Thermohydraulic Analysis of Failures in the LORELEI Test Device

The LORELEI (Light-Water One-Rod Equipment for LOCA Experimental Investigation) test device in the Jules Horowitz Reactor (JHR) is dedicated to study fuel thermo-mechanical behavior during Loss of Coolant Accident (LOCA) in power reactors. By using a Displacement Device, the fuel sample moves in the neutron flux field and generates heat according to its location. As part of the thermohydraulic design of the LORELEI test device, failure analysis is performed in order to demonstrate that the safety actions can keep the physical parameters in the required domain.
The failures of the Displacement Device and Hafnium Screen has been studied. Conservative analysis has been performed with a limiting set of boundary and operational conditions, in order to demonstrate the efficiency of the preventive actions to end each failure scenario in the adequate domain.
The Displacement Device purpose is to move the experimental setup in the neutron flux zone in order to generate the desired power (i.e. Linear Heat Generation Rate - LHGR) or temperature evolution in the fuel sample. A Displacement Device failure is defined as a loss of control in which the device starts to move forward toward the core vessel at its maximal speed.
The Hafnium screen is a neutron screen, installed on the inner side of the device holder, in order to flatten the LHGR in the fuel sample. In the Hafnium screen failure, it is assumed that the screen totally disappears, even if this failure is not realistic. In such a case where an absorber disappears, the neutron and gamma fluxes behind it increases immediately and as a consequence, the power generated in the fuel and the gamma power in the structure increase. The failure will be detected by thermocouples and in 1.5 sec safety actions will be triggered.
These two failures have been considered as incidental situations; therefore, the preventive actions must keep the physical parameters in the incidental domain. There are two thresholds in the incidental domain, the first one trigger the classified withdrawal when reaching the upper limit of the normal domain and the second triggers the preventive shut down. The results of the analyses performed in the scope of this study are temperature evolution in a Displacement Device and Hafnium Screen failures, if occurring at different instances of the experiment.
Several thermohydraulic models have been developed to simulate the Displacement Device and the Hafnium Screen failures in the different experimental phases: A model that represents the thermosyphon flow during the re-irradiation phase, a dry phase model which simulates the LOCA sequence, a reflooding model and a thermosyphon flow model for the post experiment cooling phase.
The codes that were used for the thermohydraulic models are CATHARE2 which is the French best-estimate thermal hydraulic code, COMSOL Multiphysics.
The analysis presents that for the proposed detection and action protocol, the failure consequences are acceptable.
For the Displacement Device failure: In all the experimental phases, the failure of the Displacement Device remains in the required domain as long as the preventive shut down is effective. Though, a mechanical stop block must be positioned from the core.
For the Hafnium Screen failure: the re-irradiation phase remains in the incidental domain, even if safety actions are not triggered. The dry phase and other phases will be analyzed and presented later on.