Finite Element Methods - BME701 Module III 2025-26
Module-3 BEAMS and 75866(6
Beam element
Beam is a structural member which is acted upon by a system of
external loads perpendicular to axis which causes bending that is
deformation of bar produced by perpendicular load as well as force
couples acting in a plane. Beams are the most common type of
structural component, particularly in Civil and Mechanical
Engineering. A beam is a bar-like structural member whose primary
function is to support transverse loading and carry it to the supports
A truss and a bar undergoes only axial deformation and it is
assumed that the entire cross section undergoes the same displacement,
but beam on other hand undergoes transverse deflection denoted by v.
Fig shows a beam subjected to system of forces and the deformation of
the neutral axis
Department of Mechanical Engineering, MITE 1
Finite Element Methods - BME701 Module III 2025-26
We assume that cross section is doubly symmetric and bending take
place in a plane of symmetry. From the strength of materials we
observe the distribution of stress as shown.
Where M is bending moment and I is the moment of inertia.
According to the Euler Bernoulli theory. The entire c/s has the same
transverse deflection V as the neutral axis, sections originally
perpendicular to neutral axis remain plane even after bending
Deflections are small & we assume that rotation of each section
is the same as the slope of the deflection curve at that point (dv/dx).
Now we can call beam element as simple line segment representing the
neutral axis of the beam. To ensure the continuity of deformation at any
point, we have to ensure that V & dv/dx are continuous by taking 2 dof
@ each node V & T(dv/dx). If no slope dof then we have only
transverse dof. A prescribed value of moment load can readily taken
into account with the rotational dof T .
Potential energy approach
Strain energy in an element for a length dx is given by
Department of Mechanical Engineering, MITE 2
Finite Element Methods - BME701 Module III 2025-26
But we know V = M y / I substituting this in above equation we get.
But
Therefore strain energy for an element is given by
Now the potential energy for a beam element can be written as
Department of Mechanical Engineering, MITE 3
Finite Element Methods - BME701 Module III 2025-26
Hermite shape functions:
1D linear beam element has two end nodes and at each node 2
dof which are denoted as Q2i-1 and Q2i at node i. Here Q2i-1 represents
transverse deflection where as Q2i is slope or rotation. Consider a beam
element has node 1 and 2 having dof as shown.
The shape functions of beam element are called as Hermite shape
functions as they contain both nodal value and nodal slope which is
satisfied by taking polynomial of cubic order
that must satisfy the following conditions
Applying these conditions determine values of constants as
Department of Mechanical Engineering, MITE 4
Finite Element Methods - BME701 Module III 2025-26
Solving above 4 equations we have the values of constants
Therefore
Similarly we can derive
Following graph shows the variations of Hermite shape functions
Department of Mechanical Engineering, MITE 5
Finite Element Methods - BME701 Module III 2025-26
Stiffness matrix:
Once the shape functions are derived we can write the equation
of the form
But
ie
Strain energy in the beam element we have
Department of Mechanical Engineering, MITE 6
Finite Element Methods - BME701 Module III 2025-26
Therefore total strain energy in a beam is
Now taking the K component and integrating for limits -1 to +1 we get
Department of Mechanical Engineering, MITE 7
Finite Element Methods - BME701 Module III 2025-26
Beam element forces with its equivalent loads
Uniformly distributed load
Point load on the element
Varying load
Bending moment and shear force
We know
Using these relations we have
Department of Mechanical Engineering, MITE 8
Finite Element Methods - BME701 Module III 2025-26
Solution:
Let’s model the given system as 2 elements 3 nodes finite
element model each node having 2 dof. For each element determine
stiffness matrix.
Global stiffness matrix
Department of Mechanical Engineering, MITE 9
Finite Element Methods - BME701 Module III 2025-26
Load vector because of UDL
Element 1 do not contain any UDL hence all the force term for
element 1 will be zero.
ie
For element 2 that has UDL its equivalent load and moment are
represented as
ie
Global load vector:
Department of Mechanical Engineering, MITE 10
Finite Element Methods - BME701 Module III 2025-26
From KQ=F we write
At node 1 since its fixed both q1=q2=0
node 2 because of roller q3=0
node 3 again roller ie q5= 0
By elimination method the matrix reduces to 2 X 2 solving this we
have Q4= -2.679 X 10-4mm and Q6 = 4.464 X10-4mm
To determine the deflection at the middle of element 2 we can write the
displacement function as
= -0.089mm
Department of Mechanical Engineering, MITE 11
Finite Element Methods - BME701 Module III 2025-26
Solution: Let’s model the given system as 3 elements 4 nodes finite
element model each node having 2 dof. For each element determine
stiffness matrix. Q1, Q2……Q8 be nodal displacements for the entire
system and F1……F8 be nodal forces.
Global stiffness matrix:
Department of Mechanical Engineering, MITE 12
Finite Element Methods - BME701 Module III 2025-26
Load vector because of UDL:
For element 1 that is subjected to UDL we have load vector as
ie
Element 2 and 3 does not contain UDL hence
Global load vector:
Department of Mechanical Engineering, MITE 13
Finite Element Methods - BME701 Module III 2025-26
And also we have external point load applied at node 3, it gets added to
F5 term with negative sign since it is acting downwards. Now F
becomes,
From KQ=F
At node 1 because of roller support
q1=0 Node 4 since fixed q7=q8=0
After applying elimination and solving the matrix we determine the
values of q2, q3, q4, q5 and q6.
Department of Mechanical Engineering, MITE 14
Finite Element Methods - BME701 Module III 2025-26
TRUSSES
ANALYSIS OF TRUSSES
A Truss is a two force members made up of bars that are
connected at the ends by joints. Every stress element is in either tension
or compression. Trusses can be classified as plane truss and space truss.
x Plane truss is one where the plane of the structure remain in
plane even after the application of loads
x While space truss plane will not be in a same plane
Fig shows 2d truss structure and each node has two degrees of freedom.
The only difference between bar element and truss element is that in
bars both local and global coordinate systems are same where in truss
these are different.
There are always assumptions associated with every finite element
analysis. If all the assumptions below are all valid for a given situation,
then truss element will yield an exact solution. Some of the
assumptions are:
Truss element is only a prismatic member ie cross sectional
area is uniform along its length
It should be a isotropic material
Constant load ie load is independent of
time Homogenous material
Department of Mechanical Engineering, MITE 15
Finite Element Methods - BME701 Module III 2025-26
A load on a truss can only be applied at the joints (nodes)
Due to the load applied each bar of a truss is either induced
with tensile/compressive forces
The joints in a truss are assumed to be frictionless pin joints
Self weight of the bars are neglected
Consider one truss element as shown that has nodes 1 and 2 .The
coordinate system that passes along the element (xl axis) is called
local coordinate and X-Y system is called as global coordinate
system. After the loads applied let the element takes new position
say locally node 1 has displaced by an amount q1l and node2 has
moved by an amount equal to [Link] each node has 2 dof in
global coordinate system .let node 1 has displacements q 1 and q2
along x and y axis respectively similarly q3 and q4 at node 2.
Resolving the components q1, q2, q3 and q4 along the bar we get two
equations as
Department of Mechanical Engineering, MITE 16
Finite Element Methods - BME701 Module III 2025-26
Or
Writing the same equation into the matrix form
Where L is called transformation matrix that is used for local –global
correspondence.
Strain energy for a bar element we have
U = ½ qTKq
For a truss element we can write
U = ½ qlT K ql
Where ql = L q and q1T = LT qT
Department of Mechanical Engineering, MITE 17
Finite Element Methods - BME701 Module III 2025-26
Therefore
U = ½ qlT K ql
Where KT is the stiffness matrix of truss element
Taking the product of all these matrix we have stiffness matrix for truss
element which is given as
Department of Mechanical Engineering, MITE 18
Finite Element Methods - BME701 Module III 2025-26
Stress component for truss element
The stress V in a truss element is given by
V= HE
But strain H= B ql and ql = T q
Therefore
How to calculate direction cosines
Consider a element that has node 1 and node 2 inclined by an angle T
as shown .let (x1, y1) be the coordinate of node 1 and (x2,y2) be the
coordinates at node 2.
Department of Mechanical Engineering, MITE 19
Finite Element Methods - BME701 Module III 2025-26
When orientation of an element is know we use this angle to calculate
and m as:
cosT m = cos (90 - T sinT
and by using nodal coordinates we can calculate using the relation
We can calculate length of the element as
Department of Mechanical Engineering, MITE 20
Finite Element Methods - BME701 Module III 2025-26
3
2
Solution: For given structure if node numbering is not given we have to
number them which depend on user. Each node has 2 dof say q1 q2 be
the displacement at node 1, q3 & q4 be displacement at node 2, q5 &q6
at node 3.
Tabulate the following parameters as shown
For element 1 T can be calculate by using tanT = 500/700 ie T = 33.6,
length of the element is
= 901.3 mm
Similarly calculate all the parameters for element 2 and tabulate
Department of Mechanical Engineering, MITE 21
Finite Element Methods - BME701 Module III 2025-26
Calculate stiffness matrix for both the elements
Element 1 has displacements q1, q2, q3, q4. Hence numbering scheme
for the first stiffness matrix (K1) as 1 2 3 4 similarly for K2 3 4 5 & 6
as shown above.
Global stiffness matrix: the structure has 3 nodes at each node 3 dof
hence size of global stiffness matrix will be 3 X 2 = 6
ie 6 X 6
Department of Mechanical Engineering, MITE 22
Finite Element Methods - BME701 Module III 2025-26
From the equation KQ = F we have the following matrix. Since node 1
is fixed q1=q2=0 and also at node 3 q5 = q6 = 0 .At node 2 q3 & q4 are
free hence has displacements.
In the load vector applied force is at node 2 ie F4 = 50KN rest other
forces zero.
By elimination method the matrix reduces to 2 X 2 and solving we get
Q3= 0.28mm and Q4 = -1.03mm. With these displacements we
calculate stresses in each element.
Department of Mechanical Engineering, MITE 23
Finite Element Methods - BME701 Module III 2025-26
Let Q1, Q2 …..Q8 be displacements from node 1 to node 4
and F1, F2……F8 be load vector from node 1 to node 4.
Tabulate the following parameters
Determine the stiffness matrix for all the elements
Department of Mechanical Engineering, MITE 24
Finite Element Methods - BME701 Module III 2025-26
Global stiffness matrix: the structure has 4 nodes at each node 3 dof
hence size of global stiffness matrix will be 4 X 2 = 8
ie 8 X 8
From the equation KQ = F we have the following matrix. Since node 1
is fixed q1=q2=0 and also at node 4 q7 = q8 = 0 .At node 2 because of
roller support q3=0 & q4 is free hence has displacements. q5 and q6
also have displacement as they are free to move.
In the load vector applied force is at node 2 ie F3 = 20KN and at node 3
F6 = 25KN, rest other forces zero.
Department of Mechanical Engineering, MITE 25
Finite Element Methods - BME701 Module III 2025-26
Solving the matrix gives the value of q3, q5 and q6.
Department of Mechanical Engineering, MITE 26
