Settlement of raft foundation in sloped field.

[YounEng]

Hi all!
I have a question regarding the settlement analysis of a raft foundation on a sloped field (im attaching a rough sketch). My question is, what should I take as foundation depth when I do the calculations. On one hand, on the side where the foundation depth is 7m, the load of the removed soil is greater than the load from the structure. On the side where the foundation depth is 4m, that is not the case. Is it safe to say I will have differential settlements? Should I calculate the settlements for both foundation depths and see the difference in settlements?

All the books that I've looked into have assumed that the ground level is flat. So do some software that I've tried. I'm curious what method is applied for settlement calculation in a scenario like this.

Help would be much appreciated. I hope someone in the forum has faced and knows the right method for a scenario like this. Also, if you've come across books that explain this kind of problem please let me know about the titles and authors.

 

[YounEng]

I forgot to include it, the expected loading on the foundation should be around 110-120 kPa.

 

[EireChch]

What are the slope angles above and below your site. From that picture it looks like you need to consider slope stability. Although its not to scale, so who knows. If determine that your building platform is stable then the settlement calc is the same as the one for flat ground. All the soil knows is that there is an increase or decrease in stress and this may result in settlement or heave.

There is a a lot going on on this site. 8 Storeys is a big building, 7m high retaining wall, sloping ground. You really need an expert. I woudldnt attempt to tackle all these issues with out the help of a senior engineer.

 

[YounEng]

Thank you for your answer Eire. I did check for slope stability and the factor of safety is greater than 2. The slope on the upper part is 24% while on the lower part it is 20%. I'm attaching a sketch of how the field will look after building it (marked in red). In the left side of the sketch, the load of the soil removed is more than the load of the building, while on the right side, the load of the building is about 10-15 kN more. That's why I was afraid the building could have differential settlements. I'm assuming I should I calculate for foundation depth = 4m for the entire building?

 

[hoshang]

Hi
My thought is using superimposed steps: first consider footing resting on flat ground and second consider embankment adjacent to footing. This superimposed is used to find stresses and then using theses stresses to find the settlement.

 

[fattdad]

In the realm of settlement, you'd want to know if you are dealing with ANY virgin consolidation. If the answer is, "No" then you will be addressing settlement using elastic theory (or the coefficient of recompression). Either way, the stress history will be trivial as your stress range is less than the past pressure. If the case, then run the numbers without concern for the offloading effects of grading.

In the realm of slope stability (and if there is no virgin consolidation), then when the grading is complete, you'd consider Su/P and the strength and you'd consider phi and the strength and run the numbers.

I agree that a scaled profile would help to better visualize the problem.

f-d

ípapß gordo ainÆt no madre flaca!

 

[YounEng]

Thank you for your answers.Ill create a scaled profile and upload it soon.

My main dilemma is, now that I know I should start with a level surface, should I generalize that the foundation depth is 4m, 7m, or take an average between the two? I have the dilemma because taking a foundation depth of 4m will give me different results than calculating with a foundation depth of 7m.

 

[YounEng]

I'm attching the scaled profile.

I ran the slope stability, it checks out fine, the only question is the foundation depth used for calculation (4 or 7)? And should I expect more settlements on the lower side than on the upper side (where it should be 0)?

 

[fattdad]

The question of past pressure remains. Are you dealing with soils likely to compress or consolidate? If the latter, do you anticipate ANY virgin consolidation?

If you are ONLY considering compression, figure the stresses will normalize after earthwork (i.e., when the building pad is made flat). Then the building load will instigate new elastic compression.

If there are no slope stability concerns (i.e., shear in excess of strength), then it seems you are dealing with foundation settlement and, perhaps, ultimate foundation capacity.

f-d

ípapß gordo ainÆt no madre flaca!

 

[YounEng]

I am dealing with mostly saturated clay, I anticipate consolidation. The virgin curve seems to start at the load a little over 100 kPa, so most likely yes, virgin consolidation is anticipated.

 

[fattdad]

use piling, account for downdrag (perhaps) and transfer the load to a greater depth.

f-d

ípapß gordo ainÆt no madre flaca!

 

[YounEng]

Thank you for your answer f-d. If (hypothetically) I was not anticipating any virgin consolidation from consolidation tests, would you calculate settlements for a flat surface for a foundation depth = 4m?

 

[Okiryu]

I think that you are trying to calculate elastic settlements. I expect you are trying to get some factors to plug in your elastic settlement calculations that depend on the foundation depth (Df). However, you have saturated clays and your acting stresses are in the virgin portion of your consolidation curve. Not sure about the thickness of your clay layers but I think that you may need to be more worried about long term / consolidation settlements rather than elastic settlements. Check the consolidation settlements: because your mat covers a large area, you will get different results based on the different thickness of the clay layers throughout the area that covers your mat. Talk with the structural engineer about this differential settlement so the mat can be reinforced to account for this. If your settlements are larger than the allowable ones, you may need to look at deep foundations as f-d suggested. You can consider to install drains, but because this is a large structure, I would be extra conservative and go for piles. Lastly, are there any other buildings in that area? If so, what type of foundations do they have? Are they performing well? Again, this is a large structure, so put extra care on this. Good luck!

 

[aeoliantexan]

I would start with classical consolidation settlement analysis. Find "Elastic Solutions for Soil and Rock Mechanics" by Poulos and Davis free on line.

Compute vertical stresses under the center of the right wall, the left wall, and the center of the building. Use depth increments of 5 feet to 20, then ten feet. Calculate the existing overburden stress as depth below present grade times soil unit weight, then subtract pore pressure as depth below water table times water unit weight. Calculate stress reduction due to excavation by treating the soil as a stack of rectangular, vertical-sided areas. You can use the charts for influence factors under the corners of loaded areas and use positive and negative loaded areas to get the stresses at the locations you have selected. Then calculate the stresses due to the loaded mat foundation. You can consider the mat to be flexible; then repeat for a completely rigid mat. The comparison will be interesting.

Add the backfill if it seems significant.

Add the new fill on the downhill side.

If the excavation goes below the water table, you will need to decide whether to install permanent subdrainage or a watertight basement. Then estimate the final pore water pressure at each and adjust the effective stresses.
Next use the one-dimensional consolidation equation to calculate recompression and virgin compression of each layer at each location. Recompression tends to be overestimated, so you might reduce the recompression by some factor (maybe half to two-thirds). Add the compression of the layers to get settlement at the bottom of the mat.

You may calculate elastic compression at the various points and add it in. Any virgin consolidation will dominate the differential settlement, though.

Now you can compare the predicted settlements at each side and the center. If the differential settlements look tolerable, repeat the exercise for the corners.

Then talk with the structural engineer.

This is a couple of day's work with a calculator, less if you are handy with Excel. This project is worth the effort.

 

[YounEng]

I calculated the settlements for all three points (left wall, right wall, and middle of the building). I used the existing grade for calculations, so the foundation depths were 4.0m, 5.5m, and 7.0m respectively. The results are as follows:

left wall (Df = 4m) -> Elastic settlement = 3cm, Consolidation settlement = 7.4m - (the final grade of the field will be 0.5m above the foundation level after construction is finished-as shown in the scaled sketch - the architects projected it to be at the same level with the road level - no fill).
center (Df = 5.5m) -> Elastic settlement = 0.9 cm, Consolidation settlement = 1.6cm.
right wall (Df = 7.0m) -> Elastic settlement = 0, Consolidation settlement = 0.

The soil layers underneath are uniform so I dont expect differential settlements because of soil layers. My confusion is on the determination of foundation depth and approval that it is okay to perform calculations for different points in the field for a single building when we have this kind of slope.

Is it now safe to say that differential settlements are present and piling should be designed?

 

[Okiryu]

I was expecting you to have larger settlements at the center but assuming that calculations are okay, check around in the literature the allowable differential settlement and angle distortion. I just google it and found this:

http://infohost.nmt.edu/~Mehrdad/ME442/assets/pdf/TolerableSettlementofBuildings.pdf

You can also show a plan sketch the calculated expected settlements at several locations of the mat (corners, center, left and right walls, etc) to talk with the structural engineer.

 

[Mad Mike]

I would like to know how you have a ~15m profile of "mostly saturated clay" near a hilltop- the only thick sequences of saturated clays I've encountered on slopes like this are landslide deposits. This would be straying off-topic so don't provide details unless you're concerned.

All the best,
Mike

 

[SlideRuleEra]

Is it now safe to say that differential settlements are present and piling should be designed?

No... but IMHO, it is safe to say that a raft foundation can reasonably be expected to have differential settlement issues and it is time to investigate other foundation types. Piling are probably a better solution.

Take a look at the basic issues such as:

Driven steel or concrete or cast-in-place concrete
How long are the piling
What loading per pile
Projected pile settlement under load
How many piling
Conceptual pile layout to make certain the required number of piling will "fit" the building footprint (with proper pile spacing)

http://www.SlideRuleEra.net

http://www.VacuumTubeEra.net

 

[aeoliantexan]

I assume you meant 7.4 cm of consolidation settlement, not 7.4 m. If so, you are predicting about 4 inches of differential settlement across the building. I grew up in a farmhouse where my marbles always rolled to the west wall, so I can say the slope of the floors will be noticeable, and the overall tilt will be noticeable from the outside to a sharp observer.

With that much settlement predicted, I assume that you have some normally-consolidated soil in the profile. The first thing I would do is look closely at the consolidation tests to see if the samples were reasonably undisturbed and the interpretation withstands scrutiny, If so, you still may have alternatives to a pile foundation. Perhaps you can move the building farther into the hillside to get more excavation at the downhill side, lower the building, or add a basement or a partial basement on the downhill side.

If you are not a geotechnical engineer, this would be a good time to get one involved, as well a structural engineer experienced with the design of raft foundations. There is still potential for differential settlement within the building

 

[fattdad]

If you are not triggering virgin consolidation, "Yes!" I'd just take the finish grading, apply the loads, consider attenuation, make sublayers and calculate elastic settlement ignoring the offloading effects.

I would not consider both consolidation AND compression even if you have a consolidation test showing some value of recompression ratio (or recompression index). Run the numbers using recompression ratio and get an answer. Run the numbers using soil modulus and get an answer. Compare results.

f-d

ípapß gordo ainÆt no madre flaca!