CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   ANSYS Meshing & Geometry (http://www.cfd-online.com/Forums/ansys-meshing/)
-   -   [ICEM] Blocking strategy ICEM (http://www.cfd-online.com/Forums/ansys-meshing/109535-blocking-strategy-icem.html)

krembil November 20, 2012 04:42

Blocking strategy ICEM
 
Hello,
Is this a good blocking strategy?

http://www.lecad.si/%7Etelenta/pic1.PNG

Thanks!

BrolY November 20, 2012 05:36

It looks like, but maybe you should give more information :

why do you think it would not be a good strategy ?
have you experienced some issues with your mesh ?
Quality is not good ?

krembil November 20, 2012 06:15

Hi Alexandre,

Thanks for your answer.
I am worried about the aspect ratio and angle of the cells as I change the angle of the geometry (inclination of the plates).
It is shown in the pic below.

http://www.lecad.si/%7Etelenta/pic2.PNG

diamondx November 20, 2012 11:11

i understand why you're worried. this thread will help you solve the problem:

http://www.cfd-online.com/Forums/ans...ic-trough.html

llrr November 21, 2012 06:14

Quote:

Originally Posted by diamondx (Post 393220)
i understand why you're worried. this thread will help you solve the problem:

http://www.cfd-online.com/Forums/ans...ic-trough.html

Hi diamondx, I have a similar geometry (window is open at low opeing ratio). Could you please give a little more hints about it. I read the thread that you gave, but I still could not make it succeed. Thank you.

RodriguezFatz November 21, 2012 09:52

Quote:

Originally Posted by krembil (Post 393160)
Hi Alexandre,

Thanks for your answer.
I am worried about the aspect ratio and angle of the cells as I change the angle of the geometry (inclination of the plates).

Hey Krembil,
Unfortunately, if you guide the grid around such a corner, the angel between the to adjacend edges is allways 270. Thus, the two cells have the problem: if you improve one of them, the other one gets worse.
Anyway, if you don't guide the grid around the corner, your wall cell refinement will "radiate" (however you call that in english) into your whole domain.
In both cases you are doomed.

The bad aspect ratio can be avoided by refinements, but again you get these "beams" of refinement into your whole domain.

Also: In your picture, inside the red circles, You have points where 5 edges meet. In my view, it is generally better if the angels between them have equal sizes, i.e. you split the 360 into equal parts...

diamondx November 21, 2012 11:44

Quote:

Originally Posted by llrr (Post 393384)
Hi diamondx, I have a similar geometry (window is open at low opeing ratio). Could you please give a little more hints about it. I read the thread that you gave, but I still could not make it succeed. Thank you.

I'm not sure i can be clearer than Simon's step in that thread (post#10)
Can you post some details about your geometry (pics, project ...)

Far November 22, 2012 02:14

5 Attachment(s)
As I can see from the image angle is around 45 degs for each inclined plat. So the best approach would be making the simple topology and two splits at the front and rear. In this way you can avoid the steep angle as shown in following images.

Far November 22, 2012 02:30

If the angle is greater than 45 then you can avoid sharp angle problems by using the quarter o-grid topology as mentioned by diamondx

llrr November 23, 2012 06:49

Quote:

Originally Posted by diamondx (Post 393466)
I'm not sure i can be clearer than Simon's step in that thread (post#10)
Can you post some details about your geometry (pics, project ...)

Hi diamondx, would you please have a look at this thread? I post my project file there. Thank you. I do not know the strategy to make the angle is bigger than 18 degree at the sharp wedge.
http://www.cfd-online.com/Forums/ans...-geometry.html


All times are GMT -4. The time now is 02:31.