Boundary conditions and the block-coupled solver

celestial · November 27, 2019, 15:23

Dear members

Suppose that the boundary condition at an interface between a solid and liquid is specified by an algebraic equation involving only one kind of field. Moreover, the solid and liquid meshes are not conformal so that some sort of field interpolation must be carried out in that algebraic equation. I am solving the Laplace equation in both domains subject to this common boundary condition. i.e. I solve for field x in

Ax = b where A,b depend on the scheme, solver and boundary condition.

The question is can a block-coupled solver handle this case and if so, how? Even simpler: can it when the meshes are conformal so that no interpolation is required ?

It certainly is not obvious to me after reading up on the various openFOAM workshops and tutorials relating to this solver.

A great many thanks to the viewers of this thread
Marc

HPE · November 28, 2019, 16:22

afaik, block-coupled solvers are only available in foam-extend variant of OF. might be useful particularly to dig out Jasak's Zagreb group material.

celestial · December 11, 2019, 17:04

Sorry, HPE, for my late reply, but I am also tied up in meshing problems. Actually, I was careless when I wrote openFOAM. I happen to be working with foam-extend. And I wish I could use block coupled solvers to avoid the potential problems of segregated solvers not converging. But first, I guess I will have to understand the C++ open source code related to these block coupled solvers since I find the author's material not detailed enough.

Thanks anyhow

November 27, 2019, 15:23	Boundary conditions and the block-coupled solver	#1
celestial Member Join Date: Mar 2019 Posts: 86 Rep Power: 7	Dear members Suppose that the boundary condition at an interface between a solid and liquid is specified by an algebraic equation involving only one kind of field. Moreover, the solid and liquid meshes are not conformal so that some sort of field interpolation must be carried out in that algebraic equation. I am solving the Laplace equation in both domains subject to this common boundary condition. i.e. I solve for field x in Ax = b where A,b depend on the scheme, solver and boundary condition. The question is can a block-coupled solver handle this case and if so, how? Even simpler: can it when the meshes are conformal so that no interpolation is required ? It certainly is not obvious to me after reading up on the various openFOAM workshops and tutorials relating to this solver. A great many thanks to the viewers of this thread Marc

November 28, 2019, 16:22		#2
HPE Senior Member Herpes Free Engineer Join Date: Sep 2019 Location: The Home Under The Ground with the Lost Boys Posts: 932 Rep Power: 12	afaik, block-coupled solvers are only available in foam-extend variant of OF. might be useful particularly to dig out Jasak's Zagreb group material. __________________ The OpenFOAM community is the biggest contributor to OpenFOAM: User guide/Wiki-1/Wiki-2/Code guide/Code Wiki/Journal Nilsson/Guerrero/Holzinger/Holzmann/Nagy/Santos/Nozaki/Jasak/Primer Governance Bugs/Features: OpenFOAM (ESI-OpenCFD-Trademark) Bugs/Features: FOAM-Extend (Wikki-FSB) Bugs: OpenFOAM.org How to create a MWE New: Forkable OpenFOAM mirror

December 11, 2019, 17:04		#3
celestial Member Join Date: Mar 2019 Posts: 86 Rep Power: 7	Sorry, HPE, for my late reply, but I am also tied up in meshing problems. Actually, I was careless when I wrote openFOAM. I happen to be working with foam-extend. And I wish I could use block coupled solvers to avoid the potential problems of segregated solvers not converging. But first, I guess I will have to understand the C++ open source code related to these block coupled solvers since I find the author's material not detailed enough. Thanks anyhow