BUG: decomposePar changing BCs

shock77 · November 16, 2020, 15:28

Hi,

I have noticed that decomposePar changes the boundary conditions, which is very annoying if you want to use many cores.

For e.g. it changed:

Code:

    outlet
    {
        type            waveTransmissive;
        field           p;
        psi             thermo:psi;
        fieldInf        101325;
        gamma          1.4;
        lInf            1;
        value           uniform 101325;
    }

To:

Code:

    outlet
    {
        type            waveTransmissive;
        gamma          1;
        fieldInf        101325;
        lInf            1;
        value           nonuniform 0();
    }

If I use gamma 7/5 I get gamma 7 in the end.

Can anyone tell me how to solve this problem? I use openfoam4.1.

olesen · November 18, 2020, 17:32

That is a fairly old version.
You need to see if this problem still exists in a newer version (eg OpenFOAM-v2006) if does then go to the issue tracker to describe the problem and provide a small test case.

shock77 · November 18, 2020, 18:16

I am using the version of openfoam.org.

I have tried the same with openfoam 5 and openfoam7. The issue remains the same.

olesen · November 23, 2020, 14:49

Quote:

Originally Posted by shock77

I am using the version of openfoam.org.
I have tried the same with openfoam 5 and openfoam7. The issue remains the same.

It would nonetheless be very nice if you could open an issue on the openfoam.com issue tracker with a small test case. If the problem does not occur in OpenFOAM-v2006 that would be good. If it does occur, it would be good to fix it. Just put some text in the issue like "error occurs in openfoam 7, don't know if it occurs in OpenFOAM-v2006 but @mark said that I should report with a small test case". Can see how it goes from there.

Cheers

shock77 · November 24, 2020, 04:39

Hi oelsen,

I reported the bug on https://bugs.openfoam.org/rules.php.

As it seems the bug occurs only on my computer, what is really odd, since it occurs with different versions of openfoam. What also is important, is that you have to use 1.4 and not 7/5, because otherwise its not recognizes as 1.4.

November 16, 2020, 15:28	BUG: decomposePar changing BCs	#1
shock77 Senior Member Join Date: Dec 2019 Posts: 215 Rep Power: 8	Hi, I have noticed that decomposePar changes the boundary conditions, which is very annoying if you want to use many cores. For e.g. it changed: Code: outlet { type waveTransmissive; field p; psi thermo:psi; fieldInf 101325; gamma 1.4; lInf 1; value uniform 101325; } To: Code: outlet { type waveTransmissive; gamma 1; fieldInf 101325; lInf 1; value nonuniform 0(); } If I use gamma 7/5 I get gamma 7 in the end. Can anyone tell me how to solve this problem? I use openfoam4.1.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
mapFields major bug	alchem	OpenFOAM Bugs	14	September 15, 2023 13:48
LEMOS-2.3.x inflow generator bug with decomposePar	chenyr10	OpenFOAM Pre-Processing	1	November 21, 2017 13:31
Bug in Workbench CFX	Pierre1	CFX	6	August 2, 2017 01:18
checkMesh and decomposePar crashes with salome IDEAS mesh	pajofego	OpenFOAM	0	June 23, 2011 09:43
Bug or a feature of OF15	rafal	OpenFOAM Bugs	5	July 25, 2008 06:25

November 18, 2020, 17:32		#2
olesen Senior Member Mark Olesen Join Date: Mar 2009 Location: https://olesenm.github.io/ Posts: 1,715 Rep Power: 40	That is a fairly old version. You need to see if this problem still exists in a newer version (eg OpenFOAM-v2006) if does then go to the issue tracker to describe the problem and provide a small test case.

November 18, 2020, 18:16		#3
shock77 Senior Member Join Date: Dec 2019 Posts: 215 Rep Power: 8	I am using the version of openfoam.org. I have tried the same with openfoam 5 and openfoam7. The issue remains the same.

November 24, 2020, 04:39		#5
shock77 Senior Member Join Date: Dec 2019 Posts: 215 Rep Power: 8	Hi oelsen, I reported the bug on https://bugs.openfoam.org/rules.php. As it seems the bug occurs only on my computer, what is really odd, since it occurs with different versions of openfoam. What also is important, is that you have to use 1.4 and not 7/5, because otherwise its not recognizes as 1.4.