A topic in the Open Knowledge Graph — a free, open map of 15,290 topics and the order to learn them in.
Bode plot stability analysis applies the open-loop frequency response G(jω)H(jω) to assess closed-loop stability without solving for closed-loop poles. The gain crossover frequency ωgc is where the open-loop magnitude equals 0 dB, and the phase crossover frequency ωpc is where the phase equals −180°. For minimum-phase systems in a unity feedback loop, closed-loop stability requires that the phase at ωgc exceeds −180° and the gain at ωpc is below 0 dB. These crossover relationships define the gain and phase margins, which quantify how much additional gain or phase lag the system can tolerate before becoming unstable.
Sketch asymptotic Bode plots for several open-loop transfer functions and identify crossover frequencies by hand. Compare with computed Bode plots to calibrate the accuracy of asymptotic approximations, especially near corners.
Your prerequisites give you two tools: transfer functions describe how a system maps input to output in the Laplace domain, and Bode plots show the magnitude and phase of a system's frequency response. Bode plot stability analysis combines these to answer a practical question without solving for closed-loop poles directly: will a unity feedback loop with plant G(s) be stable?
The key insight is what happens when a signal travels around the feedback loop. A sinusoidal input at frequency ω gets multiplied in magnitude by |G(jω)H(jω)| and shifted in phase by ∠G(jω)H(jω) on each trip around the loop. If the loop gain equals exactly 1 (0 dB) and the phase shift equals exactly −180°, then the signal fed back is an inverted copy of itself at full strength — which, because of the subtraction in the negative feedback summing junction, actually *adds* to the input rather than subtracting. The system reinforces itself without bound: that is instability. Bode stability analysis amounts to checking how close the system comes to this critical condition.
The gain crossover frequency ωgc is where the open-loop magnitude first crosses 0 dB. The phase margin (PM) is how far the phase at ωgc is from −180°: PM = 180° + ∠G(jωgc)H(jωgc). A positive phase margin means that at the frequency where the loop gain is 1, the phase has not yet reached −180° — there is angular "room" before instability. The phase crossover frequency ωpc is where the phase first hits −180°. The gain margin (GM) is how far the magnitude at ωpc is from 0 dB, expressed in dB: GM = −20 log₁₀|G(jωpc)H(jωpc)|. A positive gain margin means the loop gain at the critical phase is below 1 — the system would need more gain before going unstable.
Reading margins off a Bode plot is fast and visual: find the 0 dB crossing and read the phase; find the −180° crossing and read the magnitude. But the interpretation requires care. Both margins must be adequate simultaneously — a large gain margin with a small phase margin (or vice versa) still produces a poorly damped, nearly unstable design. Standard targets of GM > 6 dB and PM > 45° are rules of thumb that correspond to reasonable closed-loop damping ratios. The method only applies directly to minimum-phase systems — those with no right-half-plane zeros and no time delays — because only those systems have phase that is uniquely determined by the magnitude response. Systems with time delays or RHP zeros need the Nyquist criterion, which Bode is a simplified version of.