Hi all,

I made a converging diverging nozzle solution with density based solution. But I can not get the correct results. I have a big differance between mass flow inlet and mass flow outlet. I tried lots of thing but I could'nt the problem.

Thanks a lot for helps.

Because the flow is compressible, this difference is a reasonable result.

Probably you can try refining the mesh and see whether this condition persists. You should also try the flow inside a rectangular channel (parallel walls) to see if mass_in=mass_out.

You mean that, because of compressibility, there is no more mass conservation ?

I don't think it is a reasonable result.

Hi Ahmet

it is your code or you use a commercial CFD software ?

I would look at my boundary conditions, and also check to see if total enthalpy is being conserved. Also, try running without boundary conditions to see if your solver can maintain a uniform freestream in the absence of boundary conditions.

I use Fluent. I made better mesh but I got the same problem. I think there is mass conservation also in compressible flow.

Plot (mass_out - mass_in) as a function of time and see whether this difference is still dropping when you had taken the reading.

What you are calling "mass" and what your commercial code is conserving via the mass conservation equation may not be the same thing. Look in the manual and find the form of the continuity equation at the end of the solution procedure:

* what is numerically conserved?

* where is it numerically conserved (e.g. face or centre).

* does it include pressure smoothing terms?

* are there splitting errors at the end of an iteration?

When you check your boundary conditions the same questions apply.

In transient calculations Mass_in will not be equal to Mass_out. For steady-state calculations the inlet and outlet mass will be equal. Is your code for transient calculations?

No mention of unsteady application; assume classical steady-state Laval nozzle: Of course, the mass flow has to be conserved as Ahmet expects. No question.

Ahmet, how do you evaluate the mass flow? Could there be an error in your evaluation? Could you examine the cross-sectional mass flow at any axial location, and if yes, does the mass flow vary continuously inside the nozzle or is there a jump in mass flow right at the boundary? How well is it converged?

How big is "big", i.e. what is the relative difference between the inlet and outlet mass flows?

If this were true, it would be a HUGE embarassment for the Fluent guys, but because this is such a standard test case and surely has been evaluated many times I am guessing that there is something wrong in your setup and/or post-processing.

hi i need the information about compressible flow solution by ansys cfx.