CalculiX, use Tied contact to connect independent meshes

[UR_]

*** Off Topic ***

How about adding an option to create reaction forces for every restrained boundary condition? This would be easy to implement. Still one would need to open the dat file anyway to get the forces but it is no longer needed to edit the input file.

Increased coverage of ccx's capabilities is veeery appreciated.
A *TIED checkbox from contact task panel belongs to same wish list.
Please see this example http://mecway.com/forum/discussion/725/ ... e-calculix
Bonding meshes without using "Boolean Fragments" is really valuable.

[bernd]

A *TIED checkbox from contact task panel belongs to same wish list.
Please see this example http://mecway.com/forum/discussion/725/ ... e-calculix
Bonding meshes without using "Boolean Fragments" is really valuable.

Give him an inch and he will take a mile.

Code: Select all

Bonding meshes without using "Boolean Fragments" is really valuable. 

What would be the best way to do this in your opinion? AFAIK, for some reason simple Compounds creates glued meshes ATM.

[UR_]

Hi bernd,
as easily can be seen at linked post from Sergio, profiles are independently meshable.
Coarse and fine mesh portions can have a "TIED" contact.
The contact face does'nt disturb meshing of it's counterpart.

from above post (many thanks to Sergio for sharing)

Annotation 2019-06-07 092501.png (39.16 KiB) Viewed 641 times

EDIT:
@bernd: can you please move this posts to a new thread, it's polluting this one

[Jee-Bee]

Give him an inch and he will take a mile.

You are lost now!!

[UR_]

You are lost now!!

It's not that cruel as it looks like.
A checkbox in contact constraint's task panel and some additional lines in inp file writer.

for example see here: https://forum.freecadweb.org/viewtopic. ... ED#p188736

[Jee-Bee]

Yes it is that cruel!!
Not because of the checkbox i'm fine with that It is the units he used... @Bernd is from Swiss and i'm quite sure the have adopt the metric system

[bernd]

The broblem are the multiple meshes. All FEM tree code is only for one mesh. It is true to get started we may only need a few lines code, but for sure we will create regressions for the one mesh case ...

Anyway the good thing is this would be very useful for me too. Means we might see something in this regard. But first I need to get femconcrete branch merged and the GSoC on track ...

OFFTopic: I really would like to have these hexaeder meshes for profile steel sections ...