[fernexda]

Dear all,

I'm currently computing forces around a body, but the forces are surprisingly high, therefore I'm trying to review every aspects to check if those values are realistic. Two things are still unclear :

1.Pressure : I was wondering if the boundary and initial values of the pressure matter. What I did until now is apply a zeroGradient on the Inlet, and p=0 on the outlet. But since the simulation takes place at atmospheric conditions, shouldn't I impose p=p_atm on the outlet ? Would that make any difference ? Would it affect either the viscous forces or the pressure forces ? In what manner ?

2.Density : I don't know the influence of the density on the resulting forces (viscous and pressure forces). As far as I know, the only place where I can choose the density is in the forces dictionary where I have the two following lines :

rhoName            rhoInf;
rhoInf                1;

Should I change this value to make it equal to my fluide density (which is equal to 1.225 kg/m³) to get the real forces ?
And does this value have anything to do with the input values (boundary conditions for U, p, viscosity) ?

Thanks a lot for any answer !

Daniel

[tomf]

Hi Daniel,

I guess you use one of the incompressible solvers from the description you gave. In that case you have to specify the actual density of the fluid in the forces subDict. In transportProperties you specify the kinematic viscosity that belongs to your fluid. They are not related in the code, but they are of course in reality, so you have to make sure they correspond to the same fluid independently. In an incompressible solver the actual pressure level is irrelevant for the solution, however you have to specify an arbitrary constant to make the solution the only solution. If you specify the atmospheric pressure (instead of 0) at the outlet, the only change is that all cells have a pressure that is exactly 1 atm higher. The difference over the body would still be the same.

So to answer your questions:
1. No, only a uniform higher level, no, no
2. yes, not directly: you have to set them correctly

Regards,
Tom

[fernexda]

Hi Tom,

I forgot to specify I'm running incompressible simulations (with pimpleFoam).

You're answer is perfectly clear and it's exactly what I was looking for !

Thank you very much !

Daniel