The Repair LibraryRead · Learn · Master

Feedback Systems

Feeding the output back to the input — how negative feedback buys stability and precision, how positive feedback makes oscillators and snap-action, and how both shape every amplifier and control loop.

IntermediateLow Risk26 min read

What You Will Learn

  • You will learn what feedback is — feeding part of a system's output back to its input — and the difference between negative and positive feedback.
  • You will learn what negative feedback buys: stable, precise gain set by the feedback network, plus lower distortion and wider bandwidth, at the cost of raw gain.
  • You will learn what positive feedback does: sustain oscillation in oscillators and add hysteresis for clean, noise-immune switching.
  • You will learn how feedback runs control loops around a setpoint, and how unwanted feedback causes instability and oscillation.

What You Will Be Able To Do

  • You will be able to explain feedback and distinguish negative from positive feedback by their effect.
  • You will be able to state what negative feedback improves and the raw-gain-for-precision tradeoff.
  • You will be able to describe an oscillator and a Schmitt trigger's hysteresis as positive-feedback effects.
  • You will be able to describe a feedback control loop and recognize feedback faults like unwanted oscillation.

Required Tools

No physical tools required. This is a conceptual section.

Section Overview

Feedback — feeding part of a system's output back to its input so the output influences itself — is one of the most powerful ideas in all of electronics and control, and you've already seen it at work. The op-amp used negative feedback to tame its enormous gain; the voltage regulator used a feedback loop to hold its output steady. This section pulls the idea together. You'll see what negative feedback buys — stable, precise gain, lower distortion, wider bandwidth — and the tradeoff it demands. You'll see what positive feedback does — sustain oscillation in an oscillator, and add hysteresis for clean snap-action switching. And you'll see feedback as the basis of control loops and, when it goes wrong, of instability and unwanted oscillation. The treatment stays conceptual: the ideas and their consequences, not the stability mathematics.

Why This Matters

Feedback is the organizing principle behind a huge fraction of analog and control circuits, so understanding it turns many circuits from mysterious to readable. Nearly every quality amplifier uses negative feedback to be accurate; every oscillator, clock, and signal generator uses positive feedback to run; every regulator, thermostat, and motor controller is a feedback loop. For a repair technician, feedback faults are a recognizable and important class: an amplifier that breaks into a squeal or a low "motorboating" oscillation, an oscillator that has gone silent, a regulator or control loop that won't hold its output. Recognizing that a symptom is a feedback problem — unwanted positive feedback, a broken feedback path, a failed compensation part — points the diagnosis in the right direction. Feedback also ties together threads from across the handbook: the op-amp, the regulator, and resonance all meet here.

Required Prerequisites

No consumables required. The exercises are reasoning about feedback behavior; no parts are consumed.

  • Optional: an op-amp, resistors, and a capacitor on a breadboard to build a simple oscillator or a Schmitt trigger and watch positive feedback in action, plus a signal source and oscilloscope
  • No special hardware is required; this section is conceptual and stands on reasoning and examples

Real-World Applications

Feedback runs quietly under an enormous range of equipment. Every hi-fi amplifier uses negative feedback to keep its gain flat and its distortion low. Every crystal or RC oscillator — the clock in a computer, the tone in a signal generator, the carrier in a radio transmitter — is an amplifier wrapped in positive feedback. Schmitt triggers clean up noisy digital and sensor signals with hysteresis so a slowly changing or noisy input still produces a crisp, single transition. Thermostats, cruise controls, motor-speed governors, and every voltage regulator are negative-feedback control loops holding something at a setpoint. And the flip side shows up as trouble: an amplifier that oscillates because of stray feedback, or a power supply that becomes unstable — feedback gone wrong. Wherever a circuit corrects itself, generates a waveform, or holds a value steady, feedback is the reason.

Common Challenges

  • Reversing negative and positive. Negative feedback opposes the input and stabilizes; positive feedback reinforces the input and drives the system harder. The two do nearly opposite things, so keeping them straight is essential.
  • Thinking feedback only reduces gain. Negative feedback does lower the overall gain, but that's the price for what it buys: stability, precision, low distortion, and wide bandwidth. The "lost" gain is spent on accuracy.
  • Assuming oscillation is always a fault. Positive feedback is designed in to make oscillators and snap-action switches. Oscillation is only a problem when it's unwanted — an amplifier or supply that shouldn't be oscillating but is.

Safety Notes

Risk Level: Low. The feedback concepts here are analysis, not hands-on power work. Two practical notes.

Professional Tips Before Starting

  • Ask "does the feedback oppose or reinforce?" That one question sorts almost everything: opposing (negative) means the circuit is trying to be stable and accurate; reinforcing (positive) means it's trying to swing hard — to oscillate or snap.
  • Look for the feedback path. In an op-amp circuit it's usually a resistor (and sometimes a capacitor) from the output back to the inverting input; in a regulator it's the sense line from the output. Finding that path tells you what the circuit is regulating and where a feedback fault could hide.
  • When something oscillates that shouldn't, think unintended positive feedback: a bad compensation capacitor, a ground or supply-decoupling problem, or a stray coupling path turning wanted negative feedback into positive at some frequency.

Understanding Feedback and Its Uses

What Feedback Is

Feedback means taking part of a system's output and routing it back to its input, so the output has a say in what the system does next. That simple loop is astonishingly powerful, and it comes in two opposite flavors depending on how the fed-back signal combines with the input. If it opposes (subtracts from) the input, it's negative feedback; if it reinforces (adds to) the input, it's positive feedback. Those two do nearly opposite things, and almost every use of feedback is one or the other.

Negative Feedback and Its Tradeoff

Negative feedback — the fed-back signal opposing the input — is the stabilizing kind you met with the op-amp, where feedback drove the two inputs to be equal. Because the output is constantly fed back to counteract any deviation, the system self-corrects, and that brings a whole cluster of benefits:

  • Stable, precise gain. The overall gain stops depending on the device's large, variable, temperature-sensitive raw (open-loop) gain and instead becomes set by the feedback network — typically a pair of stable resistors. You trade an imprecise huge gain for a precise, predictable smaller one.
  • Lower distortion. Because the feedback corrects deviations from a faithful copy, the nonlinearity and in-loop disturbances the amplifier introduces are reduced. (Feedback cuts distortion, but it does not improve the noise set by the input stage — that is a separate concern.)
  • Wider bandwidth. Negative feedback flattens and extends the frequency response, widening the usable band.
  • Better impedances. It can raise input impedance and lower output impedance, so the stage loads its source less and drives its load better.

The catch is the fundamental tradeoff: you give up raw gain to gain stability, precision, and linearity. The amplifier's enormous open-loop gain is "spent" buying accuracy — which is almost always a great trade, and it's why nearly every good amplifier and every regulator relies on negative feedback. (Negative feedback was introduced in the op-amp section; here it's the general principle rather than a new definition.)

Positive Feedback: Oscillators and Hysteresis

Positive feedback — the fed-back signal reinforcing the input — pushes the system harder in the direction it's already going. Rather than settling, it wants to run away, and two hugely useful behaviors come from harnessing that:

  • Oscillators. Wrap positive feedback around an amplifier and, under the right condition, the circuit sustains a continuous waveform all by itself — an oscillator. Conceptually the requirement (the Barkhausen idea) is that the loop gain is at least 1 and the feedback arrives in phase (truly positive) at the frequency of oscillation: then any tiny disturbance is reinforced round and round the loop and grows into a steady oscillation. Oscillators are the source of every clock, signal generator, and radio carrier — often using an LC resonant circuit (Chapter 3) or a crystal to set the frequency.
  • Snap-action and hysteresis. Add positive feedback to a comparator and it becomes a Schmitt trigger: instead of one switching threshold it has two — a higher one to switch on and a lower one to switch off. That gap is hysteresis, and it makes the output snap cleanly between states and ignore noise near the threshold, so a slowly changing or noisy input still gives one crisp transition instead of a stutter of them.

The same reinforcing tendency is dangerous when it's unwanted: stray or unintended positive feedback can make an amplifier break into oscillation (a squeal, or a slow "motorboating" putt-putt) or latch up, which is why circuits are carefully designed — and sometimes compensated — to keep unwanted positive feedback from taking hold.

Feedback in Control

Feedback isn't only for signals — it's the basis of control. A control loop senses its output, compares it to a desired value (the setpoint), and corrects the difference, over and over, using negative feedback to drive the error toward zero. A thermostat senses room temperature, compares it to the set temperature, and switches heating to close the gap. A motor-speed governor senses speed and adjusts drive to hold it. And the voltage regulator from Section 4.6 is exactly this: it senses its output voltage, compares it to an internal reference (its setpoint), and adjusts to hold the output steady. Recognizing a circuit as a feedback control loop — find the sensed output, the reference, and the correcting element — makes its purpose and its failure modes clear.

Stability and Unwanted Oscillation

Negative feedback is stabilizing — usually. But feedback takes time to travel the loop, and at high frequencies that delay shows up as a phase shift. If, at some frequency, the phase shifts enough that the intended negative feedback arrives effectively in phase — positive — and the loop gain is still at least 1 there, the amplifier will oscillate at that frequency, exactly as an oscillator does. This is why amplifiers include compensation (often a small capacitor) to reduce the gain at high frequencies and keep the feedback safely negative across the band. You don't need the stability mathematics here; the takeaway is that too much feedback with the wrong phase can flip helpful negative feedback into harmful positive feedback, and a failed compensation component or a stray coupling path is a classic cause of an amplifier that has started to oscillate.

Common Mistakes

  • Swapping negative and positive feedback. Negative opposes and stabilizes; positive reinforces and drives toward oscillation or snap-action. They do opposite jobs.
  • Calling all oscillation a failure. Oscillators use positive feedback on purpose; oscillation is only a fault when it's unintended.
  • Forgetting the tradeoff. Negative feedback lowers gain in exchange for stability, precision, and bandwidth — that reduced gain is the feature working, not a defect.
  • Overlooking the feedback path in diagnosis. A broken feedback resistor, a bad compensation cap, or a lost sense line changes everything downstream; always find and check the feedback path.

Troubleshooting Guidance

Feedback faults form a recognizable family, and naming the family speeds the diagnosis. When a circuit oscillates that shouldn't — an amplifier squealing, a low motorboating putt, a supply gone unstable — suspect unwanted positive feedback: a failed compensation capacitor that no longer rolls off the high-frequency gain, a decoupling or grounding problem letting the output couple back to the input, or a stray path turning intended negative feedback positive at some frequency. When an oscillator has stopped, the loop has lost the condition to sustain oscillation: check for lost gain in the amplifier or a broken feedback path (an open component in the resonant or feedback network), since either drops the loop gain below the level needed. When a control loop won't regulate — a regulator whose output is wrong, a temperature or speed that won't hold — look at the feedback sensing: a broken sense line or a failed reference means the loop can no longer compare output to setpoint and correct, so it runs open-loop. In every case, the move is the same: identify the feedback path, decide whether the fault has broken it (loss of regulation, dead oscillator) or created unwanted feedback (unstable oscillation), and check the specific components in that path. And treat an oscillating power circuit as both a fault and a hazard — fix it rather than leave it running.

Verification & Testing Methods

Check your understanding before moving on:

  • [ ] Explain what feedback is and distinguish negative from positive feedback by their effect on the input.
  • [ ] State what negative feedback improves (stable gain, low distortion, wide bandwidth) and the raw-gain tradeoff.
  • [ ] Describe how positive feedback makes an oscillator and how it gives a Schmitt trigger its hysteresis.
  • [ ] Describe a feedback control loop (sense, compare to setpoint, correct) and one feedback fault.

Then try the practice exercises below — reasoning about feedback behavior, no powered work required.

Practice Exercises

  1. Sort the effect (5 minutes, reasoning). For each, state whether it relies on negative or positive feedback and why: (a) a hi-fi amplifier with low distortion and a precise, stable gain; (b) a radio-frequency oscillator generating a carrier; (c) a Schmitt trigger cleaning up a noisy sensor signal; (d) a thermostat holding room temperature.
  2. The tradeoff (5 minutes, reasoning). Explain in your own words why an engineer would deliberately reduce an amplifier's gain by adding negative feedback — what is gained in exchange, and why is it usually worth it?
  3. Hysteresis reasoning (10 minutes, reasoning). A comparator without hysteresis chatters (switches back and forth rapidly) when a slow, noisy signal crosses its threshold. Explain how adding positive feedback (making it a Schmitt trigger) fixes this, in terms of its two thresholds.
  4. Feedback fault triage (10 minutes, reasoning). For each symptom, name the likely feedback-fault category and one thing to check: (a) an amplifier that has started to squeal/oscillate; (b) a signal-generator oscillator that has gone dead; (c) a voltage regulator whose output no longer holds its value.

These core ideas — feedback and its two kinds, what negative feedback buys and trades, positive feedback for oscillators and hysteresis, and feedback control and its faults — are tested in the Chapter Quiz at the end of this chapter, where a score of 80% is required to continue.

Key Takeaways

  • Feedback feeds part of a system's output back to its input; it is negative when the fed-back signal opposes the input (stabilizing) and positive when it reinforces the input (driving harder).
  • Negative feedback (from the op-amp section) buys stable, precise gain set by the feedback network, lower distortion, wider bandwidth, and better impedances — trading away raw open-loop gain for that accuracy.
  • Positive feedback reinforces the input: wrapped around an amplifier with loop gain at least 1 and in-phase, it sustains oscillation in an oscillator (clocks, signal generators, radio carriers).
  • Positive feedback also gives a comparator hysteresis — two switching thresholds — making a Schmitt trigger that snaps cleanly and ignores noise near the threshold.
  • A feedback control loop senses its output, compares it to a setpoint, and corrects the difference — thermostats, motor governors, and voltage regulators are all negative-feedback control loops.
  • Feedback faults are a recognizable class: unwanted oscillation (stray/unstable positive feedback or a bad compensation part), a stopped oscillator (lost gain or a broken feedback path), or a control loop that won't regulate (broken sensing) — and an oscillating power circuit is a fault and a hazard.

Skills Learned

  • You can now explain feedback and distinguish negative from positive feedback by their effect.
  • You can now state what negative feedback improves and the raw-gain-for-precision tradeoff.
  • You can now describe an oscillator and a Schmitt trigger's hysteresis as positive-feedback effects.
  • You can now describe a feedback control loop and recognize feedback faults like unwanted oscillation.
  • You can now connect the op-amp, the regulator, and resonance as instances of the single idea of feedback.

Glossary Additions

  • feedback — routing part of a system's output back to its input so the output influences the system's own behavior; it is negative when the fed-back signal opposes the input (stabilizing and self-correcting) and positive when it reinforces the input (driving the system harder). Feedback underlies amplifiers, oscillators, and control loops.
  • positive feedback — feedback in which the fed-back signal reinforces (adds to) the input, driving the system harder in the same direction; used deliberately to sustain oscillation and to create hysteresis for snap-action switching, and a cause of instability when unwanted.
  • oscillator — a circuit that generates a continuous, repeating waveform on its own by wrapping positive feedback around an amplifier so that, with enough loop gain and in-phase feedback, oscillation sustains itself; the source of clocks, signal generators, and radio carriers.
  • hysteresis — a deliberate gap between a circuit's switch-on and switch-off thresholds, created by positive feedback (as in a Schmitt trigger), so the output snaps cleanly between states and ignores noise near the threshold instead of chattering.

Suggested Next Sections

Must read next:

  • Signal Conditioning Circuits — putting amplifiers, filters, and feedback to work preparing a real-world signal — buffering, level-shifting, scaling, and cleaning it up — for the next stage or a converter.

Recommended: