| LuckyTran |
November 8, 2021 09:20 |
Quote:
Originally Posted by LoGaL
(Post 816002)
I am not sure, the no slip condition is applied at the cell face, not at the cell center, where the velocity is actually going to be non zero. When Fluent calculates the heat flux through the wall, it uses cell centered values to reconstruct the gradient, so the conductivity is going to have a turbulent contribution.
|
Hence the need for clarification because otherwise we end up calculating things that should not even be calculated. No need to even mention flux reconstruction because that would involve properly calculating things the correct way which should be zero regardless of what the unlimited or limited reconstruction gradients even are.
To get the turbulent heat flux field (which is a vector field) you need to fetch the temperature gradient then multiply it by the turbulent viscosity and divide by the turbulent Prandtl number. Depending on your definition of turbulent heat flux (you may/may not need to divide also by the density). You can create custom field functions or a UDF.
If done using a custom field function, the temperature gradients aren't available by default. You must enable the option to retain them in memory during calculation. There are three components of the temperature gradient dT/dx, dT/dy, and dT/dz so you will need 3 custom fields. Here's a cute YT video on how to retain the temporary storage and includes the temporary gradient as an example. |
