Chapter 4 (page#168)

Continuous-Time
System Responses

4.4 Response of Second-Order Systems

2.7 Higher- Order System Response

2.5 Stability Testing

System Response

• Ist & IInd order systems if stable:

–the forced response is the steady-state

response (dc steady state)

–and the natural response is the transient response

The zero-input response of a circuit is the state of the circuit

when there is no forcing function (no current input, and no voltage
input).

• For unstable responses, "steady-state" and

"transient" are meaningless
Example: RLC Circuit with DC input
di 1
Vs  L  Ri   idt
dt C
Differentiate and divide both sides by L
Vs
Vs is a DC input; if we take its derivative, what
do we get?

d2iR di i
0 2  
2
R R 1
dt L dt LC s1      
2L  2L  LC
 2 R 1 

I S  S    0 2
 L LC  R R 1
s2      
2L  2L  LC
R 1
S2  S  0; characteristic equation
L LC
Further Simplification of Characteristics Equation

2  1  R 2 
R R 1 
R
    
s1,2      s1,2
2L  2L  LC 2L  LC  2L  

R 1  R 
2
example ( s 2  4)  0
s1, 2  j  
2L LC  2 L  s   4  j 4  2 j
1
   
R 1 2
Let   ; &
2L LC LC

s1, 2  n  j   (n ) 2

n
2
Mechanical Example of IInd order

F(s)  MS2 X  B S X  K X
B K

X(s) 1
  m / sec Mass (M)
F(s) MS 2  BS  K N

x(t)
1 f(t)
X(s) M
  m / sec
F(s) B K N
S  S
2
M M

B K
S2  S   0 characteri stic equation
M M

2
B  B  K
s1,2     
2M  2M  M
 Forcing Forcing
Time  and n
constant () Function Function

s1,2  Cn(  K2 /  1 n2 1

s) 
j 1 C(s) 
 >11  s =1 s
<1
2
 2 s  
n =0 n
2 s
s  

 Distinct Equal Complex Imaginary
real roots roots & real conjugate roots
roots X X
X X X
X
X X
Over Critically Under Un-Damped
Damped Damped Damped
s1, 2  n  jn 1   2
2.3 (p 126) RESPONSE OF SECOND-ORDER SYSTEMS

Mentioned earlier the Standard form of the II-order TF is given by:

C(s) n2
G(s)   2 ( 2.3)
R(s) s  2 n s  n2

, the damping ratio, will determine how much the system

oscillates as the response decays toward steady state.

n, the undamped natural frequency (natural frequency) , will

determine how fast the system oscillates during any transient
response.

Note: All system chracteristics of the 2nd-order system are

functions of only  and n.
2.3.6(page=136) Tr, %OS, and Ts

1. Peak time, Tp: The time required to reach the first, or

maximum, peak.

2. Percentage 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.

3. Settling time, Ts: The time required for the transient's damped
oscillations to reach and stay within 2% of the steady-state value.

4. 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.

Rise time, settling time, and peak time yield information about
the speed of the transient response.
2.3.6(page=136) Tr, %OS, and Ts

Tp
overshoot M p  
Tp  
(1) n 1   2 d
c()

ess  Steady state error  

Tr ( 2)

Td

Ts

Tr
t
Step Response
2.3.6(page=136) Tr, %OS, and Ts
overshoot M p
tp
Percentage overshoot is related (1) c()

to the damping ratio via

 ln(%OS / 100 ) tr ( 2)

 td
  ln (%OS / 100 )
2 2
ts

tr
t

 
Tp  
n 1   2 d
2.3.6(page=136) Tr, %OS, and Ts
Problem: Given the system, find J and B to yield 20% overshoot (%OS) and a
settling time (Ts) of 2 seconds for a step input of torque (t). 
τ
d 2 ( t ) d( t )
( t )  J B  K (s)  JS 2  BS  K J
dt dt
 1
Solution : 1. The transfer function is G( s )   2 B
 JS  BS  K K

B K
2. Fromthe transferfunction s2  2ns  n2  S 2  S 
J J

K B
n  and 2 n s  s
J J
4
3. From the problem statement Ts  2  or  n  2
 n

B
4. Thus, 2 n  4 
J
5. For 20% overshoot, the damping ratio () is given by equation

 ln(%OS / 100 )
  0.456 (check answer )
 2  ln2 (%OS / 100 )

4 J
From previous eqation 2 n  4    2  0.456
2n K

6. For K=5 N-m/rad,

B=1.04 N-m-s/rad and
J=0.26 kg-m2
2.3.6(page=136) Tr, %OS, and Ts

C(t)
Exponential decay jd
S-Plan
generated by real
part of the complex -=-n n
pole pair (=n)
-jd

1  n t
c( t )  1  e sin(n t  )


Sinusoidal oscillation
generated by imaginary
part of the complex pole
pair
t
Exponential decay generated by real
 
part of the complex pole pair Tp  
n 1   2 d
2.3 (p 126) RESPONSE OF SECOND-ORDER SYSTEMS…..cont

C(s) n2
G(s)   2
R(s) s  2 n s  n2

C(0)
G(0)   1 (dcgain is unity )
R(0)

G(0): DC gain of the system, will determine the size of steady

state response when the input settles out to a constant value.

We will also consider 2nd-order systems with other than unity

gain and with numerator other than a constant.
  
Tp  
n 1   2 d
2.3 (p 126) RESPONSE OF SECOND-ORDER SYSTEMS…..cont

Unit-step response for standard Second-0rder system

G(s) n2
C(s)  G(s) R(s) R(s)
s 2  2 n s  n2
C(s)

1
C( s)  G( s)
s
Step response R(s) = A/s = 1/s unit step response
n2 1
C(s) 
s 2  2 n s  n2 s
Assume poles of G(s) are complex, taking inverse LLT

1  nt
c( t )  1  e sin(n t  ) ( 2.4) Time constatnt ()  1
 n

SettlingTi me(Ts )  4  4
  cos  n
  1  2   tan 1   
   n  frequency of damped
sin usoide
Note:-

1. The impulse response of any system does give an indication of the nature of the
initial-condition (IC) response, and thus the transient response, of the system.

2. An Initial condition on a first-order system can be modeled as an impulse function

input
3. An Initial condition excitation of a Higher-order system cannot be modeled as simply
as that of the first-order system.
4. The unit-impulse response of the II-order system is given by the equation:
  2 
c( t ) ₤-1  n 
 s2  2 n  2 
 n 
1  nt
c(t )  1  e sin(n t  ).....Unit step response ( 2.4)

n  nt
c(t )  e sinn t.....impulse response

Equation impulse response is the derivative of equation 2.4
5. The impulse response of the second order system can also be considered to be the
response to certain initial conditions, with r(t)=0 i.e (R(s)=0)
0<<1 =0 -1<<0

=1 =-1

0<<1 =0 -1<<0

1.12.2 Response Terms
(From previous chapter)
Figure 1.30
Y(s) 1
T 
R(s) s

1
T
(s  aj)(s  bj)
m arg inally stable
undamped

1
T stable
( s  a)
decay exp onentially
ke at
Fig 1.30 Laplace-transform denominator root locations and corresponding time functions
1.12.2 Response Terms
(From previous chapter)

1
T
s(s  a)
un  stable

1
T
(s  a  jb)(s  a  jb)
stable  underdampe d
 1
T
(s  a  jb )(s  a  jb )
un  stable

1
T
(s  a)(s  b)
un  stable
1.12.2 Response Terms
(From previous chapter)
1
T
(s  aj)2 (s  bj)2
un  stable

1
T
(s  a) 2
stable
critically damped

1
T
( s  a) 2
un  stable
 =0
2.3.5 Un-damped 
Imaginary Roots


 0<<1
2.3.4 Under damped
Complex Roots



 =1
2.3.3 Critically damped
Real Repeated-Roots


 >1
2.3.2 Over damped
Real distinct roots

 
• The zero-input response of a circuit is the state of the circuit when there is
no forcing function (no current input, and no voltage input). Spring Mass Animations

 
Tp  
n 1   2 d
 Using Simulink to plot poles and zeros and see its effects
Self: Simulate in Matlab and change the system to stable & unstable - see its effects –
also place zeros to the right side and see its effects
Impulse Response of a Second Order system:

wn=1;
zeta=[0.2 0.5 0.7 1 2 5];
for k=1:6
num=[0 0 wn^2];
den=[1 2*zeta(k) wn^2];
sys=tf(num,den)
impulse(sys);
hold on
end
Step Response of a Second Order system:

wn=1;
zeta=[0.2 0.5 0.7 1 2 5];
for k=1:6
num=[0 0 wn^2];
den=[1 2*zeta(k) wn^2];
sys=tf(num,den)
step(sys);
hold on
end
Get equations of the above and try with this
for few zeta
Dr. Shahryar 35
Responses of generalised 2nd order
system with ω and ζ
• First solve for poles

• We get

Dr. Shahryar 36
• For each of the following transfer
functions, write by inspection, the general
form of the step response

Dr. Shahryar 37
Dr. Shahryar 38
  and n

>1 <1
s    jn
s1  s 2
1  n t
c( t )  1  k1e  t / 1
 k 2e t / 2 c( t )  1  e sin( n t  )


=1 =0
s1  s 2  s
c( t )  1  k1e  t / 1  k 2 te  t / 2
Case 1: Roots Are Complex

Over damped response:

Poles: two real poles at S1 and S2.
Natural response: two exponentials
Time constants: reciprocal of the pole locations
Case 2: Roots Are Complex

t = 0:0.01:15;
m=cos(2.828*t)+0.35*sin(2.828*t);
cc=1-exp(-t);
c=cc.*m; %Check wrong
plot(t,c)

Under damped response:

Poles: two complex poles at S1, 2  n  jn 1  
2

Natural response: damped sinusoid with exponential envelope.

Time constants of envelope: reciprocal of the pole real part
Case 3: Roots Are Complex

t = 0:0.01:3;
c=1-3*t.*exp(-3*t)-exp(-3*t);
plot(t,c)

0.9

0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

0
0 0.5 1 1.5 2 2.5 3
Case 4: Roots Are Complex

t = 0:0.01:30;
c=1-cos(-2*t);
plot(t,c)

1.8

1.6

1.4

1.2

0.8

0.6

0.4

0.2

0
0 5 10 15 20 25 30
 Step Response
1.5

Amplitude
0.5

0
0 2 4 6 8 10 12
Time (sec)

TS zeta=.5;
 
Tp   wn=[0.5 1 1.5];
n 1   2 d
for k=1:3

num=[0 0 (wn(k))^2];

Which parameter (τ ,TS,, %OS) remains constant?den=[1 2*zeta (wn(k))^2];

 Step Response
1.8

1.6

1.4

1.2

Amplitude
1

0.8

0.6

0.4

0.2

0
0 5 10 15 20 25 30
Time (sec)

Envelope ?

wn=1;

zeta=[0.1 0.3 0.5];

for k=1:3

num=[0 0 wn^2];

den=[1 2*zeta(k) wn^2];

Which parameter (τ ,TP,, %OS) remains constant? sys=tf(num,den)
Problem   1  2
400
(a) G(s)  c( t )  A  B e nt cos(n t  )
s  12 s  400
2

n2
G(s)  c(t)  A  B e6t cos(19.08t  )
s 2  2 n s  n2
900
(b) G(s)  c( t )  A  Be  t / 1  Ce  t / 2
s  90 s  900
2

c(t )  A  Be78.54t  Ce11.46t

225
(c ) G(s)  c( t )  A  Be  t / 1  C te t / 2
s 2  30 s  225

c(t )  A  Be15t  C te15t

625
(d) G(s)  c(t )  A  B cos(25t  )
s  625
2
Third-order systems (Higher Order System)

• Consider one real pole plus a pair of

complex conjugate poles

– Complex poles dominant (close to

origin), real pole non-dominant

or

–Real pole dominant, complex poles

non-dominant
Complex poles dominant:
approximate as secon-dorder
system

Real pole dominant:

approximate as first-order system
All three poles dominant:
approximation difficult

Complex conjugate poles

Dominant:
zero tends to minimize
effect of nearby real pole
Real pole dominant,
zeros tend to minimize
effects of nearby poles

Unstable due to rhp pole,

nearby zero is not useful
in canceling pole
Lhp zero will amplify overshoot

Stable system, rhp zero can

give response that starts in
opposite direction from the
steady-state resp