14 SPECTROSCOPYEUROPE SAMPLING COLUMN www.spectroscopyeurope.com VOL. 29 NO. 5 (2017) Experimental validation of a primary sampling system for iron ore pellets Karin Engström a and Kim H. Esbensen b a PhD Student & Method development engineer in sampling and analytical test methods b KHE Consulting (www.kheconsult.com) & Guest professor (Denmark, Norway, Puerto Rico) This feature forms an intermezzo in the current segment of process sampling columns. We have asked Karin Engström, Luossavaara Kiirunavaara AB (LKAB), Kiruna, Sweden to outline how industrial validation of a process sampling system takes place following ISO standards guidelines. These prescribe a rigorous procedure for comparison of a process sampling system with a “stopped belt” + manual sample extraction reference system as a means for checking for a sampling bias, as the refer- ence sampling system is considered to be fully TOS-compliant, i.e. representative. This column forms a comprehensive back- drop for the on-line alternative of variographic characterisation of the same iron ore pellet stream (which will follow in the next installment). Introduction: status of current ISO standards Primary sampling of iron ore is well established and standardised through the International Organization for Standardization (ISO). In comparison to standardisation of other mineral commodities and particulate materi- als (e.g. food/feed, pharmaceuticals), iron ore sampling standards are in close compliance with the Theory of Sampling (TOS). 1–3 Iron ore mining and processing operations apply sampling and grade control in all parts of the production value chain, from diamond drill and blast holes all the way to process sampling of slurries, pellet feed, finished pellets and at ship loading. Sampling of iron ore is stand- ardised through the international stand- ard ISO 3082: “Iron ores—Sampling and sample preparation procedures”. 4 The iron ore industry has improved its conformance to ISO 3082 over the last 10–20 years, especially regarding commercial purposes. However, there are still several areas where deviations from the standard and issues with sample representativity are common. 3 For newly constructed sampling systems or in-use systems that have been modified, ISO 3082 demands verification of the full sampling system in accordance to ISO 3086: “Iron ores— Experimental methods for checking the bias of sampling”. 5 We here report on an experimental verification and valida- tion experiment, as a base-line refer- ence to be compared with an on-line variographic sampling system QC in the following column. Fundamental Sampling Principle and basic requirements for iron ore sampling systems The Fundamental Sampling Principle (FSP) for representative sampling states that all parts of the lot must have equal probability of being selected for the sample. 6–8 This principle is equally important for primary sampling extrac- tion as for all subsequent sampling stages, i.e. during mass reduction/ sample division. ISO 3082 describes the best place for primary sample extraction to be at a transfer point between conveyor belts, where a full cross-section of the stream can be intercepted, and extracted at regular intervals. Sampling from stationary lots such as ships or stockpiles is not permit- ted by ISO 3082, as it is impossible to drive a sampling device through the lot depth and extract a full column of ore. ISO 3082 therefore recommends only to extract samples as the ore is being transported to or from a ship, stockpile, bunker or silo. The extraction of primary increments shall comply with the following regula- tions to ensure that no bias is generated (well-tested TOS principles): ■ a complete cross-section of the ore stream shall be taken when sampling from a moving stream; ■ the aperture of the sample cutter shall be at least three times the nominal top size of the ore, or 30 mm for the primary sampling and 10 mm for subsequent stages, whichever is the greater; ■ the speed of the sample cutter shall not exceed 0.6 m s –1 , unless the cutter aperture is correspondingly increased; ■ the sample cutter shall travel through the ore stream at uniform speed, both the leading and trailing edges of the cutter clearing the ore stream at the end of its traverse; ■ the lips on the sample cutter shall be parallel for straight-path samplers and radial for rotary cutters; these conditions shall be maintained as the cutter lips wear; Karin Engström