《ISO-6667-1985.pdf》由会员分享，可在线阅读，更多相关《ISO-6667-1985.pdf（16页珍藏版）》请在三一文库上搜索。
1、International Standard INTERNATIONAL ORGANIZATION FOR STANDARDIZATION.MEflYHAPOflHAR OPI-AHM3ALJVlR I70 CTAHL(APTM3AUWlORGANISATlON INTERNATIONALE DE NORMALISATION Green coffee - Determination of proportion of insect-damaged beans Cafh vert - D thus, the method described applies only to external dam
2、age caused by insects. 1 Scope and field of application This International Standard describes the types of damage caused by insects to green coffee beans and specifies a method for the determination of the proportion of insect-damaged beans in a lot of green coffee, together with the statistical use
3、 of the result obtained for technical, commercial and arbitration purposes. The method is applicable to green coffee as defined in IS0 3599. From the proportion of insect-damaged beans thus determined - estimation of the confidence limits of the true propor- tion of insect-damaged beans for a given
4、probability - estimation of the probability that the true proportion of insect-damaged beans will not exceed a given value. 5 Apparatus 5.1 Magnifying glass (recommended) A magnification of X 5 is suitable for inspection of the surface of beans. 5.2 Sample-holder (optional) The use of a sample holde
5、r having a given number of cavities, each capable of holding a whole coffee bean, facilitates counting and separation of beans. 6 Sampling See IS0 4072. 2 References IS0 3509, Coffee and its products - Vocabulary. IS0 4072, Green coffee in bags - Sampling. 3 Definition For the purpose of this Intern
6、ational Standard, the following definition applies. insect-damaged bean: A coffee bean damaged externally by insects. 4 Principle Visual examination of the external surface of green coffee beans to identity those which have been damaged by insects Determination of the observed proportion, as a perce
7、ntage, of insect-damaged beans and, if possible, identification of the species of insect responsible for the damage. 7 Damage caused by insects NOTE - Photographs are useful as a means of illustrating damage caused by insects and other types of damage visible on the external surface of the bean. Som
8、e photographic illustrations are given in annex A. 7. I Description A green coffee bean may show a great variety of damage caused by insects, from minor scars on the surface to a con- siderable number of holes and tunnels. It is usually impossible to distinguish minor damage from damage of mechanica
9、l origin and minor breakage. For the purpose of this International Standard, only major damage is considered, i.e. those beans with clearly visible holes which are characteristic of insect damage and which cannot be mistaken for other damage. If a bean shows several signs of damage, it is considered
10、 as a single damaged bean. Copyright International Organization for Standardization Provided by IHS under license with ISO Licensee=NASA Technical Standards 1/9972545001 Not for Resale, 04/21/2007 09:16:03 MDTNo reproduction or networking permitted without license from IHS -,-,- IS0 66674985 (El 8.3
11、 Determination The two insect species which are most frequently responsible for damaged and infested beans are Hypothenemus (Stephanoderes) hampei (coffee bean borer) and Araecerus fascicula tus ( coffee bean weevil 1. Operating under good lighting conditions, and preferably using the magnifying gla
12、ss (5.1), examine the entire surface of each bean in the test portion for the different types of holes characterizing insect damage. The damage caused by these insects to the external surface of beans is very specific and usually also indicates internal damage. In order to distinguish between a stai
13、n or surface defect and a cavity caused by insects, check with a pin to ascertain whether there is really a hole; such a test is particularly useful along the groove of the bean. Count only those beans which show significant penetration as insect-damaged beans. 7.2 Identification of insects causing
14、the damage In addition to assessment of the damage, it is sometimes poss- ible, and useful, to identity the species of insect causing the damage. Damage caused by H. hampei and A. fasciculatus is different in appearance and the type of damage can thus be used to identify the insect responsible. Sepa
15、rate and count the beans which show damage such as that described in 7.21 and 7.22. 9 Expression of results 7.21 Damage caused by H. hampei The observed proportion of as a percentage, is equal to insect damaged beans, expressed The damage seen on the bean surface is exit holes, as well as internal c
16、avities. usually from entry and ; x loo The holes are small (approximately neatly cut and circular 0,3 to I,5 mm in diameter), where There area. is often a blue-green stain a round the damaged surface n is the number of insect-damaged beans; N is the number of beans examined. H. hampei storage. only
17、 infests coffee beans in the field and not in 10 Statistical interpretation of results 7.2.2 Damage caused by A. fascicufatus 10.1 Estimation damaged beans of true proportion of insect- The visible damage on the bean surface is usually a bored hole, larger than that of H. hampei (from 1,O to 3,0 mm
18、in diameter) and not as neatly cut. 10.1.1 General A. fasciculatus can feed on green coffee beans during storage, if conditions are favourable, producing cavities and other exter- nal damage. . The estimate of the true proportion of insect-damaged beans in the lot is obtained from the observed propo
19、rtion of insect- damaged beans calculated as described in clause 9. The reliability of the estimate (calculation of confidence limits), for a 90 % probability, is determined by means of figure 1, from the observed proportion of insect-damaged beans and the number of beans examined. 8 Procedure 81 ,
20、Preparation of the test sample 10.1.2 Use of figure 1 Carefully mix the laboratory sample (see homogeneity of laboratory samples). annex B for a test for Find on the abscissa the observed of insect-damaged beans. proportion , as a percen tage, From this point, draw a vertical line which bisects both
21、 curves corresponding to the number of beans examined (interpolate for test portions comprising a number of beans not shown in figure 1). 8.2 Test portion Take at random at least 100 whole green coffee beans from the test sample. During counting, replace every broken bean selected by a whole bean ta
22、ken at random from the test sample. The ordinates at these points of intersection (a for the lower curve and b for the upper curve) indicate the confidence limits corresponding to the observed proportion of insect-damaged beans. NOTE - Broken beans will cause errors because a damaged bean may be cou
23、nted twice or pieces of beans may be counted as whole beans in the total. For this reason, in order to avoid any confusion in the results, the broken beans should be eliminated after being counted and it can be assumed that the proportion of insect-damaged beans subsequently determined on whole bean
24、s represents the proportion of insect- damaged beans within the laboratory sample. It can be stated, with a 90 % probability, that the true propor- tion of insect-damaged beans within the lot is between a and b. i.e. Pr (a .- z Ll : 50- -_-_- _ -_-_ -w-pp-_-_I-I_- Number of beans examined Figure 4 -
25、 Probability that the observed proportion of insect-damaged beans is from a lot in which the true proportion does not exceed 10 % Copyright International Organization for Standardization Provided by IHS under license with ISO Licensee=NASA Technical Standards 1/9972545001 Not for Resale, 04/21/2007
26、09:16:03 MDTNo reproduction or networking permitted without license from IHS -,-,- 00 (Observed proportion of insect-damaged beans) .- _ _ _ _- _ - _ _- _ _ _ - - - - - -_ - _ - - - - - .-.- - _ _._ _ -.- _ ._ -_ -_ - - _ - .-.- - -_ -. - - -_- _ - -_ _- _ _ - - _ _ _ -_ - - - -_ - - -_ _- _- _ - _
27、-.- - -_ -_ - _I -.- - _ _ ._ .-.- _ _-.- - _ _- - .-. - - - - -_ _- -.- _ -.- - - _ _ _ _._ _._ _ _ _ _ _ _ _ - _- - - _- - _- _ _- - - - _- 100 LOO 300 400 500 600 700 800 900 1000 - Number of beans examined Figure 5 - Probability that the observed proportion of insect-damaged beans is frorn a lot
28、 in which the true proportion does not exceed l2,5 % Copyright International Organization for Standardization Provided by IHS under license with ISO Licensee=NASA Technical Standards 1/9972545001 Not for Resale, 04/21/2007 09:16:03 MDTNo reproduction or networking permitted without license from IHS
29、-,-,- I (Observed proportion of insect-damaged beans) 90 80 70 -4 2 cl 60 w .a z I) : 6 50 - .- - -I_. _- - _ ._ - _ _ -.- - - -. -_ -_ - - .- _-_ _-_- - - ._ -_- - - _ - - - _ - _- -_ -_- _ _- - _- _ _ -_ .- _- -.- _ - - - - - - - - _._ _ - - -.- - - - - _ _ _ - - -_ - _ - - - . 40 30 20 -_-_-_ - _
30、 _ - -_ -_ _ - - - .-_ - - _ - - - _._ - -_ _ _ - - -.- - _ _ -_ - _ 10 1 Number of beans examined Figure 6 - Probability that the observed proportion of insect-damaged beans is from a lot in which the true proportion does not exceed 15 o/o Copyright International Organization for Standardization Pr
31、ovided by IHS under license with ISO Licensee=NASA Technical Standards 1/9972545001 Not for Resale, 04/21/2007 09:16:03 MDTNo reproduction or networking permitted without license from IHS -,-,- IS0 66674985 0 Annex A Illustrations of damaged beans A.1 Beans damaged by H. hampei A.2 Beans damaged by
32、A. fascicu/at/s ,: : _: .,:. i. . . Copyright International Organization for Standardization Provided by IHS under license with ISO Licensee=NASA Technical Standards 1/9972545001 Not for Resale, 04/21/2007 09:16:03 MDTNo reproduction or networking permitted without license from IHS -,-,- IS0 6667-19
33、85 (El ns damaged by causes other than insects Copyright International Organization for Standardization Provided by IHS under license with ISO Licensee=NASA Technical Standards 1/9972545001 Not for Resale, 04/21/2007 09:16:03 MDTNo reproduction or networking permitted without license from IHS -,-,-
34、IS0 66674985 0 Annex B Test for homogeneity of the laboratory sample The following statistical test can be used to determine whether the hypothesis of laboratory sample homogeneity should be re- jected. The test should be applied whenever there is a “suspiciously” large difference between the observ
35、ed propor- tions of insect-damaged beans in the first and second test por- tions. Let 5, be the observed proportion, as a percentage, of insect- damaged beans in test portion 1, divided by 100; fi be the sum of the numbers of insect-damaged beans in test portions 1 and 2, divided by the total number
36、 of beans examined for test portions 1 and 2; *1 be the number of beans examined in test portion 1; n2 be the number of beans examined in test portion 2. Then Z= 5, - & J$ (1 -j) Wnl + l/*2) For a 90 % significance test, if/Z1 1,645, the labora sample should be considered to be inhomogeneous. tory F
37、or a 95 % significance test, if /ZI 1,96, the laboratory sample should be considered to be inhomogeneous. If the hypothesis of laboratory sample homogeneity is rejected, a new laboratory sample should be taken from the bulk sample. 12 Copyright International Organization for Standardization Provided
38、 by IHS under license with ISO Licensee=NASA Technical Standards 1/9972545001 Not for Resale, 04/21/2007 09:16:03 MDTNo reproduction or networking permitted without license from IHS -,-,- This page intentionally left blank Copyright International Organization for Standardization Provided by IHS unde
39、r license with ISO Licensee=NASA Technical Standards 1/9972545001 Not for Resale, 04/21/2007 09:16:03 MDTNo reproduction or networking permitted without license from IHS -,-,- This page intentionally left blank Copyright International Organization for Standardization Provided by IHS under license with ISO Licensee=NASA Technical Standards 1/9972545001 Not for Resale, 04/21/2007 09:16:03 MDTNo reproduction or networking permitted without license from IHS -,-,-