Abate the noise :lots of noise in the microphone amplifier circuit when idle

Status
Not open for further replies.

Sunshine87

Newbie
Joined
Jul 1, 2016
Messages
3
Location
HongKong
I come here to get help from the experts.
I am using this BJT transistor in the circuit.
2N3904 - Central Semiconductor Corp - Transistors (BJT) - Single - Kynix Semiconductor Hong Kong Limited.
Here is the microphone that I am hooking up in the circuit.
It's an electret condenser microphone from kynix.
http://www.cui.com/product/resource/cma-4544pf-w.pdf
Here is my design
i am powering the circuit directly from 5v arduino pins.
I am following the tutorial from this link to build the amplifier. It's exactly the same circuit but with different values to make it work with microcontroller. For that, I have used the supply voltage of 5V instead of 9V as in the tutorial. I have also tweaked the values of resistors to make the correct biased voltage.
http://www.youtube.com/watch?v=Br3uOM0IVjc

The problem is that I get a lot of noise. I checked with oscillosope and it is biased correctly, i.e., midway between the 5V. But you can see a very noisy signal. I hooked the circuit to a computer speaker and you can hear the noise after you hook up the circuit. The output works correctly though, i.e, when you play a sound in the microphone you can hear clearly in the speaker but during idle there is a noise that is audible.
What is the reason of this noise and how to minimize it while the microphone is sitting idle?
I am thinking will the noise coming from any of the components in the circuit ? That's my hypothesis .If it was wrong, please ignore it. Thank you for the help.
Thanks in advance! I really appreciate any help!
 

Attachments

  • GFWhC.png
    GFWhC.png
    7.8 KB · Views: 105

prcguy

Member
Premium Subscriber
Joined
Jun 30, 2006
Messages
15,796
Location
So Cal - Richardson, TX - Tewksbury, MA
I agree the noise is probably from the 5v source. You can try isolating and connecting all the 5v source points together and bypassing them with something like 1000uf capacitor then maybe a 100 ohm resistor in series back to the 5V source.

This will make a 6dB per octave RC filter with a knee at around 1.6Hz with about 17dB roll off at 10Hz, 36dB at 100Hz, about 55dB at 1KHz and about 75dB at 10KHz. More capacitance will roll off more noise and so will more series resistance. If you change the values to a 10 ohm series resistor and 10,000 ohm capacitor you would get the same results. Since there is very little current used in the mic preamp, the 100 ohm series resistor should not affect the operating voltage or bias very much in the circuit.
prcguy


I would expect noise to be on the 5V supply line. Try using a series inductor (a coil) and a capacitor to ground: https://upload.wikimedia.org/wikipedia/commons/e/eb/Lowpass_Filter_LC.svg
 
Last edited:

cmdrwill

Member
Joined
Mar 30, 2005
Messages
3,984
Location
So Cali
There should be decoupling on the 5 volt ling at the mic/amp circuit. 100 ohm series resistor and a 100 uF to ground.

One could do a simple test, disconnect the 2.2k resistor to the mic element from the 5 volts, does the noise go away, or get reduced?
 
Status
Not open for further replies.
Top