Lecture 18: Sinusoidal Steady-State and the Exponential Trick

What this lecture covers

This lecture finishes the intuitive analysis of the underdamped RLC step response from the previous lecture, then opens a new topic: how circuits respond to sinusoidal inputs once the transient has died out. It explains why sinusoids matter (any periodic signal can be built from sums of sines via Fourier series) and why steady-state matters (in most practical circuits, such as an audio amplifier, the transient dies out quickly and what you actually observe is the long-run steady-state behavior).

The core of the lecture is a demonstration that solving directly for a cosine-driven RC circuit using ordinary differential-equation techniques becomes a trigonometric mess, and a first look at a shortcut: solving instead for a complex exponential input, then taking the real part of the result. After watching, you understand why sinusoidal steady-state analysis matters and have seen the exponential-substitution trick that the next lecture will develop into a full method.

Key ideas

• Intuitive RLC step-response sketching: the initial and final capacitor voltage can be read off by observation (the capacitor holds its voltage; after a long time it behaves as an open circuit to DC), and the direction of the very first swing can be determined by checking which way the inductor is pulling current.
• Sinusoids matter because of Fourier series: since any periodic waveform can be represented as a sum of sinusoids, understanding a linear circuit's response to one sinusoid lets you compute its response to any periodic input by superposition.
• Steady-state matters because transients die out: for most practical circuits, the homogeneous (transient) part of the solution decays to zero, so what you observe over time, such as when listening to music through an amplifier, is dominated by the steady-state response.
• Gain and phase change with frequency: a demo shows that driving an RC-loaded amplifier stage with sine waves of different frequencies changes both the output amplitude (gain drops as frequency rises) and its phase shift relative to the input.
• Direct trigonometric solution is painful: attempting a particular solution of the form A*cos(omega*t + phi) for an RC circuit driven by VI*cos(omega*t) requires expanding trigonometric identities and quickly becomes unwieldy.
• The exponential substitution shortcut: instead of solving for the cosine input directly, substitute a complex exponential input VI*e^(s*t); because exponentials keep their form under differentiation, the particular solution VP = VI / (1 + sRC) falls out with almost no algebra.
• Real part connects the two problems: since VI*cos(omega*t) is the real part of VI*e^(j*omega*t), and the circuit is linear, the real part of the easy exponential solution gives the answer to the original cosine problem — an "inverse superposition" argument.
• Complex transfer function: with s = j*omega, the particular solution becomes a complex number whose magnitude and phase, as functions of frequency, describe how the circuit scales and shifts a sinusoidal input.

Walkthrough

Finishing the RLC step-response sketch (0:32)

The lecture reviews the prior lecture's RLC circuit and completes the intuitive sketch of its underdamped step response. Using the fact that the capacitor holds its voltage and the inductor holds its current, the instructor shows that a negative initial inductor current implies the capacitor voltage must initially dip before ringing and settling at the final DC value.

Why sinusoidal drive and why steady-state (13:54)

The instructor announces a shift from time-domain transient analysis to sinusoidal steady-state analysis. He explains that sinusoids are worth studying because Fourier series let any periodic signal be expressed as a sum of sinusoids, so a linear circuit's sinusoidal response generalizes to any periodic input. He also explains that steady-state is the practically relevant regime because transient effects die out quickly in most circuits, leaving the steady-state response as what is actually observed over time.

Demo of frequency-dependent gain and phase (22:02)

A live demo drives an amplifier's RC-loaded gate node with sine waves and shows that as the input frequency increases, the output amplitude decreases and a phase shift appears between input and output, beyond the amplifier's usual inversion. The instructor uses this to motivate a formal frequency-domain analysis of the RC subcircuit.

Attempting the direct differential-equation solution (30:19)

Applying the usual four-step method to RC*dvC/dt + vC = VI*cos(omega*t), the instructor tries particular solutions of increasing complexity: a constant, then A*cos(omega*t), then A*cos(omega*t + phi). Each attempt either fails outright or leads into an expanding trigonometric expression, illustrating why this direct route becomes impractical.

The sneaky exponential substitution (36:25)

Rather than continuing the trigonometric approach, the instructor substitutes a different input, VI*e^(s*t), and finds the particular solution VP = VI / (1 + sRC) almost immediately, since differentiating an exponential returns the same exponential form. This detour, called the "sneaky approach," produces an answer quickly without indicating yet how it relates to the original cosine-driven problem.

Connecting back via the real part (43:43)

Substituting s = j*omega, the instructor shows that VI*cos(omega*t) is the real part of VI*e^(j*omega*t) by Euler's relation, and argues that for a linear circuit the real part of the easy exponential solution must equal the particular solution to the original cosine-driven equation. Working through the algebra, he arrives at VP = VI_amplitude * cos(omega*t + phi), where the amplitude and phase phi = arctan(omega*RC) come directly from the complex coefficient.

Steady-state result and frequency response preview (48:54)

The instructor notes that the homogeneous solution for an RC circuit decays as e^(-t/RC), so in steady state only the particular solution VP remains. He closes with a preview of the resulting magnitude and phase plots: output amplitude relative to input decreases as frequency increases, while the phase shifts from zero toward pi/2, setting up the next lecture's treatment of the transfer function.

Before you watch

• Review the previous lecture's RLC characteristic equation and underdamped step response, which this lecture builds on and completes.
• Be comfortable with the four-step differential equation method (particular solution, homogeneous solution, total solution, apply initial conditions) used throughout the course.
• A basic familiarity with Fourier series and Euler's relation (e^(j*theta) = cos(theta) + j*sin(theta)) helps follow the motivation and the exponential-substitution trick.

Check your understanding

1. How can you determine the initial direction (up or down) of an underdamped RLC step response just from the sign of the initial inductor current, without solving any equations?
2. Why does representing periodic signals as sums of sinusoids make it worthwhile to study a linear circuit's response to a single sinusoid?
3. Why does substituting a complex exponential input make solving for the particular solution so much easier than working directly with a cosine input?
4. Why is the real part of the response to VI*e^(j*omega*t) equal to the particular solution for an input of VI*cos(omega*t) in a linear circuit?