Must the mass of a MASS21 element always be positive? I do not want to readjust the original model after design alterations; I just want to reduce the mass in certain locations. Will a modal analysis with MASS21 elements which have negative mass, produce reliable results? |
There should be no problem for the solver as long as the total mass is positive.
|
Can solid272 match the first bending mode of pipe16? |
No, because it is similar to solid185 without enhanced strain<br>(too stiff in bending.)<br><br>For a bending application use solid273, which will match the first<br>pipe16... |
I setup a modal analysis of a cantilevered pipe using general axisymmetric elements (SOLID272), but it is only computing what would be called "Mode 0" modes, in axisymmetric-harmonic terminology. How can I get the element to produce a cantilever bending mode? |
Check to see if your KEYOPT(2) is =1, with which the element assumes axisymmetric behaviour. All you have to do is to simply increase the KEYOPT(2) to something >=3 and <=12.
|
Class 3 Error
SYMPTOM:
If a case that uses the Turbomachinery wizard in CFX-Pre has the mesh updated with a different number of blades, that is the circumferential angle of the simulation is changed, then CFX-Pre does not automatically update parameters that depend upon this information, such as the mass flow per component. Moreover, if the case is also updated in CFX-Post, then the report template will not allow for the new number of blades
PROBLEM:
This is caused by the software not being designed to allow th |
WORKAROUND:<br>The workaround is to not update the software in this way, but to repeat the Turbomachinery wizard when the number of blades is changed, and ensure that CFX-Post is loaded... |
What level of CATIA V5 licenses are required for the operation of the CADNexus/Capri gateway? |
The following CATIA V5 licenses levels have been confirmed to work correctly<br><br> ME2 Mechanical Engineering 2 Configuration<br> MD2 ... |
My MSUP (Mode Superposition) harmonic response analysis gives different results than my full harmonic response analysis. Why? |
The most likely problem is due to using an insufficient number of modes being used for the MSUP harmonic response analysis. Please see the attached example which compares displacement results for... |
How can I do a one way FSI problem with Fluent and ANSYS Classic if I have a Multiphysics license.
**** Entered By: rlange @ 01/28/2009 11:52 AM **** |
Q. How can I perform a one way FSI problem with Fluent and ANSYS Classic if I have a Multiphysics license.<br><br>A. Solve the Fluent problem and export the loads on the FSI surface... |
Class 3 Error
SYMPTOM:
The calculated values of user-defined wall functions for additional variables and mass fractions are wrong.
PROBLEM:
This is due to a coding bug i |
WORKAROUND:<br>No workaround is currently available.<br><br>FIXED IN:<br>ANSYS CFX Release 12.0.
|
Class 3 Error
SYMPTOM:
The solver may give wrong answer in parallel when using the following CEL functions:
inside()@subdomain
inside()@REGION:region (where region includes the subdomain)
PROBLEM:
If a partition boundary lies along a portion of a subdomain boundary, such that a node on the subdomain boundary is an overlap node on a partition which doesn't have any elements in the partition boundary, the CEL function is set to 1 for the overlap node. For example, if an algebraic Additional variable is set using an inside function, it will not take the correct values at the o |
WORKAROUND:<br>No workaround is currently available.<br><br>FIXED IN:<br>ANSYS CFX Release 12.0.
|
I have developed some turbine blade models consisting entirely of SHELL63 elements that produced modal frequencies that agreed well with legacy NASTRAN models. The models included both spin loads and gas loads, utilizing standard nonlinear prestress modal analysis methods. At v11, the same models are showing frequencies that are up to about 0.5% different than the v10 frequencies. I also ran a very refined tet model and produced frequencies very close to the v10 results, so I am confident in these. What has changed with regards to these models between v10 and v11? |
As part of a code enhancement, and to be consistent with Section 3.5 of the Theory Manual, the code was changed so that spin softening effects would be automatically included in stiffness... |
I would like to obtain a path plot at the outer surfaces and midplane of a shell. How can I do this? |
The SHELL command can be issued before the PDEF command to specify TOP, MID, or BOT of the shell. Please see the example... |
Class 3 Error
SYMPTOM:
For some calculations (typically those involving interphase mass transfer), where the material properties are taken from the materials library <CFXROOT>/etc/materials-extra/MATERIALS-ipmt.ccl, the liquid density used by the CFX Solver is incorrect.
PROBLEM:
Some incorrect values for density are present in the material properties file MATERIALS-ipmt.ccl. In some cases the values are significantly wrong and the error should be obvious, but for other materials the values are only wrong by a small percentage and the error will be less noticeable. A full list of the affected materials is:
COl
CO2l
C2H2l
C2H4l
Ethanol C2H6Ol
C3H6Ol
C3H8l
C4H1Ol
C5H12l
C6H6l
C6H12l
C6H14l
C7H8l
C8H18l
C9H2Ol
H2Ol
NH3l
N |
WORKAROUND:<br>Correct the liquid density in CFX-Pre before running the CFX Solver.<br><br>FIXED IN:<br>ANSYS CFX Release 12.0.
|
Class 3 Error
SYMPTOM:
For calculations involving the binary mixture C16H34vl (Hexadecane), liquid evaporation rates obtained by the CFX Solver are not correct.
PROBLEM:
This is due to incorrect entries for the Antoine Coefficients in the materials libraries (MATERIALS and MATERIALS-ipmt.ccl). The entries should be as follows:
Antoine Enthalpic Coefficient B = 1830.510 [K]*ln(10)
Antoine Pressure Scale = 1 [bar]
Antoine Reference State Constant A = 4.15357*ln(10)
Antoine Temperature Offset C = (154.4 |
WORKAROUND:<br>Correct the Antoine Coefficients in CFX-Pre for this material before running the CFX Solver.<br><br>FIXED IN:<br>ANSYS CFX Release 12.0.
|
Class 3 Error
SYMPTOM:
This problem will affect solutions which have buoyancy turbulence activated with a turbulence model which uses the omega equation. One observed symptom of this problem is that the eddy viscosity may be too large in an unstable stratified mixing layer. In other situations the effect may not be obvious.
PROBLEM:
The wrong coefficient is used in constructing the buoyancy source term in the omega equation. This affects all turbulence models which use an omega equation (k-omega |
WORKAROUND:<br>No workaround is currently available.<br><br>FIXED IN:<br>ANSYS CFX Release 12.0.
|
Class 3 Error
SYMPTOM:
When an epsilon-based Reynolds stress turbulence model is used for modelling certain flows, such as that past a blunt body, it is occasionally possible for the turbulence to disappear entirely, and for the CFX Solver to predict an unphysical laminar solution. This will often be noticeable.
PROBLEM:
This is due to the parameter ustar, in the turbulent wall function, vanishing when the turbulent kinetic energy i |
WORKAROUND:<br>No workaround is currently available.<br><br>FIXED IN:<br>ANSYS CFX Release 12.0. The fix is achieved by a combination of documentation and the inclusion of... |
Class 3 Error
SYMPTOM:
If a particle transport calculation is undertaken with transient rotor-stator interfaces, then incorrect behaviour will be predicted in two circumstances: firstly, if the speed of rotation is varying; or secondly, if a restart is carried out with the time reset.
PROBLEM:
This is because the variable that calculates the difference in circumferential position of the particles at the interface assumes that the rotation speed is constant, and that time has no |
WORKAROUND:<br>No workaround is currently available.<br><br>FIXED IN:<br>ANSYS CFX Release 12.0.1.
|
In DesignModeler, I have a model in which a beam orientation is defined. When transferred to Simulation, the mesh indicates a different orientation. When read into ANSYS, the beam orientations are given as in DesignModeler. The line bodies are defined by edges of surface bodies. What is wrong? |
This was a code defect (#78860) that has been corrected at 12.0.
|
I want to model a structure having very thin plates that join thicker solid structures. If I model the thin regions with shell elements and the thick ones with solids, how do I effect the moment transfer where they join? |
One way to effect moment transfer between the shells (that have rotational degrees of freedom) and solids (which do not) is to use contact elements (see section 9.2: "Modeling a Shell-Solid... |
Class 3 Error
SYMPTOM:
When either of the ray-tracing radiation models in CFX (Discrete Transfer or Monte Carlo) is used in conjunction with any moving mesh /sliding mesh feature (e.g. transient rotor stator), then the solution may be incorrect if any non-overlap region changes position or extent during the course of the calculation.
PROBLEM:
This is due to the fact that the ray-tracing non-overlap data structures in the solver are not updated to account for the change in non-ove |
WORKAROUND:<br>No workaround is currently available.<br><br>FIXED IN:<br>ANSYS CFX Release 12.0.1.
|
Class 3 Error
SYMPTOM:
When using the Schmehl secondary break-up model within the particle transport model, child droplet sizes in the shear break-up regime at high Weber number are incorrect. This will consequently lead to wrong particle size distributions and wrong penetration characteristics.
PROBLEM:
This is due to an incorrect too large calculation of the critical particle diameter, which is subsequently used to calculate other variables |
WORKAROUND:<br>No workaround is currently available.<br><br>FIXED IN:<br>ANSYS CFX Release 12.0.1.
|
Manually configuring CAD interfaces in 12.0 with CAD Configuration Manager In previous releases, one could configure the CAD interfaces by copying files from the installation image, and manually registering the DLLs. Is this possible in 12.0? |
Manually attempting to configure CAD interfaces is no longer recommended as the process uses the Ansys registry, not the Windows registry as with regsvr32.exe.<br><br>Use the CAD... |
I have a coarse model in which I have done a transient analysis, about 50 time points. I want the boundary conditions to be transferred to the fine model at each time point and then I want the fine model to run at each time point. So do I have to do multiple SET commands, one for each time? e.g.
SET,NEAR,,,,5
SET,NEAR,,,,10 |
In the "ANSYS Advantage" magazine Volume 1, Issue 2 of 2007 there was an article on "Submodeling in ANSYS Workbench," which you can obtain at <a target=_blank... |
I have been conducting a series of surface-to-surface radiation transient thermal analyses in ANSYS Workbench using APDL code snippets. I have run into a case now where I need to input emissivity versus temperature for each surface instead of a constant emissivity. Do you know how I can do this? I have an example Command Object for inputting enthalpy versus temperature under the geometry object but do not know how to call upon a material property for emissivity within the code for surface to surface radiation. |
Note that when you are using Command Objects, you are using ANSYS command-line features; therefore, everything in the on-line help for ANSYS applies. You will see that section 4.6.1.1 of the... |
I apply the heat transfer coefficient to my model as a function of X via a table through a Command Object. Is there a way to display HTC values on the surface of the model during post processing as a contour plot? |
There is nothing preprogrammed in Simulation to view this quantity. Simulation uses either SOLID70 or SOLID90 element types for thermal analysis. Both of these output a quantity called HFAVG,... |
We're working on porting NOx models from TASCflow, but some of the reaction rates are coming out different in CFX.
I'm reading TASCflow data into CFX (including composition and temperature variance data) so it's not a matter of having a different starting point. - JCOOPER |
The differences between Tascflow and CFX stem from how the Temperature variance equation and Thermal NO reaction rates are handled.<br><br>1. There is a difference in the production... |