Anchored Sheet Pile Wall Global Stability 2

[jaggith]

I am reviewing a failure of an anchored sheet pile wall (14 ft retained ht) along a river where the designer assumed medium dense conditions at the toe of the sheet but let the contractor stop the sheets in very soft clay, 5 to 10 feet above the actual top of the med. dense material. The wall failed in an apparent rotational or global failure with the top of wall moving inward toward shore.

The designer used the SPW911 software but did not consider possible variations in the soil strata or global failure due to consolidation of the soft clay behind the wall due to added fill surcharge and its impact on the stability.

When the fixed earth support method is used, is there a minimum depth of penetration into the medium dense material that is assumed or expected by the software?

 

[ishvaaag]

Have no idea with the software, but in pure logic it will need some degree of embedment to get to some wanted level of fixity, i.e., once elected the medium of embedment it can't stand the lever action compensating for the wanted moment at the interface but at some length of embedment. In other words, the required length surely exists as a function of the soil that provides the fixity, but it might be being derived in a per case basis by the software.

 

[InDepth]

Well I guess the question is about global stability. I think regardless of of the method of design if there is a slip plane formed it will kick out the bottom. I think you need to perform a slip plane analysis check and see what embedment is required for glabal stability (which may be larger than the required for force equilibrium of the bulkhead). Maybe our geotech experts can answer this better. There are a few good papers on the ASCE website that talk about sheet pile failures in clay.

http://cedb.asce.org/

 

[jaggith]

Two things are at issue here:

1. wall designer thinks that the software checks global stability (I don't)

2. designer was CEI on project and did nothing to make sure the tip of the wall was seated into dense material.

 

[InDepth]

Here is the link to the software. You should obtain the manual with it or call the rep. Looks like its based on:

British Steel Piling Handbook and the US Steel Sheet Piling
Design Manual. I forget if they have a global stability check.

I especially link the statement "Reports on design safety. Unique Rules Of Thumb feature ensures safe designs."

http://www.gtsoft.org/spw911.htm

Manuals link:

http://www.gtsoft.org/manuals.htm

 

[DRC1]

I have used earlier versions of SPW911 but not the current version. The earlier version did not check Global stability. Looking over the worked example in the manual it does not appear that the program checks global stability. Generaly, if global stability is an issue, I will use Stabl to check it. Can also be done by hand with method of slices.
If the designer stopped the sheets in soft clay 5-10 ft above the sand, this begs two questions 1.) did the designer assume granular material the full depth of the toe? and 2.) was the granular material not at the toe because the sheets were driven short or because the material was deeper than anticipated?
Obviously if the sheets were short, that would contribute to the failure. If the clay ran deeper than anticipated that would have an effect also as the design model for clay is different than granular materials. Most often SPT borings are the most common source of subsurface information which can give very little useful information on clay unless the designer has a lot of experience with the material. Also the propeties of the clay and the resulting load on the wall can change over time.
Finally, there are two design methodoligies for sheeting- Fixed Earth and Free Earth. Free Earth assumes that the sheeting is just long enough for the active and passive pressures and tie loads to balance. Fixed earth assumes that the sheet extends further so that the sheet can develop the cantilever capacity of the sheet, and the tie rod is providingthe required additional reaction to support the sheet. Free earth is statically determinate. Fixed earth is essentally a propped cantilever and is not statically determinate, however can be solved readily by some simplifying assumptions.Free earth has shorter sheets. Fixed earth has longer but lower section sheets and smaller tieback loads. The depth of embedment is important for both methods.
For more details see slideruleeras website for a copy of the US Steel sheetpile design manual.
From your

 

[DRC1]

I'm sorry My comments posted before I was ready.
From your post, it would appear global stbility may be an issue. Insufficent embedment may also be an issue which may be due to an unaticippated subsurface profile.
One coment you make is that the designer/CEI did nothing to make sure the tip was seated in dense material. First there is rarely nothing more done in contruction than to be sure the sheet is driven to the specified depth. Second, unlike bearing piles, simply seating the sheets in dense material will not have a significant effect on the capacity of the wall. Changes in profiles can have a radical effect on wall performance.

 

[InDepth]

Well done DRC1

 

[jaggith]

The clay did in fact run deeper at that location than the wall was designed for. The original design assumed the fixed earth method.

The contractor and CEI (original designer) just put the sheets in to the specified depth and never made an attempt to monitor rate of penetration, etc. to confirm they ended up tipped in denser sands below. From there it went downhill and the toe moved outward due to lack of passive resistance/embedment.

Do you think the designer had any responsibility as CEI to make sure the sheets actually penetrated into the materials he assumed were there in his design?

 

[DRC1]

InDepth- Thanks!

jaggith-
To answer your question, I would say typically you would not notice anyting unusual if you are driving with a vibratory hammer, the common way to install sheets. If the sheets are dropping like a rock out of the sky, then yes, you may question the soil. But sands and clays will allow reasonable quick penetration rates. The vibratory hammer does change the insitu conditions during driving, thus unlike impact driving, it is hard to assess subsurface conditions during driving.
However, The entire subsurface profile is very important to design, and the question arises how it could be off so much. Were adequate borings taken or did they guess?(it happens)If the clay was known to exist, it should have been mapped. The whole analysis of the wall would be different for a clay profile than a sand. Also sand does not require as much additional information as clay. It sounds as though the wall did not have adequate
penetration or that the clay could not develop adequate stength for the load imposed, although global stability is still a possibility. The only way to tell is to reanalyze the wall based on the clay profile. Note that when gathering properties for the clay, much of the exposed clay is probably remolded and not indicative of the strength when the system was installed. Best bet is to take borings and recover samples from an area near bu not in the failure.
Notice how no one had time or money to take these borings and tests when it was being design, but now there is plenty of time and the costs, relative to everything else are insgnificant?

 

[jaggith]

The upper river bottom clay is 1 to 2 blow count and the sands and underlying stiff marine clays are 10 to 50 blow count material so one would think you would notice the sheets slow down in rate of penetration when they reached the target layer for the tips.

They also had the ability to review the driven H-pile strut data/blow counts before they installed the sheets and figure out that they missed the boat or that the soils were highly variable in one area. Struts went to 20+ feet below where the sheet piles were arbitarily stopped.

I agree about the cost now versus the minor cost of really exploring it before.

 

[DRC1]

Could you clairify a couple of things: Who designed the sheets? Was a driven length specified? was the actual driven lenth shorter than what was specified? Was there one level of bracing or multiple levels? Not sure what you meant by Struts went 20 ft + below where sheets stopped.
Interesting problem

 

[jaggith]

Structural Engineer designed the sheets.

Driven length was specified and sheets went to the minimum depth/design length specified.

There was one level of inclined H-pile used as a lateral brace on the water side of wall near top of wall. I referred to that brace as a strut.

 

[DRC1]

Seeing how the design engineer specified a sheet length and the sheet length was installed, it will be hard to hang this on the contractor or inspector. It appears there was a difference in what was assumed for the profile and what was actually on the site. Wether this was inadequate exploration or misinterpertaion of the data would require a review of all the documents related to the design. It may not be possible, as there would need to be some data available prior to the failure, but if you can check the piont where the sturts enter the mud, you may get some (not conclusive) idea as to weather the wall failed in global stability or due to inadequate penetration. If there was a global stability failure, the point where the strut entres the soil may have translated laterally and maybe slightly upward, where as if the sheet rotated, the connetion to the wale would be disturbed, but the penetration point may not. Also along the same thought a profile of the out bord side of the wall may show the extent of soil movement, also helping to solve the question. However, mostly I have seen backcalculations to determine what part was underdesigned.