SiteMap Page 132


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

Why is Simulation 10.0 giving wrong mutual inductances for a 3 coil problem?
The result was correct in Simulation at 9.0.


At 10.0 Simulation began using the undocumented command LMATRIXW<br>to compute inductances. It is faster than the documented LMATRIX. At 10.0<br>it had a defect, 44217, that caused...

How can I find bad elements in an unconverging nonlinear solid117 model in Simulation 10.0?


The BH curves may not be the problem. When I run your model<br>I get pivot messages which is unusual. I converted all the<br>elements which use BH curves to air and got some B...

Why does the Structural Analysis Guide give a different procedure for Spectrum?

The 8.1 Dynamics course shows a 3-step process:
(1) modal analysis with expand,
(2) spectrum analysis, and
(3) postprocess including reading the .mcom file

The Structural Analysis Guide chapter on Spectrum analysis shows a 6-step method:
(1) Build the model.
(2) Obtain the modal solution (Modal analysis).
(3) Obtain the spectrum solution (Spectrum analysis).
(4) Expand the modes (Modal analysis).
(5) Combine the modes (Spectrum analysis).
(6) Review the results (/Input the jobname.mcom file).


The Structural Analysis Guide procedure is a little more efficient in that a user can expand<br>only the modes that were determined to be significant in step 3. Very few users take...

Why does PCG require so much memory on this model with contact in ANSYS10.0?


The problem is due to an inefficiency in the new preconditioner with contact (Ref: defect 44278.)<br>The problem is corrected in 11.0. For this model, setting PCGOPT,2 is a workaround.

Why isn't NUMMRG,KP merging duplicate lines, areas and volumes in ANSYS 10.0?
It did in 9.0.


This is occurring when duplicate entities share the same keypoints. A defect 30609 attempted<br>to make it possible to suppress the merging of such entities when they weren't...

How can I simulate joule heating in a stranded conductor?
The coil has temperature-dependent resistivity.


The input below is the simplest way, but it does<br>use an undocumented plane13 keyopt(1)=5. The<br>results can be checked by hand (compare the etable item<br>jheat with element...

How can I tell which cable elements aren't converging?


One way is to postprocess the unconverged solution and<br>display the disequilibrium forces (/pbc,nfor,,1or2). Large values<br>will be at nodes of elements that aren't converging.

Why do solid117 magnetic forces on conductor look wrong at 10.0?


At 10.0 the JXB force calculation was replaced with the new energy-based<br>force calculation with EMFT. Due to the use of centroidal field values instead<br>of integration point...

How can I compute harmonics for my magnetic solution?


The magnetic field values can be stored in an APDL array using a<br>path operation and then fourier coefficients can be computed. The sample<br>input and macro below demonstrate the...

Why is the FLUID136 method for obtaining squeeze-film damping giving a small
negative beta value and issuing a warning?


Squeeze-film damping is almost purely viscous and alpha damping dominates.<br>The warning message can probably be ignored. A negative value for beta damping is quite<br>common and...

How should the PDS Constant Variance Test results be interpreted?
What is the inference of the results are less than 1%?


Per development:<br><br>If the pValue is less than 1%, then the variance of the error term is not constant.<br><br>Now, let me provide a bit more extensive explanation...

Why does the ADAMS Interface in ANSYS10.0 abort when deformable MPC contact is used? This is a
very good way of supporting a master node at the center of a hole or other location where
there isn't a node.


This problem is due to the way the constraint equations being created by RBE3 are handled in the CMS<br>generation pass, particularly when one of the terms has a constraint applied to it. ...


How should you set up different wall boundary conditions for different fluids in a CFX 10.0 multiphase calculation?


<br>To be able to select different wall boundary conditions for different fluids, you need to:<br>1) set 'Fluid Dependent' option for 'Turbulence' in the...

How does one use ANSYS CFD software to simulate gas plant enclosure ventilation and safety, subject to PM84 guidelines?


Read the attached note.

In CFX 10 and CFX 11 the speed of sound in the wet steam dome is calculated as the speed of sound on the saturation line , and not in the dome.


This is true for CFX 10 and CFX 11, and will be corrected in CFX 12. The solver uses the minimum value of the speed of sound between its value at liquid saturation and the vapor saturation phases...

When writing out partial TRN files (not full TRN files) in a transient simulation, the mass flow averaging does not function correctly, as the integration point (IP) mass flows are not stored in the partial TRN files by default.


Setting the undocumented expert parameter "mfloip to trn files = t" in the CCL will cause CFX to write out also the IP mass flows to partial TRN files to enable the user to perform...

How to create several hub to shroud graphs of circumferentially averaged quantities and display them all at once in one chart in CFX-Post?
When using the Turbo mode and there the Turbo chart to create spanwise (hub to shroud) distribution plots of variables, a change in the meridional position of
the graph will delete the previous graph. Only one graph can be displayed at once.


Solution procedure:<br><br>* Create one graph after the other with the hub to shroud Turbo chart tool of CFX-Post.<br>* Each time save the graph to a CSV file by using the option...

When using an own RGP file to define a real gas / wet steam material for use in a CFX simulation, and the material is assigned to the material group "User" only, the solver crashes in the 1st iteration with unphysical values for enthalpies, temperatures and other variables.


One possible reason is the wrong setting of the thermodynamic state of the real gas material. For dry real gas or the vapor component of a dry/wet homogeneous binary mixture real gas, the...

How to calculate and export the normal velocity components on an arbitrary flat plane, which is not aligned with any of the coordinate frame axes?


To calculate the scalar value of the normal velocity on the plane:<br><br>* Create a new variable (type scalar), e.g. VelNormPlane1 = ( Velocity u * Normal X + Velocity v * Normal Y +...

The initialization of the Turbo mode of CFX-Post fails for some RSO files generated by CFX-TASCflow multiple frames simulations: the number of recognized domains is not correct


By default, CFX Post Turbo mode just recognizes 2 components from RSO files, as there are just 2 frames defined (1 stationary and 1 rotating frame), independent on the real number of components....

I have opened a model that was prepared on a different machine by someone else. I get the following error immediately upon hitting Solve:
"A general failure occurred during the solution process."

Simulation does not even write a ds.dat file. What could be causing this problem?


This error message is very non-specific and can occur due to a wide variety of problems.<br><br>In this particular case, the externally created model contained a Beta feature in the...

Class 3 Error

SYMPTOM:
When using the CEL variable 'taxis' in an expression, the resulting value is not what is expected.

PROBLEM:
Several points should be noted about the CEL variables "theta" and "taxis", and the post-processing variable "Theta".

1) The CEL variable "theta" is always atan2(y,x) in any coordinate frame. This means that it varies between -pi and +pi about the z axis and is zero at the x axis.

2) The post processing variable "Theta" (used in CFX Post), follows the convention for a "right handed" coordinate system, thus

- If the rotation axis is the z axis of a coordinate frame then "Theta" is evaluated as 'atan2(y,x)+pi'.
- If the rotation axis is the x axis of a coordinate frame then "Theta" is evaluated as 'atan2(z,y)+pi'.
- If the rotation axis is the y axis of a coordinate frame then "Theta" is evaluated as 'atan2(x,z)+pi'.

The range of "Theta" is therefore 0 to 2pi.

3) The CFX Solver always evaluates the CEL variable "taxis" using the Global Z axis in the absence of a rotation axis. This occurs even if a local Reference Coordinate Frame has been selected for use in a subdomain or boundary condition (Class 3 Error).

The range of "taxis" is currrently from -pi to +pi, when it should be 0 to 2pi.

4) CFX Post should allowusage of taxis/raxis in expressions, but it current


WORKAROUND:<br>The only workaround is to calculate the correct values of 'angle about an axis' using alternative expressions. This can be accomplished with a knowledge of the...


Do BladeEditor and TurboGrid 11.0 use similar methods for determining the periodic surfaces between adjacent blades?


<br>In ANSYS 11.0, BladeEditor and TurboGrid use similar methods for determining the periodic surfaces between adjacent blades. Different methods are still used for the portion of the...


How can you model in BladeGen 10.0 an impeller where the hub does not extend beyond the blade trailing edge?
We have a mixed pump with an impeller that has a cutoff trailing edge at the physical end of the hub, and the return channel has a rounded leading edge at the beginning of the hub surface. What happens if your real geometry does not extend beyond the trailing edge or leading edge?


<br>It is quite common in reality that the impeller may have a cutoff trailing edge at the physical end of the hub, and that the return channel may have a rounded leading edge at the...


What are the specifications and limitations for the blade curves for Bladegen 10.0? Are they similar or the same as the specifications for Turbogrid .curve data?

We are preparing a reverse engineering methodology for our customers. They have bladed components with very primitive CAD description (or in some case no CAD description at all). In order to get them to simulation we are testing the import of some curves into Bladegen. Using 3d laser scanners we are able to scan one blade (including a part of the hub and shroud). We then manipulate the stl file and are able to get curves for hub, shroud and blade profiles. We are then using the import wizard of Bladegen to read the curves and assign them to hub, shroud and blade profiles. This is the general method that we use.


<br>The specifications for Bladegen are such that the blade profiles should lie on a surface of revolution and contain a complete loop of points.<br><br>The specifications are...