Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

how to reduce LM358 preamp noise

Status
Not open for further replies.

ash1412

Member level 3
Joined
Nov 30, 2009
Messages
65
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,286
Activity points
1,693
Hi All,

I am using LM358 based preamp. I would like to know if there is a way to reduce noise after preamp? I am giving input through electret condenser mic then I preamp the signal. I got noise and also the signal voice. Can anyone help me how to remove/reduce the noise.

here's my ckt & electret mic data sheet.

Any suggestion would be highly appreciated.



Regards
Ashwani
 

Attachments

  • LM358.pdf
    9.3 KB · Views: 274
  • cz034a.pdf
    48.4 KB · Views: 250

R1 is too high in value (spec sheet is for 1 or 2.2 K). The junction of the two 10Ks that bias (R2,R5)the + inputs should be decoupled to earth with > 1MF cap. Your mic is a low impedance type, so C1 is far too low in value, try 1MF. I am not sure of the function of the buffer amp, with a low impedance source /40 impedance available at the first op-amp, what does the second IC do?
Frank
 

The second IC is an inverting opamp with a gain of about 100,000 because it is missing an input resistor. So of course it is noisy. It is also shorting the output of the first opamp.

Why would anybody use a lousy old LM358 for a mic preamp instead of using a low noise audio dual opamp like a TL072?
 
yes. I modified the ckt but noise is still there. audio quality is not good.
I am waiting for low noise opamp sample pieces.....
 

Why didn't you post your corrected schematic?
In addition to noise (hiss), the LM358 has horrible crossover distortion because it is low power and its output transistors do not have enough idle current.

You know how a sine-wave looks perfectly smooth? Here is the sine-wave at the output of an LM324 (look at the crossover distortion).
An LM324 is the quad version of the dual LM358.
 

Attachments

  • LM324 crossover distortion4.PNG
    LM324 crossover distortion4.PNG
    52.3 KB · Views: 160

Why didn't you post your corrected schematic?

Today I modified the schematic & test the results. I made the following changes in sch:
i > set for both feedback resistors to 100 K
ii> input resistors to 1K in second op-amp.

You know how a sine-wave looks perfectly smooth? Here is the sine-wave at the output of an LM324 (look at the crossover distortion).
An LM324 is the quad version of the dual LM358.

Thanks for your valuable information.

I am looking for best audio quality op-amp.
 

Today I modified the schematic & test the results. I made the following changes in sch:
i > set for both feedback resistors to 100 K
ii> input resistors to 1K in second op-amp.
Then the total gain is 10,000? It is too much.
Again, you did not post your modified schematic.
How are the results?

I am looking for best audio quality op-amp.
An OPA134 opamp and the OPA2134 dual have very low noise, distortion of only 0.00008%, a full output bandwidth to 230kHz (the LM358 barely goes to only 2kHz) and works with a supply as low as 4V.
 

here is my modified sch.



noise level little bit reduce.

---------- Post added at 18:08 ---------- Previous post was at 16:32 ----------

Is there any design or IC for acoustic echo cancellation?
 

Your 741 opamps were designed 43 years ago and have the same amount of noise as an LM358. They are never used as a preamp. You should use a low noise audio opamp instead.
In your new circuit each opamp has a gain of 100 so their total gain is 10,000 which is too much. They amplify their own hiss too much.
The output is missing a coupling capacitor.
The 10k resistors to ground at the output of each opamp do nothing and should be removed.

The +9V should have a 100uf capacitor to ground and the point where the (+) inputs of the opamps join should have a 10uf capacitor to ground.

Cirrus Logic makes a very complicated Full Duplex speakerphone IC with echo cancelling. It is the CS6422 and its datasheet is too big (a PDF with 48 pages) to upload here. I think I got it from Datasheet & Application Note Database, PDF, Circuits, Datasheets | Datasheet Archive .
 
Thanks for your detail informations.

I have drop this design. your posted design on https://www.edaboard.com/threads/208437/ is better and simple.
I have checked with TL071 IC audio quality is good but getting noise.

If you look lifesize mic, they are using LM833 amplifier and LM2903 comparator and the audio quality is excellent. I think LM2903 using for echo cancellation.

After receiving the LM833 & NE5534 samples from TI, I will check with your preamp design and getback to you all results.
 

I have checked with TL071 IC audio quality is good but getting noise.
Please do not say the word "noise" without describing the noise.
Dogs barking?
Helicopters?
Wife screaming?
Hum?
Hiss?
The TL071 is inexpensive so its noise level is not as low as a much more expensive OPA134 opamp.

If you look lifesize mic, they are using LM833 amplifier and LM2903 comparator and the audio quality is excellent. I think LM2903 using for echo cancellation.
No.
An LM2903 is a simple and old analog dual comparator. An echo canceller is much more complicated and is digital because it digitizes audio, uses a digital memory to store a copy of room echoes and has many software audio filters. Then it converts the processed digital back to analog audio. An echo canceller has a program to run it.
The video codec probably has the audio echo canceller.
 

its electronic hiss something like the sounds of radio (it's off or channel /FM signal is not available).

What could be the use of LM2903 in MIC from your point of view?
 

hi
gain=10000 its too much. High gain gives the overgain. It means the amplified signals are cut by the magnitude of power supply.
Ex: Vin*G=Vout 5mV*10000=50V but your power supply is just 9V. Therefore the amplifier's output signal will be pulse.
So you have to reduce the gain up to 50. If you want to make low noise amplifier, you should use the filters on each stage's outputs.
Add the DC block capacitor on amplifier's output.
 

If you want to make low noise amplifier, you should use the filters on each stage's outputs.
No.
You do not want to filter out the very important consonants sounds of speech that go up to 14kHz. Harmonics in music go above 20kHz.
If you filter out high audio frequencies then everything will sound like an AM radio or telephone. Speech is not only vowels. Music is not only humming.
 

Audio quality is very good with LM833 amplifier. I have tested with my PC.
 

can u suggest a circuit for when mic is mute LED blink and LED continuous on MIC is in use.

I have design a circuit, using op-amp as pulse generator o/p goes to LED and a 2 way switch connect as follow:


preamp <--. .-----9V supply
. -> pulse generator.

when mic is not in use 9V supply connect to pulse generator and when it is use connect to preamp.

Any other suggestion appreciated.
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top