[Hr_kules]

I have the, maybe, most basic question:

If i want to use wall functions for an buoyantSimpleFoam simulation, do i have to use wallfunction and prism layers, or do i just have the first cell at a distance that corresponds to a desired y+ value?

In my current simulation the results vary greatly with the expected outcome and the first cell height seems to influence the results more that i initally expected.

Apart from that during simulation if have the functionobject yPlus running, that basically always returns different values than the y+ I initally used to determine the first cell height?

Right now i am confused and sad.

Can anybody help?

[Hr_kules]

Maybe to clarify my problem a bit:

i have a pipe with 14m length and 1m diameter. It is generated from GMSH. The boundary layer is added with snappyHexMesh. For the first layer height i used the boundary layer calculator from this website: https://www.cfd-online.com/Tools/yplus.php
with following entries:
freestream velocity: 0.01305 [m/s]
Density: 998.2 [kg/m3]
Dynamic viscosity; 0.001002 [kg/ms]
Boundary layer length: 14 [m]
Desired y+ value: 35


The desired y+ value stems from the k-epsilon model i am using and ultimate goal is to compare simulation results from the dittus boelter equation for the nusselt number (Nu = 0.023*Re^0.8*Pr^0.4)


For the snappyHexMesh layer generation i run following parameters:

addLayersControls
{

    layers
    {      
        wallEntrance
        {
            nSurfaceLayers 3;
            mergeFaces false;
        }
        wallDeveloped
        {
            nSurfaceLayers 3;
            mergeFaces false;
        }        
    }

    relativeSizes             false;
    expansionRatio            1.1;
    firstLayerThickness       4.7e-2;
    minThickness              0.0001;
}

This should, or at least i thought, result in an y+ during the simulation of at least 30, however the los file of my simulation returns:

yPlus yPlus1 write:
    writing object yPlus
    patch wallEntrance y+ : min = 0.28638, max = 3.60023, average = 2.12067
    patch wallDeveloped y+ : min = 0.286411, max = 3.58798, average = 2.15099

For the first Iteration.


The log for snappyHexMesh returns following, which isn't better:

patch         faces    layers   overall thickness
                                [m]       [%]
-----         -----    ------   ---       ---
wallEntrance  16388    0.299    0.00292   2.96    
wallDeveloped 97078    0.328    0.0031    3.14

The results are also pretty bad. Anybody got an idea, how to resolve this?

[AtoHM]

It says only 0.299 layers (however that works ) were added and the stated overall thickness is way below the prescribed one for the firstlayer. Thus, the y+ value seems to be correct or at least consistent with that. It appears snappy has a hard time putting these layers in. What does the final mesh look like?

[Hr_kules]

Hey,


i found out that relaxing the mesh constraints is benefitial to the generation of layers and ran snappyHexMesh again with following parameters after bumping up the fluid velocity:

addLayersControls
{

    layers
    {      
        wallEntrance
        {
            nSurfaceLayers 2;
            mergeFaces false;
        }
        wallDeveloped
        {
            nSurfaceLayers 2;
            mergeFaces false;
        }        
    }

    relativeSizes             false;
    expansionRatio            1.1;
    firstLayerThickness       3.2e-2;
    minThickness              0.0001;
}

I have added images of the final mesh
And snappyHexMesh now displays following after generation:

patch         faces    layers   overall thickness
                                [m]       [%]
-----         -----    ------   ---       ---
wallEntrance  16388    2        0.0672    100     
wallDeveloped 97078    2        0.0672    100     

Layer mesh : cells:2065996  faces:4304480  points:436739
Cells per refinement level:
    0    2065996
Writing mesh to time constant
Wrote mesh in = 12.19 s.
Layers added in = 12.19 s.
Checking final mesh ...
Checking faces in error :
    non-orthogonality > 85  degrees                        : 0
    faces with face pyramid volume < 1e-13                 : 0
    faces with face-decomposition tet quality < 1e-30      : 0
    faces with concavity > 75  degrees                     : 0
    faces with skewness > 4   (internal) or 20  (boundary) : 0
    faces with interpolation weights (0..1)  < 0.02        : 0
    faces with volume ratio of neighbour cells < 0.01      : 0
    faces with face twist < 0.02                           : 0
    faces on cells with determinant < 0.001                : 0
Finished meshing without any errors
Finished meshing in = 14.75 s.
End

Finalising parallel run

I tried to run the case and the yPlus functionobject now displays at least somethon on the range of 30 (about 33-35).
However the desired temperature difference isn't still achieved.