[Giuseppe91]

Hello everyone, I have a problem: I simulated the case of the openfoam tutorial DLR_A_LTS, for the base case everything seems to work but when I increase the mesh in the fuel inlet duct and in general on the symmetry axis the jet speed assumes too high values ​​(not physical). I think the problem may be too dense mesh for reactingfoam. I left the boundary conditions as in the tutorial in particular u as fixedvalue. Can someone help me. Thanks in advance.

[sufjanst]

Do you start your simulation from 0 with the new mesh?

Maybe your new mesh has a bad quality. Did you run checkMesh?

When do the high velocities appear?

Does the simulation crash?


You can try to run potentialFoam first. It creates a U-Field from your boundaries and can be used as an initial flow field. potentialFoam edits the U-File in your "0"-folder. So make sure to create a backup.

After running potentialFoam you have to delete the "phi"-file in "0" and you can start reactingFoam with an initial flow-field.

[Giuseppe91]

thank you so much for your answer,
in fact it seems to be a problem of quality of the mesh, because I tried to simulate with a thicker mesh and u field has a physical behavior.
now unfortunately I have another problem I set the source of ignition via toposet. I would like to do a convergence analysis for the various meshes. in the 1x mesh the flame lights up in the 2x and 4x meshes the flame does not light up under the same conditions.
do you have any ideas to suggest?
best regards

[sufjanst]

How do you select your cells with toposet? I never worked with toposet.

I usually set high temperatures and exhaust species in a box oder cylinder for ignition with setFields.



My idea for your problem: If you select your cells in a non-geometric way (like a box or cylinder), the cells are much smaller in your 2x or 4x mesh. The ignition area is much smaller and maybe not big and hot enough to ignite the flame. I would suggest that you define a box or cylinder and patch the selected cells. I hope you understand what I mean and that toposet can deal with that.

[Giuseppe91]

I had thought about the same thing, I set toposet in the following way


actions
(
{
name ignitionCells;
type cellSet;
action new;

source sphereToCell;
sourceInfo
{
centre (0.0 0.0 0.04);
radius 0.04;
}
}
);




so I tried to lengthen the duration of the ignition source via fvOptions even at 10000 time step, the ignition sphere reaches temperatures of 3000 K.
In the 1x mesh, a 2000 time step is enough to light the flame.
Sorry if I disturb you further, I have two questions:
I have


runTimeModifiable true;

adjustTimeStep yes;

maxCo 0.4;


in this case every time step has a shorter duration of real time (physical) in the mesh to 2x than the right 1x? Do you know how I can get physical time?
Having an axisymmetric geometry I set a 1 degree angle, could it be responsible for a too small volume of ignition?

thank you so much for your help

[Giuseppe91]

fvOption file


options
{
energySource
{
type scalarSemiImplicitSource;

timeStart 0;
duration 10000;
selectionMode cellSet;
cellSet ignitionCells;
volumeMode specific;

injectionRateSuSp
{
h (1e7 0); // J/m^3/s
}
}
}