Learning Forum

[Ashen Perera]

Hi ,

I have a model of a shell and tube heat exchanger where the shell side flow is what is resolved. I have a few doubts I would like some advice as I am still at very basic level of CFD : 

1. For example : I have set my global maximum element size as 2 mm , then I apply face sizing to the shell wall (2 mm) and also to the tube wall ( 1.8 mm ) : how should I decide on the element sizes for the face sizing when refining. 
2. I would like to conduct a grid independent study, but I am still unable to obtain proper texts which provide advice on how a mesh independent study should be conducted. 
3. Also a few doubts , initally i had around 900,000 elements and my solution picked up transience , I applied sizing as above and refined after which the transient got removed ( 4.0 million elements ) is this usual. 
4. Should I apply inflation ?

Sorry for so much questions, I am trying to figure out CFD still .

TIA,

Ashen 

[SRP]

Hi,

For all the questions related to meshing, first I suggest you to check ansys innovation course on meshing: Mesh Generation using Ansys Fluent Meshing | Ansys Courses

This will help you to understand meshing properly.

And to answer your next question related to mesh independence: 

1) Run the initial simulation first on your original mesh and check that the residual error has converged, the monitor points are stable, and the imbalances are less than 1%. If not, repeat the mesh refinement process.

2) Refine the mesh globally such that the cells are finer over the entire domain after your initial simulation has satisfied the above convergence conditions. In general, we would strive for a mesh size that is about 1.5 times the initial mesh size.

3) Compared to the numbers monitor points values from both the mesh. If they match initial mesh was accurate enough to record the outcome. If not, re-fine the mesh