Negative Skin Friction

[tq3610]

Hello All,

I've got a project that has about 45 feet of granular soil underlain by overconsolidated clay from 45 to 80, and then normally consolidated clay from 80 to 165 feet (rock at 165 feet). The project will include placement of about 3 feet of fill over the site, and then a new building.

Given the 3 feet of new fill, I'm estimating the negative skin friction on two pile types (steel H piles and precast concrete piles). I have load test data from an H pile project directly across the street.

My question is: is there any technical reason not to use the available load test data to determine "site specific" skin friction, and then apply that total skin friction load to my project as negative skin friction? In other words, use the static pile load test results that will provide me with positive skin friction and apply it as a negative to my site? Given that they would be piles end bearing on rock, the negative skin friction would be applied to the full length of the pile.

Additionally, intuitively it would seem excessive to apply the NSF to the full length of the pile (165 feet) given only 3 feet of fill (if using the nearby load test data, we would be talking about approximately 125 tons of negative skin friction), but given the overall area of the new fill and looking at stress distribution with depth, I expect there to be some amount of consolidation the full depth of the clay layer.

Any insight would be much appreciated.

 

[EireChch]

i will take a pop at this.

If you place the 3ft of fill, the granular soils would settle elastically during the placement of fill. And by the time the stress distribution reaches the OC clay the net increase in pressure would be minimal which would result in minimal consolidation plus as its already OC it i would expect it to be minimal.

I dont think you need to consider NSF as the 'downward' forces have occurred prior to you piling.

 

[tq3610]

Thanks for your input!

Given the extents of the fill (for reference, let's say it's a 350' x 350' area with a building of about 150' x 150' near the center), I was considering the fill as a flexible load at ground surface. For that extent of fill, there is still some applied load down at the deeper depths (say, 0.4 x the fill load ballpark).

Regardless of that, do you have an opinion as to the use of a pile load test result for application to NSF in general?

 

[EireChch]

My first question would be are the soils the same? I dont know your geology but ive seen soil conditions vary greatly over a couple of meters. Well assume that they are so in that case i would think it is acceptable to use what ever information you have at hand once you can justify it.

The results of a static load test is more or less a graph of load vs settlement ? how to you propose to split the capacity up into end bearing and shaft capacity?

 

[tq3610]

The soils are the "same", yes. We have a lot of borings over the overall extents of the site and nearby vicinity.

The load test results are a graph of load v. strain. If you plot the load v. strain of the steel on the same plot, you can determine at what point all the load is transferred to the tip of the pile (i.e., when the deflection measured at the pile top is equal to the theoretical steel strain over the length of pile, then all the load above that point has been carried by the soil friction and any load after that point is carried by the pile tip).

I also have dynamic pile test results for the same pile that separates the side friction and tip bearing capacities.

I guess my simplified question is "is it technically correct to apply the positive skin friction value to negative skin friction." Per studies, for an end bearing pile the negative skin friction would be applied to the full length of the pile within or above the consolidating soil layer.

 

[EireChch]

Interesting idea. It would appear that you know more about the topic than i do so i am afraid i wont be able to answer your question! Hopefully some more experienced engineers can depart with their knowledge and help.

 

[SlideRuleEra]

bnereson - Am I reading this right... the rock is at a depth of 165 feet, and the piles will be steel H-pile or precast concrete and bear on the rock?

165 feet long H-piles... maybe with welded slices as driving progresses. What is the proposed size of these piles, and what hammer specs?

165 feet long precast concrete piles... I don't think so.

http://www.SlideRuleEra.net

http://www.VacuumTubeEra.net

 

[tq3610]

There will certainly be splices, that is given. There are numerous buildings surrounding this site supported on H piles and PCPs to depths of 140 to 190 feet, including a building within a couple hundred feet from this building that was constructed in the 2000s and is supported on 150-170 foot PCPs.

We haven't determined the pile sizes yet - that will depend on whether we need to account for high negative skin friction load. But a nearby building is supported on HP14x117s. That size may be a bit larger than we need for this building.

 

[Retrograde]

is there any technical reason not to use the available load test data to determine "site specific" skin friction, and then apply that total skin friction load to my project as negative skin friction?

I think if you did that you would be grossly overestimating the amount of negative skin friction.

You need to get some geotechnical advice. Given your soil profile I would not be surprised if the level of NSF would be small enough to be ignored. I agree with EireChch's initial reply.

 

[tq3610]

Thanks Retrograde. As some additional information, the neighboring building on PCPs was designed based on 1 foot of fill and 20 tons NSF.

 

[Retrograde]

the neighboring building on PCPs was designed based on 1 foot of fill and 20 tons NSF.

How did they arrive at this value of 20 tons?

 

[HENRYZAU]

I am a bit sceptical that at such a depth the clay is still NC. How was its yield stress assessed? Or did you know it's NC?

 

[tq3610]

Retrograde - I don't know how they arrived at that value. Their report said NSF could be as much as 100 tons, but they stated to minimize new fill to 1 foot and thus carried 20 tons. Regardless, I'll need my own design not based on some other engineer's design.

As far as the clay, the upper OC layer has N values around 5 bpf. The lower NC clay is WOH or N=1 bpf. Triax testing was performed.

 

[dik]

How large a structure? Can the foundations be placed in the granular material?

Dik

 

[tq3610]

Support in the upper granular soils with expected settlement was discussed but the owner did not want to proceed it that direction. Column loads up to 650+/- kips.

 

[Retrograde]

I am a bit sceptical that at such a depth the clay is still NC.

Not being a geotechnical engineer I have always wondered how you could get a normally-consolidated clay layer under an over-consolidated clay???

 

[tq3610]

There may be other reasons, but one reason is that in a past geologic age when ocean levels were lower the clay was exposed to the sun and became desicated resulting in a stiffer crust.

 

[dik]

I would think, and I'm not a geotekkie, founding in a competent granular material that is 45' thick laying over overconsolidated clay would make a reasonable foundation layer.

If the client doesn't want it, then go all out... it's his money. But, we don't want another San Francisco tower that is slowly sinking into the sunset...

Dik

 

[Retrograde]

There is some good information here:

http://www.fellenius.net/papers/095 Singapore_1984.pdf

 

[tq3610]

Thanks Retrograde. I have read that report, but in re-reading it I have noticed a statement I had previously overlooked:

"For reasons of simplicity, the shear stress along the pile is assumed to be independent of the direction of the displacement, i.e., the negative skin friction, qn, is equal to the unit positive shaft resistance, rs."