LAPLACE TRANSFORM AND ITS APPLICATION IN CIRCUIT ANALYSIS

C.T. Pan

12.1 Definition of the Laplace Transform 12.2 Useful Laplace Transform Pairs 12.3 Circuit Analysis in S Domain 12.4 The Transfer Function and the Convolution Integral

C.T. Pan

12.5 The Transfer Function and the Steady state Sinusoidal Response 12.6 The Impulse Function in Circuit Analysis

C.T. Pan

12.1 Definition of the Laplace Transform

Pierre Simon Laplace (1749-1827) :
A French astronomer and mathematician First presented the Laplace transform and its applications to differential equations in 1979.

C.T. Pan

12.1 Definition of the Laplace Transform

Definition:
L [ f (t ) ] = F ( s ) = f (t )e st dt
0

s = + j

a complex variable

The Laplace transform is an integral transformation of a function f(t) from the time domain into the complex frequency domain, F(s).

C.T. Pan

12.1 Definition of the Laplace Transform

One-sided (unilateral) Laplace transform Two-sided (bilateral) Laplace transform
L1 [ F ( s ) ] = f (t ) = 1 1 + j F ( s )e st ds j 2 j 1

Look-up table , an easier way for circuit application f (t ) F ( s )

C.T. Pan 6

12.1 Definition of the Laplace Transform

Similar to the application of phasor transform to solve the steady state AC circuits , Laplace transform can be used to transform the time domain circuits into S domain circuits to simplify the solution of integral differential equations to the manipulation of a set of algebraic equations.
C.T. Pan 7

12.2 Useful Laplace Transform Pairs

Functions
impulse step ramp exponential sine

f(t) , t>0

F(s) 1

(t )
u (t )

t
e at
sin t

1 S 1 S2
1 S +a
S2 + 2
8

C.T. Pan

12.2 Useful Laplace Transform Pairs

Functions
cosine damped ramp

f(t) , t> 0

F(s)

cos t

S S 2 + 2
1

te at

(S + a)

damped sine

e at sin t e at cos t

(S + a) (S + a)

+ 2

S +a
2

damped cosine

+ 2

C.T. Pan

12.2 Useful Laplace Transform Pairs

d L f (t ) = sF ( s ) f (0 ) d t t F (s) L - f ( ) d = 0 S L [ f (t a )u (t a ) ] = e as F ( s ) , a > 0
at L f (t ) e = F (s + a) 1 s F ( ) ,a > 0 L [ f (at)] = a a li m [ f (t ) ] = li m sF ( s )

C.T. Pan

lim [ f
t

t 0+

(t ) ] =

lim
s 0

sF (s)
10

12.2 Useful Laplace Transform Pairs

Example Use the Laplace transform to solve the differential equation.
d 2v dv + 6 + 8v = 2u (t ) dt dt 2 v(0) = 1 v '(0) = 2 Take Laplace transfrom 2 s 2V ( s ) sv(0) v '(0) + 6 [ sV ( s ) v(0) ] + 8V ( s ) = s
C.T. Pan 11

12.2 Useful Laplace Transform Pairs

2 s 2V ( s ) sv(0) v '(0) + 6 [ sV ( s ) v(0) ] + 8V ( s ) = s 2 s + 4s + 2 ( s 2 + 6s + 8)V ( s ) = s 2 s + 4s + 2 s 2 + 4s + 2 V ( s ) = = s ( s 2 + 6 s + 8) s ( s + 2)( s + 4) 1 v(t ) = (1 + 2e 2t + e 4t )u (t ) 4
C.T. Pan 12

12.3 Circuit Analysis in S Domain

(1) KCL ,

ik (t ) = 0 ,
n

for any node.

Take Laplace Transform

I k ( s) = 0 ,
n

for any node.

(2) KVL ,

vk (t ) = 0 ,
m

for any loop.

Take Laplace Transform

Vk (s) = 0 ,
m
C.T. Pan

for any loop.

13

12.3 Circuit Analysis in S Domain

(3) Circuit Component Models resistor vR (t ) = RiR (t ) VR ( s ) = RI R ( s ) I R ( s ) = GVR ( s )

C.T. Pan

14

12.3 Circuit Analysis in S Domain

inductor vL (t ) = L diL dt 1 t vL ( )d L 0
LiL (0 )

iL (t ) = iL (0 ) +

VL ( s ) = sLI L ( s ) LiL (0 ) VL ( s ) iL (0 ) I ( s) = + sL s
C.T. Pan

iL (0 ) s
15

12.3 Circuit Analysis in S Domain

capacitor iC = C dvC dt

1 sC vC (0 ) s

1 t vC (t ) = vC (0 ) + iC ( )d C 0 I C ( s ) = sCVC ( s ) Cvc (0 ) v (0 ) 1 VC ( s ) = C + IC (s) s sC

C.T. Pan

1 sC

CvC (0 )

16

12.3 Circuit Analysis in S Domain

Coupling inductors
i1 + v1 L1 M i2 L2

d i1 di2 + M dt dt + d i1 di2 v2 v2 = M + L2 dt dt v1 = L 1

V1(s)=L1SI1(s)-L1i1(0)+MSI2(s)-Mi2(0) V2(s)= MSI1(s)- Mi1(0)+L2SI2(s)-L2i2(0)

C.T. Pan

17

12.3 Circuit Analysis in S Domain

For zero initial conditions

V (s) = Z (s) I (S ) I (S ) 1 admittance @ = Y (s) = V ( s) Z (s) V ( s ) = Z ( s ) I ( s ) ohm ' s law in s domain impedance @
C.T. Pan 18

12.3 Circuit Analysis in S Domain

The elegance of using the Laplace transform in circuit analysis lies in the automatic inclusion of the initial conditions in the transformation process, thus providing a complete (transient and steady state) solution.

C.T. Pan

19

12.3 Circuit Analysis in S Domain

Circuit analysis in s domain nStep 1 : Transform the time domain circuit into s-domain circuit. nStep 2 : Solve the s-domain circuit. e.g. Nodal analysis or mesh analysis. nStep 3 : Transform the solution back into time domain.
C.T. Pan 20

12.3 Circuit Analysis in S Domain

Example Find vo(t) given vo(0)=5V

S-domain equivalent circuit

10 s +1
C.T. Pan

10 s
21

12.3 Circuit Analysis in S Domain

10 s +1 10 s

Nodal analysis 10 Vo ( s ) V ( s) V (s) s +1 + 2 + 0.5 = o + o 10 10 10 / s 25s + 35 10 15 Vo ( s ) = = + ( s + 1)( s + 2) s + 1 s + 2 vo (t ) = (10et + 15e2t )u (t ) V

C.T. Pan 22

12.4 The Transfer Function and the Convolution Integral

Given a linear circuit N in s domain as shown below

Transfer function H(s) is defined as

H(s) =
C.T. Pan

Y(s) X(s)
23

12.4 The Transfer Function and the Convolution Integral

If Y ( s ) = Vo ( s ) , X ( s ) = Vi ( s ) ; then H ( s ) = voltage gain If Y ( s ) = I o ( s ) , X ( s ) = I i ( s ) ; then H ( s ) = current gain If Y ( s ) = V ( s ) , X ( s ) = I ( s ) ; then H ( s ) = impedance If Y ( s ) = I ( s ) , X ( s ) = V ( s ) ; then H ( s ) = admittance
C.T. Pan 24

12.4 The Transfer Function and the Convolution Integral

Given the transfer funtion H(s) and input X(s) , then Y(s)=H(s)X(s) If the input is Y(s)=H(s) (t) , then X(s)=1 and

Hence , the physical meaning of H(s) is in fact the Laplace transform of the impulse response of the corresponding circuit.
C.T. Pan 25

12.4 The Transfer Function and the Convolution Integral

Y(s)=H(s)X(s) , in s-domain
y (t ) =

h(t ) x( )d @ h(t ) x(t )

in time domain

Geometrical interpretation of finding the convolution integral value at t=tk is based on :

C.T. Pan

26

12.4 The Transfer Function and the Convolution Integral

(1)Approximating the input function by using a series of impulse functions. (2)Shifting property of linear systems input x(t)!output y(t) x(t-")!output y(t- ") (3)Superposition theorem for linear systems (4)Definition of integral : finding the area
C.T. Pan 27

12.4 The Transfer Function and the Convolution Integral

(1)Input x(") is approximated using impulse functions , x(")=0 , for "<0
x( ) x(0)

instant k , x( k ) value , area 0 = 0 , , x(0) x(1 ) x( 2 ) x( 3 ) 1 =V

2 = 2V ,
0 1 2 3 4

, x(1 )V

, x(0)V

3 = 3V ,

, x( 3 )V

, x( 2 )V

x ( )

x( ) f 0 ( ) + f1 ( 1 ) + f 2 ( 2 ) + L
1

C.T. Pan

@ x(k V )V ( k V )
k =0
28

12.4 The Transfer Function and the Convolution Integral

(2) Use the linearity property
input x(0)V ( ) output (response) x(0)V h( )

x( 1 )V ( 1 ) x( 1 )V h( 1 ) M up to = tk
C.T. Pan

x( 2 )V ( 2 ) x( 2 )V h( 2 ) M

29

12.4 The Transfer Function and the Convolution Integral

(3) Use superposition theorem to find the total approximate response

$ y (tk ) = x(k V )V h(tk k V )

n k =0

n = integer[

tk ] V
30

C.T. Pan

12.4 The Transfer Function and the Convolution Integral

$ y (tk ) (4) Take the limit , #"!d", y (tk )

y (tk ) =

x( )h(t

)d

Due to causality principle , h(t-")=0 for t tk and x(")=0 for "<0

= h(tk ) x( )d
0
C.T. Pan 31

tk

12.4 The Transfer Function and the Convolution Integral

Example:
Given x(t ) = u (t ) , h(t ) = e t u (t ) , find y (4) , y (t ) =

h(t ) x( )d

y (4) =

h(4 ) x( )d
32

C.T. Pan

12.4 The Transfer Function and the Convolution Integral

Step1.
xx (! )! 1.0
0.8 0.6 0.4 0.2

4 4

2 2

hh ( ! )!
1.0 0.8 0.6 0.4 0.2

C.T. Pan

4 4

2 2

4
33

12.4 The Transfer Function and the Convolution Integral

Step2. Shift to tk=4

4 ) !! h(4 1.0 0.8 0.6 0.4 0.2

C.T. Pan

2 2

!
34

12.4 The Transfer Function and the Convolution Integral

Step3. Find the product h(4-")x(")

C.T. Pan

35

12.4 The Transfer Function and the Convolution Integral

Step4. Find the integral (area)

y (4) = e (4 ) d = e 4 e d =e 4 e
4 4 0 0

4 0

=e (e 1) = (1 e )
4

Step5. Check

C.T. Pan

1 1 1 1 a a Y(s) = H(s)X(s) = = + s +a s s +a s 1 y(t) = (1et ) , a =1 a y(4) =(1e4)

36

12.5 The Transfer Function and the Steady State Sinusoidal Response
From definition of transfer function
Y ( s) X ( s) Y (s) = H (s) X (s) H ( s) =

Assume input X(t)=Acos(wt+ ) and H(s) is given, then one can get the steady state solution without needing a separate phasor analysis.
C.T. Pan 37

12.5 The Transfer Function and the Steady State Sinusoidal Response
proof : X (t ) = A cos cos t A sin sin t A(cos s sin ) X (s) = s 2 + 2 Y (s) = H (s) X (s) K1 K1* + s j s + j + other terms due to poles under steady state : = Yss ( s ) = K1 = K1 K1* + s j s + j

H ( s) A( s cos sin ) s + j
s = j
38

C.T. Pan

1 = H ( j) Ae j 2

12.5 The Transfer Function and the Steady State Sinusoidal Response
Let H ( j) =| H ( j) | e j ( ) and take inverse Laplace transform Then i.e. then yss (t ) = A | H ( j) | cos[t + + ()] P ( X (t )) = A P ( yss (t )) = A | H ( j) | + ()

C.T. Pan

39

12.5 The Transfer Function and the Steady State Sinusoidal Response
Example : The transfer function H(s) of the circuit given below is known. Find the steady state solution of Vo(t) for the given Vg(t).

C.T. Pan

40

12.5 The Transfer Function and the Steady State Sinusoidal Response

1000( s + 5000) s + 6000s + 25 106 Vg (t ) = 120 cos(5000t + 30 ) V H ( s) =

2
C.T. Pan 41

12.5 The Transfer Function and the Steady State Sinusoidal Response
Solution : Let s = j = j5000 Evaluate H ( j5000) = = Then Voss (t ) = 120 1000( j5000 + 5000) 25106 + j5000(6000) + 25106 2 45 6

2 cos(5000t + 30 45 ) 6 = 20 2 cos(5000t 15 ) V
42

C.T. Pan

12.5 The Transfer Function and the Steady State Sinusoidal Response
nIn theory , the relationship between H(s) and H(jw) provides a link between the time domain and the frequency domain. nIn some cases , we can determine H(jw) experimentally and then construct H(s) from the data.
C.T. Pan 43

12.5 The Transfer Function and the Steady State Sinusoidal Response
Example: Find the impulse response of the following circuit.

R + vin _ C + vo _

C.T. Pan

44

12.5 The Transfer Function and the Steady State Sinusoidal Response
(a) Time domain solution
RC dvo + vo = (t ) dt At t = 0 , vo (0 ) = 0 1 (t ) dt C 0 R 1 = V RC
t

(t )

vo (t )

At t = 0+ , vo (0+ ) =

C.T. Pan

45

12.5 The Transfer Function and the Steady State Sinusoidal Response
For t > 0+ , (t ) = 0
R C + 1 vo (t ) , v(0+ ) = RC _

dvo + vo = 0 dt t 1 RC vo (t ) = e u (t ) RC RC
C.T. Pan 46

12.5 The Transfer Function and the Steady State Sinusoidal Response
(b) s-domain solution Find the transfer function
H ( s) = Vo ( s) Vin ( s)
zero I .C .

Transform into s-domain circuit

R
Vin ( s )
C.T. Pan

1 Cs

+ Vo ( s ) _
47

12.5 The Transfer Function and the Steady State Sinusoidal Response
1 Vo ( s ) 1 H ( s) = = Cs = Vin ( s ) R + 1 1 + RCs Cs -1 h(t ) = L [ H ( s )] =
t 1 RC e u (t ) RC

R
Vin ( s ) 1 Cs

+ Vo ( s ) _

Same answer.

C.T. Pan

48

12.6 The Impulse Function in Circuit Analysis

Example 1: Impulse voltage source excitation

i (t )
VO (t )

i (0 ) = 0

C.T. Pan

49

12.6 The Impulse Function in Circuit Analysis

(a) Time domain solution
At t = 0, i (0 ) = 0 i (0+ ) = 1 t VO ( x) dx L 0 V = O (A) L
VO (t )

i (0+ )

The impulse voltage source has stored energy, 2 1 2 L (i (0 )) , in the inductor as an initial current
+

in an infinitesimal moment.
C.T. Pan 50

12.6 The Impulse Function in Circuit Analysis

For t > 0+ , (t ) = 0
di + Ri = 0 , natural response dt V i (0+ ) = O A L V L i (t ) = O e t / u (t ) , = L R L

i (t )

Note that the impulse source just builds up an initial inductor current but does not contribute to any forced response.
C.T. Pan 51

12.6 The Impulse Function in Circuit Analysis

(b) s-domain solution
VO V /L = O R + sL s + R L V i (t ) = O e t / u (t ) L I ( s) =

R
VO

I (s)

sL

Same answer but much easier.

C.T. Pan

52

12.6 The Impulse Function in Circuit Analysis

Example 2: Impulse current source excitation
I o (t )

+ v(t ) , v(0 ) = 0 _

C.T. Pan

53

12.6 The Impulse Function in Circuit Analysis

(a) Time domain solution
At t = 0, v(0 ) = 0 v(0) = 0 , short circuit
v(0+ ) = 1 t I o ( x) dx C 0 I = o C

I o (t )

The impulse current source has stored energy, 2 1 2 C ( v (0 )) , in the capacitor as an initial voltage
+

in an infinitesimal moment.
C.T. Pan 54

12.6 The Impulse Function in Circuit Analysis

For t > 0+ , (t ) = 0, open circuit
dv v + = 0 , natural response dt R I v(0+ ) = o C I t v(t ) = o e u (t ) , = RC C C

v(t )

Note that the impulse current just builds up an initial capacitor voltage but does not contribute to any forced response.
C.T. Pan 55

12.6 The Impulse Function in Circuit Analysis

(b) s-domain solution Transform into s-domain circuit
Io

R
R

+ V ( s) _

1 Cs = I o / C V( s ) = I o 1 1 R+ s+ Cs RC t I v(t ) = o e u (t ) C
C.T. Pan

Same answer but much easier.

56

12.6 The Impulse Function in Circuit Analysis

Example 3: Impulse caused by switching operation The switch is closed at t=0 in the following circuit.

Note that v1 (0 ) v2 (0 )
C.T. Pan 57

12.6 The Impulse Function in Circuit Analysis

Transform into s-domain
I ( s) = Vo s +1 @ Vo Ce

sC1 sC 2 C1 C 2 Vo Ce Vo C1 Ce = , V 2(s) = = C1 + C 2 sC 2 s C1 + C 2

i(t ) = Vo Ce (t ) v2(t ) =
C.T. Pan

C1 Vo C1 + C 2
58

12.6 The Impulse Function in Circuit Analysis

At t=0 , a finite charge of C1 is transferred to C2 instantaneously. Note that , as the switch is closed , the voltage across C2 does not jump to Vo of C1 but to its final value of the two paralleled capacitors.

C.T. Pan

59

12.6 The Impulse Function in Circuit Analysis

N o te C1C 2 Vo , t > 0+ C1+ C 2 C 12 Q 1 = C 1 V 2 = Vo , t > 0+ C1+ C 2 Q 1+ Q 2 = C 1V o , t > 0 + Q 2 = C 2 V
2

A lso , a t t = 0 , Q 1 = C 1V o , Q 2 = 0 Q 1 + Q 2 = C 1V o C o n serva tio n o f ch a rg e .

C.T. Pan

60

12.6 The Impulse Function in Circuit Analysis

If we consider charged capacitors as voltage sources , then we should not connect two capacitors with unequal voltages in parallel. Due to violation of KVL , an impulse will occur which may damage the components.

C.T. Pan

61

12.6 The Impulse Function in Circuit Analysis

Example 4: Impulse caused by switching operation The switch is opened at t = 0 in the following circuit.
10 i1 3H L1 t =0 i2 L2
2H

15

100V

vo
_

At t = 0 , steady state solution 100V = 10 A 10 i2 (0 ) = 0 A i1 (0 ) =

C.T. Pan

62

12.6 The Impulse Function in Circuit Analysis

For t > 0, the S-domain circuit is
10
I ( s)

3s L1

30

100 s L2

15 Vo ( s )
2s

(100 / s ) + 30 4 2 = + 5s + 25 s s+5 60 10 Vo ( s ) = I ( s ) (2 s + 15) = 12 + + s s+5 5t i (t ) = (4 + 2e ) u (t ) .....................( A) I (s) =

C.T. Pan

v(t ) = 12 (t ) + (60 + 10e 5t ) u (t ) .....( B )

63

12.6 The Impulse Function in Circuit Analysis

10
I ( s)

3s L1

30

100 s L2

15 Vo ( s )
2s

Note t = 0 , iL1 (0 ) = 10 A , iL 2 (0 ) = 0 A t = 0+ , from ( A), iL1 (0+ ) = 6 A , iL 2 (0+ ) = 6 A Also, from ( B ), there exists 12 (t ) at vo (t ).

C.T. Pan

64

12.6 The Impulse Function in Circuit Analysis

Thus, if we consider an inductor current as a current source, then two inductors with unequal currents should not be connected in series. Due to violation of KCL, it will result in impulse voltage which may damage the components.
10 3s

30
15

I (s)
100 s
C.T. Pan

L1

Vo ( s )
L2 2s
65

SUMMARY
Objective 1 : Know the component models in s-domain. Objective 2 : Be able to transform a time domain circuit into the s-domain circuit. Objective 3 : Know how to analyze the s-domain circuit and transform the solution back to time domain.

C.T. Pan

66

SUMMARY
Objective 4 : Understand the significance of transfer function and be able to calculate the transfer function from the s-domain circuit. Objective 5 : Know the geometrical interpretation of convolution integral and be able to calculate the integral.

C.T. Pan

67

SUMMARY
Objective 6 : Know the relation between the phasor solution technique for finding sinusoidal steady state solution and the s-domain solution technique . Objective 7 : Know how to use s-domain solution technique to solve a circuit containing impulse sources or a switching circuit which may result in impulse functions.
C.T. Pan 68

SUMMARY
Chapter problems : 13.13 13.20 13.27 13.36 13.57 13.85 13.88
C.T. Pan 69