Reactimeter algorithms enable real-time evaluation of the reactivity of a near-critical reactor from the signal of a neutron detector. They only depend on a few parameters defined by the inverse point kinetic equations, and which can be calculated by most neutron transport codes. However, the neutron population measured locally by a detector may not always lead to a good estimate of the reactivity. For example, this would be the case for an in-core detector placed just next to a moving control rod. In such special cases – or when measurement precision is an issue – costly simulations are needed to provide correction factors that account for variations in the spatial distribution of the neutron flux. This paper aims to provide an auto-corrected reactimeter algorithm (ACRA) in the special cases where rapid perturbations are applied to a core close to criticality. Such perturbations were successfully performed in the MINERVE reactor using a dedicated control rod to alternatively insert and withdraw a small sample of material under study. ACRA has been validated against 3D Monte Carlo neutron transport calculations using the signals of four in-core detectors located just next to the moving sample.

Since dynamic reactivities measured on actual reactors are significantly impacted by delayed nuclear data and technological uncertainties, it is of major importance to validate at first the methodology of interpretation against state-of-the-art simulation. In this way, we set up a reference simulation of roddrop transients with a subcritical core from 3 to 16 $ in the CABRI reactor using the dynamiafforganizationc capabilities of the Monte-Carlo code TRIPOLI-4 (R). With this tool, the fission-shape evolution and the reactivity are computed for each time step. These quantities are then compared to the output of static stochastic methods based on eigenvalue computations and the Modified Source Method (MSM) to validate the hypotheses of those methods. There is a reactivity discrepancy of 2 % between the MSM model and the dynamic simulation. The fission shape is predicted with a maximum of 2 % discrepancy by the MSM method. This is significantly lower than the direct comparison to the fundamental mode, which proves the interest of a constant-delayed source hypothesis for a quasi-static approach. The nature of the rod-drop (instant or ramp-like) does not influence the fission shape and reactivity values so much, confirming the validity of the prompt-drop hypothesis. The sensitivity of reactivity to kinetic parameters is limited. These results contribute to the identification of the validity domain of the MSM method. Hypotheses investigation and sensibility studies produced pave the way to even better experimental interpretations.