Control Systems

Notes
Prepared by

Dr.D.Vanitha,
Assistant Professor in EEE Dept. of SCSVMV University

Syllabus

UNIT – I SYSTEM AND THEIR REPRESENTATION Page 2-88

Basic elements of control systems- open and close loop systems – Differential
equation - Transfer function – Modelling of Electrical systems, translational and
rotational mechanical systems – Block diagram reduction techniques – Signal flow
graphs.
UNIT – II TIME RESPONSE Page 89-128

Time response – Time domain specifications – types of input – I and II order

system response – Error coefficients – Generalized error series – Steady state error
–Effect of P,PI,PD and PID modes of feedback control ,Analysis using MATLAB.
UNIT – III FREQUENCY RESPONSE Page 129-184

Frequency response – Bode plot – Polar plot – Nyquist plot – Frequency domain
specifications from plots – Constant M and N circles – Nichol’s chart– Analysis using
MATLAB.
UNIT- IV STABILITY AND COMPENSATOR DESIGN Page 185- 243

Characteristic equation – BIBO stability - Routh Hurwitz criterion - Root locus

technique Construction of Root locus - Nyquist stability criterion – Effect of Lag,
Lead and lag-lead compensation on frequency response, Analysis using MATLAB.

UNIT- V STATE VARIABLE ANALYSIS Page 244-259

Concept of state variables – State models for linear and time invariant systems –
solution of state and output equation in controllable canonical form – concept of
controllability and observability – Effect of state feedback
 UNIT – I
SYSTEM AND THEIR REPRESENTATION
Basic elements in control systems – open and closed loop system – Electrical
analogy of mechanical and thermal systems – Transfer function – Synchros – AC and DC
Servo motors – SFG and BDR methods.
System :- A group or combination of elements connected together to perform a specific
function or task
Example :- (i) Human body system consist of digestive and respiratory system.
(ii) Universe – System, solar consist of star, solar family, planets etc.,
Control Systems :- In a system when the output quantity is controlled by varying the in
quantity is called “ Control System”
1.1 Classification of Control System:-
According to their types of components connected in the system it may be
classified as.
(i) Mechanical System
(ii) Electrical System
a. Non Linear
b. Linear
1. Continuous system
2. Discrete system
(iii) Hydraulic System
(iv) Pneumatic System
1. Linear System:- In this system the output has linear (Characteristics)
relationship with the input, if the system satisfies the Superposition and
Homogenity principle then the system is said to be “Linear System”.
2. Non Linear System:- In this, the output does not have linear relationship with
the input. It neither satisfies principle of Superposition nor principle of
Homogeneity
3. Time Variant System :- When the parameters of the mathematical model are
varying with time i.e. Non-stationary
Example:- Guidance and control of a rocket
Mass of the rocket changes with respect to time due to the depletion
of the fuel (petrol / diesel)
A d x/dt2 + B dx/dt +Ct = 0
2

4. Time – Invariant System :- It is one which the system is independent of time

Example: - A d2x/dy2 + dx/dy B + C = 0
 Continuous System i.e, It is one which the system variables are the function of
a continuous time variable “ t “
Example:- Speed control of DC motor using field control.
5. Discrete System:- It is one which system variables are the function of discrete
(not continuous ) instants of time. They are not continuous dependent on the
time.
6. Distributed system / parameter:- Significant variables in a system are
distributed in space and they vary with the spatial co-ordinates and time. This
models are called Distributed parameters / system.
Example:- T/L – R,L & C
7. Lumped Parameters / Models / System:- Substance of matter is assumed to
be lumped at some discrete points of the space.
8. Stochastic models :- In physical system we are (expect) meet some
uncertainties in such case the parameters / variable are random in function of
time and they are described by probability distribution function that are
known as “stochastic models”
9. Deterministic models: - All the quantities have definite (Known) values in the
model that is known as deterministic models.
1.1.1 Open loop system:-
In an open loop system the control action is independent of the output

In open loop system the output can be varied by varying the input
Merits:-
1. Open loop control is much simpler and less expensive.
2. No sensors are needed to measure the variables to provide feedback.
Limitations:-
1. No accuracy.
2. There is no compensation for any disturbances entering into system since it has
fixed output
3. Its performance is highly depends on the properties of the plant which some time
may vary with respect to time.
1.1.1.Closed loop system (Automatic / Feedback system):-
Control system in which the output has an effect upon the input quantity in such
a manner as to maintain the desired value of output are called “Closed loop system“
Merits:-
1. It can be designed to provide extreme accuracy at the steady order.
2. Its time response can be adjusted by appropriate design.
3. It compensates for disturbance
4. Transient response and steady state response can be controlled more
conveniently and with great flexibility.
Demerits:-
1. More complex and more expensive
2. Reduces the gain of the system
3. If the closed loop system is not properly designed, the feedback system may
lead to an oscillatory system.
1.2 Transfer Function Models:-
Transfer function:-
Definition:-
Transfer function of a time – invariant system, is the ratio of the laplace transforms of
its output and input variables, assuming zero initial conditions

Where Y(S) = Laplace transform of the output variable y(t)

R(S) = Laplace transform of the input variable r(t)
G(S) = Transfer function of the system
For Lumped linear time invariant system mathematical laws normally results in a set of
(1st and 2nd order) differential equations.
Applying Laplace transform to the differential equations results in a transfer function
model of the system.
(a) Mathematical Modelling of mechanical systems:-
It is further classified into two types based on their output
(i) Translational system (i.e) output of the system is linear displacement
(ii) Rotational system (i.e) output of the systems is rotational motion.
1.2.1.Translational system:-
The model of mechanical translational systems can be obtained by using three
elements Mass, Spring and Dash – pot.
1. The weight of the mechanical system is represented by the element Mass and it
is assumed to be concentrated at the centre of the body.
2. The elastics deformation of the body can be represented by a spring.
3. The friction existing in mechanical system can be represented by the dash-pot.
The dash-pot is a piston moving inside a cylinder filled with viscons fluid.
When a force is applied to a translational mechanical system, it is opposed by opposing
forces due to Mass, friction and elasticity of the system.
The force acting on a mechanical body are governed by Newton’s Second Law of motion.
For translational system it states that the sum of forces acting on a body is zero
[Newton’s Second law states that the sum of applied forces is equal to the sum of
opposing forces on a body]

In translational system,
x = displacement (m)
v = velocity m/s = dx/dt
a = acceleration m/s2 = dv/dt = d2x/dt2
f = applied force (input)
fm = opposing force due to Mass.
fK = opposing force due to Mass elasticity of the body (spring)
Fb = opposing force due to friction (Dash-pot) of the body
1.2.2 .Force balance equation of idealised elements (one end is fixed)
(i) Mass: - Assume B & K = 0 (only M is acting)
Mass will offer an opposing force which is proportional to the acceleration of the body
f = applied force
fm = opposing force due to mass.
fm α d2x (t)/dt2 (remove the proportionality M – constant)
fm = M d2x(t)/dt2
According Newton Law,
Applied force = opposing force
f = fm
f = M d2x(t)/dt2
take laplace f = MS2 X (S)
(ii) Dash-pot :- M & K - Neglected.
Let force applied to dash-pot.

It offers an opposing force which is proportional to the velocity of the body.

fb α dx(t)/dt
fb = B. dx(t)/dt
Newton’s law F = fB
f = B. dx (t)/dt
take laplace F = BSX (S)
(ii) Both side are free ends : (initial – final)

fb α d/dt (x1(t) – x2 (t))

fb = B d/dt [x1 (t) – x2 (t)]
take laplace F = BS[x1(S) – x2 (S)]
(iii) Elastic Element (Spring):-

M and B – Neglected
Let a force is applied on spring
The spring will offer an opposing force which proportional to displacement of the body
Fk α x(t)
Newton’s law f = KX(t) (K – Proportionality constant)
Take laplace F = KX(S)
(b) Displacement at both ends:-

fk = [x1(t) – x2(t)]
f = fk = K(x1(t) – x2(t))
take Laplace F = K [x1(S) – x2(S)]

1.2.3.Guidelines to obtain mechanical model (translational system):-

(i) write force balance equation for each mode ( each mass acts as a
node)
(ii) Draw separate free body diagram for each mass
(a)mark all forces acting on it
(b)always opposing force acts in a direction opposite to applied force
(c) Mass is move in the direction of applied force.
(iii) For each free body diagram write one differential equation by
equating the sum of applied forces to the sum of opposing forces.
1.2.4. PROBLEMS
(1) Obtain transfer function X(S)/F(S)

(i) Free body diagram

 (ii) from the diagram
f(t) = m d2x(t)/dt2 + B dx(t)/dt + K x(t)
take Laplace transform
F(S) = MS2 X(S) + BS X(S) + KX(S)
F(S) = [MS2+BS+K] X(S)
Output/input = X(S)/F(S)= 1/MS2+BS+K

(2) Determine X(S) / F(S) of the system shown in figure.

Two Mass M1 & M2 are there.

(i) Free body diagram for Mass-1-M1

f(t) = M1 d2y1(t)/dt2 + B dy1(t)/dt + K2 (y1(t) – y2 (t)) + K1 y1(t)

take laplace F(S) =MS2y1(S)+BS y1(S)+ K1y1(S)+K2y1(S)-k2y2(S)
[MS2+BS+K1+K2]Y1(S) – K2Y2(S) = F(S) ---- 1
(ii) M2 – node – 2 – Free body diagram

0 = Md2y2(t)/dt2 + K2(y2(t) – y1(S))

Take laplace MS2Y2(S) + K2Y2(S) – K2Y1(S) = 0 ---- 2
Y2(S) [M2S2+K2] – X1(S) K2 = 0
Y1(S) = Y2(S) [M2S2 + K2]/K2 ----- 3
Substitute equation (3) in the equation (1) we get
Y2(s) M2s2+K2/K2 [M1S2+BS+K1+K2] – Y2(S) K2 = F(S)
Y2S[M2S2+K2/K2 (M1S2+BS+K1+K2) – K2]
Transfer function = Y2(S)/F(S) = K2 / (M2S2+K2) (M1S2+BS+K1+K2) – K2
(3) Obtain for the mechanical system shown in the figure. Find the transfer function
X(S) / F(S)

M is the first mass i.e node – Free body diagram is

f(t) = M d2x(t)/dt2 + K2[x(t) – x1(t)]

take laplace F(S) = [MS2+K2] X(S) – X1(S) ---- 1
If more than one element connected with without mass. Assume at that point as new
node with mass value is M = 0 & displacement x1
 K1x1(t) + K2 [x1(t) – x(t)] = 0
Take laplace K1X1(S) + K2 [X1(S) – X(S)] = 0 ---- 2
X1(S) = K2 X(S) / K1+K2 ---- 3
Substitute equ.3 in equ.1
F(S) = (MS2+K2) X(S) – K22 / K1+K2 X(S)
F(S) / X(S) = [(MS2+K2) (K1+K2) – K22 / K1+K2]
Transfer function Output/ input = X(S) /F(S) = [K 1+K2 / (MS2+K2) (K1+K2) – K22]
(4).For mechanical system obtain X(S) / F(S)

(i) free body diagram for Mass – M

Md2x(t)/dt2 + B dx(t)/dt + Kx(t) + K1[(x(t) – y (t)] = 0

Take laplace transform

MS2X(S) + BSX(S) + KX(S) + K1X(S) – K1Y(S) = 0
K1Y(S) = X(S) [MS2+BS+K1+K]
Y(S) = X(S) / K1 [MS2+BS+K+K1]
2ND node [even though there is no mass force is acting at the end of the spring in b/w
input and spring consider a node with Mass is zero
 f(t) = K1(y(t) – x (t))
take laplace F(S) = K1Y(S) – K1X(S)
Substitute
F(S) = K1 X(S)/K1 [MS2+BS+K+K1] – K1X(S)
F(S) = X(S)[MS2+BS+K+K1-K1]
F(S) = X(S)[MS2+BS+K]
X(S)/F(S) = 1 / MS2+BS+K

# find the transfer function Y1(S) / F(S) = ?

In the given diagram 2 – masses are there.so 2 –nodes

At node – 1
f(t) = K2y2(t) + M2d2y2(t)/dt2 + K1[y2(t) – y1(t)] + B d/dt(y2(t)-y1(t))
take Laplace transform
F(S) = MS2Y2(S)+K2Y2(S)+K1Y2(S)+BSY2(S)-K1Y1(S)-BSY1(S)
F(S) = Y2(S) [MS2+BS+K1+K2] – Y1(S) [BS+K1] ----- 1

In node 2 , write the equation

K1[y1(t)–y2(t)] + B d/dt (y1(t)–y2(t) + M1d2y1(t)/dt2 = 0
Take Laplace transform
M1S2Y1(S) + K1[Y1(S)-Y2(S)] +BS[Y1(S)-Y2(S) = 0
Y1(S) [M1S2+BS+K1] – Y2(S) [BS+K1] = 0 ----- 2
Y2(S) = Y1(S) [M1S2+BS+K1] / BS+K1 ------ 3
Substitute equ. 3 in equ. 1
F(S) = Y1(S) [M1S2+BS+K1] [M2S2+BS+K1+K2]/ BS+K1 – Y1(S) [BS+K1]
F(S) = Y1(S) [(M1S2+BS+K1) (M2S2+BS+K1+K2] – (BS+K1)2 / (BS+K1)
Y1(S) / F(S) = (BS+K1) / (M1S2+BS+K1] (M2S2+BS+K1+K2) – (BS-K1)2

1.2.4.Mechanical Rotational Systems :-

Model of rotational mechanical system can be obtained by using three elements
(i) Moment of inertial [J] of mass
(ii) Dash pot with rotational frictional co-efficient [B]
(iii) Torsional spring with stiffness [K]
Weight of the rotational system is represented by moment of inertial of the mas. The
moment of inertial of the system (or) body is considered to be concentrated at the
centre of gravity of the body [remaining same in previous system]
When a torque is applied to a rotational mechanical system, it is opposed by
opposing torques due to moment of inertial, friction and elasticity of the system.
According to Newton’s law, torque acting on a body is zero [sum applied torques is equal
to the sum of opposing torques on a body]
In Rotational system:-
Θ – Angular displacement (rad)
dΘ/dt – Angular velocity = rad/sec
d2Θ/dt2 – Angular acceleration rad/sec2
T – Applied torque N-M
J – Moment of inertia kg.m2/rad
B – Rotational frictional co-efficient N-M/ rad/sec
K – Stiffness of the spring N-M/rad
1.2.4.1 Torque Balance equation of idealised elements:-
(i) Moment of inertia:- Assume B & K effect is zero

Opposing torque due to moment of inertia is proportional to the angular acceleration of

the body
Tj α d2Θ (t)/dt2
TJ = J d2Θ(t) /dt2
Apply Newton’s law
T = TJ
T = J d2Θ(t) /dt2
Take Laplace T(S) = JS2Θ(S)

(ii) Dash – Pot

J & K effect is zero

The opposing torque due to dash-pot is proportional to angular velocity of the body
i.e Tb α dΘ(t) / dt
Tb = B dΘ(t) /dt
According to newton’s law
T = TB
T = B dΘ(t)/dt
Take laplace T(S) = BSΘ(S)
Both ends are free ends

T = TB = B d/dt [Θ1(S) – Θ2(S)

 (iii) Spring :-
J & K effect is zero

In that a torque be applied on spring element.

Torsional spring will offer an opposing torque which is proportional to angular
displacement of this body.
i.e. TK α Θ(t)
TK = KΘ(t)
Take Laplace TK(S) = KΘ(S)
Apply Newton’s Law
T(S) = TK(S)
T(S) = KΘ(S)
Both ends are free ends.

TK α [Θ1(t) – Θ2(t)
T(S) = TK(S) = K [Θ1(S)-Θ2(S)]
(5). Obtain Θ(S) / T(S) = ?

(i) Draw free body diagram for each node (J)

T(t) = J d2Θ(t)/dt2 + B dΘ(t)/dt + KΘ(t)

Take Laplace T(S) = JS2Θ(S) + BSΘ(S) + KΘ(S)
 T(S) = Θ(S) [JS2+BS+K]
Θ(S) / T(S) = 1 / JS2+BS+K
(6). Obtain the transfer function Θ(S) / T(S) = ?

J1 & J2 – two nodes

At node – 1

J1d2Θ1(t)/dt2 + B2 d/dt[Θ1(t) – Θ(t)] + K [Θ1(t) – Θ(t)] = T(t)

Take Laplace
J1S2Θ1(S) +B2SΘ1(S) – B2SΘ(S) + KΘ1(S) – KΘ(S) = T(S)
Θ1(S) [J1S2+B2S+K] – Θ(S) [B2S+K] = T(S) ---- 1
At node – 2

J2d2Θ(t)/dt2 +B2 d/dt [Θ1(t) – Θ1(t)] + B dΘ(t)/dt +K (Θ(t) – Θ1(t)] = 0

Take Laplace J2S2Θ(S) + B2SΘ(S) - B2SΘ1(S) + BSΘ(S) + KΘ(S) - KΘ1(S) = 0 ---- 2
Θ(S) [JS2+B2S+BS+K] – Θ1(S) [B2S+K] = 0
Θ1(S) (B2S+K) = Θ(S) [JS2+B2S+BS+K]
Θ1(S) = [JS2+B2S+BS+K] Θ(S) / (B2S+K) ---- 3
Subsitute equ. 3 in equ 1
[J1S2+B2S+K] [JS2+B2S+BS+K] Θ(S) / B2S+K – (B2S+K) Θ(S) = T(S)
Take LCM
Θ(S) [J1S2+B2S+K] [JS2+(B2+B)S+K] – (B2S+K)2 / (BS2+K) = T(S)
Transfer function
Θ(S) / T(S) = (B2S+K) / [(J1S2+B2S+K) (JS2+(B2+B)S+K] – (SB2+K)2
1.3 Electrical systems
To make electrical systems, following basic components are used.
1. Resistance – R (ohms)
2. Inductance – L (hendry)
3. Capacitance - C (farad)
In this all circuit equation are framed by using Kirchoff’s laws.
By applying KVL & KCL we can obtain the following differential equation

Elements Voltage drop Ct. flowing through

V = IR I = V/R

V(t) = L.di(t)/dt I(t) = 1/L ʃ v(t).dt

V(t) = 1/C ʃ i(t).dt I(t) = C.dv(t) /dt

(7). Obtain the transfer function of the electrical network shown in figure

Voltage across the capacitor (C)

1
V2(t) = ∫ 𝑖 (𝑡)𝑑𝑡 ------ 1
𝐶
Take Laplace
V2(S) = I(S) / S.C ------ 2
I(S) = V1(S) – V2(S) ------ 3
Current in a R is equal to
V2(S) = V1(S) – V2(S) /SCR
V2(S) + V2(S) / SCR = V1(S)/SCR
V2S [1+ 1/SCR] = V1(S) 1/SCR
V2(S) / V1(S) = 1/SCR / [SCR+1/SCR]
= SCR /SCR+1 X 1/SCR
therefore transfer function = V2(S)/V1(S) = 1 / SCR+1
(8). Obtain the transfer function of electrical network shown in figure

1
V0(t) =
𝐶2
∫ 𝑖 (𝑡)𝑑𝑡
1
V0(S) = I2(S)
𝑆𝐶2
In loop-1
(L1S + 1/SC1+ R1) I1(S) - I2(S) 1/SC1 = Vi(S)
In loop - 2
-1/SC1 I1(S) + I2(S) [ R2 + 1/SC1 + 1/SC2 ] = 0
Above two equation can be written as in Matrix form
R1+SL+1/SC1 – 1/SC1 I1(S) = V1(S)
-1 / SC1 R2+1/SC1+1/SC2 I2(S) 0

I2(S) = Δ I2(S) / Δ
Δ I2(S) = R1 + SL + 1/SC1 Vi(S)
-1/SC1 0
Δ I2(S) = Vi(S) / SC1
Δ = [R1+SL+1/SC1] [R2+1/SC1+1/SC2] – 1 /S2C12
= R1R2 + R1/SC1+ R1/SC2 +R2SL + SL/SC1 + SL/SC2 + R2/SC1 + 1/S2C12
+ 1/S2C1C2 - 1/S2C12
LCM
Δ = S2[R1R2C1C2 + LC2 + LC2 ] + S3 R2LC1C2 + SC2R1 + SR1C1 + SR2C2 + 1 / S2C1C2
I2(S) = Δ I2(S) / Δ = Vi(S) SC1
= S2[R1R2C1C2+LC2+LC1] + S3R2LC1C2 + SR1C2+SR1C1 +SR2C2+1 / S2C1C2
V0(S) = I2(S) / SC2
 = Vi(S) SC2 / SC2 [S3R2LC1C2] + S2[R1R2C1C2+LC2+LC1] + S[R1C1+R1C2+R2C2] +1
VO(S)/ VI(S) = 1 / S3R2LC1C2 + S2(R1R2C1C2 + LC2 + LC1) + S(R1C1 + R1C2 + R2C2) + 1
(9). Obtain the transfer function for the following diagrams

(Ans - V0(S) / VI(S) = R1R2C1C2S2 / R1R2C1C2S2 + [R1C1+R1C2 + R2C2]S + 1)

(Ans - eo(s) / ei(s) = 1 / RCS+1)

1.4. Electro – Mechanical system

It is the combination of electrical and mechanical system.
(10). Obtain the transfer function X(S)/ E(S) for the electro mechanical system is shown
in figure.Assuming the coil has a back emf of eb = K1dx/dt and coil current i2 produce a
force fc = K2i2
On the mass – M
Electrical side – Two loop are in the figure.
Loop – 1, apply KVL
R1i1(t) + 1/C ʃ[i1(t) – i2(t)] dt = e(t) (1)
Loop – 2
L di2(t)/dt + 1/C ʃ [i2(t) – i1(t)] dt = -eb = -K1 dx(t)/dt (2)
Take laplace
RI1(S) + 1/SC [I1(S) – I2(S)] = E(S) (3)
LSI2(S) + 1/SC [I2(S) – I1(S)] = -Eb(S) = -SK1 X(S) (4)
From the equation (4)
1/SC I1(S) = SLI2(S) + 1/CS i2(S) + SK1X1(S)
1/SC I1(S) = I2(S) [SL + 1/CS] + SK1 X1(S) LCM
I1(S) = [S2LC+1] I2(S) + S2C K1X1(S) CS / CS (5)
Substitute the equation (5) in the equation (3)
Use equation (3)
[ R +1/SC] I1(S) – 1/CS I2(S) = E(S) (3)
[ R + 1/SC] [ (S2LC+1) I2(S) + S2K1C X(S)] – 1/CS I2(S) = E(S)
[S2RLC+SL+R] I2(S) + [S2K1RC+SK1] X(S) = E(S) (6)

From mechanical systems

 According to Newton’s law
fc = fm+fk+fk+fb
fc = fm+2fk+fB
put the relavant values in the above equation
K2I2(S) = [MS2+BS+2K] X(S)
I2(S) = [MS2 + BS + 2K] X(S) / K2 (7)
Sub equation (7) in the equation (6)
[S2RLC + SL + R] [(MS2+BS+2K) / K2] X(S) + [S2K1RC + SK1] X(S) = E(S)
X(S) (S2 RLC + SC + R) (MS2+BS+2K) + K2 (S2K1RC+SK1) / K2

𝑋(𝑆) K2
=
𝐸(𝑆) [S4 RCLM +LS3 [M+RCB+ S2 [RM+LS] + RC (2KL+K1+S)+ (BR + 2KL + K1K2)S + 2KR] ]

1.5. Transfer function of Armature Controlled DC motor

N α Eb / φ
N α V – IaRa / φ
Speed is directly proportional to voltage (armature) and inversely proportional to
flux in the field winding
The speed control system is an electro mechanical system. Electrical system
consist of both Armature and field circuit. But for analysis purpose, only the Armature
circuit is considered because field is excited by a constant voltage.
1.5.1. Armature Controlled DC shunt Motor
The mechanical system consist of the rotating part of the motor and load
connected to the shaft of the motor.

The equivalent circuit of armature is

 By Kirchaoff’s voltage law,
iaRa + La dia/dt + eb = Va (1)

Torque is proportional to the product of flux and armature current [ φ= constant]

T α Ia
T = KtIa (2)
The mechanical system of the motor is

J d2Θ/dt2 + B dΘ/dt = T (3)

Back emf of the dc machine is proportional to the speed (angular velocity) of shaft
eb α dΘ/dt
eb = Kb dΘ/dt (4)
take Laplace for the above equations
Ia(S)Ra + LaSIa(S) + Eb(S) = Va(S) (6)
JS2Θ(S)+ BSΘ(S) = T(S) (7)
Eb(S) = KbSΘ(S) (8)
Equating the equivalent (6) and (7)
KtIa(S) = (JS2 + BS) Θ(S)
Ia(S) = (JS2 + BS)/ Kt Θ(S) (9)
Equation (5) can be written as
(Ra+Sla) Ia(S) + Eb(S) = Va(S) (10)
Substituting for Eb(S) & Ia(S) from equation (8) & (9)
(Ra + SLa) [(JS2 + BS)/kt] Θ(S) + KbsΘ(S) = Va(S)
LCM
(Ra+Sla) (JS2+BS) + KbKtS / Kt Θ(S) = Va(S)

The required transfer function is

Θ(S) / Va(S) = Kt / (Ra+SLa) (JS2+BS) + KbKt S
= Kt / JLaS3 + (RaJ+La B) S2 + (RaB + KbKt)S
Dividing the numerator and denominator by KbKt we get
Θ(S)/Va(S) = [ (1/ Kb) / s[(JLa/KbKt)S2 + [RaJ/KbKt + LaB/KbKt]S + (RaB/KbKt) +1]
Ta = electrical time constant = La/Ra
Tm = mechanical time constsant
r = (RaB / KbKt + 1)
Θ(S)/E(S) = 1/ Kb / S (TaTmS2 + (Tm+ r.Ta)S + (r+1))

1.5.2. Mathematical model of field controlled DC motor

In field controlled DC motor Ar it is maintained at constant value and the
motor is controlled by varing the field current. The schematic diagram filed controlled
DC motor is shown in figure

N α Eb/Θ α V – IaRa / Θ
In the field controlled system, the armature voltage is kept constant and speed is
varied by varying flux of the machine. Since flux is directly proportional to the field
current, flux is varied by varying the field at
Let Rf = field resistance Ω
 Lf = field inductance H
if = field it in A
vf = field voltage V
T = developed Torque by the motor N – M
Ktf = Torque Constant N-m/A
J = Moment of inertia of motor and load Kg-m2/ rad
B = frictional co-efficient of motor and load N-m / (rad/sec)

Applying KVL
Rfif + Lf dif/dt = Vf (1)
T α φ Ia (Ia constant field mtd)
T α φ (φ α If)
T α if T = Ktf if (2)
Mechanical system of the motor is given by
J d2Θ/dt2 + B dΘ/dt = T (3)

Take laplace transform

RfIf(S) + LfSIf(S) = Vf(S) (4)
T(S) = Ktf If(S) (5)
JS2Θ(S) + BSΘ(S) = T(S) (6)
Equating (5) and (6)
Ktf If(S) = JS2Θ(S)+BS Θ(S)
If(S) = ((JS2 + BS)/ Ktf) Θ(S) (7)
Substitute equation (7) into equation (4)
 (Rf + LfS) If = Vf(S)
(Rf + LfS) ((JS2 + BS)/Ktf) ΘS = Vf(S)
Θ(S)/Vf(S) = Ktf / (Rf+SLf) S(B+JS)
= Ktf / S Rf(1+SLf/Rf) B(1+(J/B)S)
Where Km = motor gain constant = Ktf / RfB
Tf = Field time constant = Lf / Rf
Tm = Mechanical time constant = J/B
= (Ktf/ RfB) / S(1+SLf/Rf) (1+ S J/B)
Θ(S)/Vf(S) = Km / S(1+STf) (1+S Tm)

Block Diagram Representation of a Field Controlled DC Motor.

1.6. Analogous System
The system for which the differential as “analogous system”.
The basis for applying the principle of analogy is that two different similar form
are called “analogous”
Hence dynamic characteristics of analogous systems are identical. The
corresponding variables and parameters in two systems represented by equations of the
similar form are called “analogous”.
1.6.1.Electrical analogous of Mechanical translational systems:-
(i) Force – Voltage analogy:
If the dynamic characteristics of an electrical system is identical form of that
of mechanical system then the electrical system is said to be analogous of the
mechanical system. This can be easily found by comparing the mechanical models of
Electric and Mechanical systems. If the mathematical models are identical then they are
said to be analogous systems.
Elements Mechanical System Electrical System
Input Force Voltage Source
Output Velocity Current through the element
F = Kx = K ʃ V.dt e = 1/C ʃ i.dt

Item Mechanical System Electrical System (mesh

analysis)
Input (independent) Force (F) Voltage (e)
Output (dependent) Velocity (V) Current (i)
Displacement (x) Charge (q)
Dissipative element Frictional co-efficient of Resistance (R)
dash-pot (B)
Storage element Mass (M) Inductance (L)
Stiffness of Spring (K) Inverse of Capacitor 1/C

Physical Newton’s 2nd Law ∑F = 0 KVL ∑V = 0

Charge the level of Lever F1/f1 = l1/l2 Transformer e1/e2 =
independent variable N1/N2
Produce to follow the F-V analogy:-
1. Electric circuit elements in series will have same current, like wise in mechanical
system, the elements have same velocity are said to be in series.
2. Elements having same velocity in mechanical system should has same current
in the electrical system.
3. Each mode in the mechanical system corresponds to a closed loop in electrical
system. A mass is considered as a node
4. No. of meshes in the electrical system = No. of masses in the mechanical system
5. Element connected between two (nodes) masses in mechanical system is
represented as a common element between two meshes in electrical analogous system.
(ii) Force – Current Analogy:-
Mechanical system input force opposite Electrical system current source
velocity voltage across the element

Item Mechanical System Electrical System (Nodel

analysis)
Independent (Input) Force (F) Current (i)
Dependent (Output) Velocity (V) Voltage (v)
 Displacement (x) Flux (φ)
Dissipative element Frictional co-efficient Conductance G = 1/R
of dash-pot (B)
Storage element Mass (M) Capacitance (C)
Stiffness of Spring (K) Inverse of Inductance 1/L

Physical Law Newton’s 2nd Law ∑F = KVL ∑i = 0

0
Charge the level of Lever F1/f2 = l1/l2 Transformer i1/i2 = N1/N2
independent variable

Following steps to be followed while solving F – current analogy

1. In electric system, elements in parallel will have same voltage, like wise, in
mechanical system elements having same force are said to be in parallel.
2. Elements having same velocity in mechanical system should have analogous
same voltage in electrical analogous system.
3. Each mass in mechanical system corresponds to a node in electrical system.
(11).Write the differential equations governing to mechanical system shown in figure.
Draw Force – Voltage and force to current electrical analogous circuit. Verify by writing
mesh and node equations.

For Mass – 1 Free body diagram

M1 d2x1/dt2 + B1 dx1/dt + B12 d(x1-x2)/dt + K (x1-x2) = f(t) (1)

Mass – 2 Free body diagram

M2 d2x/dt2 + B2 dx2/dt + K2X2 + B12 d/dt (x2-x1) + K2 (x2-x1) (2)

d2x/dt2 dv/dt
dx/dt v
x ʃ v dt
M1 dv1/dt + B1V1 + B12 (V1-V2) + K1 ʃ (V1-V2) dt = f(t) (3)
M2 dv2/dt + B2V2 + K2 ʃV2 dt + B12 (V2-V1) + K1 ʃ (V2-V1) dt = 0 (4)
a.Force – Voltage analogy:-
1. Mechanical system – 2 nodes,
2. Force applied to Mass – 1 – M1 – represented by a voltage source in first mesh
3. M, B1, K1 and B12 are connected to node (1st)
4. M2, B12, K1, K2 and B2 connected to node (2nd)
5. K1 and B12 common between node (1) and (2)
f(t) ---- e(t)
v1 ---- i1
v2 ---- i2
M1 ---- L1
M2 ---- L2
B1 ---- R1
B2 ---- R2
B12 --- R12
K1 ---- 1/C1
K2 ---- 1/C2

Node – 1 (Mesh – 1)
L1 di1/dt +R1i1+ R12 (i1-i2) + 1/C1 ʃ (i1-i2) dt = e(t) (5)
L2 di2/dt + R2i2 + 1/C2 ʃi2 dt+R12(i2-i1) + 1/C1 ʃ (i2-i1) dt = 0 (6)
 b.Force – current analogy:-
1. Mechanical system ----- 2 – Mass
Electrical system -------- 2 – Nodes
2. Force – M1 – as current source connected to node – 1
M1, B1, K1, B12 --- first Mass
M2, K2, B2, B12, K1 --- 2nd Mass
3. Element K1 and B12 are common between mass -1 and mass – 2
f(t) = i(t)
v1(1) = v1
v(2) = v2
M1 = C1
M2 = C2
B1 = 1/R1
B2 = 1/R2
B12 = 1/R12
K1 = 1/L1
K2 = 1/L2

Rewrite the above equation (4) and (3) as

C1 dv1/dt + 1/R1 V + 1/R12 (V1-V2) + 1/L1 ʃ(V1-V2) dt = i(t) (7)
C2 dv2/dt + 1/R2 V2+ 1/L2 ʃ V2 dt + 1/R12 (V2-V1) + 1/L1 ʃ (V2-V1) dt = 0 (8)
From the equation (7) and (8)
(12).Write differential equation governing the mechanical system shown in figure.
Draw force – voltage and force – current electrical analogous circuits and verify by
writing mesh and node equations.
1.5.3. Electrical analogous of mechanical rotational systems.
Three basic elements moment of inertia, rotational dash-pot and torsional spring
that are used in modelling mechanical rotational systems are analogous to resistance,
inductance and capacitance of electrical system.
The input torque in mechanical system is analogous to either voltage or current
source in electrical system.
The output angular velocity (first derivative of angular displacement) in rotational
system is analogous either voltage or current in an element in electrical system. There
are two types of analogies.
1. Torque – Voltage analogy
2. Torque – Current analogy

i. Torque – Voltage Analogy:-

Item Mechanical system Electrical system
Input Torque (T) Current source (I)
Output Angular velocity (W) Voltage across the element
(V)
 T = KΘ = K ʃw .dt
α angular acceleration
w angular velocity
Θ angular displacement
Momen
t of
inertia

Item Mechanical system Electrical system

Torque voltage Torque current
Input independent Torque (T) Voltage (V) Current (i)
variable
Output dependent Angular velocity (w) Current (i) Voltage (e)
variable
Angular Charge (q) Flux (φ)
displacement (Θ)
Dissipative element Dash pot (B) Resistance (R) Conductance G =
(inverse of R) 1/R
Storage element Moment of inertia Inductance (L) Capacitance
(J)
Physical law Newton’s law ∑T = 0 KVL ∑V = 0 KCL ∑I = 0
Changing the level Gear T1/T2 = n1/n2 e1/e2 = N1/N2 i1/i2 = N2/N1
of independent

(13). Write differential equation governing the mechanical rotational system shown in
figure. Draw Torque – Voltage and Torque – Current electrical analogous circuit and
verify by writing mesh and node equations.
 3 nodes because 3 moment of inertia Θ1, Θ2, Θ3 ----- w1, w2, w3
Free body diagram for J1

j1d2Θ1/dt2 + K1 (Θ1 – Θ2) + B d(Θ1- Θ2)/dt = T (1)

Free body diagram for J2

J2d2Θ/dt2 + B2 d/dt (Θ2- Θ3) + B1 d(Θ2- Θ1)/dt + K1 (Θ2- Θ1) (2)

For J3

J3 d2Θ3/dt2 + K3Θ3 + B2d/dt (Θ3- Θ2) = 0 (3)

d2Θ/dt2 ------ dw/dt
dΘ/dt ----- w
Θ ----- ʃw. dt

J1 dw1/dt + B1 (w1-w2) + K1 ʃ(w1-w2) dt = T (4)

J2 dw2/dt + B2 (w2-w3) + B1 (w2-w1) + K1 ʃ(w2-w1) dt = 0 (5)
J3 dw3/dt + B2 (w3-w2) + K3 ʃw3 dt = 0 (6)
Torque – voltage analogous circuit:-
T1 --- e(t) J1 ----- L1 B1 ---- R1 K1 ---- 1/C1
W1 --- i1 J2 ----- L2 B2 ---- R2 K2 ---- 1/C3
W2 --- i2 J3 ----- L3
W3 --- i3
The above equations can be written as,
L1 di1/dt + R1 (i1-i2) + 1/C1 ʃ (i1-i2) dt = e(t) (7)
L2 di2/dt + R2 (i2-i3) + R1 (i2-i1)+ 1/C1 ʃ (i1-i2) dt = 0 (8)
L3 di3/dt + R2 (i3-i2) + 1/C3 ʃ i3 dt = 0 (9)

Vol Torque – Current analogous circuit:-

T ---- i(t) J1 --- C1 B1 --- 1/R1 K1 --- 1/L1
W1 --- e1 J2 --- C2 B2 --- 1/R2 K2 --- 1/L3
W2 --- e2 J3 --- C3
W3 --- e3
Equation 7,8 and 9 can be written as,
C1 de1/dt + 1/R1 (e1-e2) + 1/L1ʃ (e1-e2) dt = i(t) (10)
C2 de2/dt + 1/R1 (e2-e1) +1/R1 (e2-e3) + 1/L1ʃ (e2-e1) dt = 0 (11)
C3 de3/dt + 1/R2 (e3-e2) + 1/L3ʃ e3 dt = 0 (12)
1.6.BLOCK DIAGRAM REDUCTION METHOD
A system consist of number of blocks (components). The function of each
component is represented by a block. All the blocks are inter connected by the lines with
arrows indicating the flow of signals from output of one block to the input of another
block. Such a block diagram gives an overall ideal of the inter relationships that exist
among various components.
Let us consider a system with transfer function G(S). this system can be
represented by a block as

Reduction method:-
Following rules can be used for Block diagram reduction to find overall
transfer function.
The rules are framed such that any modification made on the diagram does
not alter the input and output relation.
1.6.1. Rules:-
(i) Figure relates the input and output relation that is transfer function

(ii) Take off point:-

Taking off from the input signal does not alter the input relation (signal)

(iii) Blocks in cascade:-

Several blocks connected in series (cascade), the overall transfer function
is determined as follows.

(iv) Summing point:-

Summation of two or more input signal entering in a system output of a
summing point being the sum of the entering input.
(v) Interchanging the summing point:-

(vi) Blocks in Parallel:-

By applying the rule

C(S) / R(S) = transfer function = G1(S) + G2(S) + G3(S)
(vii) Shifting of a take-off point (Branch Point) from a position before a block into a
position after the block

After applying the rule

(viii) Shifting of a take-off point from a position after a block to a position before the
block.

(ix) Shifting of a summing point from a position before a block to a position after the
block.
(x) Shifting of a summing point from a position after a block to a position before the
block.

After applying the rule

(xi) Shifting of a take-off point from a position before a summing point to a position
after the summing point.

After applying the rule

(xii) Shifting of a take-off point from a position after a summing point to a position
before the summing point.

(xiii) Elimination of Feedback

(a) Feedback Element H(S)

From the diagram

C(S) = [R(S) – H(S) C(S)] G(S)
C(S) = R(S) G(S) – H(S) C(S) G(S)
C(S)+H(S) G(S) C(S) = R(S) G(S)
 [1+H(S) G(S)] C(S) = R(S) G(S)
C(S) / R(S) = G(S) / 1+G(S) H(S)
Equivalent block

(b) Unity Feedback - H(S) = 1

C(S) / R(S) = G(S) / 1±G(S)

Equivalent block

(c) Error ratio ie E(S) / R(S)

E(S) = R(S) ± B(S) (1)

B(S) = C(S) H(S) (2)
Substitute (2) in (1)
 E(S) = R(S) ± C(S) H(S)
And also C(S) = E(S) R(S) (3)
E(S) = R(S) ± E(S) G(S) H(S)
E(S) ± E(S) G(S) H(S) = R(S)
E(S) [1±G(S) H(S) ] = R(S)
E(S)/R(S) = 1 / L ± G(S) H(S) Error Ratio
(xiv). When two or more input acts at a time on a system, the total output is obtained
by adding the effect of each individual input separately.

In order to calculate the output due to R1(S) assume R2(S) is zero

While determining the output due to R2(S), R1(S) is considered as zero

Total output = C(S) = C1(S) + C2 (S)

= R1(S) G(S) + R2(S) G(S)
Let R(S) --- Reference input signal
C(S) --- output signal
E(S) --- Error signal
G(S) --- Forward path transfer function
B(S) --- Feedback signal.
H(S) --- Feedback path transfer function
G(S) H(S) – Open loop transfer function
M(S) = C(S)/R(S) = Over all (closed loop) transfer function or control ratio
 E(S) / R(S) --- Error ratio
B(S) / R(S) --- Primary Feedback ratio
(14).Determine over all transfer function = C(S) / R(S)

(i) above diagram is re-drawn as (split the S.P)

(ii) Eliminate the two summing point

(iii) Two blocks in cascade

C(S) /R(S) = (G1+G3) G2 /(1+G2H1)
= G1G2 + G2G3 / 1+G2H1
2. Reduce the block diagram
 (ii) split summing point and shifting the summing point before the block G1 to a
position after the block G1

(ii) blocks cascade and Eliminating the Feedback (II)

C(S) / R(S) = (G1G3 +G3G3) / (1+ G1G3H1)

(15). Determine the transfer function for the closed –loop transfer function Y(S)/R(S)
(M.Gopal Page-239)
(I) Shift the take off point (I) after the block G2 into a position before the block G2

(ii) Block cascade (II) and Parallel (III)

(iii) Shifting a take off point from before the (G2G3+G4) into after the block

(iv) Eliminating the 1st Feedback

= G2G3+G4 / 1+(G2G3+G4)H2
= G2G3+G4 / 1+[G2G3H2 + G4H2]
(iv) Blocks in cascade (II) and Eliminating the Feedback loop (III)
G2H1/G2G3+G4 is feedback with the block
G1(G2G3+G4) / 1+ [G2G3H2 + G4H2]
= G1 (G2G3+G4)/1+ [G2G3H2 + G4H2]
= 1 + G1 (G2G3+G4)/1+ ( [G2G3H2 + G4H2] ) (G2H1/G2G3+G4)
= G1(G2G3 + G4) / 1 +[G2G3H2 + G4H2]
1+ [G2G3H2 + G4H2] + G1G2H1 / 1+ G2G3H2 + G4H2
= G1G2G3 + G1G4 / 1+ G2G3H2 + G4H2 + G1G2H1

(vi) Eliminating unity Feedback (IV)

H(S) = 1
G(S) = G1G2G3 + G1G4 / 1+ G2G3H2 + G4H2 + G1G2H1
G1G2G3 + G1G4 / 1+G2G3H2 + G4H2 + G4G2H1

1+ (G1G2G3 + G1G4 / 1+G2G3H2 + G4H2 + G1G2H1) 1

G1G2G3 + G1G4 / 1+ G2G3H2 + G4H2 + G1G2H1
1+ G2G3H2 + G4H2 + G1G2H1 + G1G2G3 + G1G4
/1+G2G3H2+G4H2+G1G2H1
= G1G2G3 + G1G4 / 1+G2G3H2 +G4H2+G1G2H1+G1G2G3+G1G4
C(S) / R(S) = G1G2G3 + G1G4 / 1+G2G3H2 + G4H2 + G1G2H1 + G1G2G3 + G1G4
(16). Using Block diagram reduction rules, convert the block diagram to a simple loop.

(i) shifting the take off point before the block G3 into the after the block

(ii) Two blocks G2 and G3 in cascade and feedback with H2

(iii) Eliminating feedback – II and cascade with G1 (VI)

G(S) = G2G3/ 1+G2G3H2
 H(S) = H1/G3
G(S) / 1+ G(S) H(S) = G2G3 / 1+G2G3H2
1+ (G2G3H2/1+G2G3H2) (H1/G3)
= G2G3/1+G2G3H2 / 1+G2G3H2 + G2H1 / 1+G2G3H2
= G2G3 / 1+G2G3H2 + G2H1
Cascade with G1
= G1G2G3 / 1+G2G3H2 + G2H1

(v) Eliminating the feedback loop – IV and parallel with G4

G(S) = G1G2G3 / 1+G2H1 + G2G3H2
H(S) = H1 / G3
G(S)/1+G(S) H(S) = G1G2G3 / 1+G2H1+G2G3H2
1- (G1G2G3 /1+G2H1+G2G3H1) (H1/G3) LCM
= G1G2G3 / 1+G2H1+G2G3H2
1+G2H1 + G2G3H1 – G1G2H1 / 1+G2H1 + G2G3H1
= G1G2G3 / 1+G2H1+G2G3H1 – G4G2H1
(vi) and Parallel with G4
C(S) / R(S)= G4 + G1G2G3 / 1+G2H1 + G2G3H1 – G1G2H1
(17). Deduce the transfer function of the system shown in the figure using BDR method

(i) Blocks in parallel (I) and that’s cascade with G1 (II)

(ii) Splitting the summing point (III)

(a) Eliminating parallel path IV
= 1 + G4H2
(b) Eliminating feedback loop V
G(S) = G1 (G2+G3)
H(S) = H1 H2
G(S) / 1+G(S) H(S) = G1(G2+G3) / 1+G1(G2+G3) H1 H2

(iii) Two blocks in cascade.

C(S) / R(S) = (1+G4H2) [G1(G2+G3)] / 1+G1(G2+G3) H1 H2
= (1+G4H2) (G4G2 + G1G3) / 1+G1G2H1H2 + G1G3H1H2
C(S)/R(S) = G1G2 + G1G3 + G1G2G4H2 + G1G3G4H1 / 1+G1G2H1H2 + G1G3H1H2
(18). Block diagram of a feedback control system is shown in figure. The input – output
relation is of the form
Where Y(S)= G(S) R(S)
Y(S) = Y1(S) ; R(S) = R1(S)
Y2(S) R2(S)
Find the over all transfer function using BDR method

Case (i) C1/R ratio

 In this case R2=0, consider only one input one output C1
Remove the summing point which adds R2 and need not consider G6, since
G6 is on the open path.

(i) Eliminating the unit feedback (I)

G4 / (1+G4) 1

(ii) Combining the blocks in cascade (II) and splitting the summing point (III)
(iii) Eliminating feedback loop (iv)
G(S) = G1
G(S) = G4G5H1H2 / 1+G4
G(S) / 1-G(S) H(S) = G1 / (1-G1) (G4G5H1H2/1+G4)
= G1(1+G4) / (1+G4) – G1G4G5H1H2

(iv) cascade with G2 = G1G2 (1+G4) / 1+G4 – G1G4G5H1H2

(v) Eliminating the unity feedback

G(S) = G1G2 + G1G2G4 / (1+G4) – G1G4G5H1H2
H(S) = 1
G(S)/1+G(S) H(S) = G1G2 + G1G2G4 / (1+G4) – G1G4G5H1H2
1+ (G1G2+G1G2G4/(1+G4)-G1G4G5H1H2) 1
= G1G2 +G1G2G4 / (1+G4) – G1G4G5H1H2
 (1+G4) – G1G4G5H1H2 + G1G2+G1G2G4 / (1+G4) –
G1G4G5H1H2 = G1G2 + G1G2G4 / (1+G4) – G1G4G5H1H2 +
G1G2+ G1G2G4
(vi) cascade with G3
C1/R1 = G1G2G3 + G1G2G3G4 / 1+G4+G1G2+G1G2G4 – G1G4G5H1H2

(19). Determine the ratio of C/R and C/D and total output for the system whose block
diagram is shown in the figure.

Redraw the above diagram as,

(i) Eliminating the feedback (I) and parallel blocks (II)

(ii) Blocks in cascade

 (iii) in order to obtain C/R ratio. Assume the second input is zero let D=0 (when
D=0, redraw the diagram)

(iv) Blocks in cascade and feedback H1 is eliminated.

G(S) = (G1+G2) G3G4 / 1+ H2G3
H(S) = H1
G(S) / 1+G(S) H(S) = (G1+G2) G5G4 / 1+G3H2
1+ ((G1+G2) G3G4/1+G3H2) H1 LCM
= (G1+G2) G3G4 / 1+G3H2
1+G3H2+G1G3G4H1+G2G3G4H1 / 1+G3H2
Ratio C/R = G1G3G4 + G2G3G4 / 1+G3H2+G1G3G4H1 + G2G3G4H1

(v) To determine C/P ratio consider R = 0

 (vi) redraw the diagram as,

(vii) Eliminating the feedback loop

G(S) = G4
H(S) = H1G3 (G1+G2) / 1+G3H2
C/D ratio G(S) / 1+G(S) H(S) = G4 / 1+G4 (G3H1 (G1+G2)/1+G3H2)
= G4 / 1 + G3H2 + G1G3G4H1 + G2G3G4H1 / 1+G3H2
C/D = G4 (1+G3H2) / 1 + G3H2 + G1G3G4H1 + G2G3G4H1
The total output C=C1+C2 (R=0)
C1 = (Tr) R
C2 = (Tr. Fn) D
The total output C = (G1G3G4 + G2G3G4 / 1 + G3H2 + G1G3G4H1 + G2G3G4H1) R
+
(G4+G3G4H2 / 1 + G3H2 + G1G3G4H1 + G2G3G4H1) D
(20). Find C/R ratio using BDR method
(i) Block in cascade (I) and Eliminating feedback II

(iii) Shifting the take off (H1) point before the block into after the block

(iii) Eliminating the summing point

= (H1 + 1/G2+G3)

C/R = G1G2 + G1G3 / 1+G1G2 + G1 +G1G2H1 + G1G3H1

 C/R = G1G2G3G4 / 1-G3G4H1 +G2G3H2 + G1G2G3G4H3
(22).find due to input R and disturbance D

C = R1G1G2G3 + DG3 (1+G1) / (1+G3H1) [(1+G1) G1G2G3H2]

1.7. SIGNAL FLOW GRAPH (SFG)
An alternate method for determining the relationship between system variables
called signal flow graph method, is based on a representation of the system by line
segments.
A signal flow graph is an other graphical representation of the control system.
Signal Flow Graph is one which variable are represented by the nodes of the
graph, while relationship between variables represented by the branches connecting
nodes.
1.7.1. Advantages of BDR
(i) Easy to obtain over all transfer function for whole system by simple connecting
the blocks
(ii) The function operation of the system is easily visualised by examing the block
diagram.
1.7.2. Limitations of BDR method:-
(i) Interaction of blocks:- any changes outside of the block not affected rather than
input.
(ii) Concealment of important function:- Important function (can be) being
omitted / hidden with in the walls of the blocks
(iii) Reciprocity and Non-reciprocity:-
π Network we can interchange the input and output terminal vice-
versa. The performance of the network not get altered. Known as RECIPROCITY.
For the amplifiers circuits we can’t interchange input and output terminals.
It obeys non-reciprocity
(iv) Dynamic behaviour characteristics of the system only we can get the
information no information about the construction whether electrical, mechanical or
hydraulic. All the models represented by same block.
(v) No unique from analysis point of view. Different blocks can be drawn for the
same system.
(vi) Main source of energy input not shown in block diagram representation.

Advantages of SFT Method:-

(i) Over all gain can be easily found out than that of BDR method both required
same data.
1.7.3. Basic definitions:-
(i) Node:- A node is a point representing a variable or a signal

(ii) Branch:- A signal travels along a branch from one node to another node in the
direction indicated by the branch arrow and in this process the signal gets multiplied by
the “gain” or “transmittance” of the branch
 (iii) Gain or Transmittance:- The gain acquired get by the signal when it travels
from one node to another is call transmittance.
(iv) Input node or Source:- It is a node that has only outgoing branch
(v) Output node / Sink:- it is a node that has only incoming branch
(vi) Mixed node:- It is a node that has both incoming and outgoing branches
(vii) Path:- A path is a traversal of connected branch in the direction of the branch
arrows. the path should not cross the node more than once.
(viii) Open Path:- A open path starts at a node and ends at another node
(ix) Closed Path:- Closed path starts and ends at the same node.
(x) Forward Path:- It is the path from the input node to the output node.

(xi) Forward path gain:- It is the product of the branch transmittance (gain) of a
forward path
Gain Forward path – I :- G12 G23 G34 G45
Gain Forward path – II :- G12 G23 G35 G54
(xii) Loop (individual):- It is a closed path starting from a node and after passing
through a certain part of a graph arrives at the same node without any node more than
once.
It is a path which originates and terminates at the same node.
Loop gain:- It is the product of the branch gains of a loop.
Non-touching loops:- if the loops does not have a common node then, they are said to
be non-touching loops.

1.7.4. Properties of SFG:-

1. A signal always flows along the branch in the direction of the arrow.
2. When the signal flows through branch, its strength gets multiplied by the
branch gain.
3. Value of the signal at a node is the sum of all the signal entering the node.
 4. Value of the signal at the node is presented on the every branch leaving the
node.
5. Applicable for linear system only.
1.7.5. SFG reduction method:-
SFG can be reduced by using the following methods,
(i) Using rules of SFG algebra
(ii) Mansou’s gain formula
(2nd method can e used because 1st method have time consuming and tedious)
(i). Basic Rules in SFG:- (to solve 2 marks only)
1. Branch gain:-

2. Cascade branch:-

3. Parallel branch:-

4. Mixed node:-
 5. Loop elimination (feedback)

x2 = ax1+cx3 (1)
x3 = bx2 (2)
x3 (1-bc) = ab x1
ouput / input = x3/x1 = a. b/1-bc
(ii). Manson’s Gain formula:-
Let R(S) = Input to the system
C(S) = Output of the system
Transfer function of the system is T(S)
T(S) = C(S) / R(S)
T(S) = 1 / Δ ∑nk=1 Pk Δk Mansou’s gain formula
Where Pk = forward path gain of the Kth forward path
Δ = 1 – [sum of individual loop gain] + [sum of gain products of all possible
combination of two non – touching loops] – [sum of gain products of all
possible combination of three non-touching loops]
Δk = Δ for that part of the graph which is not touching with k th forward path.
(23). Find over-all gain C(S) / R(S) for SFG as shown in figure using Manson’s gain
formula

Problem can be solved using the following steps

(i) Forward path and gain
(ii) Individual loop and gain
(iii) Non touching loop and gain
(iv) ∆ and ∆t – det
(v) Substitute in the formula
(i) forward path and forward path gain
2 forward path k= 2
1st forward path

2nd forward path

I – forward path gain = P1 = G1G2G3G4

II – forward path gain = P2 = G1G2G6
(ii) Individual loop and gain:- Five loop
(ii) Non touching loops (possible – 2

p22 = p51 p41

= -G2G3G6H3
(iv) calculation of Δ and Δk
Δ = 1 – (p11+p21) + (p31+p41+p51) + (p12+p22)
 = 1-[-G2G3H1 – H2G2 – G2G3G4H3 + G5-G2G6H3] + [-G2G5H2 –
G2G5G6H3]
Δ = 1+G2G3H1+G2H2+G2G3G4H3-G5+G2G6H3-G2G5H2-G2G5G6H3
Δ1 = 1+G2G3H1+G2H2+G2G3G4H3-G5+G2G6H3-G2G5H2-G2G5G6H3
(all loop are touch with 1st forward path, there is no non touching loops
with 1st forward path)
Δ1 = 1
Δ2 = 1+G2G3H1 + G2H2 + G2G3G4H3 – G5+G2G6H3 – G2G5H2 – G2G5G6H3
- G2G5H2 – G2G5G6H3
Δ2 = 1-G5
(v) Transfer function = 1/ Δ ∑2k=1 Pk Δk
= P1 Δ1 + P2 Δ2
= G1G2G3G4(1) + G1G2G6(1-G5) / 1+G2G3H1+G2H2 + G2G3G4H3 – G5+G2G6H3
– G2G5H2 – G2G5G6H3

(24). Obtain over all transfer function C/R from the signal flow graph shown in figure.

(i) Forward path H2 and Forward path gain (6 – Forward path ie K = 6)

1st forward path:-
Forward path -2:-

Forward path -3:-

3rd forward path gain = P31 = G!G@G7

4th forward path:-

4th forward path gain = P41 = G3G6G8

5th forward path:-
6th forward path:-

(ii) Individual loops and loop gain

LG-1 = P12 = -G4H1

LG-2 = P22 = -G5H2
LG-3 = P32 = G1G8H1H2
(iii) Non – touching loops and their gain (Minimum two)
 P12 = P11 P21 = (-G4 H1) (-G5 H2)
= G5G5H1H2
(iv) determination of Δ and Δk
Δ = 1 – [P11+P21+P31] + P12
Δ = 1 – [-G4H1 –G5H2 + G1G8H1H2] + G4G5H1H2
Δ = 1+ G4H1+G5H2 – G1G8H1H2+G4G5H1H2
Δ1 = 1+G4H1 + G5H2 – G1G8H1H2 + G4G5H1H2
Δ1 = 1+G5H2
Δ2 = 1+G4H1+G5H2 – G1G8H1H2 + G4G5H1H2
Δ2 = 1+G4H1
Δ3 = 1+G4H1 +G5H2 – G1G8H1H2 + G4G5H1H2
(All loops are touch with the 3rd path)
Δ3 = 1
Δ4 = 1
Δ5 = 1
Δ6 = 1
(v) Mansou’s gain formula
T(S) = 1/ Δ [P1Δ1 + P2Δ2 + P3Δ3 + P4Δ4 + P5Δ5 + P6Δ6]
C(S)/R(S) = G1G4G6 [1+G5H2] + G3G5G7 [1+G4H1] + G1G2G4 (1) + G3G6G8 (1)
–G3G3G7G8H1 (1) – G1G2G6G7H2
1 + G4H1 + G5H2 – G1G8H1H2 + G4 G5H1H2
(25). Find C/R Ratio

(i) forward path --- 5

P1 = G1G2
P2 = G4
P3 = G7G8
P4 = G1G5G8
P5 = G6G7G2
(ii) Loops:- L1 = G9, L2 = G3, L3 = G5G6
(iii) N.T. Loops = G3 G9
(iv) Δ = 1-G3 G9
Δ1 = 1-G9
Δ2 = 1-G3-G9-G5G6+G3G9
Δ3 = 1-G3
Δ4 = 1
Δ5 = 1
1.8. SFG from Block diagram
The following point should follow while converting block diagram into SFG
(i) Give the separate node for input and output
(ii) Also give (identify) node for each summing and take off point

(26).Using Manson’s gain formula obtain the over-all transfer function of the Block
diagram shown in the figure.
(i) Mark (give) nodes for the above block diagram

(i) No. of forward path – 1 and forward path gain (2 – forward path ; K =2)
I – forward path gain = P11 = G1G2G3
II – forward path gain = P21 = G1G2G4
(ii) Individual loop and loop gain

(iii) Determine of Non touching loops:-

All the loops are connected with the 2 forward path
(iv) Determine of Δ and Δk
 Δ = 1 – [P1+P2+P3+P4+P5]
= 1 + G1G2GH1 + G2G3H2 + G4G2H2 + G1G2G3 + G1G2G4
Δ1= 1 + G1G2H1 + G2G3H2 + G4G2H2 + G1G2G3 + G1G2G4
Δ1 = 1 and Δ2 = 1
Transfer function
C(S)/R(S) = T(S) = G1G2G3 + G1G2G4 / 1 + G1G2H1 + G2G3H2 + G2G4H2 + G1G2G3

+G1G2G4
(27). Find C/R ratio using Manson’s gain formula

Forward path – 2 GFP1 = G1G2G3 (P1)

FPG2 = G4 (P2)
Four loops’
LG1 = L1 = -G1H1
L2 = -G2H2
L3 = -G3H3
L4 = -G4H1H2H3
NT Loop (L1 & L3 --- Not connected)
Δ1 = 1
Δ2 = 1+G2H2
T(Δ) = G1G2G3 + G4 + G4H2G2 / 1+G1G1 + G2H2 + G3H3 + G4H3H2H1 + G1G3H1H3
1.9. SFG for Electrical circuit:-
Produce for converting Electrical circuit into SFG
(i) Consider circuit through series circuit
(ii) Consider voltage across the shunt element.

Voltage input and voltage output are the input and output voltages respectively
R1 --- series element circuit
= I1(S) = VI/P – V1
= I1(S) = VI/P / R1 – V1/R1 (1)
= C1 – Shunt elements, voltage across C1
V1(S) = 1/SC1 [ I1(S) – R(S)]
= I1(S)/SC1 – I2(S)/SC2 (2)
= R3 – Series element ct flowing through
R3 IS I2(S)
I2(S) = V1(S)-V2(S) / R2 = V1(S)/R2- V2(S)/R2 (3)
= C2 – Shunt element, voltage across the shunt element is V2(S)
V2(S) = 1/SC2 I2(S) (4)
(Give node names for each shunt/series input as well as output source)
(28).Construct SFG for the given SFG Electrical circuit

(1) In loop L1 there are two elements (R1+L1) series and C1 – shunt
Ct flowing through series element = I1(S) = VI-V1/R1+SL1
I1(S) = VI(S)/R1+L1S – V1(S)/R1+SL1 (1)
Voltage across the shunt element – C1
V1(S) = 1/SC1 [I1(S) – I2(S)]
= 1/SC1 I1(S) – I2(S)/SC2 (2)
(2) In Loop – 2 here have shunt (C2) and series element (R2&L2)
Ct flowing through the series element = V1(S) – V2(S) / R2+SL2
I2(S) = V2(S)/R2+SL2 – V2(S)/R2+SL2 (3)
Voltage across the shunt element = V2(S)
V2(S) = 1/SC2 [I2(S) – I3(S)]
= I2(S)/SC2 – I3(S)/SC2 (4)
(3) In third loop, have R3 is a series element and shunt element is C3
Ct flowing through R3 element is =I3(S) = 1/R3 [V2(S) – V3(S)]
I3(S) = V2(S)/R3 – V3(S)/R3 (5)
Voltage across the shunt element = V3(S) = I3(S)/SC3 (6)
V3(S) = V0(S)
 Additional Problems
(29) Using Manson’s gain formula find over all transfer function

T(S) = (G1G2G3G4G5) (1) + G1G5G6(1+G2H1) / (1+G2H1 + G2G3H2 + G5H3

+ G2G5H1H3 + G2G3G5H2H3)
(30) Using Manson’s gain formula find over all transfer function

T(S) = G1G2G3G4 + G1G2G6 – G1G2G5G6 / 1+G2G3H1 + G2H2 + G2G3G4H3

– G5 – G2G5H2 + G2G6H3 – G2G5G6H3
(31) Using Manson’s gain formula find over all transfer function C/R ratio

(i) Forward path and its gain [K = 2]

Forward path – I
 Forward path – II

(ii) Individual loop and its gain:-

(iii) Non touching loops (Minimum two loop)

 (iv) Non-touching Loops Minimum three loops

(v) Determine Δ and Δk

Δ = 1 – [sum of individual loop gain] + [sum of product of non-touching loop gain
(two possible combination] – [ sum of product of possible combination of
three non – touching loop
gain]
Δ = 1 – [L1+L2+L3+L4] + [P2N1+P2N2+P2N3] – [P3N1+P3N2]
 = 1 - [H1+G3H2 + G4H3 + H4] + [G2H1H2 + G4H1H3 + H1H4]
Δ1 = Δ2 = 1 (All loops are connected with all the nodes in the forward path)

(vi) Gain formula T(S) = 1/ Δ ∑2K=1 Pk Δk = P1Δ1 + P2Δ2 / Δ

= [G1G2G3G4G5 1] + [G1G2G3G46 1] / [1 – G3H2 – G4H3 – H1 – H4
+ G3H1H2 + G4H1H3 + H1H4 – G1H1H2H4 – G4H1H3H4]
(32). Find C/R using Block diagram reduction technique.

(i) Shifting the take off point after the block 100/s 2

(ii) Block in cascade 20(100) /s 2(s+5) feedback with 5/s

G(S) = 2000 / S3+5S
H(S) = 5/S
G(S) / 1+G(S) H(S) = 2000/S2(S+5) / 1+(2000/S2(S+5))(5/S) LCM
= 2000/S2(S+5) / S3(S+5) + 10000 / S[S 2(S+5)]
= 2000 (S) / S4+5S3+10000
 (iii) Two Blocks in cascade
= 20000S / S4+5S3+10000 is feedback by S/50
= G(S) = 20000 S / S4+5S3 +10000
H(S) = S/50
G(S) / 1+G(S)H(S) = 20000S / S4+5S3+10000 / 1+ (2000S/S4+5S3+10000) (S/50)
= 20000 S / S4+5S3+10000
50 [S4+5S3+10000] + 20000 S2 / 50 [S4+5S3+10000]
= 20000 S 50 (S4+5S3+10000)
S4+5S3+10000 50(S4+5S3+10000) + 20000 S2
= 1000000 S / 50S4+250S3+500000+20000S 2
= 1000000 S / 50S4+S50S3+20000S2+500000
(33). Write the differential equations governing the mechanical system shown in
figure.
Solution: the given system has three masses, therefore equations describing the
system are three.
Free body diagram for mass M1 is shown in figure.
Using Newton’s second law
 F(t) = Fm1+Fk1+Fb1+Fk2
= M1 d2x1/dt + B1 dx1/dt + K2(x1-x2) + K1x1
Free body diagram for Mass – M2 is shown in figure

M2 d2x2/dt2 + B2 (dx2-dt / dx3/dt) + K2(x2-x1) + K3 (x2-x3) = 0

Using Newton’s law
M2 d2x3/dt2 + B2 (dx3/dt – dx2/dt) + K3 (x3-x2) = 0
M1 d2x1/dt2 + B1 dx1/dt + K2(x1-x2) + K1x1 = f(t)
M2 d2x2/dt2 + K2(x2-x1) + K3(x2-x3) + B2(dzx2/dt – dx3/dt)
M3 d2x3/dt2 + B2 (dx3/dt – dx2/dt) + K3 (x3-x2) = 0
(34).Obtain the mechanical model of the system shown in figure

Free body diagram of M1:-

M1 d2x1/dt2 + B1 dx1/dt + K1 (x1-x2) = 0 (1)

Free body diagram of M2 :-
 M2 d2x2/dt2 + B2 (dx2/dt – dx3/dt) + K1 (x2-x3) + K3x3 = f(t) (2)
Free body diagram of M3 :-

M3 d2x3/dt2 + B2 (dx3/dt – dx2/dt) + K3x3 = 0 (3)

(35). For the spring, damper and mass system shown in figure, find the differential
equations governing the system
The forces acting on Mass M1 – shown in figure.

M1 d2x1/dt2 + B d/dt (x1-x2) + K1 (x2-x1) = 0 (1)

Forces acting on Mass M2 are shown in figure. The force balance equation is
 M2 d2x2/dt2 + K2x2 + B2 dx2/dt + K1 (x2-x1) + B1 d/dt (x2-x1) = f(t) (2)
(36).Draw signal flow graph for the block diagram in figure. Determine C/R ratio

Signal flow graph for the given block diagram is obtained

1. There is only one forward path gain = P1 = -1

2. Individual loop and loop gain
 3. Non touching loop if two possible combination is zero
4. Δ and Δk :- Determine
Δ = 1 – [sum of individual loop gain]
= 1 – [P11+P21+P31]
= 1 + G5G6G8 + G2G4G5G7G8 + G1G2G3G4G7G8
ΔK = The value of for the part of graph which is not touching the K th forward path
Δ1 = 1+ G5G6G8 + G2G4G5G7G8
5. Transfer function C/R ration
C/R = ∑Pk Δk / Δ
= P1 Δ1 / Δ
= - [1+G5G6G8 + G2G4G5G7G8] / 1+G5G6G8 + G2G4G5G7G8 +
G1G2G3G7G8
(37). for the mechanical system shown in figure draw the force voltage and force
current analogous circuits.

F(t) – e(t)
M1 – L1 B2 – R2
M2 –L2 K1 – 1/C1
 B1 – R1 K2 – 1/C2

F(t) --- i(t)

M1 --- C1 B2 --- 1/R1 Node-1 --- V1
M2 --- C2 K1 --- 1/L1 Node -2 --- V2
B1 --- 1/R1 K2 --- 1/L2

(38). Write the differential equations governing the mechanical system shown in
figure. Draw the force voltage and force current analogous circuits.
M1 dv1/dt + K1 ʃ v1 dt + K2 ʃv1-v2 dt + K3 ʃv1-v2 dt = f(t) (1)
K3 ʃv1-v2 dt + B (v2-v3) = 0 (2)
M2 dv3/dt + B(v3-v2) + K2 ʃv3-v1 dt = 0 (3)
f(t) --- e(t)
v1 --- i1 M1 --- L1 K1 --- 1/C1
v2 --- i2 M2 --- L2 K2 --- 1/C2
v3 --- i3 B ---- R K3 --- 1/C3
f(t) --- i(t)
v1 --- v1 K1 --- 1/L1 M1 --- C1
v2 --- v2 K2 --- 1/L2 M2 --- C2
v3 --- v3 K3 --- 1/L3 B ----- 1/R
 UNIT - II

TIME RESPONSE

Time response, time domain specification – types of test input, I & IInd order system
response, Error coefficients – Generalised error series – steady error – P, PI, PID (controller)
modes of feedback control.

2.1 Time response:-

The time response of a system is the output of the system as a function of time,
when subjected to a known input.

Transient and steady state response:-

The time response of a system of a system consists of two parts

(i) Transient response

(ii) Steady state response and can be represented (as) by

Total time response C(t) = Transient response ctr(t) + steady state response Css(t)

2.1.1. Transient response:-

Transfer function of the system is the portion of the total time response during
which the output changes from one state to another state, In other wards, it it the response
before the output reaching steady state value.

2.1.2. Steady State response:-

Steady state response of the system is the response of the system for a given input
after very long time. In steady state, the output response settles to its final steady state value
or steady oscillations. Figure shows the regions of transient and steady state response for
sluggish and for a process that exhibits oscillatory behaviour
2.2 TEST SIGNALS:-

IMPORTANCE:- test signals can be easily generated in test laboratories and the
characteristics of test signals resembles, the characteristics of actual input signals. The test
signals are used to pre determine the system. If the response of a system satisfactory for a
test signal, then the system will be suitable for practical applications.

Commonly used test input signals in control systems are impulse, step, ramp,
acceleration and sinusoidal signal.

2.2.1. Step signal:-

The step signal is a signal whose value changes from 0 to A and remains constant at A
for t>0 the mathematical representation of step signal is

r(t)

where r(t) = 1, t>0

u(t) = 0 ; t<0
when A = 1 ; r(t), u(t)
unit step signal
laplace transform of A(S) = R(S) = 1/S
2.2.2. Ramp signal :- ramp signal is a signal whose value increases linearly with time form an
initial value of zero at t=0. The ramp signal resembles a constant velocity input to the system.

A special case of ramp signal is unit ramp signal in which the value of ‘A’ is unity

r(t) = At ; t≥0
= 0 ; t ≤0
Laplace transform

r(t) = At1 (A=1)

=t
R(S) = 1/S2
2.2.3. Parabolic Signal :- In parabolic signal the instantaneous a parabola. The parabolic
signal resembles also stunt acceleration input to the system. A special case of parabolic signal
is unit parabolic signal in which “A” is unity.

r(t) = At2/2 ; t≥0

= 0 ; t≤ 0
Take laplace R(S) = 1/S3 (if A=1)

[ integral of step is ramp signal. Integral of ramp is parabolic signal]

2.2.4. Impulse signal:- A signal which is available for very short duration is called “Impulse
signal”. Ideal impulse signal is a unit impulse signal which is defined as a signal having zero
values at all times except at t=0. At t=0, the magnitude becomes infinite. It is denoted by S(t)
and mathematically it is expressed as

S(t) = 0 for t≠ 0

And Lt –t1ʃ+t1 8(t) dt = 1

Mathematically an impulse function is the derivatives of a step function. Laplace

transform of the impulse function is unity.

The response of the system, with input impulse signal is called weighting function of
the system. It is given by the inverse Laplace transform of the system transform function.

(it) = L-1 [ R(S) (G(S)/1+G(S)H(S)]

= L-1 [ G(S) / 1+G(S) H(S)] R(S) = 1 for impulse

2.3. Table of transform The table lists Laplace transforms of commonly encountered function
f(t), t≥0

F(S) F(t); t≥0

1 Unit impulse S(t)
1/S Unit step u(t)
1/S2 T
1/(s+a) e-at
1/(s+a)2 Te-at
a/s(s+a) 1-e-at
a/s2+a2 Sin at
s/s2+a2 Cos at
b/ (s+a)2 + b2 -at
e sin bt
(s+a) / (s+a)2+b2 e-at cos bt
 ab/s(s+a) (s+b) 1= (b/a-b) e-at- a/a-b e-bt
b-a / (s+a) (s+b) e-at-e-bt
a2/s2 (s+a) at-1+e-at
a(s+b) / s2(s+a) (a-b)/a + bt – ((a-b)/a) e-at

2.4. ORDER OF A SYSTEM

Input and output relationship of a control system can be expressed by a differential

equation. The order of the system is given by the order of the differential equation governing
the system, if the system is governed by n th order differential equation, then the system is
called nth order system. Alternatively the order can be determinate from the transfer function
of the system.

Transfer function = T(S) = K P(S) / Q(S)

Where, P(S) – Numerator polynomial

Q(S) – Denominator polynomial
K – Constant
Order of the system is given by the maximum power of “S” in the denominator
polynomial Q(S)

If Q(S) = a0sh +a1sh-1+a2sh-2 ………… + an-1s+an

Then “n” is the order of the system

If n=0 then the system is zero order system
n=1 then the system is first order system
n=1 then the system is second order system

The order can be specified for both open loop and closed loop system. Also the
value of “n” gives the number of poles in the transfer function. Hence order is given by the
number of poles of the transfer function.

2.4.1. Poles and zero form

The transfer function G(S) can also be expressed in the following form
K (S−Z1) (S−Z2) (S−Z3) ……..(S−ZM)
G(S) =
(S−P1) (S−P2) ………… (S−PM)

Where Z1,Z2,Z3…………ZM are roots of the numerator known as zero’s of the system and
P1,P2,P3 …… PM are root of the denominator known as poles of the system. The denominator
polynomial of the transfer function known as characteristic equation. Note that the roots of
the characteristics equation is poles of the transfer function G(S).

Poles and zeros may be real or complex

Conjugate poles occur in pairs

Complex planes is used to represent poles and zeros

Location of poles and zeros decides the stability of the system.

2.4.2. S – Plane

The values in G(S) is a complex variable defined by

S = σ + j Ѡ ---- (1)

Where σ - real part of S

Ѡ – Imaginary part of S

The set of all complex coordinates (σ, Ѡ) forms a plane called s-plane. The horizontal
axis of this plane is known as σ “Real axis”. Since it is represented by the line

S = σ + jo ----- (2)

The vertical axis is known as the “imaginary axis” and is represented by the line.

S = 0 + jw ---- 3

The origin is represented by the complex zero S=o+jѠ. σ

The jѠ axis (imaginary axis) divides the S-plane into two half planes. The region w to
the left of the imaginary jw-axis (for co<w<0) is termed the left half of S-plane (LHP) and the
region to the right of imaginary axis (for 0<w<w) is termed the right half of S-plane (RHP) as
shown in figure.
 The poles and zeros of the transfer function G(S) can be represented schematically in
a S-plane. The symbol “X” is used for a pole and “0” designates a zero.

2.5 Response of first order system with Unit step input

G(S) = K / TS+1 ------ first order system general equation

Where T & K are physical constant K=1

G(S) = C(S)/R(S) = output/input = 1/1+TS

C(C) = R(S) 1/(1+TS) [R(S) = unit step input]

Partial fraction,
C(S) = A/S + B/(ST+1) = 1/S(1+TS)
1/S(1+TS) = C(S) = (ST+1)A + BS/ (ST+1)S
1 = (TS+1)A + BS
Sub S = 0;
1=A
Sub S = -1/T
1 = [(-1)+1] + B(-1/T)
= -B/T
-T = B
C(t) = LC-1(S) = A/S + B /TS+1
= 1/S – T/TS+1
= 1/S – T/T(S+1/T)
= l-1 [1/S – 1/S+1/t]
= 1-e –t/T

t=0 ; c(t) = 1-e0 = 1-1 = 0

t=T ; c(t) = 1-e-1 = 0.632
t = 2T ; c(t) = 1-e-2 = 0.865
t = 00 = 1
2.6. Response of SECOND ORDER SYSTEM:-

Open loop system K / S(ST+1) ; unity feedback

Closed loop transfer function G(S) = Kῳn2 + S2+2ԐwinS + ῳn2

ῳn – undamped natural frequency

K=1

ῳd – damping frequency

Ԑ --- damping ratio

It is the ratio of the actual damping to critical damping

C(t) --- is depends on Ԑ

Characteristic equation = S2+2ԐῳnsS+ῳn2

Roots of the equation (quadratic)

S1 & S2 = -Ԑῳn±ῳn√Ԑ2-1

When Ԑ = 0 ; S1 & S2 = ±jῳn

Roots are purely imaginary and the system is undamped

When Ԑ = 1 ; S1 & S2 = -ῳῳn

Roots are real and equal and the system is critically damped

When Ԑ>1 ; S1 & S2 = - Ԑῳn±ῳn√(Ԑ2 -1)

Roots are real and unequal and system is over damped

When 0< Ԑ<1 ; S1, S1 & S2 = - Ԑῳn±ῳn√Ԑ2 -1

= - Ԑῳn ± ῳn√-1 [1- Ԑ2]

 = - Ԑῳn±jῳn√1-Ԑ2

= - Ԑῳn±jῳd

ῳd = ῳn√(1-Ԑ2) --- damped frequency of oscillation rad/sec

Roots are complex conjugate and the system is under damped

2.7. UNIT STEP RESPONSE OF SECOND ORDER SYSTEM

r(t) = u(t) (1)

Whose Laplace transform

R(S) = 1/S
C(S) = (1/S) (ῳn2 / (S2+2ԐῳnS+ῳn2) (2)
2.7.1. Case -1 :- Ԑ = 0

C(S) = (1/S) (ῳn2 / S2 + ῳn2)

C(S) = ῳn2 /S(S2+ ῳn2) =A/S + B/S2+ ῳn2)
ῳn2 /S (S2 + ῳn2) = A(S2+ ῳn2) +B(S) / S(S2+ ῳn2)nd
ῳn2 = A(S2+ ῳn2) + B(S)
S=0

ῳn2 = A ῳn2 A=φ

S = - ῳn2 (S = √-Jῳn2 = √1 √ ῳn2 = jῳn
S =Jῳ
ῳn2 = A(-ῳn2 + ῳn2) + B (jῳn)
Bjῳn = ῳn2
B = ῳn2 / jῳn
B = ῳn / j = -jῳn
B = -S
C(S) = A/S + B/(S+ jῳn2)
C(S) = 1/S + -S/(S2+ ῳn2) (3)
The time domain response :-

C(t) = L-1C(S)
= 1/S - cosῳnt (4)
2.7.2. CASE :-2 critically damped Ԑ=1

If Ԑ=0 equation (2) becomes

C(S) = ῳn2 /S(S+2ῳNs+ ῳn2) (4)
= ῳn2 / S(S+ ῳn)2
C(S) = A/S + B/(S+ ῳn) + C /(S+ ῳn)2
ῳn2 / S(S+ ῳn)2 = A(S+ ῳn)2 + B (S) (S+ ῳn) + C(S) / S(S+ ῳn)2
ῳn2 = A (S+ ῳn)2 + B(S) (S+ ῳn) + (S) (5)
sub S = 0 in equation (5)
S = - ῳn in equation (5)
ῳn2 = C (-ῳn) C = -ῳn
ῳn2 = A[S2+ ῳn2+2ῳnS] + B [S2+ ῳnS] + CS
ῳn2 = (A+B)S2 + B ῳnS + A ῳnS + CS
equating the S2 terms on bothe sides
A+B = 0 A = -B
Equating constant terms
ῳn2 = A ῳn2 A=1
S = - ῳn in equation (5)
ῳn2 = C(- ῳn) (C=- ῳn)
ῳn2 = A [S2+ ῳn2+2 ῳnS] + B [S2+ ῳNs] + CS
ῳn2 = (A+B)S2 + B ῳnS + A2 ῳnS + CS + Aῳn2
equating the S2 terms on both sides
A+B=0 A = -B
Equating constant terms
ῳn2 = A ῳn2 AA=1
=1
B = -1
Equating “S” term on both sides of the above equation
0 = A2 ῳn + B ῳn + C
0 = A12 ῳn + B-1 + C
-ῳn = C
Substituting A, B & C values in the equation
C(S) = A/S + B/S+ ῳn + C /(S+ ῳn)2
 = 1/S – 1/(S+ ῳn) - ῳn / (S+ ῳn)2
Taking inverse laplace

C(t) = 1 – e-ῳnt - ῳnte-ῳnt

1 – e-ῳnt [1+ῳnt]
2.7.3.. CASE :- III : - UNDER DAMPED SYSTEM:-

[1 < Ԑ < 1]

C(S) = (1/S) (ῳn2 / S2+2ԐῳNs+ ῳn2)

C(S) = ῳn2 / S(S2+2ԐῳNs+ ῳn2) = A/S + (BS+C) / (S2+2ԐῳNs+ ῳn2)

= ῳn2 / S(S2+2ԐῳNs+ ῳn2) = A (S2+2ԐῳNs+ ῳn2) + (BS+C) S / S(S2+2ԐῳNs+ ῳn2)

ῳn2 = A(S2+2ԐῳNs+ ῳn2) + (BS2+CS)

Equating the co efficient

→ constant

ῳn2 = A ῳn2 A=1

→ equating the coefficient of ‘ S ‘ term

A+B=0
A = -B B = -1
By equating the ‘S’ term

2ԐῳnA + C = 0
C = -2Ԑῳn
C(S) = 1/S + (-S-2Ԑῳn) / (S2+2ԐῳnS+ῳn2)
= 1/S – (S+2Ԑῳn) / (S2+2ԐῳnS+ῳn2)
Add & Subtract Ԑ2ῳn2 into 2nd part of the denominator of the equation

= 1/S – (S+2Ԑῳn) / (S2+2ԐῳnS+ῳn2+Ԑ2ῳn2) - Ԑ2ῳn2

= 1/S - (S+2Ԑῳn) / (S2+2ԐῳnS+Ԑ2ῳn2) + ῳn2 + Ԑ2ῳn2
= 1/S - (S+2Ԑῳn) / (S+2Ԑῳn)2 + ῳn2(1- Ԑ2)
wd = ῳn√(1- Ԑ2) & wd2 = ῳn2√(1- Ԑ2)

= 1/S – (S+2Ԑῳn) / (S+2Ԑῳn)2 + ῳd2

= 1/S – (S+Ԑῳn) / ((S+Ԑῳn)2 + ῳd2) + Ԑῳn/((S+Ԑῳn)2 + ῳd2)
In the 3rd part equati;on is X & / by the term ῳd
= 1/S – (S+Ԑῳn) / ((S+Ԑῳn)2 + ῳd2) + Ԑῳn ῳd / ῳd[(S+ Ԑῳn)2+ ῳd2]
 =1/S – (S+Ԑῳn) / ((S+Ԑῳn)2 + ῳd2) + (Ԑῳn/ῳd) (ῳd(S+ Ԑῳn)2+ ῳd2]
Take inverse laplace
C(t) = 1-e-Ԑῳnt cos ῳdt - Ԑῳn/ῳd e-Ԑῳnt
= 1-e-Ԑῳnt [cos ῳdt - Ԑῳn/ῳn√1-Ԑ2 sin ῳdt]
= 1-e-Ԑῳnt / √1- Ԑ2 [√1- Ԑ2cos ῳdt + sin ῳdt ]
= 1-e-Ԑῳnt / √1- Ԑ2 [√1- Ԑ2cos ῳdt + sin ῳdt Ԑ]
= 1-e-Ԑῳnt / √1- Ԑ2 [sin ῳdt cos Θ + cos ῳdt]
= 1-e-Ԑῳnt / √1- Ԑ2 [sin (ῳdt+Θ) + cos ῳdt] (sin (A+B) = sinA cosB + CosA sinΘ)
C(t) = 1-e-Ԑῳnt / √1- Ԑ2 sin(ῳdt+Θ)

Over Damped (Ԑ>1) case :- IV

C(S) = 1/S ῳn2 / S2 + 2ԐῳnS + ῳn2
When Ԑ>1 ; roots are real & different roots are obtained
S1 = -Ԑῳn + ῳn√Ԑ2-1
S2 = -Ԑῳn - ῳn√Ԑ2-1
We have
C(S) = ῳn2 / S(S+Ԑῳn-ῳԐ√Ԑ2-1) (-Ԑῳn + ῳn√Ԑ2-1
A/S + B/S+Ԑῳn-ῳn√Ԑ2-1 + C/S+Ԑῳn + ῳn√Ԑ2-1
= A(S+Ԑῳn-ῳn√Ԑ2-1) (S+Ԑῳn+ῳn√Ԑ2-1) + B(S) (S+Ԑῳn+ῳn√Ԑ2-1) + C(S) (S+Ԑῳn- ῳn√Ԑ2-1
Dr
ῳn2= A(S+Ԑῳn-ῳn√Ԑ2-1) (S+Ԑῳn + ῳn√Ԑ2-1) + B(S) (S+Ԑῳn + ῳn√Ԑ2-1) + C(S) (S+Ԑῳn+ῳn√Ԑ2-1)
S=0
ῳn2 = (Ԑῳn - ῳn√Ԑ2-1)2A
= Ԑ2ῳn2±ῳn2(Ԑ2-1)
ῳn2 = (Ԑ2ῳn2±ῳn2Ԑ2+ῳn2)A
A = ῳn2/ῳn2 = 1 A=1
B = -Ԑῳn+ῳn√Ԑ2-1
B = -1 / 2√Ԑ2-1 [Ԑ-(√Ԑ2-1)]
S = -Ԑῳn - ῳn √Ԑ2-1
C = 1 / 2√Ԑ2-1 (Ԑ+√Ԑ2-1)
 Substituting the values of A,B & C & taking inverse laplace,
C(t) = 1 – 1/2√Ԑ2-1 [Ԑ-√Ԑ2-1 e(Ԑῳn+ῳn√Ԑ2-1)t + 1/2√Ԑ2-1 [Ԑ-√Ԑ2-1 e(Ԑῳn+ῳn√Ԑ2-1)t
= 1 – e-Ԑῳnt/2√Ԑ2-1 [e(ῳn√Ԑ2-1)t / Ԑ-√Ԑ2-1] – e (-ῳn√Ԑ2-1)t / Ԑ+√(Ԑ2-1)

IMPULSE RESPONSE of second order system:-

C(S) / R(S) = ῳn2 / S2+2ԐῳnS+ῳn2 (1)

For a unit impulse input r(t), the corresponding laplace transform is unity ie R(S) = 1, then the
unit impulse response

C(S) = ῳn2 / S2+2ԐῳnS+ῳn2 (2)

Roots of the equation (s2+2ԐῳnS+ῳn2) = 0
S1&S2 = -Ԑῳn±(√Ԑ2-1)ῳn
Roots S1&S2 will be real or complex depending on the parameter Ԑ

If Ԑ>1 ; roots are real and distinct over damped

If Ԑ=1 ; roots real and equal critically damped

If Ԑ<1; roots S1 & S2 are complex under damped

Now we find the response C(t) for different values of Ԑ

CASE -1 Impulse response for Ԑ>1

S1 = -Ԑῳn + (√Ԑ2-1) ῳn
S2 = -Ԑῳn + (√Ԑ2-1) ῳn
C(S) = A/S-(-Ԑῳn + ῳn(√Ԑ2-1)) + B/ S-(-Ԑῳn - ῳn(√Ԑ2-1))
S = -Ԑῳn + ῳn(√Ԑ2-1)
A = ῳn / 2√Ԑ2-1
Similarly
S = -Ԑῳn - ῳn(√Ԑ2-1)
B = -ῳn / 2√Ԑ2-1
 C(S) =ῳn / 2√Ԑ2-1 [S-(-ῳnԐ-√Ԑ2-1) - ῳn / 2√Ԑ2-1 [S-(-Ԑῳn-√Ԑ2-1)]

Taking inverse laplace on both sides,

We have,

C(t) = ῳn / 2√Ԑ2-1 [ e(-Ԑῳn+ῳn√Ԑ2-1)t + e(-Ԑῳn-ῳn√Ԑ2-1)t

C(t) = ῳn / 2√Ԑ2-1 e-Ԑῳnt 2 sinh(ῳn√Ԑ2-1) t

CASE- II Impulse response for Ԑ<1

For Ԑ<1, the roots S1&S2 are given by

S1 = Ԑῳn+įῳn√1-Ԑ2

S2 = Ԑῳn-įῳn√1-Ԑ2

C(S) = A / S-(-Ԑῳn+įῳn√1-Ԑ2) + B / S-(-Ԑῳn+įῳn√1-Ԑ2)

C(S) = ῳn/2į√(1-Ԑ2) [[1/S-(-Ԑῳn+įῳn√1-Ԑ2)] [1/S-(-Ԑῳn-įῳn√1-Ԑ2)]

Taking inverse laplace;

C(t) = ῳn/2į√(1-Ԑ2) [e (-Ԑῳn+įῳn√(1-Ԑ2)t e (-Ԑῳn-įῳn√(1-Ԑ2)t]

= ῳn/√(1-Ԑ2) e-Ԑῳnt [eįῳn√1-Ԑ2 t – e-įῳn√1-Ԑ2 t / 2į ]

C(t) = ῳn / √(1-Ԑ2) e-Ԑῳnt sin(ῳn √1-Ԑ2 t

CASE:- III Impulse response for Ԑ=1

C(S) = ῳn2 / (S+ῳn)2 when Ԑ=1

Taking inverse laplace

C(t) = ῳn2t e-Ԑῳnt

CASE IV :- Impulse response for Ԑ=0

C(S) = ῳn2 / S2+ῳn2 when Ԑ=0

Take inverse laplace

C(t) = ῳn sin(ῳnt)

TIME DOMAIN SPECIFICATIONS

The following are the time domain specifications

 1. Rise time:- The rise time is the time required for the system to reach 100% of its
final value first time from the initial value.

2. Peak time:- Peak time is the time required for the response to reach peak of the first
over shoot.

3. Maximum Over/Peak Over Shoot:- Maximum over shoot is a measure of how much
the response exceeds

The final value following a step change. The percentage of over shoot canbe calculated
as

Mp = Peak value – final value / final value * 100

= C(tp) – C(ῳ) /C(ῳ) * 100

4. Setting time (ts). The setting time is the time required for the response to reach and
stay with in a specified (error) tolerance band of its final value. The tolerance band is usually
2% to 5%.

5. Decay Ratio:- The decay ratio is defined as the ratio of the amplitude of first two
successive peaks.

Evaluation of time domain specification

Rise Time :- consider a step response of a 2nd order system (under damped)

C(t) = 1-e-Ԑῳnt / √1-Ԑ2 sin(ῳdt+Θ) (1)

Where Θ = tan-1√1-Ԑ2/Ԑ

According to the definition of tr

C(tr) = 1

But C(tr) = 1 - e-Ԑῳntr/√1-Ԑ2 sin(ῳdtr + Θ)

e-Ԑῳntr/√1-Ԑ2 sin(ῳdtr + Θ) = 0
sin(ῳdtr + Θ) = 0
which implies that ῳdtr + Θ = nπ
 for first peak n = 1
therefore tr = π – Θ / ῳd
tr = π – Θ / ῳn √1-Ԑ2
Peak time:- (tp)
Peak time tp is found by differentiating C(t) in the equation and taking this derivative
to zero.

d(t)/dt t=tp =0

dC(t)/dt = d/dt [1 - e-Ԑῳnt/√1-Ԑ2 sin(ῳdt+Θ]

uv = v du + u dv

= [sin(ῳdt+Θ) –e-Ԑῳnt/√1-Ԑ2] +[ –e-Ԑῳnt/√1-Ԑ2 cos(ῳdt+Θ) ῳd]

= –e-Ԑῳnt/√1-Ԑ2 Ԑῳn sin(ῳdt+Θ) - ῳn√1-Ԑ2/√1-Ԑ2 e-Ԑῳnt cos(ῳd+Θ)

= ῳn e-Ԑῳnt/√1-Ԑ2 [Ԑ sin(ῳdt+Θ) – 1/√1-Ԑ2 cos(ῳdt+Θ)]
= ῳn e-Ԑῳnt/√1-Ԑ2 [sin(ῳdt+Θ) cosΘ - cos(ῳdt+Θ) sinΘ]
= ῳn/√1-Ԑ2 –e-Ԑῳnt [sin(ῳdt+Θ) – Θ]
C(t) = ῳn/√1-Ԑ2 –e-Ԑῳnt sin ῳdt
At t=tp d/dt C(t) = 0
= ῳn/√1-Ԑ2 –e-Ԑῳntp sinῳdtp =0
= sin ῳdtp =0
ῳdtp = nπ n=1
ῳdtp = π
tp = π/ῳd
tp = π/ῳn√1-Ԑ2
π – in radians
Maximum Peak over shoot (Mp)

The maximum over shoot occurs at t = tp. The final value of the output is unit

Mp = C(tp) – 1

That is Mp = 1 –e-Ԑῳntp/√1-Ԑ2 [sinῳdtp + Θ] – 1

= -e-Ԑῳn π/ῳd [sin(ῳd π/ῳd + Θ]

= -e-Ԑ π/√1-Ԑ2/√1-Ԑ2 [sin π cosΘ + cosπ sinΘ]

= -e-Ԑ π/√1-Ԑ2/√1-Ԑ2 [-sinΘ] √1-Ԑ2

Mp = -e-Ԑ π/√1-Ԑ2
The maximum percentage in over shoot is

Percentage Mp = -e-Ԑ π/√1-Ԑ2 x 100%

Settling time:- (ts) To evaluate the settling time, we must find the time for which C(t) in
equation 1 reaches and stay with in ±2% (for 2% criterions) or ±5% (for 5% criterions) of the
steady state value.

Using definition, the settling time is the time it takes for the amplitude of decaying
sinusoidal to decay 0.02 (for 2% criterion), that it,

e-Ԑῳnts/√1-Ԑ2 = 0.02

e-Ԑῳnts = 0.02√1-Ԑ2

-Ԑῳnts = ln[0.02 √1-Ԑ2]

Ts = -ln[0.02√1-Ԑ2 / Ԑῳn]

The numerator varies from 3.90 to 4.74 as Ԑ varies from 0 to 0.9

An approximitiation for the settling time for all values of Ԑ can be

ts = 4 / Ԑῳn section ln 0.02 = -4

similarly for settling time for 5%

criterion is ts = 3/Ԑῳn sec

second order system (general form)

most general form of differential equation for a second order system may be expressed
as,

1/ῳn2 = d2y(t)/dt2 + 2Ԑ/ῳn dy(t)/dt + y(t) = kx(t) (1)

Where ῳn and K depend on the characterise of the system under consideration x(t)
and y(t) are the input and output respectively, ῳn, K and Ԑ are called natural frequency, gain
and damping ratio of the system respectively.

The transfer function of the system described by equation (1) can be obtained by taking
laplace transform on both sides of the equation with all initial conditions

Set to zero that is y(0) = 0; and y(0) = 0 (2)

That is

[1/ῳn2 d2y(t)/dt2 + 2Ԑ/ῳn dy(t)/dt + y(t)] = Kx(t)

 1/ῳn2 [s2y(s) – sy(0) – y(0)] + 2Ԑ/ῳn [sy(s) – y(0)] + y(s) = Kx(s)

Since y(0) = y(0) = 0

1/ῳn2 s2y(s) + 2Ԑ/ῳn sy(s) + y(s) = Kx(s)

Y(s)/X(s) = Kῳn2/s2+2Ԑῳns+ῳn2 (3)

The above equation it’s the general form of transfer function of a second order system.

Mechanical and an electrical second order systems are shown in figure (a&b) respectively and
their transfer functions are derived as shown below

Applying KVL Applying Newton’s law

Vi(s) = Ri(t) + L di(t)/dt + -ῳʃt 1/c i(t) dt F(t) = M d2x(t)/dt2 + B dx(t)/dt + Kx(t)
Vi(s) # RI(s) + LSi(s) + 1/cs I(s) F(S) = Ms2x(s) + Bs x(s) + Kx(s)
Vi(s)/R+ZS+1/Cs = I(s) X(s)/F(s) = 1 / MS2+BS+K
Vo(t) = 1/C -ῳʃt I(T) DT
Vo(s) = I(s)/CS = [(1/M)/s2+B/M s+K/M]
= Vi(s) / C(s) [R+LS+1/CS] = K1 [(K/M)/S2+B/M S+K/M]
Vo(s)/Vi(s) = 1 / LCS2+RCS+1 = K1 = 1/K
Vo(s)/Vi(s) = [(1/LC)/(S2+R/L S+1/CC)] (2) X(S)/F(S) = K1 (K/M)/S2+B/M S +K/M (3)
Comparing equation 2 with standard second Comparing the equation 3 with standard
equation 1 second order equation 1
ῳn2 = 1/Lc ῳn2 =K/M or ῳn = √K/M
that is ῳn = 1/√LC 2Ԑῳn = B/M
2Ԑῳn = R/L Ԑ = B/2ῳnM
That is Ԑ = R/2ῳnL Ԑ = R/2 √C/L Ԑ = B/2√KM

# The parameters of Mechanical system of figure (a) are

M = 100 kg
 B = 1000 N/m/sec

K = 10000 N/m

A step force of 100N is applied to the Mass at t=0. The initial conditions are y(0) = y(0)
= 0 form the physical parameters of the system, obtain the transfer function model.
Parameters damping ratio Ԑ, undamped natural frequency ῳw, ῳd

Also obtain the step response

2nd order equation

F(t) = M d2y(t)/dt2 + B dy(t)/dt + K y(t) (1)

Apply laplace

F(s) = Ms2y(s) +Bsy(s) + Ky(s)

Y(s)/F(s) = 1 / Ms2+Bs+K

= 1 / 100S2+1000S+10000

= 0.01 / S2+10S+100

ῳn2 = 100 ῳn = 10 r/s

2Ԑῳn = 10 Ԑ = 0.5

ῳd = ῳn√1-Ԑ2

= 8.66 rad/sec

For a step input of 100N

R(S) = 100/S

Y(S) R(S) 0.01/S2+10S+100

 = (100/S) (0.01/ S2+10S+100z

= 0.01 [100 / S(S2+10S+100)]

C(t) = 0.01 L-1 [100/S (S2+10S+100)]

= 0.01 [1-e-Ԑῳnt/√1-Ԑ2 sin(ῳdt+Θ]

Tan-1(ῳn√1-Ԑ2/ῳnԐ) = 0.01 [ 1- 2/√3 e-5t sin(8.66+π/3)] Answer

# A unity Feedback system is characterised by an open loop transfer function G(S) = K/S(S+10).
Determine the gain K. so that the system will have a damping ratio of 0.5. for this value of ‘K’
let settling time, peak over shoot and time to peak over shoot for a unit step input.

G(S) = K/S(S+10) H(S) =1

C(S)/ K(S) = K/S(S+10) / 1+ KS(S+10) = K/S 2+10S+K

ῳn2 = K

2Ԑῳn = 10 Ԑ=0.5

ῳn
ῳn= =1010
K=(10)2=100 r/s

Peak time tp = π/ῳn√1-Ԑ2 = π/10√1-(0.5)2

= 0.363 sec

Peak over shoot Mp = 100 e-Ԑπ√1-Ԑ2

= 16.3%

Settling time = 4/Ԑῳn = 0.8 sec

# A second order Mechanical system is represented by the transfer function

Θ(S) = I(S) = 1/JS2+BS+K ; a step input of 10 N-M is applied to the system and results
are given below.

1. Maximum over shoot = 6%

2. Time at peak over shoot -1 sec

3. steady state value of the output 0.5 rad

Determind the values of J,B & K

Given Mp = e-Ԑπ/√1-Ԑ2 = 0.06

-Ԑπ /√1-Ԑ2 = lu (0.06)

= -2.8134

Ԑ = 0.667

tp = π/ῳn √1-Ԑ2 = 1 sec

tp = π/√1-Ԑ2 = 4.217

steady state value of the output

ess = Ltt→ῳ Θ(t) = Lts→0 S Θ(S)

given Θ(S)/I(S) = 1 / JS2+BS+K

a step input of 10 n/m that is

I(S) = 10/S

Θ(S) = 10/S(JS2+BS+K)

ess = Lts→0 S 10/S(JS2+BS+K)

ess = 0.5 = 10/k

K = 10/0.5 = 20

From the transfer function

ῳn2 = K/J = 20/J = (4.217)2

2Ԑῳn = B/J J=1.124

B = (1.1214) (2) (0.667)

B = 6.323

# obtain the unit impulse response and unit step response of a unity Feedback system whose
open loop transfer function is

G(S) = 2S+1/S2

Closed loop transfer function is

C(S)/R(S) = (2S+1)/S2 / 1+(2S+1)/S2 1 = 2S+1/(S2+2S+1)

= (2S+1)/(S+1)2 (1)
For unit impulse response R(S) = 1

C(S) = (2S+1)/(1+S)2

C(t) = L-1 [(2S+1)/(S+1)2]

= (A/S+1) + (B/(S+1)2

A=2

B=1 S = -1

C(S) = 2/(S+1) – 1/(S+1)2

C(t) = 2e-t –te-t

For unit step response R(S) = 1/S

C(S) = 1/S (2S+1)/(S+1)2

C(S) = A/S + B/(S+1) + C/(S+1)2

(2S+1)/S(S+1)2 (S+1) = A(S) (S+1)2 + B(S) (S+1) + C(S) / S(S+1)2 (S+1)

Equating S2 term

(2S+1) = A(S2+2S+1) + B(S2+S) + CS

A+B = 0

A = -B

By equating S term

2 = 2A+B+C

2 = 2-1+C C = 2-1 C=1

By equating constant term

A=1 B = -1

C(S) = 1/S – 1/S+1 + 1/(S+1)2

C(t) = 1-e-t + t e-t

# Measurements conducted on a servo mechanism shown the system response to be

C(t) = 1+0.2e-60t -1.2e-10t

Obtain an expression for closed loop transfer function. Determine the undamped natural
frequency and damping ratio

C(t) = 1+0.2e-60t -1.2e-10t

C(S) = 1/S + 0.2(S+60) – 1.2 1/S+10

= (S+60) (S+10) + [0.2 S (S+10)] – 1.2 ((S+60) S)] / S(S+60) (S+10)

= S2+60S+10S+600+0.2S2+2S-1.2S2-7.2S / S(S+60) (S+10)

= 1/S 600/(S+60) (S+10)

= 1/S 600/S2+70S+600

(i) C(S)/R(S) = R(S) 600/S2+70S+600 = ῳn2 / S2+2Ԑῳns+ῳn2

ῳn2 = 600 ῳn = 24.49 rad/sec

2Ԑῳn = 700 Ԑ = 1.429

# for a Servo mechanism represented by d2Θ/dt2 + 10 dΘ/dt – 150 C(t) = 0 where e is the
activating signal calculate damping ratio, natural frequency and damped frequency of
oscillation

d2Θ/dt2 + 10 dΘ/dt -150 e(t) = 0

d2Θ/dt2 + 10 dΘ/dt -150 [r(t) – Θ(t)] =0

take laplace

S2Θ(S) + 10S Θ(S) – 150 [R(S) – Θ(S)] = 0

Θ(S) [S2+10S+150] = 150 R(S)

Θ(S)/R(S) = 150/S2+10S+150

ῳn2 = 150 ῳn=12.25 r/s

2Ԑῳn = 10 Ԑ = 0.41

ῳd = 1.17 r/s

STEADY STATE ERROR

Steady state error is the value of the error signal e(t), when ‘t’ tends to infinity (∞). The
steady state error is the measure of system accuracy

Consider a closed loop

 R(S) = input signal

E(S) = error signal

G(S) H(S) = Feed back signal

C(S) = response

Error signal = E(S) = R(S)-C(S) H(S) (1)

Output signal = C(S) = E(S) G(S) (2)

Substitute the equation 2 in 1

E(S) = R(S) – E(S) G(S) H(S)

E(S) = [1+G(S) H(S)] = R(S)

E(S) = R(S) 1/1+G(S) H(S) (3)

Take inverse laplace

E(t) = L-1 [E(S)] = L-1[R(S) / 1+G(S) H(S)]

Let ess --- steady state error

ess = Ltt→∞ e(t) = value of e(t) when ‘t’ tends to infinity

final value theorem states that

if F(S) = L[F(t)] then

Ltt→∞ F(t) = Lts→0 S F(S)

Using final value theorem,

Steady state error, ess = Ltt→∞ e(t)

Lts→0 S E(S)

ess = Lts→0 S R(S)/1+G(S) H(S)

 when the control system is excited with standard input signal, steady state error may
be zero, constant or infinity.

Value of steady state error depends on the type number and the input signal.

The type number of a system is given by number of poles of the open loop transfer
function at the origin that is located at the origins.(it decides the steady state error)

Type -0 and step input - ess – constant

Type -1 and ramp input - ess – constant

Type -2 and parabolic input - ess – constant

Position error constant = Kp = Lts→0 G(S) H(S)

Velocity error constant = Kv = Lts→0 S G(S) H(S)

Acceleration error constant = Ka = Lts→0 S2G(S) H(S)

Where Kp, Kv and Ka are the general static error constant

Steady state errors of – type – o system

Unit step input signal R(S) = 1/S

ess steady state error = Lts→0 S R(S)/1+G(S) H(S)

ess = Lts→0 S 1/S / 1+G(S) H(S)

= 1 / 1+ Lts→0 G(S) H(S)

= 1/1+Kp

Kp = position error constant = Lts→0 G(S) H(S)

Type – 0 :-

Kp = Lts→0 G(S) H(S)

= Lts→0 K(S+Z1) (S+Z2)…….. / (S+P1) (S+P2)…….

= K Z1 Z2/P1 P2 = Kp = Constant

ess = 1/1+Kp = Constant

type -1:-

Kp = Lts→0 G(S) H(S)

 = Lts→0 K(S+Z1) (S+Z2)…….. / (S+P1) (S+P2)……. = ∞

Kp = ∞

ess = 1/1+Kp = 1/1+∞ = ∞

type – 2:-

ess = ∞

steady state error of type-1 system input signal unit ramp = 1/S2

ess = Lts→0 S R(S)/1+G(S) H(S)

R(S) = 1/S2

ess = Lts→0 S 1/S2 / 1+G(S) H(S)

= Lts→0 1/S/1+G(S) H(S)

= Lts→0 1/S+ Lts→0 S G(S) H(S)

= 1/Lts→0 S G(S) H(S)

Kv = Lts→0 S G(S) H(S)

Velocity error constant

Type -0 :-

Kv = Lts→0 S G(S) H(S)

= Lts→0 S K(S+Z1) (S+Z2)/(S+P1) (S+P2) = 0

ess = 1/Kv = 1/0 = ∞

type -1 :-

Kv = Lts→0 S G(S) H(S)

= Lts→0 S K(S+Z1) (S+Z2)/S(S+P1) (S+P2) = Z1Z2 K/ P1 P2 = Constant

= ess = 1/constant = constant

Type-2:-

Kv = Lts→0 S K(S+Z1) (S+Z2)/S2(S+P1) (S+P2) = ∞

ess = 1/∞ = 0
steady state error for type-2 and unit parabolic input signal.

R(S) = 1/S3 = parabolic input

ess = Lts→0 S R(S)/1+G(S) H(S)

ess = Lts→0 S 1/S3/1+G(S) H(S)

= Lts→0 1/S2/1+G(S) H(S)

= 1 / Lts→0 S2+ Lts→0 S2G(S) H(S)

ess = 1 / Lts→0 S2G(S) H(S)

ess = 1/Ka

Ka = acceleration error constant

= dts→0 S2 G(S) H(S)

Type-0:-

Ka = Lts→0 S2G(S) H(S)

= Lts→0 S2 K (S+Z1) (S+Z2)/ (S+P1) (S+P2)

Ka = 0

ess = 1/Ka = 1/0 = ∞

type-1

Ka = Lts→0 S2 K(S+Z1) (S+Z2)/S(S+P1) (S+P2) = 0

Ka = 0

ess = 1/0 = ∞

type-2:-

Kα = Lts→0 S2 K(S+Z1) (S+Z2)/S2(S+P1) (S+P2) = KZ1Z2/P1P2

Kα = constant

ess = 1/Ka = Constant

Input Error Steady state error

Type-0 Type-1 Type-2
Unit step u(t) Position 1/1+Kp 0 0
 ess (P) = 1/1+Kp
Unit Ramp t Velocity ∞ 1/Kv 0
ess (v) = 1/Kv
Unit Parabolic Acceleration ∞ ∞ 1/Ka
t2/2 ess(a) = 1/Ka

Dynamic error Co-efficients / Generalised error Co-efficients :-

Generalised error co-efficients gives the steady state error as a function of time.

From the GEC we can also ess find out any type of input

E(S) = R(S) 1/1+G(S) H(S) (1)

= R(S) F(S) (2)

Using convolution theorem,

State that the inverse laplace of the product of two s-domain functions is equal to
convolution of their time domain function.

L-1 [F(S) R(S)] = -∞ʃ∞ F(T) r(t-T) dT (3)

e(t) = -∞ʃ∞ F(T) r(t-T) dT

T → Dummy variable

Assume that input signal starts only t=0 (neglect -∞)

e(t) = 0ʃt=∞ F(T) r(t-T) dT (4)

expanding the function r(t-T) dT in Tylor series gives,

r(t-T) = r(t)-Tr(t) + T2/2! r(t) – T3/3! r(t)….. (-T)n/n! rn(t) (5)

steady state error is defined as = ess = Ltt→∞ e(t)

e(t) = 0ʃt=∞ F(T) [r(t)+n=1∑∞(-T)n dnr(t)/dtn] dt

ess = 0ʃ∞ F(T) [r(t)+n=1∑∞(-T)n dnr(t)/dtn] dt

= r(t) 0ʃ∞ F(T) dt+r(t)0ʃT (-T) F(T) dt + r(t) 0ʃT r2F(T) dt + …… + rK(t) (-1)K0ʃ∞TKF(T) dt+….

ess = r(t) (0+r(t) C1+ r(t)C2+…… +rK(t) ck +………….

C0 = 0ʃ∞F(T) dT (6)

C1 = -0ʃ∞T F(T) dT (7)

 C2 = 0ʃ∞T2 F(T) dT (8)

Ck = (-1)k0ʃ∞Tk F(T) dT (9)

From the definition of laplace transform

F(S) = 0ʃ∞ F(T) e-st dT

Taking the limits on both sides as S→0, we get

s→0Lt F(S) = s→0Lt 0ʃ∞ F(T) e-st dT

s→0Lt F(S) = 0ʃ∞ F(T) dT (10)

comparing equation 6 with 10

C0 = s→0Lt F(S)

Taking derivative with respect to S on both sides we obtain

s→0Lt dF(S)/ds = s→0Lt 0ʃ∞-T F(T) e-st dt

= - 0ʃT TF(T) dt (11)

Comparing 7 with 11, we get

C1 = s→0Lt dF(S)/dS

Similarly Ck = s→0Lt dKF(S)/dSK

Value of dynamic error constants can also be calculated by using static error constants

C0 = 1/1+Kp

C1 = 1/Kv

C2 = 1/Ka

ess = s→0Lt e(t)

# For a closed loop system with G(S) = 1/S+5 and H(S) =5. Calculate generalised error co
efficient and find error series

Given G(S) = 1/S+5

H(S) = 5

F(S) = 1/1+G(S) H(S) = 1/1+(1/S+5) 5 = S+5/S+10

The error co-efficients are

 C0 = s→0Lt F(S) = s→0Lt S+5/S+10 = 5/10 = 0.2

C1 = s→0Lt ds F(S) = s→0Lt (S+10)-(S+5)/(S+10)2

= s→0Lt 5/(S+10)2 = 5/100 = 0.05

C2 = s→0Lt d2F(S)/dS2 = s→0Lt [d/ds [5/(s+10) 2]]

= 5 s→0Lt [d/ds (1/s+10)2

= 5 s→0Lt [-2/(s+10)3] = -0.01

The error series is given by

e(t) = 0.2r(t)+0.05r(t)-0.01r(t)+………..

# An unity Feedback system has G(S) = 1/S(1+2S). The input to the system is described by
r(t) = 2+4t+6t2+2t3. Determine generalised error co-efficients and express the steady state
error as a function of time.

Given G(S) = 1/S(1+2S) r(t) = 2+4t+6t2+2t3

H(S) = 1

F(S) = 1/1+G(S) H(S) = 1/1+G(S)

= 1/1+1/S(1+2S) = 2S2+S/2S2+S+1

C0 = S→0Lt F(S) = S→0Lt 2S2+S/2S2+S+1 = 0

C1 = S→0Lt d/ds F(S) = S→0Lt d/ds [2S2+S/2S2+S+1]

= S→0Lt (2S2+S+1) (4S+1) – (2S2+S) (4S+1)/(2S2+S+1)2

= S→0Lt (4S+1)/(2S2+S+1) = 1

C2 = S→0Lt d2/ds2 F(S) = S→0Lt d2/ds2 [4S+1/(2S2+S+1)]

= S→0Lt (2S2+S+1)2 (4) – (4S+1) (2S2+S+1) (4s+1)/(2S2+S+1)4

= 2

C0 = 0; C1 = 1; C2 = 2

Error signal

e(t) = r(t)C0 + r(t) C1 + r(t)/2! C2 + ……. rn(t) Cn/n!

r(t) = 2+4t+6t2
 r(t) = 4+12t

r(t) = 12

e(t) = r(t)C0 + r(t) C1 + r(t)/2 C2

= (2+4t+6t2)0 + (4+12t) + 12/2 2

e(t) = 12t+16

ess = t→0Lt e(t) = ∞

# An unity feedback system is given as G(S) = 1/S(S+1). The input to the system is described
by r(t) = 4+6t+2t3. Find the generalised error co-efficient and the steady state error.

C0=0, C1 = 1, C2=0, C3=-6, C4=24

r(t) = 4+6t+2t3

r(t) = 6+6t2

r(t) = 12t

r(t) = 12

e(t) = 6t2 – 6

ess = ∞ = (S→0Lt e(t)

Another method

G(S) = 1/S(S+1) H(S) = 1 r(t) = 4+6t+2t3

F(S) = E(S)/R(S) = 1/1+G(S) = 1/1+(1/S(S+1))

= S2+S/S2+S+1

Dividing numerator by denominator polynomial

We have

E(S)/R(S) = S-S3+S4-S6

E(S) = SR(S) – S3R(S)+S4R(S)-S6R(S)+…….

S+S3+S4-S6
1+S+S2 S+S2
S+S2+S3
-S3
 -S3-S4-S5
S4+S5
S4+S5+S6
-S6

Comparing the equation with

ess = r(t) Co + r(t)C1 + r(t)C2+…..

C0 = 0; C1=1; C2/2! = 0 ; C3/3! = -1; C4/4! =1

C0 = 0; C1=1; C2=0; C3=-6; C4=24

r(t) = 4+6t+2t3

r(t) = 6+6t2

r(t) = 6+12t

r(T) = 12

ess = 6+6t2-12 = 6t2-6

ess = s→∞Lt e(t) = ∞

# Determine the generalised error co-efficients and steady state error for
G(S) = 1/S(S+1) (S+10) and H(S)=S+2 with input r(t)=6+t+t2

G(S) = 1/S(S+1) (S+10) ; H(S) (S+2)

F(S) = E(S)/R(S) = 1/1+G(S) H(S) = 1/1+[(S+2)/S(S+1)(S+10)]

= S3+11S2+10S / S3+11S2+11S+2

= 5S-22S2+94S3 …….

5S-22S2+94S3

2+11S+11S2+S3 10S+11S2+S3
10S+55S2+55S3+5S4
-44S2-54S3-5S4
-44S2-24S3+242S2-22S5
188S3+237S4+22S5
E(S)/R(S) = 5S-22S2+94S3 ……………
 E(S) = S5R(S) – 22S2R(S)+94S3R(S)
Comparing
C0=0; C1=5; C2/2!=-22; C3/3! = 94
C0=0; C1=5; C2=-44; C3=564
e(t) = 5r(t)-22r(t)+94r(t) ……
r(t) = 1+2t
¨r(t) = 1+2t
¨r(t) = 2
e(t) = 5(1+2t)-22(2)
= 10t-39
e(t) = t→∞Lt e(t) = t→∞Lt(10t-39) = ∞
# The overall transfer function of a unity feedback system is given by
C(S)/R(S) = 10/S2+6S+10 . find the values of the static error constant. Also determine the
steady state error for the input r(t) = 1+t+t 2

C(S)/R(S) = 10/S2+6S+10 = G(S)/1+G(S) H(S)

The numerator denominator by S2+6S
C(S)/R(S) =10/(S2+6S)/1+10/(S2+6S)
G(S) = 10/S(S+6)
Kp = s→0Lt G(S) = s→0Lt 10/S(S+10) = 1/0 = ∞
Kv = s→0Lt SG(S) = s→0Lt S 10/S(S+10) = 10/6 = 1.67
Ka = s→0Lt S2G(S) = s→0Lt 10S2/S(S+6) = s→0Lt 10S/S+6 = 0
F(S) = 1/1+G(S) = 1/1+10S (S+6) = S(S+6)/S 2+6S+10
C0 = s→0Lt F(S) = s→0Lt S(S+6)/(S2+6S+10) = 0
C1 = s→0Lt dt(S)/dS = s→0Lt (S2+6S+10) (2S+6) – S(S+6) (2S+6) / (S2+6S+10)2
C1 = 60/100 = 0.6
C2 = s→0Lt d2F(S)/dt2
= s→0Lt d/dt [10(2S+6) / (S2+6S+10)2]
= s→0Lt [(S2+6S+10)2 (S)–2(2S+6)(S2+6S+10)(2S+6) / (S2+6S+10)4
C2 = 0.052
e(t) = C0 r(t) + C1 r(t) + C2 r(t) …..
e(t) = 0.6 r(t) + (-0.052) r(t)
= 0.6(1+2t) + (-0.052(2))
= 0.496+1.2t
ess = t→∞Lt e(t) = ∞

Effect of adding a zero to a system

 Let us consider a 2nd order system with transfer function
C(S)/ R(S) = ῳn2/S2+2ԐῳnS+ῳn2 (1)
Now let us introduce a zero at S=-Z to the above second order closed loop system
Then we get
C(S)/ R(S) = (S+Z) ῳn2/S2+2ԐῳnS+ῳn2 (2)
adding a zero to the transfer function effects the gain of the system to make the gain
unaffected (to bring gain to unity) divide the numerator by Z now
C(S)/ R(S) = ((S+Z) ῳn2/Z)/S2+2ԐῳnS+ῳn2 (3)
To analyse the effect of adding a zero to the system, we have to find the time domain
specifications. We can write the equation (3) as
C(S)/R(S) = ῳn2/S2+2ԐῳnS+ῳn2 + S/Z (ῳn2/S2+2ԐῳnS+ῳn2) (4)
Let C(t) – step response of the system
ῳn2/S2+2ԐῳnS+ῳn2, then step response of S/Z (ῳn2/Z)/S2+2ԐῳnS+ῳn2) is
1/Z d/dt C(t). therefore the step response of the system with added zero is
Cz(t) = C(t)+1/Z d/dt C(t) (5)

 Rise time of Cz(t) is less than the system that of C(t)

 Maximum over shoot of Cz(t) is > C(t)

In summary, addition of zero to a system decreases rise time and increases the peak over
shoot.
More over, the values of peak over shoot and rise time depends on the selection of Z-
values.
The smaller values of zero, that is zero close to origin produces large over shoot where
as the large from the origin produces negligible effect on transient response.
Selecting the zeros near the origin should be avoided in design.
# for closed loop system with G(S) = 1/(S+5) and H(S) = 5
Cal.gen.error co-efficient and find error series
 F(6) = 1/1+G(S) H(S) = S+5/S+10
C0=0.2, C1= V dv ndv/V2 = (S+10) – (S+5)/(S+10)2 = 5/(S+10)2
C2 = d1/ds [5/(S+10)2] = 0.05
= (S+10)2 0 – 5(2) (S+10)21/(S+10)4
= -10/(S+10)2 = -10/100 = 0.01
# for servo mech with O.L transfer function given below explain what type of the input signal
give rise to a constant steady state error and calculation their values.
G(S) = 10/(S+2) (S+3) Kp = s→0Lt G(S) H(S)
Kp = 10/(S+2) (S+3) = 10/6 = 5/3
ess = 1/1+Kp = 1/1+5/3 = 0.375

# O.L transfer function G(S) = ῳ (S+2)/S(S+1) (S+3) H(S) = 1

Type -1 – Kv alone occure = s→0LT G(S) H(S)
= 20.2/3 = 40/3
ess = 1/Kv = 0.075 = 3/40
(ii) G(S) = 10/(S+2) (S+3) = Kp = 10/6 = 5/3
ess = 1/1+5/3 = 0.375

(iii) G(S) = 10/S2(S+1)(S+2) = Ka =10/2 = 5

ess = 1/Ka = 1/5 = 0.2

# obtain the response (i) unit step (ii) unit impulse of the following system
C(S)/R(S) = 10/S2+2S+10
Solution
C(S)/R(S) = 10/S2+2S+10
Case1:- for a unit step input
R(S) = 1/S
C(S) = R(S) = S/S2+2S+10
C(S) = 1/S(S2+2S+10)
S2+2S+10 = (-2)±√4-40 /2
= (-2)±√-36 /2
= (-2)±į6
= -1±į3
10/S(S2+2S+10) = A/S + B/S+1+į3 + C/S+1-3į
10 = A(S+1+3į) (S+1-3į) + B(S) (S+1-3į) + C(S) (S+1+3į)
PUT S=0
10 = A(1+3į) (1-3į) + 0 + 0
 10 = A(1+9)
A =1
By equating the co-efficients
10 = A[S2+2S+10) + B (S2+S-į3S) + C (S2+S+3įS)
 Compare co-efficients of S2
0 = A+B+C (1)
 Compare co-efficients of S
0 = 2A+B (1-3į) + C(1+3į) B(1-3į) + C(1+3į) = -2 (2)
 Compare co-efficients of constant
10 = A10
A=1
A+B+C = 0
B+C = -1 (3)

Solve 3 & 2 multiply the equation 3 with (1-3į)

3 = B(1-3į) + C (1+3į) = -2
3 X (1-3į) = B(1-3į) + C(1-3į) = (-1+3į)/ C(6į) = -1-3į
C = -1-3į/6į
= -(-į)-3į(-į)/6
C = į-3/6
= -3/6 + į 1/6
C = -0.5+į 0.1666
B+C = -1
B = -C-1
B = -(1+C)
= -[1-0.5+į 0.1666
= -[0.5+į 0.1666]
B = -0.5-į 0.1666
C(S) = 1/S+ -0.5-į0.1666/S+1+3į + -0.5+į 0.1666/S+1-3į
Taking inverse laplace
C(t) = 1+(-0.5-0.1666į) e(-1-3į)t + (-0.5+į 0.1667) e(-1+3į)

Case 2 :- for unit impulse input, R(S) = 1

C(S) = 10/S2+2S+10
= 10/(S+1)2 +32
C(t) = 3.33e-t sin3t
# A positional control system with velocity feedback is shown in figure. What is the response
C(t) to the input step unit. Given than Ԑ=0.5. also calculate rise time, peak time and maximum
over shoot, settling time.

Closed loop transfer function

C(S)/R(S) = G(S)/1+G(S) H(S)
G(S) = 16/S(S+0.8)n2
H(S) = (KS+1)
C(S)/R(S) = 16/S(S+0.8) + 16(KS+1)
= 16 /S2+(0.8+16K)S+16
Comparing the above equation with standard form of second order system equation,
ῳn2 = 16
ῳn = 4 rad/sec
(0.8+16K) = 2Ԑῳn
K = 2Ԑῳn-0.8/16
= 2 x (0.5) x 4 -0.8/16
K = 0.2
C(S)/R(S) = 16/S2+4S+16
 The damping ratio 0.5. hence the system is under damped and so the response of the
system will have damped oscillations.
 The roots of the characterstic polynomial will be complex conjugate
 The response in S- domain C(S) = R(S) 16/S2+4S+16
 For unit step input, R(S) = 1/s C(S) = 16/S(S2+4S+16)

Taking partial fraction

C(S) = 16/S(S2+4S+16) = A/S+BS+C/S2+4S+16
16 = A(S2+4S+16) + S(BS+C)
PUT s =0
A=1

The resides B & C are valuated by cross multiplying the following equation and equating the
co-efficients of the powers of “S”
16/S(S2+2S+16) = A/S + (BS+C)/S2+4S+16
16 = A[S2+4S+16] + S[BS+C]
16 = AS2+4AS+16A+BS2+CS
On equating the co-efficients of ‘S2’
A+B=0
A = -B
B = -1
On equating the co-efficients of ‘S’
4A+C = 0
C = -4A
C = -4
C(S) = 1/S + (-S-4)/S2+4S+16
= 1/S – (S+4)/S2+4S+4+12
= 1/S – (S+2+2)/(S+2)2+12
= 1/S – (S+2)/(S+2)2+12 – 2/(S+2)2+12 √12/√12
Taking inverse laplace
C(t) = L-1 [C(S)]
= L-11/S – (S+2)/(S+2)2+12 – 2/√12 √12/(S+2)2+12
= 1 – e-2t cos√12 t – 2/2√3 e-2t sin√12 t
C(t) = 1 – e-2t [1/√3 sin√12 t + cos√12 t]

The result can be converted to another standard form by constructing right angle triangle
with Ԑ & √1-Ԑ2

SinΘ =0.866 = √3/2

CosΘ = 0.5 = 1/2
tanΘ = 1.732
Θ = tan-1(1.732) = 60.
= 1.047 radians
The time response C(t)
C(t) = 1-e-2t [1/B x 2 x sin√12t x ½ + 2/√3 cos√12t x √3/2]
= 1-e-2t 2/√3 [sin√12t cosΘ +cos√12t sinΘ]
= 1-2/√3 e-2t [sin√12t+Θ]
= 1-2/√3 e-2t [sin√12t+1.047]
Damped frequency of Oscillation = ῳd = ῳn√1-Ԑ2
 =4√1-0.52 = 3.464 r/s
Rise time = tr = π-Θ/ῳd = π-1.047/3.464
= 0.604 sec
Peak time = tp = π/ῳd
= 0.907 sec
Percentage maximum over shoot = %Mp = e-Ԑπ/√1-Ԑ2
= 0.163 x 100
= 16.3%
Settling time ts = 5% of error
= 3/ῳnԐ = 1.5 sec
ts = 2% error = 4/Ԑῳn = 2 sec

# A closed loop servo is represented by the differential equation d 2C/dt2+8dC/dt = 64e where
C is the displacement of the output shaft, r is the displacement of the input shaft and e=r-c.
determine undamped natural frequency, damping ratio and percentage maximum over shoot
for unit step input.
d2C/dt2 + 8 dC/dt = 64e (1)
e = r-c (2)
put e=r-c
d2C/dt2 + 8 dC/dt = 64(r-c)
L(C) = C(S)
L(r) = R(S)
Taking laplace
S2C(S)+8SC(S) = 64 [R(S) – C(S)]
[S2+8S+64] C(S) = 64 R(S)
C(S) / R(S) = 64/S2+8+64
Compare with standard equation
ῳn2 = 64, ῳn =8 r/s
2Ԑῳn = 8
Ԑ = 8/2Ԑῳn = 0.5
Peak over shoot %Mp =e-Ԑπ/√1-Ԑ2 100
=e-0.5/√1-(0.5)2 100
= 16.3%

# A unity feedback system has the forward transfer function G(S) = K (2S+1)/S(5S+1) (HS) 2
the input r(t) = 1+6t is applied to the systems. Determine the minimum value of K. if the
steady state error is to be less than 0.1
 Input r(t) = 1+6t
Take laplace
R(S) = 1/S + 6/S2
Error signal E(S) = R(S)/1+G(S) H(S)
E(S) = (1/S + 6/S2) / [1+ K(2S+1)/S(5S+1) (1+S)2] X 1
E(S) = (1/S + 6/S2) / [S(5S+1) (1+S)2 + K(2S+1)/S(5S+1) (S+1)2]
E(S) = [1/S+6/S2] [S(5S+1) (1+S)2 / S(5S+1) (S+1)2+K(2S+1)]
The steady state error ess can be obtained from find value theorem.
ess = s→0Lt e(t) = s→0Lt S E(S)
= s→0Lt S [1/S [S(5S+1) (1+S)2 / S(5S+1) (1+S)2+K(2S+1)]
+ 6/S2 [S(5S+1) (1+S)2 / S(5S+1) (1+S)2+K(2S+1)]
= 0+6K
ess = 6/K
given that ess < 0.1
0.1 = 6/K
K = 60
# Determine the generalised error co-efficients and steady state error for
G(S) = 1/S (S+1) (S+10) and H(S) = (S+2) with input r(t) = 6+t+t 2

Answer :- C0=o, C1=5, C2=-44, C3=564

ess = ∞

# For unity feedback system

G(S) = 30(S+1) (S+2) (S+3)/S3 (S+5) (S+10) an input t3 u(t) is applied. Find steady state
error. [ess = 5/3]
 UNIT III
FREQUENCY RESPONSE

Frequency response – Bode Plot – Polar plot – M & N circles – Nichol’s chart –
Determination of closed loop response from open loop response – correlation b/w FDS
and TDS – programs with MATLAB

3.1. Frequency Response:-

Frequency Response is a steady state response (output) of a system when the input
to the system is a sinusoidal signal

x(t) = X sinῳt
y(t) = y sin (ῳt+φ)
the magnitude and phase relationship between the sinusoidal input and the steady state
output of a system is termed as “Frequency Response”
The response is independent of the amplitude and phase of the input signal.
Frequency response is obtained by varying the frequency by keeping the
magnitude of input signal at a constant
The transfer function T(S)S is replaced by ‘įῳ’ than resulting transfer function
will be T(įῳ) is called sinusoidal transfer function
Open loop transfer function = G(įῳ) = [G(įῳ)] [G(įῳ)]
Closed loop transfer function M (įῳ)= C(įῳ)/R(įῳ)

3.2. Frequency domain specification:-

The performance and characteristics of a system in frequency domain are specified in
terms of domain specification (The requirements of system to be designed are usually
specified in terms of these specifications)
Some of the frequency domain specifications are as follows,
1. Peak resonant (Mr)
2. Resonant Frequency (ῳr)
3. Bandwidth (ῳn)
4. Cut off rate
5. Gain margin
6. Phase margin
3.2.1. Definitions

1. Resonant Peak (Mr)

Maximum value of the magnitude of closed loop transfer function is called peak
resonant (Mr) [this increased means large over shoot and transient response]

2. Resonant Frequency:- (ῳr)

Frequency at which the resonant peak occurs is called “Resonant” frequency [related
to frequency of oscillation also indicate the speed of response]

3. Bandwidth:- (ῳb)
Bandwidth is the range of frequency for which the system gain is more than -3dB

4. Cut-off frequency:-
The frequency at which the gain is “-3dB” is called cut-off frequency

5. Cut-off rate:-
Slope of the log magnitude curve near the cut off frequency is called “cut-off rate”
indicate the ability of the to distinguish the signal from noise

6. Gain Margin (Kg)

The reciprocal of the magnitude of the open loop transfer function at phase cross
over frequency.
Kg = 1/│G(įῳpc)│ [ie dB = Kg = 20log│G(įῳpc)│
[Kg ↑, gain of the system ↑ with affecting the stability]

7. Phase cross over frequency:- (ῳpc)

The frequency at which the phase of open loop transfer function is 180’ is called
phase cross over frequency.

8. Phase Margin ɣ:-

The amount of additional phase lag at the gain cross over frequency required to bring
the system to the verge (edge) of instability.
ɣ is obtained by adding 180’ to the phase angle φ of the open loop transfer function
at gain crossover frequency
phase margin = ɣ = 180+φgc
φgc = ∟G(įῳ)ῳ =ῳpc

3.3.2. Estimation of Frequency Domain specification

1. Peak Resonant:- (Mr)

 T(S) = C(S)/R(S) = ῳn2/S2+2ԐῳnS+ῳn2 (1)
Put S =įῳ
C(įῳ)/R(įῳ) = ῳn2/(įῳ)2+ 2Ԑῳn (įῳ) + ῳn2
= ῳn2/(-ῳ)2+ į2Ԑῳnῳ + ῳn2

= ῳn2/ῳn2[-ῳ2/ῳn2+j2Ԑ ῳ/ῳn +1]

= 1 / 1- (ῳ/ῳn)2 + j2Ԑ ῳ/ῳn (2)

Let u = normalised frequency = ῳ/ῳn

T(jῳ) = 1 / 1-u2 + j2Ԑu

M = Magnitude of closed loop transfer funds
T(jῳ) = │1 / (1-u2)2 + (2Ԑu)2 │1/2 (3)
α = Phase angle of closed loop transfer function
T(jῳ) = -tan-1 (2Ԑu / 1-u2) (4)
Peak resonant is maximum value of M. this can be obtained differentiating the
magnitude equation with respect to “i.e”
dM/du = 0 │u=ur
Magnitude = M = │ 1/(1-u)2 + (2Ԑu)2│1/2
= [(1-u)2 + (2Ԑu)2]-1/2

= dM/dt = d/du = [(1-u2)2 + 4Ԑ2u2]-1/2

= -1/2[(1-u2)2 + 4Ԑ2u2]-3/2

= -[2(1-u2) (-2u) + 8 Ԑ2u] / 2[1-u2)2 + 4Ԑ2u2]3/2

= 4u[1-u2] + 8 Ԑ2u / 2[(1-u2)2 + 4Ԑ2u2]3/2

= u[(1-u2) + 4Ԑ2] / [(1-u2)2 + 4Ԑ2u2]
At u = ur dM/du = 0

4ur [1-ur2] - 8Ԑ2ur / 2[(1-ur2)2 + 4Ԑ2ur2]3/2

4ur[1-ur2] - 8Ԑ2ur = 0
4ur - 4ur3] - 8Ԑ2ur = 0
4ur [1-ur2-2Ԑ2] = 0
1-ur2-2Ԑ2 = 0
Ur2 = 1-2Ԑ2
Ur = √1-2Ԑ2
Peak resonant occurs when u = ur = √1-2Ԑ2
Mr = [ 1 / (1-u2)2 + 4Ԑ2u2]1/2 u = ur
= │ 1 / (1-ur2)2 + 4Ԑ2ur2│1/2
= │ [1 – (√1-2Ԑ2)2]2 + 4Ԑ2(√1-2Ԑ2)2│1/2
= 1 / (1-1+2Ԑ2)2 + 4Ԑ2(√1-2Ԑ2)1/2
 = 1 / (4Ԑ4+ 4Ԑ2 - 8Ԑ4)1/2
= 1 / [4Ԑ2- 4Ԑ4]1/2
= 1 / [4Ԑ2 [1-Ԑ2]]1/2
Peak resonant Mp = 1 / 2Ԑ √1-Ԑ2 (5)

2. Resonant Frequency :- ῳr
Normalised frequency = ur
= ῳr/ῳn = √1-2Ԑ2
ῳr = Resonant frequency = ῳn√1-2Ԑ2

3. Bandwidth :- ῳn
Ub = ῳb/ῳn [-3db or 1/√2]
Magnitude of closed loop transfer function is -3db corresponding range of frequency is
Bandwidth

M = 1 / [(1-ub2)2 +4Ԑ2ub2]1/2 = 1//√2 (6)

2 = (1-ub2)2 + 4Ԑ2ub2
1+ub4-2ub2+4Ԑ2ub2 = 2
Ub4 – 2ub2(1-2Ԑ2)-1 = 0
Let assume x = ub2
X2-2x(1-2Ԑ2)-1 = 0
Roots of the quadratic equation is
X = 2(1-2Ԑ2) ± 2√(1+4Ԑ4-4Ԑ2)+1
Taking +ve value only
X = (1-2Ԑ2) + √2-4Ԑ2-ԐԐ4
Ub = √x = [(1-2Ԑ2 + √(2-4Ԑ2-4Ԑ4]1/2
And also ub = ῳb/ῳn (7)
ῳb = Bandwidth = ῳn [(1-2Ԑ2) + √(2-4Ԑ2-4Ԑ4]1/2 (8)

Phase margin :- (ϒ)

open loop transfer function = G(s)
G(s) = ῳn / s(s+2Ԑ ῳn) (9)
Put S = jῳ
G(jῳ) = ῳn/ jῳ(jῳ+2Ԑῳn)
= ῳn2 / (jῳ)2+j2Ԑῳnῳ
= ῳn2/-ῳ2+ jῳῳn2Ԑ
= ῳn2 / ῳn2 [-ῳ2/ῳn2 + j2Ԑῳ/ῳn] (10)
Let u = ῳ/ῳn = Normalised frequency
G(jῳ) = 1 / -u2+j2Ԑu
Magnitude of G(jῳ) = │G(jῳ)│
│G(jῳ)│ = 1 /√u4+4Ԑ2u2 (11)
Phase angle ∠G(jῳ) = ϒ = -tan-1[2Ԑu/u2]
= -tan-1[2Ԑu/u] (12)

4.Gain gross over frequency :- ῳgc

Frequency at gain cross over frequency the magnitude of G(jῳ) is unity that is the
frequency at which the dB magnitude is zero
Let Normalised gain cross over frequency = Vgc = ῳgc/ῳn (13)
U = Ugc
│G(jῳ)│ = 1 = 1/√Ugc4+4ԐUgc2
Ugc4+4Ԑ2Ugc2 = 1
Let x = Ugc2 = x2+2Ԑ2x-1 = 0
X = -2Ԑ2 ± √4Ԑ2+1
Ugc = √x = [-2Ԑ2+√4Ԑ2+1]1/2
Take +ve values only
ϒ = 180 + tan-1 [2Ԑ/Ugc]
ϒ = 180 + tan-1 [1 / [-2Ԑ2 +√4Ԑ2+1]1/2]

(1). Determined the closed loop bandwidth, closed loop peak magnitude, gain margin and
phase margin for the following system whose open loop transfer function with unity
feedback system is G(S) = 4/S(S+1)

G(S) = 4/S(S+1)
T(S) = G(S) / 1+4(S) 1 = 4/S2+S+4
ῳn2 = 4 = ῳn = 2 r/s
2Ԑῳn = 1 = Ԑ = 0.25
Bandwidth ῳn [(1-2Ԑ2) + (4Ԑ2-4Ԑ2+2)1/2]1/2
=2[1-2(0.25)2+4(0.25)4-4(0.25)2 +2)1/2]1/2
= 2.969 r/s
Mr = 1/2Ԑ√1-Ԑ2 = 2.06559

Gain margin:- open loop phase varies from -90 to 180 and it does not cross the -180’
The gain margin is ∞
Phase margin
ɸm = 90 – tan-1 [1/2Ԑ(1+4Ԑ4)1/2 – 2Ԑ2)1/2]
= 28°

(2). Forward path transfer function of unity feedback system is G(S) = K / S(S+6). Find Peak
resonant, resonant frequency and Bandwidth of the closed loop system for the following
values of K
(i) K = 5 and (ii) K = 100
 (i) When K = 5 G(S) = 5/S(S+6)
Closed loop transfer function = K / S2+6S+K
ῳn2 = 5 = ῳn = 2.236 r/s
2Ԑῳn = 6 = Ԑ = 6/2(2.36) = 1.3416
Mr = 1 since Ԑ > 0.707
ῳr = 0 since Ԑ > 0.707
Bandwidth = ῳd = ῳn = [(1-2Ԑ2) + √4Ԑ2-4Ԑ2+2]1/2
= 2.236[(1-2(1.3416)2+√4(1.3416)4-4(1.3416)+2]
= 2.236 [-2.5997+2.78547]1/2
= 0.9637

(i) For K = 100

G(S) = 100/S(S+6)
T(S) = 100/S2+6S+100
ῳn = 10 r/s
2Ԑῳn = 6 Ԑ = 0.3
Mr = 1/2Ԑ√1- Ԑ2 = 1.747
ῳr = ῳn √1- 2Ԑ2 = 10 √1-2(0.3)2
= 9.0554
ῳd = ῳn [(1-2Ԑ2) + √4Ԑ4-4Ԑ2+2]1/2
= 10 [(1-2(0.3)2 + √4(0.3)4-(4.3)2+2]1/2
= 14.5369

(3). From the pole zero configuration of closed loop transfer function is given, determine
bandwidth, Peak resonant, normalised frequency

Only poles, no zeros

(-2±j2)
S = pole -1 (-2+2J)
S = pole -2 = (-2-2j)
(S+P1) (S+P2) = 0
(S+2-2J) (S+2+SJ) = 0
(S+2)2 – (2J)2
 S2+4+2S+4 = 0
S2+2S+8 = 0
CLOSED loop transfer function = 1/S2+2S+8

ῳn2 = 8 ῳn = √8 r/s = 2.828 r/s

2Ԑῳn = 2 Ԑ = 1/√2 = 0.707
ῳb = ῳn [1-2Ԑ + √2+4Ԑ4-4Ԑ2]
2

= 2.828 r/s
Ur = ῳr/ῳn ῳr = ῳn √1- Ԑ2 = 0.0498
Ur = 0.0176
Mr = 1/2Ԑ√1-2Ԑ2 = 1
Graphical approach:-
To find the FDS, following methods are used
1. Polar Plot
2. Bode Plot open loop transfer function ¥,kg, wpc and wgc
3. M and N circles
4. Nichol’s Plot closed loop Mr,Wb, Wr

Polar Plot :- As the frequency varies from 0 to ∞, the magnitude and phase angles varies
and hence the phasor G(jw) traces a lows in the s –plane. This lows is known as “Polar Plot”
Polar plot of a sinusoidal transfer function G(jw) is a plot of the magnitude of G(jw) vs
phase angle of G(jw) on polar co-ordinates as w varies from zero to infinity.
Minimum phase system:-
With minimum phase transfer functions. In minimum phase transfer functions, all
poles and zero will lie on the left half of S – plane.
All Pass systems are the systems with all pass transfer functions. In all pass transfer
functions, the magnitude is unity at all frequencies and the transfer function will have anti
– symmetric pole zero pattern. (ie) for every pole in the left half S-plane, there is a zero in
the mirror image position with respect to imaginary axis)
Non – minimum phase transfer frequency:- A transfer function which has low or
more zeros in the right half S-plane is known as Non-Minimum phase transfer function.

For minimum phase transfer functions, with only poles, the type number of the
system determines the quadrant in which the plar plot starts, and the order of a system
determines the quadrant in which the polar plat end
G(S) = K/S
G(S) = 1/S(ST1+1) (ST2+1)
G(S) = K / S4(TS+1)

Procedure to plot the polar graph

1. Obtain the open loop transfer function
2. Put S = jw
3. Find Magnitude of G(jw) ie │G(jw)│
4. Find phase angle of G(jw) ie ĹG(jw)
5. By varing the ῳ from 0 ∞ find │G(jw)│and ĹG(jw) and Tabulate it.
6. Plot the graph based on the tabulation
7. Obtain the frequency domain specification from the graph

The open loop transfer function of UFB system is given by G(S). H(S) = 1 / S(S+1) (S+2)
Using nuquist plot evaluate Kg, r, Wgc and Wpc
 1. G(S) H(S) = 1/S (S+1) (S+2) S=0, S=-1, S=-2 left half s plane
2. Put s=jw
G(jw) = 1/jw(jw+1) (jw+2)
3. Magnitude of G(jw) (0+jw)
│G(jw)│ = 1 /√ῳ2√1+ῳ2+√4+ῳ2
4. Phase of G(jw)ĹG(jw) = -tan-1 (ῳ/0) – tan-1(ῳ/1)-tan-1(ῳ/2)
= -90-tan(ῳ)-tan-1(0.5ῳ)

FDS are

ɸgc = -126

¥ = 180 + ɸgc = 54’

Kg = 1 / │G(jw)│ ῳ=ῳpc
= 1/0.16 = 6.25
Wgc = 0.44 r/s
Wpc = 1.42 r/s
The open loop transfer function of a UFB is given by G(S) = (1+0.2S) (1+0.025S) / S3
(1+0.0058) (1+0.001S)
Determine phase and gain margin, Wgc, Wpc using polar plot
1. G(S) H(S)= (1+0.2S) (1+0.025S) / S3(1+0.005S) (1+0.001S)
2. PUT S = jw
G(jw) = (1+0.2JW) (1+0.025JW) / (JW)3 (1+0.005JW) (1+0.001JW)
3. │G(JW)│ = √1+0.04ῳ2√1+6.25X10-4ῳ2 / √ῳ6√1+2.5X10-5ῳ2√1+1X10-6ῳ2
4. ĹG(JW) = tan-1(0.2ῳ/1)+tan-1(0.0025jw/1) – tan-1(ῳ/0)-tan-1(ῳ/0)-tan-1(ῳ/0)-tan-1
(0.005ῳ/1)-tan-1(0.00/ῳ/1)
 Wgc = 1.05 r/s
ɸgc = -255
¥ = 180+ ɸgc = -75’
Kg = ∞ │G(jw)│ = 0
For the following system sketch the polar plot
G(S) H(S) = 500 / S(S+6) (S+9)
│G(jw)│= 500/ῳ √ῳ2+36√ῳ2+81
Ĺ G(jw)│= -90-tan-1(ῳ/6)-tan-1(ῳ/9)
Gain Adjustment using polar plot
(a) To determine K for specified Gain Margin
1. Draw the polar plot with K = 1
2. Let it cuts the -180’ line at point –B
3. Corresponding gain is GB
4. To calculate GA value, the value of specified gain margin given in the problem is
“x” dB
20log 1/GA = X dB

Log 1/GA = x/20

1/GA = 10x/20
GA = 1/10 x/20
If K>1 gain is increased
K<1 gain is decreased
K = GA/GB
(b) To determine K for specified phase margin
(1) Draw the polar plot with K=1
(2) It cuts the unity circle at point –B
(3) Let the specified (given in the problem) phase margin is X
X0 =180+ɸgc(x)
ɸgc(x) = 180’-x0
(4) In the plot, the radial line corresponding to ɸgc(x) will cut the G(jw) with K=1 at
point –A and the magnitude corresponding to that point be GA
K = GB/GA = 1/GA GB=1

Sketch the polar plot for the given UFB system having open loop transfer function
G(S) = K / S(0.5S+1) (1+5S)
Determine the value of “K” so that given margin is 20dB
1. Put K=1
2. G(S) = 1 / S(1+0.5S) (1+5S)
3. Put S = jw
4. │G(jw)│ = 1 / ῳ √0.25ῳ2+1√25ῳ2+1
5. Ĺ G(jw) = -90-tan-1(0.5ῳ)-tan-1(5ῳ)
When K = 1
Gain margin = 2.5
ɸgc = -167
¥= 180+ ɸgc = 13’
GB = 0.4
GA = ?
20log 1/GA = 20 [given in the problem]
GA = 1/1020/20 = 0.1
K = GA/GB = 0.1/0.4 = 0.25
K = 0.25
Sketch the polar plot and determine the value of K so that
(1) Gain margin is 18dB
(2) Phase margin is 60’
Whose open loop transfer function is G(S) = K/S (1+0.2S) (1+0.05S)
(I) Put K = 1 G(S) = 1/S(1+0.25) (1+0.05S)
(II) Put S = jw G(jw) = 1 / jw(1+0.2jw) (1+0.05jw)
(III) │G(jw)│ = 1 / ῳ√1+0.2ῳ2√1+0.05ῳ2
(IV) ĹG(jw) = -90-tan-1(0.2ῳ)-tan-1(0.05ῳ)

given phase margin is 60’

K = 1 ; Kg = 1/0.04 = 25
ɸg = -104
¥ = 76’
Given ¥x = 60’
60’ = 180 + ɸgc(x)
ɸgc(x) = -120’
in the polar plot -120’ line cuts the graph at point A and cuts the unity circle at point B
Let GA = Magnitude of G(jw) at point A
GB = Magnitude of G(jw) at Point B
From the graph GA = 0.425
GB = 1
K = GB/GA = 1/0.425 = 2.353
(a) Give gain margin is 18dB with K = 1
Let the graph cuts the -180’ line at point B the corresponding gain is GB
 from the graph GB = 0.04
The gain margin of 28dB with K=1 has to be reduced to 18dB and so K has to be increased to
a value greater than one
Let Ga be the gain at -180’ for a gain margin of 18dB
20 log 1/GA = 18
GA = 1/1018/20
GA = 0.125
K = GA/GB = 0.125/0.04 = 3.125
BODE PLOT:-
Bode plot is a frequency response plot of the transfer function (open/closed) usually
open loop are drawn of a system Bode Plot consists of two graphs
(1) One is a Plot of the magnitude of a sinusoidal transfer function versus log ῳ
(2) The other is a plot of the phase angle of a sinusoidal transfer function versus log ῳ
Basic factors of G(jw)
1. Constant Gain K
G(S) = K if n=0 n= 0
K>1 K/S n
0 dB
2. integral factor
K / Sn if n = 0

3. Derivative factor
KSh = if n = 0

4. First order factor in denominator

G(S) = 1/(1+ST)n
5. First order factor in the numerator G(S) = (1+ST)m

Procedure for Bode Plot

 (a) Magnitude plot
(1) Convert G(S) into standard form [std form :- real part value is always unity]
(2) Put S=jw
(3) Find the corner frequencies for all the terms in the G(S) except KSh or K/SH if n=0 and
n= 0
(4) Arrange the corner frequencies in ascending order
(5) Select upper and lower cut off frequency (ωL & ωH)
Ie ωL < ωCL
ωH > ωCH
(6) Find slope for all the terms and fill the tabular coloumn

(7) Magnitude in dB – 3 cases are there

Case 1 :- KSn (if n=0) K1 = gain in standard form
(i)ω= ωL = Gain A = 20log K1(ω)n
(ii)ω= ωC1 = Gain A = 20log K1(ω)n
(iii)ω= ωC2 = Gain A = [Change in slope from ωCf to ωC2 x log ωC2/ωC1 + Aω=ω1]
(iv)ω= ωC3 = Gain A [Change in slope from ωC2 to ωC8 x log ωC3/ωC2 + Aω=ω2
ω=ωH = Gain A [Change in slope from ωC to ωH x log ωH/ωC + Aω=ω1
case 2 :- K/SN if n=0
(i) ω=ωL = Gain A = 20logK1/ωn
(ii) ω=ωC1 = Gain A = 20logK1/(ω)n
(iii) remains steps is a same as in case (i) upto ωH
Case 3 :- KSh or K/Sh if n=0
(i) ω=ωL = Gain A = 20log(K)
(i) ω=ωC1 = Gain A = 20log(K)
Remaing steps are same as in the case (i) upto ωH
(b) Phase plot:-
(i), (ii), (iii), (iv) and (v) step same in the magnitude plot.
(6) ĹG(jw) = Write the expression for phase of G(jw)
(7) for different values of ω find different angle ω=ωL, ωC1…..ω
To find stability using Body Plot
If both gain and phase margin values are positive than the systemis more stable.
Otherwise the system is unstable (even though any one is negative)

Sketch the Bode Plot showing the magnitude in decibels and phase angle in degrees as a function
of log frequency for the transfer function given by
G(S) = 5x250 / S(1+0.2S) (1+0.02S)
= (5+S) (50+S)
And hence determine the gain and phase margin of the system
(a) Magnitude Plot
1. Given G(S) is already in standard form
2. Put s=jw G(jw) = 5 / jw(1+0.2jw) (1+0.02jw)
3. Find corner frequency kept 5 & 1/S
1/(1+0.2jw) = C.F = value of real part / Imaginary Part = 1/0.2 = 5 r/s
1/1+0.02jw = C.F = 1/0.02 = 50 r/s
4. Arrange the Corner frequency in ascending order
ωu < ωC2
ωu = 5 r/s and ωC2 = 50 r/d
5. Select ωL & ωH (it should be in log Scale)
ωL < ωC1 ie 1<5 = ωL = 1 r/s
ωH > ωC2 ie 100 >50 = ωH = 100 r/s
Initude in dB (2nd Case)
ωL = A = 20Log(5/1) = 13.979 dB
ωC1 = 5 r/s ; A = 20Log 5/5 = 0
ωC2 = 50 r/s ; A = [-40 log (5/5) +0 ] = -40dB
ωH = 100 r/s ; A = [-60 log 100/50 + (-40)] = -58.06
see plot:-
write the expression for ĹG(jw)
ĹG(jw) = -90-tan-1(0.2ω)-tan-1(0.02ω)
ωC = 1 r/s = Ɵ = -102.45 13.9
ωC1 = 5 r/s = Ɵ = -164.7 0
ωC2 = 50 r/s = Ɵ = -219.3 -40
ωH = 100 r/s = Ɵ = -240.57 -58
it :- ɸgc = -137.5
¥ = 180 -137.5 = 52.5
ωgc = 5 r/s
ωpc = 17 r/s
Draw bode plot for open loop system
G(S) = 20S2/(1+0.2S) (1+0.02S)
Determine the frequency Domain specifications.
1. G(S) is in Standard form
2. Put s=jw = 20(jw)2 / (1+0.2jw) (1+0.02jw)
3. Find the Corner frequency
1/1+0.2jw = Corner frequency = 1/0.2 = 5 r/s
1/1+0.02jw = Corner frequency = 1/0.02 = 50 r/s
ωL = 1 r/s and ωH = 100 r/s
 5.Gain in dB
ω = ωL = 1 r/s A = 20log 20(ω2) = 26 dB
ω = ωC1 = 5 r/s A = 20log 20(S2) = 54 dB
ω = ωC2 = 50 r/s A = 20log 20(50/5)+54 = 74 dB
ω = ωH = 100 r/s A = 0log 100/50 +74 = 74 dB
6. ĹG(jw) = 180 –tan-1(0.2ω)-tan-1 0.02ω
Sketch the bodeplot and Act FDS
G(S) H(S) = 45000 (S+2)/(S+10) (4S+2) (S+50)
(i) convert G(S) into standard form
G(S) = 45000 2(S/2+1) / 10(1+S/10) 2(4S/2+1) 50(S/50+1)
G(S) = 90(1+0.5jw) / (1+0.1jw) (1+2S) (1+0.02S)
(ii) put s =jw
G(jw) = 90(1+0.5jw) / (1+0.1jw) (1+2jw) (1+0.02jw)
(iii) find Corner Frequency
(1+0.5jw) = ωL = corner frequency = 1/0.5 = 2 r/s
(1+0.1jw) = corner frequency = 1/0.1 = 10 r/s
(1+2jw) = corner frequency = 1/2 = 0.5 r/s
(1+0.02jw) = corner frequency = 1/0.02 = 50 r/s
(iv) arrange and find ωL and ωH
0.1 < 0.5 < 2 < 10 < 50 < 100

(vi)Magnitude in dB
ω = ωL = 0.1 r/s = A = 20log(90) = 39dB
ω = ωC1 = 0.5 r/s = A = 20log(90) = 39dB
ω = ωC2 = 2 r/s = A = [-20log(2/0.5)+39] = 27dB
 ω = ωC3 = 10 r/s = A = [0log(10/2) + 27.04] = 27dB
ω = ωC4 = 50 r/s = A = [-20log(50/10) + 27.04]= 13.06dB
ω = ωH = 100 r/s = A = [-40log(100/50) + 13.06] = 1.01 dB
(vii) ĹG(jw) = tan-1(0.5ω)-tan-1(0.1ω)-tan-1(2ω)-tan-1(0.02ω)

Draw the Bodeplot and obtain frequency domain specification for the open loop transfer function
G(S) H(S) = S2/(1+0.2S) (1+0.2S)
(i) H(S) = 1 and standard form already
(ii) Put s = jw G(jw) = (jw)2/(1+0.2jw) (0.02jw)
(iii) Find Corner Frequency
(1+0.2jw) = Corner Frequency = 1/0.2 = 5 r/s
(1+0.02jw) = Corner Frequency = 1/0.02 = 50 r/s
(iv) ωL = 1 r/s and ωH = 100 r/s

(vi)Magnitude in dB (Case -1)

ω = ωL = 1 r/s = A = 20log[1(1)2] = 0
ω = ωC1 = 5 r/s = A = 20log[(5)2] = 27.95dB
ω = ωC2= 50 r/s = A = [20log[50/5] + 27.9] = 47.9dB
ω = ωH = 100 r/s = A = [0log[100/50]+47.9] = 47.9dB
(vii) ĹG(jw) = 180-tan-1(0.02ω) – tan-1(0.2ω)
 ɸgc = 168
¥ = 180+ ɸgc = 348’
ωgc = 1 r/s
Kg = ∞
For the given open loop transfer function G(S) = K.e -0.2S / S(S+2) (S+8) find the value of K
Such that the system is stable with
(a) Gain Margin = 6 dB
(b) Phase Margin = 45’
(i) Gain adjustment Problem K = 1 and convert G(S) into standard form
G(S) = e-0.2S/S2(S/2+1) 8 (S/8+1)
G(S) = e-0.2S/16S(1+0.5S) (1+0.125S)
= 0.0625e-0.2S / S(1+0.5S) (1+0.125S)
G(jw) = 0.0625e-j0.2ω / jw(1+0.5jw) (1+0.125jw) Ae-s = Ae-jw
│ Ae-jw│= A
Ĺ Ae-jw = -ω . 180/𝜋 deg
3.Corner Frequency
(1+0.5jw) = Corner Frequency = 1/0.5 = 2 r/s
(1+0.125jw) = Corner Frequency = 1/0.125 = 8 r/s
4. ωL and ωH = 1 r/s and ωH = 10 r/s

6.Magnitude in dB
ω = ωL = 1 r/s = A = 20log(0.0625/1) = -24.08dB
ω = ωC1 = 2 r/s = A = 20log(0.0625/2) = -30.10dB
ω = ωC2 = 8 r/s = A = [-40log(8/2)-30.10] = -54.18dB
ω = ωH = 10 r/s = A = [-60log(10/8)-54.18] = -60dB
7. ĹG(jw) = -90-0.2 x ω x 180/A – tan-1(0.5ω)-tan-1(0.125ω)

Calculation of K1
Phase margin ϒ = 45’, ɸgc = ϒ-180=45-180 = -135’
With K=1, the dB gain at ɸ = -135’ is -24dB . this gain should be made to zero have to PM
of 45’. Hence every point of magnitude plot a dB gain of 24dB should be added. The
corrected magnitude plot is obtain by shifting the plot with K=1 by 24dB upwards. The
magnitude correction depends upon of frequency. Hence the magnitude of 24dB is
contributed by the term K. the value of K is calculated by equating 20log K to 24dB.
20log K = 24 K=1024/10 K = 15.84
(-24+24, -30+24, -54+24, -60+24)
0, -6, -30, -36
With K=1, gain margin is -(-33) = 33dB. But required gain margin is 6dB (given) hence
every point of magnitude plot a dB gain of 27dB should be added. This addition of 27dB
shifts the plot towards upwards. The magnitude correction is independent of frequency.
Hence the magnitude 27dB is contributed by the term K value of K is calculated.
By equating 20log K = 27dB
K = 10 28/20 = 25.12 = 22.38
(-24+27, -30+27, -54+27, -60+27)
3, -3, -27, -33

Open loop Transfer function of a UFB system is given by

G(S) = 50K/S(S+10) (S+5) (S+1)
(i) Find gain margin and phase margin
(ii) Value of steady state error co-efficient for a gain of 10dB. Value which will be make the
system maginally stable.
K = 3.2
Find the transfer function whose approximate plot is shown in figure

(i) Corner frequencies are 2.5 r/s and 40 r/s

20logK = 40+20l0g (2.5)
20logK = 47.95
K = 249.75 = 250
(ii) At ω = ωC1, 2.5 r/s, slope changes from -20 db/dec to -49db/dec. due to factor
1/1+s/2.5 = 1/(S+0.4S)
(iii) At ω = 40 r/s the slope changes from -40 db/dec to -60db/dec
Due to factor 1/1+S/40 = 1/1+0.025S
(iv) Since initial slope is -20dB/dec. it it due to the factor 1/S
(v) Open loop transfer function
G(S) = 250 / S(1+0.4S) (1+0.025S)
From the asymptotic magnitude (in dB) verses frequency (Log Scale) plot is in the figure. Find
associated function. Assume no right half plane poles and zeros present.
Transfer function = 8(1+0.5S) (1+0.25S) / S(1+0.125S) (1+0.0417S) (1+0.0278S)
Sketch the Bode plot the transfer function
G(S) = KS2/(1+0.2S) (1+0.02S). determine the value of cross over
frequency to be 5 rad/sec
K = 0.0398
M and N circle
M- circle :- it is defined as for a particular circle the value of magnitude of closed loop transfer
function is constant so these circles are known as M- Circle (or Magnitude – Circle)
N- Circle:- it is defined as for a particular circle the value of phase angle of closed loop transfer
function is constant, so that these circles are called N – Circle

C(jw)/R(jw) = M(jw) = G(jw)/1+G(jw)

G(jw) = x+jy = real and imaginary part
M(jw) = x+jy/1+x+jy

│M(jw)│ = √x2+y2 / √(1+x2)2+y2

Ĺ M(jw) = ɸ = tan-1(y/x) –tan-1y/(1+x)

Constant M Circle

M = √x2+y2 / √(1+x2)+y2

M = √(1+x2)+y2 = √x2+y2

Square on both sides

M2 (1+x)2+y2 = x2+y2
M2(x2+ 2x+ y2)= x2+y2
M2 = x2+y2-2xM2+x2M2+M2y2
M2 = (1-M2)x2 2M2x+ (1-M2)y2
By (1-M2) on both sides
(M2/1- M2) = x2-2x (M2/1- M2)+y2
Add (M2/1- M2) on both sides
(M2/1- M2)2 + (M2/1- M2) = x2+2x (M2/1- M2) + y2 + (M2/1- M2)
M4+M2(1-M2) / (1-M2)2 = [x – M2/1-M2]2 + Y2
(M2/1- M2)2 = [x – M2/1-M2]2 + Y2
(M2/1- M2)2 = (X– M2/1-M2]2 + Y2
This equation is the equation of a circle with centre at x = M2/1-M2 and y = 0
While radius = (M2/1-M2)
Sub M =1
X2(1-1)+2x+1+y2(1-1) = 0
2x+1 = 0
X = -1/2
When M=1 equation represents a straight passing through x=-1/2 and Y = 0
N circles:-
ĹN(jw) = ɸ = tan-1(y/x)-tan-1(y/1+x)
Both sides multiply by “tan”
tanɸ = tan[tan-1(y/x) – tan-1(y/1+x)]
= tan[tan-1(y/x)] – [tan(tan-1(y/1+x)] / 1+tan(tan-1(y/x) tan(tan y/1+y)
N = tanɸ = y/x – y/1+x / 1+y/x – y/1+x
N = y(1+x) – y(x) / x(1+x)/ x(1+x)+y2 / x(1+x)
N = y / x+x2+y2
N(x+x2+y2) = y
x+x2+y2 = y/N
x+x2+y2-y/N = 0
add both side by [1/4 + 1/4N2]
x2+x+y2-y/N+1/4+1/4N2 = 1/4 + 1/4N2
(x+x2+1/4) + (y2+1/4N2-y/N) = 1/4 + 1/4N2

(x+1/2)2+ (y-1/2N)2 = √(1/4+1/4N2)

Equation (4) represents family of circles for different values of N
With centre x = -1/2 and y=1/2N
And Radius R = √(1/4+1/4N2)

NICHOL’S CHART:-
It is a plot of magnitude in dB vs phase angle of open loop transfer function with “ω”
is varing parameter.
Advantages:-
1. It used to determine all closed loop domain specifications [ωb, ωr,Mp]
2. Closed loop response canbe obtained from open loop response using Nichol chart
Constructional Procedure:-
1. Convert G(S) into standard form
2. Put s=jw
3. Write the expression for │G(jw)│
4. And find │G(jw)│ in dB
5. Write the expression for ĹG(jw)
6. Find the magnitude in dB and phase angle for various values of “ω”
7. Plot the magnitude in dB vs phase angle curve in Nichol’s Chart
8. If necessary, determine frequency domain specification [Mp,ωr, ωH, ωgc, ωpc, ϒ, Kg]
9. If necessary, obtained closed loop response from open loop response.
 Measurement of Mp,ωr, ωH, ωgc, ωpc, ϒ, Kg

Using Nichol’s Chart determine FDS and also obtain closed loop response for
G(S) = 24 / S(S+2) (S+6)
(i) G(S) = standard form
G(S) = 24 / S(S/2+1)2 (S/6+1)6
(ii) Put s = jw
G(jw) = 2/jw(1+0.5jw) (1+0.17jw)
(iii) │G(jw)│= 2/ω√(0.5ω2)2+1 √(0.17ω)2+1
Magnitude in dB = 20log│G(jw)│ dB
(iv) │G(jw)│ = tan-1∞-tan-1(0.5ω)-tan-1(0.17ω)
 Kg = 12 dB
ɸgc = -147’
ϒ = 180+ ɸgc = 33’ ωb = 2.5 r/s
ωgc = 1.85 r/s Mp = 4 dB
ωpc = 3.5 r/s ωbr = 1.75 r/s
using Nichol’s Chart determine the closed loop response and FDS for
G(S) = 20/S(S+2)(S+5)
Kg = 12dB, ϒ=180+ ɸgc(-145) = 35; ωgc = 3.5 r/s; ωpc = 1.52 r/s ; ωr = 1.65 r/s; ωb=2.6 r/s and
Mp = 3 dB
For the function G(S)H(S) = 5(1+2S)/(1+4S) (1+0.25S)
Calculate gain margin and phase margin and draw the plot.
(i) Convert G(S) into standard form
G(S) H(S) = 5(1+2S)/ (1+4S) (1+0.25S)
(ii) Put s=jw
G(jw) = 5(1+2jw)/ (1+4jw) (1+0.25jw)
(iii) Corner frequencies
(1+2jw) = 1/2 = 0.5 rad/sec
(1+4jw) = 1/4 = 0.25 rad/sec
(1+0.25jw) = 1/0.25 = 4 rad/sec

ωL = 0.1 rad/sec which less than ωC1

ωL< ωC1< ωC2< ωC3< ωH
0.1 r/s < 0.25< 0.5< 4< 10 r/s
(v)Gain in dB
ω = ωL = 0.1 r/s = A = 20log(5) = 14dB
ω = ωC1 = 0.25 r/s = A = 20log(5) = 14dB
ω = ωC2 = 0.5 r/s = A = (-20log(0.5/0.25 + 14) = 8 dB
ω = ωC3 = 4 r/s = A = (0log(4/0.5 + 8) = 8 dB
ω = ωH = 10 r/s = A = (-20log(10/4 + 8) = 0 dB
(vi)Phase angle
ɸ = tan-1(2ω)-tan-1(4ω)-tan-1(0.25ω)

(vii)Gain margin (Kg)

Kg = 1/│G(jw)│ω=ωpc
Kg = ∞ because the phase plot ωpc=∞ does not touch -180’
Phase margin (ϒ) = 180+ɸgc
ωgc = 10 r/s
ɸgc = -70
ϒ = 110’
Plot the Bode diagram for the following transfer function and determine gain margin and phase
margin
G(S) = 10/S(1+0.4S) (1+0.1S)
(i) already the given G(S) in standard form and put s=jw
G(jw) = 10/jw(1+0.4jw) (1+0.1jw)
(ii) Corner frequency
(1+0.4jw) = Corner frequency = 1/0.4 = 2.5 r/s
(1+0.1jw) = Corner frequency = 1/0.1 = 10 r/s
(iii) ωL = 0.1 r/s
ωH = 50 r/s

(v)Gain in dB
ω = ωL = 0.1 r/s = A = 20log(ω/0.1) = 40 dB
ω = ωC1 = 2.5 r/s = A = 20log(ω/2.5) = 12 dB
ω = ωC2 = 10 r/s = A = [-40log(ω/2.5)+12] = -12 dB
ω = ωH = 50 r/s = A = [-60log(50/10)+(-12) = -54 dB
(vi)Phase Plot
ĹG(jw) = -90-tan-10.4ω-tan-1 0.01ω

Frequency domain specifications

 (i) ωgc = magnitude plot cut at 0 dB = 5 rad/sec
(ii) ωpc = Phase plot cut at -180’ = 5 rad/sec
ωpc = ωgc = 5 r/s
This is marginally stable system as ωpc = ωgc. We cannot get phase margin and gain margin for this
case.
The open loop transfer function of unity feedback system is given by G(S) = 1/S(S+1) (2S+1)
Sketch the polar plot and determine the gain margin and phase margin
(i) G(S) = 1/S(S+1) (2S+1)
(ii) Put s=jw
G(jw) = 1/jw(jw+1) (2jw+1)
(iii) Write the expression for │G(jw)│

│G(jw)│ = 1/ω√1+(2jw)2 √(1+ω2)

= 1/ω√1+4ω2 √(1+ω2)
(iv) Write the expression for ĹG(jw)
G(jw) = ɸ(jw) = -90-tan-1(ω)-tan-1(2ω)

Kg = 1/0.7 = 1.4286 ωgc = 0.6 r/s

ϒ= 180+ɸgc = 180’-171 = 9’ ωpc = 0.72 r/s
ϒ = +ve system is stable
Kg > 1
# Consider a unity feedback system having an open loop transfer function of “K” so that
(i) Gain margin is 20 dB
(ii) Phase margin is 30’
(1) Let K =1 G(S) = K / S(1+0.5S) (1+4S) sketch the polar plot and determine the value of
(2) Put s=jw
G(jw) = K / jw(1+0.5jw) (1+4jw)
(3) │G(jw)│ = 1 /ω√(1+0.5ω)2 √(1+(4ω)2
= 1 / ω√0.25ω2 +1√16ω2+1
(4) ĹG(jw) = -90-tan-10.5ω-tan-1 4ω
Case :-1 20log 1/GA = 20
Log 1/GA =1
1/GA = e1
1/GA = 10
GA = 0.1
K = GA/GB = 0.1/0.5 = 0.2
K = 0.2 when Kg = 20dB
Case-2:-
ϒx = 180’ + ɸgcx = 30’
ɸgcx = -150’
corresponding magnitude of the circle is 2
K = GB/GA = ½.0 = 0.50
K = 0.5 when ϒ = 30’
The unity feedback system has open loop transfer function G(S) = 20S(S+2) (S+5)
Using Nichol’s chart determine the closed loop frequency response and also determine the
frequency domain specification
(i) Convert G(S) into standard form
G(S) = 20/S(1+S/2).2 (1+S/5)5
= 2 / S(1+0.5S) (1+0.2S)
(ii) Put s= jw
G(jw) = 2 / jw(1+0.5jw) (1+0.2jw)
(iii) │G(jw)│ = 2/ω√(1+0.5)ω2√(1+(0.2ω)2
(iv) ĹG(jw) = -90-tan-1(0.5ω) – tan-1(0.2ω)

Frequency domain specification

ɸgc = -144
ωpc = 1.5 r/s
ϒ = 180 + ɸgc
= 180 -144 = 36’
ϒ = 36’
Kg = ref – (meeting point)
= 0 – (-10)
Kg = 10 dB
ωgc = 3 r/s
ωb = 2.5 r/s (from the graph)
Mr = 4 dB
ωr = 1.5 r/s
G(S) = 2 / S(1+S) (1+S/3)
Kg = 6 dB
ϒ = 22’
ωb = 1.5 r/s
 STABILITY ANALYSIS IN CONTROL SYSTEMS
1.1. AIM: Stability analysis methods have been discussed in this chapter those system are
said to be linear, time invariant (LTI), lumped and deterministic. In control
systems the term “Stability” means efficiency. In this chapter stability of LTI
systems has been analysed with two approaches. One is analytical approach and
other graphical approach. The Routh Hurwitz method is explained in analytical
approach. Root locus technic, Nyquist stability method and bode plot are used in
graphical approach. Based on above two approaches locations of roots of closed
loop transfer function, has found with open loop gain(K) is varied. For all
methods, stability of the LTI system has commented that those the roots lying in
the left half of the s-plane that system more stable. Roots are lying in the right half
of the s-plane than the system said to unstable. If the root lying in the imaginary
axis for certain values of K , the system is marginally stable. All the methods are
explained in this chapter with step by step.
1.2 Pre-test – The reader asked to take following basic MCQ before going to see Stability
Analysis.
1. What is meant by LTI ?
a. Linear Time Value
b. Linear Time Invariant ANS- (b)
2. What is meant by lumped system?
a. parameter is physically located at a point
b. Parameters values are distributed in the spatial coordinates ANS- (a)
3. S-plane is used in
a. only real numbers.
b. complex numbers ANS- (b)
4. What is meant by roots of the transfer function?
a. Value of poles
b. Value of zeros ANS- (a)
5. G(s) = 1/(s+2). The open loop transfer function G(s) is consist of
a. one pole located at -2
b. one zero located at -2 ANS- (a)
1.3 PREREQUISITE – Basic knowledge of circuit theory and electronics.
1.4 Theory Explanation
1.4.1 Introduction
 In a stable system the response or output is predictable, finite and stable for a
given finite input.
In a stable system the response or output is predictable, finite and stable for a
given input.
A system is asymptotically stable of in the absence of the input, the output tends
towards zero irrespective of initial conditions.
If a system output is stable for all variations of its parameters then the system is
called absolutely stable.
If a system output is stable for a limited range of variation of its parameters than
the system is called conditionally stable system.
1.4.2 Locations of roots and various response for unit impulse
The location of root of the closed loop transfer function with unit impulse input are shown in
the s-plane and the its time response are shown in graph.

1. M(S) = C(s)/R(s) = A/S+a in S-plane

2. M(S) = C(s)/R(s) = A/S-a . rad

3. M(S) = C(s)/R(s) = (A/S+a+jb) + (A/S+a-jb)

4. Complex conjugate roots on Right half S-plane

5. Single pair of roots on imaginary axis

6. Double pair of roots on imaginary (jw) axis

 7. Single root at the orgin

8. Double roots at the orgin

From the above plots,

Three points may be stated, regarding the stability of the system
depending on the locations of the characteristics equation.
(i) If all the roots of the characteristics equation has negative real part, then the
system is stable.
(ii) If any root of the characteristics equation has a positive real part if there is a
repeated root on the imaginary axis then the system is unstable.
(iii) If the condition (i) is satisfied except for the presence of one or more non repeated
roots on the imaginary axis, then the system is limitedly or marginally stable.
The stability of the LTI systems have checked by following
method.

1. Routh Hurwitz Criterion - Analytical method

 2. Root locus technique
3. Nyquist stability criterion - Graphical method
4. Bode plot stability
1.4.3. Routh Hurwitz Criterion (RHC) Method
In a closed loop transfer function the denominator polynomial of “S” is called
Characteristics equation.
Roots of Characteristics equation lies on the left half of S-plane. Then the system
is stable.
Let the characteristics equation
a0sn+a1sn-1+a2sn-2 ………………. +an-1 sn-n = 0
1.4.3.1 Routh Array (RA) Construction

b0 = - a0 a2
a1 a3 a1
= (a1*a2 - a0a3)/a1

b1 = - a0 a4
a1 a5 a1
= (a1*a4 - a0a5)/a1
In this way Routh Aarray (RA) is constructed up to S0 row.
Hint:-
1. Any missing terms considered as zero
 2. All elements of any row can be multiply by or divide by a positive constant to
simplify construction of Routh Array
While in construction of RA , We come across 3 cases are as follows,
1. Normal Routh Array (non-zero element in the first column of Routh Array)
2. A row of all zeros.
3. First elements or other element are not zero.
1.4.3.2 Routh Hurwitz stability Criterion:- Statement
It stated as follows “Necessary and sufficient condition for stability is that all of the
elements in the first column of Routh array be positive. If this condition is not met the
system is unstable”.
Negative sign in RA indicates that the systems is unstable. Calculate the change in
sign in RA & it corresponds to number of roots forms in the right half of the S-plane.
CASE 1:- Normal Routh Array
1. Normal one, easily construct the Routh Array
2. Sign changes are noted down
3. Number of roots lying on the right half plane and stability can be estimated.
4. If there is no change in sign in the first column of Routh Array then all the roots
lying on the left half of S-plane the system is stable.
5. If there is sign changes in first column of RA, then system is unstable and number
of roots lying on the right half of S-plane is equal to the number of sign changes,
remaining roots are lying on the left half of the S-plane.
Problem 1. Using Routh Criterion determine the stability of the system represented by the
characteristics equation S4+8S3+18S2+16S+5 = 0
4th – order system equation [ 4 roots]

In 1st the column of Routh Array, All elements are positive. There is no change in sign.
Therefore system is more stable. All four roots lies in the left half of S-plane.
CASE 2:- A row of all zeros.
This case can be analysed using following steps
1. Determine the auxiliary polynomial A(S)
2. Differentiate the A.P with respect to “S” we get dA(S)/ds
3. Row of zeros is replaced with co-efficient of dA(S)/ds equation
4. Continue the construction of array in usual manner with the new row element.
5. Number of roots on imaginary axis can be calculated from the roots of the
auxiliary polynomial A(S).
6. Remaining roots are lying on the left half of the S-plane
7. If there is no sign changes in the first column of Routh Array then all zero row
indicate roots are purely imaginary and the system is marginally stable.
Problem 2. S6+S5+8S4+6S3+20S2+8S1_16 = 0. Construct RA and determine the system
stability by using Routh Hurwitz Criterion method.

Auxiliary polynomial equation (A.E)

A.E = S4+6S2+8 = 0
Let consider s2=x
x2+6x+8 = 0
(x+4) (x+2) = 0
 x = -4, -2
S = ±√x
= ±√-2 and ±√-4
= ±j√2 & ± j√4
= ±j1.414 & ± j2
s = j𝜔 = = ±j1.414 & ± j2
∴ System oscillate with the frequency = 𝜔 = 1.414 rad/sec & 2 rad/sec
Roots of the Auxiliary equation are also roots of characteristics equation. Hence 4 roots are
lying on the imaginary axis. Remaining two roots [6-4=2] are lying on the left half of the S-
plane. The system is marginally stable.
CASE3:- first element in a row is zero.

(i) If zero is first element of a row then all elements of the next row will be infinite.
(ii) To overcome this problem let 0 is replaced by some constant let ɛ that is
0 ɛ and remaining same as in the CASE -1
(iii) Finally ɛ 0 and determine the values of the element.
Problem 3. Construct RA and determine stability of the system represented by the
characteristics equation S5+S4+2S3+2S2+3S+5 = 0

(i) There are two change in sign. That is two roots lies on the right half of
S-plane. The System is unstable.
(ii) Remaining 3 roots are lying on the left half S-plane.

Problem 4. Characteristics polynomial of a system is given by

S7+9S6+24S5+24S4+24S3+24S2+23S+1S = 0. Determine the system stability using RHC.
7th order system. ∴ systems have 7 roots.
RA construction

Calculation
Auxilary Equation = AE = S4+S2+1
dAE/dS = 4S3+2S
S3 row elements = 4 , 2 or 2 , 1
S2 = 2-1/2 ; 1-0/1
0.5 ; 1
S1 = 0.5-2/0.5 ; -1.5/5 = -3
S0 = -3-0/-3 = 1
A.E - S4+S2+1 = 0 Solution
Put S2 = x
X2+x+1 = 0
X = -1±√1-4/2 = -1±√-3/2
= -1/2±j√3/2
X = -1/2±j√3/2
S = ±√x
= ±(0.5+j0.866) and
= ±(0.5-j0.866)
Two roots lying right half of S-plane remaining(7-2) 5 roots lying on left half of S-plane

(i) System is unstable [2 – change in sign]

(ii) Two roots are lying on right half of S-plane
(iii) 4 roots lying on the imaginary axis
(iv) Rest of the one root which is lies on the left half of the S-plane

Problems 5. Determine the range of “K” for stability of unity feedback system whose open
loop transfer function is G(S) = K/S(S+1) (S+2)
Closed loop transfer function = G(S)/1+G(S) .1
= K/S(S+1) (S+2)
= 1+K/S(S+1) (S+2)
= K / [S(S+1) (S+2) + K]
Characteristics equation = closed loop transfer function denominator polynomial
[S(S+1)(S+2)+K] = 0
Expanded as S3+3S2+2S+K = 0 = Characteristics equation
System is stable (no change in sign). Hence choose K minimum value from S0 row should
+ve
From S0 row --- for system stable; K>0;
From S1 row , Maximum value of K --- for system stable (6-K)/3 > 0
6-K>0 ; value of K --- less than 6
The range of K for the system to be stable 0< K < 6

1.4.5 Root Loci / Locus technic

Root locus of a system is a plot of the roots of the system characteristics equation that is poles
of closed loop transfer function as some parameters of the system is varies.
Path taken by a root of the characteristics equation when open loop gain K is varied from 0
to ∞ is called root locus.
1.4.5.1 Guidelines for Sketching Root Loci / Locus:-
Following steps are essential to construct root loci
1. Number of root loci (N) and locate the poles and zeros into S-plane
If N = number of root loci
N = P if (P>Z)

N = Z if (Z>P)
Let P= number of poles &
Z = number of zeros
2. Starting and terminal points of Root loci
a. Starting point :- As K increases from to 0 to ∞ each root locus originate
from an open poles with gain K = 0
b.Terminating point :- either on an open loop zero or on infinity ∞ with K = ∞
Ends with open loop zeros = N-(P-Z)
End with infinity = (P-Z)
3. Asymptotes to root Loci:-
 (a) Centroid :- (P-Z) root loci which tend to infinity do so along straight line
asymptotes radiating out from a single point S=--𝜎A on the real axis called
Centroid
-𝜎A = [∑ 𝑣𝑎𝑙𝑢𝑒 𝑜𝑓 𝑟𝑒𝑎𝑙 𝑝𝑎𝑟𝑡 𝑜𝑓 𝑜𝑝𝑒𝑛 𝑙𝑜𝑜𝑝 𝑝𝑜𝑙𝑒𝑠–
∑ 𝑣𝑎𝑙𝑢𝑒 𝑜𝑓 𝑟𝑒𝑎𝑙 𝑝𝑎𝑟𝑡 𝑜𝑓 𝑜𝑝𝑒𝑛 𝑙𝑜𝑜𝑝 𝑧𝑒𝑟𝑜𝑠] / (P-Z)
(b) Asymptotes have angle
ɸA = (2q+1)/(P-Z) 180’ q=0,1 …. (P-Z-1)
4. On locus segments of the real axis
Travelling direction of negative real axis. Put test point on the (negative) real axis.
A test point of the real axis lies on the locus if the number of open loop poles plus
zeros on the real axis to the right of this point is odd.
If even the test point may not lie on the root locus point
5. On locus points of the imaginary axis
The inter sections (if any) of root loci with imaginary axis can be determined by use
of the Routh Hurwitz Criterion.
6. Angle of departure from complex poles:-
Angle of departure, ɸP of a locus from a complex open loop poles is given by
ɸP = 180+ɸ
Where ɸ is the net angle contribution at this poles of the other open loop poles and
zeros

7. Angle of arrival from complex zeros.

Angle of arrival ɸz of a locus at complex zero is given by
ɸz = 180-ɸ
 ɸ = ∑(angle made by zeros) – ∑(angle made by poles)

ɸ = Ɵ3-(Ɵ1+Ɵ2)
8. Points at which multiple (poles) roots of characteristic equation locus break away
point are the solution of dK/dS = 0
9. Break away / Break in points
Break away or Break in points either lie on real axis or exist as complex conjugate. If
there is a root locus on real axis, meting point is b/w 2 poles then there exist Break
away point.
If the root locus meets b/w two zeros then exist Break in point. Root locus on real
axis b/w poles and zeros then there may be or may not be Break away or Break in
point.
 Characteristics equation = 0
 S2+5S+K = 0
 K = - [S2+5S]
 dK/dS = 0
 find the solution of dK/dS = 0
 find value of “S” value and substitute in the equation of K
 for this values of root (gain K should be +ve and real) that root value
may be Break away or Break in point
10. The gain “K” at any point “S0” on a root locus is given by
K = Product of phasor lengths (real to scale) from “S0” to poles of F(S)
Product of phasor lengths from “S0” to zeros of F(S)
Based on the above steps help us to plot loci’ of the root. Depends upon the char.
Equation 1+G(s)H(s)=0, we will choose which steps are required.
Problem 6. Sketch the root locus whose characteristics equation is given by
𝐾
1 + 𝑆(𝑆+1) (𝑆+2) = 0

From the above equation is Characteristics equation =1+G(s)H(s)=1+K/S(S+1) (S+2)= 0

𝐾
From this G(S) = S(S+1) (S+2)

(i) Find number of root loci

 N = P if (P>Z) no of poles = 3, No of zeros = 0 N = 3
(ii) Find the values of poles & zeros
P1 --- S=0
P2 --- (S+1) =0 ∴ S = -1
P3 --- (S+2) =0 ∴ S = -2 locate the poles values on the S-plane(linear graph
sheet)
(iii) Starting & terminating point of the Root loci
a. Root locus starts at K = 0 at each poles and
b. Terminates at (i) ∞ = (P-Z) = 3 – 0 = 3
(ii) Open loop zeros = N – (P-Z) = = 3 -3 = 0
(iv) Asymptotes to root locus:-
a. Centroid = -𝝈A = -3-0/3-0 = -1
-𝜎A = -1
b. Angle ɸA = (2q+1)180/P-Z
q = 0,1 …. (P-Z-1)
q = (3-0-1)= 2
i.e three values of q = 0,1 & 2
q=0 => ɸ1 = 60°
q=1 => ɸ2 = 180°
q=2 => ɸ3 = 300°

(v) Travelling direction of the –ve real axis- Put a test point (TP) on the real axis

(vi) Intersecting point on the real axis – Break away or Break in points
C.E = 1+ K / S(S+1)(S+2) = 0
S(S+1)(S+2) + K = 0
S3+2S2+S2+2S+K = 0
S3+3S2 +2S+K = 0
K = -[S3+3S2 +2S] ------- 1
 dK/dS = -[3S2+6S+2]
3S2+6S+2 = [solution and roots]
= -b ± √b2-4ac/2a
= -b ± √36-(4.3.2) / 6
S = -6 ± √12 / 6
= -0.426 and -1.577
S = -0.426 ---- sub. In K equation (1) => K = 0.3849 --- valid
S = -1.577 ---- sub. In K equation (1) => K = -0.3849 --- invalid
(vii) Crossing point on the imaginary axis (j𝜔 𝑎𝑥𝑖𝑠)
C.E = 1+ K / S(S+1)(S+2) = 0
S(S+1)(S+2) + K = 0
S3+2S2+S2+2S+K = 0

K < 6 for stable

(i) 0 < K < 6 for stable
(ii) K = 6 --- Marginally stable
(iii) K > 6 --- Unstable
When K = 6 all the row elements must be zero
A.E = 3S2+K = 0 S3 1 2
S2 3 K
2
3S +6 = 0 S1 0 0
S2+2 = 0 S0 K
S2 = -2
S = ± √-2
S = ± j 1.414 ∴ freq. of oscillation = 1.414 rad/se
1.4.6. Nyquist Stability

𝐺(𝑠)
Closed loop transfer function = 1+𝐺(𝑠)𝐻(𝑠)

Where 𝐺(𝑠)𝐻(𝑠) = 𝑂𝑝𝑒𝑛 𝑙𝑜𝑜𝑝 𝑇𝑟𝑎𝑛𝑠𝑓𝑒𝑟 𝐹𝑢𝑛𝑐𝑡𝑖𝑜𝑛

Let P – number of poles of G(S) H(S) lying in the right half of S-Plane
Z – number of zeros of 1+G(S) H(S) lying in the right half of S-plane
N – number of encirclement of critical point (-1+jo) made by Nyquist
plot G(S) H(S)
1.4.6.1 STEP BY STEP PROCEDURE:-
1. Find P
2. Put s = j𝜔
3. Write formula for │G(j𝜔 H(j𝜔)│
4. Write formula for angle of G(j𝜔 H(j𝜔)
5. Draw the Nyquist plot with help of
(i) ω=0
(ii) ω=∞
(iii) intersection point with negative real axis

8. draw vertical line passing through critical point (-1+j0)

9. find the number of cutting points of Nyquist with critical line
 N = ± (no. of cutting point) / 2
= ± 4/2 = +2
+  ω is in clockwise
-  ω is in Anti clockwise

(iv) Fine Z = N+P

For the system to be stable ; Z = 0 if Z = 0, then the system is unstable. Now Z –
indicates number of roots lying in the right half of S – plane
Nyquist Stability Criterion:- Statement
It states that if the Nyquist plot open loop transfer function corresponding to Nyquist
column in the S-plane encircles the critical point (-1+jo) in the anti clockwise direction as
many times as the number of right half S-plane poles of G(S) H(S) closed loop system is
stable.
i.e Z = N + P
Z = 0 for N & P are equal in magnitude and opposite sign
In commonly occurring case of G(S) H(S) with number of poles in right half S-plane the
closed loop system is stable if the Nyquist plot of G(S) H(S) does not encircle the critical
point (-1+j0)
That is Z = N + P
Z = 0 for N = P = 0
Phase angle criterion is -180 = ĹG(jw) H(jw)
Magnitude criterion is 1 = │G(jw) H(jw)│
𝟐𝟎𝟎
Problem 7. G(S) H(S) = 𝑺 (𝑺+𝟓) (𝑺+𝟏𝟎) Using Nyquist stability criterion, find the stability of
the system
1. Find values of poles .P = 0, -5, -10
2. Put s = jω
G(jω) H(jω) = 200 / (ω) (jω +5) (jw+10)
3. │ G(jω) H(jω)│ = 200/ω √ω2+25 √ω2+100
4. Phase angle of G(jω) H(jω) = -90-tan-1ω/5 – tan-1ω/10
 ∠GH(jw) = -90-tan-1 ω/5 – tan-1 ω/10
-180+90 = - [tan-1 ω/5 + tan-1 ω/10]
90 = tan-1 ω/5 + tan-1 ω/10
Multiply by “tan” on both sides
tan(90) = tan [tan-1 ω/5 + tan-1 ω/10]
∞ = (tan tan-1 ω/5 + tan tan-1 ω/10) /(1- tan-1 tan-1 ω/5 + tan tan-1 ω/10)
ω ω
anything +
5 10
∞= = ω ω
0 1−( ∗ )
5 10

ω ω
Denominator = 0 =1−( ∗ )
5 10

ω2/50 = 1
ω2 = 50
ω = ± √50
ω = 7.07 r/s
6. Here ; P = 0
N = 0; Z = N+P = 0+0 = 0
Nyquist plot does not encircle the critical point (-1+j0) hence N = 0, hence this system
is stable.

1.4.7. Bode Plot Stability

Stability analysis can be done with bode plot. Draw the bode plot with magnitude &
phase angle plot. Determine gain margin (Kg) phase margin.
If both gain and phase margin values are positive than the system is more stable.
Otherwise the system is unstable (even though any one is negative)

Problem 8. Sketch the Bode Plot and determine the system stability whose the open loop
𝟏𝟐𝟓𝟎
transfer function given by G(S) =
𝑺(𝟓+𝑺) (𝟓𝟎+𝑺)

(a) Magnitude Plot

1. Find G(S) is into standard form  G(S) = 5 / S (1+0. S) (1+0.02 S)
2. Put s=jω G(jω) = 5 / jω (1+0.2jω) (1+0.02jω)
3. Find corner frequency kept 5 & 1/S
1/(1+0.2jω) = C.F = value of real part / Imaginary Part =
1/0.2 = 5 r/s
1/1+0.02jω = C.F = 1/0.02 = 50 r/s
4. Arrange the Corner frequency in ascending order
ωH < ωC2
 ωL = 5 r/s and ωC2 = 50 r/d
5. Select ωL & ωH (it should be in log Scale)
ωL < ωC1 ie 1<5 = ωL = 1 r/s
ωH > ωC2 ie 100 >50 = ωH = 100 r/s

7. Magnitude in dB (2nd Case)

ωL = A = 20Log(5/1) = 13.979 dB
ωC1 = 5 r/s ; A = 20Log 5/5 = 0
ωC2 = 50 r/s ; A = [-40 log (5/5) +0 ] = -40dB
ωH = 100 r/s ; A = [-60 log 100/50 + (-40)] = -58.06 dB
8.write the expression for ∠G(jw)
∠G(jw) = -90-tan-1(0.2ω)-tan-1(0.02ω)
Ω 𝜔(rad/sec) Ɵ │ G(jω)│ in dB
ωL = 1 -102.45 13.9
ωC1= 5 -164.7 0
ωC2 = 50 -219.3 -40
ωH = 100 -240.57 -58
Gain margin = ref – meeting point ( from the graph meeting point @-18dB)
= 0 – (-18)
= 18 dB.
The gain is positive (Kg=18dB) than the given transfer function is more stable.

1.5 Simulation Results Using MATLAB

a. Root locus
1. Sketch the root locus whose characteristics equation is given by
𝑲
𝟏 + 𝑺(𝑺+𝟏) (𝑺+𝟐) = 0

(Same problem have been solved in the theory.)

From the above equation is Characteristics equation =1+G(s)H(s)=1+K/S(S+1) (S+2)= 0
𝐾 𝐾
From this G(S) = S(S+1) (S+2) = 𝑆3 +3𝑆2 +2𝑆

While solving Root locus in MATLAB take K=1

𝟏
G(S) = 𝑺𝟑 +𝟑𝑺𝟐 +𝟐𝑺
Numerator = only constant = 1
Denominator = 𝑆 3 + 3𝑆 2 + 2𝑆 = 0
𝑆 3 𝑐𝑜𝑒𝑓𝑓𝑖𝑐𝑒𝑛𝑒𝑛𝑡 = 1
𝑆 2 𝑐𝑜𝑒𝑓𝑓𝑖𝑐𝑒𝑛𝑒𝑛𝑡 = 3
𝑆 𝑐𝑜𝑒𝑓𝑓𝑖𝑐𝑒𝑛𝑒𝑛𝑡 = 2
Constant value = 0

MATLAB Program
num=[1]
den=[1 3 2 0]
g=tf(num,den)
rlocus(g)
grid (% it is optional)
MATLAP program screen shot

OUTPUT screen shoot

b. Bode plot stability

Problem 8. Sketch the Bode Plot and determine the system stability whose the open loop
𝟏𝟐𝟓𝟎
transfer function given by G(S) = 𝑺(𝟓+𝑺) (𝟓𝟎+𝑺)

𝟏𝟐𝟓𝟎
For writing MATLAB program convert G(s) = 𝑺(𝑺𝟐 +𝟓𝟓𝑺+𝟐𝟓𝟎)
𝟏𝟐𝟓𝟎
= (𝑺𝟑 +𝟓𝟓𝑺𝟐 +𝟐𝟓𝟎𝑺)

𝟏𝟐𝟓𝟎
G(S) = 𝑺𝟑 +𝟓𝟓𝑺𝟐 +𝟐𝟓𝟎𝑺
Numerator = only constant = 1250
Denominator = 𝑺𝟑 + 𝟓𝟓𝑺𝟐 + 𝟐𝟓𝟎𝑺 + 𝟎 = 0
𝑆 3 𝑐𝑜𝑒𝑓𝑓𝑖𝑐𝑒𝑛𝑒𝑛𝑡 = 1
𝑆 2 𝑐𝑜𝑒𝑓𝑓𝑖𝑐𝑒𝑛𝑒𝑛𝑡 = 55
𝑆 𝑐𝑜𝑒𝑓𝑓𝑖𝑐𝑒𝑛𝑒𝑛𝑡 = 250
Constant value = 0

MATLAB Program

num=[1250]
den=[1 55 250 0]
g=tf(num,den)
bode(g)
grid (% it is optional)
MATLAP program screen shot

OUTPUT screen shoot

 From the MATLAB output graph Kg= Gain Matgin = ref- meeting poing
= 0 – (-21)
= 21dB
Kg is positive(21dB) , Therefore the given system is more stable.

1.6 POST-TEST MCQ

1. Closed loop ples are
a) Zeros of 1+G(S). H(s)
b) Zeros of G(s) H(s)
c) Poles of G(s) H(s)
d) Poles of 1 + G(s) H(s) Ans: (a)

2. If poles are more than zeros in G(S) F(S), then number of root locus segment is equal to

a) Number of poles
b) Number of zeros
c) Sum of poles and zeros
d) Difference of poles and zeros Ans: (a)

3. If poles of system are lying on the imaginary axis in s-plane, the system will be
a) Unstable
b) marginally stable
c) Conditionally stable
d) Unstable Ans: (b)

4. The order of the auxiliary polynomial is always:

a) Even
b) Odd
c) May be even or odd
d) None of the mentioned Ans: (a)

5. Nyquist stability criterion requires polar plot of

a) Characteristic equation
b) Closed loop transfer function
c) Open loop transfer function
d) None of these Ans: (c)
1.7 CONCLUSION

Stability analysis for LTI systems are discussed with analytical and graphical methods.
For each method more problems have been solved theoretically in step by step and the same
problems are executed in MATLAB, the corresponding MATLAB results are also presented in
this chapter.

1.8. REFERENCES

1. M.Gopal,”Control system – Principle and Design,” Tata McGraw Hill, 6th

edition, 2012.
2. Benjamin C.Kuo, “Automatic control systems,” PHI, 9th Edition 2014.
3. Nagrath I.J and Gopal M, “Control Systems Engineering”, New Age
Publishers, 6th Edition, 2018.
4. K.Ogata,”Modern control Engineering, “fifth edition, PHI, 2012.

1.9 ASSIGNMENT PROBLEMS

(i) Problems in Routh Hurwitz criterion

1. S6+S5+3S4+3S3+3S2+2S+1 = 0 . Check the system stability using RHC
2. S7+5S6+9S5+9S4+4S3+20S2+36S+36 = 0 Check the system stability using RHC
3. open loop transfer function of unity feedback control system is given by
G(S) = K /S(S+2) (S+4) (S2+6S+25)
By using RHC discuss the stability of closed loop system as a function of K.
Determine the value of “K” which will cause sustained oscillation in closed
loop system what are the corresponding oscillating frequency
(ii) Problems in Root Locus
1. The unity feedback control system has an open loop transfer function
G(S) = K/S(S2+4S+13). Sketch the root locus.
2. Sketch the root locus of system whose OC Transfer function is G(S) = K /
S(S+2)(S+4). Find the value “K” so that the damping ratio of closed loop
system is 0.5
(iii) Problems in Nyquist stability criterion
1. Check the stability of the given transfer function using Nyquist whose open
loop transfer function is G(s)H(s) = 120/[(s-2)(s+6)(s+8)]
2. Check the stability of the given transfer function using Nyquist whose open
loop transfer function is G(s)H(s) = 100/[(s-2)(s+6)(s+8)]
3. Check the stability of the given transfer function using Nyquist whose open
loop transfer function is G(s)H(s) = 50/[(s-2)(s+6)(s+8)]
(iv) Problems in Bode plot stability
1. Draw bode plot & check the system stability whose transfer function is
G(s) = 100*103(s+1)/ (s+10)(s+1000)
2. Draw bode plot & check the system stability whose transfer function is
G(s) = 200 (s+20)/s(2s+1)(s+40)
 UNIT –IV-B
COMPENSATOR DESIGN
Performance Criteria – lag, lead and lag-lead networks – compensator design using
bode plot.
Introduction to design :- The Control systems are designed to perform specific tasks.
The requirements of control system are usually specified as performance specification. The
specifications are generally related to accuracy, relative stability and speed of response.
First step in design is the adjustment of gain to meet the desired specifications. In
practical systems, adjustment of gain alone will not be sufficiently to meet the given
specification. In many cases, increasing the gain may result in poor stability or instability. In
such cases, it is necessary to introduce additional devices or components in the system to
alter the behaviour to meet the desired specifications. Such a redesign or addition of a
suitable device is called compensation. A device inserted into the system for the purpose of
satisfying the specifications is called compensator. The compensator basically introduce
pole and or zero in open loop transfer function to modify the performance of the system.
When a set of specifications are given for a system, then a suitable compensator
should be designed so that the overall system will meet the given specification.
The compensation schemes used for feedback control system is either series
compensation or parallel (feedback) compensation.

3.Compenents available parallel compensation

4. Designer’s experience
5. Economic consideration and so on.
The compensator maybe Electrical, Mechanical, Hydraulic, Pneumatic or other lyper of device or
network. Usally an electric network or electronic device serves as compensator in many control
systems. The different types of electrical or electronic compensator and are
 Lag Compensator
Lead Compensator
Lag – Lead compensator
In control systems, compensator is required in the following situations.
1. When the system is absolutely unstable, then compensation is required to stabilized the
system and also to meet the desired performance.
2. When the system is stable, compensation is provided to obtain the desired performance.
LAG COMPENSATOR:-
A compensator having the characteristics of a Lag network is called lag compensator. If a
sinusoidal signal is applied to a lag network, then in steady state the output will have a phase lag
with respect to input.
Lag compensation results in a large improvement in steady state performance but results in
slower response due to reduced Bandwidth. The alternation due to the lag compensator will shift
the gain cross over frequency to a low frequency point where phase margin is acceptable. Thus the
lag compensator will reduce the bandwidth of the system and will result in slower transient
response.
S-Plane Representation of Lag Compensator
The lag compensator has a pole at S = 1/ßT and a zero at S = -1/T. the pole – zero plot of lag
compensator is shown in figure.

Here ß > 1, so the zero is located to the left of the pole on the negative real axis. The general
form of lag compensator transfer function is given by equation (i)
G(S) = S + ZC / S+PC = S+1/T / S+1/ßt -------------- (1)
Where T > 0 and ß > 1
Zero of lag compensator Zc = 1/T ------------------ (2)
Poles of lag compensator, PC = 1/ßT = 1/ß Zc -----------(3)
From the equation (2) we get = T = 1/Zc ----- (4)
From the equation (3) = ß = Zc/Pc ----------(5)
Realisation of lag compensator using Electrical network
Lag compensator can be realised by the R-C network shown in figure.

Let Ei(S) = input voltage

Eo(S) = output voltage
Apply voltage division sure to the circuit
Eo(S) = Ei(S) (R2}+1/Sc)/R1+R2+1/Sc = Ei(S) (SCR2+1)/Sc /Sc(R1+R2)+1]/Sc
Ei(S) = (ScR2+1) / Sc(R1+R2)+1
Transfer function of the electrical network is the ratio of output voltage to input voltage
Eo(S)/Ei(S) = CR2(S+1 (R2))/C(R1+R2) [S+1/C (R1+R2)]
= (S + 1/R2C) / [R1+R2/R2] [S + 1 /[(R1+R2)1/R2] R2C2 ----- (6)
But the transfer function of lag compensator is given by
G(S) = (S+1/T) / (S+1/ßT) ----------- (7)
Comparing equation (6) & (7)
Eo(S) / Ei(S) = 1/ß (S+1/T)/(S+1/ßT)
Where T = R2C and ß = [(R1+R2)/R2] R2C ---------- (8)
The transfer function of RC network is given by equation (8)
Frequency response of lag compensator
Consider the general form of lag compensator
G(S) = (S+1/T) / (S+1/ßt) = (ST+1)/T / (SßT+1)/ßT
= ƥ(1+ST) / (1+SßT) -------- (9)
The sinusoidal transfer function of lag compensator is obtained by letting S = jw in the equation (9)
G(jw) = ß(1+jwT)/(1+jwßT) ---------------- (10)
 When ω=0, GC(jw) = ß ------------------(11)
From the equation (11) we can say that the lag compensator provides a dc gain of ß (here ß>1). If
the dc gain of the compensator is not desirable than it can be eliminated by a suitable attenuation.
Let us assume that the gain ß is eliminated by a suitable attenuation network. Now GC(jw) is given
by

GC(jw) = 1+jwT/1+jwßT = √1+(ωT)2 Ĺtan-1ωT / √1+(ωßT)2 Ĺtan-1ωßT -------- (12)

The sinusoidal transfer function in equation (12) has two corner frequencies and they are dnoteed
as ωC1 and ωC2. Here ωC1 = 1/ßT and ωC2 = 1/T
Since, ßT > T , ωC1 < ωC2

Let A = │GC(jw)│ in dB = 20log √1+(ωT)2/√1+(ωßT)2 ---------------(13)

At very low frequencies ie upto ωC1, ωT << 1 and ωßT << 1

A ⴧ 20log1 = 0
In the frequency range from ωC1 to ωC2
ωT << 1 and ωßT >> 1
at very high frequencies that is after ωC2, ωT >> 1 and ωßT >> 1

A ⴧ 20log √(ωT)2 / √(ωßT)2 = 20log 1/ß

The approximate magnitude plot of lag compensator is shown in figure. The magnitude plot
of Bode plot of GC(jw) is a straight line through 0 dB upto ωC1, then it has a slope of
-20 db/dec upto ωC2 and after ωC2 it is a straight line with a constant gain of 20log(1/ß)
Let ɸ = ĹGC(jw)
ɸ = tan-1ωT – tan-1ωßT
as ω = 0, ɸ - 0
as ω = ∞, ɸ - 0

1/ßT 1/T√ß 1/T

As ω is varied from 0 to ∞, the phase angle decreases from to 0 to a negative maximum
values of ɸm at ω=ωm, then increases from this maximum value to 0. The phase plot of lag
compensator is shown in figure above. It can be shown that the frequency at which
maximum phase lag occurs is the geometric mean of the two corner frequencies.
Frequency of maximum phase lag ωm = √ωG.ωC2
= √1/ßT 1/T = 1/T√ß
From the bode plot of lag compensator, we observe that lag compensator has dc gain unity
it offers and a high frequency gain of 1/ß 1[in dB, it is 20log 1/ß]. It means that the high
frequency noise is attenuated in passing through network and so the signal to noise ratio is
improved. A typical choice of ß = ω.
Determination of ωm and ɸm:-
The frequency ωm can be determine by differentiating ɸ with respect to ω and
equating dɸ/dω to zero as shown below.
From the equation (12) we get
Phase of GC(jw) = ɸ = ĹG(jw) = tan-1ωT – tan-1ωßT -------- (14)
Differentiating with respect to “ω”
dɸ/dω = 1/1+(ωT)2 T – 1/1+(ωßT)2 ßT ---------- (15)
ω = ωm, dɸ/dω = 0 d/dƟ tan-1Ɵ = 1/1+Ɵ
hence replace ω by ωm in equation (15) and equate to zero
1/1(ωmT)2 T– 1/1+(ωßT)2 ßT = 0
T / 1+( ωmT)2 = ßT/1+(ωßT)2
On cross multiplication we get
1+(ωßT)2 = ß [1+(ωmT)2]
(ωßT)2 - ß(ωmT)2 = ß-1
ß(ωmT)2 (ß-1) = (ß-1)
ωm2 = 1/T2ß ωm = 1/T√ß ---------(16)
the maximum phase angle ɸm can be calculated from the knowledge of ß and vice versa.
The relations b/w ɸm and ß are derived below.
From the equation (14) we get,
GC(jw) = ɸ = tan-1ωT – tan-1ωßT on taking “tan” on both sides. We get
tanɸ = tan[tan-1ωT- tan-1ωßT]
 = tan[tan-1ωT]- tan[tan-1ωßT]/ 1+ tan[tan-1ωT] tan[tan-1ωßT]
= ωT- ωßT/1+ ω2T2ß = ωT(1-ß) / 1+ß(ωT)2
At ω= ωm, ɸ = ɸm , tanɸm = ωmT(1-ß)/1+ß(ωmT)2
From equation (16) ωm=1/T√ß
tanɸm = 1/Tß T(1-ß) / 1+ß 1/T2ß T2 = (1-ß) √ß/1+1
tanɸm = (1-ß)/2√ß
maximum values of lag angle
ɸm = tan-1[1-ß/2ß] ---------------- (17)
from the equation (17) it is evident that (1-ß) and 2√ß are the two side of right angle
triangle. Hene construct a right angle triangle as shown in figure with reference diagram
sinɸm = (1-ß) /(1+ß)
sinɸm(1+ß) = (1-ß)
sinɸm + ß sinɸm = (1-ß)
ß sinɸm + ß = 1 - sinɸm
ß(1+sinɸm) = 1 - sinɸm
ß = [1 - sinɸm + 1 + sinɸm ------------------- (18)
Procedure for the design of LAG COMPENSATOR using BODE PLOT
The following steps may be followed to design a lag compensator using bode plot and
to be connected in series with transfer function of uncompensated system, G(S)
1. Choose the value of “K” in uncompensated system to meet the steady state error
requirement.
2. Sketch the bode plot for uncompensated system.
3. Determine the phase margin of the uncompensated system form the bode plot. If the
phase margin does not satisfy the requirement then lag compensation is required.
4. Choose a suitable value for the phase margin of the compensated syste,.
Let ϒd = desired phase margin as given in specifications.
ϒn = phase margin of compensated system.
Now ϒn = ϒd+Σ
Where Σ = Additional phase lag to compensate for shift in gain
Choose an initial value of Σ = 5’
5. Determine the new gain cross over frequency ωgcn. The new ωgcn is the frequency
corresponding to a phase margin of ϒn on the bode plot of uncompensated system
Let ɸgcn = phase of G(jw) at new gain crossover frequency ωgcn
ϒn = 180+ɸgcn (or) ɸgcn = ϒn-180’
The new gain crossover frequency ωgcn is given by the frequency at which the phase of
G(jw) is ɸgcn.
6. Determine the parameter, ß of the compensate the value of ß is given by the
magnitude of G(jw) at new gain crossover frequency ωgcn. Find the dB dB gain (Agcn)
at new gain crossover frequency ωgcn
Now, Agcn = 20log ß or Agcn/20 logß ß=10Agcn/ß
7. Determine the transfer function of lag compensator
Place the zero of the compensator arbitraity at 1/10th of the new gain
crossover frequency ωgcn.
Zero of the lag compensator ZC = 1/T = ωgcn/ω
T = 10/ωgcn
Pole of the lag compensator, PC = 1/ßT
Transfer function of lag compensator G(CS)= (S+1/T)/(S+1/ßT) = ß(S+1)/(1+SßT)
8. Determine the open loop transfer function of compensated system. The lag
compensator is connected in series with the plant as shown in figure.

When lag compensator is inserted in series with the plant, the open loop gain of the
system is amplified by the factor ß (ß>1). If the gain produced is not required then
attenuator with gain 1/ß can be introduced in series with the lag compensator to
fullify the gain produced by the lag compensator
9. Determine the actual phase margin of compensated system. Calculate the actual
phase angle of the compensate system using the compensated transfer function of
new gain crossover frequency ωgcn.
Let ɸgcn = phase of Go(jw) at ω = ωgcn.
Actual phase margin of the compensated system
ϒ0 = 180+ɸgcn
If the actual phase margin satisfies the given specification then the design is
accepted. Otherwise repeat the procedure from step-4 to 9 by taking Σ as 5’ more
than previous design.
# A unity feedback system has an open loop transfer function, G(S) = K/S(4+2S) (S+80).
Design a suitable lag compensator so that phase margin is 40’ and the steady state error for
ramp input is less than or equal to 0.2
1. Calculate the gain K
Given that, ess < 0.2 for ramp input,
Let ess = 0.2
 Velocity error constant ; Kv = 1/ess = 1/0.2 = 5
by definition of velocity error constant Kv = £t S G(S)
s 0
since the system is UFB H(S) = 1
Kv = £t S G(S) = £t S K/S(1+2S) K=5
s 0 s 0
2. Bode plot of compensated system
G(S) = 5/S(1+2S)
Put s=jw G(jw) = 5/jw(1+2jw)
Corner frequencies CF = ½ = 0.5 r/s
ωL = 0.1 r/s
ωH = 10 r/s
term Corner frequency slope change in slope frequency
5/jw - -20
1/1+2jw 0.5 -20 -20+(-20) = -40

# The open loop transfer function of certain unity feedback control system is given by
G(S) = K/S(S+4) (S+80). It is desired to have the phase margin to be at least 33’ and
the velocity error constant Kv = 30sec-1. Design a phase lag series compensator.
1. Calculation of gain K
Kv = 30sec-1
Kv = £t S G(S) H(S)
s 0
30 = £t S K/S(S+4) (S+80)
s 0
K = 30 x 80 x 4 = 9600
2. Bode plot for uncompensated system
(i) G(S) = 9600 / S(S+4) (S+80)
(ii) Convert G(S) into standard form 30 / S(1+0.25S) (1+0.0125S)
Put s=jw G(jw) = 30/jw(1+0.25jw) (1+j0.0125w)
(iii) Find Corner frequencies
(1+0.25jw) = 1/0.25 = 4 r/s
(1+0.0125jw) = 1/0.0125 = 80 r/s
(iv) ωL and ωH --- ωL = 1 r/s and ωH = 100 r/s
(v) term Corner Frequency slope Change in Slope
30/jw - -20
1/1+0.25jw 4 -20 -40
1/1+0.0125jw 80 -20 -60
(vi) Gain in dB
ω = ωL = 1 r/s = A = 20log30/1 = 30 dB
ω = ωC1 = 4 r/s = A = 20log30/4 = 18 dB
 ω = ωC2 = 80 r/s = A = [-40log(80/4) + 18] = -34 dB
ω = ωH = 100 r/s = A = [-60log(100/80) - 34] = -40 dB
(vii) ɸ = ĹG(jw) = -90-tan-10.25ω-tan-10.0125ω

3. determination of phase margin of uncompensated system.

ɸgc – phase of G(jw) at gain crossover frequency (ωgc)
ϒ – phase margin of uncompensated system
From the graph of uncompensated system
We get ɸgc = -168
ϒ = 180+ ɸgc = 12
The system required a phase marging of atleast 33’, but the a variable phase margin is 12’
and so lag compensation should be employed to improve the phase margin
4. choose suitable value for the phase margin of compensated system.
The desired phase margin ϒd = 33’
Phase margin of compensated system.
ϒn = ϒd+Σ
Σ = 5’ (initial) ϒn = 33+5 = 38’
5. Determine new gain crossover frequency
ωgcn = new gain crossover frequency
ɸgcn = phase of G(jw) at ωgcn
ϒn = 180+ ɸgcn = ɸgcn = ϒn -180 = 38’-180’ = -142’
From the plot we get that, the frequency corresponding to a phase of =142’ is 4.7
New gain crossover frequency ωgcn = 4/7 r/s
6. Determine the parameter, ß
From plot, we found that, the dB magnitude at ωgcn as 16dB
Gain = G(jω) in dB at ∞ = ωgcn = Agcn = 16dB
Also Agcn = 20logß = ß = 10 16/20 = 6.3
7. Determine the transfer function of lage compensator .
The zero of the compensator is placed at a frequency 1/10 of ωgcn
Zc = 1/T = ωgcn/10
Now T = 10/ ωgcn = 10/4.7 = 2.13
Pc = 1/ßT = 1/6.3 x2.13 = 1/13.419
Transfer function of lag compensator = G(S) = (S+1/T)/(S+ßT)
= ß (1+st)/(1+SßT)
= 6.3 (1+2.13S)/(1+13.419S)
 8. Determine the open loop transfer function of compensated system.
[ The gain of the compensator in nullified by introducing an attenuator in series
with the compensator shown in the figure]

open loop transfer function of compensated system.

G0(S) = 1/6.3 x 6.3(1+2.13S)/(1+13.419S) x 30 / S(1+0.25S) (1+0.0125S)
= 30 / (1+2.13S) / S(1+13.419S) (1+0.25S) (1+0.0125S)
9. Determine the actual phase margin of the compensated system.
On substituting S = jω in G0(S)
G0(jω) = 30 (1+j2.13ω) / (jω) (1+13.419jω) (1+0.25jω) (1+0.0125jω)
ɸ0 = phase of G0(jω) and ɸgco = phase of G0(jω) at ω=ωgcn
ɸ0 = tan-13.12ω – 90 – tan-113.419ω - tan-10.25ω – tan-10.075ω
at ω = ωgcmn = 4.7 r/s
ɸ0 = -147’
To make an unstable as a stable system
Actual phase margin of compensated system = ϒ0 = 180+ɸcn
= 180 -147 = 33’
Conclusion :- the actual phase margin of the compensated system satisfies the requirement. Hence
the design is acceptable.
Result:-
(i) Transfer function of lag compensator Gc(S)
Gc(S) = 6.3(1+2.13S)/(1+13.419S)
(ii) Transfer function of lag compensated system
G0(S) = 30 / (1+2.13S) / S(1+13.419S) (1+0.25S) (1+0.0125S)
LEAD COMPENSATOR :-
A compensator having the characteristics of a lead network is called a lead compensator. If
a sinusoidal signal is applied to the lead network, then in steady state the output will have a phase
lead with respect to the input.
A lead compensation increases the bandwidth, which improves the speed of response and
also reduces the amount of over shoot. Lead compensation appreciable improves the transient
response, where as there is a small change in steady state accuracy. Generally, then lead
compensation is provided to make an unstable as a stable system.
S-plane representation of lead compensator. Lead compensator has a zero
At S = 1/-T and a pole at
S=1/-T. the pole zero plot of lead compensator is shown in figure.

Here ά < 1 so the zero is closer to the origin than the pole. The general form of lead compensator
transfer function is given by the equation (1)
Gc(S) = S+ZC / S+PC = S+11/T / S+1/27 ------------- (1)
Where T > 0 and ά < 1
The zero of the lead compensator Zc = 1/T -------------(2)
Pole of the lead compensator Pc = 1/άT -------------(3)
T = 1/Zc -------------------(4)
Ά = Zc / Pc -------------------(5)
Realisation of Lead Compensator using Electrical Network.
The lead compensator can be resided by the RC network in figure.
Let Ei(S) = input voltage
Eo(S) = output voltage
In the network shown in figure, the input voltage is applied to the series combination of (R1││ C) &
R2
The output voltage is obtained across Rs. Output Voltage Eo(S) = (Voltage division)
= Ei(S) = R2 /R2 + [R1 x 1/SC / R1+SC)
= Ei(S) = R2 / R2 + R1 / R1CS+1
= Ei(S) = R2 / R2(R1(S+1) + R1 / R1 CS+1
Transfer function of electrical network the ratio of output voltage to input voltage.
Eo(S) / Ei(S) = R2 (R1C S+1) / R1R2CS+R2+R1
= R1CR2 [S+1/R1C] / R1CR2 [ S+(R1+R2)/R1CR2]
= (S+1/R1C) / S + [1 / R2/(R1+R2)] 1/R1C ----------- (6)
The general form of lead compensator transfer function is Gc(S) = (S+1/T) / (S+1/άT) ---- (7)
Comparing (6) with (7)
T = R1C and ά = R2/R1+R2
Transfer function of the RC network is similar to the general transfer function of lead compensator.

Frequency Response of Lead Compensator

Consider the general form of lead compensator
Gc(S) = (S+1/T) / (S+1/άT) = (1+ST)/T / (SάT+1)/άT = ά(1+ST)/(1+Άst) -------- (8)
The sinusoidal transfer function of lead compensator is obtained by letting s= jw in equnation (8)
Gc(S) = (1+jwT) / (1+jwάT) ά = when ω = 0, ------- (9)
The sinusoidal transfer function lead compensator is obtained by letting
Gc (jw) = ά ------------- (10)
From equation (10) we can say that the lead compensator provides an attenuation of “ά”. Here
ά < 1 if the attenuation of the compensator is not desirable then it can be eliminated by a suitable
amplifier.
Let us assume that attenuation ά is eliminated by a suitable amplifier network. Now Gc(jw)
is given by
Gc(jw) = (1+jwT) / (1+jwάT) = √1+(ωT)2 Ĺtan-1(ωT) /√1+(ωάT)2 tan-1(ωάT) ---------(11)
Sinusoidal transfer function shown in equation (11) has two corner frequency ωC1 and ωC2
ωC1 = 1/T and ωC1 = 1/άT
since T > άT, ωC1 < ωC2
let A = │Gc(jw)│ in dB = 20log √1+(ωT)2/√1+(ωάT)2 ----------(12)
at very low frequencies that is upto ωC1, ωT << 1 and ωάT << 1
A = 20log √(ωT)
The approximate magnitude plot of lead compensator is shown in figure. The magnitude
plot of Bode plot of Gc(jw) is a straight line through 0 dB upto ωC1, then it has a slope of
+20 db/dec upto ωC2 and after ωC2 it is a straight line with a constant gain of 20log(1/ά)
Let ɸ = ĹGc(jw)
ɸ = tan-1(ωT) – tan-1(ωάT) --------------- (13)
As ω --- 0 ; ɸ --- 0
ω --- ∞ ; ɸ --- 0

As “ω” is varied from o to ∞. The phase angle increases from 0 to maximum value of the
ɸm at ω=ωm, then decreases from this maximum value to 0.
It can be shown that the frequency at which maximum phase lead occur is the geometric
mean of the two corner frequencie.
Frequency of maximum phase lead ωm
ωm = √ωC1 ωC2 =√1/T 1/άT = 1/T√ά
The choice of ά is govern by the inherit noise in control system. From the Bode plot of
the lead network, we observe that the high frequency noise signals are amplified by a factor
of 1/ά , while the low frequency control signals undergo unit amplification. Thus the signal
noise ratio at the output of the lead compensator is poorer than at its input. To prevent the
signal / noise at the output from deteriorating excessively, it is recommended that the value
of ά should not be less than 0.07. A typical choice of ά = 0.1. Also it is advisable to provide
two cascade lead networks when ɸm required [(ie) phase lead required] is more than 60’

Determination of ɸm, ωm, and ά :-

The frequency ωm can be determined by differentiating ɸm with respect to “ω” and
equating dɸ/dω = 0
Phase of Gc(jw) = ɸ = Ĺ(jw) = tan-1ωT-tan-1ωάT -------- (13)
On differentiating the equation (13) with repect to “ω” and equating dɸ/dω= 0 and we get
the frequency corresponding to maximum phase lead as
Ωm = 1/T√ά ------------- (14)
Also, we can express ɸm interms of ά and ß
ά interms of ɸm
ɸm = tan-1(1-ά/2√2) --------- (15)
ά = 1-sin ɸm / 1+sin ɸm -------------(16)

Procedure for Design of Lead Compensator using Bode Plot.

Following steps may be followed to design a lead compensator using bode plot and to
be connected in series with transfer function of uncompensated system G(S)
1. The open loop gain G(S) --- K of the given system is determined to satisfy the requirement of
the error constant.
2. The bode plot is drawn for the uncompensated system using the value of K. determine from
the (previous notes)
3. Phase margin of the compensated system is determined from the bode plot.
4. Determine the amount of phase angle to be contributed by the lead network.
By using the formula given below
ɸm = ϒd-ϒ+Σ
where ɸm = maximum phase angle of the lead compensator
ϒd = desired phase margin
ϒ = phase margin of the uncompensation system.
Σ = Additional phase lead to compensate for shift in gain crossover frequency.
Choose an initial choice of Σ=5’ [if ɸm is more than 60’ then realise compensator as
cascade of two lead compensator with each compensator contributing half of the
required angle]
5. Determine the transfer function of lead compensator
Calculate ά using = ά = 1-sinɸm/1+sinɸm
From the bode plot, determine the frequency at which the magnitude of G(jw) is -20log1/√2
This frequency is ωm
Calculate “T” from the relation, ωm = 1/T√ά
Transfer function of lead compensator = Gc(S) = S+1/T / S+1Άt = ά(1+ST)/(1+άST)
6. Determine the open loop transfer function of compensated system.
The lag compensator is connected in series with G(S) as shown in figure. When the
lead network is inserted in series with plant, the open loop gain of the system is
attenuated by the factor (K1) compensator to notify the attenuation caused by the
lead compensator
Open loop transfer function of the over all system = G0(S) = 1/ά (1/S+1/T)/ (1/S+1/άT) G(S)
= 1/ά ά(1+ST)/(1+άST) G(S)
7. Verify the design:-
Finally the bode plot of the compensated system is drawn and verify whether it
satisfies the given specification. If the phase margin of the compensated system is
less than the required phase margin team repeat step 4 to 7 by taking Σ by 5’ more
than the previous design.
# Design a lead compensator for unity feedback system with open loop transfer function
G(S)=K/S(S+1) (S+5) to satisfy the following specifications
(i) Velocity error constant K >> 50
(ii) Phase margin is > 20’
1. Determine K
Given that K > 50, let Kv = 50
Kv = £t S G(S) = £t S K/ S(S+1) (S+5) = K/5
S 0 S 0
Kv x 5 = K = K = 50x5 = 250
2. Draw the Bode plot
(i) G(S) = K / S(S+1) (S+5) = 250 / S(S+1) (S+5)
= 250 / S(1+S) 5(1+S/5) = 50 / S(1+S) (1+0.2S)
Let S = jw
G(jw) = 50 / (jw) (1+jw) (1+0.2jw)
(ii) Corner Frequency
(1+jw) = ωC1 – 1/1/ = 1 r/s
(1+0.2jw) = ωC = 1/0.2 = 5 r/s
ωL = 0.5 r/s
ωH = 10 r/s
(iii) Term Corner Frequncy slope dB/dec Change in slope dB/dec
50/jw -- -20
1/1+jw 1 -20 -40
1/1+0.2jw 5 -20 -60
(iv) Gain in dB,
ω = ωC = A = 20log(50/0.5) = 40dB
ω = ωC1 = A = 20log(50/1) = 34dB
ω = ωC2 = A =[-40log(50/1]+34 = 6 dB
ω = ωH10 = A = [-60log(10/5 + 6] = -12dB
(v) The phase angle of G(jw) as a function of ω is given by
ɸ = ĹG(jw) = -90-tan-1ω-tan-10.2ω
 3. Determine the phase margin
ɸgc – phase of G9jw) at gain crossover frequency
ϒ – phase margin of uncompensated system
From the bode plot of uncompensated system we get, ɸgc is -224’
Now ϒ = 180+ɸgc = 180-224 = -44’
The phase margin of the system is negative and so the system is unstable. Hence lead
compensation is required to make the system stable and to have a phase margin of 20’
4. Find ɸm
Desired phase margin = ϒd > 20’
Additional phase lead required = Σ = 50’
Maximum lead angle ɸm = ϒd-ϒ+Σ = 20’-(-44)+5’ = 69’
Since the lead angle required is greater than 60’ we have to realise the lead compensator as
cascade of two lead compensator with each compensator providing half of the required phase lead
angle.
ɸm = 69/2 = 34.5’
5. Determine the transfer function of lead compensator
ά = 1-sin ɸm / 1+sin ɸm = 0.28
dB magnitude corresponding to ωm
ωm = 20log1/√ά= 20log 1/√0.28 = -5.5 dB
from the bode plot of uncompensated system the frequency ωm corresponding to dB gain
of -5.5 is found to be 7.8 r/s
ωm = 7.8 r/s
T = 1/ω√ά = 1/7.8√0.28 = 0.24
Transfer function of the lead compensator = Gc(S) = (S+1/T)2/(S+1/άT)2
= ά2 (1+ST)2/(1+SάT)2
= (0.28)2 (1+0.24S)2/(1+0.28 0.24S)2 = 0.0784(1+0.24S)2/(1+0.067S)2
6. Open loop transfer function of compensated system
The attenuation provided by the compensator can be retained to reduce the large value of
open loop gain, so that the unstable system can be easily brought to stable region.
Open loop transfer function of compensated system = G0(S)
= 0.0784 (1+0.24S)2 (50) / (1+0.067S)2 S(1+S) (1+0.2S)
= 4 (1+0.24S)2 / S(1+S) (1+0.2S) (1+0.067S2)
7. Draw the Bode plot of compensated system to verify the design.
(i) Put s=jw
G0(jw) = 4(1+0.24jω)2 / jω(1+ jω) (1+0.2 jω) (1+0.06 jω)2
(ii) Find Corner frequency
(1+0.24jω) = 1/0.24 = 4.2 r/s
(1+1jω) = 1/1 = 1 r/s
(1+0.2jω) = 1/0.2 = 5 r/s
(1+0.067jω) = 1/0.067 = 15 r/s
(iii) ωL = 0.5 r/s ωH = 30 r/s
(iv) Term Corner frequency slope dB/dec Change in slope
4/jω --- -20
1/(1+1jω) 1 -20 -20-20 = -40
(1+jω 0.24S) 2 4.2 +40 0
1/(1+0.2jω)2 5 -20 0 – 20 = -20
1/(1+0.067jω) 2 15 -40 -20 -40 = -60
(v) Gain dB
ω = ωL = 0.5 r/s = A = 20log(4/0.50 = 18 dB
ω = ωC1 = 1 r/s = A = 20log(4/1) = 12dB
ω = ωC2 = 4.2 r/s = A = [-40 log 4.2/1] + 12 = -13dB
ω = ωC3 = 5 r/s = A = [0 log 5/4.2] + (-13) = -13dB
ω = ωA = 15 r/s = A = [-20log 15/5 + (-13)] = -23dB
ω = ωH = 30 r/s = A = [-60log 30/15 + (-23)] = -41dB
(vi) ĹG(jw) = ɸ0 = 2tan-1(0.24ω) – 90 – tan-1ω – tan-10.2ω-2tan-1(0.067ω)
In the same semi log sheet and by using the same scales, the phase plot of compensated system is
sketched as shown.
ɸgco = phase of Go(jw) at new gain crossover frequency (ωgcn)
ϒ0 = phase margin of compensated system
From the bode plot of compensated system. We get ɸgω = -140’
ϒ0 = 180 + ɸgc0 = 40’
Conclusion :-
The phase margin of the compensated system is satisfactory. Hence the design is
acceptable.
Result:-
G(S) = 0.0784(1+0.24S)2/(1+0.067S)2 = (S+4.17)2/(S+14.92)2
G(S) = 4 91+0.24S)2 / S(1+S) (1+0.2S) (1+0.067S)2
LAG – LEAD COMPENSATOR :-
A compensator having the characteristics of lag-lead network is called lag-lead
compensator. In a lag-lead network when sinusoidal signal is applied, the both phase lag and phase
lead occurs in the output, but the different frequency regions. Phase lag occurs in the low
frequency region and phase lead occurs in the high frequency region that it the phase angle varies
from lag to lead as the frequency is increased from zero to infinity.
A lead compensator basically increases the bandwidth and speeds of the response
and decreases the maximum over shoot in step response. Lag compensation increases the low
frequency gain and thus improves the steady state accuracy of the system, but reduces the speed
of response due to reduced bandwidth.
If improvements in both transients in both and steady state response are desired,
then both a lead compensator and lag compensator may be used simultaneously rather than
introducing both a lead and lag compensator as separate dements. However it is economical to use
a single lag-lead compensator
A lag-lead compensation combines the advantages of lag and lead compensations.

Realisation of Lag-Lead Compensator (network) using Electrical network

Equating the current at node X- yields
1/R1 [ei-eo] + C d/dt [ei-eo] = I ------- (1)
Voltage eo and current I are related by
1/C2 0ʃt I dt + iR2 = eo ----- (2)
Taking laplace transform of these two equations (2) & (3) with zero initial conditions and elimating
I(S) results in the equation
[1/R1+C1S] [Ei(S) – Eo(S)] = Eo(S) / 1SC2+R2
= Eo(S)/Ei(S) = (R2+1/SC2) / (R2+1/SC2) + (R1/SC1/R1+1/SC1)
= [(S+1/R1C1) (S+1/R2C2) / (S2+1/R1C1 + 1/R2C1 + 1/R2C2)S + (1/R1R2C1C2)]
On comparing equation
T1 = R1C1 ; T2 = R2C2 ; R1,R2C1C2 = άßT1T2
1/R1C1 + 1/R2C1+1/R2C2 = 1/ßT1 + 1/T2
From the above equations we get άß = 1

DESIGN PROCEDURE FOR LAG-LEAD NETWORK:-

1. Determine the value of “K” of the given system to satisfy the specified error constant.
Generally error constant (Kp,Kv,Ka) and phase margin are the specifications given. Now
draw the bode plot of the uncompensated system with this new open loop gain K.
2. Determine the phase margin of the uncompensated system from the magnitude plot.
The following steps are given below to calculate the phase margin
(i) Find out the frequency at which the magnitude curve crosses zero decibel line
that is gain crossover frequency GCof1.
(ii) Phase margin at this frequency is determined readily from the equation of the
phase angle of ɸ(ω) by substituting ω=ωc
ϒ = 180 + ɸ(ωc)
3. Determine PMc = PMd + Σ
 Where PMc = phase margin of the compensated system.
PMd = desired or specified phase margin.
Σ = margin of safety (3’ to 15’)
Usually, phase margin PMc = 180 + ɸ(ωco)
At this phase angle ɸ(ωco) is determined readily from the equation of the phase angle of ɸ(ω).
4. To determine ά of lag section:-
If we trace out the meeting point for ωcl at magnitude plot which gives the
magnitude M in dB. This gives the ß parameter of the lag section as 20logß = M
ß = 10m/20.
[the new gain crossover frequency ωco at this phase angle ɸ(ωco) is determined
readily from the equation of phase angle of ɸ(ω).
For the design of lag section of lag-lead compensator, the selected gain crossover
frequency should be higher than ωco. The choice is made as ωgcl > ωco]
5. To find transfer function of lag section.
Now, place the upper corner frequency 1/T1 of lag section at 1/10th of ß. T1 =10/ß
Calculate ßT1
Transfer function of lag section = G1(S) = 1+ST1/1+ßT1S.
6. To find the transfer function of lead section:-
Now, we proceed to design a lead section choose ά = 1/ß. The decibels for
ωm = 20log1/√ά = M1dB
if we trace out the meeting point M1 dB line at the magnitude plot which gives
ωm
T2 = 1/ωm√ά calculate άT2
Transfer function of lead section = G2(S) = 1+ST2/1+άT2S.
7. Open loop transfer function of compensated system = Gc(S) G(S)
[ G1(S) G2(S)] G(S) transfer function of compensated system]
8. Now, take the open loop transfer function as shown above and draw one more
magnitude plot.
Next calculate the phase angle, phase margin of compensated system at new gain
crossover frequency (Gcof2) and results are verified.

# A unity feedback system has an open loop transfer function G(S) = K/S(S+2) (S+60). Design a
suitable lead lag compensator is to meet the following specifications,
(i) Phase margin is alteast 40’
(ii) Steady state error for ramp input < 0.04 rad
1. Kv = £t S G(S) =£t S K / S(S+2) (S+60) = K/120
S 0 S 0
(i) Ess = 1/Kv = 0.04 ; Kv = 25
Kv = K/120 ; K = 120.Kv = 3
G(S) = 3000 / S(S+2) (S+60) = 3000 / 2S(1+S/2) 60/(1+1/ω)
G(S) = 25 / S(1+0.5S) (1+0.0167S) ----- standard form
(ii) Corner frequency
1/1+0.5jw = 1/0.5 = 2 r/s
1/0.0167jw = 1/0.0167 = 60 r/s
 ωL = 0.1 r/s
ωH = 100 r/s
now, place upper corner frequency (1/T1) of lag section at 1/10th of the ß
T1 – 10/ß = 10/4 = 2.5
ßT1 = 4 x 2.5 = 10
transfer function of lag section G1(S)
G1(S) = 1+ST1/1+SßT1 = (1+2.5S)/(1+10S)
1. To find transfer function of lead section:-
We now proceed to design the lead section, choose ά = 1/ß = ¼ = 0.25. The decibels
for ωm = -20log1/√ά = -6dB line at magnitude plot which lie on the slope of
-40db/dec and given ωm = 10 r/s
T2 = 1/ωm√ά = 1/10√0.25 = 0.2
άT2 = 0.05
transfer function of lead section G2(S) = 1+άT2 / 1+άT2S = 1+0.2S/1+0.5S
2. Transfer function of lead – lag section Gc(S) = G1(S) G2(S)
= (1+ST1) / (1+SßT1) X (1+ST2/1+άT2S)
= (1+2.5S)/(1+10S) X (1+0.2S)/91+0.05S)
Open loop transfer function of compensated system = Gc(S) G(S)
= (1+2.5S) (1+0.2S) / (1+10S) (1+0.05S) = 25 / S(1+0.5S0 (1+0.0167S)
3. Now take the open loop transfer frequency as shown above draw one more magnitude plot.
(i) Corner frequency
(1+10S) = Corner frequency = 1/ω = 0.1 r/s
(1+2.5S) = Corner frequency = ½.5 = 0.4 r/s
(1+0.5S) = Corner frequency = 1/0.2 = 5 r/s
(1+0.05S) = Corner frequency = 1/0.05 = 20 r/s
(ii) Term Corner frequency slope Change in slope
K/S -20
1/1+0.5S 2 -20 -40
1/0.0167S 60 -20 -60
(iii) Magnitude in dB
ω=ωL = 0.1 r/s = A = 20log25/0.1 = 48 dB
ω=ωC1 = 2 r/s = A = 20log25/2 = 22 dB
ω=ωC2 = 60 r/s = A = [-40log 60/2 + 22] = -37dB
ω=ωH = 100 r/s = A = [-60log 100/60 -37] = -50 dB
2. Magnitude plot cuts the zero decibels line at 7.1 r/s. Gain crossover frequency
Gcof1 = 7.1 r/s. Phase margin at this frequency is determine readily from the equation of
phase of ɸ(ω)
ɸ(ω) = -90-tan-10.5ω-tan-10.0167ω
ω=ωc = 7.1 r/s ɸ(ωc) = -171’
phase margin (PMoc) = 180 + ɸ(ωc)
= 180 – 171’ = 9’
3. Determine PMc = Pmd+Σ
= 40’+5’= 45’
Usually phase margin PMc = 180+ ɸ(gωc)
 ɸ(ωc) = PM – 180’ = 45’-180’ = -135’
the new gain crossover frequency (ωco) at this phase angle ɸ(ωco) is determined readily from the
equation.
ɸ(ωco) = -90-tan-10.5ω-tan-10.0167ω = -135’
by solving the above equation, we get new gain crossover frequency should be higher than 1.9 r/s.
the choice is made as ωgc1 > ωco. New take ωgc = 3.5 r/s.
4. Determine ß of lag section
If we trace out the meeting point for ωgcl = 3.5 r/s. at magnitude plot which lie on
the slope of -40 db/dec and gives 12 dB. This gives the ß parameter of lag section as
20logß = 12dB
ß = 1012/20 = 4
5. To find the transfer function of lag section.
1/1+0.0167S = 1/0.0167 = 60 r/s
(iv) Term Corner frequency slope in db/dec change in slope
K = 25 -- 0
1/S -- -20 -20
1/(1+10S) 0.1 -20 -40
(1+2.5S) 0.4 +20 -20
1/(1+0.5S) 2 -20 -40
(1+0.2S) 5 +20 -20
1/(1+0.05S) 20 -20 -40
1/(1+0.0167S) 60 -20 -60
Magnitude in dB. ωL = 0.1 r/s and ωH = 100 r/s
ω=ωL = 0.01 r/s = A = 20log(25/0.01) = 48dB
ω=ωC1 = 0.1 r/s = A = 20log(25/0.1) = 48dB
ω=ωC2 = 0.4 r/s = A = [-40log(0.4/0.10 + 48] = 24dB
ω=ωC3 = 2 r/s = A = [-20log 4/0.4 + 24] = 10dB
ω=ωC4 = 5 r/s = A = [-40log (5/2) + 10] = -6dB
ω=ωC5 = 20 r/s = A = [-20log (20/5) + (-6)] = -38dB
ω=ωC6 = 60 r/s = A = [-40log(60/20) + (-37)] = -37dB
ω=ωH = 100 r/s = A = [-60log(100/60) – 37] = -50dB
the magnitude plot for compensated system acts the zero decibel ωne at 3.6 r/s
Gcof2 = 3.6 r/s ωc = 3.6 r/s
The phase margin at this frequency is determined readily from the equation of the phase
angle of ɸ(ω) by
Substituting ω = ωc = 3.6 r/s
ɸ(ω) = -90-tan-110ω+tan-12.5ω-tan-10.5ω+tan-10.2ω-tan-10.05ω-tan-10.0167ω
ɸ(ωc) = -133.6’
phase margin of compensated system = PMc = 180’+ɸ(goc)
= 180’-133.6’ = 46.4’
The phase margin of the compensated system is greater than 40’. Hence the design for lead-lag
compensator is acceptable.