Convergence in CompoundMesh with CyclicAMI
 

I am still confused what cyclicAMI means, or why my model does not converge (see tips here
https://www.cfd-online.com/Forums/op...-problems.html )


I use it for a compound mesh :

https://farm5.staticflickr.com/4918/...a850c29d_z.jpg" Periodic (Cyclic) boundary conditions are very strong tool in CFD. Meshes of different elements can be inter-connected. The boundaries have to be of equal shape and equal area. The transformation can be rotational, translational or noOrdering. " https://www.cfdsupport.com/OpenFOAM-...t/node115.html


I would guess Cyclic Boundary Conditions are just: https://en.wikipedia.org/wiki/Period...ary_conditions


The use of the boundary condition at the interface of the angled mesh and the rectangular mesh seems okay.

I am able to get convergence here:
https://www.youtube.com/watch?v=b00-xI-l0YE


If I decrease the kinematic viscosity by 4 orders of magnitude from 0.2 to 2e-5 I am unable to find inputs in my fvSolutions file that lead to convergence.

My complete exploration from Meshing to the present modelling is available here:
https://www.youtube.com/playlist?lis..._W7ozcZV_4-JBA

varying only p and U residuals success
 

I am having success with varying only p and U residuals for the kappa-epsilon model.

https://www.cfd-online.com/Forums/op...-problems.html