1、raising standards worldwideNO COPYING WITHOUT BSI PERMISSION EXCEPT AS PERMITTED BY COPYRIGHT LAWBSI Standards PublicationTechnical report of theinterlaboratory study for thedetermination of Besatz incommon wheat, rye and durumwheatPD CEN/TR 16324:2012Incorporating corrigendum November 2012ICS 67.06
2、0 Compliance with a British Standard cannot confer immunity fromlegal obligations.This Published Document was published under the authority of theStandards Policy and Strategy Committee on 30 November 2012.The UK participation in its preparation was entrusted to TechnicalCommittee AW/4, Cereals and
3、pulses.A list of organizations represented on this committee can beobtained on request to its secretary.This publication does not purport to include all the necessaryprovisions of a contract. Users are responsible for its correctapplication. The British Standards Institution 2012.Published by BSI St
4、andards Limited 2012PD CEN/TR 16324:2012 PUBLISHED DOCUMENTNational forewordThis Published Document is the UK implementation of CEN/TR 16324:2012.ISBN 978 0 580 80890 6Amendments/corrigenda issued since publicationDate Text affected30 November 2012 Implementation of CEN Correction Notice 18September
5、 2012: See EN Foreword for list ofmodificationsTECHNICAL REPORT RAPPORT TECHNIQUE TECHNISCHER BERICHT CEN/TR 16324 August 2012 ICS 67.060 English Version Technical report of the interlaboratory study for the determination of Besatz in common wheat, rye and durum wheatRapport technique relatif lessai
6、 interlaboratoire portant sur la dtermination du pourcentage dimpurets dans le bl tendre, le seigle et le bl dur Technischer Bericht des Ringversuchs zur Bestimmung von Besatz in Weizen, Roggen und Hartweizen This Technical Report was approved by CEN on 7 February 2012. It has been drawn up by the T
7、echnical Committee CEN/TC 338. CEN members are the national standards bodies of Austria, Belgium, Bulgaria, Croatia, Cyprus, Czech Republic, Denmark, Estonia, Finland, Former Yugoslav Republic of Macedonia, France, Germany, Greece, Hungary, Iceland, Ireland, Italy, Latvia, Lithuania, Luxembourg, Mal
8、ta, Netherlands, Norway, Poland, Portugal, Romania, Slovakia, Slovenia, Spain, Sweden, Switzerland, Turkey and United Kingdom. EUROPEAN COMMITTEE FOR STANDARDIZATION COMIT EUROPEN DE NORMALISATION EUROPISCHES KOMITEE FR NORMUNG Management Centre: Avenue Marnix 17, B-1000 Brussels 2012 CEN All rights
9、 of exploitation in any form and by any means reserved worldwide for CEN national Members. Ref. No. CEN/TR 16324:2012: ECEN/TR 16324:2012 (E) 2 Contents Page Foreword 3Introduction .41 Scope 52 Design of the study53 Statistical evaluation .64 Results of the determination of “broken grains“ .75 Resul
10、ts of the determination of “shrivelled grains“ 106 Results of the determination of “other cereals“ 137 Results of the determination of “grains damaged by pests“ . 168 Results of the determination of “grains in which the germ is discoloured, mottled grains“ . 199 Results of the determination of “spro
11、uted grains“ . 2210 Results of the determination of “extraneous seeds“ 2511 Results of the determination of “unsound grains“ . 2812 Results of the determination of “extraneous matter“ . 3113 Results of the determination of “husks“ 3414 Results of the determination of “ergot“ . 3715 Results of the de
12、termination of “grain impurities“ . 4016 Results of the determination of “miscellaneous impurities“ . 4317 Results of the determination of “total Besatz“ 4618 Precision data of sample set A 4919 Results of sample set A and B 53Annex A (informative) Participants . 58Bibliography . 59PD CEN/TR 16324:2
13、012CEN/TR 16324:2012 (E)CEN/TR 16324:2012 (E) 3 Foreword This document (CEN/TR 16324:2012) has been prepared by Technical Committee CEN/TC 338 “Cereal and cereal products”, the secretariat of which is held by AFNOR. Attention is drawn to the possibility that some of the elements of this document may
14、 be the subject of patent rights. CEN shall not be held responsible for identifying any or all such patent rights. PD CEN/TR 16324:2012CEN/TR 16324:2012 (E)Modifications from CEN Correction Notice 18 September 20121) In Clause 2, in the paragraph relating to sample set B, second line, delete “resp.”
15、.2) Regarding Figures 2, 5, 8, 11, 13, 16, 19, 22, 25, 28, 31, 34, 37 and 40, replace the explanation of Key X with “laboratory”.3) Replace all the occurrence of “% W/W” or “W/W %” with “w/w %” in the keys of the figures where this appears (i.e. Figure 2, 4, 5, 7, 8, 10, 11, 13, 15, 16, 18, 19, 21,
16、22, 24, 26, 27, 28, 30, 31, 33, 34, 36, 37, 39 and 40).4) In the keys of Figures 3, 6, 9, 12, 14, 17, 20, 23, 26, 29, 32, 35, 38 and 41, write for the explanation of Y1 the Mandels statistic k in italics and in lower case.5) In the heading of Figure 13, replace “six test samples” with “four test sam
17、ples”.6) In the heading of Table 22, correct “recision” with “Precision”.7) Replace Figure 42 with the new one.CEN/TR 16324:2012 (E) 4 Introduction The term “Besatz” applies to all components of a grain sample that differ from the normal basic cereal. The principle of the determination of Besatz is
18、to separate all the groups of Besatz from the normal basic cereal grains of unimpaired quality by sieving and manual selection out of a subsample and to quantify them. The amount of Besatz and its constituent groups is important for health, cleaning, milling and further processing aspects. For these
19、 reasons Besatz is a part of contracts in grain trade and also of the grain intervention system of the EU. At present a European Standard for the determination of Besatz (EN 15587) has been developed. An international interlaboratory test for the determination of Besatz, according to EN 15587 and in
20、volving 15 laboratories in 11 countries, was carried out with two common wheat, two rye and two durum wheat samples. Statistical analysis was possible for all Besatz fractions with the exception of grains overheated during drying, bunted grains, and impurities of animal origin. Very high interlabora
21、tory variation was found for the fractions mottled grains (including grains in which the germ is discoloured), unsound grains, and grains damaged by pests. The lowest interlaboratory variation was found for the fractions extraneous matter, extraneous seeds, and ergot. The coefficient of variability
22、in most cases depended clearly on the amount of the respective Besatz group, whereby the amount of the respective Besatz group was in a normal range. As a result the coefficient of intralaboratory variability for the total Besatz was not higher than 10 %. The coefficient of interlaboratory variabili
23、ty for the total Besatz was not higher than 20 %, provided the amount of mottled grains or unsound grains or insect-damaged grain is not higher than 1 % (w/w). PD CEN/TR 16324:2012CEN/TR 16324:2012 (E)CEN/TR 16324:2012 (E) 5 1 Scope The term “Besatz“ applies to all components of a grain sample that
24、differ from the normal basic cereal. It includes the following groups: broken grains, shrivelled grains, other cereals, grains damaged by pests, grains with discoloured germ, grains overheated during drying, sprouted grains, extraneous seeds, unsound grains, ergot, bunted grains, extraneous matter,
25、husks and impurities of animal origin. The amount of Besatz and its constituent groups is important for health, cleaning, milling and further processing aspects. For these reasons Besatz is a component of contracts in grain trade and also of the grain intervention system of EU. The principle of the
26、determination of Besatz is to separate all the groups of Besatz from the normal basic cereal grains of unimpaired quality by sieving and manual selection out of a subsample and to quantify them. There are various problems in the determination of Besatz: Firstly, the identification of the different g
27、roups of Besatz depends strongly on the experience and the knowledge of the investigator. Also experienced investigators can differ in their characterization of grains. Finally, one is faced with the fact that grain, even after mixing, is rarely homogenous. In other words, if a sample was divided by
28、 a sample divider into a number of portions, the amount of a specific group of Besatz in each portion could be different, even if absolutely no human or machine error occurred in each determination. These problems will result in variation of the results of the determination. An international interla
29、boratory trial for the determination of Besatz in common wheat, durum wheat and rye was accomplished with 15 laboratories in order to get information on the intra- and interlaboratory variability of the determination of Besatz. The Technical Report here describes the preparation and evaluation of th
30、e results of this interlaboratory test. 2 Design of the study Each laboratory received two kinds of sample sets. The sample set A consisted of six samples which had a mass of not less than 50g. The participants received two samples of common wheat, rye and durum wheat. The sample set A was prepared
31、from basic cereal of unimpaired quality. Certain amounts of the 14 different Besatz groups 1)were added to each sample of the set A, in the way that the two samples of the same cereal had the same composition. The composition of the spiked samples was chosen to ensure that i) laboratories obtaining
32、outlying results would be identified and ii) the influence of sample division on the precision of the method would be recognizable. The sample set B consisted of six samples which had a mass of not less than 1kg. The participants received two samples of common wheat, rye and durum wheat. The sample
33、set B was prepared from clean basic cereal, which contained small amounts of Besatz 2). One bulk sample of slightly more than 40kg of each cereal (wheat, rye, and durum) was used. After careful mixing to make the bulk sample as homogeneous as possible it was divided using a sample divider to obtain
34、40 samples. Twenty of them were directly sealed in moisture proof plastic bags. The other 20 samples were spiked with certain Besatz groups, individually mixed and the sealed in moisture proof plastic bags. In this way we achieved different concentration levels of the various Besatz groups. The conc
35、entration levels were selected to represent the range of Besatz commonly found in grain samples. The results obtained for the sample set B were used for the calculation of the precision data given in Annex D of EN 15587:2008. The participants were instructed to analyse the samples of set A only once
36、 and the samples of set B in duplicate. They had to divide the samples of set B to obtain the required size, before testing. 1) The added material of the various Besatz groups comes from numerous Besatz-investigations of our own laboratory. The material was reviewed independently by three persons wi
37、th regard to the unique assignment to a Besatz group. 2) The use of basic cereal of unimpaired quality was impossible, because the cleaning of 120 kg of grain material by manual selection was not realisable. PD CEN/TR 16324:2012CEN/TR 16324:2012 (E)CEN/TR 16324:2012 (E) 6 We sent the sample to 20 la
38、boratories and after six weeks we received results of 15 laboratories for each sample. The laboratories were instructed to report the weights of the analysed samples as well as of the separated Besatz groups. In addition, participants should calculate the results in terms of percentage weight. The o
39、btained results of sample set B showed that the method of calculation was not described clearly enough in the draft standard. Many laboratories have not taken into account that extraneous matter was removed (by sieving) before testing the two sub-samples. Only five laboratories performed the calcula
40、tions in the correct manner. The results of the other laboratories were recalculated before beginning the statistical evaluation. 3 Statistical evaluation The statistical evaluation was performed according to ISO 5725-2. This requires the examination of data with the Cochran and Mandels k test for n
41、on homogeneity of variances (5 % and 1 % level) and the examination of the data with the Grubbs and Mandels h test for deviations of laboratory mean values (5 % and 1 % level). So, the analysis of the results for consistency and outliers was done by: numerical outlier tests (Cochrans test, Grubbs te
42、st); and graphical consistency technique (Mandels statistics, k and h). The results obtained for the sample set B are presented in Clause 4 to 17. These results were used for the calculation of the precision data given in EN 15587. The single results of the duplicate determination and the calculated
43、 precision data (after outlier elimination) are always presented in a table and a figure. The results of the Mandels statistic are presented in a separate figure. Abnormal values can be identified by the comparison with the critical values which are drawn as lines in the figure. A value which was st
44、raggling (abnormal) according to one of these four statistical tests was removed only if the following two conditions were met: a) the precision data without this abnormal values differed significantly from the precision data including it, so that its elimination gave significantly better values for
45、 precision; b) the elimination of this abnormal value does not have the consequence that the ratio r / R (repeatability/Reproducibility) reaches the value 1. Only if both conditions were met, an abnormal value was eliminated. Therefore, there may be cases in which an abnormal value was not removed,
46、because the precision data were not significantly improved or the ratio r/R reached the value 1. The table which contains the collation of the test results and the calculated precision data shows also the eliminated outliers, which are underlined. The results obtained for the sample set A were stati
47、stically evaluated in the same manner as described above, but the collation of test results and precision data are not presented as detailed as above. The precision data of the sample set A are presented in Clause 18 and shown together with those of the sample set B in Clause 19. No statistical eval
48、uation could be done for three fractions: grains overheated during drying, smutty grains and impurities of animal origin. In the first two cases we did not have enough material to spike the grain samples. In the case of impurities of animal origin there was a problem with the insufficient descriptio
49、n in EN 15587. It was not clear how to handle the insects which fall through the 1 mm sieve. Some laboratories had counted these insects, others had not. PD CEN/TR 16324:2012CEN/TR 16324:2012 (E)CEN/TR 16324:2012 (E) 7 4 Results of the determination of “broken grains“ Table 1 Collation of test results, outliers and precision data on the determination of “broken grains“Sample:AABBCCDDEE FF A B C D E FLab.No1 1.676 1.517 2.935 2.498 1.202 1.269 5.512 5.639 0.120 0.630 2.158 1.690 1.60 2.72 1.24 5.58 0.38 1.922 1.9
