Sharp tip cone in cylindrical test section

jxs832 · April 19, 2018, 07:43

Hello,

I am trying to mesh my current wind tunnel configuration. The supersonic wind tunnel section is cylindrical (a big diameter pipe essentially) with a sharp tip cone moved off of the center line, therefore I cannot revolve the mesh. What would be best topology for this? The sharp tip always ends up in a pole, forcing me to some type of circular grid all around. Is there a way to avoid the pole at the center of the cone? I am trying to run this in OpenFoam. If not, any advice would be helpful. You can see some pictures of what I am describing and what my attempts have been made. I have deleted some faces and contactors near the cone so you can have a better view at the cone.

Thank you

https://imgur.com/a/8YsEzWD

dgarlisch · April 20, 2018, 09:53

Please read this and then add more information.

jxs832 · April 22, 2018, 18:48

I have added a bit more info and added a few pictures that describe the problem. Please let me know if my problem is clear now.

dgarlisch · April 23, 2018, 10:42

There are far more experienced meshers on this forum than me. However, my observations are...

The images are "far away" so I can't see the tip topology. However, I suspect the first row of cells on the cone at the tip are "illegal". You can check them with a cutting plane in Examine, Cell Type. I suggest using a shrink factor around 0.8 or 0.9.

See the cell topology page and scroll down to see QuadTri cell connectivity (Unsupported). This is a very common situation when rotating domains around a sharp tipped surface.

Someone else will need to offer a solution to this to stay with all structured blocks.

Which begs the question, is a fully structured mesh a requirement?

OpenFOAM does not require it.

Pointwise supports hybrid meshes (structured and unstructured blocks in one grid).

I imagine there is a way you could build a small unstructured block around the cone's tip that will provide a nice transition to the rest of the grid made with structured blocks without poles.

One final comment. No matter what, OpenFOAM meshes exported from Pointwise should always be run through renumberMesh before solving.

Good luck.

Josenoah · April 23, 2018, 11:06

how do i post new massege to the forum

Josenoah · April 23, 2018, 11:09

How do i post new message/thread to the forum

dgarlisch · April 23, 2018, 12:11

Quote:

Originally Posted by Josenoah

How do i post new message/thread to the forum

Click down to the appropriate forum and click the "New Thread" button at the top of the page.

Please see the attached image.

Josenoah · April 24, 2018, 10:17

Dear David,

Thank a lot i have got it.

Now i have been trying the tutorials but when i reach polyMesh command it

tells me no such a file/directory or at times gives not found.

What could be the problem and How do i need to handle this so that i can proceed to the next command.

Thanks.

dgarlisch · April 24, 2018, 10:29

Quote:

Originally Posted by Josenoah

Dear David,

Thank a lot i have got it.

Now i have been trying the tutorials but when i reach polyMesh command it

tells me no such a file/directory or at times gives not found.

What could be the problem and How do i need to handle this so that i can proceed to the next command.

Thanks.

Josenoah,

Please stop posting in this thread. Posting unrelated questions and comments is called "hijacking".

Please start a new top level post with your question and more information.

RcktMan77 · June 5, 2018, 14:55

Try this topology instead...

hconel · December 24, 2020, 13:21

hi,

i am also having a similar problem. i was not able to rotate-extrude neither of the the following domains about the x-axis:

A

B

note that i have tried quad cells adjacent to the x-axis thinking that maybe the tri cells were causing the problem. but no.

supported and unsupported cell types are given ere: http://www.pointwise.com/plugins/htm...ELL__TOPO.html
and none of my cells violate the rules:

from what i can deduce is: only a fully structured domain can be rotate-extruded about a line that is adjacent to the domain. if i am not doing something wrong, i find this somehow ridiculous.

pdp.aero · January 5, 2021, 12:22

extrude/rotate for unstructured domains are unsupported. Switch to structured grid and regenerate your mesh, you'll survive.

You may also split your domain, keep the bottom part structured and do the extrude/rotate and then build the upper part unstructured using advanced front in the solve panel with quads without doing extrude/rotate.

April 19, 2018, 07:43	Sharp tip cone in cylindrical test section	#1
jxs832 New Member Join Date: Mar 2014 Posts: 7 Rep Power: 12	Hello, I am trying to mesh my current wind tunnel configuration. The supersonic wind tunnel section is cylindrical (a big diameter pipe essentially) with a sharp tip cone moved off of the center line, therefore I cannot revolve the mesh. What would be best topology for this? The sharp tip always ends up in a pole, forcing me to some type of circular grid all around. Is there a way to avoid the pole at the center of the cone? I am trying to run this in OpenFoam. If not, any advice would be helpful. You can see some pictures of what I am describing and what my attempts have been made. I have deleted some faces and contactors near the cone so you can have a better view at the cone. Thank you https://imgur.com/a/8YsEzWD Last edited by jxs832; April 22, 2018 at 18:48. Reason: Adding more information

April 22, 2018, 18:48	Additional information	#3
jxs832 New Member Join Date: Mar 2014 Posts: 7 Rep Power: 12	I have added a bit more info and added a few pictures that describe the problem. Please let me know if my problem is clear now.

April 23, 2018, 11:06	I am new member	#5
Josenoah New Member Esemu Joseph Noah Join Date: Apr 2018 Posts: 7 Rep Power: 8	how do i post new massege to the forum

April 23, 2018, 11:09	I am new member	#6
Josenoah New Member Esemu Joseph Noah Join Date: Apr 2018 Posts: 7 Rep Power: 8	How do i post new message/thread to the forum

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
critical error during installation of openfoam	Fabio88	OpenFOAM Installation	21	June 2, 2010 03:01
OF 1.6 \| Ubuntu 9.10 (64bit) \| GLIBCXX_3.4.11 not found	piprus	OpenFOAM Installation	22	February 25, 2010 13:43
Problems in compiling paraview in Suse 10.3 platform	chiven	OpenFOAM Installation	3	December 1, 2009 07:21
Surface roughness setting for a model in wind tunnel test section	Sharon	CFX	12	March 31, 2009 13:09
open test section	white	FLUENT	0	February 25, 2005 02:33

April 20, 2018, 09:53		#2
dgarlisch Senior Member David Garlisch Join Date: Jan 2013 Location: Fidelity Pointwise, Cadence Design Systems (Fort Worth, Texas Office) Posts: 307 Rep Power: 14	Please read this and then add more information.

April 23, 2018, 10:42		#4
dgarlisch Senior Member David Garlisch Join Date: Jan 2013 Location: Fidelity Pointwise, Cadence Design Systems (Fort Worth, Texas Office) Posts: 307 Rep Power: 14	There are far more experienced meshers on this forum than me. However, my observations are... The images are "far away" so I can't see the tip topology. However, I suspect the first row of cells on the cone at the tip are "illegal". You can check them with a cutting plane in Examine, Cell Type. I suggest using a shrink factor around 0.8 or 0.9. See the cell topology page and scroll down to see QuadTri cell connectivity (Unsupported). This is a very common situation when rotating domains around a sharp tipped surface. Someone else will need to offer a solution to this to stay with all structured blocks. Which begs the question, is a fully structured mesh a requirement? OpenFOAM does not require it. Pointwise supports hybrid meshes (structured and unstructured blocks in one grid). I imagine there is a way you could build a small unstructured block around the cone's tip that will provide a nice transition to the rest of the grid made with structured blocks without poles. One final comment. No matter what, OpenFOAM meshes exported from Pointwise should always be run through renumberMesh before solving. Good luck.

April 24, 2018, 10:17		#8
Josenoah New Member Esemu Joseph Noah Join Date: Apr 2018 Posts: 7 Rep Power: 8	Dear David, Thank a lot i have got it. Now i have been trying the tutorials but when i reach polyMesh command it tells me no such a file/directory or at times gives not found. What could be the problem and How do i need to handle this so that i can proceed to the next command. Thanks.

December 24, 2020, 13:21		#11
hconel Member Hüseyin Can Önel Join Date: Sep 2018 Location: Ankara, Turkey Posts: 46 Rep Power: 7	hi, i am also having a similar problem. i was not able to rotate-extrude neither of the the following domains about the x-axis: A B note that i have tried quad cells adjacent to the x-axis thinking that maybe the tri cells were causing the problem. but no. supported and unsupported cell types are given ere: http://www.pointwise.com/plugins/htm...ELL__TOPO.html and none of my cells violate the rules: from what i can deduce is: only a fully structured domain can be rotate-extruded about a line that is adjacent to the domain. if i am not doing something wrong, i find this somehow ridiculous.

January 5, 2021, 12:22		#12
pdp.aero Senior Member Pay D. Join Date: Aug 2011 Posts: 166 Blog Entries: 1 Rep Power: 14	extrude/rotate for unstructured domains are unsupported. Switch to structured grid and regenerate your mesh, you'll survive. You may also split your domain, keep the bottom part structured and do the extrude/rotate and then build the upper part unstructured using advanced front in the solve panel with quads without doing extrude/rotate.