Overview
ISO 13292:2006 specifies experimental methods to check for bias in the sampling of copper, lead, zinc and nickel concentrates. It is intended for use when sampling is carried out in accordance with ISO 12743 and covers comparisons between a designated reference method (Method A) and a test method (Method B). Typical reference sampling is stopped‑belt (reference increments) while Method B may be mechanical sampling from a moving stream. The standard describes paired‑data protocols, statistical analysis, and practical recommendations to detect meaningful systematic differences (bias) between sampling regimes or during sample processing.
Key topics and requirements
- Paired data approach: Comparisons are based on paired sample sets; the standard requires at least 20 paired data sets as a minimum.
- Reference vs test methods: Method A is assumed unbiased (e.g., stopped‑belt or alternative transfer belt reference); Method B is the system under test (mechanical sampler, different location or processing route).
- Sample processing: Procedures for constituting sample A and sample B, similar processing and analyses (ISO 12743 referenced). Analyses are recommended under repeatability conditions to reduce analytical variance.
- Statistical framework:
- Differences di = xBi − xAi are computed, with mean d and standard deviation sd.
- Bias Detection Limit (BDL) is calculated to determine the minimum detectable bias at the 95% confidence level.
- Conventional risks: Type I (α = 0.05) and Type II (β = 0.10); a two‑sided Student’s t‑test is applied.
- Procedures to compute required sample size n when initial BDL > targeted bias δ (tables and formulae provided, including numerical examples in Annex A).
- Practical recommendations: Take increments from closely adjacent portions when testing equipment performance; take independent increments when testing sampling regimes for periodic effects. Repeat checks periodically and when equipment changes.
Applications and users
ISO 13292 is used to:
- Validate mechanical sampling systems and manual sampling procedures for concentrates.
- Compare sampling at different locations (e.g., loading vs discharge).
- Check sample processing bias and validate analytical exchange samples.
Typical users:
- Mining and concentrate producers, metallurgical QA/QC teams
- Sampling engineers, cargo surveyors and port operators
- Analytical laboratories and auditors ensuring representative sampling
Related standards
- ISO 12743 - Sampling procedures for copper, lead, zinc and nickel concentrates (normative reference used by ISO 13292)
- Annex A of ISO 13292 contains numerical examples illustrating calculations for BDL and required sample sizes.
Keywords: ISO 13292:2006, sampling bias, copper concentrates, lead zinc nickel concentrates, stopped‑belt sampling, bias detection limit, paired data, t‑test, ISO 12743.