Inverting vs non-inverting
When would you choose an inverting amplifier over a non-inverting one?
| Level | ● screening |
| Track | General board |
| Really testing | Noise gain, common-mode, input impedance. |
Going deeper
Extra explanation that is not in the book: worked numbers, where people go wrong, and how to say it in the room.
Worked example — same gain magnitude, different bandwidth. Op-amp GBW 10 MHz, gain of 2:
- Non-inverting, gain +2 → noise gain 2 → bandwidth 5 MHz.
- Inverting, gain −2 → noise gain 3 → bandwidth 3.3 MHz.
The common-mode argument, with a real failure. A rail-to-rail-input op-amp used as a follower switches between its two input pairs somewhere near the top rail. As the signal crosses that point, offset jumps — visible as distortion on a large signal. An inverting stage keeps both inputs fixed, so it never crosses over.
What inverting buys you. A summing node: Vout = −Rf·(V1/R1 + V2/R2 + …) — mixers, DAC current-to-voltage, offset injection. And gains below 1 without a divider in front.
Where people go wrong
- Putting an inverting stage directly on a high-impedance sensor — R1 loads it. Buffer first, or use non-inverting.
- Forgetting that the − input of a non-inverting stage swings with the signal, so CMRR and input range matter.
Errata
No corrections known for this question. All errata
Ask about this question
Stuck on a step, or think something is wrong? Send it with AN-03 in the subject. Good questions get answered on this page, and corrections go into the next printing.
Related
← AN-02 Inverting gain of −10, non-idealAN-04 Current source and current sink →