Scribd
Upload
24 views20 pages

FEA Knowledge 1.5

The document provides guidance on various modeling considerations when performing finite element analysis. It discusses issues like fixed supports causing singularities, applying symmetric and antisymmetric boundary conditions correctly, preventing rigid body motion with constraints, ensuring loads match reality, checking reactions, evaluating stress results, verifying models through simplified test cases, and performing sensitivity analysis. Proper modeling practices are important to obtain accurate and reliable results from finite element analysis.

Uploaded by

APURV THE Black Horse DRV
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
24 views20 pages

FEA Knowledge 1.5

The document provides guidance on various modeling considerations when performing finite element analysis. It discusses issues like fixed supports causing singularities, applying symmetric and antisymmetric boundary conditions correctly, preventing rigid body motion with constraints, ensuring loads match reality, checking reactions, evaluating stress results, verifying models through simplified test cases, and performing sensitivity analysis. Proper modeling practices are important to obtain accurate and reliable results from finite element analysis.

Uploaded by

APURV THE Black Horse DRV
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 20

Be Careful about Fixed Supports

• The Poisson effect: when


something is pulled in one
direction, it shrinks in the
perpendicular directions
• A fixed support will prevent this
shrinking, which leads to a
singularity
• Is
I th
there a potential
t ti l off ‘lift
‘lift-off’
ff’ att
the support? Might need to model
support and contact
Symmetric and Antisymmetric Constraints

• Symmetry BC: Out-of-plane


displacements and in-plane
in plane
rotations are fixed.
• Antisymmetry BC: In-plane
displacements and outout-of-plane
of plane
rotations are fixed.

• Many packages have a single


command to apply these constraints
Symmetric Constraints

pplane of symmetry
y y

deflected shape
Antisymmetric Constraints

deflected shape
Prevent rigid body motion

• For structural static, need to constrain for


t
translation
l ti and d rotation.
t ti
• Even though the beam below is in
equilibrium you need a constraint in the
equilibrium,
horizontal direction, or you will get ‘pivot
errors
errors’
Pivot Errors?

• The program doesn’t have any known values


f di
for displacements
l t iin th
thatt di
direction,
ti so it can’t
’t
solve the matrix
Prevent Rigid Body Motion

• Set loads to zero and run– if model doesn’t converge,


or results look odd, you have rigid body motion.
• Or, add density and do modal analysis
Loading

• Do the loads match real life?


• There are no point loads in real life, just really
small areas with pressures on them
• Investigate all possible combinations off loads
• Make sure you have density entered if turning
on gravity
it or d
doing
i modald l analysis
l i
• Might need to model load application
device/structure
Loading

• Consider range of load values


– Parametric analysis of different values
• Consider long-term vs. short-term loads
– Long-term analyzed for creep and fatigue
– Short-term analyzed for yielding
• Think about the load path through the
structure
St. Venant’s Principle

Two systems of loads which are statically equivalent will


produce approximately the same effect at locations
remote from the loads.

• This can be used to simplify loads or


structures
Check Reactions

• For statics, sum of forces = sum of reactions


• Can identify misplaced loads, incorrect units,
geometry mistakes, typos
• Check
C reactions at contact pairs
Check for buckling

• Occurs when part of structure is slender, and


under
d compressive i stress
t ((will
ill nott h
happen
under tension).
• Linear buckling (eigenvalue) is
UNCONSERVATIVE!
• Nonlinear buckling is more accurate
accurate, but also
more difficult (need to use full Newton-
Raphson method)
Watch Your Errors and Warnings

• Warnings about loads or constraints not


applied
li d b
because th
the node
d or element
l t are
nonexistent
• Undefined material properties
Large differences in stiffness

• Possible source of problem when the ratio of


th maximum
the i and
d minimum
i i element
l t stiffness
tiff
coefficients should be less than 1e8
• Otherwise,
Otherwise the stiffer part of the model will
‘crash through’ the less stiff part
• Can indicate problem with inputs
Evaluating Results

• Stress criteria
• Factor of safety
• Is stress greater than yield?
• Don’t assume the results are correct!
• Are the displacements in the expected
range?
• Compare to tests or theory, when possible
• Does the displaced shape make sense?
• Check out stress hotspots
Evaluating Results

• Use linearization, if needed


• Check the whole model– don’t focus so much
on one spot, you miss a problem elsewhere
• Check
C reactions against applied loads
• Check contact pairs for penetration
• Check element error to assess mesh
• Plot unaveraged stresses
Results Verification

• Use deformed animation to check loads and


l k ffor cracks
look k iin model
d l
• Combined load behavior is sometimes
difficult to predict
predict– consider separating each
load into its own load case to check
Do Test Cases

• Do a simplified case to learn about behavior


– Do 2D instead of 3D, beam instead of 3D
– Bonded contact instead of frictional
– Elastic instead of plastic or hyperelastic
– Hand calcs!
• Find pitfalls BEFORE running a model for
several days…
Sensitivity Analysis

• Process of discovering the effects of model


i
input
t parameters
t on response
• Can provide insight into model characteristics
• Can
C assist in design off experiments
• Should be subject to same scrutiny as all
V&V
Sensitivity Analysis

• A probabilistic analysis using your


model and statistical data of input
parameters to see how much
variation there is in output
• Generally requires several
analyses– very time consuming Global Sensitivity
• Positive sensitivity indicates that
increasing the value of the
uncertainty variable increases the
value of the result parameter

Local Sensitivity

You might also like