Designing Retaining Wall Around Enclosed Area

[person123]

TL;DR Summary

Are the calculations for a retaining wall different when it is around an enclosed area?

Normally when designing a retaining wall, you check for failure due to sliding, overturning, and insufficient bearing capacity. However, if I have a retaining wall which is around an enclosed pit, it doesn't seem reasonable to perform the same checks for sliding and overturning (the retaining wall is continuous concrete around the perimeter, not four separate walls). With overturning for example, you have the entire weight of the wall countering the moment in addition to the soil on the opposite side. Instead, it seems important to check for whether the wall would cave in as shown below, from a plan view. Is this correct, or is there something I'm not understanding?

 

[jrmichler]

No, you have it right. Ground forces push all four walls toward the center BECAUSE all four walls are tied together, not four separate walls. Since each wall is supported at the ends, you have the situation correctly sketched.

 

[person123]

Would the standard calculations for overturning and sliding still be applicable?

I'm not sure how familiar people would be about these calculations, so I'll just summarize the one for overturning briefly. It's done per length of wall, so all forces are force per length. The overturning moment is due to the lateral pressure from the soil. The stabilizing moment is due to the weight of the soil and wall. The ratio gives the factor of safety (which should be more than 1.5 or 2).

 

[jrmichler]

Yes, but for the entire box, not each wall. If your box is narrow, and the Earth is much higher on one side, the entire box could tip. Same for sliding.

 

[person123]

Got it. Based off the dimensions of my particular box (not very narrow, and only a mild soil slope), I think it would just look pretty absurd for the entire thing to tip over, but I could check the FoS on it.

I'm going to play devil's advocate briefly though. I thought calculating force per length might still be applicable since there's also the concern of just one section of the wall tipping over, either breaking off from the rest of the wall or else deforming. Does this seem reasonable?

 

[jrmichler]

Treat each of the four walls as a flat plate supported at two opposite ends with a distributed load. The end support condition could be simple support, fixed, or somewhere in between, depending on the exact design.

 

[person123]

Yes, that approach make sense to me. I'm not sure if I could reduce it to a 1-dimensional beam, or if I would have to keep the full complexity of a 2-dimensional plate (due to the boundary condition with the ground and the pressure varying in the vertical direction), but I also thought that general approach makes sense.

I'm pressing on the force per length check for overturning and sliding though, because those are standard checks for retaining walls. I just want to confirm whether those checks would still be relevant.

 

[jrmichler]

Start with a thought experiment. Assume a flat plate. If you stand that plate on one edge, and pile dirt against one side, it will immediately fall over. You can prevent if from falling over by:

1) Dig it down into the ground far enough to prevent it from falling over (retaining wall),
or
2) Make the bottom edge wide enough to prevent if from falling over (retaining wall),
or
3) External supports, such as placing a three sided box against it, and placing dirt against the box to prevent the box from sliding (your box).

Standard checks for overturning and sliding of retaining walls are used in your situation. They are used for the entire box, not one wall of that box.

 

[person123]

Got it. To summarize then, I guess these would be the potential failure you would have to check for:

1. Insufficient bearing capacity (same as for single retaining wall)
2. Overturning of the entire box (in my case, very unlikely)
3. Sliding of the entire box (also very unlikely)
4. Caving in of the walls of the box (I'm not sure exactly how I would check for this)

You wouldn't however, check for overturning or sliding per length of wall, as it's being supported by the other sides.

 

[anorlunda]

This may be helpful.

 

[berkeman]

Summary:: Are the calculations for a retaining wall different when it is around an enclosed area?

I have a retaining wall which is around an enclosed pit

Sorry if I missed it, but what are the dimensions of the pit and the wall? What is in the pit? Are there stairs or something for people to access it? Can you add bracing across the pit to help support the middle portions of the walls? How is any water accumulation in the pit drained?

 

[person123]

Since this is for a company (I'm trying to get some insight on the general type of problem before discussing with my PM), I don't think I can give the specifics (I'm also not sure about the stairs and water draining).

One thing I should mention though is that my rough schematic was not at all to scale to make it easier to see. Perhaps a more accurate sketch would look something like this (again not the actual dimensions), with much longer sides relative to the wall's thickness:

I noticed in the video, failure doesn't occur across the entire length of the wall; instead it's just one section of the wall failing. Wouldn't that be possible in this case as well, in which case it would make sense to check for overturning and sliding per length?