ANALYSIS OF THE DRYOUT INCIDENT IN THE OSKARSHAMN-2 BOILING WATER-REACTOR

Early in 1988 dryout of fuel rods occurred in the Oskarshamn 2 boiling water reactor. During refuelling it was observed that one corner rod was damaged in each of four fuel assemblies. These were of the SVEA design, SVEA being the trade name of the ABB Atom water cross fuel. The damaged zone covered about 180-degrees of the rod periphery facing the corner sub-channel, over a stretch of about 30 cm with the upper end just below the last downstream spacer. The dominating cause of the dryout was re-use of fuel channels for ordinary 64-rod fuel, which were located in neighbouring positions to the SVEA fuel. The re-used channels showed excessive bowing because of irradiation. This bow increased the water gap between the fuel assemblies, thus increasing the neutron moderation and the local power around one corner of the SVEA fuel. This and some other factors caused the local peaking factor for the corner rod to increase from approximately 1.04 to approximately 1.38. The flow and power conditions in the damaged fuel assemblies were calculated by means of the POLCA, PHOENIX, CASMO and CONDOR computer programs. The results of these calculations were used as a base for dryout predictions, which were carried out employing eight correlations, which are available in the open literature. The Barnett, the Becker and the Bezrukow correlations predicted the dryout power within 1%. Also the Condie & Bengston, the EPRI and the XN-1 correlations yielded very good results with accuracies of, respectively, -5.1, -2.3 and 7.3%. The Becker, the XN-1, the Bezrukow and the Condie & Bengston correlations predicted dryout to occur inside of the observed dryout zone of 30 cm length. It is concluded that the dryout in the Oskarshamn 2 nuclear power plant was not caused by any faults in the design or manufacture of the SVEA fuel, and that the re-use of fuel channels should not be permitted.