Discussion 3B @ 2019-10-06 13:32:46-07:00

EECS 16B Designing Information Devices and Systems II

Fall 2019 Discussion Worksheet Discussion 3B

1 Complex Numbers

Im{z}

y z
Imaginary Axis

|z|

θ x
Re{z}
Real Axis

Figure 1: Complex number z represented as a vector in the complex plane.

A complex
√ number z is an ordered pair (x, y), where x and y are real numbers, written as z = x + jy where
j = −1. A complex number can also be written in polar form as follows:

z = |z|e jθ

The magnitude z is denoted as |z| and is given by

p
|z| = x 2 + y2 .

The phase or argument of a complex number is denoted as θ and is given by

θ = atan2(y, x).

Here, atan2(y, x) is a function that returns the angle from the positive x-axis to the vector from the origin to
the point (x, y) 1 .
The complex conjugate of a complex number z is denoted by z (or might also be written z∗ ) and is given by

z = x − jy.

Euler’s Identity is
e jθ = cos(θ ) + j sin(θ ).
 
y
1 See its relation to tan−1 x at https://en.wikipedia.org/wiki/Atan2.

© UCB EECS 16B, Fall 2019. All Rights Reserved. This may not be publicly shared without explicit permission. 1
Discussion 3B @ 2019-10-06 13:32:46-07:00

With this definition, the polar representation of a complex number will make more sense. Note that

|z|e jθ = |z| cos(θ ) + j |z| sin(θ ).

The reason for these definitions is to exploit the geometric interpretation of complex numbers, as illustrated
in Figure 1, in which case |z| is the magnitude and e jθ is the unit vector that defines the direction.

2 Useful Identities
Complex Number Properties Complex Algebra

Rectangular vs. polar forms: z = x + jy = |z|e jθ Let z1 = x1 + jy1 = |z1 |e jθ1 , z2 = x2 + jy2 = |z2 |e jθ2 .

√ p (Note that we adopt the easier representation between

where |z| = zz = x2 + y2 , θ = atan2(y, x). We can rectangular form and polar form.)
also write x = |z| cos θ , y = |z| sin θ .
Addition: z1 + z2 = (x1 + x2 ) + j(y1 + y2 )
Euler’s identity: e jθ = cos θ + j sin θ
Multiplication: z1 z2 = |z1 ||z2 |e j(θ1 +θ2 )
e jθ − e− jθ e jθ + e− jθ z1 |z1 | j(θ1 −θ2 )
sin θ = , cos θ = Division: = e
2j 2 z2 |z2 |
Complex conjugate: z = x − jy = |z|e− jθ Power: zn1 = |z1 |n e jnθ1
1 1 θ1
(z + w) = z + w, (z − w) = z − w z12 = ±|z1 | 2 e j 2

(zw) = zw, (z/w) = z/w

Useful Relations
z = z ⇔ z is real
(zn ) = (z)n −1 = j2 = e jπ = e− jπ
π √
j = e j 2 = −1
π π
− j = −e j 2 = e− j 2
p π 1 π ±(1 + j)
j = (e j 2 ) 2 = ±e j 4 = √
2
π 1 π ±(1 − j)
− j = (e− j 2 ) 2 = ±e− j 4 = √
p
2

3 Phasors
We consider sinusoidal voltages and currents of a specific form:

Voltage v(t) = V0 cos(ωt + φv )

Current i(t) = I0 cos(ωt + φi )

where,

(a) V0 is the voltage amplitude and is the highest value of voltage v(t) will attain at any time. Similarly,
I0 is the current amplitude.

(b) ω is the angular frequency of oscillation. ω is related to frequency by ω = 2π f . Frequency f is

the number of oscillation cycles that occur per second. If T is the period of the sinusoid (that is, the
amount of time it takes for one complete cycle to occur) the frequency is f = T1 .

© UCB EECS 16B, Fall 2019. All Rights Reserved. This may not be publicly shared without explicit permission. 2
Discussion 3B @ 2019-10-06 13:32:46-07:00

V (t)
V0

T t(s)
−V0

(c) φv and φi are the phase terms of the voltage and current respectively. These capture a delay, or a shift
in time, of the sinusoid.

We know from Euler’s identity that e jθ = cos(θ ) + j sin(θ ). Using this identity, we can obtain an expression
for cos(θ ) in terms of an exponential:
1 1
cos(θ ) = e jθ + e− jθ
2 2
Extending this to our voltage signal from above:

V0 jωt+ jφv V0 − jωt− jφv V0 jφv jωt V0 − jφv − jωt

v(t) = V0 cos(ωt + φv ) = e + e = e e + e e (1)
2 2 2 2
The coefficient of the e jωt term in Equation 1 is called the phasor form of this signal:
1
Ve = V0 e jφv
2

The complex conjugate of the phasor is the coefficient of the e− jωt term in Equation 1 and is given by:
1
Ve = V0 e− jφv
2
The phasor representation is a constant that contains the magnitude and phase information of the signal. The
time-varying part of the signal does not need to be explicitly represented, because it is given by e jωt , which
is always implicit when using phasors. Phasors let us handle sinusoidal signals much more easily. They are
powerful because they let us use DC-like (16A style) circuit analysis techniques, which we already know, to
analyze circuits with sinusoidal voltages and currents.
Note: We can only use phasors if we know that all of our signals are sinusoids. We will see later in 16B
(Module 4) why this is not too restrictive a condition.
Within this standard form, the phasor domain representation is as follows. The general equation that relates
cosines to phasors is below, where Ve is the phasor.

V0 cos(ωt + φv ) = Ve e jωt + Ve e− jωt

The standard forms for voltage and current phasors are given below:

Voltage Ve = 12 V0 e jφv
Current Ie = 12 I0 e jφi

We define the impedance of a circuit component to be Z = VIe , where Ve and Ie represent the voltage across and
e

the current through the component, respectively. For capacitors and inductors, this impedance will generally
depend on jω.

© UCB EECS 16B, Fall 2019. All Rights Reserved. This may not be publicly shared without explicit permission. 3
Discussion 3B @ 2019-10-06 13:32:46-07:00

3.1 Phasor Relationship for Resistors

i
+

R v

Figure 2: A simple resistor circuit

Consider a simple resistor circuit as in Figure 2, with current being

I0 jφ jωt I0 − jφ − jωt
i(t) = I0 cos(ωt + φ ) = e e + e e
2 2

By Ohm’s law,

v(t) = i(t)R
I0 R jφ jωt I0 R − jφ − jωt
= I0 R cos(ωt + φ ) = e e + e e
2 2

In the phasor domain, for both s = + jω and s = − jω

Ve = RIe

We usually refer to the impedance of the resistor, ZR , in the phasor domain. Since ZR = R, we can also write:

Ve = ZR Ie

3.2 Phasor Relationship for Capacitors

i
+

C v

Figure 3: A simple capacitor circuit

Consider a capacitor circuit as in Figure 3, with voltage being

v(t) = V0 cos(ωt + φ )

© UCB EECS 16B, Fall 2019. All Rights Reserved. This may not be publicly shared without explicit permission. 4
 Discussion 3B @ 2019-10-06 13:32:46-07:00

By the capacitor equation, and expanding cos in the complex exponential domain:

v(t) = V0 cos(ωt + φ )
V0 V0
= e j(ωt+φ ) + e− j(ωt+φ )
2 2
V0 jφ jωt V0 − jφ − jωt
= e e + e e
2 2
jωt − jωt
= V e +V e
e e

d
i(t) = C v(t)
dt  
d V0 jφ jωt V0 − jφ − jωt
=C e e + e e
dt 2 2
V0 V0
= jωC e jφ e jωt − jωC e− jφ e− jωt
2 2

Considering the s = + jω term:

V0
Ie = jωC e jφ = jωCVe
2
This can also be derived in the sinusoid domain:

i(t) = C dv
dt (t)
= −CV0 ω sin(ωt + φ )
 !
π
= −CV0 ω − cos ωt + φ +
2
 
π
= CV0 ω cos ωt + φ +
2
 
π
= (ωC)V0 cos ωt + φ +
2

In the phasor domain,

π
Ie = ωCe j 2 Ve = jωCVe

The impedance of a capacitor is an abstraction to model the capacitor in a manner similar to a resistor in the
phasor domain. This is denoted ZC , which is given by:

Ve 1 1
ZC = = =
I
e jωC sC
for s = jω.

1. Cosine vs Sine Relation

(a) What value of φ makes the following relation true?

sin(ωt) = cos(ωt + φ )

© UCB EECS 16B, Fall 2019. All Rights Reserved. This may not be publicly shared without explicit permission. 5
 Discussion 3B @ 2019-10-06 13:32:46-07:00

Answer: Expanding sin and cos into their complex exponentials:

1 jωt 1 − jωt
sin(ωt) = e − e
2j 2j
− j jωt − j − jωt
= e − e
2 2
1 π 1 π
= e− j 2 e jωt + e j 2 e− jωt
2 2
1 j(ωt− π ) 1 − j(ωt− π )
= e 2 + e 2
2  2

π
= cos ωt −
2

2. Inductor Impedance
di
Given the voltage-current relationship of an inductor V = L dt , show that its complex impedance is ZL = jωL.
Answer:
i
+

L v

Figure 4: A simple inductor circuit

Consider a simple resistor circuit as in Figure 4, with current being

i(t) = I0 cos(ωt + φ )

By the inductor equation,

di
v(t) = L dt (t)
= −LI0 ω sin(ωt + φ )
 
π
= LI0 ω cos ωt + φ +
2
 
π
= (ωL)I0 cos ωt + φ +
2

In the phasor domain,

π
Ve = ωLe j 2 Ie = jωLIe

The impedance of an inductor is an abstraction to model the inductor as a resistor in the phasor domain.
This is denoted ZL .

Ve
ZL = = jωL
Ie

© UCB EECS 16B, Fall 2019. All Rights Reserved. This may not be publicly shared without explicit permission. 6
 Discussion 3B @ 2019-10-06 13:32:46-07:00

3. Complex Matrix Inverse

Consider a complex matrix
M = Mr + jMi
and its inverse
N = Nr + jNi

(a) Show that the inverse of M = Mr − jMi (the complex conjugate of M) is equal to N = Nr − jNi (the
complex conjugate of N).
Answer: Because N is the inverse of M, we know

I = NM = (Nr + jNi )(Mr + jMi )

= (Nr Mr − Ni Mi ) + j(Nr Mi + Ni Mr ) ⇒ I = Nr Mr − Ni Mi 0 = Nr Mi + Ni Mr

Note that we isolated the real and imaginary terms in the product. Because the product of a matrix and
its inverse must be the identity matrix, the real terms must equal the identity matrix, and the imaginary
terms must be 0.
Similarly, we try evaluating the product of NM:

NM = (Nr − jNi )(Mr − jMi )

= (Nr Mr − Ni Mi ) − j(Nr Mi + Ni Mr )
= I − j0
=I

−1
Thus, M = N = M −1 .
The inverse of the complex conjugate of a matrix is equal to the complex conjugate of the matrix’s
inverse.

4. Phasor Analysis
Any sinusoidal time-varying function x(t), representing a voltage or a current, can be expressed in the form

x(t) = Xe e − jωt ,
e jωt + Xe (2)

where Xe is a time-independent function called the phasor representation of x(t) (recall that a denotes the
complex conjugate of a). Note that 1) Xe and Xe are complex conjugates of each other, 2) e jωt and e− jωt are
complex conjugates of each other, and 3) that Xe e jωt and Xe e− jωt are also complex conjugates of each other.
The phasor analysis method consists of five steps. Consider the RC circuit below.

R i(t)

+
+
vs vC (t) C
−
−

© UCB EECS 16B, Fall 2019. All Rights Reserved. This may not be publicly shared without explicit permission. 7
Discussion 3B @ 2019-10-06 13:32:46-07:00

The voltage source is given by

 
π
vs (t) = 12 sin ωt − , (3)
4
√
with ω = 1 × 103 rad
s ,R= 3kΩ, and C = 1 µF.
Our goal is to obtain a solution for i(t) with the sinusoidal voltage source vs (t).

(a) Step 1: Write sources as exponentials: Ae e − jωt

e jωt + Ae
All voltages and currents with known sinusoidal functions should be expressed in the standard expo-
nential format. Convert vs (t) into a exponential and write down its phasor representation Ves .
Answer:
 
π
vs (t) = 12 sin ωt −
4
 
1 j(ωt− π ) 1 − j(ωt− π )
= 12 e 4 − e 4
2j 2j
12 −π π 12 π π
= e j 2 e− j 4 e jωt + e j 2 e j 4 e− jωt
2 2
− j 3π 3π
−
= 6e 4 e + 6e 4 e jωt
jωt j

The phasor is given by the term in front of e jωt

3π
Ves = 6e− j 4

(b) Step 2: Transform circuits to phasor domain for s = + jω

The voltage source is represented by its phasor Ves . The current i(t) is related to its phasor counterpart
Ie by
e − jωt .
e jωt + Ie
i(t) = Ie
What are the impedances of the resistor, ZR , and capacitor, ZC ? We sometimes also refer to this as the
"phasor representation" of R and C.
Answer:

ZR = R
1
ZC =
jωC

(c) Step 3: Cast the branch and element equations in phasor domain
Use Kirchhoff’s laws to write down a loop equation that relates all phasors in Step 2.
Answer:
The KVL equations for this circuit are:

vs (t) = vR (t) + vC (t),

© UCB EECS 16B, Fall 2019. All Rights Reserved. This may not be publicly shared without explicit permission. 8
Discussion 3B @ 2019-10-06 13:32:46-07:00

where we have denoted the voltage across the resistor as vR (t). If you expand all these quantities using
phasors (using Equation (2)), we get

Ves e jωt + Ves e− jωt = VeR e jωt + VeR e− jωt + VeC e jωt + VeC e− jωt
| {z } | {z } | {z }
vs (t) vR (t) vC (t)

Collecting together all the e jωt terms and all the e− jωt terms, the above equation can be rewritten as

(Ves − VeR − VeC )e jωt + (Ves − VeR − VeC )e− jωt = 0.

If ω 6= 0,2 it can be shown3 that both of the coefficients of e jωt and e− jωt in the above equation must
be equal to 0 for this equation to hold. That is:

Ves − VeR − VeC = 0, and

Ves − VeR − VeC = 0.

Both of the equations above have the same meaning, i.e., Ves − VeR − VeC = 0 or

Ves = VeR + VeC .

This is exactly the same KVL equation as given by vs (t) = vR (t)+vC (t), but using phasors. In the same
way, you can show that KCL is also obeyed by phasors. This conclusion implies that the standard rules
for putting together circuit equations using KCL and KVL work identically with phasors as with time-
varying notation. The only difference is that the voltage-current relationships of elements should be in
phasor form, e.g., Ie = jωCVeC = Z1C VeC .
We can apply what we’ve found above to write the circuit in the phasor domain:

ZR Ie+ ZC Ie = Ves
 
1 3π
R+ Ie = 6e− j 4
jωC

Now that we’ve shown that the phasor representation (i.e., Ie and Ve ) of our circuit is equivalent to the
time-varying representation (i.e., i(t) and v(t)), in the future we can write the KCL and KVL equations
in phasor form directly.
(d) Step 4: Solve for unknown variables
Solve the equation you derived in Step 3 for Ie and VeC . What is the polar form of Ie and VeC ? Polar form
is given by Ae jθ , where A is a positive real number.
Answer:
3π 3π
6e− j 4 j6ωCe− j 4
Ie = 1
=
R + jωC jωRC + 1
3π 3π
−j 4 1 6e− j 4
e C = j6ωCe
VeC = IZ · =
1 + jωRC jωC 1 + jωRC
2 Try working out the ω = 0 case by yourself! It’s even easier.
3 This can be shown because the functions e jωt and e− jωt are linearly independent.

© UCB EECS 16B, Fall 2019. All Rights Reserved. This may not be publicly shared without explicit permission. 9
 Discussion 3B @ 2019-10-06 13:32:46-07:00

To derive the polar form, we plug in for the values of the circuit elements:
3π 3π π 3π
j6ωCe− j 4 j6 · 103 · 10−6 · e− j 4 (e j 2 ) · 6 · 10−3 · e− j 4
Ie = = √ = √
1 + jωRC 1 + j · 103 · 3 × 103 · 10−6 1+ j 3
3π π
6 · 10−3 · e− j 4 e j 2 − j 7π
= π = 3e 12 mA.
2e j 3
3π 3π 3π
6e− j 4 6e− j 4 6e− j 4 − j 13π
VeC = = √ = π = 3e 12 V
1 + jωRC 1 + j 3 2e j 3
(e) Step 5: Transform solutions back to time domain
To return to time domain, we apply the fundamental relation between a sinusoidal function and its
phasor counterpart. What is i(t) and vC (t)? What is the phase difference between i(t) and vC (t)?
Answer:
 
e − jωt = 3e− j 7π 7π 7π
e jωt + Ie
i(t) = Ie 12 e jωt + 3e j 12 e− jωt = 6 cos ωt − mA
12
 
jωt − jωt − j 13π
jωt j 13π
− jωt 13π
vC (t) = VeC e + VeC e = 3e 12 e + 3e 12 e = 6 cos ωt − V
12
The phase difference between the vc (t) and i(t) with respect to i(t) is − π2 .

5. RLC Circuit Phasor Analysis

We study a simple RLC circuit with an AC voltage source given by
vs (t) = B cos(ωt − φ )

C
R L

− vs +

(a) Write out the phasor representations of vs (t), R, C, and L.

Answer:

B 1
Ves = e− jφ , ZR = R, ZC = , ZL = jωL
2 jωC
(b) Use Kirchhoff’s laws to write down a loop equation relating the phasors in the previous part.
Answer:

ZR Ie+ ZC Ie+ ZL Ie = Ves

 
1 B
R+ + jωL Ie = e− jφ
jωC 2

© UCB EECS 16B, Fall 2019. All Rights Reserved. This may not be publicly shared without explicit permission. 10
Discussion 3B @ 2019-10-06 13:32:46-07:00

(c) Solve the equation in the previous step for the current I.
e What is the magnitude and phase of the polar
form of I?
e
Answer:
B − jφ B − jφ
2e 2e
Ie = 1
= 1
R+ jωC + jωL R+ j(− ωC + ωL)

The magnitude of Ie is
B
2
|I|
e=q
2 1
R + (− ωC + ωL)2

The phase of Ie is  
1
∠I = −φ − atan2 ωL −
e ,R
ωC

Contributors:

• Brian Kilberg.

• Siddharth Iyer.

• Regina Eckert.

• Jaijeet Roychowdhury.

• Sidney Buchbinder.

• Anant Sahai.

• Yen-Sheng Ho.

© UCB EECS 16B, Fall 2019. All Rights Reserved. This may not be publicly shared without explicit permission. 11