Max. Differential Settlement for Isolated Footings on Clay 1

In general - "ZERO", as uneven settlement will induce additional stresses on supports and connection members. If the site consists of a soft soil layer or a layer with uneven strength, you shall consider replacing the soft layer with layers of quality fill and improving/densifying the fill by rigorous compaction. After soil improvement, uneven settlement may still occur, but with a much smaller and negligible magnitude. The most important thing is to consult with your geotechnical engineer for site preparation.

I don't believe it is in ACI 318 (although Ive never cracked the -14 edition).

The amount of differential settlement allowable depends heavily on the details of the structure being supported. It usually breaks down to how much angular distortion can be handled, more for a doctor steel structure with metal siding and less for brittle brick.

You can look to Skempton (1956) for his chart, or IS 1904 has some guidance for different construction styles and foundation types.

----
just call me Lo.

You might try the US Army Corps of Engineers 'Engineer Report No.1110-1-1904: Settlement Analysis'. It includes a section on suggested limits. However, it's mostly a matter of judgement. How much can the frame take? Or how much does the geotech expect and can you design the frame to take it?


Zero settlement - differential or otherwise - is not a plausible goal even with an unlimited budget.

@phamENG,

Uniform settlement is something your geotechnical engineer can provide guidance on, not the "uneven" settlement/differential though. "ZERO" here simply means there is no guideline on it without tedious/endless engineering analyses/evaluations, which, as I am aware, nobody wants to perform even for a simple structure.

le99 - your comment reads that you are recommending ZERO as the amount of settlement permitted.

@EireChch,

Please provide a guide/limit on differential settlement, which is the OP's question.

Most of my analyses have been to limit total settlement to around 1 inch, and differential settlement to around 1/2 inch. With vertical water tanks, we always recommend they be filled in stages and monitored to make sure they don't start to tilt, before they are placed in regular service.

Typically, the structural engineers design for some minor amount of flex within the structure.

I believe, in general, as long as sizing the spread footings strictly follows the guidelines in the geotechnical report, the differential settlement will always occur but will be kept within the tolerable/acceptable level, from both the viewpoints of the soil engineering and structural engineering. Thus, it is essentially a non-issue if there is a soil report, that is based on the competent subgrade investigation. Under this premise, a rigid number/limit is likely to add an unnecessary burden to the structural engineer rather than intended help.

AS pointed out by @Tigerguy, the total settlement given by the Geotech implies an anticipated "differential settlement" of half of the given magnitude, or worst of, the total. But I doubt any of the Geotech will say it is the limit of "differential settlement".

Yes and no.

Sure, there is no physical law that limits settlement in the same way that the capacity of a beam is limited. Some structures handle very little settlement without negative consequence, some are more flexible.

But having a set of expectations about what will be reasonable for a given structural system without requiring onerous extra material, detailing or analysis is very helpful.

And when geotechnical capacity is governed by settlement (many shallow foundations), there needs to be an understanding communicated between the geotech and structural of what approximate settlements can be accomodated. Ideally, that is communicated by direct discussion for each project, but more realistically it is communicated by traditional practice and rule of thumb assumptions in the geotech report.

----
just call me Lo.

lee99 - to specifically answer the OPs question, NO ACI-318-14 does not include differential settlement recommendation for footings on CLAY.

Tiger Guy, that must be for relatively small diameter tanks? For large tanks, which are typically flexible structures, the settlement can be hundreds of mm.

I've analyzed numerous grain bin structures for settlement on deep clays. Settlement estimates of 6 to 8 inches (150 - 200 mm) are quite common. On large diameter tanks and bins, the base is flexible, with nearly double the settlement in the center as observed at the edges.

By filling in stages, the settlement takes place at a slower rate. Some flexibility in the connections is added to the design to accommodate this movement.

The real trick is to make sure the base doesn't flex so far that the connection to the sidewalls is compromised. We covered a case study in university on one of these, it made a mess. When I've encountered sites this soft, I usually recommend the placement of some thickness of aggregate base beneath the tank floor. Just like a pavement design, add a stiff layer and improve performance.