[ankur bhatangar]

Hi guys,

I am solving flow inside pipe with varying inlet velocity profile.I am basically using Incompressible formulation and implemented velocity profile in SOLVER_DIRECT_MEAN in

void CEulerSolver::BC_Inlet(CGeometry *geometry, CSolver **solver_container, CNumerics *conv_numerics, CNumerics *visc_numerics, CConfig *config, unsigned short val_marker)

in if(incompressible) with V_inlet as a parameter varying with time.

BCs i am using is Velocity_inlet ,wall with flux 0.0 and Pressure outlet with back pressure 101325.

As a result of simulations i am getting Conservative_1 negative and zero. My working fluid is water .I also tried modify density in code by giving it direct value as 998.2 .

My concern is I don't have idea of V_Domain used in code ?what is its purpose?

How Incompressible solver is calculating Pressure?

Reasons for getting negative and zero Conservative_1 ?How can i modify Conservative_1?

[Vino]

Hi Ankur,

I once tried time varying BCs for compressible flow. But incompressible, the time is not a real time (pseudo time- if you are using AC). Plz clarify.!

Also, the pressure is normalized with free stream pressure and it is possible to get Zero and negative pressure values in incompressible flow. In incompressible, it just calculates the relative change of pressure.

[talbring]

Hi Ankur,

as Vino already wrote, incompressible flows are computed in SU2 using the artificial compressibilty method (https://math.berkeley.edu/~chorin/chorin67.pdf). This formulation is only valid for steady problems.

[ankur bhatangar]

Thanks for the information.I understood these points while i was trying with su2 few days back .Now i have shifted to openfoam for my this case.

[Vino]

Originally Posted by ankur bhatangar

Thanks for the information.I understood these points while i was trying with su2 few days back .Now i have shifted to openfoam for my this case.

Hi Ankur,

If you are finding OpenFOAM is a better choice for your problem, you can go ahead with it.

However, you can do the same thing with SU2 also. Only thing you have to take care is that your BCs should vary with real time, not with the pseudo time (you have to keep the config setting as unsteady, Dual Time Stepping and you should use real-time in your BCs)

[ankur bhatangar]

hi ,
I am calling current iter number and time step .Then finding time T= (Extiter-1)*dt; My working fluid is water and When I run the simulation it is showing negative conservative_1 and zeroes somewhere which is giving me lots of worries.I am unable to match result with exp data for same case.I don't know why it is showing such behaviour.If you have any idea then please explain.

[Vino]

Originally Posted by ankur bhatangar

hi ,
I am calling current iter number and time step .Then finding time T= (Extiter-1)*dt; My working fluid is water and When I run the simulation it is showing negative conservative_1 and zeroes somewhere which is giving me lots of worries.I am unable to match result with exp data for same case.I don't know why it is showing such behaviour.If you have any idea then please explain.

Plz do the simulation with fixed inlet velocity BC and check whether the solution is fine or not. Also plz post your config file that you are using.