CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   FLUENT (https://www.cfd-online.com/Forums/fluent/)
-   -   Mesh Indepedence Test in Ahmed, Drag Coefficient still going down (https://www.cfd-online.com/Forums/fluent/196503-mesh-indepedence-test-ahmed-drag-coefficient-still-going-down.html)

treviusss December 6, 2017 09:38

Mesh Indepedence Test in Ahmed, Drag Coefficient still going down
 
Welcome,

I am trying to compute flow around Ahmed body in 2D and 3D in Fluent 18.1. I have done triangle mesh with inflation layer around body and refinement zone close to the body to capture all the vortices near wake and so on.

Drag coefficient is the factor that I want to monitor to see if the solution is independent of the mesh

For 2D case I've done several refinement of mesh and it went to level around Cd = 0.296 and flattened out so I considered this as proper mesh

For 3D case for every refinement I'm making it seems like Cd is still going down for example:
0,292 for 3 400 000 elements mesh
0,290 for 4 100 000 elements mesh
0,286 for 5 500 000 elements mesh
0,28 for 7 300 000 elements mesh
(I'm using symmetry plane in half of the Ahmed model to reduce mesh elements)
During it I'm trying to refine both mesh in refinement zone and outside it respectively but I can't do it more because my PC is running out of RAM

I assume I'm doing something wrong but I can't figure out what is it
For Fluent settings I'm using
-Steady state
- Realizable k-epsilon turbulence model with Non-Equilibirum Wall Function
-Inlet Velocity is 40m/s with 1% Turbulent Intensity and 10 Turbulent Viscosity Ratio
-Pressure Outlet is 5% Turbulent Intensity and 10 Turbulent Viscosity Ratio
- Solution Methods : Coupled, Least Squares Cell Based, Pressure Standard, First Order Upwind for Momentum, Turbulent Kinetic Energy and Dissipration Rate for around 50 first iterations then I'm changing it for Second Order Upwind
- Controls : 50 Flow Courant Number and 0,25 for Momentum and Pressure
- I'm initializing using values for k and epsilon from Velocity Inlet and 0 Velocity (according to http://www.southampton.ac.uk/~nwb/le...ice_Ver1_2.pdf page 13)
- All other settings should be default

If anyone knows how to do it properly or any kind of piece of advice I would be really grateful and sorry for this chaotic style of writing.


All times are GMT -4. The time now is 07:34.