ISRM International Symposium 2008

5th Asian Rock Mechanics Symposium (ARMS5), 24-26 November 2008 Tehran, Iran

APPLICATION OF AE FOR DETERMINING IN SITU STRESS AT UG MINE

S. KRAMADIBRATA1, R.K. WATTIMENA1, I. KAMIL1, M.A. RAI1 and E. WIDIJANTO2

1
Department of Mining Engineering, Institut Teknologi Bandung, Indonesia
(e-mail of corresponding author: [email protected])
2
PT. Freeport Indonesia

Abstract
The stability and safe operation of any underground excavations has been the prime concern of underground mine engineers. One
of the most important parameters to obtain for underground mine planning is the in-situ principal stresses determination. The
problems with the in-situ stress determination, particularly in Indonesia, so far have been associated with technology availability,
cost efficiency and complexity in performing the in-situ test in remote regions. However, an indirect method with less cost to
determine the in-situ principal stresses which is attributed to the use of Acoustic Emission (AE) method has been used by many
and proven to be reliable as experienced at an underground gold mine in Indonesia. This paper therefore emphasizes on the
technique development of AE test on 6 cored rock specimens retrieved from a reasonably large size core rock sample with
different directions encompassing three-component principal stresses. The large size core rock sample had orientation of N58OE/-
4O and was drilled from the AB Tunnel of the Underground copper mine of PT. Freeport Indonesia. Based on the stress level at
which the Kaiser effects were detected the in-situ stresses at the AB tunnel was subsequently predicted as follows σ1 =41.4 MPa;
σ2 = 18 MPa; σ3 = 3.1 MPa.
Keywords: in-situ stress, stress memory, acoustic emission, Kaiser Effect

1. Introduction creating induced stress, is to take advantage of

the so-called Kaiser Effect of acoustic emission
Underground mining engineers have been very
(AE). This phenomenon, termed Kaiser Effect
much concerned with the safe operation of any
suggests that previously applied maximum stress
underground excavations. Prior to making up
might be detected by stressing a rock specimen
underground mine planning in-situ principal
to the stress level where there is a substantial
stresses acting at the interest region should have
increase in AE activity. Based on the Kaiser
been obtained. However, the problems with the
Effect, the previous stress can be estimated from
in-situ stress determination, particularly in
the curve of AE activity under monotonically
Indonesia, so far have been associated with
increasing stress, which is referred to the
technology availability, cost efficiency and
Uniaxial Loading Method (ULM).
complexity in performing the in-situ test in
remote regions. AE is concerned with a transient elastic wave
energy generated by rapid micro-crack growth
Underground excavation will obviously create within a material. The micro-crack propagation
induced stress in its near field domain that will is a ubiquitous phenomenon associated with
be greater than the original in-situ stress, and brittle fracture, and has provided ample evidence
even will also alter the stress direction. In the last regarding the failure process in rock. It is also
three decades, in-situ stress measurements have well understood that micro-crack begins to occur
been very important issues in the rock mechanics as the stress applied on a rock specimen exceeds
field in Indonesia, since the values obtained from its bearing capacity.
these measurements can help the rock mechanics
The Kaiser effect in rocks were first
engineers to analyze the stability of an
confirmed by [1] and followed by [2]. A decade
underground excavation in a better confidence so
later this AE technique was first applied to
that safe operation can be achieved.
measure in-situ rock stress [3]. Application of
An alternative method for determining the AE to predict in-situ stress has been done at the
stress-state at depth, remote areas, and without Laboratory of Geomechanics & Mine Equipment

379
of the Department of Mining Engineering A gravitationally loaded rock mass in which
(DME) of the Faculty of Mining Engineering no lateral strain was permitted during formation
and Petroleum Engineering (FMEPE) of Institute of the overlying strata, the value of k is
Technology of Bandung (ITB) since 2001 [4, 5, independent of depth and is given by [ν/(1-ν)],
6]. The investigations demonstrated that AE where ν is the Poisson's ratio of the rock mass.
could be used to reasonably well predict the in- This relationship was widely used in the early
situ stress of an underground opening as long as days of rock mechanics but it proved to be
retention time is taken into account. Thus, a inaccurate and is seldom used today [7].
study was then conducted to determine the A state of stress at a particular point on a
magnitude and orientation of in-situ stresses at principal plane can be expressed in the following
AB Tunnel of PT Freeport Indonesia, Papua matrix:
Indonesia.
⎡σ1 0 0⎤
2. In-Situ Stress [σP] = ⎢⎢ 0 σ 2 0 ⎥⎥ (3)
Rock at depth is subjected to stresses resulting ⎢⎣ 0 0 σ 3 ⎥⎦
from the weight of the overlying strata and from
locked in stresses of tectonic origin. As where,
implicitly explained earlier, that when an σ1: Major Principal Stress;
opening is excavated in this rock, the stress field σ2: Intermediate Principal Stress;
is locally disrupted and a new set of stresses are
induced in the rock surrounding the opening. σ3: Minor Principal Stress
Knowledge of the magnitudes and directions of Determination of the magnitude and direction
these in-situ and induced stresses is an essential of principal stress can be done using the
component of underground excavation design as, Invariant theory and especially their directions
in many cases, the strength of the rock is are estimated based on the direction cosine of
exceeded and the resulting instability can have each principal stress. The direction cosine of axis
serious consequences on the behavior of the x, y and z for each principal stress σi (i = 1, 2, 3)
excavations. must then accordingly satisfy the orthogonal
Consider an element of rock at a depth of H requirements.
below the surface. The weight of the vertical In order to determine the stress level at which
column of rock resting on this element is the the Kaiser effect is detected during the test (σKE),
product of the depth (H) and the unit weight (γ = equation (4) is then applied.
ρg) of the overlying rock mass. Hence, the
⎧σx ⎫
gravitational or vertical stress (σv) on the ⎪σ ⎪
element is estimated from the simple ⎪ y⎪
relationship: ⎪⎪ σ z ⎪⎪
[σKE] = [ lx2 ly2 lz2 2lxly 2lxlz 2lylz] ⎨ ⎬ (4)
σv = γH (1) ⎪τ xy ⎪
⎪ τ xz ⎪
The horizontal stresses (σH) acting on an ⎪ ⎪
element of rock at a depth H below the surface ⎩⎪ τ yz ⎭⎪
are much more difficult to estimate than the
Calculation of the in-situ stress of the rock
vertical stresses. Normally, the ratio of the
mass [σ] is completed based on six equations of
average horizontal stress to the vertical stress is
(5) resulted from the acoustic emission test.
denoted by the letter k such that:
σH = k.σv = k.γH (2)

380
 ⎡2
⎧ σ x ⎫ ⎢ l x1 l2y1 l2z1 2l x1l y1 2l x1lz1 2l y1lz1 ⎤ ⎧σ as the stress applied on a rock specimen exceeds
⎥ KE1 ⎫
⎪ σ ⎪ ⎢l 2 l2y2 l2z2 2l x2l y2 2l x2lz2 2l y2lz2 ⎥ ⎪σ KE2 ⎪ its bearing capacity.
⎪ y ⎪ ⎢ x2 ⎥⎪ ⎪
⎪⎪ σ z ⎪⎪ ⎢l2x3 l2y3 l2z3 2l x3l y3 2l x3lz3 2l y3lz3 ⎥ ⎪⎪σ KE3 ⎪⎪ Determination of the in-situ stresses is
⎨ ⎬=⎢2 ⎨ ⎬ (5)
⎪τ xy ⎪ ⎢l x4 l2y4 l2z4 2l x4l y4 2l x4lz4 2l y4lz4 ⎥ ⎪σ KE4 ⎪ analogous to cyclic pressure vessel testing in the
⎥
⎪ τ xz ⎪ ⎢l2 l2y5 l2z5 2l x5l y5 2l x5lz5 2l y5lz5 ⎥ ⎪⎪σ KE5 ⎪⎪ mechanical/metallurgical field. Pressure vessels
⎪ ⎪ ⎢ x5 ⎥
⎪⎩ τ yz ⎪⎭ ⎢l2
⎣ x6 l2y6 l2z6 2l x6l y6 2l x6lz6 2l y6lz6 ⎥⎦ ⎪⎩σ KE6 ⎪⎭ only emit AE after the previously applied
pressure is exceeded (Figure 2). This
Transformation of the direction cosine of the phenomenon is termed the Kaiser Effect and is
principal stress into the dip direction and dip can defined in ASTM E 610-77 as follows; “Kaiser
be made using the equations (6), (7) and (8) of Effect – the absence of detectable acoustic
the spherical coordinate system (Figure 1). emission until previously applied stress levels
are exceeded.”
λx = λ cosθ cosΦ (6)
λy = λ cosθ sinΦ (7)
λz = λ cos(90-θ) (8)
where Φ = dip direction and θ = Dip.

Fig. 2. Kaiser Effect

The AE activity is detected by use of a sensor

that is sensitive to the “sounds” (actually
transient elastic waves) emitted by material or
rock as it is being stressed. This sensor converts
the mechanical signal into an electrical signal,
which is the pre-amplified, filtered, post
amplified and counted. The typical end result is
what ASTM E 610-77 defines as the “acoustic
emission count” that is the number of times the
AE signal amplitudes exceeds a present
Fig. 1. Spherical coordinate system (r, Φ, θ). threshold during any selected portion of a test.
There are three elementary microscopic
3. Acoustic Emission mechanisms taking place during rock
deformation whereby the origin of AE can be
AE is concerned with a transient elastic wave identified [8] such as plastic deformation of
energy generated by rapid micro-crack growth constituent materials due to dislocation motions
within a material. The micro-crack propagation such as glides and climbs; sliding along pre-
is a ubiquitous phenomenon associated with existing crack surfaces and sudden advances of
brittle fracture, and has provided ample evidence crack tip.
regarding the failure process in rock. It is also
Based on this model Kaiser Effect is
well understood that micro-crack begins to occur
interpreted as follows; in elastic stage a crack,

381
pre-existing or newly formed, grows to an transmitting elastic wave and be well detected by
equilibrium configuration corresponding to the the two AE piezoelectric tranducers.
applied stress and this stress would be
memorized in the rock through the crack
configuration. The opened crack would be closed
after unloading, but not healed in an atomic
sense until the next loading. The growth process
of cracks is only responsible for AE; hence no
AE should be expected below the highest stress
in the former loading cycles. Recovery process
would correspond to the healing of pre-existing
cracks. It is therefore appropriate to define that,
the stress at which the laboratory Kaiser Effect is
observed is the previous peak stress.

4. Experimentation and Test Results

4.1. Core Rock Sample

The AE tests were performed at the Laboratory
of Geomechanics & Mine Equipment of the
DME of the FMEPE of ITB. Prior to the tests,
orientation core drilling and sample preparation
were done and followed by determination of
physical and mechanical properties of the rock
samples.
A reasonably big size of fine Diorite core rock
sample was obtained from AB Tunnel of PT Fig. 3. Core drilling orientation grip device, orientated core
Freeport Indonesia, Papua. The sample was drilling and orientation of cored rock samples for AE tests
subsequently drilled at various orientations in Table 1. Physical properties of AE rock samples.
order to avail 6 samples representing three
Sample Natural Water Porosity Void
component directions and the orientation of each Code density content % Ratio
sample is depicted in Figure 3. gr/cc %
The orientated drilling was facilitated with a Average 2.71 0.86 2.99 0.03
tool so called core drilling orientation grip Std Dev 0.08 0.08 0.22 0.00
device which was designed and manufactured in-
house (see Figure 4) and the rock sample UCS tests on 3 rock samples which were not
preparation was carried out in compliance with intended for the AE test were carried out to
the ASTM D.3967-86 and or ISRM (1985). obtain the values of UCS, elastic limit, stress
level of closing crack, Young’s Modulus and
4.2. Physical and Mechanical Properties Poisson’s Ratio. The average of those values is
The average values of the physical properties as follows: σc = 127.39 MPa; σE = 100.32 MPa;
tests of 4 (four) rock samples reveal that they can σcc = 27.07 MPa; E = 15.56 GPa and ν = 0.18. It
be considered as average solid rock type as can then be said that the rock samples are
indicated in Table 1. It is therefore not unusual to classified as medium strength [9] and have
expect that the AE test would reveal good relatively consistent stiffness.

382
4.3. Experimentation of AE Test Having the AE tests carried out, the Kaiser
Effect for each sample was detected at various
Equipment used for the AE test includes
cycles and from which the stress level was
universal compression testing machine and AE
recorded (σKE). This stress level corresponds
hardware such as two AE piezoelectric
with the stress history or in-situ stress at which
tranducers, Pre-Amplifier and software program
the sample was retrieved from. The
MISTRAS 2001. The AE test was performed by
representative σKE is obtained from taking
applying ULM on a rock sample. The load was
average of different cycles for each sample. The
given within 8-10 cycles and varied depending
values of the average in-situ stress σKE for each
on the rock strength. However, it should be
sample including their orientation are given in
borne in mind that each load or stress should not
Table 2.
exceed the average value of UCS and supposedly
within the elastic range namely between closing Table 2. The average values of σKE and their orientation
of each rock sample
cracks and yield point in which area permanent
deformation does not occur should the load is Sample
DD/D
σKE Sample
DD/D
σKE
released back to zero. The test results were code MPa code MPa
AE-01 329/00 32.24 AE-04 282/32 31.98
recorded as curves of AE activity (hits) vs. stress
AE-02 59/05 17.50 AE-05 107/39 12.65
(MPa). AE-03 239/85 12.81 AE-06 11/39 31.45
The MISTRAS 2001 AE detection and Note :DD: Dip Direction, D: Dip
analysis computer software was used to log the
stress and AE signals transmited by a load cell 5. Determination of Magnitude and
incorporated to the UCS test frame, and two Orientation of the In-Situ Stress
piezoelectric tranducers respectively. Two pre- The direction cosine for each axes of stresses on
amplifiers were hooked up between the two Cartesian coordinate system (x, y, z) for each
piezoelectrics and computer. The two tranducers rock sample are given in Table 3.
were glued at the middle of the long sides,
opposite each other, of the test specimen. The Table 3. Direction cosines of orientated rock sample
gains of preamplifier and the computer software Direction Direction
Sample Sample
were at 40 dB, whereas the the threshold gain at Cosines Cosines
the computer software was set at 45 dB. Range x 0.8571 x 0.1763
AE-01 y -0.5150 AE-04 y -0.8295
of 100-1200 kHz was used for the frequency z 0 z 0.5299
filter. Illustration of the arrangement of the x 0.5131 x -0.2272
acoustic emission test is depicted in Figure 4. AE-02 y 0.8539 AE-05 y 0.7432
z 0.0872 z 0.6293
x -0.0449 x 0.7629
AE-03 y -0.0747 AE-06 y 0.1483
z 0.9962 z 0.6293
Note: all (x2+y2+z2) = 1

Having used the data obtained from Table 3

into the equation [5] and [3] with a further
mathematical exercise the matrix tensor for the
in-situ stress can subsequently be predicted as
follows,

Fig. 4. The arrangement of the acoustic emission test

383
 ⎡ 25.0 − 8.4 14.8 ⎤ difference apparently is not uncommon as some
[σ] = ⎢⎢ − 8.4 24.3 − 6.3⎥⎥ MPa (9) factors influencing the difference, particularly in
the area of PT Freeport mine would be
⎢⎣ 14.8 − 6.3 13.2 ⎥⎦ geological structures which may have affected
the in-situ gravitational (vertical) stress.
The invariant stresses can be calculated based
on the in-situ stress [σ] data and afterwards Using the equation from [7], the k value in
checking process to satisfy the criteria given by Eq. (2) for depth less than 1000 m, can be
equation (8) is performed and the final in-situ estimated as
principal stresses are given as follow, 100 1500
+ 0.3 < k < + 0.5 (11)
σ1 = 41.4 MPa; σ2 = 18.0 MPa ; σ3 = 3.1 MPa. z z
The in-situ principal stresses can be written in which gives 0.43 < k < 2.5.
matrix form as given in equation (10):
If the rock mass condition is deemed elastic
⎡41.4 0 0⎤ (Eq. (2)), the σH will be in the range of 8.72 MPa
[σprincipal] = ⎢⎢ 0 18.0 0 ⎥⎥ MPa
– 50.69 MPa. Meanwhile the horizontal stresses
(10)
obtained from Eq. (9) are 25.0 MPa and 24.3
⎢⎣ 0 0 3.1⎥⎦ MPa. It is therefore obvious that the horizontal
stresses obtained from the calculation are within
The dip direction of each principal stress is the range that of theoretical approach.
determined and the final form of the principal
stresses is given in Table 4. Referring to [10], the ratio of (σHaverage/σv) at
depth of 751 m would be 1.606 and that of
Table 4. Principal stresses and their orientation estimated from acoustic emission is 1.876
Dip meaning that the horizontal stress is always
Principal Stress greater than that of vertical stress.
Direction Dip
Stress (MPa)
(N …°E) Other influence factors which actually could
Maximum 41.4 323.4° 28.9° have been avoided would be the absence of
Intermediate 18 62.2° 15.5° servo control in the AE tests that led to the
imprecise constant loads, noises produced during
Minimum 3.1 356.9° 56.5° the tests were high which made the Kaiser Effect
could not be observed clearly and waiting time
Requirement of the orthogonal system for the between sample collecting and the AE tests is
three component principal stresses is clearly defined as retention time. The retention time is a
conducted and having done this exercise the period during which the disappearance of stress
requirements are satisfied. memories in the samples may takes place and
this would depend on the rock type, grain size
6. Discussion isotropic, homogeneity and elasticity.
As explained earlier that the fine grain diorite Some rock types apparently do not appear to
rock sample was obtained from AB Tunnel of possess significant relationship between
PT. Freeport Indonesia Papua at depth of 751 m. retention time against the occurrence of the
Taking account the average density of the rock Kaiser Effect. The occurrence of the Kaiser
mass above of 2.71 t/m3, the theoretical vertical Effect of fine texture sandstone remains the same
stress at this point would be 20.352 MPa. On the for 7 year retention time [11], whereas the
influence of retention time on the occurrence of
other hand, the vertical stress (σz - see Eq. (9))
the Kaiser Effect on very fine sandstone and coal
obtained from the calculation is 13.2 MPa, which
is only obvious if the retention times are less
is only about 65% of the theoretical one. This
than a year (12] and 2 year respectively [11].

384
Granite which is much coarser than sandstone 323.38° E/28.93°; σ2 = 18 MPa, N 62.18° E/
and coal has only good retention time of 50 days 15.47°; σ3 = 3.1 MPa, N356.86° E/ 56.5°.
[13]. Tuff breccias and lythic tuff which are
obtained from the Pongkor underground gold Acknowledgments
mine have very short retention time – less than a Acknowledgements are due to students who did
week [5]. the serial tests of the acoustic emission and Mr.
Having learned from the foregoing evidences Sudibyo, Mr. Iwan and Mr. Suparman the
that the AE test would reveal good transmitting technicians of the Laboratory of Geomechanics
elastic wave and the rock samples are classified & Mine Equipment of the DME of the FMEPE
as medium strength and have relatively of ITB who were heavily involved in this
consistent stiffness, the influence of retention research project.
time on the Kaiser Effect occurrence for the fine
The authors would also like to thank Prof. K.
grain Diorite core rock sample can therefore be
Matsui from Kyushu University Japan and his
deemed the same to that of the fine texture
colleagues for their donation of the acoustic
sandstone meaning that the in-situ stresses
emission equipment set and MISTRAS 2001 to
determined from AE test could have been
the Laboratory of Geomechanics & Mine
appropriately accepted. Nevertheless, it is
Equipment of the DME of the FMEPE of ITB.
therefore appropriate to note at this point in time
that the outcome from the AE method warrants References
further clarification should this method is applied
to any rock samples which are obviously not 1. Mogi, K. 1962. Study of the elastic shocks
very fine grain, heterogeneous and plastic caused by the fracture of heterogeneous
materials. materials and its relation to earthquake
phenomenon. Bull. Earthquake Res. Inst., Univ.
7. Closing Remarks Tokyo, 40, 125-173.
2. Goodman, R. 1963. Subaudible noise during
The retention time apparently does not apply to compression of rocks. Geological Society of
the Diorite core rock sample or in another word; American Bulletin, 74, 487-490.
the in-situ stresses determined from AE test 3. Kanagawa, T., M. Hayashi, and H. Nakasa.
could have been appropriately accepted. 1976. Estimation of spatial geostress components
The stress levels that have been experienced in rock samples using the Kaiser effect of
by the intact rock samples obtained from the acoustic emission. Rep. No. 375017, Central
following boreholes N 329° E / 0°; N 59° E / 5°; Res. Inst. Of Electrical Power Industry, Abiko,
N 239° E / 85°; N 282° E / 32°; N 107° E / 39°; Japan.
and N 11° E / 39° are 32.24 MPa, 17.5 MPa, 4. Kramadibrata, S., H. Timbul, and M.A. Rai.
12.81 MPa, 31.98 MPa, 12.65 MPa, and 31.45 2001. Study of in-situ stress determination by
MPa respectively means of acoustic emission of uniaxial
The in-situ stress tensor matrix of the rock compression test. Proc. 3rd Asian Symposium on
mass is expressed as follow, Engineering Geology and the Environment -
Natural Resources Manage-ment for Regional
⎡ 25.0 − 8.4 14.8 ⎤ Development in Tropical Area, Jogjakarta,
[σ] = ⎢⎢ − 8.4 24.3 − 6.3⎥⎥ MPa Indonesia.
⎢⎣ 14.8 − 6.3 13.2 ⎥⎦ 5. Kramadibrata S., R. Bastiawarman, and K.
Matsui. 2004. The influence of retention time on
The magnitude and direction of the in-situ AE test of intact rock sample for determination
principal stresses at which the rock samples were in-situ stress. Proc. 2nd International Workshop
obtained are as follow, σ1 = 41.4 MPa, N on Earth Science and Technology, Kyushu,

385
Japan.
6. Kramadibrata, S. 2007. Development of
geomechanics research in Indonesia with
particular reference to time dependent behavior.
Proc. 3rd Colloquium on Postgraduate Research,
National Post-graduate Colloquium on
Materials, Minerals and Polymers, Penang,
Malaysia.
7. Hoek, E. and E. T. Brown. 1980.
Underground excavations in rock. London: Inst.
Min. Metall.
8. Kurita, K. and N. Fujii. 1979. Stress memory
of crystalline rocks in acoustic emission.
American Geophysical Union, Paper no.
8L1275, 6(1), 9-12.
9. Bieniawaski, Z.T. 1974. Estimating the
strength of rock materials. J. South African Inst.
Min. Met., March, 312 - 320.
10. Herget, G. 1988. Stresses in Rock.
Rotterdam: Balkema.
11. Seto, M., D.K. Nag, and V.S. Vutukuri.
1999. In-situ rock stress measurement from rock
cores using the acoustic emission method and
deformation rate analysis. Geotechnical and
Geological Engineering, 17.
12. Jingen, D., W. Kangping, and H.R. Zun.
1995. In-situ stress determination at great depth
by using acoustic emission technique. Proc. Int.
Conf. on Rock Mechanics (Ed. Daeman and
Schlutz), 245 – 250.
13. Li, C., and F. Nordlund. 1993. Experimental
verification of the Kaiser Effecting rocks. Rock
Mechanics & Rock Engineering, 26(4), 333-351.

386