Linear perturbation and local buckling

[aeroinspace]

Hi!

I am working on the model seen below for the critical buckling load and the corresponding buckling mode on the composite panel. pls see attached file

I am using the linear perturbation analysis. Now i need to check the results with analytical solution but there is about 100% difference in both. Should i compare the results directly given by Abaqus or not? I am applying a unit load.

Moreover how can i find local buckling load in one bay only from this Abaqus result?

regards
Yasir

 

[SWComposites]

1) what analytical solution are you using for comparison?

2) the mode shape looks odd; why is the bucking mode only at one end of the panel? what are the loads and boundary conditions?

3) the buckling load for 1 bay should be just the total load divided by 3 (assuming equal bay widths)

 

[aeroinspace]

Thanks a lot for the reply. Please see the attached image for the info you asked for. I am not sure actually why is the buckling mode at one end of the panel. Can u explain this if its possible?

 

[Nicolau]

Hi to all,

is it possible to make contact in a buckle amalysis?

regards
Nicolau

 

[gwolf2]

aeroinspace,

If you look at the contours carefully this looks (to me at least) like a 3rd order bending mode, although I am not sure why it is more pronounced at one end.

Without knowing the physical reality some of the rotation edge constraints look a little unusual - it would be worth checking.

As to the discrepancy between the analytical solution and ABAQUS linear buckling: Are you sure that your buckling equation is for bending mode 3 and not mode 1 - this will make a difference.

All panel buckling equations have some term or look-up table to represent panel edge rotational stiffness, it is common to take the average of clamped and pinned for edge stiffened panels. For this analysis you would need a better number.

I don't recognise the buckling equation which you show, it may be worth trying some other calcs e.g. ESDU, Bruhn, or Niu.

Lastly, check that your FE composite layups are properly defined and that you use suitable stiffness values in your hand calc.

 

[corus]

It may buckle at one end as that is local to the position in which the load is applied. In addition if you're applying the load to the edge of the shell then you are basically applying it offset to the centroid of each section and/or the whole of the combined section. This implies a bending load at the edge together with a direct load, hence the bending of the panels as the 1st mode as the section is less stiff in bending across each of the panels.

corus

 

[neoflight]

hello aeroinspace,

There is no way to specify buckling mode (global or local) in abaqus. it shows just the eigenvalues. We have to visually inspect the model to realize which is what...

try applying a load which is at leat 60% of the eigenvalue you got from the unit load analysis...

there are some papers which will help you with closed-form equations for buckling of stiffened plates.

caejournal.com: a place for computer aided engineering: abaqus, ansys, CAD, optimization, and much more