High Skewness in Hexcore Mesh/ Tgrid

xaero · November 7, 2009, 08:55

Hello,
I have the problem of a very high skewness at the hexcore mesh in the image below. (The model is a 3D- Scan of a Land Rover) I have wrapped the surface of this model and refined it down to a skewness of the boundary layer of 0.65. After inserting the bounding box and initializing a tetrahedral mesh upon the boundary layers, i did a hexcore mesh- which then resulted in a skewness of >.99 . Does anyone have any advice for me to bring it down? Is it because of the grubby wrapping surface? Or is there a better solution than intersecting fromn tetrahedrals to hexahedral cells?

Note:If you are wondering about the size of the bounding box: I am currently writing my first bachelor's thesis on the topic of grid generation in various software- the model is not intended to be solved in Fluent- so I kept the grid size small for faster grid generation results.

I would be very thankful for every advice given.
Greetings, Fabian

jchawner · November 7, 2009, 12:14

Fabian:

I can't help you with you question but I was wondering if you'd be willing to share the results of your thesis when it's complete. It will probably be very interesting.

ztdep · November 7, 2009, 21:33

Dear friends:
It seems that we have somthing in common. i am writing a code for the adaptive cartesian grid solver.
Since you use the cartesian grid in most of the river, why not use it at the boudnary. the cartesian grid is very easy to adapt to your bondary.
ztdep@yahoo.com.cn

xaero · November 8, 2009, 20:24

Thank you for your responses

@jchawner
Of course I can share the results of my thesis with the community. It will be completed at the end of this month. However, it might only be interesting for beginners in Tgrid and Gambit as I only compare these two programs regarding meshing possibilities and results in skewness.

@ztdep
Do you mean to apply the cartesian grid upon the bounding box? I never thought of that- but it sounds plausible

I will try it tomorrow an then post my results!

greetings, Fabian

xaero · November 10, 2009, 12:04

Hello again,
I have tried the advice of ztdep to apply a cartesian grid over the bounding box. Sadly it resulted in the same high skewness.
Then i tried the keep outer domain- function provided in the hexcore controls menu to create a bounding box. That finally worked and brought the skewness down below 0.8

Hint: Because the initial cell count for generating a outer domain is limited by 1.500.000, it has to be increased manually with the TUI command mesh/controls/hexcore/maximum-initial-cells

greets, Fabian

November 7, 2009, 08:55	High Skewness in Hexcore Mesh/ Tgrid	#1
xaero New Member Join Date: Oct 2009 Posts: 11 Rep Power: 17	Hello, I have the problem of a very high skewness at the hexcore mesh in the image below. (The model is a 3D- Scan of a Land Rover) I have wrapped the surface of this model and refined it down to a skewness of the boundary layer of 0.65. After inserting the bounding box and initializing a tetrahedral mesh upon the boundary layers, i did a hexcore mesh- which then resulted in a skewness of >.99 . Does anyone have any advice for me to bring it down? Is it because of the grubby wrapping surface? Or is there a better solution than intersecting fromn tetrahedrals to hexahedral cells? Note:If you are wondering about the size of the bounding box: I am currently writing my first bachelor's thesis on the topic of grid generation in various software- the model is not intended to be solved in Fluent- so I kept the grid size small for faster grid generation results. I would be very thankful for every advice given. Greetings, Fabian

November 7, 2009, 12:14		#2
jchawner Senior Member John Chawner Join Date: Mar 2009 Location: Fort Worth, Texas, USA Posts: 275 Rep Power: 18	Fabian: I can't help you with you question but I was wondering if you'd be willing to share the results of your thesis when it's complete. It will probably be very interesting. __________________ John Chawner / jrc@pointwise.com / www.pointwise.com Blog: http://blog.pointwise.com/ on Twitter: @jchawner

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Mesh skewness	Woo Meng Wai	FLUENT	3	June 20, 2010 11:04
[blockMesh] Mesh skewness warning how to get rid	ploceus	OpenFOAM Meshing & Mesh Conversion	2	July 1, 2009 06:01
How to control Minximum mesh space?	hung	FLUENT	7	April 18, 2005 10:38
Tgrid mesh	Samir	FLUENT	1	February 17, 2005 06:57
unstructured vs. structured grids	Frank Muldoon	Main CFD Forum	1	January 5, 1999 11:09

November 7, 2009, 21:33		#3
ztdep Senior Member p ding Join Date: Mar 2009 Posts: 427 Rep Power: 19	Dear friends: It seems that we have somthing in common. i am writing a code for the adaptive cartesian grid solver. Since you use the cartesian grid in most of the river, why not use it at the boudnary. the cartesian grid is very easy to adapt to your bondary. ztdep@yahoo.com.cn

November 8, 2009, 20:24		#4
xaero New Member Join Date: Oct 2009 Posts: 11 Rep Power: 17	Thank you for your responses @jchawner Of course I can share the results of my thesis with the community. It will be completed at the end of this month. However, it might only be interesting for beginners in Tgrid and Gambit as I only compare these two programs regarding meshing possibilities and results in skewness. @ztdep Do you mean to apply the cartesian grid upon the bounding box? I never thought of that- but it sounds plausible I will try it tomorrow an then post my results! greetings, Fabian

November 10, 2009, 12:04		#5
xaero New Member Join Date: Oct 2009 Posts: 11 Rep Power: 17	Hello again, I have tried the advice of ztdep to apply a cartesian grid over the bounding box. Sadly it resulted in the same high skewness. Then i tried the keep outer domain- function provided in the hexcore controls menu to create a bounding box. That finally worked and brought the skewness down below 0.8 Hint: Because the initial cell count for generating a outer domain is limited by 1.500.000, it has to be increased manually with the TUI command mesh/controls/hexcore/maximum-initial-cells greets, Fabian