Module 5: FEM for Two and Three Dimensional Solids

Lecture 2: Linear Strain Triangle

7

5.2.1 Element Stiffness Matrix for LST

In case of CST, it is observed that the strain within the element remains constant. Though, these
elements are able to provide enough information about displacement pattern of the element, but it is
unable to provide adequate information about stress inside an element. This limitation will be
significant enough in regions of high strain gradients. The use of a higher order triangular element
called Linear Strain Triangle (LST) significantly improves the results at these areas as the strin
inside the element is varying. The LST element has six nodes (Fig. 5.2.1) and hence, twelve degrees
of freedom. Thus the displacement function can be chosen as follows.
u   0  1 x   2 y   3 x 2   4 xy   5 y 2
(5.2.1)
v  a6  a7 x  a8 y  a9 x 2  a10 xy  a11 y 2

Fig. 5.2.1 Linear strain triangle element

Therefore, the element strain matrix is obtained as

¶u
ex = = a1 + 2a 3 x + a 4 y
¶x
¶v (5.2.2)
ey = = a 8 + a10 x + 2a11y
¶y
¶v ¶u
g xy = + = (a 2 + a 7 ) + (a 4 + 2a 9 )x + (2a 5 + a10 )y
¶x ¶y
 8

In the area coordinate system as discussed in module 3, lecture 3 we can write the shape function for
the six node triangular element as
N1 = L1 (2L1 - 1) N 2 = L 2 (2L 2 - 1) N 3 = L3 (2L3 -1)
(5.2.3)
N 4 = 4L1L 2 N 5 = 4L 2 L3 N 6 = 4L3L1
The displacement (u,v) of any point within the element can be represented in terms of their nodal
displacements with the use of interpolation function.
6
u   Ni ui
i 1
6
(5.2.4)
v   Ni vi
i 1

Using eq.(5.2.4) we can rewrite eq.(5.2.2) as,

 u1 
u 
 2
u3 
 
 N1 N 2 N 3 N 4 N 5 N 6  u4 
 0 0 0 0 0 0  u 
 x x x x x x  5
 N1 N 2 N 3 N 4 N 5 N 6  u6 
  0 0 0 0 0 0  
 y y y y y y   v1 
 N1 N 2 N 3 N 4 N 5 N 6 N1 N 2 N 3 N 4 N 5 N 6   v2 
  
 y y y y y y x x x x x x   v3 
v 
 4
 v5 
v 
 6
Or,
   B  d  (5.2.5)
Where,
 N1 N 2 N 3 N 4 N 5 N 6 
 0 0 0 0 0 0 
 x x x x x x 
 N1 N 2 N 3 N 4 N 5 N 6  (5.2.6)
 B   0 0 0 0 0 0 
 y y y y y y 
 N1 N 2 N 3 N 4 N 5 N 6 N1 N 2 N 3 N 4 N 5 N 6 
 
 y y y y y y x x x x x x 

Using Chain rule,

N1 N1 L1 N1 L2 N1 L3
 .  .  .
x L1 x L2 x L3 x
As discussed in module 3, lecture 1, we can write the above expression as,
 9

N1 b1 N1 b2 N1 b3 N1

 .  .  .
x 2 A L1 2 A L2 2 A L3
N1 b1
 .  4 L1  1
x 2 A
Similarly we can evaluate expressions for other terms and can be written as,
N1 b1 N 2 b2 N 3 b3
 .  4 L1  1  .  4 L2  1  .  4 L3  1
x 2 A x 2 A x 2 A
N 4 N 5 N 6
 4  L2b1  L1b2   4  L3b2  L2b3   4  L1b3  L3b1 
x x x
And,
N1 a1 N 2 a2 N3 a3
 .  4 L1  1  .  4 L2  1  .  4 L3  1
y 2 A y 2 A y 2 A
N 4 N5 N 6
 4  L2 a1  L1a2   4  L3a2  L2 a3   4  L1a3  L3a1 
y y y
Where,
a1  x2  x3 a2  x3  x1 a3  x1  x2
b1  y2  y3 b2  y3  y1 b3  y1  y2

The stiffness matrix of the element is represented by,

 k     B   D  B  d 
T
(5.2.7)


The, [D] matrix is the constitutive matrix which will be taken according to plane stress or plane
strain condition. The nodal strain and stress vectors are given by,
 n    x1  x 2  x 3  y1  y 2  y 3  xy1  xy 2  xy 3 
T
(5.2.8)

 n    x1  x 2  x 3  y1  y 2  y 3  xy1  xy 2  xy 3 
T
(5.2.9)

  Bn1   0 
 n     0  Bn 2  d  (5.2.10)
 Bn 2   Bn1  
Referring to section 3.3.1, using proper values of area coordinates in [B] matrix, one can find
 3b1 b2 b3 4b2 0 4b3 
1 
 Bn1    b1 3b2 b3 4b1 4b3 0  (5.2.11a)
2A
 b1 b 3b3 0 4b2 4b1 
And,
 10

 3a1  a2  a3 4a2 0 4a3 

 Bn 2    a1 3a2 a3 4a1 4a3 0 
1 
(5.2.11b)
2A
 a1 a 3a3 0 4a2 4a1 
Thus, the element stiffness can be evaluated by putting the values from eq. (5.2.11) in eq. (5.2.7).

5.2.2 Nodal Load Vector for LST

Similar to 3-node triangular element, the load will be lumped at each node which can be computed
using the earlier expression,

{F} = ò [ N ] {F }dG + ò [ N ] {F }dW

T T
S
G W (5.2.12)
G W

And for element with constant thickness,

{F} = t ò [ N ] {F }ds + t ò [ N ] {F }dA
T T
S
G W
(5.2.13)
S A

5.2.3 Numerical Example using CST

Determine the displacements at the nodes for the following 2D solid continuum considering a
constant thickness of 25 mm, Poisson’s ratio,  as 0.25 and modulus of elasticity E as 2 x 105
N/mm2. The continuum is discritized with two CST plane stress elements.

Fig. 5.2.2 Geometry and discretization of the continuum

The element 1 is connected with node 1, 3 and 4 and let assume its Cartesian coordinates are (x1, y1),
(x3, y3) and (x4, y4) respectively. If we consider nodes 1, 3 and 4 are similar to node 1, 2 and 3 in
eq.(5.1.9) then the [B] can be written as
 11

 1  2 3 0 0 0 
 B    0 0 0  1  2  3 
1 
2A
  1  2  3 1  2  3 
By introducing values of β & γ discussed in previous lecture note, we can get value of [B] as
 0 1 1 0 0 0 
1 
 B  0 0 0 3 0 3 
1500 
 3 0 3 0 1 1
For plain stress problem, putting the values of E and µ one can find the following values.
 
1  0  16 4 0 
E   4  104  
 D   1 0 
1   2  3  4 16 0 
1    0 0 6 
0 0 
 2 
Therefore the stiffness matrix for the element 1 will be
 k 1  tA  B   D  B 
T

Putting values of t, A, [B] & [D]we will get,

 750 0 -750 0 -250 250 

0 222.2222 -222.2222 -166.6667 0 166.6667 

 -750 -222.2222 972.2222 166.6667 250 -416.6667 
 k 1  4 103  0 -166.6667 166.6667 2000 0 -2000

 
 -250 0 250 0 83.3333 -83.3333 
 
 250 166.6667 -416.6667 -2000 -83.3333 2083.3333
Similarly element 2 is connected with nodes 1, 2 and 3 and global coordinates of these nodes are (x1,
y1), (x2, y2) and (x3, y3) respectively. For this element, by proceeding in a similar manner to element
1 we can calculate [B] matrix as,
 1 1 0 0 0 0 
1 
 B  0 0 0 0 3 3
1500 
 0 3 3 1 1 0 
Hence, the elemental stiffness matrix becomes,
 12

 222.2222 -222.2222 0 0 166.6667 -166.6667 

-222.2222 972.2222 -750 250 -416.6667 166.6667 

0 -750 750 -250 250 0 
 k 2  4 10  
3

0 250 -250 83.3333 -83.3333 0 
166.6667 -416.6667 250 -83.3333 2083.3333 -2000 
 
-166.6667 166.6667 0 0 -2000 2000 

By assembling the stiffness matrices into global stiffness matrix [K],

972.2222 -222.2222 0 -750 0 166.6667 -416.6667 250 
-222.2222 972.2222 -750 0 250 -416.6667 166.6667 0 
 
0 -750 972.2222 -222.2222 -416.6667 250 0 166.6667 
 
3 -750 0 -222.2222 972.2222 166.6667 0 250 -416.6667 
 K  410  0 250 -416.6667 166.6667 2083.3333 -83.3333 0 -2000 
 
166.6667 -416.6667 250 0 -83.3333 2083.3333 -2000 0 
-416.6667 166.6667 0 250 0 -2000 2083.3333 -83.3333 
 
250 0 166.6667 -416.6667 -2000 0 -83.3333 2083.3333

Now, applying equation  F    K d  , the following expression can be written.

 Fu1  972.2222 -222.2222 0 -750 0 166.6667 -416.6667 250  u1 
F  -222.2222 972.2222 -750 0 250 -416.6667 166.6667 0  u 
 u2    2
 Fu 3  0 -750 972.2222 -222.2222 -416.6667 250 0 166.6667  u3 
    
Fu 4  3 -750 0 -222.2222 972.2222 166.6667 0 250 -416.6667  u4 
    
  v1 
4 10
0
 Fv1  
250 -416.6667 166.6667 2083.3333 -83.3333 0 -2000
 
 Fv 2  166.6667 -416.6667 250 0 -83.3333 2083.3333 -2000 0  v2 
  -416.6667  
F
 v3  166.6667 0 250 0 -2000 2083.3333 -83.3333  v3 
 
F  2083.3333 v4 
 v4  250 0 166.6667 -416.6667 -2000 0 -83.3333
Putting boundary conditions u1  v1  u2  u4  v4  0 and adopting elimination technique for
applying boundary condition we get expression,
 0  972.2222 250 0  u3 
    
 0   4  10   250 2083.3333 2000   v2 
3

  2083.3333  v3 
 25000   0 -2000
Solving the above expression, the unknown nodal displacements may be obtained as follows.
v2  0.0606 mm, u3  0.0156 mm and v3  0.0612 mm.