Learning Forum

[haibin wu]

I have some doubts about the thermal boundary condition setting of wall recently. Although this may be a very simple theoretical problem, I still want to hear everyone's opinions and answers on the thermal setting of wall in fluent. To facilitate the explanation of my question, I will use the following model for illustration. The picture is a model I took, where the red color is fluid a, the yellow color is solid device b with wall thickness of 3mm, the blue color is fluid c, and the green color is solid jacket d with wall thickness of 2.5mm. Now I want to calculate the temperature during fluid flow (maybe reverse or same direction), which requires considering the thermal boundary conditions of each layer of wall (wall of fluid a, wall of device b...). Obviously, this is too complex, and in order to reduce the number of grids and computational costs, I only want to keep fluid a. However, this has caused confusion when setting the boundary conditions for the wall of fluid a. How should the wall boundary setting of fluid a be considered? I currently want to set it to convection, is this reasonable? If it's reasonable, how should I set the parameter of heat transfer coefficient, wall thickness, Free stream temperature etc. as shown in the following figure? This has really troubled me for a long time, please answer for me, thank you!!!

[Rob]

You could do the whole lot, but that will get expensive from a mesh point of view. Are you looking at pure natural convection or axial flow? 

The convection bc is a potentially good choice, but you may also want to read up on shell conduction. But, how much does the surface area (circumference) change between the edge of a, and the outer circumference of d? 

[haibin wu]

Yes, this is a convection problem, not an axial flow problem. I also do not quite understand the setting of the heat transfer boundary conditions for the shell. My current idea is to reduce the problem in the above graph to keep only fluid A, and then set it only for the boundary of fluid A. I don't know that if I only consider the boundary of fluid A, how should I set its boundary conditions? About the surface area and perimeter of fluid a, and the outer perimeter of d, I don't know what this has to do with the thermal boundary setting of wall. Could you tell me about it?thanks

[Rob]

Why not look at 2d? You set the boundary conditions via the panel in Fluent, the values are something you need to work out. 

[haibin wu]

Thanks for your suggestion, 2d can indeed help me quickly understand the temperature change caused by the setting of boundary conditions. I will try. Thank you very much for your advice

[Rob]

And you will have enough mesh to model both internal fluid zones and any surrounding material. Note, read up on the density options and especially Bousinesq for the internal fluids if they're liquids. 

[haibin wu]

You are right, I will try. Thank you for your patience!

[Rob]

You're welcome. I also advise reading up on annular cavity flow, if you blindly follow y+ guidelines and don't resolve the mesh in the bulk region or circumferentially you'll not get a good result.