A topic in the Open Knowledge Graph — a free, open map of 15,290 topics and the order to learn them in.
Underdamped second-order systems exhibit oscillatory approach to steady state with exponential decay envelope e−ζωₙt. The overshoot depends only on ζ: M_p = e^(−ζπ/√(1−ζ²)). Oscillation frequency is the damped frequency ωₙ√(1−ζ²).
When you analyzed second-order system responses, you encountered the three qualitatively different behaviors governed by the damping ratio ζ: overdamped (ζ > 1), critically damped (ζ = 1), and underdamped (ζ < 1). This topic focuses entirely on the underdamped case, which is the most common in practice and the most interesting mathematically. Underdamped systems overshoot their target and oscillate — they do not decay monotonically to steady state like their overdamped cousins.
The physical intuition is a mass-spring-damper system. If the damper is weak relative to the spring stiffness, the mass blows past its equilibrium position, springs back, and oscillates. The spring provides the restoring force that causes oscillation; the damper extracts energy each cycle, causing the oscillations to shrink. The natural frequency ωₙ controls how fast the system wants to oscillate, while ζ controls how aggressively the damper fights those oscillations. Together they determine the actual oscillation frequency: ωd = ωₙ√(1−ζ²), which is called the damped natural frequency. Notice that as ζ → 0 (no damping), ωd → ωₙ, so the undamped system would oscillate forever at ωₙ. As ζ → 1, ωd → 0, meaning oscillations slow and disappear — which makes sense, because ζ = 1 is the critically damped boundary.
The decaying oscillation has a specific shape: the decay envelope e−ζωₙt multiplies the oscillating sinusoid. Every cycle, the amplitude of the oscillation is smaller by a factor governed by how much damping occurs per period. This is why the formula for percent overshoot M_p = e^(−ζπ/√(1−ζ²)) × 100% depends only on ζ. The π/√(1−ζ²) in the exponent is precisely half the oscillation period normalized to the decay time constant — it captures how much the envelope decays in the time it takes to reach the first peak. A system with ζ = 0.5 has about 16% overshoot; ζ = 0.3 gives about 37%. Designers targeting less than 5% overshoot need ζ ≥ 0.69.
The practical importance of this analysis is in control system design. A feedback controller that is too aggressive (high gain) typically makes ζ small, leading to large overshoot and prolonged oscillation — the system hunts around its setpoint before settling. Understanding the exact relationship between ζ and overshoot lets engineers specify a desired performance (e.g., "no more than 10% overshoot") and translate that directly into a target ζ range, which then constrains the allowable controller gains. The same mathematics applies to electrical RLC circuits, mechanical vibrations, and any physical system whose dynamics reduce to a second-order ODE.