[Manel]

Hello every body ;


I am new in Fluent , and i am modelizing a jet into a 2D cavity.


I have found a problem to enter the operating conditions of the pressure and velocity with Fluent 19.2 Studient version :


In Boundary conditions i have :



Inlet

Outlet
Wall
Surfacic corps


My problem with pressure is :

When i put the inlet pressure = 10000000 pa , and i switch to Outlet pressure in order to put 101325 pa ; i find that the value of the inlet is automatically put in outlet . I am enable to specify each zone by each pressure.


My problem with velocity is :



I am also enable to put outlet velocity as a boundary condition. It ies just noticed in inlet velocity.


I am using a multiphase model with 2 phases : water and air.


The viscosity model is :



inviscid



The initialization method is :


Standard (compute from inlet)


The solution method : Pressure velocity coupling



Scheme : simple , spatial descitization : green gauss node descritization
Pressure : Presto! , i have used also : body force weignted

Momentum : Second order

Volume fraction : Geo-Reconstruct
Transient First formulation : first order implicit



When i run calculation , fluent stop .It diplay a divergence error :



Divergence detected in AMG solver: y-momentum Stabilizing pressure correction to enhance linear solver robustness.

Error at host: floating point exception


Error at Node 0: floating point exception



Sincerely , i am enable to identify the origin of the error :

may be the BC of velocity and pressure ?

What do you think please ?


Thanks you for your help

[annan]

Dear Manel,
I am not sure I understand well your probleme, but are you trying to set both velocity AND pressure at the inlet AND the outlet ?
Normally, I would recommand you to choose a velocity-inlet for the inlet and pressure-outlet for the outlet.
Hope this will help
Good luck
Annan

[Manel]

Hi ,



Thanks you for your response Annan ,



I have set operating pressure inlet =10^7 pa


V inlet =-1.5 m/s


But my problem with Fluent 19.2 studient version is :


When i want to set the operating pressure outlet (P=101325 pa) ,
i cannot , because when i do this , i find that pressure inlet has changed to

P=101325 pa.


Bu this way , the pressure gradient in my flow =0 .
I don't know if this is the real problem or if the problem is related to mesh.


What do you think ?



Have a nice day

[annan]

Dear Manel,

If I understood well, you choose velocity-inlet as boundary condition for your inlet. In this case, the pressure you specify is an Outflow Gauge pressure, which means that if your flow exits the domain at any boundary, then the pressure is equal to what you specify.

Which makes sense with what happens to your calculation. You cannot specify the velocity AND the pressure at a boundary, if you want to specify the inlet pressure then you should choose Pressure-inlet as boundary condition.

The usual set of boundary conditions is the following :
- velocity-inlet at the inlet : you specify only the velocity magnitude or components
- Pressure-outlet at the outlet : keep it at 0, which means that your flow goes to the atmosphere. By the way, the Gauge pressure is defined with respect to the operating pressure defined, the default value for the operating pressure is the atmospheric pressure, that is why you need to specify 0 and not 1atm at the outlet (Gauge pressure = P-P_operating_conditions = P-P_atm)

The total pressure gradient is not going to be equal to zero, because you have an inlet velocity, which makes your dynamic pressure (0.5*rho*v^2) is not equal to zero.

Another precision, what you specify at boundaries is Gauge pressure, and not Operating pressure. The operating pressure is defined in the Operating conditions panel.

I hope this helps, do not hesitate if any of what I explained is not clear.
Good luck,
Annan

[Manel]

Dear Annan ,



Thaks you very much.It is clear now for me.


In another side , i guess that the mesh may be the cause.


I will improve the mesh and adjust my pressure as you told me.



I will tell you when i obtain the result.


Have a nice day.


Manel

[annan]

Dear Manel,
You should check the mesh metrics (skewness, orthogonal quality ... etc.) and see if they are in a good quality range.
Good luck
Best regards,
Annan

[Manel]

Dear Hannan



Thanks you very much.You have helped me so much.


Yours faithfully ,


Manel Souguir