[Ya_Squall2010]

I am revisiting the floatingObject tutorial in OpenFOAM-3.1.x-Ext, and it seems to me that it still crashes after around 2 sec in parallel mode. It runs well all the way to the end in the serial mode though. Would be much appreciated if anyone could advise whether this long existing bug pertaining to interDyMFoam has been solved in the recent OF-ext release. Many thanks.

[Ya_Squall2010]

a bit more info:

platform: CentOS 6.5 final
compiler: GCC 4.4.7
case: /foam-extend-3.1/tutorials/multiphase/interDyMFoam/ras/floatingObject

numberOfSubdomains 4;
method scotch;

[Ya_Squall2010]

Nobody is having the same problem with me?

[wyldckat]

Hi Ya_Squall2010,

I have a couple of mental notes on this topic:

• Status on this topic sometime ago: http://www.cfd-online.com/Forums/ope...tml#post447099 - post #9
• Another bug was fixed in OpenFOAM sometime ago for this tutorial case: http://www.openfoam.org/mantisbt/view.php?id=1284

I hope you don't mind, I'll quote what we discussed via PM:

Originally Posted by wyldckat

I've added this to my to-do list. With any luck, I'll be able to look into this during the coming weekend.
I believe this is already fixed in the latest OpenFOAM version, but I don't think that the fixes have been transferred to foam-extend.

Originally Posted by Ya_Squall2010

Thanks for the reply. Yes I noticed this issued has been resolved in the foam-2.3.x but still pertains in foam-ext-3.2. Could you advise the root cause of this problem? I did some search and realized that this might have something to do with the incorrect way of pointField decomposition when a moving boundary point patch field is presented. Is that true? Thanks.

I don't think that's enough to fix the problem. In addition, there are a few problems with mapping point fields in OpenFOAM itself: http://www.openfoam.org/mantisbt/view.php?id=1530

I was going to suggest that you reported this on foam-extend's bug tracker, but I saw just now that you have already done it: http://sourceforge.net/p/openfoam-ex...ndrelease/294/


I haven't managed to look into this with more detail, but there are a few details on the release notes for 2.3.0 that be the origin of the solution:

• The primary suspect: http://www.openfoam.org/version2.3.0/mesh-motion.php
• Possible, but not very likely: http://www.openfoam.org/version2.3.0/parallel.php
• Possible: http://www.openfoam.org/version2.3.0/multiphase.php

Either way, you can access to the development history for OpenFOAM since 1.5 in the recently launched repository: https://github.com/OpenCFD/OpenFOAM-history

There is also another possibility, although this was specific to interPhaseChangeFoam: http://www.cfd-online.com/Forums/ope...tml#post568787 - post #9

As for fixing this in foam-extend, if all else fails, try signing up and participating in this: http://www.cfd-online.com/Forums/ope...ng-2016-a.html

Best regards,
Bruno

edit: Hrv answered to the related bug report on this very same day: https://sourceforge.net/p/openfoam-e...ndrelease/255/ - and I quote:

Please use tet FEM based mesh motion instead. The cell-based mesh motion uses hacks in interpolation which are so horrible that I do not think I can maintain them. Alternatively, please offer fixes in over-ride of interpolation on motion boundaries.
This kind of simulation is usually done with the Naval Hydro pack, where tet FEM motion shows it is superior. Would you consider upgrading the tutorial to a better motion solver?