Lateral Pressure For Deep Basements

[abarker]

Is anyone aware of research that might indicate a non-traditional lateral stress distribution (i.e., something other than linearly increasing with depth based on Rankine pressure distribution) might be appropriate for deep basements? I am not aware of anything, but I have a client who was told that there's less increase with depth for deep basements. He has a 40' deep basement and if he can design it for a lower stress, he likely will save substantial money on the project. I need to find out if someone is "blowing smoke" with this one. Any thoughts would be helpful!

 

[Focht3]

Hmmm,

Some have used the braced excavation pressure distribution from Peck's 1969 braced excavations paper (Mexico City ICSMFE.) Others us a K₀ uniform pressure distribution. Still others use Peck's - with some modifications.

The pressure distribution chosen must match the soil properties, wall type and stiffness, and construction sequence. Each job is unique -

Where is the project, anyway?



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?"

 

[gandersen]

abarker:

I have read research papers that describe the effect that arching has on the lateral stress distribution behind a retaining wall under active conditions. I remember seeing that at the very bottom of the wall, the lateral earth pressure was lower than the assumed Rankine Active case. Overall, the magnitude of the lateral force was about the same, but the height of the resultant was higher (I seem to remember 0.42H for some reason.

It's been more than a decade since I read these papers so I may not be able to come up with the reference easily.

Perhaps someone else also remembers.

Glen

 

[Focht3]

I don't recall that paper - but I have used the Peck pressure distribution model with a total force obtained with an equivalent fluid pressure (~Rankine) approach. You generally end up with a more conservative answer - the resultant tends to occur at the mid-depth of the wall and away from the bracing effects of the bottom floor slab. I do this when the wall will be quite stiff and the select backfill zone is chimney-like in profile.



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?"

 

[abarker]

Focht3 and gandersen,
Thank-you for the input. Focht3, this project is in Madison, Wisconsin, it's glacial deposits (on a sideslope of a ridge so generally it's lacustrine soil overlying deeper glacial till). The soil profile is lean clay and silt in the upper 1/2 then clean sand with gravel below (about lower 10' to 15' would be below the water table, but very to extremely dense. Lowest level is about 40' below exterior grade. Knowing the site constraints, I would believe the backfill will qualify as "Chimney-like". It's below grade parking, so I'd also have to believe the system will be pretty rigid (poured concrete walls with precast plank decks). However, I'll know more within a couple weeks when I recieve a set of plans and specifications for my final review of how they match my recommendations.

I had thought about Peck's braced excavation pressure distribution, but I thought that required a "top-down" type construction process where the bracing is added as the excavation advances downward. Although I'm sure the selected contractor will use his preferred construction methods, I believe the design assumes standard construction (i.e., excavate to the design lowest foundation depth then construct the building using "bottom-up" construction). Am I wrong with this thinking?

 

[Focht3]

You are generally correct. Having said that, it's important to remember that in order to develop a Rankine-type active pressure distribution a fair amount of wall (and soil) movement must occur. If the "opportunity" for movement is limited, then the geotechnical engineer must look for ways to address the circumstance. The methods can vary from the complex - FE analyses, for instance - to the relatively simple. In the end, though, it's a judgment call. As it should be.

Think about the construction sequence. Will the soils be restrained during construction, or will they be free to move as much as they "want"? I can't answer that question - only help with the thought process. Your answer to this fundamental question is a key one -



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?"

 

[abarker]

Focht3,
Great answer! No matter how "fancy" you get with the analysis, it always comes back to the fundamentals, doesn't it? I'll see what I can find out about the proposed construction sequencing. Thank-you for your thoughts on this one.

 

[cap4000]

I believe you have a clear cut case here for an "At-Rest" Ko pressure and a Rankine triangular load distribution. The Terzaghi formulas were set for "Temporary Strut" loads for "apparent" loading diagrams either rectangular or trapezoidal in shape mainly for Site Contractors. In order to for soil to active, the wall movement would have to be on the order of 0.001H to 0.004H, which I would not count on based on the multiple rigid parking level supports. The only other thing I might investigate further perhaps is to use Coulombs wall friction angle and just make a comparison to Rankine. Your right, standard construction procedures, forget the smoke and don't get to "fancy" as it only makes for potential lawsuits.

 

[GeoPaveTraffic]

I mostly agree with cap4000, don't get to fancy and I would design for full at-rest pressures. The only thing that I would add would be that the upper floors of the basement should be checked for loads higher than the at-rest contributory area. I believe that this is what Fotch3 was getting at in an earlier post.

Depending on the relative stiffness of the various building elements more or less load may arch to various points in the building. For example if one of the upper basement floors is relatively stiff, then horizontal load from lower in the structure may arch to that floor. This could result in that floor taking much more load than the contributing area of the at-rest diagram. The structural needs to know that this might happen and ensure that there is sufficient load capacity in the floor and the wall system "higher" up in the basement.

 

[cap4000]

GeoPaveTraffic

My post was with respect to the soil pressure loading diagram on a rigid basement wall versus a flexible braced cut situation. You are correct about arching and stiffness. The Structural Engineer designing the basement wall would then have to do a moment distribution analysis with perhaps fixed ends at the base and top slabs. The in between plank floors my guess would be is that they would be pinned only.
Nice job covering your rear with the Structural Engineer.

 

[Focht3]

Yes, my post was directed at the issue of load redistribution. Since my undergrad education was heavily weighted toward structural engineering and I have done a lot of monitoring of completed structures, I am very aware of the key soil/structure interaction issues. But most structural engineers aren't - so we must guide them (even when they don't know it...)



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?"

 

[abarker]

Everyone,
Thanks for the heads up from the Structural Engineering perspective. I come from a soils background and am not very strong when it comes to the redistribution of forces within a structure. I'll take the guidance any day!