2d hemisphere in wind tunnel
 

i just wanna calculate drag of a hemisphere in wind tunnel .i had a problem in first step when i tried generate mesh file in icemi . i dont know which far field is better . i tried some but i had problem in aspect ratio . if some one can help me it would be great . i must add that wind hit the flat side of hemisphere in tunnel . sry for my bad english

attach geometry file

i attached my icem file with a random farfield that i dont know if its good or not . tnx for your help

2d case..... Shouldn't it be 3d?

our teacher did it for a sphere with a bigger sphere for far field and as far as i know we do it in 2d and the axis line make it a perfect sphere . for my case i did same thing with hemisphere . for a sphere mesh had good quality but for hemishpere the flat line causes some aspect ratio problem .

My favourite post
 

I like this post very much http://www.cfd-online.com/Forums/ans...sh-fluent.html
It is classical work using ICEM.

the topic you linked is great but i feel my issue is very simple and primary

the project our teacher wanna us to do is a simple one like the one he did in class (because we are beginner and amature )
here is a link for the one he did for sphere . he did it in 5 min and asked us do it for different geometry like mine(hemisphere)
and after that exported it to fluent . and i know what to do there for drag and ...
http://encodable.com/uploaddemo/files/sphere.zip
( i just wanna know how should i take my far field for a good blocking and after that meshing ) again sry for bad english ;)

tnx a lot for your attention and time .

Ok . Done in five minutes (I wasn't counting :p.)

http://img403.imageshack.us/img403/2716/12531369.png

http://img818.imageshack.us/img818/646/30150213.png

Your case is much difficult than your teacher did.

here is how i changed the far field, of curse you need refinement so you can capture boundary layer. i suppose you know how to play with it. i'm sure there is a better blocking strategy

tnx man you helped alot . now i must look at it and figure it out how u did that ( im amature as u know :D)
thank you very very much

From left to right? Any angle of attack?

yes From left to right and no angle . btw you split it to 9 blocks ?

I just made three half ogirds. one for body, one for boundary layer and one for outer layer

I worked on what you suggested and here is what i did ( i should thank diamondx too ) . i think it has lots of problem . i exported it to fluent and when i checked quality the Minimum Orthogonal Quality was 6.62394e-01 . i dont know how i can enhance the quality .
http://uploading.com/files/c218e433/New%2Bfolder.zip/
and something else . i need check the drag coefficient from experiment with what i calculate . do you have any idea where i can find one? i search a little for it but i didnt find any .
again thanks for your help

* sry that i upload my file in different place . i just wanna some expert (like you ;) )see it with detail and help me correct mine

0.6 is very good quality. Minimum needed is 0.01 by fluent. Meshing is good, but you have very fine mesh in far field..

3D hemisphere
 

Hi, I have some problem related to this topic: I want to compute drag coefficient on a hemisphere and I know that the best way is an axisymmetric simulation. but for training I want to run a 3D simulation and need a 3D mesh.
actually I couldn't make a good one with ANSYS ICEM!
could you please help me?
Thanks.:)

Follow shpere-cube tutorial. What do you mean by good quality?

were you looking for this. Just deleted lower part of sphere!!! do you need it? I am not sure.

try this.......strategy is similar to sphere-cube tutorial