Analog Circuits

Amplifier and Signal Circuits: Feedback Amplifier Types Explained

Guide to feedback amplifier types: negative vs positive feedback, the four feedback topologies, and how each one changes gain, impedance and stability.

On this page
  1. What a feedback amplifier does
  2. Negative and positive feedback
  3. The gain formula and loop gain
  4. The four feedback topologies
  5. How feedback changes impedance
  6. Bandwidth, distortion and noise
  7. Stability: when negative feedback becomes oscillation
  8. Feedback in the circuits you’ll meet
  9. Advantages and trade-offs

What a feedback amplifier does

A feedback amplifier takes a sample of its output and returns part of it to the input. The amplifier then responds to the difference between the input signal and the returned signal. This trades raw gain for control. The overall gain ends up set mostly by the feedback network, and other properties such as distortion, bandwidth and impedance change in fairly predictable ways.

This is why an operational amplifier with a very large open-loop gain, often 100 dB or more, can be set to a gain of 10 or 100 with two resistors.

Negative and positive feedback

  • Negative feedback subtracts the returned signal from the input. The loop opposes changes, so gain drops, but stability, linearity and bandwidth usually improve.
  • Positive feedback adds the returned signal to the input. Gain rises, and the circuit can latch or oscillate. That behavior is useful in oscillators and in the Schmitt trigger.

Most amplifier circuits you build use negative feedback. Positive feedback is used on purpose in a few places, but it needs careful design.

The gain formula and loop gain

Let A be the open-loop gain of the amplifier and β (beta) the fraction of the output that is fed back. The product Aβ is the loop gain. With negative feedback, the closed-loop gain is:

Af = A / (1 + Aβ)

With positive feedback, the denominator becomes (1 − Aβ).

When the loop gain is much larger than 1, the formula simplifies to Af ≈ 1/β. The closed-loop gain then depends mainly on the feedback network, usually a pair of resistors, and much less on the amplifier’s own gain, which can drift with temperature, supply voltage and device age.

For example, take an amplifier with A = 10,000 and β = 0.1, so Aβ = 1,000. The closed-loop gain is 10,000 / 1,001 ≈ 9.99, which is very close to 1/β = 10.

The four feedback topologies

Feedback connections are named with two words. The first word says what is sampled at the output: voltage or current. The second says how the feedback signal joins the input: in series with the input voltage, or in shunt (in parallel) at the input node.

  • Voltage-series: samples the output voltage and feeds back a voltage in series with the input. The non-inverting op-amp amplifier and the emitter follower are common examples.
  • Voltage-shunt: samples the output voltage and feeds back a current in parallel with the input. The inverting op-amp amplifier is a common example.
  • Current-series: samples the output current and feeds back a voltage in series with the input. A transistor stage with an unbypassed emitter resistor is a common example.
  • Current-shunt: samples the output current and feeds back a current in parallel with the input. It appears less often in beginner circuits.

For more detail on each connection, see the four types of feedback connections.

How feedback changes impedance

The impedance changes come from the same factor, (1 + Aβ). Treat the table below as a guide, since the exact change depends on which impedance you measure and on the amplifier’s internal details.

Topology Input impedance Output impedance
Voltage-series Increases Decreases
Voltage-shunt Decreases Decreases
Current-series Increases Increases
Current-shunt Decreases Increases

In practice, voltage-series feedback suits a voltage amplifier that should load its source lightly and drive its load stiffly. Current-shunt feedback has the opposite effect on both impedances.

Bandwidth, distortion and noise

For a simple single-pole amplifier, feedback trades gain for bandwidth. The closed-loop bandwidth grows by roughly the same factor that the gain falls.

Feedback also reduces nonlinear distortion and noise generated inside the forward amplifier, by a factor similar to (1 + Aβ). It does not reduce noise or interference that enters at the input, because the feedback loop treats that signal as part of the wanted input.

In audio amplifiers, the story is more nuanced. Moderate amounts of feedback can change which distortion harmonics remain, so a single THD number does not tell the whole story.

Stability: when negative feedback becomes oscillation

The loop gain Aβ falls as frequency rises, and each stage adds phase shift. If the loop gain reaches 1 (0 dB) at a frequency where the total phase shift is 180°, the feedback turns positive at that frequency, and the circuit can oscillate.

Two margins describe how close a circuit is to that point:

  • Gain margin: how far the loop gain sits below 1 at the frequency where the phase shift reaches 180°.
  • Phase margin: how far the phase sits from 180° at the frequency where the loop gain equals 1.

Engineers check these margins with Bode or Nyquist plots. A common hobbyist symptom of poor margins is an amplifier that rings, squeals or oscillates when a load or wire changes. Check your op amp’s datasheet for recommended compensation, such as a small capacitor across the feedback resistor.

Feedback in the circuits you’ll meet

Advantages and trade-offs

Negative feedback gives you:

  • Gain that depends mainly on resistors, so it is more predictable
  • Lower distortion and better linearity
  • Wider bandwidth
  • Controlled input and output impedances

The trade-offs are:

  • Lower overall gain
  • More circuit complexity
  • A stability check every time the loop changes
  • No reduction of noise that enters at the input

Once you understand the loop gain, the formulas in this article apply to most op-amp and transistor stages you will build.

Keep exploring

Guides from other topics

All topics →