The essential prerequisite of radiation dosimetry is to provide quality assurance and documentation that the irradiation procedure has been carried out according to the specification requirement of correct calibration of the chosen dosimetry system. At the Radiation Plant of the Vinča Institute of Nuclear Sciences we compared two recommended protocols of irradiation procedures in the calibration of dosimetry systems in radiation processing: (1) by irradiation of routine dosimeters (ethanol-chlorobenzene - ECB) at the calibration laboratory and (2), by in-plant calibration with alanine transfer - dosimeters. The critical point for in-plant calibration is irradiation geometry, so we carefully positioned the phantom carrying both dosimeters in order to minimize dose gradients across the sample. The analysis of results obtained showed that the difference among determined absorbed doses for the construction of calibration curves between these two methods, (alanine vs. ECB), is less than 1%. The difference in combined standard uncertainty for each calibration procedure is 0.1%. These results demonstrate that our in-plant calibration is as good as calibration by irradiation at the calibration laboratory and validates our placement of the irradiation phantom during irradiation.