Mezzanine inside Tilt up Building - Lateral Stability (Seismic)

[Toby43]

Hi all,

It is proposed to have a storage mezzanine built into an existing single storey tilt panel building. It only partially covers the plan area of a unit and its attachment to the existing building is to 2 parallel tilt walls (at 3m elevation of 6m tall panels), which I'm happy to utilize for seismic forces in the walls in-plane direction. As proposed, only 1 line of walls/bracing is supplied perp. to existing walls, leaving an open-ended diaphragm - this leads to a demand on the new framed wall that can't be met without steel bracing and new footings cut into slab. The other alternative is to place the demand on the existing walls in out-of-plane bending between the ground floor slab and the roof bracing.
This is something I don't feel entirely comfortable with because
1) The existing walls are 150 thick, centrally reinforced with Low ductility mesh which does not meet minimum R/F requirements (all too common here in Australia - for some reason we chose to "cherry pick" ACI provisions)
2) Displacement incompatibility with roof diaphragm/bracing, which could create larger anchorage forces (indeed this may be an issue even if mezzanine can laterally support itself but is anchored to existing walls.)
3)Calculations suggest wall can cope with "elastic" design level forces (without any reduction factor), yet capacity of wall to cope with seismic forces beyond design level would be limited/brittle and mid-height hinging would in my opinion be detrimental to wall stability and the integrity of the mezzanine anchorage (failure of mezzanine gravity support).
4) Don't like the idea of "leaning" heavily loaded beams and columns on the out-of-plane stiifness/strength of panels.

Was hoping for others opinions, experiences with this structural configuration and if my worries utilizing the out-of-plane capacity of the walls are unfounded?


Thanks
Toby

 

[JAE]

I think all of your concerns are valid concerns.

Perhaps in this case, separating the mezzanine from the tilt walls would be best - i.e. an expansion joint along the mezzanine/wall edge with separation adequate for out-of-phase seismic deflections between the mezzanine and wall....square root of sum of squares. The mezzanine, then, would need its own bracing system independent of the main building.



Check out Eng-Tips Forum's Policies here:
faq731-376

 

[DETstru]

In the direction where you have two existing walls resisting seismic loads in-plane, you're all set like you said.

In the other direction:

Option 1:
Connect the diaphragm to all existing walls and design as a 3-sided box. This has complications and with existing walls I'd say it has little chance of working.

Option 2:
Connect the diaphragm to the existing wall and add a brace on the open side. This is really easy.

Option 3:
Don't connect the diaphragm to the existing wall (but still connect it to the two walls in the other direction).
Those two walls are effectively collector elements that act out-of-plane and send the load to the roof diaphragm and then the Seismic Load Resisting System. Those walls acting out-of-plane are not the Seismic Load Resisting System alone.
They must remain elastic and be designed for overstrength as a collector element. As they are existing walls, they may not have capacity and you're probably forced into Option 2.

 

[KootK]

For the purpose of what follows, I'll assume that isolating the mezzanine and providing a brace to the open side are options that have been evaluated and deemed unpalatable owing to programming and/or cost concerns. Obviously, both of those alternatives are pretty wonderful structurally.

Sticking with that assumption, I think that the key here is to recognize that, in the "bad" direction, you need some kind of system that will compete effectively in terms of stiffness with out of plane wall bending. You want to be able to say that the walls are leaning on mezzanine to some degree, rather than the other way around. Some options for that include:

1) Assuming that you've got a nice rigid mezzanine diaphragm, I actually think that the three sided box concept has legs. Much depends on your proportions in plan but I think this concept has real merit, particularly if you head into it with the intention of shielding the walls from demand.

2) Toss in some moment frames in the crap direction. One bracing line... or all of them. Whatever it takes if the project can afford this.

3) Use some beefy columns and run them straight up to the roof. This would have programming impacts of course.

4) Do #3 but only up against the wall that you're trying to shield from out of plane bending. This would have little impact on programming or cost. The columns just need to be stiffer than the wall.

It's cringe worthy seismically but one could even argue that, with these systems, you could allow the wall to fail in flexure at which point it would become two span and relying on the mezzanine for stability.



I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[Toby43]

Thanks all for your input.

I have managed to weasel in some bracing to the open side, so will enjoy my weekend that little bit more. Also existing walls turned out to be only 125 thick

I would however be interested in others opinions on the deflection incompatibility of the roof diaphragm at 6m elevation that spans full length of building with the mezzanine diaphragm at 3m elevation that only spans partial length of building. The walls adjacent to one another at the mezzanine termination point would have quite different deflection profiles out-of-plane, which I feel would cause distress to the panel or joint at mezzanine termination. My main concern is the effect it would have on the mezzanine anchorage (same reason panel "hinging" at mezzanine elevation makes me cringe also). Not convinced if low ductility mesh would make a good hinge anyhow - which brings me to another query -
As stated in my original post, here in Australia many have seemed to "cherry pick" ACI provisions regarding "slender wall" design (AS3600 is limited at best regarding out-of-plane design of slender walls), which has led to many walls not meeting minimum Reinforcement requirements (i.e. Design Moment Capacity >1.2x Cracking Moment). Indeed one of the best selling guides for design to AS3600 questions the need for minimum reinforcement, which likely only propagates the issue. In correspondence with a member of the "ACI slender wall task committee", he was adamant such a requirement must be met. On the other hand, if my memory serves me correctly, does the Canadian Code have something of the effect whereby if the moment capacity>1.33xapplied moment, then min reinforcement can be waived? Lacks robustness if you ask me.

Thanks again
Toby

 

[KootK]

Yes, Canada and ACI both have such a provision. Other things being equal, I believe that more lightly reinforced walls are more ductile. The displacement compatibility piece is tricky and favours JAE's isolation strategy.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[strucEng36]

does the Canadian Code have something of the effect whereby if the moment capacity>1.33xapplied moment, then min reinforcement can be waived? Lacks robustness if you ask me.

can you please advise where is this provision in CSA standard