Hello Maddalena,

Thanks for this test case.

Have you tested to comment the line in fanFvPatchFields.C ?
As Mike pointed out, this is the only difference between 1.5-dev and 1.6.x/1.7.x

Olivier

Hi,
no, I did not tried. However, if it has been added for stability reason, I think it is a good idea to not modify it, and define patches properly...

One more question on the subject.
I am experiencing hard time to simulate a closed loop cooling system which includes a fan. I set the fan direction as described above: choose the master and slave face as I want the air moves and set the f coefficient as 1(10). PotentialFoam confirms that my settings are OK, as shown in fan_0.png. However, when using simpleFoam (with turbulence off), the flow reverses its direction and, at time 75, it is completely on the opposite direction I want it to go, see fan_75.png. :(
I also noticed that the continuity error is a little bit too high: at time 75 I have:
This let me think that I have some problems with BC, and since all the other domains are walls or standard cyclic, I lay on the idea that I still miss something on the fan BC.
Any ideas on the reason of that?

thank you for any idea or suggestion.

mad

For everyone that will meet the same problem: the fan BC is very sensitive to the mesh quality in proximity of it. Although having a good tet mesh, the simulation never converged. The solution was to use a hexa mesh in the proximity of the fan.
Hope this help someone.

mad

How is velocity defined?
 

Hi all,

I'm using the fan bc, which is proving very useful for me. However I was wondering given the case where the velocity varies across the boundary condition which value of V is used to calculate the pressure drop (given the inputted polynomial coefficients). Is it an average value?

Thanks for your help!

what is going wrong?
 

Hello,
It seems like that problems I thought solved reappears suddenly in these days… :confused:
As usual, velocity field at the fan interface looks unrealistic: wiggles and vectors going in the wrong direction. However:

1. f coefficient is positive:1(10) -> why? see here
2. I have got hexa mesh at the fan interface -> why? see here;
3. Flow direction is ok according to potentialFoam and to the first time step of the simulation;
4. Domains and blocks are on order -> applies to pointwise-openfoam export, see here.

Any idea of what is happening over here? I have spent 9 hours today trying to understand what is going wrong, without success. Suggestions?

mad

Hi All,

I kindly ask You to read my thread http://www.cfd-online.com/Forums/ope...am1-7-1-a.html.

Thank You in advance,

Claudio

hello everyone

I'm new to the use of OpenFOAM, I have installed on my computer OpenFOAM 2.0.1, and I'm trying to model a bedroom with a fan in the center, the flow enters through a window and out through a door. Download the example of the fan, the compiler creates in the boundary conditions and fan_half1 fan_half0 patches and not that I give these values ​​in the file 0 / p

would greatly appreciate your answers
cordially
Cesar

Hello,
starting from Openfoam 2.0.0, a new BC has been implemented in OpenFOAM, see http://www.cfd-online.com/Forums/ope...essure-bc.html thread.
maddalena.

thank you very much Maddalena

If anyone could help me more, they are most grateful.

I have also problems with the creation of the mesh in OpenFOAM, I can not define the inner patch to model the fan, try making a mesh of ideas, but when using the tool converts the mesh buts ideasUnvToFoam lose some volume elements, could you help me again by favor.

thanks

Cesar.

Hi Cesar
I am sorry but I have no experience on the subject. Why do not you open a new thread here: http://www.cfd-online.com/Forums/openfoam-meshing/ ?
mad

Hello

I wanted to run this case posted above"fan.tar.gz"