|
[Sponsors] |
February 14, 2021, 03:54 |
Cell pair was created...
|
#1 |
New Member
Guillaume Beardsell
Join Date: Aug 2012
Posts: 19
Rep Power: 14 |
Hi all,
I am running simulations of IC engines (moving boundaries) and shortly after I start the simulation I get this warning many times: A cell pair was created with 16 cells at 0.00789812 0.00815266 0.00135936. This cell may reduce accuracy and/or stability. Consider refining the grid in this region. This warning will only be printed 50 times per process What does it mean and how can I fix my mesh? I tried to refine the mesh as suggested but I keep getting the warning, so I am thinking that it might have something to do with my boundaries... Cheers, Guillaume |
|
February 16, 2021, 10:25 |
|
#2 | |
Member
John Etcheverry
Join Date: Dec 2018
Posts: 56
Rep Power: 8 |
Quote:
Hi, Guillaume! When CONVERGE creates a mesh it looks at the volume of a cell cut by a surface compared to a normal neighboring cell. If the cut cell is less than 30% of the volume of the neighbor, the two are combined. If more cells are needed it combines more neighbors. If too many are needed the quality of the mesh can suffer - this is the warning you are seeing. Often, in a ICE simulation this can occur in the valve gap. Does the coordinate provided match with the valves? Also, how big is the gap in your simulation and how big are the cells in it? Thanks! - John |
||
February 16, 2021, 13:55 |
|
#3 |
New Member
Guillaume Beardsell
Join Date: Aug 2012
Posts: 19
Rep Power: 14 |
Hi John,
Thanks for your answer! Here is a slice showing the mesh around the valves. https://drive.google.com/file/d/1gmG...ew?usp=sharing The cell size is 1mm around the valves, and the valve gap is 0.2 mm initially (closed position). When fully valve is fully open, the gap is about 6mm. I don't understand what you mean when you ask if the coordinate provided match with the valves. I believe that I need to refine the mesh locally around the valve gap? |
|
February 16, 2021, 14:58 |
|
#4 | |
Member
John Etcheverry
Join Date: Dec 2018
Posts: 56
Rep Power: 8 |
Quote:
As for the point about checking if the coordinate provided is in the area of the valve - you can copy that coordinate and make a vertex at that point to see where it is. If it is in or close to the valve gap that would tell you if the valve is a possible culprit in your problem. Thanks! - John |
||
February 16, 2021, 15:48 |
|
#5 |
New Member
Guillaume Beardsell
Join Date: Aug 2012
Posts: 19
Rep Power: 14 |
John,
Indeed the coordinate of the problematic cell is very close to the valve angle. Adding embedding at the valve angle and seat solved the issue, I am not getting the cell pairing warning message anymore. Thank you for your help! Guillaume |
|
February 16, 2021, 15:51 |
|
#6 | |
Member
John Etcheverry
Join Date: Dec 2018
Posts: 56
Rep Power: 8 |
Quote:
- John |
||
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
Error when running BuoyantBoussinesqSimpleFoam | b.simpson | OpenFOAM Running, Solving & CFD | 8 | June 26, 2020 10:18 |
Neighboring cells in tetrahedral mesh | vishwesh | OpenFOAM Programming & Development | 9 | November 10, 2017 08:06 |
[snappyHexMesh] Unwanted mesh created by snappyHexMesh--baffle | bieshuxuhe | OpenFOAM Meshing & Mesh Conversion | 2 | April 23, 2014 02:59 |
[Commercial meshers] Trimmed cell and embedded refinement mesh conversion issues | michele | OpenFOAM Meshing & Mesh Conversion | 2 | July 15, 2005 05:15 |
cell to cell relation | CMB | Siemens | 1 | December 4, 2003 05:05 |