Oil Tank Shallow Foundation

[mikensr]

Hello everyone.

First post here. I came looking for some help, if possible!

I have to help in the design of a shallow foundation for an oil storage tank. Previous projects of the company where I am doing this were executed with ring foundations and, well, it would be helpful if we could use the same solution once again.

Simplified stratigraphy of the soil is as follows:

First layer: 0-3 meters, sand;
Second layer: 3-20 m, clay;
Third layer: 20-35 m, sand.

GWT: 1 - 2 meter depth.

Tank has 10 meters radius, 11 meters height.

Assuming that the shallow foundation will rest in the superficial sand layer, I would like to have some guidance in the following questions:

1) As happens in the first sand layer (drained loading), the second (clay, undrained loading) and third layers (sand, drained loading) will also experience immediate settlements due to tank loading? How can I calculate them? I guess I can use something like Boussinesq Pressure Bulbs to understand the depth of influence of tank loading. The final immediate settlement is the sum of the partial immediate settlements, calculated for each layer?

2) The clay layer has a large thickness. What is the best methodology for calculating consolidation settlements? 1D theory like thin clay layers with no horizontal movements? Or a three-dimensional consolidation theory?

3) In respect to bearing capacity, should I only concern about the superficial sand layer, where the foundation will stand? Shouldn't I check bearing capacity for the clay layer, especially due to undrained loading)? But won't its bearing capacity be much smaller than the sand one, over designing the project?

4) The ring foundations usually executed have, for example, a section of 1,30 meter (height) x 0,70m (width). The center is filled with a correctly chosen granular material, I guess.

My question is, with this kind of solution, there aren't something like "two levels" of foundations? One is the bottom plate of the tank (where centered settlements will happen) and the other is the concrete ring itself, where also settlements will happen under it, no? Is this idea "correct" or "valid"? How can I simplify it for analytical calculation and then compare it to modeling (PLAXIS)?

Thanks in advance!

 

[oldestguy]

From what I know about this forum it is not set up to design anything. However,it may be OK to advise as to some aspects. In that light, your test boring logs probably have some information on soil densities. That is needed to come up with data to figure any settlement or shearing in the sand. If possible, you may need some laboratory tests to at least get the compression characteristics of the clay zone. Then, designing the slab, calculating settlements and perhaps a stability evaluation (shear failure) would be needed. The guidance in some of the references and text-books will be needed for these if you have not done this before. Finally, you need a review by an experienced geotechnical engineer. You may have missed something, but specifically you need assurance the design is safe and will work. If there is seasonal freezing, that may affect details.

 

[BigH]

You haven't provided any material details of your strata - is the upper sand compact or dense or is it loose? Is it a clean sand or silty sand? Similarly with the clay layer - is it soft, firm or stiff? Are you in an area where earthquakes are significant?

I have been involved with a number of very large tanks - up to 150 ft in diameter and we never used concrete ring walls even when we had firm clays. We always put the tank foundation on a tank pad - usually 1.5 m thick atop the natural ground extending 3 m or so beyond the tank footprint. A slight camber upwards towards the centre will help in any differential settlement. Is your tank a fixed roof (likely) or a floating roof tank? There have been a number of threads in the past on the issue of tank foundations - use of ring walls or not. You might wish to search for these threads.

 

[mikensr]

Huge apologies for the late reply.

No geotechnical tests have been carried out (until the moment) for a proper characterization of the subsoil. The data that I have is the result of some searches that I have done about the stratigraphy of the local. I ended up idealizing a stratigraphy for the tank subsoil and my (very previous) studies are being carried out from that point.

What I know is that the location of the tank is something like an alluvial deposit, where no other construction works have been done before. Therefore, I am tempted to say that the clay is soft and the granular layer at surface is something like a silty sand (loose).

My biggest concern at the moment is to understand what I have asked in 1), in my original post: Do all the layers experience immediate settlement due to tank loading?

Because, taking into account the simple elastic theory, load influence depth is always the same because it's just governed by the geometrical characteristics of loading (in the case, tank radius). But I was wondering if, for example, by building a dense granular embankment (let's say 2 meters) to put the tank on, isn't there going to be a better "degrading" of the load? Isn't the influence depth of the load actually changing?

 

[BigH]

Have you ever heard of two layer elastic theory - or three layer? See Poulos and Davis' Elastic Solutions for Soil and Rock Mechanics. There is a site that has it for download but I can't find it off hand; I have the book.