Cantilever Retaining Wall Kern Point

[zdude]

Hello,

With respect to the design of concrete, cantilevered retaining walls, I have been wondering what the consensus of the group is.

In designing the walls, keeping the eccentricity within the middle 1/3 of the footing generally means that you have no uplift on the heel of the footing. In general having the heel come off the ground is undesirable for a variety of reasons. When I design these walls, I try to keep the eccentricity within the 1/3 point while balancing bearing pressure, sliding and overturning.

My question is in regards to seismic loads. Provided the safety factors for overturning, slide out, and allowable bearing pressure are OK, what is your thoughts on heel uplift and the eccentricity under seismic loads?

 

[msquared48]

As long as the soil at the toe is not overstressed, I have no problems with it. That's what the equations were developed for - to use if needed. It can also cause sliding issues to deal with if the heel as not engaged enough dead load.

It can tend to be an inefficient design, but sometimes that cannot be avoided due to property line and building constraints.

Mike McCann
MMC Engineering
Motto: KISS
Motivation: Don't ask

 

[bridgebuster]

I agree with Mike

 

[hokie66]

I don't worry about keeping the eccentricity within the kern for normal loading, so I definitely wouldn't have an issue with seismic. If all the statics are satisfied, and the pressures are within bounds, sticking to the kern is not necessary.

 

[weab]

I learned in college to keep the resultant within the kern for operational loads. One place I worked would allow the seismic resultant within the middle 2/3 of the footing. This means that 1/2 of the footing is in compression, 1/2 in uplift.

I haven't heard anyone say it, but I would treat this issue like ductility. Sure, the soil may be able to carry load beyond the kern, but the foundation (structure) must survive the event. It seems that if you keep the resultant within some limits, you allow for loading outside of those limits. Maybe the soil will give, but at least the foundation will not tip over. I'd rather this happen rather than having the foundation tip but the soil didn't fail.

If you design for operational and seismic loads outside of all limits, I would think that you lose "ductility", for lack of a better word.

I am wide open to any thoughts on this matter.

 

[msquared48]

If the soil does not fail, how is it going to tip?

Mike McCann
MMC Engineering
Motto: KISS
Motivation: Don't ask

 

[apsix]

If it's unstable.

 

[zdude]

Thanks for the thoughts,

I have looked at the kern point as a goal under operational loads. And typically, if you can meet the allowable bearing stress and a safety factor of 1.5 for slidng and overturning, the e is within kern.

For extreme events like a seismic event, I have approached this similar to jsdpe. The e can go outside the kern, and then there is a possibility that the heel will uplift and some soil may fill the gap leaving the wall "off-plumb".

This came up in discussion with another engineer in our office who was taught that the e should always be within the kern.

 

[msquared48]

apsix:

That's why we have factors of safety against sliding and overturning when we design them. There is no design instability if those factors are not exceeded, to include the soil bearing pressure.

Mike McCann
MMC Engineering
Motto: KISS
Motivation: Don't ask

 

[weab]

We design for factored loads. If we actually apply 1.2 DL + 1.6 LL, etc., then life should go on as if nothing happened. I'm talking about when those conditions are exceeded. We'd all like to see a ductile failure. In steel structures, steel yields, bends, and the structure may be unusable, but usually it won't fall. A concrete structure cracks, spalls, etc, but hopefully it won't fall catastrophically.

However, this is a stability issue. If loads are exceeded, the wall can pivot about the toe and tip over. It's OK if the soil is overstressed and the wall permanently leans. That would be OK. What's not acceptable in my opinion is if the wall falls over. If a design extreme loading places the resultant too close to the toe, then it may only be an incremental load increase which tips the wall. In the end, it's still a judgment call.

 

[hokie66]

Not many would agree that it is OK for the soil to be overstressed and the wall to permanently lean. But having a triangular stress distribution under the wall rather than trapezoidal does not mean that the wall will lean or that the soil will be overstressed.

 

[apsix]

msquared48

I'm not quite understanding; "There is no design instability if those factors are not exceeded, to include the soil bearing pressure", but stability and bearing pressure are separate checks.

Acceptable bearing pressure doesn't guarantee not failing a stabilty check, eg. when the wall is founded on rock.

 

[weab]

Hokie

I didn't intend to say that it is OK for anything to be overstressed or lean BY DESIGN INTENT. However, it is perfectly acceptable (and indeed preferable) for a structure to fail in a ductile manner as opposed to a catastrophic manner. I think Apsix said it better than me.