Shear buckling of a deep unstiffened plate 4

[normm]

Friends

I am analysing a steel silo made of roof, barrel, bottom hopper and columns and braces. The roof, barrel, and bottom hopper are all unstiffened plates. See attached diagram.

I have used thin shell elements and can get output contours of Nx Ny Nxy Princepal stress etc which will tell me the magnitude of membrane stresses in the plate. Obviously I will keep these values below the yield stress of the material. But I have been told that you get a problem of 'shear buckling' in the lower regions of the barrel near the column supports.

I am wondering how to check the section for shear buckling. The barrel is 160 inches tall, 120 inches dia. Supposing I know the magnitude of principal compression in the lower region, what cross-section area is to resist that?

Any of your thoughts will be appreciated.

Normm

 

[GregLocock]

Is that the same as t/r buckling - thin shell short wavelength?



Cheers

Greg Locock

SIGlease see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[normm]

Thanks Greg for your response. Talking to some people, I understand that the barrel plate will act as a deep horizontally curved beam spanning between the columns. There will be a cominations of high stresses specially near the column supports. These stresses can be reduced to principal tension and principal compression stresses. It is this principal compression that causes local buckling. My understanding is a bit vague at the moment. Your comment of t/r ( depth/thickness) will indicate the buckling capacity of the barrelplate. But I cannot make out what should be the depth. The full depth of the barrel is too unrealistic a value if I am concerned with local buckling.

 

[rb1957]

i'll preface this by saying i'm not tryng to be smart ...

shear buckling is related to shear, not compression (maybe you're hung up on the word "buckling"). thin panels buckle in shear (try it with a sheet of paper), buckling along the diagonal direction ('cause the buckles carry the shear). in your case i suspect it'd involve a lateral load case (wind ?). look up diagonal tension. i'd've thought it was something that your code addressed, that you shouldn't need someone to say "check this" ...

 

[gwolf2]

normm,

Your understanding is pretty much correct. For methods for buckling stress prediction try books by Bruhn, Niu, ESDU, or maybe HSBC for shear panel buckling allowables. It's a bit more trickey with curved panels than flat.

Reinforcement of the thin shell near the supports where the shear stresses peak will increase the buckling strength. This can be done using thickness doubler plates, or finger plates (series of tapered raised flanges radiating from the column/shell intersection).

gwolf

 

[GregLocock]

The posh phrase is "Wagner tension field". If that helps.



Cheers

Greg Locock

SIGlease see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[40818]

One of the contributing factors about a critical buckling stress for shear buckling is the effectivity of the panel edge members/stiffeners. If you have less support than simple support, then the panel will buckle earlier. A cleverwork around if you wish to keep a thin panel, is to introduce a reinforcing in the panel middle (imagine a cut-out doubler, but without the cut-out). The reinforcing acts to help prevent the onset of shear buckling.

 

[Hansmeister]

Assuming that I understand your problem, shear buckling implies that there is a torsional load on the silo. Or, maybe, you are being asked to consider a seismic case.

At any case, a silo filled with "stuff" should not expect to have shear buckling (similar to a beam shear web).

Perhaps you should get additional information, and restate the problem.