Electric Networks (ELE 302): Notes

Adam Szava
Fall 2021

Introduction
This is my compilation of notes from Electric Networks (ELE 302) from Ryerson
University. All information comes from my professor’s lectures, the textbook
Fundamentals of Electric Circuits, and online resources.

Chapter 5: Operational Amplifiers

The operational amplifier or op amp for short is an active circuit element (mean-
ing it supplies energy to the circuit) which acts as a voltage controlled voltage
source.
An op amp takes in two signals (through two pins) and does some operation
to them, then amplifies it, and then returns the signal through a new pin. Op-
erations that an op amp can do include: addition, subtractions, multiplication,
division, differentiation, and integration. Op amps are made of a complex ar-
rangement of transistors, we will consider the op amp to be a circuit element
and not concern ourselves with what exactly inside.
The symbol for an op amp is:

You can think of the op amp as this equivalent circuit:

1
As you can see, the output voltage is controlled by the voltage controlled voltage
source. The voltage is vd which is the difference between v2 and v1 , multiplied
by some A which is the amplification multiplier (also called open loop gain).

vo ≈ Avd = A(v2 − v1 )

It is approximate since there is a resistor there with a value Ro , however ideally

that resistance is negligible. More on this later.
An op amp is an active element which means it needs to be powered, this is
usually omitted in circuit diagrams but cannot be forgotten in practical appli-
cations. A diagram including the power would be as follows:

Where the numbers refer to this diagram of a real op amp:

2
 We do not care about pins 1,5, or 8. Pins 4 and 7 handle active power. Pins
2 and 3 hand input, and pin 6 handles output. If you apply a voltage to the
inverting input, it will appear with the opposite sign in the output, if you apply
a voltage to the non-inverting input it will appear with the same sign in the
output.
Next we compare ideal versus real values seen for the elements within the
op amp:
Variable Ideal Real
A ∞ 10 → 108
5

Ri ∞Ω 105 → 1013 Ω
Ro 0Ω 10 − 100Ω
The ideal Ri is infinite, this is because as Ri → ∞ =⇒ iin → 0. We want
the input current to be zero to not change the input signals. A is called the
open loop feedback due to the configuration. We will now discuss closed loop
feedback configurations.

Negative Feedback Configuration of an Op Amp

You can connect the output of the op amp to the negative input, and get a
circuit such as the following:

3
 One interesting fact we usually want to know about configurations of op
amps is the ratio vvoi . We can solve for this:

vo = A(vi − vo )
vo A
=
vi A+1
But remember A → ∞, so:
vo A
= lim
vi A→∞ A + 1

vo
= 1 =⇒ vo = vi
vi
Generally you can think of this configuration as a negative feedback loop,
which brings vo closer and closer to vi every unit of time, using the equation:

vo(new) = A(vi − vo(old) )

Assumptions
For an idea op amp in this negative feedback configuration we can make the
following assumptions:
1. The input current to either pin is zero.

i1 = i2 = 0

2. The voltage difference across the input pins is zero.

v2 − v1 = 0 =⇒ v2 = v1

4
These two equations will let you solve problems using an op amp in this config-
uration.
There is a practical limitation of real op amps which is important for lab 1.
Remember that this component needs to be powered by an outside source, and
so the output voltage cannot exceed the input voltage that powers the entire
circuit block. This voltage is called VCC , and so:

−VCC ≤ vo ≤ VCC

This means that the output of the op amp (vo ) can run in three modes in relation
to VCC :

1. Positive saturation vo = VCC

2. Negative saturation vo = −VCC
3. Linear region −VCC ≤ vo ≤ VCC

Specific Configurations of the Op Amp

Inverting Amplifier
In this configuration, the noninverting input is grounded, while the other input
is connected to the voltage source though a resistor. There is a negative feedback
loop with a resistor. The goal is to know how vo relates to vi .

5
Through the use of:
v1 = v2
ia = ib = 0
Along with KCL at the inverting input:
i1 = i2
We can derive that:
Rf
vo = − vi
R1
You can control the value of vo by adjusting Rf and R1 . Notably, the output
voltage will be of the opposite sign of the input, as in it was inverted.

Non-Inverting Amplifier
In this configuration vi is applied the noninverting input, with a resistor between
the ground and the inverting input. There is a negative feedback loop with a
resistor. The goal is to know how vo relates to vi .

6
Through the use of:
v1 = v2
ia = ib = 0
Along with KCL at the inverting input:

i1 = i2

We can derive that:

Rf
vo = (1 + )vi
r1
You can control the value of vo by adjusting Rf and R1 .

Summing Amplifier
In this configuration, three voltages are applied v1 , v2 , v3 , and the output is the
sum of the three inputs. There is negative feedback loop with a resistor.

Through the use of:

v1 = v2
ia = ib = 0
Along with KCL at the inverting input:

i1 + i2 + i3 − i = 0

We can derive that:

 
Rf Rf Rf
vo = − v1 + v2 + v3
R1 R2 R3

This is called the weighted sum of all the voltages, as it takes into account each
voltage’s resistance. In the case thatR1 = R2 = R3 = Rf then:

vo = −(v1 + v2 + v3 )

This extends to the sum of n voltages.

7
Difference Amplifier
In this configuration, the difference between two inputs are amplified. There is
a negative feedback loop with a resistor.

Through the use of:

v1 = v2
ia = ib = 0
Along with KCL at note b, we can derive that:

R2
vo = (vb − va )
R1

... and if R2 = R1 :
vo = v b − va
... and it would be called a subtractor.
This is different than a regular op amp with no negative feedback because
now you can control the amplification by controlling R
R1 .
2

Cascading Amplifiers
You can put op amps that feed into each other in a line in whatever configuration
you want. Let’s say you have three op amps in a row, which feed into each other,
all with some amplification factor of A, like the following image:

8
Then we know that:
v 0 = A1 A2 A3 v 1
Which is true generally for n cascaded op amps.

Chapter 8: Second Order Circuits

In this chapter we study second order circuits, which are circuits characterized
by a second-order differential equation. Similar to how we did first order circuits
in ELE202, here second order circuits are interested in those with two energy-
storage elements (inductors and capacitors).
Remember that in complex circuits, you only need to go through the whole
process of finding one of the voltages or currents in the circuit, and from there
you can deduce the rest using more elementary methods.

8.2 Finding Initial and Final Values

DEs in this course will always be initial value problems, as we want to describe
how a circuit is changing over time, given some initial event. We also need to
know end behavior.
In second order circuits, you need to know the initial values of i(t), v(t) but
also i0 (t), v 0 (t), depending on which circuit variable you are trying to track.
Unless otherwise mentioned, v denotes capacitor voltage, and i denotes in-
ductor current.
The best way to find these values in a second order circuit is to draw the
circuit in all of it’s states:
1. Given diagram.

2. t < 0 (DC steady state)

9
 3. t = 0 (Switch just closed)

4. t → ∞ (DC steady state)

Recall that over a capacitor:

v(0+ ) = v(0− )

As in the voltage is continuous. Or in an inductor:

i(0+ ) = i(0− )

As in the current is continuous. This means you can find the initial values of
either current of voltage by finding those values before the event at time t = 0
(before the switch closes). This is the state 2 diagram above.
Recall that for a capacitor, the following is true:

dv
iC (t) = C
dt

... and for an inductor:

di
vL (t) = L
dt
You can use these equations to solve for v 0 (0) and i0 (0). Make sure that you
find the correct current and voltage values.

8.3 The Source-Free Series RLC Circuit

In this section we analyze the following circuit:

10
 Notably this circuit has no independent sources, and so all of its energy is
that which is stored inside the inductor and capacitor at time t = 0, meaning:

v(0) = V0

i(0) = I0
Applying KV L around the loop we get:

vresistor + vL + vC = 0

We will choose to write everything in terms of the current, recalling that:

vresistor = Ri
di
vL = L
dt
Z t
dv 1
iC = C =⇒ v(t) = i(τ )dτ
dt C −∞

... which gives us: Z t

di 1
Ri + L + i(τ )dτ
dt C −∞

... differentiating both sides with respect to time to get rid of the integral:
i
Ri0 + Li00 + =0
C
... and putting it in standard form:

d2 i R di 1
+ + i=0
dt2 L dt LC
This is a second order LDE which can have three possible types of solutions.

11
 The solution to this DE depends on the solutions to it’s characteristic poly-
nomial (s1 , s2 ): s 
2
R R 1
s1 = − + −
2L 2L LC
s 
2
R R 1
s1 = − − −
2L 2L LC
These solutions are measured in nepers per second (Np/s).
We define (for source-free series RLC circuits):

R
α=
2L
1
ω0 = √
LC
... where α is called the neper frequency and ω0 is called the resonant frequency.
The particular values of α and ω0 change the way the variables will change
over time, there are three cases:
1. Overdamped when α > ω0

2. Critically damped when α = ω0

3. Underdamped when α < ω0
You can find i0 (0) by using the following equation:

di(0) di(0) 1
Ri(0) + L + V0 = 0 =⇒ = − (RI0 + V0 )
dt dt L

Overdamped
In this case our solution to the DE is:

i(t) = A1 es1 t + A2 es2 t

Where you can find the values of A1 , A2 given i(0) and i0 (0).
Both of our roots are negative and real, meaning the function will decay as
t → ∞:

12
Critically Damped
In this case our solution to the DE is:

i(t) = (A1 + A2 t)e−αt

Where you can find the values of A1 , A2 given i(0) and i0 (0).

Underdamped
In this case our solution to the DE is:

i(t) = e−αt (A1 cos(ωd t) + A2 sin(ωd t))

p
... where ωd = ω02 − α2 and you can find A1 , A2 given i(0) and i0 (0).

13
 The behavior of this network is because of the damping effect, which is the
gradual loss of the initial stored energy. This happens because of the resistor,
in fact controlling the resistor can change what state the network is in. If R = 0
and all elements ideal then the response would be a perfect sinusoid, and called
loss-less. This oscillatory response happens because of the two different types
of storage elements in the circuit which flow the energy back and forth between
the inductor and capacitor.
The critically damped case is the borderline between the underdamped and
the overdamped cases, and it decays the fastest. In most practical applica-
tions we seek an overdamped circuit that is as close as possible to the critically
damped case.

8.4 The Source-Free Parallel RLC Circuit

In this section we analyze the following circuit:

Notably this circuit has no independent sources, and so all of its energy is
that which is stored inside the inductor and capacitor at time t = 0, meaning:
v(0) = V0

14
 i(0) = I0
Applying KCL at the top node we get the following differential equation (in
terms of v(t)):

d2 v 1 dv 1
2
+ + v=0
dt RC dt LC
The characteristic polynomial of this DE has solutions of:
s 2
1 1 1
s1,2 = − ± −
2RC 2RC LC
We define (for source-free parallel RLC circuits):
1
α=
2RC
1
ω0 = √
LC
Once again, the particular values of α and ω0 change the way the variables
will change over time, there are three cases:
1. Overdamped when α > ω0
2. Critically damped when α = ω0
3. Underdamped when α < ω0
In all cases, constants A1 , A2 can be determined from the initial conditions
of v(0) and v 0 (0), which we can get from the following KVL equation:
V0 dv(0) dv(0) V0 + RI0
+ I0 + C =⇒ =−
R dt dt RC

Overdamped
In this case the solution is:
v(t) = A1 es1 t + A2 es2 t

Critically Damped
In this case the solution is:
v(t) = (A1 + A2 t)e−αt

Underdamped
In this case the solution is:
v(t) = e−alphat (A1 cos(ωd t) + A2 sin(ωd t))
Where: q
ωd = ω02 − α2

15
8.5 Step Response of a Series RLC Circuit
The step response of a circuit is obtained by a sudden application of a DC source
by means of a switch. In this section we study the following circuit:

Applying KVL around the loop when t > 0, we get the following equation:
d2 v R dv v Vs
+ + =
dt2 L dt LC LC
Which is a non-homogenous LDE, whose solution can be written in the form:

v(t) = vt (t) + vss (t)

vt (t) is called the transient response and is the same as the solution to the
source-free RLC series circuit. vss is called the steady state response and is the
final value of v(t), as in:
vss (t) = v(∞)
Which in our case:
vss (t) = Vs
For this reason our solutions to the second order RLC circuit are:
ˆ Overdamped:
v(t) = Vs + A1 es1 t + A2 es2 t

ˆ Critically Damped:

v(t) = Vs + (A1 + A2 t)e−αt

ˆ Underdamped:

v(t) = Vs + (A1 cos(ωd t) + A2 sin(ωd t))e−αt

More generally, the complete response for any variable x(t) can be found
directly through:
x(t) = xt (t) + xss (t)
Where xt (t) is the transient response which dies out over time, and xss (t) =
x(∞) and represents the steady state response of the variable.

16
8.6 Step Response of a Parallel RLC Circuit
In this section we study the following circuit:

Applying KCL at the top node for t > 0 gets us the differential equation:
d2 i 1 di 1 Is
2
+ + =
dt RC dt LC LC
Which is a non-homogenous LDE¡ whose solution can be written in the form:

i(t) = it (t) + iss (t)

it (t) is called the transient response and is the same as the solution to the
source-free RLC parallel circuit. iss is called the steady state response and is
the final value of i(t), as in:
iss (t) = i(∞)
Which in our case:
iss (t) = Is
For this reason our solutions to the second order RLC circuit are:
ˆ Overdamped:
i(t) = Is + A1 es1 t + A2 es2 t

ˆ Critically damped:
i(t) = Is + (A1 + A2 t)e−αt

ˆ Underdamped:

i(t) = Is + (A1 cos(ωd t) + A2 sin(ωd t))e−αt

8.7 General Second-Order Circuits

This section describes a method to solve a general RLC circuit for a general
variable x:
1. Determine the initial conditions:

x(0), x0 (0)

17
 2. Turn off all independent sources and find the form of the transient re-
sponse xt (t) by applying KCL and KVL. Solve the DE, with two unknown
constants, in one of the following forms: overdamped, critically damped,
underdamped.
3. Obtain the steady-state response:

xss (t) = x(∞)

4. The total response is now found as the sum of the transient response and
steady-state response:

x(t) = xt (t) + xss (t)

5. Determine constants associated with the transient response by applying

the initial conditions.

Chapter 14: Frequency Response

When doing analysis of AC circuits, we have kept the frequency of the sinusoid
constant at some value ω. In this chapter we analyse the response some element
in the circuit has as a function of the change in the input variable’s frequency
parameter. We call this the circuit’s frequency response. Take for example the
following circuit:

If the source voltage is in the form:

V (t) = A cos(ωt + φ)

... then the voltage across the capacitor would also be in the form:

VC (t) = A0 cos(ωt + φ0 )

If we vary the frequency, the amplitude and phase shift of VC would change as
well. This is because:
1
V¯C (ω) =
ˆ
̂ωC
... which is a function of ω.

18
14.2 The Transfer Function
The transfer function is a complex valued function which is the analytical tool
used to study the frequency response of a circuit.
The transfer function is the frequency-dependent ratio between a forced
function and its forcing function. For example a the voltage across a resistor
would be a forced function, while the voltage source would be its forcing function.
The forcing function (also called the input) is denoted X̄(ω). The forced
function (also called the output) is denoted Ȳ (ω). By definition the transfer
function (H̄(ω)) is:
Ȳ (ω)
H̄(ω) =
X̄(ω)
ˆ since the ω and ̂ˆ will always be
Sometimes this function is denoted H̄(̂ω)
together, and this notation emphasizes that it is a complex defined function.
There are four possible transfer functions:
1. Voltage Gain:
V̄0 (ω)
H̄(ω) =
V̄i (ω)

2. Current Gain:
I¯0 (ω)
H̄(ω) = ¯
Ii (ω)

3. Transfer Impedance:
V̄0 (ω)
H̄(ω) = ¯
Ii (ω)

4. Transfer Admittance:
I¯0 (ω)
H̄(ω) =
V̄i (ω)

Note that the output of the transfer function is a complex value and so it can
be graphed as a vector on the complex plane. As ω varies, the vector will both
rotate (φ(ω)) and scale H(ω), and written in the form:

H̄(ω) = H(ω)∠φ(ω)

This means that the plot of H̄ can be broken into two, one for magnitude and
one for phase. We will learn how to plot approximations of these using the
Bode Plot technique later on.
The transfer function will always be a ration of two complex polynomials N̄
and D̄:
N̄
H̄ =
D̄
The roots of the numerator polynomial are called zeros (ω = z1 , z2 , . . . ) of the
transfer function, and represent frequencies in which the forced function stops.

19
The roots of the denominator polynomial are called poles (ω = p1 , p2 , . . . ) of
the transfer function, and represent frequencies in which the transfer function
goes to infinity.
We now have the skills to determine the transfer function of the intro circuit
(by current division):

V̄0 ˆ
1̂ωC 1
H̄(ω) = ¯ = =
Vs ˆ
R + 1/̂ωC ˆ
1 + ̂ωRC
Which can be broken up into two real valued functions:
1
H(ω) = p , φ(ω) = − arctan(ωRC)
1 + (ωRC)2
Whose plots can be seen below:

20
We will learn how to plot these graphs in a systematic way by hand, but first
we need to learn how gain is notated.

The Decibel Scale

The unit of measurement of gain between to variables is the bel defined as:
P2
GB = log10
P1
The decibel is the most commonly variant, which is just a tenth of a bel :
P2
GdB = 10 log10
P1
Recall that gain is the measure of how much bigger or smaller one variable is in
comparison to another. One of the most useful properties of using the decibel
scale is that if kdB represents an increasing gain, −kdB represents a equivalent
decreasing gain.
If the previous definition of gain is for power, the following two are for voltage
and resistance:
V2
GdB = 20 log10
V1
I2
GdB = 20 log10
I1
Decibels are used to describe the transfer function in voltage and current
gain form as they are dimensionless.

Bode Plots
In this section we describe a systematic method to sketch an approximation to
to magnitude versus frequency, and phase versus frequency plots of the transfer
function. This is called the method of Bode plots, or just Bode plots.
The magnitude in decibels is plotted against the logarithm of the frequency,
and on a separate plot the phase in degrees is plotted against the logarithm of
the frequency. The setup would look like the following:

21
Here, the red represents the plot for the phase, while the blue markings represent
the plot for the magnitude. This will typically be done on two plots, but it is
combined here to save space.
The transfer function will always be written in the form:
  2 
K(̂ω)±1 (1 + ̂ωz1 ) 1 + ̂2β1 ω
ωk + ̂ω
ωk ···
H(w) =   2 
(1 + ̂ω
p1 ) 1 + ̂2ω
ωn + ̂ω
ωn ···

This is called standard form of the transfer function and will be the basis
of how we plot Bode plots. The idea is that we learn how to plot each factor
separately, and then add them all together to make the entire plot.
Terminology:
1. Roots of the numerator are called zeros.
2. Roots of the denominator are called poles.
The dots refer to higher and higher degree terms which can appear in the transfer
function, for this course we will only have to consider constant, linear, and
quadratic terms.
The reason we use semilog paper is that it is a lot easier to graph the log
(base 10 unless otherwise specified) of the magnitude of the function for this
reason:
K(̂ω)±1 (1 + ̂ωz1 ) · · ·
20 log10 H(ω) = 20 log ̂ω
(1 + p1 ) · · ·

̂ω ̂ω
= 20 log|K| + 20 log (̂ω)±1 + 20 log 1 + + · · · − 20 log 1 + − ···
z1 p1

22
If we just graph each term individually, we can just add them all together at
the end and we will get the semilog plot.

General Process to Plot a Bode Plot

1. Label and mark your axis correctly.
2. Write your transfer function in standard form.
3. Identify all the factors.
4. For each factor, plot the magnitude and phase plot separately. Label the
slopes.
5. Add all of the magnitude plots together to get the magnitude plot of your
transfer function.
6. Add all of the phase plots together to get the phase plot of your transfer
function.

Step 4 is the hardest part since you need to memorize the process for each
type of factor. The following is a list of them all.

How to Plot the Factors

Always in the order of magnitude then phase.
1. H(ω) = K

2. H(ω) = ̂ω

23
 In the previous image, the slope is 20dB/dec where dec refers to decade
which is a since tick on the x − axis, always 10 times bigger than the last.
1
3. H(ω) = ̂ω

̂ω
4. Simple Zero: H(ω) = 1 + z1

1
5. Simple Pole: H(ω) = ̂ω
1+ p
1

 2
̂2ω ̂ω
6. Quadratic Zero: H(ω) = 1 + ωk + ωk

24
 1
7. Quadratic Pole: H(ω) = ̂ω 2

1+ ̂2ω
ωn + ωn

Once you have plotted the factors individually, you want to combine them like
done in this diagram (the image uses s instead of ω):

The best way of doing this is by adding up the slopes in all the graphs and
denoting intervals. Then connecting it to a point you know at the start of the
interval.

14.5: Series Resonance

When evaluating the plots of the magnitude of some transfer functions, there
is sometimes a sharp peek in the Bode plot which occurs at some particular
frequency. This peek is called the resonant peek and the frequency at which it
occurs is called the resonant frequency.
Resonance is a condition in an RLC circuit in which the impedance is
purely resistive, meaning the capacitive and inductive reactants are equal in
magnitude.
The following circuit is called the series resonant circuit:

25
 If we want to find it’s resonant frequency (ω0 ), we must calculate at what
frequency is the inductive and capacitive impedances equal to each other.
1
Z̄ = R + ̂ωL +
̂ωC
 
1
Z̄ = R + ̂ ωL −
ωC
Now we take just the imaginary part since that is the inductive and capacitive
impedance, and set it to 0:
1
ω0 L − =0
ω0 C
1
ω0 = √ rad/s
LC
There are a few things to note here. First of all since the impedance of the
circuit is all resistive, this means that the entire voltage drop is across the
resistor. Effectively the inductor and capacitors are short circuits at the resonant
frequency. This also means that the voltage and current are in phase.
The average power dissipated by the circuit as a function of ω is:
1 2
P (ω) = I R
2
Vm
At ω = ω0 , I = R , so:
1 Vm2
P (ω0 ) =
2 R
We are also interested in what frequencies the circuit dissipates half of the
maximum power, also called the half-power frequencies (ω1 and ω2 ). This means
that:
V2
P (ω1 ) = P (ω2 ) = m
4R

26
Those frequencies are:
s 2
R R 1
ω1,2 =∓ + +
2L 2L LC
Additionally:
√
w0 = w1 w2
These frequencies have half the power, but around 70.7% of the maximum of
the√
graph, as you can see below. This comes from dividing the maximum value
by 2.

Bandwidth is defined as the width of the peak:

B = ω2 − ω1
The sharpness of the resonant peek is measured using the quality factor (Q).
At the resonant frequency the reactive energy of the circuit oscillates between
the inductor and capacitor, the quality factor relates the maximum energy stored
to the energy dissipated in the circuit per cycle of oscillation. In words (for one
period):
Peak energy stored in the circuit
Q = 2π
Energy dissipated by the circuit
In an equation that is:
w0 L 1
Q= =
R ω0 CR
Furthermore, we can relate the bandwidth and the quality factor:
R ω0
B= =
L Q

27
From this we can see if that the quality factor of a frequency response is the
ratio of its resonant frequency to its bandwidth. As the quality factor increases
the sharpness of the transfer function also increases, becoming more selective.

All of the equations mentioned in this section only work for the series resonant
circuit.

14.6: Parallel Resonance

The following circuit is called the parallel resonant circuit:

28
Using a similar method as before (this time using the admittance) we can con-
clude:
1
ω0 = √
LC
and... s 2
1 1 1
ω1,2 =∓ + +
2RC 2RC LC
1
B = ω2 − ω1 =
RC
ω0 R
Q= = ω0 RC =
B ω0 L

For both series and parallel resonant circuits with a quality factor higher than
10:
B
ω1,2 ≈ ω0 ∓
2

14.7: Passive Filters

A filter is a circuit that is designed to pass signals with desired frequencies and
reject others. For example when you listen to the radio, a filter is only allowing
signals of a certain frequency to reach your speakers so you only hear one radio
station at a time.
A passive filter is made of RLC elements, as in only passive elements which
require no additional energy. An active filter consists of active elements as well
as the RLC elements (meaning transistors and op amps).
There are 4 types of filters, regardless of if they are implemented with an
active or passive circuit:

1. Low-pass Filter passes low frequencies and stops high frequencies.

29
2. High-pass Filter passes high frequencies and stops low frequencies.

3. Band-pass Filter passes frequencies within a frequency band, and blocks

frequencies outside the band.

4. Band-stop Filter passes frequencies outside a frequency band, and blocks

frequencies inside the band.

30
The previous diagrams are all idealized.

Low-Pass Filters
Typically low pass circuits are constructed by taking the voltage off of a RC
circuit, like for example the following:

This is just an example of a low-pass circuit. We can calculate its transfer

function:
Vo 1
H(ω) = =
Vi 1 + ̂ωRC
Which gives the following plot. Comparing it to the ideal Low-Pass Filter
transfer function you can see how it acts as one.

31
We are still concerned with the idea of a half-power frequency however now in
the context of filters we call it the cutoff frequency or the rolloff frequency (ωc ).
Once again this will be at max√
2
or approximately 70.7% of the maximum of the
transfer function.
1
ωc =
RC
This is where half the power is dissipated in the circuit compared to the maxi-
mum.
By definition, a low-pass filter is designed to pass signals from dc (ω = 0)
to the cutoff frequency (ω = ωc ).

High-Pass Filter
Typically high pass circuits are constructed by taking the voltage off the resistor
in an RC circuit. For example the following:

32
 We can then also calculate it’s transfer function:
Vo ̂ωRC
H(ω) = =
Vi 1 + ̂ωRC
Which gives the following plot. Comparing it to the ideal high-Pass filter trans-
fer function you can see how it acts as one.

1
ωc =
RC
By definition, a high-pass filter is designed to stop filters from dc (ω = 0) to
the cutoff frequency (ω = ωc ).

Band-Pass Filters
A band-pass filter is typically constructed by taking the voltage off the resistor
in the series RLC circuit:

33
We can calculate the transfer function:
Vo R
H(ω) = = 1
vi R + ̂(ωL − ωC )

Which gives the following plot. Comparing it to the ideal band-pass filter trans-
fer function you can see how it acts as one.

The filters passes signals of a frequency between it’s two half power frequencies,
centred at ω0 :
1
ω0 = √
LC
This filter is a series resonant circuit, and so it follows the equations from that
section.

Band-Stop Filters
A band-stop filter is typically constructed by taking the voltage off the LC pair
in a series RLC circuit:

34
We can calculate the transfer function:
1
Vo ̂(ωL − ωC )
H(ω) = = 1
vi R + ̂(ωL − ωC )

Which gives the following plot. Comparing it to the ideal band-stop filter trans-
fer function you can see how it acts as one.

Once again the stop band is centred at ω0 :

1
ω0 = √
LC
In this context, ω0 is called the frequency of rejection. This filter is a series
resonant circuit, and so it follows the equations from that section.

14.8: Active Filters

There some some limitations to passive filters, namely:
ˆ They can be large due to the inductors.

35
 ˆ They cannot amplify the signal, as in the maximum gain is 1.

ˆ They only function reliably at high frequencies.

Active filters are made of resistors, capacitors, and op amps which fix many of
these issues. Active filters require a source of power to operate, and to amplify
signals.

First-Order Low-Pass Filter

In general, an active first-order low/high-pass filter is in the form:

Where components selected for Zi and Zf determine whether the filter is low/high-
pass, however one of them must be reactive.
The transfer function would be:
Vo Zf
H(ω) = =−
Vi Zi
A particular example of a low-pass filter could be:

36
The transfer function is the same as the passive low-pass filter but with a con-
stant gain:
Rf 1
H(ω) =
Ri 1 + ̂ωCf Rf
Where the cutoff frequency is:
1
ωc =
Rf Cf

First-Order High-Pass Filter

In the same general configuration as before, the following is an example of an
active high-pass filter:

37
The transfer function is the same as the passive high-pass filter but with a
constant gain:
̂ωCi Rf
H(ω) = −
1 + ̂ωCi Ri
... and the cutoff frequency is:
1
ωc =
Ri Ci

Active Band-Pass Filter

By cascading active high pass and low pass filters together, and then inputting
the signal into an inverting amplifier, we can achieve an active band-pass filter:

Fully worked with op amps, the circuit looks like the following:

We can calculate the transfer function for this circuit by multiplying together
the individual transfer functions:
Vo Rf 1 ̂ωC2 R
H(ω) = =−
Vi Ri 1 + ̂ωC1 R 1 + ̂ωC2 R
A plot of which is:

38
The low pass filter sets the upper bound ω2 :
1
ω2 =
RC1
The high pass filter sets the lower bound ω1 :
1
ω1 =
RC2
... additionally:
√
w0 = ω1 ω2
B = ω2 − ω1
ω0
Q=
B
We can determine the passband gain (K) to be:
Rf ω2
K=
Ri ω1 + ω2

Active Band-Reject Filter

By combining two signals passed through filters using a summing amplifier, we
can achieve an active band-reject filter:

39
Fully worked with op amps, the circuit looks like the following:

We can calculate the transfer function for this circuit by multiplying together
the individual transfer functions:
 
Vo Rf 1 ̂ωC2 R
H(ω) = =− − −
Vi Ri 1 + ̂ωC1 R 1 + ̂ωC2 R
A plot of which is:

We can determine the passband gain (K) to be:

Rf
K=
Ri

40
Chapter 15: Introduction to the Laplace Trans-
form
In AC domain, we used ̂ω to replace differentiation. This process however
only worked for sinusoidal input functions. We want a method to turn any
arbitrary input function into a domain where differentiation is replaced with
some simpler operation. For this purpose we use the Laplace Transform because
of the following property:
L {f 0 (t)} = sF (s) − f (0)
d
... and similar rules apply for higher order derivatives. In a sense, dt =⇒
s· in Laplace Domain. Similarly integration with respect to time is replaced
with division by s in Laplace Domain. Laplace Domain is when all the circuit
elements are rewritten by applying the Laplace Transform to all their elements.
Some notable features of the edge behaviour is:
Initial Value: f (0) = lim sF (s)
s→∞

Final Value: f (∞) = lim sF (s)

s→0
If we could have an input of 1 in the Laplace domain, then our output would
be equal to the transfer function of our circuit, since:
O(s)
H(s) =
I(s)
We then need to ask what function has a Laplace inverse of 1? This is called
the Dirac-Delta Function or the Impulse function.
(
0 t 6= a
δ(t − a) =
∞ t=a
The following is true:
L −1 {1} = δ(t)
If your input to the circuit is δ(t) then your output is the transfer function.
This is impossible in real life because it would imply infinite power because of
the discontinuous jump, but it can be approximated.
In an integral, it effectively samples the function it is being multiplied by at
the input a, as in: Z ∞
δ(t − a)f (t)dt = f (a)
0

Chapter 16: Applications of the Laplace Trans-

form
We can convert all passive elements to Laplace domain using their I − V char-
acteristics.

41
Resistors in Laplace Domain
Since:
v(t) = Ri(t)
... then:
V (s) = RI(s)
... once you take the Laplace Transform of both sides:

Inductors in Laplace Domain

Since:
di(t)
v(t) = L
dt

42
... then:
1 i(0− )
I(s) = V (s) +
sL s
... once you take the Laplace Transform of both sides and rearrange. There are
two ways to realize this in a circuit:

43
Capacitors in Laplace Domain

Since:
dv(t)
i(t) = C
dt
... then:
1 v(0− )
V (s) = I(s) +
sC s
... once you take the Laplace Transform of both sides and rearrange. There are
two ways to realize this in a circuit:

44
Impedance is defined as usual as the ratio of V (s) to I(s), and so assuming zero
initial conditions it is:
Resistor: Z(s) = R
Inductor: Z(s) = sL
1
Capacitor: Z(s) =
sC
In Laplace Domain we can use all the same circuit analysis techniques like
in DC analysis:
ˆ Node voltage analysis
ˆ Mesh current analysis
ˆ Thevenin/Norton
ˆ Voltage/Current Division
Multiplication in Laplace Domain is the same as convolution in time domain:
Z ∞
v(t) = i(τ )h(t − τ )dτ = i(t) ∗ h(t)
−∞

Chapter 13: Magnetically Coupled Circuits

13.2: Mutual Inductance
Circuits are conductively coupled if one loop affects neighbouring loop through
current conduction. These are circuit couples we have been studying so far.
Think about how when doing mesh current analysis, the mesh currents combine
to form the current between two loops.

45
 When two loops with or without contacts between them affect each other
through the magnetic field generated by one of them,they are said to be mag-
netically coupled. When two inductors are in close proximity to each other, the
magnetic flux caused by current in one coil links with the other coil, inducing a
voltage in the latter. This is called mutual inductance.
Given some inductor with some current, the magnetic flux (φ) can be through
of as:

By Faraday’s law:
dφ
v=N
dt
... where N is the number of coils in the inductor. Note that the flux is caused
by a change in current and so:
dφ di
v=N
di dt

If we define L = N dφ
di , then we get:

dφ di di
v= N =⇒ v = L
di dt dt

... which is the relationship we already know about the voltage across an induc-
tor. This inductance L is called self-inductance.
Now we consider two inductors with self-inductances L1 and L2 , being in
close proximity to each other. Note also that one inductor no current, and so
has no magnetic flux.

46
As you can see, some of the flux of inductor 1 acts only on itself (φ11 can be
thought of as flux from 1 onto 1). Then some of the flux interacts with the other
inductor (φ12 ). These fluxes must add to the total flux from the first inductor
which is φ1 :
φ1 = φ11 + φ12
Now consider the voltages across either inductor:
dφ1 di1 di1
v 1 = N1 = L1
di1 dt dt
Similarly:
dφ12 di1
v 2 = N2
di1 dt
If we define a new type inductance called mutual inductance as:
dφ12
M21 = N2
di1
Then:
di1
v2 = M21
dt
M21 is the mutual inductance of coil 2 with respect to coil 1.
Through the exact same process on the following circuit:

47
We can conclude that:
di2
v1 = M12
dt
It will be later shown that:

M21 = M12 = M

... and we define M as the mutual inductance between two inductors. This is
the ability of one inductor to induce a voltage across a neighbouring inductor,
measured in Henrys (H). We say that these inductors a mutually coupled which
can only occur when the inductors are at close proximity, and that the sources
are time-varying.
The polarity of the voltage on a single inductor is done by the passive sign
convention. The polarity of the voltages across two magnetically coupled circuits
is dependent on the orientation and physical winding of each inductor. The dot
convention is used to show the polarities. If current enters the dotted terminal
of one coil, the reference polarity of the mutual voltage in the second coil is
positive at the dotted terminal, visually that is:

48
 Inductors in series can also be mutually coupled, and can then be replaced
by just a single equivalent inductor, as follows:

49
Series-Aiding Connection

Leq = L1 + L2 + 2M

Series-Opposing Connection

Leq = L1 + L2 − 2M

13.3: Energy in a Coupled Circuit

The mutual inductance cannot be greater than the geometric mean of the self-
inductances of the coils. The extent to which the mutual inductance approaches
the upper limit is called the coefficient of coupling k given by:
M
k=√
L1 L2
Since mutual inductance cannot be negative, 0 ≤ k ≤ 1.
If the entire flux produced by one coil links another coil, then k = 1 and we
have 100% coupling and they are said to be perfectly coupled.
In general the coupling coefficient k is a measure of the magnetic coupling
between two coils. This logically depends on the distance between the two coils,
their core, their orientation, and their windings.

50
13.4: Linear Transformers
A transformer is a circuit element which takes advantage of mutual inductance.
Generally it is a four terminal device comprising two or more magnetically
coupled coils. The following is a general schematic of a linear inductor:

The coil with the voltage source is said to be primary while the coil with
the load is said to be secondary. The resistances are included to account for the
losses in the coils. The transformer is said to be linear if the coils are wound
on magnetically linear material, meaning the magnetic permeability is constant.
Most materials are magnetically linear. In other words, flux is proportional to
the current in its windings.
The function of transformers is described in the next section, for now we
would like to be able to transform a linear transformer into one of two configu-
rations.
We want to transform the following initial circuit, and values:

The first configuration is called the T circuit:

51
Where:
La = L1 − M
Lb = L2 − M
Lc = M
... or if the dots are on opposite ends use −M
Q .
The second configuration is called the circuit:

Where:
L1 L2 − M 2
La =
L2 − M
L1 L2 − M 2
Lb =
L1 − M
L1 L2 − M 2
Lc =
M
... or if the dots are on opposite ends use −M .

13.5: Ideal Transformers

An ideal transformer is one which:
1. Coils have very large reactances (L1 , L2 , M → ∞).

52
 2. Coupling coefficient is equal to unity (k = 1), this implies
p
M = L1 L2

3. Primary and secondary coils are lossless (R1 = R2 = 0).

Given some ideal transformer in the following circuit (note the two vertical
lines which indicate it is ideal):

We define n to be the turns ratio of the coils, as in:

N2
n=
N1
... where Na is the number of coils in inductor a.
We can further derive:

N2 V2 I1
n= = =
N1 V1 I2

Notice current and voltage are inversely proportional to conserve energy. There
are two simple rules to decide the sign of n:
1. If V1 and V2 are both positive or both negative at the dotted terminals,
use +n. Otherwise use −n.
2. If I1 and I2 both enter into or both leave the dotted terminals, use +n.
Otherwise use −n.

ˆ If n = 1 we call it an isolation transformer.

ˆ If n > 1 we have a step-up transformer since voltage increases from pri-
mary to secondary loops. Therefore current is decreased in the second
loop.
ˆ If n < 1 we have a step-down transformer since voltage decreases from
primary to secondary loops. Therefore current is increased in the second
loop.

53
Importantly the equivalent impedance (Zeq ) felt by the source V̄ in the
circuit above is:
ZL
Zeq = 2
n
Finally, we look at how to transform the following ideal transformer circuit
(which contains sources on either side):

We can either transform the circuit into an equivalent circuit from the perspec-
tive of V̄s1 :

Or we can transform the circuit in to an equivalent circuit from the perspective

of V̄s2 :

54
13.6: Ideal Autotransformers
Autotransformers are made of a single continuous winding with a connection
point called a tap between the primary and secondary sides.

Step-down Autotransformer

V̄1 N1 + N2 N1
= =1+
V̄2 N1 N2

55
Step-up Autotransformer

V̄1 N1
=
V̄2 N1 + N2
The main difference between an transformer and an autotransformer is that
the inductors are both conductively and magnetically coupled. Electrical isola-
tion is not possible with an autotransformer.

Chapter 12: Three-Phase Circuits

Balanced Three-Phase Elements
So far we have studied single-phase circuits. These are circuits in which all the
sources are in phase. A polyphase circuit is one in which sources are out of
phase. A two-phase circuit has two sources which are out of phase from each
other by −90◦ . A three-phase circuit has three sources, all 120◦ out of phase of
each other. This type of circuit is very useful out of the polyphase circuits, and
so it is the focus of this chapter.
Three phase circuits are created by a three phase generator such as the
following:

56
A spinning magnet called the rotor induces an AC voltage with frequency ω on
each inductor a, b, c (a0 , b0 , c0 are their respective grounds). The voltages a, b, c
are out of phase by 120◦ due to their geometric position around the circle.
There are two possible implementations of this generator in a circuit.

57
 1. Y −connected source:

2. ∆−connected source:

The following is a general plot of the voltages in relation to each other:

58
 There are two possible sequences of peak voltages (also called the phase
sequences), either:

1. Positive Sequence:

a → b → c → a → b → c → a → ...

... in this case:

Van = Vp ∠0◦
Vbn = Vp ∠ − 120◦
Vcn = Vp ∠120◦ (= Vp ∠ − 240◦ )

2. Negative Sequence:

a → c → b → a → c → b → a → ...

... in this case:

Van = Vp ∠0◦
Vbn = Vp ∠120◦ (= Vp ∠ − 240◦ )
Vcn = Vp ∠ − 120◦

Loads can also be in three-phase form, and just like sources, the can be
either in Y or ∆ configuration.
1. Y −connected three-phase load.

59
 This type of load is said to be in balanced if:
Z1 = Z2 = Z3 = ZY

2. ∆−connected three-phase load.

This type of load is said to be in balanced if:

Za = Zb = Zc = Z∆

You can use the following equation to convert a balanced three-phase load
between forms:
Z∆ = 3ZY

60
 You can also convert between ∆ and Y sources as in the following image
and equations:

Vp
Van = √ ∠ − 30◦
3
Vp
Vbn = √ ∠ − 150◦
3
Vp
Vcn = √ ∠90◦
3
There are thus four possible combinations of three-phase sources and three-
phase loads (source then load):
1. Y − Y connection.
2. Y − ∆ connection.

3. ∆ − ∆ connection.
4. ∆ − Y connection.
The following four sections of this chapter look at each of these connections
in detail.
In the following sections we define the phase voltages/current as well at the
line voltages/currents for all the different configurations. The phase voltages
are the voltages between the lines a, b, c and n in the Y configuration. The line
voltages refer to the voltage differences between the three lines. Phase current
is the current in each phase of the source or load. Line current is the current
in each transmission line.
In general, in a balanced three-phase circuit, the neutral line can be ignored
since the voltage across it is 0 and the current through it is 0.

61
12.3: Balanced Y-Y Connection
The following is a general Y − Y three phase circuit:

The following equations relate to this circuit:

Phase Voltages
Van = Vp ∠0◦
Vbn = Vp ∠ − 120◦
Vcn = Vp ∠120◦
Line Voltages √
Vab = 3Vp ∠30◦
Vbc = Vab ∠ − 120◦
Vca = Vab ∠120◦
Phase Currents
Same as line currents.
Line Currents
Van
Ia =
ZY
Ib = Ia ∠ − 120◦
Ic = Ia ∠120◦

62
12.4: Balanced Y-D Connection
The following is a general Y − D three phase circuit:

The following equations relate to this circuit:

Phase Voltages
Van = Vp ∠0◦
Vbn = Vp ∠ − 120◦
Vcn = Vp ∠120◦
Line Voltages √
Vab = VAB = 3Vp ∠30◦
Vbc = VBC = Vab ∠ − 120◦
Vca = VCA = Vab ∠120◦
Phase Currents
VAB
IAB =
Z∆
VBC
IBC =
Z∆
VCA
ICA =
Z∆
Line Currents √
Ia = IAB 3∠ − 30◦

63
 Ib = Ia ∠ − 120◦
Ic = Ia ∠120◦
You can also solve these kinds of circuits by transforming the load into a Y
topology.

12.5: Balanced D-D Connection

The following is a general D − D three phase circuit:

The following equations relate to this circuit:

Phase Voltages
Vab = Vp ∠0◦
Vbc = Vp ∠ − 120◦
Vca = Vp ∠120◦
Line Voltages
Same as phase voltages.
Phase Currents
Vab
IAB =
Z∆
Vbc
IBC =
Z∆
Vca
ICA =
Z∆
Line Currents √
Ia = IAB 3∠ − 30◦
Ib = Ia ∠ − 120◦
Ic = Ia ∠120◦

64
12.6: Balanced D-Y Connection
The following is a general D − Y three phase circuit:

The following equations relate to this circuit:

Phase Voltages
Vab = Vp ∠0◦
Vbc = Vp ∠ − 120◦
Vca = Vp ∠120◦
Line Voltages
Same as phase voltages.
Phase Currents
Same as line currents.
Line Currents
Vp ∠ − 30◦
Ia = √
3ZY
Ib = Ia ∠ − 120◦
Ic = Ia ∠120◦

12.7 Power in a Balanced System

This section is essentially just a bunch of formulas relating to power in a three
phase circuit.
To begin we look at the total instantaneous power absorbed by the load
which is the sum of the instantaneous powers in the three phases:

p = 3Vp Ip cos(θ)

65
... where θ is the lag of the current behind corresponding phase voltages.
This is a time independent quantity, and it true regardless of the topology
of the load.
The following are other formulas:
1. Average power per phase:

Pp = Vp Ip cos(θ)

2. Total average power:

√
P = 3Vp Ip cos(θ) = 3VL IL cos(θ)

3. Reactive power per phase:

Qp = Vp Ip sin(θ)

4. Total reactive power:

√
Q = 3Vp Ip sin(θ) = 3Qp = 3VL IL sin(θ)

5. Apparent power per phase:

Sp = Vp Ip

6. Complex power per phase:

Sp = Pp + ̂Qp = Vp Ip∗

7. Total complex power:

3Vp2 √
S = 3Sp = 3Vp Ip∗ = 3Ip2 Zp = ∗
= P + ̂Q = 3VL IL ∠θ
Zp

... where Zp is the load impedance per phase.

Chapter 17: The Fourier Series

The Fourier Series is a method to decompose a periodic function into the sum
of a constant term and infinitely many sinusoidal terms. This is useful in our
study of circuits, because a periodic source can be then decomposed into a DC
source, and then infinitely many phasors. For applications we can just use the
first n phasors to get an approximation.
In general these are the steps to solve a a Fourier Series question:
1. Use Fourier series to decompose any periodic function into a DC compo-
nent and many AC components.

66
 2. Solve using AC and DC techniques separately by ”turning off” one source
at a time.
3. Use the superposition principle of a linear system to add all the results
back together.

All of the theory and methods to compute the Fourier Series of a function is in
my MTH312 notes. We do it only slightly differently here:
To compute the Fourier series f (t) of a function:
∞
X

f (t) = a0 + an cos(nω0 t) + bn sin(nω0 t)
n=1

... where:
2π
ω0 =
T
Z T
1
a0 = f (t)dt
T 0
Z T
2
an = f (t) cos(nω0 t)dt
T 0
Z T
2
bn = f (t) sin(nω0 t)dt
T 0
We will also show a new way to represent it.
Given some Fourier series of a function (neither even nor odd), there will be
a corresponding cos and sin term with the same frequency. We can thus rewrite
the Fourier series in amplitude-phase form as:
∞
X
f (t) = a0 + An cos(nω0 t + φn )
n=1

... where: p
An = a2n + b2n
 
−1 bn
φn = − tan
an
... which could also be thought of as a complex number:

An = an − ̂bn = An ∠φn

This allows us to combine the two terms into a single term.

The plots of the amplitudes and the phases of each term versus frequency is
called the frequency spectrum, and the following is an example of one:

67
Now that we know this we can visually represent each term in the amplitude-
phase form of the Fourier Series as an AC voltage source, as in:

68
17.5 Average Power and RMS Values
Given some periodic input in amplitude-phase form:
∞
X
v(t) = Vdc + Vn cos(nω0 t − θn )
n=1

∞
X
i(t) = Idc + Im cos(mω0 t − φm )
m=1

69
...the average power absorbed by the circuit (P ) is:
∞
1X
P = Vdc Idc + Vn In cos(θn − φn )
2 n=1

By the definition of RMS, a function in amplitude-phase from has an RMS value

of (Frms ): v
u ∞
u 1X 2
Frms = ta20 + (a + b2n )
2 n=1 n

The average power absorbed by a resistor with a periodic current input f (t) is:
2
P = RFrms

... or with a periodic voltage across it of f (t):

2
Frms
P =
R
... or if we just choose a 1Ω resistor (known as Parseval’s Theorem):
∞
2 1X 2
P1Ω = Frms = a20 + (a + b2n )
2 n=1 n

Chapter 19: Two-Port Networks

(Chapter not covered in F2021)

Conclusion
This concludes the content in this course. I hope these notes were helpful! Good
luck in the exam!
- Adam Szava

70