RESUMO
Two methods for determining the collection efficiency of a 0.6 cm3 thimble ionisation chamber exposed to the swept electron beam of a linear accelerator Therac 20 Saturne (CGR MeV) have been compared. In one method the chamber signal has been compared to that of simultaneously exposed thermoluminescent LiF dosemeters (TLD), in the other the "two-voltage" method of Boag, adapted for swept beams, has been used. By variation of the electron energy between 20 and 13 MeV, of the focus-skin-distance (FSD) between 200 and 100 cm and of the monitor rate between 400 monitor units (m.u.) and 100 m.u. per minute, different values could be produced for the peak charge density M. The collection efficiency of the chamber, operating at a standard voltage of 250 V, decreases from 0.99 to 0.84 for a charge density increasing from 0.3 X 10(-4) C/m3 to 7.5 X 10(-4) C/m3, respectively. The maximum deviation observed between the TLD and the "two-voltage" method adopted for similar M is never more than 2% and mostly smaller than 1%. It can be concluded that, under the present experimental conditions, the calculated ionisation chamber collection efficiency is confirmed by the experimental method using TL dosimetry.