The datasheet for the PMM-3738-VM1000-R describes this component as the “world’s first and only piezoelectric MEMS microphone”. I don’t know if I would ever be so bold as to describe any technological development as the “the world’s first and only”, at least not in this modern age of constant and universal innovation. But I am in no position to cast doubt on this assertion; the folks at PUI Audio surely know much more than I do about the current state of microphone technology.
I discussed MEMS microphones in a previous article. Here’s a quick recap:
Electret microphones have a capacitive element that is sensitive to the air-pressure variations that we call sound. This capacitive element has constant charge, and its capacitance varies in response to sound. The result is that air-pressure audio signals are converted into voltage audio signals. The popularity of electret microphones indicates that this is overall a good approach—simple, cost-effective, and generally adequate in terms of performance.
MEMS microphones also use a capacitive transducer element, but it’s a MEMS transducer.
This MEMS approach allows for smaller devices and higher levels of integration. As you can see in the diagram, the microphone is right there next to the signal-processing IC. The result is a microphone device that can output buffered (or amplified) analog audio or even digital audio data.
Piezo vs. Ceramic
We know that MEMS microphones themselves are not particularly new, so what exactly is this innovation that allows PUI to make the “world’s first and only” claim?
Apparently, every other MEMS microphone uses a ceramic transducer element. I don’t think that there’s anything inherently wrong with ceramic transducers, but a complication does arise when you consider that the MEMS microphone cannot be a sealed environment. The package must include a gap that allows sound waves to enter, and the sound waves will be accompanied by those two ubiquitous aspects of the physical world: moisture and dust.
Capacitive transducer elements are not particularly robust against physical contaminants. The problem is conveyed by this screenshot taken from one of two informational animations provided by PUI Audio.
Here we have a ceramic-transducer-based MEMS microphone. Dust particles have accumulated to the point at which they are obstructing the mic’s response to sound waves. Image taken from an animation on the PUI website.
The standard solution to this dust-and-moisture susceptibility is a physical barrier, i.e., a membrane of some kind that covers the sound hole. According to PUI, the membrane comes with three disadvantages:
- higher cost
- reduced sensitivity
- altered frequency response
The piezoelectric approach results in a microphone that is inherently resistant to the effects of dust and moisture, and consequently the membrane becomes unnecessary.
The piezoelectric structure doesn’t prevent dust accumulation; rather, it allows the transducer to function more or less normally despite the presence of dust.
Piezoelectricity is a well-understood and widely utilized phenomenon, and I have not the slightest clue as to why it took so long for someone to develop a piezoelectric MEMS microphone. Obviously, there are some serious complexities involved, and it’s not surprising that the documentation for the PMM-3738-VM1000-R doesn’t elaborate on the technological breakthroughs that enabled PUI to develop a mic that (according to the datasheet) “provides superior performance and quality in all environments”.
If you want to incorporate the PMM-3738-VM1000-R into one of your designs, here’s the long list of additional components that you will need:
Seriously, though, this one qualifies for the “best typical application circuit of all time” award:
Diagram taken from the datasheet (PDF)
PUI’s product information indicates that the PMM-3738-VM1000-R offers a variety of high-performance characteristics: stable performance despite environmental variations, low noise, high dynamic range, fast startup time. I appreciate the uniformity of the frequency response over a fairly wide band:
However, I compared this response plot to that of one other MEMS mic and they look very similar. So apparently this is not something that is unique to the PMM-3738-VM1000-R or to piezoelectric technology in general.
One little cautionary note before we finish up: The PMM-3738-VM1000-R is robust against dust and moisture, but it seems rather sensitive to other forms of stress. See page 6 in the datasheet for details.
If you have any insights into the apparent difficulty of fabricating a piezoelectric MEMS microphone, feel free to share your knowledge via a comment.