Experimental and modelling results of the QUENCH-20 experiment with BWR test bundle

Abstract

The experiment QUENCH-20 with BWR simulation bundle was conducted at KIT on 9th October 2019. This test was performed in the framework of EU SAFEST project and its international access to European SA research infrastructure of the users from Swedish Radiation Safety Authority (SSM) in collaboration with Westinghouse Sweden, GRS and KTH. The test objective was the investigation of the BWR fuel assembly degradation including a control blade with B4C neutron absorbers. The test bundle represented one quarter of a BWR fuel assembly. 24 electrically heated fuel rod simulators were filled with krypton (pressure of about 0.6 MPa), whereas the holes of absorber pins were filled with helium (overpressure 0.02 MPa). According to the pre-test calculations performed with ATHLET-CD, the bundle was heated to a temperature of 1230 K at the cladding of the central rod at the hottest elevation of 950 mm. This pre-oxidation phase in steam lasted 4 h. During the following transient stage, the bundle was heated to a maximal temperature of 2000 K. The cladding failures were observed at temperatures about 1700 K and lasted about 200 s. Massive absorber melt relocation was observed 50 s before the end of the transient stage. The test was terminated by the injection of quench water with a flow rate of 50 g/s into the bundle bottom. Fast temperature escalation from 2000 to 2300 K during 20 s was observed. The mass spectrometer measured release of COx and little CH4 during the reflood as products of absorber oxidation; corresponding mass of reacted B4C was 4.3% of the total mass of B4C pins. Hydrogen production during the reflood amounted to 32 g (57.4 g during the whole test) including 10 g from B4C oxidation. The results of the post-test simulations with the AC2/ATHLET-CD code show a good agreement with the experimental observations concerning the thermal behavior. However, melting and relocation of BWR components was not calculated and the B4C oxidation was underestimated.