Lecture 2:

Review of Basic Circuit Analysis

(in time domain)

Gu-Yeon Wei
Division of Engineering and Applied Sciences
Harvard University
[email protected]

Wei Lecture 2 1
 Overview

• Reading

• Supplemental Reading
– Nilsson: Chapters 6-8
(Most textbooks on Electric Circuits have this material)

• Background
– In this lecture, we will continue to review basic circuit analysis and
focus on circuits that contain reactive elements inductors and
capacitors. Inductors and capacitors are a little more tricky than
simple resistors b/c their current/voltage relationship also depends
on time. For now, we will rely on differential equations to describe
their transient behavior. In the next lecture, we will see how to use
the Laplace transform to analyze circuits in the s-domain, which
allows us to analyze circuit characteristics w.r.t. frequency.

Wei Lecture 2 2
 L and C

• The current and voltage relationship of an inductor and capacitor are governed
by the following equations:

iL
di 1 t
vL vL = L L
dt
i L (t ) =
L ∫
t0
v L ⋅ d τ + i L (t 0 )

iC
dv C 1 t
vC vC = C
dt
v C (t ) =
C ∫ t0
iC ⋅ d τ + v C (t 0 )

• Some intuitive properties:

– current through an inductor cannot change instantaneously, but voltage
across one can
– voltage across a capacitor cannot change instantaneously, but current
through it can
Wei Lecture 2 3
 Power and Energy in L and C

• Remember, p = vi and p = dw/dt (power is the rate of change of energy)

– notice, I used w to represent energy as opposed to E to avoid confusion with electric fields

• Inductors store magnetic field energy

di L dw di
pL = Li L pL = = Li L L dw L = Li L di L
dt dt dt
– Integrate on both sides
w i 1
∫ dx = L ∫ y ⋅ dy wL =
2
Li L
0 0 2

• Capacitors store electric field energy

1
wC =
2
Cv C
2

Wei Lecture 2 4
 Energy vs. Charge

• A classic E&M problem…

– Q = C*V and we just saw that w = ½ CV2
– Let’s assume we have two capacitors. At t<0, C1 is charged by a
voltage source. At t=0+, the switch 1 (sw1) opens and switch 2 (sw2)
closes to add an additional capacitor C2 in parallel with C1.
Calculate the redistribution of charge and the redistribution of
energy at t>0.
sw1 sw2 sw1 sw2

V C1 C2 V C1 C2

t<0 t>0
– Conservation of energy vs. conversation of charge?

Wei Lecture 2 5
 Natural Response of RL Circuit

• We can use the concepts we’ve learned so far to solve for the natural response
of RL circuits. +
t=0
i

R0 L v R i L v R

• Using KVL,

di di R di R
L + Ri = 0 L dt = − idt = − dt
dt dt L i L
i (t )dx R t i (t ) R
∫i ( t 0 ) x = −
L ∫
t0
dy ln
i (0 )
=− t
L i (t ) = I0e − t /τ

τ = time constant = L/R

I0 = i(0)
• A similar analysis can be done for an RC

Wei Lecture 2 6
 Step Response of an RC Circuit

• Now, let’s find the step response of an RC circuit using the following example
circuit. t=0
A

is R C vC
i

– Summing the current around node A gives…

dv C v C dv C v i dv C
C + = is + C = s =−
1
(v C − i s R )
dt R dt RC C dt RC

dv C 1 vC (t )dx 1 t v C (t ) − i s R 1
v C − isR
=−
RC
dt ∫v C (0 ) x − i s R RC
= − ∫0
dτ ln
v C (0) − i s R
=−
RC
t

t  −
t
 for t > 0 and
v C ( t ) = i s R + (v C ( 0 ) − i s R )e
−
RC v C ( t ) = i s R  1 − e RC 
 vC(0) = 0
 

Wei Lecture 2 7
 Natural Response of Parallel RLC

• Throwing all of the components into the mix leads to a more interesting problem.
Let’s look at the natural response of a parallel RLC circuit.

R L I0 C V0 v

• I0 and V0 are initial conditions. Solve for v (sum currents)

v 1 t dv 1 dv v d 2v
+ ∫ vdτ + I0 + C =0 differentiate w.r.t t + +C 2 = 0
R L 0 dt R dt L dt

d 2v 1 dv v
+ + =0
dt 2 RC dt LC

Wei Lecture 2 8
 General Solution for 2nd Order Differential Equation

• First, let’s assume the answer is of exponential form, v = Ae st , then we can

rewrite the differential equation as
As st A st  s 1 
As 2e st + e + e =0 Ae st  s 2 + + =0
RC LC  RC LC 
• For the above equation to be a solution, A must equal 0 or the term in the
parentheses must equal 0. A cannot equal zero since this means voltage is 0 for
all time and that cannot be the case if there is some initial energy. Then, the
characteristic equation of the differential equation is
s 1
s2 + + =0
RC LC
because the roots of this quadratic equation determine the mathematical
character of v(t). The roots are…
2
1  1  1
s1, 2 =− ±   −
2RC  2RC  LC

Wei Lecture 2 9
 v (t ) = A1e s1t v (t ) = A2e s2t
• So then, either of the following two are
possible solutions

and the sum is also a possible solution v (t ) = v1 (t ) + v 2 (t ) = A1e s1t + A2e s2t

then… dv d 2v
= A1s1e s1t + A2s2e s2t 2
= A1s1
2 s1t
e + A2 s 2 s2t
2 e
dt d t
and combining the equations…
 s 1  2 s t 2 s 1 
A1s12e s1t  s12 + 1 +  + A2s2 e 2  s2 + 2 + =0
 RC LC   RC LC 
Each parenthetical term is 0 by definition,
since s1 and s2 are roots of the characteristic
equation. So, the natural response of the
RLC circuit is…
v = A1e s1t + A2e s2t

Wei Lecture 2 10
 • The behavior of v(t) depends on
– s1 and s2 which depend on R, L, and C
– Initial conditions set A1 and A2
• Three possible forms of behavior of the solution depending the R, L, and C
– overdamped
– underdamped
– critically damped
• But first, let’s define some terminology

s1, 2 = −α ± α 2 − ω 0
2
characteristic roots

1
α= Neper frequency
2RC
1
ω0 = Resonant radian frequency
LC

Wei Lecture 2 11
 Overdamped Response

• When the roots of the characteristic equation are real and distinct (α2 > ω02), the
voltage response of a parallel RLC circuit is said to be overdamped. The
solution is of the form
v = A1e s1t + A2e s2t
• To solve for A1 and A2,

+ dv (0 + )
v (0 ) = A1 + A2 and = s1 A1 + s 2 A2
dt
also,

dv (0 + ) iC (0 + ) V
= = − 0 − I0
dt C R

• Simultaneously solve the above equations.

• Substitute values to find v(t) for t>0

Wei Lecture 2 12
 Underdamped Response

• When α2 < ω02, the roots of the characteristic equation are complex, and the
response is underdamped. Rewriting the characteristic roots equation
conveniently gives…
( 2
)
s1 = −α + − ω 0 − α 2 = −α + j ω 0 − α 2
2
j = −1
s1 = −α + jω d
ωd = ω0 − α 2
2
s2 = −α − jω d
• To solve, we use Euler’s identity (other trig identities (pdf)) à e ± jθ = cos θ ± j sin θ
• Going back to the general form of the solution, we get

v = A1e s1t + A2e s2t v (t ) = A1e (−α + jω d )t + A2e − (α + jωd )t v (t ) = A1e −αt e jω d t + A2e −αt e − jωd t

v (t ) = e −αt (A1 cos ω d t + jA1 sin ω d t + A2 cos ω d t − jA2 sin ω d t )

v (t ) = e −αt [(A1 + A2 ) cos ω d t + j (A1 − A2 )sin ω d t ]

v (t ) = B1e −αt cos ω d t + B2e −αt sin ω d t

Wei Lecture 2 13
 • Note that B1 and B2 are real for an underdamped response. This is b/c A1 and
A2 are complex conjugates of each other. Using B1 and B2 just makes the
calculations easier. B1 and B2 are a function of the initial energy stored on the
capacitor and in the inductor. The two simultaneous equations to solve B1 and
B2 are…
v (0 ) = V0 = B1 and
+ dv 0 +( )
=
i C (0 + )
= −αB1 + ω d B2
dt C
+ V
where iC (0 ) + I 0 + 0 = 0 from KCL
R

• Some interesting characteristics…

– The answer is a damped sinusoid at a frequency set by ωd (where α sets
how quickly the amplitude of the sinusoid diminishes)
– As the dissipative losses decreases (i.e., R à infinity), α à 0 and ωd à ω0
• The response is a sinusoid at the resonant frequency

Wei Lecture 2 14
 • An underdamped voltage response
α sets the damping
ωd sets the frequency of oscillation

Wei Lecture 2 15
 Critically Damped Response

• When α2 = ω02 the RLC circuit is critically damped. This is the point when the
circuit is on the verge of oscillating, but ωd=0. Let’s look at the solution for a
critically damped response…
1
s1 = s2 = −α = −
2RC
However, the solution cannot take the same form as before. The equation
below cannot satisfy two independent initial conditions V0 and I0
v = ( A1 + A2 )e −αt = A0e −αt
This is because when the roots of the characteristic equation are equal, the
solution for the differential equation takes a different form…

v (t ) = D1te −αt + D2e −αt

Then, solve for D1 and D2

( )
+
v 0 = V0 = D2
( )
dv 0 +
=
iC (0 + )
= D1 − αD2
dt C

Wei Lecture 2 16
 • A critically damped response
– On verge of oscillating (but doesn’t)

Wei Lecture 2 17
 Step Response of Parallel RLC

• Now, let’s see what happens when we drive the parallel RLC with a step as
shown in the figure

iR iL iC
t=0
I R L C v

v 1 t dv
– KCL gives + ∫ vdτ + I 0 + C =I
R L 0 dt

– differentiate and get d 2v 1 dv v

+ + =0
dt 2 RC dt LC
– So, the solutions for the three forms of damped responses are the same as
before (in the natural response), but now you must take the I into account
when you solve for the coefficients.

Wei Lecture 2 18
 Example of Overdamped Step Response

iR iL iC
t=0
I R L C v

• Assume α2 > ω02 and therefore the circuit is overdamped.

So, solution is of the form (where s1 and s2 are real):

v = A1e s1t + A2e s2t

From KCL, we know that:
v dv
I = i R + iC + i L = +C + iL
R dt
Then, using the above solution for v, we get the following relationship:

i L = I + k1e s1t + k 2e s2t

Let’s now answer a few questions
– What’s the initial value of iL?
• Initially it’s 0 since there is no energy stored and when the switch closes, still 0 since current
through an inductor cannot change instantaneously
– What’s the initial value of v?
• Initially 0 and when switch closes, voltage across a capacitor cannot change instantaneously

Wei Lecture 2 19
 • Then, we get…
di L (0 )
i L (0) = I + k1 + k 2 = 0 and = v (0) = I + s1k1 + s2 k 2 = 0
dt
since we know s1 and s2, solve for k1 and k2.
• To solve for v(t)
di L
v (t ) = L
dt
• iL for three responses:

Wei Lecture 2 20
 Solutions for RLC circuits

• One can perform a similar analysis for series RLC circuits and get a
similar set of solutions.

parallel RLC series RLC

d 2v 1 dv v d 2i 1 di i
differential equation 2
+ + =0 2
+ + =0
dt RC dt LC dt RC dt LC

overdamped v (t ) = A1e s1t + A2e s 2t i (t ) = A1e s1t + A2e s 2t

underdamped v (t ) = B1e −αt cos ω d t + B2 e −αt sin ω d t i (t ) = B1e −αt cos ω d t + B2e −αt sin ω d t

critically damped v (t ) = D1te −αt + D2 e −αt i (t ) = D1te −αt + D2e −αt

Wei Lecture 2 21
 Next lecture

• So far, we’ve developed a set of tools to analyze the response of circuits in time.
Next lecture, we will pick up a set of tools that enables us to analyze circuits with
respect to frequency. This requires a transformation of the equations that govern
circuit operations from the time-domain to the frequency-domain. We will see
how Laplace transforms enable us to do so.

• Reading:

• Supplemental Reading:
– Nilsson Chapters 9, 12-13

Wei Lecture 2 22