inletOutlet boundary condition with variable inlet value
 

Hi Foamers

I was planning on using the BC inletOutlet in order to limit influx velocity into my domain and to my understanding (please correct me if I'm wrong) "value", which is applied once the flux is pointing inward towards the domain, is always applied as fixedValue, meaning that once the flux is pointing inward and I set a value of, say (1 0 0) where x1 is the surface normal, the velocity vector applied will always directly jump to (1 0 0) and might cause unrealistically high velocities.

Now what I need is a boundary condition that will behave like zeroGradient in both directions up until "value" has been reached from whereon it will behave like fixedValue for the inward direction.
Is such a BC already available, or am I out of luck?

Edit: Apologies for putting this into the wrong subsection, this should probably go under "Pre-Processing".

Hi,

I hope that I understood your question correct:
Your problem is, that you would like to define an already developed velocity profile at your inlet instead of the having a constant "wall" of velocity defined by something like (0 0 1)?

Probably it is possible to define something like that. But an easy way could be for example, to extend your model, so your velocity profile has some time to develop. Does that help you?

Best
Gerrit

Thanks for your quick reply.

Yes that is pretty much what I intend to do, I want the BC to adopt the vector field of the nearest cells, just like zeroGradient does it, but only until it reaches an upper boundary.

Unfortunately extending the model boundaries is going to be very expensive and thus unfeasible for me.
Since you're describing that as the easy way am I correct in the assumption that no such BC exists as of yet?

...Well I would not go that far, that some thing such as wished does not exist. But I do not know it - I am as well a beginner in the end.
But Google is always your friend ;-)

http://www.cfd-online.com/Forums/ope...y-profile.html

Have you tried something like this?

No, that is not at all what I'm looking for.. like I said, the BC is supposed to *copy* the flow field of the nearest cell for the inlet, just like zeroGradient, up until it reaches "value" (ideally for each vector component individually, but magnitude would be fine too), where it will behave like fixedValue for the inlet with the specified "value".
I'm not looking to replace one artificial flow with another, albeit parabolic, one, I want to retain the flow that is already present in the domain, but keep it from blowing out of proportion.

Also compiling and using a BC that is not included with the standard release of openFOAM 2.3 is unfortunately not an option I was just wondering if there's already one included that does the job.
If not I'll have to make do with inletOutlet and hope that the effect on my main flow is negligible.