code_saturne User's Forum

Hello,

we study the fluid flow in a cylindrical pipe at Re=1.76e5. The first extremity is the inlet and the other is the outlet. The turbulence model are the k-epsilon and the k-omegaSST, y+ is y+ = 2.381. The aim is to determine the head looses in the pipe in doing the substraction between outlet and inlet pressure surface integration.

We use a CGNS mesh and the Code_Saturne version is 2.0.1.

To calculate those head looses, we first used : PROPFB(IFAC,IPR(IPHAS)), but the results are bad compared with the CFX and WinDP's results (the values are very inferior). Thus we tried to use PROPFB(IFAC,IPRTOT(IPHAS)), in order to calculate the total pressure integral (maybe the head loss is leaded with total pressure in CFX) but the values are the same than with PROBFB(IFAC,IPR(IPHAS)).

So maybe the integer musn't be calculated by PROPFB?

Did anybody make such a calculation in usproj.f90 ?

We join our usproj.f90 : maybe someone could give us help about our mistake !

Best regards,

Pauline Delteil & Marie Pomarède

Hello,
The value of iphas is missing here also. Could you post (or send by mail at the support) your mesh and set up?
Mickaël

Hello Mickaël,
Thank you for your answer, and for your suggestion. Unfortunately, even if putting IPHAS = 1 is correct, that does not change our bad results. We continue our investigations (still waiting for calculation results), and in case of no more idea, we will send our files (.xml & mesh).
Many thanks,
Pauline Delteil & Marie Pomarède

Hello,

we don't find the right way to go about it. We tested different solutions to calculate the head looses like:

DO ILELT = 1, NLELT
      IFAC = LSTELT(ILELT)
      IEL = IFABOR(IFAC)
      P = P + PROPCE(IEL, IPPROB(IPRTOT(IPHAS))
ENDDO

or

DO ILELT = 1, NLELT
      IFAC = LSTELT(ILELT)
      P = P + PROPCE(IFAC, IPPROB(IPRTOT(IPHAS))
ENDDO

...etc...

but nothing give us good results.

Have you an explain, do we make a mistake somewhere?

We join our .xml and the mesh.

Best regards,

Pauline

Hello,
 
try to use RTP(IEL,IPR(IPHAS)) and then correct it with P0 term to get the total pressure.
 
Mickaël Hassanaly

Hello,

thank you for your answer but in using RTP(IEL,IPR(IPHAS)) corrected with P0(IPHAS), we obtain some results too high.

Maybe it exists an other way to obtain the pressure on the boundary faces.

Best regards,

Pauline