Measuring the Big Impact Vibration Has on Tiny Microphones

The tiny microphones used inside hearing aids can be very sensitive to vibration of the device, resulting in annoying feedback. Testing how sensitive these microphones are to vibration has been a problem that plagued engineers. In this episode, we talk to Charles King and Chris Monti of Knowles Electronics about their innovation to measure microphone vibration sensitivity.

Associated paper: Charles B. King and Chris Monti, "Microphone vibration sensitivity: What it is, why it is important, and how to measure it," Proc. Mtgs. Acoust. 50, 065001 (2022). https://doi.org/10.1121/2.0001702

Read more from Proceedings of Meetings on Acoustics (POMA).

Learn more about Acoustical Society of America Publications.

 
Music Credit: Min 2019 by minwbu from Pixabay. https://pixabay.com/?utm_source=link-attribution&utm_medium=referral&utm_campaign=music&utm_content=1022

Transcript

Copy Transcript Copied

Kat Setzer  00:06

Welcome to Across Acoustics, the official podcast of the Acoustical Society of America publications office. On this podcast, we will highlight research from our four publications. I'm your host Kat Setzer, editorial associate for the ASA.

Kat Setzer  00:25

Today we're going to talk about something very tiny that has a big impact: the vibration in miniature microphones used in hearing aids. I'm talking with Charles King and Chris Monti, whose article, "Microphone vibration sensitivity: What it is, why it is important, and how to measure it," appeared in the 50th volume of POMA and is based off a talk they gave at the 183rd ASA meeting in Nashville, Tennessee. Thanks for taking the time to speak with me today, guys. How are you?

Charles King  00:49

Great. It's good to be here. Good to talk with you.

Chris Monti  00:53

Yeah, great to be here. Thanks for having us.

Kat Setzer  00:56

Yeah, thanks for being here. So first, tell us a bit about your research backgrounds.

Charles King  01:01

For last 30 years, I've been doing some form of specialty transducer. I started doing air raid sirens and police car sirens. And then at Knowles, I've been doing high-fidelity, miniature microphones and loudspeakers. These are mostly used in hearing aids. But we also have a division that does a bunch of cell phone work, and Chris and I have also done some specialty transducers for cell phones as well.

Chris Monti  01:24

Hi, this is Chris. So thanks again for having us. So I started my career in  aerospace and defense, did that for about 10 years. That's a little bit different than being here at Knowles. I've been at Knowles a little over 10 years, and been working with Charles, actually, the entire time, I've been here, doing all kinds of new and innovative products around audio, including hearing aids, earbuds, and mobile phones. More recently, we've also been trying to expand all our capabilities into other markets, including medical technologies.

Kat Setzer  02:01

Very, very cool. So your work here is specifically with regards to microphones in hearing aids. Can you tell us a bit about these microphones? How do they work, and are there any particular considerations that need to be made in comparison to other microphones?

Charles King  02:15

So microphones basically are something that turns sound into electricity, you know. They're super sensitive, so the tiny pressure vibrations you hear in the air can move a diaphragm. And that diaphragm movement, in turn, is turned into electricity, which is processed by the, you know, rest of the audio chain. You know, for the microphones we're most talking about here, we're talking about MEMS capacitor microphones. These are physically really, really, really, really, you know, tiny pieces of silicon that make a little capacitor. And as the pressure hits the microphone, it moves it and it changes that capacitance, and that's what's electrically measured and sent down. And even though these diaphragms are really, really small, they do have mass, and they do have tiny pockets of air, and both of these respond to acceleration. So when we talk about these microphones... What I'll do here is I'll tap on my microphone, [taps microphone], which is one thing you're never supposed to do in a podcast, because it just ruins it. Podcast 101 says you don't tap it like that. But the reason I'm tapping it here is that you can hear something that sounds strange. And what we tried to describe in the Nashville paper is to say how much of that sound is coming through the vibration of the microphone, moving the mass parts, and how much of it is acoustic, just normal transmission of pressure waves, like when you're hearing my voice. And the primary reason hearing aids are concerned about this is they really want to understand and control feedback, so that's why they're interested in both mechanisms.

Kat Setzer  04:06

Okay, got it, for like that high-pitched squealing sound that you get in the hearing aids. 

Charles King  04:11

Right. Exactly, exactly.

Kat Setzer  04:14

Okay, and I guess actually, good segue here. What is feedback? And how does it impact hearing aid microphones? 

Chris Monti  04:20

So I think most people are probably most familiar with feedback from being at a concert or music venue. And really, the mechanism is very similar to hearing aid. So in a concert, there are speakers playing sound back, what they call, "monitor speakers," back at the musicians so they can hear themselves, but that same sound is being picked up by the microphone and amplified, right? That's the point of the musicians microphone is to amplify their voice so everybody can hear them. But when the speakers play back at them, picked up by the microphone again, amplified again... Even though the musician may not be singing anymore, that same sound just gets back in a positive feedback loop. And that's why it's called feedback. And it's usually kind of highlighted at a high-pitch frequency that squeals. And it's very annoying for everybody in the concert. It's usually quite loud. 

Chris Monti  05:12

The same thing exists in hearing aids. Maybe you've seen the movie, the Pixar movie Up. If you haven't seen it, you should go watch. It's a great movie. But there's a scene, I believe the dog barks, you know. And so there's a loud sound, the speaker in in the main character's ear amplifies that sound, the sound is loud enough that from the speaker, gets picked up by the microphone again, and it gets in that same kind of feedback loop. And this is really one of the biggest complaints of hearing aid users, and audiologists that prescribe those hearing aids do not like to hear their patients complain. And then subsequently, our customers, the hearing aid manufacturers, they want to make sales and audiologists are going to use the hearing aids that have the least amount of feedback. So feedback is very important in the hearing aid market and minimizing that feedback.

Kat Setzer  06:05

Right, right. Totally makes sense. Yeah. So what is microphone vibration sensitivity? And why were you interested in studying it?

Charles King  06:14

Microphone sensitivity is how efficiently sound is turned into electricity. Right? So we think about that in the in the units of volts per Pascal. Now, we also mentioned that when you shake the microphone, it also responds. So that is in the units of volts per meter per second squared, or volts per acceleration. And since we're here on Earth, we think about that in terms of volts per g, or the equivalent acceleration of being an Earth. And it's just kind of a convenient unit system that allows us to numericize these quantities, but it's basically how efficiently energy can get from sound or movement to electricity.

Chris Monti  06:58

Yeah, and in general, right, the ideal case would be that the microphones actually aren't sensitive to vibration, they're there to sense acoustic pressure. But unfortunately, they do have a little bit of mass and they do sense some acceleration as well. And the issue is, you know, I kind of described the traditional acoustic feedback path that people think of, but there are other feedback paths as well, and one of them is vibration. So when our speakers or any speaker, pretty much all the speakers on the market do move something, and as a result, they create some vibration. And if the microphone picks that up, you know, it doesn't really know what it's sensing, it doesn't know if it's sensing acoustic pressure or if the microphone's being vibrated; it generates a voltage. And so just like sound from the speaker can feedback through a pressure wave to the microphone, vibration from the speaker can look like a sound signal and create a similar kind of feedback loop. They both contribute in similar ways. And ultimately, what we want to do is be able to give this sensitivity metric to our customers, so they can compare products and understand how our products perform and even run simulation models based on the numerical measures we measure and the data we provide them.

Charles King  08:15

Yeah, and something I'll kind of add in that process, too, is that, you know, our customers want to solve those. They don't want that feedback. And so they want to know whether they're trying to solve a vibration problem, or an acoustic problem, because the ways you might do that are different.

Kat Setzer  08:32

Right. Right. That makes a lot of sense. Yeah. And so it's really sounds like it's kind of the trick of, well, like you said, there is acoustic feedback, right, but you're trying to limit the vibration sensitivity, while still having acoustic sensitivity to some extent, right?

Charles King  08:49

Yeah, I mean, ideally, you would want ideally, you want to have no vibration sensitivity, right? But remember, in hearing aids, and they're small, right? Think about a hearing aid that fits in your ear, right? So in that hearing aid, you have a microphone, you have a loudspeaker. You have a centimeter between them. And you have lots and lots and lots of gain, because some people need lots of amplification to actually hear. So like when Chris was talking in the the stage example before, you know, you always see it and you see the singer walk towards another speaker and as he gets closer, the feedback starts. In the hearing aid example, they're close to start with, and there's a lot of amplification. So they really want to understand how it's getting into the microphone so that they can control it and reduce the risk of that occurring.

Kat Setzer  09:48

Okay. Okay. So then what is intrinsic vibration sensitivity?

Charles King  09:52

But what we're interested in is the intrinsic vibration sensitivity. And that's slightly different. So if we think about the gross free field, we talked about, there's two ways sound pressure or vibration are turned into electricity, right? There's two things that cause this to... When you have a microphone, regardless of how small, if you're shaking it, it acts like a little speaker, right? It's moving and it's pushing the air. So it's creating an acoustic pressure. In addition to all of the mass and stuff that's moving inside the microphone, that's creating a vibration. So the intrinsic vibration sensitivity is the sensitivity if there is no sound pressure level, right? 

Charles King  09:52

Well, it's actually a pretty old term. Mead Killion, who used to work at Knowles, started Etymotic and is also an ASA fellow, he coined the term back in something like 1975. Right? He coined two terms, right, the first one is easier to visualize, it's what he called the gross free field vibration sensitivity. Right? And basically what that is, is if you take a microphone and you put it on the tip of a tiny stick, so basically just the microphone is there, and you shake it, the electricity that comes out is what he called the gross free field vibration sensitivity. 

Charles King  11:25

So we kind of defined it in the paper somewhat technically, and I'll kind of read this. It's "the input referred output of the microphone, subject to the vibration when the input port has zero acoustic pressure, and an absolute pressure of one atmosphere." And that's a long gangly kind of sentence, but we can kind of parse it out. So "input referred," right, that's a term and what that means is that the output of the microphone is expressed in dB SPL. So it's the equivalent sound pressure that would generate the voltage that's coming out. And then when we say in that sentence that it's "subjected to vibration," we're saying, not surprisingly, that we shake the microphone, right? We're shaking it up and down. And as we said before, it's expressed to the acceleration of one G, or 9.8 meters per second squared. And the next one is "zero acoustic pressure," right? And that means there's no loudness, or it's perfectly quiet at the entrance to the microphone. The air inside is still shaking and making sound, but at the entrance of the microphone, there's no sound pressure at all, right?  And when we say it's at one atmosphere, we're saying it's in one atmosphere: it's in air, it's not in a vacuum. So if you sucked all the air out, you would get zero pressure. But you would also have zero atmosphere. So what we're saying is, we want to know, just... The intrinsic vibration sensitivity is the output of the microphone related to just the things inside of that microphone. So the diaphragm, the air volumes, it's all of the stuff inside the microphone that is producing an output.

Chris Monti  13:15

Yeah, and I'll add, you know, there's a real reason to want to do that, right, you have to draw the line somewhere is kind of where does the signal start. And many times in application, well, 100% of times in application, the microphone's integrated to a circuit board, and then there's usually maybe a gasket and then a hole in the housing of the device, whether it's a hearing aid or something else, so there ends up being kind of a tubular structure. But most of the surfaces of the microphone, it would make sense for us to include that surface as a radiating, acoustic speaker face that's moving as the microphone's accelerated, so we had to draw the line somewhere. And so we kind of define intrinsic as starting at that entrance surface, the outside of the microphone, and that little surface of the porthole. So that's kind of our zero plane. And that's how we define intrinsic is anything that happens inside of that. And then therein lies the real challenge is how do you create zero acoustic pressure and that's, you know, Charles will get into that in a minute. 

Chris Monti  14:18

You know, the other thing I wanted to point out is there is more than one vibration sensitivity, we kind of break it down in XYZ coordinates. The MEMS manufacturing the build and construction of most microphones, tends to be very kind of Cartesian. We think of the Z axis as being, kind of we talked about, the up-and-down direction or perpendicular to the diaphragm surface and the diaphragm sitting in the XY plane. And the diaphragm itself is not very sensitive to motions in X and Y. We basically kind of assume that's zero. It's really that z direction, like where it behaves like a drum head and has some mass. In the x and y directions, here are some opportunities to generate vibration sensitivity, and that's going to be dominated by asymmetries in the acoustic structure and because that air also has a little bit of mass and sloshes around due to acceleration, but for sure, I mean, usually the y direction, which is kind of the long direction of the microphones, usually one or two orders of magnitude less sensitive. And the x direction tends to be basically unmeasurable-- very, very, very low vibration sensitivity.

Kat Setzer  15:30

Okay. Okay. So what are the challenges to measuring intrinsic vibration sensitivity?

Charles King  15:35

The main difficulty is that the zero acoustic pressure at the face of the diaphragm is a very, very unnatural condition. I mean, any room you walk into, whether it be in an anechoic room or behind a jet engine, any place you are, there's always sound of some sort. So the intrinsic vibration sensitivity definition is broken for convenience of inside the microphone and outside the microphone. But zero at that face is very, very unnatural. So historically, what people have done is they've just measured the gross sensitivity that that Killion talked about, they measure the ambient sound pressure, and they subtract those two. And the hope is that, you know, if you measure what the microphone is doing, and you measure what's close to the microphone, that ideally you get what's left over, and what's left over is the vibration sensitivity. Right? And that makes sense.

Charles King  16:42

But practically speaking, if you have one big number, and you subtract another big number, that's practically almost the same. Even if those, each of those has very, very small error, what you're going to be left with is a smaller number with a lot of error. And so the historic subtraction methods, while they're popular and have been done a lot of late, are prone to error, just because there's a lot of potential noise going on during this measurement, right? I mean, you're shaking it, so it's on top of a shaker, the shaker is making noise, you can never have the measurement microphone really, really matched to the moving microphone because one is in a moving reference plane and the other is not. So knowing that they're actually the same sound pressure for the subtraction is very difficult. So just getting to this zero acoustic pressure at the port of the microphone is a very, very hard thing to do.

Chris Monti  17:47

And one thing, one approach, you know, Charles mentioned earlier, is putting the microphone on a long skinny stick. And that's how you can get it away from other sound producing things. So the shaker itself is making a lot of sound, it might have a hissing sound, you know, that might be fairly easy to remove. But it's also acting like a speaker cone. And so one thing we did before we developed this newer approach was we were really optimizing the design of that stick. It looks kind of like a pencil that the microphone's balanced on or glued to, and you're making out of carbon fiber and all kinds of things to try to make it as stiff as possible. Because what you want to do is you want the vibration to be controlled and clean, you don't want that pencil stick to be vibrating back and forth, sideways and easily excited in another direction. So it becomes a bit of an engineering challenge. The nice thing about this approach we're talking about today is it eliminated a lot of that. We're able to mount on a nice short, very stiff structure with resonances that are, you know, well above 10 kilohertz.

Kat Setzer  18:53

Okay. Okay. Before we get into the method you sort of discussed. You talked about the subtractive method for measuring intrinsic vibration sensitivity. Are there other methods? 

Chris Monti  19:06

Yeah so there are some other methods. Another one is, so Charles will kind of describe our method, which is to create a zero pressure using two microphones, but Mead talks about creating a small cavity and positioning the microphone perfectly on center inside that cavity. If you have a symmetric closed volume, the acoustic pressure at the middle is zero when it experiences acceleration. So if you position your microphone right on that center line, you're good. One issue is it's hard to position a microphone perfectly on the center line. The most common microphones today at least don't have the port at a center line. So just adding the microphone creates an asymmetry to the cavity. There's also challenges in making sure the structure supporting the microphone is very stiff. That structure has to be symmetric inside the cavity. But that is another approach. And it also has the advantage of shielding a microphone from outside sounds, but it has a lot of the challenges I just described. 

Charles King  20:12

Yeah, in practically, you know, for the microphone topologies that existed in the 70s, the Mead process where you create a zero pressure box that you put it in is pretty good. And Chris talked about it, you know, as a, you know, in terms of masses of air inside, cancelling each other out, another way you can think about it, is just think about it in the pressure domain where, you know, if you're shaking a box, the wall on the bottom of the box presses up on the air and creates a positive pressure, right? And the wall at the top of the box is pulling away from that air and creating a negative pressure. And it turns out that there's, you know, just like the rho-g-h in a column of water pressure that this creates, you know, a perfect gradient from top to bottom, in a perfectly symmetric box, where at the center, you get zero pressure. And so, you know, Mead's technique was to try to leverage that, to put the microphone in that zero pressure environment. 

Kat Setzer  21:22

Okay. Okay. So it seems like those other techniques were pretty problematic. How did you resolve these issues with your own technique?

Charles King  21:31

Well, so again, this zero pressure concept is really important. And we talked about the box example. It's easy to imagine this square box with zero pressure in the center, right? It's easy to visualize that in your head. But it turns out that any symmetric structure will give you that zero pressure environment, right? So if I take our microphone, and I take another copy of that microphone, and I put it on top of each other, and I glue those together, what I'm going to get is a perfectly symmetrical front cavity, because everything that's going on in the bottom half is now going to be mirrored exactly on the top half, because there's this other microphone placed on top of it. And because it's perfectly symmetrical, I'm going to get that zero acoustic pressure environment. And therefore, I can do the measurement directly once I've glued these two microphones together.

Kat Setzer  22:42

What's really cool. 

Chris Monti  22:43

All right, so yeah, I mean, just to continue with what Charles was saying, when Charles first kind of had, we had a eureka moment when Charles proposed this technique in a conference room; I could still remember where we were. And you know, I immediately was trying to sketch it on the board and, and realize that it wasn't just the acoustic symmetry, but even the diaphragm itself has a symmetry because as you accelerate in, let's say, upwards, both diaphragms, one is sort of out of phase of the other because geometrically, it's flipped upside down. So both diaphragms are kind of sagging due to acceleration in the same direction. And as a result, you're not actually changing the air cavity volume that's trapped between those two diaphragms. And you're not artificially kind of increasing the pressure. If you were to just close the port in between the two microphones, there'd be a pressure build up in that small cavity underneath the diaphragm as it was moving due to acceleration, it would build up pressure, and it would change the answer you get. But with the two diaphragms, basically, one is kind of creating a pressure relief for the other. So that's a really cool approach that works both acoustically and accommodates the geometric change of the diaphragm as it undergoes acceleration.

Charles King  24:07

Another really cool thing about physically gluing them together-- and just so the listeners understand when we say "physically glue them together," you know, we put them there very, very carefully. We fixture them, and then we just douse them with epoxy, right? So they're just, they're really glued together and they're glued on a really solid object, you know? And an advantage of that process, while it sounds kind of scientifically inept, practically what you're doing is you're creating a soundproof box around these microphones. And so now the shaker can be as loud as it wants, people can be talking in the room when they do the measurement, and sound can't leak in because these microphones are facing each other. They're in their own little anechoic world coated by, you know, an anechoic batch of epoxy. So that... At this point, two of the biggest problems of the previous measurement techniques, were, you know, making sure we had that zero pressure environment and then also making sure that it stayed zero pressure and no external sound from the shakers or other parts leak in. And by using the mirroring process and gluing them together, we get rid of the two biggest problems in the measurement in previous techniques. So we have a very, very clean measurement.

Chris Monti  25:31

Yeah, and when we design a little block I mentioned before, we, you know, we're no longer beholden to having it on a long stick, because these two microphones are sealed to each other. And so we're able to make that cube very, very short. You know, I'd love to see this little cube design kind of become the standard. It certainly has become our standard. It has nice flat surfaces, we can create little pockets for the microphones to make sure they're really well registered. And being a cube, we can mount it on any side. So once we've sort of sacrificed these microphones for the cause of measurement, they can then be, you can take this whole cube mounted on its side, and accelerate in a different direction, and then mount it on a third side. We even put in features to help make sure that this block doesn't rock very easily. We kind of distributed the pressure of the mounting around the perimeter to help with that. So, yeah, this cube design, it's simple. It's pretty cheap to make. It's machined aluminum, it's about an inch-size cube, and lets us measure as many as eight microphones at the same time. So it's really nice.

Kat Setzer  26:42

That's super cool. It's kind of like scientific arts and crafts.

Charles King  26:47

Yeah, no, it definitely is. They're they're pretty looking cubes, you know, they have a nice shape.

Kat Setzer  26:55

All right. Okay, so how did you use simulations for developing and validating your new technique? 

Charles King  27:02

Well, we certainly said that it's symmetric, right? And we said, look, it's a symmetric cavity, it should be zero. But our microphones are really, really tiny, right? The port of the microphone, you know, is measured in microns, not millimeters, it's you know, 200 microns or 300 microns or 400 microns, you know, so it's a, it's a very small opening, and our microphones being some of the smallest in the world, you're not going to put another microphone to measure and find out that it's zero there. So we use FEA and simulation to validate that it is symmetric. 

Kat Setzer  27:39

What is FEA? 

Charles King  27:40

FEA is a computational technique. There's lots of programs out there. For this particular one, we use COMSOL. But basically, it breaks the geometry, you know, when we're talking about the symmetry of this part, it breaks that complex shape into little blocks. And then it solves the equations for all those blocks. So it takes physics and applies it to really irregular geometries, and give you an answer of how that physical structure is now working, you know. So it takes shape boundary condition in physics, and gives you an answer of how it works. 

Charles King  28:15

And in this particular case, you know, one thing that we really wanted to check with the FEA is that the viscosity along the boundaries of the wall didn't affect the symmetry or zero pressure condition. We thought it shouldn't, computation shows it doesn't, and it really helps validate that zero pressure at the port, where we want it to be zero pressure is in fact, zero pressure. 

Chris Monti  28:42

And I'll add a couple of little details. I mean, basically, what we learned from simulating this approach was that really, you don't actually need simulation at all. It really showed that the zero pressure condition is maintained, there's no math you need to do. It's basically a tool for just validating our assumptions and our understanding. I mean, it is kind of nice, you can parse out the different contributors, which you can't do so easily in tests. So, very simplistically, there's a diaphragm. There's the front volume, the air between the port and diaphragm, and then there's the back volume air, and it lets you visualize the difference of those and how they're behaving. And it shows that in the x direction, you would predict zero sensitivity and the y, it shows that you're predict a small sensitivity. In the z direction it did confirm the highest sensitivity. So it's basically just helping us visualize and understand. In the other approaches, like the microphone on a stick approach, it becomes actually a necessary tool. And that technique, if you really want to do it very, very well actually requires assumptions and that your simulation is right. And your actual measurement isn't just a measurement. It's a measurement, while accounting for things that you can only derive from simulation, so it becomes sort of a not, a non-pure technique, if you really want to do it well. And that's just another advantage of this technique.

Chris Monti 

 Another place we use simulations was just on the block design and make sure it was very stiff. Again, this really was just a confirmation. It was much more needed when we had the stick design to engineer that solution. Our short little cue was very easy to engineer to accommodate all of our goals, which is just the high-frequency resonances of the cube itself, and then something that doesn't rock back and forth.

Kat Setzer  30:40

Okay. So you've talked a lot about the acoustics. How do you know that the vibration measurements were right?

Chris Monti  30:46

So that's a great question, because, you know, as Charles mentioned, it's volts per g. So we've isolated the acoustics-- we're measuring volts, we're measuring the signal coming out of the microphone. We also need to measure acceleration; these vibration tables typically have an integrated accelerometer, but kind of the gold standard and measuring acceleration is a laser vibrometer. So it's an optical technique, very, very accurate. I mean, they almost are sort of intrinsically calibrated. So we get a very accurate measurement. And you can even put the laser directly on the microphone. So if our cube did have an imperfect, sort of transfer function of one to one between the input acceleration and the acceleration of the microphone, it wouldn't matter to some extent, just because you're measuring directly on the product itself. So you can directly measure the acceleration, and between the voltage out of the microphone and the data out of the laser vibrometer, you can get that Pascals per acceleration,

Kat Setzer  31:47

How did your new method end up performing compared to the previous measurement techniques? 

Charles King  31:51

They're much much, much, much.... you can put little dots there... much, much, much... it's a lot, it's a lot better, right? The previous measurements were really, really noisy. And they made sense at one kilohertz. But at any other frequency, you would see things that didn't make sense in one way or the other. And now when we measure, we get just basically a straight line, and we can measure all the way down to 100 hertz, we can measure up to 10 kilohertz, and we get clean measurements, you know, from the lowest frequency to the highest frequency. And so at any range, you know, of hearing a designer might care about, we can now produce data that is correct for the vibration sensitivity.

Chris Monti 

And I'd add, there's two kinds of ways that the measurement looks cleaner. One is it's not noisy, so you used to look at frequency versus amplitude, and it was a squiggly line, and you kind of draw a line down the middle of that. And it was pretty close to predictions of what we thought it ought to be. But you know, so not only is it noisy, but also there's always this level of uncertainty, you know. Am I subtracting the right number? I'm doing so much math, am I doing it right? Are there other errors in the reference microphone? So you just never really knew if you're getting the right answer. So not only is this much cleaner, but we really have much more confidence. It's, you know, it's kind of you can't screw it up.

Charles King  33:32

Yeah, and because of this, you know, the randomness in this noise floor is so low, we can also measure at all axes, right? We can measure at its loudest axis, which is what people mostly care about. But we can also flip it on its side and measure on that axis and get meaningful numbers and meaningful numbers that generally match our simulations, so we think they're correct. And then we mentioned that there was one orientation where you should get nothing, right? And when we measure on that, in the orientation where we should get nothing, we get an answer that's, you know, 30 DB or more down from the highest value of the answer, right? And so that that's kind of a direct measurement of the noise floor and says that we're really working on you know, solid ground because when we measure nothing, we get pretty close to nothing. 

Kat Setzer  34:24

That's exciting!

Charles King  34:27

Yeah, yeah. You know, it's just, it's hard when you're developing a test, to know that you have the right answer. But if you can put an input in there where you know what the answer is, like, you know, it should be zero and you measure zero. That builds a lot of confidence.

Kat Setzer  34:45

Right? Right. Understandably. Yeah.

Chris Monti  34:48

And there's, we know that it's a good measurement, we know that the answer in x direction is functionally zero, because there are some very small errors in the measurement technique. And the problem is the microphone does have that high sensitivity in the Z axis. So when you rotate it on its side, if for example, the shaker moves a little bit in the z direction, that's going to create a voltage. And even though you know, that's not really the direction you're intending to shake the microphone, we've kind of quantified those sources of error, maybe a little angle error, and how you've mounted the microphone on the fixture, even if it's just one or two degrees, that sensitivity in the z direction is so much higher. Well, it's non-zero, is that you cannot tell the difference between a measurement in the x direction and a measurement error, basically. And so it's not even a number that we tend to even really report we just make sure it is in fact below that noise floor. And we say it's basically zero.

Kat Setzer  35:45

Got it. Got it. So what are the next steps in this research?

Charles King  35:48

Most of the work we're talking about right now, we presented at the Nashville conference. And we also presented a paper in Chicago, where we showed all the data that we'd measured between the time in Nashville and time in Chicago. So we showed maybe 50 or 100 different measurements on different styles of microphones, did more quantification of the air. And that now what we're working on is an article for JASA where we'll combine those papers and submit it for, you know, a full peer review to the journal.

Kat Setzer  36:22

I can't wait to see that one. Do you have any other closing thoughts?

Chris Monti  36:26

Yeah, I'll just add, I mean, our hope is that after, you know, after that article, we could just kind of close the book on this one and say that should be the standard in the industry. I mean, it's not often in engineering and science that you come up with a solution that's both easier and dramatically, more accurate. And, you know, we really think we've done that here. So it's, you know, it's always just so much easier for us. Our motivation is to have everybody speaking the same language. There's no ambiguity between us and our customers in terms of what the right answer is. We've really created a nice, clean standard measurement approach. And we think that, you know, having published in a world recognized peer-reviewed journal like JASA, will kind of help demonstrate to everybody in the field that it's the best approach to use. And beyond that, just wanted to thank you so much for having us on the podcast.

Charles King  37:17

Yeah, thank you.

Kat Setzer  37:18

Yeah, thank you so much for being here. It's really amazing that you guys were able to develop this method that's, like you said, it's so simple and so much more effective, it seems like. For our listeners, there are some very helpful diagrams in the article, although I think you guys painted a very good picture of what the setup is and what your little block is like. So we'll link to the article. Yeah. Thank you again, for taking the time to speak with me today. It was really great having you guys on.

Charles King  37:42

Oh, great. It's great to be here. Thanks. Thanks again for having us.

Chris Monti  37:46

Yeah, thank you.

Kat Setzer  37:47

Thank you for tuning into Across Acoustics. If you'd like to hear more interviews from our authors about their research, please subscribe and find us on your preferred podcast platform.