Stairs Bracing Retaining Wall

[Apache1]

Hi all. I got a call on a retaining wall that someone is installing to access their basement. As shown below there will be a set of stairs down to the basement. One side of the stairwell is a new retaining wall, the other side is an existing 7' tall foundation wall below the garage. Apparently the garage fnd wall had to be that tall to get down to good bearing soil. The reason I got a call is because the builder already had another engineer detail this retaining wall, but the builder thinks it's overdone. It was engineered as a true cantilever retaining wall, so it ended up with a 12" deep x 3'-6" wide footing, #4 bar at 12" on the retaining face, #4 bar at 12" longitudinal. In my opinion I think the other engineer did what technically "should" be done on a wall like this. However, I will admit that I have done stairwell walls like this before with a 20" wide footing and #4 bar at 16" in the center of the wall, with the reasoning being that the concrete stairs provide bracing against sliding and overturning for the retaining wall as they go up. With the stairs providing bracing, lateral soil pressures that will most likely never develop, tying the new wall into the existing concrete on the side of the doorway, and the short length of the wall, I figure that a retaining wall that will triple the owner's cost is unnecessary.

Curious what thoughts you all might have on this.

 

[phamENG]

With or without the stairs, the loads will still be the same. The stairs only help to resist those loads if detailed to do so. The original engineer's design probably gave the contractor the greatest flexibility in sequencing and didn't rely on that connection. After all, what happens if those stairs need to be replaced down the road? Will they know to excavate behind the wall to prevent overloading the wall when they are pulled out? And what about the low end of the stair? A connection to the existing wall would help, but what keeps it from falling in? Some residual resistance from the stairs a few feet away forcing the wall into a complex multi-directional response? Probably. But are you going dismiss another engineer's work on judgement alone, or will you do a rigorous analysis to prove it unnecessary? (No judgement there - sometimes that's acceptable but sometimes it isn't...I usually assume it isn't when I walk into something until I can prove otherwise).

And to be honest...a 7' retaining wall with a 3.5' footing does not feel overdone to me.

My bigger concern here is the existing wall. If it wasn't designed to be a retaining/basement wall, how confident are you that it will be able to do it now? Any chance the original engineer assumed some of the load that wall might otherwise resist is finding its way into the new wall and that's why he's using a 'larger than standard' footing?

 

[Celt83]

The load path doesn't stop at the stairs if you decide to use them for bracing.

3.5' foundation for a 7' tall wall actually sounds too small for prescriptive soil values.

I'm making a thing:

http://www.thestructuraltoolbox.com

(It's no Kootware and it will probably break but it's alive!)

 

[le99]

The builder's engineer is wise to provide standalone footing and retaining wall so not to disturb the existing structure, or at least keep the disturbance to a minimum. IMO, unless the new addition is enclosed, the existing wall will experience additional soil pressure passed through the stair slab.

 

[Apache1]

what keeps it from falling in? Some residual resistance from the stairs a few feet away forcing the wall into a complex multi-directional response? Probably. But are you going dismiss another engineer's work on judgement alone, or will you do a rigorous analysis to prove it unnecessary?

Well that right there is one of my hang-ups on projects like this. I'm not aware of any analysis tool that could determine how much the stairs are helping out. Which to me, leaves only one option: Design the full length of wall as a 7' tall cantilevered retaining wall without any bracing. When I run that analysis through my retaining wall calculations I end up with a 4'-6" wide footing due to sliding (as Celt83 alluded to above). Again, I am not doubting the other engineer because as I said above, I think the 3'-6" footing he's got is what should be done but even what he has doesn't "work" if I run numbers on it. But let's ignore the garage foundation wall and say you are building two new walls down to a basement. Are you really going to put a 4' wide footing under each of those walls? With 4' between the walls you'd just end up with a massive 6' wide x 12" deep concrete pad for the full length of the stairs....that seems way beyond overdone to me.

Do we have to "prove" that everything works? If so, how is anyone building stairs down to their basement without 4' wide footings surrounding the whole thing? Even this 3'-6" wide footing doesn't work. Sorry, I'm ranting now...

 

[Apache1]

the existing wall will experience additional soil pressure passed through the stair slab.

Thanks for your comment. The garage wall is fully backfilled on the opposite side up to the slab. Could it not be said that this additional pressure is transferred to the soil on the other side of the wall?

 

[Celt83]

I believe Pham's concern stems from the construction condition of the existing wall, the garage slab on grade is likely isolated form the top of wall:

For the final condition your going to need to use the stair to actually brace the existing wall out to the new stair retaining wall. You could analyze it on a strip basis or use something like the PCA below grade tank guide, or any software package that can handle plate/shell elements. Don't forget the handrail load on top of the new foundation stem along with residential surcharge on the new foundation and garage surcharge on the existing wall.

I'm making a thing:

http://www.thestructuraltoolbox.com

(It's no Kootware and it will probably break but it's alive!)

 

[phamENG]

Do we have to "prove" that everything works?

Only if an AHJ asks you to, or a perhaps a judge. The better question is "should everything we design be provable"...and to that I would say yes. You should be able to justify everything you stamp. Do you need to run an analysis on every set of floor joists that are 8 feet long and spaced at 16" o/c? No, but if somebody asks you'd better be able to show why the work.

If this were new the two walls would probably be on a common footing.

There are plenty of 3D modeling software packages out there that would let you model this with a sloped support. Do you have RISA or a similar program? You could also look into some yield line anlaysis. It would be cumbersome, but doable by hand. Though for something this small, I suspect the larger footing will cost less than the design effort.

Since you've already proven that what the other engineer designed doesn't quite "work" based on a conservative assumption, I would pass on the project and inform the contractor that you believe the other engineer has already reduced the design as much as is reasonable in light of certain unique aspects of the job. Or something like that, anyway.

 

[phamENG]

Thanks, Celt - you're spot on there. That's how most garages are detailed around here, especially if they aren't supporting a basement wall.

 

[Apache1]

The better question is "should everything we design be provable"...and to that I would say yes

I like this and I wish it was possible, especially in residential. I feel that there are a number of things commonly done and assumed in residential work that cannot be proven (roof diaphragm shear going over peaks and through valleys out to the shear walls, knee braces on exterior columns providing lateral stability, lateral loads from tall basement foundation walls into floor diaphragm, etc.) Thanks for your input. Sounds like I need to look into some better software. Is RISA what you'd recommend? I'm not very smart, so I assume the learning curve would be so steep that I'd fall and roll back down the hill

If this were new the two walls would probably be on a common footing

So a ~6' wide footing that encompasses both walls? Wow, I've been using 2' wide footings under each wall for situations like that...Is that something you can do in RISA?

 

[le99]

@Apache1,

You have changed the loading on the existing wall from a linear varying pressure to mixed with a concentrated load at every step of the stair. As a propped cantilever, the steps will receive a considerable amount of force, and, the load is varying with the height, which can cause an adverse effect on the existing wall. All these additional forces/effects could be manageable by the existing wall, but it needs to be verified. Is the saving worth the verification effort? This question can only be answered by you. The contractors' engineer obviously said, "not".

 

[HouseBoy]

Seems to me the the existing garage wall will most likely work when everything is in place BUT.... maybe not during construction as others have suggested (until the new stairs are installed).

The newly exposed area of the existing wall does not seem very large and therefore bending stresses shouldn't be that large. 7 ft is not very tall. I'd guess that plain concrete might work for the stresses that will result. I wonder what reinforcing is present in the existing garage foundation wall. Hopefully it is similar to what they would/should have used at the garage foundation wall once it passes into the existing basement area.

I don't see the new stairs as causing some kind of adverse stresses in the existing wall. That wall is backed by the soil that is underneath the garage floor. Elevated concrete slabs sit on 4" wide concrete ledges and I don't think there is too much of a problem with the stress concentration there. Yes, the steps produce an odd geometric pattern but again, the wall height is not so great that the force will be great. It's a big increase from zero but that doesn't seem like it should be a problem.

I also don't think the original design is "too conservative" although I would definitely have considered the new wall spanning horizontally and transferring loads into the existing wall they the stairs as a viable option (yes it needs to be doweled not the existing basement wall too).

You gotta dig the same hole and form the footing. It's a little more steel and concrete but the scale of the job is not that large so the total number doesn't seem like a big number. Agree with others who have indicated the difficulties in "proving" something less will be fine. Hardly seems like it is worth the cost for that and even more concerning is the risk that someone would be taking on to try to save a small amount.

 

[mfrad]

Thanks Celt83; very useful and well organized and in one place. I bookmarked it.

 

[le99]

I think the diagram below shows the difference in pressures/forces on the existing wall when complete. I personally will do a quick and dirty evaluation to see the difference numerically.

 

[HouseBoy]

Oops! I did not understand the garage floor to be at near basement level.
I assumed the garage was at approx. 1st floor level and the foundation wall stepped down to reach the basement level.
Seems like it would improve the existing wall performance to take some soil away from it.
Still thinking that the soil load from the new stair wall is not significantly different with or without the stairs there.

 

[phamENG]

le99 - where did you get that they're lowering the garage?

HouseBoy - I'm pretty sure your interpretation about the location of the garage slab is correct. All of the OP's sketches show the garage slab up at grade level.

Apache1 - all of the things you mention are quite provable. Is it practical to prove them mathematically every time? No. But they are qualitatively understandable and can be shown to work. If you're talking about proving things that "work" in the real world but "fail" in calculations, then yes it's a bit harder. But an understanding of safety factors and the statistical reliability goals on which the codes are based help a lot. I'm fairly certain RISA would do that. It's been a few years since I used RISA. I don't use 3D modeling software very often, so while I want to get it I can't really justify the expense. There are plenty of lower cost options. IES Visual Analysis is another good one and costs a lot less. And yes, a 6ft footing would make sense. One, it's easier to justify that it'll work, and two it's a negligible cost difference. They are already paying for the truck (capacity of about 10 cu. yards of concrete) for the footing pour that's less than 2 yards anyway, and that 2' sliver down the middle adds less than a yard.

 

[le99]

If no living space behind the wall, it is a footing, which will be benefitted by removing some of the soil it was facing. I wonder about the reason for the deep footing.