Reactivity Initiated Accidents and Loss of Shutdown - 20 Years Later

Abstract

A review of the safety of Ontario’s nuclear power reactors was conducted in 1987 after the Chernobyl accident. As part of this review an analysis was performed of a Loss of Coolant Accident in a Pickering A unit with coincident failure to shutdown. This analysis showed that the power excursion was halted by channel and calandria vessel failures leading to moderator fluid displacement. The containment structure did not fail and, at worst, might suffer minor cracking at the top of the dome of the reactor building. Overall the dose consequences of such an accident were no worse than the limiting design basis dual failure event. In the intervening twenty years following this analysis, significant experimental information has been obtained that relates to power pulse behaviour. This information, together with conservatisms in the original analysis, are reviewed and assessed in this paper. In addition, the issue of reactivity initiated events in other reactor types is reviewed to identify the reactor design characteristics that are of importance in these events. Contrary to popular belief the existence of positive coolant void reactivity is not as significant a factor as it is sometimes stated to be. On balance, with appropriate design measures, no one reactor type can be claimed to be “more safe” than another. The underlying basis for this statement is articulated in this paper.