Looking at the schematic of the X-NUCLEO-AMICAM1, I’m trying to understand why a quad op-amp is used for two microphones. Why isn’t there a single input and output for each microphone? What is this configuration called, and what are the advantages of using it?
I’d really appreciate any insights or explanations!
I think the circuit is not very well drawn as it takes too long to see which signal goes where and why. The worst issue is the opamp chip is drawn as a black box with 14 pins, so you have to decipher how it would look like if the op-amps were drawn as separate 3-pin op-amp symbols.
The mic signal is first amplified with a non-inverting op-amp cirvuit, with gain of 1+(56k/2k)=29, or 29dB, and them fed to ADC positive pin.
That positive signal is also taken into an inverting op-amp circuit to invert it with gain of 10k/10k, or 0dB, and fed to ADC negative pin.
Basically the mic signal is converted into two signals that are (a) differential, (b) balanced, and (c) also symmetric for the ADC input, because that's how the ADC input wants the signal.
Basically that's what goes around in XLR audio cables between audio gear.
In reality the terms are used quite liberally, but each term means a specific thing in the system.
Basically what you have is this:
simulate this circuit – Schematic created using CircuitLab
It's a non-inverting amplifier with a gain of $$ \frac{56k}{2k}+1 = 29 $$ followed by a unity gain inverting amplifier, this gives two outputs 180° out of phase, which turns the single ended signal from the microphone into a differential signal to drive the ADC inputs shown on page 3 of the datasheet available here.
The MEMS mic requires AC coupling to high R load and inverting gain to avoid positive feedback from high output knocks that with piezo microphonics cause a parasitic high audio frequency resonance.
MP23ABS1 https://www.st.com/resource/en/datasheet/mp23abs1.pdf
High performance MEMS audio sensor single ended analog bottom-port microphone.
The MP23ABS1 has an acoustic overload point of 130 dBSPL with a typical 64 dB signal-to-noise ratio. The sensitivity of the MP23ABS1 is -38 dBV ±1 dB @ 94 dBSPL, 1 kHz
Thus a good solution to maximize SNR at a level of 94 dBSPL 1kHz is generate near 1V with 39 dB gain biased at Vdd/2=COMM on -38 dBV+/-1 means the minimum output is 0 dBV =1V
To allow the use of just about any Op Amp bias current and avoid excessively high R values that might be affected by parasitic capacitance, R=56k is the load with Zin+ is high with series 1uF rolls off near 30 Hz.
So why does the datasheet recommend and inverting amp. with gain? Why does this board uses a 56k termination to a source MEMs with 2 stages 39 dB and -1 ?
Many reasons: #1 stable SNR to parasitic feedback for phase array beam forming microphones.
