Lecture Resume

4 Node Solid Beam Element - Q4SB

Isoparametric Concept
2D Elasticity
2D Element
• When one dimension is small compared to the other two dimensions the element may be
parameterized as a two-dimensional (2D) or surface element
• We distinguish between two main categories of 2D elements:
 Plate element ─ the surface is initially flat
 Shell element ─ the surface is curved
• If the plate/shell element is subjected to transverse loading
the analyst must choose which plate theory to apply:
 Thick: t / L > 1/3 : 3D continuum theory
 Moderately thick: 1/3 > t / L > 1/10 : Mindlin plate theory
 Thin: 1/10 > t / L : Kirchhoff plate theory
Beam Models
• FE modeling of beams are usually based on:
 Euler-Bernoulli (EB) beam theory -> also called classical or engineering beam theory

 Timoshenko beam theory -> also called Mindlin-Reissner beam theory

• Mathematically, the main difference is that the EB beam requires increased order of continuity
compared to the Timoshenko beam.
• The application of the EB theory is usually restricted to situations where dimensions along the
axis of the beam are at least ten times those of the transverse (cross-section) dimensions:
t / L < 1/10
• In contrast, the Timoshenko theory includes transverse shear deformations and is applicable
when the length to cross-section dimensions are above five (when smaller the continuum theory
becomes viable) : t / L < 1/5
Beam Models - Plane EB Beam – Basics Form
Beam Models - Plane EB Beam
Finite Element Approximation
• To meet the C1-continuity requirement, Hermitian cubic shape functions are used to approximate
the transverse displacements:
Beam Models – Timoshenko Beam – Basic Form
Beam Models – Timoshenko Beam – Weak Form

• Since the highest derivatives of transverse displacement v and section rotation θ in the weak form
are only first order, both fields may be interpolated by C0 functions.
• Use of an equal-order interpolation the transverse displacement and section rotation is expressed
as:
Beam Models – Timoshenko Beam – Mixed Form
• The use of Timoshenko theory can lead to locking effects when the theory is applied to cases where
the EB theory also could be used.
• However, it is desirable to have a single formulation which remains valid throughout the range of
length to cross-section considerations and for this the Timoshenko theory should be used.
• How can Finite Element approximations which are free from transverse shear locking be developed for
the Timoshenko beam as well as plate and shell elements based on Mindlin plate theory
• one method to alleviate locking effects is to apply reduced integration, a more general approach is to
apply a mixed formulation.
• In the sequel a three-field Hu-Washizu variational form will be used to construct mixed Finite Element
Approximations for the Timoshenko beam theory.
• The three-field weak form involves two displacement components, v and θ, two forces, V and M, and
two strains, γ and κ, as given by:
Beam Models – Timoshenko Beam – Mixed Form
• To obtain exact inter-element nodal displacements for the mixed formulation we let:

where M and V are particular solutions which satisfy the governing equilibrium equations:

• In addition we enforce the strain approximations to satisfy the constitutive equations, that for the
linear elastic problem reads:

• Integrating by parts we obtain

the corresponding reduced mixed form:
Beam Models – Timoshenko Beam – Mixed Form
• Eliminating at the element level we obtain the element ‘‘stiffness’’ matrix and the consistent load
vector (which may be assembled as usual)
Methods to Overcome shear Locking
• Methods to avoid transverse shear ‘‘locking’’ of flexural (structural) elements:
1. Mixed Formulation
2. Reduced Integration
3. Assumed natural Coordinate Strains (ANS) – based on 3 field formulation
• Methods to overcome shear ‘‘locking’’ in continuum elements:
1. Selective Reduced Integration (SRI)*
2. Incompatible Element
3. Enhanced Assumed Strains – based on 3 field Hu-Washizu formulation

* J-L Batoz et al. [3] developed the SRI method along with other method on plate bending element, such as DKT
Element and HSM (Hybrid Stress Method) Element, also had compared of those 3; finally concluded that the DKT
and the HSM elements presented better results and reliability in term of triangular plate bending element
Q4 bilinear quadrilateral (Q4 FEM)
• The displacement field u and the associated strain field ε
for the four-node bilinear rectangle having eight d.o.f
may either be established directly in terms of bilinear
shape functions in non-dimensional coordinates ( ξ , η),

or in terms of generalized coordinates where the Displacement

field is expressed in terms of physical coordinates x and y:
Q4 – Strain Fields
Q4 – Displacement Patterns
Physical interpretation of the eight linearly independent displacement patterns:
Rigid Body Modes
1. Translation in x-direction:
2. Translation in x-direction:
3. Rigid-body rotation:
Constant Strain Modes
1. Strain in x-direction:
2. Strain in y-direction:
3. Shear Strain:
Bending Modes
1. in x-direction:
2. In y-direction:
Q4 – Displacement Patterns
Formulation of The 4-Node Beam Element
In xz-plane, the displacement field (Figure 1) of existing 2-node beam element, with translational and
rotational degrees of freedom, is given by:
Formulation of The 4-Node Beam Element
The 4-node beam element
• The 4-node beam element with only translational degrees of freedom at each
node is shown
The 4-node beam element
• The two rotations θ1 and θ2 at the nodes 1 and 2 can be given by the following:
• In term of transverse displacement w0 to w3

the nodal displacement vector of a 2-node beam element, {un}, with rotational and translational d.o.f,
can be expressed in terms of the global nodal displacement vector of the developed 4-node beam
element with only translational d.o.f, {un*}, as follows:
The 4-node beam element
 Isoparametric Concept
• The concepts of isoparametric element and numerical quadrature, was created to overcome
the following difficulties with the classic method for the construction of the stiffness matrix
• The isoparametric concept in one-dimension is a method of standardizing the necessary
computations to build the stiffness matrix and the right-hand side forcing vector.
Quadrilateral Flat shell Elements
• Quadrilateral flat shell element is composed from 2 elements : isoparametric 4 node quadrilateral
elements (Q4) and Discrete Kirchoff Quadrilateral (DKQ) element
 Isoparametric Q4 element to model the membrane behavior. (2 dof)
 DKQ element are use to represent the plate bending behavior (3 dof)
• It has total 24 dof, 6 dof for each node:
 2-D Isoparametric Elements
• Isoparametric finite elements are based on the parametric definition of both coordinate &
displacement functions.
• Shape function Ni are defined in local coordinates (ξ; η), these shape function are used for
interpolations of displacement (ui ,vi) components values and coordinates

• Matrix Form

Interpolation matrix for nodal values is:

Linear Quadrilateral Element (Q4)
• The coordinates at a point within the element (natural coordinates) can be obtained by interpolation of
the coordinates on the global system:

• Shape function can be used to interpolate displacements (related to membrane behavior)

• Displacement field are bilinear function over the element

• Individual shape function:
Quadratic Quadrilateral Element (Q8)
• Quadratic Quadrilateral Element is widely used element for 2‐D problems due to its high accuracy in
analysis and flexibility in modeling
• There are eight nodes for this element, four corners nodes and four mid‐side nodes, with 8 shape
function in the natural coordinate system (ξ η)

• displacement field are quadratic functions over the element. Strains and stresses over a quadratic
quadrilateral element are quadratic functions, which are better representations.
2 Dimensional shape function
 Linear Quadrilateral Element (Q4) – Stiffness Matrix Formulation
• For plane problem the strain-displacement vector contains:

• Since the geometry on isoparametric elements is defined in the natural coordinate system, the
usual derivatives like above are not available directly, it is necessary to use the chain rule of
differentiation
Linear Quadrilateral Element (Q4) – Stiffness Matrix Formulation
• By inverting the Jacobian matrix is now possible to obtain the derivatives needed in equation

Now the strain-displacement relationships can be state as,

Linear Quadrilateral Element (Q4) – Stiffness Matrix Formulation
• Now the strain-displacement relationships can be state as,

Derivatives of displacements with respect to natural coordinates (ξ, η ) can be computed by

where the strain-displacement matrix [𝐵] is given by,

Linear Quadrilateral Element (Q4) – Stiffness Matrix Formulation
• The stiffness matrix of the Q4 element is given by:

• for constant thickness, the volume integral can be obtained multiplying the thickness by an area
integral,

Where:
• t is the thickness of the element,
• the [𝐵] is the strain matrix obtained
• [𝐷] is the constitutive matrix for plane stress state
DKQ Element Stiffness Matrix Formulation
• the DKT element, was based on the Kirchoff assumptions, the shear strain energy is neglected.
The formulation of the stiffness matrix of the DKQ element is based on the following equation

where is the element strain energy due to bending

where [𝐷𝑏] is the constitutive matrix relating bending moments and forces with curvatures [𝑘],

E = modulus of elasticity of the material

𝜐 = Poisson’s coefficient
t = thickness of the element
[k] = curvatures
DKQ Element Stiffness Matrix Formulation
Batoz et al [..] made the following assumptions for the formulation of the DKQ element:
1) 𝛽𝑥 and 𝛽𝑦 vary quadratically over the element
the 𝑁𝑖(𝜉,𝜂) are the following shape functions

2) The Kirchhoff hypothesis is imposed at:

DKQ Element Stiffness Matrix Formulation
3) The variation of w along the sides is quadratic,

4) A linear variation of 𝛽𝑛𝑘 is imposed along the side,

DKQ Element Stiffness Matrix Formulation
4) A linear variation of 𝛽𝑛𝑘 is imposed along the side,
DKQ Element Stiffness Matrix Formulation
• The strain-displacement matrix [𝐵] is obtained from the component vectors of the shape functions as,

• where the Jacobian is,

• the determinant of the Jacobian is,

DKQ Element Stiffness Matrix Formulation
Quadrilateral Flat Shell Element Stiffness Matrix Formulation
• The stiffness matrix of the quadrilateral flat shell element at each node can then be
represented by:
Equilibrium Equation:
• Total force on a small portion at any given point
• Resultant forces : N(x) ; T(x) ; M(x)

dM d 3vb
• Shear Forces (T) = T  GA    EI
dx dx 3

• Bending Moment (M) = M  EI   EI d  d 2

vb
  EI 2
 EI 
dx dx
Two-Dimensional Theory of Elasticity
Reference
[1] E-M. Meghlat, M. Oudjene, H. Ait-Aider and J.-L. Batoz: “A one dimentional 4-node shear flexible beam
element for beam-to-solid modeling in mechanically jointed connections made with screws or
nails”,ECCOMAS 2012, Vienna, Austria, September 10-14, 2012.
[2] Professor Irwan Katili : Lecture Note of Finite Element Method, Universitas Indonesia
[3] Batoz, J. L., Bathe, K. J. and Ho, L. W., “A Study of Three-Noded Triangular Plate Bending Elements”,
International Journal for Numerical Methods in Engineering, Vol. 15, pp. 1771-1812, 1980.
[4] Pedro Torres Fraga: The Finite Element Method Applied to Mechanical Structures, Efficient Finite
Element Code. Dissertação de Mestre em Engenharia Mecânica. Universidade Nova De Lisboa, March 2015
[5] Kjell Magne Mathisen: “Finite element modelling of structural mechanics problem”, Lecture 10 Geilo
Winter School - January, 2012
[6] G. P. Nikishkov: Introduction To The Finite Element Method, Lecture Notes. University of Aizu
Wakamatsu, Japan, 2004