Differential Deflection of One-Way Concrete Floor Next to Rigid Wall / Shorter Span

[KootK]

Engi-Peers,

I'm designing a narrow building with a concrete core in the middle which will serve as the building's sole lateral system (see attached sketch). Originally, we had proposed framing the floors with beams and two-way slabs. In a effort to economize, we're considering using a 12", one-way composite deck system instead. Surprisingly, this system appears to be capable of spanning the entire short dimension of the building (9 m). My concerns are twofold:

1) Where the one way concrete abuts the core walls and runs parallel to them, there will be a differential deflection between the floor system and the walls of 1.5" (L/240) and;

2) Behind the core walls, there will be a shorter one-way spanning floor system. Differential deflection between the long span and short span floor systems will be about 1" where they abut.

How should one deal with this differential deflection? Can it be detailed away somehow? Can it be shown to be adequate by calculation? Am I over thinking this?

It seems to me that these problems must have already been resolved in the past for both precast hollow core and one-way pan-joist systems. While I couldn't find any relevant details or discussion for pan-joists, I found plenty for hollow core slab systems. Unfortunately, if the precast details addressed these issues at all, they did so only for initial dead load deflections (camber), not for live load or long term deflections.

With regard to this issue, PCI's Manual for the Design of Hollow Core Slabs says:

The effect of different vertical stiffnesses may be accounted for by:
1. Determining that distress will not affect the strength or performance of the system,
2. Locating vertically rigid connections near the slab supports where vertical movement is minimized, or
3. Providing allowance for vertical movement in the connection detail.

How have others been dealing with this? Part of my concern is that the "joist" members are closely spaced and no one joist is capable of supporting all that much load. If I say that the topping slab is able to span perpendicularly to the joists for a distance that justifies allowing the joists to "hang up" against the wall, that topping slab will be spanning to a joist member that likely cannot support the additional tributary load.

Thanks for your help.

Kootenay Kid

 

[slickdeals]

I am interested in what our engi-peers would come up with as this is strictly a detailing issue.

Is your drawing relatively to scale? Can you turn the direction of your floor system such that it spans left-right? You can then introduce beams running up-down at each corner and then rest the floor system on it.

 

[dik]

I would suggest that you have beams running from the core to the outside at the corners of the core. YOu need a stiff element to accommodate differential deflection, with the composite one-way action considered, it will most certainly crack at the corners of the core unless you introduce this stiff section.

Dik

 

[ron9876]

I assume that the slab and wall are not cast together in any way. I would put some type of support (shelf angle) at the deck parallel to the walls. That will help solve the differential deflection problem. Then you will need crack control steel at the four corners of the core which will help control cracking as the slab transitions from full length span to the supports at the core area.

 

[KootK]

To particular engi-peers:

@ Slickdeals:

1) The drawing is relatively to scale.
2) The span the other way is too long to do without introducing additional N-S floor beams. That would cause headroom headaches in this particular case.

@ dik: Thank you, I'll do that.

@ron: The slab could be poured between the walls, I haven't decided yet. Your solution with the angle is the most desirable outcome I think.

To all:

I'd like to solicit further discussion on point number #1 in my original post. For instance, if I go with Ron's shelf angle suggestion, how should I determine the floor load tributary to the angle? I suspect that the tributary load would be quite a bit more than the distance from the wall to the nearest rib.

Kootenay Kid

 

[msquared48]

Are you just going to use rebar in the center of the 5" slab for your drag links to the core shaft walls?

Contrary to your second answer to Slickdeals suggestions, I would still strongly recommend two composite W12 or W14 beams (5" topping slab) they would extend only 4 to 6 inches below the bottom of the 12" slab system. They copuld also serve as the drag links in the long direction. You sould still use the rebar drags in the short direction.

Mike McCann
MMC Engineering

http://mmcengineering.tripod.com

 

[KootK]

@msquared:

I will definitely run beams from the corners of the core as slick and dick have suggested. Spanning the deck from left to right is the only part of slick's recommendations that won't work in this instance. See the attached sketch for my current intent and a couple of options for the wall / slab connection detail.

I may use rebar in the topping slab as drag links. Alternately, if I can get the job done with shear friction only, perhaps I can use detail "A" without any explicit drag reinforcing. With that in mind, does it make sense to use detail "B" in the attached sketch? I prefer this to detail "A" but I suspect that detail "B" will prove more contractor friendly.

Thanks again.

Kootenay Kid

 

[msquared48]

OK. I would add extra transverse rebar, though, across the added concrete beams,snd onto the larger slab area (50% of the span length) to make the slab more two way in that area, decreasing the differential deflection problem.

Mike McCann
MMC Engineering

http://mmcengineering.tripod.com

 

[KootK]

@ MSSquared:

Sounds good. You're recommending rebar over the beams that has a length of 0.5 x beam length correct?

 

[msquared48]

No. The distance extending into the area beyond the added beams I would recommend as being half the distance from the added beam to the end of the building. I was trying to get extra reinforcing in the opposite direction to minimize the differential deflection in the slab you mentioned previously.

Mike McCann
MMC Engineering

http://mmcengineering.tripod.com

 

[hokie66]

Whether or not you analyze the system as spanning in one direction, it will try to span in two directions until it cracks. You must design for two way action at the corners of the core.

 

[BAretired]

I don't know anything about Versa Deck, but a 12" total depth spanning 9m looks too shallow to me.

BA

 

[KootK]

It is pretty shallow. The numbers work for strength and the deck supplier has some software for checking vibrations. So it's really just plain old deflection that I need to reconcile myself with.

I agree, the system will try to behave two-way locally. As far as designing for it, got any specific suggestions? Other than generic good detailing, I'm unsure how to account for the localized two-way behaviour without resorting to fancy stuff.