CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   FLUENT (http://www.cfd-online.com/Forums/fluent/)
-   -   fluid-fluid or solid-solid interface (http://www.cfd-online.com/Forums/fluent/120303-fluid-fluid-solid-solid-interface.html)

howhs July 4, 2013 05:00

fluid-fluid or solid-solid interface
 
Hi people!

I have some problem regarding fluid-fluid or solid-solid interface. From the ANSYS Mesh, I have assigned names for both sides of the domains (either both fluids or both solids). Hence, when i import it into Fluent, under the Mesh interface, there is a need for me to assign the interface so that fluent will know which face of each domain will be in interface with which face of another domain. After assigning, fluent will auto create 4 walls (including shadow walls). I have difficulty inputting the parameters in the 4 walls since the options given for the thermal conditions (heat flux, fixed temp, convection, radiation or mixed) does not have conduction (for solid-solid interface) or for fluid interface which is not applicable to any of the thermal conditions. In terms of materials, it seems like we can only choose solid materials but not fluid materials, which does not make sense for the fluid-fluid interface.

It would be greatly appreciated if anyone could assist.

Thank you!

flotus1 July 5, 2013 09:45

Why would you want a third fluid at the interface between two fluids?
From a physical point of view, that does not seem to make much sense.

A thermally conductive interface is achieved with a solid of a specified thickness as an interface condition.

howhs July 5, 2013 10:04

Thank you for your reply.
I see what you mean. But if lets say there is only interaction between the 2 fluid domains, can i just assign the thickness as zero, stating any kind of materials since there is no solid thickness at the fluids interface and at the same time, assigning the heat flux to be zero. It does not seems to make sense that heat flux is zero since it will induce adiabatic, but i cant seem to find any other options from the thermal conditions.

flotus1 July 5, 2013 10:49

1 Attachment(s)
Attachment 23211

Using the coupled option at the interface wall, you can either have a perfectly conducting wall (set thickness to zero) or a wall with a specified thermal resistance (choose appropriate material and thickness).
Using the heat flux option instead you can simulate an adiabatic wall (zero heat flux).
Is one of these options appropriate for your simulation?

howhs July 5, 2013 12:47

Thank you for the reply.
I see your point. However, whenever we have 2 faces in contact with each other (be it a solid or fluid domain), we will have to set mesh interface. Once the interface is set, Fluent will create 4 walls (one pair of walls, eg wall-1 and wall-1 shadow, and 2 additional walls). For the pair of walls (wall-1 and wall-1 shadow), we will be able to set the coupling. For the other 2 walls, we can choose the thickness of the materials (like what you have mentioned) and at the same time, choosing the thermal conditions. So i was trying to ask if we can set the thickness to zero (to allow conduction to take place if 2 domains are solids) and the heat flux as zero at the same time. I am assuming here that the heat flux will only be presence if there is any thickness. I am assuming that since the other thermal condition options does not make sense for conduction to take place. I did a verification by running a very simple model (2 cuboids, with one surface with heat flux), and it seems like heat can pass through the interface, showing that there is no adiabatic condition. I think it would be better for me to show the results, but i am out of office right now, i apologize for that! I would like someone to verify this with me so that we can better understand how Fluent works!

So that was about conduction, what if we have 2 fluids in interface with each other? I assigned the inputs (for the 4 walls as mentioned) and did another case for 2 fluid domains. From what i have got from the results, it seems like the temperature vectors does not cross over from one fluid domain to the other domain. This shows that if i set the heat flux as zero, there will be adiabatic at the interface. It would be great if anyone could asisst in this!

Greatly appreciated!

Thank you.


All times are GMT -4. The time now is 17:16.