Foundation capacities from combining footings and slabs

[chezzy6]

I am trying to replicate an old design from a seasoned engineer, but he is no longer around to ask questions, so I am looking for some opinions. In his previous design, he utilized a combination of piers/footings and slabs for the foundation design; where, in my opinion, they would not be sufficient individually. In reality, I know there is strength between the interaction, but my questions below are about how to quantify the combination on paper.

Typically, in an industrial setting, I will design the piers and footings for the pre-engineered building columns to be independent of the rest of the structure because throughout the life of the structure, it changes so much that it is not always wise to rely on other parts of the foundation to resist forces such as uplift from wind etc.

However, I feel there are situations that this is a little too conservative. For instance, in a large storage building, the use likely won't ever change, so I have the opportunity to tie the piers/footings into the building slab to help resist forces. Slabs vary from 6,8,12" with single or double layers of #5 rebar.

My question is this: if you pour the footing then pour the pier, building perimeter, and slab monolithically, how do you accurately account for the combined strength of the pier/footing and slab?

For uplift, in my mind, you would take the difference between you uplift forces and your foundation weight and the remaining forces would need to be resisted by the slab. Therefore, would you just consider the slab a "beam" on a per foot basis, multiply it by the circumference of the footing and that is your shear and moment capacity of the slab to contribute toward resistance?

For bearing, can you realistically assume the interaction between the two can combine for a larger bearing area? In theory, it would be nice to reduce the size of the footing if your slab can assist in creating a larger bearing area to assist in low allowable bearing pressure situations. However, I am skeptical of this because of the difference in bearing elevation between the bottom of slab and bottom of footing; especially when you factor in the ability to properly compact both areas identically.

As for lateral loading, I don't see many issues as the buildings are typically very large, so the weight of all the concrete is more than enough to resist the lateral kickouts of the [rigid frame] buildings.


Assume load combinations, rebar development length, all the other factors are taken into account.

 

[SlideRuleEra]

1) Typically, in an industrial setting... throughout the life of the structure, it changes so much that it is not always wise to rely on other parts of the foundation to resist forces such as uplift from wind etc.

2) However, I feel there are situations that this is a little too conservative. For instance, in a large storage building, the use likely won't ever change...

1) Good observation.

2) IMHO, don't just decided this on your own, talk to the Owner.
At our generating stations, new technology for improvements that did not exist when the units were designed and built requires "unusual" adaptations of existing structures.

 

[milkshakelake]

I wouldn't use the entire circumference for uplift resistance, because that's assuming that the whole thing is infinitely rigidly connected. I wouldn't go so far as to do an analysis on it, but maybe take 1 or 2 bays into account instead of the whole thing.

When I did industrial plants and power facilities, for ~50% of them, we only designed it for the loads at the time. Of course the use changes over time. Some owners would want something conservative so they could easily adapt to it. Most wanted the lowest cost option. This should be based on the owner's decision. The technical details, like whether to count each footing in isolation for uplift, is up to you, but should be informed using the owner's overall business goals.

There's one thing about uplift that I do to help reduce it. There's typically a load case where dead load is reduced for uplift (something like 0.6D+1.0W). For that case, I still consider the foundation weight to be 1.0D, because it's very unlikely to be reduced. Whereas the rest of the structure could have a floor deck removed, or something, so the 0.6 reduction applies to that. This isn't strictly covered in the code, but just some food for thought.

 

[Deker]

There's one thing about uplift that I do to help reduce it. There's typically a load case where dead load is reduced for uplift (something like 0.6D+1.0W). For that case, I still consider the foundation weight to be 1.0D, because it's very unlikely to be reduced. Whereas the rest of the structure could have a floor deck removed, or something, so the 0.6 reduction applies to that. This isn't strictly covered in the code, but just some food for thought.

It's a common misconception that the 0.6 factor on dead load has to do our inclination to conservatively overestimate the dead load used in design. In reality, the 0.6 factor was calibrated to produce a probability of failure that is equivalent to an LRFD design. Failure to consider the 0.6 factor on all dead used to resist uplift is a code violation, so tread lightly. See the attached statement from SEAOC on this issue as it relates to ballasted solar arrays.

 

[milkshakelake]

@Deker Didn't know that. Thanks for pointing it out.

 

[Lomarandil]

Interesting, thanks Deker. I understand SEAOC's position that the 0.6 dead load factor is not intended only to represent uncertainty in dead load. However, I have heard that part of that reduction (10%, like the old 0.9DL combinations) was due to uncertainty.

I'd really love to see that addressed specifically by the code writers.

 

[chezzy6]

I appreciate the responses. To delve deeper into the question, would treating the slab as a beam and utilizing its moment and shear capacities truly be representative of the resistance it can contribute to bearing and uplift? Are the two elements' capacities allowed to be superimposed for resistance?

In ACI 318-14, there is some verbiage in chapter 8.4 about interaction between concrete columns and slabs with/without drop panels. It is more for transferring slab moment to the column, but in my opinion, it should work both directions. I feel this can be applicable to slabs on grade connecting to footers using the same methodology: 1.5h (slab thickness) on each side of the beam (this case pier). This would allow for 1.5h + 1.5h + pier width on each side to resist moment utilizing the slab. That seems reasonable. Has anyone used this methodology in a similar design?

 

[dold]

We use the slab to resist uplift all the time. In the most conservative calculation you could just account for the weight of the soil and slab that are directly over the footing. Essentially saying the slab has zero shear capacity (hence, conservative). In practice, we have some tables/design guides that give calculated capacities for uplift resistance of the maximum 'chunk' of slab that would be mobilized to resist uplift, assuming it is poured monolithic with the footing (common on the east coast of the US since the footings are typically just a thickened portion of the slab). This takes in to account the ability of the slab to 'cantilever' out beyond the edge of the footing. To be honest I'm not sure who developed the calculations but it's what we use. But they're not too far off from what one might get from a pretty basic hand calc.

Re: 0.6DL vs 1.0DL for the footing self weight - I've experimented with this but ultimately I've been told that the 0.6DL + 0.6WL combo has a baked-in safety factor of 1.5 for uplift for the entire system. That said, it's always seemed pretty conservative to me to use 0.6DL for the footing self weight when looking at uplift due to direct suction. Seems reasonable when looking at uplift resulting from overturning of a braced frame, etc.

 

[chezzy6]

@dold

can you send me a reference to those design guides/tables?

 

[dold]

@chezzy

Unfortunately I don't have a digital copy of any of those, and I'm not sure they would exist on the web. They were paper copies that my boss distributed to us years ago and I'm no longer with that firm. And they looked to be type-written from the 70's.

Edit to add: These were specifically for designing foundations for PEMBs.

 

[MotorCity]

We never rely on the slab to resist uplift of a column, mostly because:

1. The slab is unreinforced and has negligible flexural capacity
2. You would need to tie the slab to the pier or foundation to engage any appreciable contribution to uplift resistance beyond the weight of the slab. In most cases, we avoid tying the slab to the structure because of the potential for issues created by drying shrinkage of the slab and differential movement between the slab and building.
3. The slab is usually poured sometime after the the columns of an enclosed building have most likely had their first exposure to uplift.

 

[wth]

I usually only account for the self weight of soil and the slab directly above the foundations, as the floor often is a "free floating", slab with one layer of rebar here.

I've done the slab with thicker edges once at request from the client due to site conditions.
If I recall correctly, we designed the edge as a RC beam on a spring support with no uplift capacity, and the slab as s cantilever with the necessary lenght to mobilize enough self weight to resist uplift. We also made the edge heavy enough to get a reasonable slab area.

In our code the governing load case for PEMB foundations is mostly 0.9xDL+1.5xWL.

 

[Alistair_Heaton]

Just gone against local norms with a foundation ( its a DIY build by me)

In Scotland we often use float slab ie we put footings in with a 45 deg up to the slab which has grid and link to the footings with hooks. And do a mono pour of the whole lot. Locally they do a belt foundation and then fill the floor in later.

I fired in additional rebar to point loads. as per the upper floor and the locals tend to go for a generic linear loading plan.

Its completely messing with the local design codes. As every effort they come up with to go back to soviet methods is hit with its a shear diaphragm plus foundation. If the thing moves the whole lot is going to move together.

Do you have anything to prove that? Yep you could hang 1/3rd of the building in free space and the slab isn't going to fail. Here is the numbers. Err it against local design codes. Its not actually its just you don't have codes for what the rest of the world has been doing for the last 50 years. Here is the UK as its a selfdesign and build i am running a 175% safety factor...

Err i really would advise C30 black iron rebar belt foundations with a post pour slab floor... Its in, deal with it the floor is good for 10 tons per m2 I ain't ripping it out....