CFD Online Discussion Forums

CFD Online Discussion Forums (
-   ANSYS Meshing & Geometry (
-   -   [ICEM] Wing Woes. Difficult geometry,(or inept user) Swept and tapered 3d wing (

ShowponyStuart May 10, 2013 00:16

Wing Woes. Difficult geometry,(or inept user) Swept and tapered 3d wing
Hey all,

I have been trying to mesh a 3D wing in ICEM using diamonddx's youtube vid as a guide

I must not have wrapped my head around ICEM as much as I thought because I am still having difficulty getting it to mesh at all (let alone actually get a good quality mesh)

Here is the general wing layout

This pic is the side view (looking from the outside). You can see something weird seems to be happening there.

The blocking strategy around the wing

Im not quite sure if im on the right track, but need to fix a couple of things or if I should scrap it and start from scratch.
Does anyone have any thoughts?

I have attached my files below for anyone to have a look at if they are interested.
Geometry files

And some more pictures too

Latest files

ShowponyStuart May 10, 2013 00:44

I should note that the wingtip is modeled as a *small aerofoil section (see pic). It doesn't taper down all the way. I did this for 2 reasons, the first being I wasn't entirely sure how to get it to taper cleanly in Proe and secondly, I thought having the edges would help with letting me associate stuff to the tip.

Also note, the mesh is very rough as I have been just trying to get it to work. I had planned on refining later on.

ShowponyStuart May 10, 2013 02:25

3 Attachment(s)
Alright I got mesh to behave a lot nicer. But I still feel that it is of questionable quality(even when I refine it). So Im not sure if anyone has any thoughts on my mesh.

For anyone else looking into this, all I did was follow the method in the youtube vid all the way to the end. Then I did the following to get it to start playing nice.
1. Deleted all blocks that were sitting inside the wing (pic1) should be 11 blocks to delete
2. Create iso curves (pic2)
3. Associated the corresponding curves to edges (pic3)
As simple as that. (I hate when I problem is simple but it still takes you ages to figure out lol)

FJSJ May 10, 2013 02:25

1 Attachment(s)
Hi Stuart. Sometimes I have the same error. Look at the picture. This time diamondx told me I should prove "Associate Face to Surface" tool. I say you this because Ive seen your second pic. Try to do that in this area.

FJSJ May 10, 2013 02:29

You got it :)

ShowponyStuart May 10, 2013 06:27

Does anyone know how I can make the mesh around the leading edge more square so it all flows together nicely? Or is it always going to do that because of the way its blocked?

I am trying to keep the mesh smooth and keep the elements orientated correctly with respect to the flow but Im starting to thing that this blocking strategy was the wrong approach. How should I go about tidying up that blocking? Any tips or hints.

Latest files too.

stuart23 May 10, 2013 09:47

Stuart (2),

Great work so far, although I am slightly worried what happens at your wingtip.

My advice (contrary to diamonds excellent tutorial. Sorry mate!) would be to wrap the O-Grid around the wing tip rather than propagating it to the extent of the domain.

Good Luck,


ShowponyStuart May 10, 2013 10:20

Hey Stu...well I guess you are Stu(1) seeing as you got in first :p

I absolutely agree with you, ive been staring at all those blocks for ages trying to figure out what to do with them.

So far I have made it to this stage

Im not sure the best way to show you the tip, but investigating kinda looks acceptable. But it wouldnt be a surprise in the slightest if it is completely rubbish and I just havent found a way to look at it.

Ive been staring at this block mayhem for hours now trying to work out something better. Unfortunatly this is only the second thing I have tried meshing in ICEM so far, and I still havent figured out how the blocking actually works yet haha.

If you have any ideas on how to salvage that...or if neccecary I could start from scratch (less preferred, but I want to do it right to be honest.) Here are my latest files anyways if you want a look. Be nice though, im still very...uhh special, when it comes to ICEM. haha

Far May 10, 2013 13:24

Stuart23: I agree with your idea about o-grid around wing instead of taking it into the farfield. That approach is well suited to flat tips with almost same order of chord length at tip. :D

Stuart: I am afraid, you will have to go back and make the new geometry without tip airfoil. sorry for home work ;) :p

You have sharp TE, collapse blocks behind TE.

PS: Now we have two Stuarts, hoping Stuart junior shall be as good as Stuart Senior. :)

I will take a look on files ...

ShowponyStuart May 10, 2013 13:59

I'll revisit my model see if I cant tidy that edge up, but if I do end up doing that way, is there a guide or something I can use? Im still pretty useless atm ( I may not have the skills of Stu Senior but I have perseverance. I dont like to be beaten by a program :p :D I would do well in the Tron universe. hehe)

But as it stands I wouldn't know where to start with that approach..I'll do a bit of googling though and see what I can come up with. Hopefully I find something useful. I'm running short on time for this one though, so if I cant get it solved soon (*verry soon) I will have to abandon it for now otherwise the rest of the project will go down the drain as well.

If I do figure it out, ill make up a guide for full on beginners seeing as wings seem to be a pretty popular....On that note, I have come across little guides that people done here and there. Perhaps the best of these should become stickies (or put in their own section) so people can find them easily. Just a thought.

Far May 10, 2013 14:28


If I do figure it out, ill make up a guide for full on beginners seeing as wings seem to be a pretty popular....On that note, I have come across little guides that people done here and there. Perhaps the best of these should become stickies (or put in their own section) so people can find them easily. Just a thought.
Can you show those threads or some other links which can be combined together?

ShowponyStuart May 11, 2013 01:52

Ill keep tabs on them. At the moment Im flat out getting trying to get this done, but soon as I get a breather I will try to find and collate some of the stuff (that way I can do it properly).

For now though, I cant figure out how to wrap an Ogrid around the tip like so the blocks dont propagating through the whole domain. I kind of dont get how you could actually do that, as how do all the small complicated blocks around the wing connect to a big simple block sitting next to it? (unless of course I am misunderstanding what you mean...which is quite likely haha)

I dont suppose anyone would want to knock up a tutorial quickly on it :D ...its one of those tutorials that would be helpful to heaps of people (I cant find any tutorials on 3D swept and tapered wings)

Or at least point me in the right direction.... I'm pretty stuck and running out of time. Any help would be immensely appreciated. Plus I plan to give back to the cfd-online community too, so that can only be a good thing :D

Far May 11, 2013 03:00

These are the ICEM CFD files for DLR F6 for DPW2 and probably done by The Simon. Wonderful blocking.

PS: At the moment I am recording tut on turbomachinery meshing. The purpose is to make the mesh (Full Hexa) from simple CAD model and within 30 minutes. I have in mind the tut on this geometry as well.

ShowponyStuart May 11, 2013 05:01

That is a seriously nice mesh. That blocking is great, its more or less what I had pictured in my mind, but I just cant figure out how to execute it.

Im going to see if I can copy the way he has done it.....not holding my breath though. Is it possible to do what he has done but with a sharp trailing edge and curved leading and trailing edges?

diamondx May 11, 2013 17:17

blockings of wing can be approached in many ways... that blocking was not great i should remove that video. instead i follow simon's 2d version , and apply it to a 3d geometry...

Far May 11, 2013 20:22

Ali :

Your blocking around the wing resembles to the Simon's blocking. The main problem is how to handle the 3d wind with finite thickness (wing tip blocking).

diamondx May 11, 2013 20:32

i will project everything til the far field

ShowponyStuart May 12, 2013 01:12

Hi diamondx, I agree with FAR that your blocking is not that bad for 3D wings that are a straight extrude but for wings are swept and tapered the blocking around the tip needs more work to keep it clean.

Even if you could still use similar blocking but then transition to a big block (like in the DLR F6 FAR posted earlier) it would be a lot cleaner and you wouldnt be getting any weirdly orientated or sized elements in the farfield region.

Its obviously doable, but Im just not sure how to go about it.

ShowponyStuart May 12, 2013 01:56

Just found out I have bigger problems then just my tip blocking.
This the pressure contour at about 1 back from the wing

This is the front view of a swirling strength isosurface plot (wing is middle of right hand size)

Top view of swirling strength isosurface plot

Not sure whats going on there but that simulation was with this mesh

(SST with gamma-theta transitional turbulence)

If I cant get this sorted out within the next day or so Im going to have to abandon it.

Far May 12, 2013 02:06

Go to o-grid tab and select all blocks inside the wing and click on apply. You will get o-grid around the wing.

Edit: Don't look at the results at the moment, they may be misleading. Only concentrate on good mesh and getting good convergence.

All times are GMT -4. The time now is 18:18.