MEMS麦克风的基本原理教学教材

February 2017DocID025704 Rev 2 1/20 AN4426 Application note Tutorial for MEMS microphonesIntroductionThis application note serves as a tutorial for MEMS microphones, providing general characteristics of these devices, both acoustic and mechanical, as well as summarizing the portfolio available from ST. MEMS microphones target all audio applications where small size, high sound quality, reliability and affordability are key requirements.ST's MEMS microphones are designed, developed and manufactured within ST, creating an industry-unique, vertically integrated supply chain. Both analog and digital-input, top and bottom-port solutions are available.Our best-in-class AOP and SNR make ST’s MEMS microphones suitable for applications that require a very high dynamic range, improving the audio experience in any environment. Matching very tightsensitivity allows optimizing beamforming and noise cancelling algorithms for multi-microphone arrays. Low power consumption allows extending battery life.Contents AN4426 Contents1Mechanical specifications, construction details (4)2Acoustic parameters (11)2.1Sensitivity (11)2.2Directionality (11)2.3SNR (12)2.4Dynamic range and acoustic overload point (12)2.5Equivalent input noise (13)2.6Frequency response (15)2.7Total harmonic distortion (16)2.8PSRR and PSR (16)3MEMS microphone portfolio (17)4Revision history (19)2/20 DocID025704 Rev 2AN4426 List of figures List of figuresFigure 1: MEMS microphone inside package (4)Figure 2: MEMS transducer mechanical specifications (4)Figure 3: Capacitance change principle (5)Figure 4: 4 x 5 package (5)Figure 5: 3 x 4 metal cap package - bottom port (6)Figure 6: 3 x 4 package - top port (6)Figure 7: 2 x 3 package - bottom port (7)Figure 8: Faraday cage in ST’s MEMS microphones (7)Figure 9: RF immunity simulation (8)Figure 10: EMC test setup (8)Figure 11: RF test disturbance signal with sinusoidal pattern (9)Figure 12: RF immunity test results - MP34DT04 (9)Figure 13: RF test disturbance signal @ 217 Hz burst pattern (10)Figure 14: RF immunity of analog differential microphones (10)Figure 15: Omnidirectional microphone (11)Figure 16: A-weighted filter response (12)Figure 17: Acoustic and electrical relationship - analog (13)Figure 18: Acoustic and electrical relationship - digital (14)Figure 19: MP45DT02-M frequency response (15)Figure 20: MEMS microphone portfolio (17)Figure 21: MEMS microphone notation (17)DocID025704 Rev 2 3/201 Mechanical specifications, construction detailsA microphone is a dual-die device consisting of two components, the integrated circuit andthe sensor, which are housed in a package using techniques that are proprietary to ST.ASICintegratedThe sensor uses MEMS technology (Micro-Electrical-Mechanical Systems) and it isbasically a silicon capacitor. The capacitor consists of two silicon plates/surfaces. Oneplate is fixed while the other one is movable (respectively, the green plate and the grey oneshown in the following figure). The fixed surface is covered by an electrode to make itconductive and is full of acoustic holes which allow sound to pass through. The movableplate is able to move since it is bonded at only one side of its structure. A ventilation hole,allows the air compressed in the back chamber to flow out and consequently allows themembrane to move back. The chamber allows the membrane to move inside but also, incombination with the chamber created by the package will affect the acoustic performanceof the microphones in terms of frequency response and SNR.Figure 2: MEMS transducer mechanical specificationsSo basically the microphone MEMS sensor is a variable capacitor where the transductionprinciple is the coupled capacitance change between a fixed plate (back plate) and amovable plate (membrane) caused by the incoming wave of the sound.4/20 DocID025704 Rev 2Figure 3: Capacitance change principleThe integrated circuit converts the change of the polarized MEMS capacitance into a digital (PDM modulated) or analog output according to the microphone type. Finally the MEMS microphone is housed in a package with the sound inlet placed in the top or in the bottom part of the package, hence the top-port or bottom-port nomenclature of the package. ST manufactures microphones using industry-wide techniques, but also has developed innovative packaging to achieve improved performance of the microphones. Packaging techniques will be discussed in further detail.The 4x5 package is widely used to house the digital microphone MP45DT02-M. It is a common packaging technique in a top-port configuration where the ASIC is placed under the sound inlet with glue on top (glob top) in order to protect the circuit from light and the MEMS sensor is placed beside the integrated circuit. The two silicon components are fixed to the substrate and the pads of the device are on the bottom side. The resonant chambers are identified depending on the position of each chamber with respect to the membrane and the incoming sound. In this case, considering the incoming direction of the sound, the front chamber is created by the package and the chamber inside the MEMS, behind the MEMS membrane, is the back chamber. This configuration allows protecting the MEMS from dust and particles falling into the package but results in a low SNR and frequency response with a peak in the audio band.Figure 4: 4 x 5 packageThe 3x4 package is used in ST to produce both the bottom and the top-port digital microphones, MP34DB02 and MP34DT01-M, MP34DT04/-C1, and MP34DT05. Considering the bottom configuration first, this structure is depicted in the following figure. The ASIC and the MEMS sensor are fixed to the substrate and the pads of the device are on bottom side as well. The sound inlet is obtained by drilling the substrate according to the position of the MEMS sensor. The package encloses all the components. In this configuration the front chamber is the cavity of the MEMS sensor and the package creates the back chamber. This design optimizes the acoustic performance of the microphone in terms of SNR and also allows obtaining a flat response across the entire audio band. The drawback of this solution is represented by the assembly of this microphone. Usually theDocID025704 Rev 2 5/20bottom-port microphones are soldered on the PCB. The thickness of the board modifies thevolume of the front chamber, degrading the flat response of this type of microphone (referto AN4427, “Gasket design for optimal acoustic performance in MEMS microphones” fordetails). In order to minimize the artifacts caused by this environment, a flex cable isrecommended to be used. Additionally, the bottom-port microphones have a ringed metalpad around the hole. A very careful soldering process is required to avoid dust or solderingpaste from entering in the sound port, damaging the MEMS membrane.Figure 5: 3 x 4 metal cap package - bottom portThe 3x4 top-port configuration is basically a mirrored bottom-port microphone. The ASICand the sensor are placed close to each other, the sensor is still under the sound inlet butthese two components are attached to the top of the structure, in other words, the ASICand MEMS are fixed to the package lid, not to the substrate. The pads are on the substrateand thus on the bottom side of the microphone. This configuration, covered by ST patents,allows optimizing all the benefits of the bottom-port microphone in terms of acousticperformance (i.e. maximized SNR and flat band) and all the benefits related to the top-portconfiguration during the assembly process.Figure 6: 3 x 4 package - top port6/20 DocID025704 Rev 2A smaller package, 2.5 x 3.35 mm, has been introduced in ST's product portfolio (for simplicity referred to as 2x3, see Figure 7: "2 x 3 package - bottom port"). This package is a bottom-port configuration with the same internal construction as the 3 x 4 bottom-port package and it is used for the analog differential microphones MP23AB01DM/DH and analog single-ended microphone MP23AB02B. As a result of the 2x3 bottom-port package and differential output configuration, the MP23AB01DH is the best microphone provided by ST in terms of SNR and AOP.Figure 7: 2 x 3 package - bottom portMEMS microphones housed in a plastic package are protected from radiated disturbances by embedding in the plastic package a metal shield which serves as a Faraday cage. The model in the following figure shows how the Faraday cage is implemented in ST’s plastic packages.Figure 8: Faraday cage in ST’s MEMS microphonesThe next figure shows the simulation of an electric field in open space. By applying an electric field source outside the microphone package, the Faraday cage is able to considerably attenuate the field inside the microphone structure. The temperature grade of the E field is an easy way to plot the results.DocID025704 Rev 2 7/20Figure 9: RF immunity simulationIn addition to the simulation, ST has a dedicated test to evaluate immunity, “Microphonedurability to EMC disturbances”.Microphones are subjected to RF disturbances using a proper jig with the following setup.Basically the test consists of placing the microphone under an antenna radiating adisturbance signal of 1 kHz AM modulated in the range [0.8, 3] GHz. The RF amplitudediffers depending on the frequency range according to the following criteria:•+33 dBm in the range [0.8, 2.4]•+17 dBm in the range [2.4, 3.0]8/20 DocID025704 Rev 2Figure 11: RF test disturbance signal with sinusoidal patternThe carrier of the disturbance is 1 kHz since it is an audio signal. Hence, the RF immunity of the microphone is evaluated by measuring the residual of the carrier at the output of the microphone. The next figure shows the result of the peak at 1 kHz measured when applying the RF disturbance on top of an MP34DT04.In parallel with the sinusoidal pattern, another 217 Hz burst pattern used to test the RF immunity is shown in the following figure.•The RF amplitude (power): +33 dBm•Carrier frequency: 700 MHz ~ 2.5 GHz•GSM burst frequency: 217 Hz pattern (see below)DocID025704 Rev 2 9/20Figure 13: RF test disturbance signal @ 217 Hz burst patternFigure 14: RF immunity of analog differential microphones10/20 DocID025704 Rev 2AN4426Acoustic parametersDocID025704 Rev 211/202Acoustic parameters2.1SensitivityThe sensitivity is the electrical signal at the microphone output to a given acoustic pressure as input. The reference of acoustic pressure is 1 Pa or 94 dBSPL @ 1 kHz. The sound pressure level, expressed in decibel, dBSPL=20*Log(P/Po) where Po = 20 µPa is the threshold of hearing. 20*Log(1Pa/20µPa) = 94 dBSPL • For analog microphones the sensitivity is expressed in mV RMS /Pa or dBV/Pa •For digital microphones the sensitivity is expressed in dBFSdBV ≠ dBFS. It is not correct to compare different units. As given in the above equations, dBV is in reference to 1V RMS instead of dBFS where the reference is the digital full scale.2.2 DirectionalityThe directionality indicates the variation of the sensitivity response with respect to thedirection of the arrival of the sound. MEMS microphones from ST are omnidirectional which means that there is no sensitivity change to every different position of the source of thesound in space. The directionality can be indicated in a Cartesian axis as sensitivity drift vs. angle or in a polar diagram showing the sensitivity pattern response in space.The following figure depicts the directionality in these two reference systems.Figure 15: Omnidirectional microphoneAcoustic parametersAN442612/20DocID025704 Rev 22.3 SNRThe signal-to-noise ratio specifies the ratio between a given reference signal to the amount of residual noise at the microphone output. The reference signal is the standard signal at the microphone output when the sound pressure is 1Pa @ 1 kHz (microphone sensitivity). The noise signal (residual noise) is the microphone electrical output at silence.This parameter includes both the noise of the MEMS element and the ASIC. Concerning this sum, the main contribution to noise is given by the MEMS sensor, the integrated circuit contribution can be considered negligible. Typically, the noise level is measured in an anechoic environment and A-weighting the acquisition. The A-weighted filter corresponds to the human ear frequency response.Figure 16: A-weighted filter response2.4 Dynamic range and acoustic overload pointThe dynamic range is the difference between the minimum and maximum signal that the microphone is able to generate as output. • The minimum signal is the smallest audio signal that the microphone can generate distinctly from noise. In other words, the minimum signal is equivalent to the residual noise.•The maximum audio signal is that which the microphone can generate withoutdistortion. It is also called acoustic overload point (AOP). Actually, the specification allows up to 10% in terms of distortion at the acoustic overload point.AN4426 Acoustic parameters2.5 Equivalent input noiseA microphone is a sound-to-electricity transducer which means that any output signalcorresponds to a specific sound as input. The equivalent input noise (EIN) is the acousticlevel, expressed in dBSPL, corresponding to the residual noise as output.For example, a digital microphone with a sensitivity of -26 dBFS and a 63 dB as SNR:Residual noise = -26 - 63 = -89 dBFS this sum transposed in the acoustic domain is:EIN = 94 - 63= 31 dBSPLThe following figures summarize the relationship between the acoustic and electricdomains related to each of the parameters listed above. Figure 17: "Acoustic and electricalrelationship - analog" and Figure 18: "Acoustic and electrical relationship - digital" illustratethis for analog and digital microphones, respectively.Figure 17: Acoustic and electrical relationship - analogDocID025704 Rev 2 13/20Acoustic parametersAN442614/20DocID025704 Rev 29030-90-3094dBSPL AOPS N R =63d B-26dBFSD y n a m i c R a n g e =89d BSensitivity EIN Residual noiseAN4426 Acoustic parameters 2.6 Frequency responseThe frequency response of a microphone in terms of magnitude indicates the sensitivityvariation across the audio band. This parameter also describes the deviation of the outputsignal from the reference 0 dB. Typically, the reference for this measurement is exactly thesensitivity of the microphone @ 0 dB = 94 dBSPL @ 1 kHz. The frequency response of amicrophone can vary across the audio frequency band depending on three parameters: theventilation hole, the front chamber geometry, and back chamber geometry. The ventilationhole and the back chamber geometry have an impact on the behavior at low frequencieswhile the behavior at high frequencies depends on the geometry of the front chamber only.Behavior at high frequencies can be a resonance peak caused by the Helmholtz effect.This resonance is the phenomenon of air resonance in a cavity. As a matter of fact, itdepends on the dimension of the front chamber of the microphone, representing the soundcavity in which the air resonates. A microphone with a flat frequency response is suitablewhen natural sound and high intelligibility of the system is required. The following figureshows the response of the MP45DT02-M. It shows a roll-off at low frequencies and a peakaround 18 kHz caused by the large front chamber of this microphone.Figure 19: MP45DT02-M frequency responseThe frequency response of a microphone in terms of phase indicates the phase distortionintroduced by the microphone. In other words, the delay between the sound wave movingthe microphone membrane and the electrical signal at the microphone output results in thatthis parameter includes both the distortion due to the membrane and the ASIC.DocID025704 Rev 2 15/20Acoustic parametersAN442616/20DocID025704 Rev 22.7 Total harmonic distortionTHD is the measurement of the distortion affecting the electrical output signal of themicrophone given an undistorted acoustic signal as input. THD+N is expressed as a ratio of the integer in a specified band of the power of the harmonics plus the power of noise and the power of the undistorted signal (fundamental). Equation 1TTTTTT +NN (%)=∑PPPPPPPPPP (TTHHPPHHPPHHHH HH HH )+PPPPPPPPPP (NNPPHH HH PP )NN nn−1PPPPPPPPPP (FFFFHHFFHHHHPPHHFFHHFF )Typically ST indicates the THD+N measured in the (50 Hz - 4 kHz) band for a given undistorted signal 1 kHz @ 100 dBSPL.2.8 PSRR and PSRPSRR indicates the capability of the ASIC to reject noise added to the supply voltage. To evaluate this parameter, a tone of V IN = 100 mVpk-pk @ 217 Hz (GSM switching frequency in phone applications) is added to the power supply and then the amplitude of the output is measured. The added noise can be either a square wave or sinusoidal wave. Typically the square wave is preferred since it is the worst case.PSRR is the ratio of the residual noise amplitude at the microphone output (V OUT @ 217 Hz) to the added spurious signal on the supply voltage. It is typically expressed in dB as given in the equation below: Equation 2PPPPPPPP =20 xx log �(VV OOOOOO @217TTHH )(VV IINN @217TTHH )�The capability of the integrated circuit to reject noise added to the supply voltage can also be expressed with another parameter that is the PSR. Basically it is simply a measurement of the output when noise of 100 mVpk-pk @ 217 Hz is superposed to the supply voltage. Consequently expressed in dB as given in the equation below: Equation 3 PPPPPP =20 xx log[VV OOOOOO @217TTHH ]To evaluate either the PSRR or PSR, proper sealing of the sound inlet or measurements performed in an anechoic chamber are recommended to avoid mixing the superimposed noise with that of the noise floor of the output. Finally, in the microphone datasheets PSR is commonly given instead of PSRR.AN4426 MEMS microphone portfolio3 MEMS microphone portfolioFigure 20: MEMS microphone portfolioST’s portfolio includes digital and analog microphones. The commercial products arenamed using the notation depicted in the following figure.Figure 21: MEMS microphone notationDocID025704 Rev 2 17/20MEMS microphone portfolio AN442618/20DocID025704 Rev 2The following table provides a complete overview of the microphones offered by ST. Additionally it serves as a summary for selecting the appropriate microphone among the ST portfolio as the features of both digital and analog microphones are given.Table 1: Features of MEMS microphonesParameter MP45DT02-M MP34DB02 MP34DT01-M MP34DT04 MP34DT04-C1 MP34DT05 MP23AB02B MP23AB01DM MP23AB01DH Sensitivity -26 dBFS -26 dBFS -26 dBFS -26 dBFS -26 dBFS -26 dBFS -38 dBV -38 dBV -38 dBV Directivity OmnidirectionalOmnidirectional OmnidirectionalOmnidirectionalOmnidirectionalOmnidirectionalOmnidirectionalOmnidirectionalOmnidirectionalSNR 61 dB 62.5 dB 61 dB 64 dB 64 dB 64 dB 64 dB 64 dB 65 dB AOP 120 dBSPL 120 dBSPL 120 dBSPL 120 dBSPL 120 dBSPL 122 dBSPL 125 dBSPL 130 dBSPL 135 dBSPL EIN 33 dB 31.5 dB 33 dB 30 dB 30 dB 30 dB 30 dB 30 dB 29 dB THD+N <5% @ 115 dBSPL <5% @ 115 dBSPL <2% @ 115 dBSPL <5% @ 115 dBSPL <5% @ 115 dBSPL <6% @ 120 dBSPL <2% @ 120 dBSPL <10% @ 130 dBSPL <5% @ 130 dBSPL PSR -70 dB -86 dB -70 dB -70 dB -70 dB -72 dB -70 dB -85 dB -100 dB Max. current consumption 650 µA 650 µA 600 µA 700 µA 700 µA 650 µA 220 µA 250 µA 250 µA Package dimensions 3.76x4.76x1.25mm3x4x1mm 3x4x1.06mm 3x4x1.095mm 3x4x1.095mm 3x4x1mm 2.5x3.35x0.98mm 2.5x3.35x0.98mm 2.5x3.35x0.98mm Port location Top port Bottom port Top port Top port Top port Top port Bottom port Bottom port Bottom port Operating temperature-30°C<T<+85°C-40°C<T<+85°C-40°C<T<+85°C-40°C<T<+85°C-40°C<T<+85°C-40°C<T<+85°C-40°C<T<+85°C-40°C<T<+85°C-40°C<T<+85°CAN4426Revision historyDocID025704 Rev 219/204 Revision historyTable 2: Document revision historyDate RevisionChanges09-Jan-20141Initial release14-Feb-2017 2Updated part numbers throughout documentUpdated "Introduction" and Section 1: "Mechanical specifications, construction details"Updated Figure 5: "3 x 4 metal cap package - bottom port", Figure 20: "MEMS microphone portfolio", Figure 21: "MEMS microphone notation"Added Figure 7: "2 x 3 package - bottom port",Figure 13: "RF test disturbance signal @ 217 Hz burst pattern", Figure 14: "RF immunity of analog differential microphones" Updated Table 1: "Features of MEMS microphones"AN4426IMPORTANT NOTICE – PLEASE READ CAREFULLYSTMicroelectronics NV and its subsidiaries (“ST”) reserve the right to make changes, corrections, enhancements, modifications, and improvements to ST products and/or to this document at any time without notice. Purchasers should obtain the latest relevant information on ST products before placing orders. ST products are sold pursuant to ST’s terms and conditions of sale in place at the time of order acknowledgement.Purchasers are solely responsible for the choice, selection, and use of ST products and ST assumes no liability for application assistance or the design of Purchasers’ products.No license, express or implied, to any intellectual property right is granted by ST herein.Resale of ST products with provisions different from the information set forth herein shall void any warranty granted by ST for such product.ST and the ST logo are trademarks of ST. All other product or service names are the property of their respective owners.Information in this document supersedes and replaces information previously supplied in any prior versions of this document.© 2017 STMicroelectronics – All rights reserved20/20 DocID025704 Rev 2。

一种MEMS芯片及其制作方法引言MEMS（Micro-Electro-Mechanical Systems）芯片是一种集成了微观机械部件、电学元件和电子集成线路的微型器件。

它在现代电子技术中具有广泛的应用，如加速计、压力传感器、麦克风等。

本文将介绍一种基于MEMS技术的芯片及其制作方法。

背景MEMS芯片的发展源于集成电路技术的快速进展。

通过微电子加工工艺，可以将微观机械结构与电路部件相结合，从而实现功能更加复杂的微型器件。

在MEMS芯片中，传感器是常见的元件之一，而MEMS麦克风则是其中的重要应用之一。

MEMS麦克风MEMS麦克风是一种利用MEMS技术制作的微型麦克风。

它具有体积小、功耗低、灵敏度高等优点，广泛应用于消费电子产品、通信设备等领域。

下面将介绍一种MEMS麦克风的制作方法。

制备MEMS麦克风的流程1.基底制备：首先，选择适合的基底材料，常见的有硅（Si）基底。

然后，使用光刻工艺在基底表面形成薄膜层，通常使用光刻胶和掩膜进行图案定义。

2.薄膜沉积：在基底表面沉积一层薄膜，常见的材料包括金属薄膜、多层金属膜等。

薄膜沉积可以使用物理气相沉积（PVD）或化学气相沉积（CVD）等方法。

3.薄膜刻蚀：使用光刻工艺和刻蚀工艺将薄膜层进行图案定义和刻蚀，形成MEMS麦克风的微结构。

4.封闭结构：在微结构形成后，使用封闭工艺封闭MEMS麦克风的结构，保护内部部件免受环境影响。

5.封装：将封闭的MEMS麦克风器件进行封装，通常使用注塑成型或裸芯片直接封装等方式。

制备MEMS麦克风的优势制备MEMS麦克风采用了先进的微纳加工技术，具有以下优势：•小尺寸：MEMS麦克风的尺寸小，可以实现更小型化的产品设计。

•低功耗：由于MEMS麦克风的特殊结构，功耗较低，有利于延长电池寿命。

•高灵敏度：MEMS麦克风的微结构可以实现高灵敏度的声音接收，能够捕捉到更多细节。

•可靠性高：制备过程中采用精密的工艺控制和封装技术，提高了MEMS麦克风的可靠性。

mems mic结构-回复MEMS麦克风（Micro-Electro-Mechanical Systems Microphone）是一种微型电子器件，采用微机电系统技术制造而成，用于将声音转换为电信号。

本文将一步一步回答有关MEMS麦克风结构的问题。

第一部分：引言MEMS麦克风是当今许多电子设备中的关键组成部分，如手机、耳机、音频设备等。

它的小巧尺寸和高性能使其成为一种理想的音频传感器。

在了解MEMS麦克风的结构之前，我们首先需要了解MEMS技术的基本原理。

第二部分：MEMS技术的基本原理MEMS技术是一种将微型机械系统和电子技术相结合的技术，它使用半导体制造工艺在微米级别上制造具有机械功能的结构。

根据MEMS技术的原理，MEMS麦克风结构可以被划分为以下几个部分：1.薄膜振膜MEMS麦克风的核心部件是薄膜振膜。

薄膜振膜是由一层薄的材料制成，通常是硅或聚合物。

当声波通过振膜时，振动会引起膜片的形变，从而产生电荷或电压变化。

2.电容板电容板位于薄膜振膜的下方，与薄膜振膜之间形成一个微小的电容间隙。

电容板上会施加一定的电压，形成与薄膜振膜之间的电容。

当薄膜振膜振动时，电容的值会发生变化，从而改变电信号的大小。

第三部分：MEMS麦克风的制造过程MEMS麦克风的制造过程通常包括以下几个步骤：1.硅片制备首先，通过半导体工艺将硅片切割成适当的尺寸。

这些尺寸可以根据麦克风的设计要求来确定。

2.薄膜沉积将薄膜材料沉积在硅片上，可以使用物理气相沉积或化学气相沉积等技术。

薄膜的厚度可以根据设计要求进行控制。

3.光刻和蚀刻通过光刻和蚀刻等步骤，将硅片上的薄膜制作成薄膜振膜和电容板的形状。

这些步骤使用光罩和化学蚀刻剂等工具进行。

4.封装和连接将制作好的MEMS麦克风封装在保护壳内，以保护其免受外部环境的干扰。

然后，通过金线焊接或其他技术将麦克风连接到电路板上，以传输声音信号。

第四部分：MEMS麦克风的工作原理MEMS麦克风工作时，当声波进入麦克风时，它会使薄膜振膜发生弯曲。

MEMS麦克风工作原理及应用于助听器的前景因为人口老龄化和听力丧失人群的显然增强，助听器市场不断增长，但其惹眼的形状和很短的电池寿命让许多人失去爱好。

随着听力丧失现象变得越发常见，人们将寻求越发小巧、更有效、更高品质的助听器。

助听器信号链的前端是麦克风，它检测语音和其他环境噪声。

因此，充实音频捕获可以提高信号链整体的性能并降低功耗。

麦克风是把声学信号转换为电信号以供助听器音频信号链处理的。

有许多技术可用于这种声电转换，但麦克风是其中尺寸最小、精度最高的一类麦克风。

电容麦克风中的薄膜随着声学信号而运动，这种运动引起电容变幻，进而产生电信号。

驻极体电容麦克风(ECM)是助听器中用法最广泛的技术。

ECM采纳可变电容，其一个板由具有永远电荷的材料制成。

ECM在当今助听行业声名显赫，但这些设备背后的技术自1960年月以来并无多大变幻。

其性能、可重复性以及相对于温度和其他环境条件的稳定性不是十分好。

助听器以及其他注意高性能和全都性的应用，为新型麦克风技术的进展制造了机会。

新技术应该能充实上述缺点，让创造商生产出更高质量、越发牢靠的设备。

微机电系统()技术是电容麦克风变革的中坚力气。

MEMS麦克风利用了过去数十年来硅技术的巨大长进，包括超小型创造结构、精彩的稳定性和可重复性、低功耗，全部这些都已成为硅工业不折不扣的要求。

迄今为止，MEMS麦克风的功耗和噪声水平还是相当高，不宜用于助听器，但满足这两项关键要求的新器件已经浮现，正在掀起助听器麦克风的下一波创新浪潮。

MEMS麦克风工作原理像ECM一样，MEMS麦克风也是电容麦克风。

MEMS麦克风包含一个灵便悬浮的薄膜，它可在一个固定背板之上自由移动，全部元件均在一个硅晶圆上创造。

该结构形成一个可变电容，固定电荷施加于薄膜与背第1页共4页。

MEMS 硅麦克风MEMS 麦克风采用批量化的半导体制作工艺, 具有尺寸小、性能优良、一致性高等特点, 并且易于实现阵列化, 对语音效果实现了较大的提升。

根据制造技术, 麦克风可以分为两种主要类型, 传统的驻极体麦克风和MEMS麦克风。

驻极体麦克风通常由独立的金属部件和聚合物材料制成, 尺寸较大, 不易于集成和大批量生产。

而MEMS 麦克风采用与集成电路工艺兼容的硅微加工技术制成, 尺寸较小, 比较适合集成和大规模量产, 进一步降低了生产成本, 并在性能上也得到了较大的提升.电容式MEMS 麦克风主要由两块平行的导电极板组成（包括固定极板和可动极板）, 当可动极板在声波作用下产生振动时, 改变了两极板间的距离, 从而引起电容值的变化。

那么, 通过专用集成电路（Application Specific Integrated Circuit, ASIC）可以将电容的变化转换成电压信号。

从设计的角度来说, MEMS麦克风的灵敏度取决于电学灵敏度和机械灵敏度。

其中, 电学灵敏度与偏置电压和极板面积成正比, 与极板间的距离成反比。

因此, 偏置电压越高, 极板面积越大, MEMS 麦克风的电学灵敏度就越高。

但是, 增大极板面积就意味着增大MEMS 麦克风的尺寸, 提升偏置电压就意味着增加功耗, 而且偏置电压也会受到吸合电压的限制而不能任意增大。

因此, 这就需要在尺寸、功耗、灵敏度之间找到一个平衡点, 在不增加尺寸和功耗的前提下进一步提升MEMS 麦克风的灵敏度。

MEMS 麦克风的机械灵敏度与振膜（可动电极）的刚度成正比, 一般来说, 刚度越小的薄膜在声波作用下产生的形变就越大。

因此, 减小振膜刚度可以获得更高的机械灵敏度, 但是在制作过程中, 刚度较小的振膜极易受到外界的影响产生形变甚至破裂。

而且在静电力的作用下, 振膜与固定极板之间会产生一个吸引力, 导致振膜逐渐向固定极板靠近, 当振膜与固定极板接触时的偏置电压称为吸合电压。

CMOS硅麦克风原理随着智能手机的兴起，对于声音品质和轻薄短小的需求越来越受到大家的重视，近年来广泛应用的噪声抑制及回声消除技术均是为了提高声音的品质。

相比于传统的驻极体式麦克风(ECM)，电容式微机电麦克风采用硅半导体材料制作，这便于集成模拟放大电路及ADC(∑-∆ ADC)电路，实现模拟或数字微机电麦克风元件，以及制造微型化元件，非常适合应用于轻薄短小的便携式装置。

本文将针对CMOS微机电麦克风的设计与制造进行介绍，并比较纯MEMS与CMOS工艺微导入麦克风的差异。

电容式微麦克风原理MEMS微麦克风是一种微型的传感器。

其原理是利用声音变化产生的压力梯度使电容式微麦克风的声学振膜受声压干扰而产生形变，进而改变声学振膜与硅背极板之间的电容值。

该电容值的变化由电容电压转换电路转化为电压值的输出变化，再经过放大电路将MEMS传感器产生得到电压放大输出，从而将声压信号转化成电压信号。

在此必须采用一个高阻抗的电阻为MEMS传感器提供一个偏置电压VPP，借以在MEM S传感器上产生固定电荷，最后的输出电压将与VPP及振膜的形变∆d成正比。

振膜的形变与其刚性有关，刚性越低则形变越大；另一方面，输出电压与d(气隙)成反比，因此气隙越低，则输出电压及灵敏度越优，但这都将受限于MEMS传感器的吸合电压，也就是受限于MEMS传感器静电场的最大极限值(图1)。

CMOS微机电麦克风电路设计在CMOS微麦克风设计中，电路是一个非常重要的环节，它将影响到微麦克风的操作、感测，以及系统的灵敏度。

以图2为例，驻极式电容微麦克风的感应电荷由驻极体材料本身提供的驻极电荷所产生，而凝缩式电容微麦克风则是采用从CMOS的操作电压中抽取一个偏置电压，再通过一个高阻抗电阻提供给微麦克风的声学振膜来提供固定的电荷源。

此时，若声学振膜受到声压驱动而产生位移变化，则电极板(感测端)的电压将会发生变化。

最后，通过电路放大器将信号放大，则可实现模拟麦克风的电路设计；如果再加上一个∑-∆ ADC模数转换电路，便可完成数字麦克风的电路设计(一般数字麦克风的输出信号为1比特PDM 输出)。

∙MEMS（微型机电系统）麦克风是基于MEMS技术制造的麦克风，简单的说就是一个电容器集成在微硅晶片上，可以采用表贴工艺进行制造,能够承受很高的回流焊温度,容易与 CMOS 工艺及其它音频电路相集成, 并具有改进的噪声消除性能与良好的RF 及EMI 抑制性能.MEMS麦克风的全部潜能还有待挖掘，但是采用这种技术的产品已经在多种应用中体现出了诸多优势，特别是中高端手机应用中。

目录∙MEMS麦克风的优势∙MEMS麦克风的主要参数∙MEMS麦克风的发展前景MEMS麦克风的优势∙目前，实际使用的大多数麦克风都是ECM(驻极体电容器)麦克风，这种技术已经有几十年的历史。

ECM 的工作原理是利用驻有永久电荷的聚合材料振动膜。

与ECM的聚合材料振动膜相比，MEMS麦克风在不同温度下的性能都十分稳定，其敏感性不会受温度、振动、湿度和时间的影响。

由于耐热性强，MEMS麦克风可承受260℃的高温回流焊，而性能不会有任何变化。

由于组装前后敏感性变化很小，还可以节省制造过程中的音频调试成本。

MEMS麦克风需要ASIC提供的外部偏置，而ECM没有这种偏置。

有效的偏置将使MEMS麦克风在整个操作温度范围内都可保持稳定的声学和电气参数，还支持具有不同敏感性的麦克风设计。

传统ECM的尺寸通常比MEMS麦克风大，并且不能进行SMT（表面贴装技术）操作。

在MEMS麦克风的制造过程中，SMT回流焊简化了制造流程，可以省略一个目前通常以手工方式进行的制造步骤。

在ECM麦克风内，必须添加进行信号处理的电子元件；而在MEMS麦克风中，只需在芯片上添加额外的专用功能即可。

与ECM相比，这种额外功能的优点是使麦克风具有很高的电源抑制比，能够有效抑制电源电压的波动。

另一个优点是，集成在芯片上的宽带RF抑制功能，这一点不仅对手机这样的RF应用尤其重要，而且对所有与手机操作原理类似的设备（如助听器）都非常重要。

MEMS麦克风的小型振动膜还有另一个优点，直径不到1mm的小型薄膜的重量同样轻巧，这意味着，与ECM相比，MEMS麦克风会对由安装在同一PCB上的扬声器引起的PCB 噪声产生更低的振动耦合。