Flexible Sheetpile Walls 1

[VoyageofDiscovery]

I have seen all the books explain Ka and Kp beneath the excavation. I would like to question this as I am computer modelling sheetpiles with springs.

I understand Kp as the limit of a soil's capacity to an applied pressure, and that the spring constant is limited by this value at any given depth. That being said, Ka and Kp should not exist together as Newtons Law says for every force there is an equal and opposite reaction.

This being said, as I am modelling my scenario, whether I use Ka pressures or not below the excavation depth, I do not get the pressures at the Kp limit, this makes sense.

When I use Ka pressure values below the bottom of excavation, I get spring reactions which give me pressures within Kp resistance allowing for a factor of safety of 2 but my sheetpile toe is kicking out. What gives? When I do not use Ka pressure values below the bottom of excavation, my toe does not kick out.

Driving flexible sheetpiles into level ground should result in a net zero pressure. Thus below the excavation there should be no net earth pressure as they cancel out? In an excavated situation should not the pressures below the bottom of excavation then originate from above the bottom of excavation through the soil structure interaction.

My geotechnical engineer has told me the active zone is drawn 45-phi/2 from vertical originating at the BASE of the excavation, and he still thinks in terms of Ka and Kp.

This has driven me crazy over the years.

VOD

 

[GeoPaveTraffic]

You are correct that if you drive a sheet into level ground the pressure on the sheet is equal on each side or nets to zero.

The reason that the pressure is higher on the excavation side of the sheet below the excavation is that the sheet is moving, due the unbalanced pressure above, into the excavation. As the sheet moves the springs in your model are compressed and the load increases up to the Kp value. Once the Kp value is reached the spring can continue to compress, but the load does not increase.

For your model, anytime that the sheet is moving into the soil, Kp applies to that side; when the sheet is moving away, Ka applies to that side.

 

[VoyageofDiscovery]

Hi GeoPaveTraffic,

In the case of flexible walls, below the excavation level as the sheetpile flexes either way, it only sees passive resistance springs, no active in my opinion. Thinking of a diving spring board.

VOD

 

[VoyageofDiscovery]

To clarify, my geotechnical engineer said the active zone for the excavation was from base of excavation to top of excavation at 45-phi/2 for an excavation.

VOD

 

[GeoPaveTraffic]

VOD, in general that is the case, however, if the sheets are to be in place for a "long" time active pressures can develop. Omitting them is common practice.

 

[PEinc]

Out of curiosity, why are you modelling the sheet piling? Also, passive earth pressure is a reaction, not an applied force. Hopefully, enough passive builds up to prevent overturning. For a cantilevered sheet pile wall, if you have enough passive resistance to prevent overturning, due to differing moment arm lengths between the active and passive pressures, the total passive force will be greater than the total active force. Some people think you then have a problem if you sum the forces in the x-direction. They think the wall will move backwards! Can't happen. If it did, then the higher active side would become the passive side and the lower passive side would become the active side. Can't happen.

 

[VoyageofDiscovery]

PEInc,

I am modelling the sheet piling because the conventional method gives me a depth I need to reduce. This is why I questioned this assumption of Kp & Ka pressures beneathe the excavation level. A depth I believe that can be reduced without compromising the Factor of Safety.

I am not dealing with moment arms for passive triangular limits in this particular analysis. My passive reactions are accounted for through passive springs. My results show no triangular shaped pressures on the passive side as we assume in the simplified methods to calculate depth of embedment.

VOD

 

[PEinc]

VOD,

There is not one "conventional method" to design sheet piling. There are several methods, various assumptions, and a number of computer programs for designing sheet piling. All have been used successfully even though they may give different answers. What "conventional method" did you use? Is your wall cantilevered, braced, or tiedback? How much are you trying to reduce the embedment? Tell us more about your wall. It sounds like you may be trying to reinvent the wheel.

 

[DRC1]

First of all I assure you the use of passive and active pessures do not violate any of Newton's laws. I assume you are modeling a cantilevered wall. As suport conditions change, so does the analysis. Take a dowel, mark a line on it a third from one end. The dowel is your sheet pile and the mark is your dredge line. Stand the dowel on a table. Now push the dowel slightly above the dredge line. The dowel topples,rotaing about the push point. Push the dowel at the same point and in the opposite direction at a point about half way between the botom of the dowel and the dredge line mark. The dowel stll topples, but now it rotates about the lower push point. Now (With some help)push at the same two points and at the same time push in the same direction as the first force at the very bottom of the stick. This time it should stay. In order to balance itself, the sheeting develops net passive pressures on both sides of the sheets The exac pressure distribution needs to be solved iteritvely. Teng used a simplfying solution for a more direct answer. However, a cantilevered sheet is more than an actve pressue verses a passive pressure.
The interesting thing is that as support condtions increase, the easier it is to model. I have never seen springs used for active/passive analysis. Location of moment arms are important for the analysis. A method of using the soil modulus is more suited to a spring model. The problem with soil modulus, is that it is hard to get accurate values. Even with marginal data, generaly fairly reasonable values can be assumed for active and passive coefficents.
Other concerns that your model needs to account for are water table conditions, seepage, lateral loads from sucharges, adj. foundations and embankments, support system, sheetpile, soldier pile, etc.
A good place to start is the Corp of Engineers manual on sheet pile design, EM 1110-2-2504. It is availale on line at

http://www.usace.army.mil/inet/usace-docs/eng-manuals/em1110-2-2504/toc.htm

It gives you insight into both classical anaysis and analysis by spring models using soil modulus.
Earth retention is soewhat specialized field of geotech. If you are trying to reduce yor embedment from a result of a clssical analysis, it cn be done sometimes, but you really should talk to a geotech or contractor wth experience in this field. Wall collapses can and do happen.

 

[VoyageofDiscovery]

Hi PEinc,

Thank you for your response, the conventional method is out of the USS Steel Sheet Piling Manual. The order of reduction I am getting is about 2 metres for a tie back wall about 4 metres high above dredge line. A collegue of mine did the conventional method, and then we modeled them both. Model was simple and reviewed and so the results made us question the application of loading.

DRC,
I am trying something here, your advise is somewhat patronizing. I am aware of the nuances of sheetpile design and I have narrowed it down to the application of active while questioning the passive limit from my passive spring displacements. I value your opinions in this forum and I hope my question is not a waste of time. You refer to a dowel, it is not flexible, I am modeling the stiffness of the sheetpile so it will not respond exactly the same way.

Is there certainty that flexible sheet pile walls respond to soil pressures as assumed or may we encroach in this area while ensuring a factor of safety. My comparitive analyses has made me ask this question.

Cheers

VOD

 

[DRC1]

VoyageofDiscovery,
I did not intend to be patronizing. If I offended you I appoligize, as that is surely the farthest thing from the purose of this forum. However it is difficult some times to express complicated engineering concepts in 2 or 3 written paragraphs. Sometimes it comes out well some times it does not.
Any way, my point with the dowel, is that Passive pressure develops on both sides of the sheeting and if you dont properly account for it, that could cause your analysis to show the toe kicking out. I still don't see how the active and passive pressures are modeled with springs, as that seems to indicate that the applied load may be less than full active or passive. (If this is what you are trying to do there is a good journal paper in ASCE Geotech journal a few months ago)
As for the dowel analogy, the classical method by hand first considers the stability of the sheet as a rigid body, then determines the moment capacity required of the sheet. This not to say that your method is wrong. A flexibilty approach may be amore sensible approach.I wouldnot argue that in general the classical method is conservative. My point is only that if you are using a completely different philosophical approach to the problem, it is not surprising that it does not compare exactly to the other method. I did not mean to be patronizing, nor do I mean to discourage you. reinventing a wheel is hard work, but if we end up with a better wheel that would be great.

Good Luck!

 

[miecz]

VOD

Back to your original post, it seems to me that you can have active and passive pressures at the same elevation, with the net pressure being the difference of the two. How do you figure that they can't exist together?

 

[miecz]

Instead of spring supports, what if you model tension only members on both sides of the sheet pile?

 

[VoyageofDiscovery]

Hi DRC1,
Water under the bridge, back to the post.

jmiec,
The passive nature is not argued, it is the limit of Kp because this is the failure state of the passive soil resistance in combination with the active pressure that is questioned. A soil will respond with an increasing resistance up till Kp.

To simplify, if a soil acts, an equal an opposite reaction on the passive side is necessary. Kp is merely a soil's limit.

I think using Ka and Kp together is fundamentally flawed though it may yield a conservative result and save all the iterations.

What is the use of modeling tension only members? Tension only members do not represent the force-deflection nature of soil.

 

[miecz]

VOD

A member can be used to mimic a spring, with k=AE/L. However, on second thought, this does not help solve your problem. Forget that I mentioned it.