I am working on a project to detect sounds with frequencies ranging from 3KHz to 14KHz and at approximately -100 dB. I want to record audio using a microphone connected to an Arduino uno and then process the recorded audio. Can you recommend microphones that I can use? Sensibility is very important in this project. I
-100dB is ambitious.... what's your budget?
Sound levels are normally measured with 0 dB being the threshold of human hearing, and 100 dB starts to be painfully loud.
On what do you base "-100 dB"?
Minus 100dB SPL doesn't exist on earth. 
+100dB SPL should be possible. I have a SparkFun BOB-12758 microphone board and by my estimates it should max-out (electrically clip) at about +104dB SPL.
Electrical & digital levels CAN be -100dB or less, depending on the dB reference.
the Adafruit MAX9814 Electret Microphone Amplifier is solid too (default to 'max gain' @ 60dB) and pretty cheap
Thank y'all for your answers. The budget is a maximum of $50, so even a bit more expensive option is also fine. My project replicates an experiment where multiple sounds in a house were captured and classified. The experiments in the literature shows sounds around -90 or -100dB in the spectogram, which were produced by some devices inside the house.
Given your claim to be a newly joined member, I do remember several posts with the same question recently.
Is this a class assignment or is it a new account for an old user?
I also note you are looking for a super sensible microphone. Do you really want that or a sensitive one, or maybe even a super stupid microphone?
1 Like
That digital dB level would depend on the sensitivity of the microphone and the gain of the preamp. (We don't know the dBSPL sound level.)
This would be better as a "computer project". But most tegular soundcards are noisy making -90dB impossible.
You'll probably need a pro microphone and an audio interface. (That will exceed your budget.)
0dB SPL is approximately the quietest sound humans can hear and SPL levels are positive. Digital dB levels are measured as dBFS (decibels full scale) which is the "digital maximum" so digital dB levels are usually negative.
0dBFS is the highest you can "count to" with a given number of bits. At 16-bits you can count-down to -96dBFS and below that is zero, which is -infinity dB. With 24-bits you can go-down to -144dBFS.
There is no standard calibration between SPL levels and digital levels but there is is direct correlation. i.e. If you lower the digital level by 3dB (a bigger negative number) the SPL level also drops by 3dB (a smaller positive number).
So... If you have an SPL meter you can calibrate your digital levels and add/subtract to convert the -90 or -100dB digital levels to positive SPL levels.
1 Like
Suppose that you buy a pro microphone with a S/N of 80dB and with a good audio interface, then I still don't see how this project can work.
The Arduino Uno R3 can blink a led and read buttons, but it is not good at sound processing. Not good at all.
If you want to process sounds up to 14kHz, then you better use a mini computer. For example a Raspberry Pi.
There are websites dedicated to capture nature sounds. They use microphones that can capture the smallest sounds.
Can you tell more about processing the recorded audio ? Is that on a laptop ?
2 Likes
I think a parabolic reflector would help a lot more for any microphone you try.
Yes, it is part of a school project (thesis). I want a good and sensible microphone, so I don't have problems because of the mic and I can focus on the processing part. The idea was to conect the mic to the arduino and save everything on a micro SD, then with the micro SD process in my computer.
However, I could consider use a Raspberry. I didn't think the frequency as an issue for the arduino uno. Thank you for the feedback.
Well, you will NOT be recording anything while the SD card processing running because it blocks your code until the SD card function has completed. Best to just send the data to the PC to begin with and let it store the data.
I think you need a sensitive microphone.
Please look up the difference between the words sensitive and sensible, they mean two completely different things.
so does your teacher know how difficult this is to do? The range is far too wide to achieve using affordable components. It is a matter of the digitation resolution required. You are being given an impossible assignment, no matter what platform you use.
You mentioned it’s for your “thesis” - is that a PhD level? If so what’s the main topic? Signal acquisition and processing is a science in itself (and sometimes an art).
1 Like
He did say:-
So it won't even be a undergraduate project. More like a project that will contribute towards University entry .
It's for a master thesis. In any case, I thank you all for the help and the feedback you provided. Have a good day!
For a Master's thesis, you appear to lack a great deal of the required technical background.
Posting on the Arduino forum is not the most efficient way of getting up to speed. Ask your instructors for recommendations on how to improve your understanding of the topic.
Then why call it a school when clearly it is post graduate?
I guess there is a language problem here which is why you keep not responding to being told that a "sensible microphone" is not the way to refer to a microphone. Are you using some sort of automatic web based translator? If so could you try another.
The opposite to "sensible" is "stupid". Where as the opposite of "sensitive" is "insensitive".
Maybe you could put what you post in English into your translator and see what comes out?
Even in 2024, the translation often fails. Don't take it too serious.
I'm thinking in this direction:
This topic was automatically closed 180 days after the last reply. New replies are no longer allowed.