Class 3 Error
SYMPTOM:
The calculation of the depth (characteristic element layer thickness at a GGI interface) is not carried out correctly in a parallel run. This only affects transient rotor-stator and moving mesh cases, where the interface needs to be reintersected within the parallel solver.
PROBLEM:
This is due to the solver only doing the calculation locally on each partition with the final communication step b |
WORKAROUND:<br>No workaround is available.<br><br>FIXED IN:<br>ANSYS CFX Release 12.0. A custom executable for ANSYS CFX 11.0 SP1 is available on request.
|
In Workbench 10.0, there used to be an option for the user to specify a solution folder different from the default. This is not available in Workbench 11.0, where all ANSYS batch files are located in subfolders next to the DSDB file. Is there a workaround at WB11 to specify a different solution folder? |
There is a workaround using a /CWD in a command snippet and then a /COPY in another one to copy the "file.rst" into the default WB folder for further post-processing, but<br>-... |
Editing the legend for results listed under a Solution Combination leads to completely blue or completely red contours. Is there a workaround? |
For version 11.0 and 11.0SP1, this situation can be circumvented by using named legend; for example, consider a model in which the directional deformation in the solution combination is the sum of... |
Is there a JScript command to update the geometry in Simulation using Geometry Parameter Values? |
Yes. Please try the command "DS.Script.doUpdateFromCAD();".
|
We have a large assembly made up of several subcomponents, which were individually created in separate DBs. We assemble the whole thing together by importing the subcomponents, then run the analysis and get the full set of results. The results file ends up being huge, which is cumbersome to move around and use by each subcomponent owner. How can we disassemble the results file into manageable chunks, so that the results from each component can be sent back to the owner for post-processing? |
The RESWRITE command is one possibility. You can select the areas of the model for which you would like a separate results file, then issue the RESWRITE command, making sure to give the new file a... |
What does this warning message mean?
*** WARNING *** SUPPRESSED MESSAGE CP = 464.562 TIME= 09:42:08
Element order list length 4557 does not match number of selected
elements 4632
New order vector being created. |
This message usually means that the selected set of elements has changed, and the solve command has been issued again. You will have to determine if the results are valid or not, depending on... |
A DesignSpace license does not allow 2D edge-to-edge contact in ANSYS Workbench Simulation 11.0. Is this a known defect? |
Yes, Defect 64722 was filed and corrected for version 12.0. Please also see Solution 405826 for information about a related error (Defect 29780).
|
Can you run multiple DSDB files in batch mode? |
Workbench does not have this capability.<br><br>However, the following example may be used as a workaround:<br><br>*IMPORTANT:<br>C:Program filesansys... |
The documentation for the ADRAG command indicates that the pattern of lines to be dragged may have only 2 lines sharing a keypoint. It also says that the pattern must be contiguous. I have patterns that do not conform to this. How can I over come this? |
Here are 2 simple inputs showing how to overcome both:<br><br>*************************<br>! Input dragging mulitple... |
Can I monitor integral quantities on surfaces defined in CFX-Pre as primitive regions rather than boundaries or interfaces? |
<br>Here is how you can do this. When defining an integral expression such as areaInt(Variable)@Locator, CFX-Pre will not give you the choice of selecting the primitive region as Locator,... |
When creating a surface streamline in a multidomain case, my streamlines do not pass through domain interfaces. |
The implementation of surface streamlines with version 11.0 of ANSYS CFX-Post does not extend to multiple domains. However, there is a workaround which can be performed to visualize these surface... |
During installation, the following dialog appears:
Dialog: Windows XP Setup
Please go to the Control Panel to install and configure system components. |
This is caused by missing files in the directory where the Ansys installation was extracted.<br> - Re-download and extract using Winzip.<br> - Check the box that says ¿Use... |
SYMPTOM:
The CFX Solver fails with an error similar to the following, if Efficiency Output is enabled for a calculation involving a fluid defined as a Fixed Composition Mixture:
Error in subroutine FNDVAR :
Error finding variable MASFRC_MT1_FL1
GETVAR originally called by subroutine CAL_ENTH_ISEN
No error is generated for Pure Substances or Variable Composition Mixtures.
PROBLEM:
This is due to a coding error in the CFX Solver. |
WORKAROUND:<br>Redefine your working fluid as a Pure Substance with the same effective properties as the mixture, if possible.<br><br>FIXED IN:<br>ANSYS CFX Release 12.0. A... |
Class 3 Error
SYMPTOM:
Stage Interfaces in combination with Pitch Change = None can be set up in CFX Pre and in the CFX Solver without error. The solution at the interface, however, often exhibits a single, large circumferential band (typically taking up 90% or more of the interface) where several may be expected.
PROBLEM:
This is due to a coding error in the CFX Solver, and a missing check in CFX Pre. The large circumferential band is a symptom of the fact that the CFX Solver is unable to create the bands for zero Pitch Change. Stage Interfaces should not be allowed in combination with no |
WORKAROUND:<br>No workaround is currently available.<br><br>FIXED IN:<br>ANSYS CFX Release 12.0. A custom executable for ANSYS CFX 11.0 SP1 is available on request.
|
I have a problem involving buoyant flow of an ideal gas under atmospheric conditions so that the height at a pressure boundary varies over a significant distance. When I set the pressure at the boundary, how can I specify it so that I don't see any spurious flows due to an inconsistent specification of the opening pressure? |
<br><br>Specifying the pressure at a boundary for a buoyant problem where the height in the gravity potential field varies significantly requires care, particularly for an ideal... |
In TASCFLOW and Fluent there are parameters to limit the solution temperature (T_CLIP_LOW = 273 and T_CLIP_HIGH = 3000 in TASCflow)
Are there similar parameters in CFX? |
<br><br>Below is some CCL that can be used to limit the solution temperature in CFX.<br><br>LIBRARY:<br> VARIABLE: T<br> Component Lower Bounds = 299 # set... |
How do I create non-linear springs in Simulation? |
<br>You can't do this directly but this can be done with the help of command snippets. The database attached to this article contains three models:<br><br>1) All linear... |
Class 3 Error in Beta Feature
SYMPTOM:
If materials defined through RGP-tables are used in multi-zone non-constant physics cases, then, if the reference pressure is different in the different zones, incorrect material properties can be set up as the reference pressure from the first zone will be used in calculating RGP-tables in all zones, rather than the local reference pressure.
PROBLEM:
This is due to the solver incorrectly evaluating a relative pressure for the RGP tables based on a shift |
WORKAROUND:<br>No workaround is available.<br><br>FIXED IN:<br>ANSYS CFX Release 12.0. A custom executable for ANSYS CFX 11.0 SP1 is available on request.
|
Q.
I am trying to understand the meaning of the X, Y and Z forces in the time history plots for the gauges that I have applied to a model in AUTODYN 11.0
. What is the exact meaning of these forces and are they the reaction forces?
A.
The forces X, Y and Z listed in the time history plots are not reaction forces, they are the nodal forces required after update of the stress tensor as part of the Lagrangian computational cycle which result in nodal accelerations and velocities. They are not related to the reaction forces.
If you are interested in the reaction forces you need to instead look at the stresses in the model which you can then convert to forces. The stress that is recorded is STR.AVG.1 so you will need to select this from the time history plot to provide the correct information at the gauge locations. STR.AVG.1 gives the average stress at these locations.
You will need to select this result from the output tab ' history ' select gauge variables first and then re-run the model. |
Q.<br>I am trying to understand the meaning of the X, Y and Z forces in the time history plots for the gauges that I have applied to a model in AUTODYN 11.0<br>. What is the exact... |
I have a large analysis involving many loadcases (in excess of 400). The decomposition stage seems to proceed rapidly (less than 6 hours) but then does not seem to progress very far beyond this. The analysis has been running for 8 days now. Why is it so slow? |
The high-performance incore solver has a certain requirement for memory to undertake the decomposition but may require significantly more memory than this for efficient solution when a large... |
How to implement the NX5 direct Interface for ICEM CFD 11? |
<br>The folder of the direct Interface is in the subfolder of ICEM:<br><br>..v110icemcfdwindifuglibug<br><br>e.g. C:Program FilesANSYS... |
I am using the *MSG command as the last line of my input file to retrieve some information from the analysis, but it is not working. The output windows does not even give an error; it just notes that the input file has been read. What is going on? |
The *MSG command needs to see a blank line to indicate the end of the message. Adding a couple of them to the end of your input file should resolve the issue.
|
I am trying to model bolt thread loading, but I am not sure how to simulate the combined radial and axial loading from the threads. How can I model this? |
There are at least three options:<br><br>(1) Assume the radial component is small compared to the axial component and ignore it (of course, this will depend on the geometry). The bolt... |
I need to calculate the zero-crossing frequency in Simulation. How can I do that? |
The zero-crossing frequency (also called "Statistical Average Frequency" or "Expected Frequency") is computed as the square root of the velocity variance divided by the... |
Class 3 Error
SYMPTOM:
If the turbulence control option is used in the global dynamic model control functionality, the CFX Solver may fail in several ways depending upon the turbulence model. It may give a memory management error like:
Error detected by routine PSHDIR
CDRNAM = FL1 /TURBM/TKE
CRESLT = NONE
Current Directory : /FLOW/PHYSICS/ZN1
However, for the k-epsilon turbulence model (and perhaps others), the Solver does not fail, but gives a zero equation turbulence model solution to the momentum equations, even though the k and epsilon equations themselves are solved, so that it looks as though the solution is progressing well.
PROBLEM:
This is due to an error in the s |
WORKAROUND:<br>Do not use the dynamic turbulence control, which should be unnecessary, or restart after the point of transition.<br><br>FIXED IN:<br>ANSYS CFX Release 12.0.... |