How does the snr of the Analog Devices microphone compared with that of microphones using optical interference for membrane detection? See e.g. Sensors and Actuators 75, p257-268, (1999).
Thank you for your question. I looked through the paper you attached and I think that now I understand some basis for comparing the optical interference microphone described there to our MEMS microphones. The last two pages of this paper works through the noise performance of a microphone design example. Here, it states that the theoretical SNR for a normal speech level is "about 6". On page 9, the normal speech level is defined as approximately Ps/P0 = 3e-7, which is about 64 dB SPL (P0 = ambient pressure, 101,325 Pa). The noise sources in equations 84 and 85 combine for a total noise floor of Ps/P0 = 1.34e-7, which is about 57 dB SPL.
If we compare this 57 dB SPL noise floor to our MEMS microphones, then we need to use the term "SNR" in a common way. An SNR spec for microphones typically uses a well-defined reference pressure, rather than something more ambiguious, like "normal speech level". SNR as specified in our microphone data sheets is the difference between a 94 dB SPL (1 Pa) reference pressure and the noise floor. So, for this optical interference microphone, the SNR is 94 dB - 57 dB = 37 dB. If we look at the design below equation 85, the noise floor decreases to about 43 dB SPL, for an SNR of 51 dB. We have an application note (AN-1112, pdf link) that has a more detailed description of the SNR specification for MEMS microphones, as well as other specs found in a MEMS microphone data sheet.
Analog Devices' MEMS microphones have an SNR ranging from 61 to 65 dB, depending on the specific part. Our just-released ADMP504 and ADMP521 have an industry-leading 65 dB SNR, while our first generation of MEMS microphones (ADMP4xx parts) have a 61 or 62 dB SNR. So, all of our MEMS microphones have noise performance much better than the optical interference microphones as described in this paper. A full list of our MEMS microphones products is in the selection table at the bottom of this page.
I've heard of microphone technology using optical interference before this, but I'm not sure if anything like this is being commercially manufactured. Do you know if you can buy optical interference microphones now? Thanks again.
Thank you for the promptness of your response and the detail of your comparison. It would be interesting to tweak the various parameters for the optical-interference microphone to see if this type of device could reach or exceed an SNR of 65. The examples given in the paper were meant mainly to show how to use the noise relations. I did slap together a demo device with a Mylar membrane diameter of about 2.5mm to demonstrate the principle. Qualitatively at least, that simple device worked quite well. In case you are interested, I’ve attached some 3D plots for the interference microphone that are more informative than the 2D plots given in the paper. ( I do not know of any commercial interference microphones.)
Again, thank you.
Thanks! It's interesting to read a bit about this different microphone technology. It's not something I knew much about before this. If you're able to share some of the background about what you're doing, what about your application makes an optical interference microphone attractive rather than an electro-acoustic microphone?
You asked about my background: I am now retired and doing part time MEMS consulting work. In my previous position I had responsibility for seeking out novel applications of MEMS devices with commercial value. Over a period of ten years I worked on optical modulators and blockers, wavelength sorting switches, cross connect switches, scanning mirrors, adaptive optics mirrors, coupled mechanical resonators for narrow band rf filters, safety and arming devices for rockets and for high-g munitions, infrared detectors, multiphase reluctance motors, optical microphones, sound-direction detectors, vibrating beam magnetometers, in vivo blood pressure gauges, light-driven motors…
Because the wavelength of light is comparable to the thickness of the layers in a typical MEMS structure there is a very natural and intimate connection between the two, meaning in particular that an interferometer can be incorporated easily into many types of MEMS structures. Very small displacements of a membrane then can be related to relatively large changes in the intensity of reflected monochromatic light, changes that are easily measured using a simple photo detector. Miniature interference-type microphones are predicted to have a dynamic range comparable to the best conventional high fidelity microphones. For some applications the light source is mounted directly on the structure, for others it is connected to the structure via an optical fiber. In the latter case all of the associated electronics can be located a very large distance away from the sensor, making this arrangement especially useful in surveillance applications or in noisy environments. Note that with no metallic components there is no interaction with electric or magnetic fields.
Closely related to the fiber-optic microphone is an in-vivo heart monitor that I also would like to develop. The monitor is a very small interferometric pressure sensor mounted directly on the end of an optical fiber that is inserted into the heart during catheterization. The detected signal gives very detailed information about blood pressure and pulse rate, information currently not available using standard tools. In addition to its possible comparison against an ekg trace, the high bandwidth signal also can be filtered to give audio information about valve clatter or about the onset of turbulence in blood flow. In another application the pressure gauge and a similar gauge modified to function as a gas thermometer are incorporated into an endoscope to allow the optical image to be complimented with other critical diagnostic data, for example, the local temperature near the surface of a tumor. Unfortunately even though I had the enthusiastic support of head of cardiology at a top notch medical school, my previous employer blocked me from pursuing these ideas because of liability concerns.
I am currently searching for partners with funding and access to a testing laboratory—for the microphone, the heart monitor, and for a vibrating-beam magnetometer with microgauss sensitivity. Any suggestions?
Sorry, but I don't have any specific suggestions for finding funding and lab space. Good luck, though; it sounds like you have some interesting applications for your technology.
Retrieving data ...