Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations waross on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Interaction of out of plane loading and diaphragm behavior.

Status
Not open for further replies.

Lomarandil

Structural
Jun 10, 2014
1,900
My colleague has an interesting analysis question --

A plate with significant out-of-plane loading (mostly resolved through one-way bending) also sees a moderate compressive load applied along one edge. The two adjacent edges of the plate are simply supported, with the far edge essentially free. So in-plane, it could be a very straightforward diaphragm problem. The plate has stiffening elements which can act as chords. The out-of-plane loading results in a maximum deflection approximately 1/4 the plate thickness.

Is there precedent (or guidance?) for considering or neglecting interaction or combined stresses in this scenario? It feels like there should be some scenario where those effects interact (certainly if the out-of-plane deflection were larger), but to my knowledge it isn't really considered for typical diaphragms like a building floor. To consider the plate as an ordinary compressive element seems both disingenuous and very conservative when the behavior should be dominated by shear stiffness.
 
Replies continue below

Recommended for you

I'm moderately confused by the loading description without a sketch but the buckling question at least makes sense
There was some discussion in the post-EQ lessons here about consideration for out-of-plane buckling of diaphragms under compressive loads
This is touched on in presentations such as this though there isn't much content in the slides itself to give answers
 
Fair enough, here's a quick sketch. The out of plane loading is hydrostatic over the full plate.
Plate_Loading_jxwrub.png
 
This sounds like a plate version of a beam-column, so seems like the in-plane compression load will lead to further out of plane deformation, so needs to be considered.
 
If this is a steel plate or other ductile and isotropic material, I'd be inclined to consider something like a von Mises yield criterion.

If it's concrete...then probably a form of p-delta check.

Not sure about criteria to ignore it...probably need to play with it a bit to figure out the 'when it matters' line.
 
For concrete....I really wouldn't worry about it. For steel deck, very early in my career, I use to obsess over this kind of thing.....but after dozens of pages of checks (all while my older colleagues were laughing at me [smile]) on dozens of diaphragms.....I couldn't find one that warranted a combined stress check (or much worries about buckling). You had troubles with connectors and so on before that was a issue.

That being said, IIRC, seems like I remember SDI [the Steel Deck Institute] did have some sort of (min) recommended spacing for flexural members when deck is used (kind of like blocking recommendations in NDS on wood diaphragms). Not sure if they had this issue in mind though. I'll go through my SDI manual when I get a chance and see what I can find.
 
In all that follows, I have assumed that we're talking about a steel plate.

1) I don't love the comparison to building diaphragms. If you look into that space in detail, I suspect that you'll will find that the interaction effects are given credence to some degree, even if it's rough. Wide flange chords are typically designed as beam columns; concrete slab chord steel is typically designed as additive to that provided for other purposes... etc, etc.

2) I don't think that you need an established precedent to do what you're suggesting here. This can just be an engineering judgment thing. Loosely attempting to put some meat on the bones of that statement:

a) Work out the stress resulting from in plane diaphragm flexure. If it's crazy small relative to the out of plane bending stress, don't sweat it for combined stress checks.

b) At the location of the red dot shown below, estimate the degree to which your out of plane bending will be amplified by the presence of the compression field. Just a bare bones, [1/(1-P/Pe]] amplification. If that's tiny, set the interaction aside. If it's large, deal with it appropriately.

C01_w4kb87.png
 
Steel plate is correct here -- so while I'd be curious what SDI says, don't feel obligated.

KootK's #2 is the path we were thinking of going down (I once was savvy to the ways of Von Mises, but it's been a while). We'll end up bumping the plate thickness a bit, but that's OK. I'm no aerospace engineer.
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor