0% found this document useful (0 votes)
93 views9 pages

Onded Contact Between Shell Faces in Ansys® Mechanical (Workbench) V14.5

Uploaded by

Trúc Nguyễn
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
93 views9 pages

Onded Contact Between Shell Faces in Ansys® Mechanical (Workbench) V14.5

Uploaded by

Trúc Nguyễn
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 9

ONDED CONTACT BETWEEN SHELL FACES IN ANSYS®

MECHANICAL (WORKBENCH) V14.5


Posted in Tips & Tricks - Finite Element Analysis (FEA) articles

ANSYS Mechanical (Workbench) has many settings for contact between surface body (shell)
faces. This article examines a setup to employ with bonded contact across a gap between
surface body midplanes in large deflection nonlinear analysis.

Figure 1: WB Mechanical Model with Parallel Flat Surface Bodies

Creation of a Bonded Contact Pair between Surface Body Faces


If surface bodies are created on the midplane of the thin solids that they approximate, the
surface bodies that lie on top of each other will have a gap between the midplanes. The gap
size will often be greater than the tolerance used in the automatic creation of contact pairs
when geometry is imported into Workbench Mechanical. Either the tolerance employed in the
creation of contact pairs will have to be modified, and the contact pairs regenerated, or manual
contact pairs will need to be created between overlapping surface bodies.

1
Figure 2: Surface Bodies with Thickness and Shell Mesh

Figure 2 above shows a pair of surface bodies, and the


mesh that results when they have been meshed with shell
elements. The shell thicknesses entered for the surface
bodies in the model close the gap between the surface
bodies so that they touch on a pair of faces. In this
example, there is a 4 mm gap between the surface bodies.
These surface bodies are intended to lie on the midplanes
of the shell elements, so in the example, both surface
bodies have been set to a thickness of 4 mm, as illustrated
in Figure 3 to the right. The upper image in Figure 2 shows
a thickened view of the shell elements, so that closure of
the gap is illustrated. Figure 3: Surface Body Thickness

2
As can be seen in Figure 1
above, the “Top” of each of
these surface bodies happens
to face in the –Y direction in
the global coordinate system.
Consequently, the Contact
and Target surfaces do not
face each other. To get the
contact pair to work, the
Details settings for Contact
Shell Face and for Target
Shell Face for the contact pair
can either be left as “Program
Controlled”, or set by the user
to choices of Top and Bottom,
so that the red Contact face
and the blue Target face will
look towards each other. This
is required to get the resulting
target and contact elements in
ANSYS to attach to each
Figure 4: Orientation of Contact and Target Shell Face
other without implied
Normals
penetration and solution
failure.

Continuing with the Details definition of the contact pair, the contact will be set to Bonded,
which permits both Symmetric contact with penalty-based formulations, and Asymmetric
contact with all formulations. In the present example, a Symmetric behavior is selected.

In Figure 5, the Formulation has been set


to “Augmented Lagrangian” although a
Penalty setting has also been satisfactory
in testing. The Detection Method has been
set to “Nodal-Projected Normal From

Figure 5: Definition and Advanced Contact Settings


Contact” which was found to produce deflection and stress plots that were preferable to those
resulting from alternative settings.

Figure 6 below shows the Detection Method choices in a drop-down listing. Users may want to
experiment with the consequences of these various choices. The Projected choice is a
relatively new addition to ANSYS.

Figure 6: Detection Method Alternatives

Figure 7 shows equivalent stress and Y deflection with Detection Method “Nodal-Projected
Normal From Contact”.

4
Figure 7: Equivalent Stress and Deflection with the 
“Nodal-Projected Normal From Contact” Detection Method

Contrast this with Figure 8, the consequence with “Program Determined”. The deflection plot
has similar amplitude, but the stress does not show the Figure 7 result, where “thick” region
stresses are higher at outer layers, and smaller at the bonded interface, which acts like the
mid-plane of an 8mm thick shell.

  Figure 8: Equivalent Stress and


Deflection with the “Program Determined” Detection Method

5
For this reason, the “Nodal-Projected Normal From Contact” of Figure 7 will be preferred. 

Other Detail settings in the Advanced section in Figure 5 have been left as Program
Determined in this example, although users may prefer to increase the Normal Stiffness factor,
and to set a Pinball Region manually.

Loading on the Model


The edge at one end of the surface body pair was fixed, and a force was applied to the other
end.

 Figure 9: Loads on the Surface Bodies

The force has components in the Global X and Y directions:


 

Figure 10: Force Components

The Analysis Settings for this example were set to perform a Large Displacement analysis,
with non-default Output Controls set to output contact information that would make it possible
to measure the force transmitted across the contact pair in the model. Force Results probes
are not satisfied until Contact Miscellaneous is set to “Yes”.
 

6
Figure 11: Analysis Settings--Large Displacement with Extra Output Controls

As is often the case in finite element analysis, if users are unsure whether a Large
Displacement analysis is warranted, a model can be run with Large Displacement activated,
and if Large Displacement was not required, convergence usually adds just one additional
iteration. 

Outputs from the Example Model


Many postprocessing objects were added to the test model, in order to observe the
7
consequences of contact pair options on deflections, stresses, force reactions across the
contact, and Contact Tool outputs:
 

Figure 12: Solution Objects

In Figure 12, two of the three Force


Reaction objects were suppressed
—“Contact (Underlying Element)
choice produced near-zero numbers,
and Contact (Contact Element) had
unsatisfied inputs in most cases. The
“Target (Underlying Element)” choice
produced a force result matching the
edge input forces applied.

Contact Tool results were requested


on the “Contact” side only, so
although a symmetric bonded
contact was requested, a result is
only plotted on the lower body.
Because the Top of the lower body
faces the –Y direction, results are
plotted on the -Y side face of the Figure 13: Contact Tool Results
thickened view of the lower surface
body. Figure 13 shows Status, Frictional Stress, Pressure and Sliding Distance. The pressure
and sliding distance values are non-zero because of the finite penalty stiffness values in the
contact elements.

Conclusions
After a survey of settings for a contact pair between the faces of two surface bodies, observing
stress, deflection, contact, and force reaction results in a large displacement model, it was
found that satisfactory results were seen when a face-to-face contact pair between the surface
8
bodies was set to Bonded, Symmetric, and a Detection Method of “Nodal-Projected Normal
from Contact”. The force across the contact pair could be measured with a Force Reaction
probe for the contact pair, with Extraction set to “Target (Underlying Element)”.

Figure 14: Contact Pair Inputs and Outputs

The above Figure 14 setting for Shell Thickness Effect did not have a substantial effect on
results, but was set to “Yes” to capture the offset of the contact surface from the midplane,
which might have a stronger effect with thicker shells. The Behavior entry in the Definition
section could have been “Asymmetric” or “Program Controlled” but was set “Symmetric” so
that the finite size of the contact and target elements would leave less disconnection at the
ends of the contact zone, and would cope better with variant element sizes on the contact and
target faces.

Users could try optional settings for many of the Details in the contact pair settings, including
assignment of the Contact Body and the Target Body, and manual assignment of the Contact
Shell Face, Target Shell Face, and Pinball Region size. In the present example, adequate
results were realized with many Program Controlled settings.

Convergence did not become an issue with these settings for the present Large Displacement
example. The Analysis Settings had tightened Nonlinear Controls settings for Force
Convergence and for Displacement Convergence in this example, although these are
generally at the discretion of the user.

Users should keep in mind that may of the complications of contacts with Surface Bodies and
their shell elements are avoided with Thin Surface method solid meshing with SOLSH190
elements of solid bodies.

You might also like