Plane Stress and Plane Strain

Plane Stress and Plane Strain Equations
We will now develop the four-noded rectangular plane

element stiffness matrix.
This element is an isoparametric formulation of a general
quadrilateral element (see Chapter 10).
This element is also called the bilinear rectangle because of
the linear terms in x and y for the x and y displacement
functions.
The "Q4" symbol represents the element as a quadrilateral
with four corner nodes.
Two advantages of the rectangular element over the triangular
element are:
1. ease of data input and
2. simpler interpretation of output stresses.
CIVL 8/7117 Chapter 6 - Plane Stress/Plane Strain Stiffness Equations - Part 2 1/17
The "Q4" symbol represents the element as a quadrilateral
with four corner nodes.
A disadvantage of the rectangular element is that the simple
linear-displacement rectangle with its associated straight
sides poorly approximates the real boundary condition
edges.
The usual steps outlined in Chapter 1 will be followed to
obtain the element stiffness matrix and related equations.
1. Discretize and Select Element Type
2. Select a Displacement Function - Assume a variation of the
displacements over each element.
3. Define the Strain/Displacement and Stress/Strain Relationships -
use elementary concepts of equilibrium and compatibility.
4. Derive the Element Stiffness Matrix and Equations - Define the
stiffness matrix for an element and then consider the derivation of the
stiffness matrix for a linear-elastic spring element.
5. Assemble the Element Equations to Obtain the Global or Total
Equations and Introduce Boundary Conditions
6. Solve for the Unknown Degrees of Freedom (or Generalized
Displacements) - Solve for the nodal displacements.
7. Solve for the Element Strains and Stresses - The reactions and
internal forces association with the bar element.
8. Interpret the Results
1. Discretize and Select Element Type
Consider the rectangular element shown below (all interior
angles are 90) with corner nodes 1-4 (again labeled
counterclockwise) and base and height dimensions of 2b
and 2h, respectively
{ }
1
1
2
2
3
3
4
4
u
v
u
v
d
u
v
u
v

=
`

)
2. Select a Displacement Function
For a compatible displacement field, the element dis-
placement functions u and v must be linear along each
edge because only two points (the corner nodes) exist
along each edge.
{ }
( , )
( , )
i
u x y
v x y

+ =
`
)
1 2 3 4
5 6 7 8
a a x a y a xy
a a x a y a xy
+ + +
=
`
+ + +
)
There are a total of eight generalized degrees of freedom
(a's) and a total of eight specific degrees of freedom (u
1
, v
1
at
node 1 through u
4
, v
4
at node 4) for the element.
{ }
( , )
( , )
i
u x y
v x y

+ =
`
)
1 2 3 4
5 6 7 8
a a x a y a xy
a a x a y a xy
+ + +
=
`
+ + +
)
We can proceed in the usual manner to eliminate the a's and
obtain:
( )( ) ( )( )
( )( ) ( )( )
1 2
3 4
1
( , )
4
u x y b x h y u b x h y u
bh
b x h y u b x h y u
= + +
( + + + + +

( )( ) ( )( )
( )( ) ( )( )
1 2
3 4
1
( , )
4
v x y b x h y v b x h y v
bh
b x h y v b x h y v
= + +
( + + + + +

These displacement expressions, can be expressed
equivalently in terms of the interpolation functions and
unknown nodal displacements as:
1 1 2 2 3 3 4 4
( , ) u x y Nu N u N u N u = + + +
where:
( )( )
1
4
b x h y
N
bh

=
1 1 2 2 3 3 4 4
( , ) v x y Nv N v N v N v = + + +
( )( )
2
4
b x h y
N
bh
+
=
( )( )
3
4
b x h y
N
bh
+ +
=
( )( )
4
4
b x h y
N
bh
+
=
1
N
1 2
3 4
2
N
1 2
3 4
3
N
1 2
3 4
1 2
3 4
4
N
The shape functions are visually deceiving. There is no
curvature in directions parallel to any side; however, there is a
twist due to the xy term in the element representation.
In expanded form, the equations become:
( )
( )
1
1
2
1 2 3 4 2
1 2 3 4 3
3
4
4
, 0 0 0 0
, 0 0 0 0
u
v
u
u x y N N N N v
v x y N N N N u
v
u
v

(

=
` `
(

)

)
{ } [ ]{ } N d + =
( )
( )
0
1 2 3 4 0
1 2 3 4
0
0
0
0
, 0 0 0 0
, 0 0 0 0 0
0
v
u x y N N N N v
v x y N N N N
v
v

(

=
` `
(
)

)
So that u and v will yield a constant value for rigid-body
displacement, N
1
+N
2
+N
3
+N
4
=1 for all x and y locations
on the element.
For example, assume all the triangle displaces as a rigid body in
the y direction: v =v
0
( )
0 0 1 2 3 4
( , ) v x y v v N N N N = = + + +
1 2 3 4
1 N N N N + + + =
The general definitions of normal and shear strains are:
x y xy
u v u v
x x y x
c c
c c c c
= = = +
c c c c
3. Define the Strain-Displacement andStress-Strain
Relationships
The strains over a two-dimensional element are:
{ }
x
y
xy
u
x
v
y
u v
y x
c
c c

c

c

c
= =
` `
c

)

c c
+

c c
)
Relationships
Substituting our approximation for the displacement gives:
( )
, 1 1 2 2 3 3 4 4 x
u
u N u N u N u N u
x x
c c
= = + + +
c c
, 1, 1 2, 2 3, 3 4, 4 x x x x x
u N u N u N u N u = + + +
where the comma indicates differentiation with respect to that
variable.
Relationships
The derivatives of the interpolation functions with respect to x
are:
( )( )
( )
1,
1
4 4
x
h y
N b x h y
bh x bh
c
= =
c
( )( )
( )
2,
1
4 4
x
h y
N b x h y
bh x bh
c
= + =
c
( )( )
( )
3,
1
4 4
x
h y
N b x h y
bh x bh
+
c
= + + =
c
( )( )
( )
4,
1
4 4
x
h y
N b x h y
bh x bh
+
c
= + =
c
( )( )
( )
4,
1
4 4
y
b x
N b x h y
bh y bh
c
= + =
c
Relationships
The derivatives of the interpolation functions with respect to y
are:
( )( )
( )
1,
1
4 4
y
b x
N b x h y
bh y bh
c
= =
c
( )( )
( )
2,
1
4 4
y
b x
N b x h y
bh y bh
+
c
= + =
c
( )( )
( )
3,
1
4 4
y
b x
N b x h y
bh y bh
+
c
= + + =
c
Relationships
We can write the strains in matrix form as: { } [ ]{ } B d c =
| |
( ) ( )
( ) ( )
( ) ( ) ( ) ( )
0 0
1
0 0
4
h y h y
B b x b x
bh
b x h y b x h y

= +
( ) ( )
( ) ( )
( ) ( ) ( ) ( )
0 0
0 0
h y h y
b x b x
b x h y b x h y
( + +
(
+
(
( + + +

Relationships
From equations, we observe that c
x
is a function of y, c
y
is a
function of x, and
xy
is a function of both x and y.
[ ]
x x
y y
xy xy
D
o c
o c
t

=
` `

) )
{ } [ ][ ]{ } D B d o =
( )
2
1 0
[ ] 1 0
1
0 0 0.5 1
E
D
v
v
v
v
(
(
=
(
(

For plane stress [D] is:
( )( )
1 0
[ ] 1 0
1 1 2
0 0 0.5
E
D
v v
v v
v v
v
(
(
=
(
+
(

For plane strain [D] is:
The stresses are again given as:
4. Derive the Element Stiffness Matrix and Equations
The above relationship requires:
The stiffness matrix can be defined as:
{ } { } [ ] [ ][ ]
h b
T
h b
B D B t dx dx d f

=
} }
[ ] [ ] [ ][ ]
h b
T
h b
k B D B t dx dx

=
} }
For an element of constant thickness, t, the above integral
becomes:
[ ] [ ] [ ][ ]
h b
T
h b
k t B D B dx dx

=
} }
Because the [B] matrix is a function of x and y, integration must
be performed.
The [k] matrix for the rectangular element is now of order 8 x 8.
A numerical evaluation for [k] using b =4 in., h =2 in., t =1 in.,
E =30 x 10
6
psi, and v =0.3.
This double integral was solved using Mathcad.
1.35E+10 5.49E+09 -1.69E+09 -4.22E+08 -6.75E+09 -5.49E+09 -5.06E+09 4.22E+08
5.49E+09 2.45E+10 4.22E+08 9.28E+09 -5.49E+09 -1.22E+10 -4.22E+08 -2.15E+10
-1.69E+09 4.22E+08 1.35E+10 -5.49E+09 -5.06E+09 -4.22E+08 -6.75E+09 5.49E+09
-4.22E+08 9.28E+09 -5.49E+09 2.45E+10 4.22E+08 -2.15E+10 5.49E+09 -1.22E+10
-6.75E+09 -5.49E+09 -5.06E+09 4.22E+08 1.35E+10 5.49E+09 -1.69E+09 -4.22E+08
-5.49E+09 -1.22E+10 -4.22E+08 -2.15E+10 5.49E+09 2.45E+10 4.22E+08 9.28E+09
-5.06E+09 -4.22E+08 -6.75E+09 5.49E+09 -1.69E+09 4.22E+08 1.35E+10 -5.49E+05
4.22E+08 -2.15E+10 5.49E+09 -1.22E+10 -4.22E+08 9.28E+09 -5.49E+09 2.45E+10
[k] =
Steps 5 - 7
Steps 5 through 7, which involve assembling the global stiffness
matrix and equations, determining the unknown nodal
displacements, and calculating the stress, are identical to
those in Section 6.2 for the CST.
However, the stresses within each element now vary in both the
x and y directions.
Numerical Comparison of CST to Q4 Element Models and
Element Defects.
Table 6-1 compares the free end deflection and maximum
principal stress for a cantilevered beam modeled with various all
triangular CST elements or all rectangular Q4 elements.
Element Defects.
Table 6-1 compares the free end deflection and maximum
principal stress for a cantilevered beam modeled with various all
triangular CST elements or all rectangular Q4 elements.
Element Defects.
We observe from the displacement results that the CST element
models produce stiffer models than the actual beam behavior, as the
deflections are predicted to be smaller than classical beam theory predicts.
We also observe that the CST model converges very slowly to the classical
beam theory solution.
This is partly due to the element predicting only constant stress within each
element when for a bending problem; the stress actually varies linearly
through the depth of the beam.
This problem is rectified by using the linear-strain triangle (LST) element as
described in Chapter 8.
Element Defects.
The results indicate that the Q4 element model predicts more accurate
deflection behavior than the CST element model.
The two-row model of Q4 elements yields deflections very close to that
predicted by the classical beam deflection equation, whereas the two-row
model of CST elements is quite inaccurate in predicting the deflection.
As the number of rows is increased to four and then eight, the deflections are
predicted increasingly more accurately for the CST and Q4 element models.
The two-noded beam element model gives the identical deflection as the
classical equation (o =PL
3
/3EI) as expected (see discussion in Section 4.5)
and is the most appropriate model for this problem when you are not
concerned, for instance, with stress concentrations.
Element Defects.
It has been shown in for a beam subjected to pure bending, the CST has a
spurious or false shear stress and hence a spurious shear strain in parts of
the model that should not have any shear stress or shear strain.
This spurious shear strain absorbs energy; therefore, some of the energy that
should go into bending is lost.
The CST is then too stiff in bending, and the resulting deformation is smaller
than actually should be.
This phenomenon of excessive stiffness developing in one more modes of
deformation is sometimes described as shear locking or parasitic shear.
Element Defects.
It should be noted that using a single row of Q4 elements with their linear
edge displacement is not recommended to accurately predict the stress
gradient through the depth of the beam.
Element Defects.
As mentioned previously, the CST element has constant strain and stress
within it, while the Q4 element normal strain, c
x
, and hence the normal stress,
o
x
, is linear in the y direction.
Therefore, the CST is not able simulate the bending behavior nearly as well
as the Q4 element.
The classical beam theory/bending stress equation predicts a linear stress
variation through the depth the beam given by o
x
= -My/I
As shown when comparing the principal stresses for each model, as more
rows are used, the stresses approach the classical bending stress of 20 MPa
with the Q4 approaching the classical solution much faster as indicated by
comparing the two-row solutions for Q4 and CST models.
Element Defects.
This brief description of some of the limitations in using the CST and Q4
elements does not prevent us from using them to model plane stress and
plane strain problems.
It just requires us to use a fine mesh as opposed to a coarse one, particularly
where bending occurs and where in general large stress gradients will
results.
Also, we must make sure our computer program can handle Poisson's ratios
that approach 0.5 (if that is desired, such as in rubber-like materials).
For common materials, such as metals, Poisson's ratio is around 0.3, so
locking should not be of concern.
End of Chapter 6b

Plane Stress and Plane Strain

Uploaded by

Copyright:

Available Formats

Plane Stress and Plane Strain

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Plane Stress and Plane Strain

Uploaded by

Copyright:

Available Formats

Plane Stress and Plane Strain Equations

We will now develop the four-noded rectangular plane

You might also like