Experimental investigation of two-phase flow evolution in a tight lattice bundle using wire-mesh sensor

Abstract

Abstract Tight lattice fuel assemblies have significant advantages in power density and conversion ratios. However, the dynamic characteristics of two-phase flow in tight lattice bundles, which is very important to safety analysis, are still not clarified. This study presents an experimental investigation of two-phase flow evolution in a double sub-channels tight lattice bundle. The experimental channel was up-scaled (1:2.7) compared with the prototype fuel. Phase distributions at four different axial positions, z/Dh\xa0=\xa028.95, 57.90, 86.86, and 115.8, are measured with a high spatiotemporal resolution by a 16\xa0×\xa032 wire-mesh sensor. The flow regime includes bubbly flow, cap-bubbly flow, and slug flow. In the bubbly flow, the large-scale pulsations across the gap of two sub-channels were observed, as same as the single-phase flow in a tight lattice bundle. The time-averaged void fraction shows a gap region peak in bubbly flow. In the cap-bubbly flow, the cap bubbles are confined to a single sub-channel due to the small gap. The cap bubbles are alternated arrangements in the two sub-channels and the large-scale pulsations across the gap are caused due to this special flow configuration. The time-averaged void fraction shows a core peak profile in a cap-bubbly flow. In the slug flow, the slug bubbles can across the small gap to span the two sub-channels and the time-averaged void fraction shows a core peak profile. Even though the time-averaged void fraction almost coincides at different axial positions, the phase interface structure shows a significantly different according to the virtual side view of void fraction. The accuracy of the drift-flux models has been evaluated with the experimental data. The Bestion s model underestimated the void fraction in the channel while the Chexal s model overestimated it. Clark s model gets the best prediction with an error of 10.1%.