OK, and still, a separate feed, with additional filtering and separate power and common wires to the power supply should reduce the problem. Noise in common side connections is also sometimes the problem. Current in the common conductor is often the source of noise coupling.

 

Really, the first step would have been to disconnect the microphone, and then temporarily connect the two inputs directly to each other. Then you would know if the noise was from the input circuit or internal to the amplifier module.
And certainly using the unfiltered V+ for input microphone bias is a very poor choice.

 

The noise/interference comes from both power supply and microcontroller, and the amplifier is not very kind, too. C'est la vie.

 

One more thing is that the gain equation is incorrect. The gain is much higher for audio frequencies So with a lower gain the noise would be less of a problem. You do not have a resistor in series with the input so that formula is incorrect for audio gain.

 

Gain equation? I'm not calculating gain with any of these microphone amplifier boards. They either have a fixed gain, or gain you can set via a tiny trim potentiometer.

 

IF it happens that the gain is much to high, then every bit of internal noise will be amplified a lot more. I was commenting on the circuit that I see in post #13.
Gain will have a large effect on the level of noise present in the output, no matter where the noise comes from.

 

I was commenting on the circuit that I see in post #13.

I see, that's the Adafruit MAX4466 microphone breakout board circuit, one of the ready-made breakout boards I have tested. They're all not very good. The Sparkfun equivalent is a bit better, but not much.

 

Now a question: Why are you using the non-inverting input, instead of the inverting input? The way it is shown, noise at both inputs is added and amplified. And I looked at the traces in post #1 again, and that looks like semiconductor noise that has passed thru an amplifier with limited frequency response. How much noise do you get with the microphone disconnected??

 

In post #13, the resistance of R1 is too low and is "shorting" the output of the electret mic which reduces its output level requiring more preamp gain which produces more noise. The low capacitance of C1 passes mids and low frequency noise from Vcc.

The datasheet for the mic shows a good sensitivity when your R1 is 2.2k, but then the resistance of R2 is too high resulting in not enough DC voltage at the mic resulting in additional "sensitivity reduction" as shown in the datasheet.

 

I'm not using anything, I did not design that board, Adafruit did. I am using it exactly as it is, trying to ge the best out of it without physically changing either the electret microphone amplifier board or the microcontroller, then comparing it to others (Sparkfun, DF Robot, China higher price, China rock bottom price).

The set-up from post #20 yield the best result so far for the Adafruit product, while the Sparkfun product is much less noisy right out of the box, so to speak. I assume the Sparkfun EE is better, but maybe even their PCB could be improved (beyond my skill level to judge).

 

In post #13, the resistance of R1 is too low and is "shorting" the output of the electret mic which reduces its output level requiring more preamp gain which produces more noise. The low capacitance of C1 passes mids and low frequency noise from Vcc.

The datasheet for the mic shows a good sensitivity when your R1 is 2.2k, but then the resistance of R2 is too high resulting in not enough DC voltage at the mic resulting in additional "sensitivity reduction" as shown in the datasheet.

Well, maybe you want to give Adafruit a good spanking, so they re-design that thing properly ; ) or you could change the schematic from post #13 with the right parts. Maybe some more advanced hobbyists are equipped to tinker with these SMD parts?

 

A.G. Is right, and has a great deal of insight about audio amplifiers. His advice is quite valid.

 

A.G. Is right, and has a great deal of insight about audio amplifiers. His advice is quite valid.

I don't doubt that at all. I would hope Adafruit, Sparkfun, et al. would improve their parts.