IGC 2009, Guntur, INDIA Stress Measurement in Rock Mass

STRESS MEASUREMENT IN ROCK MASS

D.V. Sarwade
Research Officer, CSMRS, New Delhi–110 016, India. E-mail: sarwadedv@rediffmail.com
K.K. Mishra
Research Assistant, CSMRS, New Delhi–110 016, India. E-mail: kkmisra_csmrs@rediffmail.com
V.K. Kapoor
Asst. Research Officer, CSMRS, New Delhi–110 016, India. E-mail: vkpr_csmrs@yahoo.co.in
Nripendra Kumar
Chief Research Officer, CSMRS, New Delhi–110 016, India. E-mail: nipendra@nic.in

ABSTRACT: Understanding and estimating the rock stress state has become increasingly important as nowadays numerical
models are mainly used as rock engineering design tool. The objective in determining rock stresses may, in many cases, be
seen as an interactive process. The type of data needed may change, depending on at what stage an underground project is in.
Further more, preliminary measurements may reveal information on the stress state that will be useful as the project advances.
Central Soil and Materials Research Station (CSMRS) is a premier institution in the investigation field of rock, soil and concrete
properties. In situ tests are conducted to know the properties of rock mass. As far as the stress measurement is concerned, there are
four methods practiced at CSMRS viz; Flat Jack, USBM Gauge, CSIRO cell and Hydro-fracturing method. The choice of the
method depends on the requirement of design. The case studies are discussed in the light of experiences gained at different projects.

1. INTRODUCTION practiced method. Later on overcoring and hydraulic-fracturing

methods were developed. In the present paper; investigation
In situ stresses released during excavation for underground
procedure, merits & demerits of all the three methods are
structures can cause rock bursting, spilling, buckling or other
explained briefly. To understand the concept of each method,
ground control problems. Knowledge of the state of in situ
a case study is also discussed.
stress is of critical importance to the design and construction
of engineering structures in a rock mass (Nripendra et al.
2004). Factors affecting the magnitude and orientations of in 2.1 Flat Jack Method
situ stress include the weight of overlying materials, geologic Flat jack method represents some of the first in situ stress
structures (on local and regional scales), tectonic forces within measuring methods available in rock mechanics. Flat jacks
the earth’s crust, residual stress and thermal stress (Hudson consist of two plates of steel that are welded together, usually
2005). The complexity of the relations between these factors of circular and square in shape. They can operate at pressures
and the in situ stress usually prohibits reliable estimation of of several thousand psi. In general, flat jack method consists
rock stress. In addition, stress cannot be measured directly and of measuring the displacement between (one or several sets
therefore rock stress determination techniques rely on the of) pins or strain gauges at the surface of an excavation created
measurement of some response viz; displacement, strain, by cutting a nearby slot. The slots can be cut by overlapping
deformation etc i.e. induced by a disturbance of the rock mass. holes or by using a large saw. A flat jack is inserted into the
This measured response of rock in a stress disturbed zone is slot, grouted in place and pressurized until the pin or strain
extrapolated, from the opening inwards, through a numerical gauge readings have returned to their original position, which is
model or analytical techniques; or measurements are made via a called as cancellation pressure. The schematic diagram of flat
drill hole that extends into an undisturbed region of the rock jack and test configuration is shown in Figure 1 (a) & (b). The
mass. cancellation pressure is used as an estimate of the stress normal
to the jack. To know the complete state of stress in the area
2. STRESS MEASUREMENT METHODS of interest, six flat jack tests need to be conducted in six
different directions.
The popular techniques practiced for stress measurement
includes Flat Jack method, Overcoring method and Hydraulic The merit of the flat jack method is that it does not require
fracturing method. All the methods are having their own knowledge of the elastic constants of the rock to determine the
merits and demerits (Fairhurst 2003). Depending upon the tangential stress at points in the walls of an excavation. On the
requirement of design, the method of investigation will be other hand, flat jack methods are limited to stress measurements
chosen. Chronologically, flat jack method is the oldest ever near the surfaces of an opening and therefore may be influenced

233
Stress Measurement in Rock Mass

by the disturbance associated with excavation of the opening, the elastic properties of the rock. Schematic diagram of the
which is a demerit. Another disadvantage of the flat jack overcoring test is shown in Figure 2. There are two types of
method is that the pressure may not be entirely transmitted overcoring methods, one using CSIRO HI cell and another
over the whole surface of the jack, in particular in the vicinity of using USBM gauge, which are discussed in the succeeding
the face welded edges. Differences between applied pressure paragraphs.
and actual overall pressure acting against the rock surfaces can
be as large as 18%. Table 1: In situ Stresses Parallel and Perpendicular to the
Drift, Rohtang Pass Tunnel Project
Test In situ Stresses
Stress
Test locations (kg/cm2)
Nos. Hor. Ver. Ratio
RD (m)
Stress Stress (K)
1&2 39.46 29.40 22.10 1.33
3&4 42.00 31.00 22.40 1.42
5&6 62.25 26.40 30.20 0.87
7&8 70.00 31.50 46.80 -
Fig. 1: Schematic Diagram of (a) Flat Jack,
(b) Test Configuration 9 & 10 72.30 38.60 31.00 1.24
11 & 12 79.75 26.10 26.10 1.00
Case Study: Rohtang Pass Tunnel Project, North Portal, 13 & 14 83.50 24.20 28.40 0.85
Himachal Pradesh
Average 29.80 26.70 1.11
The in situ rock mechanics investigation of Rohtang Pass
Tunnel Project was carried out by CSMRS in 1996. The area
2.2.1 CSIRO HI Cell
around Rohtang pass forms a part of the Higher Himalayas.
The project is situated in Rohtang Axial Zone which has a Overcoring with the CSIRO cell is generally performed at
NW-SE trend. The rock mass comprising mica schist was found depths within 30 m from working faces. The cell fits in an
to be weathered having stained joints of brown colour upto ‘EX’ diameter hole (38 mm). It is an epoxy (Araldite D type)
about 5 m. Beyond 5 m, the drift has been excavated in fresh thin-walled pipe with outer and inner diameters of 36 and 32
and hard quartz-mica schist. The strike of foliation varies mm, respectively. The cell is glued to the walls of the pilot
from N20°W-S20°E to N25°W-S25°E and dip is of the order hole using a 1 mm thick layer of glue. The epoxy and glue have
of 10° to 15° towards south western direction. a Young’s modulus and Poisson’s ratio of 3500 MPa and 0.4.
The cell contains four, three-component strain rosettes 120°
Fourteen flat jack tests were carried out in the drift at North
apart. The strain gauges are 10 mm long. Two strain gauges
Portal of Rohtang Pass Tunnel Project for determining the in
are parallel to the axis of the cell and three gauges measure
situ stresses of rock mass (CSMRS Report 1996). The average
tangential strains. The main advantage of the CSIRO cell is
vertical and horizontal stresses as determined from Flat Jack
that it can be used to determine the complete stress field in
Tests are 26.7 and 29.6 kg/cm2 respectively with a stress ratio of
one borehole only. At the same time disadvantages of the
1.11. In situ stresses measured parallel and perpendicular to
CSIRO cell are; (1) it requires long unbroken overcore (2)
the drift are tabulated in Table 1.
the cell is not recoverable until recently. Upon overcoring,
the elastic modulus of the rock is determined by biaxial
2.2 Overcoring Method testing of the recovered overcore containing the CSIRO cell.
Overcoring methods are also in practice to determine the in situ
stresses of rock mass, which involves drilling a large diameter
hole (60-150 mm) in the volume of rock, sufficiently at a
distance so that the effect of the excavation or ground surface
will be negligible. It will be followed by the small pilot hole
(usually ‘EX’ size) at the end of larger hole. The pilot and large
diameter holes must be as concentric as possible. Pilot hole
length vary between 300 and 500 mm. The large diameter hole
is resumed, partially or totally relieving stresses and strains
within the cylinder of rock that is formed. The changes in
strains or displacements are then recorded. Followed by
overcoring, the recovered overcore is often tested to determine Fig. 2: Schematic Diagram of Overcoring Test

234
 Stress Measurement in Rock Mass

Case Study: Nathpa Jhakri H.E. Project, Himachal Pradesh Table 2: Magnitude and direction of the in situ stresses,
Nathpa Jhakri H.E Project, Himachal Pradesh
The Nathpa Jhakri H.E. Project, located in Kinnaur and Shimla
districts of Himachal Pradesh, envisages harnessing of hydro Principal Magnitude Bearing Dip
power potential of river Sutlej for run-of-the river development stress (MPa) (°) (°)
scheme. The project is located in the lesser Himalaya region. Sigma 1 13.15 187 16
The main rock type found in the area was gneiss, which found
to be hard and tough. At some parts, the rock type was pre- Sigma 2 8.00 311 63
dominantly gneiss with schist bands or schist with gneiss bands. Sigma 3 4.36 90 21
Three overcoring tests using CSIRO-HI Cell were conducted Note: Final results derived after combining data of test No. 1 & 3.
to measure in situ stresses at Nathpa Jhakri H.E. Project in a
bore hole, located in the access to power house drift (CSMRS 2.2.2 USBM Gauge
Report 1994). Tests were conducted at three different depths
viz; 4.08 m, 4.57 m and 5.13 m. As the rock type was found to Overcoring with the USBM is generally performed at depths
be same at the depth 4.08 m and 5.13 m, results of these two within 30 m from working faces, in vertical boreholes.
were combined to have better understanding of stress. One Overcoring should proceed until the overcoring bit has passed
CSIRO-HI Cell contains 12 Nos. of strain gauges, while the measuring pistons for a distance of at least 150 mm.
interpreting for results 3 Nos. of strain gauges were neglected Thus, the overcore must be at least 300 mm long. However,
as to calculate six components of stress (The least square the method is difficult in jointed rocks and in fractured rocks.
method for finding out six unknown stresses from twelve A larger diameter (150 mm) hole is first drilled in the
equations and finely statistical method used to reject the strain required direction for a depth of at least one diameter of drift.
gauge on the basis of computed and measured strains have A smaller diameter hole (38 mm) is then drilled at the far end
been incorporated in computer program, as minimum nine of the borehole. A deformation gauge suitable for measuring
equations are ideally required). The findings of the test are as changes in the smaller diameter hole in three orientations is
shown in Table 2. Direction and magnitude of principal then placed inside, using special placement rods. It is then
stresses obtained from combined data of test No. 1 & 3 and overcored for a length of about 300–450 mm (approximately
test No. 2 are shown in Figures 3 (a) & (b). 2 to 3 times the diameter of the overcore hole). The change
in strain is recorded from gauge readings at every 1cm of
345 GN
15 penetration. The method does not require a dry drill hole and
330
30 can be used in holes filled with groundwater or water from
315 45
drilling. The advantages of the USBM gauge are: (1) the
300
60 gauge is recoverable (2) no cementing or gluing is required.
285 75
While the disadvantages are: (1) it requires an unbroken core
of atleast 300 mm (2) the gauge can be damaged if the core
W
E breaks (3) three non-parallel holes are necessary to calculate
255 105
the in situ stress field. Photograph of USBM drillhole
deformation gauge is shown in Figure 4.
240
120

225 135
210 150
195 165
S

3 (a)
15
Test No. 1 & 3

Test No. 2
STRESS (MPa)

10

0
SIGMA 1 SIGMA 2 SIGMA 3

3 (b)
Fig. 3: (a) & (b): Direction and Magnitude of Principal
Stresses, Nathpa Jhakri H.E. Project (H.P.) Fig. 4: Photograph of USBM Drillhole Deformation Gauge

235
Stress Measurement in Rock Mass

Case Study: Karnali (Chisapani) Multipurpose Project, Nepal pressure. The magnitude of the major secondary principal
stress can be calculated from relationships involving the fracture
The in situ test measurements by overcoring method using the
USBM gauge were carried out at Karnali (Chisapani) Multi- re-opening pressure and the shut-in pressure. An impression
purpose Project. The project envisages construction of a high packer together with a compass or a borehole scanner can be
rockfill dam located in the Karnali Gorge. The design included used to determine the orientation of the fracture. An impression
a high dam of the order of 270 m above the lowest foundation packer consists of an inflatable semi cured element wrapped
level with an underground power house of installed capacity with a replaceable, soft rubber strip. When the packer is
of about 10,000 MW. The main adit lies in Siwalik rocks which inflated, the strip is extruded into the fracture which leaves a
included sandstones, mudstones, shales, breccia, marls and permanent impression on the surface. With a known orientation
sometimes conglomerates. The beds dip uniformly upstream of the impression packer in the borehole, the direction of the
at an average angle of 57° and strike slightly obliquely across stress field can be interpreted.
the gorge in a direction N111°E. Discontinuities include four
Case Study: Larji H.E. Project, Himachal Pradesh
major joint sets, shear zones nearly parallel to bedding and
minor fault parallel to all joint sets. Hydrofracturing tests were conducted at Larji H.E. Project,
located in Kullu District of Himachal Pradesh, which envisages
Overcoring tests were conducted in the adit to power house,
the harnessing of hydropower potential of the river Beas with
in three holes using USBM deformation gauge (CSMRS
a total capacity of 126MW (CSMRS Report 1999). In the
Report 1988). The choice for selection of site for overcoring
tests at two locations (193 m & 286 m chainage) was restricted proposed underground power house area, rock type is quartz
to sandstone faces due to local geology and geometry of adits. chlorite mica schist. The strike of the rock varies between
Some joints were observed to be sufficiently continuous as N5°W–S5°E to N15°W-S15°E with dips of 30° to 50° in
the water was seen coming out from joints during drilling. south-west direction. The foliation joint is the prominent set of
joint and there are about 10 prominent sets of joints.
The results of the tests conducted at chainage 193 m indicated
that the major principal secondary stresses are nearly vertical Thirteen tests were conducted in the power house drift in seven
in all the three holes and its magnitude varies from 42.0 to 67.0 drill holes at different RD’s. The tests were carried out at
kg/cm2. The corresponding horizontal stresses vary from 18.0 to different depths in both horizontal and upward vertical
28.0 kg/cm2. The ratio of the two stresses ranging between directions (Crown of the drift). Pressure versus time curve of
0.32 to 0.57. Using three dimensional elastic analysis and the test no. 1 is shown in Figure 5. The orientation of stresses has
least square solution scheme, the principal stresses and their been calculated with respect to top of the drill hole for the
directions are worked. Major principal stress was found to be horizontal drill holes and with respect to North in the vertical
66.76 kg/cm2, intermediate principal stress as 33.35 kg/cm2 drill holes (Sarwade et al. 2008).
and minor principal stress found to be 29.26 kg/cm2. From The findings of the test can be summarized as; Average
the tests conducted at chainage 286 m any conclusions could Maximum horizontal stress (σΗ = 7.14 ΜΠα), Average
not be drawn, as either no stress relief or sudden drop without Minimum horizontal stress (σh) = 5.41 MPa, Measured
any lower bound was observed. Vertical stress (σv(Measured)) = 4.49 MPa, Estimated Vertical
Stress from overburden (σv(Estimated)) = 5.40 MPa, Ratio of
2.3 Hydraulic Fracturing Method maximum horizontal stress to measured vertical stress
The objective of the hydraulic fracturing method is to measure (σH/σv(Measured)) = 1.59 and Ratio of minimum horizontal
the state of in situ stresses through a drillhole. The pressure stress to measured vertical stress (σh/σv(Measured)) = 1.20. Ratio
applied to the borehole wall is increased to a value, where of maximum horizontal stress to estimated vertical stress
existing fractures will open or new fractures are formed. The (σH/σv(Estimated)) = 1.32, Ratio of minimum horizontal stress to
fluid pressures required to open, generate, propagate, sustain estimated vertical stress (σh/σv(Estimated)? = 1.00. Average
and reopen fractures in rock at the test horizon are measured orientation of maximum horizontal stress = N35°E.
and related to the existing stress field.
Test No. 01
Hydraulic fracturing involves the isolation of part of a borehole Location: Power House
Test depth: 6.20-6.65m
using inflatable straddle packers and the subsequent 13.80

pressurization of the hole until the wall rock fractures. If an

P ressure (MPa)

10.35
axial fracture is produced, the pressure record obtained during
the test can be used to determine the magnitudes of the 6.90

secondary principal stresses in the plane normal to the borehole 3.45

axis. During a hydraulic fracture test, pressure versus time is
recorded. 0.00
0 3 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48 51 54
Time (min)
The magnitude of the minor secondary principal stress
component can be determined directly from the recorded shut-in Fig. 5: Pressure v/s Time Curve, Larji H.E. Project (H.P.)

236
 Stress Measurement in Rock Mass

3. CONCLUSIONS CSMRS Report (1996). Report on In situ Stresses using flat

jack test conducted at Rohtang Pass Tunnel Project, North
An exact determination of in situ rock stresses is difficult as
Portal, Himachal Pradesh.
the current stress state is the end product of a series of past
geological events and is the superposition of stress components CSMRS Report (1999). Report on Hydro-fracturing Tests
Using Minifrac System Conducted at Larjji H.E. Project,
of several diverse types. In situ stresses not only vary in
Himachal Pradesh.
space but also with time due to tectonic events, erosion,
glaciations etc. Hence, in situ stresses should be measured by Fairhurst, C. (2003). “Stress Estimation in Rock: a Brief History
and Review”, International Journal of Rock Mechanics &
any of the methods and not merely estimated. However,
Mining Sciences, Vol. 40, pp. 957–973.
integrated approach should be followed to obtain more
reliable estimates of the in situ rock stresses for the design Hudson FREng., John, A. (2005). “Engineering Properties
purpose. of Rocks, Elsevier Go-Engineering Book Series, Vol. 4,
pp. 1–290.
Nripendra Kumar, Alex Varughese, Kapoor, V.K. and
REFERENCES Dhawan, A.K. (2004). In situ Stress Measurement and its
CSMRS Report (1988). Report on in situ Rock Mechanics Application for Hydroelectric Projects—An Indian
Tests Conducted at Karnali (Chisapani) Multipurpose Experience in the Himalayas, SINOROCK, Held in Chaina.
Project, Nepal. Sarwade, D.V., Mishra, K.K., Kapoor, V.K. and Nripendra
CSMRS Report (1994). Report on in situ Stress Measurements Kumar (2008). In situ Stress Measurements in Rock Mass
by CSIRO Hi Cell at Nathpa Jhakri Project, Himachal for Underground Structures: A Case Study, Proceedings
Pradesh. of World Tunnel Congress 2008, Vol. 2, pp. 679–688.

237