The quality of 3D printing depends on the properties of consumables, in particular, on the chemical composition of the powders, the size and shape of their particles. To eliminate printing defects, the working mixture of primary and secondary powders based on polyamide-12 should contain no more than 30% of the secondary powder. We present the results of studying the fractional composition of powders by the methods of statistical analysis. Digital images of polymer samples including morphological parameters of particle images were analyzed. To assess the fractional composition of the particles of primary and secondary powders, a statistical method of dimensional ranking and a differential method for determining the boundaries of fractions were used. It is shown that the particle area is the parameter most sensitive to changes in the structure of powders. The results of statistical ranking of effective particle radii are obtained proceeding from the histograms of the particle area distribution. The boundaries of the conditioned fraction are determined by the magnitude of the effective radii. A comparison of the fractional composition of primary and secondary powders, as well as calculation of the percentage of fine, working and large fractions were carried out taking into account the assessment of the fraction boundaries. It is revealed that the content of fractions of powder particles with conditioned dimensions should be about 64% of the total volume of the powder. Reduction of the amount of primary powder can lead to defects in 3D printing. The results obtained can be used to increase the degree of recovery of polyamide-12 based powders during 3D printing.