Steady state heat transfer simulation convergence

CFD10111 · January 3, 2023, 04:56

Hi,

I am quite new to CFD so some suggestions on this would be appreciated. I am trying to perform a mesh sensitivity study. At the moment, I have made an initial mesh with suitable metric values for orthogonality, skewness and aspect ratio, and the like. However my solution has not converged with this initial mesh, but could this be because my mesh has not converged? or because of other aspects such as under relaxation? So it got me thinking: should I achieve solution convergence before mesh convergence or during?

Because when doing a mesh sensitivity study, shouldn't I only change the mesh density and no other things such as the under relaxation factors? Not sure what order to do this in.

Thank you

LoGaL · January 3, 2023, 07:14

No, mesh sensitivity study has to be done only on converged solutions

CFDKareem · January 3, 2023, 10:32

I agree with LoGal.

A good quality mesh (no small/bad elements) should not cause non-convergence. A coarse mesh should still converge, but will have a large error compared to the correct solution. The goal of a mesh sensitivity study is to reduce this error, not achieve convergence.

It is much more likely that your boundary conditions, solver settings, etc. are causing the problem with convergence. Without much detail of your setup it is hard to speculate what is causing the issue.

CFD10111 · January 4, 2023, 08:50

Thank you both. This is what I needed to know

January 3, 2023, 04:56	Steady state heat transfer simulation convergence	#1
CFD10111 New Member Avon Join Date: Aug 2022 Posts: 26 Rep Power: 3	Hi, I am quite new to CFD so some suggestions on this would be appreciated. I am trying to perform a mesh sensitivity study. At the moment, I have made an initial mesh with suitable metric values for orthogonality, skewness and aspect ratio, and the like. However my solution has not converged with this initial mesh, but could this be because my mesh has not converged? or because of other aspects such as under relaxation? So it got me thinking: should I achieve solution convergence before mesh convergence or during? Because when doing a mesh sensitivity study, shouldn't I only change the mesh density and no other things such as the under relaxation factors? Not sure what order to do this in. Thank you

January 3, 2023, 07:14		#2
LoGaL Senior Member Lorenzo Galieti Join Date: Mar 2018 Posts: 373 Rep Power: 12	No, mesh sensitivity study has to be done only on converged solutions CFD10111 likes this.

January 3, 2023, 10:32		#3
CFDKareem Senior Member Kareem Join Date: Nov 2022 Location: New York Posts: 116 Rep Power: 3	I agree with LoGal. A good quality mesh (no small/bad elements) should not cause non-convergence. A coarse mesh should still converge, but will have a large error compared to the correct solution. The goal of a mesh sensitivity study is to reduce this error, not achieve convergence. It is much more likely that your boundary conditions, solver settings, etc. are causing the problem with convergence. Without much detail of your setup it is hard to speculate what is causing the issue. CFD10111 likes this. __________________ Please like the answer if it helped! Video Tutorials and Tips: https://www.youtube.com/@cfdkareem/featured

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Setting the height of the stream in the free channel	kevinmccartin	CFX	12	October 13, 2022 21:43
Domain Reference Pressure and mass flow inlet boundary	AdidaKK	CFX	75	August 20, 2018 05:37
mass flow in is not equal to mass flow out	saii	CFX	12	March 19, 2018 05:21
Solver for transonic flow?	Martin Hegedus	OpenFOAM Running, Solving & CFD	22	December 16, 2015 04:59
Constant velocity of the material	Sas	CFX	15	July 13, 2010 08:56

January 4, 2023, 08:50		#4
CFD10111 New Member Avon Join Date: Aug 2022 Posts: 26 Rep Power: 3	Thank you both. This is what I needed to know