opamp amplifier dc offset.

Good day all
I have built this circuit as part of a little audio mixer.
Im getting a 200mV DC on the output pin. Opamp is not oscillating at all, actually perfectly clean DC.
I have tried different bipolar and JFET opamps.
This is a dual opamp and both sections have the same resistors.
I also tried a resistor on pins3/5. No difference.
Can anyone help? thanks!

Audio has frequencies as low as 20Hz but your opamp circuit is amplifying DC with its voltage gain of 47 times. Add a input coupling capacitor so that its DC gain is only 1. Using an input capacitor and with the fairly low value 47k feedback resistor then the opamp should have a low input bias current. If the opamp has a high input bias current then the other input should have a resistor to ground with the same value as the feedback resistor, and a capacitor to ground.
For -3dB at 20Hz then the value of the coupling capacitor would be 8uF and for a flat response at 20Hz the capacitor value should be 40uF which is huge.

Why are the resistor values so low? I would use 470k and 10k for the same gain as you have (using a modern OPA2134 dual audio opamp) then the input impedance would be 10 times higher and the capacitor values would be 1/10th. The OPA2134 is low noise and low distortion and has Jfet inputs with almost no input bias current.

audioguru said:

If the opamp has a high input bias current then the other input should have a resistor to ground with the same value as the feedback resistor, and a capacitor to ground.

Click to expand...

Yes on the resistor, no on the capacitor. The "other input" is the non-inverting one, and needs a DC path to GND (or some other voltage) to set the output DC voltage level.

ak

AnalogKid said:

Yes on the resistor, no on the capacitor.

Click to expand...

Does he mean a bypass cap rather than series/coupling?

Maybe. I took it to be a series cap, same as what you would use in a non-inverting amp configuration from the inverting input to GND. But he might mean a cap in parallel with the bias current balancing resistor. Don't think I've seen that before.

ak

"The other input should have a resistor to ground and a capacitor to ground" so of course they are parallel, not in series. The bypass capacitor prevents interference pickup. But a modern opamp with a very low input bias current does not need them.