July/August 2024 - 115

[254 mm] at the end of construction) of
28 in (711 mm). It means that, while at
the end of construction the backfill
compressed 10 in (254 mm), about 14
months later it further compressed
nearly 18 in (457 mm). No substantial
bulging of the face was observed that
could explain the large vertical
settlement at the crest. The material
undergoing this settlement is due to
compression of the backfill.
The drainage system consisted of
four vertical risers to collect surface
water. Each riser was connected to the
other by a nearly horizontal 8 in (203 mm)
drainage pipe. The risers were relocated
by the contractor into the reinforced
soil area (from the original plan, placing
it in the retained soil zone) since it was
" easier " to construct. That is, placement
in the original locations would have
required excavation in rock. The
depression occurring around the risers
clearly indicates that the soil there was
poorly compacted. Compaction near
No substantial bulging of
the face was observed that
could explain the large
vertical settlement at the
crest.
" obstacles " like risers requires extra
attention and careful monitoring.
While formally there was a supervisor
to assure proper construction, no
record of field compaction tests was
found. Also, no report of actual quality of
backfill was found. Construction was
likely unsupervised.
And finally, borings indicate that the
fill material was truly random - more
than 50% passing #200 and boulders 12
in (305 mm) large, including some
occasional organic material.
From these observations, the following
conclusions could be made.
Generally, the backfill was poorly
compacted. Slope penetration testing
implies it was less than 90% Standard
Proctor (95% was specified). Poor
compaction around the risers resulted in
differential settlement around the risers
thus shearing the 8 in (203 mm)
horizontal pipes connected to the risers.
Video inspection confirmed this
conclusion. As a result, the surface
drainage system worked in reverse. That
is, the surface water it collected was fed
into the loose reinforced soil. The end
result was in-service compaction, an
unacceptable situation.
Piezometers indicated porewater
pressure heads of about 10 ft (3 m)
above the base of the wall. Global
stability using Bishop stability analysis
for this situation indicated a factor of
safety of about Fs=1.1. Although slope
stability analysis does not deal with
displacements, the author's experience
indicates that when Fs is between
1.1-1.2, ground movements are visible
(e.g., some outward facing movement).
Clearly, this is an unacceptable Fs value
as it indicates possible progression to
collapse. A wall in which the drainage
system works in reverse is likely to
undergo catastrophic failure with time.
It is interesting to note that in
preparation for a possible jury trial, the
attorney posed a question as to the
s igni fi cance of compact ion. To
demonstrate the impact of compaction,
the author, with the assistance of an
expert, had characterized the constitutive
behavior of typical backfill soil
used at different compaction levels,
using standard triaxial lab equipment.
This data was then input into a
numerical continuum mechanics
scheme (Finite Difference, program
FLAC), simulating the construction of
this wall. It was observed that a
threshold exists and when compaction
is less than 85-90%, compression
(settlement) exponentially increases.
Density of 95% of Standard Proctor
would have had much smaller compression.
Such demonstration is instructive
as it supports what most geotechnical
engineers know intuitively.
Lessons Learned - Georgia
This case study demonstrates that compaction
is essential to the performance
of MSE walls. This requirement is not
different from general requirements for
all geotechnical structures.
Without water infiltration the structure
would be stable. However, the
settlements would be large enough to
render a problem of serviceability. The
addition of water increases the probability
of collapse; it turns a serviceability
issue into a serious danger of collapse.
If the designed elevation of the crest
is not achieved after compacting the top
soil layer, the reason could be either
excessive settlement of the foundation
or compression of the backfill due to
poor compaction. The exact reason
should be identified before construction
of the superstructure continues.
Adding excessive backfill to
reach the target elevation should be
carefully considered as it could mask a
potential flaw due to poor compaction.
In the long run continued compression
may result in a costly failure.
As much as possible, a drainage
system should not be placed within the
reinforced soil. Otherwise, if it fails, the
wall is likely to fail as well. Reinforced
backfill is not designed to serve as a
drainage layer aimed at transporting
water. Furthermore, there is no simple
repair of a drainage system embedded
within the reinforced soil. Hence,
redundancy in drainage, ensuring that
water does not flow through the
reinforced fill, is good.
Reinforced backfill is not
designed to serve as a
drainage layer aimed at
transporting water.
DEEP FOUNDATIONS * JULY/AUG 2024 * 115
July/August 2024

Table of Contents for the Digital Edition of July/August 2024
