Feedback resistor in op-amp

[Silent Fly]

I noticed that in some circuits (e.g. Demeter Compulator) impedance converters based on op-amps use a resistor between out and (-) instead of a straight wire.

From the theoretical viewpoint it shouldn't make any difference. I was wondering if somebody knows if there is any practical reason.

In other words, can you see any reason why some people use circuit (A) instead of circuit (B)?

Thanks

feedback-op-amp.gif (3.57 KiB) Viewed 1544 times

[earthtonesaudio]

Input bias currents.

For op-amps with high input currents, datasheets typically recommend the feedback resistor to equal the source resistance when used in the non-inverting configuration. Same as when you see an inverting op-amp with a resistor from the (+) terminal to the bias voltage. This way the input bias currents are equal and you don't have weird offests or distortion happening. I think.

With high-input impedance op-amps the input currents are so small that you can safely omit these resistors.

[Silent Fly]

Input bias currents.

For op-amps with high input currents, datasheets typically recommend the feedback resistor to equal the source resistance when used in the non-inverting configuration. Same as when you see an inverting op-amp with a resistor from the (+) terminal to the bias voltage. This way the input bias currents are equal and you don't have weird offests or distortion happening. I think.

With high-input impedance op-amps the input currents are so small that you can safely omit these resistors.

Thanks for your comments earthtonesaudio.

I checked the datasheet of the LF353N (used in the Compulator). The input bias current is 50pA and the input impedance is 10E12 ohm. Strangely enough, on page 9 of the NS datasheet and on page 7 of the ST datasheet, there are 2 impedance converters without resistor... I am confused.

[analogguru]

Input bias currents.

For op-amps with high input currents, datasheets typically recommend the feedback resistor to equal the source resistance when used in the non-inverting configuration. Same as when you see an inverting op-amp with a resistor from the (+) terminal to the bias voltage. This way the input bias currents are equal and you don't have weird offests or distortion happening. I think.

With high-input impedance op-amps the input currents are so small that you can safely omit these resistors.

Thanks for your comments earthtonesaudio.

I checked the datasheet of the LF353N (used in the Compulator). The input bias current is 50pA and the input impedance is 10E12 ohm. Strangely enough, on page 9 of the NS datasheet and on page 7 of the ST datasheet, there are 2 impedance converters without resistor... I am confused.

That´s only because you trust that every pedal is well designed by a skilled electronician...

analogguru

[earthtonesaudio]

50pA is super low input current. No resistor needed. The place where you'd expect to see the resistor are on bipolar-input op-amps with much larger input currents (in milli-amps).

[Silent Fly]

Input bias currents.

For op-amps with high input currents, datasheets typically recommend the feedback resistor to equal the source resistance when used in the non-inverting configuration. Same as when you see an inverting op-amp with a resistor from the (+) terminal to the bias voltage. This way the input bias currents are equal and you don't have weird offests or distortion happening. I think.

With high-input impedance op-amps the input currents are so small that you can safely omit these resistors.

Thanks for your comments earthtonesaudio.

I checked the datasheet of the LF353N (used in the Compulator). The input bias current is 50pA and the input impedance is 10E12 ohm. Strangely enough, on page 9 of the NS datasheet and on page 7 of the ST datasheet, there are 2 impedance converters without resistor... I am confused.

That´s only because you trust that every pedal is well designed by a skilled electronician...

analogguru

You may have a point Analogguru

[Silent Fly]

50pA is super low input current. No resistor needed. The place where you'd expect to see the resistor are on bipolar-input op-amps with much larger input currents (in milli-amps).

Ok - from now on, I will follow a simple rule: "no need for a feedback resistor for unit gain unless the datasheet says it is needed".