|
LS-DYNA Helps Ensure Public Safety with Analysis of Spent Fuel Transport Cask
Figure 1. Quarter-Scale Model Plot of the UMS® Transport Cask Analysis Details LS-DYNA is used to simulate this impact. The model is shown in Figure 1, which corresponds to the NAC UMS® transport cask. The cask body for the model need only represent the flexural characteristics of the full-scale design. Actual stress evaluations indicate that the body remains elastic during the impact. The impact limiters are comprised of redwood and a series of steel gussets and plates to ensure that the redwood remains in position during the impact. Stainless steel type 304 is employed throughout the impact limiter design due to its ductility. In generating the elements for the model, care was exercised to use brick elements to minimize the problem of severely deformed elements. The material for the redwood itself corresponds to crushable foam, which uses a stress versus volumetric strain curve. Separate material testing was performed to identify this curve for a temperature range as well as different strain rates. This particular crushable foam material was modified to permit multiple strain rates to be employed as for the Fu-Chang foam model. Previous analysis and testing indicated that the crushable foam gave better correlations with testing than for the Fu-Chang model. Part of the end section of the model was complicated by the design of a trunnion in the impact limiter region. To avoid additional elements, the trunnion was modeled separately and then connected to the cask body using the option of “contact_tied_nodes_to_surface”*. The impact limiters and the body are treated as separate parts and interact via the “contact_ automatic_ single_surface” option. The typical adjustments are made to prohibit pass-through of the parts. Other than the symmetry conditions, the only boundary conditions consisted of the initial velocity (527.4 inches/second) and the body force to ensure that the total potential energy is taken into account. Analysis Results The execution of the analysis is straightforward, and the primary results required are the displacement of the cask body into the impact limiters (crush) and the maximum accelerations developed during the impact. The displacements and velocities usually do not contain significant noise in the time histories. However, the accelerations usually produce a significant level of noise since the accelerations are related to the forces acting on the elements, and the elements have a significant level of oscillations. These ocillations give no indication of the accelerations associated with either deformation of the cask or the rigid body response of the cask ends. The Butterworth filter, which has been benchmarked for impact in other applications, is used in this application. The filter frequency was identified through a separate modal analysis of the transport cask. The resulting time history trace (shown in Figure 2) was compared to the acceleration trace of the quarter-scale model testing performed at Sandia National Laboratories. The comparison was considered to be excellent. The simulation not only gave good agreement with the maximum values but reflected a similar elastic response as well. The accelerations points used in the analysis corresponded to the locations of the accelerometers attached to the quarter-scale model.
Figure 2. UMS® Quarter-scale Model, Side Drop Accelerations at Cask Top Figures 3 and 4 show a comparison of the quarter-scale model after the 30-foot impact onto the unyielding surface for the analysis model and actual quarter-scale test model respectively. The comparison indicates that the model showed similar buckling of the shells as well as the small rotation of the limiters due to the loading of the cask body on the impact limiters.
Figure 3. Predicted Impact Limiters Deformation After Side Drop
Figure 4. Deformed Impact Limiters Specimen After Side Drop Test Conclusion This analysis and test effort demonstrated that LS-DYNA could be used to predict the performance of the impact limiters using only the design dimensions and the material properties of the redwood. This is an essential step in being able to design the impact limiters and to demonstrate that they function as designed for a variety of conditions. This realizes a significant cost savings in that it limits the level of testing of the design. |
n important set of
regulations contained in Code of Federal Regulations, Part 71, gov- ern the
transport of nuclear spent fuel to ensure safety to the public. This identifies
the accident conditions that the transport cask must be designed to withstand.
The most severe condition is the impact of the transport cask with an
infinitely rigid plane from a height of 30 feet. The design criteria for the
cask and contents are ultimately based on stresses. However, the design of the
cask includes attachments, designated as impact limiters, which are designed to
limit the accelerations experienced by the cask and its contents. Impact
limiters are typically comprised of a crushable material, such as honeycomb
aluminum, redwood or balsa, and are attached to each end of the cask. It is
necessary to predict the accelerations resulting from the impact. It is
important to validate the analysis methodology. In this application it is
accomplished by performing a quarter-scaled drop test.
