Looking for microphone suggestions for recording mouth exhalations

Hello!

This is my first forum post so hopefully I'm posting in the correct discussion forum. Long story short, I'm looking for some microphone suggestions. I'm currently working on a project that is trying to record respiration with a small microphone positioned 1-2cm away from the mouth. So far I've tested two microphones with an Arduino Nano 33 IoT: The electret microphone amplifier Max 4466 from Adafruit and CUI Devices CMC-6015-47P electret condenser microphone.

Both of the above microphones work fairly well for detecting medium and strong mouth exhalations. However, I find that both struggle to record weaker exhalations that tend to be more quiet. Any advice when it comes to selecting small microphones good at recording breath/mouth-exhalations? Eventually the microphone breath sensor will be used with a VR game. I'm not an audio expert so any suggestions would be appreciated very much!

Thank you!

So. What you're really looking for is a way to record exhales, not necessarily with a mic.
The standard polygraph uses a couple of elastic bands to record the expansion and contraction of the lungs, and it is very accurate in that measurement.

POLYGRAPH Digital Polygraph-Physiograph | New Meditech | New Meditech

Yea, the bands around the chest thing is the way to go.

If the patient breathes into a tube, then a hot wire anemometer could be used.

I'm actually looking for a no-contact solution at the moment so the chest bands wouldn't work for my particular project. Ideally, I want to incorporate the breathing sensor onto a VR headset. I've also looked into the hot-wire anemometer solution but because it produces heat, it can be a bit if a hazard if placed too close to the mouth of the user (and the VR headset wearer). The microphone seemed like the easiest solution given the restraints but I'm open to other no-contact suggestions!

You could try a thermal sensor, but afaik those are the available options.

I tried using a 10k thermistor but I found that my breath added too much heat to the sensor. As a result of this added heat, short pauses between exhalations were not detected very well (i.e., the sensor did not fully cool down between breaths). I'll keep looking around for microphone solutions but thank you for the extra suggestions and the quick replies! Any help is appreciated!

Try adding a parabolic concentrator to your mic.
Got a 3D printer?

Oh that could be very interesting. I don't have a 3D printer but I can try to make a concentrator with other materials possibly. Thank you!

A small plastic shot glass or a miniature bottle lid would serve for proof of concept.

Great suggestions. I'll try that and will report back if it works!

Seems you have insufficient amplification for the microphones. Did you search the the highest gain in the associated amplifiers?

I'm not too familiar with microphones so insufficient amplification could definitely be the issue. I'm not sure how to determine the gain of the CUI device electret condenser microphone. I do know that the Max 4466 microphone has an adjustable gain pot however.

I see this in the specifications: The output is rail-to-rail so if the sounds get loud, the output can go up to 5Vpp!
Since the output is audio, or AC, how are you connecting it to the Arduino?

I currently just have the microphone connected directly to the Arduino using jumper wires. The connections are the following: (VCC--> 3.3V, GND --> GND, OUT--> A1).

I also just realized that the sensitivity/gain for the CUI device microphone is listed on the specification sheet. It's sensitivity range is approx -50dB to -44dB, typ is -47dB.

I see the specs also state :"The output will have a DC bias of VCC/2 so when it's perfectly quiet, the voltage will be a steady VCC/2 volt (it is DC coupled). ".
So with no sound, the A1 value should be about 512 or so. What are you getting with the mike covered?

I'm getting around the 512 value with no sound.

Good. Then with sound you will get values less than 512 and values greater than 512. That will tell you what to look for in the sound of breathing.

Sounds good. Thank you for your help!

Maybe a bigger electret? - 6mm electrets are pretty deaf, they are designed for close-up speech, not low level sounds. Find a microphone designed for room use rather than headset use.

You can find 25mm electret capsules, but they seem to come without the FET so cannot be used directly, alas.