|
[Sponsors] |
June 11, 2012, 11:49 |
Problem with merging edges/faces
|
#1 |
Member
Bahar
Join Date: May 2012
Posts: 46
Rep Power: 13 |
Hello everybody,
I have a problem while importing geometry from Solid Works to Gambit. As showed in figures, there are 3 blades that would be meshed. But every side edges of blades has broken in two edges while there are circles in leading edge and trailing edges. So when I mesh the airfoil it wouldn't get a smooth mesh because of difference in scale of edges. I tried to merge them but while the geometry is complex, it could not create a real face/edge. Is there any idea to resolve this problem? Thank you in advance, bahar |
|
June 12, 2012, 01:36 |
|
#2 |
Super Moderator
Maxime Perelli
Join Date: Mar 2009
Location: Switzerland
Posts: 3,297
Rep Power: 41 |
can you provide the dbs file?
__________________
In memory of my friend Hervé: CFD engineer & freerider |
|
June 12, 2012, 02:44 |
|
#3 |
Super Moderator
Maxime Perelli
Join Date: Mar 2009
Location: Switzerland
Posts: 3,297
Rep Power: 41 |
Ok I checked your dbs file, and your problem, can be resolved by using the cleaning tool (use clean short edges: first on the left) it will collapse the trailing edge.
But you have other problems: you need to connect faces 27 and 47 and you also have problem with surfaces 40 and 28: For that delete volume up, and reconstruct it with remaining surfaces (sticht faces)
__________________
In memory of my friend Hervé: CFD engineer & freerider |
|
June 12, 2012, 03:01 |
Re:problem with importing geometry in Gambit
|
#4 |
Member
Bahar
Join Date: May 2012
Posts: 46
Rep Power: 13 |
Wow! Thank you so much for all this helpful points
If I understand correctly, you mean by deleting short edges I can merging the faces. Isn't it? And about connecting faces 47 & 27, do you mean defining an interface between them in Zone section(green cube)? And problem with surfaces 40 and 28, the cylindrical faces should be moving but the frustums are stationary. I thought they would be separate volumes, what's wrong? Regards Bahar |
|
June 12, 2012, 03:08 |
|
#5 |
Super Moderator
Maxime Perelli
Join Date: Mar 2009
Location: Switzerland
Posts: 3,297
Rep Power: 41 |
make first a good mesh for steady. Then adapt your mesh for moving mesh.
-For deleting short edges, use cleaning tool: click on update (short edge), and click on auto. Gambit will automatically collapse trailing edges -For faces 47 and 27, just connect them (forced virtual) -For 40 and 28, as I said below, generate a mesh for steady state first (so delete the volume, and reconstruct it) Once your mesh is ok, compute, check if flowfield is ok, then work on the motion
__________________
In memory of my friend Hervé: CFD engineer & freerider |
|
June 12, 2012, 03:16 |
Re:problem with importing geometry in Gambit
|
#6 |
Member
Bahar
Join Date: May 2012
Posts: 46
Rep Power: 13 |
Thank you dear mAx!
I'll do all this and will announce you about the result. But there is a problem, I'm using Gambit version 2.3, I can not find here what you said about cleaning tool unfortunately. Regards, Bahar |
|
June 12, 2012, 03:26 |
|
#7 |
Super Moderator
Maxime Perelli
Join Date: Mar 2009
Location: Switzerland
Posts: 3,297
Rep Power: 41 |
I checked and Cleaning Tool was already available at release 2.1.
You can find it in Tools/Geometry Clean-Up http://cdlab2.fluid.tuwien.ac.at/LEH...ide/mg0504.htm
__________________
In memory of my friend Hervé: CFD engineer & freerider |
|
June 12, 2012, 03:33 |
Re:problem with importing geometry in Gambit
|
#8 |
Member
Bahar
Join Date: May 2012
Posts: 46
Rep Power: 13 |
Oh,yes! I just found it
I should use the first item (clean up short edge), right? Thank you very much dear mAx! Regards, Bahar |
|
June 12, 2012, 03:34 |
|
#9 |
Super Moderator
Maxime Perelli
Join Date: Mar 2009
Location: Switzerland
Posts: 3,297
Rep Power: 41 |
yes
update>auto
__________________
In memory of my friend Hervé: CFD engineer & freerider |
|
June 12, 2012, 03:36 |
Re:problem with importing geometry in Gambit
|
#10 |
Member
Bahar
Join Date: May 2012
Posts: 46
Rep Power: 13 |
OK, thank you in advance!
Regards, Bahar |
|
June 12, 2012, 03:51 |
Re:problem with importing geometry in Gambit
|
#11 |
Member
Bahar
Join Date: May 2012
Posts: 46
Rep Power: 13 |
Dear mAx,
I do it as you said, but after cleaning up I lose 3 primary edges of airfoils(button airfoils)! What's wrong? Regards Bahar |
|
June 12, 2012, 04:09 |
Re:problem with importing geometry in Gambit
|
#12 |
Member
Bahar
Join Date: May 2012
Posts: 46
Rep Power: 13 |
I did it on a new file and it worked correctly!
Thanks for your help mAx Regards, Bahar |
|
June 12, 2012, 09:14 |
Re:problem with importing geometry in Gambit
|
#13 |
Member
Bahar
Join Date: May 2012
Posts: 46
Rep Power: 13 |
Hi dear mAx,
I tried it out and there is a similar problem that Gambit couldn't convert virtual volume/face/edge of blades to real. What should I do? Regards, Bahar |
|
June 13, 2012, 01:03 |
|
#14 |
Super Moderator
Maxime Perelli
Join Date: Mar 2009
Location: Switzerland
Posts: 3,297
Rep Power: 41 |
you should use the cleaning tool, only when your topology decomposition is ready (I mean jjsut befor you start to mesh your model)
Indeed the cleaning tool, will give you finally a virtual entity and it is then no more possible to get a real entity (what you experienced) Why do you want real entities?
__________________
In memory of my friend Hervé: CFD engineer & freerider |
|
June 13, 2012, 01:19 |
Re
|
#15 |
Member
Bahar
Join Date: May 2012
Posts: 46
Rep Power: 13 |
Hi dear mAx,
I should subtract volumes to get appropriate domain as you saw. It it possible subtracting a virtual volume from a real one? And may it cause any problem in Fluent (virtual blades)? Regards, Bahar |
|
June 13, 2012, 01:21 |
|
#16 |
Super Moderator
Maxime Perelli
Join Date: Mar 2009
Location: Switzerland
Posts: 3,297
Rep Power: 41 |
no
that's why you need to keep real entities as late as you can, else you cannot handle boolean operations anymore (they are only available for real topologies)
__________________
In memory of my friend Hervé: CFD engineer & freerider |
|
June 13, 2012, 01:30 |
Re
|
#17 |
Member
Bahar
Join Date: May 2012
Posts: 46
Rep Power: 13 |
OK, Thank you in advance!
I'll do that. Regards, Bahar |
|
June 13, 2012, 04:04 |
Re
|
#18 |
Member
Bahar
Join Date: May 2012
Posts: 46
Rep Power: 13 |
Hi dear mAx,
During meshing a discontinuity occurs as showed in figure. It seems that it is as a result of existence of two virtual and aslo two real edge in a face, isn't? What should i do? Regards, Bahar |
|
June 13, 2012, 04:38 |
|
#19 |
Super Moderator
Maxime Perelli
Join Date: Mar 2009
Location: Switzerland
Posts: 3,297
Rep Power: 41 |
in the display options, make all volumes invisible, and display only blades.
Enable shaded modus, and check if the surfaces is ok (where you get degenerated cells)
__________________
In memory of my friend Hervé: CFD engineer & freerider |
|
June 14, 2012, 02:44 |
Re
|
#20 |
Member
Bahar
Join Date: May 2012
Posts: 46
Rep Power: 13 |
Hi dear mAx,
I checked the blades as you said and found out a problem in my Solid Works model. Now, I'm going to improve it and then try it again in Gambit. Thank you in advance, Regards, Bahar |
|
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
conduction problem | venkataramana | OpenFOAM | 3 | December 1, 2013 07:30 |
UDF compiling problem | Wouter | Fluent UDF and Scheme Programming | 6 | June 6, 2012 04:43 |
ICEM mesh merging problem | Serge | CFX | 2 | October 22, 2007 06:13 |
natural convection problem for a CHT problem | Se-Hee | CFX | 2 | June 10, 2007 06:29 |
Adiabatic and Rotating wall (Convection problem) | ParodDav | CFX | 5 | April 29, 2007 19:13 |