CONTROL

SYSTEMS THEORY
Transfer Function of Physical
Systems

CHAPTER 2
Objectives
 To learn about transfer functions.
 To develop mathematical models from
schematics of physical system.
Overview
 Review on Laplace transform

 Learn about transfer function

 Electric network
 Translational mechanical system
 Rotational mechanical system

 You will learn how to develop mathematical model.

 Present mathematical representation where the input,

output and system are different and separate.

 Solving problems in group and individual

Introduction
 A differential equation
 An equation that involves the derivatives
of a function as well as the function itself.
If partial derivatives are involved, the
equation is called a
partial differential equation;
If only ordinary derivatives are present,
the equation is called an
ordinary differential equation.
Introduction
 Differential equations. How to obtain?

Physical law of the process Differential equation

Examples :
Mechanical system (Newton’s laws)
Electrical system (Kirchhoff’s laws)
Introduction
Introduction
Introduction
 RLC circuit
~KVL
~Ohm’s law
Introduction
 Control
system can be represented using a
mathematical model.
 E.g. LED circuit

 Mathematical model is based on the

schematic of physical systems.
Introduction

Differential equation
t
di (t ) 1
L
dt
+ Ri (t ) +
C�
0
i (t )dt = v (t )
Introduction
• Differential equation describes the
relationship between the input and output
of a system.
d nc ( t ) d n -1c ( t ) d mr ( t ) d m -1r (t )
n
+ an -1 n -1
+�
��+ a0 c ( t ) = bm m
+ bm -1 m -1
+�
��+ b0 r (t )
dt dt dt dt

c(t) output
r(t) input
Introduction
It is easier if we can see the input and the
output clearly such as in the figure below.
Type of input test

Test waveforms used

in control systems
Laplace transform review

Table 2.1
Laplace transform table
Laplace transform review

Table 2.2
Laplace transform
theorems
Transfer function
 What is transfer function?
 Transfer function is output divided by input.

Output
= Transfer Function
Input
C ( s)
= G(s)
R( s)
Transfer function
Transfer function
 Limited to linear system
 What is linear system?
 A system that has two mathematical properties:
Homogeneity and additivity.

Homogeneity
Transfer function

Additivity
Transfer function
Example:
Find the transfer function represented by
dc(t )
+ 2c(t ) = r (t )
dt
Solution:
sC ( s ) + 2C ( s ) = R( s)
The transfer function G(s) is
C (s) 1
G ( s) = =
R(s) s + 2
Transfer function
Example:
Find the transfer function represented by
dc(t )
+ 0.5c(t ) = 0.3r (t )
dt

Solution:
sC ( s ) + 0.5C ( s ) = 0.3R ( s )

The transfer function G(s) is

C ( s) 0.3
G ( s) = =
R ( s ) s + 0.5
Transfer function
Example:
Given the transfer function for a system is
1
G (s) =
( s + 2)
The input for the system, r(t), is a unit step.
r(t)=u(t), assuming zero initial conditions.

What is the output, c(t), of the system?

Transfer function
Solution:
Refer to Table 2.1. Laplace transform for a unit step input is
1/s. We know that
Output
= Transfer Function
Input
so C ( s)
= G ( s)
R( s)
C ( s ) = R ( s)G ( s )
1 1 
= 
s  s + 2 
1
=
s( s + 2)
Transfer function
Expanding the partial fraction, we get

1/ 2 1/ 2
C (s) = -
s s+2
Taking the inverse Laplace transform (use Table
2.2) of each term,
1 1 - 2t
c(t ) = - e
2 2
Transfer function
 You are going to apply transfer function in
three types of mathematical modeling.

 Electric network
 Translational mechanical system
 Rotational mechanical system
Electric Network Transfer
Functions
 We are only going to apply transfer function to
the mathematical modeling of electric circuits for
passive networks (resistor, capacitor and
inductor).

 We will look at a circuit and decide the input and

the output.

 We will use Kirchhoff’s laws as our guiding

principles.
 Mesh analysis
 Nodal analysis
Electric Network Transfer
Functions
Component Voltage- Current- Voltage- Impedance Admittance
Current voltage charge Z(s)= Y(s)=
V(s)/I(s) I(s)/V(s)
Electric Network Transfer
Functions
Simple circuit via mesh analysis
We can obtain transfer function using
Kirchhoff’s voltage law and summing
voltages around loops or meshes. This
method is called loop or mesh analysis.
Electric Network Transfer
Functions
Example:
Find the transfer function relating the
capacitor voltage, Vc(s), to the input
voltage, V(s), in figure below
Electric Network Transfer
Functions
Solution:
Redraw the circuit using Laplace transform.
Replace the component values with their
impedance values.
Electric Network Transfer
Functions
Determine the input and the output for the
circuit. For this circuit,
Input is V(s)
Output is VC(s)

Next, we write a mesh equation using the

impedance as we would use resistor values
in a purely resistive circuit.
Electric Network Transfer
Functions

We obtain
 1 
 Ls + R +  I ( s ) = V ( s ) (2.7)
 Cs 
Electric Network Transfer
Functions
If we look at Eq.2.7, we can only find the input,
V(s). In order to calculate the transfer function we
must have the output which is VC(s).

From the circuit,

1
VC ( s ) = I ( s )
Cs
Electric Network Transfer
Functions

1
VC ( s ) = I ( s )
Cs
I ( s ) = VC ( s )Cs (2.8)

Substituting Eq.2.8 into Eq.2.7.

 1 
 Ls + R + VC ( s )Cs = V ( s )
 Cs 
VC ( s ) 1 1
= =
V (s)  1  CLs 2 + RCs + 1
Cs Ls + R + 
 Cs 
1 / LC
=
R 1
s2 + s +
L LC
Electric Network Transfer
Functions
We can also present our answer in block
diagram
Electric Network Transfer
Functions
 Solution summary

Using mesh analysis

Electric Network Transfer
Functions
Simple circuit via nodal analysis
We obtain the transfer function using
Kirchhoff’s current law and summing current
flowing from nodes. This method is called
nodal analysis.
Electric Network Transfer
Functions
Example: node

We will try to solve the previous example

but this time using nodal analysis.
Electric Network Transfer
Functions
Solution:
We will look at the current flowing in and out of the
node whose voltage is VC(s). We assume the current
leaving the node is positive and current entering the
node is negative.
VC ( s ) V ( s ) - VC ( s )
=
1 / Cs R + Ls
VC ( s ) V ( s ) - VC ( s )
- =0
1 / Cs R + Ls

VC ( s ) 1 / LC
=
V ( s) s 2 + R s + 1
L LC
Electric Network Transfer
Functions
Complex circuit via mesh analysis
To solve a complex circuit we will perform the
following steps.
1. Replace passive elements values with their
impedances.
2. Replace all sources and time variables with their
Laplace transform.
3. Assume a transform current and a current
direction in each mesh.
4. Write Kirchhoff’s voltage law around each mesh.
5. Solve the simultaneous equations for the
output.
6. Form the transfer function.
Electric Network Transfer
Functions
 Find the transfer function, I2(s)/V(s)

Output I2(s)
Input V(s)
Electric Network Transfer
Functions

Sum of 
Sum of  impedances  Sum of applied 
impedances  I ( s ) -   I ( s ) =  voltages around 
  1 common to the 2  
around Mesh 1   Mesh 1 
 two meshes 
Sum of 
impedances  Sum of  Sum of applied 
  I1 ( s ) - impedances  I 2 ( s ) =  voltages around 
common to the    
  around Mesh 2 Mesh 2 
 two meshes 
Electric Network Transfer
Functions
( R1 + Ls ) I1 ( s ) - LsI 2 ( s ) = V ( s ) (1)
 1 
- LsI1 ( s ) +  Ls + R2 +  I 2 ( s ) = 0 (2)
 Cs 

We need to solve both equation (1) and (2) to get

the value of I2(s) and V(s).

You can use substitution method or Cramer’s rule

Electric Network Transfer
Functions
It is easier if we use Cramer’s rule

( R1 + Ls ) V ( s )
- Ls 0 LsV ( s)
I 2 (s) = =
 
where

( R1 + Ls ) - Ls
=  1 
- Ls  Ls - R2 + 
 Cs 
Electric Network Transfer
Functions
Forming the transfer function, G(s), yields

I 2 ( s ) Ls LCs 2
G(s) = = =
V ( s )  ( R1 + R2 ) LCs 2 + ( R1 R2C + L ) s + R1
Example
 Solve using mesh analysis and Nodal
analysis
Mesh analysis method
 Write the mesh equations :

 Solving the equations for I2(s):

Mesh analysis method
Nodal analysis method
 Write the nodal equations :
Nodal analysis method
Mechanical Systems
 Types of Mechanical Systems
•Translational Systems
•Rotational Systems
Translational Mechanical
System Transfer Function
 We are going to model translational
mechanical system by a transfer function.
 In electrical we have three passive
elements, resistor, capacitor and inductor.
In mechanical we have spring, mass and
viscous damper.
Translational Mechanical
System Transfer Function
 In electrical we have resistance, capacitance and
inductance but in mechanical we have, spring
constant (K), viscous damper (fv) and mass (M).

 We are going to find the transfer function for a

mechanical system in term of force-displacement
(i.e. forces are written in terms of displacement)
Translational Mechanical
System
 Mass
 f(t) represents the applied force, x(t)
represents the displacement, and M represents
the mass. Then, in accordance with Newton’s
second law,

Where v(t) is velocity and a(t) is acceleration. It is

assumed that the mass is rigid at the top connection
point and that cannot move relative to the bottom
connection point.
Translational Mechanical
System
 Damper
 Damper is the damping elements and damping is the
friction existing in physical systems whenever
mechanical system moves on sliding surface. The
friction encountered is of many types, namely stiction,
coulomb friction and viscous friction force
 In friction elements, the top connection point can move
relative to the bottom connection point. Hence two
displacement variables are required to describe the
motion of these elements, where B is the damping
coefficient
Translational Mechanical
System
 Spring
 The final translational mechanical element is a
spring. The ideal spring gives the elastic
deformation of a body. The defining equation
from Hooke’s law, is given by
Translational Mechanical
System Transfer Function
Force-velocity,
force-
displacement,
and impedance
translational
relationships
for springs,
viscous dampers,
and mass
Translational Mechanical
System Transfer Function
Simple system

 Assumption
 Movement to the left is assumed to be
positive.
 positive +ve
Translational Mechanical
System Transfer Function
 Draw a free body diagram, placing on the
body all forces that act on the body either in
the direction of motion or opposite to it.

 Use Newton’s law to form a differential

equation of motion by summing the forces
and setting the sum equal to zero.

 Assume zero initial conditions, we change

the differential equation into Laplace form.
Translational Mechanical
System Transfer Function
Example:

Find the transfer function, X(s)/F(s), for the system

in Figure (a).
Translational Mechanical
System Transfer Function

 Draw the free body diagram

 Place on the mass all forces felt by the mass.
 Assume the mass is travelling toward the right.
 Use Laplace transform.
Translational Mechanical
System Transfer Function

 How do we get the equation?

 Answer:
 [sum of impedances]X(s) = [sum of applied forces]
 Movement to the left is positive
Translational Mechanical
System Transfer Function

 The equation in Laplace form is

2
( Ms + f v s + K ) X ( s ) = F ( s )
Translational Mechanical
System Transfer Function

Solving for the transfer function

X (s) 1
G (s) = =
F ( s ) Ms 2 + f v s + K
Translational Mechanical
System Transfer Function
 Complex system
 Find transfer function, X2(s)/F(s)
Translational Mechanical
System Transfer Function
a. Forces on M1 due
only to motion of
M1
b. forces on M1 due
only to motion of
M2(sharing);
c. all forces on M1
Translational Mechanical
System Transfer Function
a. Forces on M2
due only to motion
of M2;
b. forces on M2 due
only to motion
of M1 (sharing);
c. all forces on M2
Example
 Solve for G(s) = X2(s)/F(s)
Exercise
 Solve for G(s) = X2(s)/F(s)

Friction, fv
(Damper)
Rotational Mechanical
System
Rotational Mechanical
System Transfer Functions
 We are going to solve for rotational
mechanical system using the same way as
the translational mechanical systems
except

 Torque replaces force

 Angular displacement replaces translational
displacement
Rotational Mechanical
System Transfer Functions
 In translational mechanical system we
have three elements; spring, damper
and mass. In rotational mechanical
system we have; spring, damper and
inertia.
Rotational Mechanical
System Transfer Functions
Torque-angular
velocity, torque-
angular
displacement,
and impedance
rotational
relationships for
springs, viscous
dampers, and
inertia
Rotational Mechanical
System Transfer Functions
Simple system

 Assumption
 Anti-clockwise movement is assumed to be
positive
Rotational Mechanical
System Transfer Functions
Example:

Find the transfer function, θ(s)/T(s)

Rotational Mechanical
System Transfer Functions

 Draw the free body diagram

T(s)

Kθ(s)
Js2θ(s)
Θ(s) J

Dsθ(s)
Rotational Mechanical
System Transfer Functions
 We will use
 [sum of impedances]θ(s) = [sum of applied torque]
 Based on the free body diagram, the
equation of motion is
( Js 2
)
+ Ds + K q ( s ) = T ( s )
q ( s)
we know transfer function, G( s), is
T (s)
q (s) 1
= 2
T ( s ) Js + Ds + K
1
= J
2 D K
s + s+
J J
Rotational Mechanical
System Transfer Functions
 Complex system
 Find transfer function θ2(s)/T(s)
Rotational Mechanical
System Transfer Functions
a. Torques on J1
due only to the
motion of J1
b. torques on J1
due only to the
motion of J2
c. final free-body
diagram for J1
Rotational Mechanical
System Transfer Functions
a. Torques on J2
due only to the
motion of J2;
b. torques on J2
due only to the
motion of J1
c. final free-body
diagram for J2
Laplace transform review
 The Laplace transform is defined as
�

ℒ[ f (t ) ] = F ( s ) = � dt
f (t )e - st

0-

Where s = s + jw

Inverse Laplace transform, to get f(t) given

F(s), is s + j�
1
ℒ [
-1
F ( s ) ] = � dsF ( s ) e st
2p j
s - j�
Laplace transform review

Table 2.1
Laplace transform table
Laplace transform review

Table 2.2
Laplace transform
theorems
Laplace transform review
Example
Find the inverse Laplace transform of

F(s)=1/s

F(s)=1/s2

F(s)=1/(s+3)
Laplace transform review
Example:
Find the inverse Laplace transform of

1
F (s) =
Answer
s+3
We use frequency shift theorem, item 4 in Table 2.2, and Laplace
transform of f(t)=u(t), item 3 in Table 2.1

If inverse transform of F(s)=1/s2 is tu(t), the inverse transform of

F(s+a)=1/(s+a)2 is e-attu(t).

Hence, f(t)=e-3ttu(t)
Laplace transform review
Example:
Find the inverse Laplace transform of

Answer F ( s ) = 1/ ( s + 3) 2
We use frequency shift theorem, item 4 in Table 2.2, and Laplace
transform of f(t)=tu(t), item 3 in Table 2.1

If inverse transform of F(s)=1/s2 is tu(t), the inverse transform of

F(s+a)=1/(s+a)2 is e-attu(t).

Hence, f(t)=e-3ttu(t)
Laplace transform review
Can you solve this question using Table 2.1
and Table 2.2?
F ( s ) = 1 / ( s + 3)3

Or this question

s3 + 2s 2 + 6s + 7
F (s) =
s2 + s + 5
Laplace transform review
We can find the inverse Laplace transform
for complicated function by changing the
function to a sum of simpler term for which
we know the Laplace transform of each
term.

The result is called partial-fraction expansion

Partial fraction expansion
Let’s take F(s)=N(s)/D(s)

In order for us to do the partial-fraction

expansion we need to make sure the order
of N(s) is smaller than or equal to the order
of D(s).

N(s) <= D(s)

Partial fraction expansion
What will we do if

N(s) >= D(s) ?

Answer.
We will divide N(s) by D(s) successively until
the result has a remainder whose
numerator is of order less that its
denominator (polynomial division).
Partial fraction expansion
Polynomial long division
Example:
Divide x2+9x+14 by x+7

Answer
Partial fraction expansion
Divide 3x3 – 5x2 + 10x – 3 by 3x + 1

Answer:
Partial fraction expansion
Can be divided into 3 cases.

1. Roots of the denominator of F(s) are real

and distinct
2. Roots of the denominator of F(s) are real
and repeated
3. Roots of the denominator of F(s) are
complex or imaginary
Partial fraction expansion:
Case 1
Roots of the denominator of F(s) are real
and distinct
Example of an F(s) with real and distinct
roots in the denominator is
2
F (s) =
( s + 1)( s + 2)
Partial fraction expansion:
Case 1
Solution
2 K1 K2
F (s) = = +
( s + 1) ( s + 2 ) ( s + 1) ( s + 2) (2.1)

To find K1, we multiply Eq. (2.1) by (s+1),

which isolates K1. Thus
2
K1 = =2
( s + 2) s �-1

2
K2 = = -2
( s + 1) s �-2
Partial fraction expansion:
Case 1
Change value K1 and K2 gives
2 K1 K2
F ( s) = = +
( s + 1) ( s + 2 ) ( s + 1) ( s + 2)
2 2 2
F ( s) = = -
( s + 1) ( s + 2 ) ( s + 1) ( s + 2)

Using Table 2.1, the inverse Laplace

transform is ??

f ( s ) = (2e -t - 2e -2t )u (t )
Partial fraction expansion:
Case 1
Problem: Given the following differential
equation, solve for y(t) if all initial condition
are zero. Use Laplace transform.

d2y dy
2
+ 12 + 32 y = 32u (t )
dt dt
Partial fraction expansion:
Case 2
Roots of the denominator of F(s) are real
and repeated
Example of an F(s) with real and repeated
roots in the denominator is
2
F (s) =
( s + 1)( s + 2) 2

Roots of (s+2)2 in the denominator are

repeated.
Partial fraction expansion:
Case 2
 We write the partial-fraction expansion as a sum
of terms. Each factor of the denominator forms
the denominator of each term.

 Each multiple root generates additional terms

consisting of denominator factors of reduced
multiplicity.

2 K1 K2 K3 (2.2)
F (s) = = + +
( s + 1)( s + 2) 2
( s + 1) ( s + 2) 2
( s + 2)
Partial fraction expansion:
Case 2
 We can solve for K1 using method in case
1 in Eq.2.2.
2
K1 = 2
=2
( s + 2) s �-1

 Next step is to isolate K2 by multiplying

Eq. 2.2 by (s+2)2
2 2 K1
= ( s + 2) + K 2 + ( s + 2 ) K3 (2.3)
s +1 ( s + 1)
Partial fraction expansion:
Case 2
 Solve for K2 using the same method as
Case 1 in Eq.2.3.
2
K2 = = -2
( s + 1) s �-2

 To get the value of K3 we need to

differentiate Eq.2.3 with respect to s,
-2 ( s + 2) s
= K1 + K 3 (2.4)
( s + 1) ( s + 1)
2 2
Partial fraction expansion:
Case 2
 Using the method in case 1. K3 can be
found if we let s approach -2. (-2 comes
from the denominator of K3)
-2
K3 = 2
= -2
( s + 1) s �-2

 Changing K1, K2 and K3 with their

respective values
2 K1 K2 K3
F (s) = = + +
( s + 1)( s + 2) 2
( s + 1) ( s + 2) 2
( s + 2)
2 2 2
= - -
( s + 1) ( s + 2) 2
( s + 2)
Partial fraction expansion:
Case 3
Roots of the denominator of F(s) are
complex or imaginary
Example of an F(s) with complex roots in
the denominator is
3
F ( s) =
(
s s 2 + 2s + 5 )
We can expand the function in partial
fractions as 3 3
F (s) = =
( 2
s s + 2s + 5 ) s ( s + 1 + j2) ( s + 1 - j2)

K1 K2 K3
= + +
s s + 1 + j 2 s + 1 - j1
Partial fraction expansion:
Case 3
Solve for K1 using similar method in case 1.
3 3
K1 = =
(s 2
+ 2s + 5 ) s �0
5

Solve for K2 using similar method in case 1.

3 3
K2 = = - ( 2 + j1)
s ( s + 1 - j2) s �-1- j 2
20
Partial fraction expansion:
Case 3
K3 is found to be the complex conjugate of
K2
3 3
K3 = = ( 2 + j1)
s ( s + 1 + j 2) s �-1+ j 2
20

Changing K1, K2 and K3 with their respective

values
3 / 5 3 � 2 + j1 2 - j1 �
F (s) = - � + � (2.5)
s 20 �s + 1 + j 2 s + 1 - j 2 �
Partial fraction expansion:
Case 3
Inverse Laplace transform for Eq.2.5 is
3 3 �
f (t ) =- (�2 + j1) e-( 1+ j 2) t + ( 2 - j1) e-( 1- j 2) t �
�
5 20
3 3 -t ��e j 2t + e- j 2t � �e j 2t + e - j 2t � �
= - e � 4�
� �
� + 2�� �
�� (2.6)
5 20 �� 2 � � 2j ��

Using equation

e jq + e - jq e jq - e - jq
= cos q and = sin q
2 2j
Into Eq.2.6
Partial fraction expansion:
Case 3
The final result is
3 3 -t � 1 �
f (t ) = - e � cos 2t + sin 2t �= 0.6 - 0.671e -t cos ( 2t - f )
5 5 � 2 �
where f = arctan 0.5 = 26.57o