Determination of the bias in LOFT fuel peak cladding temperature data from the blowdown phase of large-break LOCA experiments

Abstract

Data from the Loss-of-Fluid Test (LOFT) Program help quantify the margin of safety inherent in pressurized water reactors during postulated loss-of-coolant accidents (LOCAs). As early as 1979, questions arose concerning the accuracy of LOFT fuel rod cladding temperature data during several large-break LOCA experiments. This report analyzes how well externally-mounted fuel rod cladding thermocouples in LOFT accurately reflected actual cladding surface temperature during large-break LOCA experiments. In particular, the validity of the apparent core-wide fuel rod cladding quench exhibited during blowdown in LOFT Experiments L2-2 and L2-3 is studied. Also addressed is the question of whether the externally-mounted thermocouples might have influenced cladding temperature. The analysis makes use of data and information from several sources, including later, similar LOFT Experiments in which fuel centerline temperature measurements were made, experiments in other facilities, and results from a detailed FRAP-T6 model of the LOFT fuel rod. The analysis shows that there can be a significant difference (referred to as bias) between the surface-mounted thermocouple reading and the actual cladding temperature, and that the magnitude of this bias depends on the rate of heat transfer between the fuel rod cladding and coolant. The results of the analysis demonstrate clearly that a core-wide claddingmore » quench did occur in Experiments L2-2 and L2-3. Further, it is shown that, in terms of peak cladding temperature recording during LOFT large-break LOCA experiments, the mean bias is 11.4 {plus_minus} 16.2K (20.5 {plus_minus} 29.2{degrees} F). The best-estimate value of peak cladding temperature for LOFT LP-02-6 is 1,104.8 K. The best-estimate peak cladding temperature for LOFT LP-LB-1 is 1284.0 K.« less