Segmental Retaining Wall Failures

[goochy]

I am involved in a situation where a wall has failed and one engineer has indicated that the wrong soil was used behind the wall. The specs called for SM soils and it appears that some MH soils were used. There are other issues as well, such as the wall was designed with a factor of safety of about 1.1. I am looking for some help into the possibility that the wall might have been adequate with the MH soils that were used. I am not a designer of these walls, but do provide testing for installers. Any sources of information for these walls and their failures would be greatly appreciated.

 

[layfieldman]

I would contact the NCMA (National Concrete & Masonry Association) for more information or contact the block manufacturer as they typically provide technical/design support. From my own research I've found that 70%-80% of SRW failures can be attributed to either poor compaction or poor drainage. Then there is also when the grid is installed in the wrong direction, wrinkles in the grid... ie. improper geogrid installation.

 

[Focht3]

What kind of wall is it? (I can't tell from the context of your post.)

What tells you that the wall was designed for silty sands (USCS SM classification)?

Where is the site? What are the wall's dimensions? How did the wall fail? (i.e. we need a lot more detail in order to help you -)

Has a claim been filed against the wall's designer?

Are you providing litigation services?



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

 

[PEinc]

goochy,

Don't get too excited about the 1.1 safety factor until you have reviewed the design calculations. It is possible that the designer factored the loads before designing the wall. Therefore, he would need a safety factor equal to or greater than 1.0.

If you want to talk to someone intimately involved with segmental retaining walls, try contacting:

Dr. Robert M. Koerner or Dr. George Koerner of Drexel University's Geosynthetic Institute (610)522-8440

Dr. James G. Collin of the Collin Group, USA (301)881-7476. He's a consultant to the MCMA (National Concrete Masonry Association).

Dr. Dov Leshchinsky from the University of Delaware

Good luck.

 

[layfieldman]

One thing I did forget to mention is that a Dr. Richard Bathurst is doing a research program at the Royal Military College in Canada. He is being funded by varius agencies which include the NCMA and FHWA. He has built 0ver 19 full size retaining walls which are fully instrumented in his labs based on a FOS of 1.0 to determine how conservative existing design practices are. From his research he has collaborated with a Dr. Tony Allen of the Washington state DOT to come up with a new empirically derived design procedure called the K-Stiffness method.

 

[cphi]

goochy,
If the wall was designed for SM soils, and MH soils were used, the construction is incorrect, and an obvious reason or at least a contribution to the failure.

That said, the constructed factor of safety with MH soils will be less than the design factor of safety of the reinforcement designed for SM soils. To determine if the actual factor of safety is adequate will require determination of the friction angle and density of the actual soils used and run the calculations with the actual soil properties with reinforcement per the original design. You will need to know what design method was used for the original design. Several different design methodologies are used, such as AASHTO, NCMA, and others. The 1.1 factor of safety you mentioned is used for several design items with seismic loading. The static factor of safety will be automatically greater.

Besides the subjects of the questions you asked, there are **MANY** other factors that could be the reason of failure.

 

[MSEMan]

I am shocked that some of my esteemed colleagues seem so blase about your problem. MH soils include a broad class of materials, many of them are inherently unsuitable for MSE walls.

To get a handle on this problem go back to your fundamentals. The reason that SM soils are specified (and these near the bottom end of acceptable soils for use with MSE walls) is that they are wholly frictional, low plasticity materials (PI < 6). Becasue of this, you can rely on acheiving effective stress transfer from the soils to the reinforcements when they are compacted. The reinforced soil structure can therefore be designed using elastic principles.

MH soils, on the other hand tend, to be poorly draining, more plastic and they rely on cohesion for a large part of their strength. Adding geogrid to an MH soil may reorient the failure plane, but do little to enhance its inherent strength. Stress transfer is unlikely to be immediate, due to the inherently high degree of pore water pressure that will build up during compaction. Additionally, MH soils may exhibit plastic creep, (and do so unpredictably and non-uniformly) so there is the very real possibility that structure movements may exceed the differential strain limits of the facing. There is no accepted rational design procedure for reinforcing plastic soils. Needless to say these materials are inherently unsuitable for MSE walls and should only be used in exceptional circumstances and after a thorough investigation of their engineerng properties. Even after such an investigation, it is likely that special construction/compaction methods would need to be employed.

The problem you are facing is caused by a fundamental error of judgment and lack of understanding on site. Unfortunately these errors are being repeated on an almost daily basis and it is a wonder to me that we aren't seeing more failures.

 

[Focht3]

Hmmm,

I'm not blase - just waiting for more details...



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

 

[MSEMan]

First of all, sorry to Focht and anyone else I offended, it wasn't intentional. What concerns me about this particular problem is in the rush to pump numbers into a computer, it is easy to forget the assumptions in the design program. The MSE design methodologies have been developed empirically, and if you don't have a good grip on the empirical background, you run the risk of applying an inappropriate model.

I am involved in the design of a major project in the Middle East where we are using SM soils with about 25% passing 63 microns. The design is AASHTO and the fill is way outside the AASHTO guidelines. None the less the mechanical properties of this fill are excellent (completely non-plastic, very high shear angle, compacts well) so we are happy for the contractor to use it (besides its free). So I'm not saying follow the spec as if it were gospel, what I'm saying is that before you "run the numbers" take a bit of time to figure out whether it is realistic to expect any sensible answers from such an exercise.

 

[Focht3]

No offense taken - we all need to vent periodically...

...take a bit of time to figure out whether it is realistic to expect any sensible answers from such an exercise...

Yup. Agreed.



Please see FAQ731-376 for great suggestions on how to make the best use of Eng-Tips Fora. See faq158-922 for recommendations regarding the question, "How Do You Evaluate Fill Settlement Beneath Structures?"

 

[cphi]

Criticism accepted, thank you.

 

[g7mann]

goochy:

As well described in the previous responses an MH soil is unsuitable for use as a compacted fill behind a segmental block wall. Unfortunately, many contractors will use whatever soil is available, and usually the cheapest [free if possible] for backfilling purposes. The primary issue with these [most] contractors who build these walls is their virtually total lack of understanding regarding the mechanism involved with a geogrid reinforced block wall. At best, they may have seen the block manufacturers' handout which, unfortunately, only typically applies to one scenario, and if is never the scenario the contractor is involved with. Nevertheless, he will blithely apply the manufacturers' "design" to his project and then wonders why it fails.

Not only will MH soils absorb water, they are unlikely to compact well, and they tend to act as a plastic or quasi plastic material and do not "grip" the geogrid reinforcing layers well. Review of teh filed compaction records should preovide some insight regarding the condition of the material when it was placed and compacted. My estimate is that it was on the wet side of optimum, probably by several points.

Another issue is the elimination of any drain system behind the compacted fil mass. If no effort is made to intercept, collect, transport and dispose of any seepage or infiltrating moisture then hydrostatic pressure can build up and act on the compacted fill mass. Simply another load condition that is often missed, particularly by contractors.

The type of block, and possibly the software used for design [precuming software was used] can also have a detrimental impact on the geogrid used. An inappropriate geogrid material, and there are some of those out there, can result in an inadequate interaction between the geogrid and the soil and this can lead to pullout failure. In my experience, when attempting to determine the most appropriate [read economical or cheap] geogrid the computed pullout resistance is almost always low. This has to be carefully evaluated to make sure that a "failure" is not built into the original design. Believe me, this happens all the time.

Lastly, was the contractor "experienced" in block wall construction? Many who say they are in reality are not. The level of expertise is quickly determined when visting the site to observe the walls' construction. Watching the contractors' staff roll out the geogrid parallel to the back of the wall so that the machine direction is incorrectly located results in a wall that is inadequately reinforced, even if the contractor chose the correct geogrid.

Hope these comments help.