Retaining Wall Design with Seismic Loading

[Chris Poulsen]

When designing retaining walls I have referred to the lateral pressures provided by the geotechnical reports. The reports have always included standard active pressure and an active pressure with seismic loading included (e.g. 55 pcf active and 85 pcf active with seismic loading). I have always designed with the seismic loading which is the higher of the two and still designed the walls to with a 1.5 safety factor against sliding and overturning. I just barely noticed a section of the building code (IBC 1807.2.3) which states an exception to the 1.5 safety factor as, "Where earthquake loads are included, the minimum safety factor for retaining walls sliding and overturning shall be 1.1".

From this, it appears that the retaining walls have been over designed at that if the larger active pressure with seismic loading included is used the safety factor can be reduced to 1.1. I have worked with several engineers and none have ever mentioned this before. Am I missing something or is it alright to use a lower safely factor when including seismic loading for the active pressure?

 

[r13]

In the past, the structures were allowed to have an overstress of 33% when wind and earthquake loads were included, which resulted in an effective safety factor equal to 1.0. The reason for that was due to transient nature of those loads. The stance of the code has since changed, that the 33% overstress was no longer applicable for design using ASD. However, since the earthquake load is scaled up in LRFD, the 1.1 safety factor seems justified (refer to IBC 1605.1.1 Stability" for more info).

 

[STrctPono]

I don't know IBC at all but AASHTO bridge code doesn't directly perform an overturning check but performs an "Eccentricity" check where you evaluate where your resultant force is and thus your kern. It's iindirectly an overturning check. When including seismic loads, AASHTO code allows a more lenient value on the kern dimension.

It seems perfectly logical that an extreme limit state check should be allowed a more relaxed factor of safety when compared to a strength limit state.

On a side note, I've never seen earthquake loads given as a trapezoidal load distribution. Mononobe-Okabe forces typically are expressed as a uniform load but I'm no geotech.

Also, do you consider the inertial weight of the wall itself?

 

[EDB9]

If the code permits a reduction in the FOS, then it is permissible. Risk definitely factors into the basis for the reduction because it isn’t a regularly occurring load like your traditional D and L loads.

Just because the code says you can, doesn’t mean you have to. I interpret the section as it states to design to a minimum of 1.1 or greater. Conservatism is never a bad thing. Do you know what soil types are behind the wall? You could also have other potential issues if you’re designing for seismic purposes as well which is why I ask.

The bottom line is, the code is what we go by. If you want to add a little margin, it never hurts. It covers you in case there are any unknowns and provides a better design, even if it’s a little beefier than it has to be.

Erica