In support of a header/feeder phenomena study, an adiabatic, near-atmospheric, air-water flow loop was commissioned simulating a single feeder of a Pressurized Heavy Water Reactor’s primary heat transport system under a postulated Loss of Coolant Accident scenario. An extensive database in representative two-phase flow conditions was collected, 750 tests in total, in order to create a two-phase flow map to be used in the more complex geometries such as header/feeder systems. The flow loop consists of two vertical test sections, for upwards and downwards flow, and one horizontal test section, each with an inner diameter of 32 mm and at least 120 diameters in length. Superficial velocities extended up to 6 m/s for the water and 10 m/s for the air. Void fraction was measured by means of quick-closing valves and a pair of wire-mesh sensors (WMS) in each test section.

Two-phase repeatability tests showed that the liquid and gas superficial velocities varied by 1.1% and 0.6% at reference conditions of 2.0 and 2.8 m/s, respectively. The corresponding void fraction measurements varied for the quick-closing valves by at most 6.8%, which indicates a low sensitivity to the closure time of the valves and an appropriate axial distance between them, and 2.3% for the WMS. For both measurement techniques, the largest variations occurred in the vertical downwards test section.

For the formal two-phase tests, over 600 distinct flow conditions were performed. The results showed that the two measurement techniques agreed within 5% at high void fractions and low liquid flow rates in vertical flow. For all other cases corresponding to the transitional or dispersed bubbly flow regime, the WMS over-estimated the void fraction by a consistent bias. An empirical correction is proposed, with a root-mean-square error of 6.6% across all tests. The void fraction map resulting from this database provides validation for the WMS measurements, a quantitative assessment of its uncertainty and range of applicability, and will be used as a reference in future tests under similar scale and flow conditions.

This content is only available via PDF.
You do not currently have access to this content.