An engineer increases the open-loop gain K of a minimum-phase control system by 3 dB. What happens to both the gain margin and the phase margin?
AGM decreases by 3 dB; PM is completely unaffected
BGM decreases by 3 dB; PM also typically decreases because the gain crossover frequency ωgc shifts to a higher frequency where phase lag is greater
CNeither margin changes — stability margins depend only on pole and zero locations, not gain
DPM decreases by 3 dB; GM is unaffected
Increasing gain K shifts the entire magnitude Bode plot upward by 3 dB. The gain crossover frequency ωgc (where magnitude = 0 dB) moves to a higher frequency. For typical minimum-phase systems, phase lag increases with frequency, so the phase at the new ωgc is more negative — PM decreases. Meanwhile, GM decreases directly by 3 dB because the magnitude at the (unchanged) phase crossover frequency ωpc is now 3 dB higher, leaving less margin before the 0 dB threshold. Both margins are affected simultaneously, which is why gain changes must be evaluated carefully during loop shaping.
Question 2 Multiple Choice
A minimum-phase control system has infinite gain margin. What does this imply about the system's phase Bode plot?
AThe system is unconditionally stable and cannot become unstable at any finite gain
BThe open-loop phase never reaches −180°, so there is no phase crossover frequency ωpc and the gain margin is undefined (infinite)
CThe closed-loop damping ratio is zero, producing sustained oscillation
DThe gain margin formula produces a division by zero, so the result is mathematically indeterminate
Gain margin is defined as GM = −20log|G(jωpc)H(jωpc)|, evaluated at the frequency where phase = −180°. If the open-loop phase never reaches −180° (which can happen for systems with limited phase roll-off, such as first- and second-order systems), there is no phase crossover frequency, and the gain margin is infinite by convention. However, this does NOT mean the system is unconditionally stable — non-minimum-phase systems, time-delay systems, and MIMO systems require more sophisticated analysis. For minimum-phase single-loop systems, infinite GM combined with positive PM does guarantee stability for all finite gains.
Question 3 True / False
Phase margin is measured at the phase crossover frequency — the frequency where the open-loop phase equals −180°.
TTrue
FFalse
Answer: False
This is a common confusion between the two margins. Phase margin is measured at the GAIN crossover frequency ωgc — the frequency where the open-loop magnitude equals 0 dB (unity gain). At ωgc, the phase margin is PM = 180° + ∠G(jωgc)H(jωgc). Gain margin, by contrast, is measured at the PHASE crossover frequency ωpc (where phase = −180°). The two margins are evaluated at two different frequencies. A good mnemonic: each margin measures how far you are from instability at the frequency where the OTHER condition for instability is already met.
Question 4 True / False
A higher phase margin generally corresponds to a more heavily damped, less oscillatory closed-loop transient response.
TTrue
FFalse
Answer: True
The relationship PM ≈ 100ζ (valid for ζ < 0.7) directly connects phase margin to closed-loop damping ratio. A 30° phase margin corresponds to ζ ≈ 0.3 (significant overshoot, ~37%); a 60° phase margin gives ζ ≈ 0.6 (modest overshoot, ~9%). A very high phase margin (e.g., 80°) means the system is overdamped — it responds sluggishly with minimal overshoot. Engineers target PM between 30° and 60° to balance responsiveness against oscillation. This is why phase margin is not merely a stability test but a primary design handle for shaping closed-loop transient performance.
Question 5 Short Answer
Explain why both gain margin and phase margin are needed to characterize stability robustness — why is one margin alone insufficient?
Think about your answer, then reveal below.
Model answer: Gain margin and phase margin measure robustness against different types of uncertainty at different frequencies. GM answers: 'how much can the gain increase before instability?' — measured at the frequency where phase lag is already at its worst (−180°). PM answers: 'how much additional phase lag can be tolerated before instability?' — measured at the frequency where the gain is already at its worst (0 dB). A system could have large GM but small PM (the gain can increase a lot, but even a small additional phase lag — from a cable delay or unmodeled dynamics — causes instability). Conversely, a system could have large PM but small GM (robust to phase variations but fragile to gain increases). Only together do the two margins characterize the full 'safety envelope' of the system against real-world uncertainties.
In practice, designers often add a third check: the distance from the Nyquist curve to the critical point (−1, 0) — the modulus margin. But for single-loop minimum-phase systems, specifying both GM > 6 dB and PM between 30–60° captures the key robustness requirements for most engineering applications. These design rules encode decades of experience about what margins are typically sufficient to survive component aging, temperature variation, and modeling errors.