interFoam: my model gets dry when changing the mesh resolution

pardoa · September 16, 2019, 14:58

Hi everyone,

I'm using interFoam to simulate the flow dynamics of an alluvial river in Switzerland. In order to find the optimal mesh resolution that allows to run faster simulations, I have being running steady-state simulations for different grid cells: 0.5x0.5x0.5m, 1x1x1m and 2x2x2m and comparing later on the simulated outputs.

Each model provides different results, which up to some extent could be considered the expected behaviour. Different cell sizes will most of the times lead to different results. However, the differences between models shouldn't be too big and, surprisingly, they are, especially for the coarsest model, which literally gets fully dry after a couple of timesteps. The 1x1x1 model also runs mostly dry most of the runtime. I wonder why this happens, since the finer model undergoes, at least in my eyes, the expected trend: there is water flowing all the time on the riverbed and reaches the steady-state after 10 timesteps.

Could this behaviour be related to the Courant number? InterFoam is a multi-phase solver and, therefore, very sensitive to this parameters. So far I have defined a Courant number of 1 for both the alpha phase and other fields. Maybe decreasing it will resolver the issue.

Hope someone can help me on this! Thanks in advance!

Álvaro

pardoa · September 17, 2019, 17:12

Hi!

Sorry to insist guys, but can anyone provide me any hint on this issue? Thanks!

Álvaro

vatavuk · September 20, 2019, 10:48

Hi Álvaro,

It seems the problem is related to boundary conditions. In the river bottom, do consider it as a wall or include some infiltration?
In the inlet normally a fixed flow rate is used.
In the outlet I think the best is to use a fixed value velocity. This condition, together with the imposed flow rate at the inlet, will give you an imposed water level at the outlet, because you have Q=V.A, if Q and V are imposed then you are also imposing the area A, that is related to the water level.

Best regards,
Paulo

September 16, 2019, 14:58	interFoam: my model gets dry when changing the mesh resolution	#1
pardoa New Member pardoa Join Date: May 2018 Posts: 29 Rep Power: 8	Hi everyone, I'm using interFoam to simulate the flow dynamics of an alluvial river in Switzerland. In order to find the optimal mesh resolution that allows to run faster simulations, I have being running steady-state simulations for different grid cells: 0.5x0.5x0.5m, 1x1x1m and 2x2x2m and comparing later on the simulated outputs. Each model provides different results, which up to some extent could be considered the expected behaviour. Different cell sizes will most of the times lead to different results. However, the differences between models shouldn't be too big and, surprisingly, they are, especially for the coarsest model, which literally gets fully dry after a couple of timesteps. The 1x1x1 model also runs mostly dry most of the runtime. I wonder why this happens, since the finer model undergoes, at least in my eyes, the expected trend: there is water flowing all the time on the riverbed and reaches the steady-state after 10 timesteps. Could this behaviour be related to the Courant number? InterFoam is a multi-phase solver and, therefore, very sensitive to this parameters. So far I have defined a Courant number of 1 for both the alpha phase and other fields. Maybe decreasing it will resolver the issue. Hope someone can help me on this! Thanks in advance! Álvaro

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Dry cooler - simplified cfd model	mech_01	FLUENT	0	September 23, 2017 16:00
[snappyHexMesh] crash sHM	H25E	OpenFOAM Meshing & Mesh Conversion	11	November 10, 2014 12:27
Journal file error	magicalmarshmallow	FLUENT	3	April 4, 2014 13:25
Warning 097-	AB	Siemens	6	November 15, 2004 05:41
Problem in TE model and cell size	Mir	Siemens	2	December 11, 2001 10:46

September 17, 2019, 17:12		#2
pardoa New Member pardoa Join Date: May 2018 Posts: 29 Rep Power: 8	Hi! Sorry to insist guys, but can anyone provide me any hint on this issue? Thanks! Álvaro

September 20, 2019, 10:48		#3
vatavuk Senior Member Paulo Vatavuk Join Date: Mar 2009 Location: Campinas, Brasil Posts: 200 Rep Power: 18	Hi Álvaro, It seems the problem is related to boundary conditions. In the river bottom, do consider it as a wall or include some infiltration? In the inlet normally a fixed flow rate is used. In the outlet I think the best is to use a fixed value velocity. This condition, together with the imposed flow rate at the inlet, will give you an imposed water level at the outlet, because you have Q=V.A, if Q and V are imposed then you are also imposing the area A, that is related to the water level. Best regards, Paulo