Computer Aided Engineering Unit 2
Computer Aided Engineering Unit 2
Meshing Techniques
Discretization:
The process of dividing the body into an equivalent number of finite elements
associated with nodes is called as discretization of a structure in finite element
analysis. Each element is associated with the actual physical behaviour of the
body. The total number of elements involved and their size variation within a
given body are matters of engineering judgement. The size of the element should
be optimal. The smallest element discretized should produce useful result and
avoid computational effort.
The body is divided into smaller elements if the results vary rapidly and into
larger elements if the results are constant. Each element consists of nodes
depending on the order of the element chosen.
The discretization of the body is done by using the mesh generation programs.
Types of Elements
Importance of Element Quality
The shape of Elements in FE Analysis must be distorted from their Ideal shapes
when meshing the irregular or complex geometric shapes.
Every element is designed to work properly within a certain range of shape
distortion. Exactly how much distortion and what type of distortion is allowed before
an element degenerate depends on factors such as element type, numerical
procedures used in the element design, and so forth.
The Ideal shape of [2D Elements] a triangular element is an equilateral triangle and
a quadrilateral; it is a square.
The Ideal shape of [3D Elements] a Tetrahedron element is a Regular or Isosceles
Tetrahedron and a Hexahedron, it is a Cube etc.,
If the actual shape that the element assumes after mapping onto model geometry
differs too much from the natural shape, the element becomes degenerated and
produces erroneous results.
Skewness:
Skewness is the Angular Measure of Element quality with respect to the
Angles of Ideal Element Types.
It is one of the Primary Qualities Measures of FE Mesh. Skewness determines
how close to ideal (i.e., equilateral or equi-angular) a face or cell is.
There are two different methods for calculating Skewness for 2D elements.
Method-1: Calculation of Skewness for Triangular/Quadrilateral Elements
(Angular Measure)
Triangular Element:
Draw a line from each node to mid-point of its opposite side, draw another line
joining mid-points of other two sides measure the angles between two lines.
Repeat the step for all the three nodes and find all six angles (Θ1 to Θ6).
Skewness is calculated by subtracting Minimum angle from 90 degree.
Quadrilateral Elements:
Draw the lines joining the mid points of opposite sides and Measure the Angle
between these two lines (Θ1 & Θ2).
Skewness is calculated by Subtracting the Minimum angles of Θ1 & Θ2 from 90
degrees.
Note: The acceptable Range of skewness is " 0 ̊ to 45 ̊ " beyond which results
may to be close to the actual solution.
Plane-1 is formed from nodes: 1,2 &4 and Plane-2 is formed from nodes: 2,3 &4
Calculate normals to Plane-1 & Plane-2 as “n1 & n2”. Find the Angle between
n1 & n2 and mark it as Θ1.
Similarly, for Plane-3 & Plane-4 find the angle between normal n3 & n4 and mark
it as Θ2.
Maximum of the two Angles Θ1 & Θ2 is Reported as "Warping Angle".
Warping Factor:
Some of the Finite Element Softwares calls the Out of Planeness of an element as
Warping Factor instead of Warping Angle.
Ideal value of Warping Factor= 0 (Acceptable: 0 to 1)
Jacobian
Ideal value = 1.0 (Acceptable > 0.6)
In simple terms, the Jacobian is a scale factor arising because of the
transformation of the coordinate system. Elements are transformed from the
global coordinates to local coordinates (defined at the centroid of every element),
for faster analysis times.
Distortion
Ideal value = 1.0 (Acceptable > 0.6)
Distortion is defined as - | Jacobian | * AreaLCS / AreaGCS
LCS – Local Coordinate system
GCS – Global Coordinate system
Stretch
Ideal value: 1.0 (Acceptable > 0.2)
For quadrilateral elements stretch = Lmin × √2 / dmax
Taper
Ideal value = 0 (acceptable < 0.5)
Chord Deviation
This helps in determining how well curvatures have been modeled. It is defined
as the distance between the mid node of an element edge to the curved surface. It
is only applicable for linear elements.
Tetra Collapse
Ideal Value = 1.0 (Acceptable > 0.1) Tetra collapse = (min of h /√A) / 1.24
(The height of the tetra element is measured from each of the four nodes to its
opposite face, and then divided by the square root of the face’s area, the minimum
of the four resulting values (one per node) is then normalized by dividing it by
1.24.)
Two types method are used to demonstrate the mesh refinement or numerical
convergence of the solution: