The classical historical epitome of an analog transducer – the audio microphone – has largely become a MEMS-based component in recent years. According to a recent IHS report “Apple Products Are Driving Market Growth for MEMS Microphones,” billions of these are now being shipped annually due to smartphone demand; the newest phones even multiple such devices for enhanced sound capture.
I generally place low credibility in market-research reports which predict the far-away future with meaningless high precision, such as one which claimed “75.4 Billion Devices Connected to the Internet of Things by 2025” (no, I am not making that up; I just don’t want to embarrass anyone here). However, I do have reasonable level of confidence in reports from credible sources which look at recent shipments or installations—they are usually good enough to allow meaningful discussion. The MEMS microphone report fits into the latter category.
I'm not sure if I should be excited or nostalgic about the MEMS microphone news. After all, the microphone is among the first widely used analog transducers. Microphone technologies and their electronic interfaces have been a major part of the analog world since the earliest days. Going way back, the basic carbon-granule version was one of the two key enablers for Bell’s telephone (the speaker was the other, of course).
That's why I was relieved when I saw a brief but fascinating article in the always-excellent Physics Today about pre-MEMS microphone technology. “The vocal microphone: Technology and practice” is a highly readable overview of how microphone development has influenced singers and their vocal techniques, an aspect of microphones about which I only knew a little (much of it wrong) with numerous misconceptions.
The article explained how different singing styles use and exploit the inherent characteristics of classic implementations such as carbon, ribbon, condenser, and moving-coil microphone designs. Although it didn’t go into interface circuitry, there is a lot of valuable design information available on line for each microphone type as well, much of it from analog component vendors.
Building an interface circuit for a microphone is a nice entry point into the broader world of analog circuitry, and a great starter project for an aspiring student. It’s a topic brewing with issues of circuitry, of course, as well as preamplifiers, providing DC (phantom) power through the audio cable, impedance issues, ground loops, isolation, and many other real-world considerations. For those students who are more interested in mechanical engineering rather than electronics, there are online features and videos on how to build various microphones. They are nice, non-trivial hands-on projects which can be tangibly demonstrated at a school science fair or maker-type event.
Will MEMS microphones supplant the classics? The article notes that the Shure 55 Unidyne has been in production since 1939. In some ways, the MEMS devices already have. I did some research and couldn't find any numbers on the number of non-MEMS microphones sold annually, but it is undoubtedly many orders of magnitude less than that of the MEMS devices used in smartphones. I'd like to think that the audio characteristics and unique attributes of the MEMS units will never come close to the best of the traditional designs.
The Shure 55 Unidyne microphone is more than just a well-known, widely recognized classic icon and image, it is still being manufactured and in widespread use, over 75 years after its introduction. (Photo: Shure Incorporated)
But it's probably prudent to “never say never” when it comes to technology advances. It wasn't very long ago when we were assured that CCD-based imaging devices would not be replaced by CMOS imagers for a very, very long time, because the performance of the former was so much better (resolution, color rendition, low-light capability, streaking and blooming issues, and speed). Fast-forward just a few years to the present, and CMOS imagers have improved amazingly across all these parameters, leaving CCD units primarily for highly specialized applications; the image-capture quality of some of the latest smartphones is tangible proof of CMOS performance.
What's your experience with microphones and their electronics? Have you ever built interfaces, either as a personal or school project, or for commercial use? Have you switched from non-MEMS to MEMS-based microphones due to cost, size, or other considerations?