Embed Size (px)
Citation preview
POSIVA OY
Working Report 2002-28
Analyses of tunnel stress failures at Pyh8salmi mine
Matti Hakala
Harri Kuula
Marko Matinlassi
Petteri Somervuori
Pauli Syrjanen
Pasi Tolppanen
May 2002
T6616nkatu 4, FIN-001 00 HELSINKI. FINLAND
Tel .. +3-58-9-2280 30
Fax +358-9-2280 3719
Working Report 2002-28
Analyses of tunnel stress failures at Pyh8salmi mine
Matti Hakala
Harri Kuula
Marko Matinlassi
Petteri Somervuori
Pauli Syrjanen
Pasi Tolppanen
May 2002
Working Report 2002-28
Analyses of tunnel stress failures at Pyh8salmi mine
Matti Hakala
Harri Kuula
Petteri Somervuori
Pauli Syrjanen
Pasi Tolppanen
Gridpoint Finland Oy
Marko Matinlassi
Outokumpu Mining Oy Pyhasalmi Mine
May 2002
Working Reports contain information on work in progress
or pending completion.
The conclusions and viewpoints presented in the report
are those of author(s) and do not necessarily
coincide with those of Posiva.
Client:
Order:
Contact persons:
Working Report 2002-28
Author:
Inspected by:
Posiva Oy
Toolonkatu 4
001 00 HELSINKI
9734/99/ AJH
November 30, 2001
Ai:::~t~:C~ " 2-ffJ.~. 2_002_ 7fr'V. Matti Hakala, Gridpoint Finland Oy
Analyses of Tunnel Stress Failures at Pyhasalmi Mine
Matti Hakala
Gridpoint Finland Oy
'/ / // ,--~ Harri Kuula
Gridpoint Finland Oy
l
~---- -~ ~- ~- ~-----~--- ~----
ANALYSES OF TUNNEL STRESS FAILURES AT PYHASALMI MINE
ABSTRACT
The aim of this project was to develop a method to map and analyse stress failure
observations, understand factors affecting stress failures, and study criteria to predict
stress-induced failures around excavations. Mapping took place at Outokumpu Mining
Oy Pyhasalmi mine at a great depth between + 1 000 - + 1400 m. The Pyhasalmi mine
was selected as a test site, because its geology and stress-strength ratio are similar to
Posiva Oy' s main investigation site for an underground nuclear waste repository at
Olkiluoto. Data was collected from newly excavated tunnels, 1700 m in total, by using
the pre-developed mapping system.
The main rock type at the mapped area is very hard volcanite with an average uniaxial
compressive strength of 220 MP a. Actually, that includes both felsic and mafic types,
but because of the very dense layering and lack of detailed rock mechanical properties,
they are handled as one type of rock. The quality of volcanite rock mass is good or very
good, GSI - value being 77. Maximum principal stress is 65 MP a to 7 5 MP a, and the
maximum elastic secondary stress around the tunnels is about 150 MPa and around the
shafts 13 5 MP a.
The majority of stress failure observations were classes of none (JO), 34 %, noise only
(Jl ), 36 %, or, light spalling (J3), 26 %. Based on the analyses, it was found that the
most important parameters inducing stress failures were the state of stress around the
tunnel and the orientation of the rock anisotropy. However, the rock type and its
strength, the shape of the tunnels, time after excavation, and the geological anomalies
also have a correlation with the stress failures.
The observed stress failures were compared with the failure predictions based on the elastic 3D secondary stresses and three statistical yield criteria. The best estimations
were found when using Read and Hoek-Brown yielding criteria. Also, the criteria of the
peak strength were used, but were found to overestimate the in situ strength of the rock.
However, the effect of stress anisotropy was found to be significant.
Keywords: stress failure, mapping, tunnel, shaft, in situ stress, rock strength, failure
analyses
PYHASALM EN KAIVOKSEN JANNITYSTILAVAURIOTIETOTULOSTEN TULKINTA
TIIVISTELMA
Tutkimustyon tavoitteena on ollut kehittaa jannitystilan aiheuttamien murtumien
kartoitus- ja analysointimenetelma, selvittaa jannitystilavaurioihin vaikuttavat tekijat
seka tutkia eri myotokriteerien sopivuutta vaurioiden ennustamiseen. Tyota varten
kartoitusta tehtiin Outokumpu Oy:n Pyhasalmen kaivoksessa syvyydella + 1000 - + 1400
m. Pyhasalmen kaivos valittiin koealueeksi, silla siella vallitsevat kiven lujuus -
jannitystila -olosuhteet ovat lahella Olkiluodossa arvioitua tilannetta. Olkiluoto on
valittu Posiva Oy:n ensisijaiseksi tutkimusalueeksi kaytetyn ydinpolttoaineen loppusij oitustilan rakentamiseksi.
Paakivilaji kartoitusalueella on erittain luja vulkaniitti, jonka yksiaksiaalinen
puristusmurtolujuus on 220 MP a. Vulkaniitti esiintyy seka happamana etta emaksisina osioina. Tassa tyossa vulkaniitit on kasitelty yhtena kivilajina, koska ne esiintyvat usein toistuvina ohuina kerroksina eika niiden mekaanisten ominaisuuksien eroja tiedeta
riittavan tarkasti. Vulkaniittikallion laatu on paasaantoisesti hyva tai erittain hyva, GSI = 77. Suurin paajannitys vaihtelee valilla 65 MPa -75 MPa. Suurin kimmoinen sekundaarijannitys louhittujen tunneleiden ymparilla on noin 150 MP a ja kuilujen
ymparilla noin 135 MPa.
Kartoitusalueen jannitystilavauriot jakautuivat paaasiassa kolmeen luokkaan; ei vauriota (JO), 34%, ei murtumia, mutta kallioaania (J1 ), 36 %, tai hilseilya (J3), 26%. Keskeiset
jannitystilavauriotekijat olivat kalliossa vallitseva in-situ jannitystila ja kiven
suuntautuneisuus, mutta myos kivilaji ja sen lujuus, louhitun tilan muoto, louhinnan ja kartoituksen valinen aika ja geologiset anomaliat korreloivat jannitystilan aiheuttamien
murtumien kanssa.
Kartoitettuja jannitystilavaurioita verrattiin kimmoisten 3D-simulointien elastisten sekundaarijannitysten ja kolmen tilastollisen murtokriteerin antamiin vauriotodennakoisyyksiin. Tunneleidensimulointimenetelmaa Paras vastaavuus saavutettiin kayttaen Readin tai Hoek-Brownin myotokriteeria. Myos huippulujuuteen perustuvaa kriteeria sovellettiin, mutta sen havaittiin yliarvioivan kallion in-situ lujuutta.
Lujuusanisotropialla havaittiin olevan merkittava vaikutus analyysin lopputulokseen.
Avainsanat: jannitysmurtuma, kartoitus, tunneli, kuilu, jannitystila, kiven lujuus,
murtoanalyysi, 3D simulointi.
1
ANALYSES OF TUNNEL STRESS FAILURES AT PYHASALMI MINE
ABSTRACT
TIIVISTELMA
TABLE OF CONTENTS ........................................................... .... ......... ....... ............... 1
LIST OF SYMBOLS ..................................................................................................... 3
NOTATIONS .............. ................................................................................................. 5
PREFACE ......................................................................... ................... ....................... 7
1 INTRODUCTION .......................... ........................ ......... ..................................... 9
2 OBJECTIVES............................................................... .................................... 11
3 SITE CHARACTERISTICS ........................ ......................... ... ........................... 13
3.1 Pyhasalmi location and mining history.......................... ........................... 13
3.2 Ore formation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
3.3 New mine................................. .. .......... ................... ........... ..................... 16
3.4 Rock types .............................................................................................. 17
4 MAPPING ................................................................................................... ...... 23
5 VISUALISATION OF MAPPING DATA .......... ................................................... 27
6 ANALYSES OF MAPPING DATA ..................................................................... 41
6.1 General .................................................................................................... 41
6.2 Stress failures in tunnels ......................................................................... 42 6.3 Stress failures in shafts ........................................................................... 51
7 ROCK MECHANICAL DATA ............................................................................ 53
7.1 Intact rock ............ ....................... .. .............. .......................... .......... ........ 53
7.2 Rock Mass ..................................................... ... ... ................................... 61
7.3 In situ state of stress ................................................... ................. ........... 63
8 STRESS STRENGTH ANALYSES....................... ............................................ 65
8.1 Cases and failure criteria ......................................................................... 65
8.2 Tunnel simulation ...... .... .......................................................................... 67
8.3 Shafts ...................... .... .............................. .............. ... .............................. 82
2
9 DISCUSSION AND CONCLUSIONS ................................................................. 95
10 RECOMMENDATIONS .................. ................................................................. 101
REFERENCES ............... .. ........................................................................................ 103
3
LIST OF SYMBOLS
Roman Letters
a constant in Hoek-Brown failure criterion ()
GSI Geological Strength Index
dd dip direction angle from north ( degrees )
dip dip angle from horizontal (degrees)
E Young's modulus
mb rock mass parameter in Hoek-Brown failure criterion ()
mi intact rock parameter in Hoek-Brown failure criterion ()
n number of values
s. d. standard deviation
si intact rock parameter in Hoek-Brown failure criterion ()
S1,2,3 major, intermediate, and minor principal stress ( Pa)
SH major horizontal in situ stress ( Pa)
Sh minor horizontal in situ stress ( Pa)
Sv vertical in situ stress ( Pa )
Q Rock mass quality classification value in Q-system
Q' Rock mass quality classification value in Q-system without the effect of
stress and ground water
Greek Letters
cr1,2,3 major, intermediate, and minor principal stress ( Pa)
crcd crack initiation stress ( Pa )
crci crack damage stress ( Pa )
crH major horizontal in situ stress ( Pa)
crh minor horizontal in situ stress ( Pa)
crv vertical in situ stress ( Pa )
at tensile strength ( Pa )
Cf'ucs uniaxial compressive strength ( Pa)
v Poisson's ratio ( )
5
NOTATIONS
CR Country rock
HB Hoek-Brown
MC Mohr Coulomb
OKP Outokumpu Mining Oy Pyhasalmi Mine
7
PREFACE
This study is part ofPosiva's "Parvi" project where one objective is to collect
information on cavern excavation and response in deep rock conditions. One special
interest is the stress strength conditions in which the rock will fail around the
excavation. The work has been commissioned and supervised by Posiva Oy and the
following organisations and people have participated in the project:
Posiva Oy
F ortum Engineering Oy
Outokumpu Mining Oy
Pyhasalmi Mine
Outokumpu Mining Oy Mining Technology Group
Gridpoint Finland Oy
Saanio & Riekkola Oy
Aimo Hautoj arvi
Pekka Anttila
Pekka Pera Marko Matinlassi Seppo Tuovinen
Pekka Lappalainen
Matti Hakala
Harri Kuula
Petteri Somervuori
Pauli Syrjanen
Pasi Tolppanen
Reijo Riekkola
Erik J ohansson Nina Sacklen
9
1 INTRODUCTION
Posiva Oy, responsible for the final disposal of spent nuclear fuel in Finland, submitted
an application in May 1999 to the Government for a policy decision in principle
concerning the final disposal at the Olkiluoto site. In December 2000, the Government
made, on the basis of this application, a favourable policy decision, which was also
ratified by Parliament on May 181h, 2001. The decision in principle does not grant a
license for the construction or operation of the facility in question, but it is needed to
judge whether the facility is in line with the overall good of society. Currently, Posiva
Oy is designing an underground investigation facility at the Olkiluoto site to
complement the site investigations. Construction will be completed in 2010. While
planning this next investigation phase, Posiva Oy is collecting information on cavern
excavation and rock mass response in deep rock conditions.
It is planned to excavate the repository for spent nuclear fuel in crystalline bedrock at
depths between 400 m and 600 m. Under these rock conditions, the in situ stress
concentrations on the excavation boundary can exceed the critical strength of the rock
and, thereby, initiate rock failure. Depending on the stress-strength ratio and rock type,
failure can vary from minor local yielding to rock burst.
The best ways to evaluate stress failures, at present, are the rule of thumb, or the
commonly-used engineering and scientific criteria. The rock strength, the in situ state of
stress, and the orientation and shape of the excavation must be known for all the criteria.
The orientation and shape of the excavation is normally well known, the in situ state of
stress can usually be measured, and the short-term strength of intact rock can also be
determined. However, the effect of size and time on rock mass strength is more
complicated to define before the excavation and monitoring.
Probably the best-known site where stress failure criteria have been developed and
tested is the URL (Underground Research Laboratory of Atomic Energy of Canada Ltd
in Manitoba, Canada). The rock in URL is massive, unfractured homogenous granite,
thus differing from the Olkiluoto site where the rock is foliated and jointed. Therefore,
additional information on valid failure criteria in the Olkiluoto type of rock mechanical
conditions is needed.
In Finland, the Outokumpu Mining Oy Pyhasalmi Mine (OKP) is the only place where
similar rock conditions to Olkiluoto are easily accessible for the study. The rock type in
Pyhasalmi is schistose, metamorphic, sparsely jointed, and heterogeneous as in
Olkiluoto, and the stress-strength ratio is equal (Figure 1-1, Hakala et al., 1998; Hakala
et al., 1999; Aikas et al., 2000). The most suitable depth for the research is the level
between 1000 - 1400 m inside the volcanite country rock that hosts the massive sulfide
ore deposit. OKP volcanite consists of felsic and mafic volcanites including pegmatite
10
ore deposit. OKP volcanite consists of felsic and mafic volcanites including pegmatite
veins and lenses. In this region, there is no large-scale mining and the in situ state of
stress and the strength of the intact rock is quite well known. However, more
information on the rock mass quality and stress failures is needed to understand the
factors affecting stress failures and select the most suitable strength criteria.
Uniaxial compressive strength ( MPa )
400 ~~~~~-------------------------------------,
300
200
100
0
--e- runnel
--e- shaft
Stableunsupported
Olkiluoto 500 m ( 400- 700) Aikas et al. , 2000
50 100
OKP 1300m ( 1200- 1400)
Hakala et al., 1998 & 1999 Minor spa/ling -light support
150
Severe spa/lingmoderate support
Heavy support required
Vety difficult to support
Based on: Hoek and Brown ( 1980 ), modified by Martin ( 1998)
200 250
Maximum secodary stress ( MPa )
Figure 1-1. Stress-strength ratio at Olkiluoto mica gneiss and Pyhiisalmi volcanite
(Hakala et al. , 1998; Hakala et al. , 1999; Aikiis et al. , 2000).
In this report, the rock mechanical mapping system, that was obtained, is introduced and the collected data summarised and analysed. Furthermore, the secondary stress state around the typical excavations is calculated, and by using different failure criteria, is
compared to observations. All the presented depth levels follow the existing system of
the Pyhasalmi mine. The effect of excavation work on the failures is excluded from the analyses.
11
2 OBJECTIVES
The main aim of this Posiva Oy- OKP co-operation research project is to collect and
analyse information on stress failures from known geological and rock mechanical
conditions, in order to develop a usable mapping and stress failure prediction method. Geological, rock mechanical, and stress failure mapping are obtained in the decline, level access, and access drifts situated in volcanites in Pyhasalmi mine (Matinlassi,
2001 ). The ends of the ventilation shafts were also mapped. All the characterised drifts were excavated by the drill and blast method while the shafts were raise-bored.
The following stages have been planned to achieve the objectives: 1) development of the rock mechanical mapping system, 2) visualise and analyse OKP stress failures and data of the rock mechanical
classification from the first mapping period, June 1999 to January 2000, 3) study factors that correlate between the failures and classified data, 4) summarise the rock mechanical testing data, 5) perform 3D-stress analyses for selected locations 6) compare the stress analyses results, using different failure criteria, to the
observed behaviour.
13
3 SITE CHARACTERISTICS
3.1 Pyhasalmi location and mining history
The Outokumpu Mining Oy Pyhasalmi (OKP) mine is located in Central Finland about
450 km north of Helsinki (Figure 3-1). The Pyhasalmi ore deposit was found in 1958.
The first estimation of reserves indicating 12.2 Mt of ore with 0.81% Cu, 2.93% Zn, and
36.8% S, was made in February 1959. Construction work started in August 1959 and
production commenced in March 1962, three and a half years after the discovery. The
mine started as an open pit and the final open pit depth of 125 m was reached in 1967.
Underground mining started the same year. The total production from the open pit was
6.8 Mt of ore and 5.6 Mt of waste rock. In 1967, after extension and automatisation of
the processing plant, the annual production was increased from 600 000 to 800 000 tons.
The current annual production from the underground mine is about 1.08 Mt. (Weihed &
Maki, 1997).
* Pyhasalmi
Figure 3-1. Location of Pyhiisalmi mine and 0/kiluoto investigation site.
3.2 Ore formation
The supracrustal rocks in Pyhajarvi are part of a Palaeoproterozoic, Svecofennian schist
belt. The NW-trending schist belt is situated between the Archaean craton in the east
and the Central Finland Granitoid Complex in the west.
14
The metavolcanic rocks in Pyhajarvi can be divided into two groups on the basis of field
observations and chemical composition (Kousa et al., 1994). The western part of the
Pyhajarvi area belongs to the Nivala gneiss complex. The Pyhasalmi and Mullikkorame
ores are situated in the eastern part of the volcanic belt (the Pyhasalmi volcanic
complex). The Pyhasalmi volcanic complex can be divided into the Ruotanen Formation
and the Mullikkorame Formation. A large gneiss area (the Kettupera gneiss), which is
intruded by syntectonic plutons is situated between the Ruotanen and Mullikkorame
formations. Both formations contain large areas of mafic volcanic rocks with pillow
lavas, sodium-rich rhyolitic volcanic rocks, and volcaniclastic. The stratigraphy of the
area is still unsolved. According to Lahtinen (1994), the Kettupera gneiss and the rhyolites in the Riitavuori and Lippikyla Mb:s are chemically similar and may be
genetically related. The age of the Kettupera gneiss and the rhyolites, 1.93 and 1.92 Ga
respectively, also suggests a common history (Kousa, 1990). Age determinations for the plagioclase porphyrites, which have been found as inclusions in the massive ore, indicate distinctly younger ages at 187 5 Ma (Helovuori, 1979). These plagioclase
porphyritic inclusions, as well as the quartz-porphyry and amphibolite inclusions, are
unaltered and represent younger intrusive dykes. Similar dykes have been found in
outcrops and drill cores outside the massive ore.
Field observations indicate that most of the mafic volcanic rocks seem to be younger
than the main part of the felsic volcanic rocks. The volcanism started with felsic
volcanism in an extensional continental margin, followed by mafic volcanism in a rifled
marine environment. Large-scale hydrothermal alteration is associated with this stage
causing strong sodium enrichment in the rhyolites. Ore formation near the centres of
mafic volcanism seems to be related to this period. Without a longer hiatus, the volcanism continued with more calc-alkaline volcanism south and west of the
Pyhasalmi volcanic complex.
The Pyhasalmi ore deposit is a typical volcanic hosted massive sulphide deposit. The volcanic rocks are mainly felsic pyroclastic rocks and coherent quartz-porphyries. Mafic volcanic rocks are coarse-grained tuffbreccias and lavas, including pillow lavas. Mafic and felsic dykes are common. A litho logical map of the Pyhasalmi volcanic complex is
shown in Figure 3-2.
15
Lake Pyhajarvi
Lithologies Mafic volcanics D Altered felsic volcanics
Pegmatite granite D Intermediate volcanics - Ore - Diabase D Intermediate tuffite D Mica gneiss
D Feldspar porphyry
D Felsic volcanics
Other symbols
~ Porphyritic granite Felsic tuffite
Granite D Skam - Gabbro/diorite D Volcanic conglomerate - Quartz diorite - Amphibolite
D Granodiorite D Altered intermediate volcanics - 0.5 0 Altered mafic volcanics
Figure 3-2. Litho logical map of the Pyhiisalmi volcanic complex
(Weihed & Maki, 1997).
Fault
Graphite schist
Seam intercalations
Mafic dyke
0.5 Kilometers
l
The stratigraphy is unclear. Polyphase deformation together with amphibolite facies
metamorphism makes the interpretation difficult. Both felsic and mafic volcanic rocks
are strongly altered near the ore. The ore body has a complex shape due to the polyphase
deformation and is surrounded by a large alteration zone. The contact between unaltered
and altered volcanic rocks can be sharp or gradational. The thickness of the altered rock
sequence at the surface is about 100 m in the west and 300 m in the east. At the deeper
16
levels, the alteration zone is only a few meters thick. Sericite-quartzites with a high
pyrite content are presented adjacent to the ore. Pennitised cordierite porphyroblasts are
common. Two zones of cordierite-anthophyllite rocks occur in the footwall , partly
within the sericite quartzites. A zone of cordierite mica gneisses, that contain portions of
a cordierite-anthophyllite rock, occurs outside the hanging wall sericite quartzites. The
folded alteration zone is over 5 km long at the surface.
3.3 New mine
In spring 1996, an ore prospecting project was started in order to find mineral reserves
below the + 1050 level. Core drillings began in June 1996. As a result of the core
drillings, the size and grade of mineralisation appeared to be auspicious and it was
decided to excavate a research tunnel. This 500 m long tunnel was completed in October
1997. An intensive core-drilling period from the research tunnel and decline then began.
Based on exploration results (22 000 m of drill core) the ore evaluation was completed
in May 1998. The probable mineral reserves between level+ 1050 and+ 1350 were
calculated to be:
Zn-mineralisation
Cu-mineralisation
3 155 000 t
11 400 000 t
Total 14 555 000 t
Cu0.9% Zn 5.1%
Cu 1.4 % Zn 1.1 % ...............................................................................................................................................................
Cu 1.3 % Zn 2.0 %
The mineralisation also contains silver 14 g/t and gold 0.4 g/t.
s 39.9%
s 45.0%
s 43.9%
According to core drillings, the ore extends to level +1410 (Figures 3-3 and 3-4). The
ore is formed as a top of fold with a massive pyrite ore core surrounded by copper ore
and zinc ore. Contacts between the ore and surrounding rocks (volcanite) are sharp.
Altered rock types in the contacts do not exist thus differing from the formation of
"old ore".
The excavation of the decline from level + 11 00 started in April 1998. The end level, + 1445, was reached in May 2000. A new hoisting shaft between the surface and level
+ 1441 has been constructed and production began on July 1st, 2001.
17
NEW MINE
Figure 3-3. Pyhiisalmi mine, the new mine area is presented in a darker colour.
3.4 Rock types
Geology
The massive Zn-Cu ore deposit at Pyhasalmi is associated with Proterorozoic volcanites.
Most of the volcanites are acid pyroclasts and porphyric rocks. The basic rocks in the
mine area are mainly coarse tuff breccias and veins. The volcanism was bimodal in
character. Both felsic and mafic volcanic rocks are strongly altered near the ore. The
thickness of the altered rock sequence at the surface is about 100 m in the west and
18
300 m in the east. At the new mine area, the alteration zone around the massive sulphide
ore is lacking. Sporadic lenses of week altered rocks can be fmed as inclusions in
massive sulphides.
z = -1100
z = -1300
z = -1400
Figure 3-4. West view of the new mine area. Copper and Zinc ores are presented in
different colors.
Acid volcanite
The felsic volcanic rocks are sodium-rich rhyolites with high Si02 content. The essential
minerals are quartz and alkaline feldspar. Schistosity is quite clear (Figure 3-5). Felsic rocks are altered near the ore at the upper part. F elsic rocks are sheared near the ore at
the deeper parts. Grain size is under 1 mm. About 60% of the rocks are felsic volcanites
in the mine scale.
19
Figure 3-5. Typical felsic volcanite with clear schistosity. The two dark layers are
mafic veins. The picture is from the + 1200 level.
Mafic volcanite
The mafic volcanic rocks are basaltic low K-tholeiites of primitive island arc type. The essential minerals are plagioclase and amphibole. The colour is very dark to black (Figure
3-6). Schictosity is quite clear near the ore but quite massive far away from the ore. The
grain size is less than 1 mm. About 40% of the rocks are mafic volcanites in the mine
scale.
Figure 3-6. Massive mafic volcanite (black) at the + 1350 level.
20
Pegmatites
The pegmatites are a minor rock type in the Pyhasalmi Mine (Figure 3-7). The
pegmatites usually occur as a vein with varying thickness. Usually the veins are 0.1-2.0
metre wide and might be very long. The pegmatites intruded in many stages, so their
composition varies slightly. The essential minerals are usually quartz and alkaline
feldspar. The grain size ranges quite considerably from 5 mm to 20 mm.
Figure 3-7. Pegmatite at the+ 1420 level.
Talc Schist
The talc schists are also minor rock types in the mine. The talc schists are very soft with perfect cleavage into laminae (Figure 3-8). The talc schists are strongly deformed and
altered and are originally were probably dolomitic limestones. In the upper parts, talc schists occur near the footwall contact, partly inside the ore. In the deeper parts, talc schist only occurs inside the ore.
21
Figure 3-8. Talc schists on the wall at the + 1125 level.
Zn-Cu-S Ore
The Pyhasalmi ore deposit is a typical volcanic hosted massive sulphide deposit. The average grade of the ore in the New Mine is 1.1% Cu, 2.1% Zn, 38% S. The composition of the Pyhasalmi ore varies both horizontally and vertically. As a rule, chalcopyrite is concentrated in the central part of the ore and sphalerite near the contact in the New Mine. The highest barite contents are generally encountered in the sphalerite-rich areas. The Pyhasalmi ore is a massive pyrite ore with 70% sulphides (Figures 3-9 and 3-1 0). The sphalerite-rich ore is in some places fmely banded and thin mylonitic bands are common. Round pyrite phenocrysts occur in the fme-grained sphalerite matrix of the mylonitic ore. A pyrite dissemination, which in some places has a breccia structure, exists around the massive ore.
22
Figure 3-9. Pyrite-copper ore in drill core 41 mm in diameter.
Figure 3-10. Drill core of Zinc-pyrite ore 41 mm in diameter.
Sericite-Quartzite
Near the ore the rhyolites are altered to sericite-rich schists with disseminated pyrite in
the upper part of the ore. In the lower part near the ore, sericite-quartzite no longer
exists. Sericite-quartzites were originally felsic volcanic rocks which can be deduced
from the Ti02/Zr ratio.
23
4 MAPPING
Data for the analysis is collected from decline tunnels, level accesses, and access drifts.
The mapping system used in this project is mainly based on the Q-classification system
(Grimstad & Barton, 1993) and the Finnish RG-classification systems (Korhonen et al.,
1974) and briefly presented here. A more detailed description of the mapping system is
presented in Matinlassi (200 1 ). In the following paragraphs, the main categories of
collected data such as rock type, rock tunneling quality index Q (visual estimate),
fracture zones and major joints, stress failures, and scanline mapping, as well as the contents of the individual databases are presented. Down the tunnels, all the data is
noted and stored as data sheets in the ExcelTM program. The starting and finishing points
are based on the metric point system in the mine.
Rock type database
Field description value • Level • Metric point
.. ~ ........... ~.!.~~~P.:.S. .. P.2~.!!! ... .2K~.9..~~---!.Y.P.~.................................................... . .......... .......... -·····--····-···-·--·····--·-·-·····-····-·····-····-·······-·--··-··-·-················-·········-·-···-··············· ... ·······························································
--~·-····· ··-~-~~~-~~ .. J?..?..~~-!~ .... ?..fE?..~~--!¥.P.~ .. -............................................................................... ·····-··········--·····-·-·-··························-·····-··· ········-·······-····-·-···-··-······-·····-··-····-····-················-··· .. ··············-·····-·-············· • . .... ~.9..~~ .... !.Y.P..~ ... t~'.~.~-~~.! .... ~.~!.~!P.~!~.2.!!2......... . ................... . .. . .... ... .. . .............................................................. .................................. . .................................................... . • Rock type fabric masstve
foliated
composite
.~ ....... ...!?..~.P .... ~!?:g!.~---~-~---~-~-~~~~-----
MO= non foliated
Ml =slightly foliated L2 =moderately foliated L3 =strongly foliated
SO= non foliated S 1 = slightly foliated S2 = moderately foliated
···············-~-~ .... =.==.=. .... ~!E9..P.:.S.!.Y ... f.9.!.~.~!.~.4.
-~------~~.P. ... ~~-~~-~!i.~~-~~-~-~~-~~-~----····-·-·-·······--·--· ·--···---·········-··························-······ ....................................... ·····--···-·······-·-·········-··--···- ···-······-·-··-··--·-·-················-·······-········-·······-···-·-··--·-·····-·
• Grain size fine-grained medium-grained
Vr (grain size) < 1 mm Vr 1-5 mm
coarse-grained Vr 5-10 mm
--··-········-··············-·--·········-····-····--·····--·-···-·-··-············-···-········--··-····· ·····---·---·-·············Y~!Y. .... ~-~~E~~=~~}E..~-~--Y..!.?:.._~_Q--~-------··-··················-·-····-·-· ·······-· ······-···-···· -·-··--• Degree of weathering unweathered RpO
slightly weathered Rp 1
highly weathered Rp2
-···············-····-·············-····-··············-·-··-···-·················-····-·····-·--·-··-·· .. -·---···--·····-···-··········-·-· · ····-···--···-~.!!.Y. .. ~-~-~!!?:.~~~.4. ......................... ~P.~---··-··· ················-·········--····--·-············································--······-············· • Remarks
24
Visual estimation o(the 0-index
Field value • Level • Metric point
--~·-·········§.!.~!.:!.~P.:.S .. P..9.i..~! ... .9..f..E.9..~~---!.Y.P~.. . ...................................................................................................................................................... . --~--- ······ -~-~~~~~---~~-~~~---?..r.~~-~-~---~!.1.'..~ ........................................................ . ......................................................................................................................................................................................... . .. ~ ........ ~.Q.P. .. {~2. ... f.!9..~ .... !h.~ ... ~.~!.!.~....... . .................................. .<.~.~~.9..~~!.~g ___ !.9. ... !!J!:!q_~/'!:§9....'.! .... f!:.'!:4. .. !2.t.~.c.!~':..!..q_~~---.!...22..fil ..... . .. ~ .......... ~~-~ ... J?.~.~! .... ~.~! .... ~~~~-~~---··············· ··········-·-··-·······-·· ·· .............................. (~.?.~.~-~~~-~~ ... !.~---·~-~~~?.-~----~-~---~ .. !.: ..... !..?.?i.J.. ............................................................................ . .. ~ .......... !.~~_jg_~P.:!. . .E9.~g~-~~-~----~~P.!.!?.~~- - -· · ······ · ·· ................................. {~-~-~.9.E4.!~g_ _!.9. ... !:?..c.!.Tt.9..'.! .... ~.t. ... q.!.:_ ..... !..2.Z1..)
(according to Barton et al. 1974)
_! .......... !.~'-j_9..~.P.:! .... :~~Y~!.~~--E.~4.~.~!.~9.P.: .... ~~P.!.!?.~!. ................. (~-~~9.E4.!.~g ___ !.9. ... !.:?..c.!..r.t.9..'.! .... ~t...qf: ...... !..2..?.1..) ...................................................... . --~···········~-~~ ---~!.~~-~-~---·~~~~.?.!.~?~.- -~~.?..!.~.~---···· ····························· · ·· · ·· (~_?.?..?..~~~.1?:~ ... !.?. .... ~.~-~~~-~-- - ·~-~- - -~!.: ..... !..?. .. ~i.J...... . .. . ................................... . .. ~ .......... ~.9..~~----~-~~-~ .... 4.~.~-~~1?..!!.9..!?.: ....... ··· ······ ····· · ······ ··· · ············ ······················· · -~-~!.~.~!.~!.~~---~!.! .... ~P.~.~~4.~.~~-~!-.~-~-~-~.4. .. .9.P.: .... ~.P.:P~! .... Y.~!.~~~---·--·· • Remarks
Fracture zones and major joints
All fracture zones and joints over 20 m in length are mapped and photographed. Only
the joints, which continue outside the tunnel, are included. The fracture zones and major
data sheet include 9 records:
Field description value
• Level • Metric point
• Location
.. ~ .......... !?.~.P.. ... ~~~!.~---~-~-- --~-~-~~-~~---··········· · ·········· ·····-······ ·-·-··· ·····-··-·····-····························································-·········-·-············· ········ ··············-·-······· ··· · ····························· · ··-· ··· · ··· ········· .. ································· ............................................ .
• ....... -.!.?..~P. .... ~~E~.~!!.9.P.: .... ~_1!. .... 4..~~-~~-~---······················-··--········ ···-···-···--···-······- ··· ··· ··· · ········ ····· ·····-····················· ····· ··-········ ··· ······ ················· ··-·· ··· ·-··········-···-· ······ -············· · ·-················· ··· · ········ ·-····· · ·········· .. ····································-• Type of failure zone one or few planar joints Ril
high fissure frequency, no fissure filling Rill densely fractured, fractures poorly filled Riiii high or dense fracturing, crushes
filled with clay
................................................................................. ~bundant g~~_ge
• ......... ~~'-...J .. ~-~~.!. .. E?.~.~~-~-~-~--- -~~~~-~-~--- ······· ········· • Ja, joint alter~!~_on number .............................................. ................................................................. . • Thickness of the fracture zone
• Joint P..!.!.i..I.lg ....... ....................................................................................... . • Remarks
RiiV RiV
25
Stress failure mapping
Stress failures are mapped only if they can be clearly noted to be induced by stress
concentrations. In some cases, it is not so straightforward. Failure locations in a tunnel
profile are also recorded. The stress failure mapping data sheet includes the following
10 records:
Field • Level
• Metric point
• ........... §.!.~~!.!?:g ... P.2!.~!. • Ending point
......................................................................
• Type of stress failure
description
none
some noise, not spalling
value
JO Jl
visual deformations or cracks in shotcrete J2
light spalling J3
spalling and rock burst J 4
······-·····---········-·-····-··-·--·-·····················--····-·-·-············-·················-····- ····-···-·· ····· ·-· · ·!?:~~:YY.E2.C?..~ ... ~~!~.! ..................... -................................ . 15
• Width of failure
-~ .......... Q.~£!!?: ... 2.f.f~!~!~--------·---·----··--·-····---··--·---···--·--······-··-····-·--·····----·--·------·-·-··--··--··------····-··--·-·-·--·-···-··-·---···---·-·----------• Location
• Breaking type failure by breakage M
failure along existing joint R
mixture ofbreakage and failure along
·································-·················· ··-·--··--·-·--··-·-······· -·-···· ·-··-······· ·····- ··-···-······-· -········· · ··~~i.~.!!.~gjg.!!?:.! ........................ -········-····-······-·······-·-·····-··-··-····-········-····-·······-.. ···-···-··········· ··-·-···-······-·············-·~!.g .............................. . • Excavation date
• Shotcrete date
• Remarks
26
Scanline mapping
Scanline mapping is performed in locations where more detailed data is required.
Photographs are also taken of the scanline mapped area. The scanline mapping data
sheet includes headline data and 13 records. In the head line, the following subjects are defined:
Field description • Level • Metric point
• Code of the drift
• Code of the scanline ···························· ······························································
• Rock type
• .......... ~.!.~!:!.!.~.K.P2.~.~! .......................................................................... ·-·····--····--··---···-····--···-·······-··-··-······-·--··-- ..................................................... ---·-·-·-·-·--------·--·--·····--······-··-····-····-··-······-····-·······-·············-·-
.. ~ .......... !?..~.P.. ... ~~.~ .... ~.~.P.. ... ~.~E~.~!.~?.~ ... ?..!. ... ~.?..~~.!.~~.~ ... .<.~.~.~.~~.~2................. ····················-···-·······-·····-·-·-······························-····-····-··········-···-········-····-········-------····-·-····-·-·········-······-··--·~----!:.~~g!.h .. .2.f.!b:.~ .. -~.~.~~i.~.~.-~.~ .. E.?:.~!!"~~·····-·-·· ··-····-···--·-·-··--·······-·····-·············--·······················-···-····--·---····-··--··-·-·----··------·-·---····---··--··---·· .. ~ ........ J.~~ ... J..?..~.~! .... ~.~! .... ~~~E-~E ...................................................................................................................................................................................................................................................................................................................................... .. _!_~QQ_:.~~.!.~~···--------·----····-······· ··· ················-·····-···-····· ········--···-·-··-·-····--·-·-·-··············-·······-·········--·--············-··-·-······--·-···-··-·--·--·····-·-·-·-·-···-···---·-····-····-·-··----·······-······-··················-·-·-··· .. -~_!~~...J..~}.~!-~~!-~E..E_~du~!.~-~~~?mber -------·--·--···-··-·-·------··-----
_! ......... §~?. .... ~.!~~~-~.-~~~~~-!.!..9...!!...f.~.~t~---·--------·----·-··---·--··------·-·····-············-·--·-···-·····-············-···-·--·-·--·---····-------·--···- ·-··--··--·-·-·-·-·-·-~!?.~~~ .. -~~~ n~~~.-~~ th~_~eol~......:gt::_·_st ________ . • Rock mass description calculated in spreadsheet based on input values
• Average Jr, joint roughness number calculated in spreadsheet based on input values • Average Ja, joint alteration number calculated in spreadsheet based on input values
13 records in the data sheet:
-·~··· ··· ··J!?. .... ~~.JE.?..~~ .... ?..!..J...?..~.~.!. ............................................................................................................................................................................................................................................................................................................................................ . -·~·········-Q~P. ... ~.~g!.~ ... ~.~ ... ~.~.~~·~·~···················-·······-································-·············-·······-············-······························ ··· ··-············-·····-··· ·······--· ···-·-··· · ·· · -··-·······················-·············-··································-·······················-··························-.. ~ ...... )?.~P. .... ~~E~.~!.~.?..~ .... ~~ .... ~.~~.~-~·~····················-····-································ ································································-···--·--·--····----··-·········-···-·-·········--·-···-···-·-··-·-····-······-···-·· ... ········-····-····-····················-···-··-·--.. ~ .......... I2.~.~! .... ~.P~.~!.~.K... . ..... ......................................................................................... ....................................... ........ ............................................ ............................................. . ............................................. ...
.. ~ .......... !.?..~.~! ... !..~~~~·--····--·-·················- ··························--····-··············-· ...................................................................................................................... . ............................................................. .
• .. I2.~.~! .... ~2~!.~.~~.~.!Y.. ................ .
• J ?.~.~t undulati~~ ............................................................................................................................................................................................................................................................................................................................... . • Jr, joiJ?:!. roughness J!Um~-~!.. ....................... .
-~ .......... !._~'. .. .J. .?..~.~! ... ~.~.!.~~.~.!~.?..n ~~~~.~~ ··········································· -·~··· · ······~2.i..~! .... 2.l?..~~~.~g········-· .......................................................................................................................................................................................................................................... .
.. ~ .......... !.?..~.~! .... ~.~.!~.~~························ ·······················-·················· ................................. ········································································································
.. ~ .......... T.Y.P.~ .... 2.f...f~!.!.~~·~····~·2.~~ ............................................. . • Remarks
27
5 VISUALISATION OF MAPPING DATA
In the following pages, a visualised overview of mapped data is shown while the more
detailed analyses are presented in Chapter 6. Visualisation was carried out using the software, Surpac2000 Version 3 .2, which is a geological modelling, mine design, and
production system. The software has applications at all stages of the mine life-cycle from resource estimation, planning and operation to remediation of the site. Originally
Surpac was developed specifically for mining but it has been found to be useful for other geotechnical design areas, for example, tunnelling. (Somervuori & Middleton,
2001; Surpac Sowtware International, 2001 ).
During the first mapping period, June 1999- January 2000, a total of2.6 km of tunnels were excavated, of which 1.7 km were mapped (Figure 5-1). A typical tunnel crosssection is 5*5 m2 with a slightly arched roof. The majority of the mapped tunnel sections
are between the+ 1200 and+ 1400 levels (Figure 5-2). The tunnels are mainly straight or
curved although there are also crossing areas and wider profiles (Figure 5-3). The
crossing area is assumed to extend 10 m away from the physical crossing.
-1050
-1085
-1120
-1155
-1190
-1225
-1260
-1295
-1330
-1365
-1400
Figure 5-1. Tunnels below the+ 1050 level coloured by the depth.
28
Figure 5-2. Excavated and mapped tunnel sections .
•
Figure 5-3. Mapped tunnel sections, crossing areas and wider profiles highlighted.
29
In the mapped tunnels the rock type is mainly volcanite but some sections of ore,
pegmatite, and sericite-quartzite also exist (Figures 5-4 and 5-5). Clear mafic or felsic
volcanite sections are rare while felsic volcanite with mafic volcanite stripes dominates .
• PEGMATITE
PEGMATITE I ORE
• ORE
• FELSIC VOLCANITE
• MAFIC VOLCANITE
• FELSIC VOLCANITE WITH MAFIC STRIPES
SERIZITE QUARTZITE I VOLCANITE
Figure 5-4. Main rock types in the mapping area. The light grey areas are not
mapped.
The volcanites are slightly or moderately foliated but the ore is massive (Figure 5-6).
The dip direction of the foliation is perpendicular to the volcanite-ore contact and the dip angle is steepest close to the ore contact (Figures 5-7 to 5-9).
..
• ORE
• VOLCANITE
• PEGMATITE
SERIZITE QUARTZITE
30
Figure 5-5. Rock types, all volcanites are included in one category .
• MO, M1
• so
• 51
52
• L2
Figure 5-6. Rock fabric.
31
y '
• 0° ... 30°
• 30° ... 60°
• 60° ... 90°
Figure 5-7. 3D-view of foliation orientation.
• 0° ... 30°
• 30° ... 60°
• 60° ... 90°
ORE BODY f
1
Figure 5-8. Top view of foliation orientation.
SG4
++
SG5
Figure 5-9.
SG10
32
SG9
+ + + + +
+ + SG9.Joints no:21
03 + +
+ + +
+ + + ++ ++ + +
t+ + + + 008
+ + + +
+
SG 7. Joints no 24 + + +*+
+ : ++
+ + + +
~ t + + +
1-+ + + +
+ + + lj.Z!..,j,--.!-!-,.--,+ + + + + 1!- + + +
+ +
+
~8. Jof~: 3~ + +
+
+
+
Stereo net projections of foliation orientation.
+
+ +
+
+
+
+
+
The volcanites has fme grain size but the ore is medium or coarse grained (Figure 5-10).
All rock types are unweathered. The mapped rock mass is sparsely jointed; the average RQD-value being 87 (Figure 5-11 ). Volcanites normally have one joint set along the
local foliation. Random jointing is typical in ore (Figure 5-12).
-~---~-----------------------
33
0
•
• <1 mm
• 1 ... 5 mm
5 ... 50 mm
• >50 mm
Figure 5-10. Grain size of mapped rocks .
• 75 ... 100
• 50 ... 75
25 ... 50
• 0 ... 25
Figure 5-11. RQD-value of mapped tunnel sections.
34
• ,
• <1 , RANDOM
• 2-3, 1 SET+ RANDOM
• 4 - 6, 2 SETS + RANDOM
9- 12, 3 SETS+ RANDOM
• 15-20, > 4 SETS, CRUSHED
Figure 5-12. Number ofjoint sets; Jn number and number ofjoint set.
The majority of mapped fractures belong to class Ril, being a planar fracture or set of
planar fractures (see table on page 25, Figure 5-13). Some Rill and Riiii class fractures
were also encountered but none with crushed filling. The fracture orientation mainly
corresponds with the foliation (Figure 5-7 - 5-9). In addition, some sub-horizontal
fractures were found (Figures 5-14 to 5-16).
. Ril
• Rill
• Ri Ill
RiiV
. RiV
35
\
Figure 5-13. Fracture types based on the Ri fracture classification system
(see Table on page 25).
\
r ;
~
* ...._..., » ~ ' ~ • 0° ... 30°
• 30° ... 60°
• 60° ... 90°
Figure 5-14. 3D-view of fracture orientations.
(
( ORE BODY
36
• 30° ... 60°
• 60° ... 90°
Figure 5-15. Top view of fracture orientations. Exceptional subhorizontal orientations
near the ore body (blue arrow) .
SG9
SG1
SG ,.--t---i..i.~ + +
+ + + +
'""$"r:-JOO~fTOl___.!+~-----i~-___,.,.,.,.-,;;;;dno : 14
SG5.Join1s no:2
SG5 (~~ CJ V
++ +
SG3
+ +
SG 7. Join1s no:
SG6
Figure 5-16. Stereonet projections of fracture orientation.
SG6. Joinls no: 6
0
37
Based on the rock mechanical mapping, the rock quality is good or very good, that is, Q'-value varies from 10 to 100 (Figure 5-17). GSI values (Eq. 5-1, Hoek et al. 1995)
vary between 70 and 90, respectively (Figure 5-18). Only a few sections have a GSI
value in the range of 60 to 70
GSI = 44 + 9ln(Q ')
• 40-100, VERY GOOD
• 10 - 40, GOOD
• 4 - 10, FAIR
1 - 4, POOR
• 0.1 - 1, VERY POOR
Figure 5-17. Q' rock mass quality value.
,
D c
( 5-l)
38
,
• 90 ... 100
• 80 ... 90
• 70 ... 80
60 ... 70
• <60
Figure 5-18. GSI rock mass quality value.
The majority of observed stress failures belong to classes 11 =noise or 13 =light
spalling (Figure 5-19). In a few locations, moderate spalling or rock burst, class 13, were
also found. The observed stress failures at tunnel cross sections or at wider profiles do
not differ considerably from the overall behaviour (Figure 5-20).
39
• ,
D JO, NONE
• J1 , NOISE ONLY
• J2, VISUAL DEFORMATION • • J3, MINOR SPALLING
J4, SPALLING OR ROCK BURST
• JS, HEAVY ROCK BURST
Figure 5-19. Stress failure observations at mapped tunnel sections.
0
JO, NONE
• J1, NOISE ONLY
• J2, VISUAL DEFORMATION
• J3, MINOR SPALLING
J4, SPALLING OR ROCK BURST
• JS, HEAVY ROCK BURST
Figure 5-20. Stress failure observations at tunnel cross sections and wider profiles.
41
6 ANALYSES OF MAPPING DATA
6.1 General
Collected data from mapping is analysed to fmd the factors, which are mainly
correlating with stress failure observations. The study is focused on the in situ state of
stress, direction of excavation, size and shape of excavation, geological structures, rock
type, rock fabric, orientation of rock foliation, and rock mass quality. Some of these
factors are connected to others and can not be studied independently.
Mapping data
A total of 1. 7 km out of the approximately 2.6 km of excavated tunnels was mapped for
this study (Figure 6-1 ). The corresponding values in the volcanic country rock are
1.3 km I 2.2 km and in the ore 0.25 km I 0.4 km. The majority of data is collected below
the + 1200 level. Mapping always follows the access tunnel directions, so, in the directions 30°, 60°, 90° and 150°, the length of mapped country rock tunnels is less than
50 m (Figure 6-2).
Sum of drift length ( m )
1400
•Excavated
1200
• Mapped
1000
800
600
400
200
0
0 0 0 CR CR CR z <=1200 200 < z <= 1300 z > 1300 z <=1200 1200 < z <= 1300 z > 1300
Ore I Country Rock, Depth range z
Figure 6-1. Total length (m) of tunnel excavation and mapping.
42
Total drift length (m)
600r-------------------------------------------------------------~
•eR_ Excavated
• cR_Mapped 500
• Ore_Excavated
0 Ore_Mapped
400 ~------------------~----------~----------------------------~
300 ~------------------~----------~----------------------------~
200 1-------------
0
0 30 60 90 120 150 180 210 240 270 300
Drift excavation direction ( degrees )
Figure 6-2. Length of tunnel excavation and mapping sections in different drift
directions; CR is country rock.
6.2 Stress failures in tunnels
Rock type versus stress failure observations
330
The majority (72%) of the tunnels were excavated in felsic and/or mafic volcanites and
50% of which is striping felsic and mafic volcanites (Volc_F _Ms in Figure 6-3).
Three stress failure classes dominate; 10 (no failure, 34%), 11(noise only, 36%) or 13 (light spalling, 26% ). Only 4% of all the observations belonged to classes 12 (visual deformations or cracks in shotcrete) and 14 (spalling and rock burst). This result might be possible, but the defmition between classes 12 to 14 is not clear enough to draw further conclusions. Therefore, classes 12 to 14 will be hereafter treated as one stress
failure class 12 .. . 14.
43
Total drift length I stress induced failure (m)
300~----------------------------------------------------------------
250
0 Volc_F (6%)
Volc_M (11%)
Volc_FM Volc_F_Ms Volc_SerQua Volc_Ore (6%) (47%) (2%) (0%)
Rock type I Contact type
Volc_Pg (7%)
Pg (2%)
Pg_Ore (0%)
Ore (14%)
Figure 6-3. Stress failure classification within mapped rock types. The amount of
each rock type is presented in parentheses. Failure classes JO ... J4 are explained in
Chapter 4.
When comparing the rock types and occurrence of all types of stress failures, the
pegmatite and pegmatite-volcanite contacts are the worst (Figure 6-4). Occurrences in
one main class are calculated by using
Isub- ctass,i x 100% , where I = length, n = number of classes. lsub- class,i
i=l,n
If class Jl (noise) is omitted then the mafic volcanite and serizitequartizite-volcanite
contacts are the worst, and the situation is best in the ore. Generally, in pegmatite 20% to 25% of the excavated tunnels do not show signs of stress failure, while in volcanites
this portion is 30% to 35%, and in the ore it is over 40%.
Location, type and extent of stress failure
The location at the tunnel surface and breaking type could be defmed for only 17% of all
J2- J4 type failure observations. Furthermore, in only a few cases was it possible to
defme the volume of a failure. Almost all of these were found at the tunnel roof. (In half
44
of these cases, the excavation direction was either 120° or 270° to the north, and the third major direction was 330°. Based on these facts, it was decided to ignore the
location and breaking type for the following analyses and use the length of mapping section instead to represent the extent of the stress failure.
Occurrence of stress induced failure ( % )
100% r;::=:=::-----------------------
. J2 .. . J4
80% ~====~-----------------------
70% 1-----------------------------
60% 1------------------
50% t--------------
40% J------
30 o/o
20 o/o
10 o/o
0% Volc_F (109m)
Volc_M (180m)
Volc_FM Volc_F _Ms Volc_SerQua Volc_Ore Volc_Pg (97m) (747m) (41m) (14m) (117m)
Rock type I Contact type
Pg (36m)
Pg_Ore (7m)
Ore (226m)
Figure 6-4. Occurrence of stress failure within the mapped rock types. The failure
classes JO ... J4 are explained in Chapter 4. The amount of each rock type is presented in
parentheses.
NOTE! All the following analyses concentrate on the volcanites, which are the dominating rock types. Although, there are some differences in stress failure observations between the different types of volcanites (Figure 6-4) they are hereafter
treated as one rock type.
Delay between excavation and mapping
A clear correlation between a delay in the mapping time after excavation to the stress
failures observed is seen; more and stronger failures are observed immediately after
excavation. In the volcanite sections, half of the mapping was done within a day of
excavation and only 15% after three days (Figure 6-5). In observations obtained within a
day of excavation, 80% of the sections show some sign of stress failure. After three
45
days, the corresponding amount is less than half. There is an equal decrease in visual and sonic stress failure observations.
60% ~-------------------------------------------------------
~ ~
~ 1/) 1/)
~ (;) "'0 Q) 0.. 0.. CO E -0 Q) C) CO "E Q)
~ Q) a..
40%
30%
20%
10%
0%
... 1 (555m)
1 .. . 2 (241m)
2 ... 3 (175m)
Days between excavation and mapping
>3 (162m)
Figure 6-5. The effect of the time period between excavation and mapping on the
observed stress failures in volcanite. The failure classes JO ... J4 are explained in
Chapter 4.
Depth level
The effect of the depth level, which should correlate with the stress magnitude, is quite
obvious when looking at the observations in volcanite within two days of excavation
(Figure 6-6). The occurrence of visible stress failures clearly increases with depth, but
the total number of stress failures does not increase monotonically. One explanation for
the result can be the unequal representation of different tunnel orientations at different
depth levels. Generally, the occurrence of stress failures seemed to be 20% higher below
the+ 1300 levels than above the+ 1200 level. The effect of the depth is practically the
same if data within one day of excavation is considered.
Tunnel shape
About 10% of mapped volcanite tunnels are in crossing areas or the profile is wider than
the typical tunnel size. These areas have 10% to 15% higher stress failure potential than
straight or slightly curved tunnels (Figure 6-7). This can be understood by higher stress
concentrations and lower confmements.
46
Occurrence of stress induced failure
100%~~~~-------------------------------------------------------------,
80% . J2 ... J4 t--------------------------------1
z <= 1200 m (104m)
1200 m< z <= 1300 m (454m)
Depth range, total mapping
z > 1300 m (225m)
Figure 6-6. The effect of the depth level on observed stress failures in volcanite;
observations within two days of excavation. The failure classes JO ... J4 are explained in
Chapter4.
Occurrence of stress induced failure
100%~~~~-------------------------------------------------------------,
80% . J2 .. . J4 1---------------------------------t
60% ~----------------------------------i
20% 1-------~
0%1----Cross-section or Widening
(95m)
Tunnel shape, total mapping
Straight or Curved (869m)
Figure 6-7. The effect of the tunnel shape on observed stress failures in volcanite;
observations within three days of excavation. The failure classes JO ... J4 are explained
in Chapter 4.
47
Direction of excavation and orientation offoliation
In Pyhasalmi vulcanite, the direction of excavation compared to the major principal
stress and orientation of the foliation can not be studied independently. The fact that the
dip direction of the foliation is towards the ore body and increases when approaching the
ore body (Figures 5-7 to 5-9) makes the situation more complex. The average trend of
the major principal stress is 115° I 295°, thus, stress condition is most favourable when
the tunnel is driven in these directions. The highest secondary stress at the tunnel roof is
induced when the tunnel is driven in a perpendicular direction to the major principal
stress; that is, in the directions 25° or 205°. The average dip direction of the foliation is
185° and 50% ofthe failure observations are between 170° to 205° and 90% between
120° to 225°.
The stress failure observations from a day after excavation and within three days of
excavation correlate with the orientation of major principal stress and dip direction of
foliation. The probability of stress failure is highest when the tunnel is excavated
perpendicular to the major principal stress or with the strike of foliation (Figures 6-8 and
6-9). In the latter case, the shear failure potential of the foliation planes is highest when
the tunnel is driven in the strike direction of the foliation. On the other hand, the lack of
data in many directions, and the change of the foliation plane dip direction, make this
conclusion even more uncertain. Occurrence of stress failures with tunnel excavation direction - volcanite, observation period is one day after excavation
Trend of the major principal stress 280°-310°
oo (32m)
180° (100m)
go• (26m)
. J2 ... J4
(~~~:) T"!n~ofthemajor pnnctpal stress 280°- 3to•
Figure 6-8. The effect of the excavation direction on the observed stress failures in
volcanite when the observation time was less than one day after excavation.
48
The result is almost the same even if the dip direction of the foliation is limited between
165° to 215° (see Figure 5-8) and all the stress failure observations related to geological
structures such as lenses, structures, and veins are rejected (Figure 6-1 0). On the other
hand, the probability of visual stress failure is decreased by some 20%.
Occurrence of stress failures with tunnel excavation direction - volcanite, observation period is three days after excavation
Trend of the major principal stress 280"- 310"
270° (91m)
o· (74m)
180° (176m)
go· (39m)
J2 ... J4
Trend of the major principal stress 28o·- 3to·
Figure 6-9. The effect of the excavation direction on observed stress failures in
volcanite when the observation time was less than three days after excavation.
Dip o(foliation
The effect of the dip angle of the foliation was studied in three tunnel excavation directions 120°, 180°, and 240° having the highest amount of data. The observations
were limited to the first two days after excavation. In these cases, the occurrence of
stress failure is approximately 15% higher with lower dip values, but the type of damage
does not correlate with the dip (Figure 6-11 ).
49
Occurrence of stress failures with tunnel excavation direction - volcanite, observation period is three days after excavation,
. J2 ... J4
-dip direction of foliation limited between 165 and 205 ,~ata with geological structures rejected
Trend of the major principal stress 280°-310°
(22m)
1so· (79m)
60° (Om)
go• (21m)
Trend of the major principal stress 280· - 310°
Figure 6-10. The effect of excavation direction on observed stress failures in volcanite
when the observation time was less than three days after excavation. The dip direction
of the foliation is limited to between 165°-205° and all data associated with local
geological structures is rejected.
Fabric of rock
The effect of fabric, that is, particle arrangement, was studied based on the same data
set, which was used previously for the dip of the foliation study. The volcanite is
massive (M), and slightly (S 1) or moderately schistose (S2). The fabric does not
correlate with the stress failure data so that it can be seen over the other factors (Figure 6-12). Neither is any correlation seen if all volcanite data is studied.
Rock mass quality
The effect of the rock mass quality was studied based on the same data set, which was
used in the two previous studies. The Geological Strength Index GSI of the volcanites
varied between 58 and 82 and the average was 69 (Figure 6-13). In two tunnel
directions, severe damage was observed in the better rock quality, and in one orientation
no correlation could be seen. If all data from volcanite tunnels is studied then more
severe stress failures are observed if the GSI value is above 65.
50
Occurrence of stress induced failure
100%r---------------------------------------------------~~~~~---------,
80% 1----:= --------------------------l . J2 ... J4 f-------1
60%
40%
20 o/o
0%
120' dip< 60
(93m)
120' dip> 60
(36m)
Ill I
180' dip< 60 (73m)
180' dip> 60
(67m)
Tunnel excavation direction, dip of foliation
240' dip< 60
(22m)
240' dip> 60
(56 m)
Figure 6-11. The effect of the dip angle of the foliation on observed stress failures in
volcanite. The observation time was less than two days after excavation.
Occurrence of stress induced failure
100% r-11------------------------------------------------------------;=====~
80% 1------------------------------------t . J2 .. J4
60% 1--~---------------------------.. ------------------------------~
40% t------
20% ~-------
0 % ~------..-
120', M 120', S1 120', S2 180', M 180', S1 180', S2 240', M 240', S1 240', S2 (13m) (93m) (44m) (27m) (97m) (38m) (34m) (53m) (13m)
Tunnel excavation direction, Fabric type of Volcanite
Figure 6-12. The effect of rock fabric on observed stress failures in volcanite. The
observation time was less than two days after excavation. M= massive, SI =slightly,
and S2= moderately schistose.
51
Occurrence of stress induced failure
100%r-.r------------------------------,.----------~~~~~w-----,
80% ~--------------------------------1 . J2 ... J4 1----1--- --1
60% ~~--~-~---~~-----~---~-----~.--~
40% ~-------------------t----
0% ~------120" 120" 120" 180" 180" 180" 240" 240" 240" < 65 65 ... 75 >75 (Om) < 65 65 ... 75 >75 (Om) < 65 65 .. . 75 >75 (Om) (22m) (45m) (18m) (61m) (85m) (7m) (38m) (42m) (8m)
Tunnel excavation direction, GSI value
Figure 6-13. The effect of rock mass quality on observed stress failures in volcanite.
The observation time was less than two days after excavation.
Other [actors
The effect of grain size or degree of weathering could not be studied because all the volcanites are fme grained and unweathered.
6.3 Stress failures in shafts
During the ftrst mapping period, there was no opportunity to proceed with the rock mechanical mapping in the shafts. Some visual observations were made from the tunnels above and below the shafts. In all cases, the stress failures were clear and were seen in the directions 10°-30° and 190°-210°. Inside a shaft of3.1 m in diameter (Figure 14), the width of the failure area was about one metre and the depth was a few dozen centimetres.
The stress failures in shafts are especially interesting because they can be considered
systematic and continuous and the effect of the excavation method is found to be minor.
These observations confmn the in situ stress measurement results, where the major in
situ stress is in the direction of 110° I 290°.
52
Figure 6-14. Stress failures in a 3.1 m wide shaft (ore pass) at + 13 7 5 level; the failure orientations are 20° (left) and 200° (right).
53
7 ROCK MECHANICAL DATA
7.1 Intact rock
For the new mine project, several uniaxial compressive and indirect tensile tests were
conducted for the main rock types (Hakala et al., 1998 and 1999). Rock mechanical
parameters like uniaxial compressive strength ( crp), crack damage stress ( O"cct), crack
initiation stress ( O"ci), tensile strength ( O"t), and the elastic parameters, Young's modulus (E) and Poisson's ratio (v), have been defined (Table 7-1). The data of intact rock is
only missing for serizite schist and serizite quartzite, but, from the previous studies of old mine area, it is known that the average uniaxial compressive strength of these rock
types varies between 40 MP a - 80 MP a.
Table 7-1. Strength and deformation parameter values for OKP intact rock types.
Rock type Young's modulus ( GPa ) Poisson's ratio
ave stdev n ave stdev n
Volcanite, Felsic 68 24 % 33 0.24 33% 33
Volcanite, Mafic 76 21 % 18 0.26 27% 18
Pegmatite 63 13 % 4 0.23 25 % 4
Ore, Zinc 98 37% 18 0.32 38% 18
Ore, Copper 139 33 % 22 0.30 31 % 25
Pyrite 120 21 % 12 0.34 45% 12
Rock type Uniaxial compressive strength ( MPa) Tensile strength ( MPa )
ave stdev n ave stdev n
Volcanite, Felsic 241 18% 24 17 30 % 32
Volcanite, Mafic 206 45 % 17 15.2 49 % 10
Pegmatite 119 10 % 4 6.8 19 % 2
Ore, Zinc 92 39 % 18 5.9 41 % 11
Ore, Copper 123 38% 22 6.1 36 % 16
Pyrite 93 24 % 12 6.4 28% 8
The following is mainly a study of the volcanites. These are the major rock types in the mapping area. No difference between felsic and mafic volcanite can be found from the
test results (Figures 7-1 to 7-3). The angle of foliation is not defined for all specimens.
However, based on the data available, some correlation with foliation and strength of
mafic volcanite has been found, but the deviation in the results is very high and so no
definite conclusions can be drawn (Figure 7-4). Because the felsic and mafic volcanites
are typically frequently layered and the deviation of the results overlap strongly, it was
decided to treat them as one volcanite rock type.
54
450 ~~~~~~~~--------------------------------------------,
400
350
300
250
200
150
100
50
0
-- -+ -- Felsic volcanite
- • -- Mafic volcanite .. :· ,• '' '' '' '' ' '
' ' ' ' .•. .~. ' ' . • ., ., '• •, ., '' '' ''
•
• • ·· -- ' ' ' :\ ! \ \ : . ', : ~--~ ,:':~ .. · .... ·... :' : -· .
/:\'·,,, • / \~./ ': .. ., .•f-·-; ·-.•. ' ' . \ \ :: ,, ,' ' '
' ... . .. •
' ' ·~:~ -· ' : • . -..--*/ . ' ' \ ' ' ' \,! •
. : • ·--.
·-:' .. I
•
0 5 10 15 20 25 30
Number of tests ( pes )
Figure 7-1. Uniaxial compressive strength of felsic and mafic volcanites.
30 ~------------------------------------------------------------------~
25
20
15
10
5
• ,. ,. •' •' :· ': '' '' '' ' '
~ '•' •: )'.
j\f :•\ • ! j\ .. i \ :'
:•. ! : ~ • ~ ::' -. .... ' ! :
\; '!( .~ ·-· ~
A ' '
~ ,' •
.. j\.,
. ~.
•
' '
·-· ~-'
•
.. •, '• '' . ! \ :' \
: I I I
' -·-·· \ ... \ .. ' ••
• •'. '' '' ' '
' ' ' ' ' ' • ~·
-- --+--- Felsic volcanite
- • - Mafic volcanite
0 ~--------------~--------~------~------~----------------._------~ 0 5 10 15 20 25 30 35 40
Number of tests ( pes )
Figure 7-2. Indirect tensile strength of felsic and mafic volcanites.
55
100 ~----------------------------------------------------------~
90
80
70
60
50
40
30
20
10
0 0
Figure 7-3.
., . . ~ : -· - ~ -·.. t-. ~ \ : ' , .~... ~ ... . ... '-! \
• :: '. : ', : : ·. :1. • -.. ' .~ · !i • ~· : ' ,.., ' .. . • • . ..
•
'·r,l \ i \ ::: '.\ t -· ·· \ ~ '.,! ~.! ',t: ' ,, ... . . . : . ..
'i •
/ ,. \ ' . ' '
·-· i \\! • I 1 1 1
~ I :
l
5 10 15 20 25
Number of tests ( pes )
/'-.. ·- ' ' . ' ' -· .. ' : •
-- -+-- Felsic volcanite
• Mafic volcanite
30 35
Young 's modulus offelsic and mafic volcanites.
0 +-------------~------------~------------~ 0 30 60 90
Foliation ( o )
40
Figure 7-4. Uniaxial compressive strength of felsic and mafic volcanites versus the
foliation angle.
56
Table 7-2 and Figures 7-5 to 7-10 summarise the statistical values of the rock mechanical parameters of volcanite. In all the figures, the cumulative probability together with the normal distribution at 10%, median, and 90% values, as well as
quartals are presented.
Table 7-2. Statistical parameters of normal distribution and quartiles for volcanic
rock.
Normal distributions Quartiles
10°/o Median 90°/o Ql Q2 Q3
Youngs modulus, E 61 75 89 64 76 85
Poisson ratio, v 0.21 0.26 0.3 0.23 0.26 0.29
Tensile strength, crt 9.3 16.5 23.8 12.1 17.2 20.4
Crack initiation stress, <Jci 63 101 138 81 96 121
Crack damage stress, crcd 106 184 261 150 172 214
Peak stren 129 222 316 172 232 261
Cumulative probability
1.0
• E
- NDISTforE
0.8
0.6
0.4
0.2
0.0 i...----------------------....1 20 40 60 80 100 120
Young's modulus ( GPa)
Figure 7-5. Deviation of Young's modulus ofOKP volcanite.
Figure 7-6.
Figure 7-7.
Cumulative probability
1.0
• n
- NDISTforn
0.8
0.6
0.4
0.2
57
ave=0.26 · I I I
n(0.1)=0.21 1
I I I I I I I
• Q3=0.29
• Q2=0.26
• Q1=0.23
0.0 ...._...._.._ ....... _._......___._....._.....__...._ _ _.._ ........ ___._..__...._.._ ....... _.___,
0.0 0.1 0.2 0.3 0.4
Poisson's ratio ( mm/mm )
Deviation of Poisson 's ratio of OKP volcanite.
Cumulative probability
1.0 ,.....-------------------...... .......,
• St
0.8
0.6
0.4
0.2
0.0 ............................................. _._ .... _ ................... _ ........................... ....._ ...................... __.
0.0 5.0 10.0 15.0 20.0 25.0 30.0
Tensile strength ( MPa)
Deviation of tensile strength Gi of OKP volcanite.
Figure 7-8.
Figure 7-9.
58
Cumulative probability
1.0 ~~~~~~----------~P----,
• Sci
- NDIST for Sci
0.8
n(0.1)=63
0.6
0.4
' 03=121
0.2
02=98
01=81
0 50 100 150 200
Crack initiation stress ( MPa )
Deviation of the crack initiation stress crci of OKP volcanite.
Cumulative probability
1.0 r:=========:;----------::::::::::~----,
0.8
0.6
0.4
0.2
0.0 0
•Sed
- NDIST for Sed
n(0.1)=106 • I
50 100
ave=184 j
I I I I I I I I ...
• I
• • • • ..
' 02=172
'01=150
150 200
' 03=214
250 300
Crack damage stress ( MPa )
350 400
Deviation of the crack damage stress Cicd of OKP volcanite.
59
Cumulative probability
1.0 ,.--------------..,....-----,
0.8
0.6
0.4
0.2
• Sp
- NDISTforSp
n(0.1)=129
ave=222
• 03=261
02=232
• 01=172
0.0 ................ .....__ ____ ............................ __ ........,_._......._._....._ ........ _..._......,
0 100 200 300 400 500
Peak strength ( MPa)
Figure 7-10. Deviation of the peak strength ap ofOKP volcanite.
For the purposes of the stress failure study, the probability envelopes for the strength
and critical stress states are defmed (Figures 7-11 and 7-12). Tensile and uniaxial
compressive strength data (see Figs. 7-7 and 7-10) is used in pairs; high tensile strength
correlates with the high uniaxial strength of the material. The cumulative probability
strength envelope presentation means that with a probability of n%, the strength
envelope is below the C.p. n% envelope. Thus, C.p.50% is an average strength envelope
and with 95% probability the strength envelope is between the envelopes C.p. 2.5% and
C.p. 97.5%. The stress envelope for crack damage ( crcd) follows the Hoek-Brown criteria with constant mi -value equal to mi for peak strength (Hoek & Brown, 1980).
The crack initiation ( crci) envelope is based on the formula cr1 - cr3 = crci (Read et al. , 1998). For peak strength, crcd and O'ci the cr1 values at cr3 = 0 are according to Figures 7-8 ... 7-10.
60
Major principal stress ( MPa )
300
- c .p.=2.5%
250 - c .p. =20%
- c .p.=40%
200 C.p. =60%
- c.p. =BO%
150 - c .p. =97.5%
100
50
0 ~.__. __ ~~~-----------------------------------30 -15 0 15 30 45 60
Minor principal stress ( MPa )
Figure 7-11. Probability of strength envelope ofOKP volcanite.
Major principal stress ( MPa )
300 ~------------~--~~~----~~--~-----------
250 Crack damage
200
150
100
50 I
: CT3 I CT1 = 5% I
--c .p. =2.5%
--c .p.=20%
--c .p.=40%
C.p. =60%
--c .p. =BO%
--- s3/s1=0.05
0 ~.__. __ ~~~--~----------------------._----~
-30 -15 0 15 30 45 60
Minor principal stress ( MPa )
Figure 7-12. Probability of critical stress states envelopes ofOKP volcanite.
61
7.2 Rock Mass
The rock mass parameters are defmed from intact rock parameters together with rock
mass classifications values, that is, joint characteristics (Hoek et al. , 1995). When the
GSI-value is above 25, the rock mass strength envelope can be defmed using the
following equations:
m6 = m;exp (GSJ -100J
28
(GSJ -100J s=exp --
9--
where
a = 0.5, constant for Hoek- Brown failure criteria
ac = uniaxial compressive strength of intact rock
(7-1)
(7-2)
(7-3)
m; = is a Hoek-Brown intact rock strength parameter, which is defined as best
fit for the intact rock tensile and compressive strength envelope. The form
of envelope is equal to Equation 7-1 assuming that s = 1 and a=O. 5.
Taking into account the deviation of the GSI value and intact rock strength values
(Figures 7-7, 7-10 and 7-13), the rock mass strength envelope is defmed using the
Monte-Carlo simulation (Figure 7-14).
62
Cumulative probability
1.0 ,...-------------------~....-----,
• GSI
- NDIST for GSI
0.8
0.6
0.4
0.2
50 55 60
n(0.1)=69 1 I I I I I I I I
65 70
mean=76 I I I I I I I I I I I I I
I
,1! . .1 l
: 0 3=79
: 02=77
0 1=72
75 80
GSI value
Figure 7-13. Deviation ofGSI-value ofOKP volcanite.
Major principal stress ( MPa )
-30 -15 0 15 30 Minor principal stress ( MPa )
85 90 95
- c .p. =2.5%
- c .p. =20%
- c .p. =40%
C.p. =60%
- c.p. =SO%
- c .p. =97.5%
45 60
Figure 7-14. Probability envelopes for Hoek-Brown rock mass strength ofOKP
volcanite analysed using the Monte-Carlo simulation; C.p = cumulative probability.
63
7.3 In situ state of stress
In the new mine area the in situ state of stress was measured in two. Firstly, the
measurements were obtained in three locations at the + 1125 level and, secondly, at the
+ 1325, + 1350, and+ 1375 levels (Ledger, 1999; Mononen, 2000). At least two
measurements were obtained at each location. At the + 1125 level, five of the seven
measurements were successful, but, at deeper levels all measurements were quite
uncertain because of core damage through disking.
At the + 1125 level, two measuring locations were in the, on different sides of the ore
body, and the third location was in the ore. The difference between the five successful
measurements was very minimal. The major and intermediate principal stresses were
almost horizontal and the minor was vertical. The average major principal stress was
65 MPa dipping 5 degrees to direction 310. The average horizontal stress ratio crH/crh was 1.6 and stress ratio CJH!crv was 2.0 (Figure 7-15).
Maximum Horizontal Stress I Vertical Stress
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 0 ~~~-6-o ~ 0 0
0 0 ,,,"'"' • o 0 00 ~ ~c 4) 0 m> 00 -/ c;tJ J~~o B( ~< • I ~ ~·· X o· il / oo Ji .
~ I ~ eon ~ •
~ oo " o -X~ iJ " ~~~ ----- ..,
I 0 e o I 0 0 r.\a .... ~ ,()~~ I
500
I ooCO ~ 0 ./'
, .. I 4 0 I • 0 ~/ I 0 c •
0 °c oo ~/,9"~t 0 0 i 0
1000
c o,. ,. e o Australia i
>J,. USA
0 I 0 1500
0 0 ,' • Canada 0 / Scandinavia
) I • I I 0 South Africa 0 I
I 0 Others I
2000 I
X Finland, Posiva I I
i4) PyhAsalmi, old ou I I
0 I I• PyhAsalmi, 1999 I
I 2500 I I
i I
Depth (m)
Figure 7-15. The stress ratio versus depth of the latest in situ stress measurements at
the Pyhasalmi mine and Posiva investigation sites together with results around the
world (Modified from Hoek & Brown, 1980).
64
In the+ 1350 level, all the measurements were done in the volcanites. The only successful measurement as well as the failed ones indicate major horizontal stress of 75 MPa in a direction of290° (Figure 7-16). The horizontal stress ratio crH/crh was 1.7
and the vertical crH/crv was 1.8.
Figure 7-16. Orientation of the major and minor principal stresses.
The stress failure observations in shafts confirms the bearing of the major principal stress to be between 280 to 310 MP a, but the magnitudes of all principal stress components at the+ 1350 level are not certain.
65
8 STRESS STRENGTH ANALYSES
8.1 Cases and failure criteria
The numerical stress strength simulations were done using a three dimensional boundary
element code Examine3D (RocScience, 1999). The volcanite rock mass surrounding the
excavations was assumed to be homogenous linear elastic. This assumption was made
because no reliable test data on the effect of the anisotropy was available. Two in situ
stress regimes were studied; magnitudes of 65 MP a or 7 5 MP a for major in situ
principal stress were used, which corresponds to the depth levels of+ 1125 and+ 1350.
The horizontal stress ratios ( crH/ crh) have a value of 1. 6 or 1. 7 and the horizontal -
vertical stress ratio ( crH/crv) was 2.0 or 1.8.
The simulated horizontal tunnels were the typical 5 m * 5 m face area with a vertical wall of 4 m height (Figure 8-1). The vertical shaft was 3.1 m in diameter (Figure 8-2). In
both cases, the length of excavations was over ten times the maximum radius to secure a
two dimensional stress state at the middle of the excavation. The excavation faces were
chamfered to avoid stress discontinuity. The development of secondary stresses with
advancing coring was monitored by setting lines of "field points" parallel to the
excavation, extending five radii from the face in both directions. In the tunnel model,
the stresses were monitored at the tunnel roof, in the upper corner, and in the middle of
the wall. In each location, there were three field point lines at the distances of 1 cm,
5 cm, and 15 cm from the excavation surface. In the shaft model, the field point lines
were on the sides of the major and minor horizontal stress and the third was between the
first and second.
Five excavation directions were simulated for the tunnel models. The angle between the major horizontal stress and tunnel axis was assumed to be 90, 60, 45, 30, or 0 degrees.
In the case of the vertical shaft model, similar studies were not needed because of the circular shape. The elastic principal stress paths of these eight tunnel simulations and two shaft simulations were compared to the following failure criteria:
1) Peak strength of intact rock.
2) Failure criteria presented by Read et al. (1998), which is based on crack damage and crack initiation strength surfaces as well as confining stress. In this criterion, the crack damage strength is assumed to be the true strength of the rock. The rock
is assumed to be damaged, if the stress exceeds the crack initiation surface and
confinement is less than 5% of maximum stress. This damage can reduce the
crack damage strength. The rate of strength reduction is not given, but can be
defined from in situ observations.
3) Hoek-Brown rock mass strength criteria (Hoek & Brown, 1997 and 1980).
66
All the failure criteria were presented in a form of probability based on laboratory test
values and mapped rock mass quality, which are presented in previous chapters (Figures 7-11 , 7-12 and 7-14).
a.)
:::: ): XX == "rxx ==:::::._ X · \ _.-:--- >< 'i'x fXI'x ----_.-:--- >< X X IX X
--------- >< X X I.X X ------- >< X X IX X ----_.-:--- >< X X I XIX
-------------~ X X I XIX ----_.-:--- >< IX X I XIX ---b.)
o4• • ••
XXX
)'\)X If\
X X X X X X X
X X IX X X 1xj~
X X XIX X IX .X f-- l-+-
X X lXlX X I XIX
X X lXlX X XIX
X X XIX X XIX
X X IXI XIX I XIX
Width 5 m Height 5 m Length 40 m
l : tl I .I
si s3
Figure 8-1. Simulation model for the tunnel (a) and the lines of .field points close to the tunnel face.
~sl
/l's2 s3
Height 16 m Diameter 3.1 m
67
a. b.
Figure 8-2. Simulation model for the shaft (a) and the lines of field points close to
the tunnel face.
8.2 Tunnel simulation
Secondary stress state at tunnel surface
The secondary state of stress refers to the situation when excavation has passed the observation point and the advancing excavation no longer affects that point. The highest compressive stresses take place at the tunnel roof and upper corner when the major
horizontal stress is perpendicular to the tunnel axis (Figures 8-3 to 8-5). The maximum
stress concentration values are 135 MPa to 155 MPa being 2.0-2.1 times the maximum
in situ stress. The minor principal stress is zero at the roof, but in the upper edge it is
2 - 5 MPa in tension. At the tunnel wall, a tension stress < 2 MPa can take place
regardless of the trend of the major horizontal stress.
68
Transient stress state at fictitious tunnel surface
Transient stress state refers to a situation when the excavation face passes the
observation point. When the tunnel is excavated by the drill and blast method, as in Pyhasalmi, all points do not follow the complete transient stress path presented in
Figures 8-3 to 8-5, instead they are shifted rapidly from one transient state to another.
But, within one cut length, the union of stress paths of points in a line parallel to the
tunnel axis forms a complete stress path.
The maximum transient deviatory stresses take place at the tunnel wall when the major
horizontal stress is at 45° to the tunnel axis (Figure 8-5). The maximum principal stress
is 310 MPa to 360 MPa, being 4. 7 times the major horizontal stress, and the minor
principal stress is about 30 MPa. There are remarkable transient tensile stresses.
Major principal stress ( MPa )
Pyhasalmi Volcanite -principal stress cr1/cr3 path 1 cm above tunnel - angle between tunnel axis and in situ cr1
stress: 90°, 45° and oo 280 -in situ: cr1 , sH =75 MPa, crH/crh=1 .7, crH/crV= 1.8
240
200
160
120
80
40
-30
- crci and crcd stress envelopes with different cumulative probabilit
-15 0 15 30
Minor principal stress ( MPa )
45 60
--lr--90°
~45°
-o--oo
--C.p. =2.5%
--C.p.=20%
--C.p. =40%
C.p. =60%
--C.p.=80%
--C.p. =97.5%
--- s3/s1=0.05
Figure 8-3. Principal stress path at the tunnel roof surface using failure criteria
presented by Read et al. (1998), Level+ 1350, CJ'H = 75 MPa.
69
Major principal stress ( MPa )
320 Pyhasalmi Volcanite - principal stress a1/a3 path 1 cm above tunnel - angle between tunnel axis and in situ a1
280 stress: 90", 45" and o•
- in situ: a1 , sH =75 MPa, aH/ah=1 .7, aH!aV= 1.8 - ad and acd stress envelopes with different
cumulative probabilit
240
200
160
120
80
40
0
-30 -15 15 30
Minor principal stress ( MPa )
45 60
--tr--900
---+--45°
--o--oo
--C.p. =2.5%
--C.p. =20%
--C.p. =40%
C.p. =60%
--C.p. =SO%
--C.p. =97.5%
--- s3/s1=0.05
Figure 8-4. Principal stress path at the tunnel edge surface based on the failure
criteria presented by Read et al. (1998), Level+ 1350, aH = 75 MPa.
Major principal stress ( MPa)
380 r-----------------------------~----~--~=:~----------~ PyMsalmi Volcanite -principal stress a1/a3 path 1 cm above tunnel - angle between tunnel axis and in situ a1
320 stress: oo•, 45• and o• -in situ: a1 , sH =75 MPa, aH/ah=1 .7, aH/aV= 1.8 - ad and acd stress envelopes with different
cumulative probabilit 280
240
200
180
120
80
40
-30 -15 0 15 30 45 60
Minor principal stress ( MPa )
--~r--soo
---+--45°
--o--oo
--C.p. =2.5%
--C.p.=20%
--C.p. =40%
C.p. =60%
--C.p. =80%
--C.p. =97.5%
--- s3/s1=0.05
Figure 8-5. Principal stress path at the tunnel wall surface with failure criteria
presented by Read et al. (1998), Level+ 1350, aH = 75 MPa.
70
Stress state at 15 cm depth from tunnel surface
Apart from the maximum secondary stresses at the tunnel roof, the highest stresses at
15 cm distance from tunnel surface are remarkably lower than at the surface (Figures
8-6 to 8-8). The transient stresses, in particular, decrease 40% to 50% from the surface
values, which means that the highest transient stresses take place close to the tunnel face
and to a very limited extent.
Major principal stress ( MPa )
320~----------------------~--~----~----------------~
Pyhasalmi Volcanite -principal stress cr1/cr3 path 15 cm above tunnel - angle between tunnel axis and in situ cr1
stress: go•, 45• and o• 280 -in situ: cr1 , sH =75 MPa, crH/crh=1.7, crH/crV= 1.8
- crci and crcd stress envelopes with different cumulative probabilit
240
200
160
120
80
40
0~~--~_.~ __ ._ __ ~--------------------------------~ -30 -15 0 15 30 45 60
Minor principal stress ( MPa )
---tr--900
--cr-o· --C.p. =2.5%
--C.p.=20%
--C.p. =40%
C.p. =60%
--C.p. =80%
--C.p. =97.5%
--- s3/s1=0.05
Figure 8-6. Principal stress path at 15 cm distance from the tunnel roof surface by
using the failure criteria presented by Read et al. (1998), Level+ 1350, an= 75 MPa.
71
Major principal stress ( MPa )
320~----------------------~----~--~~----------------~
Pyhiisalmi Volcanite - principal stress cr1/cr3 path 15 cm above tunnel - angle between tunnel axis and in situ cr1
stress: 90°, 45° and oo 280 - in situ: cr1 , sH =75 MPa, aHICJh=1 .7, aHiaV= 1.8
240
200
160
120
80
40
- crci and crcd stress envelopes with different cumulative probabilit
-30 -15 15 30
Minor principal stress ( MPa )
45 60
--tr--90°
-45°
-o--oo
--C.p. =2.5%
--C.p. =20%
--C.p. =40%
C.p. =60%
--C.p. =80%
--C.p. =97.5%
--- s3/s1=0.05
Figure 8-7. Calculated principal stress path at 15 cm distance from the tunnel edge
surface based on the failure criteria by Read et al. (1998), Level+ 1350, O'H = 75 MPa.
Major principal stress ( MPa )
320~----------------------~----~--~------------------~ ,------------~
Pyhilsalmi Volcanite -principal stress cr1/0"3 path 15 cm above tunnel - angle between tunnel axis and in situ cr1
280 stress: 90°, 45° and oo -in situ: cr1 , sH =75 MPa, aHicrh=1 .7, crHiaV= 1.8 - crci and crcd stress envelopes with different cumulative probabilit
240
200
160
120
80
40
-30 -15 15 30
Minor principal stress ( MPa )
45 60
-45°
-o--o·
--C.p. =2.5%
--C.p. =20%
--C.p. =40%
C.p. =60%
--C.p. =80%
--C.p. =97.5%
--- s3/s1=0.05
Figure 8-8. Principal stress path at 15 cm distance from the tunnel wall surface using
the failure criteria by Read et al. (1998), Level+ 1350, O'H = 75 MP a.
72
Failure criteria comparison
The stress paths, obtained by using the theories by Read et al. (1998), Hoek-Brown and
peak strength criterion, are compared with the observed stress failures (Figures 6-5, 6-6,
6-7, 8-3, 8-9 and 8-1 0). It can be seen that the criterion of the peak strength
overestimates the in situ strength, but the other two criteria present reasonable results. Therefore, the peak strength is rejected from further studies.
Based on the comparison of the mapped stress failure data with the stress failure criterion on elastic stresses, the following assumptions were made:
a) probability of visual stress failure ( J2 ... J4) was defined, based on final
secondary stress state,
b) probability of noise stress failure ( J1 ) was defined, based on transient stress state,
c) probability for noise stress failure is taken into account only if it is higher than the probability for visual stress failure,
d) yield strength reduction was ignored for the Read et al. (1998) criteria, because
the transient phase is valid only for limited discontinuous areas; that is, the
current and previous areas around the tunnel face having spacing equal to the cut length, and,
e) finally, the comparison is based on the union of the roof, corner, and wall stress
failure probabilities where higher stress failure class dominates (Figure 8-11 ).
The comparison of the resulting failure probability presented in Figures 8-11 to 8-15 with mapping results (Figures 6-5 to 6-7) shows that the Read et al. (1998) failure
criterion overestimates the strength, whereas Hoek-Brown underestimates it. At 15 cm
distance from the tunnel surface, the Hoek-Brown results are more reasonable, but the high stresses at the upper edge dominate the stress failure behaviour in the excavation
orientations 30 I 210 (Figure 8-11 b). However, the stress redistribution after stress failure is not taken into account, which would increase the stress failure potential at 15 cm depth.
Based on elastic stresses and homogenous material, the in situ failure criterion for the tunnel seems to be between the strength envelopes presented by Read et al. ( 1998) and Hoek-Brown (Hoek & Brown, 1997). It must be noted that the degree of anisotropy of
the Pyhasalmi volcanite is unknown nor was it taken into account in the simulations.
Major principal stress ( MPa )
Pyhllsalmi Volcanite - principal stress cr1/cr3 path 1cm above tunnel - angle between tunnel axis and in situ a1
stress: go• , 45" and o• 280 - in situ: a1 , sH =75 MPa, crH/crh=1 .7, crH/crV= 1.8
- HB intact rock peak strength envelopes with different cumulative probability
240
200
160
120
80
40
-30 -15
73
15 30
Minor principal stress ( MPa )
45 60
-~r-go•
--45•
-o-o·
- C.p. =2.5%
- C.p.=20%
- C.p. =40%
C.p. =60%
- C.p. =BO%
- C.p. =97.5%
Figure 8-9. Principal stress path at the tunnel roof surface with peak strength
envelopes. Level+ 1350, as= 75 MPa.
Major principal stress ( MPa )
3~r-------------------------------------~------------------~ r------------~ Pyhasalmi Volcanite -principal stress cr1/cr3 path 1 cm above tunnel - angle between tunnel axis and in situ cr1
stress: go• ' 45" and o· 280 - in situ: a1 , sH =75 MPa, crH/ah=1 . 7, aHiaV= 1.8
- HB rock mass strength envelopes with different cumulative probability
240
200
160
1~
80
40
oL---------------~----------_.----------------------------~ -30 -15 15 30 45 60
Minor principal stress ( MPa )
-tr-9o·
-o-o•
- C.p. =2.5%
- C.p.=20%
- C.p. =40%
C.p. =60%
- C.p. =80%
- C.p. =97.5%
Figure 8-10. Principal stress path at the tunnel roof surface by using the Hoek-Brown
failure criteria (Hoek & Brown, 1980). Level+ 1350, as= 75 MPa.
a)
b)
74
Probability of stress failure at tunnel roof with tunnel excavation direction - 1350 level, sH=75 MPa, 1 cm, Read et al.
Trend of the major principal stress -......_ 12o• 1 3oo· 300•
255"
• Yielded, J2 ... J4
• Damaged, J1
• Elastic, JO
120" Trend of the major
'';';~~~~S:ss
Probability of stress failure in upper corner of tunnel with tunnel excavation direction - 1350 level, sH=75 MPa, 1 cm, Read et al.
. Yielded, J2 ... J4
• oamaged, J1
Trend of the major principal stress -......_ 120• I 300• 300"
255"
• Elastic, JO
120" Trend of the major ...._ principal stress
........... 120°1300°
Figure 8-11. Probability of stress failure at the tunnel a) roof b) upper edge, and
... (continues on the next page)
c)
d)
75
Probability of stress failure at tunnel wall with tunnel excavation direction - 1350 level, sH=75 MPa, 1 cm, Read et al.
Trend of the major principal stress ......... 120· 1 3oo· 300o
255°
Probability of stress failure at tunnel periphery with tunnel excavation direction - 1350 level, sH=75 MPa, 1 cm, Read et al.
Trend of the major principal stress ......... 12o· 1 Joo• 300o
255°
• Yielded , J2 ... J4
• oamaged, J1
• Elastic, JO
120° Trend of the majo ~ principal stress . --...... 120" 1300"
• Yielded, J2 ... J4
• oamaged, J1
• Elastic, JO
120• Trend of the majo, ~ principal stress · --...... 12o· 1 Joo•
Figure 8-11 continues c) wall and d) union for the tunnel periphery based on failure
criteria presented by Read et al. (1998), Level + 1350, an = 7 5 MP a.
a)
b)
76
Probability of stress failure at tunnel periphery with tunnel excavation direction - 1350 level, sH=75 MPa, 1 cm, Read eta/.
Trend of the major principal stress ........... 12o· 1 3oo· 300o
255°
Probability of stress failure at tunnel periphery with tunnel excavation direction - 1350 level, sH=75 MPa, 1 cm, Hoek-Brown
Trend of the major principal stress ~ 120.1300° 3000
255°
• Yielded, J2 ... J4
• oamaged, J1
• Elastic, JO
120• Trend of the major
'" principal stress ........... 120" 1300.
• Yielded, J2 .. . J4
• oamaged, J1
.Elastic, JO
120° Trend of the major '" principal stress . "'- 120• I 300•
Figure 8-12. Probability of stress failure at the tunnel periphery based on failure
criteria presented by Read et al. (1998) (a) and Hoek-Brown (Hoek & Brown, 1980)
failure criteria (b). Level+ 1350, aH = 75 MPa.
a)
b)
77
Probability of stress failure at tunnel periphery with tunnel excavation direction - 1350 level, sH=75 MPa, 15 cm, Read et al.
Trend of the major principal stress '--._ 120" I 300" 300"
255"
Probability of stress failure at tunnel periphery with tunnel excavation direction -1350 level, sH=75 MPa, 15 cm, Hoek-Brown
Trend of the major principal stress ~ 120" I 300" 300"
255°
• Yielded, J2 ... J4
• oamaged, J1
• Elastic, JO
120" Trend of the major ~ principal stress . "- 120" I 300"
• Yielded, J2 ... J4
• Damaged, J1
• Elastic, JO
120" Trend of the major '9:.. principal stress ......... 120" 1300.
Figure 8-13. Probability of stress failure at 15 cm depth from tunnel periphery based on
failure criteria presented by Read et al. (1998) (a) and Hoek-Brown (Hoek & Brown,
1980) failure criteria (b). Level+ 1350, as= 75 MPa.
a)
b)
78
Probability of stress failure at tunnel periphery with tunnel excavation direction -11251evel, sH=65 MPa, 1 cm, Read eta/.
Trend of the major principal stress .......... 120• I 300" 300"
255°
Probability of stress failure at tunnel periphery with tunnel excavation direction - 1125 level, sH=65 MPa, 1 cm, Hoek-Brown
Trend of the major principal stress ............. 120" I 300" 300"
255"
• Yielded, J2 ... J4
• oamaged, J1
•Elastic, JO
120" Trend of the major ~ principal stress . .......... 120" 1300"
• Yielded, J2 .. . J4
• Damaged, J1
• Elastic, JO
120• Trend of the major ....._ principal stress . .......... 120" 1300"
Figure 8-14. Probability of stress failure at the tunnel periphery based on the failure
criterias presented by Read et al. (1998) (a) and Hoek-Brown (Hoek & Brown, 1980)
failure criteria (b). Level+ 1125, as = 65 MPa.
a)
b)
79
Probability of stress failure at tunnel periphery with tunnel excavation direction - 1125 level, sH=65 MP a, 15 cm, Read et al.
Trend of the major principal stress -......-... 120" I 300" 300"
255"
Probability of stress failure at tunnel periphery with tunnel excavation direction - 11251evel, sH=65 MPa, 15 cm, Hoek-Brown
Trend of the major principal stress -......-... 12o· 1 3oo· 300•
255"
. Yielded, J2 ... J4
• oamaged, J1
• Elastic, JO
120" Trend of the major .._ principal stress
.......... 120" 13oo·
. Yielded, J2 ... J4
• Damaged, J1
• Elastic, JO
120" Trend of the major ..._ principal stress
.......... 120. 13oo·
Figure 8-15. Probability of stress failure at 15 cm depth from tunnel periphery based on
the failure criterias presented by Read et al. (1998) (a) and Hoek-Brown (Hoek & Brown, 1980) failure criteria (b). Level + 1125, aH = 65 MP a.
80
Effect of strength anisotropy
A simple analysis of strength anisotropy was made based on the elastic simulation
results. The shear capacity on the average foliation plane (dip 60°, dd 185°) was
calculated for exaction directions of30°, 300°, 330°, and, 360° (Figure 8-16). The
results showed that the shear failure potential on the foliation plane is highest when the
tunnel is excavated along the trend of foliation and lowest when the excavation direction
is 60° from the trend of foliation. Thus, it can be assumed that the effect of anisotropy
on the probability of stress failure is similar, bringing the calculated and observed probabilities closer to each other (Figure 8-17).
81
Excavation direction 300° Excavation direction 330°
Excavation direction 0 o Excavation direction 30°
Figure 8-16. Shear failure potential on the foliation plane (dip 60, dd 185) with
different excavation directions, red = highest potential and blue = lowest potential.
Trend of major horizontal stress is 120° I 300° (see also Figure 8-17).
82
Probability of stress failure id tunnel periphery with tunnel excavation direction - 1350 level. sH=75 MPa, 1 cm, Read et al.
Trend of the major principal stress ---........ 120" I 300" 300"
Effect of anisotropy
255"
o Yielded, J2 ... J4
0 Damaged, J1
• Elastic, JO
75"
Effect of anisotropy
120" Trend of the major .._ principal stress . "'- 120" I 300"
Figure 8-17. The conceptual effect of stress I strength anisotropy on different stress
failure probabilities of Pyhasalmi volcanite. The yellow arrows and lines show that the
anisotropy increases the yield probability.
8.3 Shafts
The defmitions used for secondary and transient state of stress are given in the previous
Chapter 8.2. Unlike thetunnel excavation, every point on the shaft wall goes through the complete transient stress path because of raise boring which is a continuous excavation
method.
Secondary stress state at shaft wall
The highest compressive stresses take place on the sides perpendicular to the trend of
the major horizontal stress (Figures 8-18 and 8-19). The maximum values are 160 MPa
to 190 MPa being 2.5 times the maximum in situ stress. The minor principal stress is
zero with maximum compression, but, on the side of the major horizontal stress trend, a
relatively high tension of 8 MP a to 1 0 MP a exists.
83
Transient stress state at fictitious tunnel surface
The maximum compressive stress increases almost monotonically at the side
perpendicular to the maximum horizontal stress, but on the side parallel to the
maximum horizontal stress the maximum transient stress is 115 MPa to 135 MPa being
1.8 times the maximum horizontal stress (Figures 8-18 and 8-19). The stress failure
potential is high at the maximum transient stress because the minor principal stress is
close to zero.
Stress state at 15 cm depth from shaft wall
At a depth of 15 cm from shaft wall, the maximum compressive secondary stresses are 20% and the transient 45% lower (Figures 8-20 and 8-21). The minor principal stress is 5 MPa to 15 MPa at maximum compression and no tensile stresses exist.
Major principal stress ( MPa )
180
160
140
120
100
80
60
40
20
-30 -20 -10
84
Pyhiisalmi Volcanite - principal stress o1/o3 path 1 cm beside shaft -angle from in situ stress o1 orientation:
go•. 45• and o· -in situ: o1 = 75 MPa, oH/oh= 1.7, oH/oh= 1.8 - oci and ocd stress envelopes with different
cumulative probability
0 10 20 30 40
Minor principal stress ( MPa )
50
----go·
----6--45°
-o--o· --C.p. =2.5%
--C.p. =20%
--C.p. =40%
C.p. =60%
--C.p.=80%
--C.p. =g7.5%
--- s3/s1=0.05
Figure 8-18. Principal stress path at the shaft wall surface with failure criteria
presented by Read et al. (1998), Level+ 1350, an= 75 MPa.
Major principal stress ( MPa )
200~------------~---+--~~~--~----------------------- .----------~
180
160
140
120
100
80
60
40
20
-30 -20 -10
Pyhiisalmi Volcanite -principal stress o1/o3 path 1 cm beside shaft - angle from in situ stress o1 orientation:
go•, 45• and o• - in situ: o1 = 65 MPa, oH/oh = 1.6, oH/oh = 2 .0
- oci and ocd stress envelopes with different cumulative probability
10 20 30 40
Minor principal stress ( MPa )
50
_._go·
----6--45°
-o--o• --C.p. =2.5%
--C.p. =20%
--C.p. =40%
C.p. =60%
--C.p.=80%
--C.p. =g7.5%
--- s3/s1=0.05
Figure 8-19. Principal stress path at the shaft wall surface with failure criteria
presented by Read et al. (1998), Level+ 1125, G'H = 65 MPa.
85
Major principal stress ( MPa )
200 r-------------~---T--·--~~~~-----------------------,
180
160
140
120
100
80
60
Pyhasalmi Volcanite 40 -principal stress cr1/cr3 path 15 cm beside shaft
- angle from in situ stress cr1 orientation: go•. 45• and o•
20 -in situ: cr1 = 75 MPa, crH/crh = 1.7, crH/crh = 1.8 - crei and crcd stress envelopes with different
cumulative probability
0 ~~--~--~~~~--~------~----~------------------~ -30 -20 -10 0 10 20 30 40 50
Minor principal stress ( MPa )
--+--go•
--l:r--45°
--o--o• --C.p. =2.5%
--C.p. =20%
--C.p. =40%
C.p. =60%
--C.p. =BO%
--C.p. =g7.5%
--- s3/s1=0.05
Figure 8-20. Principal stress path at 15 cm from the shaft wall surface with failure
criteria presented by Read et al. (1998), Level+ 1350, an= 75 MPa.
Major principal stress ( MPa )
200 r-------------~--~----~~~~-----------------------, Pyhasalmi Volcanite 1 - principal stress cr1/cr3 path 15 cm beside shaft I
180 -angle from in situ stress cr1 orientation: J go•. 45• and o· ,
- in situ: cr1 = 65 MPa, crH/crh = 1.6, crH/crh = 2.0 - crci and crcd stress envelopes with different
160 cumulative probability
140
120
100
80
60
40
20
0~-L--~--~~--~--~------~----~------------------~ -30 -20 -10 0 10 20 30 40 50
Minor principal stress ( MPa )
--+--go•
--l:r--45°
--o--o· --C.p. =2.5%
--C.p. =20%
--C.p. =40%
C.p. =60%
--C.p. =BO%
--C.p. =g7.5%
--- s3/s1=0.05
Figure 8-21. Principal stress path at 15 cm from the shaft wall surface with failure
criteria presented by Read et al. (1998), Level + 1125, an = 65 MP a.
86
Failure criterion
A comparison of the stress paths with the Read et al. (1998), Hoek-Brown (Hoek &
Brown, 1997 and 1980) and peak strength criteria to the systematic visual stress failure
observations shows that the peak strength overestimates the in situ strength. The other
two criteria present reasonable results (Chapter 6.2). Therefore, the peak strength
analysis is rejected from further studies.
The interpretation of elastic stresses and failure criteria was based on the same
assumptions as in the tunnel case (Chapter 8.2). The comparison of results of failure
probability, presented in Figures 8-22 and 8-23, with visual observations (Chapter 6.2)
leads to almost the same conclusions as in the tunnel case. The failure criterion
presented by Read et al. (1998) overestimates the strength, but the Hoek-Brown method
presents quite good estimates. At a depth of 15 cm from the tunnel surface, the Hoek
Brown results are more reasonable (Figures 8-24 and 8-25). On the other hand, the
stress redistribution after stress failure is not taken into account, which would increase
the stress failure potential at the 15 cm depth.
Based on the elastic stresses and homogenous material, the in situ failure criterion for
the tunnel seems to be close to the Hoek-Brown (Hoek & Brown, 1997) strength
envelope, but the criterionpresented by Read et al. (1998) also give reasonable
estimates. It must be remembered that the degree of anisotropy of Pyhasalmi volcanite is
unknown nor was it taken account in simulations.
a.)
b.)
87
Probability of stress failure at shaft periphery with angular orientation - 1 cm from surface, 1350 level, sH=75 MPa, Read et al.
Trend of the major principal stress '-..a. 120" I 300" 300°
255°
Probability of stress failure at shaft periphery with angular orientation - 1 cm from surface, 1350 level, sH=75 MPa, Hoek-Brown
Trend of the major principal stress '-..a. 120· 1300° 300o
255°
. Yielded, J2 ... J4
• oamaged, J1
• Elastic, JO
120° Trend of the major 'Wl principal stress
........... 120.1300.
• Yielded, J2 ... J4
• Damaged, J1
•Elastic, JO
75°
120° Trend of the major ._ principal stress
........... 120" 1300°
Figure 8-22. Probability of stress failure at the shaft wall based on failure criteria
presented by Read et al. (1998) (a) and Hoek-Brown (Hoek & Brown, 1980) failure
criteria (b). Level+ 1350, aH = 75 MPa.
a.)
b.)
88
Probability of stress failure at shaft periphery with angular orientation - 1 cm from surface, 1125 level, sH=65 MPa, Read et al.
Trend of the major principal stress -.......... 120" I 300" 300°
255°
Probability of stress failure at shaft periphery with angular orientation - 1 cm from surface, 1125 level, sH=65 MPa, Hoek-Brown
Trend of the major principal stress -.......... 12o· 1 3oo· 300o
255°
• Yielded, J2 ... J4
• oamaged, J1
•Elastic, JO
120° Trend of the major .._ principal stress . " 120" 1300"
• Yielded, J2 .. . J4
• oamaged, J1
• Elastic, JO
75"
120° Trend of the major .._ principal stress . " 120" 1300"
Figure 8-23. Probability of stress failure at the shaft wall based on failure criteria
presented by Read et al. (1998) (a) and Hoek-Brown (Hoek & Brown, 1980) failure
criteria (b). Level + 1125, aH = 65 MP a.
a)
b)
89
Probability of stress failure at shaft periphery with angular orientation - 15 cm from surface, 1350 level, sH=75 MPa, Read et al.
Trend of the major principal stress --........ 12o· 1 300" 300o
255°
Probability of stress failure at shaft periphery with angular orientation - 15 cm from surface, 1350 level, sH=75 MPa, Hoek-Brown
Trend of the major principal stress --........ 12o· 1 300" 300o
255°
• Yielded, J2 .. . J4
• oamaged, J1
• Elastic, JO
120° Trend of the major ~ principal stress ........ 120" 1300"
• Yielded, J2 ... J4
• Damaged, J1
•Elastic, JO
75°
120° Trend of the major ..._ principal stress ........ 120" 1300"
Figure 8-24. Probability of stress failure at 15 cm from the shaft wall based on failure
criteria presented by Read et al. (1998) (a) and Hoek-Brown (Hoek & Brown, 1980)
failure criteria (b). Level+ 1350, an= 75 MPa.
a)
b)
90
Probability of stress failure at shaft periphery with angular orientation - 15 cm from surface, 1125 level, sH=65 MPa, Read et al.
Trend of the major principal stress ............ 120" I 300" 300°
255°
Probability of stress failure at shaft periphery with angular orientation -15 cm from surface, 11251evel, sH=65 MPa, Hoek-Brown
Trend of the major principal stress ............ 120· 1 3oo· 300o
255°
• Yielded, J2 ... J4
• Damaged, J1
• Elastic, JO
75°
120° Trend of the major ~ principal stress . ......... 120" 1300.
• Yielded, J2 ... J4
• oamaged, J1
• Elastic, JO
75°
120° Trend of the major ~ principal stress .......... 120" 1300"
Figure 8-25. Probability of stress failure at 15 cm from the shaft wall based on failure
criteria presented by Read et al. (1998) (a) and Hoek-Brown (Hoek & Brown, 1980)
failure criteria (b). Level+ 1125, aH = 65 MP a.
91
Effect ofanisotropy
Based on the elastic simulation results, a simple analysis of the strength anisotropy was
made by calculating shear capacity on the average foliation planes dipping 60°±20° with
a dip direction of 185°±20° (Figure 8-26 and 8-27). The results showed increasing stress
failure potential with as the dip became steeper, which can be understood by decreasing
normal stress when the normal of the plane of anisotropy converges with the normal of
the shaft surface. Another finding is that the highest shear failure potential on the
anisotropy plane is always on the side of the dip direction or opposite to it. In the case of
Pyhasalmi, this means that on the north or west side of the ore body the stress failure
locations on the shaft walls are rotated counter-clockwise, that is, the orientation of the
major in situ horizontal stress is not perpendicular to the failure apex orientation but to some extent is clockwise from it. This can explain the difference between the stress failure observations in orientations 10°-30° I 190°-210° and the measured major in situ
horizontal stress orientation of 130°/310°.
92
Anisotropy plane: 40° I 185°
Anisotropy plane: 60° I 185°
Anisotropy plane: 80° I 185°
Figure 8-26. Shear failure potential on the foliation plane with three different dip
values.
93
Anisotropy plane: 60°1 165°
Anisotropy plane: 60° I 185°
Anisotropy plane: 60° I 205°
Figure 8-27. Shear failure potential on the foliation plane with three different dip
direction values.
95
9 DISCUSSION AND CONCLUSIONS
Aim o[this study
The objective of this work was to collect stress failure information from known
geological and rock mechanical conditions and analyse this data in order to establish a
mapping system, understand factors affecting stress failures, and develop stress failure
prediction methods. If the method is found to be usable, it could be utilised for further
investigations at the Olkiluoto site, which is chosen as a potential site for the Posiva Oy
nuclear waste repository. The Pyhasalmi volcanites, where most of the mappings were
obtained, is found to be suitable for this purpose being schistose, metamorphic, sparsely
jointed, and heterogeneous like the Olkiluoto mica gneiss. Also the stress- strength ratio
of the rockmass is equal to that of Olkiluoto.
Data collection and management
The stress failure mapping system, that was developed, was found to be usable, but the
complex geology and massive tunnel driving limited the mapping of all the details. In
particular, the location and volume of visual stress failures were difficult to measure
from the tunnel surface, which is excavated by the drill and blast method. The detailed
measuring of noise type of stress failures was especially problematic. Furthermore, the
stress failure classification has too many classes or they are hard to define. However,
despite the tight excavation and support timetable, a total of 1700 m of tunnel out of the
total excavation length of 2600 m was mapped. The effect of the excavation on the
failures is excluded from the analyses.
During the analyses of the stress failure data, it was found that the tunnel driving
direction and any changes in tunnel profile should be recorded during the mapping time,
because the order of excavation steps in the cross-section areas is difficult to define
afterwards.
A continuous centerline for each tunnel should be defined. This would help to combine the data from different logging periods in the final analyses. This is especially important
in the case of extensive and complex tunnel networks, as in the Pyhasalmi mine.
A geological survey program, "Surpac2000", with connection of the database was used
for management, synchronising and, visualisation of the data. The mapped tunnels were
treated as boreholes and the data was recorded either as "from-to" or "point" -values.
96
Geology
The majority, 72%, of the mapping data is from tunnels in volcanites below the
+ 1200 level. The Pyhasalmi volcanite consists of felsic and mafic volcanite, but the
implementations are various, pegmatite veins and lenses being common. Most typical is
slightly or moderately foliated volcanite where mafic and felsic volcanites are densely
alternating. The average grain size is below one millimeter and no alteration exists. The
volcanites are sparsely jointed, the average RQD value is 87, and only one joint set
along the foliation exists ( Jn=2 ).
Analyses of stress failure data
Although, the mapping data includes a total of 1700 m of tunnel that turned out to be too little for statistical analyses. This is due the anisotropic and heterogenic geological
nature of volcanites in relation to various tunnel excavation directions and orientation of
the in situ state of stress.
The analyses were focused on finding factors correlating with stress failure
observations. Among the studied factors were the rock type, in situ state of stress,
direction of excavation, size and shape of excavation, orientation of foliation, geological
stuctures, rock fabric, and rock mass quality. Because many factors are dependent on
others, only those having an effect on the others were found.
In most cases, the location or length of the stress failure on the tunnel cross-section
could not be defined. Therefore, the length of stress failure was assumed to be equal to the mapping length and the location was ignored. The majority of stress failure
observations were in the classes of none ( JO ), noise only ( J1 ) or light spalling ( J3 ).
The following general trends were observed:
Rock type Occurrence Stress failure type
volcanite 33% none,JO
67% noise or minor spalling, Jl ... J4
33% minor spalling J3 ... J4
pegmatite 25% none,JO
75% noise or minor spalling, Jl ... J4
25% minor spalling J3 ... J4
ore 45% none, JO
55% noise or minor spalling, Jl ... J4
5% minor spalling J3 ... J4
volcanite contact 40% minor spalling J3 ... J4
97
Time between mapping Occurrence Stress failure type in volcanite and excavation
< 1 day 20% none,JO
80% noise or minor spa/ling, J1 ... J4
40% minor spa/ling J3 ... J4
2 to 3 days 40% none, JO
60% noise or minor spa/ling, J1 ... J4
30% minor spa/ling J3 ... J4
> 3 days 55% none,JO
45% noise or minor spa/ling, J1 ... J4
20% minor spa/ling J3 ... J4
Depth level Occurrence Stress failure type in volcanite
< 1200 50% none,JO
50% noise or minor spa/ling, J1 ... J4
20% minor spa/ling J3 ... J4
> 1300 30% none,JO
70% noise or minor spa/ling, J1 ... J4
55% minor spa/ling J3 ... J4
Tunnel shape Occurrence Stress failure type in volcanite
Straight or curved tunnel 30% none, JO
70% noise or minor spa/ling, J1 ... J4
35% minor spa/ling J3 ... J4
Crossing or wider profile 10% none,JO
90% noise or minor spa/ling, J1 ... J4
50% minor spa/ling J3 ... J4
The effects of the orientation of major horizontal stress and orientation of foliation
compared to the tunnel excavation direction could not be studied independently. In Pyhasalmi, the trend of the major horizontal stress is 120°/310°. The average dip of the
foliation plane is 64 o and its direction 185°. Based on that, the following conclusions
could be drawn:
98
Excavation Time between excavation Occurrence of minor spalling direction and mapping
_1 to aH < 1 day 100%
1 to 3 days 60%
11 to aH < 1 day 60%
1 to 3 days 45%
in term. < 1 day 30%
1 to 3 days 20%
Any geological structure such as a vein or fracture, which differs from the normal foliation of volcanite, increases the stress failure probability by 20%. Among the other factors, the dip of the foliation and rock mass quality did not have an unambiguous
correlation with the stress failures. The fabric does not correlate with the failures. Grain
size and weathering could not be studied because no variation on these parameters was
found.
During the first mapping period, there was no rock mechanical mapping in the raise
bored shafts, but some visual observations were made from above and below the
tunnels. In all cases, the stress failures were systematic and the predicted failure
directions were 10°-30° and 190°-210°. In the 3.1 m diameter shaft, the width of the
failure area was about one metre and the depth was several dozen centimetres.
The magnitude and orientation of the major in situ principal stress and the anisotropy
proved to be the most important areal factors affecting the stress failure probability. The
most unfavourable excavation directions are perpendicular to the major principal stress
and along the trend of anisotropy.
Strength and deformation parameter values of volcanites
In the tunnel scale, the layering of mafic and felsic volcanites is dense. The laboratory test data for both types of volcanites was combined, and the average deformation parameters and statistical strength envelopes were defined and used in the analyses. No reliable test results of anisotropy of the mafic/felsic volcanite composite were available
for this study.
Failure criteria
The possibility of estimating stress failures based on the elastic stresses was studied for
both the tunnel and shaft using three-dimensional boundary element simulations. The
calculated elastic stresses were compared with three different statistical failure
99
envelopes, which were 1) the crack damage strength envelope by Read et al. (1998), 2) the Hoek-Brown rock mass strength envelope (Hoek & Brown, 1997 and 1980), and,
3) the Hoek-Brown intact rock peak strength envelope. The results showed that the peak
strength envelope overestimates the in situ strength, but the other two methods result in
reasonable failure probabilities. The stress failure observation, where the excavation
along the trend of anisotropy is unfavourable, was supported by a fast and simple study
of the effect of strength anisotropy.
The more detailed qualification of failure criteria was not possible, because there were
no measured values of the extent of the failure available. Moreover, the degree of
deformation and strength anisotropy of Pyhasalmi volcanite was unknown. Based on the
tunnel observations, the anisotropy is formed by dense layering of mafic and felsic
volcanites, not from internal anisotropy of one of these rock types.
The simple shaft stability study with strength anisotropy showed that the trend of the
major horizontal stress could be clockwise from the orientation assumed by the
orientation of the stress failure observations. This result is supported by the stress
measurement results.
101
10 RECOMMENDATIONS
The lack of data was often found to be problematic in the statistical analyses. The
mapping has continued and the additional data should improve the reliability of the
analyses presented and the results achieved.
A more detailed qualification of the failure criteria involves more accurate data on the
quality and extent of the stress failure. Quantifying the extent of the stress from the
surfaces of the tunnels, which were excavated by the drill and blast method, was found to be difficult. The second mapping phase will also include data from raise-bored shafts
where the extent of the stress failure can easily be seen and measured from pre-known
diameters. Furthermore, the geological and orientational conditions are more stable
around the shafts.
The stress failure noise is originated from micro seismic events and can be localised and
quantified using a 3-D micro-seismic monitoring system. On the other hand, the
expected accuracy of the micro-seismic monitoring involves a limited monitoring area
on a scale of a few dozen metres.
The anisotropy of the rock strength seems to be an important parameter. Further study
should involve laboratory analysis of the strength and deformation anisotropy induced
by the dense layering of mafic and felsic volcanites.
Minor improvements should be made before proceeding further with the stress failure mapping. The most important thing for the analyses is the need for flags for profile
changes, cross section areas, tunnel face, geological structure, and direction of excavation. All of these can be defined from the current mapping data and survey data,
but it is very laborious.
103
REFERENCES
Barton, N, Lien, R. & Lunde, J., 1974. Engineering Classification of Rock Masses for
the Design of Tunnel Support. Rock Mechanics 6(4), 189-239.
Grimstad, E. & Barton, N., 1993. Updating ofQ-system for NMT. Proceedings of
Sprayed Concrete, 18-21 October 1993, Fagemes. Norwegian Concrete Association,
Norway.
Hakala, M., Somervuori, P. & Syrjanen, P., 1998. Rock mechanical study of the new
mine area at depth levels 1050 ... 1400. Report n:o GE0-15-001. Outokumpu Mining
Oy Pyhasalmi mine (in Finnish).
Hakala, M., Somervuori, P. & Syrjanen, P., 1999. Rock mechanical study of the new
mine area at depth levels 1050 ... 1400 - Part 11. Outokumpu Mining Oy Pyhasalmi
mine (in Finnish).
Helovuori, 0., 1979. Geology of the Pyhasalmi Ore Deposit, Finland. Economic
Geology 74, 1084-1101.
Hoek, E. & Brown, E.T., 1997. Practical estimates of rock mass strength, Int. Journal
Rock Mechanics & Mining Science & Geomechanics Abstracts. 34(8), 1165-1186.
Hoek, E. & Brown, E. T., 1980. Underground Excavations in Rock. The Institution of
Mining and Metallurgy, London.
Hoek E., Kaiser P.K. & Bawden W.F., 1995. Support of Underground Excavations in
Hard Rock. A.A. Balkema, Brookfield.
Hutchinson, D. & Diederichs, M., 1996. Cable bolting in underground mines.
Richmond, Canada, BiTech Publisher.
Korhonen, K-H., Gardemeister, R., Jaaskelainen, H., Niini, H., Vahasarja, P., 1974. Engineering geological bedrock classification. Tiedonanto 78. VTT Geotechnical
loaboratory, Espoo, Finland (in Finnish).
Kousa, J., Marttila, E, & Vaasjoki, M. 1994. Petrology, geochemistry and dating of
Paleoproterozoic metavolcanic rocks in the Pyhajarvi area, central Finland.
Geological Survey of Finland, Spec. Paper 19, 7-27.
Kousa, J., 1990. Paleoproterozoic metavolcanic rocks in the boarder zone of Savo and
Pohjanmaa, Central Finland. In Kahkonen, Y. (ed.) IGCP Project 217 Proterozoic
Geochemistry. National Working Group in Finland. Symposium, Proterozoic
Geochemistry, Helsinki '90. Abstracts, 29-30.
Lahtinen, R., 1994. Crustal evolution of the Svecofennian and Karelian domains during
2.1-1.79 Ga, with special emphasis on the geochemistry and origin of 1.93-1.91 Ga
gneissic tonalites and associated supracrustal rocks in the Rautalampi area, central
Finland. Geological Survey of Finland, Bulletin 378, 128p.
104
Ledger, A., 1999. Stress measurements in the 1125 m level. Outokumpu Mining Oy, Pyhasalmi Mine. Work report, Rock Mechanics Technology.
Martin, C.D., Kaiser, P.K. & Maybee G.M., 1998. A Failure Criteria for the Stability
of Drifts and Pillars. CIM I AGM 1998, Montreal.
Matinlassi, M., 2001. Geologinen ja kalliomekaaninen kartoitus Pyhasalmen uudessa
maanalaisessa kaivoksessa. Work report xx-01, December 2001. Posiva Oy, Helsinki.
(in press, in Finnish)
Mononen, S., 2000. CSIRO HI rock stress measurements at Pyhasalmi mine
Unpublished work report, Laboratory of Rock Engineering, Helsinki University of Technology (in Finnish).
Read, R.S., Chandler, N.A & Dzipk, E.J., 1998. In Situ Strength Criteria for Tunnel Design in Highly-stressed Rock Masses. Int. J. Rock Mechanics & Mining Science,
Vol. 35, No 3 pp. 261-278.
RocScience, 1999. Examine3D Version.
Somervuori, P. & Middleton, N., 2001. Modem Computer Applications in Mining Using Surpac Vision. In Proceedings of APCOM 2001, The 4th Regional Symposium
on Computer Applications in the Minerals Industries (Eds. P. Eloranta & P. Sarkka). September 2001, Tampere, Finland.
Surpac Sowtware International, 2001. Surpac Vision, Online: www.surpac.com.
Weihed, P. & Maki, T. (eds.), 1997. Volcanic hosted massive sulphide and gold
deposits in the Skellefte district, Sweden and western Finland. Geologian
tutkimuskeskus, Opas - Geological Survey of Finland, Guide.
Aikas, K. (Ed), Hagros, A., Johansson, E., Malmlund, H., Sievanen, U., Tolppanen, P., Ahokas, H., Heikkinen, E., Jaaskelainen, P., Ruotsalainen, P. &
Saksa, P., 2000. Engineering rock mass classification of the Olkiluoto investigation site. Report Posiva 2000-8. Posiva Oy, Helsinki.
