Abaqus dynamic explicit analysis with connector elements

[Tinni1]

Hi,

I am looking into the post-buckling behavior of cold-formed steel columns (studs) connected at both ends with a small piece of track sections.

The stud is connected to the track by a 5.5 dia tek screw. For modeling the connections I have modeled holes of 5.5 mm dia on the stud and track sections and provided connector elements with calculated screw thickness.

The stud and track assembly is loaded through a discrete rigid plate element at both ends. I have specified general contact conditions. The boundary condition is a top-hinged bottom fixed.

I am conducting a dynamic explicit analysis.

When I am looking into the failure pattern in the visualization mode, it is showing that the rigid plated has penetrated the stud. Although I am not getting any errors. Does this mean, my results will be erroneous?

I had conducted the same analysis without holes and connecter elements, and no penetration was seen. Could you please provide any direction here?

[URL unfurl="true"]https://res.cloudinary.com/engineering-com/raw/upload/v1639509984/tips/Post_buckling_behaviour_with_holes_s6hypv.docx[/url]

 

[FEA way]

Make sure that the rigid plate is properly constrained. In this case it will likely need boundary conditions fixing all degrees of freedom apart from the one towards the column.

 

[Tinni1]

Thanks for your reply.

Do you mean to say, I can not allow rotation of the top rigid plate?

But my practical loading scenario is that the top plate may undergo rotation. That's why I have allowed it.

Is this happening because of modeling the holes for the Tek screws? If so, is there any other way to model it?

As the other model, without a hole, I allowed rotation of the top plate and the penetration was not seen.

Could you provide some direction here?

Many thanks in advance!

 

[FEA way]

It’s hard to say what exactly happens in this model based solely on the final configuration. You’d have to take a closer look at the deformed shape of the model in all frames of the analysis (animation can help). Then you should notice what’s going wrong. A few runs with different boundary conditions and modeling approaches will let you find out the correct setup that matches real-life conditions and eliminates spurious effects at the same time.

 

[Tinni1]

Many thanks!

 

[g.alshamsi]

I would personally not model the holes for a CFS assembly (unless I need to model the actual damage/elongation of the hole). You can partition your elements and have connector elements with the proper behavior (cartesian should work for CFS screws). However I think the problem is in modeling the rigid plates from what I can tell. Can't really pinpoint the issue like the other users mentioned however If you could provide more details on the original configuration and the way you applied your loads/BCs in your model we can probable get to the bottom of this.

 

[Tinni1]

@g.alshamsi,

Thanks for your response.
I have modeled the rigid plate as a discrete rigid element, to idealize the end condition of an experimental setup as I do not want to model, every single element, that is not of interest and has been provided to create a real-world boundary condition and also to avoid the complexity of modeling and increased time for numerical simulation.

Could you let me know, why do you think modeling the rigid plate is a problem here?

Many thanks!

 

[g.alshamsi]

From the picture that you posted, it looks like the rigid plate is passing through your stud. At least that what I can understand from the picture.
You can try to add a reference point at the end and use a coupling constraint or a rigid body constraint to simulate the end plate.

 

[Tinni1]

Thanks for your response.

I fixed that penetration problem by changing the boundary condition of the rigid plate.