ME354A: Vibration and Control

Wednesday 12th April, 2023

Assignment 6 Solutions
Problem 1. For the system shown below, determine the values of gain K and velocity-
feedback constant Kh so that the maximum overshoot in the unit-step response is 0.2 and
the peak time is 1 sec. With these values of K and Kh , obtain the rise time and settling
time. Assume that J =1 kg-m2 and B =1 N-m/rad/sec.

Figure 1: Block diagram of a servo system

Solution 1. Determination of the values of K and Kh : The maximum overshoot is given

by,
−πζ
 
p 

Mp = e 1 − ζ2 = 0.2
−πζ
p = 1.61
1 − ζ2
=⇒ ζ = 0.456

The peak time tp is specified as 1 sec.

π
tp =
ωd
ωd = π
ωd
=⇒ ωn = p = 3.53
1 − ζ2
The transfer function for the given system is given by:
C(s) K
= 2
R(s) Js + (B + KKh )s + K

=⇒ K = Jωn2 = 12.5 N-m

1
 B + KKh
ζ= √
2 KJ
√
2 KJζ − B
Kh = = 0.178
K
The rise time is given by:
 
π−β ωd
tr = , where, β = tan−1
ωd ζωn
=⇒ tr = 0.65 sec

Settling time is given by:

4
ts =
ζωn
=⇒ ts = 2.48 sec

Problem 2. Consider the second-order system with a finite zero and unity DC gain:

24(s + z)
H(s) =
z(s + 4)(s + 6)

Determine the effect of the zero location s = −z using MATLAB on the unit-step response
when z = 1,2,..,6.

Solution 2. The step response is the inverse Laplace transform of:

1 24(s + z)
H ′ (s) = H(s) =
s z(s)(s + 4)(s + 6)

Thus, the response of the system in the time domain is given by:
   
12 −4t 12 −6t
y(t) = 1 + −3 e + 2− e
z z

If z = 4 or z = 6, one of the modes of the system is absent from the output, and the response
is first order due to the pole–zero cancellations.

2
 Figure 2: Effect of zero on transient response

The effect of the zero is most pronounced in terms of the additional overshoot for z = 1
(zero location closest to the origin). The system also has overshoot for z = 2, 3. For z = 4
or z = 6, the responses are first order, as expected. It is interesting that for z = 5, where
the zero is located between the two poles, there is no overshoot. This is generally the case
because the zero effectively compensates for the effect of the second pole, rendering the
system as first order at any given frequency.

Problem 3. For each of the following response functions, determine whether there is a
cancellation between the zero and the pole closest to the zero. Find the approximate response
where valid.
26.25(s + 4)
G1 (s) =
s(s + 3.5)(s + 5)(s + 6)
26.25(s + 4)
G2 (s) =
s(s + 4.01)(s + 5)(s + 6)
Solution 3. The partial fraction expansion of G1 (s) is:
1 3.5 3.5 1
G1 (s) = − + −
s s + 5 s + 6 s + 3.5
The residue of the pole at -3.5, which is closest to the zero at -4, is equal to 1 and is not
negligible compared to the other residues. Thus, a second-order step response approximation

3
cannot be made for G1 (s). The partial fraction expansion of G2 (s) is:
0.87 5.3 4.4 0.033
G2 (s) = − + −
s s + 5 s + 6 s + 4.01
The residue of the pole at -4.01, which is closest to the zero at -4, is equal to 0.033, about
two orders of magnitude below any of the other residues. Hence, we make a second-order
approximation by neglecting the response generated by the pole at -4.01.
0.87 5.3 4.4
G2 (s) ≈ − +
s s+5 s+6
The response G2 (s) is approximately:
G2 (t) ≈ 0.87 − 5.3e−5t + 4.4e−6t
Problem 4. For the unity feedback system shown below, specify the gain (K) and pole
location of the compensator so that the overall closed-loop response to a unit-step input has
an overshoot of no more than 18% and a 1% settling time of no more than 0.01 sec.

Figure 3: Unity feedback system

Solution 4. The transfer function of the given system is:

Y (s) 300K 300K
= 2 = 2 (1)
R(s) s + (a + 300)s + 300(K + a) s + (2ζωn )s + ωn2
Given that the maximum percentage overshoot is 18%, thus:
−πζ
 
p 
1 − ζ 2
Mp = e
=⇒ ζ ≥ 0.4791
For 1% settling time, we have:
e−ζωn ts = 0.01
4.605
ts = = 0.01 sec
ζωn
=⇒ ωn = 961.17 rad/s
Now from Eq.(1), we have:
2ζωn = a + 300
=⇒ a ≈ 621
ωn2 = 300(K + a)
=⇒ K ≈ 2458

4
Problem 5. Determine the stability of the following closed-loop transfer functions and find
the number of poles in RHP.
10
(a) G1 (s) =
s5 + 2s4 + 3s3+ 6s2 + 5s + 3
10
(b) G2 (s) =
s5 + 5s4 + 11s3 + 23s2 + 28s + 12
Solution 5. (a)

s5 : 1 3 5
s4 : 2 6 3
s3 : 0
ϵ 3.5 0
6ϵ − 7
s2 : 3 0
ϵ
42ϵ − 49 − 6ϵ2
s1 : 0 0
12ϵ − 14
s0 : 3 0 0

Since we encounter a zero in the first column, we replace it by a small positive number,
ϵ and complete the table. From the table, we notice that there are two sign changes
in the first column. Hence the system is unstable and has two poles in the right half
plane. The choice of sign for ϵ does not affect the result. If we choose ϵ as negative,
there will still be two sign changes.

(b)

s5 : 1 11 28
s4 : 5 23 12
s3 : 6.4 25.6 0
s2 : 3 12 0 → P (s) = 3s2 + 12
dP (s)
s1 : 0
6 0
0 0→ = 6s + 0
ds
s0 : 12 0 0

In this part, we encounter a row full of zeros (row s1 ). So we make an auxiliary

polynomial from the coefficients of the previous row as, P (s) = 3s2 + 12s0 . Then we
dP (s)
differentiate the polynomial, = 6s + 0, and use the coefficients from the new
ds
polynomial to continue the table. Clearly, there are no sign changes in the first column,
hence the system is stable and has no poles in the right half plane.
An entire row of zeros will appear in the Routh table when a purely even or purely odd
polynomial is a factor of the original polynomial. ’Even polynomials’ (e.g: s4 + 5s2 + 7)
only have roots that are symmetrical about the origin.
This symmetry can occur under three conditions of root position: (1) The roots are

5
symmetrical and real, (2) the roots are symmetrical and imaginary, or (3) the roots
are quadrantal.
Each case or combination of these cases will generate an even polynomial. Thus, the
row of zeros tells us of the existence of an even polynomial whose roots are symmetric
about the origin. Another characteristic of the Routh table for the case in question is
that the row previous to the row of zeros contains the even polynomial that is a factor
of the original polynomial.