interFoam 1.6.x vs 1.5 : same mesh, same bc but different results

nuovodna · April 8, 2010, 12:03

Hi, i simulated a flow around an hull ship with interFoam (only water, no air and free surface). I have results with v1.5 and they seems really good. I tried to repeat the problem setting up a complete similar case in v 1.6.x but it gives me bad pressure and velocity field (all force coeffs are wrong and the simulation stopped after two days giving me an error ). I d like to know the differences between v1.5 and v1.6.x and if i setted in the wrong way the different params of simulation. I attach a pdf with the principal files.

EDIT: contours of p and U here

http://dl.dropbox.com/u/3617688/contours.tar.gz

g.redondo · April 13, 2010, 06:56

This is interesting. Anyone else facing this problem? Has anyone fixed it?

OF-1.6.x has some really nice features for hull analysis, but unless we figure out what is going on with the results they're useless.

Regards,

Gonzalo

g.redondo · April 13, 2010, 07:21

Having a look to your files I found a couple of things that might help, but still I don't think you'll get the same results.

1. In the inlet B.C I found that setting alpha to a constant value with setFields helps convergence.

2. You should also have an atmosphere patch.

3. Try with the hull as bouyantPressure too.

Good luck, although we're still missing something. I wish there was more documentation about interFoam in 1.6.x

Gonzalo

nuovodna · April 22, 2010, 05:46

i think i have to change boundary condition on pressure : if in 1.5 i have zeroGradient, in 1.6.x i have to switch on buoyantPressure on the patch orthogonal to pressure gradient. I ll keep you updated on simulation's results. Stay tuned

nuovodna · May 3, 2010, 08:37

I changed the pressure boundary conditions and now p field seems good. But i have a local "dicontinuity" of U field on symmetry patch. How can i solve it?? I attach the screenshots

http://dl.dropbox.com/u/3617688/Ufield.png
http://dl.dropbox.com/u/3617688/UfieldMesh.png

pablodecastillo · May 4, 2010, 05:34

Hi Emanuele,

That is because unstrured mesh (tetras) is not working nice.

Change your mesh to hexa and if possible structured and you will see everything better.

openfoam_user · June 10, 2010, 07:55

Hi Emanuele,

could you put your settings for OF-1.6.x ?

Have you improve your calculation ?

Best regards,

Stephane

nuovodna · June 14, 2010, 05:27

I have set a less value of Courant number and now it works and it's running on 4 CPUs from two weeks

openfoam_user · June 14, 2010, 05:34

Hi Emanuele,

You have a very low Courant Number !

The wigley hull of Eric Paterson runs well with Courant Number 0.8.

Regards,

Stephane.

nuovodna · June 14, 2010, 05:45

Do you think i have to investigate what are wrong in other part of the case?? I thought the mesh but checkMesh is perfect...any idea??

April 13, 2010, 06:56	Interesting	#2
g.redondo New Member GRD Join Date: Jun 2009 Posts: 28 Rep Power: 16	This is interesting. Anyone else facing this problem? Has anyone fixed it? OF-1.6.x has some really nice features for hull analysis, but unless we figure out what is going on with the results they're useless. Regards, Gonzalo

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Moving mesh	Niklas Wikstrom (Wikstrom)	OpenFOAM Running, Solving & CFD	122	June 15, 2014 06:20
interFoam with irregular Mesh	luther	OpenFOAM	9	August 14, 2009 07:43
incorrect results with a tetrahedral mesh	jcmasters	OpenFOAM	5	May 20, 2009 16:44
Mesh for 3 dim Geometry	Phil	FLUENT	9	July 12, 2000 04:39
Mesh	Mignard	FLUENT	2	March 22, 2000 05:12

April 13, 2010, 07:21		#3
g.redondo New Member GRD Join Date: Jun 2009 Posts: 28 Rep Power: 16	Having a look to your files I found a couple of things that might help, but still I don't think you'll get the same results. 1. In the inlet B.C I found that setting alpha to a constant value with setFields helps convergence. 2. You should also have an atmosphere patch. 3. Try with the hull as bouyantPressure too. Good luck, although we're still missing something. I wish there was more documentation about interFoam in 1.6.x Gonzalo

April 22, 2010, 05:46		#4
nuovodna Senior Member Emanuele Join Date: Mar 2009 Posts: 110 Rep Power: 17	i think i have to change boundary condition on pressure : if in 1.5 i have zeroGradient, in 1.6.x i have to switch on buoyantPressure on the patch orthogonal to pressure gradient. I ll keep you updated on simulation's results. Stay tuned

May 3, 2010, 08:37		#5
nuovodna Senior Member Emanuele Join Date: Mar 2009 Posts: 110 Rep Power: 17	I changed the pressure boundary conditions and now p field seems good. But i have a local "dicontinuity" of U field on symmetry patch. How can i solve it?? I attach the screenshots http://dl.dropbox.com/u/3617688/Ufield.png http://dl.dropbox.com/u/3617688/UfieldMesh.png

May 4, 2010, 05:34		#6
pablodecastillo Senior Member Pablo Join Date: Mar 2009 Posts: 102 Rep Power: 17	Hi Emanuele, That is because unstrured mesh (tetras) is not working nice. Change your mesh to hexa and if possible structured and you will see everything better.

June 10, 2010, 07:55		#7
openfoam_user Senior Member stephane sanchi Join Date: Mar 2009 Posts: 314 Rep Power: 18	Hi Emanuele, could you put your settings for OF-1.6.x ? Have you improve your calculation ? Best regards, Stephane

June 14, 2010, 05:34		#9
openfoam_user Senior Member stephane sanchi Join Date: Mar 2009 Posts: 314 Rep Power: 18	Hi Emanuele, You have a very low Courant Number ! The wigley hull of Eric Paterson runs well with Courant Number 0.8. Regards, Stephane.

June 14, 2010, 05:45		#10
nuovodna Senior Member Emanuele Join Date: Mar 2009 Posts: 110 Rep Power: 17	Do you think i have to investigate what are wrong in other part of the case?? I thought the mesh but checkMesh is perfect...any idea??