Quarterly Journal of Engineering Geology, 27, 145-164. 0481-2085/94 $07.

00 © 1994 The Geological Society

A revision of the graphical method for assessing the excavatability of rock

G. S. Pettifer 1 & P. G. Fookes 2

~Consulting engineering geologist, 22b Rusham Road, Balham, London SW12 8TH, UK
2Consulting engineering geologist, 'Lafonia', l la Edgar Road, Winchester, Hampshire SO23 9SJ, UK

Abstract important, since they determine whether the disconti-

nuities are effective planes of separation, i.e. whether
the ripper tine or excavator bucket can penetrate them
A 'size-strength' graph, subdivided for various methods of
excavation, was published by Franklin, Broch& Walton in and dislodge individual rock blocks. There may be a
1971. Although this graph allows excavatability to be assessed high rate of cutting part wear in abrasive materials
rapidly, the subdivisions have become outdated as more such as fine-grained siliceous rocks; this will slow down
powerful, more efficient equipment has become available. A but not necessarily prevent a particular method of
complete revision of the graphical method, based on more mechanical excavation.
than 100 case studies, is proposed. Particular emphasis is
It is usually cheaper to break up rock masses by
placed on rock masses which can be broken up solely by
mechanical means such as ripping. Procedures are recom- ripping rather than by drilling and blasting, but
mended for obtaining geotechnical data, and the importance productivity may be lower. There may be a need to
of interpreting field observations in their geomorphological produce material of a particular size for use elsewhere
context is discussed. Consideration is also given to the effects in the earthworks programme, either directly or via a
of block shape and orientation on ripper performance, and site crusher. Where the blocks are too large, and they
adjustments to the input data are suggested. The .revised
cannot be reduced simply by driving over them with
graph is intended for all types of project but especially for
road construction, where plant mobility and flexibility are the tractor or by decreasing the spacing between ripper
important. runs, the additional breaking that is necessary may
make ripping impracticable. Blasting is often more
economic for large-scale quarrying operations, whilst
some road contracts may specify the use of techniques
Introduction such as pre-split blasting.
A number of systems have been developed for
The ease with which the ground can be excavated (i.e. assessing various aspects of excavatability, including
its excavatability) must be assessed so that civil those proposed by Weaver (1975), Kirsten (1982) and
engineering works can be planned and priced. Ground Scoble and Muftuoglu (1984). In addition, plant
preparation by ripping or blasting may first be manufacturers such as the Caterpillar Tractor Com-
required, followed by digging or scraping, loading and pany suggest procedures for assessing rippability.
transportation to dumps or stockpiles. Production Franklin, Broch & Walton (1971) published a 'size-
rates, running costs and haulage distances all have to strength' graph which related discontinuity spacing and
be taken into account when selecting plant. Working rock strength to the method of excavation required.
practices may be influenced by space limitations, The graph was subdivided into areas of digging or
environmental constraints, or by the availability of scraping, ripping, blasting to loosen and blasting to
plant, spare parts, explosives, and skilled labour. fracture on the basis of research carried out in the U K
Similar factors have to be considered when planning between 1968 and 1970. Essentially the same graph was
quarrying operations. published by Fookes, Dearman & Franklin (1971) who
The excavatability of rock depends on the geotech- emphasized that it was at the development stage.
nical properties of the material, on the method of The graphical method is particularly useful for rapid
working, and on the type and size of excavating assessments during walkover surveys, and for projects
equipment used. In mechanical excavation the cutting where borehole, seismic or laboratory test data are not
parts of the equipment must be forced into discontinui- available, but the original graph is now out of date due
ties in the rock mass or into the fabric of weak rocks. to advances in technology. A complete revision of the
It is generally accepted that discontinuity (or fracture) method has been carried out, based on published and
spacing and the strength of the intact rock are original case studies. The updated 'excavatability
particularly important properties. The separation, graph' stresses the limits of ripping using different sizes
infilling and wall strength of the discontinuities are also of tractor.
146 G.S. PETTIFER & P. G. F O O K E S

TABLE 1. Details of Caterpillar tractors considered in this study (from information in the Caterpillar Performance
Handbook, 19th edition, 1988)
Flywheel Operating
Tractor horse- mass Years Remarks
model power (tonnes) produced
D6D 140 14.3 1977-86 Most powerful D6 before introduction of D6H (165 fhp).
D7G 200 20.1 1977-86 Most powerful D7 before introduction of D7H (215 fhp).
D8K 300 32.0 1974-82 Equivalent to Komatsu D155A and Terex 8230B.
D8L 335 37.3 1982-86 Raised final drive as in 'N' series.
D9N 370 42.5 1986 on Raised final drive.
D9H 410 42.8 1974-81 Equivalent to Komatsu D355A.
D9L 460 52.0 1980-86 Raised final drive as in 'N' series.
D10 700 79.6 1978-86 Raised final drive as in 'N' series.
D1 IN 770 95.8 1986 on Raised final drive; most powerful Caterpillar tractor.

Developments in excavation methods A new generation of large hydraulic excavators, for

example the CAT 245 and Komatsu PC400, has
and equipment since 1971 extended the limits of direct digging by backhoe or
face shovel. There was some evidence from published
In most ripping operations, a single straight or curved case histories that a large face shovel might be more
shank (the tine) is pulled by tractor through the effective than ripping in moderately weak, massive
ground in order to loosen rock for removal by digging rocks. Karpuz (1990) found that hydraulic excavators
or scraping. The introduction of larger, heavier and of up to 8 m 3 bucket capacity could dig directly in rock
more powerful tractors and improvements in ripper masses, mainly marls, that would otherwise have to
design have significantly increased ripper productivity be prepared by ripping with a D8 or D9 tractor.
and extended the range of rocks that can be ripped. Similarly, large scrapers used on major earthmoving
Caterpillar (CAT) ' N ' series tractors, for example, works where there is ample turning space may be able
were introduced in 1986. These have a raised final to excavate relatively weak rocks without resort to
drive, which helps to keep the drive mechanism clean ripping. Bucket wheel excavators have a continuous
and extends the track frame, thereby increasing their mode of digging, and are therefore less adaptable than
capability. The tine is protected from wear by a other types of excavator to variable ground conditions;
detachable steel tip, and the rate of wear due to they have not been considered further in this study but
abrasive rock is one factor that will determine whether are widely used in opencast mining.
ripping is economic in marginal situations. A consider-
able amount of tip wear (e.g. replacement every 2-3
hours) may be acceptable in some circumstances.
Typical modern ripping equipment is shown in Fig. 1,
and details of tractors which have been considered in
this study are given in Table 1.

FIG. 2. Komatsu PC280 hydraulic excavator (backhoe) fitted

with Montabert 625 rock breaker. Large in situ blocks are
being fractured prior to ripping with a D9N.

Where a rock mass cannot be broken up satisfac-

FIG. 1. Caterpillar D9N tractor ripping during the early torily by ripping or direct digging, some other means
stages of excavation for a road cutting. must first be used to loosen or fracture the constituent
 EXCAVATABILITY OF R O C K : THE GRAPHICAL METHOD 147

.~.~ E,

p. .~

8
.<
-6 -o

.~

~~
•~ , ~ ~ ~ ,. ,::: ~ ~ "~
~, ~ ~ ,_,~-~..o~ .
"~ ~ '.~ ~ . . ~ ~.'~ ~"0 ~
~'~ ~ ~ _~-- ~ o
O~

~ .,-~ ~ ,.-,,

~.~.-o.-o 0~ ~ ~ ,.,
~'~ ~.~-~_ ~..~.~.~_ ~ ~ ~."~ ~ ..-,

-~:~'~..~ t:~,~ ~:~ P",,,J~ . ~ ~'~

•- ~ ~ ~,..
~ ~ ~ "~ ~ ~ ~ ~"..~
z . . . . . . . . . ~ , ~ .
148 G. S. PETTIFER & P. G. FOOKES

blocks. Cost considerations and environmental con- in ripping. Minty & Kearns (1983) modified Weaver's
straints have led to an increased use of impact rippers system for Australian conditions. Their 'geological
and hydraulic rock breakers as an alternative to light factors rating' (GFR) incorporated new numerical
blasting. Impact rippers subject the rock mass to high- ratings and reintroduced factors for groundwater
frequency impacts through the ripper tine, whilst rock conditions and for surface roughness of discontinuities
breakers (or peckers) fitted to hydraulic excavators are which were included in the original RMR system. They
used prior to normal tipping or digging. Separate multiplied GFR by seismic velocity and plotted the
machines for breaking and ripping are generally product against tractor size, in order to determine
preferred because there is less wear and tear, and worn whether ripping was likely to be either satisfactory or
parts are more easily replaced. The present study marginal. Seismic velocity is closely related to other
indicated that block size reduction using a large parameters included in these systems and Smith (1986),
hydraulic rock breaker (such as the Montabert 625 using case studies from the USA, suggested that it
shown in Fig. 2) followed by ripping with a modern should be omitted.
D9 tractor had approximately the same overall effect Kirsten (1982), also using examples from South
as ripping with a D11 tractor on its own. D1 Is require Africa, proposed an 'excavatability index', N, based on
considerably more operating space than D9s, however, the Norwegian Geotechnical Institute's 'Q' system for
and additional breaking may be necessary before the tunnelling (Barton, Lien & Lunde 1974). He also
ripped material can be handled; they are not com- suggested adjustments for discontinuity orientation in
monly available for hire in the UK. Ground prep- ripping. Kirsten's index does not include seismic
aration using rock breakers generally takes much velocity, but its usefulness is limited by the incorpor-
longer than blasting, but the method is particularly ation of RQD, which strictly can only be determined
useful where resistant rock 'knolls' occur within more from drill core, and by the inclusion of 'Q' terms
easily ripped material. which cannot be rapidly quantified. Abdullatif &
Cruden (1983) found that the RMR system gave a
better assessment of rock mass quality for excavation
purposes than the Q value.
Excavatability assessment systems Scoble & Muftuoglu (1984), from studies in British
opencast coal mines, devised a 'diggability index' based
Systems have been devised either for general assess- on discontinuity spacing, rock strength and weathering.
ments of excavatability, for specific operations such as Singh, Denby & Egretli (1987), from studies in
tipping, or for particular applications such as opencast opencast mines in the UK and Turkey, developed an
mining. Each system considers a different set of alternative rippability rating chart. They retained
geotechnical parameters (Table 2), but most include seismic velocity and weathering but introduced a
factors for the strength of the intact rock and for the parameter for abrasiveness and expressed rock strength
degree of fracturing of the rock mass. Field seismic in terms of uniaxial tensile strength. In their opinion
wave velocity, which depends on rock strength, density, discontinuity separation and gouge were adequately
fracture intensity and weathering, has been widely used covered by the weathering parameter. Excavatability
since the 1960s and is a very important factor in assessment systems have also been proposed by
several of the systems. Karpuz (1990) and by Hadjigeorgiou & Scoble (1990),
Ripper performance charts published in the Cater- specifically for use in opencast mining.
pillar Performance Handbook (e.g. 19th edition, 1988) The graph publish.ed by Franklin et al. (1971)
correlate seismic velocity values for various rock types considered only two parameters: fracture spacing
with tractor size. The field measurements are time- index, If, and point load strength index, Is. Abdullatif
consuming, and in an area of complex geology they (1982) used the graph to compare excavation case
may prove difficult to interpret. It has been suggested histories from pits and quarries in southern England.
that the published charts tend to overestimate the ease He did not have sufficient data to determine new
with which the most difficult materials can be ripped subdivisions, but found that granite and vein quartz
(e.g. Karpuz, Pasamehmetoglu, Bozdag & Muftuoglu which plotted within the 'blast to loosen' area on the
1990). 1971 graph could be ripped using a D9L tractor. Pells
Weaver (1975), using examples from South Africa, (1985) used a similar approach to evaluate the
proposed a 'rippability rating chart' adapted from the rippability of the Hawkesbury Sandstone in Australia,
Rock Mass Rating (RMR) system used for tunnel and his results indicated that moderately strong
support design (Bieniawski 1973). His main changes sandstones which plotted within the 'blast to fracture'
were the replacement of Rock Quality Designation area could in fact be tipped using a D10 tractor. More
(RQD) with seismic velocity, the introduction of a recently, Karpuz et al. (1990) have modified the graph
weathering parameter and adjustments for the effects on the basis of rippability studies at opencast lignite
of discontinuity orientation (i.e. dip and dip direction) mines in Turkey.
 EXCAVATABILITY OF ROCK" THE GRAPHICAL METHOD 149

• .f

ii i

A. Blocky B. T a b u l a r C. Flaggy
[approximately [smallest dimension [smallest dimension
equidimensional- greater than 60 mm; 20-60 mm;
largest dimension other dimensions other dimensions
not greater than at least twice at least twice
twice the smallest] the smallest] the smallest]

D. S l a t y or S h a l y E. I r r e g u l a r F. C o l u m n a r or
[smallest dimension [including Elongated
less than 20 mm; pyramidal]
other dimensions [largest dimension
at least twice at least twice
the smallest] each of the others]

Descriptions such as "blocky-tabular" may be used where block shapes are borderline

FIG. 3. Sketches illustrating various types of rock mass structure (after CIRIA/CUR 1991).

Procedures used to obtain data during structural zones, or domains, within which rock type,
strength, weathering grade, discontinuity spacing and
the present study discontinuity orientation were more or less uniform,
following procedures suggested by Anon (1977) and
Case histories for this revision of the graphical method ISRM (1981)• Strength and weathering grade were
were collected in Africa, Hong Kong and the UK. described using the criteria in the Code of Practice for
Fracture spacing and rock strength data measured in Site Investigations, BS 5930: 1981. Discontinuity aper-
the field were correlated with details of the excavation ture, infill, roughness and persistence data were also
methods, either observed on site or compiled from recorded, so that excavatability could be assessed using
information provided by site staff who were present some of the other systems.
during construction (see Appendix). Where sufficient Discontinuities were considered to be effective when
material was exposed, rock masses were divided into they clearly formed planes on which rock blocks could
150 G. S. PETTIFER & P. G. FOOKES

be dislodged, and were therefore persistent on the scale identified in most domains, and discontinuities were
of individual blocks. Joint planes healed by quartz counted along scanlines normal to the strike of each
or calcite mineralization and poorly defined bedding set. So as to give greater emphasis to the closest-spaced
planes which simply represented a change in lithology set, which would be likely to have particular signifi-
and rock strength were not effective in these terms. The cance for excavatability, individual totals were first
spacing and orientation of effective discontinuities expressed as frequencies and then their mean was
control the size and shape of the blocks. Despite converted to an equivalent spacing. The value obtained
differences in the mode of origin and structural history related to a unit volume of rock.
of rock masses, there are relatively few typical It was sometimes difficult or impossible to carry out
discontinuity patterns and hence block shapes. These the above procedures, for example where exposures
are shown in Fig. 3, based on sketches in C I R I A / C U R had a flat face so that joint sets could not be easily
(1991), with suggested limiting dimensions. Actual determined; where two sets had the same strike but
block shapes are unlikely to be as uniform as those in dipped in opposite directions; or where the jointing
the illustrations. pattern was random and the blocks were irregular.
Where these problems were encountered in double-
sided road cuttings, measurements were made along
Discontinuity spacing measurements scanlines in two orthogonal directions on each side and
then the average of all the results was taken.
Various problems arise when attempting to describe ISRM (1981) suggested that block size could be
block size in terms of a single dimension, but this is described by a volumetric joint count, Jv, the sum of
necessary in order to represent average discontinuity the number of discontinuities per metre for each set.
spacing simply. The fracture spacing index is custom- This method was used by Hadjigeorgiou & Scoble
arily determined from measurements along two ortho- (1990) for their assessment system (see Table 2).
gonal or randomly orientated scanlines, or along a Discontinuities are measured for Jv in the same way as
section of drill core. The index can be expressed either for the three-dimensional If described above, so that
as a frequency, i.e. the mean number of fractures per where there are three principal sets Jv equals 3/If.
metre, or as the reciprocal, i.e. a mean spacing in ISRM (1981) also suggested using a block size index,
metres. Alternatively, the modal spacing may be used. Ib, which is estimated by taking the average dimen-
This two-dimensional approach is likely to overlook a sions of several 'typical' joint-bounded blocks, lb is
closely spaced discontinuity set which is parallel to the therefore a modal, rather than a mean, value. This
sampling plane and thus may overestimate average approach was found to be particularly useful during
block size. the present study for small or partly obscured rock
During the present study this risk was therefore outcrops. Ib values can be compared with unit volume
reduced wherever possible by adopting a three- If values provided that the individual block dimensions
dimensional approach for Ir measurements. Three are first expressed as frequencies. The examples given
principal sets defining block characteristics could be in Table 3 demonstrate how similarities between rock

TABLE 3. Ib values for sandstone outcrops in Lesotho compared with If values for the same rocks in neighbouring road cuttings

Rock outcrops Cuttings Difference

Ib (m) If (m) If (m) A- C B- C
- - x 100 - - x 100
A B (1) C (2) C C

2.25 1.69 1.69 33% 0

3.33 2.08 2.19 52% -5%
2.50 1.85 1.89 32% -2%
3.57 2.86 3.06 17% - 6%
2.67 1.89 1.96 36% -4%
1.67 1.49 1.41 18% 6%
1.23 1.02 1.16 6% - 12%
2.25 1.89 1.86 21% 2%
1.13 0.93 0.89 27% 4%
1.50 1.39 1.36 10% 2%
Average difference: 25% _+4%
Notes:
(1) Calculated from block size dimensions expressed as frequencies.
(2) Three-dimensional fracture spacing index, calculated on a unit volume basis.
 EXCAVATABILITY OF ROCK: THE G R A P H I C A L METHOD 151

TABLE 4. Effects of block shape on If values

Discontinuity spacing (m) Block If (from If (from Ratio

shape spacings) frequencies) B Remarks
Set 1 Set 2 Set 3 -- x 100
A B A
0.2 3.0 3.0 Tabular 2.07 (0.53) 26% Use If value
0.3 0.5 2.5 Columnar I. 10 (0.52) 47% calculated from
0.3 2.0 2.1 Tabular 1.47 (0.69) 47% actual spacings

0.3 2.0 2.0 Tabular 1.43 0.69 48% Use both methods to
0.5 0.8 1.1 Blocky-tabular 0.80 0.72 90% give a range of values
0.1 0.5 0.6 Tabular 0.40 0.22 55% or a mean value

0.1 0.5 0.5 Tabular (0.37) 0.21 57% Use If value calculated
0.3 0.4 0.5 Blocky (0.40) 0.38 95% on a unit volume basis
0.05 0.2 0.3 Flaggy (0.18) 0.11 61% (frequencies)

outcrops and neighbouring cuttings may become more In spite of these difficulties, a three-dimensional
apparent once Ib values have been converted in this If value gives a more realistic impression of block
way. size than can be achieved using a two-dimensional
The method of calculating Ir on a unit volume basis approach. The method used to calculate the average
has an important limitation. For tabular blocks the fracture spacing should always be recorded, preferably
value becomes unrealistically low as the two largest with the mean or modal spacing of each set. Where
dimensions become very wide whilst the third remains more precise information is required, contoured stereo-
constant, and a similar problem may occur for plots should be used to determine the number of sets
columnar blocks. The authors' case studies did not and their orientation. Priest (1993) described statistical
include blocks with such extreme dimensions but methods for analysing orientation data and examined
Minty & Kearns (1983), for example, reported mar- discontinuity frequency and spacing in detail. Com-
ginal ripping conditions using a D355A tractor in a puter techniques may be utilized, such as those
horizontally bedded shale which had few joints. The proposed by Wang, Latham & Poole (1991) for
simplest solution is to calculate If from actual spacings, determining the complete range of in situ block sizes
rather than frequencies, where any of the principal and shapes within a rock mass.
discontinuity sets has a spacing greater than 2 m. This
emphasizes the largest block dimensions, and the value
obtained is approximately equivalent to lb. In order to Strength measurements
avoid dramatic variations in If values, the following
procedures are suggested. The strength of the intact rock was determined on site
using the point load test on irregular lumps. Test
• Where the largest dimension is greater than 2 m (i.e.
results were corrected for specimen size and shape, and
very widely spaced), calculate If from actual
expressed as the standard point load index, Iso0), using
spacings.
procedures suggested by ISRM 0985). Point load
• Where the largest dimension is between 0.6 m and
strength has also been used for other excavatability
2 m (i.e. widely spaced), calculate It both from
assessment systems (see Table 2). Some workers have
spacings and on a unit volume basis to give a range
found that point load values lower than 2.0 MPa can
which includes the 'worst case' value. Alternatively,
be unreliable, but during the present study individual
the mean of the two values can be used to give a
test results were found to be reasonably consistent for
single point on the graph.
rocks with point load values of 0.5 MPa or higher,
• Where the largest dimension is less than 0.6m,
provided that the specimens had similar weathering
calculate If on a unit volume basis.
and microstructural characteristics. Rocks weaker than
The effects of block shape on If values are illustrated this gave much more variable results and often fell
by the examples in Table 4. The difference between the apart as pressure was applied to hold them between the
two values becomes progressively less as the rock platens of the testing machine; their strength had to be
blocks become equidimensional. Values calculated by estimated using scratch and hammer tests (e.g. Anon
either method can be adjusted upwards where the 1990). This endorsed findings by Hawkins (1986), who
orientation of blocks is unfavourable for ripping (see recommended that point load values lower than
Table 5 and later discussion). 0.5 MPa should not be used.
152 G. S. P E T T I F E R & P. G. FOOKES

TABLE 5. Effects of block shape and orientation on rippability (based partly on Weaver 1975, Kirsten 1982 and Minty
& Kearns 1983)
Shape class Suggested adjustment of data
(ISRM 1981) Characteristics Significance for rippability for revised excavatability graph
Massive Few or very widely spaced Very unfavourable. None: accounted for by very high
discontinuities. If value.
Blocky Approximately equidimensional. Ripping becomes progressively None: accounted for by If and I s
Typically occurs in sedimentary more difficult as discontinuity measurements.
rocks with subhorizontal bedding spacing and the strength of the
and orthogonal joints; also some intact rock increase. Ripping may
granites (Fig. 3a). Where two of be easier where the run direction
the discontinuity sets are not is normal to the strike of any
orthogonal, blocks are rhombic. vertical discontinuities.
Tabular One dimension is considerably Favourable where the run direc- Accounted for by If value where
smaller than the other two.* Typi- tion is down-dip in inclined strata, there is no restriction on the run
cally occurs in thinly bedded sedi- particularly in very thinly bedded direction. Where the run direction
mentary rocks with widely spaced or cleaved rocks. In less thinly is parallel to the strike of the bed-
joints; also foliated metamorphic bedded strata, large slabs may ding or foliation, increase If value
rocks (Fig. 3b). Flaggy and slaty jam under the tractor. It may be by 20% (dip 10-50 degrees) or
rock masses are extreme examples possible to overcome this by rip- 40% (dip 50-90 degrees).
of this class (Figs 3c and 3d). ping in a direction at 45 degrees
(* 'thickness much less than length to the dip direction, so as to tip
or width'--BS 5930: 1981) the slabs, which may then be
broken by driving over them.
Optimum conditions for ripping
where the dominant discontinui-
ties dip at 45 degrees.
Unfavourable where the closest Increase If value by 40%.
spaced discontinuity set is sub-
horizontal (dip 0-10 degrees).
Columnar One dimension is considerably Very unfavourable where Increase If value by 50% where
larger than the other two.* Occurs dominant discontinuities are columns are subvertical.
predominantly in fine-grained subvertical.
igneous rocks (Fig. 3f).
(* 'height much greater than cross Blocks are generally well inter- Difficult conditions emphasized by
section'--BS 5930:1981) locked and difficult to dislodge. high Is values.
Irregular Wide variations of block size and Fair to unfavourable due to very Increase If value by 30%.
shape due to random joint pat- good block interlock and uncer-
tern. Occurs in some recrystallized tain behaviour in relation to run
limestones and quartzites, and in direction.
some medium- to coarse-grained
igneous rocks (Fig. 3e).
Crushed Heavily jointed to 'sugar cube' Very favourable. None: accounted for by very low
structure. If value.

Rock strength is defined in BS 5930:1981 in terms o f compressive strength tests, taking particular note o f
unconfined compressive strength, which has a linear features such as microfractures and lamination planes.
correlation with point load strength. A conversion The results indicated that the data could be c o m b i n e d
factor of 24 is widely used for point load values, but to give a single conversion factor of 20.
factors quoted in the literature range from 8 to 43. A conversion factor of 20 has been used in the
Laboratory investigations were carried out in order to present study, and this value was also considered
establish site specific conversion factors for two suites appropriate by Pells (1985) for foundation design
of rocks. The materials were tested in the saturated investigations in the Hawkesbury Sandstone. It is
condition and ranged from moderately weak sand- advisable, however, to determine a specific correlation
stones to extremely strong igneous rocks. Care was for a particular project or geological formation.
taken to compare specimens which showed a similar Conversion problems are minimized if the point load
m o d e of failure in both point load and unconfined index is used directly for the excavatability graph, but
 EXCAVATABILITY OF ROCK: THE GRAPHICAL METHOD 153

C
0
6u!q:)e~)l ,,,.._ C '--
o),eJopotu 0 o~

>- I- I >- ' 1 _ i = I -

i~i" -.. • .'~ ~:. ~,/~...,,,,. ~ . , , ~ - ~ - - - " r -I- ~ "

0
.~P,,~ • • . ' o y . |,. , ~- . ,-o| _ . U L- , L L
• • ,...-
~ If f :----
__ ~,. :~.-:,::,,o , ~ , ~ : t : : : ~ ~ . , ~-~T-~ '

• :.:..~g~. ~.,~
",t,~,,~.',~C2-----'-~p~,;~_.
:..,.>.:_.-.,~,,o
" " • ~' ."
~,c-c~ : -
I u
: ,-~
•
~,.~.~>.:.-. .,... ,. ,., ~ V T - ~ T ~ ° -c ~
~ ~ ~ " b - . . . : . , '.:~,~,,~, , ~ ~ ~ _ ~ ,- ~ "-
"~--'" ~ I '-1-- ~ '*" e-

"10 r- t-
® ~.-_
..-_( ~,- ~;:
,,-k

I ~lg~ ° I ,oo, ...,. d

~g ~.~
~ ...{~ ,,....~ ~ ~o ;>.

..,,, ~ -~

-- 0
SUOltnlos ID ,-
l=
~l!OS 6u!s!J e- e- ¢~
0
0 r~ 0
s.

~, I>1~1-1- 03
-. ", • i'.;l,I. ', ' ~'o ~
o3
r- i- ,.Q
,,,_,,,

~ ~ o
,,j=
o ~ ° 5 c.)~, 2
i I
:~.~_ ,- <I: --
- "~',: - %, "" .':?LIL. -- 0 E
,- q)
" '.~"q ll" "
0 "1o
)looJ ~, u e J e d
",I = I,-'I 1,4 m

'- o')
0
~-.~ --- .,,2- r- O~
c
°_

u o s e e s ,~Jp u! u ®
_.,;)SlJ s u o ! ; n l o s I!OS
UOSEOS 0
-- x
u! 6u!qoe_..el ~.~°u~J|s"- t- u
0 "-

; I >1--I-I=i - •10 0 •
a.
~
, , , .~:;;',--~i,,'%~-~" J
O.
0

G) "10
® ,-,

-- 3
0
([ ~,1~ ~ .c:,~,:.~ ~ . . , = . . - r - / . I
c
,. tl , , . % " - : ; ~ e.,

; ; ._o
-..::.-.~ Q ~ a)
--I r~ s.

I U J ~looJ l u e J e d

~,,~ ~. ~.~
154 G. S. PETTIFER & P. G. FOOKES

GL,
Weathering Grades
L',"..'-
-£ V-z : _ (BS 5 9 3 0 : 1 9 8 1 )
• • • • • • h i g h l y to c o m p l -
etely weathered,
-. \ \, IV-V
m•__•_ moderately
OO ~" ~'O~, \ w e a t h e r e d , III
- .:.\ \ slightly
w e a t h e r e d , II
fresh, I

E
\

10 \

12

¢-

14

16
0 2 4 6 8 10 12 14 16 18 20 22 24 26

POINT LOAD INDEX (MPa)

(I) Partly from Miglio[19811

FIG. 5. Increase of point load strength with depth in various rock weathering profiles.

field estimates or laboratory test results in terms of ing conditions (e.g. Anon 1990). Outcrops are selective,
unconfined compressive strength must still be con- and may therefore represent rock that has proved more
verted. resistant to weathering or erosion than other strata of
similar age in the area, and hence may be stronger or
more massive. The same material is likely to be
Geomorphological setting encountered in shallow cuttings for new roads. Existing
man-made exposures which are relatively deep are
It is important that field observations are interpreted likely to show a greater range of rock types, strengths,
with respect to the local geomorphology and weather- discontinuity patterns and weathering effects.
 EXCAVATABILITY OF ROCK: THE GRAPHICAL METHOD 155

Mechanical weathering of rocks is dominant in cold degraded cliff in the background was composed of
or dry climates, and typically opens existing disconti- large blocks of moderately weathered sandstone similar
nuities and forms new ones by rock fracture. Chemical to those shown at close quarters in Fig. 7. This thickly
weathering, dominant in the humid tropics, initially bedded sandstone was intersected by the next cutting
causes discoloration of the rock fabric adjacent to up the road and had to be broken up by blasting.
discontinuities and eventually leads to the complete Figure 8 shows the variety of rock types encountered
decomposition of many minerals. Mature weathering in another of the cuttings and identifies those materials
profiles comprise soil horizons overlying various grades which were tipped using a D9 tractor and those which
of weathered rock. Some typical examples are shown in were blasted; note the greater degree of weathering in
Fig. 4, based partly on illustrations by Press & Siever the dolerite compared with the sedimentary rocks.
(1986). Laterites often develop in tropical savanna
climates and form a surface or near-surface duricrust
(Fig. 4a). In more arid regions, calcium carbonate is
precipitated in the B horizon of the soil where there is
limestone bedrock (Fig. 4b), and this process may
ultimately form strong calcretes which are underlain by
much weaker materials. The transition from soil to
fresh rock tends to be more gradational in temperate
climates (Fig. 4c).
Rock strength generally increases as weathering
grade decreases. There will be either a gradual or an
abrupt change in strength with depth, depending on
the resistance of the constituent minerals and rock
fabric to decomposition or disintegration. This is
illustrated by Fig. 5, which has been compiled from
observations made by the authors and by Miglio
(1981). FIG. 6. Shallow road cutting in Lesotho ripped using a
Present-day landforms may be the result of several Komatsu D155A tractor (case study 58).
phases of weathering and erosion. Britain, for example,
appears to have experienced more intense chemical
weathering under subtropical climates in the Tertiary.
The deep profiles which developed were largely
removed by erosion during the Ice Age but have been
preserved locally, notably in parts of southwest
England, and often have periglacial or temperate
features superimposed on them.
Some of the fieldwork for the present study was
carried out in southwestern Lesotho at a latitude of
30°S and an altitude of 1500 m. The present climate is
characterized by hot, moist summers and cold, dry
winters, so that either chemical or mechanical weather-
ing may predominate. Erosion has stripped much of
the weathered material and produced a succession of
colluvium-mantled planation surfaces terminating at
low sandstone scarps or cliffs, which are thought to be FIG. 7. Sandstone outcrop in Lesotho. This material was
generally inactive at present. This landscape has been blasted in the cutting beyond (case study 77).
described in detail by Schmitz & Rooyani (1987). Road
cuttings were excavated through gently dipping blocky
or tabular sandstones and flaggy or shaly siltstones and
mudstones of Triassic age, which were locally tran- The revised excavatability graph
sected and metamorphosed by dolerite dykes. Most
domains had two principal joint sets, bedding plane The Appendix gives details of 120 case studies which
discontinuities, and occasional random joints. were used to revise the graph. Of these, 69 cases were
The moderately weak medium bedded sandstones collected by the authors, 22 were collected by
and siltstones shown in Fig. 6 were ripped using a Abdullatif (1982), and a further 29 were from studies
D155A tractor. They formed an area of relatively level by Weaver (1975), Kirsten (1982, 1988), Scoble and
ground which had a thin mantle of colluvium. The Muftuoglu (1984), Singh et al. (1987) and Karpuz et al.
156 G. S. P E T T I F E R & P. G. FOOKES

~m
_.- ==
c-- m Oi

÷0

>.-2 g e~

0
~ o ® c 1,-
O."2

;!
"2

+0 "" 0

I ¢,-q

@@

1!Iiii_:.
[,....
0 ~ ~ ¢,,,..

•t~ O .at @---- "¢:;I

4.0

O
r-
E
";== ~e - \ I.
@ ,.J
o.

ill \
O~
O~ r-,
m
0,0

.4,.0
~0
e30~
2

OO O
O~
s,,- O
.,,,..
o
e')
o~ ,,;
tO ,~
e-
OO
O
l I/)
O-"I
,.= L+O
01¢D
,%

- / .-,s
~ _u - E0
g . I:::
~ c O
~ 0

r- C
,,., . - x 6
LI.,

'o~ = \ I +o c ~

@ ~

,*..,..,..0
Z

lUUl J.HrOI3H
 EXCAVATABILITY OF ROCK: THE GRAPHICAL METHOD 157
6 I I I I I I | t I I I I I
,o,
1 I
I I
240 0 08 Blast to fracture (1971)
I
sgO 94
0 60 72 )
2
e1~o dl
i i 1,~, 08 ° 077
64 C ~ 111
0 96
046 0 38
b
I ,5~ Irla 6"~,,
2
1 •~,,,,J 08 92
71~. 11o 0 0 93
ILl
\ "e 2, 49
115 --52
,o'°'°,,~ 80 012 O 90
"z 0.6 IOA', j ~
gs\ #8 _-" "-~-..~..._..~
1~31 74 ,A
25 8, ds,,~" 0 20 011
E
o
,
I
53 440 ,o,* ,,3? o,~, 0 97 Oa9
8;,g2 2~i ,~
'°2-gsl
I
:\ 100 e )103 =

~ 0.2. \\ A58
"40
I-- 79 85
;\
_ °
.~.~~o A~,
v
>.
0.1 o~ Blast to Io o s e n (1971)
I- /

fol
Z~ 65
,~ .'.. &Is
Z
F- o.os
~27 \ ,~18...
\ -......
z 41J4 ~I05
o
f/) 45 ,\ zx55 zx"
7,
o~, \ 15ZX~63 AlO6
\
16 ,~1~ z~ \ Ak66
0.02 A
l l i i ~ i i
0.1 0.3 1.0 3.0 10.O 30.0
POINT LOAD INDEX, ISso IMPal

A Materials excavated by d i g g i n g or s c r a p i n g

A Materials ripped using a D8 or l e s s p o w e r f u l tractor,

or w i t h r i p p e r s attached to o t h e r t y p e s of p l a n t

Materials ripped using a D9 tractor

• Materials ripped using a D11,D101700fhpl, D9 f i t t e d

w i t h an i m p a c t r i p p e r , or D9 a f t e r hydraulic breaking

O Materials excavated by b l a s t i n g

Boundaries suggested by F r a n k l i n et al [ 1 9 7 1 ]
and Fookes et al ( 1 9 7 1 1

. ~ Boundary between rippable IDll I and unrippable

materials suggested by K a r p u z et al [ 1 9 9 0 1

FIG. 9. Data used to update the excavatability graph (see Appendix).

158 G.S. P E T T I F E R & P. G. FOOKES

weak- mod.weak-
(BS 5930"1981)
broken by broken in
l e a n i n g on hand by mod. strong ] strong J very str. extr. strong
s a m p l e with hitting with
ha mine r; hammer;
scratched scratched Ill
with thumb- with k n i f e
nail (Anon 1990)

I I I I I I i I I I I I I I I

I BLASTING OR
HYDR. B R E A K I N G +
RIPPING le.g D9]
)r D I G G I N G
cO
O) shovell BLASTING
v-
o,
REQUIRE[
o
¢0
o~
to
(n
m ~. " .~/,I ~ . .

_u

L \%.
\ ,.
,~
%.. \
\ \'% ".o~

\ \ %_ \@ \
\,

HARD X
~,s,j. \ foe,,, "
\ \
DIGGING ", \ \
(e.g CAT 24,* %. \ \ .
backhoe ¢r %
face shovel) X e \ \
\ ',__ % \ \
\ \ "% \
m
m
\
E
GO \ \
\
Ih EASY
> DIGGING
I i I I I I

0.1 0.3 ~1.0 3.0 10.0 30.0

POINT LOAD INDEX, Iss0 IMPal

p r e f e r a b l e to
assess strength for
these weaker rocks
Notes Ill Approx. c o r r e l a t i o n with point load index (UCS=2OIS6o-This Study
12J Where the rock mass s t r u c t u r e is u n f a v o u r a b l e for ripping,
If values should be increased (see Table 5 ) .
Jr= 3 / I f , for a rock mass with three i n t e r s e c t i n g joint sets,
where If is determined on a unit volume b a s i s (see Table 4 ) .
R Q D = 1 1 5 - 3 . 3 Jv approx. (ISRM 1981).

FIG. 10. Revised excavatability graph.

 EXCAVATABILITY OF ROCK: THE G R A P H I C A L METHOD 159

g 0
I
• .~ .~
•~ ~ ~ t~ ~ ~
o
=

q
q~
•: ~ ~ ~ ~ ~

z z ~ ~ z

¢.q =
oo

/: o
'= o ~ "~, ~ "~ "~ o

0
•r~ ~ ~ r~ ~
"-- .~.
oo
.~ ._
]= ~ ' ~ ~ ~" ~ ~..- t..E~
~ o ,,-v

•~ ~.~.~ ~ .~-~
k. •~ ~.~.~ .~ .~.= 0 .-~

8 -
~ ~.~.
~ . .--.

E
~m Ji tt
J
g ¢,q
_
t.-,

"=~~
•~ ~ ~ g ~ .'~

i i
160 G. S. PETTIFER & P. G. FOOKES

(1990). The authors' examples were mainly from road (1990) is based on a fairly limited range of rocks,
projects carried out between 1985 and 1992. Figure 9 mainly marls of various strengths. Their boundary
shows the distribution of the data and includes the between rippable and unrippable materials (see Fig. 9)
1971 subdivisions. has a similar curve to that on the original graph, but
The revised excavatability graph, with new subdivi- the present study indicates that it overestimates the
sions for the operating limits of various categories of possibility of tipping rock masses composed of either
excavating plant, is presented in Fig. 10. The axes are relatively weak very large blocks or very strong
logarithmic as in Franklin et ai.'s graph, block size is medium-sized blocks. Table 6 shows different excavata-
expressed as an average discontinuity spacing, and the bility assessments for the cutting in Fig. 8 compared
point load index is used for rock strength. Fracture with the methods actually used. The authors' revised
spacings have been expressed to two decimal places graph gives similar results in two out of three cases to
and point load values to one place; this was considered those obtained using the methods of either Kirsten
appropriate for the accuracy of the data and was (1982), Scoble & Muftuoglu (1984) or Smith (1986).
suitable for plotting values on the graph.
It is sometimes difficult to determine whether
discontinuities are truly effective, but very strong rocks
such as the acid tuff in case study 103 can be ripped
Conclusions
even though they have tight joints (i.e. 0.1-0.25mm
aperture). Rocks of this type are also likely to be The revised graphical method allows the excavatability
abrasive and this is partly taken into account by their of rock to be assessed rapidly, and is particularly
high strength. No adjustment is required for weather- suited to rippability assessments for civil engineering
ing grade since any reduction in block size or strength excavation work and small-scale quarrying operations.
due to weathering is directly considered by the graph. The graph compares average discontinuity spacing and
The effects of block shape and orientation in relation the strength of the intact rock. These are the two
to ripper run direction must be taken into account, geotechnical properties which have most influence on
however. Conditions are likely to be particularly excavatability, and they can be measured either at
unfavourable for ripping where the dominant disconti- suitable rock outcrops or from drill core. Discontinuity
nuities are either subhorizontal or subvertical, or where spacing values can be increased to allow for structural
the run direction is parallel to the strike of bedded or conditions which are unfavourable for ripping. Other
foliated rocks. Well interlocked columnar or irregular important factors such as the effectiveness of disconti-
blocks are often difficult to dislodge using any form of nuities and the abrasiveness of rock are not directly
mechanical excavation, but the task is made easier taken into account, but similar adjustments could be
where the excavation method allows working benches made to the spacing or strength values.
to be established so that the blocks have greater A three-dimensional discontinuity spacing index,
freedom of movement. Various rock mass geometries calculated on a unit volume basis or from actual
which are considered to be either favourable or spacings, should be used wherever possible since this
unfavourable for ripping are described in Table 5, gives a more realistic impression of average block size
together with adjustments that can be made to data for than a two-dimensional approach. The point load test
the graph. Based on the authors' experience, If values is the preferred method for measuring strength, but the
should be increased by 20-50% where structural strength of weak rocks has to be estimated. Local
conditions are unfavourable. landscape features and weathering patterns will give
The boundaries between areas of equal rippability clues to the character of materials which are likely to
shown on Fig. 10 curve in the opposite direction to the be encountered in excavations.
boundary between ripping and blasting on the original The graph does not necessarily resolve problems of
graph. This reflects difficult ripping conditions in equipment selection or cost, since factors other than
relatively weak rocks with widely spaced discontinuities geotechnical properties may dictate specific working
and in extremely strong rocks. The new boundaries are practices, but it can provide useful guidelines where the
essentially linear over their central sections. For rock classification of materials encountered during exca-
masses with average fracture spacings of less than 2 m vation is disputed. A comprehensive excavatability
and point load strengths lower than 10MPa, the assessment should ideally consider a full range of
equation for the boundary between extremely hard information obtained from boreholes, seismic surveys
ripping and blasting suggests that a rock mass is likely and laboratory tests, so that assessments made using
to be unrippable when I f ( m ) x ls°8(MPa)> 2.6. A different systems can be compared. None of the
similar equation indicates that rock masses within the systems is infallible, and only field trials will determine
above range can be ripped using a D9 tractor when conclusively whether a rock mass can be excavated
If x Is°'8 < 1.8. using particular methods or plant. Predictions made
The modified graph published by Karpuz et al. using the new graph need to be compared with the
 E X C A V A T A B I L I T Y OF ROCK: THE G R A P H I C A L METHOD 161

performance actually achieved under various geological Geological Society, Engineering Geology Special Publi-
and site conditions, and further revisions will be cation, 2, 59-66.
necessary as excavation technology is advanced. INTERNATIONALSOCIETYFOR ROCK MECHANICS(ISRM) 1981.
Suggested Methods for Rock Characterization, Testing
and Monitoring, BROWN, E. T. (ed.). Pergamon, Oxford.
ACKNOWLEDGEMENTS.The authors gratefully acknowledge the 1985. Suggested Method for Determining Point Load
advice and assistance given by many individuals and Strength (to replace original document published in
organizations over a number of years. A specific assignment 1972). International Journal of Rock Mechanics, Mining
with Messrs. Sandberg, Consulting Engineers, London, car- Science and Geomechanics Abstracts, 22, 51-60.
fled out for the Government of Lesotho, provided the initial KARPUZ, C. 1990. A classification system for excavation of
stimulus and framework for this study. surface coal measures. Mining Science and Technology,
11, 157-163.
, PASAMEHMETOGLU,A. G., BOZDAG, T. & MUFTUOGLU,
Y. V. 1990. Rippability assessment in surface coal
References mining. In: SINGHAL, R. K. • VAVRA, M. (eds)
Proceedings of the 4th International Symposium on Mine
Planning and Equipment Selection, Calgary. Balkema,
ABDULLATIF, O. M. 1982. Correlation Between Strength and Rotterdam, 315-322.
Fracture Spacing of Rocks and Ease of Excavation. MSc KIRSTEN, H. A. D. 1982. A classification system for
Dissertation, Queen Mary College, University of Lon- excavation in natural materials. The Civil Engineer in
don. South Africa, 24, 293-308.
- - & CRUDEN, O. M. 1983. The relationship between rock - - 1988. Case histories of groundmass characterization for
mass quality and ease of excavation. Bulletin of the excavatability. In: KIRKALDIE, L. (ed.) Rock Classifi-
International Association of Engineering Geology, 28, cation Systems for Engineering Purposes. ASTM Special
183-187. Technical Publication 984, 102-120. Philadelphia, USA.
ANON 1977. The description of rock masses for engineering MIGLIO, B. F. 1981. The engineering properties of weathered
purposes: Working Party Report. Quarterly Journal of granite from Dartmoor, South-West England. MSc Dis-
Engineering Geology, 10, 355-388. sertation, Queen Mary College, University of London.
1990. Tropical residual soils: Working Party Report. MINTY, E. J. & KEARNS, G. K. 1983. Rock mass workability.
Quarterly Journal of Engineering Geology, 23, 1-101. In: KNIGHT, M. J., MINTY, E. J. & SMITH, R. B. (eds)
BARTON, N., LIEN, R. & LUNDE, J. 1974. Engineering Collected Case Studies in Engineering Geology, Hydro-
classification of rock masses for the design of tunnel geology and Environmental Geology. Geological Society
support. Rock Mechanics, 6, 189-236. of Australia, Special Publication, 11, 59-81.
BmNIAWSKI, Z. T. 1973. Engineering classification of jointed PELLS, P. J. N. 1985. Engineering properties of the Hawkes-
rock masses. The Civil Engineer in South Africa, 15, 335- bury Sandstone. In: PEELS, P. J. N. (ed.) Engineering
344. Geology of the Sydney Region. Balkema, Rotterdam,
BRITISH STANDARDS INSTITUTION 1981. Code of Practice for 179-197.
Site Investigations: BS5930. British Standards Insti- PRESS, F. & SIEVER, R. 1986. Earth, 4th edition. Freeman,
tution, London. New York.
CATERPILLAR TRACTOR COMPANY 1988. Caterpillar Perform- PRIEST, S. D. 1993. Discontinuity Analysis for Rock Engineer-
ance Handbook, 19th edition. Caterpillar Tractor Com- ing. Chapman and Hall, London.
pany, Peoria, Illinois, USA. SCHMaTZ, G. & ROOYANI, F. 1987. Lesotho: Geology,
CIRIA/CUR 1991. Manual on the Use of Rock in Coastal and Geomorphology, Soils. National University of Lesotho,
Shoreline Engineering. Special Publication, 83. Construc- Roma, Lesotho.
tion Industry Research and Information Association, SCOBLE, M. J. & MuFruoGLU, Y. V. 1984. Derivation of a
London, U.K.; Report 154. Centre for Civil Engineering diggability index for surface mine equipment selection.
Research and Codes, Gouda, Netherlands. Mining Science and Technology, 1, 305-322.
FOOKES, P. G., DEARMAN, W. R. & FRANKLIN, J. A. 1971. SINGH, R. N., DENBY, B. & EGRETLI, I. 1987. Development of
Some engineering aspects of rock weathering with field a new rippability index for coal measures excavations.
examples from Dartmoor and elsewhere. Quarterly Proceedings of the 28th U.S. Symposium on Rock
Journal of Engineering Geology, 4, 139-185. Mechanics, Tucson, Arizona. Balkema, Boston, 935-943.
FRANKLIN, J. A., BROCH, E. & WALTON, G. 1971. Logging the SMITH, H. J. 1986. Estimating rippability by rock mass
mechanical character of rock. Transactions of the classification. Proceedings of the 27th U.S. Symposium on
Institution of Mining and Metallurgy, 80A, 1-9. Rock Mechanics, Tuscaloosa, Alabama. AIME, New
HADJIGEORGIOU,J. & SCOBLE, M. J. 1990. Ground characteri- York, 443--448.
zation for assessment of ease of excavation. In" SINGHAL, WANG, H., LATHAM,J-P. & POOLE, A. B. 1991. Predictions of
R. K. & VAVRA, M. (eds) Proceedings of the 4th block size distribution for quarrying. Quarterly Journal
International Symposium on Mine Planning and Equip- of Engineering Geology, 24, 91-99.
ment Selection, Calgary. Balkema, Rotterdam, 323-331. WEAVER, J. M. 1975. Geological factors significant in the
HAWKINS, A. B. 1986. Rock descriptions. In: HAWKINS, A. B. assessment of rippability. The Civil Engineer in South
(ed.) Site Investigation Practice: Assessing BS5930. Africa, 17, 313-316.

Received 20 July 1992; revised typescript accepted 8 November 1993

162 G. S. PETTIFER & P. G. FOOKES

Appendix. Data used to updatethe excavatabUitygraph

Case Fracture Point Method used to Excavation
study spacing load Rock type break up the rock type and
no index, index mass location
If(m) (MPa)

Abdullatif ( 1982)/Abdullatif & Cruden (1983):

1 0.65 4. I Limestone Blasting Q; Kent; Eng
2 0.46 0.8 Sandstone Ripping (t/loader) Q; Kent; Eng
3 0.51 0.3" Sandstone Ripping (t/loader) Q; Kent; Eng
4 0.57 5.1 Limestone Blasting Q; Oxfds; Eng
5 0.09 0.3* Limestone Digging (f-e/loader) Q; Oxfds; Eng
6 0.03 0.3* Limestone Digging (f-e/loader) Q; Oxfds; Eng
7 0.48 8.2 Limestone Blasting Q; Somset; Eng
8 0.74 9.5 Limestone Blasting Q; Somset; Eng
9 0.41 7.3 Limestone Blasting Q; Somset; Eng
10 0.37 8.5 Limestone Blasting Q; Devon; Eng
11 0.47 20.2 Dolerite Blasting Q; Devon; Eng
12 0.70 11.2 Granite Blasting Q; Devon; Eng
13 0.07 4.3 Shale Ripping (t/loader) Q; Devon; Eng
14 0.05 1.0 Shale Ripping (t/loader) Q; Devon; Eng
15 0.03 0.6 Shale Digging (t/loader) Q; Devon; Eng
16 0.02 0.2* China Clay Digging (D9) Q; Devon; Eng
17 0.56 3.5 Granite Blasting Q; Devon; Eng
18 0.06 0.3* Gravel Digging (f/shovel) Q; Devon; Eng
19 0.02 0.3* Ball Clay Digging (backhoe) Q; Devon; Eng
20 0.45 12.6 Dolerite Blasting Q; Devon; Eng
21 0.30 8.5 Granite Ripping (D9L) Q; Devon; Eng
22 0.20 11.2 Vein Quartz Ripping (D9L) Q: Devon; Eng
Weaver (1975):
23 3.50* 0.5* Quartzite Digging (f-e/loader) Q: S.Africa
Kirsten (1982):
24 3.00" 1.7" Conglomerate •Blasting required' -; S.Africa
Kirsten (1988):
25 0.50 3.1 t Quartzite Blasting RC; S.Africa
26 0.13 2.2t Shale Ripping (D8K) RC; S.Africa
27 0.06 0. It Resid. diabase Digging (RH12) PT; S.Africa
28 0.02 0.5t Shale Digging(RH 12) PT; S.Africa
29 0.83 1.0" Calcrete Ripping (Terex 8230B) BP; S.Africa
30 0.83 3.3* Dolomite 'Dl0 required to rip' BP:S.Africa
Singh et al. (1987)
31 0.47 (1) 1.3 Sandstone Scraping (CAT 63 l) OC; England
32 0.54 (1) 2.3 Sandstone Ripping (D8) OC; England
33 0.61 (2) 4.2 Sandstone Ripping (D9) OC; England
34 0.40 (2) 6.4 Sandstone Ripping (D9) OC; England
35 0.51 (2) 6.0 Sandstone Blasting OC; England
36 0.55 (1) 1.9 Mudstone Ripping (D9) OC; England
Pettifer & Fookes (this study)
37 0.62 1.3 Sandstone Ripping (D9) HC; Lesotho
38 1.43 4.8 Sandstone Blasting HC; Lesotho
39 0.20 8.6 Hornfels Ripping (D9) HC; Lesotho
40 0.20 5.1 Baked siltst Ripping (D9) HC; Lesotho
41 0.60 6.9 Baked siltst Blasting HC; Lesotho
42 0.20 8.1 Quartzite Ripping (D9) HC; Lesotho
43 0.72 0.7 Sandstone Ripping (D9) HC; Lesotho
44 0.37 2.5 Sandstone Ripping (D9) HC; Lesotho
45 1.18 1.4 Sandstone Ripping (D9) HC; Lesotho
46 1.36 3.2 Sandstone Blasting HC; Lesotho
47 0.04 0.3* Siltstone Scraping/digging (RH30 f/shovel) HC; Lesotho
48 0. I I 0.6* Siltstone Digging (RH30 f/shovel) HC; Lesotho
49 0.73 1.5" Sandstone Ripping (D8K) HC; Lesotho
50 0.62 1.5" Sandstone Ripping (D8K) HC; Lesotho
51 0.19 0.6 Sandstone Ripping (D6) HC; Lesotho
52 0.65 2.3 Sandstone Ripping (D155A) HC; Lesotho
53 0.43 0.9 Sandstone Ripping (D 155A) HC; Lesotho
 EXCAVATABILITY OF ROCK: THE G R A P H I C A L METHOD 163

Case Fracture Point Method used to Excavation

study spacing load Rock type break up the rock type and
no index, index mass location
If(m) (Maa)

54 1.50 0.6 Sandstone Ripping (D 155A) HC; Lesotho

55 0.04 0.4* Shale Digging (RH30 f/shovel) HC; Lesotho
56 0.47 0.4* Sandstone Ripping (D8K) HC; Lesotho
57 0.97 1.5" Sandstone Ripping (D8K) HC; Lesotho
58 0.21-0.52 0.6* Sst/siltst Ripping (D155A) HC; Lesotho
59 2.68 1.4 Sandstone Blasting HC; Lesotho
60 2.13 1.5 Sandstone Blasting HC; Lesotho
61 0.06 0.7 Shale Digging (RH30 f/shovel) HC; Lesotho
62 1.13 2. l Sandstone Blasting HC; Lesotho
63 0.03 0.7 Shale Digging (RH30 f/shovel) HC; Lesotho
64 1.47 1.8 Sandstone Blasting HC; Lesotho
65 0.07 1.0 Shale Ripping (D 155A) HC; Lesotho
66 0.02 1.6 Siltstone Ripping (D 155A) HC; Lesotho
67 0.06 0.9 Shale Ripping (D155A) HC; Lesotho
68 3.06 2.1 Sandstone Blasting HC; Lesotho
69 0.07 1.0 Siltstone Digging (RH30 f/shovel) HC; Lesotho
70 0.12 1.5 Siltstone Ripping (D6) HC; Lesotho
71 2.00 5.5 Sandstone Blasting HC; Lesotho
72 2.00 2.7 Sandstone Blasting HC; Lesotho
73 0.20 5.0* Hornfels Ripping (D155A) BP; Lesotho
74 0.47 2.5" Dolerite Ripping (D 155A) BP; Lesotho
75 0.04 4.5 Dolerite Digging (CAT 235 backhoe) HC; Lesotho
76 0.13 0.6* Sst/shale Digging (CAT 235 backhoe) HC; Lesotho
77 1.86 3.8 Sandstone Blasting HC; Lesotho
78 0.94 1.5" Sandstone Ripping (D 155A) HC; Lesotho
79 0.18 0.6* Shale Ripping (D 155A) HC; Lesotho
80 1.96 2.2 Sandstone Ripping (D155A) HC; Lesotho
81 3.00 0.1" Conglomerate Digging (MF 450 backhoe) Q; Devon; Eng
82 0.30 4.5* Sandstone Ripping (D8K) HC; Devon;Eng
83 0.20 3.5* Sandstone Ripping (D8K) HC; Devon;Eng
84 0.50-0.75 5.0* Sandstone Ripping (D9L + hammer)--couldn't rip with D9H HC; Devon; Eng
85 0.15 2.5* Argillite Ripping (D155A) Q; N. Ireland
86 0.30 4.0* Greywacke Ripping (D 155A) Q; N. Ireland
87 0.90 3.4t Calcrete Rock breaking+ tipping (D355A) BP; Libya
88 0.43 6.0* Sandstone Blasting Q; Ireland
89 0.33 18.1 Basalt Blasting Q; Scotland
90 0.72 17.3 Dolerite Blasting Q; Scotland
91 0.60 10.9 Qtz. dolerite Blasting Q; Scotland
92 0.92 9.4 Granite Blasting HC; Hong Kong
93 0.82 16.7 Welded tuff Blasting Q; Hong Kong
94 2.22 10.3 Granite Blasting Q; Hong Kong
95 0.19-0.29 12.7 Welded tuff Blasting HC; Hong Kong
96 1.50 16.8 Welded tuff Blasting HC; Hong Kong
97 0.36 13.2 Welded tuff Blasting HC; Hong Kong
98 0.04 2.5 Sandstone Digging (DH 180 backhoe) Q; S. Wales
99 0.50 8.2 Granite Rock breaking+ tipping (D9N) HC; N. Wales
100 0.25 8.2 Granite Ripping (D9N) HC; N. Wales
101 0.07 0.6* Granite Digging (CAT 245 backhoe) HC; N. Wales
102 0.26-0.42 9.9 Acid tuff Rock breaking+tipping (D9N) HC; N. Wales
103 0.22-0.26 9.9 Acid tuff Ripping (D9N) HC; N. Wales
104 0.42 5.1 Acid tuff Ripping (D9N) HC; N. Wales
105 0.05 1.4 Dolerite Digging (CAT 245 backhoe) HC; N. Wales
Scoble & Muftuoglu (1984):
106 0.03 3.3t Sandstone Digging (CAT 245 backhoe) OC; U.K.
Karpuz et aL (1990):
107 6.00 1. l Conglomerate Blasting OC; Turkey
108 1.00 3.8 Marl Blasting OC; Turkey
109 0.85 3.9 Marl Blasting OC; Turkey
1l0 0.85 4.6 Marl Blasting OC; Turkey
111 1.50 1.9 Marl Ripping (D1 l) OC; Turkey
112 1.75 1.3 Marl Ripping (D11) OC; Turkey
164 G. S. PETTIFER & P. G. FOOKES

Case Fracture Point Method used to Excavation

study spacing load Rock type break up the rock type and
no index, index mass location
If(m) (MPa)
113 1.30 1.7 Marl Ripping (D I 1) OC; Turkey
114 1.05 2.0 Marl Ripping (DI 1) OC; Turkey
115 0.70 1.5 Marl Ripping (D9) OC; Turkey
116 1.00 1.1 Marl Ripping (D9) OC; Turkey
117 1.15 1.5 Marl Ripping (D9) OC; Turkey
118 1.10 1.1 Marl Ripping (D9) OC; Turkey
119 1.10 0.9 Marl Ripping (D9) OC; Turkey
120 2.00 0.2* Marl Ripping (D9) OC; Turkey

Notes:
(1) derived from discontinuity spacing.
(2) derived from fracture index.
• denotes estimated value.
t denotes value derived from unconfined compressive strength using UCS = 20 I s (50)--this study.
Q denotes commercial quarry or pit; BP denotes borrow pit.
OC denotes opencast coal or lignite mine; PT denotes pipeline trench.
HC denotes road cutting; RC denotes railway cutting.
f-e/loader: front end loader.
t/loader: tracked loader.
f/shovel: face shovel.