Is the FORK detector a partial defect tester?

Full text: A Coordinated Technical Meeting on Spent Fuel Verification Methods held in Vienna in March 2003 stressed the need for partial defect testers for spent fuel in both wet and dry storage. One of the candidates for partial defect testing is the FORK detector. The performance of the FORK detector as a partial defect tester was investigated experimentally in the framework of a Joint Task for the IAEA by the Finnish, Swedish and Belgian Support Programmes. This was done by replacing fission chambers in the FORK detector by He-3 tubes and simulating spent fuel with fresh MOX-fuel. It was concluded that the FORK detector was not suitable as a partial defect tester, since fuel pin removal could be not be detected in an unambiguous way for all possible scenarios. However, later analysis of the results revealed that a possibility exist to interpret the measurement results unambiguously, provided that some additional measurements are performed. It was decided to perform simulation calculations for some of the configurations that were investigated during the Joint Task for validation of the correctness of the simulation methodology and to investigate the applicability of the proposed additional measurements for an unambiguous partial defect test. The calculations that were performed for the validation of the correctness of the simulation methodology first showed for some configurations high discrepancies between the calculations and measurements. The discrepancies were for some configurations as high as 10%. It was concluded that additional improvement of the calculational model should be performed in order to reduce the observed discrepancies. Additional calculations have been performed in order to reduce the discrepancies between the measurements and calculations. Acceptable results were obtained when the He-3 and ionisation tubes were positioned a few centimeters backwards, while the backends of the tubes were positioned at the same level instead of that the centres of the tubes were positioned at the same level. The observed discrepancies were at maximum 4%, while the majority of the discrepancies were less than 1%. The assumed positioning errors are considered as likely to happen, therefore they are accepted as cause of the previously observed large discrepancies. The simulation methodology is now considered as reliable and can be used for simulations of measurements on spent fuel assemblies. This paper describes the results of the simulations of spent fuel assembly measurements with the FORK detector and assesses the applicability of the proposed additional FORK measurements for spent fuel characterisation. (author)