[wongkinfeng]

How to insert the new material (PCM—Water and Ice) into CFX-PRE

PCM with a two-component model, as shown in the figure below

Screenshot 2023-10-31 003028.png

[ghorrocks]

What are you trying to do? How do the water and ice interact? Is the ice particles in the water? Or is it a water/ice slush? Is there phase change?

Note that CFX does not have a solidification or melting model.

[wongkinfeng]

Originally Posted by ghorrocks

What are you trying to do? How do the water and ice interact? Is the ice particles in the water? Or is it a water/ice slush? Is there phase change?

Note that CFX does not have a solidification or melting model.

I'm doing a simulation of a thermal energy storage tank. Screenshot 2023-10-31 175424.jpg
The spring-shaped tube is Heat transfer fluid (Air)
The cylindrical tank is PCM (Water)
I need to simulate melting and freezing processes, so I need to add a material for PCM.

[ghorrocks]

How do you plan to do this given in my last post I said:

Note that CFX does not have a solidification or melting model.

Either you are going to have to develop the solidification model yourself (that is not simple) or switch to another software which does support it.

[wongkinfeng]

Originally Posted by ghorrocks

How do you plan to do this given in my last post I said:



Either you are going to have to develop the solidification model yourself (that is not simple) or switch to another software which does support it.

My project must use CFX, so I came here to find the solution. Anyway Thanks your suggestion

[falsc233]

Originally Posted by ghorrocks

How do you plan to do this given in my last post I said:



Either you are going to have to develop the solidification model yourself (that is not simple) or switch to another software which does support it.

hi Glenn, i've seen people implementing the enthalpy-porosity method for melting/solidification problems in CFX by defining a binary mixture of liquid and solid, where the specific heat for each material (phase) is specified to account for the latent heat. A momentum source based on liquid fraction is then applied to dampen fluid motion so that there's no motion in the solid region. Is this type of approach what you mean when you say "develop the solidification model yourself"? Or were you referring to something more fundamental than this? Thank you!

[ghorrocks]

Yes, there are ways of doing it and you have described one - and you can see that there are several features you will have to develop yourself to get it to have the physics you need. You will have to make sure the latent heats are properly handled, the source term is not too strong and not too weak, that is is numerical stable and so on - so you will need a good knowledge of CFD to make it work. If you are confident you can develop this yourself then go for it.