RESUMO
To explain a difference of 0.5 % between the absorbed-dose standards of the National Institute of Standards and Technology (NIST) and the National Research Council of Canada (NRCC), Seuntjens et al. suggest the fault lies with the NIST water calorimeter being operated at 22 °C and the method with which the measurements were made. Their calculations show that this difference is due to overprediction of temperature rises of six consecutive (60)Co radiation runs at NIST. However, the consecutive runs they refer to were merely preliminary measurements to determine the procedure for the NIST beam calibration. The beam calibration was determined from only two consecutive runs followed by water circulation to re-establish temperature equilibrium. This procedure was used for measurements on 77 days, with 32 runs per day. Convection external to the glass cylindrical detector assembly performed a beneficial role. It aided (along with conduction) in increasing the rate of excess heat transported away from the thin cylindrical wall. This decreased the rate of heat conducted toward the axially located thermistors. The other sources of excess heat are the: (1) non-water materials in the temperature probe, and (2) exothermic effect of the once-distilled water external to the cylinder. Finite-element calculations were made to determine the separate and combined effects of the excess heat sources for the afterdrift. From this analysis, extrapolation of the measured afterdrifts of two consecutive runs to mid radiation leads to an estimated over-prediction of no more than about 0.1 %. Experimental measurements contradict the calculated results of Seuntjens et al. that convective motion (a plume) originates from the thermistors operated with an electrical power dissipation as low as 0.6 µW, well below the measured threshold of 50 µW. The method used for detecting a plume was sensitive enough to measure a convective plume (if it had started) down to about the 10 µW power level. Measurements also contradict the NRCC calculations in predicting the behavior of the NIST afterdrifts.
