CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   FLUENT (http://www.cfd-online.com/Forums/fluent/)
-   -   User-Defined Scalar convective flux (http://www.cfd-online.com/Forums/fluent/28273-user-defined-scalar-convective-flux.html)

 Ola Nordblom March 6, 2001 05:12

User-Defined Scalar convective flux

Hello,

I'm trying to simulate the transport of a passive scalar in a confined region with a free surface separating water and air. The free surface is modelled using Fluent's VOF model. Scalar transport specified as a species mass fraction is not an available option when using the VOF model, so I have to specify a User-Defined Scalar. One problem is, however, that some of the tracer mass will be lost due to convection across the boundary between water and air. One way to prevent this transport would be to modify the convective tracer flux with a user-defined function. The local value of the volume fraction of water could then be used to cancel the convective tracer transport to the air. The UDF manual gives no information on how to use the appropriate DEFINE macro "DEFINE_UDS_FLUX", perhaps someone can give me some advice here or a UDF example.

Ola

 Rüdiger Schwarze March 15, 2001 05:14

Re: User-Defined Scalar convective flux

DEFINE_UDS_FLUX (f,t,i): f is the index of the face on which the flux is required, t points to the thread of the face, and i is the index of the user-defined scalar. The return value of the function is the flux through face f. Argument types are: face_t f, Thread *t, int i, return value is real.

 Ola Nordblom March 19, 2001 10:41

Re: User-Defined Scalar convective flux

Thank you for your answer. I have not found out yet if the return value of the function DEFINE_UDS_FLUX should be the mass flow rate, i.e. rho*u or if it should be the flux of the scalar, i.e. rho*u*phi. Should the flux be multiplied by the face area? Do you have any idea about this? Many thanks, Ola

 Rüdiger Schwarze March 19, 2001 10:59

Re: User-Defined Scalar convective flux

Have a look at equation (8.10.3) of the Fluent manual in order to see how fluent treats a user-defined flux.

 All times are GMT -4. The time now is 01:24.