Learning Forum

[T-Ying Lin]

I'm setting up a CFX simulation for natural convection where there is a cylinder with a consistent heat flux at the center that varies axially using an equation based on z and the outer walls of the rectangular prism around it are set to 298 K. I would expect the lowest temperature to be 298 K, but it consistently produces these cold spots on the ends of the cylinder which I've set to be adiabatic. I was wondering if this was because I simply modeled this as a single body for the air and set the cylinder wall to be a the desired flux without any regard for the material of the cylinder.

When I set this up on Fluent, there is an option to set the material for the wall and, subsequently, the ends do not experience these cold spots which dip below 298K. Would I need to model this as two bodies with the cylinder and the air so that I can have a material for that wall? Or is my problem something else obvious that I'm missing? 

Thank you!

[Prashanth]

Hi,

I hope the cylinder walls are properly coupled. Also, when setting material for walls, did you use some wall thickness for it to have some effect. For more information, check 7.4.15.5.7. Thin-wall thermal resistance parameters in 23R1 user docs.

[T-Ying Lin]

I am only modeling the air around the cylinder, there is no separate body for the cylinder so I don't think the coupled walls are my issue. The wall thickness I thought  was a feature of Fluent? I didn't see an option to adjust that in the CFX gui.