Chapter 7

Root-locus Analysis of Control Systems

7.1 Introduction
Consider a feedback control system whose open-loop transfer function is
given by

K ( s + z1 )( s + z 2 ) L ( s + z Z )
G (s) H (s) = (7.1)
( s + p1 )( s + p2 ) L ( s + p P )
where K is a variable gain constant, Z is the number of finite zeros of
G(s)H(s), and P is the number of finite poles of G(s)H(s). The characteristic
equation for the system is given by
φ (s) = 1 + G(s) H (s) = 0 (7.2)
G ( s ) H ( s ) = −1 (7.3)
Recall that the poles of the closed-loop system are the roots of φ (s ) . Any
root of φ (s ) satisfies the following conditions:
1. Magnitude condition;
G ( s) H ( s) = 1 (7.4)
2. Angle condition;
∠G ( s ) H ( s ) = ±( 2l + 1)180 0
(7.5)
l = 0, 1, 2, 3, L
The root locus method analysis is a process of determining points in s-plane
satisfying the angle condition. The roots are usually plotted in the s-plane
with K as the variable parameter. The path of each root is known as the root
locus. And from the location of the roots in the s-plane, the nature of the
time response and system stability can be ascertained.
To illustrate how to make use of the angle condition, consider G(s)H(s) with
one zero at s = -z1 and three poles at s = -p1, s = -p2 and s = -p3. Thus Z =
1and P = 3.

s jω

θp3 θz1 θp2 θp1

x
p3
o
z1
x
p2 p1x σ
Point s has the phase angle given by
Z P
θ = ∠G ( s ) H ( s ) = ∑ θ zi − ∑ θ pi
i =1 i =1

If the angle θ is an odd multiple of 1800 then point s is on the system’s root
locus, i.e.
θ = ± (2l + 1)180 0
Finding the root locus over the s-plane in this manner is quite tedious.
Evan’s rules for quick sketching of the root locus are therefore very useful
and will be introduced in this chapter. First of all, simple cases are
considered by way of illustrative examples in order to expose the properties
of the root locus which Evan’s rules make use of.

7.2 Properties of the root locus

To illustrate some important properties of the root locus, we consider a few
straightforward examples.

Example 1
Let
K
G(s) H (s) = .
s +1
The system has one open-loop pole at s = 1 and no finite open-loop zero.
Therefore, the number of open-loop poles, P, equals 1 while the number of
open-loop zeros, Z, equals zero.
The characteristic equation is
K
φ ( s) = 1 + =0
s +1
or
s +1+ K = 0
⇒ s = −(1 + K )
K 0 1 3 4 10 L ∞
s −1 − 2 − 4 −5 − 11 L - ∞
Thus the root locus lies on the negative axis. It starts at s = -1 (the open-loop
pole) when K = 0 and ends at s = -∞ when K = ∞.
The root locus is sketched below. From this sketch, it can be seen that the
number of root locus branches, N, is 1.

113
 jω

K→∞ K=4 K=3 K=1 K=0

-5 -4 -3 -2
x 0 σ
-1

Example 2
Let
Ks
G(s) H (s) =
s +1
Then
Ks
φ (s) = 1 + =0
s +1

s (1 + K ) + 1 = 0
1
⇒s=−
1+ K
Note that the open-loop system has one finite zero at s = 0 and one finite
pole at s = -1. This means that both the number of open-loop poles and the
number of open-loop zeros equal 1. For various values of K > 0, the roots of
φ (s ) are
K 0 1 3 9 99 L ∞
s −1 − 0. 5 − 0.25 −5 − 0. 1 L 0
Thus the roots lie on the negative real axis between s = -1 and s = 0 as
shown below. And the number of root locus branches equals 1.

jω

K=0 K=1 K=∞

x -0.5
x
0 σ
-1

114
Example 3
Let
K
G (s) H (s) =
s ( s + 4)
The characteristic equation is
K
φ (s) = 1 + =0
s ( s + 4)
or
s 2 + 4s + K = 0
The two roots of φ (s ) are given as
s1, 2 = −2 ± 4 − K
For some values of K, the roots are tabulated below:
K 0 2 4 5 8 L ∞
s1 0 −1 −2 − 2 + j1 − 2 + j2 L - 2 + j∞
s2 - 4 −3 −2 − 2 − j1 − 2 − j2 L - 2 - j∞
The sketch is displayed below. In this case, the number of root locus
branches is 2, the number of open-loop poles is 2 while the number of open-
loop zeros is zero.

jω
↑K = ∞
3
K=8 2
K=5 1
K=0 K=2 K=4 K=2 K=0
x -2
x σ
-4 -3 -1
K=5 -1

K=8 -2
-3
↓K = ∞

115
Example 4
Let
K
G (s) H (s) =
s ( s + 1)( s + 3)
The characteristic equation is
K
φ (s) = 1 + =0
s ( s + 1)( s + 3)
or
s 3 + 4 s 2 + 3s + K = 0
The three roots of φ (s ) are given below for some values of K

K 0 0.631 5 12 40
s1 0 − 0.45 − 0.27 + j 0.98 j 3 0.5 + j 2.74
s2 - 1 − 0.45 − 0.27 − j 0.98 −j 3 0.5 − j 2.74
s3 - 3 − 3.1 − 3.55 − 4.1 -5
The root locus is shown in the s-plane below.

jω K=∞
2
K = 12
1
K=∞ K=0 K=0 K=0
x x 0
x σ
-3 -2 -1

-1
K = 12
Asymptotic line -2

K=∞

116
Note that the number of branches is 3, the number of open-loop poles is 3
and the number of open-loop zeros is zero. The open-loop poles are at s1 = 0,
s2 = -1, and s3 = -3

7.3 Rules for sketching the root locus

1) Number of branches of the root locus:
N = P if P > Z
N = Z if Z > P
Usually P > Z, and therefore, N = P.
P = number of open-loop poles
Z = number of open-loop zero
2) Starting point: The root locus starts (K = 0) from the open-loop poles.
3) Endpoints: The root locus terminates (K = ∞) either on open-loop
zeros or at infinity.
4) Locus on the real axis: A point on the real axis lies on the root locus if
the sum of open-loop poles and zeros to its right is odd.
5) Breakaway points: On the root locus between two open-loop poles,
the roots move towards each other as K increases till they coincide or
become repeated. If K is increased further, the roots become complex
conjugate pairs, i.e., they break away from the real axis. Breakaway
points may be determined by making K the subject of the
characteristic equation and then solving for the breakaway points sB
from
dK ( s )
=0
ds s = s B
6) Asymptotes (for P ≠ Z): The root locus plots are asymptotic to straight
lines which make angles with the real axis given by the
(2l + 1)180 0
θA =
P−Z
The asymptotes intercept the real axis at a point, σA , given by
σA = ∑
pi − ∑ z i
P−Z
7) jω-axis intersections: The value of K and the points at which root
crosses the jω-axis are determined by using the Routh-Hurwitz
criterion. The roots at the intersection point are imaginary.
8) Angles of Departure and Arrival: The angle of departure (from a
complex pole) or arrival (to a complex zero) is a tangent to the root
locus at the pole (or zero) concerned. It may be determined as follows:

117
 Assume a point sθ arbitrarily near the pole or zero concerned and then
apply the angle condition
Z P

∑ θ − ∑ θ = ±(2l + 1)180
i =1
zi
i =1
pi
0

where
θpi is the angle that the open-loop pole pi subtends
θzi is the angle that the open-loop zero zi subtends
The angle of departure (or arrival) at pole pj is
ψ j = ∑ θ zi − ∑ θ pi + 180 0 i≠ j
The above basic rules enable the plotting of the root locus for a given
open-loop transfer function. Determination of the value of K at a
specified point on the root locus, calculation of damping ratio ζ, and
assessment of time-response are illustrated by the following worked
examples.

Further examples

Example 5
A control system has the open-loop transfer function given by
K
G (s) H (s) =
s ( s + 4)
a) Use the above rules to plot the root locus as a function of K
b) Find the value of K such that the closed-loop system has a damping
1
ratio of
2
Solution:
(a)
1. There are two open-loop poles and no zeros. That is, P = 2, Z = 0.The
number of branches is therefore,
N=P–Z=2
2. The root loci start (K = 0) from the open-loop poles at 0 and -4.
3. There are no finite zeros, hence the branches terminate (K = ∞) at ±∞.
4. On the real axis, the root locus lies between 0 and -4.
5. Breakaway points are found as follows:
φ ( s ) = s ( s + 4) + K = 0
− K = s 2 + 4s
Hence

118
 dK
= −( 2s + 4) = 0
ds
⇒ s = −2
The breakaway point is at s = -2
6. The angle made by the asymptotes with the real axis is
(2l + 1)180 0
θA =
P−Z
Since there are two asymptotes, l = 0, 1 , and
2× 0 +1
(i) θ A1 = × 180 0 = 90 0
2−0
2 ×1 − 1
(ii) θ A 2 = × 180 0 = 270 0
2−0
The two asymptotes intersect the real axis art the centroid given by
σA = ∑
pi − ∑ zi (0 − 4) − 0
= =2
P−Z 2−0
Thus the root locus diagram is now drawn:

K=∞ jω

s = -2 +j2 K = 8 2

900
K=0 β K=0
X X
-4 2700 -2 σ

-2

K=∞

119
(b)
Now from the origin, a line of angle
1
β = cos −1 ζ = cos −1 = 450
2
is drawn. It intersects the root locus plot at a point s = -2 + j2. The value of
K at this point is determined using the magnitude condition:
G ( s) H ( s) = 1
i.e.
K
=1
s ( s + 4)
K
⇒ =1
(−2 + j 2)(−2 + j 2 + 4)
Therefore,
K = (−2 + j 2) 2 + j 2 = 8

Example 6
The open-loop transfer function of a unity feedback control system is given
by
K
G (s) H (s) =
s ( s + 1)( s + 3)
Sketch the root locus as a function of K, and determine the value of K
a) for a damping ratio of 0.5,
b) for marginal stability, and
c) at s = -4
Solution:
Three open-loop poles at s = 0, -1 and -3. Therefore, P = 3.
No finite open-loop zeros. Therefore, Z = 0.
N = P – Z = 3 = number of branches of the root locus.
The loci start (K = 0) at s = 0, -1 and -3.
The loci end (K = ∞) at infinity.
On the real axis, the root locus lies between s = 0 and -1, and also between s
= -3 and -∞.
Breakaway points:
φ ( s ) = 1 + G ( s ) H ( s ) = s ( s + 1)(s + 3) = 0
∴ K = −( s 3 + 4 s 2 + 3s )

120
and
dK
= −(3s 2 + 8s + 3) = 0 ⇒ 3s 2 + 8s + 3 = 0
ds
s1 = −0.45 and s2 = −2.21
Thus the breakaway point is sB = -0.45 since s2 is not on the root locus.
The angle of the asymptotes is given by
(2l + 1)180 0
θA = l = 0, 1, 2
3
Hence

2× 0 +1 2 ×1 + 1
θ A1 = × 180 0 = 60 0 θ A2 = × 180 0 = 180 0
3 3

2× 2 +1
θ A3 = × 180 0 = 300 0
3
The centroid of the asymptotes is given by
(0 − 1 − 3) − (0)
σA = = −1.33
3
Thus the root locus plot is as shown:

jω K=∞
2
K = 12
1
K=∞ K=0 600
K=0 K=0
x x 0
x σ
-3 -2 -1
-0.45
-1
K = 12

-2

K=∞

121
The value of K for marginal stability is obtained using the Routh-Hurwitz
criterion:
φ ( s ) = s 3 + 4 s 2 + 3s + K = 0
Routh array
s3 : 1 3
s2 : 4 K
12 − K
s1 : 0
4
s : K
0

Thus K = 12 is the value for marginal stability. It causes the s1 row to be

zero. Using the s2 row the auxiliary equation becomes
α ( s ) = 4 s 2 + 12 = 0
⇒s=±j 3
At K = 12, the system has sustained oscillations whose frequency is
ω = 3 rad/s .
The damping ratio is ζ = 0.5. Therefore, cos θ = 0.5, and θ = 600. The line
corresponding to θ = 600 is drawn from the origin. It cuts the root locus at
the point s = -0.35 + j0.6.
At the point s = -0.35 + j0.6, the value of K is found using the magnitude
condition
G ( s) H (s) = 1
⇒ K = s ⋅ s +1 ⋅ s + 3
or
K = − 0.35 + j 0.6 ⋅ − 0.35 + j 0.6 + 1 ⋅ − 0.35 + j 0.6 + 3 = 1.6
The value of K at s = -4 is also obtained using the magnitude condition as
K = − 4 ⋅ − 4 + 1 ⋅ − 4 + 3 = 12

Example 7
The open-loop transfer function of a system is
K
G (s) H (s) =
s ( s + 2 s + 2)
2

Sketch the root locus of the system as a function of K

Solution:
The open-loop poles are at s = 0, -1-j, -1+j

122
There are no open-loop zeros. Hence there are 3 branches of the locus.
The loci start (K = 0) at s = 0, -1-j and -1+j
The loci end (K = ∞) at infinity
The root locus lies on the whole of the negative real axis
There are no breakaway or break in points on the real axis.
The asymptotes which the loci approach as K→∞ are at the angles
180 0 3 × 180
θ A1 = = 60 0 θ A2 = = 180 0
3 3
5 × 180 0
θ A3 = 300 0
3
The asymptotes centroid is given by
(0 − 1 − j − 1 + j ) − (0) 2
σA = =−
3 3
The loci cross the imaginary axis at the point s = jω
The characteristic equation is
φ ( s) = s 3 + 2s 2 + 2s + K = 0
Substituting s = jω into the characteristic equation, we have
φ ( jω ) = [−2ω 2 + K ] + j[−ω (ω 2 − 2)] = 0
Solving the equation yields
ω = ± 2 and K = 4
Thus the loci cross the jω axis at s = ± j 2 when K = 4
The angle of departure from the pole at s = -1 +j is given by
ψ 1 = 0 − 1350 − 90 0 + 180 0 = −45
And for the pole at -1-j
ψ 2 = 450

123
 jω K = ∞

2
XK = 0
p1
1

600 K = 0
x
K=∞ -3 -2 -1 -600 p3 σ

-1
p2X K = 0
ψ j = ∑ θ zi − ∑ θ pi + 180
i≠ j

There is no zero; hence -2

θz = 0
θp2 = 900
θp3 = 1350 = 1800 - tan-1(1/1)
K=∞

Example 8
Sketch the root locus for the system with open-loop transfer function given
by
K ( s + 1)
G (s) H (s) =
s + 3s + 3.25
2

Show that part of the root locus is circular.

Solution:
Two open-loop poles: sp1 = -1.5+j, sp2 = -1.5-j
One open-loop pole zero: sz1 = -1
Hence only one branch goes to infinity (N = P - Z = 1)
The loci start (K = 0) at sp1 and sp2
For K = ∞, one branch terminates at sz1 the other terminates at infinity.
On the real axis, the locus lies between -∞ and -1.
Breakaway/Break-in points:
The characteristic equation is

124
 φ (s) = 1 + G ( s) H ( s) = 0
K ( s + 1)
= 1+ =0
s 2 3s + 3.25
( s 2 + 3s + 3.25)
⇒K =−
s +1

dK ⎛ ( s + 1)(2 s + 3) − ( s 2 + 3s + 3.25) ⎞
= −⎜⎜ ⎟⎟ = 0
ds ⎝ ( s + 1) 2
⎠
∴ s + 2 s − 0.25 = 0 ⇒ s1, 2 = −1 ± 1.12
2

Hence the break-in point is sB = -2.12.

Angle of Departure from the complex poles:
ψ 1 = θ z1 − θ p 2 + 180 0 = 116.56 0 − 90 0 + 180 0 = 206.56 0
or
ψ 1 = −153.44 0
Hence
ψ 2 = 153.44 0

jω
xK = 0
p1 1

K=∞
-3 -2 -1z1 0 σ
K=∞ -2.12

K=0
x
p2 -1

To prove that part of the locus is on a circle, there are two ways to go about
it.
a) The characteristic equation is
φ ( s ) = s 2 + 3s + 3.25 + K ( s + 1) = 0

125
 Using the quadratic formula,
− (3 + K ) ± (3 + K ) 2 − 4(3.25 + K )
s1, 2 =
2
Let
− (3 + K ) ± (3 + K ) 2 − 4(3.25 + K )
s1, 2 = = σ ± jω
2
Therefore,
(3 + K )
σ =− ⇒ K = −( 2σ + 3)
2
and
1
ω= 4(3.25 + K ) − (3 + K ) 2
2
4(3.25 − 2σ − 3) − 4σ 2
⇒ω =
2

4
Hence the roots satisfy the equation
4σ 2 + 4ω 2 + 8σ = 1
or
σ 2 + ω 2 + 2σ = 0.25
This is the equation of a circle with centre at σ = -1, ω = 0 and radius =
1.12

b) Alternatively all points on the locus satisfy the angle condition, i.e.,
∠G ( s ) H ( s ) = ∠s + 1 − ∠s + 1 − j − ∠s + 1 + j = 180 0
Let s = σ +jω so that
∠σ + jω + 1 − ∠σ + jω + 1 − j − ∠σ + jω + 1 + j = 180 0
That is,
ω ω −1 ω +1
tan −1 − tan −1 = tan −1 + 180 0
σ +1 σ +1 σ +1
Take tangents of both sides and simplify:
σ +1 ω +1
=
(σ + 1) 2 + ω (ω − 1) σ + 1
Re-arrangement and simplification result in an equation of a circle.

126