Int. J Rock Mech. 3,tin. S c i & Geornech..4h~zr Vol. 24. No. I, pp. 53 73. 1987 (!14'~-'~()e~2 ~;~ 33 !

) Q - 0 0 0
Printed in Great Brita:n Pergamon Journals Lid

INTERNATIONAL SOCIETY FOR ROCK MECHANICS

COMMISSION ON TESTING METHODS

SUGGESTED METHODS FOR ROCK STRESS DETERMINATION

CONTENTS
Technical Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
Method 1: Suggested Method for Rock Stress Determination Using a
Flatjack . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
Method 2: Suggested Method for Rock Stress Determination Using the
Hydraulic Fracturing Technique . . . . . . . . . . . . . . . . . . . . . . 59
Method 3: Suggested Method for Rock Stress Determination Using a
USBM-Type Drillhole Deformation Gauge . . . . . . . . . . . . . 63
Method 4: Suggested Method for Rock Stress Determination Using
a CSIR- or CSIRO-Type Cell with 9 or 12 Strain Gauges 68

Joint Co-ordinators
K. Kim (U.S.A.) and J. A. Franklin (Canada)

The following persons contributed to the preparation and reviewed the drafts of these Suggested
Methods: A. J. Bowling (Australia); P. LeComte and J. H. L, Palmer (Canada); F. Cornet (France);
R. Ribacchi (Italy); K. Suzuki (Japan); J. Charrua Graca (Portugal); W. L. van Heerden (South
Africa); B. Bjarnason, B. Leijon and D. Stephansson (Sweden); P. Egger, H. Grob and K. Kovari
(Switzerland); B. Wareham (U.K.) and J. A. Aggson, D. C. Banks, M. L. Cramer, T. W. Doe. E. C.
Gregory, B. C. Haimson, W. Miller and G. Nicholson (U.S.A.). 25 individuals, 11 countries.
 ISRM: SUGGESTED METHODS FOR ROCK STRESS DETERMINATION 55

Suggested Methods for Rock Stress

Determination
Technical Introduction

Underground rock masses are subjected to compressive opening outwards, through a numerical model or anal-
stress which increases, in general, with depth. The rate ytical techniques; or measurements must be made via a
of increase, however, varies depending on various fac- drillhole that extends into an undisturbed region of the
tors. Worldwide in situ stress data indicate that vertical rock mass. When a drillhole is used, the stress deter-
stress varies in a more predictable fashion than horizon- mination method must allow for the disturbance caused
tal stresses because vertical stress is primarily affected by by the drillhole.
the weight of the overburden. Many endeavours have been made to arrive at a
When an opening is introduced in the rock mass, the reliable means of measuring in situ stresses. Tincelin [4],
natural state of stress is disturbed locally as the rock Obert et al. [3], Haimson and Fairhurst [1] and Leeman
mass attains a new state of equilibrium. The stress [2] have made significant pioneering efforts in developing
around an opening resulting from various man-made methods of stress measurement. Of the proposed
activities is termed "'induced stress" as opposed to methods, three have received wide acceptance: they
"virgin stress" or "absolute stress," which describes the include the flatjack method, overcoring and hydraulic
original, undisturbed state of stress. This natural state of fracturing. The techniques presented in these "Suggested
stress is often referred to as simply "in situ stress". The Methods" were selected on the basis of the degree of
"Suggested Methods" presented in this publication deal acceptance and application within the geomechanics
primarily with this natural state of stress. community. This, in turn, reflects the relative acceptance
Underground in situ stress is sometimes sufficiently of the theoretical basis of the various techniques avail-
high (relative to the rock mass strength) to cause rock able and the confidence in these methods that has
bursting, spalting, buckling, heaving, or other ground evolved from laboratory investigations, field research
control problems. In such cases, knowledge of the state and engineering practice.
of in situ stress is of critical importance to the design and It is important to emphasize that rock stress deter-
construction of engineering structures in a rock mass. mination methods complement each other: each offers
Even in cases where the effects of stress are less dramatic, different advantages and disadvantages with respect to a
the optimum shape, orientation and layout of under- particular application. Also, the increasing need for
ground structures, as well as the effectiveness and ulti- accurate and reliable rock stress determinations will
mate cost of rock support systems, can be significantly ensure that measurement techniques and data analysis
influenced by the in situ stress. procedures will continue to evolve and thereby resolve
Factors affecting the magnitudes and orientations of existing difficulties and uncertainties.
in sire stress include the weight of overlying materials,
geologic structures (on local and regional scales), tec-
Received 13 January 1986.
tonic forces within the earth's crust, residual stress and
thermal stress. The complexity of the relations between
these factors and the in situ stress usually prohibits REFERENCES
reliable estimation of rock stress. In addition, stress l. Haimson B. C. and Fairhurst C, Initiation and extension of
cannot be measured directly, and therefore, rock stress hydraulic fractures in rock. Soc. Petrol. Engrs J. 7, 310--318 C1967).
determination techniques rely on the measurement of 2. Leeman E. R. The CSIR doorstopper and triaxial rock stress
measuring instruments, Proc. ISRM Syrup. on the Determination
some response (e,g. displacement, strain, deformation) of Stresses in Rock Masses, pp. 578-616, L.N.E.C., Lisbon (1969).
that is induced by a disturbance of the rock mass. This 3. Obert L., Merrill R. H. and Morgan T. A. Borehole deformation
for determining the stress in mine rock. USBM RI 5978 (1962).
measured response of rock in a stress-disturbed zone
4. Tincelin M. E. Mesures des pressions de terrains dans les mines de
(e.g. the wall of a tunnel) is extrapolated, from t h e - fer de l'est..4nnls Inst. tech. B&irn. 58, 972-990 (1952).

Method 1: Suggested Method for R o c k S t r e s s D e t e r m i n a t i o n U s i n g a F l a t j a c k T e c h n i q u e

SCOPE in an excavation. Each measurement determines stress in
one direction only and, therefore, a minimum of six
1. (a) The method is intended for the determination of measurements in independent directions are required to
rock stress parallel to and near the exposed rock surface determine the stress tensor.
56 ISRM: SUGGESTED METHODS FOR ROCK STRESS DETERMINATION

(b) The method involves the observation of the move-

ment of pairs of measuring pros located on each side of
a slot when the slot is cut and. subsequently, when
pressure ~s applied to the internal surface of the slot.
,i ¸,
(c) Measurements may be carried out even in fractured
rocks, provided a slot may be cut and remain open
throughout the process of installation of the flatjack. BLAT~AC~ .t
(d) The method may be used on materials that do not la FLAT JACK
necessarily exhibit reversible elastic properties or iso-
tropy, provided that corrections are made to validate the b TEST CONFIGURATION
results.~*

APPARATUS
2. (a) A flatjack consisting of two fiat sheets of steel ,oo- _-
plate or other suitable material welded together around
the edge to form a flat envelope of at least a 0.1 m 2 area
and incorporating a hydraulic inlet tube with con-
nections to a hydraulic hose and bleed valve. The z
selected shape o f the flatjack will depend on the method
chosen to cut the slot. 2 Great care should be taken in the
welding carried out around the edge o f the flatjack and FXCAVATIOt~ TIME -- FLATJACK P~ESSL,~E b

around the bleed tube inlet, so that the jack can expand
flexibly without leaking when installed and inflated to {c) PIN SEPARATION VERSUS SLOT EXCAVATION TIME AND FLAT JACK
PRESSURE ¢NOTtCE IDENTIFICATION OF C~LCELLATION PRESSURE, P~
full test pressure.
(b) A hydraulic pump, operated either manually or Fig. [. The flatjack test [2]. (a) Flatjack. (b) Test configuration. (c)
electrically, with an attachment to a load maintainer. Pin separation vs slot excavation and flatjack pressure (notice
identification of cancellation pressure, P~).
The pressure should be measured on gauges having an
accuracy of at least 5% of the estimated stress. The
system connected by high pressure hoses should be measuring pins, installation o f the measuring pans and
capable of maintaining any pressure within the desired cutting o f the flatjack slot.
range for a period of at least 5 min. (g) Grout, grout mixing and grout placing equipment,
(c) Two or more pairs of measuring pins grouted into if required, 3 for the installation of measuring pins and
holes drilled in the rock on each side o f the flatjack slot. the flatjack. Grout should be of strength similar to that
The typical pin is 12ram dia and 150ram in length; o f the rock being tested. Portland cement or epoxy resins
actual dimensions will depend on rock quality. The are commonly employed. The latter gain full strength
exposed end of each measuring pin and the separation more rapidly and, therefore, are usually used to anchor
(d) between measuring pins (Fig. 1) should suit the measuring pins.
measuring instrument (see paragraph 2(d)). In addition
to surface measurement pins, borehole instrumentation PROCEDURE
(stressmeters) can be installed. When the near surface
rock appears damaged by the excavation works, it is Site selection
preferable to measure the displacements at a sufficient 3. (a) In the selection of a zone of rock for testing,
depth to avoid the damaged rock. consideration must be given to the number of tests to be
(d) A demountable mechanical or electric displace- carried out in this zone. A minimum o f six tests in
ment gauge with an average gauge length of between 150 independent directions are required if one is to obtain
and 220 mm or, for larger flatjacks, 1/3 to 1/2 the size the complete stress tensor, but usually, additional tests
of the flatjack. The measurement range should be at least are carried out at any one location to enable a best fit
5ram and resolution of each reading should be to be obtained, mathematically, following assessment of
0.002 mm or better. the results. The preferred test layout in a tunnel or adit
(e) An appropriate rock drill or saw to cut the flatjack is to carry out nine tests: three in the roof. three in the
slot. 2 Slots may be formed by cutting overlapping drill- sidewall and three in the face. An example layout of slots
holes (stitch drilling), by circular saw or by wire saw. -fOr flatjack tests is illustrated in Fig. 2. The tests should
When using overlapping drillholes, these slots should be as close as possible without interfering with one
have a diameter not exceeding 40 mm and should over- another and should be a minimum of five times the
lap by 1/3 to 1/2 of the full diameter. tunnel diameter away from any other heading:
(f) Mounting frames, templates, jigs and other equip- (b) Once the general position for the test zone has been
ment to facilitate accurate drilling of holes for the determined, the excavation in the area must be carried
out with maximum care. Presplitting of the test adit is
" Numbers in superscript refer to Notes followingthe text for each suggested, followed by careful hand excavation and
method. removal of all loose material.
 ISRM: SUGGESTED METHODS FOR ROCK STRESS DETERMINATION 5":

Readings should be repeated a su~cient number of times

to achieve a repeatability of 0.005 ram.
(d) The slot is then cut. Care should be taken to
maintain the slot in the required direction and perpen-
dicular to the rock face. It is usual to cut the slot deeper
than the dimension of the flatjack and to set the loaded
area back from the rock face b~ a minimum of
25 mm. This prevents local failures of the rock during
pressurization.
(e) When coring is used, the cores should be retained,
placed side by side and photographed to record geolog-
ical features of the test area. When no cores are avail-
able, the character of the rock should be recorded by
observation of the rock face or by drilling at a location
Fig. 2. Example of the layout of slots for flatjack tests. no closer than two flatjack lengths to the test area.
(f) Further sets of displacement readines are to be
taken after cutting the slot to record the amount of slot
Selection and preparation o/" individual test locations closure and whether closure is instantaneous 5 or time
4. (a) Each test location should be in a firm, flat or dependent.
slightly concave rock surface. When struck with a drill (g) The flatjack is inserted fully into the slot and. if
steel or rod, the rock should produce a ringing sound necessary, grouted) Care must be taken not to trap
(should not sound hollow). Should no suitable location pockets of air in the grout. Should this occur, failure of
be immediately available, hand or pneumatic tool the flatjack is likely or the test results may be unreliable.
excavation must be used to prepare the test surface. (h) After the grout has set, the pressure in the flatjack
Consideration should be given to a possible modification is increased using pressure increments that are deter-
of the tunnel geometry by local overexcavation. mined from the magnitude of the displacement measure-
(b) The distance between the test location and any' ments and the control of the hydraulic pumping system.
significant geological discontinuities or irregularities on Pressure increments should allow a minimum of 10
the rock surface should be at least three times the length readings for the expected maximum pressure range.
of the flatjack slot. Tests in closely jointed rock can be (i) Readings of pin separation are to be taken at each
conducted, provided that the slot can be excavated and pressure increment. Pressure should be increased until
kept open long enough to install the flatjack. the separation of the pins is the same as before the slot
was cut. The pressure at which this is achieved, termed
Calibration the cancellation pressure, is to be determined from the
5. (a) Edge effects caused by welding, particularly for pressure versus pin separation graphs as described in
small-sized flatjacks, lead to the hydraulic pressure paragraph 7(c) below. A typical field data sheet is shown
within the jack being higher than the pressure exerted by in Fig. 3.
it on the walls of the slot. Flatjack suppliers should
measure this difference using suitable laboratory pro-
CALCULATIONS
cedures and should supply an appropriate calibration
factor with each flatjackP 7. (a) Recorded hydraulic pressures are to be corrected
(b) All pressure and displacement measuring equip- to give applied slot pressures using the edge effect and
ment is to be calibrated prior to its use in each test series. pressure gauge calibration factors discussed in para-
Calibrations are to be done by an independent testing graphs 5(a) and (b).
laboratory. (b) Slot closure and opening values are to be
calculated for each pair of pins and for each
Installation and testing sawing/pressurization increment by subtracting initial
6. (a) The long dimension of the proposed flatjack slot from subsequent readings.
should be oriented perpendicular (_+ 3 °) to the direction (c) Closure and opening for each pair of pins are to
in which the surface rock stress is to be measured. be plotted against applied pressure to determine the
(b) The pairs of measuring pins are to be located average cancellation pressure (see Fig. t(c)).
symmetrically across the marked flatjack slot location. - (d) The stress component acting perpendicular to the
The distance (d) between pins is determined by the plane of the flatjack before cutting the slot may be taken
displacement gauge. A line joining individual pins of as approximately (_+5%) equal to the average cancel-
each pair should be within 3 to normal to the slot. lation pressure, provided the pin-separation versus pres-
(c) Templates are to be placed on the prepared rock sure curves determined by a series of loading-unloading
face and the measurement pin positions described in cycles, do not show noticeable hysteresis.
paragraph 6(b) marked. The holes into which the mea- (e) The flatjack method of stress determination, with
suring pins are to be grouted are to be drilled, the pins flatjacks positioned as in paragraph 3(a), results in the
fixed in place and initial separation readings taken. determination of the disturbed stress components in the
58 ISRM: SUGGESTED METHODS FOR ROCK STRESS DETERMINATION

Test Data Sheet

Pro}ecl 3a~e

Feature Test no. _

Rock tyc~e OMentatmn

i
Eouiomeflt l Oete of
descriD~on S4full no ca~brlt~n

j
Pin lep4.'ltion medlngs

l~me P?essum

i i
i I i
h

• I l .

Fig. 3. Field data sheet.

immediate vicinity of the opening. This information can (h) The type, manufacturer and calibration informa-
be extrapolated from the opening outward to the un- tion of the displacement gauge used.
disturbed virgin stress by application of the theory of 9. The report should include the following detailed
elasticity or by numerical modelling techniques. information for each flatjack test:
(a) Initial pin displacements prior to slot excavation.
(b) Pin displacements following slot excavation (im-
REPORTING OF RESULTS mediately following slot excavation and several times
8. The report should include the following general prior to flatjack pressurization).
information: (c) A tabulation and graphic history of pin displace-
(a) A description of the test site location. ment versus flatjack pressure.
(b) Details of the test location(s) within the test site. (d) Interpretation of test results along with the
(c) Rock type and local geological structure. method(s) used to estimate the virgin stress (if applic-
(d) A description, illustrated with diagrams and able).
photographs, of the procedure and equipment used. (e) Test results showing substantial discrepancies with
Reference may be made to these "Suggested Methods," ~iher data and giving possible or probable explanations
noting departures from the recommended equipment of the causes.
and procedures.
(e) Flatjack manufacturer, specifications and cali- NOTES
brations.
(f) Diagrams of flatjack and measurement pin geo- 1. A summary of the likely causes and magnitude of
metries. errors is given in Table 1.
(g) Details of the flatjack slot excavation method and 2. When overlapping boreholes are used to cut the
problems encountered (if any). slot, the flatjack shape may be chosen with consideration
 ISRM: SUGGESTED METHODS FOR ROCK STRESS DETERMINATION 59

Table I. Errors and uncertainties in the measurement of in situ stress using t'tatjacks
Cause of error or Adjustment. calculation Liket? magnitude of
uncertainty or allowance error and or correction
(I) Disturbed ground at Careful excavation and Could invalidate the
test site testing of slot positions results completeI?
(2) Fit in slot. change Calibrations carefull? Correction 10-15% at
in contact area carried out in a modelled high stresses. Higher
during loading of slot ~ith displacement at low stresses
flatjacks measured and controlled
to rock stiffness
{3) Non-repeatable CycIing of flatjack loading Depends on rock type
stress-strain in order to determine its and in situ stress
behaviour extent. (Making allowances level. Could produce
for any' change in jack considerable errors at
characteristics if original low in situ stresses
jack dimensions not
recovered.) Further
experimental work needed
on this aspect
(4) Biaxial stress field Mathematical correction Depends on slot geometr~
based on linear elasticity but probably between 0-5°°
(5) Slot and flatjack Mathematical correction 0-5% as flatjack is
different dimensions based on linear elasticity fitted to slot in this case
(6) Measuring system Standard error for 5% with equipment noted
equipment in this paper
{7) Effect of flatjack Some influence charts 0 to > 10% depending on
rigidity available but a flatjack geometry and
mathematical study for the relative stift'nesses
equipment used is best

o f the g e o l o g y a n d f r a c t u r e spacing: T h e c h o i c e is usually 5. If the pins m o v e a w a y f r o m e a c h o t h e r o n slot

r e c t a n g u l a r , with a m i n i m u m size o f 300 x 3 0 0 m m . c u t t i n g , t h e n the r o c k stress c o m p o n e n t at the test
W h e n a saw b l a d e is to be used, the flatjack m u s t be in l o c a t i o n is tensile a n d c a n n o t be m e a s u r e d by this
the f o r m o f a c i r c u l a r s e g m e n t m a n u f a c t u r e d to the method.
s h a p e o f the b l a d e cut. T h e s a w i n g a l t e r n a t i v e is pre-
ferred, w h e n feasible. METHOD 1: B I B L I O G R A P H Y

3. G r o u t is best a v o i d e d by s a w i n g a p l a n a r slot w i t h I. Bernede J. Mesures des contraintes au verin pIat-nouvelles

s m o o t h walls o f j u s t sufficient s e p a r a t i o n to a c c o m m o - possibilities. Proc. 3rd Int. Congr. on Rock Mechanics, Vol. 2A,
pp. 433-438, Denver (1974).
d a t e the flatjack. A c i r c u l a r saw is suitable, since it 2. Goodman R. E. Introduction to Rock .~Iechanics. p. 115. Wiley,
p r o d u c e s a s m o o t h a n d u n i f o r m - w i d t h slot, but usually New York (1980).
is l i m i t e d to c u t t i n g a s e m i c i r c u l a r - s h a p e d slot o f d e p t h 3. Habib P. and Marchand R. Mesures des pressions de terrains par
l'essai de verin plat. Annls Inst. tech. Bdtim. 58, 11952).
less t h a n the saw b l a d e radius. A p a t e n t e d m e t h o d , 4. Hock E. and Brown E. T. Underground Excarations in Rock,
w h e r e the saw c a n p e n e t r a t e to a n y d e p t h by using a p. 384. IMM, London (I980).
c e n t r a l - c o l u m n d r i v e i n s e r t e d in a p r e - d r i l l e d hole, 5. Londe P. The role of rock mechanics in the reconnaissance of
rock foundations, water seepage in rock slopes and the stability
is d e s c r i b e d in the I S R M " S u g g e s t e d M e t h o d for of rock slopes, Q. Jl Engng Geol. 5, 57-127 t1973).
Deformability Determination Using Flat Jacks." 6. Obert L. and Dural W, ]. Rock Mechanics and the Design of
Structures in Rock, p. 417. Wiley, New York (1967).
4. E d g e effects m a y be s i m p l y t a k e n into a c c o u n t by
7. Rocha M., Baptista Lopes J. and DaSilva J. A new technique
e s t i m a t i n g the i n o p e r a t i v e w i d t h a r o u n d the flatjack for applying the method of the flatjack in determination of
p e r i m e t e r , s u b t r a c t i n g this f r o m the t o t a l w i d t h a n d stresses inside rock masses. Proc. Ist bzt. Congr. on Rock
Mechanics, Vol. 2, pp. 57-65, Lisbon (1966)
r e d u c i n g the a p p l i e d stress by the r a t i o o f effective
8. Tincelin M. E. Mesures des pressions de terrains dans les mines
flatjack a r e a to slot area. T h e c a l i b r a t i o n o f flatjacks in de fer de l'est. Annls hist. Tech. B~i:im. 58, 972-990 (1952).
a compression-testing machine may provide a more 9. Wareham B. F. and Skipp V. O. The use of the flatjack installed
in a saw-cut slot in the measurement of in situ stress. Proc 3rd
a c c u r a t e a s s e s s m e n t o f this a d j u s t m e n t . T h i s is p a r t i c u -
Int. Congr. on Rock Mechanics, Vol. 2A, pp. 481-488. Denver
larly so for s e m i c i r c u l a r flatjacks. (I974).

Method 2: Suggested Method for Rock Stress Determination Using the

Hydraulic Fracturing Technique
SCOPE t h r o u g h a drillhole. T h e test p r o v i d e s , in g e n e r a l , the
m a g n i t u d e s a n d d i r e c t i o n s o f the m a x i m u m a n d m i n -
1. (a) T h e o b j e c t i v e o f the h y d r a u l i c f r a c t u r i n g tech- i m u m stresses in the p l a n e p e r p e n d i c u l a r to the drillhole.
n i q u e is to m e a s u r e the state o f in s i t u stress u n d e r g r o u n d H y d r a u l i c f r a c t u r i n g is the only' r o c k stress deter-
60 1SRM: SUGGESTED METHODS FOR ROCK STRESS DETERMINATION

ruination technique that has been successfully applied to the designated test section is of suitable diameter for
deep drillholes. Thus. this technique has found applica- satisfactory seating of the packers.
tion in s~te characterization investigations wa vertical (c) The alignment and straightness of the drillhole may
drillholes from the surface. When access to the site be measured usmg an ortentation tool. if there is any
through shafts, tunnels or shallow drillholes is available. indication that deviation is excessive.
"'Suggested Methods" 1. 3 or 4 can also be applied (d) Magnetic compasses are generally used to orient
(b) Fluid pressure is applied to a test section of a cameras or televiewers for impression packers. Alternate
drillhole isolated by packers. The fluid pressures re- means of orientation must be used if the mineralogy of
quired to generate, propagate, sustain and reopen frac- the rock is likely to affect the compass reading Magnetic
tures in rock at the test horizon are measured and are rich rocks (e.g. Iron formations or basic igneous rocks)
related to the existing stress field. Directions of measured may be suspect. Gyroscopic compasses, which maintain
stress are usually obtained by observing and measuring the orientatton of the inspection device from the hole
the orientation of the hydraulically induced fracture collar, may be used.
(hydrofracture) plane.
(c) The method is. in general, more suited to mea- Packer equipment (Fig. 4) comprising
surements at depths > 50 m that are beyond the capabil- 4. (a) A system to isolate a test section of drillhole is
ities of most other techniques. It has the advantage of required. Inflatable packers, through which a water flow
requiring no advance knowledge of the elastic properties pipe runs, are used to seal the hole, enabling a test
of the rock and being able to be carried out without section to be pressurized. Double packer systems, which
difficulty below the water table. It tends to measure isolate a part of the hole. are generally used, but a single
stresses over a relatively large area, > 0.5-1.0 m dia. not packer, which isolates the base o f a hole. also may be
at a point. considered] Hydraulic or gas expansion is used to set the
(d) The method is most accurate when applied in packers and seal the test interval. The initial packer
materials whose behaviour approaches that of brittle, setting pressure depends on the packer type. If the
homogeneous, elastic, isotropic and non-porous media. interval pressure approaches the packer pressure, the
(e) The drillhole direction is assumed to be a principal packer pressure should be increased to a level sufficient
stress direction. Usually, this assumption is considered to prevent leakage past the packers.
valid from vertical holes drilled from the surface, in (b) The packers are separated by spacers.' The length
which case the vertical stress is calculated from the of the test section is chosen from the observation o f core
overburden weight. The accuracy of the test results will
be considered questionable if the drillhole direction ~a
deviates substantially (more than _ 15°), from a prin- TO PUMP.FLOWMETER. ~01
PRESSURETRANSDUCERS TO PUMP
cipal stress direction.

rO PUMP
APPARATUS

Drilling equipment GH.PRESSURE

TUBING
DRILL
2. (a) Any drilling equipment capable of producing a ROD " ~ - .
HIGH,PRESSURE
stable hole to the required test depth may be used. The . ~ ~ HOSES
hole diameter should suit the available packer equip- PRESSURE
J ~A~SDUCER
ment or vice versa. HOUSING COMPASS
(b) The drilling equipment should also be capable of |
obtaining core samples in the vicinity of the test sections
in order to evaluate drillhole rupture strength and to
examine discontinuity orientations and characteristics.

Inspection equipment !
iMPRESSION
PACKER "~'~
~ ]
~1.2 m
| -'~' \
3. (a) Knowledge of the directions of hydrofractures FRACTURE
INTERVAL*~09 m STRADDLE
is required in order to estimate principal stress direc- ~1 / . . . / / " PACKER
I
tions. Any of the following methods (all o f which are
capable of resolving to within + 5 °) may be used to find
these directions: DRILLHOLE
(i) Visual inspection using a drillhole periscope or
television camera. A video recording of a pre- ---J ~ / ORILLHOLE
/-
testing observation is advantageous for com-
parison purposes.
(ii) Examination of an acoustic televiewer image ob-
tained from reflected acoustic signals. Fig. 4. Se4aematicrepresentation of (a) hydrofracturing tool and [b)
(b) A drillhole caliper may be helpful to ensure that impression packer.
 ISRM: SUGGESTED METHODS FOR ROCK STRESS DETERMINATION 61

a n d o r drillhole wall conditions by" means of the meth- (b) Rock cuttings a n d o r cores are examined in detail
ods described in paragraph 3. A minimum length five to determine rock characteristics at the test horizons.
times the drillhole diameter is commonly recommended. The choice of packers and inflation pressures may' be
(c) The packers must provide a complete seal over the affected by rock hardness and roughness of the drillhole
full range of test pressures, with no fluid leakage from wall.
the test section. (c) It is recommended that the hole be flushed to
remove debris a n d o r the drill bit may be lowered to the
FluM injection equipment, comprising test depth to clear the passage for the packer assembly.
5. (a) A high-pressure pumping system capable of (d) It is recommended that the positions, orientations
maintaining a constant flow over the range of pressures and apertures of geological discontinuities within the test
expected during the test. section be estimated and recorded, using, for example.
The pumping system should have sufficient capacity to the core, impression packers, a downhole camera or an
overcome the friction losses in the supply rods and to acoustic televiev,'er, if practicable. This also serves as a
initiate hydrofracture. pre-test run of the fracture orientation measurement.
(b) Sufficient supply rods, tubing or hose are needed (e) It may' be advisable to run a drillhole caliper log
for the required depth of measurement. These are usu- in order to avoid placing packers in oversized sections of
ally used to lower the packer into the drillhole. the hole.
if) The packer assembly' is inserted to the prede-
Measurinz equipment, including termined depth: the depth is recorded: and the packers
6. (a) Pressure transducers for measurement of fluid are inflated to a pressure sufficient to seal against the
pressure at the surlEce or immediately above the packer. applied fluid pressure.
A surface pressure transducer is sufficient for shallow (g) When filling the injection tubing, care must be
tests or low flow rate tests that do not have head loss taken to eliminate air from the system. Trapped air
problems in the tubing. Deep tests and high flow rate greatly increases the compressibility' of the system and
tests could be enhanced by using a downhole transducer has an adverse effect on the rate of pressure build-up
as well. during test interval pressurization.
Pressure must be continuously recorded during the
testing using analogue electrical devices (chart recorder), Testing
digital recording and/or, for downhole pressure gauges, 8. (a) When pressure is monitored at the ground
mechanical recording. A pressure-time record should be surface, the pressure in the test interval is increased
plotted as the test is performed. This record is vital for slowly to ensure minimal pressure losses in the tubing.
deciding when to pump and when to shut in the test When pressure is monitored within the test section,
interval. Transducers should be calibrated against re- pressure losses are unimportant. No standard for pres-
liable standards. Downhole transducers may also be surization rate or flow rate exists: however, a c o m m o n
checked using the pressure change that occurs as the range of pressurization rates is about 0.1-2.0 MPa/'sec.
transducer is lowered through the fluid in the hole. The pressurization rate is controlled by the constant flow
Monitoring of interval pressure should be carried out rate selected. The appropriate flow rate to achieve the
with an accuracy of no less than + 2 . 5 % of the maxi- desired pressurization rate will vary depending on the
mum expected applied pressure or with an accuracy overall compressibility of the test system, which largely
acceptable to a specific project requirement. reflects the elasticity of the tubing, the length of the
(b) A pressure gauge or transducer to measure packer tubing, the compressibility of the fluids and the volume
inflation pressure with a compatible level of accuracy to of fluid in the test system. In general, deep tests with
the interval pressure gauge. large diameter tubing wilt require higher flow rates than
(c) An instrument to record fluid flow with time. short-hole tests with smaller diameter tubing. The
Surface measurement of flow is sufficient, as only sudden packer pressure should be initially set well below the
changes of flow characteristics are required. A flow anticipated breakdown pressure; the packer pressure
metre, resolving to less than 0.2 l~min, should be used. should be increased at the same rate as the injection
pressure. This procedure reduces the possibility of frac-
PROCEDURE ture initiation caused by the packer pressure. The test
interval pressure is recorded against time. As pressure
Drilling and inspection increases, both tangential and vertical effective stresses
7. (a) Choice of hole diameter and size of downhole - C a n become tensile.
hydraulic fracturing equipment may be made based on Fracture will occur if the induced tensile stress reaches
the equipment available. Following determination of the the drillhole rupture strength. Evidence of failure may be
test location and depth, a drillhole should be sunk obtained from the pressure/'time curve. The dritlhole
beneath that depth to provide the test intervals. Final fluid pressure at the moment of drillhole rupture is
choice of test zone length and depth is made based upon termed the "fracture initiation pressure" (Pr) or break-
the fracture characteristics of recovered cores or on down pressure.
inspection of the drillhole wall by an optical or acoustic (b) After injecting a volume sufficient to propagate a
logging tool. fracture length equal to about three times the drillhole
62 ISRM: SUGGESTED METHODS FOR ROCK STRESS DETERMINATION

diameter, injection is stopped and the hydraulic system

P, = ~ R 4 C T U R E I N I T I A T O N P R ~ S S U R ~
is sealed or "shut-in." yielding the "'instantaneous shut-
P. : ~ R ~ C T U R E R E O P E N I N G ~ R E S S U R E
in pressure" (P~). Additional repressurization cycles are P P~ = SHUT IN P R E S S U R E S~UT ~

used to determine the "fracture reopening pressure" (Pr) SHUT IN ' ~' P~ = S H U T tN P R E S S U R E

and additional measurements of the shut-in pressure

i '
(P0. P
P
~
7-3RMATION
PORE P R E S S U R E

As reported by Zoback and Haimson I15], hydraulic

fracturing tests can sometimes result in indistinct shut-in
pressures, and in such cases, hydraulic fractunng prac-
titioners have proposed various methods to help identify
the change in slope associated with fracture closure. 3
(c) Subsequent repressurization cycles should be con-
ducted at similar, constant flow rates; higher or lower
flow rate cycles may be added at the discretion of the
operator. Use of higher or lower flow rate cycles in the Fig. 5. Idealized hydraulic fracturing pressure record.
stress calculation must be specified and explained in the
report of results. O'max = 3P~ - Pr - Po (for subsequent repressurization
(d) The packers are deflated and the equipment is cycles).
removed from the drillhole. Care should be taken that
the packers are fully deflated before attempting to move The drillhole rupture strength (T) is determined from
them. laboratory tests, which model the hydraulic fracturing
(e) The hole inspection (paragraph 7d) is carried out process (preferable) or the tensile strength obtained from
to observe and record hydrofracture positions and orien- direct tension, or Brazilian tests. Appropriate cor-
tations. rections for the effects of sample size and test con-
(f) The "drillhole rupture strength" of the rock (T) figuration may need to be made (see Ratigan [llD. The
may be estimated from laboratory tests on core samples vertical stress is usually assumed to be the stress gener-
or may be obtained in situ by comparing the fracture ated by the depth and density of overlying rock. The
initiation pressure with subsequent fracture reopening direction of a,,a~ is m the direction of the fracture plane
pressures. and orthogonal to ami,.

CALCULATIONS REPORTING OF RESULTS

9. (a) Nomenclature used in the calculations is as 10. The report should include the following general
follows: information:
H = depth at test zone below ground level; (a) A description of the test site location.
Ph = static pressure head of fracture fluid; (b) A geotechnicat log of the test section giving all
7= mass density of rock: available information and including the r e c o r d e d geo-
P0 = initial pore water pressure; logical discontinuity characteristics and dritlhole watt
Pf = fracture initiation pressure; conditions.
Pp = pumping pressure; (c) A geological description of rock tested, including
Ps = instantaneous shut-in pressure; rock type and availability of core.
P, = fracture reopening pressure; (d) The test depth, length of the test zone and drillhole
av = vertical stress; size.
amin = minimum horizontal stress; (e) A description, illustrated with diagrams or photo-
a,,ax = maximum horizontal stress; graphs, of the procedure and equipment used. Reference
T = drillhole rupture strength of the rock. may be made to these "Suggested Methods," noting
departures from the recommended procedures.
(b) Where the pumping pressure is measured directly 11. The report should include the following detailed
at the test zone, Pr, Ps and P~ may be obtained directly depth information for each measurement location:
from the time versus pressure plot. (a) Graphs showing pumping or flow rate, injecuon
Figure 5 presents an idealized hydraulic fracturing pressure versus time for each test and a description o f the
pressure record. The general form of the pressure record --method(s) used to select Pr, Ps and P,, if distinct
depends on the relative magnitudes of the principal pressures are not obvious.
stresses. (b) Tabulated values of H, Ph, Po, Pf, P~, P , maximum
(c) When the plane o f hydrofracturing is nearly paral- and minimum horizontal stresses and stress directions.
lel to the drillhole axis, the following expressions may be (c) Representations o f fracture traces, constructed
used to obtain the principal stresses: from impression packers, on photographs a n d / o r acous-
O'min = Ps tic televiewer logs of test intervals.
am~ = T + 3Ps -- Pf -- P0 (for initial pressurization (d) A description of the method used for calculating
cycle) O'max. In case the drillhole rupture strength was deter-
 ISRM: SUGGESTED METHODS FOR ROCK STRESS DETERMINATION 63

mined by laboratory tests, the laboratory test results and 4. Gronseth J. M. Determination of the insta,".taneous shut-in
pressure from hydraulic fracturing data and its reliability as a
the method of data reduction, including equations used, measure of the minimum principal stress. Proc. 2_Zrd U.S. Syrup.
should be reported in sufficient detail. on Rock Mechanics. pp. t83-189, Berkele~ (19S21.
(e) Test results showing substantial discrepancies with 5. Haimson B. C. A simple method for estimating in situ stresses
at great depth. ASTM Special Technical Publications 554, pp.
other data and giving possible or probable explanations
156-182, Philadelphia (I974).
of the causes. 6. Haimson B. C. Near-surface and deep h,~drefracturing stress
measurements in the Waterloo quartzite. Int. J. Rock 5Iech..tlin.
Sci. & Geornech. Abstr. 17, 81-88 119_<0).
7. Haimson B. C. and Eairhurst C. Initiation and extension of
NOTES hydraulic fractures in rock. Soc. Pc,~rol. Engrx J. 7, 310-318
(1967).
1. Under some circumstances, the test interval may be 8. Hoek E. and Brown E. T. Underground Exc::::ztions in Rock,
isolated between a single packer and the base of the p. 384. IMM, London (1980).
9. Kim K. and Smith C. C. Hydraulic fracturing stress measure-
drillhole. However, the test results may then be inaccu-
ments near the Keweenaw fault in upper Michigan. Underground
rate because of stress concentrations at the base of the Rock Engineering. Proc. 13th Canadian Symp. on Rock Mechanics,
hole. Univ. of Toronto, pp. 24-30 (I980).
2. The spacer rod or tube between packers should be 10. Kim K., Dischler S. A , Aggson J. R. and Hard~ -"d. P. The state
of in situ stresses determined by' h~draulic fracturing at the
designed to carry the tension developed by application Hartford site, RHO-BW-ST-73. Rockwell Hanford Operations,
of the test pressure within the test section. Richland (1985).
11. Ratigan J. L. A statistical fracture mechanics determination of
3. The ISRM has formed a Commission on "Inter-
the apparent tensile strength in hydraulic fracture. Proc. Work-
pretation of Hydraulic Fracturing Pressure Records." shop on Hydraulic Fracturing Stress .~[easuremcn:. U.S. National
This Commission will prepare a document outlining the Committee on Rock Mechanics, Washington, DC, pp. 159-166
various interpretation procedures used by different in- (1982).
12. Roegiers J. C. and McLennan J. D. Factors influencing the
vestigators as well as the difficulties and problems prac- initiation orientation of hydraulicalb induced fractures. Proc.
titioners have encountered when attempting to interpret Workshop on Hydraulic Fracturing Stress Measurements, O F R
hydraulic fracturing pressure records. This will com- 82-1075, U.S. Geological Survey, Washington. DC, pp. 489-502
(1982).
plement the "Suggested Methods" presented herein. 13. yon Schonfeldt H. and Fairhurst C. Field experiments on
hydraulic fracturing. Soc. Petrol. Engrs J. 12(2), 69-77, also
printed in Soc. Petrol. Engrs Trans. Vol. 253 (1972).
I4. Wawersik W. R. and Stone C. M. Application of hydraulic
fracturing to determine virgin in situ stress state around waste
METHOD 2: B I B L I O G R A P H Y isolation pilot plant--in situ measurements. SAND 85-1776,
Sandia National Laboratories, Albuquerque (1985).
1. Bredehoeft L D., Wolff R. G., Keys W. S. and Shutter E. 15. Zoback M. D. and Haimson B. C. (Eds) Proc. Workshop on
Hydraulic fracturing to determine the regional in situ stress field Hydraulic Fracturing Stress Measurements, Open-File Rept
in the Piceance Basin, Colorado. Geol. Soc. Am. Bull. 87, 250-258 82-1075. U.S. Geological Survey, Washington. DC (1982a).
(1976). 16. Zoback M. D. and Haimson B. C. Status of the hydraulic
2. Doe T. W., Ingevald K., Strindell L. Leijon B., Hustrulid W., fracturing method for in situ stress measurements. Proc. 23rd
Majer E. and Carlson H. In situ stress measurements at the Stripa U.S. Syrup. on Rock Mechanics, pp. 143-156, Berkeley (1982b).
Mine, Sweden. Technical Information Rept No. 44, Lawrence 17. Zoback M. D. and Pollard D. D. Hydraulic fracture propagation
Berkeley Laboratory, Univ. of California, Berkeley (1983). and the interpretation of pressure-time records for in situ stress
3. Goodman R. E. Introduction to Rock Mechanics. p. I11. Wiley, determinations. Proc. 19th U.S. Syrup. on Rock Mechanics,
New York (1980). Stateline, pp. 14-23 (1978).

Method 3: Suggested Method for Rock Stress Determination Using a

USBM-type Drillhole Deformation Gauge

SCOPE bedding), modifications to the gauge will permit testing

to continue although alternative means of determining
1. (a) The purpose of this test is to determine rock the rock modulus must be used ~ and corrections must be
stress from measurement of changes in diameter (defor- applied to the measured diameter changes to account for
mation) of a drillhole during overcoring. The elastic the proximity to the base of the overcore hole. z
properties of the rock and the deformation measure- (c) Hole depths are usually less than 30 m although
ments are combined to calculate the stresses in the plane measurements to depths of 70 m have been made in
perpendicular to the axis of the drillhole. The test - vertical downholes.
procedure is modelled on the U.S. Bureau of Mines (d) Stress components in the plane perpendicular to
(USBM) "Borehole Deformation Gage" method de- the hole can be evaluated from the results. In tests from
scribed in the Bibliography (Obert et al. [13]). ground surface using vertical drillholes, it is c o m m o n to
(b) For this method to be successful, the rock should assume that the vertical stress is equivalent to that
not split or fracture during the overcoring process. Thus, imposed by the overburden load.
it is necessary to obtain cores of a length at least twice (e) In underground situations, tests in a single hole are
the diameter of the overcore. In cases where the core conducted to determine the stresses in a particular plane
"discs" (e.g. because of very high in situ stresses or thin of interest. Tests from three (or more) non-parallel holes
 ISRM: SUGGESTED METHODS FOR ROCK STRESS D E T E R M I N A T I O N 65

a spirit level in horizontal holes or an orientation device

.~. ,5 on the placement rod in vertical holes). The difference in
5 the readings should fall within 1 2 to 3 4 of the maxi-
.Ear 2a~ .+~
50 ° i0: mum difference range of the cantilevers. It" the difference
j
is too small, remove the gauge and add shims to the
..... c~ cantilever contact pistons. If the difference is too large,
O 0~,'/', ~4O k E t ~

,4
remove shims. With proper shimming the gauge should
fit firmly into the pilot hole without being forced. The
f
a b cantilever tips should be placed at least 150 mm, and
preferably 225 mm, past the base of the larger drillhole.
This will ensure that the presence of the larger hole does
// I~ \\\ not significantly affect the in sire stress field at the point
il Ii
\
II II \\\ of measurement•
<,. . . . . ?.~-
(g) Take up excess slack in the gauge cable and
\x
\ , '// D \ /A 30 ~ 35 ~ i 3 o
• 35 ~ '
connect the drill stem to the drill. Turn on the drilling
• " 3 ', ~ R T I C , & L
water and allow approx. 10 min to elapse before over-
DO'ANHOL5
coring. This allows the gauge to reach temperature
equilibrium.
c :d
(h) Start the overcoring with chuck speeds of approx.
120 rev min-I and a penetration rate of around
D : D,AMETER OF MINE OaENING 20 mm/min. The water pressure should not fluctuate and
should be maintained as low as possible but should be
Fig. 7. Drillhole configuration.
sufficient to allow good clearing of the cuttings. Record
the gauge readings on all cantilever axes at every 10 or
[15]). Deformation measurements should be taken out- 20 mm of penetration. A typical field data sheet is shown
side the zone of influence of underground openings; the in Fig. 8.
extent of this zone is generally taken as one diameter (D) (i) Overcoring should proceed until the overcoring bit
of the opening, as indicated in Fig. 7. Four possible has passed over the cantilever tips and beyond, for a
drilling configurations are presented in Fig. 7. distance of at least 150 mm and preferably 225 mm. The
Configurations (a), (b) and (c) have the advantage that total length ofovercore is thus 3 0 0 4 5 0 mm (approx. 2-3
the drill rig does not have to be moved for each hole. times the diameter of the overcore hole). In any case. the
Configuration (d) has converging drillholes and, thus, overcore should be long enough to obtain gauge read-
samples the smallest volume of material. ings which do not change with further overcoring. 6 A
typical plot of gauge readings during the overcoring
DriEing, gauge insertion and overcoring operation is shown in Fig. 9.
6. (a) Unless the stress close to a free face of rock is (j) Upon completion of overcoring, disconnect the
to be studied, actual overcoring of the deformation drill stem from the drill. Using the placement rods,
gauge should not begin until the larger drillhole has retrieve the gauge and remove the drill stem and the core
penetrated beyond the influence of surface irregularities barrel from the hole. Carefully determine the orientation
or openings. of the core, then use the core breaker, core shovel or core
(b) Start the hole using the large overcore bit and drill puller to remove the overcore.
to the depth at which overcoring is to begin.
(c) Remove the large core and start the pilot hole using Measuring the rock modulus
a short pilot hole starter core barrel; then, extend the 7. (a) The overcore r should be tested inside the biaxial
pilot hole for approx. 2 m. It is important that the pilot modulus chamber (Fitzpatrick [7]) as soon as con-
hole remain within + 14mm of the centre of the over- venient, preferably on site within 24 h of overcoring.
core. 5 Any fracture zones encountered by the pilot hole This applies particularly to rock cores which deteriorate
should be avoided during placing of the instrument. with time. Place the rock core inside the modulus
(d) Insert the 150 mm core barrel into the drillhole but chamber so that the actual measurement point is
do not connect the drill. Thread the gauge readout cable opposite the middle of the chamber• Orient the gauge
through the drill chuck and special water swivel. sensors in the same directions as during the overcoring.
(e) Connect the gauge readout cable to the r e a d o u t - Usually, marks are left on the walls of the pilot hole by
box and take initial readings with the gauge outside the the buttons of the gauge so that resetting at the correct
hole. Gauge calibration information is normally sup- position and orientation is simplified. If biaxial testing is
plied by the manufacturer. However, direct methods of not to be done soon after overcoring the core should be
periodically checking the calibration are available (see wrapped and sealed in heavy plastic to prevent drying.
Hooker and Bickel [9]). (b) Apply radial pressure to the rock core using the
(f) Using the placement rods, insert the gauge through hydraulic hand pump. Record drillhole diameter
the drill rods and into the pilot hole. Orient the gauge changes at various increments of applied pressure. Cal-
so that one cantilever pair lies in a known plane (using culate secant values of the biaxial elastic modulus from
R3,1MS 24t E
66 ISRM: SUGGESTED METHODS FOR ROCK STRESS DETERMINATION

Hole No Dale Or,entaI,on U,

Gage No Cah0¢at,o~ 'actor ~, _ _

Gage tacto, us

True Beating o~ Hole U3

DEFORMATION TiME
Gage Hole I IND,CATORu,
I uzREADING i Gage O. . . . . . De,. . . . . . . . I . . . . . . . . . ]
t [
-' ~Zerol U3 Set Star~ Read I R
; 1 1 I '!
g" F I c e ' I~IIS
/
~ ,j t I
[ I :
/' 1 ! / , ,
,,,,1 I I I , ' j
1

"*
8':2" ~
Ir i[ i I '

~"°" g"i ' ! ]

,o,
,o,:, I I i I! I

13 '/i" j 4 !
! 1
i
15" t
15'/*" i
18"
16111. '
17"
17'/~"
18"

Fig. 8. Field data sheet as recommendedby USBM [8].

plots of applied pressure versus drillhole diameter testing procedures that minimize the effects of elastic
changes using the unloading curves. nonlinearity can be used to determine anisotropic elastic
If possible, biaxial test pressures should be selected parameters (Aggson [I]).
such that biaxial test deformation magnitudes are com-
parable to the overcoring deformation magnitudes. CALCULATIONS
However, this is not always possible because the biaxial
test does not confine the ends of the core and failure can 8. (a) Calculations of in situ stresses involve varying
occur at relatively low biaxial pressures. Triaxial core degrees of complexity depending on whether stresses are
two or three dimensional in character and depending on
the degree of elastic anisotropy present.
-O.OOfi (b) Where the stress along the dritlhole a~s is zero(i.e.
j.f '~
stress conditions such as mightoccur close to the ground
surface or close to the wall of an opening), the defor-
o u, I mation or change in diameter (U) of the pilot hole is
given by the equation:
i o.,~ /x .....
o d p
-- U=-~[( +Q)+2(P-Q)]cos20 (1)
° t
where P and Q are the applied major and minor
~0o. ,,,., o,
MEASUREMENT
\ 1 secondary principal stresses in the plane perpendicular
FOR EACH CURVE
to the axis of the hole, d is the diameter of the pilot hole
0.020 1100 200 300 400 500
and 0 is the angle between the direction of P, the major
DEPTH OF OVERCORING BIT Imml secondary principal stress, and the direction of U, the
Fig. 9. Typical plot of drillhole deformation vs overcoring bit measured deformation. In the case of the USBM-type
position [11]. gauge which has three sensor axes, U~, U2, and U3,
 ISRM: SLGGESTED METHODS FOR ROCK STRESS DETERMINATION 67

spaced 6 0 apart in a counterclockwise direction (look- treated statistically to yield the least squares estimates of
ing into the drillhole), the major and minor principal the stress components and their standard error.
stresses and their orientations are given by the equations: (e) Young's modulus values are obtained by inserting
the biaxial chamber readings (P and U) in the following
E { -g
P=~/ {U,+ U . + U ~ ) + > S,2 [_( U - U : ) : thick-walled cylinder equation:
D: 2dP
E=
- - I U . - L ~ ) : + ( U ~ - Ut):] 1:} D:-d:" U

O =~d
E { .v"2
JUt+ U . + U 3 ) - ~ - [ ( U , - U : ) :
where:
D = diameter of the overcore;
d = diameter of the pilot hole:
P = applied radial pressure;
_;_ ( L"2 -- ~r 3 )2 _[_ ([/"1 -- ~J'l )2]1 2}
U = measured change in pilot hole diameter.
In cases where the rock is anisotropic, the measured
oo = tan ./7(u_, - values of U along variously oriented diameters will be
' 2 Ul- U,- U3 seen to differ. The effect of anisotropy on the calculated
where the angle 0e is measured from Ut to P in stresses depends on the relative magnitude of elastic
the counterclockwise direction. Depending on relative parameters and other considerations. If required, the
magnitudes of deformation, the range of this angle is degree of anisotropy can be calculated and allowed for
determined as follows: following the procedures outlined in the publications
given in the Bibliography (Amadei [2], Becker and
U : > U~ and ( U : + U ~ ) < - O") i .r
Hooker [4] and Becker [3]).
Op is in range 0°-45:

U, > U3 and ( U : + U3) > 2Ui, REPORTING OF RESULTS

Op is in range 45°-90 " 9. The report should include the following general
information:
[_.;~< C:, and (U: + [.:3) > 2U,.
(a) The locations, directions and length of drillholes.
Op is in range 90°-135 ° (b) A geotechnical log of core from each drillhole
/_,:~< [,7~ and showing the depths of measurement with particular
(U2 + U3) < 2UI,
attention given to geological and structural character-
Op is in range 135-~-180°. istics of the rock at the stress determination locations.
(c) A description, illustrated with diagrams and photo-
These equations are useful in the field for providing a graphs, of the procedure and equipment used. Reference
rough estimate of stress levels even in the three- may be made to these "Suggested Methods," noting
dimensional case. departures from the recommended equipment and
(c) In the three-dimensional case, assuming conditions procedures.
of plane strain isotropic elasticity: 10. The report should include the following detailed
depth information for each measurement location:
d(l - v-~)
U- [(P + Q) + 2(P - Q) cos 2 0 ] - ve.d (2) (a) Copies of field data sheets showing data from each
E
successful overcoring run or plots of overcoring data
where c. is the strain in the axial direction along the (e.g. Fig. 9) and the estimated values of U~. U2 and U3
drillhole. from that run.
If an estimate of a:, the axial stress, is available (e.g. (b) Plots of radial pressure versus borehole defor-
assumed superincumbent load in vertical holes from mation from biaxial modulus chamber tests or stress/
surface) then c: may be calculated since: strain curves from laboratory strain gauged rock cores.
1
(c) A tabulation of hole number, hole bearing, hole
c_. = - < . - E V ( P + Q) inclination, depth of overcore, measured values of U,, U2
and U3, orientations of Ut and measured values
assuming v, the Poisson's ratio for the rock, is known or for Young's modulus and Poisson's ratio at the test
measured by conventional laboratory techniques. The 6. location.
is applied in equation (2) to obtain more exact values of - (d) Plots and tabulations showing secondary principal
P and Q. A further iteration may be performed if so stress magnitudes and directions at the measurement
desired. locations and, if applicable, computer printouts of the
(d) The completely general case, but assuming iso- stress ellipsoid showing magnitudes and directions of the
tropic elasticity, has been treated by Panek [15]. Each set three principal stresses along with their standard devi-
of overcoring data contains three measurements of ations, standard error and correlation coefficient.
drillhole diameter change, which are oriented with re- (e) Test results showing substantial discrepancies with
spect to N and to the vertical, plus a value for Young's other data and giving possible or probable explanations
modulus and Poisson's ratio. All such data sets are of the causes.
 68 ISRM SUGGESTED METHODS FOR ROCK STRESS DETERMINATION

NOTES 7. Drillhote and core diameters r e c o m m e n d e d by the

1. Where the rock core is too short to fit inside the U S B M I H o o k e r and Bickel [9]) were selected to c o n f o r m
modulus c h a m b e r it is possible to test N X cores or pilot with compatible core testing procedures (Fitzpatrick [7]),
hole cores using strain gauges to determine the elastic Other overcore diameters and drilling procedures can be
parameters o f the rock. used in the application of this method (e.g. see Palmer
2. A solution to the problem of stress distribution and Low [14]).
a r o u n d the end o f a drillhole has been covered in H o o k e r
et al. [11].
The report gives the correction factors which should
be applied when the measured deformations are
M E T H O D 3: BIBLIOGRAPHY
influenced by the proximity o f the overcoring hole.
3. It is advisable to use at least a three-axis gauge, 1. Aggson J. R. Test procedures for noatinearb elastic stress-relief
overcores. USBM RI 8251 (1977).
because if a single axis gauge is used, three times the 2. Amadei B. Rock anisotropy and the theory of stress measure-
n u m b e r o f overcoring runs must be made. Various gauge ments. Lectures Notes in Engineering, Vol. 2. 478 pp. Springer.
styles are in extstence other than the U S B M gauge. New York (1983).
3. Becker R. M, An anisotropic elastic solution for testing stress
4. The U S B M - t y p e gauge has centralizing springs to relief cores. USBM RI 7143 (1968).
lightly support the b o d y o f the gauge at one end. 4. Becker R. M, and Hooker V. E. Some anisotropic considerations
However. the main holding force ~s provided by the m rock stress determinations. USBM RI 6965 (I967).
5. Crouch S. L. and Fairhurst C. A four component borehole
cantilever tips. If the gauge b o d y is held firmly to the deformation gauge for the determination of in situ stresses in rock
dritlhole at any other point, there is danger that, during masses. Int. J. Rock Mech. Min, Sci. 4, 209-217 (1967).
overcoring, the gauge b o d y could move axially as the 6. Duvall W. I. and Aggson J. R. Least squares calculation of
horizontal stresses from more than three diametraI deformations
rock core expands, dragging the cantilever tips over in vertical boreholes. USBM RI 8414 (1980),
the inside surface o f the hole and causing erroneous 7. Fitzpatrick J. Biaxial device for determining the modulus of
readings. elasticity of stress-rellef cores. USBM RI 6128 (1962).
8. Griswold G. B. How to measure rock pressure: new tools and
5. Eccentricity o f the pilot hole with respect to the proved techniques aid mine design. Engng Min. J. 164, 90-95
larger overcore does not affect the actual strain relief (1963).
measurements, but ~t can give rise to errors during 9. Hooker V. E. and Bickel D. L. Overcoring equipment and
techniques used in rock stress determination. USBM Circ. 8618
measurement o f rock m o d u l u s using the biaxial m o d u l u s (1974).
c h a m b e r and thick-walled cylinder equations. It is for 10. Hooker V. E. and Johnson C. F. Near-surface horizontal stresses
this reason that only a short length o f pilot hole can be including the effects of rock anisotropy. USBM RI 7224 (1969).
11 Hooker V. E.. Aggson J. R. and Bickel D. L. Improvements in
drilled ahead o f the overcored section. the three-component borehole deformation gage and overcoring
6. Where the in situ stresses are high e n o u g h to cause technique. USBM RI 7894 (1974).
discing or where the rock is so badly fractured that long 12. Kovari K., Amstad C. H. and Grob H. Ein beitrag zum problem
der spannungsmessung im fels. Proc. Int. Syrup. on Underground
u n b r o k e n cores cannot be obtained, it m a y be necessary Openings, pp. 501-512, Luzern (1972).
to use a modified gauge design which will permit the 13. Obert L.. Merrill R. H. and Morgan T. A. Borehole deformation
gauge sensors to engage the pilot hole at points close for determining the stress in mine rock. USBM RI 5978 (1962).
14. PalmerJ. H. L. and Lo K. Y. In situ stress measurementsin some
(within 15 mm) to the base o f the larger overcoring hole. near surface rock formations--ThorN& Ontario. Can. Geotech.
In such cases a correction must be applied to the J. 13(1), 1-7 (1976).
measured overcoring strains to a c c o u n t for the prox- 15. Panek L. A. Calculation of the average ground stress components
from measurements of diametral deformation of a drill hole.
imity o f the larger hole (see N o t e 2). USBM RI 6732 (1966).

M e t h o d 4: S u g g e s t e d M e t h o d for Rock Stress Determination Using a CSIR- or

CSIRO-Type C e l l w i t h 9 or 12 S t r a i n G a u g e s

SCOPE (d) Because o f difficulties in obtaining satisfactory

1. (a) This m e t h o d is intended to determine the stress strain gauge b o n d i n g when the rock is wet. this m e t h o d
tensor in rock by measuring the strains that occur in the is usually impractical for holes in water-saturated
walls o f a drillhole when the stresses are relieved by ground drilled d o w n w a r d . "'Suggested M e t h o d s " 1, 2 or
--3 m a y be used as alternatives.
overcoring.
(b) The m e t h o d is restricted to rocks whose behaviour
is not significantly different f r o m that o f a h o m o g e n e o u s , APPARATUS
perfectly elastic medium. 1
(c) The state of stress and pre-existing fracturing o f the Drilling e q u i p m e n t , including
rock at the measurement locations should be such that 2. (a) A drill and N X C (86 m m dia) or larger core
it is possible to obtain relatively tong (approx. two to barrels and bits, to advance the drillhole to the required
three times the diameter o f the overcore hole) pieces o f depth and location o f measurement. W h e n drilling from
u n b r o k e n core.-' u n d e r g r o u n d workings, the drill m a y need to be fitted
 ISRM: SUGGESTED METHODS FOR ROCK STRESS DETERMINATION 69

DRILLHOLE
t

p I L

COLLAR OF ~ ~/" ~
1

- GAUGES

'K X HORIZONTAL

Fig. 11. Strain gauge configuration for each rosette ~iewed from axis
Fig. 10. Position of the three rosette gauges. of drillhole.

with exhaust control provisions (e.g. scrubbers, etc.) and This arrangement of gauges yields an additional three
anti-spark provisions if the underground workings to six strain measurements which may be used for
present a gas ignition hazard. cross-checking or for error estimation.
(b) Coring bits and centralizing guides, to allow (c) Each individual strain gauge should have a gauge
drilling of a pilot hole (38 mm dia) into the end of the length of at least 10 ram. s
overcore hole and concentric with it.
3. A "strain cell" designed to glue a number of strain Installation equipment, including
gauges onto the sidewall of the pilot hole. The strain cell 4. (a) An installing tool into which the strain cell (Fig.
should meet the following requirements: 3'4 12) can be plugged to facilitate electrical connection of
(a) To avoid difficulties caused by the non- the various strain gauges to a multi-conductor cable and
homogeneous, anisotropic nature of most rocks, the t o hold the strain cell while it is inserted into the pilot
electric resistance strain gauges should be glued as close hole. The installing tool should also contain a device for
together as possible so that only a small volume of rock orienting the strain cell and another device (usually gas
is involved] operated) for pushing the strain gauge rosettes out into
(b) The strain gauges should be mutually orientated in contact with the sidewalls. The dimensions of the install-
such a way that six independent strain measurements ing tool should be such that it can be easily pushed into
may be obtained, (e.g. using three standard strain gauge the pilot hole. A sufficient number of installing rods and
rosettes, each consisting of three or four gauges in an a length of electrical cable and gas supply hose should
arrangement such as shown in Figs 10 and 11). be included.

ROSETTE ~LUG . , ~ ~ ~ ~. COVER UNiT

G U IOES '.~/ ~ - - ,

J'q , - ELECTICAL

X / l ~,

GLUED TO D U M M Y GAUGE STRAIN GAUGE ROSETTE

WHEN IN US = ~- y CONSISTING OF THREE PLUGS

2., ,,._ ;
: D U M M Y G&LGE UNIT
PLUGGED ~%TO BODY UNIT

COVER UNIT

Fig. 12. Exploded view of the CSIR triaxial strain cell.

70 ISRM: SUGGESTED METHODS FOR ROCK STRESS DETER.MISATIOX

(b) Cleaning materials and tools to allow the pilot hole horizontal to facilitate drainage of all drill and other
to be dried (e.g. by spraying with alcohol) and primed water from the point of installation of the cell.
to give good bonding with the strain gauge adhesive. (b) The end of the hole is flattened with the facing bit.
(c) A plug to seal the pilot hole during overcoring. (c) A 550 mm long pilot hole is drilled into the end of,
and concentric with, the 90 mm drillhole (Fig. 13b).
Measurement equipment, including (d) If the core obtained from drilling the pilot hole is
5. (a) One or several strain bridges capable of in one piece and is free from planes of vveakness, the
measuring to the nearest 5 x 10-6mm/mm. measurement can be carried out. The pilot hole core
(b) Electrical connectors and switches so that the must be carefully marked for identification later. If the
strain on each gauge may be measured by a strain bridge. core is broken, the 90mm drillhole should be drilled
(c) Temperature monitoring device (e.g. thermocouple about 0.5 m deeper and the steps in paragraphs 7b, c and
or thermistor) to monitor the temperature to the nearest d repeated until a position is found where solid un-
1°C at the sensor location when initial readings are broken core can be obtained. Intervals with fractured
taken. pilot hole core should be examined with a drillhole
6. Appropriate solvent(s), primer(s) and strain gauge periscope or TV camera to determine the extent and
adhesive(s) for the rock conditions. Strain gauges must severity of fracturing, before rejecting a potential test
remain securely bonded to the rock throughout the zone.
drilling and measuring phases for the results to be
considered valid. Installation
8. (a) The pilot hole is Rushed clean with water
through a hose or hollow drill rods. After washing, the
PROCEDURE6 part of the drillhole where the strain cell is to be
cemented is dried (e.g. with solvent) and sprayed with a
Drilling primer suitable for the adhesive to be used.’
7. (a) A hole, approx. 90 mm dia, is drilled into the (b) The strain cell is plugged into the installing tool
rock mass to the point where the stress in the rock is to which is connected by means of a multi-core cable to a
be measured, as shown in Fig. 13a. The hole should be strain bridge. The rosette gauges are carefuily cleaned
drilled as straight and as accurately as possible by using with a solvent, and a check is made to ensure that all
a long core barrel (minimum length 1.5 m) and the cores gauges are in working condition and that a balance on
should be recovered and fully logged. The hole is the bridge may be obtained.
generally drilled from underground at slightly above the (c) The operation of the strain cell is checked by
rotating the tool to make sure &hat the orienting device
is in order and, by determining if all three gauges are
pushed out, when the appropriate controls are activated.
(d) A 2- to 3-mm thick layer of adhesive is smeared
on the surface of each rosette gauge.
(e) The installing tool, with the strain cell plugged into
it, is pushed up the large diameter drillhole and into the
pilot hole by means of installing rods. Once inside the
pilot hole, the strain cell is oriented by rotating the rods,
and the gauges are then pushed into contact with the
sidewalls of the hole. Pressure on the gauges is main-
tained for sufficient time to allow full setting of the
adhesive.*

Measurements and ouercoring

9. (a) Initial strain readings are taken; the installing
tool is removed from the hole; and the pilot hole is sealed
by means of the drillhole plug.
(b) When taking readings, a balance oneach gauge
should be made twice and the average of the two
-readings should be used in the calculations. At least two
complete rounds of readings should be taken. A typical
field data sheet is shown in Fig. 14.
(c) The strain cell is then overcored and the core
containing the cell is removed from the hole. This
relieves the stresses in the rock to which the strain gauge
rosettes are attached. Further rounds of readings are
taken, as described in paragraph 9b, to determine a
Fig. 13. Illustration of the use of the CSIR triaxial strain cell. stable level of post-overcore strain for each strain gauge.
 71
ISRM: SUGGESTED METHODS FOR ROCK STRESS DETERMINATION

Hi STRESS MEASUREMENT- OVERCORING RESULTS

?,11%E LOCATION DATE

~OLE C_a*a NO ROCK TEMP

TEST N O DR~LL W A T E R T E M P

HOLE BE,~,RING COLLAR COORDS

HOLE DIP - r y e up i

EX HOLE L E N G T H D E P T H TO EX C O L L A R

CELL C L P O S I T I O N OVERCORE DIAMETER

CELL D A T A CELL N O _ _ HI - - GAUGE FACTOR _ _

SETTING DATE ~ME HRS

OVERCORE DATE ]]ME _ _ HRS

GREEN. BROWN = COMMON LEADS

GAUGE OUTPUT (MICROSTRAIN

9
I
WIRE COLOUR BLACK O R A N G E ! PURPLE GBLUE-
REEN WHITE i YELLOW RED GREY BLUE

REMARKS i A,O'I, ACirc~ i A,45:~ i BI45 ~ , B(135% BCirc~ C,GI, C{Circ) Ct45 ~ }

' i !
INITIAL R E A D I N G [ i

5 MINUTES DRILLING I
I i
WATER ! I
10 M I N U T E S D R I L L I N G
WATER
[
I i
'
I
COMMENCEMENT OF : i
DRILLING cm !
: i
om
!
[ !

i i
J ] !I ,
5 M t N U T E S A F T E R OVER.
CORE COMPLETE cm i i
!
i
I 10 M I N U T E S A F T E R OVER- ; i
CORE COMPLETE cm i [
! =
FINAL STRAIN CHANGES {
FOR D A T A R E D U C T I O N ] [

Fig. 14. Field data sheet recommended by CSIRO.

10. Measurements at greater depth m a y then be made isotropic. In cases where there is any indication that the
as required by extending the large diameter drillhole and rock is significantly anisotropic, an alternative data
repeating the operations in p a r a g r a p h s 7b to 9c. reduction method, described by Amadei [1], should be
considered.

E = A.~xG~ + A,yG, + A..a: + A>.:'rs: + A.~:r~: + A.,.vE~y.

CALCULATIONS
The A-coefficients being
11. (a) Strain relief values are c o m p u t e d by subtracting 1-v l+v
average readings before and after overcoring. Axx - 2~ 2E [cos 2w - (1 - v 2)
(b) The reference system used in these "Suggested
M e t h o d s " has the z-axis along the drillhole axis (positive x (1 - c o s 2 w ) c o s 20]
in the direction of drilling) and the x- and y-axes l-v l+v
perpendicular to the z-axis to form a right-handed A)y -- - 2E
- 2E [cos2w - (1 - v:)
system.
(c) The position of a rosette gauge on the wall of the x (1 - cos 2w) cos 20]
drillhole is identified by its azimuth from the axis
A~ 1-v [l+v )
(positive counterclockwise as viewed from the collar of - ~ ~-L-~-~- c ° s 2 w
the drillhole--see Fig. 10). The position of a strain gauge /

within a rosette is identified by the angle between the 2(1 + v) .

A,~ - - -E sm (2w cos 0)
z-direction and the direction of the strain gauge itself
(positive counterclockwise as viewed from the axis of the - 2 ( 1 + v) .
drillhole). A.~_.- E sin (2w sin 0)
(d) The following equation is used to relate the
measured strains to the c o m p o n e n t s of the stress tensor, - 2 ( 1 - v 2)
Ax, = E - (1 - cos 2w) sin 20
a.,, av, a:, r~, ~_., r.,:, in cases where the rock is reasonably
 71 ISRM: SUGGESTED METHODS FOR ROCK STRESS DETER\iISATIOS

where )t’ and B are defined in Figs 10 and I 1, E is the another method, requiring no core or for uhich shorter
Young’s modulus and 1?is the Poisson’s ratio of the rock. pieces of core would suffice. should be used.
These values are normally obtained by laboratory testing 3. The “triaxial strain cell.” devleoped by the South
of specimens taken from the cores, using the ISRM African Council for Scientific and Industrial Research
“Suggested Method for Laboratory Determination of (CSIR). fulfills the requirements of these “Suggested
Rock Deformability.” Alternatively. the elastic con- Methods”. Details of gauge orientation, etc. are specific
stants can be obtained from a biaxial or triaxial test of to a given model of strain cell and are supplied by the
the core containing the strain cell. manufacturer to assist in stress computation.
(e) Tensor components, G.~,fly, err,T,,. TV:and Tcr.which 4. The procedure and calcuIations described in these
best fit the measured data, are obtained by means of a Suggested Methods” are based on the triaxial strain cell
least squares fitting procedure. identified in Note 3. Another device that is similar in
(f) Principal stresses, Go, G?and g3, and their direction concept and fulfills the requirements of these “Suggested
cosines are then to be calculated from the tensor com- Methods” has been developed by the Commonwealth
ponents of best fit by means of elasticity equations, Scientific and Industrial Research Organization
usually with the aid of a computer (Vreede [ll]). (CSIRO) of Australia. This device, shown in Figs 15 and
16 and known as the CSIRO Hollow Inclusion (Hi) cell.
is available commercially. It consists of three, three-
REPORTING OF RESULTS component strain gauge rosettes encapsulated in a thin-
walled epoxy pipe. One of the major differences between
12. The report should include the following general
the CSIR cell and the CSIRO HI cell is that the CSIRO
information:
HI cell is permanently attached to the readout cable.
(a) The locations, directions and length of drillholes.
Thus, stress relief data is obtained and monitored during
(b) A geotechnical log of core from each drillhole
the overcoring process.
showing the depths of measurement, with particular
5. The maximum length of the strain gauge is re-
attention given to geological and structural character-
stricted by the size of the drillhole; 10 mm is the mini-
istics of the rock at locations of stress determination.
mum length that should be used.
(c) A description, illustrated with diagrams and photo-
6. The procedure described here is based on the
graphs, of the procedure and equipment used. Reference
triaxial strain cell mentioned in Note 3. If another
may be made to these “Suggested Methods,” noting
instrument is used, drillhole sizes and other dimensions
departures from the recommended equipment and
should be changed to suit this instrument.
procedures.
13. The report include the following detailed depth
information for each location of measurement:
(a) A tabulation of orientation and strain relief read-
ings for each gauge.
(b) The Young’s modulus and Poisson’s ratio values
used in the calculations and an account of how these
were determined.
(c) The six computed components of the strain relief
tensor at that location to the nearest 0.1 MPa.
(d) Standard deviation and error estimates for the
regression calculations described in paragraph 1le.
(e) Computed principal stress magnitudes and direc-
tions referred both to the drillhole axis system (para-
graphs 1lb and c) and to the global axis for the project.
(f) Test results showing substantial discrepancies with
other data and giving possible or probable explanations
of the causes.

NOTES

1. Actual rock conditions on site seldom conform to

the idealized conditions assumed for purposes of stress
computation. Whether the idealized assumptions of elas-
ticity are sufficiently close to actual conditions for stress
measurements to be worthwhile must be determined by
the project engineer. Where anisotropic modulus condi-
tions exist those strain cells containing 12 strain gauges
provide a more suitable analysis system.
2. If the rock at a measuring site is such that long
unbroken pieces of rock core cannot be obtained, then Fig. 15. Schematic layout of CSIRO hollow inclusion (HI) cell.
 ISR\!: S U G G E S T E D M E T H O D S FOR R O C K STRESS D E T E R M I N A T I O N 7;

Fig. 16, CSIRO hollow inclusion {HI) cell.

7. Water is normally used for cooling and flushing ments, Lectures Notes in En~it, eerin~, Vo[ 2, 4-8 pp. Springer,
during drilling. The surface of the hole is, therefore, wet New York I1983).
2. Duncan Fama M. E. and Pender M. J. Analysis of the hollow
alter drilling and very few strain gauge cements adhere inclusion method for measuring in situ rock stress, htt, J. Rock
to wet rock. To overcome this problem primers are Mech. $lin. Sei. & Geomech. ,4bstr. 17(3), [37-I46 (1980).
available such as Silane Coupling Agent No. A1120 or 3. Gray W. M. and Toews N. A. Optimization of the design and use
of the CSIR triaxiaI strain ceil for stress determination, field
A1 I00 from Union Carbide Corporation. A mixture of testing and instrumentation of rock. ASTM STP 554, pp.
10% (by volume) of sitane in alcohol has, for example, 1l(~134, American Society for Testing and Matenals, Philadel-
been used as a primer. phia (1974).
4. Hiltscher R. Beitrag Zur gebirgsspannungsmessung nach dem
8. The adhesive to be used depends on the conditions bohrlochboden-cntspannungsverfahren. Pro& h;t. Syrup. on the
prevailing at the site where measurements are carried Determination qf Stresses in Rock Masses. pp. 200 229, Lisbon
out. Any standard strain gauge adhesive can be used (1971).
5. Leeman E. R. The measurement of stress in rock--I, II, tII. Jl S.
provided that it bonds well to the rock and to the gauges. AJh. bzst. Min. Metall. 65, 45-I 14 and 254 284 ¢I964).
The best adhesive normally' has to be found by trial and 6. Leeman E. R. The CSIR doorstopper and triaxiat rock stress
error. measuring instruments. Proc. 1SRM Syrup. on the Determination
q/" Stresses in Rock Masses, pp. 578-616, L.N.E.C., Lisbon
A quick-setting adhesive, for which the pot life after (1969).
mixing is less than 5 min, can be used when the installa- 7. Leeman E. R and Hayes D. J. A technique for determining the
tion time is shorter than 5 rain. This means that the complete state of stress in rock using a single borehole, Proc. 1st
hzt. Congr. on Rock ),lechanics. VoI. lI, pp. 17 24. Lisbon (1966).
measurement drillhole must be short ( 1 0 m or less) and 8. Rocha M., Silverio A., Pedro J. O. and Delgado J. S. A new
the rock temperature should be below 15~C. The follow- development of the LNEC stress tensor gauge. Proc. 3rd Int.
ing quick setting adhesives may, for example, be used Congr. on Rock Mechanics, Vol. 2A, pp. 464-46-. Denver (1974).
9. Stepanv V. and Batugin S. Assessing the effect of the anisotropy
under these conditions: Philips strain gauge cement type of rocks on the accuracy of stress determination by the relief
9244 04; Hottinger quick-setting strain cement type method. Societ Min. Sci. 3, 312 315 {1967).
X-60. 10. Van Heerden W. L. Practical application of t?,e CS1R triaxial
strain cell for rock stress measurements. Proc. l S R M Syrup. on
Epoxy, a slow-setting adhesive, can, for example, be the lnz'esti~ation of Stress itt Ro('k--A~h'ance, ',n Stress ),[ea-
used when measurements are carried out in longer surement, pp. 1~5, S?dney (1976).
drillhotes and where the rock temperature is above 11. Vreede F. A. Instruction manual for the use of the CSIR triaxial
rock stress measuring equipment. CSIR Rept ME 1763, Pretoria
1 5 C (the setting time is reduced as the temperature is (1982).
increased). 12. Worotnicki O. and Walton R. d. Triaxiat hollo~ inclusion gauges
{CSIRO) for determination of rock stresses in ;itu. Proc. I S R M
METHOD 4: B I B L I O G R A P H Y Syrup. on the lncest~ation o[Stress in Rock--.qdcanees in Stress
Measurement, Supplement, pp. 1-8, S,,dney {1976)
1. Amadei B. Rock anisotrop5 and the theory of stress measure-