How to calculate the buckling load using ABAQUS eigenvalues analysis? 1

[sbaboy]

Hi,

I am trying to simulate the buckling behaviour of an aluminium plate.So,i entered the dimensions of the plate in meter(0.41*0.25*0.0032) and the young's modulus in Pascal (70*10^9).
The boundary conditions that i applied are : *At the bottom :U1=U2=U3=0.
*The top : U1=U3=0.(U2 is unconstrained!)
*Side surfaces : U1=U3=0.
Then ,I applied a unitary pressure load (=1).The result of the eigenvalue analysis was : the lowest eigen value=6.4563*10^6!!!
So, my question is : How to calculate the critical buckling force (Newton)?
Please help!

 

[sbaboy]

anyone can help me please??

 

[TGS4]

Your analysis seems to be telling you that the linear-elastic buckling load is your load multiplied by your eigenvalue.

Will your structure become plastic before it gets there?

 

[sbaboy]

That means the buckling load in this case is 6456300 N = 6456.3 KN??? it doesn't seem logical I think...

The structure doesn't get plastic , it remains elastic until the critical buckling load.

 

[TGS4]

You're going to have to give a little bit more than what you first presented - a picture would be super-helpful. How exactly is this thing loaded. At what load would it otherwise become plastic if it didn't buckle?

 

[sbaboy]

This is a picture of the plate. I hope this can help you (have a look on the first message above).

 

[TGS4]

I don't see any picture? How is your load applied? At what load would it otherwise become plastic if it didn't buckle?

 

[SWComposites]

if you put a pressure load on a flat plate, it will not buckle, it will bend.

 

[sbaboy]

Sorry, this is the picture

TGS4: Can you explain me,please,how can I know the load at which the stucture will be plastic (using Abaqus)?
SWcomposites: Thank you for replying, but i really don't understand why if I put pressure it will bend not buckle(so what should I apply? Concentrated force?)

 

[TGS4]

Don't use ABAQUS to figure out what stress level will cause plasticity - step away from the computer and pick up your pencil. Do an F/A calculation of stress and compare that to the yield stress of your material. For a sanity check you could also compare your results to Roark.

 

[SWComposites]

1) ok, so you are applying a "pressure" (stress) in the plane of the panel on the end? no? if so, then the plate will buckle under that load (assuming it is compressive)
is the "pressure" tension or compression?
what is the units for the applied "pressure"

2) this problem can be solved with a text book solution; have you calculated the buckling load using such a solution so you have something to compare the FEM result to?

 

[TGS4]

SWComposites - you raise a very important point about units. Obviously, in any analysis, your units need to be consistent. Applying only a unitary load may confuse the analyst about the units of the loads and the results.

There is probably something in Roark Table 15.2 that will get you to your textbook solution.

 

[sbaboy]

Hi,
I think the problem is solved. I was considering wrong boundary conditions and inconsistent units.
In fact, this was just an exercice for me to familiarize myself with buckling simulation on Abaqus. Now,I intend to modelize the buckling of a composite sandwich beam (dimensions 16in * 2in) which is composed of 4 glass fiber layers at the bottom , 4 at the top and a core layer (PVC A300).
This sandwich is manufactured using infusion process ( epoxy resin). The type of reinforcement used is glass fabric plies .
First I wonder if i could consider the plies as isotropic in this case (if so, how to determine the young's modulus of the fabric knowing the orthotropic one?).

Also, i am hesitating between two methods to modelize this beam :

1) Create 3 parts and assemble them ( 2 * 4 layer part + PVC core).
2)Consider the structure as composed of 9 isotropic layers and define them directly in one part using the branch "composite layups")

I hope i've explained well.

This is a picture of the experimental test that i have conducted and i want to validate numerically.

http://i44.tinypic.com/im1p1y.jpg[/IMG]]

According to the picture, can i consider the beam as clamped? Thanks for all.

 

[SWComposites]

suggest:
3)Consider the structure as composed of 9 orthotropic layers and define them directly in one part using the branch "composite layups")