Buckling of flat plates using Quad8 element

[Deleted User]

I am having difficulty trying to model flat plate buckling using the Quad8 element
(e.g., a square plate in compression similar to the Mecway sample problem having Hex20 elements). I have no problem modeling plates having Hex20 elements.
Do any sample buckling modeling having Quad8 elements exist?

John (in Seattle)

Victor

Here's the same model converted to quad8. I did this by setting the Z-coordinates of all nodes to 0 then merging them, and setting thickness in the material properties.

disla

Hi,

I'm having some difficulties too. Not with linear buckling like in the example but with the nonlinear approach + S8/S8R. (S4 works but plasticity is not properly captured).
I have seen one person solving the problem with a layered approach combined with bilinear plasticity, but Mecway doesn’t allow that combination.

REF min 9.20. :

@Victor , ¿Could we allow that combination in the future?

@johnhlaakso , I'm posting an example with analytical result for your validation. Buckling factor 112MPa.

Thanks

disla

Based on the proposed example, this picture shows the GNA solution with the layered approach (6x2mm ) and how the first three modes are triggered. Like in a spectrographic analysis , displacements show three clear snaps at (0.31s , 0.39s and 0.64s). For a linearly increasing load of 355MPa, that represents ( 110 MPa, 138 MPa and 227 MPa ) that as a first approximation agrees very well with the three first expected buckling modes ( 112MPa , 140 MPa and 233 MPa).

@Victor. Although it works very well, my goal would be to add material nonlinearity ( + later introduce imperfections to achieve the full GMNIA analisys).

Victor

@disla, I don't want to add a plastic laminate option at this stage because I think being able to input the data in the GUI would add little value compared to writing CCX cards directly. Without a redesign of how laminate data is represented, there would be a lot of duplication and complication to allow all the different ways they can be combined.

disla

Hi Victor,

I don’t have programing experience to evaluate how much work could be involved in one of these features. I was looking at one .inp I downloaded at YouTube below the referenced video and it seemed to me it just added the *plastic card. I’m talking about isotropic shells not the most general case. That means the plate is only decomposed in many layers, but all the material properties are the same in all directions.
Anyway, thanks for looking at this possibility. I will search how to write the custom cards.

[Deleted User]

Thank you Victor for your prompt sage guidance. Smashing the Hex elements to Quad elements is a slick answer to my question. (Extending such linear models to treat non-linear buckling has application to post-buckling of "brittle" carbon composite plate structures.)

disla

I have seen this is also done by the main Steel's connection design software in the market.

"The shell is divided into five integration layers through thickness of the plate at each integration point and plastic behavior is analyzed in each point. It is called Gauss–Lobatto integration. The nonlinear elastic-plastic stage of material is analyzed in each layer based on the known strains. Only the maximum stresses and strains of all layers are shown."