ME354A: Vibration and Control

Sunday 22nd January, 2023

Assignment 2 Solutions
Problem 1.

(i) The friction force between mass m and the contact surface is modelled by Coulomb
friction Fd = µk W (µk is kinetic friction coefficient). Derive the equation of motion for
the system shown in Fig. 1. Solve the equation for displacement x. Initial conditions
are given by x(0) = x0 and ẋ(0) = 0, where x0 is sufficiently large that the restoring
force in spring exceeds static friction force.

Figure 1: Spring mass system with Coulomb friction.

Solution 1. Assuming x0 is sufficiently large that the restoring force in spring exceeds static
friction force.

Figure 2: FBD of spring mass system with Coulomb friction

Let ẋ > 0

−kx − µk W = mẍ =⇒ mẍ + µk W + kx = 0

Let ẋ < 0

−kx + µk W = mẍ =⇒ mẍ − µk W + kx = 0

Combining above these equations, we have

mẍ + µk W sgn(ẋ) + kx = 0 (1)

(
1 ẋ > 0
sgn(ẋ) =
−1 ẋ < 0

1
 Take Fd = µk W . Equation(1) is non-linear and can be written in form of two linear
equation as below. These can be represented as equations for forced vibration.
mẍ + kx = −Fd for ẋ > 0 (2)
mẍ + kx = Fd for ẋ < 0 (3)
After taking x0 displacement in positive direction, equation of motion for first half cycle
(0 < t < t1 ) is given by equation(3)
k kFd
ẍ + x=
m mk
Fd
fd =
k
2 2
ẍ + ωn x = ωn fd ; x(0) = x0 ; ẋ(0) = 0 (4)
Solution to equation(4) is given by
x(t) = (x0 − fd ) cos(ωn t) + fd for 0 < t < t1 (5)
To find t1
ẋ(t1 ) = −(x0 − fd )ωn (sin(ωn t)) = 0
π
t1 =
ωn
x(t1 ) = −(x0 − 2fd )
For the second half cycle (t1 < t < t2 ) equation of motion is given by equation(2)

ẍ + ωn2 x = −ωn2 fd ; x(t1 ) = −(x0 − 2fd ); ẋ(t1 ) = 0 (6)

Solution to equation(5) is given by
x(t) = (x0 − 3fd ) cos(ωn t) − fd for t1 < t < t2 (7)
To find t2
ẋ(t2 ) = −(x0 − 3fd )ωn (sin(ωn t)) = 0
2π
t2 =
ωn
x(t2 ) = (x0 − 4fd )

Figure 3: Response due to coulomb friction

2
Problem 1.
(ii) List at least three differences between the free vibration response of a system with
Coulomb damping and the free vibration response for a system with viscous damping
whose free vibrations are under-damped.

Solution 1. (ii) Three Differences between the systems are:

(a) The magnitude of the Coulomb damping has no effect on the frequency or period of
motion while the magnitude of the viscous damping does affect ωd and period Td . An
increase in viscous damping leads to decrease in ωd and an increase in Td .

(b) The amplitude of vibration for a system with Coulomb damping decreases by a constant
amount per cycle (linear decrease) while the amplitude of vibration for a system with
viscous damping decreases exponentially.

(c) Motion ceases for a system with coulomb damping when the ampitude becomes small
enough such that the force in the elastic member is insufficient to overcome static
friction, leading to permanent displacement from equilibrium. Motion continues indef-
initely for a system with viscous damping.

Problem 2. A heavy-duty front-loading washing machine has a mass of 20 kg, a spinner

diameter of 0.8 m, and it spins at 300 rpm. The machine is mounted on a support with a
stiffness of 380 N/m and a damping coefficient of 38 Ns/m. If the maximum allowed load of
clothes is 10 kg, what is the worst case value of:
(i) vibration amplitude, and

(ii) transmitted force.

Consider this as a single-DOF system and the lump of clothes as a point mass.

Solution 2. Given:

ω = 300 rpm = 10π rad/sec , k = 380 N/m, c = 38 Ns/m, m = 10 kg, M = 20+10 = 30 kg

NOTE: M is not 20kg, it’s the mass of the whole system supported by spring and damper
i.e. sum of eccentric mass (lump of clothes) and the affecting body into consideration (empty
washing machine). Therefore, M = 30 kg.

The worst case would be when the lump of clothes is as far as possible from the center
of the spinner.
0.8
∴e= = 0.4 m
2
(i) Equation of motion for rotating unbalance mass:

M ẍ + cẋ + kx = meω 2 sin ωt

3
 Response at steady state

meω 2
x(t) = p sin(ωt − ψ)
(k − M ω 2 )2 + (cω)2

where, ψ = tan−1 ( k−M

cω
ω2
) = −0.0408 radians

meω 2
=⇒ Vibration amplitude, X = p = 0.1350 m
(k − M ω 2 )2 + (cω)2

(ii) Force transmitted to ground = Spring Force + Damper Force = kx + cẋ

meω 2
=p (k sin (ωt − ψ) + c cos(ωt − ψ)
(k − M ω 2 )2 + (cω)2
= 0.1350(380 sin (10πt + 0.0408) + 38 cos(10πt + 0.0408))
= 51.30 sin (10πt + 0.0408) + 5.13 cos(10πt + 0.0408)

Problem 3. The landing gear of an airplane can be idealized as the spring-mass-damper

system shown in Fig. 4. If the runway surface is described as y(z) = y0 cos 2πz


L
, and given
that velocity of the plane during landing is constant, v = 252 km/h, determine the values of
k and c that limit the amplitude of vibration of the airplane (x) to 0.1 m. Assume L = 1750
m, m = 2000 kg, and y0 = 0.2 m.

4
 Figure 4: Model of the landing gear

Solution 3. Note:Make following corrections in the question L = 2.8 m. Assume

value of spring constant to be k = 5e6 N/m.
Solution: If aeroplane is going with constant speed the support motion can be modelled
in the function of time. We have to find the frequency (f ) with which the runway surface
varies. f = Lv  
2πz
y(z) = y0 cos
L
Z t
z(t) = v dz = vt
0
v ω
f= =⇒ v = Lf =⇒ v = L
L 2π
2πz 2πLω
= t = ωt
L L2π
y(z(t)) = y(t) = y0 cos(ωt)

2πV
ω= = 157.08rad/s
L
k = 5e6N/m

5
 r r
k 5e6
ωn = = = 50rad/s
m 2000
r
ω
r= = 3.1416
ωn
From Fig 3(b), considering support motion, we can write equation of motion to be

ẍ + c(ẋ − ẏ) + k(x − y) = 0 (8)

with X = 0.1m, and Y = y0 = 0.2 For eq (8) transmissbility is given by

p
|X| 1 + (2ζr)2
=p
|Y | (1 − r2 )2 + (2ζr)2
p
0.1 1 + (2ζr)2
=p
0.2 (1 − r2 )2 + (2ζr)2
With r = 3.1416, from the above equation we can find

ζ = 0.794

c = 2ζmωn = 158800(N − s/m)

Problem 4. Consider the motion of a rotating unbalanced system constrained to move

vertically. The mass of the block without the rotating eccentric mass (m) is (M − m). Plot
XM
the locus of maxima for versus r (frequency ratio) for different values of damping and
em
interpret the results.

Figure 5: The rotating unbalanced system

Solution 4. Let
XM r2
f= =p
em (1 − r2 )2 + (2ζr)2

6
For local maxima
df
=0
dr
(1 − r2 )2 + (2ζr)2 4r − r2 −4r + 4r2 + 8ζ 2 r = 0

1
=⇒ r = p
1 − 2ζ 2
From the above expression, for maximum value of f
 
1
ζ ϵ 0, √
2
Expression for f maximum at different values of ζ:
1
fmax = p
4ζ 2 (1 − ζ 2 )

7
Function file: non_dim.m
%This function returns for the value of {M|X| / (me)}

function f=non_dim(r,z)
f=(r^2)/sqrt((1-r^2)^2+(2*z*r)^2);
end

Main file: q4.m

% Function non_dim is non dimesional ratio X|M|/(em)

clc;
clear all;
r=linspace(0,10,1000); % linearly distributed 1000 points between 0 to 10
for z=(0.1:0.1:1) % Here z is damping ratio
m=zeros(1,100);
for i=(1:1000)
m(i)=non_dim(r(i),z); % r is the frequency ratio

end
plot (r,m)
hold on
end

r = 1.001:0.001:10;
f_max = r.^2 ./ sqrt(r.^4 - 1);
plot(r, f_max)

Locus of Maxima

𝜁 = 0.1

𝜁 = 0.2

𝜁 = 0.3

𝑟
Problem 5. In a static firing test, a rocket is anchored to a rigid wall by a spring-damper
system, as shown in Fig. 6(a). The thrust acting on the rocket reaches its maximum value
F in a negligibly short time and remains constant until the burnout time t0 , as indicated in
Fig. 6(b). The thrust acting on the rocket is given by F = m0 v, where m0 is the constant
rate at which fuel is burnt, and v is the velocity of the jet stream. The initial mass of
the rocket is M . Its mass at any time t is given by m = M − m0 t, 0 ≤ t ≤ t0 . Given
k = 7.5 × 106 N/m, c = 0.1 × 106 Ns/m, m0 = 10 kg/s, v = 2000 m/s, M = 2000 kg and
t0 = 100 s,

(i) derive the equation of motion of the rocket, and

(ii) find the maximum steady-state 

displacement of the rocket by assuming an average
1
(constant) mass of M − m0 t0 .
2

Figure 6: Static firing test of rocket

Solution 5. (i) The FBD of the rocket for the given arrangement is:

Figure 7: FBD of the rocket during static test firing test

Writing the equation of motion for this, we get:

m(t)ẍ + cẋ + kx = F

where, m(t) = M − m0 t, and F = m0 v. Thus, the final equation of motion is:

(M − m0 t)ẍ + cẋ + kx = m0 v

9
(ii) Assuming the steady-state mass for the rocket as m = M − 12 m0 t0 , the equation of motion
can be re-written as:
1
(M − m0 t0 )ẍ + cẋ + kx = m0 v
2
Putting in the values as given in the problem,

1500ẍ + 1e5ẋ + 7.5e6x = 20000

Maximum steady state deflection for the given case can thus calculated to be:
F 20000
xs (t) = = = 0.002667m
k 7.5e6
Problem 6. Solve the differential equation 4ẍ(t) + 5ẋ(t) + 100x(t) = f (t) for the following
f (t) (assume initial conditions as x(0) = 0 and ẋ(0) = 0):
(i) sin(t) + 2 sin(3t) + 1.5 cos(7t), and

(ii) the periodic function shown in Fig. 8. Use the first three terms of the Fourier series
approximation. Take T = 2 seconds, A = 1 units.

Figure 8: Periodic function

Solution 6. (i) For a forced harmonic vibration: M ẍ + cẋ + kx = F0 sin (ωt)

Steady-state solution is: x(t) = √ F20 /k sin(ωt − ψ)
(1−r )2 +(2ζr)2
where, ψ = tan−1 k−M cω


ω2

Solving given differential equation individually for the three given harmonic terms
and then finally adding all solution.
For sin(t):
x1 (t) = 0.0104 sin(t − 0.052)

For 2 sin(3t):
x2 (t) = 0.0304 sin(3t − 0.0.230)

10
 For 1.5 sin(7t):
x3 (t) = 0.0104 sin(7t + 0.350)

x(t) = 0.0104 sin(t − 0.052) + 0.0304 sin(3t − 0.0.230) + 0.0104 sin(7t + 0.350)

(ii) The Fourier Series expression is:

∞     
X nπt nπt
f (t) = a0 + an cos + bn sin for − t1 < t < t1
n=1
t1 t1
Z t1
1
a0 = f (t) dt
2t1 −t1
1 t1
Z  
nπt
an = f (t)cos dt
t1 −t1 t1
1 t1
Z  
nπt
bn = f (t)sin dt
t1 −t1 t1

Since the function is periodic for interval: (−T /2 < t < T /2) or (−1 < t < 1)
considering this interval for fourier series approximation.
(
−1 t ϵ [−1, −1/2) ∪ (1/2, 1]
f (t) =
1 t ϵ (−1/2, 1/2)

Since the function is even function ∴ bn = 0

Z 1
1
a0 = f (t) dt = 0
2 −1

1 1
Z  
nπt 4A  nπ 
an = f (t)cos dt = sin
1 −1 1 nπ 2
∞     ∞   
X nπt X 4A  nπ  nπt
f (t) = 0 + an cos +0 = sin cos
n=1
1 n=1
nπ 2 1
4 −4 4 −4
=⇒ f (t) = cos(πt) + cos(3πt) + cos(5πt) + cos(7πt) + ...
π 3π 5π 7π
We need only first three terms:
4 −4 4
=⇒ 4ẍ(t) + 5ẋ(t) + 100x(t) = cos(πt) + cos(3πt) + cos(5πt)
π 3π 5π
Now as done in the above part, we get

=⇒ x(t) = 0.0204 cos(πt − 0.254) − 0.0016 cos(3πt + 0.182) + 0.0003 cos(5πt + 0.0883)

11
Problem 7. Determine the response of the single-DOF system: ẍ + kx = f (t) to the
transient excitation shown below (for t > τ0 ) with initial conditions as x(0) = 0 and ẋ(0) = 0
(x is the displacement from mean position).

Figure 9: Variation of forcing function with time.

Solution 7. Fig.9 forcing function can be represented as sum of three forcing functions:

(1)
ẍ1 (t) + kx1 (t) = f1 (t)
F0 t
f1 (t) =
τ0
As discussed in class  
F0 t t cos (ωn t)
x1 (t) = −
k 0.33τ0 0.33τ0
(2)
F0 (t − 0.33τ0 )
f2 (t) = −
τ0
 
F0 t − 0.33τ0 (t − 0.33τ0 ) cos (ωn (t − 0.33τ0 )
x2 (t) = − −
k 0.33τ0 0.33τ0

12
(3)
f3 (t) = −F0
F0
x3 (t) = − (1 − t cos (ωn (t − τ0 ))
k

{ẍ1 (t) + ẍ2 (t) + ẍ3 (t)} + k{x1 (t) + x2 (t) + x3 (t)} = {f1 (t) + f2 (t) + f3 (t)}

=⇒ ẍ(t) + kx(t) = f (t)

x(t) = x1 (t) + x2 (t) + x3 (t)

   
F0 t t cos (ωn t) t − 0.33τ0 (t − 0.33τ0 ) cos (ωn (t − 0.33τ0 )
x(t) = − − −
k 0.33τ0 0.33τ0 0.33τ0 0.33τ0
− (1 − cos (ωn (t − τ0 )))] (9)

13