Need a book? Engineering books recommendations...

Return to index: [Subject] [Thread] [Date] [Author]

Re:RE: Soil Pressure

[Subject Prev][Subject Next][Thread Prev][Thread Next]
Sorry for the delayed response on this thread, but I thought you might still
want some more geotechnical input on this subject.  I saw at least 6 responses
regarding the fundamental arithmetic, but what about your general approach? 
This issue has evolved considerably since the initial Mononobe-Okabe studies of
cantilever retaining walls.  The Mononobe-Okabe equation is based on a simple
horizontally accelerated sliding wedge of backfill soils.  If you really want to
immerse yourself into this subject, see "The Seismic Design of Waterfront
Retaining Structures" November 1992 by the US Army CoE and the USN (ITL 92-11
and NCEL TR-939).  For me, some additional "empirical" enlightenment comes from
CalTech, in Alexander Ortiz's "Dynamic Centrifuge Testing of Cantilever
Retaining Walls, dated August 1982.  Others have correctly established the
basics in this thread, but here are my approach comments:

(1)  Most retaining wall dynamic performance case histories are for cantilever
retaining walls, not braced basement walls.  Mononobe-Okabe is the keystone
work, which was a study a failed cantilever retaining walls.

(2)  The concept of the INVERTED right triangle (upside down equivalent fluid
pressure) is based upon the horizontal acceleration of a wedge of soil, where
centroid of the mass of the triangular wedge is roughly two-thirds (0.5 to 0.67,
commonly 0.6) up from the bottom of the wall stem.  This provides greater
applied moment on the stem, but the same shear, compared to an equivalent fluid
pressure.  I would guess this is generally accepted in the California
geotechnical community.

(3)  On the other hand, Ortiz reported in 1982, with centrifuge tests at
CalTech., that "dynamic pressures were not triangular as the Mononobe-Okabe
theory assumes, although the centroid did remain at about 1/3 the height above
the base, contradicting other investigators which state that it rises between
1/2 and 2/3 of the height."  I would guess that most geotechnical engineers
still do not agree with this.  CoE/USN suggest at the top of page 66 in the
above referenced manual that the "incremental dynamic active earth pressure
force for dry backfill" should be applied at 0.6H, measured from the base of the
stem (roughly 2/3 up from the base of the stem).

(4) I'm not sure if all this theory for cantilever retaining walls translates
well to basement walls (Seed and Whitman, etc.).  Basement walls have less
"degrees of freedom" in 3-dimension.  First, in most cases the wall should be
designed for at-rest rather than active pressures. The wall is most likely
braced to limit deflection (and hence activation of the full active soil
strength).  Second, in most cases there is not a true ground surface
acceleration on the wall, since the basement extends below "the free field"
grade.  Just take a look at all the Northridge strong motion data for buildings
with basements.  In short, if you are using at-rest pressures for basement wall
design, it is probably conservative to use Mononobe-Okabe.  OSHPD/DSA, in many
cases, has only required that the seismic increment be added to earth pressures
on basement walls when the ground was not reasonably level around the building

(5)  Here is a little empirical test that I can relate since the Statute of
Limitation has long since expired.  A very wise and successful (or at least
prolific) Structural Engineer told me in 1982, just before he retired, that I
should never recommend more than an equivalent fluid pressure of 30 pcf for
design of a basement wall in well drained sands in Southern California.  This
unnamed Structural Engineer designed many telephone facilities in Southern
California (probably most of a particular unnamed company's facilities prior to
1982).  Has anyone seen or heard of a telephone facility with a basement wall
structural failure due to earthquake ground forces?  I would guess not, but I am
curious.  These old walls were certainly tested in 1994.  While you are guessing
who this Structural Engineer is, I remind you that he has not been practicing
since 1983 (or so).

In summary, I'd say, let's not loose sight of the case histories for basement
walls (not cantilever walls outside of a basement).  Most basement wall problems
that I have seen were due to no or poor drainage rather than earthquake loading.

Finally,  and probably most importantly, I strongly recommend that you have the
Geotechnical Engineer of Record review your plans and specifications, and wall
pressure calculations, so they can verify if you have properly implemented their

Tom Benson at Lowney Associates
1785 Locust Street, Suite 10
Pasadena, CA 91106-1614
(626) 396-1490, FAX: (626) 396-1491

____________________Reply Separator____________________
Subject:    RE: Soil Pressure 
Author: <seaint(--nospam--at)>
Date:       1/27/99 3:12 PM

OK, the math works using your assumption that the peak seismic soil pressure
would be 20H whether triangular or uniformly distributed - but according to
the soils reports I've seen the seismic pressures are much higher near the
ground surface, decreasing with depth, and it sounds as if this is what your
soils report is saying too, as I recall it.

BTW, if you're using UBC '94 and getting a seismic coefficient of .12 in
Zone 4, what are you using for Rw?  If I assume C defaults to 2.75 for a
stiff bldg, Rw backs out of the equation as 10 - can't think of any concrete
bldg system which goes this high. 

        -----Original Message-----
        From:        Alex C. Nacionales [SMTP:anacio(--nospam--at)]
        Sent:        Wednesday, January 27, 1999 1:55 PM
        To:        seaint(--nospam--at)
        Subject:        RE: Soil Pressure


        Thank you for your correction of the terminology.

        The geotechs may have a better explanation. What I have is just
        a simple analysis in the absence of more complex methods.

        The 20 pcf was taken original post regarding soil pressure.

        I computed that if soil weight is 150 pcf
        for a  sandy soil and if a use a g=.134( not .4 from my previous
        I get 150x.134 = 20pcf  as additional weight due seismic effects.

        Seismic coefficients I  get for a four story RC building
        in Zone 4 is about O.12

        P = 20 pcf x 8' = 160 psf
        for invert. triangle, solving Moment at wall base
        M= 1/2x160x8x2/3= 426.6 plf
        for rectangular,
        M= 160x8x1/2= 640 plf > 426.6 plf

        Alex C. Nacionales, C.E.
        Iloilo City, Philippines