modeling de-moisturization of a plaster mould by air injection in CFX

sakalido · February 16, 2012, 12:36

Hi all,

I am trying to model the following problem and I do not know where is a good start.

I have a plaster cube which contains hollow cylinder inside all along the length of the cube.
This plaster cube is saturated with water. To be more accurate it has absorbed some percentage of water previously.
Air is injected into the cylinder to push out the water content of the plaster.
I want so see the moisture profiles inside the plaster cube.

I am trying to use the porous media models in CFX but I am not sure how to define the problem.
Should I define one main domain as plaster which is porous and then have air, water and plaster as fluids inside this domain. If so, should I consider plaster as a continuous fluid?

I will appreciate it if you could help me in giving some ideas. Until now I am unable to find anything in the literature.

Thank you,
Sara

ghorrocks · February 16, 2012, 18:12

I do not think you want to model this as a porous medium. The air does not penetrate the plaster, does it? I think you want to model the plaster as a solid with an additional variable to represent the water, and the water can move around with diffusion.

sakalido · February 16, 2012, 18:20

Hi Glenn,

Thanks for your reply.
Air penetrates the plaster and that's the way it pushes out water from the plaster. Imagine we have a plaster mould saturated in water. We inject air inside the hollow cylinder in the plaster, air penetrates from the cylinder wall into the mould and pushes away water. So we will see water bubbles coming out of the plaster mould walls.

I hope my description is clear. I appreciate your help.

Sara

ghorrocks · February 16, 2012, 18:41

In that case then, yes, a porous model makes sense. You will need to think about the physics of how the air dries the plaster. Does it push the water in front of it? (if you see water bubbles I suspect it does) Does the water evaporate into the air? What does the water do when it gets to the outside of the plaster?

sakalido · February 16, 2012, 19:01

Well, water is still inside the plaster mould (the porous media). Then, air is injected with pressure into this porous media and pushes out the water from the surface of the porous media. I think basically air replaces water. There is no major evaporation. Also, when you look at the plaster mould at this stage, you observe water coming out from the surface. Its is not exactly bubbles. I think replacement of water with air is the best description.

I do not know how I should define the domains in CFX. Do I need to have more that one domain or just one porous domain is enough? Is plaster (porous media) going to be my main domain? If so, should I define it as a "continuous fluid" or "dispersed solid"? Well the mould is actually a solid but as far as I know dispersed solid is used to define particles inside a fluid domain. So the only remaining option for me would be continuous fluid which does not sound correct to me.

What do you think?

ghorrocks · February 16, 2012, 19:12

There are tutorial examples on porous domains. Do these to see how to set up porous regions.

The physics of air pushing a liquid through a porous domain is very complex and CFX does not have a model developed to model it. I talked to a bloke who did some modelling on this - here is a paper of his (http://www.springerlink.com/content/8430t042j6gqx086/).

You are probably going to have to compromise, and work out if a free surface flow with some form of resistance from the porous medium will work, or if you can go an additional variable approach can work. There are tutorials on both of these approaches, I recommend you do them both (in fact all the tutorials) before proceeding.

Forget about how you define these things until you know how you are going to model the basic physics. Once you know the basic physics then worry about the implementation.

sakalido · February 16, 2012, 19:19

Thanks a lot Glenn, the paper helps a lot. I actually looked through the tutorials but in this version of ANSYS-CFX I am using currently (version11), there is only a catalytic converter which uses the directional loss model.
I will check the newer ANSYS versions for the tutorials you mentioned.

Thank you again for the good advice

Sara

February 16, 2012, 12:36	modeling de-moisturization of a plaster mould by air injection in CFX	#1
sakalido Member Sara S. Join Date: May 2010 Location: NJ, USA Posts: 41 Rep Power: 16	Hi all, I am trying to model the following problem and I do not know where is a good start. I have a plaster cube which contains hollow cylinder inside all along the length of the cube. This plaster cube is saturated with water. To be more accurate it has absorbed some percentage of water previously. Air is injected into the cylinder to push out the water content of the plaster. I want so see the moisture profiles inside the plaster cube. I am trying to use the porous media models in CFX but I am not sure how to define the problem. Should I define one main domain as plaster which is porous and then have air, water and plaster as fluids inside this domain. If so, should I consider plaster as a continuous fluid? I will appreciate it if you could help me in giving some ideas. Until now I am unable to find anything in the literature. Thank you, Sara

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Multiphase flow modeling: CFX or FLUENT?	Paul	Main CFD Forum	8	January 24, 2012 10:25
Problems modeling subsonic engine air inlet with flow separation	TWaung	CFX	2	March 29, 2009 19:42
Problems modeling CONTACT SURFACES with CFX Mesh	MAldaz	CFX	2	September 12, 2006 18:10
CFX for turbulence modeling	Sujit	CFX	1	August 13, 2003 07:20
Radiation modeling in CFX 5	vortex	CFX	2	August 26, 2002 17:13

February 16, 2012, 18:12		#2
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,852 Rep Power: 144	I do not think you want to model this as a porous medium. The air does not penetrate the plaster, does it? I think you want to model the plaster as a solid with an additional variable to represent the water, and the water can move around with diffusion.

February 16, 2012, 18:20		#3
sakalido Member Sara S. Join Date: May 2010 Location: NJ, USA Posts: 41 Rep Power: 16	Hi Glenn, Thanks for your reply. Air penetrates the plaster and that's the way it pushes out water from the plaster. Imagine we have a plaster mould saturated in water. We inject air inside the hollow cylinder in the plaster, air penetrates from the cylinder wall into the mould and pushes away water. So we will see water bubbles coming out of the plaster mould walls. I hope my description is clear. I appreciate your help. Sara

February 16, 2012, 18:41		#4
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,852 Rep Power: 144	In that case then, yes, a porous model makes sense. You will need to think about the physics of how the air dries the plaster. Does it push the water in front of it? (if you see water bubbles I suspect it does) Does the water evaporate into the air? What does the water do when it gets to the outside of the plaster?

February 16, 2012, 19:01		#5
sakalido Member Sara S. Join Date: May 2010 Location: NJ, USA Posts: 41 Rep Power: 16	Well, water is still inside the plaster mould (the porous media). Then, air is injected with pressure into this porous media and pushes out the water from the surface of the porous media. I think basically air replaces water. There is no major evaporation. Also, when you look at the plaster mould at this stage, you observe water coming out from the surface. Its is not exactly bubbles. I think replacement of water with air is the best description. I do not know how I should define the domains in CFX. Do I need to have more that one domain or just one porous domain is enough? Is plaster (porous media) going to be my main domain? If so, should I define it as a "continuous fluid" or "dispersed solid"? Well the mould is actually a solid but as far as I know dispersed solid is used to define particles inside a fluid domain. So the only remaining option for me would be continuous fluid which does not sound correct to me. What do you think?

February 16, 2012, 19:12		#6
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,852 Rep Power: 144	There are tutorial examples on porous domains. Do these to see how to set up porous regions. The physics of air pushing a liquid through a porous domain is very complex and CFX does not have a model developed to model it. I talked to a bloke who did some modelling on this - here is a paper of his (http://www.springerlink.com/content/8430t042j6gqx086/). You are probably going to have to compromise, and work out if a free surface flow with some form of resistance from the porous medium will work, or if you can go an additional variable approach can work. There are tutorials on both of these approaches, I recommend you do them both (in fact all the tutorials) before proceeding. Forget about how you define these things until you know how you are going to model the basic physics. Once you know the basic physics then worry about the implementation.

February 16, 2012, 19:19		#7
sakalido Member Sara S. Join Date: May 2010 Location: NJ, USA Posts: 41 Rep Power: 16	Thanks a lot Glenn, the paper helps a lot. I actually looked through the tutorials but in this version of ANSYS-CFX I am using currently (version11), there is only a catalytic converter which uses the directional loss model. I will check the newer ANSYS versions for the tutorials you mentioned. Thank you again for the good advice Sara