Wall Footing Eccentricity

[design62713]

I am trying to find out which section in ACI-318 talks about L/6 eccentricity limit for designing wall footings? Thanks in advance for help.

 

[mike20793]

It's not a code requirement, but rather a general recommendation that you keep you're eccentricity within the middle third (kern area) of the footing. This ensures that you have compressive soil stress under the entire width of the footing. The force resultant can lie outside of the middle third, but it may require special attention at the end of the footing where there is no compressive stress on the soil.

 

[design62713]

Got it. Thanks mike20793

 

[KootK]

Is your situation a property line strip footing at the bottom of a basement wall? If so, common practice is to assume a uniform soil pressure beneath the footing and resolve the eccentricity via a moment connection to the basement wall. The predominant issue is the design and detailing of the joint.

I feel like the kern business gets too much air play. It often results in unreasonably sized foundations.

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.

 

[hokie66]

KootK may be correct that the assumption of uniform soil pressure is common, but that doesn't make it right. Reinforcing the joint for the moment does not change the soil reaction.

 

[KootK]

The soil pressure doesn't need to be "right" with a bearing medium governed by either plastic behaviour or settlement concerns. Besides, it's dirt. Whether you go with a retangle, a triangle, a cycloid, or even FEM results, it's a wild ass guess at best. Might as well make it easy.

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.

 

[hokie66]

Somehow, I knew that was coming.

 

[design62713]

KootK, you are sketch is correct.

 

[KootK]

Somehow, I knew that was coming.

By now, you know full well that I'm incapable of self restraint in these situations. Can you share what your assumption regarding soil pressure would be in this situation? I really do want to know. In my experience, no reasonable footing size will work for basement property line footings unless a moment connection to the walls is employed. And, once the moment connection is thrown into the mix, I really have no idea what the "right" soil pressure distribution ought to be.

KootK, you are sketch is correct.

In my opinion, crap detailing is a much bigger issue in these situations than is correct assumptions regarding soil pressure distribution. I see the detail shown below about once a month on average and it fuels a lot of my binge scotch consumption. In my opinion, the key features of a proper design include:

1) One piece rebar crossing the footing and turning up into the wall.
2) Depth and rebar bend diameter chosen such that the last strut coming into the rebar corner could conceivably pass a strut and tie check. Usually, the proportions of these things are such that sectional shear check are questionable.
3) Some dirt side wall reinforcement continued to the nearest floor diaphragm above. I don't like to rely on lateral earth pressure that may not exist to reduce cut-off lengths.

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.

 

[hokie66]

KootK,
It is just simple statics, you know that. If the footing is the same width as the wall, the pressure is uniform. If the footing is wider and moment connected to the wall, the reaction centroid moves over, with the soil pressure becoming trapezoidal, then triangular.

 

[KootK]

It is just simple statics, you know that.

Hardly. Rather, it's extremely complex, pretty much unknowable, indeterminate statics. The soil stress distribution would depend on the interplay between:

1) The flexural stiffness of the footng.

2) The flexural stiffness of the wall.

3) The vertical stiffness of the soil below the footing.

4) The lateral stiffness of the soil behind the wall.

5) Whatever flexibility may result from the presence of drainage mat etc behind the wall.

If the footing is wider and moment connected to the wall, the reaction centroid moves over, with the soil pressure becoming trapezoidal, then triangular.

Sure. But what good is this if you never really have any idea where your soil stress "neutral axis" is located? Under the wall? 6" inside? Philadelphia?

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.

 

[msquared48]

If the slab and footing are poured monolithically, you can include top bars that extend back into the slab to counter the eccentric loading and alter the soil pressure seen.

Mike McCann, PE, SE (WA)

 

[KootK]

If the slab and footing are poured monolithically, you can include top bars that extend back into the slab to counter the eccentric loading and alter the soil pressure seen.

This is common but, frankly, also bothers me. It strikes as a gross stiffness incompatibility in most cases. People wind up relying on the flexural stiffness of a 5" slab on grade with rebar at mid-depth (d=2.5") to iron out eccentric soil stress beneath the footing. I'll be curious to here what other's think about it.

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.

 

[hokie66]

KootK,
The unknowns you listed are typical of essentially all structural engineering problems. We use models and methods, which are never precise, and don't have to be. But I will use mine, you are free to use yours, but not on my job. The assumption of uniform pressure is anathema to my way of thinking.

 

[KootK]

The unknowns you listed are typical of essentially all structural engineering problems. We use models and methods, which are never precise, and don't have to be...The assumption of uniform pressure is anathema to my way of thinking.

You can't have it both ways hokie66. On the one hand, you tell me that my uniform distribution is "anathama" to you because the soil distribution is not "right". Then, on the other hand, you tell me that there's no justification for accuracy in knowing the actual soil stress. Which is it?

And you still haven't indicated what soil stress distribution you would use or how you would determine it. Simply stating that the reaction centroid "depends" isn't saying much. If it's just simple statics, do tell. The method that you described above is only simple if you choose not to rely on the moment connection with the wall. And that will generally make for an unreasonable / impossible situation for a footing like the one being considered here.

It's also worth noting that, even with concentric footings, the assumption of uniform soil stress is also fiction. My Teng foundation book contains a figure much like that shown below.

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.

 

[hokie66]

I'll just say that I approach this in the same way I approach any cantilevered retaining wall analysis.

 

[msquared48]

I designed a eccentric monolithic footing with similar reinforcing into an 8" slab supporting a two story tilt-up concrete wall on the edge of a down slope to a railway line over 25 years ago and it has shown no distress, inside or outside of the structure. Somehow, it works - perhaps by extending the soil bearing prism into the slab area. Kinda like an "L" shaped retaining wall reinforcing pattern to spread the load.

I know that I sound like the proverbial contractor, but I am not. Trust me.

Mike McCann, PE, SE (WA)

 

[KootK]

Let's not kid ourselves: everything seems to work out in the wild. Even the shabby, ill-conceived detail that I posted above is working just fine. Geotechnical safety factors being what they are, it's entirely plausible that none of our "eccentric" strip footings really need to be any wider than the walls that they support. I will concede, of course, that an 8" slab on grade with a fighting chance of having legitimate top steel is a good deal better than the 5" SOG scenario that I described above.

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.

 

[civeng80]

Would tend to agree with hokkie on this one.

Also what if the wall is precast ?

What about tieing the ground slab to the footing ?

 

[msquared48]

My wall was precast.

Mike McCann, PE, SE (WA)