Ural State Mining University (Russia):

Kozin V. Z.,  Head of Chair, Doctor of Engineering, Professor, gmf.dek@ursmu.ru

Vodovozov K. A.,  Lecturer, gmf.opi@ursmu.ru

Additional sampling errors — a probable systematic error and an error of method, that may lead to both hidden excesses and hidden losses of valuable components at ore-dressing facilities, are considered. It is shown, that probable systematic error is caused during sampling and sample preparation, as well as in weights taking for analysis, because analyzed component mass fraction peaks are missed in different parts of material flow or sample, due to insufficient quantity of spot samples and limited mass of samples and weights, and also quality of samples mixing prior to samples' reducing. This error is characterized by a definitive dominant sign (minus for low-grade products and plus for highgrade), and may be compensated by appearance of more infrequent results with opposite sign. The cause for an error of method in performance of conventional analyses consists in rejection of results that are beyond acceptable variations limits, leading to exclusion of predominantly high results, thus, ultimately, lowering valuable components' fraction of total mass in low-grade products and increasing it in highgrade products. It is noted, that specialties of product sampling at ore-dressing facilities cause systematic deviation from true sampling results, which, accumulating, may significantly affect concentrating plant performance indicators. Ways of sampling results' approximation to true results are proposed.

1. Kozin V. Z. Oprobovaniye mineralnogo syrya (Sampling of mineral raw material). Yekaterinburg, Ural State Mining University, 2011, 316 p.
2. State standard R ISO 57251, 2, 3, 4, 5, 6 — 2002. Tochnost (pravilnost i pretsizionnost) metodov i rezultatov izmereniy (Accuracy (correctness and precision) of methods and measurements results).