Control System Synthesis by Root

Locus Method
WALTER R. EVANS
MEMBER AIEE

Synopsis: The root locus method determines all of the roots of the differential
equation of a control system by a graphical
plot which readily permits synthesis for
desired transient response or frequency
response. The base points for this plot
On the complex plane are the zeros and poles
of the open loop transfer function, which are

readily available. The locus of roots is a

plot of the values of s which make this

transfer function equal to -1 as loop gain is

increased from zero to infinity. The plot
can be established in approximate form by
inspection and the significant parts of the

locus calculated accurately and quickly by

use of a simple device. For multiple loop

systems, one solves the innermost loop
first, which then permits the next loop to be
solved by another root locus plot. The
resultant plot gives a complete picture of
the system, which is particularly valuable
for unusual systems or those which have
wide variations in parameters.
HE root locus method is the result of
an effort to determine the roots of the
differential equation of a control system
bv using the concepts now associated
with frequency response methods.' The

roots are desired, of course, because they

describe the natural response of the system. The simplifying feature of the
control system problem is that the open
loop transfer function is known as a
product of terms. Each term, such as
1/(1+Ts), can be easily treated in the
same manner as an admittance such as
1/ (R+jx). It is treated as a vector in the
sense used by electrical engineers in
sligaccircuits. The phase shift and
attenuation of a signal of the form es1
being transmitted is represented by
1/(1+ Ts) in which 8 in general is a complex number. The key idea in the root
locus method is that the values of 8
whicb make transfer function around the
loop equal to -1 are roots of the differ-

ential equation of the system.

66

The opening section in this paper,

Background Theory, outlines the over-all
pattern of

analysis. The following sec-

tion on Root Locus Plot points out the

great usefulness of knowing factors of the

open loop transfer function in finding the
roots.
roots
The graphical nature of the method
requires that specific examples be
used to demonstrate the method itself

O (s) = l

+K,GA (s)Ka (s)

The problem of finding the roots of the

differential

equation here appears in the

form of finding values of s which make

the denominator zero. After these values
are determined by the root locus method,
the denominator can be expressed in
factored form. The zeros of the function
Oo/Oi can be seen from equation 1 to be the
zeros of G;,(s) and the poles of Ga(s).
The function can now be expressed as
h
i

Go
is

(1-s/qi)(1-s/q2)

(1- s/rl) (1- s/r2) (2)

the topLis: Single Loop Example

ulder
Multiple Loop System, and Corrective

Tep constan Kpeand theeo t fo

depend upon the specific system but for
control systems y is often zero and Kc is
often 1
The full power of the Laplace Transform2 or an equivalent method now can
be used. The transient response of the
output for a unit step input, for example,
is given by equation 3

classic problem of solving an nth degree

Go(t)=1-

Networks. The topic Correlation with

Other Methods suggests methods by
which experience in frequency methods
.
can be extended to this method. The
topic Other Applications includes the

polynomial. Finally, the section on

Graphical Calculations describes the key
features of a plastic device called a

"Spirule",

which

permits

calculations to

be
plotmade from direct measurement on the

Plot.

Backgroundstituting

s=j

The over-all pattern of analysis can be

outlined before explaining the technique
of sketching a root locus plot. Thus
consider the general single loop system
shown in Figurel.
Note that each transfer function is of
the form KG(s) in which K is a static gain
constant and G(s) is a function of the
complex number. In general, G(s) has

A etit

(3)

The amplitude Ai is given by equation

A-A =-

-V

srz
The closed loop frequency response, on
the other hand, can be obtained by subinto equation 2. Fortunately, the calculation in finding Ai or
6o/60(jw) involves the same problem of
multiplying vectors that arises in making
a root locus plot, and can be calculated
quickly from the resultant root locus
plot.

in factored form. The values of s which

make the function zero or infinite can
therefore be seen by inspection and are
called zeros and poles respectively. The
closed loop transfer function can be ex-

Paper 50-il, recommended by the AIEE Feedbackand a proed by th

Control Systems pCommittecomtpe
tion at the AIEE Winter Genlerai Meeting, New
York, N. Y., January 30-February 3, 1950. Manuable for printing November 22, 1949.
WALTER R. EVANS is with North American Aviation, Inc., Downey, Calif.
The assitance givhentbyhis fellow woPrkers, K. R.

pressed directly from Figure 1 as given in

paper.

both numerator and denominator known

equation 1

Evrans-Control System Synthesis

R. M. Osborn, in the preparation of this

Jackson and
In particular, Mr. Osborn contributed the

circuit analysis example.

AJEE TRANSACTIONS

ei

Figure 1 (left). General block diagram

80

8r
'~~~~~~~~.

IU

iw

(S

Figure 3 (right). Single loop root locus

jWc

IL

LOCUS

OF S FOR
lo

00+ 01 + 0. -1

eS~.8

-_*
00

Root Locus Plot

The open loop transfer function is
typically of the form given in equation- 5.

K,MG(s)KgG6(s)
+ 03 2
K(lTs032
s(1+ Tis)[(S+0)2+ 32]

(5)

The parameters such as T, are constant

for a given problem, whereas s assumes
many values; therefore, it is convenient
to convert equation 5 to the form of
equation 6.

K,,G,(s)K#Gp(s)

K(l/T2+s)T2 [032+cW32]
s(1/T1+ s)T1[(s+ 03+jco3) (s+ U3-JCO3I

(6)
The poles and zeros of the function are
plotted and a general value of s is assumed as shown in Figure 2.
Note that poles are represented as dots,
and zeros as crosses. All of the complex
terms involved in equation 6 are represented by vectors with heads at the general point s and tails at the zeros or poles.
The angle of each vector is measured with
respect to a line parallel to the positive
real axis. The magnitude of each vector
is simply its length on the plot.
In seeking to find the values of s which

make the open loop function equal to -1,

the value -1 is considered as a vector

whose angle is 180 degrees =rn 360 degrees, where n is an integer, and whose
magnitude is unity. Then one can consider first the problem of finding the locus
of values for which the angle condition
alone is satisfied. In general, one pictures
the exploratory s point at various positions on the plane, and imagines the lines
from the poles and zeros to be constructed

_S + b3 -

-w3
- --<
3
>e 5s
/ /
1
+1
-> \'~
,- e/
4
/
_IT,
-I/Tz
/
/

O/

-C;-Jw

Figure 2. Root locus plot

1950, VOLUME 69

so that the angles in turn can be visualized. For any specific problem, however,
many special parts of the locus are established by inspection as illustrated in
examples in later sections. Surprisingly
few trial positions of the s point need be
assumed to permit the complete locus to
be sketched.
After the locus has been determined,
one considers the second condition for a
root, that is, that the magnitude of
KMG,(s)KgGp(s) be unity. In general,
one selects a particular value of s along
the locus, estimates the lengths of the
vectors, and calculates the static gain
K,,K = 1/G,(s)Gf(s). After acquiring
some experience, one usually can select
the desired position of a dominant root to
determine the allowable loop gain. The
position of roots along other parts of the
locus usually can be determined with less
than two trials each.
An interesting fact to note from equation 6 is that for very low gain, the roots
are very close to the poles in order that
corresponding vectors be very small. For
very high gain, the roots approach infinity or terminate on a zero.

Single Loop Example

Consider a single loop system such as

shown in Figure 1 in which the transfer
functions are given in equation 7.
K
KMG$(s) =
K+ G(s) = 1(7)
(1 + Tis) (1 + T2s)s'
The poles of the open loop function are
at 0, - 1/Ti and -1/T2 as represented by
dots in Figure 3.
The locus along the real axis is determined by inspection because all of the
angles are either 0 degrees or 180 degrees.
An
of angles must therefore
<r+
j
",^odd
/ /number
\
be 180 degrees as shown by the intervals
between 0 and- 1/T1, and from -1/T2
to - o. Along the ico axis, 4)o is 90 deso that 4)i must be the complement
> , ~~~~grees
of 4)i as estimated at
For very
large values of a, all angles are essentially
equal so the locus for the complex roots
finally approaches radial lines at 60 degrees.
The point where the locus breaks away
from the real axis is found by considering

a=jac.

Evans-Control System Synthesis

a value of s just above the real axis. The

decrease in Oo from 180 degrees can be
made equal to the sum of 41 and O2 if the
reciprocal of the length from the trial
point to the origin is equal to the sum of
the reciprocals of lengths from the trial
point to - 1/T1 and -1/T2. If a damping
ratio of 0.5 for the complex roots is desired, the roots ri and r2 are fixed by the
intersection with the locus of radial lines
at 60 degrees with respect to negative
real axis.
In calculating K for 8= ri, it is convenient to consider a term (1 + Tls) as a
ratio of lengths from the pole - 1/Ti to
the s point and from a to the origin respectively. After making gain K = 1/[G(s)],$=, a good first trial for finding r8
is to assume that it is near - 1/T2 and
solve for (1/T2+s). After the roots are
determined to the desired accuracy, the
over-all transfer function can be expressed
as given in equation 8.
1
/
1
1
(i(8)
r1 r2/
r1
The procedure in handling a multiple
loop system now can be explained.

Multiple Loop System

Consider a multiple loop system in
which the single loop system just solved
is the forward path of another loop, as
shown in Figure 4.
0,/0i is given in factored form by equation 8 so the roots of the inner loop now
serve as base points for the new loeus plot.
For convenience, however, neglect the
effect of the term (1-s/re) so that the
locus for the outer loop is shown in Figure
5.
The locus for the outer loop would be a
circle about the -1/T point as a center
if the effect of 00/CL were completely neglected. Actually, the vectors from the
points ri and r2 introduce net angles so
that the locus is modified as shown. The

67

Io

<

build-up rate, overshoot, natural

frequency of oscillation, and the damping

mum
r

< >

rate as effective clues in solving this

problem.

Other Applications
i

Figure 4 (above). Multiple loop block diagram

Figure 5 (right). Multiple loop root locus

angle at which the locus emerges from ri
can be found by considering a value of s
close to the point ri, and solving for
the angle of the vector ( s-ri).
Assume that the static loop gain desired is higher than that allowed by the
given system. The first modification
suggested by the plot is to move the ri
and r2 points farther to the left by obtaining greater damping in the inner loop. If
these points are moved far to the left, the
loci from these points terminate in the
negative real axis and the loci from the
origin curve back and cross the jw axis.
Moving the -1/T point closer to the
origin would then be effective in permit-

ting still higher loop gain. The next as-

pect of synthesis involves adding corrective networks.

Corrective Networks
Consider a somewhat unusual system
which arises in instrument servos whose
open loop transfer function is identified
by the poles pi and P2 in Figure 6(A). As
loop gain is increased from zero, the roots
which start from pi and P2 move directly
toward the unstable half plane. These
roots could be made to move away from
the jc& axis if 180 degrees phase shift were
added. A simple network to add is three
lag networks in series, each having a time
constant T such that 60 degrees phase
shift is introduced at p,. The resultant
locus plot is shown in Figure 6(B).
The gain now is limited only by the requirement that the new pair of roots do
not cross the jcw axis. A value of gain is
selected to obtain critical damping of
these roots and the corresponding positions of all the roots are shown in Figures
6(A) and 6(B) as small circles.
Actually, greater damping could be
achieved for roots which originate at Pi
and P2 if a phase shifting bridge were used
rather than the 3-lag networks. Its
transfer function is (3- Ts)/(1+ Ts) and is
of the "nonminimum phase" type of circuit.
Since these types of correction are
somewhat unusual, it is perhaps well to
point out that the analysis has been yenifled by actual test and application.

68

\-l/

These examples serve to indicate the

reasoning process in synthesizing a control system by root locus method. An
engineer draws upon all of his experience,
however, in seeking to improve a given
system; therefore, it is well to indicate
the correlation between this method and
other methods.

Correlation with Other Methods

The valuable concepts of frequency response methodsi are in a sense merely extended by the root locus system. Thus
a transfer function with s having a complex value rather than just a pure imaginary value corresponds to a damped sinusoid being transmitted rather than an undamped one. The frequency and gain
for which the Nyquist plot passes through
the -1 point are exactly the same values
for which the root locus crosses the jw,
axis. Many other correlations appear in
solving a single problem by both methods.
The results of root locus analysis can be
easily converted to frequency response
data. Thus one merely assumes values of
s along the jco axis, estimates the phase
angles and vector lengths to the zeros and
poles, and calculates the sum of the
angles for total phase shift and the product of lengths for attenuation. The inverse problem of determining zeros and
poles from experimental data is the more
difficult one. Many techniques are already available, however, such as drawing
asymptotes to the logarithmic attenuation curve. For unusual cases, particularly those in which resonant peaks are
involved, the conformal mapping technique originated by Dr. Profos of Switzerland is recommended.3
The transient response is described by
the poles of- the transfer function. The2
inverse problem in this-case is to locate the
poles from an experimental transient response. One might use dead time, maxi-

Evans-Control System Synthesis

Many systems require a set of simul-

equations to describe them and

~~~taneous
t

are said to be multicoupled. The corresponding block diagrams have several inputs to each loop so that the root locus
method cannot be applied immediately.
One should first lay out the diagram so
that the main line of action of the signals
forms the main loop with incidental
coupling effects appearing as feedbacks
and feed forwards. One then proceeds to
isolate loops by replacing a signal which
comes from within a loop by an equivalent
signal at the output, replacing a signal

entering a loop by an equivalent signal at

the input. One can and should keep the
physical picture of the equivalent system
in mind as these manipulations are car-

ried out.
The techniques of the root locus method
can be used effectively in analyzing electric circuits. As a simple example, consider the lead-lag network of Figure
7(A).
It can be shown that the transfer function of this network is as given in equa-

tion 9

V0

(1+RiCis)(1R2C2s)R3

(1+R2C2S)R3+
VE (1+RiCis)(1+RiCiS)Ri+
Rl [1+ (R2+Ra) C2s]
(9)
The denominator can be factored algebraically by multiplying out and finding
the zeros of the resulting quadratic. As
an alternative, it will be noted that the
zeros of the denominator must satisfy

equation 10

Pi

owl

pi

o
/\

0r-o
v

Figure 6. (A) Basic system. (B) Corrected

system

AIFE TRANSACTIONS

~~~~~~~~Figure 7 (left). (A)

Cs

Circuit diagram. (B)

Root locus
Fisure 8 (right).

R,0,

5c.

(A*R.c2

10

order term. Solve for the roots of the Several procedures are possible, but the
first loop which corresponds to the quan- over-all purpose is to successively rotate
tities in brackets above and proceed as the arm with respect to the disk through
for the multiple loop system. If each of the angles of interest. Thus for
before
(1/R1CI+a)(1/R2C2+s)R3(R1C1)(R2c2)
the
roots
close to the origin are of most adding phase angles, the disk is held
[1/(R2+Rs)C2+aj(R2+RP)C2RI
=-1 (10) interest, substitute J= l/x first and solve fixed while the arm is rotated from a pole
for root values of x. Other combinations to the horizontal, whereas the two move
repare, of course, possible because a single together in getting aligned on the next
this exre
root locus basically determines the factors pole. For multiplying lengths, the disk is
held fixed while the arm is rotated from
scheme in Figure 7(B). Theto
The two roots of the sum of two terms,
are
shown bycth two
The root locus method is thus an an- the position where a pole is on the straight
dots thereby
the
Ther
latis
alytical tool which can be applied to line to the position where the pole is on
dotsuand
cross.ma
exaccatey
could be estphicat medhor s.
other problems than control system syn- the logarithmic curve. Rotations are
thesis
for which it was developed. But in made in the opposite directions for zeros
miedlocus
The byfgraph
locus of rootsmodis.
now is simply.insimply inthan they are for poles.
a new problem one would probattacking
tervals along the negative real axis between the open loop zeros and poles as ably do well to try first to develop a
shown in Figure 7 (B). The exact location method of analysis which is natural for Conclusions
of the roots along these intervals is deter- that problem rather than seek to apply
The definite opinion of engineers using
mined in the usual way. Note that the any existing methods.*
this method is that its prime advantage
G
constant
in equation
is of the
form
R'Cot in which
R' is the10effective
value
of Graphical
is the complete picture of a system which
Calculations
effective
in
the
which
R'is
value ofthe root locus plot presents. Changing
R'Ci,
The root locus plot is first established an open loop parameter merely shifts a
R2 and Ra in parallel.
In more complicated networks, the ad- in approximate form by inspection. Any pointandmodifiesthelocus. Bymeansof
vantages of the root locus concept over sinfcn
fombinpcinAn
tlocus
the root locus method, all of the zeros and
significant point on the locus then can be potadmdfeshelu.Byensf
bec
i
checked
the
indicated
poles of the over-all function can be deby
using
techniques
particular advantage oS t retalning at all in this section. Note that only two cal- termined.
Any linear system is completely deculations are involved, adding angles and
over-all netwk
betwen
th ove-allnetwrk
multiplying
lengths.
its reFortunately,
all
of
fined
determnination, andfunction
and the parameters of individual circuit these angles and lengths can be measured sponsebyto this
input
any
particular
elements,
at the a point. Thus angles previously can be determined readily by standard
In the
classical
of
problem
finding*
the
classifereal problen
fivendingin pictured at the zeros and poles also ap- mathematical or graphical methods.
roots, the
differential
equation is given
the form of a sum of terms of successively pear at the s point but between a hori- Ref
higher order. This can be converted to
OF SERVOMECHANISMS (book),
zeros and poles. A piece of transparent
i. S.PRINCIPLES
the form shown in equation 10
G.
John Wiley and
Brown, D. P. Campbell.
paper or plastic pivoted at the s point can Sons,
New York, N. Y., 1948.
each
of
be
rotated
successively through
s'+as +bs'2+
2. TRANsIrNTs IN LINBAR SYSTBMS (book).
M.F. Gardner,J.L.Barnes. JohnWileyandSons,
-[(s+a)s+b]s+ . .. +m (11) these angles to obtain their sum.
The reader can duplicate the "spirule" New York, N. Y., 1942.
ANALYSIS OF CONTROL SYSTBMS,
This corresponds to a block diagram with two pieces of transparent paper, one 3.W. GRAPHIcAL
R. Evans. AIEE Transactions, volume 67,
with another loop closed for each higher for the disk and the other for the arm. 1948, pages 547-51.

Tesentdvectorsding
rscentedinaccordingBto

are
rootlus

broundedas.

vantagebraic metherod
reatoer
palgbrticularhadvantagerisein greatainin ats
betwmee thear
parelametensh
cirmeter
Inots,

No Discussion

1950, VOLUME 69

Evans-Control System Synthesis

69