FFD Notes David Collins Section
FFD Notes David Collins Section
FFD Notes David Collins Section
Georges Chahni
With figures from David Collins’ slides
December 2021
Contents
1 Stress and Strain 2
1.1 Engineering Stress-Strain Relations . . . . . . . . . . . . . . . 2
1.2 Shear Stress-Strain Relations . . . . . . . . . . . . . . . . . . 2
1.3 Stress Matrices . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.4 Mohrs Circle . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
2 Failure Criterion 4
2.1 The Tresca Criterion . . . . . . . . . . . . . . . . . . . . . . . 4
2.2 Von-Mises Yield Criterion . . . . . . . . . . . . . . . . . . . . 4
5 Strain - Enhanced 6
5.1 Tensors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
5.1.1 Engineering strains . . . . . . . . . . . . . . . . . . . . 6
5.1.2 Relating Tensor Strain to Stress . . . . . . . . . . . . 9
1
1 Stress and Strain
1.1 Engineering Stress-Strain Relations
The Fundamental Stress equations are:
F orce
Engineering stress : δ =
Area
∆L
Engineering Strain : ε =
L0
Combining the two equations we get:
δ
Y oungs M odulus : E =
ε
There are also other Moduli such as the shear and bulk modulus but these
will not be too relevant.
2
Figure 1: 3-D Stress State
3
Figure 2: Mohrs Circle for a Stress State
2 Failure Criterion
It is useful to be able to formulate criterions under which we can theoret-
ically evaluate whether or not a material has plastically deformed/yielded,
and two common ones are the Tresca and Von-Mises Yield Criteria. These
criteria account for a combination of shear and uniaxial stresses.
4
√
It tells us that the maximum shear would be equal to σy 3.
Solving for the above values of λ gives you the principle stresses, but of
course you can keep doing the mohrs circle stuff.
5
4.3 Plane Stress
This form of stress behaviour simply means stress is only exhibited in one
plane e.g.
σ11 0 0 0 0 0
0 σ22 0 or 0 σ22 0 (7)
0 0 0 0 0 σ33
5 Strain - Enhanced
So far we’ve defined normal strain quite well, but we must also take into
account tensor/shear strain. To evaluate tensor strains we’ll have to slightly
redefine terms like engineering strain as I will show,
5.1 Tensors
5.1.1 Engineering strains
σ ∆L
We had previously defined engineering strain as ϵ = = but this
E L
definition will not cut it. In fact we will now call engineering strain e and
define it as
∂u
∆L dx + dx − dx ∂u
ex = = ∂x =
L dx ∂x
.
For the following figure the above equation applies where ∂u is the change
in displacement(u):
6
∂ui
We can say eij = where x = (x1 , x2 , x3 ), U = (U1 , U2 , U3 )
∂xj
Note: x1 , x2 , x3 and (x, y, z) are interchangeable.
Normal strains don’t change much and are then defined as:
∂u
exx = ,
∂x
∂u
eyy = , (8)
∂y
∂u
ezz =
∂z
Shear strains are then defined a bit oddly from the below figure:
BB ′ ∂u
exy = =
AB ∂y
′
(9)
DD ∂v
eyx = =
AD ∂x
Tensor strains will steal the previous namesake of engineering strain and
will now be named ϵ, and defined as:
εxx = exx ,
εyy = eyy , (10)
εzz = ezz
7
We can use this to define tensor strain as the average of symmetric
engineering strains in the form:
1
εij = (eij + eji ) ,
2 (11)
1
wij = (eij − eji )
2
εxy = 0 for case 1 and ωij = 0 for case 2. The next leap is to notice that
e=ε+ω
1 1
Some quick maths gives: εxy = (exy + eyx ) = (γ + 0); and we come
2 2
up with the following Rule: Tensor strain must be half of engineering strain:
8
We can now define tensor strain using the above as the following matrix:
∂u 1 ∂u ∂V 1 ∂u ∂w
+ +
εxz ∂x 2 ∂y ∂x 2 ∂z ∂x
εxx εxy
1 ∂u ∂v ∂v 1 ∂V ∂W
εij = εyx εyy εyz = + +
εzx εzy 2 ∂y ∂x ∂y 2 ∂z ∂y
εzz 1 ∂u ∂w
1 ∂v ∂w
∂w
2 ∂z + ∂x +
2 ∂z ∂y ∂z
(13)
1 − 2ν
εx + εy + ε z = (σx + σy + σz ) (16)
E
Substituting one of the equations in equation set (10) into equation 12
gives the following result:
E vE
σxx = εxx + (εxx + εyy + εzz ) (17)
1+v (1 + v) (1 − 2v)
which can generalised as
E vE
σij = εij + ϵkk δij (18)
1+v (1 + v) (1 − 2v)
9
where:
(
1 if i = j (Only N ormal)
εkk = (εxx + εyy + εzz ) and δij = (19)
̸ j (Only Shear)
0 if i =
vE
Lame′ s Constant : λ = (20)
(1 + v) (1 − 2v)
and noticing
E
2G = (21)
1−v
We can now define Hookes law as:
10