Seismic Lateral Earth Pressure

Seismic Pressures on Buried and Retaining Walls
Steven F. Bartlett, Ph.D., P.E. Dept. of Civil and Environmental Engineering Farhang Ostadan, Ph.D., P.E. Bechtel, San Francisco
Objectives
Earth pressures under seismic loading
Yielding Walls Non-Yielding Walls
Compaction Induced Stress (Matt Francis, URS) Reduction of Lateral Earth Pressures (Matt Francis, URS) Panel
Bill Gordon Joergen Pilz Matt Francis Steven Bartlett
Oct. 2005 October 2004
Near Fault Seismicity
Types of Walls
Gravity Cantilever Braced Tieback Soil Reinforced Rigid Walls (Basement Walls)
Oct. 2005 October 2004 4
Seismic Behavior of Walls

Yielding Walls
Can the wall move sufficiently enough to develop the active earth pressure state?
Modes of yielding Translation (sliding) Rotation Flexible (MSE) Non-Yielding Walls

Basement walls are non-yielding walls
Do not develop the active earth pressure state
Mononobe-Okabe (M-O) Method

Effect of ground motion on retaining walls was recognized by Okabe (1924) and Mononobe (1929) following the great Kanto earthquake of 1923.
The M-O method Based on Coulombs theory of static soil pressure Pseudostatic method Valid for yielding walls
Mononobe-Okabe (M-O) Method

F. E. analyses by Pitilakis (1987) found good agreement between F. E. analyses results and Mononobe-Okabe theory and observed behavior, and that the M-O theory is satisfactory for design purposes if the wall movement was about 0.5 percent of the height of the wall.
0.05 x 20 ft = 0.1 ft deflection required for M-O method to be valid
M-O Lateral Earth Pressure

i ah*W av * W W PAE F
Fh = ah * W Fv = av * W
M-O Lateral Earth Pressure
PAE = PA + PAE
Static Earth Pressure + Earthquake Pressure
2 PAE = 1/2 H2 (1-av ) KAE AE v AE
Height of Resultant Force (h)
h = (PA H / 3 + PAE (0.6) H) / PAE
M-O Seismic Earth Pressure Coefficient
where: =tan-1 (a
10
M-O Summary
In spite of much better understanding of soilstructure interaction effects and many criticisms of the M-O method, the M-O method is widely used for building walls. In this regard, the M-O method is one the most abused methods in the geotechnical practice.
11
Seismic Lateral Earth Pressures

(Wood, 1973)
Wood developed a finite element solution for a non-yielding wall. While the solution is based on dynamic modal analysis (hence dynamic), the solution used in practice is a static solution based on horizontal body force of 1g acceleration. Need to know PGA, Poissons ratio and density of the soil to get the pressure.
12

(Wood, 1973)
13

(Wood, 1973)
Peq = H2(ah/g)Fp
14

(Wood, 1973)
15

Recent Experiments and Observations
In recent years recorded data from many underground structures have been examined. Lotung 1/4-scale model, in Lotung, Taiwan (1987).
The studies show that seismic soil pressure is:
affected by the long period part of the ground motion function of relative motion between the soil and the structure (soil-structure interaction) amplified near the resonant frequency of the backfill a function of soil nonlinearity
Many recent observations from Japan confirmed the above findings


Recent Experiments and Observations
Lotung 1/4-scale model, in Lotung, Taiwan (1987).
17

(Ostadan and White, 1997) Assumptions and Method
Assume the building basemat is founded on rock. Input ground motion at basemat elevation. The walls of the building are effectively rigid. 30 foot-embedment considered 5 percent material damping of soil Poissons ratio of soil = 1/3 Kinematic SSI is considered. Inertial SSI is not considered. The solution is derived from SSI analyses using SASSI.
18

(Ostadan and White, 1997)
19

Transfer Functions
40000 35000
Amplitude of Pressure TF
30000 25000 20000 15000 10000 5000 0 0 5 10
Vs = 500 ft/s Vs = 1000 ft/s Vs = 1500 ft/s Vs = 2000 ft/s
15
20
Frequency (Hz)

Careful examination of dynamic soil pressure response from a SSI solution shows that the pressure response is very similar to the response of a single degree-of-freedom (SDOF) system. Also, the natural frequency of the results can be normalized by the soil column frequency fsoil column = Vs / (4 x H)
21

4.5 4
Normalized Transfer Functions
Normalized Amplitude
3.5 3 2.5 2 1.5 1 0.5 0 0 0.25
Vs = 500 ft/s Vs = 1000 ft/s Vs = 1500 ft/s Vs = 2000 ft/s
A/A @ low frequency
0.5
0.75
1.25
1.5
Normalized Frequency
(F/F soil)
22

Normalized Pressure Profile (normalized to max. pressure at surface)
23

Observations from SSI analyses Response of system was controlled by stiffness at long period, damping at resonance and inertia at short period, similar to a SDOF system. The pressure amplitude and its distribution was obtained from the low frequency solution of the response.
24

To estimate the damping and the total force, the soil-wall system was subjected to 6 different input motions.
1.4 1.2 EUS - LOCAL EUS DISTANT ATC S1 WUS RG1.60 LOMA PRIETA
Spectral Acceletation (g)
0.8 0.6 0.4 0.2 0 0.1 1 10 100
Frequency (Hz)
25

The resulting force for each event was computed from the SSI analysis and the associated mass and damping were back calculated. m = 0.50 H2
(1 )(2 )
The maximum amplification of the pressure response is controlled by the radiation damping due to continuity of the soil behind the wall. The damping associated with the response is about 30% due to high level of radiation damping.

27

COMPUTATIONAL STEPS 1. Perform seismic ground response analysis (using SHAKE) and obtain the acceleration response spectrum at the basemat level in the free-field at 30% damping. 2. Obtain the total mass using: m = 0.50 H2
28

3. Obtain the seismic lateral force by multiplying the mass from Step 2 by the spectral amplitude of the free-field response (Step 1) at the soil column frequency. F = m Sa free field for fsoil
29

(Ostadan and White, 1997) 4. Calculate the max. lateral earth pressure (ground surface) by dividing the results for step 3 by the area under the normal soil pressure curve (area = 0.744 H) 5. Calculate the lateral pressure distribution verses depth by multiply the max. lateral earth pressure by the p(y) function below. p(y) = - .0015 + 5.05y - 15.84y2 + 28.25y3 - 24.59y4 + 8.14y5 where y is the normalized height (Y/H) measured from the base of the wall.
30

(Ostadan and White, 1997) The method was verified by comparing the results of the simple computational steps with the direct solution from SASSI. The verification included 4 different wall heights, 6 different input time histories and 4 different soil properties. The method is very simple and only involves free-field (e.g. SHAKE) analysis and a number of hand computational steps. The method has been adopted by building code (NEHRP 2000) and will be included in the next version of ASCE 4-98. The Ostadan-White method is by no means a complete solution to the seismic soil pressure problem. It is merely a step forward at this time.
31

(Mitigation Strategies)
32

Seismic Lateral Earth Pressure

Uploaded by

Copyright:

Available Formats

Seismic Lateral Earth Pressure

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Seismic Lateral Earth Pressure

Uploaded by

Copyright:

Available Formats

Seismic Pressures on Buried and Retaining Walls

Oct. 2005 October 2004

Near Fault Seismicity

Oct. 2005 October 2004

Seismic Behavior of Walls

Modes of yielding Translation (sliding) Rotation Flexible (MSE) Non-Yielding Walls

Mononobe-Okabe (M-O) Method

Oct. 2005 October 2004

Mononobe-Okabe (M-O) Method

0.05 x 20 ft = 0.1 ft deflection required for M-O method to be valid

Oct. 2005 October 2004

M-O Lateral Earth Pressure

M-O Lateral Earth Pressure

Height of Resultant Force (h)

h = (PA H / 3 + PAE (0.6) H) / PAE

Oct. 2005 October 2004

M-O Seismic Earth Pressure Coefficient

Oct. 2005 October 2004

Oct. 2005 October 2004