Why am I getting a message about an unknown parameter? The message is:
*** WARNING ***
Unknown parameter name _Z1. A value of 7.888609052e-31 will be used. |
This is an internal parameter used by the GUI that was unintentionally left behind.<br>The message can be safely ignored.
|
Can I get the Segalman-Reese SEQV value in Simulation 10.0
when using a command object to run a PSD analysis? It appears that the
XML equivalent stress result returned to Workbench is not using the
Segalman-Reese approach introduced in ANSYS 8.1. |
That is correct. Workbench 10.0 retrieves only the<br>component stresses and computes any requested derived<br>values from them.
|
Suppressing and un-suppressing a body requires results to be
re-solved in Simulation 10.0. That was not the case in Simulation 9.0. |
Workbench development has informed me that this<br>behavior is part of the new state engine design at 10.0. It is required<br>for reliability reasons.
|
Are the PSD Reaction Forces correct in ANSYS 9.0? |
Yes, they are written to the results file during solution<br>and can be retrieved with PRRSOL. Keep in mind, however that<br>these are 1 sigma reaction forces and a sum of such... |
Who is the piezoelectric crystal manufacturer used in VM176? |
NEC/Tokin Corp is the manufacturer of NEPEC-6.
|
What is the meaning of C0, C1, CHI2 and R2 in a PDS scatter plot? |
C0 and C1 are the coefficients that define the regression line: Y=C0 + C1(X).<br>CHI2 is the well-known Chi Square criteria. Here is a typical web site:<br><a target=_blank... |
Does ANSYS have EMI shielding examples or demos? |
EMI Shielding generally requires a transient high frequency analysis,<br>which is not available in ANSYS (or from our competitors.) It may be possible<br>to do the analysis in SPICE... |
Is ANSYS supported on the AMD Turion processor? |
We have not performed any testing on the AMD Turion processor.<br> We can not say there will not be any problems.
|
Can you please provide me with help on how to transfer a 3D part model to 2D model? |
Here is the command input which was tested on a simple model. After<br>cutting the volumes with the working plane, it selects the keypoints on the<br>z=0 plane. Then it selects the... |
Q. Is it possible to change the CFX logo in CFX-Post in ANSYS CFX 10.0?
A. I'm afraid you cannot remove the CFX logo, but you can add your own, by setting an environment variable. The process of doing this is described in the attached doc
You need to set the following environment variable:-
CFX_USER_IMAGE_DATA = 'filepath xLoc yLoc xAttach scale alphaR alphaG alphaB transparency'
where,
filepath is path to image (only ppm, png, jpeg and bmp files are currently supported)
xLoc yLoc are the horizontal and vertical locations of the image in the viewer (0-1)
xAttach: left, centre, right or none
If set to none, xLoc is used
yAttach: top, centre, bottom or none
If set to none, yLoc is used
scale: image size relataive to viewer size (0-1)
If set to 0, original pixels are shown regardless of the viewer size
alphaR alphaG alphaB: red/green/blue components (normalised to 0.-255) of alpha (the colour that will represent 100% transparency)
transparency: overall bitmap transparency (0-1) |
Q. Is it possible to change the CFX logo in CFX-Post in ANSYS CFX 10.0?<br><br><br>A. I'm afraid you cannot remove the CFX logo, but you can add your own, by setting an... |
Q. When using Natural Size in ICEM CFD Tetra, is it possible to override the mesh parameters on specified curves and/or surfaces?
A. Yes, you can do this by setting specific mesh parameters on individual curves or surfaces under the tab, Mesh ==> Set curve/surface size |
Q. When using Natural Size in ICEM CFD Tetra, is it possible to override the mesh parameters on specified curves and/or surfaces?<br><br>A. Yes, you can do this by setting specific... |
Q. I would like to define two fluid domains in a geometry meshed in ICEM-Hexa, and then assign 2 subdomains in CFX-Pre. I create different solids in Design Modeller and then bring them to ICEM for hexa meshing.
A. In ICEM-Hexa, you need to create a separate BODY (under geometry) for each of the two fluid domains eg Fluid1 and Fluid2. You can then assign the hexa mesh blocks or material to the appropriate BODY. You can do this from the right mouse on the body name in the Parts list of the object tree using:-
Add to Part ==> Blocking material, add blocks to part. Then select the appropriate block.
When you create the mesh, there will be two separate mesh volumes, one each for Fluid1 and Fluid2. These mesh volumes will then be available in CFX-Pre to assign as different domains. |
Q. I would like to define two fluid domains in a geometry meshed in ICEM-Hexa, and then assign 2 subdomains in CFX-Pre. I create different solids in Design Modeller and then bring them to ICEM for... |
Q. How can I do particle tracking of non-mass-less particles (i.e. particles with a finite mass) as a PURE postprocessing task? If I have a computed .res file, can I take it into cfxpost and generate particle tracks (not streamlines)? The particles do not modify the flow (one-way coupling), but I do not want to insert particles in cfxpre, as I do not know prior to the simulation, where to insert them. This is a steady simulation.
A. If you want to include the mass of the particles in any particle tracking option then you need to set this up in CFX-Pre. However, as you are performing steady state one-way-coupling the particle tracking option only needs to be run for the final iteration. So you can run one additional iteration from your current res file with the particle tracking options enabled. |
Q. How can I do particle tracking of non-mass-less particles (i.e. particles with a finite mass) as a PURE postprocessing task? If I have a computed .res file, can I take it into cfxpost and... |
Q. I am attempting to generate a tet mesh in ICEM CFD 5.1.
I can mesh the volume fairly coarsely at about 200000 elemets, but when I try to generate a more refined mesh of around 500000 elements using 'natural size' and 'refinement' settings I get a message that 'faces are
missoriented' and several thousand messages along the lines of 'face at 1 2 3 has one volume neighbour and no shell element'. The volume mesh is created ok but when loaded into CFX-PRE there is a surface element set created 'inside' the volume (I guess this is all the misoriented faces) which means the Pre-Processing cannot be set up correctly.
A. I would recommend that after creating the tet mesh, smooth it and then run Mesh ==> Check Mesh.
This will highlight any problems in ICEM before export. If there are problems, you can either try to fix or change the geometry and/or mesh settings. Alternatively, with more experience, you could try to fix the problem manually by creating a subset when prompted (this will allow you to locate the problem). The edit mesh tools can then be used to fix it. |
Q. I am attempting to generate a tet mesh in ICEM CFD 5.1.<br>I can mesh the volume fairly coarsely at about 200000 elemets, but when I try to generate a more refined mesh of around 500000... |
Q. I have run a steady state simulation using the SST model. I would like to know how I can get the values of the Reynolds Stresses. Is it possible to get the Reynolds Stresses through a steady state simulation using SST or do I have to run the simulation again as transient?
A. There is no implementation to obtain the Reynolds Stresses automatically at present. They will have to create expressions in CFX Post based on the effective viscosity and the velocity gradients. i.e the term is modelled by the eddy viscosity approximation so you can backout the components of the stress that way. You do not need to rerun as transient.
Alternatively, it may be of interest to run with a Reynolds Stress Model? This will then give the required stresses in each of the six directions, however, this will be more CPU and memory intensive. |
Q. I have run a steady state simulation using the SST model. I would like to know how I can get the values of the Reynolds Stresses. Is it possible to get the Reynolds Stresses through a steady... |
Q. How can I get the vorticity at the end of a run? Is it necessary to use any expert parameter?
A. To get at the vorticity you have to include the following CCL in a steady state solver run:
LIBRARY:
VARIABLE: vorticity
Output to Postprocessor = Yes
END
END
For transient runs you can include it in transient files which will force it to be written to the final results file. |
Q. How can I get the vorticity at the end of a run? Is it necessary to use any expert parameter?<br><br>A. To get at the vorticity you have to include the following CCL in a steady... |
Q. We are not able to find the Blade Editor icon on the toolbar in BladeModeler, therefore we are unable to activate it. We have the Blade Modeler and the Design Modeler licenses included in our license files.
A. To access Blade Modeler, you have to turn on the BladeEditor add-in the first time after installing it. You do this through the Tools ==>Addins menu in Workbench. Select BladeEditor and make sure the relevant boxes are checkmarked. You may need to restart Workbench if you have already started Design Modeler. This information is contained in the Blade Modeler release notes under "Installing BladeModeler". BladeEditor is turned off by default because it is a Beta feature. |
Q. We are not able to find the Blade Editor icon on the toolbar in BladeModeler, therefore we are unable to activate it. We have the Blade Modeler and the Design Modeler licenses included in our... |
Q. I have a question regarding the Turbulent Schmidt and Turbulent Prandlt number using the K-Epsilon model for a combustion application.
From the Advanced Turbulence model control on the Fluid Domains panel, it appears there is a Turbulence Flux Closure for K and E terms. Does this mean that the Schmidt and Prandtl numbers are applied to the entire set of transport equations or just the turbulence terms. We would like to apply the Schmidt and Prandtl numbers to the entire set of transport equations.
A. The data structures under Turbulence Model/Advanced Model Control are for the corresponding turbulence equations only. Turbulent Schmidt/Prandtl numbers for scalar equations may be specified by CCL as follows:
FLOW:
DOMAIN: <domain name>
FLUID MODELS:
# Enthalpy and all mass fractions share same value (specified under heat transfer model):
HEAT TRANSFER MODEL:
Option = Thermal Energy | Total Energy
TURBULENT FLUX CLOSURE:
Option = Eddy Diffusivity
Turbulent Prandtl Number = 0.9
END
END
# For additional variables (individually for each scalar):
ADDITIONAL VARIABLE: <av name>
Option = Transport Equation
TURBULENT FLUX CLOSURE:
Option= Eddy DiffusivityTurbulent SchmidtNumber = 0.9
END
END
END
END
END
**** Entered By: dsclarke @ 12/16/2005 04:21 PM **** |
Q. I have a question regarding the Turbulent Schmidt and Turbulent Prandlt number using the K-Epsilon model for a combustion application.<br>From the Advanced Turbulence model control on the... |
CFX-Post fails with fatal error:
Fatal Error:
Aborting due to unknown exception. |
This is a known problem which exists for some graphic cards. To confirm, please try running CFX-Post in software rendering mode and see if it makes a difference:<br>Unix: run CFX-Post in... |
Q)
How do I calculate the gross flow across a plane?
A)
Sometimes it is useful to know the flow across a plane in one direction only instead of the net flow.
For example it may be useful to know how much total flow is coming in across an opening.
In CFX-Post create an expression for the velocity normal to a face, e.g:
LIBRARY:
CEL:
EXPRESSIONS:
VelFaceExp=max(u*nx+ v*ny+w*nz, 0.0 [m/s])
END
END
END
where (nx,ny,nz) are the x,y,z components of the unit normal vector, that is normal to the required face.
The max function ensures that only flows in the direction of the normal give a positive result.
Then create a scalar variable using this expression:
USER SCALAR VARIABLE:VelFace
Boundary Values = Conservative
Calculate Global Range = Off
Expression = VelFaceExp
Recipe = Expression
Variable to Copy = Pressure
END
The gross flow across a plane is then given by the area integral of the variable over the required locator.
This works for bondaries and CFX-Post locators, e.g:
flow = areaInt(VelFace)@opening
flow = areaInt(VelFace)@Plane 1 |
Q)<br>How do I calculate the gross flow across a plane?<br><br>A)<br>Sometimes it is useful to know the flow across a plane in one direction only instead of the net... |
QUESTION:
Do you have a high frequency electromagnetics transmission line example comparing results obtained using the release 9.0 port definition procedure (HFPORT,,TLINE) with those using the new (at release 10.0) modal port procedure (HFPORT,,MODAL)?
ANSWER:
Input files for a model of a planar, wide, band log-periodic balun are attached to this solution along with the IEEE paper upon which the model was based. The paper presents device performance over a range of frequencies. Due to time constraints, the input files only solve at a single frequency (3GHz). Presented results from the paper in Figures 6a and 6b show that the magnitude balance and phase difference at the two output ports at 3GHz should be ~0 and ~-180 degrees, respectively. Results from the 2 supplied input files are in agreement with these results.
Further discussion:
1) tline_ports.inp: at release 9.0 and earlier, transmission line ports were defined in the manner that they are in this file. Voltage paths had to be defined in the plane of the port if s parameters are to be extracted. BFE,,JS excitation was defined some distance from the feed port. Feed ports and ports across which radiation left the computational domain needed to be 'backed' by PML ('perfectly matched layers') to absorb outgoing radiation with minimal reflection.
2) modal_ports.inp: This model uses the new modal port feature new at release 10. This procedure is MUCH simpler to use than the old TLINE port procedure. Be warned that at release 10, the macro used to calculate field distribution at modal ports (HFMODPRT.MAC) did not always work for numerically small dimensions in MKS units (sub-millimeter and smaller). A new version that works much better has been developed for release 11 and is attached to this email. Before running this model at release 10, you must replace the old HFMODPRT.MAC macro file (C:Program FilesAnsys Incv100ANSYSapdlHFMODPRT.MAC) with the one attached to this email. N |
QUESTION:<br>Do you have a high frequency electromagnetics transmission line example comparing results obtained using the release 9.0 port definition procedure (HFPORT,,TLINE) with those... |
Is PPFA supported on Windows XP x64 systems, when released with ANSYS 10.0 service pack? |
PPFA will not be supported on x64 for ANSYS 10.0 SP1, but is supported for ANSYS 11.0.
|
How can the report from the PDS response surface calculation be interpreted?
For example, what does VIF mean? |
As we perform a regression analysis to create a response surface, there are some assumptions made about the fitted response surface<br>and the content of this report is used to verify... |
Should I be concerned about the warning message below when
I run a squeeze film damping analysis with FLUID136?
*** WARNING *** CP= 132.562 TIME= 11:48:18
Beta damping multiplier: -2.356628492E-07
A negative beta damping multiplier occurs. Please check results
carefully! |
Squeeze film damping tends to be of the viscous type, which produces<br>a high alpha damping value and low beta damping value. The alpha and<br>beta damping values are computed based... |
What are the coordinates of the following points in a CCD?
a. central point (Blue round sign)
b. the N axis point (Red round sign)
c. 2N-f factorial points located at the corners (Yellow round sign)
Does it change with the type of Distribution Function used (Gaussian or
Uniform?) |
The PDDOEL documentation indicates the points are based on probability<br>so the location would be based on the design variable distribution.<br><br>If the variable has a... |