Benchmark Experiments for Steady-State Natural Convection in Fuel Rod Bundles

Abstract

Natural convection is a phenomenon in which fluid flow surrounding a body is induced by a change in density due to the temperature difference between the body and fluid. After removal from the pressurized water reactor (PWR), decay heat is removed from nuclear fuel bundles by natural convection in spent fuel pools for up to several years. Once the fuel bundles have cooled sufficiently, they are removed from fuel pools and placed in dry storage casks for long-term disposal. Little is known about the convective effects that occur inside the rod bundles under dry-storage conditions. Simulations may provide further insight into spent-fuel dry storage, but the models used must be evaluated to determine their accuracy using validation methods. The present study investigates natural convection in a 2 × 2 fuel rod model in order to provide validation data. The four heated aluminum rods are suspended in an open-circuit wind tunnel. Boundary conditions (BCs) have been measured and uncertainties calculated to provide necessary quantities to successfully conduct a validation exercise. System response quantities (SRQs) have been measured for comparing the simulation output to the experiment. Stereoscopic particle image velocimetry (SPIV) was used to nonintrusively measure three-component velocity fields. Two constant-heat-flux rod surface conditions are presented, 400 W/m2 and 700 W/m2, resulting in Rayleigh numbers of 4.5 × 109 and 5.5 × 109 and Reynolds numbers of 3450 and 4600, respectively. Uncertainty for all the measured variables is reported.