The PLLS command fails to display ETABLE items along 2D axisymmetric shell elements SHELL208. It works properly at v10. It also works fine with SHELL51 at both v10 and v11 (although this element type is undocumented at v11). Listing results do not differ between v10 and v11 for SHELL208. |
The was a defect at 11.0SP1 that has been corrected for 12.0.
|
I would like to create a rigid region of nodes. Is it possible? |
To create a rigid region, you can use CERIG (you could also just make the modulus of those elements really high, but you may run into condition issues there). The attached simple input file... |
I used the command CERIG to create some rigid links, but I am getting the following error message:
Label ROTZ is not an active DOF. The CERI command is ignored.
What does this mean, and what can I do to fix it? |
The CERIG command is trying to form a rigid region controlled from one node, but if this one node has no rotational degree of freedom, then the rigid region could not rotate. The simplest way to... |
I have a group of elements selected, and I want to find a list of unique attributes for these elements; e.g., if there are a bunch of elements with either REAL 3 or 4, I want the list {3,4}. How can I do this? |
There are a few approaches to solving this problem.<br><br><br>--------------------------------------------------<br>Approach 1: Use vectors and vector... |
Are the values reported by "CP time" accurate? |
No, it is not accurate. Even on a single core, different platforms accumulate it differently, so comparing CPU times for different machines may not be valid.<br><br>A more-accurate... |
In TurboGrid State or Session files, how can the position of Control Points be determined, e.g. for use in batch? |
The position of Control Points in TurboGrid State or Session files depends on the method of creating Control Points:<br><br> 1) Relative Definition: Existing (Master) Control Points,... |
When defining the parameter "Frozen Rotor Theta Offset" in a rotor-stator simulation involving Frozen Rotor interface, it seems to be stored in the RES file and applied on restarts from that RES file? |
The parameter "Frozen Rotor Theta Offset" is indeed stored in the RES file. Whenever a restart is performed from this RES file, and no DEF file or modified<br> CCL in RES... |
How to obtain a list of domain names, when using batch session files in CFD-Post? |
The list of domains and the names of single domains can be extracted automatically in batch session files in CFD-Post, using the following commands:<br><br>CFD-Post... |
In ANSYS CFX an error message appears indicating a too small inlet flow angle definiton for an Inlet boundary condition:
Example:
ERROR #002100024 has occurred in subroutine ASS_FLX_BDINIP.
Message:
The specified flow direction on boundaries must not be tangential
to the boundary. However, on the boundary patch Inflow
the specified flow direction is (nearly) tangential. Please do
one of the following:
(1) Check your direction specification on this boundary.
(2) Set the 'Flow Direction Linearisation' to 'Velocity Magnitude'
in your boundary specification.
This error is triggered when the angle between a face normal vector
and the specified direction vector is below a tolerance.
The default tolerance is 20 degrees, but may be modified by
setting the expert parameter 'tangential vector tolerance'. |
From ANSYS CFX 11 to ANSYS CFX 12, the default threshold value for the inlet flow angle has been changed from 5 to 20 degrees measured against the circumferential... |
During a multi-body dynamic analysis, I sometimes get the following message:
x JOINT ELEMENTS WITH LAG MULT OPTION DO NOT SATISFY JOINT CONSTRAINT CONDITIONS
What does this mean? |
The "LAG MULT" is short for "Lagrange Multiplier," which is the method used to impose the kinematic constraints in the joint elements. When trying to find a solution,... |
How to monitor the total mass of fluid in a domain. |
If it is required to monitor the total mass of a particular fluid present in a<br>domain (e.g. in a multiphase simulation) in CFX-11.0, this can easily be accomplished using a monitor point... |
Is it possible to print or plot the elastic strain energy density (SEND, ELASTIC) for a hyperelastic material? |
At ANSYS 11.0, it is not possible to extract SEND directly for hyperelastic materials. You must extract the element strain energies (SENE) to an element table first, then extract the element... |
Class 3 Error
SYMPTOM:
If an additional variable is set equal to a transient statistic (for example Temperature.trnmax), and subsequently the probe function is used to evaluate the additional variable at a point, then it is possible that the transient statistic function will be ignored, and the probe function will just give the value of the variable (in this case temperature).
PROBLEM:
This is due to the function operator not being passed to the routine evaluati |
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.
|
Q: I am an academic customer running CFX 11 SP1 on a 64-bit Linux cluster. I have a license for the Academic Research product.
When trying to run the CFX solver in parallel as part of an FSI problem the solver fails with the following error message:
====================
+--------------------------------------------------------------------+
| Job Information |
+--------------------------------------------------------------------+
Run mode: parallel run (MPI)
Host computer: n31
Par. Process: Master running on mesh partition: 1
Job started: Mon Dec 15 11:36:34 2008
Host computer: n31
Par. Process: Slave running on mesh partition: 2
Job started: Mon Dec 15 11:36:34 2008
Host computer: n31
Par. Process: Slave running on mesh partition: 3
Job started: Mon Dec 15 11:36:34 2008
Connecting to the following master process:
Host Name : n31
Port Number : 44958
No license for ANSYS CFX Solver Parallel could be found.
+--------------------------------------------------------------------+
| ERROR #001100247 has occurred in subroutine . |
| Message: |
||
| The solver is unableto continue because of licensing problems. |
| |
| A license for the following capability level could not be checked |
| out: |
| |
| PARALLEL PROCESSING |
| |
| Please carefully examine the error message output above and check |
| that: |
| |
| 1) LM_LICENSE_FILE p |
Q: I am an academic customer running CFX 11 SP1 on a 64-bit Linux cluster. I have a license for the Academic Research product.<br><br>When trying to run the CFX solver in parallel as... |
Class 3 Error
SYMPTOM:
The solver writes wrong values of the acoustic monople sources written to the cgns file.
PROBLEM:
The wall normal velocity values, used to write the acoustic monopole sources, are wrongly calculated due to a memory initializ |
WORKAROUND:<br>No workaround is currently available.<br><br><br>FIXED IN:<br>ANSYS CFX Release 12.0.
|
Q) Bilinear CZM for use with contact elements is introduced at ANSYS 11.0 (TB,CZM,,,,CBDD/CBDE). What does the C5 "artificial damping" parameter mean, and how to does it work?
A) Equation 4-361 of the ANSYS 11.0 Theory Reference shows the usage of artificial damping. Note that a minus sign is missing from the 11.0 documentation.
Using the alternate notation of P=Tn (normal traction), Equation 4-361 is the following:
P = P_final + (P_initial - P_final)*exp(t/eta)
where P is the contact pressure (normal traction) for Mode I case, t is the time interval, and eta is the C5 artificial damping parameter. P_final represents the final traction at the end of the load step (e.g., 0 for 100% damage) and P_initial is the initial traction at the beginning of the load step. A similar implementation is used for the tangential term.
ANSYS does not know what P_final may be, so Equation 4-361 is not used as-is but implemented through use of a numerically discretized approximation; the developer noted that the actual implementation is not published.
What artificial damping essentially does is to augment the bilinear traction-separation CZM curve with an exponential form. The exponential form makes the bilinear 'tip' a little more blunt, which helps in convergence since the stiffness doesn't suddenly change (going from positive to negative slope). While this form is not meant to replicate the exponential CZM model used with 20x interface elements (TB,CZM,,,,EXPO), it provides a similar effect in that the bilinear CZM curve is discontinuous whereas having a 'smoother' tip (via the 'eta' parameter) aids in convergence.
Since it is implemented through numerical discretization, the actual result is sensitive not only to the 'eta' value (which should be a small value) but also sensitive to timestep size and mesh size. |
Q) Bilinear CZM for use with contact elements is introduced at ANSYS 11.0 (TB,CZM,,,,CBDD/CBDE). What does the C5 "artificial damping" parameter mean, and how to does it... |
Class 3 Error
SYMPTOM:
The CFX Solver does not support mixing of radiation models across a solid-fluid interface with a "conservative interface flux" condition for thermal radiation. However, for example, if a case is set up with the discrete transfer model in the fluid and Monte Carlo in the solid, then neither CFX-Pre nor the CFX Solver gives any warning, but the Solver continues in an incorrect manner using the discrete transfer model everywhere, even in the solid where it should be disallowed.
PROBLEM:
This is due to poor checking of model |
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.
|
KNOWN ISSUE: Some graphics objects do not display correctly in CFX-Pre and CFD-Post |
SYMPTOM:<br>Some graphics objects may not display correctly in CFX-Pre and CFD-Post. Examples of the problems could include incorrect colors being used (e.g. for boundary markers in CFX-Pre,... |
ANSYS Workbench 11.0 sp1 crashes at startup with the following error:
*** AnsysWB Module ***
AnsysWB.exe has encountered a problem and needs to close.
We are sorry for the inconvenience.
If the customer clicks the "Click here" link, error information is shown:
*** ANSYSWB Module ***
AppName: ansyswbu.exe AppVer: 11.2007.10.1 ModName: sdkdllu.dll
ModVer: 11.2007.262.1 Offset: 0000000000003274 |
Please try the following:<br><br>- Start > Run<br>- Type: regedit<br>- Now browse to the following registry key:<br> ... |
Are there any ANSYS Workbench 11.0 examples of modeling a press-fit? |
Yes, please see the attached ANSYS Workbench Simulation 11.0 SP1 example. A solid shaft overlaps a hollow cylinder. The interference is ramped off in the first load step and an additional axial... |
Customer gets the following errors when installing the ANSYS Workbench 11.0 SP1 for UG NX6 service pack from the Customer Portal on a Windows Vista machine:
The module "C:Program FilesANSYS Incv110AISOLCad IntegrationUnigraphicsNX6IntelDSPlugInUG6U.dll" failed to load.
Make sure the binary is stored at the specified path or debug it to check for problems with the binary or dependent .DLL files.
The specified module could not be found.
The module "C:Program FilesANSYS Incv110AISOLCad IntegrationUnigraphicsNX6IntelAGPSUtilU.dll" failed to load.
Make sure the binary is stored at the specified path or debug it to check for problems with the binary or dependent .DLL files.
The specified module could not be found.
Why are the plugins not registering? |
This is happening because, for some reason, the UnigraphicsNX6 folder is being placed in an incorrect location when installing the service pack on Vista machines.<br><br>Create a... |
In Workbench Simulation, there are three Static Analyses under the same Model. Each Static Analysis has different load cases and has defined response parameters for each of them. In the Parameter view of DesignXplorer, those response parameters are listed as follows:
Total Deformation Maximum
Equivalent Stress MaximumTotal Deformation Maximum
Equivalent Stress Maximum
Total Deformation Maximum
Equivalent Stress Maximum
Each parameter has the same name for the same output type, and it is not convenient to distinguish them. Is there any way to rename the name of Response Parameters in DesignXplorer? |
The parameters need to be renamed within Simulation. When in DX, it is too late to change them.
|
I use FLUENT and am interested in optimization of CFD analysis with DesignXplorer at release 12.0. Chemical reaction equations are defined in my FLUENT model, and there are 12 input parameters. If I could use the default DOE using Central Design composite, the number of automatic design point would be 281. This makes optimization analysis quite difficult because one solution takes 10 hours. modeFRONTIER has a capability to estimate sensitivities of input parameters at first, then dropping off low sensitivity parameters in order to reduce the design points. From what I understand, DesignXplorer does not have this capability. Is it right? Is there any good strategies to do optimization analysis for his case (large solve with many parameters)? |
At 12.0, we will have a similar capability called Correlation Matrix; however, this will still require some calculations to be performed to determine the important parameters. Another way to solve... |
What is the default value of the "Linear Anisotropy Coefficient" in the P1 Radiation model formulation? Can it be changed by the user? CFX-Pre does not allow it. |
The default value is 0. It can be changed via CCL.<br><br><br> THERMAL RADIATION MODEL:<br> Option = P 1<br> SCATTERING MODEL:<br> ... |
Are there any recommended values for the Anisotropy Coefficient for the Scattering Model in coal combustion? |
Although the literature suggests that there is a large amount of scattering from coal and ash particles, most of this scattering is in fact close to the forward direction (i.e. it is very... |