[omar.asurt]

Hi everybody,
I'm working with fluent ,and I want to give inlet boundary condition from another solution . How can I connect the results of this solution to a new problem as an inlet boundary condition. Also since the problems are transient, should every time step of the previous solution match the new problem's time steps or what should I do?.
Thanks in advance .

[KevinZ09]

I'm not totally sure what it is you're trying to do? Are you trying to couple two codes, like a system and CFD code, and have them work in parallel? Or are you trying to create a Fluent simulation which has a transient inlet boundary condition for which the data comes from another source?

If the former, that's quite complicated but doable. The second is probably easier, and can be achieved either by using a UDF (see UDF manual for details) or a profile file (see section 6.6 for details).

If you're more clear I can try to help you further along, or at least guide you in the right direction.

[omar.asurt]

thanks for response , but i have two models but i run them on fluent separately
, but now i wanna to couple them, i think instead of redrawing them together then mesh this bigger one , i wanna to take the output results from the first one and enter it as inlet condition in the other , that's all about , so any idea how to do that ?

[KevinZ09]

Ok, thanks for the clarification.

I know Fluent had an option to merge two meshes in Fluent 6, but I've never done that and not sure it's still available in later versions. Either way, some info can be found here:

https://www.sharcnet.ca/Software/Flu...ug/node171.htm

Alternatively, you can create a profile file of your outlet through file --> export --> profile and then read that file into your second run on the inlet boundary. However, that's not transient data. If you want transient data, you can still create a profile file but then it'll be a constant value. I don't think you can create a transient, inhomogeneous profile file. But others can maybe chime in on that?

That said, if you can't merge the meshes easily in Fluent, I'd strongly encourage you to bite the dust and just create one domain. It's some extra work, but it's much better for the simulation than trying some ad-hoc way of "connecting" the simulations.

[LuckyTran]

Originally Posted by omar.asurt

Also since the problems are transient, should every time step of the previous solution match the new problem's time steps or what should I do?

The new problem needs inlet conditions at every time-step. So either they match inherently, some interpolation takes place, or some other black magic. Inlet conditions at every time-step must be specified, that should be clear...

Originally Posted by omar.asurt

thanks for response , but i have two models but i run them on fluent separately
, but now i wanna to couple them, i think instead of redrawing them together then mesh this bigger one , i wanna to take the output results from the first one and enter it as inlet condition in the other , that's all about , so any idea how to do that ?

Taking the outlet of one simulation and specifying it as the inlet of the other simulation does not couple them! The source simulation is still completely decoupled. The second simulation is only implicitly coupled to the first simulation (one-way coupling). If you want to couple the systems and solve the two-way coupling problem, you have to do a lot more work. In that case it is much easier to just do one big simulation.

If all you are trying to do is map transient inlet boundary conditions then you should do the profile method mentioned. There is a tutorial in the Fluent help file for transient profiles. Or do a search for transient profile help, plenty of examples.

[omar.asurt]

thanks for responds , i will take the advice , and i am also working on DPM model , port injection model , if anyone can help let me know , thanks again