*Tie versus *MPC, pin

[Kiwing]

Hello,

for connecting two parts with a matching mesh at the interface I currently use the "*tie" command, however, it seems that this does not rigidly tie these nodes together as if they were identical.

All node positions are hand-written and therefore identical. Is there a better way to pin these matching nodes together?

"*MPC pin" would require me to do that for every individual pair of nodes (requiring hundreds of lines), but seems to come close to what I want.

What is the best way to connect all degrees of freedom for corresponding nodes on two matching surfaces?

Any help is greatly appreciated!

Kind regards
Marco

 

[Mustaine3]

With a good preprocessor it should be no problem to merge these nodes. A/CAE could do that.

And I think that a Tie creates valid results, even with no direct one-to-one connection.
But if you like, you can use the MPC pin or beam, or Connectors, or Couplings, or Equations, ...

 

[salingerv]

I think as Mustaine3 said, you can create tie constraint using surfaces and not only nodes.
2nd, you can also first create a part and then partition, the interface then behaves like a tied bond.

 

[cdixonm]

I have never used a preprocessor but I am having a similar problem to this , see [URL unfurl="true"]http://www.eng-tips.com/viewthread.cfm?qid=410218[/url]. Could you use a preprocessor to merge the nodes and still assign the two sections their own material properties?

Thanks, Campbell