Immediate Settlement of Sand 2

[cmichaud75]

We have a project that has 200 feet of medium dense native clean fine to medium sand. Site development includes a tapered fill from 1 to 50 feet to achieve finish floor elevation. The fill will consisit of the same native sand cut from other areas of the site. Groundwater is greater than 100 feet deep. We have calculated anticipated immediate settlement from the weight of the fill using Schmertmann method. Considering a wide base width of 100 feet (plan area of the fill) and depending on the equation used for Es (stress-strain modulus), we are calculating up to 12 inches of settlement. Given the wide base width, the load influence penetrates very deep. I know this is supposed to be "immediate" settlement, but will settlement really happen incrementally as the fill is placed? Should I be concerned with differential post-construction settlement of the building that is placed on the extremely different fill depths? Any other general considerations for this situation?

 

[Ron]

Use settlement plates and monitor the movement. Your predictions are likely conservative.

Yes, settlement will happen incrementally as fill is placed. It is primarily elastic settlement....put load on, material moves.

 

[fattdad]

To what depth are you calculating settlement? If I recall the Schmertman method correctly, for an areal load (i.e., not a strip load) the "seat of settlement" extends to 2 times the "foundation" width. I think the size of your loaded area is inconsistent with the Schmertman approach (not sure).

How are you assigning modulus values for the settlement calculation?

Here's what I'd consider:

1. Define the soil profile (i.e., soil type, blow counts, etc.) and assign representative modulus values for the strata.
2. Consider hyperbolic modulus behavoir (i.e., modulus will increase with confining stress)
3. Look at the distribution of stresses with depth below the loaded area. Consider the center, the edge and the corners, at a minimum.
4. Integrate the stresses with depth with respect to modulus and calculate settlement (i.e., if the average stress change for a given layer is 2000 psf, the layer is 20 ft thick and the modulus value for the layer is 200 tsf, you'd calculate (2000*20/400000) 0.1 ft of settlement. This is a fundamental calculation.
5. Multiply the answer by 1.2 to account for long term affects. The 20 percent increase for long term affects is the likely extent of post-placement settlement.

+1 on the use of settlement plates.

f-d

¡papá gordo ain’t no madre flaca!

 

[BigHarvey]

Schmertmann's method is intended for shallow footings and not spread loads.
In your case, if you really have sand, a simple elastic approach will do but it's not necessary since your settlement is elastic and will happen immediately after fill placing.

 

[Drumchaser]

Sand (above 6 P.I.)is (in my opinion) the best fill material. You just have to ensure that it is "captured". No swell/shrink.
Just keep your optimum mosture + until the sand is covered with your slab or whatever. And remember that the sand will suck the moisture out of concrete placed on top if a moisture barrier (plastic etc.) is not used.

The sand fill should be placed with proper moisture and compaction. You should not have a problem.

 

[fattdad]

Does the OP concern the behavoir of the soil below the fill? My reply was focused on that, not the behavoir of the fill itself. Although settlement plates would address both issues, if used.

f-d

¡papá gordo ain’t no madre flaca!

 

[aeoliantexan]

Here is a case hisory for a sanity check: 35 feet of sand fill, many acres in area, on Missouri River floodplain with the surface clays (backswamp)removed. 150 feet of alluvium; top 100 feet is Recent, rest is probably Quaternary. Alluvium is mostly sand, loose to dense, uncorrected standard penetation resistance averages 12 blows per foot.

Settlement under 35 feet of fill was 2.4 to 4.2 inches. This was measured with settlement plates, so it is the complete settlement of the foundation with no compression of the fill included. Post-construction settlement over a few months was less than half an inch.

My rule of thumb for similar conditions is : Settlement = Fill Thickness / 100. Also E =12 N. These are averages, and variation is considerable.

 

[hollowstemhead]

Fattdad and others,

Could you comment on any concerns you would have with settlement within the actual fill mass in a situation like this?

When looking at elastic settlement of the underlying sands, what parameters and approach would you use in your model of a mass fill area, as the original poster has described?

What type of settlement plate setup have you had success with in a deep fill of this nature?

Just curious as I don't deal with deep sand sites of the nature of the posters site often.

 

[aeoliantexan]

Hollowstemhead,

On the project I described and two others like it, the settlement plates were pretty conventional: 12-inch-square by 1/4-inch thick steel plate with a 1-inch or 3/4-inch pipe coupling welded to the top and a 12-inch long pipe welded to the bottom. Dig a hole about a foot deep; level off the bottom; drive the pipe into the sand until the plate is well seated; screw on a 5-foot length of pipe; backfill the hole. Then slip a 4-foot length of two-inch pipe over the 1-inch and let it rest on some loose fill or a styrofoam donut so compression of the fill won't push the settlement plate down. Run conventional level surveys to the top of the pipe, using all the good surveying practices you learned in school. Add 5-foot lengths of both pipes as the fill rises.

Yes, some got damaged by equipment. They usually bend over about a foot below the surface and can be dug up, cut off, threaded and repaired. I fastened a steel rod to a 100-foot steel tape so I could lower it inside the pipe riser and measure the total riser length after the repair if necessary. Keep track of the elevation of the plate.

Later I changed the design to eliminate or reduce the surveying by cutting a hole in the plate, drilling a boring to a hard stratum below the compressible soil, and installing a Borros anchor or just driving a pipe into the hard layer to serve as a reference point. 1/4-inch pipe attached to the Borros anchor is run inside the 1-inch pipe. Settlement is monitored by measuring the "stick-up" of the little pipe above the big one with a ruler. These worked well for us, and saved a lot of surveying time.

The best protection for the risers is to include a provision in the specs requiring the contractor to shut down fill placement within 100 feet of a damaged instrument until it is repaired.

Surveying and instrumentation have changed a lot since then, but I would probably use the same procedures today.