Anda di halaman 1dari 8

Lab Notes PH-315

Portland State University A. La Rosa


Transfer function, Laplace transform, Low pass filter
__________________________________________________________
1. INTRODUCTION
2. THE LAPLACE TRANSFORMATION L
3. TRANSFER FUNCTIONS
4. ELECTRICAL SYSTEMS
Analysis of the three basic passive elements R, C and L
Simple lag network (low pass filter)

1. INTRODUCTION
Transfer functions are used to calculate the response C(t) of a system to a given input
signal R(t). Here t stands for the time variable.

Electronic circuit,
r(t ) mechanical system, c(t )
Input signal thermal system
Output signal
Physical system
Fig. 1 Given an input signal, we would like to know the system response C(t).

The dynamic behavior of a physical system are typically described by differential (and/or
integral) equations:
 For a given input signal R(t), these equations need to be solved in order to find C(t).
 Alternatively, instead of trying to find the solution in the time domain, each time-
variable, as well as the differential equations, can be transformed to a different
variable domain in which the solutions can be obtain in a more straightforward way;
then an inverse transform would take place the solution into the time domain

Original problem Difficult solution Solution of


r(t ), original problem
differential Eq. c (t )

Inverse
Transform transform
Problem in
Solution in
transform space
transform space
R(s) Easier solution C(s)
Algebraic operations
Fig. 2 Solving the equations in a different domain and then applying
an inverse transform to obtain the solution in the time domain
One of those transforms is the Laplace transformation

2. THE LAPLACE TRANSFORMATION L


The Laplace transform F=F(s) of a function f = f (t ) is defined by,
L
f  F

L (f) = F


F(s)   f (t ) e- st dt (1)
0

The variable s is a complex number, s = a +j.

f (t)
F F(s)

f  s

t a

Time domain Laplace domain

Example f is the unit step function


 f (t ) e
F(s )  - st dt
0

f =  1  e- st dt
0

1
= (2)
s
t
Time domain Laplace domain
Example A signal response from a system (temperature, for example) respond, after an
external; excitation has stopped, by decaying exponentially. Let’s find out how such decay is
characterized by a Laplace transformation.

f is a decaying exponential f (t )  Ae-  t


 f (t ) e
F(s )  - st dt
0

f 
-t
F(s)   Ae
0
e- st dt
A
= (2)
t s
Time domain Laplace domain

Example. As we mentioned in the introduction, the system response is governed by differential


df d2 f
equations. We would like to know then, how and transform by a Laplace
dt dt 2
df d2 f
transformation. For simplicity, and clarity, let’s use the notation: = f ’ and =f” .
dt dt 2
 If F = L ( f ) evaluate L (f ’)

 f ' (t ) e
L( f ') - st dt
s 
0

 
 f (t ) e
- st  s  f (t ) e - st dt
o 0

  f (0 )  s  f (t ) e- st dt
0

  f (0)  s [ L ( f ) ]
s
  f (0)  s F(s) Laplace transformation of (3)
the derivative
Typically, one proceeds putting the initial conditions equal to zero. (The situation with initial
conditions different than zero are added in a separate simpler procedure). Thus,
L( f ') s  s F(s) Laplace transformation of the derivative (3)’
with the initial conditions equal to zero

 If F = L ( f ) evaluate L (f ”)

 f " (t ) e
L( f ") - st dt
s 
0

  f ' (0)  s f (0)  s2 F(s) Laplace transformation of the (4)


second derivative
Typically, one proceeds putting the initial conditions equal to zero. (The situation with initial
conditions different than zero are added in a separate simpler procedure). Thus,
L( f ") s  s2 F(s) Laplace transformation of the second derivative (4)’
with the initial conditions equal to zero

Example. Sometimes the response signal of a system (the voltage across a capacitor, for
example) must be given in terms of the integral of another quantity (the integral of the
corresponding current across the capacitor). It is convenient, then, to obtain the Laplace
t
transformation of an indefinite integral g (t )  
0
f (u) du
t
 If F = L ( f ) and g (t )  0
f (u) du , evaluate L (g)
 
t
L (g) s   g (t ) e- st dt 
0
[
0
 0
f (u ) du ] e- st dt


t 1 - st  1
 -[ 
0
f (u ) du ]
s
e
0
 [
0
f (t ) ] [ 
s
e- st ] dt

1

s 
0
f (t ) e- st dt

1
 F(s) Laplace transformation of the (5)
s
indefinite integral

3. TRANSFER FUNCTIONS

Differential Eq
r(t ) governing the behavior c(t )
of the system
Input signal Output signal
Fig. 3 Schematic of the system response in the time domain

In a simple system, the output c(t) may be governed by a second order differential equation
a2 c” + a1 c’ + ao c = r (t)
Applying the Laplace transforms (3)’ and (4)’, one obtains
2
( a2 s + a1 s + ao ) C(s) = R (s)

1
C(s)  2
R(s)
a 2 s + a1 s a o

In a more general case, the differential equation may be of higher order (higher than 2). Also
the input may be composed of derivatives of a given function r=r(t). Therefore the factor
1
2
may become a more elaborated function of s.
a 2 s + a1 s a o

Thus, for a system in general,


C(s) = G(s) R(s) (6)

Notice, G=G(s) characterizes the physical system. It is called the transfer function.

It is more typical to write,

C(s)
G(s)  (7)
R(s)

from which, for a given R=R(s) the function C=C(s) can be obtained.

R(s ) G(s) C(s )


Input signal Output signal

Fig. 4 Schematic of the system response in the Laplace domain

2s  3
Example. A system is characterized by the transfer function G(s)  .
(s  1) (s  6)
Find out how the system respond to a exponentially decaying input r(t )  e- 2t .
Answer:
1
The Laplace transformation of r gives, using expression (2), R(s) =
s2

2s  3 1
The output signal, in the Laplace domain , is then given by, C(s) 
(s  1) (s  6) s  2

In a typical procedure, when possible, C(s) is re-written in the following form:


K1 K K
C(s)   2  3 , with K1, K2, K3, to be determined.
s 1 s 6 s 2

Notice, K1  C(s)(s  1) 1  0.8

K 2  C(s)(s  6) 6  - 0.3

K 3  C(s)(s  2) 2  - 0.5

Thus,

0.8 - 0.3 - 0.5


C(s)   
s 1 s 6 s 2

Now, using (2) we identify the time dependent functions these individual Laplace
transforms come from,

c( t )  0.8 e- t  0.3 e- 6t  0.5e- 2t Answer.

Recapitulating the process,

c( t )  0.8 e - t  0.3 e - 6t
r (t )  e - 2 t
 0.5e - 2t
Difficult solution
Solution of original
Original problem System problem
r(t ) Differential Eq c (t )
Integral Eq.

Laplace Inverse
Transform transform

G(s)
Problem in + algebraic operations Solution in Laplace
Laplace space space
R(s) C(s)
2s  3
1 G(s) 
(s  1) (s  6) 0.8 - 0.3 - 0.5
R(s) =
s2 C(s)   
s 1 s 6 s 2

Fig. 5 Schematic representation of the solution procedure in the previous example.

In the previous example, the transfer function was given. In the next section we will figure out
the transfer function for the case of electrical systems.
4. ELECTRICAL SYSTEMS

 Let’s analyze the three basic elements R, C and L individually

Let I  I(s) be the Laplace transform of i= i (t).


vc vL vR
C L
i R

Fig. 6 Elementary passive circuit elements

Capacitor
q( t )
vc (t) =
C
1 t using (5) 1 t 1 1
=
C  i (u) du
0
 Vc (s) =
C  i (u) du 
0 C s
I(s )

1
Vc (s)  I(s) (8)
Cs

Inductor
d i (t ) using (3)'
vL (t) = L  VL (s)  L s I(s) (9)
dt

Resistor
vR (t) = R i (t ) VR (s)  R I(s) (10)

Time domain Laplace domain

 Analysis of a simple lag network

R
vin (t) vout (t)
q
i C

Fig. 7 Low pass filter in the time domain.


q( t )
vin (t) = R i (t ) +
C
1 t using (5) 1
C 0
= R i (t ) + i (u) du  Vin (s)  R I(s)  I(s )
Cs
 1 
 R   I(s) (11)
 Cs 
On the other hand
q( t )
vout (t) 
C
1 t i (u ) du using (5) 1

C 0
  Vout (s) 
Cs
I(s) (12)

From (11) and (12)


Vout 1/Cs

Vin R  1/Cs

Vout 1
G(s)   (13)
Vin RCs  1

1
Vin (s ) G(s)  Vout (s )
RCs  1
Input signal Output signal

Fig. 8 Low pass filter in the Laplace domain.