CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   OpenFOAM (http://www.cfd-online.com/Forums/openfoam/)
-   -   Two regions, problem ? (http://www.cfd-online.com/Forums/openfoam/67106-two-regions-problem.html)

dipling August 4, 2009 07:15

Two regions, problem ?
 
Hello,

I have created two blocks which are side by side. CheckMesh outputs the following message:

*Number of regions: 2
The mesh has multiple regions which are not connected by any face.
<<Writing region information to "constant/cellToRegion"

Is it essential to connect them for solving ? Or is it unimportant ?


I think it is important, maybe anybody has some arguments why not.. ;)


How is it possible to connect these two Blocks ? With mergePairs ?

dipling August 4, 2009 09:33

1 Attachment(s)
For understanding i attached a draft...

block 1 dimension [x y z ] --> [0-10 ; 0-10 ; 0-50] hex (20 20 40)
block 2 dimension [x y z] --> [0-5; 0-5; -7 - 0] hex (15 15 10)

henrik August 5, 2009 02:15

Dipling,

yes, it is important to have a single region if you want the flow to go across.

yes, mergePairs is the keyword to use.

Henrik

holzmichel September 3, 2009 08:00

Hello all,

I have the same Problem like Dipling.
I merged my Meshes (rotor and stator) to use the interDyMFoam solver. When I run checkMesh I still have two faces which are not connected. Is it important for the ggi? I thought ggi will correct this, but all what I tried ends with floating point exception while velocity magnitude is rapidly increasing . Does anybody have some ideas for me?

Regards

Michél

chbenz September 3, 2009 08:16

hi michel,

i think ggi connects these 2 parts befor the fluxes are estimate. have a look at the tutorials in $FOAM_TUTORIALS/icoDyMFoam/mixer...
maybe this can help.

Christof

holzmichel September 3, 2009 08:54

hi Christof,

thank you first for your very fast reply :)
I looked a few times in that case to find out what I am doing wrong. I followed the steps like openfoamwiki describes about the ggi. I made two Meshes - Rotor and Stator (both has the same radius) and named the sliding patches insideSlider and outsideSlider. When I merged them the the rotor is circulating like I want. But from the beginning the velocity magnitude is increasing and the timesteps will become smaller and smaller (it looks like Courant Number mean: 9.60502e-05 max: 159.58 velocity magnitude: 6.73986e+81 deltaT = 1.41308e-85 GGI pair (outsideSlider, insideSlider) : 1.84927e+78 3.18538e+78 Diff = 1.77211e+75 or 0.0958272 % Time = 0.000650347) even if the rotation of the rotor is set to 0 in the dynamicMeshDict???

Regards
Michel

chbenz September 3, 2009 09:12

i dont know much about ggi because i dont use it up to now.
it seems to me, that ggi goes right because there are no failure. maybe your boundary conditions?
check this.

can you upload your case?

Christof

holzmichel September 3, 2009 10:38

1 Attachment(s)
here is the case. sets(insideZone / outsideZone), points, faces, etc. in constant I deleted because of the size. I also attached the checkMesh log in the archive. Maybe someone find out what is wrong. thank you.

Michel

hansel September 7, 2009 17:23

Quote:

Originally Posted by holzmichel (Post 228425)
hi Christof,

thank you first for your very fast reply :)
I looked a few times in that case to find out what I am doing wrong. I followed the steps like openfoamwiki describes about the ggi. I made two Meshes - Rotor and Stator (both has the same radius) and named the sliding patches insideSlider and outsideSlider. When I merged them the the rotor is circulating like I want. But from the beginning the velocity magnitude is increasing and the timesteps will become smaller and smaller (it looks like Courant Number mean: 9.60502e-05 max: 159.58 velocity magnitude: 6.73986e+81 deltaT = 1.41308e-85 GGI pair (outsideSlider, insideSlider) : 1.84927e+78 3.18538e+78 Diff = 1.77211e+75 or 0.0958272 % Time = 0.000650347) even if the rotation of the rotor is set to 0 in the dynamicMeshDict???

Regards
Michel

I'm not sure about your application, but having the rotor and stator the same radius would cause problems in my tests. I actually rotate one of the meshes and if they are actually the same size, then as one rotates, it crosses into the other's area. (Due to the circle being made out of flat sides.) I leave 0.0005 m between the rotor and stator, but that number depends on the mesh size and diameter. If your mesh is turning, you will need to allow a little space between them.

holzmichel September 14, 2009 05:26

Hello Steve,

thanks for your help. I tried your suggestion but the same error is given.
Maybe some other reasons for this problem.

Regards
Michél

hansel September 14, 2009 10:51

Holdmichel, set your update rate to a small number so you get many snapshots and try to continue the simulation. Maybe you will be able to tell where the velocity is starting to go wrong. I'm curious to know if it's on the moving interface or the edge of your test area.

I generally use zero velocity sides on my test areas to keep things under control. (Fixed velocity inlet, and a fixed pressure outlet.)

holzmichel September 16, 2009 03:57

well my case do not have an inlet and outlet. I want to transport the fluid in the case with the rotor. When I set dt very small like 1*10-7 or higher like 5*10-3 it makes no difference of the increasing of the velocity. I tried to set the rpm of the rotor to 0 to have no velocity but there the velocity also increase. it is really strange and makes no sense. I will try your idea to find it out and hope I will have a success.

Regards

Michél


All times are GMT -4. The time now is 02:23.