Tutorial Questions

Tutorial Question 1-1

Describe the characteristics of a composite and their advantages/disadvantages
when used as structural materials.

Answer:
• Heterogeneous – composed of diverse parts as opposed to metals which are
considered homogeneous

– Due to the multiphase nature, i.e. reinforcements and matrix

– Due to the laminated construction
• Anisotropic – its properties depend on the direction in which they are measured
For example, the stiffness and strength of the fibre reinforced composite
material is much higher in the fibre direction than those in the transverse
direction

1
 Tutorial Questions 1-1
Answer:
• High specific strength and specific stiffness (strength/stiffness to weight ratio)
• Strength/stiffness tailoring
• Exceptional formability and reduced machining
• Resistant to corrosion
• Absorbs radar microwaves
• Tapered sections & compound contours easily accomplished
• Low thermal expansion
• High resistance to fatigue damage

• Expensive raw materials

• Lack of established design allowables
• Galvanic corrosion from improper coupling with metals
• Susceptible to environmental degradation (ageing)
• Poor energy absorption and impact damage
• Linear elastic to failure
• Expensive and complicated inspection methods to detect defects and damages
• May require lightning strike protection

Tutorial Question 1-2

Identify the phases in fibre-reinforced composites and describe their functions
and characteristics.

Answer:
• Matrix
– more compliant than the other phases,
– tougher than the other phases (usually known as the reinforcement),
– supports and binds the reinforcement together,
– provides environmental protection for the reinforcement,
– transmits load from one piece of reinforcement to the other (by shear).
• Fibres
– stiffer and stronger than the matrix,
– provides the composite with its high stiffness and strength properties

2
 Tutorial Question 2-1
How is a symmetry in an anisotropic material used to reduce the number of
independent elements in the material’s stiffness or compliance matrix?

Answer:

Using the principle of symmetry:

A symmetric body experiences only symmetric strains under a symmetric stress,

and only anti-symmetric strains under an anti-symmetric stress.

A symmetric body is under only symmetric stresses when experiencing a

symmetric strain, and only anti-symmetric stresses when experiencing an anti-
symmetric strain.

Applying one symmetric stress/strain to the body and let the anti-symmetric
strain/stress equal to zero, the corresponding element of material’s
compliance/stiffness will be zero.

Tutorial Question 2-2

What are the elastic engineering constants for a full description of the constitutive
relationship of an orthotropic material?

Answer:

E1, E2, E3,

12, 13, 23,

G12, G13, G23

3
 Tutorial Question 2-3
A special type of orthotropic material is called transversely isotropic material, in
which the material exhibits isotropic characteristics in any a plane transverse to a
given direction, uni-directionally fibre reinforced composites being an example of
this type of material.

– How many independent elastic engineering constants are required to define

the constitutive relationship of a transversely isotropic material?

– What are they?

Answer:

E1, E2, 12, 23, G12

[ E 3 = E 2, 13=12, G13=G12, G23= E2/2(1+23)]

Tutorial Question 2-4

For a UD lamina of the material properties as in the Example in the notes, find
all stresses (plane stress state assumed) and strains in materials principal axes as
functions of , when it is subjected to pure shear xy=.

2 12
1
xy xy


Answer:
Stresses in the materials principal axes:

 cos 2  sin 2  2 cos  sin  

   T   
1
T  1 
  sin  2
cos 
2
 2 cos  sin  

  cos  sin  cos  sin  cos 2   sin 2  
 

 1   sin 2  2   sin 2  12   cos 2

4
 Tutorial Question 2-4

 1   sin 2  2   sin 2  12   cos 2

1 2  1  
1    21    21  sin 2
E1 E2  E1 E2 

1 2  1 
 2   12    12   sin 2
E1 E2  E1 E2 

 12 1
 12    cos 2
G12 G12

Tutorial Question 3-1

What are the basic assumptions employed in the formulation of the classical
laminate theory?
Where and why are they required?
What are the possible errors due to these assumptions?

Answer:
Normal to the laminate remains normal and undergoes no extension
Laminae are under plane stress state

Displacement pattern in the thickness direction

Constitutive relationship
To simplify the original 3D problem to a 2D one

Transverse shear strains and stresses are zero

5
 Tutorial Question 3-2
What is the major difference in the formulation of CLT between symmetric and
asymmetric laminates? What are the implications on the solution procedures
of the problem?

Answer:

[B]=0 for symmetric laminates

[B]¹0 for asymmetric laminates

No coupling between in-plane extension and bending for symmetric laminates.

In-plane and bending problems can be tackled separately.

Tutorial Questions 3-3

Find the A, B and D matrices of a [0/+45/45/90] laminate, assuming the
elastic properties as given below:

E1=138 GPa, E2=9 GPa, G12=6.9 GPa and 12=0.3.

The lamina is 0.125mm thick.

Discuss why it is called quasi-isotropic laminate and why it is only called ‘quasi’-
isotropic but not ‘isotropic’?

6
 Solution

B  1  Q'l zl2  zl21  D  1  Q'l zl3  zl31 

4 4 4
A   Q'l zl  zl 1 
l 1 2 l 1 3 l 1

Q 0 Q 45 Q  45 Q 90

 cos2  sin2   2 cos sin 
Q  T QT  T
T    sin2  cos2  2 cos sin 

cos sin  cos sin cos2   sin2  


 E1  12 E2 
1    0 
1  12 21
 12 21 
  138 .8 2.72 0 
Q     21E1 E2 E2
0    2.72 9.05 0  GPa  21   12
1  12 21 1  12 21  E1
   0 0 6.9
 
 0 0 G12 
 

Solution

138 .8 2.72 0   cos2  sin2   2 cos sin 

 
Q 0  Q    2.72 9.05 0 GPa T    sin 
2
cos2  2 cos sin 
 0 0 6.9  cos sin  cos sin cos2   sin2  
 

9.05 2.72 0  0.5 0.5 1

Q 90  2.72 138 .8 0 GPa

T 45  0.5 0.5 1 
 0 0 6.9  0.5  0.5 0 

 45 .22 31 .42 32 .44 

Q  45  T 45 Q T 45 T  31 .42 45 .22 32 .44  GPa
32 .44 32 .44 35 .6 

 45.22 31 .42  32 .44 

Q  45  T  45Q T  45T   31.42 45 .22  32 .44  GPa
  32 .44  32.44 35.6 

7
 Solution

B  1  Q'l zl2  zl21  D  1  Q'l zl3  zl31 

4 4 4
A   Q'l zl  zl 1 
l 1 2 l 1 3 l 1

z0  2t z1  t z2  0 z3  t z4  2 t

A  Q'0   t    2t   Q'45  0   t   Q '45  t  0  Q '90  2t  t 

 Q '0  t  Q '45  t  Q '45  t  Q '90  t

B  1 Q'0   t 2   2t 2  Q'45  02   t 2  Q'45  t 2  02  Q'90  2t 2  t 2 

2


1
2
    
Q'0   3t 2  Q'45   t 2  Q'45  t 2  Q'90  3t 2 
D  1 Q'0   t 3   2t 3  Q'45  03   t 3  Q'45  t 3  03  Q'90  2t 3  t 3 
3
1

 Q'0  7t 3  Q'45  t 3  Q'45  t 3  Q'90  7t 3
3


Solution
A11  A22
29 .79 8.54 0 
 
A   8.54 29 .79 0   10 6 N / m  xy  A12 / A11  0.2866
 
 0 A66  A11 1  xy  2
0 10 .63

  3.04 0  0.507 
B11  B22
 
B    0 3.04  0.507   10 3 N
  B66  B11 1   xy  2
  0.507  0.507 0 

D11  D22
 0.733 0.066 0 
   xy  D12 / D11  0.09
D   0.066 0.733 0  Nm
 
 0 0 0.109  D66  D11 1   xy  2

[B] and [D] do not show isotropic features

8
 Tutorial Questions 3-4
Find the deflection of a square [0/90] laminate as a function of x and y
subjected to the uniaxial tension Nx applied through the mid surface of the
laminate. Assume the side length and the thickness of the laminate are l and t,
respectively.

The axial bending can be suppressed by applying a bending moment Mx. Find
the magnitude and the direction of this Mx. What is the deflection in the middle
of the laminate then if the laminate is supported along the edges parallel to the x-
axis? What are the stresses in the laminate?

Answer:

Cross ply laminate A16 = A26 = B16 = B26 =D16 = D26 =0

Uniform generalised stresses

Tutorial Questions 3-4

Answer:
  x 
0
 N x   A11 A12 0 B11 B12 0
 0  A A22 0 B12 B22 0  0 
   12  y 
0 0 0 A66 0 0 B66   xy0 
     xy0   xy  0
 0   B11 B12 0 D11 D12 0   x 
 0   B12 B22 0 D12 D22 0  y 
    
 0   0 0 B66 0 0 D66   xy 
1
 N x   A11 A12 B11 B12   x0   x0   A11 A12 B11 B12  N x 
          
 0   A12 A22 B12 B22   0y   0y   A12 A22 B12 B22   0 
 
        
 0   B11 B12 D11 D12   x   x   B11 B12 D11 D12  0
          
 0   B12 B22 D12 D22   y   y   B12 B22 D12 D22   0 

2w 2w 2w constants

x   y    xy  2 0
x 2 y 2 xy

1 1
w    x x 2   y y 2  ax  by  c
2 2

9
 Tutorial Questions 3-4
The axial bending can be suppressed by applying a bending moment Mx. Find
the magnitude and the direction of this Mx. What is the deflection in the middle
of the laminate then if the laminate is supported along the edges parallel to the x-
axis? What are the stresses in the laminate?

Answer: Axial bending suppressed x  0

 N x   A11 A12 B11 B12   
0
x
   N x   A11 A12 B12   x 
0
 0  A A22 B12 B22   
0
   12 y      0
     0    A12 A22 B22   y 
M x   B11 B12 D11 D12   0 
 0   B B D   
 0   B12 B22 D12
 
D22   y     12 22 22   y 

1
 x0   A11 A12 B12   N x  M x  B11 x0  B12 y0  D12 y
 0    
 y    A12 A22 B22   0 
   B B D   0  1
 y   12 22 22    w    y y y  l 
2
1
wmiddle   y l l is the length of the square laminate
8

 '   0  z     T T      Q 

Tutorial Questions 3-5

What are the basic assumptions employed in the formulation of the First order
Shear Deformation Theory (FSDT)? What are the differences and similarities as
compared with CLT?

Answer:

– The normal to the reference surface of the laminate still remains straight
– It is allowed to rotate relative to the reference surface
– The normal strain z is negligible. z =0

The FSDT hypothesis keeps most part of the Love-Kirchhoff hypothesis as

introduced in CLT. The only thing dropped is that a normal remains
perpendicular to the deformed reference surface.

10
 Tutorial Questions 3-6
In the Example as show in the notes, what is the relationship between the
deflections obtained from FSDT and the CLT? What is responsible for the trend,
and why?

Answer:

As the thickness/length ratio increases, the difference is getting larger and larger,
FSDT producing higher deflection.

The transverse shear stiffness is responsible for this.

The increase in the thickness will increase the bending stiffness more
significantly than the transverse shear stiffness. A relatively low transverse shear
stiffness means that transverse shear strain will be significant.

The CLT assumes zero transverse shear strain which is equivalent to an infinite
transverse shear stiffness. This makes a stiffer laminate than FSDT. Therefore,
as the thickness (t ) increases, the difference between the FSDT and CLT
increases rapidly.

Tutorial Questions 4-1

Describe the maximum stress and Tsai-Wu failure criteria, using equations to
assist your description as necessary.
State the most important advantages and disadvantages of each of them.
Answer:
• Maximum stress failure criteria

Material fails when any of the following conditions is violated

i i
1 if i ³ 0 1 if i £ 0 i=1, 2, 3
 it*  ic*

i
1 i=23, 13, 12
 i*
– Tensile and compressive strengths  1*t ,  1*c ,  *2 t ,  *2 c ,  3*t ,  3*c
– Shear strengths  *23 ,  13* ,  12*
• Features: Simple, Indicating failure modes, No interactions, not accurate

11
 Tutorial Questions 4-1
Answer:
• Tsai-Wu failure criteria

Material fails when the following conditions is violated

i, j = 1, 2, ..., 6
F  Fij  i j  Fi i  1 (4, 5 and 6: shear stresses)

where 12 strength properties F11, F12, F13, F22, F23, F33, F44, F55, F66 and F1,
F2, F3 must be determined from experiments. Other coefficients Fij =0.

• Features: Interaction, Reliable, Complicated, Failure modes not indicated

Tutorial Questions 4-2

Given the strength properties of a plane stressed lamina as follows

 1*t  1280 MPa  1*c  800 MPa

 2*t  40 MPa  2*c  145 MPa  12*  73 MPa

Use the maximum stress and Tsai-Wu failure criteria to predict whether the
lamina is strong enough to take the following stresses (unit: MPa):

 1  1300  1  300  1   500  1  650

               
 2    0   2    39   2    0   2    22 
   0     0     55     40 
 12     12     12     12   

Having reached your predictions, can you rank your confidence levels on the 4
predictions and explain why.

12
 Solution

F  F11 12  2 F12  1 2  F22  22  F66 122  F1 1  F2 2

F1  1  1*t  1  1*c   4 .6875  10  4

 1*t  1280 MPa

F2  1  2*t  1  2*c  1 . 81034  10  2  1*c  800 MPa

 2*t  40 MPa
F11  1  1*t  1*c  9 .76563  10  7

 2*c  145 MPa

F22  1  2*t  2*c  1 . 72414  10  4
 12*  73 MPa
F66  1   *
12
*
12  1 .87652  10 4

F12 

1
2  2*t 
2
1   *
2t 
F1  F2    2*t 2 F11  F22   5.37109  10 6

Solution
(a) 1=1300 MPa 2=0 MPa 12=0 MPa
1 1300 2 0 12 0
  1.02  0  0
1*t 1280  2*t 40 12* 73
Maximum stress Failure due to longitudinal tension

F  F11 12  2 F12  1 2  F22  22  F66 122  F1 1  F2 2  1 .04

TsaiWu Failure

(b) 1=300 MPa 2=39 MPa 12=0 MPa

1 300  2 39 12 0
  0.234   0.975  0
1*t 1280  2*t 40 12* 73
Maximum stress No failure

F  F11 12  2 F12  1 2  F22  22  F66 122  F1 1  F2 2  1 .04

TsaiWu Failure

13
 Solution
(c) 1=500 MPa 2=0 MPa 12=55 MPa
1 500 2 0 12 55
  0.625  0   0.753
1*c 800  2*t 40 12* 73
Maximum stress No failure

F  F11 12  2 F12  1 2  F22  22  F66 122  F1 1  F2 2  1 .05

TsaiWu Failure

(d) 1=650 MPa 2=22 MPa 12=40 MPa

1 650  2 22 12 40
  0.508   0.55   0.548
1*t 1280  2*t 40 12* 73
Maximum stress No failure

F  F11 12  2 F12  1 2  F22  22  F66 122  F1 1  F2 2  1 .04

TsaiWu Failure

Solution
(1) High confidence on both MS prediction and TW prediction
(2) Medium confidence on MS prediction, high confidence on TW prediction
(3) Low confidence on MS prediction, high confidence on TW prediction
(4) Low confidence on MS prediction, high confidence on TW prediction

The most significant weakness of the maximum stress criterion is the lack of consideration of
the interactions between different stress components.
The consequence is that the prediction from it may be unreliable when strong interactions are
present among stress components in a given stress state.

(1) There is no interaction between stresses, the predictions are reliable, as a result, the
confidence on the ‘failure’ prediction as obtained is high.
(2) There is a small interaction between 1 and 2, a medium confidence for both prediction.
(3) There is a significant interaction between 1 and 12, a low confidence for MS prediction.
(4) There is a significant interaction between all three, a low confidence for MS prediction.

14
 Tutorial Questions 4-3
What are the advantages of constructing failure function from stress invariant? Has
the same argument been used anywhere else before, e.g. in a different subject?
What is an action plane? What is the significance of it?
Answer:

The coordinate system transformation does not affect the result.

Principal stresses in failure criteria for metal material.

Action plane is the plane where failure takes place to produce a fracture surface.

Only the stress components appearing on the plane contribute to the failure.

Tutorial Questions 4-4

What is the requirement on the initial stiffness of the cohesive interface in the
cohesive zone model for delamination? How is the fracture mechanics approach
indirectly included in the cohesive zone model?

Answer:

The initial stiffness of the cohesive interface should be chosen high enough to
ensure a good bonding between elastic layers, but low enough to avoid numerical
ill condition. Usually, a sensitivity study should be conducted for a proper value.

The area under the traction-separation curve is equal to the fracture energy of the
cohesive layer. The fracture energy used in fracture mechanics is thus included in
the cohesive zone model. In the mixed mode cohesive zone model, the damage
evolution law is usually controlled by the crack propagation criteria used in
fracture mechanics.

15
 Tutorial Questions 4-5
What are the similarities between the cohesive zone model for delamination and
the progressive damage models for matrix crack and fibre rupture?

Answer:

Both models use a traction/stress based criterion for damage initiation.

An equivalent separation/displacement is defined and similar damage evolution

laws are defined.

Fracture energies are used to control the completion of the damage.

Tutorial Questions 5-1

What is the 10% rule? How is it applied to very thin laminate?

Answer:

To include at least 10% of both 90º and ±45º plies in the configuration. This
reduces the possibility of splitting, parallel to the fibres, in a unidirectional
laminate

A simple 0º/90º configuration may be used for thin laminates. Inclusion of the
±45º plies may be impractical in thin laminate, minimum of 4 plies are
needed for balance and symmetry.

16
 Tutorial Questions 5-2
Should the plies with same fibre orientation be blocked together when designing
a composite laminates and why?

Answer:

No.

The plies should be finely dispersed rather than in blocks.

This increases the damage (non-critical) tolerance of the laminate and ensures
that any damage is more finely dispersed.

Tutorial Questions 5-3

Using the carpet plot data provided design a minimum thickness carbon-
epoxy laminate to carry a loading of Nx = 735kN/m, Ny = 895kN/m and Nxy
= 290kN/m. Specify the lay-up and verify that that laminate is capable of
withstanding these loads.

Answer:

No unique solution.

Refer to the example in the notes for the design procedure

17
 Tutorial Questions 6-1
A part of the outcome from the mechanics of materials approach reproduces the
result of Voigt’s bound while the other part does the Reuss’ bound. Using the
mechanics of materials approach and from the viewpoint of the minimum total
potential and complementary potential energy principles, assess the accuracy of all
the predictions from Voigt and Reuss’ bounds and explain why agreements and
disagreements arise.
Answer:

According to Mechanics of Materials, a uniform strain state in the fibre direction

gives a good approximation to that in the composite loaded uniaxially in the fibre
direction, from which E1 is measured. The same strain state is assumed for Voigt
bound. Hence, Voigt’s bound is expected to give reasonable estimate of E1.

When the composite is under transverse stress in order to measure E2, a uniform
stress state is a reasonable assumption from the viewpoint of Mechanics of
Materials. This is in consistence with assumption in the Reuss bound. Hence
gives a reasonable estimate of E2.

Because of the approximation, agreements and disagreements arise.

Tutorial Questions 6-2

Most effective properties of a composite fall between their counterparts of the
constituents, which would suggest a weighted rule of mixture produce good results.
However, 23 of UD composites seems to behave out of the step as described, as
can be seen through the data shown in the examples in the notes. Describe the
mechanism which is responsible for that.

Answer: 2

There is the tendency that the volume of the material will remain unchanged.
Under the uni-axial tension, say in 2 direction, the contraction of the material in
the 1-direction is very small due to the high stiffness of the fibre. Therefore the
contraction of the material in the 3-direction must be very large to keep the volume
of the material unchanged. That means the value of 23 of UD composites is
controlled by the above mechanism, instead of a weighted rule of mixture.

18
 Tutorial Questions 6-3
Select any three failure mechanisms as described in the notes, compare them with
their counterparts in homogenous, isotropic materials of similar ductility.
Comment on the similarities and dissimilarities.
Answer:
Failure of Composites due to Longitudinal Tension
– Matrix cracks/yields  fibre fracture/yield
– Fibre fractures/yields  matrix cracking/yield
Metal: fracture or yield

Compression Along Fibres

– Micro buckling (leads to), Kinking band (leads to), Bending fracture
Metal: fracture or yield or buckling

Interfacial Failure
– Failure of interface between fibre and matrix
– Involved in other failure modes at some stage to some extent
Metal: No this kind of failure

Usually, the failure of composites are very much different from that of metal.

19