Foundation in underconsolidated clay

[whitejuv]

I'm dealing with a mat foundation to be constructed at 3m below grade level. Dimensions are 20m x 20m. The loading from the structure will be 50 kn/m2. The removed load of soil is estimated around 60 kn/m2 (3mx20kn/m3). Normally I wouldn't expect settlements to occur, however the oedometric tests from the geotech lab indicate that the preconsolidation pressure of the soil is 35 kn/m2. Theoretically, it means that (without construction) the soil is still undergoing consolidation (ocr less than 1). soil is composed of clay with degree of saturation>95%. Modulus of elasticity is 5000 kn/m2.

Will the structure undergo consolidation settlements after the finished grade with the structure because of the ocr value?

 

[oldestguy]

Disturbed sample? In that a Newton is a unit of force,not mass, my conversion tables don't work. However, the removed soil submerged unit weight, not total mass with water,does not apply weight to soil below, since water is not involved. The density in the lab includes included water so it is not applicable. I assume the foundation will have drainage. If not. consider flotation by the water. You may need anchorage to keep it from floating.

 

[whitejuv]

Sorry about the confusion with the units, my fault.

It is a relatively undisturbed sample. I believe they used rotary drilling sampler to obtain it, not a Shelby tube.

So bottom line, I shouldn't expect settlement since the oedometer test fully submerges the sample on water while that's not the case in field conditions. Is that what you mean oldestguy?

 

[oldestguy]

Since this seems to be a fine grained soil site and "not fully consolidated", I assume the soil is saturated and there is a high natural water table. Some drying and a lower water table may make some sense using lab results. That's for figuring the unit weight of the excavated soil, maybe. A more complete description of site and lab work will help. The trouble with having field sampling by one outfit and then using that testing data by another group can lead to questions like these. Providing a plot of the lab consolidation test also would be helpful and show depth as to where it came from and actual sampling technique. Show all lab tests, if there are more. Giving only part of what is known can only confuse the issue and likely will not give useful answers here.

 

[whitejuv]

Im not authorized to post the lab reports here, but I'll sum the available data up.

The soil is composed of 51% clay, 23% silt, 23% sand, and 3% gravel. LL = 52, PL = 29, PI = 23, and LI = -0.5. The soil is classified as CH. - 1 grain size distribution and classification.

The unit weight is 20 kN/m3. The field engineer did not report they've faced the water table (drilled to a depth of 35 m). SPT N values range from 20 to 30.

2 consolidation tests (oedometer tests) were done for this clay, samples obtained at 4m and 7m. The preconsolidation pressure resulted to be 35 kN/m2 and 41 kN/m2 respectively. The virgin curve starts at around 40 kPa as loading is increased (at both samples).

For those samples, the overburden stress is 80 kN/m3 and 140 kN/m3 respectively. Based on this, calculating the OCR, i believe I'm facing underconsolidated soils (OCR less than 1).

I hope i've given a clearer picture of the problem.

Now, If we excavate to a depth of 3m, we remove a the load of 60 kN/m2. The building load will be 50 kN/m2. Even though I know the load of the removed soil is more than the load of the structures, should I expect settlements now that I know the soil is underconsolidated?

Or should I not believe the oedometer data and compute the OCR based on Mayne and Kemper 1988-Profiling OCR based on SPT and CPT? Based on Mayne and Kemper's correlations, with SPT N value from 20 - 30, the OCR will be greater than 1. If I can assume that the OCR is greater than 1 and the soil is overconsolidated, the settlements will equal 0 since we are removing more load than actually adding.

 

[oldestguy]

This modes of incomplete data so far. After completing each bore hole there should be water table readings, even times to days later if needed. Those densities are from split spoon samples? What water contents with depths? So far things given here are not complete and "not authorizing" posting a diagram is "nuts"? Who is in charge of this evaluation? If no complete water content info, get the sample and run them. A cheap test.

 

[whitejuv]

I did forget to include the water contents. The water contents for the 4m and 7m sample are 17% and 20% respectively. The densities are from relatively undisturbed samples obtained by a drive cylinder.

The company who performed the borings indicated that water level was not present at all.

 

[oldestguy]

OG back:

I'll post a chart that I have used many times with confidence.

Note how low your liquidity index is.

This indicates your effective consolidation pressure due to low moisture content gives very high load capability. That is why looking at the consolidation test plot may also show something wrong with the test. Errors are very possible there!!! That is something that seems fishy about the interpretation of that curve plot as meaning the correct value. Before getting too involved with the data you have, I'd have an experienced geotech look at EVERYTHING.

 

[oldestguy]

Og finally. If I'd be in charge of this field and lab work for this job, seeing the blow count and the FEW moisture content tests, immediately raises considerable question at even the thought of running consolidation tests. Very likely they (or it) are totally useless or unreliable considering the question that the original post raises.. More lab work or field work might be in order if what is available has ANY question as to site conditions as viewed by an EXPERIENCED geotech.

 

[BigH]

Do you have undrained shear strength values? If so back calculate p' from Su/p' =0.23 as a rough approximation of what p' should be.

 

[oldestguy]

OG once more.

Considering the low moisture content, possibly at the whole site (two tests might be representative, who knows), I suspect soil shrinkage cracks affected the consolidation test. Not showing the plot doesn't help me. Anyhow, take a look into the crack situation by a Google search. Two of many papers found are listed here which indicate there might be an explanation for the apparent assumed partial consolidation results of the consolidation test on low moisture sample.

I know of some other affects of shrinkage that are not apparent by just looking at classification tests. One being permeability.

http://users.ictp.it/~pub_off/lectures/lns018/39Taboada1.pdf

https://www.researchgate.net/publication/260812537_Shrinkage_and_desiccation_cracking_of_clay_soil

 

[Okiryu]

Based on:

- SPTs ranging from 20 to 30
- Unit weight of 20 kN/m3
- Low W% and no even close to the LL
- No groundwater

I would not expect underconsolidated conditions. I think that there was some sampling disturbance (looks like they did not use the proper tools), that affected the consolidation test results. Do you have any other strength test results? If so, you can double check your consolidation test results following BigH's suggestion.

Also, you are reporting the preconsolidation pressures in kN/m3 units (I assume this is a typo).

Anyway, talk with the drillers and the lab and verify with them the sampling and testing conditions.

 

[whitejuv]

Hi all. Thank you so much for your answers.

Yes, I do not trust the reports from the lab either.

I did not receive any unconfined compression results, but Im calculating Cu based on the equation Su/Pa = 0.06*N. For N = 20, Su = 120 kPa. With an Su of 120 kPa, using BigH's equation, the estimated preconsolidation pressure is around 500 kPpa.

I will not be designing, the foundations, my supervisor will, Im a young engineer and just wanted to know out of curiosity for future reference on what to do in cases like this.

Also, for you that have had much greater experience than myself, are underconsolidated clays common, and how have you proceeded with settlement calculations if you've faced them?

 

[oldestguy]

OK so your question came about why? Perhaps explain just where you fit in on the steps from: planning the investigation; through the tests and reports and now designing foundations. And pray tell what did the geotech say about this question you posed, if at all? I trust you now know that when you first post give EVERYTING even remotely related or a lengthy back and forth is needed to get it right.

 

[Okiryu]

I have seen underconsolidated/normally consolidated in reclaimed areas. You can start with the classic Terzaghi's theory of consolidation.

One question: how did you get the soil modulus of 5000 kPa?

 

[EireChch]

In the UK it is common to use the expression of E' = N/0.9MPA. There for for a N of 20, E' would be 18MPa.

 

[fattdad]

can you provide the lab-measured water content for the classification sample?

Do you have any unconfined compressive strength data (i.e., pocket penetrometer) data?

I'd also look at the water content in light of the LL and I'd look at the undrained shear strength in light of SHANSEP, considering some Su/P of 0.2 or thereabouts.

f-d

ípapß gordo ainÆt no madre flaca!

 

[aeoliantexan]

I agree that the low liquidity index and moderately high N values are not consistent with an underconsolidated condition. Possible explanations for the consolidation test results might include: 1) The soil is very sensitive and has been badly disturbed during sampling; 2)The sample is of slough that settled at the bottom of the boring; and 3)The consolidation machine is badly mis-calibrated.

I was under the impression that natural deposits of underconsolidated soils may be found in rapidly-depositing deltas or bays. Man's activities, such as placing fill on soft, fat clay can produce an underconsolidated condition for a few decades. I have been practicing almost 50 years in a variety of geologic conditions and don't recall ever encountering underconsolidated soil.

 

[fattdad]

My experiences with underconsolidated soil are related to contractor's claims that the primary consolidation was complete below an embankment and an attempt to build a road through hydraulic fill from a former (i.e., over 40 year ago) brick making site.

It's easy to know if the soil is underconsolidated. Just install a well or use a piezo-cone!

f-d

p.s., I carry a pocket penetrometer to the drill site to see if there is a disconnect between n-values and unconfined compressive strength. We have marine soils with N of 8 or 10 that will return PP of 3.5 to 4 tsf. Wouldn't want to think they are too soft!


ípapß gordo ainÆt no madre flaca!