— For ADINA Users Only —

This is a restricted area meant for current ADINA users only. If your license does not include current support of ADINA, please kindly exit this page.

ADINA 9.3.4 has been released.

These Useful Hints and Issues, while reported for the specific version 9.3.3, are usually also applicable to ADINA versions 9.3.2, 9.3.1 and 9.3.0.

Issue No. 46, posted February 22, 2018

ADINA Structures crashes in a nonlinear analysis when the GPFORCE result is requested for the op2 output, and the ELEMSAVE-STEPS or NODESAVE-STEPS command is used to save element or nodal results at certain steps.

As a workaround, save the node and element results at each time step, i.e., do not specify the ELEMSAVE-STEPS or NODESAVE-STEPS command. This bug is fixed in v.9.3.5.


Issue No. 45, posted February 22, 2018

The AUI fails to generate a .dat file when there are connector elements with the a-triad defined using an auxiliary point, i.e., TRIAD1OPT=AUXPT or TRIAD2OPT-AUXPT in the CONN-DEFINE command.

As a workaround, use the ORIENTATION option with an orientation vector not parallel to the element's axial direction. This bug is fixed in v.9.3.5.

Issue No. 44, posted February 22, 2018

When group-based mass-proportional loading is used in implicit dynamics with a consistent mass matrix, the results are obviously incorrect.

As a workaround, use model-based mass proportional loading. This bug is fixed in v.9.3.5.

Issue No. 43, posted February 22, 2018

ADINA Structures crashes under the following conditions:

  1. op2 file output is requested, and
  2. There is an element group of 2D solid elements with varying thicknesses, and
  3. The number of nodes per element is different for different elements in the 2D solid element group with varying thicknesses.

As a workaround, put the 2D solid elements into different groups so that in each group, the number of nodes is the same in each element and the nodes in each element are in the same r, s positions. This bug is fixed in v.9.3.5.

Issue No. 42, posted January 3, 2018

In frictional contact analysis, while the computations are correct, the outputs of the slip velocity directions have the wrong sign under the following conditions:

  1. Explicit contact analysis, or
  2. Implicit contact analysis when the friction algorithm selected by CONTACT-CONTROL FRICTION-ALGORITHM=V83 (not the default) is used.
There is no workaround. This bug is fixed in v.9.3.5.

Issue No. 41, posted December 27, 2017

The ADINA program crashes if bolt damping is used in static analysis and gravity or centrifugal loads that are calculated using a consistent mass matrix are applied in the model.

As a workaround, use a lumped mass matrix for gravity or centrifugal loads. This bug is fixed in v.9.3.5.

Issue No. 40, posted December 27, 2017

If a linear convection load is present in a model, solutions for radiosity give incorrect temperature results.

There is no workaround. This bug is fixed in v.9.3.5.

Issue No. 39, posted December 27, 2017

The ADINA program crashes when running a large model with mesh glueing, and the GPFORCE result is requested in the .op2 file output.

There is no workaround. This bug is fixed in v.9.3.5.

Issue No. 38, posted December 27, 2017

The porthole file cannot be loaded from an ADINA Structures model when the following combination of options is used:

  1. Connector element, with death time specified, and
  2. Consistent nodal forces are requested (EGROUP CONNECTOR RESULTS=FORCE or ALL).
In this case, the AUI gives the error message

ADINA porthole file mismatch, expected record "CURTRIAD", found record "OUTPUT19".

As a workaround, do not request consistent nodal forces. The bug is fixed in v.9.3.5.

Issue No. 37, posted December 27, 2017

The porthole file cannot be loaded from an ADINA Structures model when the following combination of options is used:

  1. A 3D element group in which both 4-node tetrahedral and 10-node tetrahedral elements are present, and
  2. The integration order defined by the EGROUP command is default (EGROUP THREEDSOLID TETINT =DEFAULT), and
  3. The option to save selected 2D and 3D element results is not used because PORTHOLE RESULTS-ELEMENT=NO (not the default).
In this case, the AUI gives the error message

Unexpected data in ADINA porthole file, record OUTPUT-3

As a workaround, use PORTHOLE RESULTS-ELEMENT=YES (the default). This bug is fixed in v.9.3.5.

Useful Hints and Issues
Regarding the Use of the ADINA System, v. 9.3.3

Issue No. 36, posted October 20, 2017

The AUI permits generation of .dat files when no action direction is defined for spring elements with initially coincident nodes and EGROUP SPRING NONLINEAR=GEOM. When lacking an action direction, such an element will not generate forces. Thus, the AUI should instead block the .dat file generation in this instance.

As a workaround, always use the EDATA command to define the action direction for spring elements with initially coincident nodes and EGROUP SPRING NONLINEAR=GEOM. This bug is fixed in v.9.3.4.

Issue No. 35, posted October 20, 2017

The .dat file generated by the AUI is wrong when a skew system is applied to a 6DOF spring and EGROUP SPRING SKEWSYSTEM=YES. When attempting to run the .dat file, the program will abort. The element group option SKEWSYSTEM=YES should have no effect on 6DOF springs.

Furthermore, the AUI permits generation of .dat files with a 6DOF spring orientation vector parallel to the element's local x-direction. In this instance, the local system is not uniquely defined, and the AUI should instead issue an error.

As a workaround, make sure that EGROUP SPRING SKEWSYSTEM=NO when defining a 6DOF spring element group and always ensure that the orientation vector is not parallel to the element's local x-direction. These bugs are fixed in v.9.3.4.

Issue No. 34, posted October 20, 2017

When time step blocks are used in iterative or one-way TMC heat flow solutions using the Bathe method, the porthole cannot be loaded into the AUI.

As a workaround, used the Euler backward method. This bug is fixed in v.9.3.4.

 

Useful Hints and Issues
Regarding the Use of the ADINA System, v. 9.3.2

Issue No. 33, posted October 20, 2017

The porthole file cannot be loaded in the AUI for some contact analysis models if both op2 and porthole outputs are requested. The loading stops and results cannot be displayed.

There is no workaround. This bug is fixed in v.9.3.3.

Issue No. 32, posted September 1, 2017

In some cases, a gluemesh subjected to rigid body motions causes deformations and stresses in the elements to which the gluemesh is attached.

As a workaround, make sure that each segment in the slave gluemesh surface lies entirely within the master gluemesh surface. This bug is fixed in v.9.3.3.

Issue No. 31, posted September 1, 2017

There is an error in the output of results for the 2D solid element when the orthotropic material with wrinkling is used and stresses are output in the material coordinate system. In this case, the results reported by the AUI are transposed, for example, the results output for STRESS-AA are actually the results for STRESS-CC.

As a workaround, during model definition, turn off the option to save selected element results. In the AUI menu: click Control > Porthole (.por) > Volume... and uncheck the option "Save Only Selected Results for 2D and 3D Solid Elements". The equivalent command is

PORTHOLE RESULTS-ELEMENT=NO

This bug is fixed in v.9.3.3.

Issue No. 30, posted September 1, 2017

When the option of full Newton iteration is used in mode superposition analysis and the model has linear and nonlinear element groups, incorrect results might be obtained.

As a workaround, make all element groups nonlinear (e.g., by specifying a very large death time). This bug is fixed in v.9.3.3.

Issue No. 29, posted September 1, 2017

For 2D/3D solid and shell elements with the plastic cyclic material model, in certain loading cases when the strain state goes from plastic to elastic, ADINA gives wrong strain energy results: the plastic strain energy drops to zero while the elastic strain energy jumps to a value equal to the total strain energy. The correct results are that the plastic strain energy keeps constant while the elastic strain energy equals the total strain energy minus the plastic strain energy.

The workaround is to use other plastic material models than the plastic cyclic material if the energy result is requested. This bug is fixed in v.9.3.3.

 

Useful Hints and Issues
Regarding the Use of the ADINA System, v. 9.3.1

Issue No. 28, posted July 12, 2017

When a group of bodies is subtracted from a body with the option KEEP-IMPRINT=YES, the imprints are not kept.

As a workaround, subtract the bodies one at a time. This bug is fixed in v.9.3.2.

Issue No. 27, posted July 12, 2017

For TMC analysis with the ADINA Structures program, when thermal mapping is used, (e.g., the zooming solution option), initial temperatures are not included in the second (zoomed) run, therefore the calculated thermal stresses are incorrect.

There is no workaround. This bug is fixed in v.9.3.2.

Issue No. 26, posted July 12, 2017

Concentrated masses applied on constrained rotational DOFs are not included in reaction calculations, therefore reactions are incorrect in such cases.

There is no workaround. This bug is fixed in v.9.3.2.

Issue No. 25, posted July 12, 2017

In a Steered Adaptive Meshing (SAM) model where porthole files from several SAM remeshings are loaded into the same database during post-processing, if a relatively large number of elements are picked to define a new zone, selected elements may be missed after the definition is confirmed.

There is no workaround. This bug is fixed in v.9.3.2.

Issue No. 24, posted July 12, 2017

In v. 9.3.0, a new contact traction algorithm is implemented which leads to much improved contact tractions for symmetric contact. This same algorithm is also used for self-contact, leading to much improved contact tractions for self-contact.

However, the possibility of a contact surface being used in both symmetric contact and self-contact was not considered in 9.3.0. Therefore the contact tractions of a contact surface used in both symmetric contact and self contact are not improved.

The contact traction algorithm is fixed in 9.3.2, so that the contact tractions of a contact surface used in both symmetric contact and self contact are improved.

As a workaround, make sure that contact surfaces are not used both in symmetric contact and in self-contact.

Issue No. 23, posted May 24, 2017

When only the contact tractions, and not the contact forces, are requested for output in contact groups, then the contact group does not contribute to the reaction calculations. For example, if CGROUP CONTACT3 ... TRACTIONS=YES FORCES=NO, then the contact group does not contribute to the reaction calculations.

Notice that the default is that both contact forces and contact tractions are output, so, by default, the contact group contributes to the reaction calculations.

As a workaround, request both the contact forces and the contact tractions. For example, set CGROUP CONTACT3 ... TRACTIONS=YES FORCES=YES.

This bug is fixed in v. 9.3.2.

Issue No. 22, posted May 24, 2017

In dynamic explicit analysis, if the Noh-Bathe time integration is selected, the penalty contact algorithm is used by default. Hence, the penalty algorithm stiffness parameters in the contact group dialog box should be active but they are not.

As a workaround, select the penalty algorithm instead of the default algorithm in the contact group dialog box. This bug is fixed in v. 9.3.2.

Issue No. 21, posted May 24, 2017

The results from compliant contact in 2D contact surfaces are incorrect when no surfaces in the contact group are attached to 2D solid elements.

The reasons that the results are incorrect in this case are

  1. When there are no surfaces in the contact group attached to 2D solid elements, the AUI assumes that the contact group is axisymmetric, and
  2. The type of 2D contact group (axisymmetric or planar) is used in the determination of the segment contact areas, and
  3. The contact areas are used to determine the contact compliance when the compliance feature is used.

As a workaround, make sure that at least one contact surface in the group is attached to 2D solid elements.

This bug is fixed in v. 9.3.2. In v. 9.3.2, when there are no surfaces in the contact group attached to 2D solid elements, the AUI assumes that the contact group is planar.

Issue No. 20, posted May 24, 2017

In dynamic explicit analysis with 2D contact, some of the information in the printout of contact status information is incorrect, as follows. Under the header

C O N T A C T    S T A T U S    O U T P U T    F O R    2 D    C O N T A C T

the contactor node numbers are wrong. All other information is correct.

There is no workaround. This bug is fixed in v. 9.3.2.

Issue No. 19, posted May 24, 2017

Mode shapes in the Nastran OP2 output with skew systems are wrong. Mode shapes are calculated in the skew systems and they should be converted into the global coordinate system for the OP2 output, but they are not.

There is no workaround. This bug is fixed in v. 9.3.2.

Issue No. 18, posted May 24, 2017

In ADINA Structures with radiosity calculations, the reactions are wrong when direct thermo-mechanical coupling is used.

As a workaround, use iterative or one-way thermo-mechanical coupling. This bug is fixed in v. 9.4. In v. 9.3.2, this option is blocked.

 

Useful Hints and Issues
Regarding the Use of the ADINA System, v. 9.3.0

Issue No. 17, posted May 24, 2017

When importing an IGES file into the AUI, if the x coordinate is set to 0, an input error results.

There is no workaround. This bug is fixed in v. 9.3.1.

Issue No. 16, posted May 24, 2017

In some problems involving connector element death, the porthole file cannot be fully loaded for post-processing in the AUI.

ADINA can create porthole files with element information (e.g., triad orientations) for connector elements which are inactive. When the AUI encounters information for elements which it expects to be inactive, it stops reading the porthole file.

As a workaround, rather than using element death, use a time function to scale the element's properties to zero. This bug is fixed in v. 9.3.1.

Issue No. 15, posted May 24, 2017

When the Lubby2 creep material model is used with the shell element, the ADINA program stops with an error message that states that the shell thickness is negative.

The Lubby2 creep material model is not supported in the shell element.

The workaround is to not use the Lubby2 creep material model with the shell element.

This bug is fixed in v. 9.3.1. The AUI gives an error message when the user uses the Lubby2 creep material model with the shell element.

Issue No. 14, posted May 24, 2017

When using constants in resultants in harmonic or random vibration analysis, the results are wrong.

Consider the following command sequence:

RESPONSE RANDOM RAND ...
CONSTANT CONST 1.0
RESULTANT DISP1 '<Y-DISPLACEMENT>*1.0'
RESULTANT DISP2 '<Y-DISPLACEMENT>*CONST'
ZONELIST RESPOPTION=RESPONSE RESPONSE=RAND,
                  VAR=Y-DISPLACEMENT DISP1 DISP2

The three columns of the listing should give the same result, but do not. Y-DISPLACEMENT and DISP1 give the same correct result, but DISP2 gives wrong results.

DISP2 uses a resultant in which a constant is part of the resultant definition.

As a workaround, hard-code any constant that is part of a resultant definition, if the resultant is used in harmonic or random vibration analysis. For example, DISP1 is a resultant in which all constants are hard-coded.

This bug is fixed in v. 9.3.1.

Issue No. 13, posted April 18, 2017

Results might be incorrect in an implicit analysis that is switched from an explicit analysis using the analysis switch feature (AUI command ANALYSIS-SWITCH).

There is no workaround. This bug is fixed in v. 9.3.1.

Issue No. 12, posted April 18, 2017

The ADINA Structures program stops if a bolt damping factor is specified in a frequency restart analysis.

As a workaround, remove the bolt damping factor for the frequency analysis. This bug is fixed in v. 9.3.1.

Issue No. 11, posted April 18, 2017

The plastic strain energy results are wrong for beam elements during the unloading process from a plastic state. The bug causes the plastic strain energy to become zero, and the elastic strain energy then makes up the total strain energy. The plastic strain energy should instead be constant during the unloading process and the sum of the elastic strain energy and plastic strain energy should equal the total strain energy.

There is no workaround. This bug is fixed in v. 9.3.1.

Issue No. 10, posted April 18, 2017

In explicit analysis, if element group-based mass-proportional, element group-based centrifugal loads or rotational loads (either model-based or element group-based) are defined, wrong results will be obtained if the DTMIN2 feature is used to remove elements with a very small time step.

One workaround is to use model-based mass-proportional or model-based centrifugal loads. Another workaround is to not use the DTMIN2 feature to remove elements. This issue will be fixed in a future release.

Issue No. 9, posted April 18, 2017

For the connector element, if both the local results and the consistent nodal point forces are requested (EGROUP CONNECTOR RESULTS=ALL), then the local results cannot be postprocessed using the ELPOINT and POINTLIST commands.

As a workaround, save only the local results, and do not save the consistent nodal point forces. This bug is fixed in v. 9.3.1.

Issue No. 8, posted April 3, 2017

For TMC analysis in ADINA Structures with low-speed dynamics, when iterative TMC coupling is used, the damping matrix is not properly updated after ATS cutbacks. This may result in incorrect results.

There is no workaround. This bug is fixed in v. 9.3.1.

Issue No. 7, posted April 3, 2017

In ADINA Structures, damping contributions to reactions from linear damping elements are neglected.

There is no workaround. This bug is fixed in v. 9.3.1.

Issue No. 6, posted April 3, 2017

In model-based centrifugal loadings with concentrated masses, concentrated masses are not included in load calculations for solution steps larger than one if the element birth or death option is used.

There is no workaround. This bug is fixed in v. 9.3.1.

Issue No. 5, posted April 3, 2017

The temperature plot in the AUI may not be correct at the restart time on FSI boundaries. The temperatures are calculated correctly in the solution, but are incorrectly averaged in the plots in the AUI.

There is no workaround. This bug is fixed in v. 9.3.1.

Issue No. 4, posted April 3, 2017

The ADINA Structures program may crash when a model uses multilayer shell elements with the data fitted concrete material model.

There is no workaround. This bug is fixed in v. 9.3.1.

Issue No. 3, posted April 3, 2017

In ADINA Structures, the temperature effect scaling the hyperelastic response in the three-network model, given in equations (3.17-30) of the Theory and Modeling Guide, is not included in the program.

There is no workaround. This bug is fixed in v. 9.3.1.

Issue No. 2, posted March 7, 2017

In ADINA Structures, a coding error causes the temperature-dependent three-network model stress update to converge slowly when the temperature response exponent (parameter TEMPN) is large.

As a workaround, increase the maximum number of stress update iterations or loosen the stress update tolerance. This bug is fixed in v. 9.3.1.

Issue No. 1, posted March 7, 2017

In ADINA Structures, the contact tractions may be incorrect for corner nodes of higher-order 3D contact segments. Namely, the normal contact traction is zero but the tangential contact tractions are non-zero. This error occurs for a corner node of a higher-order segment which is not in contact, but for which the adjacent nodes of at least one segment are in contact.

There is no workaround. This bug is fixed in v. 9.3.1.



For Useful Hints and Issues regarding the use of earlier versions of the ADINA System, see here

Version 9.2       Version 9.1       Version 9.0       Version 8.9                         
Version 8.8       Version 8.7       Version 8.6       Version 8.5       Version 8.4