CFD Online Discussion Forums

CFD Online Discussion Forums (
-   Main CFD Forum (
-   -   Is that physical? (

Giovanni January 8, 2008 08:08

Is that physical?
Hi! I am running a very simple open channel case, just a rectangular box, with lateral surfaces BC assumed to be wall, BC of top is Pressure outlet, and bottom's BC is wall. The inlet boundary condition is inlet vent, with velocity normal to boundary fixed (e.g. 2 m/s) and the outlet is pressure outlet, the model being used is VOF and all the boundary condition are setted with the open channel flow feature. The problem is that if I set as initial condition the heigth of the free surface and I set initial velocity field's components to zero, when I run the transient simulation immediately when the flow start I can see a soliton wave generate immediately near the inlet and start to propagate. This appens with coarse grids and also with fine grids, and with different inlet velocities, and the heigth of the wave appears to be proportional to the velocity setted in the inlet, this can be a feature of the solitons, as they velocity is c=sqrt(g*(h+A)), but I have same doubt if it is a physical feature of this scenario or if it is just numerical. So what do You think?

Giovanni January 10, 2008 12:34

Re: Is that physical?
Any suggestions? I have thought a lot about it and I have recognized where is the matter. Simply it is not physical that the water in the inlet starts suddenly to move in from rest, as a rigid body, unless it's pushed in by a piston-like machine.

otd January 10, 2008 14:32

Re: Is that physical?
Do your IC's on the pressure field satisfy static equilibrium? Does pressure at any point below the surface grow from the surface to the bottom to reflect the weight of the fluid column above that point? If not, that'll give the fluid quite a kick initially.

Sorry if that initial condition is automatic for the code you're using. I'm not familiar with the code.

Giovanni January 10, 2008 19:55

Re: Is that physical?
Hi, thank for your reply, I have tried some sensitive studies and I have found that it happens also if I impose the hydrostatic pressure profile. Giovanni

All times are GMT -4. The time now is 10:43.