)), and then the tapered walls are covered with photoresist or spin-on dielectric using a spray coating method (as shown in Figure 11(c)). Next, the photoresist or spin-on dielectric is patterned by photolithography (as shown in Figure 11(d)), and then the 124L and 124R walls are etched at right angles (as shown in Figure 11(e)). The manufacturing method described above is for illustrative purposes only, and the scope of the present invention is not limited thereto.

[0055] Figure 12 is a schematic diagram of an APG device 400 according to an embodiment of the present invention. The device 400 is modified from FIG. 7 of U.S. Patent Application No. 17 / 553,806 and is similar to the device 100 shown in FIG. 1 of the present invention. Different from the device 100, the device 400 includes only the flap pair 102 (not including the flap pair 104). The flap pair 102 is configured to perform both a modulation operation (forming a pneumatic fluctuation amplitude-modulated at the ultrasonic carrier frequency f UC (forming an air pressure fluctuation amplitude-modulated at the frequency f UC ), and a demodulation operation (forming an opening 112 in synchronization with the ultrasonic carrier wave amplitude-modulated at the frequency f UC and generating an air pulse according to the envelope of the amplitude-modulated ultrasonic air pressure change).

[0056] In FIG. 12, U104 and P104 represent the pressure profile and the air flow profile formed by the flap pair 102 in response to the modulation drive signal SM, and U102 and P102 represent the pressure profile and the air flow profile formed by the flap pair 102 in response to the demodulation drive signal ±SV. Here, it is emphasized that the demodulation drive signal is represented by ±SV, and the flap pair 102 is driven differentially to perform the demodulation operation (meaning that the demodulation drive signals +SV and -SV have the same magnitude but opposite polarities). For example, the above S101 and / or S103 may be represented by -SV and / or +SV.

[0057] In other words, the modulator and the demodulator are co-located in / as the flap pair 102. Similar to the device 100, the membrane structure 10 of the flap pair 102 of the device 400 is operated to have not only a common-mode movement for performing modulation but also a differential-mode movement for performing demodulation.

[0058] That is, the "modulation operation" and the "demodulation operation" are simultaneously performed by the same flap pair 102. As a result, the juxtaposition of the "modulation operation" and the "demodulation operation" is realized by a new drive signal wiring method as shown in FIG. 13. When the device 400 has the actuators 101A / 103A disposed on the flaps 101 / 103 and the actuators 101A / 103A have upper electrodes and lower electrodes, both the upper electrode and the lower electrode may receive the modulation drive signal SM and the demodulation drive signal ±SV.

[0059] In one embodiment, one electrode of the actuator 101A / 103A may receive the modulation drive signal SM in the common mode, and the other electrode may receive the demodulation drive signal S101(-SV) / S103(+SV) in the differential mode. For example, in the diagrams 431 and 433 shown in FIG. 13, the details of the region 430 shown in FIG. 12 are shown. As shown in the diagrams 431 and 432, the lower electrode of the actuator 101A / 103A receives the modulation drive signal SM in the common mode, and the upper electrode of the actuator 101A / 103A receives the demodulation drive signal S101(-SV) / S103(+SV) in the differential mode. A suitable bias voltage VBIAS may be applied to either the lower electrode (shown in diagram 432) or the upper electrode (shown in diagram 433), and the bias voltage V BIAS can be determined according to the actual requirements.

[0060] In an embodiment (shown in FIG. 433), one electrode of the actuator 101A / 103A may receive both the modulation drive signal SM in the common mode and the demodulation drive signal S101(-SV) / S103(+SV) in the differential mode, and the other electrode is appropriately biased. In the embodiment shown in diagram 433, the lower electrode receives the modulation drive signal SM in the common mode and the demodulation drive signal S101(-SV) / S103(+SV) in the differential mode, and the upper electrode is biased.

[0061] In the drive signal wiring method shown in FIG. 13, the applied signal of one actuator (e.g., 101A) is -SM - SV, or has -SM - SV, and the applied signal of the other actuator (e.g., 103A) is -SM + SV, or has -SM + SV in order to achieve the goal (V BIAS is not considered). It is noted that the drive signal wiring method may be modified or changed according to the actual situation / requirements. As long as the common mode signal component between the two applied signals applied to the flap pair 102 has the modulation drive signal SM (+V BIAS ), and the differential signal component between the two applied signals applied to the flap pair 102 has the demodulation drive signal SV, the requirements of the present invention are met and it belongs to the scope of the present invention. Here (or generally), the common mode signal component between two arbitrary signals a and b is represented as (a + b) / 2, while the differential mode signal component between two arbitrary signals a and b may be represented as (a - b) / 2.

[0062] Furthermore, in order to minimize the cross-coupling between the modulation operation (as a result of the drive signal SM) and the demodulation operation (as a result of the drive signal ±SV), in one embodiment, it is noted that the flaps 101 and 103 are configured as a mirror / symmetric pair in both their mechanical structure, dimensions, and electrical characteristics. For example, the cantilever length of the flap 101 needs to be equal to the cantilever length of 103, the film structure of the flap 101 needs to be the same as that of the flap 103, the position of the virtual valve 112 needs to be centered between the two support walls 110 of the flaps 101 and 103, or be equally spaced therefrom, the actuator pattern deposited on the flap 101 needs to mirror-project the pattern of the flap 103, and the metal wiring for the actuators deposited on the flaps 101 and 103 needs to be symmetric. Here, due to the mirror / symmetric pair (or the flaps 101 and 103 being mirror / symmetric), several items are named, but not limited thereto.

[0063] Figure 14 shows a set of frequency response measurement results of the physical embodiment of the device 400 in the IEC711 occluded ear emulator. The device 400 is driven using the driving method shown in the diagram 431. The Vrms for the modulation driving signal SM for the lower electrode is 6 Vrms, and the Vpp (peak-to-peak voltage) for the demodulation driving signal ±SV for the upper electrode is swept from 5 Vpp to 30 Vpp. The GRAS RA0401 ear simulator is used to measure the acoustic results. The operating frequency of the device 400 (i.e., the ultrasonic carrier frequency f UC ) is 160 kHz, and the device dimensions are designed accordingly (e.g., for C = 336 m / s, W115 ≒ λ UC = C / f UC ≒ 2.10 mm). As can be seen from Figure 14, the device 400 can generate sounds with high SPL in the low-frequency band (at least 99 dB for frequencies below 100 Hz).

[0064] Furthermore, Figure 15 shows an analysis of the measurement results of the device 400 shown in Figure 14. In Figure 15, the SPLs at 100 Hz (thick dashed line) and 19 Hz (thick solid line) in Figure 14 are plotted against Vvtop (Vpp), where Vvtop (Vpp) is the voltage between the peaks of the demodulation driving signal applied to the upper electrode as shown in the connection diagram 431. From Figures 14 and 15, it can be seen that as Vvtop increases, the SPL increases. Also, in the simulation results of the equivalent lumped circuit model of the device 100, it is shown that as the amplitude of the (valve driving or) demodulation driving signal increases, the SPL increases. Therefore, it can be understood that the volume of the sound generated by the air pulse generator of the present invention can be controlled via the amplitude of the demodulation driving signal.

[0065] Based on the results from FIGS. 14 and 15, the concept of modulator-demodulator co-location is verified, and it can be concluded that in the modulation (forming amplitude-modulated ultrasonic air pressure changes) and demodulation (forming openings synchronously to generate asymmetric air pulses) performed by apparatus 400, the APPS effect is preferably generated. Therefore, it may be possible to reduce the chamber width (e.g., W115 of apparatus 100).

[0066] For example, FIG. 16 is a schematic diagram of an APG apparatus 500 according to an embodiment of the present invention. Apparatus 500 is similar to apparatus 400, and the flap pair 102 is driven via one of the driving methods shown in FIG. 13, but is not limited thereto. Compared with apparatus 400, the chamber width W115' of apparatus 500 is reduced by half. In one embodiment, the chamber width W115' of apparatus 500 may be λ UC / 2.

[0067] Furthermore, a standing wave in the chamber such as 115 in FIG. 12 or 115' in FIG. 16 is not necessary, which means that there is no need for the chamber width (W115) to be λ UC or λ UC / 2 (related), and there is no need to form / maintain / reflect a plane wave between the side walls 111R / 111R' and 111L / 111L'. Changing the shape of the chamber to optimize other factors is free / flexible. For example, the length of the chamber can be reduced to increase the sound generation efficiency, which can be evaluated by the SPL per unit area (mm 2 ) of the apparatus.

[0068] Figure 17 is a schematic diagram of an APG device 600 according to an embodiment of the present invention. The device 600 may have sub-assemblies 610 and 640. In one embodiment, the sub-assemblies 610 and 640 may be manufactured via known MEMS processes and joined to each other via layer 620 using a bonding material or an adhesive material such as a dry film or other suitable die attachment material / method. The sub-assembly 610 itself can be regarded as an APG device (to be detailed later in Figure 26 and related paragraphs) including flap pair 102 or membrane structure 10. The sub-assembly 640 may be regarded as a cap structure.

[0069] Similar to device 500, device 600 has a flap pair 102 having flaps 101 and 103, which is driven via one of the driving methods shown in FIG. 13, but is not limited thereto. The flap pair 102 of the device 600 forms an amplitude-modulated ultrasonic air pressure change having an ultrasonic carrier frequency f UC to form an aperture 112 at a speed synchronized with the ultrasonic carrier frequency f UC and is operated to generate a plurality of air pulses toward the surroundings through the outlet according to the ultrasonic air pressure change.

[0070] Unlike device 500, a conduit 630 is formed in device 600. The conduit 630 connects the air volume above the virtual valve 112 (the slit between the flaps 101 and 103) to the external environment. The conduit 630 has a chamber 631, a passage 632, and an outlet 633 (or zones 631 to 633). The chamber 631 is formed between the membrane structure 10 and the cap structure (sub-assembly) 640. The passage 632 and the outlet 633 are formed in the cap structure (sub-assembly) 640.

[0071] The chamber 631 can be regarded as a semi-occluded compression chamber, and the air pressure in the compression chamber 631 may be compressed or rarified in response to the modulation drive signal SM of the common mode, generating ultrasonic air pressure changes / waves, which may be directly supplied to the passage 632 through the orifice 613. The passage 632 serves as a waveguide, and its shape and dimensions are optimized so that the pressure fluctuations / pulses generated within the zone / chamber 631 can propagate outward efficiently. The outlet 633 is configured to minimize reflection / deflection and maximize acoustic energy coupling to the surroundings. To achieve this, the tunnel dimensions (e.g., the width in the X direction) of the outlet 633 gradually widen towards the surroundings, and the outlet 633 may have a horn shape.

[0072] In one embodiment, the length / distance L of the conduit 630 between the opening 112 (equal to the flap pair 102 or the membrane structure 10) and the surface 650 630 is (substantially) a quarter wavelength λ UC corresponding to f UC / 4 (e.g., having a tolerance of ±10%). For example, when f UC = 192 kHz, L 630 may be 450 μm, but is not limited thereto. (Referring back to FIG. 16), an air pressure wave (as a type of air pressure change) propagates along the X direction within the chamber 115' of the device 500 (or the chamber 115 within the device 100), and it is noted that the distance between the virtual valve (opening) 112 and the sidewall surfaces 111L' / 111R' is λ UC / 4. In FIG. 17, the device 600 may be regarded as folding / rotating the air wave propagation path by 90° so as to be aligned with the Z direction, and the air wave or air pressure pulse is directly emitted towards the surroundings through the Z direction.

[0073] FIG. 18 shows a snapshot of the FEM simulated pressure profile of a device similar to device 600 according to an embodiment of the present invention. In FIG. 18, auxiliary arrows are presented to represent the polarity / sign of the pressure values. The difference between device 600 and the device shown in FIG. 18 is that at the interface between chamber 631 and passage 632, a chamfer 635 is added to sub-assembly 640, minimizing the disturbance of the air flow. In FIG. 18, the pressure in zone 631 is approximately +500 Pa, and the pressure in zone 632 near 633 is approximately -500 Pa. The brightest zones provide the pressure node planes.

[0074] The node plane in zone 632 shows an appropriate formation of wave propagation, and the space / distance between node plane 632 and the outer node plane of the device is approximately 1.2*λ / 2 (where λ = 346 (m / s) / 192 (kHz)), which is noted to be close to (and slightly larger than) λ / 2. This means that there is an uninterrupted pressure wave propagation at the speed of sound. In other words, as shown in FIG. 18, the pressure pulse or air wave generated by the membrane structure of device 600 is radiated towards the surroundings.

[0075] FIG. 19 shows the results of IEC711 closed-ear coupler SPL measurements for the frequency of physically implemented device 600. The results corresponding to demodulated drive signals ±SV with 20Vpp and 15Vpp are plotted. Also in Table 1, the parameters of devices 400 and 600 for generating the maximum SPL are compared.

[0076]

Table 1

[0077] Generally, the width W631 of the chamber 631 is λ UC / 2, e.g., in the example device 600, W631 ≈ 570 μM, and λ UC For zone 631 to perform chamber compression, the dimensions of chamber 631 are λ UC In one embodiment, the height H631 of the chamber 631 is λ UC / 5, i.e., H631<λ UC It is noted that the width (i.e., the dimension in the X direction) of the chamber 631 may be narrowed in a stepped or tapered manner from the membrane structure 10 toward the passage 632. Both of these cases are within the scope of the present invention.

[0078] 20 is a schematic diagram of an APG device 700 according to one embodiment of the present invention. Similar to device 600, device 700 has subassemblies 710 and 740 with a conduit 730 formed therein. Subassembly 710 may be fabricated by a MEMS process and may be considered an APG device. A chamber 705 is formed within subassembly 710. Subassembly 710 may itself be an APG device and may be viewed as a combination of the squeeze mode operation disclosed in U.S. Pat. No. 11,172,310, the virtual valve disclosed in U.S. Pat. No. 11,043,197, and the actuation scheme shown in FIG. 13, where U.S. Pat. Nos. 11,172,310 and 11,043,197 are incorporated herein by reference.

[0079] The conduit 730 has a chamber 731, a passage / waveguide 732, and a horn-shaped outlet 733 (or zones 731 to 733), and connects the air volume below the virtual valve 112 to the outer ambient atmosphere. Different from the device 600, the sub-assembly 740 may be formed / manufactured via techniques such as 3D printing, precision injection molding, stamping, etc. The passage / waveguide 732 has a first section that is an orifice 713 etched on the cap of the sub-assembly 740 and a second section formed within the sub-assembly 710, and a chamfer 735 may be added between them to minimize disturbances. Chambers 705 and 731 overlap. The pressure fluctuations / waves generated by the flaps 101 and 103 are directly supplied to the passage / waveguide 732.

[0080] Figure 21 is a schematic diagram of an APG device 800 according to an embodiment of the present invention. The device 800 has sub-assemblies 810 and 840. The sub-assembly 810 may have the same or a similar structure as the device 500, which can be manufactured by a MEMS process, can be regarded as an APG device, and may have flaps 101 and 103 driven by one of the methods shown in FIG. 13. A virtual valve (opening) 112 is formed. The sub-assembly 840 may be formed / manufactured via techniques such as 3D printing, precision injection molding, and precision stamping. It should be noted that the sub-assembly 810 generates a plurality of air flow pulses via a modulation (demodulation) operation.

[0081] A conduit 830 that connects the air volume below the virtual valve 112 to the surrounding external environment is formed within the device 840. The conduit 830 has a (compression) chamber 831, a passage / waveguide 832, and a horn-shaped outlet 833 (or zones 631 to 633). The compression chamber 831 is configured to convert a plurality of air flow pulses into a plurality of air pressure pulses. Specifically, the chamber 831 generates a pressure pulse ΔP n ∝P 0_n ·ΔM n / M 0_n (Equation 1), where M 0_nis the air mass in chamber 831 before the start of pulse cycle n, ΔM n is the air mass associated with the air flow pulse of pulse cycle n. Equation 1 represents converting the air flow pulse into an air pressure pulse, and the converted air pressure pulse propagates within passage / waveguide 832. In one embodiment, sub-assembly 840 within zone 831 may have a cross-sectional profile in the shape of a genuine brass mouthpiece.

[0082] Passage / waveguide 832 has an impedance that is close to, matched to, or within ±15% of that of compression chamber 831, and the outward propagation efficiency towards the periphery of the pressure pulse generated within zone 831 may be maximized. In one embodiment, the propagation efficiency may be optimized by appropriately selecting the cross-sectional area of passage 832.

[0083] In the embodiment shown in FIG. 21, the tunnel dimensions (e.g., width in the X direction) of outlet 833 gradually widen towards the periphery according to a piece-wise linear pattern (where θ1 < θ2) such that a horn shape is formed. It is noted that the horn shape of the outlet may be designed according to actual specifications. The tunnel dimensions of the outlet can be widened according to a polynomial pattern, a pure linear pattern, a piece-wise linear pattern, a parabolic pattern, an exponential function pattern, a hyperbolic pattern, etc., but are not limited thereto. As long as the tunnel dimensions of the outlet gradually widen towards the periphery, the specifications of the present invention are met and this belongs to the scope of the present invention.

[0084] To perform chamber compression in zone 831, the dimensions of chamber / zone 831 are proposed to be made sufficiently smaller than the wavelength λ UC corresponding to the operating frequency f UC For example, in an embodiment where f UC = 160 kHz and λ UC = (346 / 160) = 2.16 mm, the height H 831 may be in the range of λ UC / 10 to λ UC / 60 (e.g., H 831 = λ UC / 35 = 62 μm), and the width W815 is in the range of λ UC / 5 to λ UC / 30 (for example, W in the range of 115 μm to 350 μm 815 ), but is not limited thereto.

[0085] The membrane structure 10 sub-divides the volume of the space into a resonance chamber 805 on one side and a compression chamber 831 on the other side. Due to the nature of this sub-division, it is noted that the displacements due to the common mode movement of the flaps 101 and 103, as observed from the spaces of the chambers 805 and 831, are of the same magnitude but opposite in direction / polarity. In other words, together with the common mode movement of the flaps 101 and 103, a push-pull operation is formed, and such a push-pull operation increases the pressure difference across the flaps 101 and 103 (for example, doubles it), and thus, when the virtual valve 112 is opened, the air flow is enhanced.

[0086] Specifically, for the compression chamber 831 having a volume V1 and the resonance chamber 805 having a volume V2, the movement of the membrane / flap that brings about a volume difference DV (assuming DV << V1, V2) causes a pressure change at V1 of ΔP V1 = 1 - V1 / (V1 - DV) = -DV / (V1 - DV) ≈ -DV / V1, and a pressure change at V2 of ΔP V2 = 1 - V2 / (V2 + V) = DV / (V2 + DV) ≈ DV / V2. The pressure difference between the two volumes may be ΔP V2 -ΔP V1 = DV / (V2 + DV) + DV / (V1 - DV). When V1 ≈ V2 ≈ Va, ΔP V2 -ΔP V1 ≈ DV / (Va + DV) + DV / (Va - DV) = DV·2Va / (Va 2 - DV 2 ) ≈ 2·DV / Va ≈ 2·ΔP V2 which means that the push-pull operation can double the pressure difference between the two sub-spaces separated by the flaps 101 and 103.

[0087] FIG. 22 is a schematic diagram of an APG device 900 according to an embodiment of the present invention. The device 900 has sub-assemblies 910 and 940. The sub-assembly 910 may be manufactured by a MEMS process and may be regarded as an APG device. The sub-assembly 940 may be manufactured by 3D printing. Also, similar to the device 700 or the sub-assembly 710, the sub-assembly 940 may be regarded as a combination of the squeezing mode operation disclosed in U.S. Patent No. 11,172,310, the virtual valve disclosed in U.S. Patent No. 11,043,197, and the driving method shown in FIG. 13. In the device 900, the squeezing mode operation chamber 905 and the compression chamber 931 are separated, while in the device 700, the squeezing mode operation chamber and the compression chamber are integrated as the chamber 731.

[0088] The effects of the sub-assembly 810 and the sub-assembly 910 are similar to those in the case of air flow pulse generation, but their operating principles are different. The sub-assembly 810 utilizes resonance, while the assembly 910 utilizes the compression and rarefaction of the squeezing mode operation chamber 905 caused by the movement of the membranes (flaps 101, 103). Therefore, the chamber width W 905 no longer needs to satisfy any relationship with λ UC and thus the size of the chamber 905 may be reduced to a practical / desired extent.

[0089] FIG. 23 is a schematic diagram of an APG device A00 according to an embodiment of the present invention. Since resonance is not required, restrictions on the rectangular cross-section of the chamber, such as the chamber 905, can be eliminated, and the shape for optimizing the generation of pressure waves or the propagation of waves to the surroundings becomes more flexible. For example, the chamber A05 or the sub-assembly A40 may have a cross-section in the shape of a brass mouthpiece.

[0090] Another aspect of the apparatus A00 of FIG. 23 is “direct pressure coupling”. Instead of first passing through the orifice 913 as in the case of the apparatus 900, the pressure wave generated within the compression chamber A05 of the apparatus A00 is directly coupled to the conduit A32 and then released to the surroundings via the outlet A33. Such a direct coupling between the compression chamber and the conduit / outlet eliminates the losses caused by the orifice 913, resulting in a significant efficiency improvement over the apparatus 900.

[0091] FIG. 24 is a schematic diagram of an APG apparatus B00 according to an embodiment of the present invention. The apparatus B00 is similar to the apparatus A00. Unlike the apparatus A00, the apparatus B00 further has a (cap) structure B11, and a chamber B05 is formed between the cap structure B11 and the membrane structure 10. A push-pull operation may be performed using the chamber A05 formed on one side of the membrane structure 10 and the chamber B05 formed on the other side of the membrane structure 10, whereby the air flow pulse may be strengthened.

[0092] It should be noted that the air pulses generated by the sub-assemblies 810 and 910 may be regarded as air flow pulses, and the sub-assemblies 840 and 940 may be regarded as air flow-air pressure converters having a trumpet-shaped cross-sectional profile. On the other hand, the air pulses generated by the sub-assemblies 610, 710, A10, and B10 may be regarded as air pressure pulses, whereby a demodulated / asymmetric air pressure pulse may be directly formed and may be more efficient than the apparatuses 800 and 900.

[0093] Also, a sub-assembly having a conduit formed therein or a sub-assembly having a conduit with a trumpet-shaped cross-sectional profile, although not limited thereto, may be applied to the APG apparatuses disclosed in U.S. Pat. Nos. 10,425,732, 11,172,310, etc., filed by the present applicant, or other apparatuses such as U.S. Pat. No. 8,861,752.

[0094] FIG. 25 shows a diagram of the timing alignment of the opening of the virtual valve (VV) 112 of the APG apparatus of the present invention. In FIG. 25, the solid curve represents the flap common mode movement generated by the modulation drive signal SM, and the darkness of the background represents the acoustic resistance corresponding to the virtual valve. A darker shade means a higher resistance (VV closed, the volume in the chamber is cut off from the surroundings), and a lighter shade means a lower resistance (VV open, the volume in the chamber is connected to the surroundings).

[0095] In FIG. 25(a), the timing of the open state of the virtual valve (VV) 112 is adjusted so that the maximum value (first peak) of the pressure in the chamber is achieved, which typically occurs just before the flaps reach their most positive (first peak) common mode displacement. On the other hand, the timing of the closed state of the virtual valve 112 is adjusted so that the minimum value (second peak) of the pressure in the chamber is achieved, which typically occurs just before the flaps reach their most negative (second peak) common mode displacement. The timing alignment shown in FIG. 25(a), where the maximum opening of VV112 is aligned with the first peak of the pressure in the chamber, is for maximizing the pulse amplitude of the air flow pulse, and this may be suitable for apparatuses 100 to 500 (having a chamber but no formed conduit).

[0096] On the other hand, in FIG. 25(b), the timing of the open state of the virtual valve 112 is aligned with the maximum speed of the common mode movement of the membrane (flap) moving in the first direction, as suggested by the valve timing of a gas / piston engine in the automotive industry, while the timing of the closed state of the virtual valve 112 is aligned with the maximum speed of the common mode movement of the membrane (flap) moving in the second direction opposite to the first direction. The first direction is the direction from the membrane structure towards the surroundings. The timing alignment shown in FIG. 25(b) is for maximizing the volume of the air flow pulse, and this may be suitable for apparatuses 600, or apparatuses 700 to 900, A00, and B00 (chambers having conduits formed therein).

[0097] Figure 26 is a schematic diagram of an APG device C00 according to an embodiment of the present invention. The device C00 is similar to the APG device shown above and has flaps 101 and 103. The flaps 101 and 103 may be driven by the driving method shown in FIG. 13.

[0098] Unlike these devices, the device C00 does not have a cap structure. Compared with the aforementioned APG device, the device C00 has a simpler structure, is sufficient with fewer photolithography etching steps, eliminates complex conduit manufacturing steps, and avoids the need to couple two sub-members or sub-assemblies together. The manufacturing cost of the device C00 is significantly reduced.

[0099] Since no chamber is formed under the compressed cap structure, the sound pressure generated by the device C00 mainly results from the acceleration of the movement of the flaps (101 and 103). By matching the timing of the opening of the virtual valve 112 (responding to the demodulated drive signal ±SV) with the timing of the acceleration of the common-mode movement of the flaps 101 and 103 (responding to the modulated drive signal SM), the device C00 can generate an asymmetric air (pressure) pulse.

[0100] It is noted that the space surrounding flaps 101 and 103 is divided into two sub - spaces. One is the Z>0, i.e., +Z sub - space, and one is the Z<0, i.e., -Z sub - space. For any common - mode movement of flaps 101 and 103, a pair of acoustic pressure waves is generated, one in the +Z sub - space and one in the -Z sub - space. These two acoustic pressure waves are of the same magnitude but opposite polarities. As a result, when the virtual valve 112 is opened, the pressure difference between the two air volumes in the vicinity of the virtual valve 112 is neutralized with each other. Therefore, when the timing of the differential - mode movement reaches its peak, i.e., the timing when the timing VV112 reaches its maximum opening, it is synchronized with the timing of the acceleration of the common - mode movement that reaches its peak, and the sound pressure that should be generated by the common - mode movement is suppressed / removed by the opening of the virtual valve 112, and automatic neutralization occurs between the two sound pressures on the two opposing sides of flaps 101 and 103. Here, the two sound pressures are of the same magnitude but have opposite polarities. This means that when the virtual valve 112 is opened, the device C00 generates (substantially) zero net air pressure. Therefore, if the opening period of the virtual valve 112 overlaps with one of the (two) polarities of the acceleration of the common - mode flap movement, the device C00 can generate an asymmetrically high, single - ended (SE) or SE - like air - pressure waveform / pulse.

[0101] In the present invention, an SE (like) waveform may mean being (substantially) unipolar with respect to a certain level. The SE acoustic pressure wave may represent a waveform that is (substantially) unipolar with respect to the ambient pressure (e.g., 1 ATM).

[0102] FIG. 27 shows a diagram of the timing alignment of the opening of a virtual valve (VV) according to an embodiment of the present invention. The timing alignment method shown in FIG. 27 may be applied to the apparatus C00. In FIG. 27(a), the solid / dashed / dotted curves represent the displacement / velocity / acceleration of the common mode movement of the membranes (flaps 101 and 103) in response to the modulation drive signal SM. Similar to FIG. 25, the darkness of the background represents the acoustic resistance generated by the opening / closing operation of VV112. For the sake of explanation, the waveform of the membrane / flap movement in FIG. 27(a) is assumed (or approximately plotted) to be a sine wave with a constant amplitude, and the velocity / acceleration waveforms are the first / second derivatives of the displacement waveform. As shown in FIG. 27(a), the timing of the peak VV opening is aligned with the timing of the first peak acceleration of the common mode membrane / flap movement in the first direction, as described above. As a result of such timing alignment, automatic neutralization occurs between the two acoustic pressure waves generated in the subspaces +Z and -Z, and the net acoustic pressure is suppressed, as shown by the flat portion of the SE air pressure waveform in FIG. 27(b).

[0103] Also, as shown in FIG. 27(a), the timing at which the VV is closed is aligned with the timing of the second peak acceleration of the common mode membrane / flap movement in the second direction, where the second direction is opposite to the first direction. Since the VV is closed during / near the second peak acceleration, the sound pressure generated by the second peak acceleration of the flaps 101 and 103 can radiate away from the flaps 101 and 103, resulting in a highly asymmetric sound pressure wave, as shown by the sine half-wave portion of the SE air pressure waveform in FIG. 27(b).

[0104] It should be noted that the opening of the virtual valve 112 does not determine the intensity / amplitude of the acoustic pressure pulse, but rather determines how strong the effect of "pressure close to zero net" (or automatic neutralization) is. When the opening of the virtual valve 112 is wide, the effect of "zero net pressure" is strong, automatic neutralization is complete, the asymmetry becomes strong / clear, and a strong / distinct baseband signal or APPS effect is obtained. Conversely, when the opening of the virtual valve 112 is narrow, the effect of "zero net pressure" is weak, automatic neutralization is incomplete, and as a result of the reduced asymmetry, a weak baseband signal or APPS effect occurs.

[0105] In the FEM simulation, the device C00 can generate an SPL of 145 dB at 20 Hz. From the FEM simulation, even when the SPL generated by the device C00 is approximately 12 dB lower than the SPL (an SPL of approximately 157 dB at 20 Hz) generated by the device 600, it is observed that under the same driving conditions, the THD (total harmonic distortion) of the device C00 is 10 - 20 dB lower than the THD of the device 600. Therefore, in the simulation, the effectiveness of the device C00, an APG device without a cap structure, or an APG device without a chamber formed therein is verified.

[0106] It is noted that the description of the timing of the VV opening that is synchronized with the peak pressure in the chamber or the timing of the peak velocity / acceleration of the common - mode membrane movement implicitly suggests that an error of ±e% is acceptable. That is, the timing of the VV opening being synchronized with (1 ± e%) of the peak pressure in the chamber or the peak velocity / acceleration of the common - mode membrane movement is also within the scope of the present invention, where e% may be 1%, 5%, or 10% depending on the actual specifications.

[0107] Regarding pulse asymmetry, in FIG. 28, there are full - cycle pulses with different degrees of asymmetry (1 operating cycle T CYis shown. In the present invention, the degree of asymmetry may be evaluated by the ratio of p2 to p1. Here, p1 > p2, where p1 represents the peak value of the first half-cycle pulse having the first polarity with respect to the level, and p2 represents the peak value of the second half-cycle pulse having the second polarity with respect to the level. In the acoustic region, the level may correspond to ambient conditions such as ambient pressure (0 sound pressure) or 0 acoustic air flow, and the air pulse in the present invention may represent either an air flow pulse or an air pressure pulse.

[0108] FIG. 28(a) shows a full-cycle pulse where r = p2 / p1 > 80%. The full-cycle pulse shown in FIG. 28(a) or the full-cycle pulse where r = p2 / p1 ≒ 1 has little asymmetry. FIG. 28(b) shows a full-cycle pulse where 40% ≦ r = p2 / p1 ≦ 60%. The full-cycle pulse shown in FIG. 28(b), or the full-cycle pulse where r = p2 / p1 ≒ 50%, has a median value of asymmetry. FIG. 28(c) shows a full-cycle pulse where r = p2 / p1 < 30%. The full-cycle pulse shown in FIG. 28(c), or the full-cycle pulse where r = p2 / p1 → 0, has a high degree of asymmetry.

[0109] As described above, the higher the degree of asymmetry, the stronger the APPS effect and the baseband spectral component of the ultrasonic air pulse. In the present invention, the asymmetric air pulse represents at least an air pulse having a median value of asymmetry, meaning r = p2 / p1 ≦ 60%.

[0110] It should be noted that the demodulation operation of the APG device of the present invention is to generate an asymmetric air pulse according to the amplitude of the ultrasonic air pressure change generated by the modulation operation. From one perspective, the demodulation operation of the present invention is similar to the rectifier in an AM (amplitude modulation) envelope detector in a wireless communication system.

[0111] In a wireless communication system, as is well known, an envelope detector, which is a type of wireless AM (non-coherent) demodulator, has a rectifier and a low-pass filter. The envelope detector generates an envelope corresponding to its input amplitude-modulated signal. The input amplitude-modulated signal of the envelope detector usually has high symmetry, with r = p2 / p1 → 1. One goal of the rectifier is to convert the symmetric amplitude-modulated signal so that the rectified amplitude-modulated signal becomes highly asymmetric with r = p2 / p1 → 0. After low-pass filtering the highly asymmetric rectified AM signal, the envelope corresponding to the amplitude-modulated signal is recovered.

[0112] The demodulation operation of the present invention, which changes symmetric ultrasonic air pressure changes (r = p2 / p1 → 1) to asymmetric air pulses (r = p2 / p1 → 0), is similar to the rectifier of an envelope detector as an AM demodulator, and the low-pass filtering operation is left to the natural environment and the human auditory system (or a sound sensing device such as a microphone), and the sound / music corresponding to the input audio signal S IN can be recovered, perceived by a listener, or measured by a sound sensing device.

[0113] For the demodulation operation of the APG device, generating asymmetry is important. In the present invention, the pulse asymmetry depends on the appropriate timing of opening, which is coordinated with the movement of the membrane (flap) that generates the ultrasonic air pressure change. As shown in FIGS. 25 and 27, different APG configurations have different ways of timing coordination. In other words, the timing of forming the opening 112 is specified so that a plurality of air pulses generated by the APG device become asymmetric.

[0114] The APG device that generates asymmetric air pulses may also be applied to air pump / movement applications that may have functions such as cooling, drying, or other functions.

[0115] Furthermore, power consumption can be reduced by appropriate cell and signal path arrangements. For example, FIG. 29 shows a top view of the APG device D00 according to an embodiment of the present invention, and FIG. 30 shows a cross-sectional view of the device D00 along the line A-A' shown in FIG. 29. The device D00 has cells D01 to D08 arranged in an array. Each cell (D0x) may be one of the aforementioned APG devices (e.g., 400 to C00). In FIG. 30, the subassembly having the cap structure and the conduits formed therein is omitted for simplicity. All flaps within the device D00 are driven by the drive signal scheme 431, the upper electrode receives either the signal +SV or the signal -SV, and the lower electrode is assumed to receive SM-V BIAS is received.

[0116] In FIG. 29, the elongated rectangle along the Y direction represents the upper electrode of the flap or the actuator disposed on the flap. The background shading may represent the lower electrode of the actuator or that the lower electrode of the actuator is electrically connected.

[0117] In the device D00, the flap (e.g., 101) receiving the signal -SV and the flap (e.g., 103) receiving the signal +SV are spatially interleaved. For example, when the flap 103 of the cell D01 receives the signal +SV, it is proposed that the flap 101 of the cell D02 receives the signal -SV. This is because when the polarities of the signals +SV and -SV toggle or during the transition period of the signals +SV and -SV, the (discharge) charging current of the capacitive load flows in the X direction through the lower electrode, and the effective resistance R BT,P (where P represents the flow of the parallel current) is low due to L / W ≪ 1, and the power consumption of the device D00 is reduced. Here, L / W represents the channel length / width from the perspective of the (discharge) charging current.

[0118] On one hand, when the drive signals -SV and +SV are wired in the patterns of {+SV, -SV}, {-SV, +SV}, {+SV, -SV}, {-SV, +SV}, {+SV, -SV}, {-SV, +SV}, {+SV, -SV}, {-SV, +SV} (not shown in Fig. 29), where {…, …} indicates a pair of differential drive signals of one cell D0x, the load (discharge) charging current is in the Y direction, and the effective resistance R of the lower electrode BT,S (where S represents the flow of the series current) becomes large enough (i.e., because L / W ≫ 1, R BT,S ≫ R BT,P ), the power consumption in such a method becomes higher.

[0119] In other words, by using the wiring method shown in Fig. 29 (for example, when cells D01 and D02 are adopted), the flap 103 of cell D01 that receives the signal +SV is spatially arranged adjacent to the flap 101 of cell D02 that receives the signal -SV. When the transition periods of the signals ±SV overlap in time, the current from the lower electrode of one flap (for example, 103 of D01) does not need the device D00 to move away from the pad and then re-enter the device D00 from another pad, but directly moves to the adjacent flap (for example, 101 of D02). Therefore, the effective resistance of the lower electrode is significantly reduced, and the power consumption is also reduced.

[0120] Also, the operating frequency may be increased by incorporating a plurality (for example, two) of cells. Specifically, the air pressure pulse speaker (APPS) sound generation method using the APG device of the present invention is a kind of discrete-time sampling system. On one hand, usually, in order to achieve high fidelity, it is desirable to increase the sampling rate in such a sampled system. On the other hand, in order to reduce the required drive voltage and power consumption, it is desirable to reduce the operating frequency of the device.

[0121] Instead of increasing the operating frequency as the sampling rate of one APG device, it is efficient to achieve a high pulse / operating rate by temporally and spatially interleaving (at least) two groups (subsystems) with low pulse / operating rates.

[0122] FIG. 31 (showing a spatial arrangement) is a top view of an APG device E00 according to an embodiment of the present invention. The device E00 has two cells E11 and E12 arranged adjacent to / next to each other. The cell E11 / E12 may be one of the APG devices of the present invention.

[0123] FIG. 32 (showing a temporal relationship) shows the waveforms of two sets of (demodulation) modulation drive signals A and B intended for cells E11 and E12. The set A includes demodulation drive signals ±SV and modulation drive signal SM, and the set B includes demodulation drive signals ±SV' and modulation drive signal SM'. As shown in the embodiment of FIG. 32, the demodulation drive signals +SV' / −SV' of the signal set B are a delayed version of the demodulation drive signals +SV / −SV of the signal set A. Further, the signals +SV' / −SV' of the signal set B are the signals +SV / −SV of the signal set A delayed by T CY / 2, where T CY =1 / f UC and f UC represents the operating frequency of the cells E11 / E12. The modulation drive signal SM' of the set B may be regarded as an inversion or a polarity-inverted version of the modulation drive signal SM of the set A. The signals SM and SM' may have a relationship of SM' = -SM or SM + SM' = C, where C is a certain constant or bias. For example, if the modulation drive signal SM of the set A has negative-polarity pulses with respect to a voltage level (shown by a dashed line in FIG. 32) within a period T 22 the modulation drive signal SM' of the set B has positive-polarity pulses with respect to a voltage level (shown by a dashed line in FIG. 32) within a period T 22 .

[0124] By providing one of Group A and Group B to cell E11 and the other of Group A and Group B to cell E12, apparatus E00 may generate a pulse array having a pulse / sampling rate of 2×f UC where f UC is the operating frequency of each cell.

[0125] Figure 33 is a top view of an APG apparatus F00 according to an embodiment of the present invention. Apparatus F00 has cells F11, F12, F21, and F22 arranged in a 2×2 array. The cells in apparatus F00 may be one of the APG apparatuses of the present invention. Two of cells F11, F12, F21, and F22 may receive signal group A, and the other two cells may receive signal group B.

[0126] In one embodiment, cells F11, F22 receive signal group A, and cells F21, F22 receive signal group B. In one embodiment, cells F11, F22 receive signal group A, and cells F12, F22 receive signal group B. In one embodiment, cells F11, F21 receive signal group A, and cells F12, F22 receive signal group B. Similar to apparatus E00, the apparatus also generates a pulse array having a pulse / sampling rate of 2×f UC where f

[0127] It is noted that conventional speakers (e.g., dynamic drivers) that generate acoustic waves using physical surface motion face the problem of cancellation of front / back radiation waves. When a physical surface moves and causes a movement of the air mass, a pair of sound waves, i.e., a front radiation wave and a back radiation wave, is generated. The two sound waves cancel out most of each other, and the net SPL is greatly reduced compared to when the front / back radiation waves are measured alone.

[0128] A widely adopted solution to the problem of cancellation of front / back radiation waves is to utilize either a rear enclosure or an open baffle. Both solutions require a physical size / dimension corresponding to a wavelength of the lowest frequency of interest, e.g., a wavelength of 1.5 meters for a frequency of 230 Hz.

[0129] Compared with conventional speakers, the APG device of the present invention occupies only a few tens of square millimeters (much smaller than conventional speakers) and generates a large SPL, especially at low frequencies.

[0130] This is achieved by generating an asymmetric amplitude-modulated air pulse. The modulation part generates air pressure fluctuations of symmetric amplitude modulation through membrane movement, and the demodulation part generates an asymmetric amplitude-modulated air pulse through a virtual valve. The modulation part and the demodulation part are realized by a pair of flaps manufactured in the same processing layer, reducing the complexity of manufacturing / production. The modulation operation is executed through the common-mode movement of the flap pair, and the demodulation operation is executed through the differential-mode movement of the flap pair. The modulation operation (through the common-mode movement) and the demodulation operation (through the differential-mode movement) may be implemented by a single flap pair. Appropriate timing alignment between the differential-mode movement and the common-mode movement enhances the asymmetry of the output air pulse. Furthermore, a horn-shaped outlet or a trumpet-shaped conduit helps improve the propagation efficiency.

[0131] In summary, the air pulse generation device of the present invention has modulation means and demodulation means. The modulation means, which can be realized by applying a modulation drive signal to the flap pair (102 or 104), generates an amplitude-modulated ultrasonic acoustic / air wave having an ultrasonic carrier frequency by an acoustic signal. The demodulation means may be realized by applying a pair of demodulation drive signals +SV and -SV to the flap pair (102) or by periodically driving the flap pair (102) to form an aperture (112), and a synchronous demodulation operation is performed to shift the spectral components of the ultrasonic acoustic / air wave UAW by ±n×f UC only. As a result, the spectral components of the ultrasonic air wave corresponding to the acoustic signal are shifted to the audible baseband, and the acoustic signal is reproduced.

[0132] It will be readily understood by those skilled in the art that many modifications and variations of the apparatus and method can be made while retaining the teachings of the present invention. Therefore, the foregoing disclosure should be construed as being limited only by the boundaries of the appended claims.

[0133] (Known Documents) 1 US Patent Application Publication 2022-0225032A1 2 US Patent Application Publication 2022-0224999A1 3 US Patent 10,425,732B1 4 US Patent 7,736,324B1 5 US Patent 5,109,948A 6 US Patent 4,942,939 7 US Patent 4,646,733 8 US Patent Application Publication 2022 / 0047841A1 9 US Patent Application Publication 2014 / 0064036A1 10 US Patent Application Publication 2013 / 0279738A1 11 US Patent Application Publication 2012 / 0018244A1 12 US Patent Application Publication 2008 / 0121220A1 13 US Patent Application Publication 2005 / 0235988A1 14 US Patent Application Publication 2004 / 0024455A1 15 International Publication 2016 / 202790A2 16 US Patent Application Publication 2020 / 0059719A1 17 US Patent Application Publication 2019 / 0116417A1 18 US Patent Application Publication 2018 / 0179048A1 19 US Patent Application Publication 2016 / 0366521A1 20 US Patent Application Publication 2012 / 0032892A1 21 US Patent Application Publication 2017 / 0201192A1 22 Korean Patent Application Publication 10-2019-0116898A 23 Korean Patent 10-1901204B1 Korean Patent Application Publication No. 10-2019-0043489A Korean Patent No. 10-2093804B1 Japanese Patent Application Publication No. 2022-160366A US Patent Application Publication No. 2019 / 0238974A1 US Patent Application Publication No. 2017 / 0041708A1 US Patent Application Publication No. 2014 / 0084396A1 US Patent Application Publication No. 2014 / 0341394A1 US Patent Application Publication No. 2016 / 0059206A1 US Patent Application Publication No. 2012 / 0081337A1 US Patent No. 5,611,406A US Patent Application Publication No. 2019 / 0020944A1 Japanese Patent Application Publication No. 2022-160367A Japanese Patent Application Publication No. 2022-160368A

Explanation of Reference Numerals

[0134] 100 Air Pulse Generation (APG) Device 12 Device Layer 11 Chamber Definition Layer 124L, 124R Walls 101, 103, 105, 107 Flaps 123R, 123L Support Structures

Claims

1. An air pulse generator, having a membrane structure including a flap pair, wherein the membrane structure is operated to perform a common mode movement of the flap pair such that an amplitude-modulated ultrasonic air pressure change having an ultrasonic carrier frequency is formed, and an amplitude of the ultrasonic air pressure change is modulated according to an input audio signal, wherein the membrane structure is operated to perform a differential mode movement of the flap pair such that an aperture is formed at a speed synchronized with the ultrasonic carrier frequency, and the air pulse generator generates a plurality of air pulses according to the amplitude-modulated ultrasonic air pressure change. Air pulse generator.

2. The air pulse generator according to claim 1, wherein the plurality of air pulses are asymmetric.

3. The air pulse generator according to claim 1, wherein the plurality of air pulses are amplitude-modulated according to an input audio signal.

4. The flap pair is driven to perform the common mode movement such that the ultrasonic air pressure change having the ultrasonic carrier frequency is formed, the flap pair is driven to perform the differential mode movement such that the aperture is formed at a speed synchronized with the ultrasonic carrier frequency, and a plurality of air pulses are generated according to the ultrasonic air pressure change. The air pulse generator according to claim 1.

5. The flap pair has a first flap and a second flap, and the first flap and the second flap are formed as a symmetric pair. The air pulse generator according to claim 1.

6. The air pulse generator according to claim 1, wherein a duty factor for forming the aperture is in a range between 45% and 70%.

7. The air pulse generator according to claim 1, wherein a ratio of a resonance frequency of the flap pair to a drive frequency corresponding to a drive signal for causing the flap pair to perform the differential mode movement is greater than 1.

8. The air pulse generator according to claim 1, wherein the resonance frequency of the flap pair is greater than a frequency obtained by adding a maximum frequency of an input audio signal to the ultrasonic carrier frequency.

9. The air pulse generator according to claim 1, wherein the resonance frequency of the flap pair is less than a frequency obtained by subtracting a maximum frequency of an input audio signal from the ultrasonic carrier frequency.

10. The flap pair has a first flap and a second flap, and the first flap is driven by a demodulated drive signal. The air pulse generator according to claim 1, wherein the volume of the sound generated by the air pulse generator is controlled via the amplitude of the demodulated drive signal.

11. Further having a cap structure, A first chamber is formed between the membrane structure and the cap structure, The membrane structure is operated to perform the common mode movement such that the amplitude-modulated ultrasonic air pressure change having the ultrasonic carrier frequency is formed in the first chamber. The membrane structure is operated to perform the differential mode movement such that the opening is formed at a speed synchronized with the ultrasonic carrier frequency. The air pulse generator according to claim 1, wherein the air pulse generator generates the plurality of air pulses according to the amplitude-modulated ultrasonic air pressure change in the first chamber.

12. The cap structure has a first side wall and a second side wall, The air pulse generator according to claim 11, wherein the distance between the first side wall and the second side wall is a wavelength corresponding to the ultrasonic carrier frequency.

13. The cap structure has a first side wall and a second side wall, The air pulse generator according to claim 11, wherein the flap pair forms the opening at a central position between the first side wall and the second side wall.

14. The cap structure has a side wall, The opening is formed at a position half a wavelength away from the side wall, The air pulse generator according to claim 11, wherein the half wavelength corresponds to the ultrasonic carrier frequency.

15. The cap structure has a side wall and a ceiling, A vent is formed in the ceiling, The vent is a quarter wavelength away from the side wall, The air pulse generator according to claim 11, wherein the quarter wavelength corresponds to the ultrasonic carrier frequency.

16. The cap structure has a ceiling portion, A first vent and a second vent are formed in the ceiling portion, The air pulse generator according to claim 11, wherein the distance between the first vent and the second vent is a half wavelength corresponding to the ultrasonic carrier frequency.

17. Further having an enclosure structure, The cap structure has a ceiling, The air pulse generator according to claim 11, wherein a second chamber is formed between the enclosure structure and the ceiling.

18. The air pulse generator according to claim 17, wherein at least one outlet is formed in the enclosure structure.

19. The membrane structure has a first flap pair and a second flap pair, the first flap pair is driven to perform the differential mode movement, the opening is formed at a speed synchronized with the ultrasonic carrier frequency, and the plurality of air pulses are generated according to the ultrasonic air pressure change, the second flap pair is driven to perform the common mode movement, and the ultrasonic air pressure change having the ultrasonic carrier frequency is formed. The air pulse generator according to claim 1.

20. The first flap pair has a first demodulation flap and a second demodulation flap, the first demodulation flap and the second demodulation flap are driven to move in opposite directions, and the opening is formed. The air pulse generator according to claim 19.

21. The second flap pair has a first modulation flap and a second modulation flap, the first modulation flap and the second modulation flap are driven to move in the same direction, and the ultrasonic air pressure change is formed. The air pulse generator according to claim 19.

22. The first flap pair forms a first air wave by resonance in mode 1, the second flap pair forms a second air wave by resonance in mode 2. The air pulse generator according to claim 19.

23. The driving frequency corresponding to the driving signal for causing the first flap pair to perform the differential mode movement is half of the ultrasonic carrier frequency. The air pulse generator according to claim 19.

24. The amplitude-modulated ultrasonic air pressure change is double-sideband suppressed carrier (DSB-SC) modulated, the spectral components of the amplitude-modulated ultrasonic air pressure change are partially shifted to the baseband as a result of synchronous demodulation. The air pulse generator according to claim 1.

25. The flap pair in the membrane structure can function as a modulation part for performing the common mode movement and a demodulation part for performing the differential mode movement, and the modulation part and the demodulation part are manufactured in the same processing layer. The air pulse generator according to claim 1.

Images