Battered piles under seismic conditions 3

[ludvik]

I have recently been told that battered piles do not perform as well as straight piles under seismic conditions. Why is this?

The only justification that i have been able to find is that battered piles are stiffer laterally than straight piles and thus attract more load. They have greater lateral strength also obviously, so is there something else? Pretty much any foundation system that adds lateral strength will add lateral stiffness.

I have a bridge with very heavy superstructure and substructure that is crossing a swamp, and the top layer of soil will provide very little lateral support to the piles. Seismic will probably be the governing condition in the design, and battered piles seem to be the obvious answer.

Can any of you geotechnical experts shed any light on this for me?

Thanks,
Michael

 

[Focht3]

Where is your structure? (Aside from crossing a swamp...)

While it is possible to build a structure that is strong enough to withstand "any" strong ground motions, this is generally quite expensive. Flexible solutions can be made as safe - even safer - at less cost.

Keep in mind that the ground motion usually has a relatiely high frequency, while structures have much lower natural frequencies. For a structure supported by a flexible foundation (in an earthquake with high lateral and "small" vertical motion), the ground shakes rapidly - while the structure "rides out" the event. Very little of the ground motion is communicated to the upper parts of the structure. But when the foundation is stiff, the upper portions of the structure take a beating.

An example of this approach? The county building in San Bernadino, California. If I remember the details correctly, the structure is sitting on "sandwiches" of steel plate (and other "stuff&quot that give it essentially no lateral "strength." It has seen some moderate shaking, and I believe it had to be recentered once or twice. But no appreciable damage to the structure.

I'd avoid batter piles. But if you are in an area with deep soft soils, you need to look hard at the behavior of areas of the Bay area during the Loma Prieta quake as well as the Mexico City earthquake (1986?). The soft ground can attenuate the ground motion - causing the ground to shake at much lower frequencies. A real potential hazard -



Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[Qshake]

While I don't disagree with your overall conclusion, I do take exception to the note on frrequency of structures and ground. Typical earthquake frequencies are from less than 0.5 Hz to 20 Hz. Similarly, bridges are usually within that same range.

If on the other hand, you mean to say that earthquakes have a high frequency content I would agree with you.

 

[BigH]

There was a good paper, a long time ago, in ASCE Geo Journal although it might have been in the old SMFE Journal. They talked about the pile caps going through very rough times for battered piles (experience in 1964 Nagata(?) earthquake). Had a couple of good pictures showing the damage. I believe that the damage was related to small differences in the "time" the waves hit the pile - being battered, the wave hits at very slightly different times. In Vancouver (circa 1986) the idea is to use straight piles and "ride out the lateral movements" - use seismic bolts to keep the beams on.

Anyway - that's my thought of the day!!

 

[ludvik]

Thank you for your responses.

My bridge is just outside of New York City on long island, so its not in a major seismic zone. It is however classified as a critical bridge and has to be designed for a 2500 year earthquake. We will most likely have site specific ground motions to work with, and hopefully someone will think of the issue of soft soils affecting the frequency components.

Another problem with battered piles that was pointed out to me by someone I work with is that they develop tension under lateral load on one side. If you already have an uplift problem because of overturning type effects, this might be critical.

Thanks again.

 

[Qshake]

Please note that is some seismic areas as well as some aseismic areas governing agencies prohibit tension on piles under any circumstance. While this might result in a slightly larger pile cap, it may well be worth the effort to mitigate this type of failure. As part of the belt and suspenders, those agencies use positive connection means by way of channels or studs on the pile.

 

[Focht3]

It sounds like battered piles are out of the question for your site. For lots of reasons...

If the site is on Long Island, the soft soils may not be that thick. And the piers may bear in granite - so attenuation may or may not be an issue. You need to work very closely with your geotechnical consultant on this project.

This sounds like a major structure to me since you have a 2,500 year earthquake requirement. You will have a lot of people second-guessing your decisions.





Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[ludvik]

Yes Focht3, there will a whole slew of people second guessing my decisions. In fact i am more a second guesser than an original decision maker in this instance (i'm just doing the seismic analysis right now). It is a pretty big project. Its a 2200 foot long concrete box girder viaduct structure, with 9 spans.

The primary engineering challenge in the project is dealing with the seismic effects of the enormous mass of the concrete box and the piers on the foundation. The bedrock is actually ~700 feet below the surface at this site, so I guess there will be a fair amount of attenuation/amplification/frequency shift type effects going on. The top ~60ft are gravelly/silty sand.

 

[Qshake]

Where I work we see alot of these structures in many places. One thing that certainly doesn't help is the concrete choice for superstructure. As you noted you will have a lot of mass to worry about.

Second is the monolithic connection from superstructure to substructure which is quite congested. Of course, if you're using a girder on bearing on substructure connection then that's a different story. If the latter is the case, then perhaps uncoupling the superstructure from the substructure via a seismic bearing is best and will reduce lateral moments and shears on the foundation.

Thankfully, you have all the resources and new ductile details for the top and bottom column connections so that won't be a problem. And with them you can design for the plastic capacity of the foundation rather than an elastic response spectra, which for the 2500 year event is likely to yield very large loads.

Best of Luck to you.

 

[Focht3]

So much for bearing the piles on bedrock...


Let us know what the analyses tell you - a very interesting problem.



Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[novs]

I have talked to engineers from Caltrans about this topic.

It is my understanding that two of the reasons they would rather use larger diameter shafts/piles instead of batered piles are the following:

1. Battered piles are too rigid in the laterial resistance. This restricts lateral displacement and the release of seismic energy.

2. Loose sand when settling after a siesmic event will push down on battered piles. This may cause possible bending of the battered pile and pull out (cracking) of the pile cap.

This being said...the one job that I have seen Caltrans use batter piles, they designed for extra reinforcing in the pile cap around the batter piles to help minimize any cracking.

I hope this helps.

 

[ishvaaag]

The main reason I remember have read against the use of battered piles in foundations is that the seismic dynamical concentration of forces at the node hardpoints has been seen to smash the node, then ruining -even if partially- the foundation. But what said above also stands.

 

[ludvik]

Novs: Sands settling on the battered piles is an aspect that i haven't heard of before, though it does make sense. We did a liquifaction analysis, and in most areas it was fine, but in limited areas i think we concluded that the soil was only marginally adequate for liquifaction, so i guess there could be some potential for battered piles being loaded in an untoward fashion.