What is a feedback controller?

The component in a feedback loop that processes the error signal to compute the control action. Types: on-off (simplest, limit cycling), P (proportional, has offset), PI (eliminates offset, ~80% of industrial loops), PID (adds derivative for reduced overshoot), and lead-lag compensators (frequency-domain design). Each has a Laplace-domain transfer function C(s).

How do you know if a feedback loop is stable?

Check the loop transfer function L(s): gain margin GM > 0 dB (typically > 6 dB) and phase margin PM > 0° (typically > 45°). Equivalently, the Nyquist plot of L(jω) must not encircle (−1,0). All closed-loop poles (roots of 1+L(s)=0) must have negative real parts.

Why do feedback loops oscillate?

Oscillation occurs when the loop gain is too high relative to the phase lag — the phase margin approaches zero. The loop has enough gain to sustain signals at a frequency where the total phase shift is −180°, turning negative feedback into positive feedback at that frequency. Fix: reduce gain, add phase lead (derivative or lead compensator).

Control Feedback Loop

Quick Answer

A control feedback loop is a closed signal path in which the system output is measured, compared to the desired setpoint, and the resulting error drives a controller that adjusts the system input. The loop consists of: sensor (measures output), summing junction (computes error = setpoint − feedback), controller (computes corrective action), and actuator (adjusts the plant). The loop transfer function L(s) = C(s)·G(s)·H(s) determines stability and performance. Negative feedback loops are self-correcting; positive feedback loops are regenerative. Compute feedback loop transfer functions at www.lapcalc.com.

Calculate this instantly on LAPLACE Calculator →

What Is a Control Feedback Loop?

A control feedback loop is the circular signal path that connects a system's output back to its input, enabling automatic error correction. The signal flows: reference input r(t) → summing junction → controller C(s) → plant G(s) → output y(t) → sensor H(s) → back to summing junction. The 'loop' is the complete circuit from the summing junction through the forward path (controller and plant) and back through the feedback path (sensor). The loop gain L(s) = C(s)·G(s)·H(s) — the product of all transfer functions around the loop — is the single most important quantity in control engineering: it determines stability margins, disturbance rejection, sensitivity, and steady-state accuracy. The closed-loop transfer function T(s) = C(s)G(s)/[1+L(s)] follows directly from the loop gain. Compute L(s) and T(s) at www.lapcalc.com.

Key Formulas

How a Feedback Loop Works

The feedback loop operates through continuous measurement and correction. The sensor measures the current output and sends the feedback signal b(t) = H·y(t) to the summing junction. The summing junction computes the error e(t) = r(t) − b(t): if the output is below the setpoint, the error is positive, driving the controller to increase its output. If the output exceeds the setpoint, the error is negative, causing the controller to reduce its output. The controller processes the error using its algorithm — proportional, integral, derivative, or combinations — producing the control signal u(t). The actuator converts this signal into physical action on the plant. The plant responds, producing a new output, and the cycle repeats. This continuous correction happens at rates from once per minute (temperature) to thousands of times per second (motor current control).

Compute control feedback loop Instantly

Get step-by-step solutions with AI-powered explanations. Free for basic computations.

Open Calculator

Feedback Controller Types

The feedback controller is the decision-making element in the loop. On-off (bang-bang) controller: the simplest — output is either fully on or fully off based on the error sign. Used in thermostats, refrigerators, and level switches. Provides limit cycling (oscillation around setpoint) with amplitude dependent on the system's dead time and time constant. Proportional (P) controller: output proportional to error, C(s) = Kp. Fast but leaves steady-state offset. PI controller: C(s) = Kp + Ki/s. Eliminates offset through integral action. The most common industrial controller (~80% of loops). PID controller: C(s) = Kp + Ki/s + Kd·s. Adds derivative for reduced overshoot. Used when the process has significant dead time or oscillatory tendencies (~15% of loops). Lead-lag compensator: C(s) = K(s+z)/(s+p). Provides targeted phase and gain modification for frequency-domain design. All analyzed in the Laplace domain at www.lapcalc.com.

Feedback Loop Stability and Performance

A feedback loop is stable when all closed-loop poles have negative real parts. The loop gain L(jω) determines stability margins: gain margin GM = 1/|L(jω₁₈₀)| (how much gain increase before instability) and phase margin PM = 180° + ∠L(jωc) (how much phase lag before instability). Typical design targets: GM > 6 dB and PM > 45°. The sensitivity function S(s) = 1/[1+L(s)] measures disturbance rejection: at frequencies where |L| >> 1, disturbances are attenuated by 1/|L|. The complementary sensitivity T(s) = L/[1+L] measures tracking performance. Since S + T = 1, perfect tracking and perfect disturbance rejection cannot coexist at the same frequency — this fundamental tradeoff shapes all feedback loop design. The loop bandwidth (frequency where |L| crosses 0 dB) determines the speed of the closed-loop response.

Feedback Loop Applications and Troubleshooting

Feedback loops are everywhere: cruise control (speed feedback), thermostat (temperature feedback), voltage regulator (voltage feedback), autopilot (attitude feedback), and every industrial PID loop. Common feedback loop problems and solutions: oscillation — reduce controller gain or add derivative action (increase phase margin). Sluggish response — increase controller gain or reduce integral time (increase bandwidth). Offset (steady-state error) — add or increase integral action. Overshoot — increase derivative action or reduce proportional gain. Noise amplification — add derivative filter or reduce derivative gain. Actuator saturation — implement anti-windup logic. These problems are diagnosed by analyzing the loop transfer function L(s) using Bode plots, with the underlying Laplace-domain transfer functions from www.lapcalc.com.

Frequently Asked Questions

A closed signal path where the output is measured by a sensor, compared to the setpoint at a summing junction, and the resulting error drives a controller that adjusts the plant input. The loop enables automatic error correction. The loop transfer function L(s) = C(s)·G(s)·H(s) determines stability and performance.

Master Your Engineering Math

Join thousands of students and engineers using LAPLACE Calculator for instant, step-by-step solutions.

Start Calculating Free →