CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   Main CFD Forum (https://www.cfd-online.com/Forums/main/)
-   -   Flow in a 3D Pipe problem (https://www.cfd-online.com/Forums/main/196757-flow-3d-pipe-problem.html)

Hyllon December 14, 2017 13:16

Flow in a 3D Pipe problem
 
Hello guys, I'm new to OpenFOAM and to the Forums, so hey !

I'm trying to run a 3D flow in a cylinder. This is a simple case I think.

I made the geometry using FreeCAD. It's just a cylinder. (Radius 10 mm / Length: 200 mm)

I've used gmsh to discretize it. I added volume and a few surfaces (Inlet, Outlet and Outter_Surface).

I've used the files from cavity folder in icoFoam directory.

I've tried a lot of things, but the problem is the same: on Paraview the flow appears to be static.

I downloaded a mesh (3D pipe) from a video on Youtube and it worked with this cavity case, then I swapped back to my own mesh and the problem was there.

ANY information for the newbie will be appreciated =)

My full case is here:
https://mega.nz/#!c6gVhaga!ithwouOsD...EhVrYXMZEY6teo

P.S: English is not my native language.

juan4815 December 26, 2017 11:08

Well, I have no experience with OpenFOAM but your question is about the geometry basically, so I'll try to help you a little.
If no flow is present check these:
- Is the geometry only a volume? Maybe you added surface volumes around the cylinder volume (inlet, outlet, mantle) and the input flow 'stays' at the surface volume inlet.
- Check if by changing the condition (velocity, mass flow, volume flow, etc) you get the same result, accordingly.
- Does the software you used for meshing allow to check the mesh? As in quality, connections, etc. Maybe doing this will give a hint.

AliE December 26, 2017 14:52

Hello,

I think that the problem is that you are assigning an inlet velocity of (10 0 0) but your stearmwise direction in the geometry is "z" and so you should assing (0 0 10). Then adjust the numerical schemes since you are using a tetra mesh which is non-orthogonal and those used in the cavity tutorial are not the best options in this case.

AliE

Hyllon January 4, 2018 13:31

Quote:

Originally Posted by juan4815 (Post 676152)
Well, I have no experience with OpenFOAM but your question is about the geometry basically, so I'll try to help you a little.
If no flow is present check these:
- Is the geometry only a volume? Maybe you added surface volumes around the cylinder volume (inlet, outlet, mantle) and the input flow 'stays' at the surface volume inlet.
- Check if by changing the condition (velocity, mass flow, volume flow, etc) you get the same result, accordingly.
- Does the software you used for meshing allow to check the mesh? As in quality, connections, etc. Maybe doing this will give a hint.

Thank you juan for your reply.

Well not sure if I get what you meant but on Gmsh I added one single volume and then 3 surfaces (faces). I dont think I added volumes around a volume.

I checked the mesh using openfoam's command 'checkMesh' and it was ok.

Hyllon January 4, 2018 13:42

Quote:

Originally Posted by AliE (Post 676166)
Hello,

I think that the problem is that you are assigning an inlet velocity of (10 0 0) but your stearmwise direction in the geometry is "z" and so you should assing (0 0 10). Then adjust the numerical schemes since you are using a tetra mesh which is non-orthogonal and those used in the cavity tutorial are not the best options in this case.

AliE

Good call dude, thank you so much. Indeed i didnt noticed the flown was in Z direction. I have no ideia why though, how could I place it on X axis?

I did some changes onto the numerical scheme. I have " nNonOrthogonalCorrectors 1" and some " cellLimited Gauss linear 1;" on fvSchemes. And other stuff like that.

I think I did some boundary changes in polyMesh as well.

Now in paraView I have values on the inlet AND outlet, but the cylinder body still remains static. I'll upload my full case again.

And thank you, again.

https://mega.nz/#!Q2QxXI5A!LnMyDc4DT...flmJLp2qA6IXYM


All times are GMT -4. The time now is 20:20.