How to Calculate Wall Movement? (Active or At-Rest)

[3Fan]

I've read until I am blue in the face and can't find anything that helps me. I am not a geotechnical engineer, though understand the basics fairly well.

My situation:
150' single span plate girder bridge supported on semi-integral wall-type abutments that are roughly 30.0' tall founded on multiple rows of battered and vertical friction (pipe) piles. We initially used an equivalent fluid pressure of 40 PCF active earth pressure per out DOT's specifications. Now that we have provided our loads to the geotechnical engineer for their review of our assumptions and lateral pile analyses, they are recommending using at-rest pressure of 60 PCF since they don't think the abutment will move 0.1% * H. This equates to about 3/8" in our case.

How would one go about determining what the deflection of our abutment is? Can we just assume that for instance the lateral pile analysis show the piles actually deflect say 0.01’. Can you draw a line between the point of 0 movement on the pile up to the top of the deflected pile, then continue that line up linearly to the top of the abutment to see if we reach the 0.1%*H? The pile deflection has to go somewhere right?

I'm all about being conservative in the design and just go ahead and use the at-rest pressure, but I think we are pushing the limits of the piling as it is, actually I have a gut feeling things are going to work. Our lateral loads that are being applied to the vertical piles is pushing 50 Kips/pile (LRFD Strength 1) with the equivalent fluid pressure at 40 PCF.

Plus I want to do things the right way.

Thoughts? Guidance? Solutions?

I appreciate the opinions in advance.

 

[bridgebuster]

http://www.eng-tips.com/viewthread.cfm?qid=241796

Take a look at this thread. In the past I've always used active pressure - AASHTO allows it - but now I'm not quite sure. The question of at-rest vs. active seems to come up more often. A few years ago I did a retaining wall rehab project - 2000 LF of gravity walls, ranging from 4' to 26' tall -the geotechnical engineer said to use at-rest pressure for the stability analysis.

I also did two semi-integral bridges -like yours single spans about 150' long- we used active but a reviewer told us to change it to at-rest.

You can search the geotechnical forums here, there are several threads on this topic.

 

[VoyageofDiscovery]

Do the analysis. Show them the battered pile tip loads. Follow their recommendations after that point.

Vertical piles of 50 kips will deflect alot!

VoD

 

[VoyageofDiscovery]

Sorry jumped over the friction portion. Still seems like a lot of horizontal force, which finally tends to move long piles horizontally regardless of strength. An analysis of the bridge and piles together will be the best way.

 

[3Fan]

Thanks for the replies so far. VoD, how do you analyze the bridge and piles together?

And yes, that does seem like a whole lot of horizontal force on the vertical piles. It's going to be even more if we end up using at-rest pressure. Suggestions on eliminating some of that? I've "assumed" that the horizontal load that the battered piles can take is their axial load divided by the batter. So even if we add more battered piles, the axial load decreases per pile, thus reducing the amount of horizontal resistance.

To boot, we are trying to avoid and drive piling around the existing bridge piling, so we can't even adjust the geometry of the footing to force axial load into the battered piles.

 

[VoyageofDiscovery]

This is a standard design, speak to someone in your office or look into some foundations textbooks such Peck, Hanson and Thornburn to make sure you are analyzing the abutment piles correctly. For footings with battered piles, I typically design horizontal forces to go to the battered piles only. In California, battered piles are not acceptable anymore so there is a different approach where all piles are vertical and allowed to displace.

"I've "assumed" that the horizontal load that the battered piles can take is their axial load divided by the batter." - This is not correct, so according to that logic if you have zero load, that pile has zero capacity? Pipe piles (closed-ends, open ends) have capacity, the geotech should give you this. LRFD is one side loading, the other side capacity. Capacity should be greater than loading. At ultimate loads with piles not on bedrock, there may be deformations depending on the soil that start rotating the abutment.

If you have an unusual situation, model it on structural analysis software capable of utilizing springs to replicate soil along and adjacent to the pile. The geotech should give you these and associated ultimate limits. Then feed your results back to the geotech, mindfull that the analysis is LRFD.

HTH

VoD

 

[3Fan]

VoD

The piles are concrete filled closed end pipes.

I have analyzed the pile group based on elastic theory.  I have followed the example in Peck’s book in chapter 26.  From the Figure 26.12b, H=V(x/y).  That’s where I came up with saying the horizontal load that the battered pile can resist is the computed vertical reaction divided by the batter.  Isn't that what the example shows on page 437 with the force polygon? 

22.7/3 = 7.6 Kips
 
OK, I got what you are saying about a pile with zero load has zero capacity.  So if we drive a pile to say 300 Kips, but the pile is only has 175 Kips of load applied to it, the lateral resistance of a battered pile (4:1) would be:

300 * sin (14.04 degrees) = 72.8 Kips?  And not 175/4 = 43.8 Kips?

3Fan

 

[VoyageofDiscovery]

Yes on p437.

The pile would be good for up to 72.8 kips if the geotechs are happy with the 300.

 

[VoyageofDiscovery]

On p437 it compares the load with the overall pile resistance of 25k.

"the horizontal load that the battered pile can resist is the computed vertical reaction divided by the batter"

"can resist" should be replaced with "undergoes"

 

[3Fan]

Thanks again VoD. I'm sure you have seen the FHWA LRFD Steel Bridge Design Example. When they go through their pile design in Design Step P, it appears that they use the same procedure I had been following. On page P-62 they have a table that shows:

Total vertical load on the front row of piles = 2119.1 Kips
Batter = 0.333333
Available resistance force due to horizontal component of axial pile load = batter X vertical load on front row = 706.4 Kips
Plong = 855.0 Kips
Remaining force to be handled by bending of pile = Plong - available horizontal force = 148.6 Kips
Force per pile = 10.6 Kips

They divided the remaining horizontal force by the total number of piles (14 both vertical and battered). They did not use their controlling factored resistance that they calculated on page P-53 of 340 Kips / batter to use for their available resistance to the horizontal forces. They do go on to analyze the pile group using FB Pier with all the combined loadings to see if the results are favorable. I'm not even sure if our geotech will use FB Pier, probably just LPile.

I guess what I would like to know if how to present the loadings to the geotech.

The more I know the less I feel I really know.

 

[VoyageofDiscovery]

Just to step outside the textbook environment. The equal distribution of horizontal load to all piles when batters are present is incorrect. A batter pile is significantly stiffer than a vertical pile. Thus batters will take the load to a higher degree keeping in mind all the piles are working in parallel. In this light, as the designer, it is better to design the batters to take all the load. Based on stiffness, I believe the approach the example takes here is incorrect.

It may be good to have a one-on-one with the geotechnical engineer to see how they understand LRFD in soil mechanics with respect to deep foundation design and find out what they require to confirm their feeling that the wall is at-rest. Whatever it is, compare apples with apples, service loads for serviceability and ultimate loads for phi x failure and nothing less.

Believe me, I know that LRFD is not a method of choice for geotechnical engineers, they have been kicking and screaming all the way.

Good Luck