Risa 3D - Non-Building Structures Webinar

[LSPSCAT]

Risa 3D - Non-Building Structures Webinar

At the 52.19 mark in this video there is a discussion about interpreting the plate element results in comparison to code checks. This particular example was run with Allowable Strength Load Combinations, so the von Mises stress was compared to the yield strength with a factor of safety. (Say 50 ksi steel so .6 x Fy = 30 ksi per the example.)

Now, lets say that one had used Load and Resistance Factor Design such that factored loads had been used in the finite element analysis of the structure. The plates stresses reported are now based on the factored load combinations.

For the limit state of combined stress, considering biaxial bending or combined bending and torsion we would use phi x Fy (0.9 x Fy) and compare that that to the von Mises stress calculated based on factored loads.

Similarly, if we had a limit state of buckling due to compressive loads we would need to determine a value for phi x Fcritical and compare that to the factored compressive stress in the plate.

I want to confirm this is correct interpretation of applying the factored load combination stress results to the current AISC 14th edition code.

I realize the code is more setup to handle discrete building elements (beams, bracing, columns), however, does anyone have any references for using it along side general plate and shell element FEA formulations?

 

[JoshPlumSE]

Certain aspects of plate analysis are simple, others aren't. Now, a comparison with the Von Mises stress is really good for a comparison to initial yield. It doesn't say anything about buckling, or post yield strength.

If I'm looking at a plate under pure flexure and LRFD loads, then I usually use an allowable moment (per unit width)equal to:

Phi*Mn = 0.9*Fy*Z... Where Z is the plastic section modulus = t^2 / 4

This would be equivalent to an allowable stress of 0.9*1.5*Fy. This is because the plastic section modulus is so much greater than the elastic section modulus we would have used in the old ASD days (t^2/6).

When you start adding in the effects of bi-axial bending, shear, and axial force, the derivation of an allowable stress become more and more complicated. For a model like that shown in the webinar, I would defer to a Vessels engineer or a Tank Design engineer for a better explanation of how they would evaluate failure in those plates.

 

[LSPSCAT]

Understood regarding the use of more detailed analysis methods for the plates. What I am trying to work around is a structure that is governed by components with mixed code requirements. Things under ASME codes, not necessarily pressure vessels, but items like Below the Hook lifting devices, are allowable stress design vs. the structure which is governed by LRFD. Trying to get a more consistent approach. Most of this in the past was completed all with pure allowable stress design. However, as we get into wind loading and seismic requirements it would be nice to get a more unified approach.