A Monitoring System for the Assessment of Reactor Shield Performance

This paper describes the objectives of a shield survey, presents the results of intercomparisons and makes recommendations on the selection of a monitoring system. The performance of a reactor shield must be assessed efficiently to enable immediate repairs to be made to defective shields. Correlation of survey results with shield designer's predictions has proved difficult and the factors contributing to the disagreement have been examined. The shield designer uses flux to dose conversion factors which neglect the radiation equilibrium situation at the shield-air interface and the effect of attenuation and scattering external to the shield. The presence of the operator can significantly influence a measurement. The use of the ''maximum exposure dose'' (MED) concept is recommended for gamma photon predictions and this is compared with absorbed dose measurements in phantoms. Shield survey instruments must have a flat response over a wide range of energies. For example, many gamma dosimeters are designed for use at energies up to about 2 MeV whereas around most reactors a proportion of dose is due to 6 MeV gamma photons, with some contribution from higher energies. Tests were carried out to select suitable detectors using actual reactor operating conditions and simulated conditions at specific energies, notably 6 MeV. In practice discrepancies exceeding a factor of 3 were frequently found. A comparison was made between thermoluminescent dosimeters and film badges for the purpose of fixed position integrating dosimeters. The film dosimeter (AERE/RPS) was found to over-estimate by a factor of 2 for 6 MeV gamma radiation and in practical situations over-estimated by 20 to 80%. Thermoluminescent dosimetry is recommended for shield surveys provided that a build-up cap is used to achieve charged particle equilibrium. (author)