[TimoS]

Dear Foamers,


my simplified model has a rotating AMI pendulum in Boussinesq alpha.waves. Gravity is given in constant/g: g (0 0 -9.81); and same vector in constant/dynamicMeshDict. Pendulum refuses to rotate although very nice wave hits it and manual calculation shows it should. Looking at first output 0.1/uniform/sixDoFRigidBodyState acceleration, there is value -280.xxx in vertical direction, and it grows towards -1000.xxx in 7.5 printout. This much vertical gravity will of course lock the pendulum against any wave forces. I have rechecked several times all input files, but there is absolutely no other accelerations given. The entire dynamicMeshDict is copied from Tobias Holzmann VAWT example, and comparing it to other turbine models confirm that it is correct as far as I can understand.

Manual comparison shows, that acceleration is almost twenty times too high at the beginning, and increases along the analysis. In addition it has incorrect sign, causing a float to sink instead of moving towards surface. Still there is no reasonable explanation for the phenomena. Solver has been overInterDyMFoam and rigid body solver sixDofRigidBodyMotion. Are they correct choices? Float is modeled as a void inside rotating AMI mesh, centreOfRotation, centreOfMass and motion axis are correct and mesh remains rigid in the very small motion, which the wave can cause.


Due to limited experience I still doubt my model, but a small doubt towards a bug also starts to rise. Can somebody give a hint how to find the source of acceleration if it can be caused by a mistake in my model.


OpenFOAM version is 2212 in Windows10 wsl2/Ubuntu22.04.1LTS.