Engineering & Analysis Methods > Local Buckling & Local Postbuckling

Allowing local buckling below limit load (how to allow a local buckling MS<0)

(1/1)

HyperSizer User:
How can I tell hypersizer that for buckling check Eigen value of 2.0 is also ok (say based on testing or from literature I know that structure has post-buckling capacity up to twice as much as predicted by linear buckling) and I want that to be reflected in my optimization (in other words, I would like Von Mises stress to be still limited by Fty or Ftu depending on the situation but minimum permissible buckling Eigen value to be 2.0 instead of commonly used 1).

Phil:
I think what you are saying is that you want to "allow" an eigenvalue of 0.5, right?

Let say that for a particular panel, the predicted local buckling load is 50 lb.  But you know that panel can take 100 lb, right?  So if you put in 100 lb, you would get an eigenvalue of 0.5.  (Eigv = Buckling Load / Applied Load ) I think what you want to do is tell HyperSizer that this panel with an eigenvalue of 0.5 is ok.

With the default settings, HyperSizer will say the panel has failed if the margin of safety < 0.  This is the same as saying the panel fails if the Eigv < 1. So in the above example, for any load greater than 50 lb, HyperSizer would say the panel is failed.

However, on the failure tab, for any margin of safety, you can enter a minimum acceptable value.  If you wanted to allow an eigenvalue of 0.5, then convert this to a margin of safety (MS = Eigv - 1 = 0.5 - 1 = -0.5).  On the failure tab, right click on the local buckling failure mode that you want to assign (right click on the MS itself, or the hashed box) and select "Required Limit Margin of Safety" and then type in -0.5.  If you are working with an ultimate failure mode, like panel buckling, then select "Required Ultimate Margin of Safety".  Now when you size, if the resulting MS is greater than -0.5, then this will not be treated as a failure.  For example, say you entered a load of 75 lb.  This would result in an eigenvalue of 0.6667 and a MS of -0.3333.  However, this would not be treated as a failure, the MS would not show up as a red value, and this MS would not be used to size up the panel further.  If you entered a load of 125 lb, then this would result in an Eigv of 0.4, a MS of -0.6, and this would be treated as a failure.

I hope this helps.

JEA:
Phil,

I am looking at a bonded, two sheet, continuous stiffened panel concept.  When I have the required limit margin of safety set to zero for "closed span local buckling", it calculates and sizes to all of the selected failure modes.  However, when I set the "closed span local buckling" failure mode to a required value of -0.2 margin, it now calculates and sizes to only that failure mode.  As a result, a very low weight panel is now calculated.

What am I doing wrong?

Thanks.

Phil:
I was able to repeat this behavior and it is a bug.  It appears that under certain conditions, when a minimum required margin is set less than zero, and the returned margin is between zero and this required minimum, some of the failure analyses get skipped over.  This should be fixed for the next update.

A workaround that you can use now is to go to the Sizing Form, and from the Options dropdown menu at the top of the Sizing Form, select "Generate More Detailed Reports".  This will correct this behavior.  The only problem is that you must turn on this flag every time you close and re-open the sizing form.  Also, you have no way to activate this setting if you are running HyperFEA.

JEA:
Will do, thanks for the info.