Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations GregLocock on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Giant "donut" footing bearing capacity and settlemen

Status
Not open for further replies.

rplayer

Geotechnical
Jun 21, 2005
2
I am faced with a unique foundation arrangement and would appreciate guidance on how to analyze:

Location: Wastewater treatment plant

Project: Adding two large egg-shaped digesters and associated piping galleries. Plan view looks like a giant pair of Elton John eyeglasses - two round lenses with a large rectangle on top.

Soils: Site soils consist of medium dense to dense sand and gravel (<5% passing No.200) with occasional cobbles and boulders (Lake Missoula outwash gravels) N60 average = 25.

Existing grade is at elev 1710 feet. Top of floor slab will be at 1680 feet. An excavation 300 ft xt 200 ft will be opened up to build digesters and surrounding galleries. Outside edges will be backfilled to ~elev 1695, but inside will be open.

Digesters to be supported at their equators by large circular concrete stub walls that in turn are supported by 20-ft wide donut shaped spread footings. The stub wall diameter is 80-ft, so the diameter of the outside edge of the footing is 100 feet and inner diameter is 60 ft. Footings will be buried 5' 8' below finish grade.

The footings for the two digesters are 20-ft apart at their closest point. In addition, a 10x10 foot square column footing will go between them, leaving a 5-ft space between adjacent footing elements.

Required bearing capacity of the "donut" foundation is 8000 psf (big, liquid filled vessel).

The question is how to calculate bearing capacity (8 ksf should be doable) and settlement (<1 inch tolerable). Do the footing loads superimpose? How apart do they need to be to not superimpose?

What are some good references for complex, closely spaced footings? For "donut" foundations?

(I refer to them as donut foundations because they are so large, they don't seem to meet the ringwall criteria.)
 
Replies continue below

Recommended for you

For bearing capacity, I would suggest analyzing the donut foundations as 20-foot wide strip footings. This is obviously not the case, but it would be conservative.

For settlements, is the 1 inch criteria for immediate and long term settlements, or just long term settlement? In sand and gravel, the majority of your settlement will be immediate. Although, now that I think about it some more, most of your load comes from filling the tank, so immediate settlement would be a concern. Well, in that case, yes the footing loads are going to superimpose. You might be able to use a Bousinesq distribution for a circular footing. Calculate the influence factor for a 100-foot diameter footing and then subtract the influence of the 60-foot interior (the part of the circular footing that is not present). You should also take into account the superimposed loads from the adjacent footing. This will take quite a bit of number crunching.
 
I worked on similar foundations for LPG spheres. Annular is the donut word you are looking for. Cannot remember all the design details but they were modelled as strip footings and designed to span a 5m "soft spot".

Zambo
 
rplayer,

It seems to me that if you want to model the stress increase below the annular foundation, that you could use the Boussinesq formula for stress below a circular foundation on an elastic medium and superimpose a "negative" load using a smaller diameter circle in order to account for the actual contact area with the soil.

You would need to perform some integration to work out the stresses at multiple points relative to the radius of the circle.

After that point it should be conceptually easy to superimpose stresses due to the square footing to find the maximum applied stress and your required BC. Moreover, once you know the applied stress distribution, you could estimate the elastic settlements using a discretized soil column.

Note that the empirical relations between SPT N and Young's modulus for granular soils, as published in Bowles (1982), have been conservative in my experience for river-placed sands. If the budget is robust enough, dilatometer testing could probably dial the Es values in closer to actual values.

Good Luck!

Jeff


Jeffrey T. Donville, PE
TTL Associates, Inc.
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor