OBJECTIVE: In order to simplify the use of the corresponding tables for the quality factors of International Atomic Energy Agency TRS-398, the cross-calibration of parallel chambers is performed by comparison with a cylindrical chamber calibrated at a quality electron beam Qcross; an intermediate and random energy Qint is introduced in which the parallel chamber will be calibrated. The objective of this study is to evaluate the selection of this intermediate energy. MATERIALS AND METHODS: A parallel chamber from Scanditronix, model NACP-02 and a PTW Markus chamber were calibrated in two linear accelerators from Varian, Clinac 2100C and Clinac 23EX, in the energies of 16 and 20 MeV respectively. A Nuclear Enterprises chamber, model 2571 previously calibrated in terms of Dw at "Instituto de Pesquisas Energéticas e Nucleares", São Paulo, Brazil, for a 60Co beam was used as reference. RESULTS: The calibration factors N D,w thus obtained show a variation of 0.07% that can be considered negligible. The absorbed dose values in water determined at Clinac 2100C for a Qint of 16 MeV present a variation of 0.04% for the smallest energy (4 MeV) and of 2.6% for the largest energy (16 MeV), in relation to the TRS-381 protocol. For a Qint of 20 MeV, the observed variations are of 1.3% and 3.5%, respectively. On the other hand, the doses obtained in Clinac23EX for Qint of 16 MeV and of 20 MeV are of 1.9% and 2.0% and of 2.8% and 3.0% respectively. CONCLUSION: The choice of Qint is not so critical, particularly when no electron beams which have a R50value exactly equal to 7.5 g.cm-2 are available.

Keywords: Calibration of parallel chambers, Electron beams dosim, Dosimetry protocols