Removal of Silty Soil 1

[Gercivil]

Hello to all...I would like to ask regarding the very silty soil found in our project site if it is ok or not to remove this unsuitable material before contructing 250nos. three storey houses(ground floor area: 8m x 13m).
As per soil investigation report the sieve analysis for 3 out of 6 boreholes have an average of 20% passing percentage in .063mm sieve size from -0.5m to -2.5m depth.
I would like to have a second technical opinion since it was recommended to remove only 0.3m of unsuitable soil and replace it with suitable material. Is it enough to remove only 0.3 of the unsuitable soil?

Many Thanks in advance!

 

[Ron]

If that 0.3m of soil settles by 20% will it make a difference to your structures?

Second opinions are fine, but that second opinion should come from someone who knows the soils of the area. Having a deleterious stratum of soil under a structure can have disastrous results...or it might be a nuisance.

One must know the character of the soil before assessing whether 20% fines are ultimately an issue. In general, yes. If organic, definitely yes.

 

[SixDegrees]

You also need to consider the depth of the the foundation walls, drainage, and frost penetration, and as Ron pointed out settlements if this layer is compressible and loaded. A Glacial Till can have 20% fines and is a perfectly suitable bearing layer for a three story house.

 

[cvg]

20% fines would not be considered "very silty".What is the other 80%? If you haven't already, why don't you call your geotech and ask him this question?

 

[Gercivil]

Additional information: These 250 houses have a raft foundation. Top level of foundation is -0.4m below the proposed ground level. The project site is a reclaimed area.

Thanks for the reply:
@Ron: I inspected some of the previous projects wherein some of the houses tilted about 60mm to 70mm due to bad soil.

@Sixdegrees: Thanks for the advise..however frost in not considered in this region.

@cvg: Other boreholes had a 28.8%, 36.5% passing No.200 sieve size. We are planning to provide 22trial pits at depth 3.0m and check the sieve analysis at every 0.5meter.
There was a problem before with the Geotech and we have some doubts of their recommendation.

Is there any provision for a safe depth of fill if 1 to 3m below is a silty material or unsuitable material for low rise structures?

 

[Mccoy]

Is there any provision for a safe depth of fill if 1 to 3m below is a silty material or unsuitable material for low rise structures?

Sorry if I'm being obvious but the only a priori safe provision would be to remove all the compressible material.

 

[oldestguy]

Since the site is apparently filled material, apparently loose. If you see existing houses tilting, you need to look more at the degree of density of that material from the bore hole information. Is the stuff loose, firm or dense? That information that tells something about compressibility and shear strengths should control what you do rather than worrying about silt content.

Does not your geotech report offer recommendations on what to do there? It's difficult for members here to advise without much better info on what the material is there, in addition to silt content.

 

[Gercivil]

Mccoy & oldestguy:
I will discuss with the Geotech regarding the compressibility and shear strength of the soil...

We have one project same Geotech advise to remove the 3m depth loose soil but since it is a mass housing project the mentioned method is costly and the project is still pending.
As an alternative, the contractor proposed to remove 0.5m of loose soil and backfilled it with suitable material and the same Geotech approve the proposal?

Thanks for all the advise.

 

[dik]

If the geotech report suggests removal of 3m, then that's the prescription... unless you have other information from a competent geotekkie. It is difficult to get any degree of compaction at deeper depths... compaction effort, depending on the soil is typically effective for the top foot or two...

If the geotekkie suggests removing the top .5m and compacting the fill (type and degree of compaction as stipulated) and this is suitable for limited bearing, then you have something to design to.

Make sure the geotekkie suggests a maximum deflection/deformation and if the owner is happy, then all is (almost) well... else, maybe the project is 'canned' or find a new site.

Dik

 

[Mccoy]

Gercivil,
if total soil removal gets costly, then it could be worth to carry out a detailed study of compressibility of the surface silt, complete with settlements model of a double layer composed by superficial engineered fill and original silt.

To be more clear, that's a problem of optimization.

You would need edometric modulus of the silty soil and you can measure elastic modulus for the engineered fill layer by plate load tests.

Lab edometric moduli may be too conservative though, I worked out a similar problem recently by using Marchetti dilatometer tests.

The final result would be the optimum thickness of fill necessary to ensure an adequate degree of safety from excessive settlements.

 

[BigH]

If I could toss out another idea . . . and I am assuming that the water table isn't near the surface (that would cause additional problems on a deep excavation). Why not excavate the top 500 mm as "sacrificial" and pound the 'ell out of the material (a sand with some silt?) you have below with a heavy vibratory roller. The structures will be on a raft foundation - and I presume that you will be thickening along the wall lines and the like - so the raft should either float over possible loose zones that are left or if there is a tilt, it will be uniform one way or the other.

Another potential consideration is to dig with a backhoe some trenches logically spaced and fill with available fine rock fill - place in lifts and hit it with a hoe-pack. This, in effect, would act as "stone rows" and reinforce the footprint of the structures. - you wouldn't need to do a mass excavation in this regard. (like stone columns)

A third possibility would be to bring in some fill (say enough for 10% of the housing footprints) and preload the site (make sure you go a bit more over the average load of the raft loading). Confirm any settlement by settlement plates. When one area is done, then move the fill over to another area and start building the houses where you had preloaded.

Just a few thoughts . . .

 

[dik]

It may take a long time and an awful pile of mud to get the consolidation you are looking for...

Dik

 

[Mccoy]

I like bigH's options #1 and #2.
#2 may be tricky if silt is really loose and stability of trench walls ensue, otherwise sounds like a good remediation measure.

#1 is maybe the less costly option, you might have some quick penetration testing done (by dynamic or static means) before and after compaction, in a few representative verticals, then decide, if compaction actually occurred than 0.5 meters of compacted rockfill should suffice on top of the improved soil for that kind of structures.

Plus you have some hard data to verify it all .

 

[Gercivil]

Mccoy: For this testing the result will be for the predicted long term settlement?
BigH: In our previous project the contractor proposes #1. Before backfill of 0.5m good material we compacted the site using heavy vibrator roller.
The Geotech calculation for the settlement is only for the immediate settlement which is 23mm max.
Not necessary to check the Primary and Secondary consolidation settlement?

 

[BigH]

Again - we don't know where the water table is . . . as this material is really a silty sand (based on the description - assuming an SM soil) I don't think that consolidation settlement will be an issue. These really become important when the soil skeleton don't take the load immediately - the porewater does.

 

[Mccoy]

The Geotech calculation for the settlement is only for the immediate settlement which is 23mm max. Not necessary to check the Primary and Secondary consolidation settlement?

As bigH says, if unsaturated, the soil won't be subject to 'consolidation' phenomena entailing water expulsion unless deformation is large, depending on water content.

Besides, if saturated, the soil may posses a relatively high permeability so that water is expelled quickly and long-term settlements do not ensue. It may not though.

Is that 23 mm value based on lab or in situ tests? And which ones?

 

[Gercivil]

*The water table is 2.0m below existing ground level for the two projects - Reclaimed

Previous Project/116 houses: 0.5m soil improvement
Geotech 1st proposal remove 3m depth of loose soil but costly.
Calculation of settlement after the in situ test(plate load) prior to the proposal of the contractor......BigH/option #1 & #3(only one area was tested)

Current Project/250 houses: 0.3m soil improvement