Chapter15 PDF

15.
TRANSIENT AND ACCIDENT ANALYSES

15.1 Introduction
In the AP1000 Design Control Document (DCD) Tier 2, Chapter 15, Accident Analyses, the
applicant discussed the analysis of various design-basis transients and accidents. The
applicant used the results of these analyses in the DCD to show the conformance of the
AP1000 advanced passive plant design with General Design Criterion (GDC) 10, Reactor
Design, for fuel design limits, GDC 15, Reactor Coolant System Design, for the reactor
coolant pressure boundary (RCPB) pressure limits, and the requirements of Title 10,
Section 50.46, of the Code of Federal Regulations (10 CFR 50.46) for the performance of the
emergency core cooling system (ECCS).
The staff of the U.S. Nuclear Regulatory Commission (NRC) has reviewed the AP1000
transient and accident analyses in the DCD, in accordance with Chapter 15, Accident
Analysis, of NUREG-0800, which defines the agencys Standard Review Plan (SRP).
15.1.1 Event Categorization
The applicant assigned the initiating events to the following categories, in accordance with
Chapter 15 of the SRP and Regulatory Guide (RG) 1.70, Standard Format and Content of
Safety Analysis Reports for Nuclear Plants:
C
C
C
C
C
C
C
increase in heat removal from the primary system

decrease in heat removal by the secondary system
decrease in reactor coolant system (RCS) flow rate
reactivity and power distribution anomalies
increase in reactor coolant inventory
decrease in reactor coolant inventory
anticipated transients without scram (ATWSs)
The first category, an increase in heat removal from the primary system, includes a new event
involving inadvertent operation of the passive residual heat removal (PRHR) heat exchanger
(HX). Because this category is broader than the category of increase in heat removal from the
primary system in the SRP and RG 1.70 and reflects the AP1000 design, it is acceptable.
The applicant also grouped the design-basis events according to their anticipated frequency of
occurrence, identified as Condition Inormal operation and operational transients,
Condition IIfaults of moderate frequency, Condition IIIinfrequent faults, and
Condition IVlimiting faults. The applicants event frequency grouping is consistent with the
guidelines of RG 1.70 and the criteria of American Nuclear Society (ANS) 18.2-1973, Nuclear
Safety Criteria for the Design of Stationary Pressurized Water Reactor Plants. Condition I
events occur frequently, and their effects on the consequences of Conditions II, III, and IV
events should be considered. Condition II events may occur during a calendar year for a
particular plant. Condition III events may occur infrequently during the life of a particular plant.
Condition IV events are postulated but not expected to occur during the life of a plant.
15-1
Transient and Accident Analyses

The SRP divides the events into anticipated operational occurrences (AOOs) and postulated
accidents. The requirements of 10 CFR Part 50, Appendix A, General Design Criteria for
Nuclear Power Plants, define AOOs as conditions of normal operation and those transients
that are expected to occur one or more times during the life of a plant; therefore, AOOs
encompass the normal, moderate-frequency and infrequent events of Conditions I through III.
Chapter 15 of the SRP does not specify a category of infrequent incidents but does provide
specific acceptance criteria for those events that cannot be categorized as infrequent. Thus,
the event frequency categorization in the DCD is consistent with the NRCs licensing approach.
In DCD Tier 2, Section 15.0.1, the applicant listed the design-basis events analyzed under
Conditions II, III, and IV. These events are generally consistent with the current licensing
practice. However, the DCD lists the complete loss of forced reactor coolant flow event and the
single rod cluster control assembly (RCCA) withdrawal event at full-power conditions as
Condition IIIinfrequent faults. The event categorization is inconsistent with SRP
Sections 15.3.1 and 15.4.3, which classify the complete loss of RCS flow and the withdrawal of
an RCCA as Condition IImoderate-frequency events, with the acceptance criterion that
specifies no violation against the safety limit of the departure from nucleate boiling ratio
(DNBR). Nonetheless, the applicant analyzed the complete loss of forced reactor coolant flow
event, as presented in DCD Tier 2, Section 15.3.2, to satisfy the acceptance criteria for a
Condition II event. Thus, the staff concludes that the applicants approach for the analysis of
the complete loss of reactor coolant flow event is acceptable.
Per WCAP-9272-A, Westinghouse Reload Safety Evaluation Methodology, Westinghouse
was previously allowed to classify the withdrawal of an RCCA event as a Condition III event for
reactors that it manufactured because of the events very low probability of occurrence.
Because the manufacturer of the existing pressurized-water reactors (PWRs) also designed the
AP1000 reactor, the applicants classification of the withdrawal of an RCCA event as a
Condition III event is consistent with the current licensing practices and, therefore, is
acceptable.
15.1.2 Non-Safety-Related Systems Assumed in the Analysis
For the design-basis analysis, only safety-related systems or components can be used to
mitigate the events. In Westinghouse letter ET-NRC-93-3804, dated January 22, 1993, the
applicant responded to the staffs request for additional information (RAI) 440.31 for the AP600
review. The applicant stated that non-safety-related systems or components are assumed to
be operational in the following situations:
(a)
when assumption of a non-safety-related system results in a more limiting transient
(b)
when a detectable and nonconsequential random, independent failure must occur in

order to disable the system
(c)
when non-safety-related components are used as backup protection
15-2

For the AP1000 design, Westinghouse applied, as indicated in DCD Tier 2, Chapter 15, the
same approach as in the AP600 design, indicated above in the specified situations for nonsafety-related systems or components to be used for the analyses of the design-basis events.
The assumption in Case (a) will result in a more limiting transient and, therefore, it is
acceptable.
For Case (b), the applicant assumed continued operation of the main feedwater control system
(MFCS) in the design-basis analysis of those events not related to feedwater system
malfunction, loss of alternating current (ac) power, or turbine trip. For example, an event
involving withdrawal of an RCCA is analyzed from an at-power condition. Before the initiating
fault causing the RCCA withdrawal, the MFCS should operate and maintain steam generator
(SG) inventory. If a failure exists in the MFCS, alarms in the control room or abnormal control
system performance should detect it before the start of the RCCA withdrawal event. The staff
concludes that the assumption of MFCS continued operation is acceptable because a failure in
the MFCS is not a consequence of the initiating event, and the probability of a random,
independent failure occurring in the MFCS within the timeframe of the initiating event is
extremely low.
For Case (c), as discussed in the response to RAI 440.061 and summarized in DCD Tier 2,
Table 15.0-8, the applicant credited the following non-safety-related components as backup
protection in the design-basis analysis for the AP1000 design:
C
the main feedwater pump trip in the analysis of an increased feedwater flow event
the pressurizer heater block in the analysis of loss of normal feedwater (LONF),
inadvertent operation of core makeup tanks (CMTs), chemical and volume control
system (CVCS) malfunction that increases reactor coolant inventory, steam generator
tube rupture (SGTR), and small-break loss-of-coolant accidents (SBLOCA)
main steam isolation valve (MSIV) backup valves (including the turbine stop, control
valves, turbine bypass valves, moisture separator reheat steam supply control valve,
and main steam branch isolation valves in the analysis of inadvertent opening of SG
safety valves, steamline break (SLB), and SGTR events)
During the course of the review, the staff asked the applicant to address its compliance with
10 CFR 50.36, which specifies the criteria for the systems that are subject to technical
specification (TS) limiting conditions for operation (LCOs). Specifically, 10 CFR
50.36(c)(2)(ii)(C) requires that a TS be established for a structure, system, or component (SSC)
that is assumed to function or actuate in a design-basis analysis for the mitigation of specified
events. In its response to RAI 440.061, the applicant indicated that it complied with the 10 CFR
50.36 requirements by providing TSs to include non-safety-related systems that are credited as
backup systems in the licensing design-basis analyses. Items 7 and 27 of AP1000 TS
Table 3.3.2-1, Engineering Safeguards Actuation System Instrumentation, include applicable
modes, surveillance requirements (SRs), and trip setpoints for the main feedwater pump trip
and pressurizer heater trip, respectively. Section 3.7.2 of the AP1000 TS provides the LCOs for
the main steam branch isolation valves and the MSIV backup valves. These TSs ensure the
15-3

reliability of the non-safety-related components credited as backup systems in the design-basis
analyses. Therefore, Section 3.7.2 of the AP1000 TS is acceptable.
Based on its review, the staff concludes that crediting these non-safety-related backup
protection systems and components in the design-basis analyses is acceptable for the following
reasons:
C
The trip mechanisms of the feedwater pump trip breakers and pressurizer heater trip
breakers are simple, and the likelihood of the breaker function failure is low.
The operating data show that the turbine stop and control valves are reliable, and taking
credit for the turbine valves in the design-basis analyses for backup protection is
consistent with the staff position stated in NUREG-0138, Staff Decision of Fifteen
Technical Issues Listed in Attachment to November 3, 1976, Memorandum from
Director, NRR to NRR Staff.
The applicant has included SRs and LCOs in the TSs to ensure the reliability of the
following systems or components:
feedwater pump trip breakers and redundant pressurizer heater trip breakers
MSIV backup valves and main steam branch isolation valves
15.1.3 Chapter 15 Loss of Offsite Power Assumptions

As indicated in DCD Tier 2, Section 15.0.14, the applicant performed the Chapter 15 analysis
assuming a loss of offsite power (LOOP). The LOOP is not considered as a single failure, and
the analysis is performed without changing the event category. The assumption of the LOOP in
the Chapter 15 analysis complies with the requirements of GDC 17, Electric Power Systems,
of 10 CFR Part 50, Appendix A, which requires the analysis of AOOs and postulated accidents
assuming a LOOP. In the analysis, a LOOP is considered a consequence of an event as a
result of disruption of the grid following a turbine trip during the event.
In the case of events involving a turbine trip, the applicant assumed that a LOOP and the
resulting coastdown of the reactor coolant pumps (RCPs) occurs 3 seconds after the turbine
trip. DCD Tier 2, Section 8.2, provides the basis for the 3-second delay. That section
describes the electrical design features of the AP1000, the electrical system response to a
turbine trip, and the combined license (COL) applicant interfaces that support the 3-second
assumption. Among others, the AP1000 design provisions include the following electrical
features that support the 3-second delay:
C
Use of an output generator circuit breaker and reverse power relay, with at least a
15-second delay before tripping the breaker following a turbine trip, allows the generator
to provide voltage support to the grid and maintain adequate voltage to the RCPs for
significantly longer than the assumed 3 seconds.
The COL applicant interface item in DCD Tier 2, Table 1.8-1, Item 8.3 (that transient
stability must be maintained and the RCP bus voltage must remain above the voltage
15-4

required to maintain the flow assumed in Chapter 15 analyses for a minimum of
3 seconds following a turbine trip) ensures that, for the applicants unique grid system
configuration, a grid instability condition following a turbine trip will take at least
3 seconds before it results in a loss of power to the RCPs.
C
The COL applicant interface item in DCD Tier 2, Table 1.8-1, Item 8.3, (that the
protective devices controlling the switchyard breakers are set with consideration for
preserving the plant grid connection following a turbine trip) is especially important in
generator output circuit breaker designs to ensure that the opening of the switchyard
breakers following a turbine trip does not interrupt the backfeed offsite circuit through
the generator main stepup transformer.
This design does not use automatic transfers of RCP buses, which precludes bus
transfer failures following a turbine trip.
If a turbine trip occurs when the grid is not connected to the plant, the main generator
will be available to power the RCPs for at least 3 seconds before the generator output
breaker is tripped on generator undervoltage or exciter overcurrent.
The staff has reviewed the information on the AP1000 electrical design, as well as the COL
requirements. On that basis, and as described above, the staff has reasonable assurance that
the RCPs can receive power for a minimum of 3 seconds following a turbine trip (discussed in
Section 8.2.3.4 of this report). The staff has also reviewed the DCD Tier 2, Chapter 15,
analysis and found that the applicant considered LOOP in all of the applicable analyzed events
and applied the acceptance criteria specified in the related SRP sections for events with and
without LOOP. Therefore, the staff concludes that the applicant's approach is acceptable.
15.1.4 Analytical Methods
The analytical methods used for transient and accident analyses are normally reviewed on a
generic basis. As indicated in DCD Tier 2, Sections 15.0.11, 15.6.5.4A, and 15.6.5.4B, the
methods used for transient and accident analyses include the following computer codes:
C
TWINKLEThis multidimensional spatial neutronics code uses an implicit

finite-difference method to solve the two-group transient neutronics equations in one,
two, and three dimensions. TWINKLE has been used to calculate the kinetic response
of a reactor for transients, such as the RCCA bank withdrawal from subcritical
conditions and RCCA ejection events, which cause a major perturbation in the spatial
neutron flux distribution. As documented in WCAP-7979-P-A (proprietary) and
WCAP-8028-NP-A (nonproprietary), TWINKLE - A Multi-Dimensional Neutron Kinetics
Computer Code, issued January 1975, the NRC has approved this code for operating
Westinghouse plants and the AP600. Since the AP1000 fuel design is similar to that of
operating Westinghouse plants and the AP600 (i.e., falls within the NRC-approved
applicable range of the code), the application of the TWINKLE code to the AP1000 for
analysis of kinetic responses is acceptable.
15-5

C
VIPRE-01As documented in WCAP-14565-P-A (proprietary), and WCAP-15306-NP-A

(nonproprietary), VIPRE-01 Modeling and Qualification for Pressurized Water Reactor
Non-LOCA Thermal-Hydraulic Safety Analysis, issued October 1999, the NRC has
approved this code for the core thermal-hydraulic (T-H) analyses, determining coolant
density, mass velocity, enthalpy, vapor void, static pressure, and the DNBR distribution
along parallel flow channels within the reactor core under normal operational and
transient conditions. Since the AP1000 core design is similar to that of operating
Westinghouse plants and the AP600 (i.e., falls within the NRC-approved applicable
range of the code), the application of the VIPRE-01 code to the AP1000 T-H
calculations is acceptable.
COASTAs documented in CENPD-98-A, Coast - Code Description, issued April

1973, the NRC has approved this code for use in calculating the reactor coolant flow
coast transient for any combination of active and inactive RCPs and forward and reverse
flow in the hot- or cold-legs. The NRC approved the code for the PWRs manufactured
by the former Combustion Engineering (CE) (now merged as part of Westinghouse).
The COAST code uses equations of conservation of momentum formulated for each of
the flowpaths of the COAST model, assuming unsteady-state, one-dimensional flow of
an incompressible fluid. The equation of conservation of mass is formulated for each
nodal point. Pressure losses resulting from friction and geometric losses are assumed
proportional to the square of the flow velocity. RCP dynamics are modeled using a
head-flow curve for a pump at full speed and using four-quadrant curves, which are
parametric diagrams of pump head and torque on coordinates of speed versus flow, for
a pump at other than full speed. The COAST code is a generic code for calculating the
coastdown flow with the values of pump head curves and the pressure drop coefficients
of the RCS components as the input parameters. Since the pump head curves and
pressure loss coefficients used in the COAST code reflect the AP1000 design, the staff
concludes that the use of COAST is acceptable for the AP1000 in calculating the RCP
flow during the RCP coastdown transient.
FACTRANAs documented in WCAP-7908-A (nonproprietary), FACTRAN - A

FORTRAN-IV code for Thermal Transients in a UO2 Fuel Rod, issued June 1972, the
NRC has approved the FACTRAN code for calculations of the transient heat flux at the
surface of a rod. Since the AP1000 fuel rod design is similar to that of operating
Westinghouse plants and the AP600 (i.e., falls within the NRC-approved applicable
range of the code), the application of FACTRAN to the AP1000 heat flux calculations is
acceptable.
LOFTRANAs documented in WCAP-7907-P-A (proprietary) and WCAP-7907-NP-A

(nonproprietary), LOFTRAN Code Description, issued April 1984, the NRC previously
approved this code to allow Westinghouse to analyze system responses to non-LOCA
events for conventional Westinghouse PWRs. LOFTRAN simulates a multiloop system
using a model containing a reactor vessel (RV), hot- and cold-leg piping, SGs, and a
pressurizer. The code also includes a point kinetics model, including reactivity effects of
the moderator, fuel, boron, and control rods. The secondary side of the SG uses a
homogeneous, saturated mixture for analyses of thermal transients and a water-level
correlation for indication and control. When the applicant applied the LOFTRAN code to
15-6

the AP600 safety analysis, it modified the code to incorporate features representative of
the AP600 design which are important to modeling the non-LOCA transient analyses.
WCAP-14234, LOFTRAN and LOFTTR2 AP600 Code Applicability Document, issued
June 1997, describes the LOFTRAN modifications. WCAP-14307, AP600
LOFTRAN-AP and LOFTTR2-AP Final Verification and Validation Report, issued
August 1997, documents the test data comparisons that support the LOFTRAN
modifications.
C
LOFTTR2As documented in WCAP-10698-P-A (proprietary) and WCAP-10750-NP-A

(nonproprietary), SGTR Analysis Methodology to Determine the Margin to Steam
Generator Overfill, issued August 1985, the NRC-approved code is used to analyze an
SGTR event for conventional Westinghouse PWRs. LOFTTR2 is a modified version of
LOFTRAN with a more realistic breakflow model, a two-region SG secondary side, and
an improved capability to simulate operator actions during an SGTR event. The version
of LOFTTR2 applied to the AP600 SGTR analysis incorporated the LOFTRAN changes
to simulate passive safety features for the AP600 design. WCAP-14234 documents
these changes.
NOTRUMPThis code consists of the modeling features that meet the requirements of
Appendix K to 10 CFR Part 50. As documented in WCAP-10079-P-A (proprietary),
NOTRUMPA Nodal Transient Small Break and General Network Code, issued
August 1985, and WCAP-10054-P-A (proprietary), Westinghouse Small-Break ECCS
Evaluation Model Using the NOTRUMP Code, issued August 1985, the NRC previously
approved the NOTRUMP code for the SBLOCA analysis. WCAP-14807, Revision 5,
NOTRUMP Final Validation Report for AP600, issued August 1998, documents the
modified version of the NOTRUMP code for the AP600 application.
WCOBRA/TRAC-LBLOCAAs documented in WCAP-12945-P, Revision 2, Code

Qualification Document for Best Estimate LOCA Analysis, Volumes 1 through 5, issued
March 1998, the NRC has approved this best estimate (BE) code in a safety evaluation
dated June 28, 1996, for large-break loss-of-coolant accident (LBLOCA) analysis.
WCAP-14171, Revision 2, WCOBRA/TRAC Applicability to AP600 Large-Break
Loss-of-Coolant Accident, issued March 1998, documents the modified version of the
WCOBRA/TRAC code for the AP600 application. Westinghouse letter
NSD-NRC-97-5171 dated June 10, 1997, documents its auxiliary code, HOTSPOT,
which is updated for the AP600.
WCOBRA/TRAC-LTC and WGOTHICWCOBRA/TRAC is also used for the

post-LOCA long-term cooling (LTC) analyses. WCAP-14776, Revision 4,
WCOBRA/TRAC, OSU Long-Term Cooling Final Validation Report, issued March
1998, documents the code verification for the LTC analyses. The WGOTHIC code,
documented in WCAP-14407, Revision 3, WGOTHIC Application to AP600, issued
April 1998, is used to calculate containment boundary conditions for LBLOCA and
post-LOCA LTC. The staff previously reviewed and accepted the application of
WCOBRA/TRAC and WGOTHIC for LTC calculations, as discussed in Chapter 21 of
this report.
15-7

In support of the AP1000 application, the applicant submitted WCAP-15644-P, AP1000 Code
Applicability Report, issued March 2004, for the staff to review. WCAP-15644-P documents
the applicants assessment of the safety analysis codes that were developed and approved for
the AP600 design certification to determine their applicability for use in the AP1000 design.
The safety analysis codes include LOFTRAN, LOFTTR2, NOTRUMP, WCOBRA/TRAC, and
WGOTHIC. The staff reviewed the report and concluded in a March 25, 2002, letter from
J.E. Lyons (NRC) to W.E. Cummins (Westinghouse), Applicability of AP600 Standard Plant
Design Analysis Codes, Test Program and Exemptions to the AP1000 Standard Plant Design,
that the modified LOFTRAN and LOFTTR2 codes are acceptable for use in the AP1000
analysis with the following conditions and limitations:
C
Table 2 of the enclosure to the March 25, 2002, letter listed the transients and accidents
that Westinghouse proposed to analyze with the LOFTRAN code. The staff limited its
review of LOFTRAN usage by Westinghouse to this set. The use of the code for other
analytical purposes will require additional justification.
In the preapplication review, the staff requested that Westinghouse perform main
steamline break (MSLB) analyses for the AP1000 standard plant design. In particular,
the staff wanted to assess the ability of the code to model the resulting steam formation
in the reactor coolant loops. The applicant has provided this analysis. Chapter 21 of
this report includes a review of this material.
In addressing the staffs review question regarding compliance with the limitations imposed by
the staff on the use of the LOFTRAN and LOFTTR2 codes, the applicant provided its response
to RAI 440.054 and indicated that the codes are used only for those events identified in the
NRC letter of March 25, 2002. The applicant has submitted the MSLB analysis in DCD Tier 2,
Section 15.1.5. The analysis results demonstrate that voiding in the reactor coolant loops does
not occur and, therefore, is not a concern for the MSLB event. Because the application of
LOFTRAN and LOFTTR2 in the safety analysis for the AP1000 has complied with the
limitations imposed by the NRC staff, the staff concludes that the application is acceptable.
The applicant also provided the assessment addressing the applicability of NOTRUMP,
WCOBRA/TRAC, and WGOTHIC to the safety analysis for the AP1000 design. The staff has
reviewed the applicants compliance assessment and documented its evaluation in
Section 21.6.2 of this report for NOTRUMP, Sections 21.6.3 and 21.6.4 of this report for
WCOBRA/TRAC, and Section 21.6.5 of this report for WGOTHIC.
15.1.5 Steam Generator Middeck Plate Induced Level Measurement Uncertainty
Westinghouse has issued three Nuclear Service Advisory Letters (NSALs), NSAL-02-3 and
Revision 1, NSAL-02-4, and NSAL-02-5, which document the concerns with the Westinghousedesigned SG water-level setpoint uncertainties. NSAL-02-3 and its revision, dated February 15
and April 8, 2002, respectively, deal with the uncertainties in the SG water-level measurement
caused by the placement of the middeck plate between the upper and lower taps. These
uncertainties affect the low-low level trip setpoint (used in the analysis for events such as the
feedwater line break (FLB), ATWS, and SLB). NSAL-02-4, dated February 19, 2002, deals with
the uncertainties in the measurement created because the calculation does not reflect the void
15-8

content of the two-phase mixture above the middeck plate. These uncertainties affect the highhigh level trip setpoint. NSAL-02-5, dated February 19, 2002, deals with potential inaccuracies
in the initial conditions assumed in safety analyses affected by SG water level. The safety
analyses may not be bounding because the velocity head under some conditions may increase
the uncertainties in the SG water-level control system. The staff requested, via RAI 440.062,
the applicant to discuss (1) the AP1000 design that accounts for all the uncertainties
documented in these advisory letters in determining the SG water-level setpoints, and (2) the
effects of the water-level uncertainties on the analyses of the LOCA and non-LOCA transients
and the ATWS event.
The applicants response to RAI 440.062 stated that measurement uncertainties for the reactor
protective system and engineered safety feature (ESF) actuation system instrumentation can
be determined only when actual instrumentation is selected for the plant. The plant-specific
setpoint calculations will be completed and reviewed as part of the COL. The COL applicants
referencing the AP1000 certified design will provide a calculation of setpoints for protective
functions consistent with the methodology discussed in WCAP-14605, Westinghouse Setpoint
Methodology for Protective SystemsAP600, issued April 1996. The methodology can be
used for performing setpoint studies independent of the hardware used for the protection
system, and, therefore applies to the AP1000 design. The setpoint study will include applicable
uncertainties discussed in the referenced NSALs. Using the methodology in WCAP-14605,
plant nominal setpoints are calculated by adding the channel allowance from the setpoint study
to the setpoints used in the safety analysis.
The COL applicant should evaluate and confirm the validity of the safety analysis documented
in the DCD using plant-specific setpoints and instrument uncertainties, including the SG
middeck-level measurement uncertainty. The COL applicants should submit in the plantspecific applications the setpoint analysis and the associated safety analysis for the staff to
review and approve. This item was designated as draft safety evaluation report (DSER) Open
Item 15.1.5-1. In addressing DSER Open Item 15.1.5-1, Westinghouse provided its response
in a letter from M.M. Corletti (Westinghouse) to the NRC dated July 1, 2003, Transmittal of
Westinghouse Responses to Open Items Identified in the AP1000 Draft Safety Evaluation
Report. In the response, Westinghouse indicated that Item E of Revision 1 to RAI 440.002
addressed the COL information in DCD Tier 2, Section 7.1.6, related to instrumentation setpoint
uncertainty calculations by the COL applicant upon selection of the installed plant
instrumentation. In accordance with the current practice, the safety analyses will become plant
specific when the COL applicant performs the setpoint study to develop plant-specific TS
setpoints based on the safety analysis values, adding instrumentation uncertainty, as well as
uncertainties related to the effects of the SG middeck plate on SG-level measurement. The TS
setpoints are changed to reflect instrumentation uncertainties. Westinghouse stated, and the
staff agreed, that the plant safety analyses are not changed because the revised TS setpoints
are based on the safety analyses documented in the DCD. The COL actions to establish
appropriate plant-specific setpoints, as discussed in DCD Tier 2, Section 7.1.6, add the
appropriate plant-specific values to the TS setpoints used in the safety analyses. Therefore,
the staff concludes that no other COL action is needed to reperform any safety analyses, and
DSER Open Item 15.1.5-1 is resolved. This is COL Action Item 15.1.5-1.
15-9
15.2 Transient and Accident Analysis

The applicant presents the results of transient and accident analysis for the AP1000 design in
DCD Tier 2, Chapter 15. This section discusses the staffs evaluation of results of the analysis
and the applicants responses to the staffs RAIs. Section 15.3 of this report presents the staffs
evaluation of the analysis for radiological releases.
15.2.1 Increase in Heat Removal from the Primary System
In DCD Tier 2, Section 15.1, the applicant presented the results of its analysis of the events
involving an increase in heat removal from the primary system. The events include
(1) feedwater system malfunctions causing a reduction in feedwater temperature, (2) feedwater
system malfunctions causing an increase in feedwater flow, (3) excess increase in secondary
steam flow, (4) inadvertent opening of an SG relief or safety valve, (5) SLB, and (6) inadvertent
operation of the PRHR HX. The staff provides its evaluation of the analytical results in the
following sections.
15.2.1.1 Decrease in Feedwater Temperature (DCD Tier 2, Section 15.1.1)
Failure of a low- or high-pressure heater train may cause a decrease in feedwater temperature,
a moderate-frequency event. A reduction in feedwater temperature decreases reactor coolant
temperature, which, in turn, causes an increase in core power because of the effects of the
negative moderator coefficient of reactivity. Because the rate of energy change is reduced as
load and feedwater flows decrease, the transient initiated from zero-power conditions is less
severe than the full-power case. The applicants analysis for the limiting case is based on initial
full-power conditions with a decrease of feedwater temperature caused by the loss of one string
of low-pressure feedwater heaters. The loss of a string of feedwater heaters results in a
maximum reduction in feedwater temperature of 26.4 EC (79.5 EF). The applicants analysis
indicates that the decrease in feedwater temperature results in an increase in core power of
less than 10 percent of full power. The decrease in feedwater temperature event is bounded by
an excessive increase in secondary steam flow (a moderate-frequency event), which results in
a power increase of 12 percent. Section 15.2.1.3 of this report discusses the staffs review of
the event with an excessive increase in secondary steam flow.
15.2.1.2 Increase in Feedwater Flow (DCD Tier 2, Section 15.1.2)
Increase in feedwater flow events may be caused by system malfunctions or operator actions
that result in an inadvertent opening of a feedwater control valve. The excessive feedwater flow
reduces reactor coolant temperature, which, in turn, causes a power increase because of the
effects of the negative moderator coefficient of reactivity. The SG high-2 water-level signal trip
prevents the continuous addition of excessive feedwater by closing the feedwater isolation
valves and feedwater control valves and trips the turbine, main feedwater pumps, and reactor.
The applicant uses three codes to perform the analysis for this event. LOFTRAN calculates the
nuclear power transient, the RCS flow coastdown, and the primary pressure and temperature
transient. FACTRAN calculates the heat flux based on the nuclear power and flow from
15-10

LOFTRAN. VIPRE-01 calculates the departure from DNBR during the transient, using the heat
flux from FACTRAN and the flow from LOFTRAN.
The applicant analyzed both the no-load case and the full-power case. For the no-load
condition, the applicant assumed a feedwater control valve malfunction resulting in a step
increase to 120 percent of nominal feedwater flow to one SG. The applicant assumed a
feedwater temperature at the low value of 4.4 EC (40 EF). With the plant at no-load conditions,
the turbine is not connected to the grid. Any subsequent reactor or turbine trip will not disrupt
the grid and produce a consequential LOOP. Therefore, the applicant did not assume a LOOP
in the no-load case. The results of the analysis show that the no-load case is bounded by an
uncontrolled RCCA bank withdrawal from a subcritical or low-power startup condition, because
of a lower reactivity insertion rate than the uncontrolled RCCA bank withdrawal event stemming
from the effects of the negative moderator coefficient of reactivity. Section 15.2.4.1 of this
report provides the staffs review and approval of the analysis of an uncontrolled RCCA bank
withdrawal event.
The applicants analysis for the limiting case is based on initial full-power conditions with an
increase of feedwater flow caused by malfunction of one feedwater control valve to its
maximum capacity, resulting in a step increase to 120 percent of nominal feedwater flow to one
SG. An SG high-2 level trip signal actuates a reactor trip and an associated turbine trip. In
addressing the issue of a LOOP, the applicant assumed that a LOOP and the resulting
coastdown of the RCPs occur 3 seconds after the turbine trip. As discussed in Section 15.1.3
of this report, the assumption of a LOOP with a delay time of 3 seconds is acceptable.
The applicant has considered plant systems and equipment, discussed in DCD Tier 2,
Section 15.0.8, that are available to mitigate the effects of the event. From the viewpoint of an
SG overfilling, the worst case is a failure of the feedwater control valve in the affected SG to
close, in combination with a single failure of the feedwater isolation valve to close. In this case,
the applicant indicated in its response to RAI 440.063 that an SG high-2 level trip signal will trip
feedwater pumps and terminate the excessive feedwater flow. The staff notes that the
feedwater pumps trip is a non-safety-related system. The staff has reviewed and approved the
use of the feedwater pumps trip to terminate the excessive feedwater flow, as discussed in
Section 15.1.2 of this report.
The applicant performed the analysis using an acceptable method, and the results of the
analysis demonstrate that the limiting full-power case meets the acceptance criteria for this
moderate-frequency event. Specifically, the calculated peak RCS pressure falls below
110 percent of the RCS design pressure, and the calculated DNBRs for the transient remain
above the safety limit DNBR defined in DCD Tier 2, Section 4.4. As a result, the analysis
satisfies the acceptable criteria defined in SRP Section 15.1.2. Therefore, the staff concludes
that the analysis is acceptable.
15.2.1.3 Excessive Increase in Secondary Steam Flow (DCD Tier 2, Section 15.1.3)
An operator action or an equipment malfunction in the steam dump control or turbine speed
control may cause an excessive increase in secondary steam flow. A rapid increase in steam
flow results in a power mismatch between the reactor core power and the SG load demand.
15-11

The applicant analyzed four cases involving a 10-percent step load increase from the rated
load, using the previously approved LOFTRAN, FACTRAN, and VIPRE-01 codes and assuming
the following for each case:
Case 1minimum moderator feedback and manual reactor control

Case 2maximum moderator feedback and manual reactor control
Case 3minimum moderator feedback and automatic reactor control
Case 4maximum moderator feedback and automatic reactor control
The 10-percent step load increase is the highest load increase allowed in the range of 25 to
100 percent of full power. Each case is analyzed without taking credit for pressurizer heaters.
At the initiation of the event, the RCS pressure and temperature are assumed at their full-power
values for the DNBR calculation. The safety DNBR limit, as described in WCAP-11397-P-A,
Revised Thermal Design Procedure, issued April 1989, includes uncertainties in initial
conditions. In DCD Tier 2, Sections 15.1.4 and 15.1.5 analyze steam flow increases greater
than 10 percent, and Sections 15.2.1.4 and 15.2.1.5 of this report evaluate them.
In demonstrating the capability of the plant for the cases with automatic rod control, the
applicant took no credit for delta T trips on overpower and overtemperature. The applicant has
considered plant systems and equipment, discussed in DCD Tier 2, Section 15.0.8, that are
available to mitigate the effects of the event, and it determined that no single active failure in
these systems or equipment will adversely affect the consequences of the event. In
considering the effects of a LOOP, the applicant assumed a reactor trip with a coincident
turbine trip followed by a LOOP 3 seconds later. The LOOP primarily causes the RCPs to
coast down. Since the LOOP is delayed for 3 seconds after the turbine trip, the RCCAs are
inserted well into the core before the RCS flow coastdown begins. The resulting power
reduction compensates for the reduced flow encountered once the RCPs lose power.
Therefore, the applicants analysis indicates that the minimum DNBRs predicted during the
event will occur before flow coastdown begins.
The results of the analysis show that the calculated peak RCS pressure is less than
110 percent of the design pressure, and the calculated minimum DNBR does not violate the
safety DNBR limits. Because the analysis uses acceptable methods and the results meet the
acceptance criteria of SRP Section 15.1.3 for this moderate-frequency event, the staff
concludes that the analysis is acceptable.
15.2.1.4 Inadvertent Opening of an SG Relief or Safety Valve (DCD Tier 2, Section 15.1.4)
An inadvertent opening of an SG relief, safety, or steam dump valve may result in an increase
in steam flow. In the presence of a negative moderator temperature coefficient (MTC), the
excessive cooldown increases positive reactivity, which, in turn, increases the core power level.
In assessing the effects of the negative MTC, the applicants analysis assumes the most
negative MTC corresponding to the end-of-life rodded core with the most reactive RCCA in its
fully withdrawn position. Availability of offsite power is assumed to maximize the cooldown
effect. Because the initial SG water inventory for the no-load case is greater, the magnitude
and duration of the RCS cooldown resulting from steam releases is greater, and the associated
15-12

positive reactivity addition is, therefore, also greater. Consequently, the applicant has
determined that zero-power conditions are more limiting than at-power conditions for this
postulated event. Because the turbine is initially in the trip condition for the plant at zero power,
the consequential LOOP following the turbine trip is not considered a credible event and,
therefore, is not modeled in the analysis.
Section 15.0.8, that are available to mitigate the effects of the event. It identified the limiting
single failure as a failure of one CMT discharge valve to open. The applicant also made the
following assumptions to maximize the cooldown effects:
A typical capacity steam flow rate of 236 kilograms per second (kg/s) at
8.2 MegaPascals (MPa) (520 pounds mass per second (lbm/s) at 1200 pounds per
square inch absolute (psia)) for any single steam dump, relief, or safety valve is
assumed as the initial steam flow.
The Moody model, without consideration of the piping friction losses, is used to calculate
the steam flow.
The most reactive RCCA is assumed to be stuck out of the core after the reactor trip.
The lowest startup feedwater temperature is assumed.
Four RCPs are assumed to be operating initially.
No moisture is assumed in the blowdown steam.
Manual actuation of the PRHR system is assumed at the initiation of the event.
The applicant used the LOFTRAN code to analyze the event. During the transient, the low
cold-leg temperature S signal automatically actuates the CMT injection and the associated
tripping of the RCPs. Boron solution at 3400 parts per million (ppm) enters the RCS, providing
negative reactivity to prevent a significant return to power and core damage. Later in the
transient, as the reactor pressure continues to decrease, accumulators actuate and inject boron
solution at 2600 ppm.
The results of the analysis show that the RCS pressure remains below 110 percent of the
design pressure, and the departure from nucleate boiling (DNB) does not occur, thereby
satisfying the acceptance criteria in SRP Section 15.1.4. Therefore, the staff concludes that the
analysis is acceptable.
15.2.1.5 Steam System Piping Failure (DCD Tier 2, Section 15.1.5)
An SLB, a limiting-fault event, is defined as a pipe break in the main steam system. The steam
release during an SLB causes a decrease in the RCS temperature and SG pressure. In the
presence of a negative MTC, the RCS temperature decrease results in an addition of positive
reactivity, which increases the core power level. The SG pressure decrease initiates a reactor
15-13

trip when low pressure in the steam system produces a safeguards S signal. The S signal
initiates the actuation of the CMTs, which, in turn, initiates a trip of the RCPs. In addition, the
S signal isolates all feedwater control and isolation valves and trips the main feedwater
pumps. The low cold-leg temperature signal isolates the startup feedwater control and isolation
valves. Ultimately, the borated water from the CMTs shuts down the reactor.
The applicant used the LOFTRAN code to calculate the system transient and the VIPRE-01
code to determine whether DNB had occurred for the core transient conditions calculated by the
LOFTRAN code. The applicant analyzed a double-ended rupture at no-load conditions with no
decay heat as the limiting case. Because the SGs have integral flow restrictors with a
0.13 square meter (m)2 (1.4 square feet (ft)2) throat area, any rupture with a break greater than
0.13 m2 (1.4 ft2), regardless of location, will have the same effect on the system as a 0.13 m2
(1.4 ft2) break; therefore, the analysis assumes this limiting break area.
Because the average coolant temperature for a core tripped from at-power conditions is higher
than at no-load conditions, and energy is stored in the fuel, the RCS for a core tripped from
at-power conditions contains more stored energy than at no-load conditions. The additional
stored energy reduces the cooldown caused by the SLB. Therefore, no-load conditions are
more limiting than at-power conditions. To represent the limiting initial conditions and maximize
the cooldown effect, the applicant assumed an initial condition for the SLB analysis of zero
power with no stored energy in the fuel.
Section 15.0.8, that are available to mitigate the effects of the event. For an SLB in which a
single failure results in the failure of the MSIV in the intact SG to close, the applicant took credit
for closing the non-safety-related MSIV backup valves (including the turbine-isolation and
control valves) to avoid an uncontrolled blowdown from two SGs. The use of the MSIV backup
valves in the SLB analysis for backup protection is acceptable, as discussed in Section 15.1.2
of this report. In addition, in order to maximize the overcooling effect, the applicant made the
following assumptions:
The most reactive RCCA is in the fully withdrawn position after reactor trip.
The end-of-life shutdown margin at zero power is assumed when the accident is
initiated.
A negative moderator coefficient is assumed for the end-of-life rodded core with the
most reactive RCCA stuck out.
The Moody model without consideration of the piping friction losses, maximizing the
blowdown flow rate, is used to calculate the steam flow.
The maximum cold startup feedwater flow, plus nominal 100-percent main feedwater, is
assumed.
Four RCPs are assumed to be operating initially.
15-14
No moisture is assumed in the blowdown steam.
Manual actuation of the PRHR system is assumed at the initiation of the event to
maximize the cooldown.
Availability of offsite power is assumed to maximize the cooldown effect. The results of an SLB
with offsite power available bound the case with a LOOP for the following reasons:
The initial condition of a LOOP results in an immediate RCP coastdown, which reduces
the RCS cooldown effect and the magnitude of the return-to-power by reducing
primary-to-secondary heat transfer.
During the SLB event, actuation of the CMTs will provide borated water that injects into
the RCS. Flow from the CMTs increases if the RCPs have coasted down. Therefore,
the analysis performed with offsite power and continued RCP operation reduces the rate
of boron injection into the core, which increases the potential for the core to return to
criticality after reactor trip.
The plant protection system automatically provides a safety-related signal that initiates
the coastdown of the RCPs coincident with CMT actuation. Because this RCP
coastdown initiates early during the SLB event, the difference is insignificant in
predicting the DNBRs for cases with and without offsite power.
Because of the passive nature of the safety injection system, the LOOP will not delay
the actuation of the safety injection system.
During the event, the reactor protection system initiates a trip of the RCPs in conjunction with
actuation of the CMTs. The MSIVs fully close in less than 10 seconds from receipt of a closure
signal.
In response to RAI 440.067, which addresses the staffs concern regarding the effect of the
timing of a LOOP on the analysis of the limiting SLB case, the applicant analyzed two full-break
SLB cases initiated with the reactor at no-load conditions, one with offsite power available
throughout the event, and one with offsite power loss simultaneous with the SLB at the start of
the event. The SLB analysis shows that for the case with LOOP, the RCPs begin coasting
down at the initiation of the transient, and, for the case with offsite power available, the
protection system automatically trips the RCPs at 7.4 seconds into the transient. The results of
the analysis show that the small difference in timing of the initiation of the RCP coastdown has
no significant impact on the parameters that affect the return to power. The calculated peak
core heat flux for the case with offsite power available was slightly greater than that for the
LOOP case (3.17 percent versus 3.14 percent of the nominal full-power value). Consistent with
the results presented in the DCD, this SLB analysis confirms that the SLB event initiated from
the no-load conditions with offsite power available bounds the case with a LOOP initiated at
time zero, and the event is a limiting case.
15-15

The staff concludes that the analysis for postulated SLBs is acceptable for the following
reasons:
The applicant has used the LOFTRAN code for the system response determination and
the VIPRE-01 code for the DNBR calculations in the analysis for an SLB event.
Throughout the event, the RCS temperature remains below saturated temperatures,
confirming that the SLB analysis falls within the applicable range of the LOFTRAN code
(also discussed in Section 15.1.4 of this report).
The values used for input parameters, resulting in a maximum cooldown effect and the
greatest potential for fuel failure, are conservative.
The results of the SLB analysis have shown that the minimum DNBR remains above the
allowable safety limit DNBR, and the peak RCS pressure remains below 110 percent of
the design pressure, thus satisfying the acceptance criteria of SRP Section 15.1.5 for an
SLB analysis.
15.2.1.6 Inadvertent Operation of the PRHR (DCD Tier 2, Section 15.1.6)

The inadvertent actuation of the PRHR system may be caused by operator action or a false
actuation signal that opens the valves that normally isolate the PRHR HX from the RCS. This
moderate-frequency event causes an injection of relatively cold water into the RCS and results
in the addition of positive reactivity in the presence of a negative MTC.
The applicant considered plant initial conditions at both full power and zero power. A
comparative assessment shows that the analysis performed for the inadvertent opening of an
SG relief- or safety-valve event (discussed in Section 15.2.1.4 of this report) bounds the zeropower condition. This occurs because the latter event, a moderate-frequency event, is
analyzed assuming PRHR HX actuation coincident with SG depressurization. Therefore, the
applicants analysis for the limiting case is based on initial full-power conditions.
For this analysis, the applicant used LOFTRAN for the system response calculation, FACTRAN
for the heat flux determination, and VIPRE-01 for the DNBR calculation and assumed a
negative moderator coefficient for the end-of-life rodded core. The applicant generated the
core properties used in the LOFTRAN code for reactivity feedback calculations by combining
those in the sector with cold coolant nearest to the loop with the PRHR system with those
associated with the remaining sector. Control systems are assumed to function only when their
operation results in more severe conditions. The analysis considered cases both with and
without automatic rod control. The reactor trips on high neutron flux, and the analysis does not
credit overtemperature and overpower delta T trips.
Section 15.0.8, that are available to mitigate the effects of the event, and it determined that no
single active failure in these systems or equipment will adversely affect the consequences of
the event.
15-16

In considering the effects of a LOOP, the applicant assumed that a reactor trip and an
associated turbine trip occur at the time of peak power. A loss of power is assumed to occur
3 seconds after the turbine trip. Since the LOOP is delayed for 3 seconds after the turbine trip,
the RCCAs are inserted well into the core before the RCS flow coastdown begins. The
resulting power reduction compensates for the reduced flow encountered once the RCPs lose
power. The applicants analysis indicates that the minimum DNBRs predicted during the event
occur before flow coastdown begins. With the assumption of no reactor trip occurring during
the transient, the results show that for the limiting case (the full-power case with manual rod
control), the core power stabilizes at about 108 percent of its nominal value.
The staff finds that the assumptions used in the analysis are conservative for the reasons
stated above and, therefore, are acceptable. The results of the analysis for the limiting
full-power case with and without offsite power available show that the RCS pressure remains
below 110 percent of the design pressure, and the minimum DNBR remains above the safety
limit DNBR, thus satisfying the acceptance criteria of the SRP for moderate-frequency events.
Therefore, the staff concludes that the analysis is acceptable.
15.2.2 Decrease in Heat Removal by the Secondary System (DCD Tier 2, Section 15.2)
The applicant has analyzed transients specified in SRP Section 15.2 for cases resulting from a
decrease in heat removal by the secondary system and identified the limiting cases with regard
to the capability of the RCS boundary and fuel rod cladding to withstand the consequences of
transients. The transients include (1) steam pressure-regulator malfunction or failure resulting
in decreasing steam flow, (2) loss of external electrical load, (3) turbine trip, (4) inadvertent
closure of MSIVs, (5) loss of condenser vacuum and other events resulting in turbine trip,
(6) loss of ac power to the station auxiliaries, (7) LONF flow, and (8) feedwater system pipe
break. The staff has reviewed the applicants analyses, as discussed in Sections 15.2.2.1
through 15.2.2.8 of this report.
15.2.2.1 Steam Pressure Regulator Malfunction or Failure that Results in Decreasing Steam
Flow (DCD Tier 2, Section 15.2.1)
The AP1000 design includes no SG pressure regulators whose failure will cause a decreasing
steam flow transient. Therefore, this event does not apply to the AP1000 design.
15.2.2.2 Loss of External Electrical Load (DCD Tier 2, Section 15.2.2)
Electrical system failures may cause the loss of external electrical load, a moderate-frequency
event. Following the loss of generator load, an immediate fast closure of the turbine control
valves will occur. A decrease in heat transfer capacity from primary to secondary causes the
transient in primary pressure, temperature, and water volume because of a rapid decrease of
steam flow to the turbine, accompanied by an automatic reduction of feedwater. Reactor trips
on high pressurizer pressure, high pressurizer water-level, and overtemperature delta T signals
protect the reactor. The pressurizer and the SG safety valves may lift to protect the RCS from
overpressurization.
15-17

The turbine trip event bounds this loss of external electrical load event, because the turbine
control valves close more slowly than the turbine stop valves close as a result of a turbine trip
event. The smaller reduction in heat removal from a slower termination of steam flow will result
in a lower peak RCS pressure. Section 15.2.2.3 below discusses the staffs evaluation of the
turbine trip analyses.
15.2.2.3 Turbine Trip (DCD Tier 2, Section 15.2.3)
Signals, including generator trip, low condenser vacuum, loss of lubricating oil, turbine thrust
bearing failure, turbine overspeed, manual trip, and reactor trip, may initiate the turbine trip
event. Following a turbine trip, the turbine stop valves rapidly close, and steam flow to the
turbine abruptly stops. The loss of steam flow results in a rapid increase in secondary system
pressure and temperature, as well as a reduction of the heat transfer rate in the SGs, which, in
turn, causes the RCS pressure and temperature to rise.
The applicant performed the analysis for this event using LOFTRAN for the transient response
calculation, FACTRAN for the heat flux calculation, and VIPRE-01 for the DNBR calculation. In
the DNBR determination, initial core power, RCS pressure, and RCS temperature are assumed
to be at their nominal values consistent with steady-state, full-power operation. The DNBR limit
includes uncertainties in initial conditions as described in WCAP-11397-P-A. In maximizing the
RCS overpressurization effects, the turbine is assumed to trip without actuating the rapid power
reduction system. This assumption delays the reactor trip until conditions in the RCS result in a
trip actuated by other signals. The reactor is assumed to trip by the first reactor trip setpoint
reached on high pressurizer pressure, overtemperature delta T, high pressurizer water-level, or
low SG water-level trip signals. In addition, the analysis takes no credit for the turbine bypass
system. Main feedwater is terminated at the time of turbine trip, with no credit taken for startup
feedwater or the PRHR system to mitigate the consequences of the event. The availability of
the pressurizer safety valves is assumed to reduce the pressure increase during the transient.
In considering the effects of a LOOP, the applicant has assumed that offsite power will last for
3 seconds after the turbine trip. The applicant also has considered plant systems and
equipment, as discussed in DCD Tier 2, Section 15.0.8, that are available to mitigate the effects
of the event and it determined that no single active failure in these systems or equipment would
adversely affect the consequences of the event.
In analyzing the turbine trip event, the applicant considered both minimum and maximum
reactivity feedback cases. The applicant also considered the event with and without credit for
the effect of pressurizer spray in reducing the reactor coolant pressure. The applicant analyzed
each case with and without offsite power available. The results of the applicants analysis show
that the most limiting case analyzed is a turbine trip from full power with minimum moderator
feedback. The limiting case assumes no available offsite power and takes no credit for the
effect of pressurizer spray in reducing the RCS pressure.
The staff finds that (1) the analysis used computer codes previously approved by the NRC and
adequate assumptions to maximize the peak pressure, (2) the calculated peak RCS pressure
for the limiting turbine trip case falls below 110 percent of the RCS design pressure,
(3) pressurizer overfilling does not occur, and (4) the calculated minimum DNBR is within the
15-18

safety DNBR limit, thus satisfying the acceptance criteria of SRP Section 15.2.3. Therefore, the
staff concludes that the analysis is acceptable.
15.2.2.4 Inadvertent Closure of Main Steam Isolation Valves (DCD Tier 2, Section 15.2.4)
The inadvertent closure of steam isolation valves results in a turbine trip. The consequences of
this event are the same as those of the turbine trip event discussed in Section 15.2.2.3 of this
report.
15.2.2.5 Loss of Condenser Vacuum (DCD Tier 2, Section 15.2.5)
Loss of the condenser vacuum may result in a turbine trip and prevent steam from dumping to
the condenser. Because the applicant assumes that the steam dump is unavailable in the
turbine trip analysis, no additional adverse effects will result for the turbine trip event caused by
the loss of the condenser vacuum. Therefore, the analytical results reviewed and discussed in
Section 15.2.2.3 of this report for the turbine trip event also apply to the loss of condenser
vacuum event.
15.2.2.6 Loss of AC Power to the Plant Auxiliaries (DCD Tier 2, Section 15.2.6)
A complete loss of the offsite grid, accompanied by a turbine-generator trip, may cause the loss
of ac power, a moderate-frequency event. In terms of the removal of decay heat, this event is
more severe than the turbine trip event because, for this event, an RCS flow coastdown
accompanies the decrease in heat removal by the secondary system, which further reduces the
capacity of the primary coolant to remove heat from the core. The reactor will trip upon
reaching one of the reactor trip setpoints in the primary and secondary systems as a result of
the flow coastdown and decrease in secondary heat removal, or as a result of the loss of power
to the control rod drive mechanisms.
The applicant used the LOFTRAN code to perform the RCS system response analysis following
a plant LOOP. The analysis credits only safety-related systems to mitigate the consequences
of the event. In the system response analysis, the initial reactor power is assumed to be
102 percent of the rated power level. The ANSI 5.1-1979, Decay Heat Power in Light-Water
Reactors, decay heat data represent the core residual heat generation rate. A LOOP is
assumed to occur at the time of the reactor trip, which is actuated on a trip signal of Low
narrow-range SG level. The assumption of a LOOP coincident with the reactor trip is more
conservative than the case with the offsite power loss at time zero because of the lower SG
water inventory for heat removal at the time of the reactor trip.
In addition, the PRHR HX heat transfer coefficients are assumed to be at low values associated
with the low flow rate caused by the RCP trip. In RAI 440.074, the staff requested the applicant
to justify the adequacy of the calculated PRHR heat transfer coefficients during the loss of ac
power event. The applicants response to RAI 440.074 stated that the determination of the
PRHR heat transfer coefficients was based on the same methods discussed in Chapter 9 of
WCAP-12980, Revision 3, AP600 Passive Heat Removal Heat Exchanger Test Final Report,
issued April 1997, which the NRC previously reviewed and approved, in a letter dated
March 25, 2002, from J.E. Lyons (NRC) to W.E. Cummins (Westinghouse), for application to
15-19

the AP1000 as part of the AP1000 preapplication review. In WCAP-12980, the results of
calculations using the Dittus-Boelter correlation show that the predicted values reasonably
agree with the PRHR test data. The operating conditions for the PRHR tests include 1.1 Lpm
to 37.8 Lpm (0.3 gpm to 10.0 gpm) for the single HX tube flow, 121 EC to 343 EC (250 EF to
650 EF) for the primary temperature, and 0.446 MPa to 16.0 MPa (50 psig to 2300 psig) for the
primary water pressure. The applicant calculated the PRHR tube flow and primary temperature
during the loss of ac power event and showed that the flows and temperatures fall within the
range of the PRHR test conditions. During the loss of ac power transient, the primary pressure
increased from 15.5 MPa to 17.2 MPa (2250 psia to 2500 psia), which slightly exceeds the test
range.
Because the primary-side fluid remains in single phase during a loss of ac power event, the
applicant indicated that the impact of pressure on the primary heat transfer coefficient is much
less significant than that of temperature. In addition, the applicant performed an analysis to
address the effects of measurement uncertainties of the PRHR heat transfer coefficient on the
plant behavior. The analysis reduced the primary-side heat transfer coefficient, calculated
using the Dittus-Boelter correlation, by 25 percent. The analysis shows that reducing the PRHR
primary-side heat transfer coefficient by 25 percent results in a small reduction in the overall
PRHR heat transfer rate, and that the reduction in heat transfer delays the time of the
calculated peak RCS pressure values but does not significantly affect the magnitude of the
calculated peak RCS pressure or peak pressurizer water volume. Because the applicants
analysis shows that the calculated PRHR tube flow and primary temperature fall within the
PRHR test range, and the magnitude of the PRHR primary-side heat transfer coefficient during
the low-flow conditions of the loss of ac power events does not significantly affect the calculated
decay heat removal, peak pressure, and pressurizer water volume, the staff concludes that the
PRHR heat transfer coefficient calculated at low-flow conditions is adequate and acceptable for
use in the loss of ac power event analysis.
The applicant used the LOFTRAN, FACTRAN, and VIPRE-01 codes with the revised thermal
design procedure (RTDP) described in WCAP-11397-P-A to perform DNBR calculations. In the
analysis, initial reactor power, pressurizer pressure, and RCS temperature are assumed to be
at their nominal values consistent with steady-state, full-power operation. The analysis includes
uncertainties in initial conditions, as described in the RTDP, in determining the DNBR limit
during the transient. The functionality of the SG safety valves and pressurizer safety valves is
assumed for steam releases, and the CMTs are assumed to actuate when the PRHR HX cools
down the RCS enough to initiate a low cold-leg temperature S signal.
In considering the effects of a LOOP, the applicant assumed that the power loss and the
resulting coastdown of the RCPs occurs 3 seconds after the turbine trip. If the LOOP occurs at
the start of the event, the calculated DNBR transient will be the same as predicted for the event
involving a complete loss of RCS flow, which a LOOP initiates at the beginning of the event.
Section 15.2.3.2 of this report discusses the results of the complete loss of RCS flow event.
Section 15.0.8, that are available to mitigate the effects of the event, and it determined that the
worst single active failure is the failure to open one of the two valves in the PRHR discharge
line. During the transient, the reactor trips on the low SG water-level signal. The loss of ac
15-20

power is assumed following the reactor trip. Loss of ac power causes the RCPs to coast down.
The PRHR HX actuates on a low narrow-range SG water level coincident with low startup
feedwater flow rate, and the CMTs actuate when the PRHR HX cools the RCS enough to
initiate a low cold-leg temperature S signal.
The results of the analysis show that the calculated minimum DNBR meets the safety DNBR
limit, and the long-term PRHR heat removal capacity is sufficient to remove the decay heat. In
addition, the results show that the peak RCS pressure does not exceed the RCS pressure limit,
pressurizer overfilling does not occur, and the integrity of the RCS is maintained. Thus, the
SRP acceptance criteria for the loss of ac power are met, and the staff concludes that the
analysis is acceptable.
15.2.2.7 Loss of Normal Feedwater Flow (DCD Tier 2, Section 15.2.7)
An LONF flow event, a moderate-frequency event, may be caused by feedwater pump failures,
valve malfunctions, or loss of ac power sources. Following an event involving an LONF, the SG
water inventory decreases as a consequence of continuous steam supply to the turbine. The
mismatch between the steam flow to the turbine and the feedwater leads to the reactor trip on a
low SG-level signal. Either a low narrow-range SG water-level signal, coincident with a low
startup feedwater flow rate signal, or a low wide-range SG water-level signal actuates the
PRHR HX. The PRHR HX transfers the decay heat to the in-containment refueling water
storage tank (IRWST) and provides a continuous core heat removal capability following a loss
of normal and startup feedwater. The RCS cooldown by the PRHR leads to the actuation of a
low cold-leg temperature S signal, which activates the CMTs. The CMTs inject the cold
borated water into the RCS. Both the PRHR HX and CMTs provide heat removal capability for
long-tem decay heat removal.
The applicant performed the analysis of this event with the NRC-approved LOFTRAN computer
code. Initial reactor power is assumed to be 102 percent of the rated power. The relief of
steam in the secondary system is assumed to be achieved through the SG safety valves. Upon
initiation of the event, the RCPs are assumed to operate until they are automatically tripped by
CMT actuation on a low cold-leg temperature S signal. In the analysis, only safety-related
systems are assumed to function to mitigate the consequences of the events. A low widerange SG water level actuates the PRHR HX.
In considering the effects of measurement uncertainties, the initial temperature and pressurizer
pressure are assumed to be 4 EC and 0.446 MPa (7 EF and 50 psi) below the nominal values.
In response to RAI 440.075, related to the staffs concern about the adequacy of the initial
temperature and pressure assumed in the analysis, the applicant replied that during the LONF
event, the availability of ac power is assumed after reactor trip, and the CVCS makeup pump is
assumed to operate. A lower initial pressurizer pressure results in a slightly higher CVCS flow
rate, which is calculated with the LOFTRAN code as a function of the RCS pressure and, in
turn, results in a slightly higher peak pressurizer water level with a lower margin to pressurizer
overfilling. A lower initial RCS temperature results in a higher initial RCS mass and, thus, a
lower margin to the pressurizer overfilling. In addition, the applicant performed a sensitivity
study and showed that the effects of the initial RCS temperature and pressurizer pressure on
15-21

the plant transient behavior are insignificantly small. Therefore, the staff concludes that the
initial RCS temperature and pressure assumed in the analysis are acceptable.
worst single active failure is a failure of one of the two valves in the PRHR discharge line to
open.
resulting coastdown of the RCPs occur 3 seconds after the turbine trip for DNBR calculations.
The LOOP causes a coastdown of the RCPs. The applicant showed that the scenario of the
loss of ac power event, for which the RCPs trip instantaneously, bounds the LONF transient
event followed by the consequential LOOP after turbine trip. The analysis of the loss of ac
power event, presented in DCD Tier 2, Section 15.2.6, shows that the calculated minimum
DNBR exceeds the safety DNBR limits. Therefore, the minimum DNBR for the loss of
feedwater event also exceeds the safety DNBR limits.
The applicant performed the analyses for the peak RCS pressure and LTC for the loss of
feedwater event using approved methods. The results show that the peak RCS pressure does
not exceed the RCS pressure limit, pressurizer overfilling does not occur, and the integrity of
the RCS is maintained. For LTC, the analysis demonstrates that the PRHR can remove the
core decay heat faster than it builds up during the transient, and the long-term PRHR heat
removal capacity is sufficient to remove the decay heat. Thus, the SRP acceptance criteria for
the LONF event are met. Therefore, the staff concludes that the analysis is acceptable.
15.2.2.8 Feedwater System Pipe Break (DCD Tier 2, Section 15.2.8)
An FLB is defined as a break in a feedwater line large enough to prevent the addition of
sufficient feedwater to maintain shell-side water inventory in the SGs. The FLB may reduce the
ability to remove heat generated by the core from the RCS, because fluid in the SG is
discharged through the break, and the break may be large enough to prevent the addition of
main feedwater after the trip. Signals of high pressurizer pressure, overtemperature delta T,
low SG water level in either SG, low steamline pressure in either SG, and high-2 containment
pressure may actuate a reactor trip. During the event, either a low narrow-range SG water
level, coincident with a low startup feedwater rate signal, or a low wide-range SG water-level
signal may actuate the PRHR HX. In the long term, the PRHR HX removes the decay heat and
provides continuous core heat removal capability. A low cold-leg temperature S signal may
actuate the CMTs.
The applicant assumed a break in a feedwater line between the check valve and the SG with a
double-ended rupture of the largest feedwater line. The applicant assumed a double-ended
break area of 0.163 m2 (1.755 ft2). This break size is identified as the limiting break case
because it results in the highest water inventory in the pressurizer and the highest peak primary
pressure. In a followup to RAI 440.076, the staff requested the applicant to discuss the
analysis used to determine the limiting break case for the AP1000 design. In its response, the
applicant provided the results of a break size spectrum study for FLB events. The sensitivity
study compares the results of the limiting FLB case presented in the DCD with FLB analysis for
15-22

break sizes of 100 percent, 50 percent, 25 percent, and 10 percent of the feedwater nozzle,
assuming the break occurred at the initial transient time. The results confirmed that the DCD
case is the limiting case resulting in the highest RCS water inventory in the pressurizer and the
highest peak primary pressure.
The applicant performed the analysis of this event using the LOFTRAN computer code. The
initial power is assumed at 102 percent of the rated power. Initiation of the reactor trip is
assumed when the low narrow-range SG level setpoint is reached on the affected SG. In
minimizing the heat removal capability of the SG with the ruptured feedwater line, a saturated
liquid discharge is assumed for the break fluid until all the water is discharged from the SG with
the ruptured feedwater line. In minimizing the margin to the pressurizer overfilling, the initial
pressurizer water level is assumed at a maximum allowable value. The applicant has
considered the cases with a LOOP occurring simultaneously with the pipe break, with the
LOOP occurring during the FLB accident, and without a LOOP, and identified (in the response
to RAI 440.077) that the FLB with the LOOP occurring at the time of the break is the limiting
case, resulting in the highest RCS pressure. In the analysis for the limiting FLB case, the low
wide-range SG water level is assumed to actuate the PRHR HX, with a maximum delay time of
15 seconds to initiate automatic alignment of the PRHR HX valves. In addition, the applicant
took no credit for the high pressurizer trip, for charging or letdown, or for energy deposited in
RCS metal during the RCS heatup. During an FLB event, the ESFs required to function include
the PRHR, CMTs, and steam isolation valves.
The applicant considered plant systems and equipment, discussed in DCD Tier 2,
worst single active failure is the failure of one of the two valves in the PRHR discharge line to
open. In considering the effects of a LOOP on the DNBR calculations, the applicant assumed
that the power loss and the resulting coastdown of the RCPs occur 3 seconds after the turbine
trip.
The staff notes that the non-safety-related pressurizer spray is credited for heat removal to limit
the increase in the peak RCS pressure. In addition, the analysis assumes a low pressurizer
safety-valve setpoint. Both assumptions will result in a lower peak RCS pressure and, thus, are
not conservative. The AP600 FLB analysis made the same nonconservative assumptions.
During the previous AP600 review, the staff asked the applicant to reanalyze the FLB event and
quantify the effects of the pressurizer spray and a low pressurizer safety-valve setpoint on the
results of the FLB event. In Westinghouse letter DCP/NRC 0962, dated July 18, 1997, the
applicant replied that it had reanalyzed the event without pressurizer spray operable and with
the pressurizer safety-valve setpoint at its normal value. The confirmatory analysis showed a
peak RCS pressure of 18.08 MPa (2624 psia) and an increase of 27.6 kPa (4 psi) as compared
to the DCD case, and it confirmed that the effects of the nonconservative assumptions on the
calculated peak RCS pressure are small. Because the applicants FLB analysis documented in
the DCD shows that the RCS pressure response during an FLB event for the AP1000 design is
similar to that of the AP600 design, the effects of the nonconservatism in modeling the
pressurizer spray and pressurizer safety valves for the AP1000 FLB analysis will also be small.
In addition, the AP1000 FLB analysis shows that the calculated peak RCS pressure is less than
17.93 MPa (2600 psia). Therefore, the staff concludes that the calculated peak pressure
demonstrates that the margin (greater than 1.03 MPa (150 psi )) to the safety limit of
15-23

110 percent of the design pressure is sufficient to compensate for the nonconservative
assumptions of the pressurizer spray and pressurizer safety-valve models discussed above.
The applicant performed the FLB analysis using the LOFTRAN computer code. The results of
the analysis show that the peak pressures of the RCS and SG are below 110 percent of the
design pressures, and the pressurizer does not overfill during the transient. For LTC, the
analysis demonstrates the core coolability by showing that the PRHR removes the core decay
heat faster than it builds up. For DNBR calculations, a LOOP is assumed to occur 3 seconds
after turbine trip and cause a coastdown of the RCPs. The applicant showed that, for the
transient up to the reactor trip and complete insertion of control rods (where the minimum
DNBR occurs), the scenario of the loss of ac power event, for which the RCPs trip
instantaneously, bounds the FLB event followed by the consequential LOOP after turbine trip.
The analysis of the loss of ac power event, presented in DCD Tier 2, Section 15.2.6, shows that
the calculated minimum DNBR exceeds the safety DNBR limits. Therefore, the minimum
DNBR for the FLB event also exceeds the safety DNBR limits.
Because the applicant used NRC-approved methods, the assumptions used in the analysis are
adequate in maximizing RCS pressure and minimizing the calculated DNBRs, and the results of
the analysis meet the acceptance criteria of SRP Section 15.2.8 for the FLB break with respect
to the pressure and safety DNBR limits, the staff concludes that the analysis is acceptable.
15.2.3 Decrease in Reactor Coolant System Flow Rate (DCD Tier 2, Section 15.3)
The applicant has analyzed the transients specified in SRP Section 15.3 for cases resulting
from a decrease in RCS flow rate. The transients include (1) partial loss of forced reactor
coolant flow, (2) complete loss of forced reactor coolant flow, (3) RCP shaft seizure (locked
rotor), and (4) RCP shaft break. The applicant has also identified the limiting case with regard
to the ability of the RCS boundary and fuel rod cladding to withstand the consequences of
transients. The staff has reviewed the applicants analysis, as discussed in Sections 15.2.3.1
through 15.2.3.4 of this report.
15.2.3.1 Partial Loss of Forced Reactor Coolant Flow (DCD Tier 2, Section 15.3.1)
A mechanical or electrical failure in an RCP, or a fault in the power supply to the pumps
supplied by an RCP bus, may cause partial loss of RCS flow, a moderate-frequency event. The
low primary coolant flow reactor trip signal in any reactor coolant loop provides protection
against this event.
The applicant analyzed the partial loss of flow event using the following NRC-approved
computer codes. LOFTRAN calculates the nuclear power transient, the primary system
pressure and temperature transients, and the core flow during the transient based on the RCS
loop coastdown flow from COAST. FACTRAN calculates the heat flux transient based on the
nuclear power and flow from LOFTRAN. VIPRE-01 calculates the DNBRs during the transient
based on the heat flux from FACTRAN and the flow from LOFTRAN. The DNBR calculations
are based on the RTDP described in WCAP-11397-P-A. In the DNBR calculations, the initial
reactor power, pressure, and RCS temperature are assumed to be at their nominal values, and
the uncertainties in initial conditions are included in the DNBR limit as described in the RTDP.
15-24

In maximizing the core power and, thus, minimizing the DNBRs, the applicant used the least
negative MTC and a large absolute value of the Doppler power coefficient. The applicant
calculated the RCP flow coastdown based on RCS pressure losses and RCP characteristics.
Reactor coolant fluid momentum is neglected to obtain a low coastdown flow, which will result in
lower calculated DNBRs.
In considering the effects of a LOOP, the applicant assumed that the power loss and resulting
coastdown of the RCPs occur 3 seconds after the turbine trip. In addition, turbine trip occurs
5 seconds following a reactor trip condition. This delay to turbine trip is a feature of the AP1000
reactor trip system (RTS). The LOOP primarily causes the remaining operating RCPs to coast
down. The analysis shows that the LOOP will have no effect on the calculated minimum DNBR,
because a rapid decrease in the heat flux following a reactor trip significantly compensates for
the decrease in the RCS flow caused by a LOOP following a turbine trip, and the minimum
DNBR occurs before initiation of a LOOP. The staff finds that the applicants assumptions are
conservative and thus acceptable.
single active failure in these systems or equipment adversely affects the consequences of the
events.
Because an event involving the loss of three of the four RCPs is not credible, the applicant did
not analyze the consequences of this event. In addition, the core flow would be much lower for
an event involving the loss of two RCPs than for an event involving a loss of one RCP.
Therefore, the results for an event with a loss of two RCPs are limiting and bound the event
where only one RCP is lost. In DCD Tier 2, Sections 15.3.1 and 15.3.2 analyze and discuss the
loss of two RCPs and the loss of four RCPs, respectively.
The applicant analyzed the event using NRC-approved methods, and the results of the analysis
for the limiting case, the loss of two RCPs, show that, with and without offsite power available,
the RCS pressure will remain within 110 percent of the design pressure, and the minimum
DNBR will remain above the safety DNBR limit. The staff finds that the results of the analysis
meet the acceptance criteria of SRP Section 15.3.1 regarding the limits for the calculated RCS
pressure and the minimum DNBR. Therefore, the staff concludes that the analysis is
acceptable.
15.2.3.2 Complete Loss of Forced Reactor Coolant Flow (DCD Tier 2, Section 15.3.2)
A simultaneous loss of electrical power to all RCPs may cause a complete loss of forced flow
from the RCPs. A LOOP and the resulting loss of all forced reactor coolant flow through the
reactor core cause an increase in the average coolant temperature and a decrease in the
margin to DNB. The signals of low RCP speed or the low reactor coolant loop flow will trip the
reactor.
For the case analyzed with a complete loss of flow, the method of analysis and the assumptions
made for initial conditions and reactivity coefficients are identical to those for a partial loss of
flow, except that the RCP underspeed trip actuates a reactor trip following the loss of power
15-25

supply to all pumps at power. Section 15.2.3.1 of this report discusses the methods and
assumptions used in the analysis. The results of the applicants analysis show that the peak
RCS pressure during the transient will remain below 110 percent of the system design
pressure, and the calculated DNBR will remain above the design DNBR safety limit. Thus, the
integrity of the RCS pressure boundary is not endangered, no fuel failure is predicted to occur,
and core geometry and control rod insertability will be maintained with no loss of core cooling
capability. Therefore, the staff determines that the analysis meets the acceptance criteria of
SRP Section 15.3.2 with respect to the integrity of the RCS pressure boundary and the fuel
rods, and it concludes that the analysis is acceptable.
15.2.3.3 RCP Shaft Seizure (Locked Rotor) (DCD Tier 2, Section 15.3.3) and RCP Shaft Break
DCD Tier 2, Section 15.3.4)
An instantaneous seizure of an RCP rotor may cause RCP shaft seizure, and an instantaneous
failure of an RCP shaft may cause an RCP shaft break. Both events are classified as
limiting-fault events.
For both cases, the RCS flow through the affected reactor loop drops rapidly, leading to a
reactor trip on a low-flow signal. After the reactor trip, energy stored in the fuel rods continues
to be transferred to the coolant, causing the coolant temperature to increase and the coolant to
expand. During this period, heat transfer to the shell side of the SGs drops, because the
reduced flow results in a decreased SG tube convective heat transfer coefficient, and the
reactor coolant in the tube side cools down while the shell-side temperature increases because
of steam flow through the turbine reducing to zero upon plant trip. The rapid expansion of the
coolant in the reactor core, combined with reduced heat transfer in the SGs, causes pressure to
increase throughout the RCS. The pressurizer safety valves will open to release steam from
the pressurizer. The rapid decrease in the RCS flow also results in a decrease in the DNBR.
The analysis discussed in DCD Tier 2, Sections 15.3.3 and 15.3.4, indicates that the RCP shaft
seizure event with a LOOP bounds the RCP shaft break event with a LOOP, because the
slightly faster RCP flow coastdown for the shaft seizure event results in a lower minimum
DNBR.
The applicant analyzed the more limiting RCP shaft seizure for cases with and without offsite
power available. For cases without power available, a LOOP is assumed to occur at 3 seconds
following turbine trip. A LOOP causes a simultaneous loss of feedwater flow, condenser
inoperability, and coastdown of all RCPs. The analysis takes no credit for restoration of offsite
power before initiation of shutdown cooling.
events. The applicant analyzed this event using the LOFTRAN code for the system response
and the FACTRAN code for the heat flux calculation at the hot spot. The NRC has approved
both codes for these analyses. The low reactor coolant flow signal actuates the reactor trip.
The analysis takes no credit for the pressure-reducing effects of pressurizer spray, steam
dump, or controlled feedwater flow.
15-26

The results of the analysis show that the maximum RCS pressure remains less than
110 percent of the design pressure. The applicant also indicated, in the response to
RAI 440.080, that the calculated minimum DNBR is above the safety-limit DNBR and thus
assures no rod failure. However, for the purpose of calculating dose releases, the applicant
conservatively assumed that 16 percent of rods are damaged. The results show that the dose
release limits are met even with the assumed 16 percent of fuel rods damaged. Section 15.3 of
this report discusses the staffs evaluation of the radiological calculations.
The applicant used NRC-approved methods with results that show that the peak RCS pressure
will remain within 110 percent of the design pressure, and the radiological release will remain
within the 10 CFR 50.34(a)(1)(ii)(D)(1) limits. Therefore, the staff finds that the analysis for the
RCP shaft seizure event meets the acceptance criteria of SRP Section 15.3.3 and is
acceptable.
15.2.4 Reactivity and Power Distribution Anomalies (DCD Tier 2, Section 15.4)
In DCD Tier 2, Section 15.4, the applicant presented the analytical results of events caused by
reactivity and power distribution anomalies. The transients include (1) uncontrolled RCCA bank
withdrawal from a subcritical or low-power startup condition, (2) uncontrolled RCCA bank
withdrawal at power, (3) RCCA misalignment, (4) startup of an inactive RCP at an incorrect
temperature, (5) malfunction or failure of the flow controller in a boiling water recirculation loop
that results in an increased reactor coolant flow rate, (6) CVCS malfunction that results in a
decrease in the boron concentration in the reactor coolant, (7) inadvertent loading and
operation of a fuel assembly in an improper position, and (8) spectrum of RCCA ejection
accident. The applicant has also identified the limiting case with regard to ability of the RCS
boundary and fuel rod cladding to withstand the consequences of transients. The following
sections discuss the staffs evaluation of the analytical results.
15.2.4.1 Uncontrolled Rod Cluster Control Assembly Bank Withdrawal from a Subcritical or
Low-Power Startup Condition (DCD Tier 2, Section 15.4.1)
A malfunction of the reactor control or rod control systems may cause an uncontrolled RCCA
bank withdrawal from a subcritical or low-power startup condition. The source-range high
neutron flux reactor trip, intermediate-range high neutron flux reactor trip, power-range high
neutron flux reactor trips (low and high setting), and high-neutron flux rate reactor trip provide
protection against this event.
For the analysis of this transient, the applicant used TWINKLE for the average power
generation calculation, FACTRAN for the hot rod heat transfer calculation, and VIPRE-01 for
the DNBR calculation. The analysis assumes a conservatively low value of Doppler-power
coefficient and the least negative moderator coefficient to maximize the peak heat flux. Reactor
trip is assumed to occur on the low setting of the power-range neutron flux channel at
35 percent of full power. A 10-percent uncertainty is added to the reactor trip setpoint value.
The analysis assumes the maximum positive reactivity addition rate that exceeds that for the
simultaneous withdrawal of the combination of two sequential RCCA banks having the greatest
combined worth at maximum speed of 1.14 m/min (45 in./min). The DNBR calculation assumes
the most limiting axial and radial power shapes associated with the two highest-worth banks in
15-27

their high-worth position. The initial power level is assumed to be below the power level
expected for any shutdown condition (10-9 of nominal power). The combination of the highest
reactivity addition rate and lowest initial power produces the highest peak heat flux, resulting in
a lowest calculated minimum DNBR, and is a conservative assumption.
single active failure in these system or equipment adversely affects the consequences of the
event. Since the turbine is initially in the tripped condition for the plant at a subcritical or
low-power startup condition, a consequential LOOP following the turbine trip is not a credible
event and, thus, is not modeled in the analysis.
The results of the analysis for this event show that the maximum heat flux is much less than the
full-power value, and average fuel temperature increases to a value lower than the nominal
full-power value. The calculated minimum DNBR is above the safety DNBR limits.
The staff has reviewed the assumptions related to the reactivity worth and reactivity coefficients
used in the analysis and found that they maximize the heat flux, thereby minimizing the
calculated DNBRs, and are conservative. The staff has reviewed the calculated consequences
of this transient and found that they meet the requirements of GDC 10, in that the specified
acceptable fuel design limits are not exceeded. The applicant also meets the requirements of
GDC 20, Protection System Functions, in that the reactivity control system can be initiated
automatically so that specified acceptable fuel design limits are not exceeded. In addition, the
applicant meets GDC 25, Protection System Requirements for Reactivity Control
Malfunctions, in that a single malfunction in the reactivity control systems will not cause the
specified acceptable fuel limits to be exceeded. Therefore, the staff concludes that the analysis
satisfies the acceptance criteria of SRP Section 15.4.1 and is acceptable.
15.2.4.2 Uncontrolled Rod Cluster Control Assembly Bank Withdrawal at Power (DCD Tier 2,
Section 15.4.2)
A malfunction of the reactor control or rod control systems may cause an uncontrolled
withdrawal of an RCCA bank in the power operating range, a moderate-frequency event. Such
an event causes an increase in fuel and coolant temperature as a result of the core-turbine
power mismatch. Reactor trips, including the high neutron flux trip, overpower and
overtemperature delta T trips, and pressurizer high-pressure and pressurizer water-level trips,
provide plant protection.
The applicant performed the analyses using NRC-approved methods. The LOFTRAN code
calculates the nuclear power transient, the flow coastdown, the primary system pressure
transient, and the primary coolant temperature transient. The FACTRAN code calculates the
heat flux based on the nuclear power and flow from LOFTRAN. The VIPRE-01 code calculates
the DNBR using the heat flux from FACTRAN and the flow, inlet core temperature, and
pressure from LOFTRAN. The DNBR calculations are based on the RTDP described in
WCAP-11397-P-A. In the DNBR calculations, the initial reactor power, pressure, and RCS
temperature are assumed to be at their nominal values, and the uncertainties in initial
conditions are included in the DNBR limit as described in the RTDP. The maximum positive
15-28

reactivity insertion rate is assumed to exceed that for the simultaneous withdrawal of the
combination of the two control banks having the maximum combined worth at maximum speed.
The high neutron flux signal is assumed to occur at 118 percent of nominal full power. The
overtemperature and overpower delta-T trips include instrumentation and setpoint uncertainties,
and the delays for trip actuation are assumed to be at the maximum values. The applicant
analyzed cases with both minimum and maximum reactivity coefficients and performed a
sensitivity study of the effects of initial power levels (10-, 60-, and 100-percent power) and
reactivity insertion rates (from 1 pcm/s to 110 pcm/s) on the consequences of the event.
event. In addressing the LOOP issue, the applicant assumed that the power loss and the
resulting coastdown of the RCP flow occur 3 seconds after the turbine trip.
The results of the analysis show that the DNBR does not fall below the safety limit for all cases.
Therefore, fuel integrity and adequate fuel cooling are maintained. The calculated peak RCS
pressure will remain less than 110 percent of the design pressure. The staff finds that the
analysis meets the acceptance criteria of SRP Section 15.4.2 with respect to the integrity of the
fuel and pressure boundaries and, therefore, concludes that the analysis is acceptable.
15.2.4.3 Rod Cluster Control Assembly Misalignment (DCD Tier 2, Section 15.4.3)
RCCA misalignment incidents include one or more dropped RCCAs within the same group, a
misaligned full-length assembly, and withdrawal of a single RCCA during operation at power.
Asymmetric power distributions sensed by in-core or ex-core neutron detectors or core exit
thermocouples, rod deviation alarms, or rod position indicators can detect misaligned rods. The
deviation alarm alerts the operator to rod deviation from the group position in excess of
5 percent of span.
Section 15.0.8, that are available to mitigate the effects of the events, and it determined that no
events. In considering the effects of a LOOP, the applicant assumed that a power loss and the
resulting coastdown of the RCPs occur 3 seconds after the turbine trip.
The following sections discuss the staffs evaluation of the analysis for a dropped full-length
assembly, a misaligned full-length assembly, and withdrawal of a single RCCA during operation
at power.
15.2.4.3.1 Analysis for a Dropped Full-Length Assembly
For an event with one or more RCCAs dropped from the same group, the core power
decreases and the core radial peaking factor increases. The reduced core power and
continued steam supply to the turbine cause the reactor coolant temperature to decrease. In
the manual control mode, the positive reactivity feedback causes the reactor power to rise to
the initial power level at a reduced inlet temperature with no power overshoot. In the automatic
15-29

control mode, the plant control system (PLS) detects the reduction in core power and initiates
control bank withdrawal in order to restore the core power. As a result, power overshoot
occurs, resulting in a lower calculated DNBR. The applicant determined that the automatic
operating mode bounds the manual operating mode and is the limiting DNBR case.
The applicant analyzed the rod drop events in the automatic control mode using the nuclear
models with the computer codes described in DCD Tier 2, Table 4.1-2, for the calculation of the
hot channel factor, the LOFTRAN code for the system response, and the VIPRE-01 code for
the DNBR calculation. The results show that the calculated minimum DNBR exceeds the safety
limit DNBR for any single or multiple RCCA drop from the same group, and the peak RCS
pressure will remain less than 110 percent of the design pressure. The staff finds that the
analysis satisfies the acceptance criteria of SRP Section 15.4.3 with respect to the minimum
DNBR and peak pressure and, therefore, concludes that the analysis for the RCCA drop event
is acceptable.
15.2.4.3.2 Analysis for a Misaligned Full-Length Assembly
For RCCA misalignment situations, the applicant analyzed the two most limiting DNBR cases,
including (1) RCCA misalignments in which one RCCA is fully inserted with the rest of the
RCCAs at or above their insertion limits, and (2) a case in which a group is inserted to its
insertion limit, and a single RCCA in the group is stuck in the fully withdrawn position with the
reactor at full-power conditions. In the DNBR analysis, the initial reactor power, pressurizer
pressure, and RCS temperature are assumed to be at their nominal values consistent with
steady-state full-power operation. The radial peaking factor associated with the misaligned
RCCA for these two limiting cases is calculated using the approved methods described in DCD
Tier 2, Table 4.1-2. Uncertainties in initial conditions, as described in WCAP-11397-P-A, are
included in determining the DNBR limit during the transient. The analysis shows that the
minimum DNBR exceeds the safety DNBR limit. Therefore, the staff concludes that the
analysis is acceptable because it meets the acceptance criteria of SRP Section 15.4.3 with
respect to the fuel cladding integrity.
15.2.4.3.3 Analysis for Withdrawal of a Single RCCA
The inadvertent withdrawal of a single assembly requires multiple failures in the rod control
system, multiple operator errors, or deliberate operator actions combined with a single failure of
the rod control system. Because of the low likelihood of the event, the applicant classified the
single assembly withdrawal as an infrequent event for the AP1000 design. The event
categorization is consistent with that approved by the staff for Westinghouse operating plants
and, therefore, is acceptable. The transient resulting from such an event is similar to that
resulting from a bank withdrawal, but the increased peaking factor causes DNB to occur in the
region surrounding the withdrawn assembly. The radial peaking factor associated with the
single RCCA withdrawal is calculated using the approved methods described in DCD Tier 2,
Table 4.1-2. Uncertainties in initial conditions, as described in WCAP-11397-P-A, are included
in determining the DNBR limit during the transient. In response to RAI 440.081, the applicant
indicated that less than 4 percent of the rods in the core experience DNB during the limiting
case, an event where RCCA rod banks are at the full-power rod insertion limits, except for one
RCCA which is fully withdrawn. For the purpose of calculating dose releases, the applicant
15-30

conservatively assumed that 5 percent of the fuel rods failed. The assumption of fuel failure for
the dose release calculation is more conservative than the guidance in SRP Section 4.4, which
states that all rods that experience DNB (in this case, less than 4 percent of the fuel rods)
should be assumed to fail. Therefore, the staff concludes that the assumption is acceptable.
For the single rod withdrawal event (an infrequent event), the applicant meets the requirements
of GDC 27, Combined Reactivity Control Systems Capability, by demonstrating that the
resultant fuel damage is limited, such that control rod insertability is maintained, and no loss of
core coolability results. The DNBR calculation shows that a small fraction (4 percent) of the fuel
rods may experience cladding perforation. The dose release calculation results show that the
release acceptance criteria are met. Therefore, the staff concludes that the analysis is
acceptable. Section 15.3 of this report discusses the staffs evaluation of the radiological
consequence calculations.
15.2.4.4 Startup of an Inactive Reactor Coolant Pump at an Incorrect Temperature (DCD
Tier 2, Section 15.4.4)
Starting an idle RCP increases the injection of cold water into the core, which causes a
reactivity insertion and subsequent power increase.
Because the TSs (described in DCD Tier 2, Chapter 16, TS 3.4.4) do not allow operation with
an RCP inoperable for Modes 1 and 2, the applicant did not analyze this event at Modes 1 and
2.
15.2.4.5 A Malfunction or Failure of the Flow Controller in a Boiling-Water Reactor Loop that
Results in an Increased Reactor Coolant Flow (DCD Tier 2, Section 15.4.5)
This section does not apply to the AP1000 design.
15.2.4.6 Chemical and Volume Control System Malfunction that Results in the Boron Dilution in
the Reactor Coolant (DCD Tier 2, Section 15.4.6)
Failures of the demineralized water transfer and storage system (DWS) or CVCS because of
control system or operator error or mechanical failure cause an inadvertent boron dilution. The
CVCS and DWS are designed to limit the dilution rate to values that allow sufficient time for
automatic or operator actions to terminate the dilution before the shutdown margin is lost. The
boric acid and blended flow rates, the status of the CVCS makeup pumps, and the boron
dilution rate deviation alarm indicate the dilution rate. In Modes 1 and 2, either a rod insertion
limitlow-level alarm or an axial flux difference alarm will alert the operator to an unplanned
boron dilution event. Furthermore, when the reactor is subcritical, the available alarms and
indications to alert the operator to a boron dilution event include (1) a high flux at shutdown
alarm, (2) indicated source-range neutron flux count rates, (3) an audible source-range neutron
flux rate, and (4) a source-range neutron flux multiplication alarm. Upon any reactor trip signal,
source-range flux multiplication signal, low battery charger input voltage signal, or a safety
injection signal, a safety-related function automatically isolates the potential unborated water
from the DWS and thereby terminates the dilution.
15-31

The applicant analyzed the boron dilution event for all modes of operation. The applicant
performed the analysis using the method consistent with that employed in boron dilution event
analysis for Westinghouse operating plants. The method consists of a generic fluid mixing
model. The nodal scheme in the model includes a node to represent the RCS volume and a
flowpath to represent CVCS fluid transportation.
All cases discussed below assume a dilution flow rate of 12.6 liters per second (L/sec)
(200 gallons per minute (gpm)) of unborated water, which is the maximum makeup flow with
both makeup pumps operating (as stated in DCD Tier 2, Section 9.3.6.6.1.2).
15.2.4.6.1 Boron Dilution during Refueling (Mode 6)
Uncontrolled boron dilution is not a credible event during the refueling mode because
administrative controls isolate the RCS from the potential source of unborated water by locking
closed specified valves in the CVCS system during this mode of operation. The boric acid tank
(BAT), which contains borated water, supplies makeup water during refueling.
15.2.4.6.2 Boron Dilution during Modes 3, 4, and 5 of Operation
In Modes 3, 4, and 5, the analysis assumed a shutdown margin of 1.6 percent delta K/K (the
minimum value required by the AP1000 TSs for the shutdown modes) and the minimum initial
reactor coolant volumes. Following the AP1000 TS LCO 3.4.8 requirements, the applicant
assumed the operation of one RCP. In maximizing the effect of the boron dilution, the applicant
used the minimum amount of the water in the RCS to mix with the incoming unborated water.
For Mode 3, the minimum RCS water volume assumed is the total RCS volume without the
pressurizer and surgeline and the RV upper head volume. For Mode 4, the water volume
assumed in the analysis is the water volume of the RV without the RV upper head volume when
the normal residual heat removal system (RNS) is used to remove the decay heat. For Mode 5,
the water volume used in the analysis is the RCS water volume corresponding to the water level
at midloop operations. The source-range flux multiplication signal is assumed to actuate an
alarm in the control room and close the DWS isolation valves when the neutron flux increases
by 60 percent over any 50-minute period (per Item 15.a of AP1000 TS Table 3.3.2-1). The
analysis shows that the automatic closure of the DWS isolation valves initiated by the sourcerange flux multiplication signal occurs about 56.3 minutes after the start of the dilution for
Mode 3, 12.3 minutes for Mode 4, and 12.03 minutes for Mode 5. The results of the analysis
show that the automatic isolation of the DWS valves terminates the boron dilution and
maintains the plant in a subcritical condition. The staff determined that the analysis meets the
guidance in SRP Section 15.4.6 with respect to core criticality and concludes that it is
acceptable.
15.2.4.6.3 Boron Dilution during Startup (Mode 2)
The plant is in the startup mode only for startup testing at the beginning of each cycle. During
this mode of operation, rod control is manual. The applicant performed an analysis of an
inadvertent deboration event at initial conditions representative of the startup mode of operation
with an assumed unborated water flow rate of 12.6 L/sec (200 gpm). Following the
requirements of TSs 3.1.1 and 3.4.4, the applicant assumed an available shutdown margin of
15-32

1.6 percent delta K/K and operation of four RCPs. The initial RCS water assumed in the
analysis is the water volume included in all the RCS volumes except the pressurizer and
surgeline. Calculation of the SG tube volume accounts for 10-percent tube plugging. The
results of the analysis show that a reactor trip from a signal on the intermediate-range neutron
flux will (1) initiate closure of the DWS isolation valves (DCD Tier 2, Table 15.4-1), (2) terminate
the boron dilution, and (3) maintain the plant in a subcritical condition. Therefore, the staff
determined that the analysis meets the guidance in SRP Section 15.4.6 with respect to core
subcriticality and concludes that the analysis is acceptable.
15.2.4.6.4 Boron Dilution during Power Operation (Mode 1)
For Mode 1, the applicant analyzed both the manual mode and the automatic mode cases.
Both cases use the same initial RCS water volume as for Mode 2 discussed above. For the
manual mode case, the analytical results show that a reactor trip on the overtemperature
delta T will initiate closure of the DWS isolation valves and terminate the boron dilution without
the occurrence of a posttrip return to criticality. Because a reactor trip isolates DWS valves and
terminates the event, the subsequent LOOP assumption following a turbine trip (which occurs
immediately after a reactor trip), as required by GDC 17, will not affect the results of the
deboration event for the case in manual mode.
For the automatic mode case, the slow insertion of the control rods to avoid the reactor trip
compensates for an increase in the power and temperature caused by a boron dilution event.
Because a reactor and turbine trip does not occur as predicted in the analysis for the case in
automatic mode, the consequential LOOP following a turbine trip (as required by GDC 17) is
not a credible event and thus is not modeled in the analysis. For the AP1000 design, the
redundant pretrip alarms available to the operator for Mode 1 operation include a low-level rod
insertion limit alarm and an axial flux difference alarm. The analysis shows an available time
interval from a low-low rod insertion limit alarm, attributable to boron dilution, to loss of
shutdown margin of about 328 minutes (DCD Tier 2, Table 15.4-1). The staff finds that the
applicant has demonstrated compliance with the guidance in SRP Section 15.4.6, in that the
redundant pretrip alarms should alert the operator to the initiation of the event in sufficient time
(at least 15 minutes) to ensure detection of the boron dilution event during Mode 1 before
possible loss of shutdown margin.
The analysis shows that (1) for Mode 6, the design and procedures prevent the inadvertent
boron dilution events, (2) for Mode 1 in a manual control mode and Modes 2 through 5, the
automatic closure of the DWS isolation valves minimizes the approach to criticality and
maintains the core in a subcritical condition, thus ensuring the integrity of the fuel and RCS
pressure boundary, and (3) for Mode 1 in an automatic control mode, a number of alarms and
indications can alert an operator to a boron dilution event, and the operator has sufficient time
(328 min) to detect and terminate the event before loss of shutdown margin. Therefore, the
staff determines that the analysis has satisfied the guidance in SRP Section 15.4.6 with respect
to the operator action times and core subcriticality and, therefore, concludes that it is
acceptable.
In support of the boron mixing model used in the analysis, the applicant specified a required
minimum core flow rate. Specifically, TS LCO 3.4.8 requires that at least one RCP operate with
15-33

a total flow through the core of at least 630 L/sec (10,000 gpm) while in Modes 3, 4, and 5,
whenever the reactor trip breakers are open. The staff requested, in RAI 440.106, that the
applicant provide the basis to support the conclusion that the required core flow rate is sufficient
to provide the well-mixed flow condition assumed in the boron dilution analysis. The applicant
replied that the process of selecting 630 L/sec (10,000 gpm) included general consideration of
the results reported in NUREG/CR-2733, Experimental Data Report for LOFT Boron Dilution
Experiment L6-6, issued June 1982. As discussed in EGG-LOFT-5867, Quick-Look Report
on LOFT Boron Dilution Experiment L6-6, issued May 1982, the key parameters of the loss-offluid test (LOFT) L6 series of tests were scaled based on the characteristics of the
Westinghouse four-loop Trojan PWR. With its four cold-legs, the general configuration of the
AP1000 inlet plenum region is similar to that of a four-loop plant.
The LOFT considered two low-pressure injection system flow rates that were scaled to provide
equivalence to 189 L/sec (3000 gpm) and 378 L/sec (6000 gpm) residual heat removal (RHR)
flow rates in the Trojan plant. Typical RHR-related TSs, intended to ensure adequate boron
mixing in current Westinghouse-designed plants, allow operation in the applicable mode with a
single operating RHR pump. EGG-M-03783, DE83 013666, PWR Response to an Inadvertent
Boron Dilution Event, presented at the Third Multiphase Flow and Heat-Transfer Symposium
Workshop, April 1820, 1983, documents the results of the tests. The report indicates that, for
the 189 L/sec (3000 gpm) RHR flow equivalent case, the fluid volume in the reactor vessel was
well mixed and that the assumption of perfect mixing, though not strictly correct, is adequate for
calculational purpose. For the 378 L/sec (6000 gpm) flow equivalent case, the reported test
results show an even closer approach to perfect mixing. These results of the LOFTs have been
used to support the typical plant TSs that generally accept an RHR flow in the vicinity of
189 L/sec (3000 gpm) as sufficient to justify the perfect mixing assumption modeled in the
boron dilution analysis.
For the AP1000, the minimum core flow required by the TS exceeds the flow rates considered
in the LOFT and currently accepted as providing adequate mixing in the operating plants. In
addition, SR 3.4.8.1 places an operating speed requirement on a single RCP. Specifically, the
SR requires that, in order to be considered as an operating RCP, the single pump involved
must operate at a minimum of 25-percent rated speed, which produces a flow rate of
1,239 L/sec (19,688 gpm). This SR indicates that the total RCP flow is almost twice the
required 630 L/sec (10,000 gpm) core flow and much greater than the 189 L/sec (3,000 gpm)
value that is typically applied to operating plants. Since the general configuration of the
AP1000 inlet flow plenum region is similar to that of a four-loop Westinghouse plant, the LOFTs
that were scaled to a Westinghouse plant and are used to support the boron mixing model for
the current Westinghouse plants apply to the AP1000 for selection of the minimum core flow
rate to assure a well-mixed flow condition. In addition, the required minimum RCP flow through
the core of 630 L/sec (10,000 gpm) is much greater than the value of 189 L/sec (3,000 gpm)
that is typically applied to the operating plants and supported by the LOFT results. Therefore,
the staff concludes that the required minimum core flow gives reasonable assurance that it is
sufficient to provide the well-mixed flow conditions considered in boron dilution events that were
analyzed to address the guidance in SRP Section 15.4.6, and is therefore acceptable.
15-34

15.2.4.7 Inadvertent Loading and Operation of a Fuel Assembly in an Improper Position (DCD
The applicant indicated that, during fuel loadings, it will follow strict administrative controls to
prevent operation with a misplaced fuel assembly or a misloaded burnable poison assembly.
Nevertheless, the applicant performed an analysis of the consequences of a loading error.
The applicant used the NRC-approved methods documented in WCAP-10965-P-A, ANC:
Westinghouse Advanced Nodal Computer Code, issued September 1986, to perform the
analysis for this event. In DCD Tier 2, Figures 15.4.7-1 through 15.4.7-4, the applicant
provided comparisons of power distributions calculated for the nominal fuel loading pattern and
those calculated for four loadings with misplaced fuel assemblies or burnable poison
assemblies. The selected non-normal loadings represent the spectrum of potential inadvertent
fuel misplacement, including (1) a case in which a Region 1 assembly is interchanged with a
Region 3 assembly, (2) a case in which a Region 1 assembly is interchanged with a
neighboring Region 2 fuel assembly, (3) the enrichment error with a case in which a Region 2
fuel assembly is loaded in the core central position, and (4) a case in which a Region 2 fuel
assembly, instead of a Region 1 fuel assembly, is loaded near the core periphery.
The analysis described above shows that resulting power distribution effects will either be
detected by the startup test involving the in-core detector system (and hence be remediable) or
cause an acceptable small perturbation within the measurement uncertainty of 5 percent. The
testing requirements and the results of the analysis demonstrate that the applicant has met the
requirements of GDC 13, Instrumentation and Control, with respect to minimizing the
possibility that a misloaded fuel assembly goes undetected (and minimizes the consequences
of reactor operation in the event of inadvertent fuel misload). For the undetectable errors, the
resulting power distribution changes fall within the acceptable measurement uncertainty,
ensuring no fuel failure and satisfying the SRP Section 15.4.7 guidance. Therefore, the staff
concludes that the analysis is acceptable.
15.2.4.8 Spectrum of Rod Cluster Control Assembly Ejection Accidents (DCD Tier 2,
Section 15.4.8)
The mechanical failure of a control rod mechanism pressure housing may result in the ejection
of an RCCA. For assemblies initially inserted, the consequences include a rapid reactivity
insertion together with an adverse core power distribution, possibly leading to localized fuel rod
damage. Although mechanical provisions have been made to render this accident extremely
unlikely, the applicant has provided its analysis of the consequences of such an event. The
applicant has considered plant systems and equipment, discussed in DCD Tier 2,
single active failure in these systems or equipment adversely affect the consequences of the
events. The staff has reviewed this analysis in accordance with SRP Section 15.4.8.
WCAP-7588, Revision 1-A, An Evaluation of the Rod Ejection Accident in Westinghouse
Pressurized Water Reactors Using Spatial Kinetics Methods, issued January 1975, which the
staff has previously reviewed and accepted, documents the methods used in the analysis.
15-35

The applicant analyzed two sets of cases for the rod ejection event, one initiated at hot fullpower (HFP) and one initiated at hot zero-power (HZP). The analysis of both of these cases
uses both beginning-of-cycle (BOC) and end-of-cycle (EOC) kinetics. DCD Tier 2,
Table 15.4-3, lists the values of the initial plant parameters (power level, ejected rod worth,
delayed neutron fraction, and trip reactivity) assumed in the analysis. The analysis credits the
high neutron flux trip (high and low setting) to trip the reactor. The results show that the
calculated values of hot spot radially averaged fuel enthalpy for the four analyzed cases are
181 calories per gram (cal/g) for HFP-BOC, 104 cal/g for HZP-BOC, 170 cal/g for HFP-EOC,
and 117 cal/g for HZP EOC. These values of peak fuel enthalpy fall below the safety limit of
280 cal/g specified in SRP Section 15.4.8, Spectrum of Rod Ejection Accidents. The
calculated values also fall within the Westinghouse-specified analysis limit of 200 cal/g. In
addition, the calculated pressure surge resulting from the rod ejection does not exceed the RCS
emergency limits (Service Level C) and thus satisfies the guidance of SRP Section 15.4.8 with
respect to the RCS pressure limit.
resulting RCP coastdown occur 3 seconds after the turbine trip. The applicant has shown that
the effect of a LOOP on the calculated minimum DNBR is negligible, because a rapid decrease
in the heat flux after the control rod insertion compensates for the decrease in the RCS flow
caused by a LOOP, and the minimum DNBR occurs before initiation of a LOOP.
The analysis shows that less than 10 percent of the fuel rods experience DNB as a result of the
rod ejection event. For the purpose of calculating dose releases, the applicant conservatively
assume that 10 percent of the fuel fails. The assumption of fuel failure for the dose calculation
results in a higher radiological release dose and is conservative. Therefore, the assumption is
acceptable.
Experimental data show failure of high burnup fuels at lower enthalpy values than the fuel
enthalpy safety limit specified in SRP Section 15.4.8. However, the staff, the industry, and the
international community agree that burnup degradation in the margin to low-enthalpy fuel failure
is likely to be regained by application of more detailed three-dimensional (3-D) analysis
methods of the fuel response to rod ejection accidents. Detailed 3-D models predict that the
value of the peak fuel rod enthalpy would fall below 100 cal/gm (R.O. Meyer, R.K. McCardell,
H.M. Chung, D.J. Diamond, and H.H. Scott, A Regulatory Assessment of Test Data for
Reactivity-Initiated Accidents, Nuclear Safety, Volume 37, Number 4, OctoberDecember
1996, pages 271288). In addition, a generic analysis performed by the applicant that assumes
low-enthalpy fuel failure shows that the radiological consequences of rod ejection accidents
meet the acceptance criteria specified in Appendix A to SRP Section 15.4.8. As indicated in the
response to RAI 440.181, the applicants generic analysis is predicated on conservative
treatment of the experimental fuel data applied to existing and planned cores operating within
approved burnup limits for PWRs. Therefore, the staff concludes that, although the SRP
Section 15.4.8 fuel enthalpy safety limit may not be conservative, the generic analysis provides
reasonable assurance that radiological consequences of the rod ejection accident will not
violate the acceptance criteria in SRP Section 15.4.8 for the AP1000 core operating within the
current NRC-approved burnup limits. The staff will not accept further extension of burnup limits
until additional experimental information on fuel behavior is available to demonstrate that the
fuel cladding will satisfy the regulatory acceptance criteria used in the rod ejection analyses for
15-36

licensing applications. Section 15.3 of this report includes the review of the radiological
releases.
The applicant performed the analysis using acceptable methods. The results of the analysis
show that the calculated values of peak fuel enthalpy fall below the acceptable limit specified in
SRP Section 15.4.8, the calculated peak RCS pressure does not exceed the RCS emergency
limits (Service Level C), and the radiological consequences meet the SRP Section 15.4.8
acceptance criteria. The staff finds that the analysis meets the acceptance criteria of SRP
Section 15.4.8, with respect to the limits of the hot rod average enthalpy, RCS pressure, and
radiological consequences. Therefore, the staff concludes that the analysis is acceptable.
15.2.5 Increase in Reactor Coolant System Inventory (DCD Tier 2, Section 15.5)
In DCD Tier 2, Section 15.5, the applicant considers two cases which would result in an
increase in the RCS inventory. These cases are (1) an inadvertent operation of the CMTs and
(2) malfunction of the chemical and control system.
15.2.5.1 Inadvertent Operation of the Core Makeup Tanks during Power Operation (DCD
Operator action, a false electrical actuation signal, or a valve malfunction can cause spurious
CMT operations. The DCD presents the results of the most limiting case, a CMT inadvertently
actuated by operator error or a mechanical failure resulting in the opening of two valves in the
CMT discharge lines. During the event, the high-3 pressurizer water-level signal actuates to trip
the reactor, followed by the PRHR actuation and eventually by an S signal, which then
actuates the second CMT. The applicant analyzed the case using the LOFTRAN code and
established the following initial conditions to maximize pressurizer water level:
The reactor power is at 102 percent of nominal, the pressure is at 344.7 kPa (50 psi)
below nominal, and RCS temperature is at 3.9 EC (7 EF) below nominal.
The pressurizer spray system and automatic rod control are operable.
A least-negative MTC, a low (absolute) Doppler power coefficient, and a maximum

boron worth are assumed.
The CMT enthalpies are maximized to minimize the cooling provided by the CMTs. The
pressure drop of the CMT injection and balance lines is minimized to maximize the CMT flow
injected into the primary system. In response to RAI 440.085, the applicant indicated that
modeling high pressure drops through the PRHR loop minimizes the PRHR heat transfer
capability. A higher pressure drop limits the PRHR flow and reduces the calculated value of the
primary-side heat transfer coefficient. In addition, a maximum TS value for PRHR tube
plugging and a minimum effective heat transfer area have been assumed. The assumptions
using the higher CMT injection flow and a minimum PRHR heat transfer capability result in an
increase in the RCS temperature and RCS expansion, thus reducing the margin to pressurizer
overfilling. Therefore, the assumptions are conservative and acceptable.
15-37

Section 15.0.8, that are available to mitigate the effects of the event, and it identified that the
worst single failure is one of the two PRHR parallel isolation valves failing closed. In addressing
the issue of a LOOP, the applicant assumed that a power loss and the resulting coastdown of
the RCPs occur 3 seconds after the turbine trip.
The analysis assumes that an inadvertent opening of the CMT discharge valves, which results
in the one CMT injecting borated water, initiates the event. During the transient, the reactor is
tripped upon receipt of the high-3 pressurizer-level signal. Following reactor trip, the reactor
power drops and average RCS temperature decreases with subsequent coolant shrinkage. At
about the same time of the reactor trip, the turbine is tripped and, after a 3-second delay, a
consequential LOOP is assumed, and the RCPs are tripped. The cold-leg temperature
increases, resulting in an increased CMT injection rate, because of the increased driving head
from the density decrease in the pressure balance line. The CMT injection makes up the RCS
shrinkage, and, within 1 minute after actuation of the high-3 pressurizer-level signal, the high-3
pressurizer-level setpoint is once again reached. Initiation of the PRHR, with appropriate delay
time, is then assumed. The primary and secondary pressures increase initially because of the
assumed unavailability of the non-safety-related control systems, but eventually decrease as
the PRHR removes the core decay heat. At about 1.39 hours, the PRHR heat flux matches the
core decay heat. During this period, the pressurizer level continues to slowly increase until the
CMT recirculation decreases sufficiently to limit the mass addition to the RCS. After about
3.43 hours into the transient, the cold-leg temperature (S) setpoint is reached, and the second
CMT is actuated. The pressurizer level initially shrinks from the addition of cold borated water.
As the CMT continues to add water to the primary system, the pressurizer level begins to
increase. At 3.69 hours, the first CMT stops recirculation. At 6.06 hours, the PHRH heat flux
approaches the core heat flux. The CMTs stop recirculating at 8.52 hours into the transient.
The staff finds that the applicant used the LOFTRAN code for the analysis and adequately
identified the limiting case. The results of analysis show that no RCS water is relieved through
the pressurizer safety valves as a result of the transient. In addition, the calculated minimum
DNBR remains above the safety limit value, and the RCS and SG pressures remain below
110 percent of their respective design pressures. The staff determines that the analysis meets
the acceptance criteria of SRP Section 15.5.1 with respect to the pressure limit and core DNBR
safety limit and, therefore, concludes that the analysis is acceptable.
15.2.5.2 CVCS Malfunction that Increases Reactor Coolant Inventory (DCD Tier 2,
Section 15.5.2)
A CVCS malfunction may result in an event that increases RCS inventory. Operator action, an
electrical actuation signal, or a valve failure may cause the CVCS to malfunction. The DCD
presents the results of the most limiting case, the CVCS malfunction caused by an operator
error, resulting in the startup of two CVCS pumps to deliver the flow to the RCS. The applicant
analyzed CVCS malfunction cases using the LOFTRAN code and established the following
initial conditions to maximize the pressurizer water level:
The reactor power is at 102 percent of nominal, the pressure is 344.7 kPa (50 psi)
above nominal, and the RCS temperature is at 3.6 EC (6.5 EF) above nominal.
15-38
The pressurizer spray system is operable.
A least-negative MTC, a low (absolute) Doppler power coefficient, and a maximum

boron worth are assumed.
The initial boron concentration is chosen on the basis of an iterative analysis process,
such that the limiting case bounds the case that models explicit operator actions after
the reactor trip.
worst single failure is one of the two PRHR parallel isolation valves failing closed. In addressing
the issue of a LOOP, the applicant assumed that a power loss and resulting coastdown of the
RCPs occur 3 seconds after the turbine trip.
The analysis assumes that a CVCS malfunction that results in injection from two CVCS pumps
initiates the event. As the CVCS injection flow increases RCS inventory, the pressurizer water
volume begins increasing while the primary system cools down. The RCS temperature
decreases to reach the low cold-leg temperature setpoint and actuates an S signal, resulting
in a reactor trip. Following the reactor trip, the turbine is tripped, and, after a 3-second delay, a
consequential LOOP is assumed, and the RCPs are tripped. Soon after the reactor trip, main
feedwater lines, steamlines, and the CVCS are isolated. After a delay of 12 seconds following
the S signal, the CMT discharge valves open, and 5 seconds afterward the PRHR HX is
actuated. The operation of the PRHR HX and CMTs cools down the plant. At about 4.09 hours
into the transient, the PRHR heat flux matches the core decay heat, and at 5.61 hours the
CMTs stop recirculating.
The staff finds that the applicant used the LOFTRAN code for the analysis with adequate inputs
and appropriately identified the limiting case, and the results show that no RCS water is relieved
from the pressurizer safety valves. In addition, the calculated minimum DNBR remains above
the safety limit values, and the RCS and SG pressures remain below 110 percent of their
respective design pressures. The staff determines that the analysis meets the acceptance
criteria of SRP Section 15.5.2 with respect to the pressure limit and core DNBR safety limit.
Therefore, the staff concludes that the analysis is acceptable.
15.2.6 Decrease in Reactor Coolant Inventory (DCD Tier 2, Section 15.6)
In DCD Tier 2, Section 15.6, the applicant provided an analysis of events that may decrease the
RCS inventory. These events include (1) an inadvertent opening of a pressurizer safety valve
or inadvertent operation of the automatic depressurization system (ADS), (2) a break in an
instrument line or other lines from the reactor coolant boundary that penetrate the containment,
(3) an SG tube failure, and (4) a LOCA resulting from a spectrum of postulated piping breaks
within the RCPB. The following sections discuss the applicants analysis and the staffs
evaluation.
15-39

15.2.6.1 Inadvertent Opening of a Pressurizer Safety Valve or Inadvertent Operation of the
Automatic Depressurization System (DCD Tier 2, Section 15.6.1)
An accidental depressurization of the RCS may occur as a result of an inadvertent opening of a
pressurizer safety valve or ADS valves. During the transient, the RCS pressure rapidly
decreases and, in turn, causes a decrease in power because of the moderator density reactivity
feedback. The pressurizer level may eventually drop far enough to cause a reactor trip on a
low pressurizer-level signal.
The ADS consists of four stages of depressurization valves, which are interlocked such that
Stage 1 is initiated first, with subsequent stages actuated only after previous stages have been
actuated. The AP1000 design prohibits opening of the fourth stage valves while the RCS is at
nominal operating pressure. For inadvertent operation of the ADS valves, the applicant
considered an opening of both first stage ADS flowpaths to be the limiting case, because
operation of these valves results in a greater depressurization rate than ADS Stages 2 and 3
valves given the shorter first stage ADS valve opening time.
The applicant has also analyzed an inadvertent opening of the pressurizer safety valve. The
flow area of the pressurizer valve is smaller than the combined two ADS Stage 1 valves;
however, the safety valves open more rapidly than the ADS valves.
Normal reactor control systems are assumed not to function. The rod control system is
assumed to be in automatic mode in order to maintain the core at full power until the reactor trip
protection function is reached.
single active failure in these systems or equipment adversely affected the consequences of the
event. In addressing a LOOP, the applicant assumed that a power loss and resulting
coastdown of the RCPs occur 3 seconds after the turbine trip. The analysis shows that a
LOOP has no effect on the calculated minimum DNBR, because a rapid decrease in the heat
flux after the reactor trip compensates for the decrease in the RCS flow caused by the LOOP
(which would follow a turbine trip), and the minimum DNBR occurs before initiation of a LOOP.
To perform the analysis of these events, the applicant used LOFTRAN for the transient
response calculation, FACTRAN for the heat flux calculation, and VIPRE-01 for the DNBR
calculation. The DNBR calculations for these RCS valve opening events are performed using
the revised thermal margin procedure in WCAP-11397-P-A. Initial core power, RCS pressure,
and RCS temperature are assumed to be at their nominal values, consistent with steady-state,
full-power operation.
The staff finds that the applicant analyzed the events using acceptable methods. The analysis
shows that the overtemperature delta T reactor trip signal provides adequate protection against
the RCS depressurization events. The calculated DNBR remains above the safety limiting
value, and the RCS pressure remains less than 110 percent of the design pressure throughout
the transients. The staff determines that the analysis meets the acceptance criteria of SRP
15-40

Section 15.6.1 with respect to the pressure and core safety DNBR limits and, therefore,
concludes that it is acceptable.
15.2.6.2 Failure of Small Lines Carrying Primary Coolant Outside Containment (DCD Tier 2,
Section 15.6.2)
The reactor coolant may be released directly from a break or leak outside containment in a
CVCS discharge line or sample line. The applicants analysis has identified that the worst case
event is the double-ended break of the sample line between the isolation valve outside the
containment and the sample panel. This sample line break results in the largest release of
reactor coolant outside containment. The sample line orifices limit the maximum breakflow to
8.2 L/s (130 gpm).
Both the isolation valves inside and outside containment are open only during sampling, and the
loss of sample flow will provide an indication of the break to plant operators. A break in a
sample line releases radioactivity which will actuate area and air radiation monitors. Because
the operator sees multiple indications, the applicant assumed that, 30 minutes after initiation of
a break, the operator would isolate the sample line and terminate further release of primary fluid
discharged to the atmosphere. The assumed operator action delay time of 30 minutes is
consistent with the current operating plant design-basis analysis of a break of a small line
outside containment and, therefore, is acceptable.
The assumptions used for analysis of this event are adequate and acceptable, and the scenario
described in DCD Tier 2, Section 15.6.2, ensures that the applicant has considered the most
severe failure of piping carrying the primary coolant outside containment. In addition, the
radiological releases fall within the 10 CFR 50.34(a)(1)(ii)(D)(1) limits. Thus, the staff
determines that the analysis meets the SRP Section 15.6.2 acceptance criteria. Therefore, the
staff concludes that the analysis is acceptable. Section 15.3 of this report discusses the staffs
evaluation of the radiological release calculations.
15.2.6.3 Steam Generator Tube Rupture (DCD Tier 2, Section 15.6.3)
The SGTR accident is defined as a penetration of the barrier between the RCS and the main
steam system. The failure of an SG U-tube may cause this accident.
The analysis for the SGTR event consists of two parts, SG overfill calculation and the
calculation of the SG mass releases used to evaluate the radiological consequence.
The applicant performed an analysis with LOFTTR2 to demonstrate that the AP1000 design
features can prevent the SG from overfilling with water. To maximize the SG water increase,
the applicant identified the limiting single failure as the failure of the startup feedwater control
valve to throttle flow when nominal SG level is reached. Other conservative assumptions
maximizing the SG secondary water inventory include a high initial SG level, minimum initial
RCS pressure, LOOP, maximum CVCS injection flow, maximum pressurizer heater addition,
maximum startup feedwater flow, and minimum startup feedwater delay time. The results of
the analysis demonstrate that the AP1000 protection system and passive system design
features will prevent the SG from overfilling with water during an SGTR.
15-41

For the SG mass release calculation, the applicant performed the SGTR analysis using the
LOFTTR2 code for a case with complete severance of a single SG tube. At the initiation of an
SGTR, the reactor is assumed to be at nominal full power. The initial secondary mass is
assumed at nominal SG mass with an allowance for uncertainties. A LOOP is assumed at the
start of the event, because tripping the RCPs (resulting from a LOOP) has been determined to
maximize flashing of primary-to-secondary break flow, consequently maximizing radiological
releases. The LOOP is assumed to trip the reactor. Consistent with the assumption of a
LOOP, main feedwater pump coastdown occurs after the reactor trip, and no startup feedwater
is assumed in order to minimize SG secondary inventory and, thus, maximize secondary activity
concentration and steam release. The CVCS pumps are assumed to be loaded onto the diesel
generators. Maximum CVCS flows and the pressurizer heater addition are assumed at the
initiation of the event (even though offsite power is not available) to maximize primary-tosecondary leakage. The CVCS is assumed to isolate on the high-2 SG narrow-range level
setpoint. Because the failure of the steam dump system would result in a steam release from
the SG power-operated relief vales (PORVs) to the atmosphere following the reactor trip, the
steam dump system is assumed to be inoperable to maximize the radiological releases.
most limiting single failure is a failed-open PORV on the affected SG. The applicant assumed
that the single failure occurs coincidently with the low-2 pressurizer-level signal, maximizing the
integrated RCS-to-secondary break flow. The automatic closure of the associated block valve
on a low steamline pressure protection system signal isolates the SG PORV.
The analysis shows that after the reactor trip, low pressurizer pressure generates a safeguard
S signal. The S signal results in CMT actuation and PRHR system actuation. Opening the
SG PORVs and operation of the PRHR and CMTs decrease the primary and secondary
pressures. When the secondary pressure decreases to the low steamline pressure setpoint,
the steamline isolation valves and SG PORV block valves are closed. Following closure of the
block valves, the primary and secondary pressures and faulted SG secondary water volume
increase as breakflow accumulates. This increase continues until the SG secondary level
reaches the high-2 narrow-range level and isolates the CVCS pump. With continued RCS
cooldown and depressurization provided by the PRHR system, primary pressure decreases to
match the secondary pressure. At about 6.70 hours after the transient, the breakflow
terminates, and the system reaches a stable condition. The analysis shows that the PRHR
can remove the core decay heat and prevent the unaffected PORV from opening. During the
transient, the CMTs remain full, ADS actuation does not occur, and the SG does not overfill with
water.
During an SGTR, the RCS depressurizes as a result of the primary-to-secondary leakage
through the ruptured SG tube. The depressurization reduces the calculated DNBRs. The
analysis shows that the depressurization before reactor trip for the SGTR is slower than for the
RCS depressurization events discussed in Section 15.2.6.1 of this report. Following a reactor
trip, the DNBR rapidly increases. Thus, the staffs conclusion for the event discussed in
Section 15.2.6.1 of this report also applies to the SGTR event, in that the calculated
DNBR remains above the safety limit.
15-42

For this analysis, the applicant used the LOFTTR2 computer code together with conservative
and acceptable assumptions to maximize the primary-to-secondary leakage. The results of the
analysis show that the SG will not overfill with water, the maximum RCS will not exceed
110 percent of design pressure, and the minimum DNBR will remain greater than the safety
DNBR limit. In addition, the analysis shows that the PRHR and CMTs can achieve LTC, and
the radiological releases will remain within the limits of 10 CFR 50.34(a)(1)(ii)(D)(1). The staff
finds that the SGTR analysis meets the acceptance criteria of SRP Section 15.6.3 with respect
to the pressure and core safety DNBR limits and, therefore, concludes that the analysis is
acceptable. Section 15.3 of this report discusses the staffs evaluation of the radiological
release.
15.2.6.4 Spectrum of Boiling-Water Reactor Steam System Piping Failure Outside
Containment (DCD Tier 2, Section 15.6.4)
This section of the DCD does not apply to the AP1000 design, which is a PWR design.
15.2.6.5 Loss-of-Coolant Accident (DCD Tier 2, Section 15.6.5)
In DCD Tier 2, Section 15.6.5, Westinghouse presents the LOCA analysis results. The
applicants analyses examine SBLOCAs, LBLOCAs, and post-LOCA LTC.
The applicants LOCA analyses meet the following acceptance criteria for the calculated ECCS
performance:
The calculated peak cladding temperature (PCT) is less than 1204 EC (2200 EF).
The calculated total oxidation of the cladding is within 0.17 times the total cladding
thickness before oxidation.
The calculated total amount of hydrogen generated is less than 0.01 times the
hypothetical amount that can be generated if all of the metal in the cladding cylinders
surrounding the fuel, excluding the cladding surrounding the plenum volume, reacts.
Any calculated changes in core geometry will be such that the core remains amenable
to cooling.
After any calculated successful initial operation of the ECCS, the calculated core
temperature will be maintained at an acceptably low value, and decay heat will
be removed for the extended time required by the long-lived radioactivity remaining in
the core.
These criteria are established to provide significant margin for ECCS performance following a
LOCA. The staff finds that these acceptance criteria are consistent with the requirements of
10 CFR 50.46(b)(1)(b)(5) for ECCS performance and, therefore, are acceptable.
15-43

15.2.6.5.1 Small Breaks
The AP1000 is designed to keep the reactor core cooled and covered with water by means of
passive safety systems which do not require the start and operation of pumps to provide
makeup water. No operator action is required to actuate and control the passive protective
systems. Active systems using pumps are also available for activation and control by the
operator in the event of an SBLOCA. However, the design-basis analysis of the AP1000 design
does not credit the operation of these active systems.
During an SBLOCA, the AP1000 reactor system will depressurize to the pressurizer lowpressure setpoint, initiating a reactor trip signal. With further reduction in reactor system
pressure, the pressurizer low-pressure setpoint will be reached to actuate an S signal. The
S signal causes the opening of valves in the discharge of the CMTs and PRHR. The CMTs
will immediately begin to circulate borated water into the RV downcomer by way of the direct
vessel injection (DVI) line. Water will also begin to circulate through the PRHR HX to ensure
decay heat removal. As the reactor system drains, the CMTs provide a source of water to
replenish that lost out of the break. The S signal will also trip the RCPs, which will retain
water in the lower elevations of the reactor system and around the core and will minimize the
loss of water from the break.
As the CMTs drain, signals are sent to the ADS valves to open in a prescribed sequence. The
first three stages of ADS, which are located at the top of the pressurizer, will begin to sequence
open when the CMT water volume drops to 67.5 percent. The ADS Stage 4 valve (ADS-4)
begins its open sequence when the CMT water volume drops to 20 percent.
The action of the break, PRHR, and ADS-1, 2, and 3 causes the reactor pressure to decrease.
When the pressure reaches approximately 4.83 MPa (700 psig), the accumulator tanks, which
contain borated water pressurized with nitrogen, will inject into the RV by way of the DVI lines.
Following actuation of the ADS-4, the reactor system pressure will approach that of the
containment, permitting borated water from the IRWST to flow by gravity into DVI lines and into
the RV.
In DCD Tier 2, Section 15.6.5.4B.2, Small-Break LOCA Analysis Methodology, the applicant
described the three elements of the AP1000 SBLOCA evaluation model as the NOTRUMP
computer code, the NOTRUMP homogeneous sensitivity model, and the CHF assessment
during accumulator injection. As described in the following discussion, the applicant applied the
three elements in the analyses for the AP1000 design. The results demonstrate that the design
of the AP1000 adequately mitigates the consequences of postulated design-basis SBLOCA
events. Section 21.6.2 of this report discusses the staffs evaluation and acceptance of the
NOTRUMP code and the SBLOCA evaluation model for the AP1000.
The applicant performed SBLOCA analyses using the NOTRUMP computer code for the period
immediately after the break occurs until IRWST flow is fully established to the reactor core.
After this time, the reactor is considered to be in LTC. Section 15.2.7 of this report discusses
the applicants analyses for LTC. The NRC staff conducted its evaluation of SBLOCA for the
AP1000 design in accordance with SRP Section 15.6.5 to ensure that the acceptance criteria
listed in 10 CFR 50.46 will not be exceeded.
15-44

NOTRUMP calculates the flow of steam and water in one dimension with variable nodalization.
The code considers thermodynamic nonequilibrium between the steam and water phases.
Code features include flow regime-dependent drift-flux calculations with counter-current
flooding limitations, mixture-level tracking logic in multiple-stacking fluid nodes, and regimedependent heat transfer correlations.
The NOTRUMP analyses were made conservative by assuming decay heat at 120 percent of
the ANS 5.1-1971 Standard, as required by Appendix K to 10 CFR Part 50. The single failure
of one of the four ADS-4 valves was assumed. This failure was determined to be the most
limiting for the AP1000 design. For most of the NOTRUMP analyses, the containment was
assumed to remain at atmospheric pressure. The use of atmospheric pressure maximizes
pressure losses out of the three ADS-4 valves assumed to remain operable, which delays the
time when sustained inventory injection to the RV from the IRWST can occur. DCD Tier 2,
Section 15.0.8, discusses the plant systems and equipment that are assumed to be available to
mitigate the effects of the event.
In the DCD, the applicant described SBLOCA analyses for the following cases:
the inadvertent opening of both 10.16-cm (4-in.) ADS-1 valves (atmospheric

containment pressure)
a cold-leg break of 5.08 cm (2 in.) equivalent diameter in the loop without the
pressurizer (atmospheric containment pressure)
the double-ended rupture of a DVI line (atmospheric containment pressure)
the double-ended rupture of a DVI line (138 kPa (20 psia) containment pressure)
a cold-leg break of 25.4 cm (10 in.) equivalent diameter (atmospheric containment

pressure)
NOTRUMP did not calculate any of these breaks to cause core uncovery or core heatup. For
the analysis of the 25.4-cm (10-in.) cold-leg break, NOTRUMP calculated the core to become
highly voided during the early part of the accident when the core stored energy was removed.
NOTRUMP does not have a detailed core heatup model for hot channel evaluation. To
evaluate the core heating that might occur, the applicant performed a conservative heatup
calculation in which that portion of the core that might experience critical heat flux was allowed
to heat adiabatically until the combined flow from the two accumulators reduced the core void
fraction. This calculation resulted in a PCT of 744 EC (1370 EF), which is much less than the
1204 EC (2200 EF) limit in 10 CFR 50.46. For break sizes larger than 25.4 cm (10 in.)
equivalent diameter, even more core voiding and core heatup would be expected. The large
break sizes, evaluated in Section 15.2.6.5.2 of this report, would bound the breaks discussed
above.
In the AP1000, the hot-legs enter the RV at a lower elevation than do the cold-legs. A small
break in a hot-leg might lead to a lower RV inventory than a break in a cold-leg of the same
size. In RAI 440.098, the staff requested the applicant to perform additional SBLOCA analyses,
15-45

including hot-leg breaks, for the AP1000 design. In response, the applicant provided the
NOTRUMP predictions for the following small breaks:
a 5.08-cm (2-in.) cold-leg break in the loop with the pressurizer (atmospheric
a 5.08-cm (2-in.) hot-leg break in the loop without the pressurizer (atmospheric
the double-ended break of a cold-leg pressure balance line to a CMT (atmospheric

None of these break sizes resulted in core uncovery. The staff concludes that the applicant has
evaluated a sufficient small-break spectrum.
The double-ended severance of a DVI line represents a limiting sequence for SBLOCA
analysis, because the water from one of the two accumulators, one of the two CMTs, and one
of the two IRWST injection lines would not reach the RV, but would spill into the containment.
For this reason, the staff concentrated much of the review effort on this postulated accident.
The staff has reservations on the ability of the NOTRUMP code to conservatively predict liquid
entrainment within the upper plenum, hot-legs, and ADS-4 valves. If too little liquid entrainment
were assumed, this liquid would be available to flow back into the core and provide unrealistic
core cooling, and depressurization of the reactor system by the ADS-4 would be artificially
enhanced. Both these effects would not be conservative for safety analysis. Westinghouse
addressed this concern by performing an analysis for the double-ended DVI line break in which
all liquid leaving the core was set at the same velocity as that of the steam (homogeneous
flow). Using the homogeneous flow assumption, all liquid which reached the upper plenum
would be swept out toward the ADS-4. The homogeneous analysis did not predict significant
core uncovery. However, it predicted a lower minimum core water mass, compared to the
nonhomogeneous case. The homogeneous analysis for a postulated double-ended DVI line
break is part of the small break evaluation model for the AP1000 design.
As part of the validation of NOTRUMP, as discussed in Section 21.6.2 of this report,
Westinghouse compared NOTRUMP predications with test data from the Advanced Plant
Experiment (APEX)-1000 test facility. For most phenomena, the code compared well with the
test data. However, the code was found to be nonconservative for prediction of water in the
core in the early part of the tests simulating double-ended DVI line breaks. For this reason, the
applicant performed an analysis of the double-ended DVI line break for the AP1000 using the
Chang critical heat flux correlation (Chang, S.H., et al., A Study of Critical Heat Flux for Low
Flow of Water in Vertical Round Tubes Under Low Pressure, Nuclear Engineering and Design,
July 1991). The Chang correlation analysis demonstrated that the core will remain cooled
during this period. Use of the Chang correlation to demonstrate core cooling for the doubleended DVI line break is part of the small break evaluation model for the AP1000 design.
In performing the double-ended DVI break analyses (including the homogeneous assumption
analysis), the applicant took credit for the ADS-4 and the elevated containment pressure that
15-46

would exist as a result of the energy added to the containment atmosphere by the break. Use
of an elevated pressure increases the relieving capacity of the ADS-4 and shortens the time
before IRWST injection begins. The applicant believes that 137.9 kPa (20 psia) is the minimum
pressure that will occur within the containment building following the occurrence of a doubleended DVI line break during the time period covered by the NOTRUMP analysis. In calculating
the minimum containment pressure, the applicant used the WGOTHIC code, which the NRC
staff has reviewed, as discussed in Section 21.6.5 of this report. After discussions with the
NRC staff, the applicant made a series of conservative assumptions for computing low
containment pressures. The AP1000 minimum pressure and LTC models incorporate the
following conservative assumptions, with the detailed maximum pressure model as the
reference point:
The containment net volume was increased by a factor of 1.1.
The containment shell and the passive containment cooling system (PCS) heat structure
areas were increased by a factor of 1.1.
The remaining heat structure areas were increased by a factor of 2.1.
The Uchida correlation with a multiplier of 1.2 was used for passive heat structures
(non-PCS structures) throughout the accident.
The PCS heat and mass transfer correlation multipliers were appropriately biased to
account for the uncertainty in the experiential database, and forced convection was not
included on the PCS inner surface.
Heat transfer in dead-ended compartments below the operating deck was not turned off
at the end of blowdown.
The air gap between the steel and the concrete was reduced to zero, from the 20-mil
thickness used in the maximum pressure calculation.
The material properties for steel, concrete, air, and the inorganic zinc coating were
biased high for conservatism.
Heat transfer credit for the PCS was set to start at the beginning of the accident, earlier
than was assumed for the maximum pressure calculation.
Westinghouse maintained its treatment of ECCS spillage as implemented in 1979 (with

the acceptance of the WGOTHIC breakflow model for this evaluation).
The containment purge system was assumed to be operating at the start of the accident
and isolated on a high-pressure signal.
The initial and boundary conditions for the containment, the PCS water, and the
environment were set to minimize the calculated pressure.
15-47
The operators were assumed to actuate the non-safety-related air coolers 10 minutes
after the break occurred.
Westinghouse used similar assumptions to compute the minimum containment pressures for
ECCS evaluation of operating plants. The NRC staff reviewed the minimum back pressure
calculation and concludes that they are an acceptable adaptation of the guidelines of SRP
Section 6.2.1.5, Minimum Containment Pressure Analysis for Emergency Core Cooling System
Performance Studies, and that the results are acceptable for use in the ECCS evaluation for
the AP1000 design.
Although the core was predicted to remain covered following a double-ended DVI line break,
even if the containment pressure is maintained at 101.4 kPa (14.7 psia) as a boundary
condition for the NOTRUMP calculation, the applicant considered the 137.9 kPa (20 psia)
containment back pressure boundary condition case to be the base case for the design basis.
The LTC analysis for the design-basis double-ended DVI line break using WCOBRA/TRAC is
initialized from the NOTRUMP analysis using the 137.9 kPa (20 psia) back pressure.
As an additional check on the NOTRUMP results obtained by the applicant, the staff performed
a series of audit calculations of SBLOCAs for the AP1000 using the RELAP5 computer code.
The staff developed RELAP5, an advanced T-H simulation tool. The RELAP5 analyses used
conservative assumptions similar to those used by the applicant in the NOTRUMP analyses.
Decay heat was set at 120 percent of the ANS-5.1-1973 Standard. The ANS-5.1-1973 decay
heat standard is equivalent to the ANS-5.1-1971 standard used in the NOTRUMP analyses.
The single failure of one of the four ADS-4 valves was assumed. The containment was
assumed to remain at atmospheric pressure. The core model in the RELAP5 analyses is
somewhat more detailed than the NOTRUMP core model, in that a hot rod is modeled with a
higher heat flux than the average core. The increased heat flux of the hot rod allows for the
assessment of the possibility of fuel cladding heatup following a DNB condition or core
uncovery event.
The staff performed audit calculations for the following cases:
the inadvertent opening of both 10.16-cm (4-in.) ADS-1 valves
a cold-leg break of 5.08-cm (2-in.) equivalent diameter in the loop without the
pressurizer
a cold-leg break of 8.89-cm (3.5-in.) equivalent diameter in the loop without the
pressurizer
a hot-leg break of 8.89-cm (3.5-in.) equivalent diameter in the loop with the pressurizer
the double-ended rupture of a DVI line
a cold-leg break of 25.4-cm (10-in.) equivalent diameter
15-48

None of the breaks analyzed by the staff using RELAP5 resulted in core uncovery or cladding
heatup. RELAP5 calculated approximately the same minimum core water mass for all break
sizes. However, the analysis predicted slightly less core water mass for the double-ended DVI
line break than for the other breaks.
Operating PWRs do not have ADS valves or a PRHR to depressurize and cool the RCS
following a LOCA. Operating PWRs must cool and depressurize the reactor system using the
SGs to remove decay heat. Under this scenario, it has been postulated that water from steam
condensation within the SG tubes might flow into the lower cold-leg elevations and into the
core. The water derived from steam condensation would not be borated, so that the entry of
this water into the core might cause an increase in core power. The NRC staff does not believe
deboration will occur during an SBLOCA for the AP1000 design. Section 15.2.8 of this report
evaluates this issue.
Based on the foregoing considerations, the staff concludes that the applicants analyses for a
spectrum of small piping breaks in the reactor pressure boundary are acceptable and meet the
requirements of 10 CFR 50.46 and Appendix K to 10 CFR Part 50, and that the calculated
performance of the passive ECCS following a postulated SBLOCA is acceptable.
15.2.6.5.2 Large Breaks (DCD Tier 2, Section 15.6.5.4A.8)
The applicant performed the LBLOCA analyses using the WCOBRA/TRAC code as
documented in WCAP-12945. WCOBRA/TRAC is the Westinghouses BE T-H computer code
used to calculate T-H conditions in the reactor system during blowdown and reflood of a
postulated LBLOCA. This code consists of the BE features needed to satisfy the requirements
of 10 CFR 50.46(a)(1)(i) for a realistic code.
In addition, the applicant used WCOBRA/TRAC to analyze the post-LOCA LTC of the AP1000,
including decay heat assumptions from 10 CFR Part 50, Appendix K. Section 15.2.7 of this
report discusses the staffs evaluation and conclusions for the LTC.
The applicant used the WCOBRA/TRAC code to perform the LBLOCA analysis. DCD Tier 2,
Table 15.6.5-4, lists the initial plant physical configuration, power-related parameters, initial fluid
conditions, and RCS boundary conditions used to determine the most limiting break size.
These initial conditions are determined from the applicants sensitivity study of the worst-case
set of combinations that result in the highest limiting calculated PCT. To determine the limiting
break case, the applicant performed parametric studies for the PCT with respect to bounding
initial conditions and associated uncertainties using the methods described in WCAP-14171, to
calculate the 95th percentile PCT. The results of the analysis show that the double-ended coldleg guillotine break results in a maximum PCT and is the limiting case. In all cases analyzed,
the bounding core design values of Fq = 2.60 and FdH = 1.65 are applied to the hot rod, at
102 percent of nominal core power. Finally, it was noted that the search for the limiting
LBLOCA included the hot-leg break and the cold-leg limiting split break.
The applicant considered the plant systems and equipment that are available to mitigate the
effects of the accident, as discussed in its response to RAI 440.097, Revision 1, and identified
the limiting single failure as a failure of one CMT discharge valve to open. In modeling the
15-49

CMTs and accumulators, the applicant minimized the capability to add borated water by
assuming the failure of one CMT discharge valve to reflect the limiting single failure.
The applicant presented the results of the LBLOCA analyses in DCD Tier 2, Tables 15.6.5-5
through 15.6.5-8, and Figures 15.6.5.4A-1 through 15.6.5.4A-12. The applicant submitted
additional information in its response to RAI 440.097, Revision 1, Table 15.6.5-8, and
Revision 1 response to RAI 440.097, Figures 440.097R1-1 through 440.097R1-3. Following an
LBLOCA, the reactor trip actuates on a low-pressurizer pressure trip signal. The LBLOCA
analysis does not credit the insertion of the control rods. Within a few seconds after the
initiation of an LBLOCA, an S signal actuates on the containment high-2 pressure. As a
result, after appropriate delays, the PRHR and CMT isolation valves open, and containment
isolation occurs. The rapid depressurization of the RCS during an LBLOCA leads to the
initiation of accumulator injection early in the transient. The accumulator flow reduces CMT
delivery to the degree that the CMT level does not reach the ADS Stage 1 valve actuation
setpoint until after the accumulator tank empties, following completion of the blowdown phase.
The applicants calculations continue until the fuel rods are quenched.
The applicant used the WCOBRA/TRAC models to perform the LBLOCA analyses with
calculated PCT uncertainties derived from the effects of model-related parameters, while the
initial condition-related parameters used in the analyses are bounding and conservative.
The applicant addressed the limitations in WCOBRA/TRAC relating to the PCT for values
greater than 940 EC (1725 EF). Staff review of the sensitivity calculations required by the code
limitation indicates that the results reinforce the conservatism of the calculation.
15.2.6.5.2.1 Summary of the Large Break LOCA Analysis Results
As per 10 CFR 50.46, the AP1000 LBLOCA analysis shows a high level of probability that the
following criteria will be met:
C
The calculated PCT will not exceed 1204 EC (2200 EF).
The calculated maximum cladding oxidation will not exceed 0.17 percent of the total
cladding thickness before oxidation.
The calculated total amount of hydrogen generated from the chemical reaction of the
cladding with water or steam will not exceed 1 percent of the amount that would be
generated if the entire cladding metal surrounding the fuel (excluding the cladding
surrounding the plenum volume) were oxidized.
The calculated changes in core geometry are such that the core remains amenable to
cooling.
After successful initial operation of the ECCS system, the core temperature will be
maintained at an acceptably low value, and decay heat will be removed for the extended
period of time required by the long-lived radioactivity remaining in the core.
15-50

The calculated results of the AP1000 LBLOCA satisfy the requirements of 10 CFR 50.46 and,
therefore, are acceptable.
15.2.7 Post-LOCA Long-Term Cooling (DCD Tier 2, Section 15.6.5.4C)
This analysis establishes that (1) the core remains cooled for the duration of the LTC phase,
(2) the boron concentration in the core keeps the core noncritical, and (3) boron precipitation
will not obstruct core coolant flow.
15.2.7.1 Double-Ended Direct Vessel Injection Line Break
The applicant selected a double-ended guillotine break of the DVI line for the LTC evaluation.
The double-ended DVI (DEDVI) line break is the most limiting LTC case, in the sense that it
maximizes the decay heat generation rate and minimizes the cooling water injection head. The
case analyzed is the continuation of an SBLOCA. Maximum design resistance is applied for the
ADS-4 flowpath, the IRWST injection, and the containment recirculation. Failure of one of the
two ADS-4 valves in the PRHR loop is assumed. ADS-1, 2, and 3 valves are not modeled here,
because they do not practically impact depressurization with the ADS-4 valves opened.
The applicant used WCOBRA/TRAC for the LTC analysis. As described in Section 21.6.4 of
this report, the staff has evaluated WCOBRA/TRAC and found it to be acceptable for the
AP1000 LTC analysis. The LTC analysis uses a detailed nodalization model, which includes
4 radial channels and 17 axial nodes in each channel to represent the AP1000 core.
Initial conditions of the RCS liquid inventory and temperatures are taken from the NOTRUMP
DEDVI case to initiate the calculation shortly after IRWST injection begins, and it proceeds until
achievement of a quasi-steady state. At this time, the calculated results are independent of the
initial conditions. The WCOBRA/TRAC calculation proceeds using boundary conditions from a
corresponding WGOTHIC analysis and is carried to 10,000 seconds, until establishment of a
quasi-steady-state sump recirculation. The minimum containment flood level for this LTC
transient is 32.86 m (107.8 ft)(from the plant reference elevation), which is sufficient to inject
coolant into the RV through the broken DVI line. Likewise, the IRWST also provides sufficient
head to inject water through the intact DVI nozzle.
In the downcomer, the water level is about 5.49 m (18 ft)(from the bottom of the core), while the
core collapsed level is about 2.44 m (8 ft). Boiling in the core produces steam and a two-phase
mixture, which flows into the upper plenum. The boiling process, coupled with the changes in
the steam quality exiting the ADS-4 lines, causes pressure variations, which in turn cause liquid
flow variations entering the bottom of the core. This also results in liquid and vapor flow
variations at the top of the core. The void fraction at the uppermost two nodes is about 0.8.
Two-phase mixture continuously flows into the hot-legs, which, on average, have a collapsed
liquid level of more than 50 percent. The exit flow rates through the pressurizer- and
nonpressurizer-side ADS-4 valves are about 22.68 kg/s and 45.36 kg/s (50 lbm/s and
100 lbm/s), respectively. The broken DVI line indicates a small outward flow at the beginning of
the transient, which reverses and reaches about 22.68 kg/s (50 lbm/s) at the quasi-steady
state. The intact DVI line injects about 77.11 kg/s (170 lbm/s) at the start of the transient and
diminishes to about 29.48 kg/s (65 lbm/s) at the quasi-steady state. The WCOBRA/TRAC
15-51

analysis demonstrated that the void fraction in the top node is 0.8, and therefore the core is
wetted. The peak clad temperature for the hot rod follows saturation temperature within
-12.2 EC (10 EF), indicating that core uncovery and temperature excursion do not occur during
LTC.
A variant of this transient for a containment pressure of 101.4 kPa (14.7 psia) was calculated
using the window method and starting at 6500 seconds. The lower containment pressure
creates a higher volume of steam for the same decay heat. The reduction in containment
pressure coincides with the switchover to containment recirculation. The results also show that
the core remains covered, with PCTs remaining very near the saturation level.
15.2.7.2 DEDVI Break and Wall-to-Wall FloodupContainment Recirculation
The case of DEDVI break and wall-to-wall floodup with containment recirculation is a more
limiting case, in that it assumes that all of the spaces beneath the sump level are flooded,
corresponding to the minimum possible water level and the minimum injection head through the
broken DVI line. Flooding of the dry compartments is conservatively estimated to take 14 days.
The decay heat source is estimated for this time. DCD Tier 2, Section 15.6.5.4C.2, describes
the calculations for the initial conditions for the window at floodup conditions, including the
assumptions for an ADS-4 failure and 101.4 kPa (14.7 psia) containment pressure.
This LTC calculation was performed using the window mode methodology. The calculation was
performed with the continuous mode until establishment of a quasi-steady state at about
400 seconds of transient time. The downcomer collapsed liquid level varies from 7.0 m to
7.6 m (23 ft to 25 ft), while the collapsed liquid level in the vessel is about 4 m (13 ft). Pressure
spikes from the boiling process and changes in steam quality out of the ADS-4 lines cause
injection liquid flow rate variations through the broken and intact DVI lines of an average value
of approximately 22.68 kg/s (50 lbm/s). The top two nodes have average void fractions of
about 0.8 and 0.7, respectively. The collapsed liquid level in the hot-leg is about .5 m (1.7 ft),
while the peak clad temperature of the hot rod remains close to the saturation temperature.
The vapor mass flow rate out of the core is about 1.6 kg/s (3.5 lbm/s), and the liquid flow rate is
about 45.36 kg/s (100 lbm/s). This liquid flow out of the vessel is more than sufficient
to remove excess boron. As described in Section 15.2.7.6 of this report, the staff also
performed an independent evaluation of the AP1000 LTC behavior, which concludes that the
boric acid concentration will not reach precipitation limits. In summary, the wall-to-wall flooding
case with the lowest injection head keeps the core cooled and provides more than adequate
liquid flow to preclude boron concentration in the core.
15.2.7.3 Post Accident Boron Concentration
Westinghouse performed an evaluation of the potential for boron concentration to build up in
the core following a LOCA and during LTC. The evaluation considered short- and long-term
intervals which correspond to the times before and after ADS actuation. In the short term, the
time is relatively limited and continuous flow exists. For boron to concentrate in the vessel,
significant amounts of borated water must be dumped into the vessel. This takes place when
the cold-legs void, and the CMTs begin to inject. At a CMT level of 20 percent, the ADS-4
actuates. In this time interval, favorable conditions for boron concentration do not occur. The
15-52

applicants calculations show that a 2-in. break LOCA requires less than 16 minutes until the
ADS is actuated. For larger breaks, the time is shorter. At the decay heat levels of the
AP1000, the buildup of significant concentration requires more than 3 hours. In the long-term
interval, once the ADS-4 is actuated, liquid outflow limits the boron concentration in the core.
These calculations indicate that ADS-4 vent quality at the initiation of recirculation is about
50 percent and decreases to less than 10 percent at the time of wall-to-wall flooding. At the
maximum vent quality, boron concentration is about 7400 ppm. The maximum boron solubility
temperature is 14.4 EC (58 EF) (at 7400 ppm). The vessel is not expected to reach this
temperature. Early in the LTC phase, with high decay heat, the high vent flow velocities can
support an annular flow regime that moves water up and out the ADS-4 vent. In this case, a
larger amount of water is expelled, creating a lower boron concentration in the core and a
solubility temperature lower than 14.4 EC (58 EF). In summary, both physical considerations
and calculation results indicate that the boron concentration in the core will not reach
precipitation limits. The results of the AP1000 LTC phase are acceptable because boric acid
precipitation in the core is precluded, preventing both criticality and/or flow blockage.
Operator Actions
The water level in the AP1000 sump varies from 33.3 m (109.3 ft)(from the plant reference
elevation) at the beginning of recirculation for a non-DVI LOCA, to 31.5 m (103.5 ft) for a DVI
LOCA and wall-to-wall flooding. Westinghouse stated that, during recirculation, the operators
will be instructed to maintain the sump water level at or above the 32.6 m (107 ft) level. This
practice adds a measure of assurance that adequate recirculation hydraulic head will exist
during the LTC phase. Therefore, the core will remain covered with a two-phase mixture, and
sufficient liquid entrainment will occur to ensure that boric acid is maintained well below
precipitation limits.
Summary and ConclusionLTC Boron Precipitation
Using WCOBRA/TRAC, Westinghouse analyzed the AP1000 LTC for the limiting case (i.e.,
highest decay heat level) DVI line break LOCA. Westinghouse also analyzed a variation of this
case, which results in the lowest sump injection head through the broken DVI line (the wall-towall floodup case). The results show that sufficient ventflow (of medium- to low-quality steam)
through the ADS-4 exists to remove enough water from the core to keep the boron
concentration below 7400 ppm. The corresponding boron precipitation temperature is about
14.4 EC (58 EF), which is virtually unattainable in the vessel or the ADS-4 vent pipe. For the
floodup case, the minimum sump injection head (through the broken DVI line) is adequate to
maintain cooling and limit boron concentration. The applicant used an acceptable code to
perform the analysis, and the results showed that the core boron concentration during LTC will
not allow the core to become critical, nor will boron precipitation obstruct core coolant flow.
15.2.7.4 Additional Calculations
In Appendix H to WCAP-15644-P, Westinghouse provided additional analysis to (1) support the
conclusion above regarding LTC boron precipitation and (2) to address a staff question
regarding the instances with low decay heat that may not generate sufficient steam to support
15-53

boron removal by expulsion of liquid through the ADS-4 valves. This section reviews this
analysis.
The study consists of hand calculations and is based on a number of bounding assumptions.
For the short term (as defined in Section 15.2.7.3 of this report), this study extends the size of
the LOCA break to less than 5 cm (2 in.), with maximum makeup from the BAT. In such a
case, the ADS may not actuate, and the plant would remain in this condition for some length of
time. The hot-leg voids in about an hour, during which the CMT injects and increases the boron
concentration. The BAT continues to inject borated water until it empties. Under those
conditions, the maximum concentration is about 39,000 ppm, which could be reached in about
7 hours. In this case, the operator will likely be able to cool down and depressurize the RCS
and avoid ADS actuation. The RCS temperature will remain well above the boron solubility
temperature of about 87.8 EC (190 EF) (39,000 ppm) until the RNS cut-in temperature of
176.7 EC (350 EF) is reached. The RNS shutdown cooling operation would promote the RCS
boron mixing. On the other hand, should the failure of the operator to shut the plant down
actuate the ADS, injection from the IRWST will enter to dilute the boron concentration. In either
case, the AP1000 emergency response guidelines (ERGs) require that the RCS be sampled for
boron concentration to assure that sufficient shutdown margin exists and to prevent excessive
concentration.
For the long term, after ADS-4 activation, boron concentration depends on the amount of water
exiting from the ADS-4. A simplified model was developed, which considered the possibility of
100-percent quality steam through the ADS-4, or varying steam quality until the pressure drop
through the system equals the driving injection head. In the first case, the hot water level will
increase to the point where slugflow will be established. In the second case, the steam quality
is sufficiently low so as to maintain an acceptable boron concentration. In addition, the method
was used to estimate the capability of the core with reduced decay heat to expel water during
LTC through the ADS-4. Such instances include (1) a very low core decay heat generation rate
during the very long term and (2) the core decay heat immediately after refueling.
According to these results, with an initial core water level of 33.2 m (109 ft) and reasonable
operator action to restore the sump level, the LTC phase can be maintained for a significantly
long period of time (Westinghouse estimates this duration to be 6 years). Eventually, the decay
heat cannot sustain steam velocity or quality to expel water. On the other end (i.e., a very short
irradiation history, for which decay heat diminishes rapidly), a very small probability exists that a
LOCA will occur during the time the reactor core has been operational. However, if a fresh core
experiences a LOCA, the calculation indicates that the operator has adequate time to intervene
and add water to restore the sump water level.
In summary, the auxiliary simplified calculations support the conclusions derived with
WCOBRA/TRAC and extend the conclusion into low decay heat cases. The staff finds that
emergency operating procedures and operator action can assure LTC for extended periods of
time.
15-54

15.2.7.5 Closure of Open Item 15.2.7-1
In Revision 2 to the AP1000 DCD, the WCOBRA/TRAC LTC analysis was performed with the
same noding as in the AP600 LTC analysis, which had only two axial nodes to represent the
reactor core. However, the increased decay heat level and the increased core height to 4.3 m
(14 ft) in the AP1000 design prompted the staff to request an analysis with a nodalization of
sufficient detail to capture the spatial variation in the core void distribution. In the DSER, the
staff identified Open Item 15.2.7-1, which stated that the applicant should provide more nodes
in the vessel axial void distribution analysis during LTC to demonstrate that no possibility of
core uncovery and adiabatic fuel heatup exists.
In the later revisions to the DCD, the applicant performed the LTC analysis, described in DCD
Tier 2, Section 15.6.5.4C, based on a detailed nodalization model with 4 radial channels and
17 axial nodes to represent the reactor core. The LTC analysis results with this more detailed
noding scheme were the subject of the preceding review and show that no long-term core
uncovery occurred. The staff concludes that Open Item 15.1.7-1 is closed.
15.2.7.6 Staff Independent Post-LOCA Long-Term Cooling Calculation
This section describes the staffs independent calculation of the post-LOCA LTC behavior in the
RCS to address the AP1000 system performance to maintain core cooling and preclude boron
precipitation. During the LTC phase, the RCS will boil for extended periods of time, causing the
boric acid concentration to increase in the vessel core and upper plenum regions.
Unlike conventional nuclear steam supply systems, the AP1000 does not rely primarily on
operator action to control boric acid and prevent precipitation. Because of ADS-4 actuation, the
AP1000 design ensures that boric acid is flushed from the core and RCS during the long term,
because of the sustained entrainment of liquid from the upper plenum and hot-legs that is
expelled through the ADS-4 piping to the containment. In order to evaluate LTC system
performance, the staff developed several methodologies to show the inherent margin in the
AP1000 that ensures a continued core cooling and flushing of the boric acid from the vessel
during the post-LOCA LTC phase. The following discusses LTC behavior, models, and
analytical results dealing with the prevention of boric acid precipitation in the AP1000.
LTC and the prevention of boric acid precipitation is addressed for all break sizes, including the
double-ended DVI line break, which results in the earliest draintime for the IRWST and, hence,
the earliest start of recirculation from the sump (i.e., 2.4 hours). Because this break results in
the earliest start of recirculation, decay heat will be the highest, maximizing the pressure drop
across the ADS-4 lines. Furthermore, because liquid must always be expelled out the ADS-4
lines to control boric acid, it is also necessary to show that the two-phase level remains above
the centerline of the hot-leg. With the two-phase level above the centerline of the hot-leg, the
steam will entrain liquid from the hot-leg piping and carry it out the ADS-4 lines as long as
steaming rates are 6.8 kg/s (15 lb/s) or higher. Moreover, as the evaluation below
demonstrates, once the steaming rate drops below 6.8 kg/s (15 lb/s), the pressure drop across
the ADS-4 decreases so that the two-phase level rises to very near the top of the hot-leg. In
this condition, a two-phase mixture is expected to spill from the RCS to the containment from
the ADS-4 lines, continuing the removal of boric acid from the RCS. The steaming rate of
15-55

6.8 kg/s (15 lb/s) occurs after about 14 days following a reactor trip from full-power operating
conditions, assuming no subcooling of the entering sump injection.
To show that liquid is expelled from the ADS-4 lines, a model will be described that
demonstrates that the fluid levels in the vessel remain in the top half of the hot-leg. This
condition is necessary during the long term to assure that boric acid is flushed from the system.
That is, it will also be shown that sufficient liquid is expelled to demonstrate that more liquid
concentrate is removed than is accumulated in the core and upper plenum during extended
boiling periods lasting for upwards of 14 days.
15.2.7.6.1 Long-Term Core Cooling Evaluation
To evaluate the AP1000 long-term core cooling behavior, the staff developed a T-H model,
which consists of computing (1) the pressure drop from the upper plenum through the hot-legs
and the ADS-4 lines to the containment when single phase steam exits the RCS (maximized
pressure drop for a given mass flow rate), and (2) the hydrostatic differential head of water
between the sump/downcomer and the inner vessel containing the core and upper plenum.
The computation of this hydrostatic head differential assumes that the two-phase level in the
vessel always remains at the centerline elevation of the hot-legs. This head differential is then
compared to the pressure drop across the ADS-4 lines for a range of decay heat levels, down
to and including conditions 14 days after initiation of the LOCA. Thus, if the pressure drop
across the ADS-4 lines is less than or equal to the hydrostatic head differential between the
sump/downcomer and inner vessel, then there is assurance that the two-phase level will
always remain in the top half of the hot-leg, thereby assuring entrainment of liquid and the
removal of boric acid from the system.
The staff evaluation made the following bounding assumptions:
The containment pressure is 101.4 kPa (14.7 psia).
The water in the sump and downcomer is saturated at 101.4 kPa (14.7 psia).
The sump water level is at Elevation 31.4 m (103 ft) at the start of recirculation, which is
the minimum level or elevation calculated by Westinghouse to be at Elevation 31.5 m
(103'-5") at 14 days.
Recirculation begins at 8640 seconds (0.104 days).
Decay heat generation is based on the ANS-5.1-1971 standard increased by 20 percent
The staff used a drift-flux model to compute the void distribution in the core and upper plenum
regions of the inner vessel. Appendix A to technical report, ISL-NSAD-NRC-01-003,
Preliminary Results of the AP1000 RELAP5/MOD3.3 Analysis for the Two-Inch Cold Leg and
Main Steam Line Breaks, V. Palazov and L. Ward, Information Systems Laboratories, Inc.,
August 2001, describes this model in detail. This model was benchmarked against the thermalhydraulic test facility (THTF) bundle uncovery tests and was validated against the ACHILLES
low-pressure two-phase-level swell tests, as well as the THETIS boiloff tests, also performed at
15-56

low pressure. The benchmarks showed that the drift-flux model predicted the void distribution
(swelled level) within the experimental uncertainty for the tests.
To compute the pressure drop across the ADS-4 lines from the upper plenum at the two-phase
surface, the staff developed an additional methodology. This model consisted of a
simultaneous solution of the mass, energy, and momentum equations employing a volumeflowpath-network arrangement using a semi-implicit numerical scheme. The region from the
upper plenum through each of the hot-legs and the ADS-4 lines was modeled as a series of
parallel volumes and connecting flowpaths or junctions. The solution to the conservation
equations produced a simultaneous solution of both the system pressure distribution and
attendant mass flow rates. The equation for the time rate of change of pressure at each
volume was derived from the definition of specific volume of the fluid, the energy equation, and
the equation of state. The momentum equation contained the inertia term, the static pressure
drop between cells, the geometric and friction pressure drop, and the recoverable momentum
flux acceleration pressure losses. The time rate of change equations for pressure and
momentum form a coupled set of differential equations. Westinghouse provided the geometric
K-factors used to represent the piping and components from the upper plenum through the
ADS-4 lines to the containment. The staff model increased the frictional pressure losses by
10 percent. In addition, the least-resistant ADS-4 line was conservatively assumed to fail
closed.
The model was applied to the pressure drop data and sample problems in Crane (Flow of
Fluids Through Valves, Fittings, and Pipes, Crane Co., Technical Paper No. 14, 1988) and
shown to reproduce the pressure drop for steam flowing in pipes with various geometric and
frictional pressure losses. The model was also compared to the data of J.K. Ferrel, et al.,
(Ferrel, J.K., Two-Phase Flow through Abrupt Expansions and Contractions, TID-23394
(Volume 3), North Carolina State University, June 1966) and predicted the pressure loss across
contractions for a range of area ratios. Furthermore, this method was compared to a range of
pressures and ADS-4 steam flow rates, and it was also shown to reproduce the pressure loss
across the ADS-4 lines computed by Westinghouse using its detailed FLOAD4 network code.
Using the staff methodology described above, Figure 15.2.7.6 of this report compares the
calculated hydrostatic head differentials between the sump/downcomer and inner vessel,
covering the range of decay heat steaming rates from 25 kg/s to 5 kg/s (54.1 lb/s to 10.6 lb/s)
(1.4 h to 14 days). The start of recirculation is 8640 seconds (2.4 h), with a steaming rate in
the core of 21 kg/s (46.1 lb/s). Figure 15.2.7.6 of this report compares the head differential to
the pressure drop across the ADS-4 lines over these decay heat generation rates. The twophase level would be at the hot-leg centerline elevation only when the pressure loss across the
ADS-4 lines equals the differential head of water between the sump/downcomer and inner
vessel region. Clearly, when the ADS-4 pressure loss exceeds the differential head curve in
Figure 15.2.7.6 of this report, the pressure loss across the ADS-4 lines is high enough to
depress the two-phase level below the centerline of the hot-leg. In Figure 15.2.7.6 of this
report, the vertical line indicates this condition so that, for steaming rates of about 23 kg/s
(50 lb/s) and higher, the two-phase level will be depressed below the hot-leg centerline if only
pure steam (quality = 1.0) should exit the system. The region to the right of the vertical line
demonstrates that when only single vapor exits the ADS-4 lines, the pressure loss across the
ADS-4 lines is insufficient to depress the two-phase level below the centerline of the hot-leg.
15-57

This demonstrates that the two-phase level would be higher than the prescribed two-phase
level, which is always assumed to be at the hot-leg centerline. Because entrainment
will remove liquid from the hot-legs for steaming rates above 7 kg/s (15 lb/s) (approximately
14 days after reactor trip), liquid is expelled from the RCS for 14 days following a LOCA. In
fact, the difference between the upper and lower curve represents the excess margin in the
AP1000 design and can be interpreted to mean that the two-phase level will remain near the
top of the hot-leg from the earliest time for start of recirculation at 2.4 hours, through to decay
heat levels producing steaming rates down to 7 kg/s (15 lb/s), or 14 days after reactor trip.
Furthermore, if the minimum sump level computed by Westinghouse based on conservative
assumptions regarding containment performance is used, even more margin is available to
guarantee the expulsion of liquid from the RCS. Bounding containment calculations by
Westinghouse show that the sump level at 2.4 hours is at Elevation 32.9 m (108 ft) in the
containment (higher than the assumed Elevation 31.4 m (103 ft) in the staff analysis). Using
the sump level of 32.9 m (108 ft) at 2.4 hours which decreases to 32.6 m (107 ft) at 1 day, as
shown in Figure 15.2.7.6 of this report, at the earliest time of recirculation (2.4 h), a margin of
approximately 20.7-kPa (3-psi) in the pressure loss is required to depress the two-phase level
below the hot-leg centerline. This water is also subcooled, but the Figure 15.2.7.6 analyses did
not credit it. Subcooling would further increase the excess margin, as would higher
containment pressures, because at 2.4 hours the minimum containment pressure is 131 kPa
(19 psia). At these low pressures, an increase from the assumed 101.4 kPa (14.7 psia)
containment pressure to the minimum justified pressure of 131 kPa (19 psia) would further
reduce the pressure loss across the ADS-4 lines, since the pressure drop would decrease
because of the lower steam specific volume at the higher pressures. Clearly, excess margin
exists to cover uncertainty in the pressure losses in the ADS-4 lines (uncertainty in the CRANE
resistances and unknown variations in the line resistance set up possibly by the interactions
between two closely connected elbows or bends), as well as any additional uncertainty that may
exist in the line resistances connecting the sump to the downcomer. Figure 15.2.7.6 of this
report shows that, at the start of recirculation (2.4 h), a 20.7 kPa (3 psi) margin exists. This
estimate is conservative because it did not credit subcooling and neglected the benefit of the
higher containment pressure of 131 kPa (19 psia).
Lastly, this excess margin can also accommodate compressible effects. Crane shows that a
compressible factor of 1/0.87 should be applied to the pressure drop when the hot-leg is at
137.9 kPa (20 psia), and the steam flow rate through one hot-leg is 13.61 kg/s (30 lb/s). The
excessive margin shown in Figure 15.2.7.6 of this report more than sufficiently covers increases
in the pressure drop from compressibility effects.
In view of these results, sufficient margin in the AP1000 design exists to assure that liquid will
be expelled from the RCS through the ADS-4 lines for as long as 14 days following reactor trip.
In fact, because of the excess margin displayed in Figure 15.2.7.6 of this report, two-phase fluid
would be expected to spill from the RCS through the ADS-4 lines well beyond the 14-day period
illustrated in this graph. Even as late as 14 days, ERGs will require the operators to take
actions to secure the plant. These actions include raising the sump level to the elevation
between 33.2 m (109 ft) and 33.5 m (110 ft) by filling the containment sump with subcooled
water from the RNS. This action would ensure liquid expulsion from the RCS for an extended
period of time and would ensure that the steaming rate in the core would be terminated
15-58

because of the subcooling of the sump water and the very low decay heat generation beyond
14 days. Limiting the sump level to lower than 33.5 m (110 ft) will preclude submergence of the
ADS-4 valves, thus preventing core uncovery.
15.2.7.6.2 Boric Acid Precipitation Evaluation
The AP1000 does not require operator action to mechanically flush the boric acid from the
RCS. For larger and intermediate breaks, which result in the actuation of ADS-4, the RCS
entrains and expels sufficient liquid to maintain the boric acid content well below precipitation
limits. For very small breaks where the RCS pressure remains high for extended periods of
time to prevent ADS-4 actuation, ERGs will require the operators to depressurize the RCS or
refill the RCS with charging injection to disperse the boric acid throughout the primary system.
The staff evaluates both situations with regard to the boron precipitation concerns, beginning
with the large and intermediate break system response, followed by an assessment of the very
small breaks.
Larger and Intermediate Breaks
To assess boric acid buildup, the staff developed a simple model to compute the boric acid
accumulation in the vessel over time. The model describes the boric acid content in the core
and upper plenum as a function of time with and without the flushing of boron via ADS-4 liquid
entrainment.
Since the DEDVI line break produces the earliest IRWST draintime and, hence, the earliest
start of sump recirculation, this break is the limiting event and is used to investigate the buildup
of boric acid. The first staff calculation credits no core flushing or entrainment out the ADS-4
lines, so that a bounding calculation is performed to identify the earliest time to precipitation. In
essence, the no-flushing calculation is independent of break size because boric acid buildup
depends on decay heat. The result of the no-flushing assumption shows that boron
precipitation could occur about 5 hours into the event.
The second calculation assumes flushing through ADS-4 liquid entrainment. This analysis
includes the following bounding assumptions:
The decay heat generation is increased by 20 percent.

The mixing volume is based only on the core region.
The BAT injects at the maximum rate of 662.45 L/min (175 gpm).
The concentration of the BAT is 2.4 percent.
No entrainment or flushing is credited before IRWST drainage.
At the start of recirculation, only a 2.27 kg/s (5 lb/s) flushing flow is assumed (flow in
excess of the core boiloff rate).
The calculation credits ADS-4 liquid entrainment at the initiation of sump recirculation, which
occurs 2.4 hours into the event. At 2.4 hours or the start of recirculation, a flushing flow out the
ADS-4 lines of only 2.27 kg/s (5 lb/s) is assumed, which is the flow in excess of the decay heat
steaming rate of about 20.87 kg/s (46 lb/s). The 2.27-kg/s (5-lb/s) liquid flow out of ADS-4
valves is a very conservative bounding assumption, as the RELAP5 calculations of the DEDVI
15-59

line break show that the liquid flow out of the ADS-4 during the ADS-4 blowdown period and
LTC period well exceeds 23.13 kg/s (51 lb/s). The calculation used containment pressures,
sump levels, and sump temperatures from the bounding LTC analysis provided by
Westinghouse as boundary conditions for the determination of the liquid flow out the ADS-4
lines.
With this bounding assumption, the calculated boric acid concentration in the core over time
shows that at the start of recirculation, the excess flushing flow of 2.27 kg/s (5 lb/s) is sufficient
to reduce the boric acid concentration in the vessel. Furthermore, during the blowdown or
before IRWST injection, ADS-4 actuation would produce a liquid flow out the valves in excess
of the boiloff, which would act to reduce the boric acid concentration much earlier in this event.
Even with no credit for entrainment during blowdown, for conservatism, the boric acid
concentration does not reach the precipitation limit of 32 wt% (corresponding to the saturation
temperature at the minimum containment pressure of 137.9 kPa (20 psia) at the start of
recirculation). This calculation produces a higher boric acid concentration than that calculated
by Westinghouse. Calculations taking credit for a minimal flushing flow of 2.27 kg/s (5 lb/s)
when the ADS-4 valves first open during blowdown (i.e., about 2000 seconds) show that the
maximum concentration is 15.8 wt% (27,538 ppm), achieved shortly after the start of
recirculation. This result assumes saturated inlet conditions to maximize the boiloff rate and the
rate at which boron builds up in the core. When subcooling is credited at 48.9 EC (120 EF), the
concentration remains well below the 27,538 ppm concentration because of the decreased
boiloff rate in the core.
Based on this bounding analysis, the staff concludes that the boric acid concentration will not
reach precipitation limits following breaks in the RCS that produce ADS-4 actuation.
Very Small Breaks
In the event of a very small break with a diameter of 1.3 cm (0.5 in.) or less, the RCS could boil
for extended periods of time while remaining at elevated pressures and conditions, which would
delay actuation of the ADS-4 valves. Under these conditions, injection from the CMTs,
accumulators, and BAT could increase the boric acid concentration in the vessel. Since the
break is small with RCS pressures remaining high, the CMTs do not rapidly drain, delaying the
ADS-4 valve actuation. If boiling persists for several hours, concentrations could increase to
values that, although soluble at the high pressures and temperatures, could cause precipitation
should ADS-4 eventually actuate and rapidly depressurize the RCS. To investigate the time to
reach precipitation limits corresponding to RCS pressures at 137.9 kPa (20 psia), the staff
performed an analysis to determine the time required to increase the boric acid content in the
core to 32 wt%. Assuming the BAT injects with the boric acid content of 2.4 percent, about
3.5 hours of boiling would be sufficient to increase the boric acid content in the core to 32 wt%.
This assumption is considered to be bounding because the staff assumed saturated inlet
conditions and did not credit the fact that, as subcooled water enters the core, the boiloff rate
and the rate at which boric acid builds up in the core are significantly reduced.
As documented by Westinghouse, ERGs will instruct the operators to sample RCS boric acid
content and take actions to reduce the concentration of the injected boric acid or initiate an
early cooldown to prevent boric acid precipitation, in anticipation of ADS-4 actuation after
15-60

extended periods of boiling. The ERGs assure that precipitation limits are never exceeded
following very small breaks in the RCS. Such actions to control boric acid for the very small
breaks are directed after 1 hour, but no later than 3 hours.
15.2.8 Deboration during SBLOCAs
The staff reviewed the issue of boron dilution associated with the SBLOCA reflux condensation,
the so-called Finnish scenario. In response to RAI 440.099, Revision 1, the applicant
provided its evaluation, which concluded that the Finnish scenario is of no consequence to the
AP1000 reactor design because the AP1000 does not rely on SGs to cool the RCS during an
SBLOCA event. Consequently, the SGs should not generate any significant amount of
conservatively assumed boron-free condensate via reflux condensation over an extended
period of time during an SBLOCA event. The following describes the staff evaluation in detail.
During an SBLOCA, the SG functions as a heat source as the RCS depressurizes, rather than
a heat sink. Therefore, the differential temperature across the primary and secondary side of
the generators is such that steam from the reactor will not condense on the tubes.
The AP1000 PRHR HX becomes a dominant RCS heat sink following the generation of an
S signal during a postulated SBLOCA event. As such, the PRHR HX could become a
potential source for generating a volume of unborated coolant during an SBLOCA. Such a
scenario could lead to a reactivity excursion as a result of a restart of an RCP after the
unborated water slug had collected in the reactor coolant loop. The applicant had determined
that this scenario is not a concern for the AP1000 design for the following reasons. Specifically,
the AP1000 reactor coolant loop piping does not contain a loop seal, and thus no point exists to
collect a large slug of unborated condensate in the reactor coolant loop piping. During the
SBLOCA event, once subcooling in the RCS is lost, steam will enter the PRHR HX and will
condense on the inside of the PRHR HX tubes. Steam condensed in the PRHR is delivered to
the Loop 1 SG outlet plenum. The AP1000 loop layout does not contain an RCP crossover leg,
and the PRHR condensate will drain continuously from the SG channel head into the Loop 1
cold-legs and flow into the RV.
During the SBLOCA transient, the water in the cold-legs enters the downcomer, where it mixes
with the highly borated safety injection flow from either the accumulators, the CMTs, or both.
The relatively low flow rate of fluid from the downcomer into the core, during the post-RCP-trip
natural circulation phase of the AP1000 SBLOCA events, enables mixing to occur in the
downcomer and lower plenum. No unmixed slugs of unborated water from the PRHR can form
in the downcomer and enter the core during this scenario.
The applicant performed bounding calculations that demonstrated that it was not credible to
postulate that the boron concentration in the downcomer and lower plenum would be diluted to
a critical boron concentration for this postulated LOCA. The conclusions from these studies,
which show that boron dilution from the operation of the PRHR HX would not occur, were based
on demonstrating that the PRHR condensate would adequately mix with the water in the
downcomer and the lower plenum so that a critical boron concentration would not be reached.
15-61

The AP1000 uses a low boron core design with the boron concentration at the BOC of
approximately 1000 ppm. The low AP1000 core boron concentration significantly reduces the
potential for the PRHR to dilute the coolant in the RV to the point of criticality. Although the
AP1000 PRHR flow rate is high, the CMT flow rate, the RV downcomer, and the lower plenum
volume for the AP1000 are also high. Taking these differences into account, the AP1000
design studies show that post-LOCA boron dilution is not a concern, provided that good mixing
exists in the vessel. Analysis for the AP1000 showed that mixing in the RV downcomer and
lower plenum will counteract boron dilution in the core from PRHR operation.
NUREG/IA-0004, Thermal Mixing Tests in a Semiannular Downcomer with Interacting Flows
from Cold Legs, issued October 1986, reported the study of the mixing of high-pressure safety
injection water with primary coolant in a simulated PWR downcomer. Test #106 in
NUREG/IA-0004 considers a geometry which represents the PRHR condensate delivery
geometry into the AP1000 downcomer, namely equal flow rates of liquid entering the
downcomer through two cold-legs which are 90 degrees apart at the connection into the RV
vessel. The downcomer at the test facility is shorter in length (approximately 3.08 m (10 ft))
than the AP1000 dimension (approximately 6.16 m (20 ft) from the cold-leg bottom to the
bottom of the downcomer). The test facility, therefore, provides less than one-half of the mixing
length available in the AP1000 downcomer. The fluid velocity in the test facility cold-legs is
approximately 0.14 m/s (0.45 ft/s) for the simulated high-pressure injection (HPI) flow injection
in Test #106, as indicated by the C series of figures in NUREG/IA-0004. This is similar to the
velocity of the PRHR condensate in the cold-legs for SBLOCA scenarios. Therefore, the
parameters of Test #106 are such that the observed results provide meaningful insights into the
mixing that occurs in the AP1000 downcomer during the SBLOCA boron dilution scenarios.
The results of Test #106 illustrate that the injected plume thoroughly mixes with the resident
downcomer liquid during the 3.08-m (10-ft) fall to the bottom elevation. The Test #113 results
in NUREG/IA-0004 provide further support for AP1000 downcomer mixing.
Test #113 was run at a simulated HPI rate that is 3.6 times greater than that of Test #106, with
a 60 degree angle between the two cold-leg injection connections, as depicted in the D series
of photographs in NUREG/IA-0004. Test #113 results show mixing behavior in the downcomer
which closely resembles that of Test #106. Test #113 indicates that the sensitivity of
downcomer mixing to initial plume velocity is minor. These two tests provide compelling
evidence that the diluted boron stream in the AP1000 PRHR condensate delivery scenarios is
well mixed in the downcomer and that no unmixed slugs enter the lower plenum or core. These
test results provide additional independent technical justification that the degree of mixing which
occurs in the AP1000 downcomer during the PRHR condensate return scenarios is more than
adequate to disperse a plume of diluted boron liquid. The test results support the conclusion
that recriticality of the core is not of concern for SBLOCA scenarios.
Based on the information provided in the submittal of the AP1000 DCD, and on the analysis
performed by Westinghouse on behalf of this event, including the thermal tests mentioned in
NUREG/1A-0004, the staff finds that the analysis in support of the possible deboration from an
SBLOCA event is acceptable.
15-62

15.2.9 Anticipated Transients Without Scram (DCD Tier 2, Section 15.8)
An ATWS event is defined as an AOO (such as LONF, loss of condenser vacuum, or LOOP)
combined with an assumed failure of the RTS to shut down the reactor. On June 26, 1984, the
Commission amended the Code of Federal Regulations to include 10 CFR 50.62,
Requirements for Reduction of Risk from Anticipated Transients Without Scram (ATWS)
Events for Light-Water-Cooled Nuclear Power Plants (known as the ATWS Rule). This rule,
as amended on July 6, 1984; November 6, 1986; April 3, 1989; and July 29, 1996, requires
nuclear power plant facilities to reduce the likelihood of failure to shut down the reactor
following anticipated transients and to mitigate the consequences of ATWS events.
The equipment to be installed in accordance with the ATWS Rule must be diverse from the
existing RTS and must be capable of testing at power. This equipment will provide needed
diversity to reduce the potential for common-mode failures that result in an ATWS and lead to
unacceptable plant conditions.
For the PWRs manufactured by Westinghouse, 10 CFR 50.62(c)(1) specifies the basic
requirements of the ATWS Rule:
Each pressurized water reactor must have equipment from sensor output to final
actuation device, that is diverse from the reactor trip system, to automatically
initiate the auxiliary (or emergency) feedwater and initiate a turbine trip under
conditions indicative of an ATWS. This equipment must be designed to perform
its function in a reliable manner and be independent (from sensor output to the
final actuation device) from the existing reactor trip system.
The AP1000 design includes a control-grade diverse actuation system (DAS) to provide an
alternate turbine trip signal and an alternate actuation signal of the PRHR system for decay
heat removal, which are separate and diverse from the safety-grade RTS and PRHR normal
actuation signals. The DAS also provides a diverse scram function. Section 7.7 of this report
discusses the staffs review and acceptance of the applicants DAS design.
The AP1000 design relies on the PRHR in lieu of an auxiliary or emergency feedwater system
as its safety-related method for removing decay heat. In its letter, AP1000 Request for
Exemption, DCP/NRC-1534, dated December 3, 2002, the applicant submitted a request for
exemption from the part of the ATWS Rule, 10 CFR 50.62(c)(1), that requires auxiliary or
emergency feedwater as an alternate system for decay heat removal during an ATWS event.
The staff concludes that the applicant has met the intent of the ATWS Rule by relying on the
PRHR system to remove the decay heat and meets the underlying purpose of the rule.
Therefore, the Commission has determined that the special circumstances described in 10 CFR
50.12(a)(2)(ii) exist, in that the requirement for an auxiliary or emergency feedwater system is
not necessary to achieve the underlying purpose of 10 CFR 50.62(c)(1), because the applicant
adopted acceptable alternatives that accomplish the intent of this regulation, and the exemption
is authorized by law, will not present an undue risk to public health and safety, and is consistent
with the common defense and security.
15-63

The applicant provided the results of an ATWS analysis in AP1000 Probabilistic Risk
Assessment, Appendix A, Section A4, for the staff to review. In its analysis, the applicant used
a complete LONF event as an initial event for the ATWS analysis because the LONF event was
previously established as the limiting case (i.e., produced the maximum RCS pressure) for the
AP600.
The applicant performed the ATWS analysis with the LOFTRAN code. The AP1000 DAS is
credited to function in the analysis for ATWS cases. Specifically, the DAS is credited to actuate
a turbine trip and the PRHR when the low wide-range SG-level signal is generated. Two LONF
cases are analyzed, one for equilibrium cycle core (ECC) conditions and one for first cycle core
(FCC) conditions. The ECC case used an MTC of -12.5 pcm/EF, which is the least negative
MTC at any time in an ECC condition. The analysis shows that the peak RCS pressure is about
20.68 MPa (3000 psia), which is less than the limit of 22.06 MPa (3200 psia). For the FCC
case, the LONF analysis shows that with an MTC of -10 pcm/EF, the calculated peak RCS
reaches about 22.06 MPa (3200 psia). The MTC used in the analysis for the FCC case is less
than 60 percent of the cycle time. Therefore, the unfavorable exposure time (UET) for the FCC
case is 40 percent. The UET is the time during the fuel cycle when the reactivity feedback is
not sufficient to maintain pressure under 22.06 MPa (3200 psia) for a given reactor state. The
staff requested, in a followup question to RAI 440.014, that the applicant provide additional
justification to show that the LONF analysis is the worst case, and the selection of the MTC
used in the analysis is acceptable with respect to the acceptable UET limit.
In response, the applicant performed a probabilistic risk assessment (PRA) evaluation to
identify the frequency of the anticipated transients in DCD Tier 2, Chapter 15, and specify the
most risk-significant ATWS scenarios for the AP1000 design. Based on the results of the PRA
evaluation, the applicant identified the most risk-significant ATWS scenarios which make up
more than 95 percent of the AP1000 ATWS initiating event frequency. The applicant
performed ATWS analyses on the most risk-significant ATWS cases that, based on the results
of the DCD Tier 2, Chapter 15, non-LOCA analysis for the AP1000, may result in a significant
pressure increase during the transients. The applicant analyzed the following cases in detail to
identify the scenario that results in the least margin to the RCPB limit for the AP1000:
C
turbine trip without feedwater system operable with turbine bypass system operable
turbine trip with feedwater system operable with turbine bypass system operable
turbine trip without feedwater system operable without turbine bypass system operable
LONF event with turbine bypass system operable
LONF event without spray system, without turbine bypass system operable
LONF event with turbine bypass system operable, more realistic SG heat transfer model
complete loss of forced coolant flow with main feedwater system (MFWS) operable with
turbine bypass system operable
15-64

C
complete loss of forced coolant flow with MFWS operable without turbine bypass system
operable
complete loss of forced coolant flow induced by the loss of ac power, MFWS not
operable, no steam dump operable, turbine trip at the initiation of the transient
The results of the ATWS analysis for the above cases confirm that, for the AP1000 design, the
limiting case is the LONF event with the turbine bypass operable, resulting in the highest peak
RCS pressure.
In addressing the second concern in RAI 440.014, related to the acceptability of the MTC value
used in the ATWS analysis for the limiting case, the applicant has revised the DAS actuation
logic to improve its capability for accident mitigation in response to an ATWS event. In addition
to actuation of the PRHR and the turbine trip, the new logic actuates the CMT and RCP trip on
the low wide-range SG-level signal. Together with implementation of a new DAS logic, an
additional change has been implemented in the PLS, such that the PLS isolates the steam
dump system whenever the SG level drops below the low wide-range SG water-level setpoint.
DCD Tier 2, Section 7.1.1, includes the description of the new DAS logic.
Section 7.7.1 of this report discusses the review and acceptance of the new logic. The
applicant performed an ATWS analysis of the LONF event for the AP1000, with the new DAS
ATWS protection logic assuming an MTC of -5 pcm/EF. The value of the MTC envelops
100 percent of the AP1000 core life. The results of the ATWS show that for the limiting case,
the LONF event, the maximum calculated RCS pressure is 19.43 MPa (2818 psia), which is
within the acceptance limit of 22.06 MPa (3200 psia). The limiting ATWS case demonstrates
that the UET (i.e., the time during the fuel cycle when the reactivity feedback is not sufficient to
maintain pressure under 22.06 MPa (3200 psia) for a given reactor state) does not exist for the
AP1000 design. RAI 440.014, Revision 1, and AP1000 Probabilistic Risk Assessment,
Appendix A, Section A4, include the information on the new ATWS analysis discussed in this
section. Because (1) the ATWS analysis used the LOFTRAN code previously approved by the
NRC, (2) the limiting ATWS case is identified based on the results of the actual ATWS analysis
discussed in this section, (3) the value of the MTC used in the limiting case analysis envelops
100 percent of the AP1000 fuel cycles, and (4) the calculated peak RCS pressure for the
limiting case falls within the pressure limit acceptance criterion, the staff concludes that the
ATWS analysis is acceptable.
15.2.10 Conclusions
The staff has reviewed the safety analyses of the design-basis transients and accidents
described DCD Tier 2, Chapter 15, for the AP1000 design. Based on the evaluation discussed
above, the staff concludes that the AP1000 design meets the acceptance criteria for these
transients and accidents.
As discussed in Section 15.2.9 of this report, the staff concludes that the applicants request for
exemption to the ATWS Rule of 10 CFR 50.62 is acceptable. Specifically, the exemption
request applies to 10 CFR 50.62(c)(1), which requires auxiliary or emergency feedwater as an
alternate system for decay heat removal during an ATWS event. The AP1000 design relies on
15-65

the PRHR in lieu of an auxiliary or emergency feedwater system as its safety-related method
of removing decay heat. The staff concludes that the AP1000 design meets the underlying
purpose of the ATWS Rule by relying on the PRHR system to remove the decay heat. In
accordance with 10 CFR 50.12(a)(2)(ii) for the AP1000 design, the requirement for an
emergency feedwater system is not necessary to achieve the underlying purpose of 10 CFR
50.62(c)(1). Therefore, the exemption request is acceptable.
15.3 Radiological Consequences of Accidents

In DCD Tier 2, Chapter 15, the applicant performed radiological consequence assessments of
the following seven reactor design-basis accidents (DBAs), using the hypothetical set of
atmospheric relative concentration (dispersion) values ( /Q values) provided in DCD Tier 2,
Table 15A-5. Because all other aspects of the design are fixed, these /Q values help
determine the required minimum distances to the exclusion area boundary (EAB) and the
low-population zone (LPZ) for a given site in order to provide reasonable assurance that the
radiological consequences of a DBA will be within the dose limits specified in
10 CFR 50.34(a)(1)(ii)(D). The analyzed DBAs include the following:
C
MSLB outside containment (DCD Tier 2, Section 15.1.5)
RCP shaft seizure (locked rotor) (DCD Tier 2, Section 15.3.3)
RCCA ejection (DCD Tier 2, Section 15.4.8)
failure of small lines carrying primary coolant outside containment (DCD Tier 2,
Section 15.6.2)
SGTR (DCD Tier 2, Section 15.6.3)
LOCA (DCD Tier 2, Section 15.6.5)
fuel-handling accident (FHA) (DCD Tier 2, Section 15.7.4)
In DCD Tier 2, Chapter 15, the applicant concluded that the AP1000 design will provide
reasonable assurance that the radiological consequences resulting from any of the above DBAs
will fall within the offsite dose criterion of 0.25 Sv (25 rem) total effective dose equivalent
(TEDE), as specified in 10 CFR 50.34(a)(1)(ii)(D), and the control room operator dose criterion
of 0.05 Sv (5 rem), as specified in GDC 19, Control Room, of Appendix A to 10 CFR Part 50.
The applicant reached this conclusion by performing the following:
C
using reactor accident source terms based on NUREG-1465, Accident Source Terms
for Light-Water Nuclear Power Plants, and RG 1.183, Alternative Radiological Source
Terms for Evaluating Design Basis Accidents at Nuclear Power Reactors
relying on natural deposition of fission-product aerosol within the containment
15-66

C
controlling the pH of the water in the containment to prevent iodine evolution
using a set of hypothetical atmospheric dispersion factor ( /Q) values
The /Q values are the relative atmospheric concentrations of radiological releases at the
receptor point in terms of the rate of radioactivity release. In lieu of site-specific meteorological
data, the applicant provided a reference set of /Q values for the AP1000 design using
meteorological data that is expected to bound 70 to 80 percent of U.S. operating nuclear power
plant sites for offsite dispersion. In DCD Tier 2, Tables 2-1 and 15A-5 list the AP1000
hypothetical /Q values, and DCD Tier 2, Table 15A-6, lists the AP1000 hypothetical /Q
values for the control room.
Regulatory Evaluation
The staff evaluated the radiological consequences of DBAs against the dose criteria, specified
in 10 CFR 50.34(a)(1)(ii)(D), of 0.25 Sv (25 rem) TEDE at the EAB for any 2-hour period,
following the onset of the postulated fission product release, and 0.25 Sv (25 rem) TEDE at the
outer boundary of the LPZ for the duration of exposure to the release cloud. The staff used a
criterion of 0.05 Sv (5 rem) TEDE for evaluating the radiological consequences from DBAs in
the control room of the AP1000 design, pursuant to GDC 19 of Appendix A to 10 CFR Part 50.
The staff used applicable guidance in SRP Section 15.0.1, Radiological Consequence
Analyses Using Alternative Source Terms, and RG 1.183 in its review of the AP1000 DBA
radiological consequence analyses. RG 1.183 provides guidance on radiological consequence
analyses to those licensees of operating power reactors choosing to implement an alternative
source term pursuant to 10 CFR 50.67, which has the same regulatory dose criteria as those
specified in 10 CFR 50.34(a)(1)(ii)(D) (i.e., 0.25 Sv (25 rem) TEDE) and GDC 19 (i.e., 0.05 Sv
(5 rem) TEDE). Although RG 1.183 applies to the current operating power reactors, its
guidance on radiological acceptance criteria, formulation of the source term, and DBA modeling
is useful in the review of the AP1000 design because the AP1000 is an advanced PWR.
Technical Evaluation
The staff reviewed the radiological consequence analyses performed by Westinghouse using
the hypothetical /Q values given in DCD Tier 2, Tables 15A-5 and 15A-6. The staff finds that
the radiological consequences calculated by Westinghouse meet the relevant dose acceptance
criteria stated above. To verify the Westinghouse analyses, the staff performed independent
radiological calculations for the above DBAs using the hypothetical /Q values provided by the
applicant and the computer code described in Supplement 2 to NUREG/CR-6604, RADTRAD:
A Simplified Model for Radionuclide Transport and Removal and Dose Estimation. The
following sections describe the staffs findings.
Accident Source Terms
In SECY-94-302, Source Term-Related Technical and Licensing Issues Relating to
Evolutionary and Passive Light-Water-Reactor Designs, dated December 19, 1994, the staff
proposed to use only the coolant, gap, and early in-vessel releases from NUREG-1465 for the
radiological consequence assessments of DBAs for the passive advanced light-water reactor
15-67

(ALWR) designs. These source terms encompass a broad range of accident scenarios,
including significant levels of core damage with the core remaining in the vessel. These
scenarios define the most severe accidents from which the plant could be expected to return to
a safe-shutdown condition. The revised source terms in NUREG-1465 must be applied
conservatively in evaluating DBAs in conjunction with conservative assumptions in calculating
doses, such as adverse meteorology. Application to severe accidents may use more realistic
assumptions.
The staff considered the inclusion of the ex-vessel and the late in-vessel source terms to be
unduly conservative for DBA purposes. Such releases would only result from core damage
accidents with vessel failure and core-concrete interactions. For passive ALWRs, the estimated
frequencies of such scenarios are low enough that they do not have to be considered credible
for the purpose of meeting 10 CFR 50.34. In SECY-94-302, the Commission approved the
staff-recommended technical positions to use only the coolant, gap, and early in-vessel
releases from NUREG-1465 for the radiological consequence assessments of DBAs for the
passive ALWR designs.
The NRC issued RG 1.183 in July 2000 to provide guidance to licensees of operating power
reactors on acceptable applications of alternative source terms pursuant to 10 CFR 50.67. This
RG provides guidance based on insights from NUREG-1465 and significant attributes of other
alternative source terms that the NRC staff may find acceptable for operating light-water
reactors (LWRs). It also identifies acceptable radiological analysis assumptions for use in
conjunction with the accepted alternative source term for operating power reactors. The
applicant followed the relevant guidance in RG 1.183 for PWRs.
Post-Accident Containment Water Chemistry Management
Management of the postaccident containment water chemistry must comply with the
requirements of GDC 41, Containment Atmosphere Cleanup, and GDC 4, Environmental and
Dynamic Effect Design Bases. By minimizing the release of radioactive iodine from the
containment sump water, the water chemistry will meet the requirement of GDC 41, as it relates
to the ability of the design of containment atmosphere cleanup systems to control fission
product releases to the reactor containment following postulated accidents. By preventing
stress-corrosion cracking of stainless steel components exposed to the water accumulated in
the containment sump, the water chemistry will meet the requirement of GDC 4 that
components important to safety be compatible with the environmental conditions associated
with accident conditions, including LOCAs.
NUREG-1465 states that, after an accident, iodine entering the containment from the reactor
core is composed of at least 95 percent cesium iodide (CsI), with the remaining 5 percent
elemental iodine and a small amount of hydriodic acid. However, about three percent of
elemental iodine in contact with some organic compounds will produce organic iodides.
Therefore, the iodine in the containment will consist of 95 percent particulate iodine as CsI,
4.85 percent elemental iodine (I2), and 0.15 percent organic iodine. The composition of the
iodine in the AP1000 is consistent with the composition stated in NUREG-1465.
15-68

Iodine in the form of CsI is soluble in the containment water. However, some of it may be
converted into the elemental form, which is considerably less soluble, and will be released into
the containment atmosphere. The released radioactive iodine may leak out of the containment
and contribute to outside radiation doses. To minimize formation of the elemental iodine, the
pH of the containment water should be kept basic. Basic pH will also prevent stress-corrosion
cracking of the stainless steel components.
In the AP1000, the pH of the containment water will be between 7 and 9.5, which is the range
found acceptable in Branch Technical Position (BTP) Materials Engineering Branch (MTEB) 6-1
of the SRP. A predetermined amount of trisodium phosphate (TSP) stored in the stainless steel
baskets situated on the containment floor will maintain the pH in this range. After a LOCA,
when the containment flooding water reaches the level of the baskets, the TSP will dissolve,
and the solution of TSP will exercise buffering action and maintain sump water pH in the
required range.
Acidic and basic chemical species released to the containment from different sources in the
plant determine the pH of the containment sump water. Boric acid produces the most
significant effect on reducing containment water pH. Westinghouse identified the RCS,
IRWST, CMTs, accumulators, CVCS BAT, and spent fuel pool cooling system (SFS) cask
loading pit as sources of boric acid. Westinghouse did not include normal operating RCS
leakage, because in the AP1000 plant design such leakage would quickly drain to the waste
sump and be pumped out of the containment. Other sources of chemical species that are
formed in the containment during the 30 days following a core damage accident include
hydrochloric acid produced by radiolytic decomposition of electric cable jackets and nitric acid
produced by radiolytic formation from air dissolved in the sump water. In addition, there is
cesium hydroxide present, released from the damaged core. Because it is a strong base,
cesium hydroxide will contribute to the increase of the pH. The applicant has determined that
the baskets in the containment sump must store 12,492 kg (27,540 lb) of TSP to maintain the
pH in the range of 7 to 9.5. While 7,503 kg (16,540 lb) of TSP is needed to neutralize boric
acid, the rest will neutralize other acids existing in the containment and provide a 35-percent
safety margin, which includes 10 percent to cover possible long-term degradation of TSP.
By performing independent verifications, based on the information provided in the response to
RAI 281.004, the staff has confirmed the adequacy of the postaccident management of water
chemistry in the AP1000 plant design. The staff finds that the amount of TSP stored in the
containment sump will ensure a basic environment that will minimize the release of radioactive
iodine to the outside and will prevent stress-corrosion cracking of the stainless steel
components exposed to the containment water. The staff concludes, therefore, that the
AP1000 plant design meets the requirements of GDC 41 and 4.
Hypothetical Atmospheric Dispersion Factors
Because no specific site is associated with the AP1000 plant, Westinghouse defined the offsite
boundaries only in terms of various hypothetical atmospheric relative concentration ( /Q)
values at fixed EAB and LPZ distances. DCD Tier 2, Tables 15A-5 and 15A-6 list the
hypothetical reference /Q values used in the radiological consequence analyses for the
AP1000 design. Section 2.3.4 of this report provides the staffs discussion of the hypothetical
15-69

atmospheric dispersion factors. The staff will perform an independent assessment of
short-term (less than or equal to 30 days) atmospheric dispersion factors for potential accident
consequence analyses on a site-specific basis for a COL application that references the
AP1000 design. If site-specific atmospheric dispersion factors exceed the reference values
used in this evaluation (e.g., poorer dispersion characteristics), a COL applicant may have to
consider compensatory measures, such as increasing the size of the site or providing additional
ESF systems to meet the relevant dose limits set forth in 10 CFR 50.34 and GDC 19.
The staff had not completed its review of the control room atmospheric dispersion factors at the
time the DSER was issued. Because the radiological consequence analyses for control room
habitability use the control room atmospheric dispersion factors as an input, the staff could not
make a finding on the acceptability of the Westinghouse analyses. In addition, the staff could
not make a finding on whether Westinghouse had shown that the AP1000 design would be
expected to meet GDC 19 for the hypothetical meteorological conditions used in the DCD. With
the conclusion of the staffs review of the control room atmospheric dispersion factors,
discussed in Section 2.3.4 of this report, the staff has completed its review of the control room
habitability radiological consequences analyses. Therefore, DSER Open Item 15.3-2 is closed.
15.3.1 Radiological Consequences of a Main Steam Line Break Outside Containment
Both the staff and Westinghouse have evaluated the radiological consequences of a postulated
SLB accident occurring outside of the containment and upstream of the MSIVs. The applicant
submitted a radiological analysis for the MSLB accident in DCD Tier 2, Section 15.1.5.4. The
applicant analyzed this hypothetical accident using the following parameters:
567.8 L/d (150 gal/d) of primary-to-secondary leakage through any one SG, as specified
in the AP1000 TS
discharge of the entire mass of secondary water from one affected SG to the
environment with no iodine partitioning
The applicant also considered a coincident loss of spent fuel pool (SFP) cooling capability.
Section 15.3.9 of this report discusses the staffs review of the radiological consequences of
SFP boiling.
The staff has reviewed the applicants analysis and finds that the calculational methods used for
the radiological consequence assessment are acceptable, and the radiological consequences
calculated by the applicant meet the relevant dose acceptance criteria.
To verify the applicants assessment, the staff performed an independent radiological
consequence calculation for three scenarios for the MSLB accident. For Case 1, the most
reactive control rod was assumed to be stuck in the fully withdrawn position. The applicant
indicated, and the staff agreed, that no DNB is expected to occur. Therefore, the calculation
did not assume fuel-cladding failure. With no additional fuel failures occurring, Case 1
becomes identical to Case 2 (discussed below), and no radiological consequences are
presented for Case 1.
15-70

For Case 2, the staff assumed that a temporary increase in the primary coolant iodine
concentration (iodine spike) occurred as a result of the power/pressure transient caused by the
MSLB accident. Before the accident, the AP1000 reactor was assumed to operate at the
AP1000 TS equilibrium limit of 37 kBq/gm (1.0 Ci/gm) for dose equivalent iodine-131
(DEI-131) in the primary coolant. The iodine spike generated during the accident is assumed to
increase the release rate of iodine from the fuel by a factor of 500. This increase in the release
rate results in an increasing concentration in the primary coolant during the course of the
accident. For Case 3, the staff assumed that previous reactor operation had resulted in a
primary coolant iodine concentration equal to the maximum instantaneous AP1000 TS limit of
2.2 MBq/gm (60 Ci/gm) for DEI-131.
Tables 15.3-2 and 15.3-1 of this report provide the major parameters and assumptions used by
the staff for the MSLB accident and the results of the staffs radiological consequence analyses
for the EAB, LPZ, and control room, respectively. The offsite radiological consequences
calculated by the staff are consistent with those calculated by the applicant.
The staff concludes that the AP1000 design, as bounded by the atmospheric relative
concentrations proposed by the applicant, will provide reasonable assurance that the
radiological consequences of a postulated MSLB accident with accident-induced iodine spiking
and coincident with the loss of SFP cooling capability will not exceed a small fraction (i.e.,
10 percent or 0.025 Sv (2.5 rem) TEDE) of the dose criterion set forth in 10 CFR 50.34.
The staff also concludes that the AP1000 design, as bounded by the atmospheric relative
radiological consequences of a postulated MSLB accident with the reactor coolant at the TS
maximum value of 2.2 MBq/gm (60 Ci/gm) for DEI-131 and coincident with the loss of SFP
cooling capability will not exceed the dose criterion set forth in 10 CFR 50.34 (i.e., 0.25 Sv
(25 rem) TEDE).
In DCD Tier 2, Section 6.4.4, Westinghouse reported the results of its radiological consequence
analysis for personnel in the main control room during a design-basis MSLB, relying on the
main control room emergency habitability system (VES) to limit the radioactivity to which the
personnel may be exposed. Section 6.4 of this report describes the staffs review and
assessment of the VES. To verify the Westinghouse assessment, the staff performed an
independent radiological consequence calculation for the MSLB accident with VES operation
under high-high radiation levels. The staff finds reasonable assurance that the VES, under
high-high radiological conditions as described in DCD Tier 2, Section 6.4, can mitigate the dose
in the main control room following a design-basis MSLB to meet the dose criterion specified in
GDC 19. Tables 15.3-9 and 15.3-1 of this report provide the major assumptions used by the
staff and the resulting radiological consequence analysis results for the control room operators,
respectively.
The calculation of the dose to the control room operators does not rely on the nuclear island
nonradioactive ventilation system (VBS), a non-safety-related system, to meet the requirements
of GDC 19. Under some accident circumstances, the non-safety-related VBS would be
available to pressurize the control room and the technical support center (TSC) with filtered air
and to provide recirculation cleanup. Section 9.4 of this report describes the staffs review and
15-71

assessment of the non-safety-related VBS. In DCD Tier 2, Section 6.4.4, Westinghouse also
reported the results of its radiological consequence analysis for personnel in the main control
room and the TSC during a design-basis MSLB with the VBS available. The staff finds
reasonable the Westinghouse assertion that, if available, the VBS can mitigate the dose in the
main control room and the TSC following a design-basis MSLB to be within 0.05 Sv (5 rem)
TEDE. Table 15.3-9 of this report provides the major assumptions used by the staff and the
resulting radiological consequence analysis results for the personnel in the control room and
the TSC with VBS available.
15.3.2 Radiological Consequences of a Reactor Primary Coolant Pump Seizure (Locked
Rotor)
An instantaneous seizure of an RCP rotor rapidly reducing the primary coolant flow through the
affected reactor coolant loop, leading to a reactor trip on a low-flow signal, causes the reactor
primary coolant pump seizure accident. Westinghouse analyzed this hypothetical accident
assuming that 10 percent of the fuel elements will experience cladding failure, releasing the
entire fission product inventory in the fuel-cladding gap of these elements to the reactor coolant.
The maximum allowable 1135.6 L/d (300 gal/d) of primary-to-secondary leakage through two
SGs, as specified in the AP1000 TS, carries the activity released to the primary coolant into the
secondary coolant. The steamline safety valves or the PORVs release the activity to the
environment. The applicant submitted a radiological analysis for the reactor primary coolant
pump seizure accident in DCD Tier 2, Section 15.3.3.
The applicant also considered a coincident loss of SFP cooling capability. Section 15.3.9 of this
report discusses the staffs review of the radiological consequences of SFP boiling.
To verify the applicants assessment, the staff performed independent radiological
consequence calculations for the reactor primary coolant pump seizure accident using the
RG 1.183 source terms. Tables 15.3-3 and 15.3-1 of this report provide the major parameters
and assumptions used by the staff and the results of the staffs radiological consequence
analyses, respectively. The offsite radiological consequences calculated by the staff are
consistent with those calculated by the applicant.
radiological consequences of a postulated reactor primary coolant pump seizure accident
coincident with the loss of SFP cooling capability will not exceed a small fraction (i.e.,
analysis for personnel in the main control room during a design-basis locked rotor accident,
relying on the VES to limit the radioactivity to which the personnel may be exposed. Section 6.4
15-72

of this report describes the staffs review and assessment of the VES. To verify the
Westinghouse assessment, the staff performed an independent radiological consequence
calculation for the locked rotor accident with VES operation under high-high radiation levels.
The staff finds reasonable assurance that the VES, under high-high radiological conditions as
described in DCD Tier 2, Section 6.4, can mitigate the dose in the main control room following a
design-basis locked rotor accident to meet the dose criterion specified in GDC 19.
Tables 15.3-9 and 15.3-1 of this report provide the major assumptions used by the staff and the
resulting radiological consequence analysis results for the control room operators, respectively.
VBS, a non-safety-related system, to meet the requirements of GDC 19. Under some accident
circumstances, the non-safety-related VBS would be available to pressurize the control room
and the TSC with filtered air and to provide recirculation cleanup. Section 9.4 of this report
describes the staffs review and assessment of the non-safety-related VBS. In DCD Tier 2,
Section 6.4.4, Westinghouse also reported the results of its radiological consequence analysis
for personnel in the main control room and the TSC during a design-basis locked rotor accident
with the VBS available. The staff finds reasonable the Westinghouse assertion that, if
available, the VBS can mitigate the dose in the main control room and the TSC following a
design-basis locked rotor accident to be within 0.05 Sv (5 rem) TEDE. Table 15.3-9 of this
report provides the major assumptions used by the staff and the resulting radiological
consequence analysis results for the personnel in the control room and the TSC with VBS
available.
15.3.3 Radiological Consequences of Rod Cluster Control Assembly Ejection
The mechanical failure of a control rod mechanism pressure housing is postulated to result in
the ejection of an RCCA and drive shaft. Because of the resultant opening in the pressure
vessel, primary coolant is lost to the containment with concurrent rapid depressurization of the
reactor pressure vessel. This mechanical failure causes a rapid positive reactivity insertion,
together with an adverse core power distribution, possibly leading to localized fuel rod damage.
The applicant has assumed that 10 percent of the fuel elements will experience cladding failure,
releasing the entire fission product inventory in the fuel-cladding gap of these elements. In
addition, the applicant assumed that 0.25 percent of the fuel rods may experience fuel melting.
The applicant performed its calculations to obtain these parameters using the guidelines
provided in RG 1.77, Assumptions Used for Evaluating a Control Rod Ejection Accident for
PWRs. Therefore, the staff finds these assumptions to be acceptable. The applicant
submitted a radiological consequence analysis for the control element assembly ejection
accident in DCD Tier 2, Section 15.4.8.
15-73

The applicant assumed that the release of fission products to the environment may occur via
either of two pathways. The first pathway involves a release of primary coolant to the
containment, which is assumed to leak into the environment at the design leak rate of the
containment. In the second pathway, fission products would reach the secondary coolant via
the SGs with a maximum total allowable primary-to-secondary leak rate of 1135.6 L/d
(300 gal/d), as specified in the AP1000 TS. For both pathways, the applicant assumed that the
AP1000 reactor operated at its TS instantaneous primary coolant limit of 2.2 MBq/gm
(60 Ci/gm) for DEI-131.
To verify the applicants assessment, the staff performed independent radiological
consequence calculations for the same two pathways, as described above for the RCCA
ejection accident, using the RG 1.183 source terms. Tables 15.3-4 and 15.3-1 of this report
provide the major parameters and assumptions used by the staff and the results of the staffs
radiological consequence analyses, respectively. The offsite radiological consequences
calculated by the staff are consistent with those calculated by the applicant.
radiological consequences of a postulated RCCA ejection accident coincident with the loss of
SFP cooling capability will fall well within the dose criterion set forth in 10 CFR 50.34 (i.e.,
25 percent or 0.063 Sv (6.3 rem) TEDE).
analysis for personnel in the main control room during a design-basis RCCA ejection accident,
calculation for the RCCA ejection accident with VES operation under high-high radiation levels.
design-basis RCCA ejection accident to meet the dose criterion specified in GDC 19.
for personnel in the main control room and the TSC during a design-basis RCCA ejection
accident with the VBS available. The staff finds reasonable the Westinghouse assertion that, if
design-basis RCCA ejection accident to be within 0.05 Sv (5 rem) TEDE. Table 15.3-9 of this
consequence analysis results for the personnel in the control room and the TSC with VBS
available.
15-74

15.3.4 Radiological Consequences of the Failure of Small Lines Carrying Primary
Coolant Outside Containment
GDC 55, Reactor Coolant Pressure Boundary Penetrating Containment, contains a provision
to ensure isolation of all pipes that are part of the RCPB and penetrate the containment
building. GDC 55 also provides that small-diameter pipes that must be continuously connected
to the primary coolant system to perform necessary functions may be acceptable based on
some other defined bases. For these lines, methods of mitigating the consequences of a
rupture are necessary because the lines cannot be isolated automatically. For the AP1000
design, the two small lines in this category are
C
C
the RCS sample line

the discharge line from the CVCS to the liquid radwaste system (WLS)
No instrument lines carry primary coolant outside containment in the AP1000 design.
When boron dilution operations generate excess primary coolant inventory, the CVCS
purification flow is diverted out of containment to the WLS. Before passing outside
containment, the flow stream passes through the CVCS HXs and mixed bed demineralizer for
processing. The flow leaving the containment will be at a temperature of less than 60 EC
(140 EF). The flow from a postulated break in this line is limited to the CVCS purification normal
flow rate of 379 L/min (100 gpm). In DCD Tier 2, Section 15.6.2, considering the low
temperature of the breakflow and the reduced iodine activity because of demineralization for the
postulated break in the discharge line from the CVCS to the WLS, the applicant proposed, and
the staff accepted, that the postulated break in the RCS sample line is the more limiting event
for the radiological consequence assessment.
The RCS sample line includes a flow restrictor at the point of sample to limit the breakflow to
less than 492 L/min (130 gpm). Because the sample line isolation valves are open only when
sampling is ongoing, and there are multiple indications that a break has occurred in the sample
line, the applicant assumed, and the staff accepted, that the breakflow isolation time will be less
than 30 minutes. The applicant assumed the fluid escaping the break to be at the equilibrium
primary coolant iodine concentration limits in the AP1000 TS, with an assumed accidentinitiated iodine spike that increases the rate of iodine release from the fuel into the coolant by a
factor of 500. The staff finds this to be acceptable and in agreement with guidance on
assumptions for radioactivity released from a small line break found in SRP Section 15.6.2,
Radiological Consequences of the Failure of Small Lines Carrying Primary Coolant Outside
Containment. The applicant submitted a radiological analysis for a small line failure in DCD
Tier 2, Section 15.6.2.
The staff has reviewed the applicant analysis and finds that the calculational methods used for
calculated by the applicant meet the relevant dose acceptance criteria. To verify the applicants
15-75

assessment, the staff performed independent radiological consequence calculations for a
postulated small line break accident. The staff assumed that a temporary increase in the
primary coolant iodine concentration (iodine spike) occurred as a result of the power/pressure
transient caused by the small line break accident. Before the postulated accident, the AP1000
reactor was assumed to operate at the AP1000 TS equilibrium concentration limit of 37 kBq/gm
(1.0 Ci/gm) for DEI-131 in the primary coolant.
The iodine spike generated during the accident is assumed to increase the release rate of
iodine from the fuel by a factor of 500. This increase in the release rate results in a rising iodine
concentration in the primary coolant during the course of the accident.
Tables 15.3-5 and 15.3-1 of this report provide the major parameters and assumptions used by
the staff and the results of the staffs radiological consequence analyses, respectively. The
offsite radiological consequences calculated by the staff are consistent with those calculated by
the applicant.
radiological consequences of a postulated small line break accident coincident with the loss of
SFP cooling capability will not exceed a small fraction (i.e., 10 percent or 0.025 Sv (2.5 rem)
TEDE) of the dose criterion set forth in 10 CFR 50.34.
analysis for personnel in the main control room during a design-basis small line break accident,
calculation for the small line break accident with VES operation under high-high radiation levels.
design-basis small line break accident to meet the dose criterion specified in GDC 19.
for personnel in the main control room and the TSC during a design-basis small line break
accident with the VBS available. The staff finds reasonable the Westinghouse assertion that, if
design-basis small line break accident to be within 0.05 Sv (5 rem) TEDE. Table 15.3-9 of this
consequence analysis results for the personnel in the control room and TSC with VBS
available.
15-76

15.3.5 Radiological Consequences of a Steam Generator Tube Rupture
The applicant has evaluated the radiological consequences of a postulated SGTR accident and
provided a radiological consequence analysis for the accident in DCD Tier 2, Section 15.6.3.
calculated by Westinghouse meet the relevant dose acceptance criteria.
To verify the applicants assessments, the staff performed independent radiological
consequence calculations for two scenarios for the SGTR accident. For Case 1, the staff
assumed that a temporary increase in the primary coolant iodine concentration (iodine spike)
occurred as a result of the power/pressure transient caused by the SGTR. Before the
postulated accident, the AP1000 reactor was assumed to operate at the AP1000 TS equilibrium
iodine concentration limit of 37 kBq/gm (1.0 Ci/gm) for DEI-131 in the primary coolant. The
iodine spike generated during the accident is assumed to increase the release rate of iodine
from the fuel by a factor of 335, resulting in a rising iodine concentration in the primary coolant
during the course of the accident.
For Case 2, the staff assumed that previous reactor operation had resulted in a primary coolant
concentration equal to the maximum instantaneous concentration limit of 2.2 MBq/gm
(60 Ci/gm) for DEI-131, specified in the AP1000 TS. Tables 15.3-6 and 15.3-1 of this report
provide the major parameters and assumptions used by the staff and the results of the staffs
radiological consequence analyses for the EAB and LPZ and for the control room, respectively.
The offsite radiological consequences calculated by the staff are consistent with those
calculated by the applicant.
radiological consequences of a postulated SGTR accident with accident-induced iodine spiking
and coincident with the loss of SFP cooling capability will not exceed a small fraction (i.e.,
radiological consequences of a postulated SGTR accident with the reactor coolant at the TS
maximum value of 2.2 MBq/gm (60 Ci/gm) for DEI-131 and coincident with the loss of SFP
cooling capability will not exceed the dose criterion set forth in 10 CFR 50.34 (i.e., 0.25 Sv
(25 rem) TEDE).
analysis for personnel in the main control room during a design-basis SGTR, relying on the
VES to limit the radioactivity to which the personnel may be exposed. Section 6.4 of this report
describes the staffs review and assessment of the VES. To verify the Westinghouse
assessment, the staff performed an independent radiological consequence calculation for the
15-77

SGTR with VES operation under high-high radiation levels. The staff finds reasonable
assurance that the VES, under high-high radiological conditions as described in DCD Tier 2,
Section 6.4, can mitigate the dose in the main control room following a design-basis SGTR to
meet the dose criterion specified in GDC 19. Tables 15.3-9 and 15.3-1 of this report provide
the major assumptions used by the staff and the resulting radiological consequence analysis
results for the control room operators, respectively.
for personnel in the main control room and the TSC during a design-basis SGTR with the VBS
available. The staff finds reasonable the Westinghouse assertion that, if available, the VBS can
mitigate the dose in the main control room and the TSC following a design-basis SGTR to be
within 0.05 Sv (5 rem) TEDE. Table 15.3-9 of this report provides the major assumptions used
by the staff and the resulting radiological consequence analysis results for the personnel in the
control room and the TSC with VBS available.
15.3.6 Radiological Consequences of LOCAs
In DCD Tier 2, Section 15.6.5, the applicant analyzed a hypothetical design-basis LOCA. The
applicant concluded that certain bounding sets of atmospheric relative concentration values
(i.e., /Qs) specified in DCD Tier 2, Section 2.3, in conjunction with the use of natural deposition
of fission product aerosol within the containment and the control of the pH of the water in the
containment to prevent iodine evolution, are sufficient to provide reasonable assurance that the
calculated radiological consequences of a postulated design-basis LOCA will fall within the
relevant dose criteria established in 10 CFR 50.34 and in GDC 19.
All of the fission product releases stemming from the LOCA result from containment leakage.
The AP1000 design does not have ESF systems outside containment. Therefore, the
radiological consequence analyses do not consider leakage from the ESF systems. The
containment was assumed to leak at its design leak rate of 0.1 wt% per day for the first
24 hours and at half of that rate (0.05 wt% per day) for the remaining duration of the accident
(30 days). The AP1000 design provides neither an ESF filtration (e.g., charcoal adsorbers) nor
a safety-related containment spray system.
For the DSER, the staff had not completed its evaluation of the applicants assumptions on
aerosol removal in containment, as discussed in RAIs 470.009 and 470.011. Until the
resolution of the issues with aerosol removal in containment, the staff could not complete its
independent radiological consequence analysis of a postulated design-basis LOCA. This was
Open Item 15.3.6-1. With the resolution of DSER Open Item 15.3-1 on aerosol removal, as
discussed below, the staff could complete its independent analyses to confirm the applicants
15-78

assessment of the radiological consequences of the design-basis LOCA. Therefore, DSER
Open Item 15.3.6-1 is closed.
The AP1000 design does not provide an active containment atmosphere cleanup system.
Instead, the design relies on natural aerosol removal processes for deposition in the
containment, such as gravitational settling and plateout on containment surfaces through
diffusiophoresis and thermophoresis. Appendix 15B to DCD Tier 2 discusses the removal of
airborne activity from the containment atmosphere. The applicant has provided a containment
spray system for accident management following a severe accident as part of the AP1000 fire
protection system design (discussed in Section 19.2.3.3.9 of this report). The containment
spray system design is not safety-related and is not intended to be used during or following a
DBA. Therefore, radiological consequence assessments give the containment spray system no
credit for mitigation of radiological releases following a DBA.
The AP1000 LOCA offsite doses do not scale up from the AP600 proportionately with the
change in the power or T-H conditions in the containment. The main reason is that, given the
same leak rates, the amount of the release is a sole function of the airborne fission product
mass, which is proportional to the core inventory. The AP1000 inventory is about twice that of
the AP600 (about 70 percent higher power with a longer fuel cycle). Consequently, the removal
rate for the AP1000 must be at least twice that of the AP600 to achieve the same offsite doses.
The calculations performed by Westinghouse and the confirmatory uncertainty analysis
performed by the staff show that the current containment design does not provide for a removal
rate of twice that of the AP600. However, by adjusting the atmospheric dispersion factors
( /Qs) lower than those used in the AP600 design, the radiological consequences of the
LOCA remain acceptable. Because of this, the siting of the AP1000 design is more limited than
for the AP600.
Applying credit for aerosol removal through natural processes requires input from T-H and
aerosol behavior models. The basis document defining the revised accident source term,
NUREG-1465, does not specify an associated T-H scenario, or methodology or acceptance
criteria for aerosol removal. The alternative source term regulatory guidance, RG 1.183, also
does not specify these items. NUREG-1465 describes a source term that was derived from an
examination of a set of severe accident sequences for LWRs and is intended to be
representative or typical and does not imply a specific scenario, much less the worst case.
The determination of aerosol removal rates is simplified for a containment shown to have a
well-mixed atmosphere. The AP1000 design relies on natural circulation currents enhanced by
the PCS to inhibit stratification of the containment atmosphere. DCD Tier 2, Appendix 6A,
discusses the physical mechanisms of natural circulation mixing that occur in the AP1000.
Section 6.2.5.3 of this report provides the staffs discussion of natural circulation within the
AP1000 containment.
In DCD Tier 2, Table 15B-1, the applicant provided aerosol removal coefficient values starting
at the onset of a gap release through the first 24 hours into a DBA. The values range from
0.287/h to 1.141/h. The aerosol removal coefficients calculated by the applicant neglect steam
condensation on the airborne particles, turbulent diffusion, and turbulent agglomeration. The
assumed source aerosol size is conservatively small, because it is at the low end of the mass
15-79

mean aerosol size range of 1.5 m to 5.5 m used in NUREG/CR-5966, A Simplified Model of
Aerosol Removal by Containment Sprays, issued June 1993. Selection of a small aerosol size
would underestimate removal by sedimentation.
The aerosol removal coefficients calculated by Westinghouse for the AP1000 are based on a
single set of parameters, and Westinghouse performed sensitivity studies to determine if
changing the parameters would affect the overall removal coefficients. The staff felt that some
parameters chosen originally by Westinghouse were not conservative, as would be required for
use in design-basis analyses, nor were these parameters likely to be suitable for use as point
estimates, considering the lack of knowledge and uncertainty in some areas of post-LOCA
aerosol formation and detailed containment response.
In the past, the staff and industry evaluated aerosol removal through well-established models of
spray removal or condensation. The AP1000 application relies on natural deposition processes
that depend strongly on local T-H conditions. While gravitational settling is relatively easy to
understand, aerosol removal through diffusiophoresis and thermophoresis is much more
complex. Diffusiophoresis is associated with steam condensation on the heat sinks and
depends on the condensation steam mass flux. Thermophoresis relies only on the temperature
gradient close to the surface on which the particles would be deposited. Thermophoresis is
more subtle than the other two natural deposition processes. Because the temperature
gradient cannot be measured or easily calculated, its model uses the heat flux at the surface
divided by the thermal conductivity of the gas adjacent to the surface as an equivalent measure
of the driving force. Simultaneous occurrence of the two phoretic processes introduces an
additional level of complexity.
The Westinghouse methodology includes industrys Modular Accident Analysis Program
(MAAP) code, an integrated accident analysis program, to establish T-H boundary conditions as
an input to an aerosol code (STARNAUA). To determine the acceptability of the Westinghouse
modeling, the staff audited Westinghouse calculations of the containment removal coefficients.
The audit revealed that the heat flux used by Westinghouse included the convection, the
thermal radiation, and the decay heat from airborne fission products. Thermal radiation and
decay heat do not contribute to the temperature gradient that drives thermophoresis and their
use caused the overall aerosol removal to be nonrealistic and nonconservative. Westinghouse
recalculated the overall aerosol removal coefficients by correcting this error.
In its independent evaluation of aerosol removal coefficients, the staff considered the same
natural processes for removing aerosols from the containment atmosphere over the entire
period of an accident (30 days). These processes include the sedimentation mechanism of
gravitational settling, such as aerosol agglomeration, and the phoretic mechanisms of
diffusiophoresis and thermophoresis.
The staff contracted Sandia National Laboratory to perform quantitative analyses of
uncertainties in predicting the aerosol removal rates. The initiating event, used in the analysis
(3BE-1 sequence), is a double-ended break of a DVI 10.2-cm- (4-in.-) diameter line. The break
is assumed to be in a larger compartment of the passive core cooling system (PXS). The
sequence includes the water spillage into the PXS-B compartment from one of the lost
accumulators and one of the CMTs. The analysis assumes that one CMT and one accumulator
15-80

(in the intact DVI train) remain available to inject water into the RV, which is not sufficient to
keep the core covered. This deficiency in core cooling leads to degradation and/or melt of the
fuel. Eventually, the water level in the containment reaches the break level and spills into the
vessel, arresting the core degradation before potential failure of the vessel lower head.
As part of the staffs review, the aerosol behavior in the AP1000 containment was predicted
using the MELCOR integrated accident analysis code, which includes the MAEROS aerosol
mechanics code. The staff used the results of a fully integrated MELCOR analysis of the
AP1000 3BE-1 accident, performed by the staffs contractor, Energy Research, Inc., to develop
a simplified containment model for the uncertainty analysis. The NUREG-1465 radiological
source term for the gap release and in-vessel release phases were used in place of the source
term predicted in the fully integrated analysis. The uncertainty analysis considered those
MELCOR parameters known to affect aerosol settling and depletion to be uncertain within a
range of values, represented by an assumed distribution function. A Monte Carlo method,
which randomly samples the uncertain parameters and performs a large number of separate
MELCOR analyses, estimated the effect of the uncertain physics parameters on the aerosol
removal rate.
In its evaluation of aerosol removal rates, the staff used the containment geometry (e.g.,
volume, upward facing surface area) provided by Westinghouse and the fission product release
timing, fractions, and release rates as described in NUREG-1465.
The principal uncertainties in aerosol properties and aerosol behavior considered in the staffs
analyses included the following:
C
aerosol size and distribution

aerosol void fraction and particle shape factors
aerosol material density
nonradioactive aerosol mass
particle slip coefficient
sticking probability for agglomeration
boundary layer thickness for diffusion deposition
thermal accommodation coefficient for thermophoresis
ratio of thermal conductivity of particle to gas
turbulent energy dissipation
multipliers on heat and mass transfer to containment shell
The Westinghouse calculation of aerosol removal coefficients is based on an analysis of a

single T-H scenario and uses a single aerosol model without providing an uncertainty analysis.
The staff believes that the Westinghouse approach, though potentially acceptable, represents a
single BE result. Westinghouse used T-H conditions associated with the 3BE-1 severe
accident sequence. The staff concludes that using the T-H conditions associated with the
3BE-1 severe accident sequence represents the spectrum of accidents evaluated for the
AP1000 for the following reasons:
The conditions are representative of the 3BE accident class, which is the dominant
contributor to the core damage frequency for the AP1000.
15-81
The T-H conditions for 3BE accidents are typical of most of the analyzed sequences
because the majority of severe accident sequences analyzed for the AP1000 design are
fully depressurized and reflooded, given the highly reliable ADS.
The corresponding T-H profiles for these depressurized and reflooded cases are
sufficiently similar.
The use of a fully depressurized, low-pressure accident sequence in conjunction with

the source term described in NUREG-1465 is appropriate because the release fractions
for the source terms presented in NUREG-1465 are intended to be representative or
typical of those associated with a low-pressure core melt accident.
Therefore, the staff concludes that the 3BE-1 accident sequence is appropriate for determining
the amount of credit to give to the natural aerosol removal processes in the AP1000
containment. In its independent uncertainty analysis, the staff did not employ the T-H
calculated by Westinghouse using the MAAP code. The staffs uncertainty analysis did not
include differences between the staff and Westinghouse calculations with respect to
containment T-H and containment modeling as variables for study.
Although the choice of scenario is acceptable, the staff believes that a BE approach requires an
evaluation of the associated uncertainties. The staff used an alternative T-H code (MELCOR)
as an input to a Monte Carlo sampling (120 runs) of the above-listed parameters affecting
aerosol behavior. Engineering judgment was used to choose parameters as well as for the
range and distribution of their values, after several discussions between the staff and the
contractor. The resultant distribution of possible aerosol removal coefficients has a 95-percent
level of confidence. The 5th, 20th, 40th, 45th, 50th, 55th, 60th, 80th, and 95th percentiles, as depicted
in Figure 15.3.6-1 of this report and documented in Table 15.3.6-1 of this report, provide the
uncertainty distribution.
In the uncertainty analysis performed for the staff, the conservative lower bound (5th percentile)
aerosol removal coefficient ranges from 0.07/h to 1.26/h for the first 24 hours into a DBA
(shown in Table 15.3-8 of this report). The BE median (50th percentile) aerosol removal
coefficient ranges from 0.3/h to 1.35/h. The traditional regulatory approach is to accept a
bounding value, which would generally be the 5th percentile, as the maximum bounding value,
where there is an estimated 5 percent chance that the aerosol removal coefficients will be lower
than assumed, thereby resulting in higher calculated doses. In this particular case, however,
the staff proposes the use of the median value for the following reasons:
In an alternative source term (AST) pilot application for Perry, the staff had previously
accepted the 50th percentile value for steamline deposition, based on the NRC Office of
Nuclear Regulatory Research opinion that it is appropriate given other conservatisms
built into the other parts of the analysis.
The staff believes that the selected scenario belongs to a worst-case category.
The median values are least affected by the users sampling initial conditions.
15-82
The choice of the initial ranges and distributions of the selected parameters is highly
subjective.
Both the AST and the modeling of containment leakage have built-in conservativisms.
The dose calculation code requires yet another averaging of the aerosol removal
coefficients for the specified time periods.
The independent MELCOR-calculated aerosol removal coefficients are grouped mostly

well above the 5th percentile lower bound.
It is important to emphasize that it is the staffs judgment that the acceptance of the
50th percentile for use in DBA analyses is appropriate for this particular safety analysis because
an underlying conservative bias has been built into the staffs uncertainty methodology (e.g., the
choice of the ranges and distributions of the sampled parameters). The staff believes that
different choices of initial ranges and distributions, and/or the use of a different uncertainty
methodology, may not be acceptable.
The uncertainty analysis also indicated that, for the staffs model as implemented in MELCOR,
phoretic processes (thermophoresis and diffusiophoresis) together dominate for the first
4 hours of the LOCA and constitute 90 percent of aerosol removal. After that, gravitational
settling is the dominant removal mechanism.
Although the Westinghouse-calculated aerosol removal coefficient values lie within the staffs
uncertainty analysis upper and lower bounds for portions of the early part of the accident, they
generally do not lie below the median value calculated by the staff. While the staff does find
that gravitational settling, thermophoresis, and diffusiophoresis are physical processes that
occur in the AP1000 containment, and credit may be taken for aerosol removal through these
processes, the staff does not approve the specific Westinghouse-calculated aerosol removal
coefficient values. These values are an intermediate product used in the dose analysis and are
not subject to any regulations, per se. Staff independent analyses using the Westinghousesupplied /Qs and plant parameters result in acceptable doses, as discussed below. Thus,
while the staff and Westinghouse diverge on values for the intermediate steps in the dose
calculations, the staff agrees with the overall conclusion that the AP1000 design results in
acceptable doses.
The staff performed an independent dose analysis with the median aerosol removal coefficient
values from the staffs uncertainty analysis, along with other analysis parameters and the
bounding hypothetical atmospheric dispersion factors provided by Westinghouse, and the
results fall within the dose criteria of 10 CFR 50.34 and GDC 19. The staff performed a
sensitivity analysis, which resulted in calculated doses that remain below the regulatory
acceptance criteria for aerosol removal coefficients as low as those given in the 20th percentile
in the staffs uncertainty analysis.
The staff finds the radiological consequence analysis of the postulated DBA LOCA acceptable,
based on the Westinghouse DCD Tier 2, Chapter 15, plant parameters used in the staffs
15-83

analysis, the staff-calculated aerosol removal coefficient estimates (50th percentile, 95 percent
confidence), and the latest revision to the AP1000 /Qs, as documented in the Revision 5
response to DSER Open Item 15.3-1, dated June 21, 2004. Westinghouse will include the
revised information, including /Qs, in revision 12 of the AP1000 DCD and the staff will
confirm. With this basis, the doses meet the regulatory criteria of 10 CFR 50.34 and GDC 19
and are, therefore, acceptable.
In the DSER, the staff stated that it would perform an independent evaluation of the bounding
accident sequence and the aerosol behavior and removal rates corresponding to the selected
bounding accident sequence in the containment following a DBA. This was identified as DSER
Open Item 15.3-1. The staff has completed its evaluation and finds that the differences
between the resulting radiological consequences calculated using either Westinghouses or the
staffs aerosol removal rates are insignificant. Accordingly, analysis credit for aerosol removal
through natural processes is found acceptable, and DSER Open Item 15.3-1 is closed.
Because of the unique nature of the AP1000 design, which enhances natural aerosol removal
phenomena (such as the enhanced condensation of steam by external cooling of the
containment vessel instead of an internal containment spray), the staff has approved the use of
this T-H profile specifically for the AP1000. The NRC does not intend credit for aerosol removal
because of diffusiophoresis and thermophoresis to be generic for other plant designs, and this
practice must be approved on a case-by-case basis. Because the staff did not explicitly find the
input values for the aerosol removal coefficient used in the Westinghouse dose calculation to
be acceptable, the NRC staff must review any changes to the LOCA calculation made by a
COL applicant or licensee of an AP1000 plant.
analysis for personnel in the main control room during a design-basis LOCA, relying on the VES
to limit the radioactivity to which the personnel may be exposed. Section 6.4 of this report
LOCA with VES operation under high-high radiation levels. The staff finds reasonable
assurance that the VES, under high-high radiological conditions as described in DCD Tier 2,
Section 6.4, can mitigate the dose in the main control room following a design-basis LOCA to
meet the dose criterion specified in GDC 19. Tables 15.3-9 and 15.3-1 of this report provide
the major assumptions used by the staff and the resulting radiological consequence analysis
results for the control room operators, respectively.
for personnel in the main control room and the TSC during a design-basis LOCA with the VBS
available. The staff finds the Westinghouse assertion reasonable that, if available, the VBS can
mitigate the dose in the main control room and the TSC following a design-basis LOCA to be
15-84

within 0.05 Sv (5 rem) TEDE. Table 15.3-9 of this report provides the major assumptions used
by the staff and the resulting confirmatory analysis results.
15.3.7 Radiological Consequences of a Fuel-Handling Accident
In DCD Tier 2, Section 15.7.4, the applicant presented its analyses of the radiological
consequences of a postulated FHA. For the AP1000 design, an FHA can be postulated to
occur either inside containment or in the fuel-handling area inside the auxiliary building. If the
FHA occurs in the containment, closure of the containment purge lines based on the detection
of high airborne radioactivity can terminate the release of fission products. The applicant
assumed, in accordance with guidance in RG 1.183, that fission products are directly released
to the environment within a 2-hour period without credit for any iodine removal processes.
For the FHA, the applicant assumed that a single fuel assembly that has undergone 24 hours of
decay time is dropped, such that the activity in the gap of every rod in the dropped assembly is
released. The kinetic energy of the falling fuel assembly is assumed to break open the
maximum possible number of fuel rods using perfect mechanical efficiency. Instantaneous
release of noble gases and radioiodine vapor from the gaps of the broken rods (8 percent of
I-131, 10 percent of Kr-85, and 5 percent of other iodine and noble gas inventories in the fuel
rod) is assumed to occur, with the released gases bubbling up through the fuel pool water. The
fuel pool water has an assumed effective decontamination factor of 200 for total iodine. These
gap fractions agree with RG 1.183 guidance. The applicant assumed that iodine in the
particulate form is not volatile and, therefore, is not released. In accordance with RG 1.183
guidance, the applicant assumed that the particulate CsI is converted instantaneously to the
elemental form of iodine when it is released from the fuel into the low-pH pool water.
To verify the applicants assessments, the staff performed independent radiological
consequence calculations for the FHA occurring 24 hours after shutdown, coincident with a loss
of the SFP cooling capability. Tables 15.3-10 and 15.3-1 of this report provide the major
parameters and assumptions used by the staff and the results of the staffs radiological
consequence analyses, respectively. The offsite radiological consequences calculated by the
staff are consistent with those calculated by Westinghouse.
radiological consequences of a postulated FHA at 24 hours after shutdown with the loss of SFP
cooling capability will fall well within the dose criterion set forth in 10 CFR 50.34 (i.e., 25 percent
or 0.063 Sv (6.3 rem) TEDE).
15-85

In RAI 630.052, the NRC staff asked the applicant to justify not including a TS LCO for a decay
time limit related to the assumption used in the radiological consequences analysis of the FHA.
The applicant, in response to the RAI, stated that it had performed a sensitivity study in which
the FHA was assumed to occur 24 hours after shutdown, and the resulting doses remain below
10 CFR 50.34 guidelines. The applicant updated the DCD to include a paragraph discussing
the evaluation of the FHA, assuming a decay time of 24 hours. The applicant asserted that a
decay time LCO is not necessary because the evaluation of the FHA at 24 hours shows that the
capability of the AP1000 design to meet the regulatory dose acceptance criteria is not sensitive
to the decay time. The staff believes that, in order to exclude a TS LCO for the decay time, the
design-basis FHA dose analysis must assume a decay time that is clearly less than the time
physically needed to begin moving fuel assemblies out of the core following unit shutdown for
refueling. The staff does not consider 100 hours to be short enough. The applicant did not
revise the design-basis FHA dose analysis, which continued to include the 100-hour decay time
assumption. Additionally, in the DCD Tier 2, Section 15.7.4.5, discussion of the evaluation of
the FHA at 24 hours, the applicant did not discuss the impact of the decay time on control room
habitability from an FHA at 24 hours. The staff did not consider this issue to be resolved at the
time of issuance of the DSER. This was Open Item 15.3.7-1. To resolve this issue, the
applicant revised the design-basis FHA to assume that the fuel breached has been part of a
critical operating core 24 hours before the accident occurs. The revised analysis evaluated the
radiological consequences offsite and in the control room. The applicant also added refueling
operations TS 3.9.7, Decay Time, which requires that the reactor be shut down for at least
100 hours before movement of irradiated fuel in the reactor pressure vessel. Therefore, DSER
Open Item 15.3.7-1 is closed.
analysis for personnel in the main control room during a design-basis FHA, relying on the VES
to limit the radioactivity to which the personnel may be exposed. Section 6.4 of this report
FHA with VES operation under high-high radiation levels. The staff finds reasonable assurance
that the VES, under high-high radiological conditions as described in DCD Tier 2, Section 6.4,
can mitigate the dose in the main control room following a design-basis FHA to meet the dose
criterion specified in GDC 19. Tables 15.3-9 and 15.3-1 of this report provide the major
assumptions used by the staff and the resulting radiological consequence analysis results for
the control room operators, respectively.
VBS, a non-safety-related component, to meet the requirements of GDC 19. Under some
accident circumstances, the non-safety-related VBS would be available to pressurize the control
room and the TSC with filtered air and to provide recirculation cleanup. Section 9.4 of this
report describes the staffs review and assessment of the non-safety-related VBS. In DCD
Tier 2, Section 6.4.4, Westinghouse also reported the results of its radiological consequence
analysis for personnel in the main control room and the TSC during a design-basis FHA with
the VBS available. The staff finds reasonable the Westinghouse assertion that, if available, the
VBS can mitigate the dose in the main control room and the TSC following a design-basis FHA
to be within 0.05 Sv (5 rem) TEDE. Table 15.3-9 of this report provides the major assumptions
15-86

used by the staff and the resulting radiological consequence analysis results for the personnel
in the control room and the TSC with VBS available.
15.3.8 Offsite Radiological Consequences of Liquid Tank Failure
SRP Section 15.7.3, Postulated Radioactive Releases due to Liquid-Containing Tank Failures,
contains guidance on review of the failure of a tank containing liquid. The following regulations
provided the basis for the acceptance criteria specified in this SRP section:
GDC 60, Control of Releases of Radioactive Materials to the Environment, as it relates

to the ability of the design of the radioactive waste management system to control the
release of radioactive materials to the environment
10 CFR Part 20, as it relates to radioactivity in effluents to unrestricted areas
The failure of the most limiting (i.e., in terms of offsite radiological consequences) WLS
equipment outside the containment does not result in radionuclide concentrations in water at
the nearest potable water supply in an unrestricted area exceeding the liquid effluent
concentration limits for the corresponding radionuclides, specified in Appendix B to 10 CFR
Part 20 (Table 2, Column 2). The design of the WLS incorporates specific design features to
mitigate the effects of failure, if the WLS does not meet the above requirements of 10 CFR
Part 20.
In the AP1000 design, tanks containing radioactive fluids are located inside plant structures. In
the event of a tank failure, the floor drains would drain the liquid to the auxiliary building sump.
From the sump, the water would be directed to the waste holdup tank. Because SRP
Section 15.7.3 states that credit cannot be taken for liquid retention by unlined building
foundations, the assumption is made that release to the environment is possible.
DCD Section 15.7.3 includes a commitment for a COL applicant to perform a site-specific
offsite radiological consequence analysis, including the corresponding source term resulting
from a postulated liquid tank failure. The staff finds this commitment to be acceptable because
the assessment of offsite radiological consequences of liquid tank failures depends upon
site-specific parameters, such as the mode of transport of radioactive fluid resulting from the
failure to the region of potable water supply, the location of potable water supply, the
characteristics of the soil through which the transport occurs, and the available dilution by
waterbodies before the radioactive liquid reaches the potable water supply. The staff will
evaluate the site-specific analysis in accordance with SRP Section 15.3.7 for each COL
applicant referencing the AP1000 standard design. This is COL Action Item 15.3.8-1.
15.3.9 Radiological Consequences of Loss of Spent Fuel Pool Cooling
For the radiological consequences analysis of each DBA, Westinghouse evaluated the added
radiological consequences of SFP boiling because of the loss of SFP cooling capability.
15-87

The SFS is designed to perform the following functions:
remove heat from the SFP and the IRWST
remove radioactive corrosion and fission products from the SFP, the IRWST, and the
refueling cavity
transfer water between the IRWST and the refueling cavity for refueling operations
The system consists of redundant trains. Each train includes a pump, an HX, a filter, and a
demineralizer. However, the SFS is a non-safety-related system. Therefore, the applicant
assumed, and the staff agrees, that a loss of SFP cooling capability should be analyzed
coincident with DBAs.
The loss of SFP cooling could result in the pool reaching boiling, and a portion of the
radioactive iodine in the SFP water could be released to the environment. Without actions to
provide makeup water to the SFP, boiling is assumed to commence at 8.8 hours after loss of
SFP cooling capability. The applicant has calculated that the dose consequences from this
source are less than 0.1 mSv (0.01 rem) TEDE, both offsite and to the control room operators.
This dose is added to the dose consequences of each DBA to find the overall dose
consequences of the DBA coincident with loss of SFP cooling.
To verify the applicants assessment, the staff performed an independent radiological
consequence calculation for the loss of the SFP cooling capability. Tables 15.3-11 and 15.3-1
of this report provide the major parameters and assumptions used by the staff and the results
of the staffs radiological consequence analyses, respectively. The offsite radiological
consequences calculated by the staff are consistent with those calculated by the applicant.
15.3.10 Conclusions
The staff has reviewed the radiological consequences analyses of the DBAs described in DCD
Tier 2, Chapter 15, for the AP1000 design. Based on the evaluation discussed above, the staff
concludes that the AP1000 design meets 10 CFR 50.34(a)(1)(ii)(D) dose criteria and the offsite
dose acceptance criteria, as given in RG 1.183 for these accidents.
described in DCD Tier 2, Section 6.4, can mitigate the dose in the main control room following
DBAs to meet the dose criterion specified in GDC 19.
The staff finds it reasonable that, if available, the non-safety-related VBS can mitigate the dose
in the main control room and the TSC following DBAs to be within 0.05 Sv (5 rem) TEDE.
COL Action Item 15.3.8-1, regarding a site-specific analysis of the offsite radiological
consequences of a liquid tank failure, explains that the COL applicant should perform a sitespecific offsite radiological consequences analysis of the postulated liquid tank failure to
confirm that the plant meets the applicable regulations on radioactive waste management
15-88

systems and radiological effluents. The COL applicant should submit in the plant specific
application the radiological consequences analysis for the staff to review and approve.
15-89
15-90

Table 15.3-1 Staff-Calculated Radiological Consequences of Design-Basis Accidents
(Total Effective Dose Equivalent (TEDE))
Postulated Accident
EAB
Loss of coolant accident
190 mSv
(19 rem)
150 mSv
(15 rem)
34 mSv
(3.4 rem)
Main steamline break

outside containment
With accident-initiated
iodine spike
2 mSv
(0.2 rem)
8 mSv
(0.8 rem)
13 mSv
(1.3 rem)
With preaccident
iodine spike
<1 mSv
(<0.1 rem)
1 mSv
(0.1 rem)
9 mSv
(0.9 rem)
<1 mSv
(<0.1 rem)
<1 mSv
(<0.1 rem)
8 mSv
(0.8 rem)
<1 mSv
(<0.1 rem)
<1 mSv
(<0.1 rem)
12 mSv
(1.2 rem)
Rod ejection accident
15 mSv
(1.5 rem)
24 mSv
(2.4 rem)
11 mSv
(1.1 rem)
Fuel-handling accident
24 mSv
(2.4 rem)
10 mSv
(1.0 rem)
29 mSv
(2.9 rem)
Small line break accident
10 mSv
(1.0 rem)
4 mSv
(0.4 rem)
14 mSv
(1.4 rem)
Steam generator tube rupture

With accident-initiated
iodine spike
5 mSv
(0.5 rem)
7 mSv
(0.7 rem)
26 mSv
(2.6 rem)
With preaccident
iodine spike
10 mSv
(1.0 rem)
6 mSv
(0.6 rem)
50 mSv
(5 rem)
n/a*
<0.1 mSv
(<0.01 rem)
<0.1 mSv
(<0.01 rem)
Reactor coolant pump shaft seizure

With feedwater available
Without feedwater available
Spent fuel pool boiling
* n/a, not applicable
15-91
LPZ
Control Room

Table 15.3-2 Assumptions Used to Evaluate the Radiological Consequences of the
Main Steamline Break Accident Outside Containment
Parameter
Value
Power level, MWt
3468
Reactor primary coolant iodine concentrations

Accident-initiated iodine spike, KBq/gm DEI-131
Preaccident iodine spike, MBq/gm DEI-131
37
2.2
(1.0 Ci/gm)
(60 Ci/gm)
1.37E+5
567.8
(3.03E5 lb)
(150 gpd)
1.0
567.8
(150 gpd)
1.0
1.37E+5
5.56E+3
(3.0335E5 lb)
(1.225E4 lb)
Steam generator in faulted loop

Initial water mass, kg
Primary to secondary leak rate, L/d
Iodine partition coefficient
Steam generator in intact loop
Steam released, kg
0 to 2 hr
2 to 8 hr
Ratio of iodine release rate from fuel during
iodine spike to that during steady-state operation
500
Reactor primary coolant mass, kg
1.74E+5
Duration of accident, hr
72
Atmospheric dispersion values, sec/m3

EAB
0 to 2 hours
LPZ
0 to 8 hours
8 to 24 hours
1 to 4 days
5.1E-4
2.2E-4
1.6E-4
1.0E-4
Control room analysis parameters
Table 15.3.9
15-92
(3.84E+5 lb)

Table 15.3-3 Assumptions Used to Evaluate the Radiological Consequences of the
Reactor Coolant Pump Shaft Seizure Accident (Locked Rotor)
Parameter
Value
Power level, MWt
3468
Fraction of fuel rods failed
0.10
Fraction of core activity in failed fuel rod gap

I-131
Kr-85
Other iodines and noble gases
Alkali metals
0.08
0.10
0.05
0.12

2.2
(60 Ci/gm)
Secondary coolant mass, kg
2.75E+5
(6.06E5 lb)
Primary to secondary leak rate, kg/hr
47.3
(104.3 lb/hr)

Alkali metal partition coefficient
0.01
0.001
Steam released, kg
0 to 1.5 hr
2.94E+5
Leak flashing fraction

0 to 60 min
>60 min
0.04
0
1.68E+5
1.5

EAB
0 to 2 hours
LPZ
0 to 8 hours
5.1E-4
Table 15.3.9
2.2E-4
15-93
(6.48E+5 lb)
(3.7E+5 lb)

Table 15.3-4 (Sheet 1 of 2) Assumptions Used to Evaluate the Radiological Consequences
of the Rod Ejection Accident
Parameter
Value
Power level, MWt
3468
Peaking factor
1.65
Fraction of fuel rods failed
0.1
Fraction of fission-product inventory released to coolant

from perforated fuel rods
Iodines and noble gases
Alkali metals
0.1
0.12
Fraction of fuel rods melted
0.0025
Fraction of fission-product inventory released to coolant

from melted fuel rods
Iodines and alkali metals
Noble gases
0.5
1.0
Initial reactor coolant iodine activity, MBq/gm DEI-131
2.2
(60 Ci/gm)
Reactor coolant mass, kg
1.68E+5
(3.7E5 lb)
Duration of accident, days
30
Iodine chemical form fractions

Organic
Elemental
Particulate
0.0015
0.0485
0.95
Secondary system release path

Primary to secondary leak, kg/hr
Leak flashing fraction
Secondary coolant mass, kg
Duration of steam release from secondary system, sec
Steam released from secondary system, kg
Partition coefficient in steam generators
Iodine
Alkali metals
15-94
47.31
0.04
2.75E+5
1800
4.9E+4
0.01
0.001
(104.3 lb/hr)
(6.06E+5 lb)
(1.08E+5 lb)

of the Control Rod Ejection Accident
Parameter
Value
Containment leakage release path

Containment leak rate, % per day
0 to 24 hr
>24 hr
Airborne activity removal coefficients, hr-1
Elemental iodine
Organic iodine
Particulate iodine or alkali metals
Decontamination factor limit for elemental iodine removal
Time to reach decontamination factor limit, hr
EAB
0 to 2 hours
LPZ
0 to 8 hours
8 to 24 hours
1 to 4 days
4 to 30 days
0.10
0.05
1.7
0
0.1
200
3.1
5.1E-4
2.2E-4
1.6E-4
1.0E-4
8.0E-5
Table 15.3.9
15-95

Table 15.3-5 Assumptions Used to Evaluate the Radiological Consequences
of the Small Line Break Outside Containment Accident
Parameter
Value
Power level, MWt
3468

37
2.2
Ratio of iodine release rate from fuel during iodine

spike to that during steady-state operation
500
Reactor coolant mass, kg
1.68E+5
Duration of accident, min
30
Sample line break flow, L/m
492.1
Fraction of reactor coolant flashing
0.41

EAB
0 to 2 hours
LPZ
0 to 8 hours
5.1E-4
Table 15.3.9
(1.0 Ci/gm)
(60 Ci/gm)
(3.7E+5 lb)
(130 gpm)
2.2E-4
15-96

of the Steam Generator Tube Rupture Accident
Parameter
Value
Power level, MWt
3468

37
2.2
(1.0 Ci/gm)
(60 Ci/gm)
Steam generator in faulted loop

Initial water mass, kg
Iodine partition coefficient for breakflow
Iodine partition coefficient for secondary steaming
1.37E+5
567.8
1
0.01
0.001
(3.03E5 lb)
(150 gpd)
567.8
0.01
0.001
(150 gpd)
1.65E+5
3.24E+5
(3.64E+5 lb)
(7.15E+5 lb)
Steam generator in intact loop

Steam released, kg
0 to 2 hr
2 to 8 hr
Ratio of iodine release rate from fuel during
iodine spike to that during steady-state operation
335
1.74E+5
13.19

EAB
0 to 2 hours
LPZ
0 to 8 hours
8 to 24 hours
1 to 4 days
4 to 30 days
(3.84E+5 lb)
5.1E-4
2.2E-4
1.6E-4
1.0E-4
8.0E-5
Table 15.3.9
15-97

of the Loss-of-Coolant Accident
Parameter
Value
Power level, MWt
3468
Core activity released to the containment atmosphere, fraction

Gap Release
In-vessel Release
(00.5 hr)
(0.51.3 hr)
Nuclide Group
Noble gases
0.05
0.95
Iodines
0.05
0.35
Alkali metals
0.05
0.25
Tellurium group
0.05
Strontium and barium
0.02
Noble metals
0.0025
Cerium group
0.0005
Lanthanide group
0.0002
Organic
Elemental
Particulate
0.0015
0.0485
0.95
Primary containment leakage, weight percent/day

0 to 24 hours
>24 hours
0.1
0.05
Primary containment free volume, m3

Elemental iodine deposition removal coefficient, hr-1
Decontamination factor limit for elemental iodine removal
Removal coefficients for particulates
Accident duration, days
5.83E+4 (2.06E6 ft3)

1.7
200
Table 15.3-8
30

EAB
0 to 2 hours
LPZ
0 to 8 hours
8 to 24 hours
1 to 4 days
4 to 30 days
5.1E-4
2.2E-4
1.6E-4
1.0E-4
8.0E-5
Table 15.3.9
15-98

Table 15.3-8 Aerosol Removal Rates Used by Staff to Evaluate
Loss-of-Coolant Accident
Time Post Release

(hours)
Removal Rates
(hour-1)
0.00 to 0.37
0.37 to 0.87
0.87 to 1.37
1.37 to 1.87
1.87 to 2.37
2.37 to 2.87
2.87 to 3.37
3.37 to 6.87
6.87 to 24.00
0.945
0.540
0.430
0.600
0.855
0.585
0.575
0.480
0.430
15-99

to Control Room Operators Following a Design-Basis Accident
Parameter
Value
Accident release modeling

Locked rotor accident
Rod ejection accident
Small line break outside containment
Loss-of-coolant accident
Table 15.3-2
Table 15.3-3
Table 15.3-4
Table 15.3-5
Table 15.3-6
Table 15.3-7
Table 15.3-10
Control room free volume, m3
1.01E+3
Breathing rate of operators in control room

for the course of the accident, m3/sec
3.47E-4
Atmospheric dispersion values
Table 15.3-9a
Control room operator occupancy factors

0 to 24 hr
24 to 96 hr
96 to 720 hr
1.0
0.6
0.4
(3.57E+4 ft3)
Control room with emergency habitability system credited

Initial interval before actuation of emergency habitability system
54.5
Air intake flow, m3/min
Intake filter efficiencies
N/A
Interval with operation of emergency habitability system
Activity level at which emergency habitability system
is actuated, KBg/m3 of dose equivalent I-131
Flow from compressed air bottles, m3/min
Unfiltered inleakage, m3/min
Bottled air depletion time, hr
Interval after depletion of bottled air supply
Recirculation flow
15-100
74
1.7
0.14
72
48.1
N/A
N/A
(1925 cfm)
(2.0E-6
Ci/m3)
(60 cfm)
(5 cfm)
(1700 cfm)

to Control Room Operators Following a Design-Basis Accident
Parameter
Value
Time when compressed air restored, hr

Staff-calculated DBA control room dose results
168
Table 15.3-1
Control room and TSC with credit for supplemental air filtration mode of HVAC
Initial interval before actuation of supplemental air filtration
54.5
N/A
Time delay to switch from normal operation to filtration, sec
30
Filtered air intake flow, m3/min

Filtered air recirculation flow, m3/min
24.4
77.6
Filter efficiency, %
Elemental iodine
Organic iodine
Particulates
90
90
99
Unfiltered air inleakage, m3/min
2.55
(90 cfm)
28 mSv
4 mSv
(2.8 rem)
(0.4 rem)
4 mSv
<1 mSv
5 mSv
4 mSv
(0.4 rem)
(<0.1 rem)
(0.5 rem)
(0.4 rem)
27 mSv
29 mSv
32 mSv
14 mSv
(2.7 rem)
(2.9 rem)
(3.2 rem)
(1.4 rem)
(1925 cfm)
(860 cfm)
(2740 cfm)
Staff-calculated TEDE in control room and TSC

with credit for supplemental air filtration
Accident-initiated I spiking
Preexisting I spike
Locked rotor
With feedwater available
Without feedwater available
Rod ejection
Small line break
Accident-initiated I spiking
Preexisting I spike
Loss-of-coolant accident
15-101

Table 15.3-9a Atmospheric Dispersion Factors ( /Q) for Control Room Habitability Accident
Dose Analysis
/Q (sec/m3) at Control Room HVAC Intake for Identified Release Points

Plant Vent or
PCS Air
Diffuser
Ground
Level
Containment
Release
PORV and
Safety-Valve
Releases
Steamline
Break
Releases
Fuel-Handling
Area
02 hr
2.2E-3
2.2E-3
2.0E-2
2.4E-2
6.0E-3
28 hr
1.4E-3
1.4E-3
1.8E-2
2.0E-2
4.0E-3
824 hr
6.0E-4
6.0E-4
7.0E-3
7.5E-3
2.0E-3
2496 hr
4.5E-4
4.5E-4
5.0E-3
5.5E-3
1.5E-3
96720 hr
3.6E-4
3.6E-4
4.5E-3
5.0E-3
1.0E-3
/Q (sec/m3) at Control Room Door for Identified Release Points

Plant Vent or
PCS Air
Diffuser
Ground
Level
Containment
Release
PORV and
Safety-Valve
Releases
Steamline
Break
Releases
Fuel-Handling
Area
02 hr
6.6E-4
6.6E-4
4.0E-3
4.0E-3
6.0E-3
28 hr
4.8E-4
4.8E-4
3.2E-3
3.2E-3
4.0E-3
824 hr
2.1E-4
2.1E-4
1.2E-3
1.2E-3
2.0E-3
2496 hr
1.5E-4
1.5E-4
1.0E-3
1.0E-3
1.5E-3
96720 hr
1.3E-4
1.3E-4
8.0E-4
8.0E-4
1.0E-3
15-102

of a Fuel-Handling Accident
Parameter
Value
Power level, MWt
3468
Peaking factor
1.65
Number of fuel assemblies in core
157
Number of assemblies damaged
Reactor shutdown time before fuel movement, hr
24
Core fractions released from damaged rods

I-131
Other iodines
Kr-85
Other noble gases
0.08
0.05
0.10
0.05

Organic
Elemental
Particulate
0.0015
0.0485
0.95
Iodine effective pool decontamination factor
200

EAB
0 to 2 hours
LPZ
0 to 8 hours
5.1E-4
Table 15.3.9
15-103
2.2E-4

Table 15.3-11 Assumptions Used to Evaluate the
Radiological Consequences of Spent Fuel Pool Boiling
Parameter
Value
Initial activity in spent fuel pool, Bq

I-131
Other nuclides
1.18E+11 (3.18 Ci)

None modeled
Fuel stored in spent fuel pool from 10 years of operation (includes 68 assemblies from a
recent refueling)
Amount of I-131 diffusing into pool over 30-day period, Bq
7.18E+10
(1.94 Ci)
Initial pool water temperature, EC
49
(120 EF)
Time to initiate pool boiling, hr
8.8
Steaming rate, kg/hr

8.824 hr
2448 hr
4872 hr
7288 hr
88120 hr
120168 hr
>168 hr
7348
7257
7121
6994
6917
6772
6577
0.01

LPZ
8 to 24 hours
1 to 4 days
4 to 30 days
1.6E-4
1.0E-4
8.0E-5
Offsite breathing rate, m3/sec

8 to 24 hr
>24 hr
1.8E-4
2.3E-4
Table 15.3.9
15-104
(16,200 lb/hr)
(16,000 lb/hr)
(15,700 lb/hr)
(15,420 lb/hr)
(15,250 lb/hr)
(14,930 lb/hr)
(14,500 lb/hr)
Figure 15.3.6-1 Uncertainty Bands for Aerosol Removal Coefficients
Table 15.3.6-1 Aerosol Removal Coefficients as Calculated by Uncertainty Analysis (hr-1)
15-105

Chapter15 PDF

Uploaded by

Copyright:

Available Formats

Chapter15 PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Chapter15 PDF

Uploaded by

Copyright:

Available Formats

15.

TRANSIENT AND ACCIDENT ANALYSES

increase in heat removal from the primary system

Transient and Accident Analyses

when assumption of a non-safety-related system results in a more limiting transient

when a detectable and nonconsequential random, independent failure must occur in

when non-safety-related components are used as backup protection

Transient and Accident Analyses

Transient and Accident Analyses

15.1.3 Chapter 15 Loss of Offsite Power Assumptions

Transient and Accident Analyses

TWINKLEThis multidimensional spatial neutronics code uses an implicit

Transient and Accident Analyses

VIPRE-01As documented in WCAP-14565-P-A (proprietary), and WCAP-15306-NP-A

COASTAs documented in CENPD-98-A, Coast - Code Description, issued April

FACTRANAs documented in WCAP-7908-A (nonproprietary), FACTRAN - A

LOFTRANAs documented in WCAP-7907-P-A (proprietary) and WCAP-7907-NP-A

Transient and Accident Analyses

LOFTTR2As documented in WCAP-10698-P-A (proprietary) and WCAP-10750-NP-A

WCOBRA/TRAC-LBLOCAAs documented in WCAP-12945-P, Revision 2, Code

WCOBRA/TRAC-LTC and WGOTHICWCOBRA/TRAC is also used for the

Transient and Accident Analyses

Transient and Accident Analyses

Transient and Accident Analyses

15.2 Transient and Accident Analysis

Transient and Accident Analyses

Transient and Accident Analyses

Case 1minimum moderator feedback and manual reactor control

Transient and Accident Analyses

The lowest startup feedwater temperature is assumed.

Four RCPs are assumed to be operating initially.

No moisture is assumed in the blowdown steam.

Transient and Accident Analyses

Four RCPs are assumed to be operating initially.

Transient and Accident Analyses

No moisture is assumed in the blowdown steam.

Transient and Accident Analyses

15.2.1.6 Inadvertent Operation of the PRHR (DCD Tier 2, Section 15.1.6)

Transient and Accident Analyses

Transient and Accident Analyses

Transient and Accident Analyses

Transient and Accident Analyses

Transient and Accident Analyses

Transient and Accident Analyses

Transient and Accident Analyses

Transient and Accident Analyses

Transient and Accident Analyses

Transient and Accident Analyses

Transient and Accident Analyses

Transient and Accident Analyses

Transient and Accident Analyses

Transient and Accident Analyses

Transient and Accident Analyses

Transient and Accident Analyses

Transient and Accident Analyses

Transient and Accident Analyses

Transient and Accident Analyses

Transient and Accident Analyses

Transient and Accident Analyses