In many practical situations, it is convenient to neglect the soil bearing pressure on the underside of the heel.

For many retaining wall proportions, the soil bearing pressure will be small at the back face of the stem and end not far into the heel, so the assumption is not highly consequential.
It can be challenging to precisely determine the soil bearing pressure and distribution, especially at an ultimate state (when determining the design actions for your RC heel).
Typically, other considerations of constructability (bar spacings, bar sizes on order, labor) will be more important than providing the least weight of steel when specifying heel reinforcement. Minimizing stem and toe reinforcement is typically the focus for an efficient design.

----
just call me Lo.

Our interpretation of the AASHTO spec. is that it does not allow for the bearing resistance (the "uplift force") to be used for semi-gravity (cantilever) retaining walls, so we always design using the first loading case. The extra concrete for the thicker footing to resist shear is a bigger cost than the rebar, but both are typically fairly small in the overall project costs.

Rod Smith, P.E., The artist formerly known as HotRod10

Lo makes good points on the uncertainty of the distribution (I believe that is the reason the AASHTO spec. disallows its use) and the limited effect on the design loading if it were included.

Rod Smith, P.E., The artist formerly known as HotRod10

Thank you gentlemen for your replies.
Allow me to provide some results please in order to decide.

Factored Vu (Heel)= 1.6 x WBA + 1.2 x Wc = 3979 lbs. ⬇︎ (Down)

Vu (Bearing Resistance) = 1165 lbs ⬆︎ (Up).

So, according to above, I can ignore the 1165 from Bearing resistance and just use the 3979 lbs?

The Bearing resistance force is almost 30% of the Factored Forced applied on the Heel.

if AASHTO spec. it does not allow for the bearing resistance (the "uplift force").
Can be the same for ACI code?

The uncertainty comes in determining what is the shape of the bearing resistance. It's going to be a partially distributed load, but is it uniform, triangular, or some nonlinear distribution? Since it's bearing on a material with nonlinear stress-strain behavior (soil), a nonlinear bearing pressure distribution should be assumed. Since the shape of that distribution would be difficult or impossible to determine, the only known conservative assumption is to ignore it. The distribution of load shown in the diagrams in AASHTO are conservative for the determination of moments on the toe of the footing, but would not necessarily be conservative for determination of a reduction in moment or shear on the heel.

So yes, per our (WYDOT Bridge Design) interpretation of AASHTO you would use the 3979 lbs, with no reductions.

Rod Smith, P.E., The artist formerly known as HotRod10

This is the way I do. Be careful on load factors tracking during stability analysis.

I have always designed it according to how BridgeSmith and Lomaranandil described. So, yes, I would ignore the 1165 lbs upward force. I have always used AASHTO code. The bearing stress distribution under your footing can be either one of three conditions:

1. Uniform (Never happens in practice)
2. Trapezoidal (Typically occurs if sliding governs the design)
3. Triangular (Typically occurs if stability (overturning) governs the design)

I think most of my designs usually fall into the 3rd category. Since the bearing stress distribution is so low on the heel, it won't much matter if you consider the bearing stresses to reduce the downward weight of the soil above. Just reiterating what Lo said.

Conservatism is fine. But what is the design force in the slab below? Note that w is the upward pressure in the system.

Thanks a lot all for your help, it is very clear now.
God bless you gentlemen.

Retired13 -- the design shear is pretty straightforward there, but what is the design moment? Presuming your upward resistance is from soil, that uniform distribution may not be so uniform (even at a service state, neglecting any non-linearity).



An imperfect analogy, but it illustrates some of the uncertainty in bearing stresses. Uniform, triangular, trapezoidal bearing pressures are handy approximations, but they're less precise than a lot of other assumptions we make in structural engineering. That's even assuming a uniform bearing material itself -- no local effects due to a large boulder or soft zones due to existing hydrology.

In a case where the load distribution under the retaining wall is as trapezoidal as your first sketch shows, yes, that soil pressure probably has enough benefit to the heel design to merit consideration (where allowed by code) although I would probably still discount it's effect somewhat. However, I don't tend to see retaining walls proportioned that way.

----
just call me Lo.

If I provide 8 in of Base course over clay, the diagram of bearing pressure down above will remain as it is or it will be uniformly distributed?

XinLok,

It is good to keep in knowledge the non-linear stress distribution when dealing with different soil mediums, but don't feel been shortchanged by practically perform hand cal using straight line approximation in foundation design. If you wish, you can get more exact results by incorporate soil springs in your model, and analyze through "beam-on-elastic foundation" theory, or FEM.

The inclusion of soil pressure in heel slab design is due to the fact that the system is in structural equilibrium globally, but not locally. The design is to make up the difference in between the applied force (due to weight above) and the reactive pressure beneath the slab to maintain the equilibrium everywhere along the heel slab. I hope this make sense to you. But you might want to, as advocated by many others, ignore the soil bearing/reactive pressure to be conservative. That is perfect fine too, IMO.

Everything is very clear now.
Thank you a lot all