-
International Journal of Rock Mechanics & Mi
ical
li hs i
osto
ining
GR-1
30 Ju
e 12 S
is the wear or loss of material which the rock produces
However, in spite of its popularity, there are some
(Cerchar) de France [12]. According to the original test
is shown in Fig. 1. The testing device mainly features a
where d is the wear at diameter of the stylus tip in mm.
ARTICLE IN PRESS
Corresponding author. Tel.: +30210 772 2208;
From the above description, the Cerchar test can be
classied as an abrasive wear with pressure test [15].
1365-1609/$ - see front matter r 2005 Elsevier Ltd. All rights
reserved.
doi:10.1016/j.ijrmms.2005.06.009
fax: +30210 772 2191.
E-mail address: [email protected] (T.N.
Michalakopoulos).quartz content, and the microstructure of the
material.Abrasiveness is considered to express a
behavioralcharacteristic of the rock rather than a
fundamentalphysical or mechanical property. As a consequence, itis
difcult to implement a testing procedure with astrong, unambiguous
physical basis and a large numberof testing procedures have been
proposed.A number of different rock abrasiveness tests have
been reported [1,57]. Of these tests the Cerchar test hasgained
popularity as a relatively convenient andinexpensive method of
measuring and comparing the
vice, 1, holding the rock sample, and a steel stylus, 4,tting
into a holder, 5. On the stylus a dead weight of70N is applied, 6.
The rock sample vice can be movedunder the steel stylus through a
handwheel, 2, that drivesa screwthread of pitch 1mm/rev.To
determine the CAI value the rock is slowly displaced
by 10mm with a velocity of approximately 1mm/s. Theabrasiveness
of the rock is then obtained by measuring theresulting wear at on
the tip of the steel stylus. The CAIvalue is calculated using the
following formula [14]:
CAI 102d, (1)insert. [y] It is a function of mineralogy,
especially the
on another material such as a TBM cutter or drill bit
specication, West [13] prepared a testing device, whichTechn
The inuence of steel styabrasivenes
T.N. Michalakopoulos, V.G. Anagn
Laboratory of Excavation Engineering, Department of M
Zographou Campus,
Accepted
Available onlin
1. Introduction
During the last decades mechanical rock excavationhas gained a
signicant share in the construction of bothcivil and mining
engineering projects. The machinesused for mechanical rock
excavation are in generalcapital intensive and, in most cases, site
specic, so thata good understanding of the application
conditionsplays a vital role in maximizing machine performance.
Anumber of models have been developed during the lastyears to
evaluate machine performance under variousapplication conditions
[13]. In all such models rockabrasiveness is an intact rock
property used as a majorinput parameter [1,4].According to Atkinson
[5] the abrasiveness of a rockambiguities in the description of the
test setup in thevarious references, especially regarding the
hardness ofthe steel styli used. In this paper the inuence of the
steelstyli hardness on the CAI value is examined.
2. Testing procedure
The Cerchar test has been proposed by the Laboratoiredu Centre
dEtudes et Recherches des Charbonnagesning Sciences 43 (2006)
321327
note
ardness on the Cercharndex value
u, M.E. Bassanou, G.N. Panagiotou
Engineering, National Technical University of Athens,
5780 Athens, Greece
ne 2005
eptember 2005
abrasiveness of rocks [15,811]. The resulting
CercharAbrasiveness Index (CAI) value is used as an inputparameter
by many of the above-mentioned models forestimating mechanical
excavators performance.
www.elsevier.com/locate/ijrmms
-
A formal description of the testing procedure isprovided in the
French standard NF P 94-430-1 [14],which, to the authors knowledge,
is the only formalstandard.Al-Ameen and Waller [16] and Plinninger
et al. [17]
provide signicant details about the testing conditionsand their
inuence on the CAI value.
3. Steel styli specications
According to the French standard NF P 94-430-1 thetip of the
steel stylus applied to the rock sample isformed as a sharp conical
point of cone angle equal to901731 and 1mm length. The overall
length of the steelstylus should be greater than 15mm.The
recommended steel styli geometry is generally the
same in the various references. However, this does notapply
regarding the hardness of the steel styli. It is
obvious that any change in the hardness of the steelstyli alters
signicantly the testing conditions, thusproviding incomparable CAI
values. This has beendemonstrated by Al-Ameen and Waller [16],
whenthey modied the test specications, in order to enablethe
determination of the CAI value of certain coalsamples.In the
original test setup description [12], Valantin
recommended styli made from steel of 2000MPa tensilestrength. No
recommendation was made for the stylihardness.Suana and Peters [18]
recommend steel styli of
ARTICLE IN PRESS
Reference Tensile strength (MPa)
T.N. Michalakopoulos et al. / International Journal of Rock
Mechanics & Mining Sciences 43 (2006) 321327322Valantin [12]
2000
Suana and Peters [18] 2000
Atkinson et al. [8,9]
West [13] 2000
Atkinson [5] 2000
Al-Ameen and Waller [16]
AFNOR [14]
Plinninger et al. [17] 2000Fig. 1. The West testing device for
determining the Cerchar
abrasiveness index (reproduced from [13]).
Table 1
Recommended steel styli specicationsRockwell Hardness HRC 5456
and 2000MPa tensilestrength.Atkinson et al. [8,9] recommend styli
made of EN24
or EN25 steel, heat treated to 610 Vickers Hardness(HRC 56).West
[13] also recommends styli made of EN24 steel
with a tensile strength of 2000MPa, but heat treated toHRC
40.Atkinson [5] does not provide any recommendation
on steel type, but recommends steel of 2000MPa tensilestrength
and HRC 5456.Al-Ameen and Waller [16] follow the recommenda-
tion of Atkinson et al. with styli made of EN24steel, heat
treated to 610 Vickers Hardness, butalso provide a recommendation
for modied styli madeof EN3 steel heat treated to 225 Vickers
Hardness(HRC 19).AFNOR [14], which provides the only formal
test
description, species that the styli must be made of steelheat
treated to HRC 5456.Plinninger et al. [17] suggest the use of
115CrV4 steel
hardened to HRC 55 with a tensile strength of about2000MPa.The
various recommendations can be seen in Table 1.From the authors
experience, although it is clear
from the above that there is a general consensus for stylimade
from steel heat treated to HRC 5456, severalequipment
manufacturers, planners and contractorsrequest CAI values
determined with steel styli of HRC40, as is recommended by
West.
Hardness Steel type
HRC 5456
HRC 56 (HV 610) EN24 or EN25
HRC 40 EN24
HRC 5456
HRC 56 (HV 610) EN24
HRC 19 (HV 225) EN3
HRC 5456
HRC 55 115CrV4
-
4. Test setup and results
The objective of this study has been to investigate theinuence
of steel styli hardness on the CAI valueobtained. To accomplish
this objective the CAI valueof 73 rock samples has been determined
for twodifferent steel styli types. Of the 73 samples, 5 wereeither
extremely fragmented/weathered or too hard totest, thus results for
a total of 68 samples were obtained.The samples were of 6 different
rock types, which can beseen in Table 2.Each sample has been
subjected to 5 individual tests
with each steel styli type, and the mean CAI value andstandard
deviation of the mean corresponding to eachstyli type have been
computed.The specications of the utilized steel types can be
seen in Table 3. Styli of hardness HRC 55 were usedbecause this
is the hardness specied in the formal testdescription, and of HRC
40 because this is therecommendation by West, which is more
frequentlyused in routine Cerchar testing.The CAI values obtained
can be seen in Table 4,
where CAI55 stands for CAI values obtained with steelstyli of
HRC 55, and respectively CAI40 for HRC 40. InTable 5 the absolute
frequency of the obtained CAI
used. Moreover, it is observed that the CAI40 values
aredistributed over a wider range.
Finally, Eq. (2) has been used to relate the CAI55 andCAI40
value classication shown in Table 8.
ARTICLE IN PRESST.N. Michalakopoulos et al. / International
Journal of Rocvalues for each steel styli type can be seen. Fig. 2
showsthe corresponding bar chart.As expected, the CAI40 value was
greater than the
CAI55 value for all rock samples, demonstrating the
Table 2
Sample description
Sample type Sample nos.
(i) Gabbro 13
(ii) Pyroxenite-Gabbro 4
(iii) Mylonite 516
(iv) Recrystalized massive limestone 1724
(v) Serpentinized peridotite 2552
(vi) Limestone 5368
Table 3
Utilized steel styli nominal specications [19,20]
CAI55 CAI40
Steel type CALMAX EN24 (BS970-817M40)
HRC 55 40
C % 0.60 0.40
Ni % 1.80
Cr % 4.50 0.80
Mn % 0.80 0.70
Mo % 0.50 0.25
Si % 0.35
V % 0.20 6. Conclusions
The Cerchar test has gained popularity as a relativelyconvenient
and inexpensive method of measuring and5. Regression analysis
In Fig. 3 the plot of CAI40 versus CAI55 values ispresented. The
relation of the two CAI values has beenexamined and a linear
correlation was observed, yieldingthe following equation:
CAI55 0:110914 0:587356CAI40. (2)The predicted CAI55 values and
the corresponding
residuals for each rock sample are given in Table 4.
Thepredicted CAI55 values are plotted versus the observedones in
Fig. 4, indicating that no signicant anomaliesare present.The
regression analysis is given in Table 6, and the
analysis of variance in Table 7. Since the p-value inTable 7 is
less than 0.01, there is a statistically signicantrelationship
between CAI55 and CAI40 at the 99%condence level.The R2 statistic
indicates that the model as tted
explains 73.9251% of the variability in CAI55. Thecorrelation
coefcient equals 0.859797, indicating amoderately strong
relationship between CAI55 andCAI40. The standard error of the
estimate shows thestandard deviation of the residuals to be
0.393807. Thisvalue has been used to construct the 95% prediction
andcondence bands presented in Fig. 3.In Fig. 5 the studentized
deleted residuals are plotted
versus the predicted CAI55 values. The randompattern indicates
that the selected model describesadequately the observed data.
Moreover, the linearityand homoscedasticity of the residuals are
veried. Theabsolute studentized deleted residuals of 6
observationsare greater than 2.00 and are identied as
potentialoutliers.In Fig. 6 the quantile plot of the
standardized
residuals empirical distribution and the standardnormal
distribution can be seen. This plot veries therequired regression
assumption of normality.signicant impact of steel styli hardness on
the obtainedCAI value. The reasonable general conclusion of
thisinvestigation is that the CAI value of a rock sample
isdecreased by an increase of the hardness of steel styli
k Mechanics & Mining Sciences 43 (2006) 321327 323comparing
the abrasiveness of rocks. However, there are
-
ARTICLE IN PRESS
Table 4
Analysis of the observed CAI values and the linear model
results
Sample no. Observed CAI40 Observed CAI55 Predicted CAI55
Observed residuals Standardized residuals Studentized
deleted residuals
Mean Std deviation
of the mean
Mean Std deviation
of the mean
1 6.35 0.68 3.58 0.28 3.840620 0.260623 0.666800 0.7189372 4.55
0.28 3.66 0.30 2.783380 0.876618 2.242813 2.359460
3 2.49 0.28 2.13 0.57 1.573430 0.556570 1.423975 1.436480
4 4.27 0.39 3.05 0.15 2.618920 0.431077 1.102904 1.118210
5 1.78 0.53 1.25 0.29 1.156410 0.093593 0.239456 0.239430
6 2.85 0.49 2.06 0.19 1.784880 0.275122 0.703895 0.701103
7 0.64 0.12 0.64 0.06 0.486822 0.153178 0.391903 0.401577
8 3.08 0.49 1.73 0.28 1.919970 0.189969 0.486033 0.4832809 1.35
0.38 1.27 0.17 0.903844 0.366156 0.936804 0.949292
10 1.35 0.19 1.45 0.11 0.903844 0.546156 1.397331 1.428140
11 2.87 0.73 1.32 0.29 1.796620 0.476625 1.219437 1.22389012
2.06 0.61 2.44 0.49 1.320870 1.119130 2.863276 3.049710
13 3.51 0.13 2.67 0.08 2.172530 0.497468 1.272764 1.282000
14 3.35 0.85 2.62 0.27 2.078560 0.541445 1.385278 1.397030
15 2.08 0.61 1.30 0.19 1.332610 0.032614 0.083442 0.08311516
5.28 0.29 3.27 0.29 3.212150 0.057848 0.148003 0.152330
17 2.87 0.14 1.75 0.21 1.796620 0.046625 0.119289 0.11838118
2.54 0.21 1.78 0.16 1.602800 0.177203 0.453371 0.450868
19 2.74 0.14 1.91 0.13 1.720270 0.189731 0.485424 0.482590
20 3.63 0.29 2.61 0.24 2.243020 0.366985 0.938925 0.941311
21 4.29 0.28 1.57 0.15 2.630670 1.060670 2.713707 2.89686022
2.64 0.25 1.52 0.25 1.661530 0.141533 0.362110 0.35978723 2.64 0.22
1.70 0.16 1.661530 0.038467 0.098417 0.097696
24 2.74 0.10 1.62 0.22 1.720270 0.100269 0.256537 0.25470825
1.83 0.25 1.45 0.12 1.185770 0.264225 0.676015 0.677586
26 1.04 0.17 1.04 0.05 0.721764 0.318236 0.814202 0.829071
27 2.28 0.46 1.45 0.20 1.450080 0.000085 0.000217 0.00021628
1.58 0.35 1.22 0.08 1.038940 0.181064 0.463249 0.465120
29 1.55 0.31 1.20 0.10 1.021320 0.178685 0.457163 0.459210
30 1.73 0.54 1.37 0.18 1.127040 0.242961 0.621612 0.623550
31 5.67 0.33 3.51 0.22 3.441220 0.068779 0.175971 0.183568
32 4.67 0.20 3.76 0.24 2.853870 0.906135 2.318332 2.453100
33 5.05 0.09 3.61 0.34 3.077060 0.532940 1.363518 1.414910
34 3.91 0.37 2.16 0.32 2.407470 0.247475 0.633161 0.63438435
4.17 0.67 2.69 0.28 2.560190 0.129813 0.332124 0.333268
36 2.92 0.64 1.75 0.24 1.825990 0.075993 0.194426 0.19298437
2.64 0.16 0.79 0.17 1.661530 0.871533 2.229803 2.30173038 1.93 0.27
1.25 0.23 1.244510 0.005490 0.014045 0.014011
39 1.70 0.45 0.97 0.24 1.109420 0.139419 0.356701 0.35724940
2.08 0.29 1.12 0.33 1.332610 0.212614 0.543969 0.54303441 1.93 0.28
0.74 0.12 1.244510 0.504510 1.290781 1.30442042 2.90 0.40 1.30 0.15
1.814250 0.514245 1.315687 1.32301043 3.35 0.47 1.83 0.25 2.078560
0.248555 0.635924 0.63385444 1.40 0.31 0.69 0.13 0.933212 0.243212
0.622254 0.62753545 2.13 0.37 1.37 0.12 1.361980 0.008018 0.020515
0.020424
46 2.26 0.34 1.37 0.36 1.438340 0.068338 0.174841 0.17392147
1.42 0.39 0.64 0.08 0.944959 0.304959 0.780233 0.78793548 3.00 0.53
1.83 0.23 1.872980 0.042981 0.109966 0.10913949 1.52 0.07 0.79 0.17
1.003690 0.213695 0.546735 0.54983950 3.33 0.22 1.57 0.20 2.066810
0.496808 1.271075 1.27868051 2.36 0.14 1.19 0.13 1.497070 0.307073
0.785641 0.78448652 2.67 0.34 1.37 0.26 1.679150 0.309154 0.790965
0.78879653 3.48 0.21 1.32 0.14 2.154910 0.834912 2.136109
2.20226054 3.23 0.14 2.06 0.09 2.008070 0.051927 0.132855
0.131952
55 2.97 0.13 1.81 0.13 1.855360 0.045360 0.116054 0.11517756
3.71 0.10 2.03 0.09 2.290000 0.260004 0.665216 0.66517557 3.38 0.14
1.83 0.10 2.096180 0.266176 0.681007 0.67922158 2.87 0.10 1.93 0.22
1.796620 0.133375 0.341238 0.338905
T.N. Michalakopoulos et al. / International Journal of Rock
Mechanics & Mining Sciences 43 (2006) 321327324
-
ARTICLE IN PRESS
ed CA
f RocTable 4 (continued )
Sample no. Observed CAI40 Observed CAI55 Predict
Mean Std deviation
of the mean
Mean Std deviation
of the mean
T.N. Michalakopoulos et al. / International Journal osome
ambiguities regarding the recommended steel stylihardness.In this
paper the inuence of steel styli hardness on
the CAI value has been investigated. Sixty-eight samplesof
various rock types have been tested with steel styli ofboth HRC 55
and 40. The relation of the two CAI
59 2.92 0.11 1.42 0.18 1.825990
60 3.15 0.16 1.91 0.21 1.961080
61 2.57 0.11 1.58 0.09 1.620420
62 2.85 0.17 1.73 0.15 1.784880
63 2.89 0.21 1.86 0.10 1.808370
64 3.12 0.15 1.86 0.15 1.943460
65 3.13 0.12 2.39 0.18 1.949340
66 3.20 0.26 1.91 0.18 1.990450
67 1.93 0.36 0.91 0.07 1.244510
68 4.67 0.11 2.62 0.20 2.853870
Table 5
Absolute frequency of the observed CAI values
Value range CAI55 CAI40
0.300.50 0 0
0.501.00 8 1
1.002.00 41 15
2.004.00 19 42
4.006.00 0 9
46.00 0 1
0
10
20
30
40
50
0.3-0.5 0.5-1.0 1.0-2.0 2.0-4.0 4.0-6.0 >6.0CAI value
Abs
olut
e fre
quen
cy
CAI40CAI55
Fig. 2. Distribution of observed CAI values.I55 Observed
residuals Standardized residuals Studentized
deleted residuals
k Mechanics & Mining Sciences 43 (2006) 321327 325values has
been examined and a linear correlation wasobserved. The yielded
equation can be used to convertCAI40 values to CAI55, which are
required by the formaldescription of the test.
0.405993 1.038727 1.0392600.051084 0.130698 0.1297690.040418
0.103409 0.1026740.054878 0.140403 0.1393410.051628 0.132090
0.131088
0.083464 0.213541 0.2120470.440663 1.127429 1.130130
0.080452 0.205835 0.2044490.334510 0.855838 0.8586100.233865
0.598340 0.607421
0.00
0.50
1.00
1.50
2.00
2.50
3.00
3.50
4.00
0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00Predicted CAI55
Obs
erve
d CA
I 55
Fig. 4. Plot of predicted versus observed CAI55 values.
0.00
1.00
2.00
3.00
4.00
5.00
0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00
CAI40
CAI 5
5
Fig. 3. Plot of CAI40 versus CAI55 values.
-
ARTICLE IN PRESS
Table 6
Regression analysis
Linear model: y a bxDependent variable (y): CAI55Independent
variable (x): Observed CAI40Data sets: 68
Parameter Estimate Standard error t statistic p value
Intercept (a) 0.110914 0.132092 0.839671 0.4041
Slope (b) 0.587356 0.042938 13.6791 0.0000
Table 7
Analysis of variance
Source Sum of
squares
Degrees of
freedom
Mean
square
F ratio p value
Model 29.0188 1 29.0188 187.12 0.0000
Residual 10.2355 66 0.155084
Total 39.2543 67
Correlation coefcient 0.859797R2 73:9251%Standard error of
estimate 0.393807
-4.00
-2.00
0.00
2.00
4.00
0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00Predicted CAI55
Stud
entiz
ed d
elet
ed re
sidua
ls
Fig. 5. Plot of predicted CAI55 versus studentized deleted
residuals
values.
0.00
0.25
0.50
0.75
1.00
-3.00 -2.00 -1.00 0.00 1.00 2.00 3.00Standardized residual
value
Cum
ulat
ive
prob
abili
ty
Standardized residualsStandard normal distribution
Fig. 6. Standardized residuals normality test.
T.N. Michalakopoulos et al. / International Journal of Rock
Mechanics & Mining Sciences 43 (2006) 321327326References
[1] Howarth DF. Mechanical rock excavationassessment of
cutt-
ability and borability. In: RETC Proceedings. Littleton:
SME;
1987. p. 14564.
[2] Nilsen B, Ozdemir L. Hard rock tunnel boring prediction
and
eld performance. In: RETC Proceedings. Littleton: SME; 1993.
p. 83352.
[3] Cigla M, Ozdemir L. Computer modeling for improved
produc-
tion of mechanical excavators. Salt Lake City: Society for
Mining,
Metallurgy and Exploration Annual Meeting; 2000.
[4] Nelson PP. TBM performance analysis with reference to
rock
properties. In: Hudson JA, et al., editors. Comprehensive
Rock
Engineering, vol. 4. Oxford: Pergamon Press; 1993. p. 26191.
[5] Atkinson RH. Hardness tests for rock characterization.
In:
Hudson JA, et al., editors. Comprehensive Rock Engineering,
vol. 3. Oxford: Pergamon Press; 1993. p. 10517.
[6] McFeat-Smith I. Rock property testing for the assessment
of
tunneling machine performance. Tunnels and Tunnelling
1977;9:
2933.
[7] McFeat-Smith I, Tarkoy PJ. Site investigations for
machine
tunneling contracts. Tunnels and Tunnelling 1980;12:369.
[8] Atkinson T, Cassapi VB. The prediction and reduction of
abrasive wear in mining excavation machinery. In: Tribology
in mineral extractionwar on wear. London: IMechE; 1984.
p. 16574.
[9] Atkinson T, Cassapi VB, Singh RN. Assessment of abrasive
wear
resistance potential in rock excavation machinery. Int J Min
Geol
Eng 1986;3:15163.
[10] Atkinson T, Denby B, Cassapi VB. Problems associated
with rock material properties in surface mining equipment
selection. Trans Instn Min Metall (Sect A: Min Indus) 1986;
95:A806.
[11] Nilsen B, Ozdemir L. Recent developments in site
investigation
and testing for hard rock TBM projects. In: RETC
Proceedings.
Littleton: SME; 1999. p. 71531.
Table 8
Cerchar abrasiveness index (CAI) classication
Classication CAI55 CAI40a
Very low abrasiveness 0.300.50 0.320.66
Low abrasiveness 0.501.00 0.661.51
Medium abrasiveness 1.002.00 1.513.22
High abrasiveness 2.004.00 3.226.62
Extreme abrasiveness 4.006.00 6.6210.03
aValues estimated from Eq. (2).[12] Valantin A. Test Cerchar
pour la mesure de la durete et de
labrasivite des roches. Annexe de lexposee presente aux
Journees
de Information Techniques de creusement, Luxembourg 1973:
13740.
[13] West G. Rock abrasiveness testing for tunneling. Int J Rock
Mech
Min Sci Geomech Abstr 1989;26:15160.
[14] AFNOR. Determination du pouvoir abrasive dune
rochePartie
1: Essai de rayure avec une pointe (NF P 94-430-1). Paris:
AFNOR; 2000.
[15] ISRM. Commission on standardization of laboratory and
eld
tests. Suggested methods for determining hardness and
abrasive-
ness of rocks. Int J Rock Mech Min Sci Geomech Abstr
1978;15:8997.
[16] Al-Ameen SI, Waller MD. The inuence of rock strength
and
abrasive mineral content on the Cerchar Abrasive Index. Eng
Geol 1994;36:293301.
-
[17] Plinninger R, Kaesling H, Thuro K, Spaun G. Testing
conditions
and geomechanical properties inuencing the CERCHAR abra-
siveness index (CAI) value. Int J Rock Mech Min Sci 2003;
40:25963.
[18] Suana M, Peters Tj. The Cerchar Abrasivity Index and its
relation
to rock mineralogy and petrography. Rock Mech 1982;15:17.
[19] Uddeholm. CALMAXPlastic mould and cold work steel.
http://utab.uddeholm.com/b_97.php (accessed 27 February,
2005).
[20] NAMTEC. Low alloy steel, AISI 4340 (annealed). http://
www.namtec.co.uk/metals/samples/ferrous/sample1.html
(accessed
27 February, 2005).
ARTICLE IN PRESST.N. Michalakopoulos et al. / International
Journal of Rock Mechanics & Mining Sciences 43 (2006) 321327
327
The influence of steel styli hardness on the Cerchar
abrasiveness index valueIntroductionTesting procedureSteel styli
specificationsTest setup and resultsRegression
analysisConclusionsReferences