ML090360589

Enclosure 1 - Watts Bar, Unit 2 Severe Accident Management Alternatives Analysis

ML090360589
Person / Time
Site:	Watts Bar
Issue date:	01/21/2009
From:	Lutz R Westinghouse
To:	Office of Nuclear Reactor Regulation
References
LTR-RAM-I-08-062, Rev. 3, TAC MD8203
Download: ML090360589 (144)

Text

Enclosure 1 WBN Unit 2 Severe Accident Management Alternatives Analysis

Westinghouse Non-Proprietary Class 3

• Westinghouse To: File Date: January 21, 2009 cc:

From: Robert Lutz Your ref: N/A Ext: 412-374-4946 Our ref: LTR-RAM-I-08-062 Fax: 412-374-5099 Rev 3

Subject:

Watts Bar Unit 2 Severe Accident Mitigation Alternatives The attached is the transmittal to Tennessee Valley Authority (TVA) of the following deliverable:

Final Watts Bar Unit 2 Severe Accident Mitigation Alternatives (SAMA) analysis. The SAMA report is Attachment 1 to this letter, which is Westinghouse Proprietary Class 3. This letter report was revised to incorporate TVA comments.

Questions may be referred to the undersigned.

Author: Stephen A. Nass Risk Applications & Methods I ElectronicallyApproved*

Author: Mitch Waller Author: Robert J. Lutz, Jr.

Risk Applications & Methods I ElectronicallyApproved*

Manager: Melissa A. Lucci Risk Applications & Methods I ElectronicallyApproved*

• ElectronicallyApproved Records are Authenticated in the Electronic Document Management System

Westinghouse Non-Proprietary Class 3 Page 2 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report 1 INTRODUCTION The Watts Bar Unit 2 Final Supplemental Environmental Impact statement for the Completion and Operation of WBN Unit 2 (June 2007) was submitted to the NRC on February 15, 2008. NRC requested additional information by letter dated June 3, 2008.

By letter dated July 2, 2008, TVA committed to provide a WBN Unit 2 Severe Accident Mitigation Alternatives (SAMA) analysis consistent in scope and content with the SAMA analyses provided in support of recent license renewal applications. This report documents the development of a risk model to evaluate Unit 2 severe accidents, the identification of SAMA candidates, and a cost benefit analysis of those candidates. The results of this evaluation identify potentially cost effective hardware and procedure changes that will be considered for implementation.

2 LIST OF ACRONYMS ABGTS Auxiliary Building Gas Treatment System ABSCE Auxiliary Building Secondary Containment Enclosure AFW° Auxiliary Feedwater AOI Abnormal Operating Instruction AOT Allowed Outage Time ATWS Anticipated Transient Without Scram CCF Common Cause Failure CCS Component Cooling Water System (WBN System Designation)

CCW Component Cooling Water CDF Core Damage Frequency CT Combustion Turbine CVCS Chemical and Volume Control System DG Diesel Generator DWST Demineralized Water Storage Tank ECCS Emergency Core Cooling System EDG Emergency Diesel Generator EGTS Emergency Gas Treatment System EOP Emergency Operating Procedure EPSIL Emergency Preparedness Section Instruction Letter ERCW Emergency Raw Cooling Water (WBN System Designation)

ERG Emergency Response Guideline FPS Fire Protection System GOI General Operating Instruction HEP Human Error Probability HPCI High Pressure Injection System HRA Human Reliability Analysis HVAC Heating, Ventilation and Air Conditioning IPE Individual Plant Examination

Westinghouse Non-Proprietary Class 3 Page 3 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report IPEEE IPE for External Events ISLOCA Inter-System Loss of Coolant Accident LOCA Loss of Coolant Accident LOSP Loss of Offsite Power MCR Main Control Room MD Motor Driven MI Maintenance Instruction MSPI Mitigating System Performance Indicator NCP Normal Charging Pump PER Problem Evaluation Report PRA Probabilistic Risk Assessment PSA Probabilistic Safety Assessment PWST Primary Water Storage Tank RCIC Reactor Core Isolation Cooling RCP Reactor Coolant Pump RCS Reactor Coolant System RHR Residual Heat Removal RRW Risk Reduction Worth RWST Refueling Water Storage Tank SAMA Severe Accident Mitigation Alternative SBO Station Blackout SG Steam Generator SGTR Steam Generator Tube Rupture SI Safety Injection SQUG Seismic Qualification Users Group SW Service Water TD Turbine Driven UHI Upper Head Injection 3 METHODOLOGY The methodology selected for the SAMA assessment of Watts Bar Unit 2 (WBN2) is based on the Nuclear Energy Institute's (NEI) SAMA Analysis Guidance Document

[NEI 2005] and involves identifying SAMA candidates that have the highest potential for reducing plant risk and determining whether or not the implementation of those candidates is beneficial on a cost-risk reduction basis. The metrics chosen to represent plant risk include the core damage frequency (CDF), the dose-risk, and the economic cost-risk. These values provide a measure of both the likelihood and consequences of a core damage event. The SAMA assessment consisted of the following steps:

• Update the Watts Bar Nuclear Unit 1 (WBN1) Probabilistic Risk Assessment (PRA) Model to address peer review facts and observations (F&O), and to

Westinghouse Non-Proprietary Class 3 Page 4 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report account for Unit 2 operation. The model developed (WBN4SAMA) includes a Level 1 and Level 2 analyses of internal events including internal floods. The results of the combined Level 1 and Level 2 analysis are expressed as Release Category frequencies for input to the Level 3 analysis. The contributions of external events are incorporated as described in Section 4.3.

" Perform a Level 3 PRA analysis using the WBN4SAMA Level 2 internal events PRA output and site specific meteorology, demographic, land use, and emergency response data as input. The Level 3 analysis is performed using the MELCOR Accident Consequences Code System (WinMACCS) (Section 4.6).

" Calculate the monetary value of the unmitigated WBN Unit 2 severe accident risk using U.S. Nuclear Regulatory Commission (NRC) regulatory analysis techniques

[NRC 1997]. This becomes the maximum averted cost-risk (MACR) that is possible (Section 5).

• Identify potential SAMA candidates based on the WBN4SAMA PRA, the WBN1 Individual Plant Examination (IPE) [TVA 1992], the WBN1 Individual Plant Examination for External Events (IPEEE) [TVA 1998], and documentation from the industry and NRC (Section 6).

• Perform a Phase I SAMA Analysis by screening out SAMA candidates that are not applicable to the WBN2 design, are of low benefit in pressurized water reactors (PWRs) such as WBN2, candidates that have already been implemented at WBN2 or whose benefits have been achieved at WBN2 using other means, and candidates whose roughly-estimated cost exceeds the possible MACR (Section 7).

• Calculate the risk reduction attributable to each remaining SAMA candidate and perform a Phase II SAMA Analysis by comparing the averted cost-risk to a more detailed cost analysis to identify the net cost-benefit. PRA insights are also used to screen SAMA candidates in this phase (Section 8).

• Evaluate how changes in the SAMA analysis assumptions might affect the cost-benefit evaluation (Section 9).

• Summarize results and identify conclusions (Section 10).

4 SEVERE ACCIDENT RISK 4.1 WBN Unit 2 Level I SAMA Model The Watts Bar Unit 2 SAMA model was developed based on the latest Watts Bar Unit 1 model (WBN-REV4). Facts and Observations (F&O) from the WOG peer review performed on the Watts Bar Unit 1 PRA model were reviewed and the A and B level F&Os which may affect SAMA evaluation were identified. PRA model changes were

Westinghouse Non-Proprietary Class 3 Page 5 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report incorporated to resolve those F&Os. The resolution and status of all the A and B level F&Os are described in Section 4.4 Individual F&Os which were resolved as part of this effort include:

• Core damage arrest modeling in the Level 2 model was made consistent with the Level 1 model.

" Thermal-hydraulic evaluation of bleed and feed cooling success criteria requirements was revised to reflect updated bleed and feed cooling requirements (2 PORVs to support bleed and feed cooling with one SI pump).

• Added loss of plant compressed air initiating event LOPA.

• A sensitivity analysis of human actions was performed by setting all operator actions to failure. Numerous minor risk model changes were identified through this review.

• Detailed sequence evaluation of the top 100 model scenarios. This review was required for F&O AS-02. This review also provided the response to F&O QU-03, to verify logic for Auxiliary Feedwater (AFW) decay heat removal recovery.

This specific issue was resolved during the Watts Bar Unit 1 Rev 4 model update.

While numerous model observations were made, only ventilation system recovery was identified as a potential model change for the SAMA model.

Interviews were conducted with Sequoyah personnel to identify potential model changes required for dual unit operation. The use of Sequoyah personnel was appropriate because of their experience with dual unit operation and the similarity of the Sequoyah and Watts Bar designs. These interviews were used to establish the need for specific modeling of dual unit initiating events, beyond those modeled for Unit 1 alone, as well as the potential need to modify common systems to reflect dual unit operation. A review of Sequoyah's PRA model was also performed to identify differences in success criteria and initiating event logic for support systems. As a result of the interviews and the Sequoyah PRA model review, the following changes were incorporated into the Watts Bar SAMA Model:

• Changes to CCS to remove credit for the Unit 2 pumps from the Unit 1 model to reflect dual unit operation.

• Change to ERCW success criteria based on dual unit operation.

Following review of the various shared systems with the Sequoyah model (i.e.,

compressed air, ERCW, CCS, HVAC and AC and DC electric power), no further plant model changes to shared systems were identified as necessary for the WBN Unit 2 SAMA model.

Ns Westinghouse Non-Proprietary Class 3 Page 6 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report 4,2 WBN Unit 2 Level 2 SAMA Model The containment event tree used for the IPE was developed as a stand alone module. This event tree structure was reproduced and linked to the level 1 SAMA model. This migration to the SAMA containment event tree (CET) model included:

" Migrating split fraction rules to the new model. This included adjustment of split fraction designations for conflicting top event designations.

• Migrating interim variable (macro) assignments to the new model. This included designation of new macros for all previous CET initiating events.

• Migrating release category binning logic to the new model. Again, this included adjustment of top event designations to prevent conflict with the level 1 model top event designations.

• Translating the new model from plant damage state (PDS) initiator basis to establish level 1 model conditions for each PDS. In general, this was done by translating the previous initiators to using new macros to establish the model conditions.

• Incorporating the resulting CET module into the level 1 logic and resolving conflicting top event naming designations.

Release categories were retained from the IPE level 2 model and the binning of release categories into the four categories; Early Containment Failure, Containment Bypass, Late Containment Failure, and Intact Containment shown in Table 1, was also retained from the IPE model.

4,3 QuantitativeStrategy for ExternalEvents The SAMA PRA model is an internal events including internal flooding, at power model.

External events were evaluated in the IPEEE using seismic margins and the EPRI Fire Induced Vulnerability Evaluation (FIVE) methodologies. No vulnerabilities to external events were identified.

A multiplication factor of 2 is applied to the internal event results to account for the contribution to core damage from fire and other external events. The factor of two is based on a review of the SAMA submittals for a number of 4-loop Westinghouse plants including Wolf Creek [WCNOC 2006], Vogtle [SNC 2007], Catawba [DUKE 2001],

McGuire [2001a] and D. C. Cook [AEP 2003]. The first two were chosen because they represent recent applications while the latter three, while older applications, were chosen because they represent ice condenser plants.

Additionally, while the dominant core damage sequences will be different for seismic, fire and other external events, overall the contributions to release categories should be bounded by the internal events PRA sequences. For example, it is not expected that containment bypass sequences (SG tube ruptures and interfacing system LOCAs) will be dominant release sequences for fire and seismic initiators since these tend to result in loss

Westinghouse Non-Proprietary Class 3 Page 7 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report of power to operate and control plant equipment. Also, RCP seal LOCAs are a significant contributor to fire risk and SAMAs directed at maintaining RCP seal cooling are already considered for internal initiating events.

4.4 PRA Model Quality The A and B level F&Os from the WOG peer review performed on the Watts Bar Unit 1 PRA model are shown in Table 2. All A and B level F&Os were reviewed for impact on the SAMA analysis. The Watts Bar Unit 2 SAMA model incorporates the resolution of the A and B level F&Os described in Table 2.

4.5 WBN Unit 2 SAMA Model Results The core damage frequency result for the base case SAMA model is 1.537x10-5 , and the base case release category results are shown in Table 3.

4.6 WBN Unit 2 Level 3 SAMA Model 4.6.1 Analysis The WinMACCS computer code, Version 3.4 [NRC 2007] was used to perform probabilistic analyses of radiological impacts. The WinMACCS code is the current version of the MACCS2 code. A detailed description of the MACCS model is provided in NUREG/CR-4691 [NRC 1990]. The enhancements incorporated in MACCS2 are described in the MACCS2 User's Guide [NRC 1998].

Site-specific input parameters formed the basis for the analysis, including population distribution, economic parameters, and agricultural product. Plant-specific release data included nuclide release quantities, release timing and duration, release energy (thermal content), release frequency, and release category (i.e., early release, late release). The behavior of the population during a release (evacuation parameters) was based on declaration of a general emergency and the WBN Plant emergency planning zone (EPZ) evacuation time.

Generic input parameters given with the MACCS2 Sample Problem A, which includes the data used in NUREG 1150 [NRC 1989], supplemented the site-specific data.

This data, in combination with site-specific meteorology, were used to simulate the probability distribution of impact risks (exposure and economic cost) to the surrounding 80-kilometer (within 50 miles) population.

4.6.2 Population Distribution The population surrounding the WBN Plant site was estimated for the year 2040. The distribution was given in terms of the population at 10 distances, ranging from 0 miles to 50 miles from the plant, in the direction of each of the 16 compass points (north, north-northeast, northeast, etc.), a total of 160 segments. The population projections were determined using 2000 census population data. A map was prepared displaying county

Westinghouse Non-Proprietary Class 3 Page 8 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report and census tract boundaries for all counties partly or totally within the 50 mile boundary.

County population data for 2000 were allocated to the appropriate sectors, using census tracts to the extent feasible. For segments near the plant site, especially within 5 miles, aerial photos and TVA staff knowledge of the area were also used. The segments populations were projected for the year 2040 using growth rates from county population projections. The total projected population within 50 miles of the site was estimated to be 1,523,390 (see Table 4).

4.6.3 Economy and Agriculture Data Agriculture production information was generated using SECPOP 2000. SECPOP provides the WinMACCS model with required information on the crops season and shares (fraction of land devoted to the crop).

WinMACCS also requires spatial distribution of certain economic data (fraction of land devoted to farming, annual farm sales, fraction of farm sales resulting from diary production, property values of farm and non-farm land). SECPOP also produces this data for the site.

4.6.4 Radionuclide Release Core damage sequences that lead to containment failure (failure mode defined as bypass, early, and late) and release of radioactive materials to the environment are considered in this section. The core damage sequences from the Level 1 PRA are binned into plant damage states based on similar characteristics that control the accident progression following core damage and the timing and magnitude of fission product releases to the environment. The possible fission product releases are then binned into release categories that represent similar release magnitudes and timing. The Level 2 release categories are defined as conditional probabilities that, when combined with the plant damage state frequencies, yield release frequencies. The determination of the release characteristics for each release category is based on representative accident scenarios that reflect the post core damage behavior for the dominant sequence or sequences within a plant damage state. These core damage accident scenarios then become the major contributors to the release level categories associated with each of the containment failure modes.

The WBN2 Level 2 model is represented by a large containment event tree that is based on the NUREG-1 150 Level 2 assessment for Sequoyah. The event tree nodes and split fractions were reviewed to assure that the consequences, in terms of release frequencies, would be larger than would be expected with an updated Level 2 model. This will maximize the consequences, which in turn would maximize the economic benefits of the candidate SAMAs.

The release categories that are used in the SAMA assessment and examples of various accident scenarios leading to containment failure and/or bypass are presented below.

Westinghouse Non-Proprietary Class 3 Page 9 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report These release categories represent a consolidation of release categories from the WBN2 Level 2 PRA. The consolidation was performed to simplify the SAMA assessment by choosing the most severe release characteristics from the WBN2 Level 2 PRA for each of the three SAMA release categories. This provides the largest potential benefit in terms of fission product release prevention or mitigation for the alternatives in the Phase 1 assessment.

- Release Category I results from a reactor vessel breach with early containment failure.

- Release Category II results from a reactor vessel breach with containment bypass.

- Release Category III results from a reactor vessel breach with late containment failure.

- The remaining core damage sequences do not challenge the containment and result in an intact containment.

Table 5 shows the equilibrium reactor core radionuclide inventory at the time of a reactor trip. Table 6 provides important information on time to core damage, containment failure, and release duration.

Table 7 shows the fission product release fractions associated with each of the release categories. Table 3 provides a representation of the dominant accident scenarios that lead to each release category and the likelihood of their occurrence.

4.6.5 Evacuation Evacuation data, including delay time before evacuation, area evacuated, average evacuation speed, and travel distance, was obtained from the Tennessee Multi-JurisdictionalRadiologicalEmergency Response Planfor the Watts Bar Nuclear Plant, Annex H [TVA 2006]. For this analysis, the evacuation and sheltering region was defined as a 10-mile radial distance (the EPZ) centered on the plant. A sheltering period was defined as the phase occurring before initiation of evacuation procedures. During the sheltering period, shielding factors appropriate for sheltered activity were used to calculate doses to individuals in contaminated areas.

At the end of the sheltering period, residents would begin traveling out of the region.

Travel speeds and delay times were based on the evacuation data also found in the Tennessee Multi-JurisdictionalRadiologicalEmergency Response Planfor the Watts Bar Nuclear Plant, Annex H [TVA 2006]. General population evacuation times for the various areas within the 10-mile radius were averaged to determine an overall evacuation delay time and evacuation speed. Average evacuation speeds based on the most conservative general population evacuation times in an adverse weather condition were considered (see Table 8).

Westinghouse Non-Proprietary Class 3 Page 10 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report Based on the data cited above, an average evacuation speed of 1 meter per second following a sheltering and evacuation delay time of 45 minutes and 2.50 hours5.787037e-4 days <br />0.0139 hours <br />8.267196e-5 weeks <br />1.9025e-5 months <br /> were used.

These delay values are provided in the Tennessee Multi-JurisdictionalRadiological Emergency Response Plan for the Watts Bar Nuclear Plant, Annex H, [TVA 2006] and NUREG/CR-4551, Vol. 2 [NRC 1990]. In addition, consistent with the analysis in the NUREG-1150 evaluation of the Sequoyah Nuclear Plant, it was assumed that 99.5 percent of the population in the 10-mile EPZ would be evacuated.

For this analysis, it was conservatively assumed that persons residing farther than 10 miles away from the plant would continue their normal activities unless the following predicted radiation dose levels were exceeded. At locations where a 50-rem whole body effective dose equivalent in 1 week was predicted, it was assumed that relocation would take place after half a day. If a 25-rem whole body dose equivalent in 1 week were predicted, relocation of individuals in those sectors was assumed to take place after 1 day.

4.6.6 Meteorology Annual onsite meteorology data sets from 2001 through 2005 were used to prepare the sequential hourly data (8760 hours0.101 days <br />2.433 hours <br />0.0145 weeks <br />0.00333 months <br />) required for use in WinMACCS. The 2002 sequential hourly meteorology data was found to result in the largest risk based on sampling the population dose consequence for each year with a reference set of fission product releases and was used for all of the analyses presented below. The conditional dose from each of the other years was found to be within 20 percent of the chosen year.

The 2003 weather data set was found to result in the lowest population doses.

4.7 Severe Accident Risk Results Table 9 summarizes the risks of a severe accident (without any SAMAs implemented),

with mean meteorological conditions, within an 80-kilometer (50-mile) radius of the reactor site. The analysis assumes that a site emergency would have been declared early in the core damage accident sequence and that all nonessential site personnel would have evacuated the site in accordance with site emergency procedures before any radiological releases to the environment occurred. In addition, emergency action guidelines would be implemented to initiate evacuation of the public within 16.1 kilometers (10 miles) of the plant. The WinMACCS computer code models the evacuation sequence to estimate the dose to the general population within 80 kilometers (50 miles) of the accident. The frequency of each release category is given in Table 3. Table 10 shows the population dose risks (accident consequence multiplied by the release frequency) for each accident release category. These frequencies are based on WBN4SAMA PRA model.

Overall, the dose risk results are small. Completion and operation of WBN Unit 2 would not significantly change the risks evaluated for WBN Unit 1 because the principal change to Unit 1 accident mitigation capabilities is the loss of the Unit 2 CCS pumps as backup to the Unit 1 B Train CCS pumps, which is not risk significant. Changes to other systems, including shared systems, were found to have no significant impact on the

Westinghouse Non-Proprietary Class 3 Page 11 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report Unit 1 risks. This is consistent with the conclusions of NRC's Generic Environmental Impact Statement for License Renewal of Nuclear Plants (GEIS) [NRC 1996]. Accidents that could affect multiunit sites are typically initiated by external events. Severe accidents initiated by external events such as tornadoes, floods, earthquakes, and fires traditionally have not been discussed in quantitative terms in final environmental statements and were not considered in the GEIS [NRC 1996]. In the GEIS, however, NRC staff did evaluate existing impact assessments performed by NRC and the industry at 44 nuclear plants in the United States and concluded that the risk from beyond-design-basis earthquakes at existing nuclear power plants is small. Additionally, the staff concluded that the risks from other external events are adequately addressed by a generic consideration of internally initiated severe accidents. To account for the possible contribution of fires and other external events to the core damage frequency at Watts Bar, the internal events core damage frequency was doubled. Thus, all candidate SAMAs are evaluated using the averted costs based on doubling the core damage frequency from the internal events PRA analysis.

5 COST OF SEVERE ACCIDENT RISK / MAXIMUM BENEFIT This section explains how to monetize the severe accident consequences based on the formulas in the Nuclear Energy Institute's SAMA Analysis Guidance Document

[NEI 2005]. This analysis is also used to establish the maximum benefit that could be achieved if all risk for reactor operation were eliminated (i.e., accident consequences without SAMA implementation).

5.1 Off-Site Exposure Cost The annual off-site exposure risk was converted to dollars using the conversion factor of

$2,000 per person-rem, and discounted to present value using the following standard formula:

Wpha = C

• Zpha (1)

Where:

Wpha = monetary value of public health risk after discounting ($)

C = [ 1-exp(-rtf)]/r (years) tf = years remaining until end of facility life = 40 years r = real discount rate (as fraction) = 0.07 per year Zpha = monetary value of public health (accident) risk per year before discounting ($ per year)

Westinghouse Non-Proprietary Class 3 Page 12 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report The Level 3 analysis showed a baseline annual off-site population dose risk of about 3.30 person-rem. The calculated value for C using 40 years and a 7 percent discount rate is approximately 13.42 years. Calculating the discounted monetary equivalent of accident dose-risk involves multiplying the dose (person-rem per year) by $2,000 and by the C value (13.42). In this calculation, the delay until the initial time of operation is conservatively assumed to be zero. The calculated off-site exposure cost is estimated to be $88,541.

5.2 Off-Site Economic Cost The annual off-site economic risk was discounted to present value using the following standard formula:

Wea = C

• Zca (2)

Where:

Wea = monetary value of economic risk after discounting C = [I -exp(-rtf)]/r (years) tf = years remaining until end of facility life = 40 years r = real discount rate (as fraction) = 0.07 per year Zea = monetary value of economic (accident) risk per year before discounting

($ per year)

The Level 3 analysis showed a baseline annual off-site economic risk of $5,692.

Calculated values for off-site economic costs caused by severe accidents must be discounted to present value. This is performed in the same manner as for public health risks and uses the same C value. The resulting value is $76,365.

5.3 On-Site Exposure Cost The values for on-site (occupational) exposure consist of "immediate dose" and "long-term dose." The best estimate value provided in NUREG/BR-0184 [NRC 1997] for immediate occupational dose is 3,300 person-rem/event, and long-term occupational dose is 20,000 person-rem (over a 10-year clean-up period). The following equations are used to calculate monetary equivalents.

Westinghouse Non-Proprietary Class 3 Page 13 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report 5.3.1 Immediate Dose Wlo = R

• F *Dio

• C (3)

Where:

W1o = monetary value of accident risk avoided due to immediate doses, after discounting R = monetary equivalent of unit dose ($2,000 per person-rem)

F = accident frequency (1,54 x 10-5 events per year)

DIO = immediate occupational dose [3,300 person-rem per accident (NRC estimate)]

C = [1 - exp(-rtf)]/r (years) r = real discount rate (0.07 per year) tf = years remaining until end of facility life (40 years).

The best estimate of the immediate dose cost for WBN Unit 2 is:

W10 = 2,000 *1.54 x 10-5

• 3,300 * {[1 - exp(-0.07

• 40)]/0.07}

= $1,361 5.3.2 Long-Term Dose WLTO = R

• F

• DLTO

• C * {[1 - exp(-rm)]/rm} (4)

Where:

WLTO = monetary value of accident risk for long-term on-site doses, after discounting, ($)

R = monetary equivalent of unit dose ($2,000 per person-rem)

F = accident frequency (1.54 x 10-5 events per year)

DLTO = long-term dose [20,000 person-rem per accident (NRC estimate)]

C = [1 - exp(-rtf)]/r (years) r = real discount rate (0.07 per year)

Westinghouse Non-Proprietary Class 3 Page 14 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report tf = years remaining until end of facility life (40 years).

m = years over which long-term doses accrue (as long as 10 years)

Using values defined for immediate dose, the best estimate of the long-term dose is:

WLTO = 2,000

• 1.54 x 10-'

• 20,000 *{[1 - exp(-0.07 x 40)]/0.07}

{[1 -exp(-0.07

• 10)1/0.07

• 10}

= $5,931 5.3.3 Total On-Site Exposure The total occupational exposure is then calculated by combining equations 3 and 4 above.

The total accident related on-site (occupational) exposure risk (Wo) is:

Wo = WIo+WLTO=($l,36l +$5,931)=$7,292 5.4 On-Site Economic Cost On-site economic cost includes cleanup and decontamination cost, and either replacement power cost or repair and refurbishment cost.

5.4.1 On-Site Cleanup and Decontamination Cost The total undiscounted cost of a single event in constant year dollars (CcD) that NRC provides for cleanup and decontamination is $1.5 billion [NRC 1997]. The net present value of a single event is calculated as follows:

PVcD = [CcD/m] * {[l-exp(-rm)]/r} (5)

Where:

PVCD = net present value of a single event ($)

CCD = total undiscounted cost for a single accident in constant dollar years r = real discount rate (0.07) m = years required to return site to a pre-accident state The resulting net present value of a single event is:

PVcD = [$1.5 x 10' / 10 years] * {[1-exp(-0.07*10)]/0.07}

= $1.08 x 109 .

Westinghouse Non-Proprietary Class 3 Page 15 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report The NEI 05-01 'uses the following equation to integrate the net present value over the average number of remaining service years:

UCD = PVCD

• C (6)

Where:

UCD = total cost of cleanup and decontamination over the analysis period ($-

years)

PVcD = net present value of a single event ($1.08 x 109)

C = [1 - exp(-rtf)]/r (years) r = real discount rate (0.07 per year) tf = years remaining until end of facility life (40 years).

The resulting net present value of cleanup integrated over the license term is UCD = $1.08 x 109 * {[1-exp(-0.07*40)]/0.07}

= 1.45 x 10°0 $-years 5.4.2 Replacement Power Cost Long-term replacement power costs were determined following NRC methodology in NUREG/BR-0 184 (NRC 1997). The net present value of replacement power for a single event, PVRP, was determined using the following equation:

PVRp = [B/r] * [1 - exp(-rtf)] 2 (7)

'Where:

PVpp = net present value of replacement power for a single event, ($)

r = real discount rate (0.07) tf = 40 years (license period)

B = a constant representing a string of replacement power costs that occur over the lifetime of a reactor after an event (for a 91OMWe "generic" reactor, NUREG/BR-0 184 uses a value of $1.2E+8) ($/yr)

= $1.2 x 10'

• 1160/910 = $1.53 x 108 for WBN Power level of 1160 MWe.

Westinghouse Non-Proprietary Class 3 Page 16 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report The resulting net present value of a single event is:

2 PVRP = [$1.53 x 108/0.07] * [1 - exp(-0.07*40)]

= $1.93 x 109.

To attain a summation of the single-event costs over the entire license period, the following equation is used:

URP = [PVRP/r] * [1 - exp(-rtf)] 2 (8)

Where:

URP = net present value of replacement power over life of facility ($-year) r = real discount rate (0.07) tf = 40 years (license period)

The resulting net present value of replacement power integrated over the license term is 2

URP = [$1.93 x 109/0.07] * [1-exp(-0.07*40)]

= 2.43 x 1010 $-years 5.4.3 Total On-Site Economic Cost The total on-site economic costs are calculated by summing cleanup/decontamination costs and replacement power costs, and multiplying this value by the internal events CDF.

On-site economic cost = (1.45 x 1010 $-years + 2.43 x 1010 $-years)

• 1.54 x 10-5/year

= $595,708 5.5 Total Cost of Severe Accident Risk / Maximum Benefit The sum of the baseline costs is as follows:

Off-site exposure cost = $88,541 Off-site economic cost = $76,365 On-site exposure cost = $7,292 On-site economic cost = $595,708 Total cost = $767,906

Westinghouse Non-Proprietary Class 3 Page 17 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report The total cost risk represents the maximum averted cost risk if all risk were eliminated.

The MACR ($767,906) is based on at-power internal events contributions.

The internal events MACR is doubled to account for external events contributions. The resulting modified MACR (MMACR) is $1,535,812 and was used in the Phase I screening process.

6 SAMA Identification The list of SAMA items evaluated for WBN is given in Table 16. The process used to identify these SAMA items is described below.

The first source used to identify SAMA items is NEI 05-01 "Severe Accident Mitigation Alternatives (SAMA) Analysis Guidance Document [NEI 2005]. Generic industry SAMAs that are to be considered are the 153 items that are identified in Table 14 of NEI 05-01. Next, the license renewal applications for several recent submittals were reviewed and any SAMA items identified were added to the list of items to be evaluated.

The plants reviewed were Cook [APS 2003], Catawba [DUKE 2001], McGuire [DUKE 2001a], Wolf Creek [WCNOC 2006], and Vogtle [SNC 2007]. The review of these plant license renewal submittals resulted in the addition of 105 SAMA items (items 154 through 258) for consideration.

Identification of WBN-specific items began with a review of the original WBN Individual Plant Examination (IPE) [TVA 1992] and the WBN Individual Plant Examination for External Events (IPEEE) [TVA 1998]. The list of potential plant improvements from Section 6 of the IPE was reviewed and 12 additional SAMA items (items 259 through 270) were added. No potential improvements were identified from the IPEEE analyses.

Additional WBN-specific items included a review of the important systems and basic events. Each system and basic event with a risk reduction worth greater than 1.02 was reviewed to identify any potential SAMAs. In total, 13 new SAMA items (items 271 through 283) were generated from the importance review. Further review of the top 100 dominant sequences did not identify any additional candidate SAMAs.

As a result of the reviews described above, 283 potential SAMA candidates were identified. A complete listing is contained in Table 16.

6.1 Industry SAMA Analysis Review The SAMA identification process for WBN Unit 2 included review of the standard list of PWR SAMA candidates from NEI's Severe Accident Mitigation Alternatives (SAMA)

Analysis - Guidance Document [NEI 2005] as well as selected industry SAMA submittals. Submittals from Ice Condenser plants as well as recent 4-loop PWRs were included in the review. While many of these SAMAs are ultimately shown not to be applicable to WBN or not to be cost beneficial, they capture potentially important

Westinghouse Non-Proprietary Class 3 Page 18 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report changes not identified for WBN due to PRA modeling differences or SAMAs that represent alternate methods of addressing risk.

Phase I SAMAs were included from the following U. S. nuclear power sites:

• Cook [APS 2003]

• Catawba [DUKE 2001]

" McGuire [DUKE 2001a]

• Wolf Creek [WCNOC 2006]

• Vogtle [SNC 2007]

6.2 WBN IPE The WBN1 IPE did not identify any plant vulnerabilities. However a PRA screening was performed to examine major contributors to either the total core damage frequency or the early release frequency.

For individual initiators, single component failures, or single operator actions, potential enhancements were evaluated if they contributed more than 5 x 10-5 per reactor-year to the core damage frequency. Potential enhancements of a single system train were evaluated further if they contributed more than 1 x 10-4 per reactor-year to the total core damage frequency.

The results for Watts Bar lead to the conclusion that there were three contributors to the total core damage frequency that exceed the PRA screening criteria for consideration of potential enhancements. Loss of offsite power and the total loss of CCS initiating event categories each contribute greater than 5 x 10-5 to the total core damage frequency.

Additionally, failure of operator action to trip the RCPs in the event of a loss of CCS train A contributes greater than 5 x 10-5 to the total core damage frequency.

The options for potential enhancements were organized in terms of changes to procedural and plant hardware features.

6.2.1 ENHANCED PROCEDURES/OPERATOR ACTIONS The following procedure enhancements were suggested in the Watts Bar IPE

1. For addressing a loss of CCS train A, consideration should be given to revising AO-15, "Loss of Component Cooling Water," to facilitate stopping the RCPs on loss of CCS train A to minimize the potential for RCP seal damage due to pump bearing failure.

Westinghouse Non-Proprietary Class 3 Page 19 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report

2. Also, in the event of a total loss of CCS, clearer guidance on the desirability of cooling down the RCS prior to a seal LOCA developing to minimize the potential for seal damage should be considered. In general, additional training on the loss of CCS initiator is suggested'.

3. In the event of a loss of offsite power followed by the failure of both shutdown boards on one unit, the procedures would be enhanced by adding the guidance to align the C-S diesel generator (i.e., the fifth diesel generator) to one of the shutdown buses not powered in the accident sequence due to the loss of a normally aligned diesel generator 2. This alignment could be accommodated by including a reference to the spare diesel generator in AOI-35, "Loss of Offsite Power."

6.2.2 ENHANCED PLANT HARDWARE The following plant hardware enhancement was suggested in the Watts Bar IPE.

1. A potential improvement that could be evaluated is a plant change to provide connections for both centrifugal charging pumps, on both units, to the ERCW system for lube oil cooling in the event of a loss of CCS cooling to the associated pump. Currently, this capability is only available for centrifugal charging pump A on Unit 1. A sensitivity study shows that this could result in a decrease of about 4% in the total CDF.

6.2.3 ADDITIONAL INSIGHTS AND RECOMMENDATIONS Additional insights were presented in the IPE based on sensitivities to various scenarios.

The recommendations offered were not associated with significant plant vulnerabilities and were below the PRA enhancement criteria for further evaluation. The insights and recommendations listed below were viewed as additional considerations.

1. Enhancements to the operator training and procedures for responding to failures of support systems could potentially be beneficial, with emphasis on anticipating problems and coping.

2. Ventilation has been conservatively modeled in this study. Area ventilation is provided to the motor-driven AFW pumps and the CCS pumps from multiple systems serving the plant elevation where these pumps are located. Beyond design basis concurrent failures of the available WBN1 ventilation is assumed Later assessments of RCP seal behavior following a loss of all seal cooling shows that this time is too short to support operator actions.

2 Following completion of the IPE, it was determined that the 5 thdiesel was not cost-beneficial and completion of this feature was not pursued by TVA.

Westinghouse Non-Proprietary Class 3 Page 20 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report to impact the long-term availability of the AFW and CCS. An evaluation of the CCS/AFW area cooling requirements could be performed which could reduce this interdependence by crediting natural convection and availability of other coolers at this plant elevation.

3. In the event of a loss of ERCW, which would eventually lead to a loss of CCS cooling, additional guidance on the relationship of CCS to ERCW and the desirability of eliminating CCS loads to extend the time of suitable CCS temperatures is a potential consideration for evaluation. This could be accomplished by revising AOI-13, "Loss of ERCW," to alert the operators to shed CCS loads prior to CCS heatup.

4. During a loss of all AC, the steam generator power-operated relief valves (PORV) are to be locally operated to depressurize the steam generators, thereby cooling down the RCS. The addition of provisions for remote operation of these valves could potentially be beneficial due to the high area temperatures that may be encountered.

5. In the event of a loss of CCS cooling to the charging pumps, the time available for operation of the pumps would be limited by the loss of lube oil heat exchanger cooling. To extend the time available to protect the pumps, consideration could be given to increasing the oil capacity.

6. Losses of RCP seal cooling could potentially be reduced if the RCP thermal barrier cooling dependence on component cooling water, which is required for the charging pumps that provide RCP seal injection, could be eliminated.

7. Ventilation for the 480V board room that contains the unit vital inverters is provided by one train of ventilation. The PRA model relies substantially on recovery actions by the operators. Consideration could be given to providing two trains 3.

8. From a severe accident point of view, one potential change, for consideration, would be the delaying of containment spray operations relative to the Phase B condition. Currently, containment sprays actuate immediately in response to a Phase B condition, and air return fans (ARF) actuate after a 10-minute delay.

This is currently a requirement of the design basis LOCA where switchover to containment spray recirculation occurs prior to ice melt; thereby limiting pressure increases below containment design pressure. Modular Accident Analysis Program analyses of representative core damage sequences indicate 3 SAMA 269 describes the changes implemented for this issue.

Westinghouse Non-Proprietary Class 3 Page 21 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report that actuation of the containment sprays while ice remains in the ice condenser has little impact on severe accident containment performance and may be detrimental in that operation of the sprays rapidly depletes the inventory of the RWST, making its contents unavailable for vessel injection. Since many scenarios have successful injection for core cooling but failure at recirculation, the rapid depletion of the RWST due to spray operation accelerates the time to core damage. Therefore, an evaluation balancing the severe accident versus design basis requirements could be made.

6.3 WBN IPEEE The Watts Bar IPEEE evaluated seismic events, internal fire events, and other external events. The only vulnerability identified by the IPEEE has already been corrected as described in 6.3.3. The results of the IPEEE in the three areas is shown below.

6.3.1 SEISMIC EVENTS During the performance of the IPEEE Seismic Margins Assessment the Seismic Review Team (SRT) did not identify any significant concerns with the plant configuration control. Various minor maintenance and housekeeping issues were identified and were dispositioned and work requests (WR's) were written as needed.

No changes in maintenance, operating and emergency procedures, surveillance, staffing, or training programs were identified due to the evaluation performed for the seismic event.

6.3.2 INTERNAL FIRE EVENT No significant plant improvements were identified during the systematic evaluation of the internal fire event. The existing plant configuration and procedures adequately provide sufficient margins for the internal fire event. No changes to the physical configuration, maintenance, operating and emergency procedures, surveillance, staffing, or training programs were identified due to the evaluations performed for the internal fire event.

SAMAs 142, 143,144, 145, 146, and 256 are included in the Phase 1 analysis to specifically address potential fire risks.

6.3.3 OTHER EXTERNAL EVENTS During the systematic evaluation of the other external events, one configuration related condition was identified by the walkdown team as needing further attention:

During the walkdown, it was confirmed that Category I building entrances and exterior openings in walls and slabs are protected against tornado generated missiles which could penetrate and hit safety related equipment.

The only exception was an opening in the concrete canopy on the Unit 2

Westinghouse Non-Proprietary Class 3 Page 22 of 142 Our ref: LTR-RAM-J-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report side of the Auxiliary Building. This opening had the potential to allow tornado missiles to penetrate the Auxiliary Building from the Unit 2 area.

Thus, Problem Evaluation Report (PER) WBPER970050 was initiated to evaluate and provide necessary corrective action. The resolution was to design and install a steel shield plate over the opening to provide the required protection. That modification is complete.

No other plant improvements were identified during the evaluation as needed. The existing plant configuration and procedures adequately provide sufficient margins for the other potential severe accident external events. No other changes to the physical configuration, maintenance, operating and emergency procedures, surveillance, staffing, or training programs were identified due to the evaluations performed for the other external events.

6.4 WBN Unit 2 PRA Importance List Review The systems and basic events that have a risk reduction worth greater than 1.02 were reviewed to identify potential SAMAs. Table 11 lists the systems that have a RRW greater than 1.02 relative to CDF.

Table 12 lists the systems that have a RRW greater than 1.02 relative to LERF. Table 13 lists the basic events that have a RRW greater than 1.02 relative to CDF. And Table 14 lists the basic events that have a RRW greater than 1.02 relative to LERF.

The SAMA candidates identified through this review are identified in Table 15.

6.5 List of Phase I SAMA Candidates The initial list of SAMA candidates to be evaluated is presented in Table 16.

7 PHASE I SAMA ANALYSIS The purpose of the Phase I analysis is to use high-level knowledge of the plant and SAMAs to preclude the need to perform detailed cost-benefit analyses on them. The following screening criteria were used:

• Not Applicable: If a proposed SAMA does not apply to the WBN design, it is not retained.

• Already Implemented: If the SAMA or equivalent was previously implemented and is accounted for in the PRA model, it is not retained.

• Combined With Another SAMA: If a SAMA is similar in nature and can be combined with another SAMA to develop a more comprehensive or plant specific SAMA, only the combined SAMA is further evaluated.

Westinghouse Non-Proprietary Class 3 Page 23 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report

• Excessive Implementation Cost: If the estimated cost of implementation is greater than the modified Maximum Averted Cost-Risk, the SAMA cannot be cost beneficial and is screened from further analysis.

• Very Low Benefit: If the SAMA is related to a non-risk significant system which is known to have negligible impact on the risk profile, it is not retained Table 16 provides a description of how each SAMA was dispositioned in Phase 1. Those SAMAs that required a more detailed cost-benefit analysis are evaluated in Section 8.

8 PHASE IISAMA ANALYSIS The purpose of the Phase II analysis is to perform a cost-benefit analysis on the SAMAs that were not screened out in Phase I. The Phase I screening resulted in 18 SAMAs retained for further analysis. The risk benefit for each of these was analyzed using the PRA model described in Section 4. The cost of implementation of the SAMAs was estimated to identify those SAMAs that are potentially cost beneficial. The results of the Phase II analysis are shown in Table 17 and are described below.

SAMA 4: Improve DC bus load shedding.

==

Description:==

The SBO procedure includes shedding DC loads to extend battery availability. This SAMA evaluates the potential for enhancement to shed additional loads to extend battery life until AC power is recovered.

Risk Benefit: The risk benefit was bounded by calculating the change due to assuming AC power is always recovered prior to battery failure. The risk model was revised to set the offsite power recovery top event (OGRI) to guaranteed success. The resulting CDF is 1.493x10- 5. Calculating the averted risk cost relative to the base case using the method described in Section 5 results in a net benefit of $83,399.

Cost: The TVA estimated cost of this SAMA is $31,675.

SAMA 8: Increase training on response to loss of two 120V AC buses which causes inadvertent actuationsignals.

==

Description:==

Training is conducted on inadvertent Safety Injection, and loss of a single AC bus, however not on the loss of two 120V buses. This SAMA evaluates potential improvements in this operator training for loss of a second 120V bus.

Risk Benefit: The risk benefit was bounded by eliminating the contribution of the loss of 120V bus initiators. The risk benefit was calculated by removing the consequences of a loss of each of the single bus initiator events from the base case consequences. The resulting CDF is 1.516x10 5 . Calculating the averted risk cost relative to the base case using the method described in Section 5 results in a net benefit of $21,469.

Westinghouse Non-Proprietary Class 3 Page 24 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report Cost: The TVA estimated cost of this SAMA is $26,773.

SAMA 32: Add the ability to automatically align emergency core cooling system to recirculationmode upon refueling water storage tank depletion.

==

Description:==

Low pressure ECCS automatically aligns for recirculation from the containment sump, however the high head recirculation is manual. This SAMA evaluates potential design improvements to automatically align high head recirculation.

Risk Benefit: The risk benefit was bounded by calculating the change due to assuming that swapover to high pressure recirculation was always successful. The risk model was revised to set the top event for transfer to high pressure recirculation (RRH) to guaranteed success. The resulting CDF is 9.329xl 0-6. Calculating the averted risk cost relative to the base case using the method described in Section 5 results in a net benefit of $530,264.

Cost: The TVA estimated cost of this SAMA is $2,100,000.

SAMA 45: Enhance procedural guidance for use of cross-tied component cooling or service waterpumps.

==

Description:==

Watts Bar has the capability to cross-tie CCS trains and ERCW trains, and a flood mode procedure exists to supply CCS from ERCW by installing a spool piece. This SAMA will review procedural guidance in AOI-15 for potential upgrades.

Risk Benefit: The risk benefit was bounded by calculating the change due to assuming that ERCW alignment to charging pump cooling was always successful. The risk model was revised by setting the top event for charging pump cooling recovery (CCPR) to guaranteed success. The resulting CDF is 1.432x10-5 . Calculating the averted risk cost relative to the base case using the method described in Section 5 results in a net benefit of

$89,003.

Cost: The TVA estimated cost of this SAMA is $31,675.

SAMA 46: Add a service waterpump.

==

Description:==

An alternate pump exists that can be temporarily connected to the ERCW system to provide ERCW capability, however a permanent diesel driven 10,000 gpm pump could be installed at the IPS flush connection to provide increased ERCW availability.

Risk Benefit: The risk benefit was bounded by calculating the change due to assuming that ERCW pump LA-A was always successful. The risk model was revised to set pump lA-A to guaranteed success in alignments for top events AEBEI, AEBEX, and AEX. The

Westinghouse Non-Proprietary Class 3 Page 25 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report resulting CDF is 1.429x 10-5. Calculating the averted risk cost relative to the base case using the method described in Section 5 results in a net benefit of $102,000.

Cost: The TVA estimated cost of this SAMA is $1,042,511.

SAMA 56: Install an independent reactor coolant pump seal injection system, without dedicateddiesel.

==

Description:==

There is potential to install a small RCP seal injection pump in the PD pump room. This would be useful for loss of ERCW and loss of CCS which contributes 35% of the core damage. Suction piping, discharge piping, and power are available in the PD pump room. The current PD pump would be dismantled and a new low capacity high pressure pump would be installed. Room cooling requirements will also need to be evaluated.

Risk Benefit: The risk benefit was bounded by calculating the change due to assuming RCP seal injection is always successful when AC power is available. The risk model was revised by setting top event SE to guaranteed success when offsite power or a diesel generator is successful. Normal conditions are applied otherwise. The resulting CDF is 7.902x10- 6. Calculating the averted risk cost relative to the base case using the method described in Section 5 results in a net benefit of $675,053.

Cost: The TVA estimated cost of this SAMA is $2,400,000.

SAMA 70: Install accumulatorsfor turbine-driven auxiliaryfeedwaterpump flow control valves.

==

Description:==

The WBN turbine driven AFW pump flow control valves have a nitrogen supply that can be manually aligned. The nitrogen backup is not credited in SBO risk model. Installing accumulators would eliminate this manual action.

Risk Benefit: The risk benefit was bounded by calculating the change due to assuming the turbine-driven AFW pump level control valves (LCV) will not fail closed. The risk model was revised to set the LCV fails closed failure mode to guaranteed success in the Auxiliary Feedwater top events (AFC, AFX, and AF1OO). The resulting CDF is 1.538x10-5 . Calculating the averted risk cost relative to the base case using the method described in Section 5 results in a net benefit of $1,945.

Cost: The TVA estimated cost of this SAMA is $256,204.

SAMA 71: Install a new condensate storage tank (auxiliaryfeedwater storage tank).

Westinghouse Non-Proprietary Class 3 Page 26 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report

==

Description:==

The two unit CSTs are cross-tied so that they can supply either unit.

Installation of a new third CST would require a new pad, and piping to tie it in to the AFW supply.

Risk Benefit: There is no risk benefit for this modification. The same operator actions and procedures would be required to cross-tie the third CST as are presently available for the cross-tie of the two existing CSTs.

Cost: The TVA estimated cost of this SAMA is $1,706,586.

SAMA 87: Replace service and instrument air compressors with more reliable compressors which have self-containedair cooling by shaft drivenfans.

==

Description:==

Watts Bar is evaluating the status of the construction air compressors.

Permanent installation of this air compressor could improve the reliability of the station air system. Installation would need to consider HVAC requirements for the self-cooled compressor.

Risk Benefit: The risk benefit was bounded by calculating the change due to assuming the normal plant air system is always successful. The risk model was revised by setting top event for plant air (PD) to guaranteed success. The resulting CDF is 1.486x10-.

Calculating the averted risk cost relative to the base case using the method described in Section 5 results in a net benefit of $121,460.

Cost: The TVA estimated cost of this SAMA is $886,205.

SAMA 112: Add redundant and diverse limit switches to each containment isolation valve.

==

Description:==

Most of the containment isolation valves are air operated valves, however the ECCS valves are mostly motor operated. There is redundant valve status indication in control room. This SAMA will evaluate the number of CIVs where installation of limit switches may provide a benefit.

Risk Benefit: The risk benefit was bounded by calculating the change due to eliminating interfacing system LOCA initiating events. Interfacing LOCA due to failures other than containment isolation failure such as failure of valve disk integrity are unaffected by this change, however a maximum potential risk reduction was generated by requantifying the risk model without the ISLOCA initiating events. The resulting CDF is 1.535x10-5 .

Calculating the averted risk cost relative to the base case using the method described in Section 5 results in a net benefit of $4,565.

Cost: The TVA estimated cost of this SAMA is $691,524.

Westinghouse Non-Proprietary Class 3 Page 27 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report SAMA 136.' Install motor generatorset trip breakers in control room.

==

Description:==

Installing a low cost means for tripping the motor generator sets from the control room may reduce the risk from ATWS.

Risk Benefit: The risk benefit was bounded by calculating the change due to assuming the operator action to trip the reactor is always successful. In the WBN risk model, this operator action is modeled as part of operator action OEB. The risk model was therefore quantified with operator action OEB set to guaranteed success. The resulting CDF is 1.529x10- 5. Calculating the averted risk cost relative to the base case using the method described in Section 5 results in a net benefit of $7,397.

Cost: The TVA estimated cost of this SAMA is $241,795.

SAMA 156." Eliminate RCP thermal barrierdependence on CCW, such that loss of CCW does not result directly in core damage.

==

Description:==

Procedure AOI-7.07 provides direction to connect ERCW to CCS to supply the thermal barrier coolers. AOI-15 for loss of CCS should be revised to refer to AOI-7.07 Risk Benefit: The risk benefit was bounded by calculating the change due to assuming RCP seal injection is always successful when AC power is available. A bounding evaluation for this case was generated by revising the risk model by setting top event SE to guaranteed success when offsite power or a diesel generator is successful. Normal conditions are applied otherwise. The resulting CDF is 7.902x10-6. Calculating the averted risk cost relative to the base case using the method described in Section 5 results in a net benefit of $675,053.

Cost: The TVA estimated cost of this SAMA is $31,675.

SAMA 176: Provide a connection to alternateoffsite power source.

==

Description:==

Two 161 kV lines come into the Watts Bar switchyard from the nearby hydro plant switchyard. There are 5 redundant lines into the hydro switchyard. This SAMA would implement a design change to install an additional transmission line from the hydro plant.

Risk Benefit: The risk benefit was bounded by calculating the change due to assuming removal of grid-related failures from the frequency of loss of offsite power. From NUREG/CR-6890 (Table ES-2), grid related causes result in 1.86E-2 LOSP events per critical reactor year, compared with 3.59E-2 total LOSP frequency per critical reactor year. The risk model was revised by reducing LOSP frequency by 51.8%. The resulting

Westinghouse Non-Proprietary Class 3 Page 28 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report CDF is 1.5128x1 0 5 . Calculating the averted risk cost relative to the base case using the method described in Section 5 results in a net benefit of $42,254.

Cost: The TVA estimated cost of this SAMA is $9,126,460.

SAMA 256." Install Fire BarriersAround Cables or Reroute the Cables Away from Fire Sources.

==

Description:==

The Appendix R program rerouted permanent cables and conduits as necessary, however procedure enhancements for control of temporary cable impacts on fire protection will be reviewed. This SAMA only includes potential procedure enhancements, since hardware modifications were previously completed.

Risk Benefit: Although fire risk is not directly quantified in the risk model the benefit of enhancing the procedure controlling temporary alterations was estimated by conservatively reducing the consequences of all release categories except SGTR by 25%.

The resulting CDF is 1.144x10 5 . Calculating the averted risk cost relative to the base case using the method described in Section 5 results in a net benefit of $426,340.

Cost: The TVA estimated cost of this SAMA is $19,608.

SAMA 273: Provide a redundant path for ECCS suction from the RWST around check valve 62-504.

==

Description:==

Check valve 62-504 is a single failure point for ECCS injection and contributes 7% to CDF. This SAMA would implement a design change to install a parallel check valve with 62-504.

Risk Benefit: The risk benefit was bounded by calculating the change due to assuming check valve 62-504 is always successful. The risk model was revised to set check valve 62-504 to guaranteed success in common CVCS supply top event VS. The resulting CDF is 1.438x 10-. Calculating the averted risk cost relative to the base case using the method described in Section 5 results in a net benefit of $87,379.

Cost: The TVA estimated cost of this SAMA is $439,945.

SAM4 276: Provide an auto start signalfor AFW on loss of Standby Feedwaterpump.

==

Description:==

Incorporation of an AFW auto start signal on loss of the Standby Feedwater pump is under review. This SAMA would improve reliability of AFW for low power events (<18%) before Main Feedwater pumps are started. This SAMA is to implement a design change to install logic to start AFW on loss of flow from Standby Feedwater pump.

Westinghouse Non-Proprietary Class 3 Page 29 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report Risk Benefit: The risk benefit is only applicable to startup where a loss of SG heat sink could occur if the startup feed pump fails. The maximum benefit would be the elimination of all startup risk. This is modeled as reducing the risk for all initiators except SGTR by 1/365 assuming that startup is only performed for the equivalent of one day per year and that the startup risk is approximately equal to the at-power risk.

Therefore the averted risk cost is $5,926.

Cost: The TVA estimated cost of this SAMA is $615,605.

SAMA 2 79: Provide a permanent tie-in to the construction air compressor.

==

Description:==

The final disposition of the construction air compressor is under evaluation.

This SAMA is to implement a design change to use the construction air compressor in addition to the A, B, C and D compressors.

Risk Benefit: The risk benefit was bounded by calculating the change due to assuming the normal plant air system is always successful. The risk model was revised by setting top event for plant air (PD) to guaranteed success. The resulting CDF is 1.486x10-5 .

Calculating the averted risk cost relative to the base case using the method described in Section 5 results in a net benefit of $121,460.

Cost: The TVA estimated cost of this SAMA is $909,893.

SAMA 280: Add new Unit 2 air compressorsimilar to the Unit 1 D compressor.

==

Description:==

The final disposition of installing a compressor similar to the Unit 1 D compressor is under evaluation. This SAMA is to implement a design change to install a new compressor similar to the Unit 1 D compressor in place of current Unit 2 D compressor.

Risk Benefit: The risk benefit was bounded by calculating the change due to assuming the normal plant air system is always successful. The risk model was revised by setting top event for plant air (PD) to guaranteed success. The resulting CDF is 1.486x10-5 .

Calculating the averted risk cost relative to the base case using the method described in Section 5 results in a net benefit of $121,460.

Cost: The TVA estimated cost of this SAMA is $814,546.

9 UNCERTAINTY ANALYSIS Sensitivity cases were run for the following conditions to assess their impact on the overall SAMA evaluation:

• Use a real discount rate of 7 percent, instead of the 3 percent value used in the base case analysis.

Westinghouse Non-Proprietary Class 3 Page 30 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report

• Use the 95th percentile PRA results in place of the mean PRA results.

" Use alternate MACCS2 input variables for selected cases.

9.1 Real Discount Rate A sensitivity study has been performed in order to identify how the conclusions of the SAMA analysis might change based on the value assigned to the real discount rate (RDR). The original RDR of 7 percent has been changed to 3 percent, which could be viewed as conservative, and the MMACR was recalculated using the methodology outlined previously.

Implementation of the 3 percent RDR increased the MMACR by 81 percent compared with the case where a 7 percent RDR was used. This corresponds to an increase in the MMACR from $1,535,812 to $2,775,610.

The Phase 1 SAMA list was reviewed to determine if such a decrease in the MMACR would impact the disposition of any SAMAs. It was determined that no additional SAMAs could have been screened in the Phase 1 if an RDR of 3 percent were used in place of the 7 percent value.

The Phase 2 SAMAs are dispositioned based on detailed analysis. As shown in Table 18, the determination of cost effectiveness changed for one Phase 2 SAMA when the 3 percent RDR was used in lieu of 7 percent. However, the margin by which the SAMA becomes "cost beneficial" is small and it does not mean that this SAMA would be screened from consideration if a 3 percent real discount rate were applied in the SAMA analysis as other factors influence the decision making process, such as the 95th percentile sensitivity analysis.

9.2 95th PercentilePRA Results The results of the SAMA analysis can be impacted by implementing conservative values from the PRA's uncertainty distribution. If the best estimate failure probability values were consistently lower than the "actual" failure probabilities, the PRA model would underestimate plant risk and yield lower than "actual" averted cost-risk values for potential SAMAs. Re-assessing the cost benefit calculations using the high end of the failure probability distributions is a means of identifying the impact of having consistently underestimated failure probabilities for plant equipment and operator actions included in the PRA model. This sensitivity uses the 95th percentile results to examine the impact of uncertainty in the PRA model.

For WBN2, the results of the RISKMAN analysis of the Level 1 internal events model uncertainty analysis are provided below:

Westinghouse Non-Proprietary Class 3 Page 31 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report PARAMETER CDF per reactor-yr Mean 1.59E-05 5 percent 3.86E-06 50 percent 9.19E-06 95 percent 4.28E-05 The PRA uncertainty calculation identifies the 95th percentile CDF as 4.28E-05 per year.

This is a factor of 2.78 greater than the CDF point estimate produced by the WNB2 PRA.

For WBN2, RISKMAN model also includes an integral Level 2 model so that the impact of the Level 1 parameter uncertainty can be measured in terms of early releases. The results show:

PARAMETER LERF per reactor-yr Mean 3.80E-07 5 percent 1.05E-07 50 percent 2.37E-07 95 percent 9.83E-07 The PRA uncertainty calculation identifies the 95th percentile LERF as 9.83E-07 per year. This is a factor of 2.5 greater than the LERF point estimate produced by the WBN2 PRA.

As shown in Table 19, the determination of cost effectiveness changed for two Phase 2 SAMAs when the 95th percentile parameter uncertainty was used in lieu of the mean values. However, the margin by which the SAMA becomes "cost beneficial" is small and it does not mean that this SAMA would be screened from consideration if a 95th percentile LERF were applied in the SAMA analysis as other factors influence the decision making process.

9.3 WinMACCS Input Variations The MACCS2 model was developed using the best information available for the WBN site; however, reasonable changes to modeling assumptions can lead to variations in the Level 3 results. In order to determine how certain assumptions could impact the SAMA results, sensitivity assessments were performed on a group of parameters that has previously been shown to impact the Level 3 results. These parameters include:

• Meteorological data

• Population estimates

" Evacuation effectiveness

- Radionuclide release height

Westinghouse Non-Proprietary Class 3 Page 32 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report Meteorological data and radionuclide release height have been studied extensively (e.g.,

the Vogtle and Wolf Creek SAMA Uncertainty analyses) and have been shown to result in relatively small changes in overall risk.

On the other hand, population density and evacuation speed have been show to have the greatest effect on risk. Population density increases have been accounted for in the WBN SAMA assessments by using the projected 2040 population densities in the 50 mile radius of the WBN site. Smaller population increases would serve to reduce the cost effectiveness of various SAMA alternatives.

The impact of evacuation speed was investigated by performing a sensitivity analysis with MACCS2 where the evacuation speed was reduced from 2.2 mph (1 meter/sec) to 1.6 mph and another where the evacuation speed was increased to 3.4 mph. The results, in terms of impact on the baseline SAMA cost benefit are provided in Table 20. As shown in Table 20, the cost effectiveness of all SAMAs does not change with changes in evacuation speed. This is due to the relatively low contribution of offsite exposure cost to the overall cost as shown in Section 4.5.

10 CONCLUSIONS The benefits of revising the operational strategies in place at Watts Bar and/or implementing hardware modifications can be evaluated without the insight from a risk-based analysis. However, use of the PRA in conjunction with cost-benefit analysis methodologies provides an enhanced understanding of the effects of the proposed changes relative to the cost of implementation and projected impact on offsite dose and economic impacts.

The results of this study indicate that of the identified potential improvements that can be made at WBN, several are cost beneficial based on the methodology applied in this analysis:

" SAMA 4: Review station blackout procedures for improvements in DC load shedding.

" SAMA 45: Enhance procedural guidance for use of cross-tied component cooling or service water pumps.

• SAMA 156: Enhance procedural guidance for use of ERCW for RCP thermal barrier cooling..

- SAMA 256: Enhance procedure for controlling temporary alterations to reduce fire risk from temporary cables.

These SAMAs could be considered to be cost beneficial alone, but given the risk reduction provided by each SAMA, implementation of any one of them could make the

Westinghouse Non-Proprietary Class 3 Page 33 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report averted cost risk of implementation of the remaining SAMAs not cost beneficial as the relevant risk factors would be addressed. However, TVA commits to implementation of the identified procedure enhancements (SAMAs 4, 45, 156, and 256).

The results of the uncertainty analysis for this study indicate that one additional SAMA is cost beneficial:

- SAMA 8: Increase training on response to loss of two 120V AC buses.

TVA also commits to implementation procedure enhancements identified in SAMA 8.

Table 1 Definition and Causes of Containment Failure Mode Classes Iailure mude 1)fllu" andCas Early Involves structure failure of the containment before, during, or slightly after (within a few Containment hours of) reactor vessel failure. A variety of mechanisms can cause structure failure, including Failure direct contact of core debris with containment, rapid pressure and temperature loads, hydrogen combustion, and fuel coolant interaction (ex-vessel steam explosion). Failure to isolate containment or to provide early venting of containment after core damage also is classified as early containment failures.

Containment Involves failure of the pressure boundary between the high-pressure reactor coolant and low-Bypass pressure auxiliary system. For pressurized water reactors, steam generator tube rupture, either as an initiating event or as a result of severe accident conditions, will lead to containment bypass. In this scenario, if core damage occurs, a direct path to the environment can exist.

Late Containment Involves structural failure of the containment several hours after reactor vessel failure. A Failure variety of mechanisms can cause late structure failure, including gradual pressure and temperature increase, hydrogen combustion, and basemat melt-through by core debris. Venting containment late in the accident also is classified as a late containment failure.

Intact Involves no structural failure or bypass of the containment. If core damage occurs, fission Containment products are retained in the containment and there is no release to the environment.

Westinghouse Non-Proprietary Class 3 Page 34 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment I Final Watts Bar Unit 2 SAMA Report Table 2 Level A and B F&O Resolution IE-01 The IE database Notebook document does not completely identify all Closed - A new initiating event notebook None methodologies used in the frequency estimates or the basis for using the was created showing the basis and various methodologies. For example, the IE Data notebook only process used to calculate each initiating mentions performing frequency estimates based on plant specific and event frequency such that the results are generic data, where as in the WBN IPE report Section 3.3.1 several reproducible. Initiating events created initiator frequencies were estimated using fault tree solution. IE guidance via system fault trees are documented in should explain the process used to develop the initiating event their respective system notebooks.

frequencies and the basis for using industry accepted methodologies. The data, calculations, and results should be presented in the PSA report. The notebook does not contain enough detail to sufficiently reproduce all the results (i.e., the IPE report states certain frequencies were estimated using fault tree solution; however, in the notebook and the IPE report some of the frequencies appear to have been possibly obtained from using generic data). Detailed guidance that describes the process used and criteria for using the different methodologies should be provided such that the results can be reproduced. The data, calculations, and results should be part of the PSA report or a stand-alone calculation file. IE-06 has been combined into IE An explanation of the process used to identify and apply systematic techniques as plant specific fault tree models or FMEAs to quantify initiating event frequencies and recovery was not found in the IE documentation. Guidance should be provided that describes the process for developing initiating event frequencies and the basis for using the different methodologies. The guidance and documentation should provide sufficient detail to reproduce all results.

IE-02 PLG-1351, Initiating Event Database Notebook, Table 1-3 documents Closed - A new initiating event notebook None "Prior" Means that are updated with Plant Specific Data. There is no was created showing the basis and basis shown for the majority of the Prior Means. There is no match with process used to calculate each initiating either the PLG Data shown on Table 1-1 for PLG or NUREG/CR-5750. event frequency such that the, results are Some of the IE frequencies used are significantly higher than the NUREG reproducible. Initiating events created values (SGTR, inadvertent closure of all MSIVs). Some IE frequencies via system fault trees are documented in are significantly lower than the NUREG values (LLOCA, steam line their respective system notebooks.

breaks) but no numerical basis is shown. The bases for the calculations NUREG-5750 was used as the basis for should be provided. Document basis of Prior Means. Detailed discussion many of the prior means.

and calculation should be developed for deviation from NUREG/CR-5750 (or other referenced data sources).

Westinghouse Non-Proprietary Class 3 Page 35 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report IE-03 The initiating event analysis does not appear to have considered loss of Closed - TVA calculation WBNOSG4- An ORT success term was used HVAC as a potential initiator. Loss of HVAC was not included in the 242 shows that 6.9kV rooms remain less to show that operators would support system FMEA. Loss of HVAC was considered and dismissed in than 103.5F for at least 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> have tripped the reactor prior to a summary analysis, after the support system initiator analyses were done following loss of ventilation. Also, room heatup. This prevents a very short term action from (per Stillwell memo 1992). The current system notebook for AC power shutdown boards only supply mitigative being dependent on a very long states that room cooling is necessary for success of 6.9-kv boards, 480V plant features, not loads that would cause term heatup impact.

AC busses and 115V AC busses. If this is true, then it would seem that unit trip.

the loss of HVAC should be incorporated into (or at least further addressed in) the initiating event analysis. Clarify whether or not Loss of HVAC is an initiator for AC power rooms, CCS rooms, and ERCW rooms. Either add to the model if needed, or provide discussion in the documentation if not needed.

IE-05 The frequency for Loss of Offsite Power is updated with a Bayesian Closed - The initiating event LOSP, Loss None process. The plant specific data is listed as 0 failures in 20 years, i.e., the of Offsite Power, is based on exposure time for the switchyard is considered to be 20 years (since NUREG/CR 5750, was performed for the 1980). The claim is made, but not adequately substantiated, that the Revision 4 PSA model. The Revision 4 switchyard experience since 1980 is applicable to the current switchyard initiating event analysis documents that operation, thus allowing 20 years accumulated experience. There is no the distribution was Bayesian updated evidence provided that a) records for switchyard failures over the past 20 using RISKMAN with 0 events in 20 years were kept and are accurate (including partial failure), b) years. Note that the use of 20 years of switchyard configuration is the same now as it was during construction, data includes non operational time and c) electrical transients are the same now as in construction. Reducing the time from before commercial operation.

exposure time from 20 to 2.89 years (to reflect length of plant operation This distribution was Bayesian updated considered for PSA Rev 3) would increase the mean LOSP frequency using RISKMAN with I event in 6.25 about a factor of 3-4 depending on the assumptions of the Bayesian years. The resulting mean of 4.85E-analysis. Use 2.89 years for plant specific time for switchyard experience 02/reactor year was used in the Revision or provide stronger justification for use of an exposure time of 20 years 4 model. This analysis was compared to a for the switchyard. generic total loss of offsite power frequency from NUREG/CR-6928. The NUREG provides a frequency of 3.59E-02/reactor year. If the Bayesian update was to be performed with the updated prior generic frequency and only a 6 year plant specific data window, the resultant frequency would be lower than the Revision 4 LOSP frequency. Therefore the Revision 4 frequency is maintained.

Westinghouse Non-Proprietary Class 3 Page 36 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report IE-07 The Loss of Instrument Air initiator is not treated explicitly. It should Closed - A Loss of Plant Air initiating The Loss of Plant Air (LOPA) capture the dependency of Instrument Air and AFW flow control. Some event was incorporated into the model. initiating event was added to failures of instrument air could cause an initiator and fail AFW flow the model with a generic control. These do not appear to have been evaluated. Instrument Air to initiating event frequency of essential PSA loads is supplied by essential air and control air. The 9.81E-3. The guaranteed failure FMEA for support systems initiating events states that if control air is term for plant compressed air lost, essential air will supply loads. Loss of Instrument Air is dismissed top event PD in event tree as a special initiating event, but included as a cause of a MSIV closure, module MECH was changed to However, a loss of all air fails AFW flow control whereas the MSIV include an "INIT=LOPA" term.

closure event modeling assumes AFW flow control is operable. The initiating event LOPA was Quantify Loss of Instrument Air as an initiator. (This is needed to added to the initiating event complete the dependency analysis). group ALL and requantified at a quantification cutoff of IE-12.

Westinghouse Non-Proprietary Class 3 Page 37 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report AS-01 Accident Scenario Evaluation (Event Tree Structure): Since several Closed - The WOG 2000 Model for RCP None specific modeling issues have been identified, a comprehensive review of seal behavior following a loss of all seal the entire logic structure is recommended. The success criteria associated cooling has been implemented into Rev.

with the RCP Seal LOCA model is overly conservative. As currently 4 of the WBN PRA model (Reference the modeled, only failure of Thermal Barrier cooling to all four RCPs RCP Seal Injection and Thermal barrier combined with failure of RCP Seal Injection to any one of the four RCPs Cooler System Notebook). It appears results in a RCP Seal LOCA. With the logic modeled in this manner, that the failure to supply either seal individual pump RCP Seal failures due to loss of Thermal Barrier cooling injection or thermal barrier cooling to one to one pump combined with a loss of seal cooling to the same pump does RCP is modeled to result in failure of all not appear to be captured. This modeling technique is not representative RCP seals, even if only seal injection to of techniques currently used across industry to model RCP Seal LOCAs. only one pump is failed. This is This conservatism may result in a mis-representation of the importances obviously conservative modeling.

of non-RCP Seal LOCA related components. Additional modeling However, this conservatism should only practices associated with the handling of Common Causes Failures impact risk applications that use relative appear to be incorrect (see DE-01 for CCF concerns). This model risk measures such as RAW and Fussell-currently contains logic that has the potential to skew the RCP Seal Vesely and should not impact risk LOCA results and to also skew the importances of other plant systems. applications that use delta risk measures This may result in masking the true importances of some systems and such as SAMA. The only masking that components. Revise the RCP Seal LOCA model to ensure that all valid could potentially occur would be the combinations of failures associated with Thermal Barrier Cooling and identification of insights from the WBN Seal Injection Cooling to the same RCP result in a RCP Seal LOCA. PRA that might suggest a SAMA feature.

Suggest converting the RCP Seal LOCA model to the Westinghouse However, the evaluation of the SAMA Owners Group (WOG 2000) methodology once it is approved by the feature would not be impacted. The 21 NRC. gpm per pump leak is not a dominant contributor to core damage; the dominant seal failure leading to core damage is the 181 gpm leak with non-recovery of RCS make-up prior to battery depletion. In this case, the number of pumps experiencing the 182 gpm leak does not impact the results.

Therefore this F&O can be considered closed for the SAMA assessment.

Westinghouse Non-Proprietary Class 3 Page 38 of 142 Our ref. LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report

______ ------ Pescription~ Rshto~~tsJMdlApzt AS-02 There is no specific guidance document. The event sequences in the Closed - A detailed evaluation of the top None - The detailed evaluation original IPE which are referenced in the current revision are based upon 100 sequences was completed including of scenarios required for F&O the EOPs and AOPs in effect at the time of the original IPE. review of the event sequence logic and AS-02 has been completed.

Guidance; Accident Scenario Evaluation (Event Tree Structure): The top events The top 100 sequences This review provides the documentation of the plant model RISKMAN rule development was not represents 79% of the core damage response to F&O QU-03 sufficient for the majority of the plant model to allow determination of frequency and no issues were identified. (Verify logic for sequence 9 -

the rationale behind the development of each top events rule. The lack of A specific, rigorous evaluation of the operator recovers AFW decay documentation made it difficult to confirm the fidelity of the model rules. event tree rule structure was not heat removal at AFD 1 when Since these rules define the accident sequences and their dependencies, it performed. No specific errors were AFC is guaranteed failure).

.is critical in this type of model to carefully document and verify the identified through the closure of this or This issue was resolved during operation of the rules. Document the basis for the event tree rules and other F&Os. The large fraction of CDF the Rev 4 model update. This binning for all top events and macros; perform independent review of represented by the scenarios reviewed is shown in that previous each to confirm the basis. Perform a detailed evaluation that analyzes the provides a level of confidence in the scenario 9 or similar items no sequence of events that lead to core damage (50 top sequences minimum) reasonableness of the model results. longer appear in the model results.

The top 100 scenarios account for 79% of CDF and 28% of LERF and more than the first two decades (2.5E-6 to 8.8E-9 CDF) of ranked scenarios by CDF.

Review of this scenario list has shown that the dominant excessive LOCA scenario (RPV catastrophic failure -

scenario number 5) is assigned to a NOLERF endstate.

Subsequent review confirms that this is appropriate, given industry PTS work.

AS-03 Success Criteria and Bases: The Accident Sequence Notebook (PLG- Closed- The Rev 4 WBN PRA Success None 1339) does not completely describe the process used to 1) develop the Criteria were reviewed for accident sequences or 2) determine the success criteria associated with reasonableness against other the accident sequences. The notebook contains the statement that the Westinghouse 4-loop PWRs, including success criteria is based primarily on the FSAR Chapter 15 Analyses; the McGuire, Catawba and Cook ice

Westinghouse Non-Proprietary Class 3 Page 39 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report I ~~~Descriptioin 9R olulioiiStaius > ]lYMdelI Iiuct however, specific details identifying which Chapter 15 scenario each condenser containment plants. This is success criteria is based on are generally not provided. The rationale considered to be a valid comparison behind the individual success criteria is not provided, and references to basis. The key success criteria were specific supporting calculations (MAAP runs associated with success compared based on information in the criteria, FSAR assumptions, etc.) are not provided. A number of specific PWROG PRA Database R6. It was criteria in Table C-1 of PLG-1339 refer generally to NUREG-4550, but found that all of the WBN2 success not to particular assumptions or analyses in the study. A significant criteria, except the Bleed and Feed Issue number of references are simply left blank. This makes it difficult to raised in F&O TH-02, are similar to those check that appropriate assumptions have been made in establishing the used in the other PRAs. The Bleed and criteria. There is no guidance associated with how the rules were Feed success criteria are being assessed structured to reflect the defined success criteria. Guidance should be independently and the SAMA assessment provided that describes in sufficient detail the process used and identifies will use the new success criteria basis.

criteria for defining the accident sequences to be modeled. The bases for A more detailed comparison was the success criteria should be clear and traceable to supporting analyses completed against the Comanche Peak or assumptions. Information should be included on what the associated (CPSES) PRA Success Criteria and rules are and what they are designed to do. several instances of conservatism were found in the WBN Success Criteria:

-The very small LOCA required HP recirc for success whereas CPSES uses normal RHR or HP recirc as the success endstate, and

-The medium LOCA required 2 of 4 HP injection pumps whereas CPSES only requires 1 of 4 as a success state, Removing the conservatisms in the WBN success criteria would reduce the overall probabilities of the release category bins in the Level 2 assessment and therefore reduce the overall offsite consequences.

The impact of removing the conservatisms in the WBN2 model used to assess SAMA would be to reduce the maximum possible benefit attainable for any alternative. Thus, using conservative success criteria in the Level 2 model maximizes the possible benefit which could potentially result in additional

Westinghouse Non-Proprietary Class 3 Page 40 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report S........F&O -.. De...ipti.n.. Reso"utiWioSiatus .init M.d.

features being classified as cost-effective.

Therefore, it is concluded that the success criteria in the WBN2 PRA, except for the bleed and feed success which is being re-assessed for the SAMA assessment, are acceptable for use in the SAMA assessment without the completion of additional analyses to justify the WBN2 success criteria.

Therefore this F&O can be considered closed for the SAMA analysis.

AS-04 Accident Scenario Evaluation (Event Tree Structure): The guaranteed Closed - The applicable event trees were None success split fractions for turbine trip and reactor trip should be used only modified to correct this model deficiency.

when the occurrence of the IE ensures that these top events can be by-passed. Even though the rods have fallen and the turbine has tripped on previous trips categorized as RT or TT events, this does not ensure that these will occur on future trips. If plant conditions require a reactor trip, the likelihood that the rods are inserted must be questioned. As the frequencies for these IE are fairly large, this could lead to underestimation in the CDF contribution from ATWS and turbine trip failures. Remove RT and TT cases from these split fractions and assess the appropriateness of the remaining beneficial failures.

AS-09 Accident Scenario Evaluation (Event Tree Structure): Top Event CM Closed - Logic rules were modified to Success criteria changes were (Core Melt) - allows success when top event OB (operators align bleed questioning of BF with OB and an made to reflect the requirement and feed) and one train of either safety injection or charging is successful. injection pump for bleed and feed for two PORVs when It does not require success of top event BF (hardware for bleed and feed - success. Top event BF is anded with performing bleed and feed PORVs). Top event BF should be questioned - revise top event BF rules OB=S in CMS term. In ET module cooling using SI pumps.

to require at lease one charging pump and one safety injection pump. GTRANI, top event OB is questioned Require top event BF success when taking credit for Top OB (operators before BF, but BF is set to guaranteed align bleed and feed) success. [See F&O TH-10 for discussion on failure on OB failure. Macro BFSUPP in number of pumps required] event tree module GTRAN 1 is necessary for BF success and requires one train of charging OR one St pump.

See resolution of TH-10. This F&O can be considered closed.

Westinghouse Non-Proprietary Class 3 Page 41 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report AS-11 Accident Sequence Evaluation (Event Tree Structure): The loss of Closed - There is currently no None multiple 120VAC panels can cause spurious actuation of key plant safety dependency identified in the dependency equipment. The failure of panels 1-1 and 1-3 may have consequences matrix that would require a multiple 120 such as the spurious actuation of RHR pumps and automatic swap over VAC failure be modeled as an initiating without water in the sump (which leads to pump failure). This could also event. Also, no common mode failures lead to the automatic closure of the MSIV or other negative impacts have been identified that would associated with a spurious safety systems actuation. The failure of necessitate the modeling of loss of multiple 120VAC buses has occurred in the industry-especially due to the multiple 120 VAC failures as separate failure of the automatic transfer feature. This is not modeled as an initiating events. The potential for initiator and is not in the current set of rules for the failure of multiple secondary failure of more than one 120 inverters, post trip. If this is a valid issue, model loss of multiple VAC buses is reflected in the model 120VAC panels as an IE; model the consequences of the loss of both structure and results.

120VAC panels in the rules, post trip. If this issue can occur, then this might be a Level A significance depending on its impact to the baseline PRA and current applications. If this is not an issue, research which proves this is not a problem should be documented. If it is determined that this issue cannot occur, this is effectively a Level D significance (i.e.,

documentation of the resolution).

Westinghouse Non-Proprietary Class 3 Page 42 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report AS-12 Success Criteria and Bases - The RISKMAN rules are coded in such a Closed - Detailed review of top 100 None way that it is difficult for a reviewer to determine how the rules function, scenarios from the Rev 4 model output or verify the success criteria, the recoveries modeled, and the general have been performed and identified (and accuracy of the rules. This is particularly important for the WBN model otherwise noted) errors have been that uses "branch-everywhere" logic, in which RISKMAN rules provide corrected. No further changes required the only basis for checking the model. Correct rules are critical to ensure for the SAMA model.

that the risk model reflects the as-built as-operated plant. Improve the documentation of the rules and associated information. Consider implementing improvements such as the following: 1) Define all Macros used in plain English; 2) Eliminate the use of doubly-defined split fractions; 3) Code the rules in the SAME order as the top event list; 4)

Perform and document detailed checking of the split fraction rules; 5)

Document all bypasses as comments within the rules or use a TRUE branch everywhere tree. A detailed evaluation is recommended of the top sequences (at least the top 50) that analyzes each for the sequence of events that lead to core damage. This evaluation should document the basis for each important systems failure. For example: On a loss of ERCW all air compressors are lost. The evaluation should note these dependent failures. The sequences should then be evaluated for validity.

Invalid and unrealistic sequences will require model changes to prevent invalid sequences.

SY-03 System Model Structure (Fault Tree): The system notebooks reviewed Closed - Documentation of a systems None (Safety Injection, Chemical and Volume Control, and Main Steam) do not analysis guidance document is not provide guidance for performance of the systems analysis and do not necessary for the SAMA evaluation.

reference any external methodology documents. Guidance is important due to the complexity of identifying top events and split fractions in a consistent manner. As a minimum, the original IPE guidance for systems analysis should be referenced. An updated systems analysis methodology document could be more useful in the longer term. Consider creating a system guidance document that covers all aspects of systems modeling, specifies system designations, the failure mode identifiers and basic event coding.

Westinghouse Non-Proprietary Class 3 Page 43 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report SY-07 Guidance: The basis for the model assumptions are not directly Closed - Documentation of the sources of None referenced in the notebooks and are difficult to trace. There is some assumptions is not necessary for the inconsistency in the description of model assumptions and their impact to SAMA evaluation.

specific top events. For example, in the PORV system notebook Assumption 4 from Section 3.1, "Failure of PORV or safety valve to reseat following pressure relief will result in an isolable small LOCA."

This assumption has no direct reference to justify its use, in particular the part about safety valve failure to reseat being isolable. In the condensate and feedwater system analysis the following statement is made: "the bypass valve (FCV-2-35A) receives from a flow device." Although the references located at the end of the document probably provide this information, it would make review easier if the references are listed directly with the statements of fact. The assumptions in Section 1.2 of the Success Criteria Notebook are not traceable back to Appendix A of the IPE where they are referenced. Include more specific references in the system notebooks assumptions to facilitate traceability of information.

Westinghouse Non-Proprietary Class 3

• Page 44 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report SY-08 Systems Modeled: There is no documentation of a plant specific analysis Closed - Documentation of the basis for None for the EDG repair analysis or for AFW turbine-driven pump repair. The recovery actions is not necessary for the Onsite AC Recovery Notebook provides a table entitled "Time to SAMA analysis.

Recover a Failed Diesel Generator." The test states that this is based on a review of diesel generator failure and maintenance records collected from several plants, with an assessment of the severity of the observed failures and the experience of operations and maintenance experts. The WBN PRA staff was not able to provide the bases for the values within this table. It was stated that this may be from the Zion analysis. The AFW System Notebook uses Top Event TRP to represent the recovery of the AFW turbine-driven pump. The system notebook states that failure due to start failures is approximately 60% of all AFW turbine-driven pump failures, and the fraction of non-recoverable failures is approximately 40%. Reference 59 is shown as the source in the text but it is not listed in the reference list. The probability values used for EDG repair (time to recover a failed diesel) cannot be demonstrated as being applicable to WBN. There is a potential that the recovery probabilities are not applicable which could result in an increase in the contribution of the LOOP initiator. Perform a plant specific analysis for EDG repair or document an evaluation to show that an available analysis for another plant is directly applicable to WBN. Improve documentation for AFW turbine-driven pump recovery (reference 59 in text is not on reference list).

DA-01 Guidance/Documentation: There is no written documentation to identify Closed - A single data notebook was None which data notebook (Erin or PLG) is used for each analysis. Erin and created to identify all the data used. The PLG used similar but not identical methods. For common cause Erin source and plant specific data used to primarily used the INEEL data while PLG primarily used data from PLG- create each data variable was documented 0500. The results presented in Appendix B in the Erin notebook are not such that the results are reproducible.

cross-referenced to their basis. For example, distributional parameters were provided but the source of parameters was not provided or referenced. Both documents contain a brief discussion of what was done and some of the theoretical bases for the process. However, there is insufficient data or guidance to allow someone other than the author to reproduce the results. Each data notebook should have sufficient documentation and guidance to reproduce and update the data values.

Consideration should be given to establishing a single data notebook, following a single methodology, and using common data sources.

Westinghouse Non-Proprietary Class 3 Page 45 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report DA-02 Unique Unavailabilities or Data Modeling Issues: Plant operation with a Closed - Maintenance alignments are None SG PORV blocked was assumed to be not allowed because of the included in the model to discuss the Maintenance Rule unavailability criterion. Thus, this configuration is not block valves being closed. The closure modeled in the PRA. However, operation with a SG PORV blocked had of one and two valves are modeled.

occurred at WBN in order to permit repair of the SG PORV. Although the impact on baseline CDF/LERF is probably minimal, the condition has existed and should be included in the model. This configuration (operation with SG PORV blocked) should be incorporated into the PRA model. Alternatively, provide a more detailed assessment to demonstrate why the condition need not be included.

Westinghouse Non-Proprietary Class 3 Page 46 of 142 Our ref. LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report

- F&O, Description. ] Resolution tislMdelmjiW DA-04 Plant Specific Component Data: Table 2-2 of the ERIN data notebook Closed - A single data notebook was None lists the raw component failure data considered for the data update. This created to identify all the data used. The table shows a total of 7 failures of PRA related components, including 3 source and plant specific data used to pump failures, 1 valve failure treated as a pump (see F&O DA-03), I create each data variable was documented valve failure, and a controller wiring error that resulted in a CCF of 2 such that the results are reproducible.

compressors. The only values that were updated were PORV block valves The plant specific failures are referenced.

and no update was performed for instrumentation. As indicated in Table 2-2 only 5 of the 7 components would have been used in the update; however, a review of the events and demand data used for the update, as presented in table 2-3, shows a total of 9 component failures used in the update. These included 1 AFW turbine pump failure, IACA dryer failure, 2 MFW pump failures, 1RHR pump failure (see F&)DA-03), I ERCW pump failure and 3 CCW pump failures. The ERCW pump failure and the 3 CCW pump failures in Table 2-3 had no corresponding events in Table 2-2. No additional information is presented to permit tracing these failures. Additional review of the raw data sent to Erin indicated that there were additional failures that had not been listed in Table 2-2 which had been included in Table 2-3. The ACCESS data base list of the raw events by ERIN was reviewed and the 3 CCW pump events were there - as was an ERCW pump event. However, this list also had one additional ERCW pump failure that was not included in the failure count in Table 2-3. Further review revealed that the second ERCW pump failure was a duplicate of the first so that the data in Table 2-3 was appropriate. The reviewer concerns include the treatment of the maintenance frequency and duration data where there is a greater degree of manipulation of the raw data. The discrepancies between the raw data and the values used as input to the Bayesian update are more difficult to discern. The data error for the ERCW pumps needs to be resolved. The report should be expanded to include the raw data for maintenance frequency and duration and to show (at least example) any mathematical manipulations of the raw data needed to derive the data in Table 2-3. The data should be explicitly described to the extent that an independent reviewer can reproduce the values in Table 2-3 from the raw data. If necessary to reproduce the calculations, critical intermediate results should be included in the report.

Westinghouse Non-Proprietary Class 3 Page 47 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report F& Description, Re11iW~~iu odelIipact~

DE-01 Common Cause Treatment: There is an inconsistency in the logic for Closed - Model was revised in Rev 4 to Thermal barrier booster pumps RCP seal LOCA for top event SE in the RCP seal and thermal barrier address failure of flow to a single RCP. were added to the TB fault tree.

system notebook. An assumption for top event TB states that a single Also, for top event TB thermal barrier event consisting of common cause failure of all eight seal injection inlet booster pumps were added to the model.

check valves to re-open after loss of offsite power is used to represent all contribution from loss of seal injection. A failure to reopen any one of the eight seal injection inlet check valves will result in a loss of all seal injection for one RCP. A simultaneous loss of thermal barrier cooling for the same pump will result in a seal LOCA for that RCP. The single common cause element as modeled is not appropriate for this scenario. A review of the logic indicates that the single induced RCP seal LOCA would not be identified in this model. [See F&O AS-01 for discussion about overall induced seal LOCA model logic.] RCP Seal LOCA represents a significant fraction of the WBN CDF. The modeling of the check valve common cause failure is non-conservative and could have an impact on the CDF. The Induced RCP Seal LOCA model should be revised to correct the common cause failure logic.

DE-02 Spatial Dependencies: The treatment of spatial dependencies is Closed - A conservative flooding analysis None inconsistent and not thoroughly documented. The IPE can be used for the SAMA analysis.

internal flooding analysis used a simplified approach which relied heavily Documentation of the internal flooding on engineering judgment. The documentation of the IPE Internal analysis would not affect SAMA Flooding Analysis provides only a simplified summary of the analyses evaluation.

performed. The IPE summary did not address evaluation of flooding impacts from various pipe breaks in various locations on a room by room basis, but instead focused on impacts from a Building/Elevation standpoint. For elevations that had stairwells that propagated downwards, the internal flooding approach appears to have been to assume the impact to be negligible for that elevation, but consider potential floods originating on that elevation when evaluating lower elevations. For the lowest elevation, where a flood was modeled as occurring, it was not readily evident that the frequency used included the potential for pipe breaks at the higher elevations. In addition, the expected elevation of the flood was not specified in the IPE summary, and operator actions that may have been credited to isolate pipe breaks, and therefore limit the expected flood depths were not identified. For example, in the discussion of Auxiliary Building postulated flood, a statement is made that the passive sump is assumed to be completely

Westinghouse Non-Proprietary Class 3 Page 48 of 142 Our ref. LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report E&O ]<< DescriphnRstio _ _  ::7 1?S4hiatibWS114 Vvdeimpael overfilled and that enough water is assumed to accumulate such that the RHR and Containment Spray pumps are assumed to be failed. There is no discussion of what the maximum volume of water, and resulting maximum flood depth, was calculated to be. In addition, there was no justification as to why effects of other flood initiators on equipment at higher elevations in the Building were not modeled. The internal flooding analysis does not discuss potential spray related impacts, submergence effects on cables, equipment that would be impacted via cables/termination boxes, etc. A more detailed summary of the internal flooding analysis performed for the IPE should be maintained as part of the PRA, as the IPE summary documentation does not provide adequate information to explain the bases for the analysis. Based on the information available, the reviewers felt that there is a strong possibility that the initiating event frequencies and maximum flood depths (and therefore the impacted equipment) currently modeled may be overly conservative. Recommendations: Review the RI-ISI analysis and the HELB/MELB analysis to check that important underlying assumptions associated with the internal flooding analysis remain valid. This should include 1) review of credited operator actions (which are not identified in the current internal flooding analysis), 2) check to see if the internal flooding initiating event frequencies associated with the modeled scenarios should be updated, 3) verification that the equipment assumed failed in the internal flooding is consistent with impacts identified in the other analyses, and 4) a discussion on potential spray effects and submergence effects and why they are/are not included in the analysis.

DE-07 The ERCW system analysis does not postulate the common cause failure Closed - Plugging of ERCW traveling None of the strainers. The strainers have a motorized back wash function and screens and failure of the operator to are postulated to fail without the backwash function. However, the initiate manual backwash is addressed strainer failure is modeled as a "plug" failure mode. Because plugging is under total loss of ERCW initiating event a passive mode, as opposed to the active failure of the motorized ERCWTL. Strainer plugging is a slowly equipment, CCF was not included. CCF of both strainers will fail all evolving event with adequate time for ERCW. This mechanism may be significant to loss of ERCW initiating corrective and compensatory measures event frequency, and therefore important to CDF. Include CCF of (manual strainer rotation, etc.), relative to ERCW strainers in the ERCW system unavailability and loss of ERCW 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> mission time for plant model.

fault trees. I

Westinghouse Non-Proprietary Class 3 Page 49 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report J~>F~tY ~J. ~ 2Discn~iion '~ ~ 7&] < ~ ~3/4i~ Mokflm.pacl HR-01 A human action sensitivity study to identify sequences that, except for a Closed - A sensitivity study was I Top event ORT was set to low human error rate, would have been dominant contributors to core performed setting operator actions to guaranteed success to remove damage frequency has not been performed. The PRA provides no guaranteed failure and reviewing the top room ventilation top event (i.e.

indication of the impact on core damage frequency as a result of 50 resulting sequences. Model changes very long term) impact on a truncation of sequences with multiple human errors which when taken as were made based on this review. A very short term action when a whole result in unrealistically low human error probabilities. Perform a separate sensitivity evaluation was ESFAS is (eventually assumed) human action sensitivity analysis to identify potentially dependent actions performed to ensure that portions of the failed due to loss of room within sequences and to identify the amount of credit taken for very low event model were not shielded by being ventilation combined human action failure probabilities in the sequences. Consider quantified at 0.0. No additional scenarios 2 Top event OMU was set to establishing a lower limit for human action failure in any given sequence of interest were identified through this guaranteed failure with OSE (e.g., no less than IE-06 or other defensible cutoff) sensitivity evaluation. and RRH failed.

3 Top event RRH was set to guaranteed failure with OCD failed.

4 Top event RRH was set to guaranteed failure with OSE and OT failed.

5 Top event RRH was set to guaranteed failure with OSE and OT failed.

6 The emergency boration function and top event RRH were set to guaranteed failure with OS and ORT failed.

Westinghouse Non-Proprietary Class 3 Page 50 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment I Final Watts Bar Unit 2 SAMA Report HR-03 Calibration errors do not appear to be considered within top event RL, Closed - Common cause failures are None Initiation of RIHR Containment Sump Swapover. The containment addressed in common cause group swapover requires indication of low RWST level and high sump water RWSTMSC level. The system analysis considers the possibility of frozen RWST This has been incorporated into the instrumentation lines but does not include common cause failure of either model set of water level sensors. For the frozen lines, the system analyst assumes the operator would respond correctly and swapover the sump (I.e., no impact on HEP due to failed instrumentation). Typically, a pre-initiator error across multiple channels would be considered. Although common cause failure of multiple channels, based on collected data, is considered, pre-initiator errors are not well represented in the PRA data.

Thus, it may not be appropriate to ignore the potential for common sensor failure. Calibration errors are found to be addressed by top event TOT within the AFW system notebook. Recommendation: 1) A systematic search should be done for pre-initiator errors guided by a process and rules developed by the HRA task; 2) Human interactions identified by this search should be quantified by the HRA task not the system analyst.

[A possible way to address the noted pre-initiator would be to incorporate the impact of calibration error failing 3/4 channels into Top Event RL and other multi-channel instrumentation tops. Then requantify the HEP for manual swapover with failed instrumentation.]

HR-04 Repair is modeled for some components. However, no operator action is Closed - Repair and recovery that is None included for the start of the repaired component. It is unclear as to modeled in the PRA was reviewed and whether this action is included in the data for the fraction of start failures found to be appropriate. Additional that are recoverable. Operator action HTPR1, Start the Turbine-Driven treatment of important component Pump Given it Failed to Start due to Control or Signal Failures, (LOSP) is recovery was evaluated as potential used in top event TPR. TPR is used to represent the recovery of the SAMAs (e.g., SAMA 20, 158, 160).

turbine-driven pump from a control or signal failure or the repair of the While explicit treatment of pump restart pump. HTPR1 is under an AND Gate with the fraction of start failures would be more rigorous, the current that are unrecoverable. This implies that the repair includes the operator treatment is judged adequate for the action. The lack of consideration of specific scenario timing, confusion, SAMA model for restart of the turbine and resource loading could result in an optimistic treatment of the repair. driven AFW pump.

Clarify the treatment of the operator action for the repair. Ensure that it is consistent with the scenarios for which it is credited.

Westinghouse Non-Proprietary Class 3 Page 51 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report HR-O8 Post-Initiator Human Actions: The HRA and operator interviews were Closed - The HRA was updated to use None performed in 1992 when the plant was under construction. The input the EPRI HRA calculator in Revision 4 information for the HRA is therefore representative of construction era to the PRA procedures, training, and operators. An update should be performed to reflect current operator training and procedures, and this should be reviewed with plant Operations/Training staff.

Recommendations: Update the HRA to reflect current procedures and practices. Use a single approach if possible (i.e., either an updated FLIM or alternative methodology such as the cause-based decision tree (CBDT) method per EPRI TR- 100259.

HR-09 Pre-Initiator Human Actions: Ice condenser plants have an important Closed - The HRA was updated to use None pre-initiator human error for failure to restore the operating floor plugs the EPRI HRA calculator in Revision 4 after refueling. The failure probability for this human action is quantified to the PRA at 2.7E-7 (a very low number). This is calculated as the product of an initial human error for failure to restore of 3E-3, and two independent recovery actions of 0.025 and IE-3. The analysis was done with simplified THERP. The second recovery is credited as a completely independent action, with a failure probability lower than the original action. It is unusual for a recovery action to be more reliable than the original action. The calculation was performed in 1992; it should be updated to reflect the procedures currently in place. The failure to remove drain plugs is a single point failure for LOCA; thus this operator error could be risk significant. Assess the HEP against the current operating procedures using realistic recovery probabilities.

HR-10 Guidance: The WBN PRA uses 3 HRA methods: RO, RI - FLIM; R2, R3 Closed - The HRA was updated to use None

- THERP & EPRI-CBDT. Comparison of HEPs indicate a reduction of the EPRI HRA calculator in Revision 4 human error probabilities with each new revision. The use of three to the PRA different methods in the same PRA will produce inconsistent HEPs which could affect risk ranking and prioritization. The HRA should be performed with a consistent method.

Westinghouse Non-Proprietary Class 3 Page 52 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report

~E&O ~] .. ~ ~Descri6iion The original HRA (1992) uses 60 minutes for available timing for several

~ij~ -. ~ResolitfitiWSitzh~s . TM~ideI~Ii~5aW~

HR-11 Closed - The HRA was updated to use None HEPs related to loss of secondary side cooling. Available time for the EPRI HRA calculator in Revision 4 several other related HEPs was defined as 40 and 45 minutes. The one to the PRA. This is a symptomatic F&O hour time is based on a deterministic calculation of heat loads. The most about consistency of timing of actions in recent MAAP analysis (R3) provides a 33 minute time to start feed and the HRA and the relationship to the bleed. The HRA should use consistent times, presumably based on the success criteria. Two specific examples recent analysis. Timing of events to restore RCP seal cooling to prevent are mentioned in the F&O whose seal LOCA are based on 60 minutes. The Brookhaven model allows a resolution is documented here. Based on seal LOCA to occur at times less than 60 minutes. The assumed timing engineering judgment, these are the two in the seal LOCA model should be consistent with the assumptions in the most time sensitive operator actions in HRA. The HRA methods are time sensitive. Use of correct timing the PRA.

consistent with available analyses and models is essential. The HRA - The new time window for operator should used consistent times based on the most recent analysis. Develop actions for bleed and feed from the consistent timing for important sequences and re-quantify HEPs based on WBN2 PRA Success Criteria analyses of these times 25 minutes with at least one charging pump available and 10 minutes if only an SI pump is available was substituted into the HEP assessment for HAOBI and HAOB2. The resulting HEP did not change substantially in either case.

Based on the HRA methodology for operator actions to establish bleed and feed cooling in the HRA Calculator for Revision 4 of the PRA, shortening the time window available for operator action form 30 minutes does not change the resulting HEPs.

- The implementation of the WOG2000 model for RCP seal behavior should include a limitation of 13 minutes for restoration of RCP seal cooling. If seal cooling is restored at a time after 13 minutes, a large seal LOCA should be assumed.

Therefore, this F&O can be considered closed for the SAMA assessment if the operator action time window for Bleed and Feed and Restoration of RCP seal cooling are consistent with the values provided above.

Westinghouse Non-Proprietary Class 3 Page 53 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report HR-13 The HRA uses times of 3,4,12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> for restoration of ventilation to Closed - As evaluated by TVA An ORT success term was used shutdown board rooms. This information is taken from TI-ECS-95 calculation WBNOSG4-242, 6.9kV and to show that operators would (Sequoyah study). Reference TI-ECS-96 is a Watts Bar study but the 480V Board Room Transient have tripped the reactor prior to Temperature Analysis (RIMS T71 room heatup. This prevents a assumptions may not be applicable to the PRA. The assumptions in these very short term action from studies should be reviewed and made appropriate for the WBN PRA. 010416 804), 6.9kV board room being dependent on a very long The recovery of HVAC for the shutdown board rooms can have a temperature only reaches 103.5F by the term heatup impact.

significant impact on the CDF results. Assess room cooling consistently end of the 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> period analyzed, such for a 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> analysis with the appropriate analysis. Quantify room that loss of board room cooling would cooling recovery actions with the recovery times from the appropriate not result in component failure or analysis. unreliable operation during the PRA model mission time, such that the room ventilation top events can effectively be removed from the model.

Westinghouse Non-Proprietary Class 3 Page 54 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report

~F&O Descriptiont~iii.tts ModhIimiiad HR-15 Treatment of Dependencies: It appears each HEP is quantified Closed - A sensitivity study was 1 Top event ORT was set to independently. Although each action considers failures and successes performed setting operator actions to guaranteed success to remove directly relevant to the HI in question, there is no process in the HRA to guaranteed failure and reviewing the top room ventilation top event (i.e.

perform a systematic examination of all human actions in an individual 50 resulting sequences. Model changes very long term) impact from sequence. An example of potential human action dependencies is were made based on this review. very short term action when HAOF1, HAOF2. HAOFI, restore MFW, states that there is 45 minutes ESFAS is eventually assumed to restore MFW. This action could be asked after failure of repairing or to fail following loss of room recovering AFW (actions HAOS3, HAOS4, HTPR1)). These human cooling action failures may significantly reduce the time available to recover 2 Top event OMU was set to MFW or add to operator confusion. If the actions to restore MFW are guaranteed failure with OSE performed concurrently with that of the actions to recover AFW then the and RRH failed.

adequacy of resources would need to be determined. Establishing bleed 3 Top event RRH was set to

& feed operation following failure of actions to establish AFW or MFW guaranteed failure with OCD is another potential dependency, but this is explicitly addressed by failed.

features of the WOG Emergency Response Guidelines such that most 4 Top event RRII was set to PRAs for Westinghouse PWRs define a low-dependency or no- guaranteed failure with OSE dependency situation for these actions. It is still good practice to identify and OT failed.

the potential relationship and explain the rationale for establishing type of 5 Top event RRH was set to dependency. To be guaranteed failure with OSE consistent with accepted HRA methodology, there must be a systematic and OT failed.

process to identify, assess and adjust dependencies between multiple 6 The emergency boration human errors in the same sequence, including those in the initiating function and top event RRH events. When addressing needed enhancements to the HRA were set to guaranteed failure methodology, include the following steps to address dependent human with OS and ORT failed.

actions: I) Perform a systematic evaluation of all risk significant sequences, in which two or more post initiator actions are credited on the same sequences, including any recoveries; 2) Evaluate the potential for dependence among the actions in each sequence, document the basis for classifying the actions as dependent or independent; 3) Quantify the effects of dependencies using an accepted methodology; 4) Modify the split fractions for HRA values as appropriate to reflect dependencies.

Westinghouse Non-Proprietary Class 3 Page 55 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report

_F'._Descrzptwnl Rsoldtin Status Mde1fl v<j pmJct QU-01 The success of Top Event PI recovers the failures of both PR and SE in Closed - Top event PI recovery of both None Top Event CM. As Top Event PI only includes the PORV block valves, PR and SE in top event CM was this is not an appropriate recovery for stuck open Primary Safety Relief corrected. It is not appropriate that this Valves and RCP Seal LOCAs. This may significantly masks any recovers stuck open primary relief and contribution from primary safety relief valve failures and independent RCP seal failure. The relevant CM RCP seal failures. Break top event PR into the 4 desired functions: 1) success term has been rewritten to require PORV opens when required; 2) PORVs reclose when required; 3) PR and SE success and PI either SRVs open when required; 4) SRVs close when required. Since this successful or bypassed, such that PI does was identified through a limited review of the rules; a comprehensive NOT recover either stuck open primary review of the rules is recommended to ensure that similar errors do not relief or RCP seal failure exist.

QU-03 Sequence #9 questions the AFW decay heat removal split fraction AFD 1I Closed - The detailed evaluation of None when top event AFC is guaranteed failed. [This overestimates the worth scenarios required for F&O AS-02 has of this sequence]. This is not the appropriate split fraction for this been completed. This review also condition; the AFD split fraction where AFC is not questioned should be provided the response to F&O QU-03 used. As modeled, the worth of this sequence is overestimated. Develop (Verify logic for sequence 9 (recovers the not questioned split fractions. AFW decay heat removal at AFD 11 when AFC is guaranteed failure). This specific issue was resolved during Rev 4 model update. This is shown in that previous scenario 9 or similar items no longer appear in the model results.

QU-04 The DG top events are noted as having a dependency on the 125 VDC Closed - Diesel generators are questioned None Battery Board (top event DA & DB). The fault tree GAIV shows this in the SHARED event tree module before dependency; however, in the event tree rules success of the respective DC power is questioned in the ELECT1 top events DA or DB does not fail the DG (top event GA or GB). and ELECT2 modules, respectively.

Although the DG local batteries are sufficient for at least three starts, the Also, DG failure is much higher than 6.9kv buses still require breaker control power from the station batteries. battery failure, such that this would not Thus, this dependency should be modeled or the basis for screening be a material contributor relative to DG documented. The 125 VDC dependency should be modeled in the event failure.

tree. Fix the event tree and rules. I II

Westinghouse Non-Proprietary Class 3 Page 56 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report QU-05 An evaluation of sources of uncertainty and their effects on the analyses Closed - Documentation of the sources of None have not been performed for the current PRA model. Characterization of uncertainty are not required for SAMA.

the effects of uncertainties is an important attribute of the complete PRA, -Evaluation of how changes in analysis particularly for usage of the PRA for risk informed applications, assumptions may affect the cost-benefit Perform and document an evaluation of sources of uncertainty and their results will be performed as part of the effects on the analysis for the current PRA and for subsequent PRA SAMA analysis.

updates.

QU-06 TVA provided a truncation sensitivity evaluation for the review. Based Closed - Reference model is now None on the results of this evaluation , it appears that the current truncation quantified at IE-12 level of IE-10 does not achieve a sufficiently stable risk result, i.e., there seemed to be a significant CDF contribution remaining in the truncated residual. This can lead to errors when evaluating against absolute risk thresholds such as IE-3 in the maintenance rule and tE-6 in Reg. Guide 1.174. It is important that the truncation level used produce a set of results that capture a sufficient portion of the total. In absolute threshold evaluations use an acceptable truncation limit such that any unaccounted-for contribution is sufficiently small for the application. An example of an approach that may be used is to check that an order of magnitude drop in truncation causes a CDF/LERF change of less than 1%. It is also prudent to store at a minimum the end state and unaccounted file for each node used in the truncation sensitivity study.

Westinghouse Non-Proprietary Class 3 Page 57 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report 7

IF&~O_ I P IA*~icrltin . Resohiitr~diu. I ddilnpaci L2-0l On page 2-13 of the Level 2/LERF notebook it is stated that the Closed - The LERF / Level 2 PRA None survivability of the air return fans is based on their design basis to Analysis Notebook, Revision 1 (2000) function in LOCA containment conditions. There is no information identifies two functions for the air return provided regarding their ability to operate in conditions beyond the DBA fans (ARFs): enhancing heat removal by LOCA conditions which are of concern in the PRA. The air return fans, the ice condenser and the containment along with the igniters, are an integral part of the hydrogen control sprays and to limit hydrogen strategy for the ice condenser containments. This is especially crucial concentrations in potentially stagnant since the ultimate capacity of the Watts Bar containment is stated to be regions of the ice condenser containment.

105 psig. Many large dry containments that employ fan coolers as a post- With respect to hydrogen, the LERF /

LOCA heat removal feature rely on reducing fan speed to increase torque Level 2 Notebook documents to prevent overheating the motor in the dense mixtures that occur during assessments of hydrogen mixing with and high containment pressures. Some large dry containment plants have without the ARFs operating and predicted that the fan motors are not capable of sustained operation at concludes that there is no impact on pressure significantly above their design pressure (which is on the order containment integrity. The ARFs are not of 45 to 60 psig, i.e., substantially below the Watts Bar containment modeled in the 2000 LERF/ Level 2 ultimate capacity). When severe accident dense aerosols are also analyses as an assist to heat removal from considered, the issue may be more pronounced. Assess the capability of the ice condenser or the containment the air return fans to operate under the post core damage conditions spray and therefore it is concluded that (containment pressure and aerosol loadings) predicted in MAAP analysis the ARFs are not required for effective that are used to determine post core damage containment performance heat removal (note that the ARFs have no and fission product release characteristics. heat removal capability themselves).

Also, it is noted that the air return fans were not part of the resolution of GSI-189 (i.e., there was no recommendation to provide an alternate source of power for the ARFs to reduce challenges to containment integrity base don the NRC study in "Technical Assessment Summary for GSI-189: Susceptibility of Ice Condenser and Mark III Containments to Early Failure From Hydrogen Combustion During a Severe Accident," (2003). Because the air return fans are not required following a station blackout, it is concluded that they are not required for containment protection for

-any event. Thus, this F&O is closed with no changes to the WBN Level 2 PRA model for either the SAMA assessment or the WBN2 model update.

Westinghouse Non-Proprietary Class 3 Page 58 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report E&O' [ De6~i~iiiin ~  :: I Aesoluion Stdtu~s ModelImpact, L2-02 Level l/Level2 Interface (page 2-43): The accident sequences from Closed - This was addressed in the Rev 4 None Level I analysis that were determined to be core damage sequences update by changing the binning rules in (based on success criteria used in Level I assessment) were binned into the LERF model. Previously, for feed the various plant damage states which were then assessed in the Level 2 and bleed endstate, there was 77%

analysis. However, in some cases the subsequent MAAP analyses recovery of CDF (PDS LCI). Endstates performed in the Level 2 assessment determined that the accident BCI (MLOCA with no recirc) = 0.23 sequence did not result in core damage. This resulted in the definition of recovery, FCI (SLOCA with no recirc) a containment event tree top node (top event 3 - CV) which is used to 0.32 recovery, LNIYA and LNIYC = 0.0, consider both core damage arrested in-vessel and sequences that were whereas current LERF model has LER7 determined not to result in core damage. This implies that at least some for BCI and FCI (=0.008), LER5 for LCI of the success criteria in Level 1 analysis were not realistic. It is (=0.155), LER1 for LNIYA (=0.993, recommended that top event 3 - CV deal only with real core damage versus 0.0 for CV) and LER4 for LNIYC sequences that are arrested in-vessel. Those accident sequences that are (=0.166, versus 0.0 for CV).

determined not to result in core damage should be reconciled by updating If some of the core damage sequences do the appropriate portions of the Level I PRA. The calculated core damage not really result in core damage when frequency is conservative. Update the appropriate portions of the Level 1 analyzed with the Level 2 model, then the analysis so that all accident sequences assigned to a core damage state are core damage frequency is conservative.

true core damage sequences. This is not unexpected since some of the success criteria are bounding a number of accident sequences and the representative accident sequence for Level 2 may not be bounding for Level 1. The SAMA assessment depends only on the consequence analysis. As long as the frequency of the release category bins from the Level 2 assessment are accurate, then this F&O will not impact the SAMA assessment. A review of the treatment of the top event CV indicates that the release category bins are accurate in this respect.

Therefore, this F&O is considered closed for the SAMA assessment.

Westinghouse Non-Proprietary Class 3 Page 59 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report F"&O [ i~srpio I: --ResolujinSlaius~ Mu I~mpac~

L2-03 The Level +2 PRA does not include operator actions. 0 particular Closed - An assessment of the possible None importance to LERF is the proceduralized action to depressurize the RCS operator actions that can be considered in after the core damage has occurred per the EOPs (FR-C. 1) The only RCS the Level 2 PRA are described in WCAP-depressurization considerations in the Level 2 analysis are induced events 16657-P (Reference 8). All of the such as creep failure of the hot leg. Another human action potentially operator actions reduce the overall important to LERF is to manually open AFW discharge valves after loss probabilities of the release category bins of all instrument air. This action is included in the SBO sequences, but in the Level 2 assessment and therefore not in the LERF analysis. Both of these actions haiie significant impact reduce the overall offsite consequences.

on reducing HPME which is a large contributor to LERF. No SAMG The impact of including these operator activities have been included in the Level 2 analysis. The SAMG actions in the WBN2 model used to included both written guidance and training and their impact on the Level assess SAMA would be to reduce the 2 scenarios, particularly those that are LERF contributors, should be maximum possible benefit attainable for considered to make the analysis more realistic. Appropriate integration any alternative. Thus, not including of the WBNP EOPs and the SAMG would change the LERF considerably operator actions in the Level 2 model and would likely result in a LERF value less than 10-7. This is important maximizes the possible benefit which in PRA applications as none of the LERF measures are above the 10-7 could potentially result in additional cutoff criteria. Update the appropriate portions of the Level 1 analysis to features being classified as cost-effective.

include pre-core damage operator actions per EOPs that can impact Level Therefore, this F&O is considered closed 2 results. Update Level 2 analyses to reflect the impact of post core for the SAMA assessment.

damage EOP and SAMG activities.

Westinghouse Non-Proprietary Class 3 Page 60 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report I

F&O ~~lL. ~ DescR6,Vion.7h i

, RsifinSai& UMbeizptQ2 L2-04 Phenomena CETs/HEPs/System Considered/Success Criteria: The Closed - The WBN Level 2 model is not None hydrogen generation used in the analyses is based almost entirely on the sensitive to differences in hydrogen MAAP 3b version 17 results for the various accident sequences generation for various core damage considered. The LERF quantification is sensitive to the hydrogen accident scenarios that can be predicted detonations. MAAP 4.03 predicts considerably more hydrogen between the various computer code production for some accident sequences than does version 3B based on versions (e.g., MAAP3.0 vs. MAAP4.0).

the later version's more detailed modeling of core degradation. The most The availability of the hydrogen igniters pronounced differences are those in which the water is added to the core is the determining factor for all core after the onset of core degradation and relocation. This would occur due damage sequences. Increases or to accumulator injection due to RCS depressurization. In addition, the decreases in hydrogen generation are differences in MAAP code models related to RCS cooldown and within the capability of the igniters and depressurization can influence MAAP prediction of core recovery/core the overall uncertainties in hydrogen damage for SGTR and events with SG cooldown of the RCS. The generation for a given core damage potential for increased hydrogen generation for some accident sequences scenario. Differences in modeling SGTR may impact the Level 2/LERF results. The SGTR differences are not and RCS cooldown and depressurization /

expected to significantly change the overall results. Verify that MAAP sequences between the various versions results used in the current Level 2/LERF analysis reflect appropriate of the MAAP code are not expected to prediction of hydrogen generation and response for SGTR and sequences have an impact on the SAMA involving SG cooldown and depressurization of RCS. For future assessments. Comparisons have shown analyses and Level 2 updates, transition to MAAP 4.0 that the only impact is on the time available for operator actions and that these times are only changed by less that a few minutes out of 20 to 30 minutes.

These small changes should have insignificant changes on the subsequent HEPs for the actions. Therefore, this F&O can be considered closed for the SAMA assessment.

Westinghouse Non-Proprietary Class 3 Page 61 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report L2-05 Phenomena CETs/HEPs/System Considered/Success Criteria: The Closed - All of new assessments of the None current Level 2 assessment relies heavily on NUREG-1 150 analyses. likelihood or magnitude of the Significant additional information is available which could have an phenomena reduces the overall impact on the result of the current NUREG-1 150 based analysis. probabilities of the release category bins References include: 1) In-vessel steam explosions (NUREG- 1524, in the Level 2 assessment and therefore August 1996); 2) Thermally induced SG tube failure (EPRI); 3) Direct reduce the overall offsite consequences.

Containment Heating (NUREG/CR-6427). Review recent information The impact of including these new and consider updating Level 2 assessment to be consistent with current phenomena considerations on the WBN2 published research findings, model used to assess SAMA would be to reduce the maximum possible benefit attainable for any alternative. Thus, not including the new phenomena considerations in the Level 2 model maximizes the possible benefit which could potentially result in additional features being classified as cost-effective.

Therefore, this F&O is considered closed for the SAMA assessment.

L2-09 LERF Definition: The WOG LERF definition was chosen because it Closed - An assessment of the EALs for None contained a more detailed description that is easily applied to Level 1 WBN concludes that a General PRA. The WBN LERF definition encompasses other commonly-used Emergency would be declared more than LERF definitions with one exception - the classification of SGTR with 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> before core damage occurs for a high pressure SI success and SG isolation failure. The sequence is binned SGTR with SI available but an inability as non-LERF. The WBN LERF does not address Emergency Action to stop the loss of RCS inventory through Levels (EAL). The definition of early in LERF is related to releases the tube rupture. The SGTR event itself within 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> after notification of evacuation. Without assigning EAL's would cause an Alert declaration based to the SGTR sequence progression, it is not possible to justify the binning on the loss of one fission product barrier.

of the sequence as non-LERF. Although most other plants bin this The inability to equalize RCS and SG similar sequence as non-LERF, the EAL's must be included in the Level 2 secondary side pressure to stop the loss to justify it for WBN. The Level 2 as it stands does not support the of RCS inventory would result in SG binning of the SGTR with HPI as a non-LERF event. For SGTR with overfill and the initiation of some fission HPI, the initiation of offsite emergency actions may not occur until quite product releases (based on normal RCS late in the accident sequence; close to the time of core damage. Justify fission product inventories) which would the exclusion of the SGTR with HPI available based on timing of core escalate the declaration to at least a Site damage in relation to the EALs. Emergency. The potential loss of a second fission product barrier (the fuel rod cladding) would be diagnosed as the

Westinghouse Non-Proprietary Class 3 Page 62 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report RWST is drained and the inability to switch to ECC recirculation due to the lack of containment sump water. This would trigger a General Emergency and the potential for initiation of offsite protective actions. The remaining water in the RWST would permit several hours of continued injection prior to failure of ECCS, followed by another period of several hours before core uncovery occurred. Thus, at least 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> are available between the declaration of a General Emergency and the onset of core uncovery overheating. This neglects the obvious operator action to refill the RWST (per the EOPs on loss of ECCS recirculation) which could prevent core damage altogether.

Therefore the SGTR with HPI available is properly binned as a non-LERF sequence and this F&O can be considered closed for the SAMA assessment.

MU-01 The WBNP Configuration Control and Maintenance Desk Top Procedure Closed - Maintenance and Update None does not provide the necessary details to guide the maintenance of an as- procedures are not needed for the SAMA built, as-operated PRA. The procedure does not provide adequate detail analysis. This item is considered closed for 1) the monitoring and collecting of specific types of PRA inputs; and because TVA procedures SPP-9. I1 and

2) the process for identifying and processing open PRA issues. Changes NEDP-26 have been issued to address the in plant performance, operation, or design that affect the PRA models, issues in the finding.

assumptions, or inputs, could result in significant changes in PRA results, but the procedure does not define specific requirements for monitoring, I evaluation, and incorporating changes into the PRA.

Westinghouse Non-Proprietary Class 3 Page 63 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report MU-06 Application Re-evaluation: There is no specific guidance on how to Closed - Maintenance and Update None perform the impact assessment or ensure that impacted PSA Applications procedures are not needed for the SAMA are updated in a timely manner. Revise procedural requirements to analysis. This item is considered closed include a listing of potentially impacted PSA applications that require a because TVA procedures SPP-9.1I1 and qualitative review to ensure that the conclusions remain valid. The NEDP-26 have been issued to address the procedure should also require an update of past PSA Applications that are issues in the finding. In particular affected by the latest PSA update, and should provide guidance on how to NEDP-26 contains requirements to document that an impact evaluation has been performed. review model applications after a PRA update.

MU-07 There is no means to determine the effectiveness of the current process to Closed - Maintenance and Update None identify potential changes to the PRA as a result of plant modifications. procedures are not needed for the SAMA Document the process to evaluate plant modifications with respect to the analysis. this item is considered closed impact on the PSA. because a revision to TVA procedures SPP-9.3 has been issued to require that DCN's be reviewed for PRA impact.

Westinghouse Non-Proprietary Class 3 Page 64 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report F&O [~>~srjtol ] Resoluition Sialus Mdel U I,pqaWc.

TH-01 Success criteria analysis are primarily based on FSAR and NUREG-4550 Closed - The use of more realistic None analyses. This likely results in a somewhat conservative analysis that success criteria in place of conservative might be made less so through additional plant-specific analyses for success criteria from NUREG/CR 4550 selected scenarios. It may be prudent to review the NUREG 4550 bases and the WBN FSAR would reduce the to verify their continued applicability in light of plant changes over time overall core damage frequency which, in and changes in generally accepted state of the art methodology. turn, would reduce the overall Conservatisms should be included with care in a PRA intended for use in probabilities of the release category bins risk-informed plant applications, since it is easy to compound in the Level 2 assessment. Therefore, the conservatisms within accident sequences resulting in unrealistic overall offsite consequences would be representation of plant risk and risk contributors. Check the PRA for reduced. The impact of including more significant sources of conservatisms that may skew the PRA results. realistic success criteria considerations on the WBN2 model used to assess SAMA would be to reduce the maximum possible benefit attainable for any alternative. Thus, using conservative success criteria in the PRA model would maximize the possible benefit which could potentially result in additional features being classified as cost-effective.

The potential for conservative success criteria to mask the importance of some SAMA alternatives was also assessed by comparing the success criteria to other Westinghouse 4-loop plants. As described in the response to AS-03, the success criteria used in the WBN PRA are reasonable compared to other 4-loop Westinghouse plants. Therefore, this F&O is considered closed for the SAMA assessment.

Westinghouse Non-Proprietary Class 3 Page 65 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report F&o0 Dscription K l aResolutiaiais .odd Impact TH-03 The source of success criteria for ATWS as referenced in the Closed - Incorporated in WBN Rev 4 None documentation in Event Tree Notebook Appendix C is WCAP- 11993. update. See Pressurizer PORV and Safety The success criteria for ATWS pressure relief (per PRA Rev 2 Valves Systems Analysis Sections 1.2.3 documentation) is from NUREG/CR-4550 which uses the moderator and 3.2.3.2 and RPS Systems Analysis temperature coefficients (MTC) for critical regimes of-7pcm/F and - Section 3.1.2.

20pcm/F. These MTCs, and their implied impact on pressure relief capacity, are not consistent with the approach defined in WCAP- 11993.

If it is intended that the PSA should be based on WCAP- 11993, the model should instead use the UET approach, and split fractions reflective of the current core loading, per the WCAP. Review the ATWS modeling and reconcile the documentation with what is actually modeled. Consider implementing the WOG model per WCAP- 11993. Follow progress of the current WOG ATWS program and consider implementing the revised approach when available.

TH-06 The success criteria allow 1 of 4 HHSI/CVCS pumps for response to a Closed - Success Criteria analyses for the None small LOCA. The SI pumps have a shutoff head of -1500 psi. There is WBN2 PRA show that one HHSI pump no supporting analysis to show that 1 SI pump can provide adequate can provide adequate flow to maintain make-up flow for breaks at the low end of the SLOCA size range (e.g., RCS inventory for a 3/8 inch LOCA IF 3/8 inch or 1/2 inch break) without additional action for primary pressure AFW is operable. In this case, the RCS reduction. Provide the basis for assuming that 1 SI pump will provide pressure falls to just above the secondary adequate injection capability for the full range of small LOCAs side pressure (at the atmospheric dump setpoint) before significant RCS inventory is lost. At this lower RCS pressure, the HHSI pump can easily keep up with break flow. Therefore, this F&O is considered closed.

Westinghouse Non-Proprietary Class 3 Page 66 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report

[

F&O ~ ~ Dcscripijon 4 > Rsoluiain Sialiss ___________Impact _

TH-10 The success criteria for the ECCS pumps and PORVs for bleed and feed Closed - Success Criteria analyses for the Success criteria changes were appear to either be incorrectly stated or incorrectly incorporated into the WBN2 PRA for Bleed and Feed were made to reflect the requirement model. This might significantly affect the contribution of failure of bleed compared against the WBN PRA Rev. 4 for two PORVs when and feed to the CDF results. Top event BF is modeled in the top model (see Section 3.1 of this report) and performing bleed and feed event rules so that success of one charging pump OR one SI pump will HR-0 11resolution status. The WBN cooling using SI pumps.

allow success of top BF. However, MAAP runs for Bleed and Feed PRA model used in the SAMA appear to justify success with one train of ECCS injection pumps (one assessment was modified if necessary to charging and one SI pump) plus one PORV. Peer reviewer inspection of be consistent with the WBN2 Success the MAAP cases provided for feed & bleed cooling in the notebook Criteria analyses for the number of indicated that the MAAP-predicted ECCS flowrates at several sampled pumps and PORV required for success.

times shown in the results plots are higher than would be expected based However, the HRA was not revised based on the pump head curve data and the predicted RCS pressure at these on the assessment described in the time steps if only one pump were credited. (These were "back of the resolution of F&O HR-l11.

envelope" calculations, performed as ballpark estimates; however, they appeared to indicate that the ECCS flowrates shown in the MAAP results were higher than would be provided by a single pump). Thus, the model rules are allowing success on bleed and feed with one high head pump and one PORV, when the MAAP runs only justify two pumps and one PORV. It was also noted that some of the previous revision PRA documentation indicated that feed & bleed success requires 2 PORVs.

The success criteria for feed and bleed was stated in PLG-1339 as 2 PORV and 1/4 HHSI/CVCS pumps. If this is true, then feed and bleed should be guaranteed failure for transient initiators Loss of 6.9 kV board and Loss of 125 Vdc battery board because one PORV would be unavailable. A check of the event trees, however, shows that feed and bleed is allowed for these events. The success criteria for feed and bleed as implemented in the model appears to allow a 1/2 PORV success criterion for certain events. Either revise top event BF rules to require at least one charging and one SI pump (to match available analyses) or perform MAAP runs to show that the modeled criterion is acceptable. It is unclear whether credit could be taken for two SI pumps, but there is the potential for a basis to be developed. Review the basis for the PORV bleed and feed success criterion and the implementation of the proper success criterion in the model. Ensure that, if 2 PORVs are required, the model properly accounts for event-specific dependencies.

I

Westinghouse Non-Proprietary Class 3 Page 67 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report TVAOO Description~.~ ~ ~~ ResoluioiF .*aiu~s VModel Impact TVA-_001 WBN-ENG-01-005 Self Assessment Effectiveness Review enhancement Closed - Not needed for SAMA analysis. None El. A pending standard for PSA is in the review cycle by ASME and the The SAMA analysis is not a risk NRC. Enhancements to the WBN model and documentation will be informed application and is not required required to meet the provisions of the standard if it remains as currently to meet the PRA standard.

proposed.

TVA-002 WBN-ENG-01-005 Self Assessment Effectiveness Review enhancement Closed - Not needed for SAMA analysis. None E2. The thermal hydraulic analyses and core melt success criteria were Updating to a newer version of MAAP developed using the Modular Accident Analysis Program (MAAP) may provide more realistic success version 3B and have not been updated since the original analysis. The criteria. However basing the SAMA MAAP version 4 computer code is now available within Engineering and analysis on a conservative model will incorporates later research on core melt phenomena. A MAAP4 model result in conservative calculation of the for WBN has been developed as a task in the current PSA revision. SAMA benefit, and therefore SAMAs may be included which actually could be excluded.

TVA-1 1 Add cooling to the thermal barriers via the thermal barrier booster pumps Closed - See resolution to DE-01 Thermal barrier booster pumps to the CCS system notebook. The thermal barrier booster pumps were were added to the TB fault tree.

deleted in Revision 4 to the PSA. The reference that was the basis for this deletion is not valid and the thermal barrier booster pumps need to be added back into the model.

Westinghouse Non-Proprietary Class 3 Page 68 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report Table 3 Release Category Frequencies and Related Accident Sequences 2.638 x 10 ' Ite major accident contributors to this release event are initiated by loss ot ottsite power and the essential raw cooling water system; failure of the emergency diesels to start and/or failures in the 125-volt direct current distribution system, together with loss of secondary cooling; and no recovery before core melt.

1.192 x 10. The main contributor to this release event is initiated by a steam generator tube rupture in conjunction with either an operator error or a random failure of electrical distribution systems, leading to failure of the coolant system and failure to control the affected steam generator before core melt occurs.

I11 1.995 x 10-6 The major accident contributors to this release event are initiated by loss of offsite power and various failures in the alternating current distribution systems; no recovery of power before core melts; a reactor coolant system loss-of-coolant accident (large- and medium-sized loss-of-coolant accident); and failure to establish long-term core cooling.

IV 1.299 x 10-' Contributors to this category are core damage sequences which lead to benign releases (intact containment)

Table 4 Projected 2040 Population Distribution within 80 Kilometers (50 miles) titn LJ 1-2 2ý-3 4 45 -. '5-10 0-3 J02 00~4-0 05 N 0 18 0 0 135 2,465 1,885 2,778 4,768 6,172 18,222 NNE 0 0 18 411 185 1,536 11,762 18,766 14,502 2,547 49,727 NE 0 0 18 308 287 827 3,783 16,734 29,838 78,334 130,130 ENE 0 0 18 308 287 497 3,553 29,539 63,798 25,3831 351,832 E 0 8 431 308 616 552 11,352 18,647 30,063 44,013 105,990 ESE 0 0 0 27 41 68 6,230 20,120 5,068 3,280 34,833 SE 8 0 0 29 39 135 19,852 15,185 3,950 4,822 44,020 SSE 21 0 0 246 413 103 8,951 12,907 2,918 48,593 74,151 S 16 0 0 0 1,983 3,824 4,586 42,883 56,430 17985 127,707 SSW 0 0 21 0 0 .546 5,725 42,517 46,281 106,392 201,482 SW 0 0 0 0 0 1,051 12,978 14,499 62,307 111,795 202,630 WSW 0 6 36 59 126 711 12,791 2,837 2,840 3,372 22,778 W 0 14 22 101 90 710 3,406 5,555 2,944 5,474 18,316 WNW 0 0 22 126 79 490 2,091 4,372 5,654 20,511 33,345 NW 0 1.08 332 376 526 2,655 2,889 18,634 10,462 15,956 51,940 NNW 0 0 0 173 123 3,116 1,536 33,843 11,609 5,890 56,290 Total 1 45 154 918 2,472 4,930 19,286 11,3370 299,816 353,432 728,967 1,523,390 I1*u -.T iu" .. .. r .... inj.'1in [u 1----in .. .. 1iu -1..i ;Li ... u. Dy-

. I .OU'l.

Sote: I[SAIC o convert 22007)rom mile to kilometer mutply te vaue by Lo00.

Source

Westinghouse Non-Proprietary Class 3 Page 69 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report Table 5 Watts Bar Core Inventory

Žuclide Isolope. GroupQ ?2 & ý&

Cobalt Co-58 6 1.II E+06 Co-60 6 8.67E+05 Krypton Kr-83m 1 1.1 5E+07 Kr-85m 1 2.39E+07 Kr-85 I 1.03E+06 Kr-87 1 4.81E+07 Kr-88 I 6.66E+07 Xenon Xe-131m 1 1.05E1+06 Xe-133m 1 6.16E1+06 Xe-133 1 1.91E+08 Xe-135m I 4.05E+07 Xe-135 1 6.43E+07 Xe-138 I 1.67E+08 Iodine 1-130 2 1.93E+06 1-131 2 9.46E+07 1-132 2 1.39E+08 1-133 2 1.95E+08 1-134 2 2.16E+08 1-135 2 1.86E+08 Bromine Br-83 2 1.15E+07 Br-84 2 2.14E+07 Cesium Cs-134 3 1.66E+07 Cs-135 3 O.OOE+00 Cs-136 3 5.89E+06 Cs-137 3 1.17E+07 Cs-138 3 1.81E+08 Rubidium Rb-86 3 1.87E+05 Rb-88 3 6.83E+07 Rb-89 3 8.92E+07 Strontium Sr-89 4 9.34E+07 Sr-90 5 8.94E+06 Sr-91 5 1.16E+08 Sr-92 5 1.24E+08 Yttrium Y-90 7 9.48E+06 Y-91m 7 6.76E+07 Y-91 7 1.21E+08 Y-92 7 1.25E+08 Y-93 7 9.48E+07 Y-94 7 1.51E+08 Y-95 7 1.57E+08 Zirconium Zr-95 7 1.67E+08 Zr-97 7 1.61E+08 Niobium Nb-95 7 1.69E+08 Nb-97m 7 1.53E+08 Nb-97 7 1.62E+08 Molybdenum Mo-99 6 1.78E+08 Technetium Tc-99m 6 1.57E+08

Westinghouse Non-Proprietary Class 3 Page 70 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report Tc-99 6 O.OOE+00 Tc-101 6 1.61E+08 Ruthenium Ru-103 6 1.48E+08 Ru-105 6 1.00E+08 Ru-106 6 5.OOE+07 Rhodium Rh-103m 6 1.48E+08 Rh-105 6 9.55E+07 Rh-106 6 5.33E+07 Rh-107 6 5.77E+07 Antimony Sb-127 4 8.05E+06 Sb-129 4 3.03E+07 Sb-130 4 1.00E+07 Tellurium Te-125m 4 1.93E+04 Te-127m 4 1.33E+06 Te-127 4 7.93E+06 Te-129m 4 5.81E+06 Te-129 4 2.88E+07 Te-131m 4 1.86E+07 Te-131 4 7.99E+07 Te-132 4 1.36E+08 Te-133 .4 1.06E+08 Te-134 4 1.73E+08 Barium Ba-137m 5 1.11E+07 Ba-139 5 1.73E+08 Ba-140 5 1.73E+08 Ba-141 5 1.56E+08 Ba-142 5 1.49E+08 Lanthanum La-140 7 1.79E+08 La-141 7 1.58E+08 La- 142 7 1.54E+08 La- 143 7 1.46E+08 Cerium Ce- 141 8 1.59E+08 Ce-143 8 1.48E+08 Ce-144 8 1.29E+08 Praseodymium Pr-143 7 1.44E+08 Pr-144 7 1.30E+08 Pr-145 7 1.01E+08 Neodymium Nd-147 7 6.39E+07 Neptunium Np-239 8 1.87E+09 Plutonium Pu-238 8 3.15E+05 Pu-239 8 3.48E+04 Pu-240 8 4.38E+04 Pu-241 8 1.49E+07 Pu-243 8 2.86E+07 Americium Am-241 7 9.80E+03 Am-242 7 7.93E+06

Westinghouse Non-Proprietary Class 3 Page 71 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report Curium Cm-242 7 3.98E+06 Cm-244 7 1.61E+05 a The grouping is based on NUREG-1465.

Source [SAIC 2007]

Table 6 Release Times, Heights, and Energies for Release Categories

,Release Heiehl Warniq4g T~i"" Tie twReleasec

-'elaDurafion ~Releaise;ner&

Relesefgi' Cafny 11ietr~s/ (hoho (hours)

Or~ (i~ts I 10.00 8 10 2 28 11 10.00 20 24 4 1 III 10.00 20 30 10 3.5 a These values were taken from similar accident scenarios given in NUREG/CR-4551.

Source [SAIC 2007]

Table 7 Fission Product Source Terms

'Cteor NG s Ite Sr R11~ L Ce Ba Mo~

1 0.90 0.042 0.043 0.044 0.0027 0.0065 0.00048 0.004 0.0046 0.0065 11 0.91 0.21 0.19 0.0004 0.0023 0.07 0.00028 0.00055 0.025 0.07 1II 0.94 0.0071 0.011 0.0052 0.00036 0.00051 4.2 x 10-6 4.0 x 10-6 0.0013 0.00051 NG = Noble gases.

Source [SAIC 2007]

Westinghouse Non-Proprietary Class 3 Page 72 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report Table 8 Evacuation Times 0-to-16-kilometer (0-to-10-mile) Area 1 6-40 3-40 5- 12 2 4-23 2-41 3-47 3 4-21 2-43 5-0 4 4-10 2-36 3-41 5 4-37 2-53 4-05 6 4-25 2-45 3-54 7 4-21 2-43 3-51 8 4-25 2-45 3-54 9 3-26 2- 15 3-30 10 3 -26 2- 15 3-30 11 3 -26 2-30 3-50 12 3 -26 2-30 3-54 13 3-26 2-0 3-30 14 3-26 1 -35 3-30 15 3-20 1 -30 3-25 Total 61 -20 37-21 58-33 Average hours 5 2-29 3-54 Average speed over 10 miles 2.45 4.02 2.56 (miles per hour)

(meters per second) 1.1 1.8 1.15 Source [SAIC 2007]

Table 9 Severe Reactor Accident Annual Risks I- Early Containment Failure 2.19 x 10' 4.45 x 10' II - Containment Bypass 3.42 x 106 8.11 x l09 III - Late Containment Failure 1.16 x 106 1.78 x 109 Table 10 Annual 80-Kilometer (50-mile) Population Dose and Economic Cost Risk Poputalto'n Doke Risk Econoimic ( ost Riik Releuse Cgiegory . (penrya)(oar/a)

I - Early Containment Failure 5.78 x 10"' 1.17 x 10' 11- Containment Bypass 4.08 x 10-' 9.67 x 102 III - Late Containment Failure 2.31 x 100 3.55 x 103

Westinghouse Non-Proprietary Class 3 Page 73 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report Table 11 System Importance (RRW > 1.02) for CDF (Model Name: WBN4SAM2)

Rank' S.Stelnr SyVslen Dcescriptionw. iWot 1 ERCW ESSENTIAL RAW WATER COOLING SYSTEM 1.6398E+000 2 RHR WBN RESIDUAL HEAT REMOVAL SYSTEM 1.6386E+000 3 CVCS WBN CHEMICAL AND VOLUME CONTROL SYSTEM 1.1321E+000 4 EPS-AC AC ELECTRIC POWER SYSTEMS 1.0827E+000 5 CCS COMPONENT COOLING SYSTEM 1.0672E+000 6 EPS-DC DC ELECTRIC POWER SYSTEMS 1.0470E+000 7 RCS RCS SYSTEMS AND MISC. FUNCTIONS 1.0445E+000 8 AFW AUXILIARY FEEDWATER SYSTEM 1.0414E+000 9 VENT VENTILATION SYSTEMS 1.0246E+000 Table 12 System Importance (RRW > 1.02) for LERF (Model Name: WBN4SAM2)

Runk, ),Sw ~%Rak Sitem~~ Srstenj yste ripliont" ~kediuction~

1 ERCW ESSENTIAL RAW WATER COOLING SYSTEM 1.7756E+000 2 AFW AUXILIARY FEEDWATER SYSTEM 1.6258E+000 3 EPS-AC AC ELECTRIC POWER SYSTEMS 1.1740E+000 4 RHR WBN RESIDUAL HEAT REMOVAL SYSTEM 1.1636E+000 5 AIR WBN - PLANT COMPRESSED AIR SYSTEMS 1.0876E+000 6 RCS RCS SYSTEMS AND MISC. FUNCTIONS 1.0741E+000 7 CiS CONTAINMENT SYSTEMS 1.0616E+000 8 VENT VENTILATION SYSTEMS 1.0567E+000 9 CVCS WBN CHEMICAL AND VOLUME CONTROL SYSTEM 1.0448E+000 10 VSEQ V SEQUENCE EVENTS 1.0421E+000 11 EPS-DC DC ELECTRIC POWER SYSTEMS 1.0334E+000 12 CCS COMPONENT COOLING SYSTEM 1.0242E+000 13 SEC SECONDARY SYSTEMS AND FUCNTIONS 1.0221E+000 Table 13 Basic Event Importance (RRW > 1.02) for CDF (Model Name: WBN4SAM2)

RankA Basýic E vent Basic Elvent De .. BE Riski DHARR1 Operators fail to perform alignment for high 1.3228E+000 head recirculation 2 ERCWGLOBAL Global Failure of ERCW Pumps 1.2337E+000 3 COVFOI 0620504 Check valve 62-504 fails to open on demand 1.0734E+000 4 [PMOFR0 07000CS Common cause failure to run of CCS pumps CS, 1.0246E+000 PMOFR1 07001AA IAA, 1BB, 2BB PMOFRI 07001BB PMOFR2 07002BB]

Westinghouse Non-Proprietary Class 3 Page 74 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report Table 14 Basic Event Importance (RRW > 1.02) for LERF (Model Name: WBN4SAM2)

RAnk fiWBaIst.ic Basi Evet E'vent DescriptionReuto BERFisk I ERCWGLOBAL Global Failure of ERCW Pumps 1.2926E+000 2 FRACT 1ASNONREC Fraction I-AS failures not recoverable 1.1431E+000 3 DHARR1 Operators fail to perform alignment for high head 1.10 11E+000 recirculation 4 PDMOD23 Relief valve 0-32-512,513, 514,4906,540, or 541 1.0717E+000 opens prematurely 5 DHADS2 SGTR with isolation. Steam dumps avail. For 1.0626E+000 cooldown 6 FDHADS2 Control flag 1.0626E+000 7 CNTLKIPREEXISTL Isolation failure due to large pre-existing leaks 1.0532E+000 8 DHAMUI Operator failure to open valves59-737, 738, 511 1.0406E+000

& 742 and start 9 [PTSFSI 00301AS] Turbine pump lA-S fails to start on demand 1.0307E+000 10 COVFO1 0620504 Check valve 62-504 fails to open on demand 1.0261E+000 11 [PMSFS1MDPOO301AA Common cause failure to start of AFW pumps 1.0238E+000 PMSFS IMDPOO301BB IAA, 1BB and 1AS PTSFSI 00301AS]

12 [IDGSIAAFS] DG IA-A fails to start or run 1.0229E+000 13 [PMOFRO 06700GB] ERCW pump G-B fails during operation 1.0222E+000 14 [PMOFRO 06700EB] ERCW pump E-B fails during operation 1.0222E+000 15 [PMOFR0 06700HB] ERCW pump H-B fails during operation 1.0201E+000 16 [PMOFRO 06700FB] ERCW pump F-B fails during operation 1.0201 E+000 17 DHAOBI Operator fails to initiate bleed and feed 1.0197E+000 Table 15 SAMA Candidates Identified Through RRW Review S4A/4'iflte S,,1AMADivcu ssionSAAo Refurbish the ERCW pumps & upgrade the Improves the reliability of the ERCW pumps. 271 capacity of the current pumps.

Provide a portable diesel powered 5000 gpm Improves availability of ERCW for SBO. 272 pump as a backup to the ERCW system.

Provide a 2 MW blackout diesel generator to Improves availability of AC power during 9 power Charging Pumps, Igniters, Inverters, etc SBO.

Use a portable pump hookup to firewater system Improves availability of SG cooling. 75 to provide backup feedwater to Steam Generators Enhance procedures to prevent strainers from Improves reliability of core cooling. 198 plugging during recirculation Route ERCW to B charging pump lube oil cooler Improves reliability of charging pump. The A 262 pump design includes this capability.

Provide a redundant path for ECCS suction from Eliminates single failure potential of RWST 273 the RWST around check valve 62-504. check valve failure to open.

Cross-tie diesel generators Increased availability of on-site AC power. 12, 244 Cross-tie CCS trains with Appendix R valve. Improves availability of component cooling 45, 157,257 water.

Replace CCS pumps with positive displacement Improves reliability of CCS system. 274 pumps Provide a spare battery charger Improved availability of DC power system. 3 Provide a new inverter arrangement. Improved reliability of AC power system. 275

Westinghouse Non-Proprietary Class 3 Page 75 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report 7 ~ N4AFl S4!VM IDiscussiton( $M4'~

Provide an auto start signal for AFW on loss of Improved reliability of AFW for low power 276 Standby Feedwater Pump. events (<18%) before Main Feedwater Pumps are started.

Replace shutdown board chillers Improved reliability of shutdown board 277 HVAC.

Perform analysis to evaluate the need for Eliminate dependency requirement for HVAC. 278 ventilation to inverters, shutdown boards and ESFAS Provide a permanent tie-in to the construction air Improve availability of air system. 279 compressor.

Add new Unit 2 air compressor similar to the Unit Improve availability of air system. 280 1 D compressor.

Replace the ACAS compressors and dryers. Improve reliability of air system. 281 Enhance procedures for SGTR. Improved mitigation of steam generator tube 123, 127, 128, ruptures. 251 Enhance procedures for refill of RWST. Extend RWST capacity. 33,249 Provide cross-tie to Unit 1 RWST. Extend RWST capacity. 282 Enhance procedures for feed & bleed operation. Improve mitigation of loss of secondary 283 cooling.

Westinghouse Non-Proprietary Class 3 Page 76 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report Table 16 Phase I SAMA Candidates

.lI.-n-ULU -.- , PU--L av-alaunl ty tJ~lao1 Jul ~Ji..l%.

A.llhlll .

5 a ]pl a ti. ,Ll.J yTr capacity. during a Station Blackout (SBO). (Rev A) already installed to provide additional capacity.

2 Replace lead-acid batteries with Extended DC power availability NEI 05-01 Basis for Screening: For a plant with Excessive fuel cells. during an SBO. (Rev A) significant construction already completed, the Implementation Cost cost of implementation caused by replacing all batteries with fuel cells, including structural, electrical, and HVAC changes required, including a fuel supply which does not currently exist on site, would exceed the bounding benefit.

Combine with SAMA 174.

3 Add additional battery charger or Improved availability of DC NEI 05-01 Basis for Screening: There are currently two Already Implemented portable, diesel-driven battery power system. (Rev A) spare chargers already in place.

charger to existing DC system.

4 Improve DC bus load shedding. Extended DC power availability NEI 05-01 SBO procedure includes shedding DC loads to Retain For Phase II during an SBO. (Rev A) extend battery availability (AOl 40 Station Analysis Blackout procedure will be duplicated for Unit 2).

There is a potential for enhancement to shed additional loads to extend battery life.

Therefore this SAMA is retained for Phase II analysis.

5 Provide DC bus cross-ties. Improved availability of DC NEI 05-01 Basis for Screening: Since cross-ties are Very Low Benefit power system. (Rev A) available at the 480V supplies, and the #5 spare battery can be aligned to and supply any of the 4 buses, this SAMA has very little risk benefit.

Combine with SAMA 258.

6 Provide additional DC power to Increased availability of the 120 NEI 05-01 Basis for Screening: The #5 spare battery can Already Implemented the 120/240V vital AC system. V vital AC bus. (Rev A) supply the inverter through DC bus. I

Westinghouse Non-Proprietary Class 3 Page 77 of 142 Our ref. LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report IVAMf NS14AMIATiie *SAVA susio P.asePCominments+ ...... DS iti 7 Add an automatic feature to Increased availability of the 120 NEI 05-01 Basis for Screening: Transfer to DC supply is Already Implemented transfer the 120V vital AC bus V vital AC bus. (Rev A) already an auctioneered automatic transfer.

from normal to standby power. __

8 Increase training on response to Improved chances of successful NEI 05-01 Training is conducted on inadvertent Safety Retain For Phase II loss of two 120V AC buses which response to loss of two 120V AC (Rev A) Injection (SI), and loss of a single AC bus, Analysis causes inadvertent actuation buses. however not on the loss of two 120V buses.

signals. Therefore this SAMA is retained for Phase If analysis. Improvements in this operator training may not be a material benefit in risk reduction.

9 Provide an additional diesel Increased availability of on-site NEI 05-01 Basis for Screening: For a plant with Excessive generator. emergency AC power. (Rev A) significant construction already completed, the Implementation Cost cost of implementation ($8,500,000 to

$22,800,000, representative of similar nuclear power plants, WBN specific cost estimate

$5,000,000) would exceed the bounding benefit.

Combine with SAMA 233.

10 Revise procedure to allow bypass Extended diesel generator NEI 05-01 Basis for Screening: Diesel generator trips are Already Implemented of diesel generator trips. operation. (Rev A) bypassed on emergency start except generator differential and overspeed trips.

11 Improve 4.16-kV bus cross-tie Increased availability of on-site NEI 05-01 Basis for Screening: Procedures AOI-43.01, Already Implemented capability. AC power. (Rev A) 02, 03 and 04 provide proceduralized cross-tie capability for emergency power to any shutdown board from any diesel generator.

12 Create AC power cross-tie Increased availability of on-site NEI 05-01 Basis for Screening: AOI-43.01, 02, 03 and 04 Already Implemented capability with other unit (multi- AC power. (Rev A) proceduralized cross-tie capability for unit site) emergency power to any shutdown board from any diesel generator.

Combine with SAMA 229.

13 Install an additional, buried off- Reduced probability of loss of NEI 05-01 Basis for Screening: There are two existing Excessive site power source, off-site power. (Rev A) 161 kV connections to a nearby dam Implementation Cost switchyard above ground. The estimated cost of burying them would exceed the bounding benefit.

Westinghouse Non-Proprietary Class 3 Page 78 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report increasea avaliaonity OI on-site INt.1 UD-U1 tsasis ior 3creenlng: r or a piant wln Ezxcesslve AC power. (Rev A) significant construction already completed, the Implementation Cost estimated cost of implementation ($3,350,000 to $30,000,000, representative of similar nuclear power plants) would exceed the boundine benefit.

15 Install tornado protection on gas Increased availability of on-site NEI 05-01 Basis for Screening: A gas turbine generator is Not Applicable turbine generator. AC power. (Rev A) not available at the Watts Bar site.

16 Improve uninterruptible power Increased availability of power NEI 05-01 Basis for Screening: A design change is in Already Implemented supplies. supplies supporting front-line (Rev A) process to add 4 inverters and a spare is equipment. available.

17 Create a cross-tie for diesel fuel Increased diesel generator NEI 05-01 Basis for Screening: The capability exists to Already Implemented oil (multi-unit site). availability. (Rev A) supply any of the four 7-day tanks (one for each EDG) from either of the unit supply tanks which are cross tied. The 7-Day tanks for each diesel has a tanker truck connection to refill the tank. Therefore the intent of this SAMA is met with the current design.

18 Develop procedures for Increased diesel generator NEI 05-01 Basis for Screening: Procedures exist for Already Implemented replenishing diesel fuel oil. availability. (Rev A) maintaining long-term operation of the EDGs when necessary, including monitoring and replenishing EDG fuel oil. These procedures are used extensively in license operator initial training and license operator continuing training programs. Therefore, the intent of this SAMA is met with the current procedures and the associated operator training.

19 Use fire water system as a backup Increased diesel generator NEI 05-01 Basis for Screening: Each diesel generator has Already Implemented source for diesel cooling, availability. (Rev A) a permanent backup supply from opposite train ERCW from the other unit. Pump, equipment, and procedures are available to provide cooling water supply from the cooling tower or river. Therefore the intent of this SAMA is met with the current design.

Westinghouse Non-Proprietary Class 3 Page 79 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report S'AfAIf t ~s.AMA 'Tde S6M~~s~o u r,"c si&14t

>DisIpo')~t 20 Add a new backup source of Increased diesel generator NEI 05-01 Basis for Screening: Each diesel generator has Already Implemented diesel cooling, availability. (Rev A) a permanent backup supply from opposite train ERCW from the oiher unit. Cooling water supply from the cooling tower or river is available from an alternate pump. Therefore the intent of this SAMA is met with the current design.

21 Develop procedures to repair or Increased probability of recovery NEI 05-01 Basis for Screening: Spare breakers are Already Implemented replace failed 4 kV breakers. from failure of breakers that (Rev A) available at the shutdown boards and are transfer 4.16 kV non-emergency maintained in accordance with procedure Ml-buses from unit station service 57.01. Procedure GOI-7 provides direction for transformers. racking breakers in if needed. Therefore this SAMA is met with the current procedures.

22 In training, emphasize steps in Reduced human error probability NEI 05-01 Basis for Screening: AOIs exist for dealing Already Implemented recovery of off-site power after an during off-site power recovery. (Rev A) with SBO events, and include a high priority SBO. for steps calling for restoration of offsite power. These procedures are used extensively in license operator initial training and license operator continuing training programs.

Therefore, the intent of this SAMA is met with the current procedures and the associated operator training.

23 Develop a severe weather Improved off-site power recovery NEI 05-01 Basis for Screening: Procedure AOI-8 for Already Implemented conditions procedure. following external weather- (Rev A) tornado and other severe weather procedures, related events, exist for general site preparations and placing the plant in a safe condition depending upon severe weather conditions, and provides guidance to mitigate known vulnerabilities of equipment or systems to specific external events, including missiles generated from tornadoes or high winds and cold weather conditions. These procedures are used extensively in license operator initial training and license operator continuing training programs. Therefore, the intent of this SAMA is met with the current procedures and the associated operator training.

Westinghouse Non-Proprietary Class 3 Page 80 of 142 Our ref: LTR-RAM-L-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report 24 Bury off-site power lines. Improved off-site power NEI 05-01 Basis for Screening: The distance that would Excessive reliability during severe weather. (Rev A) be necessary to bury offsite power lines would Implementation Cost beý significant since severe weather to which transmission lines are susceptible typically affects a broad area. For a plant with significant construction already completed, the estimated cost of implementation would exceed the bounding benefit.

25 Install an independent active or Improved prevention of core melt NEI 05-01 Basis for Screening: The previous passive Excessive passive high pressure injection sequences. (Rev A) UHI system was removed from the Watts Bar Implementation Cost system. design. For a plant with significant construction already completed, the estimated cost of implementation would exceed the bounding benefit.

26 Provide an additional high Reduced frequency of core melt NEI 05-01 Basis for Screening: For a plant with Excessive pressure injection pump with from small LOCA and SBO (Rev A) significant construction already completed, the Implementation Cost independent diesel, sequences. estimated cost of implementation would exceed the bounding benefit.

27 Revise procedure to allow Extended HPCI and RCIC NEI 05-01 Basis for Screening: This is a BWR item. Not Applicable operators to inhibit automatic operation. (Rev A) PWRs do not implement the same logic for vessel depressurization in non- deliberately depressurizing the RCS upon ATWS scenarios. failure of high pressure injection to allow low pressure injection that is used in BWRs.

Therefore, this item is not applicable and is screened from further consideration.

28 Add a diverse low pressure Improved injection capability. NEI 05-01 Basis for Screening: For a plant with Excessive injection system. (Rev A) significant construction already completed, the Implementation Cost estimated cost of implementation would exceed the bounding benefit.

29 Provide capability for alternate Improved injection capability. NEI 05-01 Basis for Screening: There is a minimal Very Low Benefit injection via diesel-driven fire (Rev A) benefit from this SAMA since it does not pump. provide a recirculation path. Therefore it is not considered further. This SAMA is considered cost prohibitive relative to the potential benefit. I

Westinghouse Non-Proprietary Class 3 Page 81 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report S A- MA InMurce , ,hselLeommentsl . Dispostion 2 30 Improve ECCS suction strainers. Enhanced reliability of ECCS NEI 05-01 Basis for Screening: Watts Bar has Already Implemented suction. (Rev A) Implemented the required GSI-191 strainer improvements. This SAMA is met by the current design.

31 Add the-ability to manually align Enhanced reliability of ECCS NEI 05-01 Basis for Screening: Watts Bar has the Already Implemented emergency core cooling system suction. (Rev A) capability to manually align ECCS recirculation. recirculation. This SAMA is met with the current design.

Combine with SAMA 248.

32 Add the ability to automatically Enhanced reliability of ECCS NEI 05-01 Low pressure ECCS automatically aligns for Retain For Phase II align emergency core cooling suction. (Rev A) recirculation from the containment sump, Analysis system to recirculation mode however the high head recirculation is manual upon refueling water storage tank and the operator action is 38% of CDF.

depletion. Therefore this SAMA is retained for further analysis.

Combine with SAMA 238.

33 Provide hardware and procedure Extended reactor water storage NEI 05-01 Basis for Screening: EOPs provide directions Already Implemented to refill the reactor water storage tank capacity in the event of a (Rev A) on monitoring RWST inventory and adding tank once it reaches a specified steam generator tube rupture. water from different sources when necessary.

low level. Therefore, the intent of this SAMA is met with the current procedures.

34 Provide an in-containment reactor Continuous source of water to the NEI 05-01 Basis for Screening: For a plant with Excessive water storage tank. safety injection pumps during a (Rev A) significant construction already completed, the Implementation Cost LOCA event, since water released estimated cost of implementation would from a breach of the primary exceed the bounding benefit. There is limited system collects in the in- room in containment to install an in-containment reactor water storage containment RWST.

tank, and thereby eliminates the need to realign the safety injection pumps for long-term post-LOCA recirculation.

Westinghouse Non-Proprietary Class 3 Page 82 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report i nrotuie low pressure injection nxtenoeo reactor water storage INtl UD-UI n3asis Ior 3creenlng:

pumps earlier in medium or large- tank capacity. (Rev A) ECA-I .1 contains criteria for shutting down a break LOCAs to maintain reactor train of containment spray or low pressure water storage tank inventory. injection and high pressure injection to extend RWST storage capability. Therefore, the intent of this SAMA is met.

36 Emphasize timely recirculation Reduced human error probability NEI 05-01 Basis for Screening: Existing EOPs provide Already Implemented alignment in operator training, associated with recirculation (Rev A) directions for monitoring and conserving failure. water in the containment recirculation sump, including ensuring that maximum injection of water from the RWST occurs prior to performing swapover to containment recirculation. These procedures are used extensively in license operator initial training and license operator continuing training programs, and are practiced in the plant simulator. Therefore, the intent of this SAMA is met with the current operator training.

37 Upgrade the chemical and volume For a plant like the Westinghouse NEI 05-01 Basis for Screening: For a plant with Excessive control system to mitigate small AP600, where the chemical and (Rev A) significant construction already completed, the Implementation Cost LOCAs. volume control system cannot estimated cost of implementation to increase mitigate a small LOCA, an CVCS flow capacity would exceed the upgrade would decrease the bounding benefit.

frequency of core damage.

38 Change the in-containment Reduced common mode failure of NEI 05-01 Basis for Screening: This item only applies to Not Applicable reactor water storage tank suction injection paths. (Rev A) AP600 plants that have the RWST located from four check valves to two inside of containment. Therefore, this item is check and two air-operated not applicable and is screened from further valves, consideration.

Westinghouse Non-Proprietary Class 3 Page 83 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report safety injection pumps with the safety injection system. This (Rev A) significant construction already completed, the Implementation Cost diesel-powered pumps. SAMA was originally intended estimated cost of implementation to replace for the Westinghouse-CE System the SI pumps would exceed the bounding 80+, which has four trains of benefit.

safety injection.

However, the intent of this SAMA is to provide diversity within the high- and low-pressure safety iniection systems.

40 Provide capability for remote, Improved chance of successful NEI 05-01 Basis for Screening: WBN has capability for Already Implemented manual operation of secondary operation during station blackout (Rev A) remote manual operation of the SG side pilot-operated relief valves in events in which high area atmospheric dump valves via nitrogen a station blackout, temperatures may be encountered stations. Therefore the intent of this SAMA is (no ventilation to main steam met.

areas).

41 Create a reactor coolant Allows low pressure emergency NEI 05-01 Basis for Screening: For a plant with Excessive depressurization system. core cooling system injection in (Rev A) significant construction already completed, the Implementation Cost the event of small LOCA and estimated cost of implementation to install high-pressure safety injection larger PORVs would exceed the bounding failure. benefit.

42 Make procedure changes for Allows low pressure emergency NEI 05-01 The current EOP network provides guidance Not Applicable reactor coolant system core cooling system injection in (Rev A) for depressurizing RCS but may not be depressurization. the event of small LOCA and adequate for small LOCAs with only low head high-pressure safety injection injection available. Changes to the EOPs are failure. processed through the owners group ERG maintenance process.

Since this change to the EOPs is not within TVA's control it can not be implemented at this time and therefore a cost benefit analysis is not performed.

43 Add redundant DC control power Increased availability of SW. NEI 05-01 Basis for Screening: The Watts Bar design Already Implemented for SW pumps. (Rev A) includes two DC busses for control power for the ERCW pumps. This SAMA is met with the current design.

Westinghouse Non-Proprietary Class 3 Page 84 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report Keplace hLuuS pump motors NN tlimination ot tLt_6 oepenoency basis ior :screening: w jiN nas air cooied air-cooled motors. on component cooling system. (Rev A) motors on the ECCS pumps. This SAMA is met with the current design.

45 Enhance procedural guidance for Reduced frequency of loss of NEI 05-01 Watts Bar has the capability to cross-tie CCS Retain For Phase II use of cross-tied component component cooling water and (Rev A) trains and ERCW trains, and a flood mode Analysis cooling or service water pumps. service water. procedure exists to supply CCS from ERCW by installing a spool piece.

This SAMAwill be retained for further analysis to review procedural guidance in AOI- 15 for potential upgrades to comply with this SAMA.

46 Add a service water pump. Increased availability of cooling NEI 05-01 An alternate pump exists that can be Retain For Phase II water. (Rev A) temporarily connected to the ERCW system to Analysis provide ERCW capability, however a permanent diesel driven 10,000 gpm pump could be installed at the IPS flush connection to provide increased ERCW availability.

Therefore this SAMA will be. retained for further evaluation.

47 Enhance the screen wash system. Reduced potential for loss of SW NEI 05-01 Basis for Screening: The location of the intake Very Low Benefit due to clogging of screens. (Rev A) on the river is protected from debris therefore there is minimal benefit of this SAMA.

Combine with SAMA 202

Westinghouse Non-Proprietary Class 3 Page 85 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report

~SAMA<SAA Tijie *M4Diu 6-n ~ Vourcc ' Ph'a.eiljeirnnents ~ >ipsfo 48 Cap downstream piping of Reduced frequency of loss of NEI 05-01 Basis for Screening: To minimize the Already Implemented normally closed component component cooling water (Rev A) possibility of leakage from piping, valves, and cooling water drain and vent initiating events, some of which equipment, welded construction is used valves, can be attributed to catastrophic wherever possible. Except for the normally failure of one of the many single closed makeup line and equipment vent and isolation valves, drain lines, there are no direct connections between the CCS system and other systems.

The equipment vent and drain lines outside the containment have manual valves which are normally closed unless the equipment is being vented or drained for maintenance or repair operations. Failure of the socket welds attaching vent and drain lines to the CCS system process piping is not likely, but is more likely than failure of manual drain and vent valves to stay closed. Therefore, additional capping of the drain and vent lines provides very little additional assurance against leakage from the CCS system that may result in a total loss of CCS, and the intent of this SAMA is met with the current design.

49 Enhance loss of component Reduced potential for reactor NEI 05-01 Basis for Screening: AOI- 15 requires tripping Already Implemented cooling water (or loss of service coolant pump seal damage due to (Rev A) the RCPs immediately as a first step upon loss water) procedures to facilitate pump bearing failure. of CCS. Therefore, the intent of this SAMA is stopping the reactor coolant met with the current procedures.

pumps.

50 Enhance loss of component Reduced probability of reactor NEI 05-01 Basis for Screening: Upon receipt of any RCP Very Low Benefit cooling water procedure to coolant pump seal failure. (Rev A) seal no. I outlet temperature high alarm, AOl-underscore the desirability of 15 & 24 require an RCS cooldown after cooling down the reactor coolant isolation of the CCS path to the RCP thermal system prior to seal LOCA. barrier and isolation of RCP seal injection.

This order of actions is deemed appropriate for overall plant stabilization following a loss of CCS. Enhanced procedure will not affect the risk because of the rapid progression of the seal leak. Therefore, the intent of this SAMA is minimal benefit.

Westinghouse Non-Proprietary Class 3 Page 86 of 142 Our ref. LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report Additional training on loss ot improved success ot operator 15asis ior ,3creening: AUI-1i exists Ior a ioss component cooling water. actions after a loss of component (Rev A) of CCS, and is used extensively in license cooling water. operator initial training and license operator continuing training programs. Therefore, the intent of this SAMA is met with the current procedures and the associated operator training.

Combine with SAMA 260.

52 Provide hardware connections to Reduced effect of loss of NEI 05-01 Basis for Screening: WBN does not require Not Applicable allow another essential raw component cooling water by (Rev A) cooling to the charging pump seals. This cooling water system to cool providing a means to maintain the SAMA is not applicable to the Watts Bar charging pump seals. charging pump seal injection design.

following a loss of normal cooling water.

53 On loss of essential raw cooling Increased time before loss of NEI 05-01 Basis for Screening: AOI-13 for ERCW Very Low Benefit water, proceduralize shedding component cooling water (and (Rev A) system loss or rupture does not provide component cooling water loads to reactor coolant pump seal failure) directions to quickly implement loss of CCS extend the component cooling during loss of essential raw procedure AOI- 15 if ERCW cannot be water heat-up time. cooling water sequences. restored. AOI-13, however, does provide directions to trip all of the RCPs, isolate thermal barrier cooling, cooldown the plant and cross-tie ERCW if available.

There is minimal risk reduction for CCS load shedding since this is a timing issue for recovery of ERCW. The PRA model assumes loss of ERCW is non-recoverable within the 24 hr mission time. Therefore this SAMA has very low risk improvement benefit.

54 Increase charging pump lube oil Increased time before charging NEI 05-01 Basis for Screening: The WBN A charging Already Implemented capacity. pump failure due to lube oil (Rev A) pump design has alternate ERCW supply to overheating in loss of cooling the lube oil cooler. Therefore the intent of this water sequences. SAMA is met.

Combine with SAMA 267.

Westinghouse Non-Proprietary Class 3 Page 87 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report SMAsAMATIe ni ~ SA114.1&cussivn so~i PhIiniei 4T1 Disposition Nu~mber> ~ , .-

55 Install an independent reactor Reduced frequency of core NEI 05-01 Basis for Screening: For a plant with Excessive coolant pump seal injection damage from loss of component (Rev A) significant construction already completed, the Implementation Cost system, with dedicated diesel. cooling water, service water, or estimated cost of implementation would station blackout. exceed the bounding benefit.

56 Install an independent reactor Reduced frequency of core NEI 05-01 There is potential to install a small pump in Retain For Phase II coolant pump seal injection damage from loss of component (Rev A) the PD pump room. This would be useful for Analysis system, without dedicated diesel. cooling water or service water, loss of ERCW and loss of CCS which but not a station blackout. contributes 35% of the core damage. Suction and discharge piping and power is available in the PD pump room. Costs include dismantling current PD pump and installing new low capacity high pressure pump. Room cooling requirements will need to be evaluated. This SAMA is retained for further evaluation.

57 Use existing hydro test pump for Reduced frequency of core NEI 05-01 Basis for Screening: Watts Bar does not have Not Applicable reactor coolant pump seal damage from loss of component (Rev A) an existing hydro test pump. This SAMA is injection, cooling water or service water, not applicable.

but not a station blackout.

58 Install improved reactor coolant Reduced likelihood of reactor NEI 05-01 Unit 2 has the upgraded high temperature o- Not Applicable pump seals. coolant pump seal LOCA. (Rev A) rings in the Reactor Coolant Pumps. A new seal insert design has been proposed by Westinghouse which could eliminate seal LOCA sequences. Pending topical report approval, this alternate seal design may prove cost effective, however costs are unknown at this time.

Since this change is not within TVA's control it can not be implemented at this time and therefore a cost benefit analysis is not performed.

Combine with SAMA 232

Westinghouse Non-Proprietary Class 3 Page 88 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report Install an additional component Keduce lixkelinood o0 loss or iasis tor screenng: mne watts tsar aesign cooling water pump. component cooling water leading (Rev A) includes a swing 5th common spare CCS to a reactor coolant pump seal pump which can be powered from either train.

LOCA. There is limited space for an additional CCS pump. The cost of additional pump located in different space would be cost prohibitive.

Therefore the intent of this SAMA is met by the current design.

60 Prevent makeup pump flow Reduced frequency of loss of NEI 05-01 Basis for Screening: WBN does not have Not Applicable diversion through the relief reactor coolant pump seal cooling (Rev A) charging pump relief valves, the mini-flow valves, if spurious high pressure injection line contains two normally open MOVs with relief valve opening creates a flow power disconnected with orifices which diversion large enough to prevent recirculates flow back to the VCT. This reactor coolant pump seal SAMA is not applicable to the WBN design.

injection.

61 Change procedures to isolate Reduced frequency of core NEI 05-01 Basis for Screening: Procedure AOI-15 for Already Implemented reactor coolant pump seal return damage due to loss of seal (Rev A) loss of CCS includes instruction for isolating flow on loss of component cooling. RCP seals. Therefore, the intent of this SAMA cooling water, and provide (or is met with the current procedures.

enhance) guidance on loss of injection during seal LOCA.

62 Implement procedures to stagger Extended high pressure injection NEI 05-01 Basis for Screening: Procedure AOI-13 directs Already Implemented high pressure safety injection prior to overheating following a (Rev A) use of fire water to cool the A charging pump pump use after a loss of service loss of service water. on loss of ERCW. Therefore the intent of this water. SAMA is met with the current procedures.

63 Use fire prevention system pumps Reduced frequency of reactor NEI 05-01 Basis for Screening: WBN does not have high Not Applicable as a backup seal injection and coolant pump seal LOCA. (Rev A) pressure fire pumps. This SAMA is not high pressure makeup source. applicable to WBN.

64 Implement procedure and Improved ability to cool residual NEI 05-01 Basis for Screening: The Watts Bar design Already Implemented hardware modifications to allow heat removal heat exchangers. (Rev A) includes a CCS header cross-tie. Procedure manual alignment of the fire AOI-7.07 provided direction to use ERCW as water system to the component a cooling medium for RHR, spent fuel pit, and cooling water system, or install a sample heat exchangers. Therefore the intent component cooling water header of this SAMA is met.

cross-tie.

65 Install a digital feedwater Reduced chance of loss of main NEI 05-01 Basis for Screening: Design change is in Already Implemented upgrade.. feedwater following a plant trip. (Rev A) process to install digital feedwater control.

I I_This SAMA is met.

Westinghouse Non-Proprietary Class 3 Page 89 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report SAA.

• S, IAMA Titl* *AMA Discussion' S<kircV Phase I Commenti, *nsDi"posin

~Number 66 Create ability for emergency Increased availability of NEI 05-01 Basis for Screening: The Watts Bar design Already Implemented connection of existing or new feedwater. (Rev A) includes provisions for emergency connection water sources to feedwater and for sources to the feedwater and aux feedwater condensate systems. systems. The DWST can be connected to the CST to supply water to hotwell, condensate and then main feedwater. This lineup is proceduralized in SOI-59.01. The ERCW system can supply AFW via a hard pipe connection. There is a designed feature of AFW to swap over to ERCW supply on low level of CST. There also is provision for a flood mode spool piece connection from the fire protection system to the AFW discharge.

Procedure AOI-7.06 directs installation of this spool piece. Pump, equipment, and procedures are available to provide cooling water supply from the cooling tower or river. These emergency connections met the intent of this SAMA.

67 Install an independent diesel for Extended inventory in CST NEI 05-01 Basis for Screening: An alternate diesel driven Already Implemented the condensate storage tank during an SBO. (Rev A) pump and an alternate diesel generator are makeup pumps. available to provide this capability. The diesel fire pump is capable of makeup to the CST.

Additionally an onsite pumper truck is available for makeup to the CST. Procedures are in place to perform these actions. These means for making up to the CST meet the intent of this SAMA.

68 Add a motor-driven feedwater Increased availability of NEI 05-01 Basis for Screening: A motor-driven Standby Already Implemented pump. feedwater. (Rev A) Feedwater pump is available that can be used up to 18% load. Therefore, the intent of this SAMA is met with the current design.

Combine with SAMA 196.

69 Install manual isolation valves Reduced dual turbine-driven NEI 05-01 Basis for Screening: The WBN design has one Already Implemented around auxiliary feedwater pump maintenance unavailability. (Rev A) turbine driven AFW pump with isolation turbine-driven steam admission valves. Therefore this SAMA is not applicable valves, to WBN.

Westinghouse Non-Proprietary Class 3 Page 90 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts- Bar Unit 2 SAMA Report Install accumulators tor turoine- timinates tne neea tor iocai j ne w tN turtine anven AV W pump flow Ketamn Po driven auxiliary feedwater pump manual action to align nitrogen (Rev A) control valves have a nitrogen supply that can Analysis flow control valves. bottles for control air following a be manually aligned. The nitrogen backup is loss of off-site power. not credited in SBO risk model. Installing accumulators to eliminate this manual action may be minimal risk benefit, however this SAMA is retained for further evaluation.

71 Install a new condensate storage Increased availability of the NEI 05-01 The two unit CSTs are cross-tied so that they Retain For Phase II tank (auxiliary feedwater storage auxiliary feedwater system. (Rev A) can supply either unit. A previous estimate of Analysis tank). $300K to replace the existing PWST on Unit 1 with a stainless steel tank was used to estimate the cost of this SAMA. Installation of a new third CST would require a new pad, and piping to tie it in to the AFW supply. This SAMA is retained for further evaluation.

72 Modify the turbine-driven Improved success probability NEI 05-01 Basis for Screening: The current WBN turbine Already Implemented auxiliary feedwater pump to be during a station blackout. (Rev A) driven AFW pump is self-cooled, therefore self-cooled. this SAMA is not applicable.

73 Proceduralize local manual Extended auxiliary feedwater NEI 05-01 Basis for Screening: AOl-10 provides Already Implemented operation of auxiliary feedwater availability during a station (Rev A) guidance for local manual operation of the system when control power is blackout. Also provides a success turbine-driven AFW pump. Therefore, the lost. path should auxiliary feedwater intent of this SAMA is met with the current control power be lost in non- procedures.

station blackout sequences.

74 Provide hookup for portable Extended auxiliary feedwater NEI 05-01 Basis for Screening: The #5 battery is Already Implemented generators to power the turbine- availability. (Rev A) available to supply one channel of control driven auxiliary feedwater pump power for the turbine-driven AFW pump. An after station batteries are depleted. alternate power supply is also available for the battery charger. Therefore the intent of this SAMA is met with the current design.

75 Use fire water system as a backup Increased availability of steam NEI 05-01 Basis for Screening: The use of fire water as a Already Implemented for steam generator inventory, generator water supply. (Rev A) backup for steam generator inventory is implemented in the flood mode procedure.

Therefore the intent of this SAMA is met with the current design.

Westinghouse Non-Proprietary Class 3 Page 91 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report SA 4 -, sA* ATitl* .... I kisgiok n 1z14'O ýS;wce

• -ý" *Dispiionti I

?Number ~7 76 Change failure position of Allows greater inventory for the NEI 05-01 Basis for Screening: The condenser makeup Already Implemented condenser makeup valve if the auxiliary feedwater pumps by (Rev A) valve (valve 2-9) is normally closed, air condenser makeup valve fails preventing condensate storage operated to open, and fails closed. Therefore open on loss of air or power. tank flow diversion to the this SAMA is not applicable to WBN.

condenser.

77 Provide a passive, secondary-side Reduced potential for core NEI 05-01 Basis for Screening: For a plant with Excessive heat-rejection loop consisting of a damage due to loss-of-feedwater (Rev A) significant construction already completed, the Implementation Cost condenser and heat sink. events. estimated cost of implementation would exceed the bounding benefit.

78 Modify the startup feedwater Increased reliability of decay heat NEI 05-01 Basis for Screening: Implementation of this Excessive pump so that it can be used as a removal. (Rev A) SAMA requires a flow path around the Implementation Cost backup to the emergency isolation valves. Also for use during a station feedwater system, including blackout the Standby Feedwater pump would during a station blackout scenario, have to be powered from a diesel generator.

For a plant with significant construction already completed, the estimated cost of implementation would exceed the bounding benefit.

79 Replace existing pilot-operated Increased probability of NEI 05-01 Basis for Screening: The Watts bar success Very Low Benefit relief valves with larger ones, successful feed and bleed. (Rev A) criteria for bleed and feed is two PORVs only such that only one is required for if charging is not available. Otherwise one successful feed and bleed. PORV is sufficient. Larger valves would require piping changes, block valve changes, and analysis changes. There is a larger probability of leakage with larger valves.

Based on this, this SAMA provides little benefit for the estimated cost.

80 Provide a redundant train or Increased availability of NEI 05-01 Basis for Screening: Provisions for Very Low Benefit means of ventilation. components dependent on room (Rev A) compensatory ventilation is in place for the cooling. 480V electric board rooms and margin to room heatup limits exists in the 480V transformer room. Plant chillers are being upgraded based on Freon considerations. This SAMA is considered~not cost beneficial due to low risk benefit.

Westinghouse Non-Proprietary Class 3 Page 92 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report WSAMAr SAMA Title SAMA Discussion .our1e Phase I Commnm6Is ~ Dispoito 81 Add a diesel building high Improved diagnosis of a loss of NEI 05-01 Basis for Screening: The diesel generator Very Low Benefit temperature alarm or redundant diesel building HVAC. (Rev A) building is manned during DG starts, and louver and thermostat. shiftly operator rounds take temperature measurements per SI-2. Therefore this SAMA is considered very low benefit.

82 Stage backup fans in switchgear Increased availability of NEI 05-01 Basis for Screening: Fans are staged in the Already Implemented rooms. ventilation in the event of a loss (Rev A) 480V electric board rooms. The shutdown of switchgear ventilation. board rooms have 2 trains of cooling available. Therefore the intent of this SAMA is met with the current design.

83 Add a switchgear room high Improved diagnosis of a loss of NEI 05-01 Basis for Screening: Shiftly operator rounds Already Implemented temperature alarm. switchgear HVAC. (Rev A) check temperatures per SI-2. Shutdown board room chillers swap on high temperature and provides a control room alarm. Therefore the intent of this SAMA is met with the current design.

84 Create ability to switch Continued fan operation in a NEI 05-01 Basis for Screening: A DC powered fan is Already Implemented emergency feedwater room fan station blackout. (Rev A) installed in the AFW pump room in addition power supply to station batteries to the AC powered fan. Therefore the intent of in a station blackout. this SAMA is met with the current design.

85 Provide cross-unit connection of Increased ability to vent NEI 05-01 Basis for Screening: This is a BWR item and Not Applicable uninterruptible compressed air containment using the hardened (Rev A) not applicable to WBN.

supply. vent.

86 Modify procedure to provide Increased availability of NEI 05-01 Basis for Screening: Two of the four station Already Implemented ability to align diesel power to instrument air after a LOOP. (Rev A) non-safety related air compressors have more air compressors. capability to align to diesel power. Both of those are needed to supply full plant loads.

The 2 safety related air compressors are diesel backed. Powering the D air compressor from the emergency diesel generator was evaluated and prohibited by diesel loading. The current design of powering the two air compressors meets the intent of this SAMA.

Westinghouse Non-Proprietary Class 3 Page 93 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report

?SVAMA V1AATitle> V<.

AYTA Discussiont _ Source Phase I Comment~s Dsiito

~Number~7-87 Replace service and instrument Elimination of instrument air NEI 05-01 Watts Bar is evaluating the status of the Retain For Phase II air compressors with more system dependence on service (Rev A) construction air compressors. Permanent Analysis reliable compressors which have water cooling, installation of this air compressor needs to self-contained air cooling by shaft consider HVAC requirements for the self-driven fans. cooled compressor. This SAMA is retained for further evaluation.

88 Install nitrogen bottles as backup Extended SRV operation time. NEI 05-01 Basis for Screening: The steam generator Already Implemented gas supply for safety relief valves. (Rev A) PORVs are designed with a nitrogen bottle backup. Therefore the intent of this SAMA is met with the current design.

89 Improve SRV and MSIV Improved availability of SRVs NEI 05-01 The Main Steam System is monitored in the Already Implemented pneumatic components. and MSIVs. (Rev A) Maintenance Rule Program. Reliability improvements have been implemented such as replacing the single MSIV regulator with two regulators. Watts Bar has not experienced the MSIV sticking issues identified at other plants. Therefore, the intent of this SAMA is met with the current design.

90 Create a reactor cavity flooding Enhanced debris cool ability, NEI 05-01 Basis for Screening: For a plant with Excessive system. reduced core concrete interaction, (Rev A) significant construction already completed, the Implementation Cost and increased fission product estimated cost of implementation ($8,750,000, scrubbing. representative of similar nuclear power plants) would exceed the bounding benefit.

91 Install a passive containment Improved containment spray NEI 05-01 Basis for Screening: The source of this SAMA Excessive spray system. capability. (Rev A) is the AP600 Design Certification Review Implementation Cost submittal. For a plant with significant construction already completed, the cost of implementation ($20,000,000, representative of similar nuclear power plants) would exceed the bounding benefit.

Westinghouse Non-Proprietary Class 3 Page 94 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report

[*S'AM"= S.. MA* Title'o "A 141A MAlDiscussioe Ne I <."Disstt DSourciMPhaseCiirmen.

92 Use the fire water system as a Improved containment spray NEI 05-01 Basis for Screening: Although there are two 2- Very Low Benefit backup source for the capability. (Rev A) inch test connections (72-545 & 544) that containment spray system. could be used to connect fire water to containment spray, this lineup bypasses the containment spray heat exchangers and would not remove containment heat. It also can not recirculate water from the containment sump.

The low flow rate would be ineffective for fission product removal. Therefore this SAMA is considered very low benefit.

Combine with SAMA 170.

93 Install an unfiltered, hardened Increased decay heat removal NEI 05-01 Basis for Screening: For a plant with Excessive containment vent. capability for non-ATWS events, (Rev A) significant construction already completed, the Implementation Cost without scrubbing released fission estimated cost of implementation ($3,100,000, products. representative of similar nuclear power plants) would exceed the bounding benefit.

94 Install a filtered containment vent Increased decay heat removal NEI 05-01 Basis for Screening: For a plant with Excessive to remove decay heat. capability for non-ATWS events, (Rev A) significant construction already completed, the Implementation Cost with scrubbing of released fission estimated cost of implementation ($5,700,000, Option 1: Gravel Bed Filter products. representative of similar nuclear power plants) would exceed the bounding benefit.

Option 2: Multiple Venturi Scrubber 95 Enhance fire protection system Improved fission product NEI 05-01 Basis for Screening: Enhancements to the Excessive and standby gas treatment system scrubbing in severe accidents. (Rev A) EGTS and ABGTS filters to provide Implementation Cost hardware and procedures. scrubbing for ISLOCA source terms would exceed the bounding benefit.

EPSIL already contains instructions for spraying release points with fire water, which would provide fission product scrubbing.

96 Provide post-accident Reduced likelihood of hydrogen NEI 05-01 Basis for Screening: SAG-7 provides guidance Already Implemented containment inerting capability, and carbon monoxide gas (Rev A) for steam inerting the containment. Therefore combustion. the intent of this SAMA is met with the current design and procedures.

Westinghouse Non-Proprietary Class 3 Page 95 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report SAMA'1f <S ~T~itce . S AMAiscuission S..urc ~ Phase flcomm~ient% ~ Disp)o~s it;io 97 Create a large concrete crucible Increased cooling and NEI 05-01 Basis for Screening: For a plant with Excessive with heat removal potential to containment of molten core (Rev A) significant construction already completed, the Implementation Cost contain molten core debris, debris. Molten core debris estimated cost of implementation escaping from the vessel is ($90,000,000 to $108,000,000, representative contained within the crucible and of similar nuclear power plants) would exceed a water cooling mechanism cools the bounding benefit.

the molten core in the crucible, preventing melt-through of the base mat.

98 Create a core melt source Increased cooling and NEI 05-01 Basis for Screening: For a plant with Excessive reduction system. containment of molten core (Rev A) significant construction already completed, the Implementation Cost debris. Refractory material would estimated cost of implementation be placed underneath the reactor ($90,000,000, representative of similar nuclear vessel such that a molten core power plants) would exceed the bounding falling on the material would melt benefit.

and combine with the material.

Subsequent spreading and heat removal from the vitrified compound would be facilitated, and concrete attack would not occur.

99 Strengthen primary/secondary Reduced probability of NEI 05-01 Basis for Screening: For a plant with Excessive containment (e.g., add ribbing to containment over-pressurization. (Rev A) significant construction already completed, the Implementation Cost containment shell). cost of implementation would exceed the bounding benefit.

100 Increase depth of the concrete Reduced probability of base mat NEI 05-01 Basis for Screening: For a plant with Excessive base mat or use an alternate melt-through. (Rev A) significant construction already completed, the Implementation Cost concrete material to ensure melt- cost of implementation caused by through does not occur, reconstruction of the containment building would exceed the bounding benefit.

101 Provide a reactor vessel exterior Increased potential to cool a NEI 05-01 Basis for Screening: For a plant with Excessive cooling system. molten core before it causes (Rev A) significant construction already completed, the Implementation Cost vessel failure, by submerging the cost of implementation ($2,500,000 to lower head in water. $4,700,000, representative of similar nuclear power plants) exceeds the bounding benefit.

Westinghouse Non-Proprietary Class 3 Page 96 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report

~SAM71 ~~SAMTMMA SA AJi~i-u Y.on" ~ Sf)UM Phase I cfpnrcnmesDiosi

'.Number _________

102 Construct a building to be Reduced probability of NEI 05-01 Basis for Screening: For a plant with Excessive connected to primary/secondary containment over-pressurization. (Rev A) significant construction already completed, the Implementation Cost containment and maintained at a cost of implementation ($10,000,000 and up, vacuum. representative of similar nuclear power plants) would exceed the bounding benefit.

103 Institute simulator training for Improved arrest of core melt NEI 05-01 Basis for Screening: The estimated cost to Excessive severe accident scenarios, progress and prevention of (Rev A) upgrade the simulator to extend its capability Implementation Cost containment failure, to severe accidents is estimated as a $2 million to $5 million upgrade. The estimated cost of implementation would exceed the bounding benefit.

104 Improve leak detection Increased piping surveillance to NEI 05-01 Basis for Screening: Visual piping inspection Already Implemented procedures. identify leaks prior to complete (Rev A) is performed each outage and some forced failure. Improved leak detection outages. Inspections of the lower compartment would reduce LOCA frequency. inside the crane wall during power operation is not possible due to dose considerations.

Therefore the intent of this SAMA is met with the current inspection program.

105 Delay containment spray Extended reactor water storage NEI 05-01 Basis for Screening: Delay of containment Excessive actuation after a large LOCA. tank availability. (Rev A) spray actuation would require reanalysis of Implementation Cost safety analysis. Current safety analysis does not allow actuation delay. Cost of re-analysis and implementation would exceed the maximum benefit.

106 Install automatic containment Extended time over which water NEI 05-01 Basis for Screening: The estimated cost of Excessive spray pump header throttle remains in the reactor water (Rev A) implementing a design change including Implementation Cost valves. storage tank, when full reanalysis of the safety analysis is considered containment spray flow is not excessive cost compared to the risk benefit.

needed.

107 Install a redundant containment Increased containment heat NEI 05-01 Basis for Screening: Two containment spray Already Implemented spray system. removal ability. (Rev A) trains and two RHR spray trains provides redundancy for the containment spray function. Therefore the intent of this SAMA is met with the current design.

Westinghouse Non-Proprietary Class 3 Page 97 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report vAMA

. I /.AMoc isuiee SAM.AITiiIu PhU4 TCl4mivients I ipos~t ni on Number;, 7' .,

108 Install an independent power Reduced hydrogen detonation NEI 05-01 Basis for Screening: An alternate power Already Implemented supply to the hydrogen control potential. (Rev A) supply to the hydrogen igniters was system using either new batteries, implemented. Therefore the intent of this a non-safety grade portable SAMA is met with the current design.

generator, existing station batteries, or existing AC/DC independent power supplies, such as the security system diesel.

109 Install a passive hydrogen control Reduced hydrogen detonation NEI 05-01 Basis for Screening: For a plant with Excessive system. potential. (Rev A) significant construction already completed, the Implementation Cost estimated cost of implementation of a catalytic converter system would exceed the bounding benefit.

110 Erect a barrier that would provide Reduced probability of NEI 05-01 Basis for Screening: For a plant with Excessive enhanced protection of the containment failure. (Rev A) significant construction already completed, the Implementation Cost containment walls (shell) from estimated cost of implementation would ejected core debris following a exceed the risk benefit.

core melt scenario at high pressure.

111 Install additional pressure or leak Reduced ISLOCA frequency. NEI 05-01 Basis for Screening: Monitoring Already Implemented monitoring instruments for (Rev A) instrumentation such as: level and temperature detection of ISLOCAs. alarms, aux building radiation monitors, RHR leak indication, exist in the control room to cue operators to pipe breaks/leaks in the aux building. Procedures exist (i.e., ECA-1.2) to respond to a LOCA outside containment.

Therefore the intent of this SAMA is met with the current design and procedures.

Combine with SAMA 239.

112 Add redundant and diverse limit Reduced frequency of NEI 05-01 Most of the containment isolation valves are Retain For Phase II switches to each containment containment isolation failure and (Rev A) air operated valves, however the ECCS valves Analysis isolation valve. ISLOCAs. are mostly motor operated. The status of the valves have redundant indication in control room.

This SAMA will be retained to evaluate the number of applicable CIVs and the cost of installing limit switches.

Westinghouse Non-Proprietary Class 3 Page 98 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report SAMAI SAMAvf TI~ SAMA, Iiiýscion Source PhastieI Coment'sisoiton 113 Increase leak testing of valves in Reduced ISLOCA frequency. NEI 05-01 Basis for Screening: At WBN, valves in the Already Implemented ISLOCA paths. (Rev A) ISLOCA paths are tested in accordance with approved procedures every outage. RHR suction valves are not accessible at power.

Therefore, the intent of this SAMA is met with the current procedures.

Combine with SAMA 181.

114 Install self-actuating containment Reduced frequency of isolation NEI 05-01 Basis for Screening: CIVs that are not Already Implemented isolation valves, failure. (Rev A) required to open during an accident are generally air operated, spring to close.

Therefore the intent of this SAMA is met with the current design.

Combine with SAMA 179.

115 Locate residual heat removal Reduced frequency of ISLOCA NEI 05-01 Basis for Screening: For a plant with Excessive (RHR) inside containment, outside containment. (Rev A) significant construction already completed, the Implementation Cost estimated cost of implementation

($28,000,000, representative of similar nuclear power plants) would exceed the bounding benefit.

Combine with SAMA 178.

116 Ensure ISLOCA releases are Scrubbed ISLOCA releases. NEI 05-01 Basis for Screening: The cost of Very Low Benefit scrubbed. One method is to plug (Rev A) implementation of this SAMA has not been drains in potential break areas so estimated in detail. A minimum value of that break point will be covered $1OOK for a hardware change is assumed for with water. screening purposes. Auxiliary building releases are scrubbed by the Aux Building Gas Treatment System (ABGTS), however the ABGTS may not be sized for ISLOCA releases. RHR suction and discharge lines are in the overhead and therefore would not be submerged. Therefore this SAMA is considered very low benefit.

Combine with SAMA 237.

Westinghouse Non-Proprietary Class 3 Page 99 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report SAMA ISAZA'SMATie i.ctssion ;ourcc~ Phase I Comments r~oil 117 Revise EOPs to improve ISLOCA Increased likelihood that LOCAs NEI 05-01 Basis for Screening: ECA-1.2 for LOCA Already Implemented identification, outside containment are identified (Rev A) outside containment is current with current as such. A plant had a scenario in industry guidance. Therefore, the intent of this which an RHR ISLOCA could SAMA is met with the current procedures.

direct initial leakage back to the pressurizer relief tank, giving indication that the LOCA was inside containment.

118 Improve operator training on Decreased ISLOCA NEI 05-01 Basis for Screening: EOP network exists for Already Implemented ISLOCA coping. consequences. (Rev A) coping with ISLOCA symptoms, and are used extensively in license operator initial training and license operator continuing training programs. Therefore, the intent of this SAMA is met with the current procedures and the associated operator training.

119 Institute a maintenance practice to Reduced frequency of steam NEI 05-01 Basis for Screening: The current cost of steam Excessive perform a 100% inspection of generator tube ruptures. (Rev A) generator eddy current inspection is Implementation Cost steam generator tubes during each approximately $1million per steam generator.

refueling outage. The cost of performing 100% inspection including the cost of the added outage time would exceed the bounding benefit.

120 Replace steam generators with a Reduced frequency of steam NEI 05-01 Basis for Screening: The cost of replacing the Excessive new design. generator tube ruptures. (Rev A) steam generators at Watts Bar Unit 1 was Implementation Cost

$221,760,000. This exceeds the bounding benefit.

121 Increase the pressure capacity of Eliminates release pathway to the NEI 05-01 Basis for Screening: For a plant with Excessive the secondary side so that a steam environment following a steam (Rev A) significant construction already completed, the Implementation Cost generator tube rupture would not generator tube rupture. estimated cost of implementation would cause the relief valves to lift. exceed the bounding benefit.

122 Install a redundant spray system Enhanced depressurization NEI 05-01 Basis for Screening: Normal and auxiliary Excessive to depressurize the primary capabilities during steam (Rev A) pressurizer spray capability is available in the Implementation Cost system during a steam generator generator tube rupture. current design. The estimated cost of tube rupture. implementation of a new pressurizer spray system would exceed the bounding benefit.

Westinghouse Non-Proprietary Class 3 Page 100 of 142 Our ref. LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report Proceduralize use of pressurizer Backup method to using NEI U5-t B3asis tor ;creening: Virection for use or me vent valves during steam pressurizer sprays to reduce (Rev A) pressurizer PORV valves are already generator tube rupture sequences. primary system pressure proceduralized in EOP E-3. Therefore the following a steam generator tube intent of this SAMA is met with the current runture. Drocedures.

124 Provide improved instrumentation Improved mitigation of steam NEI 05-01 Basis for Screening: For a plant with Excessive to detect steam generator tube generator tube ruptures. (Rev A) significant construction already completed, the Implementation Cost ruptures, such as Nitrogen- 16 estimated cost of implementation of rad monitors. monitors for each steam generator would exceed the bounding benefit.

125 Route the discharge from the Reduced consequences of a steam NEI 05-01 Basis for Screening: For a plant with Excessive main steam safety valves through generator tube rupture. (Rev A) significant construction already completed, the Implementation Cost a structure where a water spray estimated cost of implementation of a new would condense the steam and structure would exceed the bounding benefit.

remove most of the fission products.

126 Install a highly reliable (closed Reduced consequences of a steam NEI 05-01 Basis for Screening: For a plant with Excessive loop) steam generator shell-side generator tube rupture. (Rev A) significant construction already completed, the Implementation Cost heat removal system that relies on estimated cost of implementation of a air natural circulation and stored cooled isolation condenser would exceed the water sources bounding benefit.

127 Revise emergency operating Reduced consequences of a steam NEI 05-01 Basis for Screening: EOPs for response to a Already Implemented procedures to direct isolation of a generator tube rupture. (Rev A) SGTR contain guidance to ensure that a faulted steam generator. faulted SG is isolated as long as an intact SG remains available. Therefore, the intent of this SAMA is met with the current procedures.

128 Direct steam generator flooding Improved scrubbing of steam NEI 05-01 Basis for Screening: Procedure E-3 directs Already Implemented after a steam generator tube generator tube rupture releases. (Rev A) maintaining level above the tubes for rupture, prior to core damage. scrubbing. Therefore, the intent of this SAMA is met with the current procedures.

129 Vent main steam safety valves in Reduced consequences of a steam NEI 05-01 Basis for Screening: The estimated cost of Excessive containment, generator tube rupture. (Rev A) design reanalysis and implementation of Implementation Cost hardware changes would exceed bounding benefit. Implementationi would also have negative consequences since the increase in containment pressure would result in containment isolation phase B which would empty the RWST.

Westinghouse Non-Proprietary Class 3 Page 101 of 142 Our ref. LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report SNMSAMA

' A.44Ti'ile ,.D Ai< .Yorree Phase I Co*m*,ents ., Dijiposti*n iNumiber~ .*9 130 Add an independent boron Improved availability of boron NEI 05-01 Basis for Screening: WBN has an independent Already Implemented injection system. injection during ATWS. (Rev A) boron injection system with multiple paths.

Therefore this SAMA is not applicable to WBN.

131 Add a system of relief valves to Improved equipment availability NEI 05-01 Basis for Screening: For a plant with Excessive prevent equipment damage from after an ATWS. (Rev A) significant construction already completed, the Implementation Cost pressure spikes during an ATWS. estimated cost of installing a relief valve system is judged to be excessive relative to the risk benefit since ATWS accounts for only 4.06 % of the total internal event CDF.

132 Provide an additional control Improved redundancy and NEI 05-01 Basis for Screening: AMSAC is installed at Already Implemented system for rod insertion (e.g., reduced ATWS frequency. (Rev A) WBN. Therefore the intent of this SAMA is AMSAC). met with the current design.

133 Install an ATWS sized filtered Increased ability to remove NEI 05-01 Basis for Screening: For a plant with Excessive containment vent to remove decay reactor heat from ATWS events. (Rev A) significant construction already completed, the Implementation Cost heat. estimated cost of implementation would exceed the bounding benefit.

134 Revise procedure to bypass MSIV Affords operators more time to NEI 05-01 Basis for Screening: This is a BWR issue. Not Applicable isolation in turbine trip ATWS perform actions. Discharge of a (Rev A) Therefore this SAMA is not applicable to scenarios, substantial fraction of steam to WBN.

the main condenser (i.e., as opposed to into the primary containment) affords the operator more time to perform actions (e.g., SLC injection, lower water level, depressurize RPV) than if the main condenser was unavailable, resulting in lower human error probabilities.

135 Revise procedure to allow Allows immediate control of low NEI 05-01 Basis for Screening: This is a BWR item. Not Applicable override of low pressure core pressure core injection. On failure (Rev A) PWRs do not implement the same logic for injection during an ATWS event, of high pressure core injection governing low pressure injection that is used and condensate, some plants in BWRs. Therefore, this item is not direct reactor depressurization applicable and is screened from further followed by five minutes of consideration.

automatic low pressure core injection.

Westinghouse Non-Proprietary Class 3 Page 102 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report SAM*W4 SA SMIil DwsinSource IP~s eý14iDip~etion

ýNu,,iber, 136 Install motor generator set trip Reduced frequency of core NEI 05-01 Installing a low cost means for tripping the Retain For Phase 1I breakers in control room. damage due to an ATWS. (Rev A) motor generator sets from the control room Analysis may reduce the risk from ATWS. This SAMA will be retained for further evaluation.

137 Provide capability to remove Decreased time required to insert NEI 05-01 Basis for Screening: Implementation of this Very Low Benefit power from the bus powering the control rods if the reactor trip (Rev A) SAMA would require reevaluation of the loss control rods. breakers fail (during a loss of of the loads on the unit boards. Training and feedwater ATWS which has rapid procedure changes is estimated to cost more pressure excursion). than the potential benefit. Therefore this SAMA is considered very low benefit.

138 Improve inspection of rubber Reduced frequency of internal NEI 05-01 Basis for Screening: The Watts Bar design Not Applicable expansion joints on main flooding due to failure of (Rev A) does not include rubber expansion joints on condenser. circulating water system the circulating water. Therefore this SAMA is expansion joints. not applicable to WBN.

Combine with SAMA 222.

139 Modify swing direction of doors Prevents flood propagation. NEI 05-01 Basis for Screening: The flood doors installed Already Implemented separating turbine building (Rev A) on elevation 708' swing so that water pressure basement from areas containing will force them closed. Therefore the intent of safeguards equipment. this SAMA is met with the current design.

140 Increase seismic ruggedness of Increased availability of necessary NEI 05-01 Basis for Screening: No vulnerabilities were Already Implemented plant components. plant equipment during and after (Rev A) identified in the Watts Bar IPEEE.

seismic events. Modifications were made to bring the plant to a 0.3 g screening value per SQUG walkdowns.

Therefore the intent of this SAMA is met with the current design.

141 Provide additional restraints for Increased availability of fire NEI 05-01 Basis for Screening: The seismic margin Not Applicable C02 tanks. protection given a seismic event. (Rev A) review for the IPEEE did not identify this vulnerability at WBN. Therefore this SAMA is not applicable to WBN.

Westinghouse Non-Proprietary Class 3 Page 103 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report SAMA `S MA'ite S.SMA-Diiciusion I Source hýV Cb1fhW' ADisjioiion 142 Replace mercury switches in fire Decreased probability of spurious NEI 05-01 Basis for Screening: The auxiliary building, Already Implemented protection system. fire suppression system actuation. (Rev A) control building, IPS and reactor building fire protection systems do not include mercury switches. Also the auxiliary building and containment building fire protection systems have fusible link operated sprinkler heads, which would require multiple link failures.

Therefore the intent of this SAMA is met with the current design.

143 Upgrade fire compartment Decreased consequences of a fire. NEI 05-01 Basis for Screening: Two and three hour Excessive barriers. (Rev A) regulatory required fire protection barriers are Implementation Cost installed and maintained. Non regulatory required two hour fire barriers are also credited in IPEEE. For a plant with significant construction already completed, the estimated cost of upgrading to 4 hour4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> fire barriers would exceed the bounding benefit.

144 Install additional transfer and Reduced number of spurious NEI 05-01 Basis for Screening: Modifications were Already Implemented isolation switches. actuations during a fire. (Rev A) implemented to mitigate spurious actuations evaluated under Appendix R, which included; installation of thermolag, rerouting conduits, or use of transfer and isolation switches in combination with manual actions. The Appendix R analysis is being reexamined for unit 2 licensing. Therefore the intent of this SAMA is met with the current design.

145 Enhance fire brigade awareness. Decreased consequences of a fire. NEI 05-01 Basis for Screening: Fire protection lesson Already Implemented (Rev A) plans and fire drills are held quarterly, and offsite live fire training is held annually.

Therefore the intent of this SAMA is met with the current training.

146 Enhance control of combustibles Decreased fire frequency and NEI 05-01 Basis for Screening: The transient combustible Already Implemented and ignition sources. consequences. (Rev A) control program and hot work permits control combustibles and ignition sources. Therefore the intent of this SAMA is met with the current procedures.

Westinghouse Non-Proprietary Class 3 Page 104 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report tCuuCCu pIUMUotttLy Ut a tat DaUS* ULt3MIMIng: rut a ptatoI WUl protection system. break LOCA (a leak before (Rev A) significant construction already completed, the Implementation Cost break). estimated cost of implementation would exceed the bounding benefit.

148 Enhance procedures to mitigate Reduced consequences of a large NEI 05-01 Basis for Screening: EOPs follow the current Already Implemented large break LOCA. break LOCA. (Rev A) owners group guidelines. Therefore the intent of this SAMA is met with the current procedures.

149 Install computer aided Improved prevention of core melt NEI 05-01 Basis for Screening: The Integrated Control Already Implemented instrumentation system to assist sequences by making operator (Rev A) System (ICS) is available to operators and the operator in assessing post- actions more reliable. Tech Support Center personnel to assist in accident plant status. assessing post-accident plant status. Therefore the intent of this SAMA is met with the current design.

150 Improve maintenance procedures. Improved prevention of core melt NEI 05-01 Basis for Screening: Maintenance Already Implemented sequences by increasing (Rev A) improvements to increase equipment reliability of important reliability have been implemented via; equipment. Maintenance rule program and MSPI, margin management procedure, AP-913 program identification of critical components.

Therefore the intent of this SAMA is met with the current procedures and maintenance practices.

151 Increase training and operating Improved likelihood of success of NEI 05-01 Basis for Screening: Operating experience is Already Implemented experience feedback to improve operator actions taken in response (Rev A) incorporated in operator training. Feedback operator response. to abnormal conditions, mechanisms are used to keep operators up to date. Therefore the intent of this SAMA is met with the current practice.

Combine with SAMA 263.

152 Develop procedures for Reduced consequences of NEI 05-01 Basis for Screening: An anti barge boom is Very Low Benefit transportation and nearby facility transportation and nearby facility (Rev A) installed at the intake structure to reduce accidents, accidents. transportation accidents. There are no identified hazardous barge shipments near the Watts Bar site. Therefore this SAMA is considered very low benefit.

Westinghouse Non-Proprietary Class 3 Page 105 of 142 Our ref. LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report SA MA A'-1Thic I lSA hkD us.sibn ouc 9'lI PhaseIiCoi~lfm1nts Disposihuiii 153 Install secondary side guard pipes Prevents secondary side NEI 05-01 Basis for Screening: For a plant with Excessive up to the main steam isolation depressurization should a steam (Rev A) significant construction already completed, the Implementation Cost valves, line break occur upstream of the estimated cost of implementation would main steam isolation valves. Also exceed the bounding benefit.

guards against or prevents consequential multiple steam generator tube ruptures. following a main steam line break event.

154 Implement procedure to open the Failure of RCP seal cooling was Cook Basis for Screening: WBN does not have Not Applicable CVCS cross-tie valve to the found to be a significant capability to cross-tie CVCS between units. It opposite unit early in the accident contributor to CDF in the loss of is cost prohibitive to add a cross-tie between response. CCW and loss of ESW events. two ASME class 2 systems. This SAMA is not The initiation of charging flow applicable to the WBN design.

from the opposite unit should provide sufficient RCP seal cooling to prevent RCP seal damage.

155 Implement loss of ESW Potentially reduces CDF due to Cook Basis for Screening: AOI-13 for ERCW Already Implemented procedure changes similar to that RCP seal LOCAs from loss of system loss or rupture, provides directions to of loss of CCW to reduce ESW. stop the RCPs, and cooldown and depressurize significance of RCP seal LOCAs. the RCS by depressurizing the steam generators. Therefore, the intent of this SAMA is met with the current procedures.

156 Eliminate RCP thermal barrier Prevents loss of RCP seal Cook Procedure AOI-7.07 provides direction to Retain For Phase II dependence on CCW, such that integrity after a loss of CCW. connect ERCW to CCS to supply the thermal Analysis loss of CCW does not result Watts Bar Nuclear Plant IPE barrier coolers. AOI-15 for loss of CCS should directly in core damage. identified that an ERCW be revised to refer to AOI-7.07 (may be connection to charging pump minimal cost). This SAMA will be retained seals could be used. for further evaluation.

Combine with SAMA 268.

157 Implement procedure guidance Potentially reduces the frequency Cook Basis for Screening: AOI-13 provides Already Implemented for use of cross-tied CCW or SW of the loss of either of these. guidance on cross-tying ERCW headers and pumps. SOI-70.01 contains instructions for CCS configurations. Therefore, the intent of this SAMA is met with the current design and current procedures.

Westinghouse Non-Proprietary Class 3 Page 106 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment I Final Watts Bar Unit 2 SAMA Report f, IA SAA TId

ý, INMA ~Discu~ssifn ~ Sý;u4ce I Fommýetls haist I!DiPo.sififsn 158 Implement procedure and Potentially improves success rate Cook Basis for Screening: AOIs exist for coping Already Implemented operator training enhancements in of operator actions after support with the loss of support systems, such as a loss support system failure sequences, system failures. of ERCW, CCS, and control air, and are used with emphasis on anticipating extensively in license operator initial training problems and coping. and license operator continuing training programs. Therefore, the intent of this SAMA is met with the current procedures and the associated operator training.

Combine with SAMA 263.

159 Improve ability to cool RHR heat Reduces probability of loss of Cook Basis for Screening: Procedure AOI-7.07 Already Implemented exchangers. decay heat removal. Options contains instructions to install ERCW flood considered include 1) performing mode spool pieces to CCS which would procedure and hardware provide alternate RHR heat exchanger modification to allow manual cooling. Therefore the intent of this SAMA is alignment of fire protection met.

system to the CCW system, or 2) installing a CCW header cross-tie.

Westinghouse Non-Proprietary Class 3 Page 107 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report

~SS+V MA cus~sion NoShrcc ~ ~ d~Uiiieii '~ fDisPosLi0ion des for improved credit to be Cook Basis for Screening: Already Implemented IC L tep oraLryHVAC.

temporary HlVA taken for loss of HVAC 1) The motor-driven AFW pumps, CCS seque nces. pumps, and ERCW pumps do not depend on HVAC systems to be operable. Therefore, the intent of this SAMA is met for these HVAC systems with the current design and current procedures.

2) The TD AFW pump, SI pumps, RHR pumps, and Containment Spray pumps do depend on HVAC systems to be operable.

There is a DC powered fan available for the TD AFW pump room. Installation of temporary HVAC would be dose prohibitive in other rooms during a LOCA.

3) There are compensatory measures and SOI-82 procedures for abnormal operation of the Diesel Generator electric board room ventilation system in the event of an equipment failure to providing alternate ventilation alignments, including use of the adjacent room exhaust fans and cross flow between rooms. Therefore, this item is not further evaluated.

161 Provide backup ventilation for the Provides enhanced ventilation for Cook Basis for Screening: There are compensatory Already Implemented EDG rooms, should their normal EDG rooms. measures and SOI-82 procedures for abnormal HVAC supply fail. operation of the Diesel Generator electric board room ventilation system in the event of an equipment failure for providing alternate ventilation alignments, including use of the adjacent room exhaust fans and cross flow between rooms. Therefore the intent of this SAMA is met with the current design.

Westinghouse Non-Proprietary Class 3 Page 108 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report SMA SI M.A Title SMADsuin, Source ~ PfDjmnAisj3O.-1j :is A~umber 162 Install an independent method of Potentially decreases frequency of Cook Basis for Screening: Sump water goes through Already Implemented suppression pool cooling, loss of containment heat removal. RHR and containment spray heat exchangers, The RHR heat exchanger is cooled by CCS and Containment Spray heat exchanger is cooled by ERCW. Therefore there is independent means of cooling the sump, and the intent of the SAMA is met.

163 Develop an enhanced drywell Provides a redundant source of Cook Basis for Screening: This is a BWR SAMA. Not Applicable spray system. water to containment to control Therefore this SAMA is not applicable to containment pressure, when used WBN.

in conjunction with containment heat removal.

164 Provide a dedicated existing Identical to the previous concept, Cook Basis for Screening: This is a BWR SAMA. Not Applicable drywell spray system. except that one of the existing Therefore this SAMA is not applicable to spray loops would be used instead WBN.

of developing a new spray system.

165 The action to turn on hydrogen Turning on the hydrogen igniters Cook Basis for Screening: Near term procedure Already Implemented igniters fails frequently due to the sooner would reduce containment changes are in progress (PER 121426) to time needed to remotely turn off failure probability for some revise the EOP network to turn on the the ice condenser air handling sequences. hydrogen igniters in E-0. Therefore the intent units, as committed to during the of this SAMA will be met with the revised original installation of the procedures.

hydrogen igniter system. This commitment will be investigated and removed if justifiable.

166 Create a water-cooled rubble bed This rubble bed would contain a Cook Basis for Screening: For a plant with Excessive on the pedestal. molten core dropping onto the significant construction already completed, the Implementation Cost pedestal, and would allow the estimated cost of implementation debris to be cooled. ($18,000,000, representative of similar nuclear power plants) would exceed the bounding benefit.

167 Enhance air return fans (ice Provide an independent power Cook Basis for Screening: 10 CFR 50.44 analysis Very Low Benefit condenser containment), supply for the air return fans, shows these fans are a negligible contribution potentially reducing containment to the containment's ability to handle a failure probability during SBO hydrogen burn. Therefore this SAMA is sequences. considered very low benefit.

Westinghouse Non-Proprietary Class 3 Page 109 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report SAMAV_ Y..4A MA q,4t11,4 SAADirdiussio Source JPhasc Cothmenets Dispo'sW4n 168 Create other options for reactor (a) Use water from dead-ended Cook Basis for Screening: The crane wall is sealed Already Implemented cavity flooding (Part a). volumes, the condensed to elevation 715' or approximately 13 feet blowdown of the RCS, or above the compartment floor. The hot leg is at secondary system by drilling elevation 718' and the maximum water level pathways in the reactor vessel post accident is elevation 721', therefore water support structure to allow will overflow to flood the reactor cavity.

drainage from the SG Therefore the intent of this SAMA is met with compartments, refueling canal, the current design.

sumps, etc., to the reactor cavity.

Also (for ice condensers), allow drainage of water from melted ice into the reactor cavity.

169 Create other options for reactor (b) Flood cavity via systems such Cook Basis for Screening: EPSIL provides direction Already Implemented cavity flooding (Part b). as diesel-driven fire pumps. to connect fire water to the containment spray test connection to fill up containment, which would result in flooding the reactor cavity.

Therefore the intent of this SAMA is met with the current design and procedures.

170 Use firewater spray pump for Provides for redundant Cook Combined with SAMA 169. Combined Containment Spray. Containment Spray method without high cost.

171 Install secondary containment For plants with a secondary Cook Basis for Screening: The ABGTS scrubs Already Implemented filtered ventilation, containment, would filter fission anything from the ABSCE. Therefore the products released from the intent of this SAMA is met with the current primary containment, design.

172 Increase containment design Reduces chance of containment Cook Basis for Screening: For a plant with Excessive pressure. overpressure failures. significant construction already completed, the Implementation Cost cost of implementation caused by reconstruction of the containment building would exceed the bounding benefit.

173 Implement procedure for Reduced SBO frequency. Cook Basis for Screening: WBN has no spare EDG Not Applicable alignment of spare EDG to for use during an SBO. An additional EDG shutdown board after LOSP and would be required before benefiting from this failure of the EDG normally SAMA. Therefore, this item is not applicable supplying it. and is screened from further consideration.

Westinghouse Non-Proprietary Class 3 Page 110 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report LOOK nasis ior 3creenmg: w orK uruers L /u io (batteries I and II) and 819658 (batteries Ill and IV) are in process to replace the existing Unit I batteries since the current batteries are reaching end of life. The cost of replacing batteries I and II is approximately $300,000.

Unit 2 batteries will also be replaced to match the Unit 1 design. Therefore the intent of this SAMA is met.

175 Create a lake water backup for Provides redundant source of Cook Basis for Screening: Supplemental cooling Excessive EDG cooling. EDG cooling, water for condenser circulating water is Implementation Cost available which originates upstream of the dam. Loss of offsite power only contributes 5% to CDF, therefore cross-tying this system to the diesel generator cooling water would exceed the potential risk benefit.

176 Provide a connection to alternate Increases offsite power Cook Two 161 kV lines come into the Watts Bar Retain For Phase II offsite power source. redundancy. switchyard from the nearby hydro plant Analysis switchyard. There are 5 redundant lines into hydro switchyard. Additional lines into the Watts Bar switchyard may exceed the maximum cost, however this SAMA will be retained for further cost-benefit analysis.

A procedure exists for backfeed when a unit is shutdown which requires the main generator links be removed.

177 Replace anchor bolts on EDG oil Millstone found a high seismic Cook Basis for Screening: The seismic margin Not Applicable cooler. SBO risk due to failure of the review for the IPEEE did not identify this EDG oil cooler anchor bolts. For vulnerability at WBN. Therefore this SAMA plants with a similar problem, this is not applicable to WBN.

would reduce seismic risk.

178 Locate RHR inside of Prevents ISLOCA from the RHR Cook Combine with SAMA 115. Combined containment, pathway.

179 Install self-actuating CIVs. For plants that don't have this, it Cook Combine with SAMA 114. Combined potentially reduces the frequency of isolation failure.

Westinghouse Non-Proprietary Class 3 Page 111 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report SAMAf SI MAýTitle SA3 IMA.DScuossibn, Source YPhiase I omiment's Dhsnosilion 180 Install relief valves in the CCW Relieves pressure buildup in Cook Basis for Screening: CCS is designed for RCS Already Implemented system CCW piping caused by an RCP pressure back to the isolation points and thermal barrier tube rupture, differential flow isolates the inlet and outlet, preventing an ISLOCA. and stops the RCP. Therefore, the intent of this SAMA is met with the current design.

181 Provide leak testing of valves in At Kewaunee Nuclear Power Cook Combine with SAMA 113. Combined ISLOCA paths. Plant, four MOVs isolating RHR from the RCS were not leak tested. Potentially reduces ISLOCA frequency.

182 Revise ISLOCA procedure to Potentially reduces ISLOCA Cook Basis for Screening: Procedure ECA- 1.2 for a Already Implemented specifically address the ISLOCA CDF. LOCA outside containment meets current sequence with the frequency that industry guidance. Therefore the intent of this was dominant in the PRA. SAMA is met with the current procedures.

183 Implement internal flood Options considered include 1) use Cook Basis for Screening: The current modeling of Very Low Benefit prevention and mitigation of submersible MOV operators, flooding concerns in the WBN PRA does not enhancements. and 2) back flow prevention in indicate a vulnerability to this item. Therefore drain lines. this SAMA is considered very low benefit.

184 Implement internal flooding Implement improvements to Cook Basis for Screening: The current modeling of Very Low Benefit improvements identified at Fort prevent or mitigate 1) a rupture in flooding concerns in the WBN PRA does not Calhoun Station. the RCP seal cooler of the CCW indicate a vulnerability to this item. Therefore system, 2) an ISLOCA in a this SAMA is considered very low benefit.

shutdown cooling line, and 3) an AFW flood involving the need to possibly remove a watertight door. For a plant where any of these apply, potentially reduces flooding risk.

185 Perform surveillances on manual Improves success probability for Cook Basis for Screening: Procedures exist to Already Implemented valves used for backup AFW providing alternate water supply perform surveillance of the ERCW supply pump suction, to AFW pumps. valves to the AFW pumps. Therefore, the intent of this SAMA is met with the current procedures.

186 Prevent overpressurization of Failure of check valve SI-151W Cook Basis for Screening: This is a Cook specific Not Applicable RHR piping by SI system. fails HPI. A redundant path, issue, and therefore not applicable to WBN.

parallel to the check valve, would improve reliability.

Westinghouse Non-Proprietary Class 3 Page 112 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report Create automatic swapover to Removes fuman error implement low pressure pump to contribution from recirculation HPSI pump piggyback operation failure.

during recirculation following RWST denletion.

188 Implement modifications to the In the IPE, failure of the Cook Basis for Screening: The Unit 2 requirements Already Implemented compressed air system (Unit 1 compressed air system was found for compressed air are being evaluated. The control air compressor) to to be a significant contributor to safety related compressors including dryers, increase the capacity of the CDF. Even though acceptable which are common to both units are being system. event tree modeling modifications replaced to meet the needs of 2 unit operation.

would lower compressed air This modification includes consideration of contributions and virtually eliminating the ERCW dependency. Therefore eliminate this vulnerability, the intent of this SAMA is met with the design evaluate cost-beneficial upgrades modification.

to the capacity of the Unit 1 control air compressor.

189 Provide an additional Improves I&C redundancy and Cook Basis for Screening: AMSAC has already Already Implemented instrumentation system for reduces ATWS frequency. been provided to reduce ATWS frequency at ATWS mitigation (e.g., WBN. Therefore, the intent of this SAMA is AMSAC). met with the current design.

190 Defeat 100 percent load rejection Eliminates the possibility of a Cook Basis for Screening: 100 percent load rejection Not Applicable capability, stuck open PORV after a LOSP, is not part of the WBN design. Therefore this since PORV opening wouldn't be SAMA is not applicable to WBN.

needed.

191 Provide self-cooled ECCS seals. ECCS pump seals are CCW Cook Basis for Screening: The WBN Charging and Excessive cooled. Self-cooled seals would SI pumps have mechanical seals which do not Implementation Cost remove this dependency. require a cooling source. The RHR pump seals are CCS cooled. Providing mechanical seals for RHR pumps would exceed the maximum benefit cost.

192 Separate non-vital buses from Some non-vital loads mixed with Cook Basis for Screening: The Watts Bar vital and Already Implemented vital buses. vital loads on load centers non-vital buses are separated. Therefore this potential cause load shedding SAMA is met by the current design.

difficulties.

193 Make CCW trains separate. Current cross-tie capability Cook Basis for Screening: This SAMA is not Already Implemented creates a potential common mode applicable to WBN. The CCS trains are failure mechanism for both trains separate.

(and both stations).

Westinghouse Non-Proprietary Class 3

• Page 113 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report SA ,A4 +S4AMA Title ~, ~S4AM D~isciaion i 5ource Phaiie ICommentsi<<<v~iD~isto

<Numfber .<,<~

194 Make ICW trains separate. Current cross-tie capability Cook Basis for Screening: The ICW system is not Not Applicable creates a potential common mode applicable to WBN.

failure mechanism for both trains (and both stations).

195 Provide a centrifugal charging Currently charging pumps are Cook Basis for Screening: WBN has two centrifugal Already Implemented pump. positive displacement pumps. charging pumps for each unit that are used for high pressure injection of borated water during emergency conditions requiring actuation of the ECCS. Therefore this SAMA is not applicable to WBN.

196 Provide a motor-operated AFW Provides redundancy for plants Cook Basis for Screening: The AFW system for Already Implemented pump. with only turbine-driven AFW each unit includes two motor-driven AFW pumps. pump trains. Therefore, the intent of this SAMA is met with the current design.

197 Provide containment isolation Potentially enhances containment Cook Basis for Screening: WBN meets the GDC Already Implemented design per GDC and SRP. isolation capability. and SRP. Therefore the intent of this SAMA is met with the current design.

198 Improve RHR sump reliability. Reduces potential for common Cook Basis for Screening: The required GSI- 191 Already Implemented mode failure of RHR due to sump modifications were implemented at unit debris in sump. 1 and will be included in unit 2.

199 Provide auxiliary building Enhances ventilation in auxiliary Cook Basis for Screening: Normal auxiliary Very Low Benefit vent/seal structure, building. building ventilation is not risk significant.

Therefore this SAMA is considered very low benefit.

200 Add charcoal filters on auxiliary Enhances fission product removal Cook Basis for Screening: The ABGTS already Already Implemented building exhaust, after ISLOCA. contains charcoal filters. Therefore the intent of this SAMA is met with the current design.

201 Add penetration valve leakage Enhances capability to Cook Basis for Screening: Temperature indication Already Implemented control system. detect/control leakage from and level detectors exist for the operators to penetration valves, detect and control leakage. Containment penetration valves are tested every outage.

Therefore the intent of this SAMA is met with the current design and operating practices.

202 Enhance screen wash. Reduces potential for loss of ICW Cook Combine with SAMA 47. Combined due to clogging of lake water screens.

Westinghouse Non-Proprietary Class 3 Page 114 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report 203 Enhance training for important The Fussell-Vesely importance Cook Basis for Screening: The PRA identified Already Implemented operator actions, list was reviewed to identify any important operator actions have been significant human errors. Those incorporated into operator training. Therefore with a F-V importance of 5E-03 the intent of this SAMA is met with the or greater are considered for current training program.

training enhancement.

Combine with SAMA 263.

204 Foldout pages are used Potentially reduces CDF related Cook Basis for Screening: Foldout pages are on the Not Applicable inconsistently by Unit to operator error in red path back of every page. WBN is not using them Supervisors. The possibility of sequences. inconsistently. Therefore this SAMA is not revising the usage of the foldout applicable to WBN.

pages will be investigated to see if diagnosis of red path conditions can be improved.

205 A clear definition of the Reduces human error related to Cook Basis for Screening: The only shared system Not Applicable coordination strategy for local cross-tie actions. between the Watts Bar units is the B train of recovery actions (e.g., between CCS. The common control room and single units during cross-tying shift manager minimizes lack of coordination.

operations) could save Therefore this SAMA is not applicable to considerable action time. WBN.

206 Implement operator training on Reduces likelihood of core melt Cook Basis for Screening: EOPs for responding to Already Implemented the impact of primary and into a failed containment, loss-of-coolant accidents and secondary side secondary system heat removal on breaks address operator actions for monitoring containment pressure response and reducing the pressure rise in containment and the possibility of containment as a result of inadequate heat removal from the failure preceding core melt. In containment. These procedures are used addition, consider procedural extensively in license operator initial training upgrades to minimize the and license operator continuing training possibility of such situations programs, and are practiced in the plant arising. simulator. Therefore, the intent of this SAMA is met with the current procedures and the associated operator training.

Westinghouse Non-Proprietary Class 3 Page 115 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report SAMA , SAM,,

V Tiiidt,

• A Discfusin, VSAM , urce* . ' Phaei7CUom I neisi "D

,Number~ ~__________

207 Implement operator training on This training will emphasize Cook Basis for Screening: Sump swapover Already Implemented the importance of a wet reactor injecting the maximum amount of instructions in the EOPs provides instructions cavity on potential fission product water possible from the RWST to to maximize RWST transfer to containment releases. the containment prior to sump and EPSIL and SAMG procedures switchover to recirculation. provides guidance to flood containment if needed. Therefore the intent of this SAMA is met with the current procedures.

208 Add protection to prevent tornado Penetration rooms are tornado Cook Basis for Screening: The Watts Bar design Already Implemented damage to RWST and penetration protected. Tornado category F2 includes a moat around the RWST to retain a rooms, and higher can generate heavy minimum amount of water in case of tank enough missiles that they could damage. The penetration room inside auxiliary impact and damage the RWST. bldg is tornado protected. Therefore this SAMA is not applicable to WBN.

209 Man SSF continuously to align At Oconee Nuclear Station a Cook Basis for Screening: This is an Oconee Not Applicable coolant makeup system for RCP dedicated operator for seals or for Nuclear Station specific item. Therefore, this seal cooling, the highest value operator action item is not applicable and is screened from could be considered. further consideration.

210 Add protection to prevent tornado Consider tornado protection for Cook Basis for Screening: The ERCW system is the Already Implemented damage causing failure of power tanks or switchgear in turbine safety related source for AFW. Switchgear for and upper surge tanks. building. Surge tanks are suction AFW pumps in auxiliary building are tornado source for emergency FW pumps. protected. Therefore this SAMA is not applicable to WBN.

211 Replace reactor vessel with Reduces core damage Cook Basis for Screening: For a plant with Excessive stronger vessel, contribution due to vessel failure. significant construction already completed, the Implementation Cost estimated cost of implementation would exceed the bounding benefit.

212 Improve seismic capacity of walls Failure of these transformers Cook Basis for Screening: The seismic margin Not Applicable near 4160/600 VAC transformers. caused by a seismically induced review for the IPEEE did not identify this failure of the walls contributed vulnerability at WBN. Therefore this SAMA approximately 25% of seismic is not applicable to WBN.

CDF. Reinforcing the walls potentially eliminates this failure mode.

Westinghouse Non-Proprietary Class 3 Page 116 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report improve seismic capacity oi mne 3eismicaiiy inuuceu iauiure ui mne lbdIS 101 OLreeIlllig. IVIUUlIIIuLlUIIb Wulu EDG fuel oil day tanks. EDG fuel oil day tanks incorporated from SQUG walkdowns to bring contributed approximately 20% of the EDG up to 0.3g seismic margin. Therefore seismic CDF. A modification to the intent of this SAMA is met with the prevent seismic impact potentially current design.

eliminates this failure mode.

214 Reinforce the seismic capacity of Seismic failure of the steel Cook Basis for Screening: For a plant with Excessive the steel structure supporting the structure supporting the auxiliary significant construction already completed, the Implementation Cost auxiliary building. building would lead to collapse of estimated cost of implementation to reinforce the building. Reinforcing the the auxiliary building to withstand beyond-building potentially precludes or design-basis earthquake levels would exceed lessens this failure mode. the bounding benefit.

215 Provide a means to ensure RCP Options to consider include using Cook Basis for Screening: Any of these options are Excessive seal cooling so that RCP seal the CVCS cross-tie, installation of considered not cost beneficial. To meet SBO Implementation Cost LOCAs are precluded for SBO a new, independently powered conditions any of these options would require events, pump, or a temporary connection a diesel backed pump. For a plant with to provide cooling to RCP significant construction already completed, the thermal barriers. Such a strategy estimated cost of implementation would would also benefit loss of ESW exceed the bounding benefit.

and loss of CCW events.

216 Improve EDG reliability. Minimizes the probability of a Cook Basis for Screening: The Watts Bar EDG Already Implemented SBO event given a LOSP. reliability meets the maintenance rule with no valid failures. WBN follows all applicable owners group, INPO TR-7-60 and industry recommendations for diesel generator preventive maintenance. PER 124298 evaluated these recommendations for implementation at WBN. Therefore this SAMA is met by the current design and operating practices.

217 Improve circulating water screens Minimizes the chance of clogging Cook Basis for Screening: Duplex screens, trash Already Implemented and debris removal, heat exchangers and condensers racks, and pre-screen improvements have been and initiating transient events, implemented to improve debris removal. The intent of this SAMA is met with the current design.

Westinghouse Non-Proprietary Class 3 Page 117 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report

~SAkA ~ 8,1 fA nile~ 14JA1 Discu'Siiin Source ~ Phias:I COinients Di~posilion4

ýNu"niberJ>IK

- 2 -

218 Improve reliability of power Reduces reactor trip frequency. Cook Basis for Screening: WBN is currently Already Implemented supplies, improving vital AC reliability, upgrading the 500kV switchyard, replacing batteries, separating vital AC for dual unit operation, installing new inverters for Unit 2, incorporating a double breaker scheme in the switchyard, and adding load tap changers.

These reliability improvements met the intent of this SAMA.

219 Improve switchyard and This initiative is to reduce human Cook Basis for Screening: WBN is implementing an Already Implemented transformer reliability, errors in the switchyard and improved double breaker scheme in 500kV alarms on plant transformers. This switchyard, adding load tap changers, and has initiative potentially lowers the eliminated the single point failure of main frequency of transient events transformer due to overpressure trips. These initiated by the electrical system. reliability improvements meet the intent of this SAMA.

220 Reduce biofouling of raw water Improves control of zebra Cook Basis for Screening: WBN treats raw water to Already Implemented systems. mussels. eliminate biofouling. This SAMA is met with current operating practice.

221 Improve reliability of main Potentially reduces transient Cook Basis for Screening: Several reliability Already Implemented feedwater pumps. initiating event frequency. improvements have been made to the main feedwater system. A design change is in process to upgrade to digital feedwater control. The main feedwater pump shaft material was upgraded. Changes were incorporated based on recommendations in availability improvement bulletins. Bentley Nevada supervisory instrumentation was installed. WBN has an extensive oil analysis program, and single point failures have been eliminated. Therefore the intent of this SAMA is met with the current design and operating practices.

222 Establish a preventive Potentially reduces flooding Cook Basis for Screening: There is a limited use of Very Low Benefit maintenance program for initiating event frequency and the expansion joints at Watts Bar and no expansion joints, bellows, and failure probability of plant indication of a vulnerability. Therefore this boots. components. SAMA is considered very low benefit.

Westinghouse Non-Proprietary Class 3 Page 118 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report SAMA IAW TiiIe -,0 SAMA1 sDii~.n . o-vPhaseTCuimnis ~ *-Ds6~to 223 Improve reliability of AFW Potentially reduces occurrence of Cook Basis for Screening: Several reliability Already Implemented pumps and valves, loss of secondary heat sink. improvements have been made to the AFW system. The AFW pumps and valves will be monitored in accordance with the maintenance rule and MSPI. There are predictive maintenance oil analysis and vibration programs. The Unit 1 EGM controller capacitor was changed out, and Unit 2 will incorporate the same design based on obsolete components. Additional changes on Unit 1 which will be incorporated into Unit 2 include governor stem changes, a new positioner and I to P converter, and short stroke LCVs to gain design margin for closure for SG isolation.

Unit 2 design is evaluating the Unit 1 corrective actions to identify additional reliability improvements. Therefore the intent of this SAMA is met with the current design and operating practice.

224 Eliminate MSIV vulnerabilities. Reduces the chance that MSIVs Cook Basis for Screening: Design changes to the Already Implemented will drift off their open seat valves and air supplies, and maintenance during low-power operations. improvements have been made on the unit I MSIVs, which will be duplicated on unit 2.

Therefore the intent of this SAMA is met with the current design.

225 Upgrade main turbine controls. Potentially reduces turbine trip Cook Basis for Screening: Since the turbine trip Very Low Benefit frequency. initiator contributes less than 1% CDF, the estimated cost of implementation would exceed the minimal risk benefit from this SAMA. Therefore this SAMA is considered very low benefit. Therefore this SAMA is considered very low benefit.

Westinghouse Non-Proprietary Class 3 Page 119 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report SA~i~ ?jSAAite,~.SAADiscu.svidn 'Sýie Phiase I6omments ~D""p stod 226 Permanent, self- powered pump This SAMA provides a means of Vogtle Basis for Screening:,The cost of this Excessive to backup normal charging pump. limiting the size of a seal LOCA. enhancement has been estimated at other Implementation Cost This SAMA would provide a self nuclear plants to be $2.7M based on a powered pump that can be conceptual design of the backup pump which automatically or rapidly aligned exceeds the bounding benefit.

to the RCP seals from the MCR.

Long term secondary side cooling can be provided through the operation of the turbine driven AFW pump using existing Vogtle procedures. This arrangement would make it possible to provide adequate core cooling in extended SBO evolutions.

227 Maintain full- time black start The combustion turbines (CTs) in Vogtle Basis for Screening: This SAMA is not Not Applicable capability of the Wilson the Plant Wilson Switchyard have applicable to WBN since a combustion turbine Switchyard combustion turbines black start diesel generators, but not available to the WBN site.

these are only verified to be operable prior to extended EDG An agreement exists with the nearby hydro AOTs. The use of the black start plant to provide power if needed per procedure diesels would be necessary to TRO-TO-SOP-10.134.

start the CTs given unavailability of offsite power at Plant Wilson.

This SAMA would add surveillance or maintenance activities to ensure the combustion turbines would be available much more often than is currently credited in the PRA model.

228 Provide enhanced structural This SAMA would provide Vogtle Basis for Screening: This SAMA is not Not Applicable protection of Plant Wilson enhanced structural protection of applicable to WBN since a combustion turbine Switchyard. Plant Wilson Switchyard such not available to the WBN site.

that it would be more likely to survive in severe weather and extreme weather events.

Westinghouse Non-Proprietary Class 3 Page 120 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report upposite u I ne current YtKA numan rsasis ror screening: ine capaoiity exists via capability. reliability assessment for this the AOI-43.01 procedure series to cross-tie action is that the cross-tie action diesel generators between units and trains.

will not succeed (i.e., HEP failure Therefore this SAMA is met with the current probability = 1.0) until at least design and procedures.

seven hours after event initiation.

Providing the ability to perform a timely 4kV AC cross-tie using an available emergency diesel generator under emergency conditions would allow operators more flexibility to operate required equipment to protect the core.

230 Permanent, dedicated generator Installation of a dedicated Vogtle Basis for Screening: An alternate power Already Implemented for one motor driven AFW pump generator for continued operation source is available capable of supplying power and a battery charger. and control of a MD AFW pump to an AFW pump and there is a spare battery would reduce the overall capable of supplying DC control power.

contribution to CDF risk. This Procedure MA-1 provides direction for generator would need to have the connecting the alternate power source.

capacity to operate a MD AFW Therefore the intent of this SAMA is met with pump and an associated battery the current design and procedures.

charger required for DC power control of the AFW pump.

231 Add bypass line around cooling Failure of the Loop CT return Vogtle Basis for Screening: This is a Vogtle specific Not Applicable tower return valves, valves results in failure of cooling feature that does not apply to WBN.

water to one of the EDGs and other systems. A bypass line around the 1668A (Loop "A")

and 1669A (Loop "B") valves that could be remotely or manually opened given failure of the existing valves could greatly reduce the CDF risk from this failure mode.

Westinghouse Non-Proprietary Class 3 Page 121 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report SAMAI S,IU-1Till" SAMIA Disus~ion S1oiircec Phase I Commneiits Dispositionl 232 Implement enhanced RCP seal For Vogtle, a dominant Vogtle Combine with SAMA 58. Combined design. contributor to the current risk profile is that without RCP seal cooling, it is assumed (based on Westinghouse and NRC consensus modeling) that an RCP seal LOCA of sufficient magnitude to require RCS injection occurs within 13 minutes. This SAMA would implement enhanced RCP seal designs that virtually eliminate this failure mode.

233 Implement alternate AC power The implementation of an Vogtle Basis for Screening: The cost of installing an Excessive source. alternate AC power source would additional EDG has been estimated to be Implementation Cost most likely take the form of an greater than $20 million in the Calvert Cliffs additional EDG. This SAMA Application for License Renewal. It was would help mitigate LOSP events similarly estimated to be about $26.09M for and would reduce the risk during both units at Vogtle. As the per unit cost of time frames of on-line EDG approximately $IOM to $13M is greater than maintenance. The benefit would the Watts Bar maximum benefit, it has been be increased if the additional DG screened from further analysis.

could 1) be substituted for any current diesel that is in maintenance, and 2) if the diesel was of a diverse design such that CCF dependence was minimized.

234 Implement automatic initiation of The implementation of an Vogtle Basis for Screening: The WBN design initiates Very Low Benefit HPI on low RCS level (after AC automatic HPI initiation system HPSI on low RCS pressure which would power recovery), would reduce the potential for result from an RCP seal LOCA. The PRA core damage from occurring model does not include operator actions to following events where ac power restore the pumps after AC power recovery is recovered, but where a seal since this sequence is dominated by recovery LOCA has already occurred. In of AC power sources. Including this operator these cases, RCS level must be action would result in limited risk benefit and restored to avoid core damage therefore is not analyzed further.

from occurring.

Westinghouse Non-Proprietary Class 3 Page 122 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report sAMA 50 vuAMA rTii" 1.1AAM&i"*hsn ' urePhu~se I Coinmenits. ~ i ipsi Number______________

235 Additional training and/or Enhanced training and/or Vogtle Basis for Screening: Revision 2 of E-2 limits Already Implemented procedural enhancement to procedure enhancements could the heat up of the RCS by adjusting intact SG implement timely RCS reduce the potential for thermally PORVs to hottest RCS hot leg temperature depressurization. induced steam generator tube and enhances the time to get to SI termination.

ruptures, thereby reducing the Therefore the intent of this SAMA is met with overall Level 2 risk contribution. the current procedures.

236 Use the hydrostatic test pump as For Vogtle, a dominant Vogtle Combine with SAMA 57. Not Applicable an alternate means of providing contributor to the current risk seal injection, profile is that without RCP seal cooling, it is assumed (based on Westinghouse and NRC consensus modeling) that an RCP seal LOCA of sufficient magnitude to require RCS injection occurs within 13 minutes. This SAMA would implement enhanced RCP seal designs that virtually eliminate this failure mode.

237 Ensure all ISLOCA releases are SAMA would scrub all ISLOCA Vogtle Combine with SAMA 116. Combined scrubbed. releases. One example is to plug all drains in the break areas so that the break location would quickly be covered with water.

238 Completely automate swap over SAMA would ensure that Vogtle Combine with SAMA 32. Combined to recirculation on RWST automatic swap over to depletion. recirculation would occur in cases where high pressure injection from the charging and SI pumps is required (compared to the current capability at Vogtle that only automates the swap over for LPI).

239 Install additional instrumentation SAMA would provide additional Vogtle Combine with SAMA 111. Combined for ISLOCA detection, confidence that detection and response to ISLOCAs could be implemented to reduce the risk from these types of events.

Westinghouse Non-Proprietary Class 3 Page 123 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report SAISA MA Tife' 7 I SAMADisculigi~n. SU~c > i~hi I Comn~itns ~ ~ ~ ipsfoI 240 Install permanent dedicated SAMA provides a means of Vogtle Basis for Screening: an alternate power Already Implemented generator for normal charging limiting the size of a seal LOCA. source capable of supplying a charging pump pump. The NCP can be automatically or exists. Therefore the intent of this SAMA is rapidly aligned to the RCP seals met with the current design.

from the MCR. This is an alternative approach to SAMA Combine with SAMA 242.

226 that provided for a backup NCP, but with similar impacts.

Long term secondary side cooling can be provided through the operation of the turbine driven AFW pump using existing Vogtle procedures. This arrangement would make it possible to provide adequate core cooling in extended SBO evolutions.

.241 Enhance procedures for ISLOCA SAMA would provide additional Vogtle Basis for Screening: ECA-1.2 for a LOCA Already Implemented response. confidence that the response to outside containment meets current industry ISLOCAs could be implemented guidance. Therefore the intent of this SAMA to reduce the risk from these types is met with the current procedures.

of events.

242 Permanent, Dedicated Generator This SAMA provides a means of Wolf Creek Basis for Screening: This requires a dedicated Excessive for the NCP with Local Operation limiting the size of a seal LOCA DG with auto start capability and auto transfer Implementation Cost of TD AFW after 125V Battery and providing primary side to meet the 13 minute criteria to prevent seal Depletion. makeup through the installation of LOCA. Additionally the DG and Charging a diesel generator that can be Pump lube oil cooling and seal cooling would rapidly aligned to the NCP from require CCS and ERCW. The estimated cost the MCR. Long term secondary of implementation of a dedicated DG would side cooling can be provided exceed the bounding benefit.

through the operation of the turbine driven AFW pump using existing Wolf Creek procedures.

This arrangement would make it possible to provide adequate core cooling in extended SBO evolutions.

Westinghouse Non-Proprietary Class 3 Page 124 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report

~SAMS, S.1MA Title ýVIAýMAD.issii4n ,;Aie Phas I(Cbvsinhe k.%'D~jin 243 Modify the Controls and An off-site diesel generating plant Wolf Creek The WBN site has an agreement with the Already Implemented Operating Procedures for Sharpe (Sharpe Station) has an agreement nearby Hydro plant to supply power when Station to allow for Rapid with Wolf Creek to provide needed per procedure TRO-TO-SOP- 10.134.

Response. power to the site in the event that This facility has black-start capability and the Wolf Creek experiences a Station procedure gives the highest priority to the Blackout. While the ten 2MW. TVA nuclear units.

diesel generators have the capacity to power the emergency loads, the time to align power to WCGS is long and is not expected to be complete before 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> after the onset of degraded AC conditions. Providing the WCGS control room with the ability to start and align these generators to the WCGS emergency buses through the switchyard would be a means of restoring power to WCGS in non-weather related LOOP events.

244 AC Cross-tie Capability. Providing the ability to perform a Wolf Creek Combined with SAMA 229. Combined timely 4kV AC cross-tie under emergency conditions would allow operators more flexibility to operate required equipment to protect the core.

Westinghouse Non-Proprietary Class 3 Page 125 of 142 Our ref. LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report

<S61A " ......... r'*SAMAD6scusstoný'

5 ýIor'n P 245 ISLOCA Isolation. The current Wolf Creek PSA Wolf Creek Basis for Screening: This is a Wolf Creek Not Applicable model does not credit operator specific SAMA. There are no known issues actions to isolate ISLOCAs using with the WBN valves. Therefore this SAMA available MOVs as it has not been is not applicable to WBN.

confirmed that those valves can isolate with RCS pressure against them. The plant engineering staff estimates that the motors could move the valves to a partially closed position before exceeding the torque limit of the valve operator. From that point, it would be possible to complete the valve closure locally assuming that the valves are accessible.

Ensuring that procedures direct this isolation in ISLOCA events is a potential means of addressing some of the ISLOCA scenarios (those where access is possible).

Alternatively, the valves could be replaced with a type that can close against RCS pressure.

246 Open Doors for Alternate DG For cases when DGHVAC fails Wolf Creek Basis for Screening: EPSIL contains Already Implemented Room Cooling. and inside air temperatures are. instructions for opening the EDG building high, the EDG room doors could room doors to provide cooling. Therefore the .

be opened to provide outside air intent of this SAMA is met with the current exchange cooling to the EDG procedures.

rooms.

Westinghouse Non-Proprietary Class 3 Page 126 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report Manual Kecirculatlon wlm Kw This SAMA is specifically related Wolf Creek Basis for Screening: The Watts Bar design Already Implemented Level Instrumentation Failure. to the failure of auto swap to includes the capability for manual recirculation mode due to the recirculation and the current EOPs require RWST level instrumentation. operators to monitor RWST and containment Because this instrumentation is sump level. Level indication would require responsible for both the auto swap multiple failures to fail recirculation initiation.

signal and the annunciator that Therefore, the intent of this SAMA is met would alert the operator that with the current design and procedures.

recirculation mode is required, the main cue that would instigate operator action is not available.

While other means of identifying the need for manual swap are available, the PSA model currently assumes that manual alignment of recirculation always fails in these scenarios because the low RWST level signal has failed. If reasonable credit is taken for the operators to use other means to diagnose the need to align recirculation mode, the importance of the level instrumentation failure is greatly reduced.

Westinghouse Non-Proprietary Class 3 Page 127 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report s' V s 1- FrMAi/,e J ýý.AMDiscu~ssio~n Source Phase icICommnt DisýPosifiorvl 248 Manual Recirculation with Auto Failure to auto swap to Wolf Creek Combine with SAMA 31. Combined Initiation Failure. recirculation mode can be caused by failure of the logic components responsible for governing the swap, by power failure to the logic, or other hardware failures.

For the majority of these cases, a cue would be available to alert the operators of the need to swap to recirculation mode; however, no credit is currently taken for manual swap to recirculation mode after auto initiation failure due to modeling complexities. If reasonable credit is taken for the operators to align recirculation mode, the importance of the scenarios including automatic swap failure is greatly reduced.

249 High Volume Makeup to the For SGTR, and ISLOCA Wolf Creek Basis for Screening; Procedure EPSIL Already Implemented RWST. scenarios where the RWST will contains guidance for refilling RWST with fire be depleted and HPI fails or the water and boric acid. Therefore the intent of sump will be unavailable for this SAMA is met with the current procedures.

recirculation mode, the addition of water to the RWST will allow for continued core cooling. A hard piped connection to the FPS is a possible means of providing this capability. '

250 Additional Instrumentation in the Early detection of a SGTR may Wolf Creek Combine with SAMA 124. Combined SG to Measure Radioactivity. increase the probability of successful isolation and mitigation.

Westinghouse Non-Proprietary Class 3 Page 128 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report NSAMIAV ~ SAMA T~it ý'$AMA ýDh~ils.%ion Sore hsI&i,nis 2;mwvip'4~

251 Additional Training on SGTR Enhanced training on detection Wolf Creek Basis for Screening: The WBN operators are Already Implemented Accidents. and mitigation of SGTR scenarios currently trained on SGTR scenarios in both may improve operator response. classroom and simulator exercises. The instruction program is continually reviewed and improved, as required. While it may be possible to further improve the SGTR training program, the results of such changes would be difficult to measure using current HRA methods.

252 SG Tube Inspection, Improved maintenance on the SG Wolf Creek Combine with SAMAs 119 & 120. Combined Replacement. tubes may reduce the frequency of tube ruptures.

253 Install SG Isolation Valves on the Installation of primary side Wolf Creek Basis for Screening: For a plant with Excessive Primary Loop Side. isolation valves provides an significant construction already completed, the Implementation Cost additional means of isolating and estimated cost of implementation would controlling an SGTR event. These exceed the bounding benefit.

valves would also eliminate the need for local action to complete a steam generator isolation after a tube rupture has occurred.

254 Alternate Fuel Oil Tank with EDG failures related to failure of Wolf Creek Basis for Screening: Failure of the fuel oil Very Low Benefit Gravity Feed Capability. the fuel oil transfer pumps are transfer pumps contributes only 2% the currently considered to be internal event CDF based on RRW review.

unrecoverable in the PSA model. Improvements in the fuel oil transfer system The installation of a large volume are judged to be a minimal risk benefit.

tank at an elevation greater than the EDG fuel oil day tanks would The cost of this enhancement has been allow for emergency refill of the estimated to be $150,000 by Wolf Creek.

day tanks in the event of fuel oil transfer pump failure.

Westinghouse Non-Proprietary Class 3 Page 129 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report IsAMOAile Wk ;AA1Afim-jsion V Souricei Ph~ssecI(7ment'is, Ai 255 Permanent, Dedicated Generator This is similar to SAMA 242, but Wolf Creek Basis for Screening: This requires a dedicated Excessive for the NCP, one Motor Driven addresses the additional scenarios DG with auto start capability and auto transfer Implementation Cost AFW Pump, and a Battery in which the TD AFW pump is to meet the 13 minute criteria to prevent seal Charger. unavailable. Increasing the LOCA. Additionally the DG and Charging capacity of the diesel generator Pump lube oil cooling and seal cooling would would be required to carry the require CCS and ERCW. The estimated cost additional load of the AFW pump of implementation of a dedicated DG would and a battery charger for long exceed the bounding benefit.

term SBO success. Fire Protection is not suggested as an alternate source of SG makeup given that it is a low pressure system and would not be available early in an accident.

256 Install Fire Barriers Around Equipment fires have the potential Wolf Creek Basis for Screening: The Appendix R program Retain For Phase II Cables or Reroute the Cables to damage safety systems that are rerouted permanent cables and conduits as Analysis Away from Fire Sources. not directly related to the original necessary, however procedure enhancements equipment fires. If cables required for control of temporary cable impacts on fire for safety system operation are protection will be reviewed. This SAMA is located above ignition sources or retained for further evaluation.

equipment to which fires may propagate, all associated safety systems depending on those cables may fail. Protecting the overhead cables or rerouting them away from equipment could reduce the consequences of fires in these areas.

Westinghouse Non-Proprietary Class 3 Page 130 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report 257 Inter-Train CCW Cross-tie for A cross-tie between the CCW Wolf Creek Basis for Screening: WBN has the capability Already Implemented Emergency Operation. loops could increase the to cross-tie CCS trains, therefore this SAMA availability of CCW flow to is met.

cooling loads. Certain failure combinations that disable CCW could be eliminated if the use of a cross-tie valve was available to provide flow to required loads.

For example, if the "A" loop CCW heat exchanger is out of service and the "B" loop of CCW has failed, the "A" loop of CCW could be used to cool the "B" loop CCW heat exchanger pending isolation of unused loads. For Wolf Creek, an entire cross-tie line with isolation valves would have to be installed, as there is no existing cross-tie.

258 Install DC Cross-tie Capability. This SAMA would improve DC Wolf Creek Combined with SAMA 5. Combined capability/flexibility in accident conditions.

259 Revise AOI-15 "Loss of Revise AOI 15, "Loss of IPE Basis for Screening: AOI-15 has been Already Implemented Component Cooling Water". Component Cooling Water," to modified to incorporate this item.

facilitate stopping the RCPs on loss of CCS train A to minimize the potential for RCP seal damage due to pump bearing failure.

260 Improve training on loss of CCS. In the event of a total loss of IPE Combine with SAMA 51. Combined CCS, clearer guidance on the desirability of cooling down the RCS prior to a seal LOCA developing to minimize the potential for seal damage should be considered. In general, additional training on the loss of CCS initiator is suggested.

Westinghouse Non-Proprietary Class 3 Page 131 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report NAMA <. 'SAJMA TI~-e ý AMA 9DiscusYsitn ~ Source ~ PhacI C, mh~eitsz7Dssio 261 Guidance to align the C-S diesel In the event of a loss of offsite IPE Basis for Screening: The cost to refurbish, Excessive generator. power followed by the failure of complete and license the spare 5th DG was Implementation Cost both shutdown boards on one estimated at -2 to 3 million in 1996. The unit, the procedures would be estimated cost exceeds the bounding benefit.

enhanced by adding the guidance to align the C-S diesel generator (i.e., the fifth diesel generator) to one of the shutdown buses not powered in the accident sequence due to the loss of a normally aligned diesel generator. This alignment could be accommodated by including a reference to the spare diesel generator in AOL 35, "Loss of Offsite Power."

262 Provide connections for A potential improvement that IPE Basis for Screening: The potential Very Low Benefit centrifugal charging pumps to the could be evaluated is a plant improvement was evaluated and there is low ERCW system. change to provide connections for benefit to aligning a second charging pump to both centrifugal charging pumps, ERCW.

on both units, to the ERCW system for lube oil cooling in the event of a loss of CCS cooling to the associated pump. Currently, this capability is only available for centrifugal charging pump A on Unit 1.

263 Enhance operator training and Enhancements to the operator IPE See SAMA 151. Combined procedures. training and procedures for responding to failures of support systems could potentially be beneficial, with emphasis on anticipating problems and coping.

Westinghouse Non-Proprietary Class 3 Page 132 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report Sourcel PhasýeI CometM , nk, DsjWiio Evaluate CCS. Ventilation has been IPE Basis for Screening: This SAMA was Already Implemented requirements. conservatively modeled in this implemented in the current PRA model.

study. Area ventilation is Therefore this SAMA was implemented.

provided to the motor driven AFW pumps and the CCS pumps from multiple systems serving the plant elevation where these pumps are located. Beyond design basis concurrent failures of the available Unit I ventilation is assumed to impact the long term availability of the AFW and CCS.

An evaluation of the CCS/AFW area cooling requirements could be performed which could reduce this interdependence by crediting natural convection and availability of other coolers at this plant elevation.

265 Revise procedures to shed CCS In the event of a loss of ERCW, IPE Combine with SAMA 53. Combined loads prior to CCS heatup. which would eventually lead to a loss of CCS cooling, additional guidance on the relationship of CCS to ERCW and the desirability of eliminating CCS loads to extend the time of suitable CCS temperatures is a potential consideration for evaluation. This could be accomplished by revising AOI 13, "Loss of ERCW," to alert the operators to shed CCS loads prior to CCS heatup.

Westinghouse Non-Proprietary Class 3 Page 133 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report 266 Provide remote-local operation of During a loss of all AC, the steam IPE Basis for Screening: Local manual operation Already Implemented steam generator PORVs. generator power-operated relief in the high temperature area is no longer valves (PORV) are to be locally required due to installation of nitrogen bottles.

operated to depressurize the steam Therefore the intent of this SAMA is met with generators, thereby cooling down the current design.

the RCS. The addition of provisions for remote operation of these valves could potentially be beneficial due to the high area temperatures that may be encountered.

267 Increase charging pump lube oil In the event of a loss of CCS IPE Combine with SAMA 54. Combined capacity. cooling to the charging pumps, the time available for operation of the pumps would be limited by the loss of lube oil heat exchanger cooling. To extend the time available to protect the pumps, consideration could be given to increasing the oil capacity.

268 Eliminate RCP thermal barrier Losses of RCP seal cooling could IPE Combine with SAMA 156. Combined cooling dependence on potentially be reduced if the RCP component cooling water. thermal barrier cooling dependence on component cooling water, which is required for the charging pumps that provide RCP seal injection, could be eliminated.

269 Provide 2 trains of cooling to the Currently, ventilation for the IPE Basis for Screening: This SAMA was Already Implemented 480V board room. 480V board room that contains implemented by a modification to provide spot the unit vital inverters is provided cooling by the alternate train to the area where by one train of ventilation. The the inverters are located. Therefore this current models rely substantially SAMA was implemented.

on recovery actions by the operators. Consideration could be given to providing two trains.

Westinghouse Non-Proprietary Class 3 Page 134 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report SI4Aije SMA ~ AMADisusionSiirde~.Phaise I Conmmens ADispos~iof 270 Delay containment spray From a severe accident point of IPE Basis for Screening: The current Watts Bar Excessive operation relative to phase B view, one potential change, for design basis calculations require sprays to Implementation Cost conditions. consideration, would be the initiate at containment phase B conditions.

delaying of spray operations This SAMA would require reanalysis of relative to the Phase B condition. Safety analysis, therefore it is considered cost Currently, containment sprays prohibitive.

actuate immediately in response to a Phase B condition, and air return fans (ARF) actuate after a 10 minute delay. This is currently a requirement of the design basis LOCA where switchover to containment spray recirculation occurs prior to ice melt; thereby limiting pressure increases below containment design pressure.

Modular Accident Analysis Program analyses of representative core damage sequences indicate that actuation of the containment sprays while ice remains in the ice condenser has little impact on severe accident containment performance and may be detrimental in that operation of the sprays rapidly depletes the inventory of the RWST, making its contents unavailable for vessel injection. Since many scenarios have successful injection but failure at recirculation, the rapid depletion .of the RWST due to spray operation accelerates the time to core damage. Therefore, an evaluation balancing the severe accident versus design basis requirements could be made.

Westinghouse Non-Proprietary Class 3 Page 135 of 142 Our ref. LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report Returbish the ERCW pumps & Improves the relial dsasis tor screening: unit 2 will oe upgrade the capacity of the ERCW pumps. refurbishing and upgrading ERCW pumps as current numns. reauired. This SAMA is met.

272 Provide a portable diesel powered Improves availability of ERCW RRW Review Basis for Screening: This SAMA has been Already Implemented 5,000 gpm pump as a backup to for SBO. implemented.

the ERCW system.

273 Provide a redundant path for Eliminates single failure potential RRW Review Check valve 62-504 is a single failure point Retain For Phase II ECCS suction from the RWST of RWST check valve failure to for ECCS injection and contributes 7% to Analysis around check valve 62-504. open. CDF. The cost of a design change, new hardware and analysis may exceed the risk reduction benefit, however this SAMA will be retained for further analysis.

274 Replace CCS pumps with positive Improves reliability of CCS RRW Review Basis for Screening: For a plant with Excessive displacement pumps. system. significant construction already completed, the Implementation Cost estimated cost of implementation would exceed the bounding benefit.

275 Provide a new inverter Improved reliability of AC power RRW Review Basis for Screening: A design change is in Already Implemented arrangement. system. process to install new inverters. Therefore the intent of this SAMA is met.

276 Provide an auto start signal for Improved reliability of AFW for RRW Review Incorporation of an AFW auto start signal on Retain For Phase II AFW on loss of Standby low power events (<18%) before loss of the Standby Feedwater pump is under Analysis Feedwater pump. Main Feedwater pumps are review. This is a low power / shutdown issue started. which is not quantitatively addressed in the current PRA risk model. This SAMA is retained for further evaluation.

277 Replace shutdown board chillers. Improved reliability of shutdown RRW Review Basis for Screening: The potential Very Low Benefit board HVAC. improvement was evaluated by reviewing the risk reduction worth (RRW) of the 6.9 kV board room ventilation and ventilation recovery. There is low benefit to these ventilation systems.

278 Perform analysis to evaluate the Eliminate dependency RRW Review Basis for Screening: Analysis evaluating the Already Implemented need for ventilation to inverters, requirement for HVAC. need for ventilation has been performed for shutdown boards and ESFAS. Unit 1 and will be updated for Unit 2.

Therefore this SAMA is met.

279 Provide a permanent tie-in to the Improve availability of air RRW Review The final disposition of the construction air Retain For Phase It construction air compressor. system. compressor is under evaluation. This SAMA Analysis is retained for further cost-benefit evaluation.

Westinghouse Non-Proprietary Class 3 Page 136 of 142 Our ref. LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report

<SAMA~ S' Il .IITiic S. 01MAsDW~sion Source Phiase I Comments~ ~sisto Number 280 Add new Unit 2 air compressor Improve availability of air RRW Review The final disposition of installing a Retain For Phase II similar to the Unit 1 D system. compressor similar to the Unit 1 D compressor Analysis compressor. is under evaluation. This SAMA is retained for further cost-benefit evaluation.

281 Replace the ACAS compressors Improve reliability of air system. RRW Review Basis for Screening: A design change to Already Implemented and dryers. replace the ACAS compressors and dryers is in progress. Therefore the intent of this SAMA is met.

282 Provide cross-tie to Unit 1 Extend RWST capacity. RRW Review Basis for Screening: For a plant with Excessive RWST. significant construction already completed, the Implementation Cost estimated cost of implementation to cross-tie the RWSTs would exceed the bounding benefit. Implementation would require analysis of technical specification implications for the opposite unit.

283 Enhance procedures for feed & Improve mitigation of loss of RRW Review Basis for Screening: Procedure FR-H 1 is Already Implemented bleed operation. secondary cooling. written to owners group guidelines and EOPs are continually updated as ERG maintenance items are issued. Operator actions required for bleed and feed operation are included on regular basis in operator requal training. The intent of this SAMA is met with the current procedures and operator training program.

Westinghouse Non-Proprietary Class 3 Page 137 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report Table 17 Phase II Analysis Results S A.Ib.

1 .. ".. s. 1 . . . . . i .Eif.... ...... ....... .. .. .. t .. .... . . .

4 Improve DC bus load shedding. $83,399 $ 31,675 2.6 Potentially Cost Beneficial 8 Increase training on response to loss of $ 21,469 $ 26,773 0.8 Not Cost two 120V AC buses which causes Beneficial inadvertent actuation signals.

32 Add the ability to automatically align $ 530,264 $ 2,100,000 0.3 Not Cost emergency core cooling system to Beneficial recirculation mode upon refueling water storage tank depletion.

45 Enhance procedural guidance for use of $ 89,003 $ 31,675 2.8 Potentially Cost cross-tied component cooling or service Beneficial water pumps.

46 Add a service water pump. $ 102,000 $1,042,511 0.1 Not Cost Beneficial 56 Install an independent reactor coolant $ 675,053 $ 2,400,000 0.3 Not Cost pump seal injection system, without Beneficial dedicated diesel.

70 Install accumulators for turbine-driven $1,945 $ 256,204 -0 Not Cost auxiliary feedwater pump flow control Beneficial valves.

71 Install a new condensate storage tank $0 $1,706,586 0 Not Cost (auxiliary feedwater storage tank). Beneficial 87 Replace service and instrument air $121,460 $ 886,205 0.1 Not Cost compressors with more reliable Beneficial compressors which have self-contained air cooling by shaft driven fans.

112 Add redundant and diverse limit switches $4,565 $ 691,524 -0 Not Cost to each containment isolation valve. Beneficial 136 Install motor generator set trip breakers $7,397 $241,795 -0 Not Cost in control room. Beneficial 156 Eliminate RCP thermal barrier $675,053 $31,675 21.3 Potentially Cost dependence on CCW, such that loss of Beneficial CCW does not result directly in core damage.

176 Provide a connection to alternate offsite $ 42,247 $9,126,460 -0 Not Cost power source. Beneficial 256 Install Fire Barriers Around Cables or $426,340 $ 19,608 21.7 Potentially Cost Reroute the Cables Away from Fire Beneficial Sources.

273 Provide a redundant path for ECCS $ 87,379 $ 439,945 0.2 Not Cost suction from the RWST around check Beneficial valve 62-504.

276 Provide an auto start signal for AFW on $5,926 $ 615,605 -0 Not Cost loss of Standby Feedwater pump. Beneficial 279 Provide a permanent tie-in to the $121,460 $ 909,893 0.1 Not Cost construction air compressor. Beneficial 280 Add new Unit 2 air compressor similar to $121,460 $ 814,546 0.1 Not Cost the Unit 1D compressor. Beneficial

Westinghouse Non-Proprietary Class 3 Page 138 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report Table 18 RDR Sensitivity Results N unthe RatioRatio 3"%

4 Improve DC bus load shedding. 2.6 4.7 No 8 Increase training on response to loss of 0.8 1.4 Yes two 120V AC buses which causes inadvertent actuation signals.

32 Add the ability to automatically align 0.3 0.5 No emergency core cooling system to recirculation mode upon refueling water storage tank depletion.

45 Enhance procedural guidance for use of 2.8 5.1 No cross-tied component cooling or service water pumps.

46 Add a service water pump. 0.1 0.2 No 56 Install an independent reactor coolant 0.3 0.5 No pump seal injection system, without dedicated diesel.

70 Install accumulators for turbine-driven -0 -0 No auxiliary feedwater pump flow control valves.

71 Install a new condensate storage tank -0 -0 No (auxiliary feedwater storage tank).

87 Replace service and instrument air 0.1 0.2 No compressors with more reliable compressors which have self-contained air cooling by shaft driven fans.

112 Add redundant and diverse limit switches -0 -0 No to each containment isolation valve.

136 Install motor generator set trip breakers -0 0.1 No in control room.

156 Eliminate RCP thermal barrier 21 39 No dependence on CCW, such that loss of CCW does not result directly in core damage.

176 Provide a connection to alternate offsite -0 -0 No power source.

256 Install Fire Barriers Around Cables or 22 39 No Reroute the Cables Away from Fire Sources.

273 Provide a redundant path for ECCS 0.2 0.4 No suction from the RWST around check valve 62-504.

276 Provide an auto start signal for AFW on -0 -0 No loss of Standby Feedwater pump.

279 Provide a permanent tie-in to the 0.1 0.2 No construction air compressor.

280 Add new Unit 2 air compressor similar to 0.1 0.2 No the Unit 1 D compressor.

Westinghouse Non-Proprietary Class 3 Page 139 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report Table 19 CDF/LERF Sensitivity Results

/ ?t, SAMA SAMA Behefli/costn BTiclet- o St Ownl,/ it Ratio Rat 4 Improve DC bus load shedding. 2.6 7.4 No 8 Increase training on response to loss of 0.8 2.2 Yes two 120V AC buses which causes inadvertent actuation signals.

32 Add the ability to automatically align 0.3 0.7 No emergency core cooling system to recirculation mode upon refueling water storage tank depletion.

45 Enhance procedural guidance for use of 2.8 7.9 No cross-tied component cooling or service water pumps.

46 Add a service water pump. 0.1 0.3 No

'56 Install an independent reactor coolant 0.3 0.8 No pump seal injection system, without dedicated diesel.

70 Install accumulators for turbine-driven -0 -0 No auxiliary feedwater pump flow control valves.

71 Install a new condensate storage tank -0 -0 No (auxiliary feedwater storage tank).

87 Replace service and instrument air 0.1 0.4 No compressors with more reliable compressors which have self-contained air cooling by shaft driven fans.

112 Add redundant and diverse limit switches -0 -0 No to each containment isolation valve.

136 Install motor generator set trip breakers -0 0.1 No in control room.

156 Eliminate RCP thermal barrier 21 60 No dependence on CCW, such that loss of CCW does not result directly in core damage.

176 Provide a connection to alternate offsite -0 -0 No power source.

256 Install Fire Barriers Around Cables or 22 61 No Reroute the Cables Away from Fire Sources.

273 Provide a redundant path for ECCS 0.2 0.6 No suction from the RWST around check valve 62-504.

276 Provide an auto start signal for AFW on -0 -0 No loss of Standby Feedwater pump.

279 Provide a permanent tie-in to the 0.1 0.4 No construction air compressor.

280 Add new Unit 2 air compressor similar to 0.1 0.4 No the Unit ID compressor.

Westinghouse Non-Proprietary Class 3 Page 140 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report Table 20 Evacuation Speed Sensitivity Results "N*. ,*" *, *'**.Ratio Raio , ,Con MRatio .usion 4 Improve DC bus load shedding. 2.6 2.6 2.6 No 8 Increase training on response to loss of 0.8 0.8 0.8 No two 120V AC buses which causes inadvertent actuation signals.

32 Add the ability to automatically align 0.3 0.3 0.3 No emergency core cooling system to recirculation mode upon refueling water storage tank depletion.

45 Enhance procedural guidance for use of 2.8 2.8 2.8 No cross-tied component cooling or service water pumps.

46 Add a service water pump. 0.1 0.1 0.1 No 56 Install an independent reactor coolant 0.3 0.3 0.3 No pump seal injection system, without dedicated diesel.

70 Install accumulators for turbine-driven -0 -0 -0 No auxiliary feedwater pump flow control valves.

71 Install a new condensate storage tank -0 -0 -0 No (auxiliary feedwater storage tank).

87 Replace service and instrument air 0.1 0.1 0.1 No compressors with more reliable compressors which have self-contained air cooling by shaft driven fans.

112 Add redundant and diverse limit switches -0 -0 -0 No to each containment isolation valve.

136 Install motor generator set trip breakers -0 -0 -0 No in control room.

156 Eliminate RCP thermal barrier 21 21 21 No dependence on CCW, such that loss of CCW does not result directly in core damage.

176 Provide a connection to alternate offsite -0 -0 -0 No power source.

256 Install Fire Barriers Around Cables or 22 22 22 No Reroute the Cables Away from Fire Sources.

273 Provide a redundant path for ECCS 0.2 0.2 0.2 No suction from the RWST around check valve 62-504.

276 Provide an auto start signal for AFW on -0 -0 -0 No loss of Standby Feedwater pump.

279 Provide a permanent tie-in to the 0.1 0.1 0.1 No construction air compressor.

280 Add new Unit 2 air compressor similar to 0.1 0.1 0.1 No the Unit I D compressor.

Westinghouse Non-Proprietary Class 3 Page 141 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report 11 REFERENCES AEP 2003 D.C. Cook, Units 1 & 2, Application for Renewed Operating Licenses, Appendix E, Environmental Report Appendices D - F. (ML033070190)

DUKE 2001 Applicant's Environmental Report Operating License Renewal Stage Catawba Nuclear Station Attachment H Severe Accident Mitigation Alternatives (SAMAs) Analysis May 2001 Final Report DUKE 2001 a Applicant's Environmental Report Operating License Renewal Stage McGuire Nuclear Station Attachment K Severe Accident Mitigation Alternatives (SAMAs) Analysis May 2001 Final Report NEI 2005 Nuclear Energy Institute (NEI) Severe Accident Mitigation Alternatives (SAMA) Analysis Guidance Document Revision A. NEI 05-01. November 2005.

NRC 1989 U.S. Nuclear Regulatory Commission (NRC).NUREG 1150 "Severe Accident Risks: An assessment for Five U. S. Nuclear Power Plants," June 1989 NRC 1990 U.S. Nuclear Regulatory Commission (NRC). MELCOR Accident Consequence Code System (MA CCS) - Model Description,NUREG/CR-469 1, Vol. 2, Division of Systems Research, February 1990.

NRC 1996 Generic Environmental Impact Statement for License Renewal of Nuclear Plants (GEIS)

NRC 1997 U.S. Nuclear Regulatory Commission (NRC). 1997. Regulatory Analysis Technical Evaluation Handbook. NUREG/BR-0 184.

NRC 1998 U.S. Nuclear Regulatory Commission (NRC). Code Manualfor MACCS2, NUREG/CR-6613, Vol. 1, prepared by D. Channing and M. L. Young, Division of System Research, May 1998 NRC 2007 WinMACCS, a MACCS2 Interface for Calculating Health and Economic Consequences from Accidental Release of Radioactive Materials into the Atmosphere. User's Guide and Reference Manual WinMACCS Version 3, July 2007 SAIC 2007 Science Applications International Corporation. "Watts Bar Nuclear Plant Severe Reactor Accident Analysis", May 30, 2007 SNC 2007 Vogtle, Units 1 and 2, License Renewal Application (ML071840360, ML071840357)

TVA 1992 "Watts Bar Nuclear Plant Unit 1 PRA Individual Plant Exam."

(ML080090324)

Westinghouse Non-Proprietary Class 3 Page 142 of 142 Our ref: LTR-RAM-I-08-062 Rev 3 January 21, 2009 Attachment 1 Final Watts Bar Unit 2 SAMA Report TVA 1998 "Watts Bar Nuclear Plant (WBNP) Individual Plant Evaluation of External Events (IPEEE) Final Report TVA 2006 Tennessee Multi-Jurisdictional Radiological Emergency Response Plan for the Watts Bar Nuclear Plant, Annex H WCNOC 2006 "Wolf Creek Generating Station, Applicant's Environmental Report; Operating License Renewal Stage," September 2006, (ML062770305)

Enclosure 2 WBN Unit 2 Listing of Open Actions Required for Licensing

1. Prior to fuel load, evaluate the potential for procedural enhancements in the Station Blackout procedures to shed additional loads to extend battery life (SAMA 4).

2. Prior to fuel load, provide procedural enhancements in AOI-15 to cross-tie Component Cooling Water (CCS) trains and Emergency Raw Cooling Water (ERCW) trains (SAMA 45).

3. Prior to fuel load, provide procedural enhancements in AO1-15 for loss of CCS to connect ERCW to supply the thermal barrier coolers (SAMA 156).

4. Prior to fuel load, provide procedural enhancements for the procedure controlling temporary alterations to reduce fire risk from temporary cables (SAMA 256).

5. Prior to fuel load, expand operator training to include response to loss of two 120-V AC buses (SAMA 8).