Mesh Convergence Analysis

yeongjia · September 8, 2016, 05:14

Hey guys, so just a quick question. I did a mesh sensitivity test on my step model in ANSYS Fluent. My main aim is to compare the velocity and pressure profile with respect to the mesh resolution at a specified location (past reattachment point).

I have specified a region in the previous steps in order to refine its mesh resolution. I ran 6 different simulations using:

101 cells
201 cells
401 cells
541 cells
801 cells
1001 cells

I have then extracted the datas from Post-CFD under results and plotted a graph accordingly.

Back to my question: I have obtained different results for my simulation on the 801 and 1001 cells (which was expected). But the simulation run of 101, 201, 401 and 541 cells seems to be exactly the same. I am wondering if this was supposed to happen. Even after plotting the graphs, the data from the 101, 201, 401 and 541 cells over lap each other.

How do I fix this, is there an explanation to this problem?

Thanks in advance,

Regards,

J

mome · September 8, 2016, 12:20

Hi,

are you sure you extracted the correct data? With such a massive variation in datapoints, there should be noticeable difference. By the way, I am surprised by these low cell counts, or did I misinterpret? Or are you just calculating a very simple problem?

Best, Mo

nvarma · September 8, 2016, 15:15

in addition to what morten asked. what turbulence model are you using? what's the y+ value for each of these meshes?

T-models switch between wall functions and resolution depending upon y+ (refer manual).

bhargavbharathan · September 8, 2016, 17:21

Hi,

It is possible that 101 to 541 cells may be one too few for your case. Have you tried with cases with more than 1001 cells?

-BB

yeongjia · September 10, 2016, 01:11

Hey guys, thank you all so much for helping out. I managed to figure out the problem. Turns out that it was just a simple error in one of the simulation runs, i somehow left out an important parameter which caused the result outcome to appear differently.

Thanks again!!

September 8, 2016, 05:14	Mesh Convergence Analysis	#1
yeongjia New Member Join Date: Aug 2016 Posts: 8 Rep Power: 9	Hey guys, so just a quick question. I did a mesh sensitivity test on my step model in ANSYS Fluent. My main aim is to compare the velocity and pressure profile with respect to the mesh resolution at a specified location (past reattachment point). I have specified a region in the previous steps in order to refine its mesh resolution. I ran 6 different simulations using: 101 cells 201 cells 401 cells 541 cells 801 cells 1001 cells I have then extracted the datas from Post-CFD under results and plotted a graph accordingly. Back to my question: I have obtained different results for my simulation on the 801 and 1001 cells (which was expected). But the simulation run of 101, 201, 401 and 541 cells seems to be exactly the same. I am wondering if this was supposed to happen. Even after plotting the graphs, the data from the 101, 201, 401 and 541 cells over lap each other. How do I fix this, is there an explanation to this problem? Thanks in advance, Regards, J

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
decomposePar problem: Cell 0contains face labels out of range	vaina74	OpenFOAM Pre-Processing	37	July 20, 2020 05:38
[mesh manipulation] Importing Multiple Meshes	thomasnwalshiii	OpenFOAM Meshing & Mesh Conversion	18	December 19, 2015 18:57
Naca 0012 (compressible and inviscid) flow convergence problem	bipulsaha	FLUENT	1	July 6, 2011 07:51
early stall, poor convergence, and mesh quality	everest	CFX	2	May 12, 2010 16:27
Icemcfd 11: Loss of mesh from surface mesh option?	Joe	CFX	2	March 26, 2007 18:10

September 8, 2016, 12:20		#2
mome Member Join Date: May 2014 Posts: 30 Rep Power: 11	Hi, are you sure you extracted the correct data? With such a massive variation in datapoints, there should be noticeable difference. By the way, I am surprised by these low cell counts, or did I misinterpret? Or are you just calculating a very simple problem? Best, Mo

September 8, 2016, 15:15		#3
nvarma Senior Member nm Join Date: Mar 2013 Posts: 100 Rep Power: 13	in addition to what morten asked. what turbulence model are you using? what's the y+ value for each of these meshes? T-models switch between wall functions and resolution depending upon y+ (refer manual).

September 8, 2016, 17:21		#4
bhargavbharathan Member Bhargav Bharathan Join Date: Jun 2015 Location: Montreal, Canada Posts: 71 Rep Power: 10	Hi, It is possible that 101 to 541 cells may be one too few for your case. Have you tried with cases with more than 1001 cells? -BB

September 10, 2016, 01:11		#5
yeongjia New Member Join Date: Aug 2016 Posts: 8 Rep Power: 9	Hey guys, thank you all so much for helping out. I managed to figure out the problem. Turns out that it was just a simple error in one of the simulation runs, i somehow left out an important parameter which caused the result outcome to appear differently. Thanks again!!