soft clay soil under residential strip footing

[gte447f]

Structural guy here with a few questions. Thanks in advance for any advice.

I designed a CMU pier and footing plan for a residential renovation involving adding a second story to an older house. The design was based on an allowable bearing pressure of 1500 psf. Now the contractor has decided to build a continuous CMU foundation wall on a strip footing, which is fine, but when the inspector came out to inspect the footing excavation prior to pouring concrete, he apparently said that the footing subgrade was too soft. So, the contractor proceeded to dig deeper looking for firmer ground. They are now down to 5 ft deep and as much as 7 ft deep in places. The inspector came back out and apparently didn't mention the soft subgrade anymore, so presumably he is satisfied that they have reached firm enough material, but he said, since the footing no longer matches the plans, he needs a revision from the engineer (i.e. me).

I am getting to the point of my question, so thanks for bearing with me. The contractor described to me that the original excavation was soft in the top 2 inches or so, based on sticking a piece of rebar into the soil, but was firm below 2 inches. So they would excavate a couple of inches (with hand tools), but then the next 2 inches would be soft, and so they would repeat the process, until now they are 5-7 ft deep. My question is why is the top 2 inches of the clay in the excavations always soft? Is it because of lack of confining pressure from the soil that has been removed?

Also, they say they have reached firm ground, but I am skeptical that I may arrive onsite to discover the same conditions (i.e. a couple of inches of soft clay in the excavation with seemingly firm material underneath). If so, what should be done? Should gravel be placed in the excavation? If so, how much?

 

[BigH]

Without knowing hardly any facts, it might be that the soil excavation at the outset was lower than the groundwater level and the softening was due to the seepage effect at the excavation surface. It is too bad that you don't have any data at the outset excavation level to confirm consistency below the "several inches" ofsoftened material. 2 to 3 inches would have little real effect on the foundation. (Could have used a Mackintosh Probe or the TRRL miniature dynamic cone to show saisfactory consistency at depth.) This might have saved a lot of excavation

It appears that you had no geotechnical invesigation so I am wondering who the inspector was (independent or city employee)? Ifthe city inspector, what are his qualifications to make such judgements? If you could prove that my earlier premise was valid, will the inspector take liability for the addirional costs?

Now the use of "gravel" - you could excavate so that a depth greater than 4x the footing width at the founding level and a width each side at the base. Then place the engineered fill and since the footing stresses would be within the engineered fill, problem is negated. Of course you should seek a geotech's advice on such remediation. There was a paper back in the late 70s in one of ASCE's geotechnical specialty publications that discusses such action.

 

[concretemasonry]

Keep in mind that a CMU foundation wall is a deep beam. In most cases, because of the loads, a strip footing's main purpose to to provide a good surface to build the masonry wall. It also generally is much wider (8" to 12" wider) than the wall because this allows the final wall to be to be built square irregardless of the accuracy of the layout of the forms. Most strip footings do not really see much flexure in either direction and it is really shear that usually works even for unreinforced strip footing because everything really is in shear. Generally the reinforcement is there to provide continuity and provide a reason to follow the plans.

I have seen two story wood frame home built on sites the variable soils and possibly some water. Usually a good builder and mason contractor will excavate the entire footing areas to a constant elevation, even if some of the soil is adequate. The reason is continuity and ease of construction even if there is more masonry or concrete wall surface and a little more material. The global benefit is the uniform soil pressure from the rigid wall/footing combination.

Dick

Engineer and international traveler interested in construction techniques, problems and proper design.

 

[gte447f]

BigH,

Your theory about the ground water level and seepage at the surface of the excavation sounds plausible, since we have had a ton of rain here in the last few months (about 35% above normal for he year so far). I would have also liked to have gotten some data before the extra excavation was done. The inspector was a city inspector, but in his defense, I don't think he instructed them to excavate to 7 ft; they made that decision on their own.

With regard to your comment about showing satisfactory consistency at depth, I have a Sowers dynamic cone penetrometer, so my thought is to use it to test for satisfactory consistency at depth. I would use Sowers' curve to correlate to SPT N value and consistency of the clay. Does this sound reasonable for a lightly loaded residential strip footing? If so, my next question is about the depth to test with the DCP... I thought the target zone was approximately 1 times footing width below the founding level based on the alpha angle of 45+phi/2, but in your comment above about engineered fill (e.g. "gravel") you seem to suggest 4x the footing width as the depth of influence of the footing. Can you please explain the 4x width of footing? Thanks.

 

[gte447f]

Dick,

Thanks for your input. I agree with your comments. I will spec a handy amount of rebar in the footing just because, but I am not really concerned about the footing stresses. I am just slightly concerned about settlement of the foundation wall after being told the subgrade is "soft".

Here's a CMU question for you... do you recommend grouting solid below grade for durability?

 

[concretemasonry]

Grouting below grade has a simple answer - It is very rarely necessarily needed, except for the reinforced cores in the masonry. Grout really does not add much to the flexural strength on the wall and it rare that the loads from thinner member above grade cannot put enough vertical load on a foundation wall to justify it structurally.

It does not add durability and arbitrarily filling all cores eliminates the befits that cores can provide. That said, some contractors will arbitrarily pour concrete into the entire wall just because the ease offsets the material cost instead of properly grouting the reinforced cores.

Dick

Engineer and international traveler interested in construction techniques, problems and proper design.

 

[gte447f]

Thanks Dick. Alright, back on topic now... BigH?

 

[BigH]

There are two considerations - one based on shear and the other based on settlement ( i.e., the depth of influence of stress increase in the underlying soil). For shear 1x footing width (or slightly less) is fine. For stress inducement in underlying soil, for square or round footings, the depth of influence is about 2x ( depth where stress increase is 10% of applied loading); for strip footings, the depth is in the order of 4x (L/D > 10)- can interpolate in between.

Fo penetrometer, the Sowers would be acceptable. What you are looking for is "constancy" from footing depth. As I believe I said, if consistency is "good" a fe inches down, just place tne footing. If the consistency is "poor", then the trench approach can be considered. What is good? What is poor? Correlations as you suggest where N is the gaged against Su is likely good enough butwith the need forsome judgment. (Many geotechnical engineers will cringe at correlating an N value to Su - I am not one of them as I consider the N value to be a tool where experince and considered thought allows one to judge its applicability.)

Anyway, theses are my thoughts.

 

[oldestguy]

This reminds me of my working with a contractor who had lots of experience. His experience was mainly running things from the office. His subject job had a basement excavation that became very soft when a skid steer was running about on it. The site was a clayey silt, being probably more silt than clay. I suspect this site of the thread is similar. With men walking back and forth doing the digging, the slight disturbance became apparent as "soft". Not really soft, but weak. The more you disturb it, the worse it gets.

It is common in my area. Routinely when I would get a call from the field saying "Hey OG, we are getting stuck while doing excavating, or some other activity". But before working, the site was stable. I'd reply, "Stop and get your front end loaders off the site. By tomorrow things will be stable again." They sometimes would wonder how I knew they used front end loaders.

They usually would not have to call back, since it fixed the problem. By recognizing this behavior for saturated silty soils, you can get the job done.

The case mentioned at the start here was quite unusual for an "experienced" contractor.

My explanation is that the voids are filled with water which was trying to get out making for an excess pore pressure and thus reduced the intergranular pressure, making for weak shear strength, temporarily. Placing some "breaker-run" rock can help.

 

[Ron]

Both BigH and CM make excellent points; however, keep in mind that even if the footing doesn't flex, it can still settle, so BigH's consideration would be more relevant to the issue.

It sounds as though your soil is highly stratified with clay lenses. If the clay lenses are stratified with sand lenses, consolidation settlement will occur faster, but not quite as fast as elastic settlement. If stratified with silty material (assuming non-organic silt) then bridging above the water table is likely.

You don't state your location, but considering your mention of the Sowers penetrometer, my guess is SE USA, as it is more commonly used here than other locations. If so, the correlations are pretty good, so no problem using them as BigH noted.

 

[fattdad]

Regarding disturbed soil at the subgrade: Sure, some measure of undercut is often warranted. I mean there are many stable soils that are saturated (i.e., way over optimum moisture content) and the simple trauma of digging can lead to disturbance. Should this lead to the "China Syndrome?" Well, not to me. . .

I first want to echo BigH's comment on the water table. Now we are not talking about optimum moisture content - as you dig deeper and deeper, you are creating the potential for a critical gradient to develop and that can lead to continuing disturbance, irrespective of how careful you excavate.

If the concern is not related to the critical gradient, then you may just have a soil that's easily disturbed. Should you keep digging and digging? No! At some point, you should be able to remove enough of the natural subgrade and construct an engineered subbase. I mean if the footing is scheduled to be 2-ft wide, the likely seat of settlement is going to be about 8 ft. So, if you have 6 ft of open-graded aggregate as backfill for the overexcavation, you'd end up with quite a stiff bearing material in proximity to the bearing surface. Will some of the stress get to the underside of this engineered subbase? Sure, but you should be able to assess the contribution of the subbase stiffness and the settlement that will develop from the natural soil that remains.

f-d

¡papá gordo ain’t no madre flaca!

 

[gte447f]

Thanks everyone for some great responses. I have some updated info after visiting the site. The footing excavations vary between approximately 4 to 6 ft deep. The soil is basically saturated clay. There is a thin layer (maybe a couple of inches) of very plastic clay in the bottom of the excavations. I think this is primarily a byproduct of the saturation and the fact that this clay at the "surface" of the excavations has been worked a lot by foot traffic since these excavations were dug by hand with guys in and out of the trench a lot. I beleive this is what OG was referring to above.

I could probe the bottom of the excavations 6"-12" with a #4 rebar using 2 hands to push it down. I tested 2 locations with the Sowers DCP. At each location and each depth that I tested, I seated the cone 2" with slide hammer blows, then did 2 successive 1.75" penetrations and recorded the number of blows for each 1.75" penetration. Here are the results:

Location #1
@ bottom of excavation (-4 ft): seat,2,3
@ 1 ft below bott of excavation (-5 ft): seat,5,7
@ 2 ft below bott of excavation (-6 ft): seat,5,7

Location #2
@ bottom of excavation (-5 ft): seat,4,5
@ 1 ft below bott of excavation (-6 ft): seat,6,7
@ 2 ft below bott of excavation (-7 ft): seat,7,9

Aside from the (2,3) recording which was in the plastic muck, the rest of the Ndcp were 5-9. Per Sowers curve, this correlates to 5-7 Nspt, and soft to medium stiff clay with qu of say .5 to 1.0 ton/ft^2 or su of 0.5 to 1.0 kip/ft^2. Does this sound reasonable?

I also did 4 Torvane tests on the vertical sides of the excavations near the bottom. The results were: Su = 0.4 ton/sf, 0.4 ton/sf, 0.5 ton/sf, and 0.6 ton/sf. These are in the range predicted by the DCP-SPT-consistency correlation above.

If I estimate q(all)=q(ult)/SF=(5.7*su)/SF, with Safety Factor (SF) equal to 3, I think I am satisfied that the soil will meet 1500 psf.

 

[gte447f]

As far as giving the builder some direction to finish these footings, I am thinking of 2 options:

1) remove any plastic muck from the bottom of the excavations, then place 12" of #57 stone in 4-6" lifts compacted with a wacker packer, then place the strip footing

2) remove any plastic muck from the bottom of the excavation, then place 36"-48" of #57 stone in 4-6" lifts compacted with a wacker packer, then place the strip footing

Option 1 addresses the bearing by using the 57 stone just to make sure there is a load path thru any plastic soft stuff at the surface of the excavation.

Option 2 addresses the bearing and confinement settlement by replacing most of the zone of influence with "engineered fill".

I like option 2, but I have seen many topics on these forums regarding long term settlement of 57 stone if it is not compacted. I know there are two sides to this discussion with some believing that 57 stone is essentially self compacting, but I personally don't believe that it is. I also don't trust this builder to place the stone in lifts and compact it, so I am leaning against option 2, unless they agree to pay for observation while they are placing the stone.

Even without replacing the entire seat of settlement with compacted fill as in option 2 above, my gut tells me that this footing will be lightly loaded enough that settlement shouldn't be an issue, but I may do some calcs just to be sure.

 

[BigH]

I really have never been involved wit 57 stone as I worked in Canada, but why not just use a road subbase material?

 

[darthsoilsguy2]

the geotech post-mortem you did looks like they may have been fine to proceed.... or should have contacted a soils firm at the least before proceeding.

The first thing i would check is the original existing conditions for the site to see if this pad was fill, if so how difficult to fill, and if so whether there were any drainage features running through it that might have also made difficulties. if it is fill, although the numbers you got are in the acceptable range, i would take the under on over/under betting that specified compaction was achieved.

I hold an extreme prejudice against #57 stone. i prefer to use it only when drainage through the fill is a desired affect. in most cases that #57 is used, i prefer compacted road stone (a.k.a. Crusher Run, Crushed Run, ABC Stone, GAB Stone, ...)