Conjugate heat exchange

alain.wokusch · September 17, 2021, 09:27

I have been told that conjugate heat transfer in OpenFOAM is not very performant. But there is a workaround or a more advanced solution available that is supposed to perform better.

Unfortunately, I can't figure out what exactly the problem is and what the proposed solution would be. Does anyone know more?

Bloerb · September 18, 2021, 09:15

This is not true. It's quite powerful, but there are some limitations. If those apply depends on your problem however. The main limitation is that the regions are not implicitly coupled. There is such a solver in foamExtend. This is however often not a problem. Another one might be speed in transient simulations where either the flow or solid domain are limiting you with their diffusion or courant number. OpenFOAM couples the regions in every time step. And solves each region in every time step. If for example your flow is solved on a 1e-6 time step due the velocities and your temperature does not change on such small scales but rather in a minute like fashion, you might face extremly long computation times without tweaks.

clapointe · September 18, 2021, 13:45

A quick note -- it looks like there's an (semi)implicit coupling option available through the dev branch (and slated for v2112 release) now too: link. I've not tested it yet, but it could be worth looking into!

Caelan

September 17, 2021, 09:27	Conjugate heat exchange	#1
alain.wokusch New Member Alain Wokusch Join Date: Jun 2015 Posts: 2 Rep Power: 0	I have been told that conjugate heat transfer in OpenFOAM is not very performant. But there is a workaround or a more advanced solution available that is supposed to perform better. Unfortunately, I can't figure out what exactly the problem is and what the proposed solution would be. Does anyone know more? Last edited by alain.wokusch; September 17, 2021 at 14:52. Reason: fix typo

September 18, 2021, 09:15		#2
Bloerb Senior Member Join Date: Sep 2013 Posts: 353 Rep Power: 21	This is not true. It's quite powerful, but there are some limitations. If those apply depends on your problem however. The main limitation is that the regions are not implicitly coupled. There is such a solver in foamExtend. This is however often not a problem. Another one might be speed in transient simulations where either the flow or solid domain are limiting you with their diffusion or courant number. OpenFOAM couples the regions in every time step. And solves each region in every time step. If for example your flow is solved on a 1e-6 time step due the velocities and your temperature does not change on such small scales but rather in a minute like fashion, you might face extremly long computation times without tweaks. alain.wokusch likes this.

September 18, 2021, 13:45		#3
clapointe Senior Member Join Date: Aug 2015 Posts: 494 Rep Power: 15	A quick note -- it looks like there's an (semi)implicit coupling option available through the dev branch (and slated for v2112 release) now too: link. I've not tested it yet, but it could be worth looking into! Caelan alain.wokusch likes this. __________________ Public git repository : https://github.com/clapointe2011/public

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Conjugate Heat transfer- Heat generating cylinder & natural convection	sandymech	OpenFOAM Running, Solving & CFD	6	July 3, 2018 14:40
conjugate heat transfer in OpenFOAM	skuznet	OpenFOAM Running, Solving & CFD	99	March 16, 2017 06:07
heat flux BC in conjugate heat transfere	skuznet	OpenFOAM Pre-Processing	8	June 24, 2014 07:00
heat flux on solid surface in conjugate heat transfer	skuznet	OpenFOAM Pre-Processing	0	December 10, 2013 01:36
Question about heat transfer coefficient setting for CFX	Anna Tian	CFX	1	June 16, 2013 07:28