CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   CFX (https://www.cfd-online.com/Forums/cfx/)
-   -   free surface flow simulation to a junction flows (https://www.cfd-online.com/Forums/cfx/92683-free-surface-flow-simulation-junction-flows.html)

emily_cx September 21, 2011 17:35

free surface flow simulation to a junction flows
 
hi everyone, I am doing a free surface modelling of two flows junction. I have some experimental data to validate the model. However, the numerical results are not as good as expected which don't have a good agreement with my lab data. So here I want to ask some of you to give me some advice to help me out.

Domain description: a main channel with 20m long and 1m wide orthogonally intersected by a branch channel at around 5m distance from the main channel's upstream. The branch channel is about 8m long and has the same width with main 1m.
Modelling description: Most of the settings are based on the Tutorial. But in this case I have two inlets, the expressions for the brach channle inlet are created the same as the main channle inlet (I am not sure this setting is right or wrong), the model is run as steady state (I read some previous comments on free surface, some people said always run transient for free surface, so would it be a source of error? but I got converged for steady state). The inlet boudary conditon is set to be mass flow rate, the outlet b.c. is static pressure with a constant water height (I know the exact tail water height), and the top of the channel is set to be openning with zero pressure gradient

Any other things I should consider for the simulation? Although I got every target achived, I really have no idea what's the problem for the non-accuracy. Can anybody give me some suggestions what I should do next?

Thank you very much!

ghorrocks September 21, 2011 19:19

It is unusual for free surface simulations to converge in steady state - it suggests to me that you may have excessive dissipation in your model. Are you using high resolution differencing?

You would have to explain what the poor agreement with experiment is for us to help you further.


All times are GMT -4. The time now is 04:32.