CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   CFX (http://www.cfd-online.com/Forums/cfx/)
-   -   upwind-high res. blending (http://www.cfd-online.com/Forums/cfx/105319-upwind-high-res-blending.html)

k.vafiadis July 27, 2012 11:19

upwind-high res. blending
 
Hello, is there any difference if someone run a upwind advection scheme, and then:
  • run a high resolution (the upwind run will be the initial value) for both momentum and continuity
  • run a high resolution for continuity - upwind for momentum (the upwind run will be the initial value) and then a high resolution for both momentum and continuity (the high resolution for continuity - upwind for momentum will be the initial value).
even if they are coupled I think that they behave different when they use different schemes and that affects the way the simulation reach convergence :confused:

Felggv July 27, 2012 13:34

I believe you can find something about it in the section Wiki CFD here at this website.

Have you read the Ansys Help Documentation?

k.vafiadis July 27, 2012 13:56

I already read the documentation but I can't find something.
I would like any suggestions from the cfdonline community experience ;)
I already run a simulation about this and I will post my conclusions!

ghorrocks July 28, 2012 08:07

The documentation describes all this. But for a fully converged steady state solution, in general it does not matter what the initial condition was. A good initial condition will converge faster, but the end result is the same. A small number of flows are bimodal (or multi-modal) and the intiial condition determines which final state they end up in.

But the final converged solution will be different if you use different advection schemes.

Note high res is not an available option for continuity. Continuity does not advection so has no need for an advection scheme. The high res option only applies to momentum and scalar equations. The important option for continuity is what P-V coupling is applied.

k.vafiadis July 29, 2012 09:23

I agree about the continuity equation. And thank you for your considerations! But what made me have this confusion was that:
solver control>equation class settings?equation class: continuity
has the option for an advection scheme and moreover it has the three choices used for advection in CFX, upwind, high resolution and specified blend factor.
http://i179.photobucket.com/albums/w...sch-upwind.jpg
My testing with this gives that the advection scheme for continuity does nothing...


All times are GMT -4. The time now is 11:17.