Driven pile ultimate capacity and downdrag

[ATSE]

Below is a static axial capacity plot for a driven pile that I just received from a nationally recognized geotechnical firm (to remain anonymous).
Soil profile is layered silty-sand then clay then silty-sand ... (near a river).
In my humble opinion, this plot is garbage and makes no sense. I am going to recommend to our project manager to fire the GE and write a new contract for another GE.
[My bias: My pile influences are primarily Fellenius and FHWA (which explains Fellenius a little bit better than Fellenius).]

Here's my questions for the seasoned geo engineers:
1) Is there any defense of the plot as shown? It appears that the GE is reducing (even going negative) the ultimate axial pile capacty due to drag forces (which he incorrectly terms downdrag).
For ultimate capacity, aren't all the soil friction vectors pointing upward?
2) Seems like a plot showing the neutral plane with a fixed pile length would be more helpful, but this is uncommon. Does anyone provide load transfer plots in their geo reports?

 

[oldestguy]

If a boring log and test values could be provided, maybe some sense can be made of it.

 

[r13]

I am curios on how to interpret the zig-zag of the compression curve. What it means when the capacity drops below zero, and again falls to zero at a deeper depth. The uplift curve makes sense though.

 

[ATSE]

oldestguy: could this plot make sense for any soil type / profile? Assume SPT values are between 5 and 45, and skin friction > 0 for all layers. Assume that the clay layers are normally consolidated. Graph is for static capacity (precluding seismic).
Here is what the graph is saying:
If the pile is 35 ft, ultimate axial capacity = 280 kips.
If the pile is 38 ft long, ultimate capacity = 0 kips.
If the pile is 21 ft long, the pile is sucked down into the middle of the earth before it is top loaded.

 

[BridgeSmith]

To me, the compression line going negative indicates that if you only drove the pile into the top silty sand layer, when the clay shrinks/collapses due to moisture changes, the layer above it sinks and drags the pile down with it - downdrag (aka negative bearing capacity). The magnitude of the uplift capacity and the downdrag force being equal makes sense, since they're both based on the friction between the pile and the soil.

What doesn't make sense to me is the near horizontal lines at 21'-22' and 36'-37', indicating that the downdrag suddenly disappears once the pile is in the clay and reappears once the pile goes below the clay. It seems to me the solid line should connect directly from where it is on the left at 21' to where it comes back to the left at 37'.

Basically, down to 37' you have no bearing capacity, and at 50' it's gone up to 90 kips. I agree with retired13 that the uplift line looks plausible. When they screw up the bearing capacity that bad, it's hard to trust any of it, though.

Rod Smith, P.E., The artist formerly known as HotRod10

 

[r13]

My guess, the compression line indicates the soil strength of each layer, so the strength can go above (strong) or below (very weak) zero. Thus, in order to know the ultimate capacity, you need to make a table to calculate the net strength for each pile length by sum the strengths and divide by the length under consideration. Don't sound user friendly, if it is true.

 

[ATSE]

BridgeSmith: Drag forces (cumulation of negative skin friction) will certainly add forces to the structural pile member (internal stresses), but do not reduce ultimate geotechnical capacity. Skin friction is an absolute term, and when we're talking ultimate compression capacity (geotechnical capacity), all the skin friction vectors are resisting the applied top load (all pointing up), and the pile is about to plunge, and the neutral plane is near the ground surface.
Downdrag is ground settlement (soil downward movement relative to the pile) and not a force.
I'm using design methodology and vernacular consistent with FHWA (I think I'm in good company ignoring the AASHTO pile method and terminology).
Also, to be clear, this discussion is just about the solid red compression line.

 

[BridgeSmith]

BridgeSmith: Drag forces (cumulation of negative skin friction) will certainly add forces to the structural pile member (internal stresses), but do not reduce ultimate geotechnical capacity.

The force added to the pile by the downdrag of the top layer must be resisted by the skin friction of the soil along the pile below the settlement zone (and/or by end bearing), does it not? Technically, I suppose you could say that the capacity is unaffected, but the load is increased. To me, it seems like the difference between subtracting a positive value or adding a negative one. The net result is the same, though; the bottom layer has to resist the superimposed structure load on the pile and the downdrag force. It appears that the graph is depicting the net capacity for various embedment depths. Once the pile has settled sufficiently, some or all of the downdrag may be mitigated. How much settlement is required will depend on the magnitude of settlement of the soil layer that is expected.

To get from the -95kips at 21' to the +90kips at 22' the pile has to move settle sufficiently to mitigate the downdrag (settle as much as the soil did) and then settle enough more to mobilize the skin friction in that top layer to resist the superimposed bearing load.

Rod Smith, P.E., The artist formerly known as HotRod10

 

[Settingsun]

I suggest asking the geotechnical engineer for an explanation. They may have valid reasons, or they may have made a mistake. If the latter, you can then decide whether to go elsewhere or whether having them 'on notice' might be enough to get properly-reviewed outputs from them. Since you have access to the project manager, see whether this was the lowest bid and by how much for future reference.

Now to speculation since we know little about the project and ground: It looks like they are relying on end bearing from 22' to 36' that increases with depth, which is then lost at 36'. There would normally be a transition zone towards the bottom which is absent in the graph. Either it has been unconservatively omitted, or the layer change is actually some distance below 36' and they've conservatively deducted end bearing at the top of the transition zone. The latter is a possibility if they don't want people to try to thread the needle by specifying toe levels in the transition zone only to have the pile over-driven later, or if they didn't do enough investigation to be sure of the transition depth across the whole site.

Only get hung up on the absolute numbers if you're a perfectionist. The graph has come out of their ground model without being touched up. The negative numbers at the top just mean don't found compression piles at that depth. Some people take the view that polishing that out of the final report is not worth it when the conclusion remains the same.

 

[r13]

Down drag arrow stops at -22' makes sense, as there is negative soil resistance from the -6' to -21' (-100 kips), then magically harder soil layer was found at -22 (+100 kips). I wouldn't trust the soil strength and drive the pile pass beyond -35' though.

 

[kieran1]

I find it very strange that the axial capacity is reduced the deeper the pile is driven?

There is almost 3 times the capacity at 35ft as there is at 50ft.

This doesn't make sense to me

Kieran

 

[r13]

I guess the line was drawn based on below count of test pile, weak soil layers were encountered at different depth.

 

[kieran1]

I'm not familiar with these capacity plots, bu how could the pile achieve the correct "set" when at 36ft the capacity is 290 Kips yet at 37ft it is zero.

I would expect some friction resistance, which completely disappears by driving another ft?

Kieran

 

[r13]

Zero driving resistance at -37' (pile advances without force), then encounter denser soil right after that point. IMO, the line needs to be correlated with boring log to determine the useful/reliable pile capacity, it does not tell the full story by its own. I wonder why the pile stopped at -50, unless they feel the layer between -27 and -36 is adequate to provide required bearing strength.

 

[Settingsun]

Is there separate treatment of settlement in the geo report, or just ultimate? I wonder whether they've done something similar to this paper, perhaps modified equation 2 so that NSF is outside the brackets.

https://www.google.com/url?sa=t&sou...FjAQegQIBhAB&usg=AOvVaw1UK7V77WJMqaJTX4rfCg7B

Also note the first sentience: negative skin friction aka downdrag. Always important in engineering (esp geotech) to remember that terms are used differently. Correct/incorrect is just an opinion when it comes to nomenclature.

 

[Okiryu]

ATSE, perhaps this is very late, but I found that graph very interesting. Attached is my interpretation of that graph. Please keep sharing interesting (unusual??) data like that with us. Thanks!

 

[ATSE]

Okiryu,
Your efforts are appreciated - thank you.
However, a pile with bearing and friction resistance having zero capacity at 38' drive depth is nonsense.

Aside: One of the downsides of an ultimate capacity graph is that it is outside the context of time.
For short term events (strength), I believe all the soil-pile skin and bearing vectors are pointing up (no drag forces). So resistance to a seismic event (no liquef) or event a 1-day flood will have high short term strength.
For long term condition (settlement below the neutral plane), the drag forces are highest and the neutral plane is lowest (fewest upward vectors). So it seems like there needs to be at lease 2 ultimate graphs; otherwise, provide allowable capacities based on events (the usual 1/3 increase would be a separate conversation). And the graphs need settlement values to accompany them.

 

[Okiryu]

ATSE, agree that the graphs should be associated with settlement values. I was just trying to understand your graph and hope you got some input from your Geotech about the considerations for that graph. If you can share with us the thoughts from your Geotech, it will be appreciated !