Lecture-2

Laplace Transform
27-03-2013
By Mr. Singappuli Wijewardana
PhD Student QMUL
(2-hours)
Learning Objectives:
1. Laplace Transform as a Mathematical Tool
2.Step function, Unit step function.
3.Theorems of Laplace Transform and Laplace
transform tables
4.Transfer Function and Characteristic equations
5. Mathematical Modelling of Circuits
6. Circuit simulation and analysis using Matlab
7. Introduction to s-plane, Poles and Zeros of a
transfer function



Laplace transform is a
mathematical Tool
Differential equations in time domain
is converted into s domain for easy
manipulation of algebra and then later
on it is converted back into the time
domain.
Fourier transform cannot be used in
some occasions as it does not
converge in certain functions.
Definition of Laplace Transform
Let f(t) is a function of time t such that f(t) = 0 for t<0
Let f(t) be multiplied by a convergence factor ,
where is real and positive,
converges then Laplace transform of f(t) is given by

[f(t)] =F(s) =

Where s = + j and j = , s is defined as a complex
variable and & are real variables.
is the symbol used to denote the Laplace transformation
F(s) = Laplace transform of f(t)



t
e
o
dt e t f
t
}

0
). (
o
} }

=
0 0
. ) ( )] ( [ dt e t f t f dt e
st st
1
Inverse Laplace Transform
Given the Laplace transform function F(s),
the operation of finding f(t) is termed the
inverse Laplace transformation and it is
denoted by:
f(t) =
-1
[F(s)]
Example-1:
U(t) is a unit step function defined as:
0
0 , 1
0
) (
) (
<
>

=
=
t
t t u
t f
f(t)
1
t=0
t
Fig.2.1
}

=
0
0
1
. )] ( [
st st
e
s
dt e t u
s se
t when
se
t When
1 1
1 , 0 , 0
1
,
0
=
|
.
|

\
|

= =

=

F(s) = [e
t
] =
F(s) = [u(t)] =
Hence, [u
s
(t)] = 1/s (Unit Step function)
Example-2:
Laplace transform of e
t
when t 0, where is a real constant
o o
o
o o
+
=

+
= =
} }

+
+
s s
e
dt e dt e e
t s
t s st t
1
0
. . .
0 0
) (
) (

Example-3:
Laplace transform of a Step Function:
If f(t) is a step function of magnitude b with the discontinuity at t=0 the function f(t)
can be written as follows:
f(t) = b u(t)
F(s) = [bu(t)

] =
s
b
s
e e b
s
be
dt be dt e t bu
s s st
st st
=
+
=

= =
} }



0 0
) 0 ( ) (
) (
0
. . ). (
When evaluating Laplace Transform, the term s is regarded as any constant that
makes F(s) convergent. From the above workout if s is positive and greater than 0
then e
-s()
= 0. When e
-s(0)
= e
0
=1 which is correct with our results. However, if s
becomes negative then immediately the value of e
-s()
= ,thus creating a situation
F(s) is not convergent. And this is against the principle definition of Laplace
transform. Therefore the function F(s) is defined only to the right of imaginary axis of
s-plane.
f(t)
b.u(t)
t=0
t
b
Fig.2.2
Theorems of Laplace Transformations
Theorem-1: Constant multiplication
If a is a constant and if F(s) is the Laplace
Transform of f(t), then

[a.f(t)] = a.F(s)

Theorem -2: (linearity)
If F
1
(s) and F
2
(s) are Laplace transform of
f
1
(t) and f2(t) respectively, then
[f
1
(t)+f
2
(t)] = F
1
(s) +F
2
(s)
Theorem-3:
If F(s) is the Laplace transform of f(t) and if
f(0) be the limit of f(t) as t approaches to 0,
then the Laplace transform of the time
derivative of f(t) is:

Theorem-4:
Laplace transform of first integral of f(t) with respect to
time is the Laplace Transform of f(t) divided by s:

) 0 ( ) ( )] ( [ ) ( )
) (
(
lim
0
f s sF t f s sF
dt
t df
t
= =

s
s F
dt t f
) (
). (
0
=
}
t

Resistors are time invarient whether it is in Laplace domain or in time domain:
Hence we can take: (resistor)= R= r
With the Ohms law we have an important ratio: V/I =R: This ratio must be same
either in time domain or in Laplace domain therefore, we can take:

V(s)/I(s) = R(s)

Capacitors:
In time domain the relationship between, current, voltage and the capacitance is:
current through a capacitor changes according to the voltage drop across it.

dt
t dv
C t i
) (
) ( =
Solving for V(t) we can
write:
C is the capacitance of
the capacitor in Farads
) (
1
) (
:
). (
1
) (
0
s I
Cs
s V
domain Laplace
dt t i
C
t v
t
=

=
}

Hence:
V(s)/I(s) is a kind of resistance(impedance) = 1/Cs for a capacitor in Laplace
domain.
(capacitor)= 1/Cs
Inductor:
We know that relationship:
) ( ) (
:
) (
) (
s LsI s V
domain Laplace
dt
t di
L t v
=

=
Once again the ratio of V(s)/I(s) = Ls : (inductor)= Ls

Theorem 5 :
The Laplace Transform of f(t) delayed by T is
the Laplace transform f(t) multiplied by e
-Ts

[f(t-T)u(t-T)]= e
-Ts
.F(s)
U(t-T) denotes the unit step function which is
shifted in time to the right by T.
Theorem 6 :
Initial Value Theorem:
If Laplace transform of f(t) is F(s) then


) ( . lim
lim
)] ( [
0
s F s
s
t
t f

=
If the limit exists
Theorem -7:
Final Value Theorem:
If the Laplace transform of f(t) is F(s), and if sF(s) is analytic
on the imaginary axis and in the right half of the s-plane
then:
) ( . lim
lim
0
)] ( [
s F s
s
t
t f

=
Convolution Theorem:
If we have Y(t) = x(t)*h(t)

h(t)
Y(t) x(t)
{ x(t)*h(t) } = X(s).H(s)
Asterisk mark denotes the
convolution. It is not the
normal multiplication that we
are talking.

But in Laplace domain it
becomes the normal
multiplication:
Y(s) = X(s) . H(s)
Convolution in the time domain becomes
multiplication in the S domain, and
convolution in the S domain becomes
multiplication in the time domain
Theorem-8:
Heaveside Expansion Formula:
If F(s) is a rational algebraic function it can be
written in the form:


) (
) (
) (
s B
s A
s F =
where A(s) and B(s) are polynomials in s:

-1
[F(s)]=
Then f(t) =

-1
is the inverse Laplace transform which we havent
introduced yet. S
i
are the n distinct roots of the
characteristic equation B(s) =0 and B signifies
differentiation with respect to s

=
n
i
i
t s
i
s B
e s A
i
1
'
) (
) (
Table:2.1- Laplace Transform
Transfer Function
(Output/Input)
The relationship between the input
and the output of a control system is
very important in control engineering
to understand the system behaviour.
The Transfer Function(T.F) of a
linear system is defined as the ratio of
the Laplace transform output to the
Laplace transform input when all initial
conditions are zero.
Conventionally the symbol used to represent
the transfer function is G(s). In Laplace
domain d/dt is replaced by s and d
2
/dt
2
is s
2
etc.
If a transfer function(T.F) is represented as

G(s)
) (
) (
s B
s A
=
Where A(s) and B(s) are polynomials in s. For a system to be
physically realizable the order of the numerator can not
exceed that of the denominator. [A(s) = output, B(s) = Input)]
Output response of a system when subjected to a changing input
variable is determined by the values of the roots of the equation
B(s)= 0. This is termed the Characteristic equation(C.E).
The roots of the C.E are referred as the poles of
the overall T.F. The number of roots is the CE is
the order of the system.
Poles and Zeros
The poles of a transfer function can be
defined as the value of Laplace transform
variable of denominator of the TF which
leads the TF to infinity.
e.g:

) (
) (
1
1
) (
s B
s A
s
s G =
+
=
Here the denominator is s+1 and s=-1 tends the G(s) to infinity.
Hence G(s) has pole at s=-1. Pole is marked as x(cross) in
the s-plane.
Zeros of a TF are the roots of the numerator or the values of s that
makes the TF to zero. It is marked as o in the s-plane.
Application of Laplace Transformations
e.g: Given the RL circuit,
determine the current as a
function of time t after the
switch is closed at t=0, by
taking all initial conditions as
zero.
e(t)
t=0
i(t)
Fig.2.3: R-L Circuit
Solution:
Applying the Kirchoffs law:
1 .
) (
) ( ) ( eq
dt
t di
L t Ri t e + =
By taking e(t) as a step input to the circuit:
) / (
/
). (
) (
1
. ) (
L R s s
L E
Ls R
L
s
L
E
Ls R s
E
s I
+
=
+
=
+
=
s
E
e
s
E
dt e t e
st
st
= =

}
0
0
. ). (
{e(t)} =
Let us take i(t) in Laplace domain as I(s) and the resistance in Laplace
domain as R. Equation 1 in the previous slide can now transform into
Laplace domain:
E/s = R.I(s) + L.s.I(s)

E/s = I(s)[R +L.s]

) (
) / ( ) / (
/
fractions partial taking by
L R s
B
s
A
L R s s
L E
+
+ =
+
) / (
) / (
) / (
/
L R s s
Bs L R s A
L R s s
L E
+
+ +
=
+
By comparing the numerator
coefficients, we can get A=E/R and B =
-E/R
(

+
=
+
=
L R s s R
E
L R s s
L E
s I
/
1 1
) / (
/
) (
By taking the Inverse Laplace transform
(

=
t
L
R
e
R
t e
t i 1
) (
) (
_______________eq2
E/s
I(s)
R-L Circuit in Laplace Domain
Ls
E/s = RI(s) + L.s.I(s)
= I(s){R +Ls}
Hence I(s) = E/s{R+Ls}


Remember the voltage applied is taken as a
step input to the system:
Plot of eq-2 using Matlab
function dcrl = RL (t,i)
%function for RL circuit calculations
clear all;
Vs=50;
R=1;
L=9;
t=1:1:100;
i0=0;

for x=1:30,
i(x)=(Vs/R)*(1-exp(-(R/L)*t(x)));
end
for x=31:100,
i(x)=i(30)*exp(-(R/L)*t(x));
end
plot (t,i)

By viewing the circuit as a voltage
divider: in Laplace domain:
Analysis of eq-2 with Multisim
RL circuit is sometimes called as RL filter or analog infinite
response electronic filters
RL can be in series or in parallel
V
in
V
L
V
R
L

I

) ( .
) (
) (
) ( .
) (
) (
s V
Ls R
R
s V
s V
Ls R
Ls
s V
in R
in L
+
=
+
=
Applying a step input voltage V: in
Laplace domain V/s
) (
.
.
) (
) (
Ls R s
Ls
V
s
V
Ls R
Ls
s V
L
+
=
+
=
Using partial fractions and inverse Laplace transform:
V
L
(t) = V. e
-Rt/L
V
R
(t) = V(1- e
-Rt/L
)
Time-Constant
These equations show that a series RL circuit has a time constant,
usually denoted by =L/R, being the time it takes the voltage across the
component to either fall (across L) or rise(Across R) to within 1/e of its
final value.
So the voltage across L will have dropped to about 37% after , and
essentially to zero (0.7%) after about 5
Time constant of a First Order System
The final value of Vc(t) =E
The time at which Vc(t)=0.63
*
final
value is called the time constant of the
system.
Final value of Vc(t)
63% of Final value of Vc(t)
Time constant
t
Vc(t)
Time constant
Time constant depends on C and R
Larger time constant means the
capacitor needs longer time to
charge.
With Multisim
First under tool bar select the DC power supply and press enter to
bring it to the circuit design window.
Double click the component to add the value.
Change the voltage to 10V
Add a ground similarly for simulation purposes.
Add a resistor from basics and double click on resistor to change
the value to 100 ohms.
From basics again select the inductor with some value for henry
and double click to change it to 1H. Click once to change it to
vertical. Now join the components with a place wire facility.
We know that the time constant is L/R = 1/100 = 0.01
Now we can simulate the circuit by using: Simulate button on the
top. Click Simulate; go to analyses: go to transient analysis: Start
time 0 is correct. End time is that we are going to see only 5x = 5
x0.01=0.05. So change it to 0.05.
If you want to add a net name for the wires connected: click on
the wire: go to properties and click the box to show the number.
Set initial conditions to zero from automatically
Click output button and select i
Now press simulate;
Transient Analysis with Multisim
RLC- Circuit
Find the transfer function of the circuit as shown below. Assume all
initial conditions are zero. e
i
(t) is the input voltage and e
c
(t) is the
output voltage across the capacitor.
e
i
(t) e
c
(t)
From KVL:
}
+ + =
t
0
). (
1 ) (
) ( ) ( dt t i
c dt
t di
L R t i t e
i
Let us assume that we need to
find the voltage across the
capacitor and that is
}
=
t
0
) (
1
) ( dt t i
c
t e
c
Applying Laplace transform to both equations:
sC
s I s E
sC
sL R s I s E
c
i
1
). ( ) (
]
1
)[ ( ) (
=
+ + =
Transfer function is the ratio of output voltage to the input:
) ( ]
1
)[ ( ) (
) (
1
). ( ) (
input
sC
sL R s I s E
output
sC
s I s E
i
c
+ + =

=
) 1 (
1
) (
) (
2
+ +
=
sCR LC s s E
s E
i
c
Transfer function or
output to the input ratio
in s-plane(Domain)
A complex function G(s) is called an analytic function in a particular region in s
plane if the function and all its derivatives exist in the region.
Example: G(s) = 1/(s+1) is an analytic function at all points in s-
plane except at s=-1( when s=-1 the function reaches to infinity.
However, G(s) = s+2 is analytic at every point in G-Plane for all
values of s.
The points in G-Plane at which the function does not exist are
known as singularities of that function.
The pole of a function at point in s-plane is the singularity of that
function.
n
n n
m m m
b s b s b
a s a s a
s G
+ +
+ +
=

1
1 0
1
1 0
...
) (
If a complex variable function G(s) can be represented as:
If we can factorize the above equation as shown below:
) ( ) )( (
) ( ) )( (
) (
2 1
2 1
n
m
p s p s p s
z s z s z s K
s G


=

K is a constant and G(s) does not exists when s=p

n
(n=1,2..)
These points in the s-plane are called of G(s)
When s =z
1
, z
2
.. G(s) reaches to zero. These points are called

and
2
) 5 )( 3 (
) 4 )( 2 ( 15
) (
+ +
+ +
=
s s s
s s
s G
Example:
The TF has poles at s=0, s=-3, and double poles at s=-5: Zeros at
s=-2,and -4.
-5 -4 -3 -2 0
2
) 5 )( 3 (
) 4 )( 2 ( 15
) (
+ +
+ +
=
s s s
s s
s G
S-plane
o
je
Learning Outcomes
Mathematical modelling of electrical
circuits
How to derive a transfer function
(Output-Input relationship)
S-plane and the time-domain
Model simulation using Simulink.
Poles and Zeros
Application of Laplace Transform to
solve differential equations