Introduction to Finite Elements

Introduction to the Stiffness (Displacement) Method: Analysis of a system of springs

FEM analysis scheme Step 1: Divide the problem domain into non overlapping regions (elements) connected to each other through special points (nodes) Step 2: Describe the behavior of each element Step 3: Describe the behavior of the entire body by putting together the behavior of each of the elements (this is a process known as assembly)

F1x k1

F2x k2

F3x

Problem Analyze the behavior of the system composed of the two springs loaded by external forces as shown above Given F1x , F2x ,F3x are external loads. Positive directions of the forces are along the positive x-axis k1 and k2 are the stiffnesses of the two springs

F1x k1

F2x k2

F3x

Solution Step 1: In order to analyze the system we break it up into smaller parts, i.e., elements connected to each other through nodes F1x 1 Node 1 d1x Element 1 k1 F2x 2 d2x Element 2 k2 F3x 3 d3x x

Unknowns: nodal displacements d1x, d2x, d3x,

F1x 1 Node 1 d1x

k1

F2x 2

k2

F3x 3

Element 1

d2x

Element 2

d3x

Solution Step 2: Analyze the behavior of a single element (spring)

2002 Brooks/Cole Publishing / Thomson Learning

Two nodes: 1, 2 d Nodal displacements: d 1x 2x f Nodal forces: f 1x 2x Spring constant: k

 2002 Brooks/Cole Publishing / Thomson Learning

,z ,y ) and global (x,y,z) coordinate systems Local (x

Behavior of a linear spring (recap)

F k 1 d k d k

x F

F = Force in the spring d = deflection of the spring k = stiffness of the spring Hookes Law F = kd

 f 1x
2002 Brooks/Cole Publishing / Thomson Learning

f 2x

Hookes law for our spring element Eq (1) Force equilibrium for our spring element (recap free body diagrams)

" k (d !d ) f 2x 2x 1x

#f "0 f 1x 2x " !f " !k (d !d ) $f 1x 2x 2x 1x

Collect Eq (1) and (2) in matrix form

Eq (2)

Element force vector

Element stiffness matrix

( ( + + k k 1 . f d % % % 1x 1x % f "kd * ' " / * ' , - k k -% % f 2x % d 2x % 0 ) & ) & # $ " $ ! " ! # " ! Element nodal #
displacement vector
f k d

Note T "k 1. The element stiffness matrix is symmetric, i.e. k 2. The element stiffness matrix is singular, i.e.,

) " k2 ! k2 " 0 det ( k

The consequence is that the matrix is NOT invertible. It is not possible to invert it to obtain the displacements. Why? The spring is not constrained in space and hence it can attain multiple positions in space for the same nodal forces e.g.,

+ %f1x ( % 1 2 - 2. +1 ( +- 2( * ' " / * '"* ' , % )f 2x % & 0- 2 2 - )2 & ) 2 & ( + 1 2 - 2. +3( +- 2( %f 1x % * ' " / * '"* ' , % )f 2x % & 0- 2 2 - )4 & ) 2 &

Solution Step 3: Now that we have been able to describe the behavior of each spring element, lets try to obtain the behavior of the original structure by assembly Split the original structure into component elements Element 2 Element 1 k1 1 2 k2 2
(1) d (1) f 1x 1x
(1) (1) d f 2x 2x (2) (2) d f 1x 1x

3
(2) (2) d f 2x 2x

(1) ( (1) ( + + k k 1 . d %f % % 1 1 1x 1x % * (1) ' " / * (1) ' , - k1 k1 - % % f 2x % d 2x % 0 ) & ) & # $ " $ ! # " ! # " ! (1 ) (1) f k (1) d

(2) ( (2) ( + + k k 1 . d %f % % 2 2 1x 1x % * (2) ' " / * (2) ' , - k 2 k 2 -% % f 2x % d 2x % 0 ) & ) & $! # # " ! #$" " ! (2)
(2) f k (2) d

Eq (3)

Eq (4)

To assemble these two results into a single description of the response of the entire structure we need to link between the local and global variables. Question 1: How do we relate the local (element) displacements back to the global (structure) displacements? F1x k2 F2x F3x k1 x 1 Node 1 d1x Element 1 2 d2x Element 2 3 d3x

(1) " d d 1x 1x (1) " d (2) " d d 2x 1x 2x (2) " d d

2x 3x

Eq (5)

Hence, equations (3) and (4) may be rewritten as

(1) ( + 1 k1 - k1 . +d1x ( %f 1x % * ' * (1) ' " / , % )f 2x % & 0- k1 k1 - )d 2x &

(2) ( + f % 1x % 1 k 2 * (2) ' " / % )f 2x % & 0- k 2

- k 2 . +d 2x ( * ' k2 , - )d 3x &

Or, we may expand the matrices and vectors to obtain

(1) ( 1 k1 ! k1 0. +d ( +f 1x 1x %d % % (1) % / , *f 2x ' " /- k1 k1 0, * 2x ' % % %0 % / 0 d , 0 0 3x ) & 0 ) & # " ! "$$ ! # " ! #$$
f
(1) e

(1) e

0 . + d1x ( + 0 ( 10 0 %f (2) % / %d % , * 1x ' " /0 k 2 ! k 2 , * 2x ' %f % % (2) % /0 - k d , k 3 x & ) 2x & ) 2 2 -# ! # " ! 0 #$$ " $$ ! "
(2)e f (2)e k d

Eq (6)
(1) e k (1) e f

Eq (7)

Expanded element stiffness matrix of element 1 (local) Expanded nodal force vector for element 1 (local) Nodal load vector for the entire structure (global)

Question 2: How do we relate the local (element) nodal forces back to the global (structure) forces? Draw 5 FBDs F1x k2 F2x F3x k1 x 1 A
B

d1x

2 C d2x
2

D3

d3x
3

F1x

(1) f 1x

(1) f 2x

2002 Brooks/Cole Publishing / Thomson Learning

F2x

(2) f 1x

(2) f 2x

F3x

(1) " 0 At node 1 : F1x - f 1x (1) ! f (2) " 0 At node 2 : F2x - f 2x 1x (2) " 0 At node 3 : F - f
3x 2x

In vector form, the nodal force vector (global)

(1) ( + F1x ( + f 1x % % % (1) (2) % F " *F2x ' " *f 2x # f1x ' %F % % f (2) % ) 3x & ) 2x &
Recall that the expanded element force vectors were
(1) e f

(1) ( +f +0 ( 1x ( 2)e (2) % %f % (1) % " *f 2x ' and f " * 1x ' %0 % %f (2) % ) 2x & ) &

Hence, the global force vector is simply the sum of the expanded element nodal force vectors + F (

% % (1) e ( 2) e F " *F2x ' " f # f %F % ) 3x &

1x

But we know the expressions for the expanded local force vectors from Eqs (6) and (7)
(1) e (1)e ( 2)e (2)e f " k d and f "k d

Hence
(1) e ( 2)e (1)e (2)e (1)e (2)e 7 " k d # k d " 5k # k 4 F"f #f 2d 3 6

F"Kd
F " Global nodal force vector d " Global nodal displacement vector K " Global stiffness matrix " sum of expanded element stiffness matrices

For our original structure with two springs, the global stiffness matrix is

0 . 1 k 1 ! k 1 0 . 10 0 , # /0 k , ! K"/ k k 0 k 1 2 2, / 1 , / / / 0 0 0, 0 - k2 k2 , 0 0 #$$ "$$ ! #$$ "$$ !

(1) e k (2)e k

1 k1 "/ /- k 1 / 0 0

! k1 k1 # k 2 - k2

0 . ! k2 , , k2 , -

NOTE 1. The global stiffness matrix is symmetric 2. The global stiffness matrix is singular

The system equations

F"Kd ! k1 k1 # k 2 - k2

imply

+ F1x ( 1 k1 % % / *F2x ' " /- k1 %F % / 0 ) 3x & 0

0 . +d1x ( % % , ! k 2 , *d 2x ' %d % k2 , - ) 3x &

F1x " k1d1x ! k1d 2x $ F2x " -k1d1x # (k1 # k 2 )d 2x ! k 2 d 3x F3x " -k 2 d 2x # k 2 d 3x
These are the 3 equilibrium equations at the 3 nodes.

F1x 1 A d1x

k1

F2x
B

k2

F3x
D3

x d3x
3

2 C d2x
2

F1x

(1) f 1x

(1) f 2x

2002 Brooks/Cole Publishing / Thomson Learning

F2x

(2) f 1x

(2) f 2x

F3x

(1) " 0 At node 1 : F1x - f 1x F2x " -k1d1x # (k1 # k 2 )d 2x ! k 2 d 3x (1) (2) !f "0 At node 2 : F2x - f 2x 1x " ! k1 8d1x ! d 2x 9 # k 2 8d 2x ! d 3x 9 (2) "0 At node 3 : F3x - f 2x (1) # f (2) "f 2x 1x (2) F3x " -k 2 8d 2x ! d 3x 9 " f 2x

(1) F1x " k1 8d1x ! d 2x 9 " f 1x

Notice that the sum of the forces equal zero, i.e., the structure is in static equilibrium. F1x + F2x+ F3x =0 Given the nodal forces, can we solve for the displacements? To obtain unique values of the displacements, at least one of the nodal displacements must be specified.

Direct assembly of the global stiffness matrix

Global F1x 1 d1x Local 1

(1) d (1) f 1x 1x

k1

F2x 2

k2

F3x 3

Element 1

d2x

Element 2

d3x Element 2 k2

Element 1 k1

2
(1) (1) d f 2x 2x

2
(2) (2) d f 1x 1x

3
(2) (2) d f 2x 2x

Node element connectivity chart : Specifies the global node number corresponding to the local (element) node numbers ELEMENT Node 1 Node 2 1 2 1 2 2 3 Local node number Global node number

Stiffness matrix of element 1

Stiffness matrix of element 2

( 2) 1 k2 k "/ 0- k 2

(1)

1 k1 - k1 . d1x "/ , k k 1 - d2x 0 1

d1x d2x

d2x d3x

- k 2 . d2x , k 2 - d3x

Global stiffness matrix d2x d3x d1x

1 k1 K"/ /- k 1 / 0 0

- k1 k1 # k 2 - k2

0 . d1x - k2 , , d2x k2 , - d3x

Examples: Problems 2.1 and 2.3 of Logan

Example 2.1
22 3 4

2002 Brooks/Cole Publishing / Thomson Learning

Compute the global stiffness matrix of the assemblage of springs shown above d2x d3x d4x d1x !1000 0 0 . d1x 1 1000
/!1000 1000# 2000 8 9 / K" / 0 !2000 / 0 0 0 !2000 0 , , d2x 8 2000#30009 !3000, d3x , 3000 - d !3000
4x

Example 2.3
3

2002 Brooks/Cole Publishing / Thomson Learning

Compute the global stiffness matrix of the assemblage of springs shown above
1 k1 / K " /-k1 / 00 -k1 k1 # k 2 # k 3 - 8 k 2 # k3 9 . , - 8 k 2 # k3 9, 8 k 2 # k3 9 , 0

Imposition of boundary conditions

Consider 2 cases Case 1: Homogeneous boundary conditions (e.g., d1x=0) Case 2: Nonhomogeneous boundary conditions (e.g., one of the nodal displacements is known to be different from zero) Homogeneous boundary condition at node 1 k2=100N/m k1=500N/m 1 2 Element 2 Element 1 d1x=0 d2x

F3x=5N 3 d3x

System equations 0
0 . 1 d1x . 1 F1x . 1 500 -500 /-500 600 -100 , / d , " / 0 , / , / 2x , / , / -100 100 , 0 0 -/ 0 d3 x , - / 05, Global Stiffness Nodal Nodal load disp matrix vector vector Note that F1x is the wall reaction which is to be computed as part of the solution and hence is an unknown in the above equation

Writing out the equations explicitly -500d 2x " F1x Eq(1) Eq(2) 600d 2 x ! 100d3 x " 0 Eq(3) !100d 2 x # 100d3 x " 5

Eq(2) and (3) are used to find d2x and d3x by solving
1 600 !100 . 1 d 2 x . 10 . / !100 100 , / d , " /5 , 0 - 0 3x - 0 1 d 2 x . 1 0.01 m . $/ ,"/ , d 0.06 m 0 3x - 0

NOTICE: The matrix in the above equation may be obtained from the global stiffness matrix by deleting the first row and column
0 . 1 500 -500 /-500 600 -100 , / , / -100 100 , 0 0 1 600 !100 . / !100 100 , 0 -

Note use Eq(1) to compute F1x =-500d 2x " !5 N

NOTICE: 1. Take care of homogeneous boundary conditions by deleting the appropriate rows and columns from the global stiffness matrix and solving the reduced set of equations for the unknown nodal displacements. 2. Both displacements and forces CANNOT be known at the same node. If the displacement at a node is known, the reaction force at that node is unknown (and vice versa)

Imposition of boundary conditionscontd.

Nonhomogeneous boundary condition: spring 2 is pulled at node 3 by 0.06 m) k2=100N/m k1=500N/m x 1 3 2 Element 2 Element 1 d3x=0.06m d1x=0 d2x

System equations 0
0 . 1 d1x . 1 F1x . 1 500 -500 /-500 600 -100 , / d , " / 0 , / , / 2x , / , / -100 100 , 0 0 -/ 0 d3 x , - / 0 F3 x , 0.06 Note that now F1x and F3x are not known.

Writing out the equations explicitly

-500d 2x " F1x

Eq(1)

600d 2 x ! 100(0.06) " 0 Eq(2) !100d 2 x # 100(0.06) " F3 x Eq(3)

Now use only equation (2) to compute d2x

600d 2 x " 100(0.06) $ d 2 x " 0.01m

Now use Eq(1) and (3) to compute F1x =-5N and F3x=5N

Recap of what we did Step 1: Divide the problem domain into non overlapping regions (elements) connected to each other through special points (nodes) Element

"k d ) Step 2: Describe the behavior of each element ( f

nodal displacement vector

Step 3: Describe the behavior of the entire body (by assembly). This consists of the following steps 1. Write the force-displacement relations of each spring in Global expanded form
e e d f "k

nodal displacement vector

Recap of what we didcontd. 2. Relate the local forces of each element to the global forces at the nodes (use FBDs and force equilibrium). Finally obtain
e F " :f

F"Kd
Where the global stiffness matrix

K " :k

Recap of what we didcontd.

Apply boundary conditions by partitioning the matrix and vectors

1K11 K12 .+d1 ( +F1 ( /K K ,*d ' " *F ' 0 21 22 -) 2 & ) 2 &
Solve for unknown nodal displacements

K22d2 " F2 ! K21d1

Compute unknown nodal forces

F1 " K11d1 # K12d2

Physical significance of the stiffness matrix F1x 1 Element 1 k1 F2x 2 Element 2 k2 F3x 3 d3x x

d1x d2x In general, we will have a stiffness matrix of the form (assume for now that we do not know k11, k12, etc) The finite element force-displacement relations:
1 k11 /k / 21 / 0 k 31

1 k11 K"/ /k 21 / 0k 31
k12 k 22 k 32

k12 k 22 k 32

k13 . k 23 , , k 33 , -

k13 . + d1 ( + F1 ( % % % % k 23 , , *d 2 ' " *F2 ' % % % % k 33 , - )d 3 & )F3 &

Physical significance of the stiffness matrix The first equation is Force equilibrium equation at node 1

k11d1 # k12 d 2 # k13d 3 " F1

Columns of the global stiffness matrix What if d1=1, d2=0, d3=0 ?
F1 " k11 F2 " k 21 F3 " k 31

While nodes 2 and 3 are held fixed Force along node 1 due to unit displacement at node 1 Force along node 2 due to unit displacement at node 1 Force along node 3 due to unit displacement at node 1

Similarly we obtain the physical significance of the other entries of the global stiffness matrix

Physical significance of the stiffness matrix In general Force at node i due to unit displacement at node j k ij = keeping all the other nodes fixed This is an alternate route to generating the global stiffness matrix e.g., to determine the first column of the stiffness matrix F1 1 d1 Element 1 k1 F2 2 d2 Element 2 k2 F3 3 d3 Set d1=1, d2=0, d3=0 x

Find F1=?, F2=?, F3=?

Physical significance of the stiffness matrix For this special case, Element #2 does not have any contribution. Look at the free body diagram of Element #1

(1) d 1x
(1) f 1x

(1) d 2x
k1
(1) f 2x

(1) " k (d (1) ! d (1) ) " k (0 ! 1) " ! k f 2x 1 2x 1x 1 1 (1) " !f (1) " k f 1x 2x 1

Physical significance of the stiffness matrix Force equilibrium at node 1 F1

(1) " k F1 =f 1x 1 (1) f 1x

Force equilibrium at node 2 F2

(1) f 2x (1) " ! k F2 =f 2x 1

F1 = k1d1 = k1=k11 F2 = -F1 = -k1=k21 F3 = 0 =k31

Of course, F3=0

Physical significance of the stiffness matrix Hence the first column of the stiffness matrix is
+ F1 ( + k1 ( % % % % * F2 ' " *! k1 ' %F % % 0 % ) 3& ) &

To obtain the second column of the stiffness matrix, calculate the nodal reactions at nodes 1, 2 and 3 when d1=0, d2=1, d3=0 Check that
+ F1 ( + ! k1 ( % % % % F " k # k * 2' * 1 2' % F % % !k % 2 & ) 3& )

Physical significance of the stiffness matrix To obtain the third column of the stiffness matrix, calculate the nodal reactions at nodes 1, 2 and 3 when d1=0, d2=0, d3=1 Check that
+ F1 ( + 0 ( % % % % * F2 ' " *! k2 ' %F % % k % ) 3& ) 2 &

Steps in solving a problem Step 1: Write down the node-element connectivity table linking local and global displacements Step 2: Write down the stiffness matrix of each element Step 3: Assemble the element stiffness matrices to form the global stiffness matrix for the entire structure using the node element connectivity table Step 4: Incorporate appropriate boundary conditions Step 5: Solve resulting set of reduced equations for the unknown displacements Step 6: Compute the unknown nodal forces