ML20095J232

Rev 0 to Design Alternatives for Sys 80+ Nuclear Power Plant

ML20095J232
Person / Time
Site:	05200002
Issue date:	04/30/1992
From:	ABB COMBUSTION ENGINEERING NUCLEAR FUEL (FORMERLY, ASEA BROWN BOVERI, INC.
To:
Shared Package
ML20095J230	List: LD-92-056, Forwards Rev 0 to Design Alternatives for Sys 80+ Nuclear Power Plant, Per Commitment in ML20095J232
References
NUDOCS 9205010196
Download: ML20095J232 (44)

Text

- ,

DESIGN ALTERNATIVES FOR THE l

SYSTEM 80+ NUCLEAR POWER PLANT l

APRIL 30, 1992 REV. 00 COMBUSTION ENGINEERING-NUCLEAR, INC. i WINDS 0R, CONNECTICUT ABB ASEA BROWN BOVEnl AO O2 A PDR

TABLE OF CONTENTS Section Title Pace

1.0 INTRODUCTION

4 2.0

SUMMARY

AND CONCLUSION 5 3.0 METHODOLOGY 7 3.1 RISK REDUCTION 7 3.2 COST ESTIMATES 8 3.3 COST BENEFIT COMPARIS0N 8 4.0 PRA RELEASE CLASSES 11 4.1 RELEASE CLASS RC7.1 12 4.2 RELEASE CLASS RC6.2 12 4.3- RELEASE CLASS RC6.4 13 4.4 RELEASE CLASS RC5.1 13 4.5 RELEASE CLASS RC5.2 14 4.6 RELEASE CLASS RC4.1 14 4.7 RELEASE CLASS RC4.2 14 4.8 RELEASE CLASS RC3.1 15 4.9 RELEASE CLASS RC2.2 15 4.10 RELEASE CLASS RC2.4 16 4.11 RELEASE CLASS RCl.4 16 4.12 RELEASE CLASS RC1.7 17 5.0 DESaGN ALTELNATIVES 19 5.1 CONT.AINMENT SPRAY 20 5.2 FILTERED VENT 20 l

5.3 DC BATTERIES AND EFWS 21 5.4 RCP SEAL COOLING 21 5.5 PRESSURIZER AUXILIARY SPRAY 22 5.6 ATWS PRESSURE RELIEF VALVES 22 5.7 CONCRETE COMPOSITION 23 5.8 REACTOR VESSEL EXTERIOR COOLING 23 5.9 H2 IGNITORS 24 5.10 HIGH PRESSURE SAFETY INJECTION 24 5.11 RCS DEPRESSURIZATION 25

6.0 REFERENCES

33 APPENDIX A: FAULT TREE ANALYSIS FOR TOTAL' LOSS OF COMPONENT COOLING WATER. 34 l

Page 1 of 35

l LIST OF TABLES 1.able Iitle Pace 2-1

SUMMARY

OF THE RISK REDUCTIONS OF THE DESIGN ALTERNATIVES 6 3-1 ECONOMIC ASSUMPTIONS FOR LEVELIZED CAPITAL COST RATE 10 4-1 RELEAst PARAMETER DATA FOR SYSTEM 80+ RELEASE CLASSES 18 5-1 FREQUENCY AND PERSON-REM ET.POSURE FOR RELEASE CLASSES 26 5-2

SUMMARY

DESCRIPTION OF RELEASE CLASSES 27 5-3 RISK REDUCTION EVALUATION FOR CONTAINMENT SPRAY 28 5-4 RISK REDUCTION EVALUATION FOR FILTERED VENT 28 5-5 RISK REDUCTION EVALUATION FOR DC BATTERIES & EFWS 29 5-6 RISK REDUCTION EVALUATION FOR PRESSURIZER AUXILIARY SPRAY 29 5-7 RISK REDUCTION EVALUATION FOR ATWS VALVES 30 5-8 RISK REDUCTION EVALUATION FOR CONCRETE COMPOSITION 30 5-9 RISK REDUCTION EVALUATION FOR VESSEL EXTERIOR COOLING 31 l

l 5-10 RISK REDUCTION EVALUATION FOR H2 IGNITORS 31 5-11 RISK REDUCTION EVALUATION FOR HPSI S2 5-12 RISK REDUCTION EVALUATION FOR RCS DEPRESSURIZATION 32 l

l 1

l Page 2 of 35

~. - - -- . .- - - - -

i ,

ACRONYMS ADV Atmospheric Dump Valves ALWR Advanced Light Water Reactor ATWS Anticipated Transient Without Scram BWR Boiling Water Reactor cal Calories CCW Component Cooling Water CET Containment Event Tree CRAC Calculations of Reactor Accident Consequences DC Direct Current DCH Direct Containment Heating EFWS Emergency feedwater System H2 Hydrogen HPSI High Pressure Safety Injection Hrs Hours HVAC Heating, Ventilation, and Air Conditioning IRWST In-Containment Refueling Water Storage Tank KAG Key Assumptions and Groundrules LOCA Loss of Coolant Accident M Meters MAAP Modular Accident Analysis Program MORV Motor Operated Relief Valve NRC Nuclear Regulatury Commission PDS Plant Damage State PRA Probabilistic Risk Assessment RC Release Class RCP Reactor Coolant Pump RCS Reactor Coolant System RHR Residual Heat Removal i SCS Shutdown Cooling System sec Second i SGTR Steam Generator Tube Rupture

SIT Safety Injection Tanks l y year 1

l l

Page 3 of 35

i ,

1.0 INTRODUCTION

The U.S. Nuclear Regulatory Commission's policy related to severe accidents requires, in part, that an application for a design approval comply with the requirements of 10CFR50.34(f). Item (f)(1)(1) requires " performance of a plant site specific [PRA] the aim of which is to seek improvements in the reliability of core and containment heat removal systems as significant and practical and do not impact excessively on the plant." Reference 1 provides the base PRA of the System 00+ plant.

The NRC also requested the ALWk participants to evaluate design alternatives that help mitigate the consequences of severe accidents. To address these requirements and requests, a review of potential modifications to the System 80+

design, beyond those included in the Probabilistic Risk Assessment (PRA), was conducted to evaluate whether potential severe accident mitigation design features could be justified on the basis of cost per person-rem averted.

This report summarizes the results of C-E's review and evaluation of eleven Design Alternatives that were considered in the System 80+ design. Improvements have been reviewed against conservative estimates of risk reductions based on the PRA and minimum order of magnitude costs, to determine what modifications are potentially attractive.

Page 4 of 35 l

l

2.0

SUMMARY

AND C0!LCLUSION The System 80+ design is an Evolutionary Advanced Light Water Reactor design with improved design features to reduce the risk of core damage and mitigate the l

consequences if core damage should occur. The design process was integrated with the PRA to ensure that the risk was very low and distributed over all of the safety related systems (i.e., no single system carries a disproportional responsibility for plant safety). The design insured that no single accident sequence dominated the plant risk and the lessons learned W previous PRAs were addressed.

l Eleven design alternatives were evaluated. These were selected based on the i

Design Alternatives evaluated for the Limerick plant2 and the results from the System 80+ PRA performed by C-E. The Design Alternative analysis used a bounding technique. It was assumed that each Design Alternative worked perfectly and completely eliminated the accident sequences that the Design Alternative was to address. This approach maximizes the benefits associated with each Design Alternative. The benefits were the reduction in risk in terms of whole body person-rems per year received by the total population around the ALWR site.

Using $1,000 per person- rem, and a levelized capital cost rate of 17.9%, this risk reduction was converted to a maximum capital benefit that was compared with capital costs.

l Table 2-1 summarizes the results of the Design Alternative analysis. The first

, column, is the percent of the total person-rem / year reduction for each design alternative. The next column, labeled capital benefit, is an equivalent present worth of the annual dose reduction. It is also the maximum amount that could be spent in capital to be cost beneficial. The third column is a rough capital cost estimate for the design alternatives. The net benefit (capital benefit - capital cost) is given in the last column.

l The System 80+ plant was designed to meet the stringent design goals in the EPRI l

Advanced Light Water Reactor (ALWR) Utility Requirements Document. The System 80+ design has a core damage frequency approximately two orders of magnitude-lower than existing plants. Therefore, the benefits of improving the existing design are significantly lower than predicted for the Limerick Plant 2. The analysis presented in this report overestimated the benefits of the Design Alternatives by assuming that they would work perfectly to eliminate the type of accident they are designed to address. Because of the small initial risk associated wi+'i the System 80+ design, none of the Design Alternatives are cost bene fici al .

Page 5 of 35

TABLE 2-1

SUMMARY

OF THE RISK REOUCTIONS OF THE DESIGN ALTERNATIVES Design' Alternative PERSON-REM CAPITAL CAPITAL NET CAPITAL RE00CTION BENE?IT* COST BENEFIT 1 CONTAINMENT SPRAY 90% $27,600- $1,500,000 -$1,472,400 2 FILTERED VENT 85% $26,300 $10,000,000 -$9,973,700  ;

3 DC BATTERIES & EFWS 69% $21,100 $2,000,000 -$1.,978,900 4 RCP SEAL COOLING 16.5% $6,034 $100,000 -$93,966

. 5 PRESSURIZER AUX 1LIARY SPRAY 6.7% $2,050 $5,000,000- -$4,997,950 6 ATWS VALVES 4.2% $1,290 $1,000,000 -$998,710 7 CONCRETE COMPOSITION. 2.5% $765 $5,000,000 -54,999,244 s

8 REACTOR VESSEL EXTERIOR C00tTNG 2.5% $765 $5,500,000 -$5,499,244 9 H2 IGNITORS 0.1% $31 $1,000,000 -$999,969 "

10 HIGH PRESSURE SAFETY INJECTION 0.004% 0 $20,000,000 -$20,000,000 11 RCS DEPRESSURIZATION 0.002% 0 $500,000 -$500,0a0 THE MAXIMUM CAPITAL COST' ASSUMES NO MAINTENANCE OR TESTING COSTS FOR THE ADDITIONAL EQUIPMENT Page 6 of 35

1 l

3.0 EJHODOLOGY The Design Alternative evaluation followed the format and procedure used by the NRC in evaluating Design Alternatives for Limerick 2. The Design Alternatives ,

were evaluated in terms of cost benefit where the cost of the additional equipment is compared with the savings in terms of a reduced exposure risk to the

general population. The savings, in person-rems per year, were converted to I

dollars using $1,000 per person-rem. The risk of the base System 80+ design is described in the System 80+ Standard Design PRA'.

l 3.1 RISK REDUCTION Risk (person-rem / year) in this analysis is the product of the frequency of core damage for each type of accident (events /y) times the consequence of the accident (person-rem / event). The total risk is the sum of the risks from all the types of accidents. For each Design Alternative, the reduction in total risk is the difference between the risk of the base System 80+ design and the risk with the Design Alternative added.

Risk is defined as the product of frequency and consequence. The frequency of core damage for various accident sequences are calculated. These sequences are then grouped (" binned") into releases classes depending on the timing of the accident and the conditions of the core, vessel, containment, and release l characteristics for the sequence. Each Design Alternative is evaluated in terms l of how it might affect each release class. For this analysis it is assumed that each Design Alternative is perfect: that is, if installed it completely eliminates all failures associated with the _ systems for which it is designed to be an alternative or addition. This implies that each Design Alternative is also tied to perfect support systems. This is a conservative upper limit approach since it overestimates the benefits associated with any design addition. If a Design Alternative is cost beneficial using this screening approach, then a more detailed analysis could be performed.

The Design Alternatives can be divided into two groups. One group prevents core damge and the other group protects the containment or reduces the releases. For -

the Design Alternatives that prevent core damage, the frequency of affected release classes was decreased to zero and the risk reduction was calculated. For example, an ideal pressurizer auxiliary spray Design Alternative is assumed -to l eliminate all core melt risk of a Steam Generator Tube Ruptere (SGTR) by always getting the plant depressurized and into shutdown cooling. Therefore tha frequency of release classes associated with SGTR was reduced to zero and a risk reduction was calculated.

Some Design Alternatives protect the containment or reduce the amount of radioactive materi ' that is released in an accident. These Alternatives reduce the consequence of he accident and therefore 6

reduce the risk (risk = frequency x consequence). Using the CRAC2 Code , the consequence in terms of dose to the general population is calculated for the ALWR site. This_ site conservatively Page 7 of 35

represents the characteristics of most potential sites'. For Design Alternatives that prevent containment failure, the releases were changed from that of a failed containment to that of an intact containment and the risk was reevaluated. For Alternatives that filter the releases, an appropriate consequence and risk reduction was calculated.

3.2 COST ESTIMATES in order to evaluate the effectivenest of the design alternatives, the benefits were compared to the costs of the Alternatives. The rough order of magnitude costs were estimated for each modification. These costs represent the incremental costs that would be incurred in incorporating the alternative in a new plant. The cost estimate for each of the modifications is given in Section 4 where the modification is. discussed.

The cost estimates were intentionally. biased on the low side, but all known or reasonably expected costs were accounted for in order that a reasonable assessment of _ the minimum cost would be obtained. Actual plant costs are. l expected to be higher than indicated in this evaluation. All costs are i referenced to 1992 U.S. dollars.

3.3 COSi BENEFIT COMPARIS0N As described in Section 3.1, the benefit of a design alternative is risk ,

reductio., which was evaluated in terms of reduced exposure of the general  :

population (in units of person-rem /y). The cost of additional equipment is in '

dollars, a one-time initial capital cost. To compare these two numbers, a common  !

measure must be used. In this analysis, the risk reduction was converted to a single capital benefit which can be directly compared with the capital cost.

The benefits of a particular modification were defined as the risk reduction to the general public. Offsite factors evaluated were limited to health effects to the general public based on total exposure (in person-rem). Conristent with the standard used by the NRC to evaluate radiological impacts, health effects costs were evaluated based on a value of $1,000 per offsite person-rem averted due to the design modification. This factor converts person-rem /y to $/y.

The annual benefit in $/y is converted to an single capital benefit using a levelized capital cost rate. Using the method and values described in Ref. 7, and the economic assumptions given in Table 3-1, a levelized capital cost rate of 17.9% is estimated. The Design Alternative results are not very sensitive to the detailed economic assumptions used in calculating a levelized capital cost rate.

Page 8 cf 35

- _ ~ . _ ._ _ . _ _ . _ . _ _ _ _ . _ . . . _ __ . _ . _ _ _ . _ . _ _ . . . . . . _ _ . . _ _ - . _

• n The offsite costs for other items, such as relocation of local residents, elimination of land use and decontamination of contaminated land are not considered. Economic losses, replacement power costs and direct accident costs incurred by the plant owner also are not considered in this evaluation. _ Again, this gives a conservative estimate of-the net benefit.

The analysis presented here conservatively neglects any annual costs associated with the operation of the Design Alternatives. These Alternatives would have to be tested and maintained at regular intervals. Regular training would also be required. In a more detailed analysis, such costs would be converted to an annual cost and be used to reduce the annual benefits.

i Page 9 of 35 ,

i 0

. - - - - . - - . . - . - . - _ .-. . . _ . - . . - ~ . . - .

3 4

TABLE 3-1 ECONOMIC ASSUMPTIONS FOR LEVELIZED CAPITAL COST RATE ASSUMPTIONS val.UE EONO (DEPT) INTEREST RATE, % 10.48 DIPT FRACTION 0.55 RETURN ON EQUITY, % 12.48 INCOME TAX RATE, % 50.0 RATE OF INFLATION, % 4.0 ANNUAL PROPERTY TAX + INSURANCE, % 2.0 TAX DEPRECIATION LIFE, YRS. 20.0 COMPONENT ECONOMIC LIFE, YRS. 24.0 RESULTING LEVELIZED CAPITAL COST RATE, % 17.9 Page 10-of 35 i

l 4.0 PRA RELEASE CLASSES In assessing the risk reduction of each Design Alternative, the potential for each Design Alternative to reduce the frequency of occurrence or the consequence of each release class (RC) is assessed. To do this, an understanding of each RC is required.

In the System 80+ PRA', the containment event analysis describes the possible accident pathways in a containment event tree (CET). This CET was developed so that each end point of an accident sequence uniquely specified the mode of containment failure and the status of the various phenomena.which have the potential to affect the source term characte:istics. Therefore, each of the accident end points is a distinct release class. - A release class (RC) can be fully characterized by the following parameters:

A) its frequency of occurrence, B) the isotopic content and magnitude of the release, C) the energy of the release, D) the time of the release, E) the dur ation of the release, and F) the location of the release.

The RC frequency is determined directly from the cumulative frequency for its respective containment event tree end point. The location of the release was assigned as follows:

1) For overpressure containment failure RCs, the release was assumed to occur at the top of the containment building. This is at an elevation of 52.8 meters above grade.

2) For containment bypass RCs initiated by an interfacing systems LOCA and for containment melt-through RCs, the release from containment occurs in the region of the auxiliary building located below the containment sphere.

The actual release to the environment occurs at grade level.

3) For all other RCs, the releases are assumed to occur at grade level.

MAAP* analyses were used to determine the isotopic content and magnitude of the source term and the time of the release. In general, releases were calculated for a period of 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> from the time of containment failure or from the time of vessel failure for containment bypass and containment isolation failure RCs.

The MAAP analyses are based on specific accident- sequences. To select the appropriate accident sequence for a specific RC, the following process was used:

1) The Plant Damage State (PDS) with the largest contribution to the RC's total frequency was determined.

2) The dominant plant-accident sequence was used for the RC. This defined the initiating event and the status of the various plant systems.

Page 11 of 35 i

3) The Containment Event Tree (CET) and the plant accident sequence definition were reviewed to determine if any special phenomenological conditions had to be specified.  !

4) A containment fallere pressure, failure time or failure condition was specified based on the RC definition.

The following sections present a brief description for each release class with a frequency greater than or equal to 1.0E-10. The numbering system and order of the release classes is developed in the PRA and represents the approximate order they appear in the CET. The data for each release class is summarized in Table 4-1.

l 4.1 RELEASE CLASS RC7.1  ;

l Release class RC7.1 covers the releases from an intact containment. These releases are attributable to design basis leakage from containment. For this analysis, a design basis leakage of 0.34 volume percent per day was used which is consistent with that used in the Chapter 15 safety analysis of CESAR-CD. The-containment leakage was adjusted to account for containment pressure variations throughout the accident. The cumulative frequency for this release class is-1.69E-6 per year.

This release class is characterized by a station blackout with battery depletion followed by late recovery of AC power and containment heat removal. The accident is initiated by a loss of offsite power with failure of the diesel generators and the alternate AC power source. The batteries supply control power for 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> and emergency feedwater is supplied by the turbine-driven emergency feedwater pumps. With battery depletion at 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br />, emergency feedwater is assumed to be lost. Cavity flood is initiated just prior to battery depletion. Vessel failure is assumed to occur at about 15.5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br />. It was assumed that the hydrogen ignitors function to control hydrogen concentrations. At 48 hours5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br />, power was recovered and containment heat removal was reestablished, thus effectively terminating the accident sequence. The releases were assumed to occur at ground level. The characteristics of this release class are summarized in Table 4-1.

i l

4.2 RELEASE CLASS RC6.2 Release class RC6.2 covers releases associated with a containment melt-through into the nuclear annex building with vaporization releases and no radioisotope scrubbing in the containment atmosphere prior to release. The cumulative release

, frequency for this release class is 1.38E-8 per year.

This release class is characterized by a large cold leg LOCA with failure of safety injection to initiate. Containment :, pray and containment heat removal are not available, and the cavity is not flooded. The hydrogen ignitors are assumd to be available for hydrogen control. Vessel failure is assumed to occur at 1.83 hours9.606481e-4 days <br />0.0231 hours <br />1.372354e-4 weeks <br />3.15815e-5 months <br /> and the corium drops into a dry cavity. By 330 hours0.00382 days <br />0.0917 hours <br />5.456349e-4 weeks <br />1.25565e-4 months <br />, the corium is assumed to have ablated through the lower subsphere region of the reactor

• building and into the shadow area of the nuclear annex building. The releases Page 12 of 35

+

b

,_y y-, , - , - _ , - - ,

pressure. No deposition of radioactive material ta tM oucler annex bl ildir.g is credited. The final release from the nuclear rex building is assumed to occur at ground level. The characteristica o r t> b release class are summarizedin Table 4-1.

4.3 RELEASE CLASS RC6.4 Release class RC6.4 cavers releases associMoo sitt a mlt-through of the subsphere of the reactor building into the exlear annex bu ild ng with vaporization and revaporization releases and no radioisotope scrutbing in the containment atmosphere prior to release. The coauiatim release fregeency for this release class is 7.28E-9 per year.

This release cli.ss is characterized by a station blact;out with tuttery depletion.

The accident is initiated by a loss of offsite power with failure of the diese!

generators and the alternate AC power source. The batteries supply control power for 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> and emergency feedwater is supplied oy the turbine-driven mergency feedwater pumps. With battery depletion at 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br />, emergency feeduter is assumed to be lost. The cavity is not flooded prior to battery depletion. The hydrogen ignitors function to control the hydrogen concentration. Vessel failure is assumed to occur at about 19.0 hours0 days <br />0 hours <br />0 weeks <br />0 months <br />. Just prior to vessel failure, a hot leg failure occurs so that the corium is ejected at low pressure into a dry cavity.

By 180 hours0.00208 days <br />0.05 hours <br />2.97619e-4 weeks <br />6.849e-5 months <br />, the corium ablates through the lower subsphere of the reactor buildin3 and into the shadow area of the nuclear annex building lccated below the reactor building. The releases are discharged through the nuclear annex building, initially at a relatively high pressure. No deposition of radioactive material in the nuclear annex building is credited. The final release from the nuclear annex building occurs at ground level. The characteristics of this release class are surmarized in Table 4-1.

4.4 RELEASE CLASS RC5.1 Release class RC5.1 covers releases associated with a late containment overpressure failure with no vaporization or revaporization releases and no radioisotope scrubbing in the containment atmosphere prior to release. The cumulative release frequency for this release class is 3.63E-8 per year.

This release class is characterized by a station blackout with battery depletion.

The accident is initiated by a loss of offsite power with failure of the diesel generaters and the alternate AC power source. The batteries supply control power for 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> and emergency feedwater is supplied by the turbine-driven emergency feedwater pumps. With battery depletion at 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br />, emergency feedwater is assumed to be lost. Cavity flood is initiated just prior to battery depletion.

The hydrogen ignitors function to control the hydrogen concentration. Vessel failure is assumed to occur at about 15.8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br />. Neither power nor containment heat removal are recovered. The containment fails due to overpressure at 128 ,

hours. The containment failure, and hence, the releases are assumed to occur at the top of the containment at an clevati n of 52.8 meters. The characteristics of this release class are summarized in Table 4-1.

[ Page 13 of 35 l

, 4.5 RELEASE CLASS RC5.2 Release class RC5.2 covers releases associated with a late containment overpressure failure with revaporization releases, but no vaporization releases and no radioisotope scrubbing in the containment atmosphere prior to release.

The cumulative release frequency for this release cla s is 6.59E-8 per year.This-release class is characterized by a station blackout with battery depletion. The accident is initiated by a loss of offsite power with failure of the diesel generator.; and the alternative AC power source. The batteries supply control power for 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> and emergency feedwater is supplied by the turbine-driven emergency feedwater pumps. With battery depletion at 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br />, emergency feedwater is assumed to be lost. Cavity flood is initiated just prior to battery depletion. The hydrogen ignitors function to control the hydrogen concentration.

Vessel failure is assumed to occur at about 18.6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br />. Just prior to vessel failure, a hot leg failure occurs so that the corium is ejected at low pressure.

Neither power nor containment heat removal are recovered. The containment fails due to overpressure at 34 hours3.935185e-4 days <br />0.00944 hours <br />5.621693e-5 weeks <br />1.2937e-5 months <br />. The containment failure, and hence, the releases are assumed to occur at the top of the containment at an elevation of 52.8 meters. The characteristics of this release class are summarized in Table 4-1.

4.6 RELEASE CLASS RC4.1 Release class RC4.1 covers releases associated with an early containment steam overpressure failure with no vaporization or revaporization releaser but with radioisotope scrubbing from the containment- atmosphere prior to release. The cumulative release frequency for this release class is 2.12E-10 per year.

This release class was characterized by a large cold leg LOCA with failure of safety injection from time zero. Containment spray and containment heat removal are available and the cavity is flooded. The hydrogen ignitors are assumed to be available for hydrogen control. Vessel failure occurred at 1.8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> and the corium dropped into a wet cavity. The containment is forced to fail at a pressure of about 25 psi at 5.6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br />. (Note: the highest containment pressure during this event was about 40 psia immediately after the initiation of the LOCA.

The containent was not failed at this point because there was no core damage and the vessel was still intact.) The containment failure, and hence, the releases-are assuaed to occur at the top of the centainment at an elevation of 52.8 meters. The characteristics of this release class are summarized in Table 4-1.

4.7 RELEASE CLASS RC4.2 Release class RC4.2 covers releases associated with an early containment overpressure failure due to direct containment heating (DCH) with no radioisotope scrubbing in the containment atmosphere prior to release. The cumulative release frequency for this release class is 2.17E-10 per year.

Page 14 of 35

This release class is characterized by a station blackout with battery depletion.

The accident is initiated by a loss of offsite power with failure of the diesel generators and the alternative AC power source. The batteries supply control power for 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> and emergency feedwater is supplied by the turbine-driven i emergency feedwater pumps. With battery depletion at 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br />, emergency ,

feedwater is assumed to be lost. The cavity is flooded just prior to battery I depletion. The hydrogen ignitors are not functioning to control the hydrogen concentration. Vessel failure is assumed to occur at about 15.8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> and the ,

corium is ejected into the cavity at high pressure. A simultaneous hydrogen burn occurs and containment was assumed to fail at the peak pressure thus produced.

Although MAAP calculates a peak pressure of about 31 psia at 16.5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br /> which is below the containment desigr pressure, the pressures potentially attainable during a DCH event could be in the 100 psia range. The containment failure, and hence, the releases are assumed to occur at the top of the containment at an ,

elevation of 52.8 meters. The characteristics of this release class are j summarized in Table 4-1.

4.8 RELEASE CLASS RC3.1 Release class RC3.1 covers releases associated with a containment overpressure failure prior to core damage with no vaporization releases and no radioisotepe scrubbing in the containment atmosphere prior to release. The cumulative release frequency for this release class is 4.82E-8 per year.

This release class is characterized by a transient involving loss of main feedwater with a subsequent failure to deliver emergency feedwater. Feed and bleed cooling is successfully established with the bleed valves being opened shortly after primary safety valve lift. However, containment heat removal is not available so the containment temperature and pressure increase until containment fails at about 36 hours4.166667e-4 days <br />0.01 hours <br />5.952381e-5 weeks <br />1.3698e-5 months <br />. Safety injection fails just after containment failure. Cavity flooding is initiated just after containment failure.

Vessel failur- is assumed to occur at 51.2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br />. The containment failure, and

, hence, the releases are assumed to occur at the top of the containment at an

(

elevation of 52.8 meters. The characteristics of this relem class are summarized in Table 4-1.

l 4.9 RELEASE CLASS RC2.2 Release class RC2.2 covers releases associated with a containment isolation failure with no vaporization or revaporization releases prior to release. The cumulative release frequency for this release class is 1.12E-9 per year.

l This release class is characterized by a small (0.02 ft )2 hot leg LOCA with failure of safety injection and emerg&y feedwater. The safety depressurization Page 15 of 35

l valves fall to open. Vessel failure is assumed to occur at 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br />, causing the corium to be initially ejected into the cavity under high pressure. Containment spray is available for containment heat removal and fission product scrubbing.

Because the containment sprays are functioning, the cavity is also flooded prior to vessel failure. At time zero, a containment isolation failure equivalent to a 6 inch diameter hole is assumed to occur. The containment isolation failure, and hence, the releases are assumed to occur at ground level. The characteristics of this release class are summarized .a Table 4-1.

l i 4.10 RELEASE CLASS RC2.4 i

Release class RC2.4 covers releases associated with a containment isolation failure with no vaporization or revaporization releases but with no radioisotope scrubbing in the containment atmosphere prior to release. Thr cumulative release frequency for this release class is 2.44E-9 per year.

This release class is characterized by a station blackout with battery depletion.The accident is initiated by a loss of offsite power with failure of the diesel generators and the alternate AC power source. The batteries supply control power for 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> and emergency feedwater is supplied by the turbine-driven emergency feedwater pumps. With battery depletion at 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br />, emergency feedwater is assumed to be lost. Cavity flooding is initiated just prior to battery depletion. The hydrogen ignitors function to control the hydrogen concentration. Vessel failure is assumed to eccur at about 15.5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br /> causing the corium to be ejected into the cavity at high pressure. Neither power nor containment heat removal are recovered. At time zero a containment isolation failure equivalent to a6 inch diameter hole is assumed to occur. The containment isolation failure, and hence, the releases are assumed to occur at l ground level. The characteristics of this release class are summarized in Table 4-1.

I 4.11 RELEASE CLASS RC1.4 Release class RCl.4 covers releases associated with a containment bypass failure with no vaporization releases and no radioisotope scrubbing in the containment atmosphere pri:" to release. The cumulative release frequency for this release class is 7.02E-9 per year.

This release class is characterized by a steam generator tube rupture with a stuck open Atmospheric Dump Valve (ADV) o: the affected generator. The accident is initiated by a tube rupture in one of the steam generators. Safety injection is successfully initiated and emergency feedwater is delivered to' the intact generator. However, the ADV on the affected generator is astumed to fail open-at about 30 minutes. As a result of this, primary inventory and-IRWST inventory continued to flow through.the ruptured tube and is exhausted to the atmosphere.

Actions to stabiliz'e the plant and terminate the accident are unsuccessful.

Page 16 of 35 i

l l

Cavity flood is manually initiated while there is still sufficient inventory in the IRWST to successfully flood the cavity. At 10.6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br />, the IRWST inventory is depleted, and safety injection flow is lost. (At this point emergency feedwater is also assumed to be l_ost.)' At 29.9 hours1.041667e-4 days <br />0.0025 hours <br />1.488095e-5 weeks <br />3.4245e-6 months <br />, core.uncovery is assumed to occur and tSe mejor releases begins. At 32.6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> the vessel is assumed to ,

fail, and the cc. bm is ejected into the cavity. The releases through the ADV are assumed to accur at ground level. The characteristics of this release class are summarized ia Table 4-1.

4.12 RELEASE CLASS RCl.7

! Release class RCl.7 covers releases associated with a containmer" bypass failure l with- vaporization releases and no radioisotope scrubbing' in -tne containment I atmosphere prior to release but with the source term attenuated due to deposition in the nuclear ar.nex building, lhe cumulative rnlease frequency.for this release class is 3.00E-9 per year.

This release class is characterized by a- failure of the check and isolation valves in one Shutdown Cooling System (SCS) line resulting in a catntrophic failure of this line outside containment (Interfacing : System LOCA), Safety injection is sue:essful. However, the primary system inventary and the IRJST inventory is being discharged outside of containment. At'3.5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br />, the IRWST inventory is depleted and safety injection flow is lost. Vessel failure occurredat 7.1 hours1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br />, and the corium is dropped into a dry cavity. The release path is through the broken SCS line into the lower levels of the nuclear annex building in the shadow region. The radioactive material then passes upward ,

through the nuclear annex building until it eventually finds a release point to the environment at ground level. The characteristics of this release class are summarized in Table 4-1.

l l

L Page 17 of 35

, ._ . ~.. . . ,. .

Table 4-1 Release Parameter Data for System 80+ Release Classes RELEASE RELEASE RELEASE RELEASE ' RELEASE CLASS START QVRATION JiEIGHT ENERGY RC7.1 15.5 Hrs 10.0 Hrs OM 1.44E4 4 cal /sec RC6.2 330.0 Hrs 1.0 Hrs 0M 3.08E+8 cal /sec RC6.4 180.0 Hrs 1.0 Hrs OM 2.?lE+8 cal /sec

• RC5.1 128.0 Hrs 1.0 Hrs 52.8 H . 64E+8 cal /sec

, RC5.2 94.0 Hrs 1.0 Hrs 52.8 M 3.83E+8 cal /sec  !

RC4.1 5.6 Hrs- 1.0 Hrs 52.8 M 1.74E+7 cal /sec RC4.2 16.5 Hrs 1.0 Hrs -52.8 M 1.40E+7 cal /sec RC3.1 36.0 Hrs. 10.0 Hrs 52.8 M 1.58E+6 cal /sec RC2.2 3.0 Hrs 10.0 Hrs 'O M 5.68E+4 cal /sec RC2.4 15.5 Hrs 10.0 Hrs 0M 9.66E+5 cal /sec-RCl.4 29.9 Hrs 10.0 Hrs 0M 2.87E+3 cal /sec RCl.7 7.1 Hrs 1.0 Hrs OM 1.52E+7 cal /sec -

l l

L l

Page 18 of 35 l

l

5.0 DESIGN ALTERNATIVES Eleven design alternatives were evaluated. These were selected based on the SAMDAs for the Limerick plant2and a review of the System 60+ PRA'. In addition, suggestions from C-E personnel with technical expertise in containment response were employed. Design alternatives from earlier plant studies were also considered.

Each release class which is described in Section 4 was evaluated for total person-rem exposure using CRAC2. Table 5-1 gives the initiating frequency, dose at 0.5 miles, and total person-rom dose for the twelve release classes with initiating frequencies greater than 1.0E-10. The risk for each release class is the product of frequency (events / year) times the total person-rem exposure per event. This product gives person rem per year and is a measure of the risk. The total risk of the dominant release classes is 5.48 person-rem /y. These results are for the ALWR site which is representative of most of the current U.S. sites'.

Table 5-2 summarizes the accident characteristics for each release class. These are the dominant sequences of the binned accidents. ror each Design Alternative, the release class was reviewed assuming that the Design Alternative worked perfectly (failure rate - 0.0). This means that each Deri n Alternative had perfect rupport systems, power supplies and heat sinks. Ia addition, for each Design Alternative, no other failure modes were considered when the Design Alternative was employed. For example, when the pressurizer auxiliary spray >

Design Alternative is employed to ensure that the primary coolant pressure.can be decreased to enter SCS operation, the SCS system is assumed to always work.

This represents an upper limit scoping analysis and maximizes the benefit of each Design Alternative. If a Design Alternative is cost beneficial in this analysis then a more detailed analysis addressing the actual failure rate of the Design Alternative can be undertaken.

The Design Alternatives can be divided into two groups. One group prevents core damage and the other group protects the containment or reduces the releases. For the Design Alternatives that prevent core damage, the frequency: of affected release classes are put equal to zero_ and the total risk reduction _is calculated.

This grcup includes the high_ capacity HPSI systems, improved DC Battery and EFWS, ATWS pressure relief valves, improved pressurizer auxiliary spray, improved

( primary depressurization system, and advanced reactor coolant pump seal injector.

l For the Design Alternatives that protect the containment, .the releases are

, changed from that of a failed containment to that of-an intect containment and l then the risk is reevaluated. These Design Alternatives include the advanced l containment sprays, filtered vent, concrete composition, reactor vessel exterior l cooling, and H2 ignitors. The person-rem exposure for an intact containment l (RC7.1) is substituted for the higher dosed when mitigation type Design Alternatives ara evaluated. For the filtered vent Design Alternative, the doses represent immediate release of the noble gasses after scrubbing (RC4.1).

l The following sections discuss each Design Alternative.

Pap 19 of 35

S 4 5.1 CONTAINMENT SPRAY A perfect containment spray system prevents the high pressure containment failures caused by slow steam pressurization (H2 burns are expected to still be possible). This system is assumed to have a perfect power supply and heat sink and work in all release classes where the containment is challenged regardless of the sequence of events or equipment failures that led t - core damage and containment challenge. These assumptions overestimate the benefits of _this design alternative. It also reduces the releases in all the release classes where no scrubbing of fission products was initially predicted. This Design Alternative reduces the dose of five of the release classes to the dose associated with an intact containment (RC7.1) but does not affect the frequencies of the events. The risk (frequency x dose) of these five RCs therefore has been reduced. The total risk is reduced by 90% (see _ Table 5-3). Using a risk conversion factor of $1,000 per person-rem, this Design Alternative would have an annual value of $4,900/y. The annual benefit of the Design Alternative could be converted to a capital benefit using the levelized _ capital cost rate .of 17.9%

developed in Section 2. The ideal containment spray system would be cost beneficial if it could be installed for less than $27,600 and have no maintenance and testing costs. Any annual operating costs would have to be subtracted from the annual risk reduction benefits.

The above analysis assumes that the system has a fcilure rate of 0.0 in terminating the accident by protecting the containment. The capital benefit is inversely proportional to the reliability of the system. For example, if the design had a conditional reliability of 0.5 in these accident sequences, then the Design Alternative would have to cost less _than $13,800 to be cost effective.

Estimating the cost to design and build a perfect containment spray system is not realistically possible. However, one option-would be to provide piping from the containment spray header to the exterior of the Nuclear Annex for a temporary hook-up of a fire truck should all containment spray and shutdown cooling pumps be unavcilable. The cost of the: additional Class 2 piping, pipe supports, valves, on-site fire truck with the required pumping capacity and pump head and building to store the fire truck is estimated to exceed $1.5 million.

5.2 FILTERED VENT The filtered vent Design Alternative prevents all slow high pressure containment l

failures and therefore reduces the doses in three release classes (see Table 5-4). Since the noble gases are not held up, the doses associated with the filtered vent were approximated with the doses associated with a scrubbed release without any fission product vaporization (RC4.1). As with all mitigation Design Alternatives, this design alternative does not affect the frequencies of the RCs.

This Design Alternative reduces the risk by 86%. Using a value of $1,000 per person-rem avoided, this Design Alternative has a benefit of $4,700/y. Using a levelized capital cost rate of 17.9%, a system with- a-capital cost of $26,300 l

would just be cost effe.ctive.

Page 20 of 35

The Swiss recently purchased a filtered vent system for one of their BWRs for approximately $3 million" for equipment along. With the building structure to house the equipment the total cost is estimated to exceed $10 million.

5.3 DC BATTERIES AND EFWS This Design Alternative addresses the release classes where emergency feedwater is lost after battery depletion during a station blackout. The System 80+ design already has an improved battery system that will carry the DC loads for 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br />.

There are still accident sequences where the batteries are depleted and emergency feedwater is lost leading to core damage. The improved DC batteries and EFWS Design Alternative is assumed to have the capability to remove decay heat using i batteries and the turbine feedwater pump for whatever time period that is required (without any failures). This Design Alternative prevents core damage

, and therefore removes six of the release classes. The risk is reduced by 69%

! (see Table 5-5). Using a $1,000 per averted penon/ ram and a levelized cost rate l or 17.9%, such a system would be cost beneficial if it cost less than $21,100.

Design of a battery system with unlimited capacity is not possible. However, to increase the existing battery capacity for the EFW pumps from the current System 80+ design capacity of 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> to 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> will require 9 times the number of current battery cells and thus approximately 9 times the space _ for building storage. The increased building space will also increase the HVAC requirements.

l The cost for the extra battery cells, building volume and increased HVAC

, requirements is estimated to exceed $2 million.

l 5.4 RCP SEAL COOLING The System 80+ employs a type of Reactor Coolant Pumps (RCP) seal which can withstand a loss of cooling and not produce a LOCA. This type of seal design has been employed in the operating C-E plants and experience has shown that the seals do not fail when seal cooling is lost *, Therefore seal failure was not modeled in the System 80+ PRA. To estimate a risk reduction associated with an improved seal cooling system, a different approach is used than used for the other Design Alternatives. It is assumed that for all station blackouts that last four hours, the seal fails and leads to core damage (frequency - 3.64E-8/y). It is also assumed that total loss of the component cooling water (CCW) also leads to seal-failure and core damage. Appendix A contains the fault tree for the analysis of total loss of CCW. Total loss of CCW requires loss of four CCW pumps or loss of the four service water pumps and would probably require a common cause failure.

l The frequency for this event is 2.25E-5/y. The containment is assumed to remain

-intact (conditional containment failure probability - 0.099) and the expected dose is 4.8E+4 (RC7.1). Therefore, the risk of seal loss leading to core damage l is 1.08 person-rem /y (2.25E-5 events /y x 4.8E+4 person-rem / event). The RCP seal-cooling Design Alternative is assumed to completely _ eliminate this ' risk and Page 21 of 35

therefore reduces the total risk of the plant by 16.5%. Using a value of $1,000 per person-rem avoided, this Design Alternative has a benefit of $1,080/y. Using a levelized capital cost rate of 17.9%, a system with a capital cost of $6,034 would just be cost effective.

The reliability of the reactor coolant pump seal cooling could be improved by adding a small dedicated positive displacement pump for diverse seal injection.

This pump would be powered from the Alternate AC Source (Combustion Turbine).

It would take suction from the boric acid storage tank and would connect to the normal supply line for seal injection inside the containment. This design addition will- provide additional diversity for RCP seal cooling and provide a seal cooling system that is not dependent on CCW. The cost of the additional pump, piping, valves, containment penetration, instrumentation, electrical cable and building space is estimated to exceed $100,000.

5.5 PRESSURIZER AUXILIARY SPRAY This Design Alternative was introduced to specifically address steam generator tube rupture (SGTR) and eliminates the one RC initiated by SGTR (RC1.4). The analysis assumes that during a SGTR, the auxiliary spray will alwc.ys depressurize the primary system to the SCS operation mode with sufficient speed and the SCS system will always remove decay heat. The elimination of the risk of SGTR in the System 80+ design has a 6.7% risk reduction (see Table 5-6). Using a $1,000 per av-ted person-rem and a levelized cost rate or 17. 9%, such a system would be co:. beneficial if it cost less than $2,050.

Designing a perfect pressurizer auxiliary spray system is not possible. However, increased reliability and diversity can be obtained by increasing the redundancy and diversity of the pressurizer spray valves and providing a diverse positive displacement charging pump that is powered from a diverse power source. The l reliability of the shutdown cooling system can be improved by providing a diverse shutdown cooling pump with a diverse power source and providing a diverse heat sink. The cost for the additional components, piping, power -supplies, instrumentation and building volume is estimated to exceed $5 million.

5.6 ATWS PRESSURE RELIEF VALVES This Design Alternative consists of a system of relief valves that can prevent any equipment damage from a primary coolant pressure spike in an ATWS accident sequence. This Design Alternative is assumed to eliminate all the ATWS core damage sequences- . The initiating frequency for each release class consists of-one or more accident sequence. The descriptions of each RC given in Section 3 and Table 4-2 are typical accident sequences that contribute to the release class frequency. ATWS events represent 55% of the initiating frequency of RC7.1 and 50% of the initiating frequency of RC4.2. The elimination of the ATWS events therefore reduces the frequency of RC7.1 by 55% and RCl.4 by 50%. This leads to a total risk reduction of 4.2% (see Table 5-7). Using a $1,000 per averted l person-rem and a levelized cost rate or 17.9%, such a system would be cost beneficial if it cost less than $1,290.

Page 22 of 35

To implement this design alternative, the safety relief valve sizes and discharge piping size would need to be increased. It may also require additional safety relief valves and thus additional safety relief valve discharge piping and supports. In addition, the size and possible the number of safety valve nozzles on top of the pressurizer would need to be increased. The cost of this design alternative is estimated to exceed $1 million.

5.7 CONCRETE COMPOSITION The containment building for System 80+ uses a spherical containment with an area below it that can be considered part of the nuclear annex building. It is assumed that in accident sequences where corium/ concrete interaction are not stopped, containment failure would lead to releases through the nuclear annex building. This Design Alternative assumes that an ideal concrete composition could be developed that prevents basemat melt-through. This would reduce two RCs (RC6.2 and RC6.4) where basemat melt-through is modeled and would produce.a 2.5%

risk reduction. Table 5-8 summarizes the risk reduction. Using a $1,000 per averted person-rem and a levelized cost rate or 17.9%, such a system would be cost beneficial if it cost less than $765.

An advanced concrete composition to prevent corium/ concrete interaction is not currently auilable. However, additional concrete could be added to increase the time before containment failure would occur. In order to increase the thickness of the concrete at the bottom of the reactor cavity the containment diameter would need to be increased. An increase in containment diameter also requires an increase in containment plate thickness. The cost of increasing the concrete thickness by two feet is estimated to exceed $5 million.

5.8 REACTOR VESSEL EXTERIOR COOLING A reactor vessel exterior cooling system is assumed to prevent vessel melt-

. through and subsequent basemat attack. It also prevents Direct Containment Heating (DCH) sequences by preventing vessel failure because it promotes cooling the corium in the vessel. This Design Alternative reduces the consequences of four RCs by reducing the doses to that of an intact containment (RC7.1). It has a dose risk reduction of 2.5% (see Table 5-9). Using a' $1,000 per averted person-rem and a levelized cost rate or 17.9%, such a system would be cost beneficial if it cost less than $765.

The current' arrangement for the IRWST will allows wetting the bottom of ' the reactor vessel through the cavity flood system. However, the water level of the IRWST is not high enough to flood the entire cavity for reactor vessel exterior cooling. Reactor vessel exterior cooling could be accomplished by ra' ting the elevation of the IRWST such that the entire reactor cavity can be flooded up the reactor flange. Raising the elevation of the IRWST would require an increase in containment diameter. In addition, the reactor vessel would have to be qualified Page 23 of 35

.}

to demonstrate that wetting the exterior will not cause a thermal shock which l would cause vessel failure. To implement- this design alternative is estimated I

to exceed $5.5 million.

l 5.9 H2 IGNITORS l

Ideal hydrogen (H2) ignitors would prevent release classes associated with containment failures from hydrogen burns or explosions. The System 80+ design already has H2 ignitors and therefore only one release class (RC4.2) has containment failure from hydrogen burning. This Design Alternative reduces the doses to that of an intact conta,nment and therefore reduces the risk by 0.1% (see Table 5-10). Such a system would have to have a negligible cost to be cost beneficial.

Providing perfect hydrogen ignitors which have no probability of failure is not possible. However, the reliabiiity of the hydrogen ignitors could be improved by either providing dedicated batteries for the existing design (glow plug ignitors) or by providing catalytic hydrogen ignitors which do not require a power source. Since catalytic hydrogen ignitors are not fully developed, possible f ailure modes, including common cause failure modes, are not known.

Therefore, they are not being selected for the System S.+ design at this time.

The addition of dedicated batteries for the hydrogen ignitors along with the additional equipment such as battery chargers and invertor and the additional building space to store this equipment is estimated to exceed $1 million.

5.10 HIGH PRESSURE SAFETY INJECTION The System 80+ design has a very reliable four train HPSI system to begin with.

The high pressure safety injection Design Alternative assumes that most sequences with HPSI failures can be aliminated (see Table 5-11). This Design Alternative eliminates RC4.1 and 2.2. For these two RCs, the core damage frequency goes to zero. It is assumed that the HPSI Design Alternative does not prevent core damage for RC3.1. In this release class, an early containment failure produces HPSI pump cavitation because the rapid depressurization caused the primary coolant to flash. This Design Alternative produces a 0.004% risk reduction and therefore would have to have a negligible cost to be cost beneficial.

Providing a perfect high pressure safety injection system is not possible; however, the reliability of the system could be improved slightly by adding two more diesel generators. The cost of adding two more diesel generators, with the j associated support systems and the building space is estimated to exceed $20 l million.

i

Page 24 of 35

)

5.11 RCS DEPRESSURIZATION ,

The System 80+ design has motor operated relief valves (M0RVs) that permit residual heat removal using the valves and HPSI pumps in a " feed and bleed" mode of operation. This Design Alternative models a perfect MORV system that permits the primary coolant system to be quickly depressurized so that the Safety Injection Tanks (SITS) and Safety Injection pumps are effective in getting coolant into the core and removing decay heat. This Design Alternative eliminates the one RC where the HPSI fails or feed and bleed fails because of failure of the four traia HPSI system. It assumes that the SITS, and RHR systems also have zero failures for these sequences and therefore the core damage frequency goes to zero for RC2.2. It is assumed that-the RCS Depressurization system does not prevent core damage for RC3.1. In this release class, an early containment failure produces HPSI pump cavitation because the rapid depressurization caused the primary coolant to fl ash. This is the least effective design alternative considered in this study with only a 0.002% dose risk reduction (see Table 5-12). Such a system would have to have negligible costs to be cost beneficial.

Designing a perfect safety depressurization systm is not possible. However, increased reliability and diversity of the syste . ' be obtained by increasing the redundancy of the safety depressurization vah J/or providing valves that are diverse. froviding the additional valves, pa. -

and instrumentation.is estimated to exceed $500,000.

I Page 25 of '35

4 .

TABLE 5-1 FREQUENCIES AND PERSON-REM EXPOSURES FOR RELEASE CLASSES RC# INITIATING PERSON-REMS TOTAL RISK FREQUENCY AT 0.5MI PERSON-REM PERSON-REM /Y 7.1 1.69E-6 25.9 4.80E+4 8.11E-2 6.2 1.38E-8 28.2 1.24E+6 1.71E-2 6.4 7.28E-9 566.0 1.66E+7 1.21E-1 5.1 3.63E-8 630.0 1.24E+7 4.50E-1 5.2 6.59E-8 2510.0 4.72E+7 3.11E-0 ,

4.1 2.12E-10 76.3 5.39E+5 1.14E-4 4.2 2.17E-10 3830.0 12.14E+7 4.64E-3 3.1 4.82E-8 1540.0 2.60E+7 1.25E-0 2.2 1.12E-9 36.7 9.01E+4 1.01E-4 l 2.4 2.44E-9 4280.0 8.76E+6 2.14E-2 1.4 7.02E-9 71900.0 5.22E+7 3.66E-1 1.7 3.00E-9 6140.0 1.94E+7 5.82E-2 Page 26 of 35

1 I

TABLE 5-2 i

SUMMARY

DESCRIPTION OF RELEASE CLASSES RM INIT TOTAL DESCRIPTION j FREQ PERSON-REM -

7.1 1.69E-6 4.80E+4 BLACK 0UT, DC DEPLETION, RESTORED CONT flEAT REMOVAL, CONTAINMENT INTACT 6.2 1.38E-8 1.24E46 1.ARCE LOCA, NO S!, NO CAVITY FLOOD, CONT MELT-THRU, i

VAPOR; "10N RELEASE, NO SCRUBBING -

6.4 7.28E-9 1.66E+7 BLACK 001, OC DEPLETION, RESTORE CONT flR, NO CAVITY FLOOD, CONT MELT-THRU, VAPORIZATION RELEASE, RE-VAPORIZATION, NO SCRUBBING 5.1 3.63E-8 1.24E47 BLACK 0UT, DC DEPLETION, CAVITY FLOOD, NO AC, NO CONT llR, LATE OVERPRESSURE F, NO VAP., NO REVAP .

N0 SCRUBBING 5.2 6.59E-8 4.72E+7 BLACK 0UT, DC DEPLETION, CAVITf FLOOD, NO AC, NO CONT HR, LATE OVERPRESSURE F, N0 VAP., REVAP., NO

, SCRUBBING l

l 4.1 2.12E-10 5.39E+5 LARGE LOCA, NO SI, CONT SPRAY & HR, CODIUM DROP l INTO WET CAVITY, EARLY STEAM OVERPRESSURE F, NO VAPORIZATION, f.o REVAP., SCRUBBING '

4.2 2.17E-10 2.14E+7 BLACK 0UT, DC DEPLETION, CAVITY FLOOD, H2 IGNITORS F, H2 BURN + OCH, EARLY OVERPRESSURE F, NO VAP., NO REVAP., NO SCRUBBING 3.1 4.82E-8 2.60E+7 LOSS MFW+EFW, FEED +BLEEC.N0 CONT HR, CONT F PRIOR 10 CM, $1 FAILS, CAVITY FLOOD, NO VAP., NO REVAP.,

N0 JRUBBING 2.2 1.12E-9 9.01E+4 SMALL !0CA, SI F, EFW F, DEP. VLVS F OPEN, CONT ISOLATION F, H0 VAP., NO REVAP., SCRUBBING l 2.4 2.441. 1 8.76E+6 BLACK 0UT, DC DEPLETION, CAVITY FLOOD, H2 IGNITORS l OK, NO AC, NO CONT HR, CONT ISOLATION F, N0 VAP.,

NO REVAP., NO SCRUBBING g

l 1.4 7.02E-9 5.22E+7 STEAM GEN TUBE RUPT, SI OK, EfW OK, ADV STUCK OPEN, PRIMARY INVENT 0RY LOST, CAVITY FLOODED, CONT BYPASS, NO VAP., NO REVAP., NO SCRUBBING 1.7 3.0F 4 1.94E+7 CHECK AND ISOL VLV F IN PHR (INTERFACE LOCA), SI OK, PRIM INVENTORY DEPLETED, DRY CAVITY, CONT BYPASS, VAP., NO SCRUBBING, DEPOSITION IN AUX BLDG Page 27 of 35 L

TABLE 5-3 RISK REDUCTION EVALUATION FOR CONTAINMENT SPRAY RC# INIT TOTAL BASE REVISED COMMENTS FREQ PERSON-REM PERSON-REM / YEAR 7.1 1.69E-6 4.80E+4 8.11E-2 8 llE-2 6.2 1.38E-8 1.24E+6 1.71E 6.62E-4 IMPROVED 6.4 7.28E-9 1.66E+7 1.21E-1 3.49E-4 IMPROVED 5.1 3.63E-8 1.24E+7 4.50E-1 1.74E-3 IMPROVED 5.2 6.59E-8 4.72E+7 3.llE-0 3.16E-3 IMPROVED 4.1 2.12E-10 5.39E45 1.14E-4 1.14E-4 4.2 2.17E-10 2.14E+7 4.64E-3 4.64E-3 H2 BURN 3.1 4.82E-8 2.60E+7 1.25E-0 2.31E-3 IMPROVED 2.2 1.12E-9 9.01E+4 1.01E-4 1.01E-4 2.4 2.44E-9 8.76E46 2.14E-2 2.14E-2 CONT OPEN 1.4 7.02E-9 5.22E+7 3.66E-1 3.66E-1 1.7 3.00E-9 1.94E47 5.82E-2 5.82E-2 5.48E-0 0.54E-0 (90% RISK REDUCTION)

TABLE 5-4 RISK REDUCTION EVALUATION FOR FILTERED VENT RC# INIT TOTAL BASE REVISED COMMENTS FREQ PERSON-REM PERSON-REH/ YEAR 7.1 1.69E-6 4.80E+4 8.llE-2 8.llE-2 6.2 1.38E-8 1.24E+6 1.71E-2 1.71E-2 6.4 7.28E-9 1.66E+7 1.21E-1 1.21E-1 5.1 3.63E-8 1.24E+7 4.50E-1 1.96E-2 IMPROVED-b.2 6.59E-8 4.72E+7 3.llE-0 3.55E-2 IMPROVED 4.1 2.12E-10 5.39E+5 1.14E-4 1.14E-4 4.2 2.17E-10 2.14E+7 4.64E-3 4.64E-3 3.1 4.82E-8 2.60E+7 1.25E-0 2.60E-2 IMPROVED 2.2 1.12E-9 9.61E+4 - 1.01E-4 1.01E-4 2.4 2.44E-9 8. 76E4 6 - 2.14E-2 2.14E-2 1.4 7.02E-9 5.22E+7 3.66E-1 3.66E-1 1.7 3.00E-9 1.94E47 5.82E-2 5.82E-2 5.48E-0 0.75E-0 (86% RISK REDUCTION)

Page 28 of 35

, . . _ _ . , e , .g , ,_ -

TABLE 5-5 RISK REDUCTION EVALUATION FOR DC BATTERIES & EFWS RC# INil TOTAL BASE REVISED COMMENTS FREQ PERSON-REM PERSON-REM / YEAR 7.1 1.69E-6 4.80E44 8.11E-2 0.0 IMPROVED 6.2 1.38E-8 1.24E+6 1.71E-2 1.71E-2 6.4 7428E-9 1.66E+7 1.21E-1 0.0 IMPROVED 5.1 3.63E-8 1.24E+7 4.50E-1 0.0 IMPROVED 5.2 6.59E-8 4.72E+7 3.llE-0 0.0 IMPROVED 4.1 2.12E.10 5.39E+5 1.14E-4 1.14E-4 4.2 2.17E-10 2.14E+7 4.64E-3 0.0 IMPROVED 3.1 4.02E-8 2.60E+7 1.25E-0 1.2EE-0 2.2 1.12E-9 9.01E+4 1.01E-4 1.01E-4 2.4 2.44E-9 8.76E+6 2.14E-2 0.0 IMPROVED 1.4 7.02E-9 5.22E+/ 3.66E-1 3.66E-1 1.7 3.00E-9 1.94E+7 5.82E-2 5.82E-2 5.48E-0 1.69E-0 (69% RISK REDUCTION)

TABLE 5-6 RISK REDUCTION EVALVATION FOR PRESSURIZER AUXILIARY SPRAY RC# INIT TOTAL BASE REVISED COMMENTS FREQ PERSON-REM PERSON-REM / YEAR 7.1 1.69E-6 4.80E+4 8.llE-2 8.11E-2 6.2 1.38E-8 1.24E+6 1.71E-2 1.71E-2 6.4 7.28E-9 1.66E+7 1.21E-1 1.21E-1 5.1 3.63E-8 1.24E+7 4.50E-1 4.50E-1 5.2 6.59E-8 4.72E+7 3.llE-0 3.llE-0 4.1 2.12E-10 5.39E45 1.14E-4 1.14E-4 4.2 2.17E-10 2.14E+7 4.64E-3 4.64E-3 3.1 4.82E-8 2.60E+7 1.25E-0 1.25E-0 2.2 1.12E-9 9.01E+4 1.01E-4 1.01E-4 2.4- 2.44E-9 8.76E+6 2.14E-2 2.14E-2 '

l.4 7.02E-9 5.22E+7 3.66E-1 0.0 IMPROVED 1.7 3.00E-9 1.94E+7 5.82E-2 5.82E-2 5.48E-0 5.llE-0 (6.7% RISK REDUCTION) -

Page 29 of 35

TABLE 5-7 RISK REDUCTION EVALVATION FOR ATWS VALVES RC# INIT TOTAL BASE REVISED COMMENTS FREQ PERSON-REM PERSON-REM / YEAR 7.1 1.69E-6 4.80E+4 8.llE-2 3.65E-2 55% IMPROVED 6.2 1.38E-8 1.24E46 1.71E-2 1.71E-2 S.4 7.28E-9 1.66E47 1.21E-1 1.21E-1 5.1 3.63E-8 1.24E+7 4.50E-1 4.50E-1 5.2 6.59E-8 4.72E+7 3.11E-0 3.llE-0 4.1 2.12E-10 5.39E45 1.14E-4 1.14E-4 4.2 2.17E-10 2.14E+7 4.64E-3 4.64E-3 3.1 4.82E-8 2.60E+7 1.25E-0 1.25E-0 2.2 1.12E-9 9.01E+4 1.01E-4 1.01E-4 2.4 2.44E-9 8.76E+6 2.14E-2 2.14E-2 1.4 7.02E-9 5.22E+7 3.66E-1 1.83E-1 50% IMPROVED 1.7 3.00E-9 1.94E47 5.82E-2 5.82E-2 5.48E-0 5.25E-0 (4.2% RISK REDUCTION)

TABLE 5-8 RISK REDUCTION EVALUATION FOR CONCRETE COMPOSITION RC# INIT TOTAL BASE REVISED COMMENTS FREQ PERS0it-REM PERSON-REM / YEAR 7.1 1.69E-6 4.80E+4 8.11E-2 8.11E-2 6.2 1.38E-8 1.24E+6 1.71E-2 6.62E-4 IMPROVED 6.4 7.28E-9 1.66E+7 1.21E-1 3.49E-4 IMPROVED 5.1 3.63E-8 1.24E+7 4.50E-1 4.50E-1 5.2 6.59E-8 4.72E47 3.llE-0 3.llE-0 4.1 2.12E-10 5.39E+5 1.14E-4 1.14E-4 4.2 2.17E-10 2.14E+7 4.64E-3 4.64E-3 3.1 4.82E-8 2.60E+7 1.25E-0 1.25E-0 2.2 1.12E-9 9.01E+4 1.01E-4 1.01E-4  ;

2.4 2.44E-9 8.76E+6 2.14E-2 2.14E-2

• 1.4 7.02E-9 5.22E+7 3.66E-1 3.66E-1 1.7 3.00E-9 1.94E+7 5.82E-2 5.82E-2 5.48E-0 5.34E-0 (2.5% RISK REDUCTION)

Page 30 of 35

TABLE 5-9 RISK REDUCTION EVALVATION FOR VESSEL EXTERIOR COOLING ,

RC# INIT TOTAL BASE REVISED COMMENTS FREQ PERSON-REM PERSON-REM / YEAR 7.1 1.69E-6 4.80E+4 8.llE-2 8.llE-2 6.2 1.38E-8 1.24E+6 1.71E-2 6.62E-4 IMPROVED 6.4 7.28E-9 1.66E+7 1.21E-1 3.49E-4 IMPROVED 5.1 3.63E-8 1.24E+7 4.50E-1 4.50E-1 5.2 6.59E.8 4.72E+7 3.llE-0 3.11E-0 4.1 2.12E-10 5.39E+5 1.14E-4 1.02E-5 IMPROVED >

4.2 2.17E-10 2.14E+7 4.64E-3 1.04E-5 IMPROVED 3.1 4.82E-8 2.60E+7 1.25E-0 1.2SE-0 '

2.2 1.12E-9 9.01E+4 1.01E-4 1.01E-4 2.4 2.44E-9 8.76E+6 2.14E-2 2.14E-2 1.4 7.02E-9 5.22E+7 3.66E-1 3.66E-1 1.7 3.00E-9 1.94E+7 5.82E-2 5.82E-2 5.48E 5.34E-0 (2.5% RISK REDUCTION)

TABLE 5-10 l RISK REDUCTION EVALUATION FOR H2 IGNITORS l RC# INIT TOTAL BASE REVISED COMMENTS FREQ PERSON-REM PERS0H-REM / YEAR l

l 7.1 1.69E-6 4.80E+4 8.11E-2 8.llE-2 6.2 1.38E-8 1.24E+6 1.71E-2 1.71E 6.4 7.28E-9 1.66E+7 1.21E-1 1.21E-1 5.1 3.63E-8 ?.24E+7 4.50E-1 4.50E-1 5.2 6.59E-8 4.72E+7 3.11E-0 3.llE-0 l 4.1 2.12E-10 5.39E+5 1.14E-4 1.14E-4

[ 4.2 2.17E-10 2.14E+7 4.64E-3 1.04E-5 IMPROVED 3.1 4.82E-8 2.60E+7 1.25E-0 1.25E-0 2.2 1.12E-9 9.01E+4 1.01E-4 1.01E-4 2.4 2.44E-9 8.76E+6 2.14E-2 2.14E-2 1.4 7.02E-9 5.22E+7 3.66E-1 3.66E-1 i 1.7- 3.00E-9 1.94E+7 5.82E 5.82E-2 5.48E-0 5.43E-0 (0.1% RISK REDUCTION)  ;

Page 31'of 35

,---,,,n- - - - , , w , -, ,, - - - r

i l

j l

TABLE 5-11 RISK REDUCTION EVALVATION FOR HPSI RC# INIT TOTAL BASE REVISED COMMENTS PERSON-REM FREQ PERSON-REM / YEAR )

7.1 1.69E-6 4.80E44 8.11E-2 8.llE-2 6.2 1.38E-8 1.24E46 1.71E-2 1.71E-2 6.4 7.28E-9 1.66E+7 1.21E-1 1.21E-1 5.1 3.63E-8 1.24E+7 4.50E-1 4.50E-1 5.2 6.59E-8 4.72E+7 3.llE-0 3.11E-0 4.1 2.12E-10 5.39E+5 1.14E-4 0.0 IMPROVED  ;

4.2 2.17E-10 2.14E+7 4.64E-3 4.64E-3 3.1 4.82E-8 2.60E+7 1.25E-0 1.25E-0 2.2 1.12E-9 9.01E44 1.01E-4 0.0 IMPROVED 2.4 2.44E-9 8.76E46 2.14E-2 2.14E-2 1.4 7.02E-9 5.22E+7 3.66E-1 3.66E-1 1.7 3.00E-9 1.94E+7 5.82E-2 5.82E-2 5.48E-0 5.48E-0 (0.004 RISK REDUCTION)

TAFiE 5-12 RISK REDUCTION EVALVATION FOR RCS DEPRESSURIZATION RC# INIT TOTAL BASE REVISED COMMENTS FREQ PERSON-REM PERSON-REH/ YEAR 7.1 1.69E-6 4.80E44 8.llE-2 8.1)E-2 6.2 1.38E-8 1.24E+6 1.71E-2 1.71E-2 6.4 7.28E-9 1.66E4 7 1.21E-1 1.21E-1 5.1 3.63E-8 1.24E+7 4.50E-1 4.50E-1 5.2 6.59E-8 4.72E+7 3.llE-0 3.11E-0 4.1 2.12E-10 5.39E45 1.14E-4 1.14E-4 4.2 2.17E-10 2.14E+7 4.64E-3 4.64E-3 3.1 4.82E-8 2.60E+7 1.25E-0 1.25E-0 2.2 1.12E-9 9.01E+4 1.01E-4 0.0 IMPROVED 2.4 2.44E-9 8.76E+6 2.14E-2 2.14E-2 1.4 7.02E-9 5.22E+7 3.66E-1 3.66E-1 1.7 3.00E-9 1.94E+7 5.82E-2 .5.82E-2 5.48E-0 5.48E-0 (0.002f. RISK REDUCTION)

Page 32 of 35 l

l  !

I -, - - -

. - . . - - .- . . _ _ - - - - _ - - . - . - - - . . - _ . - - - = - _ _ -

6.0 REFEREN Wi

1. System 83+ Standard Desian Probabilistic Risk Assp a n.1, ABB Combustion Engineering, DCRT-RS-02 REV.0, January, 1991.

2. Varga, S.A., " Supplement to the Final Environinental Statement - Limerick Generating Station, Units I and 2", Docket nos. 50-352/353 August 16, 1989.

3. Crutchfield, D.M., " Severe Accident Mitigation Alternatives for Certified Standard Designs", Docket no.52-002, Nove 6 er 21, 1991.

4. Fauski & Associates, Inc.: Modular Accident Analys1Efto, gram (MAAP_l, Atomic Industrial Forum, IDCOR Program Technical Report 16.2-3, february 1987.

5. Ritchie, L.T., et al, "CRAC2 Model Description", NUREG/CR-2LSE, March 1984.

6. Hedrick, G.E., Duke Power, Let'er to Sugnet, W.R., EPRI, "ALWR PRA o :y Assumptions and Groundrules (KAG) Reference Site CRAC2 Input Files und Narrative Duke File: ASI-1407", April 17, 1989.

7. Delene, J.G., Bowers, H.I., " Draft Proposed Power Generation Cost Methodology for NA3AP/INFCE", U.S. Department of Energy, Office of Fuel Cycle t.valuatier, November 30, 1978.

l 8. Nucleonics Week., February 20, 1992, h ge 3.

9. " Generic Issue - 23, Ewluation of the Reactor Pump Seal Integrity Issue",

Combustion Engiraering Inc., CEN-408, September,1991.

l l

1 l

Page 33 of 35

APPENDIX A l

f FAULT TREE ANALYSIS FOR TOTAL LOSS rF 1

COMPONENT COOLING WATER l

l l

1 t

l r

Page 34 of 35 i

FAULT TREE ANALYSIS FOR TOTAL LOSS OF COMPONENT COOLING WATER There are two trains of Component Cooling Water (CCW). Each train has one pump operating and one pump in stanuby. In addition, each CCW trains require service water. Each CCW train is supported by two service water pumps, one operating and one on standby. The following fault tree models total loss of CCW including common cause failures. This tree was evaluated using the CAFTA code and the resulting failure frequency was 2.25E-5/y.

l l

l Page 35 of 35 l

d I

?

I

, I l

i I

i.

t

• ^

3 t

I~

I i

FAILURE OF CCW t t

{

l t .

i I CCtK-QQ1 I  ;

-+

I I i i t

FAILURE OF CCW FAILUAE OF CCW TAAIN A TRAIN 8 i

! I C(wT *002 I ICCwiB001]

/

r ,

i b, l

_J L , , ,

FAILUAE OF TAAiN A 4 FAILUAE CF TRAIN A FAILUAE OF TRAIN 3 FAILUAE OF TRAIN B

^

CCW PUMPS ' ES= PUMPS CCW Pt SPS ESW Pt.NPS  :

i iccwraco:>] l cc= 7aco3 1 Iccwiecc2 1 Lccwreco3I r Pa e 2 Page 4 Page 7 Fege.O

/

r 1

1

i

'i FAILURE OF CCW .\ TREES \CCWSAMDA.CAF 4-15-92 Page i  ;

i i F

- + - , - , , , ,,,,..-%-, . . . ,-...-,-,,-~,:,.... .. ..m.. . . . , . . m.,,-~ . . , - . . . . . . . . - , - + . - . _ _ . . , . . . . , , . -- . . - -

l i

6 4

l i

i FA!LupE OF TrutIn a I CCW PUM*S

.gb mg pn 5

k j

1 i 1 '

FAltenE OF CCr Puka j l F AILURE OF CCd Mp6 at g 82 l

! l 1 flTnQ!O I C .* ' a ' '.t_]

7 g,

[

l 1 i 2

1 s l Comeowf 47 C00 Lim 3 CPEmattus CCa*04 CCwPONEh? COOLIN9 CMMMI COCLDO CW4 Cay $E waTEm pump A2 waTre *Lw A2 FA;Lums CF Tw3 metes Pthe ai CauSE FaruSE Cf 2 fa:LS T2 AUM & 2 frNNING CCw FAILS TO AUN FAILS f2 STam* Statcet C% PM Pes *S TO v am-GfN a ' I CC=2 CEC R C# 1 l CPN A2 I fDN*I I & CC=ENSZ' I 1

, gg .." ' O Q)

I

, , I CO *0hENT CCGLIMS DETA FaCTCa Ptg t f +

t WATER Puke a2 Fa!LW OF 3 CF 2

$ PAILS TO Stee* CCu pasart TC Staat y

ICt% 2i i CC9f* C C wasiini a A-se i l 1

FAILURE OF CCW .'\ TREES \CCWSAMDA.CAF 4-15-92 l Pace 2 - ,

I 4

i P t

4 t +

)

i .

I 1

a i

I i

OPERATING COMMCN CAUSE F AILUAE CF 2

• CF 2 At.NNING CCW

' PUMDS-

?

i ICCw2PwPACCF 1

> Page 2 ,

Pege 7 .

i.

n s t COMPONENT COOLING BETA FACTCA FCA-4 WATER PUwP A1 - CPEAATING CO* CN j FAILS TO AUN CAUSE FAILUAE OF .

t

, CCW PLWS l CoSM A1 1 1CCwomar$ I KAG11/91 A.A-3?

l l'

1 i

FAILURE"OF CCW .\ TREES \CCWSAMDA.CAF 4-15-92 Page 3 '

r e

i

-I z.

?

i f

, i l' i

=

FAILUAE OF TPAIN A i ESW PUS 8PS  !

- I

1 t p , ICCwTh0031 l' l a

. 'I I I i r 4 FAILLAE OF ESW PUwP

~

FAILUAE OF ESW PUMP ,

A2 [

A1- t

l. j f I CCmTnOGS l I_CCw!ACO7 ]

2 i

i I I i

l OPEAATING CO* MON SEAVICE WA M PUwP MO4 CAUSE SEAVICE WATEA Pthap FAILS FAILtJAE CF 2 CF 2 ,

At FAILS- CAUSE FAILUAE OF 2 .A2 OF 2 RUNNING ESW TO STAAT STANN3f E5W PUMPS t

• TO RJN 70 STAAT j ,

' PUMPS t lCESw2PMPACCF 1 I CPwdA21 I CE562NSCE .

U ICPwEA2J page 6 ,

Fege 3 2

I SERVICE WATEM PUwp A2 FAILS TO Rt4  :

4 * -

I CPan A2..]

O 9 4

.\ TREES \CCWSAMDA.CAF 4-15-92 Page 4 FAILURE OF CCW F I I

t

, . .a..-.,..._.;......_..__.-.. , __:..-.-_ - , ,_ .. , _

. 2- . - . . . _ . , ~ . . . . . . . . . . - , - - - . . . , . - - - . . , . , , -

i t

i . I a i

. i l

i l I i .

i i 4

[

4 r

'1 - [

i

' I CFEAATING C M CN A

lCaVSEFAILUAECr2 OF 2 RUNMING ESw ,

\l PUMPS

! M 5sv2 Fm4PR',CF l p ,

s PoQs 8

]

I ii 1 i

j SEAVICE NATEA PUMP BETA *ACTCA FCA At' FAILS ODEAAT;tG CO' OF 2 '

TO AUN . OF 2 ESW Pir4PS  ;

. i I CPwK

• 1.1 1CES$ PAS W ,

KAsts/st a.a-32

!- [

t

.i .

! - i

. i ,

I o 1  !

FAILURE OF CCW .\ TREES \CCWSAMDA.CAF 4-15 Page 5 i

i I

f r 4

i

' I t

4

, I e  :

r i .

i

.i . .  ;

i I

t

. I i

i  !

l t P

l J 4

! l 4,

P 1

r r ' COMMON CAUSE l i .FAILUAE OF 2 03 2 1 i

STANCBY ESW PUMP 5 J '

5 TO StAnr  ;

I CE5w2PMPSCCF j .

dage 4 Pege 8  !

l f

I1  ! [

SEAVICE WATET4 PUMD BETA FACTDA FCA CCF '

A2 FAILS CF 2 OF 2 SfANO3Y

.TO START ESW PUMPS TO START

• i

! =

.. I CPwd A 2__] ICE 5wPSSF$j I

1 i

2) ,KAG11/92 A.A-32 i i

- t s

2  :

1 i

L

-(

l l

FAILURE-OF CCW .\ TREES \CCWStMDA.CAF 4-15-92 Page 6 i

i

-,,--,m-- .._m, . . ,_,...:. , . - _ . , _ . . . . , . . , - , ~ . -,...,,.m,, ..,&...--7 . . . - - . . . ,, _ . . . . , , _ . . - _ . _ , . - , _ _ . _ . . - - .

I i

e E !

i i I t

• d 4

FAILURE OF TrulN 9 CCW PUMPS i

p 1CCwf9_002 f a

q l-1 i

FAILUAE OF CCW Pt. Pap FAILURE OF CCW PUNP l

Bi 82 ICCwiD0051 1CCwT9006 1 I I 4

1 I i 1

CPEAATING COMMON CCMPONENT CCOLING COMPONENT CC3 LING COMPONENT C0 CLING WATER PWP B2 j WATEA Pl*C 81 CAUSE FAP UAE OF 2 WATER R.PC B2 FAILS TG RUN OF 2 AUi4ING CCW FAILS TO RN FAILS TO START PUMPS I CP6kS1_j l CCw2PMcACCF ! ICPO B2. l lCP942i O A "" " O O 4

i I

l- FAILURE OF CCW .\ TREES \CCWSAMDA.CAF 4-15-92 Page 7 I

I  !

4 I  :

i i

I

' lr a FAILUAE OF TRAIN B ESW PU W ,

v f

\

, p g ICCwfS003 1  !

l 1l I 1 ,

FAILUAE CF ESW PUMP FAILUAE CF ESW PUMP Si E2 ICCwTBOO7 f. ICCwT8009 I a F~ i i CPEAATING COMMON SEAVICE WATER PUMP CCMMJN Cause 4 SEAVICE WATER PUMP l FAILUAE OF 2 CF 2 B1 FAILS -I CAUSE FAILUAE OF 2 ' 82 FAILS f

-to RUN 1 OF 2 RUNNING ESW TO START STAND 8Y ESW PUMPS

• l PUMPS TO START

- 1 CPw* B 2 i LGESw2PwPACCF I I CPd!B21 ICESw2PwP5CCF i )

2\ Page 5 Page 6 I

s 6 SERVICE WATE.A PUwp [

02 FAILS -

TO AUN icP- e2 i O .

I 5

i

}~ -

i FAILURE OF CCW .\ TREES \CCWSAMDA.CAF A-15-92 Page 8 y I

, ~m.. , ,_, , -- .-m._ .. . . ~ . .. . ~ _ _ _ _ . _ _ . _ _ _ _ _ _ _ . - . . _ _ _ . _ _ _ _ _ _ _ _ _ - -