Altair Hyperworks Hypermesh 7 Basic Training Tutorial Day2

7/21/2019 Altair Hyperworks Hypermesh 7 Basic Training Tutorial Day2

http://slidepdf.com/reader/full/altair-hyperworks-hypermesh-7-basic-training-tutorial-day2 1/153

Altair ®

HyperMesh ®

Introduction to FEA

Pre-processing, Volume II

www.cadfamily.com EMail:[email protected]

The document is for study only,if tort to your rights,please inform us,we will delete

http://www.cadfamily.com/



Altair Engineering Contact Information

Web site www.altair.com

FTP site Address: ftp.altair.com or ftp2.altair.com or http://ftp.altair.com/ftp

Login: ftp

Password: <your e-mail address>

Location Telephone e-mail

North America 248.614.2425 HyperWorks [email protected]

HyperMesh [email protected]

HyperForm [email protected]

OptiStruct [email protected]

MotionView [email protected]

HyperView [email protected]

HyperGraph [email protected]

HyperStudy [email protected]

HyperOpt [email protected]

Process Manager [email protected]

China 86.21.5393.0011 [email protected]

France 33.1.4133.0990 [email protected]

Germany 49.7031.6208.22 [email protected]

India 91.80.265885401600.44.0234 (toll free)

[email protected]

Italy 39.800.905.595 [email protected]

Japan 81.3.5396.134181.3.5396.2881

[email protected]

Korea 82.31.728.8600 [email protected]

Scandinavia 46.46.286.2052 [email protected]

United Kingdom 44.2476.323.600 [email protected]

The following countries have distributors for Altair Engineering: Australia, Brazil, Czech Republic, Greece,New Zealand, Romania, South Korea, Singapore, Spain, and Taiwan. See www.altair.com for complete

contact information.

© 2004 Altair Engineering, Inc. All rights reserved.

Altair ® HyperWorks ® 7.0 Release Notes

Trademark and Registered Trademark Acknowledgments HyperForm, HyperGraph, HyperMesh, HyperOpt, HyperShape, HyperStudy, HyperView, HyperView Player,HyperWeb, HyperWorks, MotionView, MotionSolve, Process Manager, BatchMesher and OptiStruct areregistered trademarks of Altair Engineering, Inc.

All other trademarks and registered trademarks are the property of their respective owners.

hm_basic_day2_draft.pdf, created on Oct 29, 2004






Chapter 1:

Preparing Geometry forMeshing

Proprietary Information of Altair Engineering, Inc.

HyperMesh 7.0 Introduction to FEA Pre-Processing 1

Chapter 1: Preparing Geometry for Meshing







] Realities of using CAD geometry for FEA:

• Geometry requirements for CAD are different from FEA− CAD needs an exact representation

− FEA only needs a simplified representation

− Depending on the problem, the part can be simplified for easier meshing or analysis

• Geometry can sometimes have errors when imported

] Preparing geometry for meshing allows us to:

• Correct any errors in the geometry from import

• Create the simplified part needed for the analysis

• Mesh a part all at once

• Ensure proper connectivity of mesh

• Obtain a desirable mesh pattern & quality

Preparing Geometry for Meshing

This allows much less cleanup after meshing!

] Topics:

• Import and repair CAD geometry data (topology repair)

• Create mid-surfaces

• Simplify geometry details (defeaturing)

• Modify surface topology to obtain quality mesh (topology refinement)

Preparing Geometry for Meshing

Location in the CAE Process

PreparePrepare

GeometryGeometry

Design ModelModel Analyze Visualize Report

MeshSet up for

Analysis


2 Introduction to FEA Pre-Processing HyperMesh 7.0







] Geometry page: geom cleanup, defeature, surface edit,

autocleanup, midsurface

Preparing Geometry for Meshing: Tools

] 2D page: automesh : cleanup

] Geom page of the macro menu

• Has some useful tools for working with geometry

not found elsewhere in HyperMesh

• Has shortcuts to functions in geom cleanup,

automesh : cleanup, defeature, surface edit

panels, and more.

− Shortcuts are organized by the type of entity they

work with

− Very useful for quick access to various functionswhen cleaning geometry

Preparing Geometry for Meshing: Tools











Chapter 1:


Section 1:

Repairing CADGeometry Data










Topology Repair

] Topics:• Importing geometry

• Understanding topology and topology repair

• Process of repairing topology

• Tools for repairing topology

• Hands-on practice

Topology Repair: Importing Geometry

] Import geometry data via the files -> import sub-panel

] Common types of geometry files supported:

• Unigraphics (UG Part Browser)− Import of *.prt files− Requires an installation of Unigraphics to

be available

• CATIA− import of *.model??? files− CATIA v5 license required to import CATIA

v5 files

• Pro/ENGINEER− import of *.prt and *.asm files

• IGES− Import of *.igs / *.iges files

• STEP− import of *.stp files









Topology Repair: Surface Definitions

Surface

] Fixed point (“Surface vertex”)

• Point associated with a surface

• Cannot be moved off the surface

• Can lie on a surface edge or the

“interior” of a surface

• Separates surface edges from

each other

• Forces a node to be placed at that

location during meshing

] Surface edge

•

Line associated with a surface• Defines a surface’s boundary

• Cannot be moved off the surface

• Has a fixed point on both ends

Topology Repair: What is “Topology”?

] “Topology” is how surfaces connect to adjacent surfaces of a part

• Surface connectivity is controlled by the associated surface edges

• If a surface edge is associated with more than 1 surface, those surfaces are

considered to be connected (“equivalenced”)

• Surface edges are categorized, named, and colored according to the number

of associated surfaces:

Free edge (red)

• Associated with only 1 surface• Surfaces with a free edge between

them are NOT equivalenced at thatedge

Shared edge (green)• Associated with 2 surfaces

• Surfaces are equivalenced

Non-manifold edge (yellow)• Associated with 3 or more

surfaces•Example: surfaces forming a T-

connection

• Surfaces are equivalenced

Suppressed edge (blue)•Surfaces are treated as though

combined into 1 surface









Topology Repair: Viewing topology] Topology display mode is default in the geom cleanup

and automesh panels

] Can also be accessed via vis opts on Disp page of

macro menu:

Always in topology display mode, with surface

shading added

3

Always in topology display mode, even outside the

geom cleanup and automesh panels

2

Always in component color mode (geom cleanup

and automesh panels will not use topology display)

1

default (topology display in only geom cleanup andautomesh panels)

0

] vis opts in the geom cleanup panel has control of:• Wireframe vs. shaded display of surfaces

• Visibility of free, shared, non-manifold, and suppressed

edges, and of fixed points

] HyperMesh will try to properly clean up surfaces duringimport• Some types of geometry files have surface connectivity

information which helps HyperMesh

• Geometry usually imports cleanly

] Topology Repair consists of correcting connectivityerrors between adjacent surfaces

• Possible errors include:− Unconnected adjacent surfaces

− Duplicate surfaces

− Missing surfaces

• Goal is to restore the surface data to a good representation ofthe part

Topology Repair: What is it?









] If the edge is red (free):

• Adjacent surfaces are not connected to each other (free edge pair)

• Adjacent surface is missing

] If the edge is green (shared):

• Edge is common to 3 surfaces (T-connection) & should be yellow rather

than green

• Edge is on the perimeter of the part and should be free (red) rather

than green (sliver or duplicate surface)

] If the edge is yellow (non-manifold):

• Edge is common to only 2 surfaces and should be green rather than

yellow (sliver or duplicate surface)

Topology Repair: Possible errors for topology color

] General process is to:

• Figure out what the surface connectivity of the part should be

• Observe the current display of topology colors

− Figure out what is causing the topology to be displayed this way

• Use the tools in HyperMesh that get the connectivity from what it is to

what it should be as quickly and efficiently as possible

Topology Repair: Process

Duplicate surface deletedDuplicate surface deleted

Missing surface createdMissing surface created

Free edge pair equivalencedFree edge pair equivalenced

Example:









] geom cleanup : edges sub-panel• equivalence (multiple edges at a time)

− search surfaces for pairs of free edges and combine into shared edges

• toggle (1 edge at a time)− Left mouse click on surface edges to go from free shared and shared

suppressed

− Right mouse click on surface edges to go from suppressed shared andshared free

− Free edges combined to shared if edges are within tolerance specified

• replace− combine pairs of free edges with gaps between them into a shared edge

− Control which edge to keep and which to move

] geom cleanup : surfaces : find duplicates

• identify and delete duplicate surfaces

] surface edit : filler surface• Select a closed loop of free surface edges to recreate any missing

surfaces

Topology Repair: Tools

] Understand model size & scale and determine global element size

] Set a cleanup tolerance based on the determined global element size

• Set value in options, geom cleanup, and automesh : cleanup

• Cleanup tolerance specifies the largest gap size to be closed by topology functions

• Tolerances > 15-20% of global element size can lead to significant mesh distortions

• Can change value multiple times for work on various areas of the model

] Use topology display tools to decide what needs to be cleaned

] Use equivalence to combine as many free edge pairs as possible

• Make sure surfaces are not collapsed in undesirable manner

] Use toggle to combine any remaining free edge pairs, 1 by 1

• use replace function if more control is needed

] Use find duplicates to check for any duplicate surfaces and delete them

] Use filler surface to recreate any missing surfaces

• use vis opts 3 mode from macro menu to view shaded surfaces

Topology Repair: Strategy











Exercise

Step 1: Retrieve and view the model file.

The model for this exercise is /Day_2/clip_repair.hm. Take a few moments toobserve the model using the different visual options available in HyperMesh (rotation,zooming, etc.).

Step 2: View the model in topology display and shaded mode to evaluate itsintegrity.

1. Observe where the model has incorrect connectivity and missing or duplicate surfaces.

2. Enter the geom cleanup panel on the Geom page.

Note that the surface edges are now colored according to their topology status.

3. Click the vis opts button to open the visual options menu.

Vis opts controls the display of the surfaces and surface edges. Surfaces can beshaded or wireframe, and the check boxes within this menu turn the display of thedifferent edge types and fixed points (surface vertices) on or off.

4. Uncheck everything except the free edges.

5. Move the mouse cursor off the pop-up menu to close it.

Only the free edges should be displayed at this point.

6. Observe the free edges and make a mental note of where they are.

The free (red) edges show where there is incorrect connectivity or gaps.

7. Note the locations where there are closed loops of free edges. These are locations thatprobably have missing surfaces.









Free edges indicating surface discontinuities of the clip geometry

8. In vis opts, turn the display on for only the non-manifold edges.

9. Observe the non-manifold edges and make a mental note of where they are.

The non-manifold edges show where there are more than two surfaces sharing an edge,which might be incorrect connectivity. For this part, there yellow edges completely

surrounding two areas. This tells us there are probably duplicate surfaces in theselocations.

10. In vis opts, turn on the display for all the edge types and fixed points.

11. In vis opts, toggle wireframe to shaded.

The surfaces should now appear solid rather than having only their edges displayed.

12. Rotate, zoom, and pan to locate any errors in the geometry.

13. Make a mental note of the areas to be worked on. We find:

• A surface that overhangs a round corner

• A missing surface









Surface overhanging an edge and a missing surface

• An edge that has apparently been collapsed

Area of collapsed edge









14. In vis opts, toggle shaded back to wireframe.

Step 3: Delete the surface that overhangs the round corner.

1. Enter the delete panel on the Tool page.

2. Select the overhanging surface.

3. Delete the surface.

4. Return to the main menu.

Step 4: Create surfaces to fil l large gaps in the model.

1. Use the surface edit / filler surface subpanel to fill in the areas where a surface ismissing.

2. From the Geom page, enter the surface edit panel.

3. Go to the filler surface subpanel.

Note that in the surface edit panel, the surface edges are displayed in component color,not topology display mode. Topology display mode can be activated using the macromenu.

4. On the Disp page of the macro menu, select Vis opts: 2 to display the surface edges inwire frame topology mode.

5. Back in the surface edit /filler surface subpanel, set the entity type to lines.

6. Make sure auto create is checked (active).

The auto create option simplifies the selection of the lines bounding the missingsurface. Once a line is selected, HyperMesh automatically selects the remaining freeedges that form a closed loop, and then create the filler surface.

7. Turn the keep tangency option off.

The keep tangency option looks at surfaces attached to the selected edges and tries tocreate a surface tangent to them. This helps to form a smooth transition to thesurrounding surfaces.

8. Zoom into the area indicated in the following image.









Area of missing surfaces

9. Pick one of the red lines bounding one of the holes.

HyperMesh automatically creates a filler surface to close the hole.

10. Repeat #8 to create a filler surface in the other gap.


Step 5: “ Release” the fixed points in the area of the collapsed edge.

1. Use the release tool on the geometry c leanup / fixed po ints subpanel to disconnect allof the fixed points and edges in the area of the collapsed edge.

2. Enter the geometry cleanup panel on the Geom page.

3. Go to the fixed points / release subpanel.

4. Rotate and zoom in on the area of the collapsed edge.

5. Select the point indicated in the image below to release the fixed point.

6. Two fixed points will separate, and the edges connected to them will all become freeedges.









Fixed point to be released


Step 6: Set the global geometry cleanup tolerance to .01.

1. Go to the options panel on the permanent menu.

2. For cleanup tolerance = specify 0.01 for stictching the surfaces with a gap less than0.01.


Step 7: Combine multip le free edge pairs at one time with the equivalencetool.

1. Use a tolerance of 0.01.

2. Enter the geometry cleanup panel on the Geom page.

3. Go to the edges / equivalence subpanel.

4. Select surfs >> all.

5. Verify that the cleanup to l= is set to 0.01, which is the global cleanup tolerancespecified in the options panel.

6. Equivalence the edges to combine any free edge pairs within the specified cleanuptolerance.









Most of the red free edges are combined into green shared edges. The few remainingare caused by gaps larger than the cleanup tolerance.

Step 8: Combine free edge pairs, one pair at a time, using the toggle.

You should still be in the geom cleanup / edges subpanel.

1. Go to the toggle subpanel.

2. For cleanup tol =, specify 0.1.

3. In the graphics area, click on one of the free edges shown in the image below.

4. Rotate and zoom into the area if needed. When the edge is selected, it will change fromred to green, indicating that the free edge pair has been equivalenced.

Area where free edges need to be toggled

5. Use toggle to equivalence the other edges shown in the image.

Step 9: Combine the remaining free edge pair using replace.

You should still be in the geom cleanup / edges subpanel.

1. Go to the replace subpanel.

2. From the view panel, click restore1 to bring back the saved view.

3. With the selector under retained edge: active, click on the rightmost free edge.









The selector under edge to move: becomes active automatically.

4. Select the leftmost red edge.

5. Replace the edge.

Once the line is selected, HyperMesh posts a message similar to:

Gap = (.20000). Do you still wish to toggle?

6. Click Yes to close the gap.

Edges to retain and move for replacement

Step 7: Find and delete all duplicate surfaces.

You should still be in the geom cleanup panel.

1. Go to the surfaces subpanel.2. Go to the find dupl icates subpanel.

3. Select faces >> displayed.

4. Specify 0.01 for the cleanup tol=.

5. Click find.

The header bar displays the following message, “2 faces are duplicated”.









6. Delete the identified duplicate surfaces.

Step 9: Observe the model again to identify any remaining free edges, ormissing or duplicate surfaces.

Use the topology display and shaded modes to perform this task. All of the edges in themodel should be displayed as green shared edges, indicating that we have a completelyenclosed thin solid part.

Return to the main menu.

Step 10 (Optional): Save your work

With the cleanup operations completed, now is a good time to save the model.

SummaryThe integrity of the surfaces has been repaired to what would be desirable on import. Themodel is now ready for the creation of a midsurface.









Chapter 1:


Section 2:

Midsurfacing










] For many FE analyses, parts are represented by “shell” elements

• Thickness is assigned mathematically, rather than geometrically

• Mesh is placed on the midplane of the part

] CAD geometry usually comes as a solid part

] Midsurfacing creates a layer of surfaces on the midplane which can

be directly meshed

Midsurfacing

] Midsurfaces can be created using midsurface : create on the

geom page

• solid automatically extracts midsurfaces from surfaces that enclose a

volume

− Can sometimes work if there are missing surfaces

− The greater number of missing surfaces, the less reliable the result

• between surfs creates a midsurface between 2 selected surfaces

Midsurfacing: Tools









] Once a midsurface has been created, it can be modified using tools

on midsurface : edit sub-panel.

• quick edit: allows a midsurface to be repaired by correcting where the

verticies of the surface were placed

• assign target: an extension to quick edit, and functions in a similar

fashion.

• replace edge: fill in gaps and slivers by combining one surface edge

with another. same as the one in the geom cleanup panel

• extend surface: extends two surfaces (e.g., ribs) until they intersect

• view thickness: allows review of the thickness of a midsurface using

white lines (probes) extending from each vertex of the surface

Midsurfacing: Tools for Editing Midsurfaces

] midsurface : create : solid requires an enclosed volume

• Use topology repair techniques in section 1 to get a quality, enclosed

volume of surfaces

] For complex parts, the volume of surfaces may need to be

defeatured in order to generate a good midsurface (see section 3)

] Generate the midsurface using midsurface : create : solid

• Use between surfs for areas that need more control

• Use midsurface : edit for midsurfaces that need fine tuning] View the midsurface and correct errors using the midsurface editing

functionalities

Midsurfacing: Process / Strategy

















Chapter 1: Section 2 Exercise

Generate a Midsurface

PurposeThis exercise uses CAD geometry data for a thin solid clip. Because of the small thicknessof the part, it is assumed that it will be modeled for FEA as shell elements. The elements willbe created on the mid-plane of the part.

In order to create the model, you will use the midsurface panel in HyperMesh to build a setof surfaces directly on the mid-plane of the part. This allows a mesh to be created directlyon those surfaces. Also, the midsurface that is generated can then be modified usingdefeaturing and geometry cleanup tools to obtain a desirable mesh. Defeaturing themidsurface is much easier than the volume, since there are fewer surfaces to work with.

Problem Statement

This exercise uses the model file, clip_midsurf.hm.

The surfaces in the model have no connectivity errors. This could be because the file wasimported without errors or because the errors were corrected using HyperMesh. In thiscase, errors in the topology were repaired in the previous exercise (missing surfaces are re-created, duplicate surfaces are deleted, gaps are closed, etc.). You can use either the fileyou from the previous exercise, or open the new, clip_midsurface.hm, file. Either way,the geometry is at the point where you can use the midsurface panel to generate amidsurface.

Once the file is loaded, the following actions are performed:

1. Use the midsurface panel to automatically create the midsurface.

2. Use vis opts on the macro menu to visualize the midsurface.









Exercise


This exercise uses the model, /Day_2/clip_midsurface.hm.

Step 1: Generate a midsurface from midsurface panel.

1. From the Geom page, enter the midsurface panel.

2. Go to the create subpanel.

3. Go to the solid tool.

4. With the yellow surfs selector active, select one surface from the graphics area.

5. Click extract to start the midsurface generation.

The midsurface is created, and the surfaces are organized into a new MiddleSurface component.

Step 2: Review the part’s midsurface.

1. Click the Vis opts: 3 button on the Disp page of the macro menu to shade the surfaces.

2. The midsurfaces generated for the solid sections of the model using the solid subpanelare shown in the image below.









Midsurface generated from a volume of surfaces

3. Go to the surface transparency panel on the permanent menu.

4. With the comps selector active, select a line or surface of the lvl0 component in thegraphics region.

You may need to zoom in on the model to select a valid entity.

5. Since the entity selector is set to comps, selecting a line or surface in the componentselects the entire component.

6. Click the arrow button on the right under the entity selector several times.

You should see the surfaces in the lvl0 component becoming more and moretransparent.

You can also drag the slider back and forth to control the level of transparency.

7. Rotate, zoom, and pan to visualize the midsurface.











Chapter 1:


Section 3:

Simplifying Geometry










] Depending on the analysis, certain details in the geometry may be

ignored. This may depend on:

• Importance of the part in the overall assembly

• Location of the feature relative to the area of interest in the analysis

• Size of the feature vs. the average size of the mesh being used

] Defeaturing is the removal of details in the geometry in order to

make the shape of the part simpler

Defeaturing: What is it?

Surface fillet

Pinholes

Edge fillets

] defeature panel on geom page

• pinholes searches for closed loops of free edges contained within a

surface. It looks for a range of diameters and removes them, leaving a

fixed point at the center. The pinhole does not have to be a circle.

• surf fillets deletes a surface that acts as a fillet between two other

surfaces, and then tangentially extends those to meet at a sharp corner

• edge fillets searches for rounded corners on a surface whose radii fall

within a range and squares the corner

• trim in tersect performs the same operation as edge fillets, except theuser manually indicates where square corners should be made

Defeaturing: Tools










Simplifying Geometry

PurposeThis exercise involves changing the shape of a part in order to simplify the geometry.Certain details of the shape, such as small holes or blends, may simply not be necessary forthe analysis being performed. When these details are removed, the analysis can run moreefficiently. Additionally, mesh quality is often improved as well. Changing the geometry tomatch the desired shape can also allow a mesh to be created more quickly.

Problem Statement

This exercise uses the model file, clip_defeature.hm.

The model file has geometry that has been midsurfaced. (Surfaces have been created onthe mid-plane of the part.) The model will be meshed using an element size of 2.5. You canassume a simple structural analysis will be run on the part, and thus does not require muchdetail. Based on this, there are features, which are not necessary and can be removed.

Once the file is loaded, we need to perform the following actions:

1. Mesh the clip, review the mesh quality, and determine the features to be simplified.

2. Remove surface fillets.

3. Remove edge fillets.

4. Remove pinholes.









Exercise


The model for this exercise is /Day_2/clip_defeature.hm.

Step 2: To easily work wi th the midsurface, turn of f the display of the lvl0component.

If the lvl0 component is displayed, it needs to be turned off so that you can easily work onthe midsurface geometry. (It may be on if you used the model you had open for theprevious section.)

1. Enter the disp panel on the permanent menu.

2. Set the collector type selector to comps if it is not already set.

3. Set the entity type selector to geoms.

4. Right click in the check box for the lvl0 component to turn off the display of thegeometry in that component.


Step 3 (Optional): Mesh the clip to view mesh quality before defeaturing.

1. Click per on the Disp page of the macro menu to set the graphics to performance mode.

2. Enter the automesh panel on the 2D page.

3. Go to the create mesh subpanel.

4. Set the selector type to surfs.

5. Specify 2.5 for elem size =.

6. Set the element type to mixed.

7. Switch the meshing mode from interactive to automatic.

8. Select surfs >> displayed to select all displayed surfaces.

9. Click mesh to generate the mesh preview.









Initial mesh on the clip model


Step 4 (Optional): Review the quality of the mesh.

Take a minute to look over the mesh that was created. Note the areas that have irregular,poor quality mesh. Also, use the check elems panel to evaluate the minimum length checkof the elements.

Take a minute to rotate, zoom, and pan the model to review the mesh that was created.Note the locations where the mesh was not created in neat rows and columns of quads.

1. Enter the check elems panel on the Tool page.

2. Go to the 2D subpanel.

3. In the field for length, enter a value of 1.

4. Click length to evaluate the minimum length.

Many of the elements failing the length test are located around the fillets of this model.









Elements failing the length check


6. Use the disp panel on the permanent menu to turn off the display of the elements in thelvl0 component.

Step 5: Remove the four small pinholes.

Pinholes are closed free edge loops within a surface. Pinholes do not need to be circular.

1. From the Geom page, enter the defeature panel.

2. Go to the pinholes subpanel.

3. Specify 3.0 for the maximum search diameter .

4. Select surfs >> all.

5. Click find to identify the pinholes having a diameter of 3 or less.

Notice the xP symbol at the center of the four circular holes. These symbols arehighlighted in white, indicating they are pinholes identified by HyperMesh for removal.









Pinholes identified using a 3 mm diameter

6. Click delete to remove the selected pinholes in the model.

The selected pinholes are removed and replaced by fixed points located at the center ofthe original pinholes.

Step 6: Remove all surface fillets in the clip.

1. Use the defeature / surf fillets subpanel.

2. If the surfaces are not shaded, select Vis opts: 3 from the disp page of the macromenu.

3. In the defeature panel, go to the surf fillets subpanel.

4. Change the selector for fillet profiles to surfs.

5. Select surfs >> displayed.

6. Specify2.0

for the minimum radius.7. Click find to identify all the surface fillets with radius of 2 or greater.









Surface fillets identified for removal

8. Remove the selected surface fillets.

Step 7: Automatically identify and remove rounded corners of surfaces.

You should still be in the defeature panel.1. Go to the edge fillets subpanel.

2. Select surfs >> displayed.

3. Specify 1.0 for the minimum fill et radius.

4. Set the switch under filter params to all .

5. Click find.

The edge fillets that meet the filter criteria are identified on the screen with an F symboland radial lines marking the fillet beginning and end.









Edge fillets identified for removal

6. Remove the selected edge fillets.

Most of the fillets are removed. Only one remains.

Step 8: Manually remove rounded corners of surfaces.You should still be in the defeature panel.

1. Go to the trim intersect subpanel.

2. Rotate and zoom in to the area with the remaining fillet.

3. Pick to points as indicated in the image below.









Locations to select for the trim-intersect tool


Step 9 (Optional): Save your work.

Now that the model has been simplified, it is a good time to save the model.

Summary

The model is now represented in a much simpler form that suits the analysis that will beperformed. Holes, surface fillets, and edge fillets were removed that were considered toosmall to be captured by the desired element size of 2.5.









Chapter 1:


Section 4:

Refining Topology toAchieve a Quality Mesh










] In this section we will cover:

• How surface topology affects meshing

• What is “topology refinement”?

• Tools for refining topology

• Tips for refining topology

• Hands-on practice

Topology Refinement

] Surface edges control how mesh created on adjacent surfaces interact:

Topology Refinement: How topology affects the mesh

] Automesher will ignore suppressed edges, and will not place

nodes on them

] Associated surfaces are in effect combined to form 1 larger

surface

Suppressed edges

] Automesher will place nodes along non-manifold edges

]Nodes placed along non-manifold edges are common to mesheson all associated surfaces (“equivalenced”)

Non-manifold

edges

] Automesher will place nodes along shared edges

]Nodes placed along shared edges are common to meshes onboth associated surfaces (“equivalenced”)

Shared edges

] Automesher will place nodes along free edges

]Nodes placed along free edges will not be shared with elements

on nearby / coincident edges

Free edges









] “Topology refinement” is modifying topology in order to obtain a quality

mesh

• Unlike defeaturing, this generally does not change the shape of the part

] CAD geometry has topology details that interfere with mesh quality

• Edges are created where ever there is a change in surface curvature

• Even smooth areas can be split into several faces

• The automesher will be forced to place nodes along the edges

− This may cause small element lengths, angles, aspect ratios, etc. depending on the

shape

] CAD geometry can also have surfaces with a complex outline

• Highly complex shapes can make it hard to get a quality mesh

• Adding in edges splits the surfaces into smaller and simpler regions which can

be meshed easier

• Fixed points can be added in to force a node to be placed at that location, giving

more control

Topology Refinement: What is it?

Topology Refinement: Typical actions

Suppressing

edges

Adding

edges

Removing

fixed points

Adding

fixed points

Replacing

fixed points









Topology Refinement: Example

Aft er

topology

refinement

(edges are

suppressed)

Before

topology

refinement

MeshGeometry

Topology Refinement: Tools

] geom cleanup : edges sub-panel• toggle (1 edge at a time)

− Left mouse click on surface edges to go from free shared and sharedsuppressed

− Right mouse click on surface edges to go from suppressed shared andshared free

− Free edges combined to shared if edges are within tolerance specified

• (un)suppress (multiple edges at a time)− From shared (green) suppressed (blue) or suppressed (blue) shared

(green)

] geom cleanup : fixed points sub-panel• add – add new fixed points to a surface

• replace – combine fixed points together

• suppress – delete existing fixed points

] surface edit : trim with nodes• Split a surface (add an edge) by clicking points on opposing edges









] Mesh the part

• Visually scan the part to look for poor mesh patterns

• Use element checks in order to find areas of poor element quality

] Use geom cleanup : edges : suppress with various settings for break

angle in order to suppress many edges at once

] Sometimes suppressing all edges and then unsuppressing desired edges

can be a good method

] Suppress all the fixed points before add more in

• Gets rid of as many fixed points as possible, leaving ones that are required

] Experiment!

• There is no set process, so experience is a key factor in refining topology

Topology Refinement: Tips


















Refining Topology to Achieve a Quality Mesh

Problem Statement

Once a set of surfaces has been obtained that accurately reflects the intended shape for theanalysis, further geometry cleanup may still be necessary. Topological details of thegeometry may affect the quality of the mesh created from the surfaces. These details maynot reflect any major feature of the part’s shape, and can be removed without concern.When modifying the topology affects the shape of the surfaces, a compromise must bemade between the part shape and the element quality necessary for the analysis. Othertimes, adding topological features that do not change the shape of the part may actuallyhelp create a better quality mesh.

Solution

The HyperMesh file contains surfaces that have been topologically repaired, midsurfaced,and defeatured. The clip has a few remaining topological details that will result in poormesh quality. Identify and improve these areas of the surfaces. Then create a shell meshwith an average element size of 2.5.

The following actions will be performed:

1. Mesh the part to find areas of poor element quality and/or mesh pattern.

2. Suppress small edges.

3. Split surfaces to add new edges.

4. Remove interior fixed points.

5. Replace closely placed fixed points.

6. Create final mesh.









Exercise

Step 1: Open the model file.

The model for this exercise is /Day_2/clip_refine.hm. Take a few moments toobserve the model using the different visual options available in HyperMesh (rotation,zooming, etc.).

Step 2: Create a preliminary mesh.

1. Click per on the Disp page of the macro menu to set the graphics to performance mode.



4. Set the selector type to surfs.5. Specify 2.5 for elem size =.

6. Set the element type to mixed.

7. Switch the meshing mode from interactive to automatic.


9. Mesh the surfaces.

Initial mesh on the defeatured clip model










Step 3: Review the mesh quality.

1. Take a minute to rotate, zoom, and pan the model to review the mesh that was created.

Note the locations where the mesh was not created in rows and columns of quads.


3. Go to the 2D subpanel.

4. In the field for length enter a value of 1.

5. Click the length button to evaluate the minimum length.

6. Note the elements that failed the check. The topology will be edited to correct of someof these, and the others will be left as is.


8. Use the disp panel on the permanent menu to turn off the display of the elements in thelvl10 component.

Step 4: Remove short edges by combining fixed points.

1. On the replace points : line, click moved to make it active.

2. Select the lower fixed point as indicated in the image below.

3. Once the point is selected, the retain button becomes active.

4. Select the upper fixed point as indicated in the image below.

Once the second point is selected, the two points are merged together.









Selecting fixed points to be combined

Step 5: Remove the fixed points interior to all surfaces.

You should still be in the automesh / cleanup subpanel.

1. On the remove point: line, click point to make it active.2. Select the four fixed points as shown below.

Each fixed point will be deleted as you select it.

These fixed points are left over from a defeaturing operation where small holes wereremoved. They could remain without greatly sacrificing the element quality, given theelement size used for the mesh.









Fixed points to be removed


Step 6: Add edges to the surfaces to contro l the mesh pattern.

1. Enter the automesh panel on the 2D page.2. Go to the cleanup subpanel.

3. On the split surf: line, click node to make it active.

4. Zoom into the area indicated below and select the indicated fixed point.

5. With the active selector now on line, select the line shown in the image below.

Once both the point and line are selected, an edge will be created from the location ofthe fixed point perpendicular to the line.









Select fixed point and line to split the surface.

6. Repeat #3 and #4 for the following point and line.

Select fixed point and line to split the surface.











Select fixed point and line to split the surface

Step 7: Add edges to the surfaces to control the mesh pattern.

1. Enter the surface edit panel on the Geom page.2. Go to the trim surf subpanel.

3. Select the surfaces indicated in the image below.

Surfaces to be selected for splitting

4. Click N1 to make it active.

5. Press and hold your left mouse button, and then move it over the edge indicated inFigure 9. Once over the line, the cursor will change. Release your mouse button.

6. Click two points anywhere along the edge. Do not click a third.

Nodes will be placed on the line for N1 and N2.

7. Press F4 on the keyboard to enter the distance panel.

8. Go to the three nodes subpanel.

9. Press and hold your left mouse button, and then move it over the edge of the hole, asindicated in the following image. Once over the line, the cursor will change. Releaseyour mouse button.











Surface edges to suppress by toggling

Step 9: Remesh the part.

Remesh the surfaces of the part, using the automatic mode, a size of 2.5, and the mixedmesh type.

1. Use the disp panel on the permanent menu to turn on the display of the elements in thelvl10 component.



4. Verify that elem size = is set to 2.5 and the element type is set to mixed.


6. Remesh the surfaces.









Final mesh on the clip model


Step 10: Review the mesh quali ty.

1. Take a minute to rotate, zoom, and pan the model to review the mesh that was created.Note that the mesh now consists completely of rows and columns of quads.


3. Go to the 2-d subpanel.

4. Click the length button to evaluate the minimum length.

Note the elements that failed the check. There are only two elements, and these fail thecheck because of the shape of the part. However, they are not too small compared tothe global element size, so you can leave them as is.


Step 11 (Optional): Save your work

The part is now meshed and ready to be set up for an analysis. Save the model, if desired.

Summary









Although the model started with element quality that could be considered passable, a fewactions were performed to increase the quality even further. Modifications to the topologywere made to remove very small edges, and surfaces were split to add new edges to controlthe “flow” of the mesh. This resulted in a very high quality mesh.











] What are connectors?

] Why use connectors?

] Connector terminology & concepts

] Tools for working with connectors

Connectors: Agenda

Connectors: What are they?

] HyperMesh geometric entities

that define weld locations

] Create solver independent

connections between

geometric and/or FE entities

] Can be created interactively orfrom master weld / connection

file



Chapter 2: Assemblies






Connectors: Why use them?

] Connectors hold definition of:

• Weld locations

• Linked Entities (components)

• Re-Connect Rules

] Connectors manage realized FE type

and entities

• rigids, CWELDs, MAT100s

] Allows switching between different

FE weld types

] Allow quick part reconnection /

swapping• Replacement by names or IDs

Connectors: Terminology

] Link Entities - The entities that are being connected together

• User can explicitly define link entities or specify a search tolerance

• Typically components are linked

• Can be components, elements, surfaces, nodes, or tags

] Connector Location - Where the entities are linked

• Nodes - The connector is created at the node location

• Points - The connector is created at the point location

• Lines – The connector is created for the line

− line may be split into multiple projection locations as specified by the offset,

spacing, and density values

] Connector Realization – The creation of the finite element representation

of the weld for that connector

• Rigids, springs, etc., or custom configurations such as ACMs, CWELDS, etc.










] Connector State – Whether an FE representation of a weld has been

created for that connector

• Unrealized - The initial status of the connector entity upon creation

• Realized - The status only if creation of the FE weld representation at the

connector was successful

• Failed – The status if the creating the FE weld representation at the connector

was not successful

] # of Layers - number of FE weld layers to attempt to generate for the

connector

• 2T, 3T, etc.


] Connect When: - Specifies when the link entity information is added to the

connector

• None - Just creates the connector; no links added unless added by the user

• Now - Adds the link entity information when the connector is created

• At FE Real ize - Add the link entity information to the connector when the

connector is realized into its FE representation

] Re-Connect Rule - Defines method for connector re-attachment during

part swapping/replacement• None - If a link entity is deleted, the link entity is removed from the connector

• By ID - If a link entity is deleted, the connector retains the ID of the link entity,

and will to a new entity with that ID upon realization

• By Name – Same as the by id rule except that the entity name is retained











Connectors: Tools

] Info Table – Spreadsheet utility for editing connectors

• Located in most connectors panels

• Edit link entities, re-connect rule, export mwf files, etc.









Section 1: Assemblies

Following are four exercises which will help you become familiar with connecting (welding)

an assembly of parts, using various methods, and replacing parts with newer, similar partsand updating their affected connections. The part assembly used in the exercises isdepicted in the image below. A very brief description of each exercise follows. (Theexercises are independent of each other.)

Assembly of parts to connect

Exercise 1:

Weld the two front trusses to each other at pre-defined weld points between geometrysurfaces and between shell elements.

Exercise 2:Weld the right rails to each other and to the front trusses by importing a master connectorsfile. Then weld the left rails to each other and to the front trusses by duplicating andreflecting connectors. Lastly, identify and combine pairs of adjacent welds.









Exercise 3:

NASTRAN ACM (area contact method) welds already connect the rear trusses to eachother. Change the weld type.

Exercise 4:

Replace the rear truss component Rear_Truss_1 with a new, similar part and then update

its affected welds.









Exercise 1: Weld between Geometry Surfaces and Shell Elements

Purpose

Become familiar with creating welds at pre-defined weld points between geometry surfacesand shell elements.

Problem Statement

First, weld the two front trusses depicted in the image below. Do this as follows: 1) createconnectors between their geometry surfaces at pre-defined weld points, and 2) realize theconnectors into LS-DYNA material 100 beam welds.

Weld the two front trusses

Second, weld the two front trusses to the reinforcement plate depicted in the image below.Do this as follows: 1) create connectors between their shell elements at pre-defined weldpoints and, 2) realize the connectors into LS-DYNA material 100 beam welds.









Weld the two front trusses to the reinforcement plate

This exercise uses the file f r ame_assembl y_ 1. hm.

Step 1: Retrieve and view the model file frame_assembly_1.hm .

Step 2: Create connectors between the geometry for the two front trusses atthe pre-defined weld points.

1. On the 1D page, enter the connectors module.

2. Enter the create panel.

3. On the header bar, verify that the current component is Con_Frt_Truss.

4. Switch the location: entity selector to points.

5. Select the six pre-defined weld points by selecting points >> by collector and selectingthe component Con_Frt_Truss.

6. Switch connect when: to now.

7. For connect what:, select the components Front_Truss_1 and Front_Truss_2.

8. For connect what:, toggle elems to geom.

9. Create connectors at the selected weld points.

The six connectors are yellow, indicating they are not yet realized into finite elements.They are organized as geometry (not elements) in the current component collector,Con_Frt_Truss.









Step 3: Change the display size of the connectors.

You should still be in the create panel.

1. Click vis opts.

A dialog opens.

2. For size = specify 8.

3. Return to the connectors module menu.

New connectors connecting the front truss components

Step 4: Realize the connectors in the component Con_Frt_Truss into weld

elements.

You should still be in the connectors module menu.

1. Enter the fe realize panel.

2. Select connectors >> all .

3. Verify element conf ig: is set to weld.

4. For proj tol = specify 5.

5. Realize the selected connectors into weld elements.

6. Notice the color of the connectors changes from yellow to green, indicating they arerealized into weld elements.









7. Notice fixed points were added to the surfaces at the ends of the weld elements toguarantee connectivity between the weld elements and the shell mesh that will becreated on the surfaces.

Weld element with fixed points created on the surfaces


Step 5: Create a shell mesh on the two front t russ components.

1. On the 2D page enter the automesh panel.


3. Select surfs >> by collector and select the components Front_Truss_1 and

Front_Truss_2.

4. Activate the option reset meshing parameters to.

5. For elem size = specify 10.

6. Select the mixed element type.

7. Toggle from elements to current comp to elements to surface’s comp.

8. Select the mesh mode automatic. (It currently may be interactive or QI optimized.)



Step 6: Create connectors between the shell mesh for the front trusses andthe reinforcement plate at pre-defined points.

1. On the permanent menu, enter the global panel.

2. Set the current component collector to Con_Truss_Plate.

3. Enter the connectors module.

4. Go to the create panel.

5. Verify the location: entity selector is set to points.

6. Select points >> by collector and select the component Con_Truss_Plate.

7. Verify connect when: is set to now.









8. For connect what:, select the following components:

• Front_Truss_1

• Front_Truss_2

• Reinf_Plate

9. Also for connect what:, toggle geom to elems.

10. Create connectors at the selected weld points.

The connectors are organized into the current component collector Con_Truss_Plate.


Step 7: Realize the connectors in the component Con_Truss_Plate into weld

elements.


2. Select connectors >> by collector and select the component Con_Truss_Plate.


4. Verify 5 is specified for proj tol =.

5. Click fe options.

A secondary panel appears with options for the creation of the weld elements.

6. Activate the option, snap to node.

When this option is active, if the realized finite element of the connector is coincident to anode of the shell mesh it is being connected to, the nodes are equivalenced. When theoption is inactive, coincident duplicate nodes are created for the above condition.

7. Return to the fe realize panel.











The two front trusses welded to the reinforcement plate with weld elements at the connectors


Now that the exercise is complete, save your work to a HyperMesh file or just delete it.









Exercise 2: Weld Using a Master Connectors File and Duplicatingand Reflecting Connectors

Purpose

Become familiar with defining weld locations in HyperMesh by importing a masterconnectors file. Also, become familiar with duplicating, reflecting and updating connectors tocreate welds.

Problem Statement

First, weld the two right rails to each other and to the two front trusses depicted in the imagebelow. Do this as follows: 1) import weld point data from a master connectors file, 2) createconnectors, and 3) realize the connectors into LS-DYNA material 100 welds.

Weld the two right rails to each other and to the front trusses

Second, weld the two left rails to each other and to the two front trusses depicted in theimage below. Do this as follows: 1) duplicate and reflect the connectors that were created byimporting the master connectors file, 2) update the link information for the reflectedconnectors, and 3) realize the connectors into LS-DYNA material 100 welds.









Weld the two left rails to each other and to the two front trusses

Third, combine adjacent 2T connectors into 3T connectors in order to update adjacent 2Twelds to 3T welds.



Step 2: Create connectors to connect the right rails to each other and to thefront trusses by importing a master connectors file.

1. Enter the files panel.

2. Go to the import subpanel.

3. Activate the file type weld.

4. Verify the weld option is set to connectors.

5. Import the file, r ai l s_ f r t _t r uss . mwf .

6. Notice the connectors are automatically created and are organized into the newcomponent, CE_Locations.


Step 3: Realize the connectors in the component, CE_Locations, into weld

elements.

1. From the global panel, set the current component collector to CE_Locations.










3. Select connectors >> by collector and select the component CE_Locations.


5. Verify 5 is specified for proj tol =.

6. Click fe options and verify the option snap to node is active.



Right rails welded to each other and to the front trusses with weld elements at the connectors

Step 4: Display the components Left_Rail_1 and Left_Rail_2 for elements.

1. Enter the disp panel on the permanent menu.

2. Display the components Left_Rail_1 and Left_Rail_2 for elements.


Step 5: Duplicate the connectors created from the master connectors file andreflect them.

1. Enter the collectors panel.










3. Set the collector type to comps.

4. For name = specify CE_Locations_Dup.

5. Create the component.


7. On the Tool page, enter the reflect panel.

8. Switch the entity selector to connectors.

9. Select connectors >> by collector and select the component CE_Locations.

10. Select connectors >> duplicate >> current comp.

The displayed connectors are duplicated and the duplicates are organized into thecurrent component, CE_Locations_Dup.

11. Set the plane selector to x-axis.

This is the axis normal to the plane of interest.

12. Click edit to specify a base node to reflect about.

13. Click x =.

The fields for x =, y =, and z = are activated. By default their value is 0.000, which is thebase point you want to reflect about.

14. Return to the reflect panel.

15. Reflect the connectors.


Step 6: Update the connectors for the left rails to link them to the left railcomponents.

1. Enter the connectors module.

2. Enter the add links panel.

3. Select connectors >> by collector and select the component CE_Locations_Dup.

4. Click info table to open the Connector Information Table dialog.













3. Notice that the State column indicates that all of the connectors are realized.

4. Close the Connectors Information Table.

5. Under the connectors selector, click reset to clear the selection of connectors.

6. Zoom into one of the two areas where the front trusses are connected to the rail

components.

7. Notice that at these two areas, there are pairs of adjacent connectors.

8. Click vis opts.

9. Under CE colo r , activate the option layer .

10. While still in the vis opts dialog, notice under layers that 2t (two thickness) is purple.

11. Notice the connectors are now colored purple. This means each of these connectors linktwo components.

Because the pairs of adjacent connectors create a series of two weld elements, you can

combine each pair into a single connector, which links the three components together.


Step 9: Isolate the pairs of adjacent, 2t connectors identified in the previousstep.

1. From the disp panel, turn off all components for geometry.

2. On the Tool page, enter the find panel.

3. Go to the between subpanel.

4. Switch the entity type to find to connectors.

5. Switch the entity selector to comps.

6. Select the components Front_Truss_1 and Front_Truss_2.

7. Find the connectors between these components.

8. Under the comps selector click reset.

9. Select the components Front_Truss_1 and Right_Rail_1.


11. Click reset.

12. Select the components Front_Truss_1 and Left_Rail_1.

































2. Select the Nastran user profile from the drop down menu.

The elements from which you will create connectors are a type of NASTRAN weld calledan ACM weld.

3. In the connectors module menu, enter the fe absorb panel.

The Automated Connecto r Creation and FE Absorpt ion dialog opens.

4. Set FE Configs: to custom.

5. Set FE Type: to nastran 70 acm ((T1+T2)/2).

6. Toggle Elem filter: from all to select.

7. Click the Elem filter: Elements selector twice.

A HyperMesh panel menu with an elems selector appears.

8. Select elems >> by collector and select the component Con_Rear_Truss.

9. Click proceed to return to the dialog.

10. Activate the Move connectors to FE component option.

11. Absorb the elements into connectors.

Connectors are generated at the locations of the ACM welds. They are colored purpleindicating they are realized and are 2t connectors. Also, the connectors are organizedinto the Con_Rear_Truss, the same component to which the ACMs belong.

12. Close the dialog.


14. To turn off the display for the temporary nodes, go to the Disp page on the macro menuand click Mask: Nodes.

Step 3: Isolate the 2t connectors between the Rear_Truss_2 component and

the Right_Rail_2 and Left_Rail_2 components.

1. From the disp panel, turn off the display for all geometry components.

2. On the Tool page, enter the find panel.

3. Go to the between subpanel.

4. Switch the find entity type to connectors.

5. Switch the entity selector to comps.

6. Select the components Rear_Truss_2 and Right_Rail_2.


Five connectors are found.









8. Under the comps selector click reset.

9. Select the components Rear_Truss_2 and Left_Rail_2.


Five connectors are found.


Connectors between the rear_Rr2 component and the Left_Rail_2 and Right_Rail_2

components

Step 4: Add a third link to four of the ten displayed 2t connectors.

1. In the connectors module menu, enter the add links panel.

2. With the connectors selector active, select the four connectors indicated in the imagebelow.

Connectors to select

3. For connect what:, select the component Rear_Truss_1.

4. Verify that the connect what: toggle is set to elems.

5. Activate the search tol = option and specify 5 for it.









6. Switch # of layers: to total 3.

7. Click add links to update the connectors’ definition.

The connectors are now blue as they are 3t connections.

Step 5: Unrealize the connectors in the component Con_Rear_Truss.

8. In the connectors module, enter the unrealize panel.

9. Select connectors >> displayed.

10. Unrealize the connectors.

The weld elements associated to these connectors are deleted.

11. Return to the connectors menu.

Step 6: Realize the connectors in the component Con_Rear_Truss.

1. Enter the global panel and set the current component to Con_Rear_Truss.


3. Select the connectors >> displayed.

4. Verify that element con fig: is set to weld.

5. Verify that 5 is specified for proj tol =.

6. Click fe options and verify that the snap to node option is active.

7. Realize the connectors.












Exercise 4: Swap Welded Part

Purpose

Become familiar with swapping welded parts and updating their affected connections(welds).

Problem Statement

Replace the rear truss component Rear_Truss_1 with a new, similar part and update its

affected connections (welds). Do this as follows: 1) update the connectors to use the “usename” rule, 2) delete the old part, 3) import the new part, and 4) realize the correspondingconnectors into LS-DYNA material 100 beam welds.


This part (the Rear_Truss_1 component) will be swapped


Step 2: Update the connectors to use the rule “ use name” when reconnectingparts.

1. Enter the connectors module menu.

2. Enter the add links panel.









3. Select connectors >> by collector and select the component Con_Rear_Truss.

4. Click info table to open the Connector Information Table dialog.

5. At the top of the dialog, Configure the table.

The Configure Table View dialog opens.

6. Activate the Rule option.

7. Close the Configure Table View dialog.

8. Notice the rule is specified in the Link1, Link2, and Link3 columns. In this case, the ruleis none for the selected connectors.

9. Click the select all icon to select all connectors in the table.

10. For Find what:, change the Link:Rule to use-id.

11. For Replace with :, change the Link:Rule to use-name.

12. Update the connectors.

13. Close the Connectors Information Table.

14. Click vis opts and under CE color , select realize.


Step 3: Swap the component Rear_Truss_1 for a new version of it.

1. On the Tool page, enter the delete panel.

2. Delete the component, Rear_Truss_1.


4. Notice the color of the displayed connectors changes from green (realized state) toyellow (unrealized state) for those connectors associated to the deleted component. Thisis because the deleted component was one of the connectors’ links.

5. Notice the existing finite element welds are deleted.

6. Enter the files panel / import subpanel.

7. Activate the file type option HM MODEL.

8. Import the file, r ear _t r uss_ 1_new. hm.

Step 4: Realize the connectors in the component Con_Rear_Truss.

9. Enter the global panel and set the current component to Con_Rear_Truss.










11. Select the connectors >> displayed.

12. Verify that element conf ig: is set to weld.

13. Verify that 5 is specified for proj tol =.

14. Click fe options and verify that the snap to node option is active.

15. Realize the connectors.




















Chapter 3:

Tetra Meshing



Chapter 3: Tetrameshing







] “Standard” Tetrameshing

] Volume Tetrameshing

] Quick TetraMesh Macro

Tetrameshing: Agenda

] Standard tetrameshing uses tetra mesh sub-panel

] Process:

• Generate a surface mesh of plate elements

• Check quality and connectivity of the plate elements

• Generate the tetrahedral mesh

• Delete the original surface mesh

• Edit if necessary to obtain good quality

Tetrameshing: Standard Tetrameshing









] Requirements for the shell mesh:

• Enclose one, and only one, continuous volume

• There can be no free edges.

• There can be no T-connected edges.

• There can be no duplicates in the mesh.

• Elements should not fold over and overlap each other

• Avoid very low minimum tria angles

• Avoid a large difference in size between adjacent elements

• Avoid a large difference is size between two sides of a wall thickness


] Floatable Trias:

• Adjacent tria faces on the tetrahedral mesh may have their diagonal

reversed from the shell mesh if tetras are better quality


]Fixed Trias:• Adjacent tria faces on the tetrahedral mesh always match the shell

mesh

Shell Mesh Tetra Mesh Faces

-OR-

Shell Mesh Tetra Mesh Faces









] Volume tetrameshing uses the volume tetra sub-panel

] Provides a quick method of generating a tetramesh

] Generates 2 new components:

• auto_2d_tetra_ - holds 2D shell elements

• auto_3d_tetra_ - holes 3D tetra elements

] Has some specialized options:

• Use Proximity – Creates smaller elements next to small features to

make a smooth transition from small to large

• Use Curvature – Will place more elements along curved surfaces

based on user specified settings

Tetrameshing: Volume Tetrameshing

Tetrameshing: Volume Tetrameshing

No options Use surface

curvature

Use proximity

Use surface

curvature and

proximity









] Quick Tetramesh is found on theMesh page of the macro menu

] Provides a quick method ofgenerating a tetramesh

] Maintains specified minimum elementquality criteria

• Resulting tetramesh may deviate fromthe geometry to maintain goodelement quality

• User can select “sacred elements” or“sacred surfaces” to force thetetramesh to closely follow the originalelements/surfaces

Tetrameshing: Quick TetraMesh

] Options:

• Volume Complist – Selection of components that contain the geometry to be

tetra meshed.

• Maximum reverse angle – Angle between normals of adjacent elements.

Elements that are greater than the specified value are fixed.

• Maximum feature angle – Protects nodes on corners with a feature angle

greater than the value. Helps to better maintain the geometry.

• Mesh size – Average element size of the mesh to be created.

• Minimum mesh size – Minimum allowable length for any element.

• Sacred element – Selection of elements that you have meshed to your

requirements and would like to maintain while tetrameshing the part.

• Mesh type –Trias Only or Mixed.

Tetrameshing: Quick TetraMesh


















Tetra Meshing

PurposeHyperMesh provides two methods of generating a tetrahedral element mesh. The volumetetra mesher works directly with surface geometry to automatically generate a tetrahedralmesh without further interaction from the user. Even with complex geometry, this methodcan often generate a high quality tetra mesh quickly and easily.

The standard tetra mesher requires a surface mesh of tria or quad elements as input, thenprovides you with a number of options to control the resulting tetrahedral mesh. This offers agreat deal of control over the tetrahedral mesh, and provides the means to generate atetrahedral mesh for even the most complex models.

In this exercise, both methods of generating a tetrahedral mesh are explored. It also

addresses tetrahedral element quality and re-meshing.

Problem Statement

This exercise uses the model file, housi ng. hm.

The file contains two parts defined by a volume of surfaces. The geometry has beencleaned such that surface connectivity is proper and surface edges that would cause sliverelements are suppressed.

The following functionality is discussed:

1. Volume tetra mesher

2. Standard tetra meshing

3. Checking tetra element quality

4. Re-meshing tetra elements









Exercise


The model for this exercise is housing.hm. Take a few moments to observe the modelusing the different visual options available in HyperMesh (rotation, zooming, etc.).

Only the geometry in the component, cover , is currently displayed.

Step 2: Use the volume tetra mesher and equilateral triangles to create a tetramesh for the cover.

1. On the 3D page enter the tetramesh panel.

2. Go to the volume tetra subpanel.

3. With the surf selector active, select one of the surfaces in the model.

The connected surfaces are selected automatically.

4. Verify that 2D: is set to trias and 3D: is set to tetras.

These control the type of element that will be created for the surface mesh and solidmesh of the part.

5. Verify that the use curvature and use proximity options are off.

6. For element size= specify 10.

7. Click tetmesh to create the tetra mesh.

The volume tetra mesher creates two components. One is for the shell mesh, and theother is for the tetra mesh.

8. On the Disp page of the macro menu, click Gfx: per to shade the elements.

9. Take a moment to inspect the mesh pattern that the volume tetra mesher created.









Tetra mesh from the volume tetra subpanel and equilateral triangles (2D: trias)

10. Go to the disp panel on the permanent menu to see the two new components that arecreated by the volume tetra mesher.

11. Return to the volume tetra subpanel.

12. Reject the mesh.

The mesh, along with the components that were created, is deleted.

Step 3: Use the volume tetra mesher and right triangles to create a tetra meshfor the cover.

You should still be in the tetramesh / volume tetra subpanel.

1. Select one of the surfaces in the model.

2. Select 2D: R-trias.


4. Click Gfx: per on the macro menu to shade the elements.

5. Inspect the mesh pattern that the volume tetra mesher created.

6. Compare it to the first mesh you created and note the differences. The 2D: R-triassetting tends to create tetra elements with triangular faces that are right triangles (90-45-45 angles) while the 2D: trias setting tends to create equilateral triangles (60-60-60angles).









Tetra mesh from the volume tetra subpanel and right triangles (2D: R-trias)

7. Reject the mesh.

Step 4: Use the volume tetra mesher to create a tetra mesh with moreelements along curved surfaces.

You should still be in the volume tetra subpanel.


2. Activate the option use curvature.

Additional parameters appear.

3. The option use curvature causes more elements to be created along areas of highsurface curvature. Thus, curved areas such as fillets will have more and smallerelements, which capture those features with higher resolution.

4. Verify that elem feature ang = is set to 30.

5. For min element size = specify 1.0.


7. Shade the tetrahedral elements.

8. Inspect the mesh pattern that the volume tetra mesher created.









9. Compare it to the previous meshes you created and note the differences. More elementsare created around the fillets.

Tetra mesh from the volume tetra subpanel and the option use curvature active

10. Reject the mesh.

Step 5: Use the volume tetra mesher to create a tetra mesh with moreelements around small features.

You should still be in the volume tetra subpanel.


2. Activate the use proximity option.

The use proximity option causes the mesh to be refined in areas where surfaces aresmaller. The result is a nice transition from small elements on small surfaces to largerelements on larger, adjacent surfaces.


4. Shade the elements.

5. Inspect the mesh pattern that the volume tetra mesher created. Compare it to theprevious meshes you created and note the differences. More elements were createdaround surfaces with small angles as indicated in the image below.









Tetra mesh from the volume tetra subpanel with options use curvature and use proximity


Step 6: Prepare the display to tetra mesh the hub component using thestandard tetra mesher.

1. On the permanent menu enter the disp panel.

2. Toggle from elems to geom for components.

3. Click none to turn off the display for all components for geometry.

4. Turn on the display of the geometry for the hub component.

5. Toggle from geom to elems for components.

6. Click none to turn off the display for all components for elements.

7. Turn on the display for the hub and tetras components for elements.

There are tria shell elements in the hub component. Currently, there are no elements in

the tetras component.


Step 7 (Optional): Review the connectivi ty and quality of the tria mesh tovalidate its integrity for the standard tetra mesher.









1. Use the edges and check elems panels to make sure that there are no free edges orvery small angles in the tria shell mesh.

2. On the Tool page enter the edges panel.

3. With the comps selector active, pick any tria element on the hub component.

4. Click find edges.

A message in the header bar should state “No edges found. Selected elements mayenclose a volume.”

This is desired as the tetra mesher requires a closed volume of shell elements.


6. Enter the check elements panel.

7. Verify that you are in the 2-d subpanel.

8. Identify elements having an aspect ratio greater than 5. Aspect ratio is the ratio of the longest edge of an element to its shortest edge. Thischeck helps you to identify sliver elements.

All of the hub’s shell elements pass the check; all of the elements have an aspect ratioless than 5.

9. Identify tria elements having an angle less than 20. (trias: min angle)

This check also helps you to identify sliver elements.

All the hub’s shell elements pass the check; all the elements have angles greater than

20.

The surface mesh is suitable for creating a tetra mesh.


Step 8: Create a tetra mesh for the hub using the standard tetra mesher.

1. Use the tetramesh / tetra mesh subpanel to generate the mesh.

2. Set the current component to tetras.

3. On the 3D page enter the tetramesh panel.

4. Go to the tetra mesh subpanel.

5. With the comps selector active under floatable trias/quads:, select one of the hub shellelements from the graphics area.

The floatable trias / quads option gives the tetra mesher the option to swap thediagonal for any pair of surface trias, if that will result in a better tetra mesh quality. If you









do not want to change the element connectivity on the surface mesh, use the fixed trias/ quads selector.

6. Click tetmesh to generate the tetrahedral elements.

Cut-away view of tetrahedral elements

Step 9: Check the quality of the hub’s tetra elements.

1. From the disp panel, display only the elements in the tetras component.

2. On to the Tool page, enter the check elems panel.

3. Go to the 3-d subpanel.

4. Identify the smallest element length among the displayed elements.

If the minimum length is acceptable for a target element size of 5.0, then no furtheraction is necessary.

5. Identify the smallest angle (tria faces: min angle) among the displayed elements.

If the minimum tria face angle is no less than 10°, then the mesh quality should beacceptable.

6. Identify elements having a tet collapse smaller than 0.3.









The tet collapse criteria is a normalized volume check for tetrahedral elements. A valueof one indicates a perfectly formed element with maximum possible volume. A value ofzero indicates a completely collapsed element with no volume.

The header message bar indicates that one element has a tetra collapse smaller than0.3.

Step 10: Isolate the element with the tetra collapse smaller than 0.3 and findthe elements surrounding it.

You should still be in the check elements panel.

1. With 0.3 still specified for tet collapse, click tet col lapse again.

2. Click save failed.

The element that failed the tetra collapse check is saved in the user mark, and can beretrieved in any panel using the extended selection menu.


4. On the Tool page, enter the mask panel (F5 key).

5. Set the entity selector to elems and select elems >> retrieve.

The element that was saved in the check elems panel is retrieved.

6. Select elems >> reverse.

7. Mask the elements.

Only the one tetra element that failed the tetra collapse check should be displayed.


9. On the QA page of the macro menu, click find attached.

The layer of elements that is attached to the one displayed element is identified anddisplayed.

10. Click find attached again.

The layer of elements that is attached to the displayed elements is identified anddisplayed.

The find attached macro’s functionality can be duplicated using the findpanel, find attached(fe) subpanel on the Tool page.









Step 11: Re-mesh the hub’s displayed tetra elements to improve their tetracollapse.

1. Use the tetramesh / tetra remesh subpanel to re-mesh only the displayed tetraelements.

2. On the 3D page enter the tetramesh panel.3. Go to the tetra remesh subpanel.

4. Select elems >> displayed.

5. Remesh the elements to regenerate this area of the mesh.

Note that the re-meshing operation works on only one group of elements (one volume)at a time.


7. On the Tool page enter the check elems panel.

8. Click tet collapse to find out if the tetra collapse has improved for the displayedelements.

The message in the header bar should indicate that the minimum tetra collapse is largerthan the value reported before the tetra elements were re-meshed.



Now that the tetra mesh complete, it is a good time to save the model.

Summary

A tetra mesh has been created for both parts in the file. Different procedures for tetrameshing were used. Either method can be used to mesh parts, depending on the needs ofthe analysis. Also, the tetra re-mesh function was used to show how to quickly fix thequality of tetra elements.









Chapter 4:

Creating Hexa-PentaMesh



Chapter 4: Creating Hexa-Penta Mesh







Hex Meshing: Introduction

] Hex mesh is created by “extruding” hexa / penta elements from a

shell mesh

• Allows lots of control over the hexa-penta mesh pattern

• Requires good shell meshing skills

• 2D element quality is key!

− Good quality shells lead to good quality solids

• 3D panels in HyperMesh are easy to use

− Difficulty lies in planning the strategy to mesh complex parts

Hex Meshing: Tools for Creating Hex Mesh

] Drag – Extrudes solids from

shells along a linear path

] Spin – Extrudes solids from

shells along a circular path










] Drag – Extrudes solids from shells along a linear path

] Spin – Extrudes solids from shells along a circular path


] Line Drag – Extrudes solids from shells along the path of a

selected line

] Element Offset – Extrudes solids from shells along the normals of

the selected shells










] Linear Solid – Creates layers of solids between two sets of shell

elements with the same mesh pattern and connectivity

] Solid Mesh – Creates solids in a pentagonal or hexagonal volume

defined by lines


] Solid Map

• Extrudes solids from a set of shell elements

• Define a volume using nodes, lines, surfaces, elements, or any

combination

• Automatically maps the solids into the volume

• Can automatically match connectivity with adjacent areas that have

also been mesh with solid map









Hex Meshing: Tools for Checking / Editing Hex Mesh

] Faces

• Preview and equivalence nodes within a specified tolerance

− Edges panel can also be used

• Create shell elements on every free face in the model

− Places shells in a new component called f̂ aces

− Shells can help identify connectivity problems

− Use shells in other 3D panels to create solid mesh

] Hidden Line

• Use hidden line to view a

cross section of the^faces component to

check model connectivity

Hex Meshing: Tools for Checking / Editing Hex Mesh

] Check Elems

• Use the 3-d sub-panel to check element quality of solid elements

] Split

• Hexa elements sub-panel – Splits hexa elements into 1 of 6 patterns:

• Patterns are consistent with respect to selected N1 and N2 nodes:

2 Pentas Penta-Hexa Hexa-Penta 2 Hexas 3 Hexas 4 Hexas

+0º +90º +270º+180º









Hex Meshing: Process and Strategy

] Study the geometry

• Understand the shape of the part

− Focus on small portions of complex parts

— Mask or split surfaces if necessary

− Create a cross-section of lines to help

— Use intersect panel or trim the surfaces

• Determine features that will be ignored

− Ex.: small fillets & holes

• Determine features that are critical to the analysis

− These features will probably need goodelement quality

Study the

Geometry

Determine

Methods

Form a Plan

Prepare the

Geometry

Create the

Shell Mesh

Create the

Solid Mesh


] Determine what 3D functions to use

• Mentally divide the part into different regions

• Each region must be able to be mesh using one

of the solids functions

• Look for:

− Areas where large numbers of elements can

be created with 1 function

− Identical / symmetric areas

− Intersecting features

− Areas with lots of curvature

• Look at how the region will connect with adjacent

regions

• Note what 2D mesh pattern is required

Study the

Geometry

Determine

Methods

Form a Plan

Prepare theGeometry

Create the

Shell Mesh

Create the

Solid Mesh










] Example: half sphere Study the

Geometry

Determine

Methods

Form a Plan

Prepare the

Geometry

Create the

Shell Mesh

Create the

Solid Mesh

Spin

Linear

Solid


] Form a plan

• Determine the order in which to mesh the regions

of the part

− Mesh from the inside to the outside

− Mesh smaller features before large

— Larger adjacent areas can then be

meshed to match connectivity

− Try to mesh the hardest areas first

− Sometimes start with areas where elements

can be “mass produced”

− Sometimes just mesh ANYTHING to get

started

— Starts the thinking process

Study the

Geometry

Determine

Methods

Form a Plan

Prepare theGeometry

Create the

Shell Mesh

Create the

Solid Mesh









Hex Meshing: Process and Strategy] Prepare the geometry

• Focus on a region where 1 3D meshing functionwill be used

• Determine what geometry is need to create the2D mesh needed for the 3D mesh

− DO NOT assume that geometry cleanup isnecessary!!!

− ONLY create / edit what is necessary to usethe 2D mesh function

• Perform the preparation

− Geometry cleanup

− Defeature blends & small holes if needed

− Trim surfaces− Project or create lines

− Etc.

Study the

Geometry

Determine

Methods

Form a Plan

Prepare the

Geometry

Create the

Shell Mesh

Create the

Solid Mesh


] Create the shell mesh needed for the solid mesh

• Think of what mesh pattern is required

− Don’t simply accept what the algorithms give

you

− Perform as much editing as needed

• Can use faces panel to create shells on existing

solid mesh

• Evaluate the shell mesh when created

− Make sure the pattern will lead to solid mesh

that can be connected to adjacent regions

− Good quality 2D mesh will lead to good

quality 3D mesh

Study the

Geometry

Determine

Methods

Form a Plan

Prepare theGeometry

Create the

Shell Mesh

Create the

Solid Mesh

] Element quality is key!!!










] Create the solid mesh

• This is the easy part!

• Evaluate when the mesh is created

− Make sure the mesh will be able to connect

with adjacent regions

− Check if any editing is required

— Modifications to match the geometry

– Morphing can be used here

— Splitting of elements to obtain necessary

edges

— Improvement of element quality

• Move to the next region and repeat

− Always be aware if your strategy needs to

change

Study the

Geometry

Determine

Methods

Form a Plan

Prepare the

Geometry

Create the

Shell Mesh

Create the

Solid Mesh


















Creating Hexa-Penta Mesh

PurposeFor some analyses, it is desirable to use a mesh of hexahedral and pentahedral elements.This is used for parts, which have a large thickness compared to the element size beingused, or for parts that have many features and/or changes in thickness. Castings orforgings are good examples.

This exercise introduces you to a number of HyperMesh functions that are used to createhexa-penta meshes.

Problem Statement

This exercise uses the model file, ar m_ br acket . hm.

The model geometry is a bracket designed to support a rail. It is organized into four sectionsand surfaces are split for ease of creating the shell mesh, which will be used to build thehexa-penta mesh. Mesh each of these using a different hexa meshing function. Thefollowing functions will be used:

1. Elem offset

2. Spin

3. Linear solid

4. Solid map

Once the part is meshed, it also needs to be checked for connectivity errors. The faces andhidden lines panels are used to perform this task.

Note This exercise assumes that you are familiar with using the automesh panel and itsmeshing module.









Exercise

Step 1: Retrieve and view the model file

The model for this exercise is ar m_ br acket . hm. Take a few moments to observe themodel using the different visual options available in HyperMesh (rotation, zooming, etc.).

The model is organized into four IGES layers, consisting of 1) the base, 2) the first section ofthe arm, with a constant cross section and curvature, 3) the second section of the arm, witha tapered cross section, and 4) the boss.

Step 2: Mesh the top surface of the base, including the L-shaped surface.

1. On the status bar, verify that the active component collector is base.


3. Select the surfaces on the top side of the base, including the L-shaped surface at theintersection of the base and the arm.

4. For elem size = specify 10.

5. For element type specify quads.

6. Set the meshing mode to automatic. (The current setting might be interactive or QIoptimized.)










Resulting quad mesh on base surfaces


Step 3: Create layers of hex elements for the base.

1. On the 2D page enter the elem offset panel.

2. Select the solid layers subpanel.

3. With the elems selector active, select the elements on the base.

4. For number of layers = specify 5.

5. For total thickness = specify 25.

6. Click on offset +.

The hexa mesh is created.









Hex mesh on base

Step 4: Prepare the display for meshing the arm’s curved segment.

1. Turn on the display of the geometry in the arm_curve component.

2. Press the F5 key to go to the mask panel.

3. Select elements >> by conf ig, and select the hex8 configuration.

4. Click select entities.

All of the elements with a configuration of hex8 in the model are selected.

5. Select elements >> by conf ig, and select the penta6 configuration.

6. Click select entities.

All of the elements with a configuration of penta6 in the model are selected.

7. Mask the elements.


Step 5: Create a node at the center of the arm radius.

1. Use the distance / 3 nodes subpanel in order to create the node.









The first segment of the arm can be meshed using the spin panel. This requires a nodeto be selected as the center point of rotation. The node you create here will be used asthat center point.

Temporary nodes can be created on geometry by clicking and holding the left mousebutton until the cursor changes to a box, then highlighting the geometry. Clicking on the

highlighted line creates nodes.

2. Enter the distance panel.

3. Go to the three nodes subpanel.

4. With the N1 selector active, select three nodes on one of the curved lines of the arm.

5. Press and hold the left mouse button. Move the cursor over a curved line. The cursorchanges to a square, and the line is highlighted. Release the mouse button, and the lineis selected.

6. Click at any three locations along the selected line. The active selector advances from

N1 to N2 to N3, and the locations will be selected as though there was a node there.7. Click circle center to create the node at the center.

Three nodes to create a center node


Step 6: Create hexa elements in the curved portion of the arm.

1. Use the spin panel to create the mesh. Use a density of 24 elements along the curve.

2. Use the global panel to set arm_curve as the current component.

3. Enter the spin panel on the 3D page.









4. Select the spin elems subpanel.

5. Using elems >> by window , select the plate elements within the L-shaped cross sectionof the arm.

Elements to select for spin function

6. For angle = specify 90 degrees.

7. For the direction, select the x-axis (Y-Z plane).8. For the base node, select the center node created above.

9. For on spin = specify 24.

24 layers of hex elements will be created when the plate elements are spun.

10. Click spin -.










Spin panel results

Step 7: Create faces on the hex elements.

1. On the Tool page enter the faces panel.

2. With the entity selector set to comps, select the arm_curve component.

3. Click find faces.

2-D shell elements are created on the free faces of every 3-D solid element in the

component. They are placed in a new component named ^faces.

The ^faces component is created with its visualization set to wireframe, so you will not

be able to see the new elements right away if the arm_curve component is displayed

and in shaded mode.

4. On the Disp page of the macro menu click Gfx: per to shade the elements.

You will now see the elements in the ^faces component.

Step 8: Prepare the display for meshing the second arm segment.



3. Click none to turn off the display for all geometry.

4. Display the geometry in the arm_straight component.










6. Click none to turn off the display for all elements.

7. Display the elements in the arm_straight and ^faces components.


Step 9: Mesh the L-shaped set of surfaces between the arm_straight and

boss components.

1. Enter the global panel and set the current component collector to arm_straight.

2. On the 2D page, enter the automesh panel.

3. Select the three surfaces lying on the intersection between the arm_straight and

boss components.

Note These surfaces are in the arm_straight component.

4. Set the meshing mode to interactive.

5. Click mesh to go to the meshing module.

6. From the density subpanel, adjust the densities to obtain a mesh that matches theimage below.

This mesh pattern matches the mesh pattern at the intersection of the two armsegments. This is necessary for the next step.

Densities to correspond to the mesh on the end face

7. Click return to create the elements and go back to the automesh panel.










Step 10: Use linear solid to build the mesh between the two sets of shellelements.

1. On the 3D page enter the linear sol id panel.

2. With the from: elems selector active, select the ^faces elements lying on the

intersection between the first and second arm segments.

You can select one of the elements and then select elems >> by face to select the restof the necessary elements.

3. Click the to: elems selector to make it active. Then select the shell elements betweenthe arm and boss, which you created using the automesh panel in the last step.

4. Click the from: alignment: N1 selector to make it active. Then select three nodes onone of the “from elements” you selected in #2 above.

5. Click the to: alignment: N1 selector to make it active. Then select three nodes on the“to element” corresponding to the “from element” with the three “from nodes” youselected in #4 above. Refer to the image below.

Example selection for alignment nodes

6. For density = specify 12.

7. Click solids to create the mesh.









Linear solid mesh


Step 11: Prepare the display for meshing the boss.



3. Click none to turn off the display for all geometry.

4. Display the geometry in the boss component.


6. Click none to turn off the display for all elements.

7. Display elements for the boss component.


Step 12: Create a shell mesh on the bottom of the boss.

1. Set the current component collector to boss.


3. Select the five surfaces on the bottom face of the boss.

4. Click mesh to enter the meshing module.









5. Adjust the densities to match the image below:

Mesh densities on the bottom of the boss

6. Return to the automesh panel.

The elements are created.


Step 13: Project a node to the bottom face of the boss.

1. On to the Tool page enter the project panel.

2. Go to the to line subpanel.

3. Select the node on the rightmost top vertex, as per the image below.

4. Click nodes >> duplicate.

5. For the to line select the line on the boss’ top face. Refer to the image below.









Projecting a node to a line

6. Select along vector: x- axis.

7. Project the node to the line.


Step 14: Generate hexas for the boss using the solid map panel.

1. On the 3D page enter the solid map panel.

2. Select source geom: (none).

3. Select destination geom: surf and select the top surface of the boss.

4. Select along geom: mixed.

5. Under along geom: mixed, click lines to make it the active selector.

6. Select the line indicated in the image below.

7. Click node path to make it the active selector.

8. Select nodes to define the exact location of the solid element layers, as indicated in theimage below.

A total of 13 nodes should be selected, starting at the boss mesh, and then using all ofthe nodes along the edge of the arm_straight component, ending with the node

projected to the top of the boss.









Along nodes for solid map

9. For elems to drag:, select elems >> by collector and select the boss component.

10. Click mesh.

The elements are created and the mesh on this part is completed.









Completed mesh of the arm bracket


Step 15 (optional): Check the connectivity of the model.1. On the Tool page, enter the faces panel.

2. Click comps to go to a list of components.

3. Select every component from the list, or select comps >> all.

4. Select the components to complete the selection and go back to the faces panel.

5. Click find faces.

6. Go to the disp panel on the permanent menu.

7. Turn off the display of the geometry of all components.

8. Turn off the display of the elements of all components except ^faces.

9. On the Post page enter the hidden line panel. (You can press F1 on the keyboard to goto this panel.)

10. Go to the cutting subpanel.

11. Activate the xz plane and trim plane options.

12. Click fill plot.

The faces are now displayed with a plane cutting the model in half. This is so the interiorof the model can be viewed.

13. Click near the cutting plane. Holding the left mouse button down, move the mouse backand forth.

The cutting plane moves through the model, allowing you to see if any face elementsexist on the interior of the model.

You should see that there are face elements interior to the model, between the boss andarm. You need to perform some corrections on the connectivity.









Hidden line view of faces


Step 16 (Optional): Correct the connectiv ity of the model.

1. From the disp panel, display elements for all components except for the ^faces

component.

2. On the permanent menu enter the vis panel.

3. Display the elements of the solidmap component as transparent.

4. On the Tool page enter the faces panel.

5. Select elems >> displayed.

6. Click preview equiv.

Coincident nodes on the intersection between the arm and the boss are highlighted.

7. Specify a slightly larger value for tolerance =, and click preview equiv to identify morecoincident nodes on the intersection.

8. Repeat #7 until all 60 coincident nodes have been found.

9. Click equivalence.

The nodes are replaced to the location of the lowest node ID.









10. On the Disp page of the macro menu, click Gfx: per .

All the components are reset to use the shaded visual mode.

Step 17 (Optional): Recheck the connectivi ty of the model.

Repeat Step 16 above to make sure the model is now equivalenced. If you find errors,repeat either Step 17 or Step 18 to correct the errors.


The 3-D solid mesh has now been completed. Save the model if desired.









Chapter 5:

Morphing



Chapter 5: Morphing







] What is morphing?

] How morphing works

] Typical morphing operations

] Morphing tools

Morphing – “ Agenda”

] Morphing allows you to:

• Change the shape of an existing mesh

• Modify the spatial relationship of elements in a mesh pattern

− Doesn’t change the mesh pattern at all

• Map an existing mesh to new geometry

• Save shape changes as “shape variables” for use in optimization

analysis

Morphing – What is it?

Width

IncreasedRadius

Increased



Chapter 5: Morphing






] HyperMorph uses “domains” and “handles” to control morphing

operations

Morphing – How does it work?

Local

HandlesGlobal

Handles

2D Domain

Handle

3D Domain

Handle

2D Domain on free faces

of solid elements

Global Domain

Handle

Edge Domain (red line)

] Domains divide the model into groups of elements

• A morphing operation that affects a domain will affect all elements in

that domain

• Elements can only belong to 1 domain

• Types of domains:

− Global: influences all elements in the model

− 1D: contains groupings of 1D elements

−

2D: contains groupings of 2D elements — Also found on the free faces of 3D elements

— Separates 3D domains

− 3D: contains groupings of 3D elements

− Edge: domains at the perimeter of 2D domains

— Separates 2D domains




Chapter 5: Morphing






] Handles provide the mechanism for modifying the shape of a mesh

• Handles are attached to 1 or more domains

• When a handle is moved, all elements & nodes in the attached

domains are moved proportionately

− The amount of movement depends on the location within the

domain relative to the handle

• Types of handles

− Global: handles attached to the global domain

— Used to perform large scale shape changes

− Local: handles attached to local domains (1D, 2D, 3D, & edge)

— Used to perform detailed shape changes


] Example:


Move a local

handle to the left

Handle modifies the

2D and edge

domains it is

attached to

Handle does not

modify 2D or edge

domains to which it

is not attached



Chapter 5: Morphing






] Move handles to new

locations

] Change the distance

between handles

Morphing – Typical Morphing Actions

10 15

] Change the radius of

fillets or holes

] Map to new surface

or line geometry

Morphing – Typical Morphing Actions



Chapter 5: Morphing






] The HyperMorph Module

Morphing – Tools

] HyperMorph Panels

• Domains – Create and edit the organization of elements into domains

• Handles – Create new handles

• Systems – Create and edit local coordinate axes

• Symmetry – Creates “symmetries” which allow similar domains to be

morphed in a symmetric fashion

• Morph

−

Perform modifications to model shape− Save changes as “shape variables for optimization analyses

− Define parameters for morphing operations

• Map to Geom – Projects domains / mesh to selected geometry

• Autoshape – Define multiple shape variables for optimization at the

same time

Morphing – Tools



Chapter 5: Morphing






] Disp panel (permanent menu)

• Set the collector type selector to handles, domains or symmetrys

] Mask panel

• Set the entity selector to handles, domains, or symmetrys

Morphing – Display Controls

] Domains : parameters sub-panel

• Control size of:

− handle icons

− symmetry icons

• Select colors for:

− Domains

− Symmetrys

− Faces

Morphing – Display Controls



Chapter 5: Morphing








Chapter 5: Morphing







Introduction to MorphingDomains and Handles

Purpose

To work with the morphing functionality in HyperMesh, you should first be proficientwith creating domains and handles and understand how they interact with eachother and the mesh by performing basic morphing (moving handles interactively).Once you understand these basic principles, you can then learn about othermorphing functions.

This exercise uses a very simply model of shell elements, which allows you to see

and understand the relationship of domains and handles.

Problem Statement

This exercise uses the model file, mor phi ng_ 1. hm.

As stated above, the model is a very simple model of shell elements. The exercisediscusses the following topics:

1. Automatic generation of domains and handles

2. Manually organizing elements into domains

3. Manual creation of new handles

4. Editing edge domains

5. Interactive morphing to understand domain and handle interaction



Chapter 5: Morphing






Exercise


The model for this exercise is mor phi ng_ 1. hm. Take a few moments to observethe model using the different visual options available in HyperMesh (rotation,zooming, etc.).

Step 2: Auto generate 2-D domains and handles.

6. On the Tool page enter the HyperMorph panel.

This takes you to the HyperMorph module. Within the HyperMorph module are thepanels used to define and edit morphing domains and handles, as well as the panelsthat carry out the morphing operations. The left column contains common HyperMesh

panels that are useful when morphing.

The HyperMorph module

7. In the HyperMorph module, enter the domains panel.

8. Select the create subpanel.

9. Click autogenerate.

Based on the model’s geometric features, the autogenerate function automaticallyorganizes all of the model’s elements into various domains and creates local handlesassociated with the domains.

autogenerate also creates eight global handles, one at each of the eight corners of abox enclosing the model. (The box’s sides align to the global axes.) One more globalhandle is created within the box, at the model's peak nodal density.

The screen display should look like the image below.



Chapter 5: Morphing






Model with auto generated domains and handles for morphing

Step 3: Move elements into a new 2-D domain.

You should still be in the domains / create subpanel.

1. Switch the domain type from global domain to 2D domains.

2. Use elems >> by window to select the elements indicated in the image below.

Elements to select to move into a new domain

3. Verify that the partition domains option is active.

Local handles are created for the new domain.



Chapter 5: Morphing






4. Create the domain.

You should now have two local domains.

Elements can only belong to one domain at a time. Thus, the elements you selectedwere moved into the new domain. This functionality makes it very easy to group

elements into different domains.

Step 4: Split the edge domain of the radius to have more control whenmorphing.

You should still be in the domains panel.

1. Go to the edit edges subpanel.

2. Verify that the split edge option is active.

3. With the split edge: domain selector active, select the edge domain of the part’s radius

as indicated in the image below.

The node selector automatically becomes active once the edge domain is selected.Click the domain selector to make it active and see that you selected the desired edgedomain.

Edge domain to select

4. Click the node selector to make it active.

5. Select the node on the positive Y-axis end of the radius, as indicated in the image below.



Chapter 5: Morphing






Node selection to split the edge domain of the radius

6. Split the edge domain at the node.

7. Repeat the above process to further split the edge domain of the radius, this time at thenode indicated in the image below.

Node selection to further split the edge domain of the radius

8. Return to the menu of the HyperMorph module.

Step 5: Add local handles to the 2-D domain on the part’s left side.

1. Enter the handles panel.




Chapter 5: Morphing






3. For name= enter local.

4. Click the attached to: domain selector to make it active.

5. Select the 2-D domain on the part’s left side by selecting its red icon, as indicated in theimage below.

Adding handles to a 2-D domain

6. Click the by nodes: nodes selector to make it active.

7. Select the two nodes as indicated in the above image.

8. Create the handles to add them to the 2-D domain.

9. Return to the menu of the HyperMorph module.

Step 5: Perform basic morphing to understand how domains and handlesinteract with each other and the mesh.

1. Enter the morph panel.

2. Verify that you are in the move handles subpanel.

3. (Optional) With the handles selector active, select the two handles that are on the mostpositive X-axis end of the part, as indicated in the image below.

If you select one or more handle, those handles follow the handle you drag (in step #8below).

4. Switch on domains to on plane.

5. Click the N1 selector to make it active.

6. For N1, N2, and N3, select any three nodes on the model to define a plane.

7. Click morph.



Chapter 5: Morphing






“pick handles and move to new location” appears in the message bar.

8. Click on and drag one of the selected handles to morph the part.

As you drag the handle, the mesh’s size and shape is adjusted.

9. Notice that the following occurs as the selected local handle is moved:

• The handles selected in step #3 above follow the handle you are dragging.

• All of the elements belonging to the selected local handle’s 2-D domain are affectedby moving that local handle.

• The 2-D domain’s non-selected local handles act like anchors (they do not move).

• The nodes on the edge domains and between any two non-selected local domainsdo not move.

• None of the elements in the other 2-D domain are affected.

• Once you release the mouse button, the morphing operation is complete.

Example result of morphing the model

10. Click undo.

The HyperMorph module allows for multiple levels of undo and redo for all morphingoperations. This functionality is available for any particular HyperMesh session and its

current model as long as the session and its model remain open.

11. Click reset under the handles selector.

12. (Optional) With the handles selector active, select one or more global handles.

13. Click morph.

14. Click on and drag any global handle to morph the part.



Chapter 5: Morphing






15. Notice that the following occurs as the selected global handle is moved:

• The handles selected in step #11 above follow the handle you are dragging.

• The non-selected global handles act like anchors (they do not move).

• All of the elements, local handles and edge domains are affected.



Chapter 5: Morphing








Exercise


The model for this exercise is mor phi ng_ 2. hm. Take a few moments to observethe model using the different visual options available in HyperMesh (rotation,zooming, etc.).

Step 2: Auto generate morphing domains and handles for the model.

1. Enter the HyperMorph module.

2. Enter the domains panel.


4. Click autogenerate.

Step 3: Create a new 2-D domain to set-up for morphing (altering) a fillet’sradius.

1. Verify that the domain type is still set to 2D domains.

2. With the elems selector active, select the elements shown in the image below.

Elements to select and move to a new 2-D domain.

3. Verify that the partition domains option is active.

4. Create the new 2-D domain.

5. Return to the HyperMorph module menu.



Chapter 5: Morphing






Step 4: Morph (alter) the fillet’s radius.

1. Enter the morph panel.

2. Go to the alter d imensions subpanel.

3. Switch distance= to radius=.

4. Switch the radius mode from hold center to fillet.

5. Select the two edge domains, which define the radius of the 2-D domain you justcreated, as indicated in the image below.

Once you select both domains, the value for radius= should read 133.483.

The two edge domains to select to alter the fillet’s radius

6. For radius=, specify 150.

7. Click morph to morph the part.

The radius of the arcs are changed to 150. The radius of the arcs stay blended with theadjacent edges.

Step 5: Morph (alter) a hole’s radius.

You should still be in the morph / alter dimensions subpanel.

1. Fit the model to the screen, and then zoom in on the positive X-axis end of the part.

2. Switch fillet to hold center .

3. Select the edge domain of the hole as indicated in the image below.

Once the domain is selected, the value for radius= should be 17.657.

4. For radius=, specify 20.



Chapter 5: Morphing






Edge domain to select to alter the size of the hole

5. Click morph to change the hole’s radius.

6. Return to the HyperMorph module menu.

Step 6: Morph (map) existing mesh to new geometry.

1. On the permanent menu, enter the disp panel.

2. Display on the geometry for the new_geometry component.


4. Enter the HyperMorph module.

5. Go to the map to geom panel.

6. With the map domains: domains selector active, select the edge domain indicated inthe image below.

7. Click on the map to line: line selector to make it active.

8. Select the line indicated in the image below.



Chapter 5: Morphing






The domain and line to select to map existing mesh to new geometry

9. Click auto mapping to morph (map) the mesh to the new line.

10. Repeat the above process to complete the mapping of the mesh to the new geometry.Select the two edge domains and the line as indicated in the image below.

Chapter 5: Morphing

Documents

Altair Hyperworks Hypermesh 7 Basic Training Tutorial Day2