Vibrations & Dynamics · Learn

Frequency Response & Bode Plots

Feed a system a pure sine wave and it answers with a sine wave — same frequency, but scaled and delayed. Sweep every frequency and you get the system's fingerprint: the Bode plot.

▶ Play the lab📖 Learn the theory

1The transfer function

For a second-order system, the steady response to a sinusoid is captured by the complex transfer function, with frequency ratio $r=\omega/\omega_n$:

$$ H(i\omega) = \frac{1}{1 - r^2 + i\,2\zeta r} $$

Its magnitude $|H|$ is the gain applied to the input amplitude; its angle $\angle H$ is the phase lag. A Bode plot is just these two, drawn versus frequency.

2Why decibels and log frequency

Real systems span huge ranges, so gain is plotted in decibels ($20\log_{10}|H|$) against log frequency. The payoff: multiplying gains becomes adding, so a chain of components is just the sum of their Bode plots — and the response becomes straight-line asymptotes you can sketch by hand.

3Three regimes

FrequencyGainPhase
$r \ll 1$ (low)≈ 1 (0 dB)≈ 0° — output tracks input
$r \approx 1$ (resonance)peak ≈ $1/2\zeta$−90°
$r \gg 1$ (high)−40 dB/decade rolloff→ −180° (inverted)

At low frequency the mass keeps up; near $\omega_n$ it resonates; at high frequency it can't respond and the output shrinks and flips.

4The Q factor

The sharpness of the resonant peak is the quality factor $Q \approx 1/(2\zeta)$. High $Q$ (low damping) gives a tall, narrow, frequency-selective peak — a tuning fork or a radio filter picking one station. Low $Q$ gives a broad, gentle hump that responds to a wide band — a well-damped car suspension.

▶ In the lab

Drag the operating point along the Bode curves and watch the input and output sinusoids scale and slip out of phase; lower $\zeta$ and see the peak trade width for height. Open the lab →

Key takeaways

EngineeringCandy · Learn · the theory behind the playground