[Fransje]

Hello developers,

Perhaps somewhat of a double post, as I posted a similar message in the Running/Solving section of this forum, but to no avail....

Well, my problem is that I'm trying to run a double region simulation by implementing a modified version of pisoFoam.
I think I've come quite far in my implementation, and when starting a run my solver reads the correct meshes, with the correct number of cells, the right names, correct implementation of the turbulence model, etc. So far so good.

However, when starting the time loop, OpenFOAM complains about not finding region0 (which is the standard region, but which I don't have since I named my regions differently.) I get the following error:

--> FOAM FATAL ERROR: 

    request for objectRegistry region0 from objectRegistry doubleDomain_split failed
    available objects of type objectRegistry are

2
(
region1
region2
)

I find quite strange that OpenFOAM is wanting to use region0, whereas I really specify region1 or region2 in my code, and that they are available as objectRegistries..
Would someone have an idea on how to fix this?

Kind regards,

Francois.

[Fransje]

Hum..

I've pinpointed the error origin to the command:

runTime++;

I don't know if that's of any help...

Looking forward to your replies!

Francois.

[Fransje]

And in Time.C, the problem occurs in the function:

Foam::Time& Foam::Time::operator++()

At the line:

00645     if (!subCycling_)
00646     {
00647         if (timeIndex_ == startTimeIndex_)
00648         {
00649             functionObjects_.start();
00650         }
00651         else
00652         {
00653             functionObjects_.execute();
00654         }
00655     }

Any ideas?

Kind regards,

Francois.

[Canesin]

You can just use my chtIcoMultiRegion solver with zero solid regions.

http://www.cfd-online.com/Forums/ope...egionfoam.html

[Fransje]

Dear Fabio,

Thank you for your reply and for the solver. But oh my! Is it Christmas coming early?

I'll look through it and come back to you if I have any questions.

Kind regards,

Francois.

[Fransje]

Dear Fabio,

I used the incompressible multiRegion solver you developed to try and run the case I was describing, but I encounter the same error...

The only modifications I made to the solver was to kick out the reading of the "K", "T", "Pr" etc variables since I was not planning on using them, and my programme compiled without errors.

I then used the same two region meshes I had defined earlier, and set solidRegionNames in the ./constant/regionProperties file to

solidRegionNames (    )

The simulation starts, creates all the fields properly, gives me a region dependent min and max Courant number, and then crashes again with the same error as in my original post...

Any ideas?

Kind regards,

Francois.

[Canesin]

Add a dummy region.. that do nothing.. zeroGradient in all boundaries.

[Jean El-Hajal]

Hi,

I had a problem with chtIcoMultiRegionFoam compilation with 1.7.x

In the file:

chtIcoMultiRegionFoam/derivedFvPatchFields/solidWallMixedTemperatureCoupled/solidWallMixedTemperatureCoupledFvPatchScalarField .C

just add #include "mapDistribute.H" like this:


#include "solidWallMixedTemperatureCoupledFvPatchScalarFiel d.H"
#include "addToRunTimeSelectionTable.H"
#include "fvPatchFieldMapper.H"
#include "volFields.H"
#include "directMappedPatchBase.H"
#include "mapDistribute.H"
#include "regionProperties.H"

maybe could help someone.

Jean