-
A Publication of the University of Miskolc Series A. Mining,
Volume 63, (2003)
International Conference on Safety and Environmental Aspects of
Mining
THE ROLE OF PILLARS IN SMALL
UNDERGROUND MINES
Prof . Dr. Horst W a g n e r Head Department of Mining
Engineering
Montanuniversitt Leoben (Austira)
Summary
Alpine mineral deposits tend to be of limited extent and of
complex geology. Historically pillar design was based on experience
rather than broad based engineering facts. This differs
substantially from the pillar design in extensive tabular deposits
which, based on Salamon's pioneering work, has reached a high
degree of proficiency. The basic differences in pillar design for
small alpine mines and for extensive tabular deposits are
discussed. It is shown that in the case of the former each case is
treated on its merit and that there are no common design rules. The
estimation of pillar strength constitutes a particular problem and
examples of recent developments in this area are given. It is shown
that discontinuities play an important role as far as the strength
of pillars and the layout of pillar systems is concerned. The role
of backfill is discussed and the effects of backfill on pillar
behaviour are shown. Because of the limited lateral extent of most
alpine pillar workings pillar failures tend to be stable but
situations leading to unstable pillar failures can arise in the
final stages of mining. Because of the vertical extent of some of
the alpine mineral deposits pillar workings exist on a number of
mining levels. This leads to interactions of workings in adjacent
mining horizons. The resulting difficulties are discussed and the
importance of superimpositioning of pillars is highlighted. A
number of general conclusions complete the paper.
89
-
Horst Wagner
1. INTRODUCTION
Pillars are essential elements of many mining systems. Pillars
are employed to support stoping excavations, to protect surface and
underground structures and to act as safety barriers. For many
years the design of pillar systems was entirely experience
based.
The tragic mining accident at the Coalbrook Colliery in South
Africa which caused the loss of 437 lives, when several thousand
coal pillars failed in 1960, marked the beginning of intensive
research efforts into pillar systems. Through the pioneering
efforts of Prof. M.D.G. Salamon and his colleagues at the Research
Organisation of the Chamber of Mines of South Africa the
foundations of modern pillar design principles based on probability
theory were laid1-2 Based on these principles design guidelines for
room and pillar workings were established and have found wide
application not only in South African coal mines but also in other
parts of the world3
The name of M.D.G. Salamon is not only closely linked to modern
coal pillar design but also to the important question of pillar
stability and instability and to the role of pillars in
ameliorating the rockburst hazard in deep level mining4-5 There is
no doubt that M.D.G. Salamon, more than anybody else, has made a
lasting impact on the development of pillar mining systems.
Whereas Salamon's work on pillar mining system was concerned
primarily, but not exclusively, with tabular mineral deposits of
considerable lateral extent there are many situations where pillar
mining is being carried out in mineral deposits of very limited
extent. This is particularly the case in the Alps where rather
small mineral deposits are being mined. In these situations some of
the basic concepts such as the tributary area concept are no longer
applicable. In addition the irregular shape of the deposit and in
particular the variable thickness of many of the alpine deposits
results in pillar systems which differ greatly from the classical
coal mining or typical hard rock pillar mining situations found in
tabular deposits.
In this paper a number pillar mining situations in small
Austrian underground mines will be discussed and the differences
between extensive and limited scale pillar mining applications
highlighted. In particular some of the inherent dangers resulting
from the application of limited scale mining experiences to larger
scale applications will be addressed.
90
-
The Role of Pillars in Small Underground Mines
2. FUNDAMENTAL DIFFERENCES BETWEEN LARGE SCALE AND LIMITED SCALE
PILLAR MINING
The classical approach to the design of room and pillar workings
is based on the safety factor concept and was proposed by Salamon
in 19671 In this paper Salamon states:" The average pillar load is
not easily predicted in the general case. If the area is not large
or if it contains large intact portions of the seam, pressure p
depends on many factors. Its maximum value can, however, be deduced
in a simple manner, provided the pillars are reasonably uniform.
Assuming that the whole weight of the overburden is carried by the
pillars and assuming that the weight increases 2 500 kg/m2 for
every metre of the depth*, then:
/?=(0.025H)/(l-e) (1)
where H Depth in m e Extraction ratio (lOOe = percentage
extraction) p Pillar stress in MPa.
In the case of small alpine mineral deposits the difficulty
mentioned by Salamon applies and the so called "tributary area"
concept can not be used to estimate pillar load or pillar stress.
This makes it extremely difficult to estimate the load acting on
individual pillars. The problem is aggravated further by the
irregular topography. The second difficulty arises from the fact
that, because of the limited size of deposit and the complex
geology of many of the alpine mineral deposits, there are usually
insufficient numbers of representative mining situations available
which can be used to apply the concept of back analysis which
formed the basis of the formula for the strength of coal pillars
developed by Salamon and Munro in 19672
As a consequence of these difficulties the probabilistic
approach to pillar system design based on the concept of back
analysis as proposed by Salamon in 1967 is faced with considerable
problems when applied to small alpine deposits as there is usually
insufficient field data available to follow this route. As a result
pillar system design in the majority of Austrian mines has been
done predominantly on the basis of practical experiences and
judgement. Because of the limited extent of mining pillar failures,
when they did occur, were mainly of local rather then a regional or
global nature. There were however some large scale collapses such
as the failure a whole stoping area in the lead-zinc mine Bleiberg
in the 1970's which is now closed.
" Units converted from imperial to metric units.
91
-
Horst Wagner
The basic differences between pillar systems employed in
extensive tabular mineral deposits such as for example South
African or Australian coal mines or the chrome mines of the
Bushveld igneous complex in Southern Africa and the pillar systems
found in many of the alpine mines are summarised in Table 1:
Property/parameter Overseas situations Alpine situation
Geological situation regular complex Depth shallow to medium
variable over short
distances Dip of deposit flat to slightly inclined irregular
faulting slight to moderate moderate to extensive Thickness of
deposit narrow to medium medium to large Lateral extent of
workings
large limited
1. table: Differences between overseas and alpine pillar mining
situations
As a consequence of the differences detailed in Table 1 pillar
design in alpine deposits tends to be unique in each case whereas
overseas differences between conditions in the mines operating a
particular type of deposit tend to be rather small. As a result
experiences gained in one mine can be applied in other mines.
From a pillar systems point of view the main differences are
summarised in Table 2:
System property Overseas situation Alpine situation Pillar load
based on tributary area
concept very variable, generally less than tributary area
load
Pillar material very uniform irregular and extensively
jointed
Pillar height 1 m-5 m 2 m -80 m Width/height ratio of
pillars
usually between 2 - 8 often
-
The Role of Pillars in Small Underground Mines
3. MAJOR DESIGN PROBLEMS IN ALPINE P ILLAR MIN ING
SITUATIONS
3.1 Pillar strength
The foremost problem encountered in the design of pillar systems
for alpine mining situations is the estimation of pillar strength.
For reasons detailed above back analysis of existing pillar
workings is usually not possible. Consequently pillar strength has
to be estimated on the basis of laboratory tests on rock samples
and rock mass classification. The latter is used to down-rate the
laboratory strength values to allow for the effects of jointing on
the strength of pillars. In recent years the "Geological Strength
Index" (GSI) which was introduced by Hoek in 19946has been used in
conjunction with the generalised Hoek-Brown rock mass strength
criterion' to determine the strength of the pillar material. The
effect of the width to height ratio on the strength of pillars with
W/H>1 pillar strength has been estimated using Salamon's pillar
strength formula.
where C rock mass is the strength of the rock mass according to
Hoek and Brown'
For pillars with W/H
-
Horst Wagner
The results of laboratory tests and numerical simulations show
that for very slender pillars the strength can drop to zero if the
pillars are intersected by steeply dipping discontinuities.
Figure 2 shows that regular jointing of pillars can be a common
feature in alpine mines and needs to be taken into account in
pillar design. In the case of one set of discontinuities the most
effective means of dealing with this problem is to use elongated
pillars instead of square pillars on to orient the pillars in such
a way that their long axis is aligned in the dip direction of the
joints8
Figure 2 shows the beneficial effects of aligning the long axis
of a pillar in the direction of dip of the joints (Angle=0 in
Figure 2)
Dip angb of th Joint []
Fig.2: Effect of pillar elongation and orientation on strength
of jointed pillars
-
The Role of Pillars in Small Underground Mines
Fig.3: Effects of regular system of discontinuities on pillar
behaviour in an alpine gypsum mine.
3.2 Pillar load
The estimation of pillar load in alpine mines is made difficult
by the uneven topography, the complex tectonic situation and the
irregular shape of the mineral deposits. It is only since the
availability of powerful computers and stress analysis software
that reasonable estimates of pillar loads can be made. Even now the
lack of knowledge of the primitive stresses acting in the rock mass
makes it extremely difficult to determine pillar loads accurately.
As a result of this uncertainty high nominal factors of safety have
to be employed to ensure safe mine layouts.
A particular problem is the estimation of pillar loads in
multi-level pillar workings. Such situations are not uncommon and
have caused numerous problems in the past, A particular problem has
been that for a variety of reasons pillars in different mining
horizons have not been superimposed. As a consequence very
unfavourable loading conditions have developed and in a number of
instances resulted in local collapses. Figure 4 shows a typical
example of layout problems in multi-level pillar workings.
95
-
H o r s t Wagner
Fig. 4: Extreme example of layout problems in multi-level pillar
mining. The example is from a marl mine in Tyrolia.
Figure 5 shows the results of a numerical simulations of typical
multi-level pillar mining situations found in Austrian mines.
-100 melcr
Geometry: pillar width 4m pillar height 4m
Fig.5: Interaction between pillars in adjacent mining
horizons
96
-
The Role of Pillars in Small Underground Mines
The numbers in the diagram are stress concentration factors.
From Figure 5 it is apparent that the pillar load and the stress
distribution in the sill pillars is very much dependent on the
local circumstances. As can be seen from the diagrams the thickness
of sill pillar between the individual extraction horizons is a
critical parameter as far as the interaction between pillars in
individual mining horizons is concerned.
The parametric studies have by and large confirmed the
recommendations concerning the design of multi-seam room and pillar
operations in South African collieries made by Salamon and Oravecz
in 19769
3.3 Pillar behaviour and backfill
A typical mining situation in Austria is the extraction of lens
type mineral deposits. Representative for this type of deposit are
the alpine magnesite deposits. Typically these deposits have
dimensions of 150m to 200m in length, 100m to 150 m in width and
50m to 100m in height. These deposits are usually extracted in
ascending slices using room and pillar methods and backfill. In the
final stages of extraction the magnesite pillars reach a height of
50m to 100m.
The pillars are situated in backfill which acts as the working
platform for the mining equipment and provides lateral stability
and confinement to the very slender pillars which typically have
cross sections of 7m by 7m to 7m by 14m.
The backfill itself is usually uncemented waste rock which is
compacted by the action of the mining equipment operating on top of
the fill. Relatively little is known about the strength of these
very slender pillars and the load acting on them. Experiences
however show that the workings are stable even during the final
stages of extraction that is when the ultimate height of pillars is
reached.
Figure 6 shows the effect of compaction of backfill material as
a result of operating equipment on top of backfill on the load
deformation behaviour of uncemented backfill.
Laboratory tests on model pillars show that uncemented backfill
has very little effect on the strength of pillars but is very
beneficial from the point of view of post-failure behaviour of the
pillars. Backfill is particularly beneficial in the
97
-
Horst Wagner
presence of jointing in the pillars. These findings have been
confirmed by the results of numerical simulations. Figures 7 and 8
show the results of numerical simulations of the effect of cemented
and uncemented backfill in the behaviour of jointed and unjointed
gypsum pillars of width/height ratio of 1. These simulations
confirm the results of studies by Galvin10 on the effects of
ashfill on the behaviour of coal pillars.
J / / J ^ Con paction 1 y W h M l .od*r
0 3 6 9 12 19 18 21 24 27 30 33 38
Varticai Dalormation [ % ]
Effect of Compaction by Wheel Loader an Vertical Stress and
Deformation of Uncemented Backfill (Laboratory Scale Tests)
Fig.6: Effect of equipment operation on the load deformation
behaviour of uncemented backfill
Figures 7 and 8 show that the load bearing capacity of pillar
systems is increased only if either cemented backfill is used or
the workings are almost completely backfilled. In all other
instances the main advantage of using backfill is the improved
post-failure behaviour of the pillars.
98
-
The Role of Pillars in Small Underground Mines
Vertical Da formation [V.]
Effect of Cemented and Uncemented Backfill at Fill Heights of
50% on Pillar Stress
Fig. 7: Effect of backfilling 50% of the height of the pillar
workings on the load deformation behaviour of pillars
Vert ica l De fo rmat i on [ % ]
Fig.8: Effect of filling the pillar workings almost completely
with backfill
-
Hors t Wagner
3.4 Stable or unstable pillar behaviour
A feature of pillar systems in many of the small alpine mineral
deposits is that as a result of the very limited extent of the
pillar workings and the comparatively large thickness of overburden
pillar failure where it does occur tends to be stable rather then
unstable. Figures 9 and 10 show typical cases of stable pillar
failures. This type of failure tends to be the exception in pillar
workings of substantial lateral extent.
Fig.9 and 10: Examples of a failed stope pillars
Unfortunately the potential dangers associated with this pillar
loading situation are often not fully appreciated by mining
personnel who consider the situation as perfectly stable and can
not envisage the possibility of sudden collapses as a result of
changes in the mining geometry. The latter situation can arise at
the final stages of mining when the extraction of remnant pillars
separating pillar areas of low safety factors is contemplated. In
these instances the mining spans are suddenly increased and the
stiffness of the mining layout is greatly reduced.
100
-
The Role of Pillars in Small Underground Mines
Typical examples are mines where the main development is
situated in the central part of the mineral deposit. In the final
stages of mining the pillars protecting the main development are
often the only remaining mineral reserves and there is a
considerable temptation to extract these pillars. Figure 11 gives
an idealised view of this generic problem.
4. CONCLUSIONS
Pillars play an important role in the extraction of small alpine
mineral deposits. Because of the irregular nature of the majority
of these deposits, the complex geology and the limited lateral
extent of the pillar workings there has been no systematic approach
to pillar design.
The latter has been empirical in nature and not based on sound
engineering principles. As a result of the limited extent of the
pillar workings and hence the high stiffness of the pillar layouts
pillar failures where they did occur have been usually of a stable
nature thereby creating a false sense of security. Whereas pillar
design in mines of significant lateral extent is based on the
concept of tributary area pillar load this concept is not being
applied to any significant extent in small alpine mines. As a
result of the sub-critical mining spans pillars are subjected to
only a fraction of the tributary area load. Pillar dimensions are
often constant and independent of the depth of cover. Furthermore
because of the generally low pillar
101
-
Horst Wagner
loads and the often considerable thickness of the mineral
deposits support pillars tend to have very low width to height
ratios. Frequently these are well below one. As a result
discontinuities in the rock mass play an important role in pillar
design and the layout of pillar systems. Because of the
considerable thickness of many of the deposits multi-level pillar
mining and backfill play an important role in pillar mining. Until
fairly recently the interaction of room and pillar workings in
adjacent mining horizons has been ignored.
The role of backfill has been largely to provide a working
platform for mining personnel and mining equipment and to a lesser
degree to enhance the stability of the pillar workings. Lately the
emphasis is changing and pillar stability is becoming a prominent
feature of backfill systems.
The pioneering work by Salamon in the field of the design of
pillar systems and the understanding of the conditions governing
stability and instability of pillar workings has been of great
value in the assessment of pillar systems in small alpine mines.
This knowledge has led to a reassessment of pillar systems in many
of the operating mines and in instances has resulted in substantial
changes to the pillar layouts.
102
-
The Role of Pillars in Small Underground Mines
References:
[1] Salamon, M.D.G. (1967) "A method of designing bord and
pillar workings" J. S. Afr. Inst. Min. Metall., v. 68, pp.
68-78.
[2] Salamon, M.D.G. and Munro, A.H. (1967) "A study of the
strength of coal pillars". J. S.Afr. Inst. Min. Metall. V 68 , pp.
55-67.
[3] Salamon, M.D.G., Galvin, J.M., Hocking, G. and Anderson, I.
(1996 "Coal pillar strength from back -calculation" Res. Rep. 1/96,
Department of Mining Engineering, Univ. New South Wales, 1996.
[4] Salamon, M.D.G. (1970) "Stability, instability and the
design of pillar workings". Int. J. Rock Mech. Min. Sei., v. 7, pp.
613-631.
[5] Salamon, M.D.G. and Wagner, H. (1979) "Role of stabilising
pillars in the amelioration of rockburst hazard in deep mines"
Proc. 4th Congr., Int. Soc. Rock Mech., v. 2, pp.561-566, Balkema,
Rotterdam, 1979.
[6] Hoek, E. ( 1994) "Strength of rock and rock masses" ISRM
News Journal v. 2(2), pp. 4-16
[7] Hoek, E. and Brown, E.T. (1997) "Practical estimates of rock
mass strength" Int. J. rock Mech. Min. Sei. v. 34,
pp.1165-1186.
[8] Salamon, M.D.G. and Oravecz, K.I. (1976) "Rock mechanics in
coal mining" Chamber of Mines of South Africa, Johannesburg,
1976.
[9] Galvin, J.M. and Wagner, H. (1982) "Use of ash to improve
strata control in bord and pillar workings" Proc. Strata Mechanics,
Newcastle-upon-Tyne, April 9-13, 1982
103
-
Horst Wagner
sszefoglals
Az Alpokban egyre kevesebb az svnyi anyag lelhely, s azok
geolgiai viszonyai is egyre bonyolultabbak. Trtneti okokbl a
pillreket inkbb tapasztalatokra hagyatkozva, mintsem elmletileg
megalapozva terveztk. Ez lnyegesen eltr attl a fejlett mdszertl,
ami kiterjedt telepes svnyi lelhelyekre Dr. Salamon ttr munkja
eredmnyeknt a rendelkezsnkre ll.
A tanulmny az Alpok kis bnyinak, illetve a kiteijedt telepes
lelhelyek pillr mretezsi mdszerei kzti lnyeges eltrseket ismerteti.
ltalnos tervezsi mdszerek nincsenek, minden esetben figyelembe
veszik a mltbeli tapasztalatokat. Klns sly problma a pillrek
szilrdsgi alapon val mretezse, nhny pldt bemutat a szerz e kutatsi
terlet eredmnyeibl.
A diszkontinuitsok, amint arra az rs rmutat, fontos szerepet
jtszanak a pillrek llkonysgban, s abban, hogy milyen elrendezs
szerint kell visszahagyni azokat. Megmutatja a tmedkels szerept s a
tmedknek a pillrek viselkedsre gyakorolt hatst is. Az Alpokban mkd
pillrfejtsek kis horizontlis kiteijedsek, ezrt a pillrek ltalban
llkonyak, de a bnyk bezrst meglelzen elfordulhatnak stabilitsi
problmk. Az elfordulsok gyakran jelentkeny vertiklis kiterjedse
tbbszeletes pillrfejtsek kialakulshoz vezetett. Ezrt a szomszdos
szintek hatst gyakorolnak egymsra.
A tanulmny az ebbl ered nehzsgeket is trgyalja, s rmutat annak
fontossgra, hogy az egyes szintek pillrei pontosan egyms fltt
helyezkedjenek el. A cikket ltalnos kvetkeztets zrja.
104