Amplify Microphone - Car Stereo

I've got an existing car stereo that works well, except the microphone on it is pretty dull.

Hoping to fix the issue, I removed it's onboard microphone (positive, negative wires to it) and added a new one - however this one seemed to perform worse.
I got the meter out and saw 0.4Vdc on it's output.

It's data says 1.5 to operate, so I assume that the 0.4vdc is too low.

Given I haven't done anything to the signal conditioning on board (when the microphone was removed from circuit it was reading 1.8Vdc) - can I do something off board and feed in a signal that is amplified?

i.e. could I just take the positive from the board, and off board, add a 3.3v feed into the output, to simply amplify or is it not that simple?
I've taken the +/- out of the case so I can remote mount it too, but I want that amplified a bit :slight_smile:

Presumably an electret mic insert with built-in FET? Google shows the normal sort of circuit for biasing
these devices. You can add extra amplification, but you'll have to investigate what voltage rails
are available - if reasonable then a microphone breakout board (electret + opamp + gain pot) might
be usable.

Thanks Mark,
It is an electret mic.

Am I right in thinking that if I took this:

And then took the audio terminal and ground terminal to the wires that previously went to the microphone + and -, leaving the existing circuit inside the unit alone, and adding the above where the microphone used to be, it would work, without changing anything on the board?

If you have the schematic for the car stereo it would be easier to show where the new breakout board should be attached.

Well I used the wires that used to go from the mic, and added a preamp circuit to it - this one:

That didn't solve it, and I read that swapping the 100k out for a larger value increases sensitivity - I took it to 560k- still not solved.

The situation was better with the preamp, but impossible to use for a conversation.
It's using 3.3v, so I used a 5V regulator and divided down to 3.4v.


I've got an existing car stereo that works well, except the microphone on it is pretty dull.

What car stereo has a microphone?
Can you post a picture and the model and make of the stereo please.

Sorry for my ignorance.
Tom.... :slight_smile:

All the ones that have bluetooth built in for phone calls?!

The one I have is similar to that, the base functionality is good enough, GPS works, the radio and MP3 functions are reasonable.
The mic though is crap.

As is the case with much of the items out of china, they are good at 2 of the 3 things they are meant for.

I got it for a steal, but the mic being poor defies the point of buying it (so that I can use phone hands free).

For use in a vehicle noise-cancellation is important, so a separate noise-cancelling headset might be
a good way to go?

The microphone might be useless because it has no noise-cancelling.

You need to compare it to another unit of same brand and model.

If noise cancelling is used, there will be a "noise" microphone somewhere to provide the cancelling signals.

Mounting of the radio unit will also cause problems if not firmly mounted, causing noise mechanically conducted through the radio chassis to enter the microphone.

Just some thoughts, Tom.... :slight_smile:
I note the "just good enough" reference to its performance, this may be its "not quite good enough" part of the performance spec.

Noise cancelling is interesting, would they seriously be that stupid though?

The sound is only picked up when the microphone is shouted at, and if less then shouting but louder than talking, it is only intermittent - i.e. the first syllable of a word is picked up then it just is silent..

And that was like that before I opened it up in an attempt to fix it (hasn't fixed the issue yet).

Will increasing the voltage matter? At the moment it's at 3.4, if I went to 5, would the stereo likely be affected, or do the caps remove that entirely?
If I went to 5, would it make it more sensitive?

I tried taking out their amplifier - but same result, so the problem must be further along or in the software.

I went digging for anything that resembled a mic, but there doesn't seem to be anything, given they used two wires to run the mic to the front of the unit, I would expect the same style of mic used elsewhere if they were going to do noise cancellation (its china, come on, as if noise cancellation was ever going to be there).

The cap I took out to break that circuit was 10uF - fairly high! But that made no difference, so after that cap, it went to a ribbon cable which links the face PCB to the main PCB - I doubt I'll see any signal processing on the main PCB (why have the amp part on the face, and then more on the main, makes no sense).

The audio when heard, is faint. It is also only picking up odd words out of spoken sentences.

I did some testing with the PCB itself, the multimeter registers loud sounds easily - I can see it jump to as high as 0.4V - when I blow right into the microphone at close range.

It idles at 0.003V. Playing some music 30mm away from the mic shows multimeter registering AC 0.010V.
Same music, no background noise, and it sits at 0.003V...

So I did the same test with the + leg to the stereo cut (i.e the preamp was still supplying power to the mic, the wire after the final cap on the output, which goes back to the stereo, was cut off to ensure there was nothing in the stereo loading it down.

Same result as above.

So If I think about this, the microphone is either not sensitive to picking it up (despite trying two!), or, we're not getting enough signal output from the circuit to the stereo (or multimeter in the above).

I realise it's not an oscilloscope, but, it did register the voltage change..
What voltage do I need on the output? I think what is there is too low, am I right?

By going from the 5v voltage divider to direct 5v off the regulator, and using a 1k instead of 10k resistor, audio seems better, its not great yet, but its better.

I changed the 100k resistor from the 5V supply to 560k some time back, so I'm going to experiment going back to 100k and then if all still seems poor, changing the resistor to the mic to 2.2k instead of 1k.. See if results get better or worse and I'll increase the cap on the board as there's DC on the output still I think.

The answer to this problem is - the crappy chinese stereo had a crappy microphone - not sensitive enough.
The replacement Mic I bought was a crappy Jaycar microphone, but also not sensitive enough (-69dB)
The one I bought from RS-online and wired in to an identical circuit (I did go back to 100k) works perfect.
It’s -42dB…

Easy answer - buy a SENSITIVE mic :slight_smile:

I reckon the easy solution to those who might come across this with a chinese stereo is to take out the mic the unit ships with, buy a sensitive one and stick that in it’s place, it likely would work.

In my case I had already taken removed the capacitor that filters DC power from the head unit PCB (thus ensuring the head unit wasn’t interference with the signal from my outside PCB).