JSI TRIGA fuel rod reactivity worth experiments for validation of Serpent-2 and RAPID fuel burnup calculations

Abstract

Reactivity worth of fuel rods at the JSI TRIGA research reactor was measured. Differently burned fuel rods were chosen to validate fuel burnup calculations. Two methods of measuring reactivity worth of fuel rods are used, traditional method is compared to newly introduced method using fuel rods swapping. Connection between both methods is described theoretically and the theory is validated experimentally. Fuel rod worth calculated using the newly introduced fuel rod swap method was within 1σ of worth measured using the traditional method. In addition to the recently performed experiments, weekly measurements of reactor core reactivity throughout the operational history are used for validation. The measured data were used to validate the fuel burnup and core criticality calculations. Fuel burnup calculations are performed using three different computer codes: the deterministic TRIGLAV, the Monte Carlo Serpent-2, and the hybrid RAPID. Great agreement was observed for Serpent-2 and RAPID by simulating fuel rod worth and its burnup, indicating that the fuel burnup and criticality calculations are accurate and that reactivity changes due to small burnup differences on the order of 10 pcm can be accurately simulated. In addition it was shown using ex-core detectors and large fission chamber that detector response changes due to fuel swapping are evident for fuel rod burnup differences of 20 MWd/kg. Fuel burnup calculations were further validated on excess reactivity measurements for three mixed TRIGA cores. The calculated burnup reactivity coefficient ΔρBU using Serpent-2 and RAPID was within 1σ of the measurements, showing both codes are capable of calculating burnup for different TRIGA fuel types.