Integrated Slab and retaining wall

[Chickenhawk6451]

Hi everyone,

I've been practising for about 5 years now and have done my fair share of retaining walls and have come across something interesting. To be short, the site geology is rocky with 0.5 - 1m of residual soils over the top of the bedrock. In order to design the footing I've been snookered into a situation where I need to combine my footing slab and retaining wall for this site. This is mostly due to the practicality of excavating rock in order to do a more traditional "Offset T" style retaining wall.

My question is this; It is a reasonable assumption that an integrated slab-retaining wall system would act as 1 whole system and utilise all 3 stems of the slab in resistance against sliding from forces in the retaining wall?

My logic is that each individual stem would act as a shear-key to resist the forces developed by the active soil pressure behind the retaining wall. I would consider that each stem acts as a shear key because they are sufficiently far apart that the failure plane of the soil from each stem would not affect the failure plane of the adjacent ones.

However I also acknowledge that globalised failure plane of the soil may exist, extending under the first stem and skip past the other two, leaving only 1 stem resisting the soil pressure. In this situation the slab will slide as there is insufficient depth available to develop sliding resistance from the stems.

Soil is gravelly sand with traces of silt, bulk density = 18kN/m³ and internal friction coefficient of 30°

Any guidance you have would be appreciated, hoping to create an efficient design for the client as this is a residential job so really tight budgets for all involved.

Kind regards,
Chickenhawk

 

[BridgeSmith]

That looks like a workable solution, with a couple of caveats:

1) It takes substantial movement to mobilize passive pressure - the wall may move well over an inch before passive pressure is fully mobilized.

2) It's not typical practice to count on passive pressure where there's any chance of the soil being removed (by erosion or by subsequent construction) in the future, so counting on passive pressure at the front face (left end in your sketch) is questionable.

 

[JAE]

Won’t your main sremwall rotate about the lower slab, creating passive on the OTHER side of the wall?

Where is the sremwall footing to hold up the wall?

Check out Eng-Tips Forum's Policies here:
faq731-376

 

[BridgeSmith]

JAE, it's a variation of what commonly refer to as a toe wall. Presumably, it will be designed with adequate moment capacity at all locations. The slab bearing on the soil provides the support for vertical load, which only consists of a small vertical component of the retained soil pressure and the self-weight of the stem.

 

[Chickenhawk6451]

[pre]1) It takes substantial movement to mobilize passive pressure - the wall may move well over an inch before passive pressure is fully mobilized.[/pre]

Correct, i'm hoping that utilising the weight of the structure over will prevent passive soil pressure being necessary. As always, conservative design practise is looking for the "worst case" scenario, not the most realistic one

[pre]2) It's not typical practice to count on passive pressure where there's any chance of the soil being removed (by erosion or by subsequent construction) in the future, so counting on passive pressure at the front face (left end in your sketch) is questionable.[/pre]

Agreed. I'm not sure how to approach this other than to stipulate active measures to be taken to prevent soil erosion. The site slope itself has been determined as stable by the geotechnical engineer. For now i'll ignore that stem, see what happens, design conservative

 

[BridgeSmith]

Wait, there's a structural load on the wall (other than the self-weight of the wall, which won't amount to much)? Where is the structure load applied? Is it located correctly and substantial enough to provide sliding resistance? Even if it is, you'll either need to account for the movement of the wall system necessary to reach active soil conditions (Ka) for the retained soil, or design for the at-rest soil pressure (Ko).

 

[Chickenhawk6451]

Wait, there's a structural load on the wall (other than the self-weight of the wall, which won't amount to much)?

Yes, estimated as about 6kN/m length. Its a single storey timber house over, not a large amount of weight

Where is the structure load applied?

Majority of the structural load is applied on a wall running perpendicular to the wall shown so, left to right in my sketch. This force is translated to the slab below and not to the retaining wall in question.

Is it located correctly and substantial enough to provide sliding resistance? Even if it is, you'll either need to account for the movement of the wall system necessary to reach active soil conditions (Ka) for the retained soil, or design for the at-rest soil pressure (Ko).

Not really, I've ignored it thus-far and figured any benefit form the extra weight and friction resistance will only help the system as a whole. Its conservative but i'd rather that then have the builder call me up and tell me the house has slid down the hill

 

[BridgeSmith]

Ok, I'm getting confused here. You said earlier "Correct, i'm hoping that utilising the weight of the structure over will prevent passive soil pressure being necessary." In your last post, you state "I've ignored it thus-far and figured any benefit form the extra weight and friction resistance will only help the system as a whole." Your sketch indicates the sliding resistance is generated by passive soil resistance on the stems. Can you clarify what is providing the sliding resistance for the wall system?

 

[Chickenhawk6451]

Recalculated all of the above on the basis of only 2 stems resisting but with the additional house over contributing to friction resistance also. I'm sitting pretty happy with this now, have I missed anything major?

 

[Chickenhawk6451]

Ok, I'm getting confused here. You said earlier "Correct, i'm hoping that utilising the weight of the structure over will prevent passive soil pressure being necessary." In your last post, you state "I've ignored it thus-far and figured any benefit form the extra weight and friction resistance will only help the system as a whole." Your sketch indicates the sliding resistance is generated by passive soil resistance on the stems. Can you clarify what is providing the sliding resistance for the wall system?

Sorry about that, I have been inconsistent with what I posted. I had hoped to not rely on the weight of the structure but as you pointed out, only 2 stems contributing is a more realistic scenario. In this instance, I then need to consider weight of the structure and slab also for friction resistance as passive resistance alone wont do it. My total resistance is a mixture of frictional resistance and passive soil resistance.

 

[BridgeSmith]

So long as you've made accommodation for the substantial movement necessary to mobilize the passive soil resistance (without moving the house or breaking stuff), you may be able to combine friction and and passive soil resistance. Maybe some of the geotechs here can speak to that. With axial load on drilled shafts, there is an incompatibility of the strains for side friction and end bearing, making it such that you cannot use the full values of both. I'm not sure if there would be something similar for this situation.

 

[BridgeSmith]

If you can't get a definitive answer on combining friction and passive resistance, I would suggest extending the back 2 stems deeper to provide enough passive resistance. If that gets you into the bedrock, I would likely apply the passive resistance as a point load at the bottom of the stem for the structural design of the stems, particularly the middle one. The considerably higher stiffness of the bedrock may produce an effect close to that condition.

 

[BridgeSmith]

Now that I think about it for a few more minutes, considering friction and passive resistance may double dipping, that is counting on the resistance of the same soil mass (what's under the slab between the stems) twice.

 

[Chickenhawk6451]

Now that I think about it for a few more minutes, considering friction and passive resistance may double dipping, that is counting on the resistance of the same soil mass (what's under the slab between the stems) twice.

Interesting and valid point. I've always assumed that both can be used because that's what i was taught as an graduate from a guy who had 25yrs experience in Iran, a highly seismic zone. This guy was a guru of industrial foundation design. Not to say he wasn't ever wrong but you've raised a very valid point. I'm going to check a few books i have on hand, maybe that will help clear it up

 

[MIStructE_IRE]

The books I use, primarily ‘RC Design - Mosley & Bungey’ combine friction and passive in order to resist sliding. I have always done this and have had no problems to date. I suspect had I not done this I’d be laughed at when I show others the size of base I’d need otherwise!!

 

[BridgeSmith]

Under normal circumstances, where the friction is under the heel of the wall, using the weight of the retained soil, and the passive resistance is provided by the soil in front of the wall, I would agree with the validity of using both. In this case, it would seem that it's the same soil mass for both - the soil under the footing slab between the stems. If that soil is pushed by the stem, can it still be counted on to resist the movement of the slab, or do both the slab and the soil directly under it move together, which would negate any frictional resistance?

 

[BridgeSmith]

I don't know the configuration of the structure or the remainder of its foundation, but you may be able to consider it as providing resistance to sliding. You'd have to carefully consider the entirety of the load path from the footing slab, through the connection or interface with the structure foundation, through the structure, to the other foundation elements. Typically, that would make available a substantial amount of resistance. If the wall extends beyond the structure, you'll have to consider longitudinal bending and shear on the footing slab. If you go with a rigid connection to the structure, you'd have to use Ko for the retained soil pressure and neglect the passive resistance, unless you're ok with the structure moving.

 

[Chickenhawk6451]

The structure as a whole is best described as a 3-sided box with the open side facing a downwards slope. Total dimensions are 12m x 6m. Retaining wall is on the 3 sides enclosing, open side is for vehicle access to park under the house. Retaining wall does not extend beyond bounds of the foundation. Timber house sits on top of the box. My concern is the retaining wall pushing the whole building out the "open" side of the footing down the hill

My soil mechanics books are in the office so i cant check this myself right now. I would have though the additional load from the building would act like a "stabilising surcharge" in the same way you can have a destabilising surcharge, increasing the soil resistance. Increasing the capacity because the soil cannot achieve it's failure plane due to extra weight "squeezing" it all together. Basically the stabilising surcharge is a confining load increasing the soil strength. Again, i need to check my books in the office tomorrow to see if that's realistic

Perhaps I should just do a single deep stem on the right hand side (basically, a shear key) and consider the weight of the building contributing to friction, that would eliminate the failure plane running under the 2 stems

Gah, a "simple project" they said

 

[txeng91]

Just a thought, maybe put beams running perpendicular to the wall at 4m or 6m spacing that tie into the stem, may need to widen the stem below the wall. The stem then transfers the overturning from the wall through torsion into the beams tieing into it. No passive resistance required, all friction and bearing.

 

[Chickenhawk6451]

Ok, so, after much deliberation, I have a solution that passes the "sniff test";

1) soil passive bearing reaction is ignored on all stems
2) Weight of house on slab + weight of slab + weight of trapped soil between stems was considered to provide weight to resist in friction alone

End result, a pass. Will have my checker look it over, I'm not seeing a safer assumption and this should give a more conservative result as well as ensuring serviceability criteria is met

txeng91: Essentially my design ends up being a "grid" of stems to transfer the loads adequately similar to what you're suggesting