Soil Cement Design 1

[geomane]

Where can I find a good handbook/design guide for soil-cement design? I thought I used to have one from the portland cement association, but I can only find my soil-cement inspectors manual and soil-cement construction handbook.

We are working on a project where the contractor has requested to construct a soil-cement layer in lieu of hauling in an aggregate base course. The existing soils are poorly graded, fine to medium sands (SP). The rigid pavement design was performed using PCASE, and require an equivalent modulus of subgrade reaction of around 450 pci. This equivalent modulus of subgrade reaction is measured from the surface of the soil-cement layer, so it also includes the strength of the subgrade soils. We are considering specifying a few plate load tests on some test sections/strips to verify the equivalent modulus of subgrade reaction is met. Do you see any issues with performing a plate load test on a soil-cement treated layer?

 

[Ron]

450 pci is high. You'll need a soil-cement in the 300 psi range to achieve this.

Here is a link to the old Soil Cement Laboratory Handbook produced by the Portland Cement Association but no longer available....

Soil Cement Laboratory Handbook

 

[PEinc]

Probably not exactly what you are looking for, but this is all I have.


www.PeirceEngineering.com

 

[EireChch]

I dont have a standard to refernece, however we backfilled some failed pipes with soil/cement mix of 6% cement by weight. We achieved a UCS of more than 2MPa (300psi approx)

Not sure how Ron has equated this to a subgrade modulus of 450pci though? I dont do much pavement design

 

[geomane]

Thank you for your responses!

Ron, is your estimation of unconfined compressive strength to subgrade reaction modulus based on experience? Or is this something you verify by lab testing or plate load testing?

 

[fattdad]

a CBR of 50 would relate to k=450.

I have a problem with field plate tests. They show the strength from soil suction. Clearly, the CBR shows strength in a saturated condition.

Now, where do you find a subgrade with CBR=50?

So, lets say, the poorly-graded sand with gravel has a CBR of 30? That would relate to a subgrade modulus of about 300 pci. Provide a dense-graded aggregate subbase at 12 inches (100 percent Standard Proctor) and you'd have an engineered subgrade with a design k=450.

I got these graphs!

f-d

ípapß gordo ainÆt no madre flaca!

 

[oldestguy]

Before getting too far into this, what kind of pavement system is to be built on this subgrade? In some cases that treatment will affect the pavement design and in others it has little effect.

 

[geomane]

OG, we are replacing sections of an airfield (taxiways, runway, etc). It will be a rigid pavement section, about 12 inches of concrete overlaying a 12 inch thick soil-cement treated subgrade layer built on the native subgrade souls which are poorly graded fine to medium sands. Water table is 2-3ft deep. Existing pavement sections are built with a soil-cement treated base layer.

 

[oldestguy]

Apparently with that kind of job you need to be fussy to satisfy the reviewers. However, with rigid pavement design (performance etc.) soil conditions below any pavement have little effect on the overall concrete thickness required. It's the loading's and number of them that controls.

 

[geomane]

What is the importance of unconfined compressive strength of a soil cement treated base layer for rigid pavements? Aren’t we just concerned with the modulus of subgrade reaction with rigid pavement design?

I agree with your statements fattdad. Do you typically see CBR tests performed on samples to correlate to modulus of subgrade reaction for rigid pavement design on DOT projects?

 

[ATSE]

Try this:

https://www.cement.org/docs/default-source/th-paving-pdfs/soil_cement/eb003.pdf?sfvrsn=2

 

[Okiryu]

In my area, for pavements, soil improvements contractors (or cement manufacturers) have charts that correlates cement amount per m3, CBR and unconfined compression strength for typical soils. I think that for soil-cement mixings, unconfined compression strength (qu) tests are the most important parameter because these tests are easy to conduct (and cheap) and can reach the required high compressive strengths. If you are designing for rigid pavements, the required modulus of subgrade reaction (k) can be correlated to a CBR and then go the charts to determine the amount of cement per m3 for preliminary mix design. Then we specify to field verify the k by field plate bearing tests. Since plate bearing tests are expensive, QC can be easily done using small molds to be tested for unconfined compressive strengths (I think that is one of the reasons qu is important- it makes the QC easy).

PCASE has a sensitivity tool so you can check if the 450 pci is really affecting your slab thickness. You may not get to much benefits for reducing the slab thickness specifying a high k. Specifying a high k will result in a higher cement amounts which may not be cost effective.

USACE has guidelines (

https://www.wbdg.org/ffc/army-coe/standards/crd-c655)

to get the saturated k from the field (unsaturated) k. But for soil-cement mixing I will not be worried about the adjusted saturated k.

For soil improvement methods in general, I found geotechtools (

https://geotechtools.geoinstitute.org/)

to be a good source.

 

[RGKVI]

I recommend EuroSoilStab2002 and BSEN 14679-2005. If you cant get hold of them I could email them to you.

 

[oldestguy]

On this question: "Do you see any issues with performing a plate load test on a soil-cement treated layer?"

I would not be surprised to see some pretty good numbers even with low percentages (4% or lower%). Perhaps run at different % before the exact amount is set for the job. Compact the sand first.

 

[RGKVI]

UCS, CBR and PLT are all useful for verification, UCS is the go to test to get a quick idea of the undrained shear strength but elastic modulus is very conservative because the sample is not constrained. Plate load test will give good values for Modulus of subgrade reaction (good for designing slabs) and you can correlate CBR values. A 300mm PLT on the SM platform followed by a 450mm plate on the subgrade will enable you to accurately verify a platform to BRE460 using the FPS load cases.