|
[Sponsors] |
August 6, 2014, 18:02 |
Hexa-Meshing of sharp edge
|
#1 |
Member
Robert Bischof
Join Date: Aug 2014
Posts: 56
Rep Power: 11 |
Hello together,
for my thesis i have to work with icem cfd and cfx. Therefore i'm a beginner in using both programms. I've finished the blocking in icem (after a while) and refined the mesh at various points. But cfx stops the simulation after 4 iteration (Error: "... c_fpx_handler: floating point exception: overflow" --> i thing it is a numerical problem, caused by the mesh quality) I investigated the problem and found a region, which have poor elements. I found the possibility to construct y-blocks, but still there are elements with small angles (13° - 20°). Therefor I have found no solution. Is there another possibility to mesh this region or to refine the mesh? I have to use hexa-elements and i'm not allowed to change the geometry. There are screenshots of the region and the mesh in the attachment. For your help, i would be very greatful. Thanks a lot! |
|
August 8, 2014, 08:00 |
|
#2 |
Senior Member
Sebastian Engel
Join Date: Jun 2011
Location: Germany
Posts: 566
Rep Power: 20 |
Hi Robert, is there are reason you need to mesh this small region with additional blocks? Is it possible to associate the face of the block on top to the cylinder surface? This block would be abit distorted, but generates better cells.
associate.jpg Last edited by bluebase; August 9, 2014 at 10:58. |
|
August 8, 2014, 16:48 |
|
#3 |
Member
Robert Bischof
Join Date: Aug 2014
Posts: 56
Rep Power: 11 |
Thanks for your answer.
Unfortunately, I can't access my ICEM Modell this evening. The blocking must be chosen based on the geometry. At Monday I will show some Screenshots for better understanding. Has anyone else some suggestions ? |
|
August 19, 2014, 05:38 |
|
#4 |
Member
Robert Bischof
Join Date: Aug 2014
Posts: 56
Rep Power: 11 |
For better understanding, I have added some screenshots of my inlet.
On the second picture the visibility of the front surface (of the box) is turned of, so you can the inside the pipe. There is a rectangular box, which is leading on the pipe. Because of that, there is that region, which belongs to my first post. Like I said, I'm not allowed to change the geometry. Has anybody an idea for my problem ? Greetings |
|
August 19, 2014, 15:25 |
|
#5 |
Super Moderator
|
||
August 19, 2014, 16:56 |
|
#6 |
Member
Robert Bischof
Join Date: Aug 2014
Posts: 56
Rep Power: 11 |
Today I have tried to start a job without the pipe and the bad regions (I deleted the section and started it in CFX) and the job is running. When I will start the job with the pipe and the bad regions, CFX stops with the error i wrote in the first post. So I think, that I have to change the blocking there!?
I don't know how to merge the mesh with tri elements, can someone explain? I'm sorry for that question, but thats new for me! Thank you! |
|
August 19, 2014, 17:17 |
|
#8 |
Member
Robert Bischof
Join Date: Aug 2014
Posts: 56
Rep Power: 11 |
Ok, thank you. I will look for it tomorrow
|
|
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
Trouble with sharp trailing edge | Prakash.Paudel | ANSYS Meshing & Geometry | 4 | August 12, 2011 14:01 |
[ICEM] Hole near sharp trailing edge of airplane/wing geometry | jlichtwa | ANSYS Meshing & Geometry | 2 | September 21, 2010 15:19 |
Actuator disk model | audrich | FLUENT | 0 | September 21, 2009 07:06 |
Where's the singularity/mesh flaw? | audrich | FLUENT | 3 | August 4, 2009 01:07 |
Boundary layer meshing with sharp trailing edges | Andrew Berner | FLUENT | 1 | August 28, 2007 17:39 |