The model tests were carried out by using a shaking table at IIS, the University of Tokyo. In total three
types of small-scaled retaining wall models were constructed in a 40cm-wide by 180cm-long and
87.5cm-high rigid soil container fixed to the shaking table. Both front and back side of this container
comprised transparent-tempered glass windows to make the deformation behavior of model visible.
Figs. 2, 3 and 4 show the cross-sections of three different retaining wall models which are 39.5cm in
width. They consisted of one gravity-type RW and two types of reinforced-soil RWs with a FHR
facing having two different types of reinforcements (geocell model reinforcements and geogrid model
reinforcements). The total height of retaining wall models was 50 cm assuming a scale down factor of
1/10. The gravity-type RW was constructed by a FHR facing (39.5 cm-wide× 45 cm-high× 4.5
cm-thick) connected with a spread-footing foundation (39.5 cm-wide× 5 cm-high× 20 cm-thick). A
FHR facing for reinforced-soil RWs was prepared having a size of 39.5 cm in width, 50 cm in height
and 3 cm in thickness. All of these walls were made of duralumin. The back faces of gravity-type RW,
and the bottom surfaces of the foundation directly contact with the subsoil, and the back faces of FHR
facings for geogrid-reinforced RW and geocell-reinforced RW were made rough by being covered
with a sheet of sand paper (No. 150). In order to minimize the friction between the edge of RW models
and the side wall of the soil box, a sponge tape and grease were used, and the leakage of sand from the
gap between RW models and the soil container could completely be prevented.