CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   CFX (http://www.cfd-online.com/Forums/cfx/)
-   -   Surfaces lost from ICEM to CFX-Pre 5.7 (http://www.cfd-online.com/Forums/cfx/20674-surfaces-lost-icem-cfx-pre-5-7-a.html)

Mark September 27, 2004 18:20

Surfaces lost from ICEM to CFX-Pre 5.7
 
Hi!

My model is a plate with a hole. One side of the plate is supposed to have a constant heat flux as boundary condition. Air should flow through the hole and take some heat of the plate. After meshing the geometry in ICEM two "CREATED MATERIAL" appear, standing for the meshed Volumes of the plate and the air. Now the two problems: In ICEM exist two surfaces (FRONT and BACK) of the plate. But when I open it in CFX these two surfaces don't exist anymore. There is just the "CREATED MATERIAL" and I can't differ between the surfaces by setting the boundary conditions.

The bigger problem is that after I started [Write a solver file] CFX creates for both domains (fluid and solid) a default boundary condition named "Default 1 Side 'domain name' Part 1 (or 2)". The type of boundary condition is interface and it is in conflict with the boundary condition that I already set to the "CREATED MATERIAL".

Please help me with these two problems.

Thanks,

Mark


Glenn Horrocks September 28, 2004 18:30

Re: Surfaces lost from ICEM to CFX-Pre 5.7
 
Hi Mark,

Try importing the mesh into CFX-Pre in two bits. That is delete all the solid shell and volume elements and import the fluid domain, then delete the fluid shell and volume elements and import the solid domain. CFX-Pre has no problems with the multiple domains then.

I also have had problems getting ICEM to successfully import multiple domains. You should be able to do it in one go but I have never bothered to find out how as the above method works fine. (Can anyone tell us how to do it?)

Glenn

Mark September 29, 2004 01:39

Re: Surfaces lost from ICEM to CFX-Pre 5.7
 
Hi Glenn!

Thank you for you quick answer. I tried it but than I have problems with the definition of the interfaces. I received following message:

ERROR

pgtInterfaceBoundaryListgen: unable to retrieve information for Boundary List1 in /FLOW/DOMAIN INTERFACE:INTERFACE: getStringList: unable to find value

There was a problem reading the physics CCL.

There was a problem reading the generic physics CCL from the file "h:/cfx/fluid_solid/FLUID_SOLID_fix.def". This could be because:

o The file does not exist or is not readable. o It does not contain the required data set,

["G/COMMANDS", "EVERY", 0]. o The data set contains invalid CCL.

How did you connected them?

Regards,

Mark

Glenn Horrocks September 29, 2004 18:24

Re: Surfaces lost from ICEM to CFX-Pre 5.7
 
Hi Mark,

In CFX-Pre, delete the interface and the surface patches and recreate the interface.

Glenn

Mark September 29, 2004 21:57

Re: Surfaces lost from ICEM to CFX-Pre 5.7
 
Hi Glenn!

I recreated the interfaces, but I received the same error like with direct import:

ERROR There are 2D regions that are used in more than one boundary condition.

Region "FRONT 2" appears in boundaries: "HEAT". "INTERFACE2 Side S Part 2".

In the CFX manual on page 284 (boundary conditions -> boundary details: interface) it is said that no other boundary condition than conservative interface flux is possible. Do you have any idea how to overcome this problem? (I'm going to post it also on the CFX Community site)

Regards,

Mark

Glenn Horrocks September 30, 2004 18:27

Re: Surfaces lost from ICEM to CFX-Pre 5.7
 
Hi Mark,

The error message is telling you the problem. More than one boundary condition references the region "FRONT 2". The two boundaries are "HEAT" and "INTERFACE2 Side S Part 2". The fix is obvious - remove the region FRONT 2 from the boundary it does not belong to.

Glenn Horrocks

Mark September 30, 2004 23:07

Re: Surfaces lost from ICEM to CFX-Pre 5.7
 
Hi Glenn!

Thank you, sounds easy. I tried to use the ccl-file and set a source at the interface (heat flux). It works, but I have to define the heat flux coefficient. That is not realy usefull :(

Regards,

Mark


All times are GMT -4. The time now is 14:21.