CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   OpenFOAM (https://www.cfd-online.com/Forums/openfoam/)
-   -   Small pressure gradient causes large velocity (https://www.cfd-online.com/Forums/openfoam/198978-small-pressure-gradient-causes-large-velocity.html)

TurbJet February 21, 2018 18:53

Small pressure gradient causes large velocity
 
1 Attachment(s)
Greetings,

I am using simpleFoam running a simple case. The geometry is a simple box with 4.5*4.5*1 [mm], with mesh setup of 45*45*10.

The inlet boundary condition:
p = 6011;
v = zeroGradient;

The outlet boundary condition:
p = 6010;
v = zeroGradient;

Internal field
p = 6010;
v = 0.1;

walls = no-slip

the pressure gradient of the entire flow field is small, which is 1. however the final result of velocity reaches ~18 m/s maximum. Such small pressure gradient should not create such large velocity.

I ran the exactly same case in fluent to check this, and it gave me a reasonable 1.4 m/s, which is expected.

so what's the problem with my openFoam case?

attached is y-normal cutting-surface

luther1990 February 21, 2018 22:42

According to your setup, where U is zero-gradient inlet and outlet, based on force balance, the flow field won’t converge until the pressure gradient equals to the shear force which induced by the non-slip wall. So the velocity here depends on near wall mesh size, residual accuracy and wall functions if use turbulence model.

By the way, this BC setup does not make any physical sense in a steady condition.

TurbJet February 22, 2018 12:34

Quote:

Originally Posted by luther1990 (Post 682485)
According to your setup, where U is zero-gradient inlet and outlet, based on force balance, the flow field won’t converge until the pressure gradient equals to the shear force which induced by the non-slip wall. So the velocity here depends on near wall mesh size, residual accuracy and wall functions if use turbulence model.

By the way, this BC setup does not make any physical sense in a steady condition.

yes, you are right and I figured this out... It was a stupid setup. I already changed pressure at inlet to totalPressure, velocity inlet to pressureInletVelocity, velocity outlet to inletOutlet, and everything works just fine now.


All times are GMT -4. The time now is 08:32.