AOE 3024: Thin Walled Structures

Solutions to Homework # 8

Consider a thin-walled cantilever beam as shown in the attached figure. At the tip
of the beam, a bending moment M = 3000 N-m is applied at an angle θ with respect to
the positive x-axis. Assume that the height of the web is h, the width of the flange is
b, thickness of the web is t1 , and the thickness of the flange t2 . Take b = h = 120 mm
and t1 = t2 = 3 mm.
Flange:
Section 2 y

b
E F E F
t2
G B A G B A
yG
xG h
x h1
Web: t1
Section 1

D C D C

Fig. 1 Beam’s Cross Section

For simplicity, let’s define a new parameter β as the ratio of thicknesses,

t2
β= (1)
t1
This will help us to easily linearize the equations for this thin-walled beam. For our problem
β = 1. In thin-wall assumption it is reasonable to ignore higher order thickness terms. Basically,
substitute t2 for β t1 , expand and ignore all quadratic and higher order terms in t1 . (We could have
also substituted for t1 .) Also, note that for thin-walled beam: points A and F are located at A’=
F; points E and point B are located at B’= E, and points C and D are located at C’= D.
E F b
A’
B A B’ t2

h
⇒
t1

D C C’

Fig. 2 Thin-walled assumption

1 of 7

Homework # 8
Part A. For this beam, determine the resultant deflection vector at the tip and the
mid-point as the angle θ is varied between 0◦ to 180◦ . Plot your results.
The first step is to calculate the centroid. Let’s place the origin at point B’ and calculate xG
and y G .

Section xi yi Ai
1 0 −h/2 (h)(t1 )
2 b/2 0 (b)(t2 )

b2 β t1 h2 t1
X X X  
Ai = (h + b β) t1 x i Ai = y i Ai = −
2 2

b2 β
P
x i Ai
xG = P = = 30 mm
Ai 2h + 2bβ
h2
P  
y i Ai
yG = P =− = −30 mm
Ai 2h + 2bβ

(Compare to xG = 31.1203 mm and y G = −31.1203 mm without using thin-walled assumption)

The minus sign indicates that the centroid is located below point B’. We will relocate the axis
as shown in Figure 1 and take discard the sign:

xG = 30 mm y G = 30 mm (2)

Now we proceed to calculate the second moments of area. Please see Mathematica file for details.
Only the linearized expressions are given here.

Section Ixci Iyc i Ixc yc i

1 (t1 h31 )/12 (t31 h1 )/12 ≈ 0 0

2 (t2 b3 )/12 (t32 b)/12 ≈ 0 0

Second Moment of area Ixx using thin-walled assumption is

  2
h
Ixx1 = Ixc1 + A1 − − (y G ) (3a)
2
Ixx2 = Ixc2 + A2 (y G )2 (3b)

h4 t1 4 b h3 β t1
Ixx = Ixx1 + Ixx2 ≈ + = 1.08 × 106 mm4 (3c)
12 h + 12 b β 12 h + 12 b β

(Compare to Ixx = 1.04 × 106 mm4 without using thin-walled assumption)

2 of 7

Homework # 8
 Second Moment of area Iyy using thin-walled assumption is

Iyy1 = Iyc 1 + A1 [− (xG )]2 (4a)

 2
b
Iyy2 = Iyc 2 + A2 − xG (4b)
2

4 b3 h β t1 b4 β 2 t1
Iyy = Iyy1 + Iyy2 ≈ + = 1.08 × 106 mm4 (4c)
12 h + 12 b β 12 h + 12 b β

(Compare to Iyy = 1.04 × 106 mm4 without using thin-walled assumption)

Second Moment of area Ixy using thin-walled assumption is

  
h
Ixy1 = Ixc yc 1 + A1 − − (y G ) [− (xG )] (5a)
2
 
b
Ixy2 = Ixc yc 2 + A2 [(y G )] − xG (5b)
2

b2 h2 β t1
Ixy = Ixy1 + Ixy2 ≈ = 0.648 × 106 mm4 (5c)
4h + 4bβ

(Compare to Ixy = 0.623803 × 106 mm4 without using thin-walled assumption)

Now we need to decompose the applied moment

Mx = M cos θ My = −M sin θ (6)

where M is the applied moment at an angle θ from the x-axis. Note that the bending moment is
uniform everywhere. (Note that the minus sign in My is to be consistent with the derivation of Eq.
9.17 of your text) Now using equation 9.17 from Megson’s text
Mx Ixy − My Ixx
u00 (z) = 2

(7)
E Ixx Iyy − Ixy

My Ixy − Mx Iyy
v 00 (z) = 2

(8)
E Ixx Iyy − Ixy
(Note that for the present problem the right hand side of the above equations is constant and only a
function of θ)
Integrating Eq. (7) respect to z:
Z
0
u (z) = u00 (z) dz + A (9a)
Mx Ixy − My Ixx
Z
0
u (z) = 2

dz + A (9b)
E Ixx Iyy − Ixy
Mx Ixy − My Ixx
u0 (z) = 2

z+A (9c)
E Ixx Iyy − Ixy
(9d)
3 of 7

Homework # 8
 Integrating Eq. (9) respect to z:
Z
u(z) = u0 (z) dz + B (10a)
Mx Ixy − My Ixx
Z
u(z) = 2

z dz + A z + B (10b)
E Ixx Iyy − Ixy
Mx Ixy − My Ixx z 2
u(z) = 2

+ Az + B (10c)
E Ixx Iyy − Ixy 2

The integration constants A and B are found applying the boundary conditions:

0
u (z) = 0 ⇒ A=0 (11)

z=0
u(z) = 0 ⇒ B=0 (12)

z=0

Therefore, the horizontal component of deflection using thin-walled assumption is

( )
Mx Ixy − My Ixx z2
u(z) = 2

(13a)
E Ixx Iyy − Ixy 2

3 M (3 b2 β cos θ + h (h + 4 b β) sin θ) 2
= z (13b)
2 E b3 h β (h + b β) t1

6.51042 × 10−6 cos θ + 10.8507 × 10−6 sin θ z 2 mm, for z in [mm]

= (13c)

(Compare to u(z) = (6.74673 × 10−6 cos θ + 11.2532 × 10−6 sin θ) z 2 without using thin-walled as-
sumption.) Also, recall that β = 1.

Similarly, the vertical component of deflection is

( )
My Ixy − Mx Iyy z2
v(z) = 2

(14a)
E Ixx Iyy − Ixy 2

3 M (b (4 h + b β) cos θ + 3 h2 sin θ) 2
= − z (14b)
2 b h3 (h + b β) t1

−10.8507 × 10−6 cos θ − 6.51042 × 10−6 sin θ z 2 mm, for z in [mm]

= (14c)

(Compare to v(z) = (−11.2532 × 10−6 cos θ − 6.74673 × 10−6 sin θ) z 2 without using thin-walled as-
sumption)

The magnitude and angle of the deflection vector is

p p
δt (z) = [u(z)]2 + [v(z)]2 = z 2 × 10−5 1.60123 + 1.41285 sin (2θ) mm, for z in [mm] (15)
   
−1 v(z) 180 −1 1
αt = tan = tan −0.6 − deg
u(z) π 0.9375 + 1.5625 tan θ
4 of 7

Homework # 8
(Note that the angle does not depend on the location in z)
At the tip, z = L:
utip = 14.6484 cos θ + 24.4141 sin θ mm (16)

vtip = −24.4141 cos θ − 14.6484 sin θ mm (17)

p
δtip = 810.623 + 715.256 sin(2 θ) mm (18)
 
180 −1 1
αtip = tan −0.6 − deg (19)
π 0.9375 + 1.5625 tan θ
At the middle, z = L/2:
umid = 3.66211 cos θ + 6.10352 sin θ mm (20)

vmid = −6.10352 cos θ − 3.66211 sin θ mm (21)

p
δmid = 50.6639 + 44.7035 sin(2 θ) mm (22)
 
180 −1 1
αmid = tan −0.6 − deg (23)
π 0.9375 + 1.5625 tan θ

δ HmmL
Magnitude of the resultant deflection vector

At tip, z=L
40

30

20
At mid-point, z=L/2

10

θ Hdeg L
25 50 75 100 125 150 175

α Hdeg L
Angle of the resultant deflection vector with the x- axis

75
50
25
θ Hdeg L
25 50 75 100 125 150 175
-25
-50
-75

Fig. 3 Resultant deflection vector

5 of 7

Homework # 8
 Part B. For both the tip and the mid-point of the beam, plot the angle that the
deflection vector makes with the neutral axis as the angle θ is varied.
The neutral axis is where the bending stress is zero. Therefore, we use equation 9.6 from your
text
My Ixx − Mx Ixy Mx Iyy − My Ixy
σzz = 2
x+ 2
y (24a)
Ixx Iyy − Ixy Ixx Iyy − Ixy

3 M (−3 b2 h2 β cos θ − h4 sin θ − 4 b h3 β sin θ)

= x (24b)
b3 h3 β (h + b β) t1
3 M (4 b3 h β cos θ + b4 β 2 cos θ + 3 b2 h2 β sin θ)
+ y
b3 h3 β (h + b β) t1

= (−2604.17 cos θ − 4340.28 sin θ) x + (4340.28 cos θ + 2604.17 sin θ) y MPa (24c)

Setting the stress to zero, we get an equation for the neutral axis:
My Ixx − Mx Ixy Mx Iyy − My Ixy
0 = 2
x+ 2
y
Ixx Iyy − Ixy Ixx Iyy − Ixy

My Ixx − Mx Ixy
y = − x (25)
Mx Iyy − My Ixy

2604.17 cos θ + 4340.28 sin θ

y = x
4340.28 cos θ + 2604.17 sin θ
Now, the slope of the above equation is the neutral axis angle
 2
h (3 b2 β cos θ + h (h + 4 b β) sin θ)

180 −1
ψ= tan deg (26)
π b2 β (b (4 h + b β) cos θ + 3 h2 sin θ)

The angle that the deflection vector makes with the neutral axis is given by α −ψ. The variation
of α − ψ as a function of θ is shown in Figure 4. Note that the difference is π/2 for all values of θ.
The deflection vector is thus normal to the axis.

6 of 7

Homework # 8
 ψ
α, ψ Hdeg L α

75
50

θ Hdeg L
25

25 50 75 100 125 150 175

-25
-50
-75 90°°

α−ψ Hdeg L
100
75
50

θ Hdeg L
25

25 50 75 100 125 150 175

-25
-50
-75
-100

Fig. 4 Resultant deflection angle about the neutral axis as angle θ is varied. Note that the
difference is always 90◦ .

7 of 7

Homework # 8