EC6405 CONTROL SYSTEM ENGINEERING

MAY 2015 FOURTH SEMESTER REGULATION 2013

Answer ALL questions.
PART A (10 X 2 = 20 marks)
1. Find the transfer function of the network given in figure below

Y (s ) G1(s )G2 (s )

X (s ) 1 G2 (s )

2. State Masons gain formula.

The overall gain of closed loop system can be determined by masons gain
1
formula T PK K
K

Where, T overall (or) closed loop gain

PK Path gain of Kth forward path
determinant of the graph.
= 1- (sum of all individual loop gains) + (sum of gain products of all
possible combinations of two non touching loops) (sum of gain products
of all possible combinations of 3 non touching loop)

(ie.) 1 (L1 L2 ........) (L12 L22 ........) (L13 L23 ........)

= The value of for that part of graph not touching the Kth forward path

3. How do you find the type of system?

The number of poles present at the origin of the open loop transfer function
determines the Type of the system.
5S
4. Find the unit impulse response of system H ( S ) ( S 2) with zero initial
conditions.
5( S 22) 10
H (S ) 5
( S 2) ( S 2)
2 t
h ( t ) 5 ( t ) 10 e
5. What is the use of Nichols chart?
It is used to find the closed loop frequency response from open loop
frequency response.
Frequency domain specifications can be determined from Nichols chart.
The gain of the system can be adjusted to satisfy the given specification.
6. What are the characteristics of phase lead network?
The system bandwidth is increased which is correlated to reduced rise
time and settling time and less susceptible to high frequency noise.

Prepared by Prof.S.Nagammai, HOD/EIE /KLNCE Page A.1

 EC6405 CONTROL SYSTEM ENGINEERING
The phase of the forward-path transfer function is increased in the
vicinity of the zero-gain crossover frequency. There by the phase
margin and the relative stability of the closed-loop system is increased.
7. Find the range of K for closed loop stable behavior of system with
characteristic equation
2s4 + 12s3 + 22s2 + 12s + k=0 using Routh Hurwitz stability criterion.

s4 : 2 22 k
s3 : 12 12
s2 : 20 k
s1 : 12-0.6k 0
s0 : k
For stability the following conditions are to be met:
i)(12 0.6K) 0 i.e., K< 20
ii) K>0
The range of values of K for the system to be stable is 0 K 20

8. What is the value of gain k at any given point on root locus?

Product of phasor length drawn fromopenloop polestothe point s=sa

K=
Product of phasor length drawn fromopenloop zerostothe point s=sa
9. Define observability of the system.
A system is said to be completely observable, if every state x(t0) can be
completely identified by measurement of the output y(t) over a finite time
interval.The system is completely observable if & only if the rank of
T n 1
composite matrix, Q0 [c A C ( A ) c] is n
T T T

10. State Sampling Theorem.

Shannons sampling theorem states that, a signal can be exactly
reconstructed from its samples if the sampling frequency is greater than twice the
highest frequency of the signal.
PART B (5 X 16 = 80 marks)
11.(a) (i) Explain the features of closed loop feedback control system. [4]
Features of closed loop feedback control system:
Increased accuracy: Its ability to reproduce the input accurately.
Reduced sensitivity: Its ability to reduce output variations in regard to
system characteristics and other parameter variations.
Reduced effects of nonlinearities and distortion.
Increased bandwidth.
Tendency towards oscillation or instability
Reduced effects of external disturbances or noise.

Prepared by Prof.S.Nagammai, HOD/EIE /KLNCE Page A.2

 EC6405 CONTROL SYSTEM ENGINEERING

11.(a) (ii)Derive the transfer function of system shown in fig below [12]

Solution:

The systems of equations are,

At node x1 F (t ) m1x1 K x1 x2 1
At node x2, 0 m2 x2 K x2 x1 2
Taking Laplace Transform on both sides of the above equations and substituting
numerical values,
F (S ) 2S 2 1 x1(S ) x2 (S ) 3

0 S 2 1 x2 (S ) x1(S ) 4

From equation (4) find X1(S)

x1(S ) S 2 1 x2 (S )
Substituting in equation (3) we get,

F (S ) 2S 2 1 S 2 1 x2 (S ) x2 (S )
F (S ) 2S 4

3S 2 x2 (S )

x2 (S ) 1

F (S ) 2S 3S 2
4

From equation (3) find X2(S)

x (S )
x2 (S ) 12
S 1
Substituting in equation (3) we get,

x (S )
F (S ) 2S 2 1 x1(S ) 12
S 1
S 2

1 F (S ) 2S 4 3S 2 x1(S )
x2 (S )

S2 1
F (S ) 2S 4 3S 2

Prepared by Prof.S.Nagammai, HOD/EIE /KLNCE Page A.3

 EC6405 CONTROL SYSTEM ENGINEERING
11.(b) Find the transfer function of the system shown in fig below using block
diagram reduction technique and signal flow graph technique. [16]

Solution:
Signal Flow Graph Technique
Number of forward path=1
P1 G1G2G3
Individual loop gains=4
L1 2G1G2 H1 L2 G1 L3 G2 H1 L4 G1G2G3 H2
1 1
1 2G1G2H1 G1 G2H1 G1G2G3H2
1
T P11

C(S ) G1G2G3

R(S ) 1 2G1G2H1 G1 G2H1 G1G2G3H2

Block diagram reduction technique

The given block diagram is redrawn as shown in Fig below

Step 1: Eliminating the loops

Prepared by Prof.S.Nagammai, HOD/EIE /KLNCE Page A.4

 EC6405 CONTROL SYSTEM ENGINEERING

Step 2: Eliminating the inner loop and cascading with G3

G1 G2
Let G G'
1 G1 1 G2H1
G1 G2
G
GG ' G3 1 G1 1 G2H1 3
G ''
1 2GG ' H1 G G2
1 2 1 H
1 G1 1 G2H1 1
G1G2G3 G1G2G3
G ''
1 G1 1 G2H1 2G1G2H1 1 G1 G2H1 G1G2H1 2G1G2H1
Step 3: Eliminating the outer loop

G1G2G3
C(S )

G ''

1 G1 G2H1 G1G2H1 2G1G2H1
R(S ) 1 G " H2 G1G2G3
1 H
1 G1 G2H1 G1G2H1 2G1G2H1 2
C(S ) G1G2G3

R(S ) 1 2G1G2H1 G1 G2H1 G1G2G3H2

K
12.(a)(i)A unity feedback system has the forward transfer function G ( S ) S (1 S )2 .
For the input r(t) = 1+5t, find the minimum value of K so that the steady state error
is less than 0.1 (use final value theorem) [8]
Solution:

A 1
K lt G s H s lt K
2 ; ess 0
p s0 s 0 S (1 S )
Step
1 K 1
p

K lt S G s H s lt S K 2 K
v s0 s 0 S (1 S )
A 5
ess ramp

K K
v

5
ess 0.1 0 K 50
K

12.(a). (ii) Briefly discuss about step response analysis of second order system.
[8]

Refer UNIT II material

Prepared by Prof.S.Nagammai, HOD/EIE /KLNCE Page A.5

 EC6405 CONTROL SYSTEM ENGINEERING
12.(b)(i)For the system shown in Fig. find the effect of PD controller with Td = 0.1 on
peak overshoot and setting time when it is excited by unit step input. [8]

Case i) Without derivative feedback

5
C(s) S(S+3) 5
= =
R(s) 1+ 5 S 2 +3S +5
S(S+3)
2 3
n

2 5 2.24 rad / sec

n n
2 2.24 3 0.67

M p e0.67 / 10.45 5.85%

4 4
t = = = 2.66 sec
s 0.67 2.24
n
Case ii) With derivative feedback

5
C(s) S(S+3) 5
=
S 2 S 3 5Td 5
S(S+3) d
R( s) 1 5 1 T S

Given that Td 0.1

2 5 2.24 rad / sec

n n
2 3 5Td
n
2 3 5Td 0.78
n
M p e0.78 / 1 0.61 1.97%
4 4
t = = = 2.29sec
s 0.78 2.24
n
Prepared by Prof.S.Nagammai, HOD/EIE /KLNCE Page A.6
 EC6405 CONTROL SYSTEM ENGINEERING
With derivative feedback the overshoot and settling time decreases.

12.(b) (ii)Discuss the effect of PID controller in the forward path of system. [8]
Proportional integral derivative control:

The block diagram of proportional integral derivative (PID) controller is shown in

Fig below. This mode of control action is represented by the equation
t
k de(t )
u t kc e(t ) c
i e(t ) dt k
c d
dt
uo
0

t
kc de(t )
U(t ) kc e(t )
i e(t ) dt k
c d
dt
0

k
U(S ) kc c kc dS E(S )
iS

U (S ) k
Gc (S ) kc c kc d S
E(S ) iS

In this case, the controller has three adjustable parameters namely kc , i and d .

The step response of P,PI & PID controller is shown in Fig below.

Effect of PID controller:

The proportional integral derivative control produces no offset due to the presence
of integral action and the stabilizing effect of derivative allows the gain to be
increased, so reducing the maximum deviation and increasing the speed of response
compared to P & PI controls.

Prepared by Prof.S.Nagammai, HOD/EIE /KLNCE Page A.7

 EC6405 CONTROL SYSTEM ENGINEERING
13.(a)For the following transfer function draw the Bode plot, find the gain in phase
5
margin: G ( S ) H ( S ) S (10 S )(20 S ) [16]
Solution:
0.025
G(S )H (S )
S (10.1S )(10.05 S )
0.025
G( j )
j j 0.1 11 j 0.05

Magnitude plot:

Term c CIS Slope Slope after c

before c

j
1
- -20db/dec - -20db/dec

(1 j 0.1)1 10 -20db/dec -20db/dec -40db/dec

(1 j 0.05)1 20 -20db/dec -40db/dec -60db/dec

K 0.025
Starting dB 20log n
20log 1
7.96 dB
(s ) (0.01)
Phase angle plot:

G( j) tan 1 0.1 900 tan 1 0.05

0.01 0.1 1 5 10 14.1 20 30 100 500

G( j) -90 -91 -98 -131 -162 -180 -200 -218 -253 -267

Analytical method of determining Gain and phase cross over frequencies:

To find the phase crossover frequency pc , G( j) is equated to 180

180 tan 1 0.1 900 tan 1 0.05

90 tan 1 0.1 pc tan 1 0.05 pc

Taking Tan on both sides we get,

0.1 pc 0.05 pc

1 0.1 pc 0.05 pc

1 0.005 pc 2 0 pc 14.14 rad / sec

Prepared by Prof.S.Nagammai, HOD/EIE /KLNCE Page A.8

 EC6405 CONTROL SYSTEM ENGINEERING

To find the gain crossover frequency gc , G( j ) is equated to unity.

0.025
G( j ) 1
gc 1 0.0025gc 2 1 0.01gc 2

0.000625 gc 2 1 0.0025gc 2 1 0.01gc 2

Let gc 2 a

0.000625 a 1 0.0025a 1 0.01a a3 500a 2 40000a 25 0

The roots are, a 0.0006, 400, 100

The negative roots are invalid. gc a 0.0006 0.025 rad / Sec

Prepared by Prof.S.Nagammai, HOD/EIE /KLNCE Page A.9

 EC6405 CONTROL SYSTEM ENGINEERING

13.(b) Consider a unity feedback system with open loop transfer function,
1
G f (S ) Design a lead compensator for the system so that, Kv 20 sec , with
S S 2
damping coefficient equal to 0.45 and small setting time. [16]

14.(a)Single loop negative feedback system has a loop transfer function

K ( S 6)2
G(s)H (s) . Sketch the root locus as a function of k. Find
S ( S 2 1)( S 4)
The range of k for which the system is stable.
The value of k for which purely imaginary roots exist and find the roots. [16]

Solution:

1.There are three branches of the root locus. (n=4 and m=2 so N=n=4)

2.The branches of the root locus originates with K = 0 from the open loop poles
located at s=0, j, j , and 4 resply.

3. Since there are two open-loop zeros in the finite region, two branches of root locus
terminates at the open loop zero and other two branches terminate on infinity, along
the asymptote whose angle with the real axis is given by

180[2q 1]
A ; where q = 0,1,2,. , (nm1)
(n m)

A 900 ,2700
Prepared by Prof.S.Nagammai, HOD/EIE /KLNCE Page A.10
 EC6405 CONTROL SYSTEM ENGINEERING
4. The centroid of the asymptotes is given by

Sumof poles Sumof zeros 0 j j 4 6 6

A 4
nm 2

5.The segments of the real axis between 0 and -4 lies on the root locus.

6.There is no breakaway point on the real axis.

7. Determination of angle of departure:

To find the angle of departure, let us consider the complex pole at j

The angle of departure at this open loop pole is, p 180

Where, is the angle contributed by all the other open loop poles and zeros at the
concerned open loop pole.

The angle contribution by the pole at j is,

p tan 1 j j tan 1 2 j 900

0

p tan 1 j 4 140
1

p tan 1 j 0 900
2

z tan 1 j 6 9.460
1

z tan 1 j 6 9.460
2

Angle of departure
from pole at j

p 180 - p0 p1 p2 z1 z2

p 180 90 14 90 2 9.460 5
0

The angle of departure from the complex pole at j is 5

The angle of departure from the complex pole at j is 5

8. Determination of range of values of K for stability:

The characteristic equation is, S 4S S (1 K ) S (4 12K ) 36K 0

4 3 2

Application of the Routh criterion to the above equation gives the following Routh
array:

Prepared by Prof.S.Nagammai, HOD/EIE /KLNCE Page A.11

 EC6405 CONTROL SYSTEM ENGINEERING
s4 : 1 1+K 36K

s3 : 4 4+12K 0

s2 : -8K 9K

s1 : 96K K+0.71 0

s0 : 9K

For all positive values of K the system is stable. Hence the root locus plot does not
cross the imaginary axis.

14.(b)Draw the Nyquist plot and find the stability of the following open loop transfer
K ( S 1)
function of unity feedback control system G ( s ) H ( s ) S 2 ( S 10) If the system is
conditionally stable, find the range of k for which the system is stable. [16]

SOLUTION:

In order to investigate the stability of closed loop system the Nyquist contour as
shown in Fig is used.

The open loop transfer function has two poles at origin. Hence the Nyquist contour
on s- plane enclosing the entire right half plane except the origin as shown in fig
below is used.

Prepared by Prof.S.Nagammai, HOD/EIE /KLNCE Page A.12

 EC6405 CONTROL SYSTEM ENGINEERING

MAPPING OF SECTION AB:

Along this section, S j with varying from 0 to .

K ( S 1)
G(s) H (s)
S 2 ( S 10)

K 1 j
Let S j . Therefore G( j ) H ( j ) G( j) H ( j) tan 1 180 tan 1 0.1
j 10 j
2

When 0 , G( j) H ( j) 1800

When , G( j)H ( j) 0 1800

MAPPING OF SECTION BCD:

Along this section, S Re with R and varying from 90 to 90 .

j 0 0

1
Re j j 1
K 1 Re j Re 0 1800 to1800
GH (Re j ) lim lim
Re 10 Re j 3 10
Re 1 Re j
R j 2 j R

Thus the semi circular arc of infinite radius of Nyquist contour is mapped into origin
in the G(S)H(S) plane.

MAPPING OF SECTION DE:

Along this section, S j with varying from to 0 . Mapping of this section is

the mirror image of section AB.

Prepared by Prof.S.Nagammai, HOD/EIE /KLNCE Page A.13

 EC6405 CONTROL SYSTEM ENGINEERING
MAPPING OF SECTION EFA:

Along this section, S e with 0 and varying from 90 to 90 .

j 0 0

K 1 e j
GH ( e j ) lim 1800 to1800
e j 10 e j
0 2

Thus the semi circular arc of zero radius of Nyquist contour is mapped into a circle
of infinite radius in the q(s) plane.

Conclusion:

The(-1+j0) point is not enclosed by the Nyquist contour. N=0 and P=0 Hence Z= 0.
Thus the closed loop system is stable.

15.(a) Consider a system with state-space model given below.

0 1 0

6
X 0 0 1 X
1 5 1
5 u
24
Y 1 0 0 X 0 u

Verify that the system is observable and controllable. [16]

Kalmans test:
The controllability matrix is, Qc B AB A2 B

Prepared by Prof.S.Nagammai, HOD/EIE /KLNCE Page A.14

 EC6405 CONTROL SYSTEM ENGINEERING

0 1 0 6 5
AB 0 0 1 5 24
1 5 1 24 7
0 1 0 5 24
A2 B A[ AB] 0 0 1 24 7
1 5 1 7 122
6 5 24
Qc 5 24 7

24 7 122
Qc 5408 0 (ie) its rank is r n 3. The system is completely controllable.
The observability matrix is, Q0 C T AT C T ( AT )2 C T
0 0 1 1 0
AT C T 1 0 5 0 1
0 1 1 0 0
0 0 1 0 0
( AT )2 C T AT AT C T 1 0 5 1 0
0 1 1 0 1
1 0 0
Q0 0 1 0
0 0 1
Qc 1 0 (ie) its rank is r n 3. The system is completely observable.

15. (b)Explain the functional modules of closed loop sampled data system and
compare its Performance with open loop sampled data system. [16]

Refer UNIT V material

Prepared by Prof.S.Nagammai, HOD/EIE /KLNCE Page A.15