CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   CFX (http://www.cfd-online.com/Forums/cfx/)
-   -   what is the cfx5.6 cell and face data structure? (http://www.cfd-online.com/Forums/cfx/20870-what-cfx5-6-cell-face-data-structure.html)

 Erica December 17, 2004 06:22

what is the cfx5.6 cell and face data structure?

hi every one

i have a question about the conjugate heat transfer and multiphase flow. I want to know some cell info on the interface. the cell center is on the interfae? give some advice! BR

 Glenn Horrocks December 19, 2004 17:28

Re: what is the cfx5.6 cell and face data structur

Hi,

No, all variables are stored at the nodes. This also applies at interfaces.

Glenn Horrocks

 Erica December 20, 2004 04:36

Re: what is the cfx5.6 cell and face data structur

Hi Glenn Horrocks

Thanks ! you mean that it solve hydrodynamic euations with a half volume near the interface both in solid and fluid zone?but how to couple the energy equation to determine the temperature on the interface nodes? if i can find two centers beside a face, and faces around a center? Merry christmas! Erica

 Neale December 23, 2004 22:46

Re: what is the cfx5.6 cell and face data structur

That's right. There are 1/2 control volumes at the interface. 1/2 a control volume on the fluid side is connected to a 1/2 control volume on the solid side, if it's a 1 to 1 mesh connection. If it's GGI then a 1/2 control volume on the fluid side could be connected to more than 1 half control volumes on the solid side (or vice versa, it depends on the mesh density).

There is documentation on how the implicit coupling is done in the solver theory documentation. But, the 1/2 fluid control volumes are implicitly coupled to the 1/2 solid control volumes. The two temperatures are independent (fluid and solid) are are calculated such that the flow from the solid to the fluid (or the other way, it does not matter) is exactly conserved. i.e. if 10 W flow out of the solid then 10 W go into the fluid control volume(s) it is connected to. This condition is maintained in the coefficient matrix, there is no lagging or explicit treatment.

Neale

 All times are GMT -4. The time now is 00:58.