foundation on slope 1

[longisland]

Hi,
I'm designing an apartment on a hill; half of the building is on flat ground & the other on slope. The natural terrain is roughly 1:1 slope. I checked the borelog of the adjacent site; the top layer is silty clay & silty sand followed by shale & sandstone bedrock.
I'm planning to use shallow foundation on the flat section & bored piling on the slope ground. I'ved checked the structure to ensure no tension on foundation due to lateral load.
In this case, is UU or CIU more useful & why?
How can this problem be modelled? e.g, geoslope, since the slope is modelled in one "slice" but the pilecaps are staggered.
Any other stability problem I should be worry about?
I'm allow to assign two boreholes. I'm thinking of assigning both on the slope area; is there other suggestions such as location or other field tests?

 

[Sean2]

Longisland,

I sympathize with your position. I'll deal with the issues in the order in which you raise them.

The silty clay and silty sand may be soft or loose respectively. However, I assume you have already addressed that issue, and that these strata will support your proposed structure without leading to settlements or differential settlements unacceptable to your structure. You also have not mentioned thicknesses of the various strata.

If you DO pile on the slope, beware that the piled section of the structure will hardly settle at all, while the structure founded over the silty clay is likely to undergo at least normal settlements. This will lead to differential settlement across a boundary between the piled and un-piled parts of the structure. If the clay is not stiff (as defined in BS5930), you may consider piling the entire structure.

Piling on a 1:1 slope may prove difficult. Consider how the piling rig will gain access.

In modelling the problem, I suggest you have to model two different situations. First you have the slope, loaded by the foundation on the flat area, but ignore the restraining effect of the piles IN the slope. Second, assume that the slope will apply a lateral load to the piles before the occurrence of a slope failure. You need to design your piles for the lateral loads from the soil slope (pre-failure), even though you have not allowed for any support for the slope from the piles.

If that doesn't work, you may need to use the piles to increase the factor of safety on slope stability as well as to support the structure's vertical loads.

Consider the overall stability, taking the entire slope, especially if your "flat area" is a bench cut into a much higher slope.

You should carry out a risk analysis (which you will update at every opportunity during your analysis and design). Among other things, this will tell you whether you need to design the piles to resist the forces applied by the soil in the slope sliding past them as the slope fails.

I suggest that - IF you can obtain "undisturbed" samples of the soil, you should carry out Consolidated Undrained Triaxial tests with pore pressure measurements, i.e. effective stress testing. This will yield values of the effective cohesions and friction angles. Effective values (rather than undrained values) should be used as this is the long term condition, and, in a clay, is a worse case than the undrained (short term) case. Effective stress testing takes longer and costs more than the undrained tests.

Consider what the groundwater regime is at your site. You will not establish this simply from one reading in a standpipe installed in a borehole. My rule is that I need a very good reason not to consider that ground water levels can reach ground level. (These reasons do sometimes exist, but rarely in the UK and Ireland, where I work.) In your slope, you need to know what effect groundwater will have on the stability of your slope, and this must be included in your analysis models.

I don't know how big your structure is, but WHO told you that you could only have two boreholes? I strongly recommend that you or an even more experienced geotechnical specialist working with you decide what investigation YOU think you need. Only at that stage can anyone else tell you that he can't afford an adequate investigation. You are professionally bound to advise your client then of the risks and consequences of following a different course of action from the one you recommended! (However, two boreholes might be enough.)

Without any further information on the scheme, I suggest that you need at least four boreholes, two on the flat area, and at least two in the slope, UNLESS you know that conditions are likely to be fairly uniform.

If good "undisturbed" samples prove difficult to recover, consider carrying out shear vane testing in the field. This will also give you the residual shear strength of the soil, but only its UNDRAINED strength. You may wish to ensure that you carry out laboratory tests which will give you the residual shear strength, such as the small shear box test.

I hope I haven't only told you what you already know, and that some of this will be useful.