Shear Resisting Mechanisms for Piles & Caps

[KootK]

The Canadian concrete manual has a supplement where they discuss (very briefly) the mechanisms that can be used to resist shear delivered to a pile cap assemblage. I've encountered this in the context of working on a shear wall foundation system.

Anyhow, they suggest four possible mechanisms:

1) Friction between the pile cap and soil below.
2) Passive bearing of the pile cap / shear key against the adjacent soil.
3) Passive bearing of the piles themselves against the soil.
4) Bearing of the pile cap / shear walls against the adjacent slab on grade (where available).

After listing these mechanisms, they also state that "all four mechanisms may not act concurrently". So my question is this: Which do act concurrently and in which order do they engage? I'd appreciate any insights that others may have. Additionally, does anyone know of any good references that deal with this topic?

Some of my thoughts on the issue are:

a) My guess as to the order of mechanism engagement would be as follows: 1) Friction 2) Friction gives and SOG bearing takes over 3) SOG bearing gives and passive soil bearing (piles and pile cap concurrently) takes over. Following this logic, the piles and pile cap bearing against the soil would be the only mechanisms to act concurrently. Sound reasonable?

b) If you decide to use passive bearing of the pile cap against the soil, is it wise to use a lean concrete backfill between the pile cap and the extents of the excavation to ensure proper bearing against undisturbed soil?

c) Is it really appropriate to use friction between the pile cap and the soil? With piles absorbing the bulk of the gravity load via end bearing, wouldn't the amount of friction developed between the pile cap and the soil be quite limited (and difficult to predict). My thinking is that the soil under the pile cap would tend to settle away from the underside of the cap over time. Thoughts?

 

[ishvaaag]

When you need to use piles usually the soil will be weak and friction in general an unreliable term. Hence, in general way only passive push and -if available- bracing from a SOG against say a rigid supporting soil or mountain wall at the other end of the building. So we would be in a kind of P-Delta situation against the piles, the bracing action of the SOG and passive resistance of the soil equilibrating the lateral displacement of piles.

 

[ChipB]

Soil settles away from pile cap, thereby negating any frictional contribution of the pile cap/soil interface.

Slab on grade may be removed for some other expansion in the future, thereby negating any contribution there.

They may regrade around the pile cap, cut and fill around the pile cap, or any number of other variations which would affect the passive soil conditions around the cap. Therefore, passive soil pressure can not be relied on and should not be accounted for.

Lateral pile resistance is the only value which remains and is the one that should be used.

I design mostly in heavy industrial, so others may see things differently.

 

[apriley]

In bridge design, we typically use the pile lateral capacity only, and add in passive resistance on the pile cap when needed. However, this is typically only for seismic design. For service, we use the service capacity of the piles only. You probably won't move enough to mobilize the soil to the point where you can count on passive resistance on the pile cap. I'm not sure whether it would be worth it to backfill using lean concrete in terms of cost vs. benefit.

 

[slickdeals]

I believe that using passive resistance of soil is not permitted in seismic design, correct me if i am wrong.

 

[ishvaaag]

I think to remember the Monobe-Okabe states both passive and active strengths, being an acceptable method of checking strength of walls for seismic conditions.

Thinking freely for embedded piles, it is through them that the seismic shear is passed to the building. So we have the whole piles+superestructure thrown to one side, then by vibration action back. At the moment of this back the soil will be enacting its passive action, so not counting on it would be irrealistic; other thing is that it need be evaluated at a lower value on statistical reasons and perturbed state of the soils.

Furthermore, the alternative inclined piles have in some cases buckled (I assume it will be mainly for end supported piles) and failed the pile caps, so there's no really general alternative to take somewhat into account passive strength as something that may be equilibrating a lateral force at the pilecaps under Earthquake.

 

[apriley]

slickdeals - I'm not sure about buildings, but passive pressure is definitely counted in seismic design for bridges, at least in California. CA has its own seismic code for bridges, though, and does not follow AASHTO.

ishvaaag - Piles act counter to seismic forces, so they'll resist the forces from the superstructure/wall.

 

[ChipB]

Hmm...Something for you guys to ponder:

Wind forces travel down the structure to the foundation.

Seismic forces travel from the foundation UP the structure.

Can't soil liquefy in a Seismic event? How much passive resistance does it give in this liquified state? Don't we have to design retaining walls in seismic areas against this liquified soil?

 

[KootK]

Thanks for all the helpful responses so far.

I think that Ishvaag's point is simply that, since the EQ forces are induced through the soil (as chip points out as well), it is rational to assume that the same soil is available for passive resistance.

Chip: I think that you're right for certain types of soils. If the geotech says that liquefaction is a serious problem, you probably need to start thinking raft foundation.

Ishvaag's mention of inclined piles brought to mind a related question as well. In many textbook examples, you see shear being taken by inclined piles often referred to as "Batter" piles. Mechanically, the concept is very appealing. Are these still used in practice? More common for bridges maybe?

 

[apriley]

Battered piles are used for lateral service loads. The piles don't take the load in shear, but in axial load (components). Battered piles tend to behave poorly in seismic events for many reasons, but particularly because battered piles only behave well in compression. Battered piles are used to design seat-type bridge abutments and retaining walls (earth-retaining structures), but with these structures seismic forces are different and the battered piles will not see tension.

Generally we consider seismic loads to come from the structure, NOT the soil. It may seem counterintuitive, but the mass of the structure is what causes the force.

Liquefaction may be an issue, but that's usually only an issue for the pile itself since it tends to happen in deeper layers of soil, not under 2-3 feet of fill. Scour conditions, however, would impact passive resistance against the pile cap.

 

[slickdeals]

Battered piles are not recommended for high seismic zones. Check the NEHRP commentary section for more information.

 

[KootK]

April,

Thanks for the info on the battered piles. That's helpful.

If one had battered piles on each side of an element, and only considered the contribution of the battered piles in compression, might it then be acceptable count on them for strength load cases?

For seismic:

The body forces in the structure are generated as a result of the acceleration of the masses in the structure. That acceleration is, in turn, imparted to the structure by the soil. Really, the only external force on the structure is applied by the earth against the foundation. The rest is just the internal forces developed as the structure attempts to resist relative translation between its masses and the ground.

So, if there's earth there to toss the structure around in the first place, then there should also be some earth around to count on for passive resistance. I think that's what Ishvaag was getting at. Although I should probably let Ishvaag speak for himself.

 

[ishvaaag]

KootenayKid, yes, that was more or less what I was saying, in an earthquake, the ground takes the building by the piles and then shakes it. Obviously the piles are then deflecting both ways and encountering (passive)reaction and damping on both sides. Even when starting pushing the building it is using passive push restraint, for the building inertia and rigid link to the somewhat stiff piles have them reacting the soil (as if pushing against it). With more reason this happens when once started the vibration the piles' movement is being braked by the soil: we have as well the soil reacting the thrust of the moving piles in passive push.

And of course one may have piles in soils with risk of liquefaction. In fact I checked it even for the structures of the recently named lane extension job, here that we have only .04g basic surface acceleration for normative earthquake (this maybe 8% of what coming at worst places in California), but seismic action was mandatory out of the highway being a fundamental infrastructure of the road's network. There was no problem, even in the silty soils there.

So when risk of liquefaction is a problem we have the alternative of what codes say, experience AND making an autostable foundation+superstructure outfit. Experience is what the paper on the wood piles behaving well under Hugo hurricane scouring the soils said: the wood piles behaving well if deeply embedded in foundations. This is hurricane action, and earthquake is other beast, but records of behaviour there must be. And the ship or alike outfit is available for everyone.

 

[phuduhudu]

Just to agree with ChipB, I am also working on an industrial structure and have assumed only the lateral resistance of the piles themselves causing bending in the fixed head pile which the pile is reinforced for.

 

[apriley]

Battered piles can be used successfully in service load applications, but not seismic. So if your high load case is service load, I'd say go for it - but if your battered piles end up in tension you might have a problem. To give an example - battered piles are used successfully in port applications for mooring dolphins/breasting dolphins, but the loads for these structures tend to be in one direction, so you can batter the piles in the direction of the load.

If you're not sure, I'd use vertical piles. You can count passive resistance on the pile cap if you mobilize the soil - but for service loads and low-level seismic loads this won't happen.

 

[Teguci]

I have always been a fan of battered piles. For taking lateral loads, they are very stiff and complement a shearwall bracing system very well. As for taking tension, friction piles often take an equivalent amount of axial load in both the up and down directions. Per the code, the tension piles are to be tested to twice their rated load. For end bearing piles, the uplift capacity can be greatly diminished. Any engineer looking to use battered piles should be careful about a couple of things:
- The battered pile produces a vectored load on the pile cap. The lateral resistance and the vertical resistance should not be considered separately.
- The cautions regarding battered piles in high seismic conditions is due to the production of a higher shears in the pile cap from the piles. This is never good and is to be eliminated from happening by use of the seismic overdesign factor Omega. I remember seeing a case study where a battered pile punched through a pile cap due to seismic loads.

To answer your question regarding concurrent resistance to lateral loads:
1. Axial load from battered piles
2. Bending resistance from deep piles. Use L-Pile analysis or sim (be careful with shadowing effects of piles in a row)
3. Passive resistance from pile caps (I have always included this)
4. Passive resistance from the SOG (I have always tried to not use this - susceptible to shrinkage)
5. Friction between the pile cap and ground below - NEVER use this.

 

[oldestguy]

If you fellas look at the photos in the thread "Nearly Completed building in Shanghi", you will see the piles there still are well secured in the pile caps (grade beams). Meaning there is good moment resisting potential in that connection.

Take an example of pushing a table sideways. The resistance is in the connection of legs to the top.

 

[Ltdog]

Because the soil between the piles concrection and settlement, the cap below will be disengaged from the soil.
Based on the reason, we don't tconside the friction between the pile cap and soil below.
If the adjacent soil is stiff or the backfill around the cap is satisfying the requirement of the dry density, we shall conside the the pile and its front soil may together resist the horizontal action (seismic or wind).