Real accuracy of several calibration methods for sluice gates working in the submerged orifice flow condition was determined considering water discharge from water levels and gate openings. Data were taken from three gates of the same laboratory canal covering a large operational range. Using accurate hydraulic data, most of the methods produce errors of up to ±10%. However, errors can rise up to ±40% with methods using typically recommended calibration values or constant discharge coefficients. The best results were obtained by a method based on dimensional analysis and the incomplete self-similarity theory proposed recently in the literature. Calibration with field data, in this case, produced errors not higher than ±3%. On the other side, when the gate discharge data are not available, the use of a contraction value of 0.611 within a good theoretical formulation gives very good results.