The purpose of this practice is to provide guiding principles for the use of reference standard cottons for the standardization of instruments and techniques to obtain reproducible test results (within statistical limits) when the tests are performed on samples of cotton fibers by the same or different operators at the same or different times, both within the same laboratory and in different laboratories.
Results obtained on cotton fiber tests on the same sample may vary widely due to differences in instruments and operator techniques. Some variation in test result levels can be controlled by the physical adjustment of instruments in accordance with manufacturers' instruction. Instrument adjustments are made only to cause instrument values to coincide with specified calibration points established by the manufacturer. Other variations in cotton fiber test results are caused by differences in technique used in specimen preparation, errors in reading test values from scales, variation in the reference standard, and sampling errors in specimen selection. It is necessary that the mean value of at least four sets of determinations per operator-instrument be used in order to form a sound basis for any adjustment of technique or for the use of correction factors.
In comparing results of different operators and different laboratories, for example, between the purchaser and the seller, it is essential that all results be obtained on replicate specimens. Values obtained on individual samples, as well as on a reference standard are never absolute, but have a normal variation about their mean due to heterogeneity of the sample. Therefore confidence limits of a test value are dependent upon the within laboratory variance for the test and the sampling error. The level of the results in different laboratories will be comparable only to the extent of the use of the same reference standards in both laboratories.
Any one set of determinations on a reference standard may be within the statistical limits of this reference standard cotton approximately 67 % of the time. The mean of a minimum of four sets of tests, as set out in the specific test procedure, is required to establish a reliable mean value for the determination of the test level. Caution should be exercised in the adjustment of instrument or technique on the basis of preliminary results on the working cotton standard or calibration cotton standard because of the variation within these reference standards.
When the mean value of the initial test results on a reference standard cotton falls within one unit standard deviation of the established values, testing may be started on samples of cotton fiber to be tested. If the mean value does not fall within two standard deviation units, recheck the instrument adjustments and test an additional set of test specimens from the reference standard cotton. The mean value of these two sets of determinations, where instrument checks have shown the adjustments (instruments) to be in order, may then be used as a basis for modification of technique, or if correction factors are to be used, the basis for the initiation of testing of unknown samples. Correction factors should be based on determinations made before, during, and at the completion of routine testing.
1.1 This practice covers the use of reference standard cottons for the standardization of instruments and techniques used to test cotton fibers in various laboratories.
1.2 Standardization may be achieved by application of a correction factor based on the reference standards, or by modification of the technique in use.
Note 18212;When reference standards are used to develop correction factors, or to adjust an operator's technique, no instrument calibration processes are involved. The term “Calibration
Copyright ©2024 All Rights Reserved