Frequency Response and System Analysis 3.

29

-1 0 1 2 3 4
10 2 3 4 5 678910 2 3 4 5 678910 2 3 4 5 678910 2 3 4 5 678910 2 3 4 5 678910

Degree
0
New phase margin (pm )new = 64
New Gain margin (gm)new = 30 dB
40

30

20 (10 dB)
Magnitude plot
dB

10
0 dB line
0

-10
Phase plot (-10 dB) 0
-20 -100
0
-30 -120
0
-40 -140
0
-50 (-50 dB) 0
-160
-180 dB line 0
-60 -180
(-68.07 dB) 0
-70 -200
0
-80 -220

Fig. 1
3.7 POLAR PLOT

The polar plot is a plot of the magnitude of G  j  Vs Phase angle of G  j  on polar

coordinates as  is varied from zero to infinity. It is also called Nyquist Plot.
The minimum phase systems are system with all poles and zeros on the left half of s
plane.
Typical Sketches of Polar Plot
1. Type 0, order 1

1
G(s) 
1  sT
0
-270
0 0
-180 0

0
90
3.30 Control Systems Engg.

2. Type -1; Order : 2

1
G(s) 
s 1  sT 

0
-270
0
-180 0
0

0
-90

3. Type - 0; Order -2

1
G(s) 
1  sT1 1  sT2 
0
-270
0 0
-180 0

0
-90

4. Type -0, Order - 3

1
G(s) 
1  sT1 1  sT2  1  sT3 
0
-270
0 0
-180 0

0
-90

5. Type -1; Order -3

 Frequency Response and System Analysis 3.31

1
G(s) 
s 1  sT1 1  sT2 

0
-270

0 0
-180 0

0
-90

6. Type 2, Order -4

1
G(s) 
s 2 1  sT1 1  sT2 

0
-270

0 0
-180 0

0
-90

7. Type -2; Order -5

1
G(s) 
s 2 1  sT1 1  sT2  1  sT3 

0
-270

0 0
-180 0

0
-90
3.32 Control Systems Engg.

3.7.1 Typical Sketches of polar plot on an ordinary graph and polar graph :
The typical sketches of polar plot on an ordinary graph and polar graph based on the
TYPE and ORDER of a system are given below :
Type 0 Order 2 System :
Let the loop transfer function of TYPE 0 ORDER 2 system be

1
G(s) H(s) 
1  sT1 1  sT2 
Substituting s  j  , we obtain

1
G  j  H  j  
1  j T1 1  j  T2 
The magnitude and phase angle of the system are given by

1
G  j  H  j  
1  j  T1 1  j  T2 
and

 G  j   H  j     tan  1   T1   tan 1   T2 

The magnitude and phase angle at   0 and    are given in Table 1.

Table 1 : Magnitude and Phase angle of the system.

Frequency  Magnitude G  j   H  j   Phase angle  G  j   H  j  

0 1 00

 0  1800

To determine the intersection point on real and imaginary axis :

The real and imaginary part of loop transfer function can be determined as

G  j H  j 
1 1  j  T1 1  j  T2 

1  j T1 1  j T2  1  j T1 1  j  T2 
1  2 T1  j   T1  T2 

1  2T12 1  2T2 2 
 Frequency Response and System Analysis 3.33

Therefore,
real part of loop transfer function

1  2T1T2
G R  j HR  j 
1  2T12 1  2T22 
and
Imaginary part of loop transfer function

  T1  T2 
G I  j  H I  j  
1  2T12 1  2T22 
Intersection point on imaginary axis :

Step 1 : Equating the real point of loop transfer function to zero to determine cr i.e,

G R  j  H R  j   0

2
1cr T1T2
, 0
1cr2 T12 1cr2 T22 
1
upon solving, we obtain cr  T T
1 2

1
Step 2 : Substituting cr  in the imaginary part of loop transfer function, we obtain
T1T2
the intersection of the polar plot an imaginary axis.

Intersection point on imaginary axis  G1  j   H1  j  

1

T1T2

1
Substituting  in imaginary part of loop transfer function, we obtain
T1T2

T1T2
G1  j   H1  j   
T1  T2
3.34 Control Systems Engg.

Step 3 :

1
Therefore, magnitude and phase angle of loop transfer function at   is
T1T2

T1T2
Magnitude G  j  H  j  
T1  T2

and phase angle  G  j   H  j     900

Using the magnitude and phase angle at different frequencies, we can determine the
real and imaginary parts of the loop transfer function as given in Table 2.
Table 2 : Real and Imaginary parts of the system

Frequency  Real part G R  j   H R  j   Imaginary Part G I  j   H I  j  

0 1 0
Intersection
point on imaginary

1
axis, T1T2 0  T1T2 T1  T2

 0 0
Intersection point on real axis :
Step 1 :
Equating the imaginary part of loop transfer function to zero to determine the frequency
ci ie., G1  j   H1  j    0.

ci  T1  T2 
Therefore, 0
 
1  2 T12 1  2 T22
ci ci 
Upon solving, we get ci  0 .

Step 2 : Substituting ci  0 in the real part of loop transfer function, we obtain the intersection
point of polar plot on real axis.

Therefore, intersection point on real axis  G1  j   H1  j   1

 0
 Frequency Response and System Analysis 3.35

Using the above condition and Table 2, the polar plot of Type 0 ORDER 2 system on an
ordinary graph is shown in Fig. 1(a).
G1  j  H1  j  

G  j  H  j  plane
0
0 1.0
G R  j HR  j  

T1T2
T1  T2 1

T1T2

Fig. 1(a) Polar part of the system

The magnitude and phase angle at different frequencies are given in Table 3.

Table 3 : Magnitude and Phase angle

Frequency  Magnitude G  j   H  j   Phase angle  G  j   H  j  

0 1 00

Intersection point
on imaginary axis,

1 T1T2
T1T2 T1  T2  900

 0  1800

Using Table 3, the polar plot of Type 0 Order 2 system on polar graph is shown in Fig.
1(b)
3.36 Control Systems Engg.
0
-270

G  j  H  j  plane

0
0
-180 00
T1T2
T1  T2

1

T1T2
0
-90
Fig. 1(b) Polar Plot of the system
Similarly, the polar plot on an polar graph and intersection points on the different axes
for different systems can be determined.
3.7.2 Determining the Gain K from the desired frequency Domain Specifications :
The frequency domain specifications for which the gain K can be determined are gain
margin and phase margin. The procedure for determining the gain K for the desired gm and Pm
using polar plot is not so tedious as it was using Bode plot. The following sections describe
how the gain K can be determined for desired gm and Pm using polar plot.
When the desired gain margin of the system is specified
The step by step procedure to determine the gain K for a specified gain margin is
explained below :
Step 1 : Assume k = 1 for the given loop transfer function and construct the polar
plot of the system either on the polar graph or on the ordinary graph.
Step 2 : Determine the polar crossover point, A from the polar plot. If the polar is
plotted on the polar graph, the point A refers to the point at which the polar
plot crosses  1800 line and if the polar plot is plotted on the ordinary
graph, the point A refers to the point at which the polar plot crosses negative
real axis.
Step 3 : Determine the magnitude of the loop transfer function at A from the polar
plot. Let it be G  j   H  j   A .

Step 4 : Let the desired gain margin of the system be (gm)desired dB. Let B be the
phase crossover point on the polar plot corresponding to (gm)desired. The
magnitude of the loop transfer function at B can be determined
 Frequency Response and System Analysis 3.37

by using  20log G  j   H  j     g m  .
B desired

Step 5 : Now, the gain K for the desired gain margin can be determined as

G  j  H  j  A
K 
G  j H  j B

Step 6 : with the gain K, the new polar plot can be drawn. The polar plot for different
values of K is shown in Fig. 1(a).
0
-270

G  j  H  j   plane

0
-180 B A B 00
K>1
K=1
K<1

0
-90
Fig. 1(a) Polar plot for different values of K
When the desired phase margin of the system is specified :
The step by step procedure to determine the gain K for a specified phase margin is
explained below :
Step 1 : Assume K = 1 for the given loop transfer function and construct the polar
plot of the system either on the polar graph or on the ordinary graph.
Step 2 : Determine the gain crossover point C, from the polar plot. If the polar plot
is plotted on the polar graph or on the ordinary graph, the point C refers to
the point at which the polar plot crosses unity circle.
Step 3 : Determine the magnitude of the loop transfer function at C from the polar
plot. Let it be G  j   H  j   C .

Step 4 : Let the desired phase margin of the system be (Pm)desired dB. Let D be the
gain crossover point on the polar plot corresponding to (Pm)desired .
3.38 Control Systems Engg.

Step 5 : Determine the phase angle of the system corresponding to (Pm)desired. Let it
be E deg. The value of E is determined by using (Pm)desired  1800  E 0 .

Step 6 : Draw a radial line using E deg from origin that cuts the polar plot of the

loop transfer function G  j   H  j   with K = 1 at F. The magnitude of the

loop transfer function at F can be determined from the polar plot and let it
be G  j   H  j   F .

Step 7 : Now, the gain K for the desired gain margin can be determined as,

G  j  H  j  C 1
K  
G  j  H  j  F G  j  H  j  F

Step 8 : With the gain K, the new polar plot can be drawn. The polar plot for
different values of K is shown in Fig. 1(b).
0
-270
G  j  H  j    plane

0
-180 00
0
k>1

E
F
1
k=

0
C E
k<1

F
0
-90
Fig. 1(b) Polar plot for different values of K.

Problem 8 :The loop transfer function of a system is G(s) H(s) 

K  s  1
.
s  1  0.2s  1  0.5s 
Sketch the polar plot for the system.

Solution :

1. The loop transfer function of the system

 Frequency Response and System Analysis 3.39

K  s  1
G(s) H(s) 
s 1  0.2s 1  0.5s 

2. Substituting s  j  and K = 1 in the loop transfer function, we obtain

 j   1
G  j H  j  
j  1  j0.2 1  j0.5 

3. The 3 corner frequencies existing in the system are c1  1rad sec,

1 1
c   2rad sec and c3   5rad sec .
2 0.5 0.2

4. The Magnitude and phase angle of the system are

1  2
M
 1  0.042  1  0.252

and

  tan 1     900  tan 1  0.2   tan  1  0.5 

5. For different values of frequency  , the magnitude and phase angle of the system
are calculated and tabulated in Table 1.

Table 1. Magnitude and Phase angle of the system

Frequency
 in rad/sec 0.4 0.5 0.6 0.7 0.8 0.9 1 2 4 5 7 

Magnitude M 2.6 2.2 1.8 1.6 1.5 1.3 1.24 0.7 0.35 0.3 0.2 0.1

Phase angle
 in deg. -84 -8.3 -8.2 -82.3 -82 -82 -82.8 -93 -11 -124 -136 -148

The polar plot of the given system using Table 1 is shown in Fig. 8.
3.40 Control Systems Engg.
0 0 0
0 10 360 350 0
0 340
20 350
0
0 10
0
0 0 0 0
30 340 20 330
0 0
0 330 30 0
40 320
0 0
320 40 0
0
50 Scale 310
0 1 Circle = 0.05 magnitude 0
310 50
0 0
60 300
0 0
300 60
0
70
0 290
0
290
0 70
0
0 280
80 0
0   0.8 80
280 1   0.7
0 0
90 270
0
  10  2 0
270 7  4 90
0
  0.5
0
100 260
0 0.5 0
260 100
0 0
110 250
0 0
250 1.0 110
0 0
120 240
0 0
240 1.5 120
0 0
130 230
0 0
230 0
130
0
140 2.0 220
0 0
220 0
210
0 140
150 0 0
210
0
160
0 2.5 0 200 150
0 0 190 0
0 170 180 160
200 0 0 170
0
190 180

Fig. 8
Problem 9 : The loop transfer function of a unity feedback system is given by
1
G(s) H(s)  2
. Sketch the polar plot for the system and determine
s1  s 
the gain and phase margin of the system.
Solution :
1. The Loop transfer function of a system is

1
G(s) H(s)  2
.
s 1  s 
 Frequency Response and System Analysis 3.41

2. Substituting s  j , we obtain
1
G  j  H  j   2
j  1  j  
3. The corner frequency of the given system is c1  1 rad/sec.
4. The magnitude and phase angle of the system are
1 1
M   and
  1  2   1  3 

   900  2 tan 1 
5. For different values of frequency , the magnitude and phase angle of the system
are calculated and tabulated in Table 1.
Table 1 : Magnitude and Phase angle of the system

Frequency
 in rad/sec 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 1.2 1.3

Magnitude M 2.2 1.6 1.2 1 0.8 0.6 0.5 0.4 0.3 0.26
Phase angle

 in deg. -134 -143 -151 -159 -167 -174 -180 -185 -190 -195
The polar plot of the given system using Table 1 is shown in Fig. 9.
Determination of Gain Margin and Phase Margin
The gain margin and phase margin of the system are calculated at phase cross over
frequency and gain crossover frequency respectively. The phase crossover frequency is the
frequency at which the phase angle of the system is 180 0 or  1800 . Also the gain crossover
frequency is the frequency at which the magnitude of the system is 1.
Using the above definition and Table 1, we obtain Gain crossover frequency
gc  0.7 rad sec and phase crossover frequency pc  1rad sec.

Hence gain margin and phase margin of the system are given by Gain margin,
1 1
gm   2
G  j  H  j   0.5
pc

and phase Margin, Pm  1800    G  j   H  j    gc

 1800  1590  210

3.42 Control Systems Engg.

Scale
1 circle = 0.05 magnitude 1
Gain margin = g m 

G j pc H j pc   
1
 2
0.5
pm  1800   gc  180 0  159 0  210

0 0 0
0 10 360 350 0
0 340
20 350
0
0 10
0
0 0 0 0
30 340 20 330
0 0
0 330 30 0
40 320
0 0
320 40 0
0
50 310
0 0
310 50
0 0
60 300
0 0
300 60
0
70
0 290
0
0 70
290
0
0 280
80 0
0 80
280
0 0
90 270
0 0
270  1.3 90
0

G j pc H j pc 
 1.2  0
100  1.1 0.5 260
260
0  1.0  0.9 100
0

0  0.8 0
110  0.7 250
0 0
250 1.0  0.6 110
0  0.5 0
120  0.4 240
0 0
240 1.5 120
0 0
130 230
0 0
230 0
130
0 2.0
140 220
0 0
220 0
210
0 140
150 0 0
210
0
160
0 2.5 0 200 150
0
0 170 180
0 190 0
200 0 0 160
190 180
0 170

Fig.9
 Frequency Response and System Analysis 3.43

Problem 10 : The open loop transfer function of a unity feedback system is given by
G(s)  1 
s 1  s  1  2s  : Switch the polar plot and Determine the gain
margin and phase margin.
Solution :

1
G  j  
j  1  j  1  2 j  

1

900 1  2 tan  1  1  42 tan  1 2

1
   900  tan  1   tan  1 2
 1  2 1  42

1
G  j  
 1  2 1  42 

 G  j     90 0  tan  1   tan  1 2

 0.35 0.4 0.45 0.5 0.6 0.7 1.0

G  j  2.2 1.8 1.5 1.2 0.9 0.7 0.3

G  j  -144 -150 -156 -162 -171 -179.5 -198

  1800

Gain Margin = 1.4286

Phase Margin  = 120

3.44 Control Systems Engg.

Scale
1 circle = 0.05 magnitude
1
Gain margin = K g 
G  j 
pc
1
  1.4286
0.7
p m ,  1800  gc  1800 1680  120

0 0 0
0 10 360 350 0
0 340
20 350
0
0 10
0
0 0 0 0
30 340 20 330
0 0
0 330 30 0
40 320
0 0
320 40 0
0
50 310
0 0
310 50
0 0
60 300
0 0
300 unity circle 60
0
70
0 290
0
0 70
290
0
0 280
80 0
0 80
280
0 0
90 270
0 0
270 90
0 0
100 0.5 260
0 0
260  0.7
100
0 0
110 250
0 0
250 1.0 110
0 0
120 240
0 0
240 1.5 120
0 0
130 230
0 0
230 0
130
0 2.0
140 220
0 0
220 0
210
0 140
150 0 0
210
0
160
0 2.5 0 200 150
0
0 170 180
0 190 0
200 0 0 160
190 180
0 170

Fig. 10
 Frequency Response and System Analysis 3.45

Problem 11. The open loop transfer function of a unity feedback system is given by

1 . Sketch the polar plot and determine the gain

G(s) 
s 2  1  s  1  2s 
margin a and phase margin.
Solution :

1
G(s) 
2
s 1  s 1  2s 

1
G  j 
2
 j   1  j  1  2 j  

1

21800 1  2  tan  1  1  42  tan  1 2

1
G  j  
2 1  2 1  42

 G  j     1800  tan  1   tan  1 2

 0.45 0.5 0.55 0.6 0.65 0.7 0.75 1

G  j  3.3 2.5 1.9 1.5 1.2 1.2 0.97 0.3

1

G  j  -246 -251 -256 -261 -265 -269 -278 -288

GM, Kg = 0

PM,  =  900
3.46 Control Systems Engg.

Gain margin  1 1
 0
G  j   
pc
0
pm ,  180   gc

 1800  2700

  900

0 0 0
0 10 360 350 0
0 340
20 350
0
0 10
0
0 0 0 0
30 340 20 330
0 0
0 330 30 0
40 320
0 0
320 40 0
0
50 310
0 0
310 50
0 0
60 300
0 0
300 60
0
70
0 290
0
290
0 unity circle 70
0
0 280
80 0
0 80
280
0 0
90 270
0 0
270 90
0 0
100 1 260
0 0
260 100
0 0
110 250
0 0
250 2 110
0 0
120 240
0 0
240 3 120
0 0
130 230
0 0
230 0
130
0
140 4 220
0 0
220 0
210
0 140
150 0 0
210
0
160
0 5 0 200 150
0
0 170 180
0 190 0
200 0 0 160
190 180
0 170

Fig. 11
 Frequency Response and System Analysis 3.47

Problem 12 : The open loop transfer function of a unity feedback system is given by
 1  0.2s  1  0.025s  . Sketch the polar plot and determine
G(s) 
3
s 1  0.005s  1  0.001s 
the phase margin.
Solution :

1  0.2s 1  0.025s 

G(s) 
3
s 1  0.005s 1  0.001s 

1  0.2 j  1  0.025j  

G  j 
 j  3 1  0.005j  1  0.001j  

2 2
1   0.2   tan  1 0.2 1   0.025   tan  1 0.025

2 2
3  2700 1   0.005   tan  1 0.005 1   0.001   tan  1 0.001

2 2
1   0.2  1   0.025 
G  j  
2 2
3 1   0.005  1   0.001 

 G  j    tan  1 0.2  tan  1 0.025  2700  tan  1 0.005  tan  1 0.001

 0.9 0.95 1.0 1.1 1.2 1.4 1.7

G  j  1.4 1.2 1.0 0.8 0.6 0.4 0.2

G  j  -259 -258 -257 -256 -255 -253 -249

0
Phase Margin   77
3.48 Control Systems Engg.

Scale :
1 Circle = 0.04 magnitude
pm ,  180 0  gc

 180 0  257 0
  77 0

0 0 0
0 10 360 350 0
0 340
20 350
0
0 10
0
0 0 0 0
30 340 20 330
0 0
0 330 30 0
40 320
0 0
320 40 0
0
50 310
0 0
310 unity circle 50
0 0
60 300
0 0
300 60
0
70
0 290
0
0 70
290
0
0 280
80 0
0 80
280
0 0
90 270
0 0
270 90
0 0
100 0.4 260
0 0
260 100
0 0
110 250
0
250 0.8 110
0

0 0
120 240
0 0
240 1.2 120
0 0
130 230
0 0
230 0
130
140
0 1.6 220
0 0
220 0
210
0 140
150 0 0
0 0 2.0 200 150
210 160 0 0 190
0
0
0 170 180 160
200 0 0 170
0
190 180

Fig. 12
 Frequency Response and System Analysis 3.49

Problem 13 : Consider a unity feedback system having an open loop transfer function
K
G(s)  . Sketch the polar plot and determine the value
s  1  0.5s  1  4s 
of K so that (i) Gain Margin is 20 db (ii) Phase Margin is 300 .
Solution :

K
G(s) 
s 1  0.5s 1  4s 


Let K = 1.

1
G  j  
j  1  0.5j  1  4 j  

1

2 2
900 1   0.5   tan  1 0.5 1   4   tan  1 4

1
G  j    900  tan  1 0.5  tan  1 4
2 2
 1   0.5  1   4 

 G  j     900  tan  1 0.5  tan  1 4

 0.3 0.4 0.5 0.6 0.8 1.0 1.2

G  j  2.11 1.3 0.87 0.61 0.35 0.22 0.15

G  j  -149 -159 -167 -174 -184 -193 -199

3.50 Control Systems Engg.

Scale :
1 Circle = 0.04 magnitude
1 1
GM    2.27
G  j  0.44
pc

  1800  gc

 1800  1650

 150

0 0 0
0 10 360 350 0
0 340
20 350
0
0 10
0
0 0 0 0
30 340 20 330
0 0
0 330 30 0
40 320
0 0
320 40 0
0
50 310
0 0
310 50
0 0
60 unity circle 300
0 0
300 60
0
70
0 290
0
0 70
290
0
0 280
80 0
0 80
280
G  j 
0  pc 0
90 270
0 0
270 90
0 0
100 0.4 260
0 0
260 100
0 0
110 250
0 0
250 0.8 110
0 0
120 GB 240
0 0
240 1.2 120
0 0
130 230
0 0
230 0
130
0
140 1.6 220
0
 gc1 0
220 0 G  j  210 140
0
150 gc2
0 0
210
0
160
0 2.0 0 165 200
0
150
0
0 170 180
0 190 0
200 0 0 160
190 180
0 170

Fig. 13
 Frequency Response and System Analysis 3.51

From polar plot, with K = 1

1
Gain Margin, Kg   2.27
0.44
Gain Margin, db = 20 log 2.27
= 7.12 db
0
Phase Margin,   15 .
Case (i)

Let GA be the gain at  1800 for a GM of 20 db.

1 GA
20 log  20 K 
GA GB

1 0.1
log 1 
GA 0.44

1
 10 = 0.224
GA

G A = 0.1 ( G B from polar plot)

Case (ii)

with K = 1,   150. This is has to be  to 300

300  1800  gc2

g  1500
12

Let G A  Magnitude of G  j  at point A

G B  Magnitude of G  j  at point B

From polar plot, G A  2.04 ; GB  1 ;

GB 1
K    0.49
GA 2.04