Determing Uplift Resistance of Interior Residential Foundations 3

[egc]

We have an ongoing debate around how to determine the available uplift capacity of interior residential foundations (in Florida). For example, assume you have a 6' long interior bearing wall (2x4 and well away from other foundation elements) with a combined truss uplift of 6000 lbs. Neglecting wall dead load, some argue (me) that we need to develop 6000 / 0.6 or 10000 lbs of concrete foundation and tributary slab to resist this force. Others argue that if we provide a mono footing to handle the gravity loads, that the footings and slab and building contents will be enough to resist the applied uplift. They also point out that they not aware of any structural failures of any interior mono-foundation failures during hurricane wind events.

Comments (or better, code citations)?

 

[JAE]

Two comments -

Higher wind pressures occur on the eaves and corners so I would imagine that the weak link in the load path would first fail in those regions.

I just observed a foundation settlement condition recently - this is not a wind issue but the failure mode was such that the dropped footing tried to pull the wood stud wall down with it. The trusses were hung up on interior walls so the truss was essentially pulling up on the wall. The failure mode was that the top plate to stud nailing simply pulled out of the studs (no tensile capacity for withdrawal parallel to grain) and the edge nailing on the sheating - both gypsum board and OSB both pulled through the free edge of the sheathing, separating the wall studs and sheathing from the top plate.

So with this second point, you can count on footings, slab, etc. all you want, but the top plate to stud connection probably couldn't develop it.

 

[egc]

Okay, but that really wasn't my question. We have the load path from truss to foundation covered. It's the "what happens then" that is the issue.

 

[boo1]

I design the interior load bearing walls to react the truss loads. Typically these grade beams are 8" deep x 16" wide with two #5 grade 40 rebar. The load path is truss to ground. The overturn moment (summation of moments) should be calculated, but often is omitted for one story homes.

 

[DaveAtkins]

I think if you have enough dead weight of footing to resist the uplift minus 0.6 times the superimposed dead load, you are okay. In other words, I think it is conservative to multiply the weight of the footing times 0.6. I think the 0.6 accounts for the fact that we engineers overestimate dead load of roof, floor, wall, etc. But the weight of the footing can be determined fairly accurately.

DaveAtkins

 

[egc]

Great comments guys. Thanks.

Dave,

FYI the 0.6 factor is straight from ASCE 7-XX. This code(s) does not classify dead loads by "reliability". We would love to remove the 0.6 factor but can't find the "code" justification.

Boo1,

"Truss to Ground"?

Well, what anchors the footing/slab down to the ground?

In the example case, the footing (gravity load sized) is 12" W x 16" D Mono (10' long centered under the wall segment). The slab is a 3.5" thick with fibermesh (poly).

What slab perimeter is resonable to assume contributing to the uplift resistance?

 

[UcfSE]

I would say you design the footing for the net uplift, W-0.6D. If that comes out to 6000 lbs then so be it. If you want to count slab weight then the slab needs to be checked for the uplift also, bending and shear.

I can't stand the argument that "it won't fail that way so we don't have to be thorough or design it right". If you design it right and leave the footing too small then it will fail that way.

 

[boo1]

Umm, typical FL slabs are lined with 6 mil poly so there would be little soil interaction. Would you consider the grade beam cone shearing out of the slab?

Never have I seen this addressed.

 

[egc]

Okay.

UcfSE,

If it fails (in the classical sense), then you didn't design it right <). So how would you analyze this condition if you wanted to include the slab?

boo1,

That's right on target and getting close to addressing my question. I have searched the web high/low for anything discussing this issue (and come up with zero). I have found discussions (including here) addressing eccentric moments for metal bldg perimeter footings.

What would be the general failure mode: two-way shear, flexural, etc.? Would membrane arching action extend the "effective slab area" involved until the slab/footer moved upward passed a certain point? From an engineering point of view, slab deflection is acceptable.

 

[bylar]

I am surprised at some of the answers.

The story that they never seen failure at the interior wall seem like builder’s optimism. Who determined how the house failed when all is left is ruble.
I visited Miami after Andrew and saw several commercial buildings with steel columns supporting large areas of roof and it was obvious that the slab lifted and cracked, some separated from the rest of the slab and then settled back down once their roof sheathing blew off.

You’re the designer provide enough resistance to the uplift. If you use the slab as part of that resistance then design the slab to cantilever over the footing. You will probably find that a larger footer is cheaper than trying to design the slab to take some of the load.

 

[apsix]

If the footing is part of a slab, with a vapour barrier under, I do not include any soil interaction (for obvious reasons).
If it is an isolated pad or pier (and no barrier) I include the weight of soil within a 1horz:2vertical 'cone' extending from the bottom edge of the footing.
The 0.6 dead load factor seems pessimistic, in Aust. our codes allow 0.9 which I admit can be optimistic. I downgrade it to 0.8 if it worries me. (This is for limit state design)

 

[STR04]

I would take one step back and look at your uplift pressures. What is the total tributary area for uplift? Are you using CC or MWFRS pressures? How are these trusses framing into the vertical 2x4 wall element...truss to beam to wall? What is the truss span?

 

[egc]

Bylar,

While I work at the same company, I am not the EOR of these buildings and therefore can not "dictate" design methodologies. I am only serving as a technical consultant (my main responsibility is construction plan automation).

I too visted the Dade County area after Andrew as well as Hugo (SC/NC). I do not remember seeing "interior slab" failures from wind uplift (granted, at the beaches in SC/NC it would be impossible to tell).

Regardless, I am still trying to find some quidance on reasonable parameters to utilize the slab. At this point I am telling the others that the professional "on-line" consensus is "the resistance has to be quantifiable" and suggest they do not use the slab weight beyond 1' past the thickened slab edge.

Apsix,

Okay but see previous posts, i.e, monolithic in sand, -> no soil participation. And we in the USA are stuck with the 0.6 factor.

Str04,

The reactions are MWFRS (but the trusses are designed for C-C). I assume your comments about tributary area are therefore void.

Thanks to everyone who commented. I am moving on to the really exciting stuff (like bringing Mitek Truss Profiles into ADT) <).