I have to calculate a 2d inkompressible flow with fixed transition points on an airfoil. Does anyone know how to realize forced transition in STARCD?

Hi Dirk,

Without any comment if it is usefull to do forced transition or not I would suggest to use the multiple refference frame option for that in a first step, so you can specify the physical models for each material. Try it maybe it works for your purpose. If it does not work I can imagine that there is a way to use a user subroutine to specify the turbulent paramters for say a cell type or so.

Hi Klaus,

thank you for your response. I just tried to put different material types to the laminar and the turbulent zone, but it didn't work. I got an errormessage saying you can't use two different materials in one flow without anything between them.

Another attempt was to put the turbulent viscosity of the laminar zone to zero in subroutine VISTUR while calculating the whole flow with a turbulence model. This works if you turn on swith No.12. Although the subroutine was definitly used for all laminar cells during the calculation (it wrote a control output), the result already showed no influence on turbulent viscosity and turbulent energy.

Dirk

Hi Dirk,

for the multiple refference frame you have to specify two materials with different properties. For the interface (the cells at the interface must not share the same vertices) you have to specify an pressure and a inlent boundary condition. Have a closer look at the manual under the chapter multiple refference frame explicit formulation.

Without knowing it exactly I can imagine that the turbulent viscosity routine will only take place under the const. turbulent viscosity usage.

servus, klaus

I think it may work by using the following:

1. set up different cell types for the laminar flow part and turbulent part (since it's forced transition, this is assumed to be known).

2. Activate user code for k and e so that you can fix these values to very small, also make sure that k < e so that turbulent viscosity Mt=Cm * density * k**2/e is at least one magnitude smaller than molecular viscosity.

Dear J.Y.Luo

Thank you very much for your advise, it works!

Dirk Steudel