|March 13, 2013, 03:02||
Troubles with ICEM Hexa scripting
Join Date: Mar 2013
Posts: 1Rep Power: 0
I am working on a script to enhance the blocking process in icem hexa ..
Maybe you all know the beta feature "3D Key Point Blocking".
It creates a blocking, with yet done blocks, derived from the geometry.
Bad with this feature is the thing, that it does not consider surfaces. just 3D objects.
So I want to enhance this by spliting the edges that are parallel to the surface by the corner points of the surface.
The thing is: I know the corner-point-names plus the curve-name between them. But this is no edge!!
The command to use is "ic_hex_split_grid" but this will just accept the numbers of the points at the ends of the edge and not their names.
I know this sounds like I did no researches at all but in fact I did a lot. Everywhere is written that to start you can just watch the replay. But sorry, the replay is one of the most intransparent things I EVER saw. I mean for sure the engine can form the exact command in Tcl when it knows ALL the information. When you click on an edge it knows like one bilion times more than us.
Maybe I do know too less but at the moment I am kind of disapointed by Tcl because the information I am provided by is so few.
I would be really glad if anyone can help me! Because at the moment I am even reading and encapsuling the .tin geometry file by myself in python, because I was not really able to get the needed information by Tcl.
I know ICEM is really strong on what it does, so it would be very helpful if I could script it properly.
Thanks to everyone and sorry for writing in a little negative attitude, but I had to get rid of this
|March 14, 2013, 09:04||
Retired from CFD Online
Join Date: Mar 2009
Location: Ann Arbor, MI
Blog Entries: 1Rep Power: 39
That key point blocking is probably not the best way to start a scripting because you don't know how the blocking will look. You are trying to make a script that is independent of the topology, which is very difficult.
When I script blocking, I use the top down method and make my own splits at points, etc. Since the topology is fixed I always know what edge I need to split next.
There are ways to make topology independent scripts, but you need to use queries to investigate the model. For instance there is a command to find out the location of a point and then there is another command to find the nearest curve or surface to that location. I don't recall if there is a command to find the nearest edge...
|April 24, 2013, 10:09||
Join Date: Jan 2013
Posts: 260Rep Power: 9
I just want to know if "find location" tool is only for the points. I want to make sure about it. Because I´ve been reading the Help Manual and I´ve not found that this tool can measure a "vertex location" too.. but.. when I use this option with any point gives me a small yellow point but when I use it with a vertex gives me a big yellow point. So.. I just want to make sure that gives me the real location of this vertex. Anybody know about it?
|Thread||Thread Starter||Forum||Replies||Last Post|
|[ICEM] ICEM Scripting Issues||tylerplowright||ANSYS Meshing & Geometry||17||June 3, 2012 21:23|
|[ICEM] ICEM Hexa Meshing Automation||Haasini||ANSYS Meshing & Geometry||1||April 24, 2012 22:46|
|ICEM hexa mesh built with v10 deformed in v11 (degenerated edges)||cfxicem||ANSYS Meshing & Geometry||1||October 14, 2009 15:06|
|ICEM HEXA - Where to start?||flattie||CFX||3||May 26, 2009 09:12|
|ICEM HEXA Grid||Manish||Phoenics||0||March 12, 2005 06:07|