Design Guidelinesfor Open Stope Support
Design Guidelinesfor Open Stope Support
Yves Potvin graduated from l'Univer- 1988 and is now the geornechanics program leader.
site Laval in 1982 with a [Link]. in min-
ing engineering and worked until 1983 Marty Hudyma graduated from The
as an underground mine planning en- University of British Columbia in 1986
gineer at Mines Gaspe in Murdochville, with a [Link]. (mining and mineral
Quebec. He attended The University of process engineering). In 1988, he gradu-
British Columbia from which he ob- ated from the same institute with an
tained an [Link]. in 1985 and a Ph.D. [Link]. in mining engineering (rock
in 1989. His research work was funded mechanics).
by NSERC, Noranda Technology and Currently, he is working for the
Falconbridge Ltd. and focussed on em- Centre de Technologie Noranda, relat-
J pirical design methods applicable to ing seismicity in mines to geomechan-
Canadian open stope mining condi- ics and mines design.
tions. Yves Potvin joined the oranda Technology Centre in August
plex interaction of the cable bolts with the ground conditions and Stiffness of the Support System
opening geometry will be looked at in a simplified empirical man- The stiffness of a material represents its capacity to deform
ner using the open stope stability model (Paper No. 44, 90th when submitted to stress. Stiff materials show small deforma-
Annual General Meeting of CIM, Edmonton, Alberta, 1988). tion before failing and are prone to violent failure. Materials hav-
ing a low stiffness may sustain large deformations before
collapsing. In order to obtain the maximum advantage of a cable
Design Concept bolt support system, its stiffness must be designed according to
the rock mass stiffness. If the support is too stiff, the rock mass
Prereinforcement will be restrained from deforming but the energy usually dissi-
Originally the support philosophy was to suspend the loose pated through deformation will build up and could cause the sud-
rock in the periphery of the excavation to the more competent den failure of the support system. However, if the stiffness of
and undisturbed layers remote from the opening surface. A bet- the support is very low compared with the rock mass stiffness,
ter understanding of rock mass behaviour and support systems the stabilizing effect may be insignificant.
have lead to a more efficient technique called prereinforcement. The stiffness of a cable bolt system is dependent on several
The concept of prereinforcement consists of installing the sup- factors. The density of bolting and the length of cables are im-
port prior to excavating the rock adjacent to the excavation. The portant factors and will be discussed below. The bonding strength
principal effect is to limit the displacement of the rock mass to between the grout and rock and the grout and steel also plays
small values. Fuller and Cox (1978) suggested that when prerein- a major role because they constitute the weakest points of the
forced, the rock mass displacement is limited to less than two mil- system. A full discussion on this subject is [Link] scope of
limetres. This minimizes the shear and dilation along existing this paper and can be found elsewhere (Stheeman, 1981; Jerern-
geological structures and preserves their in-situ cohesion and fric- ic and Delaire, 1983). Although there are certain variations in
tion angle. The rock mass becomes "self-supporting". cable bolting practices in Canadian mines, in general Vs-inch di-
Another interesting advantage of the prereinforcement tech- ameter cables are used with a water cement ratio less than 0.5.
nique is that blasting is done against an already reinforced sur- Sometimes additives will be used to reduce the amount of grout
face, reducing the amount of damage caused by blasting vibrations. slumping in the hole. When only a low density of cable bolts
As well, the effect of the sudden changes in the surrounding stress is employed, some operators will double the number of cables
field resulting from the change in opening geometry (after each in each hole. It is important to notice that quality control in the
blast) can be better controlled by a prereinforced rock mass. installation of cables can have a large influence on their strength.
The principal limitation in applying the prereinforcement tech- Fuller (1983) reported that the grout-steel bond can be damaged
nique to open stopes is the lack of access for cable bolt installa- by dirt or debris if the cables are not properly cleaned before
tion. Typically, open stope development is located in the footwall. their installation. On the other hand, the bond strength will in-
The mucking horizon (undercut, drawpoints) is developed first, crease when the cables are slightly rusted. The "bird cage" type
followed by the opening of the drilling horizon (overcut or drill- of cable bolts have been specially designed to improve the grout-
ing drift [s D. The ore between the two horizons can then be cable bond but are not widely used in Canada at this time.
retreated vertically or horizontally (Fig. 1). Contrary to cut-and- The designer may also adjust the stiffness of the cable sup-
fill mining, at no time during this sequence is access available to port system for specific applications, using different techniques
install cable bolts in the overcut and undercut before they are during the installation. These adjustments increase or decrease
opened. the support stiffness.
However, the use of prereinforcement may be attempted by
installing the cable bolts in the overcut as soon as possible after
it is open. A certain amount of displacement is expected to occur, Increasing the Support Stiffness
but if careful blasting is used and the time delay before the sup- A stiffer 'support system may be more efficient in zone of
port is installed is minimized, the open stope back should be effec- stress relaxation where additional reinforcement is required. This
tively supported. The advantage of having the roof reinforced for can be achieved by tensioned cable bolts. At first, tensioned cable
the heavy production blasting remains. This has become the most bolts were applied in all situations. It was later realized that the
commonly employed cable bolting procedure and it has been expansion of the rock mass after excavation in most cases was
proven effective. sufficient to naturally tension the cables. Because it is a time con-
suming technique and in most cases redundant, it is now prac-
About the concept of pre-reinforcernent, Hoek and Brown ticed only in s ecific applications such as clamping layers of
(1980) concluded that: relaxed roe ogether in a hanging wall.
The principal objective in the design of excavation sup-
port is to help the rock mass to support itself. Pre-placed
grouted reinforcing elements areprobably the most effec- Decreasing the Support Stiffness
tive means of achieving this objective and the authors have When a large amount of deformation is expected, it is .desir-
no doubt that the future will see a great increase in this able to design a cable bolt system capable of high deformation
support technique. without losing its support capacity. Matthews, Tillmann and
Worotnicki (1983) described a debonding procedure to decrease
the stiffness of cable bolts. The procedure includes the installa-
Dr. H. Miller is an associate professor at The University of British tion of supplementary anchors (swages) and artificially debond-
. Columbia and has been with the Department of Mining and Mineral ing sections of the cable with plastic tubing or paint. This
Process Engineering since 1980. He received an undergraduate degree minimizes the risk of premature strand failure due to the local-
from Cardiff University and a Ph.D. from the University of Newcastle- ized movement of a joint. Matthews, Tillmann and Worotnicki
upon-Tyne, where he also lectured from 1975 to 1980. In 1973 he be- also reported a successful applicaton of this technique in the case
came project manager of the Rock Mechanics Department of Anglo
American (Western Deeps) in South Africa. of a highly stressed crown pillar where lateral expansion was ex-
Dr . Miller is a Registered Professonal Engineer in the Province of cessive. However, none of the case histories collected for this
British Columbia. study used the above debonding technique.
Primary Secondary
Stope Stope
Length Length
Footwoll
Access
Drawpoint
Spacing
Primary Secondary
Stope Stope
length length
, I
Footwall
Access
Distance
Houloge-Orebody
FIGURE 1. Isometric view of open stope mining methods showing vertical and horizontal retreat.
4 ,
FIGURE 2(a). Uniform cable bolt pattern installed in open stope overcuts. FIGURE 3. Cable bolt support system using inclined cables and two
phases of overcut development for prereinforcement.
// \ \' ,\"'/
FIGURE 2(b). Uniform cable bolt pattern installed in open stope over-
cuts and supplemented with short rebar. FIGURE 4. Cable bolt support system using an interlaced support pattern.
the sides (S) that will be "slashed" during the second stage of drilling drifts are used as the drilling development, cable bolt de-
overcut development. At the only mine using this design, the sign has been done as shown in Figure 5. The original spans to
length of cable bolt used was 9 m with a density of bolting of be supported were very small but will become wider as the stope
0.16 cb/rn-. As in the preceding case, 2.7 m long rebar were uti- is extracted. Once again, this takes advantage of the concept of
lized in between the cables with a density of 0.44 rebar/rn-. prereinforcement. The bolt length observed in these cases were
Another modification of open stope roof bolting is shown roughly 10 m with a density of bolting of 0.2 cb/m-.
in Figure 4. Cable bolts (in this case 6 m long) are installed at
an inclination of 76 degrees in one direction for a given row while
in the next row the cables are inclined in the opposite direction. Cable Bolt Patterns for Open Stope Walls
This alternate inclination produces an interlaced pattern which The first application of cable bolts in stope walls attempted
aims at intersecting geological discontinuities at more favourable to distribute the cables as uniformly as possible over the sup-
angles. The density of bolting achieved in this type of design varies ported wall (Fig. 6). Because of the restricted access, the drill,
from 0.2 to 0.25 cb/m>, holes had to be fanned which does not necessarily produce a
When temporary pillars are left in the overcut, or parallel favourable bolt inclination for support. Only a low bolting den-
/
/
I
/
/
/
I \
/ \
\
/
I
\
\
/
I I
/ /
/
I I I
I I I
I I ------
I I I z
I I '-"
I I
I L
<lJ 100
I I I -D
, I E
I I ::J
I I I Z
I I I >,
I I -+-'
I I I .D
10
I I 0
I I I +'
(f)
I
I I I -0
I I <lJ
1.0
I I I 4=
-0
I I 0
I I :2
I I
I I I
I I I
I I I
I I I
Hydraulic Radius Cm)
I I I •. Stable Stope Surface
I I
I I • Unstable Stope Surface
I
I I 1
.•• Caved Stope Surface
I I
I I [Link]
o Stable Slopo Surface
o Unsiabie Slope Surfoco
c-,
(RQD/Jnlhydraulic radius). It can be noted that most cases plot- "0; 0.15
ting inside this zone are stable. Consequently, it appears that a o
:ii
conservative design guideline for the density of bolting is to use (5 0.10
+---'.--r--~----r--~-___,--_._-___,r__-__1
0.00
o
Cable Bolt Length (RQDjJn)
5
j Hydroulic Radius
The length of cable bolts should reach far enough into un-
disturbed ground to ensure a proper anchor. According to FIGURE 11. Design chart for cable bolt density.
numerical modelling analysis, the disturbance effect of stress in
the rock mass surrounding underground openings is a function.
of the relative size and shape of individual stope surface. Con- approximately equal to the span of opening. It also constitutes
sequently, a rough relationship is expected to exist between the an approximate average of what Canadian open stope operators
hydraulic radius of stope surfaces and the length of cable bolt have been using with success.
used. These two parameters have been plotted in Figure 12, for
the cases of supported backs in the data base, and the following
observations can be made. Cable Bolt Orientation
1. As in the bolt density plot, Figure 12 shows that the data is The design of cable bolt orientation is a three-dimensional
quite scattered. Once again it is believed that this can be attributed problem in which the optimum strike and inclination must be
to the lack of guidelines for support design and a trial and error defined. At the optimum orientation, the cable bolt system should
process. develop a maximum' strength against the forces acting on the
2. The use of cable bolts in very large open stope surfaces has cables. These forces are dependent on the potential mode of
had little success. It can be seen in Figure 12 that for hydraulic failure of the stope surface. For instance, in the case of gravity
radius exceeding ten, seven support systems out of nine have col- fall the principal forces generated on the cables are tensile. The
lapsed. cable bolts should then be oriented vertical in order to maximize
3. The minimum stope plane dimension in which cable bolts have the tensile strength acting against gravity.
been installed has an hydraulic radius of approximately three. For a sliding mode of failure, the shear force acting along
4. The minimum cable bolt length included in the data base is the potential sliding plane is the one that may induce stability
three metres. This was arbitrarily decided prior to the data col- problems. Miller (1984) based on a three-dimensional analytical
lection in order to differentiate cable bolt action and other kinds model suggested that cable bolts develop a maximum shear
of shorter rock anchors. strength when installed in the same direction as the shear plane
5. A rough and conservative guideline can be derived from the but having an inclination of 17 degrees to 27 degrees.
plot of cable bolt length and hydraulic radius. This is shown in The optimum orientation of bolting when dealing with slab-
Figure 12 by the line "L" which corresponds to a cable design bing or buckling mode of failure is perpendicular to the folia-
~ 12 "cl~
o
U la
I. IIQClt qUAllr"Y OESICHATIC-'1 .'IQl
o o steete Stope Surroce A. Very ;lOO" e -::s I. """c'c JilQ.O ;:s t'c~,.ud 0" uured.1S
o 0 0 0 !. 10 ( IM;I~jll'J 0 ). a _i.ul ",.• 1",..
o
v Coved Stopo Surface 3. Poe •• 15 - 51
0' 10 i, •••
,cd CO c .•.• lo,u,te Q.
c. r",. 50 -1S
-----------~--------------- a. Cood 1S -10'
1. iIIQO ill[c,"'.;Ils
e ee $1.I"[Link]
of S.
.•eece
i .•. 100.95.90
ae e •
e ee
'0 - 104
8 10 12 14 16 18 20 22
t. JOLNT srr HU",[R
Hydraulic Radius Cm)
A. ' •• usi ••e , 1"0 or few joint) 0.5 -1.0
a. [Link]"( see 1
FIGURE 12. Design chart for cable bolt length.
r , Thr". jOi"t se t s.
tion because it is the direction of potential movement. The C. Th'ee jo;"t UtJ plus r'ndO'll 12
he .• ,,;'y joil'{"Q
·lu9 .• r [Link]':..· •• tC 'I
The mode of failure can be easily determined using stereo- J. ('u,h.d roc ••••",:1'11;11.. 20
(O'TlQrtll'Wf)lrfU
lIr""Il'" of ••.••oCI 'o'~
whenever it is possible.
The stiffness can have a large influence on the performance FIGURE A2. Compressive stress factor according to Mathews et al.
of a support system. In most applications it is desirable to match (1980).
the stiffness of the support system with the stiffness of the rock
mass. However, for specific applications it might be useful to
increase or decrease the stiffness of the cable bolts. This can be . Difference In Strike
90· 90·
done using special cable bolting installation techniques such as 1.0 ----------------- ~:;:3
pre-tensioning or debonding. -/.
The three principal variables to be designed in a cable bolt 0.9
50·_~::::......-r::
~ /.r:
system are the density of bolting, the length of cable bolts and 50· -...;.-~ //./
L-
their relative orientations. The orientation of cable bolts should 0 0.8 --- 45"// /
be designed according to the possible mode of failure. When U /
.E /
gravity fall or slabbing are anticipated, the cable should be in- ~c:: 0.7 I /
stalled vertically. In the case of sliding, the most efficient design I /
Q)
/ 30;
is when the [Link] installed at an angle between 17 degrees .§
0.6
Vl 45:""'/ /
and 27 degrees to the shear direction. .a // I
/
Based on the compilation of actual Canadian experience, "
« 0.5 ./
I
some rough design guidelines have been proposed for the deter- c::
/
mination of density and length of cable bolts. It has been found .2
0.4 /
.3 /
that a rough relationship exists between the density of bolting c:: 30·/
and the relative size of blocks formed by jointing. Figure 11 can ·c 0.3 /
Q)
8
c. Rocic ua!;; cor.~: cnd
b .. ~ock :.l::lt Cr:71<;ZC: !J.i(;'rll
L
10 C::':l: "nila::-.
0 7
+-'
:l,. Dj sco •.•(inuou, joints
0
,. i\ol,Uj" or ; r'-o::o; •..•I,,., IJodul:o!in9 ) 0
C. Sorootl'l, undur't;ng 2
LL 6
O. Slickensidcd. undul"tj"9 1.5
I. k:1l loG jf (ht ....,~,. ~;l,ci"9 of ~he +-'
L ,"oug'" or irrco;ut~ •.• pl:o,. ••, 1.5 rdc"lI'.t joint see I, ~rc ••tcr ch:on Ja. C 5
(1)----
s,
c,
S_~h.
Slic:"~"~;ctcd. phnu'
pl,,".IIO'" 1,12
a.s
1. J,." a.s
••••sided
':0"
jOi'Hl
;:.1.,,,:0 •. ,
l:Ie vud
I;n.,.('on,.
lu"ing
for ,Iic~-
provided
E g'
+-' .-
the I' •.•
"",(io,., .re o";e,.(".,d (or ",in;",,,,,,
c. ,'fo e-ee.c :.:c:Z<: C'Or.c.:c: Hr""'itl'l.
UJ"D 4
""'.,. s;'ccz~d. ::J:=
'~(f)
"D '-../
:i. Za"" (0"'("''';1'''. doly •••;"e •.••I'
Chick f;nouo;h to prt;-ItH'I{ reell. 4: 3
•••'" I I con (.le ~ . 1.:)
0.1
FIGURE A4. Estimation of Jr and Ja ratings according to Barton et
o 5 10 15 20 , 25
al. (1974). Hydraulic Radius Cm)
L
0 7
+-'
0
0
LL 6 Acknowledgments
+-' The authors wish to express their gratitude to all the mines
C -0 5 that have provided data and time to this study.
(I) LL
E >- As well, special thanks to NSERC, oranda Research and
~~ 4 Falconbridge Limited for their continuous feedback and finan-
::J >
.~ 0 cial support of the project.
"D L
4:t.? 3
'-../
»-,
+-'
.s; 2 REFERENCES
0 BARTON, N. LIEN, R. and LU DE, J.;Engineering classification of
L
t.? rock masses for the design of tunnel support; Rock Mechanics, Vol.
6, No. 4, pp. 189-236, 1974.
FABJA CZYK, M., Review of ground support practice in Australian
underground metalliferous mines; Proc. Annual Conf. Aust. Inst.
10 20 30 40 50 60 70 80 90
Min. Metall" Melbourne, pp. 337·349, 1981.
Inclination of Stope Plane FULLER, P.G., and COX, R.H.T., Rock reinforcement design based
on control of joint displacement-a new concept; 3rd Aust. Tun-
FIGURE AS. Influence of gravity for the slabbing and gravity fall and nelling Conf., Sydney, pp. 28-35.
slabbing modes of failure. FULLER, P.G., Cable support in mining: a keynote lecture; Proc. Int.
Symp. on Rock Bolting, Abisko, Sweden, pp. 511- 522, 1983. fied cable bolt system for the support of underground openings;
HOEK, E., and BROWN, E.G., Underground Excavations in Rock; AUSL Inst. Min. Metall, Conference, Broken Hill, .S.W., Aus-
Institute of Mining and Metallurgy, London, 1980. tralia, July 1983.
JEREMIC, M.L., and DELAlRE, G.J.P., Failure mechanics of cable bolt MILLER, D.R., Design of cable reinforcement patterns to resist shear
systems; C1M tiuu., Vot. 76, No. 856, pp. 66-71, Aug. 1983. failure in open stope walls; in Stability in Underground Mininz Il
MATHEWS, K.E., HOEK, E., WYLLIE, n.c., and STEWART, A.B. Swilski and C.O. Brawner , eds., Lexington, Kentucky, Aug:
S.B.V., Prediction of Stable Excavations for Mining at Depths Below 6-8, 1984.
1000 Metres in Hard Rock; CANMET Report 802-1571, 1980. STHEEMAN, [Link], A practical solution to cable bolting problems at
MAITHEWS, M., TILLMAN, V.H., and WOROTNICKI, G., A modi- Tsumeb Mine; C1M Buff., Vot. 75, No. 838, pp. 65-77, Feb. 1982.