Module 5

EET302 Linear Control System Lecture Notes by T.G.
Sanish Kumar, EED, GEC Thrissur
NYQUIST STABILITY CRITERION

The Nyquist criterion is a semi-graphical method that determines the stability of a closed loop system
by investigating the properties of the frequency-domain plot of the loop transfer function G(s)H(s).
Using Routh stability criterion, only the absolute stability of a system can be found out. But Nyquist
stability criterion provides, in addition to the absolute stability, the information about the relative
stability of a system. (Absolute stability means whether the system is stable or unstable. Relative
stability gives the degree of stability or how close it is to instability.)
RELATION BETWEEN POLES AND ZEROS OF OPEN LOOP AND CLOSED LOOP SYSTEMS
Consider the open loop transfer function,
N (s) ( s  a )( s  b )
G (s)H (s)   where a1 & b1 are open loop poles
D ( s ) ( s  a1 )( s  b1 )
Closed loop transfer function is given by
C (s) G (s)

R (s) 1  G (s) H (s)
Closed loop poles are the roots of the characteristic equation, 1  G ( s ) H ( s )  0
N1(s)
F (s)  1  G (s)H (s) 
D1 ( s )
Poles of closed loop transfer function are given by F ( s )  0 or N 1 ( s )  0 .
Hence, poles of the closed-loop transfer function are the zeros of q(s).
1 Poles of F ( s )  1  G ( s ) H ( s ) are the open loop poles
2 Zeros of F ( s )  1  G ( s ) H ( s ) are the closed loop poles
3 For a stable system, the zeros of F(s) should not lie in the RHP.
2( s  1)
Given G ( s )  . Is the open loop system stable? Is the closed loop system stable?
( s  1)
2( s  1)
G (s) 
( s  1)
Open loop system has pole in the right-hand side of s-plane. Hence it is not stable.
2
( s  1)
C (s) 2( s  1)
  3
R ( s ) s  1  2( s  1) 1
s
3
Closed system is stable because only pole is at s = -1/3.
2( s  1)
Given G ( s )  . Is the open loop system stable? Is the closed loop system stable?
( s  1)
2( s  1)
G (s) 
( s  1)
Open loop system has pole in the right-hand side of s-plane. Hence it is not stable.
2
( s  1)
C (s) 2( s  1)
  3
R ( s ) s  1  2( s  1) 1
s
3
Page 139 of 187

EET302 Linear Control System Lecture Notes by T.G. Sanish Kumar, EED, GEC Thrissur
Closed system is unstable because there is a pole in the right-hand side of s-plane.
BASIC DEFINITIONS:
Contour map: - Contour map is a contour or trajectory in one plane mapped or translated into another
plane by a relation F(s). This process is called mapping.
For example, let us take a pole in the s-plane s o  1  j 2 and let the function be F ( s )  s 2  5 .
Substituting s o  1  j 2 in the function F ( s )  s 2  5 gives
F ( s o )  s 2  5  (1  2 j ) 2  5  2  4 j
Closed contour: - A closed contour in a complex plane is a continuous curve beginning and ending at
the same point.
Enclosed: - A point or region is said to be enclosed by a closed path if the point or region lies to the
right of the path when the path is traversed in the prescribed direction.
In figure (1), point B is enclosed; but A is not enclosed. In figure (2), point A is enclosed but B is not
enclosed.
Encircled: - A point is said to be encircled by a closed path if it lies inside that closed path. Point B in
figure (1) and (2) are encircled.
No. of encirclements N: -
Page 140 of 187

The number of clockwise encirclements of origin in figure (1) is 3 and in figure (2) is 0.
Principle of argument or mapping theorem

Let F(s) be a single-valued function that has a finite number of poles in the s-plane. Suppose that an
arbitrary closed path C is chosen in the s-plane so that the path does not go through any of the poles
or zeros of F(s), the corresponding Cs locus mapped in the F(s) plane will encircle the origin as many
times as the difference between the number of zeros and poles of F(s) that are encircled by the s-
plane locus C.
In equation form, the principle of argument is stated as

N=Z–P
where
N = number of encirclements of the origin made by the F(s) plane locus Cs
Z = number of zeros of F(s) encircled by the s-plane locus C in the s plane
P = number of poles of F(s) encircled by the s plane locus C in the s plane
For Z = 4, P = 2, N = 2 → 2 clockwise encirclements

Z = 2, P = 2, N = 0 → No encirclements
Z = 2, P = 4, N = -2 → 2 counter-clockwise encirclements
Page 141 of 187

Page 142 of 187

For stability of a closed-loop system, the zeros of the characteristic equation

F ( s )  1  G ( s ) H ( s )  0 must lie in the left half of the imaginary axis in s – plane.
If any one zero of 1  G ( s ) H ( s ) is in the right half of the s-plane then the system is unstable.
Note: Zeros of 1  G ( s ) H ( s ) are the poles of the closed-loop transfer function.
Nyquist suggested that rather than analyzing whether all the zeros are located in the left half of the s
plane, it is better to examine the presence of any one zero of 1  G ( s ) H ( s ) in the right half of the s
plane making the system unstable.
Since G ( s ) H ( s )   1 , it is convenient to choose the

G ( s ) H ( s ) plane instead of F(s) plane for investigating
the stability. Hence, the relation N = Z – P can still be
used for finding the unstable zeros of F(s) where N is the
encirclement of the point (-1 + j0) in the GH plane by the
Nyquist path CGH.
SELECTION OF NYQUIST CONTOUR

The Nyquist contour is selected such that it encloses the entire right-half side of the imaginary axis in
the s-plane. It should enclose all the poles and zeroes present on the right-half of the s-plane except the
singular points. The singular points are considered as the points lying on the imaginary axis. Hence,
these points are avoided.
If the transfer function has a pole at origin, the contour encloses all the poles and zeroes except the
origin.
If there are no poles on the imaginary axis, the contour will appear as:
Page 143 of 187

If there are poles on the imaginary axis and origin, the avoided points (origin and imaginary) in the
contour are shown below:
MAPPING A CONTOUR
A singular point (point on the imaginary axis) is not
analytical. Hence, it is generally avoided. The
Nyquist contour is mapped to determine the
encirclement of the point -1 + j0. The contour is
drawn based on the transfer function G(s)H(s).
Generally, there are four sections C1, C2, C3, and C4.
The Nyquist criterion plot is divided into four
sections so that the mapping can be easily carried out
section wise. At last, all the sections are combined to
produce the desired Nyquist plot. Let the four
sections be:
Mapping of section C1
The value of ω in section C1 ranges from 0 to +∞. The contour will be drawn in G(s)H(s) plane with
respect to the above range, and it will be the locus plot of G(jω)H(jω). For mapping section C1, calculate
the values of G(jω)H(jω) for various values of  and sketch the actual locus of G(jω)H(jω).
Page 144 of 187

The section C2 is a semi-circle of infinite radius. Therefore every point on section C2 has infinite
magnitude but the angle varies from +90˚ through 0˚ to -90˚. The mapping of section C 2 from s-plane
to G(s)H(s) plane can be obtained by taking s  Lt R e j and varying  from +90˚ through 0˚ to -90˚.
R
In the third section C3, the value of ω ranges from -∞ to zero. The locus of the third section is just the
inverse of the polar plot of G(jω)H(jω). It is the inverse of the section C1. It is given by the mirror image
of the polar plot with respect to the real axis.
The section C4 of the Nyquist contour has a semicircle of zero radius. Every point on semicircle has
zero magnitude but the argument varied from -90˚ to +90˚. Hence the mapping of section C4 from s-
plane to G(s)H(s) plane can be obtained by s  Lt  e j substituting in G(s)H(s) and varying  from
 0
-90˚ to +90˚.
NYQUIST STABILITY CRITERION
If the contour CGH of the open-loop transfer function G(s)H(s) in the G(s)H(s) plane corresponding
to the Nyquist contour in the s-plane encircles the point -1+j0 in the counter clockwise direction as
many as time as the right-half s-plane poles of G(s)H(s), then the closed-loop system is stable.
If P = number of poles of G(s)H(s) or F(s) in the right-half of the s-plane

N = number of clockwise encirclements of (-1+j0) point in the GH plane
Z = Number of zeros of F(s) = 1 + G(s)H(s) in the right half of the s-plane
To make a closed-loop system stable, there must not be any zero in the right-half of s-plane. Therefore,
Z = 0.
Z=P+N=0
If Z = P + N  0, closed loop system is unstable.
Note: Positive sign for clockwise encirclement and negative sign for anticlockwise encirclement.
50
Sketch the Nyquist plot for the open loop transfer function G ( s ) H ( s )  and hence
( s  1)( s  2)
determine the stability of the closed loop system by Nyquist criterion.
25
G ( s) H (s) 
(1  s )(1  0.5 s )
Open loop poles in right side of s–plane = P = 0
25
G ( j ) H ( j ) 
(1  j )(1  j 0.5 )
25
G ( j ) H ( j ) 
1   2  1  (0.5 ) 2
Page 145 of 187

G ( j ) H ( j )   tan 1   tan 1 0.5

Nyquist contour is
MAPPING OF C1 & C3
 G ( j ) H ( j )  G ( j ) H ( j )
0 25 0
∞ 0 -180
MAPPING OF C2
 
s  Lt Re j ;    to  clockwise
R  2 2
50 50 50
G ( s ) H ( s ) s  Re j   
(Re j  1)(Re j  2) Re j  Re j R 2 e j 2
Lt GH (Re j )  Lt  e  j 2
R   0
 Lt GH (Re j )
R 
 Lt  e  j
  0
2
 Lt  e  j
  0
2
C2 in s-plane is mapped into a circular arc of radius 0 at origin in the GH-plane.
Page 146 of 187

Meeting point on positive real axis,   0 ; G ( j ) H ( j )  25

Meeting point on imaginary axis,  G ( j ) H ( j )   tan 1   tan 1 0.5  90 ;   2 ;
25 25
G ( j ) H ( j )    11.785
1    1  (0.5 )
2 2
1  2  1  (0.5  2) 2
Point -1+j0 is not encircled by Nyquist plot. Hence, N = 0.

N = 0; P = 0
Z = P + N = 0; system is stable.
s2
Sketch the Nyquist plot for the open-loop transfer function G ( s) H ( s )  and hence
( s  1)( s  1)
s2 2(1  0.5 s )

G ( s) H (s)  
( s  1)( s  1) (1  s )( 1  s )
2(1  j 0.5 )
G ( j ) H ( j ) 
(1  j )( 1  j )
2 1  (0.5 ) 2 2 1  (0.5 ) 2
G ( j ) H ( j )  
1 2  1 2 1 2
 G ( j ) H ( j )  tan 1 0.5  tan 1   tan 1   180  tan 1 0.5  180
MAPPING OF C1 & C3
 G ( j ) H ( j )  G ( j ) H ( j )
0 2 -180˚
∞ 0 -90
MAPPING OF C2
 
R  2 2
Re j  2 Re j 1
G ( s ) H ( s ) s  Re j   
(Re  1)(Re  1) Re  Re j Re j
j j j
Lt GH (Re j )  Lt  e  j
R   0
R 
 Lt  e
 j
  0
2
2
 Lt  e
 j
  0
2
Page 147 of 187

Number of encirclements of point -1+j0, N = -1 (counter-clockwise)

P=1
50
Sketch the Nyquist plot for the open-loop transfer function G ( s) H ( s )  and hence
s ( s  5)
50 10
G ( s) H (s)  
s ( s  5) s (1  0.2 s )
10
G ( j ) H ( j ) 
j (1  j 0.2 )
10
G ( j ) H ( j ) 
  1  (0.2 ) 2
G ( j ) H ( j )  90  tan 1 0.2
MAPPING OF C1 & C3
 G ( j ) H ( j )  G ( j ) H ( j )
0 ∞ -90˚
∞ 0 -180
MAPPING OF C2
 
R  2 2
50 50 1
G ( s ) H ( s ) s  Re j   
Re j (Re j  5) Re j  Re j R 2 e j 2
Lt GH (Re j )  Lt  e  j 2
R   0
Page 148 of 187

R 
 Lt  e  j
  0
2
 Lt  e  j
  0
2
MAPPING OF C4
 
s  Lt  e j ;    to  clockwise
  2 2
50 50 10
G ( s ) H ( s ) s  e j     e  j
 e ( e  5)  e  5  e j
j j j
Lt GH ( e j )  e  j
 0
 Lt GH ( e j )
 0
 e
j
 2
2
 e
 j
 2
C4 in s-plane is mapped into a semicircle of infinite radius with angle varying from +/2 to -/2.
Number of encirclements of point -1+j0, N = 0.
Page 149 of 187

P=0
50
Sketch the Nyquist plot for the open-loop transfer function G ( s ) H ( s )  and hence
s ( s  4)( s  1)
50 12.5
G (s) H ( s)  
s ( s  4)( s  1) s (1  0.25 s )( 1  s )
12.5
G ( j ) H ( j ) 
j (1  j 0.25 )( 1  j )
12.5
G ( j ) H ( j ) 
  1  (0.25 ) 2  1   2
G ( j ) H ( j )  90  tan 1 0.25  tan 1   180  270  tan 1 0.25  tan 1 
MAPPING OF C1 & C3
 G ( j ) H ( j )  G ( j ) H ( j )
0 ∞ -270˚
∞ 0 -270
MAPPING OF C2
 
R  2 2
50 50 1
G ( s ) H ( s ) s  Re j  j j
j
 j j j
 3 j 3
Re (Re  4)(Re  1) Re  Re  Re R e
Lt GH (Re j )  Lt  e  j 3
R   0
R 
 Lt  e  j 3 / 2  Lt  e  j / 2
  0  0
2
 Lt  e  j 3 / 2  Lt  e  j / 2
  0  0
2
MAPPING OF C3
 
  2 2
50 50 12.5
G ( s ) H ( s ) s  e j     e  j (180  )
 e j ( e j  4)( e j  1)  e j  4  e j
Page 150 of 187

Lt GH ( e j )  e  j (180  )
 0
 Lt GH ( e j )
 0
 e
 j
 2
2
 e
 j
 2
C3 in s-plane is mapped into a semicircle of infinite radius with angle varying from -/2 to +/2.
Number of encirclements of point -1+j0, N = 1 (clockwise)

P=1
Z = P + N = 1 +1 = 2; system is unstable.
5( s  3)
s ( s  1)( s  1)
5( s  3) 5 / 3(1  0.333s )
G ( s) H ( s )  
s ( s  1)( s  1) s (1  s )( 1  s )
5 / 3( 1  j 0.333 )
G ( j ) H ( j ) 
j (1  j )( 1  j )
5 / 3 1  (0.333 ) 2
G ( j ) H ( j ) 
  1 2  1  2
G ( j ) H ( j )  180  tan 1 0.333  90  tan 1   (180  tan 1  )  90  tan 1 0.333
MAPPING OF C1 & C3
Page 151 of 187

 G ( j ) H ( j )  G ( j ) H ( j )
0 ∞ -90˚
∞ 0 -180
MAPPING OF C2
 
R  2 2
5(Re j  3) 5  Re j 1
G ( s ) H ( s ) s  Re j  jj j
 j j j
 2 j 2
Re (Re  4)(Re  1) Re  Re  Re R e
Lt GH (Re j )  Lt  e  j 2
R   0
R 
 Lt  e  j
  0
2
 Lt  e  j
  0
2
MAPPING OF C4
 
  2 2
5( e j  3) 5/3
G ( s ) H ( s ) s  e j    e  j
 e ( e  1)( e  1)  e j
j j j
Lt GH ( e j )  e  j
 0
 Lt GH ( e j )
 0
 e
 j
 2
2
 e
 j
 2
Page 152 of 187


P=1
Z = P + N = 1; system is unstable.
K (1  s )
Sketch the Nyquist plot for the open-loop transfer function G ( s) H ( s )  for K = 2
( s  2)( s  3)
and check whether the system is stable for this gain. Find the range of K for the system to be stable.
K (1  s )
G (s)H (s) 
( s  2)( s  3)
2(1  j )
G ( j ) H ( j ) 
(2  j )(3  j )
2 1 2
G ( j ) H ( j ) 
4  2  9  2
G ( j ) H ( j )   tan 1   tan 1 0.5  tan 1 0.33
MAPPING OF C1 & C3
 G ( j ) H ( j )  G ( j ) H ( j )
0 1/3 0˚
∞ 0 -270˚
MAPPING OF C2
 
R  2 2
2(1  Re j )  Re j 1
G ( s ) H ( s ) s  Re j   
(Re  2)(Re  3) Re j  Re j Re j
j j
Lt GH (Re j )  Lt  e  j (180  )
R   0
Page 153 of 187

R 
 Lt  e
 j
  0
2
2
 Lt  e
 j
  0
2

To find value of  and G ( j ) H ( j ) at the meeting point on real axis: -
G ( j ) H ( j )   tan 1   tan 1 0.5  tan 1 0.33  180
Hence,  = 3.318 rad/sec
2 1 2 2 1  3.318 2
G ( j ) H ( j )    0.4
4  2  9 2 4  3.3182  9  3.318 2

P=0
To find the value of K when  = 3.318 rad/s,
K 1 2 K 1  3.318 2
G ( j ) H ( j )   3.318    0.2 K
4 2  9  2 4  3.318 2  9  3.3182
N = 0 if 0.2K < 1
K<5
Hence, the system is stable for 0 < K < 5.
5
s ( s  1)( s  2)
5 2.5
G ( s) H ( s )  
s ( s  1)( s  2) s (1  s )(1  0.5 s )
Page 154 of 187


2.5
G ( j ) H ( j ) 
j (1  j )(1  j 0.5 )
2.5
G ( j ) H ( j ) 
  1    1  (0.5 ) 2
2
 G ( j ) H ( j )  90  tan 1   tan 1 0.5

MAPPING OF C1 & C3
 G ( j ) H ( j )  G ( j ) H ( j )
0 ∞ -90˚
∞ 0 -270
To find value of  and G ( j ) H ( j ) at the meeting point on real axis: -

G ( j ) H ( j )  90  tan 1   tan 1 0.5  180
Hence,   2  1.414 rad/sec
5 5
G ( j ) H ( j )    0.833
 1   4  2
2
2  1 2  4  2
MAPPING OF C2
 
R  2 2
5 5 1
G ( s ) H ( s ) s  Re j   
Re j (Re j  1)(Re j  2) Re j  Re j  Re j R 3e j 3
Lt GH (Re j )  Lt  e  j 3
R   0
R 
 Lt  e  j 3 / 2  Lt  e  j / 2
  0  0
2
 Lt  e  j 3 / 2  Lt  e  j / 2
  0  0
2
MAPPING OF C3
 
  2 2
5 5 2.5
G ( s ) H ( s ) s  e j     e  j
 e ( e  1)( e  2)  e  2  e j
j j j j
Lt GH ( e j )  e  j (180  )
 0
 Lt GH ( e j )
 0
Page 155 of 187

 e
 j
 2
2
 e
 j
 2
Number of encirclements of point -1+j0, N = 0

P=0
Z = P + N = 0 ; system is stable.
60
Sketch the Nyquist plot for the open-loop transfer function G (s) H (s)  and
( s  1)( s  2)( s  5)
hence determine the stability of the closed loop system by Nyquist criterion.
RELATIVE STABILITY USING NYQUIST CRITERION

As the Nyquist plot gets closer to -1+j0 point, the relative stability reduces and the system tends towards
instability. In frequency domain, relative stability is characterised by gain margin and phase margin.
Gain margin (GM) is defined as the reciprocal of the magnitude of the open-loop transfer function
evaluated at the phase cross-over frequency pc.
1 1
GM   where  G ( j pc ) H ( j pc )  180 
G ( j pc ) H ( j pc ) a
The frequency at which the phase angle of the open-loop transfer function is -180˚ is called the phase
cross-over frequency pc.
Page 156 of 187

1 1
Gain Margin in dB is GM  20 log10  20 log10  20 log10 a
G ( j pc ) H ( j pc ) a
If a  1 , GM is negative, the polar plot encircles the critical point (-1+j0) and the system is unstable.
If a  1 , GM is positive, the polar plot does not encircle the critical point (-1+j0) and the system is
stable.
Phase margin (PM) is the amount of phase lag that can be introduced into the system at the gain
cross-over frequency gc to drive the system to the verge of instability.
The frequency at which the magnitude of the open-loop transfer function is 1 is called the gain cross-
over frequency gc.
The value of phase margin can be obtained from the following relation,
PM  180    G ( j gc ) H ( j gc )  180    gc
For a stable system, the phase margin is positive.
If  gc   pc , system is stable.
If  gc   pc , system is marginally stable.
If  gc   pc , system is unstable.
ADVANTAGES OF NYQUIST PLOT

1 The Nyquist plot can be used for the study of stability of system with non-minimum transfer
functions.
2 The stability analysis of a closed loop system can be easily investigated by examining the
Nyquist plot of the loop transfer function with reference to the -1+j0 point once the plot is
made.
DISDVANTAGES OF NYQUIST PLOT

1 It is not so easy to carry out the design of the controller by referring to the Nyquist plot.
Page 157 of 187

DESIGN OF COMPENSATORS USING BODE PLOT
Gc ( s ) G (s)
LEAD COMPENSATOR
Lead compensator is used to improve the transient response by pole and zero placement. Here,
compensator pole is located to the left of the compensating zero.
Zero is introduced to improve the transient response but high frequency noise is amplified. Hence a
compensative pole is also placed to reduce the noise.
The pole is located at a distance 3 to 10 times the value of the zero location.
Transfer function of lead compensator is
s  zc s 1/ T
Gc ( s )   ;  1
s  pc s  1 /  T
z
  c 1
pc
1 1
Angle contribution is always positive; hence the  
T T
name lead.
s 1
Eg:- Gc ( s )  ;
s5
Gc ( j )  tan 1   tan 1 0.2
if ω=1, Gc ( j )  33.7 ; here the angle is positive; hence LEAD.
Positive angle contributed by the lead compensator shifts the root locus towards the left in the s-plane;
this results in improvement in the transient response.
A given system is stable but its transient response is unsatisfactory. Then, the root locus must be
reshaped so that it is moved farther to the left, away from the imaginary axis. This can be achieved by
a lead compensator.
Vo ( s ) R2 R2 R1 R2Cs  R2
   
Vi ( s) R2  Z1 R  R1 R1 R2Cs  R1  R2
2
R1Cs  1
1 1 1
R1 R2C ( s  ) s s
CR1 CR1 T
  
R  R2 R  R2 1 1
R1 R2C ( s  1 ) s 1  s
R1 R2C R2 CR1 T
where T  CR1 &   R2
R1  R2
The performance of a closed-loop system can be described in terms of the following frequency domain
specifications;
1) Phase Margin PM – indicative of relative stability
2) Gain cross over frequency ωgc – indicative of speed of response (ωgc is increased, ts is decreased)
Page 158 of 187

3) Error constants Kp, Kv, Ka – indicative of steady state error

If the desired specifications are not met by gain adjustment, use compensator.
RESHAPING OF BODE PLOT BY SIMPLE GAIN ADJUSTMENT

K
Consider a system G ( s )  .
s ( s  2)
Requirements are Kv  10 and PM  60
K
K v  Lt sG ( s)  10 Hence K  20
s 0 2
With K=20, G ( s)  20 . Plot Bode plot.
s ( s  2)
PM = 24°, ωgc = 4.4rad/s
If K is reduced to 2.5 to increase PM,
PM = 59°, ωgc = 1.2rad/s, Kv = 1.25
Since gc decreases, speed of response
decreases and since Kv decreases, steady
state error increases- both transient & steady
state requirements are not met.
Here, the desired specifications are not met
by simple gain adjustment.
BODE PLOT OF LEAD COMPENSATOR

1
s
T 1  sT
Gc ( s )  
s
1 1  s T
T
Since α<1, lead compensator provides an attenuation of α; hence use suitable amplifier.
1  sT
Gc ( s ) 
1  s T
1  jT
Gc ( j ) 
1  j T
1
Corner frequencies are c1  and c 2  1
T T
 rad/s A in dB
1 A=0
  c1 
T
1 1 1
c 2  A  20 log T  20 log  T  20 log
T T 
10 A  20 log T  20 log T  20 log
1

T 
  tan 1 T  tan 1 T
Page 159 of 187

1
1 20 log
10 log 

1 1
c1  c 2 
T T
1
m 
T
Φ is maximum at  m  c1c1  1 1 1
 
T T T
At m, A  20 log T  20 log 1 1 1
 T  20 log  10 log
T   
1 1 1
m  tan 1 T  tan 1 T  tan 1  T  tan 1    T  tan 1  tan 1 
T  T  
1 1  sin m
sin m  or  
1  1  sin m
DESIGN OF LEAD COMPENSATOR - STEPS
1 The open loop gain K of the given system is determined to satisfy the requirement of steady
state error.
2 After determining the value of K, draw Bode plot of uncompensated system.
3 Measure the gain cross-over frequency ωgc, and phase margin PM of the uncompensated
system.
4 Determine the amount of phase angle to be contributed by the lead network by using the
formula given below:
m   d    
where
Φm = Maximum phase lead angle of lead compensator
 d = desired phase margin
 = phase margin of uncompensated system
 = additional phase lead (margin of safety) to compensate for shift in gain cross over
frequency due to compensation
Choose an initial choice of  as 5˚.
Note: If Φm is more than 60˚, then realize the compensator as cascade of two lead compensator
with each compensator contributing half of the required angle.
5 1  sin m
Determine the parameter α of the compensator using the following equation  
1  sin m
6 1
Locate the frequency at which the uncompensated system has a gain of 10 log dB. Select

this frequency as the new gain cross-over frequency ωgc’. Measure the PM of the
uncompensated system at ωgc’. If the difference in PMs of G(jω) at ωgc & ωgc’ is less than ε,
go to next step. Otherwise choose a larger ε.
7 1
Set m   gc ' and compute parameter T of the compensator using the equation T 
m 
Page 160 of 187

8 1  sT
Transfer function of lead compensator is Gc ( s ) 
1  s T
9 1  sT
Sketch the Bode plot of the compensated system Go ( s )   G ( s)
1  s T
Design a lead compensator for the unity feedback system whose open loop transfer function is
K . The following specifications are to be satisfied. i) PM  45 , ii) K v  15 and iii)
G ( s) 
s ( s  1)
 gc  7.5rad / sec .
K
K v  Lt sG ( s )   15
s 0 1
With K=15, G ( s )  15 . Plot Bode plot.
s( s  1)
15
G ( j ) 
j (1  j )
 rad/s A in dB
 = 0.5 A = 20log15 – 20log = 29.5dB
 = c1 = 1 A = 20log15 – 20log = 23.5dB
 = 10 A = 20log15 – 20log - 20log = -16.5dB
 = gc , A = 0 20log15 – 40log = 0; gc = 3.9 rad/s
  90  tan 1 
ω 1 2 5 10
Φ -135 -153 -169 -174
gc  90  tan 1 gc  90  tan 1 3.9  166 PM = 180  166  14 
m   d      45  14  5  36
1  sin m 1  sin 36
   0.26
1  sin m 1  sin 36
1 1
A  10 log  10 log
 5.85
 0.26
A  20 log15  40 log m  5.85dB
m   gc '  5.4rad / s
1 1
T   0.36
m  5.4 0.26
1  sT 1  0.36s
Gc ( s )  
1  s T 1  0.094 s
1  sT 1  0.36 s 15
Go ( s )   G (s)  
1  s T 1  0.094 s s (1  s )
15(1  0.36 s)
Go ( s ) 
s (1  0.094 s )(1  s)
Page 161 of 187

BODE PLOT OF COMPENSATED SYSTEM

15(1  j 0.36 )
Go ( j )  c1  1; c 2  2.78; c3  10.64;
j (1  j )(1  j 0.094 )
 rad/s A in dB
  0.5 A  20log15  20log   20log15  20 log 0.5  29.5dB
 1 A  20log15  20log   20 log15  20 log1  23.5dB
  2.78 A  20log15  20 log   20log   20log15  40 log 2.78  5.76dB
  10.64 A  20 log15  20 log   20 log   20 log 0.36
 20 log15  40 log10.64  20 log 0.36  10.64  5.9dB
At    gc ' , A  20 log15  40 log  gc ' 20 log 0.36 gc '  0  gc '  5.4rad / s
gc  90  tan gc  tan 0.36gc  tan 0.094gc
1 1 1
 90  tan 1 5.4  tan 1 0.36  5.4  tan 1 0.094  5.4  133.6
ω 1 2 5 10
Φ -121 -128 -133 -143
PM = 180  133.6  46.4
All the requirements are satisfied.
Page 162 of 187

K . The following specifications are to be satisfied. i) PM  20 and ii) K v  50 .
G(s) 
s (s  1)( s  5)
K
K v  Lt sG ( s )   50
s0 5
K  250
250 50
G ( s)  
s ( s  1)( s  5) s (1  s )(1  0.2 s )
50
Sinusoidal transfer function, G ( s ) 
j (1  j )(1  j 0.2 )
 rad/s A in dB
 = 0.5 A = 20log50 – 20log = 40dB
 = c1 = 1 A = 20log50 – 20log = 33.98dB
 = c1 = 5 A = 20log50 – 20log - 20log = 6dB
 = c1 = 20 A = 20log50 – 20log - 20log - 20log0.2 = -30.1dB
 = gc , A = 0 20log50 – 40log - 20log0.2 = 0; gc = 6.3 rad/s
   G ( j )   90  tan  1   tan  1 0.2

 0.5 1 2 3 5 10 20 ∞
 -122 -146 -175 -193 -214 -238 -253 -270
At  = pc ,  = -180;    G ( j )   90  tan  1   tan  1 0.2   180  , pc = 2.24rad/s
At  = pc = 2.24 rad/s, A = 20log50 – 40log - 20log0.2 = 26.94dB
At  = gc = 6.3 rad/s,  gc   G ( j )   90  tan  1   tan  1 0.2   223 
Phase Margin  = 180 + gc = 180 – 223 = -43˚
Since the phase margin is negative, system is unstable. A lead compensator is used to make the system
stable and to have a phase margin of 20˚.
m   d      20  43  5  68
Since the lead angle required is greater than 60˚, we have to realize the lead compensator as a cascade
of two lead compensators with each compensator providing half of the required phase lead angle.
68
m   34
2
1  sin m 1  sin 34
   0.283
1  sin m 1  sin 34
1 1
A  10 log  10 log  5.482
 0.283
A  20 log 50  40 log m  20 log 0.2m  5.483dB
m   gc '  7.8rad / s
1 1
T   0.24
m  7.8 0.283
1  sT 1  0.24 s
Gc ( s )  
1  s T 1  0.068s
Page 163 of 187

(1  sT ) 2 (1  0.24s ) 2 50
Go ( s)   G (s )  
(1  s T ) 2
(1  0.068s ) 2 s (1  s )(1  0.2 s )
50(1  0.24s ) 2
Go ( s) 
s (1  0.068s )2 (1  s )(1  0.2 s )
K . The following specifications are to be satisfied. i) PM  50 and ii) K v  20 .
G(s) 
s ( s  2) 2
Page 164 of 187

LAG COMPENSATOR
Lag compensator is used when transient response is satisfactory, steady-state error is large and relative
stability is poor.
By use of lag compensator, phase margin increases, gain-cross over frequency and band-width
decreases, steady-state error decreases.
The pole is located nearer to origin.

Transfer function of lag compensator is
s  zc s 1/ T
Gc ( s )   ;  1
s  pc s  1 /  T
z
  c 1
pc 1
1
Angle contribution is always negative; hence the  
T T
name lag.
1  1 
R2  R2 C  s  
V (s) Z 2 (s) Cs  R2 C 
Gc ( s)  o   
Vi ( s ) Z 1 ( s )  Z 2 ( s ) R  R  1  1 
1 2 ( R1  R 2 ) C  s  
Cs  ( R1  R 2 ) C 
 1   1
s  s 
 R2 C 
 
T 

   1 
 s
R  R2  1 
  T 
( 1 ) s  
R2  R  R2
 ( 1 ) R 2 C 
 R2 
R  R2
T  R2 C and   1
R2
 1
s 
Gc ( s )  
T 
 1 
 s  T 
 
Note: To nullify the attenuation 1/, it is amplified with gain .
Page 165 of 187

BODE PLOT OF LAG COMPENSATOR

1
s
T 1  sT
Gc ( s )  
s
1 1  s T
T
The lag compensator provides a dc gain of . If the dc gain is not desirable, it can be eliminated by
suitable attenuation. Let us assume that the gain  is eliminated by suitable attenuation network.
1  jT
Gc ( j ) 
1  j T
1
Corner frequencies are c1  and c 2  1
T T
 rad/s A in dB
1 A=0
  c1 
T
1 A = -20logT= -20log
c 2 
T
10 A = – 20logT + 20logT = – 20log10 + 20log10 = -20log

T
  tan 1 T  tan 1 T

10 log 
1 1
c1  c 2 
T T
1
20 log

1
m 
T 
Φ is maximum at  m  c1c1  1 1 1
 
T T T 
At m, A  20 log  T  20 log 1   T  20 log   10 log 
T 
1 1 1
m  tan 1 T  tan 1 T  tan 1  T  tan 1    T  tan 1  tan 1 
T  T  
1  1  sin m
sin m  or  
1  1  sin m
DESIGN OF LAG COMPENSATOR - STEPS
state error.
Page 166 of 187

3 Determine the phase margin PM of the uncompensated system. If the phase margin does not
satisfy the requirement, then lag compensation is required.
4 Phase margin of compensated system,  n   d  
where
 = additional phase lag (margin of safety) to compensate for shift in gain cross over frequency
due to compensation
5  gcn   n  180
Determine the new gain cross-over frequency ωgcn corresponding to  gcn from Bode plot of
uncompensated system.
6 Determine the parameter  of the compensator.
dB magnitude at ωgcn = Agcn
Agcn
  10 20
7 Determine the transfer function of lag compensator.
 gcn 1
Zero of the lag compensator, zc   . (Place the zero of the compensator arbitrarily at
10 T
1/10th of the new gain crossover frequency ωgcn)
1
Pole of the lag compensator, pc 
T
Transfer function of lag compensator, Gc ( s )  1  sT
1  s T
8 1  sT
Determine the open-loop transfer function of the compensated system Go ( s )   G( s)
1  s T
1  (1  sT )
G (s)
 (1  s T )
Gc ( s)
9 Sketch the Bode plot of the compensated system and determine the actual phase margin. If the
actual phase margin satisfies the given specification, then the design is accepted. Otherwise,
repeat the procedure by taking  as 10˚.
A unity feedback system has an open loop transfer function G ( s )  K . Design a suitable lag
s (1  2 s)
compensator so that phase margin is 40˚ and the steady state error for ramp input is less than or equal
to 0.2.
1
ess   0.2
Kv
Kv  5
K
K v  Lt sG ( s )  5
s0 1
With K=5, G ( s )  5 . Plot Bode plot.
s(1  2 s )
Page 167 of 187

5
G ( j ) 
j (1  j 2 )
 rad/s A in dB
 = 0.1 A = 20log5 – 20log = 34dB
 = c1 = ½ =0.5 A = 20log5 – 20log = 20dB
=5 A = 20log5 – 20log - 20log2 = -20dB
 = gc , A = 0 20log5 – 20log - 20log2 = 0; gc = 1.6 rad/s
  90  tan 1 2
ω 0.1 0.5 1 5 10
Φ -101 -135 -153 -174 -177
gc  90  tan 2gc  90  tan 1 2 1.6  162
1
PM = 180  162  18 
System requires a phase margin of 40˚, but the available phase margin is 18˚; hence lag compensation
should be employed to improve the phase margin.
 n   d    40  5  45
 gcn   n  180  45  180  135
gcm  90  tan 1 2 gcn  135
Hence, new gain crossover frequency,  gcn  0.5 rad/sec
 gcn
Zero of the compensator, zc   0.05
10
1
zc   0.05
T
dB magnitude at  gcn  0.5 , Agcn = 20log5 – 20log0.5 = 20dB
Agcn
  10 20  10
1
Pole of the lag compensator, pc   0.005
T
1
s
Transfer function of lag compensator, G ( s )  T   1  sT  10  1  20s
1  s T
c
1 1  200 s
s
T
1 10(1  20 s ) 5
10 (1  200 s) s(1  2s )
1/  Gc ( s) G (s)
Overall transfer function, Go ( s )  5(1  20 s )

s (1  200 s )(1  2 s )
Corner frequencies, c1 = 0.005 rad/s; c2 = 0.05 rad/s; c3 = 0.5 rad/s;
 rad/s A in dB
 = 0.001 A = 20log5 – 20log = 74dB
 = c1 = 0.005 A = 20log5 – 20log = 60dB
Page 168 of 187

 = c2 = 0.05 A = 20log5 – 20log – 20log200 = 20dB

 = c3 = 0.5 A = 20log5 – 20log - 20log200 + 20log20 = 0 dB
=1 A = 20log5 – 20log - 20log200 + 20log20 - 20log2 = -12dB
 = gc , A = 0 gc = 0.5 rad/sec
  90  tan 1 20  tan 1 200  tan 1 2

gc  90  tan 1 20 gc  tan 1 200gc  tan 1 2gc  140
Actual phase margin of compensated system, 180  gc  180  140  40
Actual phase margin of the compensated system satisfies the requirement. Hence the design is
acceptable.
A unity feedback system has an open loop transfer function G ( s )  K . Design a suitable
s( s  4)( s  80)
lag compensator so that phase margin is 33˚ and Kv = 30 sec-1.
K
K v  Lt sG ( s )   30
s0 4  80
With K=9600, G ( s )  9600 30 . Plot Bode plot.

s ( s  4)( s  80) s(1  0.25s)(1  0.0125s)
Page 169 of 187

30
G ( j ) 
j (1  j 0.25 )(1  j 0.0125 )
Corner frequencies, c1 = 4 rad/s; c2 = 80 rad/s
 rad/s A in dB
=1 A = 20log30 – 20log = 29.5dB
 = c1 = 4 A = 20log30 – 20log = 17.5dB
 = c2 = 80 A = 20log30 – 20log – 20log0.25 = -34.5dB
 = 100 A = 20log30 – 20log – 20log0.25 – 20log0.0125= -40.3dB
 = gc , A = 0 20log30 – 20log - 20log0.25 = 0;
gc = 10.95 rad/s
  90  tan 1 0.25  tan 1 0.0125

ω 1 3 10 30 100 300
Φ -105 -129 -165 -193 -229 -254
gc  90  tan 0.25 gc   tan 1 0.0125gc  168
1
PM = 180  168  12 
System requires a phase margin of 30˚, but the available phase margin is 12˚; hence lag compensation
should be employed to improve the phase margin.
 n   d    33  5  38
 gcn   n  180  38  180  142
gcn  90  tan 1 0.25gcn  tan 1 0.0125gcn  142
 gcn
10
1
zc   0.46
T
dB magnitude at  gcn  4.6 , A = 20log30 – 20log – 20log0.25 = 15dB
Agcn
  10 20  5.6
1
Pole of the lag compensator, pc   0.08
T
1
s
Transfer function of lag compensator, G ( s )  T   1  sT  5.6  1  2.2 s
1  s T
c
1 1  12.5s
s
T
1 5.6(1  2.2s ) 30
5.6 (1  12.5s ) s(1  0.25s)(1  0.0125s )
1/  Gc ( s) G (s)
Overall transfer function, Go ( s )  30(1  2.2 s)

s (1  12.5s )(1  0.25s )(1  0.0125s)
Corner frequencies, c1 = 0.08 rad/s; c2 = 0.46 rad/s; c3 = 4 rad/s; c2 = 80 rad/s
Page 170 of 187

 rad/s A in dB
 = 0.05 A = 20log30 – 20log = 55dB
 = c1 = 0.08 A = 20log30 – 20log = 51dB
 = c2 = 0.46 A = 20log30 – 20log – 20log12.5 = 21dB
 = c2 = 4 A = 20log30 – 20log – 20log12.5 +20log2.2 = 2.4dB
 = 80 A = 20log30 – 20log – 20log12.5 +20log2.2 – 20log0.25 = -49dB
 = 100 A = 20log30 – 20log – 20log12.5 +20log2.2 – 20log0.25 –
20log0.0125 = -55dB
 = gc , A = 0 A = 20log30 – 20log – 20log12.5 +20log2.2 – 20log0.25 = 0;
gc = 4.6 rad/s
gc  90  tan 1 2.2 gc  tan 1 12.5 gc  tan 1 0.25 gc  tan 1 0.0125 gc  147
Actual phase margin of compensated system, 180  gc  180  147  33
Actual phase margin of the compensated system satisfies the requirement. Hence the design is
acceptable.
A unity feedback system has an open loop transfer function G ( s )  K . Design a suitable lag
s ( s  5) 2
compensator so that phase margin ≥ 70˚ and Kv = 10 sec-1.
Page 171 of 187

LAG – LEAD COMPENSATOR

When a lag-lead network is used, phase lag occurs in the low frequency region and phase lead occurs
in the high frequency region.
1
R1 
1 sC1 R1
Z1 ( s )  R1 / /  
sC1 R  1 sR1C1  1
1
sC1
1 sR C  1
Z 2 ( s )  R2   2 2
sC 2 sC 2
sR2 C2  1
V (s) Z 2 (s) sC 2 ( sR2 C 2  1)( sR1C1  1)
G (s)  o   
Vi ( s ) Z1 ( s )  Z 2 ( s ) R sR C  1 sR1C 2  ( sR2 C 2  1)( sR1C1  1)
1
 2 2
sR1C1  1 sC2
( sR2 C 2  1)( sR1C1  1)


s R1 R2 C1C 2  s ( R1C1  R2 C 2  R1C 2 )  1
2
Dividing both numerator and denominator by R1R2C1C2,
 1  1 
s  s  
 R1C1   R2 C 2 
G (s) 
 1 1 1  1
s2  s    
 R1C1 R2 C 2 R2 C1  R1 R2 C1C 2
The transfer function of lag-lead compensator is given by
 1  1  1  1
 s   s    s   s  
T T T T
Gc ( s )   1  2 
  1  2 
 1  1   1 1  1
s s   s  s  
2
  T1   T2  
 1 T  T2   T1T2
Comparing,
1 1 1 1 1 1 1
T1  R1C1 T2  R2 C 2     
 T1  T2 R1C1 R2 C 2 R2 C1  T1T2 R1 R2 C1C 2
Hence,
  1
Page 172 of 187

0  1  1
Pole-zero plot of lag-lead compensator is given by
1 1 1 1
   
 T2 T2 T1  T1
DESIGN OF LAG-LEAD COMPENSATOR - STEPS

state error.
3 Determine the phase margin PM of the uncompensated system. If the phase margin is not
satisfactory, then compensation is required.
4 Phase margin of compensated system,  n   d  
where
 = additional phase lag (margin of safety) to compensate for shift in gain cross over frequency
due to compensation
5  gcn   n  180
Determine the new gain cross-over frequency ωgcn corresponding to  gcn from Bode plot of
uncompensated system.
6 Choose the gain cross over frequency of lag compensator ωgcl > ωgcn.
7 Determine the parameter  of the compensator.
dB magnitude at ωgcl = Agcl
Agcl
  10 20
8 Determine the transfer function of lag compensator.
 gcl 1
Zero of the lag compensator, zc   . (Place the zero of the compensator arbitrarily at
10 T1
th
1/10 of the new gain crossover frequency ωgcl)
1
Pole of the lag compensator, pc 
 T1
1
s
Transfer function of lag compensator, G ( s )  T 1  sT1
1

1  s  T1
c1
1
s
 T1
9 Determine the transfer function of lead compensator.
1
Take  

1
Locate the frequency ωm at which the uncompensated system has a gain of 10 log dB.

Page 173 of 187

1
T2 
m 
1
s
Transfer function of lead compensator is G ( s )  T 1  sT2
2

1  s T2
c2
1
s
 T2
10 Determine the open-loop transfer function of the compensated system
1  sT1 1  sT2
Go ( s )   G ( s ) since   1
1  s T1 1  sT2
(1  sT1 )(1  sT2 )
G (s)
(1  s  T1 )(1  s T2 )
11 Sketch the Bode plot of the compensated system and determine the actual phase margin. If the
actual phase margin satisfies the given specification, then the design is accepted. Otherwise,
repeat the procedure by taking  as 10˚.
A unity feedback system has an open loop transfer function G ( s )  K . Design a suitable
s( s  3)(s  6)
lag-lead compensator so that phase margin is 35˚ and Kv = 80 sec-1.
K
K v  Lt sG ( s )   80
s0 3 6
K = 1440
With K=1440, G ( s )  1440 80 . Plot Bode plot.

s( s  3)( s  6) s(1  0.33s)(1  0.167 s )
80
G ( j ) 
j (1  j 0.33 )(1  j 0.167 )
Corner frequencies, c1 = 3 rad/s; c2 = 6 rad/s
 rad/s A in dB
=1 A = 20log80 – 20log = 38dB
 = c1 = 3 A = 20log80 – 20log = 28.5dB
 = c2 = 6 A = 20log80 – 20log – 20log0.33 = 16.6dB
 = 20 A = 20log80 – 20log – 20log0.33 – 20log0.167= -14.8dB
 = gc , A = 0 20log80 – 20log – 20log0.33 – 20log0.167 = 0;
gc = 11.3 rad/s
  90  tan 1 0.33  tan 1 0.167

ω 1 3 10 30 100 300
Φ -117 -161 -222 -253 -264 -268
gc  90  tan 1 0.33 gc   tan 1 0.167 gc  227
PM = 180  227   46 
To determine the transfer function of lag section:-
New phase margin,  n   d    35  5  40
Page 174 of 187

 gcn   n  180  40  180  140

gcn  90  tan 1 0.33 gcn  tan 1 0.167gcn  140
Let gcl = gain cross over frequency of lag compensator
Choose such that gcl > gcn .
Let gcl = 4 rad/sec
 gcl
10
1
zc   0.4 ; T = 2.5 sec
T
dB magnitude at  gcl  4 rad/s, Agcl = 20log80 – 20log – 20log0.33 = 23.6dB
Agcl
  10 20  15
1 0.4
Pole of the lag compensator, pc    0.027 ; T = 37
 T 15
1
s
Transfer function of lag compensator, G ( s)  T   1  sT  15  1  2.5s
1  s T
1
1 1  37 s
s
T
To determine the transfer function of lead section:-
1 1
    0.067
 15
1 1
A  10 log  10 log 11.7
 0.067
A  20 log 80  20 log m  20 log 0.33m  20 log 0.167m  11.7 dB
m   gc '  17.7 rad / s
1 1
T2    0.22
m  17.7 0.067
 T2  0.067  0.22  0.015
1  sT2 1  0.22 s
G2 ( s )  
1  s T2 1  0.015s
Transfer function of the lag lead compensator is
(1  2.5s )(1  0.22 s)
Gc ( s ) 
(1  37 s )(1  0.015s )
Transfer function of compensated system is
(1  2.5s )(1  0.22 s ) 80
Go ( s )  
(1  37 s )(1  0.015s ) s (1  0.33s)(1  0.167 s)
A unity feedback system has an open loop transfer function G ( s)  K . Design a

s (1  2 s )(1  0.5s )
suitable lag-lead compensator so that phase margin ≥ 55˚ and Kv = 30 sec-1.
Page 175 of 187

M CIRCLES (CONSTANT MAGNITUDE LOCI)

C ( s) G ( s)
M ( s)  
R( s ) 1  G ( s)
Let G ( j )  X  jY . Hence,
G ( j ) X  jY
M ( j )  
1  G ( j ) 1  X  jY
Let M be the magnitude.
X  jY
M 
1  X  jY
X 2 Y2
M2 
(1  X )2  Y 2
Rearranging,
2 2
 M2   M 
 X  1  M 2   Y  1  M 2 
2
   
This represents the equation of a circle with the centre at  M 2 , 0  and having radius M 2 .
2
1  M  1 M
For a particular value of M, a circle is obtained. For various values of M, we get a family of circles.
These circles are known as constant magnitude loci or M circles.
N CIRCLES (CONSTANT PHASE LOCI)

G ( j ) X  jY
M ( j )   
1  G ( j ) 1  X  jY
Y Y
  tan 1  tan 1
X 1 X
 Y Y 
tan   N  tan  tan 1  tan 1
 X 1  X 
Rearranging,
Page 176 of 187

2 2
 1  1  1 1
 X  2   Y  2 N   4  (2 N ) 2
This represents the equation of a circle with the centre at  1 , 1 2  and having radius 1  1 2
 2 (2 N )  4 (2 N )
. For a particular value of N, a circle is obtained. For various values of N, we get a family of circles.
These circles are known as constant phase loci or N circles.
NICHOLS CHART
When we transform M and N circles to log magnitude and phase angle coordinates, the resultant plot is
known as Nichols Chart. The critical point (-1, j0) point is mapped to the Nichols chart as the point
(0dB, -180˚). The Nichols chart contains curves of constant closed-loop magnitude and phase angle.
The Nichols chart is symmetrical about the -180˚ axis. The M and N loci repeat for every 360˚ and there
is symmetry at every 180˚ interval. The M loci are centered about the critical point (0dB, -180˚).
If we superimpose the gain-phase plot of an open-loop transfer function on Nichols chart, we get very
easily the closed-loop frequency response. Nichols chart gives the points of intersection of the gain-
phase plot of an open loop transfer function.
Nichols chart is used for the determination of the following: -
i) The complete closed-loop frequency response.
ii) Resonant peak of the closed loop system
iii) Resonant frequency of the closed loop system
iv) Bandwidth of the closed loop system
v) Phase margin
Page 177 of 187

vi) Gain margin
Page 178 of 187

Module 5

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Module 5

Uploaded by

Copyright:

Available Formats

EET302 Linear Control System Lecture Notes by T.G.

Sanish Kumar, EED, GEC Thrissur

NYQUIST STABILITY CRITERION

Page 139 of 187

Page 140 of 187

Principle of argument or mapping theorem

In equation form, the principle of argument is stated as

For Z = 4, P = 2, N = 2 → 2 clockwise encirclements

Page 141 of 187

Page 142 of 187

For stability of a closed-loop system, the zeros of the characteristic equation

Since G ( s ) H ( s )   1 , it is convenient to choose the

SELECTION OF NYQUIST CONTOUR

Page 143 of 187

Page 144 of 187

NYQUIST STABILITY CRITERION

If P = number of poles of G(s)H(s) or F(s) in the right-half of the s-plane

If Z = P + N  0, closed loop system is unstable.

Page 145 of 187

G ( j ) H ( j )   tan 1   tan 1 0.5

C2 in s-plane is mapped into a circular arc of radius 0 at origin in the GH-plane.

Page 146 of 187

Meeting point on positive real axis,   0 ; G ( j ) H ( j )  25

Point -1+j0 is not encircled by Nyquist plot. Hence, N = 0.

s2 2(1  0.5 s )

C2 in s-plane is mapped into a circular arc of radius 0 at origin in the GH-plane.

Page 147 of 187

Number of encirclements of point -1+j0, N = -1 (counter-clockwise)

Page 148 of 187

C2 in s-plane is mapped into a circular arc of radius 0 at origin in the GH-plane.

Number of encirclements of point -1+j0, N = 0.

Page 149 of 187

C2 in s-plane is mapped into a circular arc of radius 0 at origin in the GH-plane.

Page 150 of 187

Number of encirclements of point -1+j0, N = 1 (clockwise)

Page 151 of 187

C2 in s-plane is mapped into a circular arc of radius 0 at origin in the GH-plane.

Page 152 of 187

Number of encirclements of point -1+j0, N = 0.

Page 153 of 187

C2 in s-plane is mapped into a circular arc of radius 0 at origin in the GH-plane.

Number of encirclements of point -1+j0, N = 0.

Page 154 of 187

Open loop poles in right side of s–plane = P = 0

 G ( j ) H ( j )  90  tan 1   tan 1 0.5

To find value of  and G ( j ) H ( j ) at the meeting point on real axis: -

C2 in s-plane is mapped into a circular arc of radius 0 at origin in the GH-plane.

Page 155 of 187

Number of encirclements of point -1+j0, N = 0

RELATIVE STABILITY USING NYQUIST CRITERION

Page 156 of 187

ADVANTAGES OF NYQUIST PLOT

DISDVANTAGES OF NYQUIST PLOT

Page 157 of 187

DESIGN OF COMPENSATORS USING BODE PLOT

Page 158 of 187

3) Error constants Kp, Kv, Ka – indicative of steady state error

RESHAPING OF BODE PLOT BY SIMPLE GAIN ADJUSTMENT

BODE PLOT OF LEAD COMPENSATOR

Page 159 of 187

Page 160 of 187

Page 161 of 187

BODE PLOT OF COMPENSATED SYSTEM