Transient heat discharge from Phase change material : Solution strategies

yogesh03y · November 18, 2022, 07:37

Hello all,

I am trying to simulate the transient heat discharge from a phase change material (pcm) to air flowing in a channel below it. I defined my pcm as a fluid in order to simulate the solidification when it cools down. I'm using the solidification/melting model. I have defined temperature dependent properties for air, pcm and solid materials. I use the patch option after initialization for setting the initial temperature of the pcm. The computational time is too high if I solve the flow and energy equations at each iteration and time step and I am looking for solution strategies.

1. Is it possible to turn off flow solution for the pcm cell zone as I am only interested in its temperature and liquid fraction over time?

2. Is it possible to solve the energy equation at each time step but the flow equation of air not at every time step but at a desired frequency of time steps? How does the Loosely Coupled Conjugate Heat Transfer option work?

I've attached some screenshots for reference. Has anyone solved such problems before?

Thank you so much!

fredo490 · November 24, 2022, 01:16

You could try to model you PCM with a UDF. Have a couple of UDS to keep track of the key variables and then have the UDF to modify the material properties dynamically.

The entire PCM could be considered as a solid block (non matter the phase) resulting in a simple conduction equation.

yogesh03y · November 28, 2022, 15:10

Hey fredo490,

Thanks for the reply. I am interested in the solidification process and discharge of latent heat during phase change. So defining the PCM as solid is not an option. However, I figured out why I was getting poor results. With gravity enabled, it was not possible to define an operating density suitable for both air and PCM.

fredo490 · November 29, 2022, 02:29

Setting it as solid does not mean you ignore the solidification and latent heat. It just mean that you ignore all the fluid "motion" for this material. This greatly simplify your system to only consider conduction and ignore the convection part.

The solidification and latent heat can be modeled manually using simple UDF (user defined function).

yogesh03y · November 30, 2022, 05:37

Hey Fredo490,

Thanks for the information. I'll try it out.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
What are the best settings for a channel flow simulation?	Ashkan Kashani	CFX	3	October 13, 2022 21:36
Heat transfer problem with Phase Change Material	obiscolly50	Fluent Multiphase	0	August 5, 2016 09:21
Effective heat capacity - Phase change does not work!!!	papteo	CFX	8	October 31, 2013 05:15
Two-Phase Buoyant Flow Issue	Miguel Baritto	CFX	4	August 31, 2006 12:02
help needed about phase change	Yanhu Guo	Main CFD Forum	4	January 23, 2001 23:16

November 24, 2022, 01:16		#2
fredo490 Senior Member HECKMANN Frédéric Join Date: Jul 2010 Posts: 249 Rep Power: 17	You could try to model you PCM with a UDF. Have a couple of UDS to keep track of the key variables and then have the UDF to modify the material properties dynamically. The entire PCM could be considered as a solid block (non matter the phase) resulting in a simple conduction equation.

November 28, 2022, 15:10		#3
yogesh03y New Member Yogesh Join Date: Sep 2022 Posts: 12 Rep Power: 3	Hey fredo490, Thanks for the reply. I am interested in the solidification process and discharge of latent heat during phase change. So defining the PCM as solid is not an option. However, I figured out why I was getting poor results. With gravity enabled, it was not possible to define an operating density suitable for both air and PCM.

November 29, 2022, 02:29		#4
fredo490 Senior Member HECKMANN Frédéric Join Date: Jul 2010 Posts: 249 Rep Power: 17	Setting it as solid does not mean you ignore the solidification and latent heat. It just mean that you ignore all the fluid "motion" for this material. This greatly simplify your system to only consider conduction and ignore the convection part. The solidification and latent heat can be modeled manually using simple UDF (user defined function).

November 30, 2022, 05:37		#5
yogesh03y New Member Yogesh Join Date: Sep 2022 Posts: 12 Rep Power: 3	Hey Fredo490, Thanks for the information. I'll try it out.