STUDY ON THE GEOMETRIC EFFECTS OF WIRE-WRAPPED ROD BUNDLE AND U-BEND ON SINGLE-PHASE AND TWO-PHASE FLOW PHENOMENA

Flow restrictions like wire-wrapped rod bundles and U-bend geometries are critical in fast reactor cores, Pressurized Water Reactor (PWR) steam generators, and helical-coil steam generators. This study investigates their effects on single-phase and two-phase flows to enhance the understanding of thermal-hydraulic phenomena in nuclear reactor systems. A hydrodynamic test facility for a 7-pin wire-wrapped rod bundle was designed and constructed using scaling analysis. The test section consists of seven wire-wrapped stainless-steel rods, each 19.05 mm in diameter, housed in an acrylic hexagonal channel measuring 2092 mm in height. Experimental pressure drop data was collected across a wide range of Reynolds numbers. The data validated the Upgraded Cheng and Todreas Detailed (UCTD) correlation within ±10% accuracy, and the SST k-ω turbulence model was identified as the most reliable for predicting pressure drops in 7-pin wire-wrapped rod bundle through Computational Fluid Dynamics (CFD) simulations. For the U-bend study, a new experimental database was established using the existing Purdue University separate-effects test facility, featuring a 25.4 mm inner diameter pipe and a U-bend with curvature to diameter ratio of 9. Detailed local data was collected under eight test conditions at ten measurement locations, which includes void fraction, gas velocity and bubble diameter, measured with miniaturized four-sensor conductivity probes, while pressure loss was obtained using pressure transducers. Mechanistic models were developed to characterize the U-bend effects, including pressure loss, variance of void fraction, U-bend dissipation length and bubble velocity. The Lockhart-Martinelli correlations can be used to predict two-phase pressure drops across U-bend with some modifications. Variance of void fraction represents U-bend strength, dissipating exponentially with dissipation length determined by the dissipation rate. A modified Froude number is used to model variance of void fraction, dissipation rate, U-bend dissipation length and bubble velocity, with predictions generally within ±10% accuracy. Other closure models needed in the Interfacial Area Transport Equation (IATE) were also developed. Experimental data revealed strong correlation between variance of void fraction and covariance of Random Collision, modeled using the modified Froude number. Model coefficients for bubble interaction terms were determined by evaluating each region (i.e., vertical upward, U-bend, U-bend dissipation, vertical downward) using experimental data. The one-group interfacial area transport along the whole test section was evaluated using all these closure models and U-bend effects models, with deviations generally within ±15%. The study also identified limitations of existing Multiphase Computational Fluid Dynamics (MCFD) models in simulating bubbly flow across U-bend.