I am trying to simulate the dynamic lift of Semi-planing hull during high speeds. For this I have prepared setting in fluent for dynamic mesh. But I am facing negative cell volume error. After reading discussions and having some suggestions. I have tried with making a large domain, turning off the inflation layer and changing values in remeshing. Nothing worked for me. Please help.
Normally, we have to remesh the geometry with different options.
Hi Syed,
You need to focus on three aspects here: mesh model, solver and boundary conditions. It is more probable that the issue with the mesh. You need to make sure that you are enabling the mesh matching for different bodies at their conjunction for this case. This could help your problem.
Hi Syed,
Do you have moving bodies in your solution domain? If so, and if you haven't tried yet, try using contact detection to avoid contact and associated negative cell volume problems. You could either use the built-in contact detection or write a UDF to do the same job. Let me know.
Normally, we have to remesh the geometry with different options.
@Sevki Cesmeci Sir, thanks for your suggestion. But Im having the same problem i.e negative volume detected. Pls suggest me the ways to check my mesh that whether it is upto mark for dynamic meshing or not.
Hi Syed,
If I understand right, you are trying to calculate the net force on the body to lift it up. My suggestion is to estimate the maximum value of that force and using the momentum equation, determine the velocity (as a constant number). Then use that velocity through a UDF to perform the dynamic meshing and preview mesh motion. Please, note that setting the time step is critical (I'm not so familiar with smoothing and remeshing techniques, but in layering technique, Fluent cannot collapse or generate more than 1 row of cells in a single time step).
This approach will help to detect whether the issue in the cell size near the moving walls, time step or due to momentum. If you adjusted a certain max velocity of the body and the mesh moves well, then the issue belongs to the parameters controlling the momentum of the body (mass, moment of inertia, lift force and so on).
Hope this helps
Wael
@Wael Elsaady thankyou for suggestions. I will look in to this
Dear Muhammad
The face validity is an area-weighted measure of the correctness of the face normals relative to their attached cell centroid. In a good quality cell, the face normals point outwards, away from the cell centroid. In a cell with bad face validity, one or more of the face normals point inwards, towards the cell centroid. Face validity of 1.0 means that all face normals are correctly pointing away from the cell centroid. Values below 1.0 mean that some of the cell faces have normals pointing inward towards the cell centroid, indicating some form of concavity. Values of below 0.5 signify a negative volume cell. Cells with a face validity below 1.0 are considered bad. (Ref: StarCCM tutorial) Please see the attached figure, and this is what happens in 3 dimensional and volume mesh as well, and results in negative cell volume. To rectify this issue you have different options, however, they may be somewhat hard to be performed. First, you could make use of manual meshing through the Gambit that could be time-consuming but could end up with certain results, since you have control over all edges. The second option is decreasing mesh size or changing mesh type. As you can see in the attached figure splitting up the bad mesh into two cells in that setting could prevent negative cell volume. The last thing to mention is utilizing volume control over sharp edges or over curved plates and decreasing mesh size, though you have to be careful about the diffusion of results between volume control and the rest of domain, so the last mesh size in the outer layer of volume control must be set proportional with the 10 first layers of beside outer area in terms of expansion factor.
regards
I think it might be a solution to move the geometry to the positive side of the coordinates as shown in the picture below.
@Mehmet sir its already on positive side. My simulation ran for 12000 iterations. And after that it detected to negative cells.
The number of iteration is too high. The analysis must have converged before 12000. I think there is a problem with mesh quality.
Mehmet Salih Cellek Sir, Skewness is 0.7. what else i could see. If i put inflation layer. the skewness goes toward 0.9 but i read somewhere that its not an issue. we can define it as a separate cell zone.
Dear Muhammad,
You may need to check face validity and cell quality. Run a metric first, check out face validity and cell quality which must be near to one. As long as you have less-than-0.5 value in your metric, the negative cells still exist. So apart from what I sent you before, you could refine your mesh size and keep your eyes on cell quality or face validity values, once they get above 0.5 you wont have negative cells.
regrds
- Badges
- Science topic