Skip to content
Tutorials Center
My learning

Topics

Lesson 26 of 36 · Laplace Transform

Transfer Functions, Poles and Zeros

Learn what a transfer function is, how poles and zeros shape the response, how to judge stability from the s-plane and how to simulate systems with scipy.signal.

The Laplace transform of an LTI system's impulse response is its transfer function H(s) = Y(s)/X(s). It tells you everything about how the system treats any input. Because H(s) is a ratio of polynomials for systems described by ordinary differential equations, it is characterized by its zeros (roots of the numerator) and poles (roots of the denominator). Plotting those on the s-plane lets you read stability, speed and oscillation at a glance.

From differential equation to H(s)

A circuit or mechanical system such as a mass on a spring might be described by

y'' + 2ζωn y' + ωn² y = ωn² x.

With zero initial conditions the Laplace transform gives

H(s) = ωn² / (s² + 2ζωn s + ωn²).

For ωn = 2 rad/s and ζ = 0.5 this is H(s) = 4/(s² + 2s + 4). The poles solve s² + 2s + 4 = 0, so s = -1 ± j√3.

What poles tell you

Each pole at s = p contributes a term e^(pt) to the impulse response.

  • Real part of p controls growth or decay. Negative means decay.
  • Imaginary part of p controls oscillation frequency.
  • All poles in the open left half-plane means a causal system is BIBO stable: every bounded input gives a bounded output.
  • A pole on the imaginary axis gives a sustained oscillation (marginal stability).
  • A pole in the right half-plane gives growth: unstable.

Zeros do not change stability but shape the response: they create notches in the frequency response where H(jω) = 0 and can add overshoot.

Python Poles, zeros and stability

Press Run (or Ctrl + Enter) to see the result here.

Program input · what you'd type while the program runs

Type the answers before you press Run, one per line. Each time the program reads a line (with input()), it gets the next line from this box.

Frequency response from H(s)

If the poles are in the left half-plane, the frequency response is H(jω): evaluate H(s) along the imaginary axis. The magnitude |H(jω)| is the gain at frequency ω and ∠H(jω) is the phase shift. Geometrically, |H(jω)| is the product of distances from jω to the zeros divided by the product of distances to the poles, so a pole near the imaginary axis creates a resonance peak.

Time response of a second-order system

The step response of 4/(s² + 2s + 4) overshoots and rings, because the poles are complex. Standard measures:

  • Natural frequency ωn = 2 rad/s.
  • Damping ratio ζ = 0.5 (underdamped when ζ < 1).
  • Overshoot = exp(-πζ/√(1 - ζ²)), about 16.3 percent for ζ = 0.5.
  • Settling time (2 percent) ≈ 4/(ζωn) = 4 seconds.
Python Step response of a second-order system

Press Run (or Ctrl + Enter) to see the result here.

Program input · what you'd type while the program runs

Type the answers before you press Run, one per line. Each time the program reads a line (with input()), it gets the next line from this box.

The DC gain H(0) = 1 explains why the step response settles at 1, and |H| at ω = 2 is exactly 1/(2ζ) = 1: the resonance peak of this lightly damped system.

Block diagrams

Systems are often built from blocks. In series (cascade) the transfer functions multiply. In parallel they add. In a feedback loop with forward path G and feedback H, the closed-loop transfer function is G/(1 + GH). Feedback moves poles, which is how control engineers stabilize unstable plants.

flowchart LR
    X["x(t)"] --> G["G(s)"] --> Y["y(t)"]
    Y --> H["H(s)"]
    H -. "subtract" .-> G

Feedback loop: closed-loop transfer function G / (1 + G H).

Pole-zero cancellation can hide an unstable mode. If a right-half-plane pole is canceled by a zero in H(s), the system still has that unstable behavior internally; the input-output view simply cannot see it.

Check your understanding

Quick check · 4 questions

of 4 answered You scored / 4 · perfect!

  1. 1. Where must all poles be for a causal LTI system to be BIBO stable?

    Poles with negative real parts give decaying exponentials, so the impulse response is absolutely integrable.
  2. 2. The poles of H(s) = 4/(s² + 2s + 4) are...

    s = (-2 ± √(4 - 16))/2 = -1 ± j√3. The negative real part makes the system stable.
  3. 3. Two systems G1(s) and G2(s) are connected in series. The combined transfer function is...

    Cascading multiplies transfer functions because the output transform of the first block is the input transform of the second.
  4. 4. What do zeros of H(s) on the imaginary axis do?

    At a zero s = jω0, H(jω0) = 0, so a sinusoid at that frequency is canceled at the output.

Summary

  • H(s) = Y(s)/X(s) is the Laplace transform of the impulse response.
  • Poles set the natural modes: real part for decay or growth, imaginary part for oscillation.
  • A causal system is stable if and only if every pole has a negative real part.
  • Series connection multiplies, parallel adds, feedback gives G / (1 + G H).
  • scipy.signal.TransferFunction gives poles, zeros, step responses and frequency responses.

Next: The Sampling Theorem