ML19221B425
| ML19221B425 | |
| Person / Time | |
|---|---|
| Site: | 07109375 |
| Issue date: | 08/09/2019 |
| From: | Holtec |
| To: | Office of Nuclear Material Safety and Safeguards |
| Shared Package | |
| ML19221B423 | List: |
| References | |
| EPID L-2019-LRM-0006 | |
| Download: ML19221B425 (4) | |
Text
NRC RSI (Chapter 2 - Structural Evaluation)
Provide animation and analysis files for the remaining portion of the 30 ft HAC (oblique) drop and several other drop simulations.
The applicant has provided a partial submission of the 30 ft HAC oblique drop scenario results generated by LS-DYNA. Specifically, the results from the 30 ft HAC (oblique) drop scenario appear to correspond to a drop progression that has not fully terminated (results are truncated at 0.06049 seconds (frame 244))
which indicate that the cask has not made complete contact with the target.
Provide and/or update the results in the SAR that correspond to a complete 30 ft HAC (oblique) drop.
In addition, the following files are requested:
- The input files used to generate the 1 ft NCT drop (top end).
- Output files for the 30 ft C1 sensitivity analysis This information is needed by the staff to determine compliance with 10 CFR 71.71(c)(7) and 71.73(c)(1).
Holtecs Response to RSI With this RSI response, the LS-DYNA input/output files for all NCT and HAC drop simulations have been submitted to the NRC, and they have been verified for completeness.
Regarding the 30 ft HAC oblique drop scenario, the prior simulation was intentionally terminated after the cask primary impact with the rigid target since subsequent secondary or tertiary impacts of the cask with the target were not considered critical for the safety evaluation, i.e, the cask is expected to bounce off the target and drop with a much lower impact velocity.
Nonetheless, as per the request the HAC 30 ft oblique drop simulation has been extended to capture the cask secondary impact with the target surface. The latest results from the secondary impact, which are now truncated at 0.24 seconds, are documented in Appendix D of Holtec report HI-2177539, Rev. 3.
Based on the simulation results, it is observed that the results for the containment boundary stresses, the seal compression, and the deformation of the DSB parts are all governed by the primary impact. The LS-DYNA input/output files for the extended simulation, which include the primary and secondary impacts, are provided.
- 1. For the 30 ft HAC oblique drop, the primary stress results from the primary impact remain governing when compared to those associated with the cask secondary impact. Therefore, the SAR is unaffected by the extended runtime for the oblique drop scenario. The calculation package is updated to document the results arising from the cask secondary impact.
- 2. The input files for the 1 ft NCT drop (top end) are submitted to NRC staff as part of this RSI response. All previously provided output files are confirmed to be correct.
to Holtec Letter 2404014-NRC Page 1 of 4
- 3. The only missing file binout related to the 30 ft C1 sensitivity analysis is now provided with this response submittal.
to Holtec Letter 2404014-NRC Page 2 of 4
NRC OBSERVATION Justify the parameters used to model the targets concrete in LS-DYNA for the 1/4-scale prototype drop simulation.
MAT_PSEUDO_TENSOR (*MAT_016), used to represent the concrete portion of the target, incorporates values of the Lawrence Livermore National Laboratories (LLNL) target (as reported in UCRL-ID-126295) which has different concrete compressive strength and target characteristics than those of the Sandia National Laboratories (SNL) target that was effectively used for the 1/4 scale drop tests.
The applicant has not demonstrated that the LS-DYNA model used to represent the target at LLNL is representative of the one used at SNL. The applicant has also not justified the values of the parameters used in the *MAT_016 material card, such as cohesion, pressure hardening coefficients, effective plastic strain coefficients, yield stress values, etc.
The staff notes that the applicant was reminded of this concern, during the August 24, 2018, pre-application meeting. The publicly available meeting summary (ADAMS Accession Number ML18256A100) notes that the applicant shall provide additional information on the derivation of properties based on those of the test site at SNL or perform a sensitivity analysis for the variation of the concrete parameters.
This information is needed by the staff to determine compliance with 10 CFR 71.71(c)(7), 71.73(c)(1),
and 71.73(c)(3).
Holtecs Response to NRC Observation Based on the original results for the 1/4-scale prototype drop simulation, it is noted that the energy absorbed by the target pad (i.e., its internal strain energy) is only a small fraction of the total impact energy. It is also observed that this small fraction of energy is absorbed entirely by the 8-inch thick steel top plate, which rests atop the concrete mat. This implies that the cask dynamic response is not significantly influenced by the concrete material properties. However, in order to confirm this, a sensitivity simulation is performed based on the following.
Layer Thickness Material Specification and Strength LS-Dyna Material Model Steel Top Layer Same as before (see Table 7.4 of HI-2167517)
Bottom Layer 3.75 in Concrete Compressive Strength = 7500 psi Elastic material
- Mat_001 Specifically, the following two changes are considered in the sensitivity simulation:
to Holtec Letter 2404014-NRC Page 3 of 4
- 1. Modified the concrete material model from MAT_016 (MAT_PSEUDO_TENSOR) to MAT_001 (elastic).
- 2. Modified the concrete compressive strength from 6000 psi to 7500 psi.
The table below shows the comparison of deformation results for the cask corner being impacted and the impact target. Further details are furnished in the revised report HI-2167517, Revision 3.
Package Component Deformation Predicted from prior CGOC Simulation (in.)
Deformation Predicted from Sensitivity Simulation (in.)
Outer Shell Deformation Depth Measure = 2.39 Deformation Along:
Cask Longer Edge = 2.39 Cask Shorter Edge = 4.0 Depth Measure = 2.39 Deformation Along:
Cask Longer Edge = 2.39 Cask Shorter Edge = 4.0 Target Deformation Indent Length = ~ 4.22 Indent Width = 3.29 Indent Length = ~ 4.22 Indent Width = 3.29 Based on the results above, it is concluded that the cask dynamic response is not sensitive to the concrete target material model and its stiffness.
to Holtec Letter 2404014-NRC Page 4 of 4