Tech5-4 Pozo Profundo PDF

Geotechnical Engineering Journal of the SEAGS & AGSSEA Vol. 41 No.
3 September 2010 ISSN 0046-5828
RECENT ADVANCES IN PILE TESTING AND DIAPHRAGM WALL

CONSTRUCTION IN JAPAN
Kenji Ishihara
Chuo University, Hachiji, Tokyo, Japan
E-mail: Kenji Ishihara kenji-ishihara@e-mail.jp
SUMMARY: The first part of this paper consists of brief introduction of the in-situ pile loading tests that have been conducted in Japan over
the last two decades in connection with the design and construction of high-rise buildings in areas of soft soil deposits. In addition to the
conventional types of tests in which the load is applied at the top and at the toe of the pile (O-cell test), what may be called pile toe bearing
test and skin friction test is introduced. The results of these tests are described and compared with those from the conventional type of the
pile loading tests.
In-situ prototype tests are also introduced in which bearing power of Barrette type pile is compared with that of the
circular type pile. A special case of in-situ pile loading tests conducted in Singapore is also introduced in which the friction between the
circular ring-shaped concrete segment and the surrounding soil deposit was measured directly during excavation of the shaft by applying
loads up and down by jacks installed between two adjacent segments in vertical direction. The second part of this paper is a brief description
on constructions of large-diameter circular diaphragm walls that was carried out about 10 years ago for the LNG storage tank in the coastal
site in Tokyo Bay.
The construction of the large-scale Kawasaki Island in the middle of Tokyo Bay in Japan will also be introduced.
The
whole scheme and process of construction is for these two undertakings is introduced with some comments on observed behaviour of the
walls and on special precaution taken during construction.
INTRODUCTION
whether a given deposit is competent enough to mobilize the

necessary bearing power.
Some of the results of in-situ pile
In the design and practice of foundations supporting heavy-weight
loading tests recently obtained in Japan by this method are
structures such as high-rise buildings or large-scale facilities, of
introduced in this paper.
utmost importance is to know the load-carrying characteristics of

soil or rock deposits at depths as deep as 30m-50m.
In the case of the deep-seated foundation supported by the
Estimates of
diaphragm walls, the O-cell test is not applicable and some other
the load-bearing capacity for piles embedded into such deep-seated
methods need to be explored. In important projects, efforts have
deposits have been made generally by carrying out what is known
been made to measure directly the bearing power or skin friction
as the O-cell test or pile toe loading test.
In some cases, direct
within excavated pits or shafts by means of the plate loading tests
loading tests are performed by large concrete blocks or segments
performed on exposed surface of intact stiff sand, gravel and rocks.
on the intact surface of the deep deposits which are exposed during
As one of the attempts in this context, an example of in-situ loading
excavation of pits or shafts for foundation construction.
tests conducted in Singapore is introduced in this paper.
The
outcome of these in-situ tests has not unfortunately been publicized
The other topics of importance in the foundation engineering
and compiled to the extent that these data can be effectively
would be the design and construction of the large-scale diaphragm
utilized for the design practice in the future. In view of this, some
wall, particularly when it is used as circular walls without any
data from the in-situ pile loading tests performed in Japan were
bracing.
collected and presented in this paper.
distribution in the circumferential direction are important factors to
It is to be noted that the soil deposit near the surface in the area
In the roundshaped walls, the hoop compression and its
be considered with utmost precaution.
The method of design and
of industry and mega city in Japan, is generally composed of soft
results of monitoring in this regard will be introduced by quoting
soil of alluvial origin underlaid by alternative layers of stiff sand,
an example of constructing a large LNG storage tank in Tokyo.
clay, sand and gravel of diluvial origin.
It is the generally adopted
When the circular wall is constructed under the seabed, one of the
criterion that some of the stiff sand or gravel layer can be chosen as
major concerns is the safety against the bottom heave and piping
a bearing layer to support the piles.
due to high hydraulic gradient of seepage water.
Thus, the end bearing type
These issues will
piles is frequently used in Japan rather than the friction type pile.
be briefly touched on by quoting an example of the large scale
However, since the stiff bearing layer is seated at great depths of
construction of Kawasaki man-made island in the middle of Tokyo
about 20-50m, the O-cell tests have often been adopted to identify
Bay in Japan.
Geotechnical Engineering Journal of the SEAGS & AGSSEA Vol. 41 No.3 September 2010 ISSN 0046-5828
Accumulation of in-situ data on pile testing is so much both in
internally by oil pressure, creating an upward force on the shaft and
content and volume that it was difficult to overview all of them and
simultaneously an equal magnitude of force downwards to the
come up with a state-of-the-art text from a fair point of view.
bottom of the shaft.
What is written here is part of the test data which is considered
power of the pile toe is resisted by the upward force mobilized by
appropriate in the context of the present presentation.
the friction on the surface of the pile body.
There are
The downward force to monitor bearing

At the same time, it is
also many construction examples of circular diaphragm walls.
possible to monitor the shaft friction as well while the reaction is
Because of the limitation in several respects, it was not possible to
taken by the downward directed bearing power. The mutually
prepare a comprehensive overview of the current state of the
counteracting forces as above are utilized to apply two major forces,
practice.
However, it is believed that the main points of the
that is, for the end bearing test and also for the shaft resistance test.
current practice in Japan are understood via the typical examples
The upward movements of the pile at its toe and downward
introduced here.
displacement of the ground at the pile toe can be measured

independently via the two long bars with their ends attached
2.0
PILE TOE LOADING TEST
through the pipes to the top and bottom of the pressurized jack.
Thus, the upward displacement of the shaft is known along with the
2.1
General
downward movement of the pile toe, thereby permitting the

load-displacement relation to be established for the shaft resistance
The pile top loading test in which loads are applied directly on top
as well as the end bearing power.
of piles has long been the method commonly adopted to verify
The advantages of using this technique are summarized as follows.
actual bearing capacity of piles in the field. When the size of piles
1)
There is no need to provide large equipments such as heavy
is relatively small corresponding to a low magnitude of design load,
weights or reaction piles on the ground surface for applying
this method proved to be the most practical and useful tool for
the force on the pile top, making it easier and inexpensive to
assessing the load-carrying capacity of piles. However, the demand
install the test setup.
for greater bearing power has increased in recent years in unison
guages for monitoring pressure and displacements are the only
with increasing size and weight of buildings and facilities. Thus,
major apparatus for the test setup on the ground surface which
required size and embedment depth of piles have increased
are simple and installed in a limited space.
remarkably. In such situations, the top loading test has become no

longer recognized as an effective means to evaluate in-situ
performance of large piles,
2)
The oil pump and several ancillary
As far as the end bearing is concerned, the ultimate value is

measured directly, as a load of high intensity can be applied
just to the soil deposit adjacent to the pile toe.
because it requires large trestles and facilities on the ground surface
The above method, as referred to as O-cell test has been used
to apply heavy weight to the pile head, and therefore becomes
widely for evaluating the in-situ performance characteristics of
expensive.
large and long piles.
To cope with this, what is called pile toe loading test has become
In Japan, the original setup was modified to
some extent particularly in the mechanical part for applying the
a commonly used procedure. This method consists of applying up
pressure at the pile toe.
Instead of the pressure cell, a mechanical
and down forces at the tip of piles by means of a large jack which
system consisting of a large-capacity jack at the pile toe and the
has been embedded in advance at the pile toe.In its original form
measuring device controlled on the ground surface have been
proposed by Osterberg in 1984 for bored piles, a flat pancake-like
incorporated and in common use for carrying out the pile loading
cell was placed at the bottom of the drilled shaft before the concrete
tests.
was poured. After the concrete has cured, the cell was pressurized
schematically illustrated in Figure 1.
The test setup of general use in Japan for the O-cell tests is
Figure 1
2.2
Schematic diagram for pile toe loading test (Ogura et al. 1997)
Pile Toe Loading Test at Kanto Postal Bureau Site
at the pile toe through the procedure as illustrated in Figure 3.
In
the present case, a hole with a diameter of 2.2m was drilled to a

Taking advantage of the construction of the new 28-storey building
depth of 48m.
for the postal bureau office in Saitama Prefecture due north of
cement-mortar was lowered through the mud water to provide a
Tokyo, a series of pile tip loading tests was conducted by Shibuya
sound base for the pile toe and then the jack-attached reinforcement
et al. (1997) to assess the end bearing capacity of cast-in-place
cage was lowered so that the jack could be made to sit at the right
reinforced concrete piles.
position of the base.
The building was 129.8 high and had
After cleaning the bottom, a hopper laded with
Concrete was then poured to form the test
The site is located at the northern part of
pile. The area of cross section at the pile tip was Ap=3.8m2.
Ohmiya terrace area and consists of a series of soil layers of
After curing for 34days, the load was applied cyclically to the jack.
diluvial origin, as displayed in Figure 2.
One of the results of the pile toe loading tests is demonstrated in
two-storey basements.
Below the surface fill,
there exists a dense sand deposit underlaid by silty clay with a SPT
Figure 4.
N-value of about 5.
Shown in Figure 4(b) are the settlements, Sp, plotted
Below this deposit, a thick dense sand
versus the load, Pp, both monitored at the pile tip. It is known
deposit with N>50 was shown to exist at depths between 45m and
that the ultimate value of the end bearing is about 30MN and this is
60m, which was considered competent enough to act as an
mobilized when the pile tip was displaced by 50cm. The effective
end-bearing layer of the piles.
nominal bearing capacity is normally defined in Japan as the load
To substantiate its bearing capacity, a jack 1.3m in diameter

and 1.044m high with a movement stroke of 50cm was embedded
intensity at the stage when the pile is displaced by an amount equal

to 10% of the diameter of the pile tip.
Figure 2
Figure 3
Soil profile at the site of Kanto Postal Bureau (Shibuya et al. 1997)
Procedures for installing a jack at the toe of the cast-in-place reinforced concrete pile
Figure 4
Results of the pile toe loading test at the site of Kanto Postal Bureau (Shibuya et al. 1997)
In this case, by reading off a value of 23MN corresponding to

the settlement of 10% of the pile diameter, the nominal bearing
capacity
was
2
estimated
2
6.05MN/ m =605t/ m .
as
23MN/3.8
m2
In this particular test, values of upward
movement of the pile top, Uo, was also monitored on the ground
surface.
The measured uplifts at the top, Uo, and at the pile tip,
Up, are shown in Figure 4(a) as plots versus the load, Pp, applied at
the pile tip.
It can be seen that the amount of uplift at the pile tip
is greater than that at the pile top indicating the fact that the pile
body tended to shrink due to the compressive load.
To monitor the friction along the shaft, the strain gages were pasted
on the reinforcement at five depths as indicated in Figure 2.
By
reading the compressive strains of the gages, it was possible to

calculate the total magnitude of the vertical load being applied to
the cross sections at each of the five depths.
The axial load
distributions thus obtained through the depth are displayed in

Figure 5 where it is noticed that the reduction in the axial load
between two neighboring points of measurement is taken over by
the equal magnitude of friction acting on the surface of the pile
Figure 5
Axial load versus depth in the pile toe loading test at

Kanto Postal Bureau (Shibuya et al. 1997)
body.
The frictional force intensity between two adjacent points of
casing as accordingly shown in Figure 2. The friction between
measurements thus calculated is shown in Figure 6 as plots versus
Section 4 and the jack at the pile tip is known to reach its maximum
the displacement between these two points. It can be seen that the
value of 20t/m2 when it was fully mobilized. This value was almost
frictional force between the Section 1 and the top is almost equal to
equal to the friction of 17t/ m2 which had been postulated when the
zero indicating that the friction was cut off by the presence of the
design was made.
Figure 6
Frictional force intensity on the shaft versus the displacement of two neighbouring points
(Shibuya et al. 1997
2.3
Load-Settlement Curves At The Pile Top Estimated From
The Pile Toe Loading Tests
corresponding to the settlement of 50cm.

The curve predicted by the model is seen shifting rightwards by
an amount approximately equal to the toe resistance which was
The load-settlement relations, that is, Pp vs. Sp curves can be
gradually mobilized with increasing settlement.
There are many
obtained in-situ with the optimum cost as described above but what
other cases of similar studies in which the
Po So
is indeed necessary in the design of piles is the load-settlement
predicted by the
relation,
Po vs. So curve
at the top of piles. There are several
coincidence with
curve
Pp Sp relation via the model shows a good

the Po So relation actually measured at the
models proposed to estimate the relation at the top based on the
pile top. Thus, it is now considered appropriate to perform the
results of the pile toe loading test.
pile toe loading test and to estimate the
One of them is what is termed
Load transfer method proposed by Kishida and Tsubakihara

(1993).
Kanto
Using this method, the Po

Postal
Bureau
Pp Sp relation
site
was
So curve
evaluated
Po So curve at the pile
top to be incorporated in the design.
at the pile top at

based
on
the
2. 4
Compiled Results of The Pile-Toe Loading Tests
which was represented by a hyperbolic curve.
The result of the analysis is shown in Fig. 7, together with the
There have been a majority of test data of similar nature ever
Pp Sp relation obtained from the pile toe loading test.
performed in Japan at various sites to obtain the end bearing
original
It is noted that the ultimate value of the skin friction, that is,
capacity of dense sand deposits and soft rocks.
Pf 30MN , which should be equal to the upward load at the
summarized by Ogura et al. (1998) as shown in Figure 8 in terms
pile toe, can be estimated directly from the test result
of the load intensity at the pile toe,
Pp / A p ,
These were
plotted versus the
settlement,
pile tip.
Sp ,
divided by the diameter of the pile
Dp
at the
considered competent; the end bearing capacity is greater in soft
It may be seen in Figure 8 that the pile tip settlement is
rock or gravelly sand deposits as compared to that in the dense sand
much smaller for soft rock and gravelly sands than that for sand
deposits.
deposits. This implies the fact that, for the stiff deposits generally
Figure 7
The load-settlement curve at the pile top estimated from the P p-Sp relation by
using the load transfer method (Shibuya et al. 1997)
3.0
PILE TOE BEARING TEST AND SHAFT RESISTANCE
TEST
the shaft friction.
The magnitude of the force at the pile top is
slightly smaller than the upward force from jack

equal to the weight of the pile body.
by an amount
By means of this test, it is
3.1 Concepts Of The Pile Toe Bearing Test And Skin Friction
anticipated that the genuine value of end bearing capacity free from
Test
the influence of the shaft friction can be measured/obtained.

The fourth type of test is the one called Skin-friction test.
This
Shown in Figure 9 are the illustrations for four types of pile loading
type of test was proposed to cope with the criticism saying that in
tests ever performed. Figure 9(a) is the normal type of the pile top
the widely used O-cell loading test, the frictional force along the
loading test.
The pile toe loading test as described in the
shaft is mobilized in the downward direction which is opposite to
foregoing section is illustrated in Figure 9(b). This test can be
the direction activated in the normal working conditions of in-situ
carried out to know the bearing capacity of the pile toe.
However,
piles. In an attempt to shed some light on this aspect, the skin
there remains some influence in the vicinity of the pile tip, of
friction test was developed by Dr. Ogura and his associates (2005).
reducing the vertical confining stress due to the shaft friction
As illustrated in Figure 9(d), this test consists of applying an axial
directed upwards.
To eliminate this effect, if any, what is called
force, first of all, at the pile top which is approximately equal to the
Pile toe bearing test was proposed as illustrated in Figure 9(c).
upward force at the pile tip while the end-installed jack is in
In this type of test an axial load approximately equal to the upward
operation. Then, the pressure in jack is released
force at the end jack is applied in addition at the pile top to discard
Figure 8
Summary of the pile toe loading tests (Ogura et al. 1998)
There is no force acting upward from the pile toe.
Thus, the
A-pile and B-pile, were used for in-situ loading tests.
The A-pile
axial load, Po, applied at the pile top is totally resisted by the
with a diameter of 1.0m was embedded to a depth of 42m where a
friction mobilized on the surface of the shaft.
0.5m-height jack was installed. Below this depth, the pile was
This is the principle
of performing the skin friction tests.
modulated through the 5m-length to a depth of 47.5m where a sand

layer with N-value of 24 was encountered.
3.2
Pile Toe Bearing Test And Skin Friction Tests At Kaifu
For this modulated
portion near the bottom, the diameter was 1.0m but for the
Site In Aichi Prefecture
non-modulated part the diameter was 80cm.
The test site is located at Kaifu near Nagoya where the deposits
loading test by the A-pile was to know the toe bearing capacity of
consist of a series of soil layers as shown in Figure 10. The
the sand deposit.
depths at which the jack was installed on two piles A and B are also
straight pile to a depth of 44m where a 0.5m in height jack was
indicated in Figure 10.
attached.
The site is located in the flood plain area
The aim of the
The B-pile with a diameter of 0.8m was a
Below this, the modulated pile with the diameter of
composed of layers of soft silt and clay deposits with the SPT
1.0m was jointed downwards as illustrated in Figure 10.
N-value of the order of 0 to 5. There exists a gravelly sand layer
of the test by the B-pile was to examine the bearing power of the
at the depth of 48.5m.
gravelly sand layer with a N-value of 60 at a depth of 49m.
Two precast concrete piles, denoted by
The aim
Figure 9
Four types of pile loading test
Figure 10 Soil profile at Kaifu site and embedment of two piles for toe bearing
and skin friction tests (Ogura et al. 2005)
The prefabricated piles were installed in pre-bored holes after the
toe loading are, by and large, the same for the two plots of Pj versus
bottom had been cleaned up.
Sj and Sp.
Milky cement was poured and
It can also be seen in Figure 11 that the ultimate
churned with the soils at the bottom to prepare sound beds for
bearing power of 10.87MN was achieved with a settlement of
placing the piles. After the piles were put in place, the cement
17.6cm at the pile toe in the sand deposit of N=24 at the depth of
mortar was poured to fill the annular space between the precast pile
47m.
and the surrounding soils.
Measurements were made of the axial strains by means of strain

gauges pasted to the reinforcement cage through the pile length as
3.2.1
Pile Toe Bearing Test
indicated by the dotted points on the right-hand side of Figure 10.

Over the portion of the pile from the top down to the jack point, the
The results of the pile toe bearing test on the A-pile is demonstrated
axial load monitored by the strain gages was shown to take
in Figure 11 in terms of the jack-applied load Pj plotted versus the
approximately an identical value which is equal to Po Pp .
downward displacement of the pile, Sj, at the point of the jack and
the 5m-long portion from the jack point to the pile tip, the axial
also the displacement, Sp, at the pile tip.
In
It can be seen in Figure
force as measured by the strain gages at five points is displayed in
11 that the displacement at the jack point is slightly larger than that
Figure 12, where it can be seen that the axial load tends to decrease
at the pile tip.
with increasing depth from the depth 42m to 47m due to the skin
This is due to a slight compression of the pile body
taking place through the 5m-long portion between the jack point
and the pile tip.
friction mobilized over the 5m long portion above the pile toe.
However, the load-settlement curves for the pile
Figure 11
Load-displacement curves from the pile toe bearing test at Kaifu (Ogura et al. 2005)
10
Figure 12
3.2.2
Load-displacement curves obtained from the pile toe bearing tests at Kaifu (Ogura et al. 2005)
Skin Friction Test
3.2.3
Comparison Of The Toe Bearing Capacity Between The
Sand And Gravel Deposits

Following the toe bearing test on the A-pile, the skin friction test
was carried out on the same pile by reducing the jack pressure
As demonstrated in Figure 10, the B-pile was put in place in the
while keeping the pile top load unchanged.
sandy gravel layer at the depth of 49m where the SPT N-value was
The reduction in the
jack pressure was assumed to be equivalent to the increase in the

load
Po
at the pile top which is regarded as indicating the
corresponding increase in the skin friction Rd.

test is displayed in Figure 13 in terms of
Po
The result of the
or Rd plotted versus
60.
The result of the pile toe bearing test conducted on the B-pile
are shown in Figure 15 where the end bearing load intensity, q p ,

as defined by q p Pp / A p is plotted versus the vertical
displacement, Sj. at the jack point or at the pile tip,
Sp .
The
the vertical displacement, So , at the pile top or Sj at jack point. It
result of the pile toe bearing test on the A-pile shown in Figure 11
may be seen that, for a given axial load at pile top, the
is also replotted in Figure 15 in the same fashion for comparison.
displacement at the jack point, Sj, is smaller than that at the pile top,
It is obvious from Figure 15 that the end bearing load intensity
indicating that this difference corresponds to the shortening of the
q p in the gravel is much higher than that in the sand deposit.
pile body.
The depthwise distribution of the axial load through
the pile length as evaluated by the strain gage reading is displayed

in Figure 14.
It can be seen that the maximum load, Po , at the
top tended to decrease with depth and became equal to zero at the
pile tip for all the steps of loading.
11
Figure 13
Figure 14
Skin friction distributions through the toe portion of A-pile (Ogura et al. 2005)
Distribution of the axial force versus depth in the skin friction tests at Kaifu (Ogura et al. 2005)
12
Figure 15
Displacement at pile toe, Sp or at jack point Sj versus end load intensity, q p (Ogura et al. 2005)
4.0 BEARING POWER OF BARRETTE PILE AS AGAINST
Figure 17 together with the side view of the two piles.
There is a
CIRCULAR PILE
thick wind-blown deposit of clayey silt termed Kanto loam

having the SPT N-value of the order of 5 to 30 to a depth of 14.5m.
The rectangular pile called Barrette pile has been of frequent use in
Below the loam layer, there exists a stiff sandy soil deposit of
recent years particularly for installing friction piles, because of its
diluvial origin with the N-value in excess of 50.
greater amount of shaft resistance that is mobilized in comparison
installed by lowering the rebar cage into the prepared hole and
with that of the circular pile. This effect is envisioned highly
concreted.
likely as the surface area on the shaft is apparently larger in the
Strain gages were attached to the reinforcement cage of the piles to
Barrette-type pile.
measure the axial load at several depths as shown in Figure 17. The
In order to verify this aspect, in-situ pile loading tests were
tests were of normal loading type in which the load Po is applied at
conducted by Iwanaga et al. (1991) where the behaviour of the
the pile top and the settlement measured at the top.
circular pile was compared to that of the rectangular pile having an
The result of the pile top loading test on the circular pile is
identical cross sectional area.
presented in Figure 18 in which the pattern of the load-reload
the two pile loading tests.
Shown in Figure 16 is the layout of
The circular pile 1.2m in diameter had
the same cross sectional area of 1.13m
2.0 0.6m rectangular
as that of the
pile, but the latter had a surface area
which was 1.38 times larger than that of the former.
There were
Two piles were
cycles is displayed versus time on the upper left-hand side.
The
progress of the settlement with time is shown on the lower

left-hand side in Figure18.
On the basis of the time-changes thus
obtained, the load-settlement curve in cyclic loading is established
two sets of earth anchors as shown in Figure 16 which were
as shown in the lower right-hand side of the diagram.
installed to carry reaction force when the load is applied on top of
off the maximum values of load and settlement at each cycle, the
the piles.
load-settlement curve was obtained as shown in the upper
The test site was located at Tsukimino in Kanagawa
Prefecture due west of Tokyo.
The soil profile there is shown in
By reading
right-hand side.
13
The results of the pile top loading test on the rectangular pile is
embedded through 2.5~3.0m into the stiff sand layer, as shown in
demonstrated in Figure 19 in the same fashion as that of the loading
Figure 17. Based on the measured data, the equivalent axial force,
test on the circular pile.
Comparison of the load-settlement curves
Pp, mobilized by the friction at the portion of the pile toe was
in the upper right-hand side of Figures. 18 and 19 does apparently
estimated. The force Pp divided by the area Ap at the toe is plotted
indicate that the bearing power is greater for the rectangular pile
versus the settlement at the pile tip Sp in Figure 20(b). The
than for the circular pile. To demonstrate this feature more vividly,
settlement Sp was taken approximately equal to that at the pile top
the load-settlement curves for the circular pile and rectangular pile
So. As shown in Fig.ure 20(b), the ultimate load intensity for the
are shown together in Figure 20.
In Figure 20(a), the settlement,
rectangular pile was 11MPA as against 8.6MPA in the circular pile.
So, at the pile top is plotted versus the load, Po, at the pile top,
Thus, there was 1.24 times increase in the bearing power mainly by
where it can be clearly seen that the rectangular pile exhibited the
frictional resistance at he portion of the toe in the case of the
ultimate bearing power of 17MN as against 13MN in the circular
rectangular pile. It is obvious that the increase has emerged from
pile.
the fact that the rectangular pile (Barrette-type pile) has the larger
The ratio of increase in the bearing power in the Barrette
pile is about 1.3 times as compared to the circular pile.
surface area over which the frictional resistance at the toe portion
There were strain gauges equipped at the pile toe portion
Figure 16
can be mobilized, even if the cross sectional area is the same.
Layout of the pile loading tests, one on a circular and the other on a rectangular pile at Tsukimino
(Iwanaga et al. 1991)
14
Figure 17
Figure 18
Soil profile and side view of the two piles (Iwanaga et al. 1991)
Load-settlement-time relations obtained by the pile top loading test on circular pile,
Tsukimino, Kanagawa (Iwanaga et al. 1991)
15
Figure 19
Load-settlement-time relations obtained by the pile top loading test on rectangular pile,
Tsukimino, Kanagawa (Iwanaga et al. 1991)
Figure 20
Load-settlement curves from the pile loading tests on the circular and rectangular piles
at Tsukimino (Iwanaga et al. 1991)
16
5.0 SHAFT RESISTANCE IN DEEP FOUNDATIONS
The entire floor area of the building complex is enclosed by the

diaphragm walls with a thickness 0.8m to 1.2m which was
5.1 Outline Of The Project And Construction
embedded to a depth of 15m to 40m where the competent residual

soil or bouldery clay layer exists.
The diaphragm walls were
In-situ measurements of shaft resistance were carried out in the
designed to act as the permanent walls for the 3-level basement car
excavated shaft during the construction of cast-in-place reinforced
park.
concrete piles for the foundation of the United Overseas Building
section CA3 through CA6 is demonstrated in Figure 26 where it
in Singapore (UOB).
can be seen that the upper part of the piles through soft alluvial
The 66-storey building 281m high was
constructed in 1992 in the central business district of the city.
As
The side view of the caisson and bored piles along the cross
deposits (Kallang Formation) consists of the caissons enclosed by
shown in Figure 21, it is located on the right bank of Singapore
the diaphragm walls.
River about 500m from its exit to the Marine Bay.
composed of the segment of cast-in-place reinforced concrete.
as it stands now is shown in Figure 22.
The building
The lower part of the foundation was
The approximate soil
The method of constructing the caissons in the upper part consisted
profile in the east-west direction is displayed in Figure 23 where it
first in installing the diaphragm walls through water-saturated soft
can be seen that there exits a marine clay deposit about 10~30m
ground, as illustrated in Figure 27. For the CA8-pile having an
think overlaid by a surface fill and peaty clay layer.
Below the
inner diameter of 7.6m, the diaphragm wall was constructed by
marine clay, there is a residual soil deposit of medium stiffness
following the three steps of cutting 1, 2 and 3 as illustrated in
which is underlaid by the bouldery clay. The layers of alluvial
Figure 27(c).
deposits above the bouldery clay are called Kallang Formation.
soil behind, excavation was advanced in dry conditions by
According to the paper by Han et al. (1993), the bouldery clay
machines. In the course of downward excavation, the reinforced
deposit is composed of stiff to hard soil deposits interspersed with
concrete ring beams were constructed in-situ at an interval of 5~6m,
sandstone-derived boulders. The boulders varying in size and in
to provide circular bracing for the diaphragm wall. When the
shape from a few centimeters to 6m occupy about 20~30% of the
excavation reached the level of the bouldery clay deposit, that is,
total volume. Thus, the boulders are seldom in contact with each
the toe of the diaphragm wall, the cutting was carried out further
other.
The soil/clay matrix behaves essentially as a stiff to hard,
down in open conditions while protecting the exposed soil surface
heavily over-consolidated clay and it tends to soften when put in
by placing the ring-shaped cast-in-place reinforced concreting
contact with water. The undrained shear strength of the clay matrix
circular segments 40cm thick and 2m wide.
By sealing off the groundwater and retaining the
as evaluated via the pressiometer test and plate loading test was
shown to be about 1.0Mpa 10kgf/cm2.
The bouldery clay
5.2 In-Situ Measurements Of The Shaft Resistance
deposit is estimated to extend down to about 120m depth and the

majority of high-rise buildings in the district are founded on the
As the excavation proceeded through the bouldery clay deposit,
base structures constructed in this bouldery clay deposit as shown
in-situ loading tests were performed at a total of 26 spots.
in Figure 24. Construction of the foundation for the UOP Plaza is
exact locations of the loading tests conducted in the shaft CA3 and
described somewhat in details by Iwanaga et al. (1991) and
CA4 are indicated in Fig.ure 26. The scheme of the loading test at
Imamura et al. (1991).
the bottom of the current stage of excavation is schematically
The following is the excerpt from the
The
reports.
illustrated in Figure 28. The lower segment B of the concrete wall
The UOB plaza carrying a total weight of 1512MN was constructed
to be tested had been constructed against the soil wall with an open
on the foundation with 12 caissons supported by the shaft friction.
gap 75cm wide on top of the bottom segment C.
The plan view of the layout of the foundation is demonstrated on
segment A of the concrete ring for testing had also been constructed
the left side of Figure 25. The right side indicates the portion of
with an open gap of 75cm apart from the bottom of the
strutted excavation to construct the 5-storey podium attached to the
already-constructed upper segment C.
The upper
high-rise tower.
17
Figure 21
Location of the UOB Building in Singapore
Figure 22
UOB Plaza in Singapore
18
Figure 23
Figure 24
Site geology around the UOB Building in Singapore
Typical features of building foundations in the central business district in Singapore
19
Figure 25
Figure 26
Layout of the bored piles for UOB Plaza
Layout of the caissons and bored piles for the UOB Plaza Building
20
Figure 27
Construction of diaphragm walls and RC lining for foundations of UOB Building
Twelve hydraulically operated jacks were placed along the concave
shown in Figure 30(b), where it may be seen that the ultimate shaft
shelf in the annular space 80cm in height as illustrated in Figure 28.
friction was somewhere between 1.8 and 2.0MN. From the series
The setup of the jacks along the circular shelf is shown somewhat
of the loading tests as above, the ultimate value of the shaft
in details in Figure 29.
resistance was obtained for the cast-in-place reinforced concrete
Measurements of the vertical
displacement were made during the jack-driven loading by means

of several dial gauges as well as by means of the transit viewing
apparatus, as illustrated in Figure 28.
Some of the test results are demonstrated in Figure 30 in terms of
segment which is installed in contact with the bouldery clay in

Singapore.
The outcome of the tests on 26 ring segments with the diameter of
4.7, 6.2 and 6.8m is summarized in Figure 31 in terms of the shaft
the total applied load plotted versus the measured displacement.
resistance f plotted versus depth. Figure 31 shows that there is
Figure 30(a) shows the load-displacement curves obtained from the
a tendency of increasing shaft friction with increasing diameter of
tests loaded upwards to No. 9 segment and downwards to No. 10
the ring segment. However, looking over the entire test results, one
segment which were located at an elevation of 51m in the CA4
can recognize that the ultimate value of shaft friction was in excess
bored shaft, as accordingly shown in Figure 26.
of 45tf/m2.
It may be seen that the ultimate skin friction mobilized between the
the ultimate skin friction used for the original design were modified
concrete ring wall in 6.8m in diameter and surrounding bouldery
from 45tf/m2 to 35tf/m2, which conduced greatly to reducing the
clay deposit was 220tons=2.2MN for the segment No. 9 but for the
cost of bored pile construction. As a result, the length of the bored
segment No. 10 the ultimate friction would have been greater than
pile was shortened, as accordingly indicated in Figure 22.
Based on the test data as verified above, the value of
this value. The results of the tests conducted at No. 4 and No. 5
segment in the bored shaft CA2 with the diameter of 4.7m are
21
Figure 28
Scheme of the loading test for shaft resistance within the bored hole
Figure 29
Details of jack installation
22
Figure 30
Figure 31
Load-displacement curves obtained from the in-situ loading tests
Ultimate shaft friction obtained from the in-situ loading tests
23
6.0
BEARING CAPACITY OF BORED PILES RELATED
TO SPT N-VALUE
plotted in the abscissa of Figure 32.
In the case of the skin
friction, the N-value at the depth of measurements is directly used

in the abscissa as shown in Figure 33 to represent the stiffness or
A vast majority of in-situ loading tests on bored piles ever
strength of the soil materials surrounding the piles. Shown in
performed in Japan was compiled and summarized in a form of
Figure 32 with black dots are the results of in-situ measurements on
charts in the Japanese design manual of building foundations
end bearing loads for the bored piles which were designed as
(2001).
friction piles.
These are shown in Figures 32 and 33, where the ultimate
toe bearing load intensity,
Pf , and ultimate friction resistance,
f , are shown in terms of the plots against the SPT N-value.
In
the case of the toe resistance, it has been customary to take an

average N-value over the 2m section, viz. 1m above and 1m below
the tip level of piles. This averaged value is denoted by
Figure 32
N and
Summary plots of the ultimate toe bearing load intensity as function of averaged N-value (from Japanese manual)
24
Figure 33
Summary plots of the ultimate skin friction as functions of SPT N-value (from Japanese manual)
Needless to say, the value of the point resistance is shown to be
7.0
relatively small because the major part of the external load is
UNDERGROUND STORAGE TANK
carried by the skin friction.
CIRCULAR DIAPHRAGM WALL FOR
The data points indicated by open
circles in Figure 32 are those derived from the measurements on
In early 1970 techniques have evolved extensively for liquefying
the bored piles designed as end bearing piles.
natural gas and transporting it by ships over a long distance.
Since the skin
friction was not taken into account, the end bearing capacity does
naturally take large values in excess of
Pf 7MN / m .
2
Although there are considerable scatters in Figure 32, an average

line
Pf 233N is drawn through the cluster of data points,
together with two other lines corresponding to the standard
30% . It is recommended in the manual that the

relation Pf 200N be used for design purpose for the deposits
having a N -value less than 60.
deviation of
The
liquefied natural gas (LNG) carried to the site of destinations is

once stored in a large tank.
In response to demand it is
transformed back into gas again and distributed for industrial or

domestic use.
In the wake of this trend, the construction of
large-capacity storage tanks began in Japan in 1970s. In the early

stage, the tanks were constructed on the ground, but for the sake of
easiness of maintenance and safety as well attempts have been
made to construct the tanks underground.
With respect to the ultimate skin friction plotted in Figure 33
One of the early efforts in this context was the construction of the
against the SPT N-value, there also are considerable scatters in the
storage tanks in Sodegaura on the east coast of Tokyo Bay.
f 5.9N is drawn through the

cluster of data points, together with the 50% lines of deviation.
It is recommended in the manual that the relation f 2.5N be
test data, but an average line
The
location is shown in Figure 34. The tank was constructed in 1989

and had dimension 64m in diameter and 44m in depth with a
storage capacity 170,000 k .
The bird-eye view of the roof of the
used for the design of the bored piles.
tank is shown in Figure 35 and its side view is displayed in Figure
It is of interest to see where the result of the shaft friction tests in
36, together with the soil profile.
UOB building in Singapore is plotted in diagram of Figure 33.
of the circular diaphragm wall were two-fold, viz., the thickness of
Assuming that the N-value is greater than 50, the value of
the wall and the depth of its embedment.
f 350kPa 35ton / m
The critical issues for the design
is plotted in Figure 33. It is
observed that this value is located in the vicinity of the average line
deduced from the Japanese case studies.
25
7.1
Wall Thickness
37(b). Therefore, the uniform load as shown in Figure 36 (b) is

generally adopted for performing design calculation.
The second
The thickness of the wall was determined by considering three
load was the biased load acting from two directions as illustrated in
kinds of external loads expected to act on the wall.
The first is the
Figure 37(c). This kind of load is taken into account, because there
uniformly distributed load directed to the center of the circle, as
always are some unidentified sources inducing eccentric load
illustrated in Figure 37(a), which is specified from the earth
application such as sloped layers of underground deposits and
pressure at rest plus water pressure at each depth for which
non-uniformity of soil deposits. Since it is difficult, however, to
K o 0.5 was assumed. It is to be noticed that the effect of this
estimate its magnitude, it was assumed to take a value 0.2 times the
uniform pressure is exactly the same as that of the equivalent
magnitude of the uniform load.
uniform pressure acting from the four directions as shown in Figure
Figure 34
The location of the Trans Tokyo Bay Tunnel and LNG tanks
26
Figure 35
Fig. 36
Underground LNG tank at Sodegaura, Chiba, Japan
Underground LNG storage tank at Sodegaura (Goto-Iguro, 1989)
27
Figure 37
Three types of external loads in the design of the diaphragm wall
Figure 38
Stability against bottom uplift
28
When the effects of earthquake loading are to be considered, the
of excavation as shown in Figure 38.
load from one direction is assumed to act as shown in Figure 37(d).
stability check is generally made by comparing the total weight of
Considering this stage,
In the present case of the LNG tank, the magnitude of the
soil-water mass in the portion AABB with the uplift water
earthquake force was assumed to be 0.15 times the uniform load.
pressure at the bottom which is equal to w (D / 2)
As a result of the load application as mentioned above, the
usual to control the water table level inside or outside of the
thickness of the wall was determined to be 1.2m, although there are
excavation so that the factor of safety of at least about 1.03m can
some differences in the reinforcement in the direction of the depth
be maintained. The stability check for seepage-induced piping may
depending upon the computed bending moment in the body of the
need to be made by comparing the hydraulic gradient against its
diaphragm wall.
critical value. Since it is difficult, however, to evaluate the critical
. It is
value, the general practice is to design the diaphragm wall so that

7.2.
Stability Check For The Uplift And Seepage
its toe can penetrate deep enough through an impervious layer with
sufficient thickness.
In carrying out excavation in the dry condition, it is a common

practice to lower the ground water table inside to a level 1-2m
below the bottom of the excavation.
7.3.
Method Of Excavating Ditches
The ground water table
below the impervious layer is lowered outside the diaphragm wall
While there are several methods which are different in details and
to a degree necessary for reducing the uplift water pressure acting
in size, the method for excavating ditches for the diaphragm wall
at the bottom.
can be roughly classified into two groups as described below.
Anyway, the most critical time is at the final stage
Figure 39
Diaphragm wall by rebar lapping method
29
Figure 40
Diaphragm wall by concrete cutting method
7.4.
Rebar Lapping Method
The steps for constructing the wall are illustrated in Figure 39.
First of all, rectangular ditches are excavated by a cutting machine
for the A-element (advance element) at a certain interval as shown
in Step 1.
Drilling mud is filled in the ditch at this stage.
Then,
reinforcement cage (rebar cage) franked by two long steel plates is

lowered into the mud-filled ditch and concrete is cast by tremie
pipe.
At this time, the steel plate tends to be deformed outward
due to the weight of un-cemented concrete.
To prevent this, it is
common to throw gravels or crushed stones into the space outside

the steel plates as illustrated by Step 2 in Figure 39.
Then, the
excavation is made for the element B (follow element).
In this
case, both ends of the Section A are still filled with gravels, but
gravels are excavated together and the bottom is cleaned up as
shown in Step 3. Since the concrete at the element A has already
been solidified, there is no extra pressure acting on the steel plate to
induce its deformation. After the excavation, the rebar cage is
lowered into the mud and concrete is poured as illustrated in Step 4.
Figure 41
Cutting machine for the ditch
Note that the rebar in Section B is assembled so that it can fit well
30
at both ends to the already existing reinforcements projecting from
Figure 49.
the side of element A.
constructing the diaphragm wall used for temporary retaining
When this method is adopted, the cutting of the ditch is made
structure, and also for the case of circular walls where the hoop
generally by using the machine as shown in Figure 41.
compression is expected to induce high compressive loads in the
The axis
of the cutting bit is arranged vertically and by rotating the bit

horizontally, the excavation can be advanced at the bottom.
Since
the connection between element A and B is very tight, this method

is adopted for constructing diaphragm to be used as permanent
structure.
This concrete cutting method is adopted for
circumferential direction.
In the LNG storage tank, the rebar
lapping method was used, because the diaphragm wall was

intended to function as permanent structure.
No matter whichever
method is adopted, the same method can be applied for the

circular-shaped wall construction as well.
It is to be noted that
what is called circular wall is not exactly the round circle. It has a
multi-angular shape.
7.5
Concrete Cutting Method

7.6
Observed Performances
As illustrated in Step 1 in Figure 40, the ditch excavation is carried

out at element A at a certain interval.
Then, the rebar cage
The measurements were made of the earth pressure and lateral wall
without the steel plates is lowered and concrete is poured as shown
deflection at several points around the circular diaphragm wall after
in Step 2.
When the B-element is excavated in the next stage, the
the construction of the LNG tank at Sodegaura site. Shown by the
cutting is performed not only for the portion of soils but also for the
solid line in Figure 42 is the already deflected feature of the wall
hard portion of concrete on both sides, as illustrated in Step 3.
top as computed using the uniform design load in Figure 37(b).
Then, installation of the rebar and concreting is carried out as
Although the amount of deflection is not shown explicitly, it was of
shown in Step 4.
the order of 1cm.
When using this method, the ditch is excavated
Needless to say, the deflection was uniform and,
by means of the machine having the horizontal two-axes of cutting
hence, the solid line is a true circle. The dotted line in Figure 42
bit. The two cutting bit is rotated horizontally in mutually opposite
indicates the deviation of computed wall deflection by considering
directions so that the bit can cut the surround soils upwards on both
the eccentric design load shown in Figure 37(c).
walls.
A typical machine is shown in the photograph later in
Figure 42
Design loads and measured wall displacements in radial direction, LNG storage tank at Sodegaura
(Goto-Iguro, 1989)
31
The dotted line in Figure 42 indicates the deviation of computed
facility for ventilation when the highway traffic is in service.
wall deflection by considering the eccentric design load shown in
Since the Island was to be located in the middle of the sea and
Figure 37(c).
seated in the soft soil seabed at a water depth of 28m, the greatest
oval.
It is apparent that the deflected shape is slightly
The chain-dotted line in Figure 42 shows the lateral
deflection of the wall top
actuallly measured at the time when the
excavation had been completed to the targeted depth.
It may be
challenge was posed for its safe construction.
A series of steps
taken for constructing the diaphragm wall and some difficulties

encountered will be described below somewhat in details.
seen in Figure 42 that the measured distribution of the wall lies

within the two lines of deflection computed using the design loads
8.2.
Soil Conditions In Seabed
indicated in the right hand of the figure.

The soil conditions at the site were investigated by two deep
8.0
LARGE DIAPHRAGM WALL AT KAWASAKI
ISLAND IN TRANS TOKYO-BAY UNDERSEA TUNNEL
borings drilled to a depth of 135m from the sea level.
The
outcome of this investigation is displayed in Figure 44. It can be

seen that there exists a soft silty clay deposit (AC1) of alluvial
8.1
General
origin with the SPT N=0 to a depth of about 22m from the sea
The Trans Tokyo-Bay tunnel connecting Kawasaki city on the west
bottom which is underlaid by a series of alternate layers of sand
and Kisarazu city on the east was constructed during the period of
and silt of diluvial origin down further to a depth of about 45m
1989-1997 to provide the 4-lane highway across the Tokyo Bay.
from the sea bottom.
The location of the highway is shown in Figure 34.
Out of the
stiff soils with N-values in excess of 50. Since the diaphragm wall
15km long highway, the eastern and the undersea tunnel with a
had to be embedded to a sound base deposit through sufficiently
diameter of 14m was constructed for the 10km long stretch on the
impervious clay or silt layer, of most critical concern was the
west as shown in Figure 43.
Below this the deposits consist mainly of
In the middle of the tunnel, a
decision as to how deep the diaphragm was to be penetrated. The
man-made island called Kawasaki Island was constructed to
presence of the impervious layers with sufficient thickness and
provide a starter base for launching the shield tunnel to the east and
stiffness is demonstrated in Figure 45.
to the west.
This island was intended to be equipped with a
Figure 43
Plan and side view of the trans Tokyo Bay tunnel
32
Figure 44
Soil profiles at the site of Kawasaki Island
33
Figure 45
Soil profile and water-bearing layers at Kawasaki Island
Figur 46
Construction process of the man made Island
34
8.3 Construction Of The Diaphragm Wall
floor at the top.
This is called outer pier and the other one inside
is called inside pier. The next step was to drive a number of steel
The sequence of the process for constructing the Island consisted of
pipes 40cm in diameter side by side in contact in a circular
several steps as illustrated in Figure 46.
alignment along the outside rim of the inner pier and also along the
In this case, the concrete
cutting method illustrated in Figure 40 was adopted.
inside rim of the outer pier. The circular arrangement of the pipes
1)
along the inside of the outer pier is shown in the plan view of
First of all, the soft clay and soft deposits under the
seabed were stabilized to a depth of 35m below the seabed by the
Figure 48(a).
method called sand compaction pile (SCP) over the circular area
driving pipes side by side, the annular space 12.6m wide between
about 120m in diameter.
The area ratio of compaction was 30%
these two walls was filled up by dumping cement-mixed soils from
The compaction pile was constructed by
the pier. It is to be noted that the seabed deposit just underneath
means of the heavy machinery equipped on board the ship. The
the soil-cement fill had been stabilized by the technique called
annual portion of the seabed deposit with an inner diameter of
Deep Mixing Method (DMM).
100m was stabilized by the Deep Mixing Method (DMM) in which
construction is illustrated in Figure 48(c).
the soft soils were churned and mixed with cement to improve
4)
these into somewhat stiff materials. The zones of SCP and DMM
advanced downwards through the 12.6m wide cement-mixed soil
are indicated in Figure 46(a).
fill and further down to a depth of 119m from the floor level of the
2)
After the work of soil stabilization was over, steel pipe
pier. Excavation was carried out by what is called the reverse
piles 1,016m in diameter were driven by the hammer-operated
circulation method using the two-axes rotating type cutting bits as
machines aboard the ship to provide the foundation for the jacket
shown in Figure 49. The width and length of one element of
structure to be put into place later on.
excavation was 2.8m and 2.35m, respectively.
and 78.5% in plan.
The jacket frame unit
After constructing two circular walls in this way by
The feature of this phase of
From the top floor of the pier platform, excavation was
Special caution
assembled at the yard at other site was carried by a tug boat and its
was taken to maintain the in-plumb condition of excavated shaft
legs were put on the top of the pile.
One of the jacket frame units
within an accuracy of 1/1000 by using a high-precision apparatus.
in the yard is shown in Figure 47. The feature of the jacket unit
The circular diaphragm wall had an outer diameter of 103.6m and a
sitting at the site of the Island is displayed in Figure 46(b). The pier
total length of 119m counted from the working platform on the pier.
thus constructed was used as the working platform for further work
Rebar cages were lowered unit by unit and concrete was poured in
of diaphragm construction and excavation.
the bored holes.
3)
Figure 46(d).
The jacket-type platform consisted of three circular
This phase of the construction is illustrated in
piers with two outer piers connected with each other by the flat
Figure 47
Unit of jacket frame sitting on the assembling yard
35
Figure 48
Layout of the platform and piers in Kawasaki Island
36
5)
After installing the circular diaphragm walls with a
critical issue in the challenge of deep excavations below the natural
width of 2.8m, sea water inside the enclosed wall was pumped out
ground water table.
and excavation was advanced to the level of -72m below the sea
level and then framed structure of the inner jacket was dismantled
level within the diaphragm wall while lowering the ground water
and removed.
level. The dewatering and control of the water table level inside
together with the cement-mixed soil to expose the inside surface of
and outside the diaphragm wall during the excavation was
the diaphragm wall.
conducted by means of two sets of deep wells as illustrated in
schematically illustrated in Figure 46(e).
Figure 50.
One is the well points to pump water up from the two
reached the level of -28m, a huge ring-shaped hoop composed of 6
water-bearing pervious layers designated as A and B in the figure.
reinforced concrete blocks 5.8m thick and 11m high was installed
The other set was installed within the excavated pit to observe the
around the inside periphery of the diaphragm wall.
level of water during excavation.
intended to assure the tightness of the junction between blocks in
The still other was to observe
the ground water level at the deepest layer C.
The levels of water
The excavation was advanced to a targeted
The circular row of the pipes was also taken away
This phase of the construction work is

When the excavation
This was
the formation of the wall.
The feature of this installation is shown
table indicated in Figure 50 are the targeted level for the final stage
by a picture in Figure 48.
The platform of the bird-eye view is
of excavation down to the floor at the depth of -72m.
shown in Figure 52.
It is to be
noticed that the design and control of the water tables are the most
Figure 49
Two-axes rotating type cutting bits
37
Figure 50
Observation and control of water levels inside and outside of the excavated pit
Figure 51
Installation of a block hoop at the depth of -28m
38
Figure 52
8.4
Bird-eye views of the platform of the piers
Performances Of The Diaphragm Wall
important factor, because any significant deviation is associated

with a risk of eccentricity of the overall behaviour.
There were two important items to be considered in the structural
2)
design of the circular diaphragm wall.
In the lower part of the wall, a great magnitude of
One was the earth pressure
compressive stress is induced in the circumferential direction.
acting on the wall and the other was the force in circumstantial
Therefore, a special concrete with high compressive strength ought
direction which acts as support to maintain the stability of the wall.
to be cast.
As there was no bracing structure inside the wall, it was considered
was specified as 36MN/m2.
of utmost importance to make sure whether the hoop compression
In order to monitor the performances of the diaphragm walls,
was uniformly transmitted in the circumferential direction, not
several set of strain gages were attached to the rebar cages.
causing any deviation.
In the Kawasaki Island, the strength of the concrete
Also,
There are two aspects to be paid attention
the earth pressure cells and tilt meters were installed in the wall
in discussing design and performance of the circular diaphragm
before concrete was cast. Figure 53 shows the circumferential
wall.
compressive stress measured in the walls at three stages of
1)
In the upper part of the wall, distribution of the hoop
excavation
load, earth pressure and wall deflection around the circle is an

.
39
Figure 53
Figure 54
Distribution of the circumferential load intensity versus depth
Measured earth pressures at two locations at the stage of excavation to the depth of 28m
40
Figure 55
Measured earth pressure at two locations around the diaphragm wall
When the inside excavation reached the level of -28m, a huge
along the periphery of the diaphragm wall due to eccentricity of the
ring-shaped hoop was installed as indicated in Figure 53(a).
load distribution.
At
At greater depths, the compressive stress itself
this stage, the hoop load monitored at three locations was
is an important factor.
distributed versus depth as shown in Figure 53(a) with its
lower than the strength of 36MN/m2 which was achieved with the
maximum value of about 20MN/m occurring at the depth of 80m.
use of high-strength concrete.
As the thickness of the wall was 2.8m, this maximum corresponds
At the time the design was made, analyses were made to evaluate
to 20/2.8=7.14MN/m in terms of load intensity.
Shown in Figure
The measured stress was shown to be
the compressive stress in the hoop direction.
The results of the
53 (b) is the depthwise distribution of the circumferential load per
analyses are also indicated in Figure 53.
1m depth at the stage when
analysis results, the design load was specified reflecting the records
level of 40.4m.
the excavation had progressed to the
It is noted that the maximum
load of about
On the basis of the
of previous cases of diaphragm construction.
The load
38MN/m was monitored at the floor level of excavation, viz., at the
distribution thus determined for the design of Kawasaki Island is
depth of 40.4m.
also indicated in Figure 53.
The distribution of the circumferential load per
In view of the good coincidence with
1m depth at the near-final stage of 69.7m deep excavation is
the analysis and design values, actual progress of the excavation
demonstrated in Figure 53(c), where it is noted that the maximum
was considered satisfactory and to have been executed successfully.
load was about 40MN/m occurring at the bottom level of
The external force acting on the diaphragm wall was the earth
excavation.
pressure and water pressure from inside and outside as well of the
This corresponds to a compressive load intensity of

2
40/2.8=14.3MN/m =143kg/cm which is apparently smaller than

2
wall.
The earth pressure was measured by means of the pressure
the compressive strength of the concrete, 36MN/m , used in the
cells buried at various depths at four locations around the wall.
design of the diaphragm wall.
One of the features to be noticed is
The results of measurements are shown in Fig. 54 where it may
the variation of the hoop load amongst three points of
seen that the earth pressure acting on the outer face of the wall
measurements along the periphery, particularly at shallow depths.
(Figure 54(b)) was, by and large, smaller than that evaluated by the
It is noted in Figure 53 that the variation was not so significant,
theory assuming Ko=0.5. The earth pressure on the inside of the
indicating no likelihood of buckling-type deformation to take place
wall was somewhat scattered and took values in excess of those
41
evaluated by theory at two depths.
However, in the deeply
for tight connection at joints between two neighbouring elements.
embedded portion, the larger earth pressure tends to act to make the
The structural design is made to resist against the earth pressure at
wall safer against the heaving of the excavated bottom, and
rest plus water pressure acting from behind.
therefore, it was considered not to cause any serious problem.
always necessary in this case.
Distribution of measured horizontal inward displacements is shown
When the diaphragm wall is utilized in the form of circular-shaped
in Figure 55, along with the values corrected for the effects of
retaining walls, more important is the hoop compression in the
temperature.
It can be seen that, at the stage of excavation to the
circumferential direction and eccentricity in the radial deflection.
depth of 69.7m, the maximum wall deflection was of the order of
Since the hoop compression tends to increase with depth, it is a
2cm and that occurred at the bottom of the excavation.
common practice to use high-strength concrete at deeper portions.
The
Bracing system is
displacement of this order of magnitude was judged as acceptable
In this paper, the case of the LNG storage tank in Tokyo Bay was
for safe performance of the diaphragm wall.
taken up to highlight issues of importance as above.
The second
example is the construction of the Kawasaki Island in Tokyo Bay.

9
CONCLUDING REMARKS
This is unique in the sense that it has the largest-ever depth of

embedment and that it was constructed in the middle of the waters.
In a majority of cases of foundation design in Japan, it is possible
In terms of the scale and the hurdles encountered, it may be cited as
to locate a stiff layer in the deposits of dense sand, gravel or soft
one of the most difficult unprecedented challenge in the recent
rocks at depths 20-50m where a sufficient bearing capacity can be
history of civil engineering practice.
obtained.
In view of a certain degree of uncertainty in assessing
the skin friction, the piles are generally designed as the end bearing
ACKNOWLEDGEMENTS
pile for which more reliance can be put for supporting

superstructures. Under such circumstances, verification of the load
In preparing this text, information on the pile testing in Japan was
carrying capacity of a given stiff layer as to whether it can indeed
given by Dr. H. Ogura, Technical director of Japan Pile Co. The test
mobilize the design-assumed bearing capacity has been the major
data on the UOB plaza and the prototype tests in Tsukimino were
motivation to perform the in-situ pile loading tests.
offered by Dr. T. Hosoi, former executive director of Nishimatsu
As a consequence of a majority of the loading tests over the last
Co. and by Mr. Y. Ueda at Singapore office of Nishimatsu Co.
two decades, fairly reliable sets of data have been obtained and
The information on the diaphragm construction at Kawasaki Island
accumulated. These are put forward in the manual published by the
was offered by Kajima Co. and also by Dr. H. Yoshida,
Japan Association of Architecture for the design and construction
Vice-president of Chemical Grout Co. in Tokyo. Dr. T. Sueoka,
of buildings having 20 to 50 storeys.
director of engineering center of Taisei Co. also provided
For large structures such as bridges or dams, caissons or diaphragm
information on construction of the Kawasaki Island.
walls are often used as foundations to support superstructures.
In
former executive director of Tokyo Gas Co., kindly provided
such cases, loading tests are generally carried out within excavated
detailed account of the construction of the LGN storage tanks in
shafts or pits by applying loads directly to the intact surface of stiff
Sodegaura. The author wishes to extend his sincere thanks to these
soil or rocks. When the bearing layer is too deep or too difficult to
experts mentioned above who furnished the most valuable
designate such as the case of UOB building in Singapore, friction
information without which the writing of this text would not have
piles are planned and implemented, but verification of the bearing
been materialized.
capacity by means of the in-situ tests become prerequisite.
Dr. S. Goto,
As a
This paper was originally submitted and printed in the Proceedings
result of the loading tests, the pile length was reduced and this
of the Bangladesh Geotechnical Conference (BGC-2009) held in
contributed greatly to the saving of the cost.
Dhaka in December 2009. The author wishes to express his deep
Diaphragm walls have been used for various purposes as temporary
thanks to Professor A.M.M. Safiullah for giving his permission to
supports or as a part of main permanent structures.
publish this paper again in this journal in its present form.
It is utilized to
retain the soil ground for executing excavation to construct

basements of buildings, subway stations, storage tanks etc.
When
the diaphragm wall is constructed in the form of straight line in

plan for the purpose of water-proofness, precaution should be taken
42
Application of the Results of Pile Toe Loading Tests on
REFERENCES
Evaluation of Behaviour of Cast-in-Place Concrete and Pre-Cast

Goto, S. and Takahashi, Y. (1981), Construction of Large LNG
Piles, Tsuchi to Kiso, Japanese Geotechnical Society,
Inground Storage Tanks at Sodegaura Terminal, Tsuchi to Kiso,
No. 5, pp.84-89. (in Japanese)
Vol. 29, No. 1, pp. 24-28. (in Japanese)
Ogura, H., Shibuya, T., Karkee, M.B. and Saito, M. (1997), Toe
Goto, S. and Iguro, M. (1989), The Worlds First High-strength,

th
Super-Deep Slurry Wall, Proc. of the 12
Vol. 43,
Load Test on Large Diameter Cast-in-Place Pile for the New
International
Building of Kanto Regional Postal Services Bureau, Part
Conference on Soil Mechanics and Foundation Engineering, Rio de
2-Analysis and Interpretation of Results, Proc. 32nd Annual
Janeiro, Vol. 2, pp. 1487-1490.
Convention of the Japanese Geotechnical Society, Saitama City, pp.
Han K.K., Wong K.S. and Broms, B.B. (1993), Singapore
1433-1434. (in Japanese)
Bouldery Clay: The Origin, Properties and Load Tests of Bored and
Ogura, H. and Shibuya (1998), Toe Load Test on Large Diameter
Cast-in-Place Piles, Proc. 11th Southeast Asian Geotechnical
Cast-in-Place Piles, Kisoko, pp. 84-89. (in Japanese)
Conference, Singapore, pp. 523-527.
Ogura, H., Kuwayama, S., Suzuki, Y. and Yamamori, S. (2005),
Ho, C.E. and Wallace, J.C. (1993), Design and Performance of
Toe Bearing Resistance Test and Shaft Resistance Test on Large
Diaphragm Walls for a Deep Basement in Singapore, Proc. 11th
Diameter Bored Precast Pile, Proc. Annual Convention of
Southeast Asian Geotechnical Conference, Singapore, pp. 715-720.
Japanese Geotechnical Society. (in Japanese)
Iwanaga, K., Hirano, T., Ishii, T., and Ueno, T. (1991),
Photo Album Tokyo Wan Aqua-Line (1998), Japan Highway
Considerations for Earth Pressure on Shaft and Pile Loading Test
Public Corporation, and Trans-Tokyo Bay Highway Corporation.
Rearranging, Nishimatsu Technical Research Report, Vol. 14, pp.
Shibuya, T., Ogura, H., Kawamura, A. and Senno, H. (1997), Toe
108-116. (in Japanese)
Load Test on Large Diameter Cast-in-place Pile for the New
Japan Society of Architecture and Buildings, (2001), Manual of
Building of Kanto Regional Postal Services Bereau, Part 1 Load
Recommendations for Design of Building Foundations, pp.
Test Plan and the Results. Proc. 32nd Annual Convention of the
211-216.
Japanese Geotechnical Society, Saitama City, pp. 1431-1432. (in
Kishida, H. and Tsubakihara, Y. (1993), An Analytical Method for
Japanese)
Predicting the Displacement of a Pile and Soil Layers, Proc, 11th
Imamura, M., Hosoi, T., Kado, Y. and Ishii, T. (1991),
Southeast Asian Geotechnical Conference, Singapore.
Construction of Deep and Large-Diameter Caisson Type Piles,
Osterberg, J.O. (1984), A New Simplified Method for Load
Journal of Japan Society of Civil Engineers, Vol.-14, No. 427, pp.
Testing Drilled Shafts, ADSC, pp. 9-11.
289-298.
Ogura, H., Sugai, K., Kishida, H. and Yoshifuku, T. (1995),
43

Tech5-4 Pozo Profundo PDF

Uploaded by

Copyright:

Available Formats

Tech5-4 Pozo Profundo PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Tech5-4 Pozo Profundo PDF

Uploaded by

Copyright:

Available Formats

Geotechnical Engineering Journal of the SEAGS & AGSSEA Vol. 41 No.

3 September 2010 ISSN 0046-5828

RECENT ADVANCES IN PILE TESTING AND DIAPHRAGM WALL

whether a given deposit is competent enough to mobilize the

Some of the results of in-situ pile

In the design and practice of foundations supporting heavy-weight

loading tests recently obtained in Japan by this method are

structures such as high-rise buildings or large-scale facilities, of

introduced in this paper.

utmost importance is to know the load-carrying characteristics of

In the case of the deep-seated foundation supported by the

the load-bearing capacity for piles embedded into such deep-seated

methods need to be explored. In important projects, efforts have

deposits have been made generally by carrying out what is known

been made to measure directly the bearing power or skin friction

as the O-cell test or pile toe loading test.

In some cases, direct

within excavated pits or shafts by means of the plate loading tests

loading tests are performed by large concrete blocks or segments

performed on exposed surface of intact stiff sand, gravel and rocks.

As one of the attempts in this context, an example of in-situ loading

excavation of pits or shafts for foundation construction.

tests conducted in Singapore is introduced in this paper.

outcome of these in-situ tests has not unfortunately been publicized

The other topics of importance in the foundation engineering

and compiled to the extent that these data can be effectively

would be the design and construction of the large-scale diaphragm

wall, particularly when it is used as circular walls without any

collected and presented in this paper.

distribution in the circumferential direction are important factors to

In the roundshaped walls, the hoop compression and its

be considered with utmost precaution.

The method of design and

of industry and mega city in Japan, is generally composed of soft

results of monitoring in this regard will be introduced by quoting

soil of alluvial origin underlaid by alternative layers of stiff sand,

an example of constructing a large LNG storage tank in Tokyo.

clay, sand and gravel of diluvial origin.

It is the generally adopted

a bearing layer to support the piles.

due to high hydraulic gradient of seepage water.

Thus, the end bearing type

These issues will

be briefly touched on by quoting an example of the large scale

However, since the stiff bearing layer is seated at great depths of

construction of Kawasaki man-made island in the middle of Tokyo

Accumulation of in-situ data on pile testing is so much both in

internally by oil pressure, creating an upward force on the shaft and

simultaneously an equal magnitude of force downwards to the

come up with a state-of-the-art text from a fair point of view.

bottom of the shaft.

What is written here is part of the test data which is considered

power of the pile toe is resisted by the upward force mobilized by

appropriate in the context of the present presentation.

the friction on the surface of the pile body.

The downward force to monitor bearing

also many construction examples of circular diaphragm walls.

possible to monitor the shaft friction as well while the reaction is

Because of the limitation in several respects, it was not possible to