Advanced Theory of Vibrations (AE 721)

Assignment No. 4

November 21, 2020

1 Problem Statement
Given a simply supported
 πx  beam with constant E, I, and L acted upon by an external force
f (x, t) = sin sin (Ωt), find w (x, t), M (x, t) and σxx (x, t). The initial conditions of
L
the problem are w (x, 0) = w0 (x) = 0 and ẇ (x, 0) = ẇx (x) = 0.

2 Natural Frequencies and Mode Shapes

The free vibration response of a simply supported beam made with constant E and I is
governed by

∂ 2 w (x, t) 4
2 ∂ w (x, t)
+ c =0
∂t2 ∂x4
(1)
EI
c2 =
ρA
The boundary conditions for a simply supported beam in terms of w (x, t) are given by:

w (x = 0, t) = 0

∂ 2 w 

=0
∂x2 (x=0,t)
(2)
w (x = L, t) = 0

∂ 2 w 

=0
∂x2 (x=L,t)
The free vibration solution can be found using the method of separation-of-variables as

w (x, t) = W (x) T (t) (3)

In terms of W (x), the boundary conditions become

1
 W (x = 0) = 0

d2 W 

=0
dx2 x=0
(4)
W (x = L) = 0

d2 W 

=0
dx2 x=L

Substituting the assumed solution w (x, t) = W (x) T (t) in to the governing differential
equations leads to

c2 d4 W 1 d2 T
= − = −ω 2 (5)
W dx4 T dt2
Here the choice of separation constant, ω 2 , is made based on the experience that the nat-
ural frequencies comes from the temporal equation. Separating the spatial and temporal
equations yields

d4 W
− β 4W = 0
dx4

d2 T (6)
2
+ ω2T = 0
dt
ω2 ρAω 2
where, β 4 = =
c2 EI
The solution of temporal equation is given by

T (t) = A cos ωt + B sin ωt (7)

and A and B are constants that will eventually be determined by the specified initial
conditions after being combined with spatial solutions using orthogonality of mode shapes.
The solution of the spatial equation can be expressed as

W = C1 cos βx + C2 sin βx + C3 cosh βx + C4 sinh βx (8)

Applying the first two boundary condition (first two expressions of Equation 4) on spatial
solution gives

C1 + C3 = 0
(9)
−β 3 C1 + β 3 C3 = 0
which implies that C1 = C3 = 0. The spatial solution now becomes

W = C2 sin βx + C4 sinh βx (10)

Applying the last two boundary condition (last two expressions of Equation 4) on spatial
solution gives

2
 C2 sin (βL) + C4 sinh (βL) = 0
(11)
−C2 β 2 sin (βL) + C4 β 2 sinh (βL) = 0
Adding the above equations gives

C4 1 + β 2 sinh βL = 0


(12)
Since sinh βL and (1 + β 2 ) cannot be zero for positive values of β and L, C4 must be zero.
Subtracting the two expressions in Equation 11 gives

C2 1 + β 2 sin βL = 0


(13)
For non-trivial solutions of governing differential equation, sin βL must vanish. The zeros
of sin (βL) are nπ, with n = 1, 2, 3, · · · .

(βL)n = nπ n = 1, 2, 3, · · ·
nπ (14)
βn = n = 1, 2, 3, · · ·
L
Using Equation 6, we can find the natural frequencies in terms of βn .
s
n2 π 2 EI
ωn = 2 (15)
L ρA
Note that C1 = C3 = C4 = 0. If the value of C2 corresponding to βn is denoted as (C2 )n ,
we have

Wn (x) = (C2 )n sin βn x (16)

where, Wn (x) are the mode shapes of pinned-pinned beam. The solution corresponding
to n−th value of βn is

wn (x, t) = (C2 )n sin βx (An cos ωn t + B sin ωn t)

(17)
wn (x, t) = A∗n sin βx cos ωn t + Bn∗ sin βx sin ωn t
The general or total solution can be expressed as
∞
X
w (x, t) = A∗n sin βL cos ωn t + Bn∗ sin βL sin ωn t (18)
n=1

Orthogonality of Mode Shapes

The principle of orthogonality of mode shapes states that

Rl
 Wn Wm dx = 0 n 6= m



0
(19)
 Rl L
 Wn Wm dx = n=m


2
0

3
Summary
n2 π 2
r
EI
Natural frequencies: ωn = 2
L ρA

Mode shapes: Wn (x) = sin βn x

3 Forced Response
The forced response of a beam can be determined using the mode summation principle.
For this, the deflection of the beam is assumed as
∞
X
w (x, t) = Wn (x) qn (t) (20)
n=1

where Wn (x) is the n−th mode shape which satisfies the spatial differential equation
and qn (t) is the generalized coordinate in the n−th mode. By substituting the assumed
solution in to the forced vibrations equation, we obtain

∂ 2 w (x, t) ∂ 4 w (x, t)
ρA + EI = f (x, t)
∂t2 ∂x4
∞ ∞
∂2 X ∂4 X
ρA Wn (x) q n (t) + EI Wn (x) qn (t) = f (x, t) (21)
∂t2 n=1 ∂x4 n=1

∞ ∞
X d2 qn (t) X d4 Wn (x)
ρAWn (x) + qn (t) EI = f (x, t)
n=1
dt2 n=1
dx4

d4 Wn (x)
From Equation 6, EI = ρAωn2 Wn (x). Substituting in to the above equation, we
dx4
have
∞ ∞
X d2 qn (t) X
ρAWn (x) 2
+ qn (t) ρAωn2 Wn (x) = f (x, t) (22)
n=1
dt n=1

By multiplying the above equation by Wm (x), integrating from 0 to L, and using principle
of orthogonality of mode shapes, we have

ZL
d2 qm 2
2
+ ω 2 qm = f (x, t) Wm (x) dx (23)
dt ρAL
0

Since m is a dummy symbol, we can replace it any other symbol, say n.

ZL
d2 qn 2 2
+ ω q n = f (x, t) Wn (x) dx = Qn (t) (24)
dt2 ρAL
0

Substituting the expressions for Wn (x) and f (x, t) in to Qn (t), we have

4
 ZL
2 πx nπx
Qn = sin sin Ωt sin dx
ρAL L L
0
(25)
ZL
2 πx nπx
Qn = sin Ωt sin sin dx
ρAL L L
0

L
Because of orthogonality principle, the integral in the above expression will be equal
2
when n = 1, else it will be zero. So the above equation becomes
 d2 qn 2 1
 dt2 + ωn qn = ρA sin Ωt n = 1

d2 qn 2
(26)
 dt2 + ωn qn = 0
 n = 2, 3, 4, · · ·

This is essentially the same as the equation of motion of an undamped single degree-of-
freedom system. Its total solution can be expressed as

qn (t) = An cos ωn t + Bn sin ωn t + Particular Solution (27)

where the first two terms on the right-hand side represent the transient or free vibra-
tions (resulting from the initial conditions) and the third term denotes the steady-state
vibration (resulting from the forcing function).
For n = 2, 3, 4, · · · , both forcing function and initial conditions are zero, therefore, intu-
itively, we know that qn = 0 for n = 2, 3, 4, · · · . However, we will derive it mathematically
in the following.
 1
qn (t) = An cos ωn t + Bn sin ωn t + ρA(ωn2 −Ω2 ) sin Ωt n = 1

(28)
qn (t) = An cos ωn t + Bn sin ωn t
 n = 2, 3, 4, · · ·

The final step is determine arbitrary constants An and Bn using initial conditions. From
Equation 20, the forced response of a beam is given by
∞
X
w (x, t) = Wn (x) qn (t) (29)
n=1

The next step is to convert the initial condition specified on w (x, t) to initial condition
on q (t).

Displacement Initial Condition

The displacement initial condition is given as:

w (x, 0) = w0 (x) = 0 (30)

Substituting in Equation 29 gives:
∞
X
w (x, 0) = Wn (x) qn (0) = 0 (31)
n=1

5
By multiplying the above equation by Wm (x) , integrating from 0 to L, and using principle
of orthogonality of mode shapes, and replacing dummy variable m with n we have

ZL ∞ Z
X
L

w (x, 0) Wm (x) dx = Wm (x) Wn (x) qn (0) dx

0 n=1 0

ZL
L
w (x, 0) Wm (x) dx = qn (0) (32)
2
0

ZL
2
qn (0) = w (x, 0) Wn (x) dx
L
0

Since w (x, 0) = 0, qn (0) = 0.

Velocity Initial Condition

The velocity initial condition is given as:

ẇ (x, 0) = w0 (x) = 0 (33)

Substituting in Equation 29 gives:
∞
X
ẇ (x, 0) = Wn (x) q˙n (0) = 0 (34)
n=1

By multiplying the above equation by Wm (x) , integrating from 0 to L, and using principle
of orthogonality of mode shapes, and replacing dummy variable m with n we have

ZL ∞ Z
X
L

ẇ (x, 0) Wm (x) dx = Wm (x) Wn (x) q̇n (0) dx

0 n=1 0

ZL
L
ẇ (x, 0) Wm (x) dx = q̇n (0) (35)
2
0

ZL
2
q̇n (0) = w (x, 0) Wn (x) dx
L
0

Since ẇ (x, 0) = 0, q̇n (0) = 0.

Total Response
Now we have initial conditions on qn (t) (qn (0) = 0 and q̇n (0) = 0), we can solve Equation
28. Using these initial conditions, we have
  
2 Ω
qn = ρAL(ωn2 −Ω2 ) sin Ωt − ωn sin ωn t n=1


(36)


 q n (t) = 0 n = 2, 3, 4, · · ·

6
Using the above values, the total response of the beam is given by
∞
X
w (x, t) = Wn (x) qn (t)
n=1

w (x, t) = W1 (x) q1 (t)

  (37)
2 Ω
w (x, t) = sin Ωt − sin ω1 t sin β1 x
ρAL (ω12 − Ω2 ) ω1
 
2 Ω πx
w (x, t) = 2 2
sin Ωt − sin ω1 t sin
ρAL (ω1 − Ω ) ω1 L

The bending moment distribution is given by:

∂ 2w
M (x, t) = EI
∂x2
(38)
2π 2 EI
 
Ω πx
M (x, t) = − sin Ωt − sin ω1 t sin
ρAL3 (ω12 − Ω2 ) ω1 L
The bending stress is given by:

∂ 2w
σxx (x, y, t) = −Ey
∂x2
(39)
2Eyπ 2
 
Ω πx
σxx (x, y, t) = − sin Ωt − sin ω1 t sin
ρAL3 (ω12 − Ω2 ) ω1 L

Solution
 
2 Ω πx
w (x, t) = sin Ωt − sin ω1 t sin
ρAL (ω12 − Ω2 ) ω1 L

2π 2 EI
 
Ω πx
M (x, t) = − 3 2 2
sin Ωt − sin ω1 t sin
ρAL (ω1 − Ω ) ω1 L
(40)
2Eyπ 2
 
Ω πx
σxx (x, y, t) = − 3 2 2
sin Ωt − sin ω1 t sin
ρAL (ω1 − Ω ) ω1 L
s
2
π EI
ω1 =
L2 ρA