Analysis of the FIST integral tests 4DBA1, 6SB2C and T1QUV with TRAC-BFl/v2001.2

As part of the assessment of the frozen version of the PSU TRAC-BFl/v2001.2 and its qualification as a LB-LOCA and SB-LOCA code, in this work, we shall outline the comparisons between measurements and code predictions for three FIST tests: The LB-LOCA test 4DBA1, the SB-LOCA test 6SB2C and the failure to maintain water level test T1QUV. We shall study the effect of the number of axial levels in the active core as well as (in the case of the SB-LOCA test 6SB2C) the effect of the timing of the activation of the reflooding options/heat transfer package on the code predictions. Furthermore, we shall show that by using the upwinding option of some terms of the three-dimensional momentum equations, severe mass–error problems appearing in the analysis of the test T1QUV can be resolved. Generally, we shall show that although there are some differences between measurements and predictions, TRAC-BF1 captures quite well the overall behaviour of the LB-LOCA transient (depending on the number of axial nodes in the core) but underpredicts the rod surface temperatures of the SB-LOCA test 6SB2C.