RIS 2015-13, Seismic Stability Analysis Methodologies for Spent Fuel Dry Cask Loading Stack-up Configuration
| ML15132A122 | |
| Person / Time | |
|---|---|
| Issue date: | 11/12/2015 |
| From: | Todd Keene Generic Communications Projects Branch |
| To: | |
| Keene T, NRR/DPR, 301-415-1994 | |
| References | |
| TAC MF3182 RIS-15-013 | |
| Download: ML15132A122 (9) | |
ML15132A122 UNITED STATES
NUCLEAR REGULATORY COMMISSION
OFFICE OF NUCLEAR MATERIAL SAFETY AND SAFEGUARDS
OFFICE OF NUCLEAR REACTOR REGULATION
OFFICE OF NEW REACTORS
WASHINGTON, D.C. 20555-0001
November 12, 2015
NRC REGULATORY ISSUE SUMMARY 2015-13 SEISMIC STABILITY ANALYSIS METHODOLOGIES FOR SPENT FUEL DRY CASK
LOADING STACK-UP CONFIGURATION
ADDRESSEES
All holders of, and applicants for, general licenses and certificates of compliance (CoC) for an independent spent fuel storage installation (ISFSI) under Title 10 of the Code of Federal Regulations (10 CFR) Part 72, Licensing Requirements for the Independent Storage of Spent Nuclear Fuel, High-Level Radioactive Waste, and Reactor-Related Greater than Class C
Waste.
All Radiation Control Program Directors and State Liaison Officers.
The U.S. Nuclear Regulatory Commission (NRC) is also sending a copy of this regulatory issue summary (RIS) to NRC 10 CFR Part 50, Domestic Licensing and Production and Utilization Facilities, and 10 CFR Part 52, Licenses, Certifications, and Approvals for Nuclear Power Plants, licensees for information because these entities may have a general license, pursuant to 10 CFR 72.210, General License Issued, which allows persons authorized to possess or operate nuclear power reactors under 10 CFR Part 50 or 10 CFR Part 52 to store spent fuel in an ISFSI at power reactor sites.
INTENT
The NRC is issuing this RIS to share information regarding acceptable seismic stability analysis methodologies to determine seismic stability of spent fuel dry cask loading stack-up configurations. This RIS does not require any specific action or written response on the part of an addressee.
BACKGROUND INFORMATION
The stack-up configuration refers to the condition when a transfer cask containing a canister loaded with spent fuel is resting on a storage overpack. While in the stack-up configuration, the loaded canister is lowered from the transfer cask to the storage overpack. During this transfer, when the transfer cask is not attached to a single-failure-proof crane, the stack-up is free-standing and the potential exists for the stack-up configuration to become unstable and tip over during a seismic event. DEFINITIONS
Nonlinear time history analysis: Nonlinear time history analysis is an analysis which considers effects which vary with time (e.g., loading under seismic motion) and nonlinear effects such as material properties and movements in response to the applied loads.
Rocking stability: Rocking stability refers to whether the system will topple over under loading or remain standing after a seismic event.
Sliding stability: Sliding stability refers to whether the system will slide under loading or remain in place by the effect of friction after a seismic event.
Stack-up: Stack-up refers to the configuration when a transfer cask containing a canister loaded with spent fuel is resting on a storage overpack and not attached to a single failure proof crane.
SUMMARY OF ISSUE
The NRC staff has reviewed several stack-up seismic stability analysis calculations performed by licensees under, for example, 10 CFR 72.122(b) and 72.212(b). Some of the concerns that the NRC staff has identified in these reviews include:
1. Using a single time history to perform a nonlinear seismic analysis.
2. Using time histories not derived from real earthquakes.
3. Significantly altering the phasing of frequencies in the time histories.
4. Using non-conservative, low safety factors for rocking and sliding response.
5. Double counting damping in rocking analyses leading to non-conservative low responses.
6. Not benchmarking finite element models against known solutions.
7. Not evaluating the stresses in the mating device.
The RIS provides information associated with the concerns that have been identified by NRC
staff.
STACK-UP CONFIGURATION ANALYSIS
A nonlinear time history analysis is unnecessary to establish rocking stability of the stack-up configuration if a static equilibrium analysis shows that the accelerations at the center of gravity of the stack-up, considering the effects of floor flexibility and base support flexibility on frequency response, are so low that the stack-up does not pivot about the casks edge (i.e.,
there is no incipient tipping). However, if any material placed between the base of the storage overpack and the floor behaves in a nonlinear manner (i.e., has a nonlinear load-deflection curve), a nonlinear time history analysis may be necessary. Similarly, a nonlinear time history analysis is unnecessary to establish sliding stability if the same accelerations show that from static equilibrium considerations, the stack does not slide for the range of coefficients of friction applicable to the sliding surface, also considering the effects of uncertainty. For such cases, the NRC staff considers that static equilibrium provides an acceptable demonstration of kinematic stability.
Nonlinear time history analysis should be performed if the acceleration values at the center of gravity of the stack-up do not support a static demonstration of stability. As described in Standard Review Plan, NUREG-0800, Section 3.7.1,1 a minimum of five sets (each set comprised of three components: North-South, East-West, and Vertical) of ground motion time histories should be selected from real recorded ground motions when performing a nonlinear time history. Three time histories from each set should be statistically independent and the input time histories should be baseline corrected such that they yield zero final displacement and zero final velocity.
In generating the earthquake time histories, the phasing of the Fourier components associated with the real seed motions should be maintained to the maximum extent practicable. It is recognized that there will be distortion of the phase angle spectrum due to baseline correction.
There is no need to impose an arbitrary phasing requirement, since phasing is nearly random, and another equally valid time history will have completely different phasing. For this reason,
1 U.S. Nuclear Regulatory Commission, NUREG-0800 ,Standard Review Plan for the Review of Safety Analysis Reports for Nuclear Power Plants: LWR [light-water reactor] Edition, Rev. 4, Section 3.7.1, Seismic Design Parameters (Agencywide Documents Access and Management System (ADAMS) Accession No. ML14198A460). nonlinear analyses are performed using multiple time history sets with each set of time histories having different phasing.
The three-dimensional seismic (time history) analysis of the stack-up configuration should be performed for each of the five time history sets generated in accordance with Section 3.7.1 of NUREG-0800 to identify a maximum rocking angle. In addition, for each generated earthquake, the friction coefficient at the stack-up configurations base should be varied within its lower and upper bound range to account for uncertainty in the friction value. The response of the stack-up configuration should be obtained for more than two discrete values of the friction coefficients, since the worst response could come from a minimum, maximum, or intermediate value. The maximum from the mean plus one standard deviation value of the response (maximum rocking angle and maximum sliding displacement) from each discrete simulation should be defined as the computed response for the stack-up configuration, where a discrete simulation consists of five earthquake time history analyses using one friction value.
The value of the permissible angle of rotation should be equal to one-third of the critical angle of rotation of the stack-up configuration (i.e., the angle of tilt at which the center of gravity of the stack-up configuration (with the canister assumed to be at the highest elevation in the stack-up configuration) is directly over the stack-up configurations edge), consistent with, for example, American Society of Civil Engineers/Structural Engineering Institute (ASCE/SEI) Standard
43-05, Seismic Design Criteria for Structures, Systems, and Components in Nuclear Facilities.
Further, consistent with the above, the computed maximum rocking angle of the stack-up configuration should be less than the permissible angle of rotation.
The acceptance criterion for sliding should be one-third of the distance between the outer edge of the overpack and the edge of the support surface. This protects the integrity of support surfaces that provide no restraint to lateral sliding. The sliding criterion is not applicable to base support surfaces that are equipped with a physical barrier at their periphery that would prevent uncontrolled sliding. However, such physical barriers should be included in the nonlinear rocking analyses.
Analytical computer codes including the code, element types, and analysis options used in the finite element model should be validated in accordance with Spent Fuel Storage and Transportation Interim Staff Guidance 21.2
The essential features of the finite element model for stack-up consists of four components: the transfer cask with canister, the mating device, the storage overpack, and the base support structure. The transfer cask, mating device and storage overpack behave as rigid bodies.
Therefore, one way to model the stack-up is as a series of rigid bodies connected by contact elements with in-line damping at the interfaces. Such a modeling technique, however, places a dependence on the location of the contact elements and the accurate calculation of bolt element and contact stiffness, which in turn determines the accuracy of the bolt forces and impact loads produced from the analysis.
Another approach is to model all components as deformable bodies. Such models may use reduced integration hexahedral elements, which require hourglass control to inhibit zero energy modes. The hourglass control methods of Flanagan and Belytschko can be used for linear and
2 U.S Nuclear Regulatory Commission, Spent Fuel Storage and Transportation Interim Staff Guidance 21, Use of Computational Modeling Software (ADAMS Accession No. ML061080669). mildly nonlinear problems for stack-up analysis.3,4 Fully integrated elements may also be used in regions of large deformation, such as where components undergo direct impact. Thick shell elements in the critical load path (e.g., the mating device top plate, the mating device bottom plate, or the transfer cask bottom flange) should use higher-order element formulations with at least five integration points through their thickness to achieve accurate results. All contact interfaces should be modeled using standard part-to-part contact. The sliding energy should be tracked to ensure that all contact interfaces are numerically stable. Bolted connections should be modeled using a combination of higher order beam elements to simulate the bolts and one-dimensional spring elements to simulate the interface between the bolt and the bolt hole, as described, for example, in S. Narkhede, Bolted Joint Representation in LS-DYNA to Model Bolt Pre-Stress and Bolt Failure Characteristics in Crash Simulations.5 The stack-up finite element model should be described in detail (e.g., element size, type and integration order, hourglass controls, friction parameters, contact definitions, damping, etc.). The quality of the model should be sufficient to produce accurate results (e.g., sensitivity and convergence studies should be performed, as necessary, to demonstrate the quality of the model). Plots of kinetic energy, internal energy, sliding energy, and/or hourglass energy, for example, should be provided for the entire model and individual parts to show that results are within acceptable limits.
For the simulation of damping at the stack-up configuration/support-surface interface, the support-surface interface may be simulated by a set of discrete viscous dampers. The damping assigned to the interface dampers or the materials at the contact surfaces should be equal to the value necessary to provide the same (or lesser) rate of decay of the rocking amplitude of the stack-up configuration when subject to an initial tilt as that predicted by Housners classical solution.6 The initial tilt assumed in the calibration should be equal to the maximum permissible angle of tipping, and 50 percent of the maximum permissible angle of tipping. The lower of the two values of viscous damping thus determined should be used in the dynamic analysis to ensure a conservative result.
When relying on friction to restrict relative movement of the components of the stack-up configuration (i.e., the storage overpack, the transfer overpack and the mating device which joins the two), the interface friction coefficient between interfacing components should be an appropriately conservative value (e.g., for steel surfaces with an oxide layer, the lowest value is given as 0.27 in Table 3.2.3, pg. 3-22 of Marks Standard Handbook for Mechanical Engineers7).
The effect of the gap between the transfer cask and canister reduces the dynamic response of the stack-up and therefore, does not need to be modeled.
For bolt and plate stresses, the bolts joining the mating device to the transfer cask and storage overpack should meet American Society of Mechanical Engineers (ASME) Boiler and Pressure Vessel Code,Section III, Subsection NF Level D stress limits. The joint moment and shear
3 LS-DYNA Keyword User's Manual Vol. 1 LS-DYNA R7.1 May 26, 2014 (revision: 5471) Livermore Software Technology Corporation.
4 Flanagan, D.P. and T. Belytschko, A Uniform Strain Hexahedron and Quadrilateral and Orthogonal Hourglass Control, Int. J. Numer. Meths. Eng., 17, 679-706 (1981).
5 Narkhede, S, Bolted Joint Representation in LS-DYNA to Model Bolt Pre-Stress and Bolt Failure Characteristics in Crash Simulations, (11th International LS-DYNA Users Conference).
6 Housner, G.W., "The Behavior of Inverted Pendulum Structures during Earthquakes," Bulletin of the Seismological Society of America, Vol. 53, No. 2, pp 403-417, February 1963.
7 Mark's Standard Handbook for Mechanical Engineers, McGraw-Hill, Tenth Edition, (2006). should be taken as the mean-plus-one standard deviation value of the maximum moments and shear forces recorded for each discrete simulation.
Test data provided by the material manufacturer is acceptable as input data for commercially sold products used in the stack-up configuration, provided the data used in the analysis represents either the minimum or maximum material properties (as appropriate for the analysis),
where minimum and maximum are defined as the 95 percent exceedance probability, and 95 percent non-exceedance probability, respectively.
BACKFITTING AND ISSUE FINALITY DISCUSSION
This RIS describes seismic stability analysis methodologies for dry cask spent fuel storage systems which the NRC regards as acceptable for meeting NRC regulatory requirements in
10 CFR 72.212(b) and implementing provisions in individual cask Final Safety Analysis Reports,8 when preparing procedures for loading and unloading dry spent fuel storage casks at generally-licensed ISFSIs. The RIS is addressed to holders of general licenses for ISFSIs under 10 CFR Part 72. The RIS requires no action or written response on the part of these addressees.
The staff did not prepare a backfit analysis under 10 CFR Part 50, 10 CFR Part 72, or further address the issue finality criteria in 10 CFR Part 52, based on the following considerations.
The RIS does not constitute backfitting under 10 CFR Part 72 for existing ISFSI general licensees
Issuance of this RIS does not constitute backfitting as defined in 10 CFR 72.62 (the addition, elimination, or modification, after the license has been issued, of [s]tructures, systems, or components of an ISFSI or MRS [Monitored Retrievable Storage installation], or [p]rocedures or organization required to operate an ISFSI or MRS.). The information provided does not require any general ISFSI licensee to add, eliminate, or modify structures, systems, or components of a general ISFSI or MRS or the procedures or organizations required to operate them. However, this RIS addresses the underlying bases and documentation for procedures for loading and unloading spent fuel casks. The information in the RIS may cause licensees to change these procedures because they have determined that they are not in compliance with one or more provisions in 10 CFR 72.212(b). These provisions require general ISFSI licensees to, inter alia, address terms and conditions and specifications of each referenced CoC, to design the ISFSI consistent with applicable loads, including seismic loads, and to ensure that the cask and ISFSI activities such as loading and unloading are performed in accordance with the general licenses, the casks terms, conditions and specifications, and the safety analysis report for the casks. Nonetheless, the NRC has determined that issuance of the RIS does not constitute backfitting.
The RIS does not establish, recommend or suggest new safety requirements with respect to the consideration of seismic stability analysis. General ISFSI licensees are already required by existing NRC regulations, general license provisions, and their approved cask designs, to consider the site-specific seismic information when preparing procedures for loading and unloading dry spent fuel storage casks. In addition, the seismic stability analysis methodologies
8 Final Safety Analysis Reports (FSARs) for spent fuel storage casks licensed under Part 72 contain specific provisions addressing consideration of seismic loads in determining cask stability. See, e.g., FSAR for the HI-
STORM 100 Cask System, Rev. 4 (ML061040056), Chapter 2 Principal Design Criteria, pages 2.3-15 and 16. described in this RIS constitute one acceptable way for meeting an ISFSI licensees regulatory obligations in 10 CFR 72.212(b), and applicable provisions of casks utilized at each licensees ISFSI. General licensees are free to adopt other approaches, so long as licensees are able to demonstrate compliance with applicable requirements in 10 CFR 72.212, and the referenced casks terms, conditions and specifications, and the safety analysis report. Accordingly, the NRC concludes that the RIS does not constitute backfitting for general ISFSI licensees under
10 CFR Part 50, 10 CFR Part 72, or a violation of issue finality criteria in 10 CFR Part 52.
The RIS contents do not apply to holders of operating licenses under 10 CFR Part 50, or to holders of early site permits, design approvals, design certifications, and combined licenses under 10 CFR Part 52
This RIS does not apply to holders of operating licenses under 10 CFR Part 50, or to holders of early site permits, design approvals, or design certifications. Although an ISFSI licensee who is an addressee of this RIS may also be a holder of a 10 CFR Part 50 operating license or a holder of a 10 CFR Part 52 combined license, the discussion in this RIS is directed to activities and matters conducted under the authority granted by the NRCs general license for an ISFSI
and NRC-approved cask designs. The RIS is not directed to the activities controlled by either a nuclear power plant operating license under Part 50 or a combined license under Part 52 (including the matters accorded issue finality in a referenced design certification rule).
Therefore, the matters within the scope of this RIS are within the scope of matters accorded backfitting protection in the backfit rule, 10 CFR 50.109, or matters accorded issue resolution and issue finality under the applicable issue finality provisions in 10 CFR Part 52.
FEDERAL REGISTER NOTIFICATION
Although this RIS is informational and does not represent a departure from the current regulatory requirements, a notice of opportunity for public comment on the draft RIS was published in the Federal Register (80 FR 21770) on April 20, 2015, for 45 days. Four organizations and individuals provided comments, which were considered before issuance of this RIS. NRC staff reviewed all comments, and the comments and responses are available in ADAMS at Accession No. ML15188A534.
CONGRESSIONAL REVIEW ACT
This RIS is not a rule as defined in the Congressional Review Act (5 U.S.C. §§ 801-808).
PAPERWORK REDUCTION ACT STATEMENT
This RIS does not contain new or amended information collection requirements subject to the Paperwork Reduction Act of 1995 (44 U.S.C. 3501 et seq.). Existing information collection requirements were approved by the Office of Management and Budget (OMB), approval number 3150-0132.
Public Protection Notification
The NRC may not conduct or sponsor, and a person is not required to respond to, a request for information or an information collection requirement unless the requesting document displays a currently valid OMB control number.
CONTACT
Please direct any questions about this matter to the technical contact listed below.
/RA/
/RA/
Michael C. Cheok, Director
Mark D. Lombard, Director Division of Construction Inspection
Division of Spent Fuel Management and Operational Programs
Office of Nuclear Material Safety Office of New Reactors
and Safeguards
/RA A. Mohseni for/
Lawrence E. Kokajko, Director Division of Policy and Rulemaking Office of Nuclear Reactor Regulation
Technical Contact:
Gordon Bjorkman, NMSS/DSFM
(301) 415-7401
E-mail: Gordon.Bjorkman@nrc.gov
Note: A list of recently issued NRC regulatory issue summaries may be found on the NRC
public Web site, http://www.nrc.gov, under NRC Library/Document Collections.
- via e-mail
TAC MF3182 OFC
NMSS/DSFM/
SFLB
Tech Editor NMSS/DSFM/
SFLB
NMSS/DSFST/
CSTB: BC
NMSS/DUWP/
RDB:BC
NMSS/DSFM/
SFLB: BC*
NAME
JVera CHsu GBjorkman*
CAraguas*
BWatson*
MSampson DATE
07/17/2015
03/13/2014
07/07/2015
07/28/2015
07/28/2015
08/05/2015 OFC
OE*
NRR/PMDA*
OIS*
OGC*
DPR/PGCB:LA*
DPR/PGCB:PM
NAME
TMarenchin LHill TDonnell LLondon ELee (ABaxter for w/comments)
TKeene DATE
08/20/2015
08/13/2015
08/20/2015
09/17/2015
09/22/2015
09/21/2015 OFC
NRR/DPR/
PGCB:BC
NMSS/DSFM: D
NRO/DCIP: D
NRR/DPR: DD
NRR/DPR: D
NAME
SStuchell MLombard MCheok AMohseni (AMohseni for)
LKokakjo
DATE
09/22/2015
10/30/2015
10/29/2015
11/12/2015
11/12/2015