The test cases in this manual have been modeled to give reasonably accurate comparisons with experimental data wherever applicable, with a low number of elements and iterations. In some cases, even fewer elements and/or iterations will still yield an acceptable accuracy for the purpose of verifying the solution. The test cases employ a balance between accuracy and solution time. An attempt has been made to present a test case and results that are grid-independent. If test results are not grid-independent, it is due to the need to limit the run time for the test to be in the manual. Improved results can be obtained in some cases by refining the mesh, but this requires longer solution times.
Important: It should be noted that these are not validation cases of the models presented. The test cases are single instance simulations using one mesh, one turbulence model, and one scenario of the model.
The ANSYS solutions in this manual are compared with solutions or experimental data from textbooks or technical publications. Fluid-dynamics simulations have to make use of data available from experimental measurements for their verification primarily because closed-form theoretical solutions are not available for modeling the related phenomena. In this manual, test cases make use of experimental data published in reputable journals or conference proceedings for verification of the computational results. The experimental measurements for fluid-flow systems are often presented in the form of plots of the relevant parameters. The published experimental data for those cases and the corresponding simulation results are presented in graphical format to ease comparison.
Different computers and operating systems may yield slightly different results for some of the test cases in this manual due to numerical precision variation from machine to machine. Solutions that are nonlinear, iterative, or have convergence options activated are among the most likely to exhibit machine-dependent numerical differences. Because of this, an effort has been made to report an appropriate and consistent number of significant digits in both the target and the ANSYS solution. If you run these test cases on your own computer hardware, be advised that an ANSYS result reported in this manual as 0.01234 may very well show up in your printout as 0.012335271. Slightly different results may be obtained when different processor types or operating systems are used.