open channel flow-rough bed

antonio · October 12, 2010, 05:14

Hi all.

I am simulating a free surface flow where the bed is hydraulically rough (several triangles equally spaced). I have already achieved convergence (1o^-5 of residual and 0.01 of imbalance) for a case test where the

Escreva texto ou o endereço de um Web site ou traduza um documento
Cancelar

height of the triangles is very small ( I have used the multi grid solver in expert parameter).

Now I am trying to increase the height of the triangles, the height of the water and the height of the domain to see the effect of the rugosity. I have used the same type of elements in the mesh, however I am not achieving the same level of convergence. I am using k-epsilon model. Does anyone have a suggestion in order to get the same level of convergence?

Best Regards

antonio · October 12, 2010, 05:18

Sorry for the mistake. The correct post is:

Hi all.

I am simulating a free surface flow where the bed is hydraulically rough (several triangles equally spaced). I have already achieved convergence (1o^-5 of residual and 0.01 of imbalance) for a case test where the
height of the triangles is very small ( I have used the multi grid solver in expert parameter).

Now I am trying to increase the height of the triangles, the height of the water and the height of the domain to see the effect of the rugosity. I have used the same type of elements in the mesh, however I am not achieving the same level of convergence. I am using k-epsilon model. Does anyone have a suggestion in order to get the same level of convergence?

Best Regards

ghorrocks · October 12, 2010, 18:13

Why are you using the multi-grid solver expert parameter? CFX uses a multi-grid solver by default anyway and you should not change the defaults unless you have a good reason to do so.

What do you mean by rugosity? Is your simulation steady or transient?

It could be that increasing the size of the triangles is generating new flow features which your current solver settings are having problems resolving. Have you looked here: http://www.cfd-online.com/Wiki/Ansys...gence_criteria

michael_owen · October 12, 2010, 22:33

Have you post-processed the residuals to determine where and why they are high?

October 12, 2010, 05:14	open channel flow-rough bed	#1
antonio Senior Member Join Date: Jan 2010 Posts: 110 Rep Power: 16	Hi all. I am simulating a free surface flow where the bed is hydraulically rough (several triangles equally spaced). I have already achieved convergence (1o^-5 of residual and 0.01 of imbalance) for a case test where the Escreva texto ou o endereço de um Web site ou traduza um documento Cancelar height of the triangles is very small ( I have used the multi grid solver in expert parameter). Now I am trying to increase the height of the triangles, the height of the water and the height of the domain to see the effect of the rugosity. I have used the same type of elements in the mesh, however I am not achieving the same level of convergence. I am using k-epsilon model. Does anyone have a suggestion in order to get the same level of convergence? Best Regards

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Yacht in Open Channel Flow	andreimour	FLUENT	1	October 14, 2010 23:54
Modelling open channel flow.	gizeh	FLUENT	0	October 12, 2009 06:20
open channel flow with a layer of air	Naomi	CFX	1	November 15, 2008 06:34
Two-phase open channel flow	Edwin	FLUENT	3	July 11, 2006 02:21
Two-phase open channel flow	Edwin	FLUENT	0	July 11, 2006 01:36

October 12, 2010, 05:18		#2
antonio Senior Member Join Date: Jan 2010 Posts: 110 Rep Power: 16	Sorry for the mistake. The correct post is: Hi all. I am simulating a free surface flow where the bed is hydraulically rough (several triangles equally spaced). I have already achieved convergence (1o^-5 of residual and 0.01 of imbalance) for a case test where the height of the triangles is very small ( I have used the multi grid solver in expert parameter). Now I am trying to increase the height of the triangles, the height of the water and the height of the domain to see the effect of the rugosity. I have used the same type of elements in the mesh, however I am not achieving the same level of convergence. I am using k-epsilon model. Does anyone have a suggestion in order to get the same level of convergence? Best Regards

October 12, 2010, 18:13		#3
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,703 Rep Power: 143	Why are you using the multi-grid solver expert parameter? CFX uses a multi-grid solver by default anyway and you should not change the defaults unless you have a good reason to do so. What do you mean by rugosity? Is your simulation steady or transient? It could be that increasing the size of the triangles is generating new flow features which your current solver settings are having problems resolving. Have you looked here: http://www.cfd-online.com/Wiki/Ansys...gence_criteria

October 12, 2010, 22:33		#4
michael_owen Senior Member Michael P. Owen Join Date: Mar 2009 Posts: 196 Rep Power: 17	Have you post-processed the residuals to determine where and why they are high?