A topic in the Open Knowledge Graph — a free, open map of 15,290 topics and the order to learn them in.
Pole locations in the s-plane determine system stability: poles in the left half-plane give stable systems, poles on the imaginary axis give marginally stable systems, and poles in the right half-plane give unstable systems. The pole-zero plot is a graphical representation of H(s) that reveals these properties immediately.
From your study of transfer functions, you know that a system's transfer function H(s) is a ratio of polynomials: H(s) = N(s)/D(s). The zeros are the values of s where N(s) = 0 (H(s) = 0), and the poles are the values of s where D(s) = 0 (H(s) → ∞). These are generally complex numbers, living in the s-plane — a two-dimensional coordinate system where the horizontal axis is the real part of s (σ) and the vertical axis is the imaginary part (jω). The pole-zero plot places all poles (marked ×) and zeros (marked ○) in this plane, creating a visual fingerprint of the system.
Why poles determine stability. The natural response of a system — what it does when disturbed with no ongoing input — is determined entirely by its poles. Each pole at s = σ + jω contributes a natural mode of the form eσt·cos(ωt + φ). The real part σ controls whether the mode grows or decays: if σ < 0, eσt → 0 and the mode dies out (stable); if σ = 0, eσt = 1 and the mode oscillates forever without decay (marginal stability); if σ > 0, eσt → ∞ and the mode grows without bound (unstable). This is why the left half-plane — where Re(s) < 0 — is the "stable" region. A single right-half-plane pole makes the entire system unstable, regardless of where the other poles are.
Reading the pole-zero plot. Look at where the poles sit. If all poles are in the left half-plane, the system is asymptotically stable — all natural responses decay to zero. The distance of a pole from the imaginary axis (its real part magnitude |σ|) tells you how fast it decays: poles far to the left (large negative σ) correspond to fast, highly damped modes; poles close to the imaginary axis correspond to slow, lightly damped oscillations. Complex conjugate pole pairs (which always come in conjugate pairs for systems with real coefficients) produce oscillatory responses: a pair at σ ± jω_d oscillates at frequency ω_d and decays at rate |σ|. The ratio of σ to the distance from the origin (the pole magnitude) is the damping ratio ζ — a pair on the imaginary axis has ζ = 0 (pure oscillation), and a pair on the negative real axis has ζ = 1 (critically damped).
Zeros and their role. Zeros do not affect stability — they cannot send a bounded input to ∞ — but they profoundly shape the system's frequency response and transient behavior. A zero near a pole partially "cancels" that pole's contribution to the output, reducing the visibility of that natural mode. Zeros on the imaginary axis create exact notch frequencies where the output is zero regardless of input magnitude. Right-half-plane (RHP) zeros create non-minimum phase behavior: the step response initially moves in the wrong direction before settling correctly (think of backing up a trailer, where turning the wheel right first moves the trailer left). RHP zeros place fundamental limits on achievable closed-loop bandwidth in control design.
From plot to system behavior at a glance. The pole-zero plot lets you read off stability, oscillation frequencies, damping ratios, and resonance peaks without solving differential equations. A system with poles at −1 ± j5 has a natural oscillation at 5 rad/s with moderate damping (σ/|s| = 1/√26 ≈ 0.2, so ζ ≈ 0.2). Moving those poles to −0.1 ± j5 makes the system lightly damped — nearly oscillating. Moving them to +0.1 ± j5 makes it unstable. This geometric intuition builds the foundation for root locus analysis (how poles move as a gain parameter varies) and Nyquist stability analysis (whether a feedback loop stabilizes or destabilizes a plant) — both of which reason directly in the s-plane.