rADINA Elnotes - 94

UTOMATIC
YNAMIC
NCREMENTAL
ONLINEAR
NALYSIS
Release Notes
ADINA 9.4.2 July 2018
ADINA R & D, Inc.

ADINA System 9.4
Release Notes
(for version 9.4.2)
July 2018
ADINA R & D, Inc.

71 Elton Avenue
Watertown, MA 02472 USA
tel. (617) 926-5199

telefax (617) 926-0238
www.adina.com
Notices
ADINA R & D, Inc. owns both this software program system and its documentation. Both the
program system and the documentation are copyrighted with all rights reserved by
ADINA R & D, Inc.
The information contained in this document is subject to change without notice.
ADINA R & D, Inc. makes no warranty whatsoever, expressed or implied that the Program and its
documentation including any modifications or updates are free from errors or defects. In no event
shall ADINA R&D, Inc. become liable to the User or any party for any loss, including but not
limited to, loss of time, money or goodwill, which may arise from the use of the Program and its
documentation including any modifications and updates.
Trademarks
ADINA is a registered trademark of K.J. Bathe / ADINA R & D, Inc.
All other product names are trademarks or registered trademarks of their respective owners.
Copyright Notice
 ADINA R & D, Inc. 2018

July 2018 Printing
Printed in the USA
New and updated feature summary

This section lists the new and updated features that are available in ADINA System 9.4, as
compared with ADINA System 9.3.4.
There are new commands and new and changed parameters associated with the new and
updated features. The release notes refer to the commands and parameters in the command-
line formats. Further information about the new commands and new and changed parameters
can be found in the AUI Command Reference Manuals.
For user interface users, most command-line parameters have analogous fields in the dialog
boxes.
Note, when we refer to documentation, we refer to the versions of the documentation given
below in the “Available Documentation” section.
Features for all programs

Table of supported platforms, SMP program versions
Platform Operating system Fortran compiler

Linux x86_64 2.6.18 and higher, glibc 2.5 Intel ifort 11.1
and higher,
gcc 4.1.2 and higher
Linux 86_64, 2.6.32 and higher, glibc 2.12 Intel ifort 14.0.4
AVX and higher,
extensions gcc 4.4.6 and higher
Windows Windows 7, 8, 8.1, 10 Intel Visual Fortran
x86_64 14
Windows Windows 7, 8, 8.1, 10 Intel Visual Fortran
x86_64, AVX 14
extensions
1) All program versions are 64-bit, using the x86_64 architecture. The Intel 64 and AMD
Opteron implementations of the x86_64 architecture are supported.
ADINA System 9.4 Release Notes 3

Table of supported platforms, DMP program versions
Platform Operating system Fortran compiler

Linux x86_64 2.6.18 and higher, glibc 2.5 Intel ifort 11.1
and higher,
gcc 4.1.2 and higher
Linux 86_64, 2.6.32 and higher, glibc 2.12 Intel ifort 14.0.4
AVX and higher,
extensions gcc 4.4.6 and higher
1) The DMP program versions are based on OpenMPI version 3.0.0.
2) The DMP program version with AVX extensions is newly added for version 9.4.
3) The DMP program versions do not support all features of the SMP program versions, see
the DMP Users Guide for details.
Sparse solvers
The sparse solvers have been substantially improved in version 9.4. The solution time
required is often reduced by a factor of 4 or more for large models, and the memory required
is also reduced. Because the 3D-iterative solver and subspace iteration solver use the sparse
solver, the solution time required by these solvers is also reduced.
The sparse solvers also make better use of the underlying CPU architecture, so the
improvement in the sparse solvers is greater when using later CPU architectures (for example,
the Ivy Bridge or later architectures).
The sparse solvers for the DMP versions are improved in version 9.4.2, see the section
“Improvements between 9.4.1 and 9.4.2”.
ADINA Solids & Structures features

Enriched subspace iteration method
A novel extension of the Bathe subspace iteration method for the solution of generalized
eigenvalue problems in structural dynamics has been presented by K.T. Kim and K.J. Bathe,
“The Bathe subspace iteration method enriched by turning vectors”, Computers & Structures,
186 (2017), pp 11-21. The method has been implemented into the ADINA program under the
name of “Enriched subspace iteration method”.
As a first step, the method uses the standard (Bathe) subspace iteration method to perform one
iteration using standard starting vectors. The Enriched subspace method then uses a 10-step
procedure described in the above reference, where turning vectors are calculated to achieve
fast convergence. After convergence is reached, a Sturm sequence check can be performed.
4 ADINA System 9.4 Release Notes

The default number of iteration vectors, q, used is max (1.4p, p+8), where p frequencies and
associated mode shapes are to be calculated
The solution time required is often reduced by a factor of 3 or more for large models as
compared to the standard subspace method.
 Command-line:
FREQUENCIES ... METHOD={.../ENRICHED-SUBSPACE)
New robust frequencies within an interval calculations
The subspace iteration method can be employed to calculate frequencies within an interval.
Lower and upper bounds of frequencies need to be specified as well as the maximum number
of frequencies ADINA can calculate. The program then checks how many frequencies exist in
the specified interval and calculate them all if the number of existing frequencies is smaller or
equal to the maximum number of frequencies specified. If the number of existing frequencies
is larger than the specified maximum number of frequencies, then the program stops with a
message that more frequencies exist than the maximum number provided via input.
Other improvements to frequency calculations
The default number of frequencies to be calculated depends on the type of frequency

calculations, as follows:
100 if frequencies are confined to an interval and the subspace method is used.
0 if using the CMS method with Bathe subspace iteration, in which case only the reduced
system matrices are calculated in the first phase of CMS.
1 otherwise.
The default number of iterations for the subspace iteration method is increased from 24 to 48.
Improvements for SMP
The CPU time for element assembly is reduced when using SMP with multiple threads. The
improvement is made for all element types, but the improvement is greatest when using
collapsed 3-D solid elements, in particular for large 4-node tetrahedral models, and models
where a master node of a rigid link spider is connected to many slave nodes.
Time function scaling of mass and damping matrices
The mass matrix and damping matrix can be scaled by time functions when low-speed
dynamics or implicit dynamics is used. This feature can be used to control the low-speed
dynamics stabilization forces. This feature can also be used to ensure that the final converged
low-speed dynamics solution satisfies static equilibrium.

 Command-line:
AUTOMATIC TIME-STEPPING ... LSMASS-TF LFDAMP-TF
Improvements to lumped mass matrices for 3D solid elements
The lumped mass matrices for the 5-node pyramid, 13-node pyramid and 14-node pyramid 3D
solid elements are improved. In version 9.3, the element mass is divided evenly between the
element nodes, before the brick is collapsed into the pyramid, and in version 9.4, the element
mass is divided evenly between the element nodes, after the brick is collapsed into the
pyramid.
For example, in version 9.3, the 5-node pyramid is treated as a collapsed 8-node brick, in
which the apex of the pyramid is formed by collapsing the top four nodes of the brick,
therefore half of the element mass is assigned to the apex node and one-eighth of the element
mass is assigned to each of the base nodes. In version 9.4, the apex node and each of the base
nodes is assigned one-fifth of the element mass.
Improvements to Noh-Bathe method for explicit time integration
The load magnitude at the mid-step can be chosen to be either the average load for the step (as
in 9.3) or can be directly evaluated from the time functions.
 Command-line:
ANALYSIS DYNAMIC-DIRECT-INTEGRATION ...
MIDLOAD={MEAN/TIMEFUNCTION}
RBE3 elements
ADINA Structures now supports RBE3 elements. An RBE3 element constrains the motion of
a slave node to the weighted average of the motion of a set of master nodes.
 Command-line:
RBE3
Improvements to connector elements
Connector elements can reference multilinear matrices. In this case, the stiffness and/or
damping matrices are directly entered and are multilinear functions of relative displacement,
relative velocity, elastic force, damping force or temperature.
 Command-line:
EGROUP CONNECTOR ... SUBTYPE={.../MATRIX-MULTILINEAR)
CONN-PROP MATRIX-MULTLINEAR
MATRIX-NL-K
MATRIX-NL-C
MATRIX-K
MATRIX-C

Improvements to element birth-death
Version 9.4 has a number of improvements related to element birth-death. Therefore models
with birth-death might give different results in version 9.4 as compared to version 9.3,
especially if the element with birth-death contribute to the mass matrix.
One noteworthy change is the following.
In version 9.3, a DOF is marked as “attached to a dead element” if the DOF is attached to a
dead element. If the DOF has a zero pivot in the stiffness matrix, a large value is assigned to
the pivot and the equation solution continues.
However, this approach can lead to difficulties if the DOF is attached both to a dead element
and an active element. In this case, if the DOF has a zero pivot in the stiffness matrix (for
example, due to rigid-body motions of the assemblage), a large value is still assigned to the
pivot. This large value removes the (physically present) rigid body motion.
In version 9.4, a DOF is marked as “attached to a dead element” if the DOF is not attached to
any active elements. So in the above example, the DOF is unmarked and the zero pivot is
printed.
Therefore certain models that had no zero pivots in version 9.3 might have zero pivots in
version 9.4. The workaround is to identify the source of the zero pivots (for example, 6 DOF
shell elements with drilling stiffness), and to change the model to remove the zero pivots.
Thermo-plastic-cyclic material model
The thermo-plastic-cyclic material model is a new material model in version 9.4.
The thermo-plastic-cyclic material model is a temperature-dependent version of the plastic-

cyclic material model already available. All material constants are temperature-dependent.
The thermo-plastic-cyclic material model can be used with truss, 2-D solid, 3-D solid, beam,
shell and 3-D shell elements.
 Command-line:
MATERIAL THERMO-PLASTIC-CYCLIC
Improvement to thermo-plasticity with kinematic hardening
The back stress temperature correction implemented in version 9.3 is now used by default in
the thermo-plastic material models with kinematic hardening.
 Command-line:
KINEMATICS BACKSTRESS-CORRECTION={YES}

Improvements to Bergstrom-Boyce material model
The material parameters can be calibrated either in accordance to the Bergstrom-Boyce model
defined in the paper by Bergström and Boyce (1998), or in accordance with the MCalibration
software developed by Veryst Engineering.
An additional option is available for entering the flow resistance (viscous deformation rate
constant for network B).
 Command-line:
RUBBER-VISCOELASTIC BERGSTROM-BOYCE CALIBRATION TAUBASE
Improvements for plasticity, creep and viscoelastic material models
For 2D solid, 3D solid and 3D-shell elements, the large strain kinematics are now used by
default when large displacement kinematics are selected. This improvement is made by the
addition of a new parameter value in the KINEMATICS STRAINS command. When
KINEMATICS STRAINS=DEFAULT and EGROUP STRAINS=DEFAULT, small strain
kinematics are used when small displacement kinematics are selected, and large strain
kinematics are used when large displacement kinematics are selected.
This change is made because, even when the strains are relatively small (~2%), the results are
more accurate and the convergence is more robust when large strain kinematics are used.
For the other shell elements that allow large strain kinematics (single-layer MITC3, MITC4,
MITC6, MITC9, MITC16 elements), when KINEMATICS STRAINS=DEFAULT and
EGROUP STRAINS=DEFAULT, small strain kinematics are used for both small and large
displacement kinematics.
As before, the large strain kinematics can be explicitly selected or deselected using
KINEMATICS STRAINS=(SMALL or LARGE), or EGROUP STRAINS=(SMALL or
LARGE).
This change has no effect if KINEMATICS STRAINS=(SMALL or LARGE), or EGROUP

STRAINS=(SMALL or LARGE) are explicitly specified in the model input.
 Command-line:
KINEMATICS STRAINS={.../DEFAULT}

Improvements to output of results for large strain plasticity models
The Hencky strains can now be output (and is the default output) for the following plasticity
models, when used with large strain kinematics.
MATERIAL ANAND
MATERIAL CREEP
MATERIAL CREEP-IRRADIATION
MATERIAL CREEP-VARIABLE
MATERIAL GURSON-PLASTIC
MATERIAL MROZ-BILINEAR
MATERIAL MULTILINEAR-PLASTIC-CREEP
MATERIAL MULTILINEAR-PLASTIC-CREEP-VARIABLE
MATERIAL PLASTIC-BILINEAR
MATERIAL PLASTIC-CREEP
MATERIAL PLASTIC-CREEP-VARIABLE
MATERIAL PLASTIC-MULTILINEAR
MATERIAL THERMO-PLASTIC
 Command-line:
PORTHOLE ELEM-RESULT={.../STRAIN}
Improvements for contact
Nodal sliding distance
The nodal sliding distance is now computed for frictional contact. The nodal sliding distance
is accessed using variable NODAL_SLIDING_DISTANCE.
Friction delay
In version 9.4, there are three friction delay options:
No No friction delay (the default)
Frictionless Frictionless conditions are assumed in a solution step in which contact is

established. This option corresponds to the friction delay option in version
9.3.
Stabilized Stabilized frictional conditions are assumed in a solution step in which

contact is established. Stabilized friction delay is introduced in version 9.4.
In addition, a global flag is added to the CONTACT-CONTROL command so that the friction
delay option can be toggled on and off for the entire model.

 Command-line:
CONTACT-CONTROL FRIC-DELAY
CGROUP FRIC-DELAY={NO/FRICTIONLESS/STABILIZED/DEFAULT}
Contact slip velocity loading
The sign convention for contact slip velocity loading onto 3D contact surfaces has been
reversed in version 9.4. This reversal is done so that the contact slip velocity loading works
the same way for both 2D and 3D contact surfaces; contact slip velocity loading is always
addded to the nodal velocities to obtain the total contact surface velocity.
Contact slip velocity loadings can no longer be applied in explicit analysis. In previous
versions, any contact slip velocity loading is ignored in explicit analysis.
The description of contact slip velocity loading is improved in the ADINA Structures Theory
and Modeling Guide, Section 5.11.
Improvements for element group based mass-proportional loading
Mass proportional loads can now be applied to a portion of a model (i.e. group-based mass
proportional loads) in addition to the existing model-based mass proportional loads. Any
number of element groups can be included in the group-based mass proportional load
calculations. Nodal forces for group-based mass proportional loads are calculated in the same
way as for model-based mass proportional loads.
Improvements for solution accuracy indicators
The solution accuracy indicators have been improved, as follows:
An increment of work solution accuracy indicator and a force solution accuracy indicator
are now printed.
The same reference is now used in the solution accuracy indicators for both the time steps
and bolt steps.
The solution accuracy indicators are now printed for linear and nonlinear implicit
dynamics, and linear static analysis with stiffness-stabilization or shell drilling stiffness.
A warning message is now printed to the message file if the stiffness-stabilization

indicator or the shell drilling indicator are greater than 5% of the external loading.
Improvements to energy calculations
The following improvements are made to energy calculations:
Energy calculations for 3D shell elements

Energy calculations for iso-beam elements

Energy calculations for truss elements
Energy calculations for spring elements
Energy calculations for thermal strain energy
Energy calculations for creep strain energy
Energy calculations for thermal-plastic-cyclic material
Improvements for OP2 file creation
The OP2 file is now saved after each converged time step, such that the OP2 result file can be
read during the solution.
Modal participation factor analysis results are now saved to the OP2 result file.
Mode superposition analysis results are now saved to the OP2 result file.
OP2 result files can now be saved for models that use the ANALYSIS-SWITCH feature.
The solid solution results in an FSI analysis are now saved to the OP2 result file.
For models in which the OP2 result file is to be saved, the AUI now checks on data file
generation that the element label numbers are unique for all elements in the model.
Improvements for I-deas universal result file creation
Frequency analysis results are now saved to the I-deas universal result file.
Modal participation factor analysis results are now saved to the I-deas universal result file.
The solid solution results in an FSI analysis are now saved to the I-deas universal result file.
EnSight files
EnSight files can be created by ADINA Structures. Multiple EnSight files can be combined
into a single .case file. It is possible to save both the EnSight files and the porthole file in the
same solution run. Either a single EnSight file or one EnSight file per time step can be
created.
EnSight files are saved to a separate folder in the working directory.
The results to be saved to the EnSight file can now be selected using RESULTS-ELEMENT.
Multiple EnSight case files can now be merged or copied into a single case file. This feature is
useful for restart runs.

 Command-line:
ENSIGHT-CASEFILE
MASTER RESULTS={.../ENSIGHT/ENSIGHT+PORT}
RESULTS-ENSIGHT
ENSIGHT-CASEFILE
ADINA Thermal & TMC features

Improvements to radiosity analysis
The thermal system of equations is now solved using the sparse solver.
The radiosity heat flux terms in the tangent conductivity matrix can be included using a
scaling factor (0.0 = no radiosity heat flux terms, 1.0 = full radiosity heat flux terms).
A steady-state radiosity analysis can now be solved.
A more accurate radiosity calculation for simple axisymmetric models, e.g. concentric
spheres, is now included. This option can be selected using RGROUP RADIATION2 ...
AXISYMPLE=YES. In addition, the integration order in the circumferential direction can
now be chosen to be 48.
 Command-line:
RGROUP RADIATION2 ... CINT AXISIMPLE
TMC-CONTROL RAD-FACT
Membrane conduction elements
Membrane conduction elements are now supported.
Improvements to linear thermal analysis
For linear thermal problems, a linear thermal analysis is performed in one-way TMC and
iterative TMC analysis.
In version 9.3, if the structural problem is nonlinear, a nonlinear thermal analysis was always
performed, even if the thermal problem is linear.
ADINA CFD & FSI features

Reynolds-equation model
The Reynolds Equation (RE) is used to govern the pressure distribution of thin viscous fluid
films in lubrication theory. It has been implemented for FCBI-C elements in version 9.4. The
RE is seamlessly integrated into ADINA-CFD and can be used for (but not limited to) laminar

and turbulence flows and coupled with ADINA in FSI computation.
ADINA-CFD supports the RE for both smooth and rough walls. The Patir & Cheng model
(see references below) has been implemented for determining effects of roughness on
lubrication.
Ref. Nadir Patir and H. S. Cheng, “An Average Flow Model for Determining Effects of
Three-Dimensional Roughness on Partial Hydrodynamic Lubrication”, Journal of
Lubrication Technology, 100 (1978) 14-17.
Ref. Nadir Patir and H. S. Cheng, “Application of Average Flow Model to Lubrication
Between Rough Sliding Surfaces”, Journal of Lubrication Technology, 101 (1979)
220-229.
The RE is active in specified element groups and at the locations where the boundary distance
is less than the specified close/open values defined via the BOUNDARY-DISTANCE
command.
It is worth noting that RE is much faster, cheaper and more accurate than the classical Navier-
Stokes equation solution when applied to thin fluid film.
It is also worth noting that, although the velocity solution is the boundary-to-boundary
averaged quantity, the shear effects in the thin film are still included in the thermal solution,
turbulence variables, etc., as well as included in all solution output files (porthole file and
graphs *.ite file). The boundary distance that plays an important role in the RE can be saved
as an element result.
 Command-line:
FCTRL REYNOLDS-EQUATION
RESULTS-ELEMENT ... BOUNDARY-DISTANCE
Universal barotropic cavitation
The Universal Barotropic Cavitation (UBC) is a simple and effective cavitation model. It has
been implemented for the FCBI-C element in version 9.4. It can be used in the
incompressible flow module of ADINA-CFD (although the two phase flow in essence is
compressible flow).
It is worth noting that this model can be used without activating the VOF and/or heat transfer
formulations that are usually required in the caviation model. The UBC model is frequently
used together with RE (as well as other fluid formulations) when cavitation occurs.
 Command-line:
FCTRL BAROTROPIC-CAVITATION
EGROUP TWODFLUID ... MAT2 FCTRL1 FCTRL2
EGROUP THREEDFLUID ... MAT2 FCTRL1 FCTRL2

Location domains
Location domains are lists of nodes, elements, boundary condition label numbers, internal
faces or external faces, that can be used with solution graphs.
 Command-line:
LOCATION
Solution graphs
Solution graphs of convergence and solution variables can be created for FCBI-C element
results. The solution graphs are saved to the solution monitor (*.ite) file, and can be viewed
during runtime and after the solution is finished using the ADINA Graph Viewer. This
feature can be used to graphically monitor convergence and create graphs of solution results.
There are more than one hundred solution variables currently available in version 9.4. The
variables are available at nodes, element, internal/boundary faces and/or in the whole
computational domain. Some of the solution variables are
 Equation and variable convergence history for all active independent variables
 Memory usage history
 Mass rate history at inlet and outlet boundaries
 Any active solution variable history
 Shear stress, stress, reaction force, etc. on boundary face
 Heat flux and heat flow rate on boundary face
 Torque vector and its magnitude on boundary face
 Shear velocity and y-plus history in turbulence flow
 Gap status history
 Command-line:
GRAPH
Adaptive meshing using SAM
An additional parameter DISTANCE is implemented in the ADP-CONTROL command.

When DISTANCE = 0 (the default), parameters DX, DY, DZ determine both the direction
and magnitude for swept adaptive meshing. When DISTANCE > 0, parameters DX, DY, DZ
determine only the direction and parameter DISTANCE determines the magnitude.
 Command-line:
ADP-CONTROL ... DISTANCE
EnSight files
It is possible to save both the EnSight files and the porthole file in the same solution run.
Either a single EnSight file or one EnSight file per time step can be created.

EnSight files are saved to a separate folder in the working directory.
The fluid solution results to be saved to the EnSight file can now be selected using
NODESAVE-STEPS.
Additional solid and fluid solution results can now be saved to the EnSight file.
Multiple EnSight case files can now be merged or copied into a single case file. This is useful
for restart runs and runs with SAM.
 Command-line:
MASTER
RESULTS={PORTHOLE/ENSIGHT/ENSIGHT+PORT/NEU/NEU+PORT}
RESULTS-ENSIGHT
ENSIGHT-CASEFILE
Improvements to saving results
The boundary distance variable can be saved for conditional loading and Reynolds fluid
elements.
 Command-line:
RESULTS-ELEMENT … BOUNDARY_DISTANCE
AUI features
User interface improvements
New icons
The following new icons are added:
-YZ View
-XY View
-XZ View
Mesh Element Spider
Improvements to the Model Tree
Improvements to zones and zone groups
Right-click Display of a zone now displays the zone without re-zooming the display.
Right-click Hide of a zone is now supported to remove that zone from the display.

Right-click Rename is now supported for ZZ_MULTI-ZONE and for zones that are
currently displayed.
The automatically created zones are now organized into zone groups.
The Model Tree display of zones and zone groups is improved.
Improvements to categories
Right-click Display and Hide is now supported for all Model Tree categories to display
and hide that category, respectively. This can be used, for example, the display or hide all
entities with a given material model, etc.
The Model Tree labels for node sets, element sets, element-edges sets, and element-face sets
are now consistent with the set labels. For example, node set 5 is now labeled 5 in the Model
Tree.
Right-click Rename is now supported for node sets, element sets, element-edges sets, and
element-face sets.
The display of the loads and boundary conditions in the Model Tree is now improved.
Right-click Rename is now supported for loads and boundary conditions.
Model Tree lists is now limited to 10,000 items. Beyond the limit, the Model Tree summarizes
the list.
Improvements to zones and zone groups
The following right-click options are available for defining zones: Adjacent,
Attached, Auto-Chain, Edge angle [20], Invert, Attach to Nodes.
Mode Tree items can now be picked during zone definition. For example, an element group
can be picked in the Model Tree to append that element group to the zone definition.
Improvements for picking
Improvements to picking geometry
The following right-click options are available for picking lines and edges: Adjacent,
Attached, Auto-Chain, Edge angle [20], Invert.
The following right-click options are available for picking element-edges: Adjacent,
Attached, Auto-Chain, Edge angle [20], Invert.
The right-click picking options only act on the entities displayed in the current mesh plot.

Improvements to subordinate dialog boxes in Linux
Subordinate dialog boxes in Linux now open in front of the parent dialog box.
Improvements to the message window
The message window is automatically cleared when a new database is created.
Improvements to the solution graph view
A new graph viewer, ADVIEW, is supported that reads the following file formats:
ADINA Structures equilibrium iteration files (*.ite).

ADINA Structures energy files (*.egy).
ADINA CFD solution files (*.ite).
Right-click display options are supported in the graph viewer.
Improvements to the solution window
The size of the solution window can be resized by dragging the sides of the window.
Improvements to the dialog boxes in Linux
Clicking on the [x] button on the top right corner of the dialog box closes the dialog box.
In-line and block comments in the command-line parser
In-line comments can now be included using //, as in this example:
LINE STRAIGHT 1 P1=2 P2=3 // Top of model
Block comments can now be included using /* */, as in this example:
COORDINATES POINTS
CLEAR
1 1.0 0.0 0.0
/* -----------
2 0.0 1.0 0.0
3 1.0 0.0 0.0
------------ */
2 0.0 2.0 0.0
3 2.0 0.0 0.0

Improvements to command file writing
The COLORZONE command is now written to the command file.

Zone groups are now written to the command file.
Model definition
Efficiency improvements
The efficiency of importing Nastran models is improved.
The efficiency of applying fixities is improved.
Improvements to extruded lines, surfaces and volumes
Extruded lines, surfaces and volumes can be defined in terms of the distance and direction of
the extrusion.
 Command-line:
LINE EXTRUDED ... DISTANCE
SURFACE EXTRUDED ... DISTANCE
VOLUME EXTRUDED ... DISTANCE
Improvements to Parasolid geometry engine
The Parasolid geometry engine kernel is now based on Parasolid V27.1.
As a result, free-form meshes based on the Parasolid geometry engine might be different in
version 9.4 than in previous versions.
Improvements to geometry tolerances
The tolerances used in geometry and meshing commands are improved. For each tolerance, if
“0” is specified, then the tolerance distance is specified by the TOLERANCES GEOMETRIC
command.
Improvements for entities assigned to geometry points
There are several types of entities can be assigned to geometry points, for example, fixities,
skew systems, concentrated masses and dampers, and rigid links. The entire list of entities is
too long to give here.
In version 9.3, the entity is either applied to all nodes close to the point (within a tolerance), or
to the node that was chronologically last attached to the point, depending on the type of entity.
In version 9.4, by default, the entity is applied to all nodes attached to the point. The default
can be changed in order to obtain the 9.3 behavior.

 Command-line:
CONTROL ... APPLYPOINT={ATTACHED/TOLERANCE}
Improvements for geometry deletion
When geometry (points, lines, edges, faces, etc.) is deleted, the rigid links, constraints,
concentrated masses, concentrated dampers, springs, alignment elements and connector
elements associated with the geometry can optionally be deleted.
 Command-line:
MISC-OPT DELETE-REF={NO/YES}
Improvements for element edge-set, element face-set and element set definitions
Element edge-sets, element face-sets and element sets can be defined in terms of node-sets.
Element edge-sets of shell elements can now be defined using in terms of nodes, points, auto-
chaining, attached, and adjacent.
 Command-line:
ELEDGESET ... OPTION={.../NODESET}
ELFACESET ... OPTION={.../NODESET}
ELEMENTSET ... OPTION={.../NODESET}
Improvements for node set definitions
Node sets can be defined in terms of nearby geometry points
 Command-line:
NODESET ... OPTION={.../NEARBY-POINT}
Improvements to deleting a range of entities
When deleting a range of entities using the FIRST and LAST options, all entities in the
specified range are deleted, irrespective of whether FIRST or LAST entity exists in the model
or not.
Improvements to element deletion
An additional option is available that controls which nodes are deleted when elements and
element groups are deleted. When NODES or NODE-DELETE=UNUSED, nodes are deleted
only if they would be unused after the specified elements are deleted.
 Command-line:
DELETE EGROUP ALL NODES=UNUSED
DELETE ENODES NODE-DELETE=UNUSED
ELDELETE ... NODE-DELETE=UNUSED

Improvements to empty element groups
On data file generation, if an element group is empty, the AUI issues a warning message file
on data file generation but continues. In version 9.3, data file generation was stopped if an
element group was empty.
Improvements for contact surface definitions
Contact surfaces can be generated for all contact surfaces in a contact group, or for all contact
groups. Contact surfaces can be generated using mixed meshing (the default).
 Command-line:
CSURFACE ... TYPE
CSURFACE ... MESHING={V93/MIXED}
Improvements to deletion of contact meshes
Additional options are available for the deletion of contact meshes. The contact meshes can
be deleted from a contact surface, from all contact surfaces in a contact group, or from the
entire model, using a single command.
 Command-line:
CSDELETE TWO-D ... DTYPE
CSDELETE THREE-D ... DTYPE
Improvements to boundary condition definitions
A description can be added to all boundary condition definitions
 Command-line:
BOUNDARY-CONDITION ... DESCRIPTION
Improvements to spring element and axisymmetric truss defnitions
The element group number can be specified in the spring element definition and axisymemtric
truss element definition commands.
 Command-line:
SPRING POINTS ... GROUP
SPRING LINES ... GROUP
SPRING NODESETS ... GROUP
SPRING-6DOF POINTS ... GROUP
SPRING-6DOF LINES ... GROUP
SPRING-6DOF NODESETS ... GROUP
TRUSS-POINTS ... GROUP
TRUSS-LINES ... GROUP

Improvements to rigid link definitions
Rigid links can now be defined using volumes and bodies
Any combination of slave and master entity types is now allowed.
 Command-line:
RIGIDLINK ... MASTERTYPE={.../VOLUME/BODY}
Improvement to gluemesh defintions
Gluemeshes can be applied between volumes and bodies.
 Command-line:
GLUEMESH stypei=VOLUME
Improvements to cyclic boundary definitions
The list of slave and master cyclic boundaries can now be order independent. That is, the nth
entry of the slave list no longer needs to correspond to the nth list entry of the master list.
Cyclic boundaries can be defined using element-edge sets and element-face sets.
The dialog box for cyclic boundary definitions is improved.
 Command-line:
CYCLICBOUNDARY ELEDGESET
CYCLICBOUNDARY ELFACESET
Improvements to specification of active dofs
Active degrees of freedom of the reduced model for SIMPACK analysis (or boundary degrees
of freedom for the CMS method) can be specified using nodesets.
 Command-line:
DOF-ACTIVE TYPE={NODE/NODESET}
Improvements in coincidence checking for mesh generation commands
Nodal coincidence checking can be done on all boundary nodes of the meshed body, against
all nodes currently displayed in the mesh plot MESHPLOT00001.
 Command-line:
BODY REVOLVED ... NCOINCIDE={.../DISPLAY}
BODY SWEEP ... NCOINCIDE={.../DISPLAY}
GLOFTED ... NCOINCIDE={.../DISPLAY}
GSURFACE ... NCOINCIDE={.../DISPLAY}

GVOLUME ... NCOINCIDE={.../DISPLAY}

GFACE ... NCOINCIDE={.../DISPLAY}
GBODY ... NCOINCIDE={.../DISPLAY}
GHEXA ... NCOINCIDE={.../DISPLAY}
BHEXA ... NCOINCIDE={.../DISPLAY}
COPY-MESH-BODY ... NCOINCIDE={.../DISPLAY}
Improvements for moving mesh and special boundary conditions definitions
Moving mesh and special boundary conditions can be specified using geometric entities or
finite element entities.
 Command-line:
MASTER ... ALE={.../GEOM-ELEM}
Improvements for gap boundary condition definitions
Gap boundary conditions can be defined using a combination of lines and edges (or surfaces
and faces).
Gap boundary conditions can be applied to multiple bodies.
 Command-line:
BOUNDARY-CONDITION GAP ... GTYPE={.../TWO-D/THREE-D}
Improvements for leader-follower definitions
The default cone angle for all leader-followers can be specified. Leader-following pairs can
be defined using lines and edges, using surface and faces, using points, or using nodes.
 Command-line:
LEADER-FOLLOWER ... ANGLE
LFOLLOWER LINE-EDGE
LFOLLOWER SURFACE-FACE
LFOLLOWER POINT
LFOLLOWER NODE
Improvements to wall boundary condition definitions
When defining moving mesh and special boundary conditions using element entities, the wall
boundary condition can be applied to all external boundaries where a velocity or pressure
boundary condition has not been applied. This option can be used to change the default
boundary condition to a wall. If this option is not used, the default boundary condition is zero
traction.
 Command-line:
BOUNDARY-CONDITION WALL ... ALL-EXT=FREE

Improvements for body meshing
There is an option to use mixed meshing. In mixed meshing, mapped meshing is employed
when possible, otherwise free-form meshing is employed. By default, mixed meshing is used.
 Command-line:
GBODY ... MESHING={.../MIXED}
Improvements to boundary layer meshing
A quadrilateral boundary layer mesh with triangular elements outside the boundary layer can
be generated by using GFACE … NODES=3, 6, or 7.
A prismatic boundary layer mesh with tetrahedral elements outside the boundary layer can be
generated by using GBODY … NODES=4, 10, or 11.
The boundary layer thickness is now allowed to be larger than the radius of curvature of the
body edge or body face.
The boundary layer shape can be controlled using a shape distortion parameter. The lower the
shape distortion threshold, the better the quality of the boundary layer elements.
 Command-line:
GFACE … BLSHAPE
GBODY … BLSAHPE
Improvements to the placement of midside and midface nodes
A default option is now available for the placement of midside and midface nodes. This
default option automatically selects the optimal setting for the placement of these nodes.
 Command-line:
GFACE/GBODY … MIDNODES={…/DEFAULT}
GBODY … MIDFACENODES={…/DEFAULT}
Changes to meshing defaults
When meshing faces, the default is changed to mixed meshing. In mixed meshing, mapped
meshing is employed when possible, otherwise free-form meshing is employed.
When meshing bodies, the default is changed to force all bounding edges of linked faces to
have an even number of subdivisions.
 Command-line:
GFACE … MESHING=MIXED
GBODY … EVEN=LINK

Improvements for reference temperature definitions
There are two options for reference temperature definitions:
The reference temperature is applied to all nodes. The temperature of a node is the
reference temperature plus any temperature loading applied to the node. (APPLY-
REF=ALL)
The reference temperature is applied to only the nodes on which temperature loads are not
applied. The temperatures of a node without temperature loading is the reference
temperature and the temperature of a node with temperature loading is the temperature
loading. (APPLY-REF=FREE) This behavior is the default.
 Command-line:
TEMPERATURE-REFERENCE ... APPLY-REF
Improvements for radiation surface definitions
Radiation surfaces can be defined using element edge-sets or element face-sets.
 Command-line:
RADIATION-ELEMSET
Improvements to checking for duplicate elements during data file generation
The AUI now checks for duplicate connector elements on data file generation.
The first 50 duplicate elements are printed to the message file.
 Command-line:
CONTROL … DUPLICATE
Model display and post-processing

RBE3 elements
RBE3 elements can be displayed, included in zones and highlighted:
 Command-line:
HIGHLIGHT
ZONE
Geometry points in post-processing
It is now allowed to define geometry points in post-processing.

Geometry points defined in post-processing can be used in commands that reference geometry
points, see below.
 Command-line:
COORDINATES POINTS
Commands that reference geometry points, in post-processing
In previous versions of the AUI, in order to refer to geometry points, it is necessary to load a
pre-processing database (.idb file). As an improvement, it is now possible to define a
geometry point in post-processing, then the closest node to the geometry point is chosen,
without the need to load an .idb file.
The default is to use pre-processing data if an .idb file is loaded, and to use the closest node
if no .idb file is loaded.
 Command-line:
TRACERAKE GNODES ... POINTOPTION={PREPROCESSING/CLOSEST}
GNCOMBINATION ... POINTOPTION={PREPROCESSING/CLOSEST}
GNLINE ... POINTOPTION={PREPROCESSING/CLOSEST}
Enumerated variables
Some variables are enumerated, that is, they have integer values with associated string values
(for example for PLASTIC_FLAG, 1 = elastic, 2 = plastic). In listings of enumerated
variables, the string values are now listed by default. The CONTROL command can be used
to request listing of integer values.
 Command-line:
CONTROL ... LISTVARENUM
Automatically created zones
The automatically created zones are created only when the number of automatically created
zones is less than a threshold value. This change is made because for large models, when the
number of automatically created zones is large, the time and memory required to set up the
zones is also very large.
By default, the threshold value is 100.
 Command-line:
CONTROL ... AUTOZONE={YES/NO} or a threshold number

Order of nodes in a listing of local node variables
In a listing of local node variables, the results within each element can be output either in
global node number order (as in version 9.3) or in local node number order.
 Command-line:
RESULTGRID ... RESULTORDER={LOCAL/GLOBAL}
Distance along line segments
The total distance along line segments described by nodes can be listed or graphed using a
command sequence such as
NODECOMBINATION TEST
ENTRIES NODE
1
10
2
...
DATAEND
POINTLIST TEST VAR=DISTANCE
Variable DISTANCE uses the original nodal coordinates, variable DISTANCE_POSITION

uses the updated coordinates.
Listing of modal participation factors
In an analysis in which the modal participation factors are computed for the applied loads (as
in modal superposition analysis), the modal participation factors can be listed using the
MPFINFO command.
Contact variables
The NODAL_SLIP_VELOCITY variables have been moved to the “Contact” category.
Improvements to DUMPLIST command
DUMPLIST can now dump a range of lines, and in each line, dump a range of character
positions.
 Command-line:
DUMPLIST ... ENDLINE STARTCHAR ENDCHAR

Improvements between 9.4.0 and 9.4.1

Improvements in initial memory allocation
The initial memory allocation is now much smaller for problems in which there are a large
number of elements connected to a single node.
Improvements in analysis zooming
The 11-node tetrahedral elements can be used in zooming analysis.
AUI features
Model display and post-processing
Function for computing the change in a variable between two successive solution times
The new function DELTA_LOAD_STEP is used to obtain the change in a variable between
two successive solution times, as described below.
Suppose that the ADINA solver stores results at times t1 , t2 , ..., ti , ..., tn in the porthole file.
And suppose that the results are requested at time t in the AUI. Then the bracketing times
are determined as in the table on the next page.
Observe that the bracketing times correspond to a backwards difference when the requested
time equals one of the times in the porthole file.
The function DELTA_LOAD_STEP(X) returns the difference in the value of variable X

evaluated at the two bracketing times.
In particular, the function DELTA_LOAD_STEP(TIME) returns the difference of the

difference of the two bracketing times.
Therefore the ratio DELTA_LOAD_STEP(X)/DELTA_LOAD_STEP(TIME) returns the

approximate time rate of change of variable X.
The solution times t1 , t2 , ..., ti , ..., tn need not be equally spaced.
The variable X need not be stored in the porthole file for all of the solution times t1 , t2 , ...,
ti , ..., tn .

Requested time Bracketing times

t  t1 none
t  t1 none
t1  t  t2 t1 , t2
t  t2 t1 , t2
t2  t  t3 t2 , t3
... ...
t  ti ti 1 , ti
ti  t  ti 1 ti , ti 1
t  ti 1 ti , ti 1
... ...
tn 1  t  tn tn 1 , tn
t  tn tn 1 , tn
t  tn none
The variable X cannot be a constant. The variable X cannot be a resultant.
As a simple example, the resultant
RESULTANT STRAIN_RATE,
‘DELTA_LOAD_STEP(<STRAIN-XX>)/DELTA_LOAD_STEP(TIME)’
along with the command
ZONELIST VAR=STRAIN_RATE
can be used to compute the approximate strain rate for all of the solution steps.

Sparse solvers for the DMP versions
The sparse solvers for the DMP versions have been substantially improved in version 9.4.2.
The solution time required and the memory required is reduced. Because the 3D-iterative
solver and subspace iteration solver use the sparse solver, the solution time required by these
solvers is also reduced.
The sparse solvers also make better use of the underlying CPU architecture, so the
improvement in the sparse solvers is greater when using later CPU architectures (for example,

the Ivy Bridge or later architectures).

Improvements to OP2 file creation
It is now possible to specify whether the OP2 file is created after each converged time step in
which the results are saved (MASTER OP2SAVE=STEP), or at the end of the entire solution
after all time steps have converged (MASTER OP2SAVE=END).
If the OP2 file is created after each converged time step, the OP2 file can be loaded during the
solution. However, the solution time might be increased.
By default, the OP2 file is created at the end of the solution.
 Command-line:
MASTER OP2SAVE
Improvement for bolt groups
The element group number corresponding to each bolt number can now be listed in pre- and
post-processing, using the new BOLTINFO command.
 Command-line:
BOLTINFO
AUI features
Model definition
Improvement for defining element face-sets
When defining element face-sets using auto-chaining with the option EXTERNAL=GROUP,
only the element faces of the element groups listed in the table input are considered in the
auto-chain. In previous versions, all element faces were considered.

Updates to documentation
ADINA Handbook
The example on pages 66-67 demonstrating the use of transitional pyramids has been updated
to be consistent with the program’s behavior. The revised example is:
Command input 2.51 demonstrates the use of transitional pyramids between face-
linked mapped hexahedral and free-form tetrahedral meshes.
Command Input 2.51: Meshing two face-linked bodies. Body 1 is map meshed using 8 node bricks, and
body 2 is free- form meshed. Figure 2.50 shows the resulting mesh.
feprogram program=adina
body block name=1 option=centered position=vector,
cx1=0.0 cx2=0.0 cx3=0.0 dx1=1.0 dx2=1.0 dx3=1.0
body block name=2 option=centered position=vector,
cx1=0.0 cx2=1.0 cx3=0.0 dx1=1.0 dx2=1.0 dx3=1.0
*
facelink option=all
*
subdivide body 1 mode=length size=0.1
subdivide body 2 mode=length size=0.1
*
egroup name=1 type=threedsolid
egroup name=2 type=threedsolid
*
gbody 1 group=1 nodes=8
gbody 2 group=2 nodes=4 meshing=free-form pyramids=only
*
boxzone b
-10 0 -10 10 -10 10
modeldepic geometry=no
frame/meshplot zone=b
Figure 2.50 shows the results of Command input 2.51. Body 1 has been map meshed
with hexahedral elements. Body 2 is face linked (see page 14) to body 1 and is
meshed using tetrahedral elements and transitional pyramids.
To generate the red mesh containing mostly tetrahedral elements but transition
pyramids on the interface with the hexahedral mesh, as shown, the user must:
1. Specify NODES = 4 to mesh the body using tetrahedral elements.

2. Invoke free-form mixed meshing using MESHING = FREE-FORM.
3. Specify PYRAMIDS = ONLY.


Figure 2.50: Free-form mesh (in red) generated using tetrahedral elements and
transitional pyramids.
The PYRAMIDS = ONLY option generates transitional elements only on the linked
body faces that have quadrilaterals. Transitional pyramids are required for CFD
analysis to ensure a compatible mesh. In structural analysis, the PYRAMIDS =
ONLY option is not required, and tetrahedral elements can be directly attached to the
hexahedral elements without transitional pyramids.
Available documentation
The following documents are available with the ADINA System. These documents are
modified in this release of the ADINA System as described below.
Installation Notes
Describes the installation of the ADINA System on your computer. Depending on the
platform, the appropriate installation notes in pdf format can be printed or downloaded from
http://www.adina.com.
ADINA Handbook
Written as a task-oriented desktop reference, the ADINA Handbook helps users to quickly
and effectively leverage ADINA's advanced geometric modeling, meshing, and visualization
features.

Available documentation
ADINA User Interface Command Reference Manual (CRM)

Volume I: ADINA Solids & Structures Model Definition
Volume II: ADINA Thermal Model Definition
Volume III: ADINA CFD & FSI Model Definition
Volume IV: ADINA EM Model Definition
Volume V: Display Processing
Updates are made for the new and updated commands.
ADINA Primer
Problem instructions are revised for the ADINA System 9.4. Two new primer problems are
added.
Theory and Modeling Guide (TMG)

Volume I: ADINA Solids & Structures
Volume II: ADINA Thermal
Volume III: ADINA CFD&FSI
Volume IV: ADINA EM
The new features of the solution programs are described.
ADINA Verification Manual
ADINA-Nastran Interface Manual
TRANSOR for I-DEAS Users Guide
TRANSOR for Femap Users Guide

rADINA Elnotes - 94

Uploaded by

Copyright:

Available Formats

rADINA Elnotes - 94

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

rADINA Elnotes - 94

Uploaded by

Copyright:

Available Formats

UTOMATIC

ADINA 9.4.2 July 2018

ADINA R & D, Inc.

ADINA R & D, Inc.

tel. (617) 926-5199

The information contained in this document is subject to change without notice.

ADINA is a registered trademark of K.J. Bathe / ADINA R & D, Inc.

 ADINA R & D, Inc. 2018

New and updated feature summary

Features for all programs

Platform Operating system Fortran compiler

ADINA System 9.4 Release Notes 3

Table of supported platforms, DMP program versions

Platform Operating system Fortran compiler

1) The DMP program versions are based on OpenMPI version 3.0.0.

ADINA Solids & Structures features

4 ADINA System 9.4 Release Notes

New robust frequencies within an interval calculations

Other improvements to frequency calculations

The default number of frequencies to be calculated depends on the type of frequency

Improvements for SMP

Time function scaling of mass and damping matrices

ADINA System 9.4 Release Notes 5

Improvements to lumped mass matrices for 3D solid elements

Improvements to Noh-Bathe method for explicit time integration

Improvements to connector elements

6 ADINA System 9.4 Release Notes

Improvements to element birth-death

One noteworthy change is the following.

Thermo-plastic-cyclic material model

The thermo-plastic-cyclic material model is a new material model in version 9.4.

The thermo-plastic-cyclic material model is a temperature-dependent version of the plastic-

Improvement to thermo-plasticity with kinematic hardening

ADINA System 9.4 Release Notes 7

Improvements to Bergstrom-Boyce material model

Improvements for plasticity, creep and viscoelastic material models

This change has no effect if KINEMATICS STRAINS=(SMALL or LARGE), or EGROUP

8 ADINA System 9.4 Release Notes

Improvements to output of results for large strain plasticity models

Improvements for contact

Nodal sliding distance

In version 9.4, there are three friction delay options:

No No friction delay (the default)

Frictionless Frictionless conditions are assumed in a solution step in which contact is

Stabilized Stabilized frictional conditions are assumed in a solution step in which

ADINA System 9.4 Release Notes 9

Contact slip velocity loading

Improvements for element group based mass-proportional loading

Improvements for solution accuracy indicators

The solution accuracy indicators have been improved, as follows:

A warning message is now printed to the message file if the stiffness-stabilization

Improvements to energy calculations

The following improvements are made to energy calculations:

Energy calculations for 3D shell elements

10 ADINA System 9.4 Release Notes

Energy calculations for iso-beam elements

Improvements for OP2 file creation

Improvements for I-deas universal result file creation

EnSight files are saved to a separate folder in the working directory.