CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   OpenFOAM (http://www.cfd-online.com/Forums/openfoam/)
-   -   Pipeflow modelling (http://www.cfd-online.com/Forums/openfoam/89486-pipeflow-modelling.html)

GGerber June 14, 2011 08:26

Pipeflow modelling
 
Hi OpenFoamers,

I am trying to compute the pressure loss due to flow through a straight pipe. Unfortunately the computed pressures are wrong.

My model comprises the following:
A wedge shaped domain of 2m length with inlet patch, outlet patch, pipewall, wedge sides and axis of type empty.

At the inlet I specify zeroGradient for all U, k, omega. For p I specify a fixedGradient (10 [m2/s2]) and nut has type 'calculated'.

At the outlet I specify p as zero. U, k, omega and nut have zeroGradients.

The domain is initialized with U=0. k, omega and nut are given fixed values that are estimated from empirical pipeflow equations.

I use simpleFoam with SST k-omega and 40000 time-steps.

The results for the first time-step shows a near linear pressure profile along the pipe centreline, with an inlet pressure of approximately 20 [m2/s2]. However the profile is not completely linear near the inlet and in the subsequent time-steps the pressure at the inlet drops to approximately 16 [m2/s2] and the profile becomes curved.

Any advice for improving my model would be greatly appreciated.

Best regards,
George

GGerber June 15, 2011 02:06

Hi OpenFoamers,

I managed to find my model mistake.
Thanks, issue resolved.

fa123 July 16, 2013 04:59

Hi George,

i am facing exactly the same problem. I have known pressures at the inlet and outlet and used your previous settings. They seem not to be apriopriate.
Could you please tell us which BCs you used for p, U, k, omega, nu?

Thanks


All times are GMT -4. The time now is 13:14.