Files are here ...

step 5 : delete symmetry surface

You Tube Video
 

This video shows the procedure of mirroring the geometry and blocking. Also shown the subsequent steps to give the finishing touches.

http://www.youtube.com/watch?v=HwvJJ...ature=youtu.be

creation of first half blocking

http://www.youtube.com/watch?v=9OzlS...ature=youtu.be

Wow, great thread, very useful.

Just one comment: IMHO, the numerous 'finishing touches' after mirroring should not exist. The finishing work is very annoying, a mirror action should be a mirror action, no touches to add on the mirrored side. But I guess that Icem CFD is working on that...

Thanks a lot,
François

I totally agree that mirror should be mirror. Simon once told that they are working on this aspect and may be we will see a major improvement in R 14.5

hi there, i think my problem fits quite well in this thread... i've been trying to mirror a given pipe-mesh... i tried "both methods", i.e. just the mesh as mentioned further above in the thread and according to the steps described last (mirror geom, mirror block, etc)

i dont seem to be succeeding, i'm pretty helpless at this point,... i'm not a pro in icem, so maybe there's something i've missed, any help is greatly appreciated

main problems after step-by-step-procedure:
- symmetry plane seems to be un-deletable
- mesh has uncovered faces and single edges

-> screenshots of every step attached... if something is unclear, please ask, i need solve this mystery ;)

download via external platform because total file size is too big, hope i'm not infringing forum-rules, if there's a better/more suitable possibility to share large screenshots, please let me know:
https://rapidshare.com/files/1051001910/screenshots.zip

thx for every help

It is very much simple case. You should not have the problems at all.

i know i shouldn't have these problems, that's why i posted here :|
i cannot explain/understand why it doesn't work, have you seen the screenshots?

i even do everything as in the "youtube-tutorial"... MAYBE i'm doing everything right, but icem creates those "flaws"... but how should i know if that is the case

...again, any help is appreciated ;)

I also tried to reproduce your geometry (though very simple) and got the problem after mirroring the geometry and blocking. Dont know why, but trying to come up with solution.

Every time I repeat this experiment I get the same error, that is mesh is not projected on some surface. That is face is not associated with surface, which is default function in ICEM CFD. I guess O-grid (when ogrid is on touching symmetry plane) combined with mirror option get you into trouble.

The only remedy so far I have found is to do all the things manually from edge to face association. Which is very much less efficient and it would have been more economical to do the full model from Ab initio

Ca we call this a bug?

hi Far,
thanks for all the effort, it is somewhat satisfying that at least somebody else experienced this behaviour, although it is not satisfying that there is no solution yet (at least no "good" one).

as for if we can call this a bug... well, in my un-experienced opinion, i'd say i wished that a mirror function was able to mirror a half-pipe/ half-pipe-geometries. i guess your question was directed at the "public", maybe some "ICEM-people" will come across this thread. since the solution to this specific problem is not my #1 priority (+ i don't know how to solve this), i will leave it - for the time being -as is. as soon as i have new info on this subject, i will make sure it is appended here. (of course i will follow this thread though)

in the meantime... if somebody else has something to add - please do so ;) thx

YOU MADE ME A VIDEO!!! THIS IS AWESOME! :D:D:D

Thanks a lot for this guidance. I ve been on holiday therefore I could´nt appreciate your posts earlier!

Camoesas