Thermal–hydraulic analysis and code assessment for ATLAS 6-inch cold leg break (SBLOCA) test using MARS-KS

• The thermal–hydraulic analysis using MARS-KS was conducted for 6-inch cold leg break test of ATLAS.
• The assessment of the code capability to simulate the transient thermal–hydraulic behavior for SBLOCA was performed.
• Additional discussions including comparison with RELAP5 were made for the cause of core level dip at SIT injection.
• MARS-KS code has good capabilities to simulate cold leg break SBLOCA.
• However, the condensation model needs to be improved to predict more accurate results.

The thermal–hydraulic analysis using MARS-KS code was performed for 6-inch cold leg break test of ATLAS (Advanced Thermal–Hydraulic Test Loop for Accident Simulation), which was the second domestic standard problem. The calculation results were compared with experimental data to assess the code capability to simulate the transient thermal–hydraulic behavior for small break loss of coolant accident (SBLOCA). The sequence of events, except for the location of loop seal clearing (LSC) and safety injection tank (SIT) injection time was predicted well. The loop seals of 1A and 2B intermediate legs were cleared at 398 s in the experiment, while that of 1A was only cleared in the calculation at the same time. The prediction showed good agreement with the experimental data for pressurizer pressure and break mass flow rate. The sudden decrease and increase of water level at the LSC time were predicted qualitatively. After LSC, there was significant water level dip at SIT injection time which was not seen in the experiment. In addition, sensitivity study to investigate the cause of core level dip at SIT injection time was performed and discussions were made for it. In conclusion, MARS-KS code has good capabilities to simulate cold leg break SBLOCA, however, including interfacial heat and mass transfer, especially condensation model needs to be improved to predict more accurate results.