Deep Foundations - Bending Stresses

[tq3610]

I have a project that consists of a building supported on deep foundations. Lateral loads are planned by the structural engineer to be resisted by the grade beams. As part of the lateral load resistance design, the grade beams will be extended to as deep as about 8 feet. The grade beam depth results in rather significant lateral deflection required to engage full passive pressure.

My question is, would you typically go as far as to evaluate the bending stresses in the vertical (compression only) piles given the magnitude of lateral movement? In the past if not designing the deep foundations to resist lateral loads I would not go to the point of evaluating pile stresses due to lateral deflection, but this particular project may lend to doing so given the possibly greater-than-typical lateral movements?

 

[BridgeSmith]

The substructure components will all have to accommodate their anticipated movement. Those elements that are part of the seismic resisting system will have to be designed for the seismic forces and displacements they will be subjected to, maintaining their vertical load carrying capacity through ductile or elastic behavior. The acceptable level of damage to the substructure elements from ductile yielding should be discussed with the structural engineer and the owner.

 

[oldestguy]

Then comes the question. Is the total resistance equal to the sum of the individual components? Usually these are somewhat close together and then what goes on?

 

[EireChch]

tq3610 - give us a plan and cross section of foundation and the ground conditions.

Are you definitely talking about bending stress or bending moment?...I suppose they are interrelated in some way.

If you know the deflection required to achieve lateral resistance, input that displacement into a program like LPile. That would determine the bending moment and shear force along the pile length. This can be converted back (i assume) to bending stress?

 

[Lomarandil]

EireChch -- yes, bending stress and moment are two ways of talking about the same thing.

TQ -- I'd absolutely consider the lateral stresses in the pile. Whether or not the structural engineer intends to attribute any lateral load resistance to the piles, if they're along for the ride with the grade beams generating passive pressure, they're going to generate soil lateral pressures of their own. I don't know of a way to detail a pile top connection as a "roller" -- so you'll end up with at least a "pinned" connection carrying some shear (reflecting those soil pressures) at the pile top. And pile top shears can often generate significant moments in the length of the pile.

Not to create worry -- for common proportions of pile/soil stiffness, you often end up with something around 1-2% longitudinal steel. So a very reasonably solved problem. But you'd be amiss to say "the structural calls this a 'compression-only pile', so let's use 0.5% minimum steel (or an unreinforced pile) and be done."

----
The name is a long story -- just call me Lo.

 

[MotorCity]

Why are the grade beams so deep? If you are not engaging passive pressure, your pile-gradebeam arrangement will act more like a portal frame in which case you absolutely need to consider flexure in the pile.

 

[tq3610]

Grade beams are being extended to whatever depth is needed to provide the necessary lateral resistance rather than installing battered piles to resist lateral. At least that is the current path the design is considering. My understanding is that they've determined extending the grade beam is more cost effective than piles for lateral resistance.

 

[Settingsun]

But you'd be amiss to say "the structural calls this a 'compression-only pile', so let's use 0.5% minimum steel (or an unreinforced pile) and be done."

I think you'd get away with it at ultimate limit state provided the piles are sufficiently ductile: to meet seismic requirements as HotRod said; and to satisfy the lower-bound theorem you'd be relying on. You might get wide cracks at service loads though that could compromise durability (assuming concrete pile). Strictly speaking, with a zero-moment assumption the grade beams would also have to resist an additional horizontal force of pile axial force * pile slope. The grade beam designer probably doesn't want to know about that.

Doing it 'properly' would be interesting but a bit of an exercise if you're not in charge of the whole design and the pile stiffness is not insignificant compared with the passive resistance to the grade beam movement. Requires iteration to determine best split between the grade beam and pile resistance. If you just take the movement reported by the structural engineer, you can run into the trap that you increase pile size (or increase reinforcement) to take the resulting bending moment, which increases the stiffness, which increases the bending moment, which means you need a bigger pile etc...