using setFields to make a sinusodial phase condition

wschosta · June 9, 2011, 16:12

Hello foamers!

I'm working on a two phase flow simulation that will (hopefully eventually) model the kelvin helmholtz instability test. I was wondering if it's possible to set an initial phase condition other than a box using the setFields command. Currently what I'm doing is rewriting all of the alpha1 and U values from within createFields.H and I feel like this is a very roundabout way of accomplishing what I'm trying to do.

So, any suggestions on setFields? I know it goes back to setFieldsDict but I don't know what the possible parameters are.

gwierink · June 10, 2011, 09:10

Hi Walter,

Have a look at swak4Foam.

wschosta · June 13, 2011, 16:05

Hi Gijs, I have considered using the groovyBC and funkyFields parts of swak4Foam but for my purposes it would be optimal to only use the functions contained within the distributed libraries of OpenFOAM. I realize the documentation of such functions isn't as great as user made functions such as those in swak4Foam but it's a burden I have to live with.

I appreciate your input but I guess I'll just stick to setFields and the reassigning loops I've constructed.

gwierink · June 14, 2011, 01:11

Hi Walter,

Alright, well, many things are possible with the "standard" OpenFOAM pack as well. There are also other cell selection methods, to call it like that, in e.g. the cellSet utility. For instance, nearestToCell, sphereToCell, surfaceToCell, and so on. Maybe having a look there is useful. cellSet lives in: $FOAM_UTILITIES/mesh/manipulation/cellSet

Hope that is of any help.

marzbali · November 9, 2011, 14:20

Hello Gijs,

Do these functioned that you mentioned such as sphereToCell work in 2D axisymmetric simulations? Because I tried to patch a droplet in my domain and I got a cylinder instead!! I have used sphereToCell, cellSet, FunkySet and they all gave me the same shape!

Thanks a lot in advance.
Marzbali

gwierink · November 9, 2011, 18:29

Hi Marzbali,

A sphere in 2D should result in a cylinder of one cell length. If you would a like to set a cylinder specifically, there is also a cylinderToCell function. To be honest, I do not really understand the problem. In 2D you should get a circle/cylinder, right?

marzbali · November 9, 2011, 22:47

sorry I found my mistake, i didn't consider x as my axis!

June 9, 2011, 16:12	using setFields to make a sinusodial phase condition	#1
wschosta Member Walter Schostak Join Date: May 2011 Posts: 35 Rep Power: 14	Hello foamers! I'm working on a two phase flow simulation that will (hopefully eventually) model the kelvin helmholtz instability test. I was wondering if it's possible to set an initial phase condition other than a box using the setFields command. Currently what I'm doing is rewriting all of the alpha1 and U values from within createFields.H and I feel like this is a very roundabout way of accomplishing what I'm trying to do. So, any suggestions on setFields? I know it goes back to setFieldsDict but I don't know what the possible parameters are.

June 10, 2011, 09:10		#2
gwierink Senior Member Gijsbert Wierink Join Date: Mar 2009 Posts: 383 Rep Power: 18	Hi Walter, Have a look at swak4Foam. __________________ Regards, Gijs

June 14, 2011, 01:11		#4
gwierink Senior Member Gijsbert Wierink Join Date: Mar 2009 Posts: 383 Rep Power: 18	Hi Walter, Alright, well, many things are possible with the "standard" OpenFOAM pack as well. There are also other cell selection methods, to call it like that, in e.g. the cellSet utility. For instance, nearestToCell, sphereToCell, surfaceToCell, and so on. Maybe having a look there is useful. cellSet lives in: $FOAM_UTILITIES/mesh/manipulation/cellSet Hope that is of any help. __________________ Regards, Gijs

November 9, 2011, 18:29		#6
gwierink Senior Member Gijsbert Wierink Join Date: Mar 2009 Posts: 383 Rep Power: 18	Hi Marzbali, A sphere in 2D should result in a cylinder of one cell length. If you would a like to set a cylinder specifically, there is also a cylinderToCell function. To be honest, I do not really understand the problem. In 2D you should get a circle/cylinder, right? __________________ Regards, Gijs

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
VOF for single phase?	vuccj7	FLUENT	0	April 20, 2011 19:59
wall boundary condition for dispersed phase	dicara	CFX	3	October 27, 2010 04:50
Source Term used in Eulerian Model(Two phase)	Padian	FLUENT	1	May 19, 2008 03:47
OpenFOAM with IBM AIX	matthias	OpenFOAM Installation	20	March 25, 2008 02:36
a way to make lots of money quick and easy no lies	Dob	Main CFD Forum	0	October 10, 2006 16:45

June 13, 2011, 16:05		#3
wschosta Member Walter Schostak Join Date: May 2011 Posts: 35 Rep Power: 14	Hi Gijs, I have considered using the groovyBC and funkyFields parts of swak4Foam but for my purposes it would be optimal to only use the functions contained within the distributed libraries of OpenFOAM. I realize the documentation of such functions isn't as great as user made functions such as those in swak4Foam but it's a burden I have to live with. I appreciate your input but I guess I'll just stick to setFields and the reassigning loops I've constructed.

November 9, 2011, 14:20		#5
marzbali New Member Concordia_CFD Join Date: Jul 2010 Location: Canada Posts: 24 Rep Power: 15	Hello Gijs, Do these functioned that you mentioned such as sphereToCell work in 2D axisymmetric simulations? Because I tried to patch a droplet in my domain and I got a cylinder instead!! I have used sphereToCell, cellSet, FunkySet and they all gave me the same shape! Thanks a lot in advance. Marzbali

November 9, 2011, 22:47		#7
marzbali New Member Concordia_CFD Join Date: Jul 2010 Location: Canada Posts: 24 Rep Power: 15	sorry I found my mistake, i didn't consider x as my axis!