Settlement in trench from raising ground water levels.

[Abeshaka]

Wanted to see if any one has any idea the cause of the settlement seen in the trench zone. Background the trench is approx. 10 feet and 21 feet deep the ground water table is usually at 10-12 feet. dewatered every 50 feet to draw down the water table to 21 feet below grade, pipe 60" sewer pipe installed and encased in CLSM (2 feet above pipe) that that point was backfilled with Clayey Sand, 35percent passing #200 sieve, LL of 25 and PI of 11 Proctor around 121.8 @13.8, all fill was compacted to 90 percent or greater and 2to 5 percent above the optimum. Two feet of Type II compacted to 95 percent. and no problems were encountered.
When the dewatering was stopped and the groundwater levels raised up back to 10 to 12 feet below grade , 8 to 10 inches of settlement was observed in certain areas along the trench line?????

Any thoughts When we drilled borings the blow counts dropped of to 1 and 2 when we got close to the water and the fill above the water table ranged in 5 to 10 blows per foot (N60).

 

[oldestguy]

Explain the backfilling procedure. Thin layers and each thoroughly compacted? Inspection only now and then? I'd bet someone got away with some dumped in thick layers when the inspector was at lunch.

 

[Abeshaka]

The material was placed in 1 foot lifts and compacted with a Wheel attached to an excavator, the material was moistures conditioned in the lay down yard and then brought to the site, this same process is used throughout the job site approximately 3 miles, just this issue in certain areas where there is a lot of ground water flows? encountered. If the dewatering pumps are shit down the water table in the excavation rises some 8 to 10 feet within a day. Test pits dug and compactions at 90% with moistures above optimum, For example around the corner to this street, This section of street was backfilled with native for around 15 feet and type II placed( 2 feet) and Then paved with 6 inched of asphalt. This was done 4 months ago, while investigation the street that runs perpendicular to this area on Friday of last week no settlement observed. At the intersection of both these street is an shaft that the dewatering wells shut down over the weekend, over the weekend the water level rose in the shaft and to flow down the trench line and for two hundred feet you saw settlement of approx. 10 inches. We dug test pits in the area also and found the type II was still 95+ in compactions and the native select backfill was also 90 plus in compactions and tight to 8 feet, but the soils had moistures to 19.5 to 17.5 percent. But based on the proctor still achieved the 90 Percent ( 121.8pcf at 13.3%)

On Monday after the settlement the asphalt felt like it was pumping.

 

[oldestguy]

Original ground is what? Any chance that the water running to the "dewatering inlets" actually loosened soil below trench base? What was the procedure for dewatering? Sumps? well points? When was water drawn down with respect to the excavation work at those areas? Before or after excavation, during? What I am getting at is was the soil below the trench loosened because of water flow up into the trench loosening soil there? This usually results in immediate pipe settlement, but maybe the added weight of water did it. I know of a lift station that dropped a few feet due to improper dewatering earlier on.

Anther thing to try is compact samples at the percentage used for filling and also do the same at the percentage found later. Run a form of commissioner test, such as unconfined compression, CBR or other to see if only that moisture increase was the cause. Compute void percentage for each which might explain the increase in moisture permitted.

 

[aeoliantexan]

Several things are fishy. Assuming the specific gravity is 2.67, when I plot the maximum density and optimum water content, the point falls right on the zero air voids curve. If the Proctor curve is a normal shape, half the curve will be above the ZAV line; impossible. Of course, your soil may have an unusual specific gravity, but it would have to be greater than 2.80 for those numbers to make sense.

You say, "the water flowed down the trench". Did it flow in a coarse granular bedding layer? I have see cases where the backfill was dumped in and compaction started well above the spring line, and there was no soil beneath the haunches. Water flowed down that trench very well.

The behavior is what would be expected if the compaction in the lower part of the trench was very poor, like none. I would question whether the procedure described was actually followed in the areas that subsequently settled.

One more item: If the subsequent tests were taken by pushing tube samples with a drill rig, they were probably compacted in the sampling tube. I have seen results where every sample plotted on the ZAV line showing that the wet soil was compacted to 100% saturation during sampling.

 

[oldestguy]

Can the pipe be inspected for possibly settling also?