|
[Sponsors] |
August 27, 2013, 08:44 |
Delnauy technique "Bottom-up" problem
|
#1 |
Senior Member
mohamed khedr emam
Join Date: Oct 2011
Location: Egypt
Posts: 121
Rep Power: 15 |
Hi all, I spent long long time to build a good mesh of my wind turbine blade ( and its enclosure) using octree tetra technique "top down" and after I sent the mesh to Fluent I discovered that the leading edge of my blade need to has dense surface mesh so I restart and replace my ICEM technique and follow the famous Workshop "Wingbody" that depend on Delnauy technique "Bottom-up"... The "Wingbody" workshop started by setup curve sizes on leading edge , then start shell mesh, build the prism and finally fill the volume. I followed and applied carefully the same steps of the "Wingbody" workshop on my model but i found this: 1- incomplete prism on the blade surface in spite of max. prism angle was set to 180 2-if I continue to build the volume using Delnauy technique an error appear called "tri2tet exited" I attached my model at (http://www.4shared.com/rar/eT4cdfOx/modelX.html) Any help and thanks in advance
|
|
August 29, 2013, 08:51 |
|
#2 |
Senior Member
mohamed khedr emam
Join Date: Oct 2011
Location: Egypt
Posts: 121
Rep Power: 15 |
I also followed the same steps of WingBody tutorial that created by @Far http://www.cfd-online.com/Forums/ans...tra-prism.html
but the prism looks like this (pic) Any help pls. |
|
September 1, 2013, 16:57 |
|
#3 |
Senior Member
Simon Pereira
Join Date: Mar 2009
Location: Ann Arbor, MI
Posts: 2,663
Blog Entries: 1
Rep Power: 47 |
My guess is that the element quality of folding the prisms around that curve was just too low. You could lower the min quality to get more prisms, but...
When I have a sharp trailing edge, I typically create a trailing edge surface... Basically just use the surface creation tools to extrude the trailing edge surface backward for some distance... Or copy the trailing edge curves some distance backward and create a surface between the curves. Make sure the trailing edge is in a part like "TEMP". Make sure you checkbox the TEMP part for both Prism growth and Internal Wall. When you are done generating the mesh, you can either delete the "TEMP" shells from the mesh or mark them as "Internal Wall" in your solver.
__________________
----------------------------------------- Please help guide development at ANSYS by filling in these surveys Public ANSYS ICEM CFD Users Survey This second one is more general (Gambit, TGrid and ANSYS Meshing users welcome)... CFD Online Users Survey |
|
September 2, 2013, 17:09 |
|
#4 | ||
Senior Member
mohamed khedr emam
Join Date: Oct 2011
Location: Egypt
Posts: 121
Rep Power: 15 |
Thanks Simon for reply , I appreciated that
Quote:
Quote:
2- could i ask you to share a link for tutorial or video for extruded the trailing edge (i can't understand this steps) tanks in advance |
|||
September 2, 2013, 22:18 |
|
#5 | |
Senior Member
Bruno
Join Date: Mar 2009
Location: Brazil
Posts: 277
Rep Power: 21 |
It'll do both.
Quote:
When creating the mesh, set TEMP to be a internal wall and activate prisms on this wall. You can do that in the prism setup window ('Mesh > Part Mesh Setup'). Re-generate your meshing as you normally do. Cheers. |
||
September 2, 2013, 22:22 |
|
#6 |
Senior Member
Simon Pereira
Join Date: Mar 2009
Location: Ann Arbor, MI
Posts: 2,663
Blog Entries: 1
Rep Power: 47 |
1) I think the actual minimum that it will accept is 1e-6 for quality.
2) The trailing edge surface will improve the prism quality which will then allow prism to grow as you expect 3) I have posted threads on this topic before. I just went to look for the old pics I took, but couldn't find them...
__________________
----------------------------------------- Please help guide development at ANSYS by filling in these surveys Public ANSYS ICEM CFD Users Survey This second one is more general (Gambit, TGrid and ANSYS Meshing users welcome)... CFD Online Users Survey |
|
September 4, 2013, 06:00 |
|
#7 |
Senior Member
mohamed khedr emam
Join Date: Oct 2011
Location: Egypt
Posts: 121
Rep Power: 15 |
o.k i did the extruded trialing edge and followed the mentioned steps but the prism fault is the same
i uploaded the new file http://www.4shared.com/rar/bf0GN7Aa/model.html and here is a sample image of prism |
|
September 4, 2013, 07:24 |
|
#8 |
Senior Member
mohamed khedr emam
Join Date: Oct 2011
Location: Egypt
Posts: 121
Rep Power: 15 |
also I tried to increase the shell meshing elements (increase blade curves nodes) but also the prism fault occurred (the total elements reach to 3 million without volume mesh!)
|
|
September 4, 2013, 14:23 |
|
#9 |
Senior Member
Simon Pereira
Join Date: Mar 2009
Location: Ann Arbor, MI
Posts: 2,663
Blog Entries: 1
Rep Power: 47 |
The trailing edge surface should be longer (at least the element size times the number of layers) and did you make sure to set the "internal wall" and "prism" check boxes for the new surface? It needs to have triangles on it and prism needs to grow from those triangles.
__________________
----------------------------------------- Please help guide development at ANSYS by filling in these surveys Public ANSYS ICEM CFD Users Survey This second one is more general (Gambit, TGrid and ANSYS Meshing users welcome)... CFD Online Users Survey |
|
September 4, 2013, 14:53 |
|
#10 | |
Senior Member
mohamed khedr emam
Join Date: Oct 2011
Location: Egypt
Posts: 121
Rep Power: 15 |
thanks simon for reply
Quote:
yes , i upload an image for Part Mesh setup |
||
September 4, 2013, 15:06 |
|
#11 |
Senior Member
Simon Pereira
Join Date: Mar 2009
Location: Ann Arbor, MI
Posts: 2,663
Blog Entries: 1
Rep Power: 47 |
Actually, I guess you were using patch conforming mesh right? THe internal wall thing is just to prevent octree from deleting the mesh so there will be mesh for prism to grow from... If you are using another method, the key thing is to make sure you have a surface mesh in the TEMP part.
As for how to make the trailing edge surface longer, just do what you did before, but longer. If you did it with a copy transform of the curve, just copy/transform it further away. Basically, what is happening is the mesh is stair stepping away from the bad area... You want to create a trailing edge surface so that the stair stepping is happening on the trailing edge surface instead of the wing.
__________________
----------------------------------------- Please help guide development at ANSYS by filling in these surveys Public ANSYS ICEM CFD Users Survey This second one is more general (Gambit, TGrid and ANSYS Meshing users welcome)... CFD Online Users Survey |
|
September 6, 2013, 17:33 |
|
#12 |
Senior Member
mohamed khedr emam
Join Date: Oct 2011
Location: Egypt
Posts: 121
Rep Power: 15 |
Hi Simon
yes I did this (Geometry>> sweep surface), when i made the egde surface longer , the prism is done inside the blade . If i make the surface longer than the previous an error massage appear and the prism building stop. |
|
September 8, 2013, 09:34 |
prism setting...
|
#13 |
Senior Member
mohamed khedr emam
Join Date: Oct 2011
Location: Egypt
Posts: 121
Rep Power: 15 |
This is my prism setting (image) ....
Any suggestion? |
|
Tags |
wind turbine blade |
Thread Tools | Search this Thread |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
natural convection problem for a CHT problem | Se-Hee | CFX | 2 | June 10, 2007 07:29 |
Adiabatic and Rotating wall (Convection problem) | ParodDav | CFX | 5 | April 29, 2007 20:13 |
Periodic flow boundary condition problem | sudha | FLUENT | 3 | April 28, 2004 09:40 |
extremely simple problem... can you solve it properly? | Mikhail | Main CFD Forum | 40 | September 9, 1999 10:11 |
Is this problem well posed? | Thomas P. Abraham | Main CFD Forum | 5 | September 8, 1999 15:52 |