High Noise levels - electret mic - Android Head Unit mod

Thread Starter

extrememod2020

Joined Nov 30, 2021
20
Hi all,

There's a known issue with Android Head units and their poor mics so I've built this circuit using the template here. It's a mic preamp aimed at simple (external) car mics you can buy on AliExpress, eBay etc.

Capture.PNG

The amplification works great but the high-frequency noise levels are high. 50% ratio while driving but not talking. 20% ratio if drivng and talking. It sounds a lot like interference, rather than actual noise?

What can be done to reduce this?

The 12V comes from the Android Head unit that in turn gets it from the car battery. I'm using 2 variable resistors as I need to have the flexibility to adjust the resistors for different types of electret mics.

electret mic.PNG
 

DickCappels

Joined Aug 21, 2008
10,152
Just a guess: Try a lower gain amplifier stage or two. The old TL071 is pretty good for noise but the fixed gain appears to be 60 db. That does not leave much loop gain at higher audio frequencies where you are having noise issues.
 

Audioguru again

Joined Oct 21, 2019
6,673
Is the high frequency noise, Hiss? Or is the noise, a whining sound from the car's alternator?
Is the high frequency noise, Clipping Distortion because the preamp output is too high?

The gain of 150k/150 ohms= 1000 is much too high. The maximum gain needed is 200.
The cable from the electret mic to the preamp MUST be a shielded audio cable where the shield blocks interference picked up by ordinary wires acting like an antenna.
 

Thread Starter

extrememod2020

Joined Nov 30, 2021
20
Is the high frequency noise, Hiss? Or is the noise, a whining sound from the car's alternator?
Is the high frequency noise, Clipping Distortion because the preamp output is too high?

The gain of 150k/150 ohms= 1000 is much too high. The maximum gain needed is 200.
The cable from the electret mic to the preamp MUST be a shielded audio cable where the shield blocks interference picked up by ordinary wires acting like an antenna.
The testing conditions are on the desktop, so it sounds like white noise not a whine. The noise is constant, regardless of resistor values on P1 and P2. I found these values to give the optimal audio level without distortion:

P1 = 700 Ohm
P2 = 1500 Ohm

The noise levels reduce considerably as the louder the voice gets, it's almost like it's being 'suppressed'. But then it regains strength when no audio comes into the mic.

In testing conditions, the mic was moved waaay outside of the head unit but I have no shielding on the circuit itself. Feels like interference (like you have on the radio, if no station is found). Open to suggestions. Module attached.

Thank you guys


20230313_201736.jpg
 
Last edited:

LowQCab

Joined Nov 6, 2012
4,026
There may be an Automatic-Gain-Control, ( AGC ), Circuit that may be causing this problem.

The AGC Circuit "should" also have a "Noise-Gate" Function, but maybe it doesn't.

Maybe You need to really BLAST the Input to push the Noise-Floor down.

More Mic-Output amplification is always better, ( until You run into Clipping-Distortion ).

The Gain of the Mic Pre-Amp must be divided between 2, or more, Amplification-Stages,
or You are almost guaranteed to have excessive Noise.
This also creates more opportunities for "building-in" both High-Pass, and Low-Pass, Filters.
.
.
.
 

Audioguru again

Joined Oct 21, 2019
6,673
With P1 set to 700 ohms and R4 is 150 ohms, then the 150k negative feedback resistor creates a gain of 176 times which is reasonable if you are not screaming into the mic.

A TL071 audio opamp is American and has fairly low noise but yours is from China so it is probably a noisy fake.

You described "audio compression" that reduces the gain with loud signals then slowly boosts the gain back up to normal with lower signal levels. The simple preamp schematic does not have any compression but your head unit might have it.
 

Thread Starter

extrememod2020

Joined Nov 30, 2021
20
With P1 set to 700 ohms and R4 is 150 ohms, then the 150k negative feedback resistor creates a gain of 176 times which is reasonable if you are not screaming into the mic.

A TL071 audio opamp is American and has fairly low noise but yours is from China so it is probably a noisy fake.

You described "audio compression" that reduces the gain with loud signals then slowly boosts the gain back up to normal with lower signal levels. The simple preamp schematic does not have any compression but your head unit might have it.
That is what I suspect as well. As for the TL071, it's ordered from LCSC Electronics (China) but I have an option of getting one from a European site like Digikeys or CPC Farnell.

Thank you all for the replies. It sounds like I need to test with a real TL071 and then look further into the mobo of the head unit.
 
Top