2D Meshing 2D Meshing 2D Meshing 2D Meshing: When To Use 2D Elements When To Use 2D Elements
2D Meshing 2D Meshing 2D Meshing 2D Meshing: When To Use 2D Elements When To Use 2D Elements
2D Meshing 2D Meshing 2D Meshing 2D Meshing: When To Use 2D Elements When To Use 2D Elements
2D Meshing
This chapter includes material from the book “Practical Finite Element Analysis”. It also has been reviewed and has
additional material added by Matthias Goelke.
nce geometry cleanup is completed (e.g. surfaces are stitched together — no unwanted free surface edges inside the
geometry), meshing is next.
• Start with a coarse mesh and understand the modeling results; then use a finer mesh if needed.
• Try to keep mesh related uncertainties to a minimum if possible. Keep it simple as it can get more complicated on its
own.
The image above illustrates an “ugly” mesh. This chapter focuses on when to use 2D elements, how to create 2D elements of good
quality, and how to use HyperMesh to create 2D elements.
t/2
t - thickness of plate
1
- 2D Meshing -
Quite often the geometry of thin walled 3D structures, as shown in the image below, is simplified to a geometric model with lower
dimensionality. This is typically called a mid-surface model. The mid-surface
mid model is then meshed with 2D elements. Thus, there
is no need for a detailed volume mesh as the thickness of the geometry
geom is virtually assigned to the 2D elements. Mathematically,
the element thickness (specified by the user) is assigned with half in the + Z direction (element top) and the other half in the – Z
direction (element bottom).
2
- 2D Meshing -
2D element shapes
Tria Quad
The CST (Constant Strain Triangle) –An insidious survivor from the infancy of FEA, by R.P. Prukl, MFT
http://www.finiteelements.net/Papers/The%20CST.PDF .PDF (this paper is also uploaded to the Academic Blog)
The CST was the first element that was developed for finite element analysis (FEA) and 40-50 years ago it served its purpose well.
In the meantime, more accurate elements have been created ed and these should be used to replace the CST.
The Explanation
Consider a 3-noded plane stress element in the xy-plane with node points 1,
1 2 and
3. The x-deflections are u1, u2, u3 and the y-deflections v1, v2 and v3, totalling six values altogether.
The displacement function then has the following form (using six constants α ito describe the behavior of the element):
3
- 2D Meshing -
The strains in such an element are constants. We know, howeverr, that in a beam we have compression at the top and tension at
the bottom. Our single element is, therefore, not capable of modelling bending behavior of a beam, it cannot model anything at
all. Note: Three-noded elements for other applications than plane stress and strain are quite acceptable, e.g., plate bending and
heat transfer.
The Remedy
Use elements with four nodes. We then have eight constants to describe the behavior of the element:
The strain in the x-direction is now a linear function of its y-value. This is much better than for the triangular element. Use triangular
elements which also include the rotational degree of freedom about the z-axis normal to the xy-plane.
4
- 2D Meshing -
Family Of 2D Elements
1) Plane Stress
Ux Ux
Uy Uy
y Total dof = 8
Ux Ux
x
z
Practical Applications: Thin sheet metal parts, like aircraft skin, narrow
nar beams
5
- 2D Meshing -
Ux Ux
Uy Uy
y Total dof = 8
Ux Ux
x
z
Plane stress and plane strain elements are used for 2D (planner) problems.
θ θ
y y
θx θx
Uz Uz
θy θy
Total
tal dof = 12
θx θx
y Uz Uz
x
z
Ux Ux
θz θz
Uy Uy
Total
tal dof = 12
Ux Ux
y θz θz
x
z
6
- 2D Meshing -
5) Thin Shell - Thin shell elements are the most general type of element.
( Ux , Uy , Uz , θx , θy , θ z) = Uz , θx , θy + Ux , Uy , θz
U yθy U y θy
Uxθx Uxθx
U zθ z U zθ z
U yθy U y θy
Total
tal dof = 24
Uθ Uθ
y x x x x
Uzθz Uz θ z
x
z
Practical Application: Thin shell elements are the most commonly used elements.
Why is the word ‘solid’ in the name of a 2D element? This is because though the elements are planner, they actually represent
a solid. When generating a cylinder in CAD software, we define an axis of rotation and a rectangular cross section. Similarly,
for an axi-symmetric model we need to define an axis of rotation
tation and a cross section (planer mesh). The 2D planer mesh is
mathematically equivalent to the 3D cylinder.
U U
y y
Ux Ux
Uy Uy
y Total dof = 8
Ux Ux
x
z
7
- 2D Meshing -
Ф 50
1000 10,000 N
10 mm thick
Analytical Answer
The Stress Concentration Factor (SCF) is defined as = max. stress / nominal stress
Hence, the maximum stress = Stress Concentration Factor (SCF) * nominal stress = 3 N/mm2
In the first part of this study the effects of element type (quad versus
sus tria elements) on the modeling results are investigated. The
global mesh size is 100.
The boundary conditions for all models are the same: the translational
degrees of freedom (x-, y-, z- displacements=0) of all nodes along the
left edge of the model are constrained (green symbols) whereas the
nodes along the right edge are subjected to forces in the x-direction
direction (total
magnitude 10.000 N).
In order to better control the mesh pattern surrounding the hole, two so
called “washers” have been introduced (i.e. the initial surface is trimmed
by two circles with a radii of 45 mm and 84 mm, respectively.
In the stress contour plots shown below, the maximum principal stress
is depicted, respectively. Note that by default the element stresses for
shell (and solid elements) are output at the element center only. In other
words, these stresses are not exactly the ones “existing” at the hole. To
better resolve the stresses at the hole, the element stresses are output
at the grid points using bilinear extrapolation (in HyperMesh activvate the
Control cards > Global output request > Stress > Location: Corner).
Corner)
8
- 2D Meshing -
Model 1: The hole is meshed with 16 tria elements. The maximum principal stress (corner location) is 2.32 N/mm2 (the analytical
result is 3 N/mm2).
Model 2: The hole is meshed with 16 quad elements. The maximum principal stress (corner location) is 2.47 N/mm2 (the analytical
result is 3 N/mm2).
9
- 2D Meshing -
Despite the fact, that both results differ significantly from the analytical reference value of 3 N/mm2, it is apparent that the quad
elements (17 % error) are “better” than tria elements (23 % error).er All in all, both models are inappropriate when it comes to
quantitatively assessing the stresses at the hole.
Moreover, another even more important lesson to be learned is that the FEM program does not tell you that the mesh is inacceptable
– this decision is up to the CAE engineer
In the following the effect of element si ze, i.e. number of elements at the hole, on the modeling results is discussed.
182
82
- 2D Meshing -
Model 5: The hole is meshed with 16 quad elements. The maximum principal stress (corner location) is 2.47 N/mm2 (the analytical
result is 3 N/mm2).
Model 6: The hole is meshed with 64 quad elements. The maximum principal stress (corner location) is 3.02 N/mm2 (the analytical
result is 3 N/mm2).
11
- 2D Meshing -
While we have been looking at linear elements (nodes at the element corners only), Professor Dieter Pahr (University Vienna,
Austria). documented the differences between second order tria and quad elements His study is also based on a plate with a hole.
The image below is taken from his course notes: “Modeling, Verification
erification and Assessment of FEM simulation Results”.
In the figure above, the von Mises stress is depicted for linear and second order quad- and tria- elements. The vertical axes is error
(%), and the horizontal axes is number of elements at ¼ of the hole
In this image, the displacements at the hole is depicted for linear and second order quad- and tria- elements
Conclusion:
Second order quad elements behave/perform
beh best, whereas linear
trias are “problematic”. These element related effects are less
severe while looking at displacements (nodal results).
If this is so, then why not always create a very fine mesh with the maximum possible number of nodes and elements? Why is the
usual guideline for meshing 12-16 elements around holes in critical areas?
The reason is because the solution time is directly proportional too the (dof)2. Also large size models are not easy to handle on the
computer due to graphics card memory limitations. Analysts havve to maintain a fine balance between the level of accuracy and
the element size (dof) that can be handled satisfactorily with the available
a hardware configuration.
12
- 2D Meshing -
It is based on a simple exercise like the one above. The results of different mesh configurations are compared with a known
analytical answer and the one which gives the logical accuracy with a reasonable solution time is selected.
Most of the industries follow the following thumb rule for the number of elements on holes:
• critical areas = 12
• general areas = 6
During meshing, an option named “biasing” may be used. In our dayd to day life, we use the word bias the following way, “My boss is
biased, he prefers my colleague even though both of us have equal qualification and efficiency”. In the same manner, when
element length is not the same on the edge and biased towards a point, then it is known as biased meshing. Different commercial
software calculate bias differently. One of the simplest schemes is where the bias factor is a ratio of the maximum element length
divided by the minimum element length.
The above geometry was split along the diagonal and the bias was
as defined on the diagonals (at the edge point near the circular
hole).
13
- 2D Meshing -
6 6
6
6 6 6 6
6
6 6
6
Biased Mesh Exercise:
E
Distribution of elements
Nodes Elements
Stress N/mm2
(exact answer 3
N/mm2)
No Bias 1.59 168 144
Bias 5 2.13 168 144
Bias 10 2.41 168 144
Bias 15 2.56 168 144
TheBias 20
conclusion from the 2.65
previous example was that the 168
higher the number of 144
elements in the critical
area, the higher the accuracy. The example on biasing shows that even without increasing the number of
elements, one can achieve a better result just by the appropriate arrangement of the nodes and elements.
This was done at no extra computational cost.
Full Plate:
14
- 2D Meshing -
Half Plate:
y
Horizontal edge restraint: U y θ x θz (246)= 0
x
z
The advantage of symmetric boundary conditions is that the same accuracy is achieved at a lesser computational time and cost.
Symmetric boundary conditions should not be used for dynamic analysis (vibration analysis). It can not calculate anti nodes.
2) Mixed mode
3) Equilateral tria
15
- 2D Meshing -
The image above depicts a geometry that has been partially meshed.
meshed How would you “link” both meshes? Depending on whether
some single tria elements are allowed, you may use the mixed meshing method or the pure quad mesh. Both meshes are shown
below. Note the more homogeneous mesh pattern resulting from “mixed” meshing.
The mixed mode element type is the most common element type used due to the better mesh pattern that it produces (restriction:
total tria % <5). Sometimes for structural analysis or for convergence
ergence and better results for a non linear analysis, the pure
quadrilateral element meshing option is selected.
If Quads Are Better Than Trias, Why Not Always Mesh Using Only Quad Elements? Why Do Fea Software Provide The
Option For Tria Elements?
1) Mesh transition: In structural and fatigue analysis, rather than a uniform mesh, what helps is a small element size in the
critical areas and a coarse mesh or bigger elements in general areas. This type of mesh gives good accuracy with manageable
dofs. Trias help in creating a smooth mesh transition from a dense mesh to a coarse mesh.
2 ) Complex geometry: Geometry features like rib ends or sharp rp cutouts demand for the use of triangular elements. If quads are
used instead of trias, then it will result in poor quality elements.
4) Tetra meshing (conversion from Tria to Tetra): For tetra meshing, all the outer surfaces are meshed using 2D triangular
elements and then trias are converted to tetras. This methodology
thodology is discussed in detail in the next chapter.
16
- 2D Meshing -
In the figure above, the left image shows meshing with equilateral
eral trias, while the right image shows meshing with the option
R-trias (which simply splits quad elements into two tria elements).
elements)
The ideal shape for a triangular element is an equilateral triangle and is theoretically better than a R-tria element. But for the
following specific applications, R-trias have an advantage over equila
uilateral trias.
1) Tetra Meshing
17
- 2D Meshing -
The ribs are modelled using quad elements in three layers as shown
wn below and then split to R-trias. The average section thickness
is assigned to each different layer.
1 mm
t1avg = 1.5 mm
2 mm
t2avg = 2.5 mm
3 mm
t3avg = 3.5 mm
4 mm
18
- 2D Meshing -
Advantages:
1) If the geometry is changed, then the mesh will also change automatically.
au
19
- 2D Meshing -
20
- 2D Meshing -
What is acceptable.
21
- 2D Meshing -
For a better representation of the hole geometry and smooth mesh flow lines, holes should be modelled with an even number of
elements (like 6, 8, 12,16 etc. rather than 5, 7, 9 or 13).
8) Nodes should lie properly on the surface, with no deviation (and no kinks).
Switch off the element mesh lines and observe the contour (in parrticular at curvatures). Kinks as shown above are not acceptable.
22
- 2D Meshing -
9) Follow the feature lines (nodes should lie exactly on the edges).
10) Instead of a zig-zag distribution, a structured or smooth mesh is recommended (nodes aligned in a straight line)
Use of a“smooth” option, provided by most of the commercial softtware, helps in achieving systematic mesh.
11) For crash analysis, follow the mesh flow line requirement.
23
- 2D Meshing -
Not recommended for crash analysis Not recommended for structural analysis
13) For crash analysis, constant mesh size (by using trias) is preferred
erred (due to a minimum element length and a time step criteria).
Variable mesh size not recommended for crash but recommended for
or structural analysis
• drag: Builds elements by dragging a line, row of nodes, or group of elements along a vector.
24
- 2D Meshing -
• elem offset: Builds elements by offsetting a group of elements in the direction of their normals.
• spin: Builds elements by spinning a line, row of nodes, or group of elements about a vector.
• shrink wrap: Builds 2D (optionally 3D) simplified meshes of existing complex models. Further details are given further
below and in the chapter about 3D (solid) meshing.
11.9 Automeshing
The Automesh panel is a key meshing tool in HyperMesh. Its meshing module allows you to specify and control element size,
density, type, and node spacing, and also perform quality checks before accepting the final mesh.
The optimal starting point for creating a shell mesh for a part is to have geometry surfaces defining the part. The most efficient
method for creating a mesh representing the part includes using the Automesh panel and creating a mesh directly on the part’s
surfaces. The Automesh panel can be accessed from the menu bar by selecting Mesh > Create > 2D AutoMesh. A part can be
meshed all at once or in portions. To mesh a part all at once, it may
m be advantageous to first perform geometry cleanup (please
see Chapter 3 on Geometry) of the surfaces, which can be done in HyperMesh.
Provided your geometry is clean e.g. the surface of the rib is merged with the adjacent surfaces, then the resulting mesh will be
automatically compatible (all elements are connected with each other).
o
25
- 2D Meshing -
2. If you do not use surfaces, the meshing process is usually faster and uses less memory. Most of the functions are still
available and operate in the same way. Furthermore,
thermore, there are situations in which it is not possible or not desirable
to create a surface.
For either method, the module operates the same. You interactively
ely control the number of elements on each edge or side and can
determine immediately the nodes that are used to create the mesh.
mesh You can adjust the node biasing on each edge to force more
elements to be created near one end than near the other, which allows
all you to see immediately the locations of the new nodes.
You can also specify whether the new elements should be quads, trias, or mixed and whether they should be first or second order
elements.
The created mesh can also be previewed, which allows you to evaluate
e it for element quality before choosing to store it in the
HyperMesh database. While you are in the meshing module, you y can use any of viewing tools on the Visualization toolbar to
simplify the visualization of complex structures in your model.
If you use surfaces, you can specify the mesh generation and visualization options to use on each individual surface. You may
choose from several mesh generation algorithms. Mesh smoothingthing is also available and you may select the algorithm for that
operation as well.
While working with the Automesh panel you will come across the following
f options and settings:
Element size = the element size in the model may deviate from
om the specified size considerably (it depends on the size of the
surface).
Elems to surf comp vs. Elems to current comp = specifies the “storage“
orage“ place of the elements.
Start meshing, explore the meaning of the other settings latter! What may happen is that the mesh looks a bit weird …
26
- 2D Meshing -
Some surfaces apparently cause trouble. This may not actually be a problem, but a matter of your visual settings. In this example,
the geometry is still shaded, overprinting the mesh in some spots
ts. Displaying the geometry in wireframe and shading the elements
will improve the mesh “visibility“.
Note: In case you don‘t see any mesh, check the Model Browser and the status of the corresponding collector (is the elem icon
activated?)
A shrink wrap mesh can be generated as a surface mesh (using a loose or tight wrapping), or as a full-volume hex mesh by use
27
- 2D Meshing -
of the Shrink Wrap panel. This panel is located in the menu bar by
b selecting Mesh > Create > Shrink Wrap Mesh. The distinction
between surface or volume mesh is an option labeled generate solid mesh. The Shrink Wrap panel is covered in more detail in
the 3D meshing section.
In the menu, go to ‘Preferences’ > ‘Meshing options’ panel, or press the <o> key.
Specify the element size in the element size field, this specifies a default edge length for elements, in the same units that
the loaded model was created in (mm, inches, etc). This determines the default values to be used for meshing, such as
in the automesh panel.
2. What To Do When You Get The Error One Or More Surfaces Failed To Mesh? How To Retrieve The
Failed Surfaces From Meshing?
In 2D > automesh > there are two options: 1. failed surfaces,
aces, 2. unmeshed surfaces. Click on one of the options accord-
28
- 2D Meshing -
Use the Coarsen Mesh utility to simplify the mesh by combining many small elements into a smaller number of larger
ones. This dialog allows you to pick the Components that you wish to simplify, any Hard points (such as those defining a
hole or ridge) that must be preserved, a new element size, and a mesh type (Mixed or Trias-only). In both cases, you
must click the selector twice, as if you were accessing its extended entity selection menu; however, the second click
opens a temporary panel in the panel area. This panel allows
all you to select the desired components or nodes, and then
proceed in order to close the panel and return to the Coarsen Mesh dialog.
Once you set the desired options, you can “Mesh” the selected
selec components. If the results are not satisfactory, you can
“Reject” the new coarse mesh, change the options, and try t again.
Manually
Detach elements from its neighbouring mesh and remesh the selection can be a bad idea if a good connectivity is
required or the number of elements is too large.
Create surface from elements, delete the selected elements and remesh the surface, which has points on its edges
created from the surrounding mesh. Based on the difference
erence of old and new mesh sizes, can provide a very good con-
nectivity
29
- 2D Meshing -
5. How To Create A Matching Mesh On Two Opposing Identical Surfaces Or How To Project Elements On
To Surface?
Tool page >project->to surface,
You may need to duplicate the existing elements so that a copy of the elements are projected.
Select elements or components > select the elements or components using the extended entity selector > preview
equivalence > if the automatic selection based on the tolerance
olerance specified is satisfactory > click equivalence.
7. How To Make Changes To The Configuration Of Existing Elements? For Instance How To Change A 1D
Bar
Element To A Beam Element?
Go to 1D > config edit > select existing elements > select the necessary config and click on switch.
30
- 2D Meshing -
8. How To Change The Order Of Your Elements From First Order To Second Order Elements?
Go to 2D > order change > select elements and click on change order
o
9. How To Fill One Or More Holes In Your Geometry With Automatically-Generated Mesh?
31
- 2D Meshing -
Activating the “mesh” option in the “2D->automesh” panel opens up the density-mesh style-biasing-checks subpanel. In
the density subpanel, click the selector, adjust: “edge” to
o make it active.
From the graphics area, left-click an edge’s element density number to increase it by one.
Right-click an edge’s number to decrease it by one. Click and hold the mouse pointer on an edge’s number and drag the
mouse up or down to increase or decrease the number.
Overall, don’t forget that the models are built on sometimes quite
e drastic simplifications of physical phenomena. Just recently, a
nice post addressing this very topic was published on the Blog “Altair
Altair Simulate to Innovate” :
How Refined Should a Mesh be? Where are the Limits? ( frrom J. Marczyk, founder and CTO of Ontonix).
32
- 2D Meshing -
The following tutorials are part of the HyperWorks installation (check the “help-documentation”
“help for details):
• HM-3100: AutoMeshing
Recommended Videos
• Automeshing (http://altair-2.wistia.com/medias/f555xw0boc)
2.wistia.com/medias/f555xw0boc)
33
- 2D Meshing -
34
- 2D Meshing -
35