ĐẠI HỌC QUỐC GIA TP HỒ CHÍ MINH

TRƯỜNG ĐẠI HỌC BÁCH KHOA

🙞···☼···🙜

ANALOG SIGNAL PROCESSING (EE2405)

ĐỀ TÀI:
Passive Analog Filters
Giảng viên hướng dẫn: Hà Hoàng Kha
Lớp: TT02
Sinh viên thực hiện Mã số sinh viên
Phạm Chí Khoa 2251024
Nguyễn Đức Toàn 2251053
Phạm Viết Huy 1851071

Thành phố Hồ Chí Minh – 2023

 Table of Contents

I. Filters
RC filters
1. The Low-pass RC filter
II.
 Passive Analog Filters

In this research we will introduce filters and signal transmission

theory. Filters are essential components of most analog circuits and
are used to remove unwanted signals (i.e. noise) from the actual
signal. Transmission lines are essential for sending signals from one
device to another, such as from a detector to a data acquisition
module.

I. Filters
Filters are ubiquitous in analog electronic circuitry. In fact, if you
see a capacitor or an inductor in a circuit there is a good chance it
is part of a filter. Filters are frequently used to clean up (i.e. remove
high frequency noise) power supplies and remove spurious
frequencies from a signal (frequently 60 Hz, switching power supply
noise in computers, display screen noise, ground loop noise, and
Radio-Frequency (RF) pick-up).

RC Filters:
RC filters are by far the most common filters around. They are
simple to make (i.e. just a resistor and a capacitor), reliable, and
involve relatively simple design calculations.

1. The Low-Pass RC Filter

The low-pass RC filter, or integrator, is
used to remove high frequencies from a
signal. Applications include the removal of
RF pick-up noise and reducing ripple
voltages on power supplies.
 co
m
pl
ex
im
pe
da
nc
es
.
W
e
re
ca
ll
th
e
A generic RC low pass filter circuit is re
shown in figure above. We have already su
calculated its performance in the previous lts
chapter using Fourier analysis and
(equation
28):
i(ω t+ ϕ − π / 2)
V out =V C =V o sin( ϕ)e

Whe sin() 2
1  (RC) V  V eit . From these quantities we can compute
re 1 and
in
the gain and phase performance of the filter. The Gain |V out|
and
gain is defined as
the phase as      / 2 (this is just the in the exponent).
part after the t
The RC filter is just a voltage divider with complex impedances, so
we can calculate the gain easily:

| ||
V out
V¿
=
1/i ω C
=
1
=
1
||
R+1/i ω C 1+i ω RC √ 1+(ω RC )2 |
=sin (ϕ)

The phase of the output voltage with respect to the input is easily
computed and is given by
tan   tan(   / 2)  cot( )  RC (4.3)
At   1/ RC , the output voltage drops to 1/ 2 of the input voltage, and
consequently
the power transmission drops to 50% or -3dB1. At this frequency, the voltage
across the
resistor and the voltage across the capacitor are equal in   out of
amplitude, but 2
phase with the drive voltage. The average value of V across either
2

the resistor or the capacitor is down by a factor of 2 from the drive

voltage. Consequently, this frequency characterizes the RC circuit
completely and is called the 3dB frequency. We can rewrite the gain
and phase equations in terms of the 3dB frequency:

1
gain= tan φ=f / f 3 dB
√1+(f /f 3 dB )2
wit f3dB  1/ 2RC .
h
The Bode plots (log-log or semi-log plots) for the gain and phase of
the low pass filter are shown in figure 4.2, below.
Figure 4.2: Bode plots of the relative magnitude (left) and phase
(right) of the output of an RC low-pass filter.

1
A dB or decibel is a notation for quantifying a ratio of two numbers. For
power, a dB is defined as dB = 10log10(P/P0)
From this definition we can see that a ratio of 0.5 is roughly -3. Hence -3db is the same as halving a
signal.
For voltage or current, a dB is
defined as dB =
20log10(V/V0)
So at -3dB, the voltage (or current) has
dropped to its 1/√ 2 of its input value.

Past the -3 dB point, the log-log “slope” for the gain is -20 dB/Decade
or -6 dB/Octave. The low-pass RC filter is also called an integrator
1
because it integrates currents with 
frequencies 3d . In other Vout  I (see chapter 3, equation 8) for
above f B
words (t)dt currents with frequency components
C above
f3dB .

An ideal low-pass filter’s transfer function is shown. The frequency between

the pass- and-stop bands is called the cut-off frequency (ωc). All of the
signals with frequencies below ωc are transmitted and all other signals are
stopped.
In practical filters, pass and stop bands are not clearly defined, |H(jω)|
varies continuously from its maximum toward zero. The cut-off frequency is,
therefore, defined
as the frequency at which |H(jω)| is reduced to 1/√ 2 = 0.7 of its maximum
value. This corresponds to signal power being reduced by 1/2 as P ∝ V 2.

Low-pass RL filters
A series RL circuit as shown acts as a low-pass filter. For no load resistance
(“open-loop” transfer function), Vo can be found from the voltage divider
formula:
R Vo R 1
V o= V i −> H ( j ω)= = =
R+ j ω L V i R+ j ω L 1+ j(ω L/ R)
We note
1
|H ( j ω)|=
√ 1+(ω L/ R)2

It is clear that |H(jω)| is maximum when denominator is smallest, i.e., ω → 0 and

|H(jω)| decreases as ω is increased. Therefore, this circuit allows “low-
frequency” signals to pass through while “blocking” high-frequency signals (i.e.,
reduces the amplitude of the voltage
of the high-frequency signals). The reference to define the “low” and “high”-
frequencies is the cut-off frequency: “low”-frequencies mean frequencies
much lower than ωc.
To find the cut-off frequency, we note that the |H(jω)|Max = 1 occurs at ω = 0
(alterna- tively find d |H(jω)| /dω and set it equal to zero to find ω = 0 which
maximizes |H(jω)|). Therefore,

|H(jω)|max = 1
1 1
|H ( jω)|ω=ω = ,|H ( jω)|max =
c
√2 √2
2
1 1 ω c L ωc L
= −>1+( ) =2 −> =1
√ 1+(ω c L/ R)
2
√ 2 R R
Therefore,
R
ωc = 1
and H(jω)
L 1 + jω/ωc
=

Input Impedance: Using the definition of the input impedance, we have:

Vi
Zi = = jωL+ R
Ii

The value of the input impedance depends on the frequency ω. For good
voltage coupling, we need to ensure that the input impedance of this filter
is much larger than the output impedance of the previous stage. Since we
do not know the frequency of the input signal, we need to ensure that
good voltage coupling criteria is satisfied for all frequencies (or all
possible values of Zi). As such, the minimum value of Zi is an important
number. Zi is
minimum when the impedance of the inductor is zero (ω → 0).

Zi|min = R
L
Output Impedance: The output impdenace can be
found by “killing” the source and finding the equivalent
impdenace between output terminals:
Zo = jωL ǁ R
where the source resistance is ignored. Again, the value of the output
impedance depends on the frequency ω. For good voltage coupling,
we need to ensure that the output impedance of this filter is much
smaller than the input impedance of the next stage for all
frequencies,
the maximum value of Zo is an important number. Zo is maximum
when the impedance of the inductor is infinity (ω → ∞).

Zo|max = R

Bode Plots and Decibel

The voltage transfer function of a two-port network (and/or the ratio of

output to input powers) is usually expressed in Bel:
P0
| |
Number of Bels=log 10 ( )∨
Pi
V0
Vi B | |
=2 log 10
V0
Vi

because P ∝ V 2. Bel is a large unit and decibel (dB) is usually used:

| || |
V0 V0
Number of decibels=20 log 10. . ∨
Vi ||
V i dB
V log V 0
=20
. i .dB 10
Vi
V
. i.
There are several reasons why decibel notation is used:

1) Historically, the analog systems were developed first for audio

equipment. Human ear “hears” the sound in a logarithmic fashion. A
sound which appears to be twice as loud actually has 10 times
power, etc. Decibel translates the output signal to what ear hears.

2) If several two-port network are placed in a cascade (output of one

is attached to the input of the next), the overall transfer function, H,
is equal to the product of all transfer functions:

|H(jω)| = |H1(jω)| × |H2(jω)| × ...

20 log10 |H(jω)| = 20 log10 |H1(jω)| + 20 log10 |H2(jω)| + ...
|H(jω)|dB = |H1(jω)|dB + |H2(jω)|dB + ...

making it easier to find the overall response of the system.

3) Plot of |H(jω)|dB versus frequency has special properties that make
analysis simpler. For example, the plot asymptotes to straight lines at
low and high frequencies as is shown below.
Also, using dB definition, we see that, there is a 3 dB difference
between maximum gain and gain at the cut-off frequency:

[ ]
Σ
|H ( j ω c )|
20 log |H ( j ωc )|− 20 log| H ( j ω)|=20 log
| H ( j ω)|max
=20 log
( )
1
√2
≈ −3 dB

Bode plots are plots of |H(jω)|dB (magnitude) and ƒ H(jω) (phase) versus
frequency in a semi-log format (i.e., ω axis is a log axis). Bode plots
of first-order low-pass RL filters are shown below (W denotes ωc).

|H(jω)|dB

ƒ H(jω)

At high frequencies, ω/ωc ≫ 1,

|H ( j ω)| ≈
1
ω /ωc
−>|H ( j ω)|dB=20 log
[ ]
1
ω /ω c
=20 log (ω c )− 20 log (ω)

which is a straight line with a slope of -20 dB/decade in the Bode plot.
It means that if ω
is increased by a factor of 10 (a decade), |H(jω)|dB changes by -20 dB.
At low frequencies, ω/ωc 1, |H(jω)| ≈ 1 which is also a straight line in
the Bode plot. The intersection of these two “asymptotic” values is at
1 = 1/(ω/ωc) or ω = ωc. Because of this, the cut-off frequency is also
called the “corner” frequency.
The behavior of the phase of H(jω) can be found by examining ƒ H(jω)
= − tan−1(ω/ωc). At low frequencies, ω/ωc 1, ∠ H(jω) ≈ 0 and at high
frequencies, ω/ωc 1, ∠ H(jω) ≈ −90◦. At cut-off frequency, ∠ H(jω) ≈
−45◦.
General first-order low-pass filters

As we discussed before, transfer functions characterize a two-port

network. As such, it is useful to group two-port networks into families
based on their voltage transfer functions. To facilitate this grouping,
the convention is to simplify the voltage transfer function to a form
such that the “Real” part of the denominator of H(jω) is unity (i.e.,
the denominator
should be 1 + j · · · or 1 − j · · · ). As we will see later in this section, this
grouping will also
help reduce the math that we do in analyzing various circuits.
The low-pass RL filter discussed before is part of the family of first-
order low-pass filters (first order means that ω appears in the
denominator with an exponent of 1 or −1. In general, the voltage
transfer function of a first-order low-pass filter is in the form:

K
H ( jω )=
1+ j ω/ω c

The maximum value of |H(jω)| = |K| is called the filter gain. Note that the
exponent of ω
in the denominator is +1 so that |H(jω)| decreases with frequency
(thus,a low-pass filter):

|K|
|H ( j ω)|=
√ 1+(ω /ω ) c
2

| K| ω
∠ H ( jω )= tan−1 ( )
K ωc

For RL filter, K = 1, and ωc = R/L. Note that K can be negative, and in

that case, the “minus” sign adds 180 ◦ phase shift to the transfer
function as is denoted by |K|/K factor above.
 Low-pass RC filters
A series RC circuit as shown also acts as a
low-pass filter. For no load resistance
(“open-loop” transfer function), Vo can be
found from the voltage divider formula:

1/( j ω C) 1
V 0= V= V
R+1/( j ω C) i 1+ j(ω RC ) i

1
H ( jω )=
1+ j ω RC
We see that the voltage transfer function of this circuit is similar to
transfer function of a general first-order low-pass filter. So, this is a
low-pass filter with K = 1 and ωc = 1/RC. (Note: we identified the
circuit and found the cut-off frequency without doing any math!).
 We could, of course, do the math following the procedure in analyzing
the low-pass RL filter to get the same answer. (Exercise: Show this.).

Following the same procedure as for RL filters, we find input and output
Impedances
1
Zi =R+ ∧Z i ¿ min=R
j ωC
1
Z 0=R∨¿ ∧Z 0 ¿max =R
j ωC

Terminated RL and RC low-pass filters

Now let us examin the effect of a load

on the perfor- mance of our RL and RC
filters. For this example, a resistive
load is considered but the analysis can
be easily extended to an impedance
load. For example, consider the
terminated RC filter shown:

From the circuit,

'
Vo 1/( j ω C)∨¿ R L R /R
H ( jω )= = =
V i R+ [ 1 /( j ω C)∨¿ R L ] 1+ j(ω R' C)

'
with R =R∨¿ R L
 This is similar to the transfer function for unterminated RC filter but
with resistance R
being replaced by RJ . Therefore,
'
1 1 R /R
ωc= ' = ∧H ( jω )=
R C (R∨¿ R L )C 1+ j ω/ω c
1
Z 0=R∨¿ Z ¿ =R
j ω C 0 max

ωc =

We see that the impact of the load is to reduce the filter gain (K =
RJ /R < 1) and to shift the cut-off frequency to a higher frequency as RJ
= R ǁ RL < R.

1
Input Impedance: Zi =R+ j ω C ∨¿ R L Zi ¿min =R

1
Output Impedance: Z 0=R∨¿ Z ¿ =R
j ω C 0 max

We could have arrived at the same results using the the relationship
between open-loop,
Ho(jω), and terminated, H(jω), transfer functions of a two-port
network:

ZL RL 1
H(jω) = Ho(jω) 1 × 1 + jωRC
ZL + = RL + R
Zo jωC
ǁ
(Exercise: show this.) Also, note that the output impdenace of the
terminated circuit is exactly the same as the open-loop version.
Furthermore, it can be seen that as long as RL Zo or RL Zo |max =
R (our condition for good voltage coupling), RJ ≈ R and the
terminated RC filter will look exactly like an unterminated filter
– The filter gain is one, the shift in cut-off frequency disappears,
and
input and output resistances become the same as before.
Terminated RL low-pass filters

The parameters of the terminated RL filters can be found similarly:

Vo 1
Voltage Transfer Function: = , ωc = (R ǁ RL)/L.
H(jω) = Vi 1 +
jω/ωc
Input Impedance: Zi = jωL + R ǁ RL, Zi|min = R ǁ RL
Output Impedance: Zo = (jωL) ǁ R, Zo|max = R
Here, the impact of load is to shift the cut-off frequency to a lower
value. Filter gain is not affected. Again for RL Zo or RL Zo|max = R (our
condition for good voltage coupling), the shift in cut-off frequency
disappears and the filter will look exactly like an unterminated
filter.
Exercise: Derive above equations for the transfer function and input and
output impde- nacess.
 2.2 First-order high pass filters

In general, the voltage transfer function of a first-order high-pass filter is

in the form:

H(jω) K
= 1−
jωc/ω

It is a first-order filter because ω appears in the denominator with an

exponent of −1. It is a high-pass filter because |H| = 0 for ω = 0 and
|H| is constant for high-freqnecies. Paramter ωc is the cut-off freqnecy √ of
the filter (Exercise: prove that |H(jωc)| is 1/√ 2 = 0.7 of |H(jω)|Max.)
The maximum value of |H(jω)| = |K| is called the filter gain.

|K| +|K| −1 ω
|H ( j ω)|= ∠ H ( j ω)= tan ( )
√ 1+(ω /ω )c
2 K ωc

Bode Plots of first-order high-pass filters (K = 1) are shown below. The

asymptotic behavior of this class of filters is:
At low frequencies, ω/ωc 1, |H(jω)| ∝ ω (a +20dB/decade line) and ƒ H(jω)
= 90◦
At high frequencies, ω/ωc 1, |H(jω)| ∝ 1 (a line with a slope of 0) and ƒ
H(jω) = 0◦

|H(jω)|

H(jω) ƒ

High-pass RC filters freq have allow higher frequencies to

The High Pass filter uen alread pass through them, and so
allows higher frequencies to cies y seen, we can use one of these to
pass through, whilst . As capacit implement a high-pass
attenuating (blocking) lower we ors filter.
 A generic RC high-pass
filter circuit is shown in
figure. Mathematically,
high-pass filter can be
treated the same way a their
low pass cousins: they are
almost identical except that
Vout measures the voltage
drop across the resistor
instead of the capacitor.
Qualitatively, the capacitor
blocks DC and low
frequency signals. If we
treat the
high-
pass
filter as a
complex
impedan
ce
voltage
divider,
then we
obtain
immediat
ely

| ||
V out
V¿
=
R
||=
iω RC
=| ω RC
R+1/i ω C 1+i ω RC √ 1+(ω RC )2
=cos (ϕ )
tan   1/ RC
We can rewrite these using only
(f /f 3 dB )
gain= , tan φ=f 3 dB /f
√1+(f /f 3 dB )2

A series RC circuit as shown acts as a

high-pass filter. The open-loop voltage
transfer function of this filter is:

V0 R 1
H ( jω )= = =
V i R+1/( j ω C) 1− j(1/ω RC )
Therefore, this is a first-order high-pass filter with K = 1 and ωC =
1/RC. Input and output impdenaces of this filter can be found similar
to the procedure used for low-pass filters:
1
Input Impedance: Zi = R + and Zi|min = R
jωC
1
Output Impedance: Zo = R ǁ and Zo|max = R
jωC

High-pass RL filters

A series RL circuit as shown also acts as a high-pass filter.

Again, we find the open-loop tranfunction to be:

R
ωc =
L H(jω) 1
= 1 − jωc/ω

Input Impedance: Zi = R + and Zi|min = R

jωL
Output Impedance: Zo = R ǁ and Zo|max = R
jωL
2.3 Band-pass filters

A band pass filter allows signals with a range of frequencies (pass

band) to pass through and attenuates signals with frequencies outside
this range.

ωl : Lower cut-off
frequency;
ωu : Upper cut-off
frequency;
ω0 ≡√ ωi ω u :Center frequency;
B ≡ ωu − ωl : Band width;
ω0
Q ≡ :Quality factor
B

As with practical low- and high-pass filters, upper and lower cut-off
frequencies of practical
band pass filter are defined as the frequencies at which the magnitude of
the voltage transfer function is reduced by 1/√ 2 (or -3 dB) from its
maximum value.
Second-order band-pass filters:

Second-order band pass filters include two storage elements (two

capacitors, two inductors, or one of each). The transfer function for a
second-order band-pass filter can be written as
 K
H ( jω )=
ω ω0
1+ jQ( − )
ω0 ω

| K|
|H ( j ω)|=

√
2
2 ω ω0
1+ Q ( − )
ω0 ω

∠ H ( jω)=−
| K|
K
tan
−1
[(
Q
ω ω0
−
ω0 ω )]

The maximum value of |H(jω)| = |K| is called the filter gain. The lower
√
and upper cut-off
frequencies can be calculated by noting that |H(jω)|max = K, setting |
H(jωc)| = K/√ 2 and
solving for ωc. This procedure will give two roots: ωl and ωu.

1 K K
|H ( j ω)|= |H ( jω )|max= =

√
√2 √2 2 ω c ω0 2
1+Q ( − )
ω0 ωc
2
2 ω c ω0 ω c ω0
Q ( − ) =1−>Q( − )=±1
ω0 ωc ω0 ωc
 ωc ω0
=0The above equation is really two
2 2
ω c − ω0 ±
Q
quadratic equations (one with + sign in front of
fraction and one with a − sign). Solving these equation we
will get 4 roots (two roots per equation). Two of these four roots will
be negative which are not physical as ωc > 0. The other two roots are
the lower and upper cut-off frequencies (ωl and ωu, respectively):

√ √
1 ω 1 ω
ω l=ω 0 1+ 2
− 0 ω u=ω 0 1+ 2
+ 0
4Q 2Q 4 Q 2Q

Bode plots of a second-order filter is shown below. Note that as Q

increases, the bandwidth of the filter become smaller and the |H(jω)|
becomes more picked around ω0.

|H(jω)|db

∠H(jω)

Asymptotic behavior:
At low frequencies, ω/ω0 ≪ 1, |H(jω)| ∝ ω (a +20dB/decade line), and
∠H(jω) → 90◦ At high frequencies, ω/ω0 ≫ 1, |H(jω)| ∝ 1/ω (a
-20dB/decade line), and∠H(jω) → −90◦ At ω = ω0, H(jω) = K (purely
real) |H(jω)| = K (maximum filter gain), and ∠ H(jω) = 0◦.
There are two ways to solve second-order filter circuits. 1) One can
try to write H(jω) in the general form of a second-order filters and find
Q and ω0. Then, use the formulas above to find the lower and upper
cut-off frequencies. 2) Alternatively, one can directly find the
upper and lower cut-off frequencies and use ω0 ≡ √ ωl ω uto find the
center frequency and
B ≡ ωu − ωl to find the bandwidth, and Q ≡= ω0/B to find the quality
factor. The two
examples below show the two methods. Note that one can always find
ω0 and k rapidaly as
H(jω0) is purely real and |H(jω0)| = k

Series RLC
Band-pass
filters

Using
voltage
divider
formula, we
have

V0 R
H ( jω )= =
V i R+ jω L+1 /( j ω C)
R
H ( jω)=
1
R+ j(ω L − )
ωC

There are two approaches to find filter parameters, K, ω 0 ,ω u ,∧ω l

Method 1: We transform the transfer function in a form

similar to general form of the transfer function for second
order bandpass filters:
 K
H ( jω )=
ω ω0
1+ jQ( − )
ω0 ω

Note that the denominator of the general form is in the form 1 + j . . .

Therefore, we divide top and bottom of transfer function of series
RLC bandpass filters by R:
1
H ( jω )=
ωL 1
1+ j( − )
R ω RC

Comparing the above with the general form of the transfer function,
we find K = 1. To find Q and ω0, we note that the imaginary part of
the denominator has two terms, one positive and one negative (or
one that scales as ω and the other that scales as 1/ω) similar to the
general form of transfer function of 2nd-order band-pass filters
(which includes Qω/ω0 and
−Qω0/ω). Equating these similar terms we get:

Qω ωL Q L
= → =
ω0 R ω0 R
Qω0 1 1
= → Qω0 =
ω ωRC RC

We can solve these two equations to find:

√
1 ω L
ω 0= Q= 0 = 2
√ LC R /L R C
Method 2: In this method, we directly calculate the filter parameters
similar to the proce- dure followed for general form of transfer
function in page 30. Some simplifications can be made by noting: 1) At
ω = ω0, H(jω) is purely real and 2) K = H(jω = jω0).

Starting with the transfer function for the series RLC filter:
R
H ( jω )=
1
R+ j(ω L − )
ωC
We note that the transfer function is real if coefficient of j in the
denominator is exactly zero (note that this happens for ω = ω0), i.e.,
1 1
ω0 L − =0 −> ω0=
ω0 C √ LC

Also

R
K=H ( j ω= j ω0 )= =1
R
The cut-off frequencies can then be found by setting:
K 1
|H ( j ωc )|= =
√2 √ 2
2
ωc L 1
1+( − ) =2
R ωc RC

Which can be solved to find ω u∧ω l

Input and Output Impedance of Band-pass RLC filters

1 1
Zi = j ω L+ + R= j(ω L − )+ R
j ωC ωC
Zi ¿min =R occurs at ω=ω0

1
Z 0=( j ωL+ )∨¿ R −> Z 0 ¿max =R
j ωC
Wide-Band Band-Pass Filters

Band-pass filters can be constructed by putting a high-pass and a

low-pass filter back to back as shown below. The high-pass filter sets
the lower cut-off frequency and the low-pass filter sets the upper cut-
off frequency of such a band-pass filter.

| H (j) | | H (j) | 
 

c2  c1  

An example of such a band-pass

filter is two RC low-pass and
high-pass filters put back to
back. These filters are widely
used (when appropriate, see
below) instead of an RLC filter
as inductors are usually bulky
and take too much space on a
circuit-board.

In order to have good voltage coupling in the above circuit, the input
impedance of the high-pass filter (actually Zi|min = R1) should be much
larger than the output impedance of the low-pass filter (actually Zo|
max = R2), or we should have R1 R2. In that case we can use un-
terminated transfer functions:
1
H(jω) = H1(jω) × H2(jω) = jω/ω
1+ c2
 ωc1 = 1/(R1C1) ωc2 = 1/(R2C2) 1
×
1 − jωc1/ω

1 1
H(jω) =
(1 + jω/ωc2)(1 − (1 + ωc1/ωc2) + j(ω/ωc2 −
=
jωc1/ω) ωc1/ω)

Again, we can find the filter parameters by either of two methods

above. Transforming the transfer function to a form similar to the
general form (left for students) gives:

K=
1
Q=
√ω c1 /ω c 2 ω 0= √ ω c1 ω c2
1+ω c1 /ω c 2 1+ ωc 1 /ω c2
One should note that the Bode plots of previous page are
“asymptotic” plots. The real H(jω) differs from these asymptotic
plots, for example, |H(jω)| is 3 dB lower at the cut- off frequency. A
comparison of “asymptotic” Bode plots for first-order high-pass
filters are
given in page 28. It can be seen that |H1(jω)| achieves its maximum
value (1 in this case) only when ω/ωc1 < 1/3. Similarly for the low pass
filter, |H2(jω)| achieves its maximum value (1 in this case) only when
ω/ωc2 > 3. In the band-pass filter above, if ωc2 ωc1 (i.e., ωc2 ≥ 10ωc1), the
center frequency of the filter will be at least a factor of three away from
both cut-off frequencies and |H(jω)| = |H1| × |H2| achieves its maximum
value of 1. If ωc2 is not ωc1 (i.e., ωc2 < 10ωc1), H1 and H2 will not reach
their maximum of 1 and the filter
|H(jω)|max = |H1| × |H2| will be less than one. This can be seen by
examining the equation of K above which is always less than 1 and
approaches 1 when ωc2 ωc1.

More importantly, we can never make a “narrow” band filter by putting

two first-order high- pass and low-pass filters back to back. When ωc2 is
not ωc1, |H(jω)|max becomes smaller than 1. Since the cut-off frequencies
are located 3 dB below the maximum values, the cut-off frequencies will
not be ωc1 and ωc2 (those frequencies are 3 dB lower than |H(jω)|max = 1).
The lower cut-off frequency moves to a value lower than ωc1 and the
upper cut-off frequency moves to a value higher than ωc2. This can be
seen by examining the quality factor of this filter at the limit of ωc2 =
ωc1

Q=
√ ωc 1 /ωc 2 = 1
=0.5
1+ω c1 /ω c 2 1+1

while our asymptotic description of previous page indicated that when

ωc2 = ωc1, band-width becomes vanishingly small and Q should become
very large.
Because these filters work only when ωc2 ωc1, they are called “wide-
band” filters. For these wide-band filters (ωc1 ωc2), we find from
above:

K =1 .

Q= √ ω c 1 / ωc 2 ω 0=√ ω c2 ω c1
1
H(jω) 1 + j(ω/ω − ω /ω)
c2 c1
=

We then substitute for Q and ω0 in the expressions for cut-off frequencies

 (page 31) to get:

√ 1 ω ω
ω u=ω 0 1+ 2
+ 0 = 0 ( √ 1+4 Q 2+1)
4 Q 2 Q 2Q

√ 1 ω ω
ω l=ω 0 1+ 2
+ 0 = 0 ( √ 1+4 Q 2 −1)
4 Q 2 Q 2Q
Ignoring 4Q2 term compared to 1 (because Q is small),we get:

ω0 √ω c 2 ω c1
ω u= = =ω c2
Q √ ωc 1 /ω c2

For ωl, if we ignore 4Q2 term compared to 1, we will find ωl = 0. We

should, therefore, expand the square root by Taylor series expansion
to get the first order term:

ω0 1 2 ω0 2
ωu ≈ (1+ 4 Q − 1)= × 2Q =ω0 Q=ω c 2
2Q 2 2Q
What are Wide-Band and Narrow-Band Filters? Typically, a wide-band filter
is defined as a filter with ωc2 ωc1 (or ωc2 ≥ 10ωc1). In this case, Q ≤ 0.35
. A narrow-band filter is usually defined as a filter with B ω0 (or B ≤
0.1ω0). In this case, Q ≥ 10.

Example: Design a band-pass filter with cut-off frequencies of 160 Hz and

8 kHz. The load for this circuit is 1 MΩ.
As this is wide-band, band-pass filter (ωu/ωl = fu/fl = 50≫ 1), we use
two low- and high-pass RC filter stages similar to circuit above. The
prototype of the circuit is shown below:

The high-pass filter sets the

lower cut-off frequency, and the
1 MΩ load sets the out- put
impedance of this stage. Thus:
Z
o|

ma
x
=

R
1
≪
1
M
Ω

R
1
≤

1
0
0
k
Ω
 1
ωc(High-pass) = ωl = = 2π × 160 → R1C1 = 1 × 10−3kΩ
R1C1
One should choose R1 as close as possible to 100 kΩ (to make the C1
small) and R1C1 = 1×10−3 using commercial values of resistors and
capacitors. A good set here are R1 = 100 kΩ and C1 = 10 nF.
The low-pass filter sets the upper cut-off frequency. The load for this
component is the input resistance of the high-pass filter, Zi|min = R1 =
100 kΩ. Thus:

Zo|max = R2
100kΩ → R2 ≤ 10 kΩ
1
ωc(Low-pass) = ωu = = 2π × 8 × 103 → R2C2 = 2 × 10−5
R2C2
As before, one should choose R2 as close as possible to
10 kΩ and R2C2 = 2 × 10−5 using commercial values of
resistors and capacitors. A good set here are R2 = 10 kΩ
and C2 = 2 nF.

In principle, we can switch the position of low-pass

and high-pass filter stages in a wide- band, band-pass
filter. However, the low-pass filter is usually placed
before the high-pass filter because the value of
capacitors in such an arrangement will be smaller.
(Try redesigning the above circuit with low-pass and
high-pass filter stages switched to see that one
capacitor become much smaller and one much
larger.)

3) Applications
Here are some specific applications of RC filters:
Blocking Capacitor
This is a high pass filter that is used to eliminate DC. Suppose
that you want to measure small time-dependent signals that happen
to “float” on a high voltage. If you use a blocking capacitor, then
the high voltage DC will not get through to your detection
electronics, but the signal will get f3d low to insure that your entire
through. Choose signal gets through. B

Ripple Eliminator
This is a low pass filter used to build power supplies. Since
most of our power is 60 Hz AC, our DC power supplies will
convert AC to DC, but there will always be some
residual 60 Hz “ripple”. A low pass f3d set well below 60 Hz will work. You
filter with B

do not use a resistance in this case, but let the combination of the
loading resistance RL and the Thevenin resistance of the previous
components serve as your R. This usually requires a large capacitor
since RL might be quite small when you use the power supply.
If the capacitor is not big f3d will then shift to a higher frequency, and the
enough, then 60 Hz ripple B

will reappear.

Chip Supply Clean Up

Frequently the voltage which you supply to a chip component,

such as an op-amp (we will study these later in the semester), may
be “clean” when it comes out of the power supply, but will pick up
noise by the time it reaches the component. In this case a 10 – 100
nF capacitor placed at the supply leads of the component will
remove the high frequency pick-up noise.

Noise Eliminator
Any signal line is susceptible to picking up high frequency
transients; especially if there are motors or switching power
supplies (or FM radio stations!) nearby. A noise eliminator is a low
pass filter with a high value of f3dB .

Integrator
If you build an RC filter, but set the f3d much higher than the highest
value of B
frequency in your signal, the filter integrates your signal. From our earlier
analysis,
whe  f3dB , we can see that each (low) frequency voltage component will see a 
nf 2
same phase shift and its amplitude will be proportional to 1 f . This is exactly the
prescription for integration. So, a high pass filter sends high
frequencies out on the resistor, and the integral of very low
frequencies on the capacitor.
Differentiator
If you build an RC f3dB lower than the lowest frequency in your signal, the
filter with
filter differentiates your signal. From our earlier analysis, when f  f3dB ,
each (high)
frequency voltage component
will see a /2 same phase shift
and its amplitude will be
proportional to f. This is exactly
the prescription for
differentiation. So, a low pass
filter sends low frequencies out
on the capacitor, and the
derivative of high frequencies on
the resistor.
proportional to f. This is exactly the prescription for
differentiation. So, a low pass filter sends low frequencies out on
the capacitor, and the derivative of high frequencies on the
resistor.

Exercise Problems

Problem 1. Design a RLC bandpass filter with a lower cut-off frequency

of 1 kHz and a bandwidth of 3 kHz. What is the center frequency and
Q of this filter?
Problem 2. We have an amplifier that amplifies a 1 kHz signal from a
detector. The load for this amplifier can be modeled as a 50 kΩ
resistor. The amplifier output has a large amount of 60 Hz noise. We
need to reduce the amplitude of noise by a factor of 10. Design a first-
order passive filter which can be placed between the amplifier and the
load and does the job. Would this filter affect the 1 kHz signal that we
are interested in? If so, by how much?
Problem 3. The tuner for an FM radio requires a band-pass filter with a
central frequency of 100 MHz (frequency of a FM station) and a
bandwidth of 2 MHz. a) Design such a filter.
b) What are its cut-off frequencies?
Problem 4. A telephone line carries both voice band (0-4 kHz) and data
band (25 kHz to 1 MHz). Design a filter that lets the voice band
through and rejects the data band. The filter must meet the following
specifications: a) For the voice band, the change in transfer function
should be at most 1 dB; and b) The transfer function should be as
small as possible at 25 kHz, the low end of the data band.
Solution to Exercise Problems

Problem 1. Design a RLC bandpass filter with a lower cut-off frequency

of 1 kHz and a bandwidth of 3 kHz. What is the center frequency and
Q of this filter?
 The circuit prototype is:
For a 2nd order band-pass filter:

B(Hz) = fu − fl fu = 1 + 3 = 4 kHz
B(rad/s) = 2πB(Hz) = 1.88 4× 10
ωu = 2πfu = 2.51 ×4 10 ωl = 2πfl = 6.28 ×3 10
ω 0=√ ωu ω l=1.26 ×10
4

ω0 1.26 ×10
4
B(Rad /s)= −>Q= 4
=0.67
Q 1.88 ×10

For the series RLC circuit:

1 1
ω 0= −> LC= 2 1
√ LC 1 ω0
C= = −3
= 0.63 µF
10 × 10 × (1.26 × 10 4)2
Lω2 0
ω0 R ω0
Q= → = =
R/L B(rad/s) L
Q
R = LB = 10 × 10−3 × 1.88 × 104 = 188 Ω

Therefore, using commercial values, the design values are L = 10 mH,

R = 180 Ω, and
C = 0.68 µF.
Problem 2. We have an amplifier that amplifies a 1 kHz signal from a
detector. The load for this amplifier can be modeled as a 50 kΩ
resistor. The amplifier output has a large amount of 60 Hz noise. We
need to reduce the amplitude of noise by a factor of 10. Design a first-
order passive filter which can be placed between the amplifier and the
load and does the job. Would this filter affect the 1 kHz signal that we
are interested in? If so, by how much?

We want to have 1 kHz signals to go through but reduce 60 Hz

signals, so we need a high-pass filter. The prototype of the circuit is
shown below. For this circuit:

Vo 1
H(jω) =
= Vi 1 − jωc/ω

1
ωc =
RC
Zi|min = R Zo|
max = R

As the output impedance of the inverting amplifier circuit is “zero”,

we do not need to worry about the input impedance of our filter. The
output impedance of the filter is restricted by

Zo|max = R ≪ 50 kΩ → R ≤ 5 kΩ

This filter should reduce the amplitude of 60 Hz (ω60 = 2π × 60 = 120π

rad/s) signal by a factor of 10, i.e.,
 |H ( j ω= j ω60)|=|| V0
Vi 60 HZ
=
1
√1+(ω /ω
c 60
2
)
=0.1

1 + (ωc/ω60)2 = 100 1 → = ωc ≈ 10ω60 =

3751 rad/s → RC RC = 2.67 × 10
−4

Reasonable choices are R = 3.9 kΩ (to keep it below 5 kΩ) and C = 68

nF (fc ≈ 600 Hz). The impact on 1 kHz signal (ω1000 = 2000π rad/s) can
be found from:

1
1 = 0.86
2
|H(jω = jω1000)| = . = .
1+ 2 1+
(ωc/ω1000) (3751/6283)
So the amplitude of 1 kHz signal is reduced by 14% (or by -1.3 dB).
Problem 3. The tuner for an FM radio requires a band-pass filter with a
central frequency of 100 MHz (frequency of a FM station) and a
bandwidth of 2 MHz. a) Design such a filter.
b) What are its cut-off frequencies?

Because this is not a wide-band filter, the

simplest filter will be an RLC filter
as is shown. For this filter:

1 6
ω 0= =2 π 100 ×10
√ RC

√
ω L 2 π 100 ×106
Q= 0 = 2 = =50
B R C 2 π ×2 ×10 6
 Using a L = 1 µH inductor:

1 1
= 4π21016 = 4π21016 × 10−6 → C = 2.5 × 10−12 F
LC → C

Choose: C = 2.2 pF

1 1
H ( jω )= =
1+ j ω/ω c 1+ jf /f c

−6
L 2 L 10
2
=2,500 −> R = = =182 −> R1=13.5 Ω
R C 2,500C 2,500× 2 ,2 ×10 −12

Choose: R = 13 Ω (L = 1 µH and C =
2.2 pF). To find the cut-off
freqnecies, we not:

B=f u − f l=2 MHz

f 0=√ f u f l=100 MHz

Solution of the above two equations in two unknowns will give fl ≈ 99

MHz and fu ≈
101 MHz.
Problem 4. A telephone line carries both voice band (0-4 kHz) and data
band (25 kHz to 1 MHz). Design a filter that lets the voice band
through and rejects the data band. The filter must meet the following
specifications: a) For the voice band, the change in transfer function
should be at most 1 dB; and b) The transfer function should be as
small as possible at 25 kHz, the low end of the data band.

We need a low-pass filter as it should allow low-frequency signals

(voice band) to go through while eliminating high-frequency signals
(data band). The prototype of an RC low-pass filter is shown and its
transfer function is:
 1 1
H ( jω )= =
1+ j ω/ω c 1+ jf /f c

L 2 L 10
−6
fun
=2,500 −> R = = =182 −> R1=13.5 Ω
2
R C 2,500C 2,500× 2 ,2 ×10 −12
ctio
The cut-off frequency of the filter is n
not given and it should be found from lab
the specifications. First, we need the ele
change in transfer function to be at d
most 1 dB for the frequency range of “2”)
0-4 kHz. The transfer function of .
filters that satisfy this constraint is
the curve labeled “1” in the figure and
any transfer function located to the
right of this curve (such as transfer
Second, the transfer function should be as small as possible at 25 kHz.
This requires that we choose the cut-off frequency as small as
possible. Therefore, the transfer function of our filter should be curve
labeled “1” as it has the smallest possible value at 25 kHz:

20 log (|H(jf = 4 kHz)|) = −1 dB → |H(jf = 4 kHz)| =

0.891 Using the expression for H(jω), we have:

1
|H ( jf =4 KHz )|= =0.891
√1+(f / f c )2
 f
f /f c =0.509−> f c = =7.85 kHz
0.509

1 3 −5
f c= =7.85 ×10 −> RC =2.0 ×10
2 π RC

Choosing C = 1 nF, we have R = 2.0 × 104. The commercial values then

are C = 1 nF and
R = 20 kΩ.
LC Filters
RC filters are by far the simplest and the most common type
of filter found in analog circuits, however they suffer from a
relatively slow roll off of the gain: while the gain or
attenuation slope can be made steeper than -20 dB/decade,
the transition region, or knee of the curve (the region where
the gain changes from flat to a log-log slope), will always have
the same shape and frequency width.
LC filters are more complex but can be engineered to
produce much sharper features and steeper fall-off regions.
The standard design for LC filters is an LC ladder with an un-
interrupted ground line, such as in the 5 th order filters shown
in figure 4.7 and 4.8 below:
RTH C C2 C3 C4

RLOAD
Figure 4.8: 5th order high-pass LC filter
 The algebra required for computing the gain and phase Bode
plots for these filters is generally quite cumbersome, and a
computer program (i.e. Maple or Excel) is generally useful for
helping with the design. A number of web-applets can also be found
on the internet for determining all the inductor and capacitor values
that will produce the required filter performance.

Example 1: 5th order Butterworth LC low-pass filter with f =10 kHz for 50 Ω
input
3dB
impedance and 50 Ω output impedance. The relative amplitude of
the output is shown in figure 4.9, below.

Figure 4.9: Bode plot of the gain of a 5th order Butterworth

LC low-pass filter. This Butterworth filter uses C1 = C5 = 0.6946

F, C2 = C4 = 3.0642 F, C3 = 4 F,

L1 = L4 = 5 mH, and L2 = L3 = 9.397 mH. Butterworth filters have
a very flat pass-band, and a reasonably regular phase change across
the knee of the curve.

Example 2: 5th order Chebyshev LC low-pass filter with f =10 kHz for 50 Ω
input
3dB
impedance and 50 Ω output impedance. The relative amplitude of
the output is shown in figure 4.10, below.
 Figure 4.10: Bode plot of the gain of a 5th order Chebyshev LC low-pass filter.

This Chebyshev filter uses C1 = C5 = 1.125 F, C2 = C4 = 1.486 F, C3 = 1.505

F,
L1 = L4 = 0.617 mH, and L2 = L3 = 0.646 mH. Chebyshev filters
have a very sharp knee and fast cut-off, but suffer from irregular
transmission in the pass-band. They also have a highly irregular
phase variation at the knee of the filter.
The above Chebyshev and Butterworth filters can be scaled to
another frequency or load impedance with the following rules:

R load, new f 3 dB , old Rload ,old f 3 dB , new

Lnew =( )( ) Lold ,C new =( )( )C
Rload ,old f 3 db ,new R load , new f 3 dB ,old old

Some remarks on LC filters:

i. Generally, the higher the order of the LC filter, the sharper the cut-off
will be, however this usually requires a trade-off in the regularity of the
output phase.
ii. LC filters do not have resistive elements, and consequently they do
not consume power, and do not load the source signal. However,
the load of the input device into
which the filter sends its output will load the LC filter and f3d point
can shift the B

considerably – always remember to include the load resistor in your calculations.

iii. LC filters are widely used in RF circuits, where active filters do not
have the bandwidth to respond to high frequencies.
iv. One cannot construct a filter with arbitrary gain and phase profiles.
While we have treated filters in Fourier space, in the time domain
filters must obey causality. In fact, one can derive Kramers-Kronig
relations for filters.
 II. Transmission Lines
Signals are sent from one device to another, or from one part of a
circuit to another with transmission lines. The quality of your
transmission determines the quality of your signal, whether you are
connecting one device to another or a resistor to a capacitor.

LC Ladder Model of a Transmission Line

A transmission line consists of two parallel conductors separated
by a fixed distance d, which gives rise to an effective capacitance C
per unit length. The conductors will also produce a magnetic field,
which is gives the transmission line an effective inductance L per
unit length. We can model the transmission line as a repeated
network of series inductors and parallel capacitors, or LC ladder, as
depicted in the figure below.

If the LC ladder is infinite and has an impedance Z0, then if we add

an extra LC ladder “rung”, the total impedance should not change,
and we obtain the following relationship

From this relation we can extract an expression for Z0, and we obtain

Z 0=
iωL
2 √
+
L ω2 L2
C
−
4
If we consider that L and C are the inductance and capacitance,
respectively, of a short section ∆l of transmission line, then as we
take the limit ∆l0, we have L0 and C0, but L/C constant. In
this limit the equation for Z0 becomes
 Z 0=
√ L
C

where L and C are the inductance and capacitance, respectively, of

the transmission line, and Z0 is called the characteristic impedance
of the transmission line.
It may seem surprising that a network of inductors and
capacitors can have a real impedance, and consequently consume
power. The explanation for this apparent paradox is that since the
network is infinite, power is flowing from one LC ladder rung to the
next ad infinitude, so that power is constantly moving down the
transmission line, though it is
not dissipated in either the inductors or capacitors. Of course, the
power is consumed at the end of the transmission line when we
attach a load resistor.
The LC ladder model is a high frequency model of transmission
line and does not include the wire resistance which can contribute
to signal attenuation.

Transmission Line Impedance Matching

A transmission line with a characteristic impedance of Z0 should
be terminated with a load impedance of Z0 if the transmission line is
longer than 1/10 of the wavelength of the signal (recall wavelength
 c / f , where c is the speed of light in the transmission line and
f is the signal frequency). If the transmission line is not properly
terminated, then the signal will be partially reflected back towards
the source upon arrival at the load.
There are three main types of transmission lines: wires, twisted
pairs, and coaxial cables. In this section we go over their
performance characteristics.

1. Wires
Plain wires are the simplest and the cheapest transmission lines
available: they include the wires you use to connect components on
your breadboard and electrical grid power lines. If the wires are
kept parallel, then the transmission line will provide some
protection to external fields and noise at low frequencies. While
wire transmission lines are simple, they should be avoided
whenever possible since they are susceptible to interference and
high frequency pick-up. At high frequencies a wire transmission line
actually becomes an antenna (for both transmission and reception).

2. Twisted Pairs
A twisted pair of wires provides good protection from outside
fields and noise and can transmit relatively high frequency signals
without difficulty. The electro-magnetic field of a signal kept close to
the wire pair. A twisted pair can transmit analog signals at up to
250 kHz (sometimes even 1 MHz) and digital signals up to 100 MHz.
Twisted pairs are quite common and are used in many computer
communications cable, such as RJ45 Ethernet cables which has a
characteristic impedance of Z0 = 100 Ω. Twisted pair transmission
lines are easy to make: just take two wires of equal length and twist
them together!
 3. Coaxial Cables
Coaxial cable is the best form of transmission line, short of a
waveguide. In the limit of perfect conductors, signals on coaxial
cables are impervious to external fields and do not radiate either.
Coaxial cables can be used for frequencies up to 1 GHz. At high
frequencies, a significant fraction of the transmitted energy/power is
in the electric and magnetic fields that carry the signal though the
cable, instead of in the potential energy of the current carrying
electrons. The speed of light in a coaxial cable is usually 60-70% of

the speed of light in vacuum. The characteristic impedance of most

coaxial cables used in industry and labs is Z0 = 50 Ω -- coaxial cable
for cable TV is an exception, it has a characteristic impedance of 75
Ω.

Table of Contents

I/Filters…………………………………………………………………………
RC filters……………………………………………………………………
1. Low-pass RC filter…………………………………………………
Low-pass RL filter…………………………………………………
Genral first-order low past filters…………………………………..
Terminated RL low-pass filters…………………………………….
First order high pass filters…………………………………………
High pass RC filters……………………………………………….
High –pass RL filters……………………………………………..
Series RLC Band-pass filters………………………………………
Wide-Band Band-Pass filters……………………………………,
II/Transmission lines……………………………………………………………
LC-Ladder Model of a Transmission Line……………………………………
Transmission Line Impedance Matching……………………………………...