Benchmarking of Monte Carlo based shutdown dose rate calculations applied in fusion technology: From the past experience a future proposal for JET 2005 operation

The reliable assessment of dose rates is one of the key issues in operating nuclear machines. For fusion technology applications, two different computational methods have been recently developed to this end, the so-called R2S and the D1S approach. First experimental and computational benchmarks showed good results and promising prospective in application to ITER. The application to the biggest European fusion machine JET, showed some deficiencies, especially in the outer regions of the machine, where doses are lower, but predictability importance higher. This has been attributed to the very complex geometry and the poor accuracy of both the measurements and the model used in the calculation for the outer regions. A new benchmark experiment has been proposed for the next campaign of JET. Preliminary calculation results of the dose rates achievable in such experiments are reported. Those are compatible with the sensibility of the detectors that will be used even if the residual background gives a consistent contribution. In the outer part of the machine the critical issue will be the ability to describe in detail the components close to the detector in the MCNP model.