ISO 3082:1998

INTERNATIONAL ISO
STANDARD 3082
Second edition
1998-08-01
Iron ores — Sampling and sample
preparation procedures
Minerais de fer — Procédures d'échantillonnage et de préparation
des échantillons
A
Reference number
Contents Page
1 Scope .1
2 Normative references .1
3 Definitions .2
4 General considerations for sampling and sample preparation .3
4.1 Basic requirements.3
4.2 Establishing a sampling scheme .4
4.3 System verification.4
5 Fundamentals of sampling and sample preparation .5
5.1 Minimization of bias .5
5.2 Overall precision.7
5.3 Quality variation.9
5.4 Sampling precision and number of primary increments .10
5.5 Precision of sample preparation and overall precision.12
6 Methods of sampling.13
6.1 Mass basis sampling.13
6.2 Time basis sampling.15
6.3 Stratified random sampling within fixed mass or time intervals .16
7 Sampling from moving streams .16
7.1 General.16
7.2 Safety of operations .17
7.3 Robustness of sampling installation.17
©  ISO 1998
All rights reserved. Unless otherwise specified, no part of this publication may be reproduced
or utilized in any form or by any means, electronic or mechanical, including photocopying and
microfilm, without permission in writing from the publisher.
International Organization for Standardization
Case postale 56 • CH-1211 Genève 20 • Switzerland
Internet iso@iso.ch
Printed in Switzerland
ii
©
ISO ISO 3082:1998(E)
7.4 Versatility of sampling system. 17
7.5 Primary samplers. 17
7.6 Secondary and subsequent samplers. 22
7.7 On-line sample preparation . 22
7.8 Checking precision and bias. 27
7.9 Cleaning and maintenance . 27
7.10 Example of a flowsheet. 27
8 Sampling from stationary situations . 29
8.1 General . 29
8.2 Sampling from wagons . 29
8.3 Sampling from ships, stockpiles and bunkers . 30
9 Stopped-belt reference sampling. 30
10 Sample preparation . 31
10.1 Fundamentals . 31
10.2 Method of constituting partial samples or a gross sample. 33
10.3 Mechanical methods of division . 35
10.4 Manual methods of division . 39
10.5 Preparation of sample for size determination . 43
10.6 Preparation of sample for moisture determination. 43
10.7 Preparation of test sample for chemical analysis . 44
10.8 Example of sample preparation process. 47
11 Packing and marking of sample. 48
Annex A (informative) Checklist for mechanical sampling systems . 49
Annex B (normative) Equation for number of increments. 54
Annex C (informative) Alternative method of taking the reference sample . 57
Annex D (normative) Procedure for determining the minimum mass of divided gross sample for size
determination using other mechanical division methods. 62
Annex E (normative) Riffle dividers . 65
Annex F (informative) Bibliography. 67
iii
©
Foreword
ISO (the International Organization for Standardization) is a worldwide federation of national standards bodies (ISO
member bodies). The work of preparing International Standards is normally carried out through ISO technical
committees. Each member body interested in a subject for which a technical committee has been established has
the right to be represented on that committee. International organizations, governmental and non-governmental, in
liaison with ISO, also take part in the work. ISO collaborates closely with the International Electrotechnical
Commission (IEC) on all matters of electrotechnical standardization.
Draft International Standards adopted by the technical committees are circulated to the member bodies for voting.
Publication as an International Standard requires approval by at least 75 % of the member bodies casting a vote.
International Standard ISO 3082 was prepared by Technical Committee ISO/TC 102, Iron ores, Subcommittee SC 1,
Sampling.
This second edition cancels and replaces the first edition (ISO 3082:1987), together with ISO 3081:1986 and
ISO 3083:1986, of which it constitutes a collation and technical revision.
Annexes B, D and E form an integral part of this International Standard. Annexes A, C and F are for information
only.
iv
©
INTERNATIONAL STANDARD  ISO ISO 3082:1998(E)
Iron ores — Sampling and sample preparation procedures
Warning — This International Standard may involve hazardous materials, operations and equipment, and
does not purport to address all of the safety issues associated with its use. It is the responsibility of the
user of this Standard to establish appropriate health and safety practices and determine the applicability of
regulatory limitations prior to use.
1 Scope
This International Standard gives
a) the underlying theory;
b) the basic principles for sampling and preparation of samples;
c) the basic requirements for the design, installation and operation of sampling systems.
for mechanical sampling, manual sampling and preparation of samples taken from a lot under transfer to determine
the chemical composition, moisture content and size distribution of the lot. Sampling and sample preparation
procedures for physical testing are specified in ISO 10836.
The methods specified in this International Standard are applicable to both the loading and unloading of a lot by
means of belt conveyors and other ore handling equipment to which a mechanical sampler may be installed or
where manual sampling may safely be conducted.
The methods are applicable to all iron ores, whether natural or processed (for example concentrates and
agglomerates, such as pellets or sinters).
2 Normative references
The following standards contain provisions which, through reference in this text, constitute provisions of this
International Standard. At the time of publication, the editions indicated were valid. All Standards are subject to
revision, and parties to agreements based on this International Standard are encouraged to investigate the
possibility of applying the most recent editions of the standards indicated below. Members of IEC and ISO maintain
registers of currently valid International Standards.
ISO 565:1990, Test sieves — Metal wire cloth, perforated metal plate and electroformed sheet — Nominal sizes of
openings.
1)
ISO 3084:— , Iron ores — Experimental methods for evaluation of quality variation.
ISO 3085:1996, Iron ores — Experimental methods for checking the precision of sampling.
ISO 3086:1998, Iron ores — Experimental methods for checking the bias of sampling.

1)  To be published. (Revision of ISO 3084:1986)
©
ISO
2)
ISO 3087:— , Iron ores — Determination of moisture content of a lot.
ISO 4701:1985, Iron ores — Determination of size distribution by sieving.
ISO 10836:1994, Iron ores — Method of sampling and sample preparation for physical testing.
ISO 11323:1996, Iron ores — Vocabulary.
3 Definitions
For the purposes of this International Standard the definitions contained in ISO 11323 as well as those given below
apply.
3.1
lot
discrete and defined quantity of ore for which quality characteristics are to be assessed
3.2
increment
quantity of ore collected in a single operation of a sampling device
3.3
sample
relatively small quantity of ore, so taken from a lot as to be representative in respect of the quality characteristics to
be assessed
3.4
partial sample
sample, consisting of less than the complete number of increments needed for a gross sample
3.5
gross sample
sample, comprising all increments, entirely representative of all quality characteristics of a lot
3.6
test sample
sample, prepared to meet all specific conditions for a test
3.7
test portion
part of a test sample that is actually and entirely subjected to the specific test
3.8
stratified sampling
sampling of a lot carried out by taking increments from systematically specified positions and in appropriate
proportions from identified parts called strata
NOTE —  Examples of strata, based on time, mass or space, include production periods (e.g. 5 min), production masses
(e.g. 1 000 t), holds in vessels, wagons in a train or containers.
3.9
systematic sampling
selection of increments at regular intervals from a lot

2)  To be published. (Revision of ISO 3087:1987)
©
ISO
3.10
mass basis sampling
sampling carried out so that increments are taken at equal mass intervals, increments being as near as possible of
uniform mass
3.11
time basis sampling
sampling carried out so that increments are taken from free falling streams, or from conveyors, at uniform time
intervals, the mass of each increment being proportional to the mass flow rate at the instant of taking the increment
3.12
proportional sample division
division of samples or increments such that the mass of each retained divided portion is a fixed proportion of the
mass being divided
3.13
constant mass division
division of samples or increments such that the retained divided portions are of almost uniform mass, irrespective of
variations in mass of the samples or increments being divided
NOTE —  This method is required for sampling on a mass basis. "Almost uniform" means that variations in mass are less than
20 % in terms of the coefficient of variation.
3.14
split use of sample
separate use of parts of a sample, as test samples for separate determinations of quality characteristics
3.15
multiple use of sample
use of a sample in its entirety for the determination of one quality characteristic, followed by the use of the same
sample in its entirety for the determination of one or more other quality characteristics
3.16
nominal top size
smallest aperture size, within the range of the R20 Series (in ISO 565, square opening), such that no more than 5 %
(m/m) of an ore is retained on the sieve
4 General considerations for sampling and sample preparation
4.1 Basic requirements
The basic requirement for a correct sampling scheme is that all parts of the ore in the lot have an equal opportunity
of being selected and becoming part of the partial sample or gross sample for analysis. Any deviation from this
basic requirement can result in an unacceptable loss of accuracy and precision. An incorrect sampling scheme
cannot be relied on to provide representative samples.
The best sampling location to satisfy the above requirement is at a transfer point between conveyor belts. Here, the
full cross-section of the ore stream can be conveniently intercepted at regular intervals, enabling representative
samples to be obtained.
In-situ sampling of ships, stockpiles, containers and bunkers is not permitted, because it is impossible to drive the
sampling device down to the bottom and extract the full column of ore. Consequently, all parts of the lot do not have
an equal opportunity of being sampled. The only effective procedure is sampling from a conveyor belt when ore is
being conveyed to or from the ship, stockpile, container or bunker.
In-situ sampling from stationary situations such as wagons is permitted only for fine iron ore concentrates, provided
the sampling device, e.g., a spear or an auger, penetrate to the full depth of the concentrate at the point selected for
sampling and the full column of concentrate is extracted.
©
ISO
Sampling shall be carried out by systematic sampling either on a mass basis (see 6.1) or on a time basis (see 6.2),
provided no bias is introduced by periodic variation in quality or quantity. If this is not the case, stratified random
sampling within fixed mass or time intervals shall be carried out (see 6.3).
The methods used for sampling and sample preparation depend on the final choice of the sampling scheme and on
the steps necessary to minimize possible biases and obtain acceptable overall precision.
Moisture samples shall be processed as soon as possible and test portions weighed immediately. If this is not
possible, samples shall be stored in impervious air-tight containers with a minimum of free air space to minimize any
change in moisture content, but should be prepared without delay.
4.2 Establishing a sampling scheme
The procedure for establishing a sampling scheme is as follows:
a) identify the lot to be sampled;
b) ascertain the nominal top size;
c) determine the mass of increment considering the nominal top size, the ore handling equipment and the device
for taking increments;
d) specify the precision required;
e) ascertain the quality variation, s , of the lot in accordance with ISO 3084, or, if this is not possible, assume
W
"large" quality variation as specified in 5.3;
f) determine the minimum number of primary increments, n , to be taken from the lot for systematic or stratified
random sampling;
g) determine the sampling interval in tonnes for mass basis sampling or in minutes for time basis sampling;
h) determine the sampling location and the method of taking increments;
i) take increments having almost uniform mass for mass basis sampling or having a mass proportional to the flow
rate of the ore stream at the time of sampling for time basis sampling. Increments are to be taken at the
intervals determined in f) during the entire period of handling the lot;
j) determine whether the sample is for split use or multiple use;
k) establish the method of combining increments into a gross sample or partial samples;
l) establish the sample preparation procedure, including division, crushing, mixing and drying;
m) crush the sample, if necessary, except for the size sample;
n) dry the sample, if necessary, except for the moisture sample;
o) divide samples according to the minimum mass of divided sample for a given nominal top size, employing
constant mass or proportional division for mass basis sampling, or proportional division for time basis sampling;
p) prepare the test sample.
4.3 System verification
Stopped-belt sampling is the reference method for collecting samples against which mechanical and manual
sampling procedures may be compared to establish that they are unbiased in accordance with procedures specified
in ISO 3086. However, before any bias tests are conducted, sampling and sample preparation systems shall first be
inspected to confirm that they conform to the correct design principles specified in this International Standard.
Inspections shall also include an examination of whether any loading, unloading or reclaiming procedures could
©
ISO
produce periodic variations in quality in phase with the taking of increments. These periodic variations could include
characteristics such as particle size distribution and moisture content. When such cyclic variations occur, the source
of the variations shall be investigated to determine the practicability of eliminating the variations. If this is not
possible, stratified random sampling shall be carried out (see 6.3).
An example of a suitable checklist is provided in annex A. This will quickly reveal any serious deficiencies in the
sampling or sample preparation system and may avoid the need for expensive bias testing. Consequently, sampling
systems shall be designed and constructed in a manner that facilitates regular verification of correct operation.
Regular checks of quality variation and precision shall also be carried out in accordance with ISO 3084 and
ISO 3085 to monitor variations in quality variation and to verify the precision of sampling, sample preparation and
analysis. This is particularly important for new sampling systems or when significant changes are made to existing
systems.
5 Fundamentals of sampling and sample preparation
5.1 Minimization of bias
5.1.1 General
Minimization of bias in sampling and sample preparation is vitally important. Unlike precision, which can be
improved by collecting more increments or repeating measurements, bias cannot be reduced by replicating
measurements. Consequently, the minimization or preferably elimination of possible biases should be regarded as
more important than improvement of precision. Sources of bias that can be completely eliminated at the outset by
correct design of the sampling and sample preparation system include sample spillage, sample contamination and
incorrect extraction of increments, while sources that can be minimized but not completely eliminated include
change in moisture content, loss of dust and particle degradation (for size determination).
5.1.2 Minimization of particle size degradation
Minimization of particle size degradation of samples used for determination of size distribution is vital to reduce bias
in the measured size distribution. To prevent particle size degradation, it is essential to keep free fall drops to a
minimum.
5.1.3 Extraction of increments
It is essential that increments be extracted from the lot in such a manner that all parts of the ore have an equal
opportunity of being selected and becoming part of the final sample for analysis, irrespective of the size, mass or
density of individual particles. If this requirement is not respected, bias is easily introduced. This results in the
following design requirements for sampling and sample preparation systems:
a) a complete cross-section of the ore stream shall be taken when sampling from a moving stream (see 7.5);
b) the aperture of the sample cutter shall be at least three times the nominal top size of the ore, or 30 mm for
primary sampling and 10 mm for subsequent stages, whichever is the greater (see 7.5.4);
c) the speed of the sample cutter shall not exceed 0,6 m/s, unless the cutter aperture is correspondingly
increased (see 7.5.5);
d) the sample cutter shall travel through the ore stream at uniform speed (see 7.5.3), both the leading and trailing
edges of the cutter clearing the ore stream at the end of its traverse;
e) the lips on the sample cutter shall be parallel for straight-path samplers and radial for rotary cutters (see 7.5.3),
and these conditions shall be maintained as the cutter lips wear;
f) changes in moisture content, dust losses and sample contamination shall be avoided;
©
ISO
g) free fall drops shall be kept to a minimum to reduce size degradation of the ore and hence minimize bias in size
distribution;
h) primary cutters shall be located as near as possible to the loading or discharging point to further minimize the
effects of size degradation;
i) a complete column of concentrate shall be extracted when sampling iron ore concentrate in a wagon (see 8.2).
Sampling systems shall be designed to accommodate the maximum nominal top size and flow rate of the ore being
sampled. Detailed design requirements for sampling and sample preparation systems are provided in 7, 8, 9 and 10.
5.1.4 Increment mass
The increment mass required to obtain an unbiased sample can be calculated for typical sampling situations (see
5.1.4.1, 5.1.4.2 and 5.1.4.3). Comparing the calculated masses with the actual increment masses is useful for
checking the design and operation of sampling systems. If the difference is significant, the cause shall be identified
and corrective action taken to rectify the problem.
5.1.4.1 Increment mass for falling stream sampling
The mass of increment, m, in kilograms, to be taken (mechanically or manually) by a cutter-type primary sampler
I
from the ore stream at the discharge end of a conveyor belt is given by:
ql
m = .(1)
l
3,6v
c
where
q is the flow rate, in tonnes per hour, of ore on the conveyor belt;
l is the cutter aperture, in metres, of the primary sampler;
v is the cutter speed, in metres per second, of the primary sampler.
C
The minimum increment mass that can be taken, while still avoiding bias, is determined by the minimum cutter
aperture specified in 7.5.4 and the maximum cutter speed specified in 7.5.5.
For practical reasons, e.g. in the case of lumpy ore, it may be necessary for the cutter aperture to exceed three
times the nominal top size of the ore.
5.1.4.2 Increment mass for stopped-belt sampling
The mass of increment, m , in kilograms to be taken manually from a stopped-belt is equal to the mass of a
I
complete cross-section (of length l ) of the ore on the conveyor. It is given by the equation:
ql
m = .(2)
l
3 600v
B
where
q is the flow rate, in tonnes per hour, of ore on the conveyor belt;
v is the speed of the conveyor belt, in metres per second.
B
The minimum increment mass that can be taken, while still avoiding bias, is determined by the minimum length of
ore removed from the conveyor, i.e., 3d, where d is the nominal top size of the ore, in millimetres, subject to a
minimum of 10 mm.
©
ISO
5.1.4.3 Increment mass for manual sampling using spear or auger
The mass of increment, m , in kilograms to be taken from a wagon in a lot using a spear or an auger of diameter, l ,
I 3
in millimetres, is given by:
prlL
m = .(3)
l
4 000
where
r is the bulk density of the fine ore (particle size < 1 mm), in tonnes per cubic metre;
L is the depth of concentrate in the wagon, in metres.
The minimum increment mass that can be taken, while still avoiding bias, is determined by the minimum diameter of
the spear or auger, i.e., 30 mm.
This method of extracting increments is only applicable to sampling fine iron ore concentrates.
5.2 Overall precision
This International Standard is designed to attain the overall precision, b , at a probability level of 95 %, given in
SPM
table 1 for the total iron, silica, alumina, phosphorus, and moisture contents and the percent size fraction of the lot.
The overall precision for lots of intermediate mass to those shown in table 1 may be obtained by linear interpolation.
Higher precision may be adopted if required. The precision shall be determined in accordance with ISO 3085.
The overall precision, b , is a measure of the combined precision of sampling, sample preparation and
SPM
measurement, and is twice the standard deviation of sampling, sample preparation and measurement, s ,
SPM
expressed as an absolute percentage, i.e.
2 2 2
ss=+s+s .(4)
SPM S P M
2 2 2
bs==22s+s+s .(5)
SPM SPM S P M
s
W
s = .(6)
S
n
where
s is the sampling standard deviation;
S
s is the sample preparation standard deviation;
P
s is the measurement standard deviation;
M
s is the quality variation of the ore;
W
is the number of primary increments.
n
©
ISO
Table 1 — Overall precision, b (values as absolute percentages)
SPM
Approximate overall precision
Quality characteristics (b )
SPM
Mass of lot
(t)
210 000 45 000 0
to to to
270 000 70 000 15 000
Iron content 0,35 0,4 0,5
Silica content 0,35 0,4 0,5
Alumina content 0,1 0,15 0,2
Phosphorus content 0,002 0,003 0,004
Moisture content 0,35 0,4 0,5
Size –200 mm ore –10 mm fraction 3,5 4,0 5,0
mean 20 %
Size –50 mm ore
Size –31,5 +6,3 mm ore –6,3 mm fraction
mean 10 %
Size - sinter feed +6,3 mm fraction 1,75 2,0 2,5
mean 10 %
Size - pellet feed –45 mm fraction
mean 70 %
Size - pellets –6,3 mm fraction 0,7 0,8 1,0
mean 5 %
Equations (4), (5) and (6) are based on the theory of stratified sampling (see annex B for details). The number of
primary increments to be taken for a lot is dependent on the sampling precision required and on the quality variation
of the ore to be sampled. Thus, before the number of primary increments can be determined, it is necessary to
define:
a) the sampling precision, b , to be attained;
S
b) the quality variation, s , of the ore to be sampled.
W
©
ISO
When on-line sample preparation takes place within the sample plant away from the preparation laboratory, the
distinction between the terms sampling and sample preparation becomes unclear. The precision of on-line sample
preparation may be included in either the sampling precision or in the sample preparation precision. The choice
depends on how easy it is to separate the precision of secondary and tertiary sampling from that of primary
sampling. In any event, sample preparation also constitutes a sampling operation, because a representative part of
the sample is selected for subsequent processing.
The most rigorous approach is to break up the sampling standard deviation into its components for each sampling
stage, in which case equation (4) becomes:
2 2 2 2 2
s =+s sss+ + +s .(7)
SPM S1 S2 S3 P M
where
s is the sampling standard deviation for primary sampling;
S1
s is the sampling standard deviation for secondary sampling;
S2
s is the sampling standard deviation for tertiary sampling.
S3
Using this approach, the precision of each sampling stage can be separately determined and optimized, resulting in
a fully optimized sampling and sample preparation regime.
5.3 Quality variation
The quality variation, s , is a measure of the heterogeneity of the lot and is the standard deviation of the quality
W
characteristics of increments within strata for mass-basis systematic sampling. The characteristics to be selected
for determining quality variation include the iron, silica, alumina, phosphorus and moisture contents and the
percentage of a given size fraction.
The value of s shall be measured experimentally for each type or brand of iron ore and for each handling plant
W
under normal operating conditions, in accordance with ISO 3084. The quality variation of the iron ore may then be
classified into three categories according to its magnitude as specified in table 2. In the case of time basis sampling,
if the flow rate of the ore is uniform on the belt, then time basis sampling is the same as mass basis sampling and
ISO 3084 can be applied.
©
ISO
Table 2 — Classification of quality variation s (values as absolute percentages)
w
( )
Quality characteristics Classification of quality variation s
W
Large Medium Small
Iron content s Ä 2,0 2,0 > s Ä 1,5 s < 1,5
W W W
Silica content s Ä 2,0 2,0 > s Ä 1,5 s < 1,0
W W W
Alumina content s Ä 0,6 0,6 > s Ä 0,4 s < 0,5
W W W
Phosphorus content s Ä 0,012 0,012 > s Ä 0,009 s < 0,01
W W W
Moisture content s Ä2,0 2,0 > s Ä 1,5 s < 1,5
W W W
Size of –200 mm ore –10 mm fraction s Ä 10 10 > s Ä 7,5 s < 7,5
W W W
mean 20 %
Size of –50 mm ore
Size of –6,3 mm fraction s Ä 5 5 > s Ä 3,75 s < 3,75
W W W
–31,5 +6,3 mm ore mean 10 %
Size of sinter feed +6,3 mm fraction
mean 10 %
Size of pellet feed –45 mm fraction s Ä 3 3 > s Ä 2,25 s < 2,25
W W W
mean 70 %
Size of pellets –6,3 mm fraction
mean 5 %
Any ore whose quality variation is unknown shall be considered to have "large" quality variation. In this case,
measurements shall be conducted at the earliest possible opportunity in accordance with ISO 3084 to determine the
quality variation.
When separate samples are taken for the determination of chemical composition, moisture content, size distribution,
etc., the quality variation for the individual characteristics shall be adopted. When the sample is used for the
determination of more than one quality characteristic, the largest classification category for quality variation shall be
adopted.
5.4 Sampling precision and number of primary increments
5.4.1 Mass basis sampling
When the value of s is known, the number of primary increments, n , can be calculated for the desired sampling
W 1
precision, b , as follows:
S
 2s 
W
n =   .(8)
b
 
S
©
ISO
This is the preferable method of determining the number of primary increments. However, when the value of s is
W
classified in terms of large, medium or small quality variation in accordance with table 2, table 3 may be used to
determine the minimum number of primary increments required for the sampling precision, b , specified in the table.
S
The theoretical background is given in annex B. In table 3, the sampling precisions have been increased slightly for
smaller lot sizes as a trade off between sampling cost and the uncertainty in the value of the lot.
Table 3 — Example of minimum number of increments required, n , for desired sampling precision, b
1 S
Mass of lot Sampling precision Number of primary
increments
(1 000 t)
(b )
S
(n )
Fe, Al O P –200 –31,5 Pellet Quality variation
2 3
content mm or mm
SiO or content feed, large (L),
–50 mm ores,
moisture medium (M)
–45 mm
ores, –6,3
content or small (S)
fraction
mm
–10 mm
fraction
fraction
Over Up to Sinter Pellets, LM S
feed, –6,3
mm
+6,3
mm fraction
fraction
270 0,31 0,09 0,001 8 1,55 0,77 0,47 260 130 65
210 270 0,32 0,09 0,001 9 1,61 0,80 0,48 240 120 60
150 210 0,34 0,10 0,002 0 1,69 0,84 0,51 220 110 55
100 150 0,35 0,10 0,002 1 1,77 0,88 0,53 200 100 50
70 100 0,37 0,11 0,002 2 1,86 0,92 0,56 180 90 45
45 70 0,39 0,11 0,002 3 1,98 0,98 0,59 160 80 40
30 45 0,42 0,12 0,002 5 2,11 1,05 0,63 140 70 35
15 30 0,45 0,13 0,002 7 2,28 1,13 0,68 120 60 30
0,50 0,14 0,003 0 2,50 1,24 0,75 100 50 25
0 15
NOTE —  The values of n may be increased or decreased to alter the sampling precision. For example, if
the number of increments is 2n , then b will be improved by a factor of 1/ 2 = 0,71; and if it is n /2, then b
1 S 1 S
will be worsened by a factor of 2 = 1,4.
5.4.2 Time basis sampling
The minimum number of primary increments shall preferably be determined using equation (8), but table 3 may also
be used, as specified in 5.4.1.
©
ISO
5.5 Precision of sample preparation and overall precision
5.5.1 General
The precision of sample preparation depends on the choice of the preparation scheme. However, it can be
improved if sample preparation is carried out first on individual increments or partial samples at an appropriate
stage of sample preparation and then the divided increments or partial samples are combined into a gross sample.
The precision of sample preparation and measurement, b , for size determination shall be better than that specified
PM
in table 5 for each ore type.
The overall precision in terms of the standard deviation, s , where sample preparation and measurement are
SPM
carried out on the gross sample, on each of the partial samples or on each of the increments is specified below.
5.5.2 Preparation and measurement of gross sample
When a gross sample for a lot is constituted by combining all increments and n measurements are carried out on
the gross sample, the overall precision will be:
s
2 2 2 M
ss=+s+ .(9)
SPM S P
n
where
s is the precision of preparing a test sample from the gross sample.
P
5.5.3 Preparation and measurement of partial samples
When n partial samples consisting of an equal number of increments are constituted and n measurements are
3 2
carried out on each partial sample, the overall precision will be:
s
2M
s+
P
n
2 2 7
ss=+ .(10)
SPM S
n
where
s is the precision of preparing a test sample from each partial sample.
P
Further, when the above n partial samples are combined into a gross sample at an appropriate (–10 mm or less)
stage after individual sample preparation, and n measurements are carried out on the gross sample, the overall
precision will be:
2 2
s s
2 2 P1 2 M
ss=+ +s + .(11)
SPM S P2
nn
where
s is the precision of preparing each partial sample prior to constituting the gross sample;
P1
s is the precision of preparing a test sample from the gross sample.
P2
©
ISO
5.5.4 Preparation and measurement of each increment
When n measurements are carried out on each increment, the overall precision will be:
s
2M
s+
P
n
2 2 2
ss=+ .(12)
SPM S
n
where
s is the precision of preparing a test sample from each increment;
P
n is the number of primary increments.
Further, when all the increments are combined into a gross sample at an appropriate stage (–10 mm or less) after
individual sample preparation, and n measurements are carried out on the gross sample, the overall precision will
be:
2 2
s s
2 2 P1 2 M
ss=+ +s + .(13)
SPM S P2
nn
where
s is the precision of preparing each increment prior to constituting the gross sample;
P1
s is the precision of preparing a test sample from the gross sample.
P2
NOTE —  Each sample preparation stage has its own variance, so the total variance will be greater than that for a single
stage. It is desirable to use larger samples for those stages of sample preparation for which this does not greatly increase
costs. This needs to be taken into account when optimizing sample preparation schemes.
6 Methods of sampling
6.1 Mass basis sampling
6.1.1 Mass of increment
The mass of increment shall be determined according to 5.1.4.
Increments shall be taken so that they are of "almost uniform mass", i.e., the coefficient of variation of increment
masses shall be less than 20 %. The coefficient of variation, CV, is defined as the ratio of standard deviation, s ,
mass
to the mean value, m , of the mass of the increments, expressed as a percentage as follows:
100s
mass
CV = .(14)
m
For example, if the average mass of increment is to be 100 kg, the increments shall be taken in such a manner that
95 % of the increments vary between 60 kg and 140 kg, with an average of 100 kg. Provision must therefore be
made, either in the manner in which the increments are taken or by subsequent weighing and division of each
increment, to ensure that they have almost uniform mass.
To obtain increments of uniform mass, one or more of the following measures shall be taken:
a) installation of a variable-speed cutter;
b) control of the ore flow on the conveyor belt ahead of the sampling point;
©
ISO
c) installation of equipment which rejects increments of non-uniform mass and immediately restarts the primary
sampler.
If the coefficient of variation of increment masses is 20 % or greater, each increment may be subjected to division
(according to the rules of division) and the quality characteristics determined. Alternatively, divided increments of
"almost uniform mass" may be combined at an appropriate stage of division into a partial sample or a gross sample.
6.1.2 Quality variation
The quality variation shall be determined experimentally in accordance with ISO 3084.
6.1.3 Number of primary increments
The number of primary increments shall be determined in accordance with 5.4.1.
6.1.4 Sampling interval
The mass interval, D , in tonnes, between increments shall be calculated from the equation:
m
m
L
D < .(15)
m
n
where
m is the mass, in tonnes, of the lot;
L
n is the number of primary increments determined in 5.4.1.
The mass interval selected shall be smaller than that calculated above to ensure that the minimum number of
primary increments is greater than that determined in accordance with 5.4.1.
6.1.5 Methods of taking increments
Each increment shall be taken at one time by a single motion or by a complete cycle of the sampling device so that
a full cross-section of the ore stream is taken. Free fall drops of increments shall be kept to a minimum to reduce
size degradation of the ore and hence minimize bias in size distribution.
NOTES
1)  A complete cycle may involve the sampler taking a forward and return cut through the ore stream.
2)  Stopped-belt sampling may also be used to take a full cross-section of the ore stream.
The first increment shall be taken after a randomly selected tonnage has been handled within the first mass interval
after commencing the handling operation. Subsequent increments shall be taken at the fixed mass interval
determined in 6.1.4 until handling of the lot has been completed. When the calculated mass of the sample is less
than that required for testing (size determination, physical testing, etc.), the number and/or mass of the increments
shall be increased.
Either of the following two kinds of cutter may be employed for the primary sampler:
a) a fixed-speed cutter whose cutting speed is constant during the course of handling the entire lot;
b) a variable-speed cutter whose cutting speed is constant while cutting the stream but can be regulated,
increment by increment, according to the flow rate of ore on the conveyor belt.
Sampling shall be carried out at the nearest possible point to the loading or discharging facilities, preferably
immediately before or after the point of weighing.
©
ISO
6.2 Time basis sampling
6.2.1 Mass of increment
The mass of increment shall be proportional to the flow rate of the ore stream at the time of sampling. When a test
sample is prepared from each increment or partial sample, the mass of each increment or partial sample shall be
determined in order to obtain the weighted mean of the quality characteristics for the lot. Alternatively, the tonnage
of ore that the sample represents may be used to obtain the weighted mean.
6.2.2 Quality variation
When the variation in the ore flow rate is less than 20 % in terms of the coefficient of variation, ISO 3084 shall be
used to obtain an approximate value for the quality variation.
6.2.3 Number of increments
The number of primary increments shall be determined in accordance with 5.4.2.
6.2.4 Sampling interval
The time interval, D , in minutes, between increments shall be calculated from the equation:
t
60m
L
D < .(16)
t
qn
max 1
where
m is the mass, in tonnes, of the lot;
L
q is the maximum flow rate, expressed in tonnes per hour, of ore on the conveyor belt;
max
n is the number of primary increments determined in 5.4.2.
The time interval between taking increments that is selected shall be smaller than that calculated to ensure that the
minimum number of primary increments is greater than that determined in accordance with 5.4.2.
6.2.5 Methods of taking increments
Each increment shall be taken at one time by a single motion or by a complete cycle of the sampling device so that
a full cross-section of the ore stream is taken. Free fall drops of increments shall be kept to a minimum to reduce
size degradation of the ore and hence minimize bias in size distribution.
NOTES
1)  A sampler may take a forward and return cut through the ore stream in a complete cycle.
2)  Stopped-belt sampling may also be used to take a full cross-section of the ore stream.
The first increment shall be taken at random within the first time interval from the start of the handling operation.
Subsequent increments shall be taken at the fixed time interval determined in 6.2.4 until handling of the lot is
completed. When the calculated mass of the sample is less than that required for testing (size determination,
physical testi
...