Reactivity measurement using pulsed neutron source methods on VR-1 and VR-2 reactors and their verification by SERPENT code

Abstract

A series of pulse experiments were carried out on the VR-1 reactor and the newly constructed subcritical VR-2 assembly at the Czech Technical University (CTU) in Prague. The objective was to compare the results obtained from the pulse methods with the verified Monte Carlo (Serpent) model for VR-1 reactor and to investigate the potential for the process of experimental verification of the Monte Carlo model for the new VR-2 reactor. The Simmons-King’s, Sjöstrand’s, Gozani’s, and Garelis–Russell’s methods were applied to data collected from detectors located at different positions in a pulse experiment. Additionally, the Monte Carlo method was employed for the determination of the kinetic parameters essential for the comparison and application of the pulse methods. A very simple Serpent simulation of the Sjöstrand’s method was introduced. Although the used pulse source in some of its parameters (frequency, duty factor) indicated a certain limit of applicability, the expected, relatively good agreement between the experiment and the Monte Carlo model was achieved in the investigated range of reactivities for the VR-1 reactor. Therefore, it can be concluded that pulse methods in this range provide a reasonably reliable means of determining reactivity. The discrepancies observed at the VR-2 subcritical assembly indicate the necessity for a rigorous review of its existing Monte Carlo model.