Seismic Design of Segmental Retaining Walls using IBC

[NewcastleEngineer]

I am curious as to how engineers in the U.S. are designing segmental retaining walls for seismic events using IBC or ASCE7. These documents use spectral response accelerations, which appear extremely conservative. I have typically seen peak ground accelerations at either 10% exceedence in 50 years or 10% in 250 years being used by AASHTO and NCMA. I know the IBC indicates teh design methodology for retaining walls is up to the design professional, but ASCE 7 doesn't appear to be as vague.

What design coefficients (spectral accelerations or horizontal peak ground accelerations)are engineers using with segmental walls in states that have adopted the IBC or on projects referencing ASCE7? Also, I would appreciate hearing any aguments for not using spectral acceleration coefficients with segmental retaining walls. Any responses would be greatly appreciated.

 

[cap4000]

I think you have a great thread. I myself would be intersested in any responses. To give you my understanding, it goes like this. I have the AB Walls 2000 Design CD by Allan Block and with just a 0.1 or 10%g forces you have to increase the grid lengths and perhaps add more layers. This is a Big Time change from just static loads. Then if you look at graphs by Lam and Martin,1986 with 0.1g with various soil friction angles it plots the seismic active pressures Kae. There is also the Mononobe-Okabe theory. To answer you directly under a small 0.1g force and a soil friction angle of around 35 degrees their should be very little change with the typical Ka values. Proceed very carefully with segmental wall designs. Good Luck from a New Jersey PE.

 

[BigH]

I don't think that 0.1g is all that small. Your "unknown" earthquake areas - say ringing in the Dayton Oh area, Ottawa, Toronto are all in the 0.06 range. These are the "high" points for interior, if I remember my maps correctly. Only in the west coast that you get 0.17g or so for Vancouver. Of course, when you compare it to Indonesia and other Rim countries where ag is about 0.25g to 0.3g - well, that is getting large!

 

[NewcastleEngineer]

BigH:

What probabilities are you basing the seismic peak ground accelerations on? They soud like 10% probability of exceedence in 50 years based on the 0.06g value for Dayton, Ohio. What probabilities of exceedence are most appropriate for SRW design, while still being in line with the building codes?

I am still interested in hearing how engineers have dealt with the IBC and ASCE7 in SRW design. These codes only mention spectral accelerations, which result in much more conservative seismic accelerations compared to peak ground acceleration (PGA). Particularly, the 0.2 second spectral acceleration values are typically between 3 to 4 times higher than PGA values with a 10% probability of exceedence in 50 years.

CAP4000: Be careful using AB2000 software. It uses its own hybrid method of internal analyses which is inconsistent with both NCMA and AASHTO design methodologies.

Thanks for the responses.

 

[cap4000]

NewcastleEngineer: The AB Walls 2000 CD Design Literature indicates it uses the Mononobe-Okabe Earthquake formulas.
As best as I can tell the "Typical Segmental Wall" is more flexible than a rigid concrete retaining wall and as such handles earthquakes pretty well. I wonder just for curiosity how many Rigid Concrete Bridge Abutments have failed .vs. the Flexible Segmental Walls?? In New Jersey,
I use IBC the adopted code at 0.1g, not doing so leaves me open for a potential, albeit a very remote possibilty of a lawsuit. Imagine the legal grief and costs for lawyers to argue over using either AASHTO, NCMA, IBC and/or ASCE-7. My guess is that the legal fees probably would exceed the cost to build even a very "Conservative Wall Design times 3+/-.