Frequency Response of AC Circuits

Understanding how circuits behave across different input frequencies is fundamental to designing filters, communication systems, control loops, and amplifiers. This analysis, called frequency response, reveals whether a circuit passes, blocks, or amplifies specific frequency components, transforming our understanding from time-domain transients to predictable, steady-state performance. Mastering this concept allows you to predict and shape a circuit's interaction with any signal, from audio to radio waves.

From Time to Frequency: The Transfer Function

The cornerstone of frequency response analysis is the transfer function, denoted as $H(s)$ in the Laplace domain or $H(j\omega )$ in the frequency domain. It is defined as the ratio of the output signal to the input signal, expressed as a function of complex frequency ($s$) or angular frequency ($\omega$). For a circuit with a sinusoidal input, the steady-state output is also a sinusoid at the same frequency, but with a possibly different amplitude and phase. The transfer function captures this relationship: its magnitude $|H(j\omega )|$ gives the amplitude scaling, and its phase $\angle H(j\omega )$ gives the phase shift.

For example, consider a simple voltage divider. The transfer function is calculated using impedance concepts, where resistors have impedance $R$, capacitors have impedance $1/(j\omega C)$, and inductors have impedance $j\omega L$. By treating the circuit elements with their complex impedances and using standard circuit analysis techniques (like nodal analysis), you can derive $H(j\omega )$. This function is the mathematical gateway to plotting how the circuit responds to all frequencies.

First-Order Filters: RC and RL Circuits

The simplest frequency-dependent circuits are first-order systems built with a single energy-storing element (a capacitor or inductor) and resistors. These create either lowpass or highpass filter responses.

A standard RC lowpass filter places the resistor in series with the source and the capacitor in parallel with the output load. Its transfer function is: $$H(j\omega )=\frac{V_{out}}{V_{in}}=\frac{1}{1+j\omega RC}$$ The magnitude of this function is $|H(j\omega )|=1/\sqrt{1+(\omega RC{)}^2}$. At low frequencies ($\omega \to 0$), the magnitude is approximately 1 (0 dB), meaning the signal passes through. At high frequencies ($\omega \to \infty$), the magnitude approaches zero, meaning the signal is attenuated. The cutoff frequency or -3 dB frequency is ${\omega }_c=1/(RC)$, where the output power is half the input power. The phase shift goes from $0^{\circ }$ at low frequencies to $-90^{\circ }$ at high frequencies.

Conversely, an RC highpass filter swaps the positions of the resistor and capacitor. Its transfer function is $H(j\omega )=j\omega RC/(1+j\omega RC)$. It attenuates low frequencies and passes high frequencies. RL circuits exhibit analogous behavior, where the cutoff frequency depends on the $L/R$ ratio. The key takeaway is that a single reactive element creates a gentle, 20 dB per decade roll-off in the frequency response plots of magnitude and phase.

Second-Order Responses: RLC Circuits

Introducing two independent energy-storing elements (like both an inductor and a capacitor) creates a second-order system with richer behavior. RLC circuits can be configured to produce bandpass, notch (bandstop), or resonant peak responses depending on the output node selection.

A series RLC circuit with the output voltage taken across the resistor is a classic bandpass filter. It allows a specific band of frequencies to pass while attenuating those outside the band. Its transfer function is: $$H(j\omega )=\frac{V_R}{V_{in}}=\frac{j\omega R/L}{-{\omega }^2+j\omega R/L+1/(LC)}$$ This circuit exhibits resonance at the frequency ${\omega }_0=1/\sqrt{LC}$, where the impedance of the inductor and capacitor cancel, and maximum current (and thus voltage across R) flows. The sharpness of the resonance is described by the quality factor $Q={\omega }_{0}L/R$.

If the output is taken across the inductor-capacitor combination instead, the circuit becomes a notch filter, severely attenuating a narrow band of frequencies around ${\omega }_0$. These second-order responses are characterized by a steeper 40 dB per decade roll-off and are crucial in tuning circuits, like those in radio receivers or noise-cancellation systems.

Bode Plot Sketching Techniques

Manually calculating the magnitude and phase for every frequency is tedious. Bode plot techniques provide a powerful, rapid graphical method for sketching the asymptotic frequency response directly from the pole-zero form of the transfer function. The process involves two key steps.

First, factor the transfer function into contributions from real poles, real zeros, complex conjugate poles/zeros, and a constant gain term. For example, a transfer function might be written as $H(j\omega )=K\frac{(1+j\omega /{\omega }_{z1})}{(1+j\omega /{\omega }_{p1})(1+j\omega /{\omega }_{p2})}$, where ${\omega }_p$ and ${\omega }_z$ are pole and zero locations.

Second, sketch the magnitude plot by starting at a low-frequency magnitude determined by the constant $K$. As you increase frequency, each pole (in the denominator) contributes a -20 dB/decade slope change at its corner frequency, while each zero (in the numerator) contributes a +20 dB/decade slope change. The phase plot is sketched similarly, where each real pole adds a phase shift from $0^{\circ }$ to $-90^{\circ }$, centered at its corner frequency. For complex poles, the peak and sharpness depend on the damping ratio. Mastering this technique allows you to visualize a circuit's filter characteristics in seconds without a simulator.

Common Pitfalls

1. Confusing Angular Frequency ($\omega$) with Cyclical Frequency ($f$): Transfer functions are typically expressed using angular frequency $\omega$ in radians/second. The cutoff frequency ${\omega }_c=1/(RC)$ is not the same as $f_c=1/(2\pi RC)$. Always check the units in your equations and plots. A Bode plot's x-axis is almost always $\omega$ (log scale), but label it clearly.
2. Ignoring the Phase Response: Focusing solely on magnitude (gain) is a critical mistake. The phase shift is equally important, especially for stability in feedback systems and signal integrity. A filter that perfectly shapes the magnitude but introduces nonlinear phase can distort complex signals like audio or digital pulses.
3. Applying Bode Asymptotes Too Rigorously Near Corner Frequencies: The straight-line Bode asymptotes are approximations. The true magnitude at a corner frequency ($\omega ={\omega }_c$) is -3 dB, not 0 dB, for a simple pole. The true phase shift is $-45^{\circ }$, not the midpoint of the asymptotic transition. Always remember to apply these corrections for accurate sketching near these critical points.
4. Misidentifying Filter Type from Circuit Layout: It's easy to mislabel an RC circuit as highpass or lowpass by inspection. Always derive the transfer function or apply a thought experiment: At very high frequency, a capacitor acts as a short circuit and an inductor as an open circuit. Ask, "Does the output go to zero or to the input value at these extremes?" to correctly identify the filter type.

Summary

• Frequency response is characterized by plotting the magnitude and phase of a circuit's transfer function, $H(j\omega )$, versus frequency, revealing its filtering properties.
• First-order RC and RL circuits produce either lowpass or highpass responses with a cutoff frequency (${\omega }_c$) and a gentle 20 dB/decade roll-off.
• Second-order RLC circuits enable bandpass, notch, and resonant peak responses, featuring sharper roll-offs (40 dB/decade) and a resonant frequency ${\omega }_0=1/\sqrt{LC}$.
• Bode plot techniques allow for rapid manual sketching of the frequency response by using the pole-zero locations of the factored transfer function to build piecewise linear asymptotes for magnitude and phase.
• A complete analysis requires careful attention to both magnitude and phase, correct use of angular frequency, and understanding the limitations of asymptotic approximations.