Underground Singapore 2014
Deformation Modulus of Singapore Soil and Rock
Determined by Various Field and Laboratory Testing
T. Nonaka, T. Yasuda, K. Kimura, A. Hatayama, Rudi Julijanto & Tsen Soo Fui,
Kiso-Jiban Consultants Co. Ltd., Singapore Branch
K. Orihara
Kiso-Jiban Consultants Co. Ltd., Overseas Department
ABSTRACT: This paper describes deformation modulus of Marine Clay, Old Alluvium and Jurong
Formation in Singapore, determined by various field and laboratory tests. The field testing methods
include downhole seismic test and cyclic pressuremeter test carried out in boreholes. Two (2) methods
of downhole seismic testing employed are one with vibration source on the ground level and suspen-
sion method in which the vibration source is inserted in the borehole. Undisturbed samples were col-
lected from the same boreholes where the field testing was carried out. Deformation modulus of soil
sample is determined by CK0U triaxial compression test at small strain by using Hall-effect local
strain transducer and by using conventional vertical displacement transducer. Bender Element test
was also conducted on the undisturbed soil sample. The determined deformation moduli are compared
with the undrained shear strengths and SPT N-values. The deformation moduli of rocks of Jurong are
measured by suspension seismic test and by strain gauge placed on an intact rock core during uncon-
fined compression testing. The paper describes the methods used for determining the deformation
modulus. It discusses measured deformation moduli by various methods and application to the design
deformation modulus.
1 INTRODUCTION
Deformation modulus of soil and rock is one of fundamental parameters in geotechnical engineering.
Several field and laboratory test method are available to measure the deformation modulus of soil and
rock. However, the direct measurement is rarely used in determination of design modulus in Singa-
pore It may be because there are too many methods to provide a wide range of deformation modulus
and a design method has not yet been specified in the local code. It may be because the deformation
modulus is highly affected by quality of testing and the good deformation modulus can be hardly ob-
tained. In this paper, deformation modulus of soils of the Kallang Formation and the Old Alluvium
and rocks of the Jurong Formation by various field and laboratory tests, listed in Table 1, are com-
pared and also with the design modulus locally used.
Table 1. Field and laboratory tests adopted to determine deformation modulus for each soil/rock
Geological Soil/rock type Field Test Laboratory Test
Formation
Kallang Marine Clay Downhole and Suspension Bender Element
Formation PS logging CK0U triaxial compression test
Prebored Pressuremeter test with Hall-effect local strain
transducer
Old Sandy Soils with SPT Downhole and Suspension Bender Element
Alluvium N of 10 to 100 PS logging CK0U triaxial compression test
Prebored Pressuremeter test with Hall-effect local strain
transducer
Jurong Sedimentary rocks Downhole and suspension Uniaxial compression test with
Formation (Grade I to II) PS logging strain gauges
For the Kallang Formation and the old Alluvium, one (1) borehole was drilled to obtain undisturbed
soil samples by a hydraulic Piston sampler for Marine Clays and a Mazier sampler for the Old Alluvi-
um. At each undisturbed sampling depth, a prebored pressure meter test was performed. After comple-
tion of the drilling, PS logging by Downhole method and Suspension method was conducted through
PVC access pipe installed in the borehole. On the obtained undisturbed soil samples, K0 consolidated
undrained triaxial compression tests (CK0UTC) with Bender Element Test and local strain measure-
ment by Hall effect transducer were carried out.
For rocks of the Jurong Formation, rock coring and Suspension PS logging were performed in the
several boreholes. Uniaxial compression test with strain measurement by strain gauges were conduct-
ed for the selected rock core samples of Limestone. Siltstone. Tuff and Tuffaceous sandstone with
Weathering Grades I to II.
2 METHOD OF MEASUREMENTS
2.1 Seismic Wave Test
Two (2) types of seismic wave tests were carried out in this study, they are:
Downhole PS Logging Test
Suspension PS Logging Test
Seismic wave method measures the velocity of the compressional (P-wave) and shear (S-wave) prop-
agate through the soil or rock at the immediate vicinity of the borehole. The elastic properties (Pois-
son’s ratio, Young’s modulus) of the layer can be computed.
2.1.1 Downhole PS Logging Test
In Downhole PS Logging method, the receiver contains three component geophones (2 horizontal and
one vertical) firmly clamped to the borehole wall at every 1m depth intervals. Surface hammering is
usually used as the seismic source. Two shear wave (S-wave) records are obtained by striking the
plank horizontally in opposite directions. Thus, the shear wave records obtained have reversed polari-
ty. The compression wave (P-wave) record is obtained by measuring the P-wave generated by drop-
ping a weight on the ground surface. Figure 1 shows a schematic of the Downhole PS Logging meth-
od.
Figure 1 Schematic of Downhole PS Logging Test Figure 2 Schematic of Suspension PS Logging
2.1.2 Suspension PS Logging Test
Suspension PS Logging uses a downhole probe containing both a seismic source and receivers. The
logging system consists of a seismic source and two receivers built into a single probe. The source
generates a pressure wave in the borehole fluid, therefore it must be carried out under the groundwater.
At the borehole wall, this pressure wave is converted to seismic body waves (P and S), which travel
radially from the borehole wall. The seismic waves are detected by the geophones at each sensor loca-
tion. The seismic wave velocities are calculated from the difference between the arrival times record-
ed by the two receivers spaced 1m apart.
This method is intended for use in uncased boreholes filled with drilling mud or water. However, it
can be used also in cased boreholes if the coupling between casing and surrounding soil or rock is
good. Steel casing often creates problems because large amplitude, high velocity tube waves arrive
before the seismic waves. The tube wave problem is smaller with plastic (PVC) casing. Figure 2
shows a schematic of the Suspension PS Logging method.
2.2 Pressuremeter Test
The prebored pressuremeter tests were carried out using Single-cell Menard Presssuremeter Type.
In prebored single-cell pressuremeter test, uniform pressure is applied on the wall of prebored hole via
a flexible membrane as shown in Figure 3. Water is used as medium to inflate the flexible membrane.
The equipment consists of Cylinder probe with a flexible membrane, Measuring devices, i.e., volume
changes in measuring cell and pressure readout and Tubing connecting the probe with measuring de-
vice. In the tests, two (2) cycle loading was conducted to obtain unload-reload modulus Er as well as
Pressuremeter modulus Em. Some of the pressuremeter tests were performed using OYO Elastmeter.
1. Volumeter
2. Standpipe
3. Range Switching Valve Pressure Control and
4. Pressure Gauge for Measuring Cell Measuring Unit
5. Pressure Gauge for Guard Cell
6. Regulator
7. Automatic Regulator
8. Frame for Pressure gauges
9. Double Tubing
10. Probe
Figure 3 Schematic Diagram of Pressuremeter Test Figure 4 Hall effect local strain transducer
2.3 . CKoU Triaxial
CK0U triaxial compression test was conducted using an automated GDS traiaxial test system. Hall ef-
fect local strain transducers (one pear for vertical and one for radial) were put on a specimen as shown
in Figure 4. K0 consolidation was carried out under PC feedback control of stress and displacement of
the specimen. After completion of K0 consolidation, the bender element test was conducted. Then the
specimen was sheared with 0.03%/min axial strain rate.
2.4 Bender Element Test
After completion of consolidation stage of CK0U Triaxial Test, P and S-waves were measured by
bender element test. Bender element is mounted in a top cap and a pedestal as shown in Figure 5.
Figure 5 Bender element installed in top-cap Figure 6 Strain gauges installed on rock core
2.5 Uniaxial compression test on rock core, by strain gauge on intact rock
Uniaxial compression tests (UCTs) on rock cores were conducted in accordance with ASTM D7012-
07. A total of four (4) pieces of strain gauges (2 for axial stain and 2 for lateral strain) were installed
on the rock core as shown in Figure 6.
3 DEFORMATION MODULUS OF MARINE CLAY AND OLD ALLUVIUM
3.1 Deformation modulus of Marine Clay
Figure 7 presents deformation modulus obtained by pressuremeter test and CK0UTC versus un-
drained shear strength obtained by CK0UTC. The deformation modulus for CK0UTC is E50 based on
stress-strain curve obtained by the Hall effect local strain transducer. As shown in the figure, the re-
sults of CK0UTC agrees with E=250 to 400cu that commonly adopted in geotechnical design in Sin-
gapore. While deformation modulus obtained by the pressuremeter tests are much smaller and fall on
E=10 to 50cu. This could be due to disturbance of the test hole in predrilling stage. The design defor-
mation modulus is obtained in Specification for Highway Bridge by Japan Road Association to multi-
ply the pressuremeter deformation modulus Em by 4 after taking the borehole disturbance into ac-
count. However, the Em after multiplied by 4 is still very low as compared to E50.
Figure 8 presents-stress strain curves obtained by the conventional external linear transducer and the
Hall effect local strain transducer. As shown in the figure, difference between the stress-strain curves
around 50% of the peak strength is insignificant thus, in terms of E50 measurement, difference be-
tween the conventional external linear transducer and the Hall effect local strain transducer is insig-
nificant.
CK0UTC
(Marine Clay)
Figure 7 E versus cu for MC Figure 8 Comparison of External and Internal transducer
3.2 Deformation modulus of Old Alluvium
Figure 9 presents deformation modulus obtained by prebored pressuremeter test and CK0UTC versus
SPT N-value. The deformation modulus for CK0UTC is E50 based on stress-strain curve obtained by
the Hall effect local strain transducer. Orihara and Khoo (1988) reported Em=1.0N (MPa) and
Er=2.0N (MPa). Li and Wong (2001) reported Em=0.74N (MPa) and Er=3.72N. As shown in the figure,
the pressuremeter test results show the similar tendency. However, modulus obtained by CK0UTC are
much smaller than the pressuremeter tests. Figure 10 presents deformation modulus obtained by pres-
suremeter test and CK0UTC versus undrained shear strength cu obtained by CK0UTC. The modulus
by CK0UTC is E50 based on stress-strain curve obtained by the Hall effect local strain transducer. E50
of CK0UTC is about 20 to 50 times of cu obtained by CK0UTC. While deformation modulus by the
pressuremeter fall on Er=200 to 1300cu. Ratios of Er to Em range mostly from 2 to 5.
Figure 9 E versus SPT N-value for OA Figure 10 E versus cu for OA
3.3 Dynamic deformation modulus based on Seismic Wave Velocities
Figure 11 presents deformation modulus based on seismic wave velocities measured by Downhole
and Suspension PS logging and laboratory bender element test for Marine Clay and Old Alluvium.
Figure 11 Deformation modulus based on seismic velocities of Marine Clay and Old Alluvium
As shown in the figure, the seismic velocities and deformation modulus obtained by the filed and la-
boratory tests are well agreed.
Figure 12 presents the dynamic deformation modulus together with the deformation modulus obtained
by the pressuremeter tests and CK0UTC. As shown in the figure, the dynamic deformation modulus is
about one order larger than the deformation modulus obtained by the pressuremeter tests and
CK0UTC.
Figure 12 Deformation modulus determined by field and laboratory tests
4 DEFORMATION MODULUS OF ROCKS OF JURONG FORMATION
4.1 Deformation modulus determined by conventional external linear transducer and strain gauges
Figure 13 presents a typical result of the uniaxial compression test (UCT) with strain gauge measure-
ment. As shown in the figure, axial strain measured by strain gauge is much smaller than one meas-
ured by conventional external linear transducer. Figure 14 presents comparison of deformation modu-
lus determined by the strain gauge and the conventional linear transducer. As shown in the figure,
deformation modulus determined by the strain gauge is about 2 to 4 times higher than ones deter-
mined by the conventional linear transducer. This could be due to bedding error and/or system com-
pliance such as rigidity of loading system.
Figure 13 A typical stress-strain curve of UCT Figure 14 Comparison of deformation modulus de-
termined by strain gauge and conventional external lin-
ear transducer
4.2 Deformation modulus versus uniaxial compressive strength
Figure 15 presents deformation modulus determined by UCT with strain gauges and uniaxial com-
pressive strength (UCS). As shown in the figure, the results are scattered with absence of a pattern
for each weathering grade and rock type.
Figure 15 Comparison of deformation modulus determined by strain gauge and conventional external lin-
ear transducer
4.3 Deformation modulus determined based on Suspension PS logging and UCT with strain gauges
Figure 16 shows comparison of deformation modulus determined by UCT with strain gauge and sus-
pension PS logging. As shown in figure, deformation modulus determined by UCT with strain gauge
is about 2 to 5 times higher than ones determined by suspension PS logging. This could be because
deformation modulus determined by Suspension PS logging incudes discontinuity in rocks while UCT
does not since UCT is performed on Intact Rock.
Figure 16 Comparison of deformation modulus determined by UCT with strain gauge and Suspension PS
logging
4.4 Deformation modulus determined based on GSI
Deformation modulus of rock mass at each UCT sample depth are predicted using Geological
Strength Index (GSI) based on detailed rock core logging and UCT results. GSI was determined for
every 1 m core length based on detailed observation of high quality rock cores with almost 100% re-
covery and core images obtained by borehole televiewer in this study. Figures 17, 18 and 19 preset
histograms of the deformation modulus based on GSI, Suspension PS logging and UCT with strain
gauges.
Figure 17 Histograms of deformation modulus based on GSI
Figure 18 Histograms of deformation modulus based on Suspension PS logging
Figure 19 Histograms of deformation modulus based on UCT with strain gauges
GSI has considered weathering and discontinuities in rock mass. Velocity obtained in PS logging is
also considered to reflect the same. As shown in Figures 17 and 18, although modulus based on GSI
show wider distribution than dynamic modulus by PS logging, the both modulus are similar in mode
value. There is no distinct difference in deformation modulus among rock types. Deformation modu-
lus obtained in UCT on intact rock core is higher than the modulus based on GSI and PS logging.
5 FINDINGS AND APPLICATION
- For Marine Clay and Old Alluvium
o Results of field PS logging (Downhole and Suspension methods) and Bender Element
Test on K0 consolidated specimen agree well each other for Marine Clay and OA.
o Deformation modulus determined by conventional external linear transducer and Hall-
effect local strain transducer in CK0UTC are similar.
- For rocks of Jurong Formation
o Deformation modulus determined by the strain gauge is about 2 to 4 times higher than
ones determined by the conventional linear transducer in UCT.
o Deformation modulus determined by UCT with strain gauge is about 2 to 5 times higher
than ones determined by suspension PS logging. This could be because deformation
modulus determined by Suspension PS logging incudes discontinuity of rocks while
UCT does not since UCT is performed on intact rock core.
o The mode value of dynamic modulus by PS logging is similar to one of modulus deter-
mined based on GSI.
- Application to Design Deformation Modulus
o Marine Clay design modulus may be determined by E50 by CK0UTC. Deformation mod-
ulus by prebored pressuremeter tests underestimate design deformation modulus
o Old Alluvium design modulus may be determined by a reloaded modulus Er by prebored
pressuremeter test. E50 by CK0UTC underestimates deformation modulus.
o Grade I and II of Jurong Formation design modulus may be related to dynamic modulus
by PS logging as velocity by PS logging reflects weathering and discontinuity in rock
mass. Further study is required to conclude the relationship between design modulus
and dynamic modulus by PS logging.
6 REFERENCES
Li, W.W. and Wong, K. S. 2001. Geotechnical Properties of Old Alluvium in Singapore, Journal of The Institu-
tion of Engineers, Singapore, Vol. 41, No. 3, 2001, Singapore.
Orihara, K. and Khoo, K. S. 1998. Engineering Properties of Old Alluvium in Singapore and Its Parameters for
Bored Pile and Excavation Design, Proc. 13th Southeast Asian Geotechnical Conference, Taiwan, pp 545-
550.
Tan T. S., Phoon K. K., Lee F. H., Tanaka H., Locat J., Chong P. T., 2003. A characterisation study of Singapore
Lower Marine Clay, Characterisation of Engineering Properties of Natural Soils, Swets & Zeitlinger, Lisse.,
Netherland, pp. 429-454
Japan Road Association, Specification for Highway Bridge Part 4 Substructures, 2002
