Engineering Solutions is a modeling and visualization environment for NVH, Squeak and Rattle Director, Crash, CFD, and Aerospace using
best-in-class solver technology.
The Crash application offers a tailored environment in HyperWorks that efficiently steers the Crash CAE specialist in CAE model building, starting from CAD geometry and finishing with
a runnable solver deck in Radioss, LS-DYNA and PAM-CRASH 2G.
HyperWorks offers high quality tools for CFD applications enabling the engineer to perform modeling, optimization and post-processing
tasks efficiently.
Browsers supply a great deal of view-related functionality in Engineering Solutions by listing the parts of a model in a tabular and/or tree-based format, and providing controls inside the table
that allow you to alter the display of model parts.
Perform automatic checks on CAD models, and identify potential issues with geometry that may slow down the meshing
process using the Verification and Comparison tools.
Change the dimensions of existing geometry, thus changing the basic shape of solids
and other enclosed volumes.
Figure 1. Initial Dimensions
Figure 2. Modified Dimensions
Edit Dimensions
Dimensioning is accomplished with features, parameters and dimension
manipulators.
When changing several dimensions, each dimension change is performed separately using
the respective manipulator. However, if multiple dimensions are linked to the same
parameter or parameter expression, they will be updated simultaneously.
When dimensions are modified, a very limited check for mutual penetrations of the
repositioned surfaces is performed. It is the your responsibility to ensure that the
new dimensions are appropriate.
The locked end of the dimension manipulator defines the direction in which the
affected surfaces move when the dimension is modified. For a dimension to be
modified, one or both ends of the dimension manipulator must unlocked.
When dimensions cannot be modified, the locked side is set to Both and you may use
the Sides Selection advanced option to specify how the dimension should be changed,
when possible.
Create dimension feature.
In the Model Browser, right-click and select Create > Feature from the context menu.
The Feature dialog opens.
In the Name field, enter a name for the dimension.
In the Point1 and Point2 fields, use the entity selector to select two
fixed points (vertices) between the opposite surfaces where the
dimension is defined.
A dimension manipulator is then created between these fixed points
(Point1 and Point2).
In the Parameterization field, select a parameterization method.
Note:
A new parameter can be created and assigned to an existing
dimension feature at any time via the Create Parameter option in
the Entity Editorcontext menu.
Choose Create Parameter to create and
assign a new parameter to a dimension feature.
Choose Select Parameter to assign an
existing parameter to the new dimension feature.
Choose No Parameter if you do not want to
assign a parameter to the dimension feature.
Click Create.
In the Entity Editor, define dimension feature
attributes.
Edit dimension.
In the modeling window, click the dimension's
corresponding label and enter a new value.
In the Model Browser, select the parameter assigned
to the dimension feature. In the Entity Editor,
enter a new value.
Figure 3.
Dimensioning Concepts
Learn about basic dimensioning concepts, such as continuous surface offset
functionality and tolerance and accuracy.
Continuous Surface Offset Functionality
Dimensioning is based on a continuous surface offset functionality. It provides
assistance in the selection of the surfaces to offset so that a change to the
selected dimension can occur, and calculates the offset values required for each
surface to achieve the specified dimension.
The continuous offset modifies both the surfaces you selected for the offset and the
adjacent involved surfaces that must also be modified so that the result will remain
as continuous as the initial input.
These "selected" and "involved" surfaces are modified with different rules.
Selected surfaces
Offset by a constant value that is normal, or in some cases almost
normal, to the surface at each point. For example, a standalone
surface is offset by the given constant distance exactly normal to
itself. Figure 4. Normal Offset of a Standalone Surface
When the adjacent surfaces form a corner between them, the exact
normal offset will result in either disconnected surfaces or in
intersecting surfaces, for example if the offset was performed in
the opposite direction. Figure 5. Exact Normal Offset of the Adjacent Surfaces Creates a
Rupture
A continuous result that is consistent with the given offset
distance is obtained by reconciling the offset vectors of the
vertices shared by the surfaces being offset. Figure 6. Reconciled Offset Vectors at Shared Edge
Involved Surfaces
The edges of the involved surfaces that are shared with the selected
surfaces move with the selected surfaces.
The edges of the involved surfaces that do not have a common point
with the selected surfaces do not move, for example they are
locked.
The offset of the edges that connect both the moving and the locked
involved surface edges is defined by interpolation. Different
interpolation methods are available.
In general cases, the target dimension between the selected vertices is achieved by
offsetting the surfaces that contain the vertices in an infinite number of ways. To
avoid this, the following rules are implemented.
If both dimension ends (both vertices) are allowed to move, an attempt is
made to move them by the same distance whenever possible.
If possible, the dimension ends are moved in such a way that the direction
of the dimension will not change.
In the following example, the initial positions of the vertices are marked with temp
nodes to enable the changes can be easily seen. The locked state of the dimension
manipulators is indicated by the lock icons.
Note: These examples are not cumulative,
so no two images are directly related. The first image, showing the dimensions
of 3, 4, and 5, is the starting point from which all of the other examples
derive.
Figure 7. Original Model. 3 dimensions selected.
Figure 8. Dim 4 Changed to 5. Top and bottom move.
Figure 9. Dim 4 Changed to 5. Top moves, bottom is locked.
Figure 10. Dim 5 (Diagonal) Changed to 6. All sides move.
Figure 11. Dim 5 Changed to 6. Only top moves.
Figure 12. Dim 5 Changed to 6. Only right side moves.
Tolerances and Accuracy
All geometry transformation tools are numerical tools that operate with some accuracy
defined by the tolerances, such as the geometry cleanup tolerance set in the Options
panel. Curved surfaces and lines have internal structures in 3D that are invisible
to you. Significantly reducing the size of such an entity so that these structures
fall below the tolerances may result in a structure simplification that you cannot
notice at first; the structural data will be lost. When this occurs any subsequent
increase in the size will not restore the initial structures. For example, reducing
a cylinder diameter 100 times and then increasing the diameter 100 times may not
lead to the same cylinder; in some cases, a complex internal representation of the
cylinder may lead to a corrupt surface. In general, transformation of a curved
entity may result in both the simplification or complication of its internal
structure. It is therefore not recommended to perform multiple transformations on
curved entities.
Dimension Manipulators
Dimension manipulators are used to alter selected dimensions of solid
entities.
A dimension manipulator consists of:
Dimension line
A segment parallel to the line that connects the selected points, but is
shifted off the selected points for visibility. The terms manipulator
direction and manipulator ends are also used, which are the same as the
dimension line direction and the dimension line ends.
Pullout lines
Two parallel segments that connect the ends of the dimension line with
the selected points.
Lock icons
Arrow (movable) and block (locked) icons indicate the lock state of a
manipulator end.
Lock controls
Sphere handles, located near the lock icons, enable the lock state of a
manipulator end to be modified.
Display/input field
Displays the current dimension value, which can be modified or deleted.
This value can be modified or deleted. Deleting the value deletes the
the manipulator. For dimensions that are parameterized, an "&"
symbol will appear before the dimension. Editing a parameterized
dimension directly edits the parameter, or parameter expression.
Figure 13. Dimension Manipulator
Dimension Feature Attributes
Attributes associated with dimension features can be modifed in the Entity Editor.
Attribute
Action
Lock Side
Select the locked end of the dimension manipulator, which defines
the direction in which the affected surfaces move when the dimension
is modified. For a dimension to be modified, one or both ends of the
dimension manipulator must unlocked.
When dimensions cannot be
modified, the locked side is set to Both and you may use the
Sides Selection advanced option to specify how the dimension
should be changed, when possible.
Surfaces Interpolation System
Automatic
A heuristic algorithm is used to try and decide which of
the two interpolation methods to apply for each
individual, applicable involved surface.
Local
A Local Coordinate System (LSC) 2D interpolation method
that "slides" along the surfaces to determine the offset
vectors, which are then interpolated and combined into
the interpolated offset at each point. Selected surfaces
are always interpolated using this method.
Global
A global coordinate system 1D linear interpolation
method that stretches/compresses a surface
proportionally in a global 1D. Only applicable when all
of the offset vectors at the surface's vertices are
collinear and proportional to the distance parameter
along their common direction.
Figure 14. Interpolation of the Same Offset Vectors at the Vertices
for both Methods
Minimum Slide Angle
When a selected surface is offset, the involved surfaces must be
modified to keep the continuity of the model.
Surfaces can be
modified by dragging the involved surface behind the selected
surface, or by defining it as a "slider" along which the
selected surface slides.
The Minimum Slide angle
determines which method is used. If the slide angle is more than
the specified value, then the involved surface will slide;
otherwise it will drag.
When the involved surface is a
slider, the orientation of the surface does not change for
planar surfaces. However, for curved involved surfaces, the
sliding directions are defined by the tangents to the surface
where it is adjacent to the selected surface. Sliding of the
selected and involved surfaces along these directions may also
result in some change to the shape of the involved surface. Figure 15. Original Model
Figure 16. Involved Surface Dragged. Dimension modified to D=0.4.
Figure 17. Involved Surface as Slider. Dimension modified to D=0.4.
Remove Collapsed Surfaces
Remove portions of the offset surfaces that fold into themselves
or adjacent surfaces (portions of surfaces that penetrate themselves
or adjacent surfaces along the edges they are adjacent over).
For
example, suppose that the slide angle is greater than the
Minimum Slide Angle and the value in the dimension manipulator
is set to 1. If this option is off, the involved surface will
slide and ignore the self-penetration, resulting in a corrupt
model. If this option is on, the involved surface will slide as
far as possible without causing self-penetration. This may not
allow the specified dimension to be reached, but will not result
in a corrupt model.
Figure 18. Remove Collapse Surfaces Off. Dimension modified to D=1.
Figure 19. Remove Collapse Surfaces On. Dimension modified to D=1.
Another useful application is for the removal of
holes. If the hole diameter is set to 0 and this option is on,
the hole will be removed. If the option is off, a small "straw
surface" will still remain.
In general, unless it is known
that collapsed surfaces will result, it is better to keep this
option off for performance reasons, as this option has no effect
on general cases that do not result in penetration.
Sides Selection
Auto
Automatically select the surfaces to offset using the
following rules:
Surfaces adjacent to the manipulator ends are
selected if the angle between the normal to the
surface at the dimension end and the dimension
direction is less than the Max Pick Tilt.
If
surfaces are selected at both ends for the
specified Max Pick Tilt value, then the lock
control handles will allow for the manual
manipulation of the offset scenario. Figure 20. Angle between the Normal to the Surface and
the Manipulator Direction
Surfaces adjacent to the selected surfaces are
appended, provided that they are planar and the
angle along the edge over which they are adjacent
to the already selected surface is less than the
Max Expand Angle.
The total area of the selected surfaces at each
end is calculated. If the area of the selected
surfaces at one end is more than the Side
Selection Area Ratio and larger than the area of
the selected surfaces at the other end, then the
surfaces on the larger area side are unselected.
In this case, only the surfaces at the smaller
area side are used to offset. Figure 21. Side Selection Area Ratio. The image on the left has a side selection
area ratio = 3, and the image on right has a side
selection area ratio = 1.5. The bottom surface
area is twice as much as the top surface,
therefore the bottom will not move (note the lock
indicator).
The ends of the dimension lines that are allowed to move
are marked with arrows, while the locked ends are marked
with blocks. Figure 22. Example of Lock Icons
When both sides have surfaces that satisfy rule 1 above,
rule 3 can be manually overridden. In this case the lock
controls (spheres near the icons) define the offset
scenario. Clicking the lock control handles will toggle
the lock state between locked and unlocked for that end.
If a lock control state is manually specified, then rule
3 is ignored for that dimension manipulator and the Side
Selection Area Ratio option no longer applies.
When Sides Selection is set to Manual, the surfaces to
offset are selected using the Surfaces to Move selector.
The manual surface selection is then governed by the
lock state of the dimension manipulator ends.
Note: Out
of the selected surfaces, only those that are linked
to at least one of the dimension manipulator ends by
a continuous selection are actually used in the
offset.
With the manual selection, the use of Separator Lines is
also available (see the surface edit
subpanel for details).
Manual
Manually select the surfaces to offset.
With manual side selection, surfaces are selected
erroneously, and the results can be unexpected or
catastrophic. Figure 23. Dim 50.5 Changed to 70 to Move the
Wall. The three highlighted surfaces are selected in
order to change the dimension from 50.5 to 70 and
move the wall to a new position.
Advanced Considerations
Advanced considerations to keep in mind when changing the dimensions of existing
geometry.
In practice, changing of a linear dimension in a model normally implies either
stretching/compressing in the direction of the modified dimension or changing of a
diameter/radius. With dimensioning funtionality, a combination of both modification
types is provided.
In the example below, one of the two D=52 dimensions is changed to D=60. How the offset
is performed will give different results, both of which may be valid, depending on which
of the two dimension manipulators is changed.
Figure 24. Original model
Figure 25. Edge Fillet Surfaces Selected. The fillet radius is scaled.
When the value of the upper dimension manipulator is modified from 52 to 60, the edge
fillet surfaces are adjacent to the modified manipulator and are offset as selected
surfaces. As such, they are offset with the LSC interpolation, which results in a
preservation of their shape along with the change in radii.
When the value of the lower dimension manipulator is modified from 52 to 60, the edge
fillet surfaces are not adjacent to the modified dimension manipulator and are curved,
so they are offset as involved surfaces. Using automatic interpolation, it is recognized
that these two curved surfaces can be simply stretched to provide the model continuity
via the global interpolation method.
When using manual surface selection and changing the same lower dimension, a variety of
results are obtainable depending on the selected surfaces. Some of the possible results
are shown below.
Note: In each row of the three images, the first two show the initial
selection from two angles, to reveal all of the selected surfaces, while the third
shows the results of the dimension change based on those selected
surfaces.
Figure 27. Original model, 3 surfaces selected
Figure 28. Original model, 3 surfaces selected
Figure 29. Result of the dimension 52 change to 60
Figure 30. Original model, 4 surfaces selected
Figure 31. Original model, 4 surfaces selected
Figure 32. Result of the dimension 52 change to 60
Figure 33. Original model, 5 surfaces selected
Figure 34. Original model, 5 surfaces selected
Figure 35. Result of the dimension 52 change to 60
Figure 36. Original model, 6 surfaces selected
Figure 37. Original model, 6 surfaces selected
Figure 38. Result of the dimension 52 change to 60
Figure 39. Original model, 7 surfaces selected
Figure 40. Original model, 7 surfaces selected
Figure 41. Result of the dimension 52 change to 60
The following steps are used to calculate the offset values of the selected surfaces.
The required shift in the dimension manipulator direction is calculated as a
difference between the requested distance and the actual distance between the
dimension manipulator ends.
If both dimension manipulators ends are allowed to move, the required shift is
divided by two.
When Sides Selection is set to Auto, an
end is allowed to move if it belongs to a surface that is automatically
selected to move. When this can be overridden manually by you, the lock
controls appear.
When Sides Selection is set to
Manual, an end is allowed to move if it belongs
to a manually selected surface, and the surface normal at the dimension
manipulator end forms an angle with the dimension manipulator direction that
is less than arccos(0.05) (87.134016 degrees).
For example, the right
end of the dimension manipulator belongs to only the selected surface 2. The
normal to surface 2 at the right end creates a 90-degree angle with the
dimension manipulator and thus the end is not allowed to move. The left
dimension manipulator end belongs to both selected surfaces 1 and 2. The
normal to surface 1 at the left end makes a 0-degree angle with the
dimension manipulator direction, and thus the left end is allowed to
move.Figure 42. Figure 43. . In this example, the right end belongs to selected surfaces 2 and
7, with the left end belonging to selected surfaces 1 and 2. Thus,
both ends are allowed to move.
When only planar surfaces are selected, the absolute value of its normal offset
is defined as the absolute value of the required shift multiplied by the cosine
between the normal to the surface and the dimension manipulator direction. Figure 44.
For a planar surface, this provides that its shift in the dimension
manipulator direction is equal to the required shift.
When curved
surfaces are included and the Sides Selection is set to Manual, the rules of
the offset value calculations are more complex. The problem in this case
originates from the fact that a selected curved surface can provide a smooth
link between the selected planar surfaces that are tilted by different
angles versus the dimension manipulator direction. When smooth, adjacent
surfaces are offset, they must be offset by the same value to ensure
continuity of the result, because in this case it is not possible to
reconcile the different offset values as discussed earlier. This means that
the planar surfaces with a different tilt towards the dimension manipulator
direction cannot be offset by different distances, as shown above, when the
planar surfaces are smoothly linked by a selected surface.
The current
algorithm to define the offset value in the general case, for both curved
and planar surfaces, is as follows. For a selected surface adjacent to the
dimension manipulator end, its offset is calculated as shown in the image
above, based on the normal to the surface at the dimension manipulator end.
For a selected surface that is not adjacent to the dimension manipulator
end, a chain of selected surfaces that links it to the related end is
detected, and the offset is calculated along the chain, from the previous
surface to the next. The calculation along the chain is based on the
following:
If the surfaces are smoothly adjacent, the offset value is directly
passed from one surface to the next.
If the surfaces are not smoothly adjacent, the offset is calculated
in such a way that for a planar surface the result as shown in the
image above is obtained.
The problem here is that when several chains of selected
surfaces connect a selected surface with the related dimension manipulator
end, the offset results for the surface obtained along the different chains
can contradict each other. Then the dimensioning result may be corrupt.
Therefore, it is important to make appropriate manual surface
selections. Figure 45. . Surfaces 1, 2, and 4 selected, D=52 changed to D=60. Surfaces 1,
2, and 4 offset by 8. Figure 46. . Surfaces 1, 3, and 4 selected, D=52 changed to D=60. Surface 1
offset by 8, Surfaces 3 and 4 by 0. Figure 47. . Surfaces 1, 2, 3, and 4 selected, D=52 changed to D=60. The result
is corrupt.
For each selected surface the sign of the offset is defined so that it will move
in the same direction as the dimension manipulator end to which it is
related.
A surface can be related to one, and only one, of the dimension
manipulator ends. For this, first, the dimension manipulator end must be
allowed to move. Second, the surface should be linked to the dimension
manipulator end over a chain of adjacent selected surfaces. Third, in the
case when the surface is linked to both dimension manipulator ends which are
allowed to move, the surface will be related to the end that is closer to
it.
As an example, selected surface 2 will have an offset of 0, because
cos(90) = 0. The purpose for selection of this surface is just to provide a link
from the dimension manipulator ends to the other selected surfaces. Surface 1 is
at the moving dimension manipulator end, and surface 3 moves as surface 1.Figure 48.
Following the same rules, surfaces 1 and 7 are at the moving dimension
manipulator ends. Surfaces 3 and 5 move as surface 1, and surfaces 4 and 6
move as surface 7. Figure 49.