1

Lecture 1: Stiffness of Truss Members

4.1.1 Introduction
Analysis of frame structures can be carried out by the approach of stiffness method.
However, such types of structures can also be analyzed by finite element method. A unified
formulation will be demonstrated based on finite element concept in this module for the
analysis of frame like structures. A truss structure is composed of slender members pin
jointed together at their end points. Truss element can resist only axial forces (tension or
compression) and can deform only in its axial direction. Therefore, in case of a planar truss,
each node has components of displacements parallel to X and Y axis. Planar trusses lie in a
single plane and are used to support roofs and bridges. Such members will not be able to
carry transverse load or bending moment. The major benefits of use of truss structures are:
lightweight, reconstructable, reconfigurable and mobile. Configuration of few standard truss
structures are shown in Fig. 4.1.1.

Fig. 4.1.1 Configuration of various truss structures

4.1.2 Element Stiffness of a Truss Member

Since, the truss is an axial force resisting member, the displacement along its axis only will
be developed due to axial load. Therefore, using Pascals triangle, the displacement function
of truss member for development of shape function can be expressed as:

u ( x ) = 0 + 1 x =[1 x ] 0
1

(4.1.1)

Fig. 4.1.2 Axial force on the member along X axis

Applying boundary conditions as shown in Fig. 4.1.2:

At x= 0, u(0)= u 1 and at x=L, u(L) = u 2
Thus, 0 = u1 and 1 =

u 2 u1
Therefore,
L .

x
x

u ( x ) = 1 u1 + u 2 = [N ]{u}
L
L
Here, N is the shape function of the element and is expressed as:

[N ] = 1 x

x
L

L
So we get the element stiffness matrix as
=
[k ]

[ B ] [ D ][ B ] d
d [N ] 1 1
=
Where, [B ] =

(4.1.2)

(4.1.3)

(4.1.4)

L L
So, the stiffness matrix will become:
1
L
L
1
T
= [B ] E [B ]Adx = AE L
0
0
1 L

L
dx

1
AE 1 1
=
dx
L
L 1 1

Thus, the stiffness matrix of the truss member along its member axis will be:

[k ] =

AE 1 1
L 1 1

4.1.3 Element Stiffness of Truss Member with Varying Cross Section

(4.1.5)

3
Now, let us find the stiffness matrix of a pin-jointed member of length L with respect to local
axis, having cross sectional areas A1 and A2 at the two ends of the member as shown in the
figure below.

Fig. 4.1.3 Member with varying cross section

From the above figure, the cross sectional area at a distance of x from left end can be
expressed as:
A A1
(4.1.6)
Ax = A1 + 2
x
L
As it is a pin-jointed member, the displacement at any point may be expressed in terms of
u N1u1 + N 2u2 .
nodal displacement as=
Similarly the cross sectional area at any point may be represented in terms of the cross
sectional area of the two ends. Thus Ax = N 1 A1 + N 2 A2
x
x
Where the shape functions are: N1 =
1 ; N2 =
L
L
Now, the strain may be written as:

x =

N 2
u N1
1
1
=
u1 +
u 2 = u1 + u 2 = 1 [ 1
x
L
L
x
x
L

u1

1] = [B ]{u}

u 2

(4.1.7)

As the stress is proportional to strain according to Hooks law, the stress-strain relationship
will be as follows:
u1

1] = E [B ]{u}

u 2

E
x = E x = [ 1
L

(4.1.8)

Now the strain energy may be expressed as

L

1
1
1
T
T
T
T
U = x x dv = x E x Ax dx = {u} [B ] E [B ]{u}Ax dx
2V
20
20

Applying Castiglianos theorem, the force will become:

(4.1.9)

L
E
U
T
{F } =
= [B ] E [B ]{u}Ax dx = 2
L
{u} 0

[ 1

1] [ 1

u1

1]Ax dx = [k ]{d }
u 2
(4.1.10)

Thus, the stiffness matrix will be:

[k ] =

1
E
L2 1

1 L
A1 + A2 A1

L
1 0

A A1 2
E 1 1

A1 x + 2
x
x dx = 2

2L
L 1 1
0

1 1
A2 A1
E 1 1
E
(
)
A
+
=
A
+
A
1
1
2
1 1
2 2 L
L 1 1

(4.1.11)

4.1.4 Generalized Stiffness Matrix of a Plane Truss Member

Let us consider a member making an angle with X axis as shown in the figure below. By
resolving the forces along local X and Y direction, the following relations are obtained.
Fx1 Fx1 cos + Fy1 sin
=
Fx 2 Fx 2 cos + Fy 2 sin
=
Fy1 =
Fx1 sin + Fy1 cos

(4.1.12)

Fy 2 =
Fx 2 sin + Fy 2 cos

Where, Fx1 and Fx 2 are the axial forces along the member axis X . Similarly, Fy1 and Fy 2 are
the forces perpendicular to the member axis X .

Fig. 4.1.4 Inclined truss member

5
The relationship expressed in eq. (4.1.12) can be rewritten in matrix form as follows:

Fx1 cos

Fy1 sin
=
Fx 2 0

Fy 2 0

sin
cos
0
0

0
0
cos
sin

0 Fx1

0 Fy1

sin Fx 2

cos Fy 2

(4.1.13)

Now, the above equation can be expressed in short as:

{F } = [T ]{F }

(4.1.14)

Here, [T] is called transformation matrix. This relates between the global (, axis) and
member axis (, axis). Similarly, the relations of nodal displacements between two
coordinate systems may be written as:

{d } = [T ]{d }

(4.1.15)

Again, the equation stated in (4.1.5) can be generalized and expressed with respect to the
member axis including force and displacement vector as:

Fx1
1

Fy1 AE 0
=
Fx 2 L 1

Fy 2
0

0 1
0 0
0 1
0 0

0 u1
0 v1

0 u2

0 v2

(4.1.16)

Where, the nodal forces in Y direction are zero. The above equation may also be expressed in
short as:

{F } = [k ]{d }

(4.1.17)

Where, the matrices in the above equation are written with respect to the member axis. Now,
eq. (4.1.17) can be rewritten with the use of eq. (4.1.14) and (4.1.15) as given below.

Or,

[T ]{F } = [k ][T ]{d }

(4.1.18)

{F } = [T ]1 [k ][T ]{d }

(4.1.19)

6
Here, the transformation matrix [T] is orthogonal, i.e., [T]-1 is equal to [T]T. Therefore, from
the above relationship, the generalized stiffness matrix can be expressed as:

[k ] = [T ]T [k ][T ]

(4.1.20)

Thus,
cos
sin
[ k ] = 0

sin

0
co s
0
0
cos
0
sin

0
1

0 AE 0
sin L 1

co s
0

0 1
0 0
0 1
0 0

0 cos
0 sin
0 0

0 0

sin
0
co s
0
0
cos
0
sin
(4.1.21)

Or,
cos 2

sin cos
[k ] = AE
L cos 2

sin cos

sin cos

cos 2

sin 2

sin cos

sin cos

cos

sin 2

sin cos

sin cos

sin 2

sin cos

2
sin

(4.1.22)

The above stiffness matrix can be used for the analysis of two-dimensional truss problems.

0
0
sin

co s

7
Lecture 2: Analysis of Truss
4.2.1 Element Stiffness of a 3 Node Truss Member

Fig. 4.2.1 3-node truss member

Here, the displacement function using Pascals triangle can be expressed as:

0

u ( x ) = 0 + 1 x + 2 x = 1 x x 1

2
2

(4.2.1)

Applying boundary conditions:

At x= 0, u(0)= u 1 , x=L/2, u(L/2) = u and at x=L, u(L) = u
2

And solving for 0 , 1 and 2

0 = u1 , = 3u1 + 4u2 u3 and 2 =

1
L

2u1 4u 2 + 2u 3
L3

Therefore,
3x 2 x 2
4x 4x2
x 2x2
u ( x ) = 1 + 2 u 1 +
2 u2 + + 2 u 3 =
L
L
L

L
L L

[ N ]{u}

(4.2.2)

Here, N is the shape function of the element and is expressed as:

3 x 2 x 2 4 x 4 x 2
N
=
[ ] 1 + 2 2
L
L L
L

x 2x2
+ 2
L L

(4.2.3)

Now, the element stiffness matrix can be written as

=
[k ]

[ B ] [ D ][ B ] d
T

(4.2.4)

d [ N ] 3 4x
= +
Where, [ B ] =
d x L L2

4 8x

L L2

1 4x
+
L L2

So, the stiffness matrix will be:

=
[k ]

[ B ] [ D ][ B ] d = [ B ] E [ B ] Adx
T

8
3 4
+ 2

4 8
3 4
=
2 + 2


0
1
4

+ 2
16 2 24
9+

40 32 2
= 2 12 +
2
0

2
16
16

4 8

1 4
+ 2

40 32 2
2

64 64 2
16
+ 2

24 32 2
4 +
2

12 +

16 16 2
+ 2

24 32 2
4 +
2

8 16 2
1
2

(4.2.5)

After integrating the above equation, the stiffness matrix of the 3-node truss member will
become:
7 8 1
AE
[] = 8 16 8
(4.2.6)
3L
1 8 7
4.2.2 Worked Out Example
Analyze the truss shown below by finite element method. Assume the cross sectional area of
the inclined member as 1.5 times the area (A) of the horizontal and vertical members. Assume
modulus of elasticity is constant for all the members and is E.

Fig. 4.2.2 Plane truss

9
Solution
The analysis of truss starts with the numbering of members and joints as shown below:

Fig. 4.2.3 Numbering of members and nodes

The member information for the truss is shown in Table 4.2.1. The member and node
numbers, modulus of elasticity, cross sectional areas are the necessary input data. From the
coordinate of the nodes of the respective members, the length of each member is computed.
Here, the angle has been calculated considering anticlockwise direction. The signs of the
direction cosines depend on the choice of numbering the nodal connectivity.
Table 4.2.1 Member Information for Truss
Member
No.
1
2
3

Starting
Node
1
2
3

Ending
Node
2
3
1

Value
of
90
315
180

Area Modulus of
Elasticity
A
E
1.5A E
A
E

Now, let assume the coordinate of node 1 as (0, 0). The coordinate and restraint joint
information are given in Table 4.2.2. The integer 1 in the restraint list indicates the restraint
exists and 0 indicates the restraint at that particular direction does not exist. Thus, in node no.
2, the integer 0 in x and y indicates that the joint is free in x and y directions.
Table 4.2.2 Nodal Information for Plane Truss
Node No.
1
2
3

Coordinates
x
y
0
0
0
L
L
0

Restraint List
x
y
1
1
0
0
1
1

10

The stiffness matrices of each individual member can be found out from the stiffness matrix
equation as shown below.
cos 2

cos sin
[k ] = AE
L cos 2

cos sin

cos sin

cos 2

sin 2

cos sin

cos sin

cos 2

sin 2

cos sin

cos sin

sin 2

cos sin

2
sin

Thus the local stiffness matrices of each member are calculated based on their individual
member properties and orientations and written below.
1
0

0
AE
[k ]1 =
L 0

2
0

3
0

4
0

3
1
2
3
4

1
3 AE
[k ]2 =

4 2 L 1

4
1

5
1

6
1

3
4
5
6

and

5
6
1
2
0 1
0 5
1

0
0
0
1 6
AE
[k ]3 =

L 1 0
1
0 1

0
0
0 2
0
Global stiffness matrix can be formed by assembling the local stiffness matrices into globally.
Thus the global stiffness matrix are calculated from the above relations and obtained as
follows:
1
2
3
4
5
6
1
2
3
4
5
6

11

0
0
1 0
0 1
0
1

3
3
0 0

4 2
4 2

3
3
1+
[K ] = AE 0 1
L
4 2
4 2
3
3
1 0

4 2
4 2

3
3

0 0
4 2
4 2

0
0
3
3

4 2
4 2
3
3

4 2
4 2
3
3
+1
4 2
4 2
3
3

4 2
4 2
0

Fx1
F
y1
Fx 2
The equivalent load vector for the given truss can be written as:=
{F } =

Fy 2
Fx 3

Fy 3

0
0

2 P

P
0

0

Let us assume that u and v are the horizontal and vertical displacements respectively at joints.
Thus the displacement vector will be expressed as follows:
u1 0
v 0
1
u u
{d } = 2 = 2
v 2 v 2
u 3 0

v3 0

Therefore, the relationship between the force and the displacement will be:

12

Fx1
F
y1
2 P

P
Fx 3

Fy 3

0
0
0
1
1 0
0 1
0
0
0
1

3
3
3
3

0 0

4 2
4 2
4 2
4 2

AE
3
3
3
3
=
0 1
1+

L
4 2
4 2
4 2
4 2

3
3
3
3

+1
1 0

4 2
4 2
4 2
4 2

3
3
3
3

0 0
4 2
4 2
4 2
4 2

0
0

u2

v2
0

0

From the above relation, the unknown displacements u2 and v2 can be found out through
computer programming. However, as numbers of unknown displacements in this case are
only two, the solution can be obtained by manual calculations. The above equation may be
rearranged with respect to unknown and known displacements in the following form:

F k
=
F k

k d

k d

Thus the developed matrices for the truss problem can be rearranged as:
2P
P
Fx1
Fy1
Fx3
Fy3

3
3
4 2
4 2

3
3

1+
4 2
4 2

AE
=
0
0
L
0
1
3
3

4 2
4 2
3
3
4 2
4 2

.
The above relation may be condensed into following
3
2 P AE 4 2

=
P L 3
4 2

3
4 2

u
2
3 v2
1+
4 2

4 2
4 2
3
3
0 1
4 2
4 2

1 0
0
1
0 1
0
0
3
3
1 0
+1

4 2
4 2
3
3
0 0
4 2
4 2
0

u2
v2
0
0
0
0

13
The unknown displacements can be derived from the relationships expressed in the above
equation.
1

u2

v2

3
3
3

1+

AE 4 2
4 2 2 P 4 2 L 4 2

=

3 P 3 AE 3
L 3
1+
4 2
4 2
4 2

3
4 2 2 P

3 P
4 2

Thus the unknown displacement at node 2 of the truss structure will become:

8 2
u2 PL 3 +

=
3

v2 AE 3

Support Reactions:
The support reactions {Ps} can be determined from the following relation:

{ Pcs } + K {d }
{Ps } =
Where, {Pcs} correspond to equivalent loadings at supports. Thus, the support reaction of the
present truss structure will be:
0
0
0

0
AE 3
+
{Ps } =

0 L 4 2
0
3

4 2

0
0
1
8 2 3P

3 PL 3 +
=
3 2 P
AE
4 2

3
3
2P

4 2

Member End Actions:

Now, the member end actions can be obtained from the corresponding member stiffness and
the nodal displacements. The member end forces are derived as shown below.
Member 1

Fmx1

Fmy1

Fmx 2
Fmy 2

0
0 0 0 0

0 1 0 1 0
AE
PL

=

8
2
L 0 0 0 0 3 +
AE

3
0 1 0 1
3

3P

0
3P

14
Member 2

Fmx 2
F
my 2

Fmx 3
Fmy 3

8 2
1 1 1 1 3 +

3
1 1 1 1
3 AE
3 PL
=
4 2 L 1 1 1 1
AE

0
1 1 1 1
0

2P
2 P

2 P
2 P

Member 3

Fmx 3
1
F

my 3 AE 0
=

F
L 1
mx
1

Fmy1
0

0 1 0 0
0

0 0 0 0 PL 0
=

0 1 0 0 AE 0

0
0 0 0 0

Thus the member forces in all members of the truss will be:

3P

3P

2
2
=

+
F
P
P
2
2
{ m} ( ) ( ) =
2 2 P

0
0

The reaction forces at the supports of the truss structure will be:
0
3P
{FR } =
2 P
2 P
Thus the member force diagram will be as shown in Fig. 4.2.4.

Fig. 4.2.4 Member Force Diagram

15
Lecture 3: Stiffness of Beam Members
4.3.1 Introduction
A beam is a structural member which is capable of withstanding load primarily by resisting
bending. The primary tool for analysis of beam is the EulerBernoulli beam equation. Other
methods for determining the deflection of beams include "slope deflection method" and
"method of virtual work". For calculation of internal forces of beam include "moment
distribution method", force or flexibility method and stiffness method. However, all these
methods have limitations if either of geometry, loading, material properties or boundary
conditions becomes arbitrary in nature. Finite element techniques can well handle such cases
and relieve the analyzer of making simplifications to arrive approximate solutions.
4.3.2 Derivation of Shape Function
The degrees of freedom at each node for a beam member will be (i) vertical deflection and
dv
, where
(ii) rotation. For a beam member, the slope of the elastic curve is given by: =
dx
the variable v is the displacement function of the beam. As the beam has two degrees of
freedom at each node, the variation of v will be cubic and can be expressed using Pascals
triangle as:

v(x ) = 0 + 1 x + 2 x 2 + 3 x 3 = [1 x x 2

0

3 1
x
2
3

(4.3.1)

and
0

dv

=
= 0 1 2 x 3x 2 1
dx
2
3

Fig. 4.3.1 Beam element

(4.3.2)

16
Now, applying boundary conditions, the following expressions from the above relations can
be obtained:
At x=0:
0
0



V1 = [1 0 0 0] 1 ; 1 = [ 0 1 0 0] 1 ;
2
2

3
3

At x=L:
V2 = 1 L L2

0
0


3 1
2 1
;
L 2 = 0 1 2 L 3L
2
2
3
3

Thus combining the above expressions one can write:

V1

1
=

V2
2

0 0
1 0 0
0 1 0
0 1
=

1 L L2 L3 2

2
0 1 2 L 3L 3

[ A]{ }

(4.3.3)

So,
0 1
0
1
=
2 1
3 0

2
1 2 L 3L

0
1
L

0
0
L2

0
0
L3

V1 1
0
2 3
=
V2 L2
2 2
3

0
0
3
L2
2
3
L

0
V1
0
1 1

L V2
1
2
L2

0
1
2

L
1
L2

0
0
3
l2
2
3
l

0
V1
0
1 1
= [N 1
l V2
1
2
l 2

(4.3.4)

Therefore,

v(x ) = 1 x

x2

1
0
3
x 3 2
l
2
l 3

0
1
2

l
1
l2

N2

N3

V1


N 4 ] 1
V2
2

(4.3.5)

Where,
N1 = 1

3 2 2 3
x2 x3
2 2 x3
3x 2 2 x 3
;
;
and
x
+
x
N
=

+
N
=
x

x
+
N
=

2
3
4
L L2
L
L2
L3
L2
L2
L3
(4.3.6)

17
N is called shape function which interpolates the beam displacement in terms of its nodal
displacements.
4.3.3 Derivation of Element Stiffness Matrix
Now, the strain displacement relationship matrix [B] can be expressed from the following
expressions with the help of eq. (4.3.1):

d 2v
= [0
dx 2

0

1
1
6 x ] = [B ]{ } = [B ][A] {d }
2

4

1
0
Where, [ B ] [ 0=
=
0 2 6 x ] ; [ A]
1

0 0
0
1 0
0
=
; {d }
L L2 L3

1 2 L 3L2

(4.3.7)

V1

2

V2
2

From the moment curvature relationship, we can write:

M = EI = EI

d 2v
1
= EI [B ][ A] {d }
2
dx

(4.3.8)

Strain energy,
L

EI
1 T
U = [ ] [M ]dx =
2
2
0

{d } [A ] [B] [B][A ]{d }dx

L

1 T

(4.3.9)

Thus,

{F } =

[ ] [B] [B][A ]{d }dx

U
= EI A 1
{d }
0
L

(4.3.10)

So, the stiffness matrix will be:

[k ] = EI [A ] [B] [B][A ]dx = EI [A ] [B] [B]dx[A]

L

1 T

1 T

[ ]
L

Now,

0
0

L
L
0
0
B T [B ]dx = [0 0 2 6 x ]dx =
0
2
0
0

6 x
0

(4.3.11)

0
0 0
0

0 0
0
0
dx =
0 4
12 x
0

0 12 x 36 x 2
0

4L

6 L2

6 L2

12 L3

(4.3.12)

18
So,

0
[k ] = EI A 1 T
0

[ ]

0
[k ] = EI
0

0
= EI
0

0
1
0
0

4L

6 L2

3
L2
2

L
3
L2
1

6 1

0 0
3

6 L2
2

6l 3
L

0 1
[A]
6 L2

3
12 L

2
L3 0
1
0
L2
2
3 0
L
1 0

L2

4L

6 L2

1
2

L
1
L2

0
3
L2
2
3
L

0 1

0 0
3
2
6 L L2
2

12 L3 3
L

12
0
L3

6
0
2
1 = EI L

12
L
L3
1
6

L2
L2

1
2

L
1
L2

0
3
L2
2
3
L

6
L2
4
L
6
2
L
2
L

12
L3
6
2
L
12
L3
6
2
L

0
1

L
1
L2

6
L2
6
2
L
6
2
L
4

Thus, the element stiffness of a beam member is:

6 L 12 6 L
12
6 L 4 L2 6 L 2 L2
EI
[ k ] = 3 12 6 L 12 6 L
L

2
2
6 L 2 L 6 L 4 L

(4.3.13)

4.3.4 Generalized Stiffness Matrix of a Beam Member

Consider a beam member making an angle with X axis as shown in Fig 4.3.2 below. By
resolving the forces along local X and Y direction, the following relations are obtained.

19
=
Fx1 Fx1 cos + Fy1 sin
=
Fx 2 Fx 2 cos + Fy 2 sin
Fx1 sin + Fy1 cos
Fy1 =

(4.3.14)

Fx 2 sin + Fy 2 cos
Fy 2 =
M1 = M1
M2 = M2

Where, Fx1 and Fx 2 are the axial forces along the member axis X . Similarly, Fy1 and Fy 2 are
the forces perpendicular to the member axis X . M 1 and M 2 are the moment about its axis at
node 1 and 2 respectively.

Fig. 4.3.2 Inclined beam member

The relationship expressed in eq. (4.3.14) can be rewritten in matrix form as follows:
Fx1 cos

Fy1 sin
M 1 0
=
Fx 2 0
Fy 2 0

M 2 0

sin
cos
0
0
0
0

0
0
0
0
1
0
0 cos
0 sin
0
0

0
0
0
sin
cos
0

0 Fx1
0 Fy1
0 M 1

0 Fx 2
0 Fy 2

1 M 2

(4.3.15)

Now, the above equation can be expressed in short as:

{F } = [T ]{F }

(4.3.16)

20
Similarly, the displacement vector in local coordinate system (, ) may be transformed to
global (, ) coordinate system by the following relation.

{d } = [T ]{d }

(4.3.17)

The force-displacement relation in local coordinate system may be expressed as:

0
0

Fx1 0 12 EI

L3
Fy1
6 EI

0
M1
L2

=
0
Fx2 0

12 EI
Fy2 0 3
L
M
2
6 EI
0

L2

0
6 EI
L2
4 EI
L
0
6 EI
L2
2 EI
L

0
12 EI
0 3
L
6 EI
0 2
L
0
0
0
0

12 EI
L3
6 EI
2
L

6 EI
u1
L2
2 EI v1

L 1

0 u
2
6 EI v
2 2
L

4 EI 2

L
0

(4.3.18)

The matrices in the above equation are written with respect to the member axis. Now, the eq.
(4.3.18) can be rewritten as follows with the use of eqs. (4.3.16) and (4.3.17).

[T ]{F } = [k ][T ]{d }

(4.3.19)

{F } = [T ]

(4.3.20)

Or,
1

k [T ]{d }

Here, the transformation matrix [T] is orthogonal. Thus, from the above relationship, the
generalized stiffness matrix can be expressed as:

[k ] = [[T ]T [k ]][T ]

(4.3.21)

Considering = cos and = sin the above expression can be written as follows:

21

0
[ k ] = EI
0
0

0
1

0
0

0
0
0

0
0
0 0

0
0

12
0 0 0
L3

0 0
6
0

0 0
L2

0
0 0

12
0
0 3
L
0 1

6
0
L2

0
6
L2
4
L
0
6
2
L
2
L

0
0
0
0
0
0

0
12
3
L
6
2
L
0
12
L3
6
2
L

0
6

L2
2

L 0

0 0

6 0
2
L 0
4

0
0

0
0
0

0
0
0

0
0
0 0

0
0 0
0 0

0
0

0 1
0

(4.3.22)
Thus, the generalized stiffness matrix of a beam member is derived as:

12 2
L3

12
L3

6
2

[ k ] = EI L 2
12
L3

12
L3
6
2
L

12
L3
12 2
L3
6
L2
12
L3
12 2
3
L
6
L2

6
L2
6
L2
4
L
6
L2
6
2
L
2
L

12 2
L3
12
L3
6
L2
12 2
L3
12
3
L
6
L2

12
L3
12 2
3
L
6
2
L
12
3
L
12 2
L3
6
2
L

6
L2
6
L2

2
L

6
L2

6
2
L
4

(4.3.23)

22
Lecture 4: Analysis of Continuous Beam
4.4.1 Equivalent Loading on Beam Member
In finite element analysis, the external loads are necessary to be acting at the joints, which
does not happen always; as some forces may act on the member. The forces acting on the
member should be replaced by equivalent forces acting at the joints. These joint forces
obtained from the forces on the members are called equivalent joint loads. These joint loads
are combined with the actual joint loads to provide the combined joint loads, which are then
utilized in the analysis.
4.4.1.1 Varying Load
Let a beam is loaded with a linearly varying load as shown in the figure below. The
equivalent forces at nodes can be expressed using finite element technique. If w(x) is the
function of load, then the nodal load can be expressed as follows.

{Q} = [ N ]

w ( x ) dx

(4.4.1)

The loading function for the present case can be written as:
w ( x=
)

w1 +

w2 w1
x
L

(4.4.2)

Fig. 4.4.1 Varying load on beam

From eqs. (4.4.1) and (4.4.2), the equvalent nodal load will become
7 w 3w
L 2 x 3 3x 2

3 2 + 1 w( x )dx 1 + 2 L
L
L

20
20
F1 0L 3
2

x
2
x
w
w

1
2 2
M 1 L2 L + x w( x )dx 20 + 30 L

{Q} = = 0L

= 3w 7 w
3
2

2
x
3
x
1 + 2 L
F2

(
)
w
x
dx
+
20
20
0 L3
L2
M 2 L 3
w w 2
2
1
2
x
x

2 w( x )dx 30 20 L

0 L
L

(4.4.3)

23
Now, if w1=w2=w, then the equivalent nodal force will be:
wL
2

2
wL

{Q} = 12
wL
2

2
wL
12

(4.4.4)

4.4.1.2 Concentrated Load

Consider a force F is applied at a point is regarded as a limiting case of intense pressure over
infinitesimal length, so that p(x)dx approaches F. Therefore,
=
{Q}

[ N ] p ( x ) dx
=
T

N
F

(4.4.5)

Fig. 4.4.2 Concentrated load on beam

Here, [N*] is obtained by evaluating [N] at point where the concentrated load F
is applied. Thus,

[ N *]

2 x3 3x 2
3 2 + 1
L
L3

x 2 x2

L2 L + x
at dista nce a
=
3
2
2 x + 3x
L3
L2

3
2
x x

L2 L

2a 3 3a 2
3 2 + 1
L
L3

a 2a 2

L2 L + a

3
2
2a + 3a
L3
L2

3
2
a a

L2 L

(4.4.6)

24
2a 3 3a 2

3 2 + 1
L
L

F1 3
2

2
a
a

M 1 L2 L + a

Therefore, {Q} = =

3
2

2
3
a
a

F2

+
2
L3
L
M 2

3
2
a a
L2 L

(4.4.7)

Now, if load F is acting at midspan (i.e., a=L/2), then equivalent nodal load will be
F
2

FL

{Q} = 8
F
2
FL

(4.4.8)

With the above approach, the equivalent nodal load can be found for various loading function
acting on beam members.
4.4.2 Worked Out Example
Analyze the beam shown below by the stiffness method. Assume the moment of inertia of
member 2 as twice that of member 1. Find the bending moment and reactions at supports of
the beam assuming the length of span, L as 4 m, concentrated load (P) as 15 kN and udl, w as
4 kN/m.

Fig. 4.4.3 Example of a continuous beam

25
Solution
Step 1: Numbering of Nodes and Members
The analysis of beam starts with the numbering of members and joints as shown below:

Fig. 4.4.4 Numbering of nodes and members

The member AB and BC are designated as (1) and (2). The points A,B,C are designated
by nodes 1, 2 and 4. The member information for beam is shown in tabulated form as shown
in Table 4.4.1. The coordinate of node 1 is assumed as (0, 0). The coordinate and restraint
joint information are shown in Table 4.4.2. The integer 1 in the restraint list indicates the
restraint exists and 0 indicates the restraint at that particular direction does not exist. Thus, in
node no. 2, the integer 0 in rotation indicates that the joint is free rotation.
Table 4.4.1Member Information for Beam
Member
number

Starting node

1
2

Ending node

1
2

2
3

Rigidity modulus

EI
2EI

Table 4.4.2 Nodal Information for Beam

Node No.
1
2
3

Coordinates
x
y
0
0
L
0
2L
0

Restraint List
Vertical
Rotation
1
1
1
0
1
0

26

Step 2: Formation of member stiffness matrix:

The local stiffness matrices of each member are given below based on their individual
member properties and orientations. Thus the local stiffness matrix of member (1) is:
1

12 EI
L3
6 EI

L2
[k ]1 = 12
EI
L3
6 EI

L2

6 EI
L2
4 EI
L
6 EI
2
L
2 EI
L

4
6 EI
L2
2 EI

L
6 EI
2
L
4 EI

12 EI
L3
6 EI
2
L
12 EI
L3
6 EI
2
L

1
2
3
4

Similarly, the local stiffness matrix of member (2) is:

3

[k ]2

24 EI
L3
12 EI

2
= L
24 EI
L3
12 EI

L2

12 EI
L2
8 EI
L
12 EI
2
L
4 EI
L

24 EI
L3
12 EI
2
L
24 EI
L3
12 EI
2
L

12 EI
L2
4 EI

L
12 EI
2
L
8 EI

3
4
5
6

Step 3: Formation of global stiffness matrix:

The global stiffness matrix is obtained by assembling the local stiffness matrix of members
(1) and (2) as follows:
1

[K ]

12 EI
L3
6 EI

2
L
12 EI
3

= L
6 EI

2
L
0

2
6 EI
L2
4 EI
L
6 EI
2
L
2 EI
L
0
0

3
12 EI
L3
6 EI
2
L
36 EI
L
6 EI
L2
24 EI
3
L
12 EI
L2

4
6 EI
L2
2 EI
L
6 EI
L2
12 EI
L
12 EI
2
L
4 EI
L

5
0
0
24 EI
L3
12 EI
2
L
24 EI
L3
12 EI
2
L

12 EI
L2
4 EI

L
12 EI
2
L
8EI

L
0

1
2
3
4
5
6

27

Step 4: Boundary condition:

The boundary conditions according to the support of the beam can be expressed in terms of
the displacement vector. The displacement vector will be as follows
0
0

0
{d } =
2
0

Step 5: Load vector:

The concentrated load on member (1) and the distributed load on member (2) are replaced by
equivalent joint load. The equivalent joint load vector can be written as

Fig. 4.4.5 Equivalent Load

PL

P wL
2 + 2
{F } = PL wL2

12
8

wL

wL2

12

Step 6 : Determination of unknown displacements:.

The unknown displacement can be obtained from the relationship as given below:

28

{ F } = [ K ] {d }
1
{d } = [ K ] { F }
12 EI
L3

0 6 EI
0 L2
12 EI
0 L3
= 6 EI
2
2
0 L
0
3

6 EI
L2
4 EI
L
6 EI
2
L
2 EI
L

12 EI
3
L
6 EI
2
L
36 EI
L
6 EI
L2
24 EI
3
L
12 EI
L2

0
0

6 EI
L2
2 EI
L
6 EI
L2
12 EI
L
12 EI
2
L
4 EI
L

0
0
24 EI
L3
12 EI
2
L
24 EI
L3
12 EI
2
L

1
P

0
2

PL

0
8

P wL
12 EI
+ 2
L2 2
PL wL2
4 EI

L
12
8
12 EI

wL
2

L
2

8EI
2
wL

12

The above relation may be condensed into following

12 EI
2 L
=

3 4 EI
L

PL wL2
4 EI

L
8
12
L
=

2
8 EI
wL
20 EI
12

PL wL2

2 1 8
12

1 3
2

wL

12

PL wL2
2
L 4 4
=

2
3 20 EI PL + wL
8
3

PL2
wL3
2 =

80 EI 80 EI

3 =

PL
wL3
+
160 EI 60 EI

Step 7: Determination of member end actions:

The member end actions can be obtained from the corresponding member stiffness and the
nodal displacements. The member end actions for each member are derived as shown below.

Member-(1)

29
12 EI
L3
F1
6 EI
M

L L2
1
=

12 EI
F2 20 EI 3

L
M 2
6 EI

L2

6 EI
L2
4 EI
L
6 EI
2
L
2 EI
L

12 EI
L3
6 EI
2
L
12 EI
L3
6 EI
2
L

3P 3wL
6 EI
40 40
2

L
0

PL wL2
2 EI

0
= 40 40
L
3P 3wL
0
6 EI
2

2 PL wL
40
L

40

4 PL wL2
4 EI 4

20 20
L

Member-(2)
24
L2
F2
12
M

2 EI L
=

24
F3 L 2
L
M 3
12

12
L
8

12
L
4

24
L2
12

L
24
L2
12

wL 3P
12
+
0
20 40
L
PL wL2 3PL wL2

4 = 40
30

0
12
wL 3P

L PL + wL 20 40

8
3 wL2

12

Actual member end actions:

Member (1)
3P 3wL P 23P 3wL
40 40
2 40 40
F1
2

2
PL wL PL 6 PL wL
M 1 40 40 8
40
= 3P 3wL + P = 40
P
wL
17
3
F

2

+

+
M 40 40 2 40
40
2 PL wL2 PL
3PL wL2
20 20 8 40 20

Member (2)
wL 3P wL
20 + 40 2 11wL + 3P
20
F2

2
2
40
3PL wL wL 3PL wL2
M 2 40

+
30
12 =
= wL 3P + wL
40
20

F
3
9 wL 3P

M 3 20 40 2

2

2
20 40
wL

wL

12

12
23P 3wL

40
R A 40

The support reactions at the supports A, B and C are {FR } = RB = 25wL + P

2
R 40
C 9wL 3P
+
20
4

Putting the numerical values of L, P and w (P=15, L=4, w=4) the member actions and support
reactions will be as follows:

30

Member end actions:

F2 9.925

M 2 7.7

=
F3 6.075

M 3 0

Support reactions:
R A 7.425
{FR } = RB = 17.5
R 6.075
C

F1 7.425

M 1 7.4

=
F2 7.575

M 2 7.7

31
Lecture 5: Plane Frame Analysis
4.5.1 Introduction
The plane frame is a combination of plane truss and two dimensional beam. All the members
lie in the same plane and are interconnected by rigid joints in case of plane frame. The
internal stress resultants at a cross-section of a plane frame member consist of axial force,
bending moment and shear force.
4.5.2 Member Stiffness Matrix
In case of plane frame, the degrees of freedom at each node will be (i) axial deformation, (ii)
vertical deformation and (iii) rotation. Thus the frame members have three degrees of
freedom at each node as shown in Fig. 4.5.1 below.

Fig. 4.5.1 Plane frame element

Therefore, the stiffness matrix of the frame in its local coordinate system will be the
combination of 2-d truss and 2-d beam matrices:
u1

[]=

v1

u2

12

12
3

6
2

6
2

0
0

0
0

v2

0
6
2
2

0
6
2

12

3
6
2

12

3
6

(4.5.1)

4.5.3 Generalized Stiffness Matrix

In plane frame the members are oriented in different directions and hence it is necessary to
transform stiffness matrix of individual members from local to global co-ordinate system

32
before formulating the global stiffness matrix by assembly. The generalized stiffness matrix
of a frame member can be obtained by transferring the matrix of local coordinate system into
its global coordinate system. The transformation matrix can be expressed as:

cos
sin

0
[T]=
0
0
0

sin
cos
0
0
0
0

0
0
0
0
1
0
0 cos
0 sin
0
0

0
0
0
sin
cos
0

0
0

0
0
0
1

(4.5.2)

Now, the generalized stiffness matrix of the member can be obtained from the relation of
][] . Thus considering = cos and = sin the stiffness matrix in global
[] = [] [
coordinate system can be written as follows:
AE
AE

0
0

0
0
L
L

12EI
6EI
12EI 6EI
0
0
3
0 0 0 0
L3
L2
L
L2
0 0 0 0
6EI
4EI
6EI
2EI

0
0
2
2
L
L
L
L
[K] = EI 0 0 1 0 0 0

AE
0 0 0 0 AE
0
0
0
0

0 0 0 0
L
L

0 0 0 0 0 1
6EI
12EI
6EI
12EI
0
2
0
2
3
L
L
L3
L

6EI
2EI
6EI
4EI

0
2
2
0
L
L
L
L
0 0 0 0
0 0 0 0

0 0 1 0 0 0

0 0 0 0
0 0 0 0
0 0 0 0 0 1
EA
12EI
EA
12EI
2 + 3 2
3
L
L
L
L

12EI
EA
12EI
EA

2 + 3 2
L3
L
L
L
6EI
6EI

L2
L
=
EA 2 12EI 2
EA
12EI
3
+ 3

L
L
L
L

12EI
EA
12EI
EA
2

+ 3
3 2
L
L
L
L

6EI
6EI

L2
L

6EI

L2
6EI

L2
4EI
L
6EI

L2
6EI
2
L
2EI
L

EA 2 12EI 2
EA
12EI
3
+ 3
L
L
L
L
EA
12EI
EA 2 12EI 2

+ 3
3
L
L
L
L
6EI
6EI

2
L2
L
EA 2 12EI 2
EA
12EI
+ 3
3
L
L
L
L
EA
12EI
EA 2 12EI 2
3
+ 3
L
L
L
L
6EI
6EI

2
L2
L

6EI

L2
6EI

L2
2EI
L
6EI

L2
6EI
2
L
4EI
L

33
(4.5.3)

4.5.4 Worked Out Example

Analyse the plane frame shown below by the stiffness method. Assume the modulus of
elasticity of the horizontal member is 1.5 times that of the vertical member and length of
the vertical member is 1.5 times that of horizontal member. Find the bending moment and
reactions at support assuming the length, cross section area and modulus of elasticity of
vertical member as 3.0 m, 0.4 x 0.4 m2 and 2 x 1011 N/mm2, respectively.

Fig. 4.5.2 Plane frame

Solution
Step 1: Numbering of Nodes and Members
The numbering of members and joints of the plane frame are as shown below:

Fig. 4.5.3 Numbering of Nodes and Members

The members AB and BC are designated as (1) and (2). The points A, B and C are designated
by nodes 1, 2 and 3. The member information for the frame is shown in tabulated form as
shown in Table 1(a). The coordinate of node 1 is assumed as (0,0). The coordinate and
restraint joint information are shown in Table 1(b). The integer 1 in the restraint list indicates

34
the restraint exists and 0 indicates the restraint at that particular direction does not exist.
Thus, in node no. 2, the integer 0 all the restraint type indicates that the joint is in free all the
three directions.
Table 4.5.1 Member Information for Beam
Starting node
Ending node
1
2
2
3

Member number
1
2

Node no.
X
0
0
L

1
2
3

Rigidity modulus
EI
1.5EI

Table 4.5.2 Nodal Information for Beam

Coordinates
Restraint list
Y
Axial
Vertical
0
1
1
1.5L
0
0
1.5L
1
1

Rotation
1
0
1

Step 2: Formation of member stiffness matrix:

The individual member stiffness matrices can be found out directly from eqn. shown above.
Thus the stiffness matrices of each member in global coordinate system are given below
based on their individual member properties and orientations. Thus the stiffness matrix of
member (1) is:
1

12EI

(1.5L)3

0
6EI

(1.5L)2
[]1 = 12EI
(1.5L)3

6EI
(1.5L)2

AE

(1.5L)

6EI

(1.5L)2

4EI

(1.5L)
6EI

(1.5L)2

(1.5L)

AE

(1.5L)

2EI

12EI

(1.5L)3

6EI

(1.5L)2

12EI

(1.5L)3

6EI

(1.5L)2

AE

(1.5L)

6EI

(1.5L)2

2EI

1.5L
6EI

(1.5L)2

(1.5L)

(1.5L)

AE

Similarly, the stiffness matrix of member (2) is :

4

4EI

1
2
3
4
5
6

35
A(1.5 E)

L
12(1.5 E)I

0
L3

6(1.5
E)I

0
2
L
[]2 = A(1.5 E)

0
L

12(1.5 E)I
0
L3

6(1.5 E)I
0

L2

6(1.5 E)I
L2
4(1.5 E)I

A(1.5 E)
L

0
0

A(1.5 E)

2(1.5 E)I

6(1.5 E)I

L2

6(1.5 E)I
L2

2(1.5 E)I
L

6(1.5 E)I
L2
4(1.5 E)I

12(1.5 E)I

L3
6(1.5 E)I
L2

12(1.5 E)I

L3
6(1.5 E)I
L2

4
5
6
7
8
9

Step 3 : Formulation of global stiffness matrix:

The global stiffness matrix is obtained by assembling by assembling the local stiffness matrix
of member (1) and (2) as follows:
1

32EI

3
8EI

0
2
9L3
3L
2AE
0
0
3L

8EI
8EI2
0
3L
3L
32EI
8EI
0
9L3
3L2

2AE
[K] = 0

0
3L
4EI
8EI

0
3L
3L2
0
0
0

0
0
0

0
0
0

32EI
9L3

8EI
3L2
32EI
1.5EA
( 3 +
)
9L
L

8EI
3L2
1.5AE

2AE
3L

2AE
3L

18EI
)
L3
9EI
L2

18EI
L3
9EI
L2

8EI
3L2

4EI
3L
8EI
3L2
9EI
L2
8EI
6EI
( 2+ )
3L
L

9EI
L2
3EI
L

1.5EA
L

1.5AE
L

0
0

18EI
L3
9EI
2
L

18EI
L3
9EI
2
L

9
0
0

0
9EI
L2
3EI
L
0

9EI
2
L
6EI
L

Step 4: Boundary conditions:

The boundary conditions according to the support of the frame can be expressed in terms of
the displacement vector. The displacement vector will be as follows:
0
0
0

{} =

0
0
0
Here, , and indicate the displacement in X-direction, displacement in Y-direction
and rotation at point B.
Step 5: Load vector:

1
2
3
4
5
6
7
8
9

36
The distributed load on member (2) can be replaced by its equivalent joint load as shown in
the figure below.

Fig. 4.5.4 Equivalent Joint Loads

Thus, the equivalent joint load vector can be written as

0
0
0

0

2
{} = 2
12
0

2
2
12
Step 6: Determination of unknown displacements:
The unknown displacements can be obtained from the relationship of {F} = [K]{d} or
{d} =[k]-1 {F}. Now eliminating the rows and columns in the stiffness matrix and force
matrix, corresponding to zero elements in displacement matrix, the reduced matrix will be as
follows.
1
32 1.5
8
0
3 +

32
9

2 18
9
2
=
0

+ 3

2
3

8
9
8 6

+
12
2
2

Thus, the unknown displacements will be:

1 0.04327
= 10 1.7127
10

5.4978
Step 7: Determination of member end actions:
The member end actions can be obtained from the corresponding member stiffness and the
nodal displacements. The member end actions for each member are derived as shown below.

37

Member (1)
In case of member (1), the member forces will be:{ }1 = [](1) {}(1)

56.17
0
126.4 56.17
0
1

0
7110
0
0
7110
1

1
126.4
0
379.2
126.4
0
6
2 = 10 56.17
0
379.2
56.17
0

0
7110
0
0
7110
2
126.4
0
189.6
126.4
0
2
0

12
4.327 10
1.7127 1010
5.4978 1010

126.4
0

189.6
126.4
0
379.2

0.0697
1.2177

0.10479
=
0.06925
1.21661
0.20793

It is to be noted that {Fm} are the end actions due to joint loads. Hence it must be added to the
corresponding end actions in the restrained structure in order to obtain the end actions due to
the loads. Therefore, {Fm}actual are the true member end actions due to actual loading system
can be expressed as
{ } = {Fm} + {Ffm}
Where, {Ffm} are the end actions in the restrained structure. Since there is no load acting on
member (1), the actual end actions will be:

{Fm}actual =

0.0697
0
1.2177 0

0.10479 + 0 =
0.06925 0
1.21661 0
0.20793 0

0.0697
1.2177

0.10479
0.06925
1.21661
0.20793

Member (2)
In similar way, the member forces in member (2) will be {Fm}(2) = [K](2){d}(2)

38

16
0
2
0
0.284
2

2
0.426
9 0
=
10
3
16
0

0
0.284
3
0
0.426
3
0.069232
0.28325

0.54215
=
0.06923
0.283245
0.3076

0
16
0.426
0
0.853
0
0
16
0.426
0
0.426
0

0
0.284
0.426
0
0.284
0.426

0
4.327 1012

0.426
1.7127 1010

0.426 5.4978 1010

0
0

0.426
0

0.853
0

The actual member forces in the member (2) will be:

0.069232
0
0.0692
0.28325 1.5 1.2167

0.54215 0.75 0.2078

{Fm}actual =
+
=
0.06923 0 0.0692
0.283245 1.5 1.7832
0.3076 0.75 1.0576

Lecture 6 Analysis of Grid and Space Frame

4.6.1 Introduction
The property of a grid member is basically a combination of 2-d beam with torsional effect.
The plane frame is assumed to be loaded in its own plane where as loading in the grid is
normal to its plane. As a result torsional effects are included in the grid analysis. Thus the
grid member can withstand bending moment, shear force as well as torsional moment.
4.6.2 Element Stiffness Matrix for Grid Members
The degrees of freedom at each node of the grid member will be (i) vertical deformation and
(ii) rotation in two different directions.

39

Fig. 4.6.1 Degrees of freedom of grid element

Therefore, the stiffness matrix of the grid in its local coordinate system will be:
x1
GI x
L

0
[k ] = GI
x
L

z1

v1

4 EI y

L
6 EI y

L2

L3

0
L
6 EI y

6 EI y

L
12 EI y

L2

2 EI y

y2

GI x
L

0
2 EI y

0
0

GI x
L

x2

6 EI y
2

L
12 EI y
L3

L
6 EI y
L2
0

4 EI y

L
6 EI y

L2

v2

6 EI y

L2
12 EI y

L3
0

6 EI y

L2
12 EI y
L3
0

Mx1
Mz1

Fy1

(4.6.1)
Mx2
Mz2
Fy2

Here, the G is the modulus of torsional rigidity.

4.6.3 Generalized Stiffness Matrix
The generalized stiffness matrix of a grid member can be obtained by transferring the matrix
of local coordinate system into its global coordinate system. The transformation matrix can
be expressed as:
cos
sin

0
[T]=
0
0
0

sin
cos
0
0
0
0

0
0
0
0
1
0
0 cos
0 sin
0
0

0
0
0
sin
cos
0

0
0

0
0
0
1

40
Now, the generalized stiffness matrix of the member can be obtained from the relation
of [] = [] [] . Thus considering = cos and = sin the stiffness matrix in
global coordinate system can be written as follows:

0
[] =
0
0
0

0
0
0
0

0
0
1
0
0
0

0
0
0

0
0

GI x 2 4 EI y 2
+

L
L
GI x 4 EI y

L
L

6 EI y

2
L

=
GI x 2 4 EI y 2
L + L

GI x 2 EI y

L
L

6 EI y

0
0
0

0
0
0
0

0
0
1
0
0
0

GI

x
0
L
4EIy
0
0
L

0
6EIy

L2
0 0
GIx
0
0
0 L
2EIy
1 0
L

6EIy
0
L2
0 0 0
0 0 0

0 0 0
0
0
0 0 1

GI x 4 EI y

L L

GI x 2 4 EI y 2
+

L
L
6 EI y
2
L
GI x 2 EI y

+
L
L
2 EI y 2
GI

x 2 +
L
L
6 EI y
L2

6 EI y

L2
6 EI y

L2
12 EI y

L3
6 EI y

L2
6 EI y

6EIy

L2
12EIy

L3

6EIy

GIx
L

0
0

GIx
L

L2
12EIy

L3

GI x 2 4 EI y 2
+

L
L
2 EI y
GI

x +

L
L

6 EI y

L2

L2
12 EI y
L3

GI x 2 4 EI y 2
+

L
L
GI x 4 EI y

L L

6 EI y
2
L

2EIy

L
6EIy
L2

4EIy
L
6EIy
L2

L2
12EIy
L3

0
6EIy
L2
12EIy
L3
6EIy

2 EI y
GI

x +

L
L

GI x 2 2 EI y 2
+

L
L
6 EI y
2
L
GI x 4 EI y

L L

4
EI
GI x 2
y
+
2
L
L
6 EI y
L2

6 EI y

L2
6 EI y

L2

12 EI y

L3
6 EI y
2
L

6 EI y

L2

12 EI y

L3

(4.6.2)

4.6.4 Worked Out Example

Analyze the grid shown below by the stiffness method. Draw the shear force and bending
moment diagram assuming the cross sectional area and modulus of elasticity of each member
as 0.30.3 m2 and 21011 N/m2 respectively. Assume EI = 3GJ. The length of member AB
and BC is 4 m and 5 m respectively.

41

Fig. 4.6.2 Grid structure

Solution
Step 1: Numbering of Nodes and Members
The numbering of members and joints of the plane frame are as shown in the figure below:

Fig. 4.6.3 Numbering of nodes and members

The member AB and BC are designated as (1) and (2). The points A, B and C are designated
by nodes 1, 2 and 3. The member information for the grid is shown in tabulated form as
shown in Table 4.6.1. The coordinate of node 1 is assumed as (0, 0). The coordinate and
restraint joint information are shown in Table 4.6.2. The integer 1 in the restraint list indicates
the restraint exists and 0 indicates the restraint at that particular direction does not exist.
Thus, in node no. 2, the integer 0 all the restraint type indicates that the joint is free in all the
three directions.
Table 4.6.1 Member Information
Member number Starting node Ending node
1
1
2
2
2
3

42

Table 4.6.2 Member Coordinates

coordinates
Restraint list
x
z
Vertical Rotation Rotation
0
0
1
1
1
4
0
0
0
0
4
5
1
1
1

ode
No
1
2
3

Step 2: Formation of member stiffness matrix:

The individual member stiffness matrices can be found out directly. Thus the stiffness
matrices of each member in global coordinate system are given below based on their
individual member properties and orientations. As the member AB is horizontal, i.e., = 0,
the values of Cos = 1 and Sin = 0. Thus the stiffness matrix of member (1) is:
1

[k ]AB

GJ
L

=
GJ

L
0

12 EI
L3
6 EI
L2

6 EI
L2
4 EI
L

12 EI
L3
6 EI
L2

GJ
L

0
0

0
12 EI
L3
6 EI
2
L

GJ
L

6 EI
L2
2 EI
L

0
0

0
12 EI
L3
6 EI
2
L

6 EI

L2
2 EI
L

6 EI
2
L
4 EI

L
0

1
2
3
4
5
6

Assuming EI=3GJ=3K, the above equation can be written as

1

[k ]AB

K
L

=
K

L
0

36 K
L3
18 K
L2

18 K
L2
12 K
L

36 K
L3
18 K
L2

18 K
L2
6K
L

K
L

0
0
K
L
0
0

36 K
L3
18 K
2
L

0
36 K
L3
18 K
2
L

18 K

L2
6K
L

18 K
2
L
12 K

L
0

1
2
3
4
5
6

43
As the member BC member is also horizontal, the value of Cos = 1 and Sin = 0 and thus,
the stiffness matrix will be:
6

K
K
0
0

L
L

36 K
18 K
0
0
L3
L2

18 K
12 K
0
0
2

L
L
[k ]BC = K
K
0
0

L
L
36
18
K
K
0
0
3
2

L
L

18 K
6K
0
0
2
Step 3: Formation
of
global
stiffness
matrix:
L
L

8
0
36 K
L3
18 K
2
L

0
36 K
L3
18 K
2
L

18 K

L2
6K
L

18 K
2
L
12 K

L
0

6
5
4
7
8

9
The global stiffness matrix can be obtained by assembling the local stiffness matrix of
members (AB) and (BC). Now looking at the grid structure, the displacements at the fixed
supports, are known and all are equal to zero. Only the displacement at co-ordinates 4, 5, 6
are unknown. So the global system stiffness matrix, corresponding to the displacement at coordinate 4, 5, 6 will be:

12

18
[] = 2

18

2
36
36

2.65
0.72
= K 0.72 0.8505
0
1.125

3
18

0
1.125
3.2

18
2

12
+

Step 4: Boundary condition:

The boundary conditions according to the support of the grid structure can be expressed in
terms of the displacement vector. The displacement vector will be as follows

44
0
0

0

d 4
{d } = d5
d
6
0

0
0

Here, d4, d5, and d6 indicate the displacement vectors at point B.

Step 5: Load vector:
The distributed load on member (1) can be replaced by its equivalent joint load as shown in
the figure below.

Fig. 4.6.4 Equivalent load

Thus the equivalent load vector will be:

0
wL
2
wL2

12
0
{P} = wL
22
wL
12
0

0
0

45
Step 6: Determination of unknown displacements:.
The unknown displacements can be obtained from the relationship of { F } = [ K ]{d } or

{d } = [ K ] { F } .
1

Now, eliminating the rows in the force matrix, corresponding to zero

element in displacement matrix, the reduced matrix will be as follows.

0
1 4
4
2.65
0.72
0

5 = k 1 0.72 0.8505 1.125 2
6
0
1.125
3.2
16
12
0
1 0.662 1.047 0.368 2
= 1.047 3.856 1.3556 4

0.368 1.355
0.789
3
Thus, the unknown displacements will be:
4
1 1.603
5 = 5.905

6
1.658

Step 7: Determination of member end actions:

The member end actions can be obtained from the corresponding member stiffness and the
nodal displacements. The member end actions for each member are derived as shown below.
Member - AB
In case of member (AB), the member forces will be: {Fm}(AB) = [K](AB) {d}(AB)
1
1

0
0
0
0
4
4

36
18
36 18
0
0 3
1
0
43
42
4
42

18
12
18
6

0
1

0
0

0
42
4
42
4 1
1 =
1.603
1
1
2
0
0
0
0

5.905
2
4
4

1.658
2
36
18
36
18
0

43
42
43
42

18
6
18
12

0 2
2
0
4
4
4
4
Thus,

1
0.4

1.456
1

1 = 4.156
2 0.4
2 1.456
2 1.70

46

It is to be noted that {Fm} are the end actions due to joint loads. Hence it must be added to the
corresponding end actions in the restrained structure in order to obtain the end actions due to
the loads. Therefore, {Fm} Actual are the true member end actions due to actual loading system
and can be expressed as
{Fm} Actual = {Fm} + {Ffm}
Where, {Ffm}are the end actions in the restrained structure. Since there is no load acting on
member (1), the actual end action will be:
0
4

2
0.4

0.4
1.456
42
3.46

4.156 12 5.49
{ } =
+
=
0.4 0 0.40
1.456 4 0.54
1.70 2 0.34
42
12
Member - BC
In similar way, the member forces in member (BC) will be: {Fm}(BC) = [K](BC) {d}(BC)
1
1

0
0
0
0
5
4

36
18
36 18
0
0 3
2
53
52
5
52

1.658

5.905
18
12
18
6

2

0 2

0
52
5
5
5 1 1.603
2 =
1

1
0

0
0
0
0

0
3
5
5

0
3
36
18
0 36 18 0

53
52
53
52

18
6
18
12

0 2
2
0
5
5
5
5
Thus,

47
2
0.33

2
0.55
2 = 0.40
3 0.33
3 0.55
3 2.33
Since there is no load acting on member (BC), the actual end action will be:
0
0.33
0.33
0.55 0 0.55
0.40 0.40
+ 0 =

{ } =
0.33

0 0.33
0.55 0 0.55
2.33 0 2.33
Thus, the reaction forces at the support and load at the joints will be:
1

1
1
2
2 =
2
3
3
3

0.4
3.46
5.49

0
0
0
0.33
0.55
2.33

4.6.5 Analysis of Space Frame

Space frames are an increasingly common architectural technique especially for large roof
spans in commercial and industrial buildings. The rigid jointed frames such as building
frames are usually three dimensional space structures. Thus in case of certain structures like
multi-storeyed buildings, it is necessary consider 3-dimensional effects for analysis. The
space frame constitutes the final step of increasing complexity. It consists of plane frame and
grid actions. The displacement and rotation vector associated with each joint have three
components in case of space frame structures. There are six equilibrium equations associated
with each joint. The degrees of freedom at each node of the space frame member will be (i)
displacement in three perpendicular directions and (ii) rotations in three different directions.
Therefore, the degrees of freedom in each node of the member will be six as shown in the
figure below. The stiffness matrix in local coordinate system considering all possible degrees
of freedom will be as given in Table 4.6.1.

48

Fig. 4.6.5 Degrees of freedom for space frame member

Table 4.6.1 Stiffness matrix of space frame member

1
1
2
3
4
5
6
7
8
9
10
11
12

EAX
L

EAX
L

12 EI Z
L3

6 EI Z
L2

12 EIY
L3

6 EIY
L2

GI X
L

6 EIY
L2

4 EIY
L

6 EI Z
L2

4 EI Z
L

12 EI Z
L3

12 EIY
L3

0
0
6 EI Z
L2

10

11

12 EI Z
L3

12 EIY
L3

EAX
L

6 EIY
L2

GI X
L

6 EIY
L2

2 EIY
L

6 EI Z
L2

EAX
L

6 EI Z
L2

12 EI Z
L3

6 EIY
L2

12 EIY
L3

6 EIY
L2

GI X
L

GI X
L

6 EIY
L2

2 EIY
L

6 EIY
L2

4 EIY
L

2 EI Z
L

6 EI Z
L

12

6 EI Z
L2

2 EI Z
L

6 EI Z
2
L

4 EI Z
L
0

The generalized stiffness matrix of a rigid jointed space frame member can be obtained by
transferring the matrix of local coordinate system into its global coordinate system. The
transformation matrix will become a square matrix of size 1212 in this case as the degrees
of freedom for each node/joint is six.