The paper is aimed at the analysis of the ‘old wood’ effect in radiocarbon chronology of iron smelting com-plexes of South Siberia. The production sites are here set outside the settlements, and radiocarbon dating re-mains the only means of their chronological attribution. With the example of Kuyahtanar metallurgical site in the Mountain Altai, we show that the range of a series of radiocarbon dates obtained for the same iron-smelting fur-naces can span a thousand years. It has been established that such a chronological range can be explained by the use of wood from long-lived tree species for charcoal production and the resulting significant apparent age due to the ‘old wood’ effect. The essence of the effect consists in that as a tree grows, its inner annual rings die out and stop exchanging carbon with the environment, while the tree is still alive. Therefore, the pith of a long-lived tree is much older than its outer rings, and radiocarbon analysis determines not the age of the archaeologi-cal object, or time when the tree was cut, but the date of the death of that part of the tree which was sampled for the dating. Methodological recommendations have been drawn up capable of minimizing the impact of the ‘old wood’ effect in the dating of iron metallurgy sites. During the sampling, it is recommended to collect fragments of char-coal from the tree outer rings. In a case, where this is not possible, it is necessary to obtain a series of at least three dates for one site. In such a series, the latest of the three dates will be the closest to the age of the archaeological object. It is also necessary to determine the tree species used for burning the charcoal for smelting. After the tree species is identified, it becomes possible to infer its average and maximum life expectancy in the specific natural and climatic conditions, and, consequently, possible average and maximum impact of the ‘old wood’ effect. Since in the case of dating metallurgical sites the impact of the ‘old wood’ effect is virtually unavoidable, it is suggested that, in the future, the chronology of the complexes radiocarbon dated with a single sample of charcoal is revisited.