A Critical Review of Full-Scale Shaking Table Tests Conducted on Reinforced Soil Retaining Walls - Reinforced Soil Engineering: Advances in Research and Practice - page 494

Ling et al. (2000) conducted finite-element analysis to simulate the
construction response of this wall. Nonlinear elastic behavior of soils and
reinforcement, as well as the blocks-backfill soil and block-block interactions,
was simulated. Ling et al. compared the wall deformation, vertical and lateral
stresses along the wall face, and strains in the geogrid layers. The procedure was
able to give satisfactory agreement between the measured and analyzed results. A
series of parametric studies was also conducted (Ling and Leshchinsky, 2001).
There were several interesting findings from the measured and analyzed
results of this wall.
1. The study confirmed high stress concentration at the connection
between the geosynthetic layers and modular blocks (Fig. 18) ,
including the secondary reinforcement layers. Note that in a separate
parametric study, the frictional connection was used instead of the
bolting connection.
2. The results also showed nonuniform vertical stress distribution at the
bottom of the backfill, and it is the greatest at the bottom of the blocks.
The nonuniform stress can be simulated in the analysis by considering
the foundation soil instead of treating concrete floor as rigid.
3. The lateral stress measured in the concrete facing blocks was less than
that in the soil, and the earth pressure coefficient was between that of
the active and at-rest conditions.
6 GENERAL CONCLUSIONS FROM LARGE-SCALE
SHAKING TABLE TESTS
An important conclusion that may be derived from the performance of reinforced
soil is that the amplitude of lateral displacement exceeding 50mm
(corresponding to an earthquake with horizontal acceleration of 400 gal) is
considered excessive depending on the types of application. If the bridge rests
directly on the reinforced soil structure, dislocation of the bridge span is
expected. In addition, an amplification ratio of 3 or greater will induce additional
stresses to the superstructure.
The comparison between the seismic performance and design is somehow
inconclusive because of the lack of details in evaluating the soil properties. For
example, the measured and design values of the angle of internal friction and the
value used for evaluation were slightly different.
The result related to full-scale modular block walls is lacking. The 6-m wall
constructed in Japan was not a truly modular block wall because of the different
connection mechanism between the wall and reinforcement. Besides, information