SiteMap Page 121


Previous 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 Next

The ANSYS 10.0 Connection for SAT product also imports solid model data when the
WIREFRAME mode is selected, not just line entities. Is this a known error?


Yes, but the ANSYS Connection products are in maintenance mode, so this cosmetic<br> error will not be addressed. Only major errors in these products will be fixed, as they<br> ...

How to call a Windows Fortran DLL from a perl script


Problem:<br><br>In a Perl script, it is sometimes desirable to call Fortran subroutines in a<br>dynamically linked library (DLL). This may be useful in cases where a...


When running ICEM, the tetra and/or prism icons are not present. This problem has been encountered on 64 bit machines running Linux.


<br><br>Set the following environment variables to the values shown:<br><br>AI_ENV_VOLUME_MESH = 1<br>AI_ENV_PRISM = 1<br><br>It will be necessary to shut...

Setting a fully developed flow in a periodic duct.


Sometimes, it is necessary to set a fully developed flow in a periodic domain<br>(i.e. with no 'flow' boundaries). Often, the requirement is to achieve a<br>specified...



By default, the particle track coordinates are output to the results file whenever a particle track crosses an element boundary. For many applications this is sufficient, but in some cases a finer resolution is required. The default setting for the particle track file option is 'All Track Positions', however this will not output particle track coordinates to the results file at a greater resolution than the mesh.


<br><br>Particle track coordinates can be written within elements by using the appropriate particle output control settings. These are available in CFX-Pre from the output control...

Setting up a user defined particle evaporation model.


Although the built in evaporation model suffices for most applications, a user<br>defined evaporation model is sometimes required.<br><br>A user defined particle evaporation...

Q. A customer is experiencing problems with the piping module. When, say, an elbow is inserted, instead of picking up the correct pipe properties from the two elements between which the elbow is being made, it defaults to the properties of the last defined section. Thus bend radius, if set to LR or SR, is wrong and other real constant data calculated by the piping module is thrown into question. What is the cause.

A. This is a code defect (#26606). It has been corrected for 11.0 release.


Q. A customer is experiencing problems with the piping module. When, say, an elbow is inserted, instead of picking up the correct pipe properties from the two elements between which the elbow is...

Q. Some ANSYS input files created in Simulation on a Windows Platform cannot be read correctly by ANSYS on an AMD Opteron/SLES9 platform.
What is the cause?

A. ANSYS input files are being written in PC formatted files instead of a universal file.
PC formatted files have end of line markers that do not always coincide with other computer platforms.
We have to activate an ASCII protocol in Linux FTP and hope that the FTP
file transfer between PC and Linux will strip out the strange end of line characters.
Some UNIX/Linux systems may be able to detect a PC file and read it
correctly and others may not.


Q. Some ANSYS input files created in Simulation on a Windows Platform cannot be read correctly by ANSYS on an AMD Opteron/SLES9 platform.<br>What is the cause?<br><br>A. ANSYS...

I have written basic Tcl scripts for meshing/smoothing/exporting our meshes to minimize user interaction in the process. I am now looking to generalize these a bit by putting a GUI on the front end that I can input tetin name/location, whether to tet>hex, how to smooth, what name/location to output to.

Looking at the 'Programmers Guide' this seems best done building up from a form_init command. This seems quite specific to ICEM's Tcl launguage and thus I cannot find any examples of it on the net.

Please can you send me some examples for scripts that build up forms and act on their entries. Also if you can point me in the direction of any useful sites that will be much appreciated


I would probably stay away from all the form_ procedures, since they are really done in the older GUI style, and if you create your own Tk widgets, you can then run ICEM in batch, but use only...

Figure 12.2 "Examples of Singular Elements" in the ANSYS 10.0 Structural Analysis
Guide does not appear to be correct. The degenerate shape for the PLANE82 element
shows nodes I, L, and P to be coincident. Is this allowed?


In short, no, but the Fracture Mechanics documentation is going to be completely<br> rewritten in ANSYS 11.0, so this error will be corrected then. Per the Elements<br> ...

How do I set a particle injection rate at a frequency less other then every time step in a transient simulation?


In this case, particles are not injected at every time step, but are injected at every nth timestep or some other frequency. In order to accomplish this, the particle mass flow rate should be...

What does a local timestep factor (LTF) in the CFX solver mean?


<br><br>Point (a)<br>When you run local timestep factor the solver does not assemble a pseudo transient term because it is not needed. Local timestep factor is a different...

I have installed CFX 11 Preview 4 on my PC but I am not able to run CFX-Pre. I get the following pop-up when I try to either start a new simulation or open an existing simulation:

ERROR
No license for ANSYS CFX-Pre could be found

CFX-Pre from CFX 10 works fine.

In my license file I have a 1 task acfx_flotran_upgrade key and a 2 task ane3fl1 key - both of which are current and available.


To run CFX-Pre the user must have 1 acfx_flotran_upgrade key and 1 other key from the list: stfl, stfldp, stflds, ste3fl, ste3fldp, ste3flds, anfl, anfldp, anflds, ane3fl, ane3fldp, ane3flds,...

Customer would like to borrow a OneSpace interface license, but this doesn't appear in the anlic_admin borrowing tool.
Why isn't the OneSpace license available for borrowing?


We have added the option in the AnsBorrow.xml file to resolve this issue.<br><br>The corrected xml file is attached to this solution.<br>Note that this also contains the changes...

SYMPTOM:
For cases where there are very large gradients of boundary flow data on a surface, a surface plot of that information will show strange behaviour just locally to the region of large gradients. For example, in a 2-dimensional case, the data will show some non 2-dimensional behaviour in the region of large gradients. Also, quantitative calculations using this boundary data will be slightly in error.

PROBLEM:
The problem is caused by an error in the integration point to face set ordering within CFX-Post.


WORKAROUND:<br>There is no workaround.<br><br>FIXED IN:<br>ANSYS CFX 11.0

SYMPTOM:
If the user changes the solution units ifor coordinates n CCL, but does not re-write the DEF file from CFX-Pre to have the mesh re-scaled, then the units for the coordinates go wrong in the Solver.

PROBLEM:
The problem occurs because there is nothing in the def file to indicate what units the mesh is stored in, and therefore the Solver cannot tell that the user has changed units, and that there is thus an inconsistency.


WORKAROUND:<br>The workaround is to re-write a def file with the mesh scaled.<br><br>FIXED IN:<br>Not fixed yet.

When I try to plot the data created by the ANSYS hyperelastic curve fitting routine,
my ANSYS 10.0 session crashes under Windows 64-bit, but not Windows 32-bit.
Is this a known error?


Yes, this is a known error. Defect 42737: "Curve Fitting crash in ANSYS 10.0A1<br> on Windows XP x64" was filed and corrected in the April 21st build of ANSYS...

Customer says that the command line window, main menu, and standard toolbar are missing from his utility menu.
What would cause this?


In this case, the tlbrlist100.ans file had the wrong path. Once corrected, all items appeared in the utility menu.

**** Entered By: rlange @ 05/25/2006 03:28 PM ****

Q. I want to do a transient FSI problem with ANSYS and CFX but do not want to account for the effect of the deformation of the structure on the flow field, as it is expected to be small. How should I approach this?

A. The two methods to accomplish this are the CFX-Post method and the MFX method.

1. CFX-Post method: Solves the CFD problem by itself first. Then, create an ANSYS cdb file of SURFACE154 elements generated from the surface of the structural model that is to accept the CFX pressures/shear stresses loads. Import the cdb file into CFX-Post (File> ANSYS Import/Export) after the transient is completed and having saved transient results in CFX at suitable intervals, making sure that at least 'Output Boundary flows' are stored in the CFX intermediate transient results files. (see documentation on how to turn on 'Output Boundary Flows' in CFX-Pre, under 'Output Control'.

Once the Results file has been loaded and the surface imported, the load files for ANSYS are created by first accessing the particular time step result in the Time Step selector, and then exporting (File> ANSYS Import/Export). Choose a different name for the export file for each time step.

If this is done in Simulation, one need only choose CFX Pressure as the loading. The user does not have to create the 2D surfaces, as this is done automatically by Simulation. One is prompted to browse for the results file and then choose the appropriate locator and time point. To do a second time point, the load should be deleted (or suppressed) and reapplied, as it cannot be modified once specified.

2. The MFX method: in this case, one sets the ANSYS and CFX problems up as an MFX two-way problem, except that mesh motion is set to None in CFX-Pre under 'Mesh Deformation' in Domain settings. The normal two way FSI load transfer becomes a one way transfer. There will be no reference in CFX to accepting ANSYS displacements because t


**** Entered By: rlange @ 05/25/2006 03:28 PM ****<br><br>Q. I want to do a transient FSI problem with ANSYS and CFX but do not want to account for the effect of the deformation of...



I have created a mesh for one blade passage in TurboGrid 10. The nodes on periodic surface are generated with full one-to-one match. After I bring that mesh into Pre, I try to make several copies and glue them together to make a full fan model. However, Pre does not glue those copies together. Also, when I try to export the single passage with a one-to-one match on the periodic interfaces, Pre reports that a GGI connection was used instead. What is the problem?


<br><br>This can often happen in Pre when the mesh generated in TurboGrid has a well refined boundary layer. In order to force Pre to glue the assemblies together, you will need to...



When importing my curve files, I get the following warning in TurboGrid 11.0:

WARNING
HUB dimensions, 1906.249268, are outside of the ideal range (0.2-200).
Without corrective action, serious geometry and mesh generation problems may result.
Try using different Base Units for Machine Data.

WARNING
SHROUD dimensions, 1562.279419, are outside of the ideal range (0.2-200).
Without corrective action, serious geometry and mesh generation problems may result.
Try using different Base Units for Machine Data.

WARNING
Blade dimensions, 449.964568, are outside of the ideal range (0.2-200).
Without corrective action, serious geometry and mesh generation problems may result.
Try using different Base Units for Machine Data.

What should I do?


<br><br>This warning should not be taken lightly!!! It may result in surface trimming, surface intersection, projection and mesh generation errors! Base units are used by underlying...



When importing my curve files, I get the following warning in TurboGrid 10.0:

ERROR
Last blade profile is not defined as a shroud tip and is too far (0.5 of span) from the
shroud for reliable intersection. Must be at least 0.920000 of span.

What should I do?


<br><br>There are two reasons why this could happen:<br><br>1) The wrong length units were selected for the hub and/or shroud and/or profile curves. Double check these...


I'm trying to use the HP MPI parallel method for the CFX Solver in conjunction with ssh (rsh has been disabled). I keep getting the following error:

machinename: Connection refused
An error has occurred in cfx5solve:

The CFX-5 solver could not be started, or exited with return code 255. No
results file has been created.


<br>The environment variable CFX5RSH is usually used to allow the CFX Solver to operate with ssh instead of rsh:<br>CFX5RSH = ssh<br>However, HP MPI has it's own...


My Particle Tracking solution contains the variable "Particle.Wall Mass Flow Density", where "Particle" is the name of my particle material. This has negative values on the walls, is this normal?


<br>There are two Mass Flow Density variables available in CFX-Post:<br>1. "Particle.Mass Flow Density" is defined on Inlets, Outlets and Openings. A positive value...


How can I use rough walls with the SST turbulence model?


<br>You'll have to edit the CCL to use this beta feature. The CCL for a wall boundary condition in CFX-10 should look like the following:<br><br><br>BOUNDARY :...