Ansys WB Nonlinear problem

[crigido3d]

Hello everyone. I'm trying to simulate the bending of a plate in ANSYS WB.

I had to play some advanced options that not finish understand. I have some screenshots to see where I was.

I've drawn a small geometry in MD, just a test.
I want to know the support reactions plastic deformation occurs when the plate. I wish I could make a simulation like the one shown in this video, but I could not do it.

[URL unfurl="true"]http://www.youtube.com/watch?v=ebXnuNDnZvI&feature=plcp[/url]

Someone could guide me or tell me I'm doing wrong?
I made a series of tests:
I reduced the displacement of the roller than 5 mm.
I deleted the "contact sizing" and the result was:
[URL unfurl="true"]http://www.youtube.com/watch?v=M-PYBV-PQ8A&feature=youtu.be[/url]
Do not understand why the ends do not rotate. I tried using contact type "frictionless" and get the same result.
Also if I want to impose a greater displacement fails. Any help?
Greetings!

 

[IanME]

Crigido,

I noticed a couple things when reviewing your post that I believe could be the cause of your problems:

1. Mesh across cross sections of plate and cylinders is poor. 2 elements across these widths may not be enough. Especially because you were forced to remove your contact sizing mesh option to get the analysis to run, I would look into defining your mesh more carefully.

Also, is there a reason you chose to drop mid side nodes? I don't suspect this analysis is computationally intensive....

2. What is the displacement constraint you've placed on 1 face of your flat beam? I suspect this is why the beam is not "bending" at the points because you've constrained it rigidly. Could you explain this constraint?

Lastly, I suggest you review the ansys contact technology guide (

http://www1.ansys.com/customer/content/documentation/120/ans_ctec.pdf

) to gain a better understanding of the complex frictional connections you've chosen for this analysis. This could be the source of your problem but I am not well versed in these constraints.


Some things for you to think about. I hope this of some help to you.

-Ian

 

[IanME]

Please post back if you have not solved your problem as I am quite interested in your post.

-Ian

 

[crigido3d]

Hello Ian! Thank you for your comments and suggestions!
I don´t think that the problem is the mesh, I am inclined to think this is because the boundary conditions or contacts. In any case proves refining the mesh.
I have restricted movement in the direction perpendicular (Z) to the plate in two faces. Free movement in the other directions, do not look bad.
“dropped” option for the “Element Midside Nodes” this means that all the elements will be simple 8-node brick elements. I understand that Brick element technique is correct for large deformation problems as this case.
I will look at the article link and I will continue proving. I will keep you abreast of developments. If you have other suggestions let me know.
Thank you very much, greetings
Sebastian

 

[CarlesFEA]

Crigido,

It seems that geometry and loads are symmetric => if so, you could just study half of your model.
I don't think that "Adjust to touch" setting in your contact definition is to be used in this case => I'd rather leave it to Add Offset, Ramped with 0 offset (assuming that initially parts already touch each other)
You could try to slice your plate near the contact areas and redifine your contacts so no overlaping exists in the contact definition (actually, the bottom surface of the plate is used to define contact with both cylinders)

I hope that helps

Carles

 

[IanME]

Concerning mid side nodes see:

http://www.padtinc.com/blog/post/2011/02/09/mid_side_nodes.aspx

. I have not heard of the large deflection brick method. I was acting under the assumption that mid nodes could only improve the accuracy of your model.

If you are confident in your mesh I wont belabor the point anymore.

As for the 'Adjust to Touch' constraint mentioned by Carles I found this quote from another padtinc blog (great resource) : "If you have an assembly that should be
initially in contact, but isn't, you can use
the “adjust to touch” to close whatever
gaps exist. "

Does not seem applicable here as Carles pointed out.

Best of Luck,
-Ian



"All models are wrong, but some are usefull" - George E. P. Box

 

[crigido3d]

Friends, thanks for the help, but for now, I have not been able to solve it. I will continue trying. When you have an answer I will make you know. I leave the geometry of the problem in case anyone wants to try. Thank you very much.
Sebastian

 

[IanME]

Sebastion,

I'm sorry I wasn't able to provide any better help. I had actually considered modelling a mock up of your problem here at work, but I've been much to busy and we only have one WB license (and 4 people that want to use it!)

Here are some of my last ditch thoughts:

- Do you have the input/output files for the APDL analysis that was constructed successfully? I would consult these files in particular to see how the contact was formed between the bottom two rollers and the plate in bending.

- Looking again at your posted youtube video it definitely seems clear that the problem is the contact between roller and plate. Have you explored 'no separation' contact between faces? Or a frictional contact with very low coefficient of friction? Just some ideas...

- You may need to write an apdl command snippet to define this contact if the available options in workbench aren't enough. I'd again point to the Contact Technology Guide I posted in my first response for the myriad of options available. There are a lot of different contact elements.

Again I am sorry that I haven't been able to be more help. Often times problems in ANSYS appear simple and are not at all.

Best of Luck,
-Ian

"All models are wrong, but some are usefull" - George E. P. Box

 

[crigido3d]

Dear Ian, please do not worry. You have been very helpful.
I not have the input/output files for the APDL analysis, because the analysis sucssefull that is seen in this video, belongs to a user of youtube.
I've tried all the contacts and none worked. Finally I decided to remove the bottom rollers, and in its place have placed restriction of the type “simply supported” . In this opportunity I have not placed any other restrictions on the plateand and result is:

http://www.youtube.com/watch?v=pBW3UJaW1EE

I think this definitely shows that the problem is the contacts when large deformations and rotations.
I am not satisfied yet. I will keep reading and testing.
Thank you very much again, Greetings!
Sebastian


"Persistence prevails, like a stream that is temporarily blocked by boulders and then collects force enough to overflow onward."