[Tal Eluk]

Hi all
I'm using "Shell conduction" with a non zero "wall thickness" in a pipe wall
in order to simulate an isolating layer around the pipe. (fluent 14.5 -3D).

the general question is: is fluent building the shell cylindricaly around the pipe (as i hope it dose) or is this option works only for Cartesian geometries?

more specifically:
if I have a 0.1 [m] radius pipe, and i'm specifying a 0.1[m] wall thickness to the "shell conduction", will it build a cylindrical shell from r=0.1 to r=0.2 around the pipe?

Thanks,
Tal.

[LuckyTran]

Good question. A virtual mesh is generated when you use the shell conduction feature. But, that mesh does not have the usual mesh properties (x,y,z coordinates, etc). It is also displayed at the same location at the original wall. This leads me to think that the shell conduction model is still a thin-wall model (using 1d calculations), which seems to make sense otherwise you should just grid the region explicitly. Just my guess, since without having the mesh coordinates and cell properties it would be hard to debug and figure out exactly.

[Tal Eluk]

thanks for your reply LuckyTran.

creating the wall myself, and meshing it myself will solve the problem. but...
- I want to avoid recreating the geometry.
- i want to know if i can count on fluent to do it for me

[magelano]

Originally Posted by Tal Eluk

the general question is: is fluent building the shell cylindricaly around the pipe (as i hope it dose) or is this option works only for Cartesian geometries?

Sorry for the late reply, however it still may be useful for someone.

I've tested this creating many layers and watching the temperature distribution along the height of the virtual solid, and conclusion is yes, shells are built cylindrically on a pipe.