CONTROL SYSTEMS LABORATORY (18EEL66) 2020-21

Time Response Characteristics: b. By using pointer read the time domain specifications.
Rise Time (TR): Rise time is defined as the time required for the response to rise Program:
from 10% to 90% of the final value for an under damped system. Program to obtain step response of second order system
clc
% prg c(t)
num = [5];
den = [1 1 5];
Settling Time (TS): Settling time is defined as the time required for the response G = tf(num, den)
to decrease and stay within the specified percentage (2% to 5%) of its final value. kp=dcgain(G)
ess=1/(1+kp)
w = sqrt (den(3))
zeta = den(2) / (2*w)
TD=(1+0.7*zeta)/w
TS = 4/ (zeta*w)
Peak Overshoot (M P): Peak overshoot is defined as the maximum response of the TP = pi/ (w*sqrt(1-zeta^2))
system measured from the final value. TR=(pi-atan((sqrt(1-zeta^2))/zeta))/(w*sqrt(1-zeta^2))
Percentovershoot= exp(-zeta*pi/ sqrt(1-zeta^2))*100
Peak Time (TP): The time required to reach the first peak overshoot is defined as step(G)
title('Step response of c(t)')
peak time.
figure
pzmap(G)
title('pole zero map of c(t)')
1. Consider an example for the second order system with an open loop figure
transfer function as and unity feedback system H(s) = 1. Output Response:
2. The closed loop transfer function is determined by using the
equation,
3. By substituting the values of G(s) and H(s) in above equation we
get,
4. Comparing Equation (1) and (8), we get = 2.236 and = 0.224. Since
ξ < 1, the given system is under damped.
5. Write the program using GNU OCTAVE to simulate step response of the
given system and save the file and load the control package.
6. Evaluate time domain specifications :
a. Run the program. Step response appears on the screen.

DEPARTMENT OF EEE, ATMECE, MYSURU. Page 28

CONTROL SYSTEMS LABORATORY (18EEL66) 2020-21

Fig.7.1: Step Response Curve step(G)

title('Step response of c(t)')
Results:
figure
pzmap(G)
Particulars Theoretical Value Simulated Value title('pole zero map of c(t)')
Delay Time figure
Rise Time
Peak Time Output Response:
Settling Time
Peak Overshoot

Case – II: Evaluation of the effect of additional poles and zeroes on time
response of second order system.

Effect of adding pole to the transfer function:

1. If pole is added at -2 for the open loop transfer function.

.
The transfer function gets modified as,

2. The closed loop transfer function for the given system is obtained as,
Fig.7.2: Step response after adding pole to the transfer function

Results:
3. Effect of adding pole to transfer function results in
a. Increases the order of the system. Particulars Simulated Value
b. Increases the overshoot. Delay Time
c. Reduces the stability. Rise Time
d. Increases the Rise time of the step response.
Peak Time
Program:
Settling Time
clc
Peak Overshoot
% prg c(t)
num = [5];
den = [1 3 2 5];
G = tf(num, den)

DEPARTMENT OF EEE, ATMECE, MYSURU. Page 29

CONTROL SYSTEMS LABORATORY (18EEL66) 2020-21

Effect of adding zero to the transfer function: Output Response:

1. If zero is added at -2 for the open loop transfer function.

.
The transfer function gets modified as,

2. The closed loop transfer function for the given system is obtained as,

3. Comparison of location for zeroes with respect to overshoot and

damping parameters.
Time Location of Zeros
response Far away from the Zero moving Zero moves
parameters imaginary axis towards right closer to origin
Overshoot Large Reduced Almost nil Fig.7.3: Step response after adding zeros to the transfer function
Damping Very poor Improved Improves
Results:
Program:
clc Particulars Simulated Value
% prg c(t) Delay Time
num = [5 10]; Rise Time
den = [1 6 10]; Peak Time
G = tf(num, den)
Settling Time
step(G)
title('Step response of c(t)') Peak Overshoot
figure

DEPARTMENT OF EEE, ATMECE, MYSURU. Page 30

CONTROL SYSTEMS LABORATORY (18EEL66) 2020-21

Fig.7.4: Time response for

Case – III: Evaluation of effect of pole location on Stability:
Example 1(b): i.e. Pole at s = 5
1. The stability of a feedback system is directly related to the location of the
roots of the characteristic equation of the system transfer function.
2. “A linear system will be stable if and only if all the poles of the transfer Program:
function are located on the left half of the ‘S’ plane”. clc
3. Consider the below examples to study the effect of location of pole on % prg c(t)
stability of the system. num = [1];
den = [1 -5];
G = tf(num, den)
Example 1(a): i.e. Pole at s = -5 impulse(G)
title('Step response of c(t)')
Program:
figure
clc
% prg c(t) Output Response:
num = [1];
den = [1 5];
G = tf(num, den)
impulse(G)
title('Step response of c(t)')
figure

Output Response:

Fig.7.5: Time response for

Example 2(a): i.e. Pole at s = -2.5

Program:
clc

DEPARTMENT OF EEE, ATMECE, MYSURU. Page 31

CONTROL SYSTEMS LABORATORY (18EEL66) 2020-21

% prg c(t)
num = [1]; Output Response:
den = [1 5 29];
G = tf(num, den)
impulse(G)
title('Step response of c(t)')
figure

Output Response:

Fig.7.7: Time response for

Case – IV: Effect of loop gain of a negative feedback system on stability

1. Effect of loop gain of a negative feedback system on stability of a system
Fig.7.6: Time response for can be identified by knowing characteristic equation of the system and
Example 2(b): i.e. Pole at s = 2.5 applying “Routh’s Stability Criterion”.
2. Consider an open loop transfer function of a system as,
Program:
clc
% prg c(t) 3. Determine the value of ‘K’ loop gain for which the system is stable.
num = [1]; Consider the system with unity feedback H(s) = 1.
den = [1 -5 29]; 4. The closed loop transfer function is obtained as,
G = tf(num, den)
impulse(G)
title('Step response of c(t)')
figure
5. From closed loop TF, Determine characteristic equation

DEPARTMENT OF EEE, ATMECE, MYSURU. Page 32

CONTROL SYSTEMS LABORATORY (18EEL66) 2020-21

Program: Program:

clc clc
% prg c(t) % prg c(t)
num = [1 1]; num = [1 1];
den = [1 2 6]; den = [1 2 6 0];
G = tf(num, den) G = tf(num, den)
step(G) step(G)
title('Step response of c(t)') title('Step response of c(t)')
figure figure

Output Response:
Output Response:

Fig 9.2: Step Response of Type 0 System

Fig 9.3: Step Response of Type 1 System
Example-2: Consider a type 1 system given by transfer function

C ( s) ( s  1)
 Example-3: Consider a type 2 system given by transfer function
R( s) s( s  2s  6)
2
C ( s) ( s  1)
 2 2
Step response of the system can be simulated as given in the program below. R( s) s ( s  2s  6)

DEPARTMENT OF EEE, ATMECE, MYSURU. Page 41

CONTROL SYSTEMS LABORATORY (18EEL66) 2020-21

Step response of the system can be simulated as given in the program below. Observation:

Program: System Type Steady state error for unit step input
Type 0 1/1+Kp
clc
Type 1 ∞
% prg c(t)
Type 2 ∞
num = [1 1];
den = [1 2 6 0 0];
G = tf(num, den) Example-4: Consider the type 0, type 1 and type 2 systems given in examples 1, 2
step(G) and 3. The effect on steady state errors for these systems for a ramp input can be
title('Step response of c(t)') simulated by the program given below.
figure Program:

Output Response: sys1=tf([1 1],[1 2 6 ]);

sys2=tf([1 1],[1 2 6 0]);
sys3=tf([1 1],[1 2 6 0 0]);
t=0:0.1:10;
u1=t;
lsim(sys1,'r', sys2,'m', sys3,'g', u1,t)

Output Response:

Fig 9.4: Step Response of Type 2 System

Fig.9.5: Effect of ramp input on steady state errors of type 0(sys1), type
1(sys2) and type 2(sys3) systems

DEPARTMENT OF EEE, ATMECE, MYSURU. Page 42

CONTROL SYSTEMS LABORATORY (18EEL66) 2020-21

Observation: Observation:

System Type Steady state error for ramp input System Type Steady state error for Parabolic input
Type 0 ∞ Type 0 0
Type 1 1/Kv
Type 1 0
Type 2 0
Type 2 1/Ka
Example-5: Consider the type 0, type 1 and type 2 systems given in examples 1, 2
and 3. The effect on steady state errors for these systems for a parabolic input can
be simulated by the program given below. Case C: To perform a trade-off study for lead compensation
Program: The lead compensation on bode plots proceeds by adjusting the system
sys1=tf([1 1],[1 2 6 ]); error constant to a desired value.
sys2=tf([1 1],[1 2 6 0]);
sys3=tf([1 1],[1 2 6 0 0]); The phase margin of the uncompensated system is then checked. If
t=0:0.1:10; found satisfactory, the lead compensation is designed to meet the specified phase
u1=t.*t; margin. With lead compensation introduced in the system, following observations
lsim(sys1,'r', sys2,'m', sys3,'g', u1,t) are made.

Output Response:  Phase margin is increased.

 Bandwidth increases.
 Peak resonance is reduced.
 Resonance frequency is increased.
Thus in general, the effect of lead compensator increases the
margin of stability and speed of response.

Example: Consider a type 1 unity feedback system with an open loop transfer
function.

G f s  
12
S ( S  1) .

The bode plot of the given system is obtained by the program given below.
Fig 9.6: Effect of parabolic input on steady state errors of type 0(sys1), type
1(sys2) and type 2(sys3) systems

DEPARTMENT OF EEE, ATMECE, MYSURU. Page 43

CONTROL SYSTEMS LABORATORY (18EEL66) 2020-21

Program:
num = [12];
den = [1 1 0];
sys = tf (num , den)
bode (sys)
margin(sys)

Output Response:

Fig 9.8: Bode plot of a compensated system

Observation:

System Phase margin

Uncompensated system 16.4 degrees at 3.3 rad/sec
Fig 9.7 Bode plot of a type 1 system without compensation
Compensated system 42.2 degrees at 4.46 rad/sec

Consider a lead compensator whose transfer function is given by;

Conclusion:
0.377 s  1
Gc s  
0.128s  1
The open loop transfer function of the compensated system is now given by;
Outcome: At the end of experiment,
120.377 s  1
Gc s   The students will have knowledge of DC position control of DC motor and also
S ( S  1)(0.128s  1) effect of type of system and input waveform on steady state error.

DEPARTMENT OF EEE, ATMECE, MYSURU. Page 44

CONTROL SYSTEMS LABORATORY (18EEL66) 2020-21

Condition for stability through PM and GM: For a system to be stable both PM = Phase margin
gain margin and phase margins must be positive. Also ωp>ωg . WCF = phase cross over frequency
WCG = Gain cross over frequency.
Construct the Bode plot and find gain margin and phase margin and comment on These parameters will be displayed in the command window.
stability for the given transfer function. Check for the stability of the system. For a system to be stable both gain
G(s) = margin and phase margins must be positive. Also ωp>ωg .If this condition is
satisfied then the system is stable or otherwise the system is unstable.
Program:
Output Response:
num = 50;
den =[0.025 0.55 1 0];
sys = tf(num,den);
bode(sys);
margin(sys);

From the plot note down the gain margin (GM), phase margin (PM) and
the corresponding cross over frequencies. For unstable systems, GM and PM will
not be displayed correctly but can be obtained by clicking on the plot at suitable
points.

a. Click on the phase angle curve and find the frequency at which the curve
crosses the 1800 line. This gives the phase cross over frequency ωp.
b. Click on the magnitude curve and find the magnitude at ω=ω p. The gain
margin is calculated as, GM = 0 - magnitude at phase cross over frequency.
c. Click on the magnitude curve and find the frequency at which the curve
crosses 0dB line. This will give the gain cross over frequency ω g. Fig.10.1: Frequency response of the system
d. Click on the phase angle curve and find the phase angle at ω=ω g. The phase
margin is calculated as PM = phase angle at the gain cross over frequency + Conclusion:
1 0 at ω=ωg.
The GM, PM and the cross over frequencies can also be obtained using
the following function.

[GM PM WCF WCG] = margin (sys);

Where GM = gain margin in abs unit (20logGM is the GM in db)

DEPARTMENT OF EEE, ATMECE, MYSURU. Page 46

CONTROL SYSTEMS LABORATORY (18EEL66) 2020-21

Typical Graph: Effect of adding a zero on root locus plot

1. The root locus shifts towards left half of S-plane.

2. Relative stability of system is improved.

Consider a zero to be added to the given transfer function as given by

K ( s  1)
G( s) H ( s) 
s( s  1)(s  2)( s  3)
The program is modified as

num=[1 1 ];
den=[1 6 11 6 0];
sys=tf(num,den);
printsys(num,den);
[r,k]=rlocus(sys)
rlocus(sys)

Typical Graph:
Fig.11.1: Root locus curve

Tabular Column:

Theoretical Values Simulated Values

Poles
Breakaway Point
Gain
Imaginary Axis
Crossover

Fig.11.2: Root locus curve

DEPARTMENT OF EEE, ATMECE, MYSURU. Page 48

CONTROL SYSTEMS LABORATORY (18EEL66) 2020-21

Effect of adding a pole on root locus plot Conclusion:

a. Root locus shifts towards the right half of S-plane.

b. Angles of asymptotes reduce.
c. Relative stability of the system is decreased.

Consider a pole to be added to the given transfer function as given by Outcomes: At the end of the experiment,
K
G( s) H ( s)  1. The students will be able to plot root locus using root locus.
s( s  1)(s  2)(s  3)(s  4) 2. The students will able to find stability of a system using root locus.
The program is modified as
num=[1 ]; Viva Questions:
den=[1 10 35 50 24 0]; 1. What is the significance of root locus method?
sys=tf(num,den); 2. What are the rules of construction of root locus?
printsys(num,den); 3. Mention the advantages and disadvantages of root locus method.
[r,k]=rlocus(sys) 4. What is meant by Asymptotes?
rlocus(sys)

Typical Graph: Calculation/Result

Preparation Conduction Total
Analysis

Fig.11.3: Root locus curve

Signature of Faculty Member

DEPARTMENT OF EEE, ATMECE, MYSURU. Page 49