20
\$\begingroup\$

I am creating a simple circuit in order to amplify the signal from an electret microphone and then add a 2.5V bias to the signal. The problem I'm having is that the signal from the mic is distorted as it leaves the op amps.

With an oscilloscope I have measured the signal from the microphone when exposed to 440 Hz sound and the wave looks like a perfect sine just as expected.

However once the signal passes the part of the circuit to amplify the signal the sine wave starts to get slightly distorted.

After the signal has passed the circuit of the inverting summing amplifier the signal gets even more distorted.

This is the circuit I use to amplify the signal:

Amplification circuit

This circuit leads to a 20 uF capacitor and is then followed by the following circuit to add the 2.5V DC bias:

enter image description here

where all resistor values are the same. This circuit uses the same LM324 as the amplification part of the circuit.

The resulting waves at 440 Hz look as follows:

enter image description here

This is after both the amplification and the summation. The distortion of the sine wave starts happening after the amplification, and gets more obvious after the summing op amp.

I have no idea what can cause this and hope somebody can point me into the right direction.

\$\endgroup\$
5
  • 3
    \$\begingroup\$ How did you construct this? Do you have the requisite decoupling capacitors for the LM324? Are the unused opamps in the quad LM324 package floating? \$\endgroup\$
    – pipe
    Commented Sep 20, 2016 at 21:17
  • 1
    \$\begingroup\$ Try 5K or 10K from the opamp output to -V. I'll look up the reference in a moment... OK, see electronics.stackexchange.com/questions/241561/… \$\endgroup\$
    – user16324
    Commented Sep 20, 2016 at 21:19
  • 1
    \$\begingroup\$ The right word for this is "distort", not "disform". I edited your question to make it easier to search for. \$\endgroup\$
    – Adam Haun
    Commented Sep 20, 2016 at 21:20
  • \$\begingroup\$ Is this op-amp really the right choice for an audio application? \$\endgroup\$
    – zwol
    Commented Sep 21, 2016 at 14:48
  • \$\begingroup\$ @zwol good enough for my purpose, i need to analyse which frequency is the loudest in a certain sound. So for me it works as long as the shape of the wave remains roughly the same. \$\endgroup\$ Commented Sep 21, 2016 at 15:03

1 Answer 1

33
\$\begingroup\$

This is classic crossover distortion in LM324.

Add a load resistor to output until crossover distortion disappears. Start with 1K to V- or gnd if using single supply.

This occurs with capacitive loads or next stage pullup R bias to Vcc/2 perhaps. Increase R load values as alternative if so... ( by design in order to minimize idle current)

Datasheet

"For ac applications,where the load is capacitively coupled to the output of the amplifier, a resistor should be used, from the output of the amplifier to ground to increase the class A bias current and prevent crossover distortion."

Note for single supply use. R1 could be 2K to V+ and 2K/Gnd THis depends on output swing max you need. enter image description here

\$\endgroup\$
4
  • 3
    \$\begingroup\$ pulldown not up... check the drive strengths. \$\endgroup\$
    – user16324
    Commented Sep 20, 2016 at 21:19
  • 2
    \$\begingroup\$ Probably has to be to V- since he's using a dual supply, I'm guessing the quote uses Gnd for single supply apps \$\endgroup\$
    – user16324
    Commented Sep 20, 2016 at 21:28
  • 13
    \$\begingroup\$ I added a 2K resistor from output to V- after the amplification and the sine wave looks exactly as its supposed to look like! \$\endgroup\$ Commented Sep 20, 2016 at 21:37
  • \$\begingroup\$ @user16324 A pull-down is typical, but sometimes you need a pull-up, if you want to extend the output range towards V+. It's a tradeoff. TI's "Application Design Guidelines for LM324 and LM358" appnote explains it well. \$\endgroup\$ Commented Jul 1 at 15:14

Not the answer you're looking for? Browse other questions tagged or ask your own question.