CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   ANSYS Meshing & Geometry (https://www.cfd-online.com/Forums/ansys-meshing/)
-   -   [ICEM] New Feature in ICEM is required (https://www.cfd-online.com/Forums/ansys-meshing/110612-new-feature-icem-required.html)

Far December 16, 2012 01:24

New Feature in ICEM is required
 
Dear All

After working on many projects on the www.cfd-online.com using ICEM CFD. I have feeling that there should be some enhancements in the functions of ICEM CFD both in terms of blocking and geometry. But the most important feature which I think must be there is

" We can move the vertices in o-grid along it perimeter without distrubing the overall o-grid shape". At the moment vertex alignment tools do not provide this functionality.

So what you guys think, this function shall be good addition to ICEM? if so please like this post so that Simon can take notice of this ;)

PSYMN December 17, 2012 09:16

This poll option in CFD-Online is interesting. I will need to try it out myself one of these days.

For this request, please be more specific.

Do you mean that when you move an Ogrid perimeter vertex you would like an option to "maintain orthogonality"?

How would that be different than moving the vertex and then using "Blocking (tab) => Edit Block => Modify Ogrid => Reset Ogrid Orthogonality" for that edge?

Perhaps that option is what you are looking for.

Far December 17, 2012 09:28

Ok I try to to be more specific.

Take an example of simple circle. Let create block and associate edge to perimeter. Now we make the o-grid. Now we have four green edges and four black or blue edges.

As moving the green edges are concerned, they are already following the curves to whom they are associated so whenever we move the vertices they will slide along that curve. The problem comes when we want to move the vertices on black or blue edges. I wish there should be option which can restricts the motion of vertices to the main green edges. So if shape is the ellipse they will move along ellipse and if the shape is circle their motion will be circular without disturbing the length of diagonal edges (diagonal edges connects the inner and outer o-grid edges).

Moreover there be option to move parallel vertices (along length) in same fashion (extended capability)

PSYMN December 17, 2012 10:15

OK, I get it. I will check back on this Survey in January and then talk to development about this request.

In the mean time, I often use the options to Fix the motion into one plane (such as "[] Fix Z"), then you can move the unprojected internal vertices more easily.

Best regards,

Simon

diamondx December 17, 2012 11:08

@Simon

Since FAR has started this poll option... i found this interesting... when working on ICEM CFD, i sometimes come to think about some few features. some of them would have been very handy for me and i just wish they can exist. Think of them as reviews :p
i hope you don't mind getting a bunch of them, and that you will find them interesting and doable...

PSYMN December 17, 2012 12:19

@everyone...

Sure, we welcome your defect reports and feature requests and will try to implement what I can (given resource constraints). Collecting this sort of feedback is part of my job at ANSYS. We have already responded to many of defects and requests that came in thru CFD-Online.

However, please keep in mind that requests that come from a real person with a name and company associated will be given more weight than an anonymous request thru CFD-Online. For instance, I will submit Far's request with his real name and company and then toss in the survey results as supporting evidence.

Simon

mjgraf December 17, 2012 13:24

people around me know I am working with o-grids when i start uttering @#$% ogrids!!!!

LOL
:D

Quote:

Originally Posted by Far (Post 397735)
Dear All

After working on many projects on the www.cfd-online.com using ICEM CFD. I have feeling that there should be some enhancements in the functions of ICEM CFD both in terms of blocking and geometry. But the most important feature which I think must be there is

" We can move the vertices in o-grid along it perimeter without distrubing the overall o-grid shape". At the moment vertex alignment tools do not provide this functionality.

So what you guys think, this function shall be good addition to ICEM? if so please like this post so that Simon can take notice of this ;)


PSYMN December 17, 2012 13:43

@mjgraf, imagine what you would be uttering without them! :eek:

energy382 December 23, 2012 09:44

You ansys guys should put priority on smoother, as it's still a load of crap.

I'm experienced with icem hexa. But I never had a mesh, that wasn't worth after smoothing!!

Far December 23, 2012 10:12

Let me clear one thing. Simon is here to help us in private capacity and keeping in view his helpful attitude we are requesting him for these features in ICEM.

energy382 December 23, 2012 12:11

it's a poll. so i tell him what the main problems in this software are (in my point of view). that's not a complaint, this is a honest feedback.

so what's the problem?? no need to clarify anything!

mind your own business

Far December 23, 2012 12:45

Obviously the purpose of this poll is to express ideas that what feature you wish should be there in the ICEM CFD.

So Smoothing is also one of the those features and Simon told me that they are working on it (one of the top priority at ANSYS).


Everybody is welcome to express his/her ideas about features they wished should be there while working with ICEM :)

PSYMN December 23, 2012 16:35

Quote:

Originally Posted by energy382 (Post 398794)

so what's the problem??

No problem, we really appreciate the feedback. :cool:

macfly December 26, 2012 14:10

A practical feature would be an automatic 'Close Holes' option when we sweep a surface. Because during a surface sweep, it's hard to predict what will be the the new curves names, making it difficult to write a script that sweeps a surface and then close holes.

Happy holidays :D


All times are GMT -4. The time now is 07:39.