Instantaneous Leakage Evaluation of Metal Cask at Drop Impact

There have been a lot of tests and analyses reported for evaluation of drop tests of metal casks. However, no quantitative measurement has ever been made for any instantaneous leakage through metal gaskets during the drop tests due to loosening of the bolts in the containments and lateral sliding of the lids. In order to determine a source term for radiation exposure dose assessment, it is necessary to obtain fundamental data of instantaneous leakage. In this study, leak tests were performed by using scale models of the lid structure and a full scale cask without impact limiters simulating drop accidents in a storage facility, with aim of measuring and evaluating any instantaneous leakage at drop impact. Prior to drop tests of a full scale metal cask, a series of leakage tests using scale models were carried out to establish the measurement method and to examine a relationship between the amount of the lateral sliding of the lid and the leak rate. It was determined that the leak rate did not depend on the lateral sliding speeds. Drop tests of a full scale metal cask without impact limiters were carried out by simulating drop accidents during handling in a storage facility. The target was designed to simulate a reinforced concrete floor in the facility. The first test was a horizontal drop from a height of 1 m. The second test simulated a rotational impact around an axis of a lower trunnion of the cask from the horizontal status at a height of 1 m. In the horizontal drop test, the amount of helium gas leakage was calculated by integrating the leak rate with time. The total amount of helium gas leakage from the primary and secondary lids was 1.99×10−6Pa · m3. This value is 9.61×10−9% of the initially installed helium gas. The amount of leakage was insignificant. In the rotational drop test, the total amount of leakage from the primary and secondary lids was 1.74×10−5Pa·m3. This value is 8.45×10−8% of the initially installed helium gas. This value was larger than that of the horizontal drop test. Nevertheless, the amount of leakage was also insignificant. The relationship between the maximum sliding displacement of the lid and the leak rate coincided between the tests of a scale model and a full scale metal cask.