For geogrid-RS RW model (Fig. 3), geogrid reinforcements used in actual field cases were simulated
by a set of regular grids comprising longitudinal members (made of thin and narrow phosphor-bronze
strips, 0.2mm-thick and 3mm-wide) welded at nodes to transversal members (made of mild steel bar,
0.5mm in diameter) at intervals of 35 mm. To effectively mobilize friction between the reinforcement
and the backfill, sand particles were glued on the surface of the strips. Strain gauges were attached to
the reinforcements to measure the tensile force. The geogrid model reinforcement was 360 mm
(length) × 350 mm (width).
For geocell-RS RW model (Fig. 4), the square-shaped geocell model reinforcement was also set to
360 mm (length) × 350 mm (width), having six square cells in the longitudinal direction and seven
square cells in the transverse direction. The height of transverse member is 25mm and the size of each
cell is 60mm×50mm. It is made from polyester (PET) covered with PVC for protection which is a
relative weak material can be used in scaled-down model test (Kongkitkul et al., 2007). Ten layers of
reinforcements were horizontally placed at a vertical spacing of 5 cm in the backfill which was set by
reducing them to a scale of one-tenth that of actual reinforced soil retaining walls with a full-height
rigid facing constructed in Japan (R.T.R.I., 2000). Note that the geometric shape and arrangement of
reinforcements were determined by referring to those actual one, while the similitude on the properties
of geogrid materials was not considered for the purpose of measuring tensile force. The reinforcement
connections with the facing panel were designed to be perfectly rigid to prevent slippage of the
reinforcement layers at the facing (Figs. 3b and 4b).