[CFDger]

Hello,
I want to define a pressure drop observable in the ANSYS FLUENT Andjoint Solver Method. The problem is, that my "outlet" where I want to measure the total pressure loss in comparison to the inlet (which is defined as a boundary layer) is not defined as a boundary layer (in fact it is a nozzle exit).
I tried to import a .stl surface but that didn't work as the Adjoint Solver apparently only accepts boundaries as input.

Is there a workaround for this issue?

Thanks

[CFDger]

Anyone has experience with this problem?

[Indra]

Originally Posted by CFDger

Anyone has experience with this problem?

I don't really understand your problem. Could you perhaps post some images?

Calling something boundary layer can be confusing in the world of CFD. I think you meant boundary zone or wall or something.

[CFDger]

I don't mean boundary layer, that was wrong by me. I am refering to a boundary surface which only acts as an exit plane/area for post processing the information at the exit of my nozzle (which is placed inside a bigger domain possessing inlet and outlet bc). The thing is - without this interior surface I can only look at the desired flow information at the nozzle exit plane/surface in post processing. But I need the solver (Fluent) to access this area during calculation by having access to the interior surface.

I hope it is clear now. If not I will try to get a picture for you.
Thanks

[Indra]

Originally Posted by CFDger

I don't mean boundary layer, that was wrong by me. I am refering to a boundary surface which only acts as an exit plane/area for post processing the information at the exit of my nozzle (which is placed inside a bigger domain possessing inlet and outlet bc). The thing is - without this interior surface I can only look at the desired flow information at the nozzle exit plane/surface in post processing. But I need the solver (Fluent) to access this area during calculation by having access to the interior surface.

I hope it is clear now. If not I will try to get a picture for you.
Thanks

Right, i think i understand it somewhat. So the outlet has to be an outlet, it can't be an interior surface(i think so).

If you are looking at adjoint solver, you are probably trying to optimse your nozzle. In this case you can split your simulation for inside the nozzle and outside. That's the only way you can use adjoint for the nozzle.

[CFDger]

Almost, the physical outlet of the nozzle has no boundary condition at all, yet.

The nozzle itself has a boundary inlet condition, no exit condition (the exit condition is defined by the nozzle pressure ratio etc.). The nozzle exit flow enters the rest of the cylindrical domain. This domain consists of pressure ff bc and a pressure outlet bc (those do not interfere with the physical outlet of the nozzle).



I want to introduce the interior surface to the physical nozzle exit to be able to process the information stored at the nodes at that location (that way I can for example monitor the total pressure loss along the nozzle and other important design parameters).


Yes, you are right. I am trying to optimize the geometry of my nozzle with the Adjoint solver. Could you please elaborate on splitting my domain on for inside the nozzle and outside? I think thats actually that what could be the best solution.

[Indra]

Originally Posted by CFDger

Almost, the physical outlet of the nozzle has no boundary condition at all, yet.

The nozzle itself has a boundary inlet condition, no exit condition (the exit condition is defined by the nozzle pressure ratio etc.). The nozzle exit flow enters the rest of the cylindrical domain. This domain consists of pressure ff bc and a pressure outlet bc (those do not interfere with the physical outlet of the nozzle).



I want to introduce the interior surface to the physical nozzle exit to be able to process the information stored at the nodes at that location (that way I can for example monitor the total pressure loss along the nozzle and other important design parameters).


Yes, you are right. I am trying to optimize the geometry of my nozzle with the Adjoint solver. Could you please elaborate on splitting my domain on for inside the nozzle and outside? I think thats actually that what could be the best solution.

What you can do is in your CAD tool, put a surface at the nozzle exit and treat it as a mesh interface in Fluent. This means, your nozzle will have a surface at the exit and this surface will transfer all the data to outside domain. You can use this surface to get pressure in adjoint.