the macro of c_face_loop and thread_id of interface
 

Dear all,
I study on the evaporation of liquid in the heat pipes. The evpaoration interface is not traced and fixed. Interface between liquid and vapor is thought as "interface" in fluent. Sources term of mass, momentum and heat are considered on the evaporation interface with udf.
And in the source term of udf, the macro of c_face_loop was used to access the boundary of interface. There are six faces which is relant when a pair of interface are created with couple, such as: two orignal faces(boundary type: interface), two new faces(boundary type: wall) and two faces with shadow (boundary type: wall). As the interfaces are looped with c_face_loop, thread_ids of two orignal faces are retrieved and one of other four faces aren't got. And F_UDMI(or F_UDSI) with thread_id of two orignal faces can not be stored value. these errors are shown:
Error:
FLUENT received fatal signal (ACCESS_VIOLATION)
1. Note exact events leading to error.
2. Save case/data under new name.
3. Exit program and restart to continue.
4. Report error to your distributor.
Error Object: ()

I test this problem with different version fluent(6.1, 6.3 and 12.0). The result is the same.
What's the reason of this problem?
Thanks.

First F_UDMIs can only be stored on boundary threads (e.g. walls, inlets, outlets) and I guess not on "internal" face threads like interfaces.
Second interfaces are supposed to work as contact planes between seperate zones with non-conformal grids and to do all the required momentum, mass and head transfer there. If you want to do these by your own I recommend to define the liquid-gas interface as a wall seperating two distinct zones in Gambit (Fluent will seperate it into wall and wall-shadow automatically) and apply appropriate mass-, momentum- and heat-transfer conditions by udf (may take some time) on both sides i.e. for both phases. Doing so F_UDMI should be also available there.

cheers

Thanks. I have solved this problem according to your adivice.