BS ISO 12744 pdf download – Copper, lead, zinc andnickel concentrates——Experimental methods for checking the precision of sampling

BS ISO 12744 pdf download – Copper, lead, zinc andnickel concentrates——Experimental methods for checking the precision of sampling.
5 Method of experiment
5.1 Interleaved samples
Each alternate pnmary increment shall be diverted so mat pairs of interleaved samples A and B are fomiod. The number of divided increments per primary increment should be the same as for routine sampling. An example of a sampling plan for producing pairs of interleaved samples A and B Is shown in Figure 1.
5.2 Sample processing and analysis
The pairs of inlerleaved samples A and B taken In accordance with 5.1 shall be divided separately arid subqected to method 1. method 2 or method 3 sample processing and analysis as described in 5.21, 5.2.2 or
5.2.3.
5.2.1 Sample processing method 1 (see Figure 2)
The two interleaved samples A and B shall be divided separately to prepare four laboratory samples, A,, A2. and B,. 82. These laboratory samples shall each be analysed in duplicate, and the duplicates designated x,,, and x,,2tor sample A,, Zial and x,tor sample A2, x, and x212tor sample B,, and x and xfor sample B2. The eight determinations shall be run in random order, by the same analyst on the same day using the same analytical equipment. An example is given iii Annex A.
NOTE By using meThod 1, the estimated prec*ions of primary sampling. sample processing and analysis can be obtained separately.
5.2.2 Sample processIng method 2 (see Figure 3)
Sample A shall be divided to prepare two laboratory samples. A, and A2. From sample B. only one laboratory sample shall be prepared The laboratory samples shall each be analysed in duplicate, and the duplicates desi9nated x,,, and x,,2 for sample A1, X121 and X,72 for sample A2, and x, and x2 for sample B. The six determinations shall be run in random order, by the same analyst on the same day using the same analytical equipment.
NOTE By using method 2, the estimated precisions of primary sampling, sample processing and analysis can be obtained separately However, the estimated values will be less precise than those obtained using method 1.
5.2.3 Sample processing method 3 (see Figure 4)
From each of the two interleaved samples A and B, one laboratory sample shall be prepared The two laboratory samples A and B shall be analysed in duplicate, and the measurements obtained shall be designated x,, and x,2 lot sample A, and x2, and x for sample B. The four determinahons shall be run in random order, by the same analyst on the same day using the same analytical equipment
NOTE By using method 3. any the estimated precision of analysis and the combined precision of primary sampling and sample processing are obtained.
f) Conduct F-tests on the variances s, 2 and s to determine whether their differences are statistically significant using the following procedure:
1) calculate the variance ratios s2&f and 5!s
2) compare these ratios with the 95 $ confidence F-ratios given ii Table 1 for the number of degrees of freedom applicable to each variance estimate:
3) if the calculated variance ratio exceeds the F-ratio, partition the two variances into variance components. as their difference is statistically significant.
If the calculated variance ratio does not exceed the F-ratio, the variances andor cannot be meaningfully partitioned into variance components, and mora data need to be collected.