Analyses of the B and W-1810 and KRITZ-2 Critical Experiments with nTRACER

To verify the simulation capability of nTRACER for the experimental reactors, B and W-1810 and KRITZ-2 critical experiments were analyzed. By comparing with not only experimental measurements but also with the results of Monte-Carlo code McCARD, the validity of nTRACER was confirmed in the situation where the direct comparison with measurements was not appropriate. Overall, in both experiments, nTRACER solutions showed good agreements. In B&W-1810 critical experiments, out of 4 types of cores, the maximum difference of k-eff from criticality was 64 pcm. Especially, in core1 and 12 without UO2-Gd2O3, the errors were within 10 pcm. The errors of pin powers were also significantly small. For the center assembly, all RMS values were within 0.5% and the maximum error was 1.13% in the Core 5. By comparing calculation cases each other, it turned out that the modeling simplification such as negligence of the fuel rods above the water level led to the considerable errors in the nTRACER results. When run by McCARD, differences of k-eff of the models between with and without simplifications were from 295 pcm to about 1150 pcm. With the same simplified model, nTRACER results overall agree with McCARD results, with the errors within 231 pcm. The pin power errors were also reasonable. Compared with measured values detected at some specific locations, most pin power errors were less than 3.0%. In addition, compared with McCARD for the full cores, RMS and the maximum error are less than about 0.4% and 1.5% each for the cores with UO2 fuel rods and 0.5% and 1.7% each for the core with MOX fuel rods.