[kmooney]

Hi all,

I'm working on validating interDyMFoam and noticed a serious mass conservation issue with the supplied tutorial damBreakWithObstacle - 3D. It appears the the mass begins to decrease strongly as soon as cells are coarsened. Has anyone seen this or know of a bug? See attached graph...(x-axis is volume of liquid in domain - y-axis is time in seconds)

Thanks.

[kmooney]

Upon further inspection it appears that one of the side walls is specified as atmosphere and allows flow out of the domain. In other dam break tutorial set ups the atmosphere patch was at the top. I'm not sure if this is a bug or not.

Thanks.

[anmartin]

Hello,


I'm trying to simulate fluid sloshing inside a tank with InterDyMFoam, subjected to a sinusoidal lateral displacement. I start at 0,05 sec and stop at 58 sec. After that the fluid oscillates freely.


When i use the patchIntegrate p walls to obtain the vertical, I obtain a curve that start from -323 N a 0 sec but at the end of my simulation I don't obtain the same value. It seems like if I lose some fluid during the simulation, probably by the solution convergence.

I have tried to increase the nAlpahSubCycles, cAlpha and decrease maxCo to 0.25 but the problem still appears.
result_Fx_Fy.jpg
Can anybody tell me what is the reason to this effect,

[idrama]

Hallo anmartin,

I have just started to occupy myself with mass balance. Could you pleas explain in more detail what is the purpose of the integration of p over the wall? How can I use this information?

Cheers in advance

[openfoam_user]

Dear all,

I am running the damBreakWithObstacle tutorial in parallel using the Allrun script.

#!/bin/sh
# Source tutorial run functions
. $WM_PROJECT_DIR/bin/tools/RunFunctions

runApplication blockMesh
runApplication setSet -batch createObstacle.setSet
runApplication subsetMesh c0 -patch walls
runApplication setFields -latestTime
runApplication decomposePar
runApplication /opt/mpich/bin/mpirun -machinefile machines -n 4 interDyMFoam

The case seems to run fine.
I just notice 1 strange thing.

The folders 0.021 0.041 0.061 ... are present without doing the command reconstructPar. Is it normal ?
Best regards,
Stephane

[anmartin]

Hello Claus,

I have to integrate p over wall because I want to compare OF results with experimental data in order to validate the total sloshing forces and roll moment. In my test I measure the vertical, lateral and roll moment of a fuel tank filled with water which is excited with a sinusoidal lateral excitation.

Best regards,
Nueva imagen.jpg