creating transparent boundary Navier Stokes

The graphical user interface of Elmer
Post Reply
Gary R
Posts: 162
Joined: 12 Apr 2012, 07:23
Antispam: Yes
Location: Long Beach CA, USA

creating transparent boundary Navier Stokes

Post by Gary R »

This is a simplified version of my problem. The actual problem is a Navier Stokes gas flow of 2 gasses mixing with complex shaped boundaries. I have not been successful with setting the boundary condition on the original problem. I would like to be able to break up the problem into separate "zones?" and make the boundaries between them transparent. The finished analysis of the simplified version is shown below.

The .brep and .cif files are attached.
Any help will be sincerely appreciated.

Gary R
Attachments
ScreenshotTestAnalysis.png
ScreenshotTestAnalysis.png (21.81 KiB) Viewed 1141 times
case.sif
(2.4 KiB) Downloaded 96 times
NavierStokesTest.brep
(6.17 KiB) Downloaded 99 times
kevinarden
Posts: 2237
Joined: 25 Jan 2019, 01:28
Antispam: Yes

Re: creating transparent boundary Navier Stokes

Post by kevinarden »

You are getting that result because their are coincident nodes at the boundary between the bodies. Body 1 and body 2 need to share the same nodes for the boundary to be transparent. Merge coincident nodes in the mesh and it will work.
flow.png
(49.95 KiB) Not downloaded yet
Gary R
Posts: 162
Joined: 12 Apr 2012, 07:23
Antispam: Yes
Location: Long Beach CA, USA

Re: creating transparent boundary Navier Stokes

Post by Gary R »

Thanks for the reply Kevin. I corrected the node problem in my freecad sketch and re built the Elmer files. The problem persists. In addition, for some reason, the sif file now contains a Helmholtz solver that errors out at the very end of the run. It complains about not being able to find the frequency which makes no sense. I've tried to remove it manually.I've double checked all of my model settings but when I reload the project the the Helmholtz solver is reinserted into the cif file. Is this normal?

I am not at all sure what to use for a boundary condition for the input to the second solver. I used velocity 2 = 0.0 but do not think that is correct.

I've attached the new .brep and cif files.

Gary R
Attachments
case.sif
(3.34 KiB) Downloaded 97 times
NavierStokesTest.brep
(6.17 KiB) Downloaded 97 times
kevinarden
Posts: 2237
Joined: 25 Jan 2019, 01:28
Antispam: Yes

Re: creating transparent boundary Navier Stokes

Post by kevinarden »

The coincident nodes are still there. The Helmholtz equation would be on if the Helmholtz equation tab was checked active.
active.png
active.png (23.4 KiB) Viewed 1117 times
Rich_B
Posts: 421
Joined: 24 Aug 2009, 20:18

Re: creating transparent boundary Navier Stokes

Post by Rich_B »

Hello,

The following forum post may help you with coincident node removal:

viewtopic.php?p=25772#p25772

It may help if you zip and upload the ElmerGUI project folder.

Rich.
Post Reply