Investigation of Flashing Flow of Subcooled Liquid Through MicroSlits and Cracks

Abstract

When documenting a leak-before-break integrity assessment for pressurized components, it is important to demonstrate that the leak is readily detectable. This assumes the ability to predict the flaw opening area and corresponding leakage rate. The ability to estimate the leak rates from the throughwall flaws (cracks, slits, pits, frets, etc.,) is important in terms of radiological source potentially to be released into the environment as well as to the overall safe operation of nuclear power plant. In this study an experimental program and analysis methods were developed to measure and assess the chocking flow rates of initially subcooled water through micro-slits and cracks geometries. Experimental program involved chocking flow for various simulated throughwall flaw geometries for vessel pressures up to 7 MPa with various subcoolings. Measurements were performed on subcooled flashing flow through well-defined throughwall slits and cracks of thinwalled piping and vessel components with L/D <5.5, and liquid subcoolings between 14 and 510C. As subcooling increases the flashing critical flow rate also increases. However, from the comparison of data it was noticed that there wasn’t a trend that could be spotted of the effect of mass flux on subcoolings, and this appeared to be due to the small channel length, which prevents flow from developing and can essentially be thought of as a stagnant fluid that is accelerated through the crack and choking/flashing slightly downstream of the crack. Comparisons in between samples of this study and further comparison with other relevant studies in literature revealed: i. A link between lowering L/D rates and increasing Mass flux for identical conditions of subcooling and pressures. Though this relation is not noticeable for a small change in L/D of 2.5 (from 4.5 to 2), It is easily noticeable over a larger change. There appeared to be a trend which requires further work to develop a correlation. ii. The reducing importance of subcooling as a parameter affecting subcooled choked flow for small L/Ds and low channel lengths. iii. The discharge coefficients of the cracks are surprisingly high and not as small as one would expect for an orifice due to the sudden area change. Both, homogeneous equilibrium and non-equilibrium mechanistic models were developed to model two-phase critical flow through cracks and slits. A comparison of the model results with experimental data shows that homogeneous equilibrium based models significantly under predict critical flow rates in such geometries, while non-equilibrium models improve the accuracy of the predictions. The predictions using homogeneous non-equilibrium model agreed well within 10&percnt; of experimental data for chocking discharge flow from such throughwall flaws.