CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   FLUENT (https://www.cfd-online.com/Forums/fluent/)
-   -   Hypersonic flow over double wedge (https://www.cfd-online.com/Forums/fluent/116467-hypersonic-flow-over-double-wedge.html)

marilyn April 19, 2013 16:41

Hypersonic flow over double wedge
 
Hi! I want to analyze the shock interactions in a hypersonic flow over a double wedge. My Mach number is 9. I generated an unstructured mesh with about 2lakh elements. I chose such a large number because my instructor says that the mesh has to be very refined for such flows.I am new to CFD and Fluent and I am not sure in how many iterations my solution should converge. I set the residuals to 1e-5. Even after about 80000 iterations, the residual was still at 1e-2. Is this natural or am I doing something wrong? Also, how long do these computations usually take? Please help. Thanks a lot.

diamondx April 19, 2013 16:55

A not converged solution can be caused by various setting ? probably your mesh, probably you boundary conditions. have you followed any tutorial before ?

marilyn April 19, 2013 17:34

Quote:

Originally Posted by diamondx (Post 421819)
A not converged solution can be caused by various setting ? probably your mesh, probably you boundary conditions. have you followed any tutorial before ?

I could not get any tutorial specific to double wedge flow problems. But, I followed a tutorial on supersonic flow over a wedge. I think my boundary conditions are also right. I have set the wedges to wall boundary condition, otlet to pressure outlet and the farfield boundary to pressure farfield. As far as the mesh is concerned, I perform the grid check operation whenever I start the simulation. It has never shown any error.

diamondx April 19, 2013 17:50

good, how about your mesh ? are working in 3d or 2D ?? structured or unstructured ? i have work with supersonic flow, they are very very difficult to converge, did you run the simulation with 1st order then switched to second ?

marilyn April 19, 2013 17:56

Quote:

Originally Posted by diamondx (Post 421834)
good, how about your mesh ? are working in 3d or 2D ?? structured or unstructured ? i have work with supersonic flow, they are very very difficult to converge, did you run the simulation with 1st order then switched to second ?

It is a 2D mesh. and a very simple geometry. It is an unstructured mesh with local refining around a vertex. I directly ran it in the second order.

diamondx April 19, 2013 18:04

deciding what order should depend on the nature of the flow, and not the simplicity of the geometry. hypersonic should be difficult to predict. i suggest you start with a density based, first order see what you get...

marilyn April 19, 2013 18:07

Quote:

Originally Posted by diamondx (Post 421837)
deciding what order should depend on the nature of the flow, and not the simplicity of the geometry. hypersonic should be difficult to predict. i suggest you start with a density based, first order see what you get...

shall try doing that and see what happens. thanks a lot for your help :)

Far April 20, 2013 15:26

https://confluence.cornell.edu/displ...+Specification


https://confluence.cornell.edu/displ...w+Over+a+Wedge


All times are GMT -4. The time now is 06:15.