Go Down

Topic: Problem with OP Amplifier (Read 596 times) previous topic - next topic

__rian

Aug 21, 2017, 01:38 pm Last Edit: Aug 21, 2017, 01:57 pm by __rian
Hi Guys,

currently I'm trying to amplify the signal of an electret microphone with a LM358 op-amp (picture attached) and digitalize it with an ADC after that, but I'm having problems with the resistor values in my amplifier chain. The first amplifier eliminates the signal almost completely, rather than amplifying it by the inteded factor 10.

Can anybody help me please?

P.S. I already tossed the 680R resistor, that passed the microphone signal to the 2.5V bias voltage.

P.P.S. I forgot to mention the microphone has a 2200 Ohm impedance.

Thanks,
__rian

DVDdoug

Here is the schematic for the Sparkfun microphone board.

Quote
The first amplifier eliminates the signal almost completely, rather than amplifying it by the inteded factor 10.
Are you sure?  How are you measuring the output form the mic and the output from the op-amp?

Are you getting the 2.5V bias out of the 1st & 2nd stages?

Quote
P.S. I already tossed the 680R resistor, that passed the microphone signal to the 2.5V bias voltage.
I believe that's the right thing to do.

Quote
P.P.S. I forgot to mention the microphone has a 2200 Ohm impedance.
With the 2.2K bias/power resistor, you should get somewhere around 2.5V across the mic.   (It's not critical, but it shouldn't be zero or 5V.)

I'd guess our 1st stage needs more gain, and the 2nd stage less gain.   Any bias-error out of the 1st stage will be amplified by the DC coupled 2nd stage.  And/or, you might want to put a capacitor between the stages.

I don't thing it's hurting anything, but I don't think you need the 47pF feedback capacitor in the 2nd stage.  It makes a low-pass filter that becomes more "effective" (more down-into the audio range) at high gains.

A capacitor across the "bottom" bias resistor will help to minimize noise from the power supply.

MarkT

#2
Aug 21, 2017, 07:11 pm Last Edit: Aug 21, 2017, 07:18 pm by MarkT
Hi Guys,

currently I'm trying to amplify the signal of an electret microphone with a LM358 op-amp (picture attached) and digitalize it with an ADC after that, but I'm having problems with the resistor values in my amplifier chain. The first amplifier eliminates the signal almost completely, rather than amplifying it by the inteded factor 10.

Can anybody help me please?

P.S. I already tossed the 680R resistor, that passed the microphone signal to the 2.5V bias voltage.

P.P.S. I forgot to mention the microphone has a 2200 Ohm impedance.

Thanks,
__rian
Lose R6, its trashing your virtual ground completely.  You should have a decoupling cap on
your virtual ground to true ground too (10nF or 100nF or similar in parallel with R10.

Often its a better idea to generate a buffered virtual ground using another opamp as
a follower, as its then immune from pick-up in other parts of the circuit.  An unbuffered
virtual ground must only go to high impedance loads (ie the non-inverting inputs).

I've had poor experience with the LM358 above a few kHz, BTW - its doesn't seem to be able to handle
full output voltage swings (0--3V) above 5kHz, and it you drive it at 80 kHz its beyond hopeless!
[ I will NOT respond to personal messages, I WILL delete them, use the forum please ]

Go Up