Multi Region setup
 

I was wondering if someone could help me with the setup of a multiregion problem. I am trying to solve a Laplace equation in two domains, with continuity of pressure across the interface and a jump in the normal gradient (Darcy flow with two different permeabilities). I was using the conjugateCavity tutorial from foam-extend-3.0/tutorials/coupled/conjugateHeatSimpleFoam/conjugateCavity/ as a starting point and it worked without any problems.

For my own case file I have the mesh set up (passing checkMesh in both regions) and I can view it in ParaFoam - all seems OK. However if I try to view the pressure field in the slave region ("exterior") paraFoam crashes:

-------------------------------------------------------------------------

--> FOAM FATAL ERROR:

request for polyMesh interior from objectRegistry stefan failed
available objects of type polyMesh are

1
(
exterior
)


From function objectRegistry::lookupObject<Type>(const word&) const
in file lnInclude/objectRegistryTemplates.C at line 139.

FOAM aborting

-------------------------------------------------------------------------------

The reference to interior comes from the "shadowRegion interior" statement in the boundary file for the exterior region. It does not have a matching statement anywhere because the interior region is the master mesh. In conjugateCavity the statement is "shadowRegion region0". Again there is no matching statement anywhere in the case tree. I thought perhaps this was a special name and I tried that but it did not work either. I read a couple of tutorials: the report by Magnusson and the slides by Craven from OpenFOAM 6 but they both used different case setups and parameter lists than the tutorial in foam-extend-3.0. So I am a bit stuck and would appreciate any help or hints.

I am attaching the case directory. You get the current mesh setup by typing Allrun - I don't have a solver yet.

On another note - I was looking at the multiRegionHeater tutorial as well. In that case it seems you set the whole mesh up in polymesh (rather than a master polymesh and subregion polymesh dir(s)) and then add the subregion definitions with setSet. Are these equally good alternatives or are there reasons for choosing one set up over another.

Sorry for the long post

Tony

Follow up
 

I got paraFoam to show the initial pressure field.

Apparently the mistake was in constant/exterior/polymesh/boundary. I had the parameter "attached=on" in the regionCouple patch (as inMagnusson's report) but in the conjugateCavity tutorial it is set to off

Not sure what the difference is. Any enlightenment?