i have simulated a cylinder that it has four domains. these domains are concentric,and are fluid,solid, fluid and solid respectively,one of the solids generates heat and there is heat transfer between these domains, interfaces have been modeled too,but CFD post doesn't show any heat transfer, and temperature in solid with heat generation is very high.what are your suggestion for this problem?
another question : what are the effective parameters to reach conservative target?
thanks
Hi,
my suggestion is to fix in CFX-Pre the obviously existing issues in the setup of the interface conditions between the domains. Seems like interface conditions are not properly setup.
Regarding your 2nd question: I do not have the "magic silver bullet" for CFD simulations. CFD is hard work on the cycle between:
Mesh generation --> CFD Setup --> Solution --> critical results analysis --> return to beginning of the cycle with your conclusions.
Regards,
Dr. Th. Frank.
Adding some tip to Dr. Thomas Frank ´s answer: Verify if, for each interface pair, you did have set up the correct face domain pair. It is very common select the wrong face with multidomain composition - the use of "good names" is a best practice to avoid this kind of mistake.
Best Regards!
thanks,CFX defines interfaces automatically and boundaries on interfaces are "conservative interface flux" option only , i define heat generation in a solid and boundary conditions only,but temperature in the solid is very high
Question: if you try to use the automatic pairing of interfaces, are you sure that you have assured, that (e.g. due to some issues in mesh generation process) the interfaces are really at the almost identical position? Or might be the case, that there is some small, not visible "gap" between the interfaces? This would lead to the fact, that an automatic pairing of the pair of boundary faces could fail. Can you try to manually pair the desired pair of boundary faces within the interface definition?
Because what you observe is, that there is exactly NONE interface flux, since the generated heat seems to be encapsulated within the solid and is not finding a way out from the solid to the fluid region by heat conduction. This essentially only can take place, if the automatically detected overlap of boundary faces on your interface is close to zero. By the way - have you checked the information written in the CFX OUT file for the interface definitions? Any comments there, which are providing further information about the possible source of error?
Regards,
Dr. Th. Frank.
- Badges
- Science topic
- Similar topics
- Engineering
- Thermal Engineering
- Thermofluid
- Heat Transfer