Conference Technology CONTEC-2001,

Kota Kinabalu, Sabah, 11-13 October 2001

Fill Compaction and Its Consequences of Non-Compliance

S. S. Gue & S. S. Liew
Gue & Partners Sdn Bhd, Malaysia

ABSTRACT: Many failures including building failures and road pavement were the results of noncompliance to the compaction requirements of fill for its engineering purposes. The remedies are often very
costly and sometimes catastrophic. Many of these failures have been subjected to arbitration and legal suits.
The paper presents the concept of compaction and its requirements for various engineering purposes. It also
presents some case histories on the failures related to non-compliance.
1

INTRODUCTION

Quality control testing is an important aspect for

checking compliance at site. The adopted testing
procedures have to be practical and quick for site
implementation.

Soil and rock represent the most abundant and easily

available construction material on the earth. These
natural materials exist as the supporting medium
virtually for most of the civil engineering structures.
Due to rapid development, new developments on
natural undulating land away from coastal areas and
hillsites are being levelled as platform for buildings,
roads and etc. As a result, cut-and-fill earthworks
are necessary to form the platform for the structures.
When filling is involved, it becomes necessary to
consider its implications on the structures, in
particular when the fill is not properly compacted.

The consequences of non-compliance

earthwork
compaction
can
sometimes
catastrophic.

of
be

This paper discusses the abovementioned issues

and presents some case histories on failure due to
non-compliance.

Basic fundamentals of compaction are sometimes

not clearly understood by some practitioners, so as
the performance of the end products.

FUNDAMENTALS AND THEORY OF

COMPACTION

Compaction is a process to increase the density of

soil mass by mechanical means, which usually
involves rolling, vibrating, tamping or combination
of these processes. Such process is to rearrange the
soil particles with expulsion of air in the voids of soil
mass without major changes in pore water, except
for granular soil, in which water is possible be
pressed out. However, water content plays a vital
role in compacting process to achieve certain
compactness of the soil mass. Higher compactness
of the soil skeleton implies higher soil strength,
lower compressibility, lower permeability and less
susceptibility of other engineering properties to
water content alteration. These are the important
engineering features of the compacted soil mass.

Very often, there are different opinions on the

achieved engineering properties for fill materials
with respect to the specifications adopted in the
contract. In many cases, standard specification for
road works is commonly used for earthworks with
other intended functions, hence leading to
unnecessary
disputes
during
the
work
implementation and the interpretation of test results.
It is crucial to capture the important key features and
controls in the specification to ensure compliance for
the intended functions. For the contractual purpose,
there are generally three types of earthwork
specifications, namely method specification, end
product specification and performance specification,
commonly used.

Conference Technology CONTEC-2001,

Kota Kinabalu, Sabah, 11-13 October 2001

It is important to distinguish compaction from

consolidation, which is a geotechnical process of
expelling pore water from the voids as a result of
volume reduction under external static loading or the
soil self-weight.
For unsaturated or partially
saturated soil, air will be expelled in the
consolidation process.

higher compaction energy of same nature will shift

the compaction curve to the left and up from its
original location. Same energy input but different
nature of energy, such as vibratory and static rolling,
will also result to different compaction curve. Soil
type will also affect the compaction curve. Soils
with low plasticity and coarse grains will achieve
lower dry density for a given compaction effort as
compared to soils with high plasticity and fine
grains.

Using simple relationship between the specific

gravity of soil particle, void ratio and moisture
content, a series of envelope curves with different air
void ratios can be computed and plotted on graph as
shown in Figure 1 for a given compaction effort. It
is not difficult to understand that lower moisture
content in the soil skeleton will result in higher shear
resistance between the soil grains during
compaction. This is due to high contact stress in the
soil grains as a result of high surface tension
generated from the meniscus of water between the
soil grains.
When moisture content gradually
increases, the soil skeleton structures will tend to
collapse and rearrange easily to a more compact state
under compaction as the effect of surface tension
reduces with increasing water content. When the
dry density of the compacted soil mass reaches a
peak, the corresponding moisture content is called
optimum moisture content (OMC). Dry density is a
good indicator of compactness of soil mass instead
of bulk density, which will be altered with the water
content. If excessive water is present in the voids of
the soil skeleton, the large volume of the
incompressible water will act as a retardant, which
absorbs and dissipates the compaction energy and
therefore results in lower compactness. This simple
fundamental can explain why the dry densitymoisture curves in most soil types have a peak at the
OMC and are curved down at both sides of the peak.

2.8

d
id
oi id
Vo
o
rV
Ai ir V Air
A %
ro
Ze 5% 10

Dry Density, d (Mg/m3)

2.4

d = w Gs(1 - A)/(1 + w Gs)

4.5kg Rammer

2.5kg Rammer
1.6

1.2
0

10
20
Moisture Content, w (%)

30

Figure 1 Dry Density-Moisture Content

Relationship with Compaction Curves

Theoretically, the zero air voids line in Figure 1

means a fully saturated soil mass and no part of the
compaction curve can be beyond the zero air void
line. However, it is too costly and time consuming
to expel all air in the voids of soil mass to achieve
fully saturated condition by mechanical compaction
practically. Generally, there is also no engineering
requirement to do so. Nevertheless, it is possible
and practically to achieve the compaction state with
acceptable air void by compacting the soil on the wet
side of OMC, in which the wet side of the
compaction curve is more or less parallel and also
closer to the zero air void line.

LABORATORY AND FIELD TESTING

In order to quantify the degree of compaction,

laboratory test has been established as a reference to
compare with the field achievement. However, the
procedures of such tests were arbitrarily established
during its invention but have now been standardized
in the earthworks specification worldwide. The
famous laboratory compaction test is Proctor test
devised by Proctor (1933), which used mechanical
drop rammer to compact the soil in sequence of thin
layers within 1 litre cylindrical steel mould. In
general, the test results presented in the plot show the
relationship between the achieved dry density of
compacted soil over the range of moisture contents
for a certain compaction effort. Each compaction

It should be understood that all compaction curves

of a specific soil type are with respect to the amount
and nature of compaction energy. For instance,

Conference Technology CONTEC-2001,

Kota Kinabalu, Sabah, 11-13 October 2001

data point in the plot represents one compaction test

at specific moisture content. The compaction curve is
the plot of these points and indicates the maximum
dry density (MDD) with the corresponding OMC.
It should be noted that the compaction curve
established from the compaction test may not be
necessarily applicable in free draining granular soil,
such as sand, in which water can readily be expelled
during compaction process and the typical
compaction curve would be different. There are
two types of compaction tests, namely the Standard
Compaction Test and the Modified Compaction
Test, which are both defined in Part 4 of BS 1377 :
1990. The major differences between these two
tests using the same compaction mould are tabulated
in Table 1:

After the MDD and OMC of a specific soil type

have been established based on the required
laboratory compaction tests, the design engineer will
then specify the requirements for field compaction.
This will further be discussed in the later section on
specifications.
To check the quality and degree of compaction in
field, numbers of field compaction tests have been
devised with the features of simple and fast operation
for reliable and repeatable results. It is important to
determine the field dry density (FDD) and also the
corresponding moisture content of the compacted
soils to compare the computed dry density ratio or
the air void ratio with the tolerance limits in the
specification for checking of compliance. The most
common field dry density tests are sand replacement
method and nuclear density gauge of non-destructive
nature. In the determination of compacted soil
moisture content, the conventional oven dry method
is preferred with supplementary indirect field tests
using microwave oven or speedy tester. However,
proper correlation is required for meaningful
application.
Plate 1 shows the speedy tester
commonly used in earthworks operation.

Table 1 Comparison Between Standard Proctor

Test and Modified Proctor Test
Standard Compaction Modified Compaction
Test
Test
Using 4.5kg rammer
Using 2.5kg rammer
(2682kJ/m3)
(596kJ/m3)
Rammer Drop: 300mm Rammer Drop: 450mm
Compact in 3 thicker
Compact in 5 thinner
layers
layers
BS1377
BS1377
ASTM D698
ASTM D1557
The need to develop the modified compaction test
is mainly due to the availability of modern
compactor with higher compaction energy rating and
also the technical reason for higher MDD. The
compaction energy of modified compaction test is
about 4.5 times of that in the standard compaction
test.
Both tests can be compacted in either one litre
steel mould or CBR mould depending on the
application of the test results. If CBR test is needed
on compacted sample, then CBR mould should be
used for compaction and later for CBR test.

Plate 1 Speedy Tester for Quick Determination of

Moisture Content at Site

During sample preparation, special attention

should be given to soils containing particles that are
susceptible to crushing. Therefore, soils of such
nature should not be reused after compaction tests.
BS 1377 : Part 4 provides very elaborative details on
this aspect.

STATUTORY REQUIREMENT

Non-compliance of earthworks often relates to

insufficient quality assurance and control by the
contractor and inadequate site supervision by the
supervising consultant.

For free draining granular soil passing 37.5mm

test sieve, dry shaking test and dropping test are
more appropriate to establish density index for sands
and gravelly soil respectively.

Section 71 of the Street, Drainage and Building

Act, 1974 does contain a statement in relation to the
earthwork compaction. The relevant portion of the

Conference Technology CONTEC-2001,

Kota Kinabalu, Sabah, 11-13 October 2001

statement in the Bulletin Ingenieur (BEM, 2000) is

as follows;

fill, such as maximum settlement over

specific period of time after filling.

Where any building or part of a building fails,

whether in the course of construction or after
completion, or where there is any failure in relation
to any earthworks or part of any earthworks, whether
in the course or after completion thereof and the
cause of such failure is due to any one or more of the
following factors: (a) misconstruction or lack of
proper supervision during construction; (b)
misdesign or miscalculation; or (c) misuse of such
building or part of such building, or of such
earthworks or part of such earthworks, the person
responsible for (a) such misconstruction or such lack
of proper supervision; (b) such misdesign or
miscalculation; (c) such misuse, shall be liable on
conviction to a fine not exceeding five hundred
thousand dollars or to imprisonment for a term not
exceeding ten years or to both (latest update revises
the fine from RM50,000 to RM500,000, BEM,
2001).

In Malaysia, the most common used earthwork

compaction
specification
is
End
Product
Specifications. Sometimes, combinations of the
abovementioned specifications have been adopted,
but usually lead to difficulties in implementation and
disputes with contractor.
Dry Density
The dry density ratio, which is the ratio of achieved
field density to maximum dry density of either
standard or modified Proctor compaction test, is the
most common criteria being used in many earthwork
specifications. However, the rationale of specifying
certain percentage, such as 95% of MDD, may not
always be there. Such commonly used percentage
is more of the rule-of-thumb basis rather than
specially tailored figure for a specific application.
There are some guidelines for minimum standard of
compaction proposed in AS 1289 as shown in Table
2.

Building is defined in the Act including any

house, hut, shed or roofed enclosure, whether used
for the purpose of a human habitation or otherwise,
and also any wall, fence. Platform, staging, gate,
post, pillar, piling, frame, hoarding, slip, dock,
wharf, pier, jetty, landing stage or bridge, or any
structure support or foundation connected to the
foregoing.

Table 2 Guideline for Minimum Standard of

Compaction (AS 1289)
Minimum Minimum
Type of Fill
Dry
Density
Density
Index
Ratio *
Residential Fill
95%
65%
Commercial Fill to support
floor loading up to 20kPa
98%
70%
and isolated pad footing to
100kPa
Road Embankment
95%
65%
(a) 0.3m below pavement
subgrade
100%
80%
(b) < 0.3m below pavement
subgrade
* Note : All dry density ratio relate to Standard
Compaction, in accordance with AS 1289 E1.1

SPECIFICATIONS

There are generally three basic types of

specifications for compaction of the earthwork ,
namely;
a.

b.

c.

Method
Specifications:
Specify
the
procedure for the placement and compaction
of the fill, such as type and mass of
equipment, the compacting thickness or
number of passes and scarifying between
compacted layers of fine-grained soils to
eliminate lamination.
End Product Specifications: Specify certain
properties of the fill as placed and
compacted, such as the achieved dry density
ratio, placement moisture content, air voids,
CBR value or undrained shear strength.
Performance Specifications: Specify certain
aspects of the behaviour of the completed

Where landscape or other non-load bearing areas

are present in the development, it is not necessary to
achieve high level of compaction. Therefore, a
more lenient compaction specification should be
specified to reduce construction cost and time.
Nevertheless, this demands a proper site planning
during the earlier stage of design development. The
deposition of fill of various material types and the
corresponding compaction requirement should be
clearly specified based on the site zoning for its
intended functions.

Conference Technology CONTEC-2001,

Kota Kinabalu, Sabah, 11-13 October 2001

specification on building platforms. It is also a bad

practice to always base the compaction requirement
on the modified Proctor standard just because it is
newer or conservative.
This will lead to
unnecessary extra time and cost for the project. In
certain circumstances, over-compaction will not
result in satisfactory end product. Whyte, I. L. &
Vakalis, I. G. (1987) report that slickensided slip
plane in clay fill as a result of over-compaction by
pneumatic tyred roller at moisture content wet of
OMC at Trenton and Tiber Dams. Sometimes overcompaction will also cause breakage of soil or rock
particles, in which the particle grading will be
altered. This can have serious implications on
altering of compaction characteristics and also
changing the filter grading if the compacted material
is designed as filter. Plate 4 shows the breakage of
rock fragment under heavy compaction. Modified
Proctor compaction test should only be specified
when high dry density and lower compaction
moisture content are desirable.

It is also advisable to specify light compaction on

the last layer of compaction lift at the end of the day
using smooth-wheeled roller to form a thin and less
permeable layer with smooth sloping surface to
encourage surface run-off of rainfall. This can
reduce erosion by the runoff and protect the lower
compacted fill. Scarification of this top smooth
lightly compacted layer is required to remove the
lamination, particularly for fine-grained soils, before
placing the next lift as the lamination provides poor
bonding between the two compacting lifts. Plate 2
shows the smooth compacted surface of a finegrained soil. Plate 3 shows the scarifier.

Plate 2 Smooth Surface of Compacted Soils

Plate 4 Breakage of Rock Fragment under Heavy

Compaction
Moisture Content
Limits of moisture content during fill placement are
also a vital aspect to be specified. If the fill is too
wet, there will be risk of excessive consolidation
settlement and low strength or, on the other hand,
compaction on dry extreme side of OMC, the fill will
then be vulnerable to collapse settlement.
In
general, compaction on wet side will tend to produce
fill with plastic characteristics and better
deformability whereas brittle fill will be produced
with compaction on the dry side, which is
unfavourable for earth dam. Conditioning of fill to
comply with the specified limits of moisture content
around the OMC is sometimes necessary to facilitate
compaction and achieve the required dry density.

Plate 3 Scarifier Mounted to Back-Pusher

For controlling the potential of collapse
settlement, it is suggested to limit the air void
content not exceeding 5% in addition to the specified
dry density ratio (Charles et al, 1998).
It is important to understand the actual
requirement before preparing the specifications.
Specifications developed for highway embankment
may not be appropriate for fills on which buildings
will be founded. Building structures are much more
vulnerable to ground movement than road and
therefore calls for more stringent earthworks

Conference Technology CONTEC-2001,

Kota Kinabalu, Sabah, 11-13 October 2001

Loose Lift Thickness

If the loose lifts are too thick, it will result in a
dense upper crust overlying less compacted soil.
This is true regardless of the type of the used or soil
type. The loose lift thickness should be determined
such as to allow compaction energy to spread
throughout the thickness to promote the specified fill
density and low air voids content. Figure 2 shows
the difference between the thin and thick loose lift
thickness. The lift thickness is also governed by the
maximum soil particle size in terms of
compactability. If the soil particle size is larger
than or closer to the loose lift thickness, the chances
of having poorly compacted soil mass are very high,
particularly those soils around the large soil
particles, such as large rock fragments. Plate 5
shows the poor compaction around a large rock
fragment for a rockfill compaction. It is generally
accepted that the largest particle size of the
compacting soil should not be exceeding two-third of
the loose lift thickness for proper compaction.

Plate 5 Poor Compaction around Large Rock

Testing
Five points compaction curve is generally
recommended. One point compaction test should
never be allowed due to high degree of uncertainty
and variability of compacted materials. For each
source of material supplied for fill compaction,
separate compaction test should be carried out to
establish the reference for field achievement.
The frequency of field quality control testing
should be clearly specified in order to have
acceptable representative statistics. In Australia
Standard AS 3798 : 1990, guidelines and testing
frequency are specified as in Table 3. However, if
the material types in a compaction are more than one
type, then more field density tests are required.
Table 3 Guidelines to Frequency of Field Density
Test (AS 3798 : 1990)
Scope of Earthworks
Frequency of Field
Density Tests
Large Scale Operation
One test per 500m3
(Industrial lots, Road embankment)

Figure 2 Control of Loose Lift Thickness

It will be beneficial to all parties that a site trial
compaction should be conducted to determine
compacting thickness, number of passing and the
range of moisture content for a soil type in achieving
the specified compaction requirements.

Small Scale Operation

(Residential lots)
Concentrated Operation
(Gullies filling, Farm
dam)
Confined Operation
(Filling behind structures)
Trenches

One test per 200m3

One test per 100m3
One test per 2 layers
per 50m2
One test per 2 layers
per 40m

Trenter & Charles (1996) have presented an useful

model specification of engineered fill for building
purpose.

Conference Technology CONTEC-2001,

Kota Kinabalu, Sabah, 11-13 October 2001
6

ASSOCIATED ENGINEERNIG PROBLEMS IN

FILL COMPACTION AND CONSEQUENCES
OF NON-COMPLIANCE

The engineering problems associated with fill

compaction are as follows:
a.

b.
c.
d.

Total/Differential Settlements due to fill selfcompression result in downdrag on

foundation piles and alteration of drainage
gradient.
Collapse Settlement upon wetting of
compacted fill.
Weak Strength
High Permeability of Fill

Void
Plate 6

Crack

Some useful details for monitoring negative skin

friction on circular spun pile are given in a technical
paper by Gue et al (1999). Charles & Watts (1996)
and Charles et al (1998) have presented the potential
of collapse settlement in fill material and also the
adequacy and appropriateness of fixing the 95% of
standard compaction effort for engineered fill in
relation to collapse settlement.

Plate 7

Cracks on Ground Beam

Excessive platform settlement occurred a few

months after completion of the earth filling and
causing serious structural cracks on the newly
constructed ground beams, stumps, ground floor
slabs and the structural frames. More serious
distresses were observed at units on thicker fill.
Crack gauges and settlement markers were installed
during the investigation by the authors to monitor the
changes of crack width and platform settlement with
time respectively.
The results of monitoring
confirm the settlement and cracks were active. The
observed crack pattern on the wall was mostly
diagonal and initiated from the corners of the door
and window openings that were the areas of stress
concentration when distortion of the structures
occurred. Cracks on the structural frames were
mainly located at the beam-column connection and
the mid span of the beams, which have maximum
flexural stresses.

In the earthworks design, the needs of specifying

compaction requirements are to answer the above
highlighted engineering problems.
The consequences of non-compliance will be
demonstrated in the following cases histories.
6.1

Voids underneath Ground Beam

Case One : Excessive Foundation Settlement &

Short Piles

This case involved some 40 units of single-storey

terrace houses and 70 units of double-storey terrace
houses, in which platform settlement and structural
cracking were observed. Plates 6 and 7 show the
voids underneath the ground beam and the cracks at
the ground beam respectively after fill settlement.
The affected buildings were on filled ground of
varying depths up to 20m overlying the undulating
meta-sedimentary formation.
The platform was
mostly filled using the cut materials with significant
content of boulders, which subsequently posed
problems to pile installation. The existence of
boulders in the fill has been confirmed during the
additional subsurface investigation, in which a trial
pit done indicated that a boulder of about 1m in size
was discovered. The subsurface investigation also
revealed that the platform was not properly
compacted to the normal standard of engineered fill.

6.2

Case Two : Platform Settlement and Wall

Collapse

This case history was featured with many

geotechnical failures in the infrastructure works and
the buildings. The project site was located at a
previous rubber estate with undulating terrain of
meta-sedimentary formation and a few numbers of
natural water streams within the project site. As the
development area was very large in lateral extent, the
earthwork design concept was based on cut and fill

Conference Technology CONTEC-2001,

Kota Kinabalu, Sabah, 11-13 October 2001

balance and to minimize rock excavation as bedrock

was commonly expected at shallow depth below the
undulating original terrain and has a fairly consistent
weathering profile.
Therefore, the earthwork
platform was finally completed with varying fill
thickness and earth cutting. The maximum fill
thickness in this area was about 32m.
Unfortunately the filling areas were not properly
compacted as engineered fill. Many problems were
subsequently surfaced up as a result of fill
settlement, which was largely attributed to poor
compaction and collapse compression after
saturation with prolonged rainfall and rise of
groundwater level.

Gap (75mm)
Void (150mm)
Plate 8 Sagging Ground Floor Slab

Void

There were about 200 units of double-storey

terrace houses in one of the development phases
suffering from significant fill platform settlement.
These buildings were supported on 125mm125mm
timber piles. The non-suspended ground floor slab
became suspended after the platform settled and
caused the ground floor slab to sag. The partial
friction between the four sides of the floor slab and
the ground beams restrained the slab from settling.
Figure 3 shows the schematic diagram of the sagging
ground floor slab and Plate 8 shows the settling floor
slab. Voids underneath the ground floor slab was
suspected during final finishing of the floor tiles and
was further confirmed after hacking the suspected
slabs for inspection as shown in Plates 8 and 9. The
maximum gap between the center of the sagging slab
and the settling fill platform as measured at site was
about 150mm. Therefore, the total settlement of the
platform at the time of investigation was estimated to
be about 225mm, assuming the buildings did not
settle. The pattern of platform settlement coincided
very well with the original topography, in which
larger platform settlement occurs at the thicker fill
areas.

Figure 3

Plate 9 Void underneath the Ground Floor Slab

Another concern of the associated platform
settlement was the potential negative skin friction
developed at the piled foundation, although at the
time of investigation there were no serious distresses
or settlement observed on the structural frame except
some vertical tension cracks between the units on
split levels platform. The review of the topography
survey of the affected units and the piling records
reveals that some of the installed foundation piles
terminated within the fill and some penetrated few
meters into the original ground. It appeared that the
piles were installed to predetermined length. The
telltale sign of the vertical tension cracks on the
external walls of buildings on different platform
levels had indicated potential future excessive
differential settlement at areas where piles were
installed within fill of varying thickness. For those
piles penetrated into original ground, predominantly
end bearing condition could be expected, in which
negative skin friction due to platform settlement
could overstress the foundation piles beyond the
allowable design stress level of the piles but this has
yet to reach structural failure at ultimate limit
condition or showed any excessive settlement.
Figure 4 shows some pile penetration records in both
fill or cut ground and reveals that the foundation
piles were likely to be installed to length instead of
following the normal piling practice based on static
calculation from the soil properties, profile and
normal driving control using appropriate set
criterion.

Sagging Ground Floor Slab

Conference Technology CONTEC-2001,

Kota Kinabalu, Sabah, 11-13 October 2001
80

70
Reduced Level (m)

houses. However, there were no obvious distresses

yet on the building.

Building
Platform
Original
Ground Profile

60
Pile B

Filled ground
Original Ground
Original Pile's Toe
Additional Pile's Toe
Piles Adopted for Down
Drag Analyses

Pile A

50

40
0

10000

20000

30000
40000
Coordinate-X (mm)

50000

60000

Rubble Wall Base

70000

Collapsed
RC Drain

Figure 4 Pile Penetration Records

A collapse of retaining structure was also
observed at the site. The retaining structure was
about 3.5 m high pile supported rubble wall to retain
earth at the backyard of few units of apartment
structures on piles. The rubble wall was supported
on two rows of piles from 15m to 24m long at pile
spacing of 3.7m. At the toe of rubble wall, there
was a large monsoon drain of 6m wide and 3m high.
Figure 5 shows the cross section of the rubble wall
and the monsoon drain. Plate 10 shows the picture
of the completely collapsed rubble wall and the RC
drain section. Most of the platform of these units
was constructed over the previous natural stream by
earth filling. The maximum thickness of fill was
about 25m.
The failure occurred during the
monsoon month with frequent heavy raining. The
preliminary finding on the cause of the collapse was
mainly due to the brittle behaviour of the rubble wall
when subjecting to differential distortion along the
wall as a result of fill settlement. The slab
supporting the rubble wall was believed to be
initially designed as a seating plinth for the rubble
wall, but the piles was subsequently provided with
the objectives of avoiding settlement and to provide
a better support to the rubble wall.

Plate 10

Figure 5

Collapsed Monsoon Drain

Cross Section of the Rubble Wall and

Monsoon Drain

Differential settlement of filled platform could

also cause distresses to the infrastructure works.
Plate 12 shows the shear cracks with the whitish
calcite deposits on a cast-in-situ reinforced concrete
(RC) monsoon drain due to differential fill
settlement. Figure 6 shows the schematic diagram
on the cause of the shear cracks in association with
the uneven fill settlement.

However, due to the brittle behaviour of the

rubble wall, the wall cracked even at very small
distortion between the piles and the unsupported mid
span of the plinth after the platform settled. Once
the wall cracked, the stones initially binded by the
cement mortar became loose mass of stones and
collapsed. Plate 11 shows the cracks of rubble wall
due to such distortion.

Cracks

The collapsed mass piled up and surcharged the

RC monsoon drain sections causing subsequent
bending failure at the wall-base connection of the
section. The detailed analyses indicate that the
monsoon drain had sufficient resisting capacity to
retain the backfill if the rubble wall was intact. The
collapse had triggered retrogressive instability failure
of the retained platform to the units of adjacent

Plate 11

Shear Cracks on Rubble Wall

Conference Technology CONTEC-2001,

Kota Kinabalu, Sabah, 11-13 October 2001

In certain conditions, changes of platform level

due to differential settlement as a result of varying
fill thickness can potentially lead to a very
unfavourable drainage pattern, leading to a failure.
Plate 13 indicates the serious erosion of a completed
platform of the industrial lots. The area of erosion
is the location of previous stream with fill over it.
From the site observation, it is evident that
depression was formed around the previous valley
where the former stream was located.
Theis
depression forms a perfect collection point for
surface runoff and channelise the runoff to a
constrict flow running towards the edge of platform.
Unfortunately, the intercept drain is not properly
maintained and silted up. As a result, the surface
runoff has chosen to flow over the berms and erode
into gullies.
These gullies initiated from the
constricted flow on the platform eventually form a
natural earth drain and damage the monsoon drain.
Plate 13 shows the failure of monsoon drain due to
erosion.

6.3

This case history was featured with excessive

distortion of the two-storey terrace structures due to
differential settlement between the fill of varying
thickness and the cut ground. Due to the uneven
settlement of the platform across the longitudinal
axis of the terrace structures, diagonal tension cracks
were developed on the structural frame and walls.
A 150mm asbestos concrete water pipe was also
damaged and leaked due to the differential
settlement.
The leaking water subsequently
saturated the fill material and initiated collapse
settlement and further aggravated the distortion.
Figure 7 shows the schematic diagram of the failure
and the failed slopes is shown in Plate 14.

Tilted
Wall

Shear
Cracks

Shear
Cracks

Plate 12

Case Three : Distortion of Building

Figure 7

Shear Cracks on Monsoon Drain

Plate 14

6.4

Figure 6

Building Distortion due to Differential

Fill Settlement

Shear Cracks on Monsoon Drain

Failed Slope due to Differential Fill

Settlement

Case Four : Differential Settlement at Bridge

Abutment

This case history shows the differential settlement

behind a bridge abutment in Pulau Batam, Indonesia.
It can be understood that proper compaction at a
confined areas is difficult.
As a result, the
settlement between the fill sitting on sedimentary
bedrocks and the piled structure forms a depression,
which affects the riding comfort and can also pose
danger to vehicle passengers as the tyre resistance
may temporarily detached from the depressed
pavement. Plate 15 shows the depression behind
the abutment.

Soil Erosion

Collapsed
Section

Plate 13 Erosion Failure of Monsoon Drain

10

Conference Technology CONTEC-2001,

Kota Kinabalu, Sabah, 11-13 October 2001

6.

7.

8.

Plate 15 Differential Settlement behind

Abutment Fill
7

CONCLUSIONS AND RECOMMENDATIONS

It is very clear that fill compaction has important

impact on development with fill ground. If not
properly addressed during the early stages of project
implementation, the consequences can be
catastrophic and costly.
This has been
demonstrated in the few case histories presented in
this paper.

2.

3.

4.

5.

REFERENCES

AS 1289. Testing of Materials for Engineering

Purposes. Standards Association of Australia.
AS 3798 : 1990. (1990). Guidelines on Earthworks
for Commercial and Residential Developments.
Sydney, Standards Association of Australia.
BEM. (2000). Supervision of Construction of
Projects. Circular to All Professional Engineers
Ref: (20) dlm LJM 28 Jld 4, Board of Engineers,
Malaysia, 21 June 1989. Kuala Lumpur:
December 2000 issue of Bulletin Ingenieur,
Malaysia, pp35.
BEM. (2001). Supervision of Construction of
Projects. Circular to All Professional Engineers
Ref: (20) dlm LJM 28 Jld 4, Board of Engineers,
Malaysia, 21 June 1989. Kuala Lumpur: March
2001 issue of Bulletin Ingenieur, Malaysia, pp12.
Charles, J. A. & Watts, K. S. (1996).
The
Assessments of the Collapse Potential of Fills and
Its Significance for Building on Fill. Proc. Instn
Civ. Engrs Geotech. Engng: 15-28, 119, Jan 1996.
Charles, J. A., Skinner, H. D. & Watts, K. S. (1998).
The Specification of Fills to Support Buildings on
Shallow Foundations: the 95% Fixation. Ground
Engineering: 29-33, January 1998.
Gue S.S., Liew S.S. & Tan Y.C. (1999).
Instrumentation results on negative downdrag
force of abutment piles underneath the reinforced
earth wall embankment. Field measurements in
geomechanics FMFM99; Proc. 5th Intern. symp.,
Singapore, 1-3 December 1999. Rotterdam:
Balkema. (Separate print)
Proctor, R. R.. (1933).
The Design and
Construction
or
Rolled
Earth
Drains.

The following recommendations are given to

prevent and mitigate the non-compliance:
1.

Light compaction to form thin impervious

layer using smooth-wheeled roller should be
carried out on the last compacted lift at the
end of the day to facilitate surface runoff of
night rain. Scarification of the smooth
compacted surface between lifts in finegrained soils should be carried out to
eliminate lamination and poor bonding.
Field tests of statistically representative
quantity should be carried out to check for
non-compliance.
Guidelines of testing
frequency have been given.
Supervision of compaction should be
enforced for any earthworks operation, not
only to satisfy the law but also to ensure
compliance as it has significant effect to the
subsequent construction works.

It is important to carry out necessary

compaction tests on each type of material
sources to establish the target reference for
field compaction achievement.
The design engineer should only specify the
actual required compaction effort rather than
over-specifying for extra unjustifiable
comfort. Sometimes, over-compaction can
cause unexpected problems, such as
generating slickensided slip plane and
altering particle grading of compacted soil
due to particle breakage.
The loose compaction lift should be specified
to ensure minimum dry density within the
compacted lift based on the performance of
trial compaction for specific compaction
equipment.
There is a need to limit the largest particle
size to ensure proper compaction and
uniformity of the compactness of fill.
It is important to control the range of
moisture content to achieve the target dry
density under specific compaction effort.
Sometimes, conditioning of the fill materials
is required to control appropriate moisture
content at placement and compaction.

11

Conference Technology CONTEC-2001,

Kota Kinabalu, Sabah, 11-13 October 2001

Engineering News Record III, 31 August, 7, 21

and 28 September.
Trenter, N. A. & Charles, J. A. (1996). A Model
Specification for Engineered Fills for Building
Purpose. Proc. Instn Civ. Engrs Geotech. Engng:
219-230, 119, Oct 1996.
Whyte, I. L. & Vakalis, I. G. (1987). Shear
Surfaces induced in Clay Fill by Compaction
Plant. Compaction Technology, Proceedings of
the Conference by New Civil Engineer,
London:125-137.

12