Learning Forum

[rchristensen]

Hello,
I am trying to model melting of a phase change material (PCM) which is embedded with finned tubes. One unit cell of the entire geometry looks as follows. The highlighted part is the finned tube (SS 316) and the surrounding six blocks are the PCM. The innermost circle represents the heat transfer fluid that deposits the heat into the PCM for melting. The gravity acts axially (into the screen), therefore the flow is also symmetric. 

As my geometry is symmetric, I decided to divide it into six parts and just use 1/6th of the model for speeding up the simulations (see below). The flow rate through 1/6th of the circle now is 1/6th of the original value. My concern is with the boundary conditions on the edges now. 

Would a symmetric boundary condition be the appropriate choice for the highlighted faces (including the other side), along with a zero heat flux? Or do I need to have any additional boundary conditions. 

 

I would appreciate your inputs and advice. 

Thank you,
Rich

 

[DrAmine]

Symmetry is fair enough if no swirl. If swirl I will go for periodic boundary.
In your case no flow is crossing the edge? If yes then I would apply symmetry.
Neverthless even if the configuration is symmetric, flow can be asymmetric.

[rchristensen]

Thank you, Amine. 

In my case, no flow is crossing the edge. Also, the flow within the tube is fully developed laminar. So I think I will stick with a symmetric boundary condition, and see whether the results I get make sense. 

Regards,
Rich