Shear Wall Buckling

[RFreund]

I feel like I bring this up every couple years, but that's probably because I haven't found an answer I'm happy with. This is not an ideal situation, but how would you check a concrete or masonry shear wall for buckling. Imagine a situation these two situations:

An inteiror shear wall with openings on both sides. It may or may not have return walls.
An exterior wall, say a stair wall that again has no floor diaphragm to brace the wall. The end walls may or may not be concrete/masonry. See image below:

EIT

http://www.HowToEngineer.com

 

[WARose]

I had one like that some years ago: it was being loaded like a shear wall but what was loading it wasn’t really bracing it (out-of-plane) to my satisfaction. As a result, I checked it like a cantilever beam (i.e. as per the various buckling modes of a beam: LTB and so forth). And it checked out.

Indeed, the section of ACI 318-11 that addresses the required bracing spacing for flexural members says this in the commentary:

R10.4 — Distance between lateral supports
of flexural members

Tests^10.10,10.11 have shown that laterally unbraced reinforced
concrete beams of any reasonable dimensions, even when
very deep and narrow, will not fail prematurely by lateral
buckling provided the beams are loaded without lateral
eccentricity that causes torsion.

There is (of course) axial load to consider as well (and it's corresponding buckling mode).

 

[KootK]

I've been interested in this topic as well. Here's what I think I know:

1) Certainly, in high seismic events, buckling of the compressed zones of cyclically loaded shear walls has been a problem.

2) A lot of high rise buildings will have very tall lower stories where this tends to be an issue unless there's an abrupt change in wall thickness at those stories. And there usually isn't.

3) MacGregor's concrete text contains an interesting, first principles buckling check on a shear wall compression zone. If I recall correctly, I believe that he recommends h/t around 15 for high seismic. And I'm pretty sure there's something in ACI's seismic chapter limiting that ratio as well.

4) Some codes, including Canada's, enforce h/t <= 25. I've been enforcing this in at least one direction, either horizontally or vertically.

 

[WARose]

3) MacGregor's concrete text contains an interesting, first principles buckling check on a shear wall compression zone. If I recall correctly, I believe that he recommends h/t around 15 for high seismic. And I'm pretty sure there's something in ACI's seismic chapter limiting that ratio as well.

4) Some codes, including Canada's, enforce h/t <= 25. I've been enforcing this in at least one direction, either horizontally or vertically.

That's a good point. You'd certainly want to limit slenderness so that stresses beyond the elastic can develop in a seismic situation.

 

[Deker]

Check out Chapter 3 and Appendix B of this report (Link) for derivation of a method that applies to high seismic. I haven't done it, but perhaps you could modify equations 3-1 through 3-3 to apply to a non-yielded wall where seismic isn't a concern and you'd like to try an alternative approach to the h/25 limit in ACI 318-14 Section 11.3.

 

[RFreund]

Thanks for replies.

As far as h/t goes - for the interior case, assuming no return walls - Would you consider the connection to the floor slab (almost like a point support) as bracing the wall?

EIT

http://www.HowToEngineer.com

 

[jayrod12]

From just a wall stability standpoint under essentially self weight only, yes I would, however for shear wall level loading, no. I have this same situation on one of my projects (it's always the central wall in the stair core) and I neglected it's contribution to the lateral system when designing the main force resisting systems. Then I re-ran the model including the wall and designed that wall's reinforcing specifically for the amount of load it was attracting in the model just to cover myself.

I've also specified that the gap between the pre-cast stair runs and the walls be continuously grouted so I've actually got more stability bracing than I used in my design.

 

[KootK]

Would you consider the connection to the floor slab (almost like a point support) as bracing the wall?

I would. Obviously, that implies a reasonable degree of connection robustness however. More than a Simpson A35 clip. Some reasons for my recommendation:

1) You need some bracing for gravity load stability anyhow and I don't see shear wall compression edge stability demand as being that much greater. More so with heavy seismic of course. So your might as well have your shear wall back.

2) In any ULS lateral event, the last thing that I'd want is a "non-structural" wall caving into an egress route.

As you can see, opinions vary...