S I X

Stability
SOLUTIONS TO CASE STUDIES CHALLENGES
Antenna Control: Stability Design via Gain
From the antenna control challenge of Chapter 5,

T(s) =

Make a Routh table:

s3 1 198

s2 151.32 76.39K

s1 0

s0 76.39K 0

From the s1 row, K<392.2. From the s0 row, 0<K. Therefore, 0<K<392.2.

UFSS Vehicle: Stability Design via Gain

G3 ( s ) ( 0 . 25 s+ 0. 10925 ) K 1
T ( s )= =
1+G 3 ( s ) s 4 +3 . 483 s 3 +3 . 465 s2 + 0. 25( K 1 +2. 4288 ) s +0 .10925 K 1
6-2 Chapter 6: Stability

s4 1 3.465 0.10925K1

s3 3.483 0.25(K1+2.4288) 0

s2 0.10925K1 0

s1 0 0

0.25

s0 0.10925K1 0 0

For stability : 0 < K1 < 26.42

ANSWERS TO REVIEW QUESTIONS

1. Natural response
2. It grows without bound
3. It would destroy itself or hit limit stops
4. Sinusoidal inputs of the same frequency as the natural response yield unbounded responses even though
the sinusoidal input is bounded.
5. Poles must be in the left-half-plane or on the j axis, but not multiple.
6. The number of poles of the closed-loop transfer function that are in the left-half-plane, the right-half-
plane, and on the j axis.
7. If there is an even polynomial of second order and the original polynomial is of fourth order, the original
polynomial can be easily factored.
8. Just the way the arithmetic works out
9. The presence of an even polynomial that is a factor of the original polynomial
10. For the ease of finding coefficients below that row
11. It would affect the number of sign changes
12. Seven
13. No; it could have quadrantal poles.
14. None; the even polynomial has 2 right-half-plane poles and two left-half-plane poles.
15. Yes
16. Det (sI-A) = 0
 Solutions to Problems 6-3

SOLUTIONS TO PROBLEMS
1.

s5 1 4 2

s4 4 5 2
11 3
s3 4 2 0
31
s2 11 2 0
154
s1 - 0 0
341

s0 2 0 0

3 LHP, 2 RHP, 0 jω axis

2.

The Routh array for P( s )=s 5 + 6 s 3 +5 s2 + 8 s +20 is:

s5 1 6 8

s4 0 5 20
5 20
s3 − −
ε ε
s2 5 20

s 0 10

1 20
2
The auxiliary polynomial for row 4 is Q( s )=5 s +20 , with Q' (s )=10 , so there are two roots on

the jω -axis. The first column shows two sign changes so there are two roors on the right half-plane.
The balance, one root must be in the left half-plane.
6-4 Chapter 6: Stability

3.

s5 1 4 2

s4 -1 -4 -3

s3 ϵ 1 0
1
s2 ϵ -3 0

s1 1 0 0

s0 -3 0 0

2 LHP, 3 RHP, 0 jω axis

4.

s5 1 3 2
s4 -1 -3 -2
s3 -2 -3 ROZ
s2 -3 -4
s1 -1/3
s0 -4

Even (4): 4 jRest(1): 1 rhp; Total (5): 1 rhp; 4 j

5.

s4 1 11 18

s3 3 27 0

s2 2 18 0

s1 4 0 0 ROZ

s0 18 0 0

Even (2): 2 j; Rest (2): 2 lhp; Total: 2 j; 2 lhp

 Solutions to Problems 6-5

6.

s6 1 -6 1 -6

s5 1 0 1

s4 -6 0 -6

s3 -24 0 0 ROZ

s2  -6

s1 -144/ 0

s0 -6

Even (4): 2 rhp; 2 lhp; Rest (2): 1 rhp; 1 lhp; Total: 3 rhp; 3 lhp
7.

The characteristic equation is:

584
1+ =0
(s+2)(s+3)(s+ 4)(s+5)

Or
4 3 2
s + 14 s +71 s +154 s +704=0

s4 1 71 704

s3 14 154 0

s2 60 704 0

s1 -10.27 0 0

s0 704 0 0

The system has two closed loop poles in the rhp, so it is closed-loop unstable.

8.
Program:
numg=584;
deng=poly([-2 -3 -4 -5]);
'G(s)'
G=tf(numg,deng)
'Poles of G(s)'
pole(G)
'T(s)'
6-6 Chapter 6: Stability

T=feedback(G,1)
'Poles of T(s)'
pole(T)

Computer response:
ans =
'G(s)'

G=
584
-----------------------------------
s^4 + 14 s^3 + 71 s^2 + 154 s + 120
Continuous-time transfer function.

ans =
'Poles of G(s)'

ans =
-5.0000
-4.0000
-3.0000
-2.0000

ans =
'T(s)'

T=
584
-----------------------------------
s^4 + 14 s^3 + 71 s^2 + 154 s + 704

Continuous-time transfer function.

ans =
'Poles of T(s)'

ans =
-7.0657 + 3.3858i
-7.0657 - 3.3858i
0.0657 + 3.3858i
0.0657 - 3.3858i

System is unstable, since two closed-loop poles are in the right half-plane.

9.

The characteristic equation is:

K (s−1)
1+ =0
s (s +2)(s +3)

Or

s3 +5 s 2 + ( 6+ K ) s− K=0
 Solutions to Problems 6-7

s3 1 6+ K

s2 5 −K
6 K +30
s1 5 0

s0 −K 0

Therefore −5< K < 0.

10.
1
T ( s )=
2 s +5 s + s 2 +2 s +1
4 3

s4 2 1 1

s3 5 2 0

s2 1 5

s1 -23 0

s0 5

Total: 2 lhp, 2 rhp

System is unstable

11.
10
T ( s )= 7 6 4 35 2
s + 2 s −3 s −10 s −s −2 s +3 s +10

s7 1 -3 -1 3

s6 2 -10 -2 10

s5 2 0 -2 0

s4 -10 0 10 0

s3 -40 0 0 0 ROZ
6-8 Chapter 6: Stability

s2 ϵ 10 0 0
400
s1 ϵ 0 0 0

s0 10 0 0 0

Even (4): 1 rhp,1 lhp, 2jω axis. Tot: 2 rhp, 3 lhp, 2jω axis

12.

Even (6): 1 rhp, 1 lhp, 4 j; Rest (1): 1 lhp; Total: 1 rhp, 2 lhp, 4 j

13.
64
The characteristic equation is 1+ 5 4
=0 or s5 + s 4+ 0 s 3+ 0 s 2+ 64 s +64=0
s + s +64 s

s5 1 0 64

s4 1 0 64

s3 4 0 0 ROZ

s2 ϵ 64 0
−256
s1 ϵ 0 0

s0 64 0 0

Even (4): 2rhp, 2 lhp, 0 jω axis. Tot: 2rhp, 3lhp, 0 jω axis

 Solutions to Problems 6-9

14.
K (s+10)
The characteristic equation is 1+ =s 3 +5 s 2+ ( K + 6 ) s +10 K=0
s (s +2)(s +3)

s3 1 6+K

s2 5 10K

s1 6 -K 0

s0 10K 0

Stable for 0 < K < 6

15.
K (s−a)
1+ =0
The characteristic equation for all cases is s (s−b ) or s2 +( K−b)s−Ka=0 . The
Routh array is

s2 1 −K

s K−b

1 −Ka

a) a<0 , b<0 ⇒ K >b , K >0 ⇒ K >0

b) a<0 , b>0 ⇒ K >b , K >0 ⇒ K >b
c) a>0 , b<0 ⇒ K >b , K <0 ⇒ b<K <0
d) a>0 , b>0 ⇒ K >b , K <0 ⇒ No solution
6-10 Chapter 6: Stability

16.

K ( s+3 )( s +5 )
T ( s )= 2
(1+ K ) s +( 8 K −6 ) s+( 8+15 K )

For 1st column negative For 1st column positive

s2 1+K 8+15K K < -1 K > -1

s1 8K-6 0 K < 6/8 K > 6/8

s0 8+15K 0 K < -8/15 K > -8/15

Stable for K > 6/8

17.

Program:
%-det([si() si();sj() sj()])/sj()
%Template for use in each cell.
syms K %Construct a symbolic object for
%gain, K.
s2=[(1+K) (8+15*K) 0]; %Create s^2 row of Routh table.
s1=[(8*K-6) 0 0]; %Create s^1 row of Routh table.
s0=[-det([s2(1) s2(2);s1(1) s1(2)])/s1(1)...
-det([s2(1) s2(3);s1(1) s1(3)])/s1(1) 0 0];
%Create s^0 row of Routh table.
's2' %Display label.
s2=simplify(s2); %Simplify terms in s^2 row.
pretty(s2) %Pretty print s^2 row.
's1' %Display label.
s1=simplify(s1); %Simplify terms in s^1 row.
pretty(s1) %Pretty print s^1 row.
's0' %Display label.
s0=simplify(s0); %Simplify terms in s^0 row.
pretty(s0) %Pretty print s^0 row.
Computer response:
ans =

s2

[1 + K 8 + 15 K 0]

ans =

s1
 Solutions to Problems 6-11

[8 K - 6 0 0]

ans =

s0

[8 + 15 K 0 0 0]

18.

. For positive coefficients in the denominator, .

Hence marginal stability only for this range of K.

19.

The closed-loop transfer function:

K ( s +1)
T ( s )=
s +4 s 4 + Ks+ K
5

is always unstable since s3 and s2 terms are missing.

20.

s3 K 2K

S2

s1 0

S0 12-40K

For stability,
6-12 Chapter 6: Stability

21.

T(s) =

s4 1 -3 2K - 4

s3 3 K+ 3 0

s2 2K - 4 0

s1 0 0

s0 2K - 4 0 0

Conditions state that K < -12, K > 2, and K > -33. These conditions cannot be met simultaneously.
System is not stable for any value of K.

22.

s3 1 6031

s2 142 K+79002

s1 (777400-K)/142 0

s0 K+79002 0

There will be a row of zeros at s1 row if K = 777400. The previous row, s2, yields the auxiliary

equation, . Thus, s = ±j77.6595. Hence, K = 777400 yields an

oscillation of 77.6595 rad/s.

23.

T(s) =

Since all coefficients must be positive for stability in a second-order polynomial, -1 < K < 1;

- ∞< K < 1; -1 < 2K < ∞. Hence, - < K < 1.

 Solutions to Problems 6-13

24.

The characteristic equation is

K ( s+1)( s+5)
1+G ( s ) H ( s ) =1+ =0
s(s+ 2)(s+ 3)(s+ 7)

which can be expressed as a polynomial as

4 3 2
s + 12 s + ( 41+ K ) s + ( 42+6 K ) s +5 K=0

The Routh array is

s
4
1 41+ K 5K
s
3
12 42+ 6 K
s
2
450+6 K 5K
12
s 36(K +10.12)(K +51.88)
450+6 K
1 5K

To make all the entries of the first column >0, the dominant requirement is the one in the
fifth row K >0. Thus for closed loop stability 0< K < ∞ .

25.
K ( s+5)
The characteristic equation is 1+ =s 3+ 4 s 2+ ( K +3 ) s+5 K =0
s (s +1)(s +3)

s3 1 3+K

s2 4 5K
12−K
s1 4 0

s0 5K 0
[Link] for 0 < K < 12.
[Link] system will oscillate when K=12.
c. When K=12 the third row becomes a row of zeros, the auxiliary equation is
Qa ( s )=4 s +60. The poles on the jω axis are s=± j √ 15 so the oscillation frequency is
2

√ 15 rad/sec.
6-14 Chapter 6: Stability

26.
K
s(s +3)(s +7) 3 2
The characteristic equation is 1+ =s +10 s +22 s + K=0
1
1+
(s +3)(s +7)

s3 1 22

s2 10 K
220−K
s1 10 0

s0 K 0
The system is closed-loop stable for 0< K < 220. The system will oscillate when K=220.
When K=220 the third row becomes a row of zeros, yielding an auxiliary equation
2
Qa ( s )=10 s +220 . The poles on the jω axis are s=± j √ 22 so the oscillation frequency is
√ 22 rad/sec.
27.

T(s) =

s3 1 2K-4

s2 K-1 24
2
2K −6K−20
s1 0
K−1

s0 24 0

For stability, K > 5; Row of zeros if K = 5. Therefore, 4s2 + 24 = 0. Hence,  = for

oscillation.

28.
K 4 3 2
The characteristic equation is 1+ 3
=s +8 s +18 s +16 s+5 K=0
(s+1) (s +5)

s4 1 18 5K

s3 8 16 0

s2 16 5K 0
 Solutions to Problems 6-15

32−5 K
s1 2 0 0

s0 5K 0 0

The system is closed-loop stable for 0< K < 6.4. The system will oscillate when K=6.4.
When K=6.4 the fourth row becomes a row of zeros, yielding an auxiliary equation
2
Qa ( s )=16 s +32. The poles on the jω axis are s=± j √ 2 so the oscillation frequency is
√ 2 rad/sec.

29.

s3 1 201

s2 53 K+245

s1 10408-K 0

s0 K+245 0

a. System is stable for -245 < K < 10408.

b. Row of zeros when K = 10408. Therefore, 53s2 + 10653. Thus, s = ,

or = 14.18 rad/s.

30.
K
T ( s )=
s 4 + 9 s 2 +18 s 2 +12 s + K
6-16 Chapter 6: Stability

s4 1 18 K

s3 9 12 0
150
s2 K 0
9
81
s1 − 0 0
150 K +

12

s0 K 0 0

a. For stability 0 < K < 22.22.

150
b. Row of zeros when K = 22.22. Therefore, 9 s2 + 22.22. Thus, s = ± j√ 1.333 , or ω =
1.1547 rad/s.

c. The denominator of the closed-loop transfer function is: s 4 +9 s2 +18 s 2 +12 s +K .

Substituting K = 22.22 and solving for the roots yields: s = ± j 1.1547, – 6.393, and –
2.607.

31.

T(s) =

s3 1 K+1

s2 2 -K

s1 0

s0 -K 0

Stability if - < K < 0.

32.
 Solutions to Problems 6-17

s8 1 1.18E+03 2.15E+03 -1.06E+04 -415

s7 103 4.04E+03 -8.96E+03 -1.55E+03 0

s6 1140.7767 2236.99029 -10584.951 -415 0

s5 3838.02357 -8004.2915 -1512.5299 0 0

s4 4616.10784 -10135.382 -415 0 0

s3 422.685462 -1167.4817 0 0 0

s2 2614.57505 -415 0 0 0

s1 -1100.3907 0 0 0 0

s0 -415 0 0 0 0

a. From the first column, 1 rhp, 7 lhp, 0 j

b. G(s) is not stable because of 1 rhp.

33.

The characteristic equation for this system is:

K s2 + ω20 4 2 K 2 Kω0
2
1+ =0 s +( aω0 + )s + =0
m T s 2 ( s2 + aω2 ) mT mT
0 or

The Routh array is:

K Kω20
s4 1 (aω 20 + )
mT mT
K
s3 0 4 (aω 20 + )
0 2
mT
K Kω20
s 2 (aω 20 + )
-2
mT mT
Kω20 K 2
s 4 + 4( aω 20 + )
mT mT
K 2
( aω 20 + )
mT
Kω20
1
mT
6-18 Chapter 6: Stability

The second row of zeros was substituted with the coefficients resulting from differentiating the
2
4 K 2 Kω0
Q a ( s )=s +( aω20 + )s +
characteristic equation:
mT mT and
K
Q ' a (s )=4 s 3 +2(aω 20 + )
mT .

Since all the plant parameters are positive, there are two sign changes in the first column of the

Routh array. So there are two poles in the RHP, two must be in the LHP.

34.
Program:
A=[0 1 0;0 1 -4;-1 1 8];
eig(A)

Computer response:
ans =

7.4641
0.5359
1.0000

35.
Writing the open-loop state and output equations we get,

Drawing the signal-flow diagram and including the unity feedback path yields,
 Solutions to Problems 6-19

In closed loop u=r−c =r− y , substituting

x˙1=x 2
x˙2=x 2+ 2 x 3 +r − y=x 2+ 2 x 3 +r−¿ x 3 ¿=−x 1 + x 2+ x 3 +r

x˙3=−5 x 1−4 x2−3 x 3

y=x 1 +¿ x 3

In matrix form the closed loop system is

[ ] []
0 1 0 0
ẋ= −1 1 1 x+ 1 r
−5 −4 −3 0

y= [ 1 0 1 ] x

The characteristic equation is

| |
s −1 0
|s I − A|= 1 s−1 −1 =s 3+ 2 s2 −2 s+ 8=0
5 4 s +3

The Routh array:

s
3
1 -2
s
2
2 8
s -6
1 8

Since there are two sign changes in the first column the system is closed loop unstable.
6-20 Chapter 6: Stability

36.
Program:
A=[0,1,0;0,1,2;-5,-4,-3];
B=[0;1;0];
C=[1,0,1];
D=0;
'G'
G=ss(A,B,C,D)
'T'
T=feedback(G,1)
'Eigenvalues of T'
ssdata(T);
eig(T)

Computer response:
ans =

G=

a=
x1 x2 x3
x1 0 1 0
x2 0 1 2
x3 -5 -4 -3

b=
u1
x1 0
x2 1
x3 0

c=
x1 x2 x3
y1 1 0 1

d=
u1
y1 0

Continuous-time state-space model.

ans =

T=

a=
x1 x2 x3
x1 0 1 0
x2 -1 1 1
x3 -5 -4 -3
 Solutions to Problems 6-21

b=
u1
x1 0
x2 1
x3 0

c=
x1 x2 x3
y1 1 0 1

d=
u1
y1 0

Continuous-time state-space model.

ans =

Eigenvalues of T

ans =

0.2442 + 1.7764i
0.2442 - 1.7764i
-2.4883 + 0.0000i

37.
The matrix A was found to be:

[ ]
0 1 0 0
29 . 8615 0 0 0
A=
0 0 0 1
−0 . 9401 0 0 0

Using MATLAB gives:

Program:
A=[0 1 0 0;29.8615 0 0 0;0 0 0 1;-0.9401 0 0 0];
eig(A)

Computer response:
ans =
0
0
5.4646
-5.4646
6-22 Chapter 6: Stability

With one pole only in the left half-plane, one in the right half-plane, and two on the j-
axis (in this case, at the origin), we conclude that this unit is unstable and requires
stabilization.

SOLUTIONS TO DESIGN PROBLEMS

38.
T(s) =

s3 1 11.91+11K

s2 5.45+K 43.65+10K

s1 0

s0 43.65+10K 0

For stability, - 0.36772 < K < ∞. Stable for all positive K.

 Solutions to Design Problems 6-23

39.
0 . 7 K (s +0 .1 )
T (s )=
s +2 . 3s + 1. 37s2 + 0 .265 s+(0 .07 K + 0. 015 )
4 3

s4 1 1.37 0.07K+0.015

s3 2.3 0.265 0

s2 1.2548 0.07K+0.015 0

s1 0.23751 - 0.12831K 0 0

s0 0.07K+0.015 0 0

For stability, – 0.21429 < K < 1.85106

40.

The characteristic equation is given by:

1×10−6 s 2 +1 . 314×10−9 s+2 . 66×10−13
1+ K =0
s 3 +0 . 00163 s 2 +5 . 272×10−7 s+ 3. 538×10−11
Or

s3 +(0 . 00163+1×10−6 K ) s2 +(5 .272×10−7 +1. 314×10−9 K )s+(3 .538×10−11 +2 . 66×10−13 K )=0

The corresponding Routh array is:

s3 1 5 .272×10−7 +1. 314×10−9 K

s2 0 . 00163+1×10−6 K 3 .538×10−11 +2. 66×10−13 K
1. 314×10−15 ( K +1371 .6 )( K +457 . 8 )
s
0 .00163+ 1×10−6 K

1 3 .538×10−11 +2. 66×10−13 K

6-24 Chapter 6: Stability

For stability row 2 requires K >−1630 and row 4 requires K >−133 . 008 . The dominant requirement

being the latter. It is clear also that when K >−133 . 008 , the first element on row 3 is positive. So the

overall requirement for stability is K >−133 . 008 .

41.

The characteristic equation of the system is given by:

KC KC 1
1+ 2
− =0
ms +bs +k K f Ts+1 or
K f (ms 2 +bs+k )(Ts+1 )+ K f K C (Ts+1 )−K C (ms 2 +bs+k )=0 or
2 2
K f (ms +bs+k )(Ts+1 )+ K f K C (Ts+1 )−K C (ms +bs+k )=0 or
K f (mTs +(bT +m )s +(kT +b )s +k )+ K f K C (Ts+1 )−K C (ms 2 +bs+k )=0
3 2
or

K f mTs 3 +[ K f (bT +m)−K C m ]s 2 +[ K f ( kT +b)+K f K C T −K C b ] s+[ k +K f K C −K C k ]=0

Substituting numerical values the equation becomes:

5 .28×10−6 K f s 3 +[0 . 03444 K f −1. 8 ]s 2 +[1584 .78 K f −11700]s+[22500 K f −31. 5×10 6 ]=0

The Routh array is given by

−6
s3 5 .28×10 K f 1584 . 78 K f −11700

s2 0 . 03444 K f −1. 8 22500 K f −31. 5×10 6

5. 28×10−6 [ 22500 K f −31 .5×106 ]−[ 1584 . 78 K f −11700 ] [ 0 . 03444 K f −1 . 8 ]
s
−[ 0. 03444 K f −1 .8 ]

1 22500 K f −31. 5×10 6

To obtain positive quantities on the first column it is required:

5 .28×10−6 K f >0 ⇒ K f > 0

0 . 03444 K f −1. 8>0 ⇒ K f >52. 26
22500 K f −31. 5×10 6 >0 ⇒ K f >1400
 Solutions to Design Problems 6-25

[ 1584 . 78 K f −11700 ][ 0 .03444 K f −1 . 8 ]−5 . 28×10−6 [ 22500 K f −31. 5×10 6] >0

or

54 . 52 K 2f −402. 48 K f −2852 .6 K f +21060 > 0 . 1188 K 2f −166 .32 K f

or

54 . 4 K 2f −3088 . 76 K f +21060 > 0

or

K 2f −56 . 8 K f +387 .1 > 0

or

( K f −7 .92 )( K f −48 .88 ) > 0

So either
K f <7 . 92 and K f < 48 .88 ⇒ K f <7 . 92

or
K f >7 . 92 and K f > 48 .88 ⇒ K f >7 . 92

The most dominant requirement is given by the fourth row. We conclude requiring
K f >1400 .

42.

For simplification we substitute parameter values into the open loop transfer function. It becomes:

K p s+ K I 80784 s 2+ 4322.8 s +1
G ( s )= 2
s 545760

The characteristic equation 1+G ( s )=0 becomes:

K p s+ K I 80784 s2 + 4322.8 s +1
=0 or 545760 s + ( K p s + K I ) ( 80784 s +4322.8 s+1 ) =0
2 2
1+ 2
s 545760

Or
3
80784 K p s +¿

The Routh array becomes:

❑
80784 K p K p + 4322.8

4322.8 K p + 80784 K I +545760 KI

2 2
4322.8 K +545760 K p +18686599.84 K p K I + 349213075.2 K I +2359211328 K I
p
4322.8 K p +80784 K I +545760
6-26 Chapter 6: Stability

KI

It is clear from the array that the entries in the first column will be positive for all K p >0 , K I > 0

43.
Qw ( s )
GWF ( s )=
First, find the transfer function of the internal (flow-control) loop, X (s ) , then the overall
C (s) G LC (s)⋅GWF ( s )⋅Gfw (s )
T (s)= =
transfer function
R (s) 1+GLC ( s )⋅GWF (s )⋅G fw ( s )

The transfer function of the internal (flow-control) feedback loop in figure 2 is:

0. 5+2 s
Qw (s) G FC (s)×GV (s) 3 s +1 0 .5+2 s
GWF ( s )= = = =
X (s) 1+G FC ( s )×GV ( s ) 0 . 5+2 s 1. 5+5 s
1+
3 s+1

Thus, the overall system transfer function is:

( 1 . 5+5 s [ s(25 s+1) ( 2 s +2 s+1 ) ]

)
0 . 5+2 s 2
( 0 .5 +K s ) 2
C (s) G (s)⋅G ( s)⋅G (s)
LC WF fw
T (s)= = =

1 .5+5 s )[ s(25 s+1 ) ( 2 s +2 s+1 ) ]

1+ ( 0. 5 +K s ) (
R (s) 1+G ( s)⋅G (s)⋅G ( s)
LC WF fw 0 .5+2 s 2
2

( K s+0 .5 ) ×( 4 s+1 )
= s ×( 5 s+1 .5 )×(25 s+1)×( 2 s2 +2 s+1 )+ ( K s+0 .5 ) ×( 4 s+1 )
( K s+0 . 5 )×( 4 s +1 )
= ( 125 s 3 +42. 5 s 2 +1 .5 s )( 2 s2 +2 s+1 )+ ( K s+0 .5 ) ×( 4 s+1 )

The characteristic polynomial is, therefore:

P( s )=( 125 s 3 +42 .5 s2 +1. 5 s )( 2 s 2 +2 s +1 ) + ( K s +0 . 5 )×( 4 s+1 ) =

= 250 s5 +335 s 4 +213 s3 + ( 45 . 5+4 K ) s 2 + ( 3 .5+ K ) s+0.5

 Solutions to Design Problems 6-27

Hence, the Routh-Hurwitz array for the system is given by:

s5 250 213 3.5 + K 0

s4 335 45.5 + 4 K 0.5 0

s3 335×213−250×( 45 . 5 + 4 Κ ) #
335 ( 3 .5 + K ) −125 0

s2 C 0.5 0

s1 C [ 335 ( 3 .5 + K )−125 ] −0 . 5× [335×213−250× ( 45 .5 + 4 K ) ] 0

C

s0 0.5

From the s3 row: 335×213−250×( 45 . 5 + 4 Κ ) >0 , therefore K < 59.98 (4)

From the s2 row: C =

[ 335×213−250 ( 45 . 5 + 4 K ) ]× ( 45 .5 + 4 K )−335×[ 335 ( 3. 5 +K )−125 ]
335×213−250 ( 45. 5 + 4 K ) > 0.

Therefore: [ 335×213−250 ( 45. 5 + 4 K ) ]×( 45 . 5 + 4 K ) −335×[ 335 ( 3 .5 +K )−125 ] > 0.

This inequality may be expanded to:

[ 59980−1000 K ]× ( 45.5 + 4 K )−112225 K −392787 .5+41875 =

−4000 K 2 +82195 K +2378177 . 5>0 , or ( K−36 . 7339 )×( K +16 . 1852)<0 which shows

that for stability:

K < 36.73

#
The S3 row was multiplied by 335 and the S1 row was multiplied by C.
6-28 Chapter 6: Stability

C [ 335 ( 3 .5 + K )−125 ] −0 . 5× [ 335×213−250× ( 45 .5 + 4 K ) ]

For C > 0, the s1 row C is greater

than 0 if:

0 . 5×[ 335×213−250× ( 45 .5 + 4 K ) ]
[ 335 ( 3 . 5 + K )−125 ]− C >0

This inequality may be re-written as:

2
0 . 5×[ 335×213−250× ( 45 .5 + 4 K ) ]
[ 335 ( 3 .5 + K )−125 ]−
[ 335×213−250 ( 45 . 5 + 4 K ) ]×( 45 . 5 + 4 K )−335×[ 335 ( 3 .5 +K )−125 ]
>0

After algebraic manipulations this can be rewritten as:

3 2 8
−1340000 K +22845325 K +942768725 K +6.9234 × 10 >0

or

( K +0.7485 )( K +18.84 ) ( K −36.7 ) < 0

From the previous, and for positive K, we conclude that the system will be stable only if level controller’s

derivative gain, KDLC is within the range: 0 < KDLC< 36.7.

 Solutions to Design Problems 6-29

44.

[ ]
s+11. 7 −6 . 8 −61 . 6 K −7 . 7 K
3. 5 s+24 66 . 9 K −8 . 4 K
sI− A=
0 −1 s 0
−1 0 10 s
s+24 66.9 K −8.4 K 3.5 66.9 K −8.4 K
det (sI− A)=(s+11.7)|−1 s 0 |+6.8|0 s 0 |
0 10 s −1 10 s
3.5 66.9 K −8.4 K 3.5 s+24 66.9 K
−61.6 K|0 −1 0 |+7.7 K|0 −1 s |
−1 0 s −1 0 10
=(s+11.7) ( s+24)|{s 0
10 s
|−66.9 K|
−1 0
0 s
|−8.4 K|
−1 s
0 10
| }
{
+6.8 3 .5|
s 0
10 s
|−66.9 K|
0 0
−1 s
|−8.4 K|
−1 s
0 10
| }
−61.6 K 3.5| {
−1 0
0 s
|−66.9 K|
0 0
−1 s
|−8.4 K|
0 −1
−1 0
| }
{
+7.7 K 3.5|
−1 s
0 10
|−(s+24)|
0 s
−1 10
|+66.9K|
0 −1
−1 0
| }
=s 4 +35. 7 s3 +(304 . 6+59 . 2 K )s 2 +840 . 41 Ks+713 .3 K−1032. 57 K 2 =0

The Routh array is:

s1 304 . 6+59
713
. 2.3
KK

s 35 .7 840 . 41 K−1032 .57 K 2

s 35 . 66 K +30428. 6. 92 K 2 +713 .3 K
28936 . 58 K 2 +230524 . 08 K
s
35 . 66 K +304 . 6
6-30 Chapter 6: Stability

2
1 28 . 92 K +713 .3 K

Row 3 is positive if K >−8 . 54

Rows 4 and 5 are positive if K >0

So the system is closed loop stable if K >0 .

45.
The characteristic equation for this system is
1022.5 K
1+ =0
s (s +1)(s +2.5)(s +10)(s+1000)
or
5 4 3 2
s +1013.5 s +13537.5 s +37525 s +25000 s+1022.5 K=0
The Routh array is:

s
5
1 13537.5 25000
s
4
1013.5 37525 1022.5K
s
3
13500.5 25000-1.009K
s
2
0.07574K+35648.22 1022.5K
s 2
0.07642 K +13838336.8 K −891205500
−(0.07574 K + 35648.22)
1 1022.5K

For closed loop stability the last row requires K>0, which means that the fourth row will also be positive.

This makes the denominator of the 5th row negative which requires a negative numerator. The numerator of

the 5th row can be expressed as 0.07642( K −64.4)( K +1.81078 × 108) , which will be negative as

long as K<64.4. Therefore the range for closed loop stability is 0<K<64.4.
 Solutions to Design Problems 6-31

46.
The pay-loaded characteristic equation is:

s3 +( m'Igh'' (q +k k )+q g) s + k m 'Igh' ' s + m'Igh' ' ( k k +q )=0 ,

2 d 1
2
i 1

which for convenience we express as

3 2
s +a 1 s +a 2 s+a 3=0
The corresponding Routh array is

s
3
1 a2
s
2
a1 a3
s a1 a2−a3
a1
1 a3

Since it is known that all a i> 0, the condition for stability is a 1 a 2> a3
Substituting

( m' gh '
I'
( q2 + k k d ) + q1 g k
I'
> )
m' gh ' m' gh '
I'
( k k i +q1 )
Solving
m' gh ' q1 +k k i −q1 gk
>
I' k (q 2+ k k d )

47.
The characteristic equation can be obtained by calculating =0 from Mason’s Rule. Namely

∆=1+ F ( Ks G −G )=0
u
u x

Substituting
4.5
1+ 2 ( 0.4 Gu +0.4 s )=0
s + 2.4 s +2.25
or
2
s +4.2 s+ ( 2.25+ 1.8Gu ) =0
The Routh array is
2
s 1 2.25+1.8 Gu
s 4.2
1 2.25+1.8 Gu
6-32 Chapter 6: Stability

For closed loop stability G u >−1.25 .

 Solutions to Design Problems 6-33

48.
a. Applying the feedback formula to the inner loop

1 1
Ge = 2
= 2
s( s +6 s+8 )+ K 2 s s [ s +6 s+(8+ K 2 )]

Thus, the inner loop will have two equal negative real poles if K2 = 1

K1
T ( s )= 3 2
Multiplying Ge by K1 and closing the loop yields
s +6 s + 9 s+ K 1

Making a Routh table:

s3 1 9 0

s2 6 K1

s1 (54 – K1) / 6 0

s0 K1

For the system to be stable: K1 > 0 and 54 – K1 ≥ 0.

b. For the system to oscillate, the s1 row must be a zero row; e.g., K1 = 54.

Hence, the frequency of oscillation is found from: 6s2 + 54 = 0 or  = 3 rad/s.

c. If a real closed-loop pole is at s = - 5, the characteristic polynomial may be factored into

s3 + 6 s 2 + 9 s+ K 1=(s +5)( s2 +as+ b ) . Thus:

5 b = K1, 5 a + b = 9, and 5 + a = 6  a = 1, b = 4, and K1 = 20.

Hence, the second-order factor of the characteristic polynomial is: s2 + s +4

6-34 Chapter 6: Stability

ω = 2 & 2 ςωn =1  ζ = 0.25 for the dominant poles, which, therefore, are given by:
This yields: n

√
−ςωn ± j ω n 1−ς 2 =−0.5± j 1.94 . Since the pole at - 5 is 10 times farther than the

dominant poles from the j axis, the step response of this system, c(t), may be approximated by a

second-order under-damped response.

d. Thus, at K = 20, the % O.S. and settling time, Ts, are:

4
−(ςπ / √1−ς 2 ) T s=
% O.S. = e ×100= 81.1% & ςωn = 8 seconds

49.

The transfer function of the minor loop shown in Figure P5.50 was found to be:

Ω L (s ) 25 s2 + 30 s+ 312500
D( s )= =
T em (s ) s 3 + 8 .1 s 2 +62003 s+31250

Thus, the transfer function of the forward path of the system shown in Figure P 5.50 is given by:

Ω L (s ) 10 (K P s +0 .1 )(25 s2 +30 s +312500)

K M G C ( s )D( s )= =
E(s ) s (s 3 +8 . 1 s2 +62003 s +31250 )
Here, E(s) = L(s) – r(s).

The characteristic polynomial for the system, hence, is given by:

P( s )=s 4 +(8. 1+250 K P )s 3 +(62028+300 K P )s2 +(31280+3125000 K P )s+312500

Constructing the Routh table, gives:

 Solutions to Design Problems 6-35

s4 1 62028+300 K P 312500

s3 8 . 1+250 K P 31280+3125000 K P 0

s2 C 312500 0

s1 D 0 0

s0 312500 0 0

(8 .1+ 250 K P )( 62028+ 300 K P )−(31280+3125000 K P )

C=
Here:
8 .1+250 K P (1)

C (31280+3125000 K P )−312500(8 .1+250 K P )

D=
and C (2)

For the system to be stable, all of the entries must be positive. This means that:

Kp > – 0.0324, C > 0, and D > 0.

Hence, equations (1) & (2) may be re-written as:

( K P+165 . 023 )(K P +0 . 0381)¿ 0 ¿ , which gives K > – 0.0381

p

( K P+165 . 002)( K P +0 . 0381)( K P+0. 0100 )¿ 0 ¿ , which yields K > – 0.01

p

Reviewing the last two equations, we conclude that for this system to be stable Kp > – 0.01 if KI = 0.1,

which means that this system will be stable for any positive value of Kp if KI = 0.1.
6-36 Chapter 6: Stability

50.

The Characteristic Equation is given by

−520 s−10 . 3844

1+ K =0
s +2 .6817 s 2 +0 .11 s+ 0 .0126
3

or

s3 +2. 6817 s2 +0 .11 s+0 . 0126−K (520 s+10 . 3844 )=0

or

s3 +2. 6817 s2 +(0 .11−520 K )s+(0 .0126−10 .3844 K )=0

The Routh Array is:

s3 1 0 . 11−520 K

s2 2 .6817 0 . 0126−10 . 3844 K

0 .2824−1384 . 1 K
s
2. 6817
1 0 . 0126−10 . 3844 K

Thus for stability

0 .2824−1384 . 1 K
>0
2. 6817 or K <2 . 04×10−4

and

0 . 0126−10 . 3844 K >0 or K <1 .21×10−3

The intersection of both requirements gives K <2 . 04×10−4 .

 Solutions to Design Problems 6-37

51.

From the block diagram it is readily obtained:

Ω (s )
=
( K P s+ 40
s )( 0 . 2491 × (10 s +6 )
s (s +0 .5173 ) + 0 .5 × (10 s+ 6) × (s +0 .01908 ) )
( )(
R v (s )
)
K P s+ 40 0 . 2491 × (10 s +6 )
1+ 0 . 0443
s s (s +0 .5173 ) + 0 .5 × (10 s+ 6) × (s +0 .01908 ) or

Ω (s) 2. 491 ( K P s+ 40 ) ( s+ 0. 6 )
=
[
R v ( s ) s ( 6 s2 + 3. 613 s +0 . 0572 ) +0 .11035 K P s2 + ( 40+ 0. 6 K P ) s +24 ]

r
v= ω=0 .06154 ω
Noting that the change in car speed,
itot , we get the system transfer function, T(s)

= V(s)/Rv(s):

The characteristic polynomial for that system is, therefore:

P( s )= 6 s 3 + ( 3 .613+ 0. 11035 K P ) s2 + ( 4 . 4712+0 . 06621 K P ) s +2 .65

Hence, the Routh-Hurwitz for the system is given by:

s3 6 4 . 4712+0 .06621 K P 0

s2 3 . 613+0 .11035 K P 2.65 0

s1 ( 3 .613+ 0 .11035 K P )( 4 . 4712+0 . 06621 K P ) −15 .9 0 0

3 . 613+0 .11035 K P

s0 2.65 0 0
6-38 Chapter 6: Stability

For stability,
3 . 613+0 .11035 K P ≥0⇒ K P≥−32. 74 and

( 3 . 613+0 . 11035 K P )( 4 . 4712+ 0. 06621 K P )−15 . 9≥0 . That is:

0 . 00731 K 2P +0 . 73261 K P +0 . 2544456= 0 . 00731(K 2P +10 0 .22 K P +34 . 81)⇒

( K 2P+10 0 . 22 K P +34 . 81 )=( K P +99 .872 )( K P+0 . 348 )≥0

The latter condition indicates that for stability

K P≥− 99 . 872 and K P≥− 0 . 348

The intersection of these two requirements shows that for stability:

K P≥− 0 . 348 or, alternatively
K P >0
sc .

52.
Using the suggested first order Padé approximation, the characteristic equation for the
system is given by
−6
137.2× 10 (1−19.5 s)
1+ K 2 =0
(s +0.224 s+196 × 10−6 )(1+19.5 s)
Equivalently

0.4368 s3 +1.4368 s 2 + ( 26.222 ×10−3 −2.6754 ×10−3 K ) s+ ( 196× 10−6 +137.2 ×10−6 K )=0

The Routh array is

−3 −3
s
3
0.4368 26.222 ×10 −2.6754 ×10 K
−6 −6
s
2
1.4368 196 ×10 +137.2× 10 K

s −3
37.6 ×10 −3.904 ×10 K
−3

1.4368
 Solutions to Design Problems 6-39

−6 −6
1 196 ×10 +137.2× 10 K

For a positive first column in the array, the third row requires K >−1.42; the fourth row K <9.63 .
The intersection of these requirements is −1.42< K < 9.63.