For half a century, electret condenser microphones (ECM) have dominated the microphone market. During this time, they were widely used in fixed telephones (especially in the 1980s/1990s), hearing aids, headsets, GPS devices, digital recorders, mobile phones and Home Radio Service (FRS) two-way radios, and can meet voice recognition And VoIP and other applications. However, as systems become smaller and thinner, smaller and more compact ECMs are also required, which has a serious impact on its signal-to-noise ratio (SNR). In addition, the performance, repeatability and temperature stability of ECM are also challenged. Even the ECM of the same model and the same design, its sensitivity may vary greatly within the operating temperature range. Their response at the low-frequency and high-frequency ends also shows a significant difference.
Compared with ECM, MEMS microphones have obvious advantages. They have higher SNR, lower power consumption, higher sensitivity and stronger vibration resistance, and also have a smaller package, thereby saving circuit board area. In addition, they are fully compatible with contemporary installation and assembly processes. But ECM is not suitable for these.
MEMS demand exceeds ECM
Although MEMS microphones came out in the 1980s, their output did not exceed ECM until five or six years ago. The huge demand in consumer markets such as smartphones and virtual personal assistant (VPA) smart speakers has driven their shipment growth. If you look at the number of microphones in these products, you can get a good idea of the size of this market. iPhone 6s, 6s+, 7, 7+ and 8 are equipped with four MEMS microphones respectively, while Samsung Galaxy Note 8 and 9 each have two MEMS microphones. The Samsung Galaxy Home smart speaker has 8 far-field MEMS microphones, and Amazon’s Echo has 7 directional far-field microphone arrays. Although most wearable devices (such as smart watches, etc.) have a MEMS microphone, relatively new audible devices often have at least four microphones in each pair, because like most headsets or headsets with active noise reduction Like headphones, there is a second microphone on each side (used to pick up and cancel ambient noise).
According to data from Yole Development, a market research consultancy, by 2022, the total global market for MEMS microphones, ECMs, micro speakers and audio ICs will reach 20 billion U.S. dollars, of which MEMS microphones will account for approximately 1.4 billion U.S. dollars. On the contrary, the compound annual growth rate of ECM market revenue and shipment volume during the same period was -7.5% and -2.6%, respectively.
IHS Markit predicts that by 2021, the global MEMS microphone market revenue will climb to US$1.5 billion (equivalent to approximately 6.6 billion shipments). ResearchandMarkets’ forecast is more optimistic. As more new application opportunities related to the Internet of Things (IoT) and automotive systems continue to emerge, the MEMS microphone market is expected to grow to US$2.9 billion by 2024.
ECM share declined but not out
ECM has not been excluded from the market. Manufacturers are already improving the design of SNR and enhancing its features such as dust and moisture resistance. For example, PUI Audio’s high-definition ECM has high sensitivity and excellent SNR. These high-definition microphones use advanced FET and diaphragm design, and integrate a GSM buzz-blocking capacitor. They can maintain their fidelity in the frequency range of 20Hz to 20kHz, making them suitable for medical, automotive, industrial and consumer electronics product designs. The waterproof ECM protection grades provided by CUI are divided into IP57, IP65 and IP67. These 4mm diameter omnidirectional microphones have lead and end mounting configurations, operating frequencies up to 20kHz, optimal sensitivity levels up to -42dB, and SNR from 57dBA to 70dBA.
Figure 1: CUI’s waterproof ECM.
With the development of MEMS microphone technology, significant improvements in automatic speech recognition (ASR) have enabled people to use virtual personal assistants (such as Siri, Alexa, Cortana and Google Assistant) for more conversational interactions, which are voice control in smart homes The interface paved the way.
Figure 2: TDK’s InvenSense ICS-40730.
TDK’s InvenSense ICS-40730 is designed for home automation systems, smart phones, teleconference equipment, security/surveillance systems and other voice control/activation applications. It is a low-noise, bottom-port, differential analog output MEMS microphone. The device contains an impedance converter and a differential output amplifier, with SNR up to 74dBA, sensitivity of -32dBV (differential) and -38dBV (single-ended), 124dB sound pressure level (SPL) acoustic overload point (AOP), -77dBV The power supply rejection (PSR) and the sensitivity tolerance of ±2dB and other characteristics.
At the same time, InvenSense ICS-40212 analog MEMS microphone has a high dynamic range and low power consumption “AlwaysOn” mode, suitable for applications such as wearable devices, IoT hardware, and smart phones. The device can provide a low-power mode, when the power supply voltage is lower than 2V, the low-power mode is activated, and work under the conditions of 55?A current and 123dB SPL AOP. It adopts 3.5mm x 2.65mm x 0.98mm bottom port surface mount package, with strict ±1dB sensitivity tolerance and extended frequency response (35Hz～20kHz).
STMicroelectronics (STMicroelectronics) MP34DTx series adopts a top port and an EMI shielded HCLGA package design. It is a compact, low-power, omnidirectional MEMS digital microphone. These low distortion components have 64dB SNR, 122.5dB SPL AOP and -26dBFS±3dB sensitivity. Each component has an IC interface manufactured with CMOS technology, which simplifies the design of a dedicated signal conditioning circuit.
Figure 3: Infineon’s XEN SIV microphone based on MEMS technology.
Infineon’s IM69D120 and IM69D130 XENSIV omnidirectional MEMS microphones are suitable for applications requiring low internal noise (ie high SNR), wide dynamic range and low distortion. Its target applications include home automation systems, smart speakers, IoT hardware, etc. field. These microphones have pulse density modulation (PDM) output, 70dB SNR and 105dB dynamic range, and 130dB SPL AOP.
Figure 4: Knowles’ SP SiSonic MEMS silicon microphone.
Knowles’ SP SiSonic series are MEMS silicon microphones developed to meet the needs of smartphones, notebook computers and various other portable devices. These target applications require broadband audio performance and RF immunity. SP SiSonic is a high-performance, low-power MEMS microphone. These compact, bottom-port devices integrate acoustic sensors, low-noise input buffers, and sigma-delta (sigma-delta) modulators or output amplifiers, and provide a variety of Performance mode (sleep, low power, standard, etc.).
Piezoelectric technology: the next generation of MEMS microphones
The industry’s demand for higher signal-to-noise ratio and higher reliability and durability has led to the development of piezoelectric MEMS microphones. These technologies were first introduced by Vesper Technologies in the UK. The company’s VM1000 omnidirectional piezoelectric MEMS microphone has a single-ended analog output, extended dynamic range and strong SNR value (94dB SPL at 1kHz). The microphone adopts waterproof (IP57 grade), dust-proof, particle-proof and shock-proof packaging, and has a very stable performance. It is available in a package size of 3.76mm x 2.95mm x 1.1mm, users can record from the other side of the room, with excellent environmental noise cancellation and “audio zoom” functions, and more accurate sound selection when recording.
The VM1010 piezoelectric MEMS microphone with bottom port can provide voice activation for many different battery-powered smart devices. It consumes only 10?A in the listening mode, and by using the sound energy in direct contact with it, the microphone can activate the peripheral system from the complete power-off mode. With a very robust architecture, this microphone has a user-definable loudness threshold, so the voice activation mechanism can be customized to suit the required distance, and the specific background noise characteristics of the application environment can also be considered. When the surrounding environment is quiet, the system will enter a low-power “sound wake-up” mode, which can save battery energy.