9 Non-destructive Examination of Multiphase Material Distribution in Uranium Hexafluoride Cylinders. . . 85 Fig. 9.4 Aluminum plate with epoxy pattern and highlighted scan area (left), corresponding wavenumber plot (right) Fig. 9.5 3/800 aluminum plate prepared with epoxy with scan area highlighted (left), and corresponding wavenumber plot of scan region (right) not be coupled to the fill area in that region. In the cylinder tested, the UF6 was expected to be a solid block at the bottom of the fill area based on its handling and storage records. Many different scans were made of the available UF6 storage cylinder at a variety of different scan parameters. A resultant wavenumber plot of one of these UF6 scans can be seen in Fig. 9.7. In this figure, the processed image is shown on the left while the raw vibrational surface velocity image is shown on the right. A very clear low wavenumber region can be seen between the two welds, located mainly between 40 and 90 mm on the vertical axis. The difference in wavenumber between the low and high regions is relatively small compared to the difference measured in the can tests, likely due to the energy dissipation caused by the cylinder’s larger size and wall thickness (1/400). As expected, wavenumbers above and below the expected fill height were similar, and were higher than the suspected fill region, suggesting the presence of a gas behind the cylinder wall. This image reflects the expected results fairly well; the low wavenumber region begins at the lower weld and ends at approximately the hypothesized fill height. This region is very likely indicative of the presence of UF6. However, the distance between the hypothesized fill height and the weld is within the resolution distance (about 1.5 cm) found in the previous test shown in Fig. 9.4. This means that the weld could be obscuring or distorting the actual UF6 fill profile in the
RkJQdWJsaXNoZXIy MTMzNzEzMQ==