TIME RESPONSE

In this chapter you will learn the following:

How to find time response from the transfer function

How to use poles and zeros to determine the response of a control system

How to describe the transient response of first- and second-order systems

How to view the effects of nonlinearities on the system time response

How to find time response from the state-space representation
This lecture is devoted to the analysis of system transient response. First we must
describe some valuable analysis and design tools, poles, zeros and system response.
The poles of a transfer function are (1) the values of the Laplace transform variable, s,
that cause the transfer function to become infinite or (2) any roots of the denominator of
the transfer function that are common roots of the numerator.
The zeros of a transfer funtion are (1) the values of the Laplace transform variable, s,
that cause the transfer function become zero, or (2) any roots of the numerator of the
transfer function that are common roots of the denominator.
POLES AND ZEROS OF A FIRST ORDER SYSTEM : AN EXAMPLE
Given the transfer function G(s) in Figure (a), a pole exists at s=-5 and a zero exists at -
2. these values are plotted on the complex as plane in Figure(b) using an x for the pole
and a ο for the zero. Two show the properties of the poles and zeros, let us find the unit
step response of the system. Multiplying the transfer function of Figure(a) by a step
function yields C(s) = s + 2 = A + B = 2 / 5 + 3 / 5 where A = s + 2 = 2 and B = s + 2 = 3
s 5 s 5
s(s + 5) s s +5 s s +5 s →0 s → −5

2 3 − 5t
Thus, c (t ) = + e
5 5
All of these are summarized in the follwing
figure(c) and we draw the following
conclusions :
 Conclusions :
1.) A pole of the input function generates
the form of the forced response (that is,
the pole at the origin generated a step
function at the output).
2.) A pole of the transfer function
generates the form of the natural
response (that is, the pole at -5
generated e-5t).
3.) A pole on the real axis generates an
exponential response of the form e-αt
where –α is the pole location on the real axis. Thus, the farther to the left a pole is on the
negative real axis, the faster the exponential transient response will decay to zero
(again, the pole at -5 generated e-5t)
4.) The zeros and poles generate the amplitudes for both the forced and natural
response

Let us look at an examplethat demonstrates the technique of using poles to obtain the
form of the system response
Example : A system has a transfer function G ( s ) = 10 ( s + 4 )( s + 6 )
( s + 1)( s + 7 )( s + 8 )( s + 10 )
Write, by inspection, the output, c(t), in general terms if the input is a unit step.
Answer : c(t) = A + Be-t + Ce-7t + De-8t + E-10t
By now, we learned that poles determine the nature of the time response: Poles of the
input function determine the form of the forced response, and poles of the transfer
function determine the the form of the natural response. Zeros and poles of input or
transfer function contribute to the amplitudes of the components parts of the total
response. Finally, poles on yhe real axis generate exponentials responses.
FIRST-ORDER SYSTEMS
We now discuss first-order systems without zeros to define a performance specification
for such a system. A first-order without zeros can be described by the transfer funtion
given in the figure(a). If the input is a unit step, where R(s) = 1/s, the Laplace transform
of the step response is C(s), where C(s) = R(s)G(s) = a . Taking the inverse Laplace
s(s + a)
transform, the step response is given by
c(t ) = c f (t ) + cn (t ) = 1− e −at

where the input pole at the origin generated the

forced response cf(t)=1, and the system pole at
–a, as shown in Figure(b).
 c(t ) = c f (t ) + cn (t ) = 1− e −at
(*)
Let us examine the significance of parameter a, the only parameter needed to describe
the transient response. When t=1/a,
e − at = e − 1 = 0 . 37 (**) or c ( t ) t = 1 / a = 1 − e − at = 1 − 0 . 37 = 0 . 63 (***)
t =1 / a t =1 / a

We now use these equations to define three transient response specifications.

Time Constant : We call 1/a the time constant of the response. From equation (*), the
time constant can be described as the time for e-at to decay to 37% of its initial time.
Alternately, from equation (**), the time constant is the time it takes for the step
response to rise to 63% of its final value. Thus, we can call the parameter a as
exponential frequency.
Rise Time, Tr : Rise time is defined as
the time for the waveform to go from 0.1
to 0.9 of its final value. Rise time is found
by solving equation (*) and solving for
time at c(t)=0.9 and c(t)=0.1. Hence
2.31 0.11 2.2
Tr = − =
a a a
Settling Time, Ts : Settling time is
defined as the time for the response to
reach and stay within 2% of its final
value. Letting c(t)=0.98 in equation (*)
and solving for t, we find the settling time
to be 4
T =s
a
 SECOND-ORDER SYSTEMS : INTRODUCTION
Let us now extend the concepts of poles, zeros and transient response to second order
systems. Compared to the simplicity of a first ordersystem, a second order system
exhibits a wide range of responses that
must be analyzed and described. To
become familiar with the wide range of
responses before formalizing our
discussion, we take a look at numerical
examples of the second order system
responses shown in the figure. All
examples are derived from Figure(a), the
general case, which has two finite poles
and no zeros. By assigning appropriate
values to parameters a and b, we can
show all possible second,order transient
responses. We now explain each
response and show how we can use the
poles to determine the nature of the
response without going through the
procedure of a partial-fraction expansion
followed by the inverse Laplace
transform.
Overdamped Response : For this response, 9 9
C ( s) = =
s ( s 2 + 9s + 9) s( s + 7.854)(s + 1.146)

This function has a pole at the origin

that comes from the unit step input
and two real poles that come from
the system. The input pole at the
origin generates the constant forced
response; each of two system poles on the real axis generates an exponential natural
frequency. Hence the output could have been written as c(t) = K1 + K2e-7.854t + K3e-1.146t .
This response shown in figure called overdamped. We see that the poles tell us the
form of the response without the tedious calculation of the inverse Laplace transform.
9
Underdamped Response : For this response , C ( s) =
s(s 2 + 2s + 9)

This function has a pole at the

origin that comes from the unit step
and two complex poles that come
from the system. The poles that
generate the natural response are
at s = -1±j√8. The real part of the pole matches the exponential decay frequency os
sinusoids’s amplitude, while the imaginary part of the pole matches the frequecy of the
snisoidal oscillation. The sinusoidal frequency is given the name of damped frequency
of oscillation, ωd. This response shown in figure called underdamped.
 9
Undamped Response : For this response, C (s) =
s(s 2
+ 9)

This function has a pole at the origin and

two imaginary poles. The pole at the
origin generates the constant forced
response, and the two system poles on
the imaginary axis at ±j3 generate a
Sinusoidal natural response. Hence, the output can be estimated as c(t)=K1+K4cos(3t-φ)
This type of response is called undamped.
9 9
Critically Damped Response : For this response, C ( s) = =
s ( s + 6s + 9) s ( s + 3) 2
2

This function has a pole at the origin

and two multiple real poles. The input
pole at the origin generates the constant
forced response, and two poles at the
real axis at -3 generate a natural

exponential response. Hence the output can be estimated as c(t) = K1 + K2e-3t + K3te-3t

We now summarize our observations. In this section we defined the following natural
response and found their characteristics.
1. Overdamped responses : Poles : Two real at –σ1, -σ2
Natural reponse : Two exponentials with time constant equal to the reciprocal of the
pole locations, or c ( t ) = K 1 e − σ 1t + K 2 e − σ 2 t
2. Underdamped responses : Poles : Two complex at -σd±jωd
Natural response : Damped snisoidal with an exponential envelope, or
c ( t ) = Ae − σ d t cos( ω d t − φ )

3. Undamped responses : Poles : Two imaginary at ±jω1

Natural response : Undamped snisoid, or c(t ) = A cos(ω1t − φ )
4. Critically damped responses : Poles : two real at –σ1
Natural response : Two exponential term, or c ( t ) = K 1 e − σ 1 t + K 2 te − σ 1 t

Figure 4.10
Step responses
for second-order
system
damping cases
 THE GENERAL SECOND-ORDER SYSTEM
In this section we will define two physically meaningful specificaitons for second-order
system.
Natural Frequency, ωn : The natural frequency of a second order system is the
frequency of oscillation of the system without damping. For example, the frequency of
oscillation of a series RLC circuit with the resistance shorted would be natural frequency.
Damping Ratio, ζ : We define the damping ratio, ζ, to be

Exponential decay frequency 1 Natural period (second)

ζ= =
Natural frequency (rad/sec) 2π Exponential time constant

Let us now revise our description of the second-order system to reflect the new
definitions. A general second-order system can be transformed to show the quantities
ζ and ωn. Consider the general system
b
G (s) =
s 2 + as + b
Without damping, the poles would be on the jω axis, and the response would be an
undamped sinusoid. For the poles to be purely imaginary, a=0. Hence G ( s ) = s b+ b 2

By definition, the natural frequency, ωn, is the frequency of oscillation of this system.
Since the poles of the system are on jω axis at ±jω√b, ωn=√b and b= ωn2. Now, what is
the value of a?
Assuming the underdamped system, the complex poles have a real part, σ, equal to
-a/2. The magnitude of this value is then exponential decay frequency. Hence,
Exponential decay frequency |σ| a/2
ζ = = =
Natural frequency(rad/second) ωn ωn
from which a = 2ζωn. Our general second order transfer function finally looks like this:
ωn2
G (s) = 2 2
s + 2ζω n s + ω n

In the following example we find numerical values of ζ and ωn by matching the transfer
function to final equation we found above
36
Example : Given the transfer function, find ζ and ωn . G( s ) =
s 2 + 4.2s + 36
Solution : Note that ωn2=36 and ωn=6. Also, 2ζωn=42 and ζ=0.35
We have defined ζ and ωn , let us relate these quantities to the pole
location. Solving for the poles of the transfer function in equation
ωn2
G (s) = 2 2
s + 2ζω n s + ω n
yields 2
s 1, 2 = −ζω n ± ω n ζ −1
 2
s 1, 2 = −ζω n ± ω n ζ −1

From the equation above, we see that the various cases of second-order response are
a function of ζ ; they are summarized in the table below.

Figure 4.11
Second-order
response as a
function of
damping ratio
In the following example we will find the numerical value of ζ and determine the nature of
transient response
Example : For each of the systems shown in the figure, find the value of ζ and report the
kind of response expected.

Solution : First match the form of these systems to the forms shown in equations
b ωn2
G (s) = 2 and G (s) = 2 2
s + as + b s + 2ζω n s + ω n

Since a = 2ζωn and ωn=√b → ζ =a/(2√b). Using the values of a and b from each of the
systems of the figure, we find ζ=1.155 for system(a), which is overdamped, since ζ>1;
ζ =1 for system(b), which is thus critically damped; and ζ=0.894 for system(c), which is
thus underdamped, since ζ<1.
 Underdamped Second Order Systems
In addition to ζ and ωn, more specifications indigenous to the underdamped case will be
defined. These specifications are defined as follows:
1. Rise time, Tr : the time required
for the waveform to go from 0.1 of
the final value to 0.9 of the final
value.
2. Peak time, Tp : The time required
to reach the first, or maximum,
peak.
3. Percent overshoot, %OS : the
amount that the waveform
overshoots the steady-state, or
final, value at the peak time,
expressed as a percentage of the
steady-state value.

Figure 4.14 4. Settling time, Ts : The time

Second-order required for the transient’s
underdamped damped ascillation to reach and
response stay within ±2% of the steady-state
specifications value.
Evaluation of Tp, %OS and Ts : To find the value of these parameters, we use the
equations of
π − ζπ / 1−ζ 2  4
Tp =   Ts =
ωn 1− ζ 2 %OS = e ×100 ζω n

We can use the graphics in figure 4.16 to obtain the value of rise time, Tr .

Figure 4.16
Normalized rise
time vs. damping
ratio for a
second-order
underdamped
response

Let us look an example

 100
Example : Given the transfer function G ( s ) = , find Tp, %OS, Ts and Tr .
s 2 + 15 s + 100
Solution : ωn=10 and ζ=0.74. Using the related equation, Tp=0.475 second,
%OS=2.838, Ts=0.533 . Using the related graphic, normalised rise time is
approximately 2.3 second. Dividing by ωn yields Tr=0.23 second.

We now have expressions that relate peak time, percent overshoot and settling time to
the natural frequency nad the damping ratio. Now let us relate these quantities to the
location of poles that generate these characteristics.
The pole plot for a general underdamped second-order system is in figure below. We
see from the Pythagorean theorem that the radial distance from the origin to the pole is
the natural frequency, ωn, and the cosθ
= ζ. Now comparing the equations we
have evaluated formerly with the pole
location, we evaluate peak time and
settling time in terms of pole location :
π π 4 4
Tp = = Ts = =
ωn 1− ζ 2 ωd ζωn σd

where ωd is the imaginary part of the

pole and is called the damped
frequency of oscillation, and σd is the
magnitude of the real part of pole and is
exponential damping frequency
Example : Given the pole plot in figure, find ζ, ωn, Tp, %OS and Ts .

Solution : The damping ratio is given by

ζ=cosθ=cos[arctan(7/3)]=0.394. The natural
frequency, ωn, is the radial distance from the
origin to the pole, or ωn=√(72+32)=7.616 . The
peak time is
π π
Tp = = = 0.449 second
ωd 7

The percent overshoot is

−  ζπ / 1− ζ 2 
 
0
0 OS = e × 100 = 26 0
0

The approximate settling time is

4 4
Ts = = = 1.333 second
σd 3
Example : Given the system shown in figure, find J and D to yield 20% overshoot and a
settling time of 2 second for a step input of torque T(t).
Solution : First, the transfer function for the
system is G (s) =
1/ J
D K 2
s + s +
J J
K D
From the transfer function, ωn = and 2ζω n =
J J
4 D
But, from the problem statement, Ts = 2 = or ζω n = 2 . Hence 2ζωn = 4 = .
ζωn J
4 J − ζπ / 1−ζ 2 
 
Also, ζ = =2 . We know the formulation for overshoot, which is 0
0 OS = e ×100
2ω n K

If we draw ζ from this equation, we find that a 20% overshoot implies ζ=0.456. Therefore
4 J J
ζ = =2 ⇒ = 0.052
2ω n K K

From the problem statement, K = 5 N-m/rad. Combining this value with equations
D J
2ζωn = 4 = and = 0 .052 which we have found above, we get
J K

D = 1.04 N-m-s/rad and J = 0.26kg-m2

SYSTEM RESPONSE WITH ADDITIONAL POLES AND ZEROS
By now, we analyzed systems with one or two poles and with one zeros. If a system has
more than two poles or zeros we can not use the formulation we derived for calculation
of performance specifications. But, under certain conditions, a system with more than
two poles or zeros can be approximated as a second order system that has just two
dominant poles. If the real pole is five times farther to the left than two complex-
conjugate pole pair, we can say this complex-conjugate pole pair is dominant pole pair.
A way to look at the effect of an additional zerois as follows. Let C(s) be the response of
a system, T(s), with unity in the numerator. If we add a zero to the transfer function,
yielding (s+a)T(s), the Laplace transform of the response will be
(s+a)C(s) = sC(s) + aC(s)
Thus, a response of a system with a zero consists of two parts : The derivative of the
original response and a scaled version of the original response. If a is negative, an
interesting phenomenon occurs. The output may be opposite sign of input as shown in
figure. Along first 0.6 second, the output has
negative sign although the input has positive
sign (step input). A system that exhibits this
phenomenon is known as a nonminimum
phase system. If an airplane was a
nonminimum phase system, it would initially
veer left when commanded to steer right.
 EFFECT OF NONLINEARITIES UPON TIME RESPONSE
In this section we will insert nonlinearities, such as saturation, dead zone and backlash
into a system to show the effects of these nonlinearities upon the linear responses. The
responses were obtained using Simulink.
Let us assume the motor and load from a Antenna Position Control System. The input of
the system is applied voltage to motor, and the output of the system is load angular
velocity. Let us add a saturation(±5 V) to system.

As shown in figure, the effects of saturation is to limit the obtained velocity.

Let us examine the effects of deadzone nonlinearity.
Add a deadzone (-2 to +2) to system and compare the system responses with and
without the deadzone nonlinearity. Note that the input is sinusoidal.

Note that, the response begins when the input voltage to the motor exceeds a
threshold.

Let us examine the effects of backlash nonlinearity.

Add a backlash to system and compare the system responses with and without the
backlash nonlinearity. Note that the input is sinusoidal.

As the motor reverses direction, the output shaft remains stationary while the motor
begins to reverse. The resulting response is quite different from the linear response
without backlash.
 LAPLACE TRANSFORM SOLUTION OF STATE EQUATIONS
In this section, we will use the Laplace transform to solve the state equations for the
state and output vectors. Consider the state and output equations
•
x = Ax + Bu
y = Cx + Du
Taking the Laplace transform of both sides of the state equations yields
sX (s) − x(0) = AX (s) + BU (s)
In order to seperate X(s), replace sX(s) with sIX(s), where I is an n*n identity matrix, and
n is the order of the system. Combining all of the X(s) terms, we get
(sI − A) X (s) = x(0) + BU(s)
Solving for X(s) by premultiplying both sides of the last equation by (sI-A)-1 yields

X ( s) = ( sI − A) −1 x(0) + ( sI − A) −1 BU ( s )
adj ( sI − A)
= [ x(0) + BU ( s )]
det( sI − A)

Tkanig the Laplace transform of the output equation yields

Y (s) = CX(s) + DU(s)

Eigenvalues and Transfer Function Poles : We saw that the poles of the transfer
function determine the nature of the transient response of the system. Is there an
equivalent quantity in the state-space representation that yields the same information?
The roots of det(sI-A)=0 is the eigenvalues of the system matrix A. Let us show that the
eigenvalues are equal to the poles of the system’s transfer function. Let the output, Y(s),
and the input, U(s), be scalar quantities Y(s) and U(s), respectively. Further, conform the
definition of a transfer function, let x(0), the initial state vector, equal 0, the null vector.
Substituting the equation −1 −1
X ( s) = ( sI − A) x(0) + ( sI − A) BU ( s )
adj ( sI − A)
= [ x(0) + BU ( s )]
det( sI − A)

into equation Y ( s ) = CX ( s ) + DU ( s ) and solving for the transfer function Y(s)/U(s) yields
Y (s)  adj ( sI − A) 
= C   B + D
U ( s)  det( sI − A) 
Cadj ( sI − A) B + D det( sI − A)
=
det( sI − A)

The roots of the denominator of this equation are the poles of the system. Since, the
roots of X(s) and Y(s)/U(s) are identical, the system poles equal the eigenvaluses.
Hence, if a system is represented in state-space, we can find the poles from det(sI-A)=0.
The following example demonstrates solving the state equation using the Laplace
transform as well as finding the eigenvalues and system poles.
Example : Given the system represented in state,space by equations
 0 1 0  0  1 
•

x =  0 0 1  x +  0  e − t
 , x(0 ) =  0 
 − 24 − 26 − 9   1   2 
y = [1 1 0 ]x

do the following :
a. Solve the preceding state equation and obtain the output for the given input.
b. Find the eigenvalues and the system poles.
Solution : a) Remember the equation X (s) = (sI − A)−1 x(0) + (sI − A)−1 BU(s)
(s 2 + 9s + 26) s +9 1
 2 
 − 24 s + 9s s
 s −1 0   − 24s
   − 26s + 24 s 2  1
(sI − A) =  0 s − 1  ⇒ (sI − A) =
−1 , U (s) =
s
we get
s 3 + 9s 2 + 26s + 24
24 26 s + 9

( s 3 + 10 s 2 + 37 s + 29 )  X 1 (s) 
( s 3 + 12 s 2 + 16 s + 5)
Y ( s ) = [1 1 0 ] X 2 ( s )  = X 1 ( s ) + X 2 ( s ) ⇒ Y ( s ) =
X1 = − 6 .5 19 11 .5
( s + 1)( s + 2 )( s + 3 )( s + 4 ) = + −
( s + 1)( s + 2)( s + 3)( s + 4) s + 2 s + 3 s + 4
 X 3 ( s ) 
( 2 s 2 − 21 s − 24 )
X2 = where a pole at - 1 canceled a zero at - 1. Taking the inverse Laplace transform :
( s + 1)( s + 2 )( s + 3 )( s + 4 )
s ( 2 s 2 − 21 s − 24 ) y(t) = -6.5e - 2t + 19 e −3t − 11 .5e − 4 t
X3 =
( s + 1)( s + 2 )( s + 3 )( s + 4 )

b) The roots of det(sI-A)=0 give us both the poles of the system and the eigenvalues
which are -2, -3 and -4.
 TIME DOMAIN SOLUTION OF STATE EQUATIONS
We now look at another technique for solving the state equations. Rather than using the
Laplace transform, we solve the equations directly in the time domain using a method
closely allied to classical solution of differential equaitons. We will find that the final
solution consist of two parts that are different from the forced and natural responses.
The solution in the time domain is given directly by
t
x(t ) = e x(0) + ∫ e A(t −τ ) Bu (τ ) dτ
At

0
t
= Φ (t ) x(0) + ∫ Φ (t − τ ) Bu (τ ) d (τ )
0

where Φ(t)=eAt by definition, and which is called the state-transition matrix.

Notice that the first termon the right hand side of the equation is the response due to the
initial state vector x(0). Notice also that it is the only term dependent on the initial state
vector and not the input. We call this part of the response the zero-input response,
since it is the total response if the input is zero. The second term, called the convolution
integral, is dependent only on the input, u, and the input matrix, B, not the initial state
vector. We call this part of response the zero-state response, since it is total response
if the initial state vector is zero.
 Let us examine the form the elements of Φ(t) take for linear, time invariant systems.
The equation X (s) = (sI − A)−1 x(0) + (sI − A)−1 BU(s) , is the transform of Φ(t)x(0), zero input
response. Thus, for the unforced system, take the Laplace transform of x(t) yields
X(s)=(sI-A)-1x(0)
From which we can see that (sI-A)-1 is the Laplace transform of the state-transition
matrix. We have already seen that the denominator of (sI-A)-1 is a polynomial in s
whose roots are the system poles. This polynomial is found from the equation det(sI-
A)=0. Since,
 adj ( sI − A) 
[ ]
L−1 ( sI − A) −1 = L−1   = Φ (t )
 det( sI − A) 
where L-1 represents the inverse Laplace transform operator.
Example : For the state equation and initial state vector shown in the equations, where
u(t) is a unit step, find the state transition matrix and then solve for x(t).
• 0 1 0 1
x=  x (t ) + 1u(t ) , x(0) =  
 − 8 − 6    0
Solution : Find the eigenvalues using det(sI-A)=0. s2+6s+8=0, from which s1=-2 and
s2=-4. Since each term of the state-transition matrix is the sum of responses generated
by the poles ( or eigenvalues), we assume a state-transition matrix of the form
( K 1e −2t + K 2 e −4t ) ( K 3 e −2t + K 4 e −4t ) 
Φ (t ) =  − 2t − 4t − 2t − 4t 
K
 5 e + K 6 e ) ( K 7 e + K 8 e )
With Φ(t)=0, we get 8 simultaneous equations to find the values of the constants Ki’s.
After the values of all constants are calculated, we get
 1 − 2t 1 − 4t   1 −2(t −τ ) 1 −4(t −τ ) 
 ⇒ Φ(t − τ ) B =  2 e 
−2 t −4 t
( 2e + e ) ( e + e ) − e
Φ(t ) =  2 2 2 
(−4e −2t + 4e −4t ) (−e −2t + 2e −4t )   −2(t −τ )
   e ( + 2e )
−4( t −τ )

 ( 2 e −2 t − e −4 t ) 
Hence the first term of the state transition matrix will be Φ (t ) x (0) =  − 2t − 4t 
( −4e + 4e ) 
 1 1 −2t 1 −4t 
t
8 − 4 e + 8 e 
and the second term will be ∫ Φ(t − τ ) Bu(t )dτ =  1 −2t 1 −4t 
0  e − e 
 2 2 

 1 7 −2t 7 −4t 
t
8 + 4 e − 8 e 
The final result is found as x (t ) = Φ (t ) x (0) + ∫ Φ (t − τ ) Bu (τ ) dτ = 
7 − 2t 7 −4t 
0  − e + e 
 2 2 