Cantilever retaining wall or not?

[ars001]

Ok, hopefully I can explain this correctly. I have a situation where I have a medical facility that has a 14'-0" high walk-out basement. Approx. 150'x60'. CMU/Cast in place(haven't decided on cmu vs. poured) on 3 sides. Actually the wall only wraps with cmu for 40' on each side. The floor is precast concrete plank.

I'm trying to avoid providing a cantilever retaining wall along the back side. Not that I have a problem with it but will this truely act as a cantilever retaining wall with such a rigid diaphragm? Would I also need to reinforce along the inside face in this case? I'm also pressing the envelope on designing the wall as simply supported and taking my floor diaphragm load out to the side walls. Would it be reasonable to design the wall being fixed at the base and pinned at the floor to ease the reaction going into the floor diaphragm?

Thanks for the help

 

[lkjh345]

ARS001:

Working on a very similiar case right now.

I designed my back basement wall as a 'propped-cantilever' wall with a fixxed based, and a pinned top. This reduced the reaction to my 2nd floor diaphragm significantly. I then designed the floor diaphraph to take the reactions out to the side walls. Needed to add rebars in the floor topping ( in this case 3" composite deck with 3" of concrete fill ) to provide adequate tensile capacity to the diaphragm 'beam'.

Hope this helps.

 

[ars001]

lkjh345,
Thanks for the quick response. Thats probably the route I'm going to go. I shouldn't have a problem transfering my diaphragm load into my shear wall. I can do this with dowels from the precast into the wall. My tension chord for my diaphragm 'beam' shouldn't be an issue since I have a continuous beam line supporting the precast at the a walkout.

How did you reinforce your back wall. Being fixed on one end and pinned on another did you provide reinforcing on both faces of your wall?

 

[lkjh345]

Yes. I am using reinforcing on both faces. Due to some architectural precast panels coming part way down the wall, and the need to provide a ledge for them to sit on, my wall is 14" thick at the base (8" thick above the precast ledge), so, in my case, the reinforcing in not all that bad.

 

[UcfSE]

If you pin the top of the retaining wall, you're just designing a wall as you would any other with lateral soil pressure on it. When you go to a supported wall instead of a cantilever your diaphragm at the top of the wall has to take the reaction of the wall. If you use a cantilever retaining wall then youhave no reaction and hence no imposed force on the diaphragm. You would, though, have to design the diaphragm and lateral system to take the imposed deformation of the top of the wall.

I'm not sure how pinning the top reduces the reactions at the diaphragm, as lkjh mentioned, since without the top support you have no top reaction hence no imposed force.

 

[apsix]

"I'm not sure how pinning the top reduces the reactions at the diaphragm, as lkjh mentioned, since without the top support you have no top reaction hence no imposed force"

I took it to mean reduced relative to a wall pinned top and bottom.

 

[lkjh345]

apsix:

Correct. If you fix the base of the wall, the reaction at the top of the wall (to the 2nd floor diaphragm) will be less than if you use a wall pinned at the top and bottom.

In my case, I have a long, fairly shallow diaphragm to work with becasue of a two story entry lobby in the front of the building, so I am trying to reduce the load on the diaphragm as much as possible.

 

[UcfSE]

I misread it then, thanks.

 

[csd72]

It is hard geotechnically to justify that the base is truly fixed. Soil has to yield a bit before it takes up load, and this can reduce the effectiveness of the cantilever.

In this type of situation I would usually treat it as pinned top and bottom in order to get the top reaction (truth is it is somewhere in between).

Even if you assumed pinned top and bottom, it is often of benefit to put some cantilever action into the base so that the shores might be removed earlier to facilitate construction.

Dont forget that if you prop the top of the retaining wall then you must use 'at rest' soil pressure (Ko) as opposed to the 'active' soil pressure (Ka) that you use for a cantilever wall.

 

[271828]

I agree with csd72 in that it's hard to justify a truly fixed base. It's well known that a cantilevered retaining wall rotates and has lateral deflection at the top, as shown in CRSI.

I think the best solution for situations like this (questions about the wall interacting with diaphragm & other basement walls acting like shearwalls, etc.) is to design and detail the wall as a cantilever.

Assuming it's backfilled before the slab's connected, it will have already deflected so won't load the diaphragm.

If it's backfilled after the slab is in place, then you know it's ok because the cantilevered wall would've handled the load by itself. Surely it's ok with the diaphragm & other basement walls helping (adding them could make it fail somehow?!). Think lower bound theorem like used for steel connections. These are not brittle limit states, so the same theory applies.

Another advantage is that you don't really have to worry about things like when the slabs are in place like you do for a pinned-pinned basement wall. For example, I've had basement walls that were backfilled before the basement slab-on-grade was placed & cured. Probably really lucky not to have a sliding bottom of the wall. There's also no issue of bracing the top of the wall, construction sequence, etc.

I've done this probably a half-dozen times and haven't had any trouble. It was the most common practice at my old firm.

 

[271828]

Oh, I forgot one thing. When I do what I propose in my previous post, I reinforce the inside face also because there might be tension on that face depending on when the framed slab is placed.

 

[JAE]

Even if you design it as a cantilevered retaining wall, there will still be load imposed upon your upper diaphragm. This is because the initial backfilling does not constitute the entire lateral earth pressure that will develop over time on the wall.

Most geotech's will offer a design lateral pressure which is a maximum amount that can develop over the life of the wall. With fluctuations in moisture content, added fill, drainage characteristics, etc., the load on the wall may increase over time, well after the initial construction.

So designing a cantilever system to reduce loads on your diaphragm is not something I would fully rely on. As 271828 stated, "Assuming it's backfilled before the slab's connected, it will have already deflected so won't load the diaphragm" is true initially, but that same backfill can continue to develop more loading over time. I'm not aware of any way to figure out what percentage of post-backfill load will exist. The initial backfill load is NOT 100% of your design lateral pressure given you by the geotechnical engineer.

 

[271828]

JAE, I agree in theory with what you typed, but I still say it's similar to the approach taken by connection designers, the lower bound theorem. So what the diaphragm gets some of the load? The wall would resist it all if unassisted. The wall + diaphragm is stronger than the wall by itself.

Besides, we could come up with dozens of examples of elements we design assuming they take all the load, but in reality they're sharing the load with other elements. Assuming we're talking about ductile limit states, nobody in his right mind gets worked up over the loads that accidentally end up in these other elements. Of course, some engineering judgment is in order.

And finally, the proof is surely in the pudding for many hundreds (if not > 1000) of feet of wall that I'm aware of.

You don't agree? That's cool.

 

[JAE]

271828 - I agree that they share the load. You say "so what if the diaphragm gets some of the load".

I'm just reflecting on a project we once had where a building with buried walls on three sides (like the OP's description on this thread) had seriously deflected into the building, causing all sorts of distressed framing conditions throughout the walls and supported roof.

The original designer had assumed a certain level of earth pressure AND assumed that the wall would take it all. In this case, the diaphragm did take some of the total load and wasn't stiff enough to not distort under the gradual earth load which built up over time.

If you are just talking STRENGTH, then yes, you are absolutely right. The cantilever is designed for 100% of the load. It won't break. But what the OP was trying to do was reduce load on the diaphragm.

In reality, a diaphragm of stiffness k_d and a cantilever retaining wall with stiffness k_w are both participating. Force follows stiffness and if there is any value at all to k_d then there will be force sent into the diaphragm....some portion of the total.

I was just pointing out that you cannot count on the wall to take 100% of the force simply because its strong enough to do so. You have to distribute according to stiffness...not strength.

If the diaphragm takes SOME of the load (which it will) then you should, as a good engineer, account for that in your design.

 

[271828]

JAE, I agree with most of your points. Some engineering judgment is required to assess the ramifications of load going here and there.

In the building described in your situation, it sounds like a flexible (wood?) diaphragm, an under-designed wall, and/or a very tall wall which would have a large tip deflection even if designed properly as a cantilever.

For my cases, they've always been composite slab or CIP concrete floors. Basement walls usually wrapped some of the other walls also, so they were like shear walls at that level.

I guess it depends on the details of the situation.

Have a good rest of the weekend!

 

[Lion06]

271828-
I agree with JAE on this. I recently designed some masonry shearwalls (at the lowest level only) for a 3-story building (with flexible moment connections for the upper floors). The north and east sides of the building had 15' of soil retaining, while the south and west sides had virtually none.
I detailed and designed the walls as being supported at the first floor diaphragm because the masonry wouldn't work as a cantilever. This load obviously has to be taken into the diaphragm. I would definitely take some of the load into the diaphragm if I were tying it into the wall regardless of how it was designed for strength.
The point I am getting at is that even if the wall/diaphragm system is stronger than the wall only, those loads need to be accounted for in the diaphragm. Even if the wall is backfilled prior to tying the wall into the diaphragm - for reasons that JAE states as well as soil saturation, potential surcharges, I am sure I am missing a few others.
If those loads aren't accounted for (and they can be significant) the shearwalls could be underdesigned for shear and moment as well as the footing being undersized.
Also, as you point out, you must be careful to provide reinforcing for how the wall will act, not necessarily how it was designed. It is one thing to design it as a cantilever, but if it is supported at the tip and any additional load is added, there will be tension where it wasn't expected - but you did note this.

 

[271828]

StructuralEIT, your situation is different. I wouldn't do it that way either if I had masonry that wouldn't cantilever.

Geez, I should've dedicated a paragraph to qualifiers and caveats other than just stating that it required judgment.

It's fine if you guys don't agree. You have to be able to sleep at night after you design somethinig. I have many hundreds of feet of wall designed this way and I sleep fine too!

The concept is perfectly valid and safe for some systems, regardless of what you guys type, LOL!

 

[apsix]

271828
"If it's backfilled after the slab is in place, then you know it's ok because the cantilevered wall would've handled the load by itself. Surely it's ok with the diaphragm & other basement walls helping (adding them could make it fail somehow?!)."

Failure doesn't occur only because you also reinforce the inside face, and the diaphram is robust enough to resist the reaction at the top.
It may be designed as a cantilever wall, but in its permanent state it is a propped cantilever.

(NB failure does not neccessarily mean collapse)