Simulation Mechanical Forums (Read-Only)
Welcome to Autodesk’s Simulation Mechanical Forums. Share your knowledge, ask questions, and explore popular Simulation Mechanical topics.
cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 

MES Non Linear Beam structure with plates

6 REPLIES 6
Reply
Message 1 of 7
Dave_Spence
690 Views, 6 Replies

MES Non Linear Beam structure with plates

I am running an MES with non linear materials simualtion of an automotive frame which is built primarily from RHS sections.  The primary focus is for a rollover.

 

I have made a simple model of the whole structur with  beams and their appropriate cross section data entered in and i have had success with that.

 

There are some important stiffeners which are plates and need to modelled in the simulation.  I have drawn them in however i am unsure as to the element type to use as there is no "Plate" element type in the MES analysis.  

 

I have read a considerable amount of posts and autodesk help information, and i still cant find a difinitve answer as to the most appropriate solution for my problem.  Attached is a picture showing the frame and plates.

 

Any help would be much appreciated.

Regards

 

Dave
 

6 REPLIES 6
Message 2 of 7
zhuangs
in reply to: Dave_Spence

Shell element can be used for the plate parts in MES.

 

-Shoubing

Message 3 of 7
Dave_Spence
in reply to: zhuangs

Thanks for the quick reply. 

 

I have run a quick simulation and i am now finding that the plate/shell, is folding on itself as the frame rolls.  attached is two images of the setup and the folding.  here the plate is only just started to fold, but as the frame rotates more it pretty much becomes flat.  As if it is pivoting around the midpoint where it joins the beams.

 

I am assuming that i have constrained the plate to the end of the beams incorrectly.

 

Could you please shed some light on this.

 

Cheers

Message 4 of 7
zhuangs
in reply to: Dave_Spence

It seems that the shell part and the beam parts are not boned together.  You can send the model for fast investigation if you want.

 

-Shoubing

Message 5 of 7
Dave_Spence
in reply to: zhuangs

Thanks again for the reply

Could you please elaborate on boning,

Cheers
Message 6 of 7
AstroJohnPE
in reply to: Dave_Spence

In simulation (Finite Element Analysis, FEA), loads are transmitted from one part to another through the nodes. So if the nodes are not connected, or not connected correctly, unexpected things will happen.

 

For example, a majority of parts are connected together. This is done in the simulation by having different parts share the same nodes. Another way of stating that is that the meshes are identical where the parts need to be connected or "bonded" together.

 

In a linear stress analysis with surface contact, the meshes must match between the parts. Although you cannot see them, the software automatically generates elements -- "contact" or "gap" elements -- that connect the two parts together. In nonlinear/MES, surface-to-surface contact also creates invisible elements that connect the nodes on surface A with the nodes on surface B. This is how the contact loads get transmitted.

 

Now back to your model. How many nodes on the plate/shell elements are connected to nodes on the beam elements? I could not see what was happening in your image. It would be helpful if you would turn off the 3D visualization of the beam elements (right-click on the part in the browser and uncheck "3D Visualization"). And you might want to include 2 or 3 images that show how the shells are deforming over time, preferably with the model rotated as needed so that the orientation of the shells are the same in each image. (That is, eliminate the tumbling of the model which makes it harder for those of us that know nothing about your model to understand what the images show Smiley Happy.) Off hand, I cannot envision how the shells are getting flatten by the analysis.

Message 7 of 7
Dave_Spence
in reply to: AstroJohnPE

John Thanks for the reply

 

The plate elements are connected to the ends of beams at a single node, which is where i am assuming the issue is.  The plate is a bent section which is welded between to beams.  I hope the pictures provide more clarity.  

 

Also, i have another model where i have several plates in corners to represent the triangular stiffeners that are welded in and found that the expected rotation was visibly restricted as the model tried to rotate in the simulation. It is clearly due to the muyltiple plates, but i am yet unclear as to why they are affecting the rotation.

 

 

 

Regards

 

Dave

 

 

Can't find what you're looking for? Ask the community or share your knowledge.

Post to forums  

Autodesk Design & Make Report