Foundation Wall for Building on Fill

[Eng_Struct]

Hi Group,

I'm working on the design for a new expansion to an existing building. On the expansion side, the exterior grade is approximately 4 feet lower than the interior slab-on-grade elevation.

The original building, constructed around 20 years ago, uses a standard strip footing and a foundation wall with the top of the wall tied into the interior slab (see image below). What’s puzzling is that the wall shall be acting as a retaining wall, but there is no conventional heel and toe reinforcement detail.

In my experience, for a wall retaining 4 feet of soil + 4ft embedment (total 8ft retained height), I would typically design it as a retaining wall with a larger footing (heel and toe) to resist sliding and overturning. However, the original engineer stated that tying the top of the wall to the slab provides sufficient restraint, eliminating the need for a larger heel.

The owner now wants to replicate the same foundation wall system for the new expansion. I've advised that we may need to increase the footing size, but they remain unconvinced.

A few concerns I have:

• I usually ignore passive resistance within the frost depth, especially when there’s no weeping tile around the perimeter to ensure drainage.
• Without relying on passive resistance, the lateral earth pressure alone seems too large to prevent sliding at the base without an extended footing.
• It seems the wall would rotate about the top tie-point as the base slides outwards - also the detail at the top is not a moment connection to make it act as a cantilever.

Despite this, the existing system has been performing without issues for 20 years, which makes me wonder what’s actually resisting the soil pressure.

Has anyone encountered a similar detail? Can the top tie-in really eliminate the need for larger footing to resist sliding? Any insights or experience would be greatly appreciated.

 

[phamENG]

Ignoring 4' of embedment feels overly conservative to me, but I don't work in frost country so I'll let somebody else weigh in on that.

For me, the use of a retaining wall footing is a question of construction means and methods. Do they want to erect and maintain shoring for the wall until the slab is in place and sufficiently cured? If yes, then you can go with the smaller footing. If no, then it needs to be self supporting during construction, including any equipment that will be used to place and compact the soil. While the final structure cost is likely less with the former, that shoring does have a cost so it's not as cut and dry a cost savings as it may appear.

 

[milkshakelake]

Wouldn't restraint at the top reduce the sliding and overturning forces by throwing some of it into the soil? Similar to using geotextiles or tiebacks.

 

[XR250]

I usually give the contractor the option and let them decide. I am working on a similar situation currently with a residential garage on a sloped site.
I will usually show the rebar in the center of the wall so it gives them a bit of extra capacity when they decide they really don't need to shore the wall. It would normally be most efficient to have the rebar biased toward the non-fill side for a tied wall.

 

[jhnblgr]

Top restraint
Metal building pilasters spaced every so often
Active and passive soil pressures

I’d be more worried about the over turning moment on the metal building foundation. All the load paths really end up there.

 

[Eng_Struct]

Thank you everyone for the feedback.

Yes, the designer in the example I have provided appears to have specified the shoring of the wall before the placement of the slab.

My question is if I do specify shoring of the wall and tie it at the top, does the wall become simply supported with supports at the top and bottom? The top slab will need to have adequate requirements and a load path to take the tension out (either friction between the slab on grade or the slab tied to the existing slab-on-grade using hilti anchors). The bottom will need to be held back by friction at the base or passive pressure potentially.

To me, this system starts to look too much reliant on tension in slab on grade for lateral force resistance.

 

[Celt83]

What do you see as the load path for your slab on grade to provide lateral restraint to the top of the wall?

In my experience that generally won't end up working out so tend to design these conditions as true retaining walls.

 

[GaStruct]

I almost never rely on a SOG to brace the top of a RW wall. Friction resistance of the SOG against the soil is never enough (especially with a vapor barrier) and I hate dumping a ton of tension into a SOG. They are never thought of as structural, so it creates more of a risk with shoddy construction or someone coming along later and ripping it out. Plus, I don't want any cracks so I prefer to let them float.

 

[milkshakelake]

I politely disagree. It should be fairly simple to specify something like a 6" slab on grade with #4@12" each way, relying on the friction on the soil and rebar as drag elements (as long as they are properly designed for the force and properly spliced). Use lower friction coefficient for vapor barrier if needed. Those slabs aren't going anywhere. If we can't use that, then I wonder how a drag system like geotextiles can work.

 

[Eng_Struct]

What do you see as the load path for your slab on grade to provide lateral restraint to the top of the wall?

If I have to make it work, I will reinforce the slab on grade and tie the new slab using Hilti into the existing building foundation wall. Essentially the slab works in tension.

The other thing I will need to consider is the passive pressure on the outside to make it work.


I want to design it as the retaining wall but given the existing building is working in this condition, the owner does not spend money on larger footings.

 

[Celt83]

If I have to make it work, I will reinforce the slab on grade and tie the new slab using Hilti into the existing building foundation wall. Essentially the slab works in tension.

The other thing I will need to consider is the passive pressure on the outside to make it work.


I want to design it as the retaining wall but given the existing building is working in this condition, the owner does not spend money on larger footings.

the tension in the slab is only part of the load path need to terminate it somewhere

if relying on slab/ground friction you need to make sure you have enough building length and also discount a portion of weight within the active soil wedge.

 

[GaStruct]

if relying on slab/ground friction you need to make sure you have enough building length and also discount a portion of weight within the active soil wedge.

This is correct.

A 6" slab is 75 plf without friction coefficient reduction. Add that in and it never works to resist the reaction in my experience.

 

[Eng_Struct]

Sounds like for the strip footing system to work, I need:

1. Find the load path through the slab in tension either via friction or tie to the foundation wall on the opposite end. (I only have 16ft of building length so there is not much to rely on friction - I will may introduce a key at the opposite end of slab-on-grade and dowel it into the existing slab or foundation wall).

2. Also consider passive to ensure that the base does not slide and the wall acts as simply supported between top and bottom points.

 

[HTURKAK]

.....or tie to the foundation wall on the opposite end. (I only have 16ft of building length so there is not much to rely on friction - I will may introduce a key at the opposite end of slab-on-grade and dowel it into the existing slab or foundation wall).

Yes . The best reliable load path is tying to the foundation wall at opposite end. In additon you may introduce a shear key to mobilize the base friction in case of vapour barrier.

Also consider passive to ensure that the base does not slide and the wall acts as simply supported between top and bottom points.

This approach is also reasonable. The horizontal active thrust at the bottom of the wall would be resisted by ( base friction + passive thrust ).

 

[Ron247]

The original building, constructed around 20 years ago, uses a standard strip footing and a foundation wall with the top of the wall tied into the interior slab (see image below). What’s puzzling is that the wall shall be acting as a retaining wall, but there is no conventional heel and toe reinforcement detail.

From what I see, providing a reaction at the top does not eliminate the need for any other component unless statics, load paths and analysis agree it does. I have seen others say something eliminates the need for something where more accurately, it reduces the magnitude of the needed "whatever". If you have all the original design info, perform your own analysis, do not rely on theirs. Also, I did things 20 years ago, I would not do now.

However, the original engineer stated that tying the top of the wall to the slab provides sufficient restraint, eliminating the need for a larger heel.

It "may" because it reduces lateral load. The math will tell you. One thing he could not do, was increase the top force to any value he wanted without changing something in the layout or load path.

The owner now wants to replicate the same foundation wall system for the new expansion. I've advised that we may need to increase the footing size, but they remain unconvinced.

They are unconvinced, but can give you no reason why? Owners are owners, but they would also love for the cubic yard price of concrete to be the same as it was 20 years ago. I would cave-in based on my calcs, not their dissatisfaction. Problem is, you have to be tactful. Codes change; recommended practices and acceptable "optics" change all the time.

A few concerns I have:

• It seems the wall would rotate about the top tie-point as the base slides outwards - also the detail at the top is not a moment connection to make it act as a cantilever.

Remember, soil is closer to a marshmallow than it is to steel. Soil compression occurs and therefore, so does deformation. It will try to bend that top connection.

Despite this, the existing system has been performing without issues for 20 years, which makes me wonder what’s actually resisting the soil pressure.

I did better my first 20 years than I have my last 20. How often has it been subjected to design floor loads? After the addition, are the floor loads going to change?

Lastly, if you are relying on the slab, it needs to be made clear that the slab cannot be cut out or removed for X' of the wall. No parallel trenches, or underground anything without structural review.