CFD Online Logo CFD Online URL
www.cfd-online.com
[Sponsors]
Home > Forums > Pointwise & Gridgen

wing with sharp trailing edge mesh through extrusion

Register Blogs Members List Search Today's Posts Mark Forums Read

Like Tree2Likes
  • 1 Post By cnsidero
  • 1 Post By jchawner

Reply
 
LinkBack Thread Tools Display Modes
Old   February 21, 2013, 13:16
Default wing with sharp trailing edge mesh through extrusion
  #1
Senior Member
 
Ali
Join Date: Jan 2012
Location: Pakistan
Posts: 135
Rep Power: 9
Hybrid is on a distinguished road
How can I extrude a wing having sharp trailing edge?

I have come to know that by adding wake domain, the select wing domains and wake domain and extrude all of them at once. By doing so, I get a mesh with wing domains (upper and lower) extruded up and down while wake domain extruded only in one direction.

How can I get the wake domains extruded on both sides?

If one can try, can assume any wing with sharp trailing edge.
__________________
Best Regards
Ali


Hybrid is offline   Reply With Quote

Old   February 24, 2013, 17:25
Default
  #2
Senior Member
 
John Chawner
Join Date: Mar 2009
Location: Fort Worth, Texas, USA
Posts: 256
Rep Power: 11
jchawner is on a distinguished road
I'm going to attempt an answer without actually trying it myself first. When you define the initial face from which extrusion will begin (clockwise: wake, lower wing, upper wing, wake) you have to select the wake domain twice.
__________________
John Chawner / jrc@pointwise.com / www.pointwise.com
Blog: http://blog.pointwise.com/
on Twitter: @jchawner
jchawner is offline   Reply With Quote

Old   February 25, 2013, 10:38
Default
  #3
Senior Member
 
Ali
Join Date: Jan 2012
Location: Pakistan
Posts: 135
Rep Power: 9
Hybrid is on a distinguished road
I tried this but I am not able to select a domain twice in Pointwise. You can try using any generic wing with a sharp trailing edge. I wand to use NORMAL extrusion.
__________________
Best Regards
Ali


Hybrid is offline   Reply With Quote

Old   February 25, 2013, 10:50
Default
  #4
Senior Member
 
John Chawner
Join Date: Mar 2009
Location: Fort Worth, Texas, USA
Posts: 256
Rep Power: 11
jchawner is on a distinguished road
That's the danger of me answering without trying it myself. At this point your best bet is to contact our tech support team directly at support@pointwise.com.
__________________
John Chawner / jrc@pointwise.com / www.pointwise.com
Blog: http://blog.pointwise.com/
on Twitter: @jchawner
jchawner is offline   Reply With Quote

Old   February 25, 2013, 23:34
Default
  #5
Senior Member
 
Chris Sideroff
Join Date: Mar 2009
Location: Ottawa, ON, CAN
Posts: 398
Rep Power: 14
cnsidero is on a distinguished road
Ali,

The trick to getting this to work is to assemble the face manually. If your mesh, for example, had three domains (one for lower wing, one for upper wing and one for the wake) choose all three and select Create, Extrude, Normal ...

In the Assemble panel, check the Assemble Special box to expand the panel. Now click Delete All Faces to restart the face assembly process and uncheck Domain Once to allow selection of the wake domain more than once. Starting with the wake domain, pick the domains in successive order (e.g. wake, upper surface, lower surface, wake again).

When you select the wake domain again, you will be presented with the linkage assembly diagram and a list of domain-face linkages. You need to choose the linkage at the trailing edge of the airfoil. Click Use Selected Link, click Save Face and click Done. It's important to follow these steps exactly. If you did you should be in the usual Extrusion mode and the extrusion will march off of both sides of the wake domain (you should see arrows pointing from both sides).

Let me know if you get it working, Chris

Quote:
Originally Posted by Hybrid View Post
How can I extrude a wing having sharp trailing edge?

I have come to know that by adding wake domain, the select wing domains and wake domain and extrude all of them at once. By doing so, I get a mesh with wing domains (upper and lower) extruded up and down while wake domain extruded only in one direction.

How can I get the wake domains extruded on both sides?

If one can try, can assume any wing with sharp trailing edge.
Attached Images
File Type: jpg 3D-airfoil-wake-01.jpg (52.9 KB, 208 views)
onyir likes this.
cnsidero is offline   Reply With Quote

Old   February 26, 2013, 14:39
Default
  #6
Senior Member
 
Ali
Join Date: Jan 2012
Location: Pakistan
Posts: 135
Rep Power: 9
Hybrid is on a distinguished road
Thanks Chris Sideroff. It works really great.

Now a new problem, how to control the quality...extrusion fails after 3,4 steps. Sometime extruded root and tip airfoil connectors deshaped.

How can i get a nice mesh with minimum skewness and orthogonal?

How to check orthogonality in Pointwise?
__________________
Best Regards
Ali


Hybrid is offline   Reply With Quote

Old   February 26, 2013, 19:39
Default
  #7
Senior Member
 
Chris Sideroff
Join Date: Mar 2009
Location: Ottawa, ON, CAN
Posts: 398
Rep Power: 14
cnsidero is on a distinguished road
Ali,

By default, the normal extrusion only checks for negative Jacobians (aka cells with negative volume). There is additional cell quality checking you can monitor during the extrusion.

Back to your difficulty getting the extrusion to proceed. My first suggestion, and something I always watch for is mis-matched spacings in the domains of the starting face. Using my simple 3D airfoil as an example, I made sure the streamwise spacing at the trailing edge of the wing was the same as the streamwise spacing of the wake domain. This is particularly important when extruding from concave (i.e. internal) corners.

The next suggestion is to keep the initial normal spacing (init ds) about 4-6x smaller than the smallest transverse spacing in the domains of the starting face.

My last suggestion is to turn down the volume smoothing parameter to a small value (e.g. 0.05) until the extrusion proceeds away from the wall at least one BL thickness. Volume smoothing controls how much the algorithm smooths the extrusion front and I find, for tougher extrusions, the default value of 0.5 is too high. You can turn it back up once the extrusion gets away from the wall.

These are my general suggestions. If they still don't work, you're going to have to at least post a picture of your mesh so I can give more specific advice.

-Chris

Quote:
Originally Posted by Hybrid View Post
Thanks Chris Sideroff. It works really great.

Now a new problem, how to control the quality...extrusion fails after 3,4 steps. Sometime extruded root and tip airfoil connectors deshaped.

How can i get a nice mesh with minimum skewness and orthogonal?

How to check orthogonality in Pointwise?
cnsidero is offline   Reply With Quote

Old   March 3, 2013, 17:07
Default
  #8
Senior Member
 
Join Date: Nov 2010
Posts: 135
Rep Power: 8
taxalian is on a distinguished road
Send a message via Skype™ to taxalian
Hi Ali,
See my reply to the other message "[Pointwise] Boundary layer mesh" posted recently, hopefully that might also be helpful to you.

good luck.
taxalian is offline   Reply With Quote

Old   March 18, 2013, 17:17
Default
  #9
Senior Member
 
Ali
Join Date: Jan 2012
Location: Pakistan
Posts: 135
Rep Power: 9
Hybrid is on a distinguished road
Quote:
Originally Posted by cnsidero View Post
Ali,

The trick to getting this to work is to assemble the face manually. If your mesh, for example, had three domains (one for lower wing, one for upper wing and one for the wake) choose all three and select Create, Extrude, Normal ...

In the Assemble panel, check the Assemble Special box to expand the panel. Now click Delete All Faces to restart the face assembly process and uncheck Domain Once to allow selection of the wake domain more than once. Starting with the wake domain, pick the domains in successive order (e.g. wake, upper surface, lower surface, wake again).

When you select the wake domain again, you will be presented with the linkage assembly diagram and a list of domain-face linkages. You need to choose the linkage at the trailing edge of the airfoil. Click Use Selected Link, click Save Face and click Done. It's important to follow these steps exactly. If you did you should be in the usual Extrusion mode and the extrusion will march off of both sides of the wake domain (you should see arrows pointing from both sides).

Let me know if you get it working, Chris

Thanks All who reply the post.

It works but need more practice and experience on Extrusion especially Normal Extrusion.
__________________
Best Regards
Ali


Hybrid is offline   Reply With Quote

Old   February 8, 2017, 19:16
Default
  #10
New Member
 
Ahmed Hesham Saad Yassin
Join Date: Feb 2017
Location: Cairo, Egypt
Posts: 9
Rep Power: 2
AhmedSaad is on a distinguished road
Send a message via Skype™ to AhmedSaad
Hello Ali!
I wonder if you already figured out what causes mesh extrusion to fail and why leading and trailing nodes sometimes dislocated.
AhmedSaad is offline   Reply With Quote

Old   February 9, 2017, 09:23
Default
  #11
Senior Member
 
John Chawner
Join Date: Mar 2009
Location: Fort Worth, Texas, USA
Posts: 256
Rep Power: 11
jchawner is on a distinguished road
Ahmed, extrusion can fail for many reasons depending on which extrusion method you're using, the complexity of your geometry, the quality of the initial mesh, and more. I also don't know what you mean by dislocated.
__________________
John Chawner / jrc@pointwise.com / www.pointwise.com
Blog: http://blog.pointwise.com/
on Twitter: @jchawner
jchawner is offline   Reply With Quote

Old   February 9, 2017, 10:28
Default
  #12
New Member
 
Ahmed Hesham Saad Yassin
Join Date: Feb 2017
Location: Cairo, Egypt
Posts: 9
Rep Power: 2
AhmedSaad is on a distinguished road
Send a message via Skype™ to AhmedSaad
Thanks for your reply John,

I attached photos for my blade surface mesh and its minimum included Angle. When I do create, extrude, normal, I receive different errors in the first extruded step, sometimes due to some cells met the negative skew jacobian stop conditions or min positive skew jacobian.
I kept that 1st step mesh and attached photos if the created negative volumes at the root and tip of the blade. (Note: here the deformation happened in the trailing edge, In other cases I've tried, the deformation happened in the leading edge)

I tried the suggestions made by Chris and what I found in another thread but with no solution.
I think my problem within the tolerance but I am not sure. I had set an appropriate model size tolerance (as I believe) and re-import my geometry but it didn't work too.

I hope someone defines my problem.

Best Regards,
Ahmed
Attached Images
File Type: png Attributes panal.PNG (20.0 KB, 22 views)
File Type: png boundary conditions.png (10.5 KB, 26 views)
File Type: png error.PNG (19.4 KB, 23 views)
File Type: png Geometry.png (36.9 KB, 28 views)
File Type: png Min induced angle at root surface.png (79.2 KB, 27 views)
AhmedSaad is offline   Reply With Quote

Old   February 9, 2017, 10:32
Default
  #13
Senior Member
 
John Chawner
Join Date: Mar 2009
Location: Fort Worth, Texas, USA
Posts: 256
Rep Power: 11
jchawner is on a distinguished road
No images were attached. And I caution you against jumping immediately to the conclusion of tolerances.

Extrude, Normal and the fact that you're using Jacobian as a stop condition tells me you are extruding quads (if 2D) or hex (if 3D). If true, change the two Kinsey Barth smoothing factors to 3.

Look at your initial mesh and make sure that there are no large cell-to-cell jumps in cell size.
__________________
John Chawner / jrc@pointwise.com / www.pointwise.com
Blog: http://blog.pointwise.com/
on Twitter: @jchawner
jchawner is offline   Reply With Quote

Old   February 9, 2017, 10:33
Default
  #14
New Member
 
Ahmed Hesham Saad Yassin
Join Date: Feb 2017
Location: Cairo, Egypt
Posts: 9
Rep Power: 2
AhmedSaad is on a distinguished road
Send a message via Skype™ to AhmedSaad
attached the remaining photos
Attached Images
File Type: png Min induced angle at tip surface.png (76.6 KB, 26 views)
File Type: png root surface mesh.png (77.5 KB, 23 views)
File Type: png tip surface mesh.png (65.7 KB, 25 views)
File Type: png tip trailing edge after 1 step extrusion.png (39.9 KB, 23 views)
AhmedSaad is offline   Reply With Quote

Old   February 9, 2017, 10:34
Default
  #15
New Member
 
Ahmed Hesham Saad Yassin
Join Date: Feb 2017
Location: Cairo, Egypt
Posts: 9
Rep Power: 2
AhmedSaad is on a distinguished road
Send a message via Skype™ to AhmedSaad
Sorry I edited my reply and attached the photos
AhmedSaad is offline   Reply With Quote

Old   February 9, 2017, 10:38
Default
  #16
New Member
 
Ahmed Hesham Saad Yassin
Join Date: Feb 2017
Location: Cairo, Egypt
Posts: 9
Rep Power: 2
AhmedSaad is on a distinguished road
Send a message via Skype™ to AhmedSaad
I changed the two Kinsey Barth smoothing factors to 3, and still have the same problem.
AhmedSaad is offline   Reply With Quote

Old   February 9, 2017, 10:39
Default
  #17
Senior Member
 
John Chawner
Join Date: Mar 2009
Location: Fort Worth, Texas, USA
Posts: 256
Rep Power: 11
jchawner is on a distinguished road
A quick look at your images tells me the problem is at the sharp trailing edge where the extrusion direction is probably undefined (i.e. what is the normal direction to a singularity?) Because I think you're trying to extrude an O-grid topology around the airfoil.

It also looks to me as though the trailing edge is doubly defined (i.e. there are two connectors there). This means that topologically the airoil looks like an open sheet to the extruder. That initial domain must use only 1 connector at the trailing edge.

You might also consider putting a connector at the leading edge and using that as the seam instead of the trailing edge. That's a trick, but it might help.

And you might want to think about using a C-grid topology to "hide" the sharp trailing edge. Add a domain that attaches to the trailing edge and extends downstream. It will be part of your initial mesh along with the upper and lower surface of the airfoil.

Good luck.
__________________
John Chawner / jrc@pointwise.com / www.pointwise.com
Blog: http://blog.pointwise.com/
on Twitter: @jchawner
jchawner is offline   Reply With Quote

Old   February 9, 2017, 11:19
Default
  #18
New Member
 
Ahmed Hesham Saad Yassin
Join Date: Feb 2017
Location: Cairo, Egypt
Posts: 9
Rep Power: 2
AhmedSaad is on a distinguished road
Send a message via Skype™ to AhmedSaad
As I understood,
First, the O-grid topology around airfoil requires to define the normal extrusion direction of the trailing edge. How can I define it? and How it works in this video?:
https://www.youtube.com/watch?v=KOvWlgBEzo4&t=1174s
merge.jpg

Second, I checked the trailing edge connector and as it seems in the merge panel in the picture attached, it is not red.
and after I re-extrude in normal direction this time. I got a deformation in leading edge. And at the trailing edge after extrusion, It seems doubled as you mentioned.


Third, C-grid would be helpful but I don't know how to create a wake domain attached to the trailing edge (different pitch angles along the spanwise direction) Do you have any suggestions?
Attached Images
File Type: png root-leading edge.png (19.7 KB, 18 views)
File Type: png tip-leading edge.png (30.0 KB, 18 views)
File Type: png tip-trailing edge.png (72.1 KB, 16 views)
File Type: png root-trailing edge.png (33.6 KB, 15 views)
AhmedSaad is offline   Reply With Quote

Old   February 9, 2017, 11:31
Default
  #19
Senior Member
 
John Chawner
Join Date: Mar 2009
Location: Fort Worth, Texas, USA
Posts: 256
Rep Power: 11
jchawner is on a distinguished road
Your case probably requires the assistance of our Technical Support team. Email your Customer ID to them at support@pointwise.com and share this information you've showed me. Best Regards.
AhmedSaad likes this.
__________________
John Chawner / jrc@pointwise.com / www.pointwise.com
Blog: http://blog.pointwise.com/
on Twitter: @jchawner
jchawner is offline   Reply With Quote

Reply

Tags
sharp te, wing topolgy

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are On
Pingbacks are On
Refbacks are On


Similar Threads
Thread Thread Starter Forum Replies Last Post
Moving mesh Niklas Wikstrom (Wikstrom) OpenFOAM Running, Solving & CFD 122 June 15, 2014 06:20
how to set periodic boundary conditions Ganesh FLUENT 13 January 22, 2014 05:11
Trouble with sharp trailing edge Prakash.Paudel ANSYS Meshing & Geometry 4 August 12, 2011 14:01
fine mesh in wake trailing edge jose antonio ANSYS Meshing & Geometry 1 August 11, 2011 19:10
Star mesh import problem chris1980 OpenFOAM Other Meshers: ICEM, Star, Ansys, Pointwise, GridPro, Ansa, ... 20 May 8, 2006 01:07


All times are GMT -4. The time now is 02:41.