Replies: 2 comments
|
src/newTurbModels/models/kOmegaSSTtest.C After a thorough investigation, we discovered that the problem lay in the registration of the variable. If the registration template of the SAFv3 model is directly copied, it will cause significant errors when dealing with the case of using wall functions, but it generally won't cause major errors when dealing with the case of directly analyzing the boundary layer. We solved this problem by registering the variables according to the aforementioned code. It should be noted that if you have registered the turbulence model once before, after making the modifications, you need to run |
|
Thanks very much for this feedback! It is indeed a problematic setting in DAFoam. We will add a note for future developments that the new turbulence model should return proper APIs for wall function! |
Uh oh!
There was an error while loading. Please reload this page.
Dear Professor He, we have compiled a new SST model in the latest version of dafoam. However, it was discovered that even if the original SST code was completely copied in the
src/adjoint/DAModeldirectory and the name "kOmegaSST" was changed to another one, the calculation results would be different from those of the built-in SST model. We don't know why this happens. Logically speaking, after registering the names in thenewTurbModelsfolder, all the turbulence model calculations should be completed in thesrcdirectory and the calculation results should be exactly the same.But we found that this sometimes leads to significant errors.All reactions