interFoam: Wrong / non-working shear stress computation when using variable density turbulence model
Summary
For interFoam computations in the marine branch and around hydraulic structures, the treatment of turbulence near the water surface is a major problem. Several aspects have been tackled, one of them is the (wrong) transfer of k and omega from the air into the water. This can be reduced by setting "density variable;" in constant/turbulenceProperties.
Bug:
When setting "density variable;", the functionObject "forces" computes wrong forces from shear. Furthermore, the FOs "wallShearStress" and "turbulenceFields" don't work at all.
Steps to reproduce
Unpack the supplied zip file and run interFoam in both directories.
Example case
What is the current bug behaviour?
Test 1:
Run interFoam in both directories "density_uniform" and "density_variable". Compare the results in postProcessing/forces in terms of x-direction: The viscous results differ significantly. (Real world example yields more than an order of magnitude difference => unusable)
Test 2:
For "density_variable" activate the FOs "wallShearStress" and "turbulenceFields" in controlDict. The computation will fail with message "--> FOAM FATAL ERROR: (openfoam-2212) Turbulence model not found in database, deactivating"
What is the expected correct behavior?
Both variants should produce the same results. FOs should work in both variant.
Relevant logs and/or images
Environment information
OpenFOAM version : v2406|v2312|v2306|v2212 Operating system : Linux based HPC Hardware info : Same behaviour on all tested platforms Compiler : Same behaviour on all tested platforms