Prediction of reflood behavior in a bundle with deformed and un-deformed rods in ATHER 5 × 5 experiment using MARS-KS code

• Published in: Nuclear engineering and design 2022-06, Vol.392, p.111770, Article 111770
• Main Authors: Bang, Young Seok, Oh, Deog Yeon, Lee, Joosuk, Kim, Jongrok, Moon, Sang-Ki
• Format: Article
• Language: English
• Subjects: ATHER 5×5 rod bundle experiment; Cladding; Conductors; Convective heat transfer; Deformation effects; Experiments; Flow paths; Friction; Heat; Heat loss; Heat transfer; Heat transfer coefficients; Housing; MARS-KS code; Multi-dimensional modeling; Multipliers; One dimensional models; Quenching; Reflood behavior of deformed rods; Rods; Sleeves; Thermal conductors; Three dimensional models
• ISSN: 0029-5493; 1872-759X
• DOI: 10.1016/j.nucengdes.2022.111770

•We calculate the ATHER experiment simulating the reflood behavior with deformed fuel rods array using MARS-KS code.•We modeled the sleeve thermal conductor, fouling by sleeve-to-sleeve contact and the heat loss through the housing.•Difference in thermal behavior between the deformed rods and non-deformed rods was understood through the multi-dimensional calculation. A reflood experiment at ATHER 5×5 rod bundle facility having an array of deformed rods is calculated using MARS-KS code. In the experiment, the deformed rods were represented by the heater rods with sleeves on the outside. To represent properly the configuration, the sleeve thermal conductors, the steam gap between the heater rod and the sleeve, and the fouling factor into the heat transfer area by contact between adjacent sleeves are implemented. The portion into which the sleeve is installed and the portion not installed are modeled as separate heat structures. Also the heat loss through the housing of the test section is modeled. Within the range of the present calculations, imposing a convective heat transfer coefficient of 2 W/m2·K over the housing can be considered appropriate for the actual heat loss. Calculations are conducted using one-dimensional model with single channel, one with two channels, and three-dimensional model by ‘multid’ component of the MARS-KS code. From the calculation results, the predictability of the code and the present modeling scheme is discussed in terms of overall response, peak cladding temperature, and time of quenching. Sensitivity study on interfacial friction model multiplier is also performed, which shows the quenching time can be improved by applying a multiplier 1.05 to the interface friction model in the one-dimensional case, but its effect on the maximum cladding temperature and the rewet temperature seems not to be significant. The differences in phenomena between the deformed rods and un-deformed rods and their reasons are discussed. The effect of 90% of partial blockage is expected less than 30 K in terms of maximum cladding temperature for both deformed and non-deformed rods, compared to the hypothetical case of the completely un-deformed rod bundle, because cooling can be maintained by flows through the remaining unblocked flow paths.