CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   FLOW-3D (http://www.cfd-online.com/Forums/flow-3d/)
-   -   Strange influence of initial condition? (http://www.cfd-online.com/Forums/flow-3d/27273-strange-influence-initial-condition.html)

Julien October 31, 2008 05:02

Strange influence of initial condition?
 
Hi dear users,

I have observed a strange influence of initial condition that I don't understand :

I am working on a simple linear pipe which flowrate I measure using a baffle. My right boundarie condition is Outflow and I use turbulent model RNG. As initial condition I have set fluid zones with several different Xmax coordinate : the fluids ends at: - the end of the mesh - 5 m before - 10 m before - 20 m before It is the only parameter that changes. And the flowrate is very different at steady flow!! It is bigger when the distance between initial fluid and end of the mesh grows (*200%). And it is very small when initial fluid is set up to the end of the mesh. It seems to be strange as the geometry is the same...I suppose I don't understand well how the Outflow BcData works...The flowrate seems to be influenced by the velocity of the fluid when it reaches the end of the mesh? I have tried with specified pressure and continuative but it is very unstable.

Thanks

Julien


Michael Barkhudarov October 31, 2008 11:42

Re: Strange influence of initial condition?
 
Hi Julien,

What boundary condition do you have at the inlet boundary?

You are right that the outflow boundary is sensitive to how fluid arrives at it. If the flow upstream from the boundary is not well defined, then the results may be upredictable.

Julien November 3, 2008 03:45

Re: Strange influence of initial condition?
 
Hi Michael,

In fact, my pipe starts in a reservoir which is defined by a specified pressure at upstream boundary condition. I use free surface or sharp interface option.

michael barkhudarov November 4, 2008 00:54

Re: Strange influence of initial condition?
 
Julien,

Having a pressure on one side and outflow bc on the other does not fully define the problem. It should either flow rate (or velocity) and outflow, or pressure on both sides. Can you figure out what the pressure is at the outlet?

Michael

Julien November 4, 2008 03:03

Re: Strange influence of initial condition?
 
Hello Michael,

despite I achieve having good results with setting a right distance between the end of my initial fluid and the end of the mesh...it can be an a useful method I think.

But in my case what would you advise me? I modell a dam on a lack, so that I need to fix the upstream fluid height. Then I want to compare the flowrate in three pipes under the dam : I have set the same geometry for each with the same mesh. But I don't want to impose the flowrate anywhere, just let the system free...which left and right boundary condition shall I use?

With continuative it is very unstable...

Concerning the pressure at the outlet, which is a section of pipe, I don't know what the pressure can be...

Julien

michael barkhudarov November 4, 2008 10:20

Re: Strange influence of initial condition?
 
Hi Julien,

You need to position your downstream boundary so as to know something about the boundary conditions there, can't just leave it hanging there with the outflow bc. If the pipe outlets are submerged, then you can establish a hydrostatic pressure bc there by defining the fluid height. The height can be above the top of the mesh.

Michael

Julien Pralong November 6, 2008 06:03

Re: Strange influence of initial condition?
 
Ok Micheal,

good to know, I will improve my methods in that direction, and thanks for answering so quickly my questions!

Julien


All times are GMT -4. The time now is 03:07.