CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   FLUENT (https://www.cfd-online.com/Forums/fluent/)
-   -   BoundaryLayer with blowing (https://www.cfd-online.com/Forums/fluent/27820-boundarylayer-blowing.html)

 Max Nutini August 7, 2000 13:39

BoundaryLayer with blowing

How do I better simulate with Fluent a Boundary Layer with blowing or suction? Any hints?

Thanks

Max

 John C. Chien August 8, 2000 11:58

Re: BoundaryLayer with blowing

(1). I think you are talking about the boundary conditions on a wall. (2). When you have fluid going through the wall, there must be a hole or a slot. In this case, the blowing will be an inlet on the wall. And the suction will be an exit on the wall. (3). If you have arrays of holes on the surface, similar to the turbine blade cooling holes, then you need to handle each as an inet. I have done the simulation of 2-D flow over a turbine blade with ten or twelve slots at the wall. Each slot represents an inlet and requires many cells inside to provide the needed resolution. (4). The problem with the holes is that they are 3-D and will take a lot of memory to simulate many holes. Other than that, there is no basic limit to it. (5). The other extreme is the porous plate wall. In this case, what I did on my code was simply set the v-velocity to the desired value, keeping the rest of the wall condition the same. I have not try this type of porous wall using Fluent, so, I don't know whether it is easy to do or not. (6). By the way, porous wall blowing will also change the nature of the turbulent boundary layer flow. So, it is likely that the turbulence model will have to be changed also.

 Max Nutini August 9, 2000 05:21

Re: BoundaryLayer with blowing

Thanks John for your answer. I am glad to have received an answer from you! Yes, I think that that the critical part will be modelling the boundary layer developing on the blowing surface in an appropriate way. Actually, I can obviously not trust in the standard wall functions, I am wondering whether the "2layer" submodel - which as I understand does not model any wall function but solves the flow up to wall surface - can provide reliable results. - Anyway, Trusting is a basic issue in CFD... ;-) - Does anybody have experience at this regard with Fluent? Are UDF appropriate for that? Thanks everybody! Max

 All times are GMT -4. The time now is 13:08.