settlement from plate load test and real footing 6

[-Daniel-]

Hi all, can anyone help to solve my problem? Normally, Plate load test is used to verify the site soil bearing capacity. but normally the settlement is quite small. (say 5mm of settlement while applied pressure of 350 kpa, which means, if reach a settlement of 25mm, the pressure can be 1000 kpa+, not really tally with the calculation). I also heard while the footing size is bigger, applying same pressure on top, the real settlement will be bigger. Can anyone share with me the relevant reference book? Many thanks.

 

[maxim22]

Any / all soil mechanics with foundation engineering text books.

Soil mechanics in engineering practice by Karl Terzaghi.

Fundamentals of Ground Engineering by John Atkinson.

A short course in Foundation Engineering by Simons & Menzies.

 

[oldestguy]

This 91 year old geotech much doubts that will find it easy to answer the question in a text book. Reason is that we (getotechs, that I know) don't usually use plate load tests. You are right that the plate only loads a small area. Then the small area load has to be extrapolated up to footing sizes. Unless you have a detailed test boring to verify the density and settlement characteristics of all soil that might be loaded (within depth of significant stress (usually over 10% of footing pressure)) the result of some extrapolation will be rough, maybe OK and maybe not. I'd pass this question on to a local geotech, since the full info for an answer is not provided to us here. Some have (me also) will run two plate load tests of different sizes to get a rough idea about what goes on with differing sizes, AND ALSO HAVE A DETAILED TEST BORING.

 

[aeoliantexan]

Foundation Analysis and Design by Bowles has a reasonably complete discussion. Many geotechnical or structural engineers have a copy.

 

[pelelo]

In my opinion, the main problem with the load test is that is time consumuing and too expensive, therefore, usually clients do not expect to pay for these tests.

 

[fattdad]

the main problem with the plate test is it doesn't mobilize the behavior through the likely seat of settlement for the footing. It doesn't scale, in my humble opinion.

f-d

ípapß gordo ainÆt no madre flaca!

 

[-Daniel-]

to：aeoliantexan

thanks, i have that book, may i know which chapter has the answer? I'd like to have a study.

 

[Okiryu]

I think that if the soils are very consistent (moisture content, stiffness, grade of saturation, etc.) thru the entire depth of influence of the planned foundation, perhaps the plate bearing test may give you an idea of the load-settlement behavior of the expected real footings (you have to scale the values of your plate bearing tests as mentioned by others above). If you have a layered profile with several different soil properties thru the depth of influence of the planned foundation, I would forget about the plate bearing tests and rather, I would consider taking good undisturbed samples for lab testing or do some in-situ tests (pressuremeter tests).

 

[Okiryu]

Hey OG, are you already 91 ! Wooow ! I hope I can keep being enthusiastic and very responsive (and healthy) as you are when I am at that age ! Thanks again for the good and helpful posts !

 

[oldestguy]

Hey Okiryu: Well this old guy doesn't feel 91, but surely does not feel 21 either, but somewhere in between. Main problem is the exercise I do seems to miss some muscles so when out there doing things around the house, I find the daily exercises may miss something and ittakes days, not hours to get back to normal.

 

[ATSE]

I suggest not using a plate load test at all for foundation design. It's not just expensive; the data is useless, or worse, misleading.
1'-0" x 1'-0" footprint tells you almost nothing about the behavior of a 5'-0" x 5'-0" footprint.
Lab testing and simple hand calcs are the main tools for evaluating settlement and bearing capacity.

 

[aeoliantexan]

In Bowles' 5th edition, it is Section 4-12,Page 267.

I agree that the results can be misleading in stratified soils, but if you know that the soil is consistent within the depth of influence, you can factor up the results and make use of them. Some soils and highly weathered rock are not suited to sampling and lab testing, and field tests must be resorted to.

If you don't know the subsurface conditions to the depth of influence, you need to explore.

 

[geotechengineer1]

The plate load test can be quite misleading. In the end I think it might be worse than empirical or semi empirical correlations because the engineer can be fooled into thinking that what happens in the plate load test can be taken as an accurate representation of what will happen with a full scale footing.

I suppose that if your soil profile is truly a uniform, isotropic, elastic, saturated, homogeneous soil with the same composition beginning at the ground surface and extending to the depth of the zone of influence of the full size footing, it might give you a good idea. But soil is not uniform, isotropic, elastic, or homogenous.

And, the big one, how often do you have the water table at ground level? More likely, the water table is at some depth and all of the soil in the zone of influence of the plate load test is under the influence of matric suction and other fun things about partially saturated soil mechanics that make the results of the test, in essence, totally meaningless when compared to a real footing.

 

[BigH]

A short case history of a potentially misused plate load test. 10 m MSE wall to go on a soil - known for being very soft for 6 m at least. Structural engineer ordered plate load test. Test result - 350 kPa easily!! Structural engineer - just put up the wall. Oops! I, and a few others were able to counteract his recommendation. Why? He did a 300 mm plate load test in a 2 m fill that had been there for years. Built the wall in the same way we did the others in the area - wick drains and stage loading. 4 m of the first fill layer settled 400 mm. Be very careful - for me, plate load tests are good for one thing and one thing only - pavement structure - if you must.

 

[-Daniel-]

Hi aeoliantexan & all

thanks for your advice and reply.

I check with section 4.12, it is about how to determine the bearing capacity from field load tests (say plate load test). And some formula mentioned,Qult = Qplate (Bfoundation / Bplate). assume Bfoundation / Bplate = 3 (same to the design FOS) Qall = Qplate which base on most of my received test report, it can be ....say 300 kPa or even much more, as the settlement at that pressure is only 5mm--10mm, not 25mm. (Assume it is correct, i see the textbook also mentioned some limitation for this method use, but for certain case, seems like it still can be used). This allowable pressure obtained is much bigger than what calculated, and in my country, soil is not so good. (I know in US, some part of the soil is really good), so i don't really believe this result. Any other reference or correction method that i can refer to?

and my another question which this section 4.12 does not give me the answer is..... if the applied pressure for the plate load test is 300kPa and the obtained settlement is 5mm. what will be the expected soil settlement if the footing size is 1.5m x1.5m ? any reference can refer to?

many thanks to all.

 

[maxim22]

OP - text books do not allow to scale from plate test to foundation settlement because it is not good engineering unless multiple favourable ground conditions are satisfied at the specific site

 

[aeoliantexan]

Bowles has about as much on the subject as I have seen. If you can muster enough load to run 2 or 3 plate sizes to ultimate bearing capacity, you may be able to extrapolate the data using the formulas in 4.12.1. The bearing capacity of cohesionless soils should be proportional to the width. The bearing capacity of strictly cohesive soil should be independent of the width, but Housel points out that perimeter shearing stresses contribute more for small plates than large ones. For uniform soils and pressures well below ultimate, analyses and extrapolation can be attempted using formulas from elastic analyses, but confinement tends to cause higher stiffness at greater depth.

The pitfalls are numerous and serious enough that I would not recommend attempting foundation design from plate load tests unless (1)the soil is confirmed to be uniform to at least 2 times the width of the proposed footing and (2) You have a thorough understanding of the principles of both ultimate bearing capacity and settlement.

 

[-Daniel-]

Hi aeoliantexan

OK，thanks for your advice.

may i know, what is the method you will use to verify the site soil bearing capacity after compaction / construction? Say base on the soil report / SPT value or Cu value, you use allowable soil bearing capacity of 100kpa , and on site did some compaction, etc. how do you determine the site actual soil bearing capacity?

thanks

 

[aeoliantexan]

For routine local projects, We would use allowable bearing pressures for compacted clay fill depending on the degree of compaction, for example, 3000 psf for lean clay compacted to 95% of Standard Proctor density. These are based on extensive local experience. For cohesionless fill compacted to some relative density, we might use published correlations to estimate the friction angle and use traditional bearing capacity formulas after selecting the depth based on frost action and other considerations.

We would evaluate the native soils using unconfined strength for clays and SPT for sands. If the native soil is significantly weaker than the fill, we might select the fill thickness to limit the pressure at the top of the native soil to its allowable bearing pressure, or reduce the design footing bearing pressure if necessary.

For heavy loads or unfamiliar soils, we might resort to unconfined or triaxial tests on compacted fill specimens and consolidation tests on compressible clays. We might estimate elastic moduli of both fill and native soils from published correlations to calculate settlements. If compressible soils existed at depth, we would adjust the combination of bearing pressure and structural fill thickness to avoid stressing the soft soils above the preconsolidation pressure. If that was not feasible, we might preload the building site or resort to deep foundations.

 

[-Daniel-]

To aeoliantexan :

many thanks for your reply.

so if my understanding is not wrong, you will do a lab test to get the soil density after compression. say the clay layer standard density is 18 kn/m3, if you test can reach a density of 18/0.95 (or 18*0.95, sorry which one is right?) , it assumed that the soil is compacted to the design requirement, am i right?

thanks again.