Hi,

does anybody know how to copy a blocktructured mesh. for example if i need the "same" mesh 3 times?

i am able to copy the mesh and the blocking. but all the associations are gone and the mesh is not written out correctly...

thanks in advance

michael

I am afraid that that can't be prevented using Icem.

All the associations are gone, because on the other half of your geometry the curves etc. have other names in the tetin file, like for example c45 which becomes c91 when you copy it. So all your projections are a mess. The problem is that Icem doesn't associate your projections to the curve/surface names that you give, but to the curve surface names that icem gives itself (names like s01,s02 etc. in stead of "fuselage")

I've never seen anyone solving that. The same problem occurs when you use the replay button and change your tetin a slightly slightly bit (addind a curve for example), replay then suddenly doesn't work anymore, even if you use the same names, because icem, below in it's source code and tetin does NOT use the same names as soon as you change something. This make the replay button completely useless.

I hope someone has the answer for you, but I am afraid you'll have to redo all projections.

Sorry for not being a fan of Icem, neither tried any other

thanks a lot.

i am also no fan of icem. but at the moment it's the only mesher that supports our code (FASTEST) ;-(

Doing the associations again is almost impossible (or it't take at least serveral weeks - the geometry is very complex...)

regards michael

Well, let's hope I'm wrong and that someone comes up with a solution.

In general I can say that I have been doing a lot of "redo" jobs when working with icem and others around me did the same (I finally changed my job, hahaha).

You may try to find out the names of the associations and then change your tetin by hand and in a text editor (I used to do that to get rid of some icem errors, as well for the *.blk file, both are ascii), but if the geometry is very complex... ...I'm afraid not.

The best thing would be to find out only once how these *blk files and *.tin files are structured and then generating the associations directly with fortran. I did generate some tetins with self made up names like c01, c02 etc., but never managed to do that for *.blk

In fact, if someone knows how to generate this stuff, his answer is welcome.