ECM vs. MEMS microphones: Does new mean better?

Not so long ago, the sight of someone speaking to their refrigerator or electric oven may have caused you to question if they were working too hard and perhaps in need of a vacation. Incredibly, this is quickly becoming an everyday phenomenon, which is no longer questioned or inspires strange looks. Speech is fast becoming the interface of choice for users to control the IoT devices that are proliferating our daily lives. For this reason, the choice of microphone is becoming an important design consideration for these, and other electronic devices.

In this article, we review the two most common microphone technologies – micro-electromechanical system (MEMS) and electret condenser microphones (ECM) (Fig. 1). We briefly explain their principle of operation and consider their relative merits before considering what to look for when choosing between them for different applications.

Fig. 1: MEMS vs. ECM (Image: CUI Devices)

What is a MEMS microphone?

MEMS is a term used to describe highly miniaturized components and systems, that sit on a die created using a semiconductor manufacturing process. This makes a MEMS microphone small enough to be mounted on a printed circuit board (PCB) as shown in Fig. 2.

This “top-port” MEMS microphone consists of a protective metal case with a small hole that allows sound to travel into the enclosure. In a “bottom-port” version, sound travels into the enclosure from below via a hole in the PCB.

Fig. 2: Internal construction of a MEMS microphone (Image: CUI Devices)

How does a MEMS microphone work?

A MEMS microphone contains a transducer that converts sound into electrical energy. The transducer consists of a diaphragm constructed to act as a capacitor that stores an electrical voltage. Sound waves traveling into the enclosure cause the diaphragm to vibrate, varying the voltage stored in the capacitor. This tiny variable electrical signal is detected and sent to an audio pre-amplifier to make it large enough for further analog processing (Fig. 3) or conversion into a digital format using an analog-to-digital converter (ADC) – the amplifier and ADC are typically located on a separate die to the transducer.

Fig. 3:  Schematic for analog MEMS (Image: CUI Devices)

Pulse density modulation (PDM) is an encoding scheme commonly used to transmit digital audio since it only requires a clock signal to be sent along with the audio data stream, thereby simplifying the decoding of the signal at the receiver (Fig. 4).

Fig. 4: Schematic for digital MEMS (Image: CUI Devices)

In other MEMS versions, the audio signal is filtered and processed directly within the microphone enclosure using a decimation filter, allowing for direct connection to a digital signal processor (DSP) or microprocessor. This digital audio encoding scheme called I²S eliminates the need for an ADC or codec in many cases.

What are the advantages of MEMS microphones?

MEMS is a new manufacturing technology that allows for additional analog and digital circuitry (preamplifiers, etc.) to be manufactured and located in the same package as the microphone transducer. This brings several advantages:

• Firstly, circuit elements located near one another exhibit closely matched performance characteristics, including temperature stability – an important specification in applications that use arrays.
• Secondly, the smaller packages of MEMS microphones are more easily handled by ‘pick and place’ PCB mounting machines, while also requiring less board space, ultimately translating into cost savings during manufacture.

Separately, while digital MEMS microphones are inherently noise-immune, analog MEMS microphones have the advantage of low output impedance, which makes them less susceptible to interference from unwanted electrical noise. MEMS microphones are also highly tolerant to external mechanical vibration and to the extreme temperature variations that occur with solder reflow during PCB manufacturing.

What is an ECM?

The ECM operates on a similar principle as the MEMS microphone but it is constructed differently. It consists of an electret diaphragm that is separated from a pick-up plate by an air-gap dielectric, which together behave as a fixed stored-charge capacitor (Fig. 5).

Fig. 5: Internal construction of a typical ECM (Image: CUI Devices)

How does an ECM work?

Sound waves cause the electret diaphragm to vibrate, thereby varying the capacitance (ΔC). Since the stored charge (Q) is fixed, the voltage across the capacitor must vary (ΔV) in response to the vibration of the diaphragm according to the equation ΔV = Q/ ΔC. This is how it converts sound into electrical energy. The electrical voltage signal is sent to the gate of a JFET transistor (also housed within the microphone enclosure) operating as a common-source amplifier in conjunction with an external load resistor and a DC blocking capacitor (Fig. 6).

Fig. 6: Schematic for a typical ECM (Image: CUI Devices)

What are the advantages of an ECM?

ECMs can be connected to application circuits in several ways, thereby providing system designers with a great degree of flexibility. Connection options include wires, pins, solder pads, SMT, and spring contacts. Although they are physically larger than MEMS microphones in most cases, this perceived disadvantage actually provides them with one of their main strengths, namely, performance in the presence of environmental moisture and dust.

As a result, ECMs typically come with higher ingress protection (IP) ratings than MEMS microphones. They also operate over a wide range of voltages, which is an advantage in applications that have poorly regulated supplies.

Which should I choose – MEMS or ECM?

Due to their small size, electrical noise immunity, and mechanical robustness, MEMS microphones are becoming increasingly popular. However, this does not mean that they are the de-facto best choice for every application. Many legacy applications may benefit from a simple change or upgrade to their current ECM. Apart from their excellent IP ratings, which allow them to perform well in harsh environments, the intrinsic nature of ECMs also makes them an excellent choice for applications that benefit from noise-canceling or unidirectionality.

If the space constraints placed upon a design are particularly acute (such as in smartphones, wearables, hearing aid implants, etc.), or there is likely to be the need to distribute multiple microphones throughout an item of equipment (like in a VR headset for instance) then MEMS may be the better option.

Conversely, if elevated performance or resilience to challenging operating conditions are a priority, then ECM could prove to be the more appropriate route to take. Therefore, professional audio equipment, voice-controlled home assistants, voice recognition systems, and a wide range of other applications will continue to rely on these devices.

By engaging with a company that has a comprehensive understanding of the properties of both ECM and MEMS microphones, and the respective benefits of each, the optimal choice can be made.

About the author:

With an extensive knowledge of CUI Devices’ products, Ryan Smoot provides customers with a wide range of technical and application support capabilities in the field. His management of CUI Devices’ robust CAD model library further offers engineers with an invaluable resource for streamlining their product designs. When he is not assisting customers, Ryan enjoys running, the outdoors, and spending time with his wife and two kids.