The objective of this research was to evaluate the qualitative and quantitative yield of tobacco (Nicotiana tabacum L.) comparing various formulations of nitrogen-potassium fertilizers applied as sidedress treatments. Six trials were conducted in distinct flue-cured tobacco production areas in Santa Cruz do Sul region, RS, Brazil. Normal base fertilization of 640 kg/ha of the formula 10-16-10 was applied. Later, sidedress fertilization was applied, which varied between two and three applications according to the need for (N) replacement in each area. The experimental design was a randomized complete block. Management and cultural treatment followed the recommendations of the companies to which the producers were commercially linked. Tobacco was harvested four times. Sidedress applications of experimental treatments (Trt. 1. 15-0-15 contains KNO₃ [potassium nitrate] / 15-0-15 contains KNO₃ / 15-0-15 contains KNO₃; Trt. 2. 15-3-15 contains NaNO₄ [sodium nitrate] and SOP [sulfate of potash] / 15-0-15 contains KNO₃ / 15 -0-15 contains KNO₃; Trt. 3. 15-3-15 contains SOP / 15-3-15 contains SOP / 15-3-15 contains SOP; Trt. 4. 15-3-15 contains SOP / 14-0-15 contains CaNO₃ [calcium nitrate] and KNO₃ / 14-0-15 contains CaNO₃ and KNO₃) of 320 kg/ha were applied twice in two years with a third application of 160 kg/ha in the third year to compensate for leaching. The qualitative and quantitative yield data were statistically analyzed using the SASM-Agri system for the analysis of variance and the Duncan test at 5 %. Cured leaf yield was similar for all treatments. However, higher quality of the leaf from treatments 1, 2 and 3 resulted in greater income for the cured leaf amounting to 7, 8 and 6 percent, respectively, when compared to the income from treatment 4. These trials suggest an advantage in leaf value and income to the grower when KNO₃ is used as the primary source of sidedress fertilizer.