Mitigation of small-break LOCAs in pressurized water reactor systems

Any small break loss of coolant accident (LOCA) in a pressurized water reactor (PWR), including the experience at Three Mile Island Unit 2 (TMI-2) with the open relief valve, can be successfully terminated at virtually any phase in the transient. If such an event were left totally unchallenged, it would lead to significant core damage and might lead to subsequent release of radiation. This report discusses the manual and automatic actions typically available to terminate the progression of a small-break LOCA. It is recognized that the small-break LOCA covers a wide range of break sizes and potential reactor system responses. For many small-break LOCAs, early automatic safety system responses are required (and are available) to mitigate or control the initial thermal-hydraulic transients that potentially can cause early core damage (within about the first 10 minutes). The general theme of this report is to explore the operating window concept for mitigating a small-break LOCA at any point in a hypothetical accident sequence. This window involves both the time frame available for decisions and the range of options available for mitigation. The methods presented in this report for countering a small-break LOCA in a PWR use the TMI-2 system as a base for determining the types, numbers and capacities of Engineered Safety Feature (ESF) systems. Based on a preliminary analysis by NSAC, these methods point to a significantly lower probability of core melting than those in the Reactor Safety Study (WASH-1400) and a much lower probability of containment building failure. These differences result from stepping beyond the very conservative WASH-1400 assumptions and considering the operating window that is available in a PWR system for providing effective countermeasures against progression of a small-break LOCA. Several features of a small-break LOCA in a PWR strongly enhance the ability to respond effectively and to minimize the resulting damage. In discussing these features, again it is assumed that initially automatic ESF responses have occurred for those small-break LOCAs requiring such actions early in the accident. Comparison of large-break LOCA conditions makes evident the much faster and stronger countermeasures required to control the initial phases of a large-break LOCA. After that the large-break LOCA has similar flow requirements for core cooling as a larger small-break LOCA (3 to 4 inch diameter hole) which has continued leakage, as described in this report. The quantitative evaluation of the degree to which WASH-1400 overestimates the probability of progressing from a badly damaged core to core melting and containment failure is still under study by NSAC but it is clear that in general, there is a substantial margin of safety between core melt and containment building failure, and specifically, that the perception of an imminent massive release of radioactivity at TMI-2 was incorrect