Estimating settlement induced by embankment fill? 3

[grabens]

I have a site with the following soil profile:

0 to 30 feet Cemented Hard Pan [SM], SPT N = 100+/ft
30 to 50 feet Very soft clay [CH], SPT N = 2 to 4

There will be a 25 ft embankment fill placed on top of this soil profile for an MSE wall.

I am trying to determine how much stress will be carried by the hard pan (possibly by bridging) and how much will be distributed to the underlying very soft to soft clay.

I need to estimate the amount of settlement that will be induced by the embankment fill.

What would be the best approach to determine this?

 

[FixedEarth]

"Geotechnical Engineering" by Coduto 1999 +/- has a software THAT you can download and one of the programs (FILSETT) does the embankment settlement analysis. Otherwise, just create a spreadsheet using a stress distribution and Terzaghi's consolidation theory. Since the Hard Pan is much stiffer than the very soft Clay, Westergaard's stress distribution may be more applicable.

 

[Ron]

With cemented hardpan with blow counts that high, you will have to get a shear failure in the hardpan before settlement in the clay will occur. I agree with FixedEarth that Westergaard's stress distribution would be the way to go; however, Westergaard's distribution for such an application will likely show little or no stress in the clay layer. If shear failure in the hardpan occurs, then you have complete loading of the clay layer.

Check the shear strength of the hardpan. This would be a good application for elastic layer analysis.

 

[FixedEarth]

Even though your Hard Pan is moderately to highly overconsolidated and as Ron stated it will act as a Raft to distribute stresses, however, you will still get considerable settlement. See Terzaghi and Peck '67 pp 504-506 for similar discussion.

My quick calculation shows between about 5 to 7 inches of settlement for worst case soil profile. You may monitor the Fill and when your settlement assymptotes, you may proceed with your wall construction.

 

[molerat2210]

Push tubes in the clay, run consolidation tests, use stress contours provided in NAVFAC for stresses beneath an embankment (DM 7.1). Not possible to accurately estimate settlement from N-values.

 

[molerat2210]

Use a larger poisson's ratio to represent the 100 blow material.

 

[Ron]

molerat2210...why a larger Poisson's ratio for the 100 blow material? Generally a lower Poisson's ratio would apply to "stiffer" material.

 

[molerat2210]

You are right, Ron. Thanks. I hope my cred is not blown.

 

[FixedEarth]

This is just a rough approximation with assumptions of GWT at 15 ft. Notice the total settlement is only an inch to the bottom of the hard pan.

 

[grabens]

FixedEarth,

Thank you for all of your help! Would you be willing to post a copy of your spreadsheet to play around with?

Thanks,

 

[FixedEarth]

See attached.

 

[grabens]

Thanks Fixed Earth! You are awesome!

 

[FixedEarth]

grabens- For more explanations and nomenclature, refer to Geotechnical Engineering by Coduto. Good luck.

 

[fattdad]

You'd have to consider the width of the loading as well as the height of fill. I mean the areal extent of the load is what determines the stress profile with depth.

The hardpan will act as a raft of some sort, but I'm not sure I'd rely on too much extra attenuation from the hardpan.

Somehow you need to know the maximum past pressure from the clay. Pp-Po may be 1,000 psf, 500 psf or 2,000 psf and that can be a huge factor in the settlement calculation.

You need also to know Cv for the soft clay. I mean if its that thick and the drainage paths are long, the settlement response could take decades.

Run the numbers with Westergaard and Bousinesq. I already know that Westergaard will be less, but it never hurts to bracket your conclusions. . .

f-d

¡papá gordo ain’t no madre flaca!

 

[aeoliantexan]

I agree with Fattdad; you need undisturbed samples and consolidation tests. The profile seems like an unusual geologic condition. Development of the hardpan must have taken a long time. Time can add to the preconsolidation pressure. Perhaps the clay is more overconsolidated than the blow counts indicate - the settlement may be small.

On the other hand, if the clay is really normally consolidated, the differential settlement could be pretty severe when the hardpan cracks.

 

[oldestguy]

A lot depends on what sits on that 25 foot added embankment. Also, MSE walls can be built quite flexible and even tilted back considerably to allow for a lot of differential settlement. Before getting too excited about this, maybe a look at what con be done to allow it to function even if differential settlement happens to be noticeable. You might even do some geogrid reinforcing within the new fill (in addition to that needed for the MSE wall) in case Ron's shearing idea takes place.

 

[moe333]

Probably don't need a shear failure to occur through the hardpan to transmit the full load of the fill as natural fractures/fissures (in various directions) are common in soils.