High everyone！
I’m currently working on solving a fluid-structure interaction problem using openFoam+calculix+preCICE, and have encountered a divergence issue in my computation. My model consists of a small convective cooling device surrounded by strips, with the overall device dimensions on the order of 10mm. The individual strips measure around 10μm in size, and the entire fluid domain spans approximately 100mm in scale. The model is shown as follow.
Due to the significant size differences among different parts, I adopted a smaller grid size between the gaps of the small strips. The final grid drawn is shown as follows.
In order to keep the courant of the calculation small, the time step of the calculation was set to 1e-5. However, at about 0.1355s, the calculation of the fluid suddenly diverged. When checking the log file, The calculations of U, P, etc. do not diverge, it might be due to the structural deformation squeezing the mesh, causing the mesh size within the gap to be too small, which in turn leads to a sharp increase in the courant number and subsequently causes the operation of openFoam to diverge. But preCICE specified only dynamicMotionSolverFvMesh openFoam for dyamicmesh. What I should do at this time to solve the above-mentioned divergence problem?
I also found that at the last step of the operation, the time step of the structure became very strange. What causes this? I’m currently using version 3.1.2 of preCICE.
Thank you for your help!
controlDict.txt (736 Bytes)
dynamicMeshDict.txt (350 Bytes)
log_fluid.txt (2.9 MB)
log_solid.txt (4.0 MB)
precice-config.xml (3.4 KB)