CFD Online Logo CFD Online URL
www.cfd-online.com
[Sponsors]
Home > Forums > Software User Forums > ANSYS > FLUENT

Someone please help!

Register Blogs Community New Posts Updated Threads Search

Reply
 
LinkBack Thread Tools Search this Thread Display Modes
Old   January 26, 2014, 06:37
Default Someone please help!
  #1
New Member
 
Join Date: Jan 2014
Posts: 3
Rep Power: 12
DinjiP is on a distinguished road
I am extremely frustrated. I have been at this for weeks with little to no success.

I am trying to create a 2D mesh to simulate air flow over the S809 airfoil. I used coordinates from the NREL website. I have mainly tried following this tutorial: https://confluence.cornell.edu/displ...oil+-+Geometry. However, I am running into two major problems.

1) First problem, when I try to create the vertical line for the projection that will separate the domain (Under Creating Quadrants in the tutorial), I receive a Boolean error. I have figured this comes from ANSYS not being able to read the geometry of the coordinates perfectly. Therefore, I use a spline to create the airfoil from imported points instead of importing the curve directly from the coordinates file. However, I end up with a less than perfect airfoil shape.

If anyone has a better solution or an actual fix to this error, that would be great.

2) Second problem. The S809 is not symmetrical like the NACA profiles. Therefore, when I create the horizontal line, for the quadrants it separates the bottom the airfoil into two sections. If you are having trouble understanding this look at this picture. The effects of the separation is very noticeable when I try to mesh.

All in all these problems either lead to errors that cause simulation to crash or bad results when trying to validate.

I am an undergraduate student in over my head, if you could please keep in mind with your answer that I have taken no CFD courses. If anyone has any tutorials similar to the cornell tutorials for S809 that would be great. Or even a link to a sample geometry and mesh of the S809 that would help me understand would be great as well. Honestly, any help at all, would be awesome.

Thank you
DinjiP is offline   Reply With Quote

Old   January 26, 2014, 11:22
Default
  #2
Super Moderator
 
flotus1's Avatar
 
Alex
Join Date: Jun 2012
Location: Germany
Posts: 3,427
Rep Power: 49
flotus1 has a spectacular aura aboutflotus1 has a spectacular aura about
Concerning number 2, you have several options.
You could perform the split operations in a different order, e.g. first do the vertical split at the trailing edge. This way you can choose only the body "behind" the airfoil for the horizontal split.

Or you could simply do another boolean operation (unite) with the small volume that is created when you do the horizontal split.

Or you could use a bottom-up approach instead, with a separate sketch for each of the parts.
flotus1 is offline   Reply With Quote

Old   January 26, 2014, 17:49
Default
  #3
New Member
 
Join Date: Jan 2014
Posts: 3
Rep Power: 12
DinjiP is on a distinguished road
Thank you for your response Flotus1

Problem number 1 stops me from completing a vertical split, unless I offset the vertical line by .0001 meters. Which is what I did.

Now that I have done, that, how do I create an effective mesh on the dome part of the mesh, that contains the airfoil? Seeing how the tutorial calls for the creation of 4 sections, instead of the 3 I have now when I take your advice.

As for your other suggestion, the bottom up approach, I am not sure what you mean by that.
DinjiP is offline   Reply With Quote

Old   January 27, 2014, 03:57
Default
  #4
Super Moderator
 
flotus1's Avatar
 
Alex
Join Date: Jun 2012
Location: Germany
Posts: 3,427
Rep Power: 49
flotus1 has a spectacular aura aboutflotus1 has a spectacular aura about
Quote:
Now that I have done, that, how do I create an effective mesh on the dome part of the mesh, that contains the airfoil? Seeing how the tutorial calls for the creation of 4 sections, instead of the 3 I have now when I take your advice.
Ok forget about this if you have trouble creating a similar topology like in the tutorial you are using.

Did you try to merge the small part instead?
unite.jpg

Quote:
As for your other suggestion, the bottom up approach, I am not sure what you mean by that.
The tutorial guides you to create one single geometry for the whole domain in the first place and then splits it into smaller blocks. This is usually referred to as a top-down approach.
The opposite way would be a bottom-up approach where you create a separate sketch for each of the four blocks you want to have in the end.
While creating the surfaces from these 4 sketches, make sure to use the "add frozen" option. Otherwise the geometry will be merged again.
frozen.png
When you are done, select the 4 surfaces together, right-click and choose "form new part". If you dont, you will get interfaces instead of a continuous mesh.
flotus1 is offline   Reply With Quote

Old   February 5, 2014, 16:34
Default
  #5
New Member
 
Join Date: Jan 2014
Posts: 3
Rep Power: 12
DinjiP is on a distinguished road
Thanks Flotus1, your suggestion about bottom up really did the trick! You are the man!
DinjiP is offline   Reply With Quote

Old   May 24, 2016, 14:34
Default
  #6
New Member
 
J SB
Join Date: Aug 2015
Posts: 6
Rep Power: 11
fluenter2015 is on a distinguished road
Quote:
Originally Posted by flotus1 View Post
Ok forget about this if you have trouble creating a similar topology like in the tutorial you are using.

Did you try to merge the small part instead?
Attachment 28281



The tutorial guides you to create one single geometry for the whole domain in the first place and then splits it into smaller blocks. This is usually referred to as a top-down approach.
The opposite way would be a bottom-up approach where you create a separate sketch for each of the four blocks you want to have in the end.
While creating the surfaces from these 4 sketches, make sure to use the "add frozen" option. Otherwise the geometry will be merged again.
Attachment 28282
When you are done, select the 4 surfaces together, right-click and choose "form new part". If you dont, you will get interfaces instead of a continuous mesh.
Hi. I am working on a similar problem. I applied your second suggestion (the bottom-up approach) and it worked. However, when I define boundary conditions, the edges that contact the 4 surfaces (you can see them in the end link that I attach) are defined as "wall" and I canīt change them to "interior". Is there any solution?

I really appreciate any help you can provide (you or anyone)

Last edited by fluenter2015; May 27, 2016 at 12:18.
fluenter2015 is offline   Reply With Quote

Reply


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 Off
Pingbacks are On
Refbacks are On



All times are GMT -4. The time now is 16:48.