Chapter 3

Stability Analysis

LEARNING OBJECTIVES

After reading this chapter, you will be able to understand:

• Stability • Break away point
• Routh–Hurwitz criterion • Frequency response analysis
• Auxiliary equation • Frequency-Domine specifications
• Root locus • Resonant peak
• Angle and magnitude conditions • Polar plot

intRoduCtion to stABiLity Unstable: A system is unstable if its response is unbounded with a

bounded input applied.
Stability
A linear time-invariant system is stable if the output of the system Stability and Poles
is bounded for a bounded input and the output of the system tends
The system poles that are in the left half plane yield either pure
towards zero in the absence of the input.
exponential decay or damped sinusoidal natural response, which is
Stability is classified as follows:
the necessary condition for a system to be stable.
1. Absolute stability
Note 1: Stable system have closed loop transfer function with
2. Conditional stability
poles only in the left half plane.
3. Marginal stability
4. Unstable Note 2: Unstable systems have loop transfer function with at least
one pole in the right half plane or poles of multiplicity
Absolute Stability greater than 1 on the imaginary axis.
A system is absolutely stable with respect to a parameter, if the Note 3: Marginally stable system have closed loop transfer func-
system is stable for all values of that parameter. tion with only imaginary poles of multiplicity 1 and poles
in the left half plane.
Conditional Stability
A system in conditionally stable with respect to a parameter, if the Necessary Conditions for Stability
system is stable for only certain bounded ranges of values of this 1. Positiveness of the coefficients of characteristic equation is
parameter. necessary as well as sufficient condition for stability of first
and second-order system.
Marginal Stability 2. Positiveness and existence of the all coefficients of the char-
A system is marginally stable if the natural response of the system acteristic equation is necessary condition for stability of the
neither decays nor grows but remains constant or oscillates as time system.
approaches infinity.
 Chapter 3 Stability Analysis | 3.1015

Note: Roots with negative real part indicates all positive Note: If there is a single element zero in s’ row, it is consid-
coefficients in characteristic equation but all positive ered as row of all zeros.
coefficients does not indicate proofs with negative
Special case 2: When all the elements in one row of Routh’s
real part in the characteristic equation.
tabulation are zeros before the tabulation is properly termi-
nated, it indicates the following:
Routh–Hurwitz Criterion (i) There are symmetrically located roots in S-plane
Routh–Hurwitz Criterion gives the necessary and sufficient (ii) Pair of real roots with opposite signs and/or pair of
condition for all roots of polynomial to lie in the left half conjugate roots on the imaginary axis and/or complex
of the S-plane, without actually solving for the roots of the conjugate roots forming quadrates in the S-plane.
equation.
The characteristic equation of the nth-order system is Auxiliary Equation
D(s) = a0s + a1s
n n–1
+ a2s n–2
+ … + an – 1­s + an = 0 The polynomial formed by the coefficients of the row just
above the row of zeros in the Routh array is called auxiliary
Routh Array equation [A(s) = 0].
Sn a0 a2 a4 a6 Note 1: The order of the auxiliary equation is always even
• • • •

Sn–1 a1 a3 a5 a7 Note 2: The roots of the auxiliary equation also satisfy the
• • • •
original characteristic equation.
Sn–2 b1 b2 b3 b4 Note 3: Break down in the Routh table due to zero row is
Sn–3 c1 c2 c3 overcome by replacing the row of zeros with first
Sn–4 d1 d2 ⎛ dA( s) ⎞
derivative of auxiliary equation ⎜ ⎟ with
. respect to ‘s’. ⎝ ds ⎠
.
Solved Examples
S2 e1 an
S1 f1 Example 1: If a system transfer function has some poles
lying on the imaginary axis, it is
S0 an
(A) Unconditionally stable
The coefficients b1, b2 … are evaluated as follows (B) Conditionally stable
(C) Unstable
b1 = (a1a2 – a0a3)/a1;
(D) Marginally stable
b2 = (a1a4 – a0a5)/a1;
Solution: (D)
b3 = (a1a6 – a0a7)/a1 … When the poles are on imaginary axis, system is marginally
stable.
This process will continue till we get a zero as the least
coefficient in the third row. Similarly the coefficients of the Example 2: System has some roots with real parts equal to
other rows are also evaluated. zero, but none with positive real part is
The roots of the characteristic equation are all in the left (A) Absolutely unstable (B) Absolutely stable
half of S-plane if all the Coefficients of the first column of (C) Relatively stable (D) Marginally stable
the Routh’s tabulation are of the same sign.
Solution: (D)
The number of changes of signs in the elements of the
Marginally stable
first column equals the number of roots with positive real
parts or in the right half of S-plane. Example 3: Closed loop stability implies that 1 + G(s)H(s)
has only _________ in the left half of the S-plane
Special case 1: When the first term in any row of the Routh
(A) Poles (B) Zeros
array is zero while rest of the row has at least one non-zero
(C) Poles and zeros (D) Poles or zeros
term.
In this case, if zero appears as the first element of a row, Solution: (B)
the elements in the next row will all becomes infinite, to Zeros of characteristic equation are poles of the transfer
overcome this problem we replace the zero element by an function
arbitrary small positive number ‘e’ and then proceed with Example 4: None of the poles of a linear control system lie
Routh’s tabulation. in the right half of S-plane. For a bounded input the output
Finally substitute the value of e = 0 and find the values of this system
of the elements of the array which are functions of ‘e’. The (A) Could be bounded (B) Always tends to zero
resultant Routh’s array is analysed as usual. (C) Is always bounded (D) None of the above
3.1016 | Control Systems

Solution: (D) For the system to be stable, all the coefficients of ‘s’ in the
Poles are not on the right half indicates they can be on imag- characteristic equation must be positive
inary axis, so stability cannot be justified
1 + K > 0   5 – 4K > 0
Example 5: For the equation s3 – 4s2 + s + 5 = 0, the num-
K > – 1   – 4K > – 5
ber of roots in the left half of S-plane will be
(A) Zero (B) One 5
K<
(C) Two (D) Three 4
5
Solution: (B) Actual ranges of ‘K’ is –1 < K <
Routh array for s3 – 4s2 + s + 5 = 0 4
5
s3 1 1 Given K ≥ 0; 0 ≤ K < .
4 
s 2
–4 5
s 1
2.25 Example 8: The open loop transfer function of a unity
s° 5 feedback system is given below
Sign changes in first column of Routh array are 2(1 → K ( s + 4)
G(s) =
–4 → 1) poles on left half = 3 – 2 = 1 ( s + 1)( s + 2)

Example 6: The number of roots of the equation 2s4 + s3 + The range of positive values of ‘K’ for which the closed
5s + 6 = 0 that lie in the right half of S-plane is loop system will remain stable is
2
(A) Zero (B) One (A) 2 < K < 3 (B) <K<3
4
(C) Two (D) Four
2
Solution: (C) (C) 0 < K < ∞ (D) <K< ∞
4
Routh array for 2s4 + s3 + 3s2 + 5s + 6 = 0
s4 2 3 6 Solution: (C)
s3 1 5 Closed loop transfer function
s2 –7 6 G ( s) K ( s + 4)
= = 2
41 1+ G ( s) s + (3 + K ) s + ( 2 + 4 K )
s1
7
Characteristic equation of the system s2 + (3 + K)s +
s° 6 (2 + 4K) = 0
41
No. of sign changes in Routh array are 2 (1 → – 7 → ) Condition for stability is that all coefficients of ‘s’ must
No. of poles on the right half = 2 7
be greater than zero in characteristic equation
Example 7: For what range of K is the following system is 3 + K > 0   2 + 4K > 0
asymptotically stable; assume K ≥ 0
K > – 3   4K > – 2
R(S) s−4 C(S) 2
+ K>–
– s+5
4
2
\ System is stable for all value of K > –
K 4
\ Range of positive values of ‘k’ for stability is 0 < k < ∞
5 4
(A) 0 ≤ K < (B) 0 ≤ K < Example 9: A certain closed loop system with unity feed-
4 5
back has the following transfer function given by G(s) =
4 5
(C) K > – (D) K > – k
5 4 with the gain set at the ultimate value, the
s( s + 2)( s + 4)
Solution: (A) system will oscillate at an angular frequency of
Given system transfer function (A) 2 rad/sec (B) 4 rad/sec
K ( s − 4) (C) 8 rad/sec (D) 2 2 rad/sec
=
(1 + K ) s + (5 − 4 K ) Solution: (D)
Characteristic equation of the system is Characteristic equation of the system is
(1 + K) s + (5 – 4K) = 0 s3 + 6s2 + 8s + K = 0
 Chapter 3 Stability Analysis | 3.1017

System will oscillate when it is marginally stable/ from Note: Open loop gain ‘K’ corresponding to any point on
Routh array root locus can be calculated using the equation.
s3 1 8
Product of lengh of vectors from
s 2
6 K open loop poles to the point
K=
48 − K Productt of lenght of vectors form
s1 open loop zeros to the point
6
s0 K
Rules for Construction of Root Locus
System is marginally stable if 48 – K = 0 ⇒ K = 48
1. The root locus is symmetric about origin
Then Auxiliary equation is 6s2 + 48 = 0
2. Number of branches in a root locus is equal to either
s2 = – 8 the number of poles (n) or the number of zeros (m)
whichever is greater. Each branch of root locus starts
⇒ s=j2 2
form open poles (Assuming number of poles is great-
Oscillation frequency = 2 2 rad/sec er than zero) corresponding to K = 0 and terminates at
either a finite open loop zero or infinity correspond-
ing to K = ∞. ‘n’ number of branches will terminate to
Root Locus finite open loop zeros and remaining branches of root
The root locus is basically the technique of finding the locus locus (n – m) will terminate to infinity.
of roots as a single gain is changed, by solving for the roots B = P if P > Z ⇒ P – Z branches will terminate at ∞
of the characteristic equation, at each gain. B = Z if Z > P ⇒ Z – P branches will terminate at ∞
The gain that is to be varied will be open loop gain. Note P = Number of poles, Z = number of zeros
this does not mean the gain of the open loop system that is B = Number of branches of root locus
typically fixed: this refers to cascading a controller in the 3. A section of real axis lies on root locus if the total
forward path. Using the root locus method the control sys- number of open loops poles plus zeros to the right of
tem engineer can predict the effect of varying gain on the that section is odd.
open loop poles or what effect will be caused by adding 4. The angle of asymptotes and centroid:
open loop poles or open loop zeros. If P > Z, P – Z number of branches will terminate at ∞
along straight line (asymptotes) making angle with
Angle and Magnitude Conditions real axis given by
Consider the following general system 180 ( 2q + 1)
fA = ; (q = 0, 1, 2, 3, … (P – Z – 1))
C (s) G (s) P−Z
= 180 ( 2q + 1)
R (s) 1 + G (s) H (s) If Z > P ⇒ fA = ; (q = 0, 1, 2, 3, …
(Z – P–1)) Z−P
The characteristic equation of the system is obtained by
­setting the denominator of the closed loop system to zero The point of intersection of the asymptotes with the
as follows. real axis is called centroid denoted by ‘s’
1 + G(s) H(s) = 0 Sum of real part of pole −
Sum of real part of zeros
\ G(s) H(s) = –1 Centroid (s) =
P−Z
Since complex variable has both an angle and a magnitude, 5. Breakaway/in point:
we can split the above equation into two separate equations A point on root locus where multiple poles/zeros
as follows. ­exist is known as breakaway/ in point.
∠G(s) H(s) = ± 180° (2K + 1) (K = 0, 1, 2, …) The breakaway or breakin point is given by the roots
 Angle condition dK
of the equation = 0, where ‘K’ is obtained form
ds
G ( s ) H ( s ) = 1   Magnitude condition 1 + KG(S) H(S) = 0
The values of ‘s’ that satisfy the angle and magnitude condi- Note 1: Breakaway point exists if there is a root ­locus
tions are the roots of the characteristic equation (The closed on real axis between two adjacent poles.
loop poles). ONLY these values will be the roots. As we Note 2: Breakin point exists if there is a root locus
vary the gain, these values of ‘s’ that satisfy both conditions on real axis between two adjacent zeros.
will change. The resulting collection of point in S-plane are Note 3: Breakin point exists if there is a zero on real
called root locus. axis and left to that there is no root loci or
poles or zeros.
3.1018 | Control Systems

6. The angle of departure/arrival: Equating the imaginary part to zero

The angle of departure or angle of arrival is given by
w = ± 2.5 rad/sec.
Angle of departure = 180 – f
Angle of arrival = 180 + f The root locus is drawn as shown in the figure
where f =
Im(s)
⎡Sum of angles of vectors Sum of angles of vectors ⎤
⎢ to the compplex pole/zero from − to the complex pole/zero from ⎥
⎢ ⎥ –0.784
⎢⎣other poles other zeros. ⎥⎦ w = 2.5 rad/sec

7. The intersection of root locus with the imaginary 0

–3 –2 Re(s)
axis can be determined by the use of Routh criterion
[­finding poles on imaginary axis] –1.667
Example 10: Plot the root locus for a transfer function
K
G(s) =
s( s + 2) ( s + 3)

Solution: The number of poles = 3 K ( s + 2)

Example 11: G(s) = , H(s) = 1
The poles are at s = 0, s = – 2 and s = – 3 s 2 + 2s + 3
Im
Sketch the root locus
Solution: No of branches of root locus = 2
The poles are at s = – 1± j 2
–3 –2 0 Re(s) The zero is at s = – 2
–1+j 2 Im(s)

q1
Break-away Point
–2 –1 Re(s)
d 3
( s + 5s 2 + 6 s) = 0 q2
ds 
3s2 + 10s + 6 = 0 –1–j 2
s = – 0.784 and s = –2.549
The root locus starts from the conjugate poles and break in
s = –2.549 does not lie on the root locus on the real axis between –2 and –∞. One root locus ends in
s = –2, the other ends at s = –∞.
Asymptotes
q1 = ±
180
= ± 60° Asymptote
3  180
3 × 180 q1 = ± = ± 180
q3 = ± = ±180° 2 −1
3 

Centroid Angle of Departure

(0 − 2 − 3) − (0) ⎛ 2⎞
s= = –1.667 = 180 – ⎜ 90 − tan ⎟ = 145°
3−0 ⎝ 1 ⎠

Imaginary Axis Cross-over Break-away Point

K
G(jw) = dG ( s)
jω ( jω + 2) ( jω + 3)  =0
ds 
⎡ −5ω 2 (6ω − ω 3 ) ⎤
G(jw) = K ⎢ − j 3 2⎥ s = – 3.73
⎣ 25ω + (6ω − ω ) 25ω + (6ω − ω ) ⎦
4 3 2 4

 Chapter 3 Stability Analysis | 3.1019

The root locus is drawn in the following figure. Example 15: Figure shown below gives root locus of the
open loop transfer function G(s) H(s) of a system.
Im(s)
Consider the following inference drawn from the figure.
o
145
(1) It has no zero.
j 2 (2) It is a stable system.
–3.73 (3) It is a second-order system.
K=∞ K=∞
–2 –1 Re(s) Which of these inferences are correct?

–j 2
jw

s-plane
1. N – P = Z
Here P = 1
x x s
If K > 1, N = 1,
Z = 0, then the closed loop system is stable
2. If K < 1, N = 0
Z ≠ 0, then the closed loop system is unstable.
(A) 1, 2, and 3 (B) 1 and 2
K (C) 2 and 3 (D) 1 and 3
Example 13: Given G(s) H(s) = , the point
s ( s + 2 ) ( s + 5) Two poles are terminated to infinity indicates that there are
of intersection of the asymptotes of the root locus with the no zeros.
real axis is Two poles indicates the order of the system as ‘2’.
(A) 0   (B) – 2   (C) – 2.3   (D) – 3.5
Example 16: The characteristic equation of a unity-feed-
Solution: (C) back control system is given by S3 + AS2 + S + B = 0.
No. of poles (P) = 3 (0, – 2, – 5) Consider the following statements in this regard.
No. of zeros (Z) = 0
No. of asymptotes = 3 1. For a given value of B, all the root-locus branches
Centroid (Intersection of the asymptotes) will terminate at infinity for the variable ‘A’ in the
positive direction.
Σ Real part of all poles − Σ Real part of all zero 2. For a given value of B, only one root locus branch
=
P−Z  will terminate at infinity for the variable K, in the
0 − 2 − 5 − 0 −7 positive direction.
= = = –2.33. 3. For a given value, of A, all the root locus branches
3 3
will terminate at infinity for the variable ‘B’ in the
Example 14: The open loop transfer function of a unity positive direction.
feedback control system is given by Of these statement
(A) 1 and 3 are correct.
K ( s + 2) (B) 2 and 3 are correct.
G ( s) =
(
s s 2 + 2s + 1 ) (C) Only 2 is correct.
(D) Only 1 is correct.
The centroid and angles of root locus are, respectively,
2 Solution: (B).
(A) − and +60°, –60° (B) –2 and +90°, –90°
3 Example 17: The root locus of a unity feedback system is
(C) Zero and +90°, –90° (D) – 2 and +60°, –60°
shown in the following figure. The open loop transfer func-
Solution: (C) tion is given by
No. of poles = 3 (0, – 1, – 1)
Imaginary
No. of zeros = 1 (– 2)
No. of asymptotes = 2
Angle of asymptotes = +90° and –90°
Σ Real part of all poles − Σ Real part of all zero Real
Centroid = 0 x
2
K=0
=
(0 − 1 − 1) − ( −2) = −2 + 2 = 0.
2 2
3.1020 | Control Systems

K K (s + 4) Solution: (B)
(A) (B)
s ( s + 2) ( s + 4 ) s ( s + 3) ( s + 5) When the value of ‘K’ is increasing the location of poles
will tend to right-hand side of the S-plane when K > 4
K (s + 4) Ks → When K = 4 characteristic equation s2 + 3 = 0
(C) (D)
s ( s + 3) ( ) ( s + 5)
s + 3 s=±j 3=
(Poles are on the imaginary axis)
Solution: (C)
→ When K = 10 characteristic equation s2 – 6s + 3 = 0
The given root locus indicates that the open loop transfer
(s – 3)2 = 0
function has 2 poles and one zero.
s = + 3, + 3
One pole is at origin and another pole location is on the
(Poles are on RHS and equal)
right side to the zero.
→ When K > 10 poles are on RHS but not equal
So option ‘C’ is correct.
\ Option ‘B’ is the suitable root locus.
Example 18: The closed loop transfer function of a feed-
back system is given by Example 19: A control system has
C (S ) K K ( s + 5)
= G(s) H(s) =
R (S ) s + (4 − K ) s + 3
2 ( s + 2) ( s + 3) 

Which of the following diagrams represents a root locus of The break away and break in points are located,
the system for K > 0? ­respectively, at
(A) –2 and –1 (B) –1.589 and –7.5
(A) K→ ∞ (C) –2.55 and –7.5 (D) –1.5 and –6.89
jw
Solution: (C)
Characteristic equation
1 + G(s) H(s) = 0
x x s
K=0 K=0 − ( s + 2) ( s + 3)
s 2 + 5s + 6
⇒ K= =
K→ ∞ ( s + 5) s+5

dK
Breakaway or breakin points are roots of =0
(B) jw ds
(
dK ( 2 s + 5 ) ( s + 5 ) − s + 5s + 6
=
2
)
= 0
( s + 5)
2
ds
K→∞ K→∞
x x s 2s2 + 15s + 25 – s2 – 5s – 6 = 0
K=0 K=0 s2 + 10s – 19 = 0
s = –2.55, –7.449
Breakaway point is –2.55, breakin point –7.449.
(C) jw
Example 20: A transfer function G(s) has type pole zero
plot as shown in figure. Given that the steady-state gain is 3,
K→∞ K→∞
the transfer function G(s) will be
x s
K=0 K=0 x jw

s
–2 –1

(D) jw x

2 ( s + 1) 5 ( s + 1)
K→∞ (A) (B)
x x s s + 4s + 5
2
s + 4s + 5
2

15 ( s + 1) 15 ( s + 1)
K=0 K=0
(C) (D)
s + 4s + 5
2
( s + 2 )2
 Chapter 3 Stability Analysis | 3.1021

Solution: (C) Frequency-domain Specifications

Form the given pole zero plot transfer function has Resonant Peak (Mr)
1. Zero is at (–1, 0)
2. Complex poles The resonant peak Mr is the maximum value of M ( jω ) .
Only for option ‘C’ steady-state gain is ‘3’. The magnitude Mr gives indication on the relative stability
of a stable closed loop system.
15 ( s + 1) 15 For second-order system,
Lt G ( s ) = Lt = = 3.
s→0 s→0 s + 4s + 5
2
5 1 1
Mr = for ξ ≤
2ξ 1 − ξ 2
2

Frequency Response Analysis Mr = 1 for ξ >
1
The magnitude and phase relationship between the sinusoi- 2
dal input and the steady state output of a system is termed Note: A large Mr corresponds to a large maximum over
as frequency response. In linear time-invariant systems, the short of the step response.
frequency response is independent of the amplitude and
phase of the input signal Resonant Frequency (wr )
when the input of a linear time-invariant system is sinu- The resonant frequency wr is the frequency at which the
soidal with amplitude A and frequency w0. peak resonance Mr occurs.
1
r(t) = Asinw0t. For second-order system, wr = wn 1 − 2ξ 2 for ξ ≤
2
The steady-state output of a system y(t) will be a sinusoi- 1
wr = 0 for ξ >
dal with the same frequency w0 but possibly with different 2
amplitude and phase
Bandwidth (BW)
y(t) = Bsin(w0t + f)
The bandwidth (BW) is the frequency at which M ( jω )
B G ( jω ) drops to 70.7% of or 3 dB down from its zero frequency
= Transfer function = = M ( jω ) value.
A 1 + G ( jω ) H ( jω )
For second-order system
f = ∠Transfer function = ∠G(jw) – ∠[1 + G(jw) H(jw)] =
( )
1/ 2

∠M(jw) BW = wn ⎡ 1 − 2ξ 2 + 4ξ 4 − 4ξ 2 + 2 ⎤
⎣ ⎦
Bandwidth gives an indication of the transient response of
R(s) + Y(s)
∑ G(s) a control system, noise filtering characteristics and robust-
–
ness of the system.

H(s) Gain Margin (GM)

Gain margin is the amount of gain in decibel (dB) that can
Figure 1 Closed loop control system be added to the open loop before the closed loop system
becomes unstable.
1
R(s) Y(s) Gain margin = GM = 20 log10
M (jw) ∠ M (jw) ⎡⎣ M ( jωpc ) ⎤⎦

= – 20 log M ( jωpc ) dB
The ease and accuracy of measurements are some of the 
advantages of the frequency response method. Extraction The phase crossover frequency (wpc) is the frequency at
of transfer function is easy from frequency response test which phase angle becomes –180°.
than step response test (time response). The design and
parameter adjustment of the open-loop transfer function of Phase Margin
a system for specific. Closed loop performance is carried Phase margin (PM) is defined as the angle m degrees through
out more easily in frequency domain than in time domain. which the M(jw) plot must be rotated about the origin so that
The effect of noise disturbance and parameter variation are the gain cross over passes through the (–1. j0) point.
relatively easy to visualize and access through frequency
Phase margin = PM = ∠M(jwgc) – 180°
response. Nyquist criterion is a powerful frequency domain
method of extracting. The information regarding stability Gain crossover frequency (w gc) is the frequency at which
as well as relative stability of a system without the needs to M(jw) becomes 1 or decibel magnitude of
evaluate roots of the characteristic equation.
M(jw) becomes zero.
3.1022 | Control Systems

Relation Between Time Domain and Solution: (D)

Frequency Domain Characteristics Magnitude of transfer function
1. The resonant peak Mr of the closed loop frequency −ω 2 + 16
response depends on x only. when x = 0, Mr = ∞ when =
x is negative, the system is unstable, and the value of
( jω + 2) ( jω + 3)
Mr ceases to have any meaning. As x increases, Mr Magnitude of transfer function will affect the magnitude
decreases. of the system output. output becomes zero when transfer
In comparison to time response, maximum peak over- ­function magnitude is zero
shoot also depends on ‘x’ only. The peak overshoot is
−ω 2 + 16
zero if x ≥ 1. =0
2. Bandwidth is directly proportional to wn. ( jω + 2) ( jω + 3) 
→ Bw increases linearly with wn.
→ Bw decreases with increase in x for a fixed wn. –w + 16 = 0
2

For time response, rise time increases as wn ­decreases. w = 4 rad/sec

1
\ Bandwidth α . Example 23: The gain margin of a unity feedback control
Rise time
s+4
3. Band width (BW) and Mr are proportional to each system with the open loop transfer function G(s) = 2 is
s
1 1
other for 0 ≤ x ≤ . (A) 0 (B)
2 4
(C) 4 (D) ∞
Example 21: For the system shown in figure, the input
x(t) = sint. Solution: (A)
In the steady-state, the response y(t) will be Phase crossover frequency ∠G(s) = –180°
⎛ ωpc ⎞
x(t) 2 y(t) tan–1 ⎜ ⎟ – 180° = –180°
s+1 ⎝ 4 ⎠
⎛ ωpc ⎞
1
(A) sin(t –45°) (B) 2 sin(t − 45°) tan–1 ⎜ 4 ⎟ = 0
2 ⎝ ⎠
wpc = 0
1
(C) sin(t – 45°) (D) 2 sin(t – 45°) Magnitude of transfer function at phase crossover frequency
2
⎡ jω + 4 ⎤
Solution: (D) ⎢ pc ⎥ω =∞
2 2
Transfer function (T) = =
s + 1 jω + 1 ⎣ ( pc )
⎢ jω 2 ⎥ pc = 0
⎦
Input = sint [\ w = 1]
1 1
2 ⎛1⎞ Gain margin = = =0
T ∠θ = ∠ − tan −1 ⎜ ⎟ G ( jωpc ) ∞
1+1 ⎝1⎠ 
= 2 ∠–45° Example 24: The open loop transfer function of a unit
sx +1
y(t) = 1 × 2 sin (t – 45°) feedback control system B given as G(s) = . The value
s2
π
y(t) = 2 sin (t – 45°) of ‘x’ to give a phase margin of is equal to
4
(A) 0.441 (B) 0.141
Example 22: A system with zero initial condition has the (C) 1.141 (D) 0.841
closed loop transfer function
Solution: (D)
s 2 + 16
T(s) = Phase margin = 180° ∠G(jw) = 45°
( s + 2) ( s + 3)

The system output is zero at the frequency. ⎛ xωgc ⎞
180° + tan–1 ⎜ ⎟ – 180° = 45°
(A) 1 rad/s (B) 2 rad/s ⎝ 1 ⎠
(C) 3 rad/s (D) 4 rad/s xwgc = 1
 Chapter 3 Stability Analysis | 3.1023

Magnitude phase margin G ( jωpc ) = 1 In this example, the transfer function includes ‘m’ number
of zeros, ‘p’ number of poles at origin and in the mentioned
jωgc + 1 part ‘n’ number of poles. Let m = 1, n = 2, p = 1.
=1
( jω )
2
K (1 + sTz1 )
gc ⇒ G(s) =
gc

= 121/4 ( )(
s 1 + sT p1 1 + sT p 2 )
xwgc = 1 K (1 + jωTz1 )
1 G(jw) =
x = 1/ 4 = 0.841
2 
( )(
jω 1 + jωT p1 1 + jωT p 2 )
Example 25: In the G H(s) plane, the Nyquist plot of the K 1 + ω 2Tz12
π .e − s Magnitude of G(jw) =
loop transfer function G(s) H(s) = passes through the ω 1 + ω 2 T p1 2 1 + ω 2 T p2 2
s 
negative real axis at the point
(A) (–1, j0) (B) (– 0.5, j0) Magnitude of G(jw) is decibels is
(C) (– 3, j0) (D) (– 2, j0) G ( jω ) in dB = 20log G ( jω )

Solution: (D)
At the point of intersection with negative real axis, = 20logK + 20log 1 + ω 2 T p21 – 20 logw
∠G(s) H(s) = – p
⎛ π⎞ – 20log 1 + ω 2 T p1 2 − 20 log 1 + ω 2 T p2 2
− ⎜ ωpc + ⎟ = – p 
⎝ 2⎠
π The phase angle of G(jw) = ∠G(jw) = tan–1wTz1 – 90° – tan–1
wpc = wTP1 – tan–1 wTP2
2
Magnitude of the G(s) H(s) at p = wpc in the intersection Note: From the above analysis, it is clear that, when the
point with negative real axis magnitude is expressed in dB, the multiplication is
− jω converted to addition.
π e−s π e pc
G ( s) H ( s) w = wpc = = Therefore to sketch the magnitude plot, knowledge of the
s jωpc
magnitude variation of individual factors of the open loop
π π transfer function is essential. The various factors of open
= =2=
ωpc π loop transfer function are

2
1. Constant gain, K
\ Nyquist plot passes through (–2, 0).
1
or ( jω )
m
2. Poles (or zeros) at origin,
( jω )
n

Bode Plot
1
Bode plot is a graph of the transfer function of a linear, time- 3. First-order factor, or 1 + jωTz
invariant system frequency plotted with a ­log-frequency 1 + jωTp
axis, to show the system’s frequency response. It is usually a 1
combination of a Bode magnitude plot, expressing the mag- 4. Quadratic factor, 2
⎛ jω ⎞
nitude of the frequency response gain, and a Bode phase 1 + 2ξ ( jω /ωn ) + ⎜ ⎟
plot expressing the frequency response phase shift. ⎝ ωn ⎠
The standard logarithmic magnitude of open loop trans- ⎛ jω ⎞ ⎛ jω ⎞
2

fer function of G(jw) is 20 log10 G ( jω ) . The units used in or 1 + 2ξ ⎜ +

⎝ ω ⎟⎠ ⎜⎝ ω ⎟⎠
n n
this representation of the magnitude are the decibel, usually
denoted as dB.
Generally, a transfer function can be expressed in terms Constant Gain: K
of factors of its poles and zeros. The advantage of the loga- Let G(s) = K
rithmic plot is the conversion of these multiplicative factors
to additive terms. \ G(jw) = K ∠0°
Consider the general open loop transfer function.
G ( jω ) in dB = 20 log K
G(s) =
( )( )
K 1 + sTz1 1 + sTz2  (1 + sTzm )
⎛ 0⎞
sP (1 + sT ) (1 + sT )(1 + sT ) 
P1 p2 pn
f = ∠G(jw) = tan–1 ⎜ ⎟ = 0°
⎝ k⎠
3.1024 | Control Systems

|G(jw)| in dB ⎛ω⎞
Phase angle (f) = n × tan −1 ⎜ ⎟ = n × 90° 
⎝ 0⎠
K>1
Note 1: Magnitude plot of S ±n is a straight line with slop
of ±20 × n dB/decade that passes through the point
0 dB
K=1 [0 dB, 1 rad/s]
log w
Note 2: Phase angle plot of S ±n is independent of frequency
and it is constant angle of value ± 90n degrees.
K<1

First-order Factor (1+ jwT)±1

1
Note 1: The magnitude plot and phase plot of a constant Pole: open loop transfer function G(s) =
1+ sTp
‘K’ is independent of frequency and straight line.
Note 2: A constant (K) greater than unity has a positive 1
Log magnitude = 20 log = −20 log 1 + ω 2 Tp 2
value in decibels, while a number smaller than 1 + jωTp
unity has a negative value in decibels.
1
Note 3: The change in the value of gain (K) of transfer For ω << ; the asymptote is 20 log1 = 0 dB
function is increase or decrease in the magnitude Tp
plot. 1
For w >> ; the asymptote is –20 log wTp: It is a straight
Tp
Poles (zeros) at Origin (jw)±n line with slope of –20 dB/decade. This asymptote intersect
1
Pole: Open loop transfer function [G(s)] = 0 dB at the break frequency wc = 1/Tp, which is known as
sn corner frequency.
⎛ω⎞ ⎛ ωTp ⎞
Phase angle f = − n.tan −1 ⎜ ⎟ = – n × 90°
Phase angle f = − tan −1 ⎜ ⎟ = − tan (ωTp )
−1
⎝ 0⎠
⎝ 1 ⎠ 
Mag in dB 20 log |G( jω)|
At corner frequency f = –tan (wTp)
–1

s 2 = (jw)2
40 =
–tan–11 = 45°
s = jw The phase angle of the factor (1 + sTp)–1 varies from 0 to
20
w
–90° as ‘w’ is varied from 0 to infinity. The phase angle plot
0
0.1 1 10 1
crosses –45° at w = wc =
Tp
–20
s –1 = ( jw)–1 Zero: open loop transfer function
s –2 = ( jw)–2
–40 G(s) = (1 + sTz)

Log magnitude = 20 log 1 + ω 2 Tz 2

j( jw) in degrees 
1
For w < < ; the asymptote is 20 log 1 = 0 dB
s = ( jw) Tz
90° 1
For w > > ; the asymptote is 20 logwTz: it is a straight
Tz
line with slope of +20 dB/decade. This asymptote intersect
0 w 1
0 dB at the break frequency wc = , which is known as
corner frequency. Tz
1 –1
s = (jw)
–90° Phase angle f = tan–1wTz

The phase angle of the factor (1 + sTz) varies from 0 to 90°

as ‘w’ is varied from zero to infinity. The phase angle plot
Zeros: Open loop transfer function [G(S)] = sn 1
crosses 45° at w = wc = .
Log magnitude = 20 log|(jw)n) = 20n logw Tz
 Chapter 3 Stability Analysis | 3.1025

G(jw) in dB G(jw) in dB x = 0.1

x = 0.3

1
w= x = 0.5
TP
log w log w
Slope w = wr
– 20 dB/dec
Approximate plot x=1
Exact plot

(a)

G( jw) in dB
Approximate plot Figure 3 Bode plot for quadratic factor in denominator
Exact plot
Slop
Example 26: Draw the Bode plot for a system having
+ 20 dB/dec 100
log w G(s)H(s) =
s( s + 1) ( s + 2)
1
w= Find
Tz
(A) Gain margin
(B) Phase margin
(b) (C) Gain crossover frequency
(D) Phase crossover frequency
Figure 2 M
 agnitude plot for first-order pole (a)
and first-order zero (b) 50
Solution: G(jw)H(jw) =
jω (1 + jω ) (1 + 0.5 jω )
The corner frequencies are
Quadratic Factor
±1 w = 1 rad/s and w = 2 rad/s
⎡ ⎛ jω ⎞ ⎛ jω ⎞ ⎤
2

⎢1 + 2ξ ⎜ + ⎥ = Open loop transfer function = For w ≤ 1 rad/s

⎢⎣ ⎝ ω n ⎟⎠ ⎜⎝ ω n ⎟⎠ ⎥⎦ 50
G(jw)H(jw) =
⎧⎪
±1 jω
⎛ jω ⎞ ⎛ jω ⎞ ⎫⎪
2

⎨1 + 2ξ + Slope = –20 dB/decade

⎜⎝ ω ⎟⎠ ⎜⎝ ω ⎟⎠ ⎬
⎪⎩ n n ⎪⎭ |G(jw)H(jw)|dB = 20 log 50 – 20 log w
The magnitude in decibels is at w = 0.1
⎛ ⎛ ω ⎞2⎞ ⎛ ω ⎞
2 2 |G(jw)H(jw)|dB = 20 log 50 – 20 log(0.1) = 53.98 dB
= ±20 log ⎜1 − ⎜ ⎟ ⎟ + 4ξ 2 ⎜
⎜⎝ ⎝ ω n ⎠ ⎟⎠ ⎝ ω n ⎟⎠ At w = 1
|G(jw)H(jw)| = 20 log 50 = 33.98 dB
For w < < wn, the log magnitude is asymptotic to a straight
line of constant gain 0 dB and phase angle approaches to For 1 < w ≤ 2
0 deg. ⎛ω⎞ 50
For w >> wn, the log magnitude approaches ±40 log ⎜ ⎟ ; G(jw)H(jw) =
⎝ ωn ⎠ jω (1 + jω )
a straight line with slop of ± 40 dB/dec.
Asymptote intersect 0 dB at corner frequency w = wn. Slope = –20 – 20

Note: The resonant frequency is given by = –40 dB/decade

1 As w increases from 1 to 2, the reduction in gain

wr = ω n 1 − 2ξ 2 for x < ⎛ 2⎞
2 = 40 log ⎜ ⎟ = 12.04 dB
The maximum magnitude is ⎝ 1⎠
At w = 2
1 1
Mp = G ( jωr ) = for ξ <
2ξ 1 − ξ 2
2 |G(jw)H(jw)|dB = 21.94 dB
3.1026 | Control Systems

For w > 2 Solution: The line with a slope of –20 dB/decade does not
50 K
G(jw)H(jw) = pass through w = 1 rad/s, i.e. there is a term
jω (1 + jω ) (1 + 0.5 jω ) s
20 log K = –9
Slope = –40 – 20 = –60 dB/decade
As w increases from 2 to 10, the reduction in gain = K = 0.35
⎛ 10 ⎞ At w = 1 rad/sec, slope changes to 0 dB/dec indicating a
60 log ⎜ ⎟ = 41.94 dB
⎝ 2⎠ zero at w = 1 rad/sec. The term is (1 + s)
At w = 20 rad/sec, the slope changes to +20 dB/decade,
At w = 10,
⎛ s⎞
indicating a term ⎜1 + ⎟ or (1 + 0.05 s)
|G(jw)H(jw)|dB = –19.99 dB ⎝ 20 ⎠
∠G(jw)H(jw) = –90 – tan–1 w – tan–1(0.5w) At w = 40 rad/sec, the slope changes to 0 dB/dec indicat-
⎛ s ⎞
ing a term ⎜1 + in the denominator.
w ∠G(jw)H(jw)
⎝ 40 ⎟⎠
0 –90 0.35 (1 + s) (1 + 0.05s)
i.e., G(s) =
0.1 –98.6 s(1 + 0.025 s) 
0.2 –107
0.5 –130.6
Example 28: The Bode magnitude plot of H(s) =
10 4 (1 + s )
1 –161.6
1.3 –175.5 (10 + s) (100 + s)2
1.4 –179.5
1.5 –183.2 (A) H(jw) in dB
2 –198.4
40

Magnitude
54 20
Bode plot
40 –20 dB/decade
log w
–1 1 2 3
20 –40 dB/decade
–20
0.1 w –40
1 2

–100° –60 dB/decade (B) H(jw) in dB

40
Phase plot

–180° 20

–200°
log w
–1 1 2 3
Gain crossover frequency = 4.45 rad/s
–20
Phase crossover frequency = 1.40 rad/s
Gain margin = 27 dB –40
Phase margin = 53°
(C) H( jw) in dB
Example 27: Find the transfer function of the system whose
asymptotic Bode plot is shown in the following ­figure.
40
dB

20
–20 dB/decade
log w
–1 1 2 3
10 20 40
log w
–20
–9
+20 dB/decade –40
 Chapter 3 Stability Analysis | 3.1027

(D) H( jw) in dB Solution: Magnitude plot slop change at frequency ‘f1’ and
its increasing. This indicates there is a zero at f = f1.
40 G = (1 + sT1) = (1+ jwT1)

20 ⎡ω = 2π f ⎤
⎛ 2π f ⎞ ⎢ ⎥ f
= 1 + j 1 1 ⎥ ⇒ G = 1+ j
⎜⎝ 2π f1 ⎟⎠ ⎢T1 = = f1
log w ⎢⎣ ωc 2π f 1 ⎥⎦ 
–1 1 2 3
Example 30: The asymptotic Bode magnitude plot of a
–20
minimum phase transfer function is shown in figure.
–40
G( jw) Mag in dB

Solution: (A) 10 4 (1 + s ) – 40 dB/dec

Given function H(s) =
(10 + s) (10 + s) 2

10 4 (1 + s ) 20 dB
H(s) = 20 dB/dec
(1 + 0.1 s) (1 + 0.01 s)2 × 10 × 1002
0.1(1 + s ) w = 10 w (rad/sec)
H(s) = 0
w = 1.5
(1 + 0.1 s) (1 + 0.01 s)2
Corner frequencies are 1, 10 and 100
Initial magnitude = 20 log 0.1 = –20 dB 0 dB/dec
–20 dB
Magnitude starts increasing with slop of +20 dB/dec at wc = This transfer function has
1 rad, constant at w = 10 rad and decays with a slope of (A) Two poles and one zero
20 dB/dec at w = 100 rad. (B) Two poles and two zeros
Mag in dB (C) One pole and two zeros
(D) Three poles and one zero
20 Solution: Initial slop of the magnitude plot is –40 dB/dec,
log (w) indicates 2 poles of the system are at origin
0 Reduction is slop by 20 dB/dec at w = 1.5 indicates a zero.
w=1
w = 10 w = 100 Reduction in slop by 20 dB/dec at w =10 indicates
log w = 0 another zero.
log w = 1 log w = 2
20 \ Total 2 poles and 2 zeros.
Example 31: The asymptotic approximation of the log-
magnitude versus frequency plot of a minimum phase
Example 29: The function corresponding to the Bode plot system with real poles and one zero is shown in figure. Its
of figure is transfer function is
G(jw) Mag in dB Mag in dB

50 –40 dB/dec
20 dB per decade
–20 dB/dec

0 dB
f –40 dB/dec
0 f1
1
(A) G = jf / f1 (B) G =
⎛ f⎞ –20 dB/dec
⎜⎝1 + j f ⎟⎠
1
1 w (rad/sec)
(C) G = (D) G = 1 + jf / f1 5 7 25
⎛ f⎞ 0.1
⎜⎝1 − j f ⎟⎠
1
3.1028 | Control Systems

10 ( s + 5) ( s + 25) 4.4 ( s + 7) ( s + 5) (A) 3 dB

(A) (B)
s ( s + 7)
2
s ( s + 25)
2 (B) 9 dB
(C) 0.97 dB
10 ( s + 7) 4.4 ( s + 5) ( s + 25) (D) 5.4 dB
(C) (D)
s ( s + 5) ( s + 25)
2
s 2 ( s + 7) Solution: Actual magnitude of given transfer function =
⎛ jω ⎞
Solution: Transfer function has corner frequencies 5, 7 −20 log ⎜1 + ⎟
and 25 ⎝ a ⎠
Zeros are at w = 5 and w = 25 At w = 0.5a
Poles are at w = 7 ⎛ 0.5a ⎞
⇒ G ( jω ) = −20 log ⎜1 + j
⎛ s⎞ ⎛ s ⎞ ⎝ a ⎟⎠ 
K ⎜1 + ⎟ ⎜1 + ⎟
⎝ 5 ⎠ ⎝ 25 ⎠ 7 × 5 K ( s + 5) ( s + 25)
T. F = = −20 log 1 + (0.5)
=
2
s (1 + s / 7 )
2
25 s 2 ( s + 7) 

Initial magnitude = 20 log K = 10 ⇒ K = 3.16 G ( jω ) = –0.969 = – 0.97 dB
4.4 ( s + 5) ( s + 25)
Transfer function = Approximated magnitude in given plot at w = 0.5a is 0 dB
s 2 ( s + 7)
Error = 0 dB – (0.97 dB) = 0.97 dB.
Example 32: The asymptotic Bode magnitude plot of a
1
transfer function is show in figure. The error in dB
1+ s /a Polar Plot
gain at a frequency of w = 0.5a is The transfer function G(s) is a complex function and it is
given by
G(jw) in dB
G(s) = G(jw) = G ( jω ) ∠G ( jω ) = M∠f

As the input frequency is varied from 0 to ∞, the magnitude

M and phase angle ‘f’ change, the locus traced by the tip of
the phasor G(jw) is known as polar plot.
0 db a w
–20 dB/dec

G ( j ω ) ∠G ( j ω )
value

G(s) G ( jω ) ∠G ( j ω ) w→0 w→∞ Polar plot

–270°
–180° 1
1 1 –tan–1wT 1∠0 0 ∠ –90° 0°
w=• w=0
1+ sT 1+ ω T 2 2

w increase

–270°
–180°
0°
1 1 –90° ∞ ∠–90° 0 ∠– 90° w=•
s ω
w = 0 –90°

(Continued)
 Chapter 3 Stability Analysis | 3.1029

G ( j ω ) ∠G ( j ω )
value

G(s) G ( jω ) ∠G ( j ω ) w→0 w→∞ Polar plot

w=•
–270°/90°
(1+ 0) (1+∞)∠ w increase
1 + sT 1 + wT 90°
∠90° 90°
–180° w=0
o 0°

90°
w=•

s w 90° 0 ∠ 90° ∞ ∠ 90°

w=0
–180° o 0°
–270°

–180° o
w=• 0°

1 1 0
–90° – tan–1 wT ∞ ∠–90°
∠–180°
s (1 + sT ) ω 1+ ω T 2 2

–90°
w=0

–270°

–180° 1 0°
1 1 –tan–1wT1– 0 w=• w=0
1 ∠ 0°
tan–1 wT2 ∠–180°
(1 + sT1 ) (1 + sT2 ) 1 + ω 2T12 1 + ω 2T2 2

–90°

–270°

–180° w=•
–90° – tan–1wT1 – 0°
1 1 ω ∞∠ –90° 0∠–270° w=0
tan–1 T2
s (1 + sT1 ) (1 + sT2 ) ω 1+ ω T 2
1
2
1 + ω T2 2 2

–90°

–270°
–tan–1wT1 – –180° w=•
1 0°
1 tan–1wT2 – 1∠ 0° 0∠–270°
(1 + sT1 ) (1 + sT2 ) (1 + sT3 ) 1+ ω T
2
1
2
1+ ω T2 2 2
1+ ω T3 2 2
tan–wT3 w=0
–90°

–270°

–180° – tan1wT1
1 w=0
1 – ∞∠–180° 0∠–360°
s 2 (1 + sT1 ) (1 + sT2 ) ω 2 1 + ω 2T12 1 + ω 2T22 tan–1wT2 w=•
–180°

–90°

(Continued)
3.1030 | Control Systems

G ( j ω ) ∠G ( j ω )
value

G(s) G ( jω ) ∠G ( j ω ) w→0 w→∞ Polar plot

–270°

s ω 90° – tan–1 wT 0∠90° 1∠0°

–180° w = 0 w=•
1+ sT 1 + ω 2T 2
1
–90°

–270°
w=•
–180° 0°
1+ sT 1+ ω T 2 2
–90° + tan–1 wT ∞∠–90° 1 ∠0°
s ω
w=0
–90°

–270°
–180°
tan–1wT1–90° –
1 + sT1 0∠ w=•
1 + ω 2T12 tan–1wT2 – ∞ ∠ –90°
–180°
s (1 + sT2 ) (1 + sT3 ) ω 1 +ω T 2 2
1 +ω T3 2 2 tan–1wT3
Ing –90°

Note 1: Addition of a non-zero pole to a transfer function Nyquist Criterion

results in further rotation of the polar plot through
The Nyquist criterion relates the stability of a closed-loop-
an angle of –90o as w → ∞ (head of the polar plot
system to the open-loop frequency response and open-loop
shifts).
pole location. This criterion can tell us how many closed-
Note 2: Addition of a pole at origin to a transfer function
loop poles are in the right half of S-plane.
results in rotation of the polar plot at zero and infi-
The Nyquist criterion used the following concepts for
nite frequency (head and tail of polar plot) by fur-
the establishment of criterion.
ther angle of –90o.
Note 3: The effect of addition of a zero to a transfer func- 1. The poles of 1 + G(s)H(s) and the poles of G(s)H(s)
tion is to rotate the high frequency portion of the are same.
polar plot by 90o in anti-clockwise direction. 2. The zero of the 1 + G(s)H(s) are the poles of the
closed loop transfer function T(s) of the system.
↓ Start of
type-3 3. Mapping: Consider a complex number on the S-plane
system and substitute it into a function F(s), another complex
number results. This process is called ‘­mapping’.
Start of → → Start of Example: Substituting s = 4 + j3 into function F(s) =
type-2 type ’0’ s2 + 2s + 1, results in 16 + j30. We say that 4 + j3 maps
system system
into 16 + j30 through the function (s2 + 2s + 1).
Start of ↑
type-1
5. Mapping contours:
system Consider the collection of points, called a contour,
shown in figure as contour A. Also assume that
End of End of
3rd order 4th order
F (s) =
( s − Z1 ) ( s − Z2 ) ( s − Z3 )
system system
( s − P1 ) ( s − P2 ) ( s − P3 ) 
End of End of
2nd order
Contour ‘A’ can be mapped through F(s) into contour ‘B’
1st order
system system by substituting each point of contour A into the function
F(s) and plotting the resulting complex numbers. For exam-
Figure 4 S
 tart point and end point of polar plot for different ple, point ‘Q’ in S-plane maps into point ‘Q’ through the
system types and orders ­function F(s).
 Chapter 3 Stability Analysis | 3.1031

jw Q Contour A If we extend the contour to include the entire right half of

S-plane S-plane, we can count the number of right-half-plane closed
loop poles inside contour ‘A’ and determine a system’s
s stability.
Note: 
When we map the entire right half of S-plane
through G(s)H(s) instead of 1 +G(s)H(s), the result-
jw ing contour is same as mapping through 1 + G(s)
F-plane H(s), except that it is translated one unit to the left.
1 Contour B
Q So we count rotations about -1 + j0 instead of rota-
Real tions about the origin.

Statement of the Nyquist Stability

Criterion
Let us first assume that F(s) = 1 + G(s)H(s), with the picture If a contour ‘A’ that encircles the entire right half-plane is
of the poles and zeros of 1 + G(s)H(s) as shown in figure mapped through G(s)H(s), then the number of the closed
below. As each point Q of the contour ‘A’ is substituted into loop poles (Z) in the right half-plane equals the number of
1 + G(s)H(s), a mapped point results on a contour ‘B’. As we open loop poles (P) that are in the right half-plane minus the
move round contour ‘A’ in a clockwise direction, each vec- number of contour clockwise revolutions (N) around ‘–1’
tor of F(s) that lies inside contour A will appear to undergo of the mapping (i.e., Z = P – N). The mapping is called the
a complete rotation or a change in angle of 360o. On the Nyquist diagram/Nyquist plot of G(s)H(s).
other hand, each vector drawn from the poles and zeros of
jw
1 + G(s)H(s) that exist outside contour A will appear to
oscillate and return to its previous position, undergoing a
A
net angular change of 0o.
S = plane Q
S-plane
s
⇒ F(S) = 1 + G(s)H(s)
s

Contour A
Figure 5 C
 ontour enclosing right half of S-plane of
Imaginary determine stability
1+ GH = plane
Note 1: If the contour ‘A’ of the open loop transfer function
G(s)H(s) corresponding to the Nyquist contour in
Real
the S-plane encircles the point (–1 + j0) in the anti-
clockwise direction as many times as the number
of right half S-plane poles G(s)H(s), the closed
loop system is stable.
Contour ‘B ’
Note 2: No encirclement of –1 + j0 implies that the sys-
Number of anti-clockwise rotations of contour ‘B’ about tem is stable if there are no poles of G(s)H(s) in
origin (N) = P – Z the right half of ‘s’ plane; otherwise the system is
where, P =  Number of poles of 1 + G(s)H(s) inside unstable.
­contour A. Note 3: Clockwise encirclements in the Nyquist plane
Z= Number of zeros of 1 + G(s)H(s) inside indicate that the system is unstable.
­contour A. If G(s)H(s) has any poles on jw axis, the Nyquist contour
Note: Since the poles of 1 + G(s)H(s) are the poles of G(s) defined earlier cannot be used as such. Also, the S-plane
H(s) and zeros of 1 + G(s)H(s) are poles of closed contour should not pass through a singularity of 1 + G(s)
loop system, H(s). The stability in such cases is studied with modified
P = Number of open poles enclosed Nyquist contour which bypasses any jw-axis poles. This is
Z = Number of closed loop poles enclosed accomplished by indenting the Nyquist contour around the
N = Z – P = Number of closed loop poles inside the contour. ‘jw’ poles along a semicircle of radius ‘e’, where e → 0.
3.1032 | Control Systems

jw 1. Semicircular indent represented by s = lim εejθ (where

ε →0
j• ‘θ’ varied from –90o through 0o to 90o) is mapped into

jw 1 ×
⎡ 4ε e jθ ⎤ ⎛ 1 ⎞ − j 2θ
lim ⎢
ε →0 ε e θ θ jθ ⎥ = lim ⎜ ⎟ = ∞e
ε → 0 ⎝ ε 2 e j 2θ ⎠
⎣
2 j 2
( ε e j
+ 1 )( 2ε e + 1) ⎦
jw1 + ee jq
j 0+
= ∞ (∠180° → 0°→ ∠–180°).
× s
0 This part of map is an infinite circle.
j0– 2. Mapping of positive imaginary axis

jw 1
×
G ( jω ) H ( jω ) =
(1 + j 4ω )
( jω ) (1 + jω ) (1 + j 2ω )
2

–j•
For various values of ‘w’, G( jw)H( jw) is calculated
Figure 6 Indented Nyquist Contour for jw-axis
and plotted using polar plots.
open loop poles The G(jw)H(jw) – locus intersects the real axis at a
point where
Example 33: Consider a system with an open loop transfer ∠G(jw) H(jw) = –180o
function –180o – tan–1w – tan–1 2w + tan–1 4w = –180o

G (s) H (s) = 2
(4 s + 1) \ω=
1
= 0.354 rad/sec
s ( s + 1) ( 2 s + 1) 2 2

1
Find the stability of the system using Nyquist plot. ∴ G ( jω ) H ( jω ) ω = = 10.6
2 2
Solution: The given open loop transfer function has a dou- Further as w → +j∞
ble pole at origin. The Nyquist contour is intended to bypass ⇒ G ( jω ) H ( jω ) ∠G ( jω ) H ( jω ) ⇒ 0 ∠ − 270°

the origin. The mapping of Nyquist contour is obtained as
follows. as w → 0+ ⇒ |G(jw) H(jw)| ⇒ ∞∠–180o
3. The infinite semicircle of the Nyquist contour repre-
jw sented by s = lim Re jϕ (f varies from + 90o through
R →∞
j• Re jj 0o to + 90o) is mapped to
R→•

j0+
lim
(1 + 4 Re ) jϕ

= 0e − j 3ϕ
R →∞
R 2 e j 2ϕ (1 + Re ) (1 + 2 Re )
jϕ jϕ

s
j0– = 0(∠– 270o → ∠0o → ∠+270o)
ee jq
e →0
Number of counter clockwise encirclements to origin
are ‘–2’.
–j• Number of right half poles of open loop is zero.
Imaginary
Z = P – N = 0 – (–2) = +2
\ Number of poles on right half plane for closed loop
transfer function is ‘2’.
\ System is unstable.

–1+j0 R→• Example 34: Nyquist plot for the transfer function G(s) =
w = 0–
(4 + s) for positive frequencies has the form
w=+•
Real axis (A) jw (B) jw
w=–•
-10.6 j4
w = 0+
1
at w =
2 2 s
4 σ

(C) jw (D) jw

Figure 7 N
 yquist contour and corresponding
Nyquist plot 1/4 1/4
s s
 Chapter 3 Stability Analysis | 3.1033

Solution: (A) Imaginary

Given transfer function G(s) = 4 + s = 4 + jw
At w = 0 ⇒ 4 + j.0 –5 –2.
Real
–Ve 0.1
w = 10 ⇒ 4 + j10
GH-plane
w = 100 ⇒ 4 + j100
w = ∞ ⇒ 4 + j∞ 1
(A) 0.5 < K < 10 (B) K <
\ Nyquist plot is parallel to imaginary axis 5
(C) Both A and B (D) 0.5 > K > 10
jw
Solution: (C)
System gain ‘K’ should be adjusted such that the point (–1
+ j0) lies in the 0.1 to 0.2 region, because no. of encircle-
ments in this case is zero which results in stable operation
4 Real of the system.

Example 35: Which one of the following polar diagram \ 0.1k < 1 ⇒ K < 10
corresponds to a lag network? 2K > 1 ⇒ K > 0.5
(A) Img (B) Img
Range of K is 0.5 < K < 10
System is also stable if 5K < 1 [no. of encirclements will
w=0 w=• be zero in this case]
Real 1
w=0 w = • Real K<
5
Img jw
(C) Img (D)

w=• w=0 02k

Real
w=• w = 0 Real Real
5k
0.1k
Solution: (D)
Lag network offers only negative phase angles and. Let us
jw
consider a lag network example.
s +1 Example 37: The polar plot of an open loop stable system
G(jw) =
10 s + 2  as shown below the closed loop system is
imaginary
ω2 +1 ω
|G(jw)| ; ∠G ( jω ) = tan −1 tan −1 10 ω . 
100ω 2 + 1 1
w=•
Real
At w = 0 ⇒ 1∠0 –1.5
At w = 5 ⇒ 0.103∠–10.16
At w = ∞ ⇒ 0.1 ∠0o w=0

jw (A) Marginally stable

(B) Always stable
(C) Unstable with one pole on the RH S-plane
0.1 w →• 1∠0
s (D) Unstable with two poles on the RH S-plane
w=0
Solution: (D)
Complete polar plot of the given system is given in the
Example 36: The polar plot of a conditionally stable sys- ­figure No. of encirclements of (–1 + j.0) are ‘–2’
tem for open loop gain K = 1 is shown in the figure. The \ No. of open loop poles the right hand side = 0
open loop transfer function of the system is known to be
N = P – Z ⇒ Z = P – N
stable. The closed loop system is stable for
Z = 0 – (–2) = 2
3.1034 | Control Systems

\ RHS-plane poles of closed loop system are ‘2’ ⇒ Unstable No. of anti-clockwise encirclements (N) = 2.
jw No. of anti-clockwise encirclements (N) = P – Z = 2
No. of open loop poles on RHS side (p) = 0
No. of poles of closed loop system = 2
\ System is unstable.
–1 Plot (3) complete Nyquist plot is
Real
–1.5 Imaginary

w = 0+
Example 38: Consider the following Nyquist plots of loop –1 w = 0− Real p
transfer function over positive frequencies. Which of the
plots represents an unstable system?
1. Img   2. Img

w No. of counter clockwise encirclements (N) = –2

w=• No. of poles of closed loop system on RHS = 2
Real
−1 w=• \ Unstable system
w −1 Real Option (4) complete Nyquist plot is
Imaginary
w = 0−
3. Img   4. Img

w=•
Real
−1 w = •−
−1 Real
w –1 w = •+ Real

(A) (1), (2) and (3) (B) (2), (3) and (4)
(C) (1), (2) and (4) (D) (3) and (4) only w = 0+
Solution: Plot (1) Nyquist plot is
Solution: (B)
Img
No. of counter clockwise encirclements (N) = – 2
No. of RHS poles of closed loop control system (Z) =
P–N=2
\ System is unstable.
–1 Real
Example 39: A unity feedback system has the open loop
transfer function
1
G (s) =
( s − 1) ( s + 2) ( s + 3) 
No. of encirclements (N) = 0
If the open loop poles on RHS = 0 The Nyquist plot of ‘G’ encircles the origin
\ System is stable (A) Once
Plot (2) complete Nyquist plot is (B) Twice
(C) Thrice
Imaginary
(D) Never
w = 0+
Solution: (A)
No. of encirclements equals the difference ­between no. of
w = 0+ right hand side poles of G(s) and zeros.
–1 w = 0− Real N = POLTF − Z OLTF

POLTF = 1 and Z OLTF = 0 
w= 0− N = 1.
 Chapter 3 Stability Analysis | 3.1035

Example 40: Which of the following is the transfer Angle at the termination of NP is – 360o
­function of a system having the Nyquist plot shown in the Angle of termination –360o indicated system order is ‘4’.
following figure? \ System is type 2 and order ‘4’ system with no zeros.
Imaginary Example 41: In the GH-plane, the Nyquist plot of the loop
2π e −5 s
transfer function G(s)H(s) = pass through the nega-
s
tive real axis at the, point
w=+• (A) (–5, j0) (B) (–2, j0)
Real
w=−• (C) (–10, j0) (D) (–20, j0)
Solution: (D)
At the point of intersection of Nyquist plot with real axis
phase angle ∠G(s) H(s) = –180o = –p

K K 2π e −5 jω
(A) (B) ∠ = −π
s 2 ( s + 4 ) ( s + 8) s 2 ( s + 4 ) ( s + 5)
2 jω 
−1 ω
−5ω − tan = −π
K ( s + 2) K ( s + 2) ( s + 2) 0
(C) (D) 
s 2 ( s + 4 ) ( s + 8) s 2 ( s + 4 ) ( s + 5) π
−5ω − = −π
Solution: (A) 2 
Nyquist plot started at –180o angle. It indicates that the open π π
5ω = ⇒ ω =
loop system has two poles at origin. 2 10 
Magnitude and phase angle at w → 0 2π 2π
⇒ ∞∠– 180o G ( s) H ( s) ω= x = = = 20
10 ω π
Magnitude and phase angle at w → ∞ 10 

⇒ 0∠– 360o \ Nyquist plot passes through (–20, j0)

Exercises
Practice Problems 1 5. The feedback control system is fig is stable
Directions for questions 1 to 22: Select the correct alterna- C(s)
R(s) + (s − 2)
tive from the given choices. K≥o
(s + 2)2
–
1. The characteristic equation of a system is given by s6 +
3s5+ 8s4+ 18s3 + 37s2 + 75s + 50 = 0; the system is
(A) Stable. (B) Unstable. S–2
(C) Marginally stable. (D) Conditionally stable. (A) for all K ≥ 0 (B) only if K ≥ 1
2. How many roots of the characteristic equation (C) only if 0 ≤ k < 1 (D) only if 0 ≤ k ≤ 1
s6 + s5 – 2s4 – 3s3 – 7s2 – 4s – 4 = 0 6. Consider the points S1 = –3 + j4 and S2 = –3–j2 in the
S–plane. Then for a system with the open-loop transfer
lie in the left half of S-plane? k
(A) 4 (B) 5 (C) 1 (D) 6 function G(s) H(s) = is
( s +1) 4
3. A system described by the transfer function (A) S1 is on the root locus, but not S2.
1 (B) Both S1 and S2 are on the root locus.
H(s) = 3 is stable. The constraints on a
s + α s2 + k s + 2 (C) S2 is on the root locus, but not S1.
and k are (D) Neither S1 nor S2 on the root locus.
(A) a > 0, ak > 2 (B) a > 0, ak < 2
7. The gain margin (in dB) of a system having the open
(C) a > 0, ak > 0 (D) a < 0, ak < 0
loop transfer function.
4. The characteristic equation of a system is given by
2
s(s2 + 2s + 2)+ K(s + 3) = 0. The range of k for which G(s) H(s) = is
the system is stable is s ( s +1 )
(A) 0 < k < 30. (B) K > 3. (A) 0 (B) 3.01 (C) –3.01 (D) ∞
(C) 0 < k < 4. (D) 3 < K < 29.
3.1036 | Control Systems

8. The characteristic equation of a feedback control sys- 11. Which one of the following polar diagram corresponds
tem is given by s3 + 5s2 + (K+ 6)s + K = 0 In the root to a lag network?
loci diagram the asymptotes of the root loci for large (A) Im
‘K’ meet at a point in the s-plane whose coordinates are
(A) (2, 0) (B) (–1, 0)
(C) (–2, 0) (D) (–3, 0)
w=0 w=•
9. The open-loop transfer function of a system is given by
k
G(s) = the value of k which will cause Re
s ( s +1 )( s + 2 )
sustained oscillations in the closed-loop unity feed (B) Im
book system is
(A) 4 (B) 6 (C) 5 (D) 3
w=• w=0
10. A unity feedback system is given as
k (1 − s ) Re
G(s) = Indicate the correct root Locus
s( s+3 ) (C) Im
diagram.
(A) jw

w=0

Re
O s w=•

(D) Im

(B) jw

Re

s w=0 w=•

Common Data for Questions 12 and 13:

The open-loop transfer function of a unity feedback system
3 e− 2 s
is given by G(S) =
(C) jw s( s+ 2 )
12. The gain and phase cross – over frequencies in rad/s,
respectively,
(A) 0.485 and 0.632. (B) 1.26 and 0.632.
(C) 0.632 and 1.26. (D) 0.632 and 0.485.
s
13. Based on the above results, the gain and phase margins
of the system will be
(A) –7.09 dB and 87.5°.
(B) 7.09 dB and 87.5°.
(C) 7.09 dB and –87.5°.
(D) jw (D) –7.09 dB and –87.5°.
14. The loop transfer function of a closed-loop control
­system is given as
k ( s +1 )
G(S)H(S) = . The centroid of the
s s ( s + 2 )( s + 3 )
asymptotes is
(A) (–4, 0) (B) (–1, 0)
(C) (–2, 0) (D) (–3, 0)
 Chapter 3 Stability Analysis | 3.1037

15. A system has 10 poles and 2 zeroes. The slope of its 19. Loop transfer function G(s)H(s) of the magnitude plot
highest frequency asymptote in its magnitude plot is shown in the figure
(A) –100 dB/dec (B) –120 dB/dec
Mag in
(C) –160 dB/dec (D) –240 dB/dec dB
16. The polar diagram of a conditionally stable system for 60 dB
open loop gain K =1 is shown in figure. The open loop 40 dB
transfer function of the system is known to be stable. 20 dB
The closed loop system is stable for
0 1 2 3 log w
−20 dB
−40 dB

100 (0.1s + 1) 100 ( s + 10 )

2 2
–0.2
(A) (B)
–8 s 2 (0.01s + 1) ( s + 100) ( s + 1)2
(C) Both A and B (D) None of the above
–2
20. closed loop control system with transfer function
1 1 1 1 G (s)
(A) K < 5 and <K< (B) K < and < K < 5 is stable when
2 8 8 2 1 + G (s) H (s)
(A) Poles of the transfer function are on the left hand
1 1
(C) K < and 5 < K (D) K > and K < 5 side of the S-plane.
8 8 (B) Zeros of the characteristic equation are on the left
17. Pole zero plot of a loop transfer function is shown in half of the S-plane.
figure below, the breakaway/ breakin points in the root (C) Poles of the characteristic equation are on the left
locus diagram is half of the S-plane.
Imaginary (D) Both A and B.
axis 21. Polar plot of an open-loop stable system is shown in the
figure. The system is
Imaginary
axis

Real axis
−1 Real axis

(A) 1 (B) 2 (C) 0 (D) 3

18. Loop transfer function G(s)H(s) of the magnitude plot
shown in the figure
Mag in
dB
60 dB −20 dB/dec
(A) System is stable.
40 dB (B) System is unstable with one pole on the right-hand
side of S-plane.
20 dB
−40 dB/dec (C) System is unstable with two poles on the right-
hand side of S-plane.
−1 0 1 2 3 log w (D) System is marginally stable.
−20 dB
22. Which of the flowing are effects of PD controller on
−40 dB system?
1. Reduces peak overshoot.
100( S + 2) 2 10 (.01s + 1)
(A) (B) 2. Reduces raise time.
S ( S + 1) s (.1s + 1) 3. Improves damping.
100(.01S + 1) 2 (.01S + 1) 2 4. Reduces steady-state error.
(C) (D) (A) 1, 2, 3 (B) 2, 3, 4
S (.1S + 1) S (.1S + 1)
(C) 2, 3, 4 (D) 1, 3, 4
3.1038 | Control Systems

Practice Problems 2
Directions for questions 1 to 15: Select the correct alterna-
tive from the given choices.
1. Which of the following statements are ‘true’? (–1 + j 0)
   (i) Root Locus is a frequency response plot. w
w=•
(ii) The roots of characteristic equation are not a
­function of open Loop gain K.
(iii)  Root Locus technique is a tool for adjusting
the location of closed loop poles to achieve the
w=0
desired system performance.
 (iv) The exact root– locus is sketched by trial-and- (A) IG I
error procedure.
(A) i and ii –20 dB/dec
(B) ii and iii
–40 dB/dec
(C) iii and iv w
w1 w2
(D) ii, iii, and iv
2. The following statements refer to the equation P(s) + IG I
KQ(s) = 0, where P(s) and Q(s) are polynomials of s (B)
with constant coefficients. Identify the statements
which are ‘true‘. –20 dB/dec
   (i) The intersection of the asymptotes must always be
w
on the real axis. w1
–40 dB/dec
(ii) The breakaway points of the root loci must always
be on the real axis. IGI
(C)
(iii) Given the equation 1 + KG1(s)H1(s) = 0, where
G1(s)H1(s) is a rational function of s and does not
d G1 ( s ) H 1 ( s ) –40 dB/dec
contain K, the roots of are all
ds w1 w
break away points on the root loci ( – ∞ < K < ∞)
 (iv) At the break away points on the root loci the root –60 dB/dec
sensitivity is infinite.
IG I
(A) i and iv (D)
(B) i, ii, and iv –20 dB/dec
(C) ii and iii
–40 dB/dec
(D) ii, iii and iv
3. Which of the following statements are true? w
w1 w2
–60 dB/dec
   (i) Adding a zero to the function G(s)H(s) tends to
push the root loci to the left.
(ii) Adding a zero to the forward-path transfer func- 5. The Nyquist plot for the open- loop transfer function
tion will generally improve the system damping, G(s) of a unity negative feedback system is shown
and thus always reduce the maximum over shoot in the figure, if G(s) has no pole in the right half of
of the system. S-plane, the number of roots of the system characteris-
(iii) Adding a pole to G(s)H(s) has the effect of push- tic equation in the right-half of S-plane is
ing the root loci to the right. Im
  (iv) Complementary root locus (CRL) refers to root
loci with positive ‘k’.
(A) i, ii, and iii Re
−1
(B) i, ii and iv
(C) ii, iii and iv
(D) ii and iv
4. The Nyquist plot for a control system is shown in
­figure. The bode plot for the same system will be (A) 0 (B) 1 (C) 2 (D) 3
 Chapter 3 Stability Analysis | 3.1039

6. Which of the following points is NOT on the root locus 11. If the closed loop transfer function of a control system
of a system with the open-loop transfer function? s−5
is given by T(s) = , then it is
G (s) H (s) =
K ( s + 2) ( s + 3)
s ( s + 1) ( s + 3) (A) An unstable system

(B) An uncontrollable system
(A) s = − j 3 (B) s = –1.5
(C) A minimum-phase system
(C) s = –3 (D) s = –∞ (D) A non-minimum phase system
7. The figure shows the Nyquist plot of the open-loop 12. For the asymptotic Bode magnitude plot shown in the
transfer function G(s)H(s) of a system; If G(s)H(s) has following figure, the system transfer function can be
one right-hand pole, the closed loop system is
Magnitude in dB
Im
GH plane

Re 20
w=0 (−1,0)

w positive
0 w
0.001 0.1 10 1000
(A) Always stable
(B) Unstable with one closed loop right-hand pole 10 s + 1 100 s + 1
(C) Unstable with two closed loop right-hand poles (A) (B)
0.1s + 1 0.1s + 1
(D) Unstable with three closed-loop right-hand poles
100 s 0.1s + 1
K (C) (D)
8. Given G(s)H(s) = , the point of inter- 10 s + 1 10 s + 1
s ( s + 1) ( s + 3)
K
section of the asymptotes of the root loci with the real 13. The root locus of the system G(s)H(s) =
axis is s ( s + 2) ( s + 3)
has the break-away point located at
(A) –4 (B) 1.33 (C) –1.33 (D) 4
(A) (–0.5, 0) (B) (–2.548, 0)
9. The polar plot shown in the figure represents the trans- (C) (–4, 0) (D) (–0.784, 0)
fer function:
14. dB
Im(s) 160

140
w=• w=0
Re(s)

20
1

1 0.1 10 100 w
(A) G(s) =
s The approximate Bode-Magnitude plot of a minimum-
1
(B) G(s) = phase system is shown in the figure. The transfer func-
s (1 + sT ) tion of the system is
1 s + 0.1)
(C) G(s) =
(A) 10 8 ( 3

1+ sT
1 ( s + 10)2 ( s + 100)
(D) G(s) =
(1 + sT1 ) (1 + sT2 ) (B) 10 7
( s + 0.1)3
10. The open loop transfer function of a unity gain feed- ( s + 10) ( s + 100)
back control system is given by ( s + 0.1)
2
8
K (C) 10
G (s) = ( s + 10)2 ( s + 100)
( s + 1) ( s + 3) 
( s + 0.1)
3
The gain margin of the system is dB is given by (D) 10 9

(A) ∞ (B) 1 (C) 20 (D) 0 ( s + 10) ( s + 100)2

3.1040 | Control Systems

15. Consider the Bode magnitude plot shown in the figure.

(A)
( s + 10)
The transfer function H(s) is
( s + 1) ( s + 100)
20logH(jw)
10 ( s + 1)
(B)
( s + 10) ( s + 100)
0
20 dBdecade 10 2 ( s + 1)
(C)
−20 ( s + 10) ( s + 100)
103 ( s + 100 )
(D)
1 10 100 w in rad ( s + 1) ( s + 10)

Previous Years’ Questions

1. The frequency response of G(s) = 1/[s (s + 1) (s + 2)] 2. The frequency response of a linear system G( jw) is
plotted in the complex G( jw) plane (for 0 < w < ∞) is provided in the tabular form below
 [2010]
|G(jw)| 1.3 1.2 1.0 0.8 0.5 0.3
(A) Im
∠ G(jw) -130° -140° -150° -160° -180° -200°
−3/4

Re
The gain margin and phase margin of the system are
 [2011]
(A) 6 dB and 30° (B) 6 dB and -30°
w=0
(C) -6 dB and 30° (D) -6 dB and -30°
(B) w=0
Im

3. The open loop transfer function G(s) of a unity feed-

⎛ 2⎞
k ⎜ s+ ⎟
⎝ 3⎠
back control system is given as, G(s) = 2
s ( s + 2)
−3/4 Re From the root locus, it can be inferred that when k
tends to positive infinity. [2011]
(C) w=0 Im (A) Three roots with nearly equal real parts exist on
the left half of the S-plane.
(B) One real root is found on the right half of the
­S-plane.
(C) The root loci cross the jw axis for a finite value of
Re
k; k ≠ 0.
−1/6 (D) Three real roots are found on the right half of the
(D) S-plane.
Im

Re 4. The Bode plot of a transfer function G(s) is shown in

the below figure. The gain (20 log|G(s)|) is 32 dB and
–8 dB at 1 rad/s and 10 rad/s, respectively. The phase
is negative for all ω. The G(s) is [2013]
w=0
−1/6
 Chapter 3 Stability Analysis | 3.1041

8. The Bode magnitude plot of the transfer function

K (1 + 0.5s ) (1 + as )
40
G (s) = is shown below:
Gain (dB)

⎛ s⎞ ⎛ s⎞
s ⎜1 + ⎟ (1 + bs ) ⎜1 + ⎟
32 ⎝ 8⎠ ⎝ 36 ⎠
Note that-6 dB/octave = –20 dB/decade. The value of
20
a
is _________. [2014]
bK
0
1 10 100
−8 −6 dB/Octave
w (rad/s)
6 dB/Octave 0 dB/Octave
dB 0 dB/Octave −6 dB/Octave

39.8 39.8 32 32
(A) (B) (C) (D) −12 dB/Octave
s s2 s s2
0
5. In the formation of Routh–Hurwitz array for a poly- 0.01 2 4 8 24 36 w (rad/s)
nomial all the elements of a row have zero values.
This premature termination of the array indicates the 9. A system with the open loop transfer function
presence of [2014] K
G (s) =
(A) Only one root at the origin ( )
s ( s + 2) s 2 + 2 s + 2
(B) Imaginary roots
is connected in a negative feedback configuration with
(C) Only positive real roots
a feedback again of unity. For the closed loop system
(D) Only negative real roots
to be marginally stable, the value of K is _____
6. The root locus of a unity feedback system is shown in  [2014]
the figure [2014] 10. For the transfer function
5 (s + 4)
jw G (s) =
K=0 K=0
(
s ( s + 0.25) s 2 + 4 s + 25)
The values of the constant gain term and the highest
s
−2 −1 corner frequency of the Bode plot, respectively, are
 [2014]
(A) 3.2, 5.0 (B) 16.0, 4.0
The closed loop transfer function of the system is (C) 3.2, 4.0 (D) 16.0, 5.0
11. The magnitude Bode plot of a network is shown in the
C (s) K
(A) = figure [2014]
R (s) ( s + 1) ( s + 2) |G(jw)|
C (s) −K dB
Slope 20 dB/decade
(B) =
R (s) ( s + 1) ( s + 2) + K
C (s) K
(C) =
R (s) ( s + 1) ( s + 2) − K 0
1 log10 w
C (s)
1
K 3
(D) =
R ( s ) ( s + 1) ( s + 2) + K The maximum phase angle Φm and the corresponding
7. For the given system, it is desired that the system be gain Gm, respectively, are
stable. The minimum value of α for this condition is (A) −30o and 1.73 dB
________. [2014] (B) −30o and 4.77 dB
3.1042 | Control Systems

(C) +30o and 4.77 dB 15. Nyquist plots of two functions G1(s) and G2(s) are
(D) +30o and 1.73 dB shown in figure. [2015]
Im Im ∞
12. A Bode magnitude plot for the transfer function G(s)
w=∞ ω
of a plant is shown in the figure. Which one of the fol- ω=0
G1(s)
lowing transfer functions best describes the plant?
 [2015] G1(s) Re
w Re
20 log|G(j2pf)|
0    
20 (a) Im
0
w=0
−20
∞→w Re
0.1 1 10 100 1 k 10 k 100 k f (Hz)

1000( s + 10) (b) Im

(A)
s + 1000

10( s + 10) 1 Re
(B)
s( s + 1000)

s + 1000 (c) Im
(C)
10 s( s + 10)

s + 1000 Re
(D)
10( s + 10)

13. The transfer function of a second order real system with ∞

(d) Im
↑
a perfectly flat magnitude response of unity has a pole at ω
(2 − j3). List all the poles and zeroes. [2015]
(A) Poles at (2 ± j3), no zeroes.
Re
(B) Poles at (±2 − j3), one zero at origin. ω
↓
(C) Poles at (2 − j3), (-2 + j3), zeroes at (−2 − j3), (2 +
0
j3)
16. An open loop transfer function G(s) of a system is
(D) Poles at (2 ± j3), zeroes at (-2 ± j3)
14. The open loop poles of a third order unity feedback K
system are at 0, −1, −2. Let the frequency correspond- G ( s) =
s( s + 1)( s + 2)
ing to the point where the root locus of the system
transits to unstable region be K. Now suppose we For a unity feedback system, the breakaway point of
introduce a zero in the open loop transfer function the root loci on the real axis occurs at, [2015]
at −3, while keeping all the earlier open loop poles (A) -0.42
intact. Which one of the following is TRUE about (B) -1.58
the point where the root locus of the modified system (C) -0.42 and -1.58
transits to unstable region? [2015] (D) none of the above
(A) It corresponds to a frequency greater than K s
17. The transfer function of a system is Y ( s) = S + 2 the
(B) It corresponds to a frequency less than K R( s )
(C) It corresponds to a frequency K steady state output y(t) is Acos (2t + φ) for the input
(D) Root locus of modified system never transits to cos(2t)the values of A and φ respectively are [2016]
unstable region.
 Chapter 3 Stability Analysis | 3.1043

1 1 (A) once in clockwise direction

(A) ,–45° (B) ,+ 45° (B) twice in clockwise direction
2 2
(C) once in anticlockwise direction
(C) 2 ,–45° (D) 2 ,+45° (D) twice in anticlockwise direction
18. The phase cross-over frequency of the transfer func- 21. Given the following polynomial equation
tion G(S) = 100 in rad/s is [2016] S3 + 5.5S2 + 8.5S + 3 = 0, the number of roots of the
( S + 1)3 polynomial, which have real parts strictly less than –1
is ______. [2016]
19. Consider the following asymptotic Bode magnitude
plot (w is in rad/s). 22. Consider a linear time - invariant system with transfer
function
1
H(S) =
(S + 1)
If the input is Cos(t) and the steady state output is
ACos(t + a), then the value of A is _________.
[2016]
23. The open loop transfer function of a unity feedback
control system is given by
K (S + 1)
G(S) = , K > 0, T > 0
S (1 + Ts)(1 + 2S )
Which one of the following transfer functions is
best represented by the above Bode magnitude plot? The closed loop system will be stable if, [2016]
[2016] 4 ( K + 1)
(A) 0 < T <
2S K −1
(A)
(1+ 0.5S )(1+ 0.25S ) 2 4 (T + 2)
(B) 0 < K <
4(1 + 0.5S ) T −2
(B)
S (1 + 0.25S ) (C) 0 < K < T + 2
2S T −2
(C)
(1 + 2S )(1 + 4 S ) 8( K + 1)
(D) 0 < T <
4S K −1
(D)
(1 + 2S )(1 + 4 S ) 2
24. The gain at the breakaway point of the root locus of a
20. Loop transfer function of a feedback system is G(s) unity feedback system with open loop transfer func-
KS
s+3 tion G(S) = is [2016]
H(s) = Take the Nyquist contour in the (S −1)(S − 4)
s3 (s − 3)
(A) 1 (B) 2
clockwise direction. Then, the Nyquist plot of G(s)
H(s) encircles – 1 + j 0. [2016] (C) 5 (D) 9
3.1044 | Control Systems

Answer Keys

Exercises
Practice Problems 1
1. B 2. B 3. A 4. C 5. C 6. C 7. D 8. C 9. B 10. C
11. D 12. B 13. D 14. C 15. C 16. B 17. B 18. C 19. A 20. C
21. A

Practice Problems 2
1. C 2. A 3. A 4. D 5. A 6. B 7. A 8. C 9. C 10. A
11. D 12. A 13. D 14. A 15. C

Previous Years’ Questions

1. A 2. A 3. A 4. B 5. B 6. C 7. 0.618 8. 0.75 9. 5 10. A
11. C 12. D 13. D 14. D 15. B 16. A 17. B 18. A 19. A 20. A
21. 2 22. 0.707 23. C 24. A