Need a book? Engineering books recommendations...

# RE: Rigid Diaphragm supporting soil

• To: seaint(--nospam--at)seaint.org
• Subject: RE: Rigid Diaphragm supporting soil
• From: "Lutz,James" <JLUTZ(--nospam--at)earthtech.com>
• Date: Mon, 25 Oct 1999 17:16:04 -0700

```I am only aware of two methods for figuring the seismic load due to soil
loads. For cantilevered retaining walls that are free to deflect at the top
(active earth pressure), there's the Mononobe-Okabe method, which is
described in the AASHTO Standard Specifications for Bridges and numerous
foundation design texts. For unyielding walls, which is what it sounds like
you've got, the seismic forces are higher.

There are some formulas in Steven Kramer's "Geotechnical Earthquake
Engineering" (Prentice-Hall 1996) for unyielding walls which give you the
total seismic force as well as the position of the resultant. the resultant
is usually more than halfway up the wall, and the seismic soil pressure is
roughly what you would get from an equivalent fluid with a density equal to
the seismic acceleration (in g's) multiplied by the unit weight of the soil
times the depth. So if you had .3g, 120 pcf soil, and a 10 ft backfill
depth, the average seismic pressure would be .3x120x10=360 psf. The numbers
get big in a hurry for tall walls, and I've oversimplified things a bit.
Kramer's book is a good reference if you haven't seen it.

I think the above method is conservative, but there isn't much else I've run
across that you can readily hang your hat on. For non-yielding walls, I have
also seen suggestions that you use the M-O method but jack up the

Your geotech should be giving you some numbers for lateral earth pressure.

-----Original Message-----
From: Joseph M. Otto, P.E. [mailto:jmo_engineering(--nospam--at)email.msn.com]
Sent: Monday, October 25, 1999 4:53 PM
To: Seaint@Seaint. Org
Subject: Rigid Diaphragm supporting soil

I'm designing a Golf Course Clubhouse.  The site is sloped and will be cut
so that the upper level is at grade on two sides and the lower level is at
grade on two sides.  The upper roof is wood construction.  The second floor
level is steel with metal decking and a concrete floor.  Concrete and
masonry retaining walls hold back the earth where necessary.  The structure
is rectangular in shape except that it is broken at a 30 degree angle in the
middle.

It was initially proposed that I should include forces from the soil in the
lateral analysis.  The retaining wall was designed as spanning between the
rigid diaphragm at the second floor and the lower grade.

I am having difficulty deciding how to include the effects that the soil
will have in a seismic event in the lateral analysis.  One method mentioned
was to take the load applied by the soil at the second floor and divide it
by the seismic coefficient to come up with a virtual mass which could be
included in the vertical distribution of forces.  However, the soil loads
only act perpendicular to the wall in one direction and the earthquake
doesn't.

When the contribution of the soil, direct shear, and torsional shear are
included, the forces get very high.  High enough that the amplification
factor of UBC 1630.7 ===> 3.0 and the forces get even greater.

Another thought is to look at it a different way and design the walls a
cantilevers and to backfill them before the second floor diaphragm is
poured.  Then neglect the soil in the lateral analysis.

What approach do you think should be taken in the lateral analysis of this
structure.  Any input would be appreciated.