Benchmark Analysis of TRIGA Mark II Reactivity Experiment Using a Continuous Energy Monte Carlo Code MCNP

Abstract

The benchmark analysis of reactivity experiments in the TRIGA-II core at the Musashi Institute of Technology Research Reactor (Musashi reactor, 100 kW) was performed by a three-dimensional continuous-energy Monte Carlo code MCNP4A. The reactivity worth and integral reactivity curves of the control rods as well as the reactivity worth distributions of fuel and graphite elements were used in the validation process of the physical model and neutron cross section data from the ENDF /B-V evaluation. The calculated values of integral reactivity curves of the control rods were in agreement with the experimental data obtained by the period method. The integral worth measured by the rod drop method was also consistent with the calculation. The calculated values of the fuel and the graphite element worth distributions were consistent with the measured ones within the statistical error estimates. These results showed that the exact core configuration including the control rod positions to reproduce the fission source distribution in the experiment must be introduced into the calculation core for obtaining the precise solution. It can be concluded that our simulation model of the TRIGA-II core is precise enough to reproduce the control rod worth, fuel and graphite elements reactivity worth distributions.

A Review of Candidates for a Validation Data Set for High-Assay Low-Enrichment Uranium Fuels

“…A challenge with some of the reactor conversions from HEU to HALEU is establishing a converted core with an equivalent performance to their original HEU core, as expected from modeling and simulations. For example, TRIGA (Training, Research, Isotopes, General Atomics) LWR research reactors with UZrH HALEU fuel would typically have a positive computational bias in models, effectively overpredicting total core excess reactivity [25,26], while simulated worth measurements suitably represented actual values within evaluated uncertainties [27]. Comprehensive analyses indicated that the likely cause of the observed inconsistency is nuclear data deficiencies, such as for Zr and ZrH thermal scattering, which has a more pronounced impact in smaller critical systems [28] than in larger configurations.…”

Monte Carlo modelling of TRIGA research reactor

“…The calculation of the control rods' reactivity worth simulated explicitly the experiment, which was carried out by the positive period method (Matsumoto et al, 2000). Using this method, the worth of one control rod was measured in the presence of other rods used for compensating the excess reactivity.…”