MES - Wire compression - evaluating the feasibility

[Eazabella]

Dear Sir or Madam,

i recently started working with Autodesk products and really like it. I get to know more and more about how to handle them and it is quite fun as an engineer. I would like you to evaluate the feasibility of my next project:

I constructed multiple (up to 100) twisted wires (diameter of 3 mm each and lenght of 2-3 m) in Inventor. (see 1. in attachment)

I now want to compress them into a nearly triangular shape with Simulation Mechanical 2015. (see 2. in attachment)

Is this possible ? And if so then what do you think about the necessary hardware performance? I ask this question because in an Autodesk Webinar (MES Introduction) they said, "srew like structures" are very hard to simulate with MES because of the many faces.

 What is most important: Can I export the deformed model of the wires?

I am really excited about your answers and thank you in advance!

PS: I put a screenshot of a Blender sketch in the attachment so you can better understand my problem.

[arizawilmer]

If I understand correctly your problem you want to know the loads when you give the final shape to the multiline wire.

If I am OK . there is a lot of surface to surface contacts. That can be a problem of how much can take to simulate the model.

I will draw the basic parts that gives the form to the multi line wire and set all the contacts as surface to surface with the proper settings.

the basic part will  have a prescribed movement and the wire should be fixed in the extremes. That maybe is possible.

In respect to export the deformed model I know that is a old requirement on the forum that it was tall to be practically impossible

I don't know if with the NASTRAN editor now is possible. In you case i see why you need this. I hope an answer from someone of autodesk about exporting a deformed model to inventor.

[arizawilmer]

Sorry for posting again but i just found this about the exporting step  ....check the link....http://forums.autodesk.com/t5/simulation-mechanical/export-deformed-shape-to-step/idi-p/5408821 there is a solution very complicate but a possible solution.

[Eazabella]

Hey thanks for the link. I really cannot imagine why this is so difficult / not standard yet. A direct CAD export of deformations would be great.

Did you mean (by your first message) that you want to set up a model for simulation?

Thank you very much

[arizawilmer]

If i understand correctly your problem you need to know the internal loads for compressing the wires in a specific shape. As in the image I imagine that a machine has to compress the wire . You should draw the basic elements that give the shape to the wire. You can apply the prescribed motion or force to them. With the correct constrains you can get the internal forces and deformations of the wires.

I assume that you want to simulate the process of defroming your multiline wire in your desire shape. Please correct me if I am wrong. In theory should be possible. slow but possible.

[Eazabella]

I dont need to know the exact forces. I only need the deformed wire model. I guess the setup should look like this (Attachment) ... with the straight wall in the background fixed...should the wires be fixed anyhow? Because they will move towards the straight wall. What do you mean with : constraints?

Thanks for the talk 

[arizawilmer]

Yes...it is going to need to be fixed in Z direction in one of the sides to avoid any strange movement. By the look of your model ,the longest part outside of the shield can be fixed in Z.

[AstroJohnPE]

Hi Eazabella,

I suggest that you do some really simplified test models to learn what's involved, what you might be up against (maybe days to debug the model and get results), and then find out if you can get the results that you really want. (I'm not clear what results you want from the analysis.) You may want to start with a 2D model (2D elements) if possible. Certainly a strand with 100 wires, 2 to 3 meters long is not a good starting point!

Good luck.

[Eazabella]

Hi John,

I totally understand this, and have already started smaller MES simulations. 2D unfornutaley is not possible because of the twisting of the individual wires. I asked this question to know if it makes sense to get very very deep into the Simulation program (I´m ready for that of course). If someone would have said, it is not possible at all, i eventually could have switched the simulation programm (if there is any program that makes this possible).

I need the deformed model for further researches and it is important, that it is build this way.

Thank you very much. I hope I can make you understand better now.

[arizawilmer]

What John want to explain you is that you need to start with a simple model a model with straight wires .

Every time that I do a simulation outside of the normal problems I have to start with simple things because i have to bug the problem's that i would find in the way.

That if it is possible to do what you want with FEM. yes is possible, but if you only need the deformed model there is other ways to obtain that model from experiments in a much faster way.

FEM is only an aproximation and the real shape can change.

[Eazabella]

okay thank you again. i´ll try an easy model first.

What do you eman there is a more easy way to get the deformed model by experiment?