| Popis: |
Safe transportation of radioactive material using appropriately designed, certified packages ensures protection of the public and environment using rigorous system packaging requirements. This protocol was conducted to determine the effects of drop testing on a Type B shipping package (DPP-3) performance per normal and hypothetical accident conditions tests. Preparation, testing, and post-test evaluations of a Type B shipping package tested to 10 Code of Federal Regulations (CFR) 71.71 and 71.73 requirements are presented. Six packages were subjected to free drop, puncture, and thermal tests per 10 CFR 71.73. The posttest helium leak test showed a leak rate < 1 × 10−7 He-cc/sec, indicating that the containment boundary and leak-tight environment were maintained. Statistical analyses were performed to determine changes in package performance. The null hypothesis was that measurements would not change enough from pre- to post-testing to show statistical significance, which would indicate too much variability in package design. A repeated measures t-test was used to analyze drum lid and containment vessel (CV) torques. All six CV lid torques were statistically significant, but only three drum lid torques were statistically significant. Therefore, lid design caused more variability than necessary when subjected to the horizontal-and-cold, vertical-lid-down, and corner-with-lid-up drop tests. A one-way analysis of variance (ANOVA) analyzed package heights and diameters, with results on height indicating four of the six certification test units (CTUs) were statistically significant. Thus, CTU design caused more variability than necessary when subjected to horizontal-hot, horizontal-cold, corner-lid-up, and corner-lid-down tests. One-way ANOVA analysis of diameter indicated that all six CTUs were statistically significant. Thus, CTU design caused more variability than necessary when subjected to all tests. The DPP-3 has been shown to be an acceptable design for transport of radioactive materials, but statistical analysis revealed that package and test design can be improved to reduce variability in performance before and after testing. |
