Effect of ullage gas on sloshing, Part I: Global effects of gas–liquid density ratio

Abstract

Previous analytical, numerical, and experimental studies have proved that properties of ullage gas influence sloshing wave impacts and induced pressures. One of those properties is ullage gas density (ρg), also considered in dimensionless form as gas–liquid density ratio (DR). Previous studies intended to study the effects of DR sometimes mixed DR effects with gas compressibility and ullage gas pressure effects and attributed them only to DR. This study is based on experiments and is meant to focus only on the effects of DR far from impact zones which will be addressed as global effects. Effects of DR near impact zones and before detection of any compressibility effects are addressed as local effects and will be treated in part II. Quantitative and statistical comparisons will be presented in part III.Test setup consisted of two 2D tanks as transverse slices of tank 2 of a membrane LNG carrier with total capacity of 152000 m3 at scales 1:20 and 1:40 at 20% fill level. Using two liquids of water and sodium polytungstate (SPT) and different ullage gases of helium (He), air, two mixtures of sulfur hexafluoride (SF6) and nitrogen (N2), and pure SF6, all at atmospheric pressure, provided a range of DRs from 0.0002 to 0.0060.For the tested range of DR, the global effects of DR on sloshing geometry (free surface) are small when comparing at the same scale or at two different scales. Small discrepancies exist even between two exact repetitions with the same DR at the same scale. Global flow keeps the same phase regardless of tested DRs which implies that all breaking wave impacts take place at the exact same time instants, considering a small time window (usually smaller than 100 ms). This was confirmed by verifying the recorded impact times.Based on this, scaling in sloshing model tests as well as the effect of changing the ullage gas can be investigated impact by impact as opposed to the solely statistical approaches adopted so far. It also helps to track down impacts measured at full-scale (on board the ship) and to further verify whether sloshing model tests are representative for them or not. The stochastic nature of sloshing can be studied more in depth also with the help of high-speed video recordings for corresponding wave impacts. Reevaluations of the current statistical sloshing assessment methodologies can be envisaged.