CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   FLUENT (http://www.cfd-online.com/Forums/fluent/)
-   -   Open Channel Flow (http://www.cfd-online.com/Forums/fluent/47406-open-channel-flow.html)

Ryan February 21, 2008 18:27

Open Channel Flow
 
I want to model the flow around the hull of a canoe. Steady state is my only interest so I'm using an implicit solver. I don't have implicit body forces on.

I know I need to use the VOF open channel flow. I've tried modeling it several different ways (1 inlet & 1 outlet for air and water, 2 inlet faces & 2 outlet faces for separate air and water boundaries). -What is the best way to model the space? Single volume or a split volume where the waterline would be (2 zones, 1 water 1 air)?

I've tried several boundary conditions, initializations and patching, but I cannot get a simulation to converge. It will not diverge, but it just runs with no reduction in the residual. I've followed the fluent manual to no avail. (i.e. - setting inlet/outlet groups, specifying free surface & bottom level, using mass flow inlet and pressure outlets, initializing inlet phase2 faction to zero and then patching it to 1...etc).

Just drop in a suggestion, or ask questions if you want more information.

-Any pointers on the matter are greatly appreciated

vovman March 17, 2008 11:10

Re: Open Channel Flow
 
Hi, Ryan:

I am also working on a open channel problem. Unsteady simulation is running in my case. I've tried different settings like you but without good result. May you can try the unsteady simulation to simulate steady condition. But the problem should be a steady condition or you'll never get a steady solution.

As my experience, using unsteady simulation with small time step say 0.001 s, it takes a long time to let the flow from inlet to outlet. And ends in divergent when the air and water are sparking in each phases.

For flunet 6.3, it has the option of open channel. I did't succeed in using it. May you can try it.

I have google and found that University of Iowa have done some 3d open channel simulation using Fluent. I may asked them for setting the parameter for running Flunet.



All times are GMT -4. The time now is 05:13.