Restrained Retaining Wall Floor Diaphragm Transfer

[dhoward26]

I was looking through my retaining wall spreadsheet today and looking at some floor diaphragm transfers on a project. I have attached a pdf with very basic drawings and a question.

When looking at transferring the lateral load from the retaining wall through a wood floor diaphragm, does anyone ever account for the return corner taking some load, thus reducing the overall diaphragm transfer to the walls? Seams there may be room to reduce the amount of lateral load transferred so that the connections aren't so big in the side walls. I don't have a finite element program so that is out of the question for me.

I have designed a number of underground concrete tanks and rely on the horizontal transfer based on the information in the PCA Rectangular Concrete Tanks publication, but never really considered it much for a residential application.

 

[AELLC]

I don't know of any science behind this, but I would use 1/2 of the retaining height.

That looks like a very large footing, and why do use verts both faces in the wall?

 

[dhoward26]

This isn't the wall I'm designing, it's purely for visual purposes. It was an old retaining wall that I adjusted the footing on for the purposes of posing the question.

 

[briancpotter]

Interesting question.

A cursory look suggests that the value "X" will depend on the relative stiffness of the retaining wall, the side walls, and the diaphragm. Without resorting to FEA, I'm not familiar with any generic method of determining the force distribution, though I suspect someone more familiar with structural mechanics could come up with a decent approximation.

I'll think more about it and see if anything dawns on me.

Brian C Potter, PE

http://simplesupports.wordpress.com

 

[dhoward26]

briancpotter:

Thanks. It's an odd area, because part of the force has to transfer into the wall, but as you say, it's probably partly due to the stiffness of the wall.

It makes me wonder if a person were to design the first 8ft from the corner as a typical retaining wall, then in effect you should be able to ignore the load transfer to the floor diaphragm for the first 8ft from each corner. I will make another sketch of that tonight if it sounds confusing.

 

[TXStructural]

Generally, I would not consider the floor to act as a restraint for the wall. The forces involved with a rotating wall can be tremendous, and most detailing at the wall-floor connection is not sufficient. The most reliable design strategy is to design the wall as cantilever, for passive (higher) soil pressures, since any movement will be forced on the structure. Freestanding walls can be designed with lower, active pressures, since slight rotations of the top of the wall will be uneventful.

You do get help from sliding if the foundation wall goes around the perimeter (not a walk-out) and there is a concrete basement floor to transfer sliding forces from one side to the opposite side.

It is better to design only a single section, that is to say, don't change the layout around the perimeter. Simple is better than saving a few pounds of steel. I do recommend detailing horizontal corner bars around the corners and horizontal reinforcement should be provided across the inside face of the wall. There will be minor stiffness compatibility issues, especially if you have not designed for passive pressures.

 

[hokie66]

Agree with TX that I wouldn't use a wood floor to brace the top of a concrete retaining wall. My approach would be to use either a cantilevered wall, or a horizontally spanning concrete beam somewhere near the top.

 

[dhoward26]

TX: Thank you for the comments.

I have done it both ways in the past with using a cantilever system and a braced system. I usually use the braced system when the load transfer is such that I can use typical diaphragm (6" o.c. edge) nailing for the floor system. Doesn't happen all the time but I still do it. I add the horizontal steel as well to address cracking issues and stiffness. Corner detailing is a must as well.