Professional Documents
Culture Documents
Methods of Geological Engineering in Discontinous Rocks PDF
Methods of Geological Engineering in Discontinous Rocks PDF
Geoogica
• •
9 neer.ng
in Discontin ous Rocks
RICHARD E. GOOD N
Professor of Geological Engineering
University of California, Berkeley
Goodman, Richard E.
Methods of geological engineering in discontinuous rocks.
Includes index.
1. Rock mechanics. I. Title.
TA706.G66 624'.1513 75-42152
ISBN 0-8299-0066-7
v
preface
engineer can not handle the geological data correctly without the
respect for its !! fuzziness II that comes from a personal acquain tance
with geology.
I visited the sites of the Malpasset and Vajont failures in the early
such sites were inadequate and that we were, perhaps, relying too
vi Preface
behavior; but we must also search for objective tools. Since 1960,
set them forth for students, and for practicing technicians who can
leave from Berkeley. I wish to thank the John Simon Guggenheim Founda-
tion and the Regents of the University of California for this opportu-
John Bray, John Boyd, and other faculty members. And I profited from
ing Peter Cundall, Christopher St. John, Nick Barton, Tidu Maini, John
Ross Hammett, Peter Kelsall, Don Moy, Tim Harper, and Graeme Major. I
Wittke, Klaus John, Pierre Lande, and "Skip" Hendron with respect to
limit equilibrium analyses - - Dr. Leopold Muller and Karel Drozd with
Preface
Tay l o r , Edward Wil son , Hugh Trollope, and Ann Bornstein with respect
to comp u ter me t h o ds. I have also benefitted from the wo rk of past and
Ashra f Mah tab, Rudolfo de la Cruz, John Cadman, Jacques Dubois, Alain
Ergu vanl i, J ean Lu c Dessenne, and Karel Drozd. The critical comments
quite h elpful .
Tor Brek k e , W. Chin n , Lloyd Cluff, Phillip Cole, James Coul son, Walter
Thomas Lan g , P ierre Lon d e, Ken Ma tthews, J. My ung, Carlos Ospina, Marc
Ger a rdo Tarna, Jos e Tejada , Ruth Ter z aghi, Cl. Tourenq, Lloyd Underwoo d ,
a n d J oel Ver di e r.
Mechanics Equipment Co. (Glen Ellyn, Ill.), Sprague and Henwood, Inc.
Finally, I wish to thank the persons who helped me with the work
to the memory of
Parker D. Trask
j
I
xi
contents
PREFACE
1. II"TRODUCTI ON
REFERENCES 419
Figure 1-2. A concrete arch dam. Mossyrock Dam, Cowlitz River, Washington: a
doubly curved, thin arch dam 365 feet above riverbed, 606 feet above the basalt
bedrock; (courtesy of Tacoma City Light).
Introduction 5
a natural flow barrier under the line of action of the dam as fissures
within the rock mass closed in response to applied load. The Vajont
failure (Muller, 1964 and 1968) occurred when a massive landslide
moved on bedding surfaces into a relatively small reservoir, causing
overtopping and flooding. The landslide was triggered by uplift
forces associated with reservoir filling.
The large influence of discontinuities on construction operations
in rock is well illustrated by the Kukuan arch dam, designed by
Coyne and Bellier for Taiwan Power Company. This dam, 86 meters high,
was constructed in a valley cut 500 meters deep into alternating
layers of slate and quartzite. Thin clay seams containing graphite
compromise the stability of unfavorably oriented layers of the site.
The right bank (figure 1-4) is a 60 - 70 degree dip slope. To
found the dam in solid rock, it was necessary to excavate through 20
to 40 meters of loosened slabs, but conventional excavation was
undesirable because of the slide potential. Grouting and "dental
work" (localized replacement of weak rock with concrete) were un-
successful. A solution was obtained in which tunnels up to 10.7
Introduction 7
meters wide were driven well into the abutments and backfilled with
concrete. Since the tunnels cut across the bedding, they were stable.
After driving a tunnel to the full depth and width, it was concreted
to within several meters of the crown. Then, after two to three
weeks, a stone protection was laid on the concrete fill and a t un nel
was excavated above. The process was repeated until eight tunnels
had been constructed, producing a stable concrete structural abutment.
Activities in advancing the construction and utilization of
tunnels and underground chambers have also created interest in methods
of geological engineering. Investigations of tunnel sites remain
fairly primitive because the sites are long, and remain inaccessible
until construction. Some attention has been focused o n assess ing the
excavatability of the rock from tests on samples, but geological and
geophysical prediction techniques, and analytical methods to forecast
formation conditions are not yet generally available.
Techniques for investigating and anaJyzing rock behavior for
underground works such as subsurface power plant chambers (figure 1 - 5) ,
(b)
(a)
Figure '·5. (a) Oroville Dam project. The dam has sh ells of gravel wh ile th e core is
derived from a vast alluvial fan; (courtesy Calif. Dept. of Water Resources).
(b) Oroville underground power station machine hall during construction. The man
standing in the lower left gives the scale; (courtesy Calif. Dept. of Water Resources).
8 Introduction
Figure '·6. Spillway excavation on left abutment of Chivor Dam, Colombia. Notice
the truck and shovel for scale. The smooth surface of discontinuity in the middle left
was exposed during construction and caused a design change. The benches are 5 meters
wide and spaced every 10 meters; (courtesy of Ingetec Ltda., Bogota).
First . the geo logy of the site must be defined; this entails mapping
of f i e l d exposures J study of aerial photographs, and specific explora-
t ion with excavations o r drill holes . Then . the properties of the
rocks must be assessed. Here there can be different choices of
TABLE 1-1
Some Sources of Information
En g ineer i n g Ge o lo g y (Elsevier).
Geotechnique .
Proceedings of Congresses and Symposia of the International Society for Rock Mechanics*
Textbooks
(a)
(b)
(c)
Figure 2· 1. (a) High ly inter locked, crysta ll ine texture of a Mesozoic quartzite; (courtesy f Prof R. WenkL
25.5X . (b) Highly anisotropic crystal l ine texture with oriented fissures (fracture 1.ledvage) III chlorite
schist- Ho mestake M ine, S.D.; (courtesy of Dr W Chinn). 40X (c) Porous, clastic texture ' eolian sand,
stone from Olduvai gorge, Tanzan ia, consisting of poor ly sorted rock fragments and grains, some coated
with clay. Cavities occur in altered nepheline grains (N); (courtesy of Prof. R. Hay). 136K
18 Rock Classification
** Table 2-1 is based upon one devised by Karl Terzaghi and distributed
to students in his course on engineering geology at Harvard University
in the 1950's. A copy revised shortly before his death was generously
supplied by Dr. Ruth Terzaghi. A somewhat similar approach is used by
the National Institute for Road Research, South Africa, as published
by Weinert (1964); see Fookes, Dearman, and Franklin (1971).
TABLE 2-1
EF FECTS OF SATURATION ON ROCKS AND ROCK-LIKE MATERIALS
Terzaghi's Guides for Distingu ishing Rock, Weathered Rock, and Soil *
unchanged a) so l i d rock
* From Professo r Kar l Terzagh i' s 00urs e note s fo r Engine e ring Geology
a t Harv a rd Uni versi ty; i n cl ude d wi t h k ind p ermission o f Dr . Ru th
Terzagh i (w i th minor e di t ori al c h ang e s) a nd in cludi ng rev isions made
b y Kar l Terzagh i sho rtly b e f ore his death.
22 Rock Classification
ing". Materials that are not solid with a ringing sound when struck
by a hammer when first encountered should not be referred to as rock
at all, according to this scheme. Many sedimentary rocks would
accordingly be termed "soil-like rocks II in maps and reports, and the
resulting impression would be correct for the engineer.
Geological investigations must correctly diagnose a specific
soil-like condition as either inherent or localized. Weathering, and
hydrothermal alteration -- the first usually intensifying towards the
surface and the other, with depth or laterally -- may produce spotty
and variable degrees of localized softness. In contrast, some
sedimentary rocks are inherently soft either through incomplete
cementation, intense fissuring, or regional alterations; neither
"dental work" nor outright "extraction" can improve the rock condi-
tions in this case.
The distinction between rock and soil is especially important as
regards specifications for excavation contracts. So many legal
controversies have revolved about this point that agencies such as
the U.S. Bureau of Reclamation have been forced to adopt almost
comically detailed wording for contracts, as in Table 2-2. The main
ideas are that the material to be excavated is rock only if it is
both in place, (or of large mass) and solid. If it is too risky to
attempt a classification, the excavation receives one name--unclassified
excavation -- and one price throughout. This can happen in deeply
weathered materials, with their extreme variability and gradational
qualities, in soil-like soft rocks, in bedded rocks alternating in
hardness, and in very dense or cemented soils.
WEATHERING
TABLE 2-2
Classification of Excavation
According to U.S. Bureau of Reclamation
Contract Specifications
figure 2-2a, which resembles rock but has the st ren g t h o f a dense
soil. Vargas (1953), Ruxton and Berry (195 7 ), Lumb ( 19 6 2) , Dee r e
and Patton (1971), and others have described the tran si t ional state s
24 Rock Classification
(a) (b)
Figure 2-2. (a) Decomposed granite. Former joint blocks contain hard " core stones"
in their centers while relict joints are now sandy clay seams or partings in the thor-
oughly weathered rock (saprolyte). (b) The top-of-rock surface in soluble marble;
Columbia, California. The soil was removed by hydraulic monitors to obtain placer gold.
eN q - Nqo ) / (1 - Nqo )
Nq is the weig h t r a tio of quar t z to quartz + feldsp ar in the weathered
speci men , an d Nqo i s t h e c or re sp ondi n g r atio i n th e intact , un-
weathered s peci me n . N is of th e o rder of 1/3 for a fresh granite
qo
a nd increase s t owa r d 1 as t h e weathe ri ng progresses. Thus the index
26 Rock Classification
vo - Vw
K (1 )
V
o
Durability
(a)
(b)
FISSURED ROCKS
(e) (d)
F igure 2-4. (a ) Scannmg electron
microscope photo 01 a series of pores In
Ber ea sandstone. fWeinbrandt and Fatt.
19691. 102 X. fb) Scanning electron
microscope photo o t another pore In
Berea sandstone ; (We inbrandt and FaIt,
1969). 1020X . (c) Polished section of
Mont Blanc granite. show i ng Intense
ftssunng . (courtesy of Drs. M. Panet
and C. T ourenQ, Lab. des Ilonts et
Chausees. PariS). l A X. (d) M,crophoto·
graph of a fiss ure fabriC reproduced In
a plastic fil m replica of a polished
diorite surface; (COurtesy of Dr. J.
Verd ier. Coy ne and Beli ier , Par ies). SOX.
fe) Fine fissures In granite, hi gh lighted
by dye and viewed in thin sect ion;
Huelgoal gr an ite . Bri ttany; (courtesy of
(e) Drs. B. Schnei der and J. Verd ier, Coyne
and Bel1ier, PariS) .
32 Rock Classification
can b e gene r a ted b y chemi cal weac herin g , unlo ading, heati ng and
c oo li ng , and most imp ortant l y b y lo c ali z ed cr ack ing accomp a ny ing
defo rm ation. Roc ks likely to be fo un d in a fissured state are:
v o l cani c fl ow ro c k s; fo liate d met amorph i cs, espe ci ally schistose
v a riet i e s; marble; pe gmatites a nd porp h y ri t ic or hypidiomorphic
gran it ic ro cks ; g rani te s exhibitin g cleavage; serpentine; chert and
sili c eous sha les; a nd qu a rtzi te.
The d e gree of fis s ur i ng s h o u l d b e a basic compon e n t of any rock
c l as sificat i o n s cheme. It c an be char acteriz e d through direct
observ at ion , or mo r e simply throu gh index tests.
A pol ishe d s u rface will of ten enable promin e n t fissures to be
observed with t h e n aked eye. A han d lens, binocular microscope, or
be s t of all, an ore micros c op e al lows st udy of fissure distribution
in t h e p oli shed sec t ion . Tourenq (1 9 6 9) di splayed f is sures in
p ol ish ed surf a ces by prep ari ng rep l i cas of the su rf ace adapti n g
tech n i ques f or e lectron mic ro scope spec i me ns des cr ibe d by Bradley
(19 5 4) an d J acq u et and Me ncarel l i ( 1959 ) , (fi g u r e 2 -4d ) . Schneider
(196 7 ) used dyes: Fusc hine ASA (basic), Victoria Bl u e (basic ) and
Auramine J ( aci d) t o s h ow fi ne fis s u re detail in thin sect io ns viewed
in tran s mitted ligh t with a pet ro g rap h ic mi croscope, (figure 2--4e).
S tu dy of fl u i d- fi l le d e pigen eti c i nclusions , e. g. in quartz, can
a llow r elat ive ev a l ua t ion of dif f erent directi o ns of fissuring,
(Ver d ier a n d Deicha, 19 71 ). Thes e pro c es se s c a n be te d ious and for
pract ica l en g ineer i ng work it may be mo r e app ropriate t o characterize
th e de g r ee of f is suring imp lic itly.
Pa r a meters of fi s suri n g c an be d e r i ved from pressure - volume
ch an g e cu r ves , sh ear a n d l ong it udi na l wav e velocity meas u rement s ]
comp a ri s on of d i r ect a nd in dire ct t en s i l e st reng ths, and the rat io of
perme abilitie s in t en s i on and compression. Mo rl ier (1 96 8), follow i n g
wo rk of Wa l sh (1965) , s ug gested c o mput ation of the vo lume of f iss ures
--f i ss ure p oro sity f rom t he sh ap e of t h e p res sure - vo l ume change
cu rve (p versus tV ). As s h ow n in fig ure 2-5 , this curve is concave
upw a rd , b e c omi n g a symptotic to a l i ne wh ose sl o pe is defined by the
compre s s ib i li ty of t h e r ock (k = 3( 1~2 V ») ' The i n itial fis sure
p o rosi ty nf (o ) is est imate d b y the value of ~v at intercept of t h e
a sy mptote. The con cavi ty of t he cu r ve c a n also be interprete d to
34 Rock Classification
original rock
volume = V
V$I, measured
IQ x 100% (2)
V$I, calculated
TAB LE 2-3
quartz 9.6 0 . 08 6 .0 4. 1
ort h oc l ase 6. 7 0.27 5.7 3.3
plagi oc lase 8. 1 0 .28 6.3 3.5
biotite 7.0 0.25 5. 1 3.0
c a l cite 8. 1 0.30 6.7 3.4
mus c ov ite 7.9 0 .25 5.8 3.4
amphibole 12. 9 0 .28 7.2 4.0
pyrox ene 14 . 4 0.24 7.2 4 .2
olivine 20 .0 0.24 8.4 5.2
magneti t e 23. 1 0 ~ 26 7 .4 4.2
~
o
u
~
>
"--.
iii
o
a.>
E
~
>
Figure 2-6. Relative effect
of fissures and pores on the
longitudinal wave velocity
and the modulus of elastic-
ity; (Tourenq, Fourmaintreau,
Emeas./Eeale. and Denis, 1971).
TABLE 2·4
Index to Degree of Fissuring
According to the Ratio of
Transverse to Longitudinal Wave Volocities
Description
0 .6 to 0.7 f is s u red
TABLE 2·5
Tension Test Index to Fissuring
Ro c k Fi ss ure length °t *
( mrn)
°t , B
Limest on e 0. 2 1.0
Limestone 1. 5 0.45
Basalt 0.1 0 .9
Basa l t 2 t o 10 0 . 15
(a) (b)
'-4+-------0olitic
limestone
S=I
(/)
"-
~ 10-
8
150
100 12 10- 10
2~0~~0~~2~0~~~~~6~0~~80~~100
P (bars)
Figure 2-7. Radial permeability test; redrawn from Habib and Bernaix (1966).
(a) Radial permeameter (dimensions in mm). (b) Results for a porous and a fissured
rock; S = K_,1K50
± P
+ (3 )
2(1-v)
Q
K In (4)
2TT LP
TABLE 2-6
Radial Permeability Test Index to Fissuring
Rock Description of S
fissuring
DISCONTINUITIES
Rock masses invariably include numerous surfaces of real or
potential discontinuity. Though somewhat artificial, we will dis-
tinguish between discontinuities and fissures. Obviously, there is
a continuous distribution of discontinuity surfaces according to
length. However, fissures within a specimen are included in a sample
of the specimen, thus subject to meaningful inquiry in the laboratory.
Laboratory techniques for samples of larger surfaces of weakness are
developing but the results are seldom exportable to the field without
additional field observations and tests. Thus fissures can be con-
sidered as rock specimen features whereas discontinuities cannot.
A single discontinuity includes two mating surfaces and a space,
or filling. The term "joint" which has come to be used in engineer-
ing contexts for all or part of the family of discontinuities in rock
masses, is unfortunately potentially confusing for structural
engineers, who use the term joints to describe points of connection
in steel structures; in geological usage the term is applied only to
penetrative, repetitive discontinuities without appreciable shear
displacement. However, as the term joint is entrenched in its
engineering geology context, it will be retained here.
On a geological basis, we can distinguish extension and shear
joints, bedding, banding, contacts, cleavage, schistocity, foliation,
sheared zones and faults, as discussed in standard works in geology,
for example Leet and Judson (1971) and Price (1966). With reference
to mechanical and flow properties of a discontinuous rock mass, we
require considerably more information than the geological identifica-
tion. In particular, load-deformation and strength properties of
discontinuities (see Chapter 5) make specific reference to a number
of controlling quantities, including the parameters of the peak and
residual shear strength-variation with normal stress, the initial
angle of dilatancy, the normal pressure required to prevent all
dilatancy, the maximum amount a joint can close, the peak and residual
shear displacement, the tensile strength, and the normal and shear
stiffness. Though no rational formula exists for extracting the
explicit joint parameters required for an analysis from field observa-
tions, careful and detailed descriptions of the many encounters in
Rock Classification 41
out c r ops, ex cav a ti ons , and in the core box a l low the who le syst e m of
disconti nu ities in a n y pro je ct to be d i vided int o a r el atively smal l
number of types. Us uall y the fiel d des cri pt ion wil l permit reason-
able estima tes fo r certain o f the q uan t ities me ntioned and the ir
contributi on t o the rock mass cha r ac t e ri st ics. The j o int system
prop e rt ies are de r ive d from ob se rvable featur e s o f: 1 ) the discon-
tinuity surfaces ; 2) t he bl ocks they de fi ne t hro ugh t hei r repetiti on
an d intersection; 3 ) an d the p ropertie s o f t h e spac e between t h e
bl o cks. Most of t h e se feat ur e s can b e de s crib e d deterministically ,
but are better exp ressed stat is ti ca l ly through dis t ribution curves
and numer i cal in di ces.
Prop erties of t he joi n t s urface s th ems elves i nclude orientation ,
exte n t, p lan ar it y , r ou gh ne ss a nd wav in ess, a n d t he strengt h of wall
rock a s peri ti es . J o in t sy stems us ual l y di s p l a y several preferred
o ri e n tat i ons and t hi s a s pec t a l o n e re s ult s in wholly d i fferent classes
of ro ck masses. Th is sub j e c t l ends i t se l f t o st at is t ical and graphi-
cal treatment by me an s of s te re og r aphic proje c tion , which wil l be
elabo ra t ed i n Chap ter 3. Join t "e xten t ", mean i ng the t otal area or
length, us ually can n o t b e me asured dire c t ly in t h e f ield; however it
can b e estimated occas iona lly from aeria l photos (see Chapter 4).
Ro u ghness a n d wavine s s, whi ch influ e nce t he f ri ct ion a n gl e s ] di l atancy,
an d peak s h ear (Pa t ton 1 96 6 , Goodman and Dubois, 1971) refer to the
local departures fr om p l ana rity a t s ma ll and large scales resp e ctively
(fi g ure 2-8) . Th e most c onve ni e n t roughness meas ure f or rock mechan-
ics p u rp oses i s i n terms of th e l oca l an gles with respect t o the mean
pl ane through all th e hills a nd val l ey s of a j oi nt surface . Most
joints can b e r e pre se nt e d q ui te wel l b y p l ane s. P a t ton (1966)
meas u re d r ough n es s a ng les f rom e dge v iews of s u rfaces in outcrop s and
cli f f faces ; t h e require d da ta can be ob t a in e d from photo g raphs as
di s cusse d i n Ch ap ter 4. Wh en the joi n t surface i t sel f is expos ed in
outcrop , repe a t ed measur emen ts o f d i p wi ll gene rate a scattered dis-
tr i bution of va lues, who se st a n da r d devi a t ion o r mean departure may
be an estimate of t he me an r oughne ss a ng le, a s discuss e d in Chapter
5. The rough n es s ang l es i ncr ease join t s h e ar strength at l ow normal
pres sure, but at hi gh e r no rma l p ressure, t h e strength o f t he wal l
ro ck asperit ies c on t rols t he s h e ar strengt h of the joints. A good
(a)
(c)
(d)
Figure 2-8. (a) A very rough fracture surface in granite; note the perfect mating of the joint blocks across
the tracture. (b) A very rough bedding plane in limestone; voids between beds resulted trom oversliding
of asperities accompanying mass movement downslope. (c) A rough bedding plane surface; the roughness
is tormed by ripple marks preserved from the depositional surface. Photo by Dennis Lachel, (courtesy of
the Corps of Engineers). (d) A rough joint surface; the roughness is created by the intersections of the
joint with bedding; erosion has accentuated the rel ief. The rock is an argillite. Photo by Dennis Lachel,
(courtesy of the Corps of Engineers). (e) A smooth surface whose mean plane parallels bedding. Slight
roughness exists because the surface wanders from one argillite bed to another. Photo by Dennis Lachel,
(courtesy of the Corps of Engineers). (f) A very smooth, wavy surface of discontinuity undercut in the
spillway excavation of figure 1-6. This is probably an ancient shearing surface.
Rock Classif ication 43
(e) (f)
TABLE 2-1
Terminology for Spacing
of Discontinuities (after Watkins, 1910)
Figure 2-9. (opposite) (a) Sheet joints in granodiorite, Yosemite National Park, California. (b) Bedding
with tight joints in other directions. The bedding was undercut in excavation for a tunnel portal; rock
reinforcement was installed from the catwalk above the top of the excavation. Ruedi Dam, U.S. Bureau
of Reclamation. Photo by Dr. F. Heuze. (c) Closely jointed argillite with prismatic blocks; near Chivor
Dam, Colombia . (d) Bedding and two joint sets defining tabular and prismatic blocks of relatively small
size and very much larger wedge shaped blocks; near Libby Dam, Montana. Photo by Dennis Lachel,
(courtesy of the Corps of Engineers). (e) A metamorphic rock mass with three regularly spaced, tightly
closed joint sets and well developed schistocity. One well developed joint set (J1) forms the moderately
rough surface inclined towards the observer. The traces of two additional joint sets and of the schistocity
form three sets of parallel lines across the surface of J1. Near Don Pedro Dam, California. (f) Regular
columnar jointing, and completely broken, irregu larly jointed volcanic flow rock (Iatite). The columns
formed from shrinkage during cool ing, with their long axes perpendicular to the isotherms. The broken
rock represents columns that rode along on still molten material underneath. l\lear Tullock Dam, California.
46 Rock Classification
TABLE 2-8
Classes of Fau It Gouge Materials
(after Brekke and Howard, 1973)
TABLE 2-9
Classification of Rock Mass Continuity
TABLE 2-10
References to Some Engineering
Classification Systems for Rock
Ege (1968)
tunnels in granitic rocks
I
Weak Strong
Very
EH Thick
2~~----~~--~-+----+-~--~----~~-----r
VH Thick
01
E 0.6~--~--4-~~~--~~+-----~~----4-----~ C
U
H
o
0..
H Medium Vl
01 Q)
C c
u 0.2-+---4--4---~~~---=~~~~------4-----~ E
o 0..
0..
Vl 01
M Thin C
~ U
:3 U
gO.06~----~4-~--~~----+------+------+---~~
Q)
aJ
Lt 00
Very
L Thin
0.02 ~------~=-----t-----O--+---~----+---=-+-
VL VL
0.006~------~-----r------+------r------~--~"
VL L M H VH EH
0.03 0.1 0.3 1 3 10
Strength, Is ,MP
Figure 2-10. Franklin's rock
EH Extremely high L Low
quality classification; Franklin
VH Very high VL Very Low
H High EL Extremely Low et al. (1971). The strength Is
M Medium is the point load index .
Lien and Lunde (1975), for example, adjusted six parameters by means
of detailed study of 200 underground case histories, in proposing
a single numerical tunneling index -- Q.
RQD J J
(5)
r w
Q
r-
n
Ja SRF
TABLE 2-11
a) • essentially unfilled
healed 0.75
staining only; no alteration 1.0
silty or sandy coatings 3.0
clay coatings 4.0
b) filled
D. Water c o nditions Jw
dry 1.0
medium water inflow 0.66
large inflow with unfilled joints 0.5
large inflow with filled joints
which wash out 0 .3 3
high transient inflow 0. 2 - 0 .1
high continuous inflow 0.1 - 0.0 5
TABLE 2-12
After Barton, Lien, and Lunde (1975)
1.0 1.0
Q
2.0 0.83 (7)
2.0
75 1.5 1.0
Q 5 (8)
15 1.0 1.5
TABLE 2-13
Geomechanics Classification of Jointed Rock Masses
15 12 2 j 1
1 0
Dr il l core quality ROD 9 0 ~', - l ()O%. 75". - 90 % 5 - 75 ", 25"" · 50'" Z'I ' I
20 17 13
30 25 20 10
~+-------------------4-------------~-------------+-------------+-CS;I~ic
~ -en-S'~
d-ea~s-u~rla-c-eS~S-0-1t-9-0l-
, g-
e --~-
n-'M-'~-(~
~
Condition 01 jO i n ts
ve~O~o~;~II~~r~~~es S~~~~::I~~,;h s~rr~,,:s S~;~~I:a;~oUngh s~r~~es : :Gouge :, m m IhlCk OR
No seperatlOn Hard JOint wall ro c"" Soft JOint wall rock JOints open 1-5 mm JOints open 5 mm
Hard JOint wall rock Contin u o us j Oints Continuous JOints
Rat in g 25 20
10
Cla ss NO II I IV
100,- 90 90 ..... 70 70 50 SO ,. 25
Class No III IV
Cohesio n of (h e rock mas s 300 kPa 200 - l aO ,Pa 150 - 2Q!! , P a 100 . 150 , Pa 100 . "'"
TABLE 2-14
Some Factors to be Considered in
a Geotechnical Description of a Rock Mass
A. Rock material
B. Weathering Profiles
C. Discontinuities
TABL E 2·15
Input Data Form:
Geomechanics Classification of Jointed Rock Masses
S l ri k e ( tr am Iu Dip
Trace 0 f pl ane A
Upper hemisphere
pro j ec tion of A
primitive circle. This arc represents the half plane in the upper
hemisphere; the completion of the circle in the projection plane
beyond the primitive would represent the projection of the half plane
in the lower hemisphere. This fact derives from the theorem, proved
in Phillips' text on Crystallography (1946) that the projection of
a circle on the reference sphere is a circle on the projection plane .
The projection is conformal; angles measured between lines on the
sphere are preserved by the projection. As will be seen later, a
variant of the projection exists that loses this quality (the "equal
area" projection).
In dealing with problems of orientation only, i.e. where the
spatial positions of lines and planes are not considered, one may
move a line or plane parallel to itself to pass through the center
of a single reference sphere. Any plane, then, can be described by
the projection of a great circle, and any line by the projection of
a piercing point on the reference sphere. A great circle, by conven-
tion, will be assumed to represent a plane, and a point will represent
a line.
A vertical section of the reference sphere, figure 3-2a, provides
a complete description of the geometric relationships basic to the
method of projection. An upward directed line OP through the center
of the reference sphere will appear in the upper hemisphere projec-
tion at a radial distance from the center equal to r tan a/2 in the
direction of its bearing, where r is the radius of the reference
sphere and a is the complement of the angle of rise. The opposite
(tail end) of the line (-OP) projects to point -p outside the primi-
tive circle, at a radial distance equal to r cotan a/2 in the opposite
direction. If -op represents the dip vector of a plane, points p and
-p are points on the great circle representing its stereographic pro-
jection. The bisector of line (-p) (p) is the center of this great
circle which may be constructed with a compass, figure 3-2b. In
this figure, the plan view of the projection plane and the vertical
section of the reference sphere in the direction of dip have been
superimposed. It will be found that the center, q, is the same point
as the stereographic projection of a line plunging in the direction
opposite to the dip at an angle from vertical equal to twice the dip
Stereographic Projection and Joint Surveys 61
~.
2 r ctn cr/2 .. I
r ctn a/ r tan cr/
2
-p t
focus
(a)
distance pr := distance qo
stereographic projection
of sma ll circle of 20°
about E
-p
stereographic projection of
inclined plane through p and N
(dipping 20°)
primitive circle
(b)
Figure 3-2. (a) Vertical section of the reference sphere through line PO. (b) Upper hemisphere stereo-
graphic projection (plan) superimposed on the vertical section of the reference sphere through I ine PO.
62 Stereographic Projection and Joint Surveys
Stereonets
3 50 10
270 90
lA O
(a)
280 80
n. IIIIII~·
260 100
180
(b)
Stereographic Projection and Joint Surveys 65
1---- -- /2r - - - -- I
Figure 3-4. Equal area (Lambert) versus equal angle (Wulff) projections. The distance
from the center of the reference sphere to the projection of OP (referred to plane OE)
V2
is r tan (cx!2) in stereographic projection and r sin ((d2) in equal area projection.
66 Stereographic Projection and Joint Surveys
point P2' So that the radius of the resulting primitive circle will
be r, it is necessary to divide the projected lengths by 1:2 giving a
distance from the center of 1:2 r sin (a/2) for a line plunging at 90-a
from horizontal. Equatorial and polar equal area nets (Lambert nets)
are given in figure 3-5a and b. The equal area projection is not con-
formal and projections can not be made using circular constructions
as for the stereographic projection. However, all the basic con-
structions required can be performed on a tracing by sketching lines
from the nets given, exactly as presented for the stereographic
projections in examples to be discussed.
BASIC CONSTRUCTIONS
180
(a)
350 10
BlilliEF
280 80
270 90
260 100
190 170
180
(b)
Stereographic Projection and Joint Surveys 69
plot opposite -p
tion because one can switch the hemisphere, or work with the opposites
of inconveniently located pOints without difficulty. Figure 3-8
shows the relationship between upper and lower hemisphere plots of
a point and its opposite. In many problems of interest, as for exam-
ple in joint surveys, structural geology calculations, and core
orientation calculations, we do not distinguish a point from its
opposite and all required constructions can be confined to one hemi-
sphere. However, in statics, where we associate a direction with
certain lines, the entire sphere must be mapped. In this case, as
will be shown in chapter 6, we may use a single continuous projection
with focus selected carefully to reduce the size of the figure, or
we may use two side by side projections, one for the upper hemisphere,
Stereographic Projection and Joint Surveys 71
marked U.H., and the other for the lower hemisphere, marked L.H.
Figure 3-7b shows how the tracing can be used to facilitate the
plotting of a line. Mount the stereonet on cardboard and push a
thumb tack (drawing pin) through the center so that the tracing may
be revolved about the center (protect the tracing with a piece of
transparent tape to prevent enlargement of the hole with use). Mark
point B 30 degrees from the arbitrarily selected north position and
revolve the tracing to superimpose B on the equator. The angle of
rise, 20 degrees, is measured along the vertical plane represented by
the equator, counting ten great circles. The opposite to point p
can not be plotted using the net (the opposite to the lower hemisphere
projection of p is shown).
Figure 3-9a shows how to plot a plane given its strike and dip.*
Plane R strikes N 20 Wand dips 13 degrees to the NE. The strike
vector, SR' is horizontal so lies 20 degrees west of north along the
primitive circle. The other end of the strike vector (-SR) is on
the opposite end of the diameter through SR. The dip vector D and
It
* The terms strike and dip are discussed in chapter 4, page xxx
72 Stereographic Projection and Joint Surveys
plane R
primitive
(a) NORTH
SOUTH U.H. 20,0__--"'---_
nR 0
cente~r O~30
for great ~
circle
26 0
NORTH
(a)
,plane M __--..::.....
NORTH
primitive
Primitive
Figure 3-12. (a) Finding the orthograph-
ic projection of a line on a plane.
NORTH
c
R-
p rises 33°E
cp = 15 0
primitive
circle
UH
NORTH
~l
A small circle and its opposite may range over any portion of the
plane.
The construction of a small circle on the stereonet tracing is
illustrated in figure 3-13b. Points on the required small circle are
obtained by revolving the net to pass a sequence of great circles
through p, counting the required number of degrees in each direction
from p in each great circle in turn. Of course, if the conformal
stereonet is used, one can superimpose the construction of 3-13a on
the tracing to improve the accuracy of the drawing. If the small
circle lies partly in each hemisphere, as in figure 3-l3c, it may be
satisfactory for some applications to show two arcs, one being the
80 Stereographic Proiection and Joint Surveys
Pr is 51 0 to N49E
(a)
small circle about <l
through p, the locus of
p in rotation about q
Rotate p counterclockwise
120° about q to position Pr
NORTH (b)
o
U.H.
I
45 0 --- q
0
45 ---
mined. For this type of work, the precision of the counting circle
technique of contouring is questionable and one may consider using a
vector approach as discussed by Mahtab et al (1972). The mean orien-
tation of a group of unit vectors (joint normals) clustered within
a cone is determined by the orientation of the resultant R of all
the unit vectors in the cone. If all the joint normals are exactly
parallel, the magnitude of R
will equal the total number of normals
in the cluster N; as the joint directions become dispersed, the
length of R becomes less than N. Thus, as an index of dispersion of
orientations, one can use a coefficient K defined as follows*
N
K (1)
NORTH
closed curve about the mean orientation containing one half of the
total number of points, estimating its position in such a way that
the excluded points are equally distributed around the curve, if
possible; this is one estimate of the fifty percent probability curve
(P = 0.50). Rotate the diagram to orient the mean vertically (figure
3-16) . If the shape of the curves after rotation is very different
from a circle, an anisotropic Fisher distribution can be used, i.e.
one in which K varies with direction (see Shanley and Mahtab, 1975).
If the shape approximates a circle, as in figure 3-16, measure the
angle ~ from the mean orientation to the 50 percent and 95 percent
probability curves and compute K corresponding to each value from
figure 3-17, derived from the expression
K
In (1 - P) (3 )
cos (~) - 1
If the values of K.50 and K.95 are not greatly different, the
hemispherical normal distribution can be used to predict angular
distances from the mean corresponding to other probability values
(the positions of the 50 and 95 percent probability curves as drawn
can be used directly whether or not the distribution fits the hemi-
spherical normal distribution).
0 0
In the example of figure 3-16, ~50 is in the range 9.5 to 12
0 0
while ~95 is in the range 18.5 to 220. For ~50 = 11 and
0
~95 = 20.5 , figure 3-17 gives values of K respectively equal to 39
and 47. If we retain the value K =
39, we could predict for example
l 0
that the 99% probability angle (~99) is cos- (1 + In (0.01/39» = 28 .
Given the mean joint orientation in a design problem involving poten-
tial sliding, one might then consider the "design joint" to be
o
inclined 28 steeper than the mean attitude. The suggested approach
will be, however, to retain the mean joint orientation in calcula-
tions and reduce the angle of friction accordingly, as discussed in
Chapter 6.
30.-----~----~~------r_----,_----~--------._----_r----~------~
25
0
~
20
C-
o
(l)
E
(l)
..c
15
E
~
'+-- 10
~
(J'l
c:
«
5
0
5 10 20 50 100 200 500 1000 2000 5000
Dispersion Index, K
Figure 3-17. Angles of 50% and 95% probability poles from the mean of a hem ispherical normal distrib ution
After Arnold (1941), Fisher (1953), and Mahtab et al (1972) .
outcrops are biased because joints pa ral lel to the surfa ce or the
axis of the borehole are not seen. Let a be the an gle between the
normal to a joint and the axis of a borehole or t h e n orm al to a
surface outcrop, (figure 3-18). The act ua l number of intersections
with the outcrop or borehole by a se r ies of pa ra llel joints wi ll
depend on a. Only when a equals 90° will the true spacing equal the
outcrop length divided by the number of joints; similarly, onl y when a
equals 0° wil l the spacing equal the drill hole length divided by the
number of joints. However, the data can be cor re cted as shown i n fig u re
3-18. Let N be the number of joints obse rved at an gle a. We can the n
a
calculate a "correct" number of joints Nc as follows.
For borehole measurements:
N
c
Na Icos a (4a)
N Nisin
a
a Ia I (4b)
c
88 Stereographic Projection and Joint Surveys
a.
s
A Na in counting
~. Circle
~n
Surface
counting "Blind zone"
circle -20 0 < a < 20°
Figure 3-18. Correction
for bias in joint surveys;
after Ruth Terzaghi (1965).
12
13 14
Existing Maps
grade into logs, but these rarely will be found. Available general
purpose geological maps show the mapper's opinion, based upon incom-
plete evidence, on the distribution of mappable bedrock formations
and structures below the mantle of soil. The surface distributions
of mappable formations are shown by means of the surface traces of
their contacts with adjacent units. By means of structural symbols
and supplementary cross sections, subsurface information is presented
as well. Such maps are basic for all rock engineering.
Exploration of Rock Conditions 93
~~:\:~elr~~~~~y,F~~~~~~~~,
G kF th1n 1y bedd ed to th i n ly
lami n ated, very cl osely jo i n ted, slig h tl y to mo derately
we at hered, poo rly c leaved SHALE, wea k , imp e rmeable excep t along
open joints . Inter bedded wi th ve r y s ubo rdina t e grey to dark
greenish grey, fine-grained , ve r y t h i n ly bedded, th i nly lami nat e d
and cross-laminated, clo sely joi nted, s l ig ht l y to moder at el y
weathered SILTSTONE, moder a t ely s tro ng, a nd dar k gr e enish grey
medium grained, very thinl y to med ium be dded , wi th c lo se ly to
widely spaced joints slig ht ly to modera tely we athere d , SAN DSTONE,
strong.
The shale slakes on expo su re and is sui tab le for brick making.
~ ~t~~~~~~~~ le
predominates.
to map gr ou ps of beds in whi ch SA NDSTONE
Beds a r e usua lly less t ha n 12 in . thic k an d are
separated by very thin beds of s iltstone and s ha l e .
Sand stones are suitable for a ggr e ga te production.
Within the contact metamorphic aureole of th e gr a nit e , dar k gr ey ,
very pale orange to dus ky yel lowish br ow n , f i ne t o me di um g r ain ed,
thinly bedded, close ly jo i nted, s light ly t o moderate ly weathered,
hornfelsed SHALE and SAND STONE, strong , i mpe r vious e xcep t along
open joints. Local l y with fine gr ain e d bl a ck t o urmal ine
developed as selvedges up to 1 in. wi de a l ong di s co nti nuities and
with irregular quartz veins up to 2 i n. wide.
Geological Mapping
400
~~5CXJ
?
550----'''
\
2== != N
o 100 200ft.
Scole
Figure 4-2. Geologic map of a portion of the left abutment of an earth dam;
(courtesy Kaiser Engineers, Oakland).
Exploration of Rock Conditions 95
Di p vector
of plane p
TAB LE 4-1
Symbols fo r Geological Maps (courtesy U .. Geol. Survey )
40
• Contact, sh ow ing dip ...... ,~
Norma l fault; hachures on downthrown si de
90 10
I Contact, vertical (left) and overtu rned ....L. Strike and di p of bedding
Con tact, located ap prox . (give limits) ~ Strike and dip of overturned beddin g
Contact, located very approx. 90
--+ Strike of vertical bedding
,
90 Fa ult, showing dips (!) Horizontal bedding
10 20
Fa ult, located appro x . (give limits) -A. .Jo,., Und ulatory or c rumpled beds
60
- --?- -- Fa ult, existe nce uncertai n I St ri ke and dip of bedding, uncertain
Fault, projected beneath m ap ped units 70
...L ~? St rike of bedding certa in but di ps uncer ta in
Possibl e fau lt (as located from aerial Stri ke and dip of foliations
photographs) Stri ke of vertical foliati ons
<! 75 0
u
Fa ult, showin g tren d and plunge of
linear features (0, down-thrown
+++ Hori zontal foliations
Stri ke and d ip w here bedd ing
side; U, upthrown side) para llels foli ati on
Fau lt, show ing relative hor izontal Strike and dip o f jo ints (left)
move ment and veins or d ikes
Thr ust faults; T or sawteeth in Strike of vertical jo in ts (left)
u p pe r pl ate and veins or dikes
Pault zon es, sh owi ng ave . d ips Horizontal jo ints (left) and vei ns or dikes
98 Exploration of Rock Conditions
amount of dip is known but not the strike, and the surface trace is
known only at several points. Figure 4-4 shows a strike and dip
symbol plotted at the outcrop of a clay seam. Based upon the measured
plane, denoted stratum 1, one can estimate the trace of the clay seam
across the ground surface, which can then be checked by excavations
through the soil. Since a line parallel to the strike of a plane is
a contour on the plane, the clay seam should also underlie the sur~
meters) .
In these examples one observes how the trace of an inclined
plane forms a "V" pointing in the direction of dip upon crossing a
valley. The trace of a horizontal plane will follow the contours
while a vertical plane will pursue a straight course in any terrain.
A plane inclined mo~e gently than the relief provides a more complex
and often multiple trace across the surface.
Cross-sections to convey the geological structure can be drawn
in any direction by projecting surface points parallel to the strike
into the desired section. In figure 4-5, a vertical section has been
Exploration of Rock Conditions 101
drawn through Sl and S2' The apparent ang le of dip in the section
diminishes as the angle (a) between the section and the strike
approaches O. Section Sl S2' in figure 4-5 with a = 42 gives an
apparent dip (AD) of 33.8 0 . It could also have been calculated from
the relationship derived from figure 4-6.
t' (2)
a structure contour
(c)
100 foot spacing figure may represent a reasonable estimate for the
spacing of ~ey discontinuities. Field study would tell. Observe
that the lengths of particular individual fractures are more than
1400 feet. Figure 4-7d presents a similar example but one in which
spacings are more variable. Again, two approximately orthogonal
sets are expressed in the terrain. The Northeast trending joints are
deeply etched in a closely spaced zone. Figure 4-7e presents a
fracture system representative of granitic terrains. There are at
least three different sets with considerable scatter about the mean
orientation of each, as well as considerable variation in length.
Study of aerial photographs may be the only reasonable way in many
instances, to gain data about joint lengths.
Fiduci al mark
Yo
YR
----,0
_ ---, R I TR
TO
I
I I I I
I I I I
0
. t- x +x'
Figure 4-8. Parallax and coordinates of points in ove rlap ping vertica l aerial photographs.
(3 )
dh ( 4)
RQ
(5)
where H is the camera height above the principal point, and b, the
photo base, is the distance between the principal point and the con-
jugate principal point (00'). Both Hand b are averaged from the
two photographs. Let SRQ be the distance from R to Q measured on the
photograph. The ground distance between Rand Q is approximately
equal to SRQ(HQ/f). Thus, the plunge (S) of RQ is given by
(f) (dP)
Tan S RQ
(6)
(SRQ) (PQ)
Using the average photo base b and flying height H, if the point Q
is at about the same elevation as the principal point,
Tan S ( 7)
(a)
(b)
Figure 4-9. Measurement of dip using a parallax wedge: (a) th e floa t ing line pierces the ground at t he
bottom of t he sl ope; (b) the floati ng line p ierces the ground at th e top of th e dip slope.
108 Exploration of Rock Conditions
the average person 1 s eye-to-eye spacing. When placed over the stereo
pair and viewed with a lens stereoscope, the lines fuse to give the
appearance of one sloping line in space. In figure 4-9a, the sloping
line appears to pierce the ground at point Q at the top of a dip
slope. The separation of the lines on the wedge is 0.0130 feet.
In figure 4-9a, the wedge has been moved to make the floating line
pierce the base of the slope at R; the separation is now 0.0210
feet (estimating the last digit) giving a parallax difference dPRQ
of 0.0088 feet. The photo distance SRQ is 0.043 feet. The photos
were taken through a lens having a six inch focal length and the
photo base, b, averaged from the two photos is 0.317 feet (this can
not be verified on the stereopair of figure 4-9 as not enough of the
image area has been copied). Solving equation 7, the bed is dipping
at an angle of 17.5 degrees.
If the angle between the lines on the parallax wedge can be
varied while the model is viewed, the inclination of the line in
the model will appear to change. This is the principle of the
"parallax ladder"* for measurement of slopes. "Hackettrs method"
employs an analogue device to measure directly the exaggerated slope
angle in the stereo-model; the exaggeration factor is then obtained
by a graphical procedure as discussed by Ray (1960).
The most accurate measurements of attitudes will be obtained
using a plotting instrument such as a Kelsh plotter or Multiplex, in
wh~ch a tilt free stereo-model is created by projecting through film
or glass plate diapositives of the aerial photographs. The attitude
of a planar surface can be obtained by mathematical plane-fitting
to coordinates of three or more points on the surface. Ross-Brown
(1972) in fact even determined joint roughness by computing the
scatter of planes fitting four or more points in a stereo-model of
terrestrial photographs. Attitudes of discontinuity surfaces in the
model can also be measured by tilting the plotting platen into
conformity with the surface to be measured and reading the attitude
of the table with a clinometer just as it is done on the outcrop
(Ray, 1960).
Black and white vertical aerial photos like those in figure 4-7
have been taken of nearly every part of the United States by agencies
of the U.S. governmen t and are at t ai nable a t modest c ost. Table 4 -2
lists the principal sources of photographic coverage in the United
States. An order may be expedited by visiting the offices of the
agency and identifying the photo exposure numbers on an index mosaic.
Alternatively, the area desired in photos can be marked on a topo-
graphic map and forwarded with the order. Faster response can be
obtained through private aerial photo companies. Local government
agencies , such as the County Tax Assessor's Office, may possess
recent aerial photos, flown for them on contract, which might be
borrowed. For analysis of fracture patterns, high altitude photo-
graphs, e.g. 30,000 or 20,000 feet, are often more desirable than
the lower altitude photos, e.g. 5000 feet, used to make contour maps
for engineering design.
In military and space research, there has been considerable
interest in expanding the scope of photo interpretation outside of
the visible spectrum through the use of special films and artifici a l
imagery techniques (Cassines, 1972). Infra-red photographs bring
out contrast of wet and dry terrain. Color photography, though rather
expensive, can point out zones of alt e ration. Thermal imagery
produces a visual record from natural heat radiating sources, such
that warmer areas appear lighter in tone; thermal imagery can some-
times show up seepage along underground paths, subaqueous discharge,
110 Exploration of Rock Conditions
TABLE 4-2
Sources of Existing Government Aerial Photos
National Air Photo Library, Surveys and Mapping Buildings, Room 130,
615 Booth Street, Ottawa 4, Ontario, CANADA. (This is the centralized
source for Canadian air photos)
(a) (b)
Figu re 4 -10. Low sun angle aerial p hotography will d isplay shadows al ong pro min ent discontinuiti es, such
as the Wasatch Fault scarp photographed here. (a) A convent ion al vertical aerial ph otograph of a p orti on of
the Wasatch front, Utah, photo grap hed at mid-day. (b) A port ion of t he same area, photographed early in
the morning. Reproduced from Cluff and Slemmons (1 971).
112 Exploration of Rock Conditions
TERRESTRIAL PHOTOGRAPHS
Measurements on Photographs
'~--.--:::------I k
o = c amera stat ion
p = pri nci pa l point
kq' = hori zan
n ': nodi r point
L = isoce nter
Hori zo nta l ang le to q ;: H q :: k Lq '
Vertical a ngle to q :: Vq
Plun ge of camera ax i s ;: -1 k Op =8
Di stance s :
O p = foco l le ngth
p k = f to n 8
Figure 4·11 . Geometry of a Ok ,: k L ': flees 8
photograph. p n = f/ tonB
tan H f se c8 + y sin8
( 8)
q
q' (b) ,
(a)
\ \ L-------' f-'
J-\ I
\ ~ q,\
~ ----- I\
Hr
., HQ
\ ----
If::::
-- -c-.............
------ --- p- ~
i\...Horizon
n' \
\
~ I----- c-. . . . . . .~
~
\ v"-- ----
~\
~ r ~
\
L 1 -------- \ \
Yq cosB sinH q
tan V (9)
q xq
(a)
Posi tive
plane
o~----------~--------~~~
(b)
\ \ \ ,\\\\\\\\1\\11 \ TT1111111 1111 1 / 1 / / 1 / 1 I I
\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ I \ \ I I \ \ \ I I I I I
\ \ \ \ \ \ \11\\ \\11 \JIILI I I I I
\ \ \ \ \ \ \ \ \ \ \ \ \ \ \1111111 I I I I I
\
II \ \ \ 1111 I / / I
\
\ \
'\ \ \ \ TTl I 777
I
I
I
I
TTTTT!\\ \ I
°
I \
1 -r! / / I / / / , ff ! 1/ / I / I I / TlT!\\\\ 1 \ 1
lll T n l\
~H-I I I / / 11\ \1 1\ 11\ \\ 11\ \ \ \ \ \ \ \ \ \ \ \ \ \
II I II II I //II /II!I" I I I /I " / \1 \11111 1\\ \ 1 1 \ \ \ \ \ \ \ \ \ \ \ \ \ 1
I I / I I I -I 7 -I I / / / , I , , , , , I I I I I / / I I IT T1\ lT! \ 1 1 \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \
Goodman- Geologi ca l Eng-9
116 Exploration of Rock Conditions
Gnomonic projection
of a great circle - -
a plane through the
camera
y = f tan 8
Gnomonic projection
of a small circle of
latitude a
x = f ton 8
cos a
Figure 4-13. (c) Explanation of the gnomonic net.
0
is a 1 gnomonic net which can be superimposed on a photo if enlarged
or reduced accordingly. This net, plotted by computer, corresponds
to a focal length of 84.5 mm. (The focal length after any enlargement
will be found. by the relationship f y/tane where y is the distance
along the equator corresponding to an angle e.) As shown in figure
4-13c, the straight lines are the gnomonic projections of lines of
longitude -- great circles of the reference sphere having a common
intersection; they can represent any plane passing through the point
O. The curved lines are the projections of lines of latitude, i.e.,
small circles of the reference sphere; they graduate the straight
lines. To find the angle subtended at the camera between any two
Exp loration of Rock Conditions 117
~
r- p
Vr
r--
~?
~ - ro-
I
Figure 4-14. Fi nd ing the horizon from known f-f-- I V
v-
vert ical angles t o two poin ts r and q . Rotate the V
r ~
t raci ng on t he gnomo nic net, about poi nt p, to ..-- ---... ..........
~
fin d an orientation giving correct values of vr and ~
vq . Then check the hori zontal angles Hr and Hq .
(see f igure 4~12b ) .
118 Exp /orCltion of Rock Conditions
Example
(b)
Figure 4-15. (a), (b) Photographs of a road cut taken from the ends of a surveyed
base line. Horizontal and vertical angles to a number of points in each photo were
measured with a Brunton compass when the photos were taken. The enlarged focal
length (ef) is 84.5 mm. The map of figure 4-15c was made from enlargements to
248.5 mm. (c) Plan and cross section of the cut slope shown in figure 4-15a and b
by triangu lation from the two photographs.
B
'-:'0
+
~'j
Q .' .
/N
Scole of Sec llon
0 50 100 fl
I I I
0 100 200 ft
Sco le o f Pi on
po7
G ,m ero /
comerojpolnl 2
OO lum E levollon
0, 02
GEOPHYSICAL METHODS
Dobrin (1960), Heiland (1946), Jakosky (1950), Van Nostrand and Cook
(1966) and others. Many of these methods have application in
geological engineering as summarized by Griffiths and King (1965).
Here it will be sufficient to list those techniques which help to
describe geometrical and mechanical properties of discontinuities
in the rock.
Geophysical surveys usually yield two types of results. First,
they evaluate a physical property at different points within the
rock mass -- properties like the velocity of sound waves, magnetic
susceptibility, and density. This result may be a direct product of
the instrument readings, but more usually it is calculated from the
data. Secondly, the geophysical exploration will usually reveal the
distribution of the measured quantities over the map; interpreta-
tion of these data, based upon a suitable model of the geologic
structure, will often permit the geometric constants of the sub-
surface structural model to be determined. For example, data might
be interpreted to yield the depth of each layer in a multi-layered
configuration, or the width of a fault zone in a model with a tabular
fault between continuous walls.
The physical quantities measured by geophysical methods may be
of direct value in a design problem. The longitudinal wave velocity,
for example, :is used in assessing potential damage from blasting
operations. But more frequently, the real value in knowing the
measured quantity lies in its associations, such as with the rippa-
bility of rocks; rocks with sonic velocity less than 7000 feet per
second are usually excavatable by ripping, whereas rocks with
velocities greater than 10,000 feet per second will usually require
blasting (Caterpillar Tractor Co., 1966). Figure 4-16 shows how
the sonic velocity correlates with the characteristics .of weathered
granite, including the percent of "core stones" and the porosity
(Polak, 1963). The next section considers briefly some methods which
are relevant to the problem of mapping and assessing discontinuities
in a rock mass.
Seismic Methods
c~a!i!( zone 4
'lii
tI:
.E
(I)
E
~ O~------~~---~~------~~------~------~~---~~------~~~
Shot point
I
DETERMINATION OF VELOCITY
ALONG REFRACTOR (V2)
10
uQ)
E5 Vz =6900 m,lsec
~ O~------------------~~----~-----------------------~
I
..... -5
-10
uQ)
~
':""N 10
£5
"
~ O~------------------------------~---------------------------------------~
Distance
O~---------------------------------------------------------r---------------'
GEOLOGICAL PROFILE ?
'2
E
4
a.
(I)
6
o
Figure 4-17. Approximate graphical method for locating faults from seismic
refraction "reversed" profiles. Based upon Griffiths and King (1965)
Resistivity Surveys
r 2'TT a 6V I I (10 )
South North
400r-----r-----r-----r---~~--~~--~r_--~r_--_.----_.~--~
Fault
April 1972 zone
f\~
E
I 6
,',: 'q:'
~
C: , .~: ~
?:'
200 ,p Ji '\
I: .
':; ii
I •
~ , .
/ :
'iii
Q)
Il:
r
0
Q 200 800 1000
Distonce
Figure 4~ 18. Results of a fixed Wenner resistivity profile across a fault zone. From Stahl (1973)
126 Exp loration of Rock Conditions
Gravity Surveys
Magnetic Methods
The int ensi ty of the eart h's magne tic field at a point on the
surface is the sum of two vect ors -- the induced magnetic intensity
derived f r om t h e eart h 's magneti c field acting on rocks of g i ven
magnetic s u scept i bi lity, an d remanen t magneti sm in t h e rocks. Si nce
rocks vary in their magnetic suscepti b i lity and r emanent magnetism,
magnetometer surveys can map the boundaries of formations. Rocks
Exploration of Rock Conditions 127
DRilL HOLES
Qj 1.0r-----oo;;::r......::--~c____._----..._----.,.___----r__---___;----......,
~
~
....
.2
~ 0.8
c: Disc
o
U L/W: rectangles
~
2
5 06
u 0 .
If)
6~
o~
<l
g'-
t; °0.4
Cl)
~
2c:
....
'0 0 .2
.c
o
.0 O L __ _ _- L_ _ _ _~________L __ _ _ _ _ _~_ _ _ _ _ _ _ _~_ _ _ _ _ _~_ _ _ _ _ _ _ _~
Methods of Drilling
TABLE 4-3
Drilling Methods
Down the Hole Percussive Drilling - For making holes 4" to 6" in
diameter without a sample, in hard rocks. Usual depth range
120-200 feet. Can be drilled in any orientation.
Rotary Drag Bit Drilling - Very fast drilling in soft to medium hard
rocks, but without a sample. 2-3/8" to 7-1/8" diameter holes
in any orientation. Maximum depth in the range 100 to 250
feet.
penetrate all but the hardest rocks. This method was developed by
the petroleum industry where high bit pressures can be provided by
Exploration of Rock Conditions 131
the weight of the drill string. For shallow drilling, very heavy
rigs are needed. A continuous rock core is obtained using diamond
coring bits (figure 4-20) or shot drilling methods. The latter work
by turning a slotted steel cylinder on continuously supplied abrasive
steel shot; this slow expensive method is used only on very large
holes or for smaller holes in situations where diamond bits are
unobtainable. Czech engineers, for example, have used shot drilling
to explore the foundations of dam sites because they could not pur-
chase diamond bits. Large diameter shot drilled holes (36" to 60"
"calyx" holes) were used by the T. V. A., the Corps of Engineers, and
other engineering agencies to permit downhole inspection of the rock
by a geologist.* Oriented core techniques, and to a lesser extent
bore-hole cameras and television devices, have superseded calyx drill-
ing for exploration in all but very special circumstances.
Good drilling practice reduces costs and is vital in coring for
civil engineering work since the quality and completeness of the
sample are sensitive to drilling efficiency. Thrust, rotation and
flushing action must be properly adjusted according to equipment
and rock conditions: Thrust applied to the bit forces the mechanical
work to be done on the rock and not on the drill. In diamond drilling
it is the force per carat which controls; too high a value breaks
diamonds and too Iowa value polishes them. Flushing the bottom with
fluid, water or compressed air, transports the cuttings from the rock
breakage region and cools the cutting tools; an insufficient flush-
ing rate risks plugging of the annulus, indirectly causing blocking
of the drill core. The speed of rotation affects the cutting action
of the bit; it varies according to the bit design and rock type. The
optimum speed of rotation (in the range 250 to 1500 rpm) is deter-
mined by the minimum thrust. Since the controlling variable is the
tangential velocity of the diamonds, larger diameter bits must be
turned more slowly than smaller ones.
For maximum core recovery, diamond drills for civil engineering
geology should be "hydraulic feed" types, in which rotation speed
Wo ter Sw i vel
i,·rPum p
Dri ve Pipe
c D ~ c:::::, o v D
: -
Cas ing _- =- _ _-
H - 15 Dri ll
Co lumn Moun ted
Figure 4-20. Typical diamond drilling set-ups; (courtesy Joy Manufacturing Co.) .
Exploration of Rock Conditions 133
Core Barrels
TABLE 4-4
Standard Sizes of Drill Holes
and Typical Sizes of Drill Cores
86 58 72 Craelius Co.
76 48 62
66 38 52
56 34 42
Figure 4-21. Several types of diamond drill bits; (courtesy of Mindri ll, Ltd.).
136 Exploration of Rock Conditions
against the core inside the bit and returns outside of the outer
barrel. In erodible or seamy formations, the water stream can be
further separated from the core by using bottom (face) discharge bits
(figure 4-21). Longyear Co's "L" series core barrels (e.g. NXL) and
Joy, and Sprague and Henwood's "M" series (e.g. NXM) reflect improved
designs which among other things discharge the wash water closer to
the bit.
Blocking and wedging of core inside the barrel can cause core
to break, especially if it is of small diameter. Ideally, core
barrels should not be smaller than NX size to explore discontinuous
rocks for civil works.
Grinding in the barrel occurs when previously drilled core
turns against fixed core still attached to the bottom (figure 4-22).
Many feet of core can be lost in this way. Grinding can be minimized
by wi thdrawing ("pulling") core frequent ly so that the barrel is
never allowed to fill up. However, frequent interruption of "drilling
and withdrawal of all the rods to empty the barrel proves costly.
In vertical and in steeply inclined holes, "wire line" equipment can
be used (e.g. Longyear's Q series) in which the inner tube can be
uncoupled from the core barrel and pulled to the surface by a wire
Exploration of Rock Conditions 137
cable in side t he string of special hollow dril l rods. While the t ube
is be i ng e mpt i ed , a seco nd inn e r t ube is lowered and drilling con-
ti nues. If core separates along natural or new f ractures i t can
wedge against the sides o f t he barrel , blocking e ntrance of new core.
With Mindrill's S barrels and the L series barre ls, c irculation of
wat e r c uts o f f as soon as a block occurs t o warn the driller. To
pre ven t grinding, t he core should be prevented from rotating by
means o f a free swivel joint betwe en the top of the inner tube and
t he he ad o f the barrel. Ball bearing swivel type heads are now
general l y avail a bl e for al l siz e s of double tube core barrels, e . g.
t he L , M and "Large " series mentioned above , and should be selected
f or g e o l ogical engineering.
I deally, core should be removed f rom th e sample tube (inner
tube ) without disturbance to preserve mating across natural f rac t ures
a nd to preven t loss of f illing ma teria l or small rock pieces.
Longyear , Mindr i ll and other manu f acturers provide a core barr e l
eit he r with a split inner t ube or as an additional t ube sample holder
wi t hin t he inner tub e ( fig u r e 4-23a), wherein the spli t t ube is
e xt r uded hydraulically from the inn e r tube and separated t o e xpose
t h e core wi t ho ut disturbance (figure 4-23b). A t riple tube barrel
can be made b y i nserting a spli t plasti c tub e inside the inner core
barre l of a double tube t ype .
Good core recovery can be difficult to achieve in friable
mate rials . Christensen Diamond Products Co. develope d a rubber
s l ee v e core barre l which encases the core in a tigh t fitting neopre n e
sl eeve as it come s into the barrel. This is par t icularly useful
when drilling f riab l e rocks with drilling mud which can invade and
dama ge the core. Another special barrel i s Chris t ensen's "pressure
core barrel ", which tries to preserve the original por e f luid in
the core sample by seal i ng t he i nner tube as soon as it is occupied.
Good core recovery is also difficult to achieve in materials of
variable q u al ity , e. g. wh e re hard rock and soil occur in the wea t h e r-
ing profile. F or uniformly soft intervals, large diameter design
cor e barre l s ' have been equipped wi th soi l sampling extensions on
the i nne r t ube which work l ike drive sampl e rs when pushed ahead of
t he coring bit . Be rents ( 1961 ) reported a design by the Snowy
138 Exploration of Rock Conditions
(a)
Figure 4-23. (a) Longyear triple tube wireline core barrel (03). (b) Extruding the split inner tube of the 03
barrel; (courtesy Longyear Co.).
CORl: SKrrCH
I' ,,
I I I I , I i I I I I I I I I J:tl~ "!'wI !. 11f I I I I I I I Ii, I I I I I I I I r
GiY@ (i)
M ~ _=- ~~J_ ___ ~
,
u
IY '
- ~ 0- -----n--if2-
LL-L _ _
__ _
- w- ~ - ---, --- ~~-.
~_ ~'- _ _ ---1 _ _ _L- ___
A @(f) tv( M
Hole # _--"0_-..:..
4 _ _ _ _ _ Box # _ _1,--7 _
DEPrn INTE.~VAL OF BOX : 23 5 .6 to 250 . 4 INFOEHATIOr< fROM DRILLERS' LOG
CORE RUNS TNCwD'ED; - "'2.i4T t o 2'44:'"5 Drill ing Re.t~ : .1- .4 ft/min
244 . 5 to 2s4:'5 We. ter Lo ss:.....;;;,.l0;",;0'-;1_ _ _ __ -,--,-_
- - - to - - - Remarks: '...'" t " r r et urn a[ end of
core __n_ _ _ _ _ _ _ _ _ __
r u~
ROCK
ROCk. type, color, d egr':E of w6athering: Fre s h ~re·. granod ior it e
Texture, structure/foUati on: Med ium g rained. Fal i a t i o n indi s tinct a t /.4 - 50°
CHARACTER OF JOItfTS #1 #2 #3 #~
Inclination: e
59° 30° Il J5°
Remark!! :
#1 Subparall e l foli a e ion
the core box not only presents the data vividly, but provides an
interpretation and classification of the discontinuities that will
be appreciated when entering into analysis and decision making.
The potential penalties for careless or faulty logging proce-
dures are high, as the logs are often included as legal entities
available to bidders and become a part of the contract. One legal
case concerning rock excavation centered around a marginal notation
"probably needs blasting" scribbled on a log sheet in the field.
In Zambia, a construction company blamed a faulty log of a boring
at the site of an underground power house for a chain of events which
led ultimately to the forfeiture of the contract and the bankruptcy
of the company*; in this case it was alleged that biotite schist
bands within a biotite gneiss, logged uniformly as "sound gneiss",
created a falling rock condition.
~~~
(a)
II 90-0 # I
YPlanep (b)
,
I
- Core gu id e \
I
:
/ - ............ ..... ~
/'
':"
~------.....-: -:::.'
Reference
line
CD
Drilling
direc t ion - --- ............
"
+
- Gra dua ted circle
(c) Fr i c tion kno b
Figure 4-25. (a) Rosengren's reference line and the angles a and (3. (b) A simple
method of measuring (3; {1 = 2500 (courtesy of Golder and Brawner Associates,
Vancouver). (c) A goniometer for measurement of a ; (courtesy of Golder and
Brawner Associates, Vancouver).
144 Exploration of Rock Conditions
Figure 4-26. Core being referenced and described; note the continuous reference
line on the mated lengths of core laid in /Iv" troughs (courtesy of Dr. K. Matthews,
Mt. Isa Mining Co., Australia).
The downward end of the major axis of the fracture ellipse meets the
core circumference at an angle S from the reference line (measured
clockwise looking in the direction of drilling). This can be
measured routinely using a circumferential band calibrated in degrees
(figure 4-25b). Figure 4-26 shows core being referenced as above.
The angles a and S determine the orientations of any fracture in
coordinates local to the particular reference line. If the absolute
orientation is known or can be determined for any planar feature
(other than one perpendicular to the drill hole), all other planes
can be oriented absolutely.
The stereographic projection facilitates the solution. Consider
the ellipse formed by the intersection of an inclined plane of
discontinuity and a vertical cylinder (figure 4-25a). The minor
axis of the ellipse -- the strike of the discontinuity when the core
is vertical -- is the line of intersection of the plane of discon-
tinuity and the plane perpendicular to the axis of the core. Since
the latter is the plane in which S is measured, we shall call it
"the S plane". The major axis of the ellipse of intersection projects
into the B plane at 90 0 from the minor axis (the B value for a plane
is measured from the reference line to the major axis). Respecting
Exploration of Rock Conditions 145
the sign convention given , the major axis wil l be in the lower hemi-
sphere if t he hole is downward. Figure 4-27a shows a s tereographic
pr oje ction relating the angles SR and a
R for a reference plane (R)
o f kn ow n attitude in a drill hole (H) of known orientation. The S
p l an e is constructed as the p l ane normal to H. and its intersection
wit h R, together with t he measured va l ue of SR determines the position
of S = O. In this i llustration , t he drill hole H is inclined 70° to
0
t he n o rth, while the reference plane strikes N40E and dips 40 to
t he southeas t. SR = 7 o and u 0
60 .
R
Now that the orientation of the reference line in the 8 plane
has b een det e rmined , any other plane defined by values of a and S
may be oriented absolutely. For example, (figure 4-27b ) a plane PI
in t h e same s ection of core as the reference l ine of figure 4-27a,
0
has values a p = 45 and Sp = 65° . First mark the point P' at 65°
from S = 0 along the S plane . Then, in the plane BP', determin e the
posit i on of t he major axi s of the ellipse MAJ(P) located 45 0 f rom
H. Next plot one of th e minor axes of the ellipse MIN(P) at an
angle 90° from P' along the S plane (the sign does not matter). Plane
p is, de f ined by the great circle common to MIN(P) and MAJ(P) whi,ch
0
i n t he example st rikes N75E, and dips 64 t o th e South.
Th i s solution, as wel l as others to be discussed , requires a
knowledge of the orientati o n of the dril l hole. Boreholes , especially
i nclined ones, tend to wander f rom their initial orientation aft e r
a hundred feet or more, depending upon the nature of the rock ; the
problem can be acute when t he angle (a) of foliation or bedding
becomes small . Wherever core orientation is being considered , it is
good prac ti ce t o order a survey for the inclination of the dril l
h ole . This can be done by a variety of methods as discussed by
Cummin g ( 1956) .
,
J
I
NR-~90-QR I
~--__ ~ I
QR~ I
--0-__
-- ............ -.......... I
............. I
MAJ(R~ ........... j3::0
,.......... 11
/ ............. v~
/
/
e>DR
900 /
'/
/
/
/
/
MIN (A), /' j3 plane
--- -=--~---
(a)
Figure 4-27 _ (a) Absolute orientation of the reference line in a lower hemisphere
stereographic projection. (b) Absolute orientation of plane P given its lX and {3 values.
lXp = 45° and {3p = 65°
NORTH
,H //
\ //
4~o / / /'
, /'
Plane P~ \ //
(required) ',- . __ ---
MIN (P) _ _ __ ____ -~MAJ (P)
-------- \
\
\
\...
,
\
(b)
NA = N 8 W, 60° NORTH
NS=S8W,60°
0
NC = S 17 W, 50
ND=S75E,60°
0
N E = N SOW, 18
0 Hole 3
N F = N 62W, 11 90 - ex
= 46°
Hole2~
0
90 -a. =45 0
~Hole 3
o 90-a =46
0
Small circles of required radius are constructed about the hole axes
yielding 6 intersections. (In the case of drill hole 3, a part of
the cone about the upward direction of the hole is required as the
Exploration of Rock Condit ions 149
- - -- Pressure Rod
Steel Pins
Pin Cylinder
Ball Chamber
~LOCking Wedge
Spiral Spring
(b)
(a)
' ~
/
"
/ ,
:
(al
COAXIAL
BOREHOLE
(a)
DOD U
(0) (el (d ) (el
(b)
(c)
Exploration of Rock Conditions 153
N
o
5563 .. 74
0
N 45 W
5 564 -
5565 -
q q
~
y
~p ~p
(a) (b)
q q
~p ~p
q q
DETERMINATION OF PROPERTIES
Samples of Joints
e;;;?
compound
3 clay dam
4 rock outcrop
( ol (d)
--
(j)
diamond
saw cut
shallow
groove
~pore
pressure
input
heat treated
50CrNi 13
soft steel shield
ItnriiJImjr-r--++- jacket
- confining
pressure
input
fluid
TIlrITll'It--I----it---+- dis t ri bu t ing
disc
pore
~ pressure
meter
Figure 5-2. Triaxial test chamber;
cm ! !! ! scale (courtesy of Dr. F. Rummel, Ruhr
o 2 3 , 5 University, Bochum, Germany).
"potting compound" (e.g. "Dow Corning Silica Set No. 105"), after
which the mating rough surfaces are cast with hydrostone, sulfur
capping compound ("cylcap") or epoxy (figure 5-1d); this allows
repeated virgin tests with different environmental conditions, for
example different filling material characteristics (Schneider,1974).
(b)
Sh ear strength
curve f or j oint
Stress path on
Joint Figure 5-3. (a) Application of Mohr's
1st stage/ /
/2nd stj-lge circle for examination of the tendency
to slip on a discontinuity. (b) Stress
/ /3rd stage path during a multi-stage triaxial test.
/ / (c) Stress versus deformation duri ng
/ the mUlti-stage triax ial test.
(d) Deformations during the mu lti-
stage triaxial test.
(c) (d)
x tJ.z
,... ____ 1 ___ ,
v I I
tJ.x
Unbonded
strain gage
Specimen Shortening
162 Mechanical Properties of Discontinuities
(1)
T (2)
Thus the stress path during a triaxial test with a joint is as shown
in figure 5-3b. When slip is initiated at A, the confining pressure
is quickly raised from G
3,
°
0 to 3 1 and then the deviator stress is
,
raised provoking slip at B, etc. This test may therefore be termed
a "multistage" triaxial test.
The results of such a test can be useful only if the specimen
is free to slide along the joint without frictional restraint from
the ends. Rosengren (1968) showed that a friction coefficient K at
the ends adds new stress contributions on the joint of and TF , in
the first loading stage, equal to:
K °1 sinljJ cosljJ
and (3)
~z cos~ + ~x sin~
and (4)
~v ~z sin~ - ~x cos~
-1 ~u simp
where 8 cos D o < 8 < TT/2
TTD2
4 sinlJl - D ( ~u) (5 a)
rI I ~+RI ~
U t ~ ~ clamp
G) coeficient of I ri:l
friction = K I :~ rubber jacket
- rr-J-'-------"L..-_ _ --r-': rJ
I
I
I---------il
I
I
I
o-c
(~v)
not shown
/ R I
/ 21
I
r-J
I
/
I / PI / RI
,..:.J
I
I
--
I / / / ,-
!r®c1amp
Figure 5-4.Joint triaxial test with large
C I
I
///
/ /
I
I
displacement; after Rosengren (1968). I
0A N
c
and (6 )
TA c T
where
Mechanical Properties of Discontinuities 165
7TD2
P (5
{i
1 c
2
P (J D /tanlJJ
2 c
(~
D2
P
3
(5
c sinlJJ -AJ C:! a D (nu)
c
(0) (b)
(c) (d)
Figure 5-5. Controlled normal stress (a and c) and controlled normal displacement (b and d) shearing modes.
166 Mechanical Properties of Discontinuities
•
T
~
- U
~= 20 800
0'"
-0
0 ......
E 30 ~
o
Z
40
o 0.10 0.20 0.30
Shear Displacement, U (inches)
Figure 5·7. Direct shear results
for an intact specimen (1) and a
c:
specimen with an initially open ~
E-;;:; 10 ~
for the normal force or the test is con du c t ed with the b lo c k' s self
weight alone (Ruiz and Camargo , 1966 ). Shear f o rces c an be arran ge d
by jacks across a trench behind the rear face of the test bl ock. In
soft rocks such as coal , marl and soft s h a les, it may be pos s ib le to
jack into place a shear box equipp ed with cutte rs (Brawn e r et aI ,
1972) ; this provide s s up p ort for the shear bloc k duri ng i t s pre para-
tion ove rcomi ng the d amaging eff e cts of t ota l dec omp r es s ion.
Haverland and Slebir applied the n orma l force bef ore f r ee in g the side s
of the test block to avoi d total de c o mpres s ion. In situ shear tes ts
are quite expensive; because of scatter in resu l ts a few i n s i t u t e s t
results may have questionable worth unle ss supported by a l abo r atory
test program.
S ( 7)
DEFORMATIONS IN JOINTS
Normal Deformations
TABLE 5-1
Order of Magnitude of Joint Cohesion for Rock Masses
(after Hoek & Bray (1974) figure 70)
2
C (psi) C (Kg/cm )
J J
Soil < 56 < 4
o - t;
A ~v ) t (~v < Vmc ) (8)
t; ( Vmc ~v
,
"- '-
~ is the seating load. (c) Normal com-
pression of an extension fracture in a
granodiorite specimen. "-
--
/ "- ../
~
"-
/'
-
I- '-
e
(a)
0- compression
b.v
(c)
5000.-----~--·----r-----~------r---~ .-----~------~-----r------~--~
4000
.iii
-9:3000
E 2000
o
z
1000
Shear Deformations
(a)
6u
(b)
t
fj. Laboratory
o In Situ
o Model
c:
"-
If)
0..
'-
o
OJ
L
(/)
peak and residual shear strength curves can be used with acceptable
precision. In some cases, the ratio of residual to peak strength
B approaches unity as a increases.
( 1) highest (J
! 1) hi ghest (j
flu flu
(a) constant stiffness model (b) constant peak displa cemen t model
v(T)
B
6u(1 ) 6u (t )
(a)
L L L L L S= 1· L
ref--~----~----~----~----~--~----~----~
S=2 ·L
(b)
Dil
~ Roughness angle envelope
+tana
OJ lata ncy curve
3
,,
2 1.0
- tan a
Dil
41 ~
0.25r-----~.-----~rr--------,-------,-------,--------,
00
0.20~-----+~------~-------r-------+------~------~
::l
0-
"'-.
~ 0.15~------~------+-~~--~-------+------~------~
if)
if)
~
c
.9
if)
c
Q)
0.10
E
0
0.05 ~------~--~~+-----~-r~~---+------~------~
30°
0.5
20°
10°
S= n·L
E 500(mm)
20 50 100
10°
20°
0 o~#I"
'7
...t:j---
Q...~
0.5 ,1'/
30° /
/
/
40° /0
--diameter 5.5em S 1.0
-·-·-dlameter 11 em
............ diameter 21 em tan a 8a
----- diameter 42 em
(b)
LOWER HEMISPHERE
STEREOGRAPHIC PROJECTION
~1 and "'2are normal to asperity planes;
M is normal to the mean plane of the
joint. The ruled area is the roughness
figure based on apparent dips. Line N 1 '
N2 is the envelope to normals, which
has no validity here.
Figure 5-16. (a) Comparison of roughness estimate from the scatter of normals to a single rough
discontinuity Oeft figure) and actual roughness as measured by surface profiles (right figure). The
four scatter diagrams on the stereonet are envelopes to repeated measurements of attitude of a
rough joint with the compass placed on a plate of 5.5, 11, 21, or 42 cm. diameter. After Fecker
and Rengers (1971). (b) Roughness of a discontinuity formed of two sets of component planes
when profiled in different directions.
Mechanical Properties of Discontinuities 183
Most rock surfaces however are stronger when dry than when wet.
Most specimens tested by Coulson, which included granite, basalt,
gneiss, sandstone, siltstone, limestone, and dolomite, showed higher
friction after displacement of one to three centimeters, accompanied
by secondary fracture of wall rock ("Riedel shears") and formation
of gouge, figure 5-17, (see Lajtai, 1969b), particularly at normal
pressures above 500 psi. Wear of rock surfaces accompanying continued
shear displacement eventually causes the surface to be coated with
crushed material, and it becomes a new kind of specimen. In the case
of dry, unweathered rock surfaces, the new material can bring higher
friction than polished surfaces, but in moist weathered rock surfaces,
wear produces a clay film with a considerable drop in friction; for
0
example Richards (1973) found residual friction of 15 in slightly
0
weathered diorite having peak friction near 30 .
In summary, the friction properties of smooth rock surfaces vary
with micro-roughness, normal load, weathering, environmental condi-
tions, test apparatus, and testing procedure. A value for the
coefficient of friction of 0.5 to 0.6 is a reasonable assumption in
general, but values as low as 0.2 can be expected for rocks rich in
mica or other platy minerals or whose discontinuities are weathered.
Chatter marks
Figure 5-17. Characteristics of a
Compacted debris
gouge zone, artifically produced by
shearing at 1177 psi normal pressure
along sawed joints in Solenhofen
limestone; from Coulson (1972).
•
Mechanical Properties of D iscontinuities 185
- - ---'------"----- - --N
(a) (b)
the fric t ional re sis tance of rock surfaces even when the dilatancy
does not increase t h e normal stress, as discussed by Patton (1 966).
Cons ider an i d eal wedge-shaped asperity ( fi gure 5-18a) inclined i
degrees above the direction of sliding. The friction angle on the
s l idi n g surface i ts elf is ¢ , i.e.
j.l
T' N' t an ¢
)J
(9)
-
~u(tl
~u(l)
(b) shear
seams will exhibit contractancy; in the former case, once the surfaces
displace they will lock and the friction angle will be restored.
If limiting deformations govern de s ign , contractant behavior may
need to be considered. Iida and Kobayashi (1974), for example,
discuss both dilatancy and contractancy effects for computations
of stresses in dams on rock foundations.
Equation (11) can not hold at high normal st ress because the work
required to shear through asperities is less than the work to over-
r i de them. Patton (1966) found t h a t a bilinear relation (figure 5-20)
described his experimental data for shear of model joints with
regular teeth. At normal stresses less than aT
T a tan ( ¢ + i) (12a)
P ~
(12b)
T (13)
P
log(p)
0 ( 1 - a s) (v + t a n ¢w ) + as sR
T ( 14 )*
P 1 - (1 - a )
s
v t an ¢
W
+ n - 1 .l.
(1 + n o/q )2 (15)
n u
a (16a)
s
and
v tan i (16b)
o
( 17)
Filled Discontinuities
~u e/tan i (18)
c
[oj
( bJ
~u-
(3J left lateral
,--~-~[L.J left lateral shear case b;
rotational sliding [see textJ
[c 1
1969). This is because the shearing occurs at the rock wall where
pore pressures can not be dissipated and where a small displacement
creates sufficient strain to reach the residual shear strength. Since
the shear strength of filled discontinuities varies over a range from
0 0
as little as 7 , for montmorillonite fillings, to as much as 50 , for
crushed rock or preconsolidated, dry gouge, shear tests are warranted
when the strength needs to be known accurately.
I
I
I (a)
I
I b
I
I
I
I
0 a .1
!:::.v(t) dilatancy
o !:::.U [t)
(b) t
Figure 5-24. (a) Rotational friction; after Nascimento and Teixeira (1971). (b) Rotational shearing behavior
of bidilatant joints at low normal pressure; compare with figure 5-19.
Mechanical Properties of Discontinuities 195
3
2
) t an i (1 9)
n o
r
and
(2 < nr < 5)
Figure 5-25. Rupture modes observed in b iaxi al model tests with imbricated joint blocks; from Ladanvi and
Archambaul t (1972) . (a) format ion of a shear plane; () = 0; 0 3 = 28 Kg/ cm 2 (b) formation of a shear zone;
() = 30° ; 0 3 = 7 Kg/cm2 (c) formation of a kink ban d; () = 60°; 03 = 35 Kg/ cm 2
: 196 Mechanical Properties of Discontinuities
2
T 47.7 kg/cm
P
2
(J 46.9 kg/cm
Mechanical Properties of Discontinuities 197
(a) (b)
(c) (d)
Figure 5-26. Development of a k ink band; 03 = 13.5 k g/cm 2 . Fro m Ladanyi and A rchambau lt ( 1972).
Corresp onding to a =
46.9, t he peak shear st r ess wi l l be c al c u -
2
l ate d b y equa ti on s (1 4 ) to (19 ) a n d c omp ared with 4 7 . 7 kg/cm . Figure
5-27a s h ows the r e lat i ve orientations of the two joint sets an d the
mean o rientation of th e k ink band . The orientation of the blocks
r e la tive to the d i re c ti o n of s h earin g is more readil y appreciated
fr om figur e 5- 27b in which the region of the eventual kink band is
s ho wn as if i n a d i r ect shea r test; the cont in uous j oints dip i n
t h e di rec t ion of th e s he a r l o a d or in t he negat i ve d irection a ccordin g
to t h e t e rm i n ol o g y of Ha yashi a nd Kiti h ara (1 970 ). It is easy to
app r eciate why block rot ations o ccur in such a tes t. Hayashi also
198 Mechanical Properties of Discontinuities
0; 0'3
-
---
---
mean
orientation
of kink band
(a)
OJ
0-
t l ~ l l
--. _1
----
(b)
TABLE 5-2
Calculation of Peak Strength
For Kink Band of Figure 5-27
Input
0 0 0
<P11 39 , 30 , 27
2
qu 254 Kg/cm
T 28.2 (n = 9 in equation 15)
0
0
i 40 (see figure 5-27)
0
n 3
r
K2 5
IT 46.9
Intermediate Results
a 0.057
s
v 0.261
71.
52.9
48.2
(20)
a (21)
Mechanical Properties of Discon t inu i t ies 201
CJ
°1 + p
°l,a tan <P
J
(22)
and
C
J
°3 ° 3,a + P
t an <P J
° 1 + °3 °1 - °3
° c os 2 ljJ
2 2
and (2 3 )
°1 - °3
T s in 2 ljJ
2
Consider ljJ p osi tive, so l v e for K a nd in troduce tri gon o met r ic i denti-
f
ties for sin (A + B ), a nd (s in A ± si n B ) to obtain t h e condition
for limiting equi li b rium:
t an ljJ
( 25)
(° 1 compressive )
Thus we can conclude that four values of ~ (in each 180 0 range)
produce identical values of K , i.e.:
f
(26)
Equation (26) shows that the principal stress ratio to produce slip
on a joint with friction angle ~J varies markedly with the orientation
of the stresses with respect to the joint; the maximum value of Kf
(minimum strength) occurs when ~ = 45 - ~J/2.
0
This is shown in
Table 5-3, and in figures 5-28a and b, where K is plotted against
0
f
0
~ for ¢J = 20 and 40 respectively. Any set of values of 03' 01'
and ~ define a point by the polar coordinates (K = 03/01'~)'
Such a point is consistent with the criterion of slip only if it is
outside of the shaded region of the Bray diagram. During a joint
triaxial test, ~ is fixed and the test commences with K = 1. Then
as 01 - 03 is increased, K decreases and slip occurs when K = Kf .
Within the complete band - 90 < ~ < 90 0 , negative values of
K
f will be calculated from (21) when ~ > (90 - ~) or ~ < - (90 - ¢).
Since 01 is the algebraically greatest stress, with compression
positive, a negative value of K occurs when 03 is tensile. A certain
amount of tension is allowable as long as it is in a direction almost
parallel to the joint plane, i.e. 01 almost perpendicular to the
jOint plane. For a given orientation ~, slip will occur if
03 ~ 01 K f (see Table 5-3). A joint will open when cos 2~ =
(1 + K) / (1 - K). When both principal stresses are tensile, the
joint will open for all values of K.
The shear strength criterion for the rock can also be plotted
in the Bray diagram. For isotropic rock, Kf = f (03)' which can be
plotted as a series of concentric circles of radii f(03)' For example,
the Mohr-Coulomb criterion for the rock expressed in terms of princi-
pal stresses takes the form:
or
(27)
TA BLE 5-3
Li miting Principal Stress Ratio Kf = a 3 1a 1
For Slip on Jo ints in Different Orientations ( 1/J ) *
1JJ ¢: 20 ° 25 30 35 40 45 50 55 60
0 0 0 0 0 0 0 0 0 0
~~ ~\'#
1/1
\
\ CT
3
= KCT1
Mean plane
of joint
Locus
_ ton ~
Kf -tan(1/I+q,J)
Region of· ~
joint slip " '- __ 0 0 .1 0.2 0.3 0.4
I I I I I
K pas., 1/1 neg .
Scale for K = CT3 / OJ
(b)
As K is reduced at constant 1Jt:
failure IS on the joint at low 0"3 i
through the rock at high 0"3
0
-90 <1Jt<90°
ll'<K~1
Figure 5-28. (a) Bray's diagram for the Coulomb criterion for joint slip for
<l>J = 20°; 0, compressive. (b) Bray's diagram for a joint with fPJ = 40° in rock with <I> = 50°.
Mechanical Properties of Discontinui ties 205
TABLE 5-4
Kf : Radius of Rock Failure in Bray D iagram
0 3 / Qu
Kf
2
1 + (0 3/ Qu) tan ( 4 5 + ¢/2)
40 0
0
0 3 /Q u ¢ ¢ 5 0° ¢ 60
0 0 0 0
5.0
4.0
3.0
OJ -0"3
qu
2.0
0"3/Q u = 0.1
1.0
superposition can afford only a first estimate for the shear strength
anisotropy of a rock mass cut by several sets of joints, since stress
redistributions during slip introduce local block breakage as well
as rotation and buckling. However, a closer approximation to the
real behavior of multi-jointed rock systems is possible with physical
model and finite element techniques, to be discussed later.
Although the discussion of strength anisotropy caused by a set
of joints has been two dimensional, it must be appreciated that a
peak stress criterion like (20) is truly three dimensional; for in
place of (23) we can introduce the three principal stresses in (20 )
through:
12 + 2 2
m + n
° °1 °2 °3
and (28)
2 m2 2 n 2 ) - 02
T j(012 12 + +
°
2 °3
Mechanical Properties of Discontinuities 207
N
peak strength
t
criterion for joi nt
o 03 c
A ,...
':~ ~ ~~
~ liJ:N i
50 L~ pn ~
r\. 1~'"
a rn
40 ~ K
I~
B
20
o 20 40
IV
50
Figure 5-30. Application of Mohr's circles to test for slip on joints of varying
orientation according to a given peak strength criterion.
where 1, m, and n are the cosines of the angles the normal to the
joint plane makes with the directions of 01' 02' and 03 respectively*
(figure 6-37b). A quick appreciation can be gained using the three
dimensional Mohr's circle diagram, figure 5-30. Given the directions
and magnitudes of 01' 02' and 03' a joint whose normal has direction
cosines 1, m, and n has normal and maximum shear stress defined by
the coordinates of point P determined as follows. From 02 lay
-1 1
off angles ~ = cos (n) and a = cos- (1) as shown in figure 5-30
defining points E and F. P is at the intersection of an arc through E
with center D and an arc through F with center C, where D =
1/2 (01 + 02) and C = 1/2 (02 + 03). If P defines a point in the
-1
In the two dimensional case considered, ~ corresponds to cos en).
*
208 Mechanical Properties of Discontinuities
ruled area, the stress and s.trength data are incompatible, i.e. slip
occurs. Using this construction, the locus AB of the peak strength
criterion across the Mohr's circle can be transformed to cartesian axes
(~, a) and the zone of safe joint orientations the Coulomb criterion,
were presented by Jaeger and Rosengren (1969). These show that the
relative magnitude of 02 exerts a significant influence on the relative
orientations of safe joints.
6
app Ica ions of
stereographic pro·ec ·0
in mechanics of
discon i 0 s rocks
INTRODUCT ION
KINEMATICAL CONSIDERATIONS
____CJ
L
cube regutar prism tapered prism sliver
L 7
flClt slab curved plat e tetrahedron
[] ~
~
hexagonal column square column incomplete tetrahedron incomplete pris m
contact
( a)
joints
;
.I .I
shear ./
"'"
,; v
I
I
- r \
"
contact ' "\.
, -" "
,
/ - I ...... ,
.t
'" , - .; '- I
\.
t \
\
, I J
"..... - \
(c) (d)
Figure 6-2. Examples of multiple blocks (a and c) and single blocks (b and d).
( 0)
on intersection slide in 0 ( b)
rock cut region of dayl ight in g intersect ions
cut slope 5
(a) ( b)
LH LH
Figure 6-4. Simple tests for sliding along an intersection of discontinuities under
gravity loading only; after Hoek and Bray (1974).
(a) superposition of friction and daylighting tests;
(b) application of kinematic tests to a cut (S) in a rock mass.
possibilities exist.
Another construction can test the case of sliding on a single
plane, under gravity alone, because in this case the block must
translate down the dip of the plane (figur.e 6-5a). As shown in
figure 6-5b, for this mode to exist, the true dip must be contained
5 ign conventIOn
i '2 d efined as lower hemisphere Intersection of planes 1 and 2
j32 defined as the Intersection of planes 3 and 2 havln9 a bearing
___---,r--w-it_hin the half circle of the bearings of -IJ5,Ds,Ic...
3s_ . - - -_ _
Kinematics
test 1 O2daylights and
steeper than 'f'
IS
Figure 6-6. Kinematic tests for wedge sliding on a single plane under gravity loadi ng only (see f igu re 6-5) .
between the two intersection lines limiting the e xtent of the pl ane
of sliding. A line of intersection will be considered p o siti ve if
it daylights into an excavation; figure 6-6 shows how to identify
the positive direction of an intersection for the example of '"132 i n
figure 6-5a. Form the intersection of plane 3 and the cut ting S;
the dip of the cutting is denoted DS' 132 is defined t h e n as the
intersection of planes 3 and 2 directed to have a be ar i ng within the
half circle formed by the lines 13s , US' -1 3s . The k i n e ma tic require-
ments for sliding down the dip of plane 2, for example, may then b e
stated in terms of the following two tests, both of wh i ch must be
passed in order that a slide occur. (1) D2 dayl ights and i s steeper
than the friction angle ~ and (2) D2 is in the small er angle between
112 and 132 , Figure 6-6a shows an example in which the di p vector
of plane 2 passes the second test but fails the first te st and
therefore cannot slide. In the same example the d i p vector o n plane
I fails both tests and therefore cannot slide. I n fig ure 6-6b, a n
example is given in which the dip vector on plane 2 passes both te st s
216 Applications of Stereographic Projection
( a) A blocked intersecti on
, --
~
"-4---
~
(2,
Tapered with respect to faces 1,2,3 Tapered with respect to faces 3,4
and therefore can slide. In this same example, fi l , the dip of plane
1, again fails both tests.
The above kinematic tests are useful for examination of large
numbers of blocks. When a single block is under study, further tests
can be introduced by tracing the relative locations of the normals
to the supporting planes. A block will be called tapered if it grows
wider as one traces its extent behind the excavation. Such a block
cannot by itself slide forward into an excavation. Figure 6-7b gives
four examples of tapered blocks. Figure Sa shows a five-sided block
with one free surface denoted S, which is tapered with respect to
faces 3 and 4. Such a block is tapered if the normals to opposing
faces, in this case 3 and 4, make an angle of less than lSOo, as
measured in the half space outside of the rock mass. Figure 6-Sb
shows how this test can be performed on the stereonet. Since n4 is
Applications of Stereographic Projection 217
Figure 6-9. The Cross product of two vectors; sign convention-lower hemisphere.
,..
b
FIND A + B and A - B
Notation:
~ ~ /\ /\
A
~
= IAla = Aa
~ /\ /\
B = IBlb = Bb
~ ~ /\
A + B = Rl rl
~ ~ /\
A - B = R2 r2
Given: ~
A = 40 in direction
30° up to West
~
B = 20 in direction
70° up to South
- B B= 20
U.H.
Notation :
A = Ao etc
A +B = Rob rob
-
A+ C Rae fcac
R = Robe r obe
r--.
a b means plane co mmon to a and b
C with
~
the resultant in the direction r
~ac
. Adding Ra c to
~
B fixes
rabc along the great circle common to rac a nd B. Th us rabc must
lie in the intersection of the great circles common to r ab and c,
and r a n d b (figur e 6-12). A form u l a for this cons tr uc t io n can be
ac ~
written by introduc ing notation as follows: le t PQ mean the g reat
circle common to lines P and Q; and let nJ\m signify t he line of
intersection of planes nand m. Then the unit vect or r b de fi n ing
a c
the direction of A + E + C is
~/\~
r ab c r ac 0 (1)
Given:
-+
A = Aa
b,2
Find: "/
/
-+
75° 1.5 /
D = Aa" + kb b" + kc "C 1.25 /
for all values of kb and k c ' k IA
b .75
to /
Find:
--+
Components of R = Rf
in basis directions bi ' b2 , b3
Solution:
BI , B2 , B3
3 12 12 12
knowing the angles 8 and 8 , construct a triangle of for ce s to
-+ -+
3 12 " -+
decompose R into B 3 , (the component in " the di~ect i o n b 3 ) a n d R 12 ,
(the component in the plane common to b and b ). Then reading the
/"\, /\. /"\ l 2 +
angles a 1 and a 2 from r to b and b r es p ect i ve ly, dec omp o se R12
-+
12 -+ 1 2 -+ -+
into the components Bl and B2 , as shown in figure 6 - 14. Bl , B2 and
-+
B3 are the required components.
224 Applications of Stereographlc Projection
(15 oj
c
r
/
/
----------:::;oo~ :--\---100m /
edge of 2 /1\,
/ /. . . 1,2
115 ........ 2LgO 11$ /
"'" 38°..... 30 m
B~~------------~r-----~p-90m // section q, q'2
_----=:~---_-___,__---7~_12.L 80m ~'f
--~~,-----------+-~""---- \ 70 m
------~~------ 60m
A
L.. q
2
IABI = 30/si n 29° = 61 · 8
IACI = 42/sin 40° = 65'4
o 20 40 IADI = 30/sln 38° = 48·8
(15 b)
Applications of Stereographic Projection 225
of the vectors AB, AC and AD can be computed f rom the scaled lengths
on the dra wings and from the kn own plunge a ngles listed in figure
6 - l5b. To check th e accura cy of t h e construction, o n e can also
determine po int C b y constructing a vertical section in plane q
perpendicu lar to the c u t as sh own in the right half of figure 6 - 15b.
In this secti on one does n ot see the true p lun ge of t he line of
A 1
inte r section 112 b ut rathe r it s ort ho g raph ic pro jection 112 in th e
pl an e of the secti o n (the orth og raphic proje c tion of a line in a
pl a ne was shown i n figu r e 3-1 2) .
226 Applications of Stereographic Projection
./"-...
1\ 1\
ICAxDA)
0-. ....
.... .... _Or. ,
-<.::. ......60 0
'-'-, 1'5
'f-> (B"A)
I
I
I
/
/
/
I
o /
4\\0-
/
12s //
-~_~~_2_4~~ __
(DA) 112
((A)
(a)
We will need to determine the volume of the wedge ABCD and the
area of its contacting faces ABC and ACD. The volume of a tetra-
hedron can be calculated with the scalar triple product of any three
~
edges radiating from a given corner. Since we know the vectors AB,
-----4- ~
1 ~ ----+ .~
v 6" AB AC x AD (2)
North
Centroid
j~s 1,'2
DA 1'\
CA
(b)
L.H.
-+ 1
AM (3 )
4"
hedron.
-
where AM is the vector from corner A to the centroid of the tetra-
The sum of three vectors was discussed previously. In
figure 6-16b, the summation is done by a repeated operation of the
summation of two vectors; the great circle method discussed earlier
for summation of three vectors is not necessary here since all three
vectors are known absolutely.
-+
OC n n (4)
p p
--+
Line of action of force F is defined by the locus o
--+ --+
of arrows of the family of vectors OA + KF
where 0 is a known referen ce point and A is any
--+
point on the line of action of force F , (for example
its point of application).
o
(K= 0)
( b)
Figure 6-17. Position vectors defining the line of act ion of a force and the equati on of a plane.
Figure 180
n 1 are read from the stereonet and plotted as shown in the upper
right of figure 6-18b. The dot product of two vectors is a scalar
quantity; it is shown by a circle of appropriate radius about the
point D in the lower left of figure 6-1Sb. In this example, the
value DA . ~ is equal to 10.2. For a value of K = 0 we can draw
the vector DM as shown in the lower left of figure 6-18b and form its
dot product so as to produce a value of 10.2 by drawing a tangent to
Applications of Stereographic Projection 23 1
I
ygon
"- ~ Fore e pel
~
""
I
r--!-
I ..........
I
I
10(
I'--- from st ereonet
I ' 5 (fig 180)
DA = 1.9 DA 1 I
o 10 20 30 40 50
0< ,- Angle along D~M-Z
ANALYSIS OF ROlATI ON
Figure 6-19. (a) Rotation in a plane. (b) Analysis of rotation in a plane (see figure 6-19a).
(c) Stereographic projection for analysis of rotation.
(a)
----------05
//8 ,
plane 2
MR = To • IATI
~ >90
pI Stability requires MR > MA
A pplications of Stereograph ic Projection 233
from the tail of the vector dA . We will examine the moment for the
rotational sliding mode of figure 6-19a, that is rotation about an
axis (~p) perpendicular to plane p through point A. Force R pierces
plane p at point T. The moment about the axis shown, MA , is equal to
"AT x R . ~p' If np is pointed into the lower hemisphere, as in this
example, then a negative moment corresponds to kinematically possible
rotation. If we view the plane of rotational sliding as in figure
6-19b, it will be seen that rotation about A is possible only if the
angle n shown is greater than 90 0 ; n is the angle between the inter-
A A
* Personal communication.
(a) North c
t
LH
(b)
r from A a t which 1+Nl + +N2 I acts with the same moment as the sum of
+ +
Nl and N2 ·
r (6 )
T ( 7)
o
(a)
L.H.
The friction cone about the normal to Stereographic projection of friction
an inclined plane surface of sliding. cone producing a "friction circle"
If the block is at rest. the resultant
force between block and plane is
inside the cone tp is the "friction angle"
safe zone.
anisotropic
friction
roughness
'zone'
LH LH
Figure 6-22. (a) The friction circle concept. (b) Generalized friction circle for rough planes.
.p (Pl
o~--------------------~
99%
.. p
Figure 6-23. The relationship between the friction angle to be used with the mean
joint orientation and the probability of occurrence P.
240 Applications of Stereographic Projection
A sa f ety f acto r of 1 .0 c o rres ponds to the con d i t ion where the r es ul-
tant f o rce plots on t h e ci rc umfe ren ce o f t he fric ti on circle having
0
radius ¢ = 30 . To achi eve th is condi t ion, a f orce can be added to
Wto inc l i ne t h e r esu lt ant 20 0 f r om ve rtical. The magnitude of the
r ock bolt forc e re q uired to do t h i s depends upon the orientation of
the bo lt s. The r e quire d force is sm a ll e st if i t i s directed per-
pendic u la r ly to the re sultan t, t hat is, 20° abo ve the horizontal as
measure d a long t he d i r e cti on of t h e bearing of np (figure 6-24b).
Since t h is direction is in t h e upper h e mis phere, its opposite
( - B l . O) has b een s hown .
Say a fa cto r of sa fe ty of 1 .7 is des i r e d. This means that the
require d f r i c t i o n angl e (<P r eq ulr
. e d) is equal to arc tan
0
(tan <Pavai l ab l e / 1 .7) or 1 8 .5 • To achieve thi s th e rock bolt force
0
must inc line t he re su ltant 31.5 ab ove t he vert ica l alon g the
di rect ion of beari ng ~.
p
The p o lygon of f orces pres e nted in figure
6- 24b, l e ft, determines t he r e qu ired r ock bo lt forces as 1 ,400 tons
a n d 2, 100 t on s r e spe cti ve ly f o r f ac t ors of s a f e ty of 1 .0 and 1.7.
242 Applications of Stereographic Projection
(a) N
_ 0
..,.
W ~(J
00-899%
0~B1.7
-B
10
Opposites to bolting directions
. also opposite to worst acceleration
direction
Figure 6-24. (al Example
of stability analysis.
(b) Completion of stability
analysis.
Though this is the minimum force, and therefore minimizes the required
cross-sectional area of steel, it may not be the most economical
direction in which to emplace rock bolts in a particular problem;
the geometry of the block may dictate other directions for minimum
length of drilling necessary to seat the anchorage of the rock bolts
well into the support. Every specific case will require the determin-
ation of an orientation optimizing the economic factors related to
the area and expense of steel on the one hand, and the length and
cost of drilling on the other.
Now let us design the rock bolts for a given probability of
safety. To design for a probability of safety (PS) equal to 0.99,
for example, compute Wcorresponding to P = 0.98. With K = 120,
equation 2 in chapter 3 will give W(.98) equal to ISo; that is, 98%
Applications of Stereographic Projection 243
(b)
Vi
W = 4000 tons
A = 200 m2
I .. ,
100 0 tons
-
Ground acceleratio n :Xg Tot al displacement of block
Base 1 2
di splacemen t Xb = "2 Ky g tm
Ag 1 - - -- - .....
-
~
aL...
de termi ned fro m
stereo net (see f ig .6 - 22 b)
!1X =-"2 Ag to + Ag to tm - "2 Ky 9 t m
Q)
u
u
a Relative moti on between block and support
t ime ceases at time t m
Kyg tm = A 9 to
tm = ~ or V
Ky Kyg
A 9 to: V and tm= _ V
_
Ky 9
time
Compare /).X to peak displacement in a shear test
motion may come and go before the block has travelled very far, as
discussed by Goodman and Seed (1966). Hendron et al (1971) proposed
a simple method of calculating the relative movement of a rock block
once the value of its yield acceleration K g has been determined.
y
The ground acceleration and its duration, or the ground acceleration
and the particle velocity, must be known. Suppose, as in figure
6-25, the ground acceleration rises instantaneously to a value of
Ag, remains constant for time to and then drops instantaneously to
zero. The block begins to yield when the ground acceleration reaches
Ky g, and acquires velocity Ky gt ' Relative displacement between the
O
block and the ground (slip) occurs until time tm when the block ac-
quires a velocity equal to the ground velocity V = Agt O;
V
tm K g
Y
x
b
= (1/2) Ky gt m2
6.x
6.x
~: (iy - i ) (10)
R = R, +R2
= (N,n, + T, I'1r-( N2n
1 + T2 I~ 'l.)
248 Applications of Stereographic Projection
slides on
plane 2
L.H
Figure 6-27. Generalized friction circle for sliding of a wedge with two faces in contact.
Applica tions of Stereographic Projection 249
",,...--.....,,,
112 n 1 bounds t he re g i o n of kinematically poss ib le sl ip on plane 1.
A~
Similar ly th e gre at circ le 112 n 2 bounds the k i ne matically me an ingf ul
portion of the f riction circle for slip on pl an e 2 alone (fi gu re
6-27) , Between these great circ les the r e is a regi on in whi ch slip
is pos sible a long th e l i ne of int e rs e c t i on; in t hi s case, the direc-
t ion of in ci pien t s li d in g is parall el to the l in e of int e rs e c t ion
an d t he re fore the sum of the shea r f orces on each plane , at the l imi t
of slip, is p aral lel to '"112'* In a dd ition there are fo rces paral lel
to the two normals. The r e sultant f orce of the blo ck on its s up p orts,
in t he case of sliding d own the l ine of i n terse c tion , must t he ref ore
'" A A
b e composed of c omponent s paral lel to n 1 , I , a nd n and it will
12 2
acc ordingly pl o t i n the spher ic al t riang l e fo rmed by
1
, 1 , an d
12
n
n . If it is possible to slide up the line of i nt e rs ecti o n , the
2
re su ltant f or c e will h ave to lie in th e s phe rical tr i a ngle b e tween
n1 , -1 12 a nd ~ 2 (f igure 6 -2 7 ) .
At the l imit of e qu ilib ri um in the i nt e rs e cti o n mode, ful l
friction has been mob i lized in both pl anes simul t an e ously. The
re action in plane 1 is ma de up of a n ormal fo rce an d a s hear force
A 0 '"
paralle l to 112 a nd is ¢1 from n 1 . Th e r efo re t he r e a c tion in plane
1 lie s at the i ntersection of the f ri ction circle for plane 1 with
A ~'"
the great cir c le 112 n 1 · Si milar ly the r e action in plane 2 i s made
A
to the position of the resultant not only the degree of safety, but
the mode of potential slip is indicated.
(a) North
o
111
-8
(b) -N
Factor of saf et y on plane I
9- I"- <.D N N m 0 <;;t- o
II (J)
~ ~ ~ LO ~ N 0
--- - - - -
25 1.0
E I
::l
~
..0
=-= 20
V
I
FS 1 0 FS 2 0 1.85
1.12
1. 28 C\.l
::J
Q)
0-
Q) I 1,48
c
0
~ I 0.
E I c
15 1.74 0
10- ---------- >.
0 ~
'+-
"'0
2 .10 2
0
Q) (j)
1-
- 10
::J 2 .6 4 '0
0- 1-
~ 3 .57 .2
(,)
N
9- a
--
LL
0
5
a>
::J
0
>
0
0 5 10 15 20 25 30 35 ( = CPI)
Value of tpl required for limit equilibrium
(c)
o
o o
o o
II
I(TI + C + T2 + C2 ) 1/ ITI =
1 1.3
B=1 4
252 Applications of Stereographic Projection
/\
(b)
80
N
.....-
'0 60
c
9(/)
'-
(l)
0...
(/) 40
a
><
0
E
(l)
20
i
----!
0
0 20 40 60 80
T} - Angle betwee n no r mals 11 " n2 { Ram say - 1967, p. 14 1
254 Applications of Stereographic Projection
Figure 6-30. (a) A dayJighting intersection between a bedding fault and jointing
at Libby Dam, Montana. Photo by Dennis Lache l; (courtesy of the Corps of
Engineers). (b) A portion of the left abutment slide at Libby Dam after removal
of the slide debri s. T he bedding, left. strikes N 26 Wand dips 43 -5 to the west.
The joi nt su rf ace (right) st ikes N 74 E and dips 49 to the NW. Figure 6-30a was
t aken just t o the left of this view. Compare with figure 2-8d and e. Photo by
Dennis Lachel ; (courtesy of the Corps of Engineers). (c) View from the top of
the wedge of 6-30b showing the relatio nship of the slide t o the dam and reservoir.
Photo by Dennis Lachel (courtesy of the Corps of Engineers) .
TABLE 6-1
0 none none
A
nl '"
n
A
2 n3
A
I 2,3 I n 1 n3
2 23
2 1,3 2 i\ 113 A3
TABLE 6-2
nl west 0° Both
n3 - vertica I Lower
(a)
7P®
Q)ho ,@ horizontal
Face Strike Dip Friction angle Figure 6-32. (a) Example problem of a wedge
I North vertical 30° supported on three faces. (b) Lower hemi-
2 East 70° 50° sphere portion of the generalized friction
3 - HOrizontal 20° circle (the safe zone) for the example problem.
Applications of Stereographic Projection 259
NORT H
(b)
nl~____________~______+-__~
(c) NORTH
U.H.
-w
W= 10,000 tons
jection and dr aw the great circle -a b; this great circle, in the mode
....
1 region, intersects the primitive circle at point d, wh ich can b e
transferred to the lower hemisphere. The required great ci r cle
construction is completed by connecting d and a.
The completed c o nstruction shows that the orientation of the
resu l t ant force, R, is well within the safe zone. If t he wat er
pressur e is raised on face 1, a water force on the we dge wi ll poin t
in the direction of the opposite to ~l' (-~l = 123 in t h i s particu la r
problem). Along the great circle a:'ft a 14 0 rotation is requ i red to
bring the resultant force to the l i mit of stability. R, found by
addin ~ the weight force (in direction ~~) and :he thr ust of t he arch
d a m (Q) i n the di re cti on given, lies 81 fr om U1 . The polygon of
forces o f fi g u r e 6-32d shows that the required water force to initi-
ate sl ip i s 5,300 t o n s and failure will be by mode 1. Ro t at ions
have not been considered; in any p r a ctical example one cou ld u s e
the methods discussed p r eviously for detailed analysis of rot a t i ons .
created when the passive wedge moves. Figure 6-34 shows a simple
analysis for the mechanics of this case. The principal idea is to
separate the blocks along their common plane (plane 3 in this example)
which is considered as a free surface in each analysis of each block
separately. As opposed to the case in soil mechanics where the
division between the active and passive wedge has to be determined
by iterations, structural geological information determines the
direction of the interface between blocks. In contrast with soil
mechanics, there should be no debate about the proper angle of
friction for the inclination of the load transferred along the con-
necting plane, since a very small displacement is enough to mobilize
Applications of Stereographic Projection 263
ACTIVE B LO CK (A)
PASSIVE
BLOCK (P) cross secti on of blo cks
in direction of virtual
motion
Analysis:
I. Analyze acti ve block wit h
plane 3 as a f ree f ace .
Find Fp
requ ired.
2. Ana lyze passive block with
pl ane 3 as a f re e f ace ,
and with addi t iona l loa d
- Fp .
3. Safe if re sulta nt on passive
bloc k is in safe zone .
LH LH
NORTH
No.
I12 of the acti ve block meets the face o f the pass i v e b l ock . If q
is inside the face of t h is block, then movement of the passive b lock
will al l ow da y light ing of the active wedge, a nece ssary condition
for t h e t wo bl o ck slide. If t h e pas sive wedge sl ide s on a s i n gle
plane , the d i rect i on of relative s l i p on plane 3 wi ll have to be
determine d b y ite rati o n. If on the othe r h an d, b o t h t he active
an d p a ss i ve wedges move a l o ng t he ir li n e o f in te rs e c t i o n s , there is
o n ly one possib le di rection for the r el a ti ve sl i p in plane 3, a s wil l
b e shown.
Si n c e at first t h e critic a l mo de s are n ot known, an i terat ive
s o luti on will b e i l l u s i rated. In f i g u re 6- 36b , t he active wedge i s
diagramme d . We will s up p ose we do n o t kn o w t h e di r e ction of slip
on pl a ne 3 an d will determin e a n umb e r o f solut io n s f o r diff e rent
266 Applications of Stereographic Projection
A
For intersection sliding of active wedge, 112 must
(a) intersect face pqr of passive wedge.
Figure 6-36. (a) A two wedge slide. (b) Analysis of the active wedge. (c) Analysis of
the passive wedge. (d) Force polygons for analysis of the two wedge case. (e) Construc-
tion for compatible velocities of neighboring blocks, each of which slides on its line
of intersection.
possibilities. First construct the safe zone for the active wedge;
then around the normal n 3' construct a small circle of radius
equal to ~3' Points a, E: c, a, e and f are various possibilities
for the direction of the load transference between the active and
the passive wedge. Corresponding to each we can read the required
rotation a of the weight force to produce a resultant in the safe
zone of the active wedge, as tabled in figure 6-36b. Suppose ¢
App l ications of Stereographic Projec tion 267
NORTH
A
Locus of fp
ACTIVE WEDGE
a 95 16°
b 110 15 .5°
c 114 16°
d 114 16°
e 110 17 °
f 97 ISo
0
equals 20 o n all plane s ; what an gl e of f r icti on is necessary for
stab i li ty in the pas si ve wedge ?
Sin c e the ang le b e t ween the direc-
"
tio n of fp and the weight fo rce of t he active wedge i s known f or
each as sumption of di rection fp , a f o rce p ol y g o n c an be cons tr ucted,
a s shown in fi gure 6 -36d , defini n g t he magn it ude of
" A A
FP corresponding
t o each a ss umption for f (a thr ough f ) . Tr an s fe r t he opposites to
p
p o i nts a th ro ugh f onto the st e reogr a p h ic p rojection f or the passive
we dg e ( fi gure 6 -36c ) as we ll as t he opp o s i te to
~ p,
3 (point
a,
3).n n
Read the a ng l e t o Wp f r om each point jus t p l otted and construct a
s er i e s of fo rce p olygon s f o r e a c h point to determine t he rotation
of W
p
p r o duce d by each assump t ion of s lip di r ect ion in plane 3.
This h as been d one ( fi gure 36d ) for two a s sumptions o f the weight
268 Applications of Stereographic Projection
(c) NORTH
PASSIVE WEDGE
Wp = WA and Wp = WA I2
WP=WA 1.
W =W 12 (~sultant
p:c\ ) \
'y--Plane 3
"~o \
"- '-. A \ \Direction 0 slip in 3
~~5 ,~ [I~ Aplane 3]
--- 29 0
- - .----.------,-----.-------
A
fp
Angle from
Af A
IFp I Inclination
of resultant
- p to wp
wp= WA wp = W,/2
(d) (e)
The i nt ers e ct ion of the l a tter with the f ri ction circle abo u t n
A a, 3
def i nes poi n t n wh ich is the direction of the resul tan t load t r ansfer
fr om the active to t h e passive wedge. (The direction ~ correspon ds
to the d irecti on of slip in plane 3.) The wedge is un safe. An
e xp a nsion o f the fr ict ion circle about ;i 4 to 37 degre e s will p r o duce
limiti n g eq u i librium.
Fi gure 6- 36 e shows the vector triang le, used t o construc t t he
270 Applications of Stereographic Projection
Rocks differ from most other materials in that they may already
possess significant stresses before additional loadings or un loadings
are constructed. Excavation of surface or underground space leads
to stress rearrangement with concentration or spreading of the lines
of force, which, to the extent that the rock properties vary with
stress, render the rock mass non-homogeneous. If stress concentra-
tions cause extreme stress differences, the rock may break, sometimes
explosively as in deep mines in South Africa and Canada. On the
other hand, if stress reductions leave blocks of rock almost free
from their neighbors, they may fallout, as in the roof of a tunnel
at shallow depth and in the exposed corners of excavated rock walls
in complex underground openings like underground power plants (see
figure 1-5b). Since properties of joints are highly stress dependent,
a jointed rock mass with a non-homogeneous stress field will change
in character from point to point. In such rocks, we are particularly
concerned with the real possibility that a state of stress may imply
sliding or opening of individual planes. Whether or not such move-
ments will provoke rock falls depends upon the direction of the
sliding or opening tendency with respect to the kinematic freedoms
of the rock mass. Given a state of stress, we will show how to
calculate the normal and shear stresses local to a given discontin-
Applications of Stereographic Projection 271
Stress
aX T
yx T
zx
{a } TaT (11 )
xyz yx y yz
T T (J
ZX yz z
(b) NORTH
x' plane
area A
,-
,-
/
./ ""
x' plane ',\,/
/
/
/
/
/
/
/
/
/
/
",y./ I
/
I
I "y
I
I
/
I
I
I
I
I
I
I
I
I
I
I
,
1
I x y
x' 65° 42° 60° N
Xl
(c) z
x', y~ 1.' {
components
of Px'
x' plane
(12 )
(1 3)
274 Applications of Stereographic Projection
where
(T)
first we find
.422 .743
.500)
(T) -.242 -.438 .866
(
.866 -.500 o
By equation (13)
The s tereograph i c projection shows how the stresses are ori e nted on
the plane o f i nterest .
Consider the x, plane of figure 6-37b, whose direction cosines
were previously obta i ned . The x , y, and z components of the traction
~ can b e c omputed fro m e quat i on (12)
I gi v ing f or the example
x
computation
(P jP ! ) and a
I = tan (P IP . ). Then the direction of P r
I
X y X X xz x z x'X ,('""-A ~ x
is ~ located a t the i nt e r s ectio n o f gre a t circles A z and B ~
I
0
24 = 2.8. The s igns on the ste r eonet can be inte r preted as follows.
Px , produces compress ion i f i t mak e s an angle less than 90 o with x" ',
which must be visualized as the outward normal to th e wedge. Sliding
of the contiguo u s blo ck wil l tend to occur down the face of the wedge
if TX :
, max is in th e upper hemisph e re.
CONCLUSION
A
A
Upper Hemisphere
Direction of forces of
stresses on x' plane
KINEMATIC MODELS
.
pi
0 tR ~
~
0
- - ~
+
- CI\
- - -
~
'"
'"
'" !?
\!l
'" '"
; ~
~ -
::a
.,
Rig'" Toe _
Rlglil Toe
LeO Toe
0 .,"' .. ":'
~
:B
'" ~
tinuities may be pressed or cut . Each new cut c ompresses the mode l
by the width of the blade so tha t it is possible to produce a variety
of Jointing styles -- open or closed, planar or imbricated -- by
programming the cutting sequence . According to the oi l content, the
mixture may be made relatively plastic or brittle J and heterogeneous
structures can be modelled readily.
The main requirements for such a study concern cost and facility,
s i nce it is by repetitions and parameter variations that the full
(a) (b)
(c)
Figure 7-2. Procedure for conducting a base shear kinematic model study of a gravity loaded rock structure.
(a) Using a trowel , smooth out a sheet of model material
(b) Give the sheet a push to break the bond along its base.
(e) Cut the outline of the excavation (in this example, a rock cut with benches) and the system of
discontinuities.
(d) Push the model against the direction of gravity. The slope is failing by toppling, with the lower limit of
toppling defined by the discontinuity inclined towards the free surface .
280 Physical Models
JOi n t or
Fa u lt Surfa ce Mode l ing
Mal er lo f
Figure 7-3. (a) K inem atic model machine used by Go lder-Brawner Association to
stu dy strata movement in underground m in ing ventures. (b) Kinematic modelling
mach ine in th e rock mechanics laboratory at the University of California, Berkeley ,
( 1)
v u
sin(90 + <1» sin(i - <1»
or (2 )
v sin(i - <1»
u v ( sin i-cos i tan ¢)
cos <1>
Constant
rotation
constan~
velocity/'1
~
I Gravity
ecce!. No
ho,.izontal
velocity
No
rotation
Figure 7·5. Dissimilarity of base friction and gravity loaded models where momentum
is not negligible; after Dr. John Bray.
(c)
(a)
(b)
Figure 7-6. Kinematic base shear models of rock cuts with two systems of discontinuities.
(a) A cut with horizontal benches, before pushing.
(b) The same cut as (a) after pushing. Sliding of a wedge occurs, causing overturning of
loose blocks under the benches. Note the caves between joints inside the loosened
rock mass. If the wedge were to stop moving, for some reason, these caves would
be preserved as evidence of the previous slide movement.
(c) Another wedge slide in development, causing caves to from between joint blocks.
Physical Models 285
(a) (b)
(
(c) (d)
Figure 7·7. Kinematic base shear model showing the effect of excavation shape on the
deformation mode .
(a) Flat roofed opening with hor izontal bedding before "turning on gravity" by pushing
the model.
(b) Classical symmetrical flexure pattern ; the upper lines of flexural cracks extend
through eight layers, with tolerable bending in the layers above.
(c) Further propagation of cracks and the beginning of a fall from the roof.
(d) An arch shaped roof of the same span; stable.
286 Phvsical Models
(a) (b)
(c) (d)
Figure 7·8. Kinematic models of earth dams o n bedded and jointed foundations.
(a) Horizo ntal beds in the foundati on close as the dam is pushed in the direction of
the load tra nsmitted t o t he fo u ndation by the dam and reservoir. T ension cracks
form in the upstream port ion of the dam fo u ndation a nd the differential defor·
mation of the foundati o n causes crac ki ng in the embankment.
(b) A similar case, except t hat vertical jo int s in the upper part of the foundation
accommodat e the down stream defo rmat ion with o ut new tension crac king. The
dam does not crac k, perhaps because the add itional joints permit shear deforma-
tio n in the foundation which red uces the shear stress at the base of the embank·
ment. Some rotation of blocks occurs in the upstream part of the foundation.
(c) A rock-f ill da m wi t h a shea r key o n a roack mass w ith beds dipping steeply
downstream .
(d) Case (c) after push ing the dam in t he d irecti o n of the load transmi tted to the
foundation by the dam and reservoir. F lexure of t he layers in t he foundation
has occurred, lifting the d am in the downst ream part of the toe trench and in
the downstream shell as the embank ment rotates a bout its lower m id dle region.
Physical Models 287
Figure 7-9. (a) A kinematic model of an arch dam in a canyon with vertical beds
striking into the left abutment and slightly downstream. Initial condition; the arch
is made of material stiffer than the model material. (b) First stage of deformation
as the arch is pushed downstream . Shear failure of the edge of one layer has taken
place on the left abutment and flexure of the downstream layers has initiated. On
the right side, a tension crack has severed the layer under the thrust of the arch.
(c) Continued pushing of the dam downstream has increased the flexure and
initiated flexural cracking on the left abutment, while sliding of a wedge defined
by the bedding and tension cracks occurs on the right abutment. (d) The dam
swings around an axis on the right side as the flexural cracks propagate on the left side.
(a) (b)
(c) (d)
288 Physical M odels
p ( 5)
(6 )
a n d t he scal e of t ime
T ( 7)
sand and water; san d, wax, and mica; sand and clay; and plaste r
alone or mixed with sand, clay, mica, barite, lead oxide, d iatomit e ,
s awdust, or l i me. Sand and wa x mixtures tend to be plastic wh ile
p las ter or cement s and mixt u res are brittl e . Strength, def ormabil ity ,
a n d u n it weight can be va r ied ove r wi de ranges by controlling c uring
an d mi xi ng time, and mixing additional materials. For example, sand
plaster mixes are weakened by curing at 90 degrees C, or by addi n g
crushed mica; they are made more brittle with chalk or clay a dm i xtur e;
a nd adding powdered lead or barite raises their specific gravity.
Joints and other discontinuities have been simulated in man y
ways -- as rough extension fractures caused by b reaking the soli d
modelling material, as untreated or varnished saw cuts in the s olid ,
an d as thin partings of wax, grease, talc, limestone dust, wax paper ,
oiled tracing paper, lens tissue, g r aphite, and clay. It is e asy to
duplicate t he low friction and re a dy parting of important se ams a nd
faults, e.g., by grease partings, b u t it is harder to raise the
0
friction angle above 4 0 and to control dil a tancy. Krsma n o vic and
Associates developed the technique of imbricati ng joint blocks, i.e.,
offsetting blocks slightly to produce block interlocking. The
resulting joints show high strength and dilatancy at low normal
pressures but they are highly directional i n their shear be havi or.
Barton (1972) developed an indirect tension splitting device (te rmed
a "guillotine") to introduce rough extension fractures in cast pl a tes
o f model mate ri al. In contrast to cut joints, the oldest set of
s uch split discontinuities is the only one which is planar, sin c e
all younger discontinuities are offset wherever they cross t h e
earlier sets.
A simple, plane stress model study of a hypothetical un de rg round
power station has been used for several years as a class exerci se
in the rock mechanics course at Berkeley. The model, at 1:150 len gth
scale, is prepared in about one half day by placing hot mixtures of
sand, crushed mica, an d paraffin (2-4%) according to the desired
lithologic section. The model apparatus consists of a wooden b o x
o f ins i de dimensions 30 inches wi de by 25 inches high b y 4 i n ch es
t h ick (f i gure 7-10). To construct the model, the cross section i s
drawn to scale and attached to the plywood front piece (figure 7-11a).
i+i
tl i+i
rl
i-o-i
ll
l'*i
s
o
e
3
tl
!qaFl i+i i+i ;qF'
o
I
o
l@i
:r
@i i+i i*i E
o i+i @i i€.i ic€Fi R
l@ 2 o
6'
8r
i@
l@'l
tl
i+i
4, s
m
F
I
o
i+
t@
i+
,<: _ i@i
ll
i.@-i
i+
i+i \ i*i
i+i i+i
BACK VIEW DURING CONSTRUCTION BACK VIEW DURING TESTING
tlrlrtl
o lFf.
NOTES'
4. Hordwood posts
The back of the model is installed as the model is built up; then
the plywood front is removed, a grid of reference points is sprayed
over the surface, and the clear plastic front is attached. To
simulate the excavation sequence, hardwood blocks having the shape
of each excavation stage are buried in the model (figure ?-Itb) and
later withdrawn under toad, through a hole in the plywood back.
Simple studies such as this are useful for many facets of
engineering for underground and surface excavatj-ons, for example:
identifying the critical points of an excavation; studying variations
of shape and locatlon; developing rational bases for designing
supports; and putting in context the deformation readings of individ-
ual instruments . Ivluch more elaborate studies are possible, of course,
and in fact may even serve as the basis for quantitative design
decisi-ons. Figure 7-l-2 is a typical result f rom many coal mining
Physical Models 293
Figure 7-12. Flexural failure of a laminated mine roof; a model studied by Dr. Everling
at the Bergbau-Forschung, Essen, West Germany. Courtesy of Dr. Everling.
Figure 7-13. Bedded limestone in the left abutment of Grancarevo Dam, Yugoslavia;
courtesy of Dr. M. Popovic, Institute for Geotechnics and Foundation Engin., Faculty
of Civil Engin., Sarajevo Yugoslavia.
(a) (b)
Figure 7-14. (a) View of the model during construction; note the use of plumb lines
to locate points in plan; courtesy of Dr. M. Popovic, Sarajevo. (b) A gener al view of
the model; courtesy Dr. M. Popovic, Sarajevo.
Figure 7-16. Reinforcement scheme to stiffen and stren gthe n t he fo undat ion
in the left abut ment. It was designed on the basis of the results of the mode l st udy.
INTRODUCTION
THE METHOD
aE
k =- (1)
1 L
~~~~~~~~~~~~~/~I
'" . ,
. <
.
,.
•
)
--I,
rock:
~ ~u
modulus of elasticity = E
'. -, . cross sectional area = a
- length = L
.
"
"
", ~. • .,' \ . I.
~u
, .
--L __ _
W = k ~u
Example 1
I
r
1
Fext I
(3a)
( 3b )
We may term the column matrices on the left hand sid e "nodal po i n t
external force vectors" {Fe} while t h e righthand column matr i ces ma y
be termed !fnodal point d isplacement vectors!f {u}.
St ep 3. - Assemble the st r uctu r al stiffnes s matrix. The e x terna l
forces for all nodal points of the structure can now be e xpr e sse d as
the products of the nodal point displacements (un k nowns) and app ro-
priate stiffness terms. The stiffness term kpq gives the c ont ribu tion
to the force at nodal point p cause d by the displacement at no d a l p oint
q; it is the sum of all element stiffness terms linki n g p and q an d
therefore is zero if p and q have no elements i n common. For e xamp le
(figure 8- 2b), the increment of e x tern al force F2 at n oda l po i n t 2
produced by a displacement u 2 at noda l poi n t 2 fo r the stru ct u re
consi dered is k 22 u 2 where k2 2 = k~J + k~I = k A + k B·*
Considering each nodal point in tu rn we can wr ite
-k a
(4)
( 4a )
{F } + {F.} (5)
e 1
°
In the example being discussed, {F } = <-P, 0 , X> T wher e X i s the
e
reaction to the displaced and fixed b o u ndary c o ndit i on at nodal
point 3. Introducing this value for {F } and a l so substituting for
e
{F.} from (4a) yields the simultane o u s e q u ations of the structure
1
-P
( 6)
°
X
x (7a)
(7b)
and
P
u2 Y - k
b
p p
y - "k-
a kb
p
{u} y - (8 )
kb
y
-p
Fr -F -p
J
°B a2 a2 a2
The steps outlined above are the same when a more general two
d imensional structure is solved; only the element stiffness matrix
u (x,y)
x
and
u (x,y)
y
The Finite Element Method 309
a
o
lOx 0 y 0
(10)
x
I
j Figure 8-3. Constant strain triangles.
310 The Finite Element Method
o o o o o
u y,I 1 o o o o
u x,J o o o
(11)
u y,J 1 o
u x,K o
u
y,K 1 o
(lla)
(12)
XKYJ 0 0 0 0 0
0 xJYK - xKYJ 0 0 0 0
1 YJ - YK 0 YK 0 -Y J 0
(q, )-1 (12a)
0 xJY K - xKY J 0 YJ - YK 0 YK 0 -Y J
xK - xJ 0 -x K 0 x 0
J
0 xK - xJ 0 -x K 0 xJ
The determinant xJYK - xKYJ equals 2a, where a is the area of the
element. The strain components in the triangle are:
au x
E
X ax
au
E --.X
Y ay
and
au au
x + --.X
ay ax
ux(X'Y»)
(13)
( u Y (x,y)
(1JJ){cd (14)
where
~)
o 1 o o
( '¥ ) o o o o (14a)
o o 1 1
YK 0 -YJ 0
1
0 -x K 0
-x K YK XJ
(16)
(16a)
(a) (b)
y
n
trace of bedd ing
Figure 8-4. (a) Definition of local and global coordinate directions; (ex) positive.
(b) Relationship of local coordinates and elastic symmetry directions-transversely isotropic rock.
The Finite Element Method 313
-v v
sn 1 tn
E IJ.° s + IJ.o -E; IJ.° t (17)
n
~ En n
and v
-v
IJ.° n +~
st nt
E IJ.° s - E IJ.° t
t ~ En t
v v
st ts
E
s v
vsn En ns
(18)
En
v v
nt Es sn
and:
2
-v
(1 - v st sn
) - E - (1 + V )
s st
2
-v sn v
1 sn
(D) (1 + v ) (20a)
~ st En -
~
0 0
- En v 2sn En v sn (1 + v t)
s
1
(C) E (1 (20b)
m n
o o
and
2
m (1 + v s t) [1 - v s t - (2E n IE s )v sn ]
1
En 1 +
E
1
k h
"
n
E E
s
(21)
1
G
sn 2(1 + v) 1
E + k h
s
v
sn
and v
st
= v
where k n and k s are the normal and shear stiffnesses of the joints
(discussed in chapter 5) and E and v are the elastic modulus and
Poi s s on's ratio of the rock.
The stress-strain relationship in the local co-ordinate system
sn can now be wr i tten:
s,n
316 The Finite Element Method
and {flO}
s,n
2 a. 2 1
sin
cos a. 2" sin 2a.
2 2 a. 1 (23)
(T ) sin a. cos -2 sin 2a.
E
1
v 0
1 - v
E{l - v} v
(C)
(1 + v)(l - 2v) - v
1 0 (20c)
1
0 0
2(1
(K) (24a)
The Finite Element Method 317
(K) (24b)
{a} (25 )
o
x,o
o
Y,o
T
xY,o
(26)
(27)
The net load vector on the structure due to applied forces {Fe}
with initial forces {Fo} is {6F} = {Fe} - {F }' The equilibrium
o
equations of the structure then are:
Example 2
*The dimensions for the problem are of no interest · here. Any consistent
set of units may be inferred, e.g. 0, E in MN/m2 and co-ordinates in
meters; or 0, E in p.s.f., and co-ordinates in feet. Avoid mixed
units, e.g. E in p.s.i. and co-ordinates in feet.
The Finite Element Method 319
(-~
n
0 1 0 0
(L 1 ) -2 0 0 0
-2 -1 0 1 2-
1111 278
(C) 278 1111
( o o
-1 0 -2
(K)
1
4
0
1
0
-2
0
0
-1
0
1
tlll 278
0
278
1111
0
0) (-1
0\0
417 -2
0
-2
-1
1
0
0
0
0
1
0
0
2 n
0 0 2
0 2 0
F 1 0 -2 6
x, I
F 0 -2 -1 5.5
f~~ }
y,I
F 1 0 0 -5
x,J 1
F
"2 0 0 1
y,J -0.5
F 0 0 2 -1
x,K
F y,K 0 2 -5
-5.5
5
{F } + (K){u}
e 0.5
1.
5.
actual discontinuity
y 5
~----~--------------x
{u}
s,n
uK + u L u I + uJ
u
0 2 2
v + v vI + v J
K L
{E;j} (28)
2 2
vK - v v J - vI
w
L
i i
u
I
vI
I I 1
u 0 0 0 u
0 2" 2" 2" J
v
I I I J
Vo 0 0 -"2 0 0 (29)
2" 2"
uK
I I I
w 0 0 -T 0 0
T T vK
uL
vL
VT//T/7 //77701
closed
Wffffffff~
(Vo negati ve )
open
WPJ//J'iJ/A (vo positi ve )
v~J/~/~
rota ted
! open on rig ht
partly closed
on left)
( w posi ti ve)
t'l'//~J/!//0/~
slid i ng wit h
dila t ancy
( Uo positive)
P77/T/T/T///M
slid ing with
contractancy
(uo posit i ve )
1
t:.a t:.a n 1(t:.F n ,K + t:.F n, L) (30)
1
and t:.T t:.T I(t:.Fs,K + t:.F s, L) , (30a)
nS
(30b)
t:.F 0 0
s,I
£ 1
t:.F 0 -2
n,I I
t:.F 0 0
s,J
£ 1
t:.Fn , J 0 -2 -I
(31)
£
t:.F "2 0 0
s,K
£ 1
t:.F 0 2
n,K I
£
t:.F 2 0 0
s,L
£ 1
t:.Fn , L 0 -1;
2
or
(32)
or
(32a)
* I K
1 J
~---------------------------------~~
nPI7
l+--
~p -1
o 0
(33)
{L'JF}
s,n
or finally,
· The Finite Element Method 327
( aJ
( b)
Equilibrium without rotation
Equili brium wi th rotation
Figurf! 8-9. Effect of block rotation on stress state in jointed rock masses.
(B)(CJ)(LO){u} (34)
s,n s,n
where
(B) is defined by (31) and (31a),
(C J ) is defined by (33),
and
(Lo) is defined by (29) and (29a).
ks 0 k 0 -k s 0 -k s 0
s
0 2k n 0 0 0 0 0 -2k
n
ks 0 k 0 -k s 0 -k s 0
s
0 0 0 2k 0 -2k n 0 0
(K) 1:.£ n (35)
4
s,n -k 0 -k s 0 ks 0 ks 0
s
0 0 0 -2k 0 2k 0 0
n n
0 -k s 0 k 0 ks 0
s
0 -2k 0 0 0 0 0 2k n
n
328 The Finite Element Method
(36)
YJ - YI
a. = arctan
{nF} (T){nF}
s,n X,Y
and (37)
{u} (T){u}
s,n x,Y
where
cos a. sin a. 0 0 0 0 0 0
-sin a. cos a. 0 0 0 0 0 0
0 0 cos a. sin a. 0 0 0 0
0 0 -sin a. cos a. 0 0 0 0
(T)
0 0 0 0 cos a. sin a. 0 0
0 0 0 0 -sin a. cos a. 0 0
0 0 0 0 0 0 cos a. sin a.
0 0 0 0 0 0 -sin a. cos a.
(37a)
and since (T)-l = (T)T, we can rewrite (34) as:
I
t sin 20. ~
sin 20. cos a
sn,o x,o
2 2
{ao J} a
n,o sin a cos a -sin a
y,o
, s,n
0 0 0 0 T
xy,o
(39 )
or
{a 0, J} (39a)
s,n x,y
+ {Lw J} (39b)
s,n
(T)T(B)(T
(j ,
J){a}
0
(40)
x,y
{F } - {F } - (F ) = (K){u} (41)
e o,J o,R
or
Example 3
A(joint)
Deformability _properties
of A: ks =500
k n=5000
L K
...---'l~v.------...
•
I J /'
(deformability properties of 8 and C and
Figure 8-10. Example 3. initial stresses - as in Figure 8-5)
The Finite Element Me thod 33 1
A A A A
kII kIL 0 kIJ kIK 0
A A B B A A B
k LI ( k LL + k II ) k IK kLJ ( k LK +. k I J ) 0
0
B B' e 0 B
(k KJ + k )
e e
kKI (kKK ,+ k JJ ) JK k JI
A
kA
A A
kJI kJL 0 kJK 0
JJ
A A B B e A A B e
kKI (k KL + k J1 ) (kJK + k KJ ) kKJ (kKK + k JJ + k~ ) kK I
0 0 e o ,- e e
kIJ kIK k II
~
0 0 0 0 0 0 0 5000 0 -5000 0 0
-250 0 -28 -139 0 347 -250 0 945 0 - 41 7 -20 8
0 0 -20 8 -104 34 7 0 0 -5000 0 62 15 -1 39 - 11 11
0 0 0 0 -278 -2 08 0 0 - 417 -139 69 5 347
0 0 0 0 -139 -104 0 0 -208 - 1 11r 34 7 12 1 3
"
332 The Finite Element Method
FA
;1 I
F2 FA + FB
L I
F3 FB + F C
K J
{F}
(12 x 1) FA
F4 J
F5 FA + FB + Fe
K J K
FC
-+
F6 I
-1 0 0 1
0 -1 +1 5
"2"
-1 0 0 1
(~ ~) flO} -
1 0 5
0 -1 -2"
{Fo,J} (I) 1 -5 -
1 0 0 -1
a -1
0 1 +!. -5
2
1 0 0 -1
1 1 -5
0 -"2
F -1
x,l
F
y,l -5
F -5
x,2
F
y,2 -0.5
F -4
x,3
F
y,3 4.5
F -1
x,4
F -5
y,4
F
x,5 5
F 0.5
y,5
F
x,6 6
F
y,6 5.5
F
REAL BEHAVIOJR
(·constitutive lOw·)
l:Iu
Figure 8-11. Alternative iterative schemes to constrain the solution to obey the real
material behavior.
I
% :J
'~CDL®-L~
b
f(I n'
-1(2
n'
...K3
Figure 8-12. Modified joint elements.
(k I ) (ks)l
s 1
tl
(k' )
s 2 -l't2- (Cks) 1 2 + (k s )2
t; )
(k~)3
1
, (Cks)2 t2
2" + Cks)3 t~) (42)
t3
(k ') (kn)l
n 1
tl t2 )
(k~)2 -l't2-(Ckn) 1 2" + (k n )2 2"
F (43)
nlo
Tension
(j _ Fn
-7
- - Vmc
Vm
(opening)
-V
V mc ~
(44)
m
°o
2
-0
k
n,o
°o
V
o
(44a)
m ~Vmc
where °0
is the initial normal stress in the joint element; this
value of normal stiffness produces a solution with displacements
Va and Vb at nodal points a and b along a modified joint element **
then:
(45)
F
°o ~ (46)
Vm V ,Q,'
m
(F n,o )1 + 6.N
l
Initiol slope = k~ f
:_0_
0:: g'
Vm RI(Fnl,~VI)
R2 (Fn21 Ll v2)
Vm R3R4
where (47)
F F
n,o n,l
F . ~'k'~V. + (F ).
n,l n 1 n,o 1
and (47a)
where
~
~v.
1
~N. F F .
1 ~v. n,o n,l
1 +
and (47b)
~v.
1
~v.V
( (Vm
1 m
-~v.)
1
+ (V - IIV)
m 1; (48)
vm(Fn - Fn , 0)
8.v (49)
F
n
The results of the first iteration now show 8.V l negat ive, with point
Rl = 8.V l , Fn,l' The initial load for the second iteration is:
(F n,o )1 + 8.N l
where
I
k Q., (50)
n
Vm c -----~~I
Vm
---j---j---------~~~~~-------6v
F: - F
l!Jv=V m ( n Fnn,o
)
F . £'k'~v. + (F ).
n,l n 1 n,o 1
and
(F ). + (~N.) ( 50a)
n,o 1 1
where
vm(F n,l. -
k' .Q,'
F . n
0,1
1
= kn £
, '( 2~v.
1
- V +
m ~v. 1
V 2
+ (F m )./(k'£')
n,o 1 n
)
(51)
(52)
The Finite Element Method 343
Tp = fl (O')
Tr = f2 (o-)
I n m
(53)
L - L
P 0
k
s
(54)
u
r
M > 1.
For L negative,
(54a)
Let
(54b)
L - L
r
II L -L
p + u2 - u
(~u - up (_) ) ; u r (_) < ~u < u
p(-)
p r
L - L
r
IV L = L + E (~u - up (+)) ; U < ~u < u (+)
p up - U
r
p(+) r
V L = L ~u > u (55)
r' r(+)
F (56)
s,l
where k's is given by (42) and (F s,o )1 = T 0 £' (compare with (46a».
Then, as for joint opening, the initial force at node a for iteration
2 is
(F s,o ) 1 + ~Sl
with (57)
F
s,l
£'k' ~
F
S,l
. s u i + (F s,o ).1
and
(F ). + ~S. (58)
S,O 1 1
with
L.£' F .
1 S,l
L.£' (59)
1
where T.
1
is given by (55) which in turn depends upon the choice of
fl and f2 (52) and (53). Any consistent specific experimental or
346 The Finite Element Method
To or = qu (assumed) cr
qu = urlcorlfirled compressive
strength
Tr Tp (Bo 1:uB~o
+
and (60)
T .
p'
< nu <
and (61)
or nu < u r (_)
The absolute value function and the minus sign insure that the joint
thickens regardless of the sign of the shear stress, as discussed in
chapter 5.
The variation of v with a was given by equation l6b of chapter 5.
Substituting this equation in (61) with aT = qu * and considering
stresses in the modified joint element gives for the ith iteration:
CFn)i_l ) 4
, - 1 tan i
( ~ q 0
u
and (62)
At the end of the ith iteration, we will know the shear dis-
placement flu. at each nodal pOint pair. Inserting flU in (62) for
1 i
each modified joint element, we determine flV.(T).
1
If all dilatancy is prevented by the adjacent elements, there
must be external compressive forces applied to joint nodes a and b
equal in magnitude to the dilatancy multiplied by the joint stiff-
ness. Redefining the jOint stiffness as the slope of the compression
curve (43) evaluated at the previous normal stress (F . l/~')' the
n,l-
increment in initial normal stress in the joint due to dilatancy
calculated for the ith iteration is
flV.(T)
1
(fla
n,o
).
1
= ["V
mc
(62a)
F
y,a cos ex. 0 0
(63)
F 0 0 cos ex. sin ex.
x,b
Water Forces
3. Form joint element stiffness matrix for each type length and
orientation of joint element. (35) (38)
livol livol
8 (64)
vol £'e
livol ( 64a)
C £'e
w
dh
F
x,I
- dX
F y,I 1 - dh
ay-
F dh
ax
~ext}water
x,J ywa
a
"3 -3-
F dh
1 - ay- (65)
y,J
dh
F
x,K --ay
dh
1 - -ay
Fy,K
A B C D E
dh/ch -0.79 -0.06 -0.04 -0.20 -0.1,3
ohley 0. 00 .0.34 - 0.13 .0.07 .0.58
Weight Forces
The initial stresses input in the analysis may reflect the action
of self weight forces on the region under investigation. But the
residual stress input can not duplicate the action of gravity because
"loading" by an initial stress and the loading of gravity are
fundamentally different. The residual stress load is like that of a
precompressed spring; it can not continuously follow the deformations
it causes in its neighbours. Gravity, on the contrary, produces an
inertia force that pursues the deforming body. Therefore, to simu-
late the gravity action in a beam we require additional forces input
The Finite Element Method 353
as external forces at the nodal pOints; (we should also reduce the
initial stresses accordingly so that gravity is not counted twice).
Since gravity produces a force distributed over each unit of mass,
it can be treated like the seepage forces just considered. Denoting
by y the weight per unit volume of the rock (total, not buoyant),
the required applied external forces in an element are:
F 0
x, I
F
y,I -1
F 0
x,J
~ (66)
3
F
y,J -1
F 0
x,K
F -1
y,K
F k cos a.
x, I
F
y,I -1 + k sin a
F k cos a
x,J
~ (66a)
3
F
y,J -1 + k sin a.
F k cos a.
x,K
F
y,K -1 + k sin a
354 The Finite Element Method
Rock Bolts
EXAMPLE PROBLEMS
UX,o= -I
2 3
(0,0) (1,0)
Propert i es '
Rock : E = 100 MN/m 2, 11 :0
Joint: when t = - 0.1, Vmc =0.05m
Relative Displacement 2 ~
-2
-3
-4
SOlution
TABLE 8-2
!::'v (j F
x n
0 -.100
-.010 -.125
-.020 -.167
-.030 -.25
-.035 -.333
-.040 -.500
-.042 -.625
-.045 -1.0
-.046 -1.25
-.048 -2.5
-.049 -5.0
flv
--~--~~----~--~~~----~----~~----~----~r-----~O
-I
-2
Solution
Method
- - unmodified
- - - - Newton Rophson -3
- -modified (I)
........... modi fied (2)
-4
k .
I F . - Fn,o
n,l F
n,l
.
(2)
n,l t !:.V.
1
-W-m
TABLE 8-3
Start 1.0 o o
1 0.667 -1.52 -0.00333
2 0.497 -1.16 -0.00503
3 0.368 -0.93 -0.00631
4 0.264 -0.76 -0.00736
5 0.179 -0.64 -0.00821
(F . - F ) F
1 n,l n,o n,o
k . (68)
n,l £ ~v
Joint Opening
Block and joint of Example 4
0"0=-2 .0
wit h init i 01 tension in block
Co m press ion
Convergence
Converges in 22 iterations. 2.0
With accelerator applied after
2 iterations:
Relative
- - Mod. Method 1-10 iterations
-0.01 -0.02
........ Mod. Method 2-11 iterations
L------r-r+H~~~--~------~----~------~--------~0
/
I
/ /
/ / -2.0
/ I
/
/
/
/
/ Block
/
/ -4.0
/
/
B
Figure 8-21. Example 5.
5~6
0 .5 N '
CD 4 @
N O"y,o=-0.520
0.5 .. I ® 2
~\ / "A- 3 ®~ 9
(O,~ __ (1,0)
O"y,o - 0 .520
Rock properties as in Example 4
J oint properties:
V =0.05 ~ = 0.1
mc
Ks = 20 MN/m 3 Bo= 0.333
4>fL = 30° LO = 0
Fx
{MN
1.01----------iF------ii~---~----__.,.__-
0.8
06
0.4
0.2
Diverg es - Unstable
TABLE 8-4
Output Displacements in Examples SA and 7
iteration
x at node 4
u (m)
example 6A example 7
1 0.0500 0.0500
2 0.0710 0.0587
3 0.0963 0.0500
4 0.1271 0.0500
The Finite Element Method 36 1
- - -Accelerated restart
after 3 iterations
Fx
(MN)
f.OI---"-------~-___:~-____:>.~--------
Diverges - Unstable
2.0
k . (69)
S,l
I Initial state
~-------~~
~
/ stress magn. scale
I I I
X
/ I I I Results after
I ~
I 10 th itera t ion
I I Results after I
/ I 1st iteration I
I I I
l.
-- -- -_ I
/
/
--- - I
--../~
I
,c;;mlIIIlmll77l1,},.
The Finite Element Method 363
EXAMPLE 9 EXAMPLE 10
Gravity "Turn-on" Gravity -Turn-on" plus 0.5 9
Horizon tally
Mesh 0 10 20 30 40 m
I I I I I
Displacement 0 0.2 0 .4 0.6 0.8 m
MESH
89
Displacements 200 X
en
~
~ 0..04
E
~
ci
~
a.
0.02
en Converging
(5
N
·c
~ OL-~-~-L-~-~-~~-~-~~----
o 2 4 6 8 10 Iteration
Mesh 0
I
10
I
20 30 40m
I I I
!!~~ 2l
1.0
o a. 0
______
I I I I
-----
I I I
O;,.,g;ng
I I I
Displacement 0 I 234m reO 2 4 6 8 10 It eration
EXAMPLE II A EXAMPLE 12
Gravity Turn- on Gravity Turn-on plus 0 .59 Horizontally
"stable" ?
-I-- --+- 2
-f-J...;-! S 10
--+-c -f- -f- ""i- -f-
+
+-
+ -I- /J
-r----+- ........ -+-&'
~~~
""'f..c .,
~ D
/0 ·
>-[- 219
+ 24 1
+2H
+ 7S~
+ ~Sl
+
"". +
-.fO
+164 +
·1tJ
+ ,u
-$tJ +m + 271 !I()
INCREMENTAL LOADING
ONE
an illustrative
finite element prog am
PROGRAM STRUCTURE
Table A-2 shows the sequence of computations and subroutine
calls for JETTY. The main program reads the input data and calls
subroutine STIFF to assemble the structural stiffness (KS) and
load vector {R}; this is done with the aid of subroutines TRIA and
JSTIF for triangles and joints respectively. The structural stiff-
ness matrix is inverted by a library matrix inversion routine (not
listed) which uses Gauss Elimination. The inverse of the stiffness
matrix, including the boundary conditions, is stored in the array
(KS). The main iterative loop of the program is entered after
initializing the displacements {u}. The total displacement at this
point becomes {u} = (KS)*{R}. From these displacements, solid
stresses are calculated in STRESS while joint deformations, and
initial load corrections, are computed in JSTR. Initial force
increments for the next iteration are returned by JSTR in {R},
and the iterative loop is re-entered with updated displacements
{u} = (KS)*{R} + {u}.
The sequence of computations and the equations programmed are
almost identical to those presented in Chapter 8 except for the
following points. Rather than compute an updated initial load vector
Good man-Geol og ical En g - 25
372 Append~ 1
TABL E A-1
ROCK
JOINTS
TABLE A-2
Structure of the Computer Program
1. TITLE CARD
An y title message desired; it will h ea d the out p ut .
2. FIRST CONTROL CARD (7I5)
Field:
~umber of n o da l points
(2) number of elements
(3) numb e r of n o d al points in the list of fi xed po in ts
(4) number of solid material types
(5) number of joint material types
(6) number of the first iteration of this r un ( prev io us + 1 )
(7) number of iterations to be computed this run.
3. SECOND CONTROL CARD ( 5F IO.O)
Field :
~.O if gravity forces are to be computed ; ot he rw ise b lan k
( 2) t he accele r at ion of g r avity
(3) pseudo-stati c acceleration coefficient k (ac ce le r a tio n = kg )
(4) di rection of the inertia force of the applied ac celerat ion
a n angle in degrees counter- clockwise from x.
(5) the mass density of water (t h is establishes t he sys tem of
dimensions)
4. Data se t s foll ow i n any order. A hea d ing car d pre ce des e a ch
set and identifies it.
HEADING CARDS FOR DATA SETS; begin in column 1; 9-72 a r e optional.
NODAL POI NT DATA
ELEMENT DATA
HYDRAULIC DATA
RESIDUAL STRESS DATA
BOUNDARY CONSTRAI NTS
SOLID ELEMENT PROPERTIES
JOINT ELEMENT PROPERTIES
INTERMEDIATE PRINTOUT DESIRED (no data set f ollows thi s car d)
a) NODAL POINT DATA ( I 5,4FIO.O)
F i eld :
(1) nodal point number
(2 ) x coordinate
(3) y coordinate
( 4 ) applied ext ernal for ce in x d irection
(5) applied e x ternal force in y direction
Omitted nodal points a r e linearly in t e r pol a t ed wi t ho u t exte r n a l
forces. The last nodal point (highest numbe r e d ) mus t be inpu t.
b) ELEMENT DATA ( 61 5)
376 Appendix 1
Field:
(1) element number
(2) material number (joints are assigned numbers higher than
solids) .
(3) nodal point I
(4) nodal point J
(5) nodal point K
(6) nodal point L (blank for solid elements)
Nodal points are designated in counter-clockwise sequence. In
joints, the long sides must be IJ and KL (see figure 8-20).
Elements may be omitted; the numbers of omitted elements are
assigned by incrementing the corner numbers of the previous
element and the previous material number is assigned. The
highest numbered element must be input.
c) HYDRAULIC DATA (I5,3FIO.O)
Field:
(1) element number
(2) "Head" - Head is 0 for a solid element above the water table
and 1.0 for a solid element below the water table.
(3) ah/ax (negative for flow in the direction of positive x)
(4) ah/ay (negative for flow in the direction of positive y)
If elements are omitted from the list, the gradients will be
assigned equal to those on the preceding card. The highest
numbered element to receive hydraulic data must be input.
d) RESIDUAL STRESS DATA (I5,3FIO.0)
Field:
(1) element number
(2) (J
X,o
(3) (J y,o
(4) T
xy,o
Elements may be omitted, in which case the residual stresses will
be assigned equal to those on the preceding card. The highest
numbered element to receive residual stress must be input.
e) BOUNDARY CONSTRAINT DATA (8(215))
(1) nodal point
(2) 1CODE .....
if 1CODE = 0, no constraint is introduced (this is a
convenience when running multiple problems with changing
boundary conditions)
if ICODE 1, x displacement is zero ("rollers" parallel to
y)
if 1CODE 2, y displacement is zero ("rollers" parallel
to x)
if ICODE 3, x and y displacements are both zero (fixed)
(3) next nodal point
(4) 1CODE for next nodal point, etc. up to 8 nodal points per card.
At least one node must be constrained
AppendIx 1 377
Example 3
Data coding form 1 shows the input information for example 3
of Chapter 8 (figure 8-10). In order to show the complete stiffness
matrix in the output, no nodal points were fixed; however since
at least one nodal point must be assigned a boundary constraint,
node 1 was assigned ICODE equal to zero (leaving it free). The data
deck includes the card INTERMEDIATE which causes all print statements
to be executed. Output number 1 shows the formation of element stiff-
ness matrices and the assembly of the structural stiffness matrix
and load vector.
Example 4
The input information for example 4 of Chapter 8 (figure 8-19)
is listed on coding form 2. Note the use of heading cards and the
intentional omission of some residual stress cards which will be
Append~ 1 379
~
I-xj
@
a=
I-l
,
I ,
'I I
21 I
31 I
~II
SI I
1,.
£ILE""EIMtr .D~tr~
()I.
0,.
2
2,.
0,---
, I. '
'0
0,. :E ,
I
: I
I
'
I
~ l' h'
i
II I 11+1 , 5 i I I
i I
I I •
I
21 I '1 21 '51 "3 J I
7 S I I 5
b Jahk c:::._~_~ IA
"'4b~L
, P ~ r.hl IT
0,
.b~'T
"'- 0,
-- I --
I-
-- -
Z , .0 O .
~ I .0 0. '
~ 0. 5 0, '5 -
S' o. , .0
--
6 l ,. O I .0
-
7 '. 0 I .0
ELf"'E NT
I ,
D A T A. -
I
--
1 ...
-- - -
: '
:1
3 ,
J
,
2. 6
5
4-
4-
~
,
I ' I '
t- I - i- _ I - 5 J ... -
5 .2. 3 7 6 2 -
Re.Sll)
, UA~
-5
~TRe. SS
.0
b A.,. A.
I
,
~ - t .0 -
a,UNO AiR,y
, c c6,. D J:T le NS
3 3 ; ~ S 5 3 6 2- 7 ~_
I
I
NODE XORD YORD XLOAD YLOAD NODE XORD YORD XLOAD YLOAD
1 o. O. O. O. 2 O. O. O. O.
3 O. 1.00 O. O. 4 2.00 O. O. O.
5 2.00 O. O. O. 6 2.00 1.00 O. O.
ELEMENT MAT TYPE NODE LIST ELEMENT MAT TYPE NODE LIST
1 2 I 4 5 2 2 1 2 5 3 0
3 6 3 5 0
TRANSFORMATION MATRIX
• 10000E+Ol O. O.
O. .10000E+Ol -0.
O. -0. .10000E+OI
JOINT PROPERTIES
JOINT TYPE NUMBER 2
WALL ROCK COMPRESSIVE STRENGTH -.1000E+04
RATIO OF TENSILE TO COMPRESSIVE STRENGTH= .1000
SHEAR STIFFNESS .5000E+OJ
RATIO OF RESIDUAL TO PEAK SHEAR STRENGTH: .6000E+00
MAX I MUM NORMAL CLOSURF .5000E-C2
SF:ATI"'G LOAD -.1000E+Ol
FqlCTION ANGL[ OF A S~OOTH JOINT .3000E+02
DILATANCY ANGL~ .500JE+Ol
ELEMENT QF Sl ou~L S TRESSES XX, YY, XY ELEME NT RESIDUAL STRESSE S XX,YY,X Y
J -.IO OOE+ 0 2 -.S OOOE +OI -.I OO OE+Ot 2 -. 1000E+02 -.SO OOE +OI - . IOOOE+OJ
3 -.IOOOE + 0 2 -.SOOOE+OI -.IO OOE +OJ
J OI NT S TI FF NE S S - LOC AL COORD I NA TE S
.2500 E+03 O. . 250 0 E +0 3 O. -.2500E+03 O. -.2500E+03 O.
O. • 5000f +O'" O. O• O. O. O. -.5000E+0~
.2500E "0 3 O. • 2500E+03 O. -.25 00 E +0 3 O• -.250 0E .. 03 O.
D. O. O. • 5000E+04 O. - .S OOO E +04 O. O•
-.2500E +03 O. - . 2500E"03 O. . 2500E+ O.J O. . 25001:+03 O.
O. O. O. -. 5 00 0E+ 04 O. . S OO OE +O", O. O.
-.250CF "03 O. -.2500E+03 O. . 2500E+ 0 3 O. . 2500E+'J3 O.
o. -.5000E+04 O. O. O. O. O. . 5 000E+ 0~
NDDE XORD YaRD XLOAD YLOAD NOf)E XORD yaRD XLOAI.> YLOAD
I -0. -0. -0. -0. 2 1.')0 -0. -0. -0.
3 1.00 -0. -0. -0. 4 • 50 • 50 -0 • -0 •
5 -0. 1.00 -0. -0. 6 1.00 1.00 -0. -0.
7 1.00 1.00 -0. -0.
ELE"IENT MAT TYPE NODE LIST ELE"4ENT "4AT TYPE NODE LIST
I I 2 4- -0 2 I 2 6 4 -0
3 6 5 4- -0 4- 5 4 -0
5 2 3 7 6 2
JOINT PROPERTIES
JOINT TYPE NUMBER 2
WALL ROCK COMPRESSIVE STRENGTH -.1000E+-03
RATIO OF TENSILE TO COMPRESSIVE STRENGTH= .1000
SHEAR STIFFNESS .1000E+-03
RATIO OF RESIDUAL TO PEAK SHEAR STRENGTH= .6000E+00
MAX I MUM NORMAL CLOSURE • 5000E-0 I
SEATING LOAD -.IOOOE+-OO
FRICTION ANGLE OF A SMOOTH JOINT .300I)E+-02
DILATANCY ANGLE -0.
ITERATION NUMBER
ITERATIQN NUMBER 2
G">
g
~ NODE X-DISPLACEMENT Y-OI SPLACEMENT
[ I
2
-.000000
.00~009
-.000000
.000000
_.
3 .000000 .000000
'"en .002005 -.000000
"5 -.000000 -.000000
~ 6
7
.004009
.000000
.000000
.000000
ITERATION NUIoISER 10
ELE~ENT xx- STRESS YY-STRESS XV-STRESS MAJOR PRINCIPAL MINOR PRINCIPAL ORIENTATION
-.4608E+OI -.7697E-15 .8882E-15 -.142IE-I] -.4608E+OI 90.16
2 -.46 08E+0 I .147IE-32 • 1273E-I 4 O• -.~608Ef-01 90.16
3 -.4608E+OI • 7697E-15 O• .I42IE-I] -.~608E+OI 90.16
~ -.4608Ef-01 -.1 703E- 32 -.3849E-15 O. -.~608E+OI -90.16
NODE XORD YORD XLOAO YLOAO NODE XORD YORD XLOAD YLOAD
1 -0. -0. -0. -0. 2 1.00 -0. -0. -0.
1.00 -0. -0. -0. .50 .50 -0. -0.
:J
s -0. 1.00 -0. -0. "
6 1.30 1.00 -0. -0.
7 1.00 1.00 -0. -0.
ELEMENT MAT TyPE NODE LIST ELEIoIIENT MAT TyPE NODE LIST
1 1 2 -0 2 I 2 4 -0
J 1 6 5 " -0 5
6
4 -0
5 2 3 7 "
6 2 "
ELEMENT RESIDUAL STRESSES XX.YY.xy ELEMENT RES -I DUAL STRESSES Xl(.YYtl(Y
1 -.5000E+Ol -0. -0. 2 - .5000-E +0 I -0. -0.
3 -.5000E+Ol -0. -0. 4 - .5000EHl1 -0. -0.
5 -.1000E+01 --0. -0.
TRANSFORMATION MATRIX
.10000E+Ol O. O.
O. .10000E+Ol -0.
O. -0. .10000F+Ol
STRESS-STRAIN MATRIX IN GLOBAL COORDINATES
.lOOOOE+03 O. O.
O. ·lOOOOE+03 O.
O. O. .SOOOOE+02
JOINT PROPERTIES
JOINT TYPE NUMBER 2
WALL ROCK COMPRESSIVE STRENGTH -.lOOOE+03
RATIO OF TENSILE TO COMPRESSIVE STRENGTH= .1000
SHEAR STIFFNESS .1000E+03
RATIO OF RESIDUAL TO PEAK SHEAR STRENGTH= .6DOOE+OO
MAX I MUM NORIo4AL CLOSURE • SOOOE-O 1
SEATING LOAD -.IOOOE+OO
FRICTION ANGLE OF A SMOOTH JOINT .3000E+02
DILATANCY ANGLE -0.
-
O.
o. O. o. -.2500E +02 O. . 1000 E+21 O. O. O. o.
o. -. 2 500E+02 O. .25 00 E +02
- .7 5 00E +02 -. 2 500 E +02 -.7500 E + 02 • 25 0 0E+02 O. O. .3000E "03 O• -.75 00 E"02 .250 0E +02
- .7 500E + 02 -.2500E+02 O. O.
-.2500E+02 - . 7500E+02 .2 50 0 E+0 2 -.7500E +02 O. O. O. .3000E+03 .2 500E+02 -.7500E+02
-.2 500E+02 - . 7 500E+02 O. O.
• 1250E+02 -.1250E+02 O. O. O• O. -.7500E+02 .2500E+02 .1000E+ 2 1 -.2500E+02
-.1 2 50E+02 .1250E+02 O. O•
• 1250E+02 -.1250E +0 2 O. O. o. O. .2500E+02 -.7500E"02 -.2500E+ 0 2 .1000E"21
-.1250E+02 . 1250E+02 O. O.
O. O. . 1 250 E +02 .1250E +02 O. O. -.7500F+02 -.2500E+02 - .1250 E+02 -.1250E+02
• 1 750E+03 .2500E+02 - .1 0 00E+03 O•
O. o. -.125CE+02 .1250E + 0 2 O. -.2500E+02 -.25 0 0 E+02 -.7500E+02 .1250E"02 .1250E+02
.25COE+02 .1000E+21 O. -.2500E+02
O. c. O. O. O. O. O. O. O. o.
-.1000E+03 O. .1000E+21 O.
O. O. O. -.2500E+02 O. .2500E+02 O. O. O. O•
O. -.2500E+02 O. • 1000E+21
-.2500E+Ol O. .2000F+Ol O. .5000E+00 O. O. O. -.2500E+Ol O.
• 2000E+Ol O. .5000E+00 O•
UP= -.4666E-Ol UR= -.1866E+00 TORP= .4666E+Ol TORR= .2868E+Ol TOR~ -.2401E-32
OILAT= O. DELV= -.1333E-Ol
UP= -.4666E-Ol UR= -.1866E+00 TORP= .~666E+Ol TORR= .2868E+Ol TOR= -.2407E-32
OILAT= O. DELV= -.1333E-01
JOINT STIFFNESSES
O. O. O. O. O. O. O. O. .1000E+03 .2000E+03
fg
~
394 Appendix 1
PPOGRAM JETTY(I~PUT,OUTPUTpTAPE5=INPUT,TAPEo=OUTPUT,TAPE7)
c
C PPOGRAM FOR ANALYSIS OF JOINTED ROCK STRUCTURES USING THE FINITE
C ELEMENT MET~OD WRITTEN ~V OR. C. ST. JOHN WITH T~E COLLABORATION
C OF DR R F GOnD~AN, USING THE THEORETICAL DISCUSSIO~ OF CHAPTER B
C IT IS A SMALL PROG~AM WITH LIMITED HORIZONS. ITS PRINCIPAL
C PURPOSE IS FDUCATIONAL, I.E. T~ Df.MONSTRATF. HOW THF NUME~ICAL
C MODEL CAN BE CODED IN PRACTISE. THE EQUATIONS REFERRED TO IN
C THE NUMEROUS COMME~T CARDS H~REIN ARE EQUATIONS OF CHAPTE~ 8
C
C LIST OF ~AIN ARRAVS AND THEIR USES
C
C MTVPE= ELEMENT MATERIAL OR JOINT TVPE
C KS GLOBAL STIFFNESS MATRIX
C R VECTOR OF NODAL FORCES
C ~XV 2*N MATRIX EQUIVAL~NT TO R - USED FOR INPUT ANO OUTPUT
C U VECTO~ OF NODAL DISPLACEMENTS - useD JNITIALV FOR WATER FORCE
c: UXV 2*N MATRIX ~aUIVALENT TO U - USED FOR INPUT AND OUTPUT
C X VECTOR ~= X-ORDINATES OF NOnES
C V VECTOR OF V-ORDINATES OF ~ODES
C ICODE= VECTOR OF DISPLACEMFNT CO~DITIONS
C 1 - FIXED IN X DIRFCTION
C 2 - FIXED IN V DIRECTION
C 3 - FIXED IN X AND V DIRECTION
C NOD N*4 MATPIX OF NODES n~FINING ELEMENTS
C RESID= N*3 MATRIX ~F RESIDUAL STRESSES
C SIGNO= N*3 MATRIX OF INITIAL STRESSES
C C GLOBAL STRESS STRAIN MATRIX FOR SOLID ELEMENT PROPERTIES
C ALSO USE~ TO STORE JOINT ELEMENT ~ROPERTIES
C RJ VFCTOR OF MASS DENSITY OF DIFFE~ENT MATERIAL TYPES
C G SFT TO ZEPO IF GRAVITV NOT ACTIVE
C SET TO ONE IF GRAVITV ACTIVE
C KJS VECTOR OF JOINT SHEAR STIFF~ESSFS
C KJN VECTOR OF JOINT NORMAL STIFFNFSSES
C KJC; NU= NEW JOINT SHEAR STIFFNESSF.S
C KJNNLJ= NFW JOINT NORMAL STIFFNESSES
C ITFP1= NU 'IIIBER OF FIRST ITERATION THIS RUN
C ITeRN= NU"4RER OF LAST ITFRATION THIS RUN
C NOOF.S= NUMBER OF NODAL POINTS
C NCLEM= NUMBFR OF ELEMENTS
C N30UN= NUMBER OF NODES WITH DISPLACEMENT CONSTRAINTS
C NSDL = NUMRER OF SOLID ELFMENT TVPF.S
C JOINT= NUMBER OF JOINT ~LfMENT TYPES
C ITER = NUMBER OF ITERATIONS THIS RUN
C G~AV = ACCELERATION DUE TO GRAVITV
C ACCf"L= ACCELERATION NOT DUF TO GRAVITY - IN G
C ~IR DIRECTION OF ACCEL - + ANTI CLOCKWISE FROM HORllONTAL
C L~AD SFT TO ZERO. 8EC00.4[S 1 IF AN ERROR IS CNCOUNTJ::RED
C NUM VECTOR OF NODES WITH nISPLAC~MENT CONSTRAINTS
C HEAD 1.0 FOR SOLID ELEMENT ABOVE WAT~R TABLE
C 0.0 FOP SOLID ELEMeNT AELOW WATER T~9LF
C
COMrJlON /ME.:SH/ X(25) ,Y(25) ,R(50) ,U(5()) ,KS(50,50),
1 NOD(23,4),PESID(23,1),StGNO(23,3',MTVPf(23),
2 NODES,NELEM,NSOL,NFREE
Rf~L KS
C 0\1 M0 N / C; T F F / K J S ( 2 3 ) , K J N ( 2 3 , , K J S NU ( 2 3 ) , KJ "I NU ( 23 ) , I T E R 1 , I T E ~ N , I T
REAL KJS,KJN,KJSNU,KJNNU
COMMON /BOUN/ NUM(20),ICOOE(20),NBOUN
Appendix 1 395
C
DP"ENSION 8(5~), SCRTCH(lSO)
( THESE ARE NEFDcn FORTHE MATRIX INVERSION SU9ROUTINE
C
C READ THE TITLE CARD AND THEN WRIT~ IT
C
R F'&'O C 5, lOr:: ': 1 ICARD
~RITE(6,2:)O,,) If:APD
c
C READ ESSENTIAL CONTROL DATA AND wRITE IT OUT I~MEDIATELV
C
RF.ADC5,10~l' NODES,NELF~,N~OUN,NSOL,JOINT,ITERl,ITCR,G,GRAV,ACCEL,
1 DIR,PW
IF CITsRl . f O. 01 ITFRl:: 1
WRITE(~,2001)NQDE5,N~LEM,N90UN,~50L,JOINT,ITER1,ITER,G,GRAV,ACCEL,
1 DIP,RW
ITERN = [TERl+ITFR-l
C
C S~T SO~E INITIAL VALues
C
NTQT= NSOL+JOINT
NR~S=O
LBAD=O
INTEP =
PI=ATAN(1.C'*4.G
C
C SET 50"'= OF THE ~qRAVS TO ZERO BFFORE STARTING
(
NFREE=NODES+NODES
(
DO 1 1=1,50
RCf)=O.O
LJ(I)=C. ':
CONTINUE
DO 2 I=I,NELEM
KJS(I)=O.O
KJN( 1)::0.0
KJNNU( I )=0.'2
KJ SNU C I ) ::(). 0
00 2 j =1 , 3
RE SID C I , j ) :: J • C
5 I GNO ( I , J) = 0 • C
2 (OPI.ITINUE
C
C NO~ CALCULAT~ TOTAL ACCELERATION COMPONENTS - EQUATI9N 66A IN r~XT
C
DIR=DIR*PI/180.
ACELX=ACCEL*COS(OIR).GRAV
C
C G C~N TAKE VALU~ 1.0 OR 0.0 CORRESPONDING TO GRAVITV ON OR OFF
C
ACELV=I-G+ACCF.L*SIN(OIR»*GRAV
C
C q~AD NEXT OAT~ CARD -THIS MUST BE A HEADER CARD FOR A OATA SET
C OP A START / STOP / RESTART / INTER
396 Appendix 1
C
CALL INTERMEOIATE RESULTS DESIRED ON PRINTOUT
C
SINTER = 0
GO TO 10C
c
C RFAD NOO~L POINT INFORMATION
C
1~ L=I)
11 READCS,1004) N,X(N' ,V(N) ,RXV( 1 ,N) ,RXY(2,N,
c
C A SIMPLE NODAL POINT GENERATOQ AND ~RROR SFARCH
C
IF(L.Ea.o, GO TO 12
lX=N-L
DY=(YCN)-Y(L,,/ZX
OX=(X(N)-X(L')/ZX
12 L =L + 1
IF- (N ... L' 15,14,13
13 X(L)=X(L-I,+DX
V(L)=Y(L-l)+DY
GO TO 12
14 IF (NODES - N) 15,16,11
C
C NOOAL POINT DATA ERROR WQITE EqROR MESSAGE
C
15 WR I TE ( 6,20 I) 4' N
Lt;!AD=LBAD+l
Jf(N.LT.NODES' GO TO 11
C
C WRITE OUT NODE OATA
c::
16 WRITE(6,2003) (K,X(K),YCK),RXY(1,K),RXY(2,K),K=I,NODES)
GO TO 100
c
C F.~EMENT DATA - MATERIAL TYPE ANO NODAL POINTS
C
20 N.O
21 READ(S,100S) K,MAT,(NOD(K,I),I=I,4)
C
C A SIMPLE FLEMENT GENERATOR ANO F.RROR SEARCH
C
MTYPF,(K)=MAT
IF(MAT.GT.NTOT' GO TO 27
22 N=N+l
IF CK - N'27,26,23
23 Nl=N-l
MAT:;: MTYPE(Nl'
Append~ 1 397
MTVPE(N)=MAT
NOD(N,J )=NOD(Nl,l'+J
NOQ(N,2,=NOOCN1,2)+1
NOD(N,3)=NODCN1,3)+1
IF CMAT .GT. NSOL) GO TO 24
NOOCN,4)=O
GO TO 25
24 NOD(N,4)=NQD(Nl,4)+1
25 IF CK - N) 27,26,22
26 IF (NFLEM - N) 27,29,21
c
C EL~MENT ~ATA ERRQR - WRITE ERROR MESSAGE
C
27 WRITE(6,2007) K
lBAO=LBAQ+l
IF(K.LT.NELEM) GO TO 21
c
C ~PITF OUT FLEMENT DATA
C
~~ WRITE(6,?C~6) (K,MTVPECK',(NOD(K, 11,1=1,4),K=J ,NELEM'
GO TO 100
c
C READ HYORAULIC GRADIFNT DATA FOR ELEMENTS
C
30 L=O
31 READ(5,1004) N,HEAD,DHDX,OHDV
c
C ZERO ENTRIES FQR JOINTS AS RESIDUAL STRESSES WILL BE EFFECTIVE
C FOR A SOLID ELEMENT HEAD =1.0 IF BELOW WATER TABLE
c HEAD = 0.0 IF ABOVE WATER TABLE
c
C
C CALClJLATE AND THEN STORE TEPMS REQUIRED FOR EQUATION 6S
C PRESSURE GRADIENTS ARE STORED TO AVOID HAVING TO STORE
C THE WATER TABLE SWITCH
C
GW=GRAV*HEAD*RW
UXYC1,N)=DHDX*GW
UXVC2,N)=(DHDV-l.0'*GW
c
C GENERATE MISSING DATA AND CHECK FOR ERRORS
C
32 L=L+l
IF (N - L) 35,34,33
33 Ll=L-l
UXVC1,L)=UXVC1,Ll'
UXV(2,L)=UXY(2,Ll)
GO TO 32
34 IF (NELFM - N, 35,36,31
C
C HVDRAULIC GRADIENT CARD ERROR - WRITE MESSAGE
C
35 WRITE(6,2010) N
LBAD=LRAD+l
IF(N.LT.NELEM) GO TO31
C
C WRITE OUT PRESSURE GRADIENTS FOR ELEMENTS
C
36 WRITF(6,2009) (K,UXVC1,K),UXV(2.K),K=1,~LEM)
GO TO 100
c
C READ n~SIDUAL STRESS CAROS
C NRES SET T, 1 TO INDICATE THAT ~ESI)UAL STRESSES HAVE BEEN READ
398 Appendix 1
C
40 NRES;;I
L=C
41 READ(S,1004' N,RESI~(N,I"RESID(N,2"RESIDrN,3)
C
C GENERATE MISSING DATA AND THEN CHECK FOR ERRORS
C
42 L=L+I
IF (N - L, 4S,44,43
43 Ll;;:;L-1
RESID(L,l '=RESIO(Ll.l,
RESID(L,2,=RESID(LI,2)
RESID(L,3)=RESID(Ll,3)
GO TO 42
44 IF (NELEM - N, 4S,46,41
C
C RESIDUAL STRESS CARD ERROR - wRITE ERROR MESSAGE
C
4S WRITE(6,2013) N
LBAD=LBAD+l
IF(N.LT.NELEM) GO TO 41
C
C WRITE OUT RESIDUAL STRESS DATA
C
46 WRITE(6,2012) (K.RESID(K,I'.RESIO(K,2',RESID(K']',K=I,NFLEM'
GO TO 100
C
C READ LIST OF NODAL CONSTRAINTS AND THEN WRITE THEM OUT
C
SO READ(S,IOOS) (NUM(K),ICO~E(K),K=I,NBOUN)
WRITF(6,20IS) (NUM(K),ICODE(K),K=I,NBOUN'
GO TO 100
C
C REAO SOLID ELEMENT PROPERTIES AN~ THEN WRITE THEM
C
60 WRITE(6,~017)
00 69 K=I,NSOL
READ(S,1004) MAT,RO(MAT),ES,EN,GSN,PRSN,PRST,ALPHA
WRITE(6,2018~MAT,RO(MAT),ES,EN,GSN,PRSN,PRSTtALPHA
C
C F~~~ STR~SS-STRAIN MATRIX IN LOCAL CO-ORDINATES - EQUATION 208
C
RAT=EN/ES
CONST=1.C/«1.0+PRST'*(1.0-PRST-2.0*RAT*PRSN.PRS~))
E( 1 ,I ,=(ES-EN*PRSN*PRSN)*CONST
E(2,1)=EN*PRSN*(1.O+PRST)*CONST
F.(3,1)= 0.0
E(l,2'= E(2,1)
E(2,2'=EN*(1.0-PRST*PRST'*CONST
E(3,2'= 0.0
E( 1 ,3'= 0.0
f(2,])= 0.0
E(3,3)=GSN
C
IF (INTER .EQ. 0) WRITE (6,99Q) «E(J,L),L=I,3),J=I,31
999 FOQMAT (44Ha STRESS-STRAIN MATRIX IN LOCAL CO-ORDINATES/
I (SX,3E15.S"
C
C NOw FORM THE TRANSFORMATION MAT~IX - EQUATION 23.
C
ALPHA=ALPHA*PI/IBO.C
COSA=COS(ALPHA)
SINA=SIN(ALPHA)
A ppendix 1 399
T ( 1 ,1 ) =C OS A*C OSA
1( 2,1 1= S I NA*S IN A
T( 3,1 )= - 2 . 0 * SIN A* COSA
T (I, 2)= T(2 ,1)
T( 2, 2)= 1 (1 , 1 )
T( 3 , 2 ) =- T ( 3, 1)
T( I,3 ) -= S INA*COSA
T( 2,3 ) = -T(1,3)
T (3,3)= T ( l ,I)-T( 2, 1)
C
IF (I NT ER . E'O. 0) WRfTf" (0,9 98) «(T( J, L ) ,L=1,3),J=1,3)
999 FO QMAT (23 HC T RA NSFO QMA TI ON MATRIX/(5X, 3E 15.5»
C
C NOW T~ANSF ORM TO GLO BAL CO-ORD INA L SCHEME - US E H AS TFMP. STORAGE'
C
DD 66 N= I ,3
DO 66 J -:l ,3
He J,N) = C. C
DO 6 6 L=I ,3
H (J,N) =H(J,N)+E (J,L )*T ( L ,N)
66 CONTI NUE
DO 6 7 N=I,3
DO 67 J= 1 , 3
CeJ ,N,MA T )= O.O
00 67 L=1 , 3
C( J,N,MA T )=CCJ,N,MA T)+T(L,J )*H( L,N)
67 C O ~T rN UE
C
I F eINTER .CO. 0) WRITF e6,997) « C ( J, N, MA T),N-=I, 3) ,J=I,3)
997 FORM AT (44 H ~ STRE SS -ST RA I N MA TRIX I N GLOBAL COOR DI NATES/
J ( SX,3 EI5. S »
(
C (HE(K FOR ERRORS
C
IF (N S OL .GF. MAT ) GO TO 69
WRI TE(6, 2 01 Q)
LBAD =L BAD+l
69 CON T IN UE
GO TO 100
C
( RE AD J OINT PROPCqTICS AND STORF IN C
(
70 WR I TE(6,2 02C' )
NT OT = NSOL+JOINT
DO 75 J=I, J O INT
REAO(S,100 4 ) MAT ,(C J( K, MA T),K=1,8)
(
C C O~PRFSSIVE STREN GTH NEGATIVE AN~ TENSILE TO COMPPESSIVF RATIO +VE
( MAXIMU M CLOSU RE MUST RE POSITIVE
C S E AT I NG LO AD MUST ALW AYS BE NEGATIVE
C
CJ( l, MA T)=-A AS( CJ ( l ,MAT»
C J (2 ,MAT )=ABSe CJC2,MA T»
CJ(5,~AT )=AB S(CJ (5 t MA T')
(J (6 , MAT'=-ABS( C J ( 6,~AT)'
(
WR IT ::(6,2021 ) MA T, (CJ(K,M~T) , K =l ,R)
c
C CHE CK FO R FRROR S
r
IF ( NTO T .GF:. MAT) GO TO 75
WRI TE (6, 2022)
LBAD=LBAO+l
400 Appendix 1
75 CONTINUE
C
C IF THIS NOT THE FIRST RUN THEN JOI~T STIFFNESS IS TO BE REA~
C
IF(ITERI.GT.I) READ(?) KJS,KJN
GO TO l~O
C
C DATA INPUT COMPLETE - PROCEED IF NO ERRORS
C
BO IF (LBAD .EO. 0) GO TO 300
C
C ERRORS HAVE BEEN FOUND - WRITE MESS_GE
C
WRITE(6,2023)
STOP
C
C RESTART 8Y READING OATA FqOM TAPE?
C
95 RE'AD(7)
GO TO ~75
C
C CALL SUAROUTINE STIFF TO ASSEMBLE THE STIFFNESS MATRIX
C
300 CALL STIFF
C
C CALL LIBRARY SUAROUTINE TO INVERT THE STIFFN~SS MATRIX
C ~FRE~ IS THE SIZE OF THE STTFFNESS MATRIX
C IN LAPGER PROGRAMS, STIFFNESS MATRIX SHOULD ~E NARROW BANDED
C
IF (INTER .NE. 0) GO TO 350
WRITt=:(6,996)
996 FORMAT ()4HO LOAD VECTOR AND STIFFNESS MATRIX)
DO 995 I=l,NFREE
WRITE(6,994) (KS(I,J),J=l,NFREE)
994 FORMAT(2X,10E12.4)
9q5 CONTI NUE
WRITE (6,994) (R(I),I=l,NFREE)
C
350 CALL LINV3F (KS,A,1,NFREE,50,-1.0,02,SCRTCH,IER)
C
C KS IS NOW THE INVERSE OF THE STIFFNESS MATRIX
C
C
IF (INTER .NE. 0) GO TO 359
WRITE (6,993)
993 FORMAT (29HC INVERSE OF STIFFNESS MATRIX)
DO 992 I=l,NFREE
992 WRITE (6,994) (KS(I,J),J=I,NFREE)
C
C SET INITIAL STRESSES TO RESIDUAL STqESSES RFFORE STARTING
C
359 DO 360 M=l,NELEM
SIGND(M,l)=RESID(M,l)
SIGND(M,2)=RESTD(M,2)
STGNO(M,3)=RESIO(M,3)
360 CONTINUE
C
C SFT INITIAL DISPL~CEMENTS TO lERO
C
DO 37C J=I,NFRFF
U(J)=O.O
370 CONTINUE
c
Appendix 1 401
400 CONTINUE
C
C WRITE OUT NODAL DISPLACEMENTS
C
WRITF.(6,20?4) (J,UX,( 1 ,J) ,U)(Y(2,J) ,J=l ,NODES)
C
C SfT LOA~ V~CTOR TO ZERO FOR INCRE~ENTAL LOADS
C
DO 450 J=l,NFRFE
R(J)=O.O
450 CONTINUE
C
C CALL STRFS TO CALCULATE SOLID ELEMENT STRESSeS
c
CALL STRES
C
C CALL JSTR TO CALCULATF JOINT 9EHAVIOUR AND NEW LOAD VECTOR
C
IF(JOINT.GT.O) CALL JSTR
C
5eo CONTINUE
C
GO TO 100
C
C ACC=:LE'RATE WRITE NEW JOINT STIFFN=SSFS ONTO TAPE
c
600 WRITE (7) KJSNU,KJNNU
GO Tn 700
c
C SAV~ STIFFNESS WRITE ALL DATA AND INvERTFD STIFFNESS MATRIX ON TAPE
C
6S0 WRITE (7) C,X,'(,R,U,KS,NUM,NOD,RESID,SIGNO,MT,(P~,ICODE,KJS,KJN
c
7C::} STOP
C
C INPUT FORMAT STATEMENTS
C
1000 FORMAT(9A8)
IO~1 FORMAT(7IS/SFIC.OI
1004 FORMAT(IS,7FIO.O,F5.0)
1005 FORMATCI615)
C
C OUTPUT FORMAT STAT~MENTS
C
2000 FORMAT(IHl,lOX,9A8)
2001 FORMAT( 30HO NUMBER OF NODAL POINTS =,13 /
1 30H NUMBER OF ELEMENTS ::, 13 /
2 30H NUMAER OF CONSTRAINED NODES=,I) /
3 30H NUMB~R OF SOLID MATERIALS =,13 /
4 30H NUMBER OF JOINT TYPES ::, 13 /
30H NUMOER OF FIRST ITERATION =,13 /
402 Appendix 1
SUAROUTINE STIFF
C
C SUAROUTINE TO ASSFMBLF MODeL STIFFN~SS ~ATRIX AND LOAD VfCTOR
C AT ENO OF SUARDUTTNE THE MATRIX IS READY FOR INVFRSION
C
rOMMON /MESH/ X(2S),Y(2S),R(50),U(SO),KS(SC,5C),
1 NOD(23,4),RESID(23,3),SIGNO(23,3),MTYPE(23),
2 NODES,NELFM,NSOL,NFREE
RFAL KS
COMMON /BOUN/ NU~(20),ICODE(20),NAOUN
COMMON /FORC/ KSUR(8t8),F(8),LBADt~,H(Jt6)
REAL KSUA
Appendix 1 403
DIMENSION RFS(3),XT(4),YT(4),LAB(4)
C
C SET STIFF NE SS KS TO ZERO
C
LAAD = 0
00 100 J=l,NFREE
DO 1~0 l=l,NFREE
KS(I,J)=O.O
100 CONTINUE
C
C FOR EACH ELEMENT CALCULATE CONTRIAUTION TO GLOBAL STIFFNESS AND LOADS
C
DO 450 M=I,NELEM
MAT=MTYPF.fM)
C
C 15 !T ~ JOINT OR A TRIANGULAR ELEMENT
C
NUMB = 3
!F (MAT .Gr. NSr)L) NUMR 4
C
C EXTRACT NODAL COORDINATES FROM NODAL COOR~IN~TE V~CTnRS
C
250 l=l,NUMB
0(1
N="!OO (M, I )
XT( II=X(N)
YT(I)=Y(N)
250 CONTINU E
C
C SEL E CT JOINT OR TRIANGULAR STIFFN~SS SUAROUTINE
C
M2=M+M
PES(I)=RE510(M,1)
RES(2)=RESIOfM,2)
RES(3)=RESID(M,3)
IF (MAT .GT. NSOLl GO TO 270
c
[)POX=U(M2-1)
OPf)Y=U(M2)
CALL TRJAfO,MAT,XT,YT,RFS,DPDX,~PDY)
IF (LAAO .NE. 01 450, 28(')
c
27C CALL JSTIF(MAT,XT,YT,RFS)
IF (LBAD .GT. 0) GO TO 450
C
C RETAIN TRANSF~RMED RESIDUAL STRESSES
C
PE510(M,1)=RES(11
RESIDfM,2)=RES(21
RESIDfM,3)=R~S(3)
C
C NOW ADD THE ELEMENT CONTRIBUTIONS KSUO AND F TO KS AND R
C ASSFMALE VECTOR INDICAT[NG LOCATION OF ELEMfNT CONTR!BUTION
C
280 0(1 300 !=l,NUMB
J=NOD(M,I)
LAA([J=J+J-l
300 CONTINUE
C
00 400 J=l,NUMB
[I=LAEHII
[K=[+f-l
12=1 [+1
C
Goodman-Geologic al Eng .- 2·7
404 Appendix 1
C LOAD CONTRIBUTION
C
R( Ir)=R(II)+F(IK)
R( I2)=R(I2)+F( IK+l)
DO 400 J=I,NUMA
JJ=LAB(J)
JK=J+J-l
J2=JJ+l
C
C STIFFNESS CONTRIBUTION
C
KS(II,JJ)=I<S(II,JJ)+"'SUFI(IK ,J'"
KS( II ,J2)=KS(I I ,J?)+KSUA( IK ,JK+l)
KS(12,JJ)=KS(12,JJ)+KSUA(IK+l,JK
KS(I2,J2)~KS(12,J2)+KSUB(IK+l,JK+l)
4CO CONTINUE
450 CONTINUE
IF (LAAO .GT. 0) GO TO 600
C
C NOW INTRODUCE NODAL CONSTRAINTS
C
C ICDDE - FIXED IN X DIRECTION
C ICODE 2 - FIXED IN Y DIRECTION
C ICODE 3 - FIXED IN X AND Y DIRECTION
C
C SET DIAGONAL TERM TO VERY HIGH STIFFNESS
C EFFECTIVELY FIXES NODE IN REOUIRED DIRECTION
C FOR GIVEN DISPLACEMFNTS SET R(JJ)=DISPLACEMENT*1.F+20
C
DO 500 I=I,NBOUN
L=ICODE(I)
IF(L.EO.O) GO TO 500
J=NUM( I)
JJ = J + J
IF (L .EO. 2) GO TO 470
KS(JJ-l,JJ-l) = KS(JJ-I,JJ-l) + 1.0E20
IF (L .EO. 1) GO TO 500
470 KS(JJ,JJ) = KS(JJ,JJ) + I.OE20
500 CONTINUE
C
C
PETURN
C
C ER~ORS DETECTED
C
600 WRITE(6,2000)
2000 FORMAT (46HC PROGRAM STOPPED AS ERRORS HAVE HF.FN DETECTED)
STOP
C
END
SUAROUfINE TRIA(KUT,MAT,X,Y,RESID,DPDX,DPDY)
c
COMMON /CONS/ PO(lO),ACFLX,ACELY,NRES,JNTER,C(3,3,lO)
COMMON /FORC/ KSU8(B,8),F(8),LAAD,M,H(3,6)
PEAL KSUB
DIME~SION ~(4),Y(4),RFSID(3)
REAL LNORT(3,6)
C
Appendix 1 405
C
C IF THIS AREA IS ZERO OR NEGATIVE THERE IS AN ERROR IN THE DATA
C RFTUPN IF AN ERROR HAS BEEN FOUND
C
tF(AREA.GT.O.O) GO TQ 10
Lf3AD=LBAD+l
WRITEC 6 ,200G ' LB AD ,M
2000 FORMAT (7HO EPRORI3,9H, EL~MENTI5,22H HAS ~ON-POSJTIVE AREA)
RETURN
C
C FORM STRAIN DISPLACEMENT MATRIX -LNORT -EQU AT ION 16 IN TEXT
C
C NOW CALCULATE THE NECESSARY TERMS - TOGETHER wITH THEIR SIGNS
C
1~ XK=-XK/AREA
YK= YK/AREA
XJ= XJ/ARt:A
YJ=-YJ/AREA
XK J=-XK -XJ
YJK=-YJ-YK
C
C SET LNORT AS EQUATION 16
C
LN()RT(l,I)= YJK
LNORT(2,I'= o.~
LNORT(3,1)= XKJ
LNORTCl,2)= o.n
LNORT(2,2)= Xt<'J
LNORT(3,2'= YJK
LNORT(I,3)= YK
LN.)RT(2,3)= 0.0
LNORT(3,3'= XK
LNDRT(1,4'= 0.0
LNClRT(2,4,= XK
LNORT(J,41= YK
LNORT(1,5)= YJ
LNORT(2,S'= 0.0
LNORT(J,5'= XJ
LNORT(1,6'= 0.0
LN;)RT(2,6,= XJ
L NORT ( J,f>,= YJ
C
IF (I N T E R • Ea. (') WR I T EO (6 , Q Q 9 , ( ( L NO R T ( t , J) , J =I ,6' , I :: t ,3 )
99~ FORMAT (28HO STRAIN DISPLAC~MENT ~ATRIX/(5X,6E12.4»
C
406 Appendix 1
c:
C NOW CALCULATE STRESS-DISPLACEMENT TRANSFORM~TION MATRIX ( H'
e GIVEN BY C *
LNORT ,WHERE C IS THe STRESS-STRAIN ~ATRIX
C FOR THE RELEVANT MATERIAL TYP~ ( MAT'
C
DO 10C J=1,6
DO 100 1=1 ,3
H(I,J)= 0.0
DO 100 K=1,3
H(I,J)=H(I,J' + C(I,K,MAT,*LNORT(K,J'
1 CO CONTI NUE
C
IF (INTFR .EQ. 0) WRITE (6,998) «H(I,J),J=1,6',I=1,3)
999 FOQMAT (41HO ST~ESS-DISPLACE~ENT TRANSFORMATION MATRIX/
1 (5X,6E12.4',
C
C NOW PETURN IF THIS SUAROUTIN~ WAS CALLED FROM STRES
C
IF(KUT.EQ.1' RETURN
C
C ~OW COMPLETE THE FORMATION OF THE STIFFNESS MATRIX ( KSUB'
C FIRST ~ULTIPLY LNORT BY THE AREA OF THE ELE~ENT
C
AREA=AREA/~.O
00 200 J=1,6
DO 200 1=1,3
LNORT(1 ,J'=LNORT( I,J)*M~EA
2eo CONTINUE
C
C CALCULATE (LNORT) TRANSPOSE * H - EQUATION 24A IN TEXT
t
DO 300 J=1,6
DO 300 1=1,6
KSUB(I,J)=O.O
DO 300 K=1,3
KSU/3( I,J)=KSUB( I ,J'+LNOPT(K, I '*H(K,J)
300 CONTINUE
C
C CALCULATE CO~TRIBUTIONS OF GRAVITY LOADING, ACCELERATION LOADING
C AND WATER FORCES TO LOAD VECTOR -EQUATIONS 65 AND 66A
C
CONST=AREA/3.C
XCOMP=(RO(~AT'*ACELX-~PDX)*CONST
YCOMP=(RO(MAT)*ACELY-DPDY'*CONST
F( 1 )=XCOMP
F(2)=YCOMP
F(3)=XCOMP
F(4)=YCOMP
F (5)= XCOMP
F(6)=YCOMP
C
C NOW AD~ TH~ CONTRIAUTION OF THF RFSIDUAL STRESSES - - EONS. 27 AND 27A
C
IF(NRES.EQ.O' GO TO 500
DO 40C J=1,6
DO 40C 1=1,3
F(J)=FeJ)-LNORT(I,J)*RFSIO( I)
400 CONTINUE
C
5eo IF (INTER .EQ. 0) WRITE (6,997' M,«KSU8el,Jl,J=1,6),F(I),t=1,6'
997 FORMAT (29HO STIFFNESS MATRIX - ELEMENTIJ,73X,11HLOAD VECTOR/
1 (2X,6E12.4,28X,E12.4"
C
A ppendix 1 407
Rt'TU RN
E ND
C
C CONS TANTS FOR FURTHFR CALCULATION
C
COSA=DX/L
SIN A=DY / L
S2= CO S4 *S INA
CC=COS. $ COSA
SS=SIN<\*SINA
C2=C C- SS
C
C ARE THERE RESIDUAL STRESSES - IF SO , TRANSFORM THEM
C
IF (NRES .GT. 0) GO TO 80
C
C RESI DUAL S HEAR S TR FSS I S ZERQ
C
TORSN=C.O
SIGMA = o.
408 Appendix 1
GO TO 90
80 TORSN=-S2*RESID(1)+S2*RESID(2t+C2*RESIO(3)
SIGMA= SS*RESID(l t+CC*RF.:SID(2)-2.0*S2*RESIDC3)
C
C
C RFTAIN TRANSFORMED STPESSES FOR FUTURE USE
C
90 RESID(1) = TORSN
RESID(2)=SIGMA
RFSIDC3'=0.0
C
IF (INTER .Ea. 0' WRITE C6,999) TORSN,SIGMA
999 FORMAT C31HO JOINT RESIDUAL SHEAR STRESS =E12.4/31H JOINT RESIDUA
1L ~ORMAL STRESS=E12.4,
C
C JOINT STIFFNF.:SS WILL HAVE BEEN READ FROM TAPEC7, IF THIS IS AN
C ACCELERATED RESTART. IF THE NORMAL RESTART WAS USED THIS SUBROUTIN=
C WILL NOT HAVE BEEN ENTERED
C
IFCITER1.GT.l) GO TO 95
C
C CALCULATE JOINT NORMAL STIFFNESS - EQUATION 44A
C
c
C NORMAL STIFFNESS MUST LIE WITHIN LIMITS 0.01 aU/VMC AND 100 QU/VMC
C
SLOMIN=-O.01*CC1,l,MAT)/CC2,2,MAT)
SLOMAX = 10000.*SLOMIN
IF (KJNCM'.GT.SLOMAX' KJNCM'=SLOMAX
IFCKJNCM'.LT.SLOMIN) KJN(M)= SLOMIN
C
C RECOVER JOINT SHEAR STIFFNFSS
C
KJSCM'=C(3,1,MAT)
c:
C ~3W SET UP JOINT STIFFNESS IN LOCAL COORDINATES - EQUATION 35
C
95 KSO = KJS(M'*L/4.0
KNO = KJN(M'*L/2.0
DO 100 J=1,8
DO 100 1=1,8
KSUBCI,Jt=O.O
100 CONTINUE
C
C NOTE THAT THE JOINT STIFFNESS MATRIX IS WRITTEN OUT IN FULL.
C THIS PERMITS READV ALTERATION TO A NON SVMMFTRIC MATRIX.
C SINCE A ST~AIGHT INVERSION OF THE GLOBAL STIFFNESS IS CARRIED OUT
C SUCH MODIFJCATIONS wOULD CAUSF NO PROBLEMS
C
KSUB C 1 , 1 , = KSO
KSUB(3,1)= KSO
KSUB(5,ll=-KSO
KSU9(7,1,=-KSO
KSUA(1,3'= KSO
KSURC3,3)= KSO
KSUBC5,3)=-KSO
KSUBC7,3,=-KSO
KSUBCl,5,=-KSO
KSUB(~,'5'=-KSO
KSUA(5,S'= KSO
KSUAC7,5'= KSO
KSUA(1,7'=-KSO
Appendix 1 409
KSUB(3,7)=-KSO
KSUA(5,7)= KSO
KSUB(7,7)= KSO
KSUB(2,2)= KNO
KSUB(2,8,=-KNO
KSUB(4,4,= KNO
KSUB(4,6,=-KNO
KSU[316,4)=-KNO
KSUA(6,6)= KNO
KSUB(8,2)=-KNO
KSUB(8,B)= KNO
C
IF (INTER .EO. 0) WRITF (6,998) «KSUB(I,J),J=I,8),I=I,a)
998 FO~MAT 137HO JOINT STIFFN~SS - LOCAL COO~OtNATES/(2X,8E12.4»
C
C
C PER~ORM MULTIPLICATION TO TRANSFORM FROM LOCAL TO GLOBAL COORDS
C ECONO~IES ACHIEVED OV AVOIDING ZEROS IN J7A
C USE OF SVMMETRV OF STIFFN~SS MATRIx IS AVOIDED HERE
C TRANSFORM IN 2*2 BLOCKS AT A TIME
C FOQ"1 (K,*(T) IN EQUATION 38
C
DO 200 1=1,4
IK=I+I
II=IK-l
('10 200 J=1,4
JK=J+J
JJ=JK-l
Tl1=KSUB(I I ,JJ)*COSA-KSUB( I I,JK)*SINA
T 12 =K SU B ( I I, J J ) * S I NA+KSUB I I I, JK ) *COSA
T21=KSUB(IK,JJ)*COSA-KSU[3(IK,JK)*SINA
T22=KSUB(IK,JJ)*SINA+KSUA(IK,JK)*COSA
C
C PR~MULTIPLY 8V (Tl TRANSPOSF EQUATION 38
C
KSUB( I I,JJ)=COSA*Tll-SINA*T21
KSU8(II,JK)=COSA*T12-SINA*T22
KSU8ITK,JJ)=SINA*Tl1+COSA*T21
KSUB(IK,JK)=SINA*T12+COS~*T22
200 CONTINUE
C
C ~OW CALCULATF LOA~S COR~ESPONDING TO RESIDUAL STRESSES
C CHANGE TO EFFECTIVE NORMAL STRESS AV ADDING WATER PRESSURE
C TRANSFORM TO GLOBAL COOROINATES IMMEDIATELV - EQUATION 40 COMPLETED
C
SIGMA=RESID(2)
L=L/2.0
Fxt =-L*TORSN*COSA + L*SIGMA*SINA
FYI =-L*TORSN*SINA - L*SIGMA*COSA
F(I)=-FXI
F(2,=-FVI
F(3,=-FXI
F(4)=-FYI
F(S)= FXl
F(6,= FVI
F(7)= FXI
F(B)= FVI
C
IF (INTER .EQ. 0) \II"HTE (6,997) M,(KSUA(I,J),J=I,B),FII),I=l,B'
997 FORMAT (61HO JOINT STIFFNESS AND FORCE VECTOR - GLOBAL CO OROS. - E
lLEMFNTI3/(2X,8EI2.4,EI4.4')
c
410 Appendix 1
RETURN
END
SUBROUTINE STRFS
c
C SUBROUTINE TO CALCULATE SOLID CLE~ENT STRESSES FRo~ DISPLACE~ENTS
C
COMMON /MFSH/ X(2S),Y(2S),R(SO),U(50),KS(SO,50),
1 NOD(2i,4"RESI0(23,3),SIGNO(23,3),~TYPEe23),
2 NODES,NELE~,NSOL,NFREE
REAL KS
COMMON /FORC/ KSUB(8,8),F(8),LBAO,M,H(3,6)
REAL KSUA
DIMENSION DISP(6),OUMMY(3),XT(4',YT(4),SIG(6)
c
C WRITE HEADING FOR PRINTOUT OF STRESSES
C
WRITEe6,20CO)
c
C CALCUL~TE STRESS IN EACH SOLID ELEMENT IN TURN
C
00 40C M=I,NELFM
C
C IS IT A SOLIO ELEMENT OR A JOINT
C
MAT=MTYPE(M)
IF (MAT .GT. NSOL) GO TO 400
("
C ASSEMBLE VECTOR OF NODAL CUORDINATES XT AND YT
C ALSO ASSEMBLE VECTOR (DISP) OF ELE~ENT NODAL D[SPLACE~ENTS
C
DO 250 [-:: 1, J
11=1+1-1
N= NOD ( ,14, I )
JJ=N+N-l
XT(I,=X(N'
YTn )=Y(N)
DISP(lI)=U(JJ)
DISP( 11+1 )=UeJJ+l)
250 CONTINUE
C
C NOW CALL SUBROUTINF TRIA TO FORM ELEMENT STRESS-DISPLACEMENT
C TRANSFORMATION MATRIX -(H)
C DUMMY IS USED TO FILL UP UNUSED PART OF THE PARAMETER LIST
C
C
C NOW CALCULATE STRESSES - ADDING INCREMENT TO RESIDUAL STR~SSES
C EQUATION 25 IN TEXT
C
DO 300 1=1,3
SIG( I )=RES[D(M,I)
DO 30e J=1,6
SIGC I )=SIG( [)+He I,J)*OISP(J)
300 CONTINUE
C
C CALCULATE PRINCIPAL STRESSES FROM XX(SIG(IJ),YY(SIG(2»,XY(SIGe3»
C
CC=(SIGC1)+SIG(2»/2.0
BB=(SIG(1)-SIGC21)/2.0
Appendix 1 411
CR=SQRTCAB*BA+SIG(3'*SIG(3»
SIG(4)=CC+CR
SIG(S)=CC-CR
SIG(6)=28.6q~*ATAN2(SIG(3,,8B)
C
C WRITE OUT ELEMENT STRESSES
C
WRrTE (6,2001) M,SIG
400 CONTINUE
c
RE TURN
C
2000 FOP.~AT (24HO SOLID ELEMENT STRESSES//tOX,7HELE~~NT6X,q~XX-STRESS7X
I,QHYY-STRESS7X,9HXY-STQESS6X,43~MAJOR PRINCIPAL MINOR PRINCIP~L Oq
2 IE I'IT A T I ON)
2001 FORMAT (It7,SEI6.4,FI4.2)
C
END
SUBROUTINE JSTR
C
C SUBROUTINE TO CALCULATE RESPONSE OF JOINTS TO NODAL DISPLACEMENTS.
C INCREMENTAL LOADS FOP NEXT ITERATION ARE AUTOMATICALLY CALCULATED.
C VECTOR R WILL FINALLY CONTAIN INCRE~ENTAL LOADS READY FOR
C CALCULATION OF NEW INCREMENTAL DISPLACEMENTS.
C
COM~ON /MESH/ X(2S),YC2S),R(SO),UC50),KSCSO,50),
1 NOD(23,4),RESID(23,3"SIGNO(23,3),~TYPE(23),
? NODES,NELEM,NSOL,NFREE
REAL KS
COM~ON /STFF/ KJS(23),KJN(23),KJSNU(23),KJNNU(23),ITER1,ITE~N,IT
REAL KJS,KJN,KJSNU,KJNNU
COMMON /CONS/ RO(10),ACELX,ACELy,NRES,INTER,C(3,3,1~)
DIMENSION LAB(4),U5N(B),CJ(9,lO),F(B)
EQUIVALENCE (CJ( 1,1) ,C( 1,1,1»
REAL L,LHLF,KSO,KNO,MO
C
WRITE(6,2000)
C
C CONVERSION CONSTANT FOR DEGREES TO RADIANS
C
CONV= ATAN(I.0)/45.0
C
C TAKE JOINTS ONE AT A TI~E
C
DO 800 M=l,NELEM
C
C SELECT JOINT TYPE
C
MAT=~TYPECM)
c
C IS IT A JOINT - IF NOT GO TO NEXT ELEMENT
C
IF (MAT .LE. NSOL) GO TO 800
C
C RECOVER BASIC JOINT INFORMATION READY FOR CALCULATION
C
I I =NOD (M, t )
JJ=NOO(M,2)
412 Appendix 1
DX=X(JJ)-X( II)
DY='t'(JJ)-'t'(II)
C
C JOINT LENGTH AND ORIENTATION
C
L=SORT(DX*DX+O't'*D't"
COSA=DX/L
SINA=DY/L
C
C POSITION VECT3R USFD TO IN TRANFER OF DATA TO AND FRO~ GLOBAL VECTORS
C
no 200 1=1,4
J=NOD (M, I )
JJ=J+J-l
LAB([)=JJ
11=1+1
C
C TRANSFORM OISPLACEMENTS TO LOCAL COORDINATES
C
USN(I[-l'= U(JJ)*COSA+U(JJ+t )*SINA
USN(II) =-U(JJI*SINA+U(JJ+l)*COSA
2CC CUNTINUE
C
IF (INTER .EO. 0) WRITE (6,999' USN
999 FORMAT (47HO JOINT OISPLACF~ENT VECTOR - LOCAL COOROINATFS/
1 (2X,8F12.4»
C
C
C JOINT MAXIMUM CLOSURE - FOUATION 44
C
VM = -CJ(S,MAT)
IF(RESID(M,2).LT.O.1 VM = -CJ(S,MAT).CJ(6,MAT)/RES{O(M,2'
C
C JOINT SHEAR AND NORMAL STIFFNESSES
C
KNO KJN(M)
KS[) KJS(M)
C
C RATIO OF RESIDUAL TO PEAK SHFAR STRNGTH
C
80 = CJ(4,MAT)
C
C COMPPESSIVE STRENGTH OF WALL ROCK
C
au = CJ(l,MAT)
C
C RATIO OF COMPRESSIVE TO TENSILE STRENGTH LADANYI ANO ARCH TERM N
C
RAT 1/CJ(2,MATI
C
C TANGENT OF ANGLE OF FRICTION OF A SMOOTH JOINT
C
TANTH = TAN(CJ(7,MAT)*CONV)
C
C TANGFNT OF DILATATION ANGLE - ZERO CONFINING PRESSURE
C
TANI = TAN(CJ(B,MAT)*CONV)
C
C CALCULATE THE JOINT DEFORMATION - TOTAL DISPLACEMENTS - EQUATION 28,29
C
UO (-USN(1)-USN(3,+USN(S'+USN(7,)/2.0
VO (-USN(2)-USN(4)+USN(6)+USN(8')/2.0
W (USN(2)-USN(4)+USN(6)-USN(8»/L
Appendix 1 413
C
C CALCULATE TOTAL STRESS THIS ITERATION - EQUATION 3211.,25,33
C TNS=SHEAR STRESS, SN=NORMAL STRESS, MO=MOM~NT
C
TNS Kso*un +SIGNO(M,I)
SN KNO*VO +SIGNO(M,2'
MO L*L*L*KNn*W/~.O+SIGNU(M,3'
C
WPITE(6,2001IM,UD,VO,W,TNS,SN,MO
c
C ~DW TAKE EACH END OF THE JOINT IN TURN AND CH~CK THE SHEAR AND
C NORMAL FORCES FOR EACH NODAL PAIR. FIRST I,L NODES ,THEN J,K
C VAqlABLE SIGNE IS USED TO CONTROL THE SIGN OF TERMS IN THE B MATRIX
C (EQUATION 31' THIS MA TRIX IS NOT SFT UP BUT THE RELATIONSHIPS IT
C DEFINES ARE USED
C FIRST SET L TO HALF LENGTH OF JOINT
C
LHLF = L / 2.C
SIGNE ::: 1.0
TORM =0.0
FNOM O.C
FNIM 0.0
C
00 550 I::: 1,2
SIG NE ::: -SIG NE
c
C CALCULATE NODAL FORCES FOR FIRST,PREVIOUS AND CURRENT ITERATIONS
C (EQ 31 IN TEXT) THESE ARE EXTERNAL FORCES AT L AND THEN AT K
C
C
C F~IL NOR"'IAL FORCE FOR LAST ITERI\TION
C FSO INITIAL SHEAR FORCF
C FNO INITIAL NORMAL FORCF
C FSI CURRENT SHEAR FORCE
C FNI CURRE'NT NORMAL FORCE
C
FNIL SIGNO(M,21*LHLF + SIGNE*SIGNO(M,3)/L
FNO RESID(M,2)*LHLF + SIGNE*RESIO(M,3'/L
FSD RES I D ( M, 1 I * L HL F
FSI TNS * LHLF
FNI SN*LHLF + SIGNE*MO/L
C
C NORMAL FORCE PEq UNIT AqEA FOR THIS HALF OF THE JOINT IS SIGMA
C
SIGMI\ == FNI/LHLF
c
C SHEAR STRENGTH ETC. SET TO ZERO IF JOINT IN TENSION
C
IF(SIGMA.LT.O.OI GO TO 305
TORP==C.O
TORR=O.O
TOR =0.0
UP =0.0
UR =C.C
DILAT=J.O
GO TO ~60
C
C CA.CULATE PEAK SHEAR STRENGTH - EQUATIONS I~ TO 16 OF CHI\PTER 5
C F IRST CHECK IF NDR"'IAL LOADING IS ABOVE THE COMPRESSIVE STRFNGTH
C
305 IF (au .LT. SIGMA' GO TO 320
C
C TRANSITION STQESS EXCEEDED
414 Append~ 1
C
AS=l.O
VDOT=O.O
C
C LIMIT SIGMA/au RATIO TO 1.0 SO THAT PEAK SHEAR STRENGTH IS LI~ITED
C
RATIO = 1.0
GO TO 330
C
C BE_OW TRANSITION STRESS - EaUATIONS 16A AND 16~ OF CHAPTER 5
C
320 RATIO=SIG~A/aU
AS=1.C-SQRT«l.C-RATIO)**3'
VDOT=(1.0-RATIO'**4*TANI
C
C EVALUAT~ EQUATION 14 OF CHAPTER 5 TORP IS PEAK SHEAR STRENGTH
C
330 TERMl SIGMA*(l.O-AS)*(VnOT+TANTH)
TER~2 AS*QU*(SQRT(RAT+l.O)-1.0'/RAT*saRT(1.O+RAT*RATIO)
TERM3 1.O-(1.0-AS,*VDOT*TANTH
TORP=AAS(TER~1+TERM2'/TER~3
C
C IF TRANSITION STRESS EXCEEDED THEN NO PEAK-RESIDUAL BEHAVIOUR
C TORR = RESIDUAL SHEAR STRENGTH FROM EQUATION 60
C
TORR = TORP
IF (QU .LT. SIGMA) TORR = TORP*(Bn + (I.C-80)*SIG~A/aU'
C
C SE~ECT PEAK AND RESIDUAL SHEAR DISPLACEMENT - EQUATION 54 AND 54 A
C SIGN OF DISPLACE~ENT DETERMINES THE EQU4TION- SET SEL TO THE SIGN
C
AUO AAS(UO)
SEL UO/AUO
C
C UP PEAK SHEAR DISPLACEMENT
C UR RESIDUAL SHEAR OISPLACE~ENT
C
UP (SEL*TORP-RESID(M,l)}/KSO
UP (SEL*4.0*TORP-RESID(M,1»)/KSO
C
C SELECT RANGE TO WHICH SHEAR DISPLACE~ENT BELONGS
C AND CALCULATE LIMITING SHEAR STRESS CORRESPONDING TO DISPLACEMENT EQS5
C
IF (AUO .GT. ABS(UP') GO TO 400
C
C IN LINEAR R~NGE DFFINED-BY SHFAR STIFFNESS - RANGE III
C
TOR = KSO*UO + RESID(M,I)
GO TO 450
C
400 IF (AUO .GE. ABS(UR)' GO TO 430
C
C IN FALLING PORTION - RANGE II OR IV
C
TOR = SEL*(TOPP-(UO-UP}/(UP-UP'*(TORP-TORR)'
GO TO 450
C
C IN RESIDUAL PORTION - RANGE I OR V
C
430 TOR SEL*TORR
AUU ABS(UR)
C
C NOW CALCULATE DILATION - FIRST SET AUO TO ABS(UR' IF SHEAR
Appendix 1 415
C BFHAVIOUR NON - L INe AR - SEE EQU ATION 62 (WIL L BF. AAS ( UO ) IF NOT ~ESE T
C
450 DI L AT (F Nf L/ (L HLF *Q U ' -1.OI **4 *TA N I *( AU O+A 8S (F SO )/(KS O*LHLF »)
C
C CAL CULATE SHEAR FO RC E TO RE APPLIED NEXT ITERATI ON -EQU AT I ON 57
C
46 0 OELS TOR*LHLF - FS I
C
C CALCULATE WHE T HER NODAL PA IR OPE N ING OR CL OS IN G
C SET DELV TO TOT AL NORM AL DISPLACE~E NT - E aUATI O~ 4 7 A
c
I1= Y+l
IJ= 10- 11
DEL V= - USN( It '+ U5 N( [J)
I F ( DEL V) 520,5 10 ,500
C
C JOINT OPENING - EQUATION 47A
C
5eo DEL N = CDEL V/ (V M- DE LV ) + 1 .0 , *FN O - F~ f
GO TO 530
C
C JOINT UNCH AN GED
C
510 DE LN = 0.0
GO TO 530
C
C JO IN T CLOSI NG - EQUATION 50A
C
5 20 DE LN (DELV -V M* <FNI- F NO)/FNI ' *KNO*LHLF
C
C C O~PU TF EXTER NAL FORCES ON JOINT Ea UIV~ LEN T TO INCR E ASE OF RE SIDU_L
C NO RMAL STRESS - CA USED 3V DILA T I ON
C
530 DILN - b l~ A T* F NI/(V~)
C
C NOW TRA NSF ER NO DA L F ORCES TO J OI NT FO RC E VECTOR
C
F (I I) = -OEL N-OILN
F( JI -1 ) = -DELS
Fe I JI = DELN+-D ILN
FCIJ-l):;:DF.LS
C
IF (INTFR .E O.~) WR IT E (6,998) lIP, UR,TORP,TOR R ,TOR ,I) ILA T, DE LV
99B rOR MA T (5HO UP=E 1 2. 4 ,4H UQ = E1 2. 4,6H TORP =E 12 .4, 6H TORR = E1 2.4.
1 5~ TO R=E 12.4 / 8H DILA T= E12. 4 ,6H DELV=E12.4)
C
C THE F OLL OWING ARE U SED TO S AV E MEAN SHEAR STRESSFS AN D NOR MAL Fa RCE S
C FO R CALCUL~TION OF NE~ JO I NT STIFFNESSES
C
TO RM TO RM. T OR/2.0
F NIM FN IM + F~I
F NUM FNOM +- FNO
S S O CO NTI NUE
C
IF ( IN TER .EO. 0) WRITE ( 6 t 9CJ7) F
997 FOR MA T (30HC INCREME NT AL FORCES F OR J OI NT/ l X, 8F. 12 .4 )
c
C
C CA L CULATE I NI TIAL S TR E SSE S CORRESPONDI NG TO P REVIOUS PLUS
C I NCREMENT J UST C ALCULATEO ( EQUATIONS 30 I
C L IS BA CK TO FULL JOINT LENGTH
C
S IGN O ( M,l ) SIGNO( M,U +-(F( 5)+ F(7 1. /L
416 Appendix 1
conversion factors
I·
L~ ...
I ~
! LENGTH
1 inch (in) 25.4 millimeters (rnrn)
f., 1 foot (ft) 0.3048 meters (m)
1 mile 1.60934 kilometers (Km)
AREA
2 4 2
1 in 6.4516 x 10- m
2
1 ft2 0.092903 m
2
1 acre 4,046.86 m
I VOLUME
t' 1 fluid ounce (U. S. ) 2.95735 x 10- 5 m
3
1 liter 1 x 10- 3 m3
3 3
1 U.S. gallon 3.7854 x 10- m
3 3
1 U.K. gallon 4.5461 x 10- m
FLOW RATE
3
1 ft 3 /sec 0.028317 m /sec
-5 3
1 U.S. gallon/min 6.30902 x 10 m /sec
418 Appendix 2
FORCE
1 dyne 1 x 10- 5 newtons (N)
1 ounce 0.278014 N
1 pound (lb) 4.448222 N
1 kilogram (Kg) 9.80665 N
1 U.S. ton 8.89644 KN
PRESSURE
2 47.88026 N/m2
1 pound/foot (psf)
1 foot of water (at
60 degrees F) 2.9861 KN/m 2
1 pound/inch 2 (psi) 6.89476 KN/m2
2
1 Kg/cm 0.0980665 MN/m2
2
1 bar 0.1 MN/m
8 Agarwal, R.K., and Boshkov, S.H. ( 196 9), "Stresses and d isplac e-
ments around a circular tunnel in a three layer medium" I JRM&M S, V6 n 6
8 Baker, L.E., Sandhu, R.S., and Shieh, W.Y. (1969), "Appli c a tion
of elasto-plastic analysis in rock mechanics by the finite element
method, Proc. 11th Symp. on Rock Mech . , p237
6 Banks, D.C., and Strohm, W.E. (1974), "Calculations of r o ck
slide velocities" Proc. 3rd Congo ISRM, Denver, V2B, p839
4 Barr, D.J. (1969) "Use of side looking air -bo rne ra dar (SLAR)
imagery for engineering soils studies", U.S. Army Eng. Topog. Lab . ,
Ft. Belvoir, Va., Tech. Report 46-TR (AD 701-902)
Goodman-G eol ogical Eng. - 28
420 References
4 Bolstad, D.D. and Mahtab, M.A. (1974), "A Bureau of Mines direct
reading azimuth protractor", U.S.B. Mines Inf . Circ. 8617
5 Bray, J.W. (1967), "A study of jointed and fractured rock" Rock
Mech. and Eng. Geol., V5, n2 and n3
2,5 Brekke, T.L. and Howard, T.R. (1972), "Stability problems caused
by seams and faults" Proc. 1st Rapid Excav. and Tunneling Conf. (AIME)
V1, p25
2 Broch, E., and Franklin, J .A. (1972), "The point load strength
test" IJRM&MS, V9, p669
8 Call , R.O. (1971) , " Slope st ability study of the Tazadit pit,
Mauritan i a " Proc. 2nd Int. Coni. on Stability f or Open P i t Mines,
Vancouve r ( AIME)
5,7 Chappel , B.A. (1972) , uThe mechanics of blocky mat eri al" PhD
thesis , Australian Nationa l Vniv ., Canberra
5 Chenevert, M.E. and Ga t li n,. C. (1965) , " Mechani cal aniso tropies
of l a minat e d s edi me n tary rocks '" SPE J o urn al ( AIME ) , V5, p67
8 Chris t iansen, L.M. , Mis te r e k, D .L., and Bowl e s, G.F . ( 1971 ),
"Foundat i on analysis of Auburn da m s ite" F roc. Int. Symp. on Ro c k
Fracture, Nancy (ISRM) pap er 2-26
8 Cundall, P.A. (1974), "A computer model for rock mass behavior
using interactive graphics", U.S. Army, Corps of Engineers, Technical
Report MRD 2-74 (Missouri River Division)
8 Dahl, H.D. (1972), "Two and three dimensional elastic and elasto-
plastic analyses" Proc. 5th Int. Strata Control Conf., London
2,4 Deere, D.U. (1963), Tech. description of rock cores for eng.
purposes", Rock Mech. and Eng. Geol., V1, p18
2,5 Deere, D.U., and Miller, R.P. (196 6), "Eng. cl as s ification and
index properties for intact rock" Ai r Force Weap o ns La b Tech Rep . AF
WL-TR-65-116
8 Dejean, M., Mart in , F., and Ra f fou x, J .F. (~97 0 ) , "Deformat i ons
of rock galleries during driving" (i n Fren ch ), P r o c. 2nd Congo ISRM,
Belgrade, V2, paper 4-36
8 Desai, C.S., and Reese, L.C. (19 70), ' rS tr e ss- d e f orm a t ion a nd
stabili ty analyses of deep boreholes", Pr oc . 2 nd Co ng . ISRM , (Belgra d e ,
V2, paper 4-13
8 Desai, C. S., and Abel, J.F. ( 1972 ), "Intra. to the finite element
method" (Van Nostrand Reinhold )
3,4 Donn, W.L., and Shimer, J.A. (1958), "Graphic methods in struc-
tural geology" (Appleton-Century-Crofts)
4 Fanshawe, H.G., and Watkins, M.D. (1971), "A sparker survey used
to select a marine dam site in Hong Kong", Qtly J. of Eng. Geol, V4,p25
References 427
2 Franklin, J. A., Brach, E., and Walton, G . (1971 ) "Lo gg ing the
J
6 Gallico, A., and John, K.W. (1974), "Graphi c a l s tabili t y an aly ses
for rock abutments of an arch dam" Proc. 3rd Congo ISRM , Denver, V2 B ,
p884
428 References
8 Goodman, R.E. Taylor, R.L., and Brek ke, T. L . (1968), "A model for
the mechanics of jointed r ock ", J .SMFD ASCE, V94 , p637
8 Heuze, F.E., and Goodman, R.E. (1970), "The design of room and
pillar structures in competent, jointed rock, example: the Crestmore
mine, Cal.", Proc. 2nd Congo ISRM, V2, paper 4-41
1 Hoek , E., and Bray , J. ( 1974 ), " Rock s l ope e n g i nee ring II (I MM,Lon )
2,6 John, K. W. (19 62 ), "An appro ach to rock mechan i cs " J SM&FD ASCE
V88, p1
6 John, K.W . ( 1 9 70) , "Three dime nsion al s t a b ility ana l ysi s o f sl o pes
in jointed rock" Pr o c . Johan ne sb u r g Syrnp o n Ro ck Mech , South Africa
5,7 Kandau rov , I. I., Ouva r ov, L.A. , a nd Karpov , N .M . ( 1974) , I f Con-
trainte dans les mo deles des mas s if s r o che ux fi s sure s s ans pousse e
horizontale" Pro c . 3rd Cong o ISRM,De nve r, V2A, p157
4 Krumbein, w.e., and Graybi ll, F.A. ( 1 9 6 5), !IAn i ntroduct ion to
statistical models in geolo gy" (M cGraw-Hill )
V3, paper 8- 2
4,2 Leet, L.D., and Judson, S.(1971), "Physical geology, 4th edi t i on"
(Prentice-Hall)
2,5 Martin, G.R., and Millar, P.J. (1974), "Joint strength character-
istics of a weathered rock", Proc. 3rd Congo ISRM, Denver, V2A, p263
5 Men el, V. (1965 ), "Di la t ancy of rocks", Rock Me ch . and Eng. Geol.
V3, p5 8
4 Meri fi eld, P .M ., et. al. (1969 ), "Satellite imag ery of the e arth"
Ph otogrammetr ic Eng . V35, p654
4 Moye, D.G. ( 1 96 7), "Diamon d dri ll ing for foundati on exp l or a t ion" ,
Trans. Inst . o f Enginee rs, Au stra li a, V9, p 95
1 Obert, L., and Duvall, W. (1967), "Rock ~echanics and the design
of structures in rock" (John Wiley)
2,5 Piteau, D.R. (1973), "Charact eriz in g and ext rapo lating rock
joint properti es in e ng ineering practi c e", Ro ck Mech. Sup p . II, p5
5 Rale i gh , C.B., and Pate rson, M.N. (1 96 5), " Exp erime ntal deforma-
tion of serpe ntine and i ts tectonic i mplica t ion s" , J . Geoph . Res .
V70, p396 5
4 Ray, H.G. ( 1960), "Aer ial pho tog r aphs in geo lo g ic i n t erpret at ion
440 References
Proc. Int. Symp. on Rock Mech., Madrid (ISRM), paper 2-5, p159
2 Sahores, J. (1962), "Con tribut ion to the study of the mech. phen.
accompanying hydration of anhydrite" (in French), Centre d'etudes et
de Recherche de L'Industries des Liants Hydrauliques, Pub. Tech. n126
5 Serafim, J.L., and Lopes, J.B. (1961), "In-situ shear tests and
triaxial tests on foundation rocks of concrete dams l l , Proc. 5th Int.
Conf. on Soil. Mech. and Found. Eng., Paris, V1, p533
on Salt, Rau and Delwig, eds. (Northern Ohio Geol. Soc., Cleveland)
V2, p249
3 Shanley, R.J., and Mahtab, M.A. (1974), "A computer program for
clustering data points on the sphere", U.S.B.Mines Inf. Circ. 8624
4,5 Snow, D.T. (1965), "A parallel plate model of permeable fractured
media", PhD Thesis, Univ. of California, Berkeley
4,5 Snow, D.T. (1968), "Rock fracture spacings, openings, and porosi-
ties" J.SM&FD ASCE, V94, p73
4,5 Stimpson, B., Metcalfe, R.G., and Walton, G. (1970), "A new
technique for sealing and packing rock and soil samples" Qtly J. Eng
Geol. V3, p127
2 Vargas, M. (19 5 3), "S ome enginee r ing prope rti es of r e sidual clay
soils occurring in Southern Brazil", Proc. 3rd I nt . Co nf. on Soil Mech.
and Found. En g . , Zu rich, V1, p67
4 Voloshin ! V., Nixon , D . D., and Timbe r lake , L . L . ( 1 968) , "Orien ted
core, a new technique in engineering geo logy", Bull . A . E .G., V5, p37
4 Warren, C.R., Schmidt, D.L., Denny, C.S., and Dale, W.J. (1969),
"A descriptive catalog of selected aerial photographs of geologic
features in areas outside the U.S.", U.S.G.S. Prof. Paper 591
8 West J L . J., and Per ry, R. M. (1969), "Rock mechanics studies for
a high cut at a nuclear power s tation in Penn." Proc. 11th Symp. on
Rock Mech.
4 Wiebeng a , W.A. and Polak, E.J. (1969), flGreat Lake Nor t h en g in.
Geophys i ca l surveys, Tasmania, 1951-1959", Australia Dept. of National
Developme n t , Bureau o f Min. Resources, Geol, and Geophysics, Bull 101
8 Zienkiewicz, O.C., Watson, M., and King, I.P. (1968), "A numeric-
al method of viscoelastic stress analysis", IJRM&MS, VI0, p807
ABBREVI AT IONS
amphibole, 35
anhydrite, 24
aperture, 44
argillite, 42
banding, 15, 40
blasting, 8, 9
cement, 18, 46
chalk, 19, 48
Chambishi mine, 9
chert, 19, 33
claystone, 29
cleavage, 40
comminuted rock, 44
common excavation, 23
"consolidated" ro ck, 2 0
contacts, 40
cross-hole shooting , 12 3
decomposed rock, 20 , 2 1
- importance, 2, 47
- in models, 2 90
- measurement, 166
dilatometers, 157
dipmeter, 154
- laboratory, 165-168
discontinuities - aperture, 44, 153
- filling material, 44
- types, 40
drainage, 10
Subject Index 455
- cos t , 127
- methods, 129- 1 33
durabil i ty , 27-30
e xcavations - surface, 5
f ac t or o f safe ty - de f inition , 24 1
Fisher's distribution, 86
- defined by spacing, 44
- porosity, 33-36
form ation s, 47
granite, 15, 18, 22-26, 31, 33, 3 7, 39, 4 2 , 47 -48, 103, 1 23, 1 84
graphi te, 5, 46
gravi~y surveys, 1 26
Griffith theo r y, 32
g r outing, 5, 10
hematite, 127
hydrothermal alteration, 22
458 Subject Index
- columnar, 2, 43-45
- incipient, 174
- samples, 158-160
- sheeting, 44-45
Karst, 24
Kukuan dam, 5
Lambert net, 65
latite, 44, 45
limestone, 20, 24, 37, 39, 42, 48-49, 129, 184, 295
magnetite, 127
minerals, 19, 35
mining, 4
opal, 19
pegmatite, 33
petrology, 16-19
pitch angle, 76
porosity, 126
prototype, 277
pyrite, 19
pyroxene, 35
Subject Index 461
radar, 110
ripple marks, 42
- geological, 16-19
RQD, 51-52
seams, 174
Seisviewer, 154-155
sheared zones, 40
sheet joints, 44
soils, 2, 19-22
Subject Index 463
s pillways , 8
squeezing g round , 46
- friction, 237-2 4 0
- stereonets, 62-68
Stockton dam, 9
stresses - initial, 8, 10, 15, 18, 49, 270, 317-318, 329-330, 367-368
strike, 96-97
swellability, 29
swelling clay, 46
talc, 19, 46
Televiewer, 154-155
trachyte, 175
Vajont reservoir, 5
- profiles, 27
- slaking, 1, 28-29
zeolite, 19
author index
Barton, N., 50-54, it76, 180, 181, Bray, J., 8, L2, 159, 166, 169, L70 ,
r9o, 191, 196, 289, 29O 193, 2OL-2O5, 2I3, 2L4, 234, 28L
BerentS , H. P. , l-37 Brekke, T. , 29, 46, l4O, l-75
Bergh-Christensen , J., 50 Broch, E., 30
Chenevert, , 2O5
M. Everling, 2gg
Clough, R. W., 308 Fagerstrorn, H. , 191
Cluff, L. S., 111, 154 Fairhurst , C. , 189, 37 3
Kiersch, ,
G. 26 Lunde, J., 50-54
King, R. F., L24 McCracken, D. D., 334, 335
470 Author lndex
Nascimento, U. , 29 , 183 , 193 , 194 , Rosengren, K., 143, L49, 160, L62,
199 163, , L73 , 183 , 208
L64
Savage , J. F. , 1 19 Treashe r, R . , 26
Tab 0 r, D. , 1 71 We i s s, L., 8 3 , 19 6
Terzagh i , K. , 2 0 , 21 Williams , H. , 1 6
472 A u thor Index